wk? ho. , & | V.> THE PHILIPPINE JOURNAL OF SCIENCE . VOLUME 21 JULY TO DECEMBER, 1922 WITH 44 PLATES AND 11 TEXT FIGURES MANILA BUREAU OF PRINTING 1922 189549 EDITORIAL BOARD ELMER D. MERRILL, M.S., Editor R. C. McGregor, A.B., Associate Editor ALBERT H. WELLS, A.B.; GRANVILLE A. PERKINS, PH. D.; A. P. WEsT, PH.D. T. Dar JuAN, A.B., PHAR.D.: F. Accaorui, A.B.; A. gs, ARGUELLES, B.S. HowArp IRVING Coxe, PuH.D.; ALBERT rE. W. KING oe WARREN D. SMITH, PH.D.; Roy E. DICKERSON, PH.D. VICTORIANO ELICANO, B.S. Geology H. W. WADE, M.D.; Orro ScHésL, M.D. F. G, HAuGHWouT; STANTON YouNGBERG, D.V.M. Experimental Medicine Liporio GomeEz, M.D., Pu.D.; F. CALDERON, B.A. LM. VICENTE DE JESUS, M.D. Clinical Medicine W. H. Brown, PH.D.; C. F. Baker, M.A.; H. ATHERTON Leg, M.S. L. M. GUERRERO, PHAR.D. Botany ALBERT C. HERRE, PH.D.; C. F. BAKER, M.A.; S. F. Ligut, M.A. C. S. BANKs, M.A.; * D. WHARTON, M.A.; W.~ SCHULTZE Zoology H. O. Beyer, M.A.; Orto JoHNs SCHEERER, M.A. Anthropology ANNA B. BANYEA, Copy Editor CONTENTS No. 1, July, 1922 [Issued September 12, 1922.] PERKINS, GRANVILLE A. Manufacture of certain drugs for the treat- WG hl DRIP OUS soso ee hocks coceicsec One plate. CoLE, HowArpD IRVING. Manufacture of industrial alcohol and alcohol motor:fuel in’the Philippime -[elands....i0i.cc:.sicsesdeccaiise cise jeccsecsiancades One plate and three text figures. WELLS, A. H., and PERKINS, G. A. The use of Ae fumes in copra drying ARR CPE CO COREE AAD REE ETE VE pk CEPA a ate Bere Two plates. Wiliam. J: Bispiien Ger ater We a SHAW, WALTER R. Merrillosphaera, a new genus of the Volvocaceae.. Eight plates and one text figure. No. 2, August, 1922 [Issued September 29, 1922.] Sypow, H. The Amboina fungi collected by C. B. Robinson WERNER, F. Philippine mantids, or praying insects One plate. TAYLOR, EpwArpD H. Additions to the herpetological fauna of the ge FES GS Saba eine fee rats Senee marten ne Seven plates. SHAW, WALTER R. Copelandosphaera, a new genus of the Volvo- Caceac. .... Four plates and two text figures. No. 3, September, 1922 [Issued October 18, 1922.] GOMEZ, LiporRIO; BAsA, JOSE AVELLANA; and NICOLAS, CATALINO. Early lesions and the development and incidence of leprosy in the children of lepers Six plates. TAYLOR, Epwarp H. Additions to the herpetological fauna of the Pepto. Teas, | TE isc. giants Four plates. iii Page. 7 49 57 87 131 147 161 207 257 iv Contents ABRIOL, RUFINO. Correlation of death rates from certain diseases with certain economic and housing factors in the Philippine (OOS Ties [> Spears tots Css Uipiee seater MO eda AYES AER Seen ree oe ie ear Ree One text figure. WeLD, Lewis H. Notes on the Liopterinae with descriptions of new species from the Oriental Region (Hymenoptera, Cynipidae) One plate. No. 4, October, 1922 [Issued October 18, 1922.] MERRILL, ELMER D. Diagnoses of Hainan plants..............002..2.00-220.00...+- Wirt, J. C. The effect of sulphur compounds on cement CoLE, Howarp Irvinc. The use of textile fibers in microscopic qual- itative chemical analysis. V. The detection of gold by means of stannous chloride-pyrogallol viscose-silk fibers.........................- Wirt, J. C. Some generalizations on the influence of substances on cement and concrete. Second paper. ALEXANDER, CHARLES P. New or little-known Tipulidae from the Philippines (Diptera) Voss, Epuarp. Indo-Malayische Rhynchitinen (Curculionidae). I, Siebenter Beitrag zur Kenntniss der Curculioniden = Drei textfiguren. No. 5, November, 1222 [Issued November 15, 1922.] RoHwER, S. A. Philippine parasites of the family Trigonalidae........ TAKAHASHI, RYOICHI. Some PEMAR GAR ADDIUORC oon cede ec WEISE, J. Chrysomeliden der Philippinen, III MERRILL, ELMER D. Notes on the flora of southeastern China............ One text figure, No. 6, December, 1922 [Issued December 23, 1922.] MERRILL, ELMER D. Additions to our knowledge of the Bornean flora MENDOZA-GUAZON, MARIA PAz. Schistosomiasis in the Philippine PSlends Gee Five plates. ScHULTZE, W. X. Beitrag zur Coleoptera Fauna der Philippinen... Vier Tafeln. Page. 305 323 337 357 361 365 373 385 A417 421 423 491 515 535 569 THE PHILIPPINE JOURNAL OF SCIENCE VOL. 21 JULY, 1922 No. 1 MANUFACTURE OF CERTAIN DRUGS FOR THE TREATMENT OF LEPROSY ' By GRANVILLE A. PERKINS Chemist, Bureau of Science, Manila ONE PLATE The chemical investigation of leprosy drugs by the Bureau of Science ? has been continued since Brill’s latest paper, especially since May, 1920, when the work was placed in the hands of the Committee on Leprosy Investigation, a joint committee of the Philippine Health Service, the University of the Philippines, and the Bureau of Science. Since that time the Bureau of Science has prepared a number of drugs, which were used ex- perimentally at San Lazaro Hospital and Culion Leper Colony. Practically all of these have been found to be beneficial, in vary- ing degrees. The report of this work from the medical point of view, including the methods of administration and observed effects, will be published by the committee. The present paper is written from the chemist’s standpoint, and takes up the preparation and chemical composition of the drugs used. The treatment of leprosy in recent years has tended away from the old methods of oral administration and inunction toward the injection method, both intramuscular and intra- venous. Although the manner of application has changed, the * Received for publication January 28, 1922. * Brill, H. C., Hydnocarpus venenata Gaertner; false chaulmoogra, - Philip. Journ. Sci. § A 11 (1916) 75; A chemical investigation of the seeds of Pangium edule and of Hydnocarpus alcalae, ibid. § A 12 (1917) 37; Brill, H. C., and Williams, R. R., The use of chaulmoogra oil as a specific for leprosy, ibid. § A 12 (1917) 207. 187663 1 2 The Philippine Journal of Science 1922 basis of the drugs most used has remained the same for centu- ries, chaulmoogra oil. Rock® has found an old Buddhist account of “a Burmese king voluntarily exiled for leprosy about 1,000 years ago who cured himself with the [chaulmoogra] oil, and likewise effected the cure of a beautiful young woman whom he afterwards married, founding a dynasty.” CHAULMOOGRA OIL This oil, according to the present official definition, is the fixed oil from the seeds of Taraktogenos kurzii, a large tree which grows in the Assam Valley and Chittagong Hill Tracts, India. Similar oils, not correctly called chaulmoogra but perhaps quite as effective, are obtained from Hydnocarpus wightiana,* grow- ing on the Malabar coast; H. venenata,?*+ Ceylon, Deccan, and Burma; H. castanea,t Burma; H. anthelmintica,t Siam and French Indo-China; Asteriostigma macrocarpa, Travancore; H. alcalae,? Albay, P. I.; H. hutchinsonii,» Mindanao, P. I.: A. sub- falcata,’ Zambales, P. I.; and Onchoba echinata,*® Sierra Leone. " Pangium edule,? found in the southern Philippines and neigh- boring islands, gives an oil somewhat similar in composition, but the oil from Gynocardia odorata, which was long confused with chaulmoogra, is totally different, from both the chemical * and the bacteriological * standpoints. Chaulmoogra oil, like most fixed vegetable oils, is composed nearly entirely of fatty acids combined with glycerine. The principal fatty acids in the oil from Taraktogenos kurzii and a few closely related species, most of which are mentioned above, are of a peculiar type known as the chaulmoogric acid series. * Rock, J. F., U. S. Dept. Agr. Weekly News Letter 9 (1921) 1.- *Muir, E., Handbook on Leprosy. Orisa Mission Press, Cuttack, India (1921) 39. *A chemical investigation of these oils will be reported in the near future. “Goulding and Akers, Chem. Soc. Proc. 29 (1913) 197. _ "Power, F. B., and Barrowcliff, M., Journ. Chem. Soc. 87 (1905) 897. The sample of MGyneicirdhd, odorata” sodda obtained by Brill and Williams, Philip. Journ. Sci. § A 12 (1917) 211, from the Department of Agri- culture at Assam was evidently Taraktogenos or Hydnocarpus. Since the external appearance of all these seeds is somewhat similar, it is important to note the easy distinction pointed out in Muir’s Handbook, p. 39. The radical of the Gynocardia seed is lateral, while that of the Tarakto- genos and Hydnocarpus is terminal. *Unpublished work with tubercle bacilli by Dr. O. Schdbl, Bureau of Science. 21,1 Perkins: Drugs for the Treatment of Leprosy 3 The acids of this series are found nowhere else in nature, nor have they been synthesized in the laboratory. Expressed in terms of molecular structure, the distinguishing characteristic of these acids is that they contain a closed ring of. five carbon atoms. All other known fatty acids have their carbon atoms in the form of an open chain. Due to this structural peculiarity, the acids of the chaulmoogric series have the power of optical rotation, and may be detected by the polariscope. The nomenclature of the peculiar acids of chaulmoogra oil was initiated in 1879 by Moss,° who supposed chaulmoogra oil to be derived from Gynocardia odorata. He gave the name “gyno- cardic acid” to a crystalline preparation melting at about 29° C. This was later shown to be mixture, by Power and his collaborators,’° who definitely isolated from chaulmoogra oil the acids which they named chaulmoogric and hydnocarpic. They considered the glyceryl esters of these acids, together with some palmitic, to be the chief constituents, of chaulmoogra oil. The fact that some Hydnocarpus oils are very similar to Ta- raktogenos oil has led to the claim that these may be legiti- mately sold as chaulmoogra oil. If the word chaulmoogra is to be defined by its original native meaning, this claim can hardly be denied; but the modern practice of standardizing medical terms necessitates the restriction of the word to Ta- raktogenos kurzii. The question of the medicinal value of the various Hydnocarpus oils must be decided on their own merits. Chemical investigation has shown that they are very similar to chaulmoogra oil and contain at least two of the same peculiar fatty acids which have been proved to be effective in the treat- ment of leprosy. The optical rotatory power is the best present single analytical criterion of this property, though the results of McDonald and Dean“ indicate that the iodine absorption | value is also important. The writer expects to publish an analytical study of various Hydnocarpus oils in the near future. The clinical results on record are not confined to chaulmoogra oil; in fact, experts in India have in recent years preferred Hydnocarpus wightiana oil. This is encouraging to us in the Philippines, as we have no indigenous Taraktogenos kurzit, but ®* Moss, Year-book of Pharmacy (1879) 523-533. ” Power and Gornall, Journ. Chem., Soc. 85 (1904) 838, 851; Power and Less, ibid. 87 (1905) 349; Power and Barrowcliff, ibid. 884; Barrowcliff and Power, F. B., Am. Journ. Pharm. 87 (1915) 493. ™ McDonald, J. T., and Dean, A. L., Journ. Am. Med. Assoc. 76 (1921) 1472. 4 The Philippine Journal of Science 1922 we have Taraktogenos heterophylla and a commercial supply of Hydnocarpus. USE OF THE OIL IN THE TREATMENT OF LEPROSY The use of chaulmoogra oil orally and externally in leprosy has so long been favored by the medical profession that it is not necessary to dwell on the detailed reports of beneficial re- sults. Although its action has been so slow and uncertain that some physicians have had doubts of its efficacy, yet the consensus among them has been that chaulmoogra has a definite benefi- cial effect in leprosy, not produced by any other oil except per- haps the Hydnocarpus oils. This fact at once points to some specific effect of the peculiar fatty acids of chaulmoogra on the bacillus of leprosy. A remarkable confirmation of this idea is seen in the recent experiments of Walker and Sweeney,'? who have demonstrated that the sodium salts of the chaulmoogra fatty acids are about one hundred times more bactericidal to acid-fast bacteria than is phenol. They further found that so- dium salts of other fatty acids are practically inactive, and also that the action of chaulmoogra acids on acid-fast bacteria does not extend to other common pathogenic bacteria. They con- clude (p. 259) that “any therapeutic action which chaul- moogra oil may have in leprosy is due to its direct antiseptic and bactericidal action on B. leprae.” As to the “active prin- ciple” of the oil they decide that “the bactericidal activity is a function of the chaulmoogric acid series.” To explain the specificity against acid-fast bacilli they prepose the hypothesis that “these fat-elaborating bacilli attempt to utilize the chaul- moogric acids to build up their fatty capsules, and that these cyclic fatty acids contain a group or an arrangement of atoms which is toxic for the bacterial cell.” Lastly, they conclude that this toxicity is unique with the chaulmoogric acid series, and not shown by the ordinary fatty acids; for example, those found in cod-liver oil. While these four conclusions of Walker and Sweeney are based _on extensive and careful experiments, they do not agree with opinions that have been based on clinical evidence. Practical experience in treating large numbers of lepers has led to various theories, of which those of Mercado and of Rogers, respectively, will be mentioned as examples. Mercado explains the action of chaulmoogra oil as follows :"° “ Walker and Sweeney, Journ. Inf. Dis. 26 (1920) 238, “Mercado, E., Leprosy in the Philippines and its Treatment. Manila (1915) 55. 21,1 Perkins: Drugs for the Treatment of Leprosy 5 The organism repels the bacillary invasion due to the concurrence of a larger number of defensive elements as the leucocytes which possess phagocytic powers to destroy and exterminate the bacilli. In order that the organism may acquire a greater number of leucocytes for its defense, a stimulant or reactive is needed to provoke leucocytic hypergenesis, such as we are able to acquire from the presence and action of a fatty substance as the oil of chaulmoogra which also contains gynocardic acid, an alterative, acting upon the system as all other organic acids. Other leprologists assume a similar, indirect effect. Rogers, on the other hand, believes the effect to be direct, but not confined to chaulmoogra oil. He states** that there is nothing absolutely specific against leprosy in the products of chaulmoogra oil and— that the sodium salts of the unsaturated fatty acids of these two oils [chaulmoogra and cod-liver oils] act in some way on the coating of the acid-fast bacilli, that of tuberculosis having been shown to contain palmitic and other unsaturated fatty acids. During the past year [1918] sodium morrhuate has been used with very promising results in tuberculosis by several careful observers to whom I have supplied it, and morphological changes have been noted in the tubercle bacilli in the sputum. A wide field of investigation has thus been opened out, as it appears to be possible that the organisms of other chronic diseases, including that of syphilis, might possibly be broken up in the system by similar methods. By “sodium morrhuate” Rogers means the sodium salts of the principal fatty acids of cod-liver oil. He has later used successfully also ethyl esters made from cod-liver oil and so- dium salts from soya-bean oil.'® The point which seems important to Rogers is that chaul- moogric and hydnocarpic acids are unsaturated; that is, they have the power of combining with iodine, bromine, hydrogen, and oxygen by direct addition of these elements; whereas a saturated fatty acid, like palmitic, for example, is comparatively inert chemically. Unsaturation is also a property of the acids in cod-liver, soya-bean, and many other oils. While Rogers’s theory is not supported by such definite ex- perimental evidence as that of Walker, yet the fact remains that sodium morrhuate has an effect in leprosy and tuberculosis very similar to that of the chaulmoogra preparations. Walker’s conclusions are therefore not strictly correct, but they at least furnish a scientific starting point toward the solution of the problem of leprosy treatment. * Rogers, L., Brit. Med. Journ. 1 (1919) 147. * Rogers, L., Ind. Med. Gaz. 55 (1920) 127. 6 The Philippine Journal of Science 1922 METHODS OF TREATMENT BASED ON CHAULMOOGRA AND OTHER OILS Until recently chaulmoogra oil was used principally in the crude form and was administered chiefly by mouth. The dis- advantages of this method were that the nauseating properties of the oil seriously limited its use, and the action, even of large amounts, was slow. To obviate these difficulties, a number of other methods of administration were tried, in which the first real success was obtained by Mercado who used the formula (p. 54): “Camiphorated oi] 10 per? comb... seccscc- ccc Chaulmoogra oil, pure and sterile... baa 60 cc Resoreite sis Slee i I ee ee 4 grams ne i gi ee ee ee 2.50 c.c The camphorated oil is added because it “on account of its oils, helps the oil of chaulmoogra and besides that, it acts inde- pendently upon the heart and is at the same time, a recognized therapeutical agent which constantly increases phagocytic ac- tivity.” Heiser states 1° that camphor and ether were suggested by the Merck Co., to assist in the absorption of the chaulmoogra oil. The resorcine was used for its antiseptic properties. The intramuscular method is generally conceded as an ad- vance on the oral administration of chaulmoogra oil, both be- cause it avoids the nauseating effect and because it produces more-rapid results. The Mercado mixture has been found less irritating and is more easily absorbed than the pure oil injected in the same manner, although the pure oil, carefully prepared, is still injected intramuscularly in the asylums of Japan, where it is preferred on account of its more-concentrated form. While the pure oil can be used intramuscularly, the crude com- mercial oil ordinarily sold is unsuited for such use, even though applied in the form of the Mercado mixture. A widespread idea that the crude oil is more efficacious has seriously inter- fered with the intramuscular treatment of leprosy. Heiser states :17 Experience with chaulmoogra oil at San Lazaro Hospital, when admin- istered by mouth, has shown that the crude oil is much more efficacious ‘ than the refined product. When used hypodermically, there is apparently no difference whether the crude or the refined oil is used, but accurate data with regard to this point are not yet available. *U. S. Pub. Health Rept. 29 (1914) 2765. “Am. Journ. Trop. Dis. 2 (1914) 324, 21,1 Perkins: Drugs for the Treatment of Leprosy 7 The committee has made experiments using Mercado mixture prepared from chaulmoogra oil refined at the Bureau of Science, and from a high-grade oil made by Shiongi & Co., Osaka, Japan, which is either pressed from carefully selected seeds or else refined. In both cases it was found that the irritation was negligible as compared to that produced by the mixture made from ordinary commercial chaulmoogra oil. The free fatty acid was found to be one of the very irritating constituents of the crude oil. The fact that free fatty acid is more easily digested than neutral oil probably explains the superiority of the crude oil for oral administration. Using the Mercado mixture as a standard for comparison, the committee has conducted experiments with seven newer preparations, all of which are designed for more-rapid absorp- tion than occurs with the Mercado mixture. These are: Chaul- moogra ethyl esters, Muir’s E. C. C. O., sodium gynocardate A, sodium gynocardate S, chaulmoogra emulsion, sodium morrhuate, and cod-liver ethyl esters. The composition and chemistry of each of these will be considered in turn. CHAULMOOGRA ETHYL ESTERS '* ‘The mixed ethyl esters of the fatty acids of chaulmoogra oil came on the market several years ago as “antileprol” ° made by the Bayer Co. This was found by Mercado” to have no better effect than the oil when taken orally. Tests on two pa- tients were made intramuscularly without results, but in these cases only 2 per cent and 5 per cent solutions were used. The ethyl ester treatment has been used very successfully -by Dean and his collaborators at Honolulu,”! and the committee has obtained excellent results with it at San Lazaro Hospital. The ethyl esters seem to be more easily absorbed than the oil, perhaps because they form a more-mobile liquid, which is easily emulsified. They have even been used intravenously by Muir.* The method usually employed in making these ethyl esters seems to be to prepare the soap, then the fatty acids, and finally to esterify with the help of dry hydrochloric acid gas. The * With the assistance of Mr. Ernesto Paras. * Patents, D.R.P. 216,092, and B.P. 1984-1909, were granted for these ethyl esters, although Power had long before prepared them, and published aceounts of his work. *Op. cit. 43. * McDonald and Dean, U. S. Pub. Health Rept. 35 (1920) 1959; Holl- mann and Dean, Journ. Cutaneous Dis. 37: 367-373. “ Handbook on Leprosy. Orisa Mission Press, Cuttack, India (1921) 46. 8 The Philippine Journal of Science 1922 writer has found this method cumbersome on the large scale, and at present esterifies the oil directly, with the help of sul- phuric acid. Ten kilograms of crude chaulmoogra oil, 13 liters of 95 to 96 per cent alcohol, and 750 cubic centimeters of 66° Bé. com- mercial sulphuric acid are placed in a 26-liter stoneware acid carboy which has been fitted with a sheet iron steam jacket and a large glass reflux condenser. A slow current of low- pressure steam is allowed to enter at the top of the jacket, provi- sion being made at the bottom for escape of the condensed steam. The contents of the carboy are thus boiled gently for twenty hours. The reaction that slowly takes place is the well-known one of interchange of alkyl groups under the influence of a mineral acid (hydrogen ion) catalyst. The glyceryl radical in the gly- cerides of the oil changes place with the ethyl group in the alcohol, forming the desired ethyl esters and glycerol. The reaction product is syphoned into a large glass carboy, washed with about an equal volume of water and then with dilute caustic soda, just sufficient soda being added to produce a permanently alkaline reaction to phenolphthalein. The re- sulting emulsion separates fairly well after a few hours, and the upper layer is heated in a steam kettle to dry it. This dry ester mixture is then distilled at about 50 millimeters pressure in a still made from a 30-centimeter (12-inch) length of 15-centimeter (6-inch) iron pipe capped and fitted with a reducing coupling. Gas heating has been used, but we have obtained a more-rapid output by the use of an electric heater (see Plate 1) and an electrically heated still head of 5-centi- meter (2-inch) pipe, 150 centimeters (5 feet) long. Rolls of iron-wire netting are placed in the stil] head to prevent foaming over, which is the troublesome feature of this distillation. The heater is made from 20-centimeter (8-inch) stovepipe 45 centi- meters (18 inches) long and 40-centimeter (16-inch) stovepipe 52 centimeters (22 inches) long, the space between the two walls being filled with asbestos fiber. The heating units are of No. 20 chromel A wire, 500 watts each, one flat coil at the bot- tom, three coils around the 20-centimeter stovepipe, and one coil around the still head. A Crowell 2-D vacuum pump has sufficient capacity for two of these stills. The ester is allowed to enter the still by means of a tube running through the still head, in portions of about 200 cubic 21,1 Perkins: Drugs for the Treatment of Leprosy 9 centimeters, each portion being distilled before the next one is added. When properly adjusted the still delivers at the rate of 4 liters an hour. After about 50 liters have been distilled the still head is taken off and the accumulated pitch cleaned out from the body of the still. Solvents are not very effective, so the pitch is liquefied by heat and removed by mechanical means. The distilled esters contain volatile impurities which were found to be irritating. They are therefore placed in a stoneware carboy and agitated vigorously by a current of steam for two hours. Sometimes we wash the esters again with alkali at this point, but as the free fatty acid is seldom more than 1 per cent this step is generally omitted. In either case the product is freed from the accompanying water in a separatory funnel, and then filtered. The clear, almost colorless, filtered esters are next placed in small bottles, loosely corked, and heated to 150° C. Each cork is then pushed down and paraffined. In this form the ester is ready for use. The ethyl ester is used not only in the pure form but also mixed with 2 per cent iodine, which is the standard preparation used by McDonald and Dean. The iodine is added to the dry ester, in which it gradually dissolves, combining chemically. We recover alcohol from the first washings mentioned above _by neutralizing with lime, and fractionating through an iron pipe 240 by 15 centimeters (8 feet by 6 inches) filled with broken glass. The alkaline wash water is collected in a barrel and acidified to break the emulsion. A mixture of ester and fatty acid rises to the top, and is skimmed off for reésterification along with another batch of oil. About 5 per cent requires reésterification in this manner. The theoretical yield of ethyl ester is 1.05 kilograms per kilo- gram of chaulmoogra oil. A small amount of the oil remains as glyceride, which, along with unknown impurities and decom- position products from the ethyl ester, account for the accumula- tion of tar in the still. The amount of glyceride is low, and the decomposition of ethyl ester is also very small if a good vacuum is maintained. Under average working conditions 50 kilograms of oil have been found to give 2 kilograms of residue in the still. Together with the usual mechanical losses in handling, this cuts down our working yield to 90 or 95 kilograms of finished ester per 100 kilograms of chaulmoogra oil. 10 The Philippine Journal of Science 1922 COMPARISON OF HYDROCHLORIC ACID WITH SULPHURIC ACID AS CATALYST . Strong sulphuric acid has a tendency to form addition prod- ucts with unsaturated compounds, and is known to react with chaulmoogra oil. In order to ascertain whether such undesir- able action takes place under the conditions which we use for the esterification, experiments have been conducted comparing the sulphuric acid catalysis with that of hydrochloric acid, which does not have a similar combining power. The results of typical parallel experiments are shown in Table 1. In this ’ ease 1 liter of 95 per cent sulphuric acid instead of the usual 750 cubic centimeters of 93 per cent was employed as catalyst, and boiling was continued for forty-one hours to intensify the effect. The control experiment was done in the same manner, but instead of adding sulphuric acid dry hydrochloric acid gas was passed into the alcohol until it showed 5 per cent by titra- tion. Thirteen liters of 96 per cent alcohol and 10 kilograms of oil were used in each experiment. The oil was presumably pure Taraktogenos, being guaranteed by the maker, P. K. Sen, of Chittagong. TABLE 1.—Comparison of hydrochloric acid with sulphuric acid as catalyst. moogra Bate weebed with - Distilled enter: Catalyst osc. ogo ce et é HCl HeSO4 HCl HeSO4 Bpecifie gravity 80/49 Goes a es ee 0. 890 0.891 Refractive index 80° C. Dline_______.____.____ 1. 4767 1. 4575 1, 4570 1.4568 1.4566 Optical rotation, D, 100 mm oi 46.0 40 38 38.1 38.9 Iodine No., Hanus _._........--..-------...--..| 102 _ 98 94 97.5 | 9 Saponification No.2.) 198 187 190 187 188 Acidity, as-per cent oleie 22000 18.8 4.4 2.1 1.38 0.51 Freezing point of fatty acids, °C_._____._____. 29 29 30 30 31 Specific rotatory power of fatty acids dis- BOlVOd Ii S7IONG pcs cidseced. sscsas. exer ac SET 46.8 47.3 45.5 45.5 The data given in Table 1 show that sulphuric acid, under the given conditions, does not appreciably change the chaul- moogra fatty acids. The figures for the rotation of the washed esters are approximate, as they were too dark for accurate determination. The optical rotation of the hydrochloric acid esters decreased somewhat upon distillation. This decrease has been noted to occur when distillation is delayed by accidental difficulty with the vacuum, which happened in this run. The 21.1 Perkins: Drugs for the Treatment of Leprosy 11 same may be said of the occurrence of free fatty acids in the distilled product. They were produced during distillation and have no relation to the free fatty acids in the crude products, as these were washed thoroughly with alkali before distillation. The iodine numbers of the esters and the rotatory powers of the fatty acids show clearly that the use of sulphuric acid as a catalyst is permissible, even on long boiling. Sulphuric acid has obvious economical advantages as compared with hydrochloric acid gas and reduces, by its hygroscopic action, the amount of free fatty acid in the crude esters. MUIR’S E. C. C. 0. The preparation and use of this mixture is described in Muir’s Handbook on Leprosy, page 43. It is composed of Hydnocarpus wightiana ethyl esters, 1 cubic centimeter; double distilled creosote, 1 cubic centimeter; camphor, 1 gram; olive oil, 2.5 cubic centimeters. The committee has recently placed a group of patients under treatment with this drug. SODIUM GYNOCARDATE A This is the derivative of chaulmoogra oil used for subcuta- neous and intravenous injection since 1917 by Leonard Rogers.” Recently he has called it sodium hydnocarpate,”* but as it is really a mixture containing a large amount of sodium chaul- moograte there seems to be no good reason for changing the name. The name “gynocardate”’ has always been applied to a similar mixture, while “hydnocarpate” applies to a definite chemical individual. To prepare this drug, chaulmoogra oil is saponified by boiling it in a steam kettle with about four parts of water, adding one-fifth part of caustic soda in small portions. The fatty acids, which solidify at about room tem- perature, are liberated by the addition of sulphuric acid, then dried, and dissolved in alcohol. The solution is cooled in a refrigerator, which causes ‘the less-soluble acids, particularly chaulmoogric, to crystallize out. Several fractions are crystallized out, and the least soluble is mixed with sufficient of the intermediate fractions to give a product melting at 48° C. This, after neutralization with caustic soda, is a duplication of Rogers’s preparation, as closely as can be ascertained. * Rogers, L., Ind. Journ. Med. Res. 5 (1917) 277. * Rogers, L., Ind. Med. Gaz. 54 (1919) 165. 12 The Philippine Journal of Science 1922 Sodium gynocardate A is used in the form of 3 per cent sterilized solution, containing 0.5 per cent phenol, and 0.5 per cent sodium citrate to avoid clotting. SODIUM GYNOCARDATE §S This preparation was made in the same manner as sodium gynocardate A, except that the total fatty acids instead of the crystallized acids were employed. As the preparation was found to be irritating, the committee soon decided to use in its place a similar solution (sodium gynocardate D) differing only in that the fatty acids have been purified by distillation in vacuum. The irritating substances are thus removed to a large extent. CHAULMOOGRA EMULSION About the time that Rogers was beginning the intravenous injection of chaulmoogra soap solutions, Vahram”*° was using a very fine emulsion of chaulmoogra oil in the same manner. A mixture of the oii with 20 parts by weight of acacia was desiccated to cause intimate union, and then mixed with 1,390 parts of water, forming a very dilute emulsion. This prepara- tion is now on the market as “Collobiasis of Chaulmoogra,” made by the Laboratoires Pharmaceutiques de Dausse, Paris. This “Collobiasis” has been used by the Committee on Leprosy Investigation for direct infiltration in the lesions and, to a smaller extent, intravenously. A similar chaulmoogra emulsion, prepared at the Bureau of Science, has also been used for in- filtration. SODIUM MORRHUATE Cod-liver oil has not been used to any large extent in leprosy, but it has had wide application in the treatment of tuberculosis. It was probably for this reason that Rogers 2° chose cod-liver oil as a basis for a preparation similar to sodium gynocardate for use in tuberculosis. The new drug, which he named sodium morrhuate, has given good results in the treatment of this disease, and even better results in the treatment of leprosy. The composition of cod-liver oil is very complex. It contains glyceryl esters of the common fatty acids, palmitic, stearic, and oleic, together with a number of characteristic fish-oil acids *Vahram, M., Progrés Médical (1916) 19. New Orleans Med. and Surg. Journ. 69: 230. * Rogers, L., Ind. Med. Gaz. 53 (1918) 73. 21,1 Perkins: Drugs for the Treatment of Leprosy 13 (clupanodonic and others) which have not been so fully in- vestigated.2? These characteristic acids all have the property of unsaturation, which has been previously discussed in con- nection with chaulmoogra oil. The sodium morrhuate used in the experiments of the Com- mittee on Leprosy Investigation has been prepared by Rogers’s method.”* . This consists in making the sodium soap of cod-liver oil and extracting with ether to remove irritating substances. A 3 per cent solution has been used, principally by intravenous injection. COD-LIVER OIL ETHYL ESTERS This preparation has been recommended by Rogers ** as being less irritating than chaulmoogra ethyl esters. The committee obtained the opposite result, probably because of differences in the methods of preparation. The addition of 2 per cent iodine, however, seems to destroy the irritating constituents, so this mixture is now being used by the committee. CONCLUSION In conclusion, the writer wishes to point out that the success recently attained in the treatment of leprosy and tuberculosis with drugs derived from certain oils is encouraging, not only in itself but also in the prospect for improvement that it promises. If Walker’s idea of the absorption of toxic fatty substances by acid-fast bacilli is correct, it should be easy to make in the labor- atory a much more effective drug by properly combining a toxic element like arsenic or antimony with a fatty acid. If Rogers’s idea of the efficacy of unsaturation in general proves to be correct, oils may be used that are much more unsaturated than chaulmoogra oil. The conclusions as to the value of the various drugs described above must of course be deferred to the medical report of the committee, but it may be stated that the intra- muscular ethyl ester treatment has been decided upon for the main routine treatment at Culion Leper Colony. About 200 liters (40,000 doses) per month are now being manufactured at the Bureau of Science for that purpose. ™ Lewkowitsch, J., Chemical Technology and Analysis of Oils, Fats & Waxes. London 2 (1914) 424. * Rogers, L., Brit. Med. Journ. 2 (1919) 426; Ind. Journ. Med. Res., Special number, 7 (1919) 238. ” Rogers, L., Ind. Med. Gaz. 55 (1920) 127. 14 The Philippine Journal of Science SUMMARY 1. Recent developments in leprosy treatment are discussed from a chemist’s viewpoint. 2. The manufacture of chaulmoogra ethyl esters is described in detail. 3. The preparation of six other drugs, mostly following the procedure of other investigators, is described, these being the other medicines chosen by the Committee on Leprosy Investiga- tion for experimental treatment. PERKINS: DruGs For Leprosy.] [PuHiuie. Journ, Scr., 21, No. 1. PLATE 1. VACUUM DISTILLATION OF ETHYL ESTER AT THE BUREAU OF SCIENCE MANUFACTURE OF INDUSTRIAL ALCOHOL AND ALCO- HOL MOTOR FUEL IN THE PHILIPPINE ISLANDS By HowaArp IRVING COLE Chemist, Bureau of Science, Manila ONE PLATE AND THREE TEXT FIGURES INTRODUCTION Alcohol has been used for beverage purposes in the Philippine Islands since the earliest times. The production of alcohol for beverage and other purposes constitutes at the present time a very important industry, as is readily seen from Table 1 which shows the amounts of denatured (industrial) alcohol and fer- mented liquors produced annually from 1909 to 1920, inclusive. TABLE 1—Distilled liquors and malt beverages manufactured in the Philippine Islands and the taxes collected upon them. Distilled spirits. ‘ Denatured | Fermented ahs Manufac- | Tax col- | cohol. | Hauors. tured. lected. Proof Gauge liters.2 Pesos.» |Proofliters.| liters. 1909_- ANCES ef ip Smarties: Tere ee 9, 532,537 | 1,899,043 26, 672 | 3,636, 499 sb elt Sia clee i net Set Gepiperoa eae eae eese ge bee pao 10, 584,124 | 2,269, 160 183, 682 | 3,837, 000 Rr ss ec eek Soci senens yenapeevens 10, 471,299 | 2,551,501 380, 099 | 3, 887, 500 Re as geek eee wenn eeees ad aeeu 11, 426,728 | 2,718,061 | ©¢721,262 | 4,466,250 es ead ca ace coe eed en ae wandcons ines 11, 276,298 | 2, 658, 066 661,790 | 3, 663, 750 hae a es cameras eee ee 11, 847,354 | 2,729,705 820, 148 | 3,915, 000 BOG oo cncnka pee bodacadde eeacd polebieeesnwnencns 7, 164,728 | 2,248, 499 801, 428 | 3, 667, 7038 WONG ase cee soo sn ead op cu eenbw eneen 10, 350, 868 | 2, 747,396 735, 753 | 5, 720, 769 Ne ee saw atacc nha eunnescbee cament 15, 078, 758 | 4, 189, 503 966, 549 | 5, 875, 200 st + Sergei a ae aoe agate eos Sei Op in ta geet 15, 931,402 | 4,476,443 867,971 | 4,236, 200 BO a ueea cas wsaupeukecsnucuase 18, 418,612 | 4,318,895 638,217 | 4,378, 800 rs i ee een 16,019,612 | 3,976,988 | 4522,412 | 5,221, 865 @Includes denatured alcohol exported tax free and medicinal tax-free alcohol. The proof spirit referred to throughout this article is that fixed by section 3249 of the Revised Statutes of the United States which defines proof spirit as containing one-half its volume of alcohol of a density 0.7939 at 60° F. The British proof spirit contains 49.24 per cent alcohol by weight. 2 : ‘ > One peso Philippine currency equals 50 cents United States currency. € Includes tax-free alcohol used for medicinal purposes. 4 Exempt from specific tax but subject to 1 per cent on gross sales. 187663-——2 17 18 The Philippine Journal of Science 1922 A large increase in the production of industrial alcohol can be confidently looked forward to, mainly because alcohol mixtures are being used in internal-combustion engines as fuel to replace gasoline. Incidentally, there is always the possibility of pro- hibition being extended to the Philippines, which would result in decreased production of beverage alcohol and increased pro- duction of industrial alcohol. It is not the purpose of this article to discuss the production of alcohol for beverage purposes. On the other hand, a great deal of the information given applies to the manufacture of - beverages from the rectified alcohol as well as to the manufac- ture of industrial alcohol and alcohol mixtures for motor fuel. RAW MATERIALS FOR THE PRODUCTION OF ALCOHOL The largest source of raw material for the production of alcohol in the Philippine Islands is undoubtedly the saps of the various native palms; of these, the sap of the nipa palm is most used. Another large source of alcohol is the molasses obtained as a by-product in the manufacture of sugar from sugar cane. Various cereals, cassava, bamboo, and other cellulose materials are also potential sources of alcohol. In fact, the potential sources of alcohol in the Tropics are illimitable, as can be readily appreciated by any person who has traveled in tropical countries. Table 2 shows the quantity and kind of raw material used in the production of alcohol in the Philippine Islands and the amount of alcohol produced during 1919 and 1920. TABLE 2.—Quantity and kind of raw material used in the production of alcohol in the Philippine Islands in 1919 and 1920. 1919 1920 Source of alcohol. Quantity |Proof aleohol| Quantity |Proof alcohol used. produced, used. produced. Liters. Liters. Liters. Liters. Nipa sap __... 57, 770, 157 5, 887, 208 51, 730, 608 5, 148, 315 C008 ONG sss Greeks ces oaccas ose -|- 8,500, 159 456,980 | 4,309, 671 525, 643 MOUNSERO 2 a ae 11, 004, 256 6, 778, 878 18, 058, 566 9, 078, 297 Kilos. Kilos, Flared 1aclaneeec. a5 et eo eh ae Ee ee 1,378, 925 924, 888 Cen SON ee 501, 347 278, 877 516,979 326, 844 ASPMIN Sie ics eee de kee ae 2, 775 481 3, 996 1, 921 Liters. Liters Others 0000S ea ee 18, 728 16, 688 16, 206 13, 275 . ‘Total produchiots. s.r 30; Eid, O18 fo cuceie cece. 16, 019, 183 In 1910, 93 per cent of the total amount of alcohol and alcoholic beverages produced in the Philippine Islands was distilled from 21,1 Cole: Manufacture of Industrial Alcohol 19 the sap exuding from palm trees. In 1920, this value had fallen to 35.4 per cent, due mainly to the increased use of molasses. The utilization of palm saps at present is confined to a few small areas. There are great tracts well situated for exploita- tion which have not yet been touched. The cost of production of alcohol from various sources is given in Table 3. TABLE 3.—The cost in dollars," United States currency, of the various kinds of raw materials required to produce alcohol.” Cost of the raw ma- terial required to produce 1 liter. 100 proof. | 180 proof. Sugar beets at 5 dollars a ton; sugar content, 14 per cent. -.-------------- 0. 032 0. 058 Soreham stalks at 8 dollars per tons: -2220< 5 oc ses sees ee se 0. 032 0. 059 Savarcine at B25: doilars per ton. csc ke dete coca nnee- “esse 0. 028 0. 050 Beat miolnsoes Bt 16 Goss per Olas 526 cine a cine ne enc esa ee ekcee 0. 029 0. 053 Cane molasses at 0. 12 dollar per gallon¢ _______..-.----.---- ------ -------- 0, 032 0. 059 Jerusalem artichokes at 5 dollars per ton _-.-..--------.------------------ 0.027 0. 048 Cassava at 6 dotlara D6¥ tii 25.2 ces ass car tac cee de cect ce eee det ee 0.019 0,084 Potatoes at 4 dollara ner ton for culls ..5<....--5- 2+ = e- .- -5-5 0.021 0.037 Sweut-potatoes at § dollare per ton cs.2 is cs cca ete ee oe ce 0. 031 0. 056 Barley at 0.65 dollar per bushel of 48 pounds-----...--.--------. ---------- 0.041 0.074 Maize at 0.70 dollar per bushel of 56 pounds ----..--.--------------------- 0. 0387 0.066 Oats at 0.35 dollar per bushel of 32 pounds_--_--------_-----------.-------- 0,046 0. 082 Rye at 0. 80 dollar per bushel of 56 pounds. ...-...---.. ....-.-------------- 0.051 0.092 Nipa at 0. 00389 dollar per liter, 4 equivalent to 6.5 per cent alcohol - .___-- 0.015 0. 027 Coco at 0. 005 dollar per liter, equivalent to 6.5 per cent alcohol_-___--.---- 0.0384 0. 069 Coco at 0. 0062 dollar per liter, equivalent to 6.07 per cent alcohol_-___---- 0. 0488 0. 088 + 4 One dollar United States currency is equal to 2 pesos Philippine currency. >From Bull. U. S. Dept. Agr., Bur. Chem. 130 (1910) ; nipa and coco data from Gibbs, Philip. Journ. Sci. § A 6 (1911) 104. © The present price of molasses (February, 1922) is 3 centavos (1.5 cents) a gallon. 4 Pratt, D. S., Philip. Journ. Sci. § A 8 (1913) 395, estimates cost of nipa at 0.0015 dollar per liter, which gives less than one-half the value computed for cost of alcohol from nipa tuba given in the above table. MOLASSES Among the waste products utilized for the manufacture of alcohol, molasses ranks first. In the United States the chief source of industrial alcohol is cane or black-strap molasses. This product has been used for the manufacture of West Indian rum for more than two hundred years, but the utilization of molasses in large quantities for the production of industrial alcohol is a development of the last few years. For the manufacture of alcohol, molasses far surpasses any other material except palm saps in the ease with which it can be manipulated. Some 300,000,000 gallons of molasses are annually available in the West Indies. Tank steamers carry this material to the distil- 20 The Philippine Journal of Science 1922 leries located in New York, Boston, Baltimore, New Orleans, and other manufacturing centers. In 1920, 110,000,000 gallons of molasses were used for the production of industrial alcohol in the United States. The Philippine Islands sugar crop of 1921 totaled 552,027 tons... The molasses weight is about 25 per cent of the weight of the sugar produced. This gives an annual production of 23,000,000 gallons of molasses.2 Only 20 per cent of this was utilized for alcohol. Practically no alcohol was made at the centrals, the molasses being transported to the distilleries, mostly in or near Manila. Molasses is at present being run to waste in some sugar centrals because of a lack of market for it. Ne- gros Island alone produces annually 8,000,000 gallons of molasses. Production will undoubtedly be largely increased in the future. It takes from 2.5 to 3 gallons of molasses (depending chiefly on the percentage of sugar in the molasses and on the efficiency of the plant) to make 1 gallon of 190° proof (95 per cent) alcohol. At the latter conservative figure, the present annual supply of molasses in the Archipelago is a potential source of approximately 7,800,000 gallons of alcohol. Besides the fer- mentable sugar, cane molasses also contains valuable salts which can be utilized for fertilizer. In fact, the price of molasses is largely based on its fertilizing value. As far as the writer can ascertain, no attempt is being made in the Philippine Is- lands to recover the nitrogen, potash, and phosphorus com- pounds contained in the lees in the manufacture of alcohol from molasses. Theoretically, if the ash of the burned bagasse and the salts of the molasses are returned to the soil, the latter suffers no loss, as the sugar formed consists of carbon, hydrogen, and oxygen, which can be obtained from air and water. The potash and nitrogen content? is greater in Hawaiian molasses than in Philippine molasses, as is shown in Table 4. TABLE 4.—Fertilizer constituents of Philippine and Hawaiian molasses. Constituent. : Philippine.| Hawaiian. Per cent. | Per cent. POUR gn = swiss yasemh anh eeviginnruieneagicaplGsegegta tesla weap cack 1.39 3.99 PHORDNOE OS ono. ono new aw envio cashes cash a Sane scae ne 0. 88 0.21 Nitrogen. 900002023. os ee ee ee 0.21 0.64 *Commercial ton (2,000 pounds). One metric ton equals 1.1 commercial tons. : *One United States gallon equals 3.78 liters. One gallon of molasses weighs 12 pounds. * Brill, H. C., and Thurlow, L. W., Philip. Journ. Sci. § A 12 (1917) 269. 21,1 Cole: Manufacture of Industrial Alcohol at The Hawaiian cane takes up more of these constituents, be- cause the quantity available is greater due to the general use of fertilizers in the Hawaiian Islands. In many cases no fer- tilizer is added to the sugar lands of the Philippine Islands, and no attempt is made to return the fertilizer ingredients found in the ash of the bagasse and in the molasses. Thus, it is only a question of years before fertilization must be practiced or the land will become exhausted. Fermentation of the molasses—The methods of fermentation of molasses in the Philippines are crude, with consequent low efficiencies resulting. There is no distillery using modern yeast- culture machines and very few in which the dilution of the wort, the addition of acid and sulphate, the stopping of the fermenta- tion at the highest alcohol content, etc., are scientifically con- trolled. Brill and Thurlow‘ have shown that sterilization of the molasses solution or the use of good water increases the efficiency markedly. Also the dilution of the molasses to a defi- nite density (about 16.5 Brix) is essential. The addition of 2 grams of sulphuric acid and at least 0.4 gram of ammonium sulphate to every liter of ferment will increase the percentage of alcohol produced. The use of pure yeast culture is probably the greatest factor in obtaining the highest possible efficiency. They recommend 1 part of pure yeast-fermenting wort to 100 or 150 parts of ferment. The dilution of the molasses may be decreased by using the Molhant process of accustoming the yeast to fermenting in more highly concentrated solutions, with a consequent saving in heat necessary for distillation. Magné* cites the following figures with reference to the yield of alcohol according to the process employed: TABLE 5.—Yield of alcohol from molasses according to the process employed. Molasses Absolute (55 per cent Yield of al- alcohol per fermentable Method of fermentation. cohol; percent) 100 pounds | sugar) neces- of theoretical. | fermentable | sary for1gal- sugar. lon absolute alcohol. Gallons. Gallons. Spontaneously in presence of wild yeast --------- 40-60 2.9-4.4 8.5-5.4 Using compressed yeast -------------------------- 50-75 8.6-5.5 2.8-4.2 Using antisentics...6.. 050... <6. <55-- =e - er 70-85 §.1-6.2 2.5-3.0 Operating with pure yeast-___--------------------- 85-95 6. 2-6.9 2.2-2.5 * Loe. cit. *Magné, J., Pamphlet on the Manufacture of Alcohol from Molasses and Cane Juice. New Orleans, La., U. S. A. 2? The Philippine J ournal of Science 1922 It is quite possible to obtain a yield of 85 or 90 per cent, yet 75 per cent is considered good practice in the Philippine Islands. Any improvement in the methods of fermentation and distilla- tion of molasses would mean an increase of revenue to the Islands and larger profits to the manufacturer. Other raw materials for the production of alecohol_—Any starch or cellulose material can be converted into alcohol by proper treatment; consequently, the raw materials for the pro- duction of alcohol, in the Tropics at least, are practically in- exhaustible. Bamboo, cassava, the various palms, wood waste (sawdust), etc., can all be utilized; however, the nipa sap and waste molasses available will more than meet the demand for the next few years. The yields of alcohol from various sources, given in Table 6, are what may be expected in large-scale production. TABLE 6.—Approximate yield of alcohol from various sources. United States gallons of Material. . 95 per cent alcohol per ton of 2,000 pounds,* Sugar molasses 70 Sorghum stalks 13.5 Wheat | 100 Barley | : 75 Potatoes 21.5 Sweet potatoes 37.5 Cassava 42 Bananas yr 14 Sawdust (soft woods) 21.5 8 One United States gallon equals 3.78 liters. Inorganic sources of alcohol—The commercial production of alcohol from inorganic materials is a recent achievement. The general method involves the hydrogenation of acetylene to form ethylene, which is then dissolved in sulphuric acid; this mixture, containing ethyl hydrogen sulphate, is diluted and distilled, al- cohol being given off. A cheap source of power is necessary to make such a process a commercial success. A plant for the production of alcohol by this method is now in operation in Switzerland. The process is not a feasible one for the Philip- pine Islands at present. ALCOHOL AND ALCOHOL MIXTURES AS FUELS FOR INTERNAL- COMBUSTION ENGINES Although the application of alcohol as a fuel for internal- | combustion engines dates back more than two decades, it is only within the last four or five years that the problem has 21,1 Cole: Manufacture of Industrial Alcohol 293 assumed practical importance. This is due to the enormous increase in the use of automobiles in the last decade and the world-wide application of crude oil to transportation. The powers of the world are even now engaged in a struggle to monopolize the available supplies of crude oil, for our present- day commerce is dependent very largely upon an adequate supply of oil. In the United States alone the production of crude oil has increased about 140 per cent since 1909, and that of gasoline about 800 per cent; but the number of automobiles registered has risen 2,570 per cent. The consequent enormous demand for gasoline has been met, partly by increasing the boiling-point range, partly by extracting liquid hydrocarbons from natural gas, partly by cracking, and partly from stored reserves. Ac- cording to the United States Geological Survey ° the unmined supply of petroleum is, roughly, 6,000,000,000 barrels which, if the present rate of production is maintained, will be exhausted | in about thirteen years. But the present rates of production and of consumption will undoubtedly be increased, thus bringing even closer the exhaustion of our oil supply. Therefore, the finding of a substitute for gasoline is of utmost, vital, and immediate importance. We must find some less-evanescent fuel, one that is inexhaustible or almost so. There is known at present only one; namely, alcohol. The pioneer experimental work on the production of an effi- cient alcohol engine has been completed.” It has been found © that in order to use alcohol efficiently, the following alterations in the present type of engine are necessary: Increased compression.—Alcohol is more efficient in engines of low piston speed and long stroke. This is due to the slower flame propagation of alcohol vapor in comparison with gasoline vapor and its ability to stand higher compression (180 pounds per square inch) without preignition. Preheating of air and alcohol vapor.—Alcohol will not vapor- ize at ordinary temperatures. The preheating necessary can be readily done by utilizing the exhaust, with an electrical con- trivance for starting from the cold. * Mineral Resources of the United States, pt. 1 (1917). "See Lucke, C. E., and Woodward, S. M., Tests on internal combustion ‘engines on motor fuels, Bull. U. S. Dept. Agr., Off. Exp. Sta. 191 (1907). Strong, R. M., Commercial deductions from comparison of gasoline and alcohol tests on internal combustion engines, Bull. U. S. Geol. Surv. 393 (1909). + 24 The Philippine Journal of Science 1922 Reduced water-cooling circulation.—Alcohol does not give out as much heat in burning as gasoline, but needs high cylinder temperatures for efficient working. Increased jet orifice of carburetor—An alcohol engine needs the jet orifice enlarged so as to increase the fuel supply about 50 per cent. Carburetor floats——Hither a metal float or a cork float having a baked coating of boiled linseed oil should be used instead of a shellac-coated float, as shellac is soluble in alcohol. It is thus readily seen that alcohol cannot immediately replace gasoline. A fuel which can be used in the present type of engine is necessary to tide over the transition period during the change from gasoline to alcohol engines. Such a fuel is a mixture of alcohol with a more-volatile substance, such as benzol or ether. ALCOHOL MIXTURES Mixtures of alcohol with other substances have been in use for some time. The mixture most used in Europe is one of alcohol and benzol. The use of benzol makes the fuel dependent on a coal-tar industry, and such countries as South Africa, Aus- tralia, South America, the Philippines, etc., would have to im- port their benzol. Consequently there has been developed in South Africa and in Hawaii a mixture of alcohol and ether; since the ether is made from the alcohol, its manufacture is practically independent of any other industry. In South Africa a fuel named “Natalite,” or “Natilite,” has come into wide use due to the scarcity and the high cost of gasoline during the World War. It has been patented in all countries throughout the world. This mixture contains 55 per cent of rectified alcohol, 44.9 per cent of ether, and 0.1 per cent of ammonia. In 1918 over 255,000 gallons of Natalite were produced from molasses in South Africa. At the present time the largest factory is at Merebank, Natal, the capacity of which has been increased to 2,000,000 gallons per annum. ~ In 1918, due to the scarcity of gasoline in Hawaii, there was developed by Mr. Foster, of the Maui Agricultural Co., a motor fuel very similar to Natalite which has been used very suc- cessfully in the place of gasoline. It consists of a mixture of 55.55 per cent alcohol, 42.78 per cent ether, 1.11 per cent kero- sene, and 0.56 per cent pyridine; that is, 40 gallons of ether are added to 60 gallons of Formula No. 3 [to each 100 gallons ethyl alcohol add 5 gallons sulphuric ether, 2 gallons benzine (kerosene), and 1 gallon pyridine]. 21,1 Cole: Manufacture of Industrial Alcohol 25 With either of these mixtures, ether being very volatile, the engine can be started even more easily than with gasoline. The pyridine or ammonia is added to neutralize any acids that might be formed in the fuel or in the combustion of the fuel, and it also serves as a denaturant. These mixtures compare very favorably with gasoline in effi- ciency. That these fuels can even now compete with gasoline for airplane engines in the United States, where gasoline costs less than one-half what it does in the Philippines, is demon- strated by the United States Post Office Department * in its airplane mail service. Its tests show “‘a great increase in num- ber of miles per gallon, an increase in power and a very marked saving in the quantity of lubricating oil used.” Table 7 shows the consumption of alcohol fuel at various speeds, the tests covering thirty-one nonstop flights between New York and Washington. TABLE 7.—Consumption of alcohol fuel in airplane engines at various speeds. Revolutions per Gallons per hour, minute. alcohol mixture. 1,440-—1,460 : 15.9 1,475-1,480 20.1 1,500 21.5 1,520-1,525 ! 22.4 Tables 8 and 9 show the amounts of gasoline and alcohol fuel used by airplanes at varying speeds, and also the consumption of lubricating oil. TABLE 8.—Comparison of the consumption of gasoline and alcohol fuel in airplane engines at varying speeds. Gallons per hour. Revolutions per minute. Alcohol Gasoline. | jixture. sag SS Pes einem ipemeeg o peMrecag apapegupeeie ! etcneeet me ee rad ture a men a 15.9 Ti sai an eat ene oe heen e nen 24 20.1 Sa a a ha ek en ne he eiee enn Hes ee eee 24.17 21.5 These tests also proved that there was much less carbon for- mation when using alcohol fuel than when using gasoline, and the number of forced landings due to fouled spark plugs was reduced to the minimum. *Tunison, B. R., Chem. & Met. Eng. 22 (1920) 297. 26 The Philippine Journal of Science 1922 TABLE 9.—Comparison of the consumption of lubricating oil in airplane engines with gasoline and with alcohol fuel. Oil used with— Revolutions per minute. Alcohol Gasoline. wmixtute, ep Oh a eka ues CU ae Ko caw ONE Wo odek Sada cbaadicagasadtmacaul-sadspiencse 4.5 Fe a ik a ead idan oo eee pda ddcaetn Temes Seite hermes dap eeeeTes 4.65 4.2 I sce oe ae ge a aaah gata bia vedas cab ets pa ay pee ee These alcohol mixtures have been thoroughly tested, and they compare very favorably with gasoline.* The mileage obtained is equal or superior to that obtained with gasoline; there is no carbon deposit left in the cylinder; the acids formed are neutral- ized by the pyridine or ammonia present, hence there is no ° corrosion; there is less tendency toward knocking, resulting in a smoother-running engine; the heat of combustion is lower, hence the engine is less liable to overheat; in case of fire it can be extinguished with water; an engine can be started more easily from the cold than with gasoline; there is less fouling of the spark plugs and lower consumption of oil. Foster *° points out that the addition of substances that would neutralize the acids formed by the explosive combustion of the alcohol-ether fuel has incidentally often resulted in retarding the production of corrosive compounds within the fuel itself. He states in his patent: It is obvious that the production of a successful synthetic fuel for internal combustion engines involves not only the problem of protecting the engine from the products of combustion, but the far more involved problem of protecting the fuel itself from destructive action and the fuel storage and supply system from the deleterious effects resulting from such auto-destructive action of the fuel. ; The use of ammonia as a neutralizing agent in alcohol-ether mixtures is open to the great objection that ammonia is very volatile. To obtain some idea of the rate of evaporation of ammonia the following experiment was carried out: An al- cohol-ether-ammonia mixture of the same composition as Nata- lite was placed in open 60-cubic-centimeter beakers, 25 cubic centimeters in each beaker. The ammonia present was de- * Alcohol used alone is, of course, more efficient than are alcohol mixtures, when used in slow-speed engines with high compression. * Sugar News 2 (1921) 524. ™ United States patent 1,384,946.. 21,1 Cole: Manufacture of Industrial Alcohol 27 termined at certain time-intervals by titration with tenth normal hydrochloric acid. Litmus paper was used as an indicator, 20 cubic centimeters of distilled water being added to the sample before titration. The results are shown in Table 10. Text fig. 1 shows the rate of loss of ammonia from Natalite fuel when exposed to the atmosphere. It is readily seen that a less-volatile alkaline substance than ammonia must be used as a neutralizing agent. Of the many possible compounds the 100 | | ae 50 40 tee Loss of anmonia in per cent 60 10 20 30 eee « oe 6: 5 00 128" 190: FS S00 7 bee ee Exposure in minutes. Fic. 1. Percentage loss of ammonia from Natalite fuel. only ones that are obtainable in large quantities and that are _ at the same time inexpensive are pyridine and aniline. The writer has tested both Natalite and Foster’s fuel in a Buick-Six model automobile. The shellac-coated cork float, was replaced by one coated with glue. The spark was advanced to 30°. The air-fuel mixture was carefully adjusted. Table 11 shows comparative results with these fuels. 28 The Philippine Journal of Science 1922 TABLE 10.—Determination of the rate of evaporation of ammonia from a Natalite mixture exposed to the air. Temperature, 28° C, 0.1 N hydro- | 7 ed Cae | bom Duration of exposure to air. / ia boas saietaeas ammonia. mixture. Mins. ec. mgs, Per cent, i LoS SS SO Se Pewee Sim eae ators tS Sgety Mine une Oe Reread 4.3 EA eeseeye geen ct see a ec oe ee de oe er euuo cunt wee uuicanues 3.7 25.2 14.0 PU cUeakc GinUka boo Oag skews nes Sa oa eee kc cece ee Sacekk 3.3 22.5 23.0 UG ie ee a ae ee 2.3 15.7 46.4 Wicked cg eiesmu ee Ge Le ee oe eee ee ee Ss 1.3 8.9 69.6 PU eet poe don cee ee uatenaceecea suck pesieces tens occe 0.2 1.4 95.2 SAU ae oc ene sae Se See bs te vow oteue cus 0.0 0.0 100.0 TABLE 11.—Comparison of Shell gasoline with Natalite and with Foster's motor alcohol. Shell gasoline. | Natalite, _| Foster's motor MARGE DOF Gallons ics ec 13.6 11.9 12.0 AOE BOD ood eee ice nceweecasccotonee Mobiloil Arctic.| Mobiloil Arctic.| Mobiloil Arctic. Ease of start from the cold _________._____- Must be choked_| Starts readily __| Starts readily. Knocking under strain___.............-__-. PNG ssh None sos None. Btehe (sctsiud cele o pes Gotd . oro Geode Good. Carbon formation in the combustion Chamber 255 So ee Marked. 2s) Slight 5.22.52: Slight. Condition of spark plugs_-_-._.-__-.._.___. Sootedi.5 so. Clea ccc Clean. Trouble was experienced at first due to clogging of the car- buretor from dirt loosened by the motor alcohol. After a thorough cleansing of storage tank, vacuum tank, and carburetor, no further trouble was experienced. If this fuel is to be successfully launched in the Philippine Islands it will be only the part of wisdom to insist upon a thor- ough preliminary cleansing of the gasoline system and a proper adjustment to both the fuel-air mixture and the spark. The normal setting of the spark for gasoline engines is 20 to 15° ahead of “dead center.” The rate of explosion of alcohol fuel is less than that of gasoline and, consequently, the spark should be set at 25 to 30° ahead of dead center; or a safe rule to follow would be to double the advance recommended by the makers for gasoline. A lesson should be learned from Cuba’s experience with motor alcohol.” “Tillery, R. G., Louisiana Planter & Sugar Man. 67 (1921) 429. 21,1 Cole: Manufacture of Industrial Alcohol 29 Although the motor spirits industry cannot be said so far to be a complete failure in its competition with gasoline, certain facts with regard to the present consumption indicate that the demand is not increasing in the proportion that the producers had at first calculated upon. As a matter of fact, the consumption has apparently reached its peak, for the present at least, unless the manufacturers encourage it by producing a better product. This situation has been brought about chiefly by the manufacturers them- selves, through practices of their own rather than through any impos- sibility of a distiller’s being able to produce an alcoholic motor-spirit which will compare favorably with the grade of gasoline now being sold on the Cuban markets. First of all, they attempted to sell the motor- spirits at a higher price than the relative value of the product in power development or mileage obtained as compared to gasoline indicated the motor-spirits to be worth. Second, they were instrumental in having a formula adopted by the Government which the present state of equipment of their plants could meet, and as a consequence, they are turning out a product which is. inferior to all other motor-spirits which have been manufactured in other countries, and which have been determined by elaborate trials and experimental tests to compare favorably with the best grades of commercial gasoline. Third, they have not attempted to determine whether the engines in which comparative tests have been made in Cuba were properly adjusted for economic consumption of the motor- spirits and for efficient development of power, nor have they attempted to instruct the beginners in the proper manipulation of their engines during ordinary, every-day service, but rather have left it to the intel- ligence or ignorance of the chauffeurs to discover such matters for them- selves, The formula adopted by the Cuban Government was: To 1,000 parts of alcohol of not less than 95%, add 0.5 part of formal- dehyde, 8.00 parts of pyridine; and to every 100 parts of this mixture add 10 parts of gasoline or sulphuric ether, and 5 centigrams of methyl violet powder. It will be noted that either gasoline or sulphuric ether may be used and, as not one of the distilleries was equipped to make ether, gasoline was used. To add almost 10 per cent of a prod- uct with which one is competing is not logical. Furthermore, @ successful motor alcohol must have at least 40 per cent of a highly volatile liquid to compare favorably with gasoline. Cuba must éither revise her formula and make a motor fuel contain- ing at least 40 per cent of ether, or kill an industry which is capable of turning out 50,000,000’ gallons of valuable motor spirits annually. MANUFACTURE OF ALCOHOL-ETHER MOTOR FUEL | Alcohol distilleries now operating can easily make the alcohol- ether motor fuel by simply adding an inexpensive ether still and the requisite storage tanks to their present equipment. The ideal location for such a plant, however, is not in Manila but 30 The Philippine Journal of Science 1922 at the sugar centrals or in connection with a nipa swamp, thus eliminating the high cost of transportation of the raw materials and fuel. In the former case the waste steam from the central can be utilized to run the distillery; in the latter, the mangrove wood can be utilized as fuel. Cost of production.—An economical plant should have a daily capacity of at least 1,200 to 2,000 gallons of motor fuel. Smaller plants cost relatively more to erect and to operate. A plant with a daily capacity of 1,200 gallons could be built in Hawaii in 1919 for $30,000, exclusive of buildings, boilers and fuel storage tanks. The cost of operation exclusive of molasses and steam will not exceed 10 cents (U. S. currency) and should not exceed 7 cents.” A distillery of this size would require at least six fermenta- tion tanks, with a total capacity of at least 100,000 gallons; four yeast tanks, with a total capacity of 15,000 to 20,000 gallons; a yeast-culture machine; an alcohol-rectifying still, 1,200 gallons daily capacity; two ether stills and scrubbers, 350 gallons daily capacity; and the necessary storage and mixing tanks. The capacity of the storage tanks would naturally depend upon ship- ping facilities, market, etc. Another estimate '* on cost figures, for an alcohol motor-fuel plant erected in the Philippines, is as follows: Estimated cost of alcohol motor fuel plant producing 1,000 gallons per day. ' Borer Room: Dollars. 1 75-horsepower steam boiler 1,572.00 Oil and water feed pumps, general fittings, etc. 2,000.00 FERMENTING Room: 1 molasses scale 400.00 2 7,500-gallon mixing tanks 388.00 12 6,500-gallon fermenting tanks 2,164.00 3 2,000-gallon yeast tanks 212.00 3 250-gallon yeast tanks 58.75 2 pumps 400.00 STrtt Room: 1 modern copper continuous still, producing alcohol of 96-97 per cent in one operation 6,000.00 2 1,000-gallon charging tanks - 115.06 1 1,000-gallon singlings tank 57.50 DENATURING WAREHOUSE:* 2 2,000-gallon metal receiving tanks 440.00 Denaturant storage tanks 250.00 1 500-gallon denaturing tank 100.00 1 air compressor : 200.00 * Report of the Committee on Manufacture of Sugar and Utilization of By-products. Hawaiian Sugar Planters’ Association (1920). “Oscar Krenz Copper & Brass Works Inc., Sugar News 2 (1921) 591, 21,1 Cole: Manufacture of Industrial Alcohol 31 Estimated cost of alcohol motor fuel plant producing 1,000 gallons per day—Continued. ETHER Room: Dollars. 1 modern sulphuric ether generator 3,500.00 1 1,000-gallon supply tank 57.50 1 ether storage tank, metal, 1,000 gallons 144.00 FINISHED MoToR FUEL STORAGE Room: 2 6,000-gallon tanks 368.00 1 1,500-gallon mixing tank 58.00 1 pump 200.00 GENERAL: ; All piping throughout plant 2,000.00 Miscellaneous expense 1,815.25 BUILDINGS: Greatly varying as to type of construction, preparation of site, location, materials, etc., estimated at 12,500.00 Grand total 35,000.00 These figures are evolved from the result of practical experience in the erection and operation of numerous distilling plants and the particular experience of the erection of one motor fuel plant. The equipment figures and supply costs are based on current prices at San Francisco, California; plus freight charges delivered to Manila, P. I. While the technical and engineering side of * * * [these] figures have been very carefully worked out allowance must be made in the items of labor, insurance, overhead, depreciation, etc. [see pp. 32 and 33]. Those items being governed by local conditions will vary somewhat. The figures used are reasonable and conservative. In the item of buildings also there can be some variance; however, the figure used is great enough to take care of the general type of buildings that would be called for in this kind of plant. Under favorable conditions this sum would be somewhat less, where- as under conditions where there would be much excavation or grading to be done, the estimated figure would be exceeded. No allowance has been made for ground site, it being assumed that such a plant would be erected on plan- tation premises so that this would not be an item of expense. The item of Molasses also is not charged for two reasons, first because molasses at the raw sugar mill is generally considered a practical waste, and second, because of the fertilizing value of the spent slop. This alone is generally conceded to be great enough to more than offset any reasonable costs for molasses. Inasmuch as an alcohol motor fuel plant would best be placed immediately contiguous to a raw sugar mill, there should not even be transportation costs in getting molasses into the motor fuel plant. The ether can be made for less than 10 per cent over the cost of alcohol.* There are, however, two more-recent methods for manufacturing ether from alcohol which utilize the alcohol in the vapor phase. A purer product and higher yield are obtained at a lower cost than by the lead-still method. * Humboldt, E., Power 50 (1919) 420. 32 The Philippine Journal of Science 1922 The first method ** consists of superheating the alcohol vapor to the right temperature, then passing it through a chamber containing lead balls over which sulphuric acid is allowed to drop slowly. Transformation to ether occurs instantaneously, the alcohol being in contact with the sulphuric acid only moment- arily. The ether vapor is then passed through a washer con- taining caustic soda and then to a rectifying column to separate it from water and alcohol. About 3 pounds of acid and 14 pounds of soda are required for every 100 pounds (13.9 gallons) of ether made. This process is being used successfully in the Natalite factory at Merebank, Natal. The second method is a catalytic process and eliminates the use of sulphuric acid and caustic soda. The alcohol is passed through a superheater and then over alumina. acting as a catalyst. The temperature of the vapor and catalyst must be kept constant at about 200° C. If the temperature rises above 200°, ethylene is formed. By passing the outgoing vapors through tubular heat exchangers to heat the incoming alcohol, a small superheater is the only necessary heating apparatus; the amount of heat necessary for the conversion is less than 10 per cent of that required in the lead-still process. The estimated daily operating cost is given by Krenz as follows: Estimated daily operating cost of alcohol motor fuel plant producing 1,000 gallons per day. Dollars. 19.34 tons average cane molasses 000.00 No. 34 66 Be. sulphuric acid, 2.5¢ per Ib. 85 No. 14 caustic soda, 5.4¢ per pound -76 21 gals. benzene, 50.0c per gal. 10.50 103 gals. pyridine, $1.86 per gal. 19.53 40,000 gals. water, diluting, condensing, etc., pumping cost of 10c per 1,000 gals. 4.00 63 bbls. fuel oil, $7.28 per bbl. 47.32 LABOR: 1 fermenting room man and foreman, day shift 5.00 2 distillers, night and day shifts, 3.00 6.00 1 denaturer, day shift 3.00 1 ether man, day shift 3.00 2 firemen, day and night shifts, 3.00 6.00 1 laborer _ 2.50 25.50 * Annaratone Process, French patent 408,089. “Oscar Krenz Copper & Brass Works Inc., loc. cit. 21,1 Cole: Manufacture of Industrial Alcohol 33 Estimated daily operating cost of alcohol motor fuel plant producing 1,000 gallons per day—Continued. Dollars. Interest at 8% on $35,000 over 312 working days, per day 8.97 Insurance on 4 plant value, premium 38.00 per 100 per year, per day re Depreciation at 10% per year, $3,500; per day 11.21 Office expense, general overhead, etc., per day 10.00 Grand total 138.75 The above estimate shows that the cost of producing one gallon of alcohol from molasses, exclusively of the cost of molasses, is 13.87 cents (27.74 centavos). Mr. Foster computes the cost of operation to be 14 to 20 centavos, exclusive of molasses and steam. The fertilizer recovered from the lees practically covers the cost of the molasses. In the Philippines alcohol-ether motor fuel can undoubtedly be made to yield a handsome profit when sold at 50 to 70 centavos a gallon. In Natal the original Natalite company showed a profit of 38 per cent in 1920. The original 5-pound shares of the parent company are now quoted at 105 pounds Sterling. When Natalite was placed on the market in Natal at 52 cents, gasoline was selling for 1.50 dollars. To-day there are several centers where gasoline is entirely off the market.** , In the Philippines, during the World War gasoline reached as high as 3 pesos a gallon. The present price is 1.15 pesos in Manila. At a price considerably lower than this, alcohol motor fuel can more than successfully compete with gasoline. The present annual supply of molasses in the Philippines is a potential source of some 7,800,000 gallons of alcohol. The con- sumption of gasoline for 1920 was 9,160,000 gallons. Therefore, the Philippine Islands could almost supply its own needs for motor fuel by simply utilizing its present supply of molasses for such production. Since the nipa sap available is many times as great as the molasses supply, and is an even cheaper source of alcohol, the country is in a position not only to supply its own needs, but to export as well. SUMMARY The nipa palm and molasses offer cheap and easily manipu- lated sources of alcohol for use as a motor fuel. 8 Wilson, Sugar News 2 (1921) 228. 1876683——3 34 The Philippine Journal of Science 1922 Alcohol, alone, will probably not be used as a motor fuel until a new type of high-compression, slow-speed engine is developed. Alcohol mixtures to-day are being used on a large scale in the present type of gasoline engine with simple alterations. These mixtures offer many advantages and no disadvantages over gasoline and are more efficient than gasoline. Although no alcohol motor fuel is being manufactured at present in the Philippine Islands, two distilleries are installing the necessary apparatus for its manufacture. Utilization of waste molasses and of the unlimited supply of nipa sap in the production of alcohol motor fuel would not only make the Phil- ippine Islands independent of other countries for its supply of motor fuel, but would also constitute a very valuable addition to the industries of the Islands. Since modern industry largely depends on an abundant and cheap supply of motor fuel and since in time of war the price of imported fuel goes skyward and the supply is curtailed, the establishment of an industry supplying motor fuel from the natural resources of the Archi- pelago will be of immense value to the Philippine Islands. THE BURI PALM The description and uses of the buri palm, Corypha elata Roxb., can be found fully discussed in another article.2? The sap is used only to a small extent in the Islands at present and gives small promise of any large future development. In many places in the Philippines the inhabitants have every need sup- plied by this palm: bread from the pulp; sweetmeats from the young fruit kernels; houses from the leaves; hats, mats, ropes, and baskets from the leaf fibers; shoes from the bark; and sugar and wine from the sap. In India, Australia, Malaysia, and other places various other species of palms yield saps which are utilized for thaking toddy and arrack, corresponding respectively to our tuba and vino. Among the palms thus utilized may be mentioned Arenga sac- charifera Labill., the sugar palm; Borassus flabellifer Linn., the Palmyra palm; Caryota urens Linn., the fish-tail palm; Phoenix dactylifera Linn., the date palm; and Phoenix sylvestris Roxb., the wild date palm. BIBLIOGRAPHY ON ALCOHOL MOTOR FUEL ANON. Cassava as a source of industrial starch and alcohol. Bull. Imp. Inst. 17 (1919) 571. ANON. Fuel for motor transport. Engineering 110 (1920) 120. “Gibbs, H. D., Philip. Journ. Sci. § A 6 (1911) 167. 21,1 Cole: Manufacture of Industrial Alcohol 35 ANON. Motor fuel from waste molasses. Sugar 22 (1920) 335. ANON. Data concerning the advantages, production, and cost of natalite motor fuel. Int. Sugar Journ. 23 (1921) 147, 2138, 266, 386. ANON. Fuel for motor transport. Second memo. by the Fuel Research Board. London (1921). ANON. Production of 95-97% alcohol for motor fuel. Int. Sugar Journ. 23 (1921) 513. APPELL, G. M. Alcohol production from molasses. Chem. Age (N. Y.) 29 (1921) 53. BARNETT, E. B. Possible sources of home-produced motor ae Chem. Age (London) 3 (1920) 254. BAUME, NICOLARDOT, et al. Physical-chemistry researches on liquid fuels. Chimie et Industrie 3 (1920) 1761. BLACK, J. Liquid fuel mixture. U. S. patent 1,360,872. Boyp, T. A. Motor fuel from vegetation. Journ. Ind. & Eng. Chem. 13 (1921) 836. BRILL, H. C., and THuRLow, L. W. Alcohol from discard molasses in the Philippine Islands. Philip. Journ. Sci. § A 12 (1917) 267. CARSTEN, H. J. The utilization of waste molasses in the Philippine Islands. Philip. Journ. Sci. 17 (1920) 395. CHEMICAL FUEL Co. oF AMERICA. Treating alcohol for — motor fuel. British patent 159,880 (1921). CoLE, H. I. Motor fuel from Philippine waste products. Prana Padbe 6 (1922) 69. Cox, ALVIN J. Industrial alcohol and its possibility as a source of power in the Philippines. Philip. Journ. Sci. § A 4 (1909) 232. Drxon, H. B. Aleohol as a motor fuel. Journ. Soc. Chem. Ind. 39 (1920) 355R. DuMANO!IS. The advisability of using liquid fuels. Bull. Soc. Encour. Ind. Nat. 132 (1920) 436. FLINT, J., and Brown, F.G. Liquid fuel. Canadian patent 209,167 (1921). Foster, J. P. Industrial alcohol from molasses. Int. Sugar Journ. 21 (1919) 75. IpEM. Motor alcohol, its theory and use. Sugar News 2 (1921) 521. IpEM. Motor fuel. United States patent 1,384,946. FREELAND, E. C., and Harvey, W. S. Alcohol as a combustion fuel. Sugar 23 (1921) 586. GJERDE, G. K. Motor fuel. Papier Journalen (1918) No. 6, 42. GRIMWADE, W. R. Crops for the production of power alcohol. Bull. Com- monwealth of Australia, Advisory Council of Science & Ind. 7 (1918). Hayes, A. Blended motor-fuel. United States patent 1,378,858. IpEM. Liquid fuels for motors. United States patents 1,361,153 and 1,363,870. HAWAIIAN SuGAR PLANTERS’ ASSOCIATION. Report of the committee on manufacture of sugar and utilization of by-products (1920). Hiesert, H. The role of the chemist in relation to the future supply of liquid fuel. Journ. Ind. & Eng. Chem. 13 (1921) 841. Humpoitpt, E. The possibilities of new fuel mixtures for internal com- bustion engines. Power 50 (1919) 418. - Licgutroor. Power alcohol. Bull. Australian Inst. Sci. Ind. 20 (1921). Luck, C. E., and Woopwarp, S. M. Farm. Bull. U. S. Dept. Agr. 277 (1907). 36 The Philippine Journal of Science 1922 IDEM. Test-on internal combustion engines on motor fuels. Bull. P. S. Dept. Agr. Office of Exp. Sta. 191 (1907). Also Engineering News 59 (1908) 271. LYLE, T. R., et al. Power alcohol, proposals for its utilization in Australia. Bull. Advisory Council of Science & Ind. 6 (1918): Melbourne. MAGNE. Pamphlet on The manufacture of alcohol from molasses and cane juice. New Orleans, La., U. S. A. MAILHE, A., and GopoN, F. Catalytic preparation of ordinary ether in a dry manner. Bull. Soc. Chim. IV 25 (1919) 565; 27 (1920) 121. Marks, E. C. R. Motor spirit. British patents 147,398 (1919); 153,925 (1919) ; 154,867 (1919). NATALITE, British patents 24,262 (1914); 17,017 (1915) ; 375,453 (1916); 878,700 (1917). Norturop, J. H. Acetone and alcohol by fermentation with B. aceto-ethy- licum. United States patent 1,293,172. ORMANDY, W. R. Criticisms of patents on motor fuels containing alcohol. Int. Sugar Journ. 22 (1920) 402. Ip—EM. The motor fuel problem. Journ. Inst. Petroleum Tech. 5 (1918) 33. PurGoTtI, A. Employment of alcohol in explosion motors. ‘“Ethnol.” Giorn. chem. ind. applicata 2 (1920) 191, REDWOOD, BOVERTON, et al. The production of alcohol for power. Chem. Age (London) 1 (1919) 66. ScArD, F. I. Power alcohol in the United Kingdom. Int. Sugar. Journ. 23 (1921) 484, ScarRATT, A. W. The carburetion of alcohol. Journ. Soc. Autom, Eng. 8 (1921) 328. Sesti, C. G. Motor spirit. British patent 147,803 (1920). SHAVE, G. J. Benzene-aleohol experiments on omnibuses. Engineering 110 (1920) 623. STEVENS, E. W. Distilling alcohol for use in motor fuels. United States patent 1,372,465. STEVENS, E. W. Motor fuel. United States patent 1,388,531. Stronc, R. M. Commercial deductions from comparison of gasoline and alcohol tests on internal combustion engines. Bull. U. S. Geol. Survey 392 (1909). | StronG, R. M., and Stons, L. Comparative fuel values of gasoline and denatured alcohol in internal combustion engines. Bull. Bur. Mines 43 (1920). TILLERY, R. G. Use of alcohol made from final molasses as motor power in Cuba. Louisiana Planter 56 (1921) 142. IDEM. The alcoholic motor-spirits situation in Cuba. Louisiana Planter 67 (1921) 429. TUNISON, B. R. Molasses best source of industrial fuel. Sugar 22 (1920) 397. ; IpEM. The future of industrial alcohol. Chem. & Met. Eng. 21 (1919) 557. IpEM. The future of industrial alcohols. Journ. Ind. & Eng. Chem. 12 (1920) 370. WALLIN, H. Swedish motor fuel. Svensk Pappers Tid. 24 (1921) 69. SUPPLEMENT Supplementary to this paper is given a résumé of articles published by the Bureau of Science on the utilization of palm 21,1 Cole: Manufacture of Industrial Alcohol 37 saps as a source of alcohol; most of these publication are out of print. THE NIPA PALM, NIPA FRUCTICANS WURMB The nipa™ is an erect, stemless palm, the leaves and inflorescences arising from a branched rootstock, the leaves pinnate, 3 to 10 meters long. Inflorescence from near the base of the leaves, erect, brown, 1 to 1.5 or 2 meters high, bearing numerous sheathing spathes and both male and female flowers, the former lateral, catkin-like, the latter terminal in a globose head. Fruit nodding, globose, as large as a man’s head or often considerably larger, consisting of many obovoid, 6-angled, 1-celled, 1-seeded carpels, the free parts pyramidal, the pericarp fibrous, the seed large, white, hard. This species of palm grows in great abundance throughout the Archipelago in the tidewater estuaries and tidal rivers or other places flooded by the tides with brackish water. It does not thrive in either fresh- or salt-water marshes. Plate 1 shows the distribution of nipa swamps in the Islands. * %* * Rated in order of size, irrespective of their present commercial importance, the most extensive areas are situated as follows: Portions of the provinces of Pampanga and Bulacan bordering on Manila Bay, Island of Luzon. Province of Capiz, Island of Panay. The valley of the Cagayan River, Island of Panay. The valley of the Cagayan River, Island of Luzon. The valleys of the Catubig and the Gandara Rivers in the Island of Samar. The Province of Pangasinan, Island of Luzon. Surigao, Island of Mindanao. The Province of Tayabas, Island of Luzon. The other nipa areas of the Philippines do not compare in size with those listed above. In this connection it might be well to point out that the tide- water flats on which the nipa flourishes also produce various species of mangrove. The mangrove furnishes a valuable source of fuel which in virgin stands could be utilized for the distilla- tion of the nipa sap. The virgin mangle has long since disap- peared from the region around the upper part of Manila Bay. In this district cultivated mangrove swamps are found from Malabon on the east to Balanga on the west side. Nipa* and bakauan [a species of mangrove most used for firewood] are planted extensively in solid stands, but there is little planting of other swamp species. Hundreds, and may be thousands, of hectares are planted to bakauan, which is grown for firewood, and sold principally in the Manila market, this wood being preferred to almost all others for fuel. The Manila supply is inadequate, as is indicated by the remarkably complete utilization of the swamps in the vicinity, and by the use of other and inferior species. * Gibbs, H. D., Philip. Journ. Sci. § A 6 (1911) 99-206. Since reprints of this article are exhausted, part of it will be quoted in full, other parts abstracted and brought up to date. * Brown, W. H., and Fischer, A. F., Philippine Mangrove Swamps, Bull. P. I. Bur. Forestry 17 (1918) 94. 38 The Philippine Journal of Science 1922 No records seem to exist as to how, when, and where bakauan was first cultivated, but it is safe to say that the present planting methods have been in use for at least three generations. An evident reason for the cultivation is the large agricultural population on the level lands of Pampanga Province and the accessibility of the Manila market. In the municipalities of Macabebe, Guagua, Lubao, Sexmoan, and Orani there are a number of barrios along the rivers which have a population that divides its labor among the fishing, bakauan, and nipa industries. Table 3 shows us that the nipa juice is the cheapest source of alcohol available. It is also cheaper to produce alcohol from nipa than from grains or even from molasses, as it is not necessary to dilute the sap with water or even to seed it with yeast. Not only do the palm saps ferment readily, due to the wild yeast always present, but they ferment very rapidly so that the fer- mentation is often complete in less than twenty hours. The elim- ination of pure water necessary for dilution, pure yeast for seeding, addition of chemicals, and the shortening of the fer- mentation period all mean considerable saving in the production of the alcohol. There are at present only a few nipa districts utilized for the collection of nipa tuba as a source of alcohol. Most of these are in the nipa area on Luzon in Bulacan and Pampanga Prov- inces. This district extends along Manila Bay for about 32 kilometers with an average width of 5 kilometers. A consider- able portion of it is controlled by various large distilleries. Only about 30 per cent of the total area of the district is at present utilized. Some of the nipa areas near Manila have been con- verted into fish ponds which seem to yield a greater income per hectare than does nipa at present. LABOR CONDITIONS IN THE NIPA SWAMPS The best managed nipales, or nipa groves, are divided into sections of about 1 hectare, containing from 700 to 800 producing plants in every hectare, each of which is assigned to the care of one or more men. Usually two men are required to handle 1 hectare and the work of cutting the plants, gathering and transporting the tuba to the distillery and the general — care of the grove is divided between them. Since the sap not only has a recognized commercial value at the distillery, but is also greatly prized by the natives as a beverage, a close watch by the guards is necesssary. The sap, as it drops from the flower stalks, is collected in hollow joints of bamboo. Each plant is visited daily and the sap brought to the distillery in bancas of light draught which the natives paddle in and out of the waterways. Sometimes the tuba is emptied into large earthenware jars and these are transported to the distille , and, again, it is emptied directly into the boat. A nipa leaf is cut and placed in the banca over the tuba to prevent undue slopping, and thus, more or less immersed in the partially fermented tuba, the boatmen make the trip. The time for gathering the - sap is in some measure dependent upon the tides, for many of the small 21,1 Cole: Manufacture of Industrial Alcohol 39 waterways are navigable only with difficulty or else are impassable at low tide, even for such a light draught vessel as the native canoe. The collec- tion in some localities begins at 1 o’clock in the morning and is completed within 6 hours. The laborer in the nipa groves builds himself a house of nipa leaf thatch with bamboo framework, on the bank of a stream on the nipa estate. Fish, the principal article of food, are easily caught in these places. Here, usually with a large family, he lives in peace and contentment. The communities in these nipa areas are old-established ones, and undoubtedly attempts to establish new communities to-day for work in the swamps would be a difficult matter. More inducements would have to be offered than were offered in the past; medical attention, stores, and amusements would have to | be included, barrios constructed, and a water supply provided in order to obtain the laborers and keep them contented. PROPAGATION OF THE PALM The nipa palm reproduces from seed and also by the branch- ing of its roots. The original seed produces two plants, each of which in turn produces two more at the growing points formed by the branching of their roots, and so on. Hence the life of a nipa swamp is indefinite. It is said that care must be taken not to bleed both of two root-connected plants at the same time as in that case the plants will die. When the palms are raised from seed, the planting is done just before the tuberos begin the first harvest. The seeds are put in about 2 meters apart, making 100 seeds per loang (about 0.04 hectare). The plants begin to bear in about five years. By that time each plant has four growing points, about 1 meter distant from each other. HARVESTING OF THE NIPA SAP Conrado and Zobel 2 enumerate the following operations that are connected with the collection of the sap: The hauan, or “clearing up.” The sicat, or “kicking the stem.” The talog, or “cleaning the stem.” The pucao, or “working of stem to make it soft and limber.” The patit, or “cutting of the stem.” The cortes, or “daily sectioning of the stem.” The “clearing up” is usually done in July or August. The nipales are parceled to the tuberos who are required to clear * Conrado, A., and Zobel, E., Estudio de la planta llamada “nipa” de su cultivo y de sus propiedades. Imp. y lit. de “Ta Concepcion,” Marques de Comillas No. 3, Manila (1906). . 40 The Philippine Journal of Science 1922 them of underbrush, dead leaves, etc. Since the nipa sends its inflorescence up from the base and hence is near the ground, the flower stalk is conveniently situated for tapping and for gathering of the sap. The nipa bears small fruits in its fourth year but is not usually tapped until its fifth year. The stalks of the chosen fruits (nearly full-grown ones) are stripped of leaves and are kicked at the base about five times at intervals of about a week, several kicks being given each time. This is to start the flow of tuba. The stalk is then cut across near its top, usually just below the fruit, and each day a thin slice is removed to keep the wound fresh and to facilitate exudation. When a plant bears two flower stalks the usual practice is to draw sap from only one, the other being removed and the stem allowed to dry. The sap is collected in bamboo joints, called tuquils (Pam- pangan), or bombones (Spanish), which are hung upon the stem. The receptacles are about 45 centimeters high and 8 centimeters in diameter and have a capacity of about 2 liters. One stalk normally flows for about three months, but it is not uncommon for it to be entirely cut away, at least so close to the ground that it can no longer be utilized by the daily paring of small slices, long before the flow has ceased. In some districts the flower is cut before the fruit is formed, and under such circumstances the daily yield of sap is said to be increased but the period of flow reduced from three to one and one-half months, the total yield being practically the same in both cases. The plant seemingly is not affected by this treatment. A recent report by Dr. G. A. Perkins 2 gives the following data on harvesting of the sap: The tuberos are alloted bamboo from which to make tuquils and also are alloted bancas, in which they transport the tuba. These bancas are mostly hired by the company at 3 to 8 pesos per month (now about 6 pesos). About 120 tuberos are required for the first harvest (2,000 hectares). Each man handles about 1,000 producing palms. The harvest begins in July when the fruits are cut, but most fruits are reserved until the end of the floods (about August 20). The harvest ends about December 20, not all fruits being cut at the same time. Each fruit stalk flows about two and one-half months. The second harvest is from December 20 to April 10 requiring 80 tuberos who work about 800 plants each. The tuquils surviving the first harvest are used. No fruits are usually cut from April 10 to the last of June, but for the sake of having yeast for fermenting molasses a third, or “extra,” harvest is often taken by reserving fruits which could be cut in the second harvest. In 1920, 44 tuberos were employed on one * An unpublished report. 21,1 Cole: Manufacture of Industrial Alcohol 4] estate in the third harvest, each tubero handling about 500 stalks. The tuberos receive 7 centavos per tinaja (36 liters). This includes all work in preparation for harvest, but bancas and bamboos are supplied to the tubero. The only known disease of the nipa plant is absic, which is caused by insect attack on the leaves. Rain kills practically all the insects, but sometimes it is necessary to cut the infected leaves. The number of plants per hectare-—The number of plants per hectare varies within wide limits depending upon whether the swamps are cared for or overgrown, the former condition naturally giving superior plants. A conservative estimate for cultivated nipales, or nipa swamps, can be placed at 2,000 to 2,500 palms per hectare, of which 750 can be depended upon to produce fruiting stalks. Intelligent care and selection of seed undoubtedly would raise the yield of sap and the per- centage of sugar in the sap. Yield of sap.*—The estimates of the yield of sap vary within very wide limits. A distiller who has had much experience in the nipales believes that each producing plant will average 1.25 liters daily, and a chemist at one time employed by one of the distillers has stated that an average plant will flow 50 liters during the season. The yield per hectare has been siteasted: by many different writers and distillers and it is evident from a perusal of the figures that many are mere guesses. The number of plants per hectare is estimated by a chemist in the employ of one of the distillers to be 2,500. Internal Revenue Agent G. A. Ruge believes the municipalities of Abulug and Pamplona in the Province of Cagayan to have nipales containing 4,600 plants per hectare. A distiller of wide experience told me that the number is 700. If this latter figure is taken to be 700 plants producing at the same time I have no doubt that it is fairly accurate for the areas controlled by his company. He also stated that they obtained in actual practice 438 liters per hectare daily during the season. This is equivalent to 78,480 liters yearly per — hectare, a figure quite close to my estimates. Other figures, not so reliable, go as high as 225,000 liters per hectare. It is to be remembered that while wild nipa swamps may contain 4,000 plants per hectare, the percentage of producing plants and the amount of sap which each plant yields will be less than in the cultivated districts. In the uncultivated areas a small proportion of the trees is accessible; in the cultivated, the plants are thinned, and there are more waterways, and yet the yield of sap per hectare is greatly increased. A yield of 75,000 liters of tuba per hectare per year would be extremely satisfactory to the distillers, and I believe this amount is seldom reached in the best managed nipales. It should, however, be exceeded. Some operators obtain only 4,000 liters per hectare during a full season. In the year 1909 one of the largest producers obtained 8,700,000 liters of tuba from an estimated area of 1,000 hectares which is at the rate of * Gibbs, H. D., op. cit. 117. 42 The Philippine Journal of Science 1922 8,700 liters per hectare. Less than one-third of this area was producing, partly from lack of labor and partly because the greater portion of the estate had not been made accessible to the distillery. The alcohol pro- duction from this yield was about 5 per cent. The atmospheric and climatic conditions affect the flow of sap. The natives working among the plants are positive that they can predict the approach of a storm by a sudden checking, in some cases amounting to almost a stoppage, in the flow, occurring about three days before the arrival of the typhoon. It is stated that after a temporary stoppage has been produced by meteorologic conditions an increased flow will start which will balance the period of inactivity. The seasonal variation in the composition of the fresh tuba must be slight, although when received at the distilleries the differences, both in quality and quantity, are very great. This is undoubtedly due to the fact that rains dilute the juice after it has flowed from the flower or fruit stalk. The gain in volume is balanced by the loss in sugar and alcoholic content. As a result of experimental work performed by Gibbs and others, an estimate of the yield of sap per plant per season is given as 43 liters, in a nipale cared for under average condi- tions according to present methods. This means an average daily yield of 0.58 liter and a total yield of 87,000 liters per hectare per year. The average flow of sap during the season is well shown in fig. 2. The following data were obtained by Gibbs *‘ on six trees: Liters. Total flow from six trees 260.286 Greatest total flow from a single tree 49.410 Lowest total flow from a single tree 36.335 Average seasonal flow per tree 43.381 Average daily flow from six trees 3.470 Greatest daily flow from one tree ‘ 1.770 Average daily flow per tree : 0.579 The composition of the sap—From analyses made by Mr. Francisco Agcaoili, of the Bureau of Science, the following is the composition of a good quality sap: Brix 17.0 i ao Density Tp 1.020 Total solids 18.00 Ash 0.48 Acidity Trace. Sucrose 17.00 Reducing sugar Trace. An invertase proferment, or zymogen, also occurs in the sap. Changes in the composition of the sap—The inversion of the “Op. cit. 121. 21,1 Cole: Manufacture of Industrial Alcohol 43 sucrose and the alcoholic, acetic, and other fermentations begins almost immediately after the sap drops from the stem into the tuquil under ordinary local conditions of collection. When the sap arrives at the distillery the inversion is complete and the alcoholic fermentation has begun and is sometimes com- pleted. The juice is now milky in appearance and covered by a thick layer of foam. When the sap is collected in clean vessels the inversion does not begin for four or five hours. A white flocculent precipitate is slowly formed. The inversion of the | | | > w 9° °e — S “WY ee NA i * ma Lek A 400 = ¥ = ae al =| Bau o 8 we s$ 20 25 30 35 w» 45 50 55 40 65 7e | TIME IN DAYS. Fie. 2. Sap flow of nipa palm, average of six trees during one season. sugar begins at the same time. Gibbs has shown that a zymogen present in solution in the sap causes the separation of a white flocculent invertase which rapidly attacks the sucrose present. He found that if clean tuquils are used and limed with thick lime cream, the sap collected in them will remain unchanged for more than ten days. The rate of inversion of the sugar and the rate of formation of alcohol and acid are well brought out in fig. 3. 44 The Philippine Journal of Science 1922 . PERCENT Lenn & is) y o/ \> ag G Lo LIX aN ! [ \ rs-acetic . : —s 5 to 5 20 eae OO ae A ee OP Se a TIME IN HOURS Fic. 3. Spontaneous fermentation of the nipa palm sap. DISTILLERIES FOR NIPA TUBA Twelve provinces from time to time have been engaged in the industry of distilling nipa sap. The tendency has been toward a decrease in the number of distilleries operating, the primitive and inefficient distilleries being crowded out by more- modern and more-efficient plants, the production of alcohol in- creasing to some extent. In the nipa district bordering Manila Bay all the distilleries use continuous rectifying stills, producing either 100° to 120° proof alcohol or high-grade alcohol of 180° to 190° proof. Fermentation.—The nipa tuba as received by the distilleries is already partly or wholly fermented. It contains, besides the desirable wild yeast, many undesirable organisms. The tuba is placed in wooden or concrete fermentation tanks and allowed to ferment. The fermentation is usually allowed to proceed un- til the foam subsides or the temperature of the ferment falls. 21,1 Cole: Manufacture of Industrial Alcohol 45 The time allowed for fermentation varies from thirty hours to three days. It has been shown that even thirty hours is too long, much of the alcohol being changed into acetic acid by that time. Most of the large distilleries now ferment a mixture of molasses and nipa, the latter usually containing enough inver- tase and wild yeast to change the sugar in the molasses rapidly to alcohol. Molasses thus becomes a cheap source of alcohol and the distillery is often able to operate during the entire year, even during the season of small sap production. At cer- tain seasons of the year, however, the nipa sap seems to have a detrimental effect on the production of alcohol from molasses, only 1 or 2 per cent of alcohol being obtained. This is due to the killing of the yeast by the acid bacteria. In some distilleries no attempt is made to check fermentation at the highest alcohol content: This means that, while the wash is stored before going to the still, a loss of alcohol occurs due to acid fermenta- tion. In some cases, the high acidity of the beer is neutralized by lime made on the spot by burning oyster and other shells. Fermentation can be checked at the proper moment by the addition of an excess of lime to the wash, and then the latter can be stored without appreciable change to await its turn to be run to the stills. Distillation.—Distillation is usually accomplished in either pot stills or continuous rectifying stills, the former usually heated directly by wood or wood and coal fires, the latter by steam coils. The first distillate from the pot stills is collected until it has the requisite percentage of alcohol (20 to 50) desired for beverage purposes. The distillation is further continued until most of the alcohol has passed over. The pot is then emptied and the second fraction poured back to form part of the next run. ; The continuous stills give alcohol up to 100° proof. The rectifying stills yield an alcohol as high as 190° proof. The losses that occur are largely in the waste which contains some- times as high as 1 per cent of alcohol. - THE COCONUT PALM é The cultivation of the coconut palm and methods of obtaining the sap from it are fully described by Gibbs.** Large numbers or palms are utilized exclusively for their sap, which is made into the native drink vino de coco. This is a misnomer, for “8 Op. cit. 147. AG The Philippine Journal of Science vino de coco is not a wine, but a distilled liquor containing from 20 to 50 per cent of alcohol. There is more labor involved in obtaining the sap from the coconut palm than from the nipa because of the height of the - trees. To facilitate the collection bamboo poles are usually at- tached from tree top to tree top, forming bridges upon which the collector can travel from tree to tree without descending to the ground until his receptacle is full. Since the coconut blos- soms throughout the year the tuba season is continuous, and when one stalk is exhausted another is tapped. Usually not more than two stalks on a tree are tapped at the same time. The yield of sap.—The flow of sap varies with the age of the flower stalk, age and condition, of the tree, character of the soil, and climatic conditions. The rate of flow is less during the heat of the day than at night. A tree will grow about ten stalks annually, and each will run sap for about two months. A tree is in its prime when about 40 years old. An investigation of seven distilleries in Tayabas Province revealed the fact that the average daily production from 5,785 trees from April, 1909, to March, 1910, inclusive, was 0.65 liter per tree.*¢ Gibbs found, however, that trees when properly handled by a sufficient number of men would yield approximately 1.4 liters per tree daily. Composition of the sap ~ Breah sap in which no chemical change has occurred, taken from an average tree in its prime, will probably have about the following composition: Grams in 100 ce. Density 1.0700 Total solids 17.5 Acidity Trace. Ash ’ 0.40 Sucrose 16.5 Invert sugar Trace. Undetermined nitrogenous horegonede * ete. 0.60 * Op. cit. 154. ILLUSTRATIONS PLATE 1 Map of the Philippine Islands, showing distribution of nipa swamps. TEXT FIGURES Fic. 1. Percentage loss of ammonia from Natalite fuel. 2. Sap flow of nipa palm, average of six trees during one season. 3. Spontaneous fermentation of the nipa palm sap. 47 CoLEe: INDUSTRIAL ALCOHOL] [Puturre. Journ. Sctr., 21, No. 1, 7" ust ns tro° iai* ire iy. ae Cita wee ur T T T T BE i: 1 \ fem L ; | PHILIPPINE ISLANDS SCALE STATUE MILES ~ KILOMETERS, a * %» 100 = —n" ae : ? LEGEND NOS. PROVINCES TANES f /L0C0S NORTE BATANES 1S. MOUNTAIN af DENSE IPA swanp----h a APAYAO dy ithe 8 KALINGA f BonI0e LITTLE mapa swamp ----FEogd o LEPANTO SUBPROVINCES YAN BENGUET 2 CAGAAM we 2-3 HOCOSS i 4 ABRA SUBPROVINCE 4 1SABELA 5 LA UNION 6 munya VIZCAYA ? TAYABAS "uMnnnuoue SUBPROVINCE INA a BABUYANES IS. le CATANDUANES SUBPROVINCE 2/ SORSOGON wis MASBATE SUBPROVINCE MINOORO MAR PALAWAN 22 ANTIQUE ROMBLON SUBPROVINCE OO 5 OCCIDENTAL NEGROS 3 bey NEGROS S/QUIIOR SUBPROVINCE 80HOL td SURIGAO SA MIS eomerwen’ oF Mmouuac ano Sita & ZA OUSTRICTS ‘ Or ike feisty 2 CUTUAN | SUBPROVINCES gr recenes SEA \\ Se PLATE 1. THE PHILIPPINE ISLANDS, SHOWING THE DISTRIBUTION OF NIPA SWAMPS. pe THE USE OF SULPHUR FUMES IN COPRA DRYING? By A. H. WELLS Chief, Division of Organic Chemistry, Bureau of Science, Manila and G. A, PERKINS Chemist, Bureau of Science, Manila TWO PLATES A few years ago almost any grade of copra or coconut oil was salable at a good price. To-day a “Manila” oil of high rancidity and color finds a poor reception in the markets of the world. The Philippine oil mills are faced on one hand with a demand for high-grade oil, and on the other with a low-grade supply of copra from which good oil cannot possibly be made except at a prohibitive cost. This situation is thought by many to require the “cure-all” of legislation, but the writers favor the use of other methods to improve the quality of Philippine copra. The application of sound business principles to the oil and copra industries would solve the difficulty. The main cause of color and free fatty acid in coconut oil is the action of mold on the copra. Any method, therefore, of drying coconut meat that protects it from mold and contami- nation with dirt and smoke makes possible the production of good oil. The most economical method of drying for very large plantations, and at the same time one which, produces excellent copra, is by steam, using modern machinery. The economy of this method is that it saves labor, which is a very important item on large holdings. On smaller plantations, such as are found generally throughout the Philippines, the initial cost and overhead expenses of steam dryers are large items compared with the labor that they save. The problem of how to dry copra on large plantations has been rather satisfactorily handled by the firms that manufacture drying machines, some of which have agencies in Manila. Therefore, the problem of the small plantation has occupied the * Received for publication January 28, 1922. 187663——4 49 50° - The Philippine Journal of Science 1922 attention of the Bureau of Science, and the use of sulphur has been found well suited to this field. The process of sulphuring copra has been described and its advantages have been pointed out by Brill, Parker, and Yates.? The purpose of the present paper is to record certain modifica- tions that have been found advisable during several years of suc- cessful commercial use of the process in the Philippines. THE SULPHURING BOX One form of sulphuring apparatus is shown in Plate 2 fie. tc A slightly different type is shown in Plate 1, fig. 2, and Plate 2, fig. 3. The details of construction of this apparatus, which was made by Mr. H. J. Detrick, at Polo Plantation, Dumaguete, are as follows: : The sulphuring box, inside measurements 120 centimeters (4 feet) wide, 265 centimeters (8 feet, 10 inches) long, 210 centi- meters (7 feet) high, was constructed of tongue-and-groove lum- bayao, 2.5 centimeters (1 inch) thick. Hight uprights 5 by 10 centimeters (2 by 4 inches) were employed. It was erected -in a shed with galvanized iron roof and dirt floor. One length, about 5 meters (17 feet), of 60-centimeter (24- inch) cane track, and two pairs of wheels, with rough boxings, were purchased from a sugar hacienda at Bais. A frame of 5-by-10-centimeter (2-by-4-inch) and 5-by-15-centimeter (2-by- 6-inch) pieces was fitted over the wheels. The track was laid in a ditch 20 centimeters (8 inches) deep. Sixteen trays, outside measurements 90 by 230 by 10 centi- meters (3 feet by 7 feet 8 inches by 4 inches), were constructed of malatabigi about 3 centimeters (14 inches) thick. The bottoms were made of bamboo strips about 6 centimeters (2.5 inches) wide, leaving a maximum opening of 1.3 centimeters (0.5 inch). This floor rested at the ends and at three places between the ends on horizontally laid 5-centimeter (2-inch) strips. The whole was strengthened by nailing above the bam- boo directly over each of the crosspieces, except the central one, a strip of 6-by-3-centimeter (23-by-1}-inch) malatabigi. This construction produced strong ends made up of three pieces, one 10 centimeters (4 inches), one 6 centimeters (2.5 inches), and one 5 centimeters (2 inches). The 6-centimeter (2.5-inch) pieces were laid upright, like floor joists (see Plate 2, fig. 3). The capacity of this box was 3,000 nuts (without shells). * Brill, H. C., Parker, H. O., and Yates, H. S., Copra and coconut oil, Philip. Journ. Sci. § A 12 (1917) 80. 21,1 Wells and Perkins: Sulphur for Copra Drying 51 Cost of constructing the sulphuring box. Pesos. Wheels and track, second hand 20.00 2-by-4-inch lumber, 80 feet at 0.125 peso 10.00 Tongue-and-groove lumbayao, 252 feet at 0.16 peso 40.32 1g-inch malatabigi, 260 feet at 0.148 peso 38.48 Bamboos, 25 at 0.15 peso 3.75 Nails, 7 kilograms at 0.70 peso 4.90 Labor 14,40 Transportation, approximate 20.00 Total 151.85 Sulphur burner—The sulphur is burned in a pan placed on the dirt floor of the box. To utilize crude native sulphur a fiat pan, made from a kerosene can, was devised by Mr. Detrick. The top and the bottom were removed, and the sides flattened out in one piece. The edges were then bent up to form a tray 22 by 75 centimeters. The amount of sulphur to be used depends on the rainfall and conditions of drying. One kilogram for each charge is usually sufficient, and requires about four hours in burning. Pure sulphur is sold in Manila drug stores for 0.30 peso per kilogram, but the crude variety found in Silay, Occidental Ne- gros; Biliran, Leyte; Camiguin, Misamis; Camiguin, Cagayan; © and other parts of the Philippines is also suitable for this use. PREPARATION OF THE NUTS The nuts can be sulphured before removal of the shells, but it is better to sun them, if possible, for an hour or so after open- ing (see Plate 2, fig. 4); then the meat is separated from the Shell with a copra knife, and placed on the sulphuring trays. The preliminary sunning shortens the work by loosening the meat. In case the trays are made with solid sides and placed directly on each other with no air space between they are stag- gered, as shown in Plate 1, fig. 1, in order to allow circulation of the sulphur fumes. Loading the high pile of trays on the truck is facilitated by digging a shallow ditch for the track, as Shown in this photograph. The loaded truck is rolled into the box, a match applied to the sulphur, and the door closed tightly. If 1 kilogram of sul- phur is used, six hours in the box is sufficient for the full effect of the fumes. Larger charges necessitate somewhat more time, or the entrance of a little air at the bottom may be permitted, to burn the sulphur more rapidly. 52 The Philippine Journal of Science 1922 DRYING The use of sulphur has been expected by some to dry the copra. This is not the case, but sulphured coconut meat can be dried in a shed without sunshine, if necessary. Plate 2, fig. 2, shows sundried copra which had been placed in the bodega Several times on account of rain. One amacan in the left fore- ground shows plainly the ravages of mold under those condi- tions on copra which has not been sulphured. The clean white copra on the other amacans in the foreground was treated in exactly the same way except that it was sulphured soon after opening. One and one-half kilograms of sulphur per charge of 2,500 nuts has been used in this case. In regions where the rainy season is severe, drying sheds can be constructed in which trays of copra can be stacked with air space above every tray for ventilation. In such a shed sulphured copra can be dried in two or three weeks without sunshine. In regions where the rain is less severe the expense of shed construction can be avoided by sundrying on a floor, amacans, or trays. At Polo Plantation it has been found most satisfactory to spread the coconut meat thinly on light trays, which are easily stacked and covered with a small nipa roof at night or in case of rain (see Plate 2, fig. 1). The time required for drying is four days or more, depending on the weather. NATURE OF THE PRODUCT Sulphured copra, properly dried, has a lighter appearance than the whitest sundried copra, because the sulphur partially bleaches the dark outer skin of the meat as well as preserves the natural whiteness of the inside surface. The moisture con- tent is about 5 per cent, and the free fatty acid less than 1 per cent. The sulphur dioxide partly evaporates from the copra during drying, and partly oxidizes to sulphuric acid. The sulphuric acid formed does not remain in the oil, or injure it in any way, It does remain in the press cake, however. For this reason, and because there is sometimes difficulty in expressing all of the oil from very highly sulphured copra, it is better to sulphur lightly when possible. It has been found that when about a kilogram of sulphur per 3,000 nuts is used, on an average, the protection against mold is ample and no trouble is met with in expressing the oil or in using the cake for animal feed. The cake produced is, in fact, superior to the ordinary grade of cake made from rancid, moldy copra. « 21,1 Wells and Perkins: Sulphur for Copra Drying 53 The direct protection afforded by sulphur dioxide does not last more than a month, but it is well known that copra is not susceptible to mold if once thoroughly dried and kept in a rea- sonably dry place. Copra beetles, unless checked, will eventually attack any copra and rapidly destroy copra that has become moldy. They are not attracted, however, by sound copra, so that the prevention of mold by sulphuring also prevents spoilage by beetles, for a number of months, at least. WHEN TO USE SULPHUR Sulphuring should be done soon after the nuts are opened, if rainy weather is anticipated. If unsulphured copra is rained on unexpectedly, it can be sulphured even after mold has started. The mold, or insects, can always be killed, and immediate further damage prevented, but it is advantageous to sulphur as soon as possible to prevent damage and loss of oil. ’ Sulphuring is valuable chiefly as an adjunct to the sundrying process. It is not necessary if a rapidly acting hot-air dryer is used, but can be employed to supplement hot-air drying in homemade kilns. It is also of value to the copra trader, who is often obliged to redry low-grade copra, and to kill the insects and the mold that otherwise soon complete its destruction. Fig: 3, ILLUSTRATIONS PLATE 1 Sulphuring box with sixteen trays, containing the meat from about 3,000 nuts. . Sulphuring box with nine trays. These nuts (1,200) are to be sulphured before removing the shells. ' PLATE 2 Stacks of sulphured copra ready for rain. Courtesy of Mr. Detrick. Copra, sulphured and not sulphured. The amacan in the left fore- ground shows the best grade of copra that could be produced at that time without the use of sulphur. . Suphuring trays. The drying trays are made lighter than these. . Opening nuts at Polo Plantation. This lot of over 50,000 nuts was subsequently sulphured in the box shown in Plate 1, fig. 1. Courtesy of Mr. Detrick. 55 (Puivir. JourN. Sct., 21, No. 1. Copra DryInoc.] WELLS AND PERKINS: Sulphuring box with sixteen trays. i. Fig. &. a x ot ae ¢€ © with nine trays. Sulphuring box 2 Fig PLATE 1. WELLS AND PERKINS: CoprA DryYING.] (PHILIP. JOURN. Scr., 21, No. = ae tes Fig. 4. Opening nuts at Polo Plantation, PLATE 2. HISPINEN DER ALTEN WELT Von J. WEISE Herischdorf i. Riesengebirge, Germany Eine Sendung von Herrn Donckier in Paris und von Herrn T. C. Moulton aus dem Sarawak Museum, sowie die Ausbeute welche Herr Drescher in neuester Zeit auf Java machte und an meinen Collegen Herrn Reineck in Berlin schickte, lieferten hauptsachlich den Stoff zu den folgenden Beschreibungen und Anmerkungen. Auch zwei sehr hiibsche Arten von den Philip- pinen konnte ich einfiigen. Botryonopa grandis Baly. Das Weibchen, 26 Millimeter lang, hat eine einfache Hinter- brust und vor dem abgestutzten Hinterrande des letzten Abdo- minalsegmentes eine sehr grosse Grube, die doppelt so breit wie lang und einem Kreisabschnitte Ahnlich geformt ist, wihrend das kleinere Mannchen im vorderen Teile der Hinterbrust zwei nebeneinander liegende Zihnchen und im Hinterrande des letzten leicht gewélbten Bauchsegmentes eine schwache Ausrand- ung besitzt. Anisodera modesta sp. nov. Supra brunneo-rufa, subtus rufo-picea, antennis saepe nigris, articulis quinque basalibus supra, quatuor subtus subglabris, nitidis, prothorace latitudine longiore antrorsum leniter an- gustato, disco sat crebre punctato, elytris nitidulis fortiter et regulariter striato-punctatis, tenuissime costatis, apice leviter _ emarginatis. Long., 13 ad 15 mm. Borneo, Matang, April, 1902, 3200 F.; Kuching, September 19, 1899; Rook Road, August 9, 1909; Mount Saribu, Mai, 1902 (Museum Sarawak). Mit A. sheppardi am nachsten verwandt, durchschnittlich kleiner als diese und bedeutend schlanker, die Fiihler etwas diin- ner und ihre ersten fiinf Glieder an der Spitze nicht knotig ver- dickt, die Fliigeldecken ganz ahnlich stark, aber regelmassiger 57 58 The Philippine Journal of Science 1922 punktirt, mit kleinerer Spitzenausrandung, die Rippen viel feiner und die erste nie vor der Mitte ausgebildet. Schlank gebaut und massig gewoélbt, oben dunkel braunlich rot, unten pechbraun oder schwarz, fhnlich auch die Fiihler ge- farbt, Schildchen schwarz, die Nahtkante der Fliigeldecken da- hinter ein kurzes Stiick angedunkelt. Stirn ziemlich schmal, matt, zur Mittelrinne abfallend. MHalsschild um ein Drittel langer wie breit, cylindrisch, matt, die Seiten ziemlich parallel, vorn leicht convergirend, die Scheibe ziemlich dicht und krafti¢ punktirt, die Mittellinie vor der Mitte glatt und in der Regel erhoht und glanzend. Fliigeldecken schwach glanzend, stark in regelmassigen Reihen punktirt, nur die fiinfte und sechste (ganze) Reihe hinter der Mitte durch einige dazwischen tre- tende Punkte leicht gestért; die Rippen sind scharf, aber fein, die erste ist nur im letzten Drittel oder Viertel vorhanden und schwicher wie die iibrigen, die zweite und dritte sind normal. Anisodera integra sp. nov. Supra opaca, antennis, scutello corporeque subtus piceis, an- tennis articulis quinque supra, quatuor subtus subglabris, niti- dis, prothorace longiore quam lato, utrinque leniter angustato, crebre punctato, cum capite obscure ferrugineo, elytris dilute flavo-testaceis, subtiliter striato-punctatis, apice communiter ro- _ tundatis, haud emarginatis, dorso tenuiter costatis, costa prima medio longe interrupta. : 3 Long., 13 mm. Borneo, Kuching, Mai, 1900 (Museum Sarawak). Der vorigen ahnlich, die Fliigeldecken jedoch heller gefirbt wie der Vorderkérper, matt, das Halsschild etwas dichter, die Fliigeldecken feiner und flacher punktirt und an der Spitze schmal gemeinschaftlich abgerundet, ohne Ausschnitt. Unterseite, Fiihler, und Schildchen pechschwarz, Kopf und Thorax dunkel rotbraun, Fliigeldecken blass rétlich gelbbraun. Fihler kraftig, die ersten fiinf Glieder cylindrisch, oben fast glatt, glianzend, unten Glied 5 eben so dicht wie die folgenden behaart. Stirn schmal, zur Mittelrinne abfallend. Halsschild um ein Viertel linger als breit, beiderseits leicht verengt, dicht punktirt, ein kleiner gerundeter Raum der Mittellinie vor der Mitte glatt. Fliigeldecken fein und wenig tief gereiht-punktirt, die Rippen schmal und ziemlich scharf, die erste ungefahr von ein bis drei Viertel der Lange unterbrochen, die zweite im ersten Viertel erloschen, die Schulterrippe ganz. 21,1 Weise: Hispinen der Alten Welt 59 Anisodera marginella sp. nov. Nigra vel picea, fronte, prothorace (lateribus exceptis) ely- trisque ferrugineis; antennis articulis quinque supra, tribus subtus subglabris, nitidis, prothorace longiore quam lato, sub- cylindrico, sat crebre punctato, elytris sat subtiliter striato-punc- tatis, apice parum emarginatis, limbo angusto laterali postice abbreviato nigro. Long., 12 ad 15 mm. Matang, Februar 14, 1900; Sior, Juni 1, 1910 (Museum Sarawak). Wie A. modesta gebaut, aber das Halsschild noch etwas linger, die Fliigeldecken feiner und hinten verworrener punktirt, ohne recht deutliche Rippen, an der Spitze mit einer sehr kleinen, jedoch etwas tieferen Ausrandung und auf der Unterseite der Fiihler nur drei Glieder glanzend; A. testacea Gestro hat einen viel kiirzeren Thorax. Schwarz bis rétlich pechbraun, Stirn, Thorax, und Filiigel- decken blass rostrot, ein schmaler Seitensaum des Halsschildes und der Fliigeldecken (auf diesen hinten mehr oder weniger abgekiirzt) schwarz. Stirn zwischen den Augen sehr schmal, Thorax etwas mehr als um ein Drittel langer wie breit, cylin- drisch, vor und hinter der Mitte unbedeutend ausgerandet, ziem- lich dicht punktirt, matt. Fliigeldecken wenig glanzend, fein gereiht-punktirt, die Reihen hinter der Mitte meist sehr verwor- ren; von den Rippen ist nur die zweite in den hinteren zwei Dritteln angedeutet, ausserdem sind vor der Spitze kurze Stiicke der tibrigen zu bemerken. Anisodera sinuata sp. nov. Testacea vel obscure ferruginea, subtus obscurior, antennis (saepe), scutello pedibusque nigris, illis articulis primis quatuor supra, tribus subtus subglabris, nitidis, prothorace longiore quam lato, subcylindrico, minus crebre punctato, basi late de- presso, elytris sat subtiliter striato-punctatis, obsolete bicostu- latis, apice rotundatim-emarginatis. Long., 9 ad 13 mm. Matang, December 29, 1909; Januar 7 und Mai 31, 1910. (Museum Sarawak). Eine der kleineren Arten; durch die Fiihler, an denen die vier Basalglieder oben und drei unten fast kahl und glanzend sind, das lange Halsschild, und die verhdltnismidssig fein punktirten Fliigeldecken zu erkennen, welche in der Spitze einen bogenfor- 60 The Philippine Journal of Science 1922 migen Ausschnitt besitzen, der merklich linger wie der von A. sheppardi ist. Von den primaren Rippen ist die zweite schmal, niedrig, aber deutlich zu erkennen, wahrend die dritte noch schwicher ist und wenig in die Augen fallt. Der Penis bildet eine schwach gekriimmte und von oben leicht zusammenge- driickte Réhre, welche etwa so lang wie der Bauch und von der Basis aus allmihlich bis vor die Mitte der Oeffnung leicht erweitert ist, dann verengt sie schnell, gerundet, in eine breite, aufgebogene, wenig vortretende Spitze. Die Gattung Anisodera s. str. hat eine eigene Skulptur der Fliigeldecken, die aus 10.5 Punktreihen und 3 primdren Rippen besteht. Die erste Rippe, die von der Naht durch die abgekiirzte Punktreihe am Schildchen und die beiden ersten ganzen Reihen getrennt wird, ist ganz, zum Beispiel bei A. guerini, oder in der Mitte unterbrochen (A. lucidiventris), bei vielen Arten nur im letzten Drittel bemerkbar, selten voéllig geschwunden; die zweite liegt normal, zwischen der vierten und fiinften ganzen Punktreihe und ist nur in einigen Arten (A. scutellata, A. gue- rini) ganz, sonst vorn abgekiirzt; die dritte liegt vorn norma! zwischen der sechsten und siebenten Punktreihe, biegt in etwa ein Drittel der Lange etwas nach aussen und wird secundar, da sie die siebente und achte Punktreihe bis zur Spitze trennt. Auf dem hinteren Abfalle der Fliigeldecken bildet sich bei vielen Arten jedoch neben ihr noch ein kurzes Stiick der dritten pri- miaren Rippe aus. Von den ersten Fiihlergliedern sind an der Spitze knotig verdickt: Glied 1 bis 5 bei A. sheppardi, 2 bis 5 bei A. sculpticollis, 3 bis 5 bei A. whitei und 8 bis 6 bei A. scu- tellata. Nach den Beschreibungen kénnten vielleicht folgende, im Cataloge von mir unter Lissochila aufgefiihrte Arten zu Las- siochila gehéren: humilis, nasuelli, propinqua, und testacea. Genus DRESCHERIA novum Corpus valde elongatum, subparallelum, depressiusculum. An- tennae sat breves, 11l-articulatae, articulis 5 ultimis dilatatis. Frons inter antennas in cornum subquadratum producta. Pro- thorax subquadratus, pone medium leviter angustatus. Elytra prothorace vix latiora et plus quam quintuplo longiora, apice angustata et subtruncata, angulo suturali dentiformi instructa, dorso in seriebus 8.5 punctato striata, subtiliter tricarinulata, carinis internis antice longe deficientibus. Pedes breves. Diese Gattung gehért in die Gruppe der Cryptonychinen zwi- schen Octodonta und Bronthispa und unterscheidet sich von den verwandten Gattungen durch die Fliigeldecken, welche nicht 21,1 Weise: Hispinen der Alten Welt 61 10.5, sondern 8.5 Punktreihen und 3 zarte Rippen besitzen, von denen die beiden inneren nur nahe der Spitze vorhanden sind. Die Stirn ist viereckig, etwas breiter als lang, eben, jederseits von einer Augenleiste begrenzt, vorn schnell zu einem halb so breiten, missig langen Fortsatze verengt, welcher fast quad- ratisch und oben concav ist. Der Clypeus ist gross, lang, annahernd rechteckig, aber nach oben etwas verschmialert, abgeflacht, mit einer niedrigen, stumpfen Mittelleiste, fein run- zelig punktirt und dicht aufstehend behaart. Die Fiihler sind kurz und reichen kaum bis zur Basis der Fliigeldecken, Glied 1 dick, keulenférmig, 2 halb so lang und dick, 3 kleiner als 2, etwas langer wie eins der drei folgenden Glieder, die fiinf Endglieder sind erweitert, etwas zusammengedriickt, jedes wenig breiter und etwa so lang als 2, das letzte Glied eine Spur langer, mit verrundeter Spitze. Der Thorax hat ungefahr die Form wie bei einer Downesia, er ist etwa so lang als breit, mit abgerun- deten Vorder- und rechtwinkeligen Hinterecken, an den Seiten unbedeutend nach hinten verengt und dicht hinter der Mitte sanft ausgebuchtet, auf dem Riicken schwach queriiber gewdlbt. Fliigeldecken in den Schultern so breit wie der Thorax am Vor- derrande, in ein Drittel der Lange wenig eingeschniirt, dann ziemlich parallel, im letzten Viertel sanft verengt und an der Spitze gerundet-abgestutzt, mit vortretender Nahtecke; auf dem Riicken abgeflacht. Beine kraftig und sehr kurz, das vierte Tarsenglied kaum linger wie die Lappen des dritten Gliedes, Klauen einfach, gegenstindig. Ich nenne die Gattung nach Herrn Drescher, der in den letzten Jahren mit ausgezeichnetem Erfolge auf Java gesammelt hat. Drescheria reinecki sp. nov. Nigra, nitida, fronte subtiliter ruguloso-punctata, prothorace substrigoso-punctulato, linea media lata sublaevi, elytris sub- tiliter striato-punctatis. Long., 6 ad 7; lat. 1 ad 1.2 mm. Java, Mount Merbabu, Mai, 1912 (Drescher). Sehr langgestreckt, abgeflacht, glinzend schwarz. Stirn dicht und fein runzelig-punktirt, nur der Fortsatz oben glatt. Thorax auf einem breiten Mittelstreifen weitlaufig mit dusserst feinen Punkten besetzt, fast glatt, an den Seiten massig dicht punk- tulirt und vor der Mitte langsgestrichelt. Fitigeldecken mit 8.5 feinen Punktreihen, die in flachen und feinen, erst hinter der Mitte tieferen Streifen stehen. Die dritte primaire Rippe ist 62 The Philippine Journal of Science 1922 fein und niedrig, ganz, die beiden anderen sind kraftiger, aber nur im letzten Viertel vorhanden, auch die Naht is hinter der Mitte erhoht. Bronthispa javana sp. nov. Valde elongata, depressiuscula, flavo-ferruginea, nitida, an- tennis piceo-nigris, fronte ruguloso-punctata, prothorace sub- quadrato, crebre punctato, carinula obsoleta, utrinque medioque postico disci sublaevibus, elytris striato-punctatis (seriebus 5 ad 8 ante medium et seriebus 3 ad 6 juxta apicem series duas formantibus), apice bicarinatis, truncatis. Long., 9 mm. ¢ Java, Samarang (Drescher). Der mir unbekannten B. longissima von der Aru-Insel Wokan nahe verwandt, die Erweiterung der Thoraxseiten hinter dem Vorderrande ziemlich breit abgerundet und die Spitze der Flii- geldecken gradlinig abgestutzt. Sehr lang gestreckt, abgeflacht, gelblich rostrot, glanzend, die Fliigeldecken etwas dunkler, mit einem kaum merklichen bliu- lichen Anfluge versehen; Fiihler pechschwarz, Glied 1 lang und dick, fast cylindrich, 2 das kleinste, wenig linger als breit, 3 etwas langer wie 2, beide zusammen kiirzer als Glied 1, die drei folgenden Glieder unter sich ziemlich gleich, jedes langer als 8, von den fiinf breiteren, leicht zusammengedriickten Endgliedern ist Glied 7 linger als eins der folgenden und an der Innenseite, die tibrigen giinzlich, dicht, Ausserst kurz, und fein behaart. Stirn quadratisch, abgeflacht, dicht und etwas runzelig punktirt, mit feiner Mittelrinne, die auf den schmalen, stabférmigen, vorn abgestutzten Fortsatz iibergeht. Kopfschild lang, dicht absteh- end gelblich behaart. Thorax etwas langer als breit, mit convexem Vorderrande vor den Hinterecken, die durch einen Ein- schnitt dornférmig abgesondert sind, zundchst leicht erweitert, dann bis zum ersten Drittel schwach verengt, hierauf in einen abgerundeten Vorsprung erweitert, der vorn fast gradlinig verschmialert ist. Auf dem Vorsprunge liegt eine niedrige, verloschene und schriig nach innen und hinten ziehende Leiste, welche nebst einem Raume vor dem Schildchen glatt erscheint, aber unter stirkerer Vergrésserung zart punktulirt ist; die iibrigen Teile sind dicht punktirt. Die Punktreihen der Flii- geldecken sind regelmissig, aber die dritte bis sechste Reihe hinten, die fiinfte bis achte vor der Mitte auf zwei Reihen be- schrankt. Auf dem Abfalle zur Spitze sind ausser der erhdhten Naht noch die beiden ersten primiren Rippen deutlich: ausge- & 21,1 Weise: Hispinen der Alten Welt 63 pragt und die Spitze ist von der ersten Rippe der einen Decke bis zu der der andern abgestutzt. Die Zwischenstreifen sind dus- serst fein punktulirt. Xiphispa obligata sp. nov. Elongata, dilute rufo-testacea, supra subopaca, capite anten- nisque obscurioribus, elytris nigris. Prothorace sat crebre punctato lateribus pone apicem angulatis, elytris crebre striato- punctatis, carinulis primariis subtilissimis, sat obsoletis, tertia antice longe abbreviata. Long., 7 mm. Java, Salatiga. Die vorliegende Art bildet das bisher unbekannte Bindeglied zwischen den Arten von Neu-Guinea und Madagascar; sie erinnert in der Farbung am meisten an X. meijerei. Der K6r- per ist langgestreckt, wenig gewdlbt, rétlich gelbbraun, oberseits fast matt, Fliigeldecken schwarz, Kopf und Fiihler dunkel rost- rot, die Spitze der letzteren und der Hals schwirzlich. Fiihler schlank, alle Glieder deutlich getrennt, die fiinf fast kahlen Endglieder nur unbedeutend stiirker als die vorhergehenden, so dass die Art zu Xiphispa i. sp. gestellt werden muss. * Stirn quadratisch, eben, dicht runzelig punktirt, mit einer Mittel- rinne, die sich auf*das schmale, stabchenférmige, vorn abzestutzte Kopfhorn verlingert. Thorax etwas linger als breit, ziemlich dicht punktirt, die Seiten hinten fast parallel, vor der Mitte in einen kleinen stumpfwinkeligen Zahn erweitert, davor conver- girend, die Vorderecken in Form eines sehr kurzen Kegels vortretrend. Schildchen glatt. Fliigeldecken in den stumpfwin- keligen Schultern merklich breiter wie das Halsschild, sodann ziemlich von gleicher Breite, nur hinter der Schulter leicht eingezogen, am Ende gemeinschaftlich abgerundet; dicht und ziemlich stark in regelmissigen Reihen punktirt. Zwischen- streifen sehr schmalt die graden wenig héher wie die ungraden und feine Leisten bildend, von denen die dritte vor der Mitte fehlt, weil die Punktreihen 5 bis 8 daselbt auf zwei verringert sind. Prionispa magnifica sp. nov. Cuneiformis, subtus ferruginea, femoribus, tibiis antennisque pallide flavis, his articulis quatuor ultimis nigris, supra metallico- viridis, elytris flavo et cupreo marginatis, angulo postico ex- tremo lobatis, dorso carina prima pone medium, secunda medio cupreis instructis. Long., 3.5 mm. 64 The Philippine Journal of Science 1922 Luzon, Mount Maquiling (Baker). Kleiner als P. gemmata und P. pulchra, der hintere Aussen- winkel der Fliigeldecken nicht zugespitzt oder dornférmig wie bei diesen, sondern in einen schréag nach aussen und hinten gerichteten dreieckigen, an der Spitze breit abgerundeten Lappen erweitert; von P. vethi durch helle Fiihlerfarbe und die hinter der Mitte der Fliigeldecken fehlenden Hocker, von P. longicornis endlich durch die sehr abweichende Skulptur der Fliigeldecken verschieden. Die Unterseite. ist glinzend braunlich rostrot, der Bauch heller; Schenkel, Schienen, und Fiihler blassgelb, die vier Endglieder der letzteren tief schwarz. Die Oberseite ist glan- zend und lebhaft metallisch griin, Stirn und Thorax goldgelb iiberflogen, ein breiter Seiten- und Hinterrandsaum der Fiiigel- decken blassgelb, innen briunlich-kupferig angehaucht, ebenso der Ecklappen jeder Fliigeldecke und der vorhandene Rest der beiden ersten Rippen. Die erste ist ungefahr im vierten Fiinftel der Linge, die zweite im mittleren Drittel ausgepragt. Stirn sparsam punktirt, vorn in einen dreieckigen Zipfel aus- gezogen. Drittes Fiihlerglied nur etwas linger als eins der beiden ersten Glieder. Thorax dicht und stark punktirt, in der Mittellinie glatt. Die Punkte in den Reihen-der Fliigeldecken sind denen des Halsschildes Ahnlich, die der getzten Reihe aber starker. Javeta moultoni sp. nov. Elongata, nigra, fronte prothorace apice femoribusque posticis brunneo-rufis. Prothorace foveis septem parce punctatis (ante- rioribus pubescentibus), elytris fortiter striato-punctatis, tricos- tulatis. Long. vix 5 mm. Borneo, Sarawak, Baram, November 10, 1910 (Museum Sarawak). Schwarz, massig glinzend, Oberlippe rotgelb, die Stirn bréun- lich rot, Ahnlich, doch dunkler, auch das Halsschild nach der Basis hin und die Hinterschenkel; die vier ersten Bauchsegmente am Hinterrande fein gelblich gesdumt. Stirn spiegelglatt, Tho- rax kaum so lang als breit, an den Seiten sehr wenig gerundet und vorn undeutlich, hinter der Mitte merklicher eingeschnurt, oben gewélbt aber sehr uneben; das erste Drittel wird von einem etwas niedergedriickten, in der Mitte sparsam und grob punk- tirten, nach den Seiten hin dicht punktulirten und dicht kurz und tlh rg 21,1 Weise: Hispinen der Alten Welt 65 fein aufstehend behaarten! Querstreifen eingenommen, welcher nahe dem Rande in eine tiefe Grube tibergeht. Hinter dieser befindet sich je eine Ahnliche aber kahle Grube, beide stark punktirt. Im mittleren Teile der Scheibe liegen drei kleinere, mit wenigen groben Punkten besetzte Gruben, zwei am Basal- rande dicht neben einander und eine davor. Die Fliigeldecken sind fast parallel, hinter der Mitte wenig verbreitert, auf dem Riicken in regelmissigen, leicht vertieften Reihen stark und sehr dicht punktirt. Die drei secundiren Rippen liegen normal (die erste zwischen der dritten und vierten Punktreihe) und sind deutlich, wenn auch nicht besonders stark, nur die dritte ist sehr schmal und fein. Diese Art erlaube ich mir nach dem Curator des Sarawak Museums Herrn J. C. Moulton zu benennen. Coelaenomenodera cucullata Guer. ist nach der Fiihlerbildung eine Balyana, denn die vier Endglieder bilden einen dicht be- haarten matten Cylinder, an welchem die Nahte nur leicht ange- deutet sind. Die ersten sieben Glieder sind getrennt, sparsamer behaart, und massig glinzend. Glied 1 ist kurz, dick, 2 wenig kiirzer aber diinner, 3 schlank, doppelt so lang als 2 und von allen das lingste, 4 merklich kiirzer, 5 wenig kiirzer wie 4, 6 dem zweiten dhnlich, 7 so lang als 5 aber nach der Spitze ver- dickt und hier so breit als das Endglied. Der Kérper ist 9 bis 10.5 Millimeter lang, glanzend rotgelb, die Tarsen angedunkelt und das zusammengesetzte Endglied der Fiihler schwarz. Tho- rax linger als breit, grob punktirt, nur die Zwischenwande der sechs Gruben (2, 2, 2) glatt und stark glanzend. Die beiden ersten Gruben (auf der iiber den Kopf vorgezogenen und vorn abgerundeten Kapuze) sind gross, sehr tief, lang eiférmig, die folgenden vier flach und viel kleiner, quer. Fliigeldecken grob punktirt-gestreift, die Intervalle bilden feine Leisten, von denen die dritte, fiinfte, und siebente nur unbedeutend stirker als die andern sind. Coelaenomenodera cucullata var. fuscicornis var. Nov. Ueber das variiren der Coelaenomenodera ist aus Mangel an Material zur Zeit wenig bekannt und ich kann hier auch nur zwei Fille anfiihren. Im Hamburger Museum steckt ein Exem- plar mit der allgemeinen Fundortsangabe Madagaskar *Die Behaarung der vorderen Thoraxgruben von mir bereits Arch. f. Naturgesch. (1905) 98 erwahnt, scheint der Gattung Javeta wirklich eigentiimlich zu sein. 187668-——5 66 The Philippine Journal of Science 1922 (Kiderlen) , welches ich noch zu C. cucullata rechne, obwohl an den Fiihlern die beiden ersten Glieder allein braunlich rot, die folgenden (ausser dem tief schwarzen zusammengesetzten End- gliede) pechbraun sind. Auch die Kapuze ist vorn nicht gleich- missig abgerundet, sondern gerundet-abgestutzt. Ich bezeichne dies Tier als var. fuscicornis. Sodann besitze ich ein Stiick aus Madagaskar (Staudinger) von Balyana armata Gestro, welches in folgenden Punkten von der Beschreibung der einzigen Type abweicht: Die Fiihler sind ‘nicht 9- sondern regelrecht 8-gliedrig, Glied 1 bis 3 unter sich an Linge gleich, 4 bis 6 jedes eine Spur ktirzer, 7 dagegen um die Hilfte linger, nach oben verbreitert und an der Spitze auch um die Hilfte breiter als 6, das zusammengesetzte achte Glied ist wenig dicker als 7 und so lang wie Glied 6 und 7 zusammen, am Ende missig scharf zugespitzt. Die Fliigeldecken haben an der hinteren Aussenecke keinen spitzen, nach hinten gestreckten Zahn, sondern eine stumpfwinkelige Ecke, welche durch die plétzliche Verengung des abgesetzten Seitenrandes entstanden ist. Dieses Stiick diirfte das andere Geschlecht von B. armata sein, eine Art, welche von den iibrigen durch die nur von einer Grube eingenommenen Kapuze abweicht. Coelaenomenodera suturalis Guer. Die schwarze Mittelbinde des Thorax nimmt an der Basis mehr als das mittlere Drittel der Breite ein, ist davor plétzlich stark verengt und endet in einer Spitze, die weit vom Vorder- rande entfernt bleibt. Madagaskar, Maroantsetra (Staudinger). Coelaenomenodera funerea sp. nov. Elongata, subtus nigra, supra atra, prothorace crebre et forti- ter subrugoso-punctato, cucullo subtriangulari lineisque tribus obsoletis nitidis, elytris valde striato-punctatis, transversim plicatis, tricarinulatis. Long., 11 mm. Madagaskar, Tananarive. Der C. distinguenda Fairm. sehr nahe stehend und vielleicht nur eine einfarbig schwarze Form derselben, aber die Fiihler etwas linger wie die von C. femorata und die dritte (secundare) Rippe der Fliigeldecken hinten kaum schwacher wie die beiden inneren. i Die Fihler reichen bis an die Basis der Fliigeldecken und sind ziemlich dick, Glied 1 und 2 klein, sparsam behaart, glanzend, die folgenden langer und breiter, matt, 3 bis 5 an den Seiten 21,1 Weise: Hispinen der Alten Welt 67 3 deutlich linger behaart als die folgenden, Glied 8 bis 10 etwas kiirzer wie die vorhergehenden und allmahlich verdiinnt, nur durch eine leichte Einschniirung getrennt, das Endglied stumpf zugespitzt. Scheitel glatt, durch eine unregelmassig verdoppelte, _starke punktreihe von der Stirn geschieden; auf dieser laufen zwei parallele Langsleisten nach vorn bis zwischen die Fihler und schliessen einen dicht punktirten und grau behaarten Strei- fen ein. Thorax von der Basis bis vor die Mitte allmahlich schwach verengt, davor wieder erweitert, hinter den héckerartig heraustretenden Vorderrande ausgeschnitten, letzterer kapuzen- férmig weit iiber den Kopf vorgezogen, glinzend, und mit zwei lang eiférmigen Gruben besetzt. Der tibrige teil der Scheibe ist grob, sehr dicht punktirt, und hat drei feine, unregelmassige, glatte Langslinien. Fliigeldecken bedeutend schlanker wie die von C: femorata, mit dicht nebeneinander stehenden Reihen von grossen und tiefen viereckigen, durch eine Querleiste getrennten Punkten und drei ziemlich schmalen Langsrippen. Die zweite geht vorn, nahe der Basis, in eine Punktreihe iiber, die dritte ist vor der Mitte niedriger wie hinter derselben. Das einzige Stiick befindet sich in der Sammlung von Herrn Donckier. Coelaenomenodera donckieri sp. nov. _Elongata, parallela, rubra, nitida, antennis fulvis, articulis quinque ultimis dilatatis, clavam formantibus, articulis tribus apicalibus nigris; prothorace subpolito, basi impresso, cucullo magno, fere semicirculari, foveis duabus magnis, punctatis im- presso, elytris striato-punctatis. Long., 4 ad 4.5 mm. Madagaskar, Tananarive, Mahatsinjo (Donckier, Staudinger). In Grésse und Farbe der C. coccinea am ahnlichsten, aber von ihr und den anderen roten Arten durch die ‘fehlenden Rippen der Fliigeldecken und die grosse, ziemlich halbkreisférmige Ka- puze welche die vordere Hilfte des Thorax bildet und weit iiber den Kopf reicht, auf den ersten Blick zu unterscheiden. Gestreckt und parallel, lebhaft glinzend rot, Fiihler rétlich gelb, ihre drei Endglieder und oft noch die Spitze des achten Gliedes schwarz, die ersten sechs Glieder ziemlich kraftig, Glied 2 etwas linger als eins der tibrigen, 7 massig, die folgenden etwas stirker erweitert, das Endglied kurz, konisch. Halsschild in der hinteren Halfte glatt, mit einer verloschenen Grube vor dem Schildchen und zwei grésseren und tieferen Seitengruben an der Basis; die vordere Halfte wird von den beiden annahernd 68 The Philippine Journal of Science 1922 halbkreisférmigen, kraftig punktirten Gruben der Kapuze ein- genommen. Fliigeldecken in regelmassigen Reihen nicht stark punktirt, ein Stiick hinter der Basis quer eingedriickt, ausserdem noch mit einer anderen weiten Vertiefung in der Mitte zwischen der vierten und siebenten Punktreihe versehen, an deren Aus-, senrande die dritte secundire Rippe auf eine kurze Strecke ange- deutet wird. Die iibrigen Zwischenstreifen sind eben und fast glatt. Coelaenomenodera straminipennis sp. nov. Elongata, straminea, nitidissima, antennis brunneo-rufis apice infuscatis, capite, prothorace laevi, postpectore utrinque tar- sisque dilute brunnescentibus; cucullo rotundatim subquadrato, bifossulato, elytris striato-punctatis, tricarinatis, carina secunda medio interrupta, tertia antice deficiente. Long., 6.5 mm. Madagaskar, Tananarive, Mahatsinjo (Donckier). In der Kérperform mit C. pallescens iibereinstimmend und auch in der Form der Halsschildkapuze ihr 4hnlich, aber ab- weichend gefirbt, viel glanzender und durch die Skulptur der Oberseite véllig verschieden. Strohgelb oder blass rétlich gelb, oberseits stark glanzend, Kopf und Thorax bradunlich gelb, eine verloschene Makel an den Seiten der Hinterbrust und die Tarsen hell rétlich braun, das zweite bis sechste Fiihlerglied rotbraun, das siebente und achte angedunkelt, letzteres eine Spur dicker und um die Halfte langer als das vorhergehende; die folgenden fehlen. Thorax linger wie breit, hinter der Mitte ziemlich parallel, die Kapuze davor etwas schmialer, von oben betrachtet an den Seiten ziem- lich parallel, vorn iiber den Kopf vorgezogen und abgestutzt, mit breit abgerundeten Aussenecken und zwei sehr tiefen ovalen Gruben, welche durch eine massig breite glatte Zwischenwand getrennt sind. Die Scheibe dahinter ist glatt und besitzt drei Hindriicke an der Basis und (wie bei den meisten anderen Arten) iiber dem Seitenrande zwei Langsvertiefungen, die von der Scheibe durch eine fast grade Leiste getrennt werden. Die Fliigeldecken treten an der Basis schrig heraus und sind in den Schultern bedeutend breiter wie der Thorax, dahinter ziemlich parallel, jedoch hinter dem ersten Drittel durch den heraustre- tenden abgesetzten Seitenrand leicht erweitert, auf der Scheibe gereiht-punktirt, mit drei secundiren Rippen. Diese haben un- gefahr dieselbe Hohe, die erste ist im vorderen Drittel abge- flacht, die zweite kurz, in der Mitte unterbrochen, davor undeut- 21,1 Weise: Hispinen der Alten Welt 69 lich, die dritte ist l4nger wie die zweite und erlischt allmihlich vor der Mitte. Von Mahatsinjo schickte Herr Donckier noch C. chermesina, C. coccinea und C. ornata ein. Promecotheca straminipennis sp. nov. Elongata, nigra, nitida, capite, thorace scutelloque sublaevibus, elytris, flavo-albidis vel stramineis creberrime striato-punctatis, parum nitidis, femoribus posterioribus dentatis. Long., 11 mm. Neu-Pommern (Staudinger). Diese Art ist zwar dhnlich gefarbt wie P. callosa, aber viel schlanker gebaut und auf den Fliigeldecken ganz abweichend punktirt. Sehr lang gestreckt, tief schwarz, glinzend, die Flii- geldecken blass strohgelb, wenig glanzend. Kopf und Hals- schild Ausserst fein und schwer sichtbar gewirkt, letzteres mit einigen zarten Piinktchen jederseits in der vorderen Halfte, hinten tief quer eingedriickt und eingeschniirt, davor an den Seiten keulenférmig erweitert und auf dem Riicken jederseits mit einer tiefen Grube versehen. Schildchen hinten breit gerundet-abgestutzt, deutlicher gewirkt wie der Thorax. Fliigel- decken unbedeutend breiter wie der Vorderkérper, fast parallel, © nur hinter der Schulter wenig ausgeschweift, hinten gerundet- verengt und in eine gemeinschaftliche kurze und stumpfe Spitze verlingert, oben wenig gewélbt, dicht gereiht-punktirt, die Punkte von miissiger Grosse. Es sind vorn zehn Punktreihen vorhanden, die in ein Drittel der Lange auf acht oder neun verringert, gleich darauf aber wieder auf zehn vermehrt sind. Die zwei ersten Reihen sind regelmissig, die beiden folgenden Reihen vermehren sich von etwa ein Viertel der Lange ab zu- nichst in drei, spiiter in vier Reihen, wihrend die sechste und siebente sich in ein Drittel der Linge mit der fiinften verbinden und nun als eine Reihe bis vor die Spitze laufen. Eine 4hnliche Streifung der Fliigeldecken findet sich bei keiner anderen Art, denn die mit P. cyanipes verwandten Tiere haben vorn zehn, hinter der Mitte elf bis zwélf Reihen, da die dritte und vierte dort eine oder zwei Zwischenreihen erhalten. Promecotheca antiqua Ws. ist in Neu-Pommern ein lastiges Tier geworden, da sie in den letzten Jahren mehrere Palmen- pflanzungen vollstindig verwiistet hat. Sie legt ihre Eier in den Vegetationskegel der jungen Pflanzen, den die zahlreichen Larven allmihlich zerstéren und so den Stamm zum absterben bringen. Kiirzlich erhielt ich einiges Material durch meine 70 The Philippine Journal of Science 1922 Tochter Elisabeth Kaiser von der Plantage Matanatar bei Kokopo, aus dem ich ersehe, dass die Beine der Art nicht immer einfarbig gelb bleiben; es werden zunichst die Hin- terbeine, spater auch in geringerer Ausdehnung die Mittelbeine, pechbraun. Durch die folgende Tabelle hoffe ich die Bestimmung der meisten Promecotheca-Arten erleichtert zu haben: Tabelle der Promecotheca-Arten. 1. Fliigeldecken auf dem Abfalle zur Spitze halb aufstehend behaart und am Hinterrande dicht bewimpert 2 1, Fligeldecken hinten kahl, vorn mit 10, hinter der Mitte mit 9, und am Ende mit 8 miassig starken und ziemlich regelmdssigen Punktreihen (die erléschenden liegen zwischen den Reihen 5 und 8). Gelb oder gelblich rot, Tarsen schwarzlich, Bauch braun oder schwarz, Fiihler schwarz (Glied 1 ausgenommen), Fliigeldecken dunkel metallisch blau, blaugriin oder griin, einfarbig oder (ab. reichei Baly) im ersten Drittel gelb. Lange 8 bis 10 mm. Tonga und Viti-Inseln. P. coeruleipennis Blanch. 2. Halsschild glatt oder nur mit wenigen Punkten 3. 2. Halsschild stark gerunzelt. Schmutzig gelb, matt; Fiihler, Unterseite, und Beine (mit Ausnahme der Schenkelbasis) schwarz, Fliigeldecken regelmissig in Doppelreihen punktirt, mit 3 kraftigen Rippen. Lange 9 mm. Java. P. peteli Guér. 8. Fliigeldecken mit 8 regelmissigen und gleichstarken Punktreihen, die nur auf der Schulter um eine vermehrt sind 4 3. Fliigeldecken mit 10 oder mehr Punktreihen, welche im ersten Viertel bedeutend staérker als dahinter und durch kurze kraftige Langsrip- pen getrennt sind 7. 4. Fligeldecken sehr grob und tief punktirt. 5. 4. Fligeldecken verhiltnismassig fein und flach punktirt. 6. 5. Kérper tief schwarz, nur die Fliigeldecken gelb. Lange 10 bis 11 mm. Noérdliches Australien, Neu-Guinea, Neu-Pommern........ P. callosa Baly. 5. Unterseite schwarz, Vorderschenkel, Mitte der Vorderschienen, und die Oberseite riétlich gelb, Fiithler und Schildchen schwarz. Lange 11.3 mm. Nérdliches Australien. P. varipes Baly. 6. Hell rostrot, Fliigeldecken mehr gelb, Fiihler in der Regel schwarz, alle Schenkel mit einem spitzen Zahne; Thorax linger als breit, Schildchen abgestuzt, lang rechteckig. Linge 7 bis 9 mm. Philip- pinen, Palawan. Borneo. P. cumingi Baly. 6. Hell rostrot, oben fast matt; Thorax schwach quer, Schildchen hinten abgerundet, Bauch und der grésste Teil der Beine braun, violett schimmernd, Schenkel ungezihnt. Lange 10 mm. Philippinen: Bohol. P. octostriata Chap. 6. Sehr schlank, rotgelb, Fiihler und Bauch (letzterer teilweise oder ganzlich), zuweilen auch die Hinterbeine schwarz; Fliigeldecken vor der Mitte gelb, dahinter bldulich schwarz; Mittel- und Hinterschen- kel gezihnt. Lange 5.5 mm. Philippinen, Mindanao: Luzon, Los Bafios; Ticao. P. apicalis Ws. 7. Kopf und Halsschild rein schwarz 8. 7. Kopf, Halsschild, Schildchen, Unterseite, und Beine glinzend und gesattigt metallisch griin, Fliigeldecken lehmgelb bis gelblich rot, fast 21,1 Weise: Hispinen der Alten Welt 71 matt, Fihler schwarz (ab, trilbyi und scorpio Thoms.) ; oder Kopf, Thorax, Schildchen, Vorder- und Mittelbrust, und die Basis der Vorderschenkel gelblich rot, metallisch griin tiberflogen; Schildchen dreieckig, Fligeldecken im Basaldrittel mit 10 Punktreihen (die ersten 5 viel starker und tiefer als die iibrigen), dahinter mit 11 oder 12 feineren Reihen, von denen nur die beiden ersten ganz regelmissig sind. Linge 11 bis 13 mm. Philippinen, China P. cyanipes Er. 8. Schenkel ungezahnt 9. 8. Mittel- und Hinterschenkel gezihnt; tief schwarz, Fliigeldecken stroh- gelb, vorn und hinten mit 10, in ein Drittel der Linge auf sehr kurzer Strecke mit 8 Punktreihen. Lange 11 mm. Deutsch Neu- Guinea: Neu-Pommern P. straminipennis sp. nov. 9. Langgestreckt, schwarz, Beine rostrot, Hinterbrust pechbraun, Thorax in der mitte zart gewirkt, hinten und an den Seiten fein gestrichelt und sparsam punktulirt, fettig (2) oder starker glinzend (d), Fligeldecken parallel, dunkel metallischgriin, seltener blau, vorn gelb, beide Farben durch eine leicht concave Linie getrennt, die etwas hinter dem Schildchen beginnt und in der Nahe des ersten Drittels am. Seitenrande endet. Die Punktreihen der Fliigeldecken werden hinter dem ersten Viertel durch ziemlich feine Querrunzeln gestort. Lange 9 bis 11 mm. Neue Hebriden.................... P. opacicollis Gestro. 9. Massig gestreckt, Fliigeldecken nach hinten erweitert, schwarz, die drei ersten Fiihlerglieder ganz oder zum teil, Schildchen und Beine hell rostrot, Fliigeldecken gelb, das letzte Drittel schwarzblau, vorn durch eine gemeinschaftliche, schwach concav gebogene Linie begrenzt, die Punktreihen hinter der Basis durch ziemlich kriaftige Querrunzeln gestért; zuweilen ist die Mitte der Brust und des Bauches rotlich, oder die 4 hinteren Beine sind mehr oder weniger angedunkelt. Lange 7 bis 9.5 mm. Neu-Pommern. (Dieser Art miissen die mir unbe- kannten papuana und biroi Csiki sehr nahe stehen.).... P. antiqua Ws. In der Gattung Wallacea hat sowohl Baly wie Chapuis die Fiihlerbildung nicht richtig angegeben. Die ersten vier Glieder sind namlich missig glinzend, ziemlich dicht punktirt, und tragen in jedem Punkte ein iusserst kurzes weissliches Harchen ; das erste Glied ist dicker und etwas linger wie eins der drei folgenden, die sich von ihrer Basis aus nach der Spitze hin erweitern. Die sieben Endglieder dagegen sind cylindrisch, ausserst dicht und fein punktulirt und behaart, matt. Bei der typischen Art, bowringi, sind die vier ersten Glieder dunkel rostrot, die iibrigen schwarz, also nicht wie Baly sagt: “antennae black, basal joint obseure fuscous beneath.” Downesia marginicollis sp. nov. Atra, subnitida, prothorace parce punctulato in lateribus sat late marginato, elytris brunneo-rufis nitidis, basi sublaevibus tum subtiliter striato-punctatis. Long., 8.5 mm. 72 The Philippine Journal of Science 1922 China, Tschonting (Staudinger). Diese Art erinnert on D. fulvipennis, zeichnet sich aber durch den verhiltnismissig breit abgesetzten und deutlich aufgeboge- nen Seitenrand des Halsschildes und die Punktirung der Fliigel- decken aus, welche im grésseren Teile der vorderen Hialfte ziemlich erloschen ist. Schwarz, wenig glinzend, die Fliigeldecken dunkel briunlich rot, glanzender. Kopf fast glatt, mit wenigen, nur bei starker Vergrésserung sichtbaren Piinktchen nahe den Augen. Hals- schild etwas linger wie breit, nach hinten leicht verschmialert, sehr gering gewélbt, kaum bemerkbar gewirkt und weitliufig punktulirt, an den Seiten zu einer stirker punktirten, ziemlich breiten Rinne abfallend, welche den vorn verdickten Rand deutlich emporhebt. Auf den Fliigeldecken sind die Punkte der beiden inneren Reihenpaare bis hinter die Mitte fast erloschen, die des dritten Paares fehlen auf der Schulter. Die erste Punktreihe steht der ganzen Linge nach in einer Rinne, von der sich in der Mitte die zweite abzweigt, die dritte und vierte Reihe nahern sich vor der Mitte und bilden dann bis zur Basis einen flachen Streifen; die fiinfte und sechste Reihe trennen sich etwas vor der Mitte, und die beiden ausseren, deren Punkte von gleicher Starke sind, bilden von ein Drittel der Linge bis wenig hinter die Mitte nur eine Reihe. Die erste Rippe bildet bis nahe an den Abfall zur Spitze, die zweite bis zur Mitte, und die dritte im ersten Viertel, eine ziemlich breite Flache. Downesia abdominalis sp. nov. Elongata, nigra, nitida, abdomine rufo-flava, antennis sat longis, prothorace sublaevi lateribus subtiliter, marginato, ely- tris striato-punctatis, seriebus 7a et 8a ubique separatis. Long., 6 mm. Sikkim (Staudinger). Aehnlich gefirbt wie D. grandis, sehr schlank, tief schwarz, glanzend, der Bauch rétlich gelb. Fiihler langer wie in den meisten tibrigen Arten, bis auf die Schulter reichend, Glied 2 etwas langer wie 1 oder 3. Halsschild wenig linger als breit, die _ Seiten hinter den Vorderecken schnell] erweitert, sodann nach hinten allmahlich unbedeutend verengt, dusserst fein gerandet, die Scheibe fast glatt, sparsam mit sehr feinen Piinktchen be- deckt. Fliigeldecken in den Doppelreihen ziemlich fein punktirt, die beiden ersten vor der Mitte vereint, die erste Rippe bildet eine wenig deutliche schmale und niedrige Leiste, die beiden andern sind erloschen. 21,1 Weise: Hispinen der Alten Welt 73 Downesia javana Sp. nov. Elongata, subdeplanata, rufa, nitida, antennis elytrisque (his basi excepta) nigris, tibiis tarsisque infuscatis; prothorace parce obsoleteque punctulato latitudine paullo longiore, lateribus evidenter marginato, elytris sat fortiter striato-punctatis, valde tricostatis. Long., 3.5 ad 4 mm. Java (Drescher). Mit D. swmatrana nahe verwandt, aber nur halb so gross, dunkler gefirbt, die fiinf letzten Fiihlerglieder merklich dicker wie die vorhergehenden, die Punkte in den Reihen der Fliigel- decken etwas feiner, dichter, die Rippen stirker. Der K6rper ist diister rot, glinzend, Fiihler und Fliigeldecken schwarz, letztere an der Basis unbestimmt ritlich, Schienen und . Tarsen angedunkelt. Stirn glatt, mit wenigen Piinktchen zwi- schen den Augen. Thorax unbedeutend linger als breit, spar- sam und sehr fein punktulirt, schwach gewilbt, die Seiten fast parallel, nach hinten wenig convergirend und vorn etwas ge- rundet verengt, iiber der feinen Kante des Seitenrandes mit einer vertieften Laingslinie und vor dem Schildchen jederseits mit einer flachen Grube. Fliigeldecken wenig/breiter wie der Vor- derkérper, parallel, die beiden ersten Rippen hoch und kriftig, nahe der Basis verbreitert, die dritte schwiicher, die Intervalle tief, massig stark in Doppelreihen punktirt, von denen die beiden ersten in den vorderen drei Viertel der Lange, die beiden letzten in der Mitte auf eine Reihe beschrinkt sind. Downesia sumatrana Gestro sammelte Herr Drescher im Marz, 1906, bei Samarang auf Java. Die Downesia-Arten lassen sich in zwei Gruppen Teilen: 1. Die primaren Rippen der Fliigeldecken sind schmal und hoch, ihre Intervalle tief. a. Halsschild lingsrissig: Downesia strigicollis, angustata. b. Halsschild fast glatt oder sparsam punktirt: atrata, sulcata, basalis, picea, tarsata, insignis, sumatrana, javana. 2. Die Rippen der Fliigeldecken sind breit und flach oder kaum bemerkbar: Downesia gestroi, fulvipennis, grandis, elegans, andrewesi, auberti, kanarensis, marginicollis, abdominalis, Agonia tristicula sp. nov. Elongata, parallela, nigra, nitida, antennis articulis tribus ultimis flavo-albidis, basi femorum subtus macula rufa ornato, Prothorace sat fortiter punctato margine antico linea media sat lata postice abbreviata costulaque angusta utrinque laevibus; 74 The Philippine Journal of Science 1922 elytris coeruleo-nigris, fortiter striato-punctatis, tricostatis, costa tertia medio longe interrupta. Long., 6.5 mm. Borneo, Kuching (Sarawak Museum). Aehnlich gefarbt wie-A. lucida, an der Fiihlerspitze ein Glied weniger weisslich und die Beine nicht einfarbig rotgelb. Schlank, parallel, tief schwarz, glinzend, die Fliigeldecken blaulich schwarz und mit Ausnahme der beiden ersten Rippen wenig glanzend, drei Endglieder der Fiihler gelblich weiss und ein kleiner Fleck nahe der Basis auf der Unterseite der Schen- kel rétlich gelb. Thorax so lang wie breit, vor der Mitte am breitesten und kaum schmaler wie die Fliigeldecken, von hier nach vorn bis zu einem kleinen Ausschnitte hinter den Vor- derecken unbedeutend, nach hinten allmihlich etwas stirker, verengt, grob punktirt, ein Streifen am Vorderrande, in eine breite, hinten abgekiirzte Mittellinie verlingert, sowie eine stumpfwinkelige Langsleiste jederseits iiber die punktirten Stellen erhéht und spiegelglatt. Fliigeldecken stark in Doppel- _ reihen punktirt, die durch drei Rippen getrennt werden. Die beiden inneren sind ganz, vorn massig breit, nach hinten verengt, tief schwarz und spiegelglatt, die dritte ist schmal und nur nahe der Basis und vor der Spitze bemerkbar. Zwischen die dritte und vierte Punktreihe schiebt sich im ersten Drittel eine dritte Reihe ein. Agonia luzonica sp. nov. Elongata, subparallela, pallide rufo-flava, antennis, femorum apice, tibiis, tarsis, lateribus pectoris, abdominis et elytrorum (his in basi exceptis) nigris. Prothorace parce punctato, linea media subelevata carinulaque utrinque laevibus, elytris sat for- titer striato-punctatis, tricostatis. Long., 5.5 mm. Luzon (Baker 2270). Var. a. Dimidio postico elytrorum ramulis tribus antrorsum emittente nigro. Von den ahnlich gefarbten Arten durch die Rippen der Flii- geldecken verschieden, von denen die dritte ganz und fast eben so stark ist wie die beiden inneren. Gonophora chapuisi Baly, die nach der Beschreibung ebenfalls zu Agonia gehort, weicht ausserdem noch durch rotgelbe Unterseite und Beine ab. Gestreckt und ziemlich parallel, blass rétlich gelb, die Flii- geldecken fast rein gelb, die Fiihler, ziemlich die Spitzenhalfte der Schenkel, die Schienen und Tarsen, ein Streifen am Seiten- 21,1 Weise: Hispinen der Alten Welt 75 rande der Hinterbrust und des Bauches, ein Spitzensaum der Fliigeldecken, nach vorn auf der Naht bis zum ersten Viertel und am Seitenrande bis unter die Schulterbeule linienférmig verlingert, schwarz. In der var. a ist die hintere Halfte der Fliigeldecken schwarz und sendet einen gemeinschaftlichen, brei- ten, dreieckigen Zipfel auf der Naht bis zum ersten Viertel und einen schmalen Streifen am Seitenrande (innen von der dritten Rippe begrenzt) bis zur Schulterbeule. Fiihler missig lang, ihr drittes Glied um die Hialfte langer wie das zweite oder erste. Halsschild etwas breiter wie lang, an den Seiten hinten parallel, vor der Mitte schwach gerundet-verengt, auf der Scheibe ziem- lich weitliufig, nach den Seiten zu dichter punktirt, eine wenig erhéhte und beiderseits abgekiirzte Mittellinie nebst einer feinen, leicht gebogenen Lingsleiste jederseits davon glatt. Diese Leis- te begrenzt den vorderen Teil der normalen Seitengrube, die sich allm&hlich verschmdlert bis zum Schildchen hinzieht. Auf den Fliigeldecken sind die Punktreihen nicht besonders stark, aber dicht und regelmissig, nur zwischen die dritte und vierte Reihe ist vorn eine kurze Reihe eingeschoben. Agonia chinensis sp. nov. Elongata, postice leniter ampliata, lateritia, nitida, antennis nigricantibus basi ferrugineis, elytris nigris, nitidulis; protho- race parce punctato, antrorsum subangustato, elytris sat fortiter striato-punctatis, punctis antice seriebus septem, postice octo formantibus, costa tertia medio longe interrupta. Long., 5 mm. : China, Kiautschau (Staudinger). Durch Koérperform, Farbe, und Skulptur der Oberseite aus- gezeichnet. Gestreckt, nach hinten wenig erweitert, glanzend ziegelrot, Fiihler schwirzlich oder pechbraun, die beiden ersten Glieder rétlich, Fliigeldecken schwarz, missig glianzend. Thorax hinten sehr sanft, nahe den Vorderecken stirker gerundet- verengt, oben gewdlbt, sparsam kraftig punktirt, vor der Basis jederseits mit einem etwas dichter punktirten, schmalen, nach aussen verbreiterten Quereindrucke versehen. Fliigeldecken in den Schultern wenig breiter wie das Halsschild und weniger eckig wie in den verwandten Arten, sondern mehr gerundet- heraustretend, nach hinten schwach aber deutlich verbreitert, miassig stark gereiht-punktirt, dreirippig, die Rippen kriaftig, die dritte jedoch nur dicht an der Schulter und vor dem Hinterrande vorhanden. Die inneren Rippen laufen parallel, deshalb sind die ersten vier Punktreihen ganz regelmassig und vorn, zwischen 76 The Philippine Journal of Science 1922 der ersten und zweiten Rippe, nicht um eine vermehrt wie in den meisten tibrigen Arten; dagegen liegen auf der dusseren Halfte nur im letzten Viertel vier, davor drei Reihen. Agonia malangensis sp.. nov. Elongata, nigra, femoribus, tibiis, prosterno, capite (vertice excepto), prothorace (utrinque nigro-maculato) elytrorumque triente basali rufescentibus; capite laevi, prothorace parce punc- tato scutelloque nitidis, elytris minus nitidis, tricostatis, costa tertia medio interrupta. Long., 3.5 mm. Java, Malang (Staudinger). Zu den kleinsten Arten gehérig, langgestreckt, fast parallel, der Vorderkérper und die Unterseite glinzend, die Fliigeldecken fast matt. Unterseite schwarz, Schenkel, Schienen, und Pro- sternum gelblich rot, Seiten der Vorderbrust stark, die der Mittel und Hinterbrust feiner punktirt, letztere in der Mitte nebst den Bauchringen glatt, nur das Analsegment ziemlich dicht und fein punktirt. Fiihler pechschwarz. Kopf gelblich rot, Stirn spie- gelglatt, Scheitel schwarz. MHalsschild wenig breiter wie lang, an den Seiten leicht gerundet und beiderseits schwach verengt, eine Querrinne, die den Vorderrand emporhebt und die Seiten (namentlich deren grosse Grube) ziemlich dicht punktirt, neben der etwas erhéhten Mittellinie jederseits nur wenige Punkte. Fliigeldecken mit je vier regelmassigen Doppelreihen von star- ken Punkten und drei Rippen, von denen die beiden inneren gleichmassig kraftig sind, wihrend die dritte nur nahe der Schulter und hinter der Mitte deutlich ist. Agonia clavareaui Gestro. Ein Exemplar des Sarawak Museums, Borneo; Quop, Marz, 1914, stimmt in allen Stiicken mit der Beschreibung dieser Art iiberein, nur sind ausser den beiden Basalgliedern der Fiihler noch die drei folgenden (wenn auch etwas dunkler) rétlich ge- farbt. Agonia (Distolaca) crassicornis Gestro. Ist nach der Beschreibung unzweifelhaft eine Javeta, die ihn- lich wie gestroi gefarbt, aber viel kleiner ist und ein kiirzeres Halsschild hat. Ebenso kénnte pachycera Gestro zu Javeta ge- héren, da der Raum von der Naht bis zur ersten Rippe mit drei Punktreihen besetzt sein soll. 4 21,1 Weise: Hispinen der Alten Welt T7 Gonophora apicalis Baly. Diese Art wurde nach einem einzelnen Stiicke beschrieben; bei anderen Exemplaren von Luzon, Los Bafios (Baker), sind die zwei ersten Fiihlerglieder, die Basis des dritten, und die beiden Endglieder dunkel briunlich rot. Die vier Vertiefungen des Thorax (eine kurze und schrige iiber den Vorderecken und eine langere und tiefere jederseits vor dem Hinterrande) sind mit je einer kraftigen Punktreihe besetzt. Viel stirker noch sind die Punkte in den Reihen der Fliigeldecken, und deren schwarzer Spitzenfleck nimmt zuweilen etwas mehr als das letzte Drittel ein. Sein Vorderrand bildet zwischen den zweiten Rippen beider Decken eine grade Querlinie, nach aussen davon lauft er schrig nach hinten. Anisodera nigricauda Motsch. gehért nicht zu Micrispa, wie ich ? vermutete, sondern ist eine echte Gonophora, mit welcher G. taprobanae Gestro identisch ist. Das Exemplar meiner Sammlung stammt vom Originalfundorte Kandy. Gonophora (Micrispa) javana sp. nov. Ferruginea, subopaca, antennis medio flavis, lateribus post- pectoris, ano, scutello, apice et macula laterali pone medium elytrorum nigris; elytris lateribus sinuatis, costa tertia pone medium interrupta. Long., 3 mm. Java, Noesa Kambangan, Januar 10, 1913 (Drescher). In der Farbe und Zeichnung an G. minuta, in der Koérperform mehr an G. sinuata erinnernd, kleiner als beide, und von ersterer schon durch die hell gefiirbten Beine, von letzterer durch die wenig schwarz gezeichneten Fliigeldecken leicht zu _ unter- scheiden. Gestreckt und schwach gewéilbt, rostrot, die Rander und die Erhéhungen des Halsschildes nebst den Fliigeldecken noch heller und mehr gelblich, die Fiihler blassgelb, das erste Glied und die drei Endglieder rostrot, Schildchen schwarz, die Seiten der Hinterbrust, das letzte Bauchsegment, die Spitze der Fliigel- decken und eine Quermakel davor, vom Seitenrande bis zur zweiten Rippe ausgedehnt, schwdrzlich. Stirn glatt, Halsschild annahernd herzférmig, vor der Mitte am breitesten, nach vorn wenig, nach hinten stirker, verengt, die Seiten stark gerundet, ? Deutsche ent. Zeitschr. (1904) 457. 78 The Philippine Journal of Science 1922 hinten leicht eingeschniirt; auf der Scheibe sind wie gewodhnlich zwei Querstreifen starker vertieft, punktirt, die Mittellinie und ein Héckerchen jederseits davon erhéht und glatt. Schildchen hinten rechteckig, vorn erweitert. Fliigeldecken hinter der Schulter deutlich ausgeschweift, hinten wenig verbreitert, die drei Langsrippen auf jeder kraftig, die dritte an der schwarzen Makel unterbrochen, die zweite bis zwei Drittel der Lange weiter ‘yon der ersten entfernt als dahinter und durch drei Punktreihen getrennt. Dactylispa spinosa Weber vergleicht Gestro® mit seiner D. aspera und rechnet sie zu den Arten mit vier Seitendornen des Halsschildes; sie besitzt jedoch nur drei (2, 1), denn Weber sagt: “Thorax spinis quatuor erectis, acutis, furcatis et duobus lateralibus minoribus simplicibus;’’ und Fabricius: “‘Spinis qua- tuor nigris, lateralibus trifidis, intermediis bifidis.”” Zur fal- schen Auffassung dieser kleinen, 4 Millimeter langen, rostroten, auf den Fliigeldecken schwarzen Art von Sumatra ist Gestro durch den Vergleich eines Stiickes aus dem Briisseler Museum gekommen, welches von Chapuis ohne Grund fiir D. spinosa ge- halten wurde. Dactylispa flavolimbata, Sarawak, Serambu, Februar, 1913, und D. malayana, Sarawak, Banting, Mai 17, 1909 (auch von Herrn Drescher auf Java bei Noesa Kambangan im Juli, 1913 gefangen) sind nahe verwandt und bilden eine Gruppe die sich durch das Schildchen von den iibrigen Verwandten absondert. Bei letzteren bildet dasselbe eine ebene, dicht und sehr zart gewirkte matte Fliche, wahrend es in beiden oben genannten Arten zum gréssten Teile von einer glatten ovalen Quergrube eingenommen wird, die vorn nur einen schmalen, jederseits ver- breiterten, ebenen, und gewirkten Streifen frei liisst. Die drei Seitendornen des Thorax stehen auf einem gemeinsamen schlan- ken Stiele, die beiden ersten an der Spitze desselben, der dritte am Hinterrande wenig iiber der Basis. Der mittelste Dorn ist etwas langer wie die beiden anderen. Dactylispa drescheri sp. nov. Testaceo-flava, antennis piceis articulis 5 ultimis ferrugineis, elytris (limbo apicali excepto) nigris, sat nitidis, parce pilosis. Prothorace subopaco lateribus spinis tribus gracilibus, longis, proparte appendiculatis armato, elytris longe nigro-spinosis, spinis apicalibus brevioribus flavis. Long., 4 mm. * Ann. Mus. Civ. Genova (1897) 86. 21,1 Weise: Hispinen der Alten Welt 79 Java, Noesa Kambangan (Drescher). In der Farbung an D. weyersi, D. discoidalis und D. debilis erinnernd; von ersterer durch die Farbe der Fiihler und die zum Teil mit Anhingseln versehenen Thoraxdornen, von D. dis- coidalis durch die Thoraxscheibe, auf welcher der erhdhte glatte Querstreifen auf eine undeutliche, kleine, ovale Lingserhebung in der Mitte beschrankt ist, und von D. debilis durch den nicht freien dritten Seitendorn des Halsschildes zu unterscheiden. Gelb, auf Brust, Bauch, und Thorax mit rétlich brauner Bei- mischung, die Fliigeldecken schwarz, an der Spitze gelb gesiumt. Fiihler schlank, die ersten sechs Glieder pechschwarz, die etwas verdickten fiinf Endglieder dunkel rostrot; Glied 1 stark, 2 wenig schmaler, kurz, eiformig, die folgenden vier diinn, 3 we- nigstens so lang als 1, die drei folgenden kiirzer. Die Dornen der Oberseite sind verhaltnismissig lang, schlank. Der zweite Vorderranddorn des Thorax steht nicht ganz senkrecht, sondern ist schwach nach vorn geneigt und hat iiber der Mitte ein nach vorn gerichtetes Dérnchen. Von den drei Seitendornen ist der mittlere der lingste, mit zwei Dérnchen versehen, und hat mit dem ersten einen gemeinschaftlichen Basalstiel, an dessen Basis auch der dritte entspringt. Dieser ist etwas kiirzer als der zweite, aber etwas linger als der dritte Dorn; alle sind schwarz, an der dussersten Basis rétlich. Die Thoraxscheibe ist ziemlich matt, dicht verloschen und flach punktirt, mit zwei deutlichen Quer- und zwei unbestimmten Lingseindriicken, durch die ein verloschener, kleiner, kaum punktirter Raum in der Mitte em- porgehoben wird. Die Fliigeldecken sind gereiht-punktirt (in jedem Punkte ein feines, aufgerichtetes greises Hirchen) und haben ausser einigen kurzen Dornen nahe der Basis tind dicht vor dem Hinterrande noch je ungefahr dreizehn lange Dornen, wihrend der Seitenrand mit acht oder neun langen schwarzen Dornen besetzt ist. Am Hinterrande befinden sich je vier oder “fiinf kiirzere rétlich-gelbe Dornen, von denen ein grésserer mit einem kleinen abwechselt. / - Dactylispa secura sp. nov. Nigra, elytris metallico-splendentibus, antennis longis, graci- libus, apice rufescentibus, prothorace creberrime punctato opaco,. ante basin late transversim impresso, lateribus medio dilatato et bispinoso, elytris sat fortiter striato-punctatis, humeris bi-, dorso decem-lateribusque sex- vel septem-spinosis. — Long., 4 ad 5 mm. Madagaskar, Mahatsinjo bei Tananarive (Donckier). 80 The Philippine Journal of Science 1922 Von ahniicher Gestalt als D. dichroa, aber kleiner und durch die nur zweidornigen Halsschildseiten ausgezeichnet. Schwarz; Stirn, Thorax, und Schildchen matt, die Fliigeldecken glanzend, mit einem blassen Messingschimmer tiberzogen, die fiinf letzten Fiihlerglieder rotbraun. Stirn zum spiegelblanken Halse plotz- lich abfallend. Fiihler schlank, langer wie der halbe Ko6rper; Glied 1 und 3 lang, 2 kurz, 4 und 5 etwas kiirzer als 3, 6 und 7 jedes wenig kiirzer als 4, das siebente nebst den folgenden Gliedern verdickt, 8 bis 10 kurz, jedes so lang als 2, das End- glied nur wenig langer. Thorax stark quer, die Seiten nach der Mitte hin winkelig erweitert; diese Erweiterung lauft in einen kurzen Stiel aus auf dem zwei massig lange, wenig schriag, nach aussen gerichtete Dornen stehen. Unbedeutend langer ist der hintere, senkrechte Vorderranddorn, der vordere, der sich in sehr spitzem Winkel trennt, ist etwa halb solang. Die Scheibe ist sehr dicht und etwas narbig punktirt und hat zwei Quervertief- ungen, eine schwache und schmale hinter den Vorderranddornen und eine starkere vor dem Hinterrande; zwischen beiden liegt ein undeutlicher Querwulst, nach innen zu einer glatten Mittel- linie abfallend. Die Fliigeldecken sind stark gereiht-punktirt, mit schlanken Dornen, von denen die mittleren viel langer als die an der Basis und Spitze stehenden sind. Ausser einigen Basaldérnchen befinden sich auf der Schulterbeule zwei Dornen, der erste kurz, der zweite lang; auf dem ersten primiren Zwi- schenstreifen fiinf (die drei mittleren lang), auf dem zweiten einer, hinter der Mitte, auf dem dritten vier Dornen, wihrend die Seiten mit sechs oder sieben Dornen bewehrt sind. Von diesen sind der letzte oder die beiden letzten am Hinterrande sehr kurz. Dactylispa aeneipennis Duviv. : Schwarz, Fliigeldecken mit Bronceschimmer; ist dadurch ausgezeichnet dass die wenigen Dornen der Fliigeldecken nur auf den dusseren zwei Drittel stehen, nimlich drei auf der Schulter (der erste kurz, der zweite missig hoch, der dritte lang), und vier noch etwas lingere (1, 1, 2) dahinter: der erste dicht hinter ein Viertel der Linge, der zweite dicht hinter der Mitte, beide zwischen der vierten und siebenten Punktreihe, die andern zwei vor der Spitze, einer an der hinteren Aussenecke, der andere zwischen der zweiten und vierten Punktreihe. Am Seitenrande befinden sich zwei lange Dornen, genau wie die beiden ersten Scheibendornen ‘gestellt, am Hinterrande zwei Dérnchen. Madagaskar, Mahatsinjo (Donckier). 21,1 Weise: Hispinen der Alten Welt 81 Dactylispa excisa Kraatz var. repanda var. nov. Von Kiautschau erhielt ich durch Staudinger eine Dactylispa, die eine Variet’t von D. excisa bilden diirfte, auf welche ich durch den Namen repanda aufmerksam mache. Sie ist nur 4 Millimeter lang, schwarz, Fiihler, Beine, und Bauch gelb, die Dornen des Halsschildes und drei Dornen im Ausschnitte der Fligeldecken etwas dunkler, rétlich gelb, die Scheibe der Fliigel- decken mit Ausnahme der darauf stehenden Dornen glinzend rétlich pechbraun. Der vordere Dorn am Vorderrande des Thorax hat an der Basis ein dornférmiges Anhingsel, das kaum ein Drittel so lang ist wie der Dorn selbst, wihrend der Seiten- lappen nicht fiinf, sondern nur vier Dornen tragt: drei langere, die auf einer gemeinsamen breiten Basis dicht hinter einander entspringen und einen weiter abgeriickten, kiirzeren dahinter. Dactylispa adstricta sp. nov. Piceo-nigra, opaca, antennis, pedibus ventreque rufo-flavis, prothorace creberrime subvarioloso-punctato et parce brevissi- meque griseo-piloso, plagis tribus elevatis, glabris, alutaceis transversim positis, antice utrinque spinis duabus crassis ver- ticalibus, longitudinaliter dispositis et basi alte connexis, later- ibus spinis tribus rufo-flavis (1, 1, 1) armatis, elytris acute tuberculatis, crista humerali et lateribus (his paullo ante medium coarctatis) breviter spinulosis. Long., 5 mm. Japan, Harima (Staudinger). Vor der Mitte stimmt diese Art im Kérperbaue villig mit der vorigen iiberein, hinter der Einschniirung aber erweitern sich die Fliigeldecken nur ganz allmahlich nach hinten; ausser- dem sind die Fiihler kiirzer und dicker, und die ganze Oberseite ist matt pechschwarz, zart gewirkt oder gerunzelt, matt; Fiihler, Beine, und Bauch lebhaft rétlich gelb, die mittleren Fiihler- glieder 3 bis 6 mehr gelb, die drei Seitendornen des Halsschildes nebst dem langen und schmalen Lappen auf dem sie hinter einander stehen dunkler rétlich gelb. Fiihler kurz, bis zur Schulter reichend; die fiinf Endglieder bilden eine sehr schlanke Keule, an der die Glieder 8 bis 10 breiter wie lang sind. Stirn gerunzelt, ein vertiefter Streifen neben jedem Auge weisslich behaart. Halsschild um die Hilfte breiter als lang, dicht punktirt, in der Mitte mit einer Querreihe von drei gerundeten, erhéhten und nur zart gewirkten Flecken. Die beiden Dornen jederseits am Vorderrande sind schwarz, kurz, breit, senkrecht 187663——_6 82 The Philippine Journal of Science 1922 hinter einander auf einer hohen Langsleiste eingefiigt, der erste hat in der Mitte des Vorderrandes eine kleine zahnformige Erweiterung und ist etwas niedriger als der zweite; die drei Seitendornen sind etwas schlanker aber kaum linger, schrig nach aussen gerichtet, und stehen ebenfalls auf einer gemein- samen Langsleiste. Die Fliigeldecken treten an der Basis grad- linig heraus, sind in den Schultern bedeutend breiter wie der Thorax und erweitern sich dahinter bogenférmig in einen Lappen, der einem Kreissegmente dhnelt und kurz vor der Mitte endet, von hier aus erweitern sie sich allmahlich nach hinten und sind am Ende breit und fast gemeinschaftlich abgerundet. Die Seiten sind kurz, vor der Mitte dicht, dahinter weitlauf- iger, und an der Spitze dicht und sehr kurz-gezihnt, die Schulter bildet eine Leiste, die mit fiinf oder sechs Zahnchen bewehrt ist. Die Scheibe hat eine Anzahl kiirzer und dicker zahnférmiger Erhebungen. Von diesen liegen vier auf der ersten, nur angedeuteten primdren Rippe, nadhmlich ein sehr kleiner Zahn nahe dem Schildchen, ein grosser vor und hinter der Mitte, und ein kleiner vor der Spitze. Die zweite Rippe hat zwei gréssere Zahne, in und hinter der Mitte; neben dem letzteren steht ein Ahnlicher Zahn auf der dritten Rippe, welche vorn noch einige unbedeutende Erhebungen hat. Dactylispa aspera ab. matangensis ab. nov. Der Thorax ist etwas stirker punktirt als bei der Stammform von Sumatra (die Punkte anndhernd narbig, von einem erhdhten Rande umgeben) und die Fliigeldecken sind einfarbig schwarz, vorn nicht rostrot. Ausser dem vorderen Dorne am Vorder- rande des Halsschildes kann auch der hintere ein dornférmiges Anhangsel haben, oder beide sind einfach. Am Seitenrande stehen auf einem gemeinschaftlichen Stiele vier Dornen, die drei ersten unter sich gleich lang, der vierte viel kiirzer und schrig nach hinten gerichtet. Sarawak, Matang und Oya, August, 1910. Von Dactylispa longicuspis Gestro befindet sich ein Stiick im Sarawak Museum (Mount Poi, 3000 bis 3500 Fuss, April 17, 1913), mit einfarbig schwarzen Fliigeldecken. Die Art ist durch die auffallig langen, mit variablen Anhangseln versehenen drei Thoraxdornen kenntlich, von denen die beiden ersten aus einem sehr langen gemeinsamen Stiele entspringen, an dessen Basis der kurze und einfache dritte Dorn befestigt ist. 21,1 Weise: Hispinen der Alten Welt 83 Dactylispa fulvicornis sp. nov. Fulva, elytris nigris, sat nitidis, antennis elongatis, articulo tertio longissimo, prothorace crebre punctato et breviter albo- piloso, vitta media abbreviata sublaevi, lateribus bispinosis, Spinis basi conjunctis, antica appendiculata, postica multo bre- viore, elytris spinis numerosis sat longis armatis. Long., 5 mm. Borneo, Sarawak, Lawas, August, 1909. An der Bewehrung des Halsschildes kenntlich. Der hintere senkrechte Vorderranddorn ist doppelt so lang wie der vordere; an den Seiten stehen nur zwei Dornen: der erste ist fast eben so lang als der zweite Vorderranddorn, aber etwas starker und mit mehreren Dérnchen besetzt, an seiner Basis entspringt der zweite, kurze, einfache, und schrig nach hinten gerichtete Dorn. Der K6rper ist ziemlich gestreckt, rétlich gelb oder hell rostrot, die Fliigeldecken schwarz, ihr abgesetzter Seitenrand an der Spitze etwas verbreitert, und nebst den dort befindlichen kurzen Dornen rétlich gefirbt. Fiihler schlank, etwas langer wie der halbe K6rper, ihr drittes Glied so lang wie die beiden ersten zusammen. Thorax dicht punktirt und zart weisslich behaart, matt, ein gewdélbtes, beiderseits abgekiirztes Querband in der Mitte fast glatt und glinzend. Fliigeldecken mit zahlreichen schlanken Dornen besetzt, von denen der zweite in der ersten Reihe der langste ist, an den Seiten wechselt ein langer und ein kurzer Dorn ab, von zwei Drittel der Linge bis zur Nahtecke werden die Dornen allmihlich kiirzer. Dactylispa rufiventris Kraatz. Eine Reihe von Exemplaren vom Kamerunberge, Soppo, 730 m. (v. Rothkirch 1912) in den Vorriiten Staudingers hat den Bauch einfarbig tief schwarz. Sonach trifft die Bemerkung von Kraatz,* “die Art ist an ihrer Grésse und dem roten Hin- terleibe sehr leicht zu erkennen,” nicht immer zu, und ich mache auf diese abweichende Form durch den Namen D. nigriventris aufmerksam. Die Art zeichnet sich durch das Halsschild aus. Die drei Seitendornen desselben, von denen der mittelste der langste, der dritte der kiirzeste ist, sind von einander deutlich getrennt oder an der Basis nur genidhert, und die glinzende Scheibe ist sparsam punktirt. Die Punkte sind gross und flach, narbig, stehen dichter in dem weiten, an den Seiten nach vorn *Deutsche ent. Zeitschr. (1895) 197. 84 The Philippine Journal of Science 1922 gebogenen Quereindrucke vor der Basis, wahrend sie in dem vorderen Quereindrucke eine oder zwei unregelmdssige Langs- reihen bilden. Die Linge der Art schwankt zwischen 5 und 7 Millimeter; Gyllenhal* hat sie, als var. 8 seiner Hispa spinu- losa betrachtet. Hispa comata sp. nov. Testacea, opaca, supra griseo-pubescens, prothorace ramulo elongato duodecim spinoso lateraliter armato, elytris striato- punctatis dense spinulatis, in disco spinulis, crassioribus sub- infuscatis et gracilibus dilutioribus alternantibus. Long., 3 mm. Syria, Aleppo (Staudinger). Kiner kleinen H. pavida dhnlich. Réotlich gelbbraun, matt, auf dem Halsschilde sehr dicht, auf den Fliigeldecken sparsamer weisslich behaart. Fiihler ziemlich kurz, schwarzlich, Glied 2 kaum schwicher aber nur halb so lang as 1, 3 diinner und unbedeutend kiirzer als 2, 4 und 5 kiirzer als 3, 6 das kleinste, die folgenden Glieder verdickt, 7 so lang als 2, 8 bis 10 jedes kiirzer. Der Thorax erweitert sich an den Seiten in einen langen und schmalen Lappen, welcher mit zwoélf langen, feinen Dornen besetzt ist. Von diesen steht einer in der Mitte an der Basis und ist schrag nach vorn und innen gerichtet, die tibrigen bekranzen die Seiten und liegen fast horizontal; die beiden letzten sind bedeutend kiirzer wie die vorhergehenden. Wenn der letzte gespalten ist, was zuweilen und dann meist nur einseitig vorkommt, kann man dreizehn Dornen zahlen. Die Punktirung des Halsschildes wird von der dichten Behaar- ung verdeckt. Die Fliigeldecken sind gereiht-punktirt und ziemlich dicht bedornt. Die Dornen des Randes sind schlank, am Hinterrande verkiirzt, die der Scheibe bestehen aus stdr- keren, dunkler gefarbten, und etwas kiirzeren, viel diinneren hellen Dornen. Platypria moluccana sp. nov. Ferrugineo-flava, elytris glabris, nitidulis, nigris (basi ex- cepta), margine bilobatis, lobo antico sexspinoso, postico tri- spinoso. Lonz., 5 mm, Batchian, August, 1906 (Koller). In der Korperform, Grésse, und Farbe am meisten mit P. dimidiata tibereinstimmend, aber der Thoraxlappen bedeutend *Schénh. Syn. Ins. App. (1817) 8. 21,1 Weise: Hispinen der Alten Welt 85 breiter, kiirzer bedornt, und der vordere Seitenlappen der Fli- geldecken mit sechs, der hintere mit drei Dornen besetzt. Durch diese Bewehrung tritt die Art naher an P. echidna und P. squa- lida heran, weicht jedoch von ersterer durch das einfarbige Halsschild und die dunkle Fliigeldecken, von P. squalida durch unbehaarte und massig glinzende Fliigeldecken ab. Koérper rotlich gelb, die Fliigeldecken schwarz, eine Basal- binde der letzteren, ein feiner Saum zwischen den Seitenlappen und in der Spitze nebst dem matten, stellenweise dicht und sehr fein punktirten Halsschilde blass rostrot. Die Basalbinde reicht bis an die beiden ersten Scheibendornen und an den Anfang des Seitenlappens, von dem der erste Dorn und die innere sehr grosse Pore ebenfalls hell gefairbt ist, verlangert sich aber innen saumférmig auf der Naht bis an den zweiten Dorn der ersten primiren Rippe. Die Fiihler sind schlank, normal gebaut. Der Seitenlappen des Thorax ist halboval, schwach glanzend, mit fiinf Dornen besetzt, von denen die ersten vier ziemlich lang sind, waihrend der fiinfte klein und kaum langer aber stirker als der vorn befestigte Borstenkegel ist. Die Fliigeldecken besitzen am Basalrande zwei Dérnchen, auf der Schulter drei ziemlich kurze, auf der Scheibe sechs dicke und lange Dornen (2, 2, 1, 1), wovon die ersten beiden und der letzte Dorn (neben der hinteren Aussenecke) kiirzer wie die iibrigen sind. MERRILLOSPHAERA, A NEW GENUS OF THE VOLVOCACEAE By WALTER R. SHAW Of the Department of Botany, College of Liberal Arts, University of the Philippines, Manila EIGHT PLATES AND ONE TEXT FIGURE CONTENTS INTRODUCTION. MERRILLOSPHAERA CARTERI VAR, TYPICA MERRILLOSPHAERA CARTERI (STEIN) VAR. NOV. COMB. NOV. VAR, MANILANA VAR. MERRILLOSPHAERA CARTERI VAR. WEIS- NOV. MANNIA (POWERS) COMB. NOV. Description of type specimen. MERRILLOSPHAERA MIGULAE SP. NOV. Asexual cotype specimens. MERRILLOSPHAERA TERTIA (MEYER) Female coenobia. COMB. NOV. Male coenobia. MERRILLOSPHAERA AFRICANA (WEST) Forms of asexual coenobia. COMB. NOV, A smaller form of the species. DIAGNOSES OF GENUS, SPECIES, AND Cell membranes. VARIETIES, Mixed reproductive bodies. INTRODUCTION In 1859, Carter described what he supposed to be Volvox globator Ehrenberg, which he found in India, presumably in the neighborhood of Bombay. From Carter’s description, Stein (78, p. 184) recognized the plant as specifically distinct from all that had been described, and he named it Volvox carteri. Stein improperly regarded it as intermediate between the spe- cies V. globator and V. aureus of Ehrenberg, whereas in reality it is neither intermediate between those species nor even so closely related to either of them as they are to each other. One of the several members of the subfamily Volvoceae which I have collected at Manila so nearly fits Carter’s description that I shall treat it here as a mere variety. Yet this species differs so much from the earlier described species of Volvox that I shall assign it, with others, to a new genus. Carter noted the absence of protoplasmic connecting filaments between the somatic or vegetative protoplasts. There had long been in my mind a question as to the propriety of including in the genus 87 88 The Philippine Journal of Science 1922 Volvox organisms so different from V. globator and V. aureus as to lack the intercellular protoplasmic connecting filaments. The question was answered for me by the early differentiation and large size of the asexual reproductive cells of this and of a related species. Carter described the asexual reproductive cells, the gonidia, as reaching a large size before undergoing segmentation. His figure of a gonidium (’59, pl. 1, fig. 4), which is reproduced on a larger scale in Plate 7, fig. 44, shows one that had reached a diameter of 85 ». However, he did not appreciate the fact that such large gonidia, prominent in the daughter coenobia long before birth (as shown in his fig. 1, reproduced in Plate 7, fig. 42), are differentiated from the so- matogenic cells (those which by division produce the somatic or vegetative cells) at a very early stage in the development of the embryo and remain large, while the somatogenic cells be- come smaller and smaller by the repeated divisions which produce the very numerous somatic cells. Carter’s account of the asexual reproduction seems to have been based on observations which failed to include the stages that commonly occur at night. Until I had made a few obser- vations on living material through the hours of the night, my own work was in a fair way to muddle up in a description of one “species” forms that I now consider generically dis- tinct. One of these forms I have already described under the name Campbellosphaera obversa (Shaw, ’19). Its most dis- tinctive character is a migration of the gonidia during the bowl stage of the embryo from the outside to the inside through the opening, or phialopore, just before the closure which converts the bowl-shaped embryo into a closed globular colony, or coenobium. The absence of such a migration is the important character that was revealed by the nocturnal study of the species which forms the subject of this paper. Taking as a basis the characters found in the material in hand, I propose a new genus, Merrillosphaera, dedicated to Elmer Drew Merrill, the most indefatigable investigator of the flora of the Philippine Islands and the Indo-Malayan region. Believing this material cospecific with Volvox carteri Stein, I shall call it Merrillosphaera carteri (Stein) comb. nov. with the designation mani- lana for the form or variety first to be described in sufficiently full detail to furnish a basis for the recognition of the proper status of the genus. A very similar form was described by Powers (08) under the name Volvox weismannia. It is both proper and convenient to reduce this to Merrillosphaera carteri var. weis- 21,1 Shaw: Merrillosphaera 89 mannia (Powers) comb. nov. Less fully described is 2 form col- lected and prepared by Migula, and called by Klein (’89B) Volvox aureus. This evidently is not the Volvox aureus Ehren- berg described by Klein (’89A and ’90) but is a new species similar to M. carteri which I call Merrillosphaera migulae sp. nov. Of another species of Volvoz, one described by West (’10 and 18) under the name V. africanus, my Philippine material con- tains stages which fill out the life history and show its affinity with the species and varieties listed in the foregoing paragraph. It therefore becomes Merrillosphaera africana (West) comb. nov. (Volvox africanus West). A third species of Volvoxz, described by Meyer (’96) under the name JV. tertius, seems to fit into the new genus more readily than into any other, and this may be sufficient reason for clas- sifying it tentatively as Merrillosphaera tertia (Meyer) comb. nov. (Volvox tertius Meyer). Of this species we have as yet only Meyer’s description, which is in many respects very meager. All of these species are characterized by having rounded so- matic protoplasts without connecting filaments, and also by having gonidia which are differentiated at an early embryonic stage. To this extent they agree with Campbellosphaera, but they differ from the latter in lacking the migration of the gonidia through the phialopore into the coenobium. Two other species that lack interprotoplastic filaments develop their gonidia only after birth, and these I had previously assigned to two genera. Those with a reproductive area in which a small number of the cells develop into gonidia were put in the genus Besseyosphaera (Shaw, ’16), while those with a reproductive area in which all of the cells develop into gonidia were put in the genus Pleodorina (Shaw, 94). In presenting the descriptive details pertaining to the species and varieties of Merrillosphaera I shall begin with the descrip- tions of a series of specimens, selected primarily from those which have been photographed and that are supposed to represent the life history of M. carteri var. manilana, of which the first will constitute the type of the variety. Then the nearly related varieties of the literature will be taken up separately and described with the same terminology, for comparison with the variety in hand. In another paper I purpose to extend our knowledge of Mer- rilosphaera africana (West) Shaw by means of descriptions of specimens in my collections. 90 The Philippine Journal of Science 1922 Two species of Volvox proper that occur in my Philippine material more or less mixed with species of the related genera I have described in another paper (Shaw, ’22) in which a special genus was made for the European species long known as V. aureus Ehrenberg. MERRILLOSPHAERA CARTERI VAR. MANILANA VAR, NOV. DESCRIPTION OF THE TYPE SPECIMEN For the type of this variety, an asexual specimen (Plate 1, figs. 1 and 2) has been selected in order to exhibit a character which is essential to distinguish the genus from Campbello- sphaera. This character appears in embryos that are in a stage which makes it fairly plain that the gonidia are differentiated early, but do not migrate. The specimen (No. 1) is one of a batch collected from Pond C in Pasay, near the southern outskirts of Manila, September 20, 1914, and fixed at some time on the next day in picro- nigrosin. Material from this batch was mounted in glycerine on several slides and sealed with shellac. Slide 4 of this lot bears the specimen under a cover glass with glass rodlets measuring 309 to 354 yw, which were distributed for cover supports but are loose and now clustered at one side of the circle. The gly- cerine is strongly colored with picric acid. There is no record of the time when the mounts were made, but the photographs were taken June 13, 1916. So they were at the time of making measurements (July, 1919) more than three years old. The mount had then taken up no air bubble,' but the paper label of the slide showed glycerine on the edge nearest the cover, which raised a suspicion as to the security of the sealing. The slide bears a mixture of several species of Volvoceae. There are nu- merous hyphae extending into and through the glycerine; they appear to have developed in it. The type specimen measures 545 by 590 ». The upper and lower sides are seen with a difference of focal adjustment amounting to about 250 », which, with an allowance of 1.4 for the optical density of the glycerine, gives an estimated thickness for the specimen of about 350 ». The area of the somatic surface that is flattened by the cover glass is about 350 by 400 » in extent. Within this area the somatic protoplasts of the *The specimen has since been ruined, partly by escape of glycerine and formation of a large bubble, and partly by destructive movement of glass rodlets. 21,1 Shaw: Merrillosphaera 9] somatic wall are somewhat crowded and average about 9.5 bs between centers. Just outside of the flattened area the distance between cells is about 10 ». This, with an approximate mean diameter taken as 560 », gives 11,400 as the number of ceils. The protoplasts are somewhat farther apart in the anterior polar region. The somatic protoplasts are globose or oval, without con- necting filaments. They are about 5.4 » thick. Their distance from the outer membranous surface of the soma is about half a protoplastidar diameter or less. The outer surface appears quite even. The middle lamella between the protoplasts cannot be seen, nor is the inner limit of the cell membranes visible. Eight embryos are contained in the coenobium,? occupying positions in the coenobial cavity near the coenobium wall, four just ahead of the equatorial plane and the four others alternat- ing with them just back of that plane. The embryos are very darkly stained. They have developed into many-celled coenobia, in the walls of which gonidia may be seen embedded and protrud- ing. Six of the embryos can be seen to contain small numbers of reproductive cells. Two of them are not so clear. The one which is shown in Plate 1, fig. 2, is about 90 » in diameter; it has somatogenic cells of about 2.5 » width, and contains four cells of about 8 » diameter (which are not visible in the figure), grouped closely about what I take to be the site of the phialopore. It also contains eight cells about 17 » in diameter, obviously gonidia, distributed in the coenobium wall—five near the me- dian plane (two of which stand out clearly in the figure), one on the nearer side, and two on the farther side. A well-marked feature of the embryos is that their gonidia are embedded in the layer of somatic cells where they are formed. Being more cr less nearly isodiametric, and larger than the cells of the somatic layer, they protrude beyond the surface of the coenobium. ASEXUAL COTYPE SPECIMENS On the type slide are four other specimens of the same species, containing embryos in the same stage as those in the type speci- men, or slightly younger. With them are twelve asexual coenobia with gonidia and two sexual coenobia with young * The use of the term “coenobium” for the bodies of these organisms is not consistent with my view that they are properly to be regarded as individuals rather than as colonies. 92 The Philippine Journal of Science 1922 oogonia. I will describe somie of these specimens in order to bring out points that are not shown by the type specimen. Specimen 2.—This coenobium contains eight embryos and pre- sents an almost polar anterior view. Diameter, about 505 as thickness, about 365 ,; protoplasts, about 5 pa. Spacing of centers of cells: Anterior, about 20 »; posterior, about 12 »; and mean, about 15 ». Estimate of somatic cells, 4,100. In the me- dian focus there is seen a delicate line nearly parallel with the periphery of the optical section of the coenobium, forming an ellipse 280 by 300 » in diameter, lying 110 to 120 p within the outer membrane. The eight embryos lie in the zone between this line and the layer of somatic cells. These embryos are about 90 » in diameter, and those that show clearly near the median level and above it have their phialopores directed outward. Around each embryo there are two distinct lines ; these are the inner and outer limits of an embryo capsule which was formed by the wall of the gonidium from which the embryo developed. The embryo capsule touches the embryo on the side toward the center of the mother coenobium, is about 10 to 15 » from the surface of the embryo around the equatorial plane of the embryo, and approaches without touching the em- bryo on the outer side. Around the equatorial region of each embryo its capsule is about 30 p» thick, while toward the poles it thins down until at the poles it becomes very thin, as if under radial pressure between a central coenobial core and the somatic layer of the mother coenobium. All the embryos show a few protruding gonidia, and these are most evident in the profile views of the embryos. The somatic cells are very much smaller than the gonidia. The embryonic phialopores are all more or less open and have the form of a four-pointed star. This is produced by the growth about the phialopores of four marginal _lobes which for a time do not fit together, but which leave the star-shaped opening between them. The extent of the star is about one quarter of the circumference of the embryo. In some of the embryos the lobes are Slightly raised and bent outward and bear on their tips one or more larger cells that are intermediate in size between the gonidia and the somatic cells. The embryos are so far apart that their capsules are not in contact with one another. Specimen 3.—This coenobium contains eight embryos. It pre- sents a side view, with the anterior pole a little lower than the posterior pole. Dimensions, 560 by 600 wu. Spacing of cells: 21,1 Shaw: Merrillosphaera 93 Anterior, 20 »; equatorial, 13 »; posterior, 15 #3; mean, about 15 ». Estimated number of cells, 5,300. Somatic protoplasts, about 5to6,. A thick middle lamella of the somatic cells shows in a surface view, even under the low power. The embryos, about 90 p, are closed or closing. The phialopores are directed outward, and a side view of one embryo shows the lobes around the stellate opening bent outward and tipped with large cells. The inner ellipsoidal membrane, or core boundary, of the coeno- bium is 380 by 358 y», the zone between it and the outer limit of the coenobium being 115 » in front and 130 » behind. The inner lamella of the embryo capsule delimits the space about each embryo, but the outer lamella does not show. Specimen 4.—An asexual coenobium with eight embryos, 565 by 593 ». Spacing! of cells: Anterior, 20 »; equatorial, 15 p»; posterior, 13 »; estimated mean, 15». Number, 5,300. Embryos 85 to 95 y», closing and closed, with gonidia protruding beyond the surface. Lobes about the phialopore bent outward in one embryo and bearing larger cells. Only the inner lamella of the embryo capsules is visible. Specimen 5.—An asexual coenobium with six embryos and four gonidia. Oblique view. 562 by 590 ». Spacing of cells, 10 to 18 (mean 11) pp. Number, 9,800. Embryos about 80 » each, with protruding gonidia. Two of the gonidia of the parent, each about 64 » in diameter, form a posterior pair; while the two others, with diameters of 68 and 70 p», respectively, form a subposterior pair. FEMALE COENOBIA An asexual coenobium (specimen 6) with eight female daughters, which are distributed like the embryos in the type specimen, is shown in Plate 1, fig. 3. This specimen is of the same collection as the type specimen, but of a batch fixed one day later; that is, two days after collection. It is in a picro- nigrosin preparation mounted in glycerine; it was photographed in 1916, and was still in good condition in 1919. The coenobium measures 590 by 643 ». The thickness of the specimen measured about 250 ». Near by is a glass rodlet 265 » thick, which was used to support the cover glass. The somatic protoplasts are about 5.3 to 5.9 » wide. Their intercentral distances respectively are approximately these: Anterior, 21 »; equatorial, 14 »; pos- terior, 17 »; and mean, considering the relative extent of the different areas, 16 ». This gives the estimate of the number of cells, 5,250. The daughters all contain reproductive cells, which are more numerous and smaller than gonidia would be at 94 The Philippine Journal of Science 1922 the same stage. They are the female reproductive cells. One daughter, the second from the left in the upper row, measures 178 by 180 ». The somatic protoplast diameters are about 3.6 pw; the intercentral distances, about 4.9 »; and the estimated number of cells, 4,850. The oogonia are about 15 » in diameter, number about 24, and are distributed just under the somatic layer in all except the anterior quarter of the daughter, this quarter being at a higher level than the picture plane. Over each oogonium there is a vacancy in the layer of somatic cells, this space being sur- rounded by seven or eight somatic cells. Each of the other daughters shows an area lacking reproductive cells, and all the daughters have nearly the same dimensions and about the same number of somatic cells and reproductive cells. The smallest is 145 by 150 », the largest 175 by 190 ». The oogonia were counted and measured in three other daughters which are above the median plane. The counts are 27, 37, and 31, and the measure- ments 13, 15 (exceptionally 11 and 16), and 14 y, respectively. The smallest oogonia are a few of those in the coenobium that contains the largest number. The oogonia are vacuolate proto- plasts, each with a central nucleus and with parietal chromato- phores. No membranes can be seen except the outer limiting membranes of the mother and of the daughters. The position of the inner lamella of the embryo capsules seems to be marked by confinement of the cilia of the daughters to a space having a width that is much less than the length of the cilia. A nearly mature female coenobium (specimen 7) from slide 1 of the type lot is shown in Plate 1, fig. 4. This specimen measures 264 by 380 », and has a thickness, in its compressed condition under the cover glass, of about 186 ». The somatic protoplasts are 4.4 ‘to 5.3 » wide, and the distances between their centers are, approximately: Anterior, 12.5 p; equatorial, 10 »; posterior, 9 »; and mean, about 10.2 ». The estimated number of somatic cells is 4,850. Distributed in about three quarters of the coenobium, and located about 25 yu below the outer surface, there are twenty-nine reproductive cells, of which five have thickened walls which mark them as oospores. These oogonia are mostly about 27 » in diameter, two exceptional ones near the posterior pole measuring 22 and 25 n, respectively. Over each reproductive cell there is a vacancy in the layer of somatic cells, each such space being surrounded by about six somatic cells. In the reproductive region, and anterior to the equator, there are discernible five vacancies in the somatic layer without corresponding reproductive cells. These may be the t 21,1 Shaw: Merrillosphaera 95 sites of antheridia, but I do not think it likely. The oospore walls are not very strongly developed. They measure about 32 » across, except one containing a protoplast which appears to be pathologically plasmolized into a bowl shape, and this wall measures 35.6 pw. These walls are smooth, and concentric with the protoplast, except one which is eccentric. A more-mature female coenobium, one containing nearly ripe oospores, is shown in Plate 4, fig. 24. This specimen (No. 8) is in a collection taken from Pond I, less than a hundred yards from Pond C, in Pasay, October 5, 1914. It is on slide 1 of a lot of picro-nigrosin glycerine preparations made from the same collection. The cover glass is supported only by a shellac ring and has flattened the large coenobia, reducing their thickness to about 112 ». The cleft in the forward pole is a result of the cover-glass pressure. Measurements of coenobia in such a flat- tened condition are not very satisfactory, and those made of this one in 1921 do not accord with the dimensions of the photograph. The specimen now measures about 693 » wide and ' 737 » long. The average spacing of the protoplasts is about 13.3 », and the number of cells is estimated to be between 6,000 and 8,000. The protoplasts are about 6 by 7 » in diameter. There are thirty-six mature oospores, and there are two un- matured reproductive cells with them. The two dark bodies in the anterior circumpolar region are a green endophyte like a Chlorosphaera. The oospore walls shown in Plate 4, fig. 29, have a very slightly wavy appearance that is more pronounced in direct view of some spores. They measure in outside diameter about 50 », appear to be about 2.5 to 3 » thick, and contain protoplasts that are mostly about 39 » thick and in contact with the wall at one side and therefore eccentric. MALE COENOBIA A mature asexual coenobium containing male and asexual progeny is shown in Plate 1, fig. 5. This specimen (No. 9) is on the same slide as the female coenobium described in the preceding paragraph. This mother coenobium is turned away from the observer approximately 22°. On the farther side of the coenobium there is in the somatic layer a cleft about large enough for the birth of an offspring. If one of the progeny passed through that cleft it was the member of the anterior quartet located on the farther side. There remain in the an- terior quartet an asexual and two male coenobia, and the pos- 96 The Philippine Journal of Science 1922 terior quartet consists of three asexual and one male coenobia. Within the mother coenobium, near the cleft, there is an anomalous object consisting of an irregularly shaped and bent sheet of hundreds of small cells. It is like a piece torn out of the somatic layer of one of the asexual offspring, but the cells are much smaller. The mother coenobium in its flattened condition measured, in 1919, about 678 by 685 ». The corre- sponding dimensions of the photomicrograph are 650 and 660 pn, which indicates that three years of drying have brought the cover glass closer to the slide and squeezed the large specimens flatter, thus making them wider. The specimen measured about 112 » in thickness. The somatic protoplasts are about 6 to 7.1 » thick. The outer walls of the somatic cells are slightly convex, especially around the anterior pole, giving the surface of the coenobium a wavy appearance in optical section. The spacing of the somatic cells is about as follows: Forward, 22 »; in the middle, 17 »; at the back, 18 »; average, about 17.5. The consequent estimate of somatic cells is 5,480.2 Each asexual daughter contains eight gonidia, and over each gonidium there is a vacant place in the layer of somatic cells. The daughter nearest the middle of the picture is 210 » wide and 230 » long. Its somatic protoplasts are about 3.6 p» wide, and their centers are about 4.1 » apart. The number of cells in- dicated by these figures is 10,100. The gonidia are about 30 pb in diameter. There are pressure creases in the coenobium walls of two of the daughters. Of the male offspring, the lower is 240 by 280 y, the right one 240 by 290 p, and the upper one 260 by 270 ». The last of these has somatic protoplasts about 5.3 to 5.4 » thick and spaced 10 » between centers. The estimate of somatic cells is 2,500.2 Seattered throughout the periphery of the coenobial cavity are 112 or more androgonidia, the cells which produce, by dividing, the bundles or platelets of sperms, mostly with one or two of the segmentation divisions accom- plished. A few undivided ones are 10, 14, and 16 p wide. Those in the 2-celled stage are mostly about 17 by 19 », and those in the 4-celled stage about 19 » each way. The counting of the androgonidia was done by making a camera lucida sketch, in which difficulty was experienced due to overlapping of the progeny in their flattened condition. The number is evidently about the same in all, and is in the neighborhood of 120. *The estimate is probably much too high on account of the flattening of the coenobium. 21,1 Shaw: Merrillosphaera 97 A free male coenobium (specimen 10) in the same collection as the type specimen is shown in Plate 1, fig. 7. After it was photo- graphed the specimen became crushed between two glass rodlets so that the antheridia cannot be counted. The somatic proto- plasts are about 4.5 to 5 » wide and spaced about 12 p. The diameter of the coenobium being about 253 ,», the number of the somatic cells is about 1,600. The reproductive bodies are all sperm platelets that are more or less cupped and measure 14 to 19 » in diameter. They are too darkly stained to admit of ' counting the sperms. The specimen shows plainly the cilia of the somatic cells and of the sperms. Each sperm platelet is contained in a cavity about 46 » in diameter. Two free male coenobia that appear to belong to the same species are in company with the specimen shown in Plate 1, fig. 5. They are very similar to the males within this mother. These two specimens (Nos. 11 and 12) are 295 by 316 » and 316 by 324 » in diameter and have about 1,700 and 1,800 cells. The Somatic protoplasts are about 7 » in diameter and measure about 23 » wide. The antheridia number about 200 in each coenobium and are distributed without leaving a well-marked vegetative pole. They are mostly in the 16-celled stage of division, though some have divided only once or twice. On the same slide with them there is a larger male coenobium that I take to be of another species. It is larger, 390 by 418 », and has smaller cells, 6 », that are more numerous, about 2,900. The antheridia also are smaller, 18 », and more numerous, about 640. This coenobium has no vegetative pole marked by absence of reproductive cells, but the posterior half is narrowed slightly as is commonly the case in Merrillosphaera africana. FORMS OF ASEXUAL COENOBIA The photomicrographs used for illustrating this paper have been selected from about one hundred fifty that were made to serve as an aid in the task of separating the five or six species of larger Volvocaceae that occurred in various mixtures in the collections made in the neighborhood of Manila. This was done at a time when the number of species and genera was not known. The selection of material was based largely on suitability of the specimens for photography. A result of thus working simulta- neously with a number of species is that no one species was followed as far through its range of variation as might have been done had it been studied alone. Since the photographed 187663-——7 98 The Philippine Journal of Science 1922 preparations were made, large quantities of material have been collected in different years. The form selected to serve as the type of Merrillosphaera car- teri var. manilana has most commonly eight gonidia, of which four form a quartet of equidistant gonidia lying in a transverse plane a little in advance of the equatorial plane and the other four ‘form a similar quartet just back of the equatorial plane, the members of the two quartets being located on alternating longitudinal radial planes. In such cases the reproductive bodies occupy a middle zone of the coenobium and leave two polar zones without reproductive bodies. In other cases there is no empty space back of the hinder quartet. A third quartet of gonidia may be present in the hinder part of the coenobium or may be represented by fewer than all of its members. When two of these members are present they are usually on opposite sides of the polar axis. The members of the third quartet are smaller than the others, whether gonidia or daughters, and their segmentation is commonly delayed with respect to that of the others. Occasionally members of the second quartet are lacking. Some examples of the numbers and arrangements of the asexual reproductive bodies are given in the following descriptions of coenobia that have been photographed. A mature asexual coenobium containing only asexual daughters is shown in Plate 2, fig. 8. This specimen (No. 13) is one of a collection taken from Pond C on the day after the type material was collected, and fixed on the same day. When pho- tographed in 1916 the preparation in which it was included seemed in fairly good condition, but by 1919 most of the speci- mens on the slide had become disorganized. The micrograph shows a coenobium about 620 by 700 » and an average spacing of the somatic cells of about 17 ». These figures correspond with a cell number of about 5,200. Camera lucida sketches of six of the eight daughters gave eight gonidia in each. The other daughters were too obscure for sketching. The gonidia were about 20 to 24 » in diameter. A mature asexual coenobium with nine daughters is shown in Plate 6, fig. 39. This specimen (No. 14) is from the same prep- aration as the preceding one. From the photograph the number of somatic cells appears to be about 8,600. A mature asexual coenobium with ten daughters is shown in Plate 5, fig. 34. This specimen (No. 15) is also from the same collection as the preceding one. It is very much flattened by the 21,1 Shaw: Merrillosphaera 99 cover glass, and the anterior pole is turned away from the observer approximately 24°. The members of the two quartets are symmetrically arranged, and the posterior pair of smaller daughters lies almost in the middle of the picture. The total number of gonidia in the ten daughters is about seventy-seven. A mature asexual coenobium with only seven daughters is shown in Plate 6, fig. 41. This specimen (No. 16) is from the same lot as the preceding. A member of the posterior quartet is missing. A half-grown coenobium with only six gonidia is shown in Plate 2, fig. 11. This specimen (No. 17) is from the same source as the preceding. The six gonidia are about 50 to 53 » in diameter and are very symmetrically arranged in an anterior quartet and a posterior pair. An asexual coenobium with eleven daughters, all asexual, is shown in Plate 2, fig. 12. This specimen (No. 18) was taken on October 2 and fixed on the following day. The daughters are arranged symmetrically, eight larger ones being distributed in two quartets in the equatorial zone and the three remaining ones, which are smaller, as if in a quartet with a missing member, between the equator and the posterior polar region. In seven of the daughters it was possible to count eight gonidia. In the others not more than seven could be seen with certainty. An asexual coenobium with eight female daughters is shown in Plate 3, fig. 18. This specimen (No. 19) is from the same collection as the one shown in Plate 1, fig. 3, but fixed a day earlier. It differs from that one mainly by having the second quartet of daughters not so far from the posterior pole of the * ecoenobium. The oogonia in the nearest daughter numbered twenty-one and measured about 15 »in diameter. The positions of the daughters are unfavorable for showing the part of each around the anterior pole that lacks oogonia. The following specimens are from the same collection as the type specimen: Specimen 20.—Plate 6, fig. 37, shows a side view of a coenobium — with eight gonidia somewhat out of the typical positions. The gonidia are about 60 » in diameter. Specimen 21.—Plate 6, fig. 38, gives a view of a similar coe- nobium that is tilted about 45° away from the observer. The gonidia are 53 to 60 » in diameter. i Specimen 22.—Plate 5, fig. 32, gives a posterior polar view of a coenobium with ten gonidia, of which the members of the hinder pair are about 53 » and the others about 64 » in diameter. 100 The Philippine Journal of Science 1922 Specimen 23.—Plate 1, fig. 6, shows a young coenobium with eight gonidia that are mostly about 27 », one being about 25 p in diameter. Specimen 24.—Plate 3, fig. 19, shows another young coenobium with seven gonidia, of which six are about 28 » and one about 24 » in diameter. From the same pond that yielded specimens 8 and 9 (Plate 4, fig. 24, and Plate 1, fig. 5) the following specimens were obtained: Specimen 25.—Plate 2, fig. 18, shows an asexual coenobium with eight gonidia in process of division. The specimen is very much flattened by the cover glass. The reproductive body at the right in the upper row and the first and third ones in the lower row are in the 4-celled stage. The others are 2-celled. They measure about 90 » wide. . Specimen 26—Plate 5, fig. 33, shows an anomalous case in which one of the eight reproductive bodies is smaller than the others. It measures 28 » wide; the others being 46 to 50 pz. From another pond, J, on October 13, a collection was made from which the following specimens were selected: Specimen 27.—Plate 3, fig. 17, gives a side view of a typical coenobium with nine gonidia of which the larger ones, about 46 p» wide, form two characteristic quartets and a smaller one, 35 p wide, occupies one of the places of a third quartet. Specimen 28.—Plate 3, fig. 15, shows an asexual coenobium with five asexual daughters and three gonidia. It is tilted ap- proximately 12° away from the observer. The daughters are the anterior quartet and the left, distant (out of focus) member. of the posterior quartet. The mother coenobium is only very slightly compressed by the cover glass, but the daughters are flattened parallel with the coenobium wall. Each of the daugh- ters contains eight gonidia, which are about 12 » in diameter. The gonidia of the mother are about 85 » in diameter. SMALLER FORM OF THE SPECIES From a small pool, B, in Pasay, about a quarter mile from the ponds, there was taken, in company with a species of Volvoz, what appears to be a small form of Merrillosphaera carteri with smaller gonidia. I shall assume that it is such and describe the specimens that are figured. Specimen 29.—A mature asexual coenobium with eight asexual daughters is shown considerably flattened in Plate 3, fig. 20. It presents a posterior view, and near the posterior pole there is 21,1 Shaw: Merrillosphaera 101 a contracted opening through which a daughter may have been born. Each .of the daughters contains eight gonidia except the upper one, which has nine. The diameter of the mother is about 570 » (compressed to about 112 » thick), of the uppermost daugh- ter about 175 by 190 », and of the smallest about 130 by 1385 xz. The number of cells in the mother was estimated to be 3,300 (probably less), in the uppermost daughter 1,700, and in the smallest daughter 1,500. The gonidia of the uppermost daughter measure 18 to 25 », and of the smallest 18 to 19 uz. Specimen 30.—An asexual coenobium, Plate 5, fig. 36, with twelve reproductive bodies, ten of them segmented and two un- segmented gonidia. The two gonidia are in the hinder part of the coenobium and measure 54 and 54 by 57». The others have divided into 2, 4, or 8 cells, and are in process of dividing again. The number of cells in the coenobium was estimated at 5,000, though it is probably considerably less. Specimen 31.—An asexual coenobium with eight gonidia very symmetrically arranged is shown in Plate 2, fig.14. It measures about 350 by 380 » and has about 2,800 somatic cells. The goni- dia measure about 50 to 55 pz. Specimen 32.—A slightly younger asexual coenobium shown in Plate 2, fig. 9, has nine gonidia that measure about 53 p, 55 p, and 46 ». The number of somatic cells was estimated to be 2,750. - . Specimen 33.—A still younger coenobium with ten gonidia that do not conform to the typical arrangement is shown in Plate 3, fig. 16. In this the smaller gonidia are two, nearest the for- ward pole, that measure 37 and 39 », the others measuring 43 and 43 to 46 ». The number of cells in the coenobium was estimated to be 1,800. Specimen 34.—The youngest free coenobium shown of this lot is the one given in Plate 2, fig. 10. It contains twelve gonidia symmetrically arranged but with the middle quartet rather for- ward and separated from the hinder quartet. The number of somatic cells was estimated to be about 2,800. The gonidia measured 32 » in the first and second groups and 28 » in the third group. 7 On the same slide with the specimens numbered from 29 to 34 are twenty-two others of the same form. Some of these, like specimen 30, have the gonidia segmented. Two of these are noteworthy. They are: Specimen 35.—This is an asexual coenobium similar to speci- men 81. It has eight reproductive bodies very symmetrically 102 The Philippine Journal of Science 1922 arranged. One of the hindmost four is a gonidium of about 49 » diameter. The others have segmented into four and eight cells. This points to the undivided gonidium being probably full-grown. The number of somatic cells was estimated to be about 1,000. Specimen 36.—The smallest coenobium with large gonidia is one measuring 213 by 220 » and presenting an anterior view. Its somatic cells were estimated to be 730. It contains eight gonidia that are about 42 » in diameter. An undated slide of material from near the same source as the preceding small form bears three that are figured herewith. They are: Specimen 37.—An asexual coenobium with eight gonidia that are all divided, two into 2 cells and the others into 8, 16, or 32 cells, is shown in Plate 5, fig. 35. In the picture the coenobium measures 400 by 425 ». On the slide it measures only 328 by 840 ». I would think the magnification to be wrongly stated but for the fact that the reproductive bodies are about the same size as shown in the picture. Specimen 38—An asexual coenobium with eight gonidia, shown in Plate 3, fig. 21, has about 1,600 somatic cells. Specimen 39.—The smallest specimen shown is the one shown in Plate 8, fig. 22. It measured 138 by 144 », and was estimated to have about 500 somatic cells. It has seven gonidia of about 19 to 22 u. The apparent angularity of the cells in the photo- graph is an optical effect that is more marked on those that are out of focus, and probably due in part to the shadows of the cells of the farther side of the coenobium. A similar small form of the species was present in Pond Q in 1915. Some material taken from there on October 2 was stained with Bismarck brown, mounted in Venetian turpentine, and photographed. They all show considerable shrinkage and have shrunken further since the photographs were made. Specimen 40.—An asexual coenobium with eight gonidia of about 35 to 40 » diameter is shown in Plate 4, fig. 28. Specimen 41.—An asexual] coenobium with eight symmetrically arranged gonidia of about 57 » diameter is shown in Plate 4, fig. 27. In the photograph the coenobium measures 230 » wide and 220 » long. In 1921 the specimen had shrunken further to 206 » wide and 191 » long. Most of the shrinkage occurs in the membranes. Each gonidium is contained in the outer side of a gelatinous capsule that is very thin on the outer side and s0 thick in other directions as to crowd the neighbor capsules so al 21,1 Shaw: Merrillosphaera ° 103 that all form flattened surfaces of contact. There is little if any trace of a central core of the anterior part of the coenobium. If present, it is reduced to a bit about 15 » thick and 50 » wide. The somatic layer of cells has membranes so proportioned as to form a thin layer over each gonidium and a thick layer in other ‘parts of the coenobium wall. Specimen 42.—An asexual coenobium with ten daughters, shown in Plate 4, fig. 26, presents a side view with the charac- teristic arrangement of two groups of four and a hinder pair of smaller daughters. The central core of the anterior part of the coenobium seems to be present. Specimen 43.—An asexual coenobium with twelve daughters is shown in Plate 4, fig. 25. In this the four hindmost daughters are smaller than the others. CELL MEMBRANES In some Venetian turpentine preparations of material stained with Bismarck brown the cell membranes are rendered visible, partly by the stain and sometimes, apparently, partly by the absorption of water by the membranes. One specimen from Pond E, November 1, 1914, was selected to illustrate the cell membranes. Particulars in regard to this specimen follow. Specimen 44.—An asexual coenobium with eight daughters was photographed as shown in Plate 4, fig. 31, on May 15, 1916. On the following day it presented a different appearance and was again photographed as shown in Plate 4, fig. 30, and Plate 6, fig. 40. It was believed to have rotated with a slipping move- ment of the cover as a result of the slide having been left in a vertical position. The first picture seems to be a side view of a coenobium that has had the poles flattened by pressure,‘ and the others appear to be anterior polar views in which the flattened condition of the poles does not show. It is evident from all of these figures that the somatic cells form a layer, which is thin over the daughters and thick between them and over the poles. The membranes of the somatic cells form prisms with the protoplast in the outer end of each. In the thicker parts of the coenobium wall these prisms are about 18 » high; over the daughters they grade down to only a small fraction of that height. The outer ends of the prisms are slightly convex in the photographs. By further shrinkage they have become, “This flattening of the poles may have been due to great shrinkage of the parts of the coenobial core that filled the front and back of the coenobial cavity. 104 ‘The Philippine Journal of Science 1922 in 1921, very convex. So it appears that the convexity shown in the figures is probably largely a result of shrinkage. Plate 4, fig. 31, shows something of the thick capsular membranes around the daughters. MIXED REPRODUCTIVE BODIES Rarely (twice) coenobia were found that contained an anom- alous mixture of reproductive bodies. One such from Pond F was found associated with a nearly typical form of this species. It will be described next. Specimen 45.—Plate 8, fig. 28, shows a coenobium containing thirty-five reproductive bodies of three kinds. It measured in 1916, 435 by 440 » with a thickness of ‘about 300 ». Now, in 1921, its thickness is less and its dimensions are greater. The number of somatic cells was estimated to be about 4,360. Four of the reproductive bodies are asexual daughters. In one of these eight gonidia can be counted: In the others only six or seven can be seen, though there is some obscurity. Fourteen of the reproductive bodies are cells about 42 » wide, and seven- teen are smaller cells about 26 to 32 » wide. The former are vacuolate cells that I have supposed to be gonidia, and the latter are set down in my notes as oogonia. This assumption may be wrong. As before stated, this specimen is to be regarded as an anomaly. : MERRILLOSPHAERA ‘CARTERI VAR. TYPICA VAR. NOV. Carter’s description (’59, pp. 2 to 5 and 18 to 19, pl. 1, figs. 1, 3, 4, 7, 8, and 10), given under the name of Volvoz globator, will now be recast in the terminology used in the present paper. His figures of this species are reproduced herewith in Plates 7 and 8 on larger scales to make their dimensions directly com- parable with the micrographs of the Philippine material. His measurements were given in fractions of an inch. The form of the adult coenobium was described by Carter as spherical, or nearly so. The size, he stated, equals 770 ». A nearly mature asexual coenobium which he figured (Plate 7, fig. 42) measured, according to the scale given, about 744 p» broad by 765 » long. An oosporic coenobium which he figured (Plate 8, fig. 46) and stated to be the largest he had seen and about 608 » in diameter, measures 606 by 616 ». An antheridial coenobium shown in his fig. 8 (Plate 8, fig. 48) was stated to be about 270 » and the figure measures about 283 by 287 ». The diameter of the daughters before birth is shown to be at 21,1 Shaw: Merrillosphaera 105 least as much as 192 », at which time the gonidia of the daughters are about 32 to 42 » in diameter; and, according to the figure, while the gonidia in different daughters of the same parent vary, all the gonidia in any one daughter are about equal in size. An asexual coenobium in which the gonidia had not yet segmented, as indicated by the statement that the coeno- bium contained only one generation, shown in his fig. 3 (Plate 7, fig. 45), measures by the figure 637 by 643 pe The number and size of the somatic cells were not given. These cells were described ag globular and biciliated. Carter’s fig. 1 a (Plate 7, fig. 43) of such a cell shows no intercellular protoplasmic filaments, though this point loses significance when we note that his figure of a somatic cell of a true Volvor on the same plate (fig. 2 a of his plate) shows a stellate proto- plast without depicting the intercellular filaments which are characteristic of that genus. The number of the gonidia was given as generally eight, and emphasis was laid on the statement that they are distinctly visible in the daughters before birth. The distribution of these gonidia in the coenobia was not indicated more particularly than by the specification: “Daughters confined to the posterior three fourths of the sphere, the anterior fourth being empty.” The “almost mathematically regular arrangement” of the goni- dia, which was emphasized by Powers (’08, p. 153) in his de- scription of the American variety, is masked in Carter’s figures by a degree of artificiality. This artificiality of the figures is evident from the fact that, of the sixty-four gonidia in his fig. 1 and the eight gonidia in his fig. 8, no two overlap. Even the overlapping of the daughters in his fig. 1 was kept down as much as possible. With this in mind, we can appreciate the probability that the arrangement of the gonidia in Carter’s material was as regular as has been found in any variety of this species. Carter’s statement that each “daughter” (that is, gonidium) consists of an enlargement of a peripheral (somatic) cell re- ferred, for illustration, to his fig. 11, which he listed as repre- Senting another species, described in the same paper, and that happens to be a true’ Volvox, wherefore we may discount its strict applicability to the species under consideration. Acord- ing to his account (Carter ’59, p. 3), as the gonidium— enlarges, the chlorophyll and protoplasm together are seen to form an areolar structure around the internal periphery of the cells (fig. 4), which 106 The Philippine Journal of Science 1922 goes on increasing in size, and the starch-cells and chlorophyll increasing in number and quantity respectively, until a sudden re-arrangement of the gonimic contents takes place, and the whole is transformed into a globe of peripheral cells. The figure (Plate 7, fig. 44) shows a gonidium about 85 p in diameter. It represents the chloroplasts as forming a pe- ripheral network with expanded nodes containing pyrenoids. The resemblance of the nodes of the chloroplast network in this figure to the somatic protoplasts of Volvox globator (com- pare Overton, ’89, pl. 1, fig. 4) is so great that Klein (’89A, p. 196) was led to interpret the figure as representing a small coenobium of that species. The vacuolate protoplasm with the centrally suspended nucleus is not suggested by the drawing and was not noted in the description. It seems to me that Carter’s failure to observe any details of the segmentation of the gonidia and development of the resulting embryos was prob- ably due to lack of observations made on the living material late in the afternoon or in the night. Antheridial or oogonial daughter coenobia were observed by Carter in mother coenobia containing also gonidial daughters. Oogonia and antheridia were not found in the same coenobium, nor were daughter coenobia of both sexes found in the same parent. In single parents one, two, three, and even all, of the progeny were male. They were never more than half of the size of the asexual daughters. The antheridia numbered upward of one hundred in each male coenobium, and were scattered indiscrim- inately over the whole of the internal periphery. The androgonidia (called spermatic cells) were described as reaching a diameter of 40 y, his fig. 8a (Plate 8, fig. 48). They underwent division to form, each, a circular, tabular group of linear ciliated sperms arranged vertically upon the same plane. The spermatozoids were supposed to number one hundred twenty- eight in each platelet. The surface-view drawing, his fig. 10 b (Plate 8, fig. 49 6), which was cited for illustration of the antheridium of this species represents about ninety-three sperms. It is to be noted that the same drawing is used for illustration of the other species that was described in the same paper. The spermatozoids were described as linear, horn-shaped, about 11 » long; anteriorly attenuated and colorless, posteriorly greenish; provided with a pair of cilia which are attached to the anterior extremity, and with an eyespot some distance back of the tip (Plate 8, fig. 50 a, b, c). The progression of the 21,1 Shaw: Merrillosphaera 107 spermatozoids was described as “vermicular from their extreme plasticity.” The antheridial coenobia were liberated after the spermatozoids had been formed, but before they had become separated. Oogonia were observed to be present instead of gonidia in two, three, or even all, of the eight daughters in asexual coe- nobia. In such female daughters the oogonidia numbered from thirty to fifty and were scattered indiscriminately over the posterior three-fourths of their respective coenobia. The oogo- nidia were two or three times the size of the somatic cells and of a light greenish color, while the daughters remained within the parent. They became still more enlarged and of a deep dark green color a short time after the daughters had been liberated. After fertilization the oospores became surrounded by a thickened spore wall, which appeared to be “slightly wavy in its outline.” The spore wall shown in his fig. 7 a (Plate 8, fig. 47) depicting one of these spores is characterized by Stein (’78, p. 134) and by Klein (’89A, p. 196) as being slightly toothed. The diameter of the oospores was given as about 40 p, and the drawing of it measures about 42 yp. MERRILLOSPHAERA CARTERI VAR, WEISMANNIA (POWERS) COMB. NOV. Powers’s description (’08, pp. 152 to 162, 172 to 175) under the name Volvox weismannia of material collected by Professor Wolcott at Rocheport, Missouri, will now be recast for the pur- pose of comparison. The data will be taken from his photo- micrographic figures when they are not supplied by his text. It is to be noted that Powers wrongly called the gonidia ova. The form of the adult coenobium is obviously, from the figures, Spherical, or nearly so. The largest nearly mature asexual coenobium (his fig. 46) measure about 664 by 672 », and two others of about the same maturity (his figs. 37 and 55) both measure about 580 by 605 ». The size of the coenobia at the time of segmentation of the gonidia is indicated by an 8- gonidiate coenobium (his fig. 86) in which the gonidia show a radial flattening which is the first sign of segmentation. It measures about 350 ». A 10-gonidiate coenobium (his fig. 39), with the gonidia not yet showing signs of dividing, measures about 400 ». No free sexual coenobia are shown. The size of the daughters before birth reaches, as appears in his fig. 55, more than 190 », the gonidia measuring at that time more than 22 » in diameter. So far as can be seen in the figures, 108 The Philippine J ournal of Science 1922 the gonidia of the daughters before birth are about equal in size in the same daughter. This probably holds true mainly in cases in which the gonidia occur in the typical number, eight. The number of somatic cells in the coenobia was not stated by Powers. Estimates of the number of cells in the coenobia shown in ten of his figures, based on counts of cells in small, centrally located areas,® with no allowance for the error due to the flattening of the coenobia for photography, give numbers ranging from 2,500 to 7,500, numbers coming between the 11th and 18th powers of 2, in which class the real numbers probably belong. Powers did not describe the somatic cells. His figures show the distance between the protoplasts to be greatest at the anterior pole and to diminish progressively toward the posterior pole, where it is least, the ratio of the intercellular distances at the opposite poles being, for example, in a nearly mature asexual coenobium (his fig. 46) nearly 2 to 1. The number of gonidia (which Powers erroneously called ova, occasionally referring to them as primary sex cells) in the asexual coenobia he stated to be, most commonly, eight and ten, the latter number predominating. Less frequently, he stated, there are twelve. Odd numbers also occur. The gonidia are differentiated from the somatogenic cells at about the 64-celled stage of the embryonic development; or, if not, certainly at the next step. They are large globular cells, which attain diameters of about 25 » before birth (his fig. 87), and as great as 90 » before segmentation (Powers, ’08, p. 156), the largest that was figured (his fig. 41) being about 65 p in diameter. They appear to be multivacuolate with a central nucleus. _ The arrangement of the gonidia is almost mathematically regular. When the number is eight the arrangement is simplest and symmetrical; four lie equidistant from one another near * The estimates were made by applying the formula, N= pt in which N is the total number of peripheral cells, D the diameter of the coenobium in microns, n the number of cells counted in a selected area, and c a. constant, representing 4 in the formula, N— 7m a batt 9 in which a is a T the area used for the counts. In this case the constant, ¢, was given a value of 1,000 by taking an area of 3,141 square microns. The radius of a circle having this area, multiplied by 79, the indicated linear magnification of the figures, was used for cutting a hole in a card which was placed over the figures for the counts. The diameter of that hole was about 5 milli- meters. 21,1 Shaw: Merrillosphaera 109 the periphery of a transverse plane considerably anterior to the equator of the coenobium, the four others lie in the pe- riphery of a transverse plane well toward the posterior pole and alternating with the members of the anterior quartet. When ten are present, two are usually uniformly smaller and placed between the second quartet and the posterior pole on opposite sides of the polar axis in positions more or less nearly opposite to members of the anterior quartet. When twelve are present there are two quartets of larger gonidia and a posterior quartet of smaller gonidia. Deviations in the number of gonidia occur by the omission of one or more members of one or more of the quartets or pairs. In such cases the remaining members retain their typical places. The segmentation of the gonidia was not described, and the radial flattening of the gonidia which precedes segmentation was not recognized as such but was noted as a character of the “ova,” that is, the gonidia. The account of the embryonic metamorphosis is so involved with an account of an invagination and inversion process that it is impossible to decide what of it is pathological and what of it, if any, is misinterpretation of stages of normal development. Instances of invagination in young but completely closed coenobia were sought by Powers (08, p. 160), but not found. The male coenobia were called dwarf male colonies. Powers records having observed one hundred thirty-one parental coe- nobia bearing recognizable male progeny. The number of males in a parent varied from one to ten. The progeny were exclusively male in only one instance, the one shown in his fig. 57. The unborn male coenobia are about equal in size to their unborn asexual sisters, about 210 » in diameter. They are dwarf only in the sense of having no postnatal growth. It was stated that the sperms were ripened before the male coe- nobia were liberated. The number of somatic cells is small in proportion to the number of androgonidia. The sperms are formed in tabloid bundles of sixty-four and one hundred twenty- eight. The two sizes occur side by side. The sperm bundles are slightly concave, presumably on the posterior, nonciliated Side. The spermatozoids are slender and have terminal cilia. So far as it relates to female coenobia and to real oogonia Powers’s account is almost blank. This fact may have a signifi- cant bearing on the consideration of Klein’s account of the material collected by Migula, which is to be taken up in the next section of this paper. In Powers’s inventory of the contents 110 The Philippine Journal of Science 1922 of twelve coenobia he records that one daughter of No. 2 had “sixteen reproductive cells of half the size” as compared with those of its nearest sister, and No. 4 contained a “typical daughter with 18 or 20 reproductive cells.” I take these to be real female daughters. Embryonic female coenobia are prob- ably shown in his figs. 830 and 47, in which the reproductive cells, it was stated, are “probably 18 or 20” and “double the [usual] number of reproductive cells,” respectively. Of the latter case Powers said: “Their position in pairs may be par- tially noted; also their size reduced by one-half.” No free stages of the female coenobia were described, and likewise no oospores. MERRILLOSPHAERA MIGULAE SP. NOV. Klein’s description, under the name Volvox aureus (’89B), of material that had been collected and prepared by Migula in Karlsruhe, Germany, was probably less complete than it might otherwise have been, partly as a result of the treatment of the material as a mere form of a species which Klein had recently described in great detail, and partly as a result of being rep- resented by preserved material only. The most important part of this description consists of the figures of eight coenobia with various contents. The first figure shows a coenobium which differs from typical Merrillosphaera carteri in being smaller, having a smaller number of somatic cells, and having the pro- toplasts of the latter somewhat more elongated. More-marked differences are apparent in other figures. Before we attempt to summarize the characters of the coe- nobia described we have to consider the possibility or probability that certain misinterpretations were incorporated in the de- scription. In one of the figures (Klein ’98B, fig. 7) the neighbor- ing somatic protoplasts are shown connected by proptoplasmic filaments of the kind that is characteristic of Volvox aureus. Since Klein definitely stated that such connecting filaments were not visible in the preparations, the drawing evidently shows more than was visible. This leads me to question whether the four re- productive cells in this coenobium, and similar cells in five other coenobia, that were called, by Klein, “recently fertilized eggs” and drawn with very distinct and apparently thickened walls, were properly interpreted and correctly drawn. In view of the unquestionable fact that such a misinterpretation was made by Powers (’08) when he called the gonidia “ova,” it is not un- thinkable that such an error should have been made by Klein. 21,1 ° Shaw: Merrillosphaera 111 Apparently an absence of female coenobia in advanced stages of development in Powers’s material was a cause contributing to the error in hig case, and a similar condition of the material used by Klein may have contributed to the same error in his work. Nevertheless, the simplest hypothesis for us to adopt for the purposes of the present discussion is that the large re- productive cells present in several of Klein’s coenobia were in some cases gonidia, and in others the nearly ripe and the recently fertilized eggs that he called them. The form of the coenobium in four of Klein’s figures is nearly spherical, and in the other four ‘decidedly ovoid. The dimen- sions ° were given for six of the coenobia. Assuming that all of the drawings are on the same scale” I have measured those for which he did not give figures and here give the dimensions of all the coenobia in the order of the figure numbers: 290 by 325 p», 345 p, 270 p, 225 by 260 », 320 by 380 p, 290 pz, 300 », and 320 by 425 ». Of the spherical coenobia, two are practically isodiametric and two are very slightly elongated. It is the polar axis, if any, that is the longer. The somatic protoplasts were represented as round in surface view and ovoid in median optical section views of the coenobia. It was stated that the cilia were not visible under the magnifi- cation that was used for making the drawings. Measurements of the vegetative cells, by which I understand Klein to mean protoplasts, were given for all of the coenobia as follows, in the sequence of the figure numbers: 7.3 to 7.5 p, 6.4 to 7.3 p, 6.5 u, 7.3. to 8.9: 7 to TH p73. to 8 pm 64 to 7.3 4, and 6.4 p». The somatic protoplasts are farther apart at the anterior pole and the spacing is graded from one pole to the other in the drawings of all but one or two of the smaller coenobia. The number of somatic cells present was stated by Klein for all of the mother coenobia except that of his fig. 4. I have assumed that these numbers were obtained by a method which Klein stated in his first paper (‘89A, p. 146). The formula used in that procedure made no allowance for the fact that the areas occupied by the individual cells are * proportional to the areas of hexagons having diameters equal to the intercellular *The figure 260 given for the shorter dimension of the first coenobium is obviously a misprint for 290. "The scale is evidently about 152 diameters, and not 120 as stated in the explanation of the plate. * As was pointed out by Janet (12, p. 28). 112 The Philippine Journal of Science 1922 distances rather than to squares having the same diameters. Hence the numbers should be larger than given, by about 15 per cent. The numbers thus revised are, respectively: 1610, 1440, 1440, —, 2180, 1675, 1675, and 2080. The numbers of reproductive bodies present in the coenobia are 8, 11, 10, 9, 9, 10, 10, and 3, respectively. In the latter case, that of the largest coenobium, the maturity of the daughter coenobia present is such as to suggest that other daughter coenobia might have been already discharged from the mother. In all of the mother coenobia reproductive bodies are lacking in the anterior third and confined to the other two-thirds. In two coenobia the only reproductive bodies are daughters, in four they are a mixture of reproductive cells and daughters, and in the remaining two they are reproductive cells alone. These reproductive cells are mostly large, round cells with a doubly defined wall, about 55 » in diameter, which Klein called recently fertilized eggs. In one coenobium four of these so-called ferti- lized eggs are accompanied by five similar cells, somewhat smaller and without the double wall, which Klein called nearly ripe eggs. Granting that these be in reality eggs and oospores, we have here a distinct species of Merrillosphaera. Were they resting spores developed from gonidia, we would have a distinct variety of M. carteri; but if they be, as I am not now inclined to believe, merely asexual gonidia that, for one reason or another, were mistaken for eggs, then the variety is hardly distinct. The first coenobium on the plate illustrating this variety (Klein, ’89B, fig. 1) is one with eight nearly equal daughters almost perfectly symmetrically arranged in two quartets. ° One member of each quartet is a male coenobium containing andro- gonidia which are shown in the first and second stages of division. These male coenobia are of about the same size, 80 to 90 », and shape as their sisters. The latter were valled female daughter colonies. It seems to me more likely that they were in reality asexual daughters. Each of them contains a small number of relatively large reproductive cells measuring 18 to 20 ». In °A coenobium almost exactly like this one is represented diagram- matically by Janet (712, p. 117) in his fig. 14, D, under the name Volvox aureus. It contains three male coenobia and five asexual daughters that are mislabeled female coenobia, each of the latter containing eight or nine gonidia that are labeled gynogonidia. There is nothing in this figure essentially different from typical Merrillosphaera carteri. Janet’s fig. 14, F, is evidently M. migulae with a combination of reproductive bodies not duplicating anything that was figured by Klein. In this also gonidia are labeled gynogonidia, as they ard likewise in his fig. 14, E. ; 21,1 Shaw: Merrillosphaera 113 four daughters that are entirely visible the numbers of the reproductive cells shown are 6, 6, 7, and 8. It is not likely that the number present in any case was beyond the range of variation in the number of reproductivé bodies in the mother coenobia of the preparations, for in such a case it would have attracted Klein’s attention and been the subject of special remark. There is, then, no reason for regarding these daughters as other than asexual daughters, each containing well-developed gonidia, in number about the same as the eight of the mother within which the daughters were developed. In the largest, which is evidently the most mature coenobium (Klein, ’89B, pl. 3, fig. 8), from which I venture to suppose that some of the progeny had departed, there are present one male coenobium and two other coenobia which were called fe- male. The male coenobium is fully mature, about 103 » in > diameter, and contains about two hundred thirty cells *° includ- ing about sixty ripe spermatozoid bundles, each about 12.5 p wide. The so-called female coenobia measured 115 and 135 n» in diameter, have vegetative cells about 4.3 » in diameter and reproductive cells, called eggs by Klein, that are 22 to 25 » in diameter. The numbers of the latter shown in the figure are seven and nine. These numbers are probably the total numbers found present rather than the numbers of the reproductive cells seen in the nearer hemisphere of each daughter, for the reason stated in the preceding paragraph. There is nothing about them to mark them as female. In the four mother coenobia containing mixtures of repro- ductive cells and coenobia, there are altogether ten male coenobia and ten of the kind which Klein called female, but which I regard as possibly asexual. Of the six coenobia that contain progeny, five contain males and the other one contains but a single coenobium. Carter (’59) remarked, as I have already stated, that he found males and asexual coenobia in the same mother, females and asexual coenobia in the same mother, but never male and female in the same parent. A similar state of affairs in the coenobia of M. migulae would account for the absence of any female daughters in the coenobia figured, for there is but a single daughter in all of them that is not accom- panied by male coenobia. “Including the 15 per cent which I add to Klein’s estimates for the reason stated on page 111. 187663——8 114 The Philippine Journal of Science 1922 Were I to treat this form as a variety of M. carteri I would be assuming that the female coenobia of the variety contain a much larger number of reproductive cells than are contained in the asexual coenobia. This would involve the assumption that no female coenobia are shown in the figures or described, and that the so-called eggs are in reality very large gonidia. Grant- ing that the so-called fertilized eggs are actually such, then the variety has no place in M. carteri, but stands as a separate species of the genus. Male coenobia of the kind found and described in Migula’s preparations were observed by Klein (’90, p. 27) in the collections of the following season almost to the exclusion of another form of male coenobium which Klein had previously described for what he considered to be the same species, namely, Volvox aureus. After recognizing the occurrence of two forms of male coenobia, he observed that, though either form might be met with at any time during the vegetative season, as long as sexual coenobia were being formed, from May to November, nevertheless the two forms almost never occurred simultaneously in the same habitats, and when, in exceptional cases, they were so found together, it was in each case only for a short time. The male coenobia that he had described in his first paper he called, in drawing a distinction in his third paper, “normal, typical, large Sphaero- sirae.” He called them normal and typical because he took them to be like the material on which the former genus Sphaerosira was founded by Ehrenberg (’38).1! The male coenobia of the kind found in Migula’s preparations he called “small Sphaero- sirae’”’ and he further distinguished them by the name Endo- - sphaerosira. The normal Sphaerosira averaged 350 to 450 » in diameter and was found to reach even 750 ». The androgonidia in this form were hardly distinguishable from the vegetative cells at the time of birth, and the development of the antheridia was correspondingly late. The Hndosphaerosira, on the other hand, was invariably smaller, usually between 100 and 150 » in diameter, and commonly ripened its antheridia before birth, or at least had its androgonidia segmented at that time. It also contained a much smaller number (120 to 140) of vegetative cells than the normal Sphaerosira, though the number of anthe- ridia (60 to 70) was relatively larger. Klein’s figure of an Endosphaerosira (’90, fig. 32) shows a coenobium very similar * Klein (’90, p. 82) called attention to the fact that De Toni (’89, p. 536) erroneously assigned Sphaerosira volvox Ehrenb. to Volvox glabator Ehrenb., whereas it is the male coenobium of V. aureus Ehrenb. 21,1 Shaw: Merrillosphaera 115 to the male coenobium which Carter, in his fig. 8 a (Plate 8, fig. 48), had depicted in a somewhat different style and with less skill. It appears probable to me that the Endosphaerosira of Klein was improperly assigned to Volvox aureus, and that it belongs rather to one or more of the species of the genus Merrillosphaera. The sperm bundles of the Endosphaerosira, according to Klein, contained invariably thirty-two spermatozoids. MERRILLOSPHAERA TERTIA (MEYER) COMB. NOV. When Meyer (’96) was working with material that he had collected at Marburg, Germany, and that he named Volvox tertius he, apparently, had not seen Klein’s second Volvox paper (’°89B), for he made no mention of it in either his text or his list of references. The characters appearing in Meyer’s meager description, by which his material is distinguishable from Mi- gula’s material as described by Klein, are four: First, the occurrence of smaller numbers of reproductive bodies, which ranged from four to eight in a coenobium, as compared with eight to eleven; second, the occurrence of oospores in numbers greater than the numbers of asexual reproductive bodies, which were recorded as ten, twenty, and thirty, associated with four, five, and three asexually produced daughters, respectively; third, the restriction of the antheridia to the reproductive hemi- spheres of the male coenobia; and, fourth, the fact that in some daughter coenobia the gonidia had begun to divide while the daughters still remained within the mother coenobia. The latter point is shown by the following statement in Meyer’s (’96, p. 190) text: * * * Die anfangs stets hellgriinen Zellen e? beginnen sich theilweise schon in den Tochterkugeln zu theilen, kénnen aber auch in der Tochter- kugel bis ungefiahr 50 » gross werden, ohne sich zu theilen, und sich erst theilen, wenn die Tochterkugel frei wird. * * * The ranging of the number of asexual reproductive cells downward from eight, and the segmentation of some of the gonidia before the birth of the coenobia containing them, are characters by which Meyer’s species approaches Volvox africa- nus West (’10). The occurrence of oospores in the numbers ten, twenty, and thirty is indicated only by numbers in Meyer’s table of combinations of reproductive bodies found in coenobia. The data given in that table have been rearranged in Table 1 given herewith, and the three cases indicated by the letter G constitute the only information given about the ten, twenty, and thirty oospores. 116 _ The Philippine Journal of Science 1922 TABLE 1.—Combinations of reproductive bodies reported by Meyer to have been observed in his Volvox tertius, with his index leters. Daugh- Daugh- | Daugh- Total teF8 | ters with|terswith| _Oo- repro- Index letter. estos # reproduc-| antheri-| spores. | ductive gonidia, tive cells.| dia. bodies. Abe oi sing ens poe fan nwn en = a pee nena saan b a2 -| sabe oo ose] s-p-8o-+-| --bese---- 6 6 See paayins Se bce aR apelaati ey PONRRE e 2 Rea ETS Be Gs Op eons shee 6 to 8 OP ee Sse ec See Sack Pee eE ee gS Aes ee SRR PSS he hg Ane 4to8 G ic Lat a kan Gakbaaeeeees pS Ra near re Sane epe tae 10 14 ee RSI oe cock cA Ree a ae Bios scl ees aceaae 20 25 Mee 2. vattcoduajict saubicitet Ga OCSeaeee ckd be eune dy ar et eee BES a 80 33 ete SPR Seabee) Bigham ae ae Sa Re SS . Ber See Se Seta a 5 CS ee a ee gee gE ase bes Ra pans 6 FSB R Geek ean ee. Map Ce CS Ge ESS ee oy RRS SRR a See 7 Ft. cok ecb caneh bake ohn reRA ead Oa. ne ss re ee ee Bibs 6 EP pi wldwar va ete a oldnes coCess ont cie ee See See ege Sa qT Pssst eS + ay See oe - Oakes 6 S AMES PS ae BERT S EE ORR Le asas che ee eee Ree a 8 A ST RRS Sa eS Re 5 WA o cc Re a a 5 @ bitin cob entcces cancer 7 Pei nna Gab ensiw cca oe eee ol nal SOG, 4 Be era aS BRAS yee 7 We ae Se eee hy ae eo kes ee 1 For rr goad eye Werte 5 2 Ger ian Saeko yy eee eR Ne PEN ERS Oro aa ae 2 LP ey SS BIS a are ee 7 F990 ROSS Laci he Up oi AE SR Oey ayegeee name ee ye Rape wen 1 1 6 Wilaiy sciees ea bed 2 1 2 Fd Meyer did not state the number of times that the different combinations were found, nor did he indicate whether any of the numbers or combinations of numbers were met with more frequently than others. I am inclined to regard his case A as another in which gonidia were mistaken for eggs or oospores on account of their great size. Likewise the one and two oospores of his NB cases. Meyer’s attention at the time when he described this species was directed more especially to the subject of intercellular pro- toplasmic connections and cell membranes. His new species differed from two species of Volvox in which he had made a detailed study of those connections and membranes. It en- tirely lacked the protoplasmic connections between the proto- plasts of the adults, though such connections were said to be readily visible between the cells of living, unborn daughter coenobia (Meyer, ’96, p. 200, pl. 8, fig. a). It differed from the other species also in the form of the membranes of the somatic cells. His diagrammatic drawings, based on careful study of stained material, are reproduced herewith as text fig. 1. The protoplasts are shown globose in form, about 7.5 pw in diameter (Meyer, ’96. pl. 8, fig. z), and separated each from its own 21,1 Shaw: Merrillosphaera 117 outer membrane m by a thick gelatinous wall w. The space within the outer membrane of a typical cell is hemispherical in its outer half and hexagonally prismatical in its inner half, with a somewhat convex inner side. The protoplast is well within the outer half of the space. The outer membranes of the pris- matical portions of neighboring cells are united to form what has the appearance of a middle lamella that is thickened at the corners k. A peripheral membrane 7 of the coenobium covers an intercellular space system 0, that is filled with an intercellular substance, between the hemispherical parts of the somatic cells. The gonidia were described as reaching diameters as great as 50 », and dividing either before or after the birth of the coenobia containing them. Cells similar to the gonidia were Fig. 1. Protoplasts and membranes of the body cells of Merrillosphaera tertia (Meyer) Shaw. a, tangential optical section at about the level of the middle of the outer half of the cell; b, similar section at about the middle of the inner half of the cell; c, diagrammatic radial section through body cells, The peripheral lamella, p, of the coenobium covers the inter- cellular space, o, filled with an intercellular mass, between the outer lamellae, m, of the neighboring body cells. Each protoplast is separated from the outer lamella of its wall by a thick gelatinous wall, w. After Meyer. Said to become dark green and become transformed into oospores when they reached diameters of 40 to 50 ».- The fertilization Was supposed to take place while the coenobia containing the eggs were still within their mothers. The antheridia were produced in the reproductive halves of the male coenobia. The androgonidia segmented when 13 » in diameter. Each produced | a hemispherical or tabular bundle of spermatozoids while the male coenobium was yet within its mother. The oospores that Were supposed to have been seen were described as having Smooth walls. Fuller accounts of both Merrillosphaera migulae and M. tertia will be needed before we shall be in a position to make a final decision as to whether the organisms represented by these names properly belong in the same or in different species. With our 118 The Philippine Journal of Science 1922 present unsatisfactory data it is useless to attempt to predict what the ultimate systematic disposition of these forms will be. Therefore, so long as the proper status of each continues prob- lematical, it seems more conducive to an early solution of the problems to treat them separately. MERRILLOSPHAERA AFRICANA (WEST) COMB. NOV. The species described by West (’10 and ’18) from Africa under the name Volvox africanus was found in such abundance and variety about Manila that it seems desirable to reserve the description of the Philippine material for another paper. There is also in some of my material a form that is in some respects intermediate between M. africana and M. carteri that will require more study before I can decide whether to treat it as a separate species or as a variety or form of M. africana. The characters that bring M. africana into the genus are the absence of connecting filaments between the somatic protoplasts and the large size of the gonidia that are differentiated very early in the development of the coenobium. The characters that distinguish this species from M. carteri are: Coenobia smaller and ovoid; number of cells smaller; goni- dia one to four, in pairs, the members of the hinder pair smaller, differentiated very early and dividing early, generally before the birth of the coenobium and often before the birth of the mother coenobium. A diagnosis of the species based on West’s description will be given after the diagnosis of the other species in a later section of this paper. DIAGNOSES OF GENUS, SPECIES, AND VARIETIES Genus MERRILLOSPHAERA novum (Volvocaceae, Volvoceae) Body a spherical or ovoid coenobium of biciliate cells which contain chloroplasts. The cells appear to lie in the periphery of a gelatinous matrix surrounded by a hyaline envelope through which the cilia extend. Somatic protoplasts globose or ovoid, each inclosed in a thick gelatinous membrane which is more or less prismatic inform. No protoplasmic filaments connecting the protoplasts. Asexual reproduction by gonidia, differen- “This name was first published in a footnote (Shaw, ’19, p. 512) ina number of the Philippine Journal of Science that was issued March 30, 1920. 21,1 Shaw: Merrillosphaera 119 tiated in early embryonic stages,!* which sink from positions in the layer of somatogenic cells into positions directly within the coenobium. They develop to relatively large size before seg- mentation. Sexual reproduction by oospores formed from oogo- nidia which are smaller and more numerous than the gonidia and usually borne in female coenobia, and antheridia formed from androgonidia which are smaller and more numerous than the oogonidia and borne in male coenobia. Antheridia in the form of sperm bundles or platelets that may be more or less cupped. Spermatozoids elongate, probably with terminal cilia. The type species of this genus is Volvox carteri Stein, based on Volvox globator Carter, non Ehrenberg, described by Carter in Bombay, India. The first variety of the species to be de- scribed in sufficient detail to supply the essential diagnostic characters of the genus is Merrillosphaera carteri (Stein) Shaw var. manilana Shaw, described in this paper from material col- lected about Manila, Philippine Islands. Other forms are: Var. typica Shaw based on Carter’s description of the Bombay mate- rial; var. weismannia (Powers) Shaw based on Volvox weisman- mia Powers described from material obtained in Missouri, North America; and species migulae Shaw based on the description by Klein under the name Volvox aureus of material obtained from Karlsruhe, Germany. * Other species are: A somewhat doubtful one, Merrillosphaera tertia (Meyer) Shaw (Volvoxz tertius Meyer) described from Marburg, Germany, and M. africana (West) Shaw (Volvoz afri- canus West) described from Albert Nyanza, Africa, and reported by Shaw from the Philippine Islands. MERRILLOSPHAERA CARTER! (Stein) comb. nov., var. TYPICA var. nov. Plates 7 and 8. Volvox globator Carter, non Ehrenberg, in Ann. & Mag. Nat. Hist. III 3 (1859) 2-5, 18, 19, pl. 1, figs. 1, 3, 4, 7, 8, and 10. i Volvox carteri Stein in Der Organismus der Infusionsthiere. Leipzig 3* (1878) 1384. “It might more properly be said that the somatogenic cells are differ- entiated from the gonidia at a very early stage of the embryonic develop- ment of the coenobium, and that they become distinguishable by continuing to divide while the reproductive cells cease dividing and begin to grow. The view expressed by Harper (’18, p. 163) that Volvor “shows vegetative totipotence and equivalence of its cells in the growth of the colony” is obviously not applicable to any of the species of Merrillosphaera. More- over, it appears likely that the gonidia in the latter genus will be found to be representative of segments in the 8- or 16-celled stage of development, and that they become differentiated when those segments have further divided two or three times. 120 The Philippine Journal of Science 1988 Coenobia spherical; asexual reaching a diameter of 770 yn; female, 610 «4; and male, 270 ». Number of somatic cells not stated; protoplasts globose, of size not. stated. Gonidia gen- erally 8, all in a coenobium of about the same size and confined to the posterior three-fourths of the coenobium; arrangement not noted. Size of gonidia at time of differentiation not noted: reaching 40 » in daughters of about 195 » before the birth of the daughters; attaining a diameter of 85 ,» before dividing; chloroplast forming a peripheral network within the gonidium. Asexual daughters formed alone, with female daughters, or with male coenobia in the same mother coenobium. Male and female coenobia not observed in the same mother. Oogonidia 30 to 50, distributed in the posterior three-fourths of the female coenobium; reaching two or three times the size of the somatic cells in daughters yet unborn. Oospores about 40 » in diameter, with the thickened wall “slightly wavy in its outline’ or wavy toothed. Androgonidia upward of 100, scattered over the whole internal periphery of each male coenobium; reaching 40 » in di- ameter; producing antheridia before the birth of the male coe- nobium. Antheridia consisting of platelets of 128 (?) sperms. Each spermatozoid with a pair of terminal cilia and a lateral stigma; spermatozoids liberated after the birth of the male coenobium. Habitat——A pool in Bombay, India (leg. H. J. Carter, 1858). MERRILLOSPHAERA CARTERI (Stein) Shaw var. MANILANA var. nov. Plates 1 to 6. ; Coenobia spherical or slightly elongated; asexual reaching about 720 by 760 »; female, about 690 by 730 »; and male, about 260 by 280 » Number of somatic cells in asexual coe- nobia as high as 11,400; in female coenobia, 8,000; and in male coenobia, 2,500; commonly much fewer. Somatic protoplasts ovoid, about 5 to 6 » thick. Gonidia 8, about equal in size, arranged symmetrically, 4 in front of, and 4 behind, the equator, sometimes with 1 to 4 additional, somewhat smaller gonidia nearer the hinder pole. Gonidia in very young coenobia about 17 » in diameter; reaching at time of birth about 24 », and at maturity about 85 ». Asexual daughters reaching about 200 po diameter at time of birth. Male coenobia producing about 112 androgonidia. Androgonidia reaching about 16 pw; dividing before birth of the male coenobium to form sperm platelets that are more or less cupped. Female coenobia producing about 24 to 40 oogonia (oogonidia) of which none is in the front part 21,1 Shaw: Merrillosphaera 121 of the coenobium. Oogonidia about 14 » in diameter in coeno- bia of about 175 » diameter at time of birth. Oospores about 50 w in diameter; with very slightly wavy wall about 2.5 to 3 ph thick; with eccentric protoplast about 37 » wide. Habitat.—Fresh-water pools, near Manila, Philippine Islands (leg. W. R. Shaw 1914).** MERRILLOSPHAERA CARTERI (Stein) Shaw var. WEISMANNIA (Powers) comb. nov. Volvox weismannia Powers in Trans. Am. Microscop. Soc. 28 (1908) 141-175, pls. 24-26, figs. 25, 27-61. Coenobia spherical or nearly so; asexual reaching at maturity 664 by 672 »; at time of segmentation of the gonidia 350 to 400 p. Mature females not described. Male coenobia about 210 » in diameter. Somatic cells 2,500 to 7,500; form and size not stated ; intercellular distances not given, but greatest at anterior ‘pole and grading to about half as great at the posterior pole. Gonidia 8 to 12, most commonly 10; differentiated at the 64- celled or the next stage, reaching 25 » before the birth of the coenobium containing them and 90 » (the largest figured was 65 ») before segmentation. Gonidia arranged regularly in two quartets, without or with an additional pair or quartet of smaller ones from which members may be absent; occupying the pos- terior three-fifths of the coenobium. Female coenobia described only in very young stages within their mothers. Oogonia dif- ferentiated at an early embryonic stage; twice as numerous (16 to 20) as the gonidia. Male progeny in parental coenobia 1 to 10; before birth about equal in size to their asexual sisters. Androgonidia more numerous than oogonidia; scattered around the internal periphery of the male coenobia segmenting into antheridia and forming tabloid bundles of 64 and 128 sperms that ripen before the birth of the coenobia containing them. Spermatozoids slender, with terminal cilia. Oospores not de- scribed. Habitat.—A broad, shallow pond near Rocheport, Missouri, North America (leg. Wolcott, 1904). “Slide mounts of material of this species from the vicinity of Manila have been sent to Prof. Frank G. Haughwout, Bureau of Science, Manila, and to Prof. Douglas H. Campbell, Stanford University, California. Material bottled in glycerine has been sent to sixteen biologists in America and to sixteen in Europe and Asia. Duplicates of this bottled material are available for distribution from my American address: Claremont, Cali- fornia.—W. R. S. 122 The Philippine Journal of Science 1922 MERRILLOSPHAERA MIGULAE sp. nov. Volvox aureus Ehrenberg, Klein, in Ber. d. deutschen bot. Ges. 7 (1889) 48-53, pl. 3, figs. 1-10; and in Ber. d. naturf. Ges. Freiburg i. B. 5 (1890) reprint 23 and 24, 27 and 28, pl. 2, fig. 4, pl. 4, fig. $2. Coenobia spherical or ovoid; asexual usually between 250 and 350 p», reaching 320 by 425 »; female similar; male averaging 100 to 150 ». Number of somatic cells, 1,440 to 2,180. Proto- plasts ovoid, about 7 to 8 » in diameter. Gonidia 8, sometimes more; sometimes of uniform size and arranged in two quartets; attaining diameters of 25 ,» before birth of the coenobia con- taining them, and probably 62 » before segmentation. Some coenobia with gonidia and oogonidia. Female coenobia with about 8 oogonidia. Asexual, male, and possibly female coenobia formed in the same mothers, and of about the same size at time of birth. Antheridia, consisting of sperm platelets of about 12.5 » diameter, containing 32 spermatozoids, commonly mature before birth of the male coenobia. Oospores with smooth walls, 52 to 62 » in diameter. Habitat.—Karlsruhe, Germany (leg. Migula, 1888); Buch- heim, Alt-Breisach, Hochdorf, Germany (leg. Klein, 18887). MERRILLOSPHAERA TERTIA (Meyer) comb. nov, Volvox tertius :Meyer in Bot. Zeit. 54* (1896) 188-191, 200-201, 216-217, pl. 8, figs. A, Z and a, text figs. 5-7. Form and size of coenobium not stated. Number of somatic cells not given. Somatic protoplasts globose, about 7.5 u». Membranes of somatic cells (described very particularly) hemi- spherical and of a thickness equal to about one-sixth of the diam- eter of the protoplast on the outer side, and prismatic and of a thickness equal to about one and a half times the diameter of the protoplast on the inner side. Gonidia 4 to 8 in the asexual coenobia; attaining diameters as great as 50 » and dividing either before or after the birth of the coenobia containing them. Female coenobia doubtful. Three coenobia were reported to contain 4, 5, and 3 asexually produced daughters and 10, 20, and 30 oospores, respectively; that is, gonidia and oogonidia in the same coenobium. Oogonidia said to form oospores when they reach diameters of 40 to 50 ». Androgonidia produced in reproductive half of male coenobium; reach 13 » diameter; produce hemispherical or tabular bundles of sperms before the birth of the male coenobia. Oospores said to have smooth walls. Habitat.—A pool near Marburg, Germany (leg. Meyer, 18957), 21,1 Shaw: Merrillosphaera 123 MERRILLOSPHAERA AFRICANA (West) comb. nov. Volvox africanus West in Journ. Quekett Mic. Club II 11 (1910) 102-108, pl. 3, figs. 8-10; and II 13 (1918) 425-428, pl. 29, figs. 4-6. Coenobia ovoid; ranging from 295 by 345 » to 480 by 610 xz. Number of somatic cells between 3,000 and 8,000. Somatic protoplasts almost globose, 8 to 9.5 » in diameter; about 25 x» apart at anterior pole, and grading to much smaller distances at posterior pole. Gonidia 1 to 4; arranged more or less in pairs; gonidium or pair of gonidia near posterior pole smaller than the others and producing smaller daughters. Gonidia of the daughters in some cases divide and produce granddaughters containing gonidia that are the beginning of the fourth genera- tion while the daughters are still within their mother. Daughter coenobia strongly ovoid and more or less flattened by mutual pressure; granddaughters spherical. Female coenobia with 70 to 80 oospores (average 74) ; oospores with thick smooth walls, average diameter 45 ». Description of male coenobium not yet published. Habitat.—Albert Nyanza, Africa (leg. R. T. Leiper, 1907) ; Ussangu Desert, German East Africa (leg. A. W. Jakubski, 1909-10) ; fresh-water pools near Manila, Philippine Islands (leg. W. R. Shaw, 1914). . LITERATURE CITED Carter, H. J. (59). On fecundation in the two Volvoces, and their specific differences. Ann. & Mag. Nat. Hist. III 3 (1859) 1-20, pl. 1. EHRENBERG, C. G. (’38). Die Infusionsthiere als vollkommene Organis- men. Berlin and Leipzig (1838). (Not seen.) ee Harper, R. A. (’18). Binary fission and surface tension in the develop- ment of the colony in Volvox. Brooklyn Bot. Gar. Mem. 1 (1918) 154-166, pl. 2 and figs. 1-4. JANET, C. (712). Le Volvox. Limoges (1912) 151 pp. | ; KLEIN, L. (’89A). Morphologische und biologische Studien iiber die Gat- tung Volvox. Jahrb. f. wiss. Bot. 20 (1889) 183-211, pls. 10-12. KLEIN, L. (’89B). Neue Beitrage zur Kenntniss der Gattung Volvox. Ber. d. deutschen bot. Ges. 7 (1889) 42-53, pl. 3. fei KieIn, L. (90). Vergl. Untersuchungen tiber Morphologie und Biologie der Fortpflanzung bei der Gattung Volvox. Ber. d. naturf. Ges. Freiburg i. B. 5 (1890) 92 pp., pls. 2-6. Meyer, A. (96). Die Plasmaverbindung und die Membranen von Volvox, mit Riicksicht auf die thierischen Zellen. Bot. Zeit. 54+ (1896) 187- 217, pl. 8. : “is Powers, J. H. (07). New forms of Volvox. Trans, Am. Microscop. 5 27 (1907) 123-149, pls. 11-14. 124 The Philippine Journal of Science Powers, J. H. (’08). Further studies in Volvox, with descriptions of three new species. Trans. Am. Microscop. Soc. 28 (1908) 141-175, pls. 23-26. ' SHAw, W. R. (94). Pleodorina, a new genus of the Volvocinae. Bot. Gaz. 19 (1894) 279-283, pl. 27. SHAW, W. R. (’16). Besseyosphaera, a new genus of the Volvocaceae. Bot. Gaz. 61 (1916) 253-254. : SHaw, W. R. (718). Some microtechnical methods and devices. Philip. Journ. Sci. Bot. 13 (1918) 241-261. SHAW, W. R. (719). Campbellosphaera, a new genus of the Volvocaceae. Philip. Journ. Sci. 15 (1919) 493-520, 2 pls. SHAW, W. R. (’22). Janetosphaera, a new genus, and two new species of Volvox. Philip. Journ. Sci. 20 (1922) 477-508, 5 pls. 5 text figs. STEIN, F. (’78). Der Organismus der Infusionsthiere. Leipzig 3* (1878) 134, West, G. S. (710). Some new African species of Volvox. Journ. Quekett Mic. Club II 11 (1910) 99-104, pl. 3. WEsT, G. S. (718). A further contribution to our knowledge of the two African species of Volvox. Journ. Quekett Mic. Club II 13 (1918) 425-428, pls. 29 and 30. ILLUSTRATIONS [Photomicrographs of Merrillosphaera carteri (Stein) Shaw, from specimens mounted in gly- cerine except as otherwise noted, taken by W. R. Shaw and E. Cortes at the Bureau of Science, Manila.] PLATE 1 Fic. 1. Type specimen of Merrillosphaera carteri var. manilana. A coeno- bium with eight embryo daughters. x 100. 2. A portion of the coenobium wall and one of the embryo daughters of the same specimen as that shown in fig. 1. The embryo shows in profile two protruding gonidia. x 200. 3. A mature asexual coenobium of the same variety as fig. 1, con- taining eight female daughters. In four of the daughters the oogonidia counted are 24, 27, 31, and 37. x 100. 4. A nearly mature female coenobium with twenty-nine oogonidia. x 100. 5. A mature asexual coencbium somewhat flattened under a cover glass. It contains four asexual daughters and three male offspring. The daughters contain eight gonidia each, and the male coenobia each about one hundred twelve androgonidia that are mostly undivided in the hindmost one and mostly 2- or 4-celled in the two others. x 100. 6. A young coenobium with eight gonidia that are 24 to 27 u in diameter. x 100. — 7. A male coenobium with the reproductive bodies developed into sperm platelets that are more or less cupped. xX 100. PLATE 2 Fic. 8. A mature asexual coenobium of Merrillosphaera carteri var. manilana with eight asexual daughters, each with eight goni- dia so far as could be counted. xX 100. 9. An asexual coenobium of the smaller form. It has nine gonidia that measure 53, 55, and 46 ». xX 100. 10. A young asexual coenobium of the smaller form. It contains twelve gonidia very symmetrically arranged. In the forward group two are in the middle of the picture, in the second group two are at each side, and in the hindmost group two are near the middle of the picture. The gonidia of the third group are about 28 « wide, and the others about 82 4. There is a notice- ably large gap between the second and third groups of gonidia. x 100. 11. An asexual coenobium with only six gonidia. They are symmet- rically arranged in a group of four just ahead of the middle, and a pair halfway back from them to the hinder pole. They all measure about 50 to 53 # x 100. ies 126 Fic. The Philippine Journal of Science 1922 12. A mature asexual coenobium with eleven asexual daughters arranged in two symmetrical groups of four, one just ahead of, and the other just behind the middle, and three members of a hypothetical group of four smaller daughters just back of them. x 100. 18. An asexual coenobium with eight gonidia in process of division. The first three in the upper row and the second and fourth in the lower row are 2-celled. The others are 4-celled. This is from the same collection as the specimens shown in Plate 1, fig. 5, and Plate 4, fig. 24. x 100. 14. An asexual coenobium of the smaller form. It has eight sym- metrically arranged gonidia of about 50 to 55 4. x 100. PLATE 3 15. An asexual coenobium of Merrillosphaera carteri var. manilana with five asexual daughters and three gonidia. The daughters are the foremost four and the left (distant) member of the hindmost four. The gonidia are about 85 u in diameter. Each daughter contains eight gonidia that are about 12 » in diam- eter. xX 100. 16. A young asexual coenobium of the smaller form. It has ten gonidia. They are not typically arranged. Two nearest the forward pole are the smallest, or else the figure is upside down, in which case the forward group of four is midway between the equator and the forward pole. xX 100. 17. An asexual coenobium with nine gonidia in the typical arrange- ment of two fours with an odd one occupying one of the places of a hypothetical third group of four. The small gonidia 35 u, the others about 46 » wide. x 100. 18. A mature asexual coenobium with eight female daughters ar- ranged in two symmetrical groups of four. There are twenty- one oogonidia in the nearest daughter, and they are about 15 » in diameter. x 100. 19. A young asexual coenobium with seven gonidia of 24 to 28 diameter. x 100. 20. A mature asexual coencbium of the smaller form. It contains eight daughters. The uppermost of these contains nine goni- dia, the others eight. The gonidia measure from 18 to 25 4. x 100. 21. An asexual coenobium of the smaller form. It has eight gonidia and about 1,600 somatic cells. x 100. 22. An asexual coenobium of the smaller form. It has seven gonidia of about 10 to 22 », and about five hundred somatic cells. X 100. 23. An anomalous coenobium containing thirty-five reproductive bodies of three kinds. Four are daughters, all asexual, in one of which eight gonidia can be counted. In the others only six or seven can be seen, though there may be one or two more in obscurity. Fourteen of the bodies are about 42 » wide and are supposed to be gonidia, and seventeen are 26 to 32 u wide and supposed to be oogonidia. X 100. 21,1 Shaw: Merrillosphaera 127 PLATE 4 Fic. 24, A mature female coenobium of Merrillosphaera carteri var. manilana with thirty-six oospores and two immature repro- ductive cells. X 100. 25. A Venetian turpentine preparation of one of the smaller form. It contains twelve symmetrically arranged daughters of which the hindmost four are smaller than the others. The front of the coenobium is turned a little away from the observer. x 100. 26. A Venetian turpentine preparation of one of the smaller form from the same source as the foregoing. Ten symmetrically arranged daughters include two that are smaller than the others and form a hinder pair. The front of the coenobium is turned slightly toward the observer. x 100. 27. A Venetian turpentine preparation of one of the smaller form from the same source as the foregoing. It presents a side view showing eight gonidia symmetrically arranged. The go- nidia measure about 57 4. xX 100. 28. A Venetian turpentine preparation of one of the smaller form from the same source as the foregoing. It contains eight goni- dia of, about 85 to 40 uw diameter. The coenobium is turned about 48° away from the observer. X 100. 29. A portion of the same coenobium as that of fig. 24 showing two of the oospores and some of the somatic cells of the coenobium wall. In a direct view some of the spore walls appear more decidedly wavy than shown in this figure. x 400. 80. A Venetian turpentine preparation stained with Bismarck brown. A nearly front polar view of a coenobium with eight daughters that form two groups of four. The thickness of the layer of somatic cells is shown to be greater between the daughters and less over the daughters. x 100. 31. The same coenobium as shown in fig. 30, photographed after being rotated 90°. The poles had been flattened. Some of the boun- daries of the gonidial capsules can be seen. X 100. PLATE 5 Fic. 32. Rear polar view of a coenobium with ten gonidia arranged in two groups of four that are about 64 u in diameter and a hinder pair that are about 53 » in diameter. x 100. 88. An asexual coenobium with eight gonidia in two groups of four. An anomalous member of the forward four is smaller than the others, being only 28 » wide, while the others are 46 to 60 4. x 100. 84. A mature asexual coenobium of Merrillosphaera carteri var. ma- nilana with ten asexual daughters, of which there is a hinder pair of smaller ones and two very symmetrical groups of four that are about equal. The front is turned about 24° away from the observer. There are probably eight gonidia or less in each daughter. x 100. ae 35. An asexual coenobium of the smaller form. The eight gonidia have all divided into 2, 4, 8, 16, or 82 cells. X 100. 128 The Philippine Journal of Science 1922 36. An asexual coenobium of the smaller form. It contains twelve reproductive bodies. Two of the hindmost are gonidia that measure 54 » and 54 by 57 w. The others have divided into 2, 4, or 8 cells each and are in process of dividing again. x 100. PLATE 6 Fic. 37. Side view of a coenobium with eight gonidia about 60 » in diameter. x 100. 88. An asexual coenobium with eight gonidia of about 53 to 60 u diameter. The gonidia are fairly symmetrically arranged, but the coenobium is turned about 45° away from the observer. x 100. 39. A mature asexual coenobium of Merrillosphaera carteri var. ma- nilana with nine asexual daughters. They form a group of four just in advance of the middle, a group of three arranged like a quartet with one member absent just back of the middle, and a pair of smaller ones a little farther back. x 100. 40. A Venetian turpentine preparation. The same as shown in figs. 30 and 31. x 200. 41. A mature asexual coenobium with seven daughters that form a symmetrical group of four slightly in advance of the middle and a group of three arranged as if members of a group of four from which one is missing. In six of the daughters eight gonidia were counted. x 100. PLATE 7. MERRILLOSPHAERA CARTERI VAR. TYPICA 42. A reproduction of Carter’s first figure (’59, pl. 1, fig. 1) enlarged to make the magnification about x 100. A coenobium with eight daughters, each of which contains eight gonidia. An artificiality of the drawing has resulted in failure to show the symmetrical arrangement of the daughters and gonidia in groups of four. x about 100. 43. Carter’s figure (‘59, pl. 1, fig. 1 a) of a somatic cell, showing the rounded form of the protoplast and the absence of connecting filaments. Size not stated. 44. Carter’s figure (’59, pl. 1, fig. 4) of a gonidium about 85 » in diameter and nearly ready to divide. x about 400. 45. Carter’s figure (‘59, pl. 1, fig. 8) of a coenobium with eight nearly mature gonidia, one of which is shown more magnified in fig. 44. The real arrangement of the gonidia is probably not shown because of an artificiality introduced into the drawing to pre- vent any overlapping of gonidia and at the same time leave the front of the coenobium free from reproductive cells. X about 100. PLATE 8. MERRILLOSPHAERA CARTERI VAR. TYPICA 46. Carter’s figure (‘59, pl. 4, fig. 7) of a female coenobium. xX about 100. 47. Carter’s figure (‘59, pl. 1, fig. 7 a) of an oospore after fertiliza- tion. x about 650. 48. Carter’s figure (‘59, pl. 1, fig. 8 a) of a male coenobium. x about 100. 2, 1. Shaw: Merrillosphaera 129 49. Carter’s figures (’59, pl. 1, fig. 10 a and b) of antheridia. X about 650. 50. Carter’s figure (’59, pl. 1, fig. 8 b) of spermatozoids. x about 1,000. TEXT FIGURE Fic. 1. Protoplasts and membranes of the body cells of Merrillosphaera tertia (Meyer) comb. nov. After Meyer. a, tangential section at about the level of the middle of the outer half of the cell; }, similar section at about the middle of the inner half of the cell; c, diagrammatic radial section through body cells. 1876639 SHAW: MERRILLOSPHAERA ] [Pumip. Journ. Sct., 21, No. 1. Seb ite. Sa PLATE 1. MERRILLOSPHAERA CARTERI (STEIN) SHAW. SHAW: MERRILLOSPHAERA | [Puitre, Journ. Sct, 21, No. 1, PLATE 2. MERRILLOSPHAERA CARTERI (STEIN) SHAW. SHAW: MERRILLOSPHAERA ] [PuHiuie. Journ. Scr., 21, No. 1. PLATE 3. MERRILLOSPHAERA CARTERI (STEIN) SHAW. Se, 21,°No, 4. [PHILIP. JOURN. SHAW: MERRILLOSPHAERA | STEIN) SHAW. MERRILLOSPHAERA CARTER! ( PLATE 4. SHAW: MERRILLOSPHAERA | [Pumip. Journ. Sct, 21, No. PLATE 5. MERRILLOSPHAERA CARTERI (STEIN) SHAW. SHAW: MERRILLOSPHAERA | [PHIuip. JOURN. Scr., 21, No, 1. PLATE 6. MERRILLOSPHAERA CARTERI (STEIN) SHAW. MERRILLOSPHAERA | SHAW [Puimipr. Journ. Sct, 21; NO; "1. MERRILLOSPHAERA CARTERI (STEIN) SHAW. PLATE 7. SHAW: MERRILLOSPHAERA | [Pucruie. Journ. Scr., 21, No. 1. PLATE 8. MERRILLOSPHAERA CARTER! (STEIN) SHAW. THE PHILIPPINE JOURNAL OF SCIENCE VoL. 21 AUGUST, 1922 No. 2 THE AMBOINA FUNGI COLLECTED BY C. B. ROBINSON : By H. Sypow Of Berlin, Germany The Robinson collection of fungi is not large but is especially interesting because it is the first collection made in Amboina. On the whole, judging from the species that are at hand, the fungus flora of this island is closely related to that of the Philippines, as various forms referable to the Basidiomycetes and to such genera as Meliola and Asterina occurring in the Philippines are especially numerous. BASIDIOMYCETES POLYPORUS Micheli POLYPORUS BICOLOR Jungh. Amboina, Way tommo, Reliquiae Robinsonianae 2244, August 17, 1918, on a fallen log; Negri lama, Reliquiae Robinsonianae 2304, September 8, 1913, on rocks, being a zonate form. POLYSTICTUS Fries POLYSTICTUS SANGUINEUS (L.) Mey. Amboina, Batoe merah, Reliquiae Robinsonianae 2248, July 20, 1918, on decaying logs. *For other publications on Doctor Robinson’s Amboina collections see: Merrill, E. D., An Interpretation of Rumphius’s Herbarium Amboinense, Bureau of Science Publ. 9 (1917) 1-595; Reliquiae Robinsonianae, Philip. Journ. Sci. 11 (1916) Bot. 243-319; Van Alderwerelt van Rosenburgh, C.R.W.K., The Amboina Pteridophyta collected by C. B. Robinson, op. cit. 101-121, t. 5, 6; Brothérus, V. F., The mosses of Amboina, op. cit. 12 (1917) Bot. 73-80; Beccari, O., A new species of Calamus from Amboina, op. cit. 81; Radlkofer, L., A new species of Guioa from Amboina, op. cit. 83; Smith, J. J., The Amboina Orchidaceae collected by C. B. Robinson, op. cit. 249-262. Errors. 187897 131 132 The Philippine Journal of Science 1922 POLYSTICTUS AFFINIS (Nees) Fr. Amboina, Hitoe lama, Reliquiae Robinsonianae 2090, October - 11, 1918, on dead wood. POLYSTICTUS XANTHOPUS Fr. Amboina, Reliquiae Robinsonianae 2052, on dead wood. POLYSTICTUS DISCIPES Berk. Amboina, Salahoetoe, Reliquiae Robinsonianae 2108, Novem- ber 27, 1913, on old logs. POLYSTICTUS MELEAGRIS (Berk.) Cke. Amboina, Hitoe lama, Reliquiae Robinsonianae 2442, October 11, 1913, on dead wood. POLYSTICTUS OBSTINATUS Cke. forma RESUPINATA. Amboina, Batoe merah, Reliquiae Robinsonianae 2106, J uly’ 20, 1913, on decaying log. FOMES (Fr.) Cooke FOMES ADAMANTINUS Berk. Amboina, Ayer putri, Reliquiae Robinsonianae 2111, July 28, 1913, on dead trees. GANODERMA Karsten GANODERMA AMBOINENSE (Lam.) Pat. Amboina, Reliquiae Robinsonianae 2052, on dead wood. GANODERMA TORNATUM (Pers.) Bres. Amboina, Reliquiae Robinsonianae 2050, on dead trees. TRAMETES Fries TRAMETES CORRUGATA (Pers.) Bres. Amboina, Way tommo, Reliquiae Robinsonianae 2055, August 7, 1913, on fallen log. ‘TRAMETES STRIGATA (Berk,) Bres. Amboina, Paso, Reliquiae Robinsonianae 2110, October 29, 1913, on dead wood. HYMENOCHAETE Léveillé HY MENOCHAETE CACAO Berk. Amboina, Way tommo, Reliquiae Robinsonianae 2053, August 7, 1913, on dead tree. | HIRNEOLA Fries HIRNEOLA CORNEA (Ehrenbg.) Fr. Amboina, Kati kati, Reliquiae Robinsonianae 2192, October 7, 1913, on corticated fence post. 21,2 Sydow: Amboina Fungi Collected by Robinson 133 UREDINACEAE HEMILEIA Berkeley et Broome HEMILEIA VASTATRIX Berk. et Br. Amboina, Paso, Reliquiae Robinsonianae 2190, July 20, 1913, on leaves of Coffea arabica. 5 AECIDIUM Persoon AECIDIUM KAERNBACHII P. Henn. =. Amboina, town of Amboina, Reliquiae Robinsonianae 2220, July 17, 1913, on leaves of Ipomoea sp. UREDO Persoon UREDO KYLLINGIAE P. Henn. Amboina, town of Amboina, Reliquiae Robinsonianae 2217, July 25, 1913, on leaves of Kyllingia brevifolia. ASCOMYCETES MELIOLA Fries MELIOLA EUGENIAE Syd. sp. nov. Amphigena, plagulas primitus orbiculares 2-3 mm diam. dein confluendo subinde irregulares aterrimas formans; mycelium densissime reticulato-ramosum, ex hyphis obscure castaneo-brun- neis 8-10 » crassis compositum; hyphopodia capitata numeros- issima, recta, integra, 17-20 p» longa, cellula superiore 9-11 ,» lata; hyphopodia mucronata non visa; setae mycelii numerosae, rectae vel subrectae, simplices, sursum attenuatae, sed apicibus obtusis, tota longitudine opacae, 200-300 » longae, basi 9-11 pw latae; perithecia globosa, 200-220 » diam.; asci bispori; sporae oblon- gae, utrinque late rotundatae, obscure castaneo-brunneae, opacae, 4-septatae, vix vel leviter constrictae, 44-52 » longae, 22-25 » latae. Amboina, Kati kati, Reliquiae Robinsonianae 2163, October 6, 1913, on leaves of Eugenia caryophyllata. MELIOLA AMBOINENSIS Syd. sp. nov. Hypophylla, plagulas orbiculares 3-8 mm diam. formans; mycelium reticulato-ramosum, ex hyphis castaneo-brunneis 7-9 B crassis rectiusculis vel curvatis compositum; hyphopodia capi- tata numerosa, alternantia, integra, 24-32 ,» longa, cellula Superiore ovata 12-14 » lata, basali breviore stipitiformi; hy- Phopodia mucronata non visa; setae myceliales copiosae, simpli- ces, rectiusculae vel leviter falcato-curvatae, tota longitudine opacae, ad apicem semper acutae, 300-650 » longae, basi 9-10 » 134 The Philippine Journal of Science 1922 crassae; perithecia globosa, in sicco SilinoaaJ 150-200 » diam.; asci 2- ad 3-spori, fugaces; sporae cylindraceae, utrinque obtusae, castaneae, 4-septatae, non vel vix constrictae, ‘33-36 ,» longae, 13-15 » latae. Amboina, Gelala, Reliquiae Robinsonianae 2150, September 19, 1913, on leaves of Aganosma sp. MELIOLA FAGRAEAE Syd. Amboina, Hitoe messen, Reliquiae Robinsonianae 2134bis, November 5, 1913, on leaves of Fagraea speciosa. MELIOLA PACHYCHAETA Syd. sp. nov. Epiphylla, raro hypophylla vel petiolicola, plagulas orbiculares 4-8 mm diam. velutinas aterrimans formans, mycelium ex hyphis dense reticulato-ramosis et intertextis 7-10 » latis brunneis septatis (articulis 10-20 » longis) compositum; hyphopodia capitata copiosa, alternantia vel opposita, cylindracea, recta vel subrecta, 20-24 y» longa, cellula superiore majore integra vel raro leviter irregulari 10-12 » lata, hyphopodia mucronata non visa; setae mycelicae copiosae, rectae vel subrectae, simplices, tota longitudine omnino opacae, ad apicem acutae, 600-1000 » longae, basi 15-18 y» latae; perithecia gregaria, globosa, 175-250 p diam. ; asci 2- vel 3-spori, fugaces, sporae cylindraceae, utrinque obtusae, 4-septatae, leviter constrictae, brunneae, 3642 » longae, 12-15 » latae. Amboina, Batoe merah, Reliquiae Robinsonianae 2059, August 24, 1913 (type) ; Koesoekoesoe sereh, Reliquiae Robinsonianae 2246, August 7, 1913, both on leaves of Semecarpus cassuvium. Meliola pachychaeta seems to be related to'M. aliena Syd. from which it differs especially in the much larger setae. MELIOLA STEMONAE Syd. sp. nov. Epiphylla, plagulas minutas orbiculares 1-3 mm diam. for- mans; mycelium matrici dense adpressum, radians, ex hyphis brunneis 7-11 ,» crassis ramosis septatis compositum, hyphopodia capitata modice copiosa, alternantia 18-22 » longa, cellula supe- riore 15-18 yw lata, integra, recta vel curvata, cellula basali minore; hyphopodia mucronata non visa; setae mycelii copiosis- simae, rectae vel leviter curvatae, 600-800 p» longae, basi 10-12 » latae, tota longitudine opacae vel subinde ad apicem dilutiores, ad apicem dentibus 1-3 irregularibus acutis vel obtusis plus minus longis subinde brevissimis usque 20 ,» longis erectis praeditae, subinde sed raro etiam subter apicem dente singulo praeditae; perithecia gregaria, in sicco collapsa, atra, 150-250 » diam.; asci 2- ad 3-spori; sporae oblongae, utrinque rotundatae, 21,2 Sydow: Amboina Fungi Collected by Robinson .- Jae 4-setatae, ad septa leviter constrictae, fuscae, 44-51 » longae, 17-20 yp latae. Amboina, Hitoe lama, Reliquiae Robinsonianae 2230, October 8, 1913, on leaves of Stemona tuberosa. MELIOLA ROBINSONI! Syd. sp. nov. Amphigena, plerumque epiphylla, plagulas minutas 1-3 mm diam. tenues subinde confluendo -‘majores formans; mycelium dense reticulato-ramosum ex hyphis brunneis 7-10 p crassis septatis formatum ; hyphopodia capitata copiosa, alternantia vel opposita, 20-24 » longa, recta vel cellula superiore integra cur- vata 10-14 ,» lata, cellula basali brevi; hyphopodia mucronata opposita, lageniformia, usque 23 » longa; setae myceliales numerosae, rectae, opacae vel dimidia parte superiore pellucidae, 200-300 » longae, basi 7~9 » latae, ad apicem dentibus 1-4 erectis mox brevissimis mox longioribus (usque 15 ,» longis) acutis praeditae; perithecia globosa, 150-175 p diam., in sicco collapsa; asci 2- ad 3-spori; sporae oblongae, utrinque rotun- datae, 4-septatae, ad septa vix vel leviter constrictae, brunneae, 40-44 » longae, 18—20 latae. Amboina, Soja, Reliquiae Robinsonianae 211 9, August 2, 1913, on leaves of Entada phaseoloides. MELIOLA MEGALOCHAETA Syd. sp. nov. Amphigena, praecipue hypophylla, plagulas 0.5-2 cm latas primitus orbiculares dein irregulares et saepe confluendo majores formans; mycelium reticulato-ramosum, ex hyphis atro-brunneis opacis 9-12 » crassis rectiusculis compositum; hyphopodia capitata sat numerosa, integra, 26-35 » longa, cellula Superiore rotundata, plerumque recta, 16-20 ,» lata, inferiore brevi; hyphopodia mucronata solitaria vel opposita, numerosa, lageniformia, in collum plus minus abrupte producta, usque 32 p longa, basi ca. 10 p crassa; setae myceliales numerosae, rectae, rigidae, praelongae, 1,000-1,500 » longae, basi 12-17 , latae, atrae, opacissimae, ad apicem tantum dilutiores et saepe pellu- cidae, sive simplices obtuse attenuatae vel rotundatae, sive den- tibus 2-4 irregularibus erectis acutis brevissimis vel elongatis usque 15 » longis praeditae; perithecia pauca in quaque plagula, globosa, 200-250 » diam.; asci 2- ad 3-spori; sporae oblongo- ellipsoideae, utrinque rotundatae, centro crassiores, atro-brun- neae, subopacae, 4-septatae, ad septa non vel vix constrictae, 48-51 » longae, 22-24 ,» latae. Amboina, Hitoe messen, Reliquiae Robinsonianae 2078, Octo- ber 18, 1913, on leaves of Kibara moluccana. 136 The Philippine Journal of Science 1922 MELIOLA ODONTOCHAETA Syd. sp. nov. Amphigena vel caulicola, plagulas aterrimas primitus orbi- culares 2-4 mm diam. dein irregulares confluentesque formans; mycelium reticulato-ramosum, ex hyphis rectiusculis vel leviter ' undulatis brunneis 7-8 » crassis compositum; hyphopodia capi- tata copiosissima, alternantia, 22-28 » longa, integra vel subin- tegra, cellula superiore oblonga recta vel curvata 10-13 , lata, inferiore breviori stipitiformi; hyphopodia mucronata opposita, rara, 20-26 » longa, basi 8-10 » lata; setae myceliales copiosae, rectae vel leviter curvatae, 600-800 » longae, basi 10-12 1 latae, tota longitudine opacae, ad apicem dentibus 2-6 mox brevis- simis et 1-2 » tantum longis, mox elongatis usque 12 » longis erectis praeditae, paucae subinde ad apicem obtusae haud den- ticulatae tuncque dilutiores immixtae; perithecia globosa, in sicco collapsa, 200-220 » diam.; asci 2- ad 3-spori; sporae oblongae, utrinque rotundatae, brunneas, 4-septatae, 44-50 ,» longae, 16-20 yp latae, cellulis mediis latioribus. Amboina, Hitoe lama, Reliquiae Robinsonianae 21 87, October 8, 1913, on leaves of Dischidia sp. MELIOLA CANARII Syd. Amboina, town of Amboina, Reliquiae Robinsonianae 2072, July 18, 1913, on leaves of Canarium commune. MELIOLA CLAVULATA Wint. Amboina, town of Amboina, Reliquiae Robinsonianae 21 29, July 31, 1913, on leaves of Ipomoea pes-caprae. MELIOLA DESMODII Karst. et Roum. Amboina, Hitoe lama, Reliquiae Ravinaonsenis 2256, Novem- ber 1, 1913, on leaves of Desmodium gangeticum; Ayer putri, Reliquiae Robinsonianae 2180, vices 28, 1913, on leaves of: Des- modium umbellatum. MELIOLA MANGIFERAE Earle. Amboina, Hitoe messen, Reliquiae Robinsonianae 2118, No- vember 5, 1913, on leaves of Mangifera indica. MELIOLA PELLICULOSA Syd. Amboina, Paso, Reliquiae Robinsonianae 21 85, October 31, 1913, on leaves of Lumnitzera racemosa. ; MELIOLA SAKAWENSIS P. Henn. _ Amboina, Soja, Reliquiae Robinsonianae 2124, genet 4, 1913, on leaves of Clerodendron speciocissimum. 21,2 Sydow: Amboina Fungi Collected by Robinson 137 APPENDICULELLA Von Hoehnel APPENDICULELLA CALOSTROMA (Desm.) v. Hoehn. Amboina, town of Amboina, Reliquiae Robinsonianae 2415, July 23, 1913, on leaves of Rubus fraxinifolius. PARODIELLA Spegazzini PARODIELLA PARAGUAYENSIS Speg. Amboina, Batoe mera, Reliquiae Robinsonianae 2184, July 20, 1913, on leaves of Desmodium triflorum. DIATHRYPTON Sydow genus novum Mycelium superficiale, copiose evolutum, ramosum, septatum, hyphopodiatum. Perithecia ad hyphas mycelii orta, ovato-glo- bosa, astoma, cellulosa, mollia, mox et facile histolyse mucosa dissoluta, monoascigera. Asci aparaphysati, octospori. Sporae bicellulares, coloratae. DIATHRYPTON AMBOINENSE Syd. sp. nov. Amphigenum, praecipue epiphyllum, plagulas tenerrimas atro- griseas variae dimensionis formans;-mycelium copiose evolutum, ex hyphis rectis vel rectiusculis copiose et saepe opposite ramosis fuscis 4-5 » crassis septatis (articulis variae longitudinis 8-25 » longis) compositum; hyphopodia numerosa, unilateralia vel Saepius et praecipue ad hyphas primarias crassiores opposita, continua, lageniformia, 12-15 » longa, basi 5-7 » crassa; peri- thecia dense gregaria, ad hyphas mycelii orta, ovato-globosa, astoma, 45-60 » alta, mollia, tenuiter cellulosa, pariete mox et facile histolysis ope in cellulas dilute brunneas rotundatas vel angulosas 6-10 » diam. dissoluto, monoascigera; asci ovato- globosi, octospori, aparaphysati, 40-50 » longi, 35-42 ,» lati; sporae ellipsoideo-oblongae, utrinque rotundatae, medio septatae et valde constrictae, 21-26 » longae, 10-13 » latae, loculis aequa- libus vel supero subinde paullo latiore, facillime secedentibus. Amboina, Hatalai, Reliquiae Robinsonianae 2092, October 24, 1913, on leaves of Schuurmansia elegans. The new genus is closely related to Phaeoschiffnerula Theiss. from which it differs by the monoascigerous perithecia. These originate from the hyphopodia as is the case with Ph. composi- tarum Theiss.2. We have seen a few isolated brown conidia, first two-celled, and later four-celled, measuring about 20-26 by 8-9 x. Perhaps these have originated from the mycelium; * Verhandl. Zool. bot. Ges. Wien (1916) 337. 138 The Philippine Journal of Science 1922 this, however, could not be made out with absolute certainty. The brown hyphae of the mycelium are all connected by a very thin, nearly hyaline pellicle in which here and there slender, subhyaline or yellowish hyphae, 1-2 » broad, are to be seen. We have found a similar pellicle in Phaeoschiffnerula. AUERSWALDIA Saccardo AUERSWALDIA EXAMINANS (Mont. et Berk.) Sacc. Amboina, Kati kati, Reliquiae Robinsonianae 2228, October 28, 1913, on bark of decaying tree. PHYLLACHORA Nitschke PHYLLACHORA PHASEOLINA Syd. Amboina, Kati kati, Reliquiae Robinsonianae 21 54, October 6, 1913, on leaves of Vigna sp. CATACAUMA Theissen et Sydow CATACAUMA MICROPLACUM Syd. sp. nov. Stromata hypophylla, irregulariter sparsa vel aggregata, irreg- ulariter rotundata, minutissisma, punctiformia, 0.33-0.5 mm diam., subinde bina confluentia et tunc paullo majora, periphe- rice plana, opaca, centro convexa, unilocularia, clypeo epider- mali, opaco; asci clavati, paraphysati, 45-60 » longi, 11-18 » lati, octospori; sporae 1- ad 2-stichae, ellipsoideae, continuae, hyalinae, 11-13 » longae, 5-6 p latae. Amboina, Koeda mati, Reliquiae Robinsonianae 2257, on leaves of Ficus sp. Related to Catacauma apoense Syd., but differing in the smaller stromata and spores. CATACAUMA ROBINSONII Syd. sp nov. Stromata epiphylla, sparsa, plerumque orbicularia, 4-8 mm diam., opace atra, vesiculose convexa, ostiolis nitidis, multilo- cularia, haud raro stromate hypophyllo opposito sterili prae- sente; clypeo epidermali, opaco, 30-40 » crasso; asci paraphy- sati, clavati, 50-80 » longi, 11-18 » lati, octospori; sporae mono- stichae vel distichae, ellipsoideae vel oblongae, continuae, hyali- nae, 13-18 yw longae, 7-8 » latae. Amboina, Binting, Reliquiae Robinsonianae 2242, August 18, 1913, on leaves of Ficus sp. This form, which is closely related to Catacauma infectorium (Cke.) Theiss. et Syd., is characterized by the regular stromata and the comparatively large spores. 21,2 Sydow: Amboina Fungi Collected by Robinson 139 TRABUTIA Saccardo et Roumeguére TRABUTIA AMBOINENSIS Syd. sp. nov. Stromata epiphylla, valde variabilia et irregularia, minora 2mm tantum diam., majora usque 1 cm longa vel lata, plerum- que in greges usque 2 cm longos irregulariter disposita et con- fluentia, opace atra, vesiculose convexa, ostiolis nitidis, clypeo opaco ca. 30—40 » crasso; loculi copiosi, 350-420 p lati, 150-250 » alti, pariete brunneolo ca. 20 pw Crasso; asci clavati, 60-80 » longi, 11-20 ,» lati, paraphysati, octospori; sporae mono- vel distichae, ellipsoideo-oblongae, continuae, hyalinae, 17-18 , lon- gae, 8-9 u» latae. : Amboina, Koesoekoesoe sereh, Reliquiae Robinsonianae 2096, October 8, 1913, on leaves of Ficus hasskarlit. This species must be compared with Trabutia Butleri Theiss. et Syd. and 7. Elmeri Theiss. et Syd. from which it differs but slightly in the form and arrangement of the very irregular stromata. HYSTEROSTOMELLA Spegazzini HYSTEROSTOM ELLA TETRACERAE (Rud.) v. Hoehn. Amboina, Amahoesoe, Religuiae Robinsonianae 2212, Septem- ber 16, 1913, on leaves of Delima (Tetracera) Boerlagei. ARMATELLA Theissen et Sydow ARMATELLA LITSEAE (P. Henn.) Theiss. et Syd. Amboina, Hitoe messen, Reliquiae Robinsonianae 21 12, No- vember 1, 1913, on leaves of Litsea sp.; Soja and Hatalai, Reli- quiae Robinsonianae 2091, 2179, October 24, 1913, on leaves of Neolitsea amboinensis Merr. i AULACOSTROMA Sydow AULACOSTROMA PANDANI (Rostr.) Syd. Amboina, town of Amboina, Reliquiae Robinsonianae 2098, #258, July 19, 1913, on leaves of Pandanus; Gelala, Reliquiae Robtnsonianae 2107, September 19, 1913, on leaves of Pandanus humilis. ACTINODOTHIS Sydow ACTINODOTHIS PIPERIS Syd. Amboina, Gelala, Reliquiae Robinsonianae 2208, July 16, 1913, on leaves of Piper sp.; Ayer putri, Reliquiae Robinsonianae 2235, July 29, 1913, on leaves of Piper sarmentosum. 140 The Philippine Journal of Science _ 1922 ASTERINA Léveillé _ASTERINA SPECTABILIS Syd. sp. nov. Amphigena, praecipue epiphylla, plagulas primitus orbiculares 2-4 mm latas dein subinde confluendo majores formans; myce- lium copiose evolutum, ex hyphis obscure castaneis rectis op- posite ramosis 8-11 » crassis septatis (articulis 12-30 » longis) compositum, hyphopodia copiosa, fere semper opposita, saepe longa serie regulariter disposita continua, crasse cylindracea, obtusa, integra, 10-20 » longa, 9-12 pw lata; thyriothecia grega- ria, rotundata, 200-300 » diam., ex hyphis rectis 3 ,» crassis septatis (articulis 10-16 » longis) obscure brunneis contexta; asci ovato-globosi, aparaphysati, 50-80 ,» longi, 40-50 » lati, octospori, pauci in quoque thyriothecio; sporae oblongo-ellipsoi- deae, utrinque rotundatae, 1-septatae, ad septum constrictae, brunneae, leves, 35-42 » longae, 18-20 » latae, loculis aequalibus vel supero paullo latiore; pycnidia simul praesentia minora, ro- tundata, 90-140 » diam., conidia ellipsoidea vel ovata, continua, obscure castaneo-brunnea, subopaca, 24-28 w longa, 18-20 » lata. Amboina, Ayer putri, Reliquiae Robinsonianae 2126, July 28, 1913, on leaves of Flacourtia inermis. ASTERINA VENUSTULA Syd. sp. nov. Epiphylla, plagulas orbiculares tenues 3-5 mm diam. formans; mycelium tenue, matrici dense adpressum, ex hyphis fuligineis laxe ramosis leviter undulatis vel subrectig 3-4 » crassis remote septatis. (articulis 15-30 » longis) compositum; hyphopodia sat numerosa, unilateralia vel opposita, cylindracea, 10-15 » longa, 44.5 » lata, integra vel subintegra, continua, recta vel leviter curvata; thyriothecia dense gregaria, orbicularia, 150-175 p diam., ambitu copiose undulato-fimbriata, ex hyphis radiantibus fuscis ca. 3 » crassis strato singulo contexta, stellatim dehis- centia; asci ovato-globosi, aparaphysati, octospori, 32—40 » longi, 25-30 » lati; sporae oblongae, leves, brunneae, 20-23 » longae, 8-9 » latae, 1-septatae, loculis fere aequalibus. ‘ Amboina, town of Amboina, Reliquiae Robinsonianae 2088, 2252, July 18, 1913, on leaves of Averrhoa bilimbi L. ASTERINA ASSIMILIS Syd. sp. nov. Epiphylla, plagulas orbiculares 2-5 mm latas tenues formans; mycelium ex hyphis rectangulariter ramosis septatis (articulis 15-20 » longis) 5-7 » crassis obscure castaneo-brunneis com- positum; hyphopodia modice copiosa, dispersa, alternantia vel unilateralia, continua, crasse cylindracea, 10-17 p longa, 7-9 » ‘21,2 Sydow: Amboina Fungi Collected by Robinson 141 lata; thyriothecia gregaria, orbicularia, 200-300 » diam., radia- tim ex hyphis rectis ca. 3 » crassis Septatis (articulis 5-10 » longis) brunneis contexta; asci ovati vel oblongi, 60-80 » longi, 30-40 » lati, octospori, aparaphysati; sporae ovato-ellipsoideae vel ovato-oblongae, infra medium septatae, hinc loculis conspicue inaequalibus, vix vel leviter constrictae, fuligineae, leves, 24-26 m longae, 13-15 ,» latae. Amboina, Kati kati, Reliquiae Robinsonianae 2157, October 6, 1913, on leaves of Eugenia caryophyllata Thunb. This species is closely related. to Asterina ditissima Syd. from which it differs in the longer, cylindric, and less-numerous hy- phopodia, and in the conspicuously unequal-septate spores. ASTERINA DIAPHORELLA Syd. Amboina, Wae, Reliquiae Robinsonianae 2147, November 26, 1913, on leaves of Sideroxylon sp. aff. attenuatum. ASTERINA PUSILLA Syd. Amboina, Paso, Reliquiae Robinsonianae 2181, October Si, 1913, on leaves of Premna obtusifolia. ASTERINA SPONIAE Racib. Amboina, near the town of Amboina, Reliquiae Robinsonianae 2161, July 27, 1913, on leaves of Trema sp.; Hitoe messen, Re- liquiae Robinsonianae 2151, November 5, 1913, on leaves of Trema amboinensis. PRILLIEUXINA Arnaud PRILLIEUXINA LORANTHI Syd. nom. nov. Asterinella loranthi Syd. in Philip. Journ. Sci. 8 (1913) Bot. 490. (Type from Leyte, Bur. Sci. No. 15248a, on leaves of Loranthus leytensis.) Amboina, Koesoekoesoe sereh, Reliquiae Robinsonianae 2247, August 23, 1913, on leaves of Loranthus rumphii. PRILLIEUXINA MICROSPILA Syd. sp. nov. Epiphylla, plagulas minutas tenues orbiculares 1-3 mm diam. formans; mycelium ex hyphis irregularibus fuscidulis ramosis anastomosantibus septatis 3—4 » crassis haud hyphopodiatis com- positum ; thyriothecia dense gregaria, copiosa, orbicularia, 150- 200 » diam., e strato singulo hypharum subrectarum fuscidu- larum 2-3 » crassarum septatarum (articulis 4-6 ,» longis) contexta; asci ovato-globosi, aparaphysati, octospori, 35-50 » longi, 25-40 » lati; sporae oblongae, utrinque late rotundatae, fuscidulae, leves, medio septataé et valde constrictae, 21-24 B longae, 9-10 » latae, loculis aequalibus vel supero leviter latiore. 142 The Philippine Journal of Science 1922" Amboina, town of Amboina, Reliquiae Robinsonianae 2214, August 20, 1913, on leaves of Leucosyke capitellata. PRILLIEUXINA AMBOINENSIS Syd. sp. nov. Hypophylla, rarius etiam epiphylla, plagulas orbiculares vel irregulares 2-5 mm diam. formans; mycelium copiose evolutum, radians, ex hyphis septatis ramosis vel anastomosantibus obscure fuscis 4-6 » crassis torulosis haud hyphopodiatis sed hinc inde ramulos breves hyphopodia simulantes emittentibus compositum ; thyriothecia sat numerosa, gregaria, rotundata, 200-400 » diam., ambitu fimbriata, opaca, subatra, ex hyphis ca 3 » latis radiatim contexta, stellatim dehiscentia; asci ovati vel subglobosi, apara- physati, octospori, 45-60 » longi, 25-45 » lati; sporae oblongae, utrinque rotundatae, fuscae, medio septatae et valde constrictae, leves, 26-35 » longae, 13-15 » latae, loculis fere aequalibus vel supero leviter latiore. Amboina, Hoetoemoeri Road, Reliquiae Robinsonianae 2174, September 30, 1913, on leaves of Cordyline terminalis. The new species is most nearly related to the Philippine Aste- rinella calami Syd. from which it differs in the characters of the mycelium and in its smaller spores. PARASTERINA Theissen et Sydow PARASTERINA MELANOTES Syd. sp. nov. Epiphylla, plagulas orbiculares conspicuas 4-8 mm diam. for- mans; mycelium ex hyphis rectiusculis intertextis ramosis sep- tatis 8-10 » crassis atro-brunneis compositum; hyphopodia sparsa, solitaria, hemisphaerica, ca. 10 » alta et lata; thyriothecia dense gregaria, irregulariter rotundata, 400-700 » diam., ater- rima, omnino opaca, centro valde elevata, radiatim contexta, peripherice hyphis obscure castaneo-brunneis 6-8 p latis plus minus longe fimbriata; asci copiose evoluti, paraphysati, ovato- globosi, octospori, 60-85 » longi, 50-60 » lati; sporae oblongae, utrinque late rotundatae, leves, fuscidulae, 35-40 » longae, 17-19 p latae, loculis aequalibus vel supero paullo latiore. Amboina, Hitoe messen, Reliquiae Robinsonianae 2203, Oc- tober 13, 1913, on leaves of Polyosma brachyantha. CALOTHYRIUM Theissen CALOTHYRIUM VILE Syd. sp, nov. Thyriothecia nudo oculo aegre perspicua, epiphylla, in greges effusos laxe disposita, rotundata, 150-300 » diam., radiatim ex hyphis flavo-fuscidulis 2-3 » latis contexta, mycelio libero per- parco, ex hyphis undulatis vel torulosis 3—4 » crassis haud hypho- podiatis composito, hymenio plano simplici pluriascigero; asci 21,2 Sydow: Amboina Fungi Collected by Robinson 148 clavati vel clavato-saccati, octospori, aparaphysati, 50-80 » longi, 18-22 » lati; sporae oblique monostichae vel distichae, obpiri- formes vel elongato-obovatae, hyalinae, 1-septatae, non vel leviter constrictae, 18-24 » longae, 9-12 » latae, loculo supero majore. Amboina, Hitoe messen, Reliquiae Robinsonianae 2102, Octo- ber 14, 19138, on leaves of Phoebe macrophylla. LEMBOSIA Léveillé LEMBOSIA ROBINSONII Syd. sp. nov. Epiphylla, plagulas irregulares usque 1 cm diam. vel confluendo majores formans; mycelium ex hyphis rectiusculis 7-10 » crassis obscure castaneo-brunneis irregulariter reticulato-ramosis sep- tatis compositum; hyphopodia sparsa, alternantia, cylindrico- elevata, continua, integra, 12-17 » longa, 8-10 » lata; thyrio- thecia laxe gregaria, primitus rotundata et 300-500 ,» diam., mox autem exacte linearia et usque 1.25 mm longa et 200-300 » lata, centro elevata, opaca, carbonacea, ambitu fimbriata, radia- tim ex hyphis 3-4 » crassis contexta; asci ovato-globosi, 70-90 » longi, 35-55 yw lati, octospori; paraphyses distinctae, copiosae, ca. 2 » crassae; sporae oblongo-ellipsoideae, utrinque rotundatae, leves, fuligirieae, leviter constrictae, 34-38 » longae, 17-19 » latae, loculis aequalibus vel fere aequalibus. Amboina, Hitoe messen, Reliquiae Robinsonianae 2188 (type), 2158, both collected November 6, 1913, on leaves of Eugenia sp. The species differs from Lembosia eugeniae Rehm in the char- acters of the mycelium, hyphopodia, and smaller spores. It is more nearly related to Parasterina pemphidioides (Cke.) Theiss., but differs notably in the exactly lembosoid form of the thyrio- thecia and the one-celled hyphopodia. ECHIDNODES Theissen et Sydow ECHIDNODES XENOSPILA Syd. sp. nov. Epiphylla, partibus matricis atro-decoloratis et peculiariter quasi dendritice effiguratis usque 1 cm latis insidens; mycelium liberum parum evolutum, ex hyphis brunneolis ramosis anasto- mosantibus ca. 3 » crassis fuscis haud hyphopodiatis composi- tum; thyriothecia laxe disposita, plerumque recta, linearia, 350- 450 p» longa, 120-180 » lata, rima longitudinali dehiscentia, con- textu atro opaco radiato; asci ovati vel ovato-globosi, 35-45 » longi, 30-35 ,» lati, octospori; paraphyses haud numerosae, ad ‘apicem leviter incrassatae, sporae conglobatae, oblongae, leves, ex hyalino fuscae, 1-septatae, vix vel parum constrictae, utrinque rotundatae, 22-25 » longae, 9-10 » latae, cellula superiore ple- rumque leviter crassiore sed breviore. 144 The Philippine Journal of Science 1922 Amboina, Hitoe messen, Reliquiae Robinsonianae 2134, Novem- ber 5, 1913, on leaves of Fagraea speciosa, growing together with Meliola fagraeae Syd. THYROSOMA Sydow genus novum Mycelium nullum. Thyriothecia superficialia, brunnea, ex hyphis radiantibus contexta, membrana basali distincta nulla, hymenia multa polyascigera includentia, polyostiolata. Asci in plectenchymate hyalino fibroso dein evanescente siti, sessiles, octospori, aparaphysati. Sporae hyalodidymae. THYROSOMA PULCHELLUM Syd. sp. nov. Thyriothecia amphigena, praecipue hypophylla, sparsa, soli- taria, ambitu orbicularia, 1-1.5 mm diam., centro 80-100 p» crassa, plana, superficialia, centro opaca et pluristratosa, ad peripheriam tantum pellucida, ex hyphis amoene radiantibus rectis vel rectiusculis ca. 2 » crassis peripherice brunneolis cen- trum versus obscurioribus septatis (articulis 8-14 ,» longis, ad peripheriam longioribus) composita; asci sessiles, ventricosi, saccati vel elongati, 40-55 » longi, 15-18 » lati; sporae oblongae, hyalinae, circa medium 1-septatae, non constrictae, 16-18 p longae, 5-6 yp latae, cellula superiore late rotundata paullo bre- viore et latiore, cellula inferiore paullo longiore sed angustiore. Amboina, Way tommo, Reliquiae Robinsonianae 2146, 2057, November 26 and August 17, 1918, on leaves of Hrythroxylum ecarinatum. EREMOTHECA Theissen et Sydow EREMOTHECA PHILIPPINENSIS Syd. Amboina, Hitoe, Reliquiae Robinsonianae 2175, October 8, 1913, on leaves of Aporosa sphaeridophora. BYSSOGENE Sydow genus novum Ascomata in hypothallo ex hyphis asperatis brunneis ramo- sis composito superficialiter insidentia, atra, in maturitate patelliformia, extus ad basim hyphulis obsessa. Loculi uni- stratosi, in parenchymate minute celluloso siti. Asci solitarii, ovati vel oblongi, aparaphysati, octospori. Sponae muriformes, hyalinae. ; BYSSOGENE AMBOINENSIS Syd. sp. nov. Hypophylla, plagulas primitus orbiculares 3-10 mm diam. tandem saepe confluentes tuncque irregulares et majores fumoso- brunneas formans; mycelium copiosissime evolutum, densum, ex hyphis brunneis vel opace brunneis ramosis remote septatis 4-5 » crassis densissime asperatis compositum; ascomata in 21,2 Sydow: Amboina Fungi Collected by Robinson 145 mycelio superficialia, sparsa, primitus globoso-clausa, in matu- ritate patelliformia, margine elevato, atra, usque 1 mm lata, extus rugosa et praecipue ad basim hyphulis obsessa, parenchy- matica, in superiore parte minute flavo-brunnee vel brunnee parenchymatica, in inferiore parte grossius et obscurius atre vel subatre parenchymatica, cellulis 5-7 » diam.; loculi uni- stratosi, In superiore stromatum parte siti; asci ovati, oblongi vel subclavati, antice crasse tunicati, aparaphysati, 40-52 »« longi, 18-25 » lati, octospori; sporae elongato-obovatae, superne late rotundatae, basim versus leviter attenuatae, sed obtusae, transverse 3-septatae non vel vix constrictae, loculis 2-3 vel omnibus septo longitudinali iterum divisis, hyalinae, 15-17 ,» longae, 8-10 » latae. Amboina, Batoe mera, Reliquiae Robinsonianae 2105, Septem- ber 24, 1913; Ayer putri, Reliquiae Robinsonianae 2170, July 29, 1913; both on leaves of Hugenia sp. The new genus belongs to the Saccardiaceae; it is somewhat related to Calopeziza Syd., but differs especially in the black ascomata and colorless spores. CALLORIOPSIS Sydow CALLORIOPSIS GELATINOSA (Ell, et Mart.) Syd. Amboina, town of Amboina, Reliquiae Robinsonianae 2156, July 28, 1913, parasitic on Meliola on leaves of Cerbera manghas (C. odollam). PILOCRATERA P. Hennings PILOCRATERA HINDSI! (Berk.) P. Henn. Amboina, Lateri, Reliquiae Robinsonianae 2222, September 5, 1913, on dead wood. FUNGI IMPERFECTI DIPLODIA Fries DIPLODIA FRUCTUS-PANDANI P. Henn. Amboina, locality not given, Reliquiae Robinsonianae 2109, on fruits of Pandanus dubius. : ASTEROSTOMELLA Spegazzini ASTEROSTOMELLA POLYSTIGMA Syd. sp. nov. Epiphylla, plagulas primitus minutas et discretas mox autem confluentes et totam folii superficiem plus minus occupantes formans; mycelium ex hyphis rectiusculis sed dense ramosis 3-4 » crassis septatis fuscidulis compositum ; hyphopodia modice copiosa, continua, sessilia, valde et varie lobata, plerumque latiora quam altiora, 9-11 u lata, 7-8 » alta; thyriothecia copio- © 2 146 The Philippine Journal of Science sissima, dense gregaria, punctiformia, 60-85 » diam., plerumque rotundata, radiatim ex hyphis 2.5-3 » crassis rectiusculis crebre septatis fuscidulis contexta, stellatim dehiscentia; conidia con- tinua, fusca, ovata vel piriformia, 14-17 » longa, 10-13 4 lata. Amboina, town of Amboina, Reliquiae Robinsonianae 2167, August 23, 1913, on leaves of Hemigraphis (vel Peristrophe?) sp. GLOEOSPORIUM Desmaziéres et Montagne GLOEOSPORIUM ALCHORNEAE Syd. Amboina, Hitoe lama, Reliquiae Robisonianae 2143, October 8, 1913; Amahoesoe, Reliquiae Robisonianae 2237, September 18, 1913; both on leaves of Alchornea rugosa. PESTALOZZIA De Notaris PESTALOZZIA PALMARUM Cke, Amboina, Batoe mera, Reliquiae Robinsonianae 2065, July 20, 1913, on drying leaves of Cocos nucifera. CLADOSPORIUM Link CLADOSPORIUM ZIZYPHI Karst, et Roum. Amboina, Binting, Reliquiae Robinsonianae 2145, November 18, 1913, on leaves of Zizyphus jujuba. HELMINTHOSPORIUM Link HELMINTHOSPORIUM FICINUM Sace. Amboina, Soja, Reliquiae Robinsonianae 2204, 2284, August 31 and August 2, 1913; Lateri, Reliquiae Robinsonianae 2114, September 5, 1913; Wae, Reliquiae Robinsonianae 2136, No- vember 27, 1913; all on leaves of Ficus wassa. APPENDIX The following four numbers were collected by the late Doctor Robinson on the same trip, before reaching Amboina. MELIOLA SAKAWENSIS P. Henn. Boeton, Baoebaoe, Reliquiae Robinsonianae 2509, July 18, 1913, on leaves of Clerodendron (minahassae?). Celebes, Macassar, Keliquiae Robinsonianae 2466, July 41, 1913, -on leaves of Clerodendron sp. PHYLLACHORA CATERVARIA (Berk.) Sacc. Boeton, Baoebaoe, Reliquiae Robinsonianae 2507, July 18, 1913, on leaves of Ficus sp. PARASTERINA PEMPHIDIOIDES (Cke.) Theiss. Celebes, Macassar, Reliquiae Robinsonianae 2468, July 11, 1913, on leaves of Eugenia jambolana. PHILIPPINE MANTIDS, OR PRAYING INSECTS By F. WERNER Of Vienna University, Vienna, Austria ONE PLATE Through the kindness of my friend Dr. H. H. Karny, I was enabled to examine a small but highly interesting collection of mantids, consisting of sixty specimens, collected by Prof. Charles Fuller Baker, of the College of Agriculture, Los Bafios, Philip- pine Islands. For this courtesy I am much indebted to Profes- sor Baker, and hope he will further devote a part of his time to the collection of mantids, as our knowledge of the Philippine species seems still very far from perfect, as can easily be seen when not less than six species out of twenty prove to be new to science, Under the circumstances it is much too early to comment upon the geographical distribution of Philippine mantids; still many years of study will be required before we can say that we know the mantid fauna of Luzon alone. However, I hope that resident naturalists, seeing the remarkable effect of present collecting, will not cease to give attention to this interesting group of insects. In the systematic arrangement of species I have followed Giglio-Tos, whose monographic works on mantids, based on very rich material, must be regarded as the foundation for future systematic work on the insects of this family. It is significant of the paucity of Philippine material in European collections that, among the immense numbers of new species described by Giglio-Tos, hardly a half dozen are from the Philippine Islands. EREMIAPHILINA: Metallyticus violaceus Burmeister. Middle and hind femora dark at apex; abdomen entirely blue below; no red spots on head. The species has already been recorded from the same locality by Hebard.* MINDANAO, Zamboanga, 1 female. "Proc. Acad. Nat. Sci. Philadelphia 72 (1920) 18. 1878972 147 148 The Philippine J ournal of Science 1922 IRIDOPTERYGIN4= Tropidomantis tenera Stal. Eng. Resa Ins. (1858) 314. The Philippine specimens agree in every respect with speci- mens from Java. Recorded from the Philippine Islands (Zam- boanga) by Hebard.? PALAWAN, Puerto Princesa, male and female. MINDANAO, Davao, 1 female. ‘AMELIN Hystricomantis g. nov. aspera (Stal). Plate 1, fig. 1. Gonypeta aspera STAL, Gfvers. Kongl. Vetensk. Ak. Forh. Stockholm No. 10 (1877) 38. I have no doubt that this is the species very briefly described by Stal. Of the four conical tubercles on vertex the outer ones are much stronger than the median pair; besides these there are two prominent bicuspid tubercles between. the bases of antenne, and likewise a strong tricuspid tubercle near their level at inner margin of eyes. Pronotum strongly keeled, with abrupt rhom- boidal dilatation in the middle, spinose at the margins, the median keel and the short keels of the prozone diverging cephalad from the transverse sulcus; anterior cox with strong spinose outer and lower keel; anterior femora distinctly longer than cox, with concave upper margins, flat outside, with blunt longitudinal keel ; tibia short, with two outer and four inner spines, the latter increasing in length toward apex of tibia, the three outer curved; tarsus about twice as long as tibia; middle and hind legs very long and thin; wings reaching far beyond apex of abdomen, hyaline, somewhat infumated, elytra with very narrow costal area, apex washed with brown; some of the veins with dark brown points. The male seems to have been undescribed. Due to the many striking features of head, thorax, and forelegs, the species deserves to rank as representing a distinct genus of the Gonypetinz, for which I propose the name Hystricomantis. It lacks the ciliation of the costal margin of forewing and the fine denticulation of anterior femora; the antenne, how- ever, are distinctly ciliated. It is not related to the genus Myrcinus as Giglio-Tos suggested, but is related to the Oligony- cine (Haania) .* “Up. cit. 21. * According to M. Hebard the insect here described is not Gonypeta aspera Stal, but a member of the genus Haania and possibly philippina (Giglio-Tos). It should therefore be listed under Oligonycinz.—KARNY. 21,2 Werner: Philippine Mantids, or Praying Insects 149 Total length, 20.5 millimeters; pronotum, 5; elytra, 17; ante- rior femora, 6.5. LUZON, Mount Maquiling, 1 male. Hystricomantis dispar sp. nov. Plate 1, fig. 2. This nice little mantid differs, like the preceding, from all typical Gonypetine by having the antennz but not the anterior wings ciliated and the anterior femora not finely denticulated at the base of the larger spines; head broad, with prominent, round eyes and rectangular frontal scute, the upper side somewhat arched; vertex rounded, with arched upper margin; pronotum short, smooth, distinctly dilated behind the transverse sulcus, metazone about one and a half times as long as prozone, lateral denticulation very fine; abdomen somewhat broader toward end in the female, parallel-sided in the male; anterior cox smooth, with very fine spines, the interspace between the two larger ones filled with two very small ones; anterior femora strong, upper margins straight, lower angular; five outer spines, four discoidal spines; tibia strong, with ten outer and twelve inner Spines; metatarsus as long as tibia; wings reaching far beyond end of abdomen, perfectly hyaline in the male (in the female the ‘forewings are feebly infumated), costal part strongest; of the hind wings only the apex is brownish, with some darker points. Transverse veins in costal area oblique, nearly parallel, rarely divided; in the discoidal area wide-meshed, many of the trans- verse veins interrupted, the meshes nearly rectangular; body _bright yellowish brown; unicolored in the male, whereas in the female the anterior coxa has a dark reddish brown, transverse band at apex and also one on the anterior femur; basal part of the latter with dark reddish brown longitudinal streaks reach- ing to discoidal spines; all spines of anterior femora dark at their points. Length, male, 16 millimeters; female, 17; pronotum, male, 3.5; female, 4; elytra, male, 15; female, 18. ° LUZON, Mount Maquiling, male; Mount Banahao, female: Tayabas Province, Malinao, female. MINDANAO, Davao, male. According to M. Hebard this species is the same as Amantis aeta Hebard.t According to Karny the female is more inten- Sively colored than the allotype and the color pattern more extensive, as shown by his fig. 2 a. As in the meantime I have had occasion to study Hebard’s papers here mentioned, I fully agree with this statement. *Op. cit. 31. 150 The Philippine Journal of Science 1922 Gonypeta borneana Giglio-Tos. Boll. Soc. Ent. Ital. 46 (1914) Florence (1915) 155. Compared with specimens of G. punctata as defined by Giglio- Tos, from Java and Sumatra, the Philippine specimens distinctly show the specific differences. The granules on pronotum and outer portion of anterior femora are very strong and the color- ation is distinctly darker, in one specimen nearly black; only the dimensions are somewhat smaller than indicated by Giglio- Tos. MINDANAO, Zamboanga, 2 males. COMPSOMANTIN 4 Compsomantis tumidiceps (Bolivar). Bolivar, An. Soc. Esp. Hist. Nat. 19 (1890) 303 (Compsomantis) ; Kirby, Ann. & Mag. Nat. Hist. 13 (1904) 83 (Humbertiella ? brunneri) ; Werner, Verh. Zool. bot. Ges. Wien. (1916) 258 (Hapa- lomantis) ; Giglio-Tos, Boll. Soc. Ent. Ital. 46 (1914) 86; Hebard, Proc. Acad. Nat. Sci. Philadelphia 72 (1920) 36. This species was first described from the Philippines, but occurs also on Lombok and Sumbawa Islands (Kirby, Werner). The present specimen differs slightly from a female from Lom- bok, in my collection, by a somewhat longer pronotum, probably owing to its sex. There is no doubt that Giglio-Tos is perfectly right in uniting this species with Compsomantis crassiceps De Haan and Hapalomantis rufula Westwood in the same group; the union seems very natural. LuzON, Mount Maquiling, 1 specimen (probably a mee end of abdomen missing). THESPIN 2 Euchomenella heteroptera (De Haan). Bijdr. Kenn. Orth. p. 78, pl. 18, fig. 1; Saussure, Mem. Soc. Gen. 21 (1871) 48; Giglio-Tos, Boll. Soc. Ent. Ital. 47 (1915) 34, 365. This species seems not to have been found previously in the Philippine Islands, where only the rather closely related Tagalomantis manillensis Sauss. is known. It is easily dis- tinguished from that species by its unicolorous, infumated wings. The specimen agrees perfectly with De Haan’s figure, whereas the species figured as the female belongs, as Giglio-Tos has already pointed out, to quite another genus, called Mythomantis by that author. I have a specimen of the latter insect, M. confusa Westwood, from Java, that agrees in every respect with De Haan’s figure. LUZON, Tayabas Province, Malinao, 1 mai. 21,2 Werner: Philippine Mantids, or Praying Insects 151 CALIRIDIN A Leptomantis lactea (Saussure). Miopteryx lactea. SAUSSURE, Mem, Soc. Gen. 21 (1871) 125. LUZON, Laguna Province, Mount Maquiling, 1 female. Aetella bakeri Hebard. Proc. Acad. Nat. Sci. Philadelphia 72 (1920) 45. I had at first confounded this species with Leptomantis albella Burm.; but, after having received Hebard’s excellent paper, I am convinced that I had before me the new species described by him as representing a new genus. This species cannot be rare and, in any case, it appears to be widely distri- buted over the Philippines. It is easily distinguishable from Leptomantis lactea by the smaller size and the perfectly hyaline, iridescent wings, whereas that species has distinctly milky wings with white costal margin on elytre. A pair of dark spots, apparently common in L. lactea, is not frequently to be found in Aetella bakeri on the metazone of the pronotum, as far behind the transverse sulcus as the length of the prozone. LUZON, Laguna Province, Mount Maquiling, 1 female, and Los Bafios. BASILAN, 1 male. MINDANAO, Zamboanga, 1 female: Butuan and Dapitan. DEROPLATINA Deropletys philippinica sp. nov. Distinguished from all other members of the genus by the narrower pronotum, resembling somewhat in shape that of Sphodrepoda medioconstricta Wood-Mason. Prozone of prono- tum one-third the length of metazone, the latter with a longitudi- nal median keel in its posterior half. The expanded part of the prozone is not more than half the transverse diameter of the prozone itself, narrowed cephalad and with rounded anterior margin; in its anterior third the metazone is parallel-sided, widens toward its second third, and is again narrowed to its posterior end, thus forming a sinuous line on each side; wings extending much beyond apex of abdomen, the anterior ones feebly dilated toward apex, which is bluntly triangular; costal area dilated only at base, the greater portion toward the apex very narrow; stigma distinct, small, triangular; hind wings with the costal area angularly produced (more so than the fore- wings) ; middle and hind legs very slender, with small, trian- gular, somewhat denticulated lobules near apices of tibia; ex- posed parts dead-leaf-colored above and below; inner face of 152 The Philippine Journal of Science 1922 anterior femora with a large blackish spot before and some- what behind the unguicular sulcus, and three dark vertical streaks toward apex, hind wings with a light yellowish brown spot before apex which is rather dark brown, like the base, and indistinctly dark spotted; anal area with lighter transverse veins; anterior wings indistinctly mottled with darker and lighter brown; middle and hind legs broadly banded with, dark brown. MINDANAO, Davao, 1 male. Total length, 42 millimeters; pronotum, length, 15; breadth at broadest part of dilatation, 4; forewings, length, 30; breadth, 7. MANTINA Tenodera aridifolia Stoll. Giglio-Tos, Boll. Soc. Ent. Ital. 1911 43 (1912) 37; Hebard, Proc. Acad. Nat. Sci. Philadelphia 72 (1920) 50. ' LUZON, Mount Maquiling, 2 males. MINDANAO, Iligan; Da- vao, 1 female, large specimen, 10 centimeters. The Iligan specimen is somewhat darker brown, with a median longitudinal line on the pronotum. Curiously enough, this widely spread species seems not to have been found before in the Philippines. Tenodera fasciata Olivier = attenuata Stoll. Giglio-Tos, Boll. Soc. Ent. Ital. 1911 43 (1912) 45; Hebard, Proc. Acad. Nat. Sci. Philadelphia 72 (1920) 51. - Apparently new for the Philippines. MINDANAO, Zamboanga, 1 male. Hierodula vitrea (Stoll). Giglio-Tos, Mant. Esot. 5 (1912) 83. Spines of anterior cox very blunt, scarcely distinguishable. Not recorded before from the Philippine Islands. LUZON, Mount Maquiling, 1 male. Hierodula gigliotosi sp. nov. Frontal scutellum pentagonal, with parallel lateral and straight anterior sides and two feeble longitudinal keels, some- what broader than high; pronotum elongate, its metazone more than twice as long as the prozone; lateral margins very feebly denticulated at the dilatation; posterior angles and margin rounded, blackish; organs of flight extending beyond end of body, elytra with opaque, bright green costal and perfectly hyaline discoidal part; hind wings perfectly hyaline; of the internal spines of anterior femora the first (from apex) and fourth are quite black, the sixth likewise black with a black spot 21,2 Werner: Philippine Mantids, or Praying Insects 153 at base; the eighth, tenth, twelth, and fourteenth have only the base black; internal spines of tibize black at their tips, discoidal spines black at their bases; all other parts green; subgenital plate elongate, darker green, Total length, 47 millimeters; pronotum, 16; anterior wings, 32. MINDANAO, 1 male. This species is characterized by the absolute lack of spines on anterior coxe, as is normally the case in the males of the genus Ephierodula Giglio-Tos; otherwise it is a true Hierodula, closely related to H. unimaculata Oliv. from Coromandel, Bombay, and Tonkin. Hierodula patellifera Serv. Giglio-Tos, Boll. Soc. Ent. Ital. 1911 43 (1912) 95. Luzon, Mount Maquiling, a female; Los Bafios, male and female. The Los Bafios female is of the green variety, the two others, of the grayish. The species has already been recorded from the Philippines, from Luzon, by Giglio-Tos. Hebard also records it from various localities in the Philippines. Ephierodula immaculifemorata sp. nov. This fine species differs from both other species of the genus ° by the unicolorous anterior coxe and femora and by the dentic- ulated anterior coxe of ‘the male; but in all other respects it is a true Ephierodula. Frontal scute pentagonal, with rounded anterior edges; pronotum not denticulated laterally; metazone about three times as long as prozone; dilatation strong, well distinguished by a horizontal expansion of its margin, resembling somewhat the still stronger one of Rhomboderula tamolana Brancs * from New Guinea; wings extending much beyond end of abdomen ; forewings with opaque green costal and glossy, hyaline discoidal part; stigma broadly surrounded by dark brown; hind wings hyaline, infumated; anterior cox with eight feeble spines; anterior femora with four outer and fifteen inner spines ; eleven outer and fifteen inner tibial spines. Total length, 65 millimeters; pronotum, 22; elytra, 50. MINDANAO, Butuan, a male. Rhombodera ornatipes sp. nov. Plate 1, fig. 3. This fine new species is one of the group to which Rhombodera - major belongs, characterized by Giglio-Tos as having the meta- "Cf. Werner, Verh. Zool. bot. Ges. Wien (1916) 262 and 263. *Jahresb. Ver. Trencsin. Cour. 19-20 (1897) 62, pl. 1, fig. 8. 154 The Philippine Journal of Science 1922 zone of the pronotum at least two and a half times as long as the prozone, and the broadest part of the pronotum before its middle. It is, however, easily distinguished from all other members of this group by the coloration of the inner side of the anterior cox, which are deep and shining black on their apical half; also, the trochanter and a large spot at the basal hind quarter of the inner side of anterior femora are black, and there are several (about five) small black spots at the base of the inner femoral spines; the large femoral spots contain two whitish ones; the others are green; coxal spines small, irregular ; outer femoral spines four, inner sixteen, the first (from apex), fourth, sixth, an eighth black, with a black spot at their bases; wings reaching to end of abdomen, anterior ones with opaque green costal, and hyaline discoidal areas; stigma small, elongate. Total length, 49 millimeters; length of pronotum, 15; broadest part of pronotum, 6.5; length of forewings, 32. LUZON, Mountain Province, Baguio, a male. _ ARCHIMANTINZ4= Statilia haani (Saussure)? var. major var. nov. Similar to the typical form of this common, widespread species, which has been previously recorded from the Philippine Islands by Saussure and by Brunner, but larger. It is also represented in my collection, by specimens from Samar and Min- doro. One female from Mount Maquiling is much larger and distinctly darker than the other, and the wings do not reach end of body, but otherwise it is not different. Luzon, Laguna Province, Mount Maquiling, 2 males and 2 females; Los Bafios, 1 male. Statilia pallida sp. nov.® This species differs from the preceding (it also has the dark postacetabular prosternal spot) by the uniformly colored inner side of anterior femora. I have examined large numbers of S. haani from Sikkim, Annam, Penang, Perak, Sumatra, Nias, Java, Lombok, Ceram, Formosa, and Japan; and, as the coloration of mantids generally, and of S. haani especially, is fairly constant, the specific separation of these two specimens * Giglio-Tos, Boll. Soc, Ent. Ital. 1911 43 (1912) 6. * According to M. Hebard this and the preceding species are Statilia maculata. Hebard says (in litt.): “The cephalic limb coloration is no specific criterion and is subject to decided variation.” —KArRNY. Having examined more material in the meanwhile, I agree with Hebard’s opinion.— WERNER. 21,2 Werner: Philippine Mantids, or Praying Insects 155. seems to be justifiable. Color pale greenish; size about the» same as the preceding species. LUZON, Laguna Province, Mount Maquiling, a male; Los Bafios, a male. Statilia nemoralis (Saussure). Giglio-Tos, Boll. Soc. Ent. Ital. 1911 43 (1912) 8. Differs from S. haani by the shorter thorax without any trace of dark postacetabular prosternal band, the larger and more mesially situated dark band of anterior coxe (on the base of coxe in S. haani), by the broad dark band in the middle of anterior femora, and by the indistinctly margined stigma. The present specimen has the ale not tessellated, as is the case in var. infuscata Giglio-Tos, PALAWAN, Puerto Princesa, a male. ACROMANTIN A= Acromantis hesione Stal. Ofvers. Kongl. Vetensk. Ak. Forh. No. 10 (1877) 38. Hebard, Proc. Acad. Nat. Sci. Philadelphia 72 (1920) 73. Costal margin of forewings bright green, hind wings yellow- ish, infumated at apex; otherwise pale reddish brown; outside of anterior femora darker, broadly trifasciated. I at first took this species for A. javana Giglio-Tos, poorly described as the above-mentioned species. Besides, the specific name javana is preoccupied by Acromantis javana Sauss., which is an Odontomantis. The name must therefore be changed, and I propose Acromantis lili, in honor of that great student of the Mantide, whose name “Giglio” means “lily.” Total length, male, 22 millimeters; forewings, 17; female, about 27. Luzon, Laguna Province, Los Bafios, male and female. Acromantis luzonica Hebard. Proc. Acad. Nat. Sci. Philadelphia 72 (1920) 71. Resembles much a female of the Neotropical Phyllovates parallela (De Haan) in form and coloration of the wings, the anterior wings being grass green, the costal area bent down at a distinct angle, the discoidal area with an oblique dark spot; posterior wings yellow. Here we have another case of similarity between the Ethiopian Calidomantis savignyi Sauss. and the Neotropical Parastagmatoptera flavoguttata Serv. Luzon, Mountain Province, Baguio, 2 females. 156 The Philippine Journal of Science 1922 *Phyllothelys bakeri sp. nov. Near Phyllothelys westwoodi Wood-Mason, but with the spines of anterior coxze minute, and the lateral denticulation of pronotum very weak; anterior femora less slender and not fas- ciated, lobes of intermediate and posterior femora strongly undulated. Elytra distinctly hyaline on their discoidal portion, veins alternately spotted dark and white, apex with brown spots, as also the hind wings, which are more distinctly infumated than the elytra. It is possible that this species may be identical with P. decipiens Giglio-Tos, but, as the number of coxal spines is greater (fifteen) and the author says nothing about the lobes of the middle and hind legs, the identification remains doubtful. Total length, 39 millimeters; pronotum, 14.5; forewings, 23; anterior coxe, 9; anterior femora, 9.5. This new and interesting species is named after its zealous and successful discoverer, Prof. C. F. Baker. LuzON, Laguna Province, Mount Maquiling, a male. Possibly the males figured by Wood-Mason and Westwood as Phyllothelys westwoodi belong to two different species. The horn of the vertex in the figure by Westwood ® is distinctly shorter, the pronotum laterally densely spinose, the anterior femora slenderer, and the lobes of the hind leg more denticulated. HYMENOPODINA Odontomantis javana (Saussure). Mem. Soc. Gen. 21 (1879) 33, pl. 4, fig. 11. Hebard, Proc. Acad. Nat. Sci. Philadelphia 72 (1920) 77. I can find no difference between these specimens and those from Java in my collection, nor can I detect any reason for separating Odontomantis euphrosyne Stal from this species.’° LuzON, Laguna Province, Mount Maquiling, 1 female: Nueva . Vizcaya Province, Imugan, 1 female. PALAWAN, Puerto Prin- cesa, 1 male. MINDANAO, Davao, 1 female. Creobroter urbana (Fabricius). Saussure, Mem. Soc. Gen. 21 (1871) 144. - Ground color of elytra darker green, the ocelliform spot more yellowish (not reddish). ® Rev. Ins. Fam. Mant., London (1889) pl. 1, fig. 3. M. Hebard says (in litt.): “Werner has apparently not noted the differences which at least warrant geographic racial recognition.” —KARNY. LD 21,2 Werner: Philippine Mantids, or Praying Insects 157 LUZON, Mount Banahao, a female, quite typical: Mount Li- may, a female. I have in my collection specimens from Java, Annam, Tonkin, and Siam, as well as some from Ceylon; from the last-named locality only C. pictipennis Wood-Mason seems to be recorded, and I have seen a specimen from Celebes (leg. Sarasin). I am unable to find any characters distinctive from those of Creobroter meleagris Stal. ILLUSTRATION PLATE 1 Fic. 1. Hystricomantis aspera (Stal). a, head and pronotum, from above; b, head, front view; c, head, from above; d, fore parts, from side. 2. Hystricomantis dispar sp. nov. a, anterior leg of female ; b, an- terior wing of male. 3. Rhombodera ornatipes sp. nov. Antreior leg, inner side. 159 * ADDITIONS TO THE HERPETOLOGICAL FAUNA OF THE PHILIPPINE ISLANDS, I By Epwarp H. TAyLor Of Manila, Philippine Islands SEVEN PLATES New and rare frogs, lizards, and snakes have been found during recent herpetological collecting in parts of the Philippines. Although these collections were made prior to the actual public- ation of my monographs,: the descriptions could not be in- cluded therein because those books were in type. For this reason papers on these new collections will be published in the Philippine Journal of Science, of which the present is the first. The following notes from my field books show where the chief collections were made and collecting dates: NORTHERN LUZON I returned to the Philippine Islands from the United States in 1920 and arrived at Manila on March 30. On April 11, I proceeded to Baguio where I joined Prof. H. Otley Beyer and Dr. Robert L. Schofield. April 12. Specimens of Kaloula rigida sp. nov., Rana luzonensis, and Sphenomorphus jagori were collected near Baguio. April 13. Our party left Baguio for the north, going by automobile to kilometer 14. Here I collected a few specimens of Polypedates leu- comystax, Rana luzonensis, and Rana magna. We reached camp 30 in the evening. April 14. An adult specimen of Oxyrhabdium leporinum was collected. April 15. The day was spent in the vicinity of Haight’s place, Pauai, but no reptile or amphibian was seen or heard. Mr. Haight said that frogs were occasionally seen. April 16. Collecting along the trail between Haight’s and camp 88 yielded neither amphibian nor reptile. April 17. A specimen of Cornufer subterrestris was collected at kilo- meter 101. Professor Beyer left for Cervantes. April 18. A specimen of Peropus mutilatus was taken in the presidencia of Subangan. Specimens of a Mabuya were shot near the trail. * Amphibians and turtles of the Philippine Islands, Bureau of Science publication 15 [reprint from Philip. Journ. Sci.]. Manila (1921) 193 pp., 17 pls., 9 text figs. Snakes of the Philippine Islands, Bureau of Science publication 16. Manila (1922) 312 pp., 37 pls., 32 text figs. Lizards of the Philippine Islands, Bureau of Science publication 17. Manila (1922) 262 pp., 28 pls., 53 text figs., in press. 161 162 The Philippine Journal of Science 1922 April 19 and 20. I collected in the vicinity of Bontoc and was dis- appointed because of the few specimens found. Doctor Schofield saw a turtle in the market, but it had disappeared before I could get it. April 21. I left Doctor Schofield at Bontoc and set out for Balbalan, Kalinga. April 23. Near Balbalan I found a specimen of Hologerrhum philip- pinum. I purchased a specimen of the mammal Phloeomys pallidus. April 23. Near Balbalan I found a specimen of Hologerrhum philip- and as a result amphibians began to breed. I collected large series of several species, some of which are new. May 2 and 8. Balbalan to Bontoec. Very few specimens were collected. May 5. I climbed Polis Mountain. May 6. I arrived at Banaue. May 7. Banaue to Kiangan. Amphibians were heard at night near Kiangan, but I was unable to collect any of them. Two young specimens of Elaphe erythrura were collected along the trail. May 8. Kiangan to Nayon. May 9 to 12. I collected Kaloula picta and several of the commoner lizards in this vicinity. . May 13. Nayon to Kiangan. May 14. Kiangan to Tauang. May 15. Tauang to Manhuyhuy. A small frog, probably Cornufer meyeri, was found dead near the trail. May 16. Manhuyhuy to camp 59. Between kilometers 65 and 64 a large red frog jumped into the trail in front of me and from there into the cafion. This frog appeared to be an unknown species of Rana. The elevation at this place is about 2,500 meters. No frog has been recorded from so high an elevation in the Philippines. May 17. Camp 59 to Baguio. MANILA AND VICINITY During May and June, 1920, I made numerous trips about Manila, to Mount Maquiling, and to Mount Banahao, # POLILLO AND CALOTCOT ISLANDS July 8. I left Manila by rail for Hondagua where I took the steamer for Polillo. ; July 9. I arrived at Polillo. July 10 to 23. In the vicinity of the town of Polillo, I collected numerous specimens. July 24. I went to Burdeos on the east coast of the island. A snake, probably a species of Natrix, was seen on the trail, but I was unable to capture it. July 25 to 27. Some interesting specimens, including Rana merrilli sp. nov. and Typhlops cumingii (Gray), were collected near Burdeos. July 27. I embarked in a small banca for Calotcot Island, which is about 25 kilometers east of Polillo. July 28 to August 1. Calotcot Island is about 3 kilometers long and is heavily forested. There is a small clearing at the southern end. Eleven species of lizards and three species of snakes were taken on this island. No amphibians were seen. I returned to Burdeos during the night of August 1. 21,2 Taylor: Herpetological Fauna, I 163 August 2 and 8. I collected at Burdeos. August 4. I returned to the town of Polillo. August 5 to 10. I collected in the vicinity of Polillo. August 11. I went to Bislian in the southern part of the island. There I found several new species. August 12. I returned to Polillo. August 13. I left Polillo. August 14. I arrived in Manila. MINDANAO September 23 to October 6. During this period I collected in the south- ern part of Zamboanga Peninsula, for the most part in the mountains along Tumugao River, including a two days’ journey from the waterworks’ intake. Only a few species were found. One specimen of the rare black and yellow cobra, Naja samarensis, and three or four frogs of undescribed species were collected. BASILAN October 7. I collected in the vicinity of Isabela. Through the kindness of Mr. Guy Holland, a lumberman of Isabela, I was taken by launch to Port Holland, the site of a sawmill on the western coast directly opposite Great Govenen and Little Govenen Islands.2 I had visited this point in 1917 and had found several interesting specimens. At that time primeval forest covered the sites. October 8 to 14. Port Holland. I found Polypedates pardalis and Polypedates appendiculatus while collecting at night. A good breeding series of Kalophrynus stellatus was taken. While here I was very kindly entertained by Mr. Hamilton, the sawmill manager, who facilitated my collecting in no small degree. October 15. to 23. Abung-abung. Thanks to the kindness of Mr. Hamilton and Mr. Holland, I was enabled to visit a logging camp on the southern part of the island. It was an ideal collecting ground, and many specimens and species were taken, some of them unknown tome. A cecilian differing from Ichthyophis weberi was discovered and four specimens were taken. I had splendid success collecting at night. Returned to Isabela on the 23d and left for Zamboanga the same day, embarking there for Jolo on October 24, 1920. JOLO Oetober 25 to November 16. This period was spent in making col- lections in the western half of Jolo. Visits were made to Mount Tuman- tangis, Bud Daho, Indanan, Camp Romandier, and a point near the central part of the island where the Government cattle ranch is situated. I am under great obligation to Capt. Francis Link, formerly of the Philippine Constabulary, who accompanied me on many of my trips. A specimen of the rare Luperosaurus joloensis was seen, but it escaped in a tall tree. November 17. I sailed for Manila. j i ! i ic Survey *The spelling of these names is that on Coast and Geodetic maps. They are called Gouenan by the Yakans and Samals of the district. 1878973 164 The Philippine Journal of Science 1922 MOUNT MARIVELES December 26, 1920, to January 1, 1921. I collected on the northern side of Mount Mariveles. Two specimens of the rare Hologerrhum philippinum were taken, but nothing else that was noteworthy. MOUNT MAQUILING April 15 to June 1. During this period extensive collections were made on Mount Maquiling. A specimen of Oxyrhabdium, which is distinct from the two recognized forms, was taken, but it escaped from a faulty col- lecting bag. Aside from collections made by myself I have obtained small collections from southern Leyte, Sibuyan, Romblon, Batan Island, Busuanga, and Balabac. In this paper the following species are described as new: AMPHIBIANS Rana merrilli. Philautus basilanensis. Cornufer cornutus. K@gloula negrosensis. Philautus polillensis. Kaloula kalingensis. Philautus williamsi. Kaloula rigida. Philautus zamboangensis. Bufo mcgregori. LIZARDS Gekko porosus. Sphenomorphus bakeri. Gekko smaragdinus. Siaphos herrei. SNAKES Dryocalamus mecroryi. Pseudorhabdium minutum. Calamaria joloensis. Typhlogeophis ater. The following rare or little-known species are discussed: Rana everetti Boulenger. Typhlops suluensis Taylor. Megalophrys hasselti (Tschudi). Haplonodon philippinensis Griffin. Ptychozoon intermedia Taylor. Hologerrhum philippinum Ginther. Draco mindanensis Stejneger. Calamaria grayi Giinther. Typhlops cumingii (Gray). Rana merrilli sp. nov. _ Type.—No. F876, E. H. Taylor collection; collected at Bur- deos, Polillo Island, July 27, 1920, by E. H. Taylor. Description of type—Choane large, very widely separated, partially hidden by overhanging jaw; vomerine teeth in two small oval groups, lying between the choanze and barely ex- tending back as far as posterior edge of choane, separated from choanze by a distance nearly as great as that between the two groups; tongue large, deeply bifurcated, the horns rounded at their tips; head much flattened; snout moderately 21,2 Taylor: Herpetological Fauna, I 165 long, extending beyond lower lip; distance of nostril from eye double its distance from tip of snout; distance between nostrils less than interorbital distance, which is distinctly greater than width of upper eyelid; diameter of eye about equal to its dis- tance from nostril; tympanum very large and distinct, its diam- eter about four-fifths of eye and equal to the interorbital distance; loreal region slightly concave, perpendicular for some distance, then sloping abruptly to lip; canthus rostralis moder- ately distinct, rather rounded on edge; angle of mouth with a short glandular fold extending to behind tympanum, and another short fold above insertion of arm; distinct narrow dorsolateral folds begin behind eyes and continue dorsolaterally to end of body, converging slightly; upper eyelid without tu- bercles; skin finely shagreened above; chin, throat, and breast entirely smooth; belly and underside of femur distinctly gran- ulate; below anus two strong elongate glandular tubercles narrowly separated; three outer fingers with strongly widened, rounding disks on tips; inner with small, scarcely widened disks; first finger much shorter than second which is only slightly shorter than fourth; subarticular tubercles strongly developed; three strong carpal tubercles, with a few smaller tubercles scattered over palm; toes with smaller, rather pointed disks; the web between toes extends to base of disks on outer side of first, second, and third toes and on inner side of fifth toe, on inner side of second and third toes to near the disk by a very narrow margin, and to the penultimate joint of fourth; subar- ticular tubercles moderately distinct; a small inner metatarsal tubercle and a still smaller outer tubercle; hind limb brought forward the tibiotarsal articulation reaches between eye and nostril. Color in life—Above uniform olive green to lighter green on sides. No spots or markings; no bars on limbs; upper part of loreal region darker olive; a black spot immediately in front of tympanum below eye and another behind tympanum; a small yellow area behind eye ‘and another covering angle of mouth; below greenish yellow to cream with olive spots; dorsolateral glandular folds lighter than body and a slightly darker olive line below them. Variation.—The cotype, from the same immediate locality, agrees in detail with the type save that the tympanum is some- what larger proportionately and the belly immaculate. I find no vocal sacs in either specimen. The specimens are probably females. 166 The Philippine Journal of Science 1922 Measurements of Rana merrilli sp. nov. No. F876, | No. F876A, type. cotype. mm, mm, Pertes GO WONG. 5 255. eos. cogs daca te yc cto oa ee eae 88.5 89.0 WMaU OR MOMs scones agosto tens eee ot oe cet cie ee ee 11.0 12.2 cebrni Or helt ais ra ee ee os se 14.0 16.0 seth of enont ite front of ogee ooo Se 3.0 3.5 Depth of head, at tympanum -__-_- : See : 4.0 5.0 ETE OE BROW oo aS eae a ee Fee 5.5 6.0 Teie te Cpmbetl sor eg ee ee 4.8 4.5 Wismeter OL eye <5. io 4,4 4.3 Dininever of CYMIDANUM 3 a a 3.8 4,0 Forelimb --_ .. ekewac es So et aueter a oceubek eee ccs 22.0 26.0 Widltticad aible soy ee tere ec A 2.0 2.0 DOOR San ro kee 12.0 13.2 Paid TD Sos soho ee oso ne sue 60.0 64.0 WONG ese en Skee ees utah awe see taeda 18.0 19.0 Tthiassccss2s woe Se re Ae Ree ee ee 20.0 22.0 Longest toe, to metatarsal tubercle... _._- so 16.5 17.0 Remarks.—This species is differentiated from other Phil- ippine species of Rana by the very much-flattened head, the very large tympanum, and the presence of the two transverse glandular folds, or tubercles, below the anus. Specimens were discovered seated on the leaves of Pandanus trees which extended over a pool of water in an old stream bed. Two other specimens observed, one in the forest, the other in the same locality, es- caped, The species is named for my esteemed friend Elmer D. Merrill, director and botanist of the Bureau of Science, Manila. Rana everetti Boulenger. Rana everetti BOULENGER, Cat. Batr. Sal. Brit. Mus. ed. 2 (1882) 72, pl. 6; TAyLtor, Amphibians and Turtles of the Philippine Islands (1921) 68, pl. 6, figs. 1, 1a, 1b. A large specimen of this rare species was taken at Pasananka, near Zamboanga. It was found in a rain pool near the small park and water reservoir. The specimen agrees in detail with Boulenger’s admirable figure and description. The dorsolateral glandular fold is dim, but it can be traced for two-thirds the length of the body. While not mentioned in the description, Boulenger’s figure shows a tubercle between the shoulders slightly to the right of the median line. My specimen has a well-developed, conspicuous tubercle in the same place. The specimen is probably a female and consequently without vocal sacs. 21,2 Taylor: Herpetological Fauna, I 167 Measurements of Rana everetti Boulenger. mm, Snout to vent 80.0 Length of head, to angle of jaw 82.0 Width of head, at tympanum 26.0 Diameter of eye 9.1 Diameter of tympanum 7.0 Eye to end of snout 13.0 Eye to nostril 8.0 Upper eyelid 7.5 Interorbital distance 7.5 Forelimb 44,0 Longest finger with hand 25.0 Hind limb 140.0 Femur 42.0 Tibia 46.0 Longest toe, to metatarsal tubercle 35.0 Widest disk on forefoot 4.5 Remarks.—The webbing on the foot reaches the disks by a very narrow margin on the inner side of the second and third toes, and does not extend beyond the outer subarticular tubercle of the fourth. The species appears to be rare, since only this single specimen was found in the course of one month’s intensive collecting. Philautus williamsi sp. nov. Type.—No. 356, E. H. Taylor collection; collected on Polillo Island, August 12, 1920, by E. H. Taylor. Description of type-—Choanz moderately large, widely sepa- rated, not concealed by overhanging jaw; vomerine teeth want- ing; tongue moderate with two broad rounded horns posteriorly, Separated narrowly at base; no tubercle present; head very broad, nearly equal to one-half the length of body; eye large, the diameter of orbit equal to the length of snout; the width of upper eyelid equal to interorbital distance; snout short, end- ing in a very small conical projection; area about nostrils slightly raised and projecting with a depression between them; upper part of loreal region nearly perpendicular, lower part sloping obliquely to mouth; a distinct depression in front of eye not extending to nostril; tympanum small, indistinct, covered with skin, its diameter about equal to one-third that of eye; a strong fold from eye across upper part of tympanum to insertion of arm; a slight curved fold crossing angle of mouth; skin of body minutely corrugated; head with scattered tubercles, one or two tubercles on upper eyelid, two tubercles behind the eyes, the posterior largest with another pair of dim tubercles in middle 168 The Philippine Journal of Science 1922 of back; these tubercles connected with the anterior pair by a very dim fold; tibia with indistinct tubercles; skin of throat slightly corrugated; belly, breast, underside of forearm, and femur strongly granulate; fingers with widened disks, the two outer much larger; those of the two inner fingers small, and partially opposed to the other two; subarticular tubercles large, rounded, well defined; carpal tubercles ill defined; toes with pads slightly smaller than those on fingers, one-third to one- half webbed; subarticular tubercles moderately distinct; a wavy skin fold on outer side of fifth toe and part of foot; inner meta- tarsal tubercle distinct, large, flattened; outer, if present, very indistinct; the tibiotarsal articulation brought forward reaches just beyond tip of snout. Color in life-—Above yellow-brown with markings of darker brown; upper eyelid black; snout and upper part of loreal region dark brown; two light areas behind eyes; a large regular brown spot beginning on shoulders, forking in the middle of the back, is continued to groin; area between branches only slightly lighter than the branches; one broad brown bar across femur and tibia and another narrow outer one; lower part of tibia and inner part of foot pure white; posterior aspect of femur nearly white with slight marbling of brown; second tubercle behind eyes very dark; edge of upper jaw some- what lighter than lores, and a dim light cream spot below and somewhat behind eye; throat, breast, belly, underside of arm, fingers, foot, toes, and web between toes strongly powdered with cinnamon brown; two inner fingers immaculate. Measurements of the type and the cotype of Philautus williamsi sp. nov. No. 356, No. 857, . type. cotype. mm. mm. Baout to vents Sis ee A oo 21 22 Length Of head. oct oe ee eS 9 9.6 Width of tead,...0.5-2 0; 10 10 Depth of head... cece a ee 4.5 5 Forelimb .___.. Sk us Rise dee Pape Onis Gealy le 12.8 13.5 Longest finger -........... 6.8 6.3 Hind limb - 40 38 Longest toe, to metatarsal tubercle_____ : 9.6 9.5 Diameter of eye_........ Cake ON rer ee 4 4 Length of snouts ii55.22oo in ceo sake ha 4 4 Variation—There is considerable variation both in mark- ings and in the distribution of tubercles on the head and back of the four cotypes." The prominent black tubercle some 21,2 Taylor: Herpetological Fauna, I 169 distance behind the eye is present in all the specimens, but the narrow fold behind it is not so evident. Some of the specimens have numerous tubercles between the eyes and on the snout; in others they are largely wanting. Specimen 356 shows no char- acteristic mark on the back. I find no vocal sacs in any of the specimens. Remarks.—The type and three cotypes were collected in southern Polillo along the trail between Polillo and Bislian at a point near where the trail crosses the low divide. A single specimen was collected by Dr. F. X. Williams in mountains near Mauban, Tayabas, on the eastern coast of Luzon. This speci- men is rather different in color and markings, but there are no structural differences save those that may be accounted for by the state of preservation. The species is named for Dr. F. X. Williams, of the Hawaiian Sugar Planters’ Association, who collected the Luzon specimen. I believe that this is the first record of a species of Philautus for Luzon. Philautus basilanensis sp. nov. Lype.—No. 1510, E. H. Taylor collection; collected at Abung- abung, Basilan, October 15, 1920, by E. H. Taylor. Description of type-——Choanz small, widely separated, not concealed by overhanging jaw; tongue narrow, elongate, nicked behind; a slight, rounded tubercle on anterior part of tongue; openings of vocal sacs small, near angle of mouth; snout moder- ate, pointed, ending in a blunt conical point; canthus rostralis distinct, slightly rounded; loreal region perpendicular for a short distance, then sloping obliquely to edge of lip; nostrils forming raised prominences on snout, much nearer tip of snout than eye; diameter of eye equal to length of snout; eyes prom- inent, the upper lids equal to interorbital distance; tympanum small, indistinct, covered with skin, not more than one-third diameter of eye; a straight supratympanic fold from eye to insertion of arm; skin of body and limbs very strongly granular, intermixed with larger tubercles; tubercles on head and shoul- ders arranged in more or less regular lines crossing eyelid and continuing on shoulders; tubercles on femur and tibia arranged in transverse lines; numerous tubercles on eyelids, one more prominent than the others; skin on belly entirely covered with strong mosaiclike granules; lower jaw with a series of glandular tubercles around edge, with a prominent tubercle at tip of lower jaw; tips of digits of hand strongly dilated, those of the two outer fingers double the width of digit; those of the two inner fingers smaller and somewhat opposed to the two outer; sub- 170 The Philippine Journal of Science 1922 articular tubercles well defined; sole with fine granules; under- side of forearm with two or three enlarged tubercles; no trace of web between fingers; toes with enlarged disks at tips, slightly smaller than those on fingers; toes about one-third webbed; subarticular tubercles prominent; undersurface of foot and digits strongly granular; a prominent inner metatarsal tubercle; outer tubercle, if present, not distinguishable from numerous tubercles on sole of foot, which continue to heel; underside of tibia without granules; the leg being brought forward, the tibiotarsal arti- culation reaches beyond tip of snout. Color in life—Above reddish brown, variegated with lighter and darker areas but no distinct markings; tips of larger tuber- cles yellow; tip of snout lighter brown than body; lores dark; belly white to gray; the two inner fingers bright yellow; edge of upper and lower lips yellow; tubercles on lower jaw yellow; underside of hand yellow to cream, of foot reddish brown. Measurements of the type and the cotype of Philautus basilanensis sp. nov. No. 1510, | No. 1699A, type. cotype. mm. mm, REED), GUD G SiS WR ce a se ee eee 21 22 Wioth ot boead, At tompaniime ooo 9 8 Depth of head, at tympanum___ 6.6 6 Length of head, to posterior edge of tympanum-- 9 9 Diameter of eye -..--_----_- ot cea ea 4 4 Length of snout fron eye... a 4 4.1 Forelimb 22:23. SS ek Se 11.5 12 Timpest finger | 55) cele peau De dteckise Ss weweelo: 6 6 B= 07s ES 1 Sah cee rer apiece hey Aiea Gries Otugmcn cere amr Wks ore are Neat one 38 37 Longest toe, from metatarsal tubercle._.........._........-_.------------ 9.5 9 Variation.—Two cotypes, taken in the same immediate local- ity, vary little in essential characters; the granulation is not so strongly defined in either of the other specimens collected. No. 1699A has two rather strong tubercles between the eyes, which are but dimly suggested in the type. No. 1269B is light lavender-brown. A distinct black area lies between the eyes and involves part of the upper lids. A few scattered black spots are between the shoulders. Two large elongate spots are present on each side of the back. The limbs are strongly barred with reddish brown. This specimen has suffered an injury which has destroyed the side of the head and one forelimb. Remarks.—The species appears to be related to Philautus woodi Stejneger, but differs in the size and arrangement of the 21,2 Taylor: Herpetological Fauna, I 171 tubercles on back and limbs, and the greater extent of webbing between the toes. Numerous other differences are evident on a comparison of the descriptions. Specimens were captured at night by following in the di- rection of their shrill chirping cry, which for a long time was believed to come from some species of cricket. They were located with great difficulty, perched on the leaves of low shrubs. A few other voices were heard, but these sounds came from a swamp overgrown with thick jungle through which one dared not venture at night. Philautus polillensis sp. nov. Type—No. 351, E. H. Taylor collection; collected near the southern end of Polillo Island, July 12, 1920, by E. H. Taylor. Description of type—Head about as wide as body, pointed; tongue moderately long with two short horns, narrowly sepa- rated at base; choanz small, widely separated; vomerine teeth in two small rows, considerably behind the choane, separated from each other by a distance one and a half times the length of one of the groups; tongue with a raised moundlike prominence near anterior part; tip of snout conical; nostrils much nearer tip of snout than eye; diameter of orbit somewhat less than length of snout; a single large conical tubercle on upper eyelid near outer middle edge; pupil horizontal; canthus rostralis distinct; loreal region concave, sloping gently to edge of jaw; tympanum rather indistinct, its diameter one-third to one-fourth of eye; a strong fold above tympanum to near insertion of arm dimly granular; a second fold from a point above and behind tympanum to lower jaw; a row of granules from be- hind tympanum to arm, parallel to the supratemporal fold; a distinct depression between the two folds; skin above prac- tically smooth, with two small tubercles between shoulders and two pairs on back; a small tubercle on back above anus; a distinct tubercle on tip of jaw; skin of throat and breast smooth, with numerous minute pitlike depressions; belly and underside of femur strongly granular; a strong tubercle at heel; tip of each of the three outer fingers with a strongly dilated pad, nearly twice the width of digit; inner finger very small, slender, without pad; subarticular tubercles large, flat, moder- ately well defined; a single large carpal tubercle; no trace of web; tips of toes distinctly dilated, not more than one and one- half times the width of digit; fourth toe very long, third and fifth equal, barely reaching third subarticular tubercle from the 172 The Philippine Journal of Science 1922 disk on fourth toe; inner metatarsal tubercle rather large but ill defined; outer small, dim; leg brought forward, the tibiotarsal articulation reaches about halfway between eye and nostril. Color in life-—Above creamy white to yellow, slightly pig- mented with minute dots of cinnamon brown; a bar of cinnamon between eyes and dim spots about the dorsal tubercles; upper eyelids dark gray to blackish; spots on outer digits of all limbs; dim bars on tibia and femur; chin and throat yellow with very sparse peppering of brown; underside of hand and belly im- maculate; underside of leg and foot strongly peppered with brown. Measurements of the type of Philautus polillensis sp. nov. mm, Snout to vent 27 Length of head, to angle of jaw 11 Width of head, greatest 10 Diameter of eye 4 Length of snout 4.5 Forelimb 16.3 Longest finger, from wrist 7 Hind limb 42.5 Tibia 14.2 Longest toe, from metatarsal tubercle 11.2 Variation.—There are seven cotypes in the collection, all taken in the same immediate locality. These vary more or less in markings, but the coloration is identical. No. 349 (22.5 milli- meters long) strongly resembles the type in markings; the pitlike depressions in the skin of the lower jaw are not evident; the tubercles on the back are dimmer. No. 350 (20 millimeters long) is very sparsely pigmented on the back. It was almost white when taken; the skin on the chin and the throat is slightly granulate; the pigment on the back is arranged in two very dim curving lines beginning behind eye; this marking is suggeSted inthetype. No. 353 (19 millimeters long) is similar to No. 350; the sides are lighter, and the slightly curved mark- ings on the back are present. The pigment is heavier between the shoulders and as far forward as the line between the eyes. The snout and the area below the eye are distinctly lighter; the skin of the throat and the chin is smooth. No. 348 (17 millimeters long) has brown spots about the dorsal tubercles and a deep brown line behind eye to insertion of arm; the snout is very light. Nos. 352 and 354 (each 15 millimeters long) have a hair line from between the eyes to above the anus. No. 21,2 Taylor: Herpetological Fauna, I 173 355 is almost white above with a fine line from the tip of the snout to above the insertion of the arm. The distinct spots on the two outer fingers are very charac- teristic. There is some variation in the folds about the tym- panum; some of the specimens show these very dimly. In certain specimens the vomerine teeth begin at the hinder edge of the choane. Remarks.—Save for the presence of very distinct groups of vomerine teeth I should regard this species as belonging unquestionably to the genus Philautus; as it is, I have referred it to that genus with some hesitancy. However, the presence or absence of vomerine teeth is scarcely a generic distinction since we find them present or absent in the very closely related genus Polypedates. The character of the digits places it with Philautus rather than with Polypedates. The specimens were discovered concealed under and about the leaves of low-growing plants in southern Polillo. The type locality is on a trail running from the walled town of Polillo to the southeastern point of the island (known as Bislian) at a point where the trail crosses the low divide. These specimens were taken during an hour’s collecting in this locality.” Inten- sive collecting for nearly one month on the island failed to reveal the species elsewhere. Philautus zamboangensis sp. nov. Type—No. 1059, E. H. Taylor collection; collected near Pasananka, Zamboanga, Mindanao, September 26, 1920, by E. H. Taylor. Description of type—Choane large, widely separated, partial- ly concealed by overhanging jaw; no vomerine teeth; tongue rather short, forked behind; the openings of the vocal sacs elongate, nearly half the length of jaw; snout short, truncate; eye large, its diameter equal to or minutely less than its distance from end of snout; the distance between nostrils greater than their distance from eye; tympanum small, distinct, partially covered by the supratympanic fold, about one-third the diameter of eye; upper eyelid only minutely less than interorbital dis- tance; interorbital area with a raised prominence followed by a circular depressed area; a broad, shallow depression covering ‘snout; skin of body smooth above with no trace of granules; no tubercles on eyelids; skin on chin and throat smooth or with fine granulations; belly finely granulate ; numerous much-enlarged granules about anus; arm very short, the fingers with well- 174 The Philippine Journal of Science : 1922 developed, rounded pads; no web, or only a very small one, be- tween first and second fingers; second and third fingers one-third webbed, third and fourth nearly one-half webbed; subarticular tubercles strongly developed; several small tubercles on palm; a skin fold on outer side of fourth finger ending in a rounded carpal tubercle; two large tubercles at base of first finger; a broken glandular fold on outer side of forearm; first finger not extending as far as second; foot about two-thirds webbed, the web reaching to near base of disk on the outer side of second and third toes but failing to reach base of penultimate phalanx on fourth; subarticular tubercles well defined; a strong inner metatarsal tubercle on outer side of first toe near base; outer metatarsal tubercle very small; a fold on outer side of fifth toe not continued beyond tubercle; a slight skin fold on heel; hind limb brought forward the tibiotarsal articulation reaches tip of snout. Color in life—Above grayish white on head and body; a dark cinnamon brown bar between the eyes, and dark cinnamon dots scattered on back and limbs; throat cream; belly lemon; under thighs and in groin deep yellow; spots on posterior part of femur “And skin fold on heel cream; lores lighter than snout; a sharply defined cream spot below eye which involves part of lid; a dark area on tympanum, following cream spot; a black area below anus and a blackish line from heel to sole of foot; undersurface of hand yellow, of foot light brown. Measurements of Philautus zamboangensis sp. nov. mm. Snout to vent 28 Width of head, at tympanum 10.5 Length of head, to angle of jaw 10.5 Eye to tip of snout 6 Diameter of eye 4.8 Interorbital distance 4 Tympanum 1.5 Forelimb 16.5 Longest finger, from wrist 9 Hind limb 51 Femur 16 Tibia 16 Longest toe 12 Remarks.—This species appears to be related to Philautus bimaculatus (Peters) and P. montanus Taylor. From the for- mer it differs in having the openings of the vocal sacs elongate slits, nearly half the length of the jaw; in the absence of strong tubercles on the under jaw; in having a shorter arm; the distance 21,2 Taylor: Herpetological Fauna, I 175 between the nostrils being greater than their distance from eye; the undersurface of foot having subarticular tubercles; and in the lesser extent of webbing on foot. The species agree in the shape of the head and the general conformation of the hand. Philautus montanus differs from P. zamboangensis in having the first and second fingers in opposition to the third and fourth, instead of the first in opposition to the other three. The only specimen seen was discovered seated in an unfolding leaf of a low plant on the bank of Tumugao River, above the waterworks’ intake near Pasananka, Zamboanga. Shortly after capture the specimen became a deep orange color. Cornufer cornutus sp. nov. Type.—No. 764, E. H. Taylor collection; collected at Balbalan, Kalinga, Mountain Province, Luzon, April 24, 1920, by E. H. Taylor. Description of type-—Vomerine teeth in two small, more or less rounded series lying behind the inner edge of choane, widely separated from each other; choanz small, not concealed by jaw; tongue nicked behind, with a distinct papilla on anterior part; snout rather pointed in front of nostrils; canthus rostralis distinct, the lores concave, then sloping obliquely to lip; snout somewhat constricted behind nostrils; nostril a little nearer tip of snout than eye; interorbital distance distinctly wider than upper eyelid, slightly less than distance between nostrils; upper eyelid with numerous, large, rounded tubercles and a well-developed, sharp-pointed dermal spine 1 millimeter long, which projects outward in life; eye small, equal to or slightly longer than its distance from nostril; tympanum small, about two-fifths of eye; a strong supratemporal fold from behind eye to insertion of arm slightly overhanging tympanum; a tubercular fold below eye immediately in front of tympanum and a fold crossing angle of jaws; a few scattered tubercles behind eye; skin generally smooth or finely shagreened, with dim indications of tubercles on back, limbs, and sides; skin on chin and throat smooth; pos; terior part of belly and underside of femur strongly granulate; fingers with strongly dilated disks, first finger very small, the tip scarcely dilated; fourth finger longer than second ; subartic- ular tubercles large, distinct; small tubercles on palm, and large carpal tubercles; tips of toes dilated but much less so than fingers; a very strong outer metatarsal tubercle, and a dim elongate inner one; hind limb being brought forward the tibio- tarsal articulation reaches eye or slightly beyond. 176 The Philippine Journal of Science 1922 Color in life.—Blackish brown above with two putty-colored lines on side of back; small, scattered, greenish spots on back; spots on tympanum, loreal region, and top of head putty- colored; groin yellow; sides of body slightly yellow; upper part of femur yellowish green; throat and belly whitish; femur and tibia greenish yellow below. When preserved in alcohol the greenish spots appear black. The spots on the tympanum, be- low the canthus rostralis, and on the upper lip are strongly’ pronounced. Measurements of Cornufer cornutus sp. nov. Snout to vent 31.6 Length of head, to behind tympanum 11 Width of head, at tympanum 12 Depth of head ate Forelimb 20 Longest finger, to wrist 10.2 Hind limb 49 Longest toe, to metatarsal tubercle 15 Diameter of eye 3.8 Eye to tip of snout 6.2 Remarks.—The type was collected from a bush growing in the small stream which furnishes drinking water to the town of Balbalan. It was found seated on a broad leaf directly above the water. Only a single specimen of this new species of Cornufer was found in ten days’ collecting at Balbalan. This species, more than any other of the numerous species of the genus, is worthy of the name “horn-bearer.” The peculiar spine above the eye easily differentiates it from other known species of Cornufer in the Philippines. Kaloula rigida sp. nov. Type.—No. 7681, E. H. Taylor collection; collected at Bal- balan, Kalinga, Mountain Province, Luzon, April 26, 1920, by E. ° H. Taylor. Description of type—(Adult female.) Choanz large, par- tially concealed by overhanging jaw, separated from one another by a distance equal to diameter of choanz; two strong, transverse, palatal ridges immediately behind choanz, very narrowly separated medially; in front of cesophagus a wide, dermal, transverse, palatal ridge, which is preceded by a second arched ridge; tongue broadly oval, entire; snout short, truncate; rounded on edge; loreal region nearly perpendicular; diameter of eye longer than its distance from end of snout; nostrils as far forward as tip of snout, which slopes backward and down- 21,2 Taylor: Herpetological Fauna, I 177 ward to mouth; tympanum small, dimly outlined, covered with skin; a distinct fold from behind eye to insertion of arm; a dim fold in front of tympanum and another short fold behind angle of jaw; skin above uniformly corrugated save on tip of snout and lores; belly more or less smooth; a dim suggestion of granulation on ‘throat and chin and on inferior and posterior aspects of femur; tips of digits on hand slightly swollen, no wider than digits; subarticular tubercles moderately developed, large, rather flattened; carpal tubercles not prominent; toes without disks; subarticular tubercles on foot large, flattened, not strongly differentiated; a round, outer metatarsal tubercle and an elongate, blunt-edged, inner tubercle; a small but dis- tinct web between toes; leg brought forward the tibiotarsal articulation does not reach beyond insertion of arm; femur involved in body skin for more than half its length; males with internal vocal sacs. Color in life-—Above deep lavender to purple, slightly irides- cent; lighter lavender to brownish in groins and above limbs; dim traces of spots on limbs and a darker triangular area about anus; belly dirty light brown, mottled and reticulated with lighter color; chin and throat dark with fine reticulations of dirty white. Measurements of the type of Kaloula rigida sp. nov. : mm. Length, snout to vent 47 Length of head, to behind tympanum 11 Width of head posterior to edge of eye 15 Distance between posterior corners of eyes 11 Depth of head posterior to eye 7.5 Length of forelimb 30 Longest finger : 14 Length of hind limb 60 Length of longest toe, to outer metatarsal tubercle 19 Variation—Twenty-two specimens of this species were taken breeding in rain pools in the town of Balbalan, from April 28 to April 30, 1920. Most of the specimens are identical in most details with the type; a few vary in color and markings and in the condition of the skin. The females are distended with eggs, and when alive they were almost triangular in shape. Most of the specimens are colored like the type. One female (No. 838A) is light lavender-brown above with purplish black markings on the back, arranged roughly in the shape of a human being with limbs outstreched; there are two well-defined dark spots above the anus, and the limbs are distinctly spotted; the lores and the 178 The Philippine Journal of Science 1922 side of the head and body are of a uniform purplish color. The anterior aspect of the femur is also dark. Several specimens, representing a variation of this species, were taken from under stones in the garden of the Hotel Pines, Baguio, on April 12, 1920. They differ markedly from the type in having the skin very smooth and shiny. The marking on the back is similar to that of specimen 838A but with numerous, rounded, black spots scattered over the back and the sides. When first disturbed the specimens lay flat, stretched their legs out behind, and remained rigid. They were picked up in this condition and remained motionless for a considerable time. One large female, with her body much distended with eggs, has the femur involved in the body skin more than two- thirds of its length. One specimen (No. 7680, taken at Bontoc, April 20, 1920) was bright brown-red, the skin as smooth and shiny as patent leather. There are a few rounded black spots on the back. The specimen was found burrowed about 10 centimeters deep in loose earth. This species is related to Kaloula picta Duméril and Bibron, but differs in the shape of the head and the body, the size and prominence of the metatarsal tubercles, and the shorter inner toe. Numerous other differences are evident on a comparison of specimens. Kaloula kalingensis sp. nov. Type.—No. 824, E. H. Taylor collection; collected at Balba- lan, Kalinga, Mountain Province, Luzon, April 28, 1920, by E. H. Taylor. Description of type.—(Adult female, containing eggs.) Pal- atal ridges behind choanz not strongly developed, not meeting medially; choanze small, not concealed by overhanging jaw; two denticulated ridges across palate in front of cesophagus, the posterior straight, the anterior distinctly arched; tongue oblofg, entire, free behind; snout short, truncate, angular in outline from above, extending very slightly beyond lower jaw; nostrils lateral, not or scarcely visible from above, profile at the extremity of snout; distance of nostril from eye greater than its distance to edge of mouth; canthus rostralis rounded; loreal region perpendicular; eye large, length of orbit greater than length of snout; lower eyelid with an opaque, cream- colored area; interorbital distance one and one-half times the width of upper eyelid; tympanum distinct, about three-fifths of eye; a dim skin fold begins behind eye and becomes heavier 21,2 Taylor: Herpetological Fauna, I 179 behind tympanum; it continues above arm, where it becomes thickened and glandular; a strong fold behind angle of mouth below and behind tympanum, continuing across throat; a deep groove, beginning behind tympanum and continuing to behind insertion of arm, separates the folds; skin above with flattened pustular tubercles prominent on snout, lores, sides, back, and limbs; skin on chin with small granulations; skin on chest smooth, on belly and femur strongly granular; fingers strongly dilated at tips into angular pads more than twice the width of the fingers; anterior edges of pads straight or slightly concave, sides nearly parallel; first finger reaches base of pad on second; second and fourth toes equal, reaching base of pad on third; subarticular tubercles strong, two on third and fourth toes; three distinct carpal tubercles; toes with small, slightly truncate or rounded pads, not or scarcely wider than toes; a strong, oval, inner metatarsal tubercle and a smaller, rounded outer one; subarticular tubercles strongly developed; a single tubercle un- der first and second toes; two under each of the other three; toes with no or only a very slight rudiment of web; the leg brought forward the tibiotarsal articulation fails to reach tym- panum by a distance equal to diameter of tympanum. Color in life-—Bluish black above with deep red markings on neck, sides, and limbs; minute cream markings on each side of anus; belly and underside of limbs mottled with creamy white; tubercles under digits white; throat and chin blue-black. Measurements of the type and the cotype of Kaloula kalingensis sp. nov. No. 824 9, | No. 856 J, type. cotype. mm. mm. Sndat co vent .223 2 ee a oe a uae evan sa 36.5 34.0 Beastly OE ita ead deca ee ek ouesseep- one 9.5 9.5 Width of head_______.- PS PRE PSNIES CARS an hie anes eee AT ee 13.5 11.0 BVRMe Oe ae Sons nen manana aneneas dane 4.8 4.4 TIS C8 Bia ice Se sina Boas Se ee Cake Chee eee Suse Skt 4.0 3.9 Width of body; wrentast so... 2. . 655 2-5. 5-2 oe een. 25-2 eee 18.5 16.0 TR Ta ge me tome inn wo ellen oe 11.5 10.2 WUPOHIND = cbc a ae eae aa pds Laden ap saccanpensanees : 23.0 22.0 RAT Via oe oe ek Sa eR ee ne nS een cn eee wee rok eee 46.0 44.0 LLL, | MMO eRe Cie ad Saat es eee Pee mails Agape are Rae, Manan ae es at eoeton ee 15.0 13.0 li co ern ee kilt a Pl enon ar ple bhvny eelbne genase seieee nr eee MONS MRT TG BP gan i? 15.5 13.8 Variation.—A second specimen agrees with the type in prac- tically all details. It is slightly smaller, as shown in the table of measurements. The tympanum is less distinct and the de- 187897——4 180 The Philippine Journal of Science 1922 velopment of the post-temporal folds is not so marked, nor is the groove so pronounced. The spécimen is a male and has a subgular vocal sac. Remarks.—The two specimens listed above were taken in rotting logs. They were located by their loud raucous call, which differs distinctly from the call of Kaloula rigida sp. nov., found breeding in the same immediate locality. The two species are known to the Kalinga people by the name gd-ko. This species seems to be related to Kaloula baleata (Miiller). From the description and drawing of that species given by Boulenger it differs particularly in the shape of the fingers pads, the length of the third finger, the presence of tubercles on the palm, the number of subarticular tubercles under third finger and fourth toe, and in the distinctness and size of the tym- panum. I have remarked? that Meyer’s specimens of Kaloula baleata are the'only ones so far reported from the Philippines; there is a chance that an exchange of labels occurred, since Meyer collected the same species in Celebes. The Philippine specimens purport to come from Laguna de Bay. Kaloula negrosensis sp. nov. Type.—No, 538, E. H. Taylor collection; collected at Hini- garan, Negros, in April, 1915, by E. H. Taylor. Description.—Palatine bones forming straight ridges behind choane, the ridges widened medially and very narrowly sep- arated; choanz very large, separated from each other by a distance equal to less than twice the diameter of one, the outer edges somewhat hidden by the overhanging jaw; a strongly defined denticulate dermal ridge across palate in front of cesoph- agus; a second ridge only dimly defined; snout short, trun- cate, nostrils near anterior end; canthus rostralis rounded, loreal region slightly concave; eye as long*as its distance from end of snout; a strong fold begins behind eye and runs in a straight diagonal line across the dimly defined tympanum to in front of insertion of arm; a fold below tympanum behind angle of mouth, separated from tympanic fold by a distinct groove; skin above on body with distinct tubercles, practically absent laterally; skin on throat, breast, and greater part of belly smooth, some- what granular on posterior part of belly, on femur, and about anus. Fingers dilated into distinct, truncate disks, nearly a * Amphibians and Turtles of the Philippine Islands, Bureau of Science publication 15 (1921) 194. 21,2 Taylor: Herpetological Fauna, I 181 half wider than digits; first finger short, not reaching pad of second; fourth finger longer than second, reaching the last sub- articular tubercle on third; subarticular tubercles large, not clearly differentiated; three carpal tubercles very distinct; tips of toes not or but slightly widened; a strong inner metatarsal tubercle nearly the length of first toe, and a small outer one; subarticular tubercles well defined, two on each of the outer three toes; toes nearly one-third webbed. Tibiotarsal articula- tion reaches tympanum when adpressed. Color in life-—Brownish with very indistinct darker markings above; rather brownish below. Measurements of the type and the cotype of Kaloula negrosensis sp. nov. No. 588, | No. 588A, type. cot ype. mm. mm. BOUL 60 WONG oS coe ra ee ee nds eb coue esas cs 80.0 29.5 ieehg th of head. t2 aoe ee ee ee ee i cc dak 8.5 8.5 Width of head______- FES DU 5 PPA Rae Py gL Oe GW Ce a 11.0 10.0 Diamneber a Ors Se ee ee ee 8.6 3.5 Length of srgue i ee ee 3.5 3.5 Width of bias; oyeGtONt oc oie. shoes, tac a ee sae 17.0 13.0 Forelimbc2scsssc0. 20.0 19.0 Loorigeit finger os. oe ee ees ca a ee 8.2 8.0 Phe oS ee ae ee 39.5 39.5 Ae loi oc i wedded we cns os os 13.0 12.6 Longest toe, to metatarsal tubercle......-..--------------- 12.5 11. ga Variation.—The cotype, also from Negros, shows slight varia- tion; the body and head are slenderer. The fourth toe on the right foot is abnormal, being no longer than the third but much broader and with a wider pad at the tip; the markings on this specimen are distinct. There is a large brown spot on the an- terior part of the back with two narrow branches to each eye; two short, truncate, lateral branches above arms and two short posterior branches. Two equal-sized spots are separated from the ends of the posterior branches of the dorsal marking and there is another pair of spots in front of the anus. A distinct bar crosses femur, tibia, and foot, apparently continuous when the leg is folded. There is a dark brown area in the loreal region and another brown area laterally, beginning behind the tympanic fold. The belly is uniform light olive brown. Remarks.—The specimens were taken moving about at night in the street. Only these two specimens were found in my two 182 The Philippine Journal of Science 1922 years’ residence in Negros. The specimens were mentioned in my former publications on Philippine Amphibia.* : A further comparative study of the material has convinced me that it represents a species distinct from K. conjuncta. It differs from Kaloula kalingensis in the different length of the toes, the much narrower pads and their shape, in markings and coloration, and in the amount of webbing. The foot is distinctly broader in K. negrosensis. Bufo megregori sp. nov. — Type—No. 1468A, E. H. Taylor collection; collected near Pasananka, Zamboanga, September 30, 1920, by E. H. Taylor. Description of type—(Adult male.) Choanz small, nearly hidden under the overhanging jaw; tongue elongate, oval, free behind; openings of vocal sacs elongate slits; snout bluntly con- ical when viewed from above, extending beyond lower jaw and sloping obliquely downward; a distinct keel from tip of snout to mouth; canthus rostralis distinct, loreal region nearly per- pendicular, with a depression behind nostril; median area on snout with a broad shallow groove; two slight, rounded ridges on the interorbital area; eye large, prominent, much longer than its distance from end of snout; no evidence of tympanum; a strong constriction a short distance behind eye, separating head from body, represented dorsally by two distinct depressions in front of shoulders; depressions surrounded by glandular skin; snout, eyelid, temporal area, and back very strongly tubercular; upper lip rather glandular; sides and limbs strongly tubercular; belly and throat very strongly granular; fingers long, well de- veloped, the first shorter than second; finger tips swollen, not or very slightly wider than digit; subarticular tubercles dim; two large carpal tubercles; tips of toes swollen into small pads, not widened; the web extends to pads on all but fourth toe, where it reaches to base of last joint; subarticular tubercles dim; three metatarsal tubercles, a prominent, rounded, outer tubercle, and two flattened,. inner tubercles subequal in size; the membrane on outer side of first toe continues on inner side of leg; leg brought forward tibiotarsal articulation reaches to anterior corner of eye. Color in life-——Above deep olive, variegated with lighter and darker areas; throat blackish; belly muddy, underside of femur dirty white; scattered, minute, cream-colored spots on belly. ‘Philip. Journ. Sci. 16 (1920) 327; Amphibians and Turtles of the Philippine Islands, Bureau of Science publication 15 (1921) 194. 21,2 Taylor: Herpetological Fauna, I 183 Measurements of Bufo megregori 8p. Nov. mm, Snout to vent 37 Width of head 12 Length of head 13 Length of snout 4.5 Length of eye 5.2 Interorbital area é 4.3 Upper. eyelid 3.5 Forelimb 26 Longest finger 11.3 Hind limb 57 Longest toe, to metatarsal tubercle 15 Femur 18 Tibia 19 Variation.—Practically all the specimens agree in detail with the characters given in the description of the type. The sharp ridge on the end of the snout, the slight ridges between the eyes, the constriction of the neck behind the temporal region, the rounded depression behind, the absence of any indication of tympanum, the groove on the snout, the presence of strong tu- bercles over the dorsal surface, the presence of three metatarsal tubercles, and the extent of the webbing on the fourth toe are ‘characters strongly evident in every specimen. The largest specimen is 40 millimeters long. The belly is usually pinkish flesh to dirty white, marbled with darker. Certain of the speci- mens have indications of dark markings on the back and the limbs with light and dark bars. Remarks.—This species is related to Bufo muelleri Boulenger, from which it differs in the constriction of the neck, with rounded depressions in the post-temporal region (characters apparently absent in B. muelleri); in having the entire upper surface of the body very strongly tubercular (smooth in B. muelleri) ; three instead of two metatarsal tubercles; and a lesser extent of webbing between the toes. Specimens were found in Tumugao River, Zamboanga, Min- danao, about 1 kilometer above the waterworks’ intake. They were discovered clinging to spray-moistened rocks, in midstream. When disturbed they dived into the swift-flowing water and took refuge under stones at the bottom. Sixteen specimens were obtained. Found only in this locality. The species is named for Richard C. McGregor, ornithologist of the Bureau of Science. 184 The Philippine Journal of Science 1922 Megalophrys hasselti (Tschudi). Leptobrachium hasselti TscHupI, Class. Batr. (1838) 81. No. 1597A, E. H. Taylor collection; collected at Abung-abung, Basilan, October 21, 1920, by E. H. Taylor. Description—Vomerine teeth wanting; choane large, not concealed by jaw, separated by a distance equal to distance of nostril from eye ; palatal ridges behind choane low; tongue with a distinct nick; head large, about as long as broad; snout rounded, not projecting beyond lower jaw; canthus rostralis dis- tinct, not or but slightly rounding; nostril pierced halfway between tip of snout and eye; loreal region very slightly con- cave, strongly oblique; interorbital space broader than upper eyelid; eyes large, prominent, extending beyond edge of jaw in profile; tympanum feebly distinct, separated from eye by a - distance equal to one-half its diameter, slightly more than half the diameter of eye; a strong fold above tympanum to behind angle of jaw above insertion of arm; no trace of dorsolateral folds; skin on head and above eyes distinctly granulate; fingers obtuse, not swollen at tips; first finger equal to or slightly longer than second; third twice as long as second; subarticular tuber- cles irregularly elongate, sometimes coalescing into elongate ridges; two strongly defined carpal tubercles, the inner larger than the outer; toes obtuse at tips, not enlarged; subarticular tubercles irregular or forming ridges, usually distinct under longest toe; a very slight web, scarcely discernible, at base of toes; inner metatarsal tubercle moderately prominent, oval; no outer tubercle; tibiotarsal articulation reaches to near angle of jaw; foot slightly longer than head; skin on sides of body granular, with series of larger tubercles; chin and throat gran- ular; belly smooth; underside of femur slightly granular; two strongly developed, large, flat tubercles, or glands, on breast near insertion of arm. Color in life—Dark purple above with no trace of spots; the granules above slightly lighter; sides of belly with each granule cream yellow, the larger tubercles surrounded by dim black rings; underside of throat dull purplish, mottled with small dots of cream; middle of belly dirty yellow, unspotted; tubercles at base of arm yellow. Variation—A second specimen obtained in the same locality (No. 1597) agrees in practically all details save that distinct, islandlike markings of black or dark purple are evident on the back, as are spots on the lips and snout. 21,2 Taylor: Herpetological Fauna, I 185 Measurements of Megalophrys hasselti (Tschudi). No. 1597A.| No. 1597, mm, mm. Shout tO venti ii 6.cl i See ie eae 55.0 50.5 Length of head, to jaw angle. ______...___. 25.0 23.0 WIGUR OF BOBO coe cea cece cae 23.0 22.5 ength of snout. siccco ce ai ak ce te 10.5 9.5 Dismeter of 696 sc ss esis see a ele ee eS 8.0 7.5 RU CMIOI ORD WN is chase a ce ae a edad 5.5 7.0 Diameter of tympanum -_.____- Sue 4.2 4.7 SOV S U0 CHINDRNME ioe inc dn cddik vad dc dtuiaw deeds Ce dadenccedeee 3.5 3.0 Prelim bec nc suns cockciectinds seuss ect ccieccuswisdedclac saddaaen. 33.5 34.2 RIN soy aaah dies on Can oe elec mketigie Dé kennels 12.5 12.0 Hind lishh, Tront Ven’. 3. cso 5st 58.0 60.0 Feninr: -255. 21.0 20.5 FA caliatiieievatiderbecdabubaseinite ovcape 19.0 18.5 Tibia, to tip of longest toe 26.0 25.0 Remarks.—These specimens agree fairly well with the pub- lished descriptions of Megalophrys hasselti (Tschudi). The two specimens are males; both have internal vocal sacs, the openings of which are far back and concealed by skin fold. The call of the species sounds like nothing so much as the harsh raucous squawk of a chicken. The calls were heard while collecting in dense forest at night. Efforts to find them at night were fruitless. The spot was visited by daylight and after sweeping leaves from over a re cacnoraho i area the two specimens were found. This is the first record of this species for any Philippine local- ity save Palawan. I suspect that it also occurs in Mindanao. Gekko porosus sp. nov. Type.—No.—, E. H. Taylor collection; collected on Itbayat Island, Batan Islands (between Luzon and Formosa), November 21, 1921, by G. F. Lopez. Description of species——Snout moderately pointed; rostral, bent back over snout, borders nostril; posterior part depressed medially, but raised in front of nostrils; two short sutures enter rostral posteriorly in the depressed area; two large internasals border rostral, separated by a single small scale; two distinctly enlarged postnasal scales border nostril; a nar- row depression between nostrils, continuing backward on forehead where it widens; two or three rows of somewhat en- larged scales beginning at internasals and following along the elevated borders of the depression, a group of somewhat en- larged scales in front of eyes, not joined with the. other series; * 186 The Philippine Journal of Science 1922 granules on supraocular regions larger than those on interorbital or occipital areas; twelve upper labials, last one or two very small; ten lower labials; mental large, distinctly triangular, bordered behind by two large, elongate chin shields at least twice as long as wide; chin shields distinctly pointed in front and truncate behind, in contact with anterior lower labial for more than half their length and forming a mutual suture equal to nearly four-fifths their length; an enlarged scale separates the chin shields from second labial; two or three other en- ‘larged scales bordering chin shields; granules on chin and throat . uniform in size; body above covered by minute granules inter- mixed with small, rounded tubercles arranged in sixteen irreg- ular rows; only a faint suggestion of a skin fold limits the abdominal region; forty-two rows of cycloid imbricating scales on abdomen; tubercles on upper surface of legs; two rows of tubercles continue on tail for about one-third of its length; tail with annulations distinctly marked on anterior part by straight, regular, transverse rows of scales; twelve or more irregular transverse rows to each annulation; tail below with widened scales of practically uniform length and width; scales in the swollen area behind anus distinctly enlarged; occipital region with an irregularly edged, depressed area; a slight prominence above auricular opening; digits widened with undivided trans- verse lamelle below, fifteen under fourth toe of each foot; adpressed hind leg reaches beyond elbow of adpressed foreleg; scales preceding vent slightly enlarged, growing somewhat smaller immediately in front of pore scales; a very long series of preanal and femoral pore scales meeting medially in a broad angle; femoral pore scales continue to near end of femur, each scale with a distinct circular depression; eighty scales, forty on each side; ear opening irregularly oval, its diameter about one- third of eye. Measurements of the type of Gekko porosus sp. nov. : mm. Total length 111 Snout to vent 50 Tail 61 Snout to foreleg 22 Snout to eye 7 Axilla to groin 21 Foreleg 17 Hind leg 22 Diameter of eye 5 Length of head 16.5 Width of head 10 21,2 Taylor: Herpetological Fauna, I 187 Color (in alcohol, freshly preserved).—Above gray, with darker transverse blotches, somewhat darker on the sides of the neck; no stripes or markings on the head; tail distinctly barred with dirty white on the latter third, dimmer on the median third, and very indistinct or wanting on the basal third; limbs mottled with darker gray, apparently without bands; below dirty flesh white; lamellz under toes dark; no labial markings. Remarks.—The specimen was obtained from Batan Islands by Gregorio Lopez, who accompanied a relief expedition sent to the Batanes by the Government of the Philippine Islands. He stated that the specimen was caught in a house by a resident of Itbayat Island. The species is related in a general way to Gekko japonicus Duméril and Bibron and to G. swinhonis Giinther. It differs from G. japonicus in the larger number of tubercles on the back, the scalation of the forehead, the very much larger number of pore scales (which very probably represent the number of preanal and femoral pores in the males), and in the markings; G. swinhonis differs from the species here described in having no tubercles on the back and fewer preanal and femoral pore scales. From G. smaragdinus, a new species described in this paper, G. porosus differs in color and markings and in‘ the shape of the body, as well as in scalation and the number and arrangement of the pore scales. Gekko smaragdinus sp. nov. Type.—No. 260, E. H. Taylor collection; collected on Polillo Island, July 12, 1920, by E. H. Taylor. Description of type—(Adult male.) Rostral large, bent backward over snout, depressed medially but raised in front of nostril, bordering nostril; a slight suture enters rostral medially above; nostril surrounded by rostral, first labial, a supranasal, and two postnasals; an enlarged scale in contact with postnasals but not entering nostril; scales on snout equal, larger than those on occiput; upper labials larger anteriorly, becoming very small near angle of mouth, seventeen on right side, eighteen on left; the row of scales immediately above upper labials distinctly enlarged; lower labials fourteen and fifteen; mental small, rectangular; the rows of scales bordering lower labials somewhat enlarged, the two largest separated from mental by three scale rows; behind these, some distance on each side, is a row of four larger scales separated from labials by one scale row; body above covered with minute, equal-sized 1922 The Philippine Journal of Science 188 81-81 7 8°9 -7e OB gT-8t - | 9T-St | 9°8T 8°8 TZ $°LT ve 08 09 08 & |" Vvp92 61-8. | FP 9 wor «6=«| 8 SI-T | 8T-9T a 28 ‘ed LI 9% 62 9g ¥8 P |" a92 8I-8I | 8°8 2°9 oz-4t {8 PI-8t | LI-LT a 8 itd g°oT 62 ” Ft “SS scam 892 SI-L— | Lb 9°S oZ-4t (8 ST-8E | 21-91 =| ST ot 9'TZ TST 82 08 99 gL eS tai 49% ae oid “ CEN? 6. QI-81 | 6I-8T ut oC oe a ee ar ee 2 ees 8I-6l | 9°F 9 St SI-FI | 6I-8T 91 ot zz st 108 6g 89 LIT Bi yee 29% SI-6t | 6% 9 12-18 «=|F QT-ST =| LT-2T a 6 rad uy 8% ¥9 99 or | © |--"Vve9e I-81 | °F 9 oz-6t | 8 I-8T | 9T-ST vr 6 bad 1ST 9% Lg ag ait £ |--- W198 6I-6l | oF 9°9 e-83 |B SI-St | 8T-9T tI 6 9° he 8% "99 9°89 att . 192 12-12 | 6% 9 12-12 {8 SI-FI | 8T-9T aT or 9°82 9°91 2% 6% 19 06 De eae 69% 81-81 g 3°9 oZ-6t «| F ST-FI | 9T-ST I oT vB uy zs 9 £9 921 & |" ag92 6I-6I g 1'9 02-6t «(| 8 9I-St | LT-9T or 9°01 9°98 9°08 08 19 19 gIT pe |" ¥992 8T-81 g 9 1-12 [5 9T-91 | GI-9T gt 8°6 ¥ 02 9°08 ooh 9°89 POL P |" aL9 61-81 | 8°F 1'9 02-61 | 8 ST-St | 2T-9t =| 8 HT or 9% st 88 89 29 OT DS meee 19% 03-6 | SF 9 12-02 = | > PI-8— | OT-9T at 6 Fz st 08 29 gg OS B®. ame ¢9z 6I-6I | 87 9 12-08 g PI-PT 9I-ST 8T or 8% LT 28 99 Q"sg Q "P21 “Ewa rd 22-12 g 4 61-08 {8 I-91 | LT-9T LI II 9% Q°LT "Is 19 29 ae SES bans Perd 08 | 67 8°9 w-86 |B gt-st | oz-9t | 9°ST or 1°88 Q°LT 9°8z 9°89 g°gg aia 2 ae ee 99% 12-03 | 8°F L’9 61-8T 8 ST-ST 8T-81 bie | 6 9°Ss LI 12 9°89 g‘a¢ PIT “OE ragga LSZ 12-12 9 9 02-0 «|B QI-8E | 9T-¥T ut 9'OT 9% g°LT 08 9°29 9°19 wr tp a4 99% SI-6t =| 1's 9 02-08 8 Las 8I-ST 91 ra a ¥Z 9°9T 8% 29 63 Tet “Re: ar 092 “ULu “UUs "UU “ULM "WU “UU “UU “miu "UU “WU oo, Sam] : iaeauigs “s[es ‘remoryy | “soddQ | : : eaqjoure] on ofa 0} hen in a voz0d ge py ear” “Bey pury | *Zojo10,7 Pgs mall LAR Se seg "yj sue] | “xeg ‘ON OPI [euvertd | sarvog -s16a¥T ‘aou “ds snuphnimus oyyay fo szunod e008 pun szuamaimsnopy 21,2 Taylor: Herpetological Fauna, I 189 granules, with no evidence of tubercles; scales on belly larger, imbricate; no evidence of a lateral skin fold; scales on posterior part of chin and throat very small, equal; tail not annulate; preanal and femoral pores continuous, forming an arch medially, seventeen pores on each side; scales anterior and posterior to pore scales only slightly larger than other belly scales; scales directly in front of anus only slightly enlarged; digits flattened through entire length, with a single row of lamelle below, seventeen under longest digit on foreleg, twenty-one under longest toe; the lamellze on base of digits broken into from two to four scalelike elements; body extremely elongate and narrow, of equal width throughout entire length; tail tapering gradually, only slightly flattened. Color in life—Body above clear yellow-green to leaf green, more green on back than on sides and on limbs; side of head yel- lowish; numerous, rounded, black spots on head and shoulders arranged more or less regularly; a series of dim, uneven, cream spots begins behind the eye and continues halfway along the body on each side; iris of eye golden yellow; distal two-thirds of tail reddish brown, to salmon below, with series of cream yellow spots bordered by black almost circling the tail; belly whitish yellow; underside of limbs very light yellow-green to greenish yellow. * Measurements of the type of Gekko smaragdinus sp. nov. mm. Total length 133 Snout to vent 62 Tail 71 Snout to foreleg 24 Axilla to groin 34 Foreleg 19 Hind leg 25 Length of head, to auricular opening 14 Width of head 10 Greatest body width 7 Snout to eye 6.7 Eye to auricular opening 4.8 Diameter of orbit 4 Diameter of auricular opening 1.2 Variation.—There are twenty-one cotypes in the collection taken at practically the same time as the type. The variati a in measurements are due to the different ages of the specimens. There are fourteen to nineteen upper labials and thirteen to sixteen lower labials. This large difference is due to the degree 190 The Philippine Journal of Science 1922 of differentiation of the small scales that border the posterior part of the mouth, after the mouth curves upward. There are three or four small scales bordering the rostral behind. There are eighteen to twenty-two lamelle under the longest toe, the variation being chiefly in the basal count. The arrangement of the preanal and the femoral pores varies in some individuals. In a few of the specimens the line curves distinctly upward at the beginning of the femoral pores. In others no such curve is evident. There are eighteen to twenty-three pores on each side. The coloration was practically uniform in the specimens when they were first taken. Some of the specimens have more black spots on the head and the anterior part of the body than others. Remarks.—The species was discovered on a species of large Caladium. The specimen took refuge under water which was held in the petiole of a huge leaf. All other specimens were taken from Pandanus trees which were growing along small streams. The spiny-leaved trees were cut and allowed to fall in the water. The specimens were finally driven from their hiding places in the leaf axils and forced to swim to land. These spiny Pandanus trees were splendid collecting places and har- bored, besides the described species, several species of frogs and lizards, - This species belongs to the section of the genus Gekko which includes G. swinhonis and is characterized by the absence of tubercles on the back. The number and arrangement of the preanal pores and the very distinctive markings easily separate the species. Many differences are evident from a comparison of descriptions. It would appear that the species is very closely associated with the Pandanus trees. This association may account for the fact that it has not been collected before. Ptychozoon intermedia Taylor. Ptychozoon intermedia TAYLOR, Philip. Journ. Sci. § D 10 (1915) 95; Lizards of the Philippine Islands (1922) 101. A female specimen of Ptychozoon intermedia Taylor (No. 1075), taken near Zamboanga, agrees with the type specimen taken in eastern Mindanao, save that there are no rounded tubercles on the back. Each annulation on the tail is marked posteriorly with enlarged spinous tubercles, much as is the type; the granules on the neck and the back are smaller. The scales under the tail are divided into two rows of enlarged scutes, while in the type occasional scales are not divided, and others are divided into more than two parts. 21,2 Taylor: Herpetological Fauna, I 191 a Measurements of Ptychozoon intermedia Taylor. a mm. Total length 176 Snout to vent 88.5 Length of head 24.5 Depth of head : 10 Width of head 19 Axilla to groin 45 Foreleg 28 Hind leg 39 Width of lateral flap 8.5 Length of femur 15 Body width 21 Snout to orbit : 11 Length of longest finger 9.5 Length of longest toe 11.5 Remarks.—It will be seen that the measurements of the second specimen agree very well with those of the type. This specimen is dark black-brown over the greater part of the body; the deep brown, wavy lines on the back are scarcely distinguishable. On the tree from which the adult was taken, two eggs were found under bark attached to the tree. The eggs were joined together. The greatest diameter of either egg was 15.5. One egg was opened to verify beyond doubt its identity. The embryo measured 28 millimeters from snout to vent; tail, 26; the narrow flap on the tip of the tail is present as in the adult. The young _is very strongly marked in a pattern similar to that in the type. In southern Basilan certain freshly laid eggs of what appears to be this species were taken in October, 1920, but no adult was seen. In 1917 I found an egg on Buluan, a small island south of Basilan, containing a double-headed embryo of what was undoubtedly this species. I lost the specimen while swim- ming from shore to my launch across the coral reef. Draco mindanensis Stejneger. Draco mindanensis STEJNEGER, Proc. U. S. Nat. Mus. 33 (1908) 677; TAYLOR, Lizards of the Philippine Islands (1922) 128. The type and cotype of this species were collected by Dr. E. A. Mearns, at the base of Mount Malindang, western Mindanao. Both specimens were males. The two specimens here recorded are from the tip of Zamboanga Peninsula, near Zamboanga, nearly 200 kilometers distant from the type locality. A male and a female were taken, both apparently belonging to Stejneger’s Species. The sexes vary greatly in color and markings. Description of adult female—Agrees with Stejneger’s de- scription save in the following points: Rostral two and one-half to three times as long as broad, bordered by eight scales; nostril vertical in a raised, truncate, conelike scale, separated from 192 The Philippine Journal of Science 1922 a rostral by two scale rows; a median series of keeled scales with indistinct posterior diverging branches represented by two rows of keeled scales in each branch; occipital region with three differentiated scales, bordered by irregularly shaped, more or less keeled scales; a few large scales in a longitudinal row posterior to orbit; tympanum entirely hidden by small scales; thirteen upper labials (eleven on left side) ; no trace of nuchal crest; no enlarged or differentiated scales on neck; mental large, roughly triangular, not as long as rostral, but much wider; twelve lower labials; all scales on chin distinctly keeled, those along labials largest; back covered with irregular-sized, keeled scales; a few dorsolateral groups of enlarged scales; gular appendage distinctly developed, the anterior outline curving, the tip bluntly curved; near tip is a small spur about 1 milli- meter long, emerging about 1 millimeter from tip; lateral nuchal membranes strongly developed. Otherwise this specimen agrees with the type in scalation. Color in life.—Delicate yellow-green above; large paper white spots on back arranged in transverse lines; chin and underside of head pale greenish yellow; tip of gular appendage cream yellow; chin with cream dots; wings, above, blackish with nume- rous Narrow lines and small dots; below, uniform dusky, without spots. Adult male——This male specimen agrees more closely with the type than does the female just described. There are fourteen and fifteen upper labials; there is a small nuchal crest. The gular appendage is longer than in the type. Color in life-—Body yellow-olive above, mottled with areas of darker and lighter color, spots on back grayish white, in transverse rows. Wings, above, brown-red to brick red with narrow lines of minute cream dots; below, immaculate brown- red, growing purplish toward the outer, upper part; belly and underpart of limbs flesh color; gular appendage vivid orange yellow. Remarks.—The development of the gular appendage in the female is greater than that in any other Draco known in the Philippines. The small spur on the tip appears to be a normal development. The two specimens were shot from trees in heavy forest, in low mountains, at an elevation of approximately 300 meters. No other specimen was seen. 21,2 Taylor: Herpetological Fauna, I 193 Measurements of Draco mindanensis Stejneger. No. 1129, 2.|No. 1180, ¢. mm. mm. eis NG Tg : Pare eee er ryidip ee a Bart Rivar ERE NVI MDa ns nate eer vee, 265 205 SOME 20 ONG os ei oe ee eee 87 90 PW sec eis hele cclou ese skblnd ee Le, hae a ae 178 4115 With Of DOGG 62. cccccse st ig Eo a he a ee 14 12 iene th OF Heat ssi og ea ek a tees 19 19 Foreleg as Be wale eae nines pha re ee eee at 42 42 WANG TOS ee ac os es es ee 61 53 CMRP Gppedaie ces oie ee ee 9 23 Snout to eye__..___ Sigal wcinie uned ae contakunecaWaiadeam 7.5 8 ® Tip missing. Sphenomorphus bakeri sp. nov. Type.—No.—, E. H. Taylor collection; collected at Haight’s place, Pauai, Benguet, Mountain Province, Luzon, April, 1921, by Charles Fuller Baker. Description of type.——Rostral large, bent back over snout, forming a curved suture with frontonasal, broadened at base; frontonasal large, barely in contact with anterior loreal, rounded posteriorly, minutely in contact with frontal; prefrontals some- what triangular, smaller than frontonasal, separated narrowly; frontal large, four-sided, the posterior tip slightly rounded, touching two supraoculars and first superciliary; frontoparietal large, single, broader than long, touching frontal; interparietal large, well developed; parietals about one and one-half times as long as broad, forming a suture behind interparietal; no nuchals; nostrils in center of a moderate-sized nasal scale; nasal in contact with one labial; two loreals, not superimposed, anterior higher and narrower than posterior, its lower tip inserted somewhat between the first two labials; two preoculars, and two small anterior suboculars; eyelid scaled, opaque; six or seven super- ciliaries; four large supraoculars, last bordered by two smaller scales, separating it from parietal; parietals bordered by three temporals; anterior much the largest; six upper labials, fifth largest, fourth longest; tympanum scaled over and indicated by a depression (unnoticeable when freshly preserved) ; mental large; five lower labials; one azygous postmental, followed by three pairs of chin shields, only first pair in contact; twenty- eight scale rows around body; two enlarged preanal scales; ten or eleven lamellz under longest toe. 194 The Philippine Journal of Science 1922 Color (in alcohol, freshly preserved) Above brown, minutely powdered with darker brown, more pronounced medially, form- ing an indistinct dorsal line; side of head, neck, and anterior part of body bluish black but the color not solid; sides of tail scarcely darker than above; chin and throat darker, mental, postmental, and chin shields each with a large light spot; belly and underside of tail dirty white; limbs light brown with darker brown areas, and with lighter spots on posterior and anterior aspect of limbs and foot. Measurements of Sphenomorphus bakeri sp. nov. mm. Total length 38.5 Snout to vent 24 Snout to foreleg 7.5 Axilla to groin 11.5 Tail (tip regenerated) 14.5 Foreleg 6.5 Hind leg 7.5 Remarks.—This small skink was collected by Charles Fuller Baker on the mountain trail at Pauai, or Haight’s place, 58 kilo- meters north of Baguio. The elevation here is about 2,500 meters. This is the highest elevation recorded in the Philip- pines for any reptile or batrachian. The species must be rare, as my two days’ intensive collecting in this locality during April, 1920, failed to reveal it. This appears to be a diminutive species, similar in size to Sphenomorphus steerei, which it resembles in a general way. It may be differentiated from known species of Sphenomorphus by the scalation of the tympanum. When freshly preserved, no depression was noted where the tympanum is normally found, but after being removed a short time from the alcohol a depression became evident. I name the species for its discoverer, Charles Fuller Baker, dean of the College of Agriculture, University of the Philippines, in recognition of his valuable contributions to the entomological knowledge of the Philippines. Siaphos herrei sp. nov. Type.—No. 208, E. H. Taylor collection; collected on Polillo’ Island, July, 1920, by E. H. Taylor. Description of type.—Rostral large, bending backward over snout, the area visible from above equal to more than half the width of internasal; latter large, broader than long, narrowly in contact with frontal; prefrontals large, narrowly separated medially, forming sutures laterally with two frenals; frontal more than one and a half times as long as broad, diamond- 21,2 Taylor: Herpetological Fauna, I 195 shaped, with a rounding point behind touching two anterior suproaculars, its width slightly less than that of supraocular region; frontoparietals large, forming a mutual suture twice as long as suture with parietals; interparietal broadly triangular, longer than broad; parietals not nearly twice as long as wide, forming a suture behind interparietal; a large temporal borders parietal, nuchal scales strongly widened, narrowing gradually after third pair; nostril in a single nasal, which is placed diag- onally; two frenals, anterior slightly higher than posterior; two preoculars, lower largest, followed by two small scales above labials; four supraoculars, the second widest, anterior and poste- rior triangular; nine superciliaries; seven upper labials, fifth entering orbit; four enlarged temporals; mental shaped like rostral, but larger; a large postmental bordered by two labials; a pair of enlarged chin shields longer than wide, forming a long suture; these followed by two pairs of divided shields; six lower labials; lower eyelid with a large transparent scale; snout one and a half times as long as length of orbit; auricular opening well defined; tympanum moderately sunk; legs weak, the ad- pressed legs failing to meet by a considerable distance; legs with five clawed digits; about thirteen lamelle under each of the two middle fingers; twenty-three lamelle under longest toe; scales in twenty-two rows around middle of body, the two median rows much widened; two distinctly enlarged preanals. Color in life-—Above light brown, with a dark brown lateral band on each side covering parts of two scale rows; the lower part of sides, belly, and underside of head and tail flesh color. Measurements of the type and cotypes of Siaphos herrei sp. nov. No. 208 No. 207. No. 209. mm. mm. mm. TOGAl GHENT son no si pe ceeon ceonde este encueees Sociene 92.0 54.0 43.2 Rit Ge Went ooo ois conc ce kececedeausalocunanmannna demerabesens 41.0 89.0 38,2 SOIREE ck cmtwee ene s sneer 2.5 2.6 2.5 HORE WO ORE eo. Ses cate 1.0 7.0 6.7 Snout to foreleg Sea 13.5 18.5 13.5. A xille to groin 222.6022 on con es 22.0 22.0 22.0 Width of head <<... i222 cc. 4.5 5.0 4.3 IN so a i i we ecko eeee 8.5 8.0 8.0 Hind leg -__. 12.0 12.2 eine Variation.—The specimens are very similar in most details. No. 207 has the internasal and the left prefrontal partially joined as the result of an injury; a very young specimen, No. 187897——-5 196 The Philippine Journal of Science 1922 211 (snout to vent, 22 millimeters), has the interparietal en- larged and the posterior supraoculars broken abnormally. Remarks.—This species differs from the other species of the genus in having a divided frontoparietal. In general it agrees most closely with Siaphos auriculatum Taylor, of Negros, and S. kempi Taylor, of Mindoro. The species is an arboreal one and was found in the root masses of the large bird’s-nest fern Asplenium nidus and in Pandanus trees. The type specimen was captured just outside of the old wall that surrounds the town of Polillo. Six specimens were found. The species is named for my esteemed friend Dr. Albert C. Herre, chief of the division of fisheries, Bureau of Science, Ma- nila, who has kindly codperated in making collections. SNAKES Typhlops suluensis Taylor. Typhlops suluensis TAYLOR, Philip. Journ. Sci. § D 13 (1918) 257; Snakes of the Philippine Islands (1922) 61, text figs. 1 and 3. A specimen (No. 1587, E. H. Taylor collection) of this rare species, heretofore known only from the type, was captured on Basilan Island. It was found under the loose bark of a growing forest tree, 2 meters from the ground. The specimen agrees with the type in most details. The following are the measurements and scale counts of the Basilan specimen: Measurements and scale counts of Typhlops suluensis Taylor. No. 1587. Type. Total length 20s os es oe cok oan cn Seea ees mm_- 390 340 Wathic oo Sse ea es a Se eae oe eee mm... 14 13 Width of taif ..2 222. -.- ee mm 5.5 5.5 Width of body----- mm 7.6 7.4 WN Hater On Te es oo cee acca mm 6.5 5.5 ‘Fadl Wicitts im tae) Sennen. os oe ac cucuanencecuncoss times__ 2.5 2.4 Body witkth tn Bidtiy length aos ois Sas sce ccacus do.... 51 46 Tail length in body length ____ do 28 26 Remarks.—The specimen agrees with the type save that the interparietal is broken into two parts. The median ventral row of scales is clearly differentiated. The underside of the tail is darker than the rest of the underside of the body. ‘The dif- ferentiation of the median ventral series of scales of this species distinguishes it from other Philippine species of the genus. Typhlops cumingii (Gray). Onychophis cumingii GRAY, Cat. Liz. Brit. Mus. (1845) 133. Typhlops cumingii TAYLOR, Snakes of the Philippine Islands (1922) 66, text fig. 4. 21,2 Taylor: Herpetological Fauna, I 197 Two specimens of this rare species were collected on Polillo Island in July, 1920. Most of the specimens of these long- tailed species that I have collected have been found in root masses of aérial ferns. Whether or not aérial root masses are the normal habitat of Typhlops I cannot say. However, my own experience in collecting has given evidence that such is the case. In Mindanao all but two specimens of the numerous species of Typhlops that I found were collected from the root masses of aérial ferns. Accordingly, in Polillo I began a sys- tematic search for the species within the fern roots, sending my assistants into trees to cut the ferns from their resting places, then cutting the tough masses to pieces when they had fallen. From more than one hundred root masses so treated two speci- mens of Typhlops cumingii, three of Typhlops braminus, two of Haplonodon philippinensis, and a new species of lizard, Siaphos herrei, were captured. Measurements and scale counts of Typhlops cumingii (Gray). No. 300. No. 299. Total length _____-- mm 453 373 fs bps ea ney raga ts Co mr cnn ae ep ge eh Ed TON ae pales ap cad mm.. 27 25 WCU OR DGG 2 ooo ee ep mm.. 8 7 CO RNS, GO ON i eee el i lg a mm.- 6 5.3 WME OF NOOO oo oe eo ce Scone oc ee ee mm.. 6.5 5.7 ODL OE ORG oes ate areca ae ee ere anes mm 4 4 Scales on belly from mouth to vent, approximately '__------.-.------ mm.. 480 496 Seales under tad occas ca. eee ee eee one cet aee mm.. 40 39 BGS TOW EON HOUT: oc5 5.7 dn a scakaciee ce cecceeusce mm 26 24 Tediaridth fii ttl leneth: occ ce sect oe oa onc see cee .times.. 4.5 4.8 Body width in body length <<< 2220... Foose csc laa cc ccc esc ncc ed G0... 56.5 53.2 Variation.—The rostral is slightly more than one-half as wide as the head and fails to reach the level of the eye by a very slight distance. Dryocalamus mecroryi sp. nov. Type.—No. 1517, E. H. Taylor collection; collected at Abung- abung, Basilan (on the southern coast), October 23, 1920, by E. H. Taylor. Description of type—Rostral distinctly broader than high, forming its longest suture with internasals, only slightly visible above; internasals about as wide as long or a little longer, form- ing their longest sutures with prefrontals, the suture with nasals curved; prefrontals about as long as wide, broadly in contact with loreal and nasal; frontal one and two-thirds times as long as broad, longer than its distance from tip of snout, slightly 198 The Philippine Journal of Science 1922 wider than the widest part of supraocular; parietals one and two-thirds times as long as wide, forming a suture for about four-fifths their length, truncate behind; nasal large, irregular, partially divided, posterior part largest, much higher than ante- rior part; an elongate loreal, more than twice as long as wide, entering orbit; one preocular; supraocular much shorter than frontal, much wider posteriorly than anteriorly ; two postoculars, upper square, touching parietal, lower higher than upper; two anterior temporals, followed by three, then by four; the last superior temporal much the largest; three temporals bordering parietals; nine upper labials, fourth and fifth entering orbit; mental much broader than deep; two pairs of chin shields, ante- rior much longer and wider than posterior, five labials touching anterior; ten lower labials, the last lower and upper labial not well differentiated; eye moderate, its diameter nearly equal to its distance from nostril; pupil vertically elliptic; scales in seventeen smooth rows, without apical pits; ventrals, 220, keeled laterally; anal single; subcaudals, 121, in two rows; head distinct from body, flattened; tail very slender. Color in life-—Snout black, growing brown on parietals; a V-shaped white mark bordering the posterior edge of frontal: _a black mark begins on posterior edge of parietals and continues on neck; a black mark below eye continues back to eighth labial; lip light, flecked with darker; chin and neck white; anterior part of lower lip flecked with dark; body with fourteen elongate, irregularly edged, black, saddlelike blotches which are narrowed laterally, reaching the edges of the ventrals; these blotches cover twelve scales longitudinally, and are separated by smaller white rings which are four to five scales long medially but cover as many as eight scales laterally, the last two white rings with black spots; tail with nine black blotches which encircle tail toward its end; belly white. Measurements of Dryocalamus mccroryi sp. nov. mm, Total length 350 Snout to vent Foe 2h Tail 92 Width of head 8 Length of head 13 Width of neck 4.5 Remarks.—The species here described fails to agree with the generic characters assigned to Dryocalamus by Boulenger, as follows: There are seventeen scale rows instead of thirteen to fifteen; no apical pits on the scales; the dentition of the max- illary differs in that there are eleven or twelve teeth, increasing 21,2 Taylor: Herpetological Fauna, I 199 in size to eighth or ninth, and then one or two smaller teeth, followed by one or two larger. The general arrangement of the dentition resembles that of Stegonotus Duméril and Bibron, but the number of teeth is much less. In general configuration, the relation of the loreal and the preocular, and even in markings, the species appears nearest to Dryocalamus gracilis Giinther. I have not ascertained whether ‘one or two more or less distinct -tooth-like knobs on the basisphenoid”*® are present or not, as I hesitate to mutilate the type. The specimen was taken lying quietly in sunlight at the base of a stump in the forest, only a few meters from the sea. Only a single specimen was found. The species is named for Mrs. Ida M. McCrory, of Manila, who has assisted me greatly in making collections. Haplonodon philippinensis Griffin. Haplonodon philippinensis GRIFFIN, Philip. Journ. Sci. § D 5 (1910) 212, text fig. 1, pl. 1; TAytor, Snakes of the Philippine Islands (1922) 126, pl. 9, fig. 18. Two specimens of this rare snake were taken on Polillo Island in July, 1920. They were found in the root masses of the aérial fern Asplenium nidus. Two other specimens ob- served in similar localities escaped. The measurements and scale counts follow: Measurements and scale counts of Haplonodon philippinensis Griffin. No. 319. No. 320. TR a ea te ecu gnuunca peberameneke mm.. 550 333 Mnoue to VONG ck. eo ee ae ee mm-. 394 242 a i eat oct ees ce ee eee ee, mm... 156 91 Width of heed cu5cficst Ae ae eccl a e dea mm.. 9 7 Ttenar tik 8 Tice i as aici el -mm. 15 il PV CRO oe eo ene eens sO en uaeas ah ptenn 210 209 Pubewudals cisco sc cscs 121 122 Remarks.—In general conformation the specimens agree with the type. No. 319 has the loreal broken in two on the right side, leaving two preoculars, and on the left the fifth labial is broken, making a third preocular; the nasal appears to be partially divided. There are 88 spots on the dorsal surface. In No. 320 the loreal is broken on both sides, leaving two pre- oculars, the loreal not entering the eye. There are 96 spots along the back. A single specimen has just been received from Itbayat, Batan Islands. It agrees with No. 320 in having the loreal broken *Boulenger, Cat. Snakes Brit. Mus. 1 (1893) 369. 200 The Philippine Journal of Science 1922 on both sides. The specimen is larger and the head is propor- tionately broader than in either of the Polillo specimens. Hologerrhum philippinum Giinther. Hologerrhum philippinum GUNTHER, Cat. Col. Snakes (1858) 186; TAYLOR, Snakes of the Philippine Islands (1922) 116, pl. 7... During my past two years’ collecting, specimens of this rare snake have been found in four localities: Polillo Island (Nos. 297 and 298); Kalinga, northern Luzon (Nos. F735, F937, F932); Mount Mariveles, Bataan Province, Luzon (Nos. 1781 and 1782) ; and Mount Maquiling, Laguna Province, Luzon (No. 1873). Eight specimens were taken altogether. Measurements and scale counts of Hologerrhum philippinum Giinther. No. | Sox.| Total |Snoutto| gn, | Zength | Width ofl yentrate| Suen mm, mm, mm, mm, mm. Pil aaescsavbakcwis 324 256 68 14 8 136 47 298. 178 142 36 9.5 5 188 48 Oe oe 2 360 286 74 14 8 151 49 so Fog CGL Ay SURG BS 2 402 825 77 15 8 151 46 DOE icin satin cone 9 354 287 67 14 ‘: 154 46 st. 1 Seeger niied steamer] Saaeee ge 305 249 56 13 5 150 43 Piles Sie eae ge eS 274 228 246 12 a 153 42 SelOnsatusceupene wee lerse. 210 165 45 9 6 146 49 8 Tip missing. Variation.—The markings agree very well with those given in the drawing of the type; the color varies rather markedly. -No. 735, bright reddish to orange brown, darker anteriorly; belly bright reddish salmon, lighter anteriorly ; underside of head dusky, with milk-white spots; stripe on side of head cream yellow. The specimen was found crawling in the open along a path. Nos. 1781 and 1782 when taken were reddish brown above, below dirty white, growing pinkish anteriorly. These specimens were taken under rocks in the edge of a small brook. No. 1873 was taken under a log. The cream line on the side of the head has a pink area below it. Dim lines of minute yellow dots are present on each side of the anterior dorsal part of body; belly uniform coral to red. The blackish dots on the outer edge of ventral scales are associated with a small milk- white spot. Pseudorhabdium minutum sp. nov. Type.—No. F772, E. H. Taylor collection; collected at Bal- balan, Kalinga, Mountain Province, Luzon, April 25, 1920, by E. H. Taylor. 21,2 Taylor: Herpetological Fauna, I 201 ? Description of type——Rostral very narrow,.small, higher than wide, minutely visible above, not as wide as mental; internasals five-sided, small, about one-third the size of prefrontals, the sutures with nasal and prefrontal equal, forming their shortest sutures with loreal; prefrontals five-sided, entering eye, sutures with frontal and loreal nearly equal; the sutures formed with internasals transversely straight, those with frontal together form a slight angle; frontal broader than long, the anterior edge only very slightly angular, not reaching anterior to eyes; parie- tals more than twice as long as wide, forming a mutual suture for more than half their length; nostril pierced in a partially divided nasal; postnasal present on right side (fused with loreal on left) ; loreal large, much elongate, entering eye on right side, failing to do so by a short distance on left side; supraocular small, as long as eye, longer than broad; a small postnasal; five upper labials, third and fourth entering eye, fifth very large, touching parietal; two temporals posterior to fifth labial border- ' ing parietal; a scale directly behind fifth labial might be re- garded a sixth labial save that it appears to be behind angle of mouth; mental moderate, touching anterior chin shields, which are nearly three times as long as wide; posterior chin shields about half as long as anterior, forming a mutual suture for less than half their length; five lower labials, three touching anterior chin shields; ventrals, 140; anal single; subcaudals, 15; scales smooth, in fifteen rows, without apical pits. Color in life—Deep ultramarine with yellow-green to blue iridescence; a spot on the chin cream; ventrals each with a bluish band followed by a lighter area; parietals darker than body; a few yellowish spots in front of anus; underside of tail colored as on back. Measurements of Pseudorhabdium minutum sp. nov. mm. Total length 150 Snout to vent 139 Tail 11 Width of head 5 Length of head, to parietals 6 Width of body 5 Remarks.—The species is related to Pseudorhabdiwm mena- marz Taylor but differs in numerous characters. The tail of this new species is proportionately much shorter, with fewer ventrals; the markings and color are different; the frontal is shorter and truncate in front, and the rostral is smaller. It is impossible to tell whether the presence of a postnasal is the 202 The Philippine Journal of Science 1922 normal condition sr not. Only the type was found. It was taken under a small log, in deep forest, immediately behind the town of Balbalan. Typhlogeophis ater sp. nov. Type.—No. 1103, E. H. Taylor collection; collected near Pasa- nanka, Zamboanga, Mindanao, September 28, 1920, by E. H. Taylor. Description of type.—Snout pointed, rostral higher than broad, distinctly visible above; internasals small, about equal to one- fourth the prefrontals, their mutual suture less than one-third the length of that between prefrontals, in contact with second labial; prefrontals large, touching two labials laterally, and ocu- lar scale; no loreal; no preocular; frontal much broader than . long, the anterior edge forming a straight transverse line on a level with eyes; frontal about one-half the length of parietals, very broad, in contact with ocular; parietals elongate, nearly twice as long as wide, forming a mutual suture for more than half their length; nostril between first labial, which is fused with the anterior nasal, and a small nasal; five upper labials, third and fourth bordering ocular, which covers eye; the scale has a rounded, transparent prominence on anterior part, through which the eye is distinctly visible; no postocular distinct from ocular; no anterior temporals; one large posterior temporal bor- dering parietals, with two enlarged scales below it; no supra- oculars; mental small, three times as wide as deep; two pairs of chin shields, the anterior about three and one-half times the size of second pair; three labials touching chin shields (four on right side) ; five lower labials. Ventrals, 113, not keeled or angular ; subcaudals, 33; anal single; tail slender, pointed; scales smooth, in 15 rows, without apical pits. Measurements of Typhlogeophis ater sp. nov. mm. Total length © 173 Snout to vent 143 Tail 30 Width of head 4.5 Length of head 8 Width of body 4 Color in life-——Uniform blackish brown, somewhat iridescent above ; belly and region under tail slightly lighter blackish brown; head colored like body. Remarks.—The genus Typhlogeophis was created by Giinther for T. brevis Giinther, a species founded on a single specimen collected by A. Everett, supposedly on Mindanao or Dinagat 21,2 Taylor: Herpetological Fauna, I 208 Island. The species here described differs markedly from T. brevis. In T. ater there are 40 ventral scales less on the belly and 18 more subcaudals; the frontal is wider than long; the eye is visible through the transparent ocular; there is a striking difference in the relative sizes of the chin shields; the tail is long and slender; the scales and shields are not white-edged. The ocular scale appears as a fusion of two elements, the post- ocular part is distinct, and there is a slight depression between this and the rounded, moundlike, transparent part covering the eye. ’ My specimen was taken in a small brook that empties into Tumugao River just above the waterworks’ intake near Pasa- nanka, Zamboanga. It was found under a partly submerged log in wet earth. It was very active, and escaped over a waterfall. It was found again only after a long, diligent search. Calamaria joloensis sp. nov. Type.—No. 1855, E. H. Taylor collection; collected in cen- tral Jolo, October 30, 1920, by E. H. Taylor. Description of type.—Rostral broader than deep, distinctly visible from above; prefrontals large, longer than wide, form- ing lateral sutures with nasal, two labials, and preocular; frontal six-sided, very pointed behind, large, distinctly longer than its distance from end of snout, longer than wide, at least three and one-half times the width of supraocular; parietals large, form- ing a mutual suture for little more than half their length; nasal extremely small, triangular; one small preocular; supraocular not quite twice as long as wide; one small postocular; no anterior temporals; diameter of eye slightly less than its distance from mouth; five upper labials, third and fourth entering eye, fifth very large, forming a long suture with parietals; a single pos- terior temporal; two scales following parietals distinctly en- larged; mental small, in contact with chin shields; four lower labials; chin shields partially grown together ; scales in 13 rows; ventrals, 120; subcaudals, 14; anal single. Color in life—Above blue-black to purplish black, highly iri- descent; below lavender to black with lighter cream-colored areas along anterior part of belly. Measurements of the type of Calamaria joloensis sp. nov. mm, Total length 150 Snout to vent 140 Tail 10 Width of head <. 204 The Philippine Journal of Science Remarks.—This single specimen was taken on Jolo Island, under a rock in a small dry brook. I do not believe an apology is necessary for adding this species to the already very large assemblage of species of Calamaria. The very low ventral count and the absence of marking easily differentiate it from other Philippine species of the genus. The species seems to be nearer Calamaria prakkii Lidth de Jeude, from which it differs in color, the very much shorter tail, and the very much smaller number of subeaudals. Calamaria grayi Giinther. Calamaria grayi GUNTHER Cat. Col. Snakes (1858) 6; TAYLOR, Snakes of the Philippine Islands (1922) 184. A specimen of this rare species (No. 1034, E. H. Taylor collec- tion) was taken at Zamboanga, near Pasananka, in a small stream that enters Tumugao River above the waterworks’ intake. It was found under a rock which was at the water’s edge. The species varies from Boulenger’s description, in that the head is flattened and distinct from the neck, and the frontal is twice the width of the supraoculars. Color in life-—Head dirty white to yellow-cream oite a brown band between the eyes and involving the eye; grayish markings on the prefrontals and the rostral; the anterior part of the body has grayish rings, one scale wide dorsally and covering three or four ventrals below; toward the posterior part of the body the light rings are broken, and a series of gray spots follows the median line; below, the posterior part of the body is like the anterior. Ventrals, 191; subcaudals, 18; anal single. Measurements of Calamaria grayi Giinther. mm, Total length 195 Snout to vent 183 Tail 12 Width of head 5 Width of neck 4 Remarks.—In my monograph on Philippine snakes ° I state that only the types have been collected. This statement is doubtless erroneous, since Steindachner * records Calamaria phil- ippinica, which is regarded as a synonym of this species. * Snakes of the Philippine Islands, Bureau of Science publication 16 (1922) 312. ™Verh. Zool. Bot. Ges. Wien 17 (1867) 13, figs. 4-6, a paper which I have not seen. Fie. 1. Fic. 1. Fig. 1. ILLUSTRATIONS PLATE 1 Philautus basilanensis sp. nov. Photograph of cotype (No. 1699A). Actual size of specimen, snout to vent, 22 millimeters. - Philautus basilanensis sp. nov. Photograph of type (No. 1510). Actual size, snout to vent, 21 millimeters. . Philautus williamsi sp. nov. Photograph of cotype from Polillo (No. 358). Actual size, snout to vent, 15 millimeters. . Philautus williamsi sp. nov. Photograph of type from Polillo (No. 356). Actual size, snout to vent, 21 millimeters. . Philautus williamsi sp. nov. Photograph of cotype from Polillo (No. 359). Actual size, snout to vent, 15 millimeters. . Philautus williamsi sp. nov. Photograph of cotype from eastern coast of Luzon. Actual size, 15 millimeters. . Philautus zamboangensis sp. nov. Photograph of type (No. 1059). Actual size, snout to vent, 28 millimeters. PLATE 2. PHILAUTUS POLILLENSIS SP. NOV. Photograph of cotype (No. 353). Actual size, snout to vent, 19 millimeters. . Photograph of cotype (No. 352). Actual size, snout to vent, 15 millimeters. . Photograph of type (No, 351). Actual size, snout to vent, 27 millimeters. The folds across the snout and between the eyes are not normal. . Photograph of cotype (No. 350). Actual size, snout to vent, 20 millimeters. PLATE 3 Kaloula kalingensis sp. nov. Photograph of type (No. 824). Actual size, snout to vent, 36.5 millimeters. . Kaloula kalingensis sp. nov. Photograph of cotype (No. 856). Actual size, snout to vent, 34 millimeters. . Kaloula negrosensis sp. nov. Photograph of cotype (No. 538A). Actual size, snout to vent, 29.5 millimeters. The photograph fails to show very characteristic markings on the back. . Kaloula negrosensis sp. nov. Photograph of type (No. 538). Actual size, snout to vent, 30 millimeters. . Kaloula rigida sp. nov. Photograph of cotype from Baguio (No. 710). Actual size, snout to vent, 46 millimeters. . Kaloula rigida sp. nov. Photograph of type (No. 768) from Ka- linga. Actual size, snout to vent, 47 millimeters. 205 206 FIG. Fig. FI. The Philippine Journal of Science PLATE 4 1. Megalophrys hasselti (Tschudi). Photograph of a specimen from Abung-abung, Basilan (No. 1597A). Actual size, snout to vent, 55 millimeters. 2. Bufo megregori sp. nov. Photograph of type (No. 1468A). Actual size, snout to vent, 37 millimeters. 8. Bufo megregori sp. nov. Photograph of cotype (No. 1468B). Actual size, snout to vent, 37 millimeters. PLATE 5 1. Gekko smaragdinum sp. nov. Photograph of type specimen. Ac- tual length, body and tail, 133 millimeters. The very characteris- tic black spots on the anterior part of the body are scarcely evident in the photograph. 2. Gekko porosus sp. nov. Photograph of type. Actual length, body and tail, 111 millimeters. PLATE 6. DRYOCALAMUS MCCRORYI SP. NOV. . 1. Drawing of head of type, side view. xX 3. 2. Drawing of head of type, top view. xX 3. 8. Photograph of type specimen. Actual length, 350 millimeters. — PLATE 7 1. Siaphos herrei sp. nov. Drawing of head of type, top view. xX 4. 2. Calamaria joloensis sp. nov. Drawing of head of type, side view. xX 4. 3. Calamaria joloensis sp. nov. Drawing of head of type, top view. x 4. 4, Pseudorhabdium minutum sp. nov. Drawing of head of type, side view. x 4. 5. Pseudorhabdium minutum sp. nov. Drawing of head of type, top view. xX 4. 6. Typhlogeophis ater sp. nov. Drawing of head of type, side view. x 4. 7. Typhlogeophis ater sp. nov. Drawing of head of type, top view xX 4. TAYLOR: HERPETOLOGICAL FAUNA, I.] [PHILIPp. JouRN. Sct., 21, No. 2 PLATE 1. NEW SPECIES OF PHILAUTUS. TAYLOR: HERPETOLOGICAL Fauna, I.] [PHILIP. JOURN. Scr., 21, No. 2 PLATE 2. PHILAUTUS POLILLENSIS SP. NOV. TAYLOR: HERPETOLOGICAL FAUNA, I.] [Purmip. Journ. Sci., 21, No. PLATE 3. NEW SPECIES OF KALOULA. TAYLOR: HERPETOLOGICAL FAUNA, I.] [Puiip. Journ, Sct, 21, No. 2. PLATE 4. PHILIPPINE TOADS. TAYLOR: HERPETOLOGICAL FAUNA, I.] [PHILip. JourRN. Sct, 21, No. 2. PLATE 5. NEW SPECIES OF GEKKO. 9 No. SCI, Zi, JOURN, [PHIuire. TAYLOR: HERPETOLOGICAL FAUNA, I.] PLATE 6. DRYOCALAMUS MCCRORY! SP. NOV, [Puiuip. Journ. Scr., 21, No. 2. TAYLOR: HERPETOLOGICAL FAUNA, I.] NEW PHILIPPINE REPTILES. PLATE 7. COPELANDOSPHAERA, A NEW GENUS OF THE VOLVOCACEAE By WALTER R, SHAW Of the Department of Botany, College of Liberal Arts, University of the Philippines FOUR PLATES AND TWO TEXT FIGURES CONTENTS INTRODUCTION, COPELANDOSPHAERA SPERMATOSPHAB- COPELANDOSPHAERA DISSIPATRIX. RA, Description of type specimen. DIAGNOSES OF GENUS AND SPECIES. Other asexual coenobia. THE RELATIONSHIPS OF COPELANDO- Young coenobia. SPHAERA. Maturing sexual coenobia. Keys To THE GENERA AND SPECIES Observations on living speci- OF THE VOLVOCEAE, mens. The manner of birth, A freak coenobium. INTRODUCTION In collections of the Volvocaceae made in the vicinity of Manila during the latter months of 1914 and 1915 there occurs a new member of this family that has the general appearance © of a Volvox. However, its cells lack the protoplasmic connecting strands that are characteristic of Volvox. ‘This, supplemented by other differences, I take as ground for considering it as not properly to be classed as a Volvox. It differs from its nearest affinity on one side, Besseyosphaera, established by Shaw (16) and based on the “second form of Volvox’” described by Powers (07), in having its gonidia (the asexual reproductive cells) differentiated before the completion of the growth divisions and segmented before the birth of the coenobia in which they occur. It differs from its affinity on another side, Merrillo- sphaera, established by Shaw (’22B) based on Volvox carter Stein (’78) which in turn was originally described by Carter (59) under the name Volvor globator, in having the gonidia not differentiated in an early stage of the embryonic development, not becoming very large before segmentation, and not being symmetrically arranged in pairs or fours. s 207 208 The Philippine Journal of Science 1922 It may be distinguished from still another similar genus, Campbellosphaera, described by Shaw (719), by the absence of a vacancy in the layer of somatic cells of a mother over the gonidia and the daughters developed from the gonidia. In that genus the gonidia are not formed from cells in the coenobium wall at the places where they are to grow and develop into daughters, but come from the outside of the coenobium to take up places within. In the species which forms the subject of this paper the gonidia are differentiated from the somatic cells late in the prenatal development of the asexual daughters in which they are formed. As they grow they sink below the level of the somatic protoplasts, leaving a vacancy in the layer of protoplasts. This new species will be treated as the type of a new genus, to be called Copelandosphaera* as a token of recognition of the work of Edwin Bingham Copeland on the phylogeny of the ferns of the Oriental Tropics. The first specimens observed were living, and some of them appeared to be setting free their vegetative cells or somatic protoplasts, whence the name dissi- patrix that will be applied to the species. The use of this name is not intended to imply that the behavior to which it refers is habitual; nevertheless, it may serve to direct attention to a possible recurrence of the phenomenon in this or other species of the family under circumstances that may admit of following the history of the detached vegetative cells. COPELANDOSPHAERA DISSIPATRIX GEN. ET SP. NOV. DESCRIPTION OF TYPE SPECIMEN For the type of Copelandosphaera dissipatrix a mature asexual specimen, represented by Plate 2, fig. 6, and Plate 3, fig. 18, has been selected. It is in a Venetian turpentine mount of material that was collected at Pasig, near Manila, in August, 1914. The collection from which the slide was prepared was numbered XVI.2- The material was stained with a combination of nigrosin and Bismarck brown. *The use of this name was forecast in a footnote of an earlier paper (Shaw 719, p. 513). * The slide bearing the type specimen is in my possession. Slide mounts of material collected not far from the type locality in the same year have been sent to Prof. F. G. Haughwout, Bureau of Science, Manila, P.‘I., and to Prof. D. H. Campbell, Stanford University, California. Material from the same locality, bottled in glycerine, has been sent to thirty-two biologists in the Northern Hemisphere. Duplicates of this bottled material are available for distribution from my American address: Claremont, Cali- fornia. 21,2 Shaw: Copelandosphaera 209 The specimen is a globose coenobium containing nine daughters that fill the middle and hindmost thirds of the mother and cause the wall of the rear half of the mother to bulge out into the form of large, closely set bosses.? The specimen is com- pressed under the cover glass to about 400 » and measures 1,050 mn wide and 1,015 » long. The number of: cells in the mother was estimated at 52,600, the average spacing of the protoplasts being about 8.5 ». The front of the coenobium is turned away from the observer about 25°. The somatic protoplasts are ovoid or ellipsoidal. In the front of the coenobium they measure about 5 by 8». At the sides and back they are somewhat smaller. The thickness of the space occupied by the layer of somatic cells with their membranes is about 14 to 19 » in different parts of the front, and about 10 » at the back over the daughters. The nine daughter coenobia are arranged in a way that leaves one vacant space in the back part of the group. Four daughters that lie in the median optical section of the mother press the wall of the latter out to form bosses that are of a height equal to one quarter or more of the diameters of the daughters. The daughters are all more or less ellipsoidal, the smallest measur- ing 275 by 315 » and the largest 315 by 365 », the average dimensions being 298 by 322 ». An estimate of the number ‘of cells in the smallest daughter gave 15,000, the protoplasts being about 2.5 » in diameter and the spaces between about 1 x. All of the daughters are asexual. The largest contains ten gonidia of about 21 » diameter. The smallest contains 2-celled embryos of about 21 » diameter and 4-celled embryos of about 22 » diameter. The largest embryos are 8-celled ones measur- ing 30 yw. The numbers of the reproductive bodies in the daughters are: 6, 7, 8, 9, and 10 reproductive bodies in 2, 2, 3, 1, and 1 daughters, respectively. The most-advanced embryos, 8-celled, are in the four daughters that are nearest the front of the mother. The thickness of the layer of somatic cells of a daughter with their membranes can be made out by direct microscopic obser- vation of the daughter that bulges out on the right side of the mother. There is a line parallel with the peripheral boundary of the daughter and about 10 to 12 below it. This is shown in *This bossed condition of the mature asexual coenobium is partly, if not wholly, a result of the manner in which the shrinkage produced by the mounting medium has taken place. 210 The Philippine Journal of Science 1922 fig. 1. The protoplasts are in less than the outer half of the space between these two limits. Where a gonidium or embryo occurs in the median optical section of the daughter the inner boundary of the somatic membranes is deflected inward to form a lenticular thickening just thick enough to include the repro- ductive body between its inner limit and the layer of somatic protoplasts. Such a 2-celled embryo measured 18 ,» thick and about 23 » wide. This embryo, then, if my interpretation of . the line 7’ in fig. 1 be correct, occupies the center of a lenticular thickening of the body wall of the daughter that is equal in thickness to the radial dimensions of the peripheral lamella, the somatic protoplasts, and the embryo combined. Fic, 1. Optical section through mother and daughter coenobia of the type specimen in Vene- tian turpentine. In the right half of the figure the invisible bounding lamellae of the separate cells are drawn as they are supposed to be. The somatic wall, m, of the mother with its protoplasts, p, is included between its outer membrane, o, and its inner membrane, 7. The wall, d, of the daughter between its outer, o’, and inner, i’, membranes contains the layer of smaller protoplasts and the gonidium, g. Semidiagrammatic. X 1,000. None of the 8-celled embryos are in positions that are favor- able for showing the inner limit of the membranes of the coe- nobial wall. Embedded in the coenobium wall are numerous endophytic algae of a species probably belonging to the genus Chlorosphaera Klebs. They are most numerous on the front of the coenobium and are present in a great variety of stages. OTHER ASEXUAL COENOBIA A less-mature coenobium (specimen 2) with daughters nearly or fully formed is shown in Plate 2, fig. 9. This is on the same slide as the type specimen. The coenobium measures 535 by 525 », and the daughters range from 140 by 174 to 190 by 200 p. There are ten daughters. They are more nearly in two cir- cles of five each than in fours or pairs, but the members of a 21,2 Shaw: Copelandosphaera 211 circle are not all in the same plane. The daughters have ap- parently completed their growth divisions, and gonidia can be seen in all but two; these two I believe to be sexual daughters. Measurements of somatic cells were made in three daughters and of gonidia in eight. For the somatic cells the measurement is space per cell obtained by measuring a straight row of five cells and dividing by five. The gonidia in any one daughter are of about the same size. Examples showing the extremes are: : Width of somatic cells 2.9, 3.2, 3.2 «# Diameter of gonidia 5.0, 8.6, 11.0 4 The gonidial protoplasts are sunk below the level of the so- matic protoplasts. Over each daughter on the nearer side of the mother the vacancy in the layer of somatic protoplasts of the mother forms an elongated space that is surrounded by seven or eight protoplasts. Specimen 2A.—On the same slide with the previously de- scribed specimens there is a very much shrunken mother coe- nobium with six daughters that are plump. One of "these is asexual, and five are sexual. The somatic cells of the mother measure about 5 » wide. The asexual daughter measured about 225 by 255 p, consisted of about 23,000 cells about 3 » wide, and contained five gonidia that ranged from 18 to 21 » in di- ameter. The sexual’ daughters ranged in size from 230 by 245 to 230 by 265 », and in number of cells from about 13,000 to about 26,000. The reproductive bodies measured about 9 » in diameter in the sexual daughters, and in one they were esti- mated to number about 190. Most of the micrographs used to illustrate the species were made from material collected at Pasay, near Manila, from a pond called J for labeling material from that source. The material was collected October 138, 1914, and mounted in glyc- erine. The photographed specimens will be described first. Specimen 3.—Plate 2, fig. 8. This is a young coenobium with fifteen embryonic daughters in about the 64-celled stage. The photograph shows that the phialopores of the embryos, each of which is directly under the vacant space in the somatic layer of the mother, are still open. The mother measures about 346 y» both ways. The protoplasts are about 4 to 5.5 » wide. Their spacing is about 10 », and the number of cells is about 4,300. The daughters measure about 29 to 43 ,» in diameter. Specimen 4.—Plate 2, fig. 10. This coenobium, shown on a smaller scale, contains eight more-advanced embryos. The | 187897——-6 212 The Philippine Journal of Science 1922 mother measures 560 by 604 ». The cell spacing is about 9 p, and the number of cells is about 14,700. The embryo daughters measure from 62 to 75 » and have their cells about 7 » wide. The number of cells in these embryos is between 256 and 512. Specimen 5.—Plate 2, fig. 7. This coenobium, which also is shown on a smaller scale, contains four embryo daughters. The mother measures about 472 by 517 ». The cell spacing is about 12 », and the number of cells is about 6,000. The daughters measure about 125 by 118 », being wider than long, and their average cells measure about 6 by 9 » and have very thin walls. The number of their cells is about 1,024; that is to say, ten divisions have been accomplished. The capsules of the daugh- ters are separated from their somatic cells by distances of about 10 » around the equators, but are in contact with both poles of the daughters. Specimen 6.—Plate 3, fig. 11. A coenobium with ten embryo daughters. The preparation bearing this and the next specimen dried up before measurements were made. From the picture we find the dimensions to be about 590 by 620 »; the spacing of the cells, about 9 »; and the number of cells, about 16,000. The embryo daughters are about 100 and 110 » in diameter. Specimen 7.—Plate 4, fig. 16. This is a more-mature coeno- bium with only six daughters. Measurement of the photograph gives dimensions of 630 by 700 » and cell spacing about 12 x. The number of cells is about 10,900. The largest daughter measures about 200 by 230 uz. Specimen 8.—Plate 4, fig. 14. A less-mature coenobium with ten daughters. The mother measures about 680 by 720 ,; it has cells spaced about 12 » and numbering about 12,000. The daughters are about 130 to 160 » in diameter. The unclosed phialopores of the two nearer daughters can be seen in the picture. In all observed cases this opening is directly under- neath the vacant space left in the somatic layer by the cell that became the gonidium. YOUNG COENOBIA In both of the collections from which the specimens already described were selected there occur sexual coenobia. In the glycerine preparations from Pond J they are present in a great variety of stages, from unborn to nearly mature. They occur in the same mother with asexual daughters in ‘various pro- portions. Measurements of some of these will serve to show 21,2 : Shaw: Copelandosphaera 213 the condition of the two kinds of coenobia before and at the time of birth. . Specimen 9.—A very nearly mature coenobium (not figured) with one sexual and five asexual daughters is very much shrunken, though the daughters are not. The sexual daughter measures 257 by 287 y, has somatic protoplasts about 3.5 », and numerous oogonidia of about 12 ». The largest asexual daughter mea- sures 272 » each way and has six gonidia of which some are undivided and others divided into two cells. These gonidia measure about 20 to 22 ». The smallest daughter measures 215 by 236 » and has seven gonidia that measure about 18 pe Specimen 10.—An overmature coenobium (not figured) with two remaining daughters, both asexual. The mother is much snrunken and has large holes in the wall. The daughters are terete. One daughter presents a side view and measures 280 by 310 », and contains ten embryos that are 8-celled and mea- sure about 25 ». The embryos of the other daughters are of the same number, size, and stage. Specimen 11.—A glycerine preparation of material from Pond E in Pasay on October 12, 1914, bears an abundance of material of this species of which one specimen should be noted here. It is a very mature, somewhat shrunken coenobium (not figured) with three remaining daughters caught at the time of fixing the material in the act of passing out, each through its own hole in the wall of the mother. About one-fourth to one-third of each is out. The two sexual coenobia are coming out for- ward end first, and the asexual one is coming out hind end first. The asexual daughter measures 345 by 385 » and contains seven embryos that are 8- or 16-celled and measure 29 p» wide. The sexual daughters measure 365 by 440 » and 370 by 460 p» and have numerous oogonidia of about 15 to 17 pw and of about 14 », respectively. In the young sexual coenobia that have been described the reproductive cells are all of about the same size in each coeno- bium, and have apparently been formed from cells of the body layer after the last cell division of growth, for the vacant space left when the oogonidial protoplast sank below the level of the somatic layer is not much larger than that occupied by each neighboring somatic cell. The gonidia, on the other hand, appear to have been differentiated before the last division, but precisely when is still an unanswered question. 214 The Philippine Journal of Science 1922 MATURING SEXUAL COENOBIA We will pass to a consideration of sexual coenobia that are approaching maturity and describe the one* that is represented by Plate 1, figs. 1 to 3. This and the two sexual coenobia that are shown on the next plate are in the lot of glycerine preparations that were made from the Pond J material from Pasay, October 13, 1914. Specimen 12.—Plate 1, fig. 1, shows a sexual coenobium with antheridia very nearly mature. This specimen is a slightly ovoid coenobium about 610 by 660 ». The number of cells forming the coenobium was estimated to be about 17,600. The protoplasts of the cells are ovoid and measure about 6 by 7 yu in the front and 4 by 5 » in the middle and back of the coeno- bium. .The distance between the protoplasts is greatest in the front and grades to the minimum at the back. The oogonidia when counted in a camera lucida sketch were found to be 127. They are distributed in about three-fourths of the length of the coenobium and are slightly more crowded in the hindmost quarter than elsewhere. They are globular and measure about 28 » in diameter. Their outer sides are only about 10* below the outside surface of the coenobium. The antheridia that can be seen are ten, and they are not all grouped in pairs like the four shown in the photograph. Each antheridium consists of a platelet about 37 » wide and 10 » thick, slightly dished, con- sisting of about 256 closely packed sperms. The vacant spaces in the coenobium wall over the two antheridia that can best be seen are large enough to have seven protoplasts around them. They are, if different, larger than the spaces above the oogonidia, most of which seem to have not much more than six protoplasts around them. Specimen 13.—Plate 2, fig. 4, shows a nearly mature sexual coenobium on a smaller scale. The picture was taken with a focus below the middle of the specimen, and the oogonidia that show most plainly are those of the farther side. The coeno- bium measures 760 by 840 yn, the spacing of the cells is about 10.7 », and the number of cells about 19,600. Few antheridial sites can be distinguished in the coenobium wall. The oogo- “This one had been selected, at the time when the figures were made up into plates, to serve as the type of the species, but it is now thought that the characters of the genus are better shown by mature asexual coenobia with their contained daughters. *This number was obtained by finding that the diameter of a platelet is about equal to 18 sperms in a straight row. 21,2 Shaw: Copelandosphaera 215 . nidia are dense and, except one, without spore walls. They are about 46 » in diameter. The number present is 130. One oospore wall had developed around a reproductive body some- what smaller than the others, about 41 » wide, that is of a more orange color than the others. . The wall is yellowish and wavy and not very thick. The protoplast has not contracted within the wall of this spore. Specimen 14.—Plate 2, fig. 5, shows a large sexual coenobium on the same small scale used for the preceding figure. It measures 870 by 973 y»; has its cells spaced about 10.7 #, and contains about 25,800 cells. There appear to be two empty antheridial sites on the nearer and one on the farther side. The oogonidia are about 46 » wide and are all dense and without spore walls. They number about 112. Specimen 15.—In the same Venetian turpentine preparation with the type specimen there is a sexual coenobium that is figured in Plate 4, fig. 15. It is very much shrunken. It con- tains eighty-one reproductive bodies of which all but ten have more- or less-developed, smooth spore walls from which in some cases the dense portion of the protoplast has contracted. They measure about 42 to 43 ». Some of them are shown on a larger scale in Plate 3, fig. 12. OBSERVATIONS ON LIVING SPECIMENS The paucity of recorded observations in my notes on living specimens of this species is due largely to the fact that in most collections the species was accompanied by several more-puzz- ling species of the same family. Descriptive data from these notes will be given to supplement those obtained from material in glycerine and Venetian turpentine. Specimen 16.—A living specimen in a collection from Pasig, August 4,1914. This was an asexual coenobium with ten daugh- ters and measured 900 by 950 ». It was observed to have no protoplasmic connections between the cells. The surface of the coenobium was smooth. The protoplasts of the mother measured about 6 » wide and were estimated to number 13,000. Measured daughters were 220 by 270 y», 260 by 300 », and 270 by 300 ». The cells of the daughters were about 4 » in diameter, and the gonidia of the daughters about 18 » in diameter. Specimen 17.—On August 5, 1914, observations were made on living material that had been collected at Pasig the day before. “In this respect it differed from Campbellosphaera, which was repre- sented in the same collection. 216 The Philippine Journal of Science 1922 Many of the coenobia showed numerous bald spots of irregular form and varying size. From these spots the cells were gone. Around the borders of these spots some cells appeared to be partly loosened and about to escape. In the morning specimen 17, with very large daughters, was placed in a glass ring under a cover glass. By 1.30 in the afternoon cells had disappeared from a very large area, and the water in the ring was full of green, motile cells of about the same size. Specimen 18.—August 5, 1914. In material collected the day before and kept in a watch glass an asexual coenobium with eight daughters measured 1,630 by 1,860 » and was estimated to contain 20,200 cells. The daughters were all in the hindmost half of the mother and were about 370 by 440 ». The somatic protoplasts of the daughters were about 4 » wide and almost in contact with one another. The number of these protoplasts was estimated to be 14,820. The gonidia in the daughters were about 28 » in diameter. Specimen 19.—August 5, 1914. In the same material as the preceding a sexual coenobium from which the anterior fifth was gone measured (restored) 620 by 660 ». It contained oogonidia distributed beneath the remaining portion of the coe- nobium wall. The cells of the coenobium were ovoid and meas- ured about 5 » wide. The number of cells was estimated to be about 19,450. The oogonidia were about 25 » wide, and their number was estimated to be about 500.’ Specimen 20.—August 20, 1914. In material from Pasig a coenobium with nine daughters measured 1,950 by 2,250 ». The lateral protoplasts measured 4.5 to 5 », and they were 15 to 20 » apart. The number of cells was estimated to be about 25,900. The daughters were all of about the same size, 450 by 530». The reproductive bodies in the daughters were all embryos of about the same age and size. They measured about 30 » wide and were 16- or 32-celled spheres, each with an opening (phialopore) of about the size of one cell. Specimen 21.—December 15, 1917. From living material from ‘Pond E in Pasay one sexual coenobium was mounted in a hanging drop. It measured 800 by 900 » and consisted of about 10,700 cells. It contained forty-eight orange oospores and four green oogonidia, of which none were in the forward "It was observed that there were about 4 of the oogonidia per area 90 sq. # which corresponds to about 623 for the whole spheroid. From this number 123 was deducted for the anterior fifth which presumably con- tained no reproductive cells. 21,2 Shaw: Copelandosphaera 217 third of the coenobium. The oospores had a smooth outer wall. about 46 » in diameter and a smooth inner wall about 42 pw in diameter. The space between the membranes appeared to be occupied by a substance with large vacuoles. THE MANNER OF BIRTH Each daughter coenobia makes its exit from the mother through a separate opening formed in the wall of the mother. This is well illustrated by specimen 11, already described. After the departure of the daughter the opening remains with a smooth outline. This birth of the daughters through separate holes is in striking contrast with the manner of birth in another species that was first observed in the same habitat with Cope- landosphaera. In that one, the species of Campbellosphaera, as has been recorded elsewhere (Shaw, ’19, p. 494), the daughters nearest the hinder pole mature first and are liberated one by one through an opening in the back of the mother, through which all the other daughters pass in turn. A FREAK COENOBIUM oe On the same slide with the Pond E material that includes specimen 11 with an abundance of mature and other coenobia there is one specimen, a sexual coenobium with oogonidia a little over half grown, with a deep equatorial constriction, which at first sight appears to have been produced by a birth that was long interrupted when the daughter was halfway out. This coenobium is widened and shortened to 610 by 560 ». A normal coenobium of about the same age jis about 30 » longer than broad. The furrow is about 70 » deep on the nearer side of the coenobium, and the sides of the furrow are in contaet for some distance from the bottom. The median section effect is that of a closed furrow between two rounded ridges. The furrow is slightly in advance of the equator, and there is one row of reproductive cells in front of it, all the others being on the furrow or behind it. The coenobium wall, then, consists of two nearly equal parts, each with a flange on the inner side of the line of union, the parts being united by the inner edges of the flanges. Some of the measurements of coenobia and numbers of somatic and reproductive cells are given in Table 1. Certain of the figures greatly exceed the general range of those recorded, yet the table will serve its purpose even though some of the extreme figures are not above suspicion of error. 218 The Philippine Journal of Science 1922 TABLE 1.—Sizes of coenobia and numbers of cells in Copelandosphaera dissipatrix gen. et sp. nov. ASEXUAL COENOBIA. Specimen No. ee Somatic cells. tes ongpie é Be BO rear eee eee aes et cme eee aeeee 2, 250 25, 900 9 eee oak aaah S pdupeeanes bases 1, 860 20, 000 8 gt te Rant eg Oi tia Sg eS, yes Ge = 1,050 52, 600 9 20 cc cabwiucuspabuadcaenb its a eet Me aoc 950 18, 000 10 Seis ecescbettldeiece b720 12, 000 10 f Be aes Rien sees 6700 10, 900 6 Oa ie ana ee i a a a ew Me os ww ee b 604 14, 000 8 o b 590 16, 000 10 Se Gs ae a san ncee ee cous b517 6, 000 4 SEXUAL COENOBIA. Bee pcsaua teen cen nds Wephoy uecee aseoe Gc Gaes coe bos b973 25, 000 112 WO cocks antec os ace bg40 19, 600 130 Toei ee se 660 19, 000 500 ew arse ean du pac usuucounuctsuacueaedcacewacuesa b 660 17, 600 127 Be nace eee ce ae Causa SURE Te i a a b 6380 11, 500 60 Siti Scena ee cuGh amas eeucct esse as 900 10, 700 52 es ecg a ee ee ee b530 10, 700 47 BO is ciasanin ease albaignrs sienna des arian naie aca a b545 9, 600 64 Be tila hs were aan auee bale eee ear ee nie eee 6540 9, 200 50 8 Venetian turpentine preparation. > Glycerine preparation. © Not described, COPELANDOSPHAERA SPERMATOSPHAERA (POWERS) COMB. NOV. Powers (’07) gave a description of a form that he distin- guished as his “first form of Volvox” based on material collected from a shallow remnant of a prairie pond containing consider- able alkali. In the following year he extended the description to embrace the characters of material received from the State of Washington, from Missouri, and possibly from Louisiana, and named it Volvox spermatosphara Powers (’08). The name was emended by West (’10) to V. spermatosphaera. Powers’s description is the most complete ever written by the author of any new species of the Volvocaceae, and his beautiful photomi- crographic figures mark the beginning of a new epoch in the study of the members of this family. Our knowledge in this field will not begin to be satisfactory until we have similar photo- graphic illustrations of all the known species of the group. The shape of the coenobia, described by Powers (’08, pp. 145 and 151) as uniformly and strongly oval, might possibly better be characterized as strongly ellipsoidal, for they do not have one end regularly larger than the other as is commonly the case in 21,2 Shaw: Copelandosphaera 219 Volvox africanus West (’10) -which appears in West’s photo- micrographic figures and in photographs of some of my own material. The coenobia of that species are commonly more truly egg-shaped in having one end, the forward one, larger than the other, and are properly described as ovoid. In Cope- landosphaera spermatosphaera the ellipsoidal form is strongly marked in the young coenobia, and is evident before birth, except in the sperm spheres, which are spherical (Powers 07, p. 129). The size of the mature coenobia in Powers’s first collection of this species ranged between 500 and 1,000 pw. Material subse- quently collected in Nebraska gave maximum dimensions for mature coenobia with vegetative or mixed content of 600 to 650 », and for coenobia with oospores alone of about 500 pw. The smallest coenobia containing mature sperm spheres were as small as 150 » or even smaller. The most frequent size of mature coenobia in a large collection from Missouri was not far from 350 ». In the larger mothers the size of the daughters at birth was about 250 »; in smaller ones they frequently escape at 100 » or less. The number of cells in the coenobia was estimated by Powers to be between 1,000 and 3,000. When coenobia with smaller numbers of cells produce daughters with larger numbers, the progeny, though present in medium or smaller numbers, fill the entire cavity of the mother and stretch the maternal coenobial wall out of its original shape. This was illustrated by figures of overcrowded mothers (Powers ’08, pl. 24, figs. 20, 23, and 26). When, on the other hand, the number of cells in the daughter coenobia is the same as or less than the number in the mother there is ample room for the progeny in the posterior half or three-fifths of the mother. The somatic cells of the adult coenobia were stated by Powers to range from 6 to 10 », most of them being of the smaller size. The distance separating them was given as usually 28 to 40 uz, though sometimes as great as 50 ». The somatic cells in the daughters at the time of birth were stated to be about 5 By and Powers found that they increase to nearly or quite full size by the time the gonidia begin to divide. In shape the somatic cells were stated to resemble most the figures which Meyer gave for his Volvox tertius (Powers ’07, p. 139, Meyer ’96, pl. 8, fig. Z), and like them they showed no signs of any connecting protoplasmic filaments between the neighboring protoplasts. From such observations as he made 220 The Philippine Journal of Science 1922 on the cell membranes, Powers received the impression that these also resemble those of V. tertius. A well-developed stigma in each of the somatic cells about the anterior pole of the coenobium was observed by Powers (708, p. 150), and in the cells farther from that pole the stigmata were proportionately smaller, until a little back of the coenobial equator they were no longer visible as colored bodies. Powers was unable to decide whether at the posterior pole they were really quite absent, or whether one of a group of small colorless granules was the homologue of a pigment body. No recogniz- able stigma could be seen about the posterior pole even with an immersion lens. The stigmata about the anterior poles were visible in the daughters before birth. Reproductive cells of one, two, or three kinds occur, accord- ing to Powers, in the same coenobium in numbers ranging from one to twenty-five, the numbers being about the same regard- less of what kind or kinds were present. These reproductive cells are evidently differentiated at about the 256-celled stage of the development of the daughter coenobium. That such is the case appears from a comparison of the reproductive cells of a daughter coenobium with the cells of a sperm sphere found in the same mother (Powers ’07, pl. 13, figs. 13 and 14). This sperm sphere consists of about 256 roundish cells, the maximum number for such a structure. The reproductive cells in the neighboring daughter coenobium are very similar in size and shape to these sperm-sphere cells, a fact which indicates that they were formed at the 256-celled stage and remained undivided while the somatogenic cells divided three times, producing about 1,800 somatic cells. The number of reproductive cells in the daughter coenobia within four mother coenobia was determined by Powers (’07, p. 134) and they were found to occur with frequencies as given in Table 2. TABLE 2.—Distribution of reproductive cells in the daughters contained in four mother coenobia of Copelandosphaera spermatosphaera. Number of reproductive cells. Number of daughters. 15 | 16 | 17 | 18 | 19 | 20 | 21 | of | 28 | 24 b {| eee er ee sae EB lonecas 3 2 a e? 1 * he Do cea con wn deb en wneeenbae own seenenieneemn 4 2 3 pe ES i A RR 6... ------ ooo ono oe neo - - 5 =| = 1 2 Ohicstt A ee tc ee Bo ee col sce eens ee seoeesaraeee AEs Rod SON Skee Ey Dee EES ee SCOR gy 2 2 21,2 Shaw: Copelandosphaera 221 The highest number of reproductive cells found after a mod- erate search was twenty-five. In material consisting of coe- nobia averaging smaller in size than the first lot the average number of reproductive bodies is somewhat smaller, and Powers (08, p. 145) reported a marked tendency for the coenobia to contain exactly eight oogonidia (“ova”), but the number was frequently increased to eleven or more. The various combinations of reproductive bodies present in one hundred coenobia that were sufficiently mature for dis- tinguishing between the different kinds, taken as they occurred in the preparations, were recorded by Powers with the results that are given in Table 3. Sperm spheres alone in mother coe- nobia, though not found among any of this hundred, were other- wise of not infrequent occurrence. The largest number of daughter coenobia observed in one mother was twenty-two. The largest number of oospores observed in one coenobium was reported as nineteen. The largest number of sperm spheres recorded was fourteen, and these were accompanied by two vegetative coenobia. TABLE 3.—Combinations of reproductive bodies in 100 matured coenobia of Copelandosphaera spermatosphaera. Number of reproductive bodies, Number of mother coenobia. daughters. Oospores. spheres, Wa csree dates Sete ee ea Re 6s gg ee ee Bie 2S Bees ee Sere: BORA So cre ots: 1-11 so re ee mee eR Es AEN a RRR NG oe 2-13 1-7 1-8 The gonidia have, accordingly” at the time of differentiation diameters of about 9 » (Powers ’07, pl. 13, fig. 13), at the time of birth they measure about 15 to 18 » (07, p. 188), and at the time of segmentation, which is always considerably later than birth, they reach about 30 to 36 uw. Some rare cases illus- trated by Powers (’07, pl. 13, fig. 18) of gonidia in which seg- mentation was delayed until they had reached a size of 51 pe are open to the suspicion of belonging to some other species. The segmentation of the gonidia was described as proceeding more slowly at first and then rapidly until the somatic cells are produced with diameters of about 3 ,, long before the young coenobia have closed to form complete spheres (Powers 07, p. 132). From this time on the somatic cells of the young coe- 22? The Philippine Journal of Science 1922 nobia increase slowly in size throughout the entire period of growth. The oogonidia, according to Powers, are reproductive cells like the gonidia, with which they may be mixed in the same coe- nobium. The highest number counted in coenobia which con- tained them alone was nineteen. They become larger than the gonidia, reaching 51 to 54 yp. The oospores, according to Powers’s photographs, develop a thick smooth wall within which the somewhat smaller proto- plast is concentrically located. The name androgonidia I will here apply to those reproductive cells which give rise to the sperm spheres. They occur mixed with the other reproductive cells in various proportions. They were described by Powers as about 1 to 3 » smaller than the gonidia at the time of segmentation, and gave evidence of having cytoplasmic and nuclear structure different from that of the gonidia. They were found to segment more tardily than the gonidia to the extent that in their early stages the sperm spheres are one or two cell divisions behind the vegetative coenobia that develop from the gonidia. The sperm spheres formed from the androgonidia, according to Powers, are Eudorina-like spheres of 32, 64, 128, or 256 cells, all of which proceed forthwith to divide and form sperm plate- lets, there being no somatic or vegetative cells among them. They have only rudimentary cilia and no power of locomotion. During the development of the sperm spheres the diameter of the constituent cells does not fall below 6 », whereas in the vegetative spheres the somatic cells become as small as 3 py, though the reproductive cells appear to be about the size of the cells in the sperm spheres. The form of the sperm spheres, as before stated, was described as spherical, except when they were deformed by pressure. A sperm platelet consisting usually of thirty-two spermato- zoids though sometimes of sixty-four, is formed from each cell of the sperm sphere, the number of sperms being the same in all the platelets of a sphere. The sperm sphere with its plate- lets is functionally a compound antheridium and may be con- veniently so styled. The spermatozoids were described by Powers as compact, with spherical nuclei and terminal cilia (’07, p. 134). 21, 2 Shaw: Copelandosphaera 223 DIAGNOSES OF GENUS AND SPECIES Genus COPELANDOSPHAERA novum (Volvocaceae, Volvoceae) Body a spherical or ellipsoidal coenobium of biciliate cells that contain chloroplasts. The cells appear to lie in the periph- ery of a gelatinous matrix surrounded by a hyaline envelope through which the cilia protrude. Somatic protoplasts globose or ovoid, each inclosed in a thick gelatinous membrane that is more or less prismatic in form. No protoplasmic \filaments connecting the protoplasts. Asexual reproduction by gonidia that are differentiated in late embryonic stages of the coenobia producing them. The gonidia develop to relatively moderate — size before segmentation. Oogonidia and androgonidia in the same coenobia with gonidia or in sexual coenobia. Antheridia consisting of sperm platelets or of sperm spheres compounded of sperm platelets. Spermatozoids with terminal cilia. The type species of this genus is the new species Copelando- sphaera dissipatriz, described herewith from the Philippine Islands. One other species takes a place in this genus; namely, Volvox spermatosphara Powers, which was described from western North America. The diagnosis of that species given herewith is a revision of Powers’s original definition and is made in accord with the details of his description and figures. COPELANDOSPHAERA DISSIPATRIX sp. nov. Plates 1 to 4. Coenobia spherical or ellipsoidal; asexual 1,000 w, More or less (2,250 » recorded), about 300 by 350 p at birth; sexual 620 by 660 », more or less (973 recorded). Number of cells usually between 9,000 and 25,000 (52,600 recorded). Proto- plasts globose to ellipsoidal, about 4 by 5 » to 5 by 6 p in diameter; about 3 » wide at time of birth; spaced 8 to 20 p apart (center to center). Stigmata large in cells about the anterior pole, gradually decreasing in size backward. Repro- ductive cells confined to the posterior half, three-fifths, or three- fourths of the coenobia. Gonidia in asexual coenobia; 4 to 15; not arranged in pairs or fours; differentiated at about the stage before the last two or three growth divisions; reaching about 20 to 22 » and segmenting to produce embryos that are 8- to 16-celled by the time of birth or segmenting after birth. Early stages of embryos globose; later embryonic stages oblate; daughters becoming prolate before birth. Oogonidia and andro- 224. The Philippine Journal of Science 1922 gonidia in the same sexual coenobia. Oogonidia mostly be- tween 50 and 150; smaller and appearing later than the gonidia, in sister coenobia; differentiated at about the time of the growth divisions; about 18 » at time of birth; reaching about 46 ». Oospores with smooth or slightly wavy walls. Andro- gonidia few; about the same size as the oogonidia and scattered among them; each forming a sperm platelet of 256 sperms. Spermatozoids probably with terminal cilia. Habitat.—Fresh-water pools near Manila, Philippine Islands (leg. W. R. Shaw, 1914). COPELANDOSPHAERA SPERMATOSPHAERA (Powers) comb. nov. “First form of Volvox,” Powers in Trans. Am. Microscop. Soc. 27 (1907) 124-140, 146-148, pls. 11-18, figs. 1-17. Volvox spermatosphara Powers in Trans. Am. Microscop. Soc. 28 (1908) 142-151, 171-172, pls. 23 and 24, figs. 1-23 and 26. Volvox spermatosphaera (Powers) West in Journ. Quekett Mic. Club II 11 (1910) 101. Coenobia spherical or ellipsoidal; 150 to 1,000 » (more com- monly 250 to 600 ») at maturity; about 250 » or less at time of birth. Somatic cells about 1,000 to 3,000; protoplasts globose to ovoid; 6 to 10 yw in diameter (mostly nearer the smaller size); about 5 » at time of birth; spaced in adults 28 to 40 or even 50 » apart. Stigmata large in the cells about the anterior pole, gradually decreasing in size backward, and becoming invisible not far back of the coenobial equator. Reproductive cells 1 to 25, whether gonidia, oogonidia, andro- gonidia, or mixtures of two or three kinds in the same coeno- bium ; confined to posterior half or three-fifths of the coenobium; differentiated from somatogenic cells at about the 256-celled stage; about 15 to 18 » at time of birth. The gonidia segment after birth, when they reach about 30 to 36 y», and produce vegetative coenobia. The oogonia mature at about 51 to 54 un. Oospores with a smooth wall around a concentrically placed zygote. Androgonidia 1 to 3 » smaller than the gonidia, and of different constitution; segmenting more tardily; each producing a hollow sphere of 32 to 256 weakly ciliated cells that segment forthwith and produce each a platelet of 32 or, sometimes, 64 sperms. The resulting compound antheridia or “sperm spheres” measure about 80 to 180 », have no power of locomotion, and usually mature after fecundation of the eggs in the same coe- nobium. Spermatozoids compact, with spherical nuclei and termina] cilia. Habitat.—The shallow remnant of a prairie pond containing considerable alkali, Nebraska (J. H. Powers, 1903?); shallow 21,2 Shaw: Copelandosphaera 225 bodies of water, Nebraska (J. H. Powers, 1904?) ; Washington State (Elda R. Walker, 1904?) ; a shallow pond, near Rocheport, Missouri (R. H. Wolcott, 1904); and probably New Orleans, Louisiana (E. Foster, 1904?). THE RELATIONSHIPS OF COPELANDOSPHAERA It has been pointed out by Crow (18) that Stephanosphaera and Volvox bear, in the characters of their cell membranes, cell connections, contractile vacuoles, and chloroplasts, a strong resemblance to the unicellular genus Sphaerella, whereas the other multicellular genera of the Volvocaceae resemble in these characters the unicellular genus Chlamydomonas. According to Crow, the relationships of these genera would be best ex- pressed by grouping them into two families, the Sphaerellaceae and the Chlamydomonadaceae, that present two lines of evolution on parallel lines. Such an arrangement would bring into the Chlamydomonadaceae in this sense Gonium, Pandorina, Stepha- noon, and Eudorina of lower organization, Pleodorina and Bessey- osphaera of intermediate organization, and Copelandosphaera, Merrillosphaera, and Campbellosphaera of higher organization. While such a disposition may represent the course of phylogeny, nevertheless the subfamily Volvoceae, embracing all the mul- ticellular Volvocaceae, will be useful for practical purposes. A key to the genera and species of Volvoceae, which is here- with presented, will serve to show the position of Copelando- sphaera among its relatives. The nearest relative to Copelandosphaexa on the side of lower organization is Besseyosphaera, which has its gonidia not dif- ferentiated until after the birth of the coenobia. Although the sexual reproduction of the latter genus is not known, we can, without difficulty, picture it as something intermediate between that of Copelandosphaera and that of Pleodorina ecali- fornica as described by Chatton (’11), and not greatly dif- ferent from either. The nearest relatives on the side of higher organization are the species of Merrillosphaera. The relationship is so close that some people may prefer to reduce them all to one genus. This would be undesirable at the present time, for it is almost certainly true that the assumption that all of Volvox is com- prised in two or at most three species was, until the time of Powers’s work, a deterrent to the proper study of such forms of this group as were found. In identifying them, if the form in hand had rounded protoplasts it was set down as Volvox aureus, and if no protoplasmic connections could be seen it was 226 The Philippine Journal of Science 1922 assumed that they were probaly too fine to be visible with the magnification used or with the fixing and mounting media em- ployed. Until the Volvocaceae have been studied in all quarters Campbellosphaera Janetosphaera obversa aurea Merrillosphaera Volvox carteri rousseleti M. migulae V. barberi M. africana V. merrilli x i M. tertia V. perglobator < Copelandosphaera - globator spermatosphaera | Cc. dissipatrix Besseyosphaera powersi' Pleodorina ‘ californica P. illinoiensis Platydorina caudata Se Eudorina elegans Stephanoon coo Ra a Gonium pectorale Pandorina morum Spondylomorum quaternarium Stephanosphaera pluvialis G. sociale G. lacustre Chlamydomonas Sphaerella Carteria Fic. 2. The phylogeny of the Volvoceae. All the genera above the horizontal line are of this subfamily except Stephanosphaera and Spondylomorum, The latter are sometimes so treated for convenience. The genera below the horizontal line are unicellular. of the globe it will be conducive to more discriminating investi- gation of the members of the group if the species remain seg- regated in their several genera. Subgenera are not equally 21,2 Shaw: Copelandosphaera 227 adapted to this end. There is reason to believe that, as was pointed out by Powers (’08 p. 170), European workers have encountered in their own region species that have not received due recognition until found elsewhere. My present view, in December, 1921, of the relationships of Copelandosphaera is indicated by the diagram of the phylogeny of the Volvoceae given in fig. 2. The broken line from Stepha- nosphaera to Volvox is a reminder of the view advanced by Crow (18) that Volvox and Janetosphaera are allied to Sphaerella rather than to Chlamydomonas. According to his view the species shown here should be represented as forming two genetic trees instead of one, KEYS TO THE GENERA AND SPECIES OF THE SUBFAMILY VOLVOCEAE Key to the genera. 1. Coenobium a plate of biciliate cells. 2. Cells all facing with one side of plate............... Gonium. 2. Some cells facing with each side of plate Platydorina. 1. Coenobium globose, ellipsoidal or ovoid. ; 2. Cells on equatorial belt, facing outward Stephanoon. 2. Cells in entire surface of spheroid. 3. Cells all alike. 4. Inner sides of cells pyramidal Pandorina. 4. Cells globose , aes Eudorina. 3. Vegetative cells differentiated. 4. No vegetative cells in reproductive area.......: Pleodorina. 4. Vegetative cells in reproductive area. 5. Gonidia differentiated after birth Besseyosphaera. 5. Gonidia differentiated before birth. 6. No intercellular protoplasmic strands. 7. Gonidia not differentiated in young embryos. Copelandosphaera, 7. Gonidia differentiated in young embryos; becoming very large. 8. Gonidia not migratory in embryos........ Merrillosphaera. 8. Gonidia migratory in embryos............... Campbellosphaera. 6. Cells connected by intercellular, protoplasmic strands, 7. Cells round; without separate inner walls.... Janetosphaera. 7. Cells stellate; with separate inner walls................. .. Volvox. Key to the species. GONIUM 1. Cells 16 G. pectorale Mill. 1. Cells 4. : 2. Cilia vibratile throughout. : .. @ sociale (Duj.) Warm. 2. Cilia vibratile only in terminal half... G. lacustre West. PLATYDORINA Cells 16 or 32, arranged in a horseshoe-shaped plate, those of the two faces intercalated; posterior end with 3 or 5 tails.............. P. caudata Kofoid. 187897-——7 228 The Philippine Journal of Science 1922 STEPHANOON Cells 16, in two alternating rows on the equatov........ S. askenasii Schewk. PANDORINA : Cells 16 or 32, crowded, each with a single chromatophore and pyrenoid. P. morum Bory. EUDORINA Cele. 16, Soe Gr Gb isi ec teens geiesi .... E, elegans Ehrenb. PLEODORINA Cells 32, rarely 16 or 64; gonidia not more than twice the diameter of the vegetative cells which constitute the anterior quartet. P. illinoisensis Kofoid. Cells nearly 128, rarely 64, 32, or fewer; gonidia about two to three times the diameter of the vegetative cells; vegetative cells constituting the greater part of the anterior hemisphere.................. P, californica Shaw. BESSEYOSPHAERA Cells about 1,000; gonidia 10 to.78, distributed in two-thirds to four-fifths of the coenobium, developed in daughters after birth. B. powersi (Powers) Shaw. COPELANDOSPHAERA Cells 9,000 to 25,000 or more; gonidia 4 to 15; sperm platelets in monoecious COONOUIS oi C. dissipatrix sp. nov. Cells 1,000 to 3,000; gonidia 1 to 25; sperm a. pintalele in sperm spheres. C. spermatosphaera (Powers) comb. nov. MERRILLOSPHAERA 1. Coenobia strongly ovoid; gonidia in pairs; those of the posterior pair the smaller M. africana (West) Shaw. 1. Coenobia subglobose or ellipsoidal; gonidia mostly about equal and in fours. 2. Gonidia and oogonidia in separate coenobia.... M. carteri (Stein) Shaw. 2. Gonidia and oogonidia in the same coenobia. 3. Oospores of about the same number as the gonidia. M. migulae Shaw. 3. Oospores more numerous than gonidia........ M. tertia (Meyer) Shaw. CAMPBELLOSPHAERA Gonidia 8 or less; anterior gonidia usually the smaller when not all are alike C. obversa Shaw. * _‘ JANETOSPHAERA Oospores with eccentric double walls J. aureus (Ehrenb.) Shaw. VOLVOX : 1. Oospore walls angularly wavy V. globator (L.) Ehrenb. 1. Oospore walls crenate..... Se ‘.....-. VW. perglobator Powers. 1. Oospore walls spinose. 2. Sexual coenobia dioecious ao V. rousseleti West. 2. Sexual coenobia monoecious. 3. Somatic protoplasts large and broad V. merrilli Shaw. 3. Somatic protoplasts small and narrow........................ V. barberi Shaw. , 21,2 Shaw: Copelandosphaera 2929 LITERATURE CITED CARTER, H. J. (’59). On fecundation in the two Volvoces, and their specific differences. Ann. & Mag. Nat. Hist. III 3 (1859) 1-20, pl: 1. CHATTON, E. (’11). Pleodorina californica a Banyuls-sur-mer. Son cycle évolutif et sa signification phylogénique. Bull. Sci. France et Belg. VII 44 (1911) 309-381, pl. 7. CoHN, F. (’75). Die Entwickelungsgeschichte der Gattung Volvox. Fest- schr. z. Gépperts 50 jahrig. Doktorjubilaum. Breslau (1875) 34 pp., 1 col. pl. Crow, W. B. (’18). The classification of some colonial chlamydomonads. New Phytologist 17 (1918) 151-159. ' EHRENBERG, C. G. (’38). Die Infusionsthiere als vollkommene Organismen. Berlin and Leipzig (1838). (Not seen.) JANET, C. (712). Le Volvox. Limoges (1912) 151 pp. JANET, C. (’14). Note préliminaire sur l’oeuf du Volvox globator. Limoges (1914) 12 pp. KLEIN, L. (’89A). Morphologische und biologische Studien iiber die Gat- tung Volvox. Jahrb. wiss. Bot. 20 (1889) 133-211, pls. 10-12. KLEIN, L. (’89B). Neue Beitrage zur Kenntniss der Gattung Volvox. Ber. deutschen bot. Ges. 7 (1889) 42-53, pl. 3. KLEIN, L. (’90). Vergl. Untersuchungen iiber Morphologie und Biologie der Fortpflanzung bei der Gattung Volvox. Ber. naturf. Ges. Freiburg i. B. 5 (1890) 92 pp., pls. 2-6. : Korow, C. A. (98). On Pleodorina illinoisensis, a new species from the plankton of the Illinois River. Bull. Ill. State Lab. Nat. Hist. 5 (1898) 273-298, pls. 36, 37. Koror, C. A. (99). On Platydorina, a new genus of the family Volvocidae, from the plankton of the Illinois River. Bull. Ill. State Lab. Nat. Hist. 5 (1899) 419-440, pl. 38. MERTON, H. (’08). Ueber den Bau und die Fortpflanzung von Pleodorina illinoisensis Kofoid. Zeitschr. wiss. Zool. 90 (1908) 445-477, pls. 27, 28, 1 col. MEYER, A. (’95). Ueber den Bau von Volvox aureus Ehrenb. und Volvox _ globator Ehrenb. Bot. Centralbl. 63 (1895). Meyer, A. (96). Die Plasmaverbindung und die Membranen von Volvox, mit Riicksicht auf die thierischen Zellen. Bot. Zeit. 54* (1896) 187-217, pl. 8. * OverTON, E. (’89). Beitrag zur Kenntniss der Gattung Volvox. Bot. Centralbl. 39 (1889) 39 pp., 4 pls. Powers, J. H. (707). New forms of Volvox. Trans. Am. Microscop. Soc. 27 (1907) 123-149, pls. 11-14. Powers, J. H. (’08). Further studies in Volvox, with descriptions of three new species. Trans. Am. Microscop. Soc. 28 (1908) 141-175, pls. 23-26. SHAW, W. R. (’94). Pleodorina, a new genus of the Volvocineae. Bot. Gaz. 19 (1894) 279-283, pl. 27. SHAw, W. R. (’16). Besseyosphaera, a new genus of the Volvocaceae. Bot. Gaz. 61 (1916) 258, 254. SHAw, W. R. (’18). Some microtechnical methods and devices. Philip. Journ. Sci. Bot. 13 (1918) 241-261. 230 : The Philippine Journal of Science SHAw, W. R. (719). Campbellosphaera, a new genus of the Volvocaceae. Philip. Journ. Sci. 15 (1919) 493-520, 2 pls. SHAW, W. R. (’22A). Janetosphaera, a new genus, and two new species of Volvox. Philip. Journ. Sci. 20 (1922) 477-508, 5 pls., 5 text figs, SHAw, W. R. (’22B). Merrillosphaera, a new genus of the Volvocaceae. Philip. Journ. Sci. 21 (1922) 87-129, 8 pls., 1 text fig. STEIN, F. (’78). Der Organismus der Infusionsthiere. Leipzig 3* (1878) 134, WEsT, G. S. (’10). Some new African species of Volvox. Journ. Quekett Micr. Club II 11 (1910) 99-104, pl. 3. ‘West, G. S. (716). Algae. Cambridge 1 (1916) 181. West, G. S. (718). A further contribution to our knowledge of the two African species of Volvox. Journ. Quekett Micr. Club II 13 (1918) 425-428, pls. 29-30. ILLUSTRATIONS [Photomicrographs of Copelandosphaera dissipatrix sp. nov. from specimens mounted in glyc- erine and Venetian turpentine, taken by W. R. Shaw and E, Cortes at the Bureau of Science, Manila.] PLATE 1 Fic. 1. Sexual coenobium, mounted in glycerine, containing about one hun- dred twenty-seven oogonidia and seven antheridia. Four of the antheridia appear in the photograph, two presenting a surface view and two an edge view. xX 100. 2. A portion of the same specimen showing the two antheridia that present a surface view and four of the oogonidia. Most of the somatic cells are out of focus in the photograph, though some near the ends of the figure, along the line between the dark and light protoplasts, are in focus. X about 200. 3. A portion of the same specimen on a larger scale. The scale at the right is a part of a stage micrometer scale with smallest divisions of 10 u» that was photographed on the same plate and with the same adjustment of apparatus that was used for the specimen. X about 405. PLATE 2 Fic. 4. A nearly mature sexual coenobium in glycerine. The oogonidia on the lower side are more nearly in focus than those on the upper side. Number of oogonidia counted, 130. x 50. 5. A nearly mature sexual coenobium in glycerine. Number of oogo- nidia counted, 112. x 50. ; 6. An asexual coenobium, the type specimen, in Venetian turpentine. This is shown on a larger scale in Plate 3, fig. 13. x 50. 7, An asexual coenobium in glycerine. It has four daughters in about the 1,024-celled stage. x 50. 8. An asexual coenobium in glycerine. It has fifteen embryonic daughters in about the 64-celled stage. The photograph shows the open phialopores of those embryos that face them toward the observer. x 100. 9. An asexual coenobium in Venetian turpentine. It has ten daughter coenobia. Of these eight contain gonidia and in two no repro- ductive cells are yet differentiated. These are probably sexual daughters. x 100. 10. An asexual coenobium in glycerine. It has eight embryonic daughters that are in the 256- or 512-celled stage. X 50. PLATE 3 Fig.11. An asexual coenobium in glycerine. It contains ten embryonic daughters. X 100. 12. Part of a sexual coenobium in Venetian turpentine. It shows some immature oospores and some flask-shaped unicellular algae that are embedded in the membranes of the somatic cells. X 400. -281 232 The Philippine Journal of Science 13. The type specimen of Copelandosphaera dissipatrix in Venetian turpentine. It contains nine daughters which in turn contain gonidia or embryos of two or four cells. A number of flask- shaped unicellular algae are embedded in the membranes of the somatic layer. x 100. PLATE 4 Fic. 14. An asexual coenobium with ten daughters, in glycerine. x 100. 15.-A sexual coenobium in Venetian turpentine, very much shrunken, ° of which a part is shown in Plate 3, fig. 12. It contains eighty- one reproductive bodies. Of these seventy-one have begun to develop spore walls. X 100. 16. An asexual coenobium with six daughters, in glycerine. x 100. TEXT FIGURES Fic. 1. Optical section through mother and daughter coenobia of the type specimen, in Venetian turpentine. 2. The phylogeny of the Volvoceae. SHAW: COPELANDOSPHAERA. | ee o¢ ne Fs . a i®@ ats o* * e° Oe «* $ . Poy PLATE 1. COPELANDOSPHAERA DISSIPATRIX SP. NOV. SHAW: COPELANDOSPHAERA. ] [Purmrp. Journ. Sctr., 21, No. 2. PLATE 2. COPELANDOSPHAERA DISSIPATRIX SP. NOV. SHAW: COPELANDOSPHAERA. | [PHILIp. JOURN. Scr., 21, No. 2. PLATE 3. COPELANDOSPHAERA DISSIPATRIX SP. NOV. SHAW: COoPELANDOSPHAERA. ] [PHiLip. JouRN. Sct., 21, No. 2. PLATE 4. COPELANDOSPHAERA DISSIPATRIX SP. NOV. THE PHILIPPINE JOURNAL OF SCIENCE VOL. 21 SEPTEMBER, 1922 No. 3 EARLY LESIONS AND THE DEVELOPMENT AND INCI- DENCE OF LEPROSY IN THE CHILDREN OF LEPERS By LiIBorRI0 GOMEZ Of the Bureau of Science, Manila; Member of the Committee on Leprosy Investigation and Joszk AVELLANA BASA and CATALINO NICOLAS Of the Philippine Health Service; Physicians to the Leper Colony, J Culion Island, P. I. SIX PLATES Since 1906 the lepers in the Philippine Islands have been segregated on Culion Island with the view to reducing and finally exterminating leprosy. Thus, a colony for lepers has been established which at present contains a population of about five thousand. At the beginning of the establishment of the colony children conceived outside were born there; since then, on account of inter-marriage among lepers, the number of births in the leper colony has been steadily increasing. In 1916 a limited number of healthy children were isolated in a separate building known as the “Negative Children’s House,” located away from the leper population. Also, a small number were sent away from Culion Island and adopted by relatives or other persons who guaranteed their ability to take care of them: However, on ac- count of the inadequate facilities for isolation, the immense majority of the children born of leper parents were permitted to live with them and among other lepers. There were born in the colony, since 1906 to December 31, 1921, inclusive, 689 children. Of this number 333 died and 51 188431 ' 933 234 The Philippine Journal of Science 1922 were sent away from Culion, 3 of whom were returned later to the colony. There were residing at the end of 1921, 308 children born in the colony (both parents of practically all of them being lepers) the majority of whom were living among lepers; a limited number were isolated in the Negative Children’s House after having been in intimate contact with lepers for varying lengths of time. The children now living in the Culion Leper Colony, born of leper parents and raised among lepers—that is, in an ideal environment for infection—offer favorable material for the study of the early lesions and the development of leprosy. This paper is the result of the study of the records in the colony, and repeated clinical and bacteriological examinations made during the period between June, 1921, and March, 1922, of all children born up to December 31, 1921, inclusive, and now residing in the colony. MORBIDITY AND MORTALITY As a rule the healthy children of lepers show the same phys- ical and mental development as do the children of healthy parents. No records were kept in the colony of minor ailments among the children, but the general observation is that they are subject to the diseases of childhood that occur among children of the same age in healthy communities, and that they are not partic- ularly susceptible to any childhood disease. On examination of the children we were struck by the fre- quency of diseases of the skin other than leprosy. Over one- third of the number examined showed active or recent lesions of an itch of some kind, which in many cases persists for years and becomes generalized all over the body and is eczematous in nature. There are also a few others suffering from tinea, yaws, and other skin diseases. No comparison can be made with the incidence and nature ’ of skin diseases in nonleper communities, as no data are avail- able; but it seems that the “itc ” among the children of lepers is rather rebellious to treatment. The mortality among children of lepers during the first years of the establishment of the colony was very high; none of the children born in 1906 and 1907 survived. The mortality stead- ily decreased in subsequent years on account of better care and accommodation of the population. In the City of Manila from 1905 to 1918, inclusive, deaths among children under 1 year of age averaged 41.3 per cent of the total number of births. (still- births excluded).(6) In the Culion Leper Colony, up to De- 21,3 Gomez et al.: Early Lesions of Leprosy 935 cember 31, 1921, there were 290 deaths, or 42 per cent, among children under 1 year of age out of a total of 689 births, ex- cluding stillbirths. This is practically the same percentage of infantile deaths as occurs in a nonleper population. Table 1 gives an idea of the ages and the causes of death of children of lepers born in the colony. These data have been gathered from the colony records; the diagnoses cannot be con- sidered as very accurate, as no autopsies of such children were made. TABLE 1.—Causes of death among children born of leper parents in the colony, by age groups and by causes of death, from May, 1906, to December 31, 1921, inclusive. Age groups, years. Cause of death. ar ai Total. 0-1 1-2 2+ JASOB XIG NEOHRUOLUH «cock cane ce co oo opt ~ ng ca cmap ees Dh eee eusecuaten: 1 Beriberi .2525<.=.=..2 Posi he etese uecut see an eetet 2leSse 16 “ag Pothe par Srl 18 Bronchitis; acute, c2cessece os eat ces 25 kt 4 pF eae Bea 15 Br h 1 RE oe Pye vis 10 pie ae Sea 11 Congenital debility (including athrepsia, malnutrition, end maraetins) 062252225. caer es es ee cde 95 4 1 100 Diarrhosa and enteritis 2sc2ceece2eia sie So eck Barre ee | ers Su.. 3 Diphtheria sos. kn. es a scones Se he eae wn ene iecfeneaee re “slereree 1 1 Dysentery, bactliary 2... 602. - 2.2 ae 10 6 2 18 Gastritis (ooo he acne 10 lsidsdcdcaatecds cco 10 Gastrdenteritis, acute... ._..+.---.-++.+.------------------ 31 4 1 36 Gastroenteritis, chronic. --..------------------------------ 6 3 9 Infantile convulsions........../..---=-------------------- 49 49 Enfieehiee 250 |. 2 ieee de Se ec eee ace 3 Ce 7 Ill defined dite cbgee amet e Ges 24 “2 = 24 VE PS Cg ti ie conse es Demonia png aren nr ye Gus x 2 2 5 5 Teg 1 | pap pupae ace ese egies Suan es macet eeyenegeeaee 4 1 2 7 Niniietles 285s 2G ce Se Be eee 5 2 2 9 Prac os dc i as os cae ake ec liv ecen > seoeen Boiss cope] cc te. ee 3 PUR os oe daca ck ns cues eundee i eacaaes om BRE Saree SS ASS aeadee 3 Umbitical tetas ret at Bite aceon aoc 2 Tonslitls (2305 eR Re Ee PATO ey sah A empre lee cE A en wee 1 Waviclotd Lecce ohn Bole plewe de Seco nn heii ceewewen=| sew meee wt la ens-- eens BE 1 "Dotale. 6252 Se eh hee ekenlen gine aenene 290 31 12 333 _ On examining Table 1 it will be noticed that infantile debility (including malnutrition, athrepsia, and marasmus) holds the first place among the causes of death; then follow the gastro- intestinal disturbances and infantile convulsions. According to the annual reports of the Director of Health of the Philippine Islands, during the period from 1904 to 1920, there were, in the City of Manila, 8,451 deaths, or 14.6 per cent, from congenital! debility out of a total of 57,782 deaths from all diseases of 236 The Philippine Journal of Science 1922 children under 1 year of age. According to Table 1 there were 100 deaths, or 34 per cent, due to infantile debility out of a total of 290 deaths from all causes in children under 1 year of age. Therefore, the proportion of deaths from infantile debility is relatively higher than in nonleper communities, which may be presumed to be due to the weakened condition of the parents. The increase in deaths from infantile debility is partly com- pensated by the small number of deaths from infantile beriberi, which is one of the important causes of infant mortality in nonleper populations. This is due to the use of unpolished rice in the diet in the colony. THE SITE OF THE EARLY LESION Our data regarding the site of the initial lesion of leprosy are based on comparison of the physical and bacteriological exami- nations of the skin and nose. Bacteriological examination was made of the skin of all children showing suspicious or charac- teristic lesions, and in practically all of these the secretion and scrapings from the nose have also been examined. Dr. Felisa Nicolas, of the eye, ear, nose, and throat department of the Phil- ippine General Hospital, on a visit to Culion, kindly made inspec- tion of the noses of twenty-six children who either were nonlepers or showed suspicious or definite lesions of leprosy in the skin. The findings of Doctor Nicolas, compared with results of the physical and bacteriological examinations of the skin lesions and the nose, are shown in Table 2. The physical examination of the skin comprised, besides in- spection, the testing for tactile and thermal anzsthesia and the determination of areas of anhidrosis by the injection of pilo- carpine solution. Tactile anzesthesia was tested by the light ap- plication of the point of a pin and asking the child to point out the location where the pin was applied. Thermal anzsthesia was tested by applying alternately hot water and cold water in test tubes; this test was applied to older children who could give intelligent answers. To test the anhidrosis a solution of 1 milligram of pilocarpine nitrate to 1 cubic centimeter of water was prepared, and subcutaneous injection was made of an amount that was proportionate to the age of the child as figured by the formula of Young,' on the basis of 5 milligrams as the average dose for the adult; and, in addition to the pilo- carpine, hot tea was given to drink, and the child wrapped in a 412 + age on = the denominator of a fraction the numerator of which is 1. 21,3 Gomez et al.: Early Lesions of Leprosy 237 blanket and examined after the breaking out of copious sweat. The sensibility to pain was not tested because, except in extreme cases, it is hard to determine from the answers of the children the dissociation of the tactile and the pain sensations. The physical examination of the nose was made with the aid of a nasal speculum, in ordinary daylight, which naturally did not show the hind portions of the nasal cavity. The specimen for bacteriological examination was obtained by scraping about the edge of the suspicious lesion in the skin; or, in the case of the nose, by scraping the suspicious lesion noticed on the physical inspection; or, in the absence of such lesion, the specimen was obtained from the site at about the junction of the cartilaginous and bony portions of the septum. The scraping was made with a sharp knife until a small amount of serum and blood was obtained; this material was smeared over a glass slide, fixed by heat, stained by steaming carbol- fuchsin, washed in water, and decolorized with Gabbett’s methyl- ene blue. TABLE 2.—Results of physical and bacteriological examinations of the skin and nose of children of lepers, November, 1921. Examination of skin. Examination of nose. No. tinhe. Age. Bacte- Bacte- PB cen. rmena Physical. riologi- Physical. riologi- cal. cal. the Ae ts 10: | Nepntive..... oe _ Negative oo. — Nonleper. TS oe 7 | Negative; noscabies_| — Erosion of mucosa, ae Do. anterior portion septum, both sides. 8|B.M..| 6] Negative__.-.--.-...| — | Right sidenegative;| — Do. small ulcer sep- tum, left side, anterior margin of quadrilateral car- tilage. 4} A.C... Fi OG. ae Slight erosion, mu- Do. cosa, both sides of septum. §| A.N.. 8 | White area on right a Shallow ulcer, right ra Suspicious. side; scabies; no side, anterior por- pilocarpine test tion of septum; left made. side negative. 6|L. de 8| White particles on| — | Left side appar-| — Do. ja C. face and neck and ently negative; chin, which are not small nodule right anzsthetic and side septum, ante- show sweating on riormargin quadri- pilocarpine injec- lateral cartilage. tion. 238 The Philippine Journal of Science 1922 TABLE 2.—Results of physical and bacteriological examinations of the skin and nose of children of lepers, November, 1921—Continued. Ini- , tials. Examinatién of skin. Examination of nose. Age. Physical. Bacte- riologi- cal. Physical. Bacte- riologi- Diagnosis, type of leprosy. 10 ll 4 15 eC. A.V.S. gH i Saet M. de Bs: Ds M.Z _. MC. White area on right loin; no anesthesia and no anhidrosis. White spot over _ vight loin showing center, which does not sweat on pilo- carpine test. Whitish, nonanzs- thetic areas which do not sweat on face, left but- tock, and right thigh. Whitish, anhidrotic area on left loin; apparently flushed face like sunburn, but no anzsthesia. White area above right knee; pri- mary yaws; pilo- earpine test un- satisfactory. White _anhidrotic spots all over the body, which show no tactile anzs- thesia. Small whitish areas all over body with reddish centers, especially on limbs; Whitish, anzsthe- tic, anhidrotic area on back of left thigh. Diffuse reddish area from nates to foot. of left leg, both legs flushed; white anzsthetic areas on inner side of right thigh. Negative, both sides_ do. Left side negative; ulcer right side of septum, anterior portion. Uleer both sides of septum, anterior portion. Negative, both sides do. Suspicious. Macular. 21,3 Gomez et al.: Early Lesions of Leprosy 239 TABLE 2.—Results of physical and bacteriological examinations of the skin and nose of children of lepers, November, 1921—Continued. No. Ini- tials. Age. Examination of skin. Examination of nose. Physical. Bacte- riologi- cal. 16 17 18 21 Pi locs G.M.. A RON... Ws Rasa E.S... J.C. Yrs. =~ 1 10 10 10 White anesthetic spot on left cheek and anesthesia on right ulnar region; pilocarpine test unsatisfactory. White anzsthetic area on back of left side and another similar area about 2.5 centimeters in diameter on right thigh; no sweating in white area. Advanced nodular case. Reddish, slightly in- filtrated right ear; anzesthesia on left ulnar region; skin on left arm and el- bow thickened and reddish; diffuse reddish area on right forearm, Marked nodular case. Maculo-anesthetic spot on back which does not sweat. Well-developed nod- ular case. Physical. Bacte- riologi- eal, Diagnosis, type of leprosy. Right side negative; small depression anterior portion septum, left side. Uleer anterior por- tion septum, right side; with atrophic changes of the mucosa of the turbinate on both sides. Left side infiltra- tion of middle septum and corre- sponding portion of middleturbinate producing adhe- sion between the two surfaces; ero- sion at nasal ori- fices. Left side negative; small ulcer right side septum, ante- rior portion. Ulceration and infil- tration of right inferior turbinate; | two small nodules anterior portion of septum, left side. Left negative; small ulcer right side septum, infiltrated area right side of septum, anterior portion; left side negative. ; Maculo-anzs- thetic. Nodular. Maculo-nod- ular. Nodular. Nodular. 240 The Philippine Journal of Science 1922 TABLE 2.—Resulis of physical and bacteriological examinations of the skin and nose of children of lepers, November, 1921—Continued. Examination of skin. Examination of nose, No. iain: Age. Bacte- Bacte- Bilge (ewig Physical. riologi- Physical. riologi- cal. cal. Yr. : 3) AC. 8 | Rounded white + Inner aspect right +. | Macular. patches with red ala nasi with ulcer; centers on face, superficial ulcer anterior portion septum, left side; corresponding por- tion of right side slightly swollen and with scab. #4 EB R=. 7| White anzesthetic o- Small nodule ante- a2 Maculo-anzs- patches alloverthe rior portion sep- thetic. body; many of the tum, right side; spots show reddish thickened right centers. Some of ala. the spots show sweating and others do not, 25 UP 5 | White anesthetic + Nothing abnormal — Do. and anhidrotic on right side; ves- spot on back. tibuleleftside with scales. 26 PB. 5 | Some doubtful spots + Negative .c25 ce era Infiltration. and a slight infil- tration of right cheek; unsatisfac- tory pilocarpine test. On examining Table 2 one notices that there exist bacterio- logically negative erosions and ulcers in nonleprous children; that some children, showing suspicious, bacteriologically negative lesions in the skin, show erosions, ulcerations, or nodules which are invariably bacteriologically negative; that children who show skin lesions which are bacteriologically positive are not necessarily bacteriologically positive in the nose and that the nose in these cases may either show ulcerations (No. 5, A.N.) or not (No. 25, F.R., and No. 26, P.B.); that in cases in which the bacteriological findings of the nose were positive there were definite lesions in the septum; and that, invariably, bacteriolog- ically positive lesions were found in the skin in cases which are positive bacteriologically in the nose. Table 3, which summarizes the types of lesions and bacterio- logical findings in the skin and nose of the twenty-four children 21,3 Gomez et al.: Early Lesions of Leprosy 941 born in the colony who. at the present writing are unquestionably lepers, confirms the findings in the preceding table, in that in all of the twenty-four cases lesions of the skin were found positive while only thirteen were bacteriologically positive in scrapings from the septum of the nose; the other eleven were bacteriologically negative in this site. Our records show two children (No. 13, A.C. and No. 12, A.M., Table 3) who, on former bacteriological examinations, were positive only in skin lesions, and now they are positive bacterio- logically in the skin and nose. Two other children (G.D. and P.I.), showing suspicious lesions in the skin, were repeatedly examined bacteriologically with negative results and, after an interval of six months, became positive bacteriologically, only in the skin so far. Bacteriological but not physical examination was made of the noses of sixty-one other children, from 1 to 12 years old, with negative results. Their skin was also negative bacterio- logically and showed lesions; such as flushed skin of the legs only, suspicious or characteristic white patches in some of which there was a definite reddish tint indicating progressive changes, or definite anesthesia either in white patches or in regions of the body which show no skin blemishes. Considering the above data, one might conclude that the early lesion is located in the skin and not in the nose. Our results are at variance with the theory of Sticker,(7) who claimed that the nose was the site of the initial lesion of leprosy, and are in conformity with the results obtained by Brinckerhoff and Moore,(1) who concluded that the routine examination of the nasal septum and the nasal secretions is not an efficient method for the detection of leprosy in its incipient stage. THE NATURE OF THE EARLY LESION It is natural that in young children it is hard to obtain re- liable history as to subjective sensations, and the main reliance on manifestations of the disease must be looked for in the sub- jective signs in the skin. We have found that the commonest manifestation of the disease is the appearance of macular lesions, which are whitish patches resembling morphea spots and which we refer to here as “macula alba.” The macula alba are whitish light fawn in color, level with the skin, and have a smooth, nonscaly surface and irregular edges of from 1 to 10 centimeters wide. They appear singly and gradually increase in size and number, or they may break out in 24? The Philippine Journal of Science 1922 crops. The commonest locations are the trunk, buttocks, thighs, arms, and face. They simulate tinea of the skin, commonly called “pafio blanco,” a common affection among Filipinos, but from which they can be easily distinguished on account of the rounded shape, small size, and elevated furfuraceous surface of the tinea lesions when these are early and discrete; also, scrapings from tinea lesions treated with 10 per cent caustic potash solution show fungi, whereas no fungi are found in the white patches of leprosy. They are differentiated from vitiligo or leucoderma because in these there is often an increase of pig- mentation on the border of the patches producing a sharp con- trast between the lesion and the surrounding skin, which is not noticed in leprosy. In the early macula alba there is no loss of hair, no change in the activities of the sudoriferous glands as tested by the injection of pilocarpine, and no change in the sensibility to touch or to heat or cold. The majority of the macula alba are bacteriologi- cally negative, but in older patches, especially those showing a reddish tint, Bacillus lepre may be found. In some of the white patches that did not show tactile anexs- thesia there was noticed an inability to distinguish thermal differences, and in patches where both tactile and thermal anzesthesia existed the thermal was found to be more extensive. The test for tactile discrimination—that is, ability to distin- guish whether the affected region is touched by one or by two points—was not systematically applied, on account of the un- reliability of the answers of children. Injections with pilocarpine nitrate and administration of hot tea in an effort to determine the absence of sweat in certain portions of the body, especially in connection with blemishes and white areas of the skin, were not always satisfactory; as the child, being covered, had the tendency to move and wipe away the moisture. It was noticed in connection with the white areas that in some cases, after injection, the white area was absolutely smooth and dry, whereas the skin in the immediate vicinity was moist; in other cases the white areas were dry, but the imme- diately surrounding skin was also dry and the corresponding area of the skin on the other side of the body was also dry. In other cases the white areas showed definite sweating similar to the surrounding skin. Another early objective sign also seen in the leprosy of children is the shiny and flushed appearance of the legs occurring coin- cident with, or independent of, the white patches. Also, in- ¢ 21,3 Gomez et al.: Early Lesions of Leprosy 243 filtration of the ears or of parts that are frequently subject to friction and traumatism, such as the elbows and knees, may be the first manifestation of the disease. In some of the older children we were able to elicit a history of paresthesia, hypesthesia, and anesthesia as the first and, frequently, only manifestation of leprosy. Such disturbances were frequently noticed first on the peroneal side of the leg and the ulnar side of the forearm or on the external maleolus and the back of the elbow joint; and, occasionally, on close examination one can notice also isolated patches of anesthesia in other parts of the body. We were not able to see or obtain a clear history of bulle formation, which is one of the first manifestations of leprosy, pointed out by Dyer(3) and by Gwyther. (4) Hopkins(5) has already noted that the macular lesion is the most constant of the early manifestations in all types of skin leprosy and that such a macule is usually unaccompanied by symptoms, such as pain or itching, that would likely attract the attention of the patient. We cannot state that all the white patches that we noticed in children are leprous, in the absence of other functional nervous disturbances and other definite leprous lesions. It is significant, however, that such blemishes are so prevalent among the children of lepers, about one-fourth of whom, as noted in Table 5, show this type of lesion with or without concomitant functional nervous changes. On the other hand, we have seen definite evidences of progressive changes in these whitish patches; that is, in the course of time they became reddish, infiltrated, and bacteriologically positive, as shown in Plate 5, figs. 1, 2, and 3. We have also found some white patches (Plate 2, figs. 2 and 3), which were bacteriolog- ically positive, and which on their physical appearance could not be distinguished from bacteriologically negative patches (Plate 1, figs. 1, 2, and 3; and Plate 2, fig. 1). DEVELOPMENT OF THE DISEASE After the first lesions, such as white areas, anzsthesia, flushed legs, and infiltrations, the disease progresses by the aggravation of the preceding conditions. The white areas increase in num- ber, may remain white or may acquire reddish edges (Plate 4, fig. 1; and Plate 5, fig. 3) or centers (Plate 5, figs. 1 and 2), may become more or less elevated, and previously bacteriologi- cally negative patches may become positive. The infiltrations become more accentuated, and nodules appear on the ear, the » 244 The Philippine Journal of Science 1922 face, and other parts of the body. The skin of the legs becomes dry and scaly, suggesting ichthyosis. Single ulcers break out on the legs and the feet, which are chronic and difficult to heal. Anesthesia may be noticed in connection with other lesions, but occasionally this is about the only definite sign of leprosy present, the patient being bacteriologically negative and showing neither thickening of the superficial nerves nor other physical appearance indicative of trophic disturbances. The development of leprosy is suggested by the type of lesions that appear in children, according to their age, as shown in Table 3. TABLE 3.—Positive lepers, arranged in order of age. x. Bacteriology. | We. | textilale. or Age. | Sex. Lesions — according to order of | birth. ese pig Skin. | Nose. | } Yrs. PPG ot see 3/ M pe Macules white and one red macule __-__---. ay — 2\A.deL -}| 1917 “Tor One whicmacule. 202052 3.22 st + + SRR bry A= OO: Mieiien white oan oc Sa 5 5 raf 41 MG. 4h 19iT 4| M | Macules white and red_-_-_-.--....--------- - = OS ve: Saute Ree i Gy My Maecues White 225 cea a8 os? 6G. D2" 1316 5, F | Macules white and red; nodules --._--_--.-- + = $43 oy AS 6{ M | Macules white ._-_.- ss + agp Sih P as 1915 6) Pes Ps Fr Sapper es Se eee es ae —- 5} B.R | 194 7| M | Macules white and red; infiltration; anes- | + 23 thesia. a0 PP 1914 T1:°F. |sMacules- white and red2cus.2soce SS tgs + = IER. ye 4} ome | WMaculen white: ooo) cece cans nag een nance so = 122) A.M....| 1918 8| F | Macules white; anesthesia -.-----.------~-- is ss A.C...) 1918 8| F | Macules whiteand red .-__-.---.--~------..- + az a Bebo svc 1912 9.17 Mct Noduletss sess es es + “b 16} G.M....} 1912 9| M | Macules white; infiltration....-..-.-...----- + 3 16} J;C_..-.| “1912 9| F | Nodules; macules white and red; anzsthesia + + 17 | AW ..-.| 1912 9| M | Anzsthesia; nodules._......------..-------- + + 18} BiB cs.) 100)-|. 20) Mi. | Macules white ...-.2 nse .cc--be cies sone + es 19 oak, fee 104. Mi. | Aneethesin: nodules... o.oo sc al * = 20 | R.N..--| 1911 10{| F | Infiltration; anzsthesia___......--.......--- ag a CLAS hoe ae: 11 {| F {| Macules white; nodules; anzsthesia-._--.-- + a2 22; D.S----| 1909} 12] F | Macules white; anwsthesia _._._...--..----- + — 23 | E.R .-..| 1909 12| F | Macules white; nodules; anzsthesia--_--.-- + + 2) FB. 008 13 | M | Anzsthesia; nodules; macules white --.---- os + Table 3 has been arranged according to the correlative age of the children, and the types of lesions are listed in the order of their appearance as far as could be ascertained. It shows that the majority of the younger children, while bacteriologically positive in the skin, are negative in the nose; also that, as a general rule, the younger children show only macular lesions and, as the children grow older, infiltrations, anzesthesia, and 21,3" Gomez et al.: Early Lesions of Leprosy 245 nodules appear; and that, with the exception of two cases (FS. and J.C.), when macular lesions (usually whitish) coexist with other lesions, the former appear first; in four cases there were no macular patches and in two of these (J.R. and A.C.) anzs- thesia preceded the nodules, in one there were infiltration and anesthesia, and in another only nodules. The early lesion and the development of leprosy are also illus- trated by the following records of children who are bacteriolog- ically positive lepers: I—C.B. (Table 3, No. 1), Filipino, 3 years old, male. Born in Culion. Father nonleper, living outside. The mother and maternal grandfather are lepers living in Culion. One sister 15 years old, one brother 13 years old, and one sister 8 years old are nonlepers living outside. He has been living since birth within the leper population. In June, 1921, this child was negative physically and bacteriologically. In February, 1922, he showed several white areas on the left cheek, right loin, right buttock, external surface of right thigh, posterior surface of left thigh, posterior surface of right arm, ulnar side of right forearm, and back of left arm, and two slightly elevated areas on the surface of the left thigh. There was one white area on the anterior surface of the left forearm which showed an elevated, infiltrated, distinctly red center (Plate 5, fig. 1). The skin of both legs -was flushed. There was no anesthesia. There were some scars due to scabies. The nose was negative on physical examination. According to the mother the first spot that appeared was the one on the left cheek, about eight months ago; three months later the area on the right buttock and the rest of the patches appeared in rapid succession, practically at the same time. The reddening and infiltration of the center of the white patch on the left forearm was noticed only about one month ago. The bacteriological examination in February, 1922, showed bacilli in the white patches, but much more numerous in the red center of the area of the left forearm. ‘The nasal septum was bacteriologically negative. Il—G.D. (Table 3, No. 6), Filipino, 5 years old, female, born in Culion in 1916. Both parents lepers, mother dead, father alive and residing in the Culion Leper Colony. One sister, 8., nonleper, was sent away from Culion; another, L., 4 years old, now in Culion, has suspicious lesions which are bacteriologically negative. G. was isolated in the Negative Children’s House at the age of 1 year and 6 months, and was discharged two and a half years later, on account of the development of white nonanzsthetic areas, suspicious of leprosy. In June, 1921, she showed whitish areas all over the body and flushed, shiny, scaly legs; the white areas and the nose were bacterio- logically negative. In September and November, 1921, the patches remained about the same (Plate 4, fig. 1), the scrapings from the spots and from the nose were still negative for B. leprx; injection of 0.001 gram pilocarpine nitrate and administration of hot tea failed to elicit sweating in the white areas, In February, 1922, the white areas all over the body showed reddish elevated borders (Plate 5, fig. 3), the face was reddish, and there were sma]] nodules on both ears. The legs were still flushed and scaly. The 2AG The Philippine Journal of Science * 1922 nodules of the ears and the reddish white patches were found bacteriolog- ically positive, and the septum of the nose was negative. The physical examination of the nose in February, 1922, showed catarrh, with the left side of the septum thickened, and the right side smooth and normal. A previous examination in November, 1921, gave negative findings on both sides of the septum of the nose (Table 2, No. 12). IlI.—A.C. (Table 3, No. 13), Filipino, 8 years old, female, born in Culion in 1913. Both parents living and lepers, residing in the Culion Leper Colony. There are two sisters: P., 7 years old, nonleper, and C., 6 years old, and one brother S., 4 years old, who are considered clinically as lepers although they are bacteriologically negative. A. was isolated in the Negative Children’s House when 3 years old, and from there she was discharged after two and one-half years on account of the appearance of white patches which were suspicious of leprosy. There was no record made of the bacteriological findings at the time of her discharge. At present she shows white areas, with red circular patches in the center, on both cheeks and chin (Plate 5, fig. 2); one extensive, faint, whitish area on the left scapular region; and several white, reddish areas on the left ulnar region. There are no nodules or infiltration of the ears. On the left ulnar region and on the posteroexternal surface of both legs there is both thermal and tactile anesthesia. The legs are reddish and scaly, and there is ecthymatous itch over elbows, buttocks, thighs, and feet. The nose, examined in November, 1921 (Table 2, No. 23), showed inner aspect of right ala nasi with ulcer; superficial ulcer, anterior portion of septum, left side; corresponding portion of right side slightly swollen and with scab. The bacteriological examination in June, 1921, showed the nose negative and the skin positive; in November, 1921, both the nose and the skin were positive for B. leprz. . IV—A.M. (Table 3, No. 12), Filipino, 8 years old, female. Born in Culion. Both parents were lepers, now dead. No brothers or sisters. There is no reliable history obtainable on account of the death of the parents, but she says that the white areas appeared at the same time. Her condition on February 20, 1922, showed whitish areas of different shapes and sizes on chest, abdomen, back, and loin. The buttocks and legs showed the same color as the whitish areas, indicating that they consist of white confluent areas. The face was red and infiltrated. Both ears showed small nodules. The legs were red, dry, and scaly. There was ul- ceration on the right leg. The left forearm was also scaly. The white areas were not anesthetic, and in November, 1921, they showed sweating on administration of pilocarpine and hot tea. There is slight anesthesia on the anterior surface of the left forearm, both to temperature and to touch. The infiltration of the face and the nodules on the ears were not present on former examinations in June and September, 1921, when her picture was taken (Plate 4, fig. 8). There is also ecthymatous scabies on both legs and abdomen. The examination of the nose showed the right side of the septum with nodulelike swelling, the apex of which seems eroded and ulcerated, the left side of the septum being normal. Bacteriological examinations repeatedly made, June to November, 1921, showed the skin positive and the nose negative. On February 20, 1922, 21,3 Gomez et al.: Early Lesions of Leprosy IAT scrapings from lesions of the skin and the septum of the nose were pore for B. lepre. V.—E.R. (Table 3, No. 23), Filipino, 12 years old, female. aon in Culion. Both parents, now dead, were lepers. No brothers or sisters. First lesions that appeared were two whitish areas on the back; later such areas became numerous and extensive; then nodules appeared, first on both elbows and knees, then on the hands and feet and, later, on the ale nasi, chin, and ears (Plate 6, fig. 1). Her condition on February 18, 1922, was briefly as follows: Nodules on ears, ale nasi, chin, both elbows, and forearms, and nodular masses on hands and knees. Infiltrations on cheeks, both forearms, nates, thighs, legs, and feet. Extensive white areas on trunk and arms. Legs dry and scaly. Inguinal glands on both sides enlarged. Axillary glands not pal- pable. There was no falling out of eyebrows. The white areas in parts were denuded of hair, but in most cases the hair was normal. Thermal and tactile anesthesia on the ulnar region, both sides, and on both calves; thermal anezthesia only on both elbows and tendo achillis; no thermal or tactile anesthesia on the white areas. The nose shows scabs and ulceration on both sides of the septum and on the left ala nasi; there was bleeding on detaching the scabs. The bacteriological findings were positive for the white areas, nodules, and the septum of the nose. VI—F.S. (Table 3, No. 24), Filipino, 18 years old, male. Born in Culion. Both parents, now dead, were lepers. No brothers or sisters. The disease commenced several years ago with anesthesia on the elbows and on the knees. Later there was infiltration on both ears, and then the appearance of nodules. Two months ago numerous white areas appeared on the trunk (Plate 6, fig. 2). There has been no prodromal chill and fever; nor were there any paresthetic symptoms. — At present (February, 1922) there are typical nodules on the ears and on the right side of the ale of the nose. There is infiltration on both cheeks. The femoral glands are greatly enlarged, and the inguinal and axillary glands are palpable. The small finger of the right hand is thickened but is not anesthetic to touch. Both ulnar nerves are slightly thickened, but there is no tactile anesthesia on the forearms. There is thermal and tactile anesthesia on the thighs and in the neighborhood of the knees. The trunk and chest show white nonanzsthetic areas. The skin on both arms and legs is dry and scurfy. The bacteriological examination was positive of smears made from the white areas, nodules, the juice aspirated from the left inguinal gland, and the septum of the nose. The type of suspicious or definite leprosy lesions occurring among brothers and sisters of different ages is indicative of the period of development of leprosy and suggests that the white patches in the skin of children are either leprous spots or pre- cursors of definite leper manifestations. The following notes on children of the same parents serve to verify this statement: 1—“I.” E., 11 years old, female, was previously suspicious, showing a thickening in ‘the region of the Darwinian tubercle of the ear and a whitish area on the left side of the face which was slightly reddish; 2AS The Philippine Journal of Science 1922 later both the ear infiltration and the face area disappeared after injections with ethyl ester of chaulmoogra oil and sodium cacodylate. P., 7 years old, female, is a positive leper (Table 8, No. 10). G., 3 years old, female, is negative. 2—“C.” J., 9 years old, female, is a positive leper (Table 3, No. 16). D., 8 years old, female, is a suspicious leper, showing white area on the right loin and shiny legs. M., 7 years old, female, is negative. 3.—‘N.” R., 10 years old, female, is a positive leper (Table 3, No. 20). B., 7 years old, male, is negative. 4—“C.” A., 8 years old, female, is a positive leper (Table 3, No. 13). P., 7 years old, female, is a nonleper. C., 6 years old, female, shows an extensive whitish area on and above right knee which is anesthetic to temperature but not to touch. S., 4 years old, male, shows whitish areas on back, above both knees, and on the right knee. 5.—“D.” G., 5 years old, female, is a positive leper (Table 3, No. 6). L., 4 years old, female, shows small white spots on the right cheek and right arm, and flushed, shiny legs. S., 3 years old, female, sent away from Culion when about 5 months old. 6.—“R.” There are five living children of a leper mother and a healthy father. J., 30 years old, female, and M., 20 years old, female, are lepers living now in Culion. Three younger brothers, aged 10, 9, and 7 years, respectively, are nonlepers and live in Tarlac Province. The mother, living in Culion, has borne another child by a leper father and this one, M. G., 4 years old, male, is now a positive leper (Table 3, No. 4). Leper lesions occasionally disappear spontaneously, as has been observed in one child, P.B., 5 years old, male, who was isolated in the Negative Children’s House and later was discharged for suspicious leprosy. In June, 1921, he presented a reddish infil- trated area on the left cheek which was_bacteriologically positive, but on February 20, 1922, and without treatment, the infiltration was negative. In our records there are four other children that showed, on first examination, bacteriologically negative white patches, but eight months later some or all of the patches disappeared. THE EFFECT OF SEGREGATION ON THE DEVELOPMENT OF THE DISEASE Since 1913 fifty-one children were transferred from Culion for adoption outside of the island. The age of the majority on transfer was between 4 months and 3 years; only one was 4 years, and another was 10 years of age. The children at the time of transfer were negative, physically and bacteriologically. So far three have been returned to the colony: two on account of leprosy, and one on account of the request of the father who is in the colony, as he thought the child was not being properly taken care of. The two children returned for leprosy were the following: G.M., male, born in Culion February 1, 1912; transferred away on March 23, 1914, at the age of 2 years and 2 months; on 21,8 Gomez et al.: Early Lesions of Leprosy 249 September 7, 1921, was admitted to San Lazaro Hospital, Manila, as a leper, maculo-anesthetic type, and returned to Culion on November 17, 1921, at the age of 9 years. A.C., born in Culion on May 26, 1912; transferred on April 23, 1914, at the age of 1 year and 11 months; returned to the colony about May 17, 1917, at the age of 5 years, and at the present writing the disease is of the nodular and anesthetic type. Beginning with the year 1916, an effort has been made to segregate physically and bacteriologically negative children in a house located on the site reserved for the healthy residents of the colony. Since the implantation of this policy in March, 1916, to October 6, 1921, there were isolated 83 children, of which 22 remain negative, 25 have died, 5 were transferred outside of the colony, 5 were returned to the leper population for other diseases, and 26 were returned for leprosy or suspicious signs of leprosy. No records were made of the physical and bacterio- logical findings at the time of discharge of the children. TaBLE 4,—Children isolated at the Negative Children’s House as nonlepers and later returned to the colony on account of definite or suspicious signs of leprosy. [(+) Positive clinically and bacteriologically.] Age when definite or No. | _ Initials. Dete of birth, | fAsen dame Age wham, | wosphcloer |. eitioa. ° leprosy deve Y. om fe Fs me a |. Pee ee: et, 1| J. RL (+) .2.| March 30, 1911.___--_- 0s 1 5 1 26 65 2) 0. 8 2|A.T. (+)..-! February 19, 1910....| 11 10 12 ee a 6 it 38) -0~ 6:15 3} P.1. (+)_..) June 25, 1914 -.-.--:.- 7°66 P96 Oy OF ie Be: T 4|B.R. (+).--| June 14, 1914 -------- 1 AS 2 + 3 2 10 ()} 0 9 9 5 | B.S. (+)---] September 8, 1911 ....| 10 3 28 4 10 19 6 (2) 2m 6 | R.N. (+)_-.] October 17, 1911 -_.--. 10 2 14 4 6 12 & 6-18) -.4 3-2 y a eos eee January 20, 1919__.-.- 211 11 b-<4 0 Sf» 18-1. 8 18 EE en eee June 18, 1914 _....---- 7° 6 18 1 38 26 Sli 6 AS tivo... September 8, 1918....| 8 3 23 210 9 ‘6 or: 8 10} i deG@s.cz.: February 20,1913 __..| 8 10 11 e-3' 0.921) £2 11 | A. V.S. (+)_| November 8, 1913__--- oar ees 4-8 2 6.710! 8 3-18 12 | F.R. (+)-_._} August 21, 1916___._-- . 2 : 0.9 S$ 42! 2 42 13 | G.D. (+).._| April 8, 1916.___-_---- 5 8 28 oe £9 Tl f ££ BB 4) ASNee2 August 15, 1913__-----. 8 4 16 2 7S & 6 8} 2275. & “45 | A. Gs (+)-._| April 21, 19182,2----2- 8.8 10 8 2 26 BS $874..8 824 16 1h os November 11, 1913 ---| 8 1 20 2414 4.41 1) 3 7 yet eeamamiees February 2, 1915._---- 6 10 29 5 are Weg | | S$ it 161 2°39 2 18 | P.B. (+)-...| July 18, 1916-__.....-- 5 6 18 i 2555 8 (9 St 2 2 19) Me April 26, 1915_.------- 58 0 10 29 $8 2.29.2 By Ane eee January 16, 1909.----- 12 11 15 2 &) 0 'R Ae ys 21 | J.C. (+)__-_| February 24, 1912_.--- ei TT) 4 4:3 €? 2; 8 2 oF) MOS April 14, 1913_....-..- peu tue Cog -8t 4 2° 8 188431——2 950 The Philippine Journal of Science 1922 The 26 children discharged for leprosy or suspicious signs of leprosy were repeatedly examined during the period from June, 1921, to February, 1922, and 4 were found nonlepers, 11 showed suspicious signs of leprosy but were bacteriologically negative, and 11 were clinically and bacteriologically positive. A tabula- tion is given (Table 4) to show the age at which the suspicious or definite leprosy developed and the period of segregation before such development, that would indicate the minimum period of time at which a latent leprosy may manifest itself. Glancing over Table 4, one notices that children isolated at as early an age as one year (No. 12) developed later signs of leprosy confirmed by bacteriological examination; and that definite leprosy may appear within a period of between four months (No. 1) and three years and three months (No. 21) after isolation. The two cases transferred outside of Culion and later returned as positive probably developed the disease much earlier than the date of their return, as they no doubt escaped detection for a long time. Several months or years may elapse, therefore, after contact or association with lepers has ceased, before suspicious or definite signs of leprosy develop. INCIDENCE OF LEPROSY AMONG THE CHILDREN OF LEPERS Denney (2) has already published data regarding the incidence of leprosy among the children of lepers born in Culion. Our figures give the results of a general survey made in February and March, 1922, of all the children born and living in Culion up to December 31, 1921, and take into account other details not previously considered. The total number of children exam- ined was 308, of which 153 were males and 155 females. The results are shown in Table 5. Both parents of practically all the children examined were lepers. The children have been classified into positive lepers, clinical lepers, suspicious, and negatives or nonlepers. Positive lepers are those who -show definite clinical signs of leprosy and in whom Bacillus lepre has been demonstrated. Clinical lepers are those who are negative bacteriologically, but who show at least one of the following symptoms: (a) Anesthesia in the ulnar or peroneal region which cannot be attributed to other diseases ; or (b) whitish patches which show definite anzsthesia, either thermal only or both thermal and tactile; or reddish tint indica- tive of progressive changes; or which definitely show no sweat- ing on pilocarpine injection, in contrast with the surrounding normal skin and its corresponding symmetrical area on the other 21,3 Gomez et al.: Early Lesions of Leprosy 951 side of the body. The suspicious cases are those that are also negative bacteriologically, and show whitish patches which are not anesthetic and which show definite or doubtful sweating on pilocarpine injection, or they may show only flushed shiny legs. Negatives, or nonlepers, are those who are bacteriologically negative and on whom no suspicious blemishes or lesions are found in the skin. ~ TABLE 5.—Incidence of leprosy in the children of leper parents. Children examined. Positive lepers. Clinical lepers. Age in years. ; Males. | Fe- | Total.|Males.|,5@. | Total. | Males.| fic, | Total. FP; ot. P. ot Under 1-.-. 15 17 Ber ies hc ype hdl onc ecaslsncpandalosocus|acdetece = OR Ea 17 BESSA Spi se, PaRiameee UESAPASS We: 2 Ree gn ay Portes: 13 15 Di cies Sos dag ee dat akiveldeoe dusk 1 1| 3.57 , EST 25 15 40 pe Penge em 1 1 Rg Se eerie i aS sl ead ol iinmies Ee sciscenuee 19 20 89 1 2 3 7.69 5 2 7\ 17.9 Cie 9 8 17 1 1 yo RSs oc Fy | RSE eee eae Rees RO ee Ee na LS ne rae 15 21 36 1 1 2 5.383 1 3 4/11 , eepeeerre 12 12 24 2 I 3] 12.50 1 1 2/ 8.33 BREE ees 6 16 22 2 2 9.09 1 3 4/18 Sie 6 3 9 3 1 4| 44.44 ieee Bi SSS he 5 3 8 2 1 Ie I eR: ne a ea Wa scka oe celen dee 4 4 1 5 BG ae Bee 42 = nes 1 4 ae | 2 2} 40 1 1 2 | 40 35 cer te Bat sees 2 1 1} 50 eis Total__ 153 155 308 12 12 24 19 9 ll 20} 6.4 Suspicious. Negatives. Age in years. Males. | Females. Total. Males. | Females. Total. P. et. P60: Under 1 lz A pe, SOT 15 17 82 | 100 p badieg te te hence nae 3 3 6] 11.904 22 4 86 | 85.71 Bet Sa 2 2 4 14.27 11 12 23 82.14 Svichwis camrecmd 9 4 13.| 32.5 15 11 26| 65 Bilas oaks ek hae 5 6 11 28.2 9 9 18 46.15 5 me, Seb eats eet eres s 4 6 35.29 6 3 9 52.94 Ce ee 7 3 10 21.7 6 14 20 55. 55 7... cusaguammcneice ts py Mey eee Ti s...4:17 8 10 18 75 Beer cee 1 4 5 22.72 4 ys 11 50 Oi 555 hte yo) ER erie Geib, eS EE Se 3 2 5 55. 55 AOE Pa ei 1 1 2 25 2 a 3 37.5 ph RRR ee RRS meee prapege 3 3 75 ook oe | ea cds Cow [onde eee hee] chunac bse leowset ones x 1 20 18. Sp eS eee RB bee wk ye Bers Saee 1 50 Total c55. : 31 27 68 | 18.8 102 104 206| 66.8 252 The Philippine Journal of Science 1922 Analyzing the data in Table 5, we notice that the positive lepers number 24, or 7.79 per cent of the total number of children (308) examined; of these 12 are males and 12 are females. The clin- ical lepers numbers 20, or 6.4 per cent; the suspicious, 58, or 18.8 per cent; and the nonlepers, 206, or 66.8 per cent. The distribution between sexes is practically the same, except in the clinical lepers among whom there is a greater number of females. The incidence of the disease is greater as the children grow older. Taking the cases of undoubted leprosy, the positive lepers, and separating the children into three age groups in order to get average data for comparison, we find that there are 19 children from 10 to 13 years of age, inclusive, of which 7, or 36.8 per cent, are positive lepers; 108 children from 5 to 9 years, inclusive, of which 13, or 12 per cent, are positive lepers ; and 181 children under 5 years of age, of which 4, or 2 per cent, are positive lepers. The youngest child found positive was 3 years old; the youngest child only clinically a leper was 2 years old; and the youngest children with suspicious blem- ishes were 1 year old. SUMMARY AND CONCLUSIONS The children born of leper parents show about the same sus- ceptibility to other morbid conditions as children of nonleper parentage. The mortality on account of congenital debility is higher than in children born in nonleper populations, but this greater mortality due to congenital debility is counterbalanced by the lesser mortality due to infantile beriberi, a disease that has practically disappeared in Culion on account of the exclusive use of unpolished rice as the staple article of diet. Our data indicate that the most-frequent recognizable site of the early lesion of leprosy is the skin, and that the infection through this route is greatly favored, presumably, on account of the great prevalence of skin diseases among the children, which offer anatomical conditions favorable to the invasion of the lepra bacillus. ‘| The most-frequent recognizable early lesion of leprosy is macular lesion of the skin. In Filipino children this manifests itself in whitish, fawn-colored patches, which at the beginning may not show the bacillus of leprosy or disturbances that may be attributed to the innervation of the affected region, such as diminution in the sense of touch, thermal discrimination, sensi- bility to pain, or disturbance in the secretory power of the sweat glands. Later, when nervous disturbances become appa- 21,2 Gomez et al.: Early Lesions of Leprosy 253 rent, the sense of thermal discrimination—that is, ability to distinguish between hot and cold—is the first one affected. The white patches, macula alba, are either precursors of other cutaneous or nervous leprous manifestations, or they them- Selves develop into definite progressive and bacteriologically positive leprous lesions with concomitant thermal and tactile anesthesia and anhidrosis. In a few instances we noticed the disappearance, without treat- ment, of leprous lesions that were either negative or positive bacteriologically. _ Children that have lived in intimate contact with lepers for varying lengths of time may develop leprosy, several months or years after they were isolated from lepers. The shortest period in our series was four months, and so far the longest period was three years and three months after isolation. The survey made of 308 children of leper parents born and living at the time in Culion Island show 24, or 7.79 per cent, positive lepers and 78, or 25 per cent, with suspicious or definite signs of leprosy though not bacteriologically positive. The leprosy in children affects both sexes about equally. The inci- dence of leprosy is in direct ratio to the age of the child: the older the child the greater the incidence. The youngest child found to be a positive leper was 3 years old, but suspicious blemishes have been found in children as young as 1 year of age. REFERENCES 1. BRINCKERHOFF, W. R., and Moors, W. L. Studies upon Leprosy, IV. Upon the utility of examination of nose and nasal secretions for the , detection of incipient cases of leprosy. U. S. Government Printing Office, Washington (1909); also U. S. Public Health Bull. 27 (1909). 2. DENNEY, O. E. A statistical study of leprosy in the Philippine Islands. Journ. Am. Med. Assoc. 69 (1917) 2171. 3. Dyer, ISADORE. The dermatologic aspects of leprosy. Journ. Am. Med. Assoc. 61 (1913) 950. 4. GwyTHER, A. Is there a primary lesion in leprosy? Ind. Med. Gaz. 50 (1915) 41. 5. Hopkins, RALPH. Early manifestations of leprosy. New Orleans Med. and Surg. Journ. 70 (1917-1918) 56. 6. Population and Mortality of the Philippine Islands, Philippine Census, Bull. No. 4, Manila (1918) 86. 7. SticKER, GEoRG. Mittheilungen iiber Lepra nach Erfahrungen in Indien und Egypten. Miinch. med. Woch. 44 (1897) 1063, 1103. ILLUSTRATIONS [Photographs by E. Cortes, Bureau of Science.] PLATE 1. SINGLE SUSPICIOUS, WHITISH; BACTERIOLOGICALLY NEGATIVE AREAS Fic. 1. J.B., 2 years old. Whitish area on the right buttock. Photo- graphed on March 3, 1922. 2. White area noticed one year ago on the left cheek of B.B., 4 years old. Photographed on March 3, 1922. 3. White area on buttock, four years’ duration, on A.T., 7 years old, female. Photographed on September 13, 1921. PLATE 2. FURTHER PROGRESS OF WHITE AREAS Fic. 1. A.L., 3 years old. White area on the external surface of the left thigh, of about eight months’ duration, and white area on the left cheek, which are at present bacteriologically negative. Vaccination mark on the left arm. Photographed on March 3, 1922. 2. A. de L., 4 years old, female. White area on the right loin which is bacteriologically positive and is the only leprous lesion in this child. Photographed on March 3, 1922. 3. T.R., 6 years old, male. White areas on the back of the left arm of two years’ duration which are bacteriologically positive. They are the only leprous lesions found in this case. Adjoining the areas, there are lesions of chronic “itch” which might have pre- disposed to the development of the leprous lesions. Photographed on March 3, 1922. PLATE 3. DEVELOPMENT OF FLUSHED, SHINY, AND SCALY LEGS Fic. 1. L.D., female, 4 years old, bacteriologically negative at the time. Photographed on September 13, 1921. 2. G.D., female, 5 years old, sister of L.D. in fig. 1, also bacteriolog- ically negative at the time. Photographed at the same time as fig. 1, September 13, 1921. 8. A.M., female, 8 years old, who is bacteriologically positive. The scaliness of the legs is more marked. A trophic ulcer is also shown on the right leg. Photographed on September 13, 1921. PLATE 4. MULTIPLE WHITE LEPROUS AREAS Fic. 1. D. sisters. They are the children whose legs are shown larger in Plate 3, figs. 1 and 2. The older girl, G., five years old, shows white areas all over the body, and the younger one, L., 4 years old, faint whitish areas on the right side of chin and right arm. Both bacteriologically negative at the time. Photographed on September 13, 1921. 255 256 Fic, 1. Fie6. 1. The Philippine Journal of Science . F.B., 7 years old. White areas all over the body which are bac- teriologically positive and on injection of pilocarpine show no sweating. These patches appeared about one year ago and were preceded by a white area on the right cheek which was noticed when he was hardly over 1 year old. Photographed on March 3, 1922. . A.M., 8 years old. White areas of different sizes and shapes all over the body which appeared in crops and are bacteriologically positive. Ulceration on right leg and scaliness of the skin of both legs is shown more plainly in Plate 3, fig. 3. Photographed on September 13, 1921. PLATE 5. PROGRESSIVE CHANGES IN THE LEPROUS PATCHES C.B., 3 years old, male. White areas on face and chest and one white area on anterior surface of the left forearm which recently showed reddening and infiltration and on which the number of bacteria found are more numerous than on the white spots. Photographed on March 3, 1922. . A.C., 8 years old. White areas on cheek, with red Bute which dovelbned later, indicating progressive changes. Bacteriologic- ally positive in these lesions and in the nasal septum. Photo- graphed on September 13, 1921. . This is the older girl represented in Plate 4, fig. 1, who was examined and photographed about six months later. Now she shows reddening of the edges of the white areas, flushing of the face, and nodular infiltration of the ears. The red margins of the macules and the nodules of the ear are bacteriologically positive, although the septum of the nose still remains negative. Photographed on March 3, 1922. PLATE 6. ADVANCED STAGES OF LEPROSY E.R., 12 years old. Nodular lesions in the ears, chin, and ale nasi, and white areas on the chest which preceded the infiltra- tive and nodular lesions. Bacteriologically positive in the skin lesions and nose. Photographed on March 3, 1922, 2. F.S., 138 years old. Nodules on both ears and infiltration on both cheeks. White areas on the chest appeared recently, after the development of infiltration and nodules. The skin of the chest and arms also shows scars and active lesions of an itchy disease of the skin other than leprosy. Bacteriologically positive in the skin and nose. Photographed on March 3, 1922. GOMEZ ET AL.: EARLY LESIONS oF LEPROSY. ] [Puiuie. Journ. Sctr., 21, No. 3. PLATE 1. SINGLE, SUSPICIOUS, WHITISH BACTERIOLOGICALLY NEGATIVE AREAS. GOMEZ ET AL,.: EARLY LESIONS OF LEPROSY. ] [PuILip. JouRN. Scr, 21, No. PLATE 2. FURTHER PROGRESS OF WHITE AREAS. GOMEZ ET AL,: EARLY LESIONS oF LEPROSY.] [PHIuip. JOURN. Sci., 21, No. 3. PLATE 3. DEVELOPMENT OF FLUSHED, SHINY, AND SCALY LEGS. GOMEZ ET AL.: EARty Lestons or Leprosy.] [PHILIP. JouRN. Sctr., 21, No. PLATE 4. MULTIPLE WHITE LEPROUS AREAS. -: EARLY Lesions or Leprosy. ] : [Puiuip. JourN. Sct., 21, No. 3 PLATE 5. PROGRESSIVE CHANGES IN THE LEPROUS PATCHES. GOMEZ ET AL.: EARLY LESIONS oF LEPROSY.] [Puiuie. Journ. Scr., 21, No. 3 PLATE 6. ADVANCED STAGES OF LEPROSY. ADDITIONS TO THE HERPETOLOGICAL FAUNA OF THE PHILIPPINE ISLANDS, II By Epwarp H. TAYLOR Of Manila, Philippine Islands FOUR PLATES This paper contains descriptions of the following species be- lieved to be new: AMPHIBIANS Rana igorota, Cornufer rivularis. Rana yakani. Cornufer montanus. Rana tafti. Cornufer subterrestris. Micrixalus diminutiva. — Polypedates linki. LIZARDS Sphenomorphus beyeri. Tropidophorus stejnegeri. SNAKE i Natrix barbouri. The following species are listed with notes on color and scala- tion and other pertinent data: AMPHIBIANS : Oxyglossis levis Giinther. 3 Polypedates pardalis (Giinther). Rana suluensis Taylor. — Polypedates appendiculatus (Giin- Rana luzonensis Boulenger. ther). Rana sanguinea Boettger. - Kalophrynus stellatus Stejneger. Rana erythrea (Schlegel). ._ Bufo philippinicus Boulenger. « Staurois natator (Ginther). Megalophrys stejnegeri Taylor. Cornufer corrugatus (Duméril). LIZARDS Lepidodactylus divergens Taylor. Brachymeles gracilis (Fischer). Lepidodactylus aureolineatus Tay- Brachymeles schadenbergi lor. (Fischer). Sphenomorphus luzonensis (Bou- Brachymeles vermis Taylor. lenger). Dibamus argenteus Taylor. SNAKES Natrix lineata (Peters). Zaocys luzonensis Giinther. Natrix dendrophiops (Giinther). Holarchus meyerinkii (Stein- Oxyrhabdium modestum (Dumé- dachner). ril and Bibron). Psammodynastes pulverulentus Oxyrhabdium leporinum (Giin- (Boie). ther). 957 258 The Philippine Journal of Science 1922 Boiga dendrophila divergens Tay- Doliophis philippinus (Giinther). lor. Naja naja philippinensis Taylor. Hemibungarus calligaster (Wieg- Naja naja samarensis (Peters). mann). Trimeresurus wagleri wagleri Hemibungarus meclungi Taylor. (Boie). Rana guerrerot Taylor has been relegated to the synonymy of Rana luzonensis Boulenger. The species was established on immature specimens but the study of a large series of larve, young, half-grown, and adult specimens obtained in 1920 makes this action imperative. Under Natrix barbouri sp. nov., I have endeavored to show that Natrix crebripunctata (Wiegmann) should be eliminated from lists of Philippine snakes, since it is beyond question a name based on a young specimen of Natrix spilogaster (Boie). AMPHIBIANS Oxyglossis levis Giinther. Oxyglossis levis GUNTHER, Cat. Batr. Sal. Brit. Mus. (1858) 6, pl. 1, fig. A; TAytor, Amphibians and Turtles of the Philippine Islands (1921) 30. I obtained speqimens in Kalinga and Polillo; on Mount Mari- veles, Mount Banahao, and Mount Maquiling; at Zamboanga; and on Basilan. In each of these localities the species occurred in abundance. I failed to find specimens in Jolo. I have re- ported specimens from “Sulu Archipelago”: but fail to find in my records any mention of collections in Sulu. Usually where the species occurs it is very common and is the first species one picks up. It probably does not occur in Jolo. Philippine specimens of this species appear to be larger than those from southeastern Asia. Many of my specimens are more than 50 millimeters long from snout to vent. Occasional specimens are more than 60 millimeters long. Published records from extra-Philippine localities give much smaller measure- ments. The tips of the digits appear to be more widened in ,Philippine specimens, Rana suluensis Taylor. Plate 1, fig. 2. Rana suluensis TAYLOR, Philip. Journ. Sci. 16 (1920) 264; Amphi- bians and Turtles of the Philippine Islands (1921) 65. Two specimens of Rana suluensis were taken along a small stream in central Jolo, near the Government cattle ranch. The specimens agree very well with the type save in the presence of very distinct dorsolateral folds. Certain young cotypes, how- vr * Loe. cit. 21,3 Taylor: Herpetological Fauna, II 259 ever, had narrow folds. As the type has not been figured I include a photograph of one of the Jolo specimens, somewhat enlarged. The type locality is Tawitawi, an island to the south- west of Jolo. Rana luzonensis Boulenger. = Rana luzonensis BOULENGER, Ann. & Mag. Nat. Hist. VI 17 (1896) 401; TAyLor, Philip. Journ. Sci. 16 (1920) 254; Amphibians and Turtles of the Philippine Islands (1921) 55. Rana guerreroi TAYLor, Philip. Journ. Sci. 16 (1920) 255; Amphi- bians and Turtles of the Philippine Islands (1921) 56, A number of specimens which I have referred to Rana luzo- nensis Boulenger were taken at Baguio and at various places along the trail to Bontoc. Larveze were obtained near the Bon- toc-Lepanto boundary from an old Bontoc fisherman, who had a small tin nearly filled with the larve. He had taken them for food, from a mountain stream near the trail. Larve and newly transformed young were taken at the town of Bontoc in the river and among the stones forming the walls of the rice paddies. A study of this series of larve and young and adult specimens has convinced me that my Rana guerrero? is the young of Rana luzonensis. Measurements of Rana luzonensis Boulenger. No. 662. No. 713, mm, mm. IED OD WOR oo. conn nke a Banke oceans eceucuenassmeeknieaes 58 53 Leehreh of RONG 2... See tuseeees. 21 20 Widtlr of headcs' 3.06664 Soc. ee ictievean sence. 5 eee 17 14 Tiheatabel ib eric oes ce by 63- ac cosa ass coe seach ergeancane 6.7 6.5 yee O8G mf sont i oe i ec caas coca nupentenneaen ancien 9.5 9 TAPE 0 Hom oo cc. reece nde ce peep eneeen ela 6 5.3 AIDDGE CVCING oiide oo oh nhs Cochin oh weak ok dp tien sees ewasendicanies 5 5 Interorbital space -------.----. ins eas tie 5 5 Ce ee Se cane eae da kel eee aae man anen ls nuewe mie 39 34 Longest Tinger, With hand <2... 2-25 eect en ene cacceseeneesecenn: 19 15 RECN CD ks cut cuhiuee cud pew hk coun kcages Uostee need nena debames te tasks adie 109 97 WS ce oe Pcs accek Cie can os 0 enue de ea So ceeaecascettes 32 30 THD coca ds. The type has one hind limb measuring 11 millimeters, the other 13 millimeters. Remarks.—No. 1172 agrees with the type in the scalation of the head. However, it has twenty-six instead of twenty-four scale rows around the body. The length of the leg is contained in axilla-to-groin distance approximately four times. No. 1173 agrees with the type in scalation, save that the frontal and frontonasal are more broadly in contact. The post- parietal body scales on the left side are fused, forming an 21,3 Taylor: Herpetological Fauna, II 289 elongate nuchal. The length of the leg is contained in the axilla- to-groin distance four and six-tenths times. Scales in twenty- four rows around body. No. 1520, collected at Abung-abung, differs from the type in having the parietal broken into two scales, as have certain speci- mens of Brachymeles vermis Taylor and Sphenomorphus bipa- rietalis Taylor, both from the Sulu region. The first pair of chin shields is separated by a small single shield. Scales in twenty-four rows around body. This species probably does not attain as large a size as the other pentadactyl species of the genus. It is a burrowing form and is usually found under logs or stones. The lizards of this species move with great rapidity. It is difficult to grasp them because of their extremely smooth scales. Brachymeles schadenbergi (Fischer). Eumeces (Riopa) schadenbergi FiscHER, Jahrb. wiss. Anst. Ham- burg 2 (1885) 87, pl. 3, fig. 2. Brachymeles schadenbergi TAYLOR, Philip. Journ. Sci. § D 12 (1917) 268; Lizards of the Philippine Islands (1922) 249. Twenty-one specimens of this species were taken at Zambo- anga, Basilan, and Jolo. They agree very well in color and markings. A single specimen was collected in southern Leyte, by Gregorio Lopez. The head scales are variable. Thus, of the twenty-two speci- mens, twenty-one have the supranasals in contact and one has them separated. In seventeen specimens the parietals form no suture; in five specimens they do. Seven specimens have the fourth labial entering the orbit; fifteen, the fourth and fifth. The character of the chin shields is apparently invariable, save in one Jolo specimen, which has the second pair wider than the first pair. Brachymeles vermis Taylor. Brachymeles vermis TAYLOR, Philip. Journ. Sci. § D 13 (1918) 255, fig. 10; Lizards of the Philippine Islands (1922) 258, fig. 53. Eight specimens of this small wormlike form were collected on the Government cattle ranch near the central part of Jolo Island. They were found under leaves and trash along a small stream. All the specimens agree in the arrangement of the chin shields; that is, having the first pair separated and the second pair wider than the first pair, separated by a single scale. All have the postmental in contact with a single labial. In the conformation of the head scales they agree in general 290 The Philippine Journal of Science 1922 with the type. The nasals are broadly separated in the eight specimens; the prefrontals are separated in seven, in contact in one; the frontoparietals are separated in the eight specimens, the parietals separated in seven, and forming a suture in one. Small nuchals are present in all the specimens. In a single specimen, the parietals are broken as shown in the drawing of a cotype.° Dibamus argenteus Taylor. Plate 4, fig. 2. Dibamus argenteus TaAyLor, Philip. Journ. Sci. § D 10 (1915) 107, pl. 1, figs. 11 and 12; Lizards of the Philippine Islands (1922) 261. Several specimens belonging to this species were collected on Basilan at Port Holland. Specimens were found under rotting logs in a cut-over area. All of the specimens show the irregular silver blotches on the body. I have already reported, in my work on the lizards of the Philippine Islands, the Negros and Papahag specimens that are listed in the table. Measurements and scale counts ~ Dibamus argenteus Taylor. Scales No. Locality. | Collector. | Sex. |Length | Tail. | Scale | Upper bordering|Length rietal. mm, mm, mm, | 1216_.....---| Port Hioll- | E.H.Taylor g 132 | 19.5 22 1 Bisjees and, Basi- | lan. REID. ucosts se CO see.) seaus Ge see g 126 18.5 22 1 © foscrccak $2002. Savi Ge sest | ated Gos ccu.s roy 88 | 13.0 22 1 6| 2 Be As tates G0 pi cuclusee HO wens fof 123 19.0 22 1 6 3.6 | pbs: aatesmpentace) OLN OO cci5.-[onkne pt ea ae rol 122 19.5 22 i 5 4.0 ee 7s eee Negros ---.-.|----- d0icace: ¢ 154 | 28.0 22 1 Fae fo?” eras PapahagIs- |__--- Oe wicnd J 120| (2) 22 1 41 8.6 land, Sulu. The limbs of this species appear to be larger and better de- veloped than in Dibamus novae-guineae Duméril and Bibron. In Boulenger’s illustration the limbs appear to be covered *° with one large terminal scale and a single pair of scales above or, at most, three scales. In my specimens the limb has a large, rather pointed, terminal scale with three or four pairs of smaller scales; the preanals, too, are different, consisting of two elongate scales separated by two or three scales, the tips of these over- lying a sharply pointed scale which lies directly on the edge of the anus. In the females the arrangement of the preanals is * Taylor, loc cit., fig. 10d. ” Fauna Malay Peninsula; Rept. & Batr. (1912) ee 26 A and B. 21,3 Taylor: Herpetological Fauna, II 291 also different. The tail length of this species is contained in the body length an average of six and six-tenths times. SNAKES Natrix barbouri sp. nov. Tropidonotus crebripunctatus BOULENGER (non Wiegmann), Cat. Snakes Brit. Mus. 1 (1893) 262. Natrix crebripunctata TAytor, Snakes of the Philippine Islands (1922) 91. Type.—No. 939, E. H. Taylor collection; collected at Balbalan, Kalinga, Luzon, April 26, 1920, by E. H. Taylor. Description of type——(Adult male.) Rostral much broader than high, slightly visible above; internasals longer than broad, longer than prefrontals; latter broader than long, much broader than internasal; frontal bell-shaped, the anterior edge nearly a straight transverse line; length of frontal equal to or slightly less than supraocular, sightly longer than its distance from end of snout; supraoculars elongate, slender, at least two and one- half times as long as broad; parietals only very slightly longer than frontal, the width equal to three-fourths the length; nasal divided, anterior part lower and longer than posterior part; loreal nearly square, much lower than nasal; two preoculars, the upper broader than the lower (three on right side) ; eye large, its diameter equal to its distance from anterior edge of anterior nasal scale; four postoculars; two anterior temporals, the lower much the larger, touching two postoculars; three equal-sized posterior temporals; nine upper labials, the fourth, fifth, and sixth entering orbit; ten lower labials, five touching anterior chin shields; mental broadly triangular, much wider than deep, not as wide as rostral; anterior chin shields forming a suture their entire length, not three-fourths as long as second pair, which are separated their entire length; scales in 19 rows, all strongly keeled, with apical pits; ventrals, 169; anal, divided; subcaudals, 107; tail long, slender, ending in a sharp point. Color in life—Gray above with transverse indistinct bands of black, broken laterally by yellowish spots; head olive without markings ; neck dark without a yellow spot; upper labials yellow, edged with black above and a small spot on each of the first three labials; lower labials, chin, and throat white; a row of black dots on outer edge of ventrals strongly defined to tip of tail; outer edges of ventrals marked with gray on anterior part of belly; on latter half the entire belly also gray; under tail gray with edges of subcaudals yellowish. 292 The Philippine Journal of Science 1922 Measurements of the type and the cotype of Natrix barbouri sp. nov. Type, 3. | CNSR mm. mm, "Total levi th sso ki Soe ec cas cae eee aucaeenaeere etek cymes ee 803 820 Sot to @CONG re 8S oe a ee a oe ee ee eon eee 553 560 Well one cece copes peaks sb ee caceceneaneeaees soso ee os seeeiek eee : 250 260 Width trl Teh cc a ss eg ede ew cen ne 11.5 13 Tpertir of WOR soo ee ae a emcee net weer reo tae cade sas tone 20 20 Diameter of eye ..... 022.2... --- o 5 os eon es += -- 5 -- === 5 - - = 6 6 | Eye to end of snout _.._.--..---------------------------------------------- 6.4 6.3 = Remarks.—In a previous publication “ I stated in a footnote: “T strongly suspect that N. crebripunctata Wiegmann is indeed N. spilogaster. I believe further that Boulenger’s species of this name is a distinct species.” After obtaining a copy of Wiegmann’s type description, and comparing it with young and half-grown specimens of Natriz spilogaster (Boie) from Manila, and after obtaining specimens of a Natrix which is undoubtedly Boulenger’s Tropidonotus crebripunctatus and comparing these with Wiegmann’s type description, I have no hesitancy in placing Tropidonotus crebripunctatus of Wiegmann as a synonym of Na- trix spilogaster (Boie) and making a new species of the Natrix I have found, and which appears to be the same as that which Boulenger calls T. crebripunctatus. Variation.—There are only two specimens in my collection and a very young one in the collection of the College of Agri- culture at Los Bajios that I believe belong to this species. The cotype is likewise from Balbalan, Kalinga. It differs somewhat in scale counts from the type, as follows: Ventrals, 163; subcaudals (tip of tail missing), 95; preoculars, 3; post- oculars, 2 on right, 3 on left side. In other respects the speci- men agrees with the type in scalation. The scale counts for the College of Agriculture specimen are: Ventrals, 160; subcaudals, 100; preoculars, 2; postoculars, 3; temporals, 2+2 and 243. It likewise agrees with the type save that the head is proportionately larger, as is natural in. very young specimens. In color the cotype differs from the type in having a very narrow yellow line on the median dorsal surface, broken by narrow black spots less than half the length of the intervening yellow. This continues some distance on the anterior part of the body. The entire latter half of the belly is gray, as well as “Snakes of the Philippine Islands (1922) 92. 21,8 Taylor: Herpetological Fauna, II 293 the subcaudal area. This vertebral stripe is mentioned in Bou- lenger’s description ** as being present. In comparing these specimens with Wiegmann’s description I find that none of the three specimens has the white nuchal spot (“der Nacken dunkler mit einem weissen Fleck’), which is invariably present in Natrix spilogaster; the two small spots on the parietal scales are wanting (‘“‘die beiden Punkte der Occi- pitalschilder, die auch bei andern Kielnatter-Arten in Jugend- alter vorkommen’’), and there is only a single irregular row of black dots on the outer edge of the ventrals instead of in- numerable black dots arranged in transverse rows as is true in Natriz spilogaster. (“Die Bauchseite ebenfalls weisslich, aber mit unzdhligen schwarzen Punkten bedeckt, die auf den ein- zelnen Bauchschildern in Querreihen zu 6-10 stehen”.) The ventral and subcaudal counts of Wiegmann’s specimen (148 and 96) are well within the range for N. spilogaster, but much lower than in the species described, in which they are 163 to 169 for ventrals and 100 to 107 for subcaudals. The species is named for Dr. Thomas Barbour, the eminent herpetologist of the Museum of Comparative Zodlogy at Harvard College, Cambridge, Mass. Natrix lineata (Peters). Tropidonotus lineatus PrTERS, Mon. Berl. Ak. (1861) 686. Natriz lineata TayLor, Snakes of the Philippine Islands (1922) 92. I collected three specimens of this snake along Tumugao River, Zamboanga, Mindanao, and seven specimens on Basilan, at Port Holland and Abung-abung. In general they differ from speci- mens collected in Agusan, Mindanao, in color and markings. Most of the Agusan specimens are dull black above with no markings distinguishable save in very young specimens. In all ten specimens the color is light olive above, with a regular net- work of black markings and an indistinct series of yellowish - dots on the sides. The ventrals and subcaudals have a large dusky area near the middle. The head is deep brown. The strongly defined yellow labial line is present in all the speci- mens. All of them show a dark nuchal band with a light, ir- regular, nuchal spot. No. 14038, the largest specimen collected, was found at night near an old unused well where Polypedates pardalis were breed- ing. In life the colors were as follows: Top of head and anterior part of body reddish brown, gradually becoming olive on the # Loc. cit. 294 The Philippine Journal of Science 1922 posterior part of body and tail; line on lip brownish white; chin and throat white; neck spot white; belly with dirty olive mark- ings; outer edges of ventrals pinkish. Most of the specimens were taken under rocks along small rivers. The known ventral range of the species is 132 to 142; the subcaudal, 61 to 73. Measurements and scale counts of Natrix lineata (Peters). bere ke Bee A oe See es a SS : Piece S3i/ a) aa a 2; Hist 2 oS tes No. | Locality.|Collector.| 9 a a\8 a g Bia tae tele a uw |g ie eS Oe ee Sl ee toe et oe Si Bl s/ 2/8/41 8 8/8| 3 | ¢|8 a ei 8S 8 3 EB BH 6 | & 3 ai 5 dees bee] a end Ge bh be os mm.| mm. 1015__| Zambo- | E. H.| o& | 463 | 120 | 189 | 73 | 2 27°35 10 |3,4,5 5/1178 anga. Tay- ; lor. - 1046 ..|__..do_._|.--.do-..| ¢& | 412 | 107 | 186 | 68 | 2 2; 3 10 |3,4,5 6 |1+38 1016 --|_...do ...}----do---| co | 305]. 78 | 187) 68 | 2 Bi8 os 10 |3, 4,5; GO rcates. 4,5 1408 _.| Po rt|----do_.-| 9 | 598 | 140 | 182 | 61| 2| 1-2) 3] 8] 9-10 /3,4,5| 45/1+3 Hol- land, Basi- lan. 1480 _.| Abung- j---.do-..; yg | 266| 62) 134 | 64| 2 AS ee 10 |3,4,5 |. 4-5, 1+3 abung, Basi- lan. 1481 __|_...do_--|....do_..| 9 | 480; 105 | 186 | 64| 2 see ae 10; 4,5 6 j1+3 1590A |....do -._|-..-do...| oc’ | 550 | 140 | 183 | 68 | 2 2+ B18 10 3, 4,5; 6 |1+3 4,5 1590B_|....do__.|.._.do-..| yg | 260 | 57 | 183 | 62| 2] 12) 3| 8 10 |3,4,5 6} 1-+3 1590C_|_...do -..|_...do ...| ye | 258 | 60 | 182 | 67} 2)| 8-2) 3 8 10 |8,4,5 5 /1+8 1590D.|_...do _..|....do_..| 2 | 463 | 120 | 182 | 69} 2 21°34 3 10 |3,4,5 5 148 Natrix dendrophiops (Giinther). Tropidonotus dendrophiops GUNTHER, Ann. & Mag. Nat. Hist. V 11 (1888) 186, fig. Natrix dendrophiops dendrophiops TAYLoR, Snakes of the Philippine Islands (1922) 95. One specimen of this rare species was collected at Port Hol- land, Basilan. Its presence was discovered by hearing the cry of a frog which it had just caught and was beginning to eat. It was the only specimen seen. Natriaz lineata (Peters) and Natrix auriculata (Giinther) were taken in the same locality. Color in life-——Head and anterior fifth of body olive to olive brown, gradually more brownish on the second fifth; latter three-fifths of body olive to dull olive brown. On the second fifth distinctly maroon to red-brown on sides, more pronounced 21,3 Taylor: Herpetological Fauna, II 295 on the edges of the ventrals; the yellow spots in the black cross- bands are strongly pronounced, continuing some distance on the tail, less distinct on the neck. No spots on the parietals and no nuchal spot; black marks on lower labials at the sutures; spots on the first three upper labials and on the sixth. Scalation.—Nine upper labials, ten lower labials; one pre- ocular on left, two on right; three postoculars; temporals two, followed by three, the lower anterior very large; scales in seventeen rows; tip of tail missing; anal, double; ventrals, 168. This is the first record of the snake from Basilan. This species differs from Natrix barbouwri sp. nov. in the larger eye and the dentition. Oxyrhabdium modestum (Duméril and Bibron). Stenognathus modestus DUMERIL and BIBRON, Erp, Gén. 7 (1854) 504. Oxyrhabdium modestum TaAytor, Snakes of the Philippine Islands (1922) 100. I collected three specimens of this snake in 1920; two near Zamboanga, and one on Basilan Island. Gregorio Lopez col- lected a specimen at Cabalian, Leyte. The specimens from Zam- boanga and Basilan have rather shorter, wider heads than the Leyte specimen has, and a higher average of ventrals, the counts for the two Zamboanga specimens being much higher than any recorded counts. The known range is: Ventrals, 162 to 189; subcaudals, 49 to 66. The chin shields of the Leyte specimen are longer and narrower than those of the other specimens. The Zamboanga specimens were found in a rotted tree stump. The Basilan specimen collected at Abung-abung was found at the base of a small palm tree. Measurements and scale counts of Oxyrhabdium modestum (Duméril and Bibron). % Ad ¢ 128 ‘ > <= : . my o o a na Foot ; ei 3)a| 8 188] €) No. | Locality. | Collector. : eta le é 23 2 g elslele| 2 ii} 3) a aig e ee a le 2 | Bao 2 Sia | el. eee j mm, | mm. i 1017. .... Zambo-| BH. Tay-| 2 -} |. 7] -180| 02) 8) 7) EB, By 8 anga. lor. yy Reena erect BO cde do .. 9 482 80 | 186 57 7 |. 7-6 4-5 | 4-5 2 108... Basilan___|..... do....| 9 | 340} G2] 174| 88 8 Ede, i ep ee 0000_.___ Leyte_.... Gregorio| ¢& 500 98 | 165 57 | 93 y ee sg See Seen Lopez. 5:6 296 The Philippine Journal of Science 1922 Oxyrhabdium leporinum (Giinther). Rhabdosoma leporinum part., GUNTHER, Cat. Col. Snakes Brit. Mus. (1858) 12. Oxyrhabdium leporinum TAyYLor, Snakes of the Philippine Islands (1922) 103. Three specimens are in the collection, one collected by myself along the mountain trail near Haight’s (Pauai) ; the other two were collected by Father F. Sanchez, 8. J., at Baguio, and pre- sented to me. No. 697 when taken was light olive to yellow- green above, dusky below. This specimen differs from the other two in having the chin shields together, forming a circle. The shields are shorter and wider than normal. No. 384 is a young specimen with a broad light area behind the eye which narrows above and crosses the occiput on the posterior edges of the parietals. The body is crossed by thirty-seven narrow yellow lines not wider than a single scale. Measurements and scale counts of Oxyrhabdium leporinum (Giinther). {3 oe $3 _ = m a i ; a| 9/8) 2) 84 ; No. | Locality. | Collector. | 3 | é|3/3)/2/ 8/8318) 4 s| 8 21 Sts fepoes aay et sg 9 ek “A ee ee a @ = Eee eae ee ee ee ee a ee eee ion] © betas ee n a mm. | mm. 697-| Mountain | E.H.Taylor| ¢ | 640| 92} 175| 42} 7] 6| 45 4} 15| 142 trail, Ben- guet. 547_| Baguio,|Father| 92} 540] 90| 169] 49] 7| 6] 45 4} 15] 2+2 Benguet. Sanchez. 1+2 | eh Fy RSS docs ye | 235] 36] 167| 45| 7] 6) 4,5 4| 15] 142 Zaocys luzonensis Giinther. Zaocys luzonensis GUNTHER, Proc. Zool. Soc. London (1873) 169; TayLor, Snakes of the Philippine Islands (1922) 1385, pl. 12, figs. 1 and 3; pl. 18, figs. 1 and 2. Three specimens of this rare snake were obtained; one in northern Kalinga at Balbalan, one in Polillo Island, and the third was presented to me by Mr. O. W. Pflueger of the School of Forestry at Los Bafios. The Polillo specimen is very young and differs so markedly in coloration that I supected when I collected the specimen that it was new. Variation—The Kalinga specimen is brown above on the anterior part of the body, each scale with a black area or edged with black. The middle third of the body is lighter 21,3 Taylor: Herpetological Fauna, II 297 brown, the scales more deeply edged with black; in fact, the black is predominant. The latter third has the black still more prominent, with groups of flowerlike yellow-brown spots on each side of the median line of the back. The tail is entirely black save on the basal third, which has a few spots arranged in two rows on the dorsal surface. Chin and anterior third Measurements and scale counts of Zaocys luzonensis Giinther. 4 & | } 5 2) sia a CS PARES ae : ; No. | Locality. sl 413 83 2 4 4: E a “ 3 (oS ae hag Cae $i Bees 8 Bee g ¥ r= S28 ie | Se) EAs & | a Fa | o Ca) a@/| 9/3 |s8| 2! o]'s & | 9 © 8 n -l Hiri alalPle ie Ay | A ee | an emo | itil "ea audi Salat Mad Seo? CaP Pies mm, mm.| oe Kalinga ....| d | 1,640 | 480 192/122) 2/98) 9| 5/12] 8 242) 14 8 24241 f 1 | 295-.--- Polite: 2c 508 | 188 | 202/124 2| 8| 9} 5| 2/ 2 2+2| 14 | 000... Los Bajfios -.|-_.-.- 2,060 | 580 | 201 126 | 2|8is9! 5/382] 2 242) 14 of belly flesh color; median third flesh color anteriorly, each ventral edged with black and growing heavier posteriorly ; poste- rior third of belly and under tail deep black. The snout is light brown; the top of the head black-brown with no markings on labials. The specimen from Los Bajios differs in having much less black on the posterior part of the body. The yellow-brown spots continue to near the tip of the tail. The young Polillo specimen is gray, reticulated with black. On the anterior part of the body there are some dim white bars. The anterior part of the belly is flesh-colored; the posterior part and underside of tail black. Holarchus meyerinkii (Steindachner). Holarchus meyerinkii TAYLOR, Snakes of the Philippine Islands (1922) 139, I found a single young specimen of this species near Indanan, Jolo. I unearthed it from under débris collected about the base of a large tree. When exposed it crawled very slowly. Description—(No. 1710.) The specimen agrees with that figured in my Snakes of the Philippine Islands in the scalation of the head, save that the nasal is entirely divided instead of only partially. The scale counts are: Ventrals, 157; anal, single; 188431——-5 298 The Philippine Journal of Science 1922 subcaudals, 46. There are eight lower labials on the left side and seven on the right. Four lower labials touch the first pair of chin shields on the right side, and five on the left. Total length, 160 millimeters; tail, 27. Color in life—Above striped, dull salmon pink and brown; a pair of broad, dark brown stripes, each covering two whole and two half rows of scales, separated by a salmon pink, | straight-edged, median stripe, covering one whole and two half ‘rows of scales; the inner edge of the brown stripes on either side igs deep black, and the outer edge with numerous black ’ dots; below the broad stripes laterally is a flesh-colored stripe covering two half scale rows; below this another brown, black- edged stripe, also covering two half rows of scales; below this another dull flesh-colored line covering parts of two scale rows, below which is another line, nearly black, covering parts of two scale rows; another flesh pink line follows, covering part of the outer scale row and the edges of the ventral, the ventrals with elongate spots on each side forming a narrow black line; belly flesh color to coral pink; a curved, rather broad line crossing head involving eyes; the median brown lines meet on the frontal ; a line across the angle of mouth; deep black spots on the anterior ventrals. . Remarks.—The specimens whose scale counts are available give the following variation: 156 to 162 for the ventrals, average 158; subcaudals, 43 to 48, average 45. It will be noted that this form differs from Holarchus octolineatus (Schneider) not only in the lower ventral count but also in a very much lower subcaudal count. Psammodynastes pulverulentus (Boie). Psammophis pulverulenta Bog, Isis (1827) 547. Psammodynastes pulverulentus TAYLOR, Snakes of the Philippine Islands (1922) 209. Fifteen specimens were collected in Zamboanga, Basilan, and Jolo. I failed to find it in the other localities in which I made collections. This widely distributed species varies greatly in color, but in the material examined I cannot separate any varieties. The table shows the average ventral counts for five Basilan specimens as 170; the average for three Jolo specimens as 152; for the seven specimens from Zamboanga, 163. Thereis less variation in the subcaudal counts. The known range for ventral and subcaudal counts is 151 to 179, and 53 to 69, re- spectively. Females are darker for the most part than males. 21,8 Taylor: Herpetological Fauna, II 299 Table of measurements and scale counts of Psammodynastes pulverulentus (Boie). it pre g .|a)¢)4g 2|$ No. | Locality. | Collector.| & | #/ 3/3 8 Z roles Be ” al Be ee pee 3] @ = a E ee eee Be ee eo eee mm mm, i See ae Zambo-| E.H.Tay-| o | 315| 58| 168, 58| 283| 2| 2 si 7 anga. lor. pi eee Beet eae eae do....| ye | 290| 61| 160| 62) 2] 2 eee be MW ee Bae do....| d | 354] -75| 156| 65] 2] 2] 2] 8| 7 Wikies he atl cake do - 9-4 0064 184108) 80] 24. <8) 8 BES SIT pe NEcnn Pages do....| & | 89| 64| 161| 944; 2] 2] 1 gi 7 Wks ered Become dost 9°) el OT ie) me 8S eo 8 MA abcele. Peete | amare do....| 2 | 490] 89] 167] 68 2 2 2 8) 7 MED ssc: Basilan___|____. do itet.<951 | 1204/10 bt 8 Be 81.4 Wiha ori do....|- ye | 194] 94| 10} o:| 2] 2| 2! 8] 7 WE eeS oes a PF Te 8 1 ee os ee eT 8 7 ji) Seger en Cree dats do....| & | 345! 58| 174] 55 1 Sea ele Ses HRS pS eee do....| ye | 275| 55| 160] 63 Adis 1 g4:24 TS. Je aS Ee eae Bae! aa Slee See Ses see ee | gs EBs nea Cee eres de) Sa Or | oT 8 | ee 48 WG. cal ee) BO. deads Go ci) 92 S08 [78 (I | Bt). 2 tow 8 Tip of tail missing. Boiga dendrophila divergens Taylor. Boiga dendrophila divergens TAYLOR, Snakes of the Philippine Islands (1922) 201. / Two specimens of this recently described form have been taken; one (No. 301) in Polillo, the other (No. 2006) on Mount Maquiling, near the School of Forestry. The gray color of No. 2006 is strongly pronounced; the black color of the body merely borders the yellow crossbands. No. 301 has the loreal entering the eye; in No. 2006 the loreal is separated from the orbit by the preocular ; the second pair of chin shields is smaller. Scale counts—No. 301: Ventrals, 221; supcaudals, 94. No. 2006: Ventrals, 227; subcaudals, 95. No. 301 has 74 yellow bars; No. 2006 has 100. Remarks.—The fact that the specimens of Boiga dendrophila from Luzon and Polillo (the Samar record * is doubtful) show more resemblance to the Palawan form than to the Mindanao form is rather inexplicable. Using reptiles and amphibians as criteria there is very little evidence to show that Palawan has ever had any land connection with any part of the Philippines east of the Sulu Sea. In consequence, we would expect that ® Taylor, loc. cit. 300 The Philippine Journal of Science 1922 a Luzon form would show more similarity to the well-defined color form of Boiga dendrophila latifasciata which occurs in Min- danao. It is, of course, not improbable that both subspecies occur in Mindanao. However, no specimen of B. d. divergens has been found there. Hemibungarus calligaster (Wiegmann). Elaps calligaster WimGMANN, Nova Acta Acad. Leop.-Carol. 1 17 (1835) 253, pl. 20, fig. 2. Hemibungarus calligaster TAYLOR, Snakes of the Philippine Islands (1922) 269. Three specimens of this species were obtained at Los Banos; one through the courtesy of Prof. Charles Fuller Baker, dean of the College of Agriculture, the other two from Prof. O. W. Pflueger, director of the School of Forestry. No. 40 presents an interesting variation in the relative length and width of the body, as shown by the table. It varies also from the normal _ in having the first lower labials separated and the mental in -eontact with the first pair of chin shields. This condition is also present in No. 41. The very low subcaudal count and the high ventral count of No. 40 are also extraordinary. The tail is doubtless abnormal; there is no evidence that the shortening is due to injury. It is presumed that these specimens were collected low on Mount ‘Maquiling, about the grounds of the College of Agri- culture and the School of Forestry. Measurements and scale counts of Hemibungarus calligaster (Wiegmann) . No. 40, ? .|No. 54, oo, 41, ¢. Width.of body: ics osc. 532532 ves 3 es Sc oak ee eee mm.. 6 9 9 With of én 6.52: o0 25555602 25. kL ean noes + +2 mm... 5.5 8.5 q TiO ein cote wee ces a2 ok ee eee een ee mm 460 508 455 UWE ec ee sa a oe a ak le eee re eee mm 18 26 25 Wentrale soca eens Ss Co te Venn een Us Cee fae eae 256 224 228 Sublénndalé 2605 5255S ee ee ae 12 20 20 Lower lablalg 200 ke eee ws eee ae 5 q 6 Labials touch chin thields 2 2.2222 a ee, 3 4 3 Hemibungarus meclungi Taylor. Hemibungarus mceclungi TAYLOR, Snakes of the Philippine Islands (1922) 272, pl. 33, fig. 3; pl. 34, figs. 3 and 4. A specimen of this rare snake was taken on the road to Bislian on Polillo. However, it was attacked by ants when killed and certain of the head scales were destroyed. The characteristic shape of the frontal plate is clearly evident. Total length, 310 21,3 Taylor: Herpetological Fauna, II 301 millimeters; tail, 26. Anal, single; ventrals, 204; subcaudals, rat Color in life-——The body is nearly black above with very in- distinct dotted bands of cream on neck; belly banded with intense black and coral red, the red lighter on neck. The black bands below, each incorporating an elongate narrow white stripe. The species is obviously distinct from Hemibungarus calli- gaster and is very probably confined to Polillo or the Polillo group of islands. Doliophis philippinus (Giinther). Callophis intestinalis var. philippina GUNTHER, Rept. Brit. India (1864) 349. Adeniophis philippinus MryEr, Sitzb. Ber. Ak. Wiss. Berlin 36 (1886) 614, Doliophis philippinus TAYLOR, Snakes of the Philippine Islands (1922) 277, pl. 35, figs. 1 and 2. A single specimen of this rare species was collected in the mountains lying back of the Zamboanga waterworks intake. It was found near the summit, under a decayed log. When dis- turbed the tip of the tail was turned up to display the bright red spots beneath. Description of specimen.—(No. 1100, E. H. Taylor collection.) - Female. Total length, 390 millimeters; tail, 26; width of head, 4.5; length, 7.5. Ventrals, 260; subcaudals, 21; anal, single. Color in life-—Above brown with a black median stripe cover- ing three whole and two half rows of scales, inclosing longitu- dinal brown spots much longer than the interspaces between them; on either side of the black stripe are narrow brown stripes covering parts of two scale rows; below these a narrow black line which merges in the large black blotches on the belly; black blotches separated by pink to flesh white blotches, which reach the third outer scale row on side; spots below tail bright red. Naja naja philippinensis Taylor. Naja naja philippinensis TAYLOR, Snakes of the Philippine Islands (1922) 265. I have obtained two adult specimens of the Philippine brown cobra; one (No. 869) from Los Bafios, presented to me by Prof. C. F. Baker, dean of the College of Agriculture, and a second specimen, collected near Antipolo by Mrs. R. M. McCrory. The Los Bafios specimen has a very much longer head than the Antipolo specimen. The measurements are 52 and 43 milli- meters, respectively. 302 The Philippine Journal of Science Measurements and scale counts of Naja naja philippinensis Taylor. No. 869, ¢.|No. 000, &. BSCR RN ee dees caus RGN Ee owe wa awe Saas coNy, mm_.- 1, 423 1,120 Pail iclsc cesses pass te ca Se res ce cl bese ee cece. mm.-- 195 140 Ventrale icc nds SS ees CR ae 190 184 He bene0s le. oss es es es oo sn Ee nek en gkn seuakenctes 46 40 Upper lablaie cs ae Pec ce eee 7 7 Lower labials. - cis. joie 2 eeses cole sep sens tee a. see sean nee 8-9 Scale rows: OGRE sissc suse a eee oe eg ee ee kao sede eee 25 23 Body siskehc are ee a aaa ae esis 21 21 Naja naja samarensis (Peters). _ Naja tripudians var. samarensis PETERS, Mon. Berl. Ak. (1861) 690. Naja naja samarensis TAYLOR, Snakes of the Philippine Islands (1922) 259. A single specimen of this subspecies was collected near a trail which follows Tumugao River, some 20 kilometers from Zam- boanga. Color in life-—The specimen is uniform black; the skin be- tween scales, bright yellow. None of the scales has yellow - spots. Chin yellow. Beginning with the fifth ventral the fol- lowing twenty scales are deep black, gradually becoming cream. Outer edge of ventrals of most of the scales with dusky spots. — This subspecies is not rare in Zamboanga; several of the snakes have been seen or killed on the golf links. Measurements and scale counts of Naja naja samarensis (Peters). Length (mm.) : 1,010 Tail (mm.) 165 Ventrals 173 Subcaudals 48 Seale rows on neck 21 Scale rows around body 19 Upper labials fj Lower labials 8 Labials touching first chin shields 4 Trimeresurus wagleri wagleri (Boie). Trimeresurus wagleri wagleri TAytor, Snakes of the Philippine Islands (1922) 298. A specimen of this subspecies has been recently obtained from Balabac Island. It is uniformly green above, with two series of very small white dots on the back. The tail is brown on the tip, and has several larger white dots. Ventrals, 142; anal, single; subcaudals, 51. The subocular is separated from the labials. Fig. 1. Fig, FIG. FIG. 2. ILLUSTRATIONS PLATE 1 Rana yakani sp. nov.; photograph of a cotype specimen from Basilan, about natural size. . Rana suluensis Taylor; photograph of specimen from Jolo, enlarged. . Micrixalus diminutiva sp. nov.; photograph of cotype specimen from Zamboanga; actual length, snout to vent, 20 millimeters. . Micrixalus diminutiva sp. nov.; photograph of cotype specimen from Zamboanga; actual length, snout to vent, 20 millimeters. * PLATE 2 . Rana yakani sp. nov.; photograph of No. 1027, male, a cotype specimen from Zamboanga; actual length, snout to vent, 48 millimeters. . Micrixalus diminutiva sp. nov.; photograph of the type specimen, from Zamboanga; actual length, snout to vent, 21 millimeters. . Micrixalus diminutiva sp. nov.; photograph of a specimen from Jolo; actual size, snout to vent, 20 millimeters. Note the lesser extent of webbing on hind foot. PLATE 3 . Rana igorota sp. nov.; photograph of a cotype from Kalinga, about natural size. . Polypedates linki sp. nov.; photograph of the type specimen. PLATE 4 . Tropidophorus stejnegeri sp. nov.; photograph of the type. The characteristic markings on the back fail to appear in the photo- graph. : : Dibamus argenteus Taylor; anal region, from a Basilan specimen; enlarged. 3. Cornufer rivularis sp. nov.; photograph of the type specimen. 4, Cornufer montanus sp. nov.; photograph of the type specimen. 303 TAYLOR: HERPETOLOGICAL FAUNA, II.] [Puruip. Journ. Sct., 21, No. PLATE 1. TAYLOR: HeRPETOLOGICAL FAUNA, II] [Puitipe. Journ. Scr., 21, No. 3. PLATE 2. TAYLOR: HERPETOLOGICAL FAUNA, II.] [Puiuip. Journ. Sct, 21, No, PLATE 3. 3. 0, , SN Sc. JOURN, [ PHILIP. TAYLOR: HERPETOLOGICAL FAUNA, II.] PLATE 4. CORRELATION OF DEATH RATES FROM CERTAIN DISEASES WITH CERTAIN ECONOMIC AND HOUSING FACTORS IN THE PHILIPPINE ISLANDS 1 By RUFINO ABRIOL ONE TEXT FIGURE INTRODUCTION In the work of a health officer there constantly arises a multi- plicity of problems which cannot possibly be correctly solved without the application of modern statistical methods. Essen- tial among these methods is correlation. ‘By the use of this technique, the most complicated problems, which could be at- tacked in no other way, may be solved.”(7) While other statis- tical methods are being applied to public-health problems with considerable frequency, it can reasonably be said that this is not generally done with the method of correlation. In the solu- tion of biological, anthropological, meteorological, social, and economic problems the theory has found application, but in the domain of vital statistics little use has been made of it. It is gratifying to realize that in its meager use in vital statis- tics it has thrown light on numerous fundamental as well as surprising interrelationships between various phenomena; for example, compare Pearl’s work on the epidemiology of in- fluenza. (6, 8) In considering the general health conditions of a community, such as the prevalence of disease, malnutrition, infant mortality, general morbidity, etc., it is generally assumed that there must be a certain relation between these and the housing conditions or economic factors. It is possible that scientific analysis would show that these accepted relations are not so intimate as is generally believed. Also, there is the possibility that the simi- larity of conditions might not hold universally. As the material herein finds its source in the Philippine Islands, it should be * Papers from the department of biometry and vital statistics, School of Hygiene and Public Health, Johns Hopkins University, No. 49. 305 306 The Philippine Journal of Science 1922 borne in mind that variation in conditions doubtless is in some degree responsible for the results here presented. Concretely, in the Philippine Islands, of necessity, the dwell- ings are of a construction far different from that found in Amer- ica or Europe. Whereas in occidental countries warmth, to- gether with protection, must be considered in the construction of homes, in the Philippines, since there exists no need of heat- ing the houses, a loose construction is common that affords per- fect ventilation. Thus we find a fundamental difference. The size of the family per housing unit is generally taken as a criterion of “crowding.” In America the evil of this is mostly felt in centers of population, and it is especially among the poorer class and foreign element that this condition flourishes. In the Philippines it must be acknowledged that the same num- ber of people could live under a given roof, the ratio of which would be considered “crowding” in America, and not have the same injurious effect on the health of the Filipinos, due to the difference in climate, ventilation, etc. In fact, with the poorer class, which constitutes the bulk of the population, it is a com- mon occurrence to find several families not only under the same roof but also living in the same room. Consequently, it would appear that, with such close relationships, the possibility of interdetrimentalness would be greater than in the United States ; but the neutralizing effect of the facts that dwellings cannot be closed and that they have at all times perfect ventilation must be considered. Therefore, though the figures may show the index of “crowding” in the United States to check numerically that of the Philippines, yet upon application of the correlation method to death rates the resulting effects may be found differ- ent, due to other factors. Such was actually found to be the case. 5 FACTORS CORRELATED In the development of this idea, by the method of correlation, certain Philippine data were analyzed. For simplicity and with a view to further work along this line, the diseases selected as best suited for this problem are malaria, smallpox, Asiatic chol- era, dysentery, leprosy, beriberi, and pulmonary tuberculosis; and, lastly, all causes of death combined were considered. The diseases named are those adopted in the standard classification of the International Nomenclature of Causes of Death. Also, this classification is strictly adhered to in the annual reports of the Philippine Health Service, and was furthermore adopted in the Philippine Census of 1918. The death rates from these dis- 21,3 Abriol: Death Rates and Economic Factors 307 eases per 1,000 population in the various provinces have been correlated with certain economic and housing factors. Thus, should the results warrant following up this problem, there would be a satisfactory basis for comparative work. SOURCE AND NATURE OF DATA The data in this work are drawn from the Philippine Census of 1903, taken in March of that year. Therefore, the figures on economic factors, dwellings, and families represent those existing during the early part of 1908. For that year only par- tial and rather incomplete data on vital statistics are included, while for 1902 comprehensive figures are available. As the census gives figures on the economic, dwelling, and family factors for the early part of 1903, it is considered that the statistics on mortality for 1902 are sufficiently closely related in point of time. Therefore, the data of vital statistics for 1902 and the figures for the economic, dwelling, and family factors for 1903 are herein analyzed. In considering the provinces with reference to their popula- tion and relative wealth, a difficulty presents itself. The ob- jective is the finding of the ratio between the wealth of the provinces and their respective population that will give a con- crete measure for statistical use. As the determination of the income of the people is not feasible, the valuation of taxes and personal property and real estate are primarily considered. It seems fairly reasonable to assume that the taxes paid by a com- munity furnish a measure of its wealth. Likewise, the valuation of the personal property and real estate must serve a similar purpose. However, the question that arises is: To what degree can these elements be properly regarded as the measure of the economic status of the province and its people? Total taxes paid or the total value of personal property and real estate does not seem to furnish as true an index of the wealth of the com- munity as when these values are considered with respect to the individuals inhabiting the province. Consequently, the figures have been taken on a per capita basis, and both taxes and property valuations have been used for the correlations. ~The property values include personal and real property as given in the census. The taxes do not include customs collec- tions; they do, however, cover the amount of taxes paid for insular, provincial, municipal (town or city), and road purposes. Both the property and the tax values were estimated in the various parts of the Islands by tax assessors, appraisers, the \ 308 The Philippine Journal of Science 1922 presidentes,? or by the owners of the property. Therefore, al- though the figures do not represent actual valuations, they can reasonably be accepted as good approximations, accurately com- parable among themselves. The word “family” is herein used in the sense in which it is employed in the United States Census; that is, “a group of per- sons living under one roof, or in one suite or apartment.” (2) No distinction is made, however, between public and private families. A dwelling is considered as one roof, sheltering one or more families. Certain provinces have been omitted, because of incomplete and unreliable records. Manila, being a city, is also omitted for the simple reason that conditions obtaining there are not comparable with those of the provinces. This is particularly true in respect to the family factor, in which are included a large number of institutions such as prisons, colleges, hospitals, ete. Thirty-nine provinces are included in this work, namely: Abra. Masbate. Albay. Mindoro. Ambos Camarines. Misamis. Antique. Nueva Ecija. Bataan. Nueva Vizcaya. Batangas. Occidental Negros. Bohol. Oriental Negros. Bulacan. Palawan. Cagayan. Pampanga. Capiz. Pangasinan. Cavite. Rizal. Cebu. Romblon. Tlocos Norte. Samar. Tlocos. Sur. Sorsogon. Tloilo. Surigao. Isabela. Tarlae. Laguna. Tayabas. La Union. Zambales. Leyte. Zamboanga. Marinduque. THE CORRELATION METHOD A brief explanation of the correlation method may be of advan- tage to those not familiar with it. This method was first in- troduced on a sound basis by Galton,(4) and later elaborated by Edgeworth,(3) Pearson,(9) Yule,(11) and others. “Two characteristics are said to be correlated when there is a tend- ency for the changes in the value of one to depend on the * The presidente is an official corresponding to the mayor in an Amer- ican township or city. Abriol: Death. Rates and Economic Factors 21,3 309 changes in the value of the others.”(5) For instance, were death rates and taxes correlated in the various provinces, it would be found that the high death rates occur in the provinces where the taxes are high. This is positive correlation. If, on the other hand, the high death rates were found in provinces where taxes are low, there would also be correlation, but in this case a negative correlation. and average number Laguna 13.37 .... .... 6.48 Masbate Bataan 8.99 .... ....6.36 Surigao Rizal 1.88: 6.32 Occidental Negros Bulacan ( es Misamis Tayabas GAS. Mindoro Batangas 6.78 . Nloilo Antique 6.29 . Zamboanga Pampanga 6.07 . Romblon Ambos Camarines 5.61. Marinduque Cavite 5.52 Albay Nueva Ecija §.41. Antique Zambales 5,30 . Oriental Negros Noilo 5.02 . Bohol Tarlac 4.89 . Palawan Marinduque 4.88 . Ambos Camarines Pangasinan 4,83 . La Union Occidental Negros 4.58... Cebu Mindoro 4.30 ... Samar Capiz 4.22. Cayiz Cagayan 3.90 Sorsogon Isabela 3.74. Leyte Nueva Vizcaya 3.74 . Pampanga Abra 3.52 Bataan Albay 3.40 Tayabas Romblon 3.87 Abra Sorsogon 3.25 .. Cagayan Tlocos Sur 3.00 ... Pangasinan Tlocos Norte 2.76 . Batangas Misamis 2.66 . Zambales Oriental Negros 2.66 . Ilocos Sur La Union 2.61. Nueva Ecija Surigao 2.18 . Isabela Bohol 2.01 ... Tarlac Masbate 1.65... locos Norte _ Palawan 1.63 . Nueva Vizcaya Cebu 1.53 ... Bulacan Leyte 1.53 ... Rizal Samar 1.21... Cavite Zamboanga 1.05 ... Laguna Fra. 1. Correlation between death, rate per 1,000 population from tuberculosis (left column) of persons to a dwelling (right column) in certain provinces of the Philippine Islands. The application of the theory of correlation involves numerous mathematical computations. The simplest method enables one to determine roughly the degree of correlation between two 310 The Philippine Journal of Science 1922 variables by a simple diagram, such as is shown in fig. 1, with- out entailing mathematical calculations. Such a method, how- ever, does not enable one to measure the degree of association or correlation. From mere inspection of fig. 1, one may deduce that the higher rates correspond to, or, to use the statistical term, are “correlated” with, the lower figures in the dwellings. The underlying principle of the method of correlation is the measurement of the association between two values in terms of the coefficient of correlation. This coefficient is expressed as follows: S (#, . 2) N o, 92 N12 pam a= where 1,. = the correlation coefficient X, and X, = the variables correlated x, and x,— the deviations from the means of X, and X,, re- spectively ; > = summation o, and c, =the standard deviations of X, and X, N = number of observations (variates). The coefficient may vary from 0 to 1, in either the positive or the negative direction, depending upon the nature of the correlation. The numerical value of this coefficient expresses the degree of association, approaching complete correlation as the value approaches unity. In other words, if the value is either —1 or +1, the correlation is perfect; that is, for every change in one variable, there is a definite and constant propor- tional change in the other. o, and o, are derived according to the following formula: Hee (2?) oN N = number of pairs >= summation of x. oC Before conclusions can be drawn from the coefficient of corre- lation it is necessary to know something about the probable error. The probable error is a conventional measure of the reliability of results. > It is a constant so chosen that when its value is added to and subtracted from the result obtained or the numerical conclusion reached, it is exactly an even chance that the true result or conclusion lies either inside or outside the limits set by the probable error in the plus and minus direction. 21,3 Abriol: Death Rates and Economic Factors 311 For example, if it is stated that the mean age of death of persons dying in Baltimore is 39.83 + 2.60 years, it means that the mathematical probability that the true average age falls between 37.23 years (39.83 — 2.60) and 42.48 years (39.83 + 2.60) is exactly equal to the mathe- matical probability that the true age falls outside those limits. (7) The significance of our results, therefore, can be judged only after considering their relation to their probable errors. Bowly and Hooker(6) consider that when the correlation coef- ficient is six or more times the probable error it is practically certain that the two factors in question are associated; that is, the coefficient is significant. Raymond Pearl(7) considers a coefficient not significantly different from 0 when it is not more than four times its probable error. This basis of interpreting the probable error and coefficient will be considered as the crite- rion of significant results in this paper. ANALYSIS OF RESULTS The death rates from the diseases under consideration in the thirty-nine provinces are each correlated with the values of the personal property and real estate per capita, with taxes per capita, and with the average number of persons to a dwell- ing. Table 1 shows the means and the standard deviations of the factors entering into the correlation.. Table 2 shows the correlation coefficients. Casual examination of this table reveals a most interesting and rather surprising result, in as much as the coefficients are for the most part contrary to what was expected. For a closer scrutiny and analysis, each of the diseases is considered separately. TABLE 1.—Means and standard deviations of death rates from certain diseases and certain economic and housing factors. Variable. Mean. se ace SG Gr TESS Troi Bll COUSON 5. 65 fo. Sok hole ee ae 61.57 +2.63 | 24.38 +1.86 PS Sg OE asa ap: Soa eel See Se eee op nace negra eee Se 17.19 £1.07 9.87 +0. 75 WOMUMITIONE oe cee 2 hes wok van ccbenceusucateasesusdcuoceenncs 1.99 +0.35 8.28 40.25 MEWUTNG COOLOER ics see, 22 sea ec re ws weigh eon 18.80 £1.45 | 13.40 +1.02 Dpbentebires cei a ee ci es ork scart Gey Seen 2.55 +0.21 1.98 40.15 SIODRORG bo csi ceka cache ceo nce pec cwalaea Lace lows ceas oe soca 0.01740. 002 | 0.01740. 001 WOOTTON a aes Se ee 0.67 +£0.06 0.55 +0. 04 Pulmonary fubdrcilobis: 2005 oe as 4.35 +0.26 2.42 +0.19 Real O6t0te. 655 Gisecciceecc acd usenet tees ic al i a ek apap 78.24 +6.19 | 57.32 44.38 ROG sick dood subs ancccucs bck wud uiinue sacs nkeee erie elenes 0.96 +0.07 0.68 +0. 05 WMD Fe es re ee ee ee ae 4.70 +£0.06 0.52 +0. 04 PU TELL. gipamncshimge Stag Gane nen gett nd GRMN US ey AI orgy Sats MERU peg 5.55 +0. 06 0.58 +0.04 312 The Philippine Journal of Science 1922 TABLE 2.—Coefficients of correlation between death rates from certain diseases and certain economic and housing factors. | Correlated with— | Death rate from— Personal prop- : : : | creyand'real | Taxes. | Papegin | Reecnasm eee ee rt ee Saee ee | tht} egtibethtcvaluseeun este +0,1840.10 | +0.36+0.09| —0.52+0.08| —0.82+0.10 BOON asia Se ee +0,120.11} +0.2640.10| —0.22+0.10| —0,05+0.11 (einatipie cL, Ss sk ~0,08+0.11 | —0.1040.11| +0.1740.10| —0.09+0.11 | Asiatic cholera_.....------------- +0.0140.11; +0.4240.09) —0.4940.08| —0.10+0.11 1. Dysenteryccce.a--<9 3 teens 40.1140.11 | —0.12+0.11| —0.3940.09} —0.31+0.10 Yaprely 0 be ee —0,1940.10 | —0.04+0.11} —0.0940.11 | —0.28:0.10 [= Bevibert zou: S2k. 2s eee +0.1340.11; 0.114011) +0.0640.11 | +0.03+0.11 | Pulmonary tuberculosis..-.------ +0.3740,09 | +0. 0640.11) —0.4240.09 | —0.58+0.08 All causes—The coefficient with personal property and real estate is +-0.18, which, considered with its probable error, + 0.10, is not significantly different from 0. The coefficient with taxes is just four times its probable error and, therefore, prob- ably though not surely significant. The coefficient with the average number of persons to the family is high, revealing a negative association of the rates with the number of persons in the family; in other words, high death rates occur in prov- inces where families are small. Asiatic cholera.—The coefficients with taxes and with number of persons to the family are both significant. It is judged from this that the death rates from cholera are higher among the people in provinces where higher taxes per capita are paid. It furthermore means that the rates are inversely proportional to the number of persons in the family. There is, however, an apparent anomaly in the death rates having a negative corre- lation with the number of persons in the family and no correla- tion with the number of persons in the dwelling, considering the fact, alluded to above, that families and dwellings are factors of practically the same nature. Dysentery.—Economic factors show no association with death rates from dysentery. Number of persons in the family, how- ever, appears to be slightly negatively associated with the rates. Pulmonary tuberculosis—In this disease, as in all causes, Asiatic cholera, and dysentery, the association with the number of persons in the family is apparent. Unlike any of the others, however, the correlation with the number of persons in the dwelling is high and negative; in fact, the coefficient for this is the highest obtained in the whole series. This suggests that, ~~ 21,3 Abriol: Death Rates and Economic Factors 313 owing to the general social condition obtaining in the Phil- ippine population, the usual relations respecting familial infec- tion are somewhat upset. Taxes, as a factor, show no associa- tion. The correlation coefficient with personal property and real estate is shown in Table 2 as being slightly more than four times the probable error. The actual figures, carried to four decimal places, however, are +0.3708 with a probable error of +0.0932, the former being slightly less than four times the latter. The coefficient, therefore, does not certainly indicate any association of death rates with real estate. Malaria, smallpox, leprosy, and beriberi.—None of the co- efficients for these diseases has any significance. DISCUSSION The results presented must be regarded as entirely prelimi- nary to more-detailed statistical analysis. There is involved an element of spurious arithmetical correlation due to the fact that the data correlated are rates having the same denominator for both deaths and economic and social variables. However, it is believed that the preliminary gross results here presented have considerable suggestive value. Yule(12) has argued that in such death-rate correlations the gross coefficients (which in- clude the so-called spurious element) have real significance be- cause they represent what is really the actuality. Green- wood’s(1) study of death-rate correlations of cancer and diabetes appears to bear out this contention. Pending further analysis by the methods of partial correlation, to eliminate wholly the spurious element, the present conclusions should be accepted with caution and reservation. In the consideration of the various correlations it is deemed necessary to note just what may be the effects of civilization upon a people like the Filipinos, presenting as they do a popula- tion of heterogeneous character. Down through history, the acquisition of civilization by the so-called primitive peoples has been followed by a general weakening of their natural powers. Attention is drawn to its effects upon the American Indians, for instance. Since they, as nations, possessed no acquired specific immunity against the diseases that follow closely in the path of civilization, they easily succumbed to them. Another effect, closely allied, is the influence upon the size of the families. It is not the general trend of sociological development that the less-enlightened peoples have larger families than those endowed with the benefits of culture and civilization ? 188431——6 314 The Philippine Journal of Science 1922 Furthermore, it is generally conceded that the people of the better and wealthy class, who naturally represent the portion of the population possessing higher culture and civilization, appear to have lower resistance to disease than the lower and poorer class, who are hardy and live close to nature. The families of the latter class are large; many families live under the same roof; however, due to the climate, they do not cling to the dwelling as do the better and wealthier class. Thus, a large number of persons in the family or dwelling among these people is only an apparent crowding and not a real one. None of the diseases except Asiatic cholera exhibits association with either real estate or taxes per capita. In cholera there seems to be a definite correlation with taxes, indicating that in the Philippine Islands the death rates from this disease are found highest in the provinces where the population pays rela- tively higher taxes. If the amount of taxes per capita in a province can be considered as a fair index of the wealth of the population, the conclusion that can be tentatively drawn from this coefficient is that the wealthier the people the more liable they are to infection from cholera. This seemingly puzzling phenomenon is precisely what would be expected if we were to consider the effects of civilization as referred to above. Ex- amination of the data reveals the fact that where death rates are highest, the provinces concerned possess a population of a higher culture than do those which present lower death rates. _ The results with the housing factors are likewise puzzling. If the coefficients are to be taken as a true measure of correlation between the death rates and the housing factors, at least as far as all causes, Asiatic cholera, dysentery, and pulmonary tuberculosis are concerned, higher rates are found in provinces where the families are smaller, and in the case of tuberculosis, also, where the average number of persons to a dwelling is lowest. Again, the effects of civilization are offered as a possible explanation of this phenomenon. It has been suggested above that civilization tends to make the families smaller; likewise, it is pointed out that morbidity increases with the advance of civilization. It may be, then, the combined effects of these factors that influence the correlation of death rates with the housing variables. It is to be remembered that the poorer and less-enlightened class is the one that possesses larger families and dwellings with a larger number of occupants. It must be remembered, too, that only numerical crowding exists, and that the dwellings are well ventilated. Then there is the 21,3 Abriol: Death Rates and Economic Factors 315 possibility that the more crowded the dwellings are the greater will be the tendency for the people to live out of doors. They are, consequently, the hardy type, the outdoor-loving people, and therefore the class that possesses higher resistance to disease. This assumption is apparently borne out by the fact that in the provinces in which the death rates are lower and the families larger the population belongs to that class. TABLE 3.—Deaths from certain principal causes in certain provinces, Phil- ippine Islands. Data for 1902, Philippine Census, 1908. Cause of death. Province. . BS aaa ae sf All | Malaria. | Small- | Asiatic | Dysen- | Lep- | Beri- | Tubereu- causes. pox. cholera. tery. rosy. | beri. losis. DIG Coe 2e os 1, 296 398 1 464 41 0 6 133 MS ig se 11, 564 38, 108 1, 289 + 480 279 2 187 815 Ambos Camarines__| 10,872 2, 634 1,189 1,144 327 1 494 1,309 Antiques. ico. 7,411 3, 181 173 827 372 1 19 826 Bateaniis cosa: 4, 479 618 8 2, 252 218 0 29 406 Batangas --___-._--_- 34, 257 13, 216 44 10, 383 1, 784 2 282 1,747 Bohol. oc:ss..tae 10, 851 2,748 425 4, 883 279 3 30 542 Bulacatis-.-25245 13, 015 2, 041 248 2,494 413 1 198 1, 602 Cagayam.__--..-..- 9, 783 2, 724 2,418 1,375 249 8 57 557 Caples thence 15, 564 3, 254 694 5, 903 1,339 4 263 949 Cavite 223 25 oo 8, 983 2, 073 289 1, 893 374 2 87 744 Lo Bele ae unpre oe ee 20, 920 7, 020 1,269 5, 800 216 20 673 1,001 Nlocos Norte -__.___- 10, 657 2, 236 233 8,907 1, 057 1l 28 488 Nocos Sur_..---.---- 12, 694 1,214 4] 8,187 293 cae 13 521 11 | pa ed eames 51, 158 16, 572 338} 21,772 645 5 168 2, 028 PRRDMD is So 4, 584 1,357 569 861 267 1 23 257 RARONG cies sss 15, 918 4,877 2 4,768 806 3 241 1, 987 Se OnIGR oo 8, 821 1, 008 449 4, 647 696 4 30 333 Beers 2d 17, 016 5,011 2, 465 8,618 269 8 381 594 Marindugue -..._._- 8, 222 1, 466 0 265 156 0 43 252 Maabatein. oo 1,881 811 18 664 9 0 7 72 Mindoro .......0.2.. 2, 466 621 113 854 226 0 66 139 Misamib vac. i5.ic6.z 9, 038 4,961 0 2, 386 155 3 190 361 Nueva Ecija____...- 8, 487 2, 553 121 1,642 332 5 75 720 Nueva Vizeaya-_.... 1, 228 611 1 433 14 0 1 60 Occidental Negros_.| 24, 461 8, 220 17 12,710 837 2 72 1,390 Oriental Negros_--.- 5, 972 2,185 77 1,875 110 8 92 492 Palawan i266 oo 664 160 0 185 26 0 19 47 Pampang 14, 383 38, 247 127 2, 672 502 1 60 1,352 Pi i 24, 701 6, 531 104 9, 586 1,028 22 86 1,905 pT Be ae ieee 11, 883 1, 648 148 8, 370 458 3 187 1, 167 Romblee. 3 1, 422 531 0 164 30 2 33 178 nn ea eae 18, 614 3, 918 1,789 | 4,226 253 9} 235 320 Sorsogon 2.252265 4.380 998 82 810 157 1 92 891 Surtgacec.. gk 8,304 1,200 35 1, 142 25 1 62 216 Perles tik 7, 088 2, 453 55 1, 464 143 1 84 653 Tayabas .._......... 9,418} 3,501 1| 1,569 “680 1} 254) 1,075 Zambales __.__-_---- 6, 772 1,398 896 2,180 584 2 16 537 Zamboanga SRR ann Ee 1,559 866 0 752 64 0 nies 41 All provinces -...... 425,781 | 117,504 14,831 | 134,052 15, 708 144 | 4,834 28,207 316 The Philippine Journal of Science 1922 SUMMARY The coefficients obtained indicate that the death rate from Asiatic cholera are positively correlated with the taxes paid per capita in the provinces. In the provinces where higher taxes are paid, the death rates from this disease are higher. TABLE 4.—Death rate per 1,000 population from certain principal causes in certain provinces of the Philippine Islands. Computed from Tables 8 and 5. is | Cause of death. | Province. : Tu- | All | Ma- | Small-|Asiatic|Dysen-| Lep- | Beri- | pepoy- causes.) laria. | pox. |cholera.) tery. | rosy. | beri. | j,i, Abe s: Coes Se ces 34,26 | 10.52 0.038] 12.27 1.08 | 0.000 0.16 |} 3.52 AES Raul wee coke ora 48.80} 12.96} 5.17} 2.00} 1.17] 0.008; 0.78) 8.40 Ambos Camarines._._..-------- 46.57 | 11:28 5.09 4.90 1.40 | 0.004 2.12 | 6.61 Antiree 2 a 56.47 | 24.24 1,32 6.30 2.83 | 0.008 0.14} 6.29 ete «oS aS 99.17 | 13.57 0.07 | 49.86 4.83 | 0.000 0.64 | 8.99 BOtgneGGe roo sons cece 132.93 | 51.28 0.17 | 40.29 6.92 | 0.008 1.09; 6.78 MBA St ee 40,30 | 10.21 1.58 | 18.14 1.04} 0.011 0:11 |. :261 ei as SS 58.28 9.14 cis ba beg 8 ey 1.85 | 0.004 0.89} 7.17 Cagay: 68.50 | 19.07} 16.98 9. 63 1.74 | 0.056 0.40 | 3.90 CilG kg Boe. 69.14 | 14.46 3.08 | 26.22 5.95 | 0.018 11%) 4:23 One... SS 66.65 | 15.38 2.14) 14.05 2.77 | 0.015 0.65; 5.52 Cebts 2 Shi a ae $2.00 | 10.74 1.94 8.87 0.33 | 0.031 1.038; 1.53 Tlonon Notté 2233 A 60.28 | 12.65 1.82 | 22.10 5.98 | 0.062 0.16 | 2.76 Hooon Sars: oh 73.04| 6.98} 0.02] 47.11} 1.69) 0.040; 0.07) 3.00 Iloilo. eS ae 126.64 | 41.03 0.08 | 53.90 1.60 | 0,012 0.42 | 5.02 leehelt jo 66.63 | 19.73 8.27 | 12.52 3.88 | 0.015 0.83 | 3.74 Rewanee b5. ccc 107.12 | 32.82 0.01} 32.05 5.42 | 0.020 1.62 | 13.37 tet Dales. oe eee 69. 03 7.89 3.51 | 36.36 5.45 | 0.031 0.23 | 2.61 Leyte 43.75 | 12.88 6.34 9.30 0.69 | 0.021 0.98 | 1.53 Mirinduwse 62.85 | 28.87) 0.00| 5.13) 3.02/ 0.000) 0.83) 4.88 Miia te a 43.07 | 18.57 0.41} 15.20) 0.21; 0.000 0.16; 1.65 Mindoro ‘ 76.30 | 19.22 8.50 | 26.42 6.99 | 0,000 2.04 4.30 Misamis ---- 66.71 | 36.62} 0.00| 17.61] 1.14} 0.022!) 1.40) 2.66 Nueva Beiiisso2i5 2 tS 63.81} 19.20} 0.91| 12.35] 2.50| 0.038 0.56; 5.41 Nueva Vizcaya. .... .-....--..-- 76.63 | 31.89} 0.06| 27.02} 0.87] 0.000; 0.06| 3.74 Occidental Negros -...-.-_.--- 80.55 | 10.60} 0.06; 41.86) 2.76) 0.007| 0.24) 4.58 Oriental Negros-__..........-... $2.30 | 11.55 0.42 | 10.14 0.59 | 0.043 0.50} 2.66 Palaw at ae, 5.55 0.00 6.41 0.90 | 0.000 0.66 | 1.63 Pampang 64.60} 14.58| 0.57) 12.00, 2.25! 0.004/ 0.27] 6.07 Pengesie..oSe 62.61 | 16.55 0.26 | 24.17 2.61; 0.056 0.22 | 4.83 Rizal 80.02 | 11.10 1.00 | 22.69 8.05 | 0.020 1.26 | 7.86 Romblon. -.- F 26.91 10.05 0.00 3.10 0.57 | 0.038 0.62 | 3.37 Sawer onic age nieces 51,27 | 14.74 6.74} 15,91 0.95 + 0.034 0.88! 1.21 SordowoWi wise oo ee, 36.36 8.29 0.68 | 2.57 1.30 | 0.008 0.76 | 8.25 Sorines 5 eae 83.27) 12.08) 0.35) 11.50] 0.25] 0,010; 0.62] 2.18 DMPA oii ne 53.09 | 18.37} 0.41] 10.97| 1.07] 0.007] 0.25) 4.89. TAYVADRS <5. -2. .<-se conc nesasens 62.68 | 23.30; 0.01); 10.44) 4.53} 0.007 1.69} 7.15 FeO WS 6.4.0 55s. vss icenenees 66.80 | 13.79 3.91 | 21.50 5.76 | 0.020 0.16 | 5.30 Gertie ooo os ae 39.79} 9.34] 0.00}; 19.19| 1.63} 0.000; 0.03) 1.05 Ail provinces ............-2.-.-. 63.19 | 17.44| 2.20} 19.89| 2.88] 0.021} 0.72] 4.19 21,3 Abriol: Death Rates and Economic Factors 817 In the case of all causes, Asiatic cholera, dysentery, and pul- monary tuberculosis, the higher death rates are found in the provinces where the average number of persons in the family is lower. In pulmonary tuberculosis alone there is a significant TABLE 5.—Civilized population, total dwellings and families, and average number of persons to a dwelling and a family in certain provinces of the Philippine Islands. Data for year 1908, Philippine Census, 1908. Average persons | * per— Province. Population.| Dwelling. | Families. | Dwelling.| Family. pi lik sg oe ee oS a eee tese 87, 828 7,177 9, 072 5.27 4,17 Alby ols eo tae aseds 239, 434 39, 720 50, 563 6.03 4.74 Ambos Camarines - .--------------------- 283, 472 40, 410 48, 826 5.78 4.78 Antique........--...-.--. ~----- <--<-+---- 131, 245 21, 842 24, 701 6.01 5.31 Moatente. 20. 605 oe SK os ee ee ee 45, 166 8, 432 9, 928 5.36 4,55 Batangas ........2-2-22--- ----25-<--- soe 257, 715 49, 538 63, 831 5. 20 4.04 Holl 2: oes cs es ee 269, 223 45, 687 46, 901 5.89 5.74 Bulacam si.c.5 5 fo hs eke sce 223, 327 47, 212 49, 266 4.73 4.53 Cawayan ii fis Bs a wee 142, 825 27,297 28, 328 5.23 5.04 Capiz en aes et ances 225, 092 89, 544 48, 769 5.69 4.62 Cavite 126 222-25 ces -ehspce ees 134, 779 29, 124 83, 853 4.63 3.98 Cebu E < 653, 727 118, 481 116, 069 5.76 5.63 Ilocos Norte --..-------------------------- 176, 785 35, 811 43, 773 4.94 4.04 Tlocos Sur---- ae ate 178, 800 34, 124 44, 180 5.09 3.98 ROG 86 ei RS, eee 403, 932 65, 795 94, 454 6.14 4.28 Taehele (3535 sooo ee os de dasa ane 68, 793 13,814 14, 627 4. 98 4.70 Laguna -..---..---------------------+----- 148, 606 32, 308 87,217 4,60 3.99 Te Unidl soo eis Ch ede ee 127, 789 22,117 28, 030 5. 78 4.56 Tab te 6 sha o ee enseeee 388, 922 69, 526 71, 659 5.59 5.43 Marinduaue -.----------: --- ------------- 51,674 8, 550 9, 445 6. 04 5.47 Mian at os oon. ass Sen ca ot eee 43, 675 6, 790 8, 493 6.43 5.14 Ady eee oe ok cdi ee enn ae 82, 318 5, 236 7, 282 6.17 4.47 Misathté <<. -.-.- 20. .c5 = assoc 185, 473 21, 443 25, 611 6.32 5.29 Nueva Wella 22-3220: -.-.eaae--- 182, 999 26, 630 30, 664 4.99 4.34 Nueva Vizcaya -------------------------- 16, 026 3, 269 4,470 4.90 3.59 Occidental Negros-.--------------------- 303, 660 48, 030 68, 027 6.32 4,44 Oriental Negros ------ 184, 889 30, 889 36, 970 5.99 5.00 Dain Wario is sock So ace tewe= 28, 852 4, 928 6, 420 5.85 4.49 Pampanga -..----------------------------- 222, 656 40, 994 47, 257 5.48 4.71 : Pangasinan ...<2..400--ic.----<2-— ~~ 394, 516 75, 652 92, 080 5.21| 4.28 Steel <2 8 ee ne 148, 502 81, 466 31, 957 4,72 4.65 Romblon ...-------------------------+---- 52, 848 8, 726 11, 046 6.08 4.78 Samar 2.25.5. ---2- 25.2. 22 ------ eens 265, 549 46, 428 56, 121 5.72 4.73 Sorsog ee 120, 454 21, 205 238, 582 5.68 6.11 Surigao ...--- ------2---2------- ----<2 20" 99, 298 15, 620 17, 265 6. 36 6.75 Warlne ..c..c0c-c--epctawss<--- - + y o a “ 346 The Philippine Journal of Science 1922 RUTACEAE EVODIA HAINANENSIS sp. nov. Arbor parva (4 m alta fide McClure), partibus junioribus leviter hirsutis; ramis teretibus, glabris, rubro-brunneis, circiter 3 mm diametro, parcissime lenticellatis, ramulis leviter hirsutis; foliis pinnatis, 25 ad 35 cm longis, foliolis 9, membranaceis, oblongo-lanceolatis, tenuiter obtuse acuminatis, basi acutis, superioribus 11 ad 15 cm longis, 2.5 ad 4 cm latis, inferioribus brevioribus, nervis utrinque 9 ad 15, patulis, curvatis, distinctis, anastomosantibus, reticulis laxis, obscuris, supra olivaceis, sub- glabris, vel ad costa pubescentibus, subtus perspicue atro-glan- dulosis, ad costa leviter hirsutis, petiolulis circiter 3 mm longis; cymis terminalibus, circiter 7 cm longis et latis, trichotomis, leviter hirsutis; coccis plerumque 4, obovoideis, 5 mm longis, brevissime apiculatis, leviter rugosis, obscure glandulosis. Neg Chi Leng, McClure 8449, December 7, 1921. On slopes, altitude about 1,000 meters, known to the Lois as ngau tau shue. A species belonging in the group with Evodia rutaecarpa Hook. f. & Th., well characterized by its differently shaped, narrower, nearly glabrous leaflets, and its short inflorescences. EUPHORBIACEAE BREYNIA ROSTRATA sp. nov. Frutex vel arbor parva, glabra, ramis ramulisque teretibus, laevis, in siccitate atro-brunneis; foliis chartaceis vel subcoria- ceis, rigidis, oblongo-ovatis, 2 ad 3 cm longis, 1 ad 1.5 cm latis, basi acutis ad obtusis, apice acuminatis, in siccitate atro-oliva- ceis, utrinque concoloribus vel subtus leviter pallidioribus, nervis utrinque 4 vel 5, tenuibus, arcuato-anastomosantibus, obscuris, supra plerumque leviter impressis, secondariis obsoletis; petiolo circiter 2 mm longo; stipulis lanceolatis, acuminatis, petiolo subaequantibus; calycis circiter 6 mm diametro, sepalis majori- bus orbiculari-ovatis, minoribus leviter anguste ovatis, leviter accrescentibus, sub fructu reflexis; fructibus globosis, glabris, laevis, circiter 7 mm diametro, perspicue rostratis. Ng Chi Leng, McClure 8516, December 22, 1921. In forested ravines, altitude 900 meters. A species much resembling Breynia fruticosa Hook. f. in its general appearance, distinguished however by its very strongly reflexed calyx segments and its prominently rostrate fruits, the stout rostrum about 2 mm long, formed by the persistent styles. 21,4 Merril: Diagnoses of Hainan Plants 347 % ANTIDESMA HAINANENSE sp. nov. Frutex 1.5 m altus, ramulis sordide pubescentibus, ramis tere- tibus, cinereis, glabris; foliis chartaceis, oblongis ad oblongo-lan- ceolatis, 10 ad 15 cm longis, 3 ad 4 cm latis, olivaceis, nitidis, supra costa excepta glabris, subtus ad costa nervisque pilosis, apice perspicue acuminatis apiculatisque, basi acutis, nervis utrinque circiter 7, distantibus, arcuato-anastomosantibus, supra impressis, subtus cum reticulis primariis laxis valde perspicuis; stipulis filiformibus, usque ad 5 mm longis, pilosis; petiolo dense pubescente, 4 ad 8 mm longo; racemis fructiferis axillaribus, solitariis, 3 cm longis, perianthii segmentis 4, oblongo-lanceo- latis, acuminatis, 1 mm longis, pubescentibus; fructibus obovoi- deis vel suborbicularis, compressis, 7 ad 8 mm longis, obtusis, glabris, leviter rugosis, pedicellis circiter 1 mm longis, bracteolis persistentibus, petiolo subaequantibus. Ng Chi Leng, McClure 8702, December 20, 1921. On forested slopes, altitude 700 meters. A form first identified by me with Antidesma apiculatum Hemsl. but differing from that species in very numerous char- acters, notably in its smaller, fewer-nerved leaves which are pilose on the nerves beneath, shorter pedicels, much shorter racemes, and differently shaped fruits. MALLOTUS MACLUREI sp. nov. § Axenfeldia. Frutex circiter 1.5 cm altus, ramis teretibus, glabris, ramulis obscure puberulis; foliis suboppositis, in paribus inaequimagnis, haud granulari-glandulosis, chartaceis, in siccitate pallidis, supra glabris, subtus ad costa leviter pilosis, subcaudato-acuminatis, deorsum leviter angustatis, basi truncato-rotundatis vel subcor- datis, nervis utrinque 7 vel 8, perspicuis, irregulariter dentatis, foliis majoribus usque ad 13 cm longis et 4.5 cm latis, minor- ibus 6 cm longis et 2 cm latis; petiolo foliis majoribus 1 ad 2 cm longo; infructescentiis simplex, racemosis, circiter 6 cm longis, capsulis 10 ad 12 mm diametro, granulari-glandulosis, echinatis, echinis sparsis, submollis, 2 ad 3 mm longis, curva- tis, glabris. Ng Chi Leng, McClure 8558, December 20, 1921. In forested ravines, altitude 600 meters. This species is allied to Mallotus esquirolii Lév. from which it differs in its smaller leaves which are not strictly opposite, and which are truncate-rounded or even slightly cordate at the base. The capsules are not densely echinate, the spines being distinctly scattered. qn j_wW 348 The Philippine Journal of Science 1922 ICACINACEAE GOMPHANDRA HAINANENSIS sp. nov. Frutex vel arbor parva, ramulis junioribus exceptis glabra; ramis teretibus, cinereis, in siccitate rugosis, 2 ad 3 mm dia- metro, glabris, ramulis adpresse subferrugineo-pubescentibus ; foliis glaberrimis, membranaceis vel subchartaceis, anguste oblongis, olivaceis, nitidis, 9 ad 14 cm longis, 2 ad 3 cm latis, obtuse acuminatis, basi acutis, nervis utrinque 5 ad 8, valde distantibus, supra subobsoletis, subtus distinctis, arcuato-anasto- mosantibus, reticulis laxis; petiolo 1 cm longo; fructibus oblongo- ellipsoideis, obtusis, leviter inaequilateralibus, circiter 13 mm longis et 7 mm diametro, pedicellatis, pedicellis 3 mm longis. Ng Chi Leng, McClure 8546, December 20, 1921. On forested slopes, altitude about 700 meters. This is apparently most closely allied to Gomphandra cam- bodiana Pierre, from which it differs in its entirely glabrous leaves and its conspicuously arched marginal nerves. GONOCARYUM MACLUREI sp. nov. Arbor parva, glabra, ramulis tenuibus, olivaceis, 2 mm dia- metro; foliis coriaceis, oblongis ad oblongo-ellipticis, olivaceis, nitidis, 7 ad 10 cm longis, 3 ad 4.5 em latis, brevissime obtuse acuminatis vel obtusis, basi obtusis ad late acutis, nervis utrinque 5 vel 6, distantibus, tenuibus, obscure arcuato-anastomosantibus, reticulis obscuris; petiolo circiter 1 cm longo; fructibus ellip- soideis vel oblongo-ellipsoideis, laevis, nitidis, in siccitate nigris, 3 ad 4 cm longis, 1.8 ad 2.4 cm diametro, apice leviter depressis apiculatisque. Ng Chi Leng, McClure 8606, December 16, 1921. On partly forested slopes, altitude 700 meters. | The first representative of the genus to be found in China, its alliance being with Gonocaryum subrostratum Pierre and more especially with G. harmandianum Pierre, from both of which it is distinguished by its smaller, differently shaped leaves. SABIACEAE MELIOSMA ANGUSTIFOLIA sp. nov. Arbor parva, circiter 5 m alta, inflorescentiis exceptis glabra, ramulis ultimis 7 mm diametro; foliis pinnatis, circiter 25 cm longis, foliolis circiter 19, lanceolatis, integris, coriaceis, 5 ad 8 em longis, 1.5 ad 2 cm latis, utrinque subaequaliter angustatis, acute acuminatis, basi acutis, leviter inaequilateralibus, supra pallide olivaceis, nitidis, subtus pallidioribus, nervis utrinque 8 ad 10, obscuris, arcuato-anastomosantibus, petiolulis 5 ad 7 mm 1124 21,4 Merril: Diagnoses of Hainan Plants 349 longis; paniculis terminalibus, erectis, pedunculatis, foliis aequantibus, ramis primariis 5 ad 6 cm longis, patulis vel ad- scendentibus, rigidis, plus minusve hirsutis; fructibus globosis vel obovoideis, 4 ad 5 mm longis, glabris, breviter pedicellatis. Ng Chi Leng, McClure 8507, December 22, 1921. In forested ravines, altitude 950 meters. A species strongly characterized by its narrow, lanceolate, en- tire, sharply acuminate, obscurely nerved leaflets. RHAMNACEAE RHAMNELLA HAINANENSIS sp. nov. Frutex glaber, vel ramulis junioribus parcissime pubescen- tibus, circiter 3 m altus, ramis teretibus, tenuibus, in siccitate pallidis, ramulis olivaceis, vix 1 mm diametro; foliis chartaceis, oblongo-ellipticis, 3.5 ad 8 cm longis, 1.5 ad 3.4 cm latis, deorsum integris, sursum obscure serrulatis, nitidis, pallide olivaceis, basi late acutis ad subrotundatis, apice distincte et plerumque obtuse acuminatis, nervis utrinque 5, tenuibus, distinctis; petiolo 3 ad 5 mm longo; fructibus plerumque solitariis, oblongo-ovoideis, 1-locularis, circiter 1 cm longis et 5 mm diametro, acutis; pedi- cellis circiter 8 mm longis. Ng Chi Leng, McClure 8358, December 6, 1921. In shaded ravines, altitude about 800 meters. Among the six hitherto known representatives of this genus the present species seems to be nearest to Rhamnella wilsonii Schneider. PASSIFLORACEAE _ ADENIA MACLUREI sp. nov. Suffruticosa, scandens, glabra, ramis in siccitate pallidis, 2 ad 3 mm diametro, leviter striatis; foliis chartaceis, oblongis, integris, pallide olivaceis, utrinque concoloribus nitidisque, 10 ad 12 cm longis, 4.5 ad 6 cm latis, subacutis ad obscure acumi- natis, basi late acutis ad subrotundatis, auriculato-glandulosis, nervis utrinque plerumque 4, distinctis, anastomosantibus, reti- culis laxis, petiolo 2 ad 2.5 cm longo; cirrhis rigidis, 5 ad 6 cm longis; fructibus anguste ellipsoideis, circiter 8 cm longis, 3- valvis, nitidis, laevis, basi acutis, apice obtusis, solitariis, pedun- culo 5 ad 6 cm longo; seminibus circiter 8 mm longis, compressis, rugosis. Notia, McClure 8242, October 27, 1921. In dry thickets. The special characters of this species are its oblong, subacute, noncordate, entire leaves. It belongs in the group with Adenia acuminata (Blume) King. 350 The Philippine Journal of Science 1922 MYRTACEAE 415% EUGENIA MACLUREI sp. nov. § Syzygium. : Frutex glaber, 2 m altus, ramis tenuibus, teretibus, ramulis 1 mm diametro, angulatis; foliis oblongis ad oblongo-ellipticis, chartaceis, olivaceis, 2 ad 4 cm longis, 0.5 ad 1.5 cm latis, utrin- que subaequaliter angustatis, basi acutis, apice obtusis vel ob- scure obtuse acuminatis, supra minutissime puncticulatis, costa impressa, subtus paullo pallidioribus, nervis utrinque numerosis, tenuibus, obscuris, primariis quam secundariis vix magis dis- tinctioribus; petiolo 1 mm longo; inflorescentiis axillaribus ter- minalibusque, quam foliis brevioribus, paucifloris, racemosis vel subumbellatis, breviter pedunculatis ; fioribus circiter 1 cm longis, brevissime pedicellatis, calycis circiter 8 mm longis, truncatis, teretibus, apice circiter 3 mm diametro, deorsum gradatim an- gustatis; filamentis brevibus, petalis calyptratim connatis, deci- duis, calyptra circiter 3 mm diametro. Taai Uen, McClure 7678, October 26, 1921. In dry thickets. A species belonging apparently in the group with Hugenia flu- viatilis Hemsl. but with smaller leaves and very differently shaped flowers. : ERICACEAE «J. © RHODODENDRON HAINANENSE sp. nov. | Frutex erectus, multiramosus, ramulis fastigiatis, adpresse hirsutis; foliis linearis ad lineari-lanceolatis, confertis, 1.5 ad 2 cm longis, circiter 3 mm latis, subcoriaceis, utrinque subaequa- liter angustatis, basi cuneatis, apice acutis vel leviter acumi- natis, supra subolivaceis, subtus glaucescentibus, utrinque pilis paucis albidis instructis, nervis utrinque 3 vel 4, tenuibus; flori- bus depauperato-fasciculatis, 3 cm longis, purpureis, pedicellis hirsutis; calycis lobis 5, ellipticis ad elliptico-ovatis, 3 mm longis, rotundatis, ciliatis; corolla glabra, vel deorsum utrinque puberula, lobis subaequalibus, 1.5 cm longis, tubo subaequantibus; staminibus 10, filamentis deorsum cinereo-pubescentibus, inae- qualibus; ovario dense piloso; capsulis ovoideis, 6 mm longis, valvis incrassatis, rectis, vetustioribus glabris. Hainan, Miss K. L. Schaeffer, January, 1921. On rocks along streams in the eastern part of the island. A species apparently belonging in the group with Rhododen- dron simsii Planch., strongly differentiated, however, by its very narrow leaves which are distinctly glaucous beneath. Ur als ise 21,4 Merrill: Diagnoses of Hainan Plants 351 MYRSINACEAE ARDISIA MACLUREI sp. nov. § Bladhia. Suffruticosa, haud ramosa, repens, caulis 4 ad 15 cm longis, radicans, rugosis, 2 mm diametro, deorsum glabris, sursum vil- losis; foliis alternis vel superioribus subverticillatis, oblongo- obovatis ad oblongo-ellipticis, olivaceis, 4 ad 6 cm longis, 2 ad 3 cm latis, obtusis, basi cordatis, irregulariter dentatis, utrinque pilis paucis articulatis elongatis instructis, costa utrinque dense subfurfuraceo-villosis, nervis utrinque 9 ad 12, cum reticulis dis- tinctis; petiolo dense villoso, 5 ad 10 mm longo; inflorescentiis subterminalibus, subumbellatis, villosis, pilis numerosis articu- latis patulis subfurfuraceis instructis, pedunculo circiter 4 cm longo; floribus paucis, plerumque 2 vel 3, bracteolis linearis ad anguste lanceolatis, 3 ad 5 mm longis, pedicellis 5 ad 7 mm longis, villosis ; sepalis anguste lanceolatis, 3 ad 4 mm longis, plus minusve ciliatis ; fructibus globosis, 6 mm diametro, glabris. Ng Chi Leng, McClure 8668, December 15, 1921. In forested ravines, altitude 1,200 meters. A very strongly marked species of the section Bladhia, dis- tinguished at once from all hitherto described forms by its cor- date leaves. SAPOTACEAE SIDEROXYLON ROSTRATUM sp. nov. | Arbor circiter 10 m alta, floribus exceptis ores: ramis ramulisque teretibus, tenuibus, ramis lenticellatis; foliis char- taceis, oblongo-ellipticis, 10 ad 15 cm longis, 4.5 ad 5 cm latis, olivaceis, nitidis, subtis pallidioribus, glaberrimis, utrinque sub- aequaliter angustatis, basi acutis, apice acuminatis, nervis utrinque 8 ad 10, distinctis, arcuato-anastomosantibus, reticulis laxis; petiolo 1 ad 1.5 cm longo; floribus junioribus ferrugineo- pubescentibus, paucis, axillaribus; fructibus ovoideis, circiter 3.5 cm longis, 2 cm diametro, minute subferrugineo-pubescentibus glabrescentibus, perspicue rostratis, sepalis kahit cir- citer 3 mm longis, subglabris. Ng Chi Leng, McClure 8559, December 20, 1921. In forested ravines along streams. A species well characterized by being entirely glabrous except - the flowers (buds) and fruits, and in its rostrate fruits. It resembles the Philippine Sideroxylon villamilii Merr. in its vege- tative characters but has very different fruits. 352 The Philippine Journal of Science 1922 EBENACEAE DIOSPYROS CARDIOPHYLLA sp. nov. Frutex vel arbor parva, ramulis et subtus foliis perspicue ferrugineo-villosis, ramis teretibus, glabris, ramulis atro-brun- neis; foliis brevissime petiolatis, subcoriaceis, olivaceis, nitidis, oblongo-ovatis ad oblongo-ellipticis, 8 ad 10 cm longis, 2.5 ad 4.5 cm latis, basi late rotundatis, distincte cordatis, apice acu- minatis, supra glabris, nervis distincte impressis, subtus per- spicue ferrugineo-villosis praesertim ad costa nervisque, nervis utrinque circiter 10, adscendentibus, perspicuis, curvato-anasto- mosantibus; petiolo 2 ad 4 mm longo; floribus axillaribus, soli- tariis, sessilibus, sepalis persistentibus, lanceolatis, acuminatis, 7 mm longis, chartaceis, extus dense ferrugineo-villosis ; fructi- bus ovoideis, brunneis, nitidis, parce villosis, 10 ad 12 mm longis, subacutis, 1-locellatis, 1-spermis, albumine aequabile. Neg Chi Leng, McClure 8349, December, 1921. On slopes, al- titude 650 meters. A species resembling in many characters Diospyros eriantha Champ. and manifestly allied to it. It is at once distinguishable by its longer, more-numerously nerved, cordate leaves which are conspicuously ferruginous-villous beneath. APOCYNACEAE WRIGHTIA HAINANENSIS sp. nov. Arbor parva, floribus exceptis glaberrima; foliis lanceolatis ad oblongo-lanceolatis vel oblongis, 8 ad 13 cm longis, 1.5 ad 4.5 em latis, caudato-acuminatis, basi acutis, chartaceis, oliva- ceis, nitidis, nervis utrinque 5 vel 6, curvato-adscendentibus, distinctis; cymis paucifloris, brevibus, floribus albidis; calycis segmentis orbiculari-ovatis, 1 mm longis, corollae tubo brevio- ribus ; petalis ellipticis ad elliptico-oblongis, 1 cm longis, utrinque puberulis, squamis coronae 25, filiformibus, glabris, 3 ad 4 mm longis; antheris lanceolatis, pubescentibus, 5 mm longis; folli- culis circiter 35 em longis, 7 mm diametro, lenticellatis. Ng Chi Leng, Kai Fong, and Fong Mok Szee, Hongkong herb. | 425 (type), 446, McClure 8685. In forested ravines. The first specimen cited, in flower, was distributed as a Raw- wolfia; the second, in fruit, as a Villaris. McClure’s specimen, in fruit, was from forested ravines, altitude 750 meters. The species is well characterized by being entirely glabrous, except for its puberulent flowers; by its elongated follicles; and by its corona being divided into 25 filiform, glabrous, subequal segments. It belongs in the group with Wrightia laevis Hook. f. " othe 2) 21,4 Merrill: Diagnoses of Hainan Plants 353 CONVOLVULACEAE ERYCIBE HAINANENSIS sp. nov. Frutex scandens, circiter 10 m altus, ramis teretibus, circiter 5 mm diametro, dense castaneo-villosis; foliis chartaceis ad subcoriaceis, ellipticis ad oblongo-ellipticis, 15 ad 18 cm longis, 6 ad 8 cm latis, perspicue acuminatis, basi obtusis ad subro- tundatis, utrinque subconcoloribus, pallide olivaceis, nitidis, supra glabris, puncticulatis, costa impressa, subtus ad costa dense ferrugineo-villosis ceteroquin leviter villosis, nervis utrin- que circiter 9, reticulis laxis, vix conspicuis; petiolo densissime villoso, circiter 1 cm longo; inflorescentiis subracemosis, axilla- ribus terminalibusque, densissime ferrugineo-villosis, usque ad 12 cm longis; fructibus ellipsoideis, glabris, circiter 2 cm longis, in siccitate nigris, sepalis persistentibus orbiculari-reniformibus, ferrugineo-villosis, 3 ad 3.5 mm diametro, margine ciliatis. Ng Chi Leng, McClure 8547, December 20, 1921. On forested slopes, altitude about 700 meters. This is not referable to any of the forms hitherto described from or accredited to China, and does not conform with the characters of any of the species of Indo-China. It probably belongs in the group with E'rycibe magnifica Prain of the Malay Peninsula, although it is very different from that species. BIGNONIACEAE RADERMACHERA HAINANENSIS sp. nov. Arbor glabra, ramis ramulisque griseis, rugosis; foliis pin- natis, 5-foliolatis, circiter 10 cm longis, foliolis chartaceis, oblongo-ovatis, 4 ad 6.5 cm longis, tenuiter acuminatis, utrinque minutissime puncticulatis, nervis tenuibus, utrinque 5 vel 6; inflorescentiis axillaribus, paucifloris, racemosis vel depauperato- paniculatis, quam foliis multo brevioribus; floribus flavidis 4 cm longis, calycis cylindraceis, obliquis, 1.8 cm longis, breviter 3-lobatis; corollae tubo extus glabro, deorsum (1.5 cm) angus- tato et intus villoso, sursum expanso, lobis late reniformi-ovatis, 10 mm latis; folliculis usque ad 40 cm longis, circiter 5 mm diametro, seminibus 12 mm longis. Hoihow, McClure 7648, October 12, 1921, in dry sandy waste places at low altitudes, with the local name hung fa. The species is also represented by Hongkong herbarium No. 453 collected in Hainan October 22, 1893, without locality. A species radically different from the few representatives of the genus known from China, strongly characterized by its short, 5-foliolate, simply pinnate leaves; its greatly abbreviated, few-flowered inflorescences; and its cylindric, oblique calyx tube. 189203-———2 354 The Philippine Journal of Science 1922 GESNERIACEAE “? OREOCHARIS FLAVIDA sp. nov. Herba perennis, acaulis, petiolis et subtus foliis densissime rufescente lanuginosis; foliis longe petiolatis, in siccitate char- taceis vel subcoriaceis, oblongo-ellipticis vel ellipticis, usque ad 16 cm longis et 7 cm latis, apice rotundatis, basi plerumque rotundatis, distincte inaequilateralibus, margine crenatis, supra hirsuta, nervis utrinque circiter 6, subobsoletis; petiolo usque ad 12 em longo; inflorescentiis folia subaequantibus, pedunculo deorsum rufescente lanuginoso, supra subglabro; floribus sub- umbellatis, flavidis, circiter 2 cm longis, pedicellis tenuibus, circiter 2 cm longis; sepalis lanceolatis, acuminatis, subliberis, 5 mm longis, ciliato-pilosis; corolla subcylindrica, extus glabra, intus leviter pubescens, subaequaliter 5-lobata, lobis suborbicu- laris, 6.5 mm longis, leviter ciliatis; filamentis glabris, 6 mm longis, antheris horizontaliter dehiscentibus, subhippocrepifor- mibus; capsulis junioribus linearis, glabris, 2 cm longis. Ng Chi Leng, McClure 8598, December 16, 1921. On damp rocks on forested slopes, altitude 1,400 meters. This species conforms to the characters of Oreocharis in habit, appearance, and all other characters except in its some- what horseshoe-shaped, horizontally dehiscent anthers, a char- acter that might lead some to erect a new genus for this par- ticular species. The dense, woolly indumentum entirely covers the lower surfaces of the leaves. ACANTHACEAE | (0% STROBILANTHES MACLUREI sp. nov. Herba erecta, usque ad 0.5 m alta, plus minusve hirsuta, caulis ut videtur deorsum procumbens, glabris, 2 mm diametro, ramis tenuibus, leviter hirsutis; foliis in paribus inaequimagnis, oblongo-ovatis, acuminatis, basi acutis, crenatis, olivaceis, supra cystolithis numerosis instructis, utrinque parce hirsutis, subtus pallidioribus, nervis utrinque 4 vel 5, perspicuis, majoribus usque ad 9 cm longis et 3.5 cm latis, petiolo 1 ad 2.5 cm longo; inflorescentiis subcapitatis, terminalibus et in axillis superior- ibus, paucifloris, bracteis obovatis ad oblongo-obovatis, usque ad 1 cm longis, foliaceis, persistentibus, perspicue hirsutis; floribus subcaeruleis, 3 cm longis, sepalis anguste lanceolatis, acuminatis, 6 mm longis, perspicue albido-ciliatis, corollae tubo deorsum angusto, glabro, sursum ampliato, lobis brevibus; ovario glabro, apice barbato; antheris 2.5 mm longis. Ng Chi Leng, McClure 8578 (type), 8480, December 21, 1921. In forested ravines, altitude 700 to 800 meters. 21,4 Merrill: Diagnoses of Hainan Plants 355 A species apparently most closely allied to Strobilanthes debilis Hemsl. from which it is distinguished by its larger crenate leaves. ; RUBIACEAE LASIANTHUS HAINANENSIS sp. nov. Frutex 2 ad 3 m altus, petiolis junioribus stipulisque parcis- sime pubescentibus exceptis glaber; ramis teretibus, laevis, circiter 3 mm diametro, internodiis 6 ad 8 cm longis; foliis oblongis, olivaceis, nitidis, chartaceis, 12 ad 16 cm longis, 3.5 ad 5 cm latis, -utrinque angustatis, basi acutis, apice tenuiter acuminatis, nervis primariis utrinque circiter 15, distinctis, rectis, anastomosantibus, nervis marginalis leviter arcuatis; petiolo 10 ad 12 mm longo; stipulis oblongis, pubescentibus, acuminatis, 2.5 mm longis; fructibus axillaribus, paucis, obo- voideis ad oblongo-obovoideis, sessilibus, 12 mm longis. Ng Chi Leng, McClure 8569, December 3, 1921. In forested ravines, altitude about 1,400 meters. quay In vegetative characters this species is very similar to Lasian- thus laevigatus Blume from which it is at once distinguished by its solitary or fascicled sessile fruits. COMPOSITAE GYNURA MACLUREI sp. nov. Herba erecta, subglabra (partibus junioribus parce furfuraceo- — pubescentibus) ; foliis mebranaceis, olivaceis, ovatis ad oblongo- ovatis, 7 ad 11 cm longis, 3 ad 7 cm latis, glabris, olivaceis, subtus pallidioribus, margine grosse irregulariter repando-den- tatis, dentibus acutis ad apiculatis, apice acutis ad breviter acuminatis, basi truncato-rotundatis, leviter inaequilateralibus, nervis utrinque 5 vel 6, vix perspicuis; petiolo 2.5 ad 4 cm longo; stipulis semihastatis, 7 mm longis, grosse lobato-den- tatis; inflorescentiis pedunculatis, terminalibus, 15 ad 17 cm longis, subpaniculatis; capitulis 1.3 ad 1.5 cm longis, multifloris, bracteolis exterioribus linearis, 2 ad 3 mm longis, interioribus 10 ad 12 mm longis, 2.5 mm latis, glabris, tenuiter acuminatis. Ng Chi Leng, McClure 8588 (type), 8626, December 21, 1921. On rocks in forests and on burned-over slopes, altitude 600 to 750 meters. Apparently well characterized by its glabrous, coarsely re- pand-toothed, ovate to oblong-ovate leaves, and its semihastate auriculate stipules or stipulelike appendages. The ultimate re- ticulations are not elongated, and contain numerous free veinlets. THE EFFECT OF SULPHUR COMPOUNDS ON CEMENT . By J. C. Witr Formerly of the Bureau of Science, Manila Some papers on the effect of sodium sulphide and the effect of calcium sulphate? on cement were published by me several TABLE 1.—The effect of sodium sulphide on tensile strength. [Pounds per square inch. Each value is the average of six tests.] NEAT. Sulphur, : a g s 3 2 = g z Fd Cement. grams per| @ Si 8i3j,21% Fy E $ | § ect wii ei a1 Si 2) & | el 2 A (a) 353 | 595 | 685/ 673| 688| 711] 688| 662! 576 | 650 A 0.097 | 307] 586 | 642] 606] 594| 696] 757/| 610! 530| 528 A 0.97 | 300] 506/ 548| 542] 635] 660] 6es| 661| 571 | 675 Pe Se 4.80 | 155] 300] 573| 560] 555] 615| 617| 557] 305 | 527 A. P 9.70 | 190| 268| 391] 406] 477| 488] 528/ 530/| 528 | 445 B- (a) 837} 617| 651| 649| 586] 621| 636| 591| 411 | 457 B 0.097 | 411} 622/ 611] 618} 605| 573| 573| 597| 501 | 565 B 0.97 | 388] 579/ 599] 581| 547] 601] 580! 567| 475 | 587 8S ee 4.80 | 340] 475) 567| 569| 564] 542] 543| 466| 454 | 454 . 2 Se eae 9.70 | 262] 460} 495] 526] 417| 457| 446| 475] 365 | 397 c ‘ (a) 330 | 586| 679| 644| 686] 680| 695| 644| 558 | 653 _ AN aan s 0.097 | 412| 577. 666 | 652] 681| 608| 613} 554 | 675 c a ee 0.97 | 407| 568| 605| 676+ 563| 652/| 602! 595| 565 | 657 C.. 4.80 | 355| 461| 452| 503| 601| 625] 630| 622| 472 | 590 SERS ie Pe ee 9.70 | 841] 438) 410] 400| 582] 572| 597| 552| B40 | 42 tava: ais: (a) 848 | 633 | 722] 677] 710] 695] 692| 617| 529! 546 et Ae Sa 0.097 | 835| 687| 726] 685| 716| 725| 698! 702| 670 | 630 PO Rapa de saan 0.97 | 308) 564| 655) 642| 712! 788) 717} 642| 560 | 648 Wi 8 ge 4.80 | 214] 467/ 630) 581| 631| 599| 717| 616| 598 | 577 Bis gs hear eae, 9.70 | 255] 553| 584] 584] 536| 595] 571| 563] 502 | 446 | Se, (3) 888 | 628 731} 720| 706| 677| 706| 685! 559 | 597 S22 Sea 0.097 | 334| 638/ 736| 676] 721| 686| 738| 649| 647 | 698 Geos ee an 0.97 | 840| 609} 645 | 604| 666| 665/ 636| 642/| 564 | 592 E., 4.80 | 222] 456/ 588| 589] 633] 630] 518| 576/ 515/550” Ee 9.70 | 268/ 519| 589| 516] 506| 489] 517! 5e0| 506 | 447 : _@ Water. * The effect of sulphide on cement, Philip. Journ. Sci. § A 11 (1916) 273. The work suggested a later paper: The action of sodium sulphide on ferric oxide, Journ. Am. Chem. Soc. 43 (1921) 7384. * The effect of calcium sulphate on cement, first paper, Philip. Journ. Sci. § A 12 (1917) 183; second paper, Philip. Journ. Sci. 14 (1919) 221; Concrete 16 (1920) 82, 357 358 The Philippine Journal of Science 1922 TABLE 1.—The effect of sodium sulphide on tensile strength—Continued. 1:3 STANDARD OTTAWA SAND. a2 n aL a Pa ia] n mn a Cement. - ome » b a a 3S % z z g g ries HESS PETRA 2S A pitisties eee (a) ..----| 238 |, 883 | 860] 878| 899| 383) 882] 420] 881 G asebechi eget aeeeye ern s 0007 fc. 209 | 292 | 299) 356 | 337] 383) 323) 390 | 349 7 Yeohixgae ease eer ese: OVE pss. - 174 | 287| 320| 833) 346) 331) 365) 427 | 376 Bo weuckcus Sea esewees 4. fessm8 234 | 240} 288 | 259| 259} 268] 305| 876 | 885 eS $270: {2 124| 124| 258| 204] 246) 243| 256| 311 | 297 B > C5 ee eee 248 | 308] 3825| 319| 323} 284] 380] 382 /| 329 p ; Pee etewees yet alrueee et 0007 |xa.--. 233 | 280} 399} 309| 305) 304| 320/ 346 | 315 b: en arenes ae, Oe OcOT c[Sentee! 280 | 258 | 382 | 3805| 304| 309] 323| 871 | 805 Be ee 400 to 174| 252 | 290) 279| 277) 263] 292) 389) 311 4, Gener cee aNRaen so 145 | 218 | 259| 266| 253 | 257| 257) 824 | 296 C..-. ----------------=" (@). |------ 282 | 380 | 428| 402} 450) 394) 425| 451 | 417 OG. So a 0.097 |------ 253 | 357] 3818) 401| 482| 452 | 3842) 361 | 425 |} jae Gene eer ens 0.97 248 | 3805 | 290] 410| 393] 3873] 402] 878 | 884 Cc We: 4.80 [asks 174| 155 |. 205] 855 | 306] 889| 845| 879 | 344 ny RE eee ee ce eS ae ‘6 bse 168 | 251|. 157| 320| 339| 288} 383] 374 | 340 D 22222-2222 520-552=4 Qo fees 294} 409} 441 | 402) 418 | 399| 886 | 420 | 374 Dit Ais. sik: Re 0.097 |------ 239 | 843 883 | 881] 411| 896| 394] 425 | 393 AS ore eS DOT p.cse: 240} 829} 348] 392] 408} 364| 405| 450 | 340 L2G 1 4.80 |----- _ 162 | 237} 288} 319| 320| 827| 3823| 370 | 349 1 i ea ey ee eg ee a PATG. joie. » 144 226 265 | 299; 292; 316) 350| 315 | 317 B.S ae O55 hee. 331 | 407| 414) 402| 488 367) 842 |e 378 | 375 BP al 05007 |-2e8.- 265 819} 844] 361] 368| 825| 339] 448 | 342 Bis Sh O9F LAL: 874 | 304| 851] 359] 360| 885 | 385] 368 | 889 WS seek a.00 |. 38s. 179 | 279| 267); 319} 317] 287} 307| 364 | 316 a DO Lan. 157 | 227| 249| 271| 260} 261} 265| 9818 | 268 - r 8 Water. years ago. In connection with the first two papers, tensile- strength specimens were stored in water, to be broken at various periods up to five years. All of these specimens have now reached maturity and the results are given in Tables 1 and 2. The first two papers contained such details as the chemical analyses and physical tests of the cements, the amounts of sul- phide or sulphate added, and a discussion of the results obtained up to that time. Consequently this material will not be repeated ‘here. : | The five-year periods in Table 1 show some interesting results. As shown by the early tests, concentrations of sodium sulphide not exceeding 1 gram of sulphur (as sulphide) per liter do not seriously affect the tensile strength; in fact, in most cases the strength is increased. With the higher concentration, however, the strength is lowered in every case. The maximum loss is 82 per cent with the neat and 37 per cent with the mortar. \ 21,4 Witt: Effect of Sulphur Compounds on Cement 359 In the first paper it was pointed out that the percentage loss in tensile strength of the neat specimens, due to the highest concentration of sulphide, varied with the iron content of the cement. In general the same relationship persists after five years, as is shown in Table 3, though Cement B is an exception. TABLE 2.—The effect of calcium sulphate on tensile strength. [Pounds per square inch. Each value is the result of six tests.] NEAT. Sul- Neat. 1:3 standard Ottawa sand. anny: : 5 an n n n Gee | Goo | eae) aie et} ere eT Se el 4 ince- | @ a i} S $ $ aj S S $ $ mente | (3° Ioan 5 B > Ae ee & 5 > > Ld Nn ~* © ri fon} L ed Nn oO wo nm oO P. ct Aes 1.08 | 527] 680] 743] 648] 755| 715 | 226] 285| 877] 385} 415 | 400 we Si 1.52 | 578 | 648| 788| 707| 766| 774| 216 | 288| 364] 411| 423/ 447 ge eer 2.19 | 507| 631| 743| 769| 730| 580| 188| 309| 390] 413 | 447 /| 488 an eed 2.81| 418| 623| 690| 778| 720| 687! 142| 250| 345| 375| 400| 417 yas gee an 9.18 | 465] 593 | 697| 684| 749] 662| 142) 197] 226|) 201} 195 | 308 Boece 1.14] 569] 612} 600} 594] 548| 4571 218/| 300| 333] 308| 326 | 379 Bo cccesereures 1.59 | 535] 580| 584] 640] 606] 572/| 206] 282] 281] 281 302/| 850 B ......----..| 1.89 | 553} 618| 685| 619| 576| 456, 268] 347/ 358| 828| 366 | 871 Be 4.55 | 509] 584| 613} 610| 541/ 568] 136 $85 | 856 | 427 | 437 B kssees 9.92 | 528| 602} 648| 677| 666| 693} 143] 187| 170| 150| 160 | 208 Cac eal 384 |: B86. | STO once 680 | 695 |_.____ co Be 1 ae 450 | 394 C....-.-------| 2.10} 586 | 650| 636| 650| 649| 570| 241) 351/| 409 | 398] 423 | 413 O22 8.02 | 495 | 620| 687] 656] 684] 503| 218] 346| 426] 433] 501] 416 Dor Scr ane 4.86 | 416] 569| 711} 623] 592] 665/| 128] 187] 182] 333] 414 | 484 2 eon peace 9.54] 898) 491| 646] 688) 576| 555| 112] 170] 244| 164/ 171 | 192 De a LB4i G88 | 728 [2 695 | 692 |__._.. 294} 409 |__.... 418 | 899 |__._. De: 2.14} 665] 629| 661| 649} 650| 577| 267/| 387| 480| 3890] 386 | 351 5? Semana 8.380 | 648] 678 | 783} 693 | 697| 563] 261| 399] 469] 436]! 470 | 339 jt Leama, 5.00] 455] 672] 628| 716] 686| 553/| 161] 282/ 470| 456 | 473 | 368 Dia a 10.20| 425] 658 | 644] 696/ 596| 474] 149] 218] 229] 187/] 156 | 113 yc rears Sree 10h) OS 1 EL i] OT 1 6 [cs 831 | 407 438 | 867 p PEDERI S 2.05 | 681/ 664] 639] 655| 684| 458| 276] 3884] 899| 3859] 403 | 401 Fea! 2.73 | 692| 682] 751| 615| 618| 618| 829| 875 | 443| 395| 441 | 415 y patnlese rr 4.79 | 465| 531] 662] 695| 691| 640} 184| 408] 512] 462) 505 | 462 tA A nen 9.58 | 413] 517| 599| 686) 540] (*) | 187] 162} 154) 118| 98} (*) b BEE EE aN Se 1.46 | 650| 628] 603] 704) 645| 655| 217| 816) 361| 3878] 388 | 347 | goon ee 2.00| 768/ 814/ 831| 840| 760) 818] 181| 268| 317| 368| 854 | 311 fo 8.29| 674/ 713| 785| 781| 776| 695| 233] 330] 389| 359] 369} 382 b G Seqinr neat 5.00 | 421| 648] 623] 660] 698| 758| 194] 331 | 407] 444| 465 | 406 rat agree Paes 9.11 | 385] 489| 654| 591] 611] 578] 157| 317| 4380 404 | 314 « These specimens were swollen, cracked, and partly disintegrated, and could not be tested. The identification marks on five-year briquettes of the calcium sulphate series were illegible when the specimens reached maturity, so the three-year results are the last that can be considered. Table 2 shows that, for the three-year period, the presence of as much as 5 per cent of sulphuric anhydride did 360 The Philippine Journal of Science not cause any serious loss in strength. In some instances there was an increase. The highest percentage of sulphuric anhy- dride, however, caused a decided reduction in strength. The three-year briquettes made with Cement E were cracked and partly disintegrated and could not be tested. TABLE 3.—Variation of percentage loss in tensile strength with the iron content, Tensile strength of five-year neat bri- ° coaites, Saeees Ferric el Corre- evide sponding Cement. (Fez f Loss. loss for Os) Sodium sul- early m phide solu- period, Water.| tion con- taining 9. g. per liter. Lbs.per; Lbs. per \Lbs. per P. ct. | sq.in. sq. in. sq.in. | P. et. P, ct. Ba ee sa 1,12 546 446 100} 18,31 18. 56 Be i eho ee ra, 1,34 457 897 60} 18.18 24.07 WO ase csttnns BiG wecSesnk ase 1.38 597 447 150} 25.13 26.12 Peete 5 Sake. ibe hed 659 445 214| 32.47 47,29 THE USE OF TEXTILE FIBERS IN MICROSCOPIC. QUALITATIVE CHEMICAL ANALYSIS Vv. THE DETECTION OF GOLD BY MEANS OF STANNOUS CHLORIDE-PYROGALLOL VISCOSE-SILK FIBERS * By Howarp Irvine Cote Chemist, Bureau of Science, Manila The reduction of a solution of gold chloride by a mixture of stannic and stannous chlorides leads to the formation of a red or violet color in the gold solution. This color is due to the precipitation of finely divided gold on the stannic hydroxide. It is called purple of Cassius because A. Cassius wrote a pam- phlet entitled De Auro, describing its preparation, in 1685. In 1904 Donau ’ applied this color reaction as a test for gold. He prepared silk fibers in the following manner: Raw-silk fibers were purified by treatment with 10 per cent sodium or potassium hydroxide, for from three to four hours, and then thoroughly washed. The fibers were treated with a dilute solution of stannous chloride and pyrogallol, washed several times with water, and dried between filter paper. Donau states that silk fibers thus prepared last two or three days, but if exposed longer to the air they become dark colored in the gold solution. He made the test for gold as follows: The gold was dissolved in aqua regia, and the solution was evaporated to dryness and taken up in water. A drop of this solution was placed on a slide and a fiber, treated as described above, was dipped vertic- ally into the drop from a tiny wax cone. The fiber was then examined under the microscope. Gold turned the fiber red or red-violet. In continuing the study of the use of textile fibers in micro- scopic qualitative analysis I have endeavored to make a test fiber for gold that would remain stable indefinitely and still be sen- sitive to the presence of a minute amount of gold. The prep- *For Parts I to IV see Journ. Ind. and Eng. Chem. 9 (1917) 969; 10 (1918) 48. : *Donau, J., Monatsh, 25 (1904) 545. 861 362 The Philippine Journal of Science 1922 aration of such impregnated fibers involved a study of the kind of fiber to be employed, the concentration of the solutions to be used, the length of time of dyeing, and the temperature of the bath. The fibers tested were wool, cotton, flax, silk, and viscose silk. The fibers were impregnated with a mixture of stannous chloride and pyrogallol. In a test solution of gold, the prepared fibers gave the following colors: Wool, blue to gray; cotton, pink; flax, silk, and viscose silk, dark red to violet. Irrespective of the color produced, viscose silk is the fiber best suited for use in microscopic qualitative analysis. The follow- ing tests were therefore made on viscose silk: Fibers were dyed for one, ten, and twenty minutes in cold and in boiling solutions of stannous chloride and pyrogallol. The concentra- tion of each of the two chemicals was varied from 1 per cent to. 20 per cent. The best results were obtained in a solution of 10 per cent stannous chloride and 10 per cent pyrogallol. The effect of the addition of hydrochloric acid to the bath was tested and is shown in Table 1. TABLE 1.—Color of impregnated viscose-silk fibers in gold solution prepared with varying concentrations of stannous chloride, pyrogallol, and hydrochloric acid. Concentration of chemical. a Time of Hydrochlo- prcigenorie OE SA root oesets. | Stpeaee terivanl|- hen, | ee ted. Per cent. | Per cent. | Per cent. | Minutes. 0 5 10 10 | Pale red. 1 10 0.1 10 | Very pale red. 1 10 1 10 | Red. 1 10 10 10 | Good red. 5 10 10 10 | Very good red, Weakened fi- 10 10 10 10 ber. 0 10 10 10 | Color too light. 5 10 20 10 | Good red. A bath containing over 5 per cent concentrated hydrochloric acid or 10 per cent stannous chloride tended to weaken the fiber. Immersion for more than ten minutes in 10 per cent stannous chloride solution seriously weakened the fiber. Stannous chloride made by dissolving 10 grams of tin in 100 cubic centimeters concentrated hydrochloric acid and adding 10 ‘per cent pyrogallol gave a fairly good fiber when the silk was allowed to remain in the solution only one minute. Two minutes 21,4 Cole: Detection of Gold 363 weakened the fiber so that it could not be handled without break- ing. In the experiments discussed in this paper, when stannous chloride solution is mentioned it means that a filtered solution of stannous chloride crystals dissolved in water was used. Viscose silk boiled for ten minutes in 0.5 per cent, 1 per cent, and 10 per cent solutions of tannic acid gave a faint blue color- ation in gold solution. Fibers treated with a solution of stannous chloride and quinol (hydroquinone) gave almost no color with gold solution. Fibers treated with stannous chloride and resorcinol gave a pink coloration with gold solution. All the fibers prepared as above mentioned were left exposed to the air and light for six months and tested from time to time. No marked decrease in sensitivity was noted even after six months’ exposure to air. It was noted, however, that the fiber after several days’ exposure to air turned dark blue in concentrated gold solution, but red to purple in dilute solution. Freshly prepared fibers gave red to purple in either dilute or - concentrated gold solution. The final method of preparation of the test fibers was as follows: Ten grams of stannous chloride crystals were dissolved in 95 cubic centimeters distilled water and 5 cubic centimeters concentrated hydrochloric acid and filtered; 10 grams of pyro- gallol were added to the filtrate. The viscose-silk fibers were heated in this solution for ten minutes on the water bath. The fibers were then removed and washed well with water and dried between filter paper. The test for gold with the stannous chloride-pyrogallol fiber is best carried out as follows: Place upon an object slide a drop of the solution (neutral) to be tested. Insert into it a fiber 5 millimeters long for only part of its length. Move the preparation on the stage of the microscope until a portion of the fiber both in and out of the drop falls within the field of vision. Note the change of color if any. Let the drop evapo- rate to dryness spontaneously. A red to purple color denotes gold. Strong solutions of gold immediately turn freshly prepared fibers dark red, and older fibers dark blue; weaker solutions give a red or blue color in the fiber only upon evaporation of the drop to dryness. Alkali prevents the formation of the color. Mineral acids retard the formation of the color. Reducing and oxidizing agents in general interfere. Ferrous and ferric salts, 364 The Philippine Journal of Science potassium ferro- and ferricyanides interfere. Chromates and bichromates tend to give a brown tone to the red color. Silver nitrate turns the fiber yellow to brown even in the ab- sence of gold; but, unless silver is present in excessive amounts, it does not mask the red color produced when gold is present. Platinum salts do not color the fiber. Permanganates color the fiber brown. Ammonium molybdate and phosphomolybdate turn the fiber a light blue color, readily distinguishable from the dark blue or bluish red obtained when gold is present. None of the other basic or acid radicles interfere. In order to express numerically the sensitivity of the fibers to gold solutions the following method was adopted: Solutions of decreasing concentration of gold were used. A drop of constant size was obtained by means of a No. 29 platinum wire in the form of a loop 1 millimeter in diameter. The drop de- livered by this loop contained 0.00045 cubic centimeter. Start- ing with a solution of gold producing a decided color change in the fiber when the test drop is evaporated to dryness, solu- tions of gold of decreasing concentrations were tested until a concentration was reached which failed to give a color in the fiber. The most-dilute solution, one drop of which will produce a color change in the fiber upon evaporation of the test drop, gives the degree of sensitivity. This limit may be expressed either in terms of grams per liter or as the absolute amount contained in the drop used. The most-dilute solution which gave a good color change in the fiber in a drop containing 0.00045 cubic centimeter was one containing 0.05 gram gold per liter. Therefore, the absolute amount of gold detectable by this method is 0.000,022 milligram, or 0.022 micromilligram (0.001 milligram — 1 micromilligram). This test for gold is applicable not only in cases where the ordinary tests are used but also when the latter cannot be used; that is, in exceedingly minute amounts of material or upon subjects which must not be disfigured in the testing. SUMMARY 1. Viscose-silk fibers dyed with stannous chloride and pyro- gallol give a very sensitive microscopic method for the detection of gold. 2. The test fibers, prepared as described, are stable. No de- crease in sensitivity was noted at the end of six months. 3. A drop of solution containing 0.000,022 milligram of gold gives a positive test for gold by this method. SOME GENERALIZATIONS ON THE INFLUENCE OF SUBSTANCES ON CEMENT AND CONCRETE SECOND PAPER By J. C. Witt Formerly of the Bureau of Science, Manila This paper is one of a series published as a result of re- search started by me at the Bureau of Science in 1915 to in- vestigate the effect of some inorganic substances on cement and concrete.t. The work was confined to neat and mortar specimens with the idea of later trying the effect of the most-active sub- stances on concrete. Since sulphur compounds were frequently mentioned in the literature as destructive agents, the work was begun with these—first sulphides and then sulphates. The ef- fect of sulphides was found to be related to the iron content of the cement, and this led to an investigation of the reaction between sodium sulphide and ferric oxide.2. Also another prob- lem related to the effect of inorganic salts was taken up; namely, the solubility of cement in water. The work with sulphides and sulphates led to a study of the effect of a number of other salts ¢ on the setting time and tensile strength, and it is chiefly with this phase of the work that the present paper is concerned. A review of the literature revealed that while many papers had been published little definite infor- mation was available, and the subject as a whole showed little progress. No uniform method in making tests had been fol- lowed by the various investigators, and consequently there was no basis for comparing their results. Solutions were made from eleven salts (four concentrations of each) and these solutions were used in making setting-time _tests, and 1:3 standard Ottawa sand briquettes with four cements. The briquettes were stored in water and broken at the end of 7 days, 28 days, 180 days, 1 year, and 5 years. Each * Philip. Journ. Sci. § A'11 (1916) 278; 12 (1917) 183; 14 (1919) 221; 21 (1922) 357. * Journ. Am. Chem. Soe. 43 (1921) 734. * Philip. Journ. Sci. § A 13 (1918) 147; Concrete 16 (1920) 13. * Philip. Journ. Sci. § A 13 (1918) 29. This will be called the “first paper” in the discussion that follows. 365 366 The Philippine Journal of Science 1922 TABLE 1.—The effect on tensile strength. [Pounds per square inch. All briquettes are 1:3 standard Ottawa sand. Each value is the average of three tests.] Cement I. Cement II. Salt and normal- apg osereee: 7 days. |28 days. aoe 1 year. |5 years.| 7 days. |28 days. Pott l year. 5 years. a c ; ‘Water... 2. 267 359 365 366 310 280 351 373 372 820 Sodium chlo- : ride: 6635. 214 264 246 268 825 202 288 269 255 278 0.40 se 205 296 335 883 333 210 292 316 821 292 POS oss 248 880|- 401 873 402 275 807 379 381 362 Los: 267 387 421 360 317 285 295 359 860 396 Zine chloride: Ob oe 235 275 295 263 259 210 267 282 269 248 Oi os 212 230 329 335 330 252 827 344 835 311 0:00.25 263 393 467 449 367 284 396 473 501 404 js) Noe oapeenee 244 344 476 472 887 250 333 430 497 355 Copper chlo- ride: 0:06... 218 294 831 265 275 183 254 270 240 248 E205 ees 221 326 308 333 827 203 262 372 847 338 oo) Bae Sigslen 140 870 450 467 | 446 153 861 418 352 331 a IE 1 aoe eae 176 319 474 465 360 244 333 407 465 326 Sodium ni- trate: 6... 206 290 316 269 278 211 238 272 278 290 (8 Ut Sipe a at 195 298 310 316 310 231 301 336 308 312 ON 272 401 487 406 374 233, 307 362 357 340 1002.2. 187 313 886 359 306 196 290 339 325 288 Potassium ni- trate: OORwiscks 209 261 283 276 296 190 255 273 266} 290 OWaascaecd 215 280 364 815 350 227 326 B47 348} 884 Lg 1 Speer 197 266 361 366 352 208 293 338 348 319 Oe 217 300 337 340 310 207 289 315 304 280 Ammonium nitrate: C£..c. 221 288 290 287 320 241 268 287 259 315 OMe ss 249 306 348 312 839 242 | § 288 313 838 298 6 233 318 415 392 382 299 365 418 846 872 pA | | ree ee ee 404 439 411 369 292 390 387 399 855 Sodium sul- phate: : G0 210 277 288 291 307 223 286 827 292 330 O20.2 es 239 316 337 355 324 239 286 340 331 345 Usa 200 308 870 334 325 246 355 364 867 866 BO 306 385 449 408 330 300 356 421 874 295 Zine sulphate: OG occcnu 186 266 288 256 262 258 323 313 295 333 ae U Reser 251 318 358 344 326 231 811 859 331 876 Osc 809; 430} 469/ 495| 388/ 283] s66| 4751 505| 380 = A | ea ee 107 385 524 462 873 263 386 466 548 363 21,4 Witt: Cement and Concrete 367 TABLE 1.—The effect on tensile strength—Continued. Cement I. Cement II. Salt and normal- ity of solution. 180 180 7 days. |28 days. days. l year. |5 years.|7 days. |28 days. days. 1 year. |5 years. Copper _ sul- phate: 006i. 252 830 851 852 347 250 845 361 820 | 844 iy || eater 205 829 858 880 826 219 295 308 321 336 0: BO sca 108 866 495 413 406 191 340 461 448 393 ES | t Saeen 56 280 413 449 358 156 362 394 363 850 Sodium bicar- bonate: 1g ena 282 839 365 320 828 242 262 276 257 335 as || emenere ere 191 287 848 343 834 270 287 310 854 318 20} earnest 201 815 406 398 805 224 811 370 357 339 LOO. ada. 254 335 415 485 330 251 328 323 314 265 Potassium bi- carbonate: Occ 234 331 357 814 838 226 249 288 235 374 Pee a | Pecoraro 197 270 330 320 386 243 321 358 339 347 0. BO see coe 253 869 421 364 827 243 803 394 355 | 875 pO, caer or 272 343} 3884 366 308 234 308 - 846 351 398 Cement III. Cement IV. |Salt and normal- ity of solution. 7 180 180 days. |28 days. days. 1 year. |5 years.| 7 days. saere dave. 1 year. |5 years. Water 321 389 418 459 353 218 297 373 877 351 Sodium chlo- ride: 0. 05. 268 830 343 330 320 182 264 823 325 | 831 O10. 208 290 355 317 321 153 218 313 331 297 2 | Eee 233 808 349 347 321 149 290 324 819 | 3806 LO iccad. 803 878 378 334 270 197 236 324 348 308 Zine chloride: Obes 265 345 372 821 328 196 265 356 327 355 O.1Geicc 278 821 420 408 876 | 156 249 315 386 284 US | Seana are 293 397 428 427 393 231 294 417 388 386 sees F Seranieic 391 368 365 341 295 197 307 486 427 309 Copper chlo- ride: 0 GB ses. ick 216 390 352 322 317 170 203 365 329} 348 OS ce 230 319 409 399 845 141 230 321 339 | 344 | Sener 249 348 450 470 875 99 226 367 408 290 p A: Saree 247 403 453 425 317 185 284 895 438 | 325 Sodium ni- trate: DW esc ce 271 352 854 826; 305 166 232 342 819 | 295 5 Peeing 213 339 877 375 377 130 176 807 829 | 320 De ekas aan 194 302 340 833 {> 250 107 168 291 303 302 Pe eee 179 299 874 825 294 132 232 336 | 333! 290 368 The Philippine Journal of Science 1922 TABLE 1.—The effect on tensile strength—Continued. Cement III. Cement IV. Salt and normal- ity of solution. 7 days. (28 days. Gon: 1 year. |5 years.|7 days. |28 days. Pa 1 year. |5 years. | Potassium ni- trate: Gakic 280 337 341 335 302 166 298 805 346 0: 305545552 235 336 363 345 327 136 182 291| 318 | 325 0; bbe 168 245 827 833 301 116 179 309 836 | 350 LU 249 310 316 361 302 139 203 319 845 300 Ammonium nitrate: 0.002225... 267 329 326 314 313 169 247 345 831 | 319 Ua | cone sey 224 325 325 861 348 133 195 298 807 327 O:S0s2-c oo 208 340 373 389 275 127 199 334 356 322 a | Ean ane 272 330 469 289 250 171 268 366 409 325 Sodium _ sul- phate: 0,08 2s.c23- 302 410 487 328 265 179 228 321 825; 312 0.1022... 3. 266 824 837 350 347 150 220 316 845 | 395 Lg 0 | ih een 184 281 370 893 307 140 205 845 333 312 fH Senin * 282 345 430 338 290 223 323 396 889 370 Zinc suphate C06 263 385 392 311 323 201 262 341 353 359 O10 267 834 370 884 323 183 248 349 818} 312 8 eee 250 365 442 441 311 226 338 417 408 | 400 2.00 eke 800 413 452 412 825 272 367 419 538 375 Copper _ sul- phate: Oe s 2c, 277 366 373 838 815 182 281 361 355 368 ey Us Dome rnees, 252 352 351 378 339 165 235 821 403 321 0,503... c 168 : 311 393 898 340 120 284 391 418 364 ROO ees 258 423 526 366 230 185 316 400 461 | 310 Sodium bicar- bonate: Descccc. 278 342 349 304 296 171 246 839 883 317 oe SEN 221 308 327 829 298 147 227 845 335 824 O.B0sen 5: 173 264 368 338 274 195 147 283 280 285 LUO aca, 226 307 365 345 286 185 234 247 336 300 Potassium bi- carbonate: US Sapien: 280; 363) 3870} 816; 307| 207! 236] 351] 9348| 365 0.10........| 251] 382} 365/ 367, 295! 165| 216| 339| 319! 336 0. B0up cd 188} 256| 3683} 352] 276 98} 109; 274, 825] 374 Ls 246 809 | 360 869 295 152 247 277 812 | 260 test was made in triplicate, 2,700 briquettes in all. The first paper included the results up to 180 days. The remaining specimens have now been broken, and all the results, including those previously published, are given in Table 1. A list of the salts and the concentrations of the solutions as shown in Table 4 of the first paper is repeated here as Table 2. 21,4 Witt: Cement and Concrete 369 TABLE 2.—Data on solutions employed. "aie Solution ado Salt. nar ages centi- | for nor- f - 7 | by anal-| 2! om-| Of salt, | tiverad-| tee eens ysis), |*!8tency.* "| teal. ical. dg. cm, Sodium chloride .....-.-------.-.---- 0.05 | 0.0029 22) 0.0640| 0.0252 0.0388 iO ee re 0.10} 0.0058 23 0.1889/ 0.0527] 0.0812 De: ik: 0.50} 0.0291 23} 0.6695 0.2684] 0.4061 [1 ES EES er ee Cee eae we 1.00} 0.0582 22| 1.2810) 0.5040| 0.7770 __ ee a eae ace alas 0.05 | 0.0084 22} 0.0753) 0.0361 | 0.0892 De 0.10} 0.0068 23| 0.1575 | 0.0756 | 0.0819 Weg! i oceris alia cre, 0.50} 0.0342 23| 0.7874/ 0.3779 | 0.4095 , 3 Eales ale a a 1.00} 0.0685 22| 1.5065|/ 0.7226| 0.7839 Copper chloride 2.2.02 202 5 0.05 0. 0084 22 0.0741 0.0851 | 0.0890 Die icstac li: domino ane 0.10; 0.0067 23} 0.1549 0.0782 |" 0.0817 Do ccidoawecds pete s 0.50} 0.0387 23} 0.7745| 0.36683 | 0.4082 Do. soon hemne SRT 1.00! 0.0674 22| 1.4824/ 0.7011] 0.7818 Sodium nitrate 2220. ee 0. 05 0.0043 22 0. 0937 0.0258 | 0.0684 js MEMES Gee Soe E RT a oe) 0.10 | 0.0085 23 | 0.1959| 0.0530] 0.1419 Dodissnsan.sokalts cutie 0.50} 0.0426 23 0.9795 | 0.2650 | 0,7145 i es oe ae 1.00| 0.0852 22| 1.8789! 0.5069] 1.3670 Potasium nitrate 0.05 | 0.0051 22| 0.1116} 0.0482| 0.0684 Do. 0.10 | 0.0102 | - 23 0.2384) 0.0904) 0.1480 Do 0.50 | 0.0508 23} 1.1670} 0.4520] 0.7150 Do. Sh 1.00} 0.1015 22| 2.2382| 0.8646 | 1.3686 Ammonium nitrate 0.05 | 0.0040 | 22} 0.0883} 0.0199 | 0.0684 Do. Tee Bie 0.10 | 0.0080 23} 0.1846] 0.0416 | 0.1430 ps a aay ae 0.50/ 0.0401 23 | 0.9230! 0.2079 | 0.7151 Do....- 1.00 | 0.0803. _ 22) 1.7659) 0.3978 | 1.3681 Sodium sulphate ___. _.__-..._._____- €.05 | 0.0086 22| 0.0785 | 0.0254! 0.0581 iia e sorte tor 0.10} 0.0071 23} 0.1643| 0.0582) 0.1111 Do. ae ie 0.50} 0.0857 23} 0.8215| 0.2661 | 0.5554 TR cl ead is A ae ge eae ters 1.00} 0.0714 22} 1,5719| 0.5091 | 1.0628 WAC alphetes 2 a 0.05 | 0.0040 22| 0.0886! 0.0859] 0.0527 a os eae ee Sf 0.10} 0.0081 23 | 0.1852| 0.0751 { 0.1101 Do. Sen 0.50/| 0,0408 23 0.9260} 0.8753 | 0.5507 We i ke 1.00 0.0806 23} 1.7721] 0.7181] 1.0540 Copper sulphate .__.....__-_________ 0.05 | 0.0040 22} 0.0881! 0.0351, 0.0530 Wo. Qase is ete es 0.10 | 0.0080 23 | 0.1843| 0.0734] 0.1109 Deco. 0.50} 0,¢401 23} 0.9215 | 0.8672] 0.5543 165.25. “i 1.00 | 0.0801 22; 41.7628] 0.7025] 1.0603 Sodium bicarbonate>______.__._____ 0.05 | 0.0020 22; 0.0447] 0.0125! 0.0816 Dex sens: 0.10 | 0.0041 23} 0.0985 | 0.0262] 0.0661 Do a . 0.50 | 0.0203 22| 0.4472| 0.1259! 0.3168 C1” Gea ge AN AR ona Mee Pa 1.00} 0.0407 22} 0.8947| 0.2509] 0.6827 Potassium bicarbonate>_____________ 0.05 | 0.0024 22| 0.0538] 0.0185 | 0.0318 Dieriterk eric csc bicwccaysce 0.10! 0.0049 23 0.1126} 0.0449} 0,0663 Pi parc ge ce 0.50} 0.0245 23} 0.5629| 0.2246| 0.3319 Sas ane ee 1.00} 0.0489 22/ 1.0769] 0.8706} 0.6365 ® The normal consistency values given in this column are averages for the four cements. In nearly every case the variation was only 1 or 2 per cent, In gauging Cement IV with 0.50 N copper sulphate and with 0.50 N potassium bicarbonate, however, the normal consistency was unusually high, being 26 and 28 per cent, respectively. These two values were not counted in the averages given here. > The concentrations of these solutions are based on the amount of COs; present. They are only half the indicated concentrations with respect to alkalinity. : 18920833 370 The Philippine Journal of Science 1922 All four cements when gauged with water alone show a de- crease in tensile strength between the l-year and the 5-year periods. The general effect of the various salts is to lower the tensile strength—usually less than 50 pounds. In some cases, there is an increase in strength. The results are not sufficiently regular to permit any definite conclusions as to the relative effect of the various ions present. Cement III suffered the greatest loss in strength. A comparison of the chemical] analysis of the four cements offers no clue as to why this should be the case. In the physica] tests this cement is higher in fineness than any of the others. Up to one year its tensile strength is much higher than that of the others, but it falls off considerably between the first and the fifth years. The results given in this paper and also those obtained by other investigators since this work was carried out indicate that in general—in the case of concrete, at least—ordinary inorganic substances are less likely to affect the strength of cement when they are used in the mixing water than when present in water to which hardened concrete is exposed. In fact, this is true of many substances, both organic and inorganic. Recently a series of concrete specimens was made, using a number of industrial waste waters.’ These waters contained large quan- tities of various substances in solution, and it was expected that they would show very erratic results. However, in most cases the specimens already broken have not shown any ill effects. At first it seems peculiar that a given water should have less effect when mixed with concrete than when brought into contact with the finished concrete because, of course, there is much more intimate contact in the former case. In the tests made in the Bureau of Science, a normal solution of sodium sulphate had little or no effect on either of the four cements used, while concrete specimens exposed to much more dilute solutions of the same salt may become completely disintegrated. The ex- planation for this is found in the relative quantities of the foreign substance present. The amount of a substance in solu- tion that can be added to cement through the agency of the mixing water is relatively small. For example, when briquettes were made, using normal sodium sulphate solutions as the mix- ing water, the sulphuric anhydride added was only approximately 1.4 per cent of the calcium oxide present in the cement. When concrete is exposed to waters containing sulphates in the field, usually the amount of sulphuric anhydride present in relation *From a paper by D. A. Abrams which has not yet been published. 21,4 Witt: Cement and Concrete 371 to the calicium oxide is unlimited. A number of disintegrated concrete tiles recently examined contained sulphuric anhydride as high as 50 per cent of the calcium oxide present. In the first paper it was stated that— The great variety of uses which modern industry is finding for concrete is continually presenting new problems for research and likewise increasing the importance of work that was completed at a time when the theoretical side was perhaps the only one under consideration. * * * This field of research is becoming more important year by year, because of the many new demands being made on the material. Sewer and drain pipe, storage tanks for various liquids, and even boats are now made of concrete. There- fore it is not suprising that the material should be called upon to resist conditions which were not known a few years ago. These statements are even more applicable to-day than they were five years ago. Notwithstanding the retarding influence of the World War, when labor, fuel, equipment, and transporta- tion were all difficult to obtain, more cement has been produced in the United States during the last two years than during any other two years in the history of the industry. Develop- ment of new uses for cement has kept pace with the increase in production of the material and we are approaching that time when, according to Sunderland,* everyone will have the privilege of studying in a fireproof concrete schoolhouse, living in a city paved and drained with concrete, traveling all over America on concrete roads, and finally being buried in a concrete coffin. During the last few years there has been great activity in the manufacture of various substances to modify the properties and behavior of cement. There are hardeners, set accelerators, antifreeze mixtures, waterproofing compounds, and paints and coatings for decoration and protection against various sub- stances. There are compounds for making cement more econom- ical, for increasing the early strength, for increasing the ultimate strength, for increasing density, and for decreasing density. Without discussing the merits of these compounds, it may be readily seen that they introduce many new problems. * Address by L. T. Sunderland, president of the Portland Cement Asso- ciation, May 16, 1921. tes NEW OR LITTLE-KNOWN TIPULIDAD FROM THE PHILIPPINES (DIPTERA) By CHARLES P. ALEXANDER Of Urbana, Illinois A small but very interesting collection of Philippine crane flies has been received from Prof. Charles Fuller Baker, of the College of Agriculture, University of the Philippines, and is discussed in this paper. Types of all unique specimens have been returned to Professor Baker and so indicated in the text. I have been permitted to retain the types of certain other species, but in all cases duplicate material of these are in Professor Baker’s collection. My sincere thanks are extended to Professor Baker for the privilege of studying this important collection of Philippine Tipulide. LIMNOBIIN 42 Conosia irrorata (Wiedemann). Limnobia irrorata WIEDEMANN, Aussereur. zweifl. Ins. 1 (1828) 574. Conosia irrorata VAN DER WULP, Tijdschr. v. Entomol. 23 (1880) 161. LUZON, Laguna Province, Los Bafios (Baker) ; a male and a female. Epiphragma bakeri sp. nov. General coloration light brown; legs dark brown, the knees and tarsi yellowish; wings yellow with a heavy dark brown pattern; a series of about a dozen cross veins and spurs in cell C; ventral segments yellow, the incisures narrowly dark brown. Male.—Length, 9.5 millimeters; wing, 11.3. Rostrum pale brown; palpi broken. Antenne broken. Head brown, more fulvous on the vertex. Mesonotal prescutum light reddish brown, the disk uniformly darker brown, margined anteriorly and laterally by a light gray line; scutum injured by pin, the lobes dark brown, paler laterally; scutellum injured by pin; postnotum brown basally, more pruinose on posterior half. Pleura almost uniformly brown. Halteres long and slender, pale brown, the knobs darker. Legs with the coxe and trochanters pale brown; 373 374 The Philippine Journal of Science 1922 femora dark brown, paler basally, the extreme tips pale yellow; tibize dark brown, the extreme bases pale yellow; tarsi pale _ brownish yellow. Wings light yellow with a very heavy, non- ocellate brown pattern, distributed as follows: Large blotches at base of cells R and M; at origin of Rs; along the cord and outer end of cell lst M, and at the tips of all the longitudinal veins; a series of about ten dark brown spots in cells C and Sc; most of the remainder of the wing disk clouded with paler brown, but all of the darker brown areas surrounded by narrow, clear yellow borders; outer ends of cells Sc,, 2d R,, R., R,, M,, 2d M., M,, Cu,, and lst A with conspicuous pale yellow spots; veins dark brown, paler in the yellow areas. Venation: Sc, at the extreme tip of Sc,, the latter punctiform; each of the dark spots in costal cell with a spur or complete supernumerary cross vein, there being about a dozen of these beyond h; Rs long, feebly angulated at origin; R,-+, a little longer than the basal deflection of Cu,; r at tip of R,; inner ends of cells R, and R, in alignment; cell Ist M, large, the inner end arcuate; petiole of cell M, short, a little longer than R,+-,; m a little shorter than the outer deflection of M, ; basal deflection of Cu, slightly sinuous, at midlength of cell 1st M,; arcular vein completely obliterated. Abdominal tergites with the basal half, or slightly less, of each dark brown, the remainder slightly paler, thus producing an indistinct bicolorous appearance; sternites conspicuously light yellow with the caudal margin of the segments narrowly but conspicuously dark brown; intermediate sternites with the ex- treme bases likewise darkened; eighth segment and hypopygium dark brown. : LUZON, Mountain Province, Benguet, Pauai (Haight’s place), altitude about 2,400 meters (Baker). Type returned to Profes- sor Baker. Epiphragma bakeri is a very interesting species, distinguished from all allied forms by the series of supernumerary cross veins in the costal cell. Similar cross veins occur in the eriopterine genera Conosia van der Wulp and Clydonodozus Enderlein, but were not known in the Hexatomini. The strongest one of these cross veins has persisted in the other known species of Epi- phragma. This handsome crane fly is dedicated to Prof. Charles Fuller Baker in appreciation of his untiring efforts to make known the remarkable insect fauna of the Philippines. Eriocera mindanaoénsis sp. nov. General coloration reddish, the abdomen orange with the ter- minal segments faintly darkened; mesonotal prescutum reddish 21, 4 Alexander: Little-known Tipulidz (Diptera) 375 gray with dark red stripes; pleura pale reddish brown, faintly striped longitudinally with dark brown; legs brownish black; wings very long and narrow, strongly suffused with brown; cell M, present. Male.—Length, 20 millimeters; wing, 20, its greatest width, 4,4, Female.—Length, 22 millimeters; wing, 17. Male.——Rostrum brown; palpi dark brown. Antenne dark brownish black, short. Head dark brownish gray; vertex behind antennal bases light gray pruinose; immediately caudad of the antennal bases a low, black, vertical tubercle whose cephalic margin is truncated, the lateral angles prominent; immediately behind this is a second tubercle, concolorous with the remainder of the head. Mesonotal prescutum reddish gray with three, confluent, dark red stripes ; remainder of mesonotum obscure reddish brown, the posterior margin of the scutellum faintly pruinose. Pleura pale reddish brown; a distinct, longitudinal, dark brown stripe on mesepisternum; mesosternum and dorsopleural region likewise darker than the ground color of the pleura. MHalteres short, dark brown, base of stem narrowly pale. Legs with the coxz shiny dark brownish red; trochanters dark brown; remainder of legs entirely brownish black; posterior legs broken. Wings very long and narrow, more than four times as long as wide, strongly suffused with yellowish brown, brighter in the costal region ; stigma lacking; veins brown. Venation: Sc, ending just before r, Sc, some distance from tip of Sc,, the latter alone a little shorter than the basal deflection of Cu,; Rs only a little longer than R; r—-m on R,+, about two and one-half times its length beyond the fork; petiole of cell M, a little shorter than cell; basal deflection of Cu, just beyond midlength of cell 1st M,; in the right wing of the type, R, just before its tip bends down to R, in a punctiform contact; in the left wing of the same specimen there is a supernumerary cross vein in cell R, just beyond r. Abdomen with tergites 1 to 4 bright orange; 5 to 7 orange with the median area faintly darkened; 8 and 9 uniformly brown; sternites similarly orange, but the terminal segments even less distinctly darkened; hypopygium small. Female.—Similar in most respects to the male. Head darker; a clear gray line surrounding the eyes; vertical tubercles present but less developed; prescutal interspaces and a capillary median line brown, dividing the red prescutal stripes; mesosternum not 376 The Philippine Journal of Science 1922 so distinctly darkened; no supernumerary cross veins in cell R, and petiole of cell M, longer than the cell; ovipositor with the valves long and slender. _ MINDANAO, Bukidnon Subprovince, Tangcolan (Baker). The allotypic female has been returned to Professor Baker. Eriocera mindanaocénsis is related to E. ferruginosa van der Wulp (Sumatra and Java). In the latter species the antennal flagellum is light colored, the femora and tibiz are yellow with dark apices, and the abdominal tergites are margined laterally with dark brown. ‘ Eriocera perennis polillensis subsp. nov. Female.—Length, about 22 millimeters; wing, 16.8. Differs from typical Eriocera perennis Osten Sacken, as fol- lows: The first scapal segment of antenne is gray pruinose like the head. Mesonotum shiny black. Legs with the femora dark brown, the tips black. Wings dark brown, the proximal end of cell 1st A pale; cell 2d A not pale at base. Petiole of cell M, longer than the cell. Abdominal crossbands on tergites. 2, 3, and 4 only; segment 5 entirely dark. POLILLO (Edward H. Taylor). Type returned to Professor Baker. : Eriocera perennis subcostata subsp. nov. Female.—Length, about 17 millimeters; wing, 12.3. Differs from typical Eriocera perennis Osten Sacken, as fol- lows: Rostrum and palpi black. Antenne with the basal segment gray ; second scapal and basal two flagellar segments yellow; ter- minal flagellar segments black. Head clear light whitish gray, the gene becoming infuscated. Mesonotal prescutum grayish brown with three clear-cut, con- fluent black stripes; remainder of thorax velvety black. Wings dark brown; base of wing broadly bright yellow, this color including about the basal third of cells C and Sc, and the cells proximad of arculus; basal fourth of cell 2d A pale; a conspicuous paler yellow crossband before the cord, extending from R into the base of cell Cu,; cell Sc paler than the remainder of the disk, connecting the basal and discal pale areas ; veins black, yellow in the pale areas. Venation: Sc, ending opposite mid- length of the basal section of R,; r about one and one-half times to twice its length beyond the fork of R,+,; inner end of cell 1st M, nearly square; petiole of cell M, shorter than cell; basal deflection of Cu, at midlength of cell 1st M.,. 21,4 Alexander: Little-known Tipulide (Diptera) 377 Abdomen velvety black, the bases of segments 2 to 6 broadly obscure orange yellow, becoming more obscure and more re- stricted in extent on the posterior segments; segment 7 entirely black; segments 8 and 9 orange; valves of ovipositor black. MINDANAO, Bukidnon Subprovince, Tangcolan (Baker). Type returned to Professor Baker. Eriocera chalybeicincta sp. nov. General coloration velvety black; antennal flagellum obscure brownish yellow, darker apically ; wings brownish black, the anal cells paler; cell M, lacking; basal deflection of Cu, near outer end of cell 1st M,; abdomen velvety black, tergites 2 to 7 with shiny, metallic blue rings. Male.—tLength, 14 millimeters; wing, 13.5. Rostrum and palpi black. Antenne short; scapal segments black; basal three flagellar segments obscure brownish yellow, the terminal segments passing into dark brown; flagellar seg- ments gradually decreasing in length outwardly. Head velvety black with abundant, elongate, black sete. Thorax entirely deep velvety black. Halteres short, dark brown. Legs with the coxe and trochanters velvety black; femora and tibie dark brown, the tips black; tarsi black; legs comparatively stout. Wings brownish black with violaceous reflections; cells M and Cu adjoining vein Cu much paler in one wing of type but not in the other; anal cells grayish; a small pale streak near outer end of cell R; centers of cells 1st M, and Cu, indistinctly pale; veins black. Venation: Se, ending just. before r; Rs about one-third longer than R; R,+, about two- thirds the deflection of R,+-,; basal section of R, about equal to terminal section of R, ; basal section of M,+-, about two-thirds the second section; cell M, lacking; basal deflection of Cu, close to or at the outer end of cell 1st M,; Cu, and outer deflection of Cu, subequal. Abdomen with the first tergite black; tergite 2 velvety black with a broad, shiny, metallic blue ring before midlength ; tergites 3 to 7 with a similar basal ring, more or less telescoped beneath the preceding segment; remainder of abdomen, includ- ing hypopygium, black. MINDANAO, Kolambugan (Baker). Eriocera chalybeicincta is related to E. morosa Osten Sacken (Celebes), from which it is distinguished by the coloration of the body and the venation, as see the position of the outer deflection of Cu,. 378 The Philippine Journal of Science 1922 Eriocera vittipennis sp. nov. General coloration light gray ; thorax with black stripes; wings whitish subhyaline, the veins broadly seamed with dark brown to produce a streaked appearance; Sc, longer than the basal deflection of Cu,; r-m beyond midlength of cell 1st M,; cell M, lacking; basal deflection of Cu, at outer end of cell lst M,; ab- domen black, the basal and terminal segments rufous; ovipositor with the valves long and straight. Female.—Length, about 14 millimeters; wing, 10.2. Rostrum and palpi dark brown.’ Antennz with the first scapal segment elongate, dark brown, dusted with light gray; second scapal segment and basal segment of flagellum yellowish brown, the terminal flagellar segments passing into darker brown. Head light gray, provided with abundant, long, black bristles; no distinct vertical tubercle. Pronotum light gray. Mesonotal prescutum very light gray with three black stripes, the median stripe very broad in front, near midlength tapering rapidly to a point before the suture, conspicuously split by a pale gray vitta; lateral stripes small but conspicuous, widely separated from the intermediate stripe; scutum light gray, the centers of the lobes blackened; scutellum black, the posterior margin narrowly grayish; postnotum gray. Pleura gray, the mesosternum and mesepisternum with more blackened areas. Halteres small, brown. Legs with the coxe dark, dusted with gray; trochanters reddish brown; femora dark brown, the bases narrowly obscure yellow, less distinct on midfemora; tibiz and metatarsi light brown, the tips narrowly darker brown; terminal tarsal segments dark brownish black; posterior legs missing. Wings whitish subhyaline, very con- spicuously streaked longitudinally with dark brown, all the veins being broadly seamed with this color, the pale centers to the cells a little narrower than the seams except in cells 1st R,, R, and M; cells C and Sc brown; cells Cu, Ist A and 2d A more uniformly grayish brown; cells R and M narrowly inter- rupted by a brown wash near midlength; veins dark brown. Venation: Sc, ending opposite r, Sc, some distance from the tip of Sc,, the latter longer than the basal deflection of Cu,; r at fork of R,+,; R, about two to three times R,+,; R,+, and the deflection of R,+, subequal; basal section of M,+. one-half longer than the second section ; inner end of cell 1st M, pointed; cell M, lacking; basal deflection of Cu, almost opposite the outer end of cell 1st M,, about equal to Cu.. 21,4 Alexander: Little-known Tipulide (Diptera) 379 Abdomen with the basal segment black, dusted with gray; tergites 2 and 3 bright rufous, gray pruinose basally on the sides and with a broad, median black stripe; tergites 4 to 7 black, dusted with gray; segments 8 and 9 rufous; sternites 1 to 3 and 8 and 9 rufous; sternites 4 to 7 black. Ovipositor with the valves very long and slender, rufous at base, passing into horn color at the gently upcurved tips of the tergal valves. MINDANAO, Bukidnon Subprovince, Tangcolan (Baker). Eriocera vittipennis is abundantly distinct from all described species of this large and complex genus. TIPULIN 4 Scamboneura plumbea sp. nov. General coloration dark plumbeous, dusted with gray; wings faintly grayish brown, cell Sc and the stigma brown; abdominal segments cross-banded with reddish orange and blue-black. Female.—Length, 17 millimeters; wing, 15. Frontal prolongation of head dark brownish black, narrowly obscure reddish on the dorsolateral region; nasus conspicuous, black; palpi dark brown, the elongate terminal segment paler on basal half. Antenne black, including the scape, the tip of the second scapal segment narrowly paler. Head black, the vertex obscurely reddish behind the antennz, the color extended as a narrow border to the eyes. Prothorax dark plumbeous. Mesonotal prescutum and scutum dark plumbeous, only the humeral region very restrictedly obscure reddish; scutellum obscure yellow; postnotum plum- beous, the lateral sclerites more testaceous. Pleura plumbeous with a light, blue-gray bloom. Halteres dark brown, base of the stem brightened. Legs with the coxae blue-gray; trochanters yellow; femora dark brown, narrowly brighter at base; tibie and tarsi brownish black. Wings with a faint grayish brown tinge, a little darker beyond the cord; cell Sc and the small stigma darker brown; veins dark brownish black. Venation: Se, ending opposite the base of r-m, Sc, at its tip; Rs short but distinct; R,+, about equal to the deflection of R,+,, in align- ment with R,+.; r at fork of R,+,; tip of R, atrophied; de- flection of R,+, directed basad as in the genus, subequal to r-m; forks of cells M,, M,, and M,, deep; basal deflection of Cu, about two-thirds of its length beyond the fork of M. The stigma entirely fills the small cell 1st R.. Abdomen with the basal tergite plumbeous; segment 2 red- dish, darker dorsomedially, the apical two-fifths shiny blue- 880 The Philippine Journal of Science 1922 black; segments 3 and 4 with a little less than the basal half of each reddish orange, darker dorsomedially, the remainder shiny blue-black; segments 5 and 6 reddish orange, darker dorsomedially; segments 7 and 8 blue-black; segment 9 and ovipositor obscure brownish orange. Ovipositor with the tergal valves elongate, compressed, the tips rather obtusely rounded. LUZON, Mountain Province, Benguet, Pauai (Haight’s place), altitude about 2,400 meters (Baker). Scamboneura plumbea is very different from the two Phil- ippine species of the genus hitherto described. There are now five species of Scamboneura, all from Amboina, Java, and the Philippines. The two species from East Africa described as species of this genus by Speiser and by Riedel are not mem- bers of the genus in the sense of Osten Sacken. One, at least, belongs to the subgenus Trichodolichopeza Alexander, of the genus Dolichopeza Curtis. Pselliophora perdecora sp. nov. General coloration obscure orange, the abdominal segments beyond the second shiny black; legs with the femora bright yellow, the tibize and tarsi black; wings dark brown, the base rather broadly bright yellow. Female.—Length, about 20 millimeters; wing, 20. Frontal prolongation of the head orange, the nasus faintly darker; basal segments of palpi yellow with brown setz; ter- minal segment yellow, the extreme tip darkened. Antenne brownish orange throughout; first flagellar segment subequal to segments 2 and 3 combined. Head obscure orange. Thorax entirely obscure orange, without markings. Halteres yellowish orange. Legs with the cox and trochanters concol- orous with the pleura; femora bright yellow, the extreme apices very indistinctly darker; tibiz and tarsi black, unmarked. Wings broad, dark brown, the base rather broadly and conspicu- ously bright yellow, this including about the basal fifth of cells C and Sc, the extreme bases of cells R, M, Cu, and 1st A, about the basal third of cell 2d A and all the cells proximad of ar- culus; an indistinct pale area immediately proximad of the stigma; a linear white streak in cell R near midlength; a similar but smaller streak near the outer end of cell M (lacking in one wing of type) and a third streak near outer end of cell 1st A; veins dark brown, yellow in the flavous basal area; obliterative breaks at the end of Rs and base of R,+,, basal section of M,+, and the second section of M,. Venation: Rs much longer than R,+,, its course subsinuous; R,+, longer 21,4 Alexander: Little-known Tipulide (Diptera) 381 than the ultimate section of R,; cell M, rather broadly sessile; m-—cu distinct. Abdomen with segments 1 and 2 orange, the extreme lateral apical angle of tergite 2 darkened; remainder of abdomen shiny black, only the tips of the ovipositor deep reddish. Valves of ovipositor comparatively long and slender. MINDANAO, Bukidnon Subprovince, Tangcolan (Baker). Pselliophora perdecora belongs to the incunctans group of the genus, an aggregation of large, showy species that contains some of the most beautiful crane flies known. Pselliophora incunctans (Walker) and P. ineunctans ochrifemur Enderlein of Celebes, P. praefica Bezzi and P. tripudians Bezzi of the Philip- pines, and possibly other species belong to this group. Pselliophora praefica Bezzi. Pselliophora praefica Brzzi, Philip. Journ. Sci. § D 12 (1917) 110. The undescribed female of this species may be described as follows: Length, 21 millimeters; wing, 19. Head black, the gene and occiput deep reddish brown. Other characters entirely as in male. Ovipositor with the valves black, only the extreme tips reddish. MINDANAO, Kolambugan (Baker). Allotype. Female. The allotype has been returned to Professor Baker. Bezzi’s type male was likewise from Mindanao. In the collec- tion of the State Natural History Museum in Vienna, Brunetti ' found a male and a female of this species from Palawan. These he considered to represent a variety of P. incunctans (Walker), but in my opinion there can be no doubt but that P. praefica is a distinct species. Pselliophora tripudians Bezzi. Pselliophora tripudians Bezzi, Philip. Journ. Sci. § D 12 (1917) 111. A female agrees very closely with Bezzi’s description except that the specimen is smaller (length, 17 millimeters; wing, 16). The foreleg, lacking in Bezzi’s type, is very different from the other legs and may be described as follows: Femora entirely light yellow, the tip not darkened; tibize yellow, the tip a little infuscated; tarsi entirely brownish black. LUZON, Laguna Province, Los Bafios (Baker). This specimen has been returned to Professor Baker. "Rec, Indian Mus. 6 (1911) 239. 382 The Philippine Journal of Science 1922 In the collection of the Paris Museum, a somewhat broken male specimen was found which is surely the male of this species. It is herewith described as allotypical. Male.—Similar to the female, differing as follows: Basal seg- ments of antennz yellow, passing into brown toward the tips; flabellations of antennze dark brown. Mesonotum entirely orange, including also the coxe and trochanters of all the legs; the forelegs are broken but the other legs are entirely as in the female. Abdomen black, the hypopygium conspicuously and abruptly reddish orange, only the caudal margin of the ninth tergite and the pleural appendages dark. LUZON, Manila, No. 246-1861. Allotype, male, in the collec- tion of the Paris Museum. Pselliophora tripudians is evidently allied to, but distinct from, P. incunctans (Walker) of Celebes. The thoracic pat- tern is paralleled in the two sexes of the two species. Pselliophora pumila sp. nov. General coloration black, the mesonotal prescutum yellow with three grayish brown stripes; all tibiz ringed basally with white; wings grayish brown; cells C and Se dark brown; a brown clouding along the cord; abdomen black, the caudal margin of tergites 2 to 4 narrowly yellow. - Male—tLength, 10 millimeters; wing, 11. Frontal prolongation of head dark brown; palpi concolorous. Antenne dark brown throughout; flabellations of antennz con- spicuously clothed with white hairs. Head dark brown. Pronotal scutum dark brown; scutellum dark brown me- dially, the lateral portions yellow. Mesonotal prescutum yellow with three conspicuous brown stripes that are dusted with gray; median stripe very broad; lateral stripes small, confluent with the median stripe near suture; pseudosutural fovese small, brown; scutum and postnotum black; scutellum black with con- spicuous, elongate, yellow bristles and delicate transverse rows of microscopic yellow hairs. Pleura dark brown, the extreme dorsopleural region faintly yellowish. Halteres dark brownish black, the base of the stem indistinctly yellow. Legs with the coxe and trochanters black, femora black, about the basal half of the posterior femora more testaceous; tibie black with a conspicuous white ring just beyond the base, this broadest and most distinct on the posterior tibie; tarsi black. Wings clouded with grayish brown; costal region dark brown, including cells C and Sc; paler brown clouds along the cord and along vein Cu; veins dark brown. Venation: Sc, ending about opposite two- 21,4 Alexander: Little-known Tipulide (Diptera) 883 thirds Rs; R,+-, short, about equal to the distal section of ‘oe much shorter than the basal deflection of Cu, ; cell M, very short- petiolate; a slight fusion of M, and Cu,. Abdomen shiny black, the caudal margins of tergites 2, 3, and 4 rather narrowly but conspicuously yellow. Hypopygium comparatively large and complex, black, the ninth segment elongate-cylindrical and tilted at an angle to the remainder of the abdomen. LUZON, Mountain Province, Benguet, Pauai (Haight’s place), altitude about 2,400 meters (Baker). Pselliophora bicinctifer sp. nov. © General coloration black; antennal flagellum with only nine segments; mesonotal prescutum obscure yellow with three brownish black stripes; halteres yellow; tibize dark brown with a basal and a subterminal yellowish ring; wings pale grayish yellow, the cells beyond the cord dark brown; abdomen deep rufous, the tergites trilineate with black, the sternites with only the lateral margins narrowly darkened. Female.—Length, about 17 millimeters; wing, 15. Frontal prolongation of the head dark brownish black, clothed with conspicuous golden yellow and black bristles interspersed ; palpi brown, the third segment pale at base and apex; terminal segment black beyond the base. Antennz with the first seg- ment brownish black, the remainder bright orange yellow; flagellum with only nine segments, the first segment shorter than the second and third combined ; Segments 2 to 5 narrow at base, the proximal face dilated; terminal segments flattened, disklike, greatly crowded. Head dark brownish black. Pronotum dark brownish black. Mesonotal prescutum ob- secure yellow, with three conspicuous brownish black stripes, the median stripe narrowed behind but attaining the suture; scutal lobes with brownish black centers; median area in front with a triangular shiny area immediately behind the suture; scutellum brown; postnotum dark brown. Pleura dark velvety brownish black. Halteres light yellow. Legs with the coxe velvety brownish black; trochanters brownish black, the tips fading into reddish; femora yellow, the tips broadly and con- spicuously blackened; tibize yellow, the tips narrowly dark brown; a conspicuous dark brown annulus occupies the mid- center of each tibia, restricting the ground color to a narrow basal and a slightly broader subapical yellow ring; metatarsi yellow, the tips black; remainder of tarsi black; inner face of claws microscopically denticulate. Wings with the cells proxi- 384 The Philippine Journal of Science mad of the cord light grayish yellow, the broad apex brown; stigma still darker brown; cell C darkened except for a narrow space at either end; cell Sc dark except for a narrow space at ~ base; a conspicuous dusky cloud along vein 1st A; base of wing brighter yellow; a band along the cord (including base of cells ist R,, lst M,, M,, and Cu, and the outer ends of cells R and most of M, especially along vein Cu) much clearer yellow than the remainder of the basal cells; anal cells pale; veins dark brown, yellow in the pale areas, including the entire length of vein Cu. Venation: Rs arcuated; R,+, about equal to the distal section of R,; R, at base perpendicular to R,+-,; cell M, barely sessile; cell 1st M, irregularly pentagonal, none of the faces exactly parallel; in one wing of type a supernumerary cross vein in the base of cell M,; fusion of M,+, and Cu, punctiform. : Abdomen with basal tergite deep rufous, the caudal margin black; remaining tergites deep rufous, trilineate with black, the narrow median stripe slightly interrupted at the posterior margin of the segments; sternites deep rufous, the lateral mar- gins narrowly darkened. Ovipositor with the tergal valves straight, yellowish horn color. Luzon, Mountain Province, Benguet, Pauai (Haight’s place), altitude about 2,400 meters (Baker). Type returned to Profes- sor Baker. Nephrotoma ortiva (Osten Sacken). Pachyrrhina ortiva OSTEN SACKEN, Berliner Ent. Zeitsch. 26 (1882) 93. LUZON, Laguna Province, Los Bafios (Baker). A female that agrees closely with Osten Sacken’s type female, except in the following characters: Length, 15 millimeters; wing, 12. Scutum yellow, each lobe completely traversed by a shiny black longitudinal stripe which leaves not only the median area yellow, as stated by Osten Sacken, but also the posterior lateral portions of each lobe; scutellum black; caudal margin of lateral sclerite of postnotum black. Abdominal segments 2 to 6 with a broad, black, terminal band that is slightly, if at all, attenuated laterally, the band occupying about the posterior half of each segment. INDO-MALAYISCHE RHYNCHITINEN (CURCULIONIDA) I, SIEBENTER BEITRAG ZUR KENNTNISS DER CURCULIONIDEN Von EDUARD Voss Waldsiedlung Spandau, Germany DREI TEXTFIGUREN Die mir durch Herrn Hofrat Prof. Dr. K. M. Heller, Dres- den, freundlichst zur Bearbeitung iiberwiesene Ausbeute von Prof. C. F. Baker, Manila, muss als recht ergiebig bezeichnet 120° 125° T T T , , § 20 SEE : ; /roaney > $8 sii : ‘CEL EBES ad rh dealt 120° 125° Fic. 1. Karte der Philippinen mit eingetragenen Sammelpunkten. 1892083——4 j 385 386 The Philippine Journal of Science 1922 werden. Sie bildet einen wichtigen Beitrag sowohl zur Kennt- niss der indomalayischen Fauna wie auch im _ besonderen der Subfamilie Rhynchitine. Die nachstehend beschriebenen Arten gehéren der Deporaus und Eugnamptus Gruppe an. Na- mentlich die erste Gattung ist recht artenreich vertreten und bringt eine Anzahl neuer bisher nicht oder wenig bekannter Formen. Die Hugnamptus-Arten dagegen treten hier mehr zuriick und scheinen in den rezenten Formen in Central Amerika ihr Hauptverbreitungsgebiet zu haben. Schon an dieser Stelle darf darauf hingewiessen werden, dass beide Gattungsgruppen in morphologischer Hinsicht eine auffallende Parallelentwicklung genommen haben und wahrscheinlich auf einen gemeinsamen Ursprung zuriickzufiihren sein werden; das Aufspaltungsgebiet diirfte in Indonesien zu suchen sein. Genus SCOLOCNEMUS Kirsch Scolocnemus pilosiusculus sp. nov. Fig. 2. Kopf viel langer als breit; auf dem Scheitel und der Stirn kraftig runzlig, seitlich und hinten nur sehr fein und zerstreut punktiert; Stirn flach, Augen sehr stark vorgewolbt; Schlafen schwach gerundet, fast parallel und langer als der Augenlangs- durchmesser. Kopf wenig tief, rinnenformig abgeschniirt; Hals kurz und schwach gerundet, zur Basis verbreitert. Riissel so lang wie der Kopf, wenig gebo- gen; von der Seite gesehen, oben vor der Spitze bucklig erhoben. Fihler* kurz vor der Spitze des Riissels eingelenkt; Fiihlerfurche breit und tief und zur Spitze auf die Oberseite iibertretend. Auf der Basalhilfte stumpf gekielt, im tibrigen kraftig und dicht punktiert. Halsschild langer als breit; kraftig runzlig (wirbelartig) punktiert; in der Mitte mit feiner Langsfurche. Seitlich massig stark gerundet, zum Vorderrand etwas stirker als zur Basis verschmilert, hier wie dort schmal gerandet. Schildchen langer als breit. Flii- geldecken gut anderthalbmal so lang wie breit, seitlich fast parallel, hinter den Schultern etwas eingezogen und hinter dem Fig. 2. Scolocnemus pilosiusculus sp. nov. *An dem typischen Exemplare fehlen die Fiihler bis auf den kurzen gedrungenen Schaft. Das Tier ist aber so charakteristisch, dass die Erkennung ohne weiteres méglich ist. 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 387 Schildchen schwach eingedriickt. Punktfurchen stark und tief, Punkte sehr dicht stehend, teilweise viereckig; Zwischenriiume sehr schmal, kaum ausgebildet; hier und da findet ein feiner Punkt Platz. Tibien schlank, alle kraftig gebogen. Mittelti- bien am starksten gebogen; Vordertibien hinter der Mitte mit starkem, etwas einwirts gebogenen Dorn bewehrt. Ebenso sind die Vordercoxen auf der Spitze mit einem ldffelartigen Hocker besetzt. Die Mittelbeine sind kiirzer als die iibrigen. Farbung braunlich rot. Das Abdomen, die Tibien, Hinter- schenkel, Riissel an der Seite (anscheinend auch die Fiihler), und die Apicalhalfte der Fliigeldecken sind geschwirzt. Be- haarung der Decken hinten reihig, greis, schwach aufstehend; im tibrigen ist das Tier nur sparsam und kurz behaart, lediglich das Abdomen ist mit etwas langeren greisen Haaren besetzt. Linge, 4 Millimeter (s. r.). ° BORNEO, Sandakan (13938), 1 Exemplar. Genus DEPORAUS Leach Uebersicht der Untergattungen. a*, Hintertibien des Mannchens abnormal lang, im letzten Drittel kraftig einwarts gebogen; Riissel des Weibchens langer als Kopf und Hals- schild zusammen; Schaftglied langgestreckt, den Vorderrand der Augen erreichend Megalarodepus subg. nov. a*, Hintertibien von normaler Lange, Riissel des Weibchens nicht langer als Kopf und Halsschild zusammen; Schaftglied kiirzer, die Augen nicht erreichend. b*. Riisselspitze mit 2 aufstehenden scharfen Héckern besetzt oder mit vorgezogenem, der Lange nach gekieltem Lappen versehen. Capylarodepus subg. nov. 6’. Riisselspitze ohne scharfe Hicker, einfach. c*, Riissel vollstindig flachgedriickt; von oben gesehen vorn sehr breit, hinten stark verengt; Fiihler kurz vor der Riisselbasis eingelenkt. Kopf gross, von der Seite gesehen breiter als das Halsschild. Platyrhynchites subg. nov. c*. Riissel nicht der ganzen Linge nach flachgedriickt. d*. Kopf mehr oder weniger kraftig ringsum abgeschniirt. e*. Augen meist stark vorgewélbt, oft halbkugelig vorstehend (selten flacher gewélbt; dann sind sie auf der Stirn ein- ander genihert und der Kopf ist tief vom Hals abge- schniirt) ; Schlafen in der Regel kiirzer als der Augenlangs- durchmesser. Farbung rotgelb bis rotbraun, bisweilen sind die Fligeldecken mehr oder weniger geschwarzt oder das Halsschild trigt ein schwarzes Lingsband. : Arodepus Heller i. }. e*. Augen flach, wenig vorgewélbt ; Schlafen so lang oder linger als die Augen; Hals sehr kurz oder ganz im Halsschild versenkt. Farbung schwarz oder blau. 388 The Philippine Journal of Science 1922 f*. Der vorletzte Punktstreifen vereinigt sich mit dem Rand- streifen in der Gegend der Hinterhitiften. Hypodeporaus subg. nov. f?*. Der vorletzte Punktstreifen liuft parallel dem Randstreifen bis zur Spitze der Decken as . Deporaus s. s. d*, Kopf nicht. deutlich abgeschnirt......, Dannlindincehun subg. nov. Deporaus (Megalarodepus) tibialis sp. nov. Minnchen.—Kopf langer als breit, massig stark abgeschniirt; ziemlich fein und sehr dicht, fast runzlig punktiert. Stirn etwas eingedriickt und so breit wie der Riissel vor der Basis. Augen ziemlich kriftig vorgewolbt; Schlafen nicht ganz so lang wie der Augenlingsdurchmesser, parallel, und kaum zur Hin- schniirung zugerundet; Hals halb so lang wie die Schlifen. Riissel so lang wie der Kopf mit Hals, gebogen, oben gebuckelt; auf der Basalhalfte gekielt, seitlich undeutlich punktiert ge- furcht; vor der Spitze kraftig verbreitert. Fiihler in der Mitte des Riissels eingelenkt. Schaftglied Keulenférmig, dreimal so lang wie breit; das erste Geisselglied lang oval, doppelt so lang wie breit; das zweite bis vierte Glied so lang wie das Schaft- glied ; das fiinfte und sechste Glied etwas kiirzer, und das siebente Glied wenig kiirzer als die vorhergehenden beiden. Das erste und zweite Glied der Keule gleich lang, etwa doppelt so lang wie breit, das dritte Glied mit dem deutlich erkennbaren Endglied langer.. Halsschild quer, zum Vorderrand stark zuge- rundet ; massig stark und dicht punktiert. Schildchen so lang wie breit. Fliigeldecken anderthalbmal so lang wie breit; seitlich zunachst parallel, dann starker verbreitert. Punktstreifen sehr kraéftig und tief, Zwischenriiume sehr schmal und fein punktiert. Hinterschenkel viel langer als die Vorder- und Mittelschenkel ; die Hintertibien fast doppelt so lang wie die Mittelschienen, kriftig gebogen, wahrend die anderen flach gebogen sind. Das hin- tere erste Tarsenglied ist wenigstens gleich zwei Drittel der Lange der Mitteltibien. _ Weibchen.—Stirn etwas breiter als der Riissel vor der Basis; Schlafen halb so lang wie der Augenlingsdurchmesser. Riissel sehr lang und im ersten Drittel gebogen, langer als Kopf und Halsschild zusammen; oben glanzend und nicht deutlich sicht- bar punktiert. Fiihler im basalen Viertel eingelenkt; Fiihler- furche flach und nach vorn iiber die Einlenkungsstelle nicht hinausgefiihrt. Schaftglied sehr lang, die Spitze erreicht fast die Augen. Beine normal gebildet. Farbung rotbraun; Apicalhalfte des Riissels sowie der grés- sere hintere Teil der Fliigeldecken schwarz. Auch die Tibien sind mit den Tarsen mehr oder weniger angedunkelt. Behaar- ung kurz, halbaufstehend; einige zerstreute Haare stehen lin- 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 389 ger auf. Unterseite lang anliegend greis, doch nicht sonderlich dicht behaart. Lange, 4.5 bis 5.5 Millimeter (s. r.). . BORNEO, Sandakan, 1 Parchen. Deporaus (Capylarodepus) galerucoides Heller. : Deporaus nigrilineatus sp. nov. Unter den Arten mit Doppelhécker auf der Riisselspitze auf- fallig durch die mit breitem schwarzem Nahtband und schwar- zen Seiternrandern gezeichneten Fliigeldecken. Kopf seicht eingeschniirt; auf dem Scheitel starker und dichter, seitlich und hinten feiner und zerstreuter punktiert. Stirn schwach eingedriickt, breiter als der Riissel vor der Basis. Schlafen parallel, fast so lang wie der Augenlangsdurchmesser. Riissel so lang wie das Halsschild auf dem Riicken, schwach gebogen, zur Spitze mdssig verbreitert. Fiihler kurz vor der Mitte des Riissels unter einer hoéckerigen Wulst eingelenkt. Schaftglied kréftig, doppelt so lang wie breit; das erste Geissel- glied schwiacher, cylindrisch; das zweite Glied anderthalbmal so lang wie das Schaftglied, verkehrt kegelformig; die naichsten Glieder etwas kiirzer als das zweite Glied, in der Lange jedoch kaum verschieden. Das erste und zweite Glied! der Keule gleich lang, jedes so lang wie das zweite Geisselglied; das dritte Glied langer und schwicher, schlank, zugespitzt. Halsschild so lang wie breit, seitlich kraftig gerundet; vor der Basis etwas einge- zogen und hier breiter gerandet; zum Vorderrand stirker ver- schmilert und feiner gerandet. Punktierung oben ziemlich kraf- tig und dicht, seitlich fein und zerstreut; glanzend. Schildchen so lang wie breit, an der Spitze tief dreieckig ausgeschnitten. Fliigeldecken fast doppelt so lang wie breit; in der Mitte seit- lich etwas eingezogen. In den Punktfurchen stehen im hellen Feld massig starke Punkte um mehr als die grésse ihres Durch- messers von einander entfernt. Zwischenraume gewdélbt und fein punktiert. Vorder- und Mitteltibien gerade, Hintertibien etwas linger und leicht gebogen. Farbung der Kérperunterseite bis auf die gelbbraunen Seiten schwarz; ebenso die Seiten der Mittel- und Hinterschenkel, das Schildchen, ferner der Seitenrand und die Naht der Fliigeldecken breit schwarz gefirbt; Beine, Fiihler, und der restliche Teil der Decken gelb; im tibrigen ist das Tier rotgelb bis rétlich braun gefirbt. Behaarung der Fliigeldecken kurz und wenig auf- gerichtet; die Seitenteile der Brust sind lang anliegend silber- greis behaart, die Haare verdecken den Untergrund. Abdomen, 390 The Philippine Journal of Science 1922 Beine, und die Unterseite des Riissels sind lang abstehend behaart. Lange, 7.5 Millimeter (s. r.). BoRNEO, Sandakan, 1 Tier. SUMATRA, Tandjong Poera (R. Heintze ges.), 1 Exemplar. Das Exemplar von Borneo ist namentlich im Halsschild etwas dunkler, stellenweise schwarz gefarbt; die Punktierung ist auf dem Halsschild in der Nahe der Mittelfurche schwach verrunzelt. Weitere grundlegende Unterschiede waren nicht festzustellen. Deporaus rugicollis sp. nov. Kopf nur seicht abgeschniirt, fein und zerstreut punktiert; matt. Augen weniger vorgewolbt als bei den nichsten Arten; die Entfernung zwischen Augenhinterrand und Halsschildvor- derrand grésser als der Augenlangsdurchmesser. Schlifen parallel. Stirn flach und breiter als der Riissel vor der Basis. Dieser kurz und gedrungen, kaum so lang wie der Kopf, matt. Fiihler in der Mitte des Riissels eingelenkt. Schaftglied nur wenig linger als breit; das erste Geisselglied so lang wie das Schaftglied, schwacher, fast walzenférmig; das zweite Glied annahernd so lang wie das Schaft- und das erste Geisselglied zusammen. Halsschild linger als breit, seitlich nur schwach gerundet, am Vorderrand wenig schmaler als an der Basis, hier wie dort schmal gerandet. Punktierung kraftig querrunzlig. Schildchen so lang wie breit, an der Spitze stumpf dreieckig ausgeschnitten. Fliigeldecken nach hinten schwach verbreitert. Hinter dem Schildchen quer eingedriickt. Streifen tief ge- furcht, runzlig; Zwischenriume sehr schmal. Vordertibien ge- rade; Mittel- und Hintertibien kriftiger gebogen. Hinterbeine erheblich linger als die Mittelbeine. Farbung rotbraéunlich, die Schenkelspitzen geschwirzt. Be- haarung der Decken kurz, fast anliegend; Abdomen nur sehr kurz, abstehend und zerstreuter behaart. Lange, 5.5 Millimeter. BORNEO, Sandakan (13942), 1 Exemplar. Deporaus iliganensis sp. nov. Dem vorigen sehr ahnlich, doch ist der Kopf kraftig und dicht punktiert, die Augen sind starker vorgewolbt; Entfernung zwischen Augenhinterrand und Halsschildvorderrand fast so gross wie der Augenlingsdurchmesser. Riissel etwas linger als der Kopf, schwach gebogen. Fiihler lang und schlank, in der Mitte des Riissels eingelenkt ; Schaftglied kraftig, doppelt so lang 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 391 wie breit; das erste Geisselglied fast cylindrisch, mindestens doppelt so lang wie breit; das zweite Glied fast so lang wie das Schaft- und erste Geisselglied zusammen; das dritte Glied wenig kiirzer als das zweite Glied; das vierte bis siebente Glied wiederum etwas kiirzer und untereinander fast gleich lang. Keule kraftiger, das erste und zweite Glied gleich lang und so lang wie das zweite Geisselglied; das dritte Glied der Keule viel langer. Halsschild kaum so lang wie breit, seitlich kraftig gerundet; zum Vorderrand viel starker verschmalert als zur Basis, an ersterem breiter, an letzterer schmal gerandet. Punktierung sehr kraftig, runzlig; die Mitte der Scheibe fein langsgefurcht. Punktstreifen der Fliigeldecken tief gefurcht, die Punkte mit einander verbunden, sodass die einzelnen Punkte nicht erkennbar werden. Zwischenradume feiner runzlig punk- tiert. Hintertibien schwach gebogen; Schenkel und Schienen innen lang zottig behaart. Farbung rotgelb, Seitenteile der Mittelbrust stellenweise geschwarzt. Behaarung der Decken kurz und halbaufgerichtet ; Brust und Seitenteile lang anliegend, der iibrige Korper langer abstehend behaart, zum Teil zottig. Lange, 6 Millimeter. MINDANAO, Iligan (13954), 1 Exemplar. Deporaus sandakanensis sp. nov. Der vorhergehenden Art sehr ahnlich, doch stehen die Punkte in den Punktstreifen deutlich von einander entfernt, die Zwi- schenraume sind staérker verrunzelt, und die Behaarung ist dichter, anliegend. Von Deporaus rugicollis ist sie hauptsachlich durch den staérker und dichter punktierten Kopf zu trennen. Schlifen parallel, Augen kraftig vorgewélbt; Stirn schwach eingedriickt, langsrunzlig punktiert. Riissel des Mannchens kaum so lang wie der Kopf, des Weibchens etwas linger, schwach gebogen. Mannchen: Riissel zwischen der Fiihlereinlenkung etwas buckelig und langsgefurcht, zur Spitze kraftig verbreitert. Fiihler in der Mitte des Riissels eingelenkt. Schaftglied verkehrt kegelfé6rmig, doppelt so lang wie breit; das erste Geisselglied etwas schwicher, fast cylindrisch, doppelt so lang wie breit; das zweite Glied anderthalbmal so lang wie das erste Glied ; das dritte und vierte Glied ein wenig kiirzer als das zweite Glied; die restlichen Glieder fast gleich lang und wenig kiirzer als das dritte und vierte Glied. Weibchen: Riissel der ganzen Linge nach oben glatt, glinzend; seitlich ist die Wulst durch eine missig starke Punktreihe scharf abgegrenzt. Fiihler im basalen 392 The Philippine Journal of Science 1922 Drittel des Riissels unter einem seitlich vorspringenden Hocker eingelenkt; Schaftglied fast dreimal so lang wie breit; das erste Geisselglied schwacher, gut doppelt so lang wie breit; das zweite und dritte Glied gleich lang und erheblich langer als das erste Glied; die nachsten Glieder etwas kiirzer und untereinander gleich lang. MHalsschild so lang wie breit, seitlich schwach gerundet, an der Basis nur wenig breiter als am Vorderrand. Basis und Vorderrand breit gerandet. Punktierung auf der Scheibe kraftig runzlig, seitlich sehr fein zerstreut. Fliigel- decken seitlich parallel. Streifen furchenartig, doch stehen die Punkte deutlich von einander getrennt. Zwischenriume schmal und runzlig punktiert. Farbung gelblich braun. Behaarung der Decken kurz, fast anliegend, auf dem Halsschild kurz aufstehend; auf der Brust lang anliegend, den Untergrund jedoch nicht verdeckend. Abdo- men, Beine, und Unterseite des Riissels linger abstehend behaart. Lange, 4.5 bis 5 Millimeter. BORNEO, Sandakan (13943, 13952), 1 Parchen. Deporaus exophthalmus Heller. LuzON, Los Bafios. Deporaus pilifer sp. nov. Kopf massig stark abgeschniirt, kraftig runzlig punktiert, Augen ziemlich stark vorgewélbt. Riissel des Mannchens so lang wie das Halsschild, des Weibchens so lang wie Kopf und Hals- schild zusammen, gebogen; hinten rauh punktiert, vorn glatt und glanzend. Mannchen: Fiihler kurz. hinter der Mitte unterhalb einés seitlichen Buckels eingelenkt. Schaftglied iiber dreimal so lang wie breit; das erste Geisselglied oval, etwas schwacher, kaum doppelt so lang wie breit; das zweite Glied schwach verkehrt kegelférmig und so lang wie das Schaftglied; das dritte und vierte Glied gleich lang, kiirzer als das zweite Glied; das fiinfte Glied nur etwa halb so lang wie das zweite Glied; das sechste Glied linger als das fiinfte Glied, und das siebente Glied wenig kiirzer als das vorhergehende sechste. Keule massig kraftig. Weibchen: Fiihler im basalen Viertel des Riissels eingelenkt. Schaftglied sehr lang, die Spitze fast den Vorderrand der Augen erreichend. Das erste Geisselglied kiirzer als die nachfolgenden, die in der Linge wenig verschie- den sind. Halsschild so lang wie breit, stark runzlig punktiert; seitlich ziemlich kraftig gerundet, zum Vorderrand mehr als zur Basis verschmilert; an der Basis schwach gerandet.. Schildchen so lang wie breit, an der Spitze nicht ausgebuchtet 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 393 und hier bewimpert. Fliigeldecken nicht ganz doppelt so lang wie breit, nach hinten zu wenig verbreitert, sehr stark punk tiert. Zwischenriume nicht ausgebildet. ; Farbung braunschwarz, Riissel und Halsschild mehr _ rot- braun. Behaarung lang abstehend; auch das Abdomen ab- stehend, die Brust mit Seitenteilen jedoch anliegend behaart. Lange, 4 bis 4.5 Millimeter. ; SINGAPORE (13925), 2 Minnchen und 1 Weibchen. Durch die aufstehende Behaarung der Fliigeldecken, die starkeren Punktstreifen, und den runzlig punktierten Kopf von den nichstverwandten Arten leicht zu trennen. Deporaus (Arodepus) angustifrons sp. nov. Kopf linger als breit, sehr kriftig abgeschniirt, fein und nicht ganz dicht punktiert, im hinteren Teil dusserst fein querge- wellt. Augen wenig vorgewélbt und die Schlifen fast so lang wie der Augenlangsdurchmesser ; letztere sind kraftig gerundet zur Einschniirung verengt. Der Hals ist stark kegelférmig zum Halsschildvorderrand wieder erweitert. Augen auf der Ober- seite einander genihert; Stirn daher schmaler als der Riissel, ungefurcht. Riissel so lang wie der Kopf, geknickt; zur Spitze schwach verbreitert, oben glanzend und feiner punktiert. Fiih- ler hinter der Riisselmitte eingelenkt. Schaft- und erstes Geis- selglied fast gleich lang, doppelt so lang wie breit, das zweite und dritte Glied gleich lang, jedes wenig kiirzer als Schaft- und erstes Geisselglied zusammen; das vierte bis siebente Glied in der Linge wenig verschieden, etwas kiirzer als die beiden vorhergehenden. Keule kraftig, die zwei ersten Glieder kaum doppelt so lang wie breit. Halsschild etwa so lang wie breit, seitlich schwach gerundet und stark konisch von der Basis nach vorn verschmilert; ziemlich fein und dicht punktiert. Basal- rand fein, Vorderrand breiter gerandet. Fliigeldecken nach hinten schwach verschmilert. Punktstreifen kraftig, die Punkte stehen von einander etwas entfernt; Zwischenriume sehr schmal, fein punktiert. Farbung schwarz; Vorder- und Mittelhiiften, die Schenkel, und das Abdomen rétlich braun. Fliigeldecken seitlich, hinten, und an der Naht schwarz, im iibrigen braun. . Behaarung der Decken kurz, wenig aufgerichtet. Die Seitenteile der Mittel- und Hinterbrust sind dicht und lang anliegend silbergreis, das Abdomen diinner behaart. Linge, 3.5 Millimeter. SINGAPORE (13906), 1 Exemplar. 394 The Philippine Journal of Science 1922 Von allen verwandten Arten mit tief abgeschniirtem Kopf an den flachen auf der Stirn einander geniherten Augen leicht zu unterscheiden, Deporaus penangensis sp. nov. Kopf kraftig abgeschniirt, 4usserst fein und zerstreut punk- tiert. Stirn so breit wie der Riissel vor der Basis, flach und gefurcht; Augen gross und vorquellend. Riissel des Mannchens so lang, des Weibchens linger als der Kopf, gebogen, zur Spitze verbreitert; besonders auf der vorderen Halfte glatt und glan- zend, beim Minnchen fein, beim Weibchen nicht deutlich punk- tiert. Fiihler hinter der Mitte, beim Weibchen im basalen Viertel eingelenkt. Mannchen: Schaftglied gut doppelt so lang wie breit; das erste Geisselglied oval, langer als breit; das zweite Glied etwas linger als das Schaftglied; das dritte Glied langer als das zweite Glied; das vierte und fiinfte Glied so lang wie das zweite Glied; das sechste und siebente Glied etwas kiir- zer. Keule missig stark, das erste und zweite Glied gleich lang, langer als breit; das dritte Glied doppelt so lang wie das zweite; Endglied ziemlich deutlich abgesetzt. Weibchen: Schaft- glied linger, wohl dreimal so lang wie breit; das dritte Geissel- glied erheblich langer als das zweite Glied; im tibrigen sind die Fiihler wie beim Mannchen gebildet. Halsschild kaum breiter als lang, seitlich kraftig gerundet, zur Basis wenig, zum Vor- derrand stark verschmilert; hier breiter, an der Basis feiner gerandet. Punktierung miassig kraftig und ziemlich dicht. Fliigeldecken gut anderthalbmal so lang wie breit. Punkt- streifen ziemlich stark, Zwischenriume fast so breit wie die Streifen, feiner punktiert. Tibien gerade. Farbung briunlich rot; Fiihler, Tibien, und Tarsen, die Spitzen der Schenkel, Seitenrainder, Spitze, und Naht der Fliigeldecken schwarz gefarbt, die Fliigeldecken im iibrigen hellbraun. Be- haarung der Decken kurz und etwas aufstehend. Seitenteile der Hinterbrust dicht anliegend silberweiss behaart. Lange, 3.5 bis 4 Millimeter. PENANG, 1 Parchen. Deporaus nigriceps sp. nov. Von Deporaus penangensis, dem die Art sehr nahe steht, hauptsachlich in folgenden Merkmalen verschieden: Riisseb ktirzer als der Kopf, kurz vor der Basis kriaftig nach unten geknickt; glanzend und unpunktiert. Das erste Glied der Fiihlerkeule so lang wie das letzte Glied, Keule deutlich ab- 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 895 gesetzt. Halsschild breiter als lang. Zwischenriiume der FIlii- geldecken nicht ganz so breit wie die Streifen. Farbung rotgelb. Mittel- und Hinterbrust, der Kopf, die Spitzen der Klauenglieder und die Rander sowie die schmale Naht der Decken schwarz. Behaarung der Decken greis, kurz, und teilweise erhaben. Unterseite dichter und lang anliegend silbergreis behaart. Lange, 3.4 Millimeter (s. r.). BoRNEO, Sandakan, 1 Exemplar. Deporaus apicalis sp. nov. Kopf massig stark und dicht punktiert, Augen ziemlich kraf- tig vorgewolbt. Schlafen viel kiirzer als der halbe Augenlings- durchmesser; parallel und dann plotzlich tief und schmal ein- geschnirt; Hals parallelseitig und seitlich schwach gerundet. Riissel etwas linger als der Kopf, schwach gebogen; seitlich stark punktiert; Spitze etwas verbreitert. Fiihler wenig hinter der Riisselmitte eingelenkt. Schaftglied dreimal so lang wie breit; das erste Geisselglied langlich oval, gut doppelt so lang ° wie breit; das zweite Glied etwas linger; das dritte Glied so lang wie das erste; das vierte, sechste, und siebente Glied wieder etwas kiirzer als das dritte, das fiinfte Glied am kiirzesten, etwa anderthalbmal so lang wie breit. Keule ziemlich gestreckt; das erste und zweite Glied fast doppelt so lang wie breit, das dritte Glied langer und kraftig zugespitzt. Halsschild so lang wie breit, stark rundlich, vorn viel schmaler als an der Basis, ziemlich fein und dicht punktiert; Scheibe linienférmig langs- gefurcht. Basis breit gerandet, Vorderrand sehr fein gerandet und bewimpert. Fliigeldecken nach hinten schwach geradlinig verschmalert. Punktreihen missig stark, Zwischenriume brei- ter als die Streifen. Vordertibien gerade, Mittel- und Hinter- tibien wenig gebogen. Farbung rotgelb; Riissel oben, Vordertibien, Spitze der Hin- tertibien, und die Tarsen, die Fiihler, einzelne Flecken auf der Hinterbrust, die Spitzenhalfte und die Seiten der Fliigeldecken schwarz. Behaarung der Decken kurz greis, fast anliegend; Unterseite silberweiss, doch weniger dicht behaart. Linge, 4 Millimeter. BORNEO, Sandakan, 1 Exemplar. Deporaus taeniatus sp. nov. Kopf langer als breit, ziemlich kraftig und dicht punktiert; Augen weniger stark gewélbt, Schlifen langer als der halbe Augenlangsdurchmesser. Riissel viel linger als der Kopf, gebo- 396 The Philippine Journal of Science 1922 gen, glinzend, seitlich fein punktiert, zwischen der Einlenk- ungsstelle der Fiihler mit Lingsgriibchen. Fihler in beiden Geschlechtern hinter der Riisselmitte eingelenkt. Schaftglied gestreckt, fast viermal so lang wie breit; das erster Geissel- glied linglich oval, gut halb so lang wie das Schaftglied; das zweite und dritte Glied etwas linger als das erste, schwacher, ver- kehrt kegelférmig; die naichsten Glieder allmahlich an Lange abnehmend, beim Weibchen etwas gedrungener als beim Mann- chen; letztes Glied noch deutlich linger als breit. Keule ziem- lich gedrungen, das erste und zweite Glied langer als breit, das dritte Glied am lingsten. Halsschild linger als breit, stark konisch, seitlich schwach gerundet; vorn viel schmaler als an der Basis; ziemlich fein und dicht punktiert. Fliigeldecken seitlich parallel; Punktstreifen massig stark, tief eingerissen ; Zwischenriume breiter als die Streifen, dicht fein punktiert. Farbung schwarz; Fiihler rotgelb oder dunkelbraun, ebenso die Beine mit Ausnahme der Vordertibien und -Tarsen, die ‘ Vorder- und Mittelhiiften, das Abdomen und ein vor der Basis der Fliigeldecken befindliches, nach innen verschmalertes und durch die Naht schmal unterbrochenes Querband rotgelb ge- firbt. Behaarung der Decken greis, kurz, wenig erhoben. Die Seitenteile der Mittel- und Hinterbrust sowie das Schildchen anliegend silberweiss behaart. Lange, 3 Millimeter. BorNEO, Sandakan (13940/41), 1 P&rchen. Deporaus rugiceps sp. nov. Kopf sehr seicht eingeschniirt, sehr stark und runzlig punk- tiert. Stirn breiter als der Riissel an seiner schmalsten Stelle; gefurcht. Riissel kraftig, so lang (Mannchen) oder etwas langer (Weibchen) als der Kopf, fast gerade; zur Spitze kraftig verbreitert. Fiihler kurz hinter der Mitte des Riissels einge- lenkt; sehr diinn, gestreckt. Schaft- und erstes Geisselglied kraftig, anderthalbmal so lang wie breit; zweites Geisselglied fast so lang wie Schaft- und erstes Glied zusammen; die nidchs- ten Glieder allmahlich und wenig an Lange abnehmend; das siebente Glied noch mehr als doppelt so lang wie breit. Auch die Keule ist langgestreckt; das erste und dritte Glied etwa so lang wie das zweite Geisselglied, das mittlere ein wenig kiir- zer. Halsschild kaum langer als breit, seitlich schwach gerundet, zum Vorderrand etwas mehr als zur Basis verschmilert, an ersterem nicht deutlich, an letzterer sehr fein gerandet. Punk- tierung von gleicher Starke wie die des Kopfes, runzlig. Flii- 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 397 geldecken schmal. Punktstreifen breit und tief, Zwischenriume sehr schmal. Farbung schwarz, Abdomen dunkelbraun. Beine teilweise etwas heller braun. Fliigeldecken oben dunkelbraun. Behaar- ung auf den Decken fast anliegend; unten dicht silberweiss, auf dem Abdomen langer abstehend behaart. Lange, 4 Millimeter. SINGAPORE (13927), 1 Parchen. Deporaus nigrifrons Heller. Luzon, Los Bajios. - Deporaus dimidiatus Heller. LUZON, Mount Maquiling. Deporaus longiceps sp. nov. Kopf viel linger als breit und dadurch von den verwandten Arten leicht zu unterscheiden; Scheitel kriftig und dicht, hinten etwas feiner punktiert. Schlafen parallel und fast so lang wie der Augenlangsdurchmesser. Augen missig stark vorgewilbt, Kopf mit Augen kaum breiter als das Halsschild an seinem Vor- derrand. Riissel so lang wie der Kopf, gebogen; auf der Basal- halfte scharf gekielt, zwischen der Fiihlereinlenkung, die in der Mitte des Riissels erfolgt, gefurcht. Fiihler lang und schlank. Schaftglied etwas linger und kraftiger als das erste Geisselglied, letzeres. linger als breit, oval; zweites Glied erheblich linger als das Schaftglied; das dritte. bis sechste Glied in der Linge wenig verschieden, doch kiirzer als das zweite Glied ; das siebente Glied so lang wie das erste. Keule gestreckt, das erste und zweite Glied gleich lang, iiber doppelt so lang wie breit; das dritte Glied am langsten. Halsschild etwas linger als breit, seitlich schwach gerundet, der Vorderrand schmaler als die Basis. Punktierung missig kraftig und dicht, doch nicht runzlig. Vorderrand ziemlich breit und tief, Basalrand fein gerandet. Fliigeldecken seitlich parallel, hinter den Schultern ein wenig eingezogen. Punktstreifen sehr kriftig, sodass die Zwischenraume nicht zur Entwickelung kommen. Vorder- und Hintertibien gerade, Mitteltibien schwach gebogen. Farbung des Riissels und des Kopfes (mit Ausnahme der Sei- ten), des Halsschilds, der Basis der Decken, der Hiiften und teilweise der Schenkel rotgelb bis rétlich braun; im iibrigen schwarz gefarbt. Behaarung der Decken vorn sehr kurz, hinten langer anliegend. Unterseite anliegend silberweiss behaart, doch verdeckt die Behaarung nicht den Untergrund. Lange, 4 Millimeter. BoRNEO, Sandakan (13939), 1 Exemplar. 398 The Philippine Journal of Science 1922 Deporaus signatus sp. nov. Kopf langer als breit, fein und entfernt stehend punktiert, sehr seicht eingeschniirt, Stirn ungefurcht. Augen massig stark vorgewolbt; Schlafen parallel, seitlich schwach gerundet, nicht ganz so lang wie der Augenlingsdurchmesser. Riissel wenig langer als der Kopf, schwach gebogen; oben glatt, seitlich punktiert. Fiihler in der Mitte des Riissels eingelenkt. Schaft- glied und erstes Geisselglied gleichlang, gedrungen, kriftig, lan- ‘ger als breit; das zweite Glied langer als das erste Glied; das dritte und vierte Glied gleich lang, etwas kiirzer als das zweite Glied; das fiinfte bis siebente Glied in der Lange wenig verschieden und wiederum etwas kiirzer als die vorherge- henden zwei. Keulenglied 1 und 2 von gleicher Linge, andert- halbmal so lang wie breit, leztes Glied etwas langer. Hals- schild konisch, vorn also viel schmaler als an der Basis und seitlich miassig gerundet. Am Vorderrand nicht, Basalrand sehr fein gerandet. Punktierung fein und massig dicht. Fii- geldecken seitlich parallel, hinter den Schultern seicht einge- zogen. Punktstreifen mdssig kraftig, Zwischenriume flach und so breit wie die Streifen, fein und dicht punktiert. Tibien schwach gebogen, Mitteltibien kraftig und gedrungen. Farbung braunrot. Fiihlerkeule schwarzlich; Mittel- und Hintertibien und -tarsen braunschwarz. Ein breiter Mittel- streifen auf dem Halsschild ist tiefschwarz. Behaarung der Decken gelblich und fast anliegend; Unterseite sehr kurz und diinn, auf den Hinterbrust-Epimeren etwas gereiht behaart. Linge, 3.8 Millimeter. BORNEO, Sandakan (13953), 1 Exemplar. Deporaus nigritibialis sp. nov. Kopf massig stark abgeschniirt; auf dem Scheitel kriftig, im iibrigen fein punktiert. Schlafen parallel und so lang wie der halbe Augenlangsdurchmesser. Augen vorgewélbt, Kopf mit Augen jedoch kaum breiter als der Halsschildvorderrand. Riissel wenig linger als der Kopf, im basalen Drittel gebogen. Filer in der Riisselmitte eingelenkt. Schaftglied wenig linger als das erste Geisselglied, dieses anderthalbmal so lang wie breit; das zweite Glied erheblich linger als das erste; die nachfolgenden Glieder in der Lange vom ersten Glied wenig verschieden. Keule lang und kraftig, das erste und zweite Glied derselben doppelt so lang wie breit, das dritte Glied etwas linger. Halsschild so lang wie breit, seitlich gerundet, am Vorderrand etwas mehr ver- schmalert als an der Basis, hier wie dort gerandet. Punktier- 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionidz) 399 ung kraftig und dicht, teilweise runzlig. Punktstreifen der Fliigeldecken sehr kraftig. Zwischenriume sehr schmal und fein punktiert. Farbung rotgelb bis rotbraun; Fiihler, Tibien, und Tarsen, sowie groésstenteils die Decken mit Ausnahme der Basis schwarz. Behaarung kurz, greis, fast anliegend. Unterseite sehr diinn und sparsam behaart. : Lange, 4.2 Millimeter. BASILAN, 1 Exemplar. Deporaus maculiger sp. nov. Kopf ziemlich fein und dicht punktiert, Stirn sehr seicht gefurcht. Augen gross, doch nur miassig stark gewédlbt. Schlafen parallel und dann zur kriaftigen Einschniirung verrun- det. Russel linger als der Kopf, schwach gebogen; an der Basis gekielt, seitlich punktiert ; zur Spitze schwach erweitert. Fiihler hinter der Riisselmitte eingelenkt; schlank und diinn. Schaft- und erstes"Geisselglied kriftig, letzteres etwas kiirzer als das erste und wenig langer als breit; das zweite und dritte Glied so lang wie das Schaftglied, gestreckt; die nichsten Glieder kiirzer. Keule diinn, das lezte Glied etwas langer als die vorhergehenden. Halsschild kaum langer als breit, seitlich schwach gerundet, am Vorderrande mehr als an der Basis verschmilert; an letzterer fein gerandet, Vorderrand ungerandet. Punktierung miassig stark und sehr dicht. Fliigeldecken nach hinten seitlich sehr schwach geradlinig verbreitert. Punktstreifen sehr kriftig, Zwischenraiume schmaler als die Streifen, fein punktiert. Der vorletzte Streifen vereinigt sich in der Mitte mit dem Rand- streifen, Zwischenraum schmal. Farbung gelbbraun. Riissel, Fiihler, ein Fleck auf dem Kopf, und Halsschild dunkler; ebenso die Partie um das Schildchen. Fliigeldecken seitlich hinten braunschwarz gefarbt. Behaarung auf den Decken wenig erhoben. Unterseite diinn und anliegend behaart. Lange, 2.5 Millimeter. BASILAN, 1 Stiick. Bei einem Zweiten Exemplar von Borneo, Sandakan, ist das Halsschild etwas runzlig punktiert, doch sind weitere Unter- schiede kaum festzustellen. Das Tier ist etwas grésser (3.2 Millimeter) (13937). Deporaus distinctus sp. nov. Kopf langer als breit, missig stark und nicht ganz dicht punktiert; Stirn seicht gefurcht. Augen gross, vorgewdlbt, 400 The Philippine Journal of Science 1922 Schlifen so lang wie der halbe Augendurchmesser, Einschniirung miissig. Riissel wenig langer als der Kopf, seicht gebogen; Basalhilfte stumpf gekielt, seitlich ziemlich kraftig punktiert. Fiihler des Mannchens in der Mitte, des Weibchens hinter der Riisselmitte eingelenkt. Schaft- und erstes Geisselglied gleich lang, etwa doppelt so lang wie breit; zweites Glied fast so lang wie die vorhergehenden Glieder zusammen ; das dritte Glied um ein Drittel kiirzer als das zweite Glied; das vierte Glied etwas kiirzer als das dritte; das fiinfte bis siebente Glied etwas langer als breit. Keule kraftig, das zweite Glied wenig kiirzer als das erste und wenig linger als breit. Halsschild kaum langer als breit, seitlich schwach und gleichmassig gerundet, am Vorder- rand wenig schmaler als an der Basis, an letzterer fein, an ersterer breiter wulstig gerandet. Punktierung kraftig und sehr dicht, die Mitte der Scheibe fein linienférmig gefurcht. Fliigeldecken seitlich parallel. Punktstreifen kraftig, Zwi- schenraiume schmaler als die Streifen und fein punktiert. Vor- letzter abgekiirzter Punktstreifen kurz und gebogen; dazwischen befinden sich noch zwei Punkte von der Starke der Punkte in den Reihen. Tibien gerade. Farbung rotgelb; Tibien und Tarsen braunschwarz, Fiihler schwarz. Behaarung gelblich, wenig erhoben. Unterseite sehr kurz und diinn behaart. Lange, 3.8 bis 4 Millimeter. MINDANAO, Kolambugan und Dapitan, 1 Parchen. Deporaus nigricornis Heller. LuZzON, Laguna, Los Banos. Deporaus nigricornis forma palawana f. nov. Wahrend bei der Nominatform das Halsschild runzlig punk- tiert ist, ist dies bei einem Tier von Palawan nicht der Fall. Die Beine sind ferner ganz schwarz und zwischen dem Rand- und Supplementstreifen befindet sich keine eingeschlossene Punktreihe. Médglicherweise handelt es sich hier auch um eine selbststandige Art; bei dem geringen oe lasst sich das allerdings schwer feststellen. PALAWAN, Puerto Princesa (5938), 1 Exemplar. Deporaus testaceus sp. nov. Kopf etwa so lang wie breit, Stirn seicht langsgefurcht, miis- sig stark und ziemlich dicht punktiert. Augen gross, vorge- wolbt. Riissel wenig linger als der Kopf, schwach gebogen, zur Spitze etwas verbreitert; oben glanzend, seitlich punktiert. Fiihler des Mannchens kurz vor der Spitze, des Weibchens in der 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionidz) 401 Mitte des Riissels eingelenkt. Schaftglied wenig linger als das erste Geisselglied, das zweite Glied am lingsten, fast doppelt so lang wie das erste; das dritte bis fiinfte Glied kiirzer und wenig in der Lange untereinander verschieden; das siebente Glied des Mannchens noch deutlich langer als breit, beim Weibchen so lang. wie breit. Keule nicht ganz so lang wie die Geissel, mittleres Glied etwas kiirzer als das erste und dritte Glied. Halsschild so lang wie breit, seitlich schwach gerundet, zur Basis wenig, zum Vorderrand kraftiger verschmilert; kraftig dicht runzlig punktiert. Fliigeldecken seitlich fast parallel. Punktstreifen kraftig, Zwischenriume sehr schmal, fein punktiert. Farbung rotlich gelb, Fiihler schwach gebraunt. Behaarung gelblich, fast anliegend, die Abdominalsegmente linger abstehend behaart. Lange 2.5 bis 2.8 Millimeter. BORNEO, Sandakan, 1 Parchen. Deporaus fusculus sp. nov. Kopf hinter den Augen kraftig und tief abgeschniirt; fein. punktiert, auf dem Scheitel zerstreuter, hinten dichter. Kopf mit Augen breiter als der Vorderrand des Halsschildes. Riissel in beiden Geschlechtern linger als der Kopf, beim Weibchen etwas langer als beim Mannchen; gebogen, an der Basis stumpf gekielt, im tibrigen oben glatt und nicht erkennbar punktiert, seitlich dagegen ziemlich kraftig und dicht punktiert. Fiihler des Weibchens kurz hinter der Mitte, beim Mannchen kurz vor dem basalen Drittel eingelenkt. ‘ Mandibeln innen tief doppelt. gezihnt, aussen sehr fein, beim Mannchen stumpfer doppelzah- nig. Schaftglied mehr als doppelt so lang wie breit, etwas linger als das linglich ovale erste Geisselglied; das zweite und dritte Glied fast walzenférmig, diinn, und so lang wie das Schaftglied ; das vierte und fiinfte Glied gleich lang und etwas ktirzer und kraftiger als das zweite und dritte Glied; das sechste und sie- bente Glied viel kraftiger und noch deutlich langer als breit. Keule kriftig und gedrungen. Halsschild etwa so lang wie breit, fein und missig dicht punktiert; seitlich gerundet, zum Vorder-_ rand viel stirker verschmiilert als an der Basis. Fliigeldecken seitlich fast parallel. Punktstreifen kraftig, Zwischenraume wenig schmaler, schwach gewolbt und fein punktiert. Vorder- tibien beim Minnchen innen sehr fein geziahnt. - Farbung rotbraun; Fiihler schwarz, ebenso die Tarsen und manchmal die Tibien und die Riisselspitze. Behaarung gelblich, 1892035 402 The Philippine Journal of Science 1922 sehr kurz und etwas aufgerichtet; Unterseite nur sehr diinn behaart. Linge, 3 bis 3.6 Millimeter. MINDANAO, Iligan, Dapitan, und Butuan (4431), 2 Mannchen und 3 Weibchen. Deporaus sulcifrons sp. nov. Kopf etwas breiter als lang; hinten fein und weniger dicht, vorn etwas kriftiger und dichter punktiert. Stirn etwas breiter als der Riissel an seiner schwichsten Stelle, ziemlich tief gefurcht. Einschniirung recht kriftig, Schlifen zugerundet, kiirzer als der halbe Augenlingsdurchmesser. Riissel in beiden Geschlechtern erheblich linger als der Kopf, seicht gebogen, zur Spitze massig verbreitert ; auf dem Riicken glatt, seitlich kraftig gereiht punk- tiert. Fiihler bein Mannchen in der Mitte, beim Weibchen hin- ter der Riisselmitte eingelenkt. Schaftglied nicht doppelt so lang wie breit; das erste Geisselglied so lang wie das Schaft- glied, langlich oval; das zweite und dritte Glied langgestreckt und gleich lang, langer als das erste Glied; das vierte bis siebente Glied ein wenig kiirzer und untereinander gleich lang. Das erste Glied der Keule so lang wie das Schaft- und erste Geissel- glied zusammen, das zweite Glied kiirzer; das dritte Glied so lang wie das erste. Halsschild ein wenig breiter als lang, seit- lich gerundet, zur Basis kaum verengt, fein gerandet; zum Vorderrand mehr verschmdlert, hier eingeschniirt und ge- wulstet. Punktierung missig kraftig und wenig dicht, die Mitte der Scheibe fein der Linge nach gefurcht. Fliigeldecken seit- lich fast parallel, nur unerheblich nach hinten verbreitert. Punktstreifen kriftig und breit, Zwischenréume sehr schmal und fein punktiert. Farbung rotgelb; Spitzen der Tibien und Tarsen gebraunt; Fiihler schwarz. Bahaarung gelblich, halbanliegend. Unter- seite wenig deutlich behaart. Lange, 3.5 bis 3.8 Millimeter. BORNEO, Sandakan. (13955), 1 Parchen. Dem Deporaus fusculus am niachsten stehend; wihrend aber bei diesem das Schaftglied mehr als doppelt so lang wie breit ist, ist es bei D. sulcifrons nur etwa anderthalbmal so lang wie breit; das Halsschild ist kraftiger punktiert und die Farbung heller als bei fusculus. Deporaus serratissimus sp. nov. Kopf kraftig abgeschniirt, Schlafen parallel und etwa halb so lang wie der Augenlangsdurchmesser; auf dem Scheitel ziem- 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 4038 lich kraftig und dicht, hinten feiner punktiert. Riissel wenig langer als der Kopf, schwach gebogen, seitlich kraftig reihig punktiert. Fiihler kurz hinter der Mitte eingelenkt. Schaft- und erstes Geisselglied etwa von gleicher Lange, gedrungen, linger als breit; das zweite Glied nicht deutlich langer als das dritte, erheblich langer als das erste Glied, die nachsten Glieder etwas kiirzer werdend; letztes Glied noch deutlich linger als breit. Keule kraftig; das erste Glied wenig linger als das mitt- lere, das letzte am langsten. WHalsschild so lang wie breit, auf der hinteren Halfte seitlich fast parallel, dann stark gerundet nach vorn verschmalert und vor dem Vorderrande eingeschniirt. Punktierung kriftig und namentlich seitlich verrunzelt. Flii- geldecken seitlich fast parallel; Punktstreifen ziemlich kriftig, Zwischenraume schmaler als die Streifen und fein punktiert. Tibien fast gerade, aussen sehr fein gezihnt. Farbung gelblich braun, Keule dunkler. Behaarung der Decken kurz, fast anliegend; Bauch abstehend behaart. Linge, 3 Millimeter. BORNEO, Sandakan, 2 Exemplare. Deporaus basilanensis sp. nov. Kopf so lang wie breit, ziemlich fein und dicht punktiert; Stirn seicht gefurcht. Augen klein, kraftig gewélbt; Schlifen parallel, mit kleinem Radius zur verhiltnissmissig wenig tiefen Einschniirung gerundet verengt; linger als der halbe Augen- lingsdurchmesser. Riissel des Mannchens so lang, des Weib- chens etwas langer als der Kopf, schwach gebogen, auf der Basal- halfte stumpf gekielt, seitlich kraftig punktiert. Fiihler des Mannchens in der Mitte, des Weibchens kurz hinter der Mitte eingelenkt. Schaft- und erstes Geisselglied gleich lang, langer als breit; das zweite und dritte Glied gleich lang und langer als das erste Glied, verkehrt kegelférmig; das vierte bis siebente Glied wiederum gleich lang und so lang wie das erste Glied. Keule kraftig, die Glieder gleich lang; die beiden ersten andert- halbmal so lang wie breit, das letzte schwacher. Halsschild so lang wie breit, seitlich schwach gerundet und am Vorderrand etwas mehr als an der Basis verschmilert, an letzterer fein, am Vorderrand wulstférmig gerandet. Punktierung massig stark und sehr dicht. Fliigeldecken seitlich parallel. Punktstreifen kraftig, Zwischenraume schmal und fein punktiert. Farbung briunlich rot; Fiihler, Tibien, und Tarsen schwarz. Behaarung auf den Decken kaum erhoben, Unterseite sehr diinn und kurz behaart. 404 The Philippine Journal of Science 1922 Linge, 3 bis 3.2 Millimeter. BASILAN, 1 Parchen. Deporaus discretus sp. nov. Kopf langer als breit, kraftig abgeschniirt, Sehlafen parallel, so lang wie der halbe Augenlangsdurchmesser. Stirn und Scheitel kraftig und dicht, im iibrigen fein und zerstreut punktiert. Augen ziemlich gross und vorgewdlbt. Riissel etwas linger als der Kopf, gebogen; oben glinzend, auf der Basalhalfte stumpf gekielt. Fiihler in der Riisselmitte eigelenkt, kraftig und gedrungen. Schaft- und ertes Geisselglied ungefahr gleich lang und gut anderthalbmal so lang wie breit; das zweite Glied so lang wie das erste; das dritte Glied um ein Drittel langer als das zweite Glied; das vierte bis siebente Glied so lang wie das erste. Glied 1 und 2 der Keule gleich lang, anderthalbmal so lang wie breit, letztes Glied linger. Halsschild so lang wie breit, seitlich ziemlich kraftig gerundet, zum Vorderrand etwas mehr als zur Basis verschmilert, doch nicht eingeschniirt. Basis fein gerandet. Punktierung massig stark und nicht ganz dicht; dazwischen dusserst fein mattiert punktiert. Fliigel- decken nach hinten missig verbreitert. Punktstreifen kraftig, Zwischenriume etwa so breit wie diese und fein punktiert. Farbung gelbbriunlich; Fiihler schwarz, doch sind die Wur- zeln und Spitzen der ersten Glieder gelbrot gefirbt. Behaarung kurz, halbaufgerichtet. Lange, 2.8 Millimeter. MINDANAO, Iligan, 1 Exemplar. Deporaus curtirostris sp. nov. Kopf etwa so lang wie breit, massig kraftig und nicht dicht punktiert. Stirn mit flacher Grube versehen; Augen ziemlich gross, kraftig vorgewélbt, beim Mannchen kriftiger als beim Weibchen; Kopf mit den Augen breiter als der Vorderrand des Halsschildes. Riissel des M&nnchens so lang, des Weibchens langer als der Kopf, auf dem Riicken stumpf gekielt, fast gerade, zur Spitze schwach verbreitert. Mandibeln des Minn- chens aussen nicht, des Weibchens fein gezahnt. Fiihler des Mannchens in der Mitte, des Weibchens im basalen Drittel ein- gelenkt. Mannchen: Schaftglied sehr kurz und kriftig, so lang wie breit; das erste Geisselglied wenig schwdicher und wenig langer als das Schaftglied; das zweite Glied am schwichsten, verkehrt kegelférmig, so lang wie das erste Glied; die nichsten Glieder kraftig und gedrungen; das siebente Glied so lang wie 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionidz) 405 breit. Keule kraftig und so lang wie die Geissel; das erste und zweite Glied gleich lang; das dritte Glied mit dem deutlich erkennbaren Endglied doppelt so lang wie das zweite Glied, auch etwas schwiacher. Weibchen: Fiihler etwas schlanker, letztes Geisselglied noch deutlich linger als breit; letztes Glied der Keule nur anderthalbmal so lang wie das zweite. Hals- schild kaum breiter als lang, seitlich missig gerundet, zum Vor- derrand staérker als zur Basis verschmilert. Punktierung massig stark, dicht; Scheibe mit feiner Lingsfurche. Punkte der Streifen auf den Fliigeldecken kraftig und dicht, Zwischen- raume schmal und gewilbt, fein punktiert. Farbung rotbraun; Fiihler, Tibien, und Tarsen schwarz. Be- haarung gelblich, diinn; auf den Decken und auf der Unterseite wenig erhoben, auf dem Halsschild sehr kurz aufgerichtet. Lange, 2.8 bis 3.6 Millimeter. BORNEO, Sandakan (13946), 2 Mannchen und 4 Weibchen. Deporaus affinis sp. nov. Kopf kraftig abgeschniirt, Schlafen fast so lang wie der halbe Augenlangsdurchmesser. Punktierung ziemlich kraftig und dicht, Stirn nur flach gefurcht. Augen gross und vorgewilbt. Riissel kurz und kraftig, kiirzer als der Kopf, geknickt. Fiihler kurz hinter der Riisselmitte eingelenkt. Schaftglied langer als breit; das erste Geisselglied wenig schwacher und so lang wie das Schaftglied; das zweite Glied am langsten, anderthalbmal so lang wie das erste; das vierte Glied deutlich etwas langer als das dritte Glied und die nichsten Glieder in der Lange wenig verschieden. Keule missig stark, das erste Glied anderthalbmal so lang wie breit, deutlich langer als das zweite und so lang wie das dritte Glied, an welchem ein Endglied deutlich sichtbar ist. Halsschild etwas linger als breit; die basale Halfte seitlich pa- rallel, dann nach vorn gerundet verengt und vor dem Vorderrand breiter gerandet; Basalrand schmal gerandet. Punktierung fein und dicht. Fliigeldecken seitlich parallel. Punktstreifen ziem- lich kraftig, Zwischenraume kaum schmaler als die Streifen. Vorletzter abgekiirzter Streifen, schwach gebogen, vereinigt sich erst hinter der Mitte der Decken mit dem Randstreifen. Farbung gelbbraun, Unterseite etwas heller als die Oberseite. Behaarung kurz und schwach erhoben, Unterseite sehr diinn behaart. Lange, 3 Millimeter. BoRNEO, Sandakan, 1 Exemplar. A406 The Philippine Journal of Science 1922 Deporaus monticola sp. nov. Kopf langer als breit, sehr fein und zerstreut punktiert; Stirn flach, kraftiger und dichter punktiert, breiter als der Riissel vor der Basis in der Nahe der Fiihlereinlenkung. Augen miassig vorgewolbt, Kopf mit Augen breiter 4ls das Halsschild; Schlafen linger als die Augen, zunichst parallel, dann konisch zugerundet. Kinschniirung kraftig, Hals schmal und kaum zur Basis verbrei- tert. Riissel kiirzer als der Kopf, geknickt; von der Fiihlerein- lenkung, seitlich gesehen, stark verschmilert; von oben gesehen von der Basis zur Einlenkungsstelle konisch verengt, dann zur Spitze stark verbreitert, hier doppelt so breit wie an der diinn- sten Stelle. Fihler etwas hinter der Mitte des Riissels einge- lenkt. Schaft- und erstes Geisselglied langlich oval, letzteres etwas kiirzer, das Schaftglied fast doppelt so lang wie breit; das zweite Glied langgestreckt, diinn, linger als das Schaftglied; das dritte Glied etwas kiirzer als das zweite, dagegen wenig langer als die nachfolgenden Glieder; das siebente Glied ist noch gut doppelt so lang wie breit. Keule ziemlich kriftig, gestreckt; das erste und zweite Glied jedes so lang wie das Schaft- und das erste Geisselglied zusammen; das dritte Glied langer. Halsschild langer als breit und nur so breit wie der Kopf; kraftig und sehr dicht punktiert; Seiten fast parallel, am Vorderrand kraftig eingeschniirt, Basis fein gerandet. Flii- geldecken langer als breit, nach hinten etwas verbreitert. Punktstreifen kraftig, Zwischenriume schmaler als die Streifen, feiner punktiert. Farbung gelblich braun, Unterseite und Beine rotgelb; Keule geschwirzt. Behaarung greis, halb aufstehend. Lange, 2.5 Millimeter. Luzon, Mount Banahao, 1 Exemplar. Deporaus impressipennis sp. nov. Kopf hinter den Augen nur sehr seicht eingeschniirt; fein und dicht punktiert. Augen gross, halbrund vorquellend. Riissel kraftig, ktirzer als der Kopf; von der Seite gesehen gerade, oben mit einem Buckel, an dem die Fiihler eingelenkt sind; zur Spitze massig verbreitert, auf der Basalhalfte doppelt gefurcht. Fiih- ler in der Mitte des Riissels eingelenkt. Schaftglied nur andert- halbmal so lang wie breit; das erste und Zweite Geisselglied so lang wie das Schaftglied, das erste Glied langlich oval, das zweite Glied verkehrt kegelférmig; das dritte bis siebente Glied in der Lange wenig verschieden, doppelt so lang wie breit. Keule langgestreckt, letztes Glied etwas linger als das erste und 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 407 zweite Glied. Halsschild seitlich kaum gerundet, konisch, zum Vorderrand viel starker verengt als zur Basis; vor dem Vorder- rand ringsum eingeschniirt, der Rand wulstig aufgeworfen. Punktierung fein und zerstreut. Schildchen viereckig, so lang wie breit; hinten tief dreieckig bis dicht zur Basis ausgeschnit- ten. Fliigeldecken etwas linger als breit, seitlich parallel, viel breiter als das Halsschild; Punktstreifen kraftig und tief, Zwisch- enriume sehr schmal, dachférmig, auf den schragen Seiten fein punktiert. Farbung rotgelb, Unterseite dunkler; Fliigeldecken briunlich rot. Behaarung der Unterseite sehr diinn und anliegend, der Fliigeldecken etwas dichter. Lange, 3.5 Millimeter. BORNEO, Sandakan (13944), 2 Tiere. Deporaus (Platyrhynchites) basalis sp. nov. Fig. 3. Kopf etwas langer als breit, kraftig abgeschniirt, ausserst fein und zerstreut punktiert, breiter als das Halsschild; Augen missig stark vorgewolbt. Riissel nicht ganz so lang wie der Kopf, an der Basis so breit wie die Stirn, vorn stark verbreitert, fast so breit wie lang. Von der Seite gesehen sehr schmal, dadurch flach gedriickt erscheinend; schrag abwarts gebogen, oben nicht skulptiert. Fiihler in nichster Nihe der Riisselbasis eingelenkt. Schaftglied schwach keulenférmig, fast doppelt so lang wie breit ; das erste Geisselglied langer als breit; das zweite Glied schwiicher, wohl dreimal so lang wie breit; das dritte Glied etwas kiirzer; das vierte bis siebente Glied in der Lange wenig verschieden, aber noch deutlich linger als breit. Keule missig stark, gedrungen. Halsschild wenig linger als breit, seitlich nur schwach gerundet; ziemlich kraftig runzlig punktiert, auf der Scheibe seicht lingsgefurcht. Basis fein gerandet. Fliigeldecken nach hinten gleichmissig schwach verbreitert. Punkt- streifen kraftig, Zwischenriume schmal. Tibien schlank, fast gerade, aussen fein gesiigt. Erstes und zweites durchgeh- endes Abdominalsegment vollstandig verschmolzen; nur seitlich bildet ein .. . pevoraus (Platyrhyn- feines Griibchen den rudimentaren Teil chites). basslis sp. nov. einer Trennungsnaht. Farbung gelbbraun; Hinterbrust und Tarsenspitzen dunkel- braun. Behaarung tiberall sehr fein und diinn. 408 The Philippine Journal of Science 1922 Linge, 2.4 Millimeter. LUZON, Mount Maquiling (5989), 1 einziges Exemplar. Deporaus (Hypodeporaus) conicirostris sp. nov. Kopf linger als breit, kraftig abgeschniirt, fein und sehr dicht punktiert; Augen wenig vorgewolbt, Schlafen so lang wie der Augenlingsdurchmesser. Riissel kaum so lang wie der Kopf, nur sehr schwach gebogen. Vor der Basis ist er, seitlich einge- schniirt. Fiihler beim Minnchen in der Mitte, beim Weibchen kurz hinter der Mitte des Riissels eingelenkt. Schaft- und erstes Geisselglied gedrungen, wenig langer als breit; das zweite Glied am langsten und schwichsten, linger als das erste Glied; die nichsten Glieder werden zur Spitze hin allmahlich ktirzer, das siebente Glied ist nur so lang wie breit. Keule gedrungen. Halsschild langer als breit, kraftig dicht und runzlig punktiert ; seitlich gerundet, zum Vorderrand stark verschmalert, an der Basis fein gerandet. Fliigeldecken linger als breit, nach hin- ten verbreitert. Punkte der Streifen kraftig, Zwischenraume ‘schmal und fein punktiert. Farbung schwarz, Mandibeln rétlich. Behaarung auf den Decken kurz, greis, halbaufstehend; Unterseite etwas linger .anliegend, doch nicht dicht behaart. Lange, 2 Millimeter. Luzon, Mount Maquiling und Los Bafios, 1 Parchen. Deporaus subrugaticollis sp. nov. Dem Deporaus conicirostris sehr ahnlich. Wahrend aber bei diesem die Punktstreifen der Fliigeldecken breiter als die Zwischenriiume und letztere sehr fein punktiert sind, sind sie bei der vorliegenden Art gleich breit; die Zwischenraume sind dicht und etwas starker punktiert. Das zweite Glied der Fiihler- keule ist langer als breit; der Kopf kraftiger und sehr dicht punktiert; Riissel etwas Manger als der Kopf; Halsschild so lang wie breit. _ Farbung schwarz. Behaarung wie bei dem vorigen. Lange, 2.5 Millimeter. BASILAN und BoRNEO, Sandakan (13936), 2 Exemplare. Deporaus (Pseudodeporaus) pullatus sp. nov. Kopf langer als breit, massig kraftig und dicht pambetteet wenig deutlich abgesetzt. Augen flach gewoélbt, wenig aus der Kopfwélbung vortretend; Schlifen kaum kiirzer als der Augen- langsdurchmesser, parallel. Augen auf der Stirn einander ge- nihert, sodass die Entfernung zwischen ihnen nur halb so breit 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionidz) 409 ist als der Riissel vor der Basis. Riissel kaum langer als der Kopf, schwach und gleichmassig gebogen. Fiihler in der Mitte des Riissels eingelenkt. Schaft- und erstes Geisselglied gleich lang, langer als breit; zweites Glied schwacher, verkehrt kegel- foérmig, aber von gleicher Lange mit den vorhergehenden. Die ndchsten Glieder etwas kiirzer, das letzte noch deutlich langer als breit. Keule ziemlich kraftig. Halsschild wenig langer als breit, seitlich nur schwach gerundet; Vorderrand nur wenig schmaler als die Basis, kraftig quergerunzelt. Fliigeldecken langer als breit, seitlich parallel. Punktfurchen ziemlich kraf- tig; Zwischenriume schmaler als die Streifen und fein, dicht punktiert. Farbung bliulich schwarz. Behaarung iiberall fein greis, an- liegend und wenig dicht. Lange, 3 Millimeter. PENANG, 1 Exemplar. Deporaus periscelis sp. nov. Kopf linger als breit, massig kraftig und dicht punktiert, nur undeutlich abgesetzt. Augen gross und kraftiger vorgewoélbt als bei Deporaus pullatus; Schlifen mehr als halb so lang wie der Augenlangsdurchmesser. Stirn gut halb so breit als der Riissel vor der Basis. Dieser linger als der Kopf, schwach gebogen, kurz vor der Spitze kriftiger verbreitert. Fiihler vor der Riis- selmitte eingelenkt; beim Mannchen weiter zur Spitze als beim Weibchen; Schaft- und erstes Geisselglied gleich lang und gleich kraftig, linger als breit; zweites Glied langgestreckt, fast so lang wie das Schaft- und das erste Geisselglied zusammen ; das dritte Glied kiirzer; das vierte und fiinfte Glied gleich lang und wenig kiirzer als das dritte Glied; das sechste und siebente Glied gleich lang und wieder etwas kiirzer als das vierte und fiinfte Glied. Halsschild wenig linger, missig stark quergerunzelt; seitlich schwach gerundet, beim Miannchen zum Vorderrand et- was mehr als zur Basis verschmilert; Basis fein, Vorderrand breiter gerandet. Fliigeldecken langer als breit, seitlich paral- lel; Punktstreifen massig kraftig; Zwischenraume breit, sehr dicht und fein punktiert. — Farbung pechbraun bis blau; Basalhalfte der Schenkel durch- scheinend gelb bis braun. Abdomen rotlich braun ( Nominat- form) oder schwarz wie der tibrige Korper (f. nigriventris n.). Behaarung der Unterseite dichter weiss, Fliigeldecken: gelblich oder greis behaart. Linge, 3 bis 3.5 Millimeter. 410 The Philippine Journal of Science 1922 Die Nominatform, 4 Exemplare, von LUZON, Los Bafios: Mount Limay. MINDANAO, Dapitan (7342). Die Variation von Bor- NEO, Sandakan (13935), 1 Exemplar. Genus DEPASOPHILUS novum Kopf halsartig abgeschniirt; Augen gewolbt, aus der Kopf- woélbung vortretend. Riissel gedrungen, schwach gebogen. Fiihler ungeknickt, 1l-gliedrig, Keule gestreckt. Fltigeldecken langer als breit, seitlich parallel, hinten einzeln abgerundet, rei- hig punktiert. Pygidium von den Decken vollstindig unbedeckt, das letzte Dorsalsegment jedoch bedeckt bis auf ein zwischen den inneren Abrundungsradien der Fliigeldecken liegendes Drei- eck. Vorderhiiften hart am Vorderrand des Prosternums steh- end. Alle Tibien gebogen, aussen fein reihig mit Sagezihnen besetzt; ebenso aussen die Schenkel auf der Spitzenhalfte fein reihig gehdéckert. Das erste Tarsenglied linger als das zweite, das dritte Glied gelappt, Klauen gespalten. Erstes und zweites durchgehendes Abdominalsegment verschmolzen. Abgekiirzter Nahtstreifen auf den Fliigeldecken fehlt, vorletzter Punkstreifen vereinigt sich mit dem Randstreifen in der Gegend der Hinter- hiiften. Hierher einstweilen die folgende Art: Depasophilus bakeri sp. nov. Kopf ziemlich kraftig abgeschniirt, auf dem Scheitel mit eini- gen kraftigen Punkten besetzt, im tibrigen jedoch fein und zer- streut punktiert. Stirn nur flach eingedriickt und so breit wie der Riissel vor der Basis. Augen des Mannchens stirker vorgewolbt und die Schlafen fast so lang wie der Augenlangsdurchmesser ; beim Weibchen stehen die Augen etwas weniger vor und die Schlafen sind kiirzer. Riissel des Mannchens so lang, des Weib- chens langer als der Kopf, nur schwach gebogen; oben glinzend und unpunktiert, seitlich beiderseits mit einer Punktfurche ver- sehen; zur Spitze ist der Riissel nur schwach verbreitert. Fiih- ler des Mannchens in der Mitte, des Weibchens hinter der Mitte eingelenkt. Schaft- und erstes Geisselglied wenig linger als breit; das zweite und dritte Glied gleich lang, gestreckt, etwas linger als das erste Glied; das vierte und fiinfte Glied etwas kiir- zer ; das sechste und siebente Glied noch langer als breit. Hals- schild kaum Janger als breit, fein und dicht punktiert, seitlich schwach gerundet, zum Vorderrand stark verschmilert; hier und an der Basis nicht erkennbar gerandet. Fliigeldecken seit- 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 411 lich parallel; Punkte der Streifen vorn kriaftiger als hinten; Zwischenraume so breit wie die Streifen, fein und dicht punk- tiert. Alle Tibien gleichmissig gebogen (die Hintertibien namentlich beim Weibchen jedoch schwicher) ; Schenkel und Schienen aussen sehr fein gesigt. Farbung rotbraunlich; Fiihler an der Spitze geschwiarzt. Behaarung nur diinn, gelblich, anliegend. Lange, 2.5 bis 3 Millimeter. BORNEO, Sandakan, eine Anzahl Exemplare. Ein Tier vom gleichen Fundort mit schwarzen Fiihlern, Tibien, und Tarsen, mit seichter abgesetztem Kopf, und etwas langerer und dichterer Behaarung midge als nigritibialis subsp. nov. bezeichnet sein. Genus EUGNAMPTUS Schoénherr Eugnamptus longicollis sp. nov. Kopf mit den grossen vorquellenden Augen breiter als das Halsschild, Schlafen kurz und zur Abschniirung zugerundet, Hals lang; Punktierung missig stark und dicht. Riissel ktirzer als der Kopf, gebogen, zur Spitze verbreitert; der Lange nach glanzend gekielt, seitlich rauh punktiert. Fiihler hinter der Mitte des Riissels eingelenkt. Schaftglied gut dreimal so lang wie breit, keulenfoérmig; das erste Geisselglied linglich oval, anderthalbmal so lang wie breit; das zweite und dritte Glied etwas kiirzer als das Schaftglied; das vierte und fiinfte Glied kiirzer, nur etwa doppelt so lang wie breit; das sechste und siebente Glied wieder linger, doch etwas kiirzer als das zweite und dritte Glied. Keule ziemlich kraftig, das zweite Glied etwas kiirzer als das erste Glied; das dritte Glied fast doppelt so lang wie das zweite Glied. Halsschild langer als breit, seitlich leicht gerundet, vor dem Vorderrand plétzlich abgeschniirt und hier ein Stiick parallel verlaufend; vor der Basis leicht eingeschniirt und dann etwas erweitert. Punktierung sehr kraftig und runz- lig. Fliigeldecken nach hinten gleichmassig schwach verbreitert. Punkte der Streifen kraftig und dicht;! Spatien kaum so breit wie diese und gewidlbt, weit entfernt stehend fein punktiert. Farbung braunlich rot; Fiihler schwarz, ebenso die Tibien und die Spitze der Decken. Unterseite pechbraun, Hiiften und Tar- sen rotgelb. Behaarung gelblich, doppelt; iiberall sehr lang aufstehend und kiirzer halbaufgerichtet. Lange, 4 Millimeter. BORNEO, Sandakan, 1 Exemplar. 412 The Philippine Journal of Science 1922 Eugnamptus flavicornis Sp. nov. Kopf mit den kraftig vorgewélbten Augen ein wenig breiter als das Halsschild, doch sind die Augen nicht ganz so gross wie bei longicollis; Punktierung sehr fein und zerstreut. Riissel etwas kiirzer als der Kopf, schwach gebogen, auf dem Riicken stumpf gekielt, seitlich punktiert ; zur Spitze verbreitert und vorn stark verrunzelt. Fiihler hinter der Mitte des Riissels eingelenkt, lang und diinn. Schaftglied verkehrt kegelférmig, gut viermal so lang wie breit; das erste Geisselglied langlich oval, kaum halb so lang wie das Schaftglied; Glied 2 bis 7 sehr diinn, fast walzenférmig, in der Lange wenig verschieden und etwas kiirzer -als das Schaftglied. Keule wenig kraftig, das letzte Glied nur unerheblich linger als das erste und zweite Glied. Beim Mann- chen sind die Fiihler etwas gedrungener und die Keule kraftiger. Halsschild kaum linger als breit, zum Vorderrand etwas kraf- tiger als zur Basis abgeschniirt; sehr stark und dicht runzlig punktiert, seitlich schwach gerundet. Fliigeldecken seitlich fast parallel. Punkte der Reihen kraftig und dicht, Zwischenraume schmal und gewilbt, fein punktiert. Hintertibien an der Spitze mit einem Dorn. > Farbung rétlich gelb. Spitzen der Decken und bei einem Exemplar auch die Seiten geschwiarzt. Behaarung gelblich, dop- pelt; lang abstehend und kiirzer halb aufgerichtet. Linge, 4.5 Millimeter. BoRNEO, Sandakan, 1 Parchen. Subgenus Eugnamptobius novum Wihrend bei saimtlichen echten Eugnamptus-Arten der vor- letzte Punktstreifen an der Seite der Fliigeldecken parallel dem Randstreifen bis zur Spitze lauft, ist er bei der zu dieser Unter- gattung gehérigen Art kurz vor der Mitte in der Nahe der Hinterhiiften abgekiirzt und vereinigt sich hier mit dem Rand- streifen. Die Spitze des Epistoms ist sichtbar und tritt tren- nend zwischen die Hinterhiiftsegmente und Fliigeldecken. Eugnamptus (Eugnamptobius) insularis sp. nov. Kopf langer als breit; Punktierung kraftig und runzlig. Be- haarung auf dem Scheitel fein anliegend, dazwischen langer aufstehend. Riissel so lang wie der Kopf, kriftig gebogen, auf dem Riicken glinzend gekielt, seitlich runzlig punktiert. Fih- ler kurz hinter der Mitte des Riissels eingelenkt; Fiihlerschaft- glied etwa dreimal so lang wie breit, schwach keulenférmig; das erste Geisselglied langlich oval, halb so lang wie das Schaftglied ; 21,4 Voss: Indo-Malayische Rhynchitinen (Curculionide) 413 das zweite und dritte Glied schwach, langgestreckt; die nachsten Glieder zur Spitze allm&hlich an Stirke zu- und an Linge ab- nehmend. Keule langgestreckt; das zweite Glied kiirzer als das erste; das dritte Glied am langsten. MHalsschild etwas linger als breit, seitlich schwach gerundet, vor dem Vorderrand abge- schniirt; Punktierung sehr kraftig und dicht. Fliigeldecken seitlich parallel. Punkte der Streifen kraftig und dicht, Spatien schmal und gewdlbt, feiner punktiert. Decken in der Hohe der Hinterhiiften etwas ausgeschweift, Spitze des Epistoms sichtbar und der vorletzte Streifen abgekiirzt. Episternum der Hinter- brust kriftig, dicht und runzlig punktiert.: Farbung blauschwarz, Fiihler gelbrot, Keule geschwarzt. Be- haarung gelblich, doppelt: lang abstehend, struppig und kirzer, halbaufgerichtet. Lange, 4.2 Millimeter. PENANG (13930), 1 Weibchen. ILLUSTRATIONEN TEXTFIGUREN Fig. 1. Karte der Philippinen mit eingetragenen Sammelpunkten. 2. Scolocnemus pilosiusculus sp. nov. 3. Deporaus (Platyrhynchites) basalis sp. nov. 415 THE PHILIPPINE JOURNAL OF SCIENCE VoL. 21 NOVEMBER, 1922 No. 5 PHILIPPINE PARASITES OF THE FAMILY TRIGONALIDAS By S. A. RoHWER Custodian of Hymenoptera, United States National Museum In 1874 Westwood described a species of Trigonalys from Min- danao. Since then no other species has been recorded from any of the Philippine Islands. Recently Prof. C. F. Baker has forwarded three specimens of trigonalids for identification. These were collected at Surigao, Mindanao, and represent two species. One of them is the species that Westwood described, and the other two represent a curious new genus of the sub- family Nomadininez, which has been named after Professor Baker and is described herewith. Genus BAKERONYMUS novum Genotype, Bakeronymus typicus sp. nov. ' Belongs to the subfamily Nomadinine Schulz and may be separated from the only other genus of the group by the number of antennal joints, shape of the head, etc. Male (female unknown).—Antenna 14-jointed, thickening apically, joints without tyloiden; head transverse, seen from in front more than two times as broad as high, smooth and shining; eyes very. large, occupying the entire side of head, their inner margins distinctly converging below; antennz inserted about one-fourth the distance from the lower margin of head, separated from each other by a distance twice as great as the distance from one antenna to inner eye margin; clypeus small, not defined dorsally, inserted between the antenne, so if its dorsal 189287 . 417 418 The Philippine Journal of Science 1922 margin were defined it would about touch a line drawn tangent to the dorsal margin of the antennal foss; labial palpi 2- jointed, the joints long; maxillary palpi 4-jointed, the basal joint very small, the three following elongate; mandibles large, with four inner teeth, inserted much dorsad of a line drawn tangent to the lower margin of the eyes, their fosse separated from the antennal fosse by a narrow strip of chitin; thorax distinctly sculptured; notauli complete; mesepisternum divided by a trans- verse suture; propodeal spiracle short oval; legs slender; hind basitarsus subequal in length with the following joints; venation similar to Nomadina as figured by Schulz; abdomen depressed, flat dorsally, somewhat angulate ventrally; sternite without protuberances. - Bakeronymus typicus sp. nov. Male.—Length, 7.75 millimeters. Third antennal joint dis- tinctly shorter than the fourth, which is longer than the fifth or sixth; fifth and sixth subequal in length; head with a broad median impression, which extends from the clypeus to the occiput; on the vertex this impression forms a broad U-shaped area; impression narrowest at the anterior ocellus, which is situated in it and occupies most of the width of the impression ; thorax coarsely coriaceous; suture in front of scutellum feebly foveolate; abdomen finely shagreened, especially apically, first two segments slender, the first nearly parallel-sided, the second gradually widening posteriorly. Black; scape, spot on mandi- bles, mark on clypeus, inner margins of eyes broadly and con- nected with an oblique line extending to lateral ocelli, an elongate spot above each antenna, posterior orbits, spot on occiput medianly, spot on dorsal angle of pronotum, spot on side of pronotum, anterior lateral angles of prescutum, scutellum except oblique lateral face, most of metanotum, posterior portion of propodeum, oblique line on mesepisternum, poorly defined bands of tergites 2 to 5 inclusive, poorly defined bands on sternites 2 to 4 inclusive yellow; legs yellowish, four posterior cox except lateral spot, most of four hind femora, apical two- thirds of hind tibize, and hind tarsi except base black; wings hyaline with a distinct cloud in the radial cell; venation pale brown. The paratype has a lateral spot on the first sternite and the bands on the third and fourth sternites interrupted medianly. The paratype also differs in venation, but it agrees so perfectly in other important characters that I believe the venation is ab- normal. The difference is: The presence of the second intercu- 21,5 Rohwer: Philippine Parasites of the Trigonalide 419 bitus, thus forming a small second cubital cell instead of a large one as in the type and the complete absence of the third inter- cubitus. Type locality—Surigao, Mindanao, Philippine Islands. Type.—Catalogue No. 24728, United States National Museum. Described from two males (one type) received from C. F. Baker. Lycogaster lachrymosa (Westwood). Trigonalys lachrymosa WESTWoopD, Thesaurus Ent. Oxoniensis (1874) 128, pl. 23, fig. 5. Lycogaster lacrimosa SCHULZ, Gen. Ins. fasc. 61 (1907) 12. This species was originally described from material collected by Semper in Mindanao. Professor Baker has forwarded a female from Surigao, Mindanao, which agrees well with the original description. SOME MALAYAN APHIDIDAi By RyoicH! TAKAHASHI Of the Entomological Laboratory, Department of Agriculture, Government Research Institute, Formosa The Malayan Aphidide recorded in this paper are represented by specimens sent to me by Mr. S. Kiyatake, who collected them in the vicinity of Johore. I here express my sincere thanks to Mr. Kiyatake for this material. Only about ten species of aphids have been recorded from the Federated Malay States, and most of them were described from Singapore by P. van der Goot.' THE KNOWN MALAYAN APHIDID/: Macrosiphoniella ¢itricola v. d. G. Oregma muiri v. d. G. Fullawayella varicola (v. d. G.). Oregma rhapidis v. d. G. Aphis bambusae Full. Oregma singaporensis v. d. G. Aphis malvoides v. d. G. Oregma sundanica v. d. G. Greenidea roepkei (v. d. G.). Cerataphis lataneae (Boisd.). Glyphinaphis bambusae v. d. G. SPECIES OF APHIDS COLLECTED BY S. KIYATAKE, IN THE VICINITY OF JOHORE, MALAY PENINSULA Aphis gossypii Glover. Many wingless viviparous females and nymphs of winged ones were collected on Hibiscus rosa-sinensis in September and De- cember, 1920. Hitherto unrecorded from the Malay Peninsula. Aphis medicaginis Koch. - Many wingless viviparous females were collected on a plant of the Leguminose in March, 1921. Hitherto unrecorded from this peninsula. Aphis shirakii Takah. Special Bull. Agri. Expt. Sta. Formosa 20 (1921) 58. Some wingless and winged viviparous females were collected on the leaves of Melastoma candidum in December, 1920. The specimens sent by Mr. Kiyatake exactly agree with the Formosan ones. In Formosa this species is very common. Hitherto unrecorded from the Malay Peninsula. 1 Tijdschrift voor Entom. 60 (1918) 112-127. : 421 422, The Philippine Journal of Science Genus SETAPHIS van der Goot Cerciaphis THEOBALD, Bull. Entom. Research 11* (1920) 70, is a synonym of Setaphis v. d. G. The species of this genus have been found in Java, Africa, and the Malay Peninsula, but no species is known to occur in Formosa. Setaphis viridis v. d. G. Contrib. 4 la Fauna des Indes néerland 1 * (1918) 158. Mr. Kiyatake collected some wingless viviparous females and a nymph of the winged form on an unknown plant in March, 1921. The specimens from Johore differ slightly from the original description as follows: Wingless viviparous female——Yellowish green; antenne pale yellowish green, with the apex of the fourth joint and the fifth dusky; legs and abdominal tubercles pale yellowish green; eyes dark reddish (color notes from specimens preserved in alcohol) ; body oval, lacking hairs; head provided with a few very small hairs; front wide; frontal tubercles almost absent; eyes very small, composed of three facets; antenne short, 5-jointed, im- bricated, with three or four very small hairs; the third joint lacking sensoria; the fourth with a circular apical sensorium of medium size; the relative length of joints as follows: III, 50-52; IV, 25-30; V, 37 (22+15)—41 (23+18). Rostrum somewhat stout, reaching the middle coxxz; body lacking lateral tubercles; spiracles not protuberant. Abdomen just above the cauda provided with a pair of long and slender, hornlike tuber- cles, projecting horizontally far beyond the caudal apex, which is a little shorter than the fifth antennal joint, striate, slightly curved, sharply pointed and almost twice as long as the hind tarsi, provided with a short bristle at the apex; cornicles very short, cone-shaped, of lachnid appearance, slightly striate, pro- vided with one or two very small bristles; cauda not distinct, with some rather long, very stout bristles; legs provided with a few short bristles, tarsi rather short, roughly imbricated, with very long bristles at the apex. Length of body, 1.9 millimeters ; antenne, 0.9. This species is hitherto unrecorded from the Malay Peninsula. CHRYSOMELIDEN DER PHILIPPINEN, III Von J. WEISE Herischdorf, i. Riesengebirge, Germany Die hier besprochenen Tiere erhielt ich durch Herrn C. F. Baker, College of Agriculture, University of the Philippines in Los Bafios, welcher seit mehreren Jahren griindlich und mit bestem Erfolge die dortigen Coleopteren sammelte und zur Zeit wohl einen grossen Teil der im nérdlichen Luzon vorkommenden aufgefunden haben diirfte. Lema globicollis Baly ab. albidipennis ab. nov. Mir liegt zwar augenblicklich kein ostindisches Stiick der Art zum Vergleiche vor, doch glaube ich nicht zu irren, wenn ich » die Philippinenform dazu stelle. Sie hat nicht dunkelblaue son- dern weissliche Fliigeldecken, mit einem braunen, metallisch blau iiberflogenen und auf die Nahtkante beschrankten Saume. Das Halsschild hat keine Seitengruben sondern dicht vor der Basis eine breite, missig tiefe Querrinne, die aussen verbreitert und verfiacht ist. Sie ist A4usserst dicht und fein punktirt, die Schei- be vor ihr auf einem weiten Raume iiber den Vorderecken, am Vorderrande und auf zwei bis drei Langsreihen in der Mitte stark punktirt, so dass nur zwei Langsstreifen iibrig bleiben, welche hinten dusserst fein punktirt, vorn fast glatt sind. LUZON, Laguna, Los Bajos. Lema concinnipennis Baly. Lema haemorrhoidalis WEISE, Horae Soc. Ent. Ross. (1889) 57. Ein sehr kleines Exemplar, bei dem das Halsschild sehr fein und ungleichmdssig punktirt und der Bauch mit Ausnahme des vorderen Zipfels am ersten Segmente bradunlich rot gefarbt ist. MINDANAO, Iligan. Lema torulosa Lacordaire. Bei einem vielleicht frischen Exemplare ist der K6rper rat- lich gelbbraun, nur Schienen und Tarsen sind angedunkelt. Die Art ist der folgenden ausserordentlich nahe verwandt. Luzon, Laguna, Los Bafios (288). 423 424 The Philippine Journal of Science 1922 Lema coromandeliana Fabricius. Diese Philippinenform (var. philippina), hat die Seitenstiicke der Mittel- und Hinterbrust stets so hell gefarbt wie die iibrigen Teile der Unterseite, und bei dem Miinnchen einfache Mittel- schienen. In der ab. rufipes sind auch die Beine rot, die Schie- nen an der Spitze leicht, die Tarsen stirker angedunkelt. LUZON, Laguna, Los Bajios, Mount Maquiling; Tayabas, Ma- linao. Lema cyanoptera Lacordaire. Die Stiicke sind 6 bis 7 Millimeter lang, dunkel rostrot bis rotgelb, die Spitze der Schienen und die Tarsen meist angedun- kelt. Fiihler blass gelb, an der Basis mehr oder weniger weit braunlich rot, Fliigeldecken metallisch dunkelblau bis bliulich griin, Seitenstiicke der Mittel- und Hinterbrust dicht weisslich behaart. Thorax linger als breit, hinter der Mitte mit einer feinen, deutlichen Querfurche, die an den Seiten in eine massig breite, glatte, oben von einer Kante begrenzte Quergrube abfallt und die Seiten tief einschniirt. Letztere erweitern sich davor bis zur vorderen Borstenpore nur massig in schwachem Bogen und sind dann schnell gerundet-verengt. Die Scheibe erscheint glatt, hat aber etwa vier weitlaufige und sehr unregelmissige Langsreihen von Piinktchen in der Mitte. Die Reihen der Flii- geldecken bestehen vorn, bis zu dem schwachen Quereindrucke hinter der Basis, aus stirkeren, weitlaufig gestellten, dahinter aus dichter stehenden feinen Punkten und gehen vor der Spitze, an den Seiten schon hinter der Mitte, in Punktstreifen iiber. Bei einer Abanderung von Cagayan auf Mindanao (4673), hat die Spitze der Fliigeldecken einen ziemlich breiten rotgelben Saum, dessen vordere Grenze einen concaven Bogen bildet; die Schienen sind mit Ausnahme der dussersten Basis schwirzlich, die Tarsen aber rotgelb, nur das Klauenglied angedunkelt — ab. semperi Jacoby. LUZON, Laguna, Los Bafios: Tayabas, Malinao, 670. Lacordaire gibt zwar die Lange der Lema cyanoptera mit 34 lin. = 8.3 Millimeter, und Jacoby die von semperit mit 2.5 lin. = 5.6 Millimeter an; trotzdem halte ich beide fiir dieselbe Art, die an dem verhiltnissmissig schmalen, an der vorderen Borst- enpore abgerundeten Halsschilde zu erkennen ist. *Ann. Belg. (1893) 267. 21,5 Weise: Chrysomeliden der Philippinen, III 425 Lema bakeri sp. nov. Sat elongata, flavo-ferruginea, nitida, pectore, abdominis basi plus minusve et femoribus posticis nigris, tibiis infuscatis. Pro- thorace pone medium transversim sulcato et valde coarctato, antrorsum fere rectilineatim dilatato, dorso medio subtilissime subseriatim punctulato, elytris subtiliter striato-punctatis, intra basin obsolete transversim impressis, serie abbreviata fere nulla. Long., 5.5 ad 7 mm. LUZON, Laguna, Los Bafios, Mount Makiling, Mount Banahao. Ab. a. Pedibus fulvis, femoribus posticis macula nigra. Ab. 6, juvenilis. Corpore subtus pedibusque omnino fulvis (1449). Ab. ¢, tincta. Ut in b, sed elytris (apice excepto) vittisque _ duabus abbreviatis prothoracis nigris (279). In der Farbung sehr verinderlich, sonst aber in Form, Srtbine, und Skulptur mit Lema cyanoptera iibereinstimmend, von ihr durch das tiefer eingeschniirte, davor stirker und fast gradlinig bis zur vorderen Borstenpore erweitert und hier fast abgestutzte Halsschild zu unterscheiden. Der Korper ist glinzend und hell gelblich rot, Seiten der Mittelbrust, die Hinterbrust, das erste und die Seiten der zwei oder drei Bauchsegmente nebst den Hinter- schenkeln schwarz, die Schienen und zuweilen auch die Spitze der Mittleschenkel angedunkelt. Bei der Abinderung a ist von der schwarzen Farbe der Hinterschenkel nur noch eine Langs- makel in der Basalhalfte der Aussenseite iibrig geblieben; sie leitet zu der hellsten Form, der ab. juvenilis iiber, deren Kérper einfarbig gelblich rostrot gefarbt ist. Die dunkelste Form, die ab. tincta, hat auf dem Thorax vor der Querfurche jederseits eine schmale schwarze Langsbinde, welche den Voderrand nicht erreicht; ebenso sind die Fliigeldecken schwarz, mit einem brei- ten, vorn ziemlich gradlinig begrenzten gelben Spitzensaume. Bei einem Stiicke ist die Unterseite nebst den Beinen einfarbig hell gefarbt; ein anderes hat eine schwarze Quermakel jederseits auf der Hinterbrust, sowie einen kleinen gerundeten schwarzen Fleck an der Aussenseite der Hinterschenkel vor der Spitze. Tabelle der Lema-Arien. 1. Augen nicht ausgerandet, Thorax stark punktirt, dicht vor der Basis durch eine schmal muldenférmige, an den Seiten verbreiterte und verflachte Querfurche schwach eingeschniirt, ohne Seitengruben. Ostindien, Philippinen L. globicollis Baly. 1. Augen tief ausgerandet, Thorax hinter der Mitte durch eine Querrinne, die sich jederseits zu einer Grube ausbildet, kraftig eingeschniirt.... 2. 426 The Philippine Journal of Science 1922 2. Auf dem glatten Thorax liegt vor der normalen Querfurche eine zweite, die oft in der Mitte verflacht oder unterbrochen ist.................2-...-..--- 3. 2. Halsschild nur mit der normalen Querfurche, vor dieser in drei bis vier unregelmassigen Liangsreihen fein oder dusserst fein punktirt........ 4, 3. Stirn gelblich behaart, Fliigeldecken rot oder rotbraun, einfarbig oder an der Seite mit einer schwarzen bis dunkelblauen Langsbinde, die sich nach hinten allm&hlich verbreitert und in zwei Drittel der Lange die Naht erreicht. Philippinen ...................... L. torulosa Lacordaire. 8. Stirn weisslich behaart, Fliigeldecken schwarz, violett, blau oder griin. Ostindien, Hinterindien, Sunda Inseln, Philippinen. L. coromandeliana Fabricius. 4, Fihler schwarz, Stirnbeulen punktirt und behaart, Seitengruben des Thorax punktirt, abgekiirzte Punktreihe der Fliigeldecken vertieft, aus starken Punkten gebildet, die von denen der benachbarten Reihen kaum verschieden sind; sie endet in einer grubenférmigen Vertiefung neben der Naht. China, Japan, Philippinen.... L. concinnipennis Baly. 4. Fihler gelb, Stirnbeulen kahl und nebst der Seitengrube des Thorax glatt, abgektirzte Punktreihe der Fliigeldecken kaum bemerkbar; sie besteht aus verloschenen Piinktchen unmittelbar neben der Naht und lauft bis zu dem schwachen Quereindrucke hinter der Basis fort........ 5. 5. Die Seiten des Halsschildes verbreitern sich vor der Einschniirung massig und sind an der vorderen Borstenpore abgerundet. Phil- EG, Oe SS Ba 2 sek Wie he te ee Se L. cyanoptera Lacordaire. 5. Halsschildseiten vor der Einschniirung fast gradlinig stark erweitert, sie bilden an der vorderen Borstenpore einen Winkel, der kleiner als ein rechter ist. Philippinen.. mh L. bakeri sp. nov. Crioceris distigma sp. nov. Ferruginea, nitida, pectore, abdomine pedibusque quatuor posticis nigris, antennis sat brevibus flavescentibus, scutello glabro, elytris basi evidenter-, postice subtilius striato-punctatis, singulo ante medium puncto magno nigro. Long., 8 ad 9 mm. LUZON, Laguna, Los Bafids (275, 1586). Crioceris unipunctata Lacordaire. Crioceris unipunctata LACORDAIRE, Mon. (1845) 574; BALY, Trans. Ent. Soc. London (1865) 35. ; Ab.a@. impicta. Puncto nigro elytrorum nullo. Ab. b. Ut in cto sed pectore, abdomine pedibusque ferrugineis. Die Crioceris unipunctata Fabricius von Java hat die Fiihler, Unterseite, und die vier verdickten Hinterschenkel schwarz, kann also nicht die Philippinen-Art sein. Crioceris luzonica sp. nov. Ferruginea, nitida, antennis sat elongatis, nigris, tibiis infus- eatis. Prothorace subcylindrico, medio eo arctato, dorso laevi uniseriatim punctulato, scutello glabro, elytris striato-punctatis, 21,5 Weise: Chrysomeliden der Philippinen, III 427 intervallis laxe uniseriatim punctulatis, vix rugulosis, externis convexiusculis. Long., 6.5 mm. LuZzON, Mount Banahao. Die kleinste Art der Philippinen und die einzige bei welcher die Punktreihen der Fliigeldecken hinter dem leichten Querein- drucke nur wenig schwicher als davor, in ihrer ganzen Lange sehr deutlich, die Zwischenstreifen mit einer weitlaufigen Reihe feiner Piinktchen und mit kleinen verloschenen Runzeln versehen sind, wodurch die Scheibe einen geringeren Glanz erhiilt. Rostrot, die Schienen, das letzte Tarsenglied nebst den Klauen, und ein unbestimmter Fleck auf dem Riicken der vier Hinter- schenkel schwarzlich, die Fiihler schwarz, Glied 1 pechbraun. Die Fiihler sind etwa halb so lang als der Korper und die einzelnen Glieder vom fiinften ab viel langer als breit. Thorax cylindrisch, in der Mitte eingeschniirt, an der vorderen Borsten- pore abgerundet und hier kaum so breit wie an der Basis, glatt, vorn mit einer ziemlich weitlaufigen Langsreihe von Piinktchen. Crioceris gracilicornis sp. nov. Ferrugineo-rufa, nitidissima, abdomine antennisque flaves- centibus, his in basi, femoribus in disco apicem versus, tibiis tarsisque infuscatis, scutello pubescente, elytris ante medium nigris, laevibus, basi striato-punctatis. Long., 7, mm. MINDANAO, Agusan, Butuan, Durch schlanke Fiihler ausgezeichnet, die wenigstens zwei Drittel der Koérperlainge erreichen, ziemlich diinn und nur leicht zusammengedriickt sind; ihre Glieder 5 bis 8 sind eine Spur brei- ter wie die vorhergehenden und folgenden Glieder, das zweite ist sehr klein, 3 fast doppelt so lang, 4 wenig langer, die iibri- gen unter sich fast von gleicher Lange, jedes ziemlich so lang wie 3 und 4 zusammen. Von ahniicher Képerform als Crioceris distigma, nur das Halsschild vorn bedeutend breiter. Rot, stark glanzend, Bauch und Fiihler gelblich, Glied 1 und 2 der letzteren rotbraun, 3 und 4 schwirzlich, der Riicken der Schenkel nahe der Spitze nebst Schienen und Tarsen angedunkelt, wenig mehr als die vordere Halfte der Fliigeldecken schwarz. Thorax quadratisch, in der Mitte durch eine glatte Quergrube jederseits eingeschniirt, auf der Scheibe glatt, mit einigen kaum sichtbaren Piinktchen die in der Mitte zwei unordentliche Langsreihen bilden. Schildchen behaart. Fliigeldecken hinter der Basis leicht quer vertieft, glatt, nur die abgekiirzte und die vier folgenden Punktreihen vor der Mitte vorhanden. 428 The Philippine Journal of Science 1922 Crioceris ? nucea Lacordaire. Ich bin im Zweifel ob die Exemplare von Los Bafios und dem Mount Maquiling zu dieser Art geh6éren, da sie nur 8 Millimeter lang, hinter der Schulter etwas eingeschniirt, und auf dem Schildchen, der Mitte der Vorderbrust und den Seitenstiicken der Mittel- und Hinterbrust dicht gelblich weiss behaart sind. Eine Deutung auf die hellbeinige ab. b. von Crioceris distigma, zu der die Beschreibung verleiten kénnte, ist ausgeschlossen, weil Lacordaire auf dem Thorax zwei Liangsreihen von Piinktchen erwahnt. Crioceris philippinensis Jacoby. Von Herrn Baker auf dem Mount Banahao (2422) und bei Baguio, Benguet (6102) gesammelt, ist die bis jetzt bekannte grésste Art der Philippinen und an den schwarzen Schienen, Tarsen, und Fiihlern (nur das Basalglied ist gelbich rot), sowie dem weiten Quereindrucke der Fliigeldecken leicht kenntlich, welcher die gegend an der Basis bedeutend emporhebt und tiefer wie bei den verwandten Arten ist. Von den beiden Punkt- reihen des Thorax lassen sich unter starker Vergrésserung kaum noch einige Piinktchen vor der Mitte erkennen, auch die Fliigel- decken sind glatt bis auf einige Reste der ersten vier oder fiinf Punktreihen in und nahe dem Quereindrucke. Ein Mannchen von Baguio, Provinz Benguet, hat stark gebogene Mittelschienen und einfarbig schwarze Fiihler. Jacoby’s Abbildungen? von Crioceris philippinensis und anderen Arten sind wahrscheinlich von dem Zeichner des Briis- seler Museums angefertigt, der auch meine Tafeln zu den Hispinen* gezeichnet hat. Sie sind unzuverliassig und nicht viel besser als die bekannten Groschen-Bilderbogen aus+ Neu Ruppin bei Gustav Kiihn. Alle thier besprochenen Arten haben die Seitenstiicke der Hinterbrust dicht weisslich oder gelblich behaart, wahrend daneben die Brust glatt ist. Tabelle der Crioceris-Arten. 1. Schildchen kahl, der Thorax hat ausser einigen Piinktchen an den Seiten nur eine Reihe dusserst feiner Punkte in der Mitte... a. 1. Schildchen behaart, Thorax in der Mitte wenigstens mit zwei Reihen ausserst feiner, zuweilen schlecht bemerkbarer Punkte................ 8. 2. Fithler gelb, kurz, Glied 5 das lingste, 8 bis 10 jedes nur so lang als breit oder wenig linger, die Punktreihen der Fliigeldecken auf der inneren Halfte vor der Mitte deutlich, aussen und hinten ziemlich erloschen. Linge, 8 bis 9 mame...e.c.-cccccccicccscccc C. distigma sp. nov. *Genera Insectorum 23 (1904). ‘* Heft 125 (1911). 21, 6 Weise: Chrysomeliden der Philippinen, III 429 2. Fiihler schwarz, lang, Glied 5 bis 10 unter sich fast gleich, jedes viel langer als breit, die Punktreihen der Fliigeldecken bis zur Spitze deutlich und die 4usseren Intervalle leicht gewolbt. Linge, 6.5 CEE Lg a ea tS) keke PUN er aces fled: .. C. luzonica sp. nov. 3. Fihler sehr schlank,.gelb, die Basis dunkel, von den Gliedern 5 bis 11 jedes mehr als doppelt so lang wie breit. Lange, 7 mm. ; C. gracilicornis sp. nov. 3. Fihler kraftig, von den Gliedern 5 bis 11 jedes wenig langer wie breit.. 4. 4, Fihler gelb, kaum zusammengedriickt, ihre Glieder vom fiinften ab cylindrisch, dicht behaart, matt, Stirnbeule kurz dreieckig. Lange, 8 mm..... Sispsviaslibcpiwictscindsatscokemacetoes Cc. ? nucea Lacordaire. 4, Fihler schwarz (Glied 1 oft rotgelb), deutlich zusammengedriickt, Glied 1 und 2 fast glatt, glinzend, die folgenden dicht behaart, matt, 5 bis 11 jedes von der Basis zur Spitze erweitert. Stirnbeule lang. BT a acs oh hig cn adem C. philippinensis Jacoby. Pedrillia luzonica sp. nov. Elongata, subparallela, testaceo-flava, antennis ab articulo tertio nigris, tibiis tarsisque infuscatis, prothorace elytrisque creberrime punctatis et pubescentibus. Long., 4.5 mm. LuZON, Mount Maquiling, Laguna (6167). Schlank gebaut und fast parallel, hell rétlich gelb, Schienen und Tarsen leicht angedunkelt, die Fiihler vom dritten Gliede ab schwarz, normal gebaut.. Glied 1 das lingste, 2 das kleinste, 3 etwas kiirzer als 1, 4 und die folgenden allm&hlich wenig kiirzer. Hals glatt, Stirn zwischen den Augen missig dicht, sehr fein punktirt, und weisslich behaart. Halsschild nahe der Basis eingeschniirt, ohne Quereindruck, die Seiten davor gerundet, die Scheibe dusserst dicht und fein punktirt, dicht, sehr fein anlie- gend weisslich behaart, die Harchen nach hinten gerichtet. F'lii- geldecken etwas breiter als der Vorderkérper, dusserst dicht und eine Spur runzelig punktirt, A4hnlich wie der Thorax aber viel kraftiger behaart. Klauen mit Basalzahn. Aspidolopha manilensis Weise. | Luzon, Mount Banahao (2880). Aspidolopha bakeri sp. nov. Oblongo-ovalis, subtus coeruleo-nigra, sat dense brevissimeque cinereo-pubescens, palpis antennisque nigris, his basi testaceis, capite aeneo, clypeo parce punctulato, vertice magno, convexo, laevi, prothorace, scutello elytrisque laete ferrugineis, nitidis, his crebre subtiliter hinc inde subseriatim punctatis. Long., 6.5 mm. Luzon, Mount Banahao (2774, 4738). An der glinzend und lebhaft rostroten Farbe der Oberseite zu erkennen, nur der Kopf ist metallisch dunkelgriin, die Unterseite nebst den Beinen blaulich schwarz, ziemlich dicht und 4380 The Philippine Journal of Science 1922 zart greisbehaart. Fiihler schwarz, ihre drei ersten Glieder rét- lich gelbbraun. Kopfschild sparsam und verloschen punktulirt, der grosse, gewolbte Scheitel glatt, der dazwischen liegende drei- eckige Raum an den Augen dusserst dicht und fein runzelig punk- tirt, behaart. Halsschild und das grosse dreieckige Schildchen glatt, die Fltigeldecken dicht, fein, und namentlich auf der in- neren Halfte unregelmassig gereiht-punktirt. Das Pygidium ist im oberen Teile gleichmassig und schwach gewdlbt, die Spitze nach vorn umgeschlagen. Aspidolopha semperi Lefévre. Verhaltnismassig schlank gebaut, schwarz, Halsschild und Fliigeldecken gelblich rot, ersteres zuweilen mit einer Querreihe von vier dunklen Fleckchen, die Fliigeldecken mit einer kleinen schwarzen Schultermakel und einem grossen Querflecke hinter der Mitte an der Naht, vom Seitenrande viel weiter entfernt. Fiihler und Beine roétlich gelbbraun, die Schenkel meistens, zuweilen auch die Schienenspitze angedunkelt. Bei den mir zu-. ganglichen Stiicken ist der Clypeus glatt und hat jederseits zwei Gruben am Rande, eine gréssere und tiefe oben, eine kleinere und sehr flache darunter. Da Lefévre die Art, “late ovata’ und den Clypeus “remote fortius punctatus’ nennt, habe ich vielleicht nicht die richtige Aspidolopha semperi vor mir. Luzon, Laguna, Mount Maquiling (1595). Aspidolopha congrua sp. nov. Oblongo-ovata, subtus testaceo-flava, griseo pubescens, meta- sterno antennisque fuscis, his in basi, ore, pedibus, prothorace elytrisque fulvis, nitidis, capite supra antennas nigro-aeneo, scutello maculisque quatuor in elytro singulo crebre subtilissime punctato nigris, 2, 2 collocatis. Long., 5 mm. PALAWAN, Puerto Princesa. Kleiner wie die vorige, auf den Fliigeldecken feiner punktirt und mit je vier schwarzen Flecken versehen; der erste, auf der Schulterbeule, fast doppelt so lang als breit, der zweite, im verloschenen Eindrucke hinter der Basis, quer, von der Naht und Makel 1 ziemlich gleichweit entfernt, die beiden anderen in einer graden Querreihe hinter der Mitte, dreieckig, mit der Spitze einander zugekehrt, 3 neben der Naht, 4 nahe dem Seitenrande. Es ist sehr wahrscheinlich, dass diese Makeln bei ihrer Vergrés- serung in zwei Querbinden zusammenfliessen. Der Kopf ist iiber der Fiihlerwurzel metallisch griinlich schwarz, darunter rotgelb, der gewodlbte Scheitel und der helle vordere Teil des Clypeus 21,5 Weise: Chrysomeliden der Philippinen, II 431 glatt, der dunkle Teil desselben dicht, der etwas vertiefte und zart behaart dreieckige Raum neben den Augen viel dichter und feiner punktirt. Thorax und Schildchen glatt, Fiihler angedunkelt, die ersten drei oder vier Glieder ahnlich wie die Beine rotgelb. Aspidolopha semicincta sp. nov. Sat oblongo-ovata, albido-lutea, subtus dense argenteo-pubes- cens, antennis (basi excepta), pectore, tibiis apice tarsisque nigricantibus, femoribus apicem versus, capite dense punctato, maculis binis prothoracis et tribus (1, 2) elytrorum praeterea limboque subtili pone medium nigro-aeneis. Long., 5.5 mm. LUZON, Mount Banahao. Ausgezeichnet durch die sehr helle Grundfarbe und die metallisch dunkelgriine Zeichnung des Kérpers, sowie durch die Skulptur der Stirn. Blass weisslich gelb, unterseits dicht und sehr kurz silberweiss behaart, Mittel- und Hinterbrust, die Spitz- enhalfte der Schienen, die Tarsen und Fiihler schwarzlich, die vier Basalglieder der letzteren rétlich gelb, die Seiten des Bau- ches leicht angedunkelt, der Riicken der Schenkel in der Spitzen- halfte, der Kopf, zwei Makeln des Thorax, das Schildchen, drei Makeln und ein feiner Seitensaum jeder Fliigeldecke hinter der Mitte metallisch dunkelgriin. Jede Makel des Thorax besteht aus einem grésseren Teile dicht vor der Basis, einem breiten Quer- streifen ahnlich, und einem kleineren Stiicke, das vor der inneren Vorderecke aus nach vorn und innen lauft. Die erste Makel der Fliigeldecken ist gross, quer, und nimmt die Schulterbeule und einen Raum dahinter ein; sie ist hinten leicht ausgebuchtet; die beiden anderen Makeln liegen in einer Querreihe unmittelbar hinter der Mitte. Sie sind kleiner, gerundet, unter sich etwas weiter entfernt wie jede vom Rande, die iussere hingt mit dem Spitzensaume zusammen. Augen und Stirn sind linger wie bei den vorigen Arten, die Stirn dicht, das Kopfschild weitlaufiger und etwas starker punktirt. Thorax glanzend, unter starkerer Vergrésserung dusserst dicht und fein punktulirt, mit einigen grosseren Punkten dazwischen. Schildchen glatt, Fliigeldecken sehr dicht, fein und teilweise gereiht-punktirt, weniger glinzend. Der Penis bildet eine gekriimmte, flach gedriickte Réhre, die am Ende gerundet-abgestutzt und mit einem kleinen winkeligen Ausschnitte versehen ist. | Bei dem Mannchen der Gattung Aspidolopha ist das letzte Bauchsegment in der Mitte abgeflacht und sparsam behaart, beim Weibchen hat es dort eine grosse und tiefe Eigrube. 432 The Philippine Journal of Science 1922 Tabelle der Aspidolopha-Arten. 1. Kérper fast einfarbig metallisch blau oder blaulich griin, Pygidium neben dem aufgebogenen Seiten- und Hinterrande muldenférmig vertieft, in der Mitte zu einem dreieckigen Wulst ansteigend, dessen héchste Stelle eine schwache Lingskante bildet ................ A. manilensis Weise. Hierher auch die heller gefarbte und stirker punktirte A. phil- ippinensis Lefévre. 1. Halsschild und Fliigeldecken gelb bis rostrot, einfarbig oder mit dunkler Zeichnung. Pygidium in der Mitte nicht wulstartig erhéht............ 2. 2. Bauch dunkel, schwarz oder blaulich bis griinlich schwarz.................... 3. 2; Bauch’ Rell, braunwsen pele oon ea 4, 8. Thorax, Schildchen, und Fligeldecken einfarbig gelblich rot, Beine schwarz und 4hnlich der Unterseite blaulich sutalfachinanernd, A. bakeri sp. nov. 3. Thorax und Fliigeldecken gelblich rot, in der Regel dunkel gezeichnet, Beine rotgelb, die Schenkel und zuweilen auch die Schienenspitze mehr oder weniger angedunkelt. = A. semperi Lefévre. 4. Augen und Stirn lang, verhaltnismassig schmal, letztere dicht punktirt, nur in der Mittellinie des Scheitels glatt........... A. semicincta sp. nov. 4, Stirn und Augen kurz, der grosse gewdlbte Scheitel und der vordere breite Querstreifen des Clypeus glatt.......0...... A. congrua sp. nov. Aetheomorpha palawanica sp. nov. Oblongo-ovalis, subtus flava, medio postpectoris pedibusque plus minusve infuscatis, supra fulva, nitida, antennis (basi ex- cepta), vertice, maculis quinque limboque laterali elytrorum ni- gris, his antice seriatum punctatis, postice sublaevibus. Long., 4 mm. PALAWAN, Puerto Princesa. ‘Langlich eiformig, unten gelb, die Hinterbrust (ohne die Sei- tenstiicke) schwarzlich, das Analsegment, der Riicken der Schenkel und die Spitzenhalfte der Schienen nebst den Tarsen leicht angedunkelt. Kopf und Halsschild rétlich gelb, Oberlippe und die Stirn tiber dem Clypeus schwarz, ebenso das Schildchen, welches aber hell gesiumt ist. Fliigeldecken etwas heller und mehr gelb als der Thorax, ihre Epipleuren, ein Saum dariiber und fiinf Makeln auf jeder schwarz; die erste kurz, quer, bildet einen Saum an der Basis, der innen an der vierten Punktreihe endet; die zweite und dritte liegen in etwa ein Drittel der Lange; 2 ist gemeinschaftlich, klein, gerundet; 3, etwas weiter nach hinten geriickt, bildet einen Querstreifen von der fiinften Punktreihe bis in den Seitensaum. Die beiden folgenden sind zu einer etwas schrég von innen nach hinten und aussen ziehenden und in der Mitte eingeschniirten Querbinde vereint, reichen bis drei Viertel der Lange und lassen eine gerundete gemeinschaftliche Quermakel vor dem schwarzen Hinterrande frei. Der Kopf ist glatt, der Clypeus bildet eine grosse sechseckige Fliche, die 21,5 Weise: Chrysomeliden der Philippinen, III 4383 durch eine Querrinne iiber der Mitte der Augen von der Stirn geschieden und davor winkelig eingedriickt ist. Thorax doppelt so breit wie lang, mit gerundeten Seiten nach vorn verengt, oben fast glatt, in dem Basaleindrucke vor dem Schildchen einige feine Punkte. Fliigeldecken nach hinten etwas verbreitert, ziem- lich kraftig in nicht ganz regelmiissigen Reihen punktirt, die von der Mitte ab dusserst fein werden und hinten ganz erloschen. Fiihler schwarz, die drei ersten Glieder rotgelb, Glied 3 das . kleinste, 4 fast doppelt so breit, die folgenden stark quer, nach der Spitze allmiéhlich leicht verschmilert. Aetheomorpha luzonica sp. nov. Elongata, albido-flava, nitida, capite, prothorace scutelloque rufescentibus, antennis (basi excepta), vertice, sutura limboque laterali elytrorum, pectore, ano pygidioque nigris, tibiis apicem versus tarsisque infuscatis; capite prothoraceque laevibus, ely- tris subtiliter seriatim punctatis. Long., 3.6 mm. LuzON, Laguna, Mount Maquiling. Gestreckt, nach hinten etwas erweitert, missig gewdlbt, oben glinzend, unten weisslich gelb, Fiihler (ausgenommen die ersten drei rostroten Glieder), Mittel- und Hinterbrust, der grésste Teil des letzten Bauchsegmentes, und das Pygidium schwarz, die Schienen nach der Spitze hin nebst den Tarsen angedunkelt. Kopf in den Augen breiter wie der Vorderrand des Thorax, rot- gelb, glatt, Scheitel schwarz, Clypeus kaum von der Stirn ge- trennt, Augen gross, gewélbt. Thorax glatt, rotgelb, mehr als um die Halfte breiter wie lang, nach vorn schwach und fast grad- linig verengt. Schildchen rostrot. Fliigeldecken gelblich weiss, die Nahtkante und ein miassig breiter Saum am Seiten- und Hin- terrande schwarz, der Riicken in fast regelmissigen Reihen fein punktirt. Erstes Tarsenglied so lang, an den Hinterbeinen langer wie das dritte Glied. Aetheomorpha gemina sp. nov. Elongata, flava, capite prothoraceque sublaevibus, rufescenti- bus, antennis apicem versus infuscatis, pygidio nigro, elytris subtilissime striato-punctatis, nigro-limbatis, limbo suturali basi et pone medium dilatato. Long., 3 ad 3.5 mm. PALAWAN, Puerto Princesa (4739). Var. a. Abdomine infuscato, elytris nigris, singulo maculis duabus flavis. Der vorigen sehr Ahnlich, eine Spur kurzer gebaut, Kopf und Thorax einfarbig rotgelb, ersterer in den Augen nur so breit 189287———2 434 The Philippine Journal of Science 1922 wie der Vorderrand des Thorax, letzterer weniger glatt, unter starker Vergrésserung ziemlich dicht verloschen punktulirt, die Fliigeldecken abweichend gezeichnet. Der Kopf bildet eine Flache die zwischen den Augen einen seichten Langseindruck besitzt. Fiihler kurz, vom fiinften Gliede ab stark erweitert und mehr oder weniger angedunkelt. Thorax fast doppelt so breit als lang, nach vorn verengt, die Seiten wenig gerundet. Schildchen . rostrot. Fligeldecken gelblich weiss, ringsum schwarz gesaumt, der Nahtsaum am schmalsten, aber vorn dreieckig dicht hinter der Mitte makelformig erweitert. Diese Makel dehnt sich in der var. a zu einer gemeinschaftlichen Querbinde aus, die in der Mitte jeder Decke verengt ist; die Fliigeldecken sind nun schwarz, mit je zwei gelben gerundeten Quermakeln vor der Mitte und vor der Spitze; Brust und Bauch sind schwirzlich. | Fliigeldecken dusserst fein gereiht-punktirt. Das erste Tarsen- glied aller Beine ktirzer wie das dritte, das letzte Bauchsegment gleichmassig schwach queriiber gewélbt (Mannchen), oder in der Mitte mit einer weiten und nicht tiefen Eigrube (Weibchen). Tabelle der neuen Aetheomorpha-Arten. 1. Langlich eif6rmig, Fliigeldecken vor der Mitte kraftig in Reihen punk- tirt, die dahinter fast erléschen...: A. palawanica sp. nov. 1. Gestreckt, Fliigeldecken fein gereiht-punktirt.........c...cccecceceeeseeeeceeceeee 2; 2. Augen gross, seitlich tiber die Vorderecken des Thorax hervorquellend, Schienen (ohne die Basis) und Tarsen dunkel........ A. luzonica sp. nov. 2. Augen seitlich nur so weit wie die Vorderecken des Halsschildes reich- end, Beine gelb A. gemina sp. nov. Chlamys bakeri sp. nov. Sat elongata, atra, subopaca, antennis basi fulvis apice cum labro ferrugineis, prothorace ruguloso-punctato, medio valde gibboso, gibbere subconico, canaliculato, elytris striato-punctatis, singulo tuberculis octo, posticis duobus sat altis armato, pygidio evidenter quadrisulcato. Long., 2.8 ad 3 mm. LUZON, Benguet, Baguio. Das Halsschild steigt in der Mitte zu einem sehr hohen, fast konischen Hocker auf, der oben eine durchgehende Langsrinne besitzt, am vorderen Abfalle jederseits von mehreren punktirten Langsstreifen und hinten von 4hnlichen unregelmassigen und kurzen Querstreifen durchzogen ist. Fliigeldecken fast recht- eckig, vorn am breitesten und hier kaum breiter als die Tho- raxbasis, bis zur Hinschniirung vor der Mitte etwas verengt, sodann parallel, am Ende bei der Betrachtung von oben grade abgestutzt, auf der Scheibe abgeflacht, in Reihen punktirt, die durch etwa acht Hocker unterbrochen und gestért werden. Die 21,5 Weise: Chrysomeliden der Philippinen, III 435 Lage dieser Erhebungen ist ungefihr dieselbe wie bei den meisten tibrigen Arten: Auf der ersten primaren Rippe liegen zwei, ein kleiner nahe dem Schildchen und ein hoher in der Mitte. Die zweite, mit zwei bis drei Punktreihen besetzte Rippe, tragt vier Hocker, einen kleinen auf der vorspringenden Basis innen von der Schulterbeule und drei grossere, zwei schrig neben den Héckern der ersten Rippe, der dritte und héchste vor der hinteren Nahtecke; die beiden folgenden liegen in einer Schragreihe mit dem zweiten Hicker der ersten Rippe und dem dritten der zweiten Rippe, der siebente ist missig hoch, der achte héher, er verdeckt bei der Ansicht von oben die in Wirk- lichkeit abgerundete hintere Aussenecke der Fliigeldecken ; Brust und Bauch sind mit grossen, aber flachen, blatternarbigen Punkten versehen, auf dem Pygidium liegen vier weite Langs- furchen. Das kleinere Minnchen ist im mittleren Teile des letz- ten Bauchsegmentes abgefiacht, ohne Punkte und nahe dem Vor- derrande mit zwei neben einander liegenden, gekérnelten Héck- ern besetzt, wihrend das Weibchen dort eine grosse, ziemlich flache Eigrube trigt. Chlamys philippinensis sp. nov. Sat elongata, nigra, opaca, antennis basi, palpis labroque ferrugineis, prothorace creberrime punctato, medio gibboso, gibbere transverso, subrotundato, elytris striato-punctatis, minus alte tuberculatis. Long., 2.6 ad 3.2 mm. LuZON, Mount Banahao. LEYTE, Tacloban. Etwas breiter gebaut wie die vorige und nicht so tief schwarz gefarbt, matt, der Hécker des Halsschildes breiter, niedriger, oben breit abgerundet und von einer breiten, flachen Lingsrinne durchsetzt, alle Hicker der Fliigeldecken niedrig, aber die Stiicke der priméren Rippen zwischen ihnen kraftiger; diese Skulptur ahnelt der von Chlamys spilota Baly. Tabelle der neuen Chlamys-Arten. 1. Thorax fast so hoch wie breit, der Hécker annahernd konisch, hintere Aussenecke der Fliigeldecken von oben betrachtet rechtwinkelig, die vier Lingsfurchen des Pygidiums miassig tief. Oberseite rein schwarz nd GEWES PARSE «QW C. bakeri sp. nov. 1. Thorax viel breiter wie hoch, der Hécker oben gerundet, hintere Aus- senecke der Fliigeldecken von oben betrachtet breit abgerundet, die vier Lingsfurchen des Pygidiums flach. Oberseite schwarz, matt. C. philippinensis sp. nov. Coenobius coerulescens sp. nov. Niger, supra nigro-coeruleus, nitidus, antennis basi pedibus- que piceis, prothorace laevi juxta medium transversim sulcato, 436 The Philippine Journal of Science 1922 basi medio triangulariter producta, elytris striato-punctatis. Long., 2.8 mm. LUZON, Laguna, Los Bajfios, 1' Weibchen. Durch die Bildung des Thorax ausgezeichnet. Derselbe ist vollig glatt, auch die normale Punktreihe an der Basis fehlt, aber der Rand ist dafiir jederseits deutlich dicht und fein ge- kerbt und in der Mitte stark dreieckig gegen das Schildchen vorgezogen. Wenig hinter der Mitte befindet sich eine vertiefte Querlinie, die nach den Seiten hin breiter und tiefer wird, so dass letztere aus zwei gewélbten Flachen bestehen. Schildchen lang oval, glatt, hinten zugespitzt. Fliigeldecken in Reihen punktirt, die siebente Reihe wie gewodhnlich sehr kurz, die achte feiner und weitliufiger als die iibrigen punktirt, die zweite und dritte vorn, die neunte und zehnte ginzlich streifenférmig vertieft. Der K6érper ist schwarz, Thorax und Fliigeldecken dunkel blau, stark glanzend, die ersten sechs Fiihlerglieder nebst den Beinen pechbraun, teilweise rotlich. Coenobius monticola sp. nov. Flavescens, nitidus, antennis apice leviter infuscatis, protho- race sublaevi, pone medium transversim striis basi profundis. Long., 2.5 mm. LuZzON, Laguna, Mount Maquiling, 1 Weibchen. ‘Blass gelb, stark glinzend, Thorax und Schenkel etwas rét- licher, das siebente bis neunte Fiihlerglied an der dussersten Spitze, die beiden letzten weiter angedunkelt. Thorax wie bei dem vorigen, aber die Basis nur wenig gegen das Schildchen vorgezogen und mit einigen weitlaufig und unregelmissig an- geordneten Punkten besetzt, welche den Rand nicht emporheben, aber schon die regelmissige Punktreihe der folgenden Arten andeuten. Schildchen lang oval. Fliigeldecken fein punktirt- gestreift, die sechs ersten Streifen nahe der Basis, die beiden letzten ginzlich stirker vertieft. Coenobius brevicornis sp. nov. Subovalis, niger, nitidus, antennis brevibus basi pedibusque flavis, prothorace antrorsum rotundatim-angustato, sat dense punctato, utrinque leviter transversim impresso, serie subbasali fortiter punctato instructo, elytris punctato-striatis, intervallis vix convexis, laevibus. Long., 1.8 mm. Luzon, Laguna, Mount Maquiling, 1 Weibchen. Kurz oval, massig gewélbt, gliinzend schwarz, die ersten sechs Fiihlerglieder und die Beine rétlich gelb. Fiihler kurz, jedes 21,5 Weise: Chrysomeliden der Philippinen, III 437 der erweiterten sechs Endglieder nicht langer als breit. Kopf- schild dicht und fein runzelig punktirt. Thorax fast doppelt so breit wie lang, von der Basis aus in schwacher Rundung verengt, die Scheibe ziemlich dicht punktirt, im dusseren Dritte] jeder- seits mit einer seichten, schrigen Querrinne, hinter welcher die etwas gewoélbte Fliche bis zu den Hinterecken weitlaufiger punktirt ist. Die Punkte der normalen Reihe vor dem leisten- formigen Hinterrande sind wenig starker wie die auf der Scheibe. Fltigeldecken mit den Seiten des Halsschildes in einer Flucht bis vor die Mitte erweitert, sodann Ahnlich verengt, hin- ten breit gerundet-abgestutzt, mit verrundeter Nahtecke, oben schwach gewélbt, punktirt-gestreift, die Intervalle breit, kaum gewolbt, glatt. Diese kleine Art ist an den kurzen Fiihlern, der Skulptur, und dem schwachen Seiteneindrucke des Thorax kenntlich, welcher bei der Ansicht von oben keine Einschniirung hervorruft. Coenobius bicolor sp. nov. Subcylindricus, niger, nitidus, capite, antennis basi, pedibus prothoraceque laete flavo-rufis, hoc sublaevi, transversim im- presso, elytris punctato-striatis, intervallis haud convexis, lae- vibus. Long., 2.3 mm. Luzon, Laguna, Mount Maquiling, 1 Mannchen (6136). Durch die Farbung, die an den japanischen Coenobius sulci- collis erinnert, ausgezeichnet. Der Vorderkérper nebst den fiinf ersten Fiihlergliedern und den Beinen ist lebhaft gelblich rot, der Hinterkérper schwarz, die Oberseite stark gliinzend. Fiihler schlank, von den etwas stirkeren sechs schwarzen End- gliedern ist jedes mehr als doppelt so lang wie breit. Clypeus ziemlich dicht punktirt. Thorax etwas breiter als lang, nach vorn zusammengedriickt-verengt und die Seiten nahe der Mitte durch einen Quereindruck deutlich eingeschniirt. Letzterer ’ reicht von einer Seite zur andern, ist aber im mittleren Teile flach, an den Seiten tief. Die Scheibe ist bis auf wenige undeut- liche Piinktchen vor dem Schildchen glatt; der Basalrand wird durch eine vertiefte Punktreihe leistenformig emporgehoben, ist nebst der Punktreihe schwarz gefarbt und in der Mitte in einem sehr flachen stumpfen Winkel gegen das lang elliptische Schildchen vorgezogen. Die Fliigeldecken sind ziemlich parallel, kaum breiter und nur um die Halfte linger als das Halsschild punktirt gestreift, die glatten Intervalle fast eben, nur die beiden dausseren gewilbt. 438 The Philippine Journal of Science 1922 Coenobius bakeri sp. nov. °. ‘ Breviter ovalis, convexus, fulvus, nitidus, antennis gracilibus articulis sex ultimis nigris, prothorace sublaevi, utrinque pro- funde oblique impresso, elytris punctato-striatis, fascia basali et apicalis nigro-coerulis. Long., 3 mm. PALAWAN, Puerto Princesa (4740). Die bis jetzt bekannte grésste Art der Philippinen und durch die Zeichnung leicht kenntlich. Der Korper ist sehr kurz eifér- mig, oben leuchtend gelblich rot, glanzend, unten mehr gelb, die sechs verbreiterten Endglieder der schlanken Fiihler schwarz, eine gemeinschaftliche Querbinde an der Basis und Spitze der Fliigeldecken metallisch dunkel blau. Die Basalbinde nimmt ungefahr das erste, die Apicalbinde das letzte Viertel der Lange ein, beide verlangern sich aber aussen (die vordere breiter wie die hintere) bis sie sich in der Mitte des Seitenrandes schmal verbinden. Clypeus stark, aber nur missig tief und dicht punk- tirt. Von den sechs Endgliedern der Fiihler ist jedes doppelt so lang wie breit. Thorax um die Hialfte breiter als lang, annih- ernd halbkugelig, jederseits mit einer tiefen Schrigfurche nahe der Mitte, welche die Seiten einschniirt; beide Furchen sind oben durch einen weiten, verloschenen Eindruck verbunden. Die vordere Halfte ist dusserst fein, die hintere etwas stirker punk- tirt, die basale Punktreihe kraftig, der Hinterrand in der Mitte schwach gegen das lang elliptische Schildchen vorgezogen. In der Gattung Coenobius hat das Halsschild einen von der Scheibe abgesetzten, oft hell gefirbten, und von zwei (ofter punktirten) Querlinien eingefassten, tiefer liegenden Vorderrand; dieser ist in der vorliegenden Art linger wie bei den tibrigen und Ausserst zart quer liniirt. Fliigeldecken um ein Drittel langer als das Halsschild, ziemlich fein, vorn stirker als hinten punktirt- gestreift, die Intervalle einzeln dusserst fein punktulirt, fast eben, nur die drei dusseren gewdlbt. Coenobius gilvus sp. nov. ¢. Subcylindricus, pallide flavus, nitidus, antennis gracilibus, ar- ticulis sex ultimis infuscatis, prothorace ante basin punctulato, medium versus profunde oblique impresso, elytris punctato-stria- tis, intervallis laevibus exterioribus subconvexis. Long., 2 mm. PALAWAN, Puerto Princesa. Eine kleine blassgelbe, brdunlich oder ritlich angelaufene Art mit sechs schlanken, verdickten, schwirzlichen Endgliedern der Fiihler, von deren fiinf hellen Basalgliedern das dritte fast so lang wie das erste ist, wihrend die andern viel kiirzer sind. Clypeus dicht runzelig punktirt. Thorax an der Basis kaum 21,5 Weise: Chrysomeliden der Philippinen, III 439 breiter wie in der Mitte lang, nach vorn verengt, glatt, in der hinteren Halfte nicht dicht, sehr fein punktulirt, jederseits mit einer schmalen, tiefen Schraégrinne, welche die Seiten einschniirt. Fliigeldecken etwas linger wie das Halsschild, nach hinten etwas verengt, oben punktirt-gestreift, die mittleren Streifen hinten abgeschwiicht, die Intervalle glatt, die inneren kaum, die dus- seren deutlich gewdlbt. Coenobius flaviventris sp. nov. Breviter-subcylindricus, niger, nitidus, prothorace postice punctato utrinque profunde et oblique impresso, elytris vix coeruleo indutis ante medium punctato-striatis, postice subtilius striato-punctatis, intervallis punctulatis, labro, antennis basi, pedibus abdomineque flavis. Long., 2.6 mm. LuzoN, Mount Banahao, 1 Weibchen (2794). Breit cylindrisch, glanzend schwarz, die Fliigeldecken mit ein- em kaum merklichen blaulichen Anfluge, die Oberlippe, die ersten fiinf Fiihlerglieder, Beine, und Hinterleib gelb. Fiihler schlank, Glied 3 langer als 4 und so lang wie 5, die folgenden wenig linger aber etwas dicker, jedes mindestens doppelt so lang als breit. Clypeus massig dicht punktirt. Thorax breiter wie lang, nach vorn verschmialert, jederseits mit einer Schrigfurche, wel- che die Seiten einschniirt, die hintere Hilfte mit Ausnahme der Mittellinie und der Seiten ungleichmassig und nicht dicht punk- tirt, die Punkte nehmen nach vorn an Starke ab. Fliigeldecken in den Schultern am breitesten, nach hinten wenig verengt und am Ende gerundet-abgestutzt, mit verrundeter Nahtecke, vorn punktirt-gestreift, der erste Punkt jedes Streifens grubenfér- mig, hinten sehr fein gereiht-punktirt, die Zwischenstreifen breit, eben, die inneren einreihig, die 4usseren verworren punk- tulirt, der letzte schwach gewélbt. Coenobius fulvifrons sp. nov. 2. Subcylindricus, niger, nitidus, capite, antennis basi pedibus- que rufo-flavis, prothorace postice in medio parce punctulato, utrinque profunde oblique impresso, elytris subtiliter striato- punctatis, intervallis vix perspicue punctulatis. Long., 2.5 mm. Luzon, Benguet, Baguio (6132). . Cylindrisch, glinzend schwarz, der Kopf, die ersten drei bis fiinf Fiihlerglieder, und die Beine rotlich gelb. Kopfschild etwas uneben, ziemlich dicht punktirt. Fiihler schlank, Glied 3 so lang wiel. Thorax kaum um die Halfte breiter als lang, nach vorn verengt, jederseits mit einer tiefen Schragfurche welche die Seiten einschniirt, in der Mitte der hinteren Halfte zerstreut 440 The Philippine Journal of Science 1922 und sehr fein punktulirt. Fliigeldecken nach hinten unbedeutend verschmilert, fein gereiht-punktirt, der erste Punkt jeder Reihe grubenférmig, die beiden ausseren Reihen vertieft, ihre Inter- valle leicht gewolbt, die iibrigen abgeflacht, alle mit einer Reihe wenig bemerkbarer Piinktchen. Coenobius pusillus sp. nov. @. Breviter ovalis, niger, nitidus, labro, antennis (apicem versus infuscatis) pedibusque flavis, prothorace sublaevi, valde trans- verso, utrinque sat profunde impresso, elytris subtilissime punc- tato-striatis, intervallis sublaevibus leviter convexis. Long., 1.8 mm. LUZON, Benguet, Baguio (6131). Kurz oval, massig gewélbt, schwarz, glinzend; Oberlippe, Beine, und Fiihler gelb, letztere missig lang, die Glieder vom zweiten ab unter sich an Linge ziemlich gleich, aber 2 dick, 3 bis 5 sehr diinn, die folgenden verbreitert, 6 bis 8 an der Spitze, die folgenden ganzlich angedunkelt. Kopfschild rotbraun, klein, fast glatt. Halsschild doppelt so breit wie lang, nach vorn in schwacher Rundung verengt, oben ziemlich glatt, nur in der Mitte vor dem Hinterrande einzeln und verloschen punktulirt, davor jederseits mit einer ziemlich kurzen und massig tiefen Schragfurche, welche die Seiten leicht einschniirt. Fliigeldecken an den Seiten leicht gerundet, mindestens doppelt so lang wie das Halsschild, dusserst fein punktirt-gestreift, Intervalle glatt, eben, die drei fiusseren gewiolbt. Coenobius ingenuus sp. nov. ¢. Subcylindricus, niger, nitidus, antennis basi, labro pedibusque flavis, prothorace minus dense evidenter punctulato transver- sim impresso, elytris striato-punctatis. Long., 2 mm. PALAWAN, Puerto Princesa. Schlanker gebaut wie die vorige, schwarz, glinzend; Ober- lippe, die ersten fiinf Fihlerglieder, und die Beine gelb. Fiihler schlank, jedes der sechs Endglieder fast doppelt so lang wie breit. Halsschild quer, nach vorn allmahlich verschmilert, nicht dicht fein punktirt, die Punkte etwas kleiner wie die der Basalreihe. Hinter der Mitte liegt eine feine Querrinne, die an den Seiten tiefer wird und etwas nach vorn biegt. Fliigeldecken wenig breiter wie das Halsschild, nach hinten eine Spur verengt, fein punktirt-gestreift, die Reihen bis zur Spitze deutlich, ihre Inter- valle kaum gewédlbt. 21,5 Weise: Chrysomeliden der Philippinen, III 441 Tabelle der Coenobius-Arten. tieften Punktreihe welche den Basalrand leistenférmig emporhebt.... 3. 2. Halsschild und Fliigeldecken dunkel blau, ersteres glatt. Linge, 2.8 mm ee C. coerulescens sp. nov. 2. Oberseite gelb, vor der Thoraxbasis einige Punkte welche die normale Punktreihe andeuten. Lange, 2.5 mm................ C. monticola sp. nov. 3. Fihler auffallig kurz, nur die Hinterecken des Thorax erreichend, letzterer ziemlich dicht punktirt. Lange, 1.8 mm. C. brevicornis sp. nov. 3. Fuhler schlank, die Hinterecken des Thorax bedeutend tiberragend.... 4. 4. Fligeldecken hell gefirbt mit zwei gemeinschaftlichen dunkelblauen Querbindens Lange; 3. mci... isccsseccdccsshiseescce C. bakeri sp. nov. © Fae coed occ sce ae Pic nc bet) | a a COR ERIE REDON SP Ls ROR CD aR DRE Ge SGN, 5. 5. Kérper einfarbig gelb. Linge, 2 mm e C. gilvus sp, nov. 5. K6érper anders gefarbt...... RiSts ai itinevet svemsdiae nian dete toates ei, (ado 6. 6. Vorderkérper und Beine gelblich rot, Hinterkérper schwarz. Lange, 2.8 mM 2. , .. C. bicolor sp. nov. 6. Thorax und Fliigeldecken einfarbig schwarz a i, 7. Bauch und Beine gelb. Lange, 2.6 mm............ C. flaviventris sp. nov. 7. Bauch schwarz paacarauas 8. 8. Beine schwarz, nur die Vorderbeine, die Spitze der vier Hinterschie- nen und die Tarsen gelbbraun. Linge, 2 mm.... C. manilensis Weise. 8. Alle Beine gelb 9. 9. Kopf hell gelb. Lange, 2.6 mm C. fulvifrons sp. nov. 9. Kopf dunkel 10. 10. Thorax glatt, Punktstreifen der Fliigeldecken hinten verloschen. Lange, 1.8 mm C. pusillus sp. nov. 10. Thorax deutlich punktirt, Punktstreifen der Fliigeldecken nach hinten wenig abgeschwiicht Cc. ingenuus sp. nov. Cryptocephalus laevissimus Suffrian. Das Mdnnchen ist 3, das Weibchen 3.5 Millimeter lang, blass rétlich gelb, glinzend, Mittel- und Hinterbrust, selten auch noch die Grube zum Einlegen der Schenkel in die Seitenstiicke der Vorderbrust schwarz, die Tarsen angedunkelt. Ein schwarzer Schulterfleck ist héchst selten vorhanden. Das kleinere Mann- chen hat ein in der Mitte leicht abgefiachtes Analsegment. Luzon, Laguna, Los Bafios, Mount Maquiling (2891). Melixanthus intermedius Suffrian. - Die Fliigeldecken haben dieselbe Farbe wie die vorige Art, aber zwei Flecke auf jeder, naher der Naht als dem Seitenrande, schimmern dunkel durch, die Zwischenstreifen sind bei dem vorliegenden Mannchen hinten kaum gerunzelt und eben so 44? The Philippine Journal of Science 1922 glinzend als vorn, die Schulterbeule ist innen nicht deutlich abgesetzt. PALAWAN, Puerto Princesa (6124). Melixanthus (Anteriscus) palawanus sp. nov. Subcylindricus, ferrugineus, nitidus, subtus (femoribus excep- tis) flavescens, mandibulis antennarumque articulis septem ultimis subdilatatis nigris, labro rufo, elytris ante medium sub- tiliter striato-punctatis, postice sublaevibus. Long., 4.2 mm. PALAWAN, Puerto Princesa. Einem Melixanthus cryptocephalus im Koérper- und Fiihler- bau ahnlich, aber die Klauen mit einem Basalzahne. Cylin- drisch, unten braunlich gelb, Schenkel und Oberseite rostrot, glinzend, Oberlippe dunkel rot, Mandibeln und die erweiterten sieben Endglieder der Fiihler schwarz. Clypeus von der Stirn undeutlich geschieden, beide bilden eine fast ebene Fliche und sind fein und zerstreut punktirt; die Stirn hat in der Mitte ein Griibchen. Halsschild fast halbkugelig, um die Hilfte breiter als lang, fast glatt. Fltigeldecken vor der Mitte fein in regel- massigen Reihen punktirt, welche dahinter allmihlich immer feiner werden. Melixanthus (Anteriscus) bakeri sp. nov. Subcylindricus, subtus fulvus, antennis elicits septem ulti- mis nigris, supra coeruleus subaeneo-micans, nitidus, elytris striato-punctatis, singulo fascia fulva ante medium. Long., 5.3 mm. LUZON, Mount Banahao (2881). Etwas grésser, namentlich breiter wie die vorige, unten rétlich gelb, die erweiterten sieben letzten Fiihlerglieder schwarz, oben gesittigt metallisch blau, bei gewissem Lichte griinlich tiberflogen, glinzend, jede Fliigeldecke mit einer gelb- lich roten Querbinde. Von dieser liegt nur ein sehr schmaler Streifen hinter der Mitte, der gréssere Teil vor derselben; sie ist innen massig breit, mit ziemlich parallelen Rindern, und reicht nahe an die Naht heran, aussen verschmilert sie sich von der sechsten Punktreihe an und endet an der zehnten Reihe. Das Kopfschild ist fein runzelig, die Stirn sparsam punktirt. Thorax quer, halbkugelig, fast glatt, unter starker Vergrésserung dicht, in der Mitte feiner als an den Seiten punk- tulirt. Die Punktreihen der Fliigeldecken sind kriftig, regel- miassig, aber in der Mitte der hinteren Hilfte bedeutend abge- schwacht. 21,5 Weise: Chrysomeliden der Philippinen, III 443 Tabelle der Melixanthus-Arten. 1. K6rper nach hinten verengt, Fihler kurz, Glied 6 bis 10 quer, Vorder- hiften vom Vorderrande des Prosternums entfernt, zwischen ihnen und den Vorderecken des Thorax eine grosse, dreieckige, dicht punk- tirte Grube (Melixanthus i. sp.). Rétlich gelb, glanzend, Kopf und Thorax etwas dunkler und mehr rot gefarbt, die erweiterten sechs Endglieder der Fiihler schwarz... M. intermedius Suffrian. 1. Korper cylindrisch, Fiithler lang, die Hinterhiiften bertihrend, von den erweiterten Gliedern 6 bis 10 jedes linger als breit, Vorderhiiften sehr schmal vom Vorderrande des Prosternums getrennt (subg. Ante- POON eSB a catiphte id sy, Gs ee 2. 2. Oberseite lebhaft rostrot................... M. (Anteriscus) palawanus sp. nov. 2. Oberseite metallisch blau, eine Querbinde der Fliigeldecken rot. M. (Anteriscus) bakeri sp. nov. Basilepta bakeri sp. nov. Laete metallico-viridis, nitidissima, antennis nigricantibus articulis 4 primis, palpis pedibusque rufis, tarsis infuscatis in dorso viridi indutis, femoribus apice viridibus posticis subden- tatis; prothorace medio minus dense, latera versus crebre et fortius punctato, lateribus angulatis, elytris infra basin pro- funde transversim impressis, subtilissime striato-punctatis punctis intra impressionem basalem sat fortis. Long., 4 mm. LuzON, Laguna, Mount Maquiling (2092). In Grésse und Habitus dem Basilepta cumingi ahnlich, aber leuchtend metallisch griin, etwas goldig schimmernd gefirbt, Mandibeln schwarz, Fiihler schwirzlich, die beiden ersten Glie- der nebst Tastern und Beinen gelblich rot, die beiden folgenden Glieder braunlich rot, an den Schenkeln der abgeschniirte Spitz- enteil ahnlich griin wie der Kérper, auch der Riicken griinlich schimmernd, die Tarsen angedunkelt, der Bauch lebhaft goldig, an der Spitze kupferig. Fiihler schlank, Glied 3 etwa so lang wie 2, doch diinner. Scheitel langsrissig punktirt, ein Raum davor, in der Mitte der Stirn, glatt, unten mit einer kurzen Mittelrinne, der untere Teil der Stirn nebst dem Clypeus nicht dicht punktirt. Thorax etwas breiter wie lang, die Seiten von der Basis bis hinter die Mitte schwach divergirend, davor starker convergirend, beide Teile durch eine stumpfwinkelige Ecke ge- schieden, die Scheibe nicht dicht und stark, an den Seiten viel dichter und starker punktirt, vorn zu einer regelmassig und dicht punktirten Querrinne abfallend, welche einen miassig breiten, glatten, tiefer liegenden Streifen am Vorderrande begrenzt. Fliigeldecken breiter wie der Thorax, mit hoher Schulter- und Basalbeule, fast glatt, die Punktreihen nur in der Nahe des tiefen Quereindruckes (die sechste auch bis zur 444 The Philippine Journal of Science 1922 Basis und die letzte giinzlich) tief und deutlich, sonst wie abge- schliffen und nur noch unter starkerer Vergrésserung bemerkbar. Schenkel massig verdickt, die hintersten fein stumpf gezahnt. Basilepta gemmata sp. nov. Oblongo-ovalis, sat convexa, minus nitida, subtus aenea, pedi- bus obscure rufis violaceo splendentibus, tarsis antennisque nigris, his basi testaceis, supra aeneo-viridis, scutello violaceo, capite prothoraceque creberrime fortius punctatis, elytris den- sissime foveolatim punctatis (punctis subseriatis viridi-auri- chaleeo splendentibus), singulo maculis duabus purpureis. Long., 3.5 mm. LEYTE, Tacloban. Von den iibrigen Arten der Philippinen durch die sehr dichte und grobe Punktur der Oberseite abweichend. Die Unterseite ist metallisch griinlich schwarz, etwas messingglinzend, Beine rotbraun, Schenkel violett iiberflogen, Tarsen und Fiihler schwarzlich, an diesen die vier ersten Glieder rétlich braun. Die Oberseite erscheint bei der Ansicht mit blossem Auge fast matt, aber schon bei geringer Vergrésserung treten die Gruben der Fliigeldecken als goldig griin leuchtende Punkte hervor. Kopf und Thorax sind ausserst dicht, stark, und tief punktirt, die Zwischenraéume bilden feine Leisten; ahnlich, nur grésser, sind die Punkte auf den Fliigeldecken, die zu dicht neben einander liegenden, in der vorderen Hialfte verworrenen, dahinter fast regelmadssigen Reihen geordnet sind. Das Halsschild erweitert sich von der Basis aus schwach und erreicht etwas hinter der Mitte die grésste Breite, hier biegen die Seiten plétzlich um, bilden einen zahnférmigen stumpfen Winkel und convergiren etwas starker nach vorn. Die schwarzlich purpurfarbenen Makeln jeder Fliigeldecke, eine in ein Viertel der Lange nahe der Naht, die andere dicht hinter der Mitte am Seiterrande, sind klein. Die Beine sind massig lang, die Schenkel in der Mitte verdickt und die hintersten mit einem leicht zu iibersehenden Zahnchen versehen. Basilepta severa sp. nov. Ovalis, convexa, subtus nigra, tibiis (basi excepta), tarsis palpisque fulvis, antennis fuscis basifulvis, capite aurichalceo vel cupreo, prothorace elytrisque coeruleo-viridibus vel coeruleis, illo fortiter punctato lateribus subrotundatis, elytris striato-punc- tatis, punctis pone medium evanescentibus, callo humerali et basali subpolitis, femoribus muticis. Long., 2.5 ad 3 mm. 21,5 Weise: Chrysomeliden der Philippinen, III 445 Luzon, Laguna, Mount Maquiling; Mount Banahao (6130). Durch die Farbung ausgezeichnet. Die Unterseite ist schwarz, die ersten vier Fiihlerglieder, die Spitzenhilfte der Vorderschienen, die folgenden Schienen: mit Ausnahme der Ba- sis nebst allen Tarsen gelblich rot; Kopf und Vorderrand des Thorax goldig griin bis kupferrot, die Scheibe des letzteren griin schwach messingglinzend, die Fliigeldecken griin mit blaulich- em Schimmer bis gesittigt blau. Stirn fast glatt, Clypeus ein- zeln punktirt. Halsschild stark queriiber gewélbt, nach vorn etwas zusammengedriickt verengt, die Seiten gerundet, die Scheibe in der Mitte miassig dicht und kraftig punktirt, nahe den Seiten dichter und viel starker. Fliigeldecken regelmissig gereiht-punktirt, die zweite und dritte Reihe sind ganz, die fol- genden reichen hoéchstens bis zur Mitte, die Basalbeule ist fast glatt. Beine ziemlich kurz, Schenkel unbewehrt. Basilepta assimilis sp. nov. Ovalis, convexa, subtus nigra, pectore nigro-aeneo, palpis antennisque fulvis, his apice infuscatis, femoribus basin versus rufo-piceis ; supra obscure viridi-aenea, nitida; prothorace minus dense-, latera rotundata versus crebrius et fortius punctato, ely- tris striato-punctatis, punctis in callo basali et in apice evanes- © centibus. Long., 3 mm. Leyte, Tacloban (4748). Den groéssten Stiicken der vorigen ahnlich, der Thorax grdésser, breiter, und stirker gewoélbt, und der Korper abweichend ge- farbt. Unten schwarz, Taster und Fiihler rétlich gelb, die vier oder fiinf Endglieder der letzteren gebraiunt, Schenkel an der Basis etwas gerétet, Brust und die Oberseite dunkel metallisch grin, Oberlippe schwarz. Stirn sparsam und ziemlich fein, der Clypeus dichter und stirker punktirt. Halsschild kissenartig gewolbt, an den Seiten schwach gerundet und vorn nur massig verengt, in der Mitte der Scheibe wenig dicht und tief, an den Seiten stirker, tiefer, und dichter punktirt, etwas runzelig. Flii- geldecken wenig breiter wie das Halsschild, regelmidssig in Rei- hen punktirt, die auf der Basalbeule fast erloschen, hinter der Mitte allm&hlich abgeschwacht sind. Beine ziemlich kurz, Schenkel ungezahnt. Basilepta splendida sp. nov. Ovalis, convexa, subtus nigra, antennis (apice fuscis) pedi- busque rufo-testaceis, dorso femorum apice infuscato, supra nigro-aurichalcea, nitidissima, capite prothoraceque laevibus, hoc paullo ante basin latissimo, antrorsum fortiter angustato, apice 446 The Philippine Journal of Science 1922 anguste marginato, elytris striato-punctatis, punctis ante apicem evanescentibus. Long., 2.3 mm. PALAWAN, Puerto Princesa. An der stark glinzenden, dunkel messingfarbigen Oberseite und dem vorn sehr fein gerandeten Halsschilde stets wieder zu erkennen. Die Unterseite ist schwarz, Fiihler schwirzlich, die ersten fiinf bis sieben Glieder nebst den Beinen rotbraun, der Riicken der Schenkel wenigstens in der Spitzenhilfte angedun- kelt, Taster rotgelb. Kopf glatt. Thorax vor der Basis erwei- tert und in etwa vier Fiinftel der Lange am breitesten, hier ab- gerundet und gradlinig sehr stark nach vorn verschmialert, die Scheibe massig gewdlbt, fast glatt, doch sind bei starker Ver- grésserung einzelne Piinktchen wahrzunehmen. Fliigeldecken nach hinten leicht verengt, oben in regelmdssigen Reihen punk- tirt, die auf der Basalbeule abgeschwacht, aber erst nahe der Spitze verloschen sind. Schenkel einfach. Basilepta forticornis sp. nov. Ovata, convexa, laete ferruginea, nitida, antennis articulis septem ultimis valde dilatatis cum mandibulis et tarsis nigris; fronte sublaevi, clypeo prothorace parce punctulatis, hoc lateribus rotundatis, elytris pone basin obsolete impressis, striato-punctatis, punctis in basi et pone medium subtilissimis, intervallis parce vix perspicue punctulatis. Long., 5 mm. LUZON, Laguna, Mount Magquiling (4734). Kiformig, mit der gréssten Breite hinter der Mitte der Flii- geldecken, am Ende breit abgerundet, lebhaft rostrot und stark glanzend, die sieben Endglieder der Fiihler, Mandibeln, und Tarsen schwarz. Stirn fast glatt, nur der Scheitel jederseits nahe den Augen punktirt, auch das Kopfschild mit einigen Piinkt- chen. Fiihler kaum langer wie der halbe Kérper, die vier ersten Glieder kahl, glinzend, die iibrigen dicht behaart, matt, ziemlich stark erweitert, jedes um die Halfte linger als breit, die beiden Endglieder eine Spur kiirzer, Glied 3 und 4 diinn, 3 kiirzer als 2 und viel kiirzer als 4. Thorax um die Halfte breiter wie lang, nahe der Mitte am breitesten und nach vorn etwas stairker wie nach hinten in gleichmassiger Rundung verengt, am Seiten- und Vorderrande glatt, auf dem iibrigen Teile zerstreut, fein und verloschen punktirt. Fiigeldecken mit elf regelmdssigen Punktreihen und breiten, ebenen, einzeln und schwer sichtbar punktulirten Intervallen. Von den Punktreihen ist die kurze erste nebst dem hinteren Teile der zweiten und der Anfang der sechsten vertieft, die Punkte selbst sind in der Nahe des sehr flachen Quereindruckes hinter der Schulter nicht stark, an der 21,5 Weise: Chrysomeliden der Philippinen, III A447 Basis und hinter der Mitte abgeschwicht, vor der Spitze fast erloschen. Durch die starken Fiihler, Kérper- und Thoraxform von den ibrigen Arten abweichend, aber wohl nicht generisch ver- schieden. Basilepta luzonica sp. nov. Oblongo-ovalis, sat convexa, testaceo-flava, nitida, prothorace medio punctulato, lateribus angulatis, elytris obsoletissime stria- to-punctatis, sutura infuscata. Long., 2.2 mm. LUZON, Laguna, Mount Maquiling. In Grésse und K6érperform der Basilepta thoracica Lefévre am ahnlichsten, aber durch die Thoraxform und die Skulptur der Oberseite sehr verschieden. Langlich oval und miassig gewolbt, blass rétlich gelb, der Thorax etwas mehr gerétet wie die Fliigeldecken, glanzend, die Naht der Fliigeldecken vorn pechschwarz, hinter der Mitte rostrot gesiumt. Kopf mit einigen zerstreuten Punkten. MHalsschild vor der Basis gradlinig erwei- tert, sodann plotzlich etwas stirker und fast gradlinig nach vorn verengt, so dass etwas hinter der Mitte ein annadhernd zahn- formiger stumpfer Winkel entsteht; die sonst glatte Oberflaiche hat im mittleren Teile einige ziemlich weitlaufig stehende, sehr flache aber nicht besonders feine Punkte. Auf den Fliigeldecken sind die Punktreihen sehr verloschen und selbst in und neben dem Quereindrucke hinter der Basis schlecht bemerkbar, nur die erste Reihe liegt in einem feinen Streifen. Schenkel unge- zahnt. Basilepta palawanica sp. nov. Suboblongo-ovalis, convexiuscula, testaceo-flava, nitida, pro- thorace laevi paullo ante basin latissimo, deinde apicem ver- sus valde angustato, elytris subtiliter sed evidenter striato-punc- tatis, sutura antice brunnescente. Long., 2.3 mm. PALAWAN, Puerto Princesa. Etwas breiter gebaut wie die vorige, und von ihr durch die Form des glatten Halsschildes sowie die deutlichen Punktreihen der Fliigeldecken bedeutend abweichend, einem kleinen einfarb- igen Stiicke der Basilepta binotata Lefévre am 4hnlichsten. Blass rotlich gelb, glinzend, die Naht der Fliigeldecken vorn leicht gebriunt. Kopf und Thorax glatt, letzterer ist nahe der Basis in einer verrundeten winkeligen Ecke am breitesten, von hier nach hinten wenig, nach vorn fast gradlinig stark verengt. Die Punktreihen der Fliigeldecken sind deutlich, auf 448 The Philippine Journal of Science 1922 der Basalbeule abgeschwicht, nahe der Spitze verloschen, die erste Reihe liegt in einem Streifen und verbindet sich vorn durch eine regelmissige Querreihe von Punkten hinter der Basalkante mit der fiinften Reihe. Beine ziemlich kurz mit einfachen Schenkeln. Tabelle der Basilepta-Arten. _ A, METALLISCH GEFARBTE ARTEN 1. Kopf und Halsschild glatt, letzteres an der Spitze scharf jedoch 4usserst schmal gerandet. Schwarz messingfarbig, der grosste Teil der Fiihler und die Beine dunkel rotbraun. Lange, 2.3 mm. Palawan. B. splendida sp. nov. . Kopf und Thorax deutlich punktirt. 2 . Die ganze Oberseite dusserst dicht punktirt, die Punkte des Vorder- kérpers stark, die der Fltigeldecken | grubenférmig bis hinten hin von gleicher Grésse und nur teilweise gereiht. Lange, 3.5 mm. Leyte. B. gemmata sp. nov. 2. Fliigeldecken in regelmassigen Reihen oder Streifen punktirt................ 3. 3. Thorax dicht und ziemlich stark, nadelrissig punktirt, an den Seiten gerundet. Unten griinlich blau, oben violettblau, Oberlippe pech- schwarz, Taster und Basis der schwarzen Fiihler rotgelb bis rotbraun. tae, 25 Wek. | BOGOR acs ie B. philippinensis Lefévre. . Thorax nicht dicht punktirt.. 4. . Halsschild an den Seiten winkelig, Basalbeule der Fligeldecken stark.. 5. . Halsschild an den Seiten gleichmassig gerundet. 6. . Lebhaft griin, Halsschild in der Mitte nicht dicht, an den Seiten dicht und stark punktirt, Punktreihen der Fliigeldecken sehr fein, nur die Hinterschenkel mit einem Zahnchen. Linge, 4 mm. Luzon. B. bakeri sp. nov. 5. Sehwarzlich griin, Quereindruck der Fligeldecken violett, Thorax ziem- lich weitléufig, die Fliigeldecken stark und in Reihen punktirt, die auf der Basalbeule feiner, hinter der Mitte fast erloschen sind, alle Schenkel spitz gezihnt. Linge, 4.5 mm. Luzon... B. cumingi Baly. 6. Schenkel geziihnt. Brust und Oberseite metallisch prfin, Beine rétlich gelbbraun, Bauch und Basis der schwarzen Fiihler rotlich gelb, Thorax weitldufig stark punktirt, jederseits mit zwei Gruben. Linge, 5.5 mm. Luzon B. semperi Lefévre. G. Bchemieel wergrens lyri be sao cp eek peepee enna ts jageneeen gio onmngoie 7. 7. Unterseite braunlich rot, Oberseite glanzend violett, Thorax in der Mitte weitlaufig, nach den Seiten dichter ziemlich stark punktirt, Fliigeldecken hinter der Basis leicht quer vertieft, gereiht-punktirt, Punkte hinten fast erloschen. Lange, 6 mm. Luzon. B. janthina Lefévre. . Unterseite schwarz, Brust zuweilen metallisch griin ss 8. 8. Oberseite und Brust metallisch dunkel griin. Lange, 3 mm. “Leyte. B. assimilis sp. nov. 8. Kopf und Halskragen goldgriin oder kupferrot, Thorax lebhaft metal- lisch griin, die Fliigeldecken ebenso oder gesattigt blau, Fiihlerbasis, Spitzenhalfte der Schienen, und die Tarsen rétlich gelb. Linge, 2.5 Te WAU. LOSON neice oi pepe etree B. severa sp. nov. ne aor & Ww ~] 21,5 Weise: Chrysomeliden der Philippinen, III 449 B, ROTLICH GELB ODER ROT GEFARBTE ARTEN 1. Kérper hinten breit abgerundet, lebhaft rostrot, Seiten des Thorax gleichmissig gerundet, Fiihler ziemlich dick. Lange, 5 mm. Luzon. B. forticornis sp. nov. 1. Kérper hinten schmal abgerundet, Fiihler schlank 2. 2. Die grésste Breite des Halsschildes liegt etwas hinter der Mitte zwischen den stumpfwinkeligen Seiten, die nach vorn mehr als nach hinten convergiren, die Scheibe ist in der Mitte weitlaiufig punktirt, an den Seiten glatt; Punktreihen der Fliigeldecken sehr fein. Lange, 2.2 mm. Luzon B. luzonica sp. nov. 2. Die grésste Breite des Halsschildes liegt wenig vor den Hinterecken.... 3. 8. Thorax in der Mitte glatt, an den Seiten (ausgenommen ein Streifen am Vorderrande) dicht, kraftig, etwas runzelig punktirt. Einfarbig rétlich gelb. Linge, 2.38 bis 3 mm. Luzon........ B, thoracica Lefévre. Pid ge ig > a. MARCA CUeaRige i Wagh Fe «seme SiN Slap Breen ee A. 4. Thorax an der breitesten Stelle winkelig, schmal abgerundet. Rétlich gelb, Naht der Fliigeldecken vorn etwas gebraunt, die erste Punktreihe lduft vorn ganz regelmassig nach aussen hinter der Basalkante und verbindet sich mit der fiinften. Lange, 2.3 mm. Palawan. B, palawanica sp. nov. 4. Thorax an der breitesten Stelle kaum winkelig sondern breit abgerundet, die erste Punktreihe der Fliigeldecken endet an der Basis und ist dort nicht mit der fiinften verbunden. Rétlich gelb, die inneren Punktreihen der Fliigeldecken dunkel durchscheinend, die Naht und eine Makel hinter der Basalbeule schwarz. Linge, 3 bis 3.3 mm. Luzon, Bohol. B. binotata Lefévre. Nodina luzonica sp. nov. Subrotundata, convexa, subtus nigra, supra nigro-coerulea, nitida, antennis basi fulvis, capite aenescente minus crebre verticeque dense aciculatim punctatis, prothorace sat crebre subtiliter latera versus ‘paullo fortius punctato, elytris regula- riter striato-punctatis. Long., 2.56 mm. LUZON, Laguna, Mount Maquiling. Gerundet, wenig linger als breit, gewolbt, schwarz, die ersten zwei oder drei Fiihlerglieder rotgelb. Kopf griinlich messing- farbig angehaucht, Thorax, Schildchen, und Fliigeldecken schwarzblau, glanzend. Die Stirn ist vereinzelt und fein, der Clypeus dichter und starker punktirt, der Scheitel langsrissig. Thorax doppelt so breit wie lang, von der Basis zur Mitte schwach, davor stirker gerundet verengt, in der Mitte der Scheibe ziemlich dicht und fein, nach dem Seitenrande hin dichter und etwas stirker punktirt. Fitigeldecken in zwolf Reihen fein punktirt; die erste und elfte Reihe kurz, die neunte vorn und hinten abgekiirzt, alle Reihen (mit Ausnahme der zehnten) werden hinter der Mitte feiner, bleiben aber bis zum Ende ziemlich deutlich. 189287——3 450 The Philippine Journal of Science 1922 Nodina santula sp. nov. Late ovata, convexa, subtus nigra, palpis, antennis pedibus- que rufo-testaceis, supra aeneo-viridis, aurichalceo-micans, nitida, prothorace minus dense subtilissime, latera sat rotundata versus crebrius et fortius punctato, elytris striato-punctatis, punctis apicem versus deletis. Long., 1.4 ad 2 mm. Luzon, Laguna, Los Bafios, Mount Maquiling; Mount Bana- hao (4762, 6153), Leyte, Mindanao. Var. a. Femoribus posticis piceis. | Mit Nodina pusilla Motsch. am nichsten verwandt, aber der Thorax nur an den Seiten dicht punktirt. Gerundet, unten schwarz, oben metallisch dunkel griin mit Messingschimmer, stark glinzend, Taster, Fiihler, und Beine rotlich gelbbraun. Kopf fein zerstreut punktirt, der Clypeus deutlich von der Stirn getrennt. Halsschild bedeutend breiter wie lang, vorn zusam- mengedriickt und starker als hinten queriiber gewolbt, in der Mitte wenig dicht fein punktirt, an den Seiten dichter und stirker, die Seiten sind gerundet und convergiren hinten wenig, vorn viel mehr. Fliigeldecken in den Schultern kaum breiter wie der Thorax, bis zur Mitte unmerklich erweitert, dann verengt und hinten gemeinschaftlich schmal abgerundet, regel- missig gereiht punktirt. Die Punkte sind vorn fein und erléschen hinter der Mitte allm&hlich, nur die beiden ersten Reihen lassen sich bis zur Spitze verfolgen. Der Ausschnitt an der Spitze der Hinterschienen ist sehr undeutlich. Zuweilen sind. die Hinterschenkel pechbraun. Tabelle der neuen Nodina-Arten. 1. Oberseite schwarzblau, Fiihlerbasis rotgelb, neunte Punktreihe der Fliigeldecken deutlich. Linge, 2.5mm, Luzon...... N. luzonica sp. nov. 1. Oberseite griinlich schwarz bronceschimmernd, neunte Punktreihe der Fliigeldecken héchstens durch einige Punkte angedeutet. Linge, 1.4 bis 2mm. Luzon, Leyte, Mindanao N. santula sp. nov, Phaedroides philippinensis Lefévre. Ein Hauptmerkmal dieser Gattung liegt in der Bildung der Vorderbrust-Episternen. Diese werden seitwarts durch eine seichte, glatte Rinne begrenzt iiber der ein scharf gerandeter, umgeschlagener, paralleler Lingsstreifen des Thorax liegt, der wie dieser lebhaft metallisch griin gefarbt ist und vorn in eine weit iiber den Vorderrand der Episternen vorgezogene Spitze (die Vorderecken des Thorax) endet. Die einzige bis jetzt bekannte Art ist oberseits metallisch griin, blau oder kupferrot, 21,5 Weise: Chrysomeliden der Philippinen, III 451 die unterseite schwarz, oft griinlich oder bliulich schimmernd, die unter sich ziemlich gleichlangen vier ersten Fiihlerglieder rotbraun, die folgenden etwas verdickt, schwarz. Kopf und Halsschild ziemlich dicht und massig fein punktirt, Fliigeldecken mit kraftigen Punkten, die in ungefihr zwanzig nicht iiberall regelmassig Reihen gestellt sind. Besonders unregelmdssig pfiegt die dritte und vierte Reihe zu sein. LUZON, Benguet, Baguio (6122). . Genus PAGRIA Lefévre Die beiden bisher auf den Philippinen gesammelten Arten verbreiten sich von Japan tiber die Sunda-Inseln und das Fest- land bis Ostindien; es sind: Tabelle der Pagria-Arten. 1. Stirn sparsam und sehr fein punktirt, Augenrinnen breit, innen von einer hohen Kante scharf begrenzt. Thorax weitlaufig, in der Mitte viel feiner wie an den Seiten punktirt. Unterseite rostrot bis pech- schwarz, Fihler und Beine hell rétlich gelb, Kopf und Halsschild rotbraun, oft zum Teil angedunkelt, Fliigeldecken gelb, ein Naht- und Seitensaum sowie eine Querbinde in dem Eindrucke hinter der Basal- beule schwarz. Diese Binde ist auf jeder Decke oft in zwei breite und lange Spitzen nach hinten und in eine Linie auf der sechsten punkt- reihe nach vorn ausgezogen. Linge, 2 bis 2.3mm. Los Bafos, Mount Maquiling (2441) ... P. signata Motschulsky. | 1. Stirn ziemlich dicht punktirt, Augenrinnen massig breit, innen nicht scharf begrenzt, Halsschild ziemlich dicht und fast gleichmissig punktirt. Schwarz. Kopf und Thorax oft pechbraun, Fihler, Beine, und ein kurzer Lingsstrich an der Basis jeder Fliigeldecke innen von der Schulterbeule blass rétlich gelb. Linge, 1.8 bis 2mm. Luzon: Los Bajios, Mount Maquiling; Benguet, Baguio; Tayabas, Malinao (6129) P. flavopustulata Baly. Aendert ab. Fliigeldecken hell braunlich gelb, einfarbig, oder ein Naht- und Seitensaum schwirzlich. ab. bicolor. Tabelle der Scelodonta-Arten von den Philippinen. 1. Hinterbrust an der Seite mit einem kahlen, glanzenden, leistenférmigen Langsstreifen, neben dem ein 4hnlicher Streifen der me ei lie; : : i oped und deren Episternen gleichmiassig weisslich behaart. Ober- seite einfarbig braun kupferig, glinzend. Linge, 3.5 bis 4mm. La- zon: Los Bafios (647), Palawan §. dillwyni Stephens. 2. Oberseite matt, dusserst dicht runzelig punktirt, dunkel gefleckt.. eae 3. 2. Oberseite glinzend, einfarbig metallisch griin, blau oder kupferig braun, Fliigeldecken in Reihen punktirt, von denen die vier inneren nahe der Basis mehr oder weniger verdoppelt, die vier folgenden etwas unre- gelmissig, die ausseren ‘ganzlich, und die iibrigen vor der Spitze in 452 The Philippine Journal of Science 1922 Streifen gestellt sind. Linge, 4 bis 4.5 mm. Mindanao: Dapitan und Tliggan (4748) -rescssseecesrrensescsenseceesssetettes §. dispar Lefévre. 3. Metallisch lebhaft griin, oberseits grosstenteils mit schwarzen oder purpurfarbenen Makeln bedeckt, Schildchen, Schenkel, und Schienen leuchtend kupferrot, Tarsen schwarzlich, die ersten sechs Fiihlerglieder griin, die erweiterten folgenden blaulich schwarz, Vorder- und Mittel- prust nebst dem Bauche messingfarbig. Lange, 5 mm. Luzon, Min- danao §. curculionoides Westwood. 8. Diister schwarzlich braun mit einem blassen Metallschimmer, die ersten sechs Fiihlerglieder, Schenkel, und Schienen dunkel kupferig, Fli- geldecken mit drei wenig hervortretenden kupferig-schwarzen Quer- binden. Lange, 4 bis 5 mm. Luzon: Los Banos (24,).------c-ece--s-eceee §. aeneola Lefévre. Pagellia sexmaculata sp. nov. Oblongo-ovalis, subtus flava, lateribus pectoris, femoribus apicem versus basique tibiarum nigris, capite prothoraceque rufis, hoc fortiter punctato, elytris profunde punctato-striatis, nigris, singulo maculis tribus transversis flavis. Long., 2.5 mm. Luzon, Laguna, Mount Maquiling. Die kleinste und schlankste Art der Gattung. Sehr lang eiférmig, Fiihler, Taster, Beine, und Bauch gelb, letzterer vorn mehr gebraunt, die Mitte der Brust rotbraun, die Seiten nebst der oberen Hilfte der Schenkel und der Schienenbasis schwarz, Kopf, Thorax, und Schildchen rot, Fliigeldecken schwarz, jede mit drei gelben Quermakeln: die erste an der Basis vor dem Quereindrucke, dehnt sich tiber drei Intervalle von der zweiten bis in die fiinfte Punktreihe aus; die zweite, unmittelbar hinter dem Quereindrucke, liegt zwischen der dritten und sechsten Punktreihe und verschmiilert sich nach aussen; die dritte, vor der Spitze, ist dreieckig, vorn und innen gradlinig begrenzt, hinten und aussen zugespitzt. Kopf wenig dicht und kraftig punktirt. Halsschild so lang als breit, stark queriiber gewolbt und nach vorn abfallend, an den Seiten massig gerundet, nach vorn und hinten gleichmissig verschmilert, die Scheibe in der Mitte dicht und wenig stirker wie die Stirn, an den Seiten viel dichter punktirt. Fliigeldecken an der Basis in grader Schrag- linie heraustretend und in den Schultern breiter wie das Halsschild, bis hinter die Mitte wenig erweitert, dann schnell verengt und am Ende schmal gemeinschaftlich abgerundet, mit den normalen tiefen dreizehn Punktstreifen, von denen der erste und zwolfte sehr kurz sind und der neunte und zehnte, vorn verbunden, erst hinter der Schulter beginnen. Die inneren vier Intervalle sind miassig breit, gewdlbt, die folgenden schmal, hoch, und scharf. . 21,5 Weise: Chrysomeliden der Philippinen, III 458 Pagellia butuanensis sp. nov. Oblongo-ovalis, dilute testacea, nitida, prothorace latera versus subinfuscato, creberrime subtiliterque strigatim punctato, juxta marginem anticum alutaceo et parce punctulate, elytris infra basin obsoletius impressis, profunde punctato-striatis, inter- vallis convexis. Long., 3 mm. MINDANAO, Butuan. Den kleinen Stiicken von Pagellia suturalis sehr ahnlich, hell rotlich gelbbraun, das Halsschild wenig glinzend, nach den Seiten hin etwas angedunkelt; fein und sehr dicht langsrunzelig punktirt, die Runzeln linger und kraftiger wie die von P. sutu- ralis, am Vorderrande nur zart gewirkt und zerstreut punktulirt. Fliigeldecken glinzend, hinter der Basis kaum merklich quer eingedriickt, tief punktirt-gestreift, mit gewdlbten Intervallen, von denen das neunte bis elfte schmal und rippenférmig ist. Pagellia suturalis Lefévre. Auf Luzon die haufigste Art, in der Farbung von fast einfar- big bradunlich gelb bis pechschwarz variirend, jedoch bleibt die Basalbeule der Fliigeldecken blass gelblich gefirbt, selbst in den hellsten Stiicken. Die mir vorliegende dunkelste Form ist pech- schwarz, Fiihlerbasis, Tarsen, ein Aussensaum des Bauches, und wenigstens die vordere Halfte der Basalbeule auf den Fliigel- decken rétlich gelbbraun. Die hinteren zwei Drittel des Halsschildes sind sehr dicht, fein punktirt, und kurz langsrun- zelig, fast matt, ein Streifen vor der Basalrinne glatt und das vordere Drittel zart gewirkt, weitliufig punktirt, glanzend. Die Piinktchen werden nach aussen und lings des Seitenrandes stark und tief. Die Gattung Pageilia ist durch die auffallig stark punktirt- gestreiften Fliigeldecken und deren leistenformige Intervalle, sowie durch die tief herabgebogenen Seiten des Halsschildes aus- gezeichnet; der Punktirung nach scheint P. foveolata Lefévre nicht hinein zu gehéren, dagegen die Lindinia lefevret Jacoby, die mit P. suturalis verwandt sein diirfte. Luzon, Laguna, Mount Maquiling, Mount Banahao (2093). Tabelle der Pagellia-Arten. 1. Fliigeldecken mit deutlicher Basalbeule, die hinten durch einen kraftigen Eindruck emporgehoben wird ... z. 1. Fliigeldecken hinter der Basis kaum bemerkbar eingedriickt. 5. 2. Halsschild glanzend, stark queriiber gewélbt, vorn so breit wie hinten, annihernd cylindrisch, dicht und ziemlich grob punktirt, Fligel- decken schwarz, jede mit drei blassgelben Quermakeln (1, 1, 1). Lange, 2.5 mm. Luzon ss P. sexmaculata sp. nov. 454 The Philippine Journal of Science 1922 2. Halsschild vorn viel schmaler als hinten, auf der Scheibe abgeflacht und sehr dicht, ziemlich fein, mehr oder weniger langsrunzelig punktirt, fast matt 3. 8. Thorax auf einem Querstreifen vor der Basalrandlinie glatt, nahe dem Vorderrande zerstreut punktirt und nicht langsrunzelig wie auf dem dahinter liegenden Teile. Ro6tlich gelbbraun bis pechschwarz, die Naht und die verdickten Endglieder der Fiihler schwarzlich, die Basalbeule der Fliigeldecken stets heller als die tibrige Scheibe. Lange, 3 bis 3.5mm. Luzon P. suturalis Lefévre. Hierher vielleicht Lindinia lefevrei Jac., bei der die inneren Punkt- streifen der Fligeldecken nahezu unpunktirt sein sollen. 8. Thorax bis an alle Rander runzelig punktirt 4, 4. Halsschild am Vorderrande weniger dicht langsrunzelig punktirt wie in der Mitte. Fligeldecken rétlich braun mit einigen helleren Makeln. Lange, 3 mm. Luzon P. acuticosta Lefévre. 4. Halsschild in der Mitte sehr dicht und fein narbig punktirt, neben dem Vorder- und Seitenrande stirker und langsrunzelig punktirt. Fli- geldecken rétlich gelb mit je drei schwarzen Makeln (1, 1, 1). Lange, 38 mm. Luzon P. signata Weise. 5. Fliigeldecken gleichmiassig tief punktirt-gestreift. Lange, 3 mm. Min- danao: Butuan P. butuanensis sp. nov. 5. Fliigeldecken vorn in starken Reihen punktirt, hinter der Mitte fast glatt. Lange, 3 mm. Luzon P. foveolata Lefévre. Genus ICOGRAMMA novum Corpus subovatum dense brevissimeque pubescens. Caput exsertum frons deplanata, antennae gracilis, dimidio corporis breviores, articulis 6 ultimis paullo incrassatis. Prothorax brevis, antrorsum angustatus, lateribus haud marginatis. Ely- tra striis punctatis circiter viginti impressa. Prosternum valde transversum, episternis antice concavis. Pedes sat graciles, femoribus parum incrassatis, muticis, tibiis quatuor posteriori- bus subemarginatis. Unguiculi fissi. Mit Malegia verwandt, jedoch durch das sehr kurze und breite Prosternum und die enge, neben einander liegenden, ungefahr zwanzig Punktstreifen der Fliigeldecken verschieden. Die Beine sind ziemlich kurz, schlank, Schenkel in der Mitte etwas verdickt, der Ausschnitt an der Spitze der vier Hinterschienen kurz und flach, dicht bewimpert. Icogramma lineigera sp. nov. Testacea, subopaca, dense brevissimeque albido pilosa, pedibus antennisque flavis, his apice infuscatis, elytris creberrime punc- tato-striatis, intervallis angustis, convexis, uniseriatum punc- tulatis. Long., 2 ad 2.5 mm. 7 ie Puerto Princesa. Luzon, Benguet, Baguio (4750, 135). 21,5 Weise: Chrysomeliden der Philippinen, III 455 Annahernd eiférmig, massig gewolbt, blass rétlich gelbbraun, ‘ziemlich matt, Beine, Taster, und Fiihler gelb, an letzteren die fiinf Endglieder angedunkelt, der Bauch zuweilen schwirzlich. Kopf massig gross, Stirn fast eben, dusserst zart punktulirt und anliegend weisslich behaart. Thorax breiter wie lang, nach vorn verengt, an den Seiten wenig gerundet, ohne Randleiste, stark queriiber gewolbt und 4ahnlich wie die Stirn punktulirt, die Piinktchen aber durch die dichte anliegende Behaarung verdeckt. Die Harchen sind weisslich und von der Mittellinie schrag nach hinten und aussen gerichtet. Schildchen rechteckig. Fliigel- decken in den Schultern bedeutend breiter wie der Thorax, bis hinter die Mitte ziemlich parallel, dann verengt und schmal ab- gerundet, auf dem Riicken dicht gestreift und in den Streifen mit einer dichten Reihe von Punkten besetzt, von denen jeder ein sehr kleines, weisses, schuppenférmiges Harchen tragt, waihrend die Schmalen, leistenformigen Intervalle mit einer Reihe von etwas langeren, aber viel feineren und weniger hell gefarbten Harchen besetzt sind. Vorderhiiften weit getrennt, das Pro- sternum daher kurz, breit, nach vorn und hinten abfallend. Icogramma obscura sp. nov. 3 Nigra, opaca, pube grisea brevissima dense vestita, antennis _ (apice vix infuscatis) pedibusque flavis, elytris crebre subtili- terque striato-punctatis, intervallis latera versus convexiusculis. Long., 1.8 ad 2 mm. PALAWAN, Puerto Princesa. Kleiner und etwas schlanker als die vorige, oberseits gleich- miissig dicht, sehr kurz und fein greisbehaart, die an der Spitze kaum angedunkelten Fiihler nebst den Beinen gelb, die Fliigel- decken feiner in dichten Reihen punktirt, welche innen kaum, aussen etwas vertieft und hier mit massig gewélbten Interval- len versehen sind. Aulexis gracilicornis sp. nov. ¢. Elongata, pallide testaceo-flava, supra dense subtilissime et brevissime griseo pubescens, subopaca, antennis gracilibus, corporis longitudine parum brevioribus. Long., 4 ad 5 mm. LUZON, Benguet; Baguio (6141). Femina (?): Antennis nigris, articulis duobus primis flavis. Stirn nur missig dicht fein punktirt, Kopfschild langer wie bei den iibrigen Arten, wenig quer, fast quadratisch, nicht dicht und bedeutend stirker punktirt als die Stirn. Fihler des Minnchens so lang wie der Korper, Glied 3 mindestens doppelt 456 The Philippine Journal of Science ‘gles so lang als 2 und fast so lang als 1, die folgenden unter sich ziemlich von gleicher Lange, jedes wenig linger als 8, die fiinf Endglieder unbedeutend verdickt. Bei einem mit dem Mann- chen zusammen gefangenen Weibchen, welches ich nur fraglich zu dieser Art stellen kann, sind die Fihler etwas kiirzer, die Endglieder merklich starker, Glied 3 schwirzlich, die folgenden schwarz, von den Gliedern 6 bis 10 jedes kiirzer wie Glied 4 oder 5. ‘Thorax normal gebaut, vor dem zweiten Quereindrucke nicht dicht und dusserst fein, dahinter dichter und etwas starker punktirt. Die Punktirung der Fliigeldecken ist dichter und stirker wie die an der Thoraxbasis. Korper blassgelb mit einem leichten rétlich braunen Anfiuge. Aehnlich lange Fiihler besitzt Aulexis longicornis Jacoby von Sumatra, aber bei ihr ist Glied 3 so kurz wie 2. Aulexis brevicornis sp. nov. Subelongata, flavo-rufa, sat nitida, antennis brevibus, articulis 5 ultimis incrassatis elytrisque nigris, his cum prothorace obso- lete biimpresso sat crebre subtilissimeque punctatis. Long., 2.5 ad 3 mm. PALAWAN, Puerto Princesa. -Ab. a. Supra omnino brunneo- flava. : Eine kleine, durch den stirkeren Glanz der Oberseite und kurze’ Fiithler ausgezeichnete Art, die sich von der dhnlichen Aulexis pusilla Lefévre schon durch die schwachen, verlosch- enen Quereindriicke des Thorax entfernt. Weniger schlank gebaut wie die vorhergehende Art, gelblich rot, Glied 5 und 6 der Fiihler angedunkelt, die folgenden Glieder nebst den Fli- geldecken schwarz. Die Fiihler reichen bis hinter die Schulter, Glied 2 und 6 sind klein, 4 wenig, 3 und 5 deutlich linger, die folgenden verdickt, 7 so lang wie 1, 8 bis 10 jedes kurz, an der Spitze fast so breit wie lang. Clypeus quer, sparsam, die Stirn etwas dichter punktulirt und behaart. Thorax quer, von der Basis bis zur Mitte erweitert, davor ausgerandet-verengt, die Seitenzihnchen sehr klein, verloschen, die Scheibe vorn 4us- serst fein, hinter der Mitte wenig starker und nicht besonders dicht punktulirt, mit zwei angedeuteten Quereindriicken. Fiii- geldecken bedeutend stirker wie das Halsschild vor der Basis punktirt, jedoch weitlaufiger wie in den tibrigen Arten, daher glanzender als diese. Bei der ab. a, die vielleicht aus nicht ganz ausgereiften Stiicken besteht, ist die Oberseite einfarbig braunlich gelb. . 21,5 Weise: Chrysomeliden der Philippinen, III 457 Aulexis flavopilosa Lefévre. LuzOoN, Mount Banahao; Paete (2793). Aulexis luzonica Lefévre. LuzON, Laguna, Mount Maquiling: Benguet, Baguio (6142). Die Gattung Aulexis lasst sich mit Sicherheit an dem ge- streckten annihernd cylindrischen, oberseits dicht anliegend und auf den Fliigeldecken ausserdem noch sparsamer aufgerichtet- behaarten Korper, sowie dem Seitenrande des Thorax erkennen, welcher nahe der Mitte in der Regel in einen dreizdhnigen Vorsprung erweitert, davor ungerandet ist, waihrend an den Seiten der Scheibe zwei Quereindriicke hintereinander liegen. Das Kopfschild endet vorn in zwei abgeflachte und durch einen bogenformigen Ausschnitt getrennte Zahnchen, von denen jedes in .eine Vertiefung der Oberlippe eingreift. Die Fliigel sind schwarzlich, metallisch griin oder kupferrot iiberflogen. Der Hinterrand des letzten Bauschsegmentes ist entweder in der Mitte gradlinig abgestutzt oder leicht gerundet-vorgezogen (Weibchen) oder bogenférmig ausgerandet (M&annchen). Die schwer zu unterscheidenden Arten verbreiten sich von Ostindien tiber die Kiistenlinder und die Sunda-Inseln bis zu den Philippinen ; ihre Grésse und Farbung scheint sehr veranderlich Zu sein. Tabelle der Aulexis-Arten. 1. Oberseite dicht und verhaltnismassig lang goldgelb oder gelblich behaart, so dass die Skulptur fast ganz verdeckt wird 2. 1. Die Behaarung der Oberseite ist zwar dicht, aber fein, kurz, und ver- deckt die Punktirung nicht... —- 2. Oben schwarz, unten rotbraun, Brust angedunkelt bis schwarz, Ober- lippe, Taster, Fiihler, und Beine rétlich gelbbraun. Lange, 4.5 bis 6 mm. Luzon, Paete A. flavopilosa Lefévre. 2. Rotbraun, nur die sieben Endglieder der Fiihler und die Fliigeldecken schwarz. Linge, 6.8 mm. Ph