Or PUY \A\\\ oe THE PHILIPPINE JOURNAL OF SCIENCE f | VOLUME XIV JANUARY TO JUNE, 1919 WITH 49 PLATES AND 30 TEXT FIGURES BUREAU OF PRINTING 919 == é MANILA (> ~ EDITORIAL BOARD ELMER D. MERRILL, M.S., Editor R. C. McGrecor, A.B., Associate Editor ALBERT H. WELLS, A.B.; J. R. WRIGHT, Pu.D.; A. P. West, PH.D. T. Dar JUAN A.B., PHaR.D.; F. AGcaori, A.B. ; F. D. Reyss, B.S. A. 8. ArGtELLES, B.S.; VicToRIANO ELICANO, B.S. Chemistry, Physics, and Geology H. W. Wabg, M.D.; Orro ScHiésu, M.D. F. G. HAuGHWouT; STANTON YOUNGBERG, D.V.M. Experimental Medicine Lrpor1o Gomez, M.D., Pu.D.; F.. CALDERON, B.A., L.M. VICENTE DE JESUS, M.D. Clinical Medicine W.#H. BRowN, Pu.D.; C. F. BAKER, M.A.; H. S. Yares, M.S., Pu.D. O. A. REINKING, B.S.A., M.S.; L. M. Guerrero, PHar.D. Botany E. H. Tayzor, A.B.; R. P. Cowes, Pu.D, C. F. Baker, M.A.; S. F. LicHT, M.A.; C. S. BANKs, M.A. L. D. WHARTON, M.A.; W. SCHULTZE; H. O. BEYER, M.A. Zoblogy and Ethnology A. B. BANyEA, Copy Editor CONTENTS No. 1, January, 1919 {Issued July 3, 1919.] Weis, A. H. The physiological active constituents of certain Phil- ippine-medicinal plants +: Wd.25.. secede Bae ce hae aa One plate. SALEEBY, N. M. The treatment of human beriberi with autolyzed yeast ci 418 2 |] Nae OSE Ae ar Ria eee Peed io tecr t Mores) peya mr eee nope ee espe | DENNEY, OSWALD E. A photographic study of leprosy. Four plates. LANTIN, Pepro T. A comparative study of different methods of treatment of typhoid fever.......................-...--- ee Ten text figures. . MERRILL, E. D., and WaAprE, H. W. The validity of the name Dis- ' eomyces for the genus of fungi variously called Actinomyces, Streptothrix, and Nocardia BANKS, CHARLES S. Iontha ida, a new Philippine noctuid One plate. COCKERELL, T..D. A. The social bees of the Philippine Islands............ CocKERELL, T. D. A. The Philippine bees of the family Nomadidz.... SHuFELDT, R. W. The osteology of the giant gallinule of the Philip- -pines, Porphyrio pulverulentus Temminck: With notes on the osteology of Tachybaptus philippensis (Bonnaterre) and Hydro- phasianus chirurgus (Scopoli) Five plates. TAyYLor, Epwarp H. New or rare Philippine reptiles... Two plates and four text figures. TAYLOR, Epwarp H. Ipon fisheries of Abra River. REINKING, Otto A. Phytophthora faberi Maubl.: The cause of coco- nut bud rot in the Philippines......... te ~ ‘Three plates. Proceedings of the Manila Medical Society.............----------------1--eeeeeee No. 2, February, 1919 [Issued September 5, 1919.] BANKS, CHARLES S. Phlebotomus nicnic, a new species, the first Phil- ippine record for this genus One plate. iii Page. 13 19 105 127 131 153 155 163 iv Contents . BANKS, CHARLES S. The bloodsucking insects of the Philippines........ COCKERELL, T. D. A. The Prosopide, or obtuse-tongued bees, of the Pe AUMNNCN a COCKERELL, T. D. A. The Philippine bees of the families Anthopho- erm ee I CARREON, MARCIANO. The absence of both hind legs below the femur Be Cs ge One plate. HAUGHWOUT, FRANK G., and DE LEON, WALFRIDO. On the ingestion of erythrocytes by Pentatrichomonas sp., found in a case of I a ap ee One plate. Wirt, J.C. The effect of calcium sulphate on cement: Second paper.. Two text figures, PANGANIBAN, C. S., and Scuést, O. Experience with methylena blue- eosin lactose agar in searching for Bacillus dysenteriz in stools... MERRILL, E. D. Notes on the flora of Sumatra No. 3, March, 1919 [Issued September 18, 1919.] Acosta-SIson, Honoria, and CALDERON, FERNANDO. Pelvimetry and cephalometry among Filipino women and newborn babies: Made on one thousand two hundred thirty-seven cases Five text figures. AGuitar, R. H. The lumbang-oil industry in the Philippine Islands... Fett, E. P. New Philippine gall midges ee ie Ao Wen eT Wet EbUE Is TeleediNckEccawka ne cauks Beccart, O. The palms of the Philippine Islands Three plates. eg aN een MOE Re et RASA E DWEkNT Sheed bbe Usa: 5d6 ce No. 4, April, 1919 [Issued September 19, 1919.] MERRILL, E. D. New or noteworthy Philippine plants, XV No. 5, May, 1919 {I d December 1, 1919.] ST ae ores Tone Se een eealne sinks SaN ey enw es he deeneen apt chy ia ty Three text figures. dyna hevahntctataieitemethete: ta hit Cn ee PN + We wewecccece Papua, REGINO G. Cystolithiasis among Filipinos in association with dietetic deficiency One text figure, SORA RC SE Hane eRe Sth ee ehebsnkbecuk aun 195 201 207 221 235 239 251 253 275 287 295 363 365 459 Contents Yates, Harry §. The growth of Hevea brasiliensis in the Philippine PERT sti) gpupeee oe cae ree tense hele SPATE UC AGETEN Sipe Sue dear ar "runt ep Ge WAM e Sha Nr Oe TO One text figure. : UICHANCO, LEOPOLDO B. A sctigleed and systematic study of Phil- pee to egy Cs k ge «811 aptconaeeere it pent esspencny uberis em ah Mes ery ot) marstise en ne rp Fifteen plates. COCKERELL, T. D. A. The sphecodine bees of the Philippine Islands.... No. 6, June, 1919 [Issued December 16, 1919.} Kinc, ALBERT E. W. The mechanical properties of Philippine bast- 1S 90 1, Same ee asa ae sre tort en ele ay orm EP Tremere er oer karen tO” ferret Pee Five plates and two text fais Lee, H. ATHERTON, and YATES, Harry S. Pink disease of citrus........ Seven plates and two text figures. ERRATA st a es ES eee en ate eee OE Suid vas cewctn cae TT a or See eens : Pies ANNOUNCEMENT With the beginning of 1919 the sections of the Philippine Journal of Science will be combined, and the publication will be issued as a monthly; each number will be larger than the present bimonthly issues. The policy with regard to the char- acter and scope of the material published will be continued. The Journal is intended to be as it has been, the scientific organ of the Philippine Government, and will be devoted to the scien- tific and commercial interests of the tropics. Its aim will be to collect and publish in one place original scientific informa- tion and material relating to the Philippine Islands. Suitable articles will receive prompt publication in the Philippine Journal of Science, and specialists working on Philippine material will continue to receive hearty codperation. The subscription price will be 5 dollars United States cur- rency per annum. A refund of 2 dollars will be made to anyone that has forwarded a subscription for the four sections of the Journal at the old rate of 7 dollars. Any sum received as a subscription to a single section. will be credited, and a bill will be rendered for the balance due at the new rate. The Philippine Journal of Science in its new form will be sent in exchange to all periodicals and institutions with which exchange relations are in force. THE EDITORS. MANILA, P. I., December 31, 1918. 162323 THE PHILIPPINE JOURNAL OF SCIENCE VoL. XIV JANUARY, 1919 No. 1 THE PHYSIOLOGICAL ACTIVE CONSTITUENTS OF CERTAIN PHILIPPINE MEDICINAL PLANTS: III By A. H. WELLS (From the Laboratory of Organic Chemistry, Bureau of Science, Manila) ONE PLATE CONTENTS ARCANGELISIA FLAVA (LINN.) CASSIA SIAMEA LAM. Merr. (A. LEMNISCATA BECcc.). GEODORUM NUTANS AMES. MENISPERMACEAE. CORIARIA INTERMEDIA MATs. Berberine has been found in the following plants: Berberis aquifolium. (1) Caulophyllum thalictroides. (14) Jateorhiza palmata. (2) Leontice thalictroides.(15) Hydrastis canadensis. (3) Berberis aetnensis. (16) Menispermum canadense. (4) Cocculus palmatus.(17) Argemone mexicana. (5) Chelidonium majus. (18) Coptis trifolia.(6) Styllaphorum diphyllum.(18) Xanthoxylum americanum. (7) Coscinium fenestratum. (19) Berberis asiatica.(8) Xanthorriza apiifolia. (20) Berberis aristata.(8) Xanthoxylum senegalense. (21) Berberis lycium. (8) Thalictrum flavum. (22) Podophyllum peltatum. (9) Ptelea trifoliata. (23) Berberis vulgaris. (10) Xylopia polycarpa. (24) Coptis tecta.(11) Geoffrayra inermis. (25) Xanthoxylum clava-herculis. Evodia glauca. (26) (12) Jeffersonia diphylla. (27) Coeloeline polycarpa. (13) Toddalia asiatica. (28) ARCANGELISIA FLAVA (LINN.) MERR. (A, LEMNISCATA BECC.). MENISPERMACEAE Philippine names: Albustra, abutra (Ilocano) ; lagtal, lagtan, albotra (Visayan); lagtang, bute, ligtang, suma (Tagalog) ; suma (Pampango). 162323 2 The Philippine Journal of Science 1919 The names albustra, abutra, albotra, and bute are all corrup- ' tions of the Spanish-American name abuta, which was brought from Mexico by the Spaniards. These names are also applied to Anamirta cocculus W. & A. of the same family, which has much smaller fruits and white, not yellow, wood. It is a coarse, woody vine, the stems up to 10 centimeters in diameter, with a characteristically yellow wood. The species is widely distributed in the Philippines, occurring in most, if not all, of the larger islands and provinces. However, it is local and widely scattered in thickets and forests, occurring especially on rocky slopes. No data as to its abundance were obtainable. Chemical examination.—The soft porous wood was freed from the thin outer bark, finely ground while fresh, and macerated with 95 per cent ethyl alcohol. The solution resulting from this exhaustion was quickly evaporated under reduced pressure to one-fifth its volume; it was then cooled and salts made with hydrochloric, nitric, sulphuric, and hydriodic acids. A few grams of the pure alkaloid were obtained by an aqueous extrac- tion of portions of the ground wood in a neutralized solution. Repeated extraction and evaporation with alcohol, and final crys- tallization from neutral ethyl acetate in a vacuum desiccator over calcium chloride, gave beautiful golden yellow crystals of the berberine base. When compounds of this ‘nature are crys- tallized in a vacuum desiccator over sulphuric acid it is notice- able that traces of acid sulphates are formed. This was observed particularly when the acetone compound was placed in such a desiccator. The compound was identified as berberine. Large quantities of salts of the alkaloid were easily crystal- lized out with nitric, sulphuric, and hydrochloric acids. Upon heating, the sulphate melts to a brown amorphous mass, slightly soluble in water; it has an intense yellow color and a bitter taste, is soluble in alcohol and in hot water, and is almost insol- uble in ether and benzene. The solution allows precipitation with alkaloidal reagents and also with potassium iodide solu- tion. In concentrated sulphuric acid it gives an olive green color; a fragment of sodium nitrate added to the sulphuric acid solution gives a characteristic violet color. The acetone com- pound was obtained by the method of acetone treating the sul- phate of the compound in an alcoholic solution. The resultant compound, which is represented as C,,H,,NO,: C,H,O, was also obtained by treating & hot solution of 50 grams of the crystalline berberine sulphate in 1 liter of water. Five hundred grams eg pag { ~~ XIV, 1 Wells: Philippine Medicinal Plants: III 3 of acetone were added to this solution; after mixing, it was made alkaline with sodium hydroxide. This is a method of Gaze.(29) A greenish yellow crystalline substance separates therefrom (see Plate I). The pure crystalline berberine ob- tained by the decomposition of the berberine acetone compound is dark golden brown, but it is slightly lightened in color by purification with alcohol. Quantitative determinations made on the plant substance show approximately 4.8 per cent of pure alkaloid in the moisture- free material. Combustion of the alkaloid, when freed from as much water of crystallization as is possible without apparent decomposition, gave the following results: Determination No. 1.—0.15 gram of substance gave 0.0788 gram of water and 0.3876 gram of carbon dioxide. Determination No. 2.—0.15 gram of substance gave 0.0771 gram of water and 0.3854 gram of carbon dioxide. Theoretically, the combustion of water-free berberine gives carbon 71.64 per cent and hydrogen 5.07 per cent. Determina- tion No. 1 shows carbon 70.47 per cent and hydrogen 5.84 per cent. Determination No. 2 shows carbon 70.47 per cent and hydrogen 5.63 per cent. Nitrogen found in the alkaloid amounted to 4.03 per cent. Arcangelisia flava (Linn.) Merr. is believed to have a larger percentage of alkaloid than any other plant found in the Phil- ippines; and, should the therapeutic value of berberine become very important, even a small plant for the recovery of this al- kaloid would develop into a paying industry. The above cal- culations, together with the qualitative tests expressed, are suffi- cient to establish the Per of berberine in Arcangelisia flava (Linn.) Merr. CASSIA SIMEA LAM. RissbweNeeae) In the Philippines this tree occurs in cultivation only. It is commonly planted as a shade tree in Manila and in other large towns of the Archipelago. It is known by its Spanish name acacia, which is also applied to other trees of the same fam- ily. Cassia florida Vahl is a synonym. I can find no record of the tree serving any other useful purpose in the Philippine Islands than that of giving shade. : A search of the literature revealed no record of any active principles found in this plant. However, a recent communica- tion from Father Algue, director of the Philippine Weather 4 The Philippine Journal of Science 1919 Bureau, states that the feeding of fresh branches and pods of this tree had caused the death of many of their hogs. For this reason a chemical analysis was made of the pods, branches, and leaves and the presence of a poisonous alkaloid established. 0.1015 gram of alkaloid gave 0.0725 gram of water and 0.2463 gram of carbon dioxide, which gave 7.93 per cent hydrogen and 66.18 per cent carbon; 0.1000 gram of alkaloid gave 0.0685 gram of water and 0.2385 gram of carbon dioxide, equal to 7.61 per cent of hydrogen and 65.04 per cent of carbon; 0.2016 gram of alkaloid gave 0.1415 gram of water and 0.4971 gram of carbon dioxide equal to 7.80 per cent of hydrogen and 67.25 per cent of carbon. Percentages of carbon, hydrogen, and nitrogen in alkaloid from Cassia siamea. Carbon. | Hydrogen.} Nitrogen. WMI Be Socar art iat eas spo es ohn 66.18 7.93 5.39 NE Bi Rit it ae ce 65. 04 A } Ss SE EES MON Pi ipspencr eons aca 67. 25 7.80 5.60 Le nea Ey RISC ORNE a Aaah a ent 198. 47 28.34 10. 99 BIONRG ks aac ice a SS 66. 16 1.78 5.495 These figures furnish data for an empirical formula for the alkaloid found in Cassia siamea, namely C,,H,,NO,. With in- traperitoneal injections of 1 cubic centimeter of a 5 per cent solution of the hydrochloride large guinea pigs readily exhibited symptoms of poisoning, usually resulting in death. GEODORUM NUTANS AMES Philippine names: Camaog, lubi lubi (Visayan) ; cola (Taga- log, from the Spanish word cola, meaning “glue’”’), This plant is a terrestrial orchid that is found in thickets and in uncultivated open lands at low altitudes throughout the Phil- ippines, generally as scattered individuals, but occasionally abundant. Its only known use is that of an adhesive. The rhizomes are split and the pulp rubbed upon the surfaces to be glued. The plant is used throughout the Islands as a glue in the manufacture of stringed instruments. The water extraction of the plant gives a gum which is separated by precipitation with alcohol. This gum has exceptional strength and lasting power. The moisture content of the plant is 79.5 per cent. The yield of dry gum on the fresh plant is about 14 per cent, and on the moisture-free Sample, 68.8 per cent. No substances of therapeutic value were found. {> a? XIV,1 Wells: Philippine Medicinal Plants: III 5 CORIARIA INTERMEDIA MATSUMURA This is the only representative of the small family Coriariaceae known from the Philippines and occurs at high altitudes in Mountain Province, Luzon; it is called buacat by the Igorots of Benguet, who claim that the fruit is inedible because it is poisonous. A glucoside is found to be present in very small quantities in the leaves and fruits of this plant. One hundred kilograms of the fresh plant give 69 grams of crude glucoside, which can be extracted by water or squeezed from the young shoots. The extract is treated in the usual manner with lead acetate, and the filtrate is freed from excess lead by hydrogen sulphide and evaporated to a sirup. From this the glucoside is extracted by alcohol and precipitated by ether. Two cubic centimeters of plant extract, representing 20 grams of the fresh plant, intraperitoneally injected into a guinea pig weighing 700 grams, resulted in death after thirty minutes. Coriaria is known in New Zealand as “toot-plant.” Landsay (30) has the following to say regarding this plant: During a tour through the New Zealand provinces in 1861-1862, I was struck with the abundant evidences, which everywhere presented themselves, of the ravages produced among the flocks and herds of the settlers by the Toot-plant, one of the most common indigenous shrubs of these islands. In many cases of losses by individual settlers brought under his [their ?] notice, the amount of loss from this source alone had been from 25 to 75 per cent. In Otago, particularly, were such losses felt during the height of the gold mania there, from July to December, 1861: the traffic between Dunedin and Tuapeka gold-fields required the service of large numbers of bullocks, a great proportion of which were lost by Toot- poisoning. * * * 1. The Toot-poison belongs to the class of Narcotico-irritants. a. Its action on man includes the following symptoms:—coma, with or without delirium; sometimes great muscular excitement or convulsions; the details differing in different individuals; during convalescence, loss of memory, with or without vertigo. 6. In cattle and sheep, they include vertigo, stupor, delirium, and convulsions; curious staggerings and gyrations; frantic kicking, and racing or coursing; tremors. 2. The poisonous portion of the plant, (a) to man, is generally the Seed, which is contained in a beautiful, dark-purple, luscious berry, resembling the blackberry, which clusters closely in rich pendent racemes, and which is most tempting to children; occasionally the young Shoots of the plant, as it grows up in spring: (b) to cattle and sheep, in almost all cases, is the young Shoot, which is tender and succulent, resembling in appear- ance and taste the similar state of Asparagus. CONCLUSIONS Arcangelisia flava (Linn.) Merr. (A. lemniscata Bece.) con- tains berberine and shows approximately 4.8 per cent of pure alkaloid based on moisture-free wood. The plant material is 6 The Philippine Journal of Science "1919 soft and porous, and contains but small amounts of extractive matter that interferes in the recovery of the alkaloid. The re- covery of the alkaloid is simple and inexpensive, showing the plant to be an excellent source of the drug. Cassia siamea Lam. contains an alkaloid with the formula C,,H,,.NO,. The rhizomes of Geodorum nutans Ames contain about 14 per cent of a water-soluble adhesive. Coriaria intermedia Mats. contains @ poisonous glucoside in its leaves and fruit. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) REFERENCES Pommerehne, Arch, Pharm. (1895), No. 233, 127; Hare, Caspari, Rusby, National Standard Dispensatory. Lea & Febiger, Phila- delphia (1908), 307; Allen’s Commercial Organic Analysis, 4th ed. (1912), 6, 551. Bédeker, Ann. d. Chem. u. Pharm. (1841), No. 66, 384; (1849), No. 69, 40; Hare, Caspari, Rusby, op. cit., 361; Allen’s Commercial Organic Analysis, loc. cit. Mahla, F., Am. Journ. Sci. (1862), II, 33, 43; Allen’s Commercial Organic Analysis, loc. cit.; Hare, Caspari, Rusby, op. cit., 824; Perrins, J. D., Journ. Chem. Soc. London (1862), 15, 339. Hare, Caspari, Rusby, op. cit., 1004; Allen’s Commercial Organic Analysis, loc. cit.; Gordin, Arch. Pharm. (1902), No. 240, 146, contending absence of berberine. Schlotterbeck, J. O., Proc. Am. Pharm. Ass. (1901), 247. Mayer, F. F., Am. Journ. Phar. (1863), 35, 289; Perrins, J. D., loc. cit.; Gross, E. Z., Journ. Chem. Soc. London (1873); Allen’s Com- mercial Organic Analysis, loc. cit. Hare, Gaspari, Rusby, op. cit., 1720. Stavart, L. W., Phar. Journ. (1848), 7, 74; Waring, E. D. Phar- macopea of India (1868), 12; Dymock, W., The Vegetable Materia Medica of Western India. Frubner & Co., London 2d ed. (1885), 37. Mayer, F. F., op. cit., 97. Buchner and Herverger, Ann. d. Chem. u. Pharm. (1837), No. 24, 228; Rudel, Arch. Pharm. (1891); No. 229, 631; Allen’s Com- mercial Organic Analysis, loc. cit, Perrins, J. D., op. cit., 841. Chevallier and Pelleton, Journ. Chim. Med. (1826), 2, 314; Perrins, J. D., op. cit., 342. Stenhouse, J. W., Ann. d. Chem. u. Pharm. (1855), No. 95, 108; (1858), No. 105, 360; Pharm. Journ. Trans. (1855), 14, 455. Pommerehne, loc. cit. Mayer, F. F., Journ. de Pharm. et de Chim. (8) 46,.496. Some authors claim that Leontice thalictroides is identical with Caulo- phyllum thalictroides. Perkin, A. G., Chem. Soc. Trans. (1897), 71, 1198. Bédeker, Ann. d. Chem. u. Pharm. (1841), No. 66, 384; Chem. Gaz. VI,.No. 69, 40; Gordin, loc. cit., contradicting Bédeker. Schlotterbeck, J. O., Am. Journ. Pharm. (1902), 74, 584, hes Le | ef XIV, 1 (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) Wells: Philippine Medicinal Plants: III 7 Perrins, J. D., Ann. d. Chem. u. Pharm. (1852), 83, 276; Journ Chem. Soc. London (1862), 15, 389. Idem, Journ. Chem. Soc. London, (1862), 15, 340. Geacosa and Soave, Am. Journ. Pharm. (1890), 62, 500. Wood, Remington and Sadlter, U. S. Dispensatory. J. B. Lippincott & Co., Philadelphia, 19th ed., 1678. Steere, Am. Journ. Pharm. (1867), 39, 337. Stenhouse, Pharm. Journ. Trans. [8], 46, 495. Gastel, Jahresbericht iiber die Forstchritte der Chemie und ver- wandter Theile andere Wissenschaften (1866), 480. Martin, Pharm. Journ. Trans. [8], 13, 337. Gordin, loc. cit., contradicting Mayer. Brill and Wells, Philip. Journ. Sci., Sec. A (1917), 12, 167. Gaze, Arch. d. Pharm. (1890), No. 228, 607. Landsay, W. Lauder, Journ. Bot. (1863), 1, 247. ILLUSTRATION PLaTe I Fic. 1. Berberine-acetone compound. x 110. 2. Quick crystallization of the hydrochloride of the alkaloid. 9 {Puruir. Journ. Sct., XIV. No. 1. WELLS: MEDICINAL PLANTS, III.] Berberine-acetone compound. oil) phe be Fig. 2. Quick crystallization of the hydrochloride of the alkaloid. PLATE lt. THE TREATMENT OF HUMAN BERIBERI WITH AUTOLYZED YEAST EXTRACT By N. M. SALEEBY (Manila) The extract of rice polishings has become recognized as one of the most efficient remedies for the treatment of infantile beri- beri: However, it does not markedly improve cases of long standing. | Williams and Saleeby ? have shown that the hydrolyzed ex- tract does have a prompt and decisive action on these more chronic cases, but that it must be administered only under the direct supervision of the physician on account of its poisonous qualities when given in excessive doses. These investigators believe that the poisonous nature of the hydrolyzed extract is due to the choline present. That polyneuritis in fowls is improved by treatment with yeast has been known for some time. Cooper found that no poisonous effects followed the daily dose of 30 cubic centimeters of autolyzed yeast extract to chickens and that the administra- tion of 100 cubic centimeters did no harm. In order that this extract might be tried on human beriberi, in 1917 and 1918 the Bureau of Science prepared and delivered to me in 50 cubic centimeter bottles an autolyzed yeast extract and requested me to try its efficacy when given to human pa- tients. Brewers’ yeast was obtained from the San Miguel and the Oriental breweries, of Manila, separated from the adhering beer and placed in an incubator at a temperature of 35° C. until it had assumed a mushlike consistency (about forty-eight hours) due to autolysis. The mass was then filtered through paper, washed on the filter, and the filtrate concentrated under a partial vacuum at a temperature below 60° C. to a volume of about one-third of the original. * Albert, J., Philip. Journ. Sci., See. B (1915), 10, 81. The Bureau of Science, Manila, manufactures about 400 liters of this extract annually for use in the Philippines. ? Williams, R. R., and Saleeby, N. M., Philip. Journ. Sci., See. B. (1915), 10, 99. * Cooper, E. A., Bio-chem. Journ. (1914), 8, 250. | 11 12 The Philippine Journal of Science A concise summary of my findings when the autolyzed yeast extract was given to human patients is as follows: 1. About two score cases were treated; five were children below 2 years of age; ten were treated in hospital; all others were out-patients. 2. Adults were given from 15 to 40 cubic centimeters three times a day. Children were given from 2 to 4 cubic centimeters every three hours, or six times a day. Bigger doses did not seem to give better results. No sign of poisoning was observed. 3. Only acute and uncomplicated symptoms of beriberi were observed under treatment. Chronic nerve, muscular, or cardiac lesions were actually unaffected. 5. All acute peripheral symptoms of neuritis were affected quickly. Marked results were noted in less than three days, and a week’s treatment seemed to give full relief in mild acute cases. Treatment was generally followed up for two weeks at least. 6. Infantile-beriberi symptoms were relieved with comparative rapidity. £dema yielded quickly, and nutrition improved at once. 7. No special diet was prescribed. Patients were given re- gular hospital diets in accordance with the state of their digestion. 8. Children receiving the extract continued to nurse at the mother’s breast. 9. The effect of the autolyzed yeast extract used is similar to that produced by the hydrolyzed extract of rice polishings; it seemed weaker, however. yr A PHOTOGRAPHIC STUDY OF LEPROSY: . By Oswatp E. DENNEY (Chief, Culion Leper Colony, Culion, P. I.) FOUR PLATES The great variety of typical lesions of leprosy has rendered it impracticable completely to illustrate the disease in the aver- age textbook; it seemed desirable, therefore, to present a pho- tographic study showing the progression of the external lesions. The clinical manifestations of leprosy have been so fully described by the many students of the disease that discussion at length in the present paper is not necessary. The disease is generally considered to manifest itself in two distinct types— the “nodular” and the “maculo-anesthetic.” A combination of symptoms of these two is recognized as the “mixed” type. In this study the two principal types only will be considered; the mixed type, obviously, may present any composite picture of the nodular and maculo-anzsthetic types. Since the manner of entrance into the body of the causative agent of the disease (Mycobacteriwm lepre, or Bacillus leprz Hansen) is unknown, the initial lesion cannot be demonstrated with certainty; however, the earliest recognized lesion can be illustrated (Plate I, figs. 1 to 9; and Plate II, figs. 10 to 15). The progression of the lesions of the nodular type as indicated in the above series may be reproduced in practically any part of the body. The photographable manifestations of uncomplicated maculo- anesthetic leprosy are mainly confined to the macules, which appear chiefly on the back and face, but which may appear on any part of the body; the deformities that result from motor- nerve destruction; and the trophic ulcers that result from dis- turbed nutrition. The deformities of uncomplicated maculo- anesthetic leprosy in the hands are illustrated by Plate II, figs. 16 and 17; and Plate III, figs. 18 and 19. Plate III, fig. 20, illustrates the typical perforating ulcer of the foot; this lesion is considered as diagnostic of leprosy by the natives in many parts of the world. * Received for publication, September 6, 1918. 13 14 The Philippine Journal of Science Plate III, fig. 21, illustrates the typical, sluggish, neurotrophic ulcers, which although almost incurable appear to have a strong resistance to the ordinary pyogenic infections. Here, too, is shown the curious parchmentlike skin frequently seen among anesthetic lepers. The macules of the maculo-anesthetic type are reproduced with difficulty, particularly in the dark-skinned races, because of the lack of contrast (Plate III, figs. 22 to 25). Another lesion of leprosy that, due to its great prevalence among the Filipino lepers, deserves special description is the condition of bone necrosis and absorption commonly seen in the feet and hands. The condition is generally described as of the maculo-anesthetic type and is probably not entitled to classifi- cation as a special type. The condition may be seen rarely among the pure nodular types, commonly among the maculo- anesthetic and mixed types, and occasionally among lepers who fail to show coincident evidence of progressive nodular or maculo- anesthetic symptoms. The complete series of changes exterids over a number of years and therefore may be followed with great difficulty in individual cases. The sequence appears to be much as shown in Plate IV, figs. 26 to 38. The process of absorption as indicated above usually does not continue without complications from trauma and secondary in- fections; the bones being near the surface and unprotected by the normal connective tissue pads of the finger tips, sinuses frequently form, from which necrotic bone is discharged. An instance is recalled of a patient, since dead, in whom the five finger nails of one hand remained attached in a much distorted condition at the site of the radiocarpal articulation, the bones of the entire hand having been absorbed without rupture of the skin from trauma. It is not to be inferred from this series of photographs that each case of leprosy progresses by rule to a definite classical physical picture; there is perhaps no disease so capricious in the manifestations of its clinical progress. ILLUSTRATIONS PLATE I Fig. 1. A maculo-papular lesion of three years’ duration on the cheek of a 6-year-old Filipina, which up to the present time is her only demonstrable lesion of leprosy. The spot, about 2 centi- meters in diameter, is slightly elevated above the normal skin surface; it is pink and contains numerous small nodules in which the causative organism can be found. A lesion of this character may remain practically unchanged for years; may almost entirely disappear, leaving a pale or pigmented, smooth spot; or it may grow progressively larger with the coincident appearance elsewhere of similar spots, as in fig. 2. 2. Maculo-papular lesions of leprosy on each cheek and on the chin. 3. An apparent latency of the macules of leprosy; which, in this case, are neither elevated nor nodular, but are smooth and faintly pigmented and as such cannot be distinguished from the _ numerous tinea, etc., that are frequently met with in the | tropics. In this and in the succeeding illustrations may be seen the progressive infiltration of the tissues over the inner half of the superciliary ridge, giving rise to overhanging eyebrows, suggesting a frown. Mention is frequently made of the loss of the hair of the outer half of the eyebrows as an early and diagnostic symptom of leprosy; this is not present in the cases shown in figs. 1 to 3. 4. A case of leprosy, relatively older than those of figs. 1 to 3, showing a thinning of the outer half of the eyebrows as well as the early nodular infiltration of the nasal ale and the helix and lobule of each ear. 5. A somewhat more advanced case of leprosy with a nodular infiltra- tion extending over the face. The nodules are small with the exception of one on the chin. Complete loss of the hair of the outer half of the eyebrows is here shown. A small amount of edema of the eyelids is present. ; 6. The nodules are larger and more prominent than in the preceding eases. This and figs. 7 to 9 illustrate the loss of normal contour of the nose from the destruction of cartilaginous tissue. . 7. In addition to the diffuse infiltration of the face, there are well-marked infiltration of the lips and slightly pendulous ear lobules. 8. The nodules, as discrete tumors, are surrounded by apparently normal tissue. A considerable loss of the cartilage of the tip of the nose has caused a marked flattening of this appendage. 9. Illustrating shotlike tubercles of leprosy as discrete nodules, scattered over the entire face; the arms and the legs, and to a lesser extent the trunk, showed similar tumors. 1628232 15 16 Fie. Fig. The Philippine J ournal of Science 1919 PLATE II 10. Showing a not uncommon condition in leprosy, in which the nodules themselves undergo necrosis with subsequent absorp- tion, leaving smooth cicatrices. Necrosis of the cartilage of the left ear and of one nodule on the left cheek is here in progress. 11. Showing the partial coalescence of large nodules on the cheeks 12 and chin; this condition in advanced cases of leprosy gives rise to the leonine facies. . A profile view of a case of early leontiasis, showing a diffuse thickening of the skin of the face. This is an excellent example of pendulous ear lobules. 18. A slightly more advanced case of leontiasis, showing the loss of the normal lines of expression of the face, notably in the distorted lips. 14, A still farther advanced case of leontiasis; the general thickening of the tissues of the face has given rise to a “death-masklike”’ appearance. Leprous cachexia is well marked in this woman. 15. A fairly typical case of leontiasis. This may be considered as 16. the terminal manifestation of uncomplicated nodular leprosy in the face. A typical “ring and little finger deformity,” the earliest recogniz- able deformity of this tyne of leprosy in the hands. Both fingers can be straightened with only moderate force and at this stage retain some function. 17. A moderate contraction in all the fingers in leprosy; both thumbs, 18 in this case, appeared to be normal. At the instant the picture was being taken, a moderate pressure was being exerted on the fingers to show the approximate amount of available extension. PLATE III . A more nearly complete contraction of the fingers than in fig. 17; here, too, the limit of extension is shown. The cedema present in the fingers of the left hand is transient and is due to an infection in the palm of the hand. 19. An example of the claw hand of leprosy, the classical main en 20. griffe; the typical atrophy of the interossei is indicated in the left hand. The condition of this hand may be considered as the terminal manifestation of anzsthetic leprosy of the hands in uncomplicated cases. Unfortunately, the disease rarely stops with this deformity. Due to the coincident anesthesia, the hands are subject to trauma, particularly to burns, which may be very extensive and are usually painless. Other trauma and neurotrophic changes result in various deformities. . The typical perforating ulcer of the foot; this lesion is considered by the natives in many parts of the world as diagnostic of leprosy. 21. The typical, sluggish, neurotrophic ulcers of leprosy, which although almost incurable appear to have a strong resistance to the ordinary pyogenic infections. The curious parchmentlike skin frequently seen among anesthetic lepers is shown. AIV FIG. FIG. 22 23. 24, "25. 26. 27. 28. 29. 30. 51. 82. 33. Denney: Photographic Study of Leprosy 17 A single, large patch on the back of a leper having a slightly pigmented, rugged, central area and a slightly erythematous periphery. An almost diffuse outcropping of small erythematous macules of leprosy, not unlike those of secondary syphilis. The coalescence of numerous macules of leprosy, giving rise to curious circinate designs. A somewhat unusual herpeslike series of macules of leprosy. PLATE IV An undeformed hand, aside from the slight spindle shape of the fingers; the proximal portions being about normal in appear- ance, the distal phalanges being smaller and pointed. An obviously deformed hand, the spindle shape being greatly exaggerated. Unquestionably in this case of leprosy the osseous tissues are not alone in the retrogressive process. The process of absorption is in progress, and there is some distortion of the digits from coincident nerve involvement. The thumb of each hand in this case is unattacked. A more advanced stage; in this case the nails may be seen in their deformed state almost at the point of articulation of the first and second phalanges. The proximal phalanges are almost completely absorbed, the thumb of the right hand alone having escaped the process. A similar irregularity in the absorption of the fingers, the middle finger of the left hand being unabsorbed. The peculiar “telescoped” appearance occasionally seen, in which the process of absorption of the osseous tissue has proceeded more rapidly than that of the other tissues. The partial absorption of the metacarpals of the left hand; the distorted finger nails remain as horny projections on the palmar aspect of the hand. The right hand shows a long scar at the site of the metacarpo-phalangeal articulation, the result of amputation of infected fingers. The left hand shows no scars since the skin has remained unbroken during the entire process. DENNEY: STUDY Or LEPROSY.] {Puiuip, Journ. Scr., XIV, No. 1. Fig. 1. Fig. 2. Fig: 3. Fig. 4. Fig. 5. Fig. 6. Fig. 7. Fig. 8. Fig. 9. PLATE lI. DENNEY: STUDY or LEPROSY.] {Puitte. Journ. Scr., XIV, No. 1. Fig. 13. Fig. 14. Fig. 15. Fig. 16. Fig. 17. PLATE Il. DENNEY: STUDY OF LEPROSY, ] {Puinte, Journ. Sct., XIV, No. 1. Fig. 21. Fig. 22. Fig. 23, Fig. 24. Fig. 25, PLATE ill. DENNEY: STUDY oF LEPROSY.] [PuHitte, Journ. Sct., XIV, No, 1. Fig. 26. Fig. 27. Fig. 28. Fig. 29. Fig. 30. Figs 3A. Fig. 32. Fig. 33. PLATE IV. A COMPARATIVE STUDY OF DIFFERENT METHODS OF TREATMENT OF TYPHOID FEVER* By Pepro T. LANTIN (From the Department of Medicine, College of Medicine and Surgery, University of the Philippines, and the Clinics of the Philippine General Hospital, Manila) TEN TEXT FIGURES CONTENTS INTRODUCTION. VARIETY OF TREATMENT—Continued. PRESENTATION OF THE CLINICAL Foreign protein—Continued. CASES. Nonsensitized vaccine. Period of observation. Intravenous. Source of material. Peptone. Type of epidemic. Intramuscular. Morbidity and mortality. Milk, Selection of material. Intramuscular. Clinical grouping of cases. Colloidal preparation. VARIETY OF TREATMENT. Colloidal gold (colibiase). Foreign protein. Intravenous. Sensitized vaccine. DISCUSSION. Intravenous. Specificity of the treatment. Controls. Nonspecificity of the treatment. Intramuscular. SUMMARY AND CONCLUSIONS. INTRODUCTION Life is short and art long; opportunity is fleeting; experiment is dangerous and decision is difficult.—First Aphorism of Hippocrates. The question of the therapy of typhoid fever has been given considerable attention by scientists for the last three years. It now forms a new field of scientific research, among both clin- icians and laboratory workers. Medical literature teems with reports of the favorable results obtained by different observers of the merits of the various methods employed; namely, the use of (a) foreign proteins and (b) colloidal preparations. These methods of treatment may be said to be still in the experimental stage. After three years of careful trial and observation, the views expressed by leading investigators with particular refer- ‘Thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Tropical Medicine, in the Graduate School of Tropical Medicine and Public Health, University of the Philippines, February 21, 1918. 19 20 The Philippine Journal of Science 1919 ence to the action of vaccines are many, and opinion is far from unanimous. The claims made by some, that the curative effect of vaccines is due to their specific action, have been held to be untenable, so that many look upon the question as still unsettled. It is my intention to make the present paper the basis of my contribution to this much-disputed question, although conscious of the fact that it requires some temerity to enter a field already occupied by so many distinguished American and European writers. I hope, however, to set forth some facts in connection with these methods of treatment and then to present the con- clusions reached through my observations. PRESENTATION OF THE CLINICAL CASES Period of observation.—The present investigation covers a period of seven months, beginning August, 1917, and extending to February, 1918. During that time there were admitted to the medical department in the Philippine General Hospital nine- ty-eight cases of typhoid fever, all Filipinos. Both sexes are represented in the series. The minimum age was 14 years. Another series of twenty cases of typhoid fever has not, been included with these ninety-eight cases for the reason that some were in a dying condition on admission and, therefore, short-lived in the hospital; and others were admitted in a con- valescent state, in which they showed marked defervescence. In connection with these convalescent cases it was felt that the effects of the treatment under consideration in this paper might lead to erroneously favorable interpretations. Source of material—tThe source of the cases studied was the district of Manila. Relatively few cases came from Rizal, Ca- vite, Bulacan, and Laguna Provinces. Type of epidemic.—Many of the patients on admission showed marked toxemia, frequently associated with pneumonic, cardiac, or renal complications; while other cases were admitted in a condition of collapse, either with intestinal perforation or acute intestinal hemorrhage. It can be fairly stated, then, that we were dealing with a severe type of epidemic. It should be fur- ther stated that, with very few exceptions, these typhoid cases belonged to the laboring class of the community, whose hygiene and dietary were very deficient. Upon falling ill many of them remained in their homes for a number of days without any med- ical attendance, and only applied to the hospital for admission when their condition had become very serious. In these ninety-eight cases males predominated over females. The sex incidence is shown in Table I. XIV, 1 Lantin: Treatment of Typhoid Fever VAI TABLE I.—Sex incidence in typhoid fever. Sex. Cases. Deaths. Per cent. MiBIO 6S aah ee Sa ee 65 14 21.5 GING Soria eae ey bee aie ee 33 5 15.1 Weel is 2 ee 98 19 | 19.38 | : ee cued Morbidity and mortality—Out of these ninety-eight cases there were nineteen deaths, a mortality of 19.38 per cent, a very discouraging if not alarming figure. It should be remembered, however, that many of the cases showed little prospect’ of re- covery, on account of their serious condition when admitted to the hospital. In fact, ten of the nineteen cases that died were considered hopeless on admission. Excluding these ten hopeless cases, therefore, reduces the mortality rate to 10.22 per cent. McCrae(31) gives 25 per cent as the mortality rate for British troops in South Africa; Curschmann,(5) 9 to 12 per cent; Os- ler,(837) 7 to 20 per cent in hospital practice, and 5 to 12 per cent in private practice; Rogers,(42) 16.3 per cent for white troops in the tropics and 26 per cent for Indians. In the Phil- ippines Chamberlain(3) places the mortality rate at 17.65 per cent for Filipgios and 16.8 per cent for white troops in the tropics. Gutierrez,(12) of the Philippine General Hospital, in analyzing one hundred twenty-five cases of typhoid fever in 1913, gives 20.43 per cent; after excluding ten hopeless cases he gives a mortality rate of 13.13 per cent. Selection of material.—Selection of cases is fraught with no less difficulty. The varying degree of severity of infection in each case, and the fact that some individuals came early for treatment while others came later, rendered comparison a dif- ficult task. However, particular emphasis was laid on the se- verity of the individual case on admission. The cases were then grouped. It should be stated that the diagnoses in all these ninety-eight cases were established on clinical findings, and lab- oratory examination was then resorted to for confirmatory pur- poses. In each instance search was made for malarial parasites. Clinical grouping of cases.—These ninety-eight cases were divided into groups according to the apparent condition of the individual case on admission or after one or two days’ obser- vation in the ward. ‘Thus, cases that presented high fever and delirium, with or without complications, were classified as severe and toxic; those that presented high fever, with or without complications, but not apparently poisoned, were classified as 22 | The Philippine Journal of Science 1919 severe and nontoxic; while those with moderately high fever, without complications, and not apparently poisoned, were classi- fied as mild. These classifications are, of course, more or less arbitrary and conventional. Classifications were made under these headings as follows: TaBLE I1.—Classification of cases of typhoid fever. Cases. Severe and toxic 26 Severe and nontoxic 47 Mild 25 Total 98 VARIETY OF TREATMENT FOREIGN PROTEIN Bacteria, being composed of protoplasmic matter, are con- sidered to be protein substance. In fact, Robertson,(40) and Jobling, Petersen, and Eggstein,(17) have called this substance “bacterial protein.” My view is that any organism, living or Haida becomes foreign protein when introduced into the ost. Typhoid fever, so far as I am aware, is the only disease in the treatment of which protein has been extensively employed. It has been used either in homologous or heterogonous forms. In the great majority of instances, however, vaccines in the form of foreign protein have been used. These consisted of either living sensitized organisms, dead sensitized organisms, or non- living nonsensitized organisms. Sensitized organisms are bacilli that have been treated with immune serum, while non- sensitized nonliving organisms are killed bacilli that have not been treated with immune serum. SENSITIZED, OR TREATED, VACCINE ADMINISTERED INTRAVENOUSLY Vaccine treatment dates from 1893. Eugene Fraenkel(7) was the pioneer in this field. He employed a vaccine of killed typhoid bacilli, subcutaneously administered, for therapeutic purposes in typhoid fever, and he claimed to have modified the course of the disease. His work received scant attention from early investigators. Nine years passed before reports of similar observations came from Petruschy(38) in 1902. Then followed the work of Pescarole and Quadrone(39) in 1908. Ichikawa, (15) in 1914, used the intravenous method of administration of living sensitized typhoid vaccine and secured excellent results. His success awakened interest in this form of treatment: and since then, according to Gay and Chickering, (11) it has received care- ful trial at the hands of such leading investigators as Koranyi, XIV, 1 Lantin: Treatment of Typhoid Fever 23 Biedl, Eggerth, Sladek and Kotlowski, Boral, Holler, Lowy, Luksch, and Wilhelm and Myer. Variety of sensitized typhoid vaccine.—So far as I am aware, there are four kinds of this type of vaccine; namely, (a) Ichikawa’s living attenuated sensitized vaccine, (0) Besred- ka’s(1) living sensitized vaccine, (c) Gay’s(9) sediment vaccine, and (d) sensitized killed vaccine. Dosage.—The dose administered intravenously varies accord- ing to the kind of vaccine. Thus, of Besredka’s vaccine there have been used from 100 to 300 millions; Ichikawa’s vaccine, from 200 to 300 millions; Gay’s vaccine, from 0.2 to 0.4 milli- gram (150 to 300 millions). The dosage of killed sensitized vaccine is from 50 to 800 millions as indicated by Jobling ; (17) 70 to 300 millions, by Dithorn and Schultz ;(6) and equal parts of typhoid and paratyphoid bacilli (500 millions) is the amount employed by Robertson.(40) I have employed a dosage of from 250 to 1,000 millions of killed sensitized typhoid bacteria in- travenously, increasing the doses every three or four days ac- cording to the case. Contraindications.—According to eminent writers on the sub- ject, weakness of the heart, presence of pneumonia, and intesti- nal hemorrhage are contraindications for vaccine treatment. Thus Sladek and Kotlowski(46) believe that it increases peris- talsis of the intestines, with subsequent danger of hemorrhage and perforation. My experience leads me to believe that vaccine therapy is as yet of very limited applicability, and I do not feel that I am in a position to confirm or controvert their observa- tions. However, I have employed this treatment in all condi- tions, except in cases of intestinal hemorrhage, and so far there have been no untoward effects that could be attributed directly to the treatment. : This series comprises fifty-one cases of typhoid admitted during the height of the epidemic, of which thirty cases were subjected to intravenous injection of polyvalent sensitized typhoid vaccine, while twenty-one cases were used as controls. The clinical grouping of cases given vaccine treatment intra- venously is shown in Table III. TABLE I11.—Condition of patients before treatment. Cases. Severe and toxic 9 Severe and nontoxic 7 Mild 30 Total 24 The Philippine Journal of Science 1919 Method of treatment.—The view that sensitized typhoid vac- cine is considered less toxic than nonsensitized typhoid vaccine seems to be the consensus among investigators at present. Cecil,(2) Garbat,(8) and Nichols(35) have shown by experiments with animals that such is the case. Cecil, however, has found that the lethal dose of sensitized typhoid vaccine is three times as large as that of the nonsensitized bacilli. Reasoning from this point of view, since McWilliams(34) has used intravenously 500 millions of plain nonsensitized killed typhoid bacilli as the initial dose, the safe limit of dosage for sensitized organisms would be 1,500 millions; I have used, therefore, 250 to 500 millions as my dose and, in a very few cases, 1,000 millions. I have never exceeded 1,000 millions, even in repeated doses. The vein selected for intravenous injection is the median basilic, this vessel being commonly prominent, although the injection may be given in any other vein provided the site is rendered aseptic. The injection should be done under rigid . asepsis and preferably by the closed method. The results of the treatment of thirty cases by intravenous injection of sensitized vaccine are shown in Table IV. TABLE 1V.—Sensitized typhoid vaccine administered intravenously. Cases. Days | Dura- | Dura- Classification by results. Sivene pee er Somot | oe. Mild. \Severe.| and ment. | ment. | ease. toxic. Days. | Days. Aborted ._._ x 4 8 0 * 8.2 5.1} 18.8 FROIN ao oak 2 12 0 14 11.5 14.4 | 25.9 Relatively unaffected _.......__..__.__ 0 0 8 8 8 25.8 | $3.8 Deaths o ae 0 0 6 i ETS ed ae een premiers Crisitz a ao EK OPO EE OER Peas , SS Rane Sancta Rae Ae LRG a es chm ee onuinnkwtwccunee:. BO feecencchcwcck ticsewsey sub luclgaldamtierenpspetgs Sacer 6 eset Sn ee 24.1 In Table IV the cases have been separated into groups according to the results; namely, (a) aborted type, (b) benefited type, and (c) relatively unaffected type. There were seven aborted cases out of thirty cases treated intravenously, with an average duration of treatment of 5.1 days, in contrast to seven days observed by Gay.(9) McCrae(32) in analyzing fifteen hundred cases found 0.1 per cent that re- covered by crisis. In my series, I found 28.33 per cent of aborted cases. It is interesting to note that the aborted cases were observed during the first week, and that they have a direct XIV, 1 Lantin: Treatment of Typhoid Fever 25 relation to intravenous injections. This abortive type was ob- served in mild cases in the majority of instances and in severe cases in a few instances. The temperature record of one of the abortive cases is shown in the accompanying chart (fig. 1). iS 6th day lth. day. th day. 9th day. | se PM. A.M. | P 4} ALM. P M. A. M™. RM.\| AN P| 10\2 6 10} 2 6 10\2 6 10\246 8 lo/A2 46 B/0I2\2 6126 lle sl} 1 | sity S {1 LB ft = SRN & NS NN SS 8S +o sen bs aot || sett Tl tore) Th 7 xs se) Bd 377 \ \ LY TOo 36 \ Fic. 1. Temperature chart of B. Y. Abortive type of typhoid fever, treated with sensitized typhoid vaccine intravenously. The benefited type was observed during the second week. As a rule the temperature drops, either by crisis or by lysis, fol- lowing each successive intravenous injection; but afterward it rises again temporarily, then becoming permanently normal. The average duration of treatment was 14.4 days in my cases, while in those of Gay it was 15.8 days. Examples of the ben- efited cases are shown in the accompanying charts (figs. 2, 3, and 4). The relatively unaffected type was observed in advanced cases. Each injection was followed by a temporary drop in temperature, which subsequently in some cases resumed the original level at which it was before the injection, while in others it became slightly remittent or intermittent. The course of the disease was slightly altered. The average duration of the disease was 25.3 days in this series. Symptomatology.—From thirty minutes to one hour after the intravenous injection of vaccine, the patient as a rule feels a slight chilly sensation as a prodromal symptom. This lasts for a few minutes; then follows the real shaking of the entire body. This rigor shows varying degrees of intensity according to in- dividual susceptibility, and may last for from fifteen to thirty- 96 The Philippine Journal of Science 1919 five minutes. During this period the pulse and the respiration are slightly increased in rate, and the skin is pale and feels cold; the temperature is usually high, reaching its maximum height in from three to six hours. It then falls, generally either by crisis or by lysis, or else it may become intermittent or re- mittent. There are cases, however, in which the temperature drops by crisis or lysis without chills or a preceding rise in temperature. Such cases are relatively rare. The crisis may BS 13th day. 14th day. (5th day, | 6th day S| Am. PM. A.M. |PM\|AMILPMIAM S| 2 6/012|\2 468 10/2\2 4 6 B/0\2 6 Wl? 6/126 102 6 10 31. aS ES REN SPF SS Sy] y Siz SIS QSi8 SES § SSS SMM ESES 40" SSP sip 39° = 2. NS 38; ws 6 6 \ Nei role la 37{- 7 oad 5 Fig. 2. Temperature chart of E. .N. Benefited type of typhoid fever, treated with sensitized typhoid vaccine intravenously. /4.th day. (5th day. (6th day. ‘7th day. AM PM.| AM IPM. A. MM, PM, AM P i, 15/02 6 102 6 142 6/0122 46 81122 46 BIO? 6102 6106 ar Ss ty bihahe = TSptaig : Shige cps S SSS = Ou. bs 3g le 36 (@; , ont nN. ba oe 37 on Fic. 3. Temperature chart of M. L. Benefited type of typhoi sensitized : d fever, trea typhoid vaccine intravenously, on XIV, 1 Lantin: Treatment of Typhoid Fever 27 If th day. } 12 th day. 13 th day. 14th day 15 th doy. 16th day. AMT eM |AM.| OP OM. AM: P.M. AMP M.|AM| PM, AM.[P M. iS 26 Wi26 1026 026 Wi2)\24+6 8 1212 46 810l2\2 6 12 6 0\2 6 10i'2 6/126 WiI26 » Fe Sit = S = hi S Sane 31] 8 = S ese |i] |B PSNI 5 § #1 re St et te } se secs Bis o sed aA cn SE ae ‘sr ‘an Ri v Sor tT 37) ; Oj N. o O Noo Fic. 4. Temperature chart of N. C. Benefited type of typhoid fever treated with sensitized typhoid vaccine intravenously. reach a normal or a subnormal level about twelve hours after injection. The rigor is followed by thirst, cold perspiration, and relaxation, but accompanied by a feeling of marked alleviation of subjective symptoms. In the majority of instances the pa- tient, previously sleepless, is able to sleep soundly the following night. | The temperature may remain permanently at the normal level, and in this event no further injection is necessary; but, if the temperature should rise again for a period of three or four days, another injection is needed with increasing dose. My ex- perience is that two or three doses are sufficient to produce the desired result; if without result, further injections seem to be of no value. According to Gay and Chickering,(11) and MeWilliams, (34) there is a leucopenia during the rise of temperature, and hyper- leucocytosis during the apyretic period. They were able to ob- serve these phenomena by frequently examining the blood of the patients. I was unable to make frequent examinations of the blood of the patients injected, but examination of the blood twenty-four hours after injection revealed a slight increase in the leucocyte count. I have observed that a moderate reaction is very necessary to produce beneficial results. Similar observations have been confirmed by Gay(9) and claimed by Leake. (24) To avoid errors in interpretation of results, it is highly advisable to base one’s judgment on the effects of treatment; that is to say, not from the subjective symptoms, but rather from the objective phenomena, such as shortening of the duration of the fever or, 28 The Philippine Journal of Science 1919 in other words, the occurrence of the abortive type in relation to injection. Complications.—Excluding eight cases, in which complications were already present on admission into the hospital, six de- veloped complications during treatment. In ninety-eight cases treated by Gay,(9) he observed thirteen that developed com- plications; namely, pyelitis, one case; lobar pneumonia, one; laryngitis, two; bronchopneumonia, two; toxemia, one; hzemor- rhage, four; and perforation, two. Table V shows the complications that were observed during treatment and before treatment was begun. TABLE V.—Complications before and during treatment. Devel- | Devel- oped oped Complication. a before | during treat- treat- ment ment PUMICE ans oro esa oo ae oe at Oo oe ee 8 3 Ri Ceebisinh HOiOPTNee es so a eee oe © 0 b2 Br hop WM ah ea icg ke SSE EN odes wh as oa dbuwa vated saul obab ink 1 1 RUQUAE CROONER bs ge ee ae ee 9 6 i eae * Relapses, three cases. » One case had hemorrhage four days after injection ; another case, during convalescence. Early investigators, Krumbhaar and Richardson, (22) Waitz- felder, (48) and Robertson,(40) held the belief that vaccine therapy diminishes the complications. But Table V seems to show that it has no advantageous effect on the complications as compared with my controls. Whittington,(49) who has had considerable experience with vaccine therapy, reports that he observed a higher percentage of complications (49.5 per cent) under vaccine treatment, while his controls showed 46 per cent. His observation is based upon one hundred fifteen cases treated with vaccine and another one hundred fifteen under classical treatment. Relapses.—In my thirty cases, treated with vaccine, three had relapses, lasting only a few days. Gay advised subcuta- neous injection of vaccine a few days after the intravenous, so as to prevent relapse. Again, Whittington found a higher per- centage of relapses in cases treated with vaccine (10.4 per cent), in contrast to 7.8 per cent observed in his controls. In my controls I had five relapses (23.9 per cent), a higher number than shown by the cases that had undergone vaccine treatment (10 per cent). This difference in results may be due to the limited series that I have had so far. XIV,1 Lantin: Treatment of Typhoid Fever 29 Concomitant disease.—On examining these thirty cases, two cases of tuberculosis were found. One was a case of incipient pulmonary tuberculosis that later became active during the course of the disease. It seemed that vaccine treatment ad- ministered intravenously had done harm to the patient; because, after the injection and consequent shaking up of the body, it was noticed that traces of blood were found in his sputum. This might be a mere coincidence, but it should be regarded with suspicion. It seems advisable to exercise prudence and caution in the use of vaccine for patients suffering from pulmonary tuberculosis. The other case was a tuberculous cervical aden- itis. In this particular case vaccine injection was not followed by any untoward effects. Mortality.—The mortality in these thirty cases is 20 per cent, a very high rate. It should be remembered, as already stated, that these patients belonged to the laboring class who did not realize the advantage of early hospitalization. Consequently they had remained in their respective homes without adequate treatment and care, only coming to the hospital at a time when medical attention would be of little avail because of the advanced stage and seriousness of the cases. In fact, there were three cases in which I considered the outlook on admission as hopeless on the grounds of marked toxemia and pulmonary complica- tions. By excluding these three hopeless cases, the mortality is reduced to 10 per cent. Autopsy findings will reveal the causes of death, as shown in the following protocols: Autopsy findings (sensitized intravenous treatment). File No. 1957: D. H. Anatomic diagnosis: Bilateral lobar pneumonia with abscess and gangrene of the lung; typhoid enteritis; acute parenchymatous degeneration of the viscera; chronic fibrous pleurisy. File No. 1880: V. C. Autopsy findings: Typhoid ulcerated enteritis, healing; acute hzmor- rhagic bronchopneumonia; acute pyonephritis; parenchymatous de- generation of the liver; fatty degeneration of the heart; acute myocarditis. File No. 1875: S. C. Anatomic diagnosis: Intestinal hemorrhage; acute ulcerative entero- colitis, typhoid; acute splenitis, mesenteric; parenchymatous degen- eration of the heart and liver; acute parenchymatous nephritis with some interstitial nephritis; focal necrosis of the liver; emacia- tion; ascariasis; trichuriasis. File No. 1985: F. J. ; Anatomic diagnosis: Beginning lobar pneumonia, right; typhoid enteritis; acute parenchymatous nephritis; acute parenchymatous degeneration of the heart and liver; calcareous nodule in the left lung; ascariasis. 30 The Philippine Journal of Science 1919 The above necropsy records show that no known treatment would have prevented the fatal outcome. Undoubtedly, these patients died of what are commonly known as “accidents of typhoid.” Gay and Chickering,(11) in 1915, had 9 per cent mortality in fifty-three cases, treated with their vaccine; Gay,(9) in 1917, had 6.6 per cent mortality in ninety-eight cases. In my series the mortality is 10 per cent, after excluding the three hopeless cases. Ichikawa,(15) however, in 1914, had 12 per cent mor- tality in eighty-two cases; but his controls-showed 30 per cent mortality, while my controls showed 23.8 mortality. McWil- liams (34) has this to say about the rate of mortality: The percentage mortality in itself offers but little evidence as to the danger or efficacy of the treatment, as it will necessarily vary according to whether only mild cases, only severe cases, or all types of cases are subjected to the injections. The highest death rate, 23 per cent., was observed by Paulicek. His patients were soldiers who had suffered from severe exposure in cold weather, and they died in most instances from a complicating pneumonia. The high mortality of 23 per cent. represents, he says, a reduction by 20 per cent. from what was obtained without the injection of vaccine. It has already been mentioned that the death rate in Ichikawa’s control cases was more than twice as high as in the treated cases. Of far greater importance than the percentage mortality is the question as to how many, if any, of the deaths could be attributed directly to the ion of the vaccine. Five cases (treated by Boral, by Csernel and arton, by Lowy, Luksch and Wilhelm, and by Paulicek) showed intestinal hemorrhage following the injection and terminated fatally. One case (of Sladik and Kotlowski) resulted fatally from hemorrhage into the thyroid gland. One of Eggerth’s patients developed a hemorrhage from the lungs and died three hours after the injection. One of Biedl’s had hemorrhage from the nostrils before the intravenous injection of vaccine was given; after the treatment the hemorrhage recurred and could not be stopped. ~ CONTROLS The fact is well known to all who have had experience with the disease that, strictly speaking, no two cases of typhoid fever are of equal severity. But this does not mean that we should disregard entirely the value of controls; for, although the results are not necessarily equal, yet, taking them as a whole, the factors concerned are minimized and the results thus obtained are approximate, if not exact. By the term control, as used in this paper, I mean the classi- fication of typhoid patients into groups of cases showing a more or less equal degree of infection, as a basis of comparison with other groups of patients, similarly affected by the same disease but undergoing different kinds of therapy. That is to say, the XIV, 1 Lantin: Treatment of Typhoid Fever 31 controls are grouped conjointly with the cases that are subjected to the intravenous treatment with sensitized vaccine. These control cases were subjected to classical treatment, which consists in the application of general hygiene and care; the administration of light, nourishing food; the application of hydrotherapy, either cold or tepid; and absolute rest. As an accompaniment to this treatment the patient may be given cardiac stimulation, soda enemata (1.5 per cent solution) every other day, and occasional, small doses of urinary antiseptics. The patient may be given 1,000 cubic centimeters of proctocly- sis daily of normal saline solution and should be encouraged to drink plenty of water as an aid to disintoxication. This series consists of twenty-one cases. Their clinical grouping is shown in Table VI. TABLE V1.—Condition of patients before treatment. Cases. Severe and toxic 6 Severe and nontoxic 11 Mild 4 Total 21 In the sixteen cases remaining after five deaths are excluded the average duration of the disease was 27.7 days, while in the cases treated with sensitized vaccine, as previously stated, it was 24.1 days. Gay,(10) in using similar treatment, found that the average duration of the disease was 27.6 days. McCrae, (33) in analyzing fifteen hundred cases of typhoid, found that the average duration was thirty-one days. Complications—Out of these twenty-one cases there were eight cases that developed complications, two of which showed them on admission. Table VII shows the variety of the com- plications: TABLE VII.—Complications in eight cases of typhoid fever. Intestinal hemorrhage Cases. Lobar pneumonia (present on admission) 2 Bronchopneumonia 3 Perforation : 8 Total In connection with these complications, it is interesting to compare the complications that developed only during vaccine treatment with those that appeared in the controls as shown in Table VIII. 1623233 32 The Philippine Journal of Science 1919 TABLE VIII.—Complications in control cases. Cases. Controls (after excluding 2 cases in which complications were present on admission) 6 Sensitized vaccine treatment 6 The above comparison differs somewhat from the results ob- tained by other investigators, such as Gay(9) and Waitz- felder,(48) who have claimed that vaccine therapy diminishes the incidence of complications. My results, however, are some- what similar to those of Whittington,(49) who found that well- controlled cases develop a smaller number of cases of compli- cations. The discrepancy between my observations and those of Whittington may be due to the fact that he observed a greater number of controls than I had available for treatment. Mortality.—The rate of mortality is very discouraging, if not alarming. There were five deaths out of twenty-one cases, a mortality of 23.8 per cent. If we exclude the two cases that I considered hopeless on admission, the rate of mortality will be reduced to 14.3 per cent. Only two cases were autopsied out of five deaths, and the anatomic diagnoses show the following: Anatomic diagnoses (autopsy findings of control cases). File No. 1749: I. §. ) Anatomic diagnosis: Acute ulcerative enteritis (typhoid) ; suppurative peritonitis following perforation; acute parenchymatous degeneration of the viscera; cardiac dilatation; oedema of the meninges. File No. 1834: B. C. Anatomic diagnosis: Acute enteritis (typhoid), congestion, and begin- ning necrosis; acute splenitis; acute lymphadenitis, mesenteric; acute dilatation of the heart; acute parenchymatous nephritis; parenchymatous degeneration of the heart and liver ; focal necrosis of the liver; chronic caseous lymphadenitis, bronchial; chronic fibrous pulmonary tuberculosis; emphysema, interstitial; ascariasis. Again, I have to state that the untreated control cases of Ichikawa(15) showed a mortality rate of 30 per cent in com- parison with my mortality of 14.3 per cent after two hopeless cases were excluded. SENSITIZED VACCINE ADMINISTERED INTRAMUSCULARLY Early in 1911 Metchnikoff and Besredka (27) advocated the use of living sensitized typhoid vaccine intramuscularly for pro- phylactic purposes and claimed to have secured excellent results thereby. Followers of this school have adopted that practice and have even gone so far as to use this form of vaccine as a XIV, 1 Lantin: Treatment of Typhoid Fever 33 therapeutic agent, and they have drawn others into the ranks. Among the leading investigators in this field, according to Gay . and .Chickering,(11). are Ardin-Delteil, Negre and Raynoud, Boinet, Delearde and Leborgne, Sable, Netter, Roques and Al- faro, Feistmantel, and Garbat. In this series there were ten cases treated with sensitized typhoid vaccine intramuscularly, and the series was treated just after the series of cases treated intravenously with sensitized vaccine. TABLE IX.—Condition of patients before treatment. Severe and toxic Severe and nontoxic Mild bet Q Total Mode of treatment.—The sensitized typhoid vaccine, as stated previously, was administered intramuscularly. The dosage em- ployed was from 250 to 1,000 millions every three or four days. Gay advised 800 millions every other day and Garbat 500 millions every five or six days. The results of this treatment upon the course of the disease are shown in Table X. TABLE X.—Sensitized typhoid vaccine administered intramuscularly. j Cases. Dey § | Dura- | Dura- Classification by results. Severe | severe Total. — peed 3 i povthed mild. | 224 |" and ment. | ent. toxic. toxic. ‘ Days. ; Days. AbOtte -sis e ee See eas 4 0 0 4 10 $.2| 13.2 Bene tee fe. Se as aa Ses 0 2 0 2 16 11.0) 27.0 Relatively unaffected .._-.-....---.--.- 0 1 0 1 11| 32.0} 43.0 PONG a er ee 8 Se ak meee nies Seu tens Deeg Cridias. 22) oe eee es ee te ee ee pe PERE aor) © einer Siar oie RR RSE See inp ee Ie mh Renae Mn as Oy Geren on bier ha! alert S ae eet PVerage oo a oo a a ee aes capes 27.7 x. Thus, it may be seen that there were four abortive cases, all observed in the beginning of the second week, and all of them mild, in which the average duration of the disease was 13.2 days; two benefited cases, during the third week, both of the severe and nontoxic type, with twenty-seven days as the average dura- tion of the disease; and one relatively unaffected case of the same type as the benefited cases, with forty-three days as the average duration of the disease. The total average duration of 34 The Philippine Journal of Science 1919 the disease was 27.7 days. Fall of temperature by crisis was _ observed in only one case and fall by lysis in three cases, all in cases of the aborted type. Symptomatology.—The intramuscular injections were not followed by the train of symptoms observed in the case of the intravenous injections. However, there was a slight rise of temperature, appearing in the majority of cases from twelve to twenty-four hours following injection, and rarely after this time. The temperature in such instances may fall by lysis and, very seldom, by crisis. During the course of the treatment it was observed that a mild reaction was necessary in each case to obtain the desired result. Complications.—The complications are summarized in Table XI. TABLE XI.—Complications before and during treatment. Before | During treat- treat- ment. ment. Lobar p Moan tibia esta ebe sauces a 1 1 Bronchc Pp Bee ek anal eU ne dcwsesuskoudSececeuucccabescauee ee 0 1 OIRO Sa a a nie ty ec 0 0 NN aa rah an enh bea) Sees aE ve a 1 2 There were two cases that developed complications during the treatment, giving a rate of 20 per cent; one case was lobar pneumonia and the other bronchopneumonia. Mortality.—The total mortality in these ten cases was three, a death rate of 30 per cent, which is exceedingly high. Ex- cluding two cases admitted in a very serious condition, the mortality rate would be 10 per cent. The deaths that occurred did not follow immediately after injection, but two days later. Only two cases were autopsied; the findings are shown below: Autopsy findings (sensitized intramuscular treatment). File No. 1968: J. P. Anatomic diagnosis: Typhoid enteritis (healing); acute parenchy- matous degeneration, viscera; beginning lobar pneumonia, right; chronic fibrous pleurisy ; oedema of the meninges, File No. 1914: V. H. Anatomic diagnosis: Bilateral bronchopneumonia; typhoid enteritis (healing) ; cardiac dilatation; acute parenchymatous degeneration, viscera; acute cholecystitis, It will be seen, therefore, that the patients died of “typhoid accidents,” XIV, 1 Lantin: Treatment of Typhoid Fever 85 Garbat(8) in his paper quotes the results obtained by some investigators who employed sensitized typhoid vaccine intramus- cularly, thus: Boinet (1913), 15 cases; Szecsy, 112 cases; and Garbat (1911-1912), 17 cases. All of these authors observed good results. NONSENSITIZED, UNTREATED VACCINE, ADMINISTERED INTRAVENOUSLY Following the report by Ichikawa(15) in 1914, on intravenous injection of sensitized typhoid vaccine in the treatment of typhoid fever, many investigators used his method, but they em- ployed a vaccine of the plain bacilli killed by heat. Among Ichikawa’s leading followers according to Gay and Chickering (11) are Thiroloix and Bardon, Kraus and Mazza, Kraus, Biedl, Csernel and Marton, Rhein, Reibmayr, Mazza, Holler, Lowy, Luksch and Wilhelm, Paulicek, Dithorn and Schultz, and McWilliams. All of these investigators observed beneficial re- sults following the use of this form of vaccine. There were nine cases in this series treated with nonsen- sitized vaccine, administered intravenously. However, the patients in this series had been treated at a different time, following the treatment of the series that had undergone treat- ment by intramuscular injection of sensitized vaccine. The clinical grouping of these cases is shown in Table XII. TABLE XII.—Condition of patients before treatmert. Cases. Severe and toxic 1 Severe and nontoxic 6 Mild 2 Total 9 The results of this treatment are illustrated in Table XIII. TABLE XIII.—Nonsensitized vaccine administered intravenously. Days | Dura tetas Classification by results. —— Gave Total. pee ty Lowney —< Mild. nee Fc ment. | ment. toxic Days. | Days. PI eae 2 3 0 5 11.8 8.9 | 15.1 Pencdhited 352 eee 0 3 0 3 15.3 9.6 | 24.9 Relatively unaffected <.:.....s<-2+2-<--|.--.---=|2--.5-]--3- +255 0 biccoctubieeepeecianeencds UAB visi cate ie Sax tewkcencake<|oenwinlawewceen 1 1 hicewexcctebeceweatenes sous UE re etiednnianneenrmweclareenree 5 Ge eal Pe eee Mee ee Tigi a cra keen denne dion ean nn amen o ne paenenees pie EDC, SERIE ORE: HY aS AVORERO ooo Ks Ef aie sens eae sounhueaas Con cee camealesemeees 20.0 36 The Philippine Journal of Science 1919 . As shown in Table XIII there were five aborted cases, ob- served during the second week of the disease, of which two were mild and three were severe and nontoxic, with an average dura- tion of the disease of 15.1 days. On the other hand there were three benefited cases, all of which were severe and nontoxic, with an average duration of 24.9 days. The total average duration for the aborted and benefited cases was twenty days, in contrast to 24.1 days, the total average duration for the cases treated with sensitized vaccine intravenously. The difference in results may be accounted for by the difference in the severity of the cases, the period of observation, and the relatively fewer cases in the series. An illustration of the abortive type is shown in the accompanying charts (figs. 5 and 6). S| 5th day Cth day. 7th day 8th day. | 9thday y A.M. | A.M.| RM. A. M. EO AY REE A Te S| [2 6/02 6 /0ol6 /012\2 4 6 B 0122 4 6 8102 102 6 Ole 6 /0 be] +1 1}4 8 $ + Wiou 8 SL" 3 vel a e Xs ) 31S x = w Si | [ays ais] ja |s 8 S > do “apa es ~ S Si] Ss] 1 8S 8 & 8 : SG Sil iG} |S =I 8g A i PPro Fo! 38% i? 0 Ol | OO 37° -O Ot O Fic. 9. Temperature chart of J. R. Abortive type of typhoid fever, treated with colloidal gold intravenously. Widal negative till discharge. day. Sth day. 9th Jay. /0th day. Mth day. 42 th day. 3 th day. AT ACMA 7 a 5 a 1A.MIE MIA MIP MIA.MSNP M.. 26 10|\2 6 10\2 6 10\2 6 IW2 6 10\2,6 10 > = Do) = h = NP siX Ny ak sik NS of? sik Nin n siz Mp n six Nh n sk Shy % MaRS 8, Centigrade Widal = PSec.intravenous —|\% NS a hd N SN 37 Do. *#Ic. 10. Temperature chart of I. R. Showing benefited type of typhoid fever, treated with colloidal gold intravenously. Symptomatology.—Letulle described the symptoms that ap- peared after the intravenous injection as consisting of a tem- porary rise in temperature preceded by chills, occurring from fifteen to forty minutes after injection, and followed by fall in temperature, sweats, general weakness, and alleviation of subjective symptoms. With my limited opportunities for ob- servation, so far, I can only say that none of my cases developed chills. There is the possible exception of one patient who felt a slight sensation of chilliness, although his temperature showed a slight rise. Whether this difference of observation is due to racial insusceptibility of Filipino patients to the after effects of the treatment or whether the preparation on hand was defective are matters for future investigation. . Other symptoms observed were slight leucocytosis, temper- ature fall by lysis, remitting or intermitting temperature, and, rarely, fall by crisis. On the following day amelioration of the subjective symptoms may be marked. Contraindications.—Letulle does not believe that delirium and high fever, tachycardia, small pulse, or albuminuria are contra- indications for the intravenous injection. According to him the kidney continues to function well, for polyuria appears after injection. He observed neither intestinal hemorrhage nor per- foration. It is not improbable that nephritis may be produced after a long-continued use of this preparation, because metals generally have an irritating effect upon the renal epithelium. This con- ception is supported by the experiment undertaken by Schébl in which he found marked nephritis in the kidneys of animals experimented upon with intravenous injection of colloidal silver preparation. * Chief of the serum section of the Philippine Bureau of Science. XIV, Lantin: Treatment of Typhoid Fever 45 In conclusion, Letulle has stated that this preparation is anti- thermic, antitoxic, and without danger to the patient. He bases his conclusions on the disappearance of typhoid symptoms such as stupor, etc. Complications.—The complications observed during the treat- ment are shown in Table XXII. TABLE XXII.—Complications before and during treatment. Lobar pn@imonitisss a5 cocel ss ech eked 3s id netEbeeNeE ietiewt 1 1 Intestinal h rears Se Se ge Be 2 Sl aieatnaaenbinna sus 0 a2 AE oi eg na ok aoa dan CebL ui ps uwasddadiade dkwonWaspucsedns 1 3 4 One case developed two days after injection and the other six days after injection. We see that there were three cases showing complications during the treatment, two of which were of intestinal hzemor- rhage and one of lobar pneumonia. The fact that the cases of hemorrhage did not appear early—that is, not until after two days following the injection—seems to show that the treatment did not induce this complication. The proportion of complica- tions during the treatment of the series was found to be 20 per cent. Mortality.—Out of fifteen cases in this series there were two deaths, a mortality rate of 13.3 per cent. Excluding one hope- less case, the mortality rate was 6.6 per cent. Autopsy findings of the two cases that were autopsied are shown as follows: Autopsies (colloidal gold treatment). File No. 2682: A. C. Anatomic diagnosis: Lobar pneumonia; typhoid enteritis; parenchy- matous degeneration of the viscera. File No. 2603: H. G. Anatomic diagnosis: Ulcerative enteritis, typhoid, healing; subacute splenitis; focal necrosis of the liver; chronic parenchymatous nephritis; chronic suppurative bronchitis; cloudy swelling of the liver and heart; trichuriasis; chronic pleurisy; emaciation. Undoubtedly the two fatalities resulted from complications. DISCUSSION . SPECIFICITY OF THE TREATMENT Vaccine therapy has been studied extensively during recent years. The practical application of this biologic principle to the therapy of typhoid fever has its clue in the fundamental 46 The Philippine Journal of Science 1919 phenomena of Ehrlich’s side-chain theory. The undoubted success of prophylactic vaccine seemingly has given rise to un- restrained enthusiasm on the part of early investigators and has lead them to advocate its use for therapeutic purposes. This treatment, therefore, is the outcome of conscientious study in immunology and it is not a surprise that its followers should approach the question from that point of view. Early in the history of vaccine therapy, Jenner(16) discovered the efficacy of the treatment. Fraenkel(7) employed it thera- peutically for typhoid fever with good results. Wright(50) has advocated and emphazied the specificity of its action. Waitz- felder(48) has advanced the view that the treatment is “logical, scientific, and exact. It measures up to and beyond the expectant plan of treatment, the one in present use, in that it reduces mortality and shortens the period of illness.” Rodet (41) em- ployed immune typhoid serum injected intravenously in typhoid cases with excellent results. Ichikawa(15) used the intravenous injection of living sensitized typhoid bacilli with excellent re- sults. Gay(9) and Garbat(8) in using sensitized vaccine have upheld the specific theory. Mechanism of cure.—To explain the exact mechanism of cure with this form of treatment is exceedingly difficult, if not im- possible. However, the interpretation of the mechanism of the action of vaccine has been given most serious consideration by other investigators, who have advanced hypotheses which we may review. Thus, Garbat(8) believes that the curative effect of vaccine is due to the production of antiendotoxins as the result of stimulation by endotoxins that had been liberated into the blood- stream after the complement in the blood had combined with the bacilli. Ichikawa,(15) on the other hand, is inclined to the supposition that in the case of intravenous injection followed by chill, rise of temperature, and then crisis, we have phenomena of anaphylaxis, followed by antianaphylaxis, Gay and Chicker- ing (11) believe that the cure is due to specific hyperleucocytosis and increased formation of antibodies in the blood circulation. Koranyi(20) has observed increased opsonic index in the blood after vaccine injection. NONSPECIFICITY OF THE TREATMENT It is the tendency among workers along these lines to interpret the complex action of vaccine therapy through phases of im- munology and the specificity of its action. The Specific theory has been regarded by the dissenters as untenable, XIV, 1 Lantin: Treatment of Typhoid Fever 47 Rumpf (43) observed very early the nonspecificity of the treat- ment. By injecting Bacillus pyocyaneus in typhoid he could procure results quite as favorable as those secured by Fraenkel after the injection of typhoid bacilli. Kraus(21) used colon vaccine intravenously in similar cases, and gained equally good results. On the other hand, Kraus(21) used typhoid vaccine in cases of puerperal infections with similarly beneficial results. Even Ichikawa(15) found that his paratyphoid patients were benefited by the injection of typhoid vaccine. Furthermore, I have observed that the use of intravenous injections of plain killed typhoid vaccine in any form of arthritis produced remark- able results in the majority of cases. This is in agreement with the observations of Miller(28) and Thomas. (47) Ludke, (26) in accordance with his view that bacteria consti- tute nothing more than a foreign protein in the animal economy of the host, employed another kind of protein of nonbacterial origin. He used deutero-albumose in typhoid, also with good results. Miller and Lusk(29) used 1 to 2 cubic centimeters of a 4 per cent solution of proteose intravenously in typhoid and ob- served similarly good results; 20 per cent of their cases so in- jected recovered by crisis after a single injection. Nolf(86) noted beneficial results in the use of peptone solution, injected either intravenously or intramuscularly. Sax! (44) successfully treated typhoid patients by intramuscular injection of sterile milk. In this study both peptone solution and sterile milk have been used by me with success. There is evidence, however, that foreign protein is not the only substance used successfully in combating the disease. Thus Letulle and Mage,(25) Gay,(10) and Labbe(23) employed col- loidal-gold preparations with good results. In this investigation colloidal gold was also used, and satisfactory results were ob- served; as may be seen, 20 per cent of the cases in my series aborted. Mitlander(30) used salt solution intravenously in three hundred cases of typhoid fever, preceded intravenously by 1 cubic centimeter of 20 per cent caffein and 10 per cent camphor, and marked improvement followed. ; Mechanism of cure.—Investigators in other fields have ques- tioned the specificity of vaccine therapy, a claim which they declare to be without foundation. To attack a disease with its own weapon, so to speak, seems rather paradoxical. For, then, this question has been asked—and it still remains without a satis- factory answer: Are we justified, in acute infectious disease, in introducing in any manner into a host more of the same toxin 16238283———4 48 The Philippine Journal of Science 1919 from the effects of which he is already suffering? The view that there could be no possible stimulation of antibodies in a patient already suffering from acute infection with the typhoid organism has been given much thought in the effort to explain the me- chanism of abortive cures, because it has been observed that there is no immediate change or increased concentration of anti- bodies in the patient after injection of vaccine.(18) In the case of pneumonia Cole,(4) at least, has not been able to produce immunity in animals rapidly enough to be of therapeutic value within six or seven days. The observations of other writers that there is a slight increase of Widal titer have been questioned on the ground that the Wright(28) opsonic index seems now- adays to be an unreliable guide, for the simple reason that the immunologist has no better method for determining immune body formation. It has been observed in experimental work on animals that there is increased formation of nonspecific ferments, such as protease and lipase.(19) It is believed the protease has no action on bacteria, but that it hydrolyzes toxic protein into simpler and nontoxic substances, and that the lipase becomes lipoprotein against organisms in the final analysis. Whether the thera- peutic application of other substances would mobilize the same nonspecific antiferments remains to be seen, although it is not improbable that such a thing may occur. That all improve- ments observed in bacteriotherapy might be wholly due to such Specific action is, in my opinion, doubtful; because the frequent occurrence of relapses forms one of the objections to the belief that bacteriotherapy is specific. Having set forth the two conflicting ideas in regard to the probable mechanism of cure, I am led to believe that the exact details of its mechanism are not fully known, and that in order to work out this problem a thorough codperation between clinicians and laboratory workers is very necessary to allow justifiable interpretations of results. Now, it might be asked, What form of therapy should be recommended in treating typhoid fever with the methods em- ployed in this investigation? This is an exceedingly difficult, if not impossible, question to answer. It should be remembered that, in order to answer it satisfactorily, all the cases in each series should be placed under similar circumstances; that is, each series should consist of the same number of patients, who should be of the same sex and approximate age, and treatment should be administered simultaneously ; the disease in each case should XIV,1 Lantin: Treatment of Typhoid Fever 49 be of equal degree of severity and occur during the same period of epidemic, and the same season of the year. When a physi- cian has the opportunity to handle a large number of cases at one time, and observe as closely as possible the factors enume- rated above, then and only then will a definite statement seem warranted. When laboring as I have done, under unfavorable circumstances, in which many of the ideal and necessary factors could not be obtained, to recommend a particular treatment as being the best would be premature, and unfair both to the ex- pounders of the several systems of therapy and to the treatment itself. There are, however, some clinical grounds for maintain- ing the view that any form of treatment will have beneficial effects, provided a moderate reaction can be produced. Let me lay emphasis on the principle of the early application of the treatment. The patients respond better during the early stage of the disease. In advanced cases the treatment is seldom ‘productive of good results. SUMMARY AND CONCLUSIONS For a period of seven months, extending from August, 1917, to February, 1918, ninety-eight cases of typhoid fever were admitted to the medical department in the Philippine General Hospital. Twenty additional cases were admitted during that period, but these were not included in the present paper, because they were already either convalescent or in a dying condition on admission. These ninety-eight cases were all Filipinos, sixty- five males and thirty-three females, ranging in age from 14 years up. The majority of the cases belonged to the working class and, as a rule, exhibited low powers of resistance, conse- quent on “their mode of life, under-development, and limited diet.” The mortality rate in these ninety-eight cases is 19.38 per cent; but, excluding the hopeless cases, the death rate was 10.22 per cent. These cases were classified according to their severity, the cases being grouped in series. Each series naturally comprised an unequal number of patients, was treated at different periods of the epidemic, and received different kinds of treatment. The treatment of the cases has been carried under two general groups; namely, (a) treatment with foreign protein, which in- eludes sensitized and nonsensitized vaccine, peptone, and milk; (b) treatment with colloidal preparation, such as colloidal gold. Table XXIII shows graphically the comparative results of the different treatments. 50 The Philippine Journal of Science ms TABLE XXIII.—Comparative results of different treatmenis. | | Total ; Mode of administra- | Pa- | Sverage , Variety of treatment. Ae see tents. | a Deaths. | ease PS aa cota or ae ad ee Days. |Number.' Per cent. Sensitized vaccine .._____.______. Intramuscular__-_____- 30 | 24,1 6 20 IASNT A a ee Contes oo 21 | 27.7 5 23.8 Sensitized vaccine _._.__.________ Intramuscular____.._- 10 27.7 3 30 Nonsensitized vaccine -.....____. Intravenous ___.._.._. 9 | 20.0 1 a ee Penne cst oh ests Sle Intramuscular_______. 5 | Py a 0 0 POA Se pu ecenomec see oe ea CSS Saga a ETC 8 | 21.0 2 25 (onoMsl wold’. 23h a Intravenous __________ 15 | 22.0 2 13.3 SORE Sader thn torso as bek ae ee ee 98 | oe ok | a lanpake eae In the investigation here discussed I have employed some of the treatments that I believe are stamped with the march of progress. After conscientious deliberation of the observations gained, both from my limited personal experience and from the experiences of other investigators in this field, I am inclined to believe that the nonspecific theory with regard to the action of vaccine as used in this investigation seems plausible, but that the exact details of the mechanism of action of each kind of treatment administered is not yet fully understood. However, it has been observed that a moderate reaction, following the administration of any of the treatments discussed, produced beneficial effects. ACKNOWLEDGMENTS To Prof. Ariston Bautista, chief of the department of medicine, University of the Philippines, I acknowledge sincerely my high obligations for financing this investigation, and placing at my disposal all the material available in the department; also to Drs. Luis Guerrero, A. G. Sison, and Elias Domingo, for their encouragement in this work: to Prof. F. G. Haughwout and Dr. Otto Schébl, for their help and suggestion; to my as- sociates, Drs. Facundo Esquivel and Trinidad Banuelos, for their valuable codperation; to Dr. Jose Hilario, for the preparation of the sensitized vaccine; to Dr. Ricardo Fernandez, for the translation of the French literature on the subject; and finally to the department of pathology whose autopsy records I have been permitted to use in this study. XIV,1 Lantin: Treatment of Typhoid Fever 5l PRwWnN ee - REFERENCES . BESREDKA. Cited by McWilliams (84). _ Ceci, R. L. Journ. Inf. Dis. (1915), 16, 26. . CHAMBERLAIN, W. P. Philip. Journ. Sci., Sec. B (1911), 6, 317. . Cots, R. I. Trans. Congr. Am. Phys. & Surg. for 1916. The Tuttle, Morehouse and Taylor Co., New Haven, Conn. (1917), 10, 138. . CURSCHMANN, H. Nothnagel’s Encyclopedia of Practical Medicine, edited with additions by W. Osler. W. B. Saunders Co., Philadelphia (1902), 383. _ DITHORN and ScHULTZ. Cited by Jobling, Petersen and Eggstein (17). . FRAENKEL. Cited by Hektoen (13). . Garpat, A. L. Med. Rec. (1916), 90, 1145. . Gay, F. P. Journ. Lab. Clin. Med. (1917), 2, 785. . IpeM. Cited by Gay (9). . Gay, F. P., and CHICKERING, H. T. Arch. Int. Med. (1916), 17, 303. GuTIERREZ, P. Philip. Journ. Sci., Sec. B (1914), 9, 375. . HEKToEN, L. Trans. Congr. Am. Phys. & Surg. for 1916, 10, 110. . Hetos, J. E. Am. Med. (1916), 22, 162. _ IcHIKAWA. Cited by McWilliams (34). . JENNER. Cited by Helms (14). . JoBLING, J. W., PETERSEN, W., and Ecoste1n, A. A. Journ. Exp. Med. (1915), 22, 603. _Ipem. Journ. Am. Med. Assoc. (1915), 65, 515. _ Ipem. Ibid. (1916), 66, 1753. . Koranyi. Cited by Gay (9). . Kraus. Cited by Jobling, Petersen and Eggstein (17). ; _ KRUMBHAAR and RICHARDSON. Cited by Gay (9). . LABBE. Cited by Gay (9). . LEAKE, J. P. Journ. Am. Med. Assoc. (1917), 69, 631. LETULLE,’M. M., and MAGE, M. Bull. Acad. Med. (1914), 72, 42. _ LupKE. Cited by Miller (28). . METCHNIKOFF and BESREDKA. Cited by Gay and Chickering (11). . Miter, J. L. Journ. Am. Med. Assoc. (1917), 69, 765. . Mruuer, J. L., and LUSK, F. B. Journ. Am. Med. Assoc. (1916), 66, 1765. _ MitLaANpeR. Abstracted in Journ. Am. Med. Assoc. (1916), 66, 1321. . McCrag, T. Osler’s System of Medicine. Lea Brothers & Co., Phila- delphia and New York (1907), 2, 207. . IpeM. Cited by Gay (9). _ Ipem. Cited by Gay and Chickering (11)._ _ McWituiams, H. I. Annual Report of the Health Officer of the Port of New York. Albany (1916), 98. . Nicuots, H. J. Journ. Exp. Med. (1915), 22, 780. _ Nour, P. Abstracted in Journ. Am. Med. Assoc. (1917), 68, 1349. . Oster, W. The Principles and Practice of Medicine. D. Appleton and Company, 8th ed. (1914), 38. _ Perruscuy. Cited by Gay and Chickering (11)... . QuapRONE. Cited by Gay and Chickering (11). 52 40. 41. 42, 43. 44, 45. 46. 47. 48. 49. The Philippine Journal of Science ROBERTSON, W. E. WN. Y. Med. Journ. (1916), 103, 777. RopeT, A., and LEGRIFFOUL. Presse Medicale, Paris (1910), 18, 969. Rocers. Cited by Gutierrez (12). Rumpr. Cited by Miller (28). SAXxL. Cited by Miller (28). IpEM. Cited by Miller and Lusk (29). SLADEK and KotLowsky. Cited by Jobling, Petersen and Eggstein (17). THomas, H. B. Journ. Am. Med. Assoc. (1917), 69, 407. WAITZFELDER, E. N.Y. Med. Journ. (1916), 103, 407. WHITTINGTON, T. H. Lancet (1916), 1, 759. 50. WRIGHT. Cited by Jobling, Petersen and Eggstein (17). | Pig? Ak 2. 3. 10. ILLUSTRATIONS TEXT FIGURES Temperature chart of B. Y. Abortive type of typhoid fever, treated with sensitized typhoid vaccine intravenously. Temperature chart of E. N. Benefited type of typhoid fever, treated with sensitized typhoid vaccine intravenously. Temperature chart of M. L. Benefited type of typhoid fever, treated with sensitized typhoid vaccine intravenously. . Temperature chart of N. C. Benefited type of typhoid fever, treated with sensitized typhoid vaccine intravenously. . Temperature chart of J. A. Abortive type of typhoid fever, treated with nonsensitized typhoid vaccine intravenously. . Temperature chart of C. G. Abortive type of typhoid fever, treated with nonsensitized typhoid vaccine intravenously. . Temperature chart of F. Vv. Abortive type of typhoid fever, treated with peptone intramuscularly. . Temperature chart of E. A. Illustrating the effect of intramus- cular injection of milk. . Temperature chart of J. R. Abortive type of typhoid fever, treated with colloidal gold intravenously. Widal negative till discharge. Temperature chart of I. R. Showing benefited type of typhoid fever, treated with colloidal gold intravenously. 53 THE VALIDITY OF THE NAME DISCOMYCES FOR THE GENUS OF FUNGI VARIOUSLY CALLED ACTINO- ‘“MYCES, STREPTOTHRIX, AND NOCARDIA By E. D. MERRILL and H. W. WanbrE (From the Botanical and Bacteriological Sections of the Biological Labora- tory, Bureau of Science, Manila) The nomenclature of the group of fungi the pathogenic mem- bers of which produce the various actinomycoses, so-called, has been the subject of a confusion that resulted from an unusual combination of circumstances. For some time it was a moot question whether the organisms were of bacterial or of fungous nature, in part because of erroneous conceptions of their mor- phology, which is complex and variable, and differs widely in different strains; even yet opinions differ as to whether or not the forms involved should be included in a single genus. One of the types, a saprophyte, Streptothrix foersteri Cohn, was for a time erroneously included in a genus of the higher bacteria, while the first pathogenic species described, Actinomyces bovis Harz, having been recognized as a fungus, was given a different generic name. The question was further complicated by the fact that both names had long before been employed for entirely different organisms. Since then some authors have held one invalid, some the other, and some have rejected both. Other names have been misapplied from time to time, while new ones have been proposed, the list now including a total of ten. As is too frequently the case, the systematist and the pathol- ogist have tended to ignore the work and the viewpoint of one another. Medical writers, who almost exclusively have been con- cerned with the study of these organisms and consequently the use of their names, have been.very prone to choose these from the viewpoint of convenience and local custom rather than to recognize and adhere to the rules of nomenclature by which modern biologists are bound. On the other hand, botanists have overlooked or ignored—and they still do this—names and de- scriptions that have, in sincerity but without the formality cus- tomary with themselves, been published by medical writers. It is to consider the matter from both viewpoints in an effort to determine the actually correct designation that we have collabor- ated in a review of the vicissitudes of nomenclature that this group has undergone. e 56 The Philippine Journal of Science 1919 HISTORICAL The phase of the confusion in which the characteristic of true branching in these organisms was not appreciated began in 1875, when Cohn(18) described, among others, two organisms that he made the types of new genera. One, which he named Clado- thriz dichotoma, a colorless, filamentous plant found abundantly in water containing decomposing alge, was characterized by a false branching that he compared to that of certain alge; the other, which, apparently in ignorance of Corda’s(19) previous use of the same generic name, he called Streptothrix foersteri, was a branching filamentous organism said to have been found first by Graefe and then by Foerster in concretions in the lachrymal canal of man and classified by Waldeyer as Leptothrix buccalis. Cohn did not accept this conclusion, the mode of branching sug- gesting the mycelium of fungi. It has been pointed out by Sauvageau and Radais(56) that the distinction between his Cladothrix and Streptothrix was so clear to Cohn that in the text he did not even compare them; furthermore, that his illus- trations of them are quite distinct. They quote his diagnoses: Cladothriz—n. g. filamenta leptothricoidea tenerrima, achroa, non ar- ticulata, stricta vel subundulata, pseudo-dichotoma. Streptothriz—n. g. filamenta leptothricoidea tenerrima, achroa, non ar- ticulata vel anguste spiralia, parce ramosa. In his summary, howevers Cohn did not clearly differentiate them. According to Migula(44) he put them together among organisms showing false branching, although indicating uncer- tainty as to Streptothriz by an interrogation point: Zellfaden durch falsche Astbildung verzweigt, Faden cylindrisch, farblos—Cladothrix Cohn. Streptothria? Cohn’s later understanding of the morphology of the latter is evident from Israel’s article cited below. However, it is hardly to be suspected from this arrangement that, as is now generally recognized, the dividing line between the higher bacteria and the lower fungi separates these two genera. Bollinger, in 1876, demonstrated the fungous nature of the granules, or “drusen,” from the lumpy jaw of cattle. Attempts at cultivation and inoculation had been without result. In the following year(10) he published a description in which he stated - that Harz, to whom he had submitted fresh material, had con- cluded that the ray fungus (Strahlenpilz) belonged to the mold fungi and that it was related to Botrytis, Monosporium, and Polyactis; the name Actinomyces bovis was proposed for it. xiv,1 Merrill and Wade: The Validity of Discomyces 57 Rivolta, (53) in 1878, changed the generic name to Discomyces. After amplifying the descriptions of the granules (corpuscoli discoidi) that he had made in 1868 and 1875, he said in part: E vero chi i corpuscoli discoidi compressi si risolvono in pennelli od in ventagli fatti di rami e ramoscelli, mar percio non si ponno dire raggiati. Questa parola in storia naturale ha un senso ben determinato. Il com- plesso dei dischi che ci rappresenta, se si voule, un micelio, non ha la forma raggiata, e per consequenza non si puo denominar raggiata o come venne detto actinomyces, e nemmeno si debbono indicare i danni o le lesioni che produce con la parola actinomicosi. Il solo nome conveniente, a mio avviso, sarebbe quello di discomyces bovis, e con la parola sarcomicosi si potrebbero indicare le lesioni che produce vel corpo del bue. Harz (28) then published a separate description of the fungus, rejecting Rivolta’s change. Israel, (29) in 1878, used Actinomyces, but called attention to the similarity between the organism found in lesions in man and Cohn’s Streptothrix foersteri, a resemblance which, he said, Cohn himself had confirmed. Perroncito,(49) although himself employing Actinomyces, quoted a communication from Professor Garovaglio, director of the Cryptogamic Laboratories of the University of Padua, in which its previous use by Meyen (42) was noted. Rivolta (54) later declared that he was willing to accept Actino- myces bovis, but added that one could, nevertheless, form a group of pathogenic discomycetes containing: (1) Actinomyces bovis Harz; (2) Discomyces pleuriticus canis familiaris Rivolta; and (3) Discomyces equi Rivolta and Micellone. The second is now Cladothrix canis Rabe. (1898), and the third is known as a Micrococcus (M. botryogenes Rabe., M. ascoformans Johne, etc.). The first is, therefore, the only one of these organisms remaining in Rivolta’s genus, as thus amplified by him, and is the type of the genus, both as originally published and as later amplified. During this period systematists, who placed these organisms among the bacteria, denied the generic validity of Cohn’s Strepto- thriz. Winter,(61) Zopf, (63) Schroter, (58) and Baumgarten (5) considered it to be a synonym of Cladothriz. Schroter included, in the same family, the genus Actinomyces, this being apparently the first recognition of Harz’s organism in systematic classi- fication. Baumgarten concluded that the ray fungi belonged among the pleomorphic higher bacteria in the genus Cladothriz. MacFadyean(39) agreed that the organisms of actinomycosis probably belonged to the Schizomycetes; he held that the oc- currence of clublike elements in the granules was not of specific value because inconstantly formed. 58 The Philippine Journal of Science . 1919 Macé(87) also confused the genera, but in a new fashion. In 1888 he erroneously described for Cladothrix dichotoma a process of true branching and adopted(38} this generic name for the ray-fungus group. Sauvageau and Radais hold that he had never had the true Cladothrix under observation. Affanassiew(1) at first called the organism of actinomycosis Bacterium actinocladothriz, but in the following year, 1889, Affanassiew and Schulz(2) gave the term Actinocladothrix gen- eric rank. The only evidence that we have encountered of the use of this name by anyone else is the mention, without refer- ence, of “Actinocladothrix nocardt,” in an article by Haass. (27) De Toni and Trevisan, in Saccardo’s Sylloge Fungorum, (20) accepted these organisms as belonging to the Schizomyce- taceze. In the Cladothricee: “Sporae (arthrosporae) in fila- mentis normalibus obvenientes. Filamenta pseudo-ramosa” they included Sphaerotilus, Cladothriz, and a genus that they called Nocardia Trevisan: “‘Filamenta evaginata. Arthrosporae trans- formatione cocci singuli ortae.” In this genus they included Streptothrix Cohn, non Corda; Actinomyces Harz, non Meyen; and Discomyces Rivolta, five species being defined. The descrip- tion of these organisms as falsely branching was, of course, erroneous. In 1890 Almquist(3) and Gasperini(23) described certain or- ganisms that they identified as species of Cohn’s Streptothrix. Kruse held that these species fell, with the organism of actinomy- cosis, into Zopf’s Cladothrix group. Rossi-Doria (55) soon de- scribed six new species of Streptothrix from the air and classed Actinomyces bovis Harz, which he is said to have renamed Strep- tothriz actinomyces, with them. Kruse(31) later also employed Streptothriz, differentiating it from Cladothriz. From cases of actinomycosis in man Bostroem(11) repeatedly cultivated an organism that differed distinctly from that culti- vated by Israel. He concluded that it belonged to the Cladothrix group of the Schizomycetes and pointed out that it might be related to, or even identical with, Streptothrix foersteri Cohn. Griiber, (26) in 1891, described as Micromyces hofmanni an organism that subsequent authors have included in the group under discussion. Sauvageau and Radais’s(56) discussion of the confusion of Cohn’s Cladothrix and Streptothria has been referred to. They believed that the two were distinct; that Cladothriz, the most differentiated of the Bacteriacez, was falsely branched; and that Streptothriz, a true though very low hyphomycetous fungus, xiv.1 Merrill and Wade: The Validity of Discomyces 59 to which the organism of actinomycosis belonged, showed true branching. They concluded that the latter really belonged to Oospora Wallroth (1831), but that, whether or not this was correct, it was necessary to discard Streptothrix Cohn because of Corda’s use of this name in 1839. De Toni and Trevisan’s description of Nocardia as falsely branching was incorrect, for although Nocard(47) had originally so described his “Bacille de farcin,”’ Metchinkoff had found that it was a true-branching orgahism. Kanthack (30) accepted Oospora and created the name Oospora indica for the parasite of Madura disease, having demonstrated the identity of actinomycosis and of certain myce- tomas. Lehmann and Neumann, (33) in 1896, introduced Myco- bacterium as a family name for a group that they considered intermediate between the Hyphomycetes and the Schizomycetes, but rather more closely related to the former, and at first adopted Oospora as the generic name for the organisms under discussion. Gasperini,(24) in 1894, proposed the use of Actinomyces to include the whole group, discarding Streptothria; he listed eight- een species. Berestnew, (6) in 1897, accepted Actinomyces as valid and later(7) called attention to Gasperini’s publication, which apparently had been overlooked. Lachner-Sandoval, (32) in 1898, pointed out the invalidity of Oospora in this connection and also adopted Actinomyces. Levy (84) reviewed the question, concluding that all the described types were generically related and that Actinomyces was the proper designation for them. He did not note Rivolta’s original application of Discomyces. Leh- mann and Neumann, in the second (1899) edition of their work, substituted the family name Actinomycetes Lachner-Sandoval for their own Mycobacterium, the pathogenic forms placed in the genus Oospora now becoming Actinomyces. This broader application of the term to the entire group is not now widely accepted, though Mallory, (40) after Gasperini, employed it ten- tatively, and Babes(4) and other German authors still use it. Migula, in his earlier (1895) classification, (43) included these organisms among the higher bacteria, in his family Chlamydo- bacteriacese. He separated Streptothriz Cohn from Cladothriz, giving it a much modified diagnosis. In Cladothrix Cohn he included C. bovis (Harz) Migula (Actinomyces bovis Harz) and C. foersteri (Cohn) Schroter (Streptothrix foerstert Cohn), thus perpetuating the error of the earlier systematists. As already noted, Macé had adopted this generic name, although from a different viewpoint. Later (45) Migula modified this genus rad- 60 The Philippine Journal of Science 1919 ically, removing those species that are now recognized to belong to the fungi. Engler, in his Syllabus, (21) included Harz’s organism in the genus Sphaerotilus as “Sph. (Actinomyces) bovis,” thus adding a new name to the list of synonyms. He had not revised this grouping in the fifth (1907) edition. Discomyces Rivolta was shown to be the correct designation for the genus by Blanchard(9) who, stimulated by Levy’s and Berestnew’s articles, reviewed the question of nomenclature. in adopting this term he had changed his earlier opinion, for he ~ had previously(8) employed Nocardia. His argument is based on accepted principles and should carry conviction. Previous to this the term had been practically ignored. It is true that Sheube(57) cites Nocard and then Blanchard as having advo- cated this term for Discomyces (Streptothrix) indica, but we have been unable to find any publication by Nocard in which it is used; on the other hand, in the third (1903) edition of Nocard and le Clainche’s Maladies Microbiennes des Ani- maux,(48) Actinomyces is used in connection with actinomy- cosis and Streptothrix with “farcin du boeuf.” Gedoelst(25) evidently accepted Blanchard’s decision, for he designated the genus Discomyces Rivolta 1878, and the organism of actinomy- cosis Discomyces bovis (Harz 1877) Rivolta 1878. Stitt(59) is apparently the only American authority who has adopted this name. Brumpt(12) in a discussion of the mycetomas, used Discomyces and still subscribes to it, for in discussing organisms presented in 1913 by Pinoy(51) to the Société de Pathologie exotique (Paris) as Nocardia he used the former rather than the latter term. Manson,(41) in subscribing to Brumpt’s classifica- tion of the mycetomas, also used the same nomenclature. Cas- tellani and Chalmers(13) employed Discomyces in 1910, although they later discarded it. A new name was introduced by Ligniéres and Spitz, (35) who called a subtype of this group Actinobacillus. In a later arti- cle(36) they acknowledged the strict propriety of Blanchard’s argument in regard to the application of Discomyces to the general group, although they continued to use Streptothriz. During this period certain German authors had adopted Acti- nomyces and Streptothriz as separate genera. This is exempli- fied by Petrusky’s(50) classification in which they are placed in a family which he terms Trichomycetes. Wright(62) believed that Actinomyces should be retained for the organism of actino- mycosis, which he emphatically maintained should be differ- entiated generically from other organisms of the group. He x1v.1 Merrill and Wade: The Validity of Discomyces 61 rejected de Toni and Trevisan’s objection that Meyen had given the name to another organism as an unreasonably strict interpre- tation of the principles of botanical nomenclature. On the other hand, for the rest of the group he rejected Streptothriz and accepted Nocardia. This subdivision of the group has been adopted in several American bacteriological textbooks. How- ever, Chester (16) adopted the first classification of Lehmann and Neumann, except that Oospora was replaced by Streptothriz Cohn. Clements(17) included these organisms among the Schizo- ‘mycetes; he followed Migula’s earlier classification, except that Nocardia was substituted for Streptothrix Cohn, emend. Mig. Foulerton(22) argued that, since the other names that had been proposed had dropped out of use, only Streptothrix and Actinomyces remained to be considered. He chose the former because, although Corda had used it in 1834, it had become uncer- tain to what particular organism the term had been applied; further, a committee of the Pathological Society of London in 1899 had recommended the term “streptotricosis” as the ap- propriate clinical designation for the infection. Musgrave and Clegg (46) acknowledged that N: ocardia was probably more strictly correct, but “chiefly because of usage, and therefore somewhat arbitrarily, tentatively accepted Streptothrix * * *”? They suggested the possible advantage of substituting an entirely new name, Carteria (Carterii, sic!), evidently hoping that by this means further controversy might be eliminated. Pinoy has divided the group into Nocardia, which is to include most of the species, and Cohnistreptothriz, said to be designed to replace Cohn’s invalid Streptothrix. The article that con- tains his argument is probably one by Pinoy and Morax, (52) which is not available to us. According to Chalmers and Chris- topherson(15) the characteristics of this genus are preference for anaérobiosis, difficulty of cultivation, and nonproduction of arthrospores; in it they include Cohn’s Streptothrix foerstert and Israel’s Actinomyces from man (Streptothriz israeli Kruse, 1896). Vuillemin, (60) as a result of the adoption by the 1910 meeting of the International Botanical Congress at Brussels of a program for the next congress that included the determination of the point of departure for the nomenclature of the Schizomycetes and the elaboration of lists of nomina conservanda for these organisms, has recently published a revised generic classifica- tion, which was intended to be submitted for consideration at the scheduled London (1915) meeting of the congress. In an appendix to this work he includes the family Microsiphones, 62 The Philippine Journal of Science 1919 composed of genera to certain of which organisms such as the “bacillus” of tuberculosis and the “bacillus” of diphtheria are as- signed. For the genus under discussion he adopts Nocardia Trevisan, which he recommends for inclusion in the list of nomina conservanda. He says, in effect, that systematic botany need not concern itself with the “medical genus” Dvsco- myces, in which Rivolta combined, without mycological signifi- cance, the parasites of actinomycosis, botryomycosis, and canine pleurisy, nor with the genus Cohnistreptothrix, founded by Pinoy upon bacteriological grounds, and that Nocardia remains the valid name for the genus. As the Congress did not meet in 1915, Vuillemin’s recommendations have not yet been acted upon. Castellani and Chalmers have substituted, without discussion, Nocardia for Discomyces in the second edition of their work. (14) They remark that there are many points in favor of Pinoy’s subdivision of the genus, which probably would be soon generally accepted. The most recent discussion of this question is in a study of actinomycotic mycetoma by Chalmers and Christopherson, (15) who enumerate sixty-three species of Nocardia and eleven of Cohnistreptothrix. They argue the validity of Nocardia on the grounds: (1) that it is the oldest name against which no objec- tions can be raised; (2) that it has been formally adopted by the Botanical Section of the First International Congress of Pathology ;' (8) that there are objections to the other names in use. They eliminate Discomyces, because: Discomyces was used by Rivolta in 1878 merely as a trivial name, and though it has not been applied to any other genus, still the word Disco- mycetaceae was introduced in 1886 by Fries for a large fungal group and has come into general use, and therefore has the double claim of priority and general use, and as its type genus should bear the name Discomyces, confusion is bound to arise if the same term is retained for the generic name of Bollinger’s organism. The value of these objections will not be discussed at this point. SUMMARY The source and present status of the various names that have been applied to the organisms of this group may be summarized as follows: Cladothrix Cohn (1875). This name was used as generically *This probably refers to the Congrés international de pathologie com- parée, organized by the Société de pathologie comparée, the first and as yet only meeting of which was held at the Faculty of Medicine, University of Paris, in 1912. xiv,1 Merrill and Wade: The Validity of Discomyces 63 valid over Streptothrix Cohn (1875) by Winter (1884) and other systematists, the distinction not being understood. The organism of actinomycosis was informally assigned to this genus by Bostroem, Baumgarten, and others, formally by Migula (1895). Cladothrix Cohn is a different type of organism and the name is, therefore, inapplicable. | Streptothrix Cohn (1875), non Corda (1839). This name was applied by Cohn to a true-branching organism but was placed in his classification as doubtfully synonymous with Cladothriz. The resemblance of the fungus of actinomycosis in man to it was noted by Israel (1878) ; the name was adopted in 1890 by Almquist and by Gasperini for nonpathogenic air organisms, and in 1891 it was adopted by Rossi-Doria for that of actinomycosis. For a time this was probably the most widely used name for the group. At present it is frequently applied to the group minus the organism of actinomycosis. It is unquestionably in- valid in either connection because previously applied by Corda (1839) to an organism distinct from those under consideration. Actinomyces Harz (1877) non Actinomyce Meyen (1827). This name was applied by Harz to the fungus of “lumpy jaw” of cattle, by Gasperini (1894) to the entire group, replacing Streptothriz, and accepted in this application by Berestnew (1897), Lachner-Sandoval (1898), and others. It is now used by many writers, particularly the German and the American, as a valid name for the organism of actinomycosis only; it is seldom used in the more general sense. As is shown below, this name is invalid because published in connection with an entirely different organism by Meyen (1827). Bacterium Ehrenberg 1830. Affanassiew (1888) is said for a time to have called the organism of actinomycosis Bacterium actinocladothriz. This designation is manifestly inapplicable. Actinocladothrix Affanassiew and Schultz (1889). This was proposed as a generic name by Affanassiew and Schultz in 1889 for the organism of actinomycosis. It did not receive the con- sideration to which, being of even date with the widely adopted Nocardia, it was certainly entitled. Micromyces Griiber (1891). This name was applied by Griiber to an Actinomyces-like organism that he called M. hofmannt. This organism cannot be distinguished from the general group under consideration. Oospora Wallroth (1833). This was adopted by Sauvageau and Radais (1892), who concluded that the group belonged to Wallroth’s genus. Lehmann and Neumann (1896) adopted this view, but later abandoned it, Lachner and Sandoval (1898) 1623235 64 The Philippine Journal of Science 1919 having shown that Oospora Wallroth is an organism entirely different from those under discussion. Sphaerotilus Kuetz. (1883). This name was adopted by Engler for the group including Cladothrix (Streptothrix) foerstert Cohn, with which he included Actinomyces bovis Harz. This disposition was undoubtedly due to the old misapprehension as to the distinction between Cladothrix and Streptothriz. Actinobacillus Ligniéres and Spitz (1902). This name was ap- plied by Ligniéres and Spitz to a supposed subtype of this group. The distinction has not been recognized, and by most authors the name is considered a synonym. Carteria Musgrave and Clegg (1908). The adoption of this new name (as “Carterii”) was tentatively suggested by Mus- grave and Clegg as possibly advantageous for the purpose of avoiding further controversy, although they did not definitely advocate this highly informal procedure. Nocardia Trevisan (1889). This name was adopted by de Toni and Trevisan to cover the entire group. Blanchard used it for a time in its original application and Wright (1894) adopted it for nonpathogenic strains only. As many other authors use it in one sense or another, of late it has gained much prestige. Vuillemin, and Chalmers and Christopherson have recently adopted it for the entire genus. The validity of this name we deny on the grounds indicated in the discussion that follows. Discomyces Rivolta (1878). This name was definitely sub- stituted by Rivolta for Actinomyces, with the change of name of Bollinger’s fungus to Discomyces bovis. It was practically ignored until Blanchard (1900) argued its priority over Nocar- dia. Subsequently Gedoelst, Brumpt, Manson, Stitt, and for a time Castellani and Chalmers, adopted it. As indicated in the discussion this name is clearly valid over Actinomyces and all subsequent names. DISCUSSION Before considering the validity of Discomyces for this group over Nocardia and Actinocladothriz, it is necessary to emphasize the invalidity of two older terms that have gained general recognition. Streptothriz Cohn (1875) is invalidated by Streptothrix Corda (1839). It has been argued that there is doubt as to what organism is referred to by Corda’s name. This is apparently not the case for in recent years several new species have been described in Corda’s genus. Streptothrix Corda is unques- xiv,1 Merrill and Wade: The Validity of Discomyces 65 tionably a valid, recognized genus, and Streptothriz Cohn must fall. Actinomyces was used by Harz with but a very limited knowledge of the organism to which he applied it, evidently without suspecting its possible relationship to Cohn’s Strepto- thrix foersteri and probably without being aware of Meyen’s use of the name. Whether or not this newer application is valid, as most writers seem at least tacitly to agree, depends on the validity of its preémption by Meyen. That it is valid is evident from the following transcription from the original publication : Actinomyce Sporidochia, cellulis hyalinis simplicibus enormiter et multipliciter ramificantibus sporis impletis, substantiae uniformi gelatinosa hyalina induta. Actinomyce Horkelii R. forma irregulari sphaeroidea, gelatinosa duritie ad basin augente usque ad consistentiam cartilaginosam, colore hyalino-subcoeruleo. Hab. in pinguedine et pleuris animalium aquae submersis, autumno prope Coloniam Agrippinam. Zum Schlusse wage ich noch, etwas itiber das beginnende Wachsthum dieses Pilzes zu sagen. Der Pilz ist nicht eine Krankheitsform eines Organismus, sondern er ist ein eigener Organismus, ein eigenes Leben unabhangig von seinem Mutterboden, aber dennoch von demselben be- schrankt. It is to be noted that Meyen used the name Actinomyce. While by some the use of this form might conceivably be argued not to invalidate Actinomyces, the derivation of the two is identi- cal, and the argument cannot hold. Actinomyce horkelii Meyen is now an organism of uncertain status. Although it was described by Meyen as a fungus, the description apparently ap- plies to one of the colonial Cyanophycee. The genus is not recognized in either mycological or algological literature. How- ever, the description of both the genus and the species is indis- putably valid, and in the present connection the question of its identity is unimportant. In being validly published, it invali- dates the further use of the same name for another group of organisms in the plant kingdom. From the foregoing it is evident that by the accepted principles of botanical nomenclature both of these names are preoccupied. To deny on this ground either of them and yet accept the other, as has been done, is inconsistent. Recommendation of “strepto- tricosis” by a committee of the Pathological Society of London cannot be accepted as competent to validate Streptothriz, nor ean the adoption by the Botanical Section of the First Inter- national Congress of Pathology validate Nocardia. It is true 66 The Philippine Journal of Science : 1919 that through formal adoption by the proper bodies nomina con- servanda are validated; however, neither of the societies men- tioned has authority to do this in botany. There remains to be considered, then, the name next applied to this genus. This was published by Rivolta, in 1878, when he proposed, definitely and distinctly, to substitute Discomyces for Actinomyces.2. The reason for which he did this is an in- valid one; he believed that Actinomyces was not properly de- scriptive of the organism and, unhampered by rules of ‘nomenclature, adopted Discomyces as preferable in this respect. He was undoubtedly not aware of the fact that the former had been used before, but it is on this ground rather than that on which he advanced his new name that Discomyces is valid. Vuillemin, and more recently Chalmers and Christopherson, in advocating Nocardia as the valid generic name, hold that Rivolta’s use of Discomyces was trivial and without botanical significance. We do not agree with this argument, which is clearly refuted by Rivolta’s original paper. Here he distinctly proposes Discomyces bovis as the name for the organism called Actinomyces bovis by Harz in a manner that must be acknowl- edged as valid from the viewpoint of botany, even though it is not in conventional form and was advocated on irrelevant, in- adequate grounds. Therefore, it is in no sense a “medical genus,” as Vuillemin asserts. The fact that subsequently | Rivolta erroneously referred other organisms to this genus has no bearing on the case. His original application of it was to the organism of Bollinger and Harz alone, which is, therefore, the type of the genus. Nor does the fact that, to propitiate Harz, Rivolta later agreed to accept Actinomyces affect the question. As Blanchard pointed out, a name once introduced ?'The definite manner in which this substitution was made has been generally ignored, possibly because of the inaccessibility of the original paper, it having been published in an Italian veterinary journal. The rarity of this publication is exemplified by the difficulty that we have had in consulting it. The 1878 volume of Clinica Veterinaria was found to be missing from the set of this periodical in the Surgeon-General’s library in Washington, whereupon Mr. P. L. Ricker, of the United States Depart- ment of Agriculture, to whom we had applied, requested it from Mr. B. B. Woodward, librarian of the British Museum. He, not finding the publi- cation in that library, forwarded the request to Mr. F. Bullock, of the Royal College of Veterinary Surgeons, through whose kindness a separate of Rivolta’s article was forwarded to Washington, where photostat repro- ductions were made, one of which Mr. Ricker forwarded to us. To these gentleman we express our appreciation. xiv.1 Merrill and Wade: The Validity of Discomyces 67 is no longer the property of its originator to withdraw or modify at will. Finally, to argue, as do Chalmers and Christopherson, that Discomycetacez, a group name, invalidates Discomyces as a generic name in the connection in which Rivolta used it, on the ground that the type genus of Fries’s Discomycetacee, published in 1836, should bear the designation Discomyces, in- dicates an erroneous conception of the principles of nomencla- ture and priority in technical names; a family name such as Discomycetacee cannot invalidate the generic name Discomyces any more than a generic name can invalidate a similar specific name. This generic name was new with Rivolta, and there is no valid objection to its adoption in taxonomy. Nocardia is no longer to be considered. Both it and Actino- cladothriz of Affanassiew and Schultz, the publication of which seems to have been completely ignored,. were proposed eleven years later and fall as synonyms of Discomyces Rivolta (1878), which genus is typified by Discomyces bovis (Harz) Rivolta. The question of division of the group is a different matter. It is our conception that the group, exhibiting as it does wide differences among the species, should be subdivided. However, neither the characters upon which separation was advocated by earlier writers (granule formation in tissues, club-ended fila- ments in the granules), nor those advanced by Pinoy (difficulty of cultivation, anaérobiosis, absence of arthrospores), ‘seem to be convincing for generic distinction. Granules may be formed in animal lesions by a variety of these organisms, and club formation is a variable feature even in typical actinomycosis. Anaérobiosis and difficulty of cultivation are not generically dis- tinctive botanically, nor so considered for other groups of mi- crodrganisms from the bacteriological viewpoint. Furthermore, these features characterize both the strains described by Israel and those studied by Wright. Should it appear desirable to di- vide the genus, this will probably be done on the basis of mor- phologic rather than metabolic differences. REFERENCES (1) AFFANASSIEW. St. Petersb. med. Wochenschr. (1888), 13, 76 and 83. Cited by Blanchard (1910). (2) AFFANASSIEW and SCHULTZ. Third Congress of Russian Physicians, St. Petersburg, Subsection for Bacteriology. Ref.: Centralbl. f. Bakt., etc., 1. Abt. (1889), 5, 683; also Baumgarten’s Jahresber. (1889), 5, 398... (3) AtmQuist, E. Zeitschr. f. Hyg. (1890), 8, 189. 68 (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) The Philippine Journal of Science 1919 Bases. In Kolle and Wassermann’s Handbuch der pathogenen Mikroorganismen. 2d ed. Jena (1913), 5, 365. BAUMGARTEN, P. Lehrbuch der pathogenischen Mykologie. Braun- schweig (1886-1890), 2, 869. BERESTNEW. Aktinomykose und ihre Erreger. (Diss.) Moskau (1897). Ref.: Centralbl. f. Bakt., etc., 1. Abt. (1898), 24, 706. IpEM. Centralbl. f. Bakt., etc., 1. Abt. (1899), 26, 390. BLANCHARD, R. In Bouchard’s Traité de Pathologie générale (1895), 2, 649-932. Cited by Brumpt. é IDEM. Arch. parasit. (1900), 3, 193. BoLLiIncer, O. Centralbl. f. d. med. Wiss. (1877), 15, 481; and Deutsche Zeitschr. f. Thiermed. u. vergl. Path. (1877), 3, 334. BostrorM. Beit. z. path. Anat. u. z. allg. Path. (4891), 9)-1, Brumpt, E. Arch. parasit. (1906), 10, 489. CASTELLANI, A., and CHALMERS, A. J. Manual of Tropical Medicine. London (1910), 615. IpEM. Ibid., 2d ed. (1913), 812. CHALMERS, A. J., and CHRISTOPHERSON, J. B. Ann. Trop. Med. & Parasit. (1916), 10, 223. CHESTER, F. D. A Manual of Determinative Bacteriology. New York (1901), 360. CLEMENTS, F. E. The Genera of Fungi. Minneapolis (1909), 7. CoHN, F. Untersuchungen iiber Bakterien, II, Beit. z. Biol. d. Pflanzen, Bd. I. Breslau (1875). Quoted by Migula and others. Corpa, A. C. J. Pracht-Flora europaeischer Schimmelbildungen. Leipzig and Dresden (1839), 28. Cited by Saccardo, Sylloge Fungorum (1886), 4, 282, and others. DE ToNI, J. B., and TREVISAN, V. In Saccardo, P. A., Sylloge Fun- gorum (1889), 8, 927. ENGLER, A. Syllabus der Pflanzenfamilien. 3d ed. Berlin (1903), 5. FouLertTon, A. G. R. In Allbutt and, Rolleston’s System of Medicine (1906), 2, pt. 1, 302. GASPERINI, G. Ann. de microgr. (1890), 2, 449. Ref.: Centralbl. f. Bakt., etc., 1. Abt. (1891), 9, 59. IpeEM. Proc. Eleventh Internat. Cong. of Med. Rome. Ref.: Cen- tralbl. f. Bakt., ete., 1. Abt. (1894), 15, 683. GEDOELST, W. Les Champignons Parasites de l’Homme et des Ani- maux Domestiques. Lierre and Bruxelles (1902), 155. GRUBER, M. Trans. VIIth Internat. Cong. of Hyg. and Demog. London (1891) ; Section II, Bacteriology (1892), 2, 65. Also Ref.: Miinch. med. Wochenschr. (1891), 37, 653, Haass, E. Centralbl. f. Bakt., etc., 1. Abt. (1906), 40, 180. Harz, O. C. Jahresber. d. k. Centr.-Thierarzneischule z. Miinchen (1879), 125. Ref.: Jahresber. d. gesam. med. (1879), 1, 627. ISRAEL, J. Vireh. Arch. (1878), 74, 15. KANTHACK, A. A. Journ. Path. & Bact. (1893), 1, 140. Kruse, W. In Flugge’s Mikroorganismen. 3d ed. (1896), 48, LACHNER-SANDOVAL. Ueber Strahlenpilze. Eine bacteriologische- botanische Untersuchung. (Diss.) Strashourg (1898). Cited by Blanchard (1900), and others. xiv,1 Merrill and Wade: The Validity of Discomyces 69 (33) LEHMANN, K. B., and NEUMANN, R. O. Atlas und Grundriss der Bakteriologie und Lehrbuch der speciellen bacteriologischen Diag- nostik. Miinchen, 1. Auf. (1896), Teil IT. (34) Levy, E. Centralbl. f. Bakt., etc., 1. Abt. (1899), 26, 1. (35) Licnizres, J., and Spirz, G. Bull. Soc. centr. Méd. veter. (1902), 487. Ref.: Centralbl. f. Bakt., etc. Ref. (1902), 32, 781. (36) IpEM. Centralbl. f. Bakt., etc. Orig. (1904), 35, 294, and 452. (37) Mac&, E. Compt. rend. Acad. Sci. (1888), 106, 1622. (38) IpEM. Traité pratique de Bactériologie. Paris (1889). Cited by Sauvageau and Radais. ; (39) MacFapyEan, J. Brit. Med. Journ. (1889), 1, 1339. (40) Matxory, F. B. Bost. Med. & Surg. Journ. (1895), 132, 296. (41) Manson, P. Tropical Diseases. 4th ed. New York (1907), 759. (42) Meyen, J. Linnaea (1827), 2, 441. (43) Micuta, W. In Engler and Prantl, Die natiirlichen Pflanzenfamilien. Leipzig, Teil I, Abt. ia, (1895), 1. (44) Ipem. System der Bakterien. Jena (1896), 1, 21 et seq. (45) Ipem. Ibid. (1900), 2, 1033. (46) Muscrave, W. E., and CLecc, M. T. Philip. Journ. Sci., Sec. B (1908), 3, 470. (47) Nocarp, E. Ann. Inst. Past. (1888), 2, 293. (48) Nocarp, E., and LE CLAINCHE, E. Les Maladies microbiennes des Animaux. 3d ed. Paris (1903), 2. (49) Prerroncito, E. Deutsche Zeitschr. f. Thiermed. u. vergl. Path. (1879), 5, 33. (50) Perrusky. In Kolle and Wassermann’s Handbuch der pathogenen Mikroorganismen. Jena (1903), 2, 832; and ibid., 2d ed. (1913), 5, 267. (51) Pinoy, E. Bull. Inst. Past. (1913), 11, 927 and 977. (52) Pinoy and Morax. Bull. Soc. Ophthalmol. Paris (1911), No. 3. Cited by Pinoy (1913). (58) Rivotta, 8. Clinica Veterinaria (1878), 204 (Nos. 7, 8, and 9). Also abstracted and discussed by Bollinger in Deutsche Zeitschr. f. Thiermed. u. vergl. Path. (1879), 5, 110. Also quoted by Blan- chard (1900). (54) IpEM. Giornale Anat. Fisiol. e Patol. (1884), 16. Cited by Blan- chard (1900). (55) Rosst-Dorta, T. Ann. R. Univ. Roma (1891), 1, 399. Ref.: Baum- garten’s Jahresb. (1891), 7, 348. (56) SAUVAGEAU, C., and Rapais, M. Ann. Inst. Past. (1892), 6, 242. (57) ScHEUBE, B. Die Krankheiten der warmen Lander. Jena (1900), 625. (58) ScHROTER. Die Pilze, Bd. III. In Kryptogamenfiora von Schlesien. Bakterien (1886). Quoted by Migula (1896), 1, 30. (59) Srirt, E. R. Practical Bacteriology, Blood Work and Animal Para- sitology. 4th ed. Philadelphia (1918), 158. (60) VumttemiIn, P. Ann. Mycolog. (1918), 11, 512. (61) Winter, G. In Rabenhort’s Kryptogamen-Flora von Deutschland, etc., Die Pilze. Leipzig (1884), 1, 1 Abt., 38. (62) WricHT, J. H. Journ. Med. Res. (1904), 13, 349. (63) Zopr. Die Bakterien. 1. Aufl. (1884); 2. Aufl. (1885). Quoted by Migula (1896), 1, 26. IONTHA IDA, A NEW PHILIPPINE NOCTUID By CHARLES S. BANKS* (From the Department of Entomology, College of Agriculture, University of the Philippines, Los Bafios) ONE PLATE LEPIDOPTERA: HETEROCERA NOCTUIDA: QUADRIFIN 4? Genus IONTHA Doubleday ? Iontha ida sp. nov. Male.—Head dark brown; eyes nearly black; apical segment of palpus ringed with ochraceous scales at middle, antenne reddish; body and wings dorsad and ventrad a faded, dark, yellowish brown, the wings above darker apicad and highly bronze-irides- cent; a darker reniform, median, subcostal spot on fore wing circumscribed by a very few heliotrope scales and punctured centrad by one or two ochraceous scales; a faint brown zigzag line connects it with caudal margin, at which point is a very tiny patch of heliotrope scales; a dark brown dot, similarly circumscribed, five-eighths of the distance from base of wing to reniform spot; a submarginal row of six dark brown dots from apex to anal angle on outer margin. Each of these dots lies within an area of diffused heliotrope scales forming more or less of a circle around the individual spot but all running together to form an indistinct marginal band; ectad of this a thin, dark brown wavy line; marginal cilia alternately white and brown, the white being subjacent to the respective submar- ginal dots. Hind wings with a similar pale heliotrope band containing about five distinct, dark brown dots, almost obsolete at outer angle but growing more marked toward anal angle, which is much produced and lobed laterad and somewhat re- curved (See Plate I, fig. 1). Fore wings ventrad uniformly brown, irrorated with heliotrope on apical area and with a 1 Professor of entomology and chief of the department. The Entomologist (1842), 298, 1 figure. 71 72 The Philippine Journal of Science 1919 subcostal, ill-defined, darker brown, transverse spot beyond me- dian area; marginal cilia as on dorsal’ surface. Hind wings ventrad slightly darker iridescent brown, strongly irrorated with heliotrope scales over costal and outer areas and with a row of linear, heliotrope dots extending from costa obliquely across the cephalic two-thirds of wing to a line from base to anal lobe. Of these dots the costal is most prominent and almost white. These dots have faint counterparts on the dorsal surface but the latter extend farther toward inner margin, which is heavily fringed with pale brown cilia; the basal, median, and inner areas of the hind wing and the basal area of the fore wing on their dorsal surfaces are heavily clothed with long hair of a golden brown hue. Tegule concolorous with wings. Ventral surface of thorax and abdomen whitish ochraceous, growing darker toward apex of abdomen, so that sixth abdominal sternite has a diffused, longitudinal, brown, median stripe, and seventh is nearly all brown, with sublateral suffusions of helio- trope. Anal tuft nearly as long as abdomen, very dark iri- descent brown, the iridescence more marked ventrad. Fore and mid legs with femora and tibize brown ectad, creamy white entad; tarsi brown with creamy white at articulations. Hind legs uniformly brown, very hirsute, the tibie and tarsi having very dark brown hairs, lying in the same plane, lengthening toward apex of tibia and growing shorter toward apex of tarsus. Total length (including palpi and anal tuft), 48 millimeters; width of head, 5.25; length from front of head to apex of ab- domen, 30; length of anal tuft, 18. Expanse of wings, 60 milli- meters; length of fore wing, 28; length of hind wing, 23; length of antenna, 20; length of hind leg, 23; hind tibiotarsal length, 17; tibial fringe, 4. Luzon, Laguna, Los Bafios, P. I.; July 8, 1915 (Charles S. Banks). A single specimen flew into my house at night and, rapidly half crawling, half flying, around on the table, was cap- tured without injury. Type, No. 18468, male, in the entomological collection, College of Agriculture, Los Bajios, P. I. This species is closely related to Jontha umbrina Doubleday,* from which it differs most strikingly in having the anal angle of the hind wing lobed and much reflexed ectad; in having the heliotrope submarginal band extended to anal angle on hind wings and equally as well marked as on the fore wings, and in the alternately pale and dark brown fringe of the fore wings. ~ *The Entomologist (1842), 298, 1 figure. XIV, 1 Banks: A New Philippine Noctuid 73 Its very striking appearance differentiates it at once from all other noctuids that I have seen, and its graceful actions are not easily forgotten when once they have been witnessed. I dedicate this beautiful insect, the first and only one of its genus ever taken in the Philippines, to my mother, Ida Randolph Banks. Its closest ally, lontha wmbrina Doubleday, is recorded from Sylhet, in northern India, by all authors and from Singapore by Swinhoe. Hampson‘ gives Borneo in addition but, as he cites no collector, this may be an error. “Fauna of British India, Moths (1894), 2, 541, fig. 301. ILLUSTRATION PLATE I Iontha ida sp. nov., natural size. a, dorsal aspect; b, ventral aspect; c, lateral aspect. 15 BANKS: IONTHA IDA.] [PHILte. Journ. Sct., XIV, No. 1. PLATE |. IONTHA IDA SP. NOV., NATURAL SIZE. THE SOCIAL BEES OF THE PHILIPPINE ISLANDS By T. D. A, CockERELL (University of Colorado) The social bees of the Philippines are included in three families, easily distinguished as follows: Anterior wings with reduced venation; small, stingless bees.... Meliponide. Anterior wings with three submarginal cells; larger bees, with sting in females and workers. ge is 1. Eyes hairy; marginal cell long LDOROY. HOON coe cet ectigeeeeg ss Apide. Eyes naked; large hairy bees (HOMbiS Bete) Bombide. MELIPONIDZ A large family of social bees, abundant in the tropics of both hemispheres, but absent from the temperate parts of the north- ern hemisphere, though extending south of the tropics in Aus- tralia. The only Philippine genus is the following: Genus TRIGONA Jurine The record of T. leviceps Smith is probably erroneous. The following species are known to occur: Base of abdomen bright ferruginous, the following segments intense black, abruptly contrasting palavanica Cockerell. Abdomen ‘not thas bieolredicca ge oe co ee i si 1. Larger, the worker 6.5 to 7 millimeters long; abdomen ferruginous. luteiventris Friese. Smaller, worker about 3.75 millimeters long; abdomen dark. biroi Friese. Trigona palavanica Cockerell. Trigona palavanica COCKERELL, Ann. & Mag. Nat. Hist. (1915), VIII, 16, 2. PALAWAN, Puerto Princesa (from Baker). Trigona luteiventris Friese. Trigona luteiventris FRIESE, Résult. L’Expéd. Sci. Néerlandaise a la Nouvelle-Guinée. Leiden (1900), 5, Zoologie, 358. PALAWAN; also Perak. Trigona biroi Friese. Trigona biroi FRIESE, Termés. Fiizetek (1898), 21, 429. Philippines and New Guinea, according to Friese. Philippine Islands (C. R, Jones). 17 78 The Philippine Journal of Science 1919 LuZON, Los Bafios (Baker): Bataan, Lamao (P. J. Wester), at flowers of Nipa fruticans, February 26, 1916. I take the opportunity to make known some species obtained by Professor Baker in Penang and Singapore: Trigona ambusta sp. nov. Worker.—Length, 8.5 millimeters; anterior wing, 8 milli- meters. Head large, shining black, the clypeus ferruginous, broadly suffused with dusky above and at sides; antennez dark, except basal third of scape, which is bright ferruginous; man- dibles simple; front with fine brown pruinose tomentum; a band of stiff black hair behind ocelli; mesothorax and scutellum fer- ruginous, the latter with two dark marks and the mesothorax with dusky lines anteriorly; surface of mesothorax with very short thin rich fulvous tomentum, and anteriorly with black hairs; scutellum with short stiff black hairs; metathorax hairy at sides, but the broad central portion polished, shining black; pleura black in middle, reddish around sides, and red below; tegulz castaneous; wings fuliginous, with the apical field broadly reddish hyaline; stigma and nervure dusky reddish; legs black, the coxe and trochanters red; abdomen black (extreme base of first segment red), narrow, compressed, shining, with dark hair at apex. SINGAPORE (Baker 9067). Allied to T. lacteifasciata Cameron, from Borneo, but with black femora, basal part of wings dark, and other differences. It is also related to T. thoracica Smith, differing in the color of the wings, which Smith describes as flavo-hyaline in his species. Trigona atripes Smith, a variety differing a little in the color of the legs, comes from Penang Island (Baker 9068). The following species with black head and thorax were obtained by Baker on Penang or at Singapore: Mesothorax dull, bordered with fulvous hair (Penang). fulvomarginata sp. nov. (9073). Moacthcias not thus bordered..... DH 1. Small species, with red scape 3; Larger species; scape black, at most red at extreme base....................--- 3. 2. Tegule ferruginous (Singapore) valdezi sp. nov. (9074). Teguiz black (Penang).. -- penangensis sp. nov. (9075). 3. Larger; transverse-cubital nervures barely indicated (Singapore). busara sp. nov. (9072). Smaller; transverse-cubital nervures distinct.... 4. 4. Scutellum bare (Penene) na bakeri sp. nov. (9069). Scutellum conspicuously hairy (Singspors). itama sp. nov. (9071=type; 9070). XIV, 1 Cockerell: Social Bees of the Philippines 79 Trigona fulvomarginata is very close to T. ventralis Smith and has the abdomen whitish at base and beneath as in ventralis. It differs by the dusky wings and the bright fulvous hair bor- dering mesothorax and scutellum. The scape is pale at the extreme base, and the face has short grayish white hair. Trigona valdezi and penangensis belong to the iridipennis and biroi series. They differ at once from iridipennis by the dusky wings. The wings of penangensis are less produced apically than those of iridipennis, and the abdomen is pure black. Trigona biroi is larger than penangensis and has darker wings. Trigona valdezi is 5 millimeters long, but penangensis is not over 4. The abdomen of valdezi is brown, palest basally; that of penangensis is pure black. Trigona valdezi is also close to T. leviceps Smith, but differs by the black femora, tibie, and middle and hind basitarsi. Trigona busara is about 7 millimeters long, robust, with dusky wings; stigma and nervures dilute sepia; face and front cov- ered with cinereous pile; scutellum with much black hair; pleura with mouse-colored tomentum above, grading into cine- reous below; legs black. Trigona bakeri and T. itama are much alike, about 6 milli- meters long, with dilute fuliginous wings, noticeably darker than those of, JT. busara. The front mesothorax and abdomen are shining, but in bakeri the mesothorax is extremely smooth and polished, in itama distinctly dullish. The legs are black in both. Trigona busara, bakeri, and itama all have the abdomen shining black. The following, described by Smith from Singapore, are not represented in the collection: Trigona fimbriata, T. leviceps, and T. thoracica. Smith described four others from Mount Ophir; one of them (7. atripes) was found on Penang. It is a fulvous insect, quite unlike the others here described. APIDA I recognize only a single genus, though the segregates pro- posed by Ashmead may be considered subgenera. Genus APIS Linnzus Large species, workers about 16 to 18 millimeters, with eyes somewhat converging above; second recurrent nervure joining third submarginal cell very near its apex * a caioussescigiabac Subgenus Megapis Ashmead. Medium-sized species, typified by the common honey bee; second recurrent nervure not going so near end of third submarginal cell. Subgenus Apis Linnzus. Small species, workers about 8 millimeters........ Subgenus Micrapis Ashmead. 1623236 80 The Philippine Journal of Science 1919 Subgenus Megapis Ashmead Basal half of abdomen clear LOPTYUMINOUES 2 eens alas dorsata Fabricius. Abdomen black, with a band of,white tomentum at base of second segment. binghami Cockerell. Subgenus Apis Linneeus Length of worker, 10 to 13 millimeters; labrum black... mellifera Linnzus. Length of worker, 9 to 11 milimeters; labrum and more or less of clypeus Pees POGUE ee indica Fabricius. Subgenus Micrapis Ashmead One species; labrum and clypeus dark florea Fabricius. Apis dorsata Fabricius. Apis dorsata Fasricius, Ent. Syst. (1798), 2, 328. Listed by Ashmead. Apis binghami Cockerell. Apis binghami COCKERELL, Canad. Entom. (1906), 166 (zonata Smith, preoccupied). . LUZON, Mount Banahao (Baker); Bacoor (P. L. Stangl) ; reported by Ashmead. ses Apis mellifera Linnzus. Apis mellifera LINNZ&vs, Syst. Nat. (1758), 10, 576 (later called mellifica by Linnzus). . Presumably occurs only as a domesticated insect. Apis indica Fabricius. Apis indica Fasrictus, Ent. Syst. Suppl. (1798), 274. LUZON, Los Bajios. MINDANAO, Dapitan (from Baker). The Philippine specimens seen by me have the abdomen banded con- spicuously with black, and belong to the race nigrocincta Smith. The form unicolor Latreille, with black abdomen, has been re- ported by Ashmead from Cagayan and Alcala. Ashmead reports nigrocincta from Manila (Stanton). Apis florea Fabricius. Apis florea FABRICIUS, Mant. Ins. (1787), 1, 305. Said to occur in the Philippines. I have none from the Islands. BOMBIDA Genus BOMBUS Latreille Body covered with pale hair, some black intermixed on abdomen; wings BIMING coc cccicccie cc, bes -- Mearnsi Ashmead. XIV, 1 Cockerell: Social Bees of the Philippines R1 Hair of head nearly all black, of thorax above black, but on pleura fulvous; abdomen with first two segments yellow-haired, the others with black; Wings Tualiginoig scenes sees irisanensis Cockerell. Bombus mearnsi Ashmead. Bombus mearnsi ASHMEAD, Proc. U. S. Nat. Mus. (1905), 28, 959; COCKERELL, Ann. & Mag. Nat. Hist. (1905), VII, 16, 393. MINDANAO, Mount Apo, 6,000 feet (HF. A. Mearns). Type in United States National Museum. Bombus irisanensis Cockerell. Bombus irisanensis COCKERELL, Ann. & Mag. Nat. Hist. (1910), VIII, 5, 416, Luzon, Benguet, Irisan (collector unknown). Type in Brit- ish Museum. Structurally resembles B. swmatrensis Ckll., from Sumatra, but the colors are quite different. Bombus is known - from Java (B. rufipes Lep.), Sumatra (B. senex Snell., B. rufipes melanopoda Ckll., and B. sumatrensis Ckll.), and the Philippines ; _ but not yet from Borneo. THE PHILIPPINE BEES OF THE FAMILY NOMADIDA# By T. D. A. COCKERELL (University of Colorado) The Nomadide are represented in the Philippine Islands by species of Nomada Scopoli, a genus very widely spread over both hemispheres. The Philippine species are all small and look like small wasps. They are parasitic, presumably in the nests of the various species of Halictus. Those so far recog- nized may be separated thus: Genus NOMADA Scopoli Anterior wings with two submarginal cells..............---.-..-.------:e---eeeeee 1. Anterior wings with three submarginal cells..... Sa 4. 1: Meéesothorax of female black 2.2.0.2. cc pervasor sp. nov. Mesothorax of female red, at least at sides.............. z. "S- Pront black, except along orbits u.......2.. cn. attrita sp. nov. I GG ie Seek i 3. 3. With a yellow spot at each side of second abdominal segment. makilingensis Cockerell. Without such yellow spots Ns jsabcdeWinpaddamtidideclebousscubtes palavanica sp. nov. a ee in ek 5. Females 6. 5. Third antennal joint short, little longer than broad.* banahaonis Cockerell. Third antennal joint long, much longer than broad. . mindanaonis Cockerell. 6. With a conspicuous yellow spot at each side of second abdominal segment. exheredans sp. nov. Waicnont 80th Bow i eee i re 7. Mesothorax black, at most with a little red at sides 8. Mesothorax red, with at most a blackish discal shade...............-.....---------- 9. 8. First abdominal segment with a broad red band; hind tibie red. concessa sp. nov. First abdominal segment black, with small red spots; hind tibize black, with the ends red bakeri Cockerell. 9. Hind margins of abdominal segments suffused with dusky; flagellum dull red beneath en PERE LE LOE OE ACCOR RAG FE lusca Smith. Abdomen usually clear red; flagellum black........ mindanaonis Cockerell. * The second joint is very small and partly hidden in the apex of the first, so it is possible to mistake the third for the second. 83 84 The Philippine Journal of Science 1919 Nomada pervasor sp. nov. Male (type).—Length, about 6 millimeters; black, with the following chrome-yellow markings: Basal part of mandibles (which have no inner tooth), labrum, lower margin of clypeus, small triangular areas at lower corners of face, claviform streak above eyes, scape in front except at base (but suffused with reddish), tubercles, tegule (except a dark spot), scutellum (which is bilobed), apical part of femora (especially in front), tibiz (except a dusky mark, and on hind tibize the middle half dark, except a stripe behind), anterior tarsi and middle basitarsi, large spots on each side of second and third abdominal seg- ments, smaller (paler) marks on fourth, a briefly interrupted band on fifth, and a broad complete band on sixth. Middle of face and a transverse mark on pleura reddish; apical plate of abdomen pale ferruginous, very deeply notched; flagellum long, black; third. antennal joint a little longer than fourth; meso- thorax very densely and coarsely punctured; wings with only two submarginal cells, the second transverse-cubital nervure lacking; apical part of wings strongly infuscated; basal nervure going far basad of transverse median; abdomen polished and- shining, first segment entirely black. Female.—Similar to the male in most respects; a large, quad- rate, subapical yellow patch (more or less emarginate anteriorly) on abdomen. LUZON, Benguet, Baguio (Baker), 1 male, 3 females. Nomada attrita sp. nov. Female.—Length, a little over 4 millimeters; differing from N. makilingensis thus: Smaller; front black, with a red band along each orbit; mesothorax with middle third or rather more black, lateral parts dusky red; abdomen beyond first segment suffusedly blackened. MINDANAO, Butuan (Baker collection). Nomada makilingensis Cockerell. Nomada makilingensis CocKERELL, Ann. & Mag. Nat. Hist. (1915), VIII, 15, 263. Luzon, Laguna, Mount Maquiling (Baker). Nomada palavanica sp. nov. Nomada mindanaonis, variety, CockErett, Ann. & Mag. Nat. Hist. (1915), VIII, 15, 4. Female.—Length, about 5.5 millimeters; differing from N. makilingensis by the absence of yellow spots on second abdo- minal segment, the black flagellum, and the dusky hind legs. XIV, 1 Cockerell: Philippine Nomadidz 85 PALAWAN, Puerto Princesa (Baker collection). I was evi- dently in error in regarding this as a variety of N. mindanaonis ; the first recurrent nervure joins the second submarginal cell much nearer the base than in that species. Nomada banahaonis Cockerell. Nomada banahaonis CocKERELL, Ann, & Mag. Nat. Hist. (1915), VIII, 15, 264, Luzon, Mount Banahao (Baker). The male varies in size; length, 4.5 to 6 millimeters. Nomada mindanaonis Cockerell. Nomada mindanaonis COCKERELL, Ann. & Mag. Nat. Hist. (1915), VIII, 15, 265. MINDANAO, Dapitan (type locality) ; Davao. PALAWAN, Puer- to Princesa. All from Baker. The tegule of the male vary to red. The females vary, Davao specimens showing more or less dusky shades on the abdomen. It becomes difficult to separ- ate females of this species from N. lusca, but I have no male of lusca from Luzon. It is quite possible that N. mindanaonis is only a local race of lusca, differing in average rather than absolute characters. The type is a male. Nomada exheredans sp. nov. Female.—Length, about 5.5 millimeters; red with blackish markings, the second abdominal segment with a large round yellow spot on each side; mandibles simple; front, occiput, and cheeks black, but a red band along orbits; scape long, red, with a dusky spot at apex behind; flagellum very long, reddish black, the basal half red beneath; third antennal joint about as long as fourth; mesothorax dusky red, with a broad median black stripe and obscure sublateral ones; white hair patches on meso- thorax dense and bright; tegule clear red; wings dusky at apex; basal nervure going a short distance basad of transverse median; second submarginal cell large; legs’ clear ferruginous, only the hind tarsi dusky; abdomen with broad dusky bands; venter clear red. LEYTE, Tacloban (Baker collection). Nomada concessa sp. nov. Female.—Length, about 6 millimeters; black,, marked with yellow and red; closely related to N. bakeri, but differing thus: Mesothorax very distinctly shining between the punctures, its lateral margins reddish; first abdominal segment broader, and with a broad band; hind tibiz red. MINDANAO, Dapitan (Baker collection). 86 The Philippine Journal of Science Nomada bakeri Cockerell. Nomada bakeri COCKERELL, Ann. & Mag. Nat. Hist. (1915), VIII, 15, 2638. LuZON, Mount Maquiling (Baker). Nomada lusca Smith. Nomada lusca SmitH, Cat. Hymenop. Insects British Museum, pt. 2 (1854), 248; BincHAM, Fauna British India, Hymenoptera (1897), 1, 465; COCKERELL, Trans. Amer. Ent. Soc. (1905), 31, 318; COCKERELL, Ann. & Mag. Nat. Hist. (1915), VIII, 15, 263. Luzon, Los Bafios; Benguet, Baguio (Baker). This species was briefly described by F. Smith from the Philippine Islands, the particular island not stated. I have examined Smith’s type in the British Museum. Bingham recorded the species from “Sikhim; Tenasserim; Ceylon; Philippines;” remarking how- ever that the variety he described appeared to be intermediate between the Philippine N. lusca and the Indian N. adusta. I have little doubt that true N. lusca is confined to the Philippines, the Indian specimens belonging to one or more distinct species. Bingham was inclined to lump species in Nomada; thus he placed N. subpetiolata Smith as a synonym of N. adusta Smith, but Meade-Waldo later reéstablished it as a valid species. THE OSTEOLOGY OF THE GIANT GALLINULE OF THE PHILIPPINES, PORPHYRIO PULVERULENTUS TEMMINCK WITH NOTES ON THE OSTEOLOGY OF TACHYBAPTUS PHILIPPENSIS (BONNATERRE) AND HYDROPHASIANUS CHIRURGUS (SCOPOLI) By R. W. SHUFELDT (Major, Medical Corps, U. S. Army, Washington, D. C.) FIVE PLATES Before describing the skeleton of this big paludicoline bird, it will be as well to present a brief history of what we know of its habits, distribution, taxonomy, and other matters of interest. For this history I am indebted to Mr. Richard C. McGregor, ornithologist of the Bureau of Science at Manila, who has kindly furnished me with the following notes: Porphyrio pulverulentus Temminck. Porphyrio pulverulentus TEMMINCK, Pl. Col. (1826), 5, Pl. 405; SHARPE, Cat. Bds. Brit. Mus. (1894), 23, 207; Hand-list (1899), 1, 109; McGrecor, Man. Phil. Bds. (1909), 81. Distribution.—Philippine Islands. ® At the time that Sharpe wrote volume 23 of the Catalogue of Birds, the British Museum possessed only two specimens of the Philippine blue gallinule; these are listed as follows: a. Ad. sk, Manila, Luzon. Hugh Cuming, Esq. [C.]. b. Juv. sk. Manila. Gould Coll. It is very unlikely that either of these was collected in the vicinity of Manila. That they came from Laguna de Bay is probable, for the species can be found at many points along the shores of that lake, but it is nowhere as abundant as the moorhen (Gallinula chloropus). Since the Catalogue of Birds was written, this gallinule has been collected in Bohol, Mindanao, and Mindoro, as well as in Luzon. It is probable that it occurs on all of the larger islands where there are suit- able lakes with shallow reed-filled water near the shores. The food of Porphyrio, as indicated by the stomach contents of speci- mens collected at Paete, Laguna, Luzon, consists of fresh-water snails, small insects, seeds, and vegetable matter that could not be determined. When startled or surprised this gallinule stands erect, as if at attention. From this and its conspicuous red bill the Spaniards called it “artillero.” The most conspicuous external features of this bird are the heavy bill and long, heavy feet. It has a striking resemblance to the pictures of Notornis, next to which the genus is placed in Sharpe’s Hand-list. See the text figure of Notornis in Knowlton, Birds of the World, page 325, and in Newton, Dictionary of Birds, page 592. You will notice that the 87 88 The Philippine Journal of Science 1919 frontal shield in the specimens of P. pulverulentus extends much farther backward than is represented by Newton for either Notornis or for Por- phyrio. If you happen to have a copy of Egyptian Birds, by Charles Whymper, London, Adam and Charles Black, 1909, you will find, facing page 168, some nice studies of Porphyrio madagascariensis that might pass for the Philippine Porphyrio if done in black and white. In times past, in various journals here and in Europe, I have published complete accounts of the skeletons of all the American Gruide, or cranes; a full description of the skeleton of Aramus vociferans; and the same of the majority of our Rallide, or rails, gallinules, and coots. These papers and monographs are now so well known to ornithotomists and to many ornithologists, that it will not be necessary to cite them by title in the present connection. This also applies to such descriptions as I have published on the osteology of paludicoline birds of the Old World and elsewhere. At the present writing I have been unable to obtain the skeleton of an American gallinule; so I shall compare the bones of the Philippine giant gallinule with the corresponding ones in the skeleton of a coot (Fulica americana). Such material has kindly been loaned me by the Division of Birds of the United States National Museum. (No. 19710, adult male ?) There are but few marked differences to be discovered when we come to compare the skeletons of the soras, the cranes, the short-billed rails, the gallinules, and the coots, or other closely allied forms in the same group. Still there are some interesting points to be noticed along such lines, and they are of generic as well as specific significance. Most of them, it would seem, pertain to the skull rather than to any other part of the skeleton. OSTEOLOGY OF PORPHYRIO PULVERULENTUS The skull—As compared with Fulica, there is a general lack of pneumaticity in the entire skeleton of this big gallinule, which is corroborative evidence that the demand for its being a good flier is considerably less; moreover, it points to the fact that its relation to Notornis is much nearer than any of its con- geners of the allied groups of the Rallide. This reduction of the amount of air gaining access to the inner recesses of the bones is well exemplified in the skull, as compared with that part of the skeleton in the coot; consequently we find it to be, in the gallinule, thicker, denser, darker, and proportionately heavier in comparison—a condition which is largely extended to other parts of the skeleton, as will be seen further on in this description. XIV, 1 Shufeldt: Osteology of Porphyrio 89 Viewed upon its superior aspect, the skull of Porphyrio will be seen to be very broad in its interorbital area, generally convex, with almost complete reduction of the superorbital glandular fosse, and, posteriorly, faintly differentiated from the parietal area of the cranium. The craniofacial line is not as strongly marked ag it is in Fulica, though on both the naso-premaxillary sutures are distinctly in evidence on this superior view of the skull. When we come to regard the cranium of this gallinule laterally, we find a number of very striking differences, as compared with what obtains in our coot. In the former, the superior mandible is proportionately far more massive, broader, and deeper from above downward. Furthermore, its decurvature is more pro- nounced, and it is carried more abruptly to a sharper apex. The osseous roof of the mouth is not as open as it is in the coot, while the external narial apertures in this Porphyrio are rela- tively, as well as actually, very much smaller and of an elliptical outline (Plate II, fig. 10). A lacrymal bone in our subject differs very considerably from that element of the cranium in Fulica; for in the first-mentioned bird its superior portion is curved and elongate, making close articulation with the external margins of the frontal and nasal. It is pneumatic, while the foramen is usually in the lower portion. This latter is rather broad, thin, curved, and pointed below; a small, free ossicle brings it in articulation with the infero- external angle of the thin, oblong, though thoroughly ossified pars plana of the same side. In the coot the lacrymal is trian- gular for its lower portion—triangular and very thin—being produced as a spiculalike point below. It is separated by a | wide interval from the pars plana, or ethmoidal wing. The latter is a thin lamina of bone presenting several peculiarities. Its superoexternal angle is produced forward as a slender process in contact with the under surface of the frontal, while inter- nally, immediately below this same surface, an extensive elliptical foramen is formed for the passage of the nerve to the rhinal chamber. In both the Porphyrio and the coot the interorbital septum is almost entirely lacking in bone, which also applies to the anterior cranial wall above and behind it. Both birds have the foramen rotundum circular and complete. Porphyrio pulverulentus has the zygomatic bar very straight, rather broad, and transversely much compressed, rendering its upper and lower edges sharp. We > The Philippine Journal of Science 1919 On the side of the cranium the crotaphyte fossa is well defined (Plate II, fig. 10), which is not the case in Fulica. Both species have a pronounced postfrontal process, which points directly downward in either species; while in Fulica the squamosal process is spiculiform. In the Philippine gallinule it is broad and transversely flattened, constituting a cdnspicuous character of the side of the skull. These birds have the osseous meatus of the ear very open, which admits, in the dried skull, of a complete view of the interior parts, the nature of the articulation of the head of the quadrate, the Eustachian passage, and other characters. This is also true of the coot; but in that bird the posterior wall of the osseous meatus is not nearly as much thickened, either relatively or actually, as it is in the big gallinule of the Philip- pines. Posteriorly, in either species, the supra-occipital prom- inence is conspicuously developed, and is not pierced by twin foramina as it is in many birds. A well-defined occipital ridge is present and very distinctly marked as a bounding line to this region. Above, on either side, the bounding line of the crotaphyte fossa runs into it; the points where the two lines meet it are 13 millimeters apart. The external surface of the cranial vault is flatter than it _ is in the coot and presents a pair of parietal eminences placed side by side. Beyond this the frontal region, superiorly, is smooth and convex from before backward, merging gradually with the nasal on either side, but terminating abruptly and mesially in a transverse line over the premaxillaries; the latter are thoroughly united, with all sutural traces absorbed. A quadrate has a large, inturned, blunt orbital process, with an extensive articulation for the mandible, it being triangular in form, with an articular, convex facet at each angle. A pterygoid is short and not much twisted upon itself; it is broadened anteriorly, and the two bones meet in the median line, when normally articulated as they are in life. The palatines are very intimately in contact in the median line, below the rostrum of the sphenoid, as far forward as the union with the bifurcation of the vomer, which latter fuses with them. The hinder part of either palatine below has an inner and an outer crest running forward and parallel with each other. Above, the broad orbital surface, however, is smooth and flat. Beyond this either bone mounds up in a peculiar way to meet the lower part of the ethmoidal wing and lacrymal bone of the same side. Still more anteriorly a palatine fuses with the maxillopalatine; the latter is separated from the fellow XIV,1 Shufeldt: Osteology of Porphyrio 91 of the opposite side by quite an interval. The vomer is keeled inferiorly; it is slightly spreading above and is pointed at its free anterior apex. Fulica has all of these bones far more delicately constructed and slenderer. In this species the prepalatines are extremely slender rods, the postpalatine parts being short, and far removed, on either side, from the pars plana. Moreover, a maxillopalatine is a shell-like bone, distinct from its fellow, being attached at the point of union of the zygoma, the nasal, the prepalatine, and the dentary part of the premaxillary. It is elliptical in outline, with its major axis parallel to the slender prepalatine, while its outer surface is concave, and the mesial correspondingly convex. The external narial openings in the coot are very large and elongate, which is not the case in Porphyrio (Plate II, fig. 10). As in many birds, the mandible is a V-shaped bone, with deep, thin rami and extensive symphysis in this species. There are no postarticular processes present; in fact, the hinder aspects of the articular ends of this mandible are substantially flat surfaces, only slightly concaved and lying in planes perpendi- cular to the long axis of the bone. The free margins, both above and below, are moderately rounded, while the dentary ones above are cultrate, terminating in a subacute apex anteriorly. There are two foramina in either ramal part; that is, the usual splenial one, which is elliptical in outline, and another, smaller one, posterior to it, halfway between the first and the articular extremity, which is circular in outline. Posteriorly, the ramal portion of the bone is thin, it being con- siderably heavier for its anterior portion. Most of the former moiety exhibits some pneumaticity, and this condition is present in a good part of the cranium and associated bones of the palatal region and face; it is also true of the ossa quadrata. The ear bones have been lost. In Fulica we also find a V-shaped mandible, with a number of its general characters agreeing with the corresponding ones in Porphyrio. The former, however, has the bone more elongate, narrower, and with a shorter symphysis. Finally, its ramal sides are not nearly so deep nor so thin. Then, too, in Fulica there is a special character not found in Porphyrio, which is well worthy of notice. It consists of little semicircular plates of bone attached, one on either side, just above the splenial foramen on the superointernal margin of the ramus. These platelets are directed horizontally and toward the median line. My im- pression is that each is covered with the horny theca of the 3 92 The Philippine Journal of Science 1919 lower jaw and is capable, in life, of being moved up and down in the vertical direction. They are not present in Porphyrio; and what their special function is in life is difficult to conjecture. (Specimen No. 19710, collection United States National Museum bird skeletons.) The hyoid bones (Plate I, fig. 4) are extremely slender and elongate, and the glossohyal remains in cartilage throughout life, while the very minute urohyal is prolonged by a short, threadlike extension. This part of the skeleton of the coot at hand has been lost, so no comparisons can be made. It is fair to presume, however, that the skeletal parts of the tongue in these two birds are very similar. The trunk skeleton—There are thirteen vertebre in the cervical division of the spinal column of Porphyrio wherein the pleurapophyses are not free; in the fourteenth they are small, and are found to articulate freely with the vertebra, while in the fifteenth each rib of the pair is long and as slender as a needle. They do not reach the sternum, nor are epipleural appendages present upon them. The fourteenth vertebra has some of the characters of a leading dorsal, and these are still more pro- nounced in the fifteenth vertebra (Plate III, fig, 11). The cervical vertebre in Fulica are far more delicately fash- ioned than they are in the big gallinule of the Philippines. Proportionately, they are more elongate, with the pleurapophyses of the fourth to the eighth, inclusive, needlelike and long; in Porphyrio, relatively as well as actually, they are shorter and blunter. Only the second, third, and fourth have low, blunt, neural spines upon them, while in the fourteenth and fifteenth these processes are conspicuous and resemble the dorsal neural spines, only they are not so broad anteroposteriorly. Hzemapo- physes are present on the atlas and on the next two vertebrz behind it; they then disappear, to be seen again on the eleventh, twelfth, thirteenth, and fourteenth, where they are thin lamine of bone in the median line and transversely compressed. The carotid canal is open for its entire length; while the vertebral canal, on either side, is completely surrounded by bone in the third to thirteenth cervicals, inclusive. In both Porphyrio and Fulica the first two dorsal vertebrz possess hemal spines resembling those of the cervical vertebre; they are short and transversely compressed. On the other hand, all of the vertebre in this division of the spinal column possess very large, thin, quadrilateral neural spines, which increase in size from first to last as we proceed in the anteroposterior direc- tion. Their free superior margins are slightly thickened; and ee +, A) Shufeldt: Osteology of Porphyrio 93 from the second to the eighth dorsal vertebra—the latter the last of the series—they are in contact at all of their superior angles, each vertebra with the one next behind it. The anterior and posterior margins of these neural spines: are concave in outline and sharp. This, in the articulated skeleton, leaves elliptical vacuities among them (Plate III, fig. 11). There are linking metapophyses on the superior outer extremities of the transverse processes of all these dorsal vertebre; while in Fulica the neural spines above are lashed together through the ossifi- cation of the tendons of the muscles of the back. As will be noted from Plate III, fig. 11, of the present article, the dorsal ribs of Porphyrio are long and slender, as in the case of all true coots and gallinules, with costal ribs, or hemapo- physes, to correspond with them. The leading five support epi- pleural appendages—six in Fulica. There is always a pair of slender pelvic ribs; but the costal ribs of this pair do not articulate with the sternum. The pelvis.—Porphyrio has a pelvis that is shorter and broader than it is in Fulica and in the gallinules—relatively with respect to the latter and actually with respect to the former. This increased breadth is especially marked in the postacetabular ‘portion of the bone; it is also of denser and heavier build in the big gallinule of the Philippines than in the other forms mentioned. The preacetabular portion is elongate, narrower, and markedly concaved on the external iliac surfaces, which in Porphyrio fuse with the superior margin of the presacral vertebre for its full length, completely closing up the iliac-neural posteriorly. Curiously enough, in the coots and gallinules the inner margin of the middle third of the preacetabular part of either ilium is concave and sharp, thus sweeping below the superior margin of the sacral crista, between the anterior third of the crest and the posterior, which latter is opposite the acetabule. At this interval the inner margins of the ilia are not in contact with * When the present paper had been finished up to this point, there came to hand a “complete” disarticulated skeleton of a Gallinula (No. 18889, United States National Museum). It is from an adult individual collected by Dr. W. L. Abbott on Ile St. Louis, of the Seychelles. The skull, hyoid arches, vertebre, and ribs in this species of gallinule resemble those parts of the skeleton in Fulica more than they resemble the corresponding bones in the skeleton of Porphyrio. It is to be noted, however, that the curious processes on the mandible found in Fulica are not present in this Gallinula; while the cranium, on the other hand, agrees in all respects with that of Fulica, and consequently exhibits the same differences when we come to compare it with the cranium of Porphyrio. 94 The Philippine Journal of Science 1919 the sacral crest, a little open slit being present as a consequence; this is also the arrangement in Gallinula. In most respects, the postacetabular portion of the pelvis in this Gallinula agrees in its characters with the corresponding ones as we find them in Porphyrio. There are two striking characters, however, which agree, upon comparison, in Fulica and Gallinula and depart from the corresponding ones in Porphyrio; they are these: The posterior-inferior angle of the ischium in the coot and in the gallinule is drawn downward and outward as a conspicuous process; this angle is truncate in Porphyrio. Again, the distal free extremity of the postpubic rod in Fulica and in Gallinula is bent downward at an angle with the rest of this rod, the bending taking place at a point opposite the posterior-inferior angle of the ischium of the same side. Turning to the skeleton of Porphyrio, we note that this angle of the ischium on either side is truncated, and that the postpubic rod is not bent at any part of its length (Plate III, fig. 11). In all the species of this coot-gallinule group of birds there is a very remarkable character on the inferior side, or ventral aspect, of the hinder part of the pelvis; it consists of a deep, pocketlike recess, or concavity, apparently formed by the exten- sion, on either side, of the ischium. Mesiad, it thoroughly cods- sifies with the last three vertebre of the pelvic sacrum, and its anterior free margin beneath is sharp and concave in outline. The completed osseous pockets thus formed are deep and capa- cious, occupying very nearly one-half of the postacetabular con- cavity of the pelvic basin. Each extends as far forward as the posterior margin of the rather large subcircular ischiadic fora- men on either side. There are seven small free caudal vertebre in Porphyrio to eight in F'ulica. In addition to these in each bird there is a more or less inconspicuous quadrilateral pygostyle. When duly artic- ulated, these vertebree form an arc, which is concave along its superior line (Plate III, fig. 11). The shoulder girdle and sternum.—tThere is a great similarity in the characters with respect to all the bones of this part of the skeleton in the bird forms referred to in the present paper. In Porphyrio a scapula, in the articulated skeleton, reaches almost as far back as the pelvis; it is narrow, flat, pointed, and markedly curved, being convex along its inner border and correspondingly concave along its outer one. In Fulica this is much narrower, longer, and less curved and does not seem to articulate with the os furcula, whereas it does so, extensively, in Porphyrio. XIV, 1 Shufeldt: Osteology of Porphyrio 95 A coracoid in our present subject is a stout, straight bone, much expanded for its sternal moiety, which part is concave posteriorly and nearly flat anteriorly. As in Fulica and Galli- nula it develops a sharp process at the outer inferior angle of its shaft, and the two bones in none of these genera meet in the coracoidal groove of the sternum in the articulated skeleton. All three bones of the arch assist in forming, superiorly, the “tendinal foramen.” The very delicately formed “fourchette” is a U-shaped bone in all of these paludicoline birds (Plate III, fig. 11), and lacks anything like a hypocleidium. At the median point of the arch below, however, there is usually developed a minute process on the upper side, directed upward. None of the bones of the pectoral arch in any of these genera is pneumatic, which is also true of the sternum; in fact there is little or no pneumaticity of any part of the skeleton, as we find it among the various genera of the marsh birds. Indeed, this is what we would expect in the case of fowls that make so little use of such powers of flight as they possess. There is no mistaking the sternum of any species of this in- teresting group, and the characters of the bone are much the same throughout. The anterior border of the keel in Porphyrio slopes away posteriorly, more than it does in Gallinula and the coots; but beyond this the differences are barely of generic rank (Plate III, fig. 11). The sternal body is narrow and much concayved on its dorsal aspect. Markedly prominent, the quadrilateral “costal proces- ses” are flaring and truncated superiorly. A very small manu- brium is present, and the costal grooves are practically con- tinuous with a median notch above them on the superior border. The “carina” is well developed, being concave on its anterior border and convex on its inferior, the “carinal angle” being acute. There are six articular facets upon either costal border in Porphyrio and Fulica, but apparently only four or five in the Gallinula from the Seychelles Islands. The midxiphoidal process, with the carina running the entire length of it on its ventral aspect, is bluntly pointed posteriorly, and an isosceles triangle in outline, the somewhat blunt angle being acute. This xiphoidal portion of the bone is profoundly one-notched upon either side; the notch being triangular, with the angle pointing anteriorly. This gives rise to a long, lateral xiphoidal process on either side; each process is narrow, of uniform width, and somewhat expanded at its free extremity. Plate II, fig. 7, gives some idea of the ossifications that take place in " 162323-——-7 96 The Philippine Journal of Science 1919 trachea; they are quite ordinary, which is likewise true of such ossifications as are met with in the larynx. The pectoral limb.—All the bones of either limb in Porphyrio are entirely nonpneumatic, and this is apparently the case in Fulica, Gallinula, and their near congeners in various parts of the world. In Gallinula, when the skeleton is carefully cleaned and bleached, the humeri—as is the case in all the bones—are extremely light and creamy white; but they present no pneu- matic foramina at the sites where, when present, they occur in the vast majority of birds. All the bones of the limbs in Porphyrio, even when carefully prepared, are dark and greasy— at least this is the case with those before me, and they were prepared at the United States National Museum.2 In their morphology the pectoral limb bones in Porphyrio and Fulica americana are wonderfully similar, the several bones being somewhat shorter and correspondingly slenderer in the latter species. The humeral shaft in the coot is a shade stouter. In Porphyrio the bone has an extreme length of 7.8 centimeters; in Fulica, 7.5 centimeters. Were these two bones found fossil, they would never be described by any com- petent avian paleontologist as having belonged to represent- atives of different genera. In the humerus of the coot the pneumatic fossa is deeper and somewhat more circumscribed ; and this, apart from the difference in length, is the only distin- guishing character of any consequence. The form of the humerus in Porphyrio is well shown in Plate I, figs. 1 and 6. It will be noted that the radial crest is very low (fig. 6); and the notch, or valley, between the humeral head and the thickened proximal portion of the ulnar crest overarching the pneumatic fossa is notably deep and character- istically conspicuous, ; Radius is nearly straight and inclined to be slender, while the ulna exhibits considerable curvature, thus insuring, in the articulated skeleton, a rather broad “interosseous space.” Radiale and ulnara of the carpus present the usual ornithic characters and articulations. The shafts of the carpometacarpus are long and slender (fig. 1), and the pollex phalanx supports a free claw at its distal end; but no such claw occurs on the terminal phalanx of the index digit. The general characters of the pelvic limb may be well seen in Plate II, fig. 8. *I am inclined to believe that Mr. Scollick made no attempt to degrease | or bleach this skeleton. XIV, 1 Shufeldt: Osteology of Porphyrio 97 The head of the femur is very small in proportion to the size of the remainder of the bone; and the pit for the ligamentum teres is extensive and rather deep. Rising somewhat above the summit of the bone, the trochanter major is very broad across its outer aspect. While bowed to the front, the shaft of the bone has a curious though very slight turn in it about its middle (fig. 8). Distally, the condyles are large, and the fibular notch in the outer one is notably deep. As a matter of fact, the skeleton of the pelvic limb of Porphyrio is not only big-boned for the size of the bird; but the individual bones are long, with conspicuous characters at their extremities. No patella is to be seen in this limb, and this is also true of the coots and gallinules. Porphyrio has a very long tibiotarsus as well as tarsometa- tarsus, and this is also true of the joints of pes (Plate IT, fig. 8). In the first-named segment of the skeleton of the leg, the cnemial process is rather conspicuously developed, it being confined to the upper part of the shaft. The “ectocnemial projection” ter- minates in a little hook at its lower angle. Fulica has the en- tocnemial process wonderfully developed; it not only rises well above the summit of the shaft of the bone, but also projects far forward and to some extent downward. Often the fascia at- tached to its inferior border ossifies to some considerable extent, especially the outer margin of it, running into the antero-inner surface of the shaft at a point below the “fibular ridge” on the other side. The “ridge” referred to is well developed in all of these paludicoline birds, distinctly so in Porphyrio, while in it, as in all of them, the fibula is very weak, and partly ligamentous below its articulation with this projection. At the distal end of the tibiotarsus, on the anterior aspect, we may note above the condyles the usual ossified tendinal bridge crossing the longitud- inal groove in that locality. eb Taking the unusual development of the proximal extremity of this bone into consideration, it is worthy of note that the condylar end is, relatively speaking, not so markedly enlarged; that is, the condyles, though of good size, are not strikingly bulky. As is usually the case, the outer one is the larger and the rounder in outline, and, anteriorly, the thicker in its trans- verse diameter. To some extent they project beyond the surface of the shaft, posteriorly, thus continuing the intercondylar valley in that locality. Hypotarsus of the tarsometatarsus is well developed, but in Porphyrio appears to be neither grooved nor pierced for the passage of tendons. It is both once-grooved and once-pierced 98 The Philippine Journal of Science 1919 in Fulica and in Gallinula, though feebly so in either case. Por- phyrio has simply a shallow, central depression there to guide the passage of the tendons of the muscles coming down from the leg. . Our big gallinule has the shaft of the long tarsometatarsus somewhat flattened on its anterior and posterior aspects, and these surfaces are strongly grooved for their entire lengths, particularly on the hinder aspect, where the tendinal gutter runs the entire length of the straight shaft of this element of the pelvic limb. A large accessory metatarsal is present, being hinged to the shaft by strong ligament at its usual site in ordinary birds; it © supports the big phalanx and claw of hallux. All the other phalangeal joints of the three anterior toes of pes are lengthy and of considerable caliber, as will be seen by referring to Plate II, fig. 8. This brief review of the osteology of Porphyrio shows it to be a form possessing many of the characters of the typical paludicoline fowls; at the same time it exhibits not a few others that are distinctly generic in kind, especially those that have been described above as pertaining to the skull and to the pelvis. NOTES ON THE OSTEOLOGY OF TACHYBAPTUS PHILIPPENSIS (BONNATERRE) Early in August, 1917, when Mr. McGregor sent me the skele- tons of Porphyrio pulverulentus, he also sent for my examination rough skeletons of Tachybaptus philippensis (Bonnaterre) and Hydrophasianus chirurgus (Scopoli); one each of these two species has been cleaned for me by Mr. Scollick, of the United States National Museum, an assistance for which I have to thank Dr. James E. Benedict, chief of exhibits of that institution. When Mr. McGregor sent me these skeletons, he also included brief life histories of Tachybaptus and Hydrophasianus, which I may some day use in another connection; the present brief notes refer, as their title indicates, simply to their osteology. The specimen of Tachybaptus philippensis, an adult male from Paete, Laguna, Luzon, P. I.; March 26, 1917 (McGregor), is No. 226033, United States National Museum. The entire skele- ton of this bird presents the characters of those found in that part of the anatomy of any small average grebe. Indeed, they are quite typical, as I find to be the case upon comparing the various bones with the corresponding ones in the skeleton of Podilymbus podiceps (No. 17272, United States National XIV, 1 Shufeldt: Osteology of Porphyrio 99 Museum) and those of Colymbus cornutus (No. 17873, United States National Museum). Taken as a whole, the skull of Tachybaptus agrees much better with the skull of Colymbus than it does with that of Podilymbus—a fact due principally to the shortening and broad- ening of the superior mandible and jaw in the latter form. The space between the superior peripheries of the orbits in the frontal region is quite as broad in the Philippine grebe as in the dabchick, while in the horned grebe it is much narrower. Tachybapius presents a peculiar character in the pterygoids not to be found in the skulls of the two species with which it is here being compared; this consists in their being, in the case of either pterygoid, gradually broadened out by thin lamine that pass from the quadrate end to the palatine of the same side, the broadest part being the distal third. In Colymbus these bones are very slender rods, with scarcely a trace of laminar expansion. The hyoid arches seem to have been lost in these skeletons except in Tachybaptus, where they have been preserved with the trachea. The glossohyal is rather short and broad, the urohyal being very slender. The hypobranchials are long, and the ceratobranchials very short, while the two together are reduced almost to hairlike proportions, so exceedingly slender are they in caliber. Throughout its length the trachea is simple in structure, and the numerous rings are entire and thoroughly ossified from first to last. The true grebes seem to possess eighteen cervical vertebre in the spinal column to seventeen of Podilymbus; possibly one may have been lost in the skeleton of the latter bird at hand, but I am inclined to think not. Tachybaptus has the first pair of ribs on the nineteenth ver- tebra; they are without costal ribs, though the epipleural ap- pendages are well developed and long. These do not ankylose with their ribs anywhere in the series, and they are found upon the next following four pairs of ribs, all of which latter have true costal ribs. There are two pairs of pelvic ribs; the first pair reaches the sternum through costal ribs, while the second does not. Behind these a very delicate pair of floating hzema- pophyses is seen. None of these possess unciform appendages. We find the same arrangement in Podilymbus podiceps, while in Colymbus cornutus there are nine pairs of ribs, the leading seven pairs having unciform processes. The first two pairs do not possess costal ribs; and the last two pairs, which are pelvic 100 The Philippine Journal of Science 1919 ribs, are without appendages. Only the first pair of these pelvic ribs has costal ribs, while an aborted floating hzmapophysis, long and slender, completes the series. Morphologically, the grebe’s pelvis is well known—compar- atively speaking as well as actually; it is shorter in Tachybaptus than in either Colymbus or Podilymbus; in the latter genus it is extremely long and strikingly narrow. Nothing of particular note marks the small caudal vertebre in this little grebe of the Philippines. In regard to its shoulder girdle, or pectoral arch, it comes nearer Podilymbus than it does to Colymbus, especially in the matter of the U-shaped furcula; for there is no hypocleidium present except in the case of Colymbus cornutus, where it is a well-developed peg of bone directed backward. The sternum presents all the usual characters as we find them in the grebes generally; it is nonpneumatic, and the mid- xiphoidal notch is notably shallower than in the dabehick and most other American grebes. This is also true of the lateral sternal notches, while the lateral xiphoidal processes are stouter and shorter than we find them among the birds last mentioned. As among the Colymbide generally, the long bones of the arm and hand in Tachybaptus are completely nonpneumatic, while in form they are long and slender. While the usual “sigmoid curve” is to be seen in the humerus, it is not nearly as well marked as we find it in many other groups of birds. Then, too, the general characters of the proximal extremity of the bone are more or less feebly developed—a fact that causes no‘surprise, in as much as these birds are feeble fliers as com- pared with many other aquatic species. There are no claws on the phalangeal joints, and the carpometacarpus is long and slender. Judging from appearances in the prepared skeleton, it would seem that the femur and the superior halves of the bones of the leg might be to some extent pneumatic, which is not the case in the grebes found in North America. Otherwise, the morpho- logy of this part of the skeleton in Tachybaptus is entirely in agreement with that of any small grebe; this especially applies to the patella and other features of the knee joint. As in other grebes the inner trochlea of the tarsometatarsus is situated pos- teriorly on the shaft of the bone, and the accessory metatarsal, with its joint and claw (hallux), is found well up the shaft above it. The ungual joint of the midanterior toe is much expanded and shell-like, while in the case of those of the other ® XIV, 1 Shufeldt: Osteology of Porphyrio 101 two toes, though thin and flattened, they are no wider than the joint next behind them in the case of’ either toe; this is also characteristic of the American grebes. NOTE ON THE SKELETON OF HYDROPHASIANUS CHIRURGUS (SCOPOLI). Hydrophasianus possesses no special osteological characters not found in any typical. jacana; as, for example, in Jacana spinosa (Linnzus), the Mexican jacana. This is what we would look for, in as much as it is but a good average representative of the Jacanidz of the world’s avifauna. Its skull is to some considerable extent pneumatic, including the mandible, while no other part of the skeleton enjoys that condition. There is an unusual resemblance in its skull to some of the sandpipers, and, to a slighter degree, this is likewise to be observed in the remainder of the skeleton; though this does not apply, to be sure, to the greatly lengthened digits of pes. The sternum is somewhat narrow and elongate, being pro- foundly once-notched on either side, which gives rise to long, slender xiphoidal processes, and a rather broadish midxiphoidal part with slightly projecting outer angles. As the skeleton of this species has probably not been heretofore figured—and certainly not from a photograph made direct from the dried skeleton—I am here giving sych a plate, which I per- sonally prepared from the skeleton forwarded me by Mr. Richard C. McGregor, and which now forms a part of the collection of bird skeletons in the United States National Museum. ILLUSTRATIONS {All of the figures of the plates are of natural size and are reproduced from the author's HIG: . 1. a Laie 10 photographs made direct from the specimens.] PLATE I Left pectoral limb of Porphyrio pulverulentus Temminck, seen upon palmar aspect. The bones figured on Plates I to III illustrating this paper are all from the same _ individual. Adult male, No. 226035, collection of the United States National Museum. Paete, Laguna, Luzon, P. I. . The mandible of Porphyrio pulverulentus, dorsal view. . The skull of Porphyrio pulverulentus, superior aspect. See Plate Il, fig. 10. . The hyoid arches seen from above; the cartilaginous parts of the glossohyal have been allowed to remain by the osteologist who prepared the skeleton at the United States National Museum. . The first eight cervical vertebre of Porphyrio pulverulentus, normally articulated and seen from above. . The right humerus seen upon anconal aspect. PLATE II Right lateral view of the trachea of Porphyrio pul. erulentus, with the bronchial extremity and a few rings above it removed. . Left pelvic limb of Porphyrio pulverulentus seen upon outer aspect. The patella does not exist in this species, its place being taken by a strong fibrous ligament. . The leading eight cervical vertebre of Porphyrio pulverulentus seen from below; they are normally articulated as in life, and the upper view of them is shown in Plate I, fig. 5. Right lateral view of the skull of Porphyrio pulverulentus, with mandible dissociated. See Plate I, figs. 2 and 3. PLATE IIT Left lateral view of the trunk skeleton of Porphyrio pulverulentus, with all its bones normally articulated. This gives a side view of the remainder of the cervical vertebre, the anterior ones being shown in Plate I, fig. 5, and Plate II, fig. 9. PLATE IV Fig. 1. Right lateral view of the skull of Hydrophasianus chirurgus (Scopoli); adult female, natural size. Mandible detached. Paete, Laguna, Luzon, P. I. Collection of the United States National Museum. 2. Hyoid arches of the specimen shown in fig. 1. 8. Trachea of the specimen shown in fig. 1. 4. Right pectoral limb of the same individual, seen upon palmar aspect. 5. Right lateral view of the trunk skeleton of the same specimen as in the other figures. 103 104 The Philippine Journal of Science PLATE V Outer aspect of the skeleton of the left pelvic limb of Hydrophasianus chirurgus. From the same specimen as shown in Plate IV. The patella is extremely small and is here shown in its tendon, occupying the extreme upper apex. SHUFELDT: OSTEOLOGY OF PoRPHYRIO.] [Puiute. JourN. Sct., XIV, No. 1. 6 . PLATE |. PORPHYRIO PULVERULENTUS TEMMINCK. SHUFELDT: OSTEOLOGY OF PORPHYRIO.] [Puiur. Journ. Sct., XIV, No. 1. PLATE Il. PORPHYRIO PULVERULENTUS TEMMINCK, SHUFELDT: OSTEOLOGY OF PoRPHYRIO. | [PuHiuip. Journ. Sct., XIV, No. 1. PLATE Ill. PORPHYRIO PULVERULENTUS TEMMINCK, SHUFELDT: OSTEOLOGY OF PORPHYRIO. | [Puturpe. Journ. Sct., XIV, No. 1. PLATE IV. HYDROPHASIANUS CHIRURGUS (SCOPOLI). SHUFELDT: OSTEOLOGY or PoRPHYRIO. | ({Purure. Journ. Sct., XIV, No. 1. Fy ei wi PLATE V. HYDROPHASIANUS CHIRURGUS (SCOPOLI). NEW OR RARE PHILIPPINE REPTILES By Epwarp H. TAYLOR (From the Section of Ichthyology, Biological Laboratory, Bureau of Science, Manila) TWO PLATES AND FOUR TEXT FIGURES In this paper I am describing five new species of snakes and five new lizards from the Philippine Islands. One species is founded on specimens in the Bureau of Science collections, two on specimens in the Santo Tomas Museum of Manila, and the remainder on specimens in my own collections. The species described and their respective type localities are as follows: ae SNAKES Typhlops luzonensis. Mount Magquiling, Laguna, Luzon. Typhlops manilz. “Philippines.” Typhlops longicauda. Bunawan, Agusan, Mindanao. Typhlops rugosa. Bunawan, Agusan, Mindanao. Trimeresurus mcgregori. Batan Island, Batan Islands. LIZARDS Lepidodactylus naujanensis. Naujan Lake, Mindoro. Gekko mindorensis. Pocanil, Mindoro. Siaphos kempi. Naujan Lake, Mindoro. Sphenomorphus lednickyi. Aroroy, Masbate. Sphenomorphus lanosi. “Philippines.” SNAKES Typhlops luzonensis sp. nov. Type.—No. 109, E. H. Taylor collection; collected on Mount Maquiling, Laguna, Luzon, May 12, 1915, by E. H. Taylor. Description of type—Head rather flat, broader than neck, the lower jaw not or scarcely visible in lateral profile; snout rounded, projecting, rather truncate, the end only slightly less deep than head on a level with eyes; portion of rostral visible above much longer and a little wider than the part below, failing to reach the level of the eyes by a minute distance, and minutely less than one-half the width of head; prefrontal forming a suture with rostral little less than one-third its width, larger than frontal, its longest sutures with the supraoculars; frontal, the smallest upper head scale, forming equal sutures with inter- 105 » 106 The Philippine Journal of Science 1919 parietal and prefrontal; supraocular about the same size as parietal, its lower point barely reaching eye; parietals some- what narrowed on their lower end; nasal completely divided; nasal suture arises from second labial and after passing nostril reaches rostral in a line horizontal to upper edge of nostril; nasals not in contact behind rostral; preocular reaching above level of eyes, about as broad as ocular, in contact with two labials below; its edge crosses over middle of eye; two postoculars only slightly differentiated from body scales; first labial very small, in contact with anterior nasal only; second labial nearly three time as large as first, touching both nasals and preocular; third labial more than twice as large as second, and a little larger than fourth; lower jaw narrow, about five scales on lower jaw between fourth upper labials; eye a visible black spot, very small, with no pupil evident ; about 338 scales from head to vent, 10 subcaudally; tail ending in a small spine, Color in life-—Above reddish olive brown; below yellowish brown. Each scale with a darker yellowish brown area, giving body a checkered appearance on close examination ; rostral, nasal, and labials on underside of snout yellowish white. Measurements of Typhlops luzonensis 8p. nov. Total length (mm.) 260 Tail (mm.) 4 Width of body (mm.) 4.5 Width of head (mm.) + £5 Width of tail (mm.) 4 Width of body in total length (times) 58 Width of tail in tail length (times) 1 ruber it differs in having the preocular in contact with two labials, the nasal completely divided, and the length greater in propor- tion to the width of body. (Here the width of body is contained in the total length 58 times, while in T. ruber, it is contained 36 times.) From T. kraalii it differs in having the rostral much more than one-third the width of the head Jn contact with two labials; the color is also somewhat different. Typhlops manile sp. nov. Type.—Specimen in Santo Tomas Museum, unnumbered ; la- beled “Filipinas ;” locality and collector unknown. XIV, 1 Taylor: New or Rare Philippine Reptiles 107 Description of type-—Snout rounded in front, projecting; a distinct depression crossing head in region of eyes. Rostral nar- rowed at a point on snout between nostrils, distinctly longer than wide below; rostral little more than one-third the width of head; nasals not in contact behind rostral; rostral reaching level of eyes; prefrontal rather large, narrowly in contact with frontal; supra- oculars large, their lower end not touching eye; frontal slightly smaller than prefrontal, about the same shape; parietals round- ing, a little broader than deep, smaller than the supraoculars; interparietal enlarged; nasal not completely divided; the suture issues from the second labial, then makes a backward deflection, widening the anterior part of nasal; preocular narrowed above, reaching above level of eye, but scarcely reaching below level of nostril, abruptly widened below eye, its posterior suture not crossing eye; nasal is much wider than either preocular or ocular; a small subocular scale in contact with second and third labials; preocular touches second labial behind this intercalated scale; ocular widens abruptly on a level with eye; it extends higher than preocular; first labial elongate; second higher and shorter, of nearly the same bulk; third very large, three or four times as large as second, reaching to near the top level of nostril, larger and higher than fourth labial (third labial on one side is fused with subocular); three scales border ocular behind; eyes very small but distinct; nostril comparatively large; lower jaw very narrow, in its middle scarcely two-fifths the width of head. Tail ends in a blunt spine; twelve scales under tail in longitudinal line; snout projecting 2 millimeters. Color in alcohol.—Reddish brown, darker on posterior two- thirds of body; anterior part rather grayish brown. Head dis- tinctly marked with darker and lighter areas; snout yellowish. Below lighter yellowish, each scale with a slightly darker area. Measurements of Typhlops manile sp. nov. Total length (mm.) 280 Tail (mm.) 5 Width of tail (mm.) 5.2 Width of body (mm.) 5 Width of head (mm.) a Scale rows Remarks.—This unique specimen was found in the collection of the Santo Tomas University, Manila. The container was labeled “Filipinas” with no indication as to the locality from which it came. This species has no close affinities among other species of the genus in the Philippines as characterized by the 108 The Philippine Journal of Science 1919 presence of a subocular. It belongs to the division of the genus of which 7. ater and T. inornatus are members, but differs from all in the very much larger number of scale rows. Typhlops longicauda sp. nov. Plate i figs; Type.—No. R 99, E. H. Taylor collection; collected at Buna- wan, Agusan, Mindanao, July 15, 1913, by E. H. Taylor. Description of type.—Head rather broader than neck, broadly oval in outline; snout with a sharp horizontal cutting edge, moderately projecting, not or but scarcely hooked in profile; rostral not as wide below as above, somewhat narrowed between nostrils, and failing to reach level of eye by more than one-half the depth of prefrontal; the latter wider than deep, larger than frontal, the suture formed with it larger than that with rostral which is only about one-fifth its width; frontal about as wide as deep, equal to the parietals which are about the size of the body scales and scarcely differentiated from them; interparietal somewhat larger than frontal; the supraocular diagonal, the lower point reaching the anterior level of eye, but failing to reach the horizontal level by a distance equal to its distance from nasal; two nasals, the anterior very small, the suture di- viding them arises from first interlabial suture; preocular nar- rower and much shorter than ocular, in contact with two labials below; ocular large, with a slight, rounded prominence over eye; eye and pupil distinct: four postoculars between parietal and fourth labial; four labials, first and second smallest, subequal in size; third more than twice as large as second and about one- half of fourth; scales in 26 rows; about 430 scales in a longi- tudinal line to above vent; 40 scales in a row on underside of tail. Color in life-—Above light yellowish brown, gradually be- coming lighter below. Head lighter; laterally a distinct, broadly rectangular, lighter spot, including the eye and reaching the mouth. Each ventral scale has a regular darker brown area. Measurements of Typhlops longicauda Sp. nov. Total length (mm.) ~ 840 Tail (mm.) 34.5 Width of head (mm.) 5.5 Width of body (mm.) : 6 Width of tail (mm.) 4.75 Body width in length (times) 56.6 Tail width in tail length (times) 72 Remarks.—This species has a very marked, apparently normal, enlargement of the pelvic region, which suggests a greater de- XIV, 1 Taylor: New or Rare Philippine Reptiles 109 velopment of the pelvic bones or the rudimentary leg bones. The tail in this species is believed to be comparatively longer than in any other of the extremely numerous species of this genus. The type specimen was obtained from the trunk of a small tree, which was bored full of tunnels by large black ants. It probably feeds on the larve of ants. There are eleven specimens in my collection. Typhlops rugosa sp. nov. Type.—No. R 97, E. H. Taylor collection; collected at Buna- wan, Agusan, Mindanao, July 14, 1913, by E. H. Taylor. Description of type——Head rough, the anterior outline broken by depressions between scales along the sutures. Rostral a little longer than wide above, not reaching to level of eyes, more than “one-third the width of head; below the enlarged part about as deep as wide, forming a slightly curved hook on snout; rostral dimly granular; prefrontal a little wider and somewhat smaller than frontal, its posterior point reaching a little beyond level of eyes; frontal as broad as long, the distance from oculars equal to distance between nasals, larger than interparietal; supra- oculars larger than frontal, wider than deep; parietals much larger than frontal, separated by an interparietal, which is smaller than frontal; parietal not twice as wide as long; nasal with a swollen prominence about and above nostril, which gives the anterior head outline a roughened appearance; nostrils latero- inferior, not visible from above; nasal cleft issues from first labial and barely passes beyond nostril, not wholly dividing the _ scale; preocular not as wide as, and much shorter than, nasal, in contact with two labials; eyes dim, barely outlined; two postoculars, the inferior largest, in contact with fourth labial; four upper labials, fourth largest, first and second smallest, subequal in size; scales in 26 rows; tail ending in a sharp spine; 479 scales in a dorsal longitudinal line. Color in life-——-Above brownish to golden yellow, slightly lighter beneath. Very little distinction between the two colors as they merge gradually on the sides. Measurements of Typhlops rugosa sp. nov. Total length (mm.) 395 Tail (mm.) 23 Width of head (mm.) 7.5 Width of body (mm.) 8 Width of tail (mm.) 6.5 Body width in length (times) 50 Tail width in tail length (times) 8.54 110 The Philippine Journal of Science 1919 Remarks.—Two other specimens besides the type were taken, an adult and a young. These two specimens were forwarded to Dr. Lawrence E. Griffin at the University of Pittsburg. They have not been at hand for comparison. All were taken in masses of fern roots growing in high forest trees. This species has no close affinities among the Philippine species unless T’. mindanensis be regarded as such. From this it differs in the size of the frontal, which is larger than the prefrontal in T. mindanensis. One has two, the other three, labials touching the nasal. In one the head is very rough; in the other, com- paratively smooth. Many other differences are obvious on a com- parison of the two descriptions. The roughness of the head is not unlike that in T. rossii and T. regine, but here the resem- blance ceases. Trimeresurus megregori sp. nov. Fig, 1. Type.—No. 748, Bureau of Science collection; collected on Batan Island, Batanes group (lying between Luzon and For- mosa), June 12, 1907, by Richard C. McGregor. Description of type.—Rostral a little wider than high, slightly narrower at top, visible above as a narrow line, bordered be- hind by a rectangular scale, dis- tinctly enlarged, which separates the two much enlarged supra- nasals; latter not or barely in contact with rostral, separated from anterior supraocular by three (four on right side) scales; two enlarged supraoculars fol- lowed by one or two small scales above eye; nasal single, large, triangular, visible above as a narrow line, the nostril, which is vertically oval, pierced near the lower border; canthus ros- tralis sharp, formed by the edge of nasal, the narrow elongate loreal following nasal, and supe- rior preocular; facia] pit sur- z 8, Hie oe A? He 4ope fie o f if Fic. 1. Trimeresurus megregori sp. nov., ° . from the type, X 1. a, head, lateral OUNded by second labial, which view; b, head, dorsal view; ¢, chin. forms anterior border of pit, and XIV, 1 Taylor: New or Rare Philippine Reptiles 111 middle and lower preoculars, which are much elongated; three preoculars; a narrow elongate subocular, as long as orbit; two or three postoculars; ten supralabials, first small, triangular, nar- rowly in contact with rostral; second high, reaching almost to canthus rostralis; third much the largest, broadly in contact with subocular; fourth and fifth scales separated from subocular, each by a single scale; temporal scales distinctly enlarged, larger than, or as large as, posterior labials; mental broadly triangular, wider than rostral; eleven lower labials, first, seventh, and eighth largest; a pair of large chin shields, much longer than wide, broadly in contact, bordered by three labials; 5 pairs of scales between chin shields and first ventral; 28 scales from angles of mouth across occiput; 13 scales between supraoculars; 29 scale rows on neck (at seventh ventral) ; 21 on body; ventrals, 175; 56 subcaudals; anal single; temporal scales perfectly smooth, body scales slightly keeled on the eight or ten median rows. Head rather angular, flattened above and depressed in supra- ocular region. Tail prehensile. Color in life.—Above bright yellow with a darker yellow lateral streak; tail with a few small reddish brown spots near tip (in alcohol entire snake almost paper white with practically no trace of marking). Measurements of type and cotype of Trimeresurus mcgregori sp. nov. Type. Cotype. r mm. mm. Total length 865 702 Tail 120 100 Head width 25 25 Head length 36 33 Eye to end of snout 12 92 Eye to mouth 6 5 Supraocular width 16 14 Length of eye 4.5 4 Width of eye 3.2 2.8 Variation.—A second specimen from the same locality shows a certain amount of variation. The scale counts are as follows: Ventrals, 179; subcaudals, 59; scale rows on neck (at seventh ventral), 29; body, 21 rows; scales between eyes, 13; upper labials, 11-10; lower labials, 12; three scales behind supranasals — bordering the rostral. On the right side the third labial does not touch the subocular; the lateral stripe covers one whole and a half scale rows. Remarks.—This species belongs to the T. gramineus Shaw group, which includes T. flavomaculatus and T. halieus, of the 1623238—8 112 The Philippine Journal of Science 1919 Philippines. It is differentiated from that group, however, by the striking color with no dark markings, a larger number of scales on snout and supraocular region, and larger unkeeled temporals; the supranasals are larger and more clearly dif- ferentiated. Mr. McGregor, its discoverer and for whom I take pleasure in naming it, states that it is not rare on Batan Island. In a memorandum dated June 12, 1907, he says: Our party went to the summit of the mountain. On the return a large yellow snake was found resting at about 2 meters from the ground coiled on some leaves that had lodged among the thick stems of a kind of large grass. The snake was struck with an alpen-stock and fell to the ground. In attempting to put a string on its neck I was scratched by the fangs, between the last two joints of my thumb. Mr. H. G. Ferguson immediately made several cuts across the wound with a pocket knife and tied a string around the thumb. My hand and forearm were swollen by evening. The swelling subsided within a couple of days. There was very little pain, and no further trouble was experienced. LIZARDS Hemidactylus luzonensis Taylor. Plate I, fig. 2, b and e. Hemidactylus luzonensis TAYLOR, Philip. Journ. Sci., Sec. D (1915), 10, 93. - This species was first described from a single mutilated female specimen. I have since examined several other specimens of this rare lizard. Four were presented to me by Dr. Edward S. Ruth, of the University of the Philippines. Several other specimens have been collected for the Bureau of Science col- lection. Description of species.—( Adult male, No. 1620, E. H. Taylor collection; collected at Manila in 1916, by Edward S. Ruth.) Head flattened; snout rather oviform, more than twice as wide as deep, elongate, little less than twice the diameter of eye, one and one-half times the distance of eye from auricular opening; latter distinctly oblique on one side, rather vertical on the other; rostral squarish, upright, with a median notch and a cleft nearly half the depth of the scale; nostril pierced between rostral, first labial, two postnasals, and a supranasal; supranasals sepa- rated by two scales; ten upper labials, the last two very small; nine lower labials; mental triangular followed by two pairs of chin shields, first pair touching one labial and forming a long median suture; second pair in contact with first pair and two labials, but separated from each other by three scales; scales * Compared with Stejneger’s description of a Formosan specimen. XIV, 1 Taylor: New or Rare Philippine Reptiles 1138 bordering upper and lower labials somewhat enlarged; scales on chin and throat small, those on abdomen imbricate and larger; snout covered with uniform granules, larger than those on occi- put or body; occiput with scattered tubercular granules, rather conical; body with about sixteen irregular rows of trihedral tubercles; latter present on limbs; tail but slightly depressed, with whorls of sharply keeled spines marking annulations, about eight spines in transverse rows at base of tail; below enlarged, broadened, imbricate scales; regenerated part of tail without spines; limbs moderate, with digits all clawed, without any trace of webs, the distal phalanx rising from near end of toe; latter long, compressed, much deepened near end; nine divided lamellz under longest toe, seven under longest finger; a distinct lateral fold from axilla to groin; a short series of femoral pores, five on one side, six on other side. Color in aleohol.—The specimen is light yellow-brown with no evidence of marking save a dark spot on the snout. Measurements of Hemidactylus luzonensis Taylor. mm. Total length (tail regenerated) . 123 Width of head 10.5 Length of head 18 Snout to vent 58 Foreleg 22.5 Hind leg 29.5 Variation.—The variation noted among specimens is largely in the markings. The live specimens examined usually exhibited a series of large dim dark blotches on the back, alternating with lighter markings. In most specimens the spiny tubercles on the back were white (see Plate I, fig. 2,b and c). The skin above the auricular opening forms an indistinct flap or fold (scarcely noticeable in preserved specimens), which is usually held distended in living or freshly killed specimens; this char- acter is very evident and is shown in Plate I, fig. 2, b and ¢. The fold of skin on sides of body is prominent in living specimens and is shown clearly in the figure mentioned. Remarks.—Evidently a rare form; most closely allied to H. depressus but easily distinguished by the longer snout, the char- acter of the tail-and the preanal pores, and the absence of webs on the feet. Known only from the Philippines. Lepidodactylus naujanensis sp. nov. Type.—No. 2006, E. H. Taylor collection ; collected April 25, 1916, at Naujan Lake, Mindoro, by E. H. Taylor. Description of type.—Rostral low, more than twice as broad 114 The Philippine Journal of Science — as high, not entering nostril; the latter surrounded by first labial and five nasal scales, forming a rounded prominence; postnasal bordering second labial, largest of the five; supranasals separated from each other by two series of three scales, all about the same size ag nasal scales; a rather distinct groove on front end of snout; 13-14 upper labials, 13 lower labials; angle of mouth extends scarcely behind posterior vertical of eye; mental scarcely larger than adjacent scales, chin covered with a large number of somewhat enlarged scales, about 35, which fill all the space in front of a line drawn across jaw between the fifth lower la- bials, those bordering labials largest; granules on snout larger than those on body; back and sides covered with minute granular scales with scattered, slightly larger, spinelike scales, which are yellow; these also occur on the back part of head and neck; tail rather cylindrical, with a distinct lateral denticulated fringe, the annulations marked by an enlarged spinelike scale in the lateral fringe; scales on upper surface of tail distinctly larger than those on body, those on underside still larger; scales on belly imbricate, much smaller than those under tail; a long series of 25 preanal and femoral pores forming a median sharp angle; pores elongate in Shape; limbs well developed, the ad- pressed hind leg reaches the wrist of the adpressed foreleg; web between toes and fingers very rudimentary; 14 lamelle under longest toe; toes much wider at end than at base, the basal lamelle rather scalelike; inner digits on limbs well de- veloped, without claws; eye nearer ear than end of snout; ear slightly nearer eye than foreleg. Color in life-—Reddish brown above with dim, zigzag, darker marking of brown across back ; lighter at base of tail; a dark line between eyes and another on nose; sides dark with minute yellow spots; a more or less distinct row of yellow dots borders belly ventrolaterally; belly canary mixed with brown scales ; underside of tail yellow at base, grayish at tip; lower part of eye dark. Measurements of Lepidodactylus naujanensis sp. nov. mm. Length 74 -Snout to vent 84 Snout to foreleg 12 Tail é 40 Axilla to groin 16 Foreleg 10 Hind leg 13.5 Width of head 5.1 Length of head 9.3 XIV, 1 Taylor: New or Rare Philippine Reptiles 115 Variation.—Two other adult specimens were taken in the type locality. The following variations are in evidence. One specimen, a female, has 14-13 upper labials; 12 lower labials; the internasal scales are arranged with one large median scute, with a pair of smaller scales on each side; the angle of mouth fails distinctly to reach the posterior vertical of eye; no preanal pores, but a series of 12 enlarged scales on each side, angular medially, representing the pore scales; a series of distinct black spots on each side of tail just above the lateral fringe. The other specimen, also a female, has the internasal scales similar to those of the type save that the second row has 4 instead of 3 scales; upper labials, 18-14; lower labials, 12; calcareous de- posits are present on the sides of neck. On the opposite side of Mindoro, near Sablayan, another adult male specimen was taken. It agrees remarkably well with the type. The row of seales bordering the lower labials are dis- tinctly larger than other chin scales. Preanal pores 12-11 in broad angular series; the body is gray with bronze-brown, irregular, zigzag markings. Young.—Six pairs of small eggs were found attached to various trees under bark. These were brought to Manila, and with no special care ten young were hatched. These were very lively, but owing to my inability to obtain suitable food all died soon or were preserved. A male and female emerged from each pair of eggs; they were a uniform grayish brown; the largest meas- ured 32.5 millimeters in length, the smallest, 29 millimeters. The eggs are smooth, white, flattened on the two sides where they were attached to each other and to the trees. A great many of these eggs were found, but most of them were de- stroyed in removing them from their resting places. Remarks.—This species is related to L. christiani Taylor, as shown by the arrangement of the nasals and the separation of the nostril from the rostral. It differs, however, in the develop- ment of the web between the toes; there is no skin fold on the femur, and the tail is essentially different. The type and cotype, from Naujan Lake, Mindoro, were found under leaves of small climbing vines on trees. Other specimens observed escaped. The habitat is especially similar to that of the small Siaphos kempi, described from the same locality. On two occasions the two species were observed on the same tree. This species is probably common in the type locality. Gekko mindorensis sp. nov. Fig. ye Type.—No. 499, E. H. Taylor collection; collected at Pocanil Point, Mindoro, May 4, 1916, by E. H. Taylor. 116 The Philippine Journal of Science 1919 Description of type.—Head moderately large, rather oviform; distance from eye to snout somewhat longer than distance from eye to auricular opening; latter at least one-half diameter of eye, oval, distinctly oblique; rostral large, wider than deep, with a suture above, medially; nostril large, obliquely oval, surrounded by rostral, first labial, two subequal supranasals, and a postnasal; rostral forming the longest side of nostril; a single scale inserted between the two anterior supranasals; twelve upper labials; ten lower labials ; the line of mouth makes Cc Fic. 2. Gekko mindorensia sp. nov., X 2. a, chin; b, head, side view; ¢, anal region. a sharp angle upward immediately below eye; loreal regions covered with rather enlarged granules, largest immediately in front of eye; a row of enlarged granules bordering upper labials above; interorbital region and medial area on snout depressed, covered with much smaller granular scales; occipital region covered with minute granules interpersed with indistinct larger granules; mental triangular, followed by two elongate chin shields nearly three times as long as wide; a few enlarged pentagonal scales behind and at sides of chin shields; throat covered with equal-sized granules, larger than those on dorsal surface. Body rather slender, covered above with minute gran- . XIV, 1 Taylor: New or Rare Philippine Reptiles 117 ules, intermixed with numerous rounding granules; an in- distinct lateral fold; about 30 longitudinal rows of scales between the folds; preanal and femoral pores continuous, 27-28 on.each side, slightly separated medially, a total of 55 pores. Tail re- generated wholly; above covered with fine granules, with no trace of annulations, below with irregular large broad scales ; two enlarged tubercles at base of tail. Extremities moderate; toes not greatly dilated, about 16 lamelle under longest toe; no rudiment of web evident. Color in life-——Above almost uniform olive brown with but very little variation delineating the lighter and darker areas. No markings evident; below yellowish white. Measurments of Gekko mindorensis sp. nov. mm. Total length : 162 Tail, regenerated 80 Snout to foreleg 34.5 Snout to vent 82 Snout to ear opening : 22 Foreleg 30 Hind leg 40 Variation—Eight other specimens have been studied, taken by myself about Naujan Lake, Mindoro. These specimens are all young; they are darker than the type, with blackish stripes across the backs and on the tails. Two males in the lot have 62 and 64 pores, respectively. Remarks.—This species is related to Gekko monarchus, from which it differs as shown in the table of measurements. Sev- eral specimens of that species are at hand for comparison. Comparative measurements of Gekko mindorensis and G. monarchus. G. mindorensis. G. monarchus. Snout to vent (mm.) 82 85 Foreleg (mm.) 26 29 Hind leg (mm.) 35 40 Width of body (mm.) 16 18 Width of head (mm.) 18 20 Diameter of eye (mm.) 6.5 5 Diameter of auricular opening (mm.) 3.3 2 Interorbital measurement (mm.) 6.75 8 Preanal and femoral pores 54 33 Scale rows across belly 30 45 Two males of equal size are compared. The ear-opening of G. mindorensis is larger, equaling half the eye. The eye itself is much larger than that of G. monarchus; the interorbital distance is less; there is a much larger number of preanal and 118 The Philippine Journal of Science 1919 femoral pores, the number of scale rows across the belly is about fifteen less; the tubercles are fewer and much less prominent and the very characteristic markings of G. monarchus are wanting. The type specimen was captured with the assistance of Mr. Clark Burks, of Sumagui, Mindoro; two eggs were found at the same time. Oshima? has described Gekko kikuchii from Botel Tobago, Formosa, which seems to vary from Gekko monarchus in much the same way as does the present species. Although no speci- mens of the species are at hand for comparison the following differences are evident: Gekko mindorensis has many more femoral pores, 55 to 64, while 48 are recorded in G. kikuchii; the limbs are longer in the present species, and the basal web is wanting on the feet in both sexes. Siaphos kempi sp. nov. Fig. 3. Type.—No. 2016, E. H. Taylor collection; collected April 23, 1916, at Naujan Lake, Mindoro, by E. H. Taylor. Description of type.—Head short, snout blunt, not especially depressed; rostral covering tip of snout, the part visible above much less than the depth of snout at the posterior border of rostral; no supranasals; frontonasal large, the suture with rostral equal to, or smaller than, that with nasal, and slightly larger than that with frontal; prefrontals much reduced, separ- ated, not in contact with first supraocular; frontal moderate, about equal to frontoparietal in length, as long as its distance from end of snout, in contact with two supraoculars and first superciliary; frontoparietals fused in a single scale a little broader than long; interparietal triangular, a little wider than frontal, but shorter; parietals forming a suture behind inter-. parietal ; four pairs of nuchals; nasal pierced medially by nostril, touching a single labial; two frenals, both higher than wide; four supraoculars, last longest, second widest; eight superci- liaries; several much enlarged temporals; seven upper labials, fifth below center of eye; five or six lower labials; mental larger than rostral; ear opening obliquely oval, moderately large; eye nearer tip of snout than ear opening; limbs small, barely touching when adpressed; digits not or but slightly compressed ; 20 lamellze under fourth toe; two enlarged preanals; median row of scales under tail slightly widened; 22 scale rows around body ; scales smooth, the median dorsal rows widest. * Philip. Journ. Sci., Sec. D (1912) 7, 241. XIV, 1 Taylor: New or Rare Philippine Reptiles 119 Color in life-—Above silvery olive to brown, somewhat lighter on tail; laterally a broad dark stripe from eye to near end of tail, becoming somewhat lighter on tail, bordered above and below with a narrow greenish silver line; head uniform dark brown; belly dirty greenish white with a few small specks of dark color under chin; underpart of tail creamy white; limbs mottled brown and yellowish. Fic. 3. Siaphos kempi sp. nov., from the type, X 4. a, head, dorsal view; b, head, lateral view; c, chin. : : Measurements of Siaphos kempi sp. nov. - mm. Total length 93 Tail 53 Axilla to groin 21 Snout to vent 40 Snout to foreleg 14.5 Foreleg 9 Hind leg 14 Variations.—Two other specimens were taken in the same locality ; both agree with the type in markings save that both are lighter above; they are practically identical in scalation. Remarks.—This species appears to be intermediate between Leiolepisma and Siaphos. I have referred it to the latter rather than to the former genus because of its closer superficial resem- blance to the other Philippine members of that genus and the fewer prefrontals and shorter limbs; the ear opening, how- ever, is large and distinct. The known Philippine species of Leiolepisma have the divided frontoparietal; the three known species of Siaphos have the frontoparietal single. The species is named for Ollie C. Kemp, Mangyan agent in Mindoro, who accompanied me on the collecting trip to Naujan Lake and assisted in making collections. . The species was found living under the leaves of small, close- clinging vines on trees. When these vines were loosened at the base of the tree and torn down, the lizards were revealed on the 120 The Philippine Journal of Science 1919 sides of the trees at some distance from the ground and they immediately took refuge much higher up the tree; thus many specimens escaped. Only three specimens, brought down with the vines, were captured. This species differs markedly from the two small species Siaphos infralineolatum Peters and Siaphos quadrivitatum Peters in size, markings, and the presence of an auricular opening. From the recently described Siaphos auriculatum Taylor, which it resembles greatly in markings, it differs in the presence of small prefrontals, and it probably does not grow so large. Many other differences are obvious on a comparison of descriptions and figures. Known only from the type locality. Sphenomorphus lednickyi sp. nov. Fig. 4. Type.—No. R 1992, E. H. Taylor collection; collected in June, 1917, on Masbate Island, by Victor E. Lednicky. Description of type.—Rostral only slightly visible from above, forming a broad, rather curved suture with fronto- nasal; latter much broader than deep, broader behind than in front, in contact with anterior frenal; prefrontals large, broadly in contact, touching both frenals laterally, not in contact with first supra- ocular; frontal as wide as, or slightly less than, the supraocular region, in contact with first superciliary and two supra- oculars, narrowed behind; frontoparietals distinct, rather elongate, broadly in con- tact; interparietal much longer than wide, with a distinct eyespot; parietals forming “Fic. 4. Sphenomorphus leq. @ Suture behind interparietals; no nuchals nickyi sp. nov., No. 291, present; five supraoculars, last very small, e head, donal tives g first deepest, second widest; nostril head, lateral view. pierced in a single nasal; no supranasals; anterior frenal not as high as nasal, much higher than wide; second frenal wider at top than bottom, lower than anterior; three preoculars superimposed, between first superciliary and third labial; six upper labials, fourth and fifth under eye, but separated from orbit by several scales; third labial smaller than second; five lower labials; mental small, followed by a very wide postmental; three pairs of chin shields, only first pair in contact; four enlarged temporals, that bordering parietal largest; auri- cular opening very large; tympanum very superficial; 40 rows XIV,1 Taylor: New or Rare Philippine Reptiles 121 of scales around belly; scales on sides in longitudinal rows, all rows subequal in size; preanal scales distinctly enlarged; limbs well developed, the adpressed hind leg reaching wrist of foreleg; latter brought forward reaches middle of eye; 22 lamelle under ° longest toe; scales below tail scarcely broadened; eye a little nearer end of snout than ear; latter nearer foreleg than end of snout. Color.—Head and body variegated brown above with dark areas over supraocular region and a median row of irregular black dots, extending somewhat on tail; a heavy black stripe begins behind eye and continues as a broken series of irregular spots along side of body and tail above legs; legs brown, with black variegations and a light spot on knee; toes barred with black; throat and labials muddy white; belly light cream; tail below with dark flecks; a few cream yellow spots above and below the black stripe on neck. Measurements of Sphenomorphus lednickyi sp. nov. Type. Cotype. mm. mm. Total length 106 114 Snout to vent 46 50 Snout to foreleg 18 18.5 Tail 60 64 Axilla to groin 27 28 Width of head : 7 8 Length of head 8.6 | 8.5 Foreleg 15 14 Hind leg 21 22.5 Variation.—A second specimen from the same locality agrees in scalation, but the head is broader, and the broad black stripe is almost wanting along the body. The regenerated tail has broad scales above and below. Remarks.—The two specimens were collected in Masbate, near the Aroroy gold mines, by Mr. Victor E. Lednicky. He states that they appear to be plentiful in that locality. Superficially this species resembles Sphenomorphus curtirostris, but the latter has the frontoparietal single, fourteen lamelle under longest toe, more upper labials, and the nasal followed by superimposed fre- nals. It is a larger species. J take pleasure in naming the species for its discoverer. Sphenomorphus Ianosi sp. NOV. Plate II. Type.—Santo Tomas Museum, Manila, specimen unnumbered, collector unknown. Probably from Luzon. Description of type.—Habit lacertiform; snout narrower and 122 The Philippine Journal of Science i9ie longer than in Sphenomorphus jagori; rostral much wider than high, much narrowed laterally in front of first labial below nasal; frontonasal little broader than deep, forming a straight suture with rostral, laterally in contact with upper anterior frenal, posteriorly in contact with frontal; prefrontals rectan- gular, separated in the middle, in contact laterally with upper anterior and posterior frenals, and a single superciliary; fron- tal much elongate, narrowed behind and in front; distance between supraocular regions about one-third the width of the region above one eye; frontal little shorter than fronto- parietal and interparietal length, in contact with two supra- oculars; frontoparietals slender, pointed in front, forming a common suture much more than half their length; interparietal nearly as broad as frontal, but not nearly as long, as large as frontoparietals; parietals large, in contact behind interparie- tals, touching last supraocular and two small postoculars; no nuchals; nasal quite large, the scale placed diagonally, touching only one labial; two anterior frenals superimposed, the two not as high as nasal, but higher than the very large frenal following, which is much wider at top than bottom, touching two labials; two preoculars, the lower largest, followed by two moderately large scales below eye, the second partly wedged between fifth and sixth labials; ten superciliaries; five large supraoculars followed by two paired scales, the first supraocular more than a third of the length of the supraocular region, second widest; nine upper labials, sixth below orbit, sixth and seventh largest; five distinctly enlarged temporals, the largest bordering parietal: eight or nine lower labials; one unpaired postmental; two paired chin shields, first pair in contact, second separated by one scale; anals enlarged ; 23 lamelle under longest toe; ear opening mode- rate, vertically oval; 42 scale rows around middle of body. Ad- pressed hind leg reaches beyond elbow of adpressed foreleg; adpressed foreleg reaches anterior corner of eye. Color in alcohol.—Above rich, reddish brown traversed by about thirteen indistinct bands of light, black-edged dots on back, and continuing on tail; behind eye to above foreleg is a large, broad, dark brown stripe bordered below by a white stripe, which begins on second labial and continues back to ear, then widens and takes a backward and downward course to foreleg; below this there is a rather irregular mottled area of brown; lower labial flecked with brown; limbs above mottled with elongate light spots ; sides somewhat lighter, flecked with brown; a brown XIV,1 Taylor: New or Rare Philippine Reptiles 123 area above hind limb; below yellowish; there is a very dim row of brown spots along upper lateral region. Measurements of Sphenomorphus llanosi sp. nov. mm. Total length 148 Tail (end regenerated) 60 Foreleg 16 Hind leg 26 Axilla to groin 27.5 Snout to foreleg 21.5 : Snout to eye 5 Eye to ear 5.2 Ear to foreleg 8.5 Width of head 9 Remarks.—Closely related to S. jagori, but differs in the fol- lowing points: The first frenal is divided and the scales super- imposed; the second frenal is proportionally much larger ; there are five instead of four large supraoculars; the frontoparietals are slenderer and extend farther forward; the coloration is also distinctive; there are more rows of scales about the body. Unfortunately the collections in the Santo Tomas Museum are unnumbered, and for the most part are without authentic local- ities. It is highly probable that this species is from Luzon, and perhaps the northern part of the island. Only a single specimen, the type, is present in the collection. I take pleasure in naming this species in honor of Father Florencio Llanos, director of the University of Santo Tomas, who has generously permitted me to study the collections in the museum of the university and to describe the new species. ILUSTRATIONS PLATE I Fic. 1. Typhlops longicauda sp. nov., from the type, xX 5. b, dorsal view; c, ventral view. 2. a, Hemidactylus frenatus Duméril and Bibron; 6b and c, Hemi- dactylus luzonensis Taylor. PLATE II a, lateral view; Sphenomorphus llanosi sp. nov., from the type, about x 1.5. TEXT FIGURES Fic. 1. Trimeresurus mcgregori sp. nov., from the type, x1. a, head, lateral view; 6, head, dorsal view; c, chin. 2. Gekko mindorensis sp. nov., X 2, a, chin; 6, head, lateral view; c, anal region. 8. Siaphos kempi sp. nov., from the type, x 4. a, head, dorsal view; b, head, lateral view; c, chin. 4. Sphenomorphus lednickyi sp. nov., X 3. head, lateral view. 125 a, head, dorsal view; }, [Puiir. Journ. Sct., XIV, No. 1. TAYLOR: NEw or RARE PHILIPPINE REPTILES. } « Pn ae . 6, Saen one Hemidactylus luzonensis Taylor. Fig. 2. a, Hemidactylus frenatus Duméril and Bibron; b and Cc, Fig. 1. Typhlops longicauda sp. nov., from the type, x 5. PLATE I. TAYLOR: NEW oR RARE PHILIPPINE REPTILES. ] ; [PuHILip. Journ. Scr., XIV, No. 1. PLATE Il. SPHENOMORPHUS LLANOSI SP. NOV., FROM THE TYPE, ABOUT x 1.5. IPON FISHERIES OF ABRA RIVER By Epwarp H. Taytor (From ‘the Section of Ichthyology, Biological Laboratory, Bureau of Science, Manila) On the western coast of Luzon as well as in numerous other parts of the Philippines there are several annual runs of the young of a small fish of the family Gobiide, known to the Ilocanos as ipon or hipon.: At stated seasons they appear along the shore and attempt the ascent of the rivers, where they attain adult size. The adults reach a length of from 8 to 10 centimeters; the young when taken are usually less than 2 centimeters in length. In spite of the very small size of these fishes the catch in the Philippines is so large that the industry . is of marked economic importance. The ipon fisheries of Abra River, which are typical of the ipon fisheries of other places along the Luzon coast, are prob- ably the best illustration of this important industry. Abra River rises in the highlands dividing Kalinga Subprov- ince, Mountain Province, from Abra Subprovince, Ilocos Sur. It is fed by several fairly large streams, which carry much water during the rainy season. It emerges from the mountainous pla- teau of Abra through a narrow gap in the coast range about 12 kilometers from the sea. Here on the coastal plain it spreads out in fan shape, to lakelike dimensions, entering the sea through several mouths. In its upper reaches it probably attains con- siderable depths, since several soundings of more than 10 meters’ depth were made. However, it frequently descends over long stretches of gravel and small bowlders, with a depth of from 1 to 2 meters. During the dry season the depth at the various mouths is slight, probably averaging about a meter. The pro- vincial road crosses the river very near the mountain gap. Here the river is about 0.4 kilometer wide, and the depth does not exceed 2.5 meters during December. The ipon of Abra River that were taken during November were the young of a small goby which when full grown does not attain a length of more than 10 centimeters. During the greater 1In many of the Filipino dialects the names ipon and hipon are applied to small shrimps or prawns, species of which are found in both fresh and salt water. 1623289 127 128 The Philippine Journal of Science 1919 part of the year it is found in the fresh water of lakes and rivers. At certain seasons it descends to the sea to spawn, probably about the month of June, for the young ipon make their first appearance along the coast during September, rarely in August. . According to substantiated report the young arrive near the coast nine days after the first full moon in each month from September until February. Fishing begins along the coast as soon as the fish are sighted, in water from 2 to 10 meters deep. The sea catch is of superior quality and brings the highest prices. However, shore fishing is profitable for only three days, because during these three days the fish have entered the river and may be caught by nets and traps. The river fishing lasts several days after the shore fishing has ceased to be profitable. The value of the fresh-water catch quickly diminishes. The fresh water has a tendency to turn the living fish dark, and as the flesh darkens it acquires a somewhat bitter taste. The fish when full grown has a fair flavor and when fried crisp is quite palatable. After a few days the river catch has lost entirely its commercial value, and the fishermen await the next monthly run. The number of adults captured for food is negligible. The two most important methods of taking the fish are by means of a special net, known as daclis de ipon, and a small trap, known locally as sarep. The former is the more important. It is a large affair, averaging 60 to 100 meters in length and 8 to 12 meters in.depth. It is composed of many pieces of abacaé or maguey cloth sewn together lengthwise and resembles a huge piece of patchwork, varying in shade from yellow to brown. Brails are attached at the ends; the tops are well corked; the bottoms are leaded heavily, and the leads are concealed by several long ropes running lengthwise and securely fastened together. These ropes seem to prevent the bottom of the net from becoming readily entangled with objects on the sea bottom. The net is used near the river mouth in water as deep as 12 or 14 meters. During a run of the ipon, a good haul will often bring in as many as 10 cavans;* the usual haul is less. From January 1 to November 1, 1916, the town of Caoayan lying at the mouth of the river, licensed one hundred thirty-six of these nets, daclis de ipon (or chinchorros para ipon, as desig- nated on the tax ordinance). Each net of this type is taxed annually 10 pesos,* thereby bringing to the tax income of the * One cavan equals 75 liters. *One peso Philippine currency equals 50 cents United States currency. XIV, 1 Taylor: Ipon Fisheries of Abra River 129 town 1,360 pesos per annum. Each net over 60 meters in length has a marketable value of 300 pesos or more; thus the nets represent a property value of at least 40,800 pesos. The sarep is a small cylindrical trap made of finely woven bamboo and is about 1 meter long by 25 centimeters in diameter; at one end it tapers to a narrow cyclindrical mouth stopped by a cork about 5 centimeters in diameter. The other end is furnished with a funnel-shaped mouth, which extends within for about one-third of the length. These traps are set in the river near the various mouths where the water is shallow or where there are riffles, as many as two hundred being placed side by side in certain designated places. The places where sarep may be used are fixed and classified by the municipal councils. These places are designated as “first class” and “sec- ond class.” For a first-class location the minimum tax is 12 pesos per annum; but these places are usually sold at auction, the best localities bringing as much as 40 pesos each. The minimum for the second-class locations is 6 pesos per annum. The auctioning of locations is frequently the occasion of much dissension among the fishermen, at times resulting in fights. The catch for the most part is made into the popular fish food, known as bagoong.! To make bagoong the fish are laid in layers in large earthen jars called tinajas, a layer of salt between two layers of fish. This mixture is then stored away and allowed to ferment into an evil-smelling substance, which is greatly relished by many Filipinos. * The liquid portion obtained from the mixture is used for a sauce and is usually eaten with rice, while the more solid part of the ferment is heated or fried and eaten as ordinary fish. The proportion of salt and fish varies; according to Seale* the proportion is two parts of fish to three parts of salt for certain varieties of bagoong. This is certainly a much higher percentage of salt than is used or required for the bagoong made of ipon. : The fisheries at the mouth of the river are in the hands of the’ two municipalities of Santa and Caoayan. In the two towns the taxes on the fishing industry alone amounted to 3,224.39 pesos for 1915, and the collections for 1916 up to December 1 220.15 pesos. Not all of this represents tax on amounted to 3,2 ’ the ipon fishery, but certainly a large part of it does, since a number of preparations made oes ic term applied to ; Bagoong is a gener ded guinamos in Visayan from small fish or shrimp. It is usually known as dialects. © Seale, Alvin,’ Philip. Journ. Sci., See. D (1914), 9, 4. ’ 180 The Philippine Journal of Science according to a previous paragraph the taxes on the daclis de ipon in the town of Caoayan alone, not including the sarep localities, amounted to at least 1,360 pesos. A conservative estimate of the value of the ipon fisheries of Abra River based on the average of the catches of individual fishermen, on the estimates of various officials and traders, and on the taxes for the nets and traps for catching ipon, amounts to well over 350,000 pesos annually. Various estimates made by the municipal treasurers and secretaries were invariably between 400,000 and 500,000 pesos. These last estimates are too large, save during exceptional years, as the size of the catch varies from year to year. To use this fish for canning would be feasible. Without doubt such a canned product would find a ready sale in interior towns. Possibly it might take the form of a paste such as the anchovy paste now found in European and American markets, or the ipon might be preserved in a piquant sauce and used as a relish with curry or preserved in oil and used the same as salmon and other fish products. Appended is a statement obtained from the Provincial Treas- urer of Ilocos Sur of tax receipts for towns along Abra River. TABLE I.—Statement of collections on fisheries of the municipalities along Abra River. Municipality. ; 1914 | 1915 | 1916 ee Pesos. Pesos. Pesos. Bopmonrriacd,, A ir a a es 21. 55 21.75 17.05 | Nov. 30. Bucay; Adres 62. 56 70, 20 43.00 | Oct. 31. Caonyah, Tloces Sur: io cae 1, 908.90 | 2,087.16 | 2,833.31 ; Nov. 30. Dolores, ‘A bra.5 0553555 5S ea es os 6.91 14, 88 7.13 | Sept. 30. Piotiojan, Abe oo oie ee 10. 00 10. 00 11.00 | Oct. 31. Pilar, ADra oe ee 33. 60 14.60 |_._...._..| Sept. 30. Santa, Ilocos Sur -.----...------2.2--- 2-5, SES Cees 1,048.49 | 1, 187.28 986.04 | Nov. 30. Vayu, Abra a2 ios Sa aa 44,00 62. 09 53.00 | Nov. 30. Two towns of Iloccs Sur at mouth of river___._._...._.| 2,957.89 | 8,224.39 | 2,220.15 | 99,396.93 Six towns on upper river in Abra ..._.____.___________. 177.62 193. 13 181.18 | 501.93 ® Total for three years. PHYTOPHTHORA FABERI MAUBL.: THE CAUSE OF COCONUT BUD ROT IN THE PHILIPPINES By Otto A. REINKING (From the Agricultural Experiment Station, College of Agriculture, Los Baiios) ; THREE PLATES The cause, and the method of spread, of coconut bud rot, which produces severe losses in all parts of the tropics where coconuts are cultivated have been subjects of investigation. Va- rious organisms have been considered responsible for the disease. Because of this diversity of opinion, extensive work has been done by the writer to determine, if possible, the organism re- sponsible for the disease in the Philippine Islands. Although these studies are still in progress, sufficient evidence has been obtained to make it seem advisable to present in a preliminary paper the most important conclusions that have been reached. Before adequate control measures can be devised to reduce the damage done by this disease, it is of the utmost importance that the causal organism and its methods of dissemination be definitely known. The work already done has been sufficient to establish the identity of the organism responsible for the disease in the Philippine Islands. HISTORY OF THE DISEASE IN THE PHILIPPINES _In March, 1908, the first authentic and reliable investigations into the bud-rot situation in the Philippines were conducted by Copeland.(5) The disease was reported as prevalent in the coconut sections of Laguna and Tayabas Provinces, being es- pecially severe in the regions surrounding Lilio and Nagcarlan; and some cases were observed on the mountain slopes near the town of Tayabas. In the vicinity of Nagearlan, it had been reported as being serious for a period of ten years preceding these studies. Due to these investigations ordinances were enacted in Laguna Province whereby all the infected trees were to be cut down and burned, in order to check the advance. In the latter part of the same year Byars,(3) of the Philippine Bureau of Agriculture, visited the affected regions for “the in- 131 182 The Philippine Journal of Science 1919 spection of these provinces with a view to eradicating the bac- terial disease known as ‘bud-rot of the coconut.” On this trip Byars found the disease to be most prevalent around the base of Mount Banahao in the municipalities of Nagcarlan, Lilio, and San Pablo. It was reported that more than two thousand trees had been destroyed in one of the barrios of Nagcarlan. In March, 1911, Roxas(11) in a report on the cultivation of coconut mentions the bud rot as being probably due to bacteria. Cevallos,(4) in 1911, discussed control measures for the bud rot that had been observed some years ago. In 1912 Barrett(1) wrote of the infection as being by far the most serious of all the fungus or bacterial diseases that trouble the coconut planter. In 1916 the first extensive means for combating the bud rot were inaugurated by the Bureau of Agriculture under the di- rection of Mackie.(8) The work “was designed to furnish in- formation as to the spread and the prevalence of disease and, if possible, to ascertain the pathogenic agent, and the means of distribution.” The inspections were confined largely to the coconut belt in central Luzon, in Laguna and Tayabas Provinces. New ordinances were passed, and the Bureau of Agriculture was charged with inspection and destruction of infected trees. Little was learned, however, concerning the cause and manner of its spread. A bacterium was reported to have been isolated from affected trees, and from palm weevils that attack the trees, and an attempt was made to prove that the insects carried the or- ganism and spread the disease. No inoculation studies were performed. Wester, (12) in a paper on coconuts, cited Reinking, of the College of Agriculture, as having isolated a bacterium which produced disease when inoculated into healthy trees. The latter, in a report on Philippine economic-plant diseases, (9) under the discussion of bud rot, stated that a diseased condition could be produced by a specific bacterium. From this resumé of the literature in the Philippine Islands it may be seen that very little work has been reported on the causal organism, and that the general opinion has been that a bacterium was the cause of the disease. Johnston,(7) after an extensive piece of work on coconut bud rot, arrived at the conclusion that the disease in the West Indies was caused by a bacterium similar in most respects to Bacillus coli (Escherich) Migula. Butler(2) has presented evidence to prove that the disease on palms in India was produced by Pythium palmivorum Butler. In other sections of the world various fungi have been associated with the disease. For the past two years the writer has made a study of the XIV, 1 Reinking: Coconut Bud Rot in the Philippines 183 rot with the aim of determining the causal organism. Until the present investigations were started, no accurate and definite study was made of the organism which causes the coconut bud rot in the Philippines. The report as here presented is only a preliminary one; a complete work on the bacterial and fungus phases will be prepared at a later date. Preliminary tests indi- cated that the disease might be due entirely to bacteria; but more extensive studies proved that bacteria could not account for the severity of the disease observed under field conditions. On March 11, 1919, from a diseased tree kindly sent in by the Bureau of Agriculture, a fungus was isolated which was proved to produce disease in all seedlings inoculated. This work, taken in connection with inoculation studies with Phytophthora faberi Maubl. isolated from cacao pods, appears to the writer to furnish conclusive evidence that the primary agent in the production of the infection is a fungus; and that bacteria, while always associated with severe cases of the disease, are to be regarded as secondary agents. DISTRIBUTION OF THE DISEASE In the Philippines, according to the Bureau of Agriculture reports for 1916 by Mackie, (8) the disease was found in Laguna, Tayabas, Batangas, Pangasinan, Tarlac, Albay, and Ambos Ca- marines Provinces; and it was reported as occurring in Misamis, Capiz, Samar, and Bohol. The report of the Bureau of Agri- culture for the year 1917-18, submitted by Mr. G. Merino, states that the disease is also prevalent in Zamboanga. The infection is most severe and widespread in those sections of the Islands in which the climatic conditions afford a very moist at- mosphere. Thus coconut bud rot is very abundant in the upper bout Mount Banahao. The studies of the world has been given by Copeland(6) and by Johnston. (7) NATURE OF THE DISEASE General diagnosis.—The first symptom is a withering of the youngest daetded leaf, followed by the leaf turning brown. Gradually the next younger leaves wither and turn brown, until the entire central group is affected. At this stage the central leaves may be easily pulled out. Frequently, in advanced Soap they fall over (Plate I, figs. 1, 2, and 3). About this ree central portion is a fringe of older leaves, which are pe ect y healthy and remain upon the tree for months after infection. 184 The Philippine Journal of Science 1919 It appears that trees are most commonly infected when they first come into bearing. The young nuts on a bearing tree attacked remain small and fall off prematurely. Internal symptoms are very characteristic. The fungus ap- parently gains entrance into the soft tissue, usually called the “cabbage,” through the youngest leaves. In the early stages a longitudinal section of the bud shows that the disease may start in young leaves, at a point where they begin to unfold (Plate II, fig. 1). At this point a spotting of the leaf is first noticed; then the organism works downward, causing a soft rot and browning of the group of unfolded leaves. The upper exposed portions of these die and turn brown, due to the rotting beneath. The rot advances downward to the growing point and then spreads into the soft tissue below. From here it invades the ' woody parts, usually not penetrating farther than from 5 to 10 centimeters. In the early stages no discoloration is produced in the growing point and “cabbage,” but a dark red to brownish line always marks the limits of the advance in the woody parts (Plate II, fig. 4). The organism does not penetrate readily into the old leaf sheaths surrounding the young, tender, developing portions (Plate III, figs. 1, 2, and 3). The rot is commonly checked when it reaches the firmer tissues of the trunk though, in advanced cases, it may penetrate about 20 centimeters (Plate Ill, figs. 1 and 2). The affected portion in the trunk may be- come greatly softened; this is shown by the fact that the finger can be pushed into the diseased part. There is present a putrid, somewhat sour odor. The most advanced stages are charac- terized by the change of the white “cabbage” into an ill-smelling, semiliquid mass. A portion of the trunk below becomes a softened group of fibers. Spread and loss.—The disease spreads very rapidly from tree to tree, but the manner in which this is accomplished has not _ been thoroughly investigated. Wind and insects are probably the most important agencies. In one barrio under observation, fifty-eight new infections appeared within one year after an inspection in which all attacked trees found had been cut down and burned. The disease may have started from trees unob- served during the first examination. The trees were located in the upper extremities of the coconut region on the slopes of Mount Banahao, where the atmosphere is very moist and where thick planting is practiced. Both conditions are favorable for the development and spread of the organism. Up to September 30, 1918, according to reports of the Bureau of Agriculture furnished the writer by Mr. José Sanvictores, XIV, 1 Reinking: Coconut Bud Rot in the Philippines 135 12,813 coconut trees had been found affected in Laguna, Taya- bas, Pangasinan, and Zamboanga Provinces. The largest num- ber was found in the first two provinces. Since each tree is valued at not less than 10 pesos, the entire loss, not including early losses and the cost of inspection and burning of affected coconut trees, has been 128,130 pesos. The number of trees destroyed will undoubtedly be lessened by a thorough system of inspection and eradication of diseased trees. Future examina- tions ought to show a lower proportion of infection. INVESTIGATION OF THE DISEASE FIELD STUDIES Careful study has been made of more than thirty typical cases in the field. These diseased trees were cut down and the buds opened for observation. Judging from macroscopic examina- tion the infection appeared to be due to bacteria. Portions of infected trees were collected and placed in sterile vials for trans- porting to the laboratory. In addition to these studies, work -was done with diseased specimens sent to the College of Agri- culture by the Bureau of Agriculture. LABORATORY STUDIES Bacterial.—During the first part of the investigation micro- scopic examinations, in the majority of cases, failed to reveal the presence of mycelia. Bacteria, however, were always present in abundance. Diseased pieces, which had been collected under sterile conditions in the field and placed immediately in sterile vials, usually developed no fungi; but they were completely over- grown with bacteria. Later studies have indicated that, in these cases, the bacteria multiplied rapidly and destroyed the slower- growing fungus that probably was present. The only fungi observed in the earlier studies were saprophytic forms. Cultures were obtained by cutting and plating small pieces from all parts of infected trees, specimens being taken from the tip of the unfolded infected leaves down to the growing point and the woody tissue below. A mixed culture of bacteria was present in the majority of these cases. In very young stages of infection, and in tissues into ‘which infection has ad- vanced farthest, only one kind of bacterium may be present. Such cases were usually not obtained, probably because sapro- phytic bacteria are soon washed down into the infected parts and there find a favorable place for development. Since no specific fungus was isolated in the first studies, and bacteria 136 The Philippine Journal of Science 1919 were always present, inoculation experiments were conducted with them in order to test their virulence. . The inoculations were made chiefly with seedling coconuts from 60 to 180 centimeters in height. They were prepared by stripping off the outermost leaves. Then the outside of the portion to be inoculated was washed with mercuric chloride, 1 to 1,000, and after ten minutes was rinsed with sterilized water. With a sterile scalpel a stab was made into the growing point, and a pure culture of the bacterium introduced with a platinum needle. The injured portion was then covered with melted paraffin. Controls were prepared in the same manner except that ‘sterile water was used in place of the bacteria. About three hundred inoculations have been made in this manner, and in a number of instances cases of bud-rot were produced (Plate II, figs. 5 and 6). Employing this method, saprophytic bacteria were eliminated, because of their failure to produce disease. After the elimination of these saprophytic forms, there remained one distinct kind of bacterium that would produce disease under certain favorable conditions, such as severe injury of the grow- ing point, and excessive dampness. Inoculations were not re- peatedly successful, but when the exact requirements were met a number of positive infections could be obtained. Bacteriolo- gical tests have shown that the organism thus isolated and used for inoculations was similar to Bacillus coli (Escherich) Mig., and apparently is identical with that considered by Johnston(7) to be the cause of coconut bud rot. Authenticated cultures of Bacillus coli (Escherich) Mig. were obtained from Dr. L. R. Jones, of the University of Wisconsin, and also from Dr. O. Schébl, of the Philippine Bureau of Science. The cultures from the United States were isolated from man, and those secured in the Philippines, from man and _ horse. Under extremely favorable conditions a bud rot could be pro- duced with each of these cultures. The infections produced from the first inoculations were very slight; but the organism, after it was reisolated and then used for reinoculation, appeared to produce a more rapid and severe case of rot. This may indicate that the bacteria increase somewhat in virulence after passing through a weakened tissue. Cytological studies also have shown the presence of bacteria in the tissues of the coconut. Sections of a typical case were made from parts of the young leaves near the growing point, from portions of the growing point, and from the woody tissue. These sections show that the organism is present not only in the parenchymatous tissue, but also in the vascular system. XIV, 1 Reinking: Coconut Bud Rot in the Philippines 137 The bacteria are present in the xylem tubes of the young leaves, and of all the parts down to the woody tissue. This probably accounts for the rapid spread of the bacteria. Thus bacterial studies have proved that under certain very favorable conditions Bacillus coli (Escherich) Mig., and a simi- lar organism isolated from coconut trees, may produce disease. A summary of the entire bacterial investigations has indicated that, although the bacteria are always present and account for the destruction of a portion of the weakened tissues, they can not explain the prevalence and rapid spread of the disease. Further researches have been conducted in order to ascertain whether or not a fungus is present and is the primary agent in the production of the disease. Phytophthora faberi Maubl. studies.—While the investigations of coconut bud rot were in progress, studies were also being conducted with Phytophthora faberi Maubl. isolated from cacao. The latter organism proved not only to cause the black rot of cacao pods, the rot of papaya fruit, and the canker of cacao and Hevea rubber, but also to produce an infection in a number of other hosts as well. Table I gives a summary of these inoculations. TABLE I.—Infection experiments with Phytophthora faberi Maubl. isolated from the black rot of cacao pods. Seedlings killed by the organism: Annona muricata L. Guanabano; soursop. Hevea brasiliensis (H. B. K.) Muell. et Arg. Para rubber. Spondias lutea L. — Theobroma cacao L. Cacao. | Infection confined to wounds: Annona glabra L. Artocarpus odoratissima Blanco. Marang. Artocarpus integra (Raderm.) Merr. Jack fruit. * Averrhoa bilimbi L. Camias. Eugenia jambolana Lam. Duhat; lumboy. Lansium domesticum Jack. Lansones. Fruit rot produced: Carica papaya L. Papaya. Theobroma cacao L: Cacao. it of this fungus, and since Because of the omnivorous hab : A ‘og in the East, notably India, have reported various countries in duced by a Phycomy: cete, that coconut bud rot has been pro , Pythium palmivorum Butler, (2) it was deemed. advisable to try inoculations with Phytophthora faberi Maubl. isolated from ca- cao pods. As is shown by the inoculation experiments foam in Table II, a large percentage of the trees became disease cal 1388 The Philippine Journal of Science 1919 TABLE II.—Coconut inoculation experiments with Phytophthora faberi Maubl. from cacao. Observations. Tree.; Date. Character of inoculation. 5 = phe aieg No rot. 1918. : 1918. 1918, 2 Oct 4 | Bint; damp conmpet. 6s oe ef Ons NS PRES y oseas + Jee Ase i ee Get Ib Bag ucla RS SS so ee es aes + B4555500 ch ONS ccktebven im aa chan ddawet cuonetssatae ate 6 ccc sees Bi NOW bef Cy (REEVE Je SEO a EE on in SSO SD Novy. 22 bs 22 3 + “6° Aug 16 | Stab: outside in shade2.-6c She WJOUE de Totacccee 7 | Aug. 26 |_____ 5 eae popes tea RSIS aS Ae pl Soy Ree eae Sept. 20 tosis: ae 8 | Nov. 18 |____- NU ae ae ee ae ae Wee A eas - 9 | Dec. 10 |. 0020.55 ee aR Sey ei Sy RR ee Dee. 26} ccenctece + a0 T5500 255 S25 GOSS ree eeu boc et SOO coh ew es > : a11 | Nov. 18 |_____ Co eae aa a Sep Ra UE NS ise Pee este 66:16 te: € 1919 1919. ; 12 | Jan. 18 |_____ SEES SERRA oer SETS CE 8 en + ; BS eet petrs eaten Os SSE Sa, Sue See ese cS Sree SA tise ccxeece + 3 1919, : Bg Rene. Son as ee 2 RRO Ss Sateen kL eRe cy eee UNE Pea iene srs gan. 16) is 1918. 1918. 15 | Nov. 20 |____. GO oe oe oss ages oe Nov--06 {25 cs. a 16'}s. do 222} 23 WOrcGi sec oni ete oe es SSeOO sebtee eras 2 1918, SR bade Dh Lote eames Sater Sek eile make aM ORS Spee s PONENTS Oe meee ES Nov. 26 18 | Nov. 22 |____. 4022555 ane aR CG ewe Sos Sos a Dec. 10 |___- bie Gore» 1 pak eee 2 URE EE NEP Ste San Ri ener OER CE I Deen Tee + Ls RSS “peepee Beate Sea tao er eg oe a ere es De PCS Dec. 10 O23) 00 OO acc sccs See Nee ee a a DOG Ae ise scc 1919. i 1919. 22) Jan: 2b GO regu ea sa eh ea ae ee aon 28 aa + ye Senge» (cases, Sear OOS esto ees eee ae jaw cowhesiadaubestsnes cousce 1919. O26 1.2186 cot Osc ee ean 20 fe 25 | Jan. 8}._... Diewcu i] 2, AL ae ge bud — & I 3 ey 8 3 ° 5 o 3 pA Rpt aces OY April7, 1919_| Severe rot. Le ee 2 12 nen EPS ey AG eee i aera ee ee ey G0 ac None. 1 | April3, 1919_| Injury, damp chamber -__.-._.-..-...-._-___- April 7, 1919_| Severe infection. YB Bee cn Lpeagen ae RE nes Siiceg kc TSCA ARES Etat ie SSSR eerste eae ce Tr aebente Te Do. 2 Beamon do __ CLP SER ee) aN Aine Ae, ee BY: 7 gaseen Do. , a Se do _. ON ees ee eas eens 6 feos Slight infection. gc 2 eres» pesca aioe Aeneas eo Sap eager Pe hE RC EON Rt OG 2255 No infection. ® Control. The inoculations clearly show that the fungus will cause a rot of cacao and papaya fruit, and a severe infection and the death of Hevea rubber seedlings. Since the same hosts are attacked by the strain of Phytophthora faberi Maubl. isolated from cacao fruits it can be stated, from the evidence furnished by cultural studies, that the organism from cacao is identical in this respect with the strain of Phytophthora isolated from coconut bud rot. MORPHOLOGY OF PHYTOPHTHORA FABERI MAUBL. ISOLATED FROM COCONUT GROWTH ON VARIOUS MEDIA The fungus grows well on potato agar, oat meal agar, corn meal, and macerated young corn. A dense white, more or less felty growth, is produced on each. In the small number of tests thus far conducted the spore formation appeared to be slightly different on each medium. Chlamydospores are produced in - XOV,1 Reinking: Coconut Bud Rot in the Philippines 1438 abundance by young cultures on oat meal agar, and conidia are only sparingly formed. In older cultures more conidia are produced. On corn meal the reverse was true in the prelimi- nary tests. These tests also indicated that on potato agar and on macerated corn the chlamydospores appeared before the conidia. MYCELIUM The mycelium is white, producing a dense mass in pure cul- ture. In young cultures it is nonseptate and granular. In older ones septa may be produced. Protoplasmic streaming is common. The submerged mycelium is more or less gnarled, while the aérial is straight. Branching is abundant. The width varies from about 3 to about 8 microns. CONIDIOPHORES Conodiophores are produced in great abundance in pure cul- ture, especially on corn meal. Special culture methods must be employed in order to show them in their best condition. Mate- rial showing excellent conidiophore production may be obtained in sterilized Van Tieghem cells, by placing a few spores on a film of agar that has been put on the flamed cover slip. Each coni- diophore may bear from one to fifteen or more conidia. A coni- dium is produced at the tip of a conidiophore; the latter then continues its growth by pushing the conidium to one side, and produces another conidium at its tip. By a continuation of this process a bunch of spores is finally formed. The conidiophores vary in size, ranging approximately from 180 to 645 microns in-length and from 3 to 6 microns in width. CONIDIA The conidia are produced terminally as described above. They are elliptic to ovate, and possess very prominent raised terminal papille. These spores are pale yellow to colorless, and densely granular, usually having a large vacuole. Older conidia possess more granules that are in groups. Measurements of conidia.—The measurements conform with those of Phytophthora faberi Maubl. as determined by Rosen- baum.(10) Measurements were made from two-day-old cultures on corn meal. The lengths and widths of two hundred spores were measured. The results are presented in Table V, which gives the class in microns and the number of conidia out of two hundred, both for length and width, that falls into each class. 1623828310 - * The Philippine Journal of Science 1919 TABLE V.—Summary of measurements of conidia of Phytophthora faberi Maubl. from coconuts. [Corn meal cultures; age 2 days.] Class. Conidia in each class. Length. Width. tite Se a ise ee a eee Bo Se ee ee i a nen ee We th 6c es wit ctbuccusdascemeeenies 6 ee oe ae ee ei ewSti.. Senos eer ee OF 6.5986 ae aa ee 93:6 0b Oi ea re ea ee 66-0676. 00S ere 6 00.6. a ee a. 00.6 0044-40. 6x). once ec sen ee oe We ir A es certs a as ieee 47.5 to 49, 49... 21.2.2 -22----- +--+ +4-- +--+ -------------------+-------- 2 -+---- +--+ 49.610 51,49... ee in sh eh eee one oe ede ee SY.5 66 68. 40 2 on a eee aa ene ae eee cnc nee-u- ee OO; 6 05 Ob: 40s os ae core epee Row de ohh sn as: Saas each eeeeuan coeeues iE OE A8 ss se sarss a aet a o oy cans cae cane eee O70 te O00 oo inks pede ntacdee suse sawp des abies bias Sane wena e ue pipes GOB 06 TAOS oe cde ~ Sern stent wk Soe eee eeasleul hana pie dew andlee eran _ nw m™ pb 3 on anem een oandB ao BP EReEBewoartnooond 8 Scooocosco eo sos occ oa nm BNBBEw From Table V it can be readily seen that the conidia vary in length from 27.5 to 69.49 microns, and in width from 21.5 to 41.49 microns. In length the majority of the spores falls into ~ the classes between 39.5 and 53.49 microns, the largest number falling in classes between 43.5 and 53.49 microns. In width the majority falls into classes between 27.5 and 37.49 microns, the largest number being in class 31.5 to 33.49 microns. Table VI presents the arrangement in classes of the ratios of the length to the width of the conidia. XIV, 1 Reinking: Coconut Bud Rot in the Philippines 145 TABLE VI.—Arrangement in classes of the ratio of the length to the width of the conidia, showing the limits of variation. [Corn meal] cultures; age 2 days.] Class. on class. 85 to 0.94 0 95 to 1.04 1 1.05 to 1.14 6 1.15 to 1.24 11 1.25 to 1.84 36 1.85 to 1.44 40 1.45 to 1.54 28 1.55 to 1.64 24 1.65 to 1.74 23 1.75 to 1.84 il 1.85 to 1.94 13 1.95 to 2.04 5 2.05 to 2.14 1 2.15 to 2.24 0 2.25 to 2.84 1 2.35 to 2.44 0 Total 200 The class of ratio values into which the greatest number of conidia fell was 1.35 to 1.44. The mean ratio of length to width would thus have an approximate value of 1.4. These figures correspond closely with those obtained for Phytophthora fabert Maubl. by Rosenbaum. (10) Germination of the conidia.—Germination takes place by the production of either germ tubes or swarm spores. Every coni- dium is potentially a sporangium; its method of germination is influenced greatly by its environment. Germination by germ tubes is by far the commoner method under cultural conditions. From one to five germ tubes may be produced; these apparently may develop from any part of the surface of the conidium. Up to the present time, no swarm-spore formation has been ob- served in the cultures obtained from coconut; but in those iso- lated from cacao germination by this method was frequently produced. During the first trials it was found impossible to obtain germination by swarm spores. In the months of Feb- ruary and March, at which time the nights are cool, swarm spores were produced abundantly in old cultures on macerated 146 The Philippine Journal of Science 1919 corn. Swarm spores were readily obtained in Van Tieghem cells by growing the organism on very dilute agar or in hang- ing drops of water placed on sterile cover slips. Just before formation, there appears a rearrangement of the protoplas- mic granules. The swarm spores are then produced within the sporangium. The papilla finally breaks off, and the spores escape. No vesicle formation has been observed. The spores near the opening escape one by one. Each rests for the period of a moment on the outside, and then swims off rapidly. As goon as a few spores have escaped, those remaining swim about actively within the sporangium and escape one at a time. The swarm spores are kidney-shaped. They swim about for a time by means of two flagella and then come to rest. At this stage they become spherical and may germinate after a few hours by the production of a germ tube. CHLAMYDOSPORES Chlamydospores are produced directly from the mycelium, usually terminally, but sometimes intercalarily. They are spher- ical, granular like the conidia, but with a slightly deeper yellow shade. In pure cultures they are produced in great abundance on oat meal or potato agar. Measurements of chlamydospores.—The size of the chlamy- dospores is one of the criteria by which the species are separated. Two hundred measurements were made of the spores from cul- tures 3 to 10 days old on oat meal agar, and one hundred mea- surements of spores from cultures 6 and 7 days old on macerated corn. A summary of these measurements, grouped in classes, appears in Table VII. The chlamydospores vary in diameter from 15.5 to 57.49 mi- crons, the largest number falling in class 39.5 to 41.49 microns. The mean diameter is, therefore, more than 35 microns, being approximately 39 microns. These measurements correspond closely with those for Phytophthora faberi Maubl., found in the classification prepared by Rosenbaum.(10) There was little or no difference in size between the chlamydospores produced by the fungus growing on oat meal agar and those produced on macerated corn. Germination of chlamydospores.—In Van Tieghem cells pre- pared with hanging drops of distilled water, corn meal extraci, pure agar, or potato agar, direct germination will take place within twenty-four hours. From one to twelve germ tubes may arise from one spore. xIv,i Reinking: Coconut Bud Rot in the Philippines 147 TABLE VII.—Summary of measurements of chlamydospores of Phytoph- thora faberi Maubl. from coconut bud rot. Chlamydospores in each class, ae Oat meal — culture; 3 corn cal eto — ture; 6 to * |Tdaysold Me BBD t015, 40 rik Boe eae Sa Wie oa ea eke He eee 0 0 wie Gee we A Ee 5 sitll eh tee app rte sDomeive ys etios FED TOLER GIN ie Pete RANA Co Wie nee Re MS 0 1 Be O 00 10. Bene Fa ee ge AE Ld Sh ee ceawdeonke 0 2 19: Sto Bite oo ne eee kt aca ee eee eal eiedn whee 1 2 91. 6 40 FB Oo oo octesisiinn iene wed mink aman Cauins anus news eins~ Deal ascwniage 0 1 OS Be BO Ae ee ea A en ee Gane bee ehat ag ai aidan: f 3 25.6 WO S149 once ace eee nica nde a ee oe eaeewendes wens 0 2 OT; 6 50 SON eo ea eo ee ed ee ee ae 8 0 0; Br AO ie sg, Bak aeeunadatan ipa tehaks chee ducm ake nee 2 8 SIG Wee G0 oo eee ee ele gels cade een cau 12 5 SRW te Oba a a eee Se Ee ps 2 8 Biz SOB TG ea as 5 a a a Se ee ee ee ae eae 20 15 SOT Th Sak es Aes a ee pre ee ik ge mene ne CON ane deh Minin oclemcieesiw ims 15 9 BO EA A Soo ag ical ake ace nds basbies cue ucda at eanea esse le setae eres 52 24 Es We Ge es ss eats ce ci wadaueansbadbcapanadeaaen 15 15 Ts ca ae ee SE ins Spa caine dee nda sia pewuubws, 41 10 ce i ia a ig Ga rg ha oe ese ks ee So he te 13 2 ye Mie A oe se ale a ce akg enewnbawe Ue Gasp ensue ognee 18 2 OR EO ow ee as Na Sl ci wa ban ap awe Senda e Copan pee len geaume 1 0 61666 G0 40 oer eee a cons tea poe ead ee eee cab ees ceeien 1 0 BGG be Be: Os a a ea edo oe rd atone adap weit ace e nu slee ceuent 1 0 BO6 0607. 80 oso aia ee orc ann sew en eet tla deh wean one Gen 2 1 OI ta BO 4Y ooo ce eee cc algae se seatetcanse carn etee 0 0 thy SO SE Ba seaeegen nam nous eeee ewes Seat waamer en matone 200 105 SEXUAL BODIES No sexual bodies have been observed in diseased portions of coconuts or in pure cultures. The absence of antheridia places the fungus at once in the faberi group, according to the classi- fication of Rosenbaum. (10) TAXONOMY In determining the species isolated from coconut bud rot, the methods employed by Rosenbaum(10) have been followed. Ac- cording to his key, which is given below, the fungus is included in the faberi group and is the species Phytophthora fabert Maubl. AAA. Faberi group.—Antheridium entirely unknown or its relation to the oogonium not yet determined, chlamydospores absent or present. P. faberi, P. jatrophae. B. Chlamydospores large, mean diameter more than 35 #. 148 The Philippine Journal of Science 1915 C. Mean diameter of chlamydospores 38.98 «, mean ratio of length to width of conidia 1.47........ 10. P. fabert. BB. Chlamydospores small, mean diameter-less than 35 uv. C. Mean diameter of chlamydospores 32.89 u, mean ratio of length to width of conidia 1.28...11. P, jatrophae. SUMMARY 1. Enormous losses, amounting to thousands of pesos each year, are produced by coconut bud rot. As shown by the Bu- reau of Agriculture reports, the disease is most prevalent in Laguna, Tayabas, Pangasinan, and Zamboanga Provinces. It is most abundant in very humid sections and in thickly planted groves, both of which conditions are: found on the slopes of Mount Banahao. Field studies show that the spread may be extremely rapid during favorable weather. 2. An organism similar to Bacillus coli (Escherich) Mig., and other saprophytic bacteria are associated with the disease. Under certain conditions, such as a host weakened by severe injury, the former organism and Bacillus coli (Escherich) Mig. isolated from man or horse, may in inoculation experiments produce disease. 3. A summary of the entire bacteriological work done by the present writer, including approximately three hundred inocu- lations, has indicated that, while the bacteria are always present and are a factor in destroying the weakened tissues, they can- not account for the initiation of the disease or its prevalence and rapid spread. 4. Phytophthora faberi Maubl. isolated from cacao produces = typical bud rot of coconut seedlings and of mature coconut ees, 5. A fungus isolated from a typical field case of coconut bud rot was found to be identical with the Phytophthora faberi Maubl. isolated from cacao. 6. Phytophthora faberi Maubl. isolated from the field case of coconut bud rot produced in all inoculated seedlings a typical infection. 7. Phytophthora faberi Maubl. isolated from the field case of coconut bud rot produced disease in coconut seedlings, cacao fruit, Hevea rubber seedlings, and papaya fruit. The same species of fungus isolated from cacao fruit produced disease in coconut seedlings and mature trees, cacao fruit and stem, Hevea rubber seedlings and mature trees, and papaya fruit. 8. A morphologic and taxonomic study of the organism isolated from coconut has proved that it is Phytophthora faberi Maubl., as described by Rosenbaum. (10) XIV, 1 Reinking: Coconut Bud Rot in the Philippines 149 e 9. From these researches it can be stated with certainty that Phytophthora faberi Maubl. causes coconut bud rot; bacteria are apparently, in the majority of cases, always secondary, but are concerned with destroying the weakened tissues. 10. By proving that the fungus causing coconut bud rot is identical with the organism which produces black rot of cacao pods, canker of cacao, fruit rot and canker of Hevea rubber, and rot of papaya fruit, it becomes evident an entirely new series of controls will have to be devised. Phytophthora faberi Maubl. may grow readily, under favorable conditions, as a saprophyte also, on dead portions of coconut, cacao, and papaya. RECOMMENDATIONS 1. Trees when once severely infected never recover. The mode of growth of the palms and the nature of the disease make it impossible to cure trees already badly affected. 2. Systematic inspection, condemning and burning of all dis- eased coconut trees, as carried on by the Bureau of Agriculture, should be continued. | 3. All parts of diseased trees must be burned; otherwise the organism will live as a saprophyte on dead. matter, and then spread to healthy trees. 4. Clean cultivation ought to be practiced in all groves. 5. Under no circumstances should coconuts be interplanted with cacao or papayas. 6. If coconuts are planted near diseased Hevea rubber, pre- cautions should be taken to avoid the spread of the disease. 7. Trees in new groves must be planted 10 meters apart each way. This spacing is one of the most satisfactory means of control against bud rot, and at the same time tends to give the highest production of nuts. BIBLIOGRAPHY (1) Barrett, O. W. Diseases of the coconut. Philip. Agr. Rev. 5 (1912) 262-263. (2) Butier, E. J. The bud rot of palms in India. Memoirs of the : Department of Agriculture in India 3 (1910) 221-280. (3) Byars, A. F. Coconuts in Laguna and Tayabas Provinces. Philip. Agr. Rev. 1 (1908) 516-520. (4) CEVALLOoS, FELIPE O. Control of diseases and pests by cultural methods. Philip. Agriculturist and Forester 1 (1911) 86-88. (5) CopELAND, EpwIN BINGHAM. Bud rot of the coconut. Philip. Agr. Rev. 1 (1908) 210-220. (6) IpeM. The coco-nut. London, Macmillan and Co. (1914) 43-63. (7) JoHNSTON, JOHN R. The history and cause of the coconut bud rot. U. S. Dept. Agr. Bur. Plant Industry Bull. 228 (1912) 1-175. 150 The Philippine Journal of Science e (8) Macktz, D. B. A summary of the work of the pest control section for the year 1916. Philip. Agr. Rev. 10 (1917) 128-145. (9) REINKING, Otto A. Philippine economic-plant diseases. Philip. Journ. Sci., Sec. A, 13 (1918) 192-196. (10) RosenBAum, J. Studies of the genus Phytophthora. Journ. Agr. Research 8 (1917) 283-276. (11) Roxas, MANUEL L. The cultivation of coconut. Philip. Agriculturist and Forester 1 (1911) 57-60. (12) Westrr, P. J. The coconut, its culture and uses. Philip. Agr. Rev. 11 (1918) 5-57. ACKNOWLEDGMENT The writer is deeply indebted to members of the Bureau of Agriculture for the kind assistance they have given toward ob- taining bud-rot material and data concerning inspection work; to Dr. S. F. Trelease for reading, and giving many valuable suggestions in, the manuscript; and to his students E. Roldan, F. D. Luistro, F. M. Clara, and L. S. Clemente for their kind assistance with some of the inoculations, ILLUSTRATIONS {The plates are from a publication prepared by the writer on Philippine economic-plant diseases Fig. Fig. Fig. in the Philippine Journal of Science, Sec. A, 18 (1918) 192-274.] PLATE I 1. Coconut bud rot. Old infection. Entire central group of leaves killed and some fallen over. 2. Coconut bud rot. Central leaves killed and some fallen over. Outer older leaves healthy. 3. Coconut bud rot. Diseased central bud fallen over. PLATE II 1. Coconut bud rot. Young infection, showing unfolded tips of leaves newly diseased. From this point the disease advances downward into the growing point and more woody portion. 2. Coconut bud rot. Old infection. Entire “cabbage” and growing point softened. 3. Coconut bud rot. Old infection. Rotted portion just above’ grow- ing point. 4. Coconut bud rot. Old infection. Characteristic brownish stripe, showing limits of infection in the wood. 5. Coconut bud rot. Disease produced in an injured seedling with a pure culture of bacteria similar to Bacillus coli (Escherich) Mig. 6. Coconut bud rot. Disease produced in an injured seedling with a pure culture of bacteria similar to Bacillus coli (Escherich) Mig. PLaTE III 1. Coconut bud rot. Young infection. The fungus entered in young leaves at top. Note brownish line of demarcation at the limits of the advance in the woody parts. 2. Coconut bud rot. Young infection starting in at young unfolded leaves at top. 8. Coconut bud rot. Portion just below growing point in “cabbage” and young wood. Young infection. eo 1 > +» LV, Nos 1. [Puiuie. JourRN. Sct Coconut Bup Rot IN THE PHILIPPINES. ] REINKING: Fig. 3. Fig. 2. COCONUT BUD ROT, Fig. 1. PLATE I. RRPINKING: CocoNut Bup Rot IN THE PHILIPPINES. ] [PHILIP. JOURN. Sci., XIV, No. 1. PLATE Il. COCONUT BUD ROT. REINKING: CocoNut Bup Rot IN THE PHILIPPINES. ] [Puiuie. Journ. Scr., XIV, No. 1. PLATE Ill. COCONUT BUD ROT. REVIEWS Concerning some | Headaches and Eye Disorders | of Nasal Origin | by | Green- field Sluder, M. D, | Clinical professor and director of the department of laryngology and rhinology, Washington University Medical School, | St. Louis. | St. Louis | C. V. Mosby Company | 1918. 272 pages, with 115 illustrations and index. Impotence and Sterility | with | Aberrations of the Sexual Function | and | Sex-gland Implantation | By | G. Frank Lydston, M. D., D. C. L. | [3 lines of titles] | The Riverton Press | Chicago | 1917. 333 pages, illus- trated. Cloth, $4. Naval Hygiene | by | James Chambers Pryor, A. M., M. D. | medical in- spector, United States Navy; master of arts in hygiene | Johns Hopkins University; head of department of hygiene, | U. S. Naval Medical School; professor of preventive | medicine, George Washington Uni- versity | published with approval of the Surgeon General, U. S. Navy | and | by permission of the Navy Department | Philadelphia | P. Blakistcen’s Son & Co. | 1012 Walnut Street. Cloth, 507 pages, with 153 illustrations and index, $3 net. Modern Chemistry and | Chemical Industry of | Starch and Cellulose | (with reference to India) | by | Tarini Charan Chaudhuri, M. A. | professor of chemistry, Krisnath College, Berhampore (Bengal); formerly | Government Research Scholar in chemistry; author of “Sir William | Ramsay as a Scientist and Man,” etc. | Calcutta: Butterworth & Co. (India), Ltd., 6 Hastings St. | Winnipeg: Butterworth & Co. (Canada), Ltd. | Sydney: Butterworth & Co. (Australia), Ltd. | London: Butter- worth & Co., Bell Yard, Temple Bar. | Medical Publishers | 1918 | All rights reserved. Pp. i—viii + 1-156, including index. Price, Rs. 3/12/-net. Abstracts | of | Surgery | An abstract of the war literature of general | sur- gery that has been published since | the declaration of war in 1914 | prepared by the Division of Surgery, Surgeon- | General’s Office | St. Louis | C. V. Mosby Company | 1918 | 434 pages. Cloth, $4. As shown by the Preface, the “preparation of these abstracts, in common with many of the other early war activities, was an emergency war measure * * *, The volume must of necessity be regarded merely as a condensed text for ready reference. Most of the abstracts have been used through the courtesy of Surgery, Gynecology and Obstetrics, the Journal of the American Medical Association, the Medical Record, the Military Surgeon, and the New York Medical Journal. Some articles in the British Medical Journal and in Surgery, Gyneco- logy and Obstetrics were so fundamental that they were abstracted with a minimal amoynt of paraphrase.” - Genitourinary Diseases | and Syphilis | by | Henry H. Morton, M. D., F. A. C. S. | [10 lines of titles] | fourth edition, revised and enlarged | with 330 illustrations and 36 full-page colored plates | St. Louis | C. V. Mosby Company | 1918. Price, $7. osby pany | , ae Issued with THE PHILIPPINE JOURNAL OF SCIENCE, Vol. XIV, No. 1, January, 1919. PROCEEDINGS OF THE MANILA MEDICAL SOCIETY REGULAR MONTHLY MEETING, OCTOBER 7, 1918 MINUTES OF THE MANILA MEDICAL SOCIETY The meeting was called to order at 8.40 p. m. in the Philip- pine General Hospital by President F. W. Vincent. Twenty members were present. The minutes of the previous meeting were read and approved. Dr. Miguela Gemil was elected an active member of the society. The following resolution, presented by Dr. R. B. Gibson, was adopted: Whereas the National Guard of the Philippine Islands is urgently in need of medical officers at the present time. Therefore, be it resolved that the Philippine Islands Medical Association and its component organization, the Manila Medical Society, in joint session, urge those available individual members thereof, who have not yet offered their services to the Guard, to at once apply for service with the Guard as a patriotic duty and for the honor of the medical profession of these Islands, and to use what personal influence they may command to persuade other medical men to this end. The following papers from the surgical department of the College of Medicine and Surgery were read and freely discussed: 1. A Study of a Series of Cases of Splenectomy, by Dr. Poten- ciano Guazon. 2. Report of a Case of Schistosoma in the Appendix, by Dr. Potenciano Guazon. 3. Ureteral Obstruction, by Dr. José Eduque. 4. Demonstration of Some Important Cases of Fractures, by Dr. Ricardo Fernandez. There being no further business, the meeting adjourned. D. DE LA PAZ, Secretary-Treasurer, Manila Medical Society. SCIENTIFIC PROGRAM A STUDY OF A SERIES OF CASES OF SPLENECTOMY By Dr. PorenciANo GuAZON Several cases on whom splenectomy was performed by the department of surgery were described. The technic and special precautions to be observed were given in detail, and the post- operative treatment and prognosis were discussed. 155 156 The Philippine Journal of Science A CASE OF SCHISTOSOMA IN THE APPENDIX By Dr. PoTENCIANO GuUAZON Structures, seemingly schistosoma ova, were found in a section of an appendix. The possibility of the schistosoma as a cau- sative factor in appendicitis is suggested. URETERAL OBSTRUCTION By Dr. José EpuquE Several cases of ureteral obstruction were reported. The diagnosis was made by Roentgen-ray photographs. Recoveries. followed operation. The occurrence of this condition is com- moner than may be supposed, and is characterized by acute attacks of pain which may be relieved when the patient is kept in bed for a few hours. Skiagrams illustrating the cases were exhibited by Dr. Ricardo Fernandez. A DEMONSTRATION OF SOME IMPORTANT CASES OF FRACTURES By Dr. Ricarpo FERNANDEZ A demonstration of skiagrams of some complex cases of frac- tures was given, R. B. GIBSON, Editor of the Proceedings, Manila Medical Society. Issued with THE PHILIPPINE JOURNAL OF SCIENCE, Vol. XIV, No. 1, January, 1919. PROCEEDINGS OF THE MANILA MEDICAL SOCIETY REGULAR MONTHLY MEETING, NOVEMBER 4, 1918 MINUTES OF THE MANILA MEDICAL SOCIETY The meeting was called to order at the Philippine General Hospital at 8.45 p. m. by Doctor Gibson, in the absence of the officers and councillors. _On motion, duly seconded and carried, Doctor Schébl took the chair. Doctor Gibson acted as Secretary pro tempore. Twelve members and six guests were present. The minutes of the pre- ceding meeting were not read. The following scientific program was presented, and the papers were read, with the exception of that of Doctor Gon- zalez who was not present: 1. Clinical Observations on 178 Cases of Dysentery, by Drs. J. Albert and J. Tirona. Discussion by Dr. H. W. Wade. 2. A Case of the Pseudomeningitic Form of Infantile Beriberi, — _ py Dr. T. C. Arvisd. 3. A Case of Subcutaneous Emphysema Complicating Acute Bronchopneumonia, by Dr. Alberto Tupas. 4. Codperation between Hospital Ward, Dispensary, and So- cial Service Department, by Dr. J. Gonzalez. Discussion by Dr. J. Fabella. 5. Demonstration of Skiagrams of Beriberi Heart in Children, by Dr. R. Ferndndez. The meeting was adjourned at 10.55. . R. B. GIBSON, Secretary pro tempore, Manila Medical Society. SCIENTIFIC PROGRAM CLINICAL OBSERVATION ON 178 CASES OF DYSENTERY By Drs. José ALBERT and J. TIRONA The records of the department of pediatrics for the last two years show a mortality of 54 per cent in the dysentery cases entered, thus making it the most serious type of cases which the pediatrics staff encounters. The incidence and virulence of the infections is increased during the rainy season. The largest number of cases occurred in children under 2 years of age, and the next highest in those under 5 years; of seventy-three cases 157 158 The Philippine Journal of Science under 2 years of age, twenty-four were infants less than a year old (one case only 6 days old). Almost 50 per cent of the chil- dren of 5 years or younger died with pulmonary complications. The usual period of the disease is from three weeks to two months. Five cases had a history of previous or existing cases of dysentery in the family. Bacteriological examinations of the stools for B. dysenteriz were rather unsatisfactory, negative reports being received for many cases clinically dysenteric. Five cases were positive for Entameba histolytica. Twenty- eight cases were heavily infected with ascaris, ten had trichu- riasis, and eight had a double infection with these parasites. There was one case each of anchylostomiasis, oxyuriasis, and blastomycosis, and two having trichomonas. Four cases were complicated with paratyphoid, and one was positive for both pa- ratyphoid and typhoid infections as shown by blood cultures. Treatment is partly symptomatic; purgatives and saline colonic irrigations are given also when acute toxemia is present, and intramuscular or intravenous injections of antidysenteric serum are administered, when available, to the severe cases. A CASE OF THE PSEUDOMENINGITIC FORM OF INFANTILE BERIBERI By Dr. T. C. Arvist Pseudomeningitie cases of beriberi in infants have recently been reported to this society by Doctor Albert. Another case, 6 months old, is presented. Symptoms were slight fever, drow- siness, vomiting, cyanosis, convulsions, no crying, and ptosis. Skiagrams showed an enlarged right heart. Improvement and recovery followed treatment with vitamine preparations (tiki- tiki extract), which is confirmatory of the diagnosis. DEMONSTRATIONS OF SKIAGRAMS OF BERIBERI HEART IN CHILDREN By Dr. Ricarpo FERNANDEZ The method of determining the size of the heart from skia- grams was described. Plates exhibited demonstrated the hyper- trophy in beriberi cases, R. B. GrBson, Editér of the Proceedings, Manila Medical Society. Issued with THE PHILIPPINE JOURNAL OF SCIENCE, Vol. XIV, No. 1, January, 1919. PROCEEDINGS OF THE MANILA MEDICAL SOCIETY REGULAR MONTHLY MEETING, DECEMBER 2, 1918 MINUTES OF THE MANILA MEDICAL SOCIETY The meeting was called to order at 8.40 p. m. in the Philip- pine General Hospital. President F. W. Vincent and nine members were present. The minutes of the October and November meetings were read and approved. Drs. Pedro T. Lantin and Alva D. Cook were elected active members in the society. The society approved a plan to hold a symposium on the present epidemic of influenza at the coming annual meeting. The following program was presented, and the papers were read and discussed: | 1. Cataract Cases and Operations, by Dr. A. R. Ubaldo. 2. Unusual Complication after Grattage of the Lids for Tra- choma, by Dr. A. R. Ubaldo. : 3. Mastoid Operations, by Dr. H. E. Velarde. 4. Cases of Tonsillitis in the Dispensary of the Philippine General Hospital, by Dr. F. Nicolas. 5. Teratoma of the Maxillary Antrum, by Dr. H. E. Velarde. 6. Skiagrams of Cases from the Department of Eye, Ear, Nose, and Throat, by Dr. R. Fernandez. The meeting was adjourned at 10.50 p. m. D. DE LA PAZ, Secretary-Treasurer, Manila Medical Society. CATARACT CASES AND OPERATIONS By Dr. A. R. UBALDO “Fifty operations for cataract included twelve cases of double cataract, one congenital, one with a dislocated lens, and three traumatic. The ages of the patients varied from 4 to 90 years. The methods employed, including the advantages of a preliminary iridectomy, combined removal of the iris and lens, or extraction of the lens alone, were presented. In one case of double cata- ract, one eye was operated with iridectomy, the other sub- 162323-——11 159 160 The Philippine Journal of Science 1919 sequently without; vision after the recovery was 20/80 and 20/40, respectively. The author favors the upper corneal section. . Postoperative treatment, complications, and accidents were gen- erally discussed; prolapse of the iris with iritis was the most frequent postoperative complication in the series of cases. AN UNUSUAL COMPLICATION AFTER GRATTAGE OF THE LIDS FOR TRACHOMA By Dr. A, R. UBALDo A female patient, 26 years old, had trachoma for three years. Three weeks before operation the condition became more acute. Examination showed the upper fornix of the right eye covered with granulations. Three hours after operation (with local anesthesia, using cocaine powder) the patient complained of severe pain in both eyes. Treatment consisted of ice compresses, morphine, bromides, and aspirin. .The following day the patient had severe pain in the right eye; the lids were swollen and were opened with a retractor. An unusual keratitis with peculiar striations was observed. Both eyes were affected. There was no ciliary injection and no tension of the eyeball. On the third day pain continued in the right eye. Atropine dilated the left irisonly. An cedematous condition of the nimbus (sclero-corneal union) with beginning suppuration was observed. The condition lasted from two to three weeks. The right eye became entirely blind. MASTOID OPERATIONS ; By Dr. H. E. VELARpE The writer reviewed the common operative procedures and postoperative treatment. Of eighty-five cases considered, sixty recovered, eleven improved, four did not improve, five died, and five were discharged, results comparing favorably with statis- tics from other hospitals. TONSILLITIS By Dr. F. Niconas The etiology, character, treatment, and complications were discussed for the acute and chronic cases. The commonest forms of tonsillitis treated were the acute catarrhal and follicular types. The condition is most common in males, in patients from 14 to 25 years of age, and it occurs most frequently in July and August. XIV,1 Proceedings of the Manila Medical Society. 161 TERATOMA OF THE MAXILLARY ANTRUM Z By Dr. H. E. VELARDE The case described was that of a 14-year-old girl. The con- dition was of several years’ standing, there being an increasing prominence of the right face. The right maxilla in the region of the antrum of Highmore was found to be prominent and bulging, extending to the side of the nose. There was no in- flammation nor was there tenderness on pressure. A skiagram showed the presence of a tooth in the antrum. A radical op- eration was performed, the antrum being opened by a blow on a chisel, the membranous sac (containing fluid) was incised, and the sac and the single rooted tooth buried in the roof of the antrum and floor of the orbit were removed. An opening into the nose was left for drainage. The pathological examination indicated that the tumor mass was of the nature of an enamel organ or adamantinoma. The location is unusual. SKIAGRAMS OF CASES FROM THE DEPARTMENT OF EYE, EAR, NOSE, AND THROAT By Dr. Ricarpo FERNANDEZ An instructive series of plates was presented showing the locations of coins in the esophagus and lower tract, seeds (coated with bismuth subnitrate) in the intestine, an automo- bile (cracker) in the cesophagus, three cases of foreign bodies in the eye which were located by double exposures with the eyes directed first to the right and then to the left, a shot in the orbit, and sinus and mastoid infections. R. B. GIBSON, Editor of the Proceedings, Manila Medical Society. THE PHILIPPINE JOURNAL OF SCIENCE VoL. XIV FEBRUARY, 1919 No. 2 PHLEBOTOMUS NICNIC, A NEW SPECIES, THE FIRST PHILIPPINE RECORD FOR THIS GENUS By CHARLES S. BANKS* (From the College of Agriculture, University of the Philippines, Los Banos) ONE PLATE Although the genus Phlebotomus has been recorded from In- dia’ and Ceylon? no species of this genus has ever been noted as from the Philippines; in fact, casual collecting during a period of many years has resulted in the amassing of a con- siderable number of species of the Psychodide, none of which has been determined or described and among which there are sure to be found several new species in this very interesting group of the Diptera. The present paper is written especially to bring to scientific attention a new species of the genus as a serious factor in human existence in this part of the world and as a not improbable agent in disease transmission. For a number of years I have had accounts of a tiny fly which bites at night but which is “too small to be seen,” and it was only after experiencing the bite of this pest in the summer of 1915 that I was able definitely to assign to a given fiy the term “nicnic” which is the Tagalog name of a “tiny fly too small to be seen.” I have therefore decided to give the vulgar name permanence as well as definiteness, by using it to designate a new species of Phlebotomus which was particularly abundant at the College of Agriculture campus about the middle of July in 1915 and which has been awaiting description since that time. Professor of entomology and chief of the department of entomology, University of the Philippines. * Brunetti, E., Fauna Br. Ind., Dipt. Mem. (1912), 202. 164001 163 164 The Philippine Journal of Science 1919 Phlebotomus nicnic sp. nov. Male and female.—Grayish ochraceous to brownish buff, with slight silvery reflections at ends of some of the squamose hairs which so abundantly cover the body. Seen by transmitted light head and thorax are honey yellow, abdomen and legs are buff. Head.—Long ellipsoidal to end of clypeus which is slightly concave basad, and inflated apicad, with numerous conical tu- _bercles on its apical third. Proboscis one-half length of entire head. Palpi with first and second segments subequal, the former curved basad; third slightly longer and thinner, fourth and fifth subequal and nearly filiform, the latter bulbous basad. Antennse with first segment cyathiform, second spherical, third three times length of fourth. Segments 8 to 10 slightly fusiform, segments 11 to 15 cylindrical, segment 16 obovoid fusiform. Each segment bears a subbasal regular whorl of curved hairs and other scattered groups most of which are three-fourths the length of the segment or longer than it. Eyes dark brown to black, nearly circular in outline and seven-eighths the occipito- gular diameter. Hairs on head and thorax erect and of a length ' equaling the head diameter, Thorax.—Pronotum a tiny squamose sclerite slightly proflexed over base of head. Distance. in profile, from mesonotal dorsi- meson to apex of coxe equal to length of abdomen. Length over mesonotum slightly less than this. Scutellum well rounded in profile, metanotum nearly plane. Upright hairs abundant. Abdomen.—Cylindrical in male, slightly obovoid in female, with semierect hairs as long as the segments and evenly scattered over the tergites and sternites. Legs.—Long, slender, coxee and femora sparsely covered with long coarse hairs, femora, in addition, and remaining segments with a thickset covering of very fine recumbent downy hairs, Ungues rectangulate at their basal fifth, thence straight or scimitar-shaped apicad. Wings.—Linear ovate or subspatulate, twice length of abdo- men, their greatest width one-third their length in both sexes, very hairy on all veins, especially costa, sixth vein, and anal margin. Auxiliary vein ends midway between costa and first longitudinal, touching neither, but causing a decided curvature in latter. Petiole of first fork of second longitudinal equals an- terior branch of fork. Fourth longitudinal forks at middle of wing. All veins except third longitudinal enter wing margin nearly perpendicularly, angulating before entrance. All veins about equidistant over wing surface. Both anterior and poste- e 7 b XIV,2 Banks: Phlebotomus nicnic 165 rior cross veins very evident, the former being at origin of third — longitudinal and the latter at that of fifth. Seventh longitudinal not present or, if so, confused with fold in base of anal margin. Halteres very stout, with large, dark knob and paler stem. Genitalia.—_Hypopygium of male twice length of last abdo- minal segment, ventral styles fleshy, straight, or slightly curved, setose along sides and at rounded apex. Harpes asymmetrically spatulate and with four stout, curved spines along apex. Penis slender, constricted before apex which is obconical. Ovipositor of female with ventral lobes broadly ovoid and setose, dorsal lobes minute, setose. Length.—Owing to the humpbacked attitude of this insect the measurements are taken from cephalic end of mesonotym to caudal end of abdomen and from tip of proboscis to dorsum of mesonotum. Male: Thorax—abdomen, 1.083 millimeters; proboscis—thorax, 0.59; total, 1.673. Female: Thorax-abdomen, 1.317 millimeters; proboscis— thorax, 0.85; total, 2.167. Male: Length of wing, 1.44 millimeters; female, length of wing, 1.74. LuzoN, Laguna, Los Bafios (Charles S. Banks). Type.—Male and female, No. 18492 in entomological collec- tion, College of Agriculture, Los Bafios, P. I. Several addi- tional specimens collected at the same time and place are labeled as paratypes. The habits of this insect, which is a very vicious biter, are described in another paper dealing with the bloodsucking in- sects of the Philippines. It is believed that the nicnic breeds in the kitchen drains of this vicinity, and as soon as time will permit an endeavor will be made to rear it. ILLUSTRATION PLATE I. PHLEBOTOMUS NICNIC BANKS SP. NOV. . Head of female, showing peculiar clypeus, first segments of antenner, and form of palpus. . Profile of clypeus, showing conical portions of hairs which have been left when main part has been broken off. Apical segment of fore tarsus of female, showing rectangular ungues. Antenna of male, showing three obconical distal segments. . Antenna of female, showing single obconical distal segment, much larger than that of male. . Wing denuded, showing peculiarities of venation. . Genitalia of male, profile. . Genitalia of female, profile. . Genitalia of female, ventral aspect, showing peculiar form of sternite of last abdominal segment, 167 BANKS; PHLEBOTOMUS NICNIC.] [PHILIP. JOURN. Sct., XIV, No. 2. y g st SERIE RNase, wean <_| <= AE uaa LEE i Ss SRS

pBr9qnz7103Uy ---- eS ee serene -o-- wenn | eee eee - +e -- we--- Sg tahrg wy Acerca coy gin aie anc chaa tater nce ates miatctipcstginaerepenacaiead | ha AOLIOYSOT 99°L Chnbn ame nels ened mt etloonebn enna ee ee aeclennece —- Re Re RE REE ERR EEE ERR ERR 18771388 aOllejuUy eS eee eee ed es ne -- a te AS ee ee en et rr er Ge wororere<<= TBIOBBOIYIS] ie foes |---- aneeene on on bt aes ~ateiererencuccantanesannnannnn+=9"" 21 39,8000299U8 30700 tte | sor | sor Oewr fete Per rami ahs | a ba. Waves 2 mRNA TAC de Rh 4's ONAN RBM TIE at Ira dh are Ee ee ene eter en -9 oe" 9 Sabooepaeg ppt Su 609 "8% €2°08 =| 89°82 ae eh On ae eae bance tee ee ee ame RE ee ete sents any ce | POOLE, L°S | STS "Fe bE "93 Pe. 4) Ore 0°82 62 62 Be Oe ee a aa ae en ee ea ns Oa 9°32 | Bes"es TO'WS =| 99°82 66 “8% 8°92 9% 9% ve me rrnene rene te” beer are menor eee are e kt te, oe pa SIS Acosta-Sison and Calderon “TZ. aerouet) ordmats ba ial, *[SULION | “qoMor | * “uruIn “pye ‘suoWUy Apoomy, @ | ‘PesI4V “707 0UTe| Ap . Teusouqy,, ic bi 3 t : : i ‘orBoN {8331 mor | yysiy jueuLier oy XIV, 3 MS, Sra ae 70M ea Sines P 258 The Philippine Journal of Science 1919 an = ~SINSS S siss lad L ate MY Suge SSN SNS SSNS fo Lie Interspinous diameter (measured from tip). f\2 is Fig. The easiest method of bringing out the mixture of types is by plotting measurements mn: es graphic way, so that the resulting curve will show the proportion or per- centage of individual cases of any and all particular types. The measurements of the one thousand two hundred thirty-seven pelves on which the present paper is based have been plot- ted in such’ graphic curves. For the whole number of cases. the curves for each measure- ment are shown in figs. 1 to 5. This investigation has shown that we are able to isolate three major, and two minor, types of Philippine pelvis, as has just been demonstrated. To associate any of these types of pelvis with par- ticular racial varieties of man would be folly in the present state of our knowledge (or rather ignorance) of the other races of Oriental peoples, who surround the Philip- pines. Anthropologists _ be- lieve that the different racial types of India, China, Japan, Polynesia, and Melanesia—in addi- tion to the late Spanish » = a , ee XIV, 3 Acosta-Sison and Calderon: Pelvimetry 259 mixture from Europe—have all influenced our composite pop- ulation. If it were possible to compare the measurements made in our investigation with the considerable series of others from all the different racial types just mentioned, we might hope to arrive at certain tentative conclusions. However, since practically no pelvic measurements from the regions mentioned sia] «& 5 | age PAC) Ea eZ} [e| a Tieig|2 Py, Z SD) 6S |z au raratd 3 Si Bou 3 == $] SI a 3 bieic} zi P15 7; = 2 Bi S Pita cit 274 5 4 4 1) 3 2) he oe I =" ‘ 8 o é g ai z ry g ca FS ‘ata A | viel © i723 ee a 8 * 4 ii 3 7 = Lois 3g ee iva a , _ wie] Ota £ Siale rae Bag tole 0 Eau te SHS hj S/O/S[alo/4/S!9/o a HS ROMSRIO Sosa S| ba bial ) SSSss IN PO RON NOISE OE lOu Sep PIopI Nels ba bd ice OE GGT wy SBEERRRENSEERRERERAESNSEL ET Ti t tii LEET 260 The Philippine Journal of Science 1919 KISKRARAAKRIR ‘4 RAAKRRARARS {) ~ 2 i Bi ee come jE 2245) a Z } £ c etude iis pd J.) ciel ipl lal AL aie Sled S be a ~ iN an Fic. 3. Baudelocque’s diameter. have been published, the correlation of Philippine types with those of other oriental peoples must await the results of further investigations.® | 2 On examining Table III, we find that the diagonal conjugates measuring from 11 centimeters to 12.4 centimeters have the greatest number of cases (which form 85 per cent of the total * Though unable to draw conclusions from our pelvic measurements, it may be proper to state here that anthropologists have demonstrated from other sources that the predominant Philippine racial types are two: First, a type known as the Indonesian, which came into the Islands from the south and is related to the races of India; and, second, a Mongol-Chinese type, which came into the Islands chiefly from the north. The so-called Malayan Filipinos of to-day are essentially a mixture of these two types, with the addition of certain minor elements. The most important of the latter is a type represented by immigrants from Melanesia. Other minor types are of course the historical immigrations of Chinese and of Spanish Europeans. It is possible, then, that our five pelvic types correspond to these racial elements; though for the present such a conclusion must be regarded as mere speculation. (For discussions of Philippine racial types, see Beyer, Cole, Bean, Koeze, Keane, Montano, and Virchow.) ‘+ XIV, 3 Acosta-Sison and Calderon: Pelvimetry 261 BRS 7 11010 r S | } ri i Hes : i 2 Ss AT g if i" HBF | i iy 2 i Z i Ze y HHS x 3 ECECSCCee che SUSI SSS Aa teint sd dapdeaciceed st i Lit ASS SS Re ee ~~) SN ~~ Fic. 4. Intertuberal diameter. | als i aPC ; AAs é LO} OS fais) a I [ § ih Be Hi} Ry aa (9) Os q [4] | | OF | 14S 3 ‘A BIG Fy iI wo < “ 2 210 il i] li i fio 8. Dae ae | ~ Shida sy | H EN = | BaRe 7 SIN 5) Bea ~ $°st "BIABAd BUEDBI J |--~-~ TI ‘WOIsIDA pale. Suan he ae re Pee Pee : ‘L ‘Zo[qno MorzeU “gq '°O°T | T ‘sdeor0f pr eh 'TL st of 86 ST 03 2°aT ‘T“d'O°U “Tt ‘Bisdurepog j----- % ‘sdeor03 prt | POT T'L6 $8 6 Wea te te ee PZT 93.2 ‘T ‘peoy qa pusy jo wIIq WURIMIODUOD 0} eNp 23e7Is puodeS pesuo[oIg “*T ‘eisdurepoq ‘e“d ‘O° ee L ‘UOisISA OTBPOd') py T 9T "S96 BLO RS Bo" SECS gaa onch bs bala oe eg 6°IT 91 9°IT ‘T ‘Aquq Sig *T ‘etaoid ¥QU908[ I “[ ‘a384s puoves pasuojoig |~~~----~ % ‘sdao10y MOTT | ‘T ‘e3u3s puoses posuojorg |--~---~ g ‘sdoo10F SIT | *p ‘o8e}8 puooes pesuojo1g | ; “| ‘esdejoid pusy yy *g*O°WUh) g ‘sdaoi0J pry : tata be >) 208 Pe ae a prea ormaicgs eiaaie ee VIE C8 IT *T ‘woreseuy }|---"-~~ g ‘sdeor0oy MOT *g ‘adeqs puooses pesuofoi1g TeV 20H ‘I ‘pReu Ziq’ d'O"l % ‘g1adid Bjus0B[d |----~ p ‘UOIsIAA DITBpog "uolzIsod eszeasuery, |~-~-~ T ‘WOIszeA OI[epog *BI}IOUL OULIEy) [-~77 > TOROS ei A ca uw wehbe des ook : ! *e8B]S puoves pesuolorg | --~--~~ I ‘sdaos0y plc | 9F - uM ot o a *Y¥ ‘WW’ 0} 03818 puoves poSucjorg |~-~-~-~"] “sdeoroy MOT] LL Ol TOU |" oe eeeoees PHN| Or late au tee agi "erjazaul autre} |----~ z ‘uOIsI9A OITepog} 6 £3 "69 ot FT &1 FOL 03 OL “gaeng "99 ‘d ‘mM “H “9 t : “10qBy "uoryerodo Joy SUOT}VOIpUT *suoieredGQ, “10qB] Snosuszu0dg re eae, “sasty ‘quouleinsts py ‘saspo OTg9 S@undiyynwm ur ogm snoaunjuods fo Rhouanbaw{ pun Rouawryfa fo mara fo yuod ay) wots ayninfuoo pnuohmg— III ATAVL L Baeae ‘~ XIV, 3 Acosta-Sison and Calderon: Pelvimetry 263 TABLE I1V.—Diagonal conjugate from the point of view of efficiency and spontaneous labor in primipare. Length s A pontaneous ; Indications for Measurement. Cases. od - tabew. Operations. operation, cm. H. m.| P. et. | Cases. RS 66 10.48 2555s ccce 9; 15 35; 88.88 8 | Low forceps, 1...) Prolonged labor. 10.8 to 18,9 -ex:is 21| 18 42| 97.76 19 aid forceps,1-.-| Do. Mid forceps, 1 .-- Do. Dry labor, 1. ies 92115 11] 97.88 90 | Mid forceps, 2....|)Meconium in vagina, (baby presenting by the vertex) 1. 338 06-1159 ece 74\18 50/ 98.64 73 | Mid forceps, 1..--| Prolonged labor. Pie ie bocce 34| 14 52] 94.01 32 | Mid forceps, 2---- Do. | 12. 6 t040 6255s 6| 15 02 100 6 fin cece cama iene, Comparing Table IV with Table III on the multipare, one is struck with the apparently greater percentage of spontaneous la- bor in primipare. The cause of this is that in primiparous cases, our procedure has been to let nature do all it can before instru- mental intervention so as to allow sufficient molding of the head. Moreover, many of the operative cases in primipare are not shown here for the reason that we made all of our measurements during the puerperium, and in order to avoid infection and relaceration of the repaired wounds, a regrettable occurrence in many of our earlier cases, we refrained from measuring the diagonal conjugate in patients who had been delivered instru- mentally. | It is interesting to note, however, that the number of cases whose diagonal conjugate measures between 11 and 12.4 centi- meters forms 87 per cent, which is a slightly greater proportion than that of the corresponding measurements in the multipare. PELVIC OUTLET An effort was first made to determine the relationship of the intertuberal diameter and the length of the second stage of labor, but the results soon demonstrated that this measurement when considered alone is of little significance. In cases of contracted intertuberal diameter where a prolonged second stage was ex- pected it was found that, though the baby was not undersized, oftentimes the duration of the second stage was even shorter than the average normal. In these cases the posterior sagittal 264 The Philippine Journal of Science 1919 diameter had a compensatory lengthening, so that the index of the posterior plane of the pelvic outlet was either normal or larger than normal. This confirms the view of Williams, Daniels, and others, that a contracted intertuberal diameter does not neces- sarily mean a prolonged or operative labor so long as there is a corresponding lengthening of the posterior sagittal diameter. Williams has stated that a spontaneous labor is exceptional with the following measurements (head, average size) : Transverse outlet. Posterior sagittal. em. cm. 8.00 7.5 7.00 8.0 6.5 8.5 6.0 9.0 5.5 10.0 but that, with these measurements, a safe delivery through the accustomed channel may be effected by instrumental means. Daniels, however, basing his calculations on the average normal measurements reported by Klein—namely, intertuberal diameter, 11 centimeters; posterior sagittal, 9.95 or 10 (Daniels uses 10 instead of 9.95)—has established the following rule: That the index of the posterior plane of the pelvic outlet in normal cases should be 55, as in the following equation: Intertuberal diameter (11) x posterior sagittal (10) 2 == 55.0 or Intertuberal diameter * & : == 55. 2 This index, according to Daniels, may be as low as 33.3 and no operation needed but perhaps the use of forceps. Following the formula of Daniels, we find that in Filipino women, the normal index of the posterior plane of the pelvic outlet is much lower than what he considers to be normal for American women. It is 11 Xx 8.5 “== 46.75 or 46.8; and an index as low as 31 may end in spontaneous delivery. te ‘XIV,3 Acosta-Sison and Calderon: Pelvimetry 265 Taste V.—The relationship of the index of the posterior plane of the pelvic outlet to the length of the second stage and birth weight of the baby. “9 - bite! orm: grams 0 and spon- tp Average| birth Index of posterior plane of pelvic outlet taneously! probe) birth j|weightof in multipare. delivered. ane at weight | baby per Cases, pho of baby. minute of 651. i second stage. Hm | 9. Te ee a is a eee seas he paca sida heneye 124|0 57.75 3; 48.1 ee Be cr cc radenunes a eagheeapemendae 890 | 1 00 2, 996 49. 98 Bak 0 0 a ie os cis Seeeee eee 126; 0 44 8,085 69.0 Oe 00 TR Oech oss ck own cdc cvauas ees ence aleavecwencetan 11/0 47 8, 041 65.0 On examining Table V one notices the close relationship be- tween the weight of the baby, the index of the posterior plane, and the length of the second stage of labor. TABLE VI.—Cases with prolonged second stage of labor; 43 cases; not included in Table V. Multiparez. Aver- Aver- age length (Aver | of Index of the poste- ss Z 4 rior plane of pelvic Cases) .,, of a bieth weight Complications. Operations. outlet. nt ari stage weight.) minute of of labor. second stage. HH. m. g Prolapse of hand with vertex, 1----- Placenta previa, 1_.- Mid forceps, i. $1 to 40.9.2 cccccc., 10; 8 84) 8,188 | 18.4 4h. O. P., 1-222... few forceps, 7. R. O. T., 1---.------- Podalic version, 1. Wes Lan h deme ccecs Prolonged labor, 5_-- IE 8 Ae SOG Bie ceegcoet ot . vO Ps Boas iids Placenta previa, 1___ He OoT. dgeecacaee Transverse position, 41 to 50.9 25| 2 44| 3,089; 18.8|) 2 nieecpliorgne Poe [ict <1 Se geketeenapea eee . AH mee er eee : ne SG SG "Roe Mid forceps, 5. Prolapes arm with Low forceps, 3. Werte, fo sas Uterine inertia, 1___- Ory labOr SD occu Prolonged labor, 3..- BR OP; 4c BI to 67.9.--------. 8| 2 10/93,448| 26,5 Placenta previa, 1...) pogatic version, 5. Fis babyy Bcc. ccan< 2 One baby weighed 6,320 grams. 266 The Philippine Journal of Science 1919 The close relationship of the baby weight and malposition or malpresentation with the length of the second stage and the size of the index of the posterior plane of the pelvic outlet is here emphasized. TABLE VII.—Cephalic measurements in multipare and primipare; total cases, 1,237. [Measurements are in centimeters.] Multiparz, 780 cases. Primipare, 457 cases. Measurement, Average.| Least. |Greatest.!Average.| Least. | Greatest. Occipit tal di LE eRe Gy rear 12.78 pe ESS 14.5 pb Hey wg 11.0 14.4 Occipito-frontal diameter ---_-_..------ 11.29 9.8 13.5 10. 88 9.5 12.1 Suboccipito-bregmatic diameter- ---- 9.614 8.2 10.9 9.39 8.4 10.5 Biparietal diameter --_-..-.---------- 9.031 7.5 10.2 8.9 4.9 10.0 Bitemporal diameter ----..---------- 7. 992 6.9 9.0 7.89 Fok 9.1 Occipito-mental cireumference---..-- 36. 711 30.9 43.4 86.5 33.0 41.2 a 40 Occipito-frontal circumference -_--_-- 83. 189 28.7 nL Occipito-bregmatic circumference--| 31.158 27.3 $5, 8 tis cho Fo eee aL. M. A. dL. O. A. Table VII shows a slight diminution of the cephalic diameters in primiparz as compared with those of the multipare. This TABLE VIII.—Showing the relative duration of labor and length and weight of babies in multipare and primiparz; 1,237 cases. ‘ Multiparz, 730. Primipare, 457. Average.) Least. |Greatest.|Average| Least. | Greatest. BRO oo ck concer coe bee aries years__ 27.07 16 45 18. 87 15 45 Length of labor, spontaneous i: OLA ee hrs. min __ 9.01 1.00 48.00 16. 00 4.20 72.00 Length second stage spontaneous : LO. Ale See ee 0. 52 0.05 4,00 2.12 0.15 4,00 Length of baby, female; 360 of multipared: | 2 see em_. 48.98 44 53 Length of baby, male; 420 48. 98 44 61 of multipars 2-2 - << sxcac om.3] 49.35 43.0 55 Weight of baby, female, grams.-| 2,972.59 2, 728 Weight of baby, male -.--.----- do-_-.| 3,039.08 {1,700 —— 2, 831 { x Weight of baby, both sexes_...do_._.| 3,005.83 | 1,700 4, 500 2, 831 1, 260 4, 130 Weight of babies of both sexes whose mothers stayed more than one week in hospital before par- turition 2.0.5 .65 scuscawedeed do....} 3,144 Pate SERN bed XIV,3 Acosta-Sison and Calderon: Pelvimetry 267 diminution is most marked in the fronto-occipital and suboc- cipito-bregmatic diameters whereas the mento-occipital dia- meter is almost the same in both cases. The figures in Table VIII are self-explanatory. As in other nationalities, the multipare have a shorter duration of labor, and longer and heavier babies than those of the primipare ; also, the female babies are smaller than the male. The influence on the birth weight of the baby of a week’s sojourn in the hos- pital before confinement is here well shown. The difference in the average birth weight of the babies whose mothers stayed one week or more in the hospital before parturition and those whose mothers entered the hospital on the day of, or a few days _ before, confinement is 139.85 grams in favor of the former. TABLE [X.—Average length of labor and duration of second stage in normal spontaneous deliveries in multipare. These form 92.807 per cent of 780 cases of multipare. Length of labor. sigs emia Position. Cases. ' Aver- Long- | Aver- |Short-} Long- age. Least. est. age. est. est. H. m.\H. m.|H. m.\H mH m.|H, m. BU Ge Cae pe eS anaes Chan wemene 614} 9 01:1 00); 48 00,0 520 10/4 00 ON “hg gaa See en grape nrc ae em 75|12 05/2 00| 48 00) 1 02/0 15 | 8 00 Oe ee 5| 23 84|6 40/60 00/1 33]... eal ta 06 Po oe eee 13 | 28 55/4 20/72 00/1 31/0 20/3 00 Os ek a a ee ene 6/27 14/|7 00! 72 00 1 40/0 40 | 3 00 Breech presentation. .-..----- we aie Sosa 7/10 38'}8 00/15 00/1 18 0 20 | 0 20 From Table IX if is evident that, as in women of other coun- tries, the duration of labor and of the second stage in anterior vertex presentations is shorter than in the posterior or trans- verse vertex presentations. Moreover, the length of labor in left occipito-anterior position is shorter than the duration of nor- mal labor as taught in most American obstetrical text-books. The data in Table X indicate that the frequency of the dif- ferent foetal presentations is much the same as in other coun- tries; only that, in vertex presentations, there is a relatively high percentage of those which occur in the left, those occurring in the right being correspondingly diminished. The twenty-seven Cesarean sections performed in the obstet- 268 The Philippine Journal of Science 1919 TABLE atheimapee of various presentations and positions in the 780 cases of multipare. Number. | Per cent. Vertex presentation: eS Os ee aw ca a ee + ei ge Hue a Sone Rekapeneee esas 6384 |} 84 R.O.A eee re ol ay ean ek cages See Fre goer ead 82| 10.9 Tc OB canes is ee ied SR QA ene owned aera oneness 7) 0.98 OS Se eee Powe ices eens owas Ae 17 2.26 ROe ee ae a ee ees 9 1,19 Da God eden it she ete bee hee eee epee 3 0.39 Total ois sds solia Ena See A 752 | 96.41 Face prosantation .c. <2... on Suse we ee eee eee te 2 0. 256 Transverse position --.-.---...----.------+-- ---+---2 ---2-- 22-25-55 ene sas 6 0. 76+ Breech presentation. ....---. 2-2 -osne Jae ease SGuiehwsdecseestaen 20; 2,36+ 3 TABLE XI.—Operations in multipare. Average length — Operation. Cases. Position. Cases. Second Labor. stage. H. m.|H. m. High forceps __.__--- 2 es oe i Os. CU SS ee 1 |[27 | 4 30 Ri Be a Ses eee ee 4 s R. 0. A enw 2 Mid forceps .....-.--| 12 L, O. P ---...----------------------------- 1 : “Ss ae re aueenoemencn acs =: 3 17 38,2 31 5 IES ig geet Sct one Pane eps aa eR AERC 1 See ore eC oec aaa seo os beque sous 1 Be ica ele enact Sasoe sxe 9 Ma MR gsc ew ckscebuncnnkwaca 2 Coe be a om pe er eee 26 86/8 12 Lem toteene BiLo.T ge Sr ia 2 | a er ait cioe cake eo 1 ‘Transverse position. ...._.......--...--... 6 L. 0. A., placenta previa_._..........---- 4 L, 0. A., vertex born with hand --------- 1 A L. O. A., uterine inertia, big head --._..- 1 Podalic version___-__ 20 115 80|1 40 < R. O.'A., placenta previa__._._.._-.--.-.- 4 NOW os eo an eae ie cease 1 nO: Pe a S hg OR as sa ae ps R.S.A 4 Breech extraction. _. Whee £2 g {fil 38) 0 47 : Lei P 3 XIV,3 Acosta-Sison and Calderon: Pelvimetry 269 TABLE XII.—Operations in primipare. Average length of— Operation. Cases, Position. Cases. Labor. — H. m.|H. m i Oc Be esti cccens 11 Bath A ie 1 yO og Wk BEF MEE Soo ae eels ee 21 L. O. P ...---------------- 2 = 2) Ss. $s Ar es ee ciaeee 3 Db Ost ee eS 2 i OST ne 2 Es OS Biv nnsg cSiagt aun 11 Ro ca cae a ges 1 Mid forceps---...-.------------------ Sih 0. Poe ai 1|$s2 34/4 09 fe-O. Tse eee 5 5 Bie 6 Bes gine tak atari sR 3 Podalic version... ..-..--..---.------ g [Py As. Ano --=-- oon . 5 15|2 30 i be Pee ee 1 Breech extraction -..---------------- 7 {L. S. A ------------------- 4 hie 05} 1 338 TR ae Gia eeearense ee 3 | rical department of the Philippine General Hospital and one other similar operation performed at another hospital (Roxas) are not included in this report, for only those cases whose babies were born alive per vaginam were measured. Moreover, in all Cesarean sections made on Filipino women, only on two occa- sions was the disproportion of pelvis and child the indication. One was a case of a primipara whose baby had an enormously large hydrocephalic head whose mento-occipital diameter mea- sured 19 centimeters; and the other, a case of osteomalacia (Roxas). The rarity of instances in which pelvic contraction, other than that caused by osteomalacia (which was found only once), can be an indication to Cesarean section is emphasized by the fact that of the twenty-seven Cesarean sections performed in the Philippine General Hospital prior to April, 1918 (Rustia) , twenty-four were performed on Filipinas for an indication (ex- cept in the case of the primipara with a hydrocephalic fcetus, if that should be excluded) other than the disproportion between — the foetal head and the pelvis or pelvic contraction. Whereas the only Cesarean sections performed on American private pa- tients (three cases) were necessary on account of dystocia caused by contracted pelvis. 164891——2 270 The Philippine Journal of Science 1919 CONCLUSIONS 1. The Filipino pelvis is of a type different from either the white American or the Negro pelvis, and its average normal measurements are similar to those of the generally contracted pelvis of the white American. 2. The index of the posterior pelvic plane in cases of con- tracted intertuberal diameter is important in the determination of the probable outcome of labor. 3. Contracted pelvis, except in cases of osteomalacia, is rarely an indication to Cesarean section among Filipino women. 4. There is practically no difference in the measurements of the pelvic diameters of multiparz and primapare. 5. The newborn babies of multiparze are longer and heavier than those of the primipare. 6. The babies of mothers who stayed in the hospital for one or more weeks before delivery are heavier than those whose mothers entered the hospital at the time of labor. 7. The male babies are in greater number and are longer and heavier than the female babies. 8. Labor is longer in primipare than in multipare. ACKNOWLEDGMENTS We wish to thank all the doctors and nurses of the obstetrical staff of the Philipine General Hospital for their cordial assist- ance in making it possible to measure the cases. We thank also Dr. M. Tracy for her kind help, and in particular we wish to extend our gratitude to Mr. H. O. Beyer for his valuable sugges- tions in the interpretation of many of our findings. REFERENCES Acosta-Sison, H. Pelvimetry and cephalometry among Filipinas. Philip. Journ, Sci. § B 9 (1914) 493-497. BEAN, RoBert B. The Racial Anatomy of the Philippine Islanders. J. B. Lippincott Company, Philadelphia (1910). See also various articles by this author in the Philippine Journal of Science and the American Anthropologist (1908-1915). Bayer, H. OTLEY. Population of the Philippine Islands in 1916. Part II. Philippine Education Co., Manila (1917). IpEM. Ethnographic grouping in the Philippines. Manila (1918). (An unpublished manuscript placed at our service.) BippLe, A. G. A new method of measuring the pelvic outlet. Am. Journ. Obstetrics (1912) 14. CoLE, Fay Cooper. The wild tribes of Davao district, Mindanao. Field Museum of Natural History, Chicago (1913) No. 170. See also various shorter articles by this author in the Philippine Journal of Science and the American Anthropologist (1908-1911). XIV, 8 Acosta-Sison and Calderon: Pelvimetry 271 Daniets, C. D. A new and original method of calculating the required posterior-sagittal diameter of the outlet in a lateral contraction of the pelvis. Am. Journ, Obstetrics, 74, No. 2. Dr Souza, D. H. The measurements of the pelvis with special reference to obstetrics prediction. Biometrika 9 (1918) 486. Epear, J. C. The Practice of Obstetrics. P. Blakiston Co. Emmons, A. B. A study of the variations in the female pelvis based on observations made on 217 specimens of the American Indian squaw. Biometrika 9 (1913) 34. Hirst, B. C. A Text Book of Obstetrics. W. B. Saunders Co. (1918). HuntincTon, J. The true value of certain pelvic measurements. Am. Journ. Obstetrics (1917) 76. Rust1a, G. Cesarean section—report of the cases performed in the service of the Philippine General Hospital. Memorias y Comunicaciones de la Cuarta Asamblea Regional de Médicos y Farmacéuticos de Filipinas (1918). Wituiams, J. W. Frequency, etiology and practical significance of con- tractions of pelvic outlet. Surg. Gyn. and Obstet. 8 (1909) 619. IpeM. The funnel pelvis. Am. Journ. Obstetrics 64, No. 403. IpEM. Obstetrics Text-Book. D. Appleton & Co. (1917). ILLUSTRATIONS TEXT FIGURES Fig. 1, Interspinous diameter (measured from tip). 2. Intercristal diameter (between upper borders of the iliac bones). 3. Baudelocque’s diameter. 4, Intertuberal diameter. 5. Diagonal conjugate, primipare and multipare. 273 THE LUMBANG-OIL INDUSTRY IN THE PHILIPPINE ISLANDS By R. H. AGUILAR (Bureau of Science, Manila) The manufacture of lumbang oil in a crude way antedates existing records. It has been mostly carried on by Chinese, who have made no attempt to improve the methods of produc- tion, and the oil that they express is dark in color and of a disagreeable odor. The oil industry in the Philippine Islands is undergoing a radical change, for the manufacturers have adopted the motto of larger production and better quality. Some of the large local concerns, hitherto devoting themselves exclusively to the man- ufacture of coconut oil, are now turning their attention to other oils, among which lumbang holds an important place. There are several points to be taken into consideration, if the quality of the latter is to be improved and the industry placed on a commercial basis. The object of this paper is to discuss some of these points with a view to more profitable production. There are two kinds of lumbang nuts known in the Philippine Islands from which oils are obtained; lumbang bato (Aleurites moluccana)* and lumbang banucalag (Aleurites trisperma). Lumbang bato is of wider distribution, more abundant, and better known than lumbang banucalag, and the attention of manufacturers at present is devoted to the production of oil from the former. Lumbang banucalag is almost unknown in the Manila market, and when the word “lumbang” is employed it is generally taken to mean lumbang bato. The Chinese, who are the largest dealers in lumbang oils,? are not willing to handle the banucalag variety, because it is supposed that contact causes skin eruptions;* but, in my opinion, there is no founda- tion for this belief. ? Richmond, G. F., and Rosario, M. V., Philip. Journ. Sci. § A 2 (1907) 441-443. ?For further data see Brill, H. C., and Agcaoili, F., Philip. Journ. Sci. § A 10 (1915) 118. me * Aguilar, R. H., Philip. Journ. Sci. § A 12 (1917) 236. 275 276 The Philippine Journal of Science 1919 The supply of banucalag nuts is irregular and until a definite supply can be depended upon, the extraction of oil from these nuts would hardly be profitable as a separate industry; but no doubt it could be carried on profitably in connection with the extraction of lumbang bato oil. The close similarity * between lumbang banucalag oil and Chinese wood oil, and the fact that the latter is so highly appreciated® by United States paint manufacturers, constitute inducements for the production of banucalag oil. Both varieties of lumbang fruits mature and drop from the trees in June and July, and the best time for gathering them is during July and August. Information obtained from local Chinese merchants indicates that the gathering of lumbang nuts could be. much extended in Mindanao, more especially in the Davao Gulf district, where the nuts are known by the name “biao,” and where now large quantities go uncollected. Heretofore, merchants have hesitated to invest capital in the lumbang-oil industry for they believed insufficient nuts could be obtained to warrant the establishment of a factory. On the other hand a Manila firm in September, 1910, wrote with regard to this oil: With proper attention the Philippines would be in a position to export the oil profitably. Proper machinery installed near San Pablo, Laguna, would make profitable a large export business, for there seem to be sufficient lumbang seeds that could be secured from Laguna, Tayabas and Batangas provinces. An estimate of the yearly Philippine crop of lumbang nuts is very difficult to make. It would not be legitimate to base such on the annual production of lumbang oil, because the por- tion of the crop now used for oil manufacture is very small. In 1911 one manufacturer claimed that there was a sufficient available supply with proper machinery to produce 5,000 kilo- grams of oil per day; this amount would represent about 9,000 kilograms of kernels or about 26,000 kilograms of dry unshelled nuts. On this basis the available nuts would supply a modern oil factory with a monthly capacity of 234 metric tons of nut kernels. The Bureau of Forestry is encouraging the planting of lum- bang trees throughout the Archipelago, and it is estimated that approximately a half million trees are being planted annually. Therefore, an increasing supply of raw material for the man- ‘Richmond, G. F., and Rosario, M. V., Philip. Journ. Sci. § A 2 (1907) 443. * Drugs, Oils, and Paints 30 (1914) 207. tv XIV, 3 Aguilar: The Lumbang-Oil Industry 277 ufacture of lumbang oil is assured. The number of lumbang trees in the Islands is not definitely known, but the information at hand would seem to indicate that it is possible to secure an abundant supply of lumbang nuts and to make of the lumbang- oil industry a profitable business. In the purchase of lumbang nuts care should be taken to base the price upon the dry nuts. Freshly gathered nuts, especially lumbang banucalag, contain from 10 to 22 per cent of moisture. The moisture content can be estimated as follows: For lumbang bato, take about 1 kilogram or more of repre- sentative nuts, and place them in an oven at 90° C. to 95° C. for from two to three hours. For lumbang banucalag, the same quantity may be taken, but the temperature should be carefully regulated between 65° C. and 70° C. for ten hours or more. This is extremely important, as the kernels of the banucalag nuts harden slightly and assume a darker color at a higher temperature. Lumbang bato nuts may be stored for a year or more in a cool dry place without their undergoing an appreciable change in the amount or the composition of the oil; but lumbang banu- calag nuts stored for the same length of time, probably due to oxidation, will be greatly affected in both the amount and the composition of the oil, with an increase in its acidity. No mold or fungus was observed on any of the nuts stored. Due to the comparatively thin shell the kernels of lumbang banucalag nuts can be easily separated from the shell after cracking; but considerable difficulty was experienced in trying to find some adequate means of separating the kernel from the shell of lum- bang bato nuts; nuts gathered over a year ago showed, when . opened, that the kernel still adhered firmly to the shell. Steaming has been used successfully with certain varieties of coconuts, but is unsuccessful in this case and, furthermore, darkens the kernel and the oil. The commonest procedure now in use is to crack the nuts and pick out the kernels by means of a pointed instrument, a very tedious operation. LUMBANG BATO (ALEURITES MOLUCCANA) Biau (Misamis, Davao); lumbang (Rizal, Laguna, Zamboanga, Batan- gas); lumbang bato (Cavite). Lumbang is the name officially adopted by the Bureau of Forestry. The fruit of Aleurites moluccana is fleshy, ovoid, dark, dull green, 5 to 6 centimeters long, and contains one or two seeds. The seed is about 3 centimeters long and 2.5 centimeters broad. It has a very hard, rough, rigid shell about 2.5 millimeters thick. 978 The Philippine Journal of Science 1919 This contains a white, oily, fleshy kernel consisting of a very thin embryo surrounded by a large endosperm. This is in turn covered by a thin, white, papery seed coat. This thin seed coat adheres firmly to both the shell and the kernel, so that the kernel is separated from the shell with difficulty. The importance of lumbang bato oil has been known since 1873 when Prof. Julius Wiesner,* reporting on The Foreign Plant Stuffs in Industrial Use at the Vienna Universal Exhibition in that year, said: A source of oil not hitherto regarded by the European oil industry is the Bankul nut (from Aleurites triloba), which deserves to be brought into use, on account not only of the low price of the raw material, but also from the high quality of the oil. In the Philippines the oil is now being used in the caulking of vessels, in the manufacture of soft soaps, and as a substitute for linseed oil in the local manufacture of paints; also there is a demand much in excess of the present production for this oil from United States paint manufacturers. Therefore, the great value of the oil for industrial purposes is now well established, and its extraction based upon a more systematic method should be carefully considered. . Among the various procedures now in use for the purpose of minimizing the work involved in removing the kernel from the shell of lumbang bato nuts, none of which has been found wholly . Satisfactory, are the following: In some localities the Chinese place large quantities of nuts on the ground, cover them with straw and after burning the straw immediately sprinkle the nuts with cold water. They claim that . with this method the nuts burst. In Laguna, Tayabas, and Ba- tangas Provinces, the nuts are placed in tanks of boiling water and left there for from five to six hours. This loosens the kernel, and when sufficiently cool the nuts are cracked and the kernels are separated from the shells. These two methods produce brown kernels from which only brown oil can be expressed. In Moro Province, along the coast of Davao, the nuts are dried in the sun until the kernels loosen sufficiently, which may be ascertained by occasionally cracking a few nuts for trial. The drying takes from five to ten days or more, depending upon the condition of the weather; the nuts are then cracked and the kernels removed. This process is very slow, although the kernel usually comes out whole and is of the best quality. “Andes, L. E., Vegetable Fats and Oils, 2d ed. Greenwood and Co., London (1902) 6. XIV, 8 Aguilar: The Lumbang-Oil Industry 279 The method that has been developed in the Bureau of Science laboratory reduces somewhat the time and the labor involved in the preparation of the kernel and has no injurious effect on the oil. The nuts were heated in an oven at 95° C. for from three to four hours, dumped rapidly into cold water, and left over- night. The next morning the shells had burst, and the kernels were separated without difficulty. To eliminate the time and labor involved in removing the kernel from the shell, it has been suggested to crush and grind the nuts, and then express the oil from the ground mixture of shell and kernel. This procedure has been tried in several places and seems to be reasonable from the standpoint of economy of time and labor; however, I am more inclined to favor the method of extracting the oil direct from the kernel, because (a) about 20 kilograms more oil per ton of nuts can be recovered from sepa- rated kernels than when the shell is ground with them, and (b) the fertilizing value of the cake, which Chinese manufacturers now sell at a good profit, will be very much reduced when shell is included in it, as will be shown later. An idea of the commercial advantage of either of these methods over the other can be obtained by comparing the product of one factory utilizing kernels with that of another factory utilizing nuts for the expression of oil. Before describing these two methods of procedure, it is inter- esting to note the variation existing in the kernel content of the different lots of lumbang bato nuts from Cavite Province. This variation is probably due to the age of the nut-bearing trees and the length of time the nuts lay uncollected on the ground. TABLE I.—Moisture content of nuts and percentage of kernels referred to the weight of dry nuts. ‘4 Kernels | Fresh | Moist-| Dry | from |Kernels nuts. wate nuts. dry tate : nuts. g. Pict. g. g. P, wt; 8,000 19| 2,430 845 | 34.77 2, 000 22| 1,560 561 | 35.96 2,000} 676 | 83.80 Steere, Caters 2, 000 652 | 32.60 2, 000 685 | 34.25 2, 000 10} 1,800| 589 | 32.73 DP VOPEGS io. ca etalon eccse $4. 02 In November, 1918, the price in Manila of lumbang bato nuts was 50 pesos per ton; kernels, 244 pesos per ton; and oil, 8.66 280 The Philippine Journal of Science 1919 pesos per 5-gallon gasoline can. The average weight of 18.927 liters (5 gallons) at ordinary temperature (30° C.) is 17.33 kilograms. Using the above data as a basis of calculation the following results are obtainable for 1 metric ton of kernels: (1) Factory utilizing nut kernels for the expression of oil. Pesos. Yield, 560 kilograms of oil, expressed at a pressure of 800 kilograms per square centimeter; selling at 8.66 pesos per can 279.83 Yield of press cake, 440 kilograms; selling at 25 cen- tavos per kilogram 110.00 Total 389.83 Cost of 1 ton of kernels 244.00 Gross profit 145.83 (2) Factory utilizing whole nuts crushed for the expression of oil, Pesos. Yield 501.24 kilograms of oil, 20 kilograms of oil per ton of nuts less than in the former process, at a pressure of 800 kilograms per square centimeter; selling at 8.66 pesos per can 250.47 2.988 tons of nuts (based on the average given in Table I), necessary to produce 1 ton of kernels, at 50 pesos per ton 146.90 Gross profit 103.57 The fertilizing value of the cake from this process is very much reduced, as is shown in Table III. It is doubted if a market for it can be obtained; furthermore, the oil from this process, while the same chemically, is relatively dirtier and harder to clarify than is that from the former. The above results conclusively show the advantage of the first method over the second from a commercial standpoint, especially until a definite market is established for the cake from the latter. As many factors enter into any calculation of the cost of opera- tion, it is difficult to furnish approximate and satisfactory data for the two processes. If the two factories were to operate in the same locality, under identical conditions, there is no doubt that the cost of production for the first factory would be rela- tively less. One large item of expense would lie in the fact that a factory utilizing nuts would need heavier crushers and larger presses and consequently more power, besides having to handle a greater bulk of material, thus lowering the daily output of oil and the general efficiency of the whole system. To form an idea of the quality of the lumbang kernels sold XIV, 3 Aguilar: The Lumbang-Oil Industry 281 in the market at prices ranging from 18 to 19 pesos per 75 kilograms, four different samples were secured from which the oil was expressed and analyzed; the results obtained are presented in Table II. TABLE II.—Quality of kernels sold in Manila. Oil. Sample No. Kernel color. Res | ased on Acid kernel. Color. value, P. ct fe RAB na ing ee Rat A SANG oe ee 55 Light colored ._.._---- 27.63 eaten ie aaa caus Disa ty g1 ereing anmeeenr ones 50 | Light brown--_.------- 224,75 a ee aes Oe eclacnkn RE RE OER ESC Bi IN ee ee 266. 53 Dts oe ca hema a ee Gh. ccs ve ese G8 42S C6 Se 272. 28 «Milligrams KOH used to neutralize 1 gram of the oil. Table III shows an analysis of oil expressed from fresh kernels as compared with two grades of oil obtained from the market. TABLE III.—Constants of three samples of oil. Oils obtained from Oil expressed the market. from fresh kernels. GradeI.| Grade II. UAH. oe ease a ne wcones light colored | brown | dark brown Spetite wravity at 1b, 60 Co ss os oc 0.9261 | 0.9253 0, 9237 Saponification value ___._____. aig See ee gah Cee eR A 188 198 194 elie inner oe oe ee a eo Ca seacte ae 154 157 160 FO chet 2 aetna Stair Vinee etn 7 aie ein Cea vem per een 0.55 64. 25 106. 48 The results shown in Table III indicate clearly that the pre- paration of the kernel is one of the most important points to be taken into account in the manufacture of lumbang oils. Pro- longed heating at high temperatures produces brown kernels, and consequently a brown oil that is high in free fatty acid, as shown by the acid values of samples 2, 3, and 4, in Table Il. Long storageof the kernel also causes deterioration in the value of its oil content; for it is readily attacked by small black beetles, thus reducing the available amount of oil; and the oil content becomes more and more acid without any perceptible change in the ap- pearance of the kernel or in the oil extracted. Sample 1, Table II, has the same appearance as the kernel prepared in the labor- atory. Freshly prepared kernel, therefore, is the best material for the expression of oil. To determine the change in the amount of free fatty acids of 282 The Philippine Journal of Science 1919 the oil content of kernels due to storage, a sample of fresh kernels was secured and divided into two portions. The first portion was expressed immediately and gave an oil with an acidity of 0.55. The second portion was stored in a cool, open, dry place for a month, when the oil was expressed, and found to have an acidity of 5.32. The fertilizing value of the cake from the kernels as well as that of cake obtained from crushed nuts are shown in Table IV. TABLE 1V.—Fertilizing value of lumbang bato cake from kernels and crushed nuts. Cake Cake from from crushed kernels,4 note Per cent, | Per cent. PADIMEUROS boos cases doa ose oa ee ee ee 11.138 8.46 Witrowen (N4@) oo5. oe ccc ois.csk kena Ss ee 8. 86 1,25 Potash (is0) oS pinkdas Sac eeee clue ee ee 1.67 0. 68 Phoapnorus (PsO8) ois: 3 2h oe ie ee ae he 1,02 0. 25 * Price in November, 1918, 25 centavos. per kilogram. >» No market established. LUMBANG BANUCALAG (ALEURITES TRISPERMA) Baguilumbang, balucalad (Laguna); banucalag, lumbang banucalag, lumbang gubat (Cavite); balucanag (Batangas); lumbang (Oriental Ne- gros, Camarines). Baguilumbang is the name officially adopted by the Bureau of Forestry. The fruit is 5 to 6 centimeters in diameter, somewhat rounded and obscurely angled, usually 3-celled, the cells 1-seeded, tardily dehiscent. The seed is somewhat ellipsoid, slightly flattened, brown, rather smooth, and has a rather brittle shell about 0.5 millimeter thick. This contains a white, oily, fleshy kernel, con- sisting of a very thin embryo surrounded by a large endosperm. This in turn is covered by a thin, white, papery seed coat. When dry the kernel with the thin seed coat shrinks slightly away from the shell so that the shell and the kernel are easily separated. The close similarity between Chinese wood oil and lumbang banucalag oil’ and the important place Chinese wood oil-is now occupying in the paint industry of the world,* clearly show the industrial possibilities of banucalag oil. The kernel of lumbang banucalag nuts is not so difficult to separate from the shell as is that of lumbang bato. When the oe Richmond, G. F., and Rosario, M. V., Philip. Journ. Sci. § A 2 (1907) * Wilson, E. H., Bull. Imp. Ins. 11 (1913) 18, 454. XIV, 3 Aguilar: The Lumbang-Oil Industry 283 nut is dry the kernel shrinks somewhat, and it can be easily separated after cracking the shell. No banucalag kernels are being sold in the market, and the industry of lumbang banucalag at present is very limited. This may be due to two important reasons, namely : (1) The banucalag kernels when stored undergo a change, probably due to partial oxidation, as the result of which the oil becomes much more difficult to express and assumes a darker color. (2) The banucalag oil when expressed, unless kept in air-tight containers, becomes rancid and foul smelling. These facts constitute the main drawbacks to the production of lumbang banucalag oil; but, if certain precautions are strictly followed, there is no doubt that the industry of lumbang banu- calag oil in connection with lumbang bato oil may become a profitable enterprise. Freshly gathered nuts, free from moisture, produce kernels of the best quality, and the yield of oil by expression at 800 kilograms per square centimeter may be as much as 56 per cent of the weight of the kernels. This amount represents about 35 per cent of the weight of the dry nuts, whereas the oil obtained from lumbang bato kernels represents only about 19 per cent. The banucalag oil so prepared is of very good quality, light amber in color, and altogether satisfactory in its physical and chemical properties. If the oil is placed in air-tight containers while fresh, it will keep longer and no appreciable change in its composition will be noticed. On the other hand, unshelled nuts that were kept for sixteen months underwent so great a change in their oil value that the yield by expression was reduced from 56 per cent to 40 per cent of the weight of the kernel, and the oil was high in free fatty acids and much darker in color. The results obtained are shown in Table V. TABLE V.—Constants of banucalag oil. ——] oe Oil expressed Oil expressed! ing kept pen phew from the ker-} for 16 stored ta & anak nels of fresh! months in a cool. dey nuts. in a well- ie for 16 stopper-| PRY the. led bottle. ™on Per cent of oil based on the weight of kernels._..___....- 56 Siopeiics 40 BB 1 RR eas I DE Pt NSIC arti ae ce ae rE TNR light amber | amber | dark amber meets eravity at 16,67 O20 0.9362 | 0.9885 0, 9441 Hehonmoktion Gales 191 192 200 Ona Wilt 5 a 166 170 160 MOUS Wea EO es Sie he ean ete 2,22 5.51 49.14 284 The Philippine Journal of Science 1919 The results shown in Table V demonstrate conclusively the advisability of expressing the oil from the kernels of freshly gathered nuts, and of avoiding as much as possible long storage. These precautions should be strictly followed to ensure the pro- duction of a clear, light-colored oil, low in free fatty acids. The properties that make this oil so sensitive are the ones that make it valuable. The excellent drying qualities of this oil have been already reported.® Lumbang. banucalag nuts may be crushed and finely ground in an oil mill and the oil expressed directly from the crushed nuts. However, this procedure is even less commendable than in the case of lumbang bato; in the first place, the oil thus ob- tained is dirty, highly contaminated with shell particles, and darker colored; in the second place, the acid value is relatively higher, as shown in the following test: From one sample lot of banucalag nuts two samples of oil were taken; in one case the oil was expressed from the kernel after separating the shell, and in the other case from the crushed nuts without separating the shell from the kernel. After one week the sedimentation by gravity was complete, and the two samples of oil were clear. The acid value of the first was 3.65; that of the second, 10.65. The better quality of oil obtained when extraction was made > from the kernels fully justifies the relatively small amount of additional work required to separate the kernel from the shell of lumbang banucalag nuts. Clear oil may be obtained, either by sedimentation or through filter presses, and kept in air-tight containers for storage. The fertilizing value of lumbang banucalag cake compares very favorably with that of lumbang bato cake, as will be seen by a comparison of Tables VI and IV. TABLE VI.—Fertilizing value of lumbang banucalag cake from kernels and crushed nuts. Cake | fromm from | crushed kernel. itis Per cent.| Per cent. Molature 22.25... 2 SR yee nat Ey ete reas REN a 1.67 9.45 Nitrogen (Na) 222522622 a ee eee ee 6.20 2.99 Potash (E20)... ee 1.79 0.90 Phosphorus (POs). ce eae ani 1.18 0.95 ° Aguilar, R. H., Philip. Journ. Sci. § A 12 (1917) 235. XIV, 3 Aguilar: The Lumbang-Oil Industry 285 SUMMARY AND CONCLUSIONS The lumbang oil industry, if properly developed, will become a profitable business. It is much less desirable for a factory devoted to the manu- facture of lumbang oils to purchase kernels than the nuts from which to prepare its own kernels, because oil from kernels that are being sold in the market is almost always of lower grade than that from freshly prepared ones. No appreciable change can be noticed in the available oil con- tent of uncracked lumbang bato nuts stored for over a year; but the content and the quality of the oil of unshelled banucalag nuts stored for the same length of time is considerably reduced. In the latter case even the shell is insufficient to protect the oil in the kernel from oxidization by the air, and shelled kernels dete- riorate very rapidly. Both oils after expression may be kept satisfactorily in proper containers. In view of these facts, a lumbang-oil factory en- gaged in the commercial expression of both bato and banucalag oils, in order that the maximum yield and best quality of both bato and banucalag oils may be obtained, preferably should com- mence the season with the expression of banucalag oil and store the bato nuts until the other species is finished. 164891——-3 NEW PHILIPPINE GALL MIDGES By E. P. Frew.r (State Entomologist, Albany, New York) This paper is supplemental to an earlier one;* it not only de- scribes and records the food habits of a number of Philippine gall midges, but also establishes the occurrence in the eastern tropics of Ctenodactylomyia Felt or a closely related genus, previously known only in subtropical America. This was to be expected, since studies on distribution show that certain sub- tropical genera have a range which would suggest their probable occurrence in all warmer climates where food plants permit their existence. This record is analogous to the discovery of species of the genus Aplonyzx in such widely separated parts of the world as the Mediterranean region and the vicinity of Salt Lake, Utah—localities apparently agreeable to the host plants as well as to the insects. This collection, like the preceding, was received through the courtesy of Prof. Charles S. Banks, chief of the department of entomology of the College of Agriculture, University of the Phil- ippines, who collected some of the species, as detailed below. Mr. L. B. Uichanco reared a number of species from various galls; the galls, I understand, are to be described in detail in another paper. Ctenodactylomyia antidesme sp. nov. Female.—Length, 1.5 millimeters. Antennz extending to the base of the abdomen, sparsely haired, reddish brown, the basal segments reddish orange, probably of 14 segments, the fifth cylin- drical, its length two and one-half times its diameter. Palpi presumably triarticulate, first segment irregularly and broadly oval, second a little longer, slenderer, third three times as long as its width and somewhat dilated. Eyes black, holoptic. Mesonotum pale yellowish, sparsely clothed with coarse sete. Scutellum translucent yellowish. Postscutellum yellowish orange. Abdomen reddish orange, distal segments rather thickly clothed *Felt, E. P., New Philippine gall midges, with a key to the Itonidide, Philip, Journ. Sci. § D 13 (1918) 281. 287 288 The Philippine Journal of Science 1919 with coarse, dark sete or scales. Wings hyaline, their length about twice their width, third vein uniting with margin well be- fore apex, fifth vein obsolescent distad and joining margin at distal fourth, its branch near basal third. Membrane rather thickly clothed with dark hairs. Halteres whitish transparent basad, fuscous apicad. Coxe and femora basad yellowish orange. Femora distad and tibiz dark straw. Tarsi mostly fuscous, the anterior pair with narrow, indistinct, yellowish annulations at articulations. Claws quadridentate, there being at least three minor stout teeth nearly as long as, though more slender than, the main claw. Pulvilli about one-half length of claws. Ovi- positor about one-half length of abdomen. The terminal lobes five times as long as wide and sparsely setose. Type.—Cecid. a2881, New York State collection. LuZON, Laguna Province, Mount Maquiling, October 2, 1917, College of Agriculture accession No. 18157 (L. B. Uichanco). The single female, described above, was reared from leaf galls on Antidesma leptocladum Tul. The species is tentatively placed in this genus, though the wings are decidedly broader than in the type species, and there is a possibility that it should be referred to a new genus. Such action is deferred until both sexes can be secured. Lasioptera falcata sp. nov. Female.—Length, 1.5 millimeters. Antenne extending to near base of abdomen, sparsely haired, dark brown, of 24 segments, the fifth as long as its diameter. Terminal segment slightly produced and narrowly oval. First segment of palpi short, irregular; second about three times as long as its diameter; third nearly one-half longer, slenderer; fourth a little longer and slenderer than third. Mesonotum reddish brown, apparently denuded. Scutellum reddish orange. Postscutellum dark brown. Abdomen mostly dark brown, first segment margined caudad with a narrow white line, the others though badly rubbed show submedian whitish spots; terminal segment yellowish. Wings hyaline, third vein uniting with costa near distal third. Halteres yellowish. Coxz yellowish orange. Legs mostly dark brown. Ovipositor nearly as long as body, terminal lobe about four times as long as wide, slightly curved, sparsely setose, and with an oval group of heavy, stout, hastate spines basad and a group of rather long, moderately slender, chitinous hooks. Type.—Cecid. a2887, New York State collection. Luzon, Laguna Province, Los Bafios, January 3, 1917, College of Agriculture accession No. 18183 (Uichanco). XIV, 3 Felt: New Philippine Gall Midges 289 The single female, reared from a stem gall on a wild cucurbit, is presumably undescribed, and there is little question that it is the producer of the gall. The insect is remarkable because of the unusually long, somewhat falcate, terminal lobes of the ovipositor. Asphondylia grewie sp. nov. Male.—Length, 1.5 millimeters.. Antennz probably nearly as long as body, sparsely haired, basal portion, at least, dark brown, presumably of 14 segments, fifth with a length four times its diameter and with moderately high circumfila. Palpi probably triarticulate. Mesonotum dark reddish brown. Scutellum and postscutellum yellowish brown. Abdomen dark brown, sparsely haired. Wings unusually broad and rather thickly clothed with fuscous hairs. Third vein unites with margin near apex, fifth at distal third, its branch at basal fourth; fork unusually long. Halteres fuscous yellowish. Legs dark brown. Claws long, moderately stout, pulvilli as long as claws. Genitalia: Basal clasp segment short, stout, and bidentate, with a distinct rudi- mentary tooth near the bottom of the excavation; dorsal plate divided, the lobes irregularly triangular, lobes of ventral plate’ rounded. Type.—Cecid. a2880, New York State collection. LUZON, Laguna Province, Mount Maquiling, September 23, 1917, College of Agriculture accession No. 18137 (Uichanco). The one male described above was reared from a leaf gall on Grewia stylocarpa Warb. The species is remarkable in this genus because of the* unusually broad wings, these having a length nearly one-half greater than the diameter. The specimen was badly broken and rumpled. Contarinia saltata sp. nov. Male.—Length, 1.25 millimeters. Antennz one-fourth longer than body, thickly haired, reddish brown, of 14 segments, fifth having stems with a length equal to, and one-half greater than, their diameters, respectively. Basal enlargement subglobose, distal enlargement somewhat produced, both with moderately long circumfila; basal portion of terminal segment with the stem short, distal enlargement irregularly fusiform and tapering slightly to a very broad, broadly rounded apex. Length of first. segment of palpi about two and one-half times its diameter; second a little longer, slenderer; third a little longer than second; fourth as long as third and somewhat dilated apicad. Mesonotum dark reddish brown. Scutellum reddish orange. Postscutellum 990 The Philippine Journal of Science 1919 dark brown. Abdomen reddish brown. Wings hyaline, moder- ately narrow, third vein uniting with margin at apex. Halteres yellowish basad, reddish apicad. Coxe and legs mostly reddish brown. Claws long, slender, strongly curved apicad, pulvilli as long as claws. Genitalia: Basal clasp segment very short and stout; terminal clasp segment moderately long and stout, dorsal plate short, broad, narrowly incised, lobes obliquely truncate, lateral angles broadly rounded; ventral plate deeply and trian- gularly emarginate, lobes somewhat divergent and tapering to a narrowly rounded vertex; style moderately long, tapering strongly, narrowly rounded apicad. Type.—Cecid. a2884, New York State collection. LuZzON, Laguna Province, Los Bajios, February 6, 1918, College of Agriculture accession No. 18180 (Uichanco). A series of specimens was taken from a spider’s web on a leaf of Andropogon sorghum Linn. It is possible that this insect lives upon Andropogon, though it is very different from Con- tarinia sorghicola Coq., which is well known in America. It is described here because it is believed that the peculiar characters will make the recognition of this species easy and, therefore, no confusion need arise from a description not based in part upon a food-plant record. Bremia macrofilum sp. nov. Male.—Length, 1 millimeter. Antenne one-half longer than body, thickly haired, dark brown, of 14 segments, fifth with stems having a length one and one-half and two and one-half times their diameters, respectively. Basal enlargement sub- globose, distal enlargement oval, its length one and three-fourths times its diameter. Dorsal loops on basal circumfilum produced nearly to distal enlargement and dorsal loops of distal circum- filum extending to apex of basal enlargement of next segment. Terminal segment greatly produced, basal portion of stem with a length about five times its diameter, distal enlargement pyri- form, its basal third rather strongly constricted. Appendage long, irregular, and slightly fusiform. First segment of palpi with a length about three times its diameter; second a little long- er, stouter; third as long as second; fourth one-half longer than third and slenderer. Mesonotum dull dark brown. Scutellum yellowish brown. Postscutellum dark brown. Abdomen dark brown, almost black, sparsely haired; genitalia yellowish. Wings hyaline. Halteres yellowish basad, reddish apicad. Coxe dark brown. Legs mostly reddish brown. Claws moderately long, slender, strongly curved, the anterior ones unidentate, pulvilli XIV, 3 Felt: New Philippine Gall Midges 991 rudimentary. Genitalia: Basal clasp segment moderately long, stout ; terminal clasp segment long, slender, swollen basad; dorsal plate deeply and triangularly emarginate, the lobe sparsely setose, tapering to a narrowly rounded apex; ventral plate apparently very long and slender and with a somewhat curved, quadrate, obliquely truncate, lateral process at the basal third, the length of which is approximately one-third that of the entire organ. The distal portion of the ventral plate beyond this process is slender and a little longer than the basal part; harpes chitinized, indistinct; style rather short, stout, obtuse apicad. Female.—Length, 1.5 millimeters. Antenne a little shorter than body, sparsely haired, dark brown, of 14 segments, fifth with a stem one-fourth the length of the cylindrical basal enlarge- ment. The latter is two and one-half times as long as its diam- eter. It bears a sparse subbasal whorl of long, stout sete and a median band of shorter, slender, curved setz. Circumfila unusually heavy, terminal segment produced, length of basal portion four times its diameter, appendage knoblike. Mesonotum dark brown, submedian lines sparsely haired. Scutellum yel- lowish fuscous. Postscutellum dark brown. Abdomen mostly dark brown, basal segments with a dark orange hue. Legs mostly dark brown, fourth tarsal segment of anterior legs and distal tarsal segment of middle pair of legs white in certain lights. Ovipositor short, the lobes roundly triangular and sparsely setose, otherwise as in the male. Type.—Cecid. a2888, New York State collection. MINDANAO, Lanao District, Kolambugan, June 10, 1914, College of Agriculture accession No. 18184 (C. S. Banks). The midges described above were accompanied by the statement that they were caught on a spider’s web. They were dancing on the web, by the thousand, on a bright sunshiny afternoon. This peculiar species has unusually heavy circumfila in the female, and the male genitalia present striking characteristics. Arthrocnodax copre sp. nov. Male.—Length, 0.75 millimeter. Antenne a little longer than body, thickly haired, reddish brown, of 14 segments, fifth having stems with a length one-half and one and one-half times their diameters, respectively. Basal enlargement subglobose, distal broadly ovoid, both with moderately long and unusually heavy circumfila; basal enlargement of terminal segment subglobose, stem relatively short, distal enlargement broadly cylindrical, apex almost truncate. First segment of palpi short, irregular; second with a length three times its width; third a little shorter 292 The Philippine Journal of Science Ses than second, more dilated; fourth one-half longer than second. Mesonotum shining dark brown. Scutellum and postscutellum dark reddish brown. Abdomen yellowish fuscous. Wings hya- line, third vein uniting with costa well before apex; fifth obso- lescent distad, joining posterior margin at distal fourth, its branch at basal third. Halteres translucent whitish basad, fus- cous apicad. Coxe dark brown. Legs mostly yellowish brown. Claws rather long, evenly curved, slender, pulvilli rudimentary. Genitalia: Basal clasp segment moderately long, stout; terminal clasp segment rather long, swollen basad, dorsal plate deeply and triangularly emarginate, lobes triangular and sparsely setose; ventral plate moderately long, broad, broadly rounded apicad; style long, stout, and tapering to a narrowly rounded apex. Female.—Length, 0.75 millimeter. Antenne a little shorter than body, rather thickly haired, yellowish brown, of 14 seg- ments, fifth with a stem about one-half the length of cylindrical basal enlargement, the latter about three times as long as its diameter. First segment of palpi short, irregular; second some- what dilated, its length over twice its diameter; third one-half longer than second, slenderer; fourth a little longer than third and more dilated. Mesonotum dark reddish brown. Scutellum yellowish, darker basad. Postscutellum yellowish. Abdomen deep red. Halteres yellowish basad, fuscous apicad. Legs mostly yellowish fuscous, distal tarsal segments somewhat darker. Ovipositor short, lobes with. a length two and one-half times width and sparsely setose. Type.—Cecid. a2883, New York State collection. The coloration in this species is evidently variable, another male having a shining reddish brown mesonotum, yellowish scutellum and postscutellum, and dark reddish abdomen. A second female had a shining, very dark brown, almost black mesonotum. LuzON, Laguna Province, cs Bafios, December 6, 1917, College of Agriculture accession No. 18173 (C. S. Banks). A series of midges was reared from copra (dried coconut meat) kept in the laboratory for breeding Necrobia rufipes de Geer. The glass vessel had not been opened for a month. The charac- teristic short stems of the flagellate antennal segments and the heavy circumfila lead me to place this species here, though the poorly developed pulvilli would indicate an affinity with Silves- trina Kieff. or Planodiplosis Kieff., the latter known only by the female. It is probable that the larve are scavengers, as is the case with A. aphiphila Felt and some other American species. XIV,3 Felt: New Philippine Gall Midges 293 Itonida paederie sp. nov. Male—Length, 1 millimeter. Antenne probably twice the length of body, thickly haired, pale yellowish, probably with 14 segments, fifth having basal portion of stem with a length one-fourth greater than its diameter, distal portion three times as long as its diameter; basal enlargement subglobose, terminal enlargement subcylindrical, somewhat expanded distad and with a length one and three-fourths times its diameter, circumfila moderately long, terminal segment missing. First segment of palpi moderately long, irregular, length of second three times its diameter; third one-half longer and slenderer than second; fourth as long as third and somewhat dilated. Scutellum reddish brown. Postscutellum and abdomen mostly yellowish brown, abdomen thickly clothed with long hairs. Wings hyaline, mem- brane rather thickly haired, third vein uniting with costa just beyond apex, fifth obsolescent at the distal third, its branch at the basal third. WHalteres translucent yellowish. Legs mostly dark straw. Claws moderately long, rather strongly curved, simple, pulvilli about three-fourths length of claws. Genitalia: Basal clasp segment rather long, slender; terminal clasp segment moderately long, stout and curved; dorsal plate moderately long, broad, deeply and triangularly emarginate, interior margin with a distinct setose process at basal third; lobes triangular and sparsely setose; ventral plate long, broad, deeply and roundly emarginate, lobes slender and setose apicad; style long, moder- ately stout and broadly rounded apicad. Female.—Length, 1.5 millimeters. Antenne probably nearly as long as body, rather thickly long haired, mostly reddish brown, of 14 (?) segments, fifth with a stem nearly three-fourths length of cylindrical basal enlargement. The latter is about three times as long as its diameter. First segment of palpi short, irregular; second narrowly ovoid, with a length three times its diameter; third one-half longer than second; fourth a little longer than third. Mesonotum yellowish brown, sparsely long haired. Scu- tellum brownish yellow. Postscutellum and abdomen reddish brown, the latter sparsely haired. Halteres yellowish basad, fuscous apicad. Coxe yellowish brown. Legs mostly dark brown. Ovipositor short, terminal lobes sparsely setose, their length about three times their width. Ventral lobe much shorter and broadly rounded apicad. Type.—Cecid. a2882, New York State collection. Luzon, Laguna Province, Los Bafios, October 13, 1917, College of Agriculture accession No. 18165 (Uichanco). 294 The Philippine Journal of Science The specimens were reared from leaf galls on Paederia tomen- tosa Blume. Cecidomyia philippinensis sp. nov. Female.—Length, 2 millimeters. Antennz about one-half the length of body, rather sparsely haired, reddish brown, of 14 segments, fifth with a stem one-fourth the length of the cylin- drical basal enlargement, which latter is about three and one- half times as long as its diameter; a rather sparse whorl of stout sete basad and a somewhat thick band of slenderer setze subapicad; basal portion of terminal segment with a length about two and one-half times its diameter, the distal portion long and digitate. Each palpus consisting of one narrowly ovoid segment. Eyes holoptic. Mesonotum dull reddish brown, submedian lines yellowish, thickly haired. Scutellum pale yellowish. Postscutellum brownish yellow. Abdomen mostly dark red, thickly haired. Wings hyaline, slender, the length being nearly three times the width, subcosta united to costa by a rather distinct chitinized area and ending near basal third. Third vein nearly straight and joining margin well beyond apex of wing, fifth vein uniting with posterior margin at distal fourth, its branch near basal half. Halteres yellowish basad, reddish apicad. Coxe pale yellowish. Femora distad reddish brown, tibiz and tarsi mostly pale straw. Claws wanting. Ovipositor short, terminal lobes rather broadly ovoid and some- what thickly setose. Type.—Cecid. a2889, New York State collection. Luzon, Laguna Province, Mount Maquiling, September 3, 1917, College of Agriculture accession No. 18316 (Uichanco). The one female described above was accompanied by the following statement: “It is doubtful whether this is the cause of the numerous galls on the leaves of this plant (Spatholobus philippinensis Merr.), as this insect was bred out of a narrow thistle-shaped case, attached to the leaf and not from the galls.” Despite the above, there is a fair probability that this is the gall producer; and, as there is a definite food-plant record, I have described the species from a somewhat mutilated specimen. Since the characters are sufficiently marked, there should be comparatively little difficulty in establishing the identity of the insect later. THE PALMS OF THE PHILIPPINE ISLANDS By O. BEccaRi (Florence, Italy) THREE PLATES The species of palms at present known to be indigenous to the Philippines number one hundred twenty, not including a few semiwild or cultivated forms. Of that number but about a dozen are species of relatively wide geographic distribution, all the others being endemic forms. The nonendemic element is in part derived from species growing in swamps or at the mouths of rivers, along the coasts of neighboring countries, and are the following: Oncosperma filamentosa, Oncosperma horrida, Caryota mitis, Licuala spinosa, Nipa fruticans, Korthalsia laci- niosa, and three species of Calamus. The following are also nonendemic Philippine palms: Actinorhytis calapparia, Arenga saccharifera, Corypha elata, Metroxylon Rumphii, Caryota Rum- phiana, Livistona cochinchinensis, and L. rotundifolia, some of which may have been introduced by man or by other means. Heterospatha elata, which is common in the Philippines, is reported as growing also in the Moluccas, but I have not seen specimens of it collected in those islands. In fact the genus Heterospatha is represented in the Philippines by four species and may be considered as a Polynesian element in the Philip- pine flora. With this exception and that of Adonidia, discussed below, all the Philippine endemic species belong to genera more frequently found in the Malay Peninsula, the Malayan islands, and in Cochinchina; these genera are Areca, Pinanga, Arenga, Caryota, Orania, Ptychoraphis, Oncosperma, Livistona, Kor- thalsia, Zalacca, Plectocomia, Calamus, and Daemonorops. Only one genus appears to be exclusively Philippine, Adonidia, mentioned above, which belongs in a group of palms very rich in genera and species in the Papuan and Polynesian regions, but only scantily represented in southern Asia and in Malaya proper. e It is rather surprising that some common Malayan genera are either not at all or at most very poorly represented in the Philippines. Thus the genus Jguanura is entirely wanting. In Licuala, a gefius with numerous species in the primeval Malayan ; 295 296 The Philippine Journal of Science 1919 and Papuan forests as undergrowth palms, only the widespread and littoral L. spinosa has been found in the Philippines, where it occurs in the more southern islands of the Archipelago. Other southern Asiatic or Malayan genera of which represen- tatives are wanting in the Philippines are: Pholidocarpus, Nenga, Cyrtostachys, Bentinckia, Gigliolia, Zalaccella, Calo- spatha, Pigafetta, Ceratolobus, Plectocomiopsis, and Myrialepis. The genus Eugeissonia also, which is largely represented in Borneo, seems to be unrepresented in the Philippines. . Areca and Pinanga, among the Arecinaceae, and Calamus and Daemonorops, among the Lepidocaryeae, furnish two-thirds of the species of palms known as indigenous to the Philippines and form the great bulk of its palm flora. The genus Areca is particularly remarkable, as out of the total number of thirty-six species known to me at present no less than ten are very characteristic Philippine ones. In this number is also included Areca Catechu, of which we have good grounds for believing that it may have acquired its actual specific characters in some part of the Philippines. Even leaving out of consideration the fact that a variety of Areca Catechu (var. silvatica) has been found growing as a real forest plant in Palawan, there occur in the Philippines other fine species of Areca, closely related to A. Catechu, although cer- tainly specifically distinct from it. Of this number are A. macrocarpa, A. parens, A. Whitfordii, and perhaps A. Camari- nensis. Areca Catechu var. longicarpa, which may perhaps represent a distinct species, is also, apparently, a wild-growing plant in Polillo. The group to which all the above-mentioned species belong has no representative elsewhere; and if we consider A. Catechu as a palm introduced into the Philippines, where can we look for the allied species, from which we may suppose it to have been derived? As is more fully shown in the detailed treatment of the genus Areca, several species besides those mentioned above are indigenous to the Philippines. It is plainly evident, therefore, that the genus Areca has found in the Philippines more favorable evolutionary conditions for the plasmation of its endemic forms than elsewhere, and especially for the species related to Areca Catechu. In the genus Pinanga eighty species are now known to me (the descriptions of some of them not yet published), and 25 per cent of these are endemic Philippine species. However, this genus, unlike Areca, presents no conspicuous or anomalous Philippine representatives differentiating them much from the common Malayan types; but several are fine large palms deserv- XIV, 2 Beccari: The Palms of the Philippine Islands 297 ing the name of trees; such as Pinanga insignis, P. batana- nensis, P. sibuyanensis, and P. speciosa. Among the small forms P. maculata alone shows close affinities with the Malayan P. disticha. The genus Orania has also found in the Philippines very favorable conditions for evolution and is there represented by several forms, all of which, however, are apparently derived from two species belonging to two different and widely separated regions; namely, O. macrocladus, of the Malay Peninsula, and Orania moluccana and O. regalis, of the Moluccan and the Papuan regions. In Caryota forms are found that are but slight modifications of Malayan species, and the same may be said of the genus Arenga. The variety philippinensis of Caryota Rumphiana may, perhaps, be considered as an endemic species of the second degree, but its great affinity with the other forms of that poly- morphic palm, which is widely spread throughout the entire Malayan region, is plainly evident. Arenga saccharifera is, I think, to be considered as an in- troduced and naturalized plant. Oncosperma filamentosa, the elegant gregarious palm which so frequently gives so special a feature to the vegetation of the swampy estuaries in Borneo and Sumatra, is found in the Phil- ippines only in Palawan. Oncosperma horrida is another Ma- layan type from which the two endemic species O. platyphylla and O. gracilipes have apparently been derived. Livistona rotundifolia and L. cochinchinensis, two nonendemic ‘species, are widely distributed in the Philippines, and their presence in the Archipelago may be accounted for by the trans- port of their fruits from Indo-China, Celebes, and the Moluccas through the agency of birds or perhaps by that of flying foxes. Livistona rotundifolia, however, has given rise to some local forms in the Philippines. Livistona Merrillit and L. Robinson- iana are endemic. The Philippine Lepidocaryeae number fifty-five, of which thirty-six belong to Calamus, thirteen to Daemonorops, three to Korthalsia, one to Plectocomia, and one to Zalacca. Metroxylon Rumphii, which seldom produces fertile fruits, is probably an introduced plant in Mindanao and other islands in the southern part of the Archipelago. ‘ The Philippine species of Calamus, with only three exceptions, are endemic. The nonendemic species are C. symphysipus, a common plant in Celebes; and C. ornatus and C. Diepenhorstii, 298 The Philippine Journal of Science 1919 the last two rather common littoral Malayan species, although represented in the Philippines by distinct varieties. The thirteen species of Daemonorops are also, with the excep- tion of D. Margaritae, of southern China, all endemic, but several of them are closely related to species growing in neigh- boring countries. Thus D. Gaudichaudii, D. affinis, and D. gra- cilis are related to the common D. hystrix; D. ochrolepis approaches D. Rumphii, of the Moluccas; and D. virescens is allied to D. longipes, a common littoral Malayan species. Korthalsia laciniosa is also a common littoral plant in southern Asia, but K. scaphigeroides and K. Merrillii are endemic, al- though related to Malayan species. The only Zalacca known from the Philippines, Z. Clemensiana, belongs to the group which have the fruits covered with spicu- liferous scales, as Z. edulis and allied species. Plectocomia Merrillii is evidently a derivation from the Bor- nean P. Muellerii. On the whole we are led to conclude that the Philippine palms have in great measure originated in the Archipelago, but their phylogeny may be traced to species growing chiefly in Borneo, Celebes, the Moluccas, and Indo-China, excluding about a dozen species, which possess adaptations for easy dissemination, and for this reason are palms of rather wide geographic distribution. In addition a Polynesian element is represented by Adonidia and Heterospathe. The palm flora of the Philippines has, therefore, no individual- ity as to genera, the only genus peculiar to the Archipelago being the monotypic new genus Adonidia. In the Philippines a few large genera have varied so much as to give origin to numerous species, quite different in that respect from what has happened in Polynesia, where monotypic or oligotypic genera are numerous, and where no genus occurs that contains a great number of species. ENUMERATION OF THE PALMS INDIGENOUS TO, OR NATURALIZED IN, THE PHILIPPINE ISLANDS Areca Catechu Linn. (var. com- Areca parens Becc. munis). ‘= Areca Ipot Becc. Areca Catechu var. silvatica Bece. Areca Ipot var. polillensis Becc. Areca Catechu var. batanensis Becc. Areca camarinensis Becc. Areca Catechu var. longicarpa Becc. Areca Caliso Becc. Areca macrocarpa Becc. Areca costulata Becc. Areca Whitfordii Becc. Areca Vidaliana Becc. Areca Whitfordii var. luzonensis Areca Hutchinsoniana Becc. Becc. Pinanga maculata Porte. XIV, 3 Pinanga geonomaeformis Becc. Pinanga modesta Beve. Pinanga isabelensis Becc. Pinanga heterophylla Becc. Pinanga Barnesii Becc. Pinanga Copelandi Becc. Pinanga Curranii Becc. Pinanga philippinensis Becc. Pinanga Elmerii Bece. Pinanga urdanetana Bece. Pinanga samarana Becce. Pinanga urosperma Becc. Pinanga rigida Becc. Pinanga Woodiana Bece. Pinanga sclerophylla Becce. Pinanga negrosensis Bece. Pinanga insignis Bece. (forma ty- pica). Pinanga insignis var. gasterocarpa Bece. Pinanga insignis var. leptocarpa Becc. Pinanga Bece. Pinanga batanensis Becc. Pinanga basilanensis Becc. Pinanga speciosa Bece. Pinanga sibuyanensis Becc. Actinorhytis calapparia Wendl. Drude. Heterospathe elata Scheff. Heterospathe sibuyanensis Becc. Heterospathe philippinensis Becc. Heterospathe negrosensis Becc. Ptychoraphis microcarpa Bece. Ptychoraphis intermedia Becc. Ptychoraphis Elmerii Bece. Ptychoraphis cagayanensis Becc. Adonidia Merrillii Bece. Oncosperma filamentosa Blume. Oncosperma horrida Scheff. Oncosperma platyphylla Bece. Oncosperma gracilipes Becc. Orania philippinensis Scheff. Orania philippinensis var. sibuya- nensis Bece. Orania paraguaensis Bece. Orania decipiens Becc. Orania decipiens var. ensis Becc, Orania decipiens var. montana Becc. Orania rubiginosa Bece. insignis var. Loheriana mindanao- Beccari: The Palms of the Philippine Islands 299 Arenga saccharifera Labill. Arenga mindorensis Becc. Arenga tremula Blanco. Arenga Ambong Becc. Caryota Rumphiana var. philip- pinensis Becc, Caryota Rumphiana var. oxyodonta Becc. Caryota Cumingii Lodd. Caryota Merrillii Bece. Caryota majestica Linden. Caryota mitis Lour. Cocos nucifera Linn. Nipa fruticans Wurmb. Phoenia Hanceana Planch. var. phil- ippinensis Bece. Licuala spinosa Wurmb. Livistona cochinchinensis Mart. Livistona Merrillii Becc. Livistona rotundifolia Mart. Livistona rotundifolia var. luzonen- sis Becc. Livistona rotundifolia var. mindo- rensis Becce. Livistona Robinsoniana Becc. Corypha elata Roxb. Metroxylon Rumphii Mart. Zalacca Clemensiana Becc. Plectocomia Elmerii Bece. Korthalsia scaphigeroides Becc. Korthalsia Merrillii Becc. Korthalsia laciniosa Mart. Korthalsia squarrosa Becc. Calamus mollis Blanco. Calamus mollis var. major Becc. Calamus mollis var. palawanicus Bece. Calamus Meyerianus Schauer. Calamus Blancoi Kunth. Calamus melanorhynchus Becc. Calamus filispadix Becc. Calamus Diepenhorstii var. exulans Becc. Calamus ornatus var. philippinensis Becc. : Calamus discolor Mart. Calamus discolor var. negrosensis Bece. Calamus Cumingianus Becc. Calamus simphysipus Mart. Calamus Merrillii Bece. (forma typica). ee 300 Calamus Calamus Becc. Calamus Calamus Merrillii var. Nanga Bece. Merrillii var. Merrittianus Fouworthyi Becc. manillensis H. Wendl. Calamus trispermus Becc. Calamus Arugda Becc. Calamus vinosus Becc. Calamus Moseleyanus Becc. Calamus Mindorensis Becc. Calamus multinervis Becc. Calamus grandifolius Becc. Calamus Jenningsianus Bece. Calamus Samian Becc. Calamus megaphyllus Becce. Calamus Elmerianus Becc. Calamus mitis Beec. Calamus Reyesianus Becc. Calamus spinifolius Becc. Calamus viridissimus, Becc. Calamus microsphaerion Becc. Calamus microsphaerion var. spino- sior Becce. Calamus ramulosus Bece. Calamus Vidalianus Bece. Calamus siphonosphathus Mart. Calamus siphonosphathus var. sub- laevis Beecc. Calamus siphonosphathus var. oligo- lepis major Becce. The Philippine Journal of Science 1919 Calamus siphonosphathus var. oligo- lepis minor Bece. Calamus siphonosphathus var. poly- lepis Bece. Calamus siphonosphathus var. bata- nensis Becc. : Calamus dimorphacanthus Bece. Calamus dimorphacanthus var. mon- talbanicus Becc. Calamus dimorphacanthus var. zam- balensis Becc. Calamus microcarpus Becce. Calamus microcarpus var. diminutus Bece. Calamus haleonensis Bece. Calamus discolor Bece. Daemonorops Margaritae var. pala- wanicus Bece. Daemonorops virescens Bece. Daemonorops ochrolepis Bece. Daemonorops urdanetanus Bece. Daemonorops pedicellaris Bece. Daemonorops Loherianus Bece. Daemonorops pannosus Becce. Daemonorops oligolepis Becc. Daemonorops Clemensianus Becc. Daemonorops Gaudichaudti Becce. Daemonorops affinis Becc. Daemonorops Curranii Becce. Daemonorops gracilis Becc, NATURALIZED SPECIES Areca Catechu Linn. Actinorhytis Calapparia Wendl. & Drude. Arenga saccharifera Labill. Cocos nucifera Linn. Metroxylon Rumphii Mart. NONENDEMIC SPECIES Heterospathe elata Scheff. Oncosperma filamentosa Blume. Oncosperma horrida Scheff. Nipa fruticans Wurmb. Licuala spinosa Wurmb. Livistona cochinchinensis Mart. Livistona rotundifolia Mart. Korthalsia laciniosa Mart. Calamus Diepenhorstii var. exulans Bece. ; Calamus ornatus var. philippinensis Bece. ; Calamus simphysipus Mart. Daemonorops Margaritae var. pala- wanicus Becce. Caryota mitis Lour. Caryota rumphiana Mart. Corypha elata Roxb. - ARECA Linnaeus Ten species of Areca are at present known from the Philip- pines; all, with the exception of A. Catechu, are endemic. There are, moreover, several distinct varieties, for generally XIV, 3 Beccari: The Palms of the Philippine Islands 301 — speaking the same species is rarely represented by individuals offering an absolute uniformity of characters, when inhabit- ing different islands, even when these are not separated by a wide extent of sea. There are some large species specially worthy of remark, for they are so much like A. Catechu, as to render it uncertain whether they are to be regarded, at times, as simply varieties of that palm or as distinct species; they are A. macrocarpa, A. Whitfordti, and A. Catechu var. longicarpa. The first two are really forest plants, and apparently also the last is a forest plant. Areca Catechu var. silvatica possibly repre- sents the true wild plant of that widely cultivated species (of which the true place of origin has not been ascertained), for in the Philippines various forms of Areca occur so closely related to A. Catechu as to afford good ground for belief that it was in those Islands that A. Catechu finally assumed the specific char- acters it now exhibits. In support of this opinion I would ob- serve that in no other part of southern and eastern Asia or Malaya is any species of Areca to be found which in any way approaches A. Catechu in specific characters, whereas in the Philippines an entire group of species exists closely related to it. Areca parens is another fine species, endemic in the Phil- ippines, which, although endowed with characteristic and easily appreciable diagnostic characters, evidently belongs to the group of A. Catechu. The fruits of Areca Ipot and A. camarinensis in their prin- cipal characters and general appearance are also very much like those of A. Catechu, but these two species belong to the group that produces female flowers, and then fruits, densely grouped around the main axis of the spadix, and are therefore related to the species for which was established the subgenus Balano- carpus, which includes A. glandiformis, of the Moluccas; A. ma- crocalyx, of the Papuan Islands, and a few others. Areca costulata is no doubt related to A. (Mischophloeus) paniculata Scheff., a palm growing in northern Celebes and in the Moluccas; not only on account of its male flowers having the cyathiform calyx slightly 3-toothed and with a solid base, but also in its general habit. Areca Caliso is somewhat related to A. costulata, from which, -however, it differs by the calyx of the male flower being divided into three small distinct sepals; its nearest allied species is probably A. oxycarpa, of Celebes. Areca Vidaliana and A. Hutchinsoniana are of the type of the very widespread A. triandra, not only on account of the male flowers being inserted in pairs on unilateral notches of the flor- iferous branches, but also because the male flowers have six 1648914 802 The Philippine Journal of Science 1919 stamens, but A. triandra has 3-androus flowers; on that account A. Vidaliana and Hutchinsoniana agree with A. laosensis, of Indo-China, which also, although of the general habit of A. triandra, has male flowers with six stamens. Conspectus of the Philippine species. a’. Floriferous branches bearing only one or very few female flowers, sessile on their basal thickened part, and the geminate male flowers in distichous alternate indentures in their upper slender part. b’. Spadices spreadingly twice or thrice branched; palms with tall slender stems and relatively large fruits. (§ Euareca.) ce’. Male flowers narrowly lanceolate (unknown in A. Whitfordii) ; rudimentary ovary conspicuous, trifid, as long as, or longer than, the stamens; anthers acute or acuminate. Female flowers with broadly imbricate sepals about as broad and long as the petals. Fruit with the pericarp finally entirely disintegrating into very fine and soft fibers. d’. Fruit having the mesocarp considerably thicker at both ends than at the sides, and the seed inserted considerably above the base. e’. Seed having the vascular bundles.of the integument arching on the sides, and strongly anastomosing immediately from its base, even on the raphal side 1. A. Catechu. f'. Fruit orange-red; globose-ovoid or ovoid-ellipsoid, not more than one-third or one-fourth longer than broad (4 to 5 em long, 3 to 4 cm broad). Seed subglobose with a more or less flattish base................ A. Catechu (forma communis.) f’. Fruit ovoid-ellipsoid, rather ventricose, smaller than usual (4 cm long and 8 em or a little less broad); seed globose- depressed or broader than high, the base flat. A. Catechu var. silvatica. f*. Stems thicker and shorter than in forma communis; spadix denser and with shorter floriferous branches. A. Catechu var. batanensis. f*. Fruit narrowly ellipsoid; twice, and even more, as long as broad (5.5 to 7 cm long, 2.5 cm broad). Seed ovoid-conical with a blunt apex and flat base, slightly longer than broad. A. Catechu var. longicarpa. e’. Seed having the vascular bundles of the integument arising straight, almost erect, from the raphal side and slightly branching. Fruit elongate-ellipsoid, twice as long as broad (7 cm long, 3.2 to 3.5 em broad). Seed ovoid-conical with a blunt apex 2. A. macrocarpa. @. Fruit ellipsoid-fusiform, twice and more as long as broad, having the mesocarp not much thicker at the ends than at the sides. Seed placed nearly in the middle of the pericarp and equally narrowing to both ends. e. Fruit 4 to 4.5 cm long, 1 to 2 em broad............ 3. A. Whitfordii. e’. Fruit larger, 5.5 em long, 2 em broad. : A. Whitfordii var. luzonensis. ce. Male flowers relatively large, ovoid-subtrigonous or trapezoidal; rudimentary ovary small, slender, entire, subulate, shorter than XIV, 8 Beccari: The Palms of the Philippine Islands 303 the stamens; anthers very obtuse. Sepals of the female flower considerably smaller than the petals. Fruit large, ovoid, about 6 cm long, 3.5 cm broad, the pericarp fibrous in its outer half, woody in the inner half. Seed broadly ovoid, its vascular bundles very numerous, parallel and almost undivided, ascending from the raphal side 4. A. parens. b’. Spadices simply branched. Female flowers clustered around the main axis, solitary, sessile or nearly so, at the base of the branchlets; the latter slender and bearing alternately distichous male flowers. (§ Balanocarpus.) (Of A. camarinensis the detached fruits only are known, and it is doubtfully placed here.) c’. A palm with the habit of Areca Catechu but smaller (stem about 4 m high, 7 to 12 em in diameter); fruiting spadix dense, cylindraceous-oblong, about 14 cm long, 6 to 7 em thick. Fruit ovoid, very similar to that of A. Catechu, 5 cm long, 3 cm broad. Seed globose with rounded (not flat) base and with the vascular bundles of the integument very close together, much anastomosing and forming very narrow loopholes all around the seed. 5. A. Ipot. d'. A smaller plant, the stem 5 cm in diameter, the spadix smaller, with fewer female flowers, and forming a shorter mass. A. Ipot var. polillensis. ec. Fruit ovoid, 4 to 5 em long, 3 cm broad. Seed conical-ovoid; the vascular bundles of the integument forming a uniform network all around the seed with lozenge-shaped loopholes. Otherwise the fruit is similar to that of A. Catechu........ 6. A. camarinensis. a’. Floriferous branches bearing several female flowers on their basal parts, gradually narrowing above and bearing male flowers only in pairs on alternating notches. Low palms with relatively small or medium- sized fruits. (§ Arecella.) b’. Spadix simply branched, with thickish floriferous branches appressed to the main axis, and bearing in their basal part numerous, approximate, alternate, female flowers. Male flowers hexandrous, the calyx with three small, distinct sepals; anthers acute. Fruiting perianth cupular, truncate, the petals exactly equaling the sepals. Fruit ellipsoid, 3 to 3.5 em long, 20 to 22 mm thick, the pericarp entirely dissolving into very fine, soft fibers................ 7. A. Caliso. b*. Spadix twice loosely branched; floriferous branches slender, bearing in their basal part three or four alternate, rather distant, female flowers, and in the upper and slenderer part alternate male flowers. Calyx of the male flowers subpedicelliform, shortly 3-dentate with a solid base; anthers bifid at the apex. Fruit small, pluricostulate, ellipsoid, the pericarp formed by only two layers of rigid com- planate fibers 8. A. costulata. a’. Spadix diffusely, two or three times branched, the floriferous branches bearing one or more female flowers in their lower part, and above male flowers in pairs in unilateral notches. Low slender palms having very small male flowers, with the calyx completely divided into three sepals. Fruit small. (§ Arecopsis.) b*. Male flowers 2.5 to 3 mm long; stamens 6; rudimentary ovary as long as the stamens, divided into three points. Female flowers ovate, obtuse, 8 mm long, 4.5 to 5 mm broad. Fruit small, oblong- 804 The Philippine Journal of Science 1919 ellipsoid or subfusiform, 17 to 19 mm long and broad; pericarp rather thin, the mesocarp formed by only two layers of slender, but rigid, flattened, parallel fibers....! 9. A. Vidaliana. b*. Male flowers smaller than in A. Vidaliana (2 mm or a little longer) ; stamens 6; rudimentary ovary as long as the stamens, trifid. Fe- male flowers ovoid, conical, narrowing to an acute point, larger than in A. Vidaliana, 13 to 14 mm long, 7 mm broad at the base. Fruit also larger, ellipsoid-elongate or subfusiform, about 3 cm long, ine Ne ee kk Si i, 10. A. Hutchinsoniana. 1. ARECA CATECHU Linn. (forma COMMUNIS). Commonly cultivated throughout the Archipelago and occa- sionally spontaneous. The specimens collected in Mindanao on the edge of the forest at Katajan, Mount Malindang, Misamis Province, For. Bur. 4717 Mearns & Hutchinson were probably from a semiwild plant. ARECA CATECHU Linn. var. BATANENSIS Becc. Areca Catechu (non Linn.) Becc. in Philip. Journ. Sci. 3 (1908) Bot. 340. BATAN ISLANDS, Batan, Santo Domingo de Basco, Bur. Sci. 3834 Fénizx, along a mountain stream. According to the collector this is 7 to 8 m high with a trunk 25 cm in diameter. It is therefore a plant with a relatively shorter but thicker trunk than the common Areca Catechu. The spadix is also denser than in that palm and with shorter flori- ferous branchlets, the latter only 10 to 12 cm in length. I have not seen the fruits of this apparently very local variety. ARECA CATECHU Linn. var. SILVATICA Becc. Trunk 10 to 12 m high, 10 to 15 cm in diameter. Leaves 2 to 2.5 m long with about 30 pairs of leaflets. Spadix diffuse, with very elongate and slender floriferous branchlets. Male flowers 4 mm long, in shape exactly as in the type, with 6 stamens and the rudimentary ovary divided into 3 acuminate points. Female flowers relatively narrow, tapering above to a conical and rather acute point, 12 to 14 mm long, 7 mm thick. Fruit reddish-yellow, ovoid-ellipsoid, broadly conical in the upper part, 4 cm long (including the perianth), 28 mm thick. Seed (placed almost in the center of the fruit) globular-depressed or slightly broader than high (16 mm high, 22 mm broad), reg- ularly rounded above and flattish in its basal part; embryo conical; rumination and network of the surface of the integu- ment exactly as in the type; walls of the pericarp 3 mm thick at the sides (in the dry fruit). Fruiting perianth 2 cm broad at the mouth. XIV, 3 Beccari: The Palms of the Philippine Islands 805 PALAWAN, near Lake Manguao, in a narrow valley along a small stream in the primeval forest, Merrill 9447, April, 1913. This differs from the common form in having smaller fruits with the seed globular-depressed and in its more acuminate fe- male flowers. As to the possibility of this being a really wild form of Areca Catechu, Mr. Merrill writes to me from Manila under date of October 2, 1914, as follows: As to Areca Catechu the specimen that I collected in Palawan is the only one that I have seen growing in primeval forests in all my travels in the Philippines. At the place where found the plants, few in number, were growing in a forested ravine along a small stream at a place where an old and apparently much traveled native trail crossed the stream. I strongly suspect that the trees that I found in this place originated from seeds accidentally left there by natives. The species is of course very common and very widely distributed in the Philippines, but normally is found only in the settled areas and in cultivation. It is, however, dis- tinctly self seeded and one frequently finds young plants in thickets, fence rows, etc., that have not been purposely planted, but which have originated from seeds fallen from neighboring trees. ARECA CATECHU Linn. var. LONGICARPA Bece. in Philip. Journ. Sci. 6 (1911) Bot. 229. | POLILLO, Bur. Sci. 10470 McGregor, October-November, 1909. It is not known whether or not the specimens were collected from cultivated or wild plants. This is a very remarkable variety on account of its elongate- ellipsoid fruit, almost equally tapering at both ends, and with a conical apex, varying from 5.5 to 7 cm in length (including the perianth) and from 2.5 to 3 cm in diameter, or more than twice as long as broad. The pericarp is 3 to 4 mm thick at the sides, and is solid in most of its lower half, the seed being inserted very high, or only a little less than half the entire length of the pericarp. The mesocarp is not so filled with fibers as in the other forms of Areca Catechu, and in its central and basal part the fibers extend through a very soft and spongy tissue. The seed is 24 mm long (in one specimen), has a nearly or quite flat (22 mm broad) base, and suddenly assumes a conical shape, yet it is blunt at the apex; the vascular branches of the integ- ument start from the raphe at an angle of about 45° and form a close network as in the typical form. The male flowers are 6 mm long and 1.5 mm broad, or somewhat longer, narrower, and more acuminate than in forma communis. The fruits are said to be yellow. It seems to represent rather a subspecies of Areca Catechu than a simple variety of that plant. 306 The Philippine Journal of Science 1919 2. ARECA MACROCARPA Bece. in Philip. Journ. Sci. 4 (1909) Bot. 601. MINDANAO, Zamboanga District, Port Banga, For. Bur. 9108 Whitford & Hutchinson, January, 1908. _ This seems to be a close ally of Areca Catechu, but unfortu- nately its male flowers are not known for a precise comparison with the latter. Its fruit, however, is considerably larger than that of A. Catechu, from which it is further distinguishable by its being equally tapering at both ends, twice as long as broad; by the seed inserted very high, or somewhat above the lower third of the entire fruit; by the ovoid-conical seed with flat base; and by the vascular branches of the integument ascending in great numbers from the raphe and arching on the sides at an angle of 45°, forming there a rather close network. In the only seed I dissected, the embryo was globular-depressed, where- as it is conical in A. Catechu and all its varieties. The fruiting perianth is distinctly campanulate. 8. ARECA WHITFORDI! Becc. in Philip. Journ. Sci. 2 (1907) Bot. 219. MINDORO, in the semiswampy forests called “guipa” by the natives, near sea level, Bongabon River, Whitford 1372, Tagalog, bungan gubat. I formerly supposed that Areca Whitfordii was related to the species having male flowers with three stamens; but now that I am in possession of additional material, although even now incomplete, it appears to me that its real affinities are with A. Catechu and especially with the variety longicarpa, because the floriferous branchlets of A. Whitfordii bear, as does A. Catechu, alternate and not unilaterally geminate male flowers as is the case in A, triandra and allied species; although the male flowers of typical A. Whitfordii are as yet unknown. It differs, how- ever, from A, Catechu var. longicarpa especially in its ovoid- ellipsoid seed, narrowed at both ends, symmetrically placed in the central part of the fruit, of which it occupies almost the entire length, the pericarp extending below and above the seed only 7 mm. ARECA WHITFORDI! Becc. var. LUZONENSIS Becc. var. nov. This differs from the type only in its larger fruits. The fruits seen by me, not thoroughly mature but apparently having ettained their definitive dimensions, are ellipsoid-fusiform, 5.5 cm long, equally narrowed at both ends, and 2 cm in diameter in their central part. The seed, from the size and the shape of the endocarpal cavity, is apparently equally narrowed at both ends, 28 mm long and 16 to 17 mm through at its middle; the portion of solid pericarp remaining above and below the seeds is 1 cm in length. The leaves have 20 to 25 pairs of leaflets. XIV, 8 Beccari: The Palms of the Philippine Islands — 307 Luzon, Nueva Vizcaya Province, vicinity of Dupax, Bur. Sci. 11884 McGregor. The collector notes that it is a really wild, and not a cultivated, plant and that its fruits are used as a substitute for those of A. Catechu. Not differing from the above is For. Bur. 17770 Curran from Cagayan Province, Luzon, where it is common in overflowed river swamps. 4. ARECA PARENS Becce. sp. nov. Major, A. Catechu similis. Frondium segmenta majuscula conspicue bicostulata, falcato-acuminata. Spadix duplicato- ramosus, ramis floriferis in parte basilari crassiusculis, superne zigzag sinuosis. Flores ¢ pro rata majusculi, ovato-subtrigoni vel trapezoidei, 7 mm longi, 4 mm lati; calyce*minuto, sepalis triangularibus, acuminatis ; staminum filamentis brevibus; anthe- ris linearibus, obtusis; ovarii rudimento parvo subulato. Flores 2 ovati, 12 ad 15 mm longi, sepalis quam petala conspicue bre- vioribus. Fructus ovato-ellipticus, inter majores, 6 et ultra cm longus, 3.5 cm crassus, utrinque aequaliter attenuatus; pericarpio ad latera 5 ad 6 mm spisso, in dimidia inferiori parte lignoso, caetero fibris tenuibus percurso. Semen prope tertiam inferio- rem partem insertum, ovoideo-conicum, 28 mm longum, 22 mm crassum, integmenti ramificationibus vascularibus numerosis, pa- rallelis, indivisis, a raphe adscendentibus, superne et in facie antirapheali tantum anastomosantibus. Albumen lamellis rectis numerosis descentibus ruminatum, Apparently of the general habit and dimensions of Areca Ca- techu and with the same kind of leaves. One of the leaflets is about 1 m long and 8 cm broad, very strongly bicostulate and with an acuminate falcate point. The spadix has the same kind of branching as that of Areca Catechu; its ultimate divisions or floriferous branches are 15 to 20 cm long, 3 mm thick at their bases, angular, strongly zigzag-sinuous with only one female flower in their basal part but covered elsewhere with distinctly distichously alternate male flowers; the latter are in pairs at every flexure and furnished with two very small triangular brac- teoles; they are asymmetrical, irregularly ovoid-subtrigonous or trapezoidal by mutual pressure, and relatively broad (7 mm long, 4 mm broad) ; the calyx has three, very small, triangular- acuminate, dorsally keeled sepals; petals pergamentaceous, rather rigid, ovate, bluntish; stamens 6, the filaments terete, short; anthers linear, more or less sinuous, bluntish, with parallel cells, and dark connective; rudimentary ovary very small, subu- late, scarcely as long as the filaments. Female flowers sessile on a horizontal cushion, the latter furnished with a small rim representing the floral bract; they are ovate, obtuse, 12 to 15 mm long, 10 mm broad; sepals broadly triangular, acute or 808 The Philippine Journal of Science 1919 acuminate, considerably smaller than the petals; the latter, one- third or more longer than the sepals, very broadly obtuse or else cucullate at the apex. Fruit regularly ovoid-ellipsoid, equally narrowed at both ends, 63 mm long and 35 mm through (in two specimens) or a little less than twice as long as broad. The pericarp is 5 to 6 mm thick at the sides, and only in its outermost half at maturity does it disintegrate into capillary fibers, its inner half remaining woody; the apex in its internal part is also woody, but the base is entirely fibrous. The seed is inserted a little below the lower third of the entire fruit, ovoid with a flattish base and a very obtuse apex, 28 mm long, 22 mm broad; the vascular branches of the integument are very numerous and arise parallel and undivided at an acute angle from the base, anastomosing only above and on the ventral or antiraphal side; the rumination of the albumen is produced by numerous straight and descending lamella. Embryo broadly ovoid. LuzON, Camarines Province, For. Bur. 21712 Pefas, Soriano, & Abella- nosa, April 23, 1914, with the Bicol name tacohtoh. This is a wild-growing species approaching Areca Catechu, although doubtless specifically distinct from it. The spadix, in flower, has the appearance of that of A. Catechu, but its flori- feroys branches are more robust and strongly zigzag-sinuous in the male flower-bearing portion. The male flowers are larger and broader than those of Areca Catechu and differ also in having very blunt anthers and a short rudimentary ovary. Fur- thermore, the female flowers of A. parens differ also from those of A. Catechu especially in the sepals being considerably shorter than the petals, the latter being usually blunt and more or less cucullate at apex. The fruit of Areca parens resembles that of A. macrocarpa, but differs from the fruit of all the species ap- proaching A. Catechw in having the outer half of the pericarp disintegrable into soft fibers only, the inner half being woody. The seed is also characterized by the very numerous vascular branches of the integument lying close together, parallel and undivided laterally and on the raphal side, and by the lamellz which cause the rumination of the albumen being straight and inclined toward the base of the seed. 5. ARECA IPOT Becc. in Leafl. Philip. Bot. 2 (1909) 639. Areca Catechu Linn. var. humilis Blanco, Fl. Filip. (1837) 716, ed. 2 (1845) 495, ed. 3, 3 (1879) 120. Luzon, Tayabas Province, Sampaloc, For. Bur. 10178 Curran (cultivated specimen); Lucban, Elmer 9292, local name bunga ipot. XIV, 3 Beccari: The Palms of the Philippine Islands 809 This is chiefly distinguishable by its simply branched spadix having all the female flowers crowded around the main axis and by the male flowers being geminate or solitary in alternate notches of the slender branchlets, each of which starts from the base of a female flower. The ovoid fruit closely resembles that of Areca Catechu and the seed also is similar, but in A. Ipot the vascular ramifications of the integument of the seed are more numerous, more anastomosing, and form a closer network than in A. Catechu, and the rumination of the albumen is formed by more numerous and straighter lamellae, all converg- ing toward the center of the seed. The pericarp entirely dis- integrates into soft capillary fibers, except in the innermost part for the extent of about a millimeter. ARECA IPOT Becc. var. POLILLENSIS Bece. in Philip. Journ. Sci. 6 (1911) Bot. 229. PoLILLo, Bur. Sci. 10463, 10469 McGregor, Bur. Sci. 6988 Robinson, along streams at about 25 meters’ elevation (apparently a wild plant). This differs from the type in being a smaller plant, with a stem about 4 m high and 4 to 5 cm in diameter, and in having a shorter spadix. 6. ARECA CAMARINENSIS Becc. sp. nov. — Gracilis, caudice circiter 8 cm diametro, internodis longius- culis (7 cm longis in specimene suppetenti). Frondium (circiter 2m longum) segmenta numerosa, utrinque 8 ad 10, aequidis- tantia, alterna vel subopposita, 4 ad 6 cm inter se dissita, angusta, elongata, falcato-sigmoidea, 2- vel 3-costulata, costulis conspicuis, complanatis, 2 mm elevatis; segmentis inferioribus et intermediis acuminatissimis, majoribus 60 ad 75 em longis, 2.5 ad 5 cm latis, nonnullis superioribus apice attenuatis et falcatis, attamen inciso-dentatis, duobus terminalibus 6- vel 7-costulatis, brevioribus, circiter 30 cm longis, 4 ad 5 em latis, apice truncatis et duplicato-inciso-dentatis, dentibus brevibus. Petiolus longiusculus, subteres. Rachis compressa, in parte basilari subtus rotundata, apicem versus compressa et subtus plana, superne angulo elevato acuto percursa, marginibus acutis. Spadix * * *, Fructus rubri, ovati, 4 ad 5 cm longi, 3 cm crassi, utrinque rotundati, areola stigmatiferia plana, minime elevata, coronati; mesocarpio crasso, fibris tenuibus numero- sissimis percurso; endocarpio tenuiter lignoso, intus albescenti et laevi; semine conoideo, 2 cm longo, 15 mm diametro, apice obtuso, basi truncato, distincte reticulato, lamellis approximatis rectis, descendentibus ruminato. ae : 310 The Philippine Journal of Science 1919 LuzoN, Camarines Province, Goa, on the lower slopes of Mount Isarog, For. Bur. 14268 Aguilar, April, 1909, Bicol mono. I have seen one leaf and one fruit only. It seems, however, to be a very peculiar and interesting palm, apparently belonging to the section Balanocarpus, although its true affinities remain somewhat uncertain; neither the entire spadix nor the male flowers are known. The fruit somewhat resembles that of Areca Catechu and of A. Ipot, but it differs from both by the vascular bundles of the integument which form a distinct and rather regular network with 1.25 mm wide lozenge-shaped loopholes. From A. Catechu the fruit also differs in the numer- ous and approximate lamellae of the rumination which con- verge toward the center in a straight and descending course. 7, ARECA CALISO Becc. in Leafl. Philip. Bot. MS. MINDANAO, Davao District, Mount Apo, Todaya, at about 1,000 meters’ elevation, Elmer 11898, September, 1909, locally known as caliso: Agusan Subprovince, Mount Urdaneta, along the Dalahcan River, Elmer 13814. BILIRAN, Bur. Sci. 18875 McGregor. LEYTE, Dagami, Bur. Sci. 15388 Ramos, August, 1912; in this specimen the fruit is a little larger than in the Mindanao plant. 8. ARECA COSTULATA Bece. sp. nov. Mediocris, caudice circiter 5 m longo. Folia inaequaliter pin- natisecta, segmentis rigide papyraceis, 2-3-costulatis, falcatis, vel in apice duplicato-bifidis, majoribus 60 cm longis, 7 cm latis. Spadix patule laxeque ramosus; ramulis floriferis in parte basi- lari, flores 2 ad 4 remote alterne ferentibus, superne zigzag- sinuosis et floribus masculis geminis alterne onustis. Flores masculi majusculi, 6 ad 8 mm longi; calyce campanulato, basi plaeno, breviter 3-dentato; staminibus 6; antheris apice pro- funde bipartitis; ovarii rudimento inconspicuo papillaeformi. Fructus ovoideo-elliptici, 3 cm longi, 13 mm crassi, conspicue longitudinaliter pluricostulati; pericarpio coriaceo, 2 mm spisso, fibris rigidis biseriatis percurso. Perianthium fructiferum campanulatum, non marcescens. Apparently of medium or rather small size. Stem (slender?) 5 m high. The leaves have very unequal and inequidistant, firmly papyraceous leaflets, the latter with very robust main nerves (costae) and sharp secondary nerves, which render both surfaces distinctly striate; all nerves are sprinkled, espe- cially on the lower surface, with very minute glanduliform papillae, The lower leaflets in the specimen before me are 3-costulate, 60 cm long, 7 em broad, base broad, apex falcate; upper leaflets smaller, 2- to 3-costulate, double-toothed at the apex, the primary teeth 4 to 6 cm long, the secondary teeth XIV, 3 Beccari: The Palms of the Philippine Islands sll short, acute; uppermost leaflets narrow, straight, their second- ary teeth short, acute. Spadix loosely twice branched; primary branches 20 to 30 cm long, bearing few, gradually diminishing, floriferous branchlets; the latter have a naked basal part 3 to 4 cm in length, then bear three or four alternate female flowers and above them the male flowers, which are alternately inserted in pairs at every indenture on the rather elongate, filiform, zigzag-sinuous upper part of the branchlets. Female flowers ovoid, coriaceous, 8 to 10 mm long; corolla very slightly longer than the calyx, the petals surpassing the sepals solely by their small,.apical, valvate, 1.5 mm. long, triangular, bluntish apices; floral bracts small. Male flowers lanceolate, acute or acuminate, 6 to 8 mm long; calyx campanulate, shortly and broadly 3-toothed, solid in its lower part and flat at the base; corolla about four times as long as the calyx; petals per- gamentaceous, lanceolate-acuminate; stamens 6, very slightly shorter than the petals; filaments terete, unequal, some of them at times as long as their respective anthers, the latter linear, slightly sinuous, deeply bipartite at apex and with the cells deeply separated at the base; rudimentary ovary inconspicuous, papillaeform. Fruits ovoid-ellipsoid, slightly ventricose in the middle, equally tapering to both ends, 3 cm long, 18 mm broad, boldly longitudinally striate or pluricostulate (when dry), ter- minated by a circular areola 3 mm in diameter bearing the remains of the stigmas. Pericarp coriaceous, not quite 2 mm thick, formed by two layers of relatively very robust and hard fibers (the cause of the ribs on the dry fruit). Seeds ovoid- oblong, rounded above, deeply ruminated. Fruiting perianth campanulate, not accrescent, the apices of the sepals and petals unchanged (not marcescent). LeyTE, Dagami, Bur. Sci. 15286 Ramos, August, 1912, in forests. So far as I can judge from the very fragmentary specimens at my disposal, this seems to be a plant of modest dimensions. It is especially characterized by its relatively firm leaflets, which are very minutely papillose-punctulate on the nerves, especially on the lower surface; by its diffusely branched spadix; by the relatively large male flowers, having a subpedicelliform, cam- panulate, shortly 3-toothed calyx, solid in its lower part; by the stamens with relatively long filaments and anthers, which are linear and deeply incised at their apices; by the inconspicuous rudimentary ovary; by the floriferous branches bearing few distant flowers in their basal part and alternate pairs of male flowers in their zigzag-sinuous upper parts; by the campanulate 312 The Philippine Journal of Science 1919 fruiting perianth unchanged; and by the small, ellipsoid, plu- ricostulate fruits, having in the pericarp a double series only of very rigid robust fibers. 9. ARECA VIDALIANA Bece. in Philip. Journ. Sci. 2 (1907) Bot. 222. Areca mammillata Becc. op. cit. 220. PALAWAN, Vidal 3955 in Herb. Bece. Areca mammillata Becc. was also collected in Palawan, For. Bur. 3816 Curran, 1906, in swampy places along the Sariban River; Taytay, Merrill, Phil. Pl. 1464, May, 1913. I previously considered the plant collected by Curran in Pala- wan and named by me Areca mammillata to be a species distinct from A. Vidaliana, because the specimens at that time at my disposal presented some remarkable differences in the leaves, the segments being 3- or 4-costulate in A. Vidaliana and 1-cos- tulate in A. mammillata; but after the inspection of more com- plete material I have found that in A. Vidaliana the character derived from the number of main costae in the leaflets is a very variable one, and that the flowers and fruit offer no differences in the two plants. Areca Vidaliana is very similar in habit to A. triandra, but its male flowers are always 6-androus, and not 8-androus as is the case in the latter. 10. ARECA HUTCHINSONIANA Becc. nom. nov. Areca mammillata Becc. var. mindanaoensis Becc. in Philip. Journ. Sci. 4 (1909) Bot. 602. A slender palm. Stem about 3 cm in diameter. Leaves about 1 m in length. Leaflets numerous, subequidistant, 4 to 6 cm apart on each side of the rachis, usually bicostulate, but at times 1- to 4-costulate, falcate-sigmoid, slightly attenuate at the base, and gradually narrowed upward to a long, linear, caudate tip, when unicostate, 2 em, when bicostulate, 4 to 4.5 cm broad; the 3- or 4-costulate leaflets are, of course, propor- tionally larger; the two of the apex are united by their bases, truncate and almost praemorse, and with as many incisions as there are costae; the resulting primary teeth are shallowly and obtusely 2-toothed. Spadix broadly panicled, twice branched, its pedicellar part very short and flattened, 10 to 15 mm long; the main branches, after the fall of the male flowers, are spread- ing and devaricate; the lowest branches are the largest, 15 to 20 cm long or more, and are unilaterally divided into several small secondary or floriferous branchlets, each of which carries in its basal part one or two sessile female flowers, and is very slenderly filiform in its upper part; this part is unilaterally and very closely notched, and carries two male flowers at each notch. The male flowers are very small, 2 mm or slightly longer, lan- ceolate, acute or bluntish; the calyx is very small, trigonous, xIV,8 Beccari: The Palms of the Philippine Islands 313 and almost entirely divided into three acuminate sepals; the corolla is about four times as long as the calyx; stamens 6, slightly shorter than the petals; anthers linear-lanceolate, acute- apiculate; rudimentary ovary represented by three filiform bodies as long as the stamens. Female flowers ovoid, obsoletely angular, rounded or flattish at the base, gradually attenuate above, acute, 13 to 14 mm long, 7 mm broad at the base; sepals broadly ovate, minutely ciliate, obsoletely keeled and slightly thickened below the bluntish or only slightly acute apices; petals imbricate, broadened at both sides in their basal part, and sud- denly contracted into a thick, triangular, acute, valvate apex, the latter, 3 to 4 mm long, protruding entirely above the calyx; ovary oblong, attenuate below; stigmas 3-gonous, spreading. Fruit very narrowly ellipsoid-oblong or subfusiform, diminish- ing considerably toward both ends, about 3 cm long, 9 mm in diameter. Pericarp thin, formed by only 2 or 3 layers of slen- der, parallel, complanate fibers, which render the surface of the dry fruit very finely striate longitudinally. Seed 13 mm long, 6.5 mm thick. Fruiting perianth cyathiform, 12 mm high. MINDANAO, Zamboanga District, Port Banga, in forests at about 20 meters above sea level, For. Bur. 9141 Whitford & Hutchinson (specimen with fruits). BASILAN, near the southwest end of Mindanao, Mount Balanting, at about 100 meters above sea level, For. Bur. 4019 Hutchinson, February, 1906, a palm 3 meters high, flowers white, fruit white; very common, local name pisa. Apparently also referable to A. Hutchinsoniana is Vidal 4065 (in Herb, Beccari), from Siassi, one of the smaller islands in the Sulu Archipelago, southwest of Basilan. The description above, as regards the fruit, is derived from Whitford & Hutchinson 9141, otherwise from Hutchinson 4019. It is closely related to Areca Vidaliana, from which it differs in its smaller male and larger female flowers and, especially, in its larger fruits. PINANGA Blume Conspectus of the species. a‘. Very slender plants having simple fiabellate leaves or only one or two segments on each side of the rachis. b'. Leaves mottled (at least those of young plants), deeply bilobed, other- wise entire 1. P. maculata. b?. Leaves with a terminal, deeply bilobed flabellum and 1 or 2 acinaciform segments on each side of the rachis. Spadix simple. Fruit dis- tichous. 2. P. geonomaeformis. a*. Leaves having few (4 to 9), very unequal, inequidistant segments on each side of the rachis. Spadix branched. Small plants with slender stems 1 to 3 cm in diameter. b*. Fruits biseriate. 814 The Philippine Journal of Science 1919 c’. Leaves short with very few (4 or 5), unequal, sigmoid-acinaciform, 3- to 7-costulate segments on each side of the rachis, ashy- puberulous underneath, at least in newly expanded leaves. Spadix with 8 or 4 branches only. Fruits exactly distichous, narrowly ovoid, with a conical point, 15 to 16 by 6 to 7 mm. Seed caudic- ulate at the base; vascular branches of the integument rather numerous, nearly simple. 3. P. modesta. ec’. Leaves having on each side of the rachis 8 or 9, narrow, 2- to 4-costulate, distant, ensiform-subfalcate, very acuminate segments, which aré glabrous underneath. Spadix with 5 or 6 branches. Fruits exactly distichous, narrow, conical-subfusiform, broadest below their middle, 16 to 18 by 6 mm; seed elongate, conical in its upper part; vascular branches of the integument very few (3 or 4), almost simple 4. P. isabelensis. b?. Fruits 3-seriate. Leaves having on each side of the rachis 7 or 8 very inequidistant, unequal, 1- to 4-costulate, narrowly falcate- sigmoid, long-acuminate segments, which are glabrous underneath. Spadix with few (5) triquetrous branches. Fruit small, 11 to 12 by 5 to 5.5 mm, narrowly ovoid-ellipsoid, acute, equally narrowed to both ends. Seed ovoid, with a conical point; vascular branches of the integument 8 or 9, slightly anastomosing.. 5. P. heterophylla. a*, Leaves with more numerous segments. : b*. Segments ashy-puberulous underneath; plants of moderate size. ec’. Fruits 3-seriate. Segments relatively not very numerous, very unequal, inequidistant, 3- to 7-costulate, deeply incised into 3 to 7 bifid laciniae, the latter falcate, acuminate. Spadix with few (5 or 6) branches. Fruits relatively large, ovoid-ellipsoid, 2.5 to 3.3 cm long. Fruiting perianth low, cupular, not contracted at the mouth. Stem 3 to 5 em in diameter.................... 6. P. Barnesii. ce. Fruits 2-seriate. d@’. Segments rather numerous, very unequal, 1-pluricostulate, sig- moidal; when with more than one midcosta then deeply cleft at the apex into falcate-acuminate points. Spadix with the upper branches spirally scattered. Fruit obovoid, 18 to 20 by 13 mm. Seed spherical, with a horizontal embryo fovea. Fruiting perianth very shallowly cupular or almost explanate. 7. P. Copelandii. @. Segments numerous, 2- or 3-costulate, almost straight, not deeply incised at the apex, the divisions 2-toothed, the teeth acute. Fruit ovoid-ellipsoid, obtuse, small, 15 by 9 mm. Seed with a very oblique embryo fovea. Fruiting perianth shallowly cup- ular, with nearly vertical walls, not contracted at the mouth. 8. P. Curranii. b*. Segments glabrous underneath. c’. Spadix with rather few branches. @. Fruits distichous. Small or medium-sized plants. e*, Stem 2 to 5 cm in diameter. Leaf-sheaths densely covered with a grayish tomentum. Segments rather numerous, slightly sigmoid, acuminate, usually 3-costulate and relatively broad, or else narrow and 1-costulate. Spadix with few, scattered XIV, 8 o- 4 Beccari: The Palms of the Philippine Islands 815 or subdistichous branches. Fruit narrowly ovoid-ellipsoid, acuminate, 11 to14by5to8mm. Fruiting perianth cupular, contracted at the mouth 9. P. philippinensis. . Very similar to the preceding. Stem 2 to 4 m high, 2 to 5 cm in diameter. Leaf-sheaths covered with appressed rusty scales (not tomentose). Segments rather numerous and sub- equidistant; usually 1-costulate, slightly falecate acuminate or nearly straight. Spadix with a few scattered spreading branches. Fruit ovoid-ellipsoid, 12 to 14 by 7 to 8 mm. Fruiting perianth cupular, contracted at the mouth. 10. P. Elmeriti. . Stem 2 to 3 cm in diameter. Leaf-sheaths sprinkled with dark- purple scales. Segments rather numerous, equidistant, 8 to 9 cm apart on each side of the rachis, 3-costulate, ensiform, about 60 cm long, 4 to 4.5 em wide, paler or subglaucous, and not sprinkled with microlepidia underneath, the apices acu- minate-caudate and very slightly falcate. Spadix with 14 or 15 spirally alternate branches. Immature fruits fusiform, 14 to 15 mm long (ovate-ellipsoid at complete maturity?). Fruiting perianth contracted at the mouth..11. P. urdanetana. . Of medium size (?). Segments lanceolate-ensiform, acuminate, unequal, straight, 60 cm long, 3.5 cm wide, their lower surface in the dry specimens reddish-brown and densely sprinkled with extremely minute, light-colored dots (microlepidia). Spadix with strongly flattened branches. Fruit thickly fusi- form, equally narrowing to both ends, 15 to 18 mm long, and about 1 cm thick. Fruiting perianth very low, contracted at the mouth 12. P. samarana, . Of medium size. Segments very numerous, equidistant, ap- proximate, long, narrow, slightly falcate, very acuminate, all unicostulate. Spadix with 5 or 6 much flattened branches. Fruit relatively large, ovoid-ellipsoid, conical-mammillate at the apex, the base narrow and subpedicelliform when dry, 23 to 832 mm long, 13 to 14 mm thick. Seed ovoid, suddenly prolonged at the base into a slender caudiculum; embryo fovea broad and slightly oblique; vascular branches of the integument simple on the raphal or dorsal side and anas- tomosing anticously. Fruiting perianth truncate, not con- tracted at the mouth...... 13. P. urosperma. d@’. Fruits 3-seriate, at least in the lower part of the branches. . e’, Segments straight, not faleate at their apices. Plants of medium size. Fruits small. f'. Segments numerous, very approximate, inserted at a very acute angle, unicostulate, lanceolate-ensiform, quite straight, concolorous, the apex bifid, its divisions acuminate. Spadix with several spirally inserted branches. Fruits 3-seriate from the base to the end of the branches, ovoid-ellipsoid, the apices mammillate, 13 to 15 by 6 to 7 mm. Seed globular-ovoid, rounded above, not caudiculate at the base; embryo fovea oblique; vascular branches of the integument few (5 or 6), not or only very slightly divided. Fruiting 316 The Philippine Journal of Science 1919 perianth very low, 1.5 mm high, 4 mm broad, contracted at the mouth 14, P. rigida. f’. Segments numerous, equidistant, 6 to 7 cm apart on each side of the rachis,.ensiform, 1- or 2-costulate, straight, rigid, concolorous, very acuminate, bifid at the apex. Spadix with several, triquetrous, spirally inserted branches. Fruits 3-seriate in the lower part of the branches, bifarious near the end, small, 12 to 15 by 7 to 8 mm, ovoid-ellipsoid; seed ovoid, not caudiculate at the base; embryo fovea very oblique; vascular branches of the integument 5, all slightly anastomosing. Fruiting perianth low, 2 mm high, 4 mm broad, slightly contracted at the mouth. 15. P. Woodiana. f*'. Segments numerous, equidistant, thickish and rigid, 1-cost- ulate, concolorous, narrow, very long-acuminate. Spadix with several, 3-gonous, spirally inserted branches. Fruit broadly ovoid, 18 by 8 to 9 mm. Seed broadly ovoid; embryo fovea almost horizontal; vascular branches of the integument 8 or 9, of which two are undivided and pass over the apex, and 2 or 8 on each side are arched and slightly anastomosing. Fruiting perianth 2 mm high, 4 mm broad, contracted at the mouth........ 16. P. sclerophylla. e*. Segments falcate at their apices, numerous, equidistant, rigid, concolorous, narrow, very long-acuminate, 1- or 2-costulate. Spadix with several spirally inserted branches, trigonous in their lower part and flattened above. Fruit 3-seriate in the lower part of the branches, and bifarious above, rather narrowly ovoid-ellipsoid, narrowing a good deal to both ends, 12to13 by 6mm. Seed caudiculate at the base; embryo fovea slightly oblique. Fruiting perianth contracted at the mouth. 17. P. negrosensis. e. Spadix large with numerous branches inserted spirally at different levels. Large arboreous plants. @. Fruits biseriate. e’. Segments very numerous, uniform, equidistant, ensiform, quite straight, rigid, very acuminate, very strongly 2-costulate, more or less deeply bifid at their apices, subconcolorous, very finely granulate-scabrid on the secondary and tertiary nerves on the lower surface. Fruits rather large, ovoid-ellipsoid, narrowing to both ends, 24 to 25 by 13 to 14 mm; pericarp containing several layers of capillary fibers. Seed obsoletely caudiculate at the base, the embryo fovea very broad, oblique. Fruiting perianth 3.5 mm high, 8 mm broad, not or only very slightly contracted at the mouth................ 18. P. insignis. f'. Fruit shorter than in the species and more ventricose, 20 to 22 by 13 to 14 mm P. insignis var. gasterocarpa. f°. Fruit narrower than in the species or thickly fusiform, 20° to 22 by 8 to 9 mm P. insignis var. leptocarpa. f. Fruit very broadly ovoid, 20 to 22 by 13 to 15 mm. Pericarp with rigid fibers, rendering the surface of the fruit striate. Leaflets dusty-subglaucous underneath, P. insignis subsp. Loheriana. XIV, 3 Beccari: The Palms of the Philippine Islands 317 e’. Segments very numerous, unicostulate, very approximate by twos on each side of the rachis, ensiform, quite straight, very acuminate. Fruits ovoid, broad at the base and suddenly apiculate-mammillate, 20 by 12 mm. Seed ovoid, blunt; vascular branches of the integument simple on the raphal side, elsewhere rather closely anastomosing. Fruit- ing perianth broadening at the mouth........ 19. P. batanensis. e’. Segments elongate-lanceolate, 3-costulate, 1 meter long or more, spadix with slender, strongly flattened, pendulous branches. Fruit of medium size, 20 by 12 mm, slightly obovoid or ovoid-olivaeform; pericarp somewhat fleshy, traversed by a few slender fibers. Seed relatively small, 13 by 8 to 9 mm. Fruiting perianth deeply cupular or subcampanulate, not contracted at the mouth, 4 mm high, 6 mm broad. 20. P. basilanensis. e*. Segments equidistant, large, straight, 2- to 3-costulate, of a rather herbaceous texture, the basilar and the intermediate segments acuminate, the upper with as many not very deep incisions as there are costae, and with the resulting divisions shortly 2-toothed. Fruits small, 15 by 9 mm, obovoid, rounded above. Seed oblong, the embryo fovea very slightly oblique; vascular pene of the integument almost simple. 21. P. speciosa. d’. Fruit 3-seriate. Segments ensiform, strongly bicostulate, the apex divided into two straight points. Fruits rather large, ovoid-ellipsoid, with a conical apex, 25 to 28 by 15 to'17 mm. Seed broadly ovoid, rounded above, the embryo fovea some- what oblique; vascular branches of the integument much branched and forming a network all around the seed. Fruit- ing perianth somewhat contracted at the mouth. 22. P. sibuyanensis. 1. PINANGA MACULATA Porte in Illustr. Hort. 10 (1863) 92, 188, t. 361, Belgique Hort. (1863) 236; H. Wendl. in Kerch. Les Palms 253; Drude in Bot. Zeit, (1877) 6387, t. 5, f. 12-12 (ovulum); Becc. Male- sia 3 (1886) 145, et in Perkins Fragm. Fl. Philip. (1904) 48, et . in Webbia (1905) 325; Curtis’s Bot. Mag. t. 8011. PALAWAN, along Ewiig River, Merrill 712 (sterile specimen). Of this fine small Pinanga I have seen only the sterile speci- men cited above. The leaves appear distinctly mottled even in the dry condition and have a petiole about 20 cm long; the blade, 30 cm long, has 9 or 10 main costae on each side of the rachis, tapers gradually to the base, and is divided down to its lower one-third into two lobes. Fertile specimens received from the botanic garden at Bui- tenzorg, Java, under the name of Pinanga maculata, do not seem to me to differ from P. disticha; they have the uniformly green blade parted only to the middle, and the petiole is shorter and thicker than in the Palawan plant. Probably the leaves of young 164891——5 318 The Philippine Journal of Science 1919 plants only are variegated and possibly also in gardens the very similar Pinanga disticha may also go under the name of P. ma- culata. In fact Ridley‘ actually reduces P. maculata to P. dis- ticha. I think, however, that true Pinanga maculata of the Phil- ippines is distinguishable from the Malayan P. disticha by the leaves having a longer petiole and the blade being more deeply bilobed. The flowers and the fruit probably offer other diag- nostic characters, but I have not seen fertile specimens of the wild Philippine plant. 2. PINANGA GEONOMAEFORMIS Bece. in Philip. Journ. Sci. 4 (1909) Bot. 602. Luzon, Tayabas Province, For. Bur. 10155 Curran: Laguna Province, San Antonio, Bur. Sci. 20474 Ramos: Rizal Province, Pantao, Loher 7055 in Herb. Kew. and Becc. 3. PINANGA MODESTA Becc. in Philip. Journ. Sci. 2 (1907) Bot. 223. MINDANAO, Davao District, Copeland: Lanao District, Camp Keithley, Mary Strong Clemens 487: Bukidnon Subprovince, Bur. Sci. 15751 Féniz. BASILAN, For. Bur. 3987 Hutchinson. 4. PINANGA ISABELENSIS Bece. sp. nov. Gracilis, caudice 2 ad 2.5 cm diametro. Frondium vaginis squamuloso-pubescentibus, segmentis aequidistantibus, utrinque 8 vel 9 inter se remotis, elongato-ensiformibus et subfalcatis, acuminatissimis, 2- ad 4-costulatis, subtus glabris vel vix secus nervos papillosis; spadicis ramis paucis (5 vel 6); fructibus exacte distichis, anguste ovoideo-conicis, vel in dimidia inferiori parte latiusculis et superne conico-attenuatis, 16 ad 18 mm longis, 6 mm latis; semine elongato, apice conico, fovea em- brionali obliqua; integumenti ramis vascularibus 3 vel 4, fere indivisis; perianthio fructifero cupulari truncato. Luzon, Isabela Province, Biocbian Bay, Bur. Sci. 10660 McGregor, in beach forest, about 50 meters from the sea. A slender plant, the stem 2 cm in diameter, with short inter- nodes (3 to 4cm long). The wood is very hard, and it is appa- rently a slow-growing plant, as the surface of the stem is covered with crustaceous lichens in its lower and older parts. Leaf-sheaths finely and softly tomentose. The only leaf seen by me is 90 cm long in the piniferous part; the petiole is 20 cm long, slightly scurfy, or sprinkled with brown scales; the segments are subopposite, very few (6 or 7) and 8 to 9 cm apart on each side of the rachis; two or three of the lowest segments are very narrow; the intermediate ones are elongate-ensiform, *(Mat. Fl. Mal. Penins. Monocot.°2 (1907) 189). XIV, 3 Beccari: The Palms of the Philippine Islands 319 slightly sigmoidal and faleate above, very long and slenderly acuminate, 3- or 4-costulate, about 50 cm long and 3.5 to 4.5 em broad, reddish-brown underneath in dry specimens and under a strong lens appear finely hairy-papillose along the nerves; the two terminal segments are narrowly linear, 2 to 2.5 cm broad and 30 cm long, truncate and shortly incised-toothed at the apex. Spadix small with a very few (5 or 6) flattened, about 13 cm long, floriferous branches. Fruits exactly biseriate, elongate- ovoid-conical, 16 to 18 mm long, 6 mm thick, having their broad- est part below the middle and the base suddenly contracted into a caudiculum, the latter extending into the perianth. Seed ovoid- conical, 11 mm long, 5 mm thick, with the areola of the embryo oblique; the vascular bundles of the integument are 3 or 4 only, very slightly anastomosing. Fruiting perianth cupular, trun- cate, not contracted at the mouth, 2 mm high, 3.5 mm wide. 5. PINANGA HETEROPHYLLA Bece. sp. nov. Gracilis, caudice 1.5 cm diametro. Frondium segmentis utrin- que 7 vel 8, inaequidistantibus, valde inaequalibus, 1- ad 4-costu- latis, falcato-sigmoideis, longe acuminatis, chartaceis, subtus glabris; spadicis ramis paucis (5) triquetris; fructibus 3-seriatis, parvis, anguste ovato-ellipsoideis, utrinque aequaliter attenuatis, acutis, 11 ad 12 mm longis, 5 ad 5.5 mm latis; semine ovato, superne conico; integumenti ramis vascularibus 8 ad 10, modice anastomosantibus. NEGROS, Mount Maripari, For. Bur. 17346 Curran, September, 1909, in forests... Slender, stem 15 mm in diameter. Leaf-sheaths elongated, 22 cm in one specimen, the mouth very oblique (not split) for the extent of 5 to 6 cm, strongly striate, and more or less covered with rusty-brown scales; the ligule at the base of the petiole is membranaceous, triangular, acuminate, 12 to 15 mm long; petiole 20 to 25 em long, 3-gonous, also covered with rusty scales; the piniferous part is about 85 cm long, and has 7 or 8 segments on each side; the segments are very unequal, inequidistant, often separated by vacant spaces as much as 8 cm long; the lower surface is glabrous and paler than the upper, narrowly falcate- sigmoid, very acuminate; the intermediate segments are 35 to 40 cm long and of very variable breadth, some of them being 1-costulate, and only 7 cm broad, while others are 2- to 4-cos- tulate and 2 to 4 cm broad; the two terminal segments are united by their bases and are shorter and broader than the others, pluricostulate, incised-toothed, the teeth divided again into two short, narrow, acuminate, secondary teeth. The spadix has very 320 The Philippine Journal of Science 1919 few branches (5 in the specimen at hand), spreading, trique- trous, 7 to 9 cm long. Fruits 3-seriate, small, narrowly ovoid- ellipsoid, equally attenuate at both ends, their apex acute, 11 to 12 mm long, 5 to 5.5 mm fhick. Seed ovoid, conical in its upper part, 7.5 mm long, 5 mm thick, noncaudiculate at the base; areola of the embryo small, slightly oblique; vascular branches of the integument 8 to 10, ascending from the base, slightly anastomosing. Fruiting perianth cupular, truncate, slightly contracted at the mouth. This species resembles Pinanga Elmerii, even as to its dimen- sions, but the fruits are 3-seriate. 6. PINANGA BARNESI! Bece. in Webbia 1 (1905) 320, Philip. Journ. Sci. 3 (1908) Bot. 340, 6 (1911) Bot. 229, Pinanga Barnesii var. macrocarpa Bece. in Philip. Journ. Sci. 2 (1907) Bot. 227. Luzon, Bataan Province, Lamao River, For. Bur. 122 Barnes (type specimen), For. Bur. 2762 Meyer: Tayabas Province, Lucban, Elmer 7924; Kabibihan, Bur. Sci. 18248 Ramos: Cagayan Province, For. Bur. 17756 Ramos: Rizal Province, Bur. Sci. 123419 Ramos: No precise locality Loher 7069, 7068 in Herb. Kew.: Union Province, Mount Tonglon, Loher 7067 in Herb. Kew. POoLiLio, Bur. Sci. 6937 Robinson, Bur. Sci. 10466 McGregor. BABUYAN ISLANDS, Camiguin, Bur. Sci. 4144 Fénix. Minporo, McGregor 275 (var. macrocarpa). The size of the fruit in the cited specimen is very variable as many are immature. 7. PINANGA COPELAND! Becce. in Webbia 1 (1905) 317. MINDANAO, Davao District, Todaya, Copeland 1283 Herb. Manila (type specimen), Elmer 10467: Butuan Subprovince, Weber 1134, Bur. Sci. 15918 | Fénix: Agusan River, Merrill 7281; Mount Urdaneta, Elmer 18875. NE- cros, Gimagaan River, Whitford 1669. Luzon, Camarines Province, Phil. Pl. 1594 Ramos. 8. PINANGA CURRANII Becc. in Philip. Journ. Sci. 4 (1909) Bot. 226. - PALAWAN, near Puerto Princesa, For. Bur. 3515 Curran. 9. PINANGA PHILIPPINENSIS Becc. Malesia 3 (1887) 180, Perk. Fragm. Fl. Philip. (1904) 48, Webbia 1 (1905) 44. Luzon, Sorsogon Province, Mount Bulusan, Vidal 3950, 4064, in Herb. Bece. (type specimens): Bataan Province, Mount Mariveles, Whitford $33, Elmer 6825, Merrill $316 (specimens very similar to the typical ones) : Rizal Province, Mount Matulid, Loher 7065 in Herb. Kew.: Tayabas Prov- ince, Mount Banahao, Elmer 9800 (with very narrow, 1-costate, falcate segments): Cagayan Province, For. Bur. 11817 Klemme: Laguna Province, Mount Maquiling, Bur. Sci. 18679 Ramos: Benguet Subprovince, Bur. Sct. 12907 Fénix. The last four are with difficulty distinguishable from P. elmerii Becce. 10. PINANGA ELMERI! Bece. in Webbia 1 (1905) 322, Philip. Journ. Sci. 3 (1908) Bot. 341. e XIV, 3 Beccari: The Palms of the Philippine Islands 821 BaBuyYAN ISLANDS, Camiguin, Bur. Sci. 4149 Fénix. Luzon, Benguet Subprovince, Baguio, Elmer 6067, For. Bur. 10891 Curran, Bur. Sci. 12907 Fénix: Ifugao Subprovince, Mount Polis, Bur. Sci. 19782 McGregor (a slender plant with very narrow leaflets, perhaps representing a distinct variety) : Ilocos Norte Province, Mount Piao, For. Bur. 14011 Merritt & Darling: Nueva Vizcaya Province, Bur. Sci. 11870, 20204, 20205 McGregor: Bataan Province, Mount Mariveles, Merrill 3846 (typical), For. Bur. 1578 Borden, Whitford 130: Rizal Province, Bur. Sci. 13418 Ramos: Laguna Province, Mount Maquiling, Bur. Sci. 17119 Robinson: Zambales Province, Mount Tapulao, For. Bur. 8263 Curran & Merritt: Tayabas Province, Mount Pular, Bur. Sci. 19416 Ramos: Without precise locality, Loher 7064, 7063 in Herb. Kew. Mrnporo, Mount Halcon, Merrill 5555. Leyte, Dagami, Bur. Sci. 15259 Ramos. Pinanga Elmerii is very closely related to P. philippinensis, and at times is distinguishable from it only with difficulty. 11. PINANGA URDANETANA Becc. sp. nov. in Leafl. Philip. Bot. MS. MINDANAO, Agusan Province, Camilanan Peak, Mount Urdaneta, Elmer 14137. > 12. PINANGA SAMARANA Bece. sp. nov. Mediocris. Frondium segmentis intermediis rectis (apice mi- nime falcato) inaequalibus, lanceolato-ensiformibus, 6 ad 7 cm inter se dissitis, 2-3-costulatis, 60 cm longis, 3.5 ad 6 cm latis, in sicco brunneis et subtus punctulis minutissimis pallidis con- spersis; spadicis ramis valde complanatis; fructibus bifarius, crasse fusiformibus vel ovoideo-ellipticis, nonnihil utrinque atte- nuatis ac acutis, 15 ad 18 mm longis et circiter 10 mm crassis; perianthio fructifero parvo, 15 mm alto, 3 mm lato et in ore parum constricto. Samar, Mount Cauayan, Bur. Sci. 17585 Ramos, March 29, 1914. This is very incompletely known, as of it I have seen only a small portion of a leaf and fragments of the spadix bearing not thoroughly mature fruits. In the dry specimens the leaflets have the upper surface green, and the lower very minutely dotted and of a reddish-brown color. In general habit Pinanga samarana resembles P. urdanetana, but the fruits of the former are larger than those of the latter in which the leaflets of the dry specimens are brownish above and grayish-glaucescent un- derneath, without any traces of the very minute and numerous dots (microlepidia) peculiar to P. samarana. 18. PINANGA UROSPERMA Becc. in Philip. Journ. Sci. 3 (1908) Bot. 341. BaBUYAN IsLANDS, Camiguin, Bur. Sct. 4044 Fénizx. 14. PINANGA RIGIDA Becc. in Leafl. Philip. Bot. 2 (1909) 642. NEGROs, Cuernos Mountains, Elmer 10187. . 399 The Philippine Journal of Science 1919 15. PINANGA WOODIANA Becc. in Philip. Journ. Sci. 4 (1909) Bot. 604. MINpDoRO, Mount Halecon, Merrill 5680. MINDANAO, Davao District, Mount Apo, Elmer 10485 (the fruit a little larger than in the Mindoro specimen). 16. PINANGA SCLEROPHYLLA Bece. in Philip. Journ. Sci. 4 (1909) Bot. 603. MINDoRO, Mount Haleon, Merrill 4468. 17. PINANGA NEGROSENSIS Becce. in Leafl. Philip. Bot. 2 (1909) 642. Negcros, Dumaguete, Cuernos Mountains, Elmer 10030. 18. PINANGA INSIGNIS Becce. in Philip. Journ. Sci. 2 (1907) Bot. 223. MINpDoRO, Bongabong River, Whitford 1288 (type specimen). MINDANAO, Agusan Province, Mount Urdaneta, Elmer 13950: Surigao Province, Bolster 222. LuzoN, Nueva Vizcaya Province, Bur. Sci. 11283 McGregor: Laguna Province, San Antonio, Bur. Sci. 15111 Ramos: Tayabas Province, Lucban, Elmer 8028, Bur. Sci. 13286 Ramos: Rizal Province, Siya Bundoc, Loher 7059, Herb. Kew. (All the Luzon specimens agree very well with the typical Mindoro form.) LEYTE, Wenzel 534, perhaps representing a local variety with narrower and more approximate fruits, but the fruits are very immature. PINANGA INSIGNIS Becc. var. GASTEROCARPA Becc. Luzon, Laguna Province, Mount Maquiling, For. Bur. 7791 Curran & Merritt, Loher 7093, Herb. Kew. NecGros, For. Bur. 17348 Curran; Faraon, For. Bur. 18560 Meyer & Foxworthy. MAsBATE, For. Bur. 11555 Rosen- bluth. PINANGA INSIGNIS Becc. var. LEPTOCARPA Becc. NeEGros, Mount Silay, Whitford 1621. PINANGA INSIGNIS Bece. subsp. LOHERIANA Bece. Luzon, Laguna Province, Mount Banahao, 1,000 meters’ altitude, Loher 7092, Herb. Kew. and Becc. 19. PINANGA BATANENSIS Bece, in Philip. Journ. Sci. 3 (1908) Bot. 340. BATAN ISLANDS, Batan, Santo Domingo de Basco, Bur. Sci. 3841 Fénix. 20. PINANGA BASILANENSIS Bece. sp. nov. Major; frondium segmentis intermediis ultra metralibus, 9 cm latis, elongato-lanceolatis, 3-costulatis, rectis; spadicis ramis gracilibus, fructiferis pendulis, valde compressis; fructibus e ra- morum basi usque ad apicem distichis, ovoideo-olivae formibus, aut laeviter obovatis, 2 em longis, 12 mm latis; pericarpio non- nihil carnoso, fibris tenuibus paucis percurso; semine pro rata parvo, 15 mm longo, 8 ad 9 mm crasso, oblongo, apice rotundato; perianthio fructifero profunde cupulari vel subcampanulato in ore non coarctato, 4 mm longo, 6 mm lato. Apparently a large species of the dimensions of P. insignis. The only leaf-segment seen by me is strongly 3-costulate, 106 em long and 9 cm broad in its intermediate part, elongate- XIV, 3 Beccari: The Palms of the Philippine Islands 923 lanceolate, straight, narrowed a little in its lowest part but, nevertheless, attached to the rachis by a broad base; it narrows also in its upper part, and has the upper margin prolonged into an acuminate straight point, whereas the lower margin, toward the apex, is marked by several narrow teeth, the upper teeth gradually surpassing in length those below; in the dry state the segment is concolorous on both surfaces, the lower surface is glabrous, but is covered (occasionally?) with small granu- lations pertaining to the parenchyma, which are wanting or are much less visible on the upper surface. The spadix is ap- parently large and has numerous slender, floriferous branches, about 30 cm long; the latter are strongly flattened and bear the fruits in two series, even on their lowest parts. Fruit ovoid- olivaeform or very slightly obovoid, apex round, slightly nar- rowed toward the base, 2 cm long, 12 cm thick; the pericarp contains a few slender fibers and has the appearance of having been rather fleshy at maturity. Seed oblong, rounded above, relatively small, 13 mm long, 8 to 9 mm thick; the embryo area slightly oblique; the vascular branches of the integument rather numerous, but slightly anastomising. Fruiting perianth rather deeply cupular-subcampanulate, broadening a little at the mouth, 4 mm high and 6 mm broad. BASILAN, Bur. Sci. 16119 Reillo. This is related to Pinanga insignis, but is distinguishable by the fruiting perianth being slightly widened, not contracted, at the mouth and relatively very deep and nearly campanulate ; by the fresh mature fruit being pulpy and with few fibers; by the leaf-segments having the lower margin near the apex incised- toothed, the teeth looking like ascending steps (this character is probably present in some segments only). 21. PINANGA SPECIOSA Becc. in Webbia 1 (1905) 316. MINDANAO, Davao District, Todaya, Copeland 1265, Elmer 10484, locally known as sadawag: Lanao District, Camp Keithley, Mary Strong Clemens $74: Zamboanga District, For. Bur. 9271 Whitford & Hutchinson: Agusan Province, Cabadbaran, Elmer 13739, 13941, 14023, locally known as salacon and saraway. A large palm with a relatively slender stem, 8 to 10 m high, 10 to 12 em in diameter. Leaves large, the segments numerous, uniform, equidistant, 7 to 10 cm apart, of a not very rigid structure (perhaps rather herbaceous when fresh), glabrous, straight, lanceolate-ensiform, narrowed somewhat toward the base; the lower and intermediate segments very long-acuminate, usually 2-costulate, 1 to 1.2 m long, 8 to 10 cm wide, the upper 324 The Philippine Journal of Science 1919 ones gradually shorter, not very deeply 2- or 38-lobed at the apex, the lobes shortly 2-toothed. Spadix large, divided into several, spirally alternate, strongly flattened floriferous branches, the lower of which are 30 to 40 cm long, and the upper gra- dually shorter. The fruits are exactly distichous throughout the entire length of the branches, small, obovoid, apex round, base (in the part included in the perianth) acute, 15 to 17 mm long and 9 mm thick. Seed oblong, rounded above, 10 to 11 mm long, 7 mm thick, with the areola of the embryo very slightly oblique; vascular branches of the integument almost simple. Fruiting perianth shallowly cupular, very slightly broadening at the mouth, 2.5 to 3 mm high, 5 mm broad. This was first described from very incomplete material. The various specimens with fruits, received later, correspond very well to the typical ones, which had flowers only. All the her- barium specimens I have examined have acquired a uniform yellowish tint on both surfaces, are not very rigid but rather flexible, and look as if they had been rather herbaceous in the living plant. 22. PINANGA SIBUYANENSIS Bece. sp. nov. Inter majores, subelata, caudice ad 9 m longo, 15 cm diametro; frondium segmentis numerosis, aequidistantibus, late ensifor- mibus, usque ad apicem rectissimis, utrinque virentibus, subtus paullo pallidioribus, validissime bicostulatis, apice inaequaliter bifidis; spadicis ramis numerosis, spiraliter insertis; fructibus 3-seriatis, majusculis, 25 ad 28 mm longis, 15 ad 17 mm latis, ovato-ellipticis, utrinque aequaliter attenuatis; semine late ovato, superne rotundato; fovea embryonali nonnihil obliqua; integu- menti ramis vascularibus numerosis, undique reticulatim anas- tomosantibus ; perianthio fructifero. ad faucem nonnihil coarctato. SIBUYAN, Mount Giting-Giting, Elmer 12425. _ This is allied to Pinanga insignis, but is apparently a smaller plant yet with larger fruits, and with the vascular branches of the integument forming a network all around the seed. ACTINORHYTIS Wendland et Drude ACTINORHYTIS CALAPPARIA Wendl. et Dr. in Linnaea 39 (1875) 184. MINDANAO, Davao District, Todaya, Elmer 11238, with the local name tangalo. : A fine but almost useless palm, very widely spread in the Malay Archipelago and in the Moluccas, but probably of foreign origin in the Philippines. XIV, 3 Beccari: The Palms of the Philippine Islands 325 HETEROSPATHE Scheffer Conspectus of the species. a’. Large trees. b%. A tree 8 to 10 m high. Leaflets with strong secondary nerves, the midrib without paleolae underneath. Spadix three times branched, floriferous branches slender. Fruit globular, 7 to 7.5 mm in diam- eter, excentrically apiculate, the surface granulose from short scattered sclerosomes. Seed spherical 1. H. elata. b?. A tree as much as 9 m high, 12 cm in diameter. Leaflets having rather distinct secondary nerves, and the midrib furnished under- neath with conspicuous brown paleolae. Spadix three times branched; floriferous branches thickish (2.5 mm thick). Fruit ovoid, 1 em long, 7 mm thick, having the point conical and slightly oblique and the surface shagreened by linear sclerosomes. Seed globose-ovoid, blunt : 2. H. sibuyanensis. a’. Shrubs or small trees. b*. Stem slender, 1 to 3 m high, 2 to 3 em in diameter. The largest leaflets 25 to 30 cm long, 10 to 15 mm broad, secondary nerves faint. Spadix twice branched in its basal part, simply branched above. Fruit ovoid, 10 to 11 mm long, 6 mm thick, very suddenly, and nearly centrally, apiculate, the surface closely shagreened by con- spicuous, shortly fusiform sclerosomes. Seed globose-ovoid, blunt. 3. H. philippinensts. b*. More robust than the preceding, 3 to 5 m high. Stem 4 to 5 cm in diameter. Leaflets 35 to 40 cm Jong, 2 to 2.5 cm wide, the secondary nerves rather distinct. Spadix twice branched. Fruit ovoid-ellipsoid, narrowing above to a conical, nearly symmetrical point, 9 to 11 mm long, 5 mm thick. Seed ovoid, acute. 4. H. negrosensis. 1. HETEROSPATHE ELATA Scheff. in Ann. Jard. Bot. Buitenz. 1 (1876) 141, 162; Becc. in Webbia 1 (1905) 328; Farnsworth Philip. Bur. Ed. Bull. 54 (1915) 66, 67, plate. Luzon, Tayabas Province, For. Bur. 10352 Curran: Camarines Province, For. Bur. 10481 Curran: Cagayan Province, For. Bur. 17064 Curran: La- guna Province, Mount Maquiling, For. Bur. 19749 Villamil. CrBu, Piper 4. MINDANAO, Zamboanga District, For. Bur. 9030 Whitford & Hutchinson: Davao District, Todaya, Elmer 11968. Mtnporo, Bongabong River, Whitford 1895. Local names: Saguisi, sagisé, seguise, niog-niogan, tagisi, balantug, salaniog. 2. HETEROSPATHE SIBUYANENSIS Becc, in Leafl. Philip. Bot. MS. SIBUYAN, Mount Giting-Giting, at about 230 meters’ elevation, Elmer 12850, local name bilis. This is the representative of Heterospathe elata in Sibuyan. It differs from H. elata in the leaflets being furnished with con- spicuous paleolae on the lower surface; in the fruit being ovoid, with an obliquely conical upper part; and in the seed being slightly longer than broad (not spherical). 326 The Philippine Journal of Science 1919 3. HETEROSPATHE PHILIPPINENSIS Becc. in Philip. Journ. Sci. 4 (1909) Bot. 610. Ptychoraphis philippinensis Becc. in Ann. Jard. Bot. Buitenz. 2 (1885) 90, Webbia 1 (1905) 47, Cuming Exsicc. No. 1476 in Herb. Webb. at Florence. Luzon, Benguet Subprovince, For. Bur. 10856 Curran, For. Bur. 18019 Merritt: Ifugao Subprovince, Mount Polis, Bur. Sci. 19784, 19785 McGregor: Nueva Vizcaya Province, Bur. Sci. 8198 Ramos: Camarines Province, For. Bur. 21752 Fischer: Rizal Province, Mount Matulid, Loher 7054. LEYTE, Dagami, Bur. Sci. 15260 Ramos. 4. HETEROSPATHE NEGROSENSIS Bece. in Philip. Journ. Sei. 4 (1909) Bot. 611. NeEcrOS, Mount Silay, Whitford 1539; Cuernos Mountains, Elmer 9439, 10147; Mount Maripari, For. Bur. 13639 Curran & Foxworthy. Bur. Sci. 9443, 9358 Robinson are doubtfully referable to Heterospathe negrosensis ; both were collected at about 900 meters’ altitude on Mount Binuang, In- fanta, Tayabas Province, Luzon. In regard to 9358 the collector notes that it is a plant 2 m high with the trunk 7 cm in diameter (this is probably the diameter of the upper part of the stem when covered with the mass of petiolar bases) and of the other that it is 4m high. PTYCHORAPHIS Beccari Conspectus of the species. a’. Fruit longer than broad. . b*. Of medium size. Leaflets furnished underneath with a few paleolae on the midrib only. Spadix twice branched. Fruit narrowly ovoid, tapering above to a slightly oblique, conical point, 10 to 12 mm long, 5 mm thick. Seed ovoid-ellipsoid, subacute, 7 mm long, 4 mm thick; vascular branches of the integument loosely anastomosing. 1. Pt. microcarpa. 6’. Of medium size. Leaflets furnished underneath with paleolae, often on three nerves. Spadix twice branched. Fruit ovoid-ellipsoid, not or very slightly and asymmetrically obtuse-acuminate, 12 mm long, 7 mm thick. Seed ovoid, rounded at both ends; vascular branches of the integument very closely anastomosing. 2. Pt. intermedia. b*. Robust, stem about 15 m high. Spadix thrice branched. Leaflets rigid, without paleolae on the lower surface ; secondary nerves well marked; margins somewhat thickened. Fruit narrowly ovoid, tapering above to a conical, slightly oblique point, 12 to 13 mm long, 5 mm broad. Seed ovoid-ellipsoid, acute, 9 mm long, 4.5 mm thick; vascular branches of the integument loosely anastomosing. 8. Pt. Elmerii. a’. Fruit spherical. Stem about 6 m high. Leaflets long-acuminate, the point slightly faleate. Spadix thrice branched, floriferous branches 12 to 15 cm long, 2.5 to 8 mm thick. The bracteoles of the female flower form a regular cupular caliculum. Fruiting perianth shallowly cupular. Fruit spherical, 7 mm in diameter, with the remains of the stigmas nearly central and apical, the surface not granulose. Seed spherical, 5.5 mm in diameter 4. Pt. cagayensis. XIV,3 Beccari: The Palms of the Philippine Islands 327 1, PTYCHORAPHIS MICROCARPA Becc. sp. nov. : Mediocris; frondium segmentis numerosissimis, rectissimis, majoribus 70 cm longis, 3.5 cm latis, costa media tantum subtus paleolis induta; spadice duplicato-ramoso, ramis floriferis rigidis, crassiusculis; fructibus anguste ovoideis, superne vix oblique co- nice attenuatis, 10 ad 12 mm longis, 5 mm crassis; semine ovato-elliptico, subacuto, 7 mm longo, 4 mm crasso; integumenti ramificationibus vascularibus laxe anastomosantibus. Apparently of medium size. Leaves rather large, perhaps about 3 m long, very regularly pinnate; petiolar part smooth, strongly convex underneath, deeply channeled above (at least in its uppermost part) ; rachis also convex, smooth and glabrous underneath, 2.5 cm wide in its basal part. Leaflets very num- erous, equidistant, ensiform, quite straight, attached to the rachis by a rather narrow base, with the margins there strongly reduplicate, very long acuminate in their upper part to a slender undivided point, which however has a tendency to split down the middle, firmly papyraceous, glabrous and subconcolorous on both surfaces, the margins slightly thickened; midrib almost equally prominent on both surfaces, provided, on the lowér surface, especially near the base, with distant, brown, linear paleolae; the latter at times present (on each side of the mid- costa) on a secondary nerve, which is usually stronger than the other; tertiary nerves not very sharp, being immersed in the parenchyma; transverse veinlets obsolete; intermediate seg- ments up to 70 cm long, and 3.5 cm wide; the lowest segments considerably smaller; the upper ones gradually diminish in length and breadth; all, however, are of one shape: Spadix over 1 meter in length, but about one-half of it consists of an undivided, strongly flattened, peduncular part; the latter is 17 to 18 mm wide, the margins obtuse, covered with a very dark tobacco- colored, furfuraceous scurf; panicle diffuse, twice branched, its primary divisions short, angular, and subdivided into 2 or 3 floriferous branches; the latter are rigid, thickish (4 mm thick) throughout their entire length (not including the flowers), 15 to 22 cm long, uniformly covered with glomerules from the base to the apex, each composed of three flowers. Spathes two, falling to pieces at the time of fruiting, the outer, or basal, spathe coriaceous, brittle; the second inserted on the peduncular part 15 cm above the other. Glomerules not very regularly yet rather closely arranged, although often with vacant spaces inter- posed, around the branches, on very shallow scrobiculi. Male flowers * * *, Female flowers * * *. Fruiting perianth 5 mm high, the petals twice as long as the sepals; the bracts of the 328 The Philippine Journal of Science 1919 female flowers form a shallow truncate cup, 3 mm wide. Fruit narrowly ovoid, 10 to 12 mm long, 5 mm broad, slightly asym- metrical, being faintly obliquely conical in its upper part, the apex marked by a minute mammillaeform areola, bearing the remains of the stigmas; the surface is dull and even (not roughened by underlying sclerosomes), the epicarp thinly crus- taceous, the mesocarp with 3 or 4 layers of strongly flattened, yet soft, fibers, the endocarp very thin, pellicular. Seed ovoid-ellip- soid, rather acute, 7 mm long, 4 mm thick; hilum sharply im- pressed all along one side; testa brown and polished; the vascular branches of the integument form a loose network all around the seed; albumen radiately ruminate; embryo basal. CAMIGUIN DE MISAMIS, a small island north of Mindanao, Bur. Sci. 14678 Ramos. This species is closely related to Ptychoraphis intermedia, but is distinguishable by its smaller fruits, considerably narrowing in their upper part, and by the ellipsoid, acute seeds having the vascular branches loosely anastomosing. 2, PTYCHORAPHIS INTERMEDIA Bece. in Leafi. Philip. Bot. MS. MINDANAO, Agusan Province, between Duras and Canilan Peaks, Mount Urdaneta, over 1,000 meters’ altitude, Elmer 13663, September, 1912, Manobo name marighoy. BILIRAN, Bur. Sci. 18107 McGregor, June, 1914, this specimen agreeing very well with that from Mindanao. This species is closely related to Ptychoraphis microcarpa, from which it differs in its larger spadix having longer and thicker fructiferous branches and in the leaflets having paleolae not only on the midcosta but also on three nerves underneath, but more especially in the shape and size of the fruit and in the seed, which is rounded at both ends and marked by the close network of the vascular branches of the integument. Both Ptychoraphis microcarpa and Pt. intermedia are more closely related to Pt. angusta, of the Nicobar Islands, than to Pt. singaporensis, of the Malay Peninsula and Singapore. 8. PTYCHORAPHIS ELMERII Becc. comb. nov. Heterospathe Elmerii Bece. in Leafi. Philip. Bot. 2 (1909) 646. NeEGrRos, Dumaguete, Cuernos Mountains, Elmer 9559, altitude about 1,000 _meters, local name Dilisan. 4. PTYCHORAPHIS CAGAYENSIS Becc. comb. nov. Heterospathe cagayensis Bece, in Philip. Journ. Sci. 4 (1909) Bot. 611. Luzon, Cagayan Province, For. Bur. 12286 Klemme, locally known as dumayaca. : An imperfectly known palm, somewhat uncertain in its generic position, but greatly resembling Heterospathe elata in its fruit; XIV, 3 Beccari: The Palms of the Philippine Islands 329 the fruit of the latter, however, has a smooth (not granulose) surface, the pericarp apparently being without sclerosomes. For this reason I have thought proper to transfer it from the genus Heterospathe to Ptychoraphis. The fruit is spherical, almost centrally apiculate, 7 mm in diameter, and has the hilum conspicuously marked all along one side. The spadix is thrice branched, and its fruiting branches are thicker than in Hete- rospathe elata (2.5 to 3 mm in diameter), almost torulose; the scrobiculi are approximate, have a raised rim, and the floral bracts form a shallow cup; the fruiting perianth is very shal- lowly cupular and does not have a pedicelliform appearance. The leaflets have slightly falcate apices with no paleolae under- neath on the midrib. ADONIDIA Beccari genus novum Normanbyae sp. Becc. in Philip. Journ. Sci. 4 (1909) Bot. 606. After a more complete revision of all the palms belonging to the Ptychospermeae I now think it advisable to establish an autonomous genus for the Philippine palm which was formerly considered by me as a member of the Australian genus Nor- manbya, in consideration of the fact that, notwithstanding the great similarity in the flowers and fruit, Adonidia differs greatly from Normanbya Muellerii in the leaves, the vegetative charac- teristics being of great importance in the delimitation of the genera of the subtribe of Ptychospermeae. In Adonidia the leaflets are regularly set, narrow, elongate, acuminate, bifid at the apex, 1-costulate, and with a strong nerve at both margins; whereas in Normanbya the leaflets are very peculiarly arranged in groups, each group resulting, apparently, from the spliting of one leaflet into several, which yet remain united by their bases; furthermore, the resulting leaflets are narrow, have a truncate praemorse apex, and have no marginal nerves. The separation of Adonidia from Normanbya gives us a genus peculiar to the Philippines and the only endemic one in the family. 1. ADONIDIA MERRILLII Becc. comb. nov. Normanbya Merrillii Becc. in Philip. Journ. Sci. 4 (1909) Bot. 606, t. 30, $1. PALAWAN, in sandy soil of the woods fronting the beach south of Brooke’s Point, Elmer 12708, February, 1911, locally known as oring-oring; Apulit Island, Taytay Bay, on the coast of Palawan, growing abundantly, often gregariously, on steep limestone slopes, altitude 10 to 100 meters, Merrill 9415, May, 1918. The species is commonly cultivated in Manila for ornamental purposes. te 330 The Philippine Journal of Science 1919 ONCOSPERMA Blume Conspectus of the species. a’. Gregarious. Floriferous branches of the spadix numerous, long, slender, and inserted at different levels on the rachis. Male flowers with 6 stamens. Fruit small, spherical, 11 to 12 mm in diameter. 1. O. filamentosa. a’. Stem very tall, solitary. Spathe very densely covered with criniform spines. Spadix with numerous, long, floriferous branches, which gradually narrow from a very thick base to a slender apex. Fruit large, spherical, 20 to 22 mm in diameter........................ 2. O. horrida. a‘, Rather large and growing in clumps. Spadix with rather short and thick floriferous branches; the latter clustered together and very closely scrobiculate. Fruit spherical or very slightly longer than broad, 15 to 16 mm in diameter, with the remains of the stigmas placed laterally, about half way or a little above.................. 3. O. platyphylla. a‘. Trunk slender. Spadix with relatively few, clustered, thickish, very closely scrobiculate, floriferous branches. Spathe not densely spinose. Fruit spherical, 14 to 15 mm in diameter, with nearly apical remains of the stigmas. Fruiting perianth 12 mm in diameter... 4. O. gracilipes. 1. ONCOSPERMA FILAMENTOSA BI. Rumphia 2 (1836) 97, t. 82, 103; Bece. in Philip. Journ. Sci. 4 (1909) Bot. 610. PALAWAN, Caranagan River, For. Bur. 3790 Curran; Brooke’s Point, Elmer 12662. Quite identical with the Malayan plant. 2. ONCOSPERMA HORRIDA Scheff. in Nat. Tijdsch. Nederl. Ind. 32 (1871) 189; Bece. in Philip. Journ. Sci. 4 (1909) Bot. 610. MINDANAO, Zamboanga District, San Ramon, Copeland 1626: Agusan Province, Cabadbaran, Elmer 13886: Davao District, Todaya, Mount Apo, Elmer 11876. PoutLLo, Bur. Sci. 9277 Robinson. This is slightly different from the Malayan plant, but it seems to me not sufficiently so to consider it as even a variety. In the specimens I have examined the spines that cover the spathe are criniform or considerably slenderer than in the typical Malayan form. The fruit also has the pericarp slightly thinner, and the remains of the stigmas are less prominent in the dry fruit than in the Malayan plant. 3. ONCOSPERMA PLATYPHYLLA Bece. in Philip. Journ. Sci. 4 (1909) Bot. 609. Necros, Gimagaan River, Whitford 1670, For. Bur. 17847 Curran. 4. ONCOSPERMA GRACILIPES Bece. in Philip. Journ. Sci. 2 (1907) Bot. 228, 4 (1909) Bot. 610. Luzon, Tayabas Province, Atimonan, Merrill 4010; Guinayangan, Bur. Sci. 13213 Ramos; between Paete and Piapi, For. Bur. 10148 Curran. BiuirRAN, Bur. Sct. 18490 McGregor. The type specimen is Merrill 4010, of which it is stated that the trunk was 7 to 10 m high. Curran’s field note indicates that the plant grows in clumps, and that the trunk is 10 m high and 12 cm in diameter. XIV, 3 Beccari: The Palms of the Philippine Islands 831 ORANIA Blume Representatives of Orania are rather common in the forests of the Philippines, especially in the more southern islands of the group, but their specific delimitation is often difficult and un- certain, as individuals apparently belonging to the same species seldom exhibit constant characters, when growing in separate islands, or even in remote localities in the same island. This is not confined to Orania, but is an almost general occurrence in other genera of Philippine palms. After a careful study of the specimens that offer the greatest divergence of characters, especially in the flowers and the fruits, I have concluded that there exist in the Philippines three primary species of Orania; namely, Orania philippinensis, O. rubiginosa, and O. decipiens (O. paraguanensis is perhaps only a derivation from O. philippinensis) ; but numerous variations are presented by these main types, especially in the size and shape of the fruits and in the thickness of their mesocarps, so as to render not inconsistent the opinion that in the Philippines we have only one “synspecies” of Orania, represented by several “micro- species” the latter however derived, we may suppose, by the in- terbreeding of two foreign elements. Orania philippinensis is apparently derived from the Papuan O. regalis, of which varieties or allied species exist also in the Moluccas; whereas the more typical form of O. decipiens is plainly allied to the Malayan O. macrocladus. : Conspectus of the species. a’. Floriferous branches ultimately glabrous. b’. Male flowers angular, lanceolate, 6 to 8 mm long, 3 to 3.5 mm broad, or about twice as long as broad, having the stamens one-third to one-half shorter than the petals; anthers linear-oblong. Female flowers broadly ovate-trigonous, the calyx cupular, very low; petals triangular, subaequilateral. Fruit spherical or very slightly nar- rowed at the base, usually 5.5 to 6 cm in diameter, at times somewhat less; mesocarp about 5 mm thick : 1. O. philippinensis. ce’, Fruit exactly spherical, larger than in the species (6.5 cm in diameter), yet with a thinner mesocarp (3.5 to 4 mm thick). O. philippinensis var. sibuyanensis. b*. Male flowers narrow, linear, 8 mm long, 2 mm broad, or about four times as long as broad, having the stamens nearly as long as the petals and the anthers very narrowly linear. Fruit spherical, smaller than in O. philippinensis, 4.5 to 4.7 cm in diameter, the meso- carp 3.5 to 4 mm thick 2. O. paraguanensis. b*. Male flowers narrowly linear. Fruit slightly narrowing to the base, or slightly pyriform, considerably smaller than in the preceding species, 4 to 4.5 cm long, 85 to 87 mm thick; kernel spherical, extended at the base into a broadly obconical blunt point; mesocarp 332 The Philippine Journal of Science 1919 relatively thick (3 to 4 mm) and furnished with many short and stout woody fibers. Seed about 25mm in diameter.. 3. O. decipiens. ce. Fruit smaller than in the species (837 mm long, 31 to 32 mm thick), but always more or less narrowing to the base; mesocarp also thinner (2.5 mm thick)................ O. decipiens var. mindanaoensis. ¢. Fruit spherical, not narrowing to the base, 42 mm in diameter; mesocarp 4 to 5 mm thick. O. decipiens var. montana. a?. Floriferous branches more or less permanently rusty-tomentose. Male flowers very narrow and long, 1 cm long, 2 to 3 mm broad. Female flowers pyramidate-trigonous, acuminate, twice as long as broad, calyx campanulate. Fruit subpyriform, 40 to 45 mm long, 30 to 38 mm thick; mesocarp thin, 1.5 to 2 mm [Act Sepieoeteroy Romer ert 3 4. O. rubiginosa. 1. ORANIA PHILIPPINENSIS Scheff. ex Becc. in Ann. Jard. Bot. Buit, 2 (1885) 156, t. 14; Becc. in Webbia 1 (1905) 335. Orania Palindan Merr. in Govt. Lab. Publ. 27 (1905) 88, Philip. Journ. Sci. 1 (1906) Suppl. 32. Palindan Blanco Fl. Filip. (1887) 444 (partly ?). | Caryota Palindan Blanco Fl. Filip. ed. 2 (1845) 518 ? Luzon, Bataan Province, Mount Mariveles, For. Bur. 1610 Borden, this number with mature spherical fruits agreeing entirely with Scheffer’s typical specimens: Tayabas Province, For. Bur. 10410 Curran, fruit slightly smaller than in Borden’s specimen and slightly narrowed below, 5.3 cm long, 4.5 em thick: Laguna Province, Calauan, Bur. Sci. 12391 McGregor: Zambales Province, For. Bur. 8413 Curran & Merritt: Cagayan Province, For. Bur. 17068 Curran; the last three specimens have immature fruits. ORANIA PHILIPPINENSIS Scheff. var. SIBUYANENSIS Bece. var. nov. SipuyANn, along Patao River, Mount Giting-Giting, Elmer 12066, local name banga. A tree about 5 m high, the stem 12 cm in diameter. Spadix about 1.50 m long; one spathe, persistent and coriaceous, is 1.40 m long, lanceolate, 5 cm broad, slightly rusty-furfuraceous outside. Fruit of a citron-yellow color. 2. ORANIA PARAGUANENSIS Becc. in Webbia 1 (1905) 335. : PaLAWAN (Paragua), Taytay, Merrill 9269, Merrill Phil. Pl. 1816; Separation Point, Merrill 869 (type specimen). This is the representative in Palawan of O. philippinensis and perhaps should be considered as a variety of that species. It was first described by me from incomplete specimens, the fruits attached on very thick (perhaps abnormal?) branchlets (Mer- rill 869). Other specimens subsequently collected also by Mer- rill in Palawan at Taytay (1316 and 9269) have floriferous branches not differing from those of the typical O. philippi- nensis from Luzon, but have the male flowers much narrower than in that form, linear, 8 mm long, 2 mm thick, with stamens as long as the petals, and very narrowly linear anthers. The fruits are somewhat smaller than in the Luzon plant, 45 to 47 mm in diameter, with the mesocarp 3.5 to 4 mm thick. XIV, 38 Beccari: The Palms of the Philippine Islands 333 3. ORANIA DECIPIENS Bece. in Philip. Journ. Sci. 4 (1909) Bot. 614. Minporo, Bongabong River, at about 70 meters above the sea level, For. Bur. 4120 Merritt, Tagalog and Visayan name banga. ; This is distinguishable by its relatively small fruit, yet having a thick mesocarp. In the size and the shape of the fruit O. de- cipiens could be mistaken for O. macrocladus, and among the Philippine species, for O. rubiginosa, but in these the mesocarp is rather thin, whereas it is relatively very thick in O. decipiens. ORANIA DECIPIENS Becc, var. MINDANAOENSIS Bece. var. nov. MINDANAO, Zamboanga District, Port Banga, For. Bur. 9179 Whitford & Hutchinson. Fruit a little smaller than in the species, 37 mm long, 31 to 32 mm thick, slightly narrowing to the base; mesocarp 2.5 mm thick. ORANIA DECIPIENS Becc. var. MONTANA Bece. var. nov. MINDANAO, Davao District, Mount Apo, at about 1,000 meters’ altitude, Elmer 11881: Agusan Province, Cabadbaran (Mount Urdaneta), at about 830 meters’ altitude, Elmer 13970. The fruit is exactly spherical or else slightly broader than high, 42 mm in diameter; mesocarp 4 to 5 mm thick. In Elmer 13970 the male flowers are very narrowly linear, 7 to 9 mm long, 2 mm broad. 4. ORANIA RUBIGINOSA Becc. sp. nov. Frondium rachi et spadicis parte axilli et ramis ferrugineo- scrophulosis; floribus masculis angustissimis, 1 cm _ longis; floribus foemineis trigono-pyramidatis, acuminatis; fructibus globoso-subpyriformibus, quam in O. philippinensi minoribus, mesocarpio tenui (1.5 ad 2 mm spisso). Apparently a smaller plant than Orania philippinensis, 6 to 8 m high, the stem 10 to 15 cm in diameter, the leaves also smaller and with fewer and narrower leaflets, the latter about 30 on each side of the rachis. Base of the leaves and rachis covered, especially underneath, with a rusty, very appressed indumen- tum, evanescent in age. The leaflets have on each side of the midrib 2 or 3 very strong secondary nerves, especially prominent on the lower surface near the base; the intermediate segments are 60 to 70 cm long and 3.5 to 4 cm broad, very obliquely truncate, and erose-toothed at the apex, and prolonged along the upper margin into a somewhat elongated point; in the upper leaflets the apex is truncate, irregularly lobulate and erose- toothed; the lower surface is as usual whitish and is, further- more, sprinkled, especially in newly expanded leaves, with minute rusty scales (microlepidia), and has the upper margin marked by 164391——-6 834 The Philippine Journal of Science 1919 a rusty furfuraceous band. The spadix has more or less per- manently rusty-furfuraceous, 10 to 20 cm long or less, floriferous branches, bearing on their lower parts a few female flowers, each of which is accompanied, as usual, on each side, by a male flower; the upper parts of the branches present male flowers only. The male flowers are very narrow, 1 cm long and 2 to 3 mm broad, sinuous, acuminate; calyx very shortly campanu- late, sharply 3-toothed; petals about five times longer than the calyx, narrowly lanceolate, with the tips acuminate and more or less faleate; stamens 6, slightly shorter than the petals; fila- ments short; anthers linear, sinuous, emarginate at the apex. Female flowers pyramidal-trigonous, acuminate, 6 to 8 mm long; calyx relatively large, shallowly cupular, with three, broad, acuminate teeth, but latter splitting into three broad lobes; corolla about two and one-half times as long as the calyx; petals elongate-triangular, acuminate, 6 to 7 mm long, 3 mm broad at their bases; staminodia 6, subulate, not bearing rudi- mentary anthers. Fruiting perianth measuring about 9 mm be- tween the tips of the petals. Fruit globose, slightly diminishing toward the base, or subpyriform, 4 to 4.5 cm long, 3 to 3.8 cm broad; mesocarp relatively thin, 1.5 to 2 mm thick; kernel, when detached from the mesocarp, with a blunt, not clearly defined prominence at the base; walls of the endocarp 0.33 mm thick at the sides. Seed globular with a flattish base, the vascular branches of the integument few, ascending, sinuous, not or very slightly divided. Embryo very slightly and excentrically apical. Luzon, Cagayan Province, on the east coast, Bur. Sci. 10575 McGregor, For. Bur. 17259, 17192 Curran: Isabela Province, Biocbian Bay, Bur. Sci. 10658 McGregor: Tayabas Province, Tagcauayan, Bur. Sci. 13390 Ramos. This is easily distinguishable from Orania philippinensis by the general rusty coating which covers the leaf-rachis and the axial parts of the spadix; by its long and narrow male flowers, three to four times longer than broad; by the trigonous, acu- minate female flowers having a campanulate calyx, and the. petals elongate-triangular, twice as long as broad, at least; and by the subpyriform fruit, considerably smaller than that of O. philippinensis, and with a thin mesocarp. Apparently also referable to O. rubiginosa is Loher 7090 collected at Angilog, Rizal Province, Luzon (Herb. Kew.), of which the collector notes that it is a tree 3 to 4 m high with a thick stem. In this specimen the leaflets are larger than in the type, and the fruits also are somewhat larger, 5 cm long and 4 cm thick, and have the mesocarp 2.5 mm thick. XIV, 3 Beccari: The Palms of the Philippine Islands 885 ARENGA Labillardiére Conspectus of the species. a’. Leaflets elongate, narrow, having smooth or remotely and minutely toothed margins, the secondary nerves parallel, all starting from the base. b’. Large, with solitary stems. Fruit more or less turbinate, 3 cm or more in diameter. Male flowers with very numerous stamens; BHthENS BYISUALEAiiick. oi Si hh client 1. A. saccharifera. b*. Relatively small and caespitose. Male flowers with 20 to 30 stamens; anthers blunt or slightly apiculate. Fruit spherical, 15 to 18 mm in very boldly toothed, or lobed; secondary nerves divergent from the rachis at different levels. b*. Caespitose; stem attaining 2 to 3 m in height and 15 cm in diameter. Male flowers with a rounded top. Stamens about 150. Fruit longer than broad, rounded at both ends.. 3. A. tremula. b?. Stem short and thick, about 30 cm in diameter. Male flowers apiculate. Stamens about 100. Fruit spherical........ 4. A. Ambong. 1. ARENGA SACCHARIFERA Labill. in Mém. Inst. Fr. 4 (1801) 209; Griff. Palms Brit. Ind. (1845) 164, t. 135A; Becc. in Perkins Fragm. Fl. Philip. (1904) 48, Philip. Journ. Sci. 6 (1911) Bot. 229. Saguerus saccharifer Wurmb in Verh. Bat. Genootsch. 1 (1779) 350; Blume Rumphia 2 (1836) 128, t. 123-4. Luzon, Tayabas Province, Lucban, Elmer 7644; Guinayangan, Merrill 2048. Pouitto, Bur. Sci. 9041 Robinson, Bur. Sci. 10468 McGregor. BILIRAN, Bur. Sci. 18941 McGregor. 2. ARENGA MINDORENSIS Becc. in Perkins Fragm. Fl. Philip. (1904) 48, Webbia 1 (1905) 49. Mrnporo, Calapan, Merrill 1790 (type), For. Bur. 8768 Merritt, locally known as dumayaca. LuzON, Tayabas Province, Atimonan, For. Bur, 10203 Curran; Mauban, Bur. Sci. 19468 Ramos, with the female flowers well developed, the staminate ones not yet open: Bataan Province, For. Bur. 5469 Curran: Laguna Province, Calauan, Bur. Sect. 12395 McGregor. MINDANAO, Davao District, Mount Apo, Elmer 11192, a more robust plant than the Mindoro and Luzon form but otherwise presenting no peculiar characters. 3. ARENGA TREMULA Becc. in Philip. Journ. Sci. 4 (1909) Bot. 612. Caryota tremula Blanco Fl. Filip. (1837) 744; Kunth Enum. ig ae (1841) 549. Didymosperma tremulum Wendl. et Dr. in Kerch. Les Palms (1878) 243. Luzon, Tayabas Province, For. Bur. 10213, 10280 Curran, very common, Bur. Sci. 1737 McGregor. Patawan, Victoria Peak, Bur. Sci. 735 Fox- worthy; near Irauan (altitude, 200 meters) For. Bur. 3542 Curran, Brooke’s Point, Elmer 12596. BatasBac, Merrill 5372. 4. ARENGA AMBONG Bece. in Philip. Journ. Sci. 2 (1907) Bot. 229. Wallichia oblongifolia (non Griff.) Bece. in Webbia 1 (1905) 328. 836 The Philippine Journal of Science 1919 MINDANAO, Zamboanga District, Garcia 658, Copeland 1681, Merrill 8367. Moro name ambong. CARYOTA Linnaeus The genus Caryota is represented in the Philippines by few species, which however seem rather variable, and which, so far as I can determine from dry, very often incomplete, herbarium specimens, present, at times, forms with ambiguous or uncertain characters and are therefore difficult to identify. The doubt is not altogether to be excluded that occasionally some of the ambiguous forms may be of hybrid origin. Two principal species are frequently met with throughout the Philip- pines, one of which, a large and fine palm, is to be considered, I think, not specifically distinct from the very widely spread and very variable Caryota Rumphiana; the other, considerably smaller, is an endemic species, C. Cumingii, which at times pre- sents forms not easily distinguishable, at least in the herbarium, from C. mitis Lour., a plant common in Cochinchina and in the Malayan region. Specimens of Caryota mitis, agreeing in every respect with the Cochinchina plant, have been collected in Pa- lawan; C. Cumingii, however, is always solitary, not suboliferous. Caryota Merrillii and C. majestica are species of which I have only an incomplete knowledge; both are apparently re- lated to C. Cumingii or perhaps are varieties of that, and prop- erly to be understood they require the examination of more complete material than I have had at my disposal or, better still, an examination of the living plants in their native country. Caryota Rumphiana var. oxyodonta Becc., collected only once and established on a single incomplete specimen, is another form of which more complete material is desired. Conspectus of the species. a’. Large trees. Stem tall, solitary. Fruit 1- or 2-seeded. Fruiting peri- anth 10 to 11 mm in diameter. Male flowers large, 15 to 17 mm long, . with numerous stamens. b*. Leaflets of the full-grown plant long and narrow, having the upper margin at times very obsoletely, yet at times rather sharply, and very unequally toothed, and the lower margin much produced into a taillike point. Male flowers with 40 to 60 stamens. Stem up to 30 to 40 cm in diameter............ 1. C. Rumphiana var. philippinensis. b*. Leaflets having the upper margin deeply and acutely toothed, the teeth long, narrow, acuminate, and very close together. Male flowers with 27 to 30 stamens. A smaller plant than var. philippinensis. C. Rumphiana var. oxyodonta. a’. Of medium size. Fruit always 1-seeded. Male flowers (where known) small and with few stamens. b*. Stem solitary. Male flowers with 6 to 9 stamens only. XIV, 3 Beccari: The Palms of the Philippine Islands 337 ct. Trunk 5 to 8 m high, 10 to 20 cm in diameter. Leaflets erect- spreading, dimidiate-rhomboidal. Male flowers 6 to 7 mm long, with 9 stamens. Fruiting perianth 6 to 7 mm in diameter. Fruit 12 to 17 mm in diameter, spherical. Seed with a chestnut- brown polished surface. Branches of the spadix strongly hairy- scurfy 2. C. Cumingii. ce. Leaflets very spreading or ‘horizontal, frequently opposite, very sharply toothed. Male flowers * * *. Fruit spherical, 12 mm in diameter. Seed nearly spherical, 8 to 9 mm in diameter, with a black, even, and polished surface 3. C. Merrillit. c*. Leaflets ascending, very narrow and very deeply and sharply toothed. Male flowers having 6 stamens only. Fruit 11 to 12 mm in dia- meter. Seed slightly broader than high, 9 mm broad, of a shiny chestnut-brown color, the surface slightly grooved. 4. C. majestica. b'. Soboliferous or with stems in clusters, about 4 m high and 10 cm in diameter. Male flowers 8 to 12 mm long, with 12 to 16 stamens. Fruit 15 to 16 mm in diameter, frequently broader than high. Fruiting perianth 8.5 to 9 mm in diameter-...............-.-....- 5. C. mitis. 1. CARYOTA RUMPHIANA Mart. var. PHILIPPINENSIS Becc. Caryota Rumphiana (vix Mart.) Becc. in Perkins Fragm. FI]. Philip. 1 (1904) 48, Webbia 1 (1905) 331. Caryota wrens (non Linn.) Blanco FI. Filip. (1837) 740 (pro parte). Luzon, Tayabas Province, For. Bur. 7850 Curran & Merritt; Lucban, Elmer 9801: Laguna Province, Santa Maria Mavitac, For. Bur. 10064 Curran; Mount Maquiling, For. Bur. 19961 Villamil: Manila, Merrill 8037 (cultivated). Mrnporo, Bongabong, Whitford 1878. MInpANAo, Agusan Province, Mount Urdaneta, Elmer 13623, Manobo name pogahan: Davao District, Todaya (Mount Apo), Elmer 10940, Bagobo name pola. CARYOTA RUMPHIANA Mart. var. OXYODONTA Bece. Luzon, Laguna Province, Santa Maria Mavitac, For. Bur. 10045 Curran, The specimen of this variety seen by me has dimidiate-rhom- boidal leaflets, much more acutely toothed than in the species, the teeth being very close together, very narrow, and very acu- minate, and on the average 10 mm long; the lower margin of the leaflets is prolonged into an acuminate, not very long point; the largest leaflets on the sides of the ultimate divisions of the rachis are 18 to 20 cm long and 4 to 5 em broad; the basal divisions are broader and less acuminate. The floriferous branches are very similar to those of var. philippinensis and of the same dimensions. The male flower-buds are globose, 7 mm in diameter with the corolla not yet produced outside the — calyx, but even in that condition they show that they would have acquired the dimensions usual to var. philippinensis. The stamens are numerous. According to the collector the plant had a trunk 7 m high and 10 cm in diameter, and leaves 2 to 8 m long. 338 The Philippine Journal of Science 1919 2. CARYOTA CUMINGI! Lodd, ex Mart. Hist. Nat. Palm. 2:195 (2d ed.); Becc. in Perk. Fragm. Fl. Philip. (1904), Webbia 1 (1905) 331. Curtis’s Bot. Mag. III 25 (1869) ¢. 5762; Farnsworth Philip. Bur. Ed. Bull. 54 (1915) 59, with plates. Caryota urens (non Linn.) Blanco (partly) Fl. Filip. (1837) 740, ed. 2 (1845) 510, ed. 3, 3 (1879) 141 t. 349. Caryota sp. Cuming exsicc. No. 1915; Vidal Phan. Cuming (1885) 153. LuzON, Manila, Gaudichaud in Herb. Delessert, Merrill 8036, cultivated: Tayabas Province, For. Bur. 10348 Curran: Union Province, Bauang, Elmer 5647. MINvDoRO, For. Bur. 6216 Merritt, local name apican. PANay, Miagao, Vidal 4067 in Herb. Bece. Guimaras, For. Bur. 119 Gamill, local name patisan. MINDANAO, Zamboanga District, Banga, For. Bur. 9077 Whitford & Hutchinson. PALAWAN, Bur. Sci. 873 Foxworthy; Taytay, Merrill 9360. The Mindanao and Palawan specimens do not seem to be such typical representatives of Caryota Cumingii as those from Luzon. 8. CARYOTA MERRILLI! Becc. in Webbia 1 (1905) 383. Luzon, Pangasinan Province, Bautista, Merrill 2880. This is known only from the type collection, a rather imper- fect specimen; it may perhaps be but a variety of C. Cumingii. 4. CARYOTA MAJESTICA Lind. Illustr. Hort. 28 (1881) 16 (name only) ; Bece. in Perk. Fragm. Fl. Philip. (1904) 48, Webbia 1 (1905) 334. Caryota wrens (non Linn.) foliis angustioribus Blanco FI. Filip. (1837) 740. Luzon, Rizal Province, Bosoboso, Merrill 1892, Merrill 8490, local name anibung. Perhaps Vidal 3949 (Herb. Becc.) from Tarlac (sterile specimen) is also referable to this species. 5. CARYOTA MITIS Lour. Fl. Cochinch. (1790) 569. PALAWAN, For. Bur. 3888, 4148 Curran. These specimens are indistinguishable from typical specimens of C. mitis collected in Cochinchina; the male flowers are larger than in C. Cumingii and with 12 to 18 stamens. I consider also Elmer 12606 collected at Brooke’s Point, Palawan, with the local name bato as referable to C. mitis. Of this it is noted that the stems are usually 3 to 5 in a cluster, which is one of the essential characteristics of C. mitis. NIPA Rumphius NIPA FRUCTICANS Wurmb in Verh. Bat. Genootsch. 1 (1779) 349. LuzoN, Cagayan Province, For. Bur. 17265 Curran. BiuiraAn, Bur. Sci. 18700 McGregor. Pou, Bur. Sci. 9061 Robinson, Bur. Sci. 10462 McGregor. PALAWAN, Bur. Sci. 278 Bermejos. Along tidal streams throughout the Philippines, in places gre-_ garious over very large areas, forming the so-called “nipales” or nipa swamps. XIV, 3 Beccari: The Palms of the Philippine Islands 839 Cocos Linnaeus cocos NUCIFERA Linn. Sp. Pl. (1753) 1188. Cultivated throughout the Philippines both in the coastal region and in the interior in those regions where there is no prolonged dry season. PHOENIX Linnaeus PHOENIX HANCEANA Naud. var. PHILIPPINENSIS Becc. in Philip. Journ. Sci. 3 (1908) Bot. 339. BaTAN ISLANDS, Sabtan, Bur. Sci. $744 Fénix. The collector’s note gives the diameter of the trunk of this variety as 45 cm, which he. afterwards explained was an error for its circumference. LICUALA Rumphius LICUALA SPINOSA Wurmb in Verh. Bat. Genootsch. 2 (1780) 469. PALAWAN, Merrill 1468, Bur. Sci. 836 Foxworthy, Bur. Sci. 269 Bermejos. BALABAC, Bur. Sci. 448 Mangubat. CULION, Merrill 543. In thickets at low altitudes, in some regions growing in mud immediately back of the mangrove swamps. LIVISTONA R. Brown Conspectus of the species. a'. Leaves irregularly parted into primary 2- to 6-costulate segments; secondary segments 1-costulate, very deeply parted into two very long flaccid laciniae. Petiole armed, especially in its lower portion, with very robust spines. Flowers sessile and in small groups on the branchlets. Fruit globose or very slightly reniform, bluish even when dry, 11 to 15 mm in diameter 1. L. cochinchinensis. a?. Leaves entire in their central part, and with the periphery more or less deeply divided into always unicostulate segments. Flowers soli- tary, spirally inserted around the branchlets. b'. Flowers relatively large, 4 to 4.5 mm long. Leaves of adult plants having unarmed or, at times, slightly spinose petioles. The dry mature fruit spherical, 22 to 23 mm in diameter, witha very dark brown polished surface. The young fruits are slightly oblong and narrow a little toward the base......... . 2. L. Merrillii. b?, Flowers very small, at most 2 mm in diameter. ce. Petioles of the adult plant spinose in their basal part, unarmed elsewhere. Spadix composed of three main inflorescences, free from their bases and all issuing from a common flattened spathe; upper spathes very tightly sheathing throughout, truncate at the mouth, and, as are all the other parts of the spadix, reddish-brown when dry. Fruit spherical even when young, dark-violaceous when fresh, quite black when dry.........--------------- 8. L. rotundifolia. Forma typica (not yet found growing in the Philippines) is especially characterized by the seed having the intrusion of the raphe penetrating only two-thirds of the albumen. 340 The Philippine Journal of Science 1919 d@’. Fruit 2 cm in diameter. Seed traversed completely from base to apex by the intrusion of the raphe. Leaves of very young plants having the petioles armed, in their bagal part, with conspicuous spines, as much as 15 to 20 mm in length, the leaves of adult plants with the central segments shortly bifid. L. rotundifolia Mart. var. luzonensis. @. Fruit 12 to 15 mm in diameter, the kernel alone 10 to 13 mm in diameter. Seed 8 to 10 mm in diameter, more or less traversed by the intrusion of the raphe. Central segments of the adult leaves shortly bifid at their apices. L. rotundifolia var. microcarpa. d*. Fruit of medium size. Central segments of the adult leaves parted into two 15 to 20 em long laciniae. L. rotundifolia var. mindorensis. c’. Petioles of leaves in the adult plant unarmed, at least in their upper part; in young plants armed with very small spines. The mature fruit yellowish orange when fresh, yellowish brown when dry. Spathes straw-colored, slashed at the mouth. 4. L. Robinsoniana. 1. LIVISTONA COCHINCHINENSIS Mart. Hist. Nat. Palm. 3 (1838) 319; Hook. f. Fl. Brit. Ind. 6 (1892) 434; Bece. Malesia 3 (1886) 69. Livistona spectabilis Griff. in Caleutta Journ. Nat. Hist. 5 (1845) 336, Palms Brit. Ind. (1845) 130, t. 266, C. Livistona inaequisecta Bece. in Philip. Journ. Sci. 4 (1909) Bot. 616. Luzon, Cagayan Province, For. Bur. 17145, 17802, 17156 Curran, local name tarao, For. Bur. 14762 Darling: Laguna Province, Santa Maria Mavitac, For. Bur. 10079 Curran; Cavinti, Loher 7058, 2. LIVISTONA MERRILLII Becc. in Perk. Fragm. Fl. Philip. (1904) 45, Webbia 1 (1905) 339. Livistona Whitfordii Becc. in Webbia op. cit. 341, Philip. Journ. Sci. 4 (1909) Bot. 615. Luzon, Cagayan Province, Mount Narig, For. Bur. 17811 Curran, com- mon at 200 meters’ altitude, forming small groves in Shorea forests, trunk 20 m by 60 cm; For. Bur. 17278 Curran, on rocky coral limestone hills near the seashore, altitude 60 meters, locally known as ballang, the leaves used for thatching houses and for raincoats, the stems for bows and for flooring: Tayabas Province, Guinayangan, Merrill 2071 (type), For. Bur. 10189 Curran; Atimonan, Whitford 731 (type of L. Whitfordii Becc.): Zambales Province, For. Bur. 5876 Curran, locally known as telsis. 38. LIVISTONA ROTUNDIFOLIA Mart var. LUZONENSIS Becc. Livistona rotundifolia (vix Mart.) Becc. in Leafl. Philip. Bot. 2 (1909) 647, Luzon, Zambales Province, For. Bur. 5834 Curran: Albay Province, For. Bur. 15322 Aguilar, local name anahao or bulus: Tayabas Province, Lucban, Elmer 9293 (in flower): Union Province, Castillo (specimen with mature fruits), Loher 7070 in Herb. Kew.: Benguet Subprovince, Sablan, Bur. Sci. 12687 Fénix (leaves from a young plant), Merrill 37992 *For the uses of the anahao palm see Philip. Bur. Ed. Bull, 54 (1917) 68, 69, plates. e XIV, 3 Beccari: The Palms of the Philippine Islands 841 LIVISTONA ROTUNDIFOLIA Mart. var. MICROCARPA Becc. Livistona microcarpa Becc. in Philip. Journ. Sci. 2 (1907) Bot. 231. PALAWAN, Caranugan River, For. Bur. 3784 Curran, in river swamps. MINDANAO, Davao District, a few miles west of Digas at Todaya (Mount Apo), Elmer 11967, local name balla. LIVISTONA ROTUNDIFOLIA Mart. var. MINDORENSIS Bece. Livistona mindorensis Becc. in Philip. Journ. Sci. 4 (1909) Bot. 615. Minporo, Bongabong River, For. Bur. 4108 Merritt, very common, Tagalog name panobao. 4. LIVISTONA ROBINSONIANA Becc. in Philip. Journ. Sci. 6 (1911) Bot. 230. PoLILLo, Bur. Sci. 9265 Robinson, Bur. Sci 10471 McGregor, Tagalog pilig. Luzon, Laguna Province, Cavinti, Loher 7056, in Herb. Kew. CORYPHA Linnaeus CORYPHA ELATA Roxb. FI. Ind. ed. 2, 2 (1882) 176; Griff. Palms Brit. Ind. (1845) 112, t. 220 D. Corypha Gebanga Bl. Rumphia 2 (1836) 59, t. 97, 98 et 105. Livistona Vidalii Becc. in Webbia 1 (1905) 343. Corypha umbraculifera (non Linn.) Vidal Sinopsis Atlas (1883) 91, t. 98; F.-Vill. Novis. App. (1883) 281, incl. var. cubang and sylvestris. Luzon, Pangasinan Province (cultivated), For. Bur. 8410 Curran & Merritt; Union Province, Naguilian, Bur. Sci. 13000 Fénix: Tayabas Prov- ince, Lucban, Elmer 9294: Pampanga Province, Arayat, Garcia 63 (leaf from a very young plant, the type of Livistona Vidalii Becc.). SIBUYAN, Mount Giting-Giting, Elmer 12567. MINDORO, Bongabong River, For. Bur. 4121 Merritt. MINDANAO, Elmer 11965, with the local name serrar. This species is the buri palm. Leaf strips from it are used in the Philippines for making different sorts of hats * and other valuable commercial materials.* The specimens from Mindoro have floriferous branchlets larger than usual, as much as 40 cm in length. ' Corypha elata Roxb. is distinguishable from the allied species by the trunk, which in the full-grown plant is very high and relatively slender and is marked all around by a slightly de- pressed spiral trace of the insertion of the fallen leaves; it is also characterized by its pyramidate inflorescence representing from one-fourth to one-fifth of the entire plant, and by its globular fruit, 20 to 23 mm in diameter, born on pedicels 3 to 5 mm long. Specimens of Corypha collected on Biliran Island by Mc- Gregor in June, 1914, Bur. Sci. 18720, have a secondary branch * See Robinson, C. B. Philippine hats in Philip. Journ. Sci, 6 (1911) Bot. 106, 113, t. 7, 8. : “See Farnsworth, C. G., Philip. Bur. Ed. Bull. 54 (1915) 62, 63, plates. 342 The Philippine Journal of Science 1919 of the inflorescence divided into several very short (3 to 5 cm long) fructiferous branchlets, and the fruits are smaller than usual (16 to 18°mm in diameter, the seed 12 mm in diameter). I do not know if this specimen is to be considered an abnormal one, or if it represents a constant characteristic; in the latter case it ought to be considered as a species distinct from Corypha elata, METROXYLON Rottboell METROXYLON RUMPHII Mart. Hist. Nat. Palm. 3 (1838) 214, 318, t. 102, 159; Beee. in Ann. Bot. Gard. Calcutta 127: t. 105. MINDANAO, Davao District, Todaya, Elmer 11160, local name lumbia: Surigao Province, Catel, Merrill 5448. ZALACCA Reinwardt ZALACCA CLEMENSIANA Bece. in Philip. Journ. Sci. 4 (1909) Bot. 618, Ann. Bot. Gard. Calcutta 127: t. 49. MINDANAO, Lanao District, Camp Keithley, Mrs. Clemens 1109; Davao District, Todaya, Elmer 11879, local name lacaubi. PLECTOCOMIA Martius et Blume ‘PLECTOCOMIA ELMERII Becc. in Ann. Bot. Gard. Calcutta 127: t. 22. MINDANAO, Davao District, Todaya, Elmer 11887. KORTHALSIA Blume Conspectus of the species. a’. Spikes amentiform with very closely crowded flowers and appressed spathels. b’. Leaf-sheaths produced at the base of the petioles into an inflated elongate-elliptic ocrea; leaflets more or less nearly white under- neath 1. K. scaphigeroides. b’. Leaf-sheaths produced at the bases of the petioles into a closely sheathing, densely spinous ocrea. ce’. Slender; leaflets of the upper part of the fertile plant small, rhomboidal, green on both surfaces or slightly paler beneath than above 2. K. Merrilli. ce’. Robust; leaflets large, cuneate-rhomboidal or trapezoidal, sharply double-toothed, paler beneath than above................ 3. K. laciniosa. a’. Spikes of squarrose appearance, the spathes scarious and not appressed; leaves furnished with an elongate, cornet-shaped ocrea which is truncate at the apex; leaflets cuneately rhomboidal, white under- neath = 4. K. squarrosa. 1. KORTHALSIA SCAPHIGEROIDES Becc. in Philip. Journ. Sci. 4 (1909) Bot. 619, Ann. Bot. Gard. Calcutta 127: t. 67. MINDANAO, Zamboanga District, For. Bur. 4816 Hutchinson: Butuan Sub- province, Agusan River, Merrill 7313. BASILAN, For. Bur. 6106 Hutchinson. 2. KORTHALSIA MERRILLII Becc. in Ann. Bot. Gard. Calcutta 12°: tu Bhs PALAWAN, Malampaya Bay, Merrill 9410. XIV, 3 Beccari: The Palms of the Philippine Islands 343 3. KORTHALSIA LACINIOSA Mart. Hist. Nat. Palm. 3 (1838) 211, 348. Korthalsia sp., Vidal No. 4066; Becc. in Philip. Journ. Sci. 4 (1909) Bot. 621. Luzon, Sorsogon Province, Mount Bulusan, Elmer 16672, July, 1916, in flower. Vidal’s sterile specimen (4066), also from Sorsogon, is doubtless this species, to which is also referable another sterile specimen collected at Laguimanoc, Tayabas Province, Luzon, by Klemme, November, 1904. Local name daanan (which is also the Malay name for several species of Korthalsia). Possibly to Korthalsia laciniosa Mart. also belongs a sterile specimen collected in Polillo, Bur. Sci. 10464 McGregor. 4. KORTHALSIA SQUARROSA Bece. in Philip. Journ. Sci. 4 (1909) Bot. 620, Ann. Bot. Gard. Calcutta 12°: t. 108. PALAWAN, near Iwahig, For. Bur. 4185 Curran. BALABAC, Merrill 5384. CALAMUS Linnaeus Conspectus of the species. a’. Leaves noncirriferous (the rachis not prolonged into a filiform, clawed or aculeate appendix). b'. Female flowers and fruits sessile or nearly so; that is, not furnished with a distinct pedicel derived from the lengthened involucrophore. c'. Leaflets almost equally green on both surfaces. @. Leaflets narrow, linear or linear-lanceolate, 1- to 3-costulate. e’. Spadices shortly flagelliferous, about as long as the leaves; fruits small, ovoid; seeds with equable albumen. f'. Leaf-sheaths armed with slender straight spines; primary spathes also spinulous; leaflets very numerous; spathels of the female spikelets very short, bracteiform. A very variable plant, of which it is difficult to establish well- defined varieties, as one. merges into the other by inter- mediate forms 1. C. mollis. . g'. Sheathed stem usually 15 to 20 mm in diameter; leaf- sheaths more or less densely spinous; leaves 50 to 80 cm long C. mollis (forma typica). g’. Robust; sheathed stem 2.5 to 3 cm in diameter; leaves up to 1.2 m long. C. mollis var. major. g’. Slender; sheathed stem 12 to 15 mm in diameter; leaf- sheaths almost spineless.......... C. mollis var. palawanensis. f’. Leaf-sheaths and _ spathes unarmed; leaflets numerous; spathels of the female spikelets very short, bracteiform. 2. C. meyenianus. f. Very slender; leaflets very few and very inequidistant; spathels of the female spikelets shortly infundibuliform. 3. C. Blancoi. e. Spadices (male and female) extremely long, and flagelliform, considerably longer than the leaves. f'. Leaflets sparingly spinulous on three nerves above, the midrib alone minutely hairy-spinulous underneath; female spadix with thickish spikelets drawn together around the main axis; fruit nearly spherical (138 to 14 by 10 mm), with a broad, blunt, black beak; seed pitted-ruminate. 4. C. melanorhynchus. ° 344 The Philippine Journal of Science 1919 f. Leaflets having three slightly bristly nerves on the upper surface and covered throughout on the lower surface with numerous fulvous bristles; female spadix with slender, very spreading spikelets; fruit small, globose-ovoid (11 to 12.5 by 7 mm), with a narrow beak; seed pitted-ruminate. 5. C. filispadiz. f*. Leaflets with two bristly lateral nerves and the midrib smooth on the upper surface, the midrib bristly and the lateral nerves smooth underneath; spadices very loosely branched; male spikelets short, comblike; fruit spherical, 15 to 16 mm in diameter; seed very deeply ruminate through- out 6. C. Diepenhorstui var. exulans. d. Leaflets lanceolate, 5-costulate; very robust; leaflets large, equi- distant; fruit large, ellipsoid, 8.5 cm long, 2.3 mm thick; seed quadrangular 7. C. ornatus var. philippinensis. ec’. Leaflets conspicuously discolorous, green above, white underneath; leaf-sheaths flagelliferous; spadix flagelliferous at its apex; primary spathes much lacerated in their upper part. d'. Leaflets bristly on three nerves above and on the midrib alone beneath : 8. C. discolor. d*, Leaflets without bristles or nearly so on the upper surface, densely sprinkled with numerous subspiny bristles beneath. C. discolor var. negrosensis. b?. Female flowers supported by a distinct pedicel derived from the elonga- tion of the involucrophore; leaves of the upper part of the plant having the apices with gradually diminishing, pluricostulate leaflets, and the rachis clawed and subcirriferous. ce’. Leaflets distinctly grouped, broadly oblanceolate and suddenly apic- ulate, slightly paler below than above; spikelets branched; fruit pisiform 9. C. Cumingianus. c. Leaflets not grouped, lanceolate, gradually acuminate, more or less covered underneath with a very thin, adherent, ochraceous coating; spikelets simple, elongate; fruit pisiform........ 10. C. simphysipus. . a*, Leaves having the rachis prolonged into a clawed cirrus. b*. Male and female spadices having the spikelets provided with a very , distinct pedicellar part which is inserted at the bottom of the spathes. ce’. Very robust; leaf-sheaths covered with slender spiculae, the latter individually distinct or more or less confluent by their broadened bases; leaflets more or less furnished with long bristles, especially on the midrib underneath; secondary spathes coriaceous, entire; fruit spherical, 10 to 12 mm in diameter; secondary spathes smooth 11. C. Merrillii (forma typica). d@’. Secondary spathes prickly................ C. Merrillii var. Merrittianus. d*. Secondary spathes smooth or nearly so; a smaller plant and with slenderer spikelets than in the species, the fruit also smaller (9 mm in diameter) ..............C. Merrillii var. Nanga. ce’. Moderately large; leaf-sheaths very densely covered with blackish uniform bristles; leaflets with long bristles on three costae on both surfaces; fruit spherical, 10 to 12 mm in diameter. ; 12. C. Foxworthyi. XIV, 8 Beccari: The Palms of the Philippine Islands 345 b?. Male and female spadices having sessile spikelets inserted at or near the mouths of their respective spathes. - ¢, Primary spathes elongate and closely sheathing. d'. Fruit containing three seeds. e'. Robust; leaflets large, subequidistant, lanceolate, long-acum- inate, plicate-pluricostulate; spikelets thickish, as much as 15 to 16 cm long; fruit spherical, 14 to 17 mm in diameter. 18. C. manillensis. e’. A smaller plant; leaflets broadly lanceolate, shortly acuminate, subequidistant in the full-grown plant, in pairs on each side of the rachis in young plants; fruit obovoid, conspicuously beaked 14. C. trispermus. da’. Fruit 1-seeded. e’. Leaflets equidistant or nearly so. f'. Leaflets broadly lanceolate, pluricostulate. g'. Two female flowers at every spathel with a neuter one in- terposed between the two. ht, Female spadix very dense and with short branches; spikelets short and with few flowers; female flowers relatively large, 6 mm long; immature fruits fusiform; fruiting perianth campanulate ...............- 15. C. Arugda. }2. Female spadix very diffusely branched; spikelets elongate and with numerous flowers; fruit globose-ovoid; fruit- ing perianth shortly pedicelliform............ 16. C. vinosus. g’. One female flower only at each spathel, with a neuter flower at its side. ht. Fruit small, pisiform; seed pitted, the albumen equable or nearly so. i. Leaflets narrowly elliptic-lanceolate, equally narrowed at both ends, more or less spinulose on some nerves above, smooth underneath; fruit 8 to 9 mm in dia- meter, having squarrose scales in twelve longitud- inal series 17. C. Moseleyanus. ?. Leaflets lanceolate, acuminate, more or less spinulous on some nerves above, smooth underneath; fruit 6.5 mm in diameter, having appressed scales in 18 to 20 longitudinal series -...............-.-- 18. C. mindorensis. #. Leaflets lanceolate, very long-acuminate without bris- tles or spines on either surface; fruit globose-ovoid, 6 mm in diameter, shortly conical-ovoid, and having squarrose scales; leaf-sheaths quite unarmed. 19. C. multinervis. h?. Fruit rather large; seed with a deeply ruminated albumen. #, Leaflets large, broadly lanceolate, 40 to 42 cm long, 4 to 4.5 em wide, pluricostulate and with the nerves smooth on both surfaces; fruit spherical, 2 cm in diameter 20. C. grandifolius. 7. Leaflets 5-costulate, elliptic-lanceolate, 22 to 25 cm long, 30 to 32 mm wide, with the nerves smooth on both surfaces; fruit ovoid-ellipsoid, conspicuously beaked, 25 mm long, 18 mm thick. 21. C. Jenningsianus. 346 The Philippine Journal of Science 1919 f?. Leaflets very narrowly lanceolate, 3-costulate; leaflets more or less bristly-spinulous on three nerves above and smooth underneath; fructiferous spikelets curved-scorpioid; fruit spherical, shortly and obtusely beaked, 12 to 13 mm in diameter; albumen deeply ruminated; leaf-sheaths armed SPRGEE CUSEOTIN, GOINGS sii si ics Seon ccccen 22. C. Samian. e’. Leaflets conspicuously inequidistant. f’. Leaflets more or less distinctly geminate on each side of the rachis, 5-pluricostulate, oblong or lanceolate, the leaflets of each pair parallel, that is, not approximate by their bases and not divaricating; fruiting perianth pedicelliform, the fruit itself furnished with a short, pedicelliform or necklike involucrophore. - Robust; leaflets very large, pluricostulate oblong-spathulate with smooth nerves on both surfaces; 35 to 45 cm long, 6 to 10 cm wide; fruit spherical, 18 mm in diameter; leaf-sheaths unarmed .............2...00..2.-.... 23. C. megaphyllus. g- Rather slender; leaf-sheaths armed with slender spines; leaflets elliptic-lanceolate or oblanceolate, 15 to 25 cm long, 3 to 6.5 cm wide, with five slender costae almost smooth on both surfaces .. 24. C. Elmerianus. g’. Slender; leaf-sheaths unarmed; leaflets oblanceolate- elliptic or oblong-spathulate, 20 to 24 cm long, 5 to 6.5 em wide, smooth on both surfaces; fruit globose, 9 to 10 mm in diameter. 25. C. mitis. f’. Leaflets 5-costulate, those of each pair very approximate by their bases and divaricate; female spikelets having the involucrophorum (where known) not the least pedicelli- form or necklike, but immersed within its spathel. g. Of medium size; leaf-sheaths strongly spinous; leaflets elliptic-lanceolate, 22 to 25 cm long, 6 to 6.5 em wide, very frequently furnished with one or two rigid spines on the midrib above near the base, otherwise smooth on both surfaces; fruiting perianth obconical, almost spreading 26. C. Reyesianus. g’. Slender; leaf-sheaths smooth or very sparingly spinulous; leaflets elliptic-lanceolate, 10 to 16 em long, 2 to 3.5 cm wide, usually furnished with a few erect needlelike spines on some of the five costae on the upper surface and on the margins or else entirely smooth; fruiting perianth almost explanate; fruit spherical, 10 to 11 mm in diameter. 27. C. spinifolius. g’. Leaf-sheaths 2.5 cm in diameter, armed with short spines; leaflets deep green and smooth when dry, almost equally e shiny on both surfaces, lanceolate-elliptic, quite devoid of hairs or spinules even at the apex and on the margins, occasionally furnished above with a robust spinule on the midrib near the base; male spikelets flattened- pectinate, with contiguous flowers and very approximate bracteiform spathels. 28. C. viridissimus. f’. Leaflets in distant groups; the latter composed of more than two leaflets on each side of the rachis; leaf-rachis smooth. 1 g . XIV, 8 Beccari: The Palms of the Philippine Islands 347 g’. Slender; leaf-rachis smooth above; leaflets in groups of 2 to 4 on each side of the rachis, lanceolate, very long-acuminate to a filiform tip, 3- sub 5-costulate, 20 to 30 cm long, 15 to 20 cm wide; female spadix very diffusely paniculate; spikelets filiform; fruit very small, spherical, 5 mm in diameter............ 29. C. microsphaerion. h’. Leaf-sheaths unarmed; leaflets glabrous on both surfaces. C. microsphaerion (forma typica), h’. Leaf-sheaths strongly armed with short spines; leaflets slightly bristly-spinulous on one to three nerves on the upper surface...... C. microsphaerion var. spinosior. g°. Rather slender; leaflets in distant groups of 3 to 9 on each side of the rachis, linear, 1-, sub 3-costulate, smooth on both surfaces, 20 to 25 cm long, 1 cm wide; female spadix very diffuse and much branched; spikelets fili- form; fruiting perianth shortly pedicelliform; fruit very small, globose 30. C. ramulosus. f*. Leaflets very inequidistant, yet not distinctly grouped on each side of the rachis, the latter strongly prickly above, at least in its lower portion; leaflets elongate, linear- lanceolate, rigid, 3-costulate, more or less spinuliferous on the upper surface, smooth underneath, 30 to 32 cm long, 20 to 25 mm wide; leaf-sheaths about 2 cm in diameter, armed with scattered, pale, acicular spines; male and female spadices shorter than the leaves, simply and spread- ingly branched..... 31. C. Vidalianua. c’. Primary spathes very loosely sheathing, usually short, and more or less inflated in their upper part. The species of this group are difficult to discriminate if the specimens are not with mature fruits; the male spadices alone do not offer appreciable characters for specific distinction. ad’. Fruit very small, having convex scales, the latter only slightly or not at all grooved along the center and with the points not appressed or subsquarrose. e’. Leaf-sheaths armed with scattered slender spines or almost smooth; fruiting perianth pedicelliform, terete. f’. Leaf-sheaths armed with scattered slender spines; primary spathes aculeolate; fruit ovoid or subobovoid, 6 mm long, 3.5 to 4 mm thick, the scales arranged in fifteen longi- tudinal series............ 32. C. siphonospathus (forma typica). f°. Leaf-sheaths almost spineless; primary spathes smooth; fruit with scales arranged in fifteen longitudinal series. C, siphonosphathus var. sublaevis. f°. Fruit with scales in twelve longitudinal series; leaflets with five bristly nerves on the upper surface. C. siphonospathus var. oligolepis major. f*. Smaller; fruit with scales in twelve longitudinal series; leaflets with three bristly nerves on the upper surface. C. siphonospathus var. oligolepis minor. f*. Primary spathes aculeolate; fruit with scales in eighteen longitudinal series............C. siphonospathus var. polylepis. f’. Primary spathes very slightly inflated; fruit elongate-ellip- 348 The Philippine Journal of Science 1919 soid, 10 to 11 mm long (including the perianth), 5 mm thick; scales in fourteen or fifteen longitudinal series. C. stphonospathus var. batanensis. e’. Leaf-sheaths very densely armed, at least in their upper part, with ascending unequal spines. f’. Leaf-sheaths armed with elongate, ascending, very narrowly laminar spines; the elongate ligula densely armed with similar spines; petiole and rachis armed irregularly with unequal spines; leaflets with rigid bristles on the midrib alone above, smooth underneath; margins conspicuously spinulous-ciliate; fruit small, ovoid-ellipsoid, 8 to 9 mm long, 5 mm thick; fruiting perianth campanulate, sub- pedicelliform........ 33. C. dimorphacanthus (forma typica). g'. Leaf-sheaths armed with unequal long spines, some of which are very slender and criniform, others laminar; the very elongate ocrea is also armed with similar spines; leaflets with rigid bristles on three nerves above and smooth underneath; margins closely and finely ciliate-spinulous. C. dimorphacanthus var. montalbanicus. g. Leaf-sheaths very densely armed, in their upper part mostly, with very rigid subcriniform spines; leaflets very rigid, furnished on the upper surface with distant coarse bristles on the midrib alone, the lower surface smooth, margins coarsely spinulous; fruit larger than in the species, globose, 13 mm long, 10 mm thick; supported by the short terete pedicelliform perianth. C. dimorphacanthus var. zambalensis. @, Fruit covered by strongly gibbous scales, very deeply grooved along the center, and with very appressed points. e*. Fruit ovoid or subglobose-ovoid, 8 to 12 mm long, including the short, terete, supporting perianth, and 5 to 8 mm thick, obtusely beaked. f’. Leaflets numerous, elongate, 10 to 15 mm wide; leaves with subequidistant leaflets, at least in their lower part, and more or less grooved above. 34. C. microcarpus (forma typica) . f. Very slender; leaflets very narrow, not numerous, and very inequidistant; spadix small..C. microcarpus var. diminutus. e’. Fruit ovoid or subovoid, minutely beaked, 17 mm long, includ- ing the short, terete, pedicelliform perianth, and 12 mm through 35. C. halconensis. e’. The primary spathes at first enveloping the partial inflorescences, then splitting longitudinally and opening flat, becoming laminar and finally falling in decay; leaflets conspicuously discolorous, © green above and with a chalky coating underneath; fruit small, ovoid or globose-ovoid, 8 to 9 mm long, including the short, terete, pedicelliform perianth, and 5 mm thick .......... 36. C. bicolor. 1. CALAMUS MOLLIS Blanco FI. Filip. (1837) 264; ed. 2 (1845) 184; ed. 3, 1 (1877) 329; Kunth Enum. Pl. 3 (1841) 594; Mart. Hist. Nat. Palm. 3 (1838) 886; Walp. Ann. 3 (1853) 486, 5 (1858) 831; x1v,s Beecari: The Palms of the Philippine Islands 349 Mig, Fl. Ind. Bat. 3 (1855) 123; Bece. in Rec. Bot. Surv. Ind. 2 (1902) 204, Perk. Fragm. Fl. Philip. 1 (1904) 46, Webbia 1 (1905) 345, Philip. Journ. Sci. 3 (1908) Bot. 342, Ann. Bot. Gard. Calcutta 11 (1908) 212, t. 61, 62 and Suppl. 21; C. B. Robinson in Philip. Journ. Sci. 6 (1911) 117; Farnsworth gars Bur. Ed. Bull. 54 (1915) 19; 72-73 with fig. Calamus Haenkeanus Mart. 1. c. 3: 212 (1st ed.) and 337; Kunth Enum. Plant 3 (1841) 211; Walpers Ann. 3 (1852) 488, 5 (1858) 831; Mig. Fl. Ind. Bat. 3 (1855) 127. Calamus usitatus (not of Blanco) Mart. op. cit. 340; Vidal Sinopsis Atlas (1883) ¢. 98, f. D. Calamus sp. Cuming No. 1478; Vidal Phan. Cuming Philip. (1885) 154. Luzon, Camarines Province, Cuming 1478: “Manilla,” Gaudichaud in Herb. Deless.: Rizal Province (Morong) Vidal 1939, Herb. Kew.: ; Bosoboso, Loher 1372, 1367, Herb. Kew., Bur. Sci. 381 Ramos; Montalban, Toker 7081, Herb. Kew.; Antipolo, For. Bur 389 Ahern’s collector: Bataan Province, Mount Mariveles, Warburg 12506, Herb. Berol., For. Bur. 6372 Curran, For. Bur. 3026 Borden, For. Bur. 2481 Barnes: Butheaii Province, Baliuag, Bur. Sci. 9571 Robinson: Nueva Ecija Province, For. Bur. 8484 Curran: Ilocos Norte Province, Bur. Sci. 7728 Ramos: Bontoc Subprovince, For. Bur. 17034 Curran: Nueva Vizeaya Province, Dupax, Bur. Sci. 11218, 11822 McGregor: Cagayan Province, For. Bur. 17273 Curran, local name barit. BaBuyAN IsLANDS, Camiguin, Bur. Sci. 4032 Fénix. Bu.IRAN, Bur. Sci. 18634 McGregor. MINDANAO, Surigao Province, Bolster 347: Zamboanga District, For. Bur. 4819 Hutchinson: Davao District, Todaya, Mount Apo, Elmer 11969, local name arit. CALAMUS MOLLIS Blanco var. MAJOR Becc. in Webbia 1 (1905) 345, Ann. Bot. Gard. Calcutta 11: Suppl. 28; Merr. in Philip. Journ. Sci. 1 (1906) Suppl. 31. Luzon, Bataan Province, Lamao River, Mount Mariveles, For. Bur. 2499 Meyer, local name uay. CALAMUS MOLLIS Blanco var. PALAWANICUS Bece. in Philip. Journ. Sci. 2 (1907) Bot. Gard. Calcutta 11: Suppl. 22. PALAWAN, For. Bur. 3618 Curran, Bur. Sci. 609 Foxworthy; Puerto Prin- cesa, Bur. Sci. 191, 196 Bermejos; Brooke’s Point, Elmer 12607. Barely distinguishable from the type; the leaflets, however, are frequently almost equidistant. 2. CALAMUS MEYENIANUS Schauer in Nova Acta Acad. Caes. Nat. Cur. 19 (1843) Suppl. 1: 425; Becc. in Rec. Bot. Surv. Ind. 2 (1902) 217, Ann. Bot. Gard. Caleutta 11 (1908) 215, t. 68. Luzon, Pangasinan Province, near the village of San Mateo on Mount Masiquie, Callery in Herb. Paris; Umingan, Bur. Sci. 17755 Otanes: Nueva Vizcaya Province, vicinity of Dupax, Bur. Sci, 11442 McGregor. 8. CALAMUS BLANCO! Kunth Enum. Pl. 3 (1841) 595; Mart. Hist. Nat. Palm. 3 (1849) 343; Walpers, Ann. 3 (18583) 492, 5 (1858) 832; Mig. Fl. Ind. Bat. 3 (1855) 139; Bece. in Rec, Bot. Surv. Ind. 2 (1902) 204. 164391——-7 350 The Philippine Journal of Science 1919 Calamus gracilis (not of Roxb.) Blanco Fl. Filip. (1837) 267, ed. 2 (1845) 186, ed. 3, 1 (1877) 332. Calamus brevifrons Mart. op. cit. 3: 388; Mig. op. cit. 127. Calamus parvifolius Vidal Phan. Cuming. Philip. (1885) 154, non Roxb. Luzon, Albay Province, Cuming 1225: Bataan Province, Mount Mari- veles, Loher 1876: Zambales Province, For. Bur. 887 Maule. LEYTE, Palo, Elmer 7282. 4. CALAMUS MELANORHYNCHUS Bece. in Ann. Bot. Gard. Calcutta 11: Suppl. 30, ¢. 16. MINDANAO, Davao District, Mount Apo, Elmer 11708. 5. CALAMUS FILISPADIX Bece. in Philip. Journ. Sci. 6 (1911) Bot. 230, Ann. Bot. Gard. Caleutta 11 (1908) Suppl. 28, t. 15. Calamus Hookerianus Becc. in Philip. Journ. 5 (1909) Bot. 621 (non in Ann. Bot. Gard. Caleutta 11: 226). Luzon, Albay Province, Adlumay Hills, For. Bur. 10680 Curran (with young fruits; type specimen): Tayabas Province, Kabibihan, Bur. Sci. 18128 Foxworthy & Ramos (2 specimen in flower); Rizal Province, Mon- talban, Loher 7084 in Herb. Kew. (this specimen was erroneously mentioned by me under C. Diepenhorstii var. exulans Becc. in Ann. Bot. Gard. Calcutta, Suppl. 51). Po.inLo, Bur. Sci. 10467 McGregor. PALAWAN, Puerto Prin- cesa, Elmer 12769; Taytay, Merrill 7251 (sterile specimen) (the Palawan plant perhaps represents a distinct variety). MINDANAO, Agusan Province, Mount Urdaneta, Elmer 13949 (perhaps representing a distinct variety). 6. CALAMUS DIEPENHORSTII Mig. var. EXULANS Becc. in Philip. Journ. Sci. 5 (1909) Bot. 627, 6 (1911) Bot. 230, Ann. Bot. Gard. Calcutta 11: Suppl. 51 (excl. Loher 7054). PALAWAN, Taytay, Merrill 9357; Mount Victoria, Bur. Sci. 756 Foz- worthy, POoLiLLo, Bur. Sci. 9111 Robinson. 7. CALAMUS ORNATUS Blume var. PHILIPPINENSIS Becc. in Ann. Bot. Gard. Calcutta 11 (1908) 370. Calamus maximus Blanco Fl. Filip. ed. 1 (1837) 265, ed. 2 (1845) 185, ed. 3, 1 (1877) 331; Kunth Enum. 3 (1841) 595; Mart. Hist. Nat. Palm. 3 (1849) 343; Walp. Ann. 3 (1853) 492, 5 (1858) 882; Miq. Fl. Ind. Bat. 3 (1855) 138. LUZON, without definite locality, Loher 1887 Herb. Kew.: Bataan Prov- ince, Mount Mariveles, Whitford 343, 502, For. Bur: 212 Barnes, For. Bur. 2489 Borden: Tayabas Province, Lucban, Elmer 7625; Unisan, Reyes: Laguna Province, San Antonio, Bur. Sci. 16619 Ramos: Cagayan Province, For. Bur. 17252 Curran (3 flowers): Rizal Province, Montalban, Loher in Herb. Kew., Tagalog name limuran or alimoran. MINporo, Bongabong River, For. Bur. 3911, 3913 Merritt. Necros, For. Bur. 6089 Everett. BASILAN, For. Bur. 6107 Hutchinson. Poo, Bur. Sci. 9266 Robinson, Bur. Sci. 10461 McGregor, Tagalog name limuran. MINDANAO, Davao Dis- trict, Mount Apo, Elmer 11236. 8. CALAMUS DISCOLOR Mart. Hist. Nat. Palm. 3: 212 (1st ed.) and 841; Kunth Enum. Pl. 3 (1841) 212; Walp. Ann. 3 (1853) 491, 5 (1858) 832; Mig. Fl. Ind. Bat. 3 (1855) 1386; Becc. in Ann. Bot. XIV, 8 Beccari: The Palms of the Philippine Islands 851 Gard. Caleutta 11 (1908) 495, t. 228, and Suppl. 125, ¢. 71, Leafi. Philip. Bot. 2 (1909) 649. Calamus Lindenii Rodigas Illustr. Hort. 30 (1883) 157, t. 489; Ridley in Journ. Str. Branch Roy. As. Soc. 44 (1905) 200. Luzon, Tayabas Province, Lucban, Elmer 9299: Laguna Province, San Antonio, Bur, Sci. 12085 Ramos, Tagalog hamlis. CALAMUS DISCOLOR Rodigas var. NEGROSENSIS Becc. in Philip. Journ. Sci. 4 (1909) Bot. 6385, Ann. Bot. Gard. Calcutta 11: Suppl. t. 72. NEGROS, Cadiz, For. Bur. 12432 Danao, For. Bur. 20893 Tamesis. §. CALAMUS CUMINGIANUS Bece. in Record Bot. Surv. Ind. 2 (1902) 210, Ann. Bot. Gard. Calcutta 11 (1908) 348, t. 142. Calamus sp. Vidal Phan. Cuming. Philip. (1885) 18, 154. Luzon, Tayabas Province, Cuming 762 in Herb. Kew. To this species apparently also belongs Elmer 13646, collected at Cabadbaran, Mount Urdaneta, Agusan Province, Mindanao. 10. CALAMUS SIMPHYSIPUS Mart. Hist. Nat. Palm. 3: ¢. he Be fi dd Bece. in Ann. Bot. Gard. Calcutta 11 (1908) 346, #4. 147. MINDANAO, Agusan Province, Cabadbaran (Mount Urdaneta), along Can- tagan Creek at about 900 meters’ altitude, Elmer 18902, Manobo name balanog. 11. CALAMUS MERRILLI! Becc. Ann. Bot. Gard. Calcutta 11 (1908) 3890, t. 167, and Suppl. 78, t. 41, Webbia 1 (1905) 347, Philip. Journ. Sci. 4 (1909) Bot. 629. Calamus maximus (not of Blanco) Bece. in Perk. Fragm. FI. Philip. (1904) 45. Luzon, Rizal Province, Bosoboso, Merrill 1893 (type specimen); Mount Matulid (1,200 meters’ altitude), Loher 7074 in Herb. Kew. (male speci- men); Montalban, Loher 7076 in Herb. Kew. (fruit): Laguna Province, For. Bur. 20654 Villamil. MrinpaNao, Lake Lanao, Camp Keithley, Mrs. Clemens 1112, 1124: Agusan Province, Cabadbaran, Mount Urdaneta, Elmer 18926, Manobo name palasan: Davao District, Todaya, Mount Apo, Elmer 11885, Bagobo name acab-acab. The specimens from Mindanao represent a transition to var. Merrittianus. Apparently it grows also in Basilan Island, judging from a sterile specimen collected by Hutchinson, For. Bur. 6098. CALAMUS MERRILLII Becc. var. NANGA Becc. in Ann. Bot. Gard. Cal- cutta 11: Suppl. 79. MINDANAO, Davao District, Mount Apo, Elmer 11874, 11110. CALAMUS MERRILLII Becc. var. MERRITTIANUS Bece. in Ann. Bot. Gard. Caleutta 11: Suppl. 78 t. 42. Calamus Merrittianus Becc. in Philip. Journ. Sci. 2 (1907) Bot. 233, 4 (1909) Bot. 629. Mrnporo, Bongabong River, For. Bur. $912 Merritt. 352 The Philippine Journal of Science 1919 12. CALAMUS FOXWORTHY! Becc. in Ann. Bot. Gard. Calcutta 11: Suppl. 81, t. 45. PALAWAN, Victoria Peak, Bur. Sci. 690 Foxworthy. 13. CALAMUS MANILLENSIS’H. Wendl. in Kerch. Les Palm. (1878) 287; Becc. in Rec. Bot. Surv. India 2 (1902) 215, Webbia 1 (1905) 349, Ann. Bot. Gard. Calcutta 11 (1908) 413, t. 226, f. 1, and Suppl. 98, t. 55. Daemonorops manillensis Mart. Hist. Nat. Palm. 3: 3380, t. 175, VIII, f. 1-8; Walp. Ann. 3 (1853) 480, 5 (1858) 829. LuZON, “Manilla,’”’ Gaudichaud in Herb. Delessert: Nueva Vizcaya Prov- ince, Dupax, Bur. Sci. 18994 McGregor. MINDANAO, Davao District, Mount Apo, Elmer 11714, 10560: Agusan Province, Mount Urdaneta, Elmer 14133, 14011, 14178. 14. CALAMUS TRISPERMUS Becce. in Perk. Fragm. Fl. Philip. (1904) 46, Ann. Bot. Gard. Calcutta 11 (1908) 412, ¢. 180, and Suppl. 97, Webbia 1 (1905) 349. Luzon, Rizal Province, Antipolo, Merrill 1645; Montalban, Loher 7071 in Herb. Kew. 15. CALAMUS ARUGDA Bece. in Philip. Journ. Sci. 4 (1909) Bot. 622, Ann. Bot. Gard. Caleutta 11: Suppl. 90, ¢. 50. Luzon, Cagayan Province, Lalloc, For. Bur. 6649 Klemme. 16. CALAMUS VINOSUS Becc. in Leafl. Philip. Bot. MS. MINDANAO, Agusan Province, Mount Urdaneta, Elmer 14158. A well-characterized species, resembling C. multinervis, but offering the very uncommon characters of two female flowers at every spathel with a neuter one interposed, and the wine red color of the ripe fruits. 17, CALAMUS MOSELEYANUS Bece. in Record Bot. Surv. Ind. 2 (1902) 211, Ann. Bot. Gard. Caleutta 11 (1908) 396, t. 171, and Suppl. 82, Webbia 1 (1905) 348. MALANIPA near Basilan, Moseley in Herb. Kew. MINDANAO, Zamboanga District, San Ramon, Hallier: Davao District, Mount Apo, Elmer 11886. 18. CALAMUS MINDORENSIS Becc. in Ann. Bot. Gard. eugicter 11: Suppl. $2.4. $6; MINDORO, Balete, on Baco River, McGregor 309, For. Bur. 6217 Merritt, local name tumalin. 19. CALAMUS MULTINERVIS Bece. Ann. Bot. Gard. Caleutta 11: Suppl. 88, t. 49. MINDANAO, Davao District, Mount Apo, Elmer 11955 (male plant), Ba- gobo name balala; and in the same district, south of the Baruring River, Elmer 11791 (specimen with young fruits), Bagobo name ubbli. 20. CALAMUS GRANDIFOLIUS Bece. in Philip. Journ. Sci. 4 (1909) Bot. 629, Ann. Bot. Gard. Calcutta 11: Suppl. 94, t. 58. Luzon, Laguna Province, Mount Banahao, Loher 7088 in Herb. Kew.: Tayabas Province, Infanta, Bur. Sci. 9448 Robinson. XIV, 3 Beccari: The Palms of the Philippine Islands 853 21. CALAMUS JENNINGSIANUS Becc. in Philip. Journ. Sci. 4 (1909) Bot. 623, Ann. Bot. Gard. Calcutta 11: Suppl. 91, t. 51. MiInpDoRO, Mount Halcon, For. Bur. 4400 Merritt. 22. CALAMUS SAMIAN Becc. Ann. Bot. Gard. Calcutta 11: Suppl. 92, te 62; MINDANAO, Davao District, Mount Apo, Elmer 11836, Bagobo name samian. 23. CALAMUS MEGAPHYLLUS Becc. in Ann. Bot. Gard. Calcutta 11: Suppl. 66, t. 35. MINDANAO, Davao District, Mount Apo, Elmer 11878: Agusan Province, Mount Urdaneta, Elmer 13542. 24. CALAMUS ELMERIANUS Becce. in Leafl. Philip. Bot. 2 (1909) 647, Ann. Bot. Gard. Calcutta 11: Suppl. 69, t. 37. Luzon, Tayabas Province, Lucban, Elmer 9298 (type specimen). MIN- DANAO, Davao District, Mount Apo, Elmer 11756; Agusan Province, Mount Urdaneta, Elmer 14166; Agusan River, Merrill 7289. (The Mindanao plant perhaps represents a distinct variety.) 25. CALAMUS MITIS Becc. in Philip. Journ. Sci. 3 (1908) Bot. 341, Ann. Bot. Gard. Calcutta 11: Suppl. 68, t. 36. BATAN ISLANDS, Batan, Bur. Sci. 3817 Féniz. 26. CALAMUS REYESIANUS Becc. in Philip. Journ. Sci. 2 (1907) Bot. 237, Ann. Bot. Gard. Calcutta 11: Suppl. 86, t. 48. Luzon, Tayabas Province, Unisan, Reyes; Tagcauayan, Bur. Sci, 13312 Ramos. Probably a sterile specimen collected by Hutchinson, For. Bur. 6105, in Basilan Island is referable to this species. 27. CALAMUS SPINIFOLIUS Becce. in Rec. Bot. Surv. Ind. 2 (1902) 202, Ann. Bot. Gard. Calcutta 11 (1908) 410, t. 178, 179, Suppl. 95, Webbia 1 (1905) 348. PANAY, Iloilo Province, Igbaras, Vidal 3954 in Herb. Kew. and Beccari (sterile specimen). Luzon, Pampanga Province, Arayat, Loher 1873, Herb. Kew., Garcia 59, local name curacling (sterile): Bataan Province, For. Bur. 17309 Curran (male specimen), For. Bur. 1454 Ahern’s collector (fruiting specimen), For. Bur. 20947 De Leon. 28. CALAMUS VIRIDISSIMUS Becc. Ann. Bot. Gard. Calcutta 11: Suppl. 84, t. 47. MINDANAO, Davao District, Mount Apo, Elmer 11988, Bagobo name, acal. Probably the following sterile specimens also belong to C. viridissimus: MINDANAO, Zamboanga District, For. Bur. 4820, 5180 Hutchinson. 29. CALAMUS MICROSPHAERION Becc. in Perk. Fragm. Fl. Philip. (1904) 45, Webbia 1 (1905) 349, Ann. Bot. Gard. Calcutta 11 (1908) 548, t. 204, and Suppl. 113. CULION, Halsey Harbor, Merrill 507. 354 The Philippine Journal of Science 1919 Apparently a sterile specimen collected by Hutchinson in Mindanao, For. Bur. 4818, also belongs to this species. In this specimen the sheathed stem is only 8 to 9 mm in diameter, and the sheaths are provided with some very small, scattered, tuber-. culiform spines. CALAMUS MICROSPHAERION Becc. var. SPINOSIOR Bece. var. nov. Sheathed stem 2 to 3 cm in diameter; leaf-sheaths densely armed with scattered rather robust spines 5 to 12 mm long and having broad subbulbous bases. Leaflets bristly-spinulous on 1 to 3 costae above. PALAWAN, Taytay, Merrill 9858 (with d flowers), 9247 (with fruits). 80. CALAMUS RAMULOSUS Bece. in Perk. Fragm. Fl. Philip. (1904) 46, Webbia 1 (1905) 349, in Ann. Bot. Gard. Calcutta 11 (1908) 454, t. 205. LUZON, Tayabas Province, Guinayangan, Merrill 2070. 31. CALAMUS VIDALIANUS Becc. in Rec. Bot. Surv. Ind. 2 (1902) 212, Webbia 1 (1905) 850, Ann. Bot. Gard. Calcutta 11 (1908) 464, t. 121, and Suppl. 114, t. 64. Calamus horrens ? Vidal Rev. Pl. Vasc. Filip. (1886) 280 (not of Blume). Luzon, Tayabas Province, Unisan, Vidal 933, in Herb. Kew. (type specimen) ; Guinayangan, Bur. Sci. 13214 Ramos (db specimen in flower) : Rizal Province, Montalban, Loher 7087 in Herb. Kew., Tagalog bogtong (¢ specimen in flower): Bataan Province, For. Bur. 20948 De Leon: Nueva Ecija Province, For. Bur. 22162 Alvarez (dé). 32. CALAMUS SIPHONOSPHATHUS Mart. Hist. Nat. Palm. 3: 342; Walpers Ann. 3 (1853) 491, 5 (1858) 832; Miq. FI. Ind. Bat. 3 (1855) 187; Becc. in Rec. Bot. Surv. Ind. 2 (1902) 218, Webbia 1 (1905) 350, Ann. Bot. Gard. Calcutta 11 (1908) 471, and Suppl. 115. Calamus sp. Vidal Rev. Pl. Vasc. Filip. (1886) 330 (No. 931 and 1942). Calamus siphonospathus is a very variable plant; it is, how- ever, easily distinguishable by its spadices sheathed in gradually diminishing, subimbricate, loosely-tubular, subinflated spathes; by their dense and short, supradecompound, female inflorescen- ces; and'by the very small fruits. The inflated spathes of this Calamus are, apparently, ant-harboring organs. CALAMUS SIPHONOSPATHUS Mart. forma TYPICA Becc. in Ann. Bot. Gard. Caleutta 11 (1908) 473. Luzon, “Manila,” Perrottet in Herb. Delessert: Rizal Province, Bosoboso, Merrill 1891, Loher 1364 in Herb. Kew.; Montalban, Loher 7089 in Herb. Kew. (2) and 7077 (d): Isabela Province, Cabagan River, For. Bur. 18558 Alvarez (3), Negrito name jusi. XIV, 8 Beccari: The Palms of the Philippine Islands 855 I am not quite sure that the sterile specimen, Garcia 60, mentioned in “Asiatic Palms” really is C. siphonospathus. CALAMUS SIPHONOSPATHUS Mart. var. SUBLAEVIS Becc. in Webbia 1 (1905) 852, Ann. Bot. Gard. Calcutta 11 (1908) 474. LuzoN, Bataan Province, Lamao River, Whitford 308, For. Bur. 1577 Borden, For. Bur. 2635 Meyer, For. Bur. 6225, 6514 Curran: Ilocos Norte Province, Mount Piao, For. Bur. 13985 Merritt & Darling. CALAMUS SIPHONOSPATHUS Mart. var. FARINOSUS Becc. Ann. Bot. Gard. Caleutta 11 (1908) 474. Luzon, Rizal Province, Montalban, Loher 7088 in Herb. Kew. CALAMUS SIPHONOSPATHUS Mart. var. OLIGOLEPIS (major) Becc. in Webbia 1 (1905) 353, Ann. Bot. Gard. Calcutta 11 (1908) 475, f. 2415. Calamus inflatus Warb. in Herb. Berol.; Becc. in Perk. Fragm. FI. Philip. (1904) 45. Luzon, Tayabas Province, Sampaloc, Warburg in Herb. Berol. CALAMUS SIPHONOSPATHUS Mart. var. OLIGOLEPIS (minor) Becc. in Webbia 1 (1905) 358, Ann. Bot. Gard. Calcutta 11 (1908) 476, t..21Z2 Luzon, Isabela Province, Malunu, Warburg in Herb. Berol. CALAMUS SIPHONOSPATHUS Mart. var. BATANENSIS Becc. in Philip. Journ. Sci. 3 (1908) 842, Bot. Ann. Gard, Calcutta 11: Suppl. 115. BATAN ISLANDS, Batan, Bur. Sci. 3611 Féniz. 33. CALAMUS DIMORPHACANTHUS Becc. in Rec. Bot. Surv. Ind. 2 (1902) 214, Ann. Bot. Gard. Calcutta 11 (1908) 479 t. 219, Suppl. 117, Philip. Journ. Sci. 4 (1909) Bot. 631. PANAY, Iloilo Province, Igbaras, Vidal 3956, in Herb. Becc. and Kew., local name taguiti. Luzon, Union Province, Loher 1871 in Herb. Kew.: Benguet Subprovince, Loher 1370 in Herb. Kew.; Mount Santo Tomas, Elmer 6238. Perhaps a variety of, this species is Bur. Sci. 20559 Ramos, from San Antonio, Laguna Province, with ¢ flowers; the leaf- lets are very inequidistant and have three nerves covered with bristly spinules above, and the midrib minutely spinulous underneath. CALAMUS DIMORPHACANTHUS Bece. var. MONTALBANICUS Becc. in Philip. Journ. Sci. 4 (1909) Bot. 631, Ann. Bot. Gard. Calcutta 11: Suppl. 117, t. 118. Luzon, Rizal Province, Montalban, Mount Batay, Loher 7085 in Herb. Kew. CALAMUS DIMORPHACANTHUS Becc. var. ZAMBALENSIS Becc. in Philip. Journ. Sci. 4 (1909) Bot. 632, Ann. Bot, Gard. Calcutta 11: Suppl. 118. 3 : Luzon, Zambales Province, For. Bur. 8412 Curran & Merritt. 856 The Philippine Journal of Science 1919 84. CALAMUS MICROCARPUS Bece. in Rec. Bot. Surv. Ind. 2 (1902) 218, Webbia 1 (1905) 355, Ann. Bot. Gard. Calcutta 11 (1908) 477, t. 218, and Suppl. 116, Philip. Journ. Sci. 4 (1909) Bot. 627, 6 (1911) Bot. 230. Luzon, Camarines Province, Vidal 3952 in Herb. Becc. and Kew.: Rizal Province, Montalban, Loher in Herb. Kew.: Laguna Province, Mount Maquiling, For. Bur. 18305 Tamesis, For. Bur. 20668 Villamil: Tayabas Province, Bur. Sci. 12826 Foxworthy. MINDANAO, Lanao District, Lake Lanao, Mrs. Clemens 1232: Agusan Province, Cabadbaran, Elmer 13551: Davao District, Todaya, Mount Apo, Elmer 10676. PoLILLo, Bur. Sci. 9131 Robinson, Bur. Sci. 10465 McGregor. Apparently a specimen with male spadices, from Abulug River, Cagayan Province, Luzon, For. Bur. 1 1625 Fischer also belongs to C. microcarpus. CALAMUS MICROCARPUS Bece. var. DIMINUTUS Bece. var. nov. Very slender, sheathed stem 8 to 10 mm in diameter. Leaves with very narrowly linear, very long-acuminate, very inequi- distant, and more or less distantly grouped leaflets, the latter 4 to 5 mm wide. Spadix very short. Fruit globular, 8 mm in diameter. Luzon, Laguna Province, San Antonio, Bur. Sci. 16599 Ramos. I am uncertain if we have to do with a case of occasional slenderness of Calamus microcarpus or with a really distinct and constant variety or even a subspecies of it. Very similar to the preceding specimen is another, which is a little larger and with a male spadix. The sheathed stem is 11 mm in dia- meter; the leaflets are 5 to 6 mm wide. Kabibihan, Tayabas Province, Bur. Sci. 13240 Ramos. 85. CALAMUS HALCONENSIS Bece. in Philip. Journ. Sci. 4 (1909) Bot. 633, Ann. Bot. Gard. Caleutta 11: Suppl. 116, t. 65. MrInDoRO, Mount Halcon, For. Bur. 4399 Merritt. 36. CALAMUS BICOLOR Becce. in Ann, Bot. Gard. Calcutta 11. Suppl. 126, t. 73. MINDANAO, Davao District, Mount Calelan, Mount Apo, Elmer 10618, 10541, Bagobo names sambonotan and lassee or rassee. DAEMONOROPS Blume Conspectus of the species. a’. Cymbospatha. Spadix, ¢ and 2, contracted. Primary spathes cymbi- form, beaked, the outermost completely inclosing the inner ones. : : 1. D. Margaritae var. palawanicus. a’. Piptospatha. Spadix, ¢ and 9, elongated. Inner primary spathes gradually longer than the outermost. b*. Leaflets very inequidistant, ensiform, the largest 40 to 50 cm long, 2.5 to 3 em wide; fruit ovoid-ellipsoidal, 25 mm long, 16 to 17 mm thick, very shortly pedicellate.. 2. D. virescens. XIV, 3 Beccari: The Palms of the Philippine Islands 357 b*. Leaflets equidistant. ce’. The mouths of the leaf sheaths unarmed. d’. Leaflets lanceolate-ensiform, the largest 30 to 45 em long, 3 to 3.5 cm wide, the midrib only sparsely bristly below, smooth above or else minutely spinulous near the apex; fruit spherical, mammillate-beaked, 18 to 20 mm in diameter..... 3. D. ochrolepis. @. Leaflets less than 3 cm wide, bristly on three to five nerves above. e’. Leaflets 30 cm long, 15 to 16 mm wide (the largest), bristly on three nerves above, and on the midrib only underneath; the axis of the spadix and spikelets coated with a rusty- brown scurf; fruit carried on a pedicel 8 to 10 mm long, globose-ovoid, obtusely mammillate-beaked. 4. D. urdanetanus. e*. Leaflets very narrowly lanceolate, 20 to 23 em long, 14 to 18 mm. wide (the largest), bristly on three nerves above, and with a few long bristles on the midrib only underneath. Male flowers very long and slender (12 mm long). : 5. D. Loherianus. e*. Leaflets 30 cm long, 15 to 20 mm wide (the largest), bristly on five nerves above, but only on the midrib beneath, fruit 12 to 17 mm long, 9 to 11 mm through, ovoid-ellipsoid, carried on a pedicel 4 to 6 mm long...........-..........--. 6. D. pedicellaris. d’. Leaflets less than 8 cm wide, having the midrib alone spinulous, on only one or on both surfaces. e’. Leaflets 80 to 32 cm long, 2 to 2.5 em wide (the largest), having the midrib alone spinulous on both surfaces; axis of the spadix and spikelets densely coated with a copious brown felt; fruit ovoid-ellipsoid, blunt-mammillate, carried on a thick, 5 to 6 mm long pedicel 7. D. pannosus. e*. Leaflets 30 to 40 cm long, 2 to 2.5 em wide (the largest), smooth or nearly so on the upper surface, underneath the midrib alone remotely spinulous; axial parts of the spadix and spikelets coated with adherent rusty-brown scurf; fruit globose and obtusely mammillate, 17 to 18 mm in diameter, with a few well-conformed scales, and carried on a pedicel 10 to 12 mm long. 8. D. oligolepis. e. The mouths of the leaf-sheaths armed with erect spines, longer than those on the body. ad’. Fruit large, over 2 cm in diameter; leaflets narrowly ensiform, 40 to 42 cm long, 13 to 15 mm broad (the largest), spinulous on three nerves above and bristly on the midrib alone beneath; fruit spherical, 20 to 24 mm in diameter.... 9. D. Clemensianus. d’. Fruit less than 2 cm in diameter. e’. Leaflets linear-ensiform, 35 to 45 cm long, 16 to 24 mm broad (the largest), with three bristly nerves on the upper surface, underneath the midrib alone or, occasionally, also three nerves bristly; partial inflorescences and spikelets spreading; fruit globose, mammillate-beaked, often slightly depressed, 15 to 18 mm in diameter........ 10. D. Gaudichaudii. e. Leaflets narrowly ensiform, 55 to 60 cm long, 20 to 22 mm broad (the largest), almost smooth above and with only a few short bristles on the midrib underneath; partial in- 858 The Philippine Journal of Science 1918 florescences and spikelets inserted at a very acute angle; fruit globose, conically beaked, 12 mm in diameter. ; 11. D. affinis. é. Leaflets lanceolate-ensiform, 33 to 40 cm long, 17 to 20 mm broad, bristly on three to five nerves on the upper surface and on the midrib alone underneath; spikelets spreading; fruit globular or shortly ovoid, minutely beaked, 12 mm in diameter. 12. D. Curranii. e*. Leaflets linear-lanceolate, 25 to 28 cm long, 12 to 14 mm broad (the largest), sparingly spinulous on three nerves above and underneath with only a few bristles on the midrib from the middle upward; fruit small, broadly ovoid-ellipsoid, 12 mm long, 9 mm thick. ” XIV, 4 Merrill: Noteworthy Philippine Plants, XV 391 . glabrous, smooth, black when dry, about 1.7 cm long, apex very obscurely rostrate, the pedicels black when dry, 2 to 2.5 cm long. LUZON, Ilocos Norte Province, Burgos, Bur. Sci. 27333 Ramos, March 10, 1917, in forests near streams. This species is manifestly allied to Phaeanthus ebracteolatus Merr.; it is distinguished especially by its indumentum. PHAEANTHUS VILLOSUS sp? nov. Frutex erectus, perspicue et molliter villosus; foliis chartaceis, oblongis, usque ad 22 cm longis, tenuiter caudato-acuminatis, supra glabris, subtus villosis, nervis utrinque circiter 20, per- spicuis; pedicellis villosis, 7 ad 9 cm longis; floribus solitariis, circiter 4.5 cm longis, petalis interioribus caudato-acuminatis. A shrub, the branches, lower surface of the leaves and the pedicels softly and densely villous. Leaves chartaceous, oblong, 16 to 22 cm long, 6 to 8 cm wide, the apex slenderly caudate- acuminate, base acute to somewhat rounded, the upper surface glabrous, shining, the lower surface densely and softly villous on the midrib and nerves; lateral nerves about 20 on each side of the midrib, distinct; petioles densely villous, 7 to 10 mm long. Flowers yellow, extra-axillary, solitary, their densely villous pedicels 8 to 9 cm long. Sepals and the three outer petals minute, pubescent, triangular-ovate, about 1 mm long. Inner petals coriaceous-fleshy, oblong-elliptic, about 4.5 cm long, to 2.3 cm wide, somewhat pubescent outside and with slender longitudinal nerves, base somewhat narrowed, apex subcaudate- acuminate. Anthers numerous, oblong, about 3 mm long, the connectives oblique, somewhat concave. Carpels numerous, _ oblong, about 3 mm long, somewhat pubescent ; style club-shaped, pubescent, about 1.5 mm long. PANAY, Capiz Province, Mount Macosolon, Bur. Sci. 30775 Ramos & Edafio, April 23, 1918, along small streams in damp forests. This species is well characterized by its rather dense soft indumentum, which is found on most parts of the plant. It is most closely allied to Phaeanthus pubescens Merr., from which it differs essentially in its much more numerous lateral nerves. POLYALTHIA Blume POLYAETHIA DOLICHOPHYLLA sp. nov. § Eupolyalthia. Frutex subglaber; foliis brevissime petiolatis, chartaceis, oblanceolatis ad oblongo-oblanceolatis, usque ad 50 cm longis, 11 ad 18 cm latis, basi perspicue cordatis, subamplexicaulibus, 392 The Philippine Journal of Science 1919 nervis utrinque circiter 18, valde perspicuis; floribus axillaribus, pedicellatis, circiter 4 cm longis; calycis lobis 1.5 ad 2 cm longis; petalis crasse coriaceis, exterioribus circiter 1.5 cm latis, inte- rioribus angustioribus, infra contractis, basi leviter inflatis; carpellis anguste oblongis, pubescentibus, ovulis 2, superpositis. A shrub, 2 to 3 m high, glabrous except the petioles and the midribs on the lower surface toward the base. Leaves char- taceous, oblanceolate to oblong-oblanceolate, about 50 cm long, 11 to 18 cm wide, the upper surface grayish when dry, shining, the lower surface brownish, apex somewhat acuminate, narrowed below the middle to the abruptly rounded and prominently cor- date base, the base 5 to 8 cm wide, the basal lobes somewhat amplexicaul; lateral nerves about 18 on each side of the midrib, very prominent on the lower surface, curved, arched, anasto- mosing; petioles very stout, 8 to 10 mm long, pubescent. Flowers yellow or reddish-yellow, axillary, solitary, at maturity about 4 em long, their pedicels up to 1.5 cm long, each with an ovate, acuminate, basal bracteole about 1 mm long. Calyx-lobes ovate to oblong-ovate, acute or obtuse, chartaceous, 1.5 to 2 cm long. Petals about 3.5 cm long, thickly coriaceous, glabrous, the outer three about 1.5 em wide, the inner three about 1 em wide, acute or obtuse and distinctly narrowed below the middle and above the slightly inflated base. Stamens indefinite, the anthers oblong, 2.8 mm long. Carpels numerous, narrowly oblong, pubescent, the stigmas densely pilose; ovules 2, superposed. . PANAY, Capiz Province, Jamindan, Bur. Sci. 30983 (type), 36873 Ramos & Edaio, April and May, 1918, along small streams in forests. This remarkable species is prominently characterized by its unusually large leaves which are narrowed below to the abruptly rounded and prominently cordate, somewhat amplexicaul base; the petioles are unusually short, some of the leaves being sub- sessile. It has no close alliance among the Malayan species known to me. PSEUDUVARIA Miquel PSEUDUVARIA GRANDIFLORA sp. nov. Arbor circiter 5 m alta, plus minusve cinereo-pubescentibus, novellis dense ferrugineo-pubescentibus; foliis oblongis, usque ad 20 cm longis, obtuse acuminatis, basi acutis ad subrotundatis, nervis utrinque 8 ad 11, curvato-adscendentibus, perspicuis; flo- ribus axillaribus, numerosis, fasciculatis, longe pedicellatis, cir- citer 1.5 cm longis. A tree, about 5 m high, the branches dark grayish-brown, XIV, 4 Merrill: Noteworthy Philippine Plants, XV 393 rugose, glabrous, the branchlets rather densely cinereous-pubes- cent, the young tips ferruginous-pubescent. Leaves chartaceous, oblong, 13 to 20 cm long, 4 to 5.5 em wide, pale-olivaceous and somewhat shining on the upper surface when dry, the midrib pubescent, the lower surface paler, pubescent on the midrib and lateral nerves, base acute, apex blunt-acuminate; lateral nerves 8 to 11 on each side of the midrib, curved-ascending, prominent, anastomosing; petioles pubescent, 5 to 7 mm long. Flowers numerous, axillary, fascicled, about 1.5 cm long, usually two or three in each fascicle, their pedicels pubescent, 1.5 to 2.5 cm long. Sepals reniform, rounded, pubescent, about 1.8 mm long and 2 mm wide. Outer three petals suborbicular-ovate, pubescent, rounded, 3.5 to 4 mm long; inner three petals arched, the claw about 9 mm long, pubescent externally, glabrous within, the limb rhomboid, about 10 mm wide, pubescent externally and on the margins, glabrous within. Stamens very numerous, about 1 mm long. Female flowers unknown. LUZON, Ilocos Norte Province, Bangui, Bur. Sci. 27514 Ramos, March 9, 1917, in damp forests at low altitudes, Ilocano name apnit. This species is readily distinguished from the only other known Philippine species, Pseuduvaria philippinensis Merr., by its very much larger flowers. UVARIA Linnaeus UVARIA PANAYENSIS sp. nov. Frutex scandens, floribus dense griseo-puberulis exceptis gla- ber vel subglaber; foliis oblongis, chartaceis ad subcoriaceis, usque ad 28 cm longis, tenuiter acuminatis, basi cordatis, nervis utrinque circiter 15, perspicuis; floribus pallidis, circiter 4.5 cm diametro; sepalis reniformibus; petalis coriaceis, utrinque dense cinereo-puberulis, late ovatis, 2.5 cm longis; carpellis lineari- oblongis, glabris, pauciovulatis. A scandent shrub, nearly glabrous except the densely grayish- puberulent flowers. Branches glabrous, the branchlets with very few, scattered, ferruginous, scarcely stellate hairs. Leaves oblong, firmly chartaceous or subcoriaceous, 25 to 28 cm long, 8 to 9 em wide, entirely glabrous, grayish on both surfaces when dry, the apex slenderly and sharply acuminate, base rounded and distinctly cordate; lateral nerves prominent, curved, anas- tomosing, about 15 on each side of the midrib, the reticulations lax; petioles 5 mm long or less, black when dry. Flowers about 4.5 cm in diameter, apparently pale or yellowish, their pedicels about 3 cm long, glabrous or with very few, short, ferruginous, 394 The Philippine Journal of Science 1919 obscurely stellate hairs, each with a broadly ovate, ferruginous- pubescent, 3 to 4 mm long bract at or below the middle. Sepals reniform, coriaceous, about 6 mm long, 9 mm wide, externally cinereous-pubescent. Petals coriaceous, densely cinereous-pu- bescent on both surfaces, the outer three broadly ovate, obtuse, about 2.5 cm long, 2 cm wide, the inner three somewhat smaller. Anthers very numerous, linear-oblong, 3.5 to 4 mm long, the con- nectives rounded-truncate, not oblique, very slightly produced. Carpels many, linear-oblong, glabrous, including the stigma 3 mm long; ovules few. PANAY, Capiz Province, Jamindan, Bur. Sci. 31367 Ramos & Edano, May 22, 1918, in forests along small streams. This species is manifestly allied to Uvaria leytensis (Elm.) Merr., from which it is readily distinguished by its distinctly cordate leaves and its cinereous-puberulent flowers. UVARIA MACGREGORII sp. nov. Frutex scandens, inflorescentiis exceptis glaber; foliis char- taceis, oblongo-ovatis et ovato-ellipticis, usque ad 12 cm longis, griseo-olivaceis, apice obtusis, basi rotundatis et plerumque levi- ter cordatis, nervis tenuibus, utrinque circiter 10; floribus cir- citer 2 cm diametro, rubro-brunneis; sepalis margine ferrugineo- pubescentibus; petalis late ovatis, circiter 1 cm longis, margine ferrugineo-pubescentibus ceteroquin glabris; carpellis anguste oblongis, glabris. A scandent shrub, glabrous except the axillary buds on the young branchlets, and parts of the inflorescences. Leaves firmly chartaceous, oblong-ovate to ovate-elliptic, 8 to 12 cm long, 3.5 to 6 cm wide, glabrous on both surfaces, grayish-olivaceous when dry, slightly shining, somewhat narrowed upward to the obtuse apex, base broadly rounded and usually slightly cordate; lateral nerves slender, not prominent, about 10 on each side of the mid- rib, anastomosing; petioles 5 mm long, black when dry, glabrous. Flowers dark reddish-brown, about 2 cm in diameter (somewhat immature) in short, usually 2-flowered, leaf-opposed inflores- cences, the peduncles 3 mm long or less, the pedicels 5 to 10 mm long, rugose when dry, the subtending bracts oblong, 3 to 4 mm long, the bracts and pedicels with few, scattered, ferruginous hairs, the latter soon becoming glabrous. Sepals broadly ovate to reniform, about 4 mm long, rounded or subacute, coriaceous, rugose when dry, their margins ferruginous-pubescent. Petals broadly ovate about 1 cm long, obtuse to subacute, densely fer- ruginous-pubescent on the margins and toward the tip, otherwise glabrous. Anthers numerous, narrowly oblong, 1.5 mm long, the connectives oblique, slightly produced. Carpels numerous, XIV, 4 Merrill: Noteworthy Philippine Plants, XV 395 narrowly oblong, glabrous, 1.5 mm long; ovules usually 4; styles 0.5 mm long. PANAyY, Antique Province, Batbatan Island, opposite Culasi, Bur. Sci. 32249 McGregor, June 29, 1918, fairly common in thickets at low altitudes. This species rather closely resembles Cuming 1607 from Mi- samis Province, Mindanao, which has been referred to Uvaria ovalifolia Blume but which apparently is not referable to Blume’s species. Among the other characters the present species differs from Uvaria ovalifolia in its somewhat smaller, fewer-nerved leaves; in being nearly glabrous; and in its entirely glabrous carpels. It is well characterized by its sepals and petals being nearly glabrous except for their densely ferruginous-pubescent margins and apices. LAURACEAE CRYPTOCARYA R. Brown CRYPTOCARYA RAMOSII sp. nov. Arbor parva, ramulis inflorescentiisque exceptis glabra; foliis oblongis ad oblongo-lanceolatis, subcoriaceis, 6 ad 11 cm longis, acuminatis, basi acutis, nervis utrinque 7 vel 8, perspicuis, re- ticulis ultimis confertis, utrinque obscure foveolatis; fructibus ellipsoideis, 1.5 cm longis, nigris, nitidis, glabris, laevis, 1.5 cm longis. A tree, 5 to 6 m high, glabrous, except the young branchlets and inflorescences which are ferruginous-pubescent. Branches reddish-brown, lenticellate. Leaves oblong to oblong-lanceolate, subcoriaceous, usually pale when dry, shining, 6 to 11 cm long, 1.5 to 4 cm wide, narrowed upward to the acuminate apex, base acute; lateral nerves usually 7 or 8 on each side of the midrib, rather prominent on the lower surface, pinnately arranged, curved, anastomosing, the ultimate reticulations very close, both surfaces very shallowly and minutely foveolate; petioles 5 to 8 mm long. Panicles in the uppermost axils, narrow, rather few-flowered, 6 to 8 cm long, the branches rather few, short, not exceeding 1.5 cm in length. Flowers subsessile, ferruginous- pubescent, about 4 mm long, the tube about 2 mm long. Pe- rianth-lobes elliptic-ovate, as long as the tube. Fruit ellipsoid, black and shining when dry, 1.5 cm long, glabrous, smooth. LuzON, Ilocos Norte Province, between Bangui and Claveria, Bur. Sci. 38071 Ramos, August 30, 1918, in forests at low altitudes. _ The description of the flowers is from Bur. Sci. 33346 Ramos, 396 The Philippine Journal of Science 1919 from Mount Palimlim, this specimen differing slightly from the type in its relatively somewhat shorter and slightly fewer-nerved leaves. The alliance of this species is manifestly with Crypto- carya vidalii Merr., from which it is easily distinguished by its much smaller fruits. CRYPTOCARYA EUPHLEBIA sp. nov. Arbor parva, ramulis et petiolis et inflorescentiis dense cas- taneo-pubescentibus; foliis oblongis ad oblongo-ellipticis, sub- coriaceis, usque ad 10 cm longis, costa excepta glabris, olivaceis, nitidis, apice breviter acuminatis, basi acutis ad subrotundatis, utrinque, praesertim subtus, minute et dense foveolatis; nervis utrinque 6 ad 8, supra impressis, subtus valde priminulis, sub- adscendentibus, vix anastomosantibus; paniculis axillaribus ter- minalibusque sub fructu usque ad 10 cm longis; fructibus in siccitate nigris, depresso-globosis, laevis, glabris, 10 ad 12 mm diametro. A small tree, the branchlets, petioles, and inflorescences dense- ly and finely castaneous-pubescent. Branches terete, glabrous, lenticellate, smooth, brownish. Leaves subcoriaceous, oblong to oblong-elliptic, 6 to 10 cm long, 2.5 to 5 em wide, olivaceous, shining, glabrous except for the somewhat pubescent midrib, the apex abruptly and distinctly acuminate, base acute to somewhat rounded, both surfaces, but especially the lower one, densely and shallowly foveolate; lateral nerves 6 to 8 on each side of the mid- rib, impressed on the upper surface, very prominent on the lower surface, somewhat ascending, slightly curved, not or but very obscurely anastomosing; petioles 8 to 10 mm long. Panicles axillary and terminal, in fruit up to 10 cm long. Fruits black, glabrous, smooth, shining, depressed-globose, 10 to 12 mm in diameter. CATANDUANES, Bur. Sci. 20576 Ramos, November 20, 1917, on the forested slopes of Mount Mariguidon. This species is well characterized by its shallowly and densely foveolate, glabrous, conspicuously nerved leaves and its de- pressed-globose fruits. In the latter character it differs espe- cially from Cryptocarya oblongata Merr. It seems to be most closely allied to Cryptocarya zamboanguensis Merr. and C. inter- media Elm. CRYPTOCARYA ELLIPTIFOLIA sp. nov. Arbor parva, inflorescentiis exceptis glabra; foliis ellipticis ad elliptico-ovatis, chartaceis, subolivaceis, nitidis, usque ad 20 cm longis, apice perspicue et obtuse acuminatis, basi leviter decur- rento-acuminatis, nervis utrinque 5, distantibus, distinctis, cur- . XIV, 4 Merrill: Noteworthy Philippine Plants, XV 897 vatis, vix elevatis; paniculis sub fructu axillaribus, circiter 7 cm longis; fructibus globosis, circiter 1.5 cm diametro, in siccitate nigris, nitidis, glabris, leviter longitudinaliter striatis. A small tree, 6 m high fide Ramos, entirely glabrous except the inflorescences, which are apparently somewhat pubescent. Branches and branchlets smooth, terete, slender, smooth, brown- ish-olivaceous. Leaves elliptic-ovate to elliptic, chartaceous, sub- olivaceous, shining, 15 to 20 cm long, 7 to 9 cm wide, not at all foveolate, the apex prominently and obtusely acuminate, the acumen about 1.5 cm long, stout, the base somewhat decurrent- acuminate; lateral nerves 5 on each side of the midrib, prom- inent but scarcely projecting, somewhat curved, reticulations slender, distinct, rather close; petioles 1 to 2 cm long. Panicles in fruit axillary, about 7 cm long, the branchlets sparingly pu- bescent. Fruits globose, black when dry, shining, faintly ridged longitudinally, glabrous, about 1.5 cm in diameter. CATANDUANES, Bur. Sci. 30330 Ramos, December 9, 1917, in forests along small streams back of Calolbong. A species well characterized by its rather large, conspicuously acuminate leaves; lateral panicles; and rather large, globose, somewhat ridged, glabrous fruits. CRYPTOCARYA DENSIFLORA Blume Bijdr. (1825) 556; Koord. & Val. Bijdr. Boom. Java 10 (1904) 213. PANAY, Jamindan and Mount Macosolon, Bur. Sci. .31317, 33435 Ramos & Edafio, April and May, 1918 (leaves not glaucous beneath). NerGrRos, For. Bur. 23666 Hinolan, March, 1915. MINDANAO, Lanao, Mrs. Clemens 1001, s. n., April and July, 1907. This species has previously been reported from Java, where it is common and widely distributed, and from the Malay Penin- sula. The Philippine material cited above closely matches a series of specimens from Java and the Malay Peninsula, and also conforms to the descriptions of Blume’s species. The speci- mens collected by Mrs. Clemens were described by me in 1908 as a new species of Beilschmiedia, but the description was for- tunately never published. Cryptocarya laevigata Elm., the type of which is a specimen with very immature fruits, is scarcely to be distinguished from Blume’s species. NEOLITSEA Merrill NEOLITSEA LANCEOLATA sp. nov. Arbor circiter 6 m alta, ramulis junioribus et petiolis dense adpresse pubescentibus, ramis glabris, teretibus, ramulis et foliis verticillatis; foliis lanceolatis, chartaceis ad subcoriaceis, in sic- citate brunneis, nitidis, usque ad 7 cm longis et 1.5 cm latis, basi 1658688 898 The Philippine Journal of Science 1919 acutis, apice tenuiter acuminatis, junioribus ad costa villosis, vetustioribus glaberrimis, penninerviis, nervis primariis utrin- que circiter 10; fructibus axillaribus, fasciculatis, parvis, gla- bris, pedicellis adpresse pallide villosis, circiter 5 mm longis. A tree, about 6 m high, the young branchlets and petioles densely pale or pale-ferruginous appressed-pubescent. Branches glabrous, terete, the branchlets verticillate, 3 to 6 at a node. Leaves verticillate, chartaceous to subcoriaceous, brownish when dry, lanceolate, 5 to 7 cm long, 1 to 1.5 em wide, narrowed below to the acute base and above to the slenderly acuminate apex, shining, smooth, the older ones entirely glabrous, the younger ones pale appressed-villous along the midrib on both surfaces; nerves all pinnately arranged, the primary ones about 10 on each side of the midrib, slender, curved, somewhat ascending, ob- scurely anastomosing, the reticulations not prominent, the ulti- mate ones close and very obscurely foveolate; petioles 5 to 8 mm long, rather densely pubescent. Fruits fascicled on the ultimate branches below the leaves, 3 to 6 in a fascicle, the pedicels rather stout, somewhat thickened upward, pale appressed-pubescent, the persistent calyx-tube somewhat funnel-shaped, truncate, glabrous, about 5 mm in diameter. CATANDUANES, Mount Mariguidon, Bur. Sci. 30231 Ramos, November 27, 1917, in forests near the summit of the mountain. This species is well characterized by its verticillate branchlets and its small, lanceolate, penninerved leaves. It does not appear to be very closely allied to any of the previously described species of the genus. NEOLITSEA PAUCINERVIA sp. nov. Species N. zeylanicae affinis. Arbor circiter 7 m alta, ramis et foliis glabris, ramulis dense pubescentibus; foliis alternis, oblongo-lanceolatis, subcoriaceis, in siccitate brunneis, nitidis, usque ad 8 cm longis et 3 cm latis, tenuiter acuminatis, basi acutis, 3-plinerviis, nervis lateralibus utrinque circiter 3, dis- tinctis, curvato-anastomosantibus, adscendentibus, obscure anas- tomosantibus; fructibus axillaribus, ovoideis vel subglobosis, in siccitate nigris, rugosis, nitidis, circiter 5 mm diametro, calycis lobis 4, lanceolatis, acuminatis, villosis, pedicellis nigris, deci- due villosis. A tree, about 7 m high, the branches smooth, terete, glabrous, the branchlets and petioles densely pubescent with short hairs. Leaves alternate, rather more numerous near the tips of the branchlets and here sometimes pseudo-verticillate, oblong-lanceo- late, subcoriaceous, brownish and shining when dry, both sur- Ae XIV, 4 Merrill: N oteworthy Philippine Plants, XV 899 . faces densely and shallowly foveolate, not at all glaucous, entirely glabrous, 7 to 8 cm long, 2 to 8 cm wide, narrowed below to the acute base and above to the slenderly acuminate apex, the base 3-plinerved, thé lateral basal nerves leaving the midrib about 5 mm above the base of the leaf and extending to or beyond the middle, the lateral ones above the base 2 or 3, slender, distinct, curved-ascending, obscurely anastomosing; petioles 5 to 8 mm long. Fruits axillary and in the axils of fallen leaves, fascicled, 5 to 12 in a fascicle, ovoid to subglobose, about 5 mm in diam- eter, black and shining when dry, somewhat wrinkled. Pedi- cels about 5 mm long, black, covered with a brownish, more or less deciduous, appressed, villous indumentum, the persistent calyx-lobes lanceolate, acuminate, villous, about 2 mm long. LUZON, Sorsogon Province, Mount Kililibong, Bur. Sci. 23316 Ramos, August 17, 1915, in the mossy forest. This species, among the Philippine forms, is probably as closely allied to Neolitsea villosa Merr. as any, but differs in its few- nerved, glabrous, not glaucous leaves. Its true alliance seems to be with N, zeylanica Merr., from which it is distinguished by its nonglaucous leaves and its villous pedicels. It also closely re- sembles some forms of the Javan N. cassiaefolia Merr. SAXIFRAGACEAE POLYOSMA Blume POLYOSMA VILLOSA sp. nov. Frutex, subtus foliis et ramis et inflorescentiis dense molliter villosus ; foliis verticillatis, 4-natis, oblongis ad oblongo-ellipticis, chartaceis ad subcoriaceis, 7 ad 10 cm longis, apice acutis, mar- gine distanter dentatis; nervis utrinque 12 ad 15, patulis, dis- tinctis; petiolo 1.5 ad 2 cm longo; racemis axillaribus, 5 cm longis, paucifloris; fructibus ellipsoideis, 8 mm longis, plus min- usve pubescentibus, bracteolis subaequalibus, circiter 1 mm longis. ; A shrub, the branches, branchlets, petioles, inflorescences, and the lower surface of the leaves, especially on the midrib and nerves, densely and softly villous. Leaves verticillate, usually 4 at each node, oblong to oblong-elliptic, chartaceous to subcoria- ceous, 7 to 10 cm long, 2.5 to 4 em wide, dark-olivaceous and shining when dry, the base rounded to acute,. apex acute, some- times with a very short mucro, the margins conspicuously villous, coarsely and distantly toothed, the upper surface more or less pubescent on the midrib and nerves; lateral nerves 12 to 15 on each side of the midrib, spreading at right angles, distinct, anas- 400 The Philippine Journal of Science 1919 tomosing, the reticulations lax. Petioles 1.5 to 2 cm long. Ra- cemes axillary, solitary, about 5 cm long, few-flowered, pubescent. Fruits ellipsoid, about 8 mm long, somewhat pubescent, their pedicels about,3 mm long, the subtending bracteoles subequal, oblong-ovate, pubescent, about 1 mm long. Persistent calyx lobes ovate, acute, 1 mm long. LuzoN, Ilocos Norte Province, Mount Palimlim, Bur. Sct. 33340 Ramos, August 21, 1918, on forested slopes, altitude about 900 meters. This species apparently belongs in the group with Polyosma verticillata Merr., from which it is specially distinguished by its dense indumentum and its lateral, not terminal, few-flowered racemes. POLYOSMA LONGIPETIOLATA sp. nov. Frutex, partibus junioribus inflorescentiisque exceptis glaber; foliis verticillatis, 4- vel 5-natis, oblongo-ellipticis ad anguste oblongo-obovatis, coriaceis, in siccitate nigris, 8 ad 13 cm longis, integris, basi acutis, apice breviter lateque acuminatis, nervis utrinque 8 ad 11, distinctis; petiolo usque ad 6 cm longo; racemis terminalibus, usque ad 20 cm longis; fructibus anguste ellipsoi- deis, 8 mm longis, bracteolis valde inaequalibus, interioribus quam lateralibus duplo longioribus, 3 ad 5 mm longis. A shrub, about 5 m high, glabrous except the very young parts and the inflorescences. Branches terete, dark reddish-brown, smooth, glabrous, the branchlets more or less angled, the younger ones appressed-pubescent with very short hairs. Leaves verti- cillate, 4 or*5 at each node, oblong-elliptic to narrowly oblong- obovate, coriaceous, glabrous, the upper surface nearly black when dry, the lower surface dark-brown, 8 to 13 cm long, 2.5 to 5 cm wide, entire, base acute, apex very shortly and broadly acuminate, the acumen obtuse and with a blunt apiculus; lateral nerves 8 to 11 on each side of the midrib, spreading, distinct, anastomosing. Petioles unusually long, up to 6 cm in length, when young pubescent, soon becoming glabrous. Racemes ter- minal, in fruit up to 20 cm long, more or less pubescent; pedicels 3 to 5 mm long. Fruits narrowly ellipsoid, about 8 mm long, slightly pubescent, the subtending bracteoles very unequal, pu- bescent, linear-oblong, the middle one 3 to 4 mm long, at least twice as long as the two lateral ones. Persistent calyx-lobes triangular-ovate, acute, 1 mm long. Luzon, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 33259 sent August 20, 1918, on forested slopes, altitude about 1,000 meters. XIV, 4 Merrill: Noteworthy Philippine Plants, XV 401 Among those species with verticillate leaves Polyosma longi- petiolata is strikingly characterized by its glabrous branches; unusually long-petioled, glabrous, entire leaves; and its very unequal bracteoles. PITTOSPORACEAE CITRIOBATUS A. Cunningham CITRIOBATUS JAVANICUS Boerl. & Koord. in Ic. Bogor. 1 (1901) t 77. LUZON, Batangas Province, Mount San Pedro, Bur. Sci. 22343 Ramos, August 20, 1914, in thickets at low altitudes. With the exception of this one species the small genus Citrio- batus is confined to Australia. The Philippine specimens were distributed unidentified, as without flowers I could not determine the group to which they belonged. I am now, however, able to name the material by a direct comparison with a series of five specimens from Java, including duplicates of three numbers of Koorders’s collection on which the species was based. The Philippine specimens agree with the Javan ones in all respects. PITTOSPORUM Banks PITTOSPORUM PSEUDOSTIPITATUM sp. nov. Frutex glaber, ramulis tenuibus; foliis verticillatis, 3- vel 4- natis, anguste lanceolatis, 4 ad 6 cm longis, utrinque angustatis, apice tenuiter acuminatis, nervis tenuibus; fructibus terminali- bus, fasciculatis, tenuiter pedunculatis, obovoideis, 2-valvis, basi tenuiter stipitatis, stipite circiter 4 mm longo, partibus semini- feris 6 ad 7 mm longis, 6 mm diametro. A glabrous shrub, about 3 m high, the branches and branch- lets slender, terete. Leaves verticillate, usually 3 or 4 at a node, coriaceous, narrowly lanceolate, 4 to 6 cm long, 6 to 13 mm wide, narrowed below to the acute or acuminate base and above to the very slenderly acuminate apex, the tip with a distinct cartilaginous mucro, the upper surface olivaceous to black when dry, shining; lateral nerves slender, scarcely more distinct than are the reticulations; petioles about 5mm long. Fruits fascicled at the tips of the branchlets, 2 to 4 in a fascicle, their pedicels slender, about 1 cm long, the capsule obovoid, black when dry, 2-valved, the lower part narrowed into a distinct pseudostalk, the thickened portion of the capsule 6 to 7 mm long, about 6 mm in diameter, the entire capsule, including the pseudostalk, about 1 cm long. Luzon, Ilocos Norte Province, Mount Nagapatan, Bur. Sci. 83127 Ramos, August 8, 1918, on dry slopes, altitude about 700 meters. * 402 The Philippine Journal of Science 1919 This species is readily distinguished by its narrowly lanceo- late, very sharply acuminate and cartilaginous-mucronate leaves as well as by its terminal, fascicled fruits, these being obovoid and much narrowed below the middle to form a very distinct pseudostalk. PITTOSPORUM ACUMINATISSIMUM sp. nov. Frutex epiphyticus vel pseudo-epiphyticus, glaber, ramulis te- nuibus; foliis chartaceis, oblongis ad oblongo-ellipticis, nitidis, tenuiter acute acuminatis, usque ad 8 cm longis, nervis utrinque 7 ad 9, tenuibus; inflorescentiis solitariis, lateralibus et termina- libus, brevibus, 3-floris; floribus circiter 12 mm _ longis, ovario dense villoso; fructibus leviter compressis, 2-valvis, ellipsoideis, circiter 23 cm longis, valvis crasse carnoso-coriaceis, in siccitate rugosis, An epiphytic or pseudoepiphytic shrub, entirely glabrous ex- cept the densely pubescent ovaries, the ultimate branches slender, 2 to 3 mm in diameter. Leaves chartaceous, oblong to oblong- elliptic, brownish-olivaceous when dry, 5 to 8 cm long, 2 to 4 cm wide, the apex sharply and slenderly acuminate, base acute; lateral nerves 7 to 9 on each side of the midrib, slender as are the reticulations; petioles 2 to 3 cm long, slender. Inflorescences usually solitary, lateral, rarely terminal, usually 3-flowered, their peduncles 5 mm long or less, the pedicels 5 to 10 mm long. Calyx shallowly cup-shaped, about 4 mm in diameter, the lobes broadly ovate, obtuse, 1.5 to 1.8 mm long. Petals 5, narrowly oblong, 11 mm long, 3.5 mm wide, obtuse. Filaments 4 to 4.5 mm long. Ovary oblong-ovoid, densely villous; style glabrous, 1.5 mm long. Capsules ellipsoid, somewhat compressed, 2-valved, about 2.3 cm long, 1.8 cm wide, the valves thick, HenES COINS rugose when dry. PaNAY, Capiz Province, Mount Macosolon, Bur. Sci. 30745 Ramos & Edafio, April 21, 1918, on trees in forests. This form has the same habit as Pittosporum resiniferum Hemsl. and is manifestly closely allied to that species. It differs in its smaller leaves and its usually solitary, lax, few-flowered inflorescences; in Hemsley’s species the flowers are borne in dense fascicles, while in the present species they are in depau- perate, 3-flowered, usually solitary cymes. PITTOSPORUM GLABERRIMUM sp. nov. Frutex glaberrimus, ramulis tenuibus; foliis verticillatis, 3- — vel 4-natis, oblongo-lanceolatis, utrinque angustatis, acutis vel leviter acuminatis, nervis utrinque 6 vel 7, tenuibus, distinctis; XIV, 4 Merrill: Noteworthy Philippine Plants, XV - 403 capsulis terminalibus, solitariis, ellipsoideis vel obovoideis, 2- valvis, 2.5 cm longis, pedicellis 1 cm longis. An entirely glabrous shrub, about 3 m high, the ultimate branchlets slender, terete, about 1.5 mm in diameter. Leaves verticillate, usually 3 or 4 at a node, oblong-lanceolate, shining subcoriaceous, those of each whorl more or less unequal, 3 to 6 cm long, 1 to 2 cm wide, subequally narrowed below to the acute base and above to the acute or slightly acuminate apex; lateral nerves slender, distinct, 6 or 7 on each side of the midrib, anastomosing, the reticulations lax; petioles 3 to 5 mm long. Capsules terminal, solitary, ellipsoid to somewhat obovoid, 2- valved, about 2.5 cm long, 1.5 to 2 cm in diameter, black when dry, orange-yellow when fresh, their pedicels about 1 cm long. Luzon, Ilocos Norte Province, Mount Palimlim, Bur. Sci. $3309 Ramos, August 21, 1918, on forested slopes, altitude about 1,000 meters. The alliance of this species is apparently with Pittosporum odoratum Merr., from which it is at once distinguishable by its solitary terminal fruits. ROSACEAE PHOTINIA Lindley PHOTINIA SERRULATA Lindl. in Trans. Linn. Soc. 13 (1821) 103 (excl. syn. Crataegus glabra Thunb.); Hemsl. in Journ. Linn. Soc. Bot. 33 (1887) 263; Rehder & Wilson in Sargent Pl. Wils..1 (1912) 184. _ Luzon, Benguet Subprovince, Pauai, Bur. Sci. 4425 Mearns, Bur. Sci. 31560 Santos, Sandkuhl 258, altitude about 2,300 meters, in flower May to July, in fruit in February, Igorot name itangan, This species has not previously been reported from the Phil- ippines. The specimens cited above conform very closely with our series from Formosa, and from Kwangtung, Fokien, and Yunnan Provinces, China. CONNARACEAE CONNARUS Linnaeus CONNARUS CASTANEUS sp. nov. - Frutex scandens, perspicue castaneo-pubescentibus; foliis cir- citer 30 cm longis, foliolis plerumque 7, coriaceis, oblongo- ellipticis, leviter acuminatis, 11 ad 18 cm longis, subtus densissime | ferrugineo-pubescentibus; infructescentiis paniculatis, termin- alibus, usque ad 25 cm longis, folliculis obovatis, compressis, leviter inaequilateralibus, 6 cm longis, extus densissime castaneo- pubescentibus, intus leviter villosis. A scandent shrub, the branchlets, petioles, infructescences, 404 The Philippine Journal of Science 1919 and fruits densely castaneous-pubescent, the leaflets beneath densely ferruginous-pubescent. Leaves about 30 cm long, the leaflets usually 7, alternate, coriaceous, oblong-elliptic, somewhat acuminate, base rounded, 11 to 18 cm long, 5 to 7 cm wide, the upper surface greenish, more or less pubescent with deciduous pale hairs; lateral nerves usually 7 on each side of the midrib, prominent on the lower surface, anastomosing, the reticulations rather lax; petiolules 5 to 8 mm long, densely castaneous-pubes- cent. Panicles terminal, in fruit up to 25 cm long, the primary branches up to 15 cm long. Fruits compressed, obovate, slightly inaequilateral, about 6 cm long and 4 cm wide, externally very densely castaneous-pubescent, inside sparingly villous with pale hairs. Luzon, Camarines Province, Paracale, Bur Sci. 33600 Ramos & Edafio, November 30, 1918, in damp forests at low altitudes, the fruits red, the lower surface of the fresh leaves red. A species strongly characterized by the dense castaneous indu- mentum on the branchlets, petioles, infructescences, and fruits; and by the leaflets being densely ferruginous-pubescent on the lower surface and reddish-brown in contrast with the green upper surface. The indumentum is composed entirely of simple hairs. ROUREA Aublet ROUREA LUZONIENSIS sp. nov. Frutex scandens, glaber; foliis circiter 10-foliolatis, 8 ad 12 em longis, foliolis 2 ad 4 em longis, subcoriaceis, ovatis ad ob- longo-ellipticis, obtusis vel obscure breviterque acuminatis, basi plerumque rotundatis; infructescentiis axillaribus, solitariis, ra- cemosis, 8 ad 10 cm longis, fructibus paucis, tenuiter pedicella- tis, subaequilateralibus, oblongo-ovoideis, 8 ad 10 mm longis. A scandent glabrous shrub, about 3 m high. Leaves 8 to 12 em long, about 10-foliolate; leaflets subopposite and alternate, subcoriaceous, ovate to oblong-elliptic, 2 to 4 cm long, 1.2 to 2.3 cm wide, base rounded, usually somewhat inequilateral, apex ob- tuse to shortly acuminate, both surfaces rather pale when dry, shining. Inflorescences axillary, solitary, racemose, 8 to 10 cm long, few-flowered. Fruits oblong-ovoid, nearly equilateral, acute or slightly apiculate, 8 to 10 mm long, their pedicels 6 to 10 mm long. Persistent sepals broadly ovate, glabrous, the outer ones acute or slightly acuminate, the inner ones broadly rounded, about 3.5 mm long. Luzon, Camarines Province, Paracale, Bur. Sci. 33534 Ramos & Edano, November 29, 1918, in dry thickets at low altitudes. XIV, 4 Merrill: N oteworthy Philippine Plants, XV 405 This species is allied to Rowrea erecta (Blanco) Merr., from which it is especially distinguished by its fewer, differently shaped, not slenderly acuminate leaflets; and its smaller, nearly- equilateral, slenderly pedicelled fruits which are racemosely arranged. LEGUMINOSAE TRIFIDACANTHUS Merrill TRIFIDACANTHUS UNIFOLIOLATUS Merr. in Philip. Journ. Sci. 12 (1917) Bot. 269. Pods straight, about 2 cm long, usually composed of three or four 1-seeded joints, flat, thin, reticulate, sparingly pubescent, the upper suture nearly straight, slightly indented between the joints, the lower suture rather deeply indented; joints usually 6 to 7 mm long, about 4 mm wide, the valves chartaceous or thinly coriaceous. Luzon, Ilocos Norte Province, Burgos, Bur. Sci. 32925 Ramos, July 18, 1918, in dry thickets at low altitudes. The original description of this new genus and species was based on flowering specimens, but the second collection, and this from the type locality, shows that I was correct in placing it in the Papilionatae-Hedysareae-Desmodiinae, the Desmodiinae being indicated in the original description with doubt. It is manifestly most closely allied to Desmodium, from which, among other characters, it is distinguished by its characteristic spines. INDIGOFERA Linnaeus INDIGOFERA HENDECAPHYLLA Jacq. Coll. 2 (1788) 359, Ic. Rar. 3 (1768-93) t. 570; Baker in Hgok. f. Fl. Brit. Ind. 2 (1876) 98 (endecaphylla) ; Craib in Notes Bot. Gard. Edinb. 8 (1918) 68. Indigofera pusilla Lam. Encycl. 3 (1789) 248. Luzon, Ilocos Norte Province, Burgos, Bur. Sci. 32770 Ramos, August 5, 1918, in open dry places at low altitudes. India to southeastern China, tropical and South Africa, and Madagascar; not previously reported from the Philippines. ORMOSIA Jackson ORMOSIA ORBICULATA sp. nov. § Toulichiba, Macrodisca. Arbor, ramulis inflorescentiisque ferrugineo-pubescentibus ex- ceptis glabra; foliis usque ad 30 cm longis, rhachibus productis ; foliolis 7, subcoriaceis, glabris, oblongis ad oblongo-ellipticis, 8 ad 15 cm longis, apice acuminatis, nervis utrinque circiter 8; pani- culis circiter 16 cm longis; floribus numerosis, circiter 2 cm 406 The Philippine Journal of Science 1919 ‘longis, calycis circiter 8 cm longis; ovulis 2; leguminis suborbi- cularibus ad ovatis, usque ad 7 cm longis, basi late rotundatis, apice apiculatis, valvis crassis, lignosis; seminibus solitariis, ellipsoideis, exarillatis, 2 cm longis. A tree, attaining a height of at least 10 m, glabrous except the young branchlets, buds, and inflorescences, which are rather densely ferruginous-pubescent. Leaves up to 30 cm long, the rachis extended beyond the distal pair of leaflets; leaflets 7, subcoriaceous, oblong to oblong-elliptic, pale when dry, somewhat shining, 8 to 15 cm long, 3.5 to 7 cm wide, base acute to rounded, apex distinctly acuminate, the acumen usually blunt; lateral nerves about 8 on each side of the midrib, distinct on both surfaces as are the reticulations; petiolules about 5 mm long. Panicles about 16 cm long, densely ferruginous-pubescent, term- inal, the primary branches few, the lower ones up to 6 cm long. Flowers white, about 2 cm long. Calyx externally densely fer- ruginous-pubescent, about 1.5 cm long, the tube broad, up to 8 mm long and 1 cm wide, the lobes oblong, usually obtuse, about 8 mm long, slightly pubescent inside. Standard broadly obovate, about 2 cm long, the claw stout, 4 to 5 mm long. Filaments glabrous. Ovary stipitate, oblong, very slightly pubescent with scattered hairs, soon becoming glabrous, the stipe stout, about 5 mm long; style glabrous, curved, about 1.3 cm long; ovules 2. Pod thick, woody, suborbicular to ovate, up to 7 cm long and 6 cm wide, base broadly rounded, apex prominently apiculate, the valves up to 1 cm in thickness. Seed solitary, ellipsoid, not arillate, red, shining, about 2 cm long. Luzon, Ilocos Norte Province, Burgos, Bur. Sci. 32775 Ramos (type), July 17, 1918; Pasuquin, For. Bur. 22094, 25094 Paraiso, July 24, 1914, and February 18, 1916, on forested slopes, ascend- ing to an altitude of 300 meters. Ilocano name panapotien. This species closely resembles Ormosia macrodisca Baker, of the Malay Peninsula, but the flowers are much larger and the single specimen that I have with seeds shows no basal aril. From Ormosia paniculata Merr. it is at once distinguished by its en- tirely glabrous leaves. ORMOSIA CLEMENTIS sp. nov. § Toulichiba, Macrodisca. Arbor circiter 10 m alta, glabra (inflorescentiis ignotis) ; foliis circiter 40 em longis, foliolis 5, rhachibus productis, folio- lis chartaceis ellipticis ad oblongo-ellipticis, usque ad 20 cm longis, abrupte acute acuminatis, basi acutis, nervis utrinque 8 ad 10, distinctis; leguminis ligneis, subellipticis, circiter 7 cm longis, apice rotundatis vel obscurissime et late apiculatis vel XIV, 4 Merrill: Noteworthy Philippine Plants, XV 407 acutis; seminibus plerumque 2, exarillatis, coccineis, basi nigris, subellipticis, circiter 2 cm longis, 1.5 ad 2 cm latis, nitidis. A tree about 10 m high, glabrous (inflorescences unknown). Leaves about 40 cm long, the rachis produced above the upper pair; leaflets chartaceous, rather pale when dry, elliptic to oblong-elliptic, 9 to 20 cm long, 3.5 to 8.5 cm wide, the apex rather abruptly and sharply acuminate, base acute; lateral nerves 8 to 10 on each side of the midrib, distinct, anastomosing. Pods woody, subelliptic, about 7 em long, brown when dry, the valves recurved after dehiscence, base and apex rounded or the apex obscurely and broadly apiculate or merely subacute. Seeds usually 2, not arillate, crimson, with a black base, subellipsoid, about 2 cm long, 1.5 to 2 em wide, hard, shining, slightly rugose when dry. MINDANAO, Lanao District, Camp Keithley, Mrs. Clemens 1139, July, 1907, altitude about 900 meters. This species is probably most closely allied to Ormosia orbi- culata Merr., but the pods are usually 2-seeded and the seeds have a conspicuous black base. ORMOSIA BASILANENSIS sp. nov. § Toulichiba, Macrodisca. Arbor magna, partibus junioribus et inflorescentiis plus mi- nusve ferrugineo-pubescentibus, ramis circiter 1 cm diametro; foliis circiter 40 cm longis, rhachibus et petiolis et subtus foliolis ad costa ferrugineo-pubescentibus, rhachibus productis; foliolis 9, oblongis ad oblongo-ellipticis, subcoriaceis, 12 ad 17 cm longis, basi acutis, apice acuminatis, nervis utrinque circiter 8; legu- minis orbiculari-ovatis, leviter inaequilateralibus, circiter 7 cm longis, valvis crassis, lignosis, apice breviter crasseque acumi- natis; seminibus solitariis, 8 cm longis, ellipsoideis, rubris, nitidis, exarillatis. A large tree, the younger parts, inflorescences, and leaves more or less ferruginous-pubescent, the leaflets only sparingly pubescent beneath along the midrib, the branches rather stout, about 1 em in diameter, dark colored when dry, glabrous. Leaves about 40 cm long, the rachis and petioles more or less ferruginous-pubescent with rather long hairs, the rachis prolonged beyond the distal pair of leaflets; leaflets 9, oblong to oblongo-elliptic, subcoriaceous, shining, 12 to 17 cm long, 4.5 to 7 em wide, subequally narrowed to the acute base and the dis- tinctly acuminate apex; lateral nerves about 8 on each side of the midrib, distinct, ascending; petiolules 5 to 7mm long. Pods orbicular-ovate, slightly inequilateral, about 7 cm long and wide, base rounded, apex with a short stout acumen, the valves woody, 408 The Philippine Journal of Science 1919 much thickened, much swollen opposite the solitary seed. Seed about 3 cm long, ellipsoid, red, shining, not arillate. BASILAN, Maluno, For. Bur. 17893 Rafael, November 18, 1912, Hollier 508, January, 1904, in forests, altitude about 140 meters. This species like the others here described resembles Ormosia macrodisca Baker, but the seeds are not arillate. Among the Philippine species it is perhaps most closely allied to Ormosia orbiculata, but the leaflets are more numerous, subequally nar- rowed at both ends, acute at the base, and sparingly pubescent beneath along the midrib. ORMOSIA GRANDIFOLIA sp. nov. § Toulichiba, Macrodisca. Arbor circiter 10 m alta, inflorescentiis foliisque plus minusve pubescentibus; foliis circiter 50 cm longis, rhachibus productis; foliolis 7, terminalibus late obovatis, lateralibus ellipticis, 13 ad 17 cm longis, subcoriaceis, apice abrupte obtuseque acuminatis, basi rotundatis, vel terminalibus acutis, nervis utrinque circiter 8; leguminis ellipticis, 6 ad 7 cm longis, apice late rotundatis, valvis crassis, lignosis; seminibus solitariis ellipsoideis, cocci- neis, 3 cm longis, exarillatis. A tree, about 10 m high, the inflorescences, petioles, and leaf- lets on the lower surface more or less pubescent, the indumentum on the leaflets obscure. Leaves about 50 cm long, the rachis extended beyond the distal pair of leaflets; leaflets 7, the terminal one broadly obovate, the lateral ones elliptic, 13 to 17 cm long, 8 to 10 cm wide, subcoriaceous, rather pale when dry, shining, the base usually rounded or that of the terminal leaflet acute, apex very abruptly and shortly acuminate; the lateral nerves about 8 on each side of the midrib, prominent; petiolules about 1 cm long. Pods elliptic, 6 to 7 cm long, about 5.5 cm wide, apex broadly rounded, scarcely apiculate, the valves thick, woody, recurved, apparently one-seeded. Seed solitary, ellipsoid, scarlet, shining, somewhat rugose when dry, 3 cm long, 2.2 cm wide, slightly compressed, not arillate. Luzon, Camarines Province, Magdoronganon River, For. Bur. 21172 Alvarez, March 23, 1914, in forests near the river, alti- tude about 100 meters. This species manifestly belongs in the same group as Ormosia orbiculata, but has much larger, differently shaped leaflets, which are somewhat pubescent beneath; differently shaped pods, which are broadly rounded and scarcely apiculate at their apices; and much larger seeds. The seed is not at all arillate. Ormosia basilanensis Merr. differs in its more numerous leaflets and its elliptic pods, which are broadly rounded at their apices. XIV, 4 Merrill: Noteworthy Philippine Plants, XV 409 DESMODIUM Desvaux DESMODIUM ORMOCARPOIDES DC. Prodr. 2 (1825) 327; Gagnep. in Not. Syst. 3 (1916) 256; Merr. Interpret. Herb. Amb. (1917) 267, non auct. plur. Desmodium ormocarpoides Desv. in Mém. Soc. Linn. Paris 1825 (1826) 307, non auct. plur. Hedysarum adhaerens Poir. in Lam. Encycl. Suppl. 5 (1817) 15, non Vahl. Desmodium dependens Blume ex Mig. Fl. Ind. Bat. 1* (1855) 248. JOLO, Tagasip, F. L. Rola 118, May 27, 1917, in teak forests at low altitudes. This Malayan species is new to the Philippines, the specimens agreeing perfectly with material from Celebes and Amboina. As Gagnepain notes all modern botanists have misinterpreted Desmodium ormocarpoides DC., an examination of the type showing that it is the form described by Blume as Desmodium dependens. Desmodium ormocarpoides of modern authors, in- cluding the Philippine specimens previously so named, is Des- modium zonatum Miq. GERANIACEAE ERODIUM L’Héritier ERODIUM CICUTARIUM (Linn.) L’Hérit. ex Ait. Hort. Kew. 2 (1789) 414, Geranium cicutarium Linn. Sp. Pl. (1753) 680. Luzon, Benguet Subprovince, Pauai, Bur. Sci. 31669 Santos, May 9, 1918, in waste places, altitude about 2,300 meters. A weed of wide distribution in temperate regions, unquestionably an introduced plant here. RUTACEAE EVODIA Forster EVODIA MELIAEFOLIA (Hance) Benth. Fl. Hongk. (1861) 58. Megabotrya meliaefolia Hance in Walp. Ann. 2 (1851-52) 259. Boymia glabrifolia Champ. in Hook. Kew. Journ. Bot. 3 (1851) 330. Eurycoma dubia Elm. Leafl. Philip. Bot. 2 (1908) 481. LUZON, Benguet Subprovince, Pauai, Bur. Sci, 31968 Santos, June, 1918, altitude 1,800 meters, with the local name galiuy- guiuan. NeGRos, Cuernos Mountains, Elmer 10120 (type of Eurycoma dubia Elm.). P The second collection of this species in the Philipines induced me to make a critical study of it, as it was perfectly evident that the form described by Mr. Elmer as Eurycoma dubia represented no simarubaceous plant. I cannot distinguish the Philippine form from Evodia meliaefolia Benth., which was previously known from southern Japan, the Liu Kiu Islands, 410 The Philippine Journal of Science 1919 Formosa, and southern China. It is the only pinnate-leaved Evodia known from the Philippines. EVODIA SESSILIFOLIOLA sp. nov. Arbor usque ad 10 m alta, partibus junioribus inflorescentiis- que plus minusve villosis; foliis plerumque 3-foliolatis, interdum 1-foliolatis, foliolis chartaceis ad subcoriaceis, sessilibus, oblongo- obovatis, usque ad 20 cm longis, basi cuneatis, apice obtusis, subtus punctatis, nervis utrinque circiter 13, subtus perspicuis; inflorescentiis axillaribus, pedunculatis, usque ad 10 cm longis; floribus 4-meris, petalis circiter 2.5 mm longis. A tree, 8 to 10 m high, the younger parts and the inflores- cences more or less villous, the indumentum grayish, the older parts soon glabrous. Leaves mostly 3-foliolate, sometimes 1-foliolate ones on the same branchlets with the 3-foliolate ones, their petioles 5 to 7 cm long; leaflets sessile, chartaceous to sub- coriaceous, brownish or rather pale when dry, somewhat shining, obovate to oblong-obovate, 9 to 20 cm long, 4 to 10 cm wide, the mature ones glabrous, beneath distinctly punctate, apex obtuse to rounded, base of the terminal leaflet usually cuneate, equi- lateral, of the lateral ones usually obtuse and distinctly inequi- lateral; nerves about 13 on each side of the midrib, prominent on the lower surface, anastomosing, the reticulations prominent. Inflorescences axillary, peduncled, up to 10 cm long, paniculate, rather many flowered. Flowers white, 4-merous, their pedicels up to 2 mm in length, densely villous. Sepals oblong, 0.5 mm long. Petals ovate, about 2.5 mm long. Disk densely villous. Fruit of 3 or 4 cocci, glabrous, the individual cocci about 3 mm in diameter. Luzon, Benguet Subprovince, Pauai, Bur. Sci. 31788 Santos (type), April 22, 1913, altitude about 2,400 meters, with the Igorot name itapan, Mrs. Clemens s. n., January, 1915; Baguio, Williams 1531, November 30, 1904, Sandkuhl 86, July, 1913. Among the Philippine species the alliance of this form is with Evodia semecarpifolia Merr., and among the extra-Philip- pine forms with E. lunur-ankenda (Gaertn.) Merr. It is distin- guished from both by its sessile leaflets. The combination of 8-foliolate and 1-foliolate leaves on the same branchlets is rather unusual in the genus. MELIACEAE AGLAIA Loureiro AGLAIA DIFFUSIFLORA sp. nov. § EHuaglaia. Arbor parva, plus minusve brunneo- vel castaneo-lepidotis ; foliis alternis, circiter 60 cm longis, 7-foliolatis, foliolis oblongo- XIV, 4 Merrill: Noteworthy Philippine Plants, XV 4j1 ellipticis, circiter 25 cm longis, chartaceis, obtusis vel obtuse acuminatis, nervis utrinque circiter 16; paniculis diffusis, laxis, foliis subaequantibus, multifloris, ramis inferioribus usque ad 30 cm longis; floribus racemose dispositis, 5-meris, calycis lobis orbicularibus, petalis obovato-ellipticis, 1.4 mm longis, tubo sta- mineo libero. A small tree, the branchlets and inflorescences densely covered with closely appressed brown or castaneous scales with similar scattered ones on the lower surface of the leaflets. Leaves al- ternate, about 60 cm long, 7-foliolate; leaflets alternate or the upper pair opposite, chartaceous, rather pale when dry, oblong- elliptic, somewhat inequilateral, about 25 cm long, 9 to 11 cm wide, the apex obtuse or somewhat obtusely acuminate, base usually rounded on one side and acute on the other; lateral nerves about 16 on each side of the midrib, prominent; petiolules 1.5 to 2.5 cm long. Panicles diffuse, about 50 cm long, the branches spreading, the lower ones up to 30 cm in length, many flowered, the flowers laxly and racemosely arranged on the ulti- mate branchlets, their pedicels 1 to 2 mm long. Flowers white, 5-merous. Calyx-lobes orbicular, rounded, 0.5 mm in diameter. Petals elliptic-obovate to broadly elliptic, 1.4 mm long. Sta- minal-tube turbinate, glabrous, about 1 mm long, crenulate, the crenulations minutely puberulent, free, glabrous. Anthers 5, inserted below the rim of the tube, about 0.5 mm long, their tips minutely exserted. - PANAY, Capiz Province, Jamindan, Bur. Sci. 31098 Ramos & Edano, May 16, 1918, in damp forests. This species is allied to Aglaia pallida Merr., from which it is distinguished by its larger leaflets, the lepidote indumentum not being at all stellate, and by its very much longer petiolules. ANACARDIACEAE SEMECARPUS Linnaeus f. SEMECARPUS SUBSESSILIFOLIA sp. nov. Species S. gigantifoliae affinis, differt inflorescentiis terminal- ibus, inflorescentiis hypocarpiisque cinereo-pubescens. Frutex vel arbor parva, simplex; foliis oblongo-oblanceolatis, usque ad 50 cm longis et 16 latis, acuminatis, deorsum longe angustatis, basi rotundatis vel obtusis, nervis utrinque 30 ad 35, perspicuis; paniculis circiter 20 cm longis; fructibus oblongis, circiter 1 cm longis. A shrub or a small unbranched tree, glabrous. except the in- florescences. Leaves oblong-oblanceolate, about 50 em long, 12 to 16 cm wide, firmly chartaceous, shining, the upper surface 412 The Philippine Journal of Science : 1919 olivaceous, the lower pale when dry, apex shortly acumin- ate, gradually narrowed in the lower two-thirds to three-fourths to the narrow and abruptly rounded or obtuse base, which is at most 2.5 em wide; midrib very stout, the primary lateral nerves 30 to 35 on each side of the midrib, somewhat curved, anas- tomosing, prominent, the reticulations rather lax, distinct, the secondary nervules leaving the nerves at about right angles; petioles very stout, about 1 cm long or less, panicles terminal, about 20 cm long, more or less pubescent, the primary branches few, up to8cm long. Immature fruits oblong, about 1 cm long, glabrous; hypocarp somewhat turbinate, about 4 mm long, black when dry, cinereous-pubescent. PANAY, Capiz Province, Mount Madiaas, Bur. Sci. 30711 Ramos & Edaiio, May 12, 1918, in forests. This species is manifestly allied to Semecarpus gigantifolia F.-Vill. and like that species is unbranched. It differs essentially in its terminal, not cauline, inflorescences which are more or less pubescent and in its distinctly pubescent hypocarps. SEMECARPUS FERRUGINEA sp. nov. Arbor parva, ramulis inflorescentiisque ferrugineo-pubes- cens; foliis oblanceolatis ad oblongo-oblanceolatis, coriaceis, nitidis, acuminatis, basi angustatis, subtus glaucis, nervis utringue 12 ad 14, perspicuis; paniculis terminalibus, usque ad 28 cm longis; fructibus oblique ovoideis, glabris, 1.2 cm longis, hypocarpio ferrugineo-pubescens. A small tree, the branchlets, inflorescences, and hypocarps more or less ferruginous-pubescent. Leaves oblanceolate to oblong-oblanceolate, coriaceous, 12 to 16 cm long, 3 to 4.5 cm wide, apex slightly acuminate, somewhat narrowed below to the cuneate or obtuse base, the upper surface brownish-olivaceous, shining, glabrous except the pubescent midrib, the lower surface glaucous, slightly pubescent along the midrib and nerves; lateral nerves 12 to 14 on each side of the midrib, curved, anastomosing, the reticulations rather lax, distinct, the primary nervules leav- ing the nerves at nearly right angles; petioles stout, 7 to 10 mm long, sulcate on the upper side, pubescent. Panicles terminal, in fruit up to 28 cm long, the rachis and branches densely ferrug- inous-pubescent, the primary branches few, scattered, the longer ones about 6 cm in length. Fruits obliquely ovoid, glabrous, black and somewhat shining when dry, about 1.2 cm long, the very young ones with a few ferruginous hairs; hypocarps 5 to 7 mm long, ferruginous-pubescent, the indumentum somewhat deciduous; pedicels 5 to 8 mm long. XIV, 4 Merrill: Noteworthy Philippine Plants, XV 413 PANAY, Capiz Province, Mount Macosolon, Bur. Sci. 30756 Ramos & Edafo, April 22, 1918, on forested slopes. This species resembles Semecarpus philippinensis Engl., to which it is manifestly allied, but from which it is easily distin- guished by its ferruginous-pubescent panicles and hypocarps. ONCOCARPUS A. Gray ONCOCARPUS OBOVATA (Elm.) comb. nov. Dichapetalum obovatum Elm. Leafi. Philip. Bot. 2 (1908) 483. Luzon, Tayabas Province, Mount Banahao, Elmer 7931 (type), May, 1907, in fruit; Quisumbing 1346, May, 1918, with staminate flowers. While it has long been realized by me that the form described by Mr. Elmer could be no dichapetalous plant, no study of it was previously made on account of the incomplete material rep- resenting it. Staminate material from near the type locality shows that it is an anacardiaceous plant and is referable to Onco- carpus. It differs radically from the other Philippine species of this small genus in its very short inflorescences, the staminate ones being reduced to few-flowered, axillary fascicles; the stam- inate flowers are moreover 5- to 6-merous, with 5 calyx-lobes and 6 petals and stamens. The petals are oblong-lanceolate, re- ticulate, villous externally, about 3 mm long, 1 to 1.2 mm wide, obtuse or acute; the filaments are unequal in length and glabrous; the rudimentary ovary densely villous. In the type collection, which was distributed as Pterospermum, the single fruit I have seen is 1-celled and 1-seeded, although it is described as 2-celled with one seed in each cell. The pedicel is only slightly thickened in fruit. The alliance of this species is manifestly with Onco- carpus densiflora Merr. ore PARISHIA Hooker f. PARISHIA OBLONGIFOLIA sp. nov. Arbor, inflorescentiis ferrugineo-tomentosis exceptis glabra; foliis usque ad 40 cm longis, foliolis 15 ad 17, coriaceis, aequi- lateralibus, oblongis ad oblongo-lanceolatis, acuminatis, usque ad 18 cm longis, nitidis, nervis utrinque circiter 20; infructescentiis quam foliis paullo longioribus, calycis accrescentibus usque ad 10 em longis, ferrugineo-tomentosis, lobis anguste oblongis, obtusis, usque ad 9 cm longis et 1.5 cm latis, tubo inflato, in siccitate brunneis, fructibus densissime villosis. A tree, with ultimate branches thickened, rugose, dark-brown, about 1.5 cm in diameter. Leaves up to 40 cm long; leaflets 15 to 17, coriaceous, oblong to oblong-lanceolate, 11 to 18 cm long, 165368——4 414 The Philippine Journal of Science 1919 8 to 6 cm wide, acuminate, base acute, shining, brown when dry, the lower surface paler than the other; lateral nerves about 20 on each side of the midrib, distinct; petiolules 7 to 10 mm long. Panicles equal to or longer than the leaves, ferruginous- tomentose. Accrescent calyx up to 10 cm in length, the lobes 4, ‘narrowly oblong, up to 9 cm long, about 1.5 cm wide, obtuse, ferruginous-tomentose; at first membranaceous becoming coria- ceous, the inflated base brown when fresh, the lobes more or less pink toward their tips, when dry uniformly brown. Young fruits ovoid, very densely ferruginous-villous with spreading hairs about 2 mm long. PANAY, Capiz Province, Jamindan, Bur. Sci. 31044 Ramos & Edano, May, 1918 (type); Antique Province, Culasi, Bur. Sci. 82219 McGregor, May 24, 1918, on forested hillsides, altitude 900 meters. The same species is represented by a sterile specimen from Sibuyan Island, For. Bur. 27040 Sajor, November 7, 1917, with the Visayan name bulalog. This species is the second one of the genus to be found in the Philippines and is readily distinguished from Parishia malabog Merr. by its narrower, differentially shaped, equilateral, more numerously nerved leaflets and by its ferruginous-tomentose in- florescences and accrescent, much larger calyces, the lobes of which are much wider than are those of Parishia malabog. ICACINACEAE VILLARESIA Ruiz and Pavon VILLARESIA PHILIPPINENSIS sp. nov. Arbor parva, inflorescentiis exceptis glabra; foliis crasse co- riaceis, rigidis, ellipticis ad elliptico-oblongis, olivaceis, nitidis, usque ad 11 cm longis, integris, acuminatis, basi plerumque le- viter inaequilateralibus, acutis, nervis utrinque circiter 4, per- spicuis, petiolo 2 ad 3 em longo; inflorescentiis terminalibus, 5 ad 8 cm longis, solitariis vel binis, ramis brevissimis; floribus circiter 6 mm longis, confertis; fructibus ovoideis, 2.5 ad 3 cm longis, inaequilateralibus. A small tree, 4 to 5 m high, entirely glabrous except the in- florescences. Leaves thickly coriaceous, rigid, elliptic to elliptic- oblong, olivaceous and shining on both surfaces when dry, 6 to 11 cm long, 3 to 6.5 cm wide, entire, the apex rather prominently acuminate, base usually slightly inequilateral, acute; lateral nerves usually 4 on each side of the midrib, curved-ascending, arched-anastomosing, prominent on the lower surface, the reti- culations lax, distinct; petioles black when dry, 2 to 3 cm long. Inflorescences terminal, 5 to 8 cm long, solitary or sometimes XIV, 4 Merrill: Noteworthy Philippine Plants, XV 415 in pairs, narrow, more or less hirsute with short, somewhat ferruginous hairs, the branches numerous, rather crowded, race- mosely arranged, 1 cm long or less, each bearing at the apex 5 to 8, densely crowded, cymosely arranged, sessile flowers. Sepals 5, subelliptic, rounded, imbricate, 1.5 to 2 mm long, sparingly hirsute. Petals 5, narrowly oblong, white, glabrous, 5 to 6 mm long, the tip inflexed, the margins slightly imbricate throughout, prominently keeled inside along the median portion. Filaments free, 5 mm long; anthers broadly ovoid, 1 mm long. Ovary gla- brous, ovoid, 1-celled; ovules 2, pendulous; style glabrous, about 3 mm long; stigma minute, obscurely 2-lobed. Fruit ovoid, glabrous, shining, 2.5 to 3 cm long, distinctly inequilateral, black, the pericarp 1 to 1.5 mm thick. Seed deeply sulcate. Luzon, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 32267 (type), 33308 Ramos, August 20, 1918, on forested slopes, al- titude about 1,000 meters. This species is apparently allied to Villaresia suaveolens (Blume) Val., the type of the genus Pleuropetalon Blume and of the genus Chariessa Miq., from, which it is perhaps best dis- tinguished by its long petioles; I have seen no specimen of Blume’s species and all of the published descriptions are in- complete. VILLARESIA LATIFOLIA sp. nov. Arbor alta, inflorescentiis exceptis glabra; foliis crasse coria- ceis, nitidis, obovatis ad oblong-obovatis, usque ad 18 cm longis et 11 cm latis, integris, abrupte et brevissime acuminatis, basi acutis ad decurrento-acuminatis, nervis utrinque 5 vel 6, petiolo 1 ad 1.5 em longo; infructescentiis anguste pyramidatis, 10 ad 15 em longis, ramis inferioribus usque ad 5 cm longis; fructibus ovoideis, aequilateralibus, 2 cm longis. A large tree, glabrous except the somewhat hirsute inflores- cences. Leaves thickly coriaceous, shining, obovate to oblong- obovate, 12 to 18 cm long, 6 to 11 cm wide, entire, the apex broad, abruptly and shortly acuminate, base acute to decurrent- acuminate, equilateral; lateral nerves 5 or 6 on each side of the midrib, curved-ascending, anastomosing, the reticulations lax; petioles 1 to 1.5 cm long. Infructescences terminal, solitary, narrowly pyramidal, 10 to 15 cm long, tripinnately paniculate, the primary branches racemosely arranged, numerous, rather crowded, the lower ones up to 5 cm in length, the uppermost ones simple, about 1 em long. Fruits (not quite mature) ovoid, equi- ’ Jateral or nearly so, brown and shining when dry, about 2 cm Jong. Seed deeply longitudinally sulcate. 416 The Philippine Journal of Science 1919 SAMAR, Catubig River at Pinipisakan, Bur. Sci. 24557 Ramos, March 21, 1916, in damp forests near the river at low altitudes, with the Visayan name malaampipi, distributed as an uniden- tified representative of the Menispermaceae. The type of the genus Villaresia is a tropical American plant, and after a critical study of the two Philippine forms I am in entire agreement with Bentham and with Valeton in reducing to it the genus Chariessa Miquel, which was based on Pleuwrope- talon Blume (non Plewropetalum Hook.), the type of which was a Javan specimen. Engler maintains Chariessa as a genus dis- tinct from Villaresia chiefly on the basis that the petals of the latter are imbricate and those of the latter are valvate. Blume described the petals of the Javan form as imbricate, while in Villaresia philippinensis the petals are distinctly although not prominently imbricate throughout. There appears to be no other character by which the two genera can be distinguished. The genus Villaresia has about ten known species in tropical America, one in tropical Africa, one in Java, two in Australia, one in Polynesia, and two in the Philippines. MIQUELIA Meissner MIQUELIA PHILIPPINENSIS sp. nov. Frutex scandens, ramulis junioribus hirsutis; foliis membra- naceis, ovatis, 7 ad 11 cm longis, basi late rotundatis et plus minusve cordatis, apice acuminatis, margine integris vel repan- dis; inflorescentiis ¢ axillaribus, pedunculatis, floribus umbella- tim dispositis; fructibus immaturis oblongo-ellipsoideis, 1 ad 1.3 cm longis, longe stipitatis, A scandent shrub, the young branchlets hirsute. Leaves mem- branaceous, ovate, olivaceous, 7 to 11 cm long, 5 to 9 cm wide, base broadly rounded and usually shallowly cordate, apex some- what acuminate, margins entire or somewhat repand; lateral nerves 5 or 6 on each side of the midrib, slender, distinct, the basal ones scarcely distinguishable from the others, reticulations lax; petioles pubescent when young, soon becoming glabrous, 3 to 7 cm long. Pistillate inflorescences axillary, their peduncles about 3 cm long, each bearing usually about 7 umbellately ar- ranged flowers, their pedicels 5 mm long, each subtended by a lanceolate, acuminate, sparingly hirsute bracteole, 1 to 1.5 mm long. Sepals oblong, obtuse, 2.5 mm long. Fruits (immature) slightly pubescent, oblong-ellipsoid, 1 to 1.3 cm long, narrowed below into a pseudostalk as long as the fruit proper. LuZON, Ilocos Norte Province, Burgos, Bur. Sci. 32885 Ramos, August 6, 1918, in rocky forests at low altitudes. XIV, 4 Merrill: Noteworthy Philippine Plants, XV 417 This is the second species of the genus to be found in the Philippines and differs from Miquelia cumingii notably in its umbellately arranged, not capitate, pistillate flowers. SAPINDACEAE ALLOPHYLUS Linnaeus ALLOPHYLUS STENOPHYLLUS sp. nov. Frutex erectus, ramulis foliisque plus minusve pubescens, ramis glabris, lenticellatis; foliis 5-foliolatis, foliolis chartaceis, angustissime lanceolatis, 5 ad 10 cm longis, 1 ad 1.5 cm latis, tenuiter acuminatis, margine distanter serratis, supra subglabris, subtus ad costa nervisque leviter pubescentibus, axillis dense barbatis; racemis axillaribus, 7 ad 5 cm longis; fructibus sub- ellipsoideis, circiter 5 mm longis. A small erect shrub, the branches glabrous, brownish, dis- tinctly lenticellate, the branchlets paler and somewhat appressed- pubescent. Leaves palmately 5-foliolate, their petioles 4 to 6 cm long, appressed-pubescent with short hairs; leaflets chartaceous, very narrowly lanceolate, 5 to 10 cm long, 1 to 1.5 cm wide, slenderly acuminate, base decurrent-acuminate, margins rather distantly serrate, the upper surface pale-olivaceous, shining, gla- brous or very slightly pubescent especially on the midrib, the lower surface paler, sparingly pubescent on the midrib and nerves, densely bearded in the axils; petiolules of the central leaflets 5 to 10 mm long, the lateral leaflets subsessile. Racemes axillary, solitary, sparingly pubescent, 5 to 7 cm long. Fruits rather numerous, subellipsoid, brown when dry, very slightly pubescent, about 5 mm long. LUZON, Ilocos Norte Province, Burgos, Bur. Sci. 32839 Ramos, August 29, 1918, in dry thickets at low altitudes. This species manifestly belongs in the group with Allophylus dimorphus Radlk., from which I do not consider that A. quinatus Radlk, can be specifically distinguished. Allophylus stenophyllus differs from all forms of the above species in its very narrowly lanceolate leaflets, which are nearly glabrous on both surfaces and which attain a length of 10 cm but do not exceed 15 cm ~ in width. GUIOA Cavanilles GUIOA PARVIFOLIOLA sp. nov. Frutex, partibus junioribus foliisque plus minusve pubescens; foliis 10 cm longis, foliolis plerumque 6, anguste oblongis, obtusis, coriaceis, 2 ad 4 cm longis, 6 ad 15 mm latis, basi angus- tatis, apice obtusis, vetustioribus supra glabris, nervis utrinque 418 The Philippine Journal of Science 119 8 ad 10; capsulis late obovatis, 3-lobatis, circiter 1 cm longis et 1.3 cm latis, lobis late rotundatis, apice retusis apiculatisque. A shrub, the younger parts and the young leaves more or less pubescent, the older leaves glabrous or nearly so, at least on the upper surface. Leaves 10 cm long or less, the rachis and petio- lules puberulent; leaflets usually 6, narrowly oblong, coriaceous, 2 to 4 cm long, 6 to 15 mm wide, obtuse, narrowed below to the somewhat decurrent-acuminate base, the upper surface dark- olivaceous, glabrous, the lower surface much paler and, at least when young, more or less pubescent with scattered short hairs; lateral nerves 8 to 10 on each side of the midrib, prominent on the lower surface; petiolules 2 to 5 mm long. Infructescences axillary, 3 to 5 cm long, pubescent, the branches few. Capsules broadly obovate, 3-lobed, apex more or less retuse and distinctly apiculate, about 1.3 cm wide and nearly 1 cm long, the lobes broadly rounded, red, glabrous. Seeds obovoid, smooth, shining, about 6 mm long. LuzoN, Ilocos Norte Province, Mount Nagapatan, Bur. Sci. 83187 Ramos, August 8, 1918, on dry slopes, altitude about 700 meters. The alliance of this species is manifestly with Guioa obtusa Merr., from which it radically differs in its very much smaller, relatively much narrower leaflets. ELAEOCARPACEAE ELAEOCARPUS Linnaeus ELAEOCARPUS ILOCANUS sp. nov. § Dicera. Arbor parva inflorescentiis densissime ferrugineo-pubescens; foliis ellipticis ad obovato-ellipticis, glabris, olivaceis, nitidis, 4 ad 7 cm longis, basi acutis, apice breviter obtuse acuminatis, margine distanter apiculato-crenatis, nervis utrinque circiter a; distinctis; racemis numerosis, solitariis, plerumque e axillis de- foliatis, 5 cm longis; floribus 5-meris, 8 mm longis; sepalis lanceo- latis, pubescentibus; petalis glaberrimis, usque ad 4 laciniatis, laciniae circiter 20; staminibus circiter 25, antheris oblongis, obtusis, apice parce ciliatis; ovario pubescente, 3-loculare; fruc- tibus ellipsoideis, 2.5 cm longis, 1-locellatis, A tree, about 5 m high, the inflorescences very densely ferru- ginous-pubescent, the branches. terete, glabrous, the very young branchlets either more or less pubescent or glabrous. Leaves elliptic to obovate-elliptic, olivaceous and shining on the upper surface, the lower surface paler, 4 to 7 cm long, 2 to 3.5 cm wide, base acute, apex shortly and obtusely acuminate, margins distantly apiculate-crenate; lateral nerves about 7 on each side XIV, 4 Merrill: Noteworthy Philippine Plants, XV 419 of the midrib, prominent; petioles glabrous, 1 to 1.3 cm long, racemes numerous, solitary, chiefly from the axils of fallen leaves, about 5 cm long, usually 6- to 10-flowered. Flowers pink, about 8 mm long, 5-merous, their pedicels 5 to 6 mm long, densely pubescent. Sepals lanceolate, 7 mm long, densely pubescent. Petals entirely glabrous, 7 to 8 mm long, cuneate, split to about the middle into three primary divisions, these again divided, the ultimate laciniae about 20, slender. Stamens about 25, fi- laments 1 to 1.5 mm long, anthers narrowly oblong, 1.5 to 1.8 mm long, minutely scabrid, the cells slightly unequal, obtuse, sparingly ciliate at their tips. Ovary ovoid, densely pubescent, 3-celled, Fruit ellipsoid, about 2.5 em long, black when dry, one-celled. LUZON, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 33271 Ramos, August 21, 1918, on forested slopes near the summit, altitude about 1,000 meters. This species somewhat resembles both Elaeocarpus cumingit Turcz. and E. forbesii Merr., differing radically from the former in its very densely pubescent racemes and sepals and from the latter in its much longer racemes and entirely glabrous petioles. SLOANEA Linnaeus SLOANEA JAVANICA (Migq.) Szyszyl. ex K. Schum in Engl. & Prantl. Nat. Pflanzenfam. 3° (1890) 5; Koord. & Val. Bijdr. Boom. Java 1 (1894) 239, Atlas Baumart. Java 3 (1914) f. 433 (after Miquel). _Phoenicospermum javanicum Mig. Ann. Mus. Bot. Lugd.—Bat. 2 (1865) 68, t. 3. Luzon, Cagayan Province, Pamplona, For. Bur. 14713 Darling, March 17, 1909; Pamplona, For. Bur. 16988 Bacani, March, 1907: Laguna Province, Cavinti, For. Bur. 19665 Racelis, August, 1912, with mature fruits. MINDoRO, Ibolo River, For. Bur. 11482 Merritt, May, 1908, sterile. A tree, 10 to 25 m high, in forests up to an altitude of 600 meters. This species has previously been known only from Java, where it is apparently very rare, as until recently it was known only from trees cultivated in the botanical garden at Buitenzorg; it is now known from Depok, near Buitenzorg, Koorders 42807, a duplicate of Koorders’ specimen being before me. The Philip- pine specimens have somewhat larger and more prominently acuminate leaves than the Javan material, while the fruits are a little larger. In the absence of flowers I can detect no differen- tial characters of sufficient importance to warrant the separa- tion of the Philippine form from the Javan one. The genus is new to the Philippines. 420 The Philippine Journal of Science 1919 Szyszylowicz does not make the transfer of Phoenicospermum javanicum Miq. to Sloanea in his paper “Zur Systematik der Tiliaceen” * as indicated by Koorders and Valeton, but merely reduces the genus Phoenicospermum to Sloanea. STERCULIACEAE PTEROSPERMUM Schreber PTEROSPERMUM MEGALANTHUM sp. nov. Arbor parva, subtus foliis et ramulis et inflorescentiis densis- sime ferrugineo-stellato-tomentosis; foliis oblongis, inaequila- teralibus, integris, acuminatis, supra glabris, nitidis, usque ad 15 cm longis; inflorescentiis plerumque 2-floris, floribus longe pedicellatis, circiter 6.5 cm longis; sepalis anguste lanceolatis, 6 cm longis et 1 cm latis; petalis oblique obovatis, 3 cm latis, quam sepalis paullo longioribus. A tree, 7 to 8 m high, the branchlets, petioles, inflorescences, and the lower surface of the leaves densely stellate-tomentose with ferruginous hairs. Branches at first stellate-pubescent, soon becoming glabrous, nearly black when dry. Leaves inequi- lateral, oblong, chartaceous to subcoriaceous, 9 to 15 cm long, 3 to 6.5 cm wide, entire, or the margins obscurely undulate, rather sharply acuminate, the upper surface olivaceous, glabrous and shining when dry, the basal lobe of the broader side broadly rounded, of the narrower side rounded to acute, the leaf some- what obliquely cordate; lateral nerves about 6 on each side of the midrib, excluding those extending into the broader basal lobe; petioles densely ferruginous-stellate-pubescent, 5 to 10 mm long. Inflorescences in the uppermost axils and subterminal, usually 2-flowered, the peduncles 1.5 cm long or less, the pedicels in an- thesis 3 to 4 cm long, densely ferruginous-pubescent; bracts linear-lanceolate, entire, deciduous. Flowers white, fragrant, about 6.5 cm long, 5-merous. Sepals coriaceous, narrowly lan- ceolate, about 6 cm long, 1 cm wide, slightly acuminate, externally densely ferruginous-stellate-pubescent, inside densely subap- pressed fulvous-hirsute. Petals obliquely obovate, membrana- ceous, glabrous, about 6.5 cm long, 3 em wide. Staminal column glabrous, 1.5 cm long. Staminodes 3 cm long. Filaments and anthers each 1.5 cm long, sometimes 2, sometimes 3 between each pair of staminodes. Ovary ovoid, densely ferruginous-villous; style glabrous, 3 em long. : PANAY, Capiz Province, Jamindan, Bur. Sci, 31269 Ramos & Edano (type), April 28, 1918; Ibahay, For. Bur. 25359 Achacoso, *Engl. Bot. Jahrb. 5 (1885) 427-457. XIV, 4 Merrill: Noteworthy Philippine Plants, XV 421 April 7, 1916, along streams in open forests, altitude about 100 meters. ; This species has almost exactly the vegetative characters of Pterospermum blumeanum Korth. (P. javanicum Jungh.) and the second specimen cited above, which has only immature buds, was originally so identified. While it is manifestly closely allied to Korthals’s species, it differs remarkably in its larger flowers, somewhat longer and much broader sepals, and in its much larger petals, which exceed the sepals in length. DILLENIACEAE DILLENIA Linnaeus DILLENIA MEGALOPHYLLA sp. nov. § Wormia. Arbor, inflorescentiis exceptis glabra, ramulis ultimis 1.5 ad 2 cm diametro; foliis coriaceis, ovatis ad elliptico-ovatis, usque ad 1 m longis, apice rotundatis, basi angustatis, margine undu- latis vel obscure dentatis, nervis utrinque 15 ad 18, perspicuis, lamina decurrente, basi conduplicatis ; inflorescentiis axillaribus, longe pedunculatis, paucifloris, circiter 40 cm longis, dense pubes- centibus; sepalis coriaceis, obovatis ad late ellipticis, 3 cm longis, extus molliter dense fulvo-pubescentibus ; carpellis plerumque 8. A tree, about 7 m high, nearly glabrous except the inflores- cences, the ultimate branches 1.5 to 2 cm in diameter. Leaves coriaceous, ovate to elliptic-ovate, up to 1m long and 60 cm wide, apex rounded, base gradually narrowed, margins somewhat un- dulate and obscurely toothed, glabrous on both surfaces, what is apparently the petiolar part of the leaf with the lamina contin- uous but with much slenderer nerves than the leaf proper, ex- tending to the very base of the petiole, conduplicate, the upper surface for a distance of about 2.4 cm on each side of the midrib closely appressed and with a modified, brown, apparently absorb- ent epidermis, this region in some leaves inhabited by colonies of ants, the lower surface of this conduplicate area rather softly pubescent; lateral nerves above the modified petiolar part of the lamina very prominent, 15 to 18 on each side of the midrib. In- florescences in the uppermost axils, long-peduncled, few-flowered, sparingly branched above, about 40 cm long, densely and softly pubescent; pedicels 3 to 5 em long. Sepals coriaceous, obovate to broadly elliptic, rounded, concave, about 3 cm long, softly and densely fulvous-pubescent externally. Petals not seen. Sta- mens apparently equal or subequal. Carpels usually 8, glabrous, not twisted, the styles up to 3 cm long. Immature carpels coria- ceous, about 3 cm long, apparently purplish, dehiscent; ovules about 11 in each carpel; immature seeds with a small basal aril. 422 The Philippine Journal of Science mie LUZON, Ilocos Norte Province, between Bangui and Claveria, Bur. Sci. 33034 Ramos, August 21, 1918, in forests near the sea at low altitudes. This species is remarkable for its exceedingly large leaves, which attain at least 1 meter in length. It belongs in the group with Dillenia suffruticosa Grieff. and like that species has a por- tion of the upper surface of the decurrent basal part of the lamina conduplicate and supplied with a modified, apparently ab- sorbent, epidermal tissue. Among the Philippine species it is most closely allied to Dillenia papyracea Merr., of Basilan, from which it is readily distinguished by its very much larger leaves, softly pubescent inflorescences and sepals, and fewer carpels. SAURAUIA Willdenow SAURAUIA SANTOSII sp. nov. Arbor 7 ad 9 m alta, S. eleganti affinis, differt floribus major- ibus, 1.5 ad 2 cm diametro, confertis, foliis subtus densissime fulvo-ferrugineo-lanatis. A tree, 7 to 9 m high, the branches brown, setose, and densely ferruginous-tomentose, the setae comparatively few, filiform, 4 to 7 mm in length, ascending. Leaves thickly coriaceous, rigid, oblong, 13 to 18 cm long, 5.5 to 7 cm wide, obtuse, the base round- ed, cordate, often somewhat oblique, the margins with small, thick, rounded teeth, these often terminating in filiform, decid- uous setae, the upper surface brown, glabrous, shining, rugose- pustulate, the nerves impressed, the lower surface very densely fulvous-ferruginous-lanate, the indumentum covering the whole surface, midrib, and nerves, the midrib and nerves also with few, scattered, filiform setae; lateral nerves 25 to 30, prominent, projecting; petioles 2 to 4 cm long, setose and tomentose. In- florescences axillary, long-peduncled, cymose, the flowers crowd- ed, the peduncles 10 to 12 cm long, densely tomentose and rather sparingly setose, the cymes compact, 4 to 5 cm in diameter. Flowers white, 1.5 to 2 cm in diameter, their pedicels up to 5 mm in length, setose and tomentose. Sepals broadly ovate, obtuse or rounded, 5 mm in diameter, the outer ones thickly coriaceous, densely villous and setose, the setae slender, up to 2 mm in length, the inner ones thinner, rather sparingly villous and not setose. Petals oblong-obovate, rounded, about 11 mm long and d mm wide. Ovary glabrous; styles 2, free to the base, 4 mm ong. LUZON, Benguet Subprovince, Pauai, Bur. Sci. 31937 Santos, May 16, 1918, on slopes, altitude about 2,200 meters, This species is manifestly allied to Sawrauia elegans (Choisy) XIV, 4 Merrill: Noteworthy Philippine Plants, XV 423 F.-Vill., from which it is readily distinguished by its dense indu- mentum, its compact cymes, and its much larger flowers. The two styles are constant in all the flowers examined, Saurawia elegans normally having three styles, although in some flowers they are reduced to two. FLACOURTIACEAE HOMALIUM Jacquin HOMALIUM RAMOSII sp. nov. § Myriantheia. Arbor, inflorescentiis exceptis glabra; foliis oblongis ad ob- longo-ellipticis, subcoriaceis, integris nitidis, 6 ad 10 cm longis, basi acutis, apice obtuse acuminatis, nervis utrinque circiter 7; inflorescentiis axillaribus terminalibusque, racemosis vel depau- perato-paniculatis, usque ad 10 cm longis; floribus fasciculatis, 5-meris, 6 ad 7 mm diametro, petalis oblongis ad oblongo-obo- vatis; staminibus 15. A small tree, glabrous except the inflorescences, the branches terete, brownish. Leaves oblong to oblong-elliptic, subcoriaceous, shining, entire, 6 to 10 cm long, 3 to 4.5 cm wide, subequally narrowed to the acute base and acuminate apex, the acumen blunt, short; lateral nerves about 7 on each side of the midrib, arched, anastomosing, distinct as are the reticulations; petioles 5 to 8 mm long. Racemes axillary and terminal, solitary or forming sparingly branched inflorescences, the individual ones up to 10 cm long, cinereous-pubescent, the flowers mostly fas- cicled at the nodes, usually 2 or 3 in a fascicle, their pedicels 3 to 4 mm long, jointed with the calyx; bracteoles narrowly lan- ceolate, about 2 mm long. Flowers white, 5-merous, 6 to 7 mm in diameter. Sepals and petals about 2.5 mm long, pubescent, rather densely so inside, oblong to narrowly oblong-obovate. Stamens 15, three opposite each petal, their filaments about 2 mm long, glabrous. Ovary very densely pubescent, styles about 1.8 mm long, villous below. Luzon, Ilocos Norte Province, between Bangui and Claveria, Bur. Sci. 82985 Ramos, August 12, 1918, on slopes, altitude about 450 meters. This species has much the vegetative characters of Homalium multiflorum Merr. but belongs in a different section of the genus and is most closely allied to H. samarense Merr., from which it differs in its much shorter indumentum on its inflorescences; its smaller flowers, which are fascicled and not solitary along the racemes; and its entire leaves. In the present species occasional leaves are found presenting very few greatly reduced marginal teeth, but most of the leaves are quite entire. — 424 The Philippine Journal of Science > 1919 BEGONIACEAE BEGONIA Linnaeus BEGONIA COLLISIAE sp. nov. § Diploclinium. Herba, pedunculis foliisque utrinque ad nervis plus minusve ferrugineo-villosis; foliis longe petiolatis, in ambitu late ovatis, usque ad 20 cm longis, inaequilateralibus, basi lateraliter cor- datis, apice leviter acuminatis, profunde subpalmatim lobatis, lobis numerosis, oblongis, 2 ad 5 cm longis, irregulariter et grosse dentato-serratis vel lobulatis; inflorescentiis longe pedun- culatis, dichotome ramosis, parce glandulosis, paucifloris; floribus circiter 3 cm diametro; sepalis late ovatis, circiter 11 mm latis;. petalis sepalis aequantibus, 6 ad 7 mm latis. An herb, from short, rather stout, prostrate rhizomes about 1cmindiameter. Leaves long-petioled, in outline broadly ovate, 15 to 20 cm long, 10 to 15 cm wide, when dry submembranaceous, the upper surface olivaceous, somewhat shining, the lower brown- ish, both surfaces more or less ferruginous-villous on the nerves, when fresh the nerves beneath red, inequilateral, subpalmately and conspicuously lobed, the lobes oblong, 2 to 5 cm long, 2 to 3 cm wide below, all acute or acuminate and irregularly and coarsely dentate-serrate or lobulate, the primary lobes usually about 10, the base of the leaf laterally cordate, the sinus narrow, 2 to 4 cm deep, the basal lobes broadly rounded, often overlap- ping; petioles brown when dry, ferruginous-villous. Peduncles nearly glabrous, up to 20 em in length, the younger parts spar- ingly glandular, dichotomously branched at the apex, the pri- mary branches about 4 cm long, few-flowered. Staminate flowers pink and white, about 3 cm in diameter. Sepals broadly ovate, about 14 mm long, 11 mm wide, rounded, with few, small, widely scattered, brown glands. Petals oblong-obovate, as long as the sepals, 6 to 7 mm wide. Stamens many, the anthers narrowly obovoid, 0.8 mm long, equaling or shorter than the filaments. Pistillate flowers and fruits unknown. PANAY, Capiz Province, Libeacao, Bur. Sci. 31469 Ramos & Edaio, May 5, 1918, on stream banks in forests. This species is remarkably distinct from all previously de- scribed Philippine ones in the section Diploclinium, and is strongly characterized by its deeply lobed leaves, the lobes being usually about 10 in number, up to 5 cm in length, and coarsely and irregularly toothed and lobulate, BEGONIA LANCILIMBA sp. nov. § Diploclinium. Herba glaberrima; foliis longe petiolatis, membranaceis, oblongis ad oblongo-lanceolatis, sub-aequilateralibus, usque ad XIV, 4 Merrill: Noteworthy Philippine Plants, XV 425 13 cm longis, basi rotundatis, leviter cordatis, apice tenuiter caudato-acuminatis, margine irregulariter dentatis; petiolo 15 cm longo; inflorescentiis foliis subaequantibus, paucifloris; floribus é@ circiter 3 cm diametro. An entirely glabrous herb, the rhizomes creeping, simple, about 5 mm in diameter, supplied with numerous, lanceolate, caudate-acuminate, brown stipules, 1 to 1.5 cmin length. Leaves pale-green on both surfaces when fresh, when dry membrana- ceous, somewhat shiny, oblong to oblong-lanceolate, nearly equi- lateral, 7 to 13 cm long, 2.5 to 4 cm wide, base rounded and very slightly cordate, narrowed upward from the lower one-third to the slenderly caudate-acuminate apex, the margin irregularly toothed, a few of the teeth rather large, most of them small; basal nerves about 8 ascending, a few lateral nerves leaving the midrib at and above the middle; petioles about 15 cm long. Pe- duncles of the staminate inflorescences 19 to 24 cm long, usually with 2 short branches at the apex, each bearing about 2 rela- tively large white flowers about 3 cm in diameter. Sepals 2, oblong-elliptic rounded, about 1.8 mm long, 10 mm wide. Petals narrowly oblong, equaling the sepals, 3 to 3.5 mm wide, rounded. Stamens about 35, the anthers oblong, about 1.5 mm long; fila- ments about 1 to 2 mm long. Panay, Antique Province, Culasi, Bur. Sci. 32232 (type), 32570, 82286 McGregor, June 14 and July 14, 1918, on forested slopes, altitude 900 to 1,000 meters. This species is not closely allied to any previously described Philippine form and is very strongly characterized by being en- tirely glabrous; by its nearly equilateral, oblong-lanceolate to lan- ceolate, caudate-acuminate leaves which are slightly cordate at the base; by its long peduncles; and its few, comparatively large, white flowers. BEGONIA OBTUSIFOLIA sp. nov. § Diploclinium. Herba prostrata, petiolis et subtus foliis ad nervis adpresse ferrugineo-pubescentibus; foliis longe petiolatis, oblongis, inae- quilateralibus, 7 ad 10 cm longis, apice obtuse rotundatis, basi leviter oblique cordatis, margine late lobatis vel undulatis, lobis rotundatis; inflorescentiis quam foliis longioribus, paucifloris; floribus ¢ circiter 3 cm diametro; capsulis aequaliter 3-alatis, junioribus circiter 14 mm latis et 10 mm longis, apice subtrun- catis, basi rotundatis. Rhizomes slender, creeping, the petioles and nerves on the lower surface of the leaves appressed ferruginous-pubescent. 426 The Philippine Journal of Science 1919 Leaves oblong to oblong-ovate, distinctly inequilateral, 7 to 10 cm long, 3.5 to 4.5 cm wide, the upper surface glabrous, some- what shiny, somewhat narrowed upward to the obtusely rounded apex, the base somewhat obliquely cordate, the sinus shallow, narrow, the margin on the narrower side usually with 1 or 2 short obtuse lobes, on the broader side with from 2 to 4 similar lobes or undulations ; basal nerves usually 6, slender, distinct; pet- ioles slender, 10 to 13 em long. Peduncles 18 to 24 cm long, glabrous, dichotomously branched at the apex, the primary branches up to 3 em in length, few-flowered. Flowers pink, the sepals of the staminate ones broadly ovate, rounded, about 14 mm long, 11 mm wide. Petals equaling the sepals, narrowly oblong, 4mm wide. Pistillate flowers as large as the staminate ones, the young capsules equally 3-winged, about 14 mm wide and 10 mm long including the wings, base rounded, apex some- what truncate. PANAY, Capiz Province, Mount Macosolon, Bur. Sci. 30808, Ramos & Edaito, April. 19, 1918, on stones in damp forests. In the form and characters of its leaves this species is remote from all other described ones of the section and is easily recog- nized by its usually oblong, rounded, obtuse leaves, which have a few short rounded lobes or undulations on each side. BEGONIA RUBRIFOLIA sp. nov. § Diploclinium. Herba prostrata, rhizomate brevibus, petiolis et subtus foliis ad nervis plus minusve ferrugineo-lanatis; foliis integris, inae- quilateralibus, late ovatis, chartaceis, in siccitate utrinque rubris, in vivo supra viridis, subtus rubris, usque ad 9 cm longis, basi lateraliter cordatis, palmatim 8-nerviis, nervis conspicuis, apice late acutis; infructescentiis 20 ad 30 cm longis, quam foliis multo longioribus, dichotome ramosis; capsulis paucis, inaequa- liter 3-alatis, circiter 12 mm longis et 2.3 mm latis, apice trun- catis, basi rotundatis, — A prostrate herb, the rhizomes short, stout, brown, glabrous. Leaves chartaceous, when fresh green on the upper surface, red beneath, when dry uniformly reddish-purple on both surfaces, the upper surface glabrous, the lower densely white- or grayish- punctate, rather densely ferruginous-lanate on the nerves, broadly ovate, inequilateral, entire, 8 to 10 cm long and nearly as wide, the apex broadly acute, base laterally cordate and palmately 8-nerved, the nerves conspicuous, margins ciliate, the sinus narrow, basal lobes broadly rounded; petioles sparingly ferruginous-lanate, 7 to 11 em long; stipules oblong-ovate, some- XIV, 4 Merrill: Noteworthy Philippine Plants, XV 427 what ciliate, about 1 cm long, and with a few fimbriate-ciliate paleae. Infructescences long-peduncled, 20 to 30 em long, nearly glabrous, dichotomously branched at the apex of the peduncle, the primary branches about 5 cm long. Capsules few, including the wings about 12 mm long and 23 mm wide, inequally 3- winged, one wing about twice as large as the other two, the apex subtruncate, base broadly rounded, the outer upper angles of the wings rounded or obtuse. PANAY, Antique Province, Culasi, Bur. Sci. 32430 McGregor, June 7, 1918, altitude about 1,000 meters. _ This species is allied to Begonia copelandii Merr. and B. alva- rezti Merr., but its indumentum is very different from both. The leaves are characteristically reddish-purple on both sur- faces when dry. BEGONIA SERPENS sp. nov. § Diplocliniwm. Herba prostrata, repens, leviter ciliatis, internodiis 1 ad 3 em longis; foliis breviter petiolatis, irregulariter rhomboideis, inaequilateralibus, 1.5 cm diametro, basi leviter cordatis, apice acutis vel acuminatis, margine irregulariter lobatis; inflorescen- tiis paucifloris, circiter 8 cm longis; floribus ¢ circiter 1.5 cm diametro. A very small, slender, prostrate, sparingly appressed-ciliate herb, the rhizomes 2 mm or less in diameter, attaining a length of at least 25 em. Leaves irregularly rhomboid, inequilateral, base more or less cordate, apex shortly acuminate, 1.5 to 2.5 em in diameter, margins shallowly and irregularly lobed, the upper surface usually mottled with gray, glabrous, the lower surface sparingly ciliate on the nerves with appressed brown hairs, the margins more or less ciliate with short spreading hairs; petioles 1 to 1.5 cm long, sparingly ciliate; stipules brown, 5 to 6 mm long, oblong-ovate, with a long slender tip. Peduncles slender, about 3 cm long, usually 2-flowered. Staminate flowers usually pink, about 1.5 cm in diameter. Sepals broadly obovate, rounded, 7 to 8 mm long, 6 to 7 mm wide. Petals equaling the sepals, oblong-obovate, rounded, 4 mm wide. Stamens about 35; anthers oblong-obovoid, 0.8 mm long; filaments 1 mm long or less. Panay, Antique Province, Culasi, Bur. Sci. 32588 (type), $2541 McGregor, May and June, 1918, on damp forested slopes, altitude 500 to 900 meters. This species is as closely allied to Begonia nigritarum Steud. as to any other but is readily distinguished by its smaller leaves, short petioles, and short, very few-flowered inflorescences as 428 The Philippine Journal of Science 1919 well as by its very slender, greatly elongated rhizomes; the internodes vary from 1 to 3 cm in length. BEGONIA PANAYENSIS sp. nov. § Petermannia. Herba suffruticosa, erecta, ramosa, glabra, circiter 1.5 m alta; foliis in siccitate membranaceis, nitidis, oblongis, usque ad 20 cm longis, inaequilateralibus, apice angustatis, tenuiter acuminatis, basi rotundatis, haud cordatis, margine distanter‘et irregulariter dentatis; inflorescentiis terminalibus, cymoso-paniculatis, usque ad 17 cm longis; floribus ¢ numerosis, circiter 1.5 cm diametro; sepalis orbicularis, 7 ad 8 mm diametro; petalis 0; capsulis ellipticis vel obovato-ellipticis, circiter 2.5 cm longis et 2 cm latis, aequaliter 3-alatis, apice truncato-rotundatis, basi vix an- gustatis, rotundatis. An erect, branched, glabrous, suffrutescent herb, about 1.5 m high, the stems and branches stout. Leaves when dry mem- branaceous, shining, inequilateral, oblong, 16 to 20 cm long, 4 to 6 cm wide, narrowed upward to the slenderly acuminate apex and somewhat narrowed below to the rounded base, or the narrower side of the lamina acute and the broader side rounded, not at all cordate, margins distantly and irregularly dentate, or that of the narrower side often nearly entire except near the apex; nerves prominent, ascending; petioles stout, 1 to 1.5 cm long; stipules membranaceous, oblong to oblong-lanceolate, acuminate, 1.5 to 2 cm long. Inflorescences terminal, cymose- paniculate, up to 17 em in length, the staminate flowers numer- ous, the pistillate flowers apparently few and only at the base of the inflorescence, none present on the type, but represented by mature capsules. Staminate flowers pink, about 1.5 cm in diameter. Sepals orbicular, 7 to 8 mm in diameter. Petals none. Stamens about 35, the anthers narrowly oblong to oblong- obovoid, 1.2 to 1.5 mm long, longer than the filaments. Cap- sules-one or two at the base of the inflorescence, including the three equal wings about 2.5 cm long, 2 cm wide, elliptic to obovate-elliptic, apex rounded-truncate, base scarcely narrowed, usually rounded; pedicels 1.5 to 2 cm long. PANAY, Antique Province, Culasi, Bur. Sci. 32309 McGregor, June 8, 1918, in forests along cool mountain streams, altitude about 250 meters. , This unusually robust species is manifestly allied to Begonia longistipula Merr., from which it is distinguished by its larger leaves and flowers and by its much larger, differently shaped capsules. XIV, 4 Merrill: Noteworthy Philippine Plants, XV 429 THYMELAEACEAE PHALERIA Jack PHALERIA PLATYPHYLLA sp. nov. Frutex glaber Phaleria cumingiae affinis differt foliis multo majoribus, circiter 25 cm longis et 9 ad 11 cm latis, floribus caulinis, fasciculatis. An entirely glabrous shrub, the branches reddish-brown when dry. Leaves oblong to oblong-ovate, firmly chartaceous, about 25 cm long, 9 to 11 em wide, brownish-olivaceous when dry, shining, apex slenderly acuminate, base broadly rounded; lateral nerves about 10 on each side of the midrib, distinct; petioles 1 cm long or less. Flowers white, 4-merous, in fascicles on the trunk, about 3 cm long. Perianth-lobes 4, elliptic, rounded, 8 to 9 mm long, 6.5 mm wide. Filaments 8, somewhat exserted. Immature fruit ovoid-ellipsoid, somewhat rostrate, 1.5 cm long. PANAY, Capiz Province, Mount Macosolon, Bur. Sci. 30759 Ramos & Edaio, April 22, 1918, in forests. This species has the vegetative characters of Phaleria perrot- tetiana F.-Vill., but the flowers are entirely glabrous and are moreover borne in fascicles on the trunk. It is distinguished from Phaleria cumingii F.-Vill. by its much larger leaves and cauline inflorescences. MYRTACEAE CLOEZIA Brongniart and Gris CLOEZIA URDANETENSIS (Elm.) comb. nov. Photinia urdanetensis Elm. Leafi. Philip. Bot. 8 (1915) 2802. MINDANAO, Agusan Province, Mount Urdaneta, Elmer 13694, September, 1912, on exposed forested ridges in the mossy forest, altitude about 1,700 meters. The genus Cloézia has hitherto been known only from New Caledonia, where it is represented by six species. The discovery of a representative of it in the Philippines adds another genus to the now remarkable list of genera that are known only from the Philippines and the islands to the south and southeast of the Archipelago, including Celebes and the Moluccas generally, New Guinea, New Caledonia, and northeastern Australia. The present species was described by Mr. Elmer as a representative of the rosaceous genus Photinia, but a critical examination of the type collection shows that it agrees even to the most minute details with the myrtaceous Cloézia.* In the original description *Brongniart & Gris in Ann. Sci. Nat. Bot. V 2 (1864) 134. 165368——5 430 The Philippine Journal of Science i919 of Mr. Elmer the stamens are described as “about 10,” but I find them to vary from 15 to 20. The ovary is two-thirds to three-fourths superior, 3-celled, each cell with about 6 ovules ascending from the inner basal angle. The apical gland of the anthers, or continuation of the connective, is unusually conspi- cuous, being longer than the anther-cells. The fruits of the Philippine species are as yet unknown. The leaves are distinctly Eugenia-like and are glandular-punctate. — : TRISTANIA R. Brown TRISTANIA OBLONGIFOLIA sp. nov. Species 7.. decorticatae similis, differt foliis junioribus utrinque dense ferrugineo-pubescentibus, inflorescentiis et flo- ribus dense fulvo-pubescentibus, staminibus circiter 35. Arbor circiter 5 m alta, ramis teretibus, glabris; foliis coriaceis, ob- longis, in siccitate brunneis, usque ad 9 cm longis, utrinque - subaequaliter angustatis, basi cuneatis, apice leviter obtuse acu- minatis, nervis primariis utrinque circiter 20, tenuibus; inflores- centiis axillaribus, corymbosis, circiter 3 cm longis; capsulis glabris, subovoideis, circiter 5 mm longis. — A tree, about 5 m high, the older parts glabrous, the younger leaves very densely ferruginous-tomentose on both surfaces, the inflorescences densely fulvous-tomentose. Branches brown, smooth, terete, glabrous, the younger branchlets dark-brown. Leaves oblong, coriaceous, brown when dry, shining, of about the same color on both surfaces, glandular beneath, 6 to 9 cm long, 1.8 to 2.4 cm wide, subequally narrowed to the cuneate base and to the slightly acuminate apex, the tip obtuse or subobtuse; primary lateral nerves about 20 on each side of the midrib, slender, about equally distinct on both surfaces, not much more prominent than are the secondary ones; petioles 5 mm long or less. Inflorescences axillary, about 3 cm long, rather stout, few- flowered, densely fulvous-pubescent. Flowers about 4 mm long, their pedicels stout, pubescent, 1 to 2.5 mm long, the buds ellip- soid. Calyx ovoid, the lobes 5, ovate, obtuse, 1.2 mm long. Stamens in 5 phalanges of about 7 each, inflexed in bud. Infruc- tescences glabrous, 4 to 5 cm long. Capsules somewhat ovoid, dark-brown, glabrous, about 5 mm long. Luzon, Tayabas Province, Mount Binuang, Bur. Sci. 28617 Ramos & Edafio, May 28, 1917, on forested slopes, altitude at least 400 meters. In the vegetative characters, other than the very densely fer- ruginous-tomentose young leaves, this species greatly resembles XIV, 4 Merrill: Noteworthy Philippine Plants, XV 431 Tristania decorticata Merr. It differs remarkably from that species in its densely pubescent, larger flowers and its stamens being arranged in phalanges of about 7 each, not in groups of threes. MELASTOMATACEAE MEDINILLA Gaudichaud MEDINILLA OBLANCEOLATA sp. nov. Frutex, ramulis et inflorescentiis et foliis subtus ad costa ner- visque perspicue setoso-ciliatis; foliis ternatis, breviter petio- latis, chartaceis, oblongo-oblanceolatis, usque ad 38 cm longis, tenuiter acuminatis, deorsum valde angustatis, penninerviis, nervis utrinque circiter 6, perspicuis; inflorescentiis lateralibus, circiter 10 cm longis, pedunculatis, cylindraceo-paniculatis, ramis numerosis, circiter 2 cm longis, patulis, dichotomis; floribus 4-meris, petalis oblongo-obovatis, circiter 13 mm longis, stami- nibus aequalibus. A shrub, the stems about 2 cm in diameter, the young branch- lets, inflorescences, and the midrib and nerves on the lower sur- face more or less setose-ciliate with long, spreading, pale-brownish hairs, the ciliae up to 5 mm in length. Leaves verticillate, 3 at each node, chartaceous, oblong-oblanceolate, 35 to 38 cm long, about 10 cm wide, the apex slenderly acuminate, narrowed in the lower one-half. or two-thirds to the abruptly obtuse base which is 1 cm wide or less; lateral nerves about 6 on each side of the midrib, distant, ascending, somewhat curved, the primary reticulations subparallel, distant, distinct; petioles very stout, 1 em long or less. Inflorescences lateral from the trunks, about 10 cm long, solitary, peduncled, paniculate the primary branches numerous, densely arranged, spreading, about 2 cm long, dichot- omous, sparingly ciliate, the bracteoles oblong, about 2 mm long. Flowers 4-merous, red, their pedicels about 5 mm long, petals oblong-obovate, 13 mm long. Stamens 8, equal, their filaments . 8 mm long; anthers lanceolate-acuminate, 7 mm long, the dorsal appendage less than 1 mm long, the anterior ones short, curved. Fruit fleshy, glabrous, ovoid or obovoid, truncate, about 1 cm long. Panay, Capiz Province, Mount Madiaas, Bur. Sci. 30703 Ramos & Edafio, May 14, 1918. This species is entirely distinct from all hitherto described forms and is strongly characterized by its elongated, ciliate- setose indumentum, which is dense on the young branchlets and scattered on the midrib and nerves; by its ternate, elongated, 432 The Philippine Journal of Science 1919 oblong-oblanceolate, slenderly acuminate leaves which are much narrowed below and very shortly petioled; and by its cauline, paniculate, rather dense inflorescences, the panicles being sub- cylindric in outline. MEDINILLA HASSELTII Blume in Flora 14 (1881) 513; Cogn. in DC. Monog. Phan. 7 (1891) 586. PANAY, Capiz Province, Jamindan, Bur. Sci. 31215 Ramos & Edaiio, May 19, 1918, in forests. Malay Peninsula, Sumatra, Borneo, and Java; not previously recorded from the Philippines. MEMECYLON Linnaeus MEMECYLON RAMOSII sp. nov. Frutex glaber, circiter 2 m altus, ramis ramulisque teretibus; foliis distincte 3-nerviis, oblongis, chartaceis ad subcoriaceis, usque ad 20 cm longis, basi subacutis, apice tenuiter acuminatis; nervis lateralibus inter nervulos transversales plus minusve ar- cuatis, distinctis, nervis transversalis utrinque circiter 14 3 pe- tiolo 1.5 ad 2 cm Jongo; infructescentiis axillaribus, cymosis, circiter 3 cm longis; fructibus ovoideis vel subglobosis, 6 ad 7 mm diametro. An erect glabrous shrub about 2 m high, the branches and branchlets terete, the latter pale when dry, about 1 mm in dia- meter. Leaves oblong, chartaceous to subcoriaceous, 15 to 20 cm long, 4.5 to 6 cm wide, the upper surface olivaceous when dry, the lower surface pale; marginal nerves distinct, more or less arched between the tips of the transverse ones, the latter about 14 on each side of the midrib, prominent on the lower sur- face; petioles 1.5 to 2 cm long. Infructescences axillary, cy- mose, about 3 cm long, branched from the base or several cymes in a fascicle. Fruit ovoid to globose, 6 to 7 mm in diameter, um- bellately arranged on the ultimate branchlets. Luzon, Ilocos Norte Province, between Bangui and Claveria, Bur. Sei. 32962 Ramos, August 29, 1918, in forests at low alti- tudes. This species belongs in the group with prominently nerved leaves, somewhat arched marginal nerves, and terete branches and branchlets. It is distinguished from all of the other Philip- pine species of this group by its unusually long petioles. MEMECYLON STENOPHYLLUM 5p. nov. Frutex erectus, circiter 1 m altus, glaber, ramis ramulisque tenuibus, teretibus; foliis angustate-lanceolatis, subcoriaceis, XIV, 4 Merrill: Noteworthy Philippine Plants, XV 483 usque ad 11 cm longis et 1.3 cm latis, acuminatis, basi abrupte rotundatis, sessilibus vel brevissime petiolatis, in siccitate bul- lato-rugosis, nervis utrinque circiter 25, patulis, supra leviter immersis, reticulis obsoletis; infructescentiis terminalibus, pe- dunculatis, 3 ad 4 cm longis, ut videtur paucifioris, fructibus globosis, circiter 7 mm diametro. An erect glabrous shrub, about 1 m high, the branches and branchlets slender, terete, the latter about 1 mm in diameter, the internodes 2 to 4 cm long. Leaves coriaceous, greenish-oliva- ceous, shining, narrowly lanceolate, 7 to 11 cm long, 6 to 13 mm wide, narrowed upward to the somewhat acuminate apex, the base abruptly rounded, sessile or subsessile, the midrib and lat- eral nerves somewhat impressed on the upper surface, the former very prominent beneath, the upper surface especially somewhat bullate-rugose on account of the impressed nerves; lateral nerves slender, spreading, about 25 on each side of the midrib, the reticulations obsolete. Infructescences terminal, peduncled, 3 to 4 em long, usually trichotomously branched, the ultimate branchlets 1 cm long or less. Fruits globose, black when mature, about 7 mm in diameter, their pedicels about 2 mm long. LUZON, Bulacan Province, Angat, Bur. Sci. 34101 Ramos & Edafio, March 1, 1919, on dry forested slopes at low altitudes. This species does not closely resemble any previously described form, being remarkably well characterized by its very narrow, sessile or subsessile leaves, which are gradually narrowed up- ward to the somewhat acuminate apex, the base being abruptly rounded; when dry the leaves are characteristically bullate- rugose by the more or less impressed lateral nerves. OTANTHERA Blume OTANTHERA STRIGOSA sp. nov. Frutex erectus, circiter 1 m altus, ramis ramulisque teretibus, parce adpresse strigosus, ramulis tenuibus; foliis oblongis, usque ad 9 cm longis, chartaceis, utrinque subaequaliter angustatis, basi acutis vel acuminatis, apice acuminatis, supra perspicue ad- presse strigosis, subtus, nervis parce strigosis exceptis, glabra; nervis 5; inflorescentiis plerumque 3-floris, bracteis lanceolatis, acuminatis, 5 ad 7 mm longis; floribus 5-meris, calycis perspicue penicellatis, setis patulis, 2 ad 3 mm longis, lobis anguste ob- longis, tubo aequantibus; petalis circiter 1.5 cm longis, obovatis. An erect shrub, about 1 m high, the branches and branchlets slender, terete, brownish when dry, sparingly appressed strigose. = 434 The Philippine Journal of Science 1919 Leaves oblong, chartaceous, 5 to 9 cm long, 1.5 to 2.5 em wide, subéqually narrowed to the acute or acuminate base and to the acuminate apex, chartaceous, the upper surface olivaceous when dry, conspicuously appressed strigose, the lower surface paler, glabrous except for the sparingly appressed strigose nerves, occasionally the primary reticulations with very few short pro- cesses; nerves 5, distinct, the inner pair reaching the apex, the marginal pair evanescent above the upper two-thirds. Inflores- cences terminal, usually 3-flowered, the pedicels 1 to 1.5 cm long; bracts lanceolate, acuminate, strigose, 5 to 7 mm long. Calyx- tube ovoid, 7 to 8 mm long, conspicuously penicillate, the setae simple, spreading, 2 to 3 mm long, somewhat scattered; calyx lobes 5, as long as the tube, narrowly oblong, setose. Petals purplish, obovate, about 1.5 cm long, their apices sparingly peni- cillate. Stamens 10, equal; filaments 6 mm long; anthers lan- ceolate, 5 mm long, the anterior basal appendages nearly round, glandlike, the connectives not at all produced. LUZON, Ilocos Norte Province, between Bangui and Claveria, Bur. Sci. 33079 Ramos, August 29, 1918, on forested slopes near small streams at low altitudes. This species is readily distinguished from its congeners by its appressed-strigose branches, branchlets, and leaves, the latter being nearly glabrous on the lower surface except for the sparingly strigose nerves. ASTRONIA Noronha ASTRONIA BRACHYBOTRYS sp. nov. Frutex vel arbor parva, ramulis junioribus et inflorescentiis leviter castaneo-furfuraceis; foliis coriaceis, rigidis, viridibus, oblongo-ellipticis, minute verruculosis, 5 ad 6 cm longis, utrinque subaequaliter angustatis, basi 3-nerviis, apice acuminatis; pani- culis brevissime pedunculatis, 2 ad 3 cm longis, paucifloris, ramis usque ad 1 cm longis; floribus ¢ urceolatis, calycis tubo 2.5 mm longo, lobis reniformibus. A shrub or small tree, the very young branchlets and inflores- cences somewhat castaneous-furfuraceous, otherwise glabrous, the branches and branchlets terete, rather slender. Leaves cori- aceous, rigid, oblong-elliptic, green on both surfaces, usually brownish-olivaceous when dry and minutely verruculose, 5 to 6 cm long, 2 to 3 em wide, subequally narrowed to the acute base and to the distinctly acuminate apex, base prominently 3-nerved and with an additional very slender marginal pair, the latter not more prominent than are the transverse nervules, the pri- XIV, 4 Merrill: Noteworthy Philippine Plants, XV 435 mary lateral nerves anastomosing with the midrib in the acu- men, when young the nerves beneath with few, scattered, furfuraceous-castaneous scales, soon becoming glabrous; petioles 1.5 to 2 cm long, glabrous. Panicles terminal, shortly pedun- cled, rather few-flowered, 2 to 3 cm long, the primary branches few, 1 cm long or less, the flowers somewhat crowded at the apices of the branchlets. Perfect flowers shortly pedicelled, ur- ceolate, the calyx-tube about 2.5 mm long, the limb spreading and about 4.5 mm in diameter, the lobes reniform. Petals obo- vate, about 3 mm long. Staminate flowers similar to the pistil- late ones but the calyx-tube broadly cup-shaped. LUZON, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 83279 (type), 33278, 33266 Ramos, August 21, 1918, in forests, altitude about 1,000 meters. This species is allied to Astronia pachyphylla Merr., A. ver- ruculosa Merr., and A. pauciflora Merr., being distinguished from the first by its smaller, differently shaped leaves and its castaneous-furfuraceous inflorescences and it further lacks the characteristic large bracts of A. pachyphylla; the leaves and flowers are much smaller than are those of A. verruculosa; while A. pauciflora is distinguished, among other characters, by its densely lepidote branchlets and inflorescences. ARALIACEAE SCHEFFLERA Forster SCHEFFLERA OBTUSIFOLIA sp. nov. § Euscheffiera, Heptapleurum. Frutex erectus, ramulis inflorescentiisque exceptis glaber; foliis brevissime petiolatis, palmatim 7- ad 9-foliolatis, foliolis coriaceis, oblongo-obovatis, integris, 8 ad 15 cm longis, apice plerumque rotundatis, basi acutis; inflorescentiis breviter pedun- culatis, ramis primariis racemose dispositis, 18 ad 24 cm longis, leviter pubescentibus; umbellulis numerosis, racemose dispositis, 8- ad 12-floris pedunculis usque ad 1 cm longis; floribus 5- meris; fructibus ellipsoideis ad oblongo-ellipsoideis, 5 mm longis, 5-sulcatis. An erect shrub, about 2 m high, glabrous except the some- what pubescent branchlets and slightly pubescent inflorescences. Leaves very short-petioled, 7- to 9-foliolate; leaflets coriaceous, shining when dry, oblong-obovate, entire, 8 to 15 cm long, 4 to. 6.5 em wide, the apex usually rounded, narrowed below to the acute base; primary lateral nerves 7 to 10 on each side of the midrib, the nerves and reticulations distinct on both surfaces; -petiolules 3.5 to 8 cm long; petioles stout, 1 to 1.5 cm long. In- 436 The Philippine Journal of Science 1919 florescence terminal, shortly peduncled, the peduncle and axis 4 to 7 cm long, the former glabrous or nearly so, the bracts subtending the branches deciduous; primary branches 8 to 15, racemosely arranged, 18 to 24 cm long, slightly pubescent, spreading-ascending; umbels numerous, racemosely arranged on the primary branches, 8- to 12-flowered, their peduncles up to 1 cm in length; the pedicels in flower, 2 to 3 mm long, in fruit somewhat longer. Flowers 5-merous, white. Calyx obconic, truncate, about 2.5 mm in diameter. Petals 5, ovate, acute, 2 mm long, somewhat 3-nerved. Filaments 3.5 to 4 mm long. Fruits ellipsoid to oblong-ellipsoid, about 5 mm long, 5-angled or sulcate, 5-celled. LUZON, Ilocos Norte Province, Burgos, Bur. Sci. 32912 (type), 32901 Ramos, July, 1918, on rocks in dry thickets at low alti- tudes. Bur. Sci. 33207 Ramos from Mount Nagapatan, Ilocos Norte Province, Luzon, perhaps represents a form of the same species but the leaflets are smaller, while the petioles vary from 2 to 6 cm in length; this number was collected at an altitude of about 800 meters, growing on tree trunks. This species is most closely allied to Schefflera brevipes Merr. and is one of the very few Philippine species with short petioles. It is distinguished from S. brevipes by its smaller and much fewer-nerved leaflets; its much shorter inflorescence-axis, and rather long-pedicelled, never subsessile, umbels. SCHEFFLERA ALVAREZII sp. nov. § Euschefilera, Heptapleurum. Frutex scandens, glaber, ramulis tenuibus; foliis 6-foliolatis, foliolis oblongis ad oblongo-obovatis, integris, subcoriaceis, usque ad 9 cm longis, apice tenuiter acuminatis, basi acutis, nervis utrinque 6 ad 8, subpatulis, perspicuis; inflorescentiis termina- libus, ramis primariis circiter 4, usque ad 25 cm longis; umbel- lulis circiter 5-floris, racemose dispositis; floribus 5-meris. An entirely glabrous vine, the ultimate branches 4 mm in diameter or less. Leaves palmately 6-foliolate; leaflets subco- riaceous, oblong to oblong-obovate, entire, pale when dry, shining, 6 to 9 cm long, 2 to 3.5 em wide, the apex slenderly acuminate, base acute; lateral nerves 6 to 8 on each side of the midrib, spreading or only slightly ascending, prominent on the lower surface as are the reticulations; petiolules 1 to 2.5 cm long; petioles 6 to 7 cm long. Inflorescence terminal, the rachis 2 em long or less, bearing usually 4 primary branches up to 25 em long. Flowers 5-merous, about 5 in each umbel, their pedi- cels about 3 mm long, the umbels widely scattered and racemosely XIV, 4 Merrill: Noteworthy Philippine Plants, XV 437 arranged on the primary branches, their peduncles 4 to 5 mm long. Calyx obovoid, truncate, about 1.2 mm long, petals oblong ovate, acute, 2.6 mm long, ovary 5-celled. MINDANAO, Lanao District, on the Lanao—Cotabato trail, For. Bur. 25240 Alvarez, March 24, 1916, altitude 1,100 meters. This species is manifestly allied to Schefflera caudatifolia Merr., from which it differs especially in its flowers being arranged in umbels rather than in fascicles; in its much shorter rachis and much longer primary branches; while the primary nerves _ of the leaves are spreading rather than ascending. SCHEFFLERA GLOBOSA sp. nov. § Fuscheffiera, Heptapleurwm. Frutex scandens, ramulis et infructescentiis stellato-tomen- tosis, indumento brunneo, ramis incrassatis; foliis 6-foliolatis, longissime petiolatis, foliolis oblongis, firmiter chartaceis, inte- gris, usque ad 44 cm longis, apice tenuiter acuminatis; inflores- centiis terminalibus, rhachibus leviter productis, ramis primariis ~ 3 ad 5, usque ad 20 cm longis, racemose dispositis; fructibus globosis, esulcatis, 5-locellatis, 5 mm diametro, umbellulis race- mose dispositis, breviter pedunculatis, fructibus confertis brevis- sime pedicellatis. A scandent shrub, glabrous except the young branchlets, bracts, and infructescences, which are brown stellate-tomentose, the ultimate branches up to 1 cm in diameter. Leaves palmately 6-foliolate, their petioles inflated and clasping at the base, stout, 30 to 35 cm long; leaflets oblong, firmly chartaceous, entire, gla- brous, shiny, 22 to 44 cm long, 7 to 10 cm wide, the base rounded to subacute, the apex slenderly acuminate; lateral nerves 15 to 20 on each side of the midrib, distinct as are the reticulations, anastomosing, petiolules 5 to11cmlong. Inflorescence terminal, the rachis somewhat produced, the primary branches 3 to 5, 16 to 20 cm long, racemosely disposed, each subtended by an oblong- lanceolate, acuminate, subcoriaceous bract about 3 cm in length. Fruits globose, about 5 mm in diameter, very slightly furfura- ceous, not sulcate, 5-celled, terminated by the 5 nearly sessile stigmas, arranged in short-peduncled, scattered umbels, 9 to 10 fruits crowded on each umbel, their pedicels 1 to 2 mm long, the peduncles 2 to 7 mm long, the pedicels and peduncles with the same type of indumentum as the rachis. PANAY, Capiz Province, Jamindan, Bur. Sci. 30897 Ramos & Edafio, May 21, 1918, on tree trunks in forests along streams. This species in vegetative characters strongly resembles Schef- flera digitata (Blanco) Merr. (S. machranta Merr.), but differs radically in its very short pedicels and 5-celled fruits. The ® 438 The Philippine Journal of Science 1919 brown stellate-tomentose indumentum on the inflorescences; the globose, nonsulcate, short-pedicelled, crowded fruits; and the very long petioles are characteristic. SCHEFFLERA PLATYPHYLLA sp. nov. § Euschefflera, Heptapleurum. Frutex scandens, inflorescentiis brunneo-tomentosis; foliis 5-foliolatis, late oblongo-ovatis, integris, usque ad 34 ecm longis et 17 cm latis, apice abrupte acuminatis; inflorescentiis termina- libus, ramis primariis circiter 12, usque ad 40 cm longis, racemose dispositis; floribus minutis, 5-meris, numerosis, confertis, um- bellulis globosis, circiter 6 mm diametro, valde numerosis, race- ‘ mose dispositis, pedunculis 5 ad 10 mm longis. A vine, glabrous except the rather densely brown-tomentose inflorescence, the indumentum obscurely stellate, the ultimate branches about 1 cm in diameter. Leaves palmately 5-foliolate, their petioles about 15 cm long; leaflets broadly oblong-ovate, coriaceous, entire, 30 to 34 cm long, 15 to 17 cm wide, the upper surface olivaceous, shining, the lower somewhat brownish when dry, the base broadly rounded, the apex rather abruptly acumi- nate, the acumen up to 2 cm in length; lateral nerves about 25 on each side of the midrib, distinct as are the reticulations; petiolules stout, 5 to 8 mm long. Inflorescence terminal, the rachis stout, up to 8 cm long, bearing about 12 elongated, race- mosely arranged, primary branches up to 40 cm in length, the subtending bracts oblong-ovate, up to 2.5 cm long. Flowers minute, white, 5-merous, crowded in globose umbels about 6 mm in diameter, 30 to 40 flowers in an umbel, the umbels racemosely arranged on the primary branches, abaut 75 on each branch, their peduncles 5 to 10 mm long; pedicels 1 to 2 mm long. Calyx pubescent, turbinate, truncate, about 1.2 mm long. Petals 5, oblong-ovate, 1.2 mm long. PANAY, Capiz Province, Jamindan, Bur. Sci. 30897 Ramos & Edano, April 17, 1918, on tree trunks in forests along streams. This species has an indumentum similar to that of Schefflera globosa Merr. and is manifestly allied to that species, from which it differs radically in its much larger, wider leaflets; in the nu- merous, elongated, primary branches of the inflorescence; and in its very numerous, densely many-flowered, longer-peduncled umbels. Both species apparently belong in the group with Schefflera clementis Merr. SCHEFFLERA SANTOSII sp. nov. § Euschefflera, Heptapleurum. Frutex scandens (vel erectus ?), glaber, ramis in siccitate rugosis; foliis 3- ad 5-foliolatis, brevissime petiolatis, petiolo 1 ad 2 cm longo; foliolis subcoriaceis, in siccitate olivaceis, nitidis, e XIV, 4 Merrill: Noteworthy Philippine Plants, XV 439 ellipticis ad oblongo-ellipticis, breviter acuminatis, integris, 6 ad 9 cm longis, nervis utrinque circiter 12, perspicuis; inflorescentiis terminalibus, rhachibus elongatis, ramis primariis 4 ad 8, race- mose dispositis, usque ad 20 cm longis; umbellis distantibus, racemose dispositis, 8- ad 12-floris; floribus parvis, 5-meris. A scandent (or erect ?), glabrous shrub, 3 to 4 m in length, the branches grayish, when dry wrinkled, the ultimate ones 5mm in diameter or less. Leaves 3- to 5-foliolate, their petioles rela- tively short, 1 to 2 cm long, inflated and clasping at the base. Leaflets subcoriaceous, entire, olivaceous, shining, elliptic to oblong-elliptic, 6 to 9 cm long, 3 to 5 cm wide, subequally nar- rowed to the acute base and to the shortly acuminate apex; petiolules 5 to 20 mm long; lateral nerves about 12 on each side of the midrib, distinct and somewhat projecting on the lower surface, anastomosing, spreading-ascending, the reticulations lax, distinct. Inflorescence terminal, the rachis 5 to 7 cm longy the primary branches 4 to 8, racemosely disposed, alternate, up to 20 cm in length. Umbels racemosely arranged on the primary branches, 8- to 12-flowered, their peduncles up to 12 mm in length, the pedicels 1 to 4 mm long. Flowers 5-merous, white or purplish, calyx about 2 mm in diameter. Petals 5, oblong-ovate, 2 mm long. Luzon, Benguet Subprovince, Pauai, Bur. Sct. 31722 Santos, June 1, 1918, in thickets near streams, altitude about 2,000 meters. This species is well characterized by its very short-petioled leaves, in this character resembling Schefflera brevipes Merr., but differing radically from that species in its much fewer and smaller leaflets. . SCHEFFLERA PANAYENSIS sp. nov. § Cephaloschefflera. Arbor, partibus junioribus plus minusve furfuraceis, foliis junioribus dense stellato-tomentosis, cito glaberrimis, ramis in- crassatis; foliis 10- ad 12-foliolate, longissime petiolatis, foliolis coriaceis, oblongis, usque ad 17 cm longis, tenuiter caudato-acu- minatis, nervis utrinque 10 ad 12, perspicuis; capitulis racemose dispositis, globosis, immaturis 1 ad 1.5 cm diametro, fructibus sulcatis, 5-locellatis. A tree, about 5 m high, the very young parts densely furfura- ceous, the very young leaves densely stellate-tomentose, soon becoming entirely glabrous. Ultimate branches about 1.5 cm in diameter, rugose, marked with very large petiolar scars. Leaves 10- to 12-foliolate, their petioles about 25 cm long; leafiets oblong, coriaceous, shining, 10 to 17 cm long, 3 to 3.5 cm wide, 440 The Philippine Journal of Science 1919 entire, base rounded to obtuse, apex slenderly caudate-acumi- nate ; lateral nerves 8 to 10 on each side of the midrib, anastomos- ing, distinct, as are the reticulations; petiolules 6 to 8 cm long. Branches of the inflorescence 40 to 45 cm long, stout, glabrous, bearing numerous, racemosely arranged, dense, globose, glabrous heads, which in young fruit are from 1 to 1.5 cm in diameter, their peduncles 1 to 1.5 em long. Young fruits 5-sulcate, 3 to 4 mm long, 5-celled. PANAY, Capiz Province, Mount Madiaas, Bur, Sci. 30736, Ramos & Edafio, May 12, 1918, in the mossy forest. This species resembles Schefflera catanduanensis, but has longer inflorescence branches and is furthermore very easily distin- guished by its peduncled heads. BOERLAGIODENDRON Harms BOERLAGIODENDRON CAUDATUM sp. nov. Frutex circiter 2 m altus; foliis longe petiolatis, palmatim 7- ad 9-foliolatis, foliolis nitidis, lanceolatis, valde inaequalibus, 10 ad 23 cm longis, tenuiter caudato-acuminatis, basi decurrento- acuminatis, majoribus plerumque utrinque 1- vel 2-lobatis, mar- gine irregulariter perspicue dentatis ; petiolo circiter 40 cm longo, basi cristis pectinatis 2 vel 3 instructis, dentibus deflexis; inflo- rescentiis sessilibus, circiter 15 cm diametro, radiis primariis circiter 25; capitulis ¢ circiter 1 cm diametro, floribus confertis brevissime pedicellatis, 4-meris. An erect shrub, about 2 m high, the leaf-bearing part of the stem at least 1 cm in diameter. Leaves long-petioled, palmately 7- to 9-foliolate; leaflets subcoriaceous, olivaceous and shining when dry, lanceolate, very unequal in size, 10 to 23 cm long, slenderly caudate-acuminate, base decurrent-acuminate, the larger ones usually deeply lobed, the lobes 1 or 2, rarely 3, ex- tending nearly to the midrib, the margins with irregular, rather prominent, usually incurved teeth; petiolules of the larger leaf- lets up to 3 cm in length, the smaller leaflets sometimes nearly sessile; petioles about 40 em long, 4 to 5 mm in diameter, the basal 1 cm supplied with 2 or 3 pectinate crests, their teeth defiexed up to 7 mm long. Inflorescence terminal, sessile, about 15 cm in diameter, composed of about 25 primary branches, each subtended by an oblong to oblong-ovate, coriaceous bract about 1.5 cm long, the primary branches about 4 cm long, glabrous, then dichotomously or trichotomously branched, the bracts subtending the branches oblong to oblong-lanceolate, about 1 cm long, some- what setose-furfuraceous, central branch 1 to 1.5 cm long, gla- brous, bearing a globose head of sterile flowers about 1 cm in XIV, 4 Merrill: Noteworthy Philippine Plants, XV 441 diameter, the bracteoles subtending the sterile flowers oblong, acute to obtuse, 4 to 5 mm long, coriaceous, somewhat keeled, more or less furfuraceous; lateral branches about 4 cm long, straight or somewhat geniculate at the middle, here supplied with a pair of coriaceous, somewhat setose-furfuraceous bracts, 4 to 5 mm in length; heads ellipsoid, dense, about 1 cm long, many- flowered, each flower subtended by an obovate 3 to 4 mm long bracteole, which is thickened in the median portion and rather prominently brown fimbriate-ciliate on the back and margins, and more or less inclosing the flower; pedicels 1 mm long or less. Calyx obovoid, 2 to 2.5 mm long, truncate, narrowed below, ob- scurely 4-angled by compression, the angles rounded. Ovary 4-celled. Luzon, Ilocos Norte Province, Mount Palimlin, Bur. Sci, 33308 Ramos, August 21, 1918, on forested slopes near the sum- mit, altitude about 1,000 meters. This strongly marked species is most closely allied to Boerlagio- dendron catanduanense Merr., from which it is easily distin- guished by its pectinate petiolar crests; its much longer petioles ; and its few-lobed caudate-acuminate leaflets, the lobes extending nearly to the midrib and ascending rather than divaricate. The inflorescence is also much larger and further differs in its fur- furaceous-setose bracts and very different sterile and fertile heads. ' ERICACEAE VACCINIUM Linnaeus VACCINIUM ILOCANUM sp. nov. Frutex epiphyticus, inflorescentiis exceptis glaber ; foliis ellip- ticis, crasse coriaceis, nitidis, integris, 6 ad 8 cm longis, basi acutis, apice breviter abrupteque acuminatis, nervis utrinque circiter 4, tenuibus, inferioribus adscendentibus; racemis axillar- ibus, solitariis, 2 ad 4 cm longis, pubescentibus; bracteis per- sistentibus, ovatis ad elliptico-ovatis, 5 ad 6 mm longis, acutis vel acuminatis; corolla glabris, 6 mm longa; filamentis barbatis; antheris oblongis, 1.1 mm longis. An epiphytic shrub, entirely glabrous except the inflorescences, the branchlets about 3 mm in diameter, reddish-brown when dry. Leaves elliptic, thickly coriaceous, olivaceous and shining when dry, entire, 6 to 8 cm long, 3.5 to 5 cm wide, margins somewhat recurved, the lower surface somewhat paler than the upper and distinctly glandular-punctate, the base acute, apex acute to shortly and abruptly acuminate; lateral nerves about 4 on each side of 442 The Philippine Journal of Science 1919 the midrib, slender, the lower two pairs from near the base ascending, reaching at least to the upper two-thirds of the leaf ; petioles very stout, reddish-brown, 5 to 8 mm long. Racemes axillary, and sometimes terminating short lateral branches, 2 to 4 cm long, the rachis, bracts, pedicels, and calyces more or less - pubescent with short spreading hairs; pedicels 3 to 4 mm long; bracts persistent, ovate to elliptic-ovate, 5 to 6 mm long, usually acute, sometimes slightly acuminate. Calyx-tube almost obsolete, the lobes triangular, acute, pubescent, 1.5 mm long. Corolla oblong-ovoid, glabrous, narrowed upward, about 6 mm long, 3.5 mm in diameter below the middle, red, the lobes short, broadly ovate, recurved. Filaments densely bearded, about 1 mm long; anthers oblong, about 1.1 mm long, the terminal tubes very short, not narrowed, opening by oblique pairs. Style stout, glabrous, 3mm long. Top of the ovary very slightly pubescent. Luzon, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 33372 Ramos, August 21, 1918, growing on trees on forested slopes near the summit, altitude about 3,100 meters. The alliance of this species is manifestly with Vaccinium platyphyllum Merr., from which, among other characters, it is distinguished by its smaller, fewer-nerved leaves, shorter pedi- cels, and only slightly pubescent top of the ovary. DIPLYCOSIA Blume DIPLYCOSIA GLABRA sp. nov. Frutex epiphyticus, glaber, vel ramulis junioribus parcissime setosis ; foliis ellipticis ad oblong-ellipticis, acutis vel brevissime apiculatis, basi acutis, usque ad 6 cm longis, in siccitate olivaceo- brunneis vel pallidis, nitidis, nervis utrinque 2, tenuibus, supra leviter impressis, subtus obsoletis; floribus axillaribus, solitariis vel fasciculatis, pedicellatis, bracteolis late ovatis, 1.5 mm longis, margine leviter ciliatis, calycis circiter 2 mm longis, lobis late ovatis, obtusis, 1.5 mm longis. A glabrous, erect, epiphytic shrub, attaining a height of from 3 to 4 m, the very young branchlets sometimes with a very few slender setae. Leaves coriaceous, elliptic to oblong-elliptic, 2 to 6 cm long, 1 to 2.5 em wide, brownish-olivaceous or pale when dry, shining, the lower surface with small scattered, brownish glands, the base acute, the apex usually acute and ending in a short, stout, blunt apiculus; lateral nerves 2 on each side of the midrib, slender, slightly impressed on the upper surface, obsolete on the lower surface, curved, the upper pair leaving the midrib at or near the middle of the leaf, the reticulations obsolete; petioles XIV, 4 Merrill: Noteworthy Philippine Plants, XV 443 about 2mm long. Flowers axillary, solitary or few in a fascicle, their pedicels about 5 mm long, the apical bracteoles connate, broadly ovate, acute or slightly acuminate, margins sparingly ciliate, 1.5 mm long. Calyx about 2 mm long, the lobes broadly ovate, obtuse, 1.5 mm long, their margins slightly ciliate. Corolla not seen. Fruit soft, fleshy, bluish-black, when dry brownish or black, about 5 mm long. LUZON, Benguet Subprovince, Pauai, Bur. Sct. 31784 Santos April. 22, 1918 (type), Bur. Sci. 8509 McGregor, June, 1909, For. Bur. 14434 Darling, January, 1909, Copeland, s. n., May, 1913, in the mossy forest, altitude about 2,300 meters. Igorot name dugui-is. Among the Philippine species this is most closely allied to Diplycosia calelanensis Elm., from which it is readily distin- guished by its nerves being obsolete on the lower surface of the leaves. Among the extra-Philippine species its alliance is mani- festly with the Javan Diplycosia heterophylla Blume, but its leaves are different in shape, never acuminate, with shorter petioles, while the venation is distinctly different. MYRSINACEAE ARDISIA Swartz ARDISIA ILOCANA sp. nov. § Akosmos. Frutex, ramulis inflorescentiisque castaneo-pubescentibus ex- ceptis glaber; foliis oblongis, membranaceis ad chartaceis, 6 ad 12 cm longis, basi acutis, apice leviter acuminatis, integris, utrinque glandulis perspicuis instructis, nervis utrinque 15 ad 20, tenuibus, subobscuris, reticulis obsoletis ; inflorescentiis bipin- natim paniculatis, 8 ad 10 cm longis, pedunculatis, plerumque e axillis defoliatis; floribus racemose dispositis, circiter 8 mm diametro, calycis lobis orbicularibus, leviter ciliatis, petalis punc- tato-glandulosis, leviter acuminatis. An erect shrub, the branchlets and inflorescences distinctly castaneous-pubescent, otherwise glabrous. Leaves oblong, mem- branaceous to chartaceous, 6 to 12 cm long, 2 to 3.5 cm wide, base acute, apex somewhat acuminate, somewhat olivaceous and slightly shining when dry, both surfaces with numerous, rela- tively large glands which are distinctly visible to the naked eye and which occur at fairly regular intervals on all parts of the leaf; lateral nerves 15 to 20 on each side of the midrib, very slender, not prominent, obscurely anastomosing, the reticulations obso- lete or nearly so; petioles pubescent when young, soon becoming 444 The Philippine Journal of Science 1919 glabrous, 5 to 10 mm long. Panicles lateral, from the ultimate branchlets, axillary and in the axils of fallen leaves, peduncled, bipinnate, 8 to 10 cm long, the flowers racemosely arranged on the primary branches, their pedicels 3 to 4mm long. Flowers white, about 8 mm in diameter. Calyx lobes orbicular, rounded, glan- dular-punctate, margins ciliate. Corolla lobes oblong-elliptic, glandular-punctate, glabrous, 4 mm long, somewhat acuminate. Anthers lanceolate, acuminate, 2.56 mm long, the connectives slightly glandular. LUZON, Ilocos Norte Province, between Bangui and Claveria, Bur. Sct. 32982 Ramos, August 12, 1918, altitude about 500 meters. This species is well characterized by its numerous, relatively large, regularly spaced glands, which are distinctly visible to the naked eye on both surfaces of the leaf. Its alliance is with Ardisia racemoso-panniculata Mez, from which it is distinguished not only by its very evident glands but also by its very slender obscure nerves and practically obsolete reticulations. ARDISIA LONGIPETIOLATA sp. nov. § Pyrgus. Frutex glaber, ramis 6 ad 8 mm diametro; foliis confertis, anguste oblongis ad oblongo-ellipticis, usque ad 30 cm longis, integris vel obscurissime denticulatis, obtusis vel acutis, per- spicue glanduloso-punctatis, nervis utrinque circiter 12; petiolo 4 ad 7 cm longo; paniculis circiter 30 cm longis, ramis inferiori- bus 9 ad 12 cm longis, floribus subumbellatim dispositis, sepalis punctatis, margine ciliatis; fructibus circiter 1 cm diametro. A glabrous shrub, the branches terete, reddish-brown, 6 to 8 mm in diameter. Leaves pseudoverticillate, usually about 6 crowded at the end of each branch, narrowly oblong to oblong- elliptic or oblong-oblanceolate, 23 to 30 cm long, 6 to 10 cm wide, entire, or very obscurely toothed, apex obtuse to acute, base rounded, acute, or somewhat acuminate, when dry grayish or greenish olivaceous, shining, the lower surface rather densely glandular-puncticulate with dark-colored glands; lateral nerves usually about 12 on each side of the midrib, prominent, curved, anastomosing, the reticulations distinct; petioles 4 to 7 cm long, somewhat triangular. Panicles terminal, peduncled, about 30 cm long, the lower primary branches 9 to 12 cm long, the flowers subumbellately arranged at the tips of the primary branches. Pedicels about 1.5 cm long, thickened upward. Persistent calyx about 7 mm in diameter, the lobes ovate, acute, about 3 mm long, glandular-punctate, the margins ciliate. Fruits gla- XIV, 4 Merrill: Noteworthy Philippine Plants, XV 445 brous, globose, slightly glandular, red when mature, about 1 cm in diameter. Luzon, Camarines Province, Bur. Sci. 33492 (type), 33554 Ramos & Edafo, December, 1918, in damp forests at low alti- tudes between Paracale and Mambulao. This species belongs in the group with Ardisia perrotietiana A. DC., from which it is readily distinguished by its larger leaves and elongated petioles. AMBYLANTHOPSIS Mez AMBYLANTHOPSIS CRASSIFOLIA sp. nov. Frutex erectus, glaber, ramulis incrassatis, cylindricis, cir- citer 1 cm diametro, cicatricibus multis ornatis; foliis crassis- sime coriaceis, integris, oblongis ad anguste oblongo-obovatis, 10 ad 18 cm longis, minute punctatis, obtusis, nervis utrinque 10, haud perspicuis, admodum_ subobsoletis; inflorescentiis axillaribus, plerumque racemosis, usque ad 7 cm longis; flori- bus 4-5-meris, 6 ad 7 diametro, petalis ellipticis ad oblongo- ellipticis, filamentis petalis aequilongis. An erect glabrous shrub, about 2 m high, the ultimate branch- lets much thickened, cylindric, about 1 cm in diameter and marked with numerous, large, petiolar scars. Leaves very thickly coriaceous, entire, oblong to narrowly oblong-obovate, 10 to 18 cm long, 3.5 to 4 cm wide, obscurely and minutely glandular-punctate, the apex obtuse, narrowed below to the acuminate base; midrib very stout, the lateral nerves about 10 on each side of the midrib, not prominent, sometimes nearly obsolete; petioles stout, 6 to 15 mm long. Inflorescences in the upper axil, up to 7 cm in length, the flowers usually race- mosely arranged, sometimes in very depauperate panicles. Flowers red, 4- and 5-merous, 6 to 7 mm in diameter, their pedicels about 2 mm long. Calyx about 4 mm in diameter, the lobes oblong-ovate, obtuse, somewhat glandular-punctate, gla- brous. Corolla lobes elliptic to elliptic-oblong, obtuse, 4 mm - long, somewhat punctate, united for the lower 1 mm. Filaments as long as the petals, the anthers longitudinally dehiscent. Ovary and style 2.5 mm long, the ovary ovoid, the style as long as the ovary. Luzon, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 33288 (type), 33381 Ramos, August 21, 1918, on forested slopes, altitude about 1,000 meters. This species is entirely different from the other described forms of the genus Ambylanthopsis, but by definition must be 1653686 446 The Philippine Journal of Science 1919 placed in this genus unless a new one be proposed for it. It cannot be placed in the genus Discocalyx on account of its elongated filaments, which equal the corolla lobes in length. In the type specimen the much thickened, prominently scarred, ultimate branchlets attain a length of at least 15 cm and are distinctly thicker than the smooth or nearly smooth branches which bear them. The species is otherwise very prominently characterized by its very thickly coriaceous, entire, obscurely nerved leaves, which are crowded toward the apices of the branchlets and by its usually racemose inflorescences. - SAPOTACEAE PALAQUIUM Blanco PALAQUIUM GLABRIFOLIUM sp. nov. Arbor novellis et floribus exceptis glabra; foliis subcoriaceis, anguste oblongo-obovatis, usque ad 14 cm longis, apice perspicue acuminatis, basi angustatis, cuneatis vel decurrento-acuminatis, nervis utrinque circiter 20, tenuibus; floribus solitariis vel fas- ciculatis, sepalis coriaceis, exterioribus glabris, interioribus leviter pubescentibus, circiter 3.5 mm longis; ovario puberulo. A tree, 6 to 8 m high, entirely glabrous except the growing tips of the branchlets, the inner sepals, and the ovary. Leaves subcoriaceous, rigid, dark-brown when dry, shining, the lower surface paler than the upper, narrowly oblong-obovate, 10 to 14 cm long, 3.5 to 5 em wide, apex rather conspicuously acu- minate, the acumen blunt, narrowed in the lower one-half to the cuneate or somewhat decurrent-acuminate base; primary lateral nerves slender, about 20 on each side of the midrib, anastomosing, slightly more conspicuous than are the secondary nerves and reticulations; petioles 1.5 to 2 cm long. Flowers solitary or fascicled, from the axils of fallen leaves, their pedi- cels about 1 cm long, in age entirely glabrous. Outer sepals ' glabrous, thickly coriaceous, somewhat rugose when dry, very broadly ovate, 2.5 to 3 mm long, about 4 mm wide, the inner sepals ovate, obtuse or acute, about 3.5 mm long, slightly pubes- cent on the median portion, the margins thin, glabrous. Sta- mens and corolla not seen. Ovary slightly puberulent, 6-celled; style about 9 mm long, glabrous. PANAY, Capiz Province, Mount Macosolon, Bur. Sci. 30820 Ramos & Edaio, April 22, 1918, on forested slopes. This species is most closely allied to Palaquium dubardii Elm., from which it is distinguished by its conspicuously acuminate XIV, 4 Merrill: Noteworthy Philippine Plants, XV 447 leaves, which are much narrowed below, and by its much more numerous nerves. The characteristic indumentum of the genus is entirely wanting except on the growing tips of the branch- lets and on parts of the flowers. The fruits with the specimens are globose or ovoid, about 2.5 cm in diameter, but are abnormal in that when young some insect has deposited its eggs in them and the developing larvae have destroyed the seeds, the entire fruit, except the parts destroyed by the larvae, being filled with a dense, rather brittle, somewhat woody substance. SYMPLOCACEAE SYMPLOCOS Jacquin SYMPLOCOS BRACHYBOTRYS sp. nov. Arbor parva, glaberrima;:foliis coriaceis, integris, oblan- ceolatis, 5 ad 8 cm longis, obtuse acuminatis, deorsum angustatis, basi cuneatis, nervis utrinque 15 ad 18, tenuibus; racemis axillaribus, 1 ad 1.5 cm longis; floribus brevissime pedicellatis, circiter 8 mm diametro, calycis tubo circiter 1 mm longo, lobis orbiculari-ovatis, quam tubo brevioribus; fructibus cylindraceis, oblongis, circiter 1 cm longis. An entirely glabrous tree, about 5m high. Leaves coriaceous, entire, oblanceolate, 5 to 8 cm long, 1.5 to 2.5 cm wide, blunt- acuminate, much narrowed below to the cuneate base, suboliva- ceous, shining; lateral nerves slender, 15 to 18 on each side of the midrib, anastomosing, the reticulations lax, distinct; petioles 1 cm long or less. Racemes axillary, few-flowered, black when dry, 1 to 1.5 cm long. Flowers about 8 mm in diameter, pedicels 1 mm long or less, the subtending bracteoles elliptic-ovate, deciduous, about 1 mm long. Calyx-tube some- what narrowed below, cylindric, 1 mm long, the lobes 5, some- what spreading, orbicular-ovate, rounded, about 0.8 mm long, the inner ones slightly smaller than the outer three. Petals oblong to oblong-elliptic, obtuse, about 4 mm long. Stamens numerous, free, as long as the petals. Fruit oblong, nearly terete, about 1 cm long and 5 mm in diameter. LUZON, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 33277 Ramos, August 22, 1918, on forested slopes, altitude about 900 meters. _ This species closely resembles Symplocos purpurascens Brand and is apparently closely allied to that species. It is, however, very easily distinguished by its very short inflorescences and its much more numerous lateral nerves. 2 448 The Philippine Journal of Science 1919 LOGANIACEAE GENIOSTOMA Forster GENIOSTOMA PACHYPHYLLUM sp. nov. Frutex vel arbor parva, glabra, ramulis et foliis in siccitate nigris; foliis coriaceis vel subcoriaceis, anguste oblongis ad oblongo-lanceolatis, 5 ad 8 cm longis, basi acutis, apice acutis vel obscure acuminatis, nervis utrinque 5 vel 6, tenuibus, reti- culis subobsoletis; fructibus axillaribus, solitariis, ovoideis vel ellipsoideis, circiter 7 mm longis, breviter pedicellatis. A glabrous shrub or small tree, the branches terete, straw- colored, the young branchlets black when dry as are the leaves and petioles. Leaves coriaceous or subcoriaceous, brittle, nar- rowly oblong to oblong-lanceolate, 5 to 8 cm long, 1.2 to 2.5 cm wide, subequally narrowed to the acute base and to the acute or slightly acuminate apex; lateral nerves slender, 5 or 6 on each side of the midrib, obscurely anastomosing, the reticula- tions obsolete or nearly so; petioles about 5 mm long; stipules _ short, truncate. Fruits axillary, solitary, ovoid or ellipsoid, about 7 mm long, black when dry, their pedicels about 5 mm long; persistent sepals broadly ovate, rounded, their margins slightly ciliate. LuzON, Ilocos Norte Province, Mount Palimlim, Bur. Sct. 83316 Ramos, August 21, 1918, on forested slopes, altitude about 1,000 meters. This species is distinguished among the Philippine forms by its glabrous, slenderly nerved, brittle, coriaceous or subcoriaceous leaves, the reticulations being obsolete or nearly so, and by its solitary, axillary fruits. The whole plant, except the branches, is characteristically black when dry. APOCYNACEAE ALYXIA Banks ALYXIA RETUSA sp. nov. Frutex scandens, inflorescentiis leviter pubescentibus exceptis glaber, ramulis incrassatis, 5-angulatis; foliis verticillatis, 5- natis, crassissime coriaceis, obovatis ad oblongo-obovatis, 5 ad 8 em longis, apice late rotundatis retusisque, margine revolutis, nervis tenuibus, indistinctis; cymis axillaribus, breviter pedun- culatis, circiter 3 cm longis; fructibus usque ad 3 cm longis, partibus seminiferis solitariis vel binis, ellipsoideis, 10 ad 12 mm longis, 8 mm diametro. A scandent shrub, glabrous except the sparingly pubescent inflorescence, the branches stout, up to 8 mm in diameter, rugose a XIV, 4 Merrill: Noteworthy Philippine Plants, XV 449 when dry, the younger ones distinctly 5-angled. Leaves verti- cillate, 5 at each node, very thickly coriaceous, obovate to oblong- obovate, 5 to 8 cm long, 2.5 to 4 cm wide, the apex broadly rounded and retuse, base acute to decurrent-acuminate, margins somewhat revolute, the midrib impressed on the upper surface, very prominent on the lower surface; lateral nerves numerous, slender, indistinct; petioles 1 to 2 cm long. Cymes axillary, somewhat pubescent, shortly peduncled, about 3 cm long, ex- cluding the fruits. Flowers apparently few, the persistent sepals ovate to oblong-ovate, obtuse, slightly pubescent, coriaceous, about 2.5 mm long. Fruits shortly pedicelled, either 3 cm long and constricted in the middle, with two ellipsoid seed-bearing parts, or not constricted and consisting of a single seed-bearing part, the seed-bearing parts 10 to 12 mm long, about 8 mm in diameter. LUZON, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 33330 Ramos, August 21, 1918, on forested slopes, altitude about 1,000 meters. This species is strongly characterized among those with very thick coriaceous leaves by the leaves being broadly rounded and retuse at their apices. RAUWOLFIA Linnaeus RAUWOLFIA MEMBRANACEA sp. nov. Frutex erectus glaber; foliis membranaceis, oblongis ad oblongo-lanceolatis, acuminatis, 7 ad 11 cm longis, nervis utrinque circiter 10, tenuibus; cymis axillaribus terminalibusque, paucifloris, 4 ad 6 cm longis, fructibus ellipsoideis, leviter com- pressis, 1 cm longis, in siccitate rugosis. An erect glabrous shrub. Leaves membranaceous, olivaceous, oblong to oblong-lanceolate, slenderly acuminate, base acute, oli- vaceous and shining when dry, 7 to 11 cm long, 2 to 3 cm wide; lateral nerves slender, about 10 on each side of the midrib, petioles 7 to 12 mm long. Cymes axillary and terminal, pedun- cled, 4 to 6 cm long, apparently few-flowered, the persistent calyx-lobes lanceolate, acuminate, about 2 mm long. Fruits somewhat ellipsoid, slightly compressed when dry, about 1 cm long and 6 mm wide, somewhat narrowed to the acute base and usually obtuse apex, rugose when dry. Luzon, Ilocos Norte Province, Mount Nagapatan, Bur. Sci. 83214 Ramos, August 9, 1918, on dry slopes, altitude about 700 meters. This species is not closely allied to any previously described 450 The Philippine Journal of Science 1919 Philippine form. It is well characterized by its membranaceous, slenderly acuminate, few-nerved leaves. TABERNAEMONTANA Linnaeus TABERNAEMONTANA ECARINATA sp. nov. Frutex glaber; foliis charteceis, olivaceis, oblongis ad oblongo- ellipticis, 7 ad 14 cm longis, apice late obtuseque acuminatis, nervis utrinque 8 ad 10; infructescentiis circiter 6 cm longis, haud ramosis, folliculis 2 gerentibus; folliculis 2.5 ad 3 cm longis, inaequilateralibus, ecarinatis, apice rostratis; seminibus circiter 18. A glabrous shrub, the branches slender, terete. Leaves char- taceous, olivaceous, shining, equal, oblong to oblong-elliptic or oblong-lanceolate, 7 to 14 cm long, 3 to 7 cm wide, base often slightly inequilateral, somewhat acuminate, apex broadly and obtusely acuminate; lateral nerves 8 to 10 on each side of the midrib, distinct, anastomosing, the reticulations slender, lax; petioles about 5 mm long. Inflorescences apparently very few- flowered, the peduncles in fruit up to 6 cm long, simple, each bearing a single pair of follicles but usually presenting one or two scars of fallen pedicels. Persistent sepals ovate, about 2.5 mm long. Corolla-tube about 12 mm long, the lobes spreading, narrow, about 1.5 cm long. Follicles yellow, 2.5 to 3 cm long, 1.5 to 2 cm wide, inequilateral, smooth, not at all keeled, some- what compressed, one side nearly straight, the other semi- circular, apex rather prominently rostrate. Seeds up to 18 in each follicle, irregular, 7 to 8 mm long. LuzoN, Ilocos Norte Province, between Bangui and Claveria, Bur. Sci. 33022 (type), 33082 Ramos, August, 1918, in forests along small streams at low altitudes. This species is strongly characterized by its rostrate, rather many-seeded follicles, which are not at all keeled. In vegetative characters it resembles some forms of Tabernaemontana panda- caqui Poir. but differs entirely in its fruit characters. CONVOLVULACEAE IPOMOEA Linnaeus IPOMOEA POLYMORPHA R. & S. Syst. 4 (1819) 254. Ipomoea heterophylla R. Br. Prodr. (1810) 487, non Ortega, 1800. Convolvulus brownii Spreng. Syst. 1 (1825) 612. Convulvulus robertianus Spreng. op. cit. 5 (1828) 192. Luzon, Ilocos Norte Province, Burgos, Bur. Sci. 32945 Ramos, July 25, 1918, in open dry places at low altitudes: Union Prov- XIV, 4 Merrill: Noteworthy Philippine Plants, XV 451 ince, San Fernando, Lete 108, 175, July and September, 1916, in open sandy places near sea level, with the Ilocano name camarin. This species has been previously reported only from Austra- lia and Formosa, but the Philippine specimens cited above agree very closely with the descriptions and with Australian material kindly supplied by Mr. J. H. Maiden. In vegetative characters the Philippine specimens present great variation, as do the Aus- tralian ones. Iam further of the opinion that Henry 1112 from Formosa represents a somewhat dwarfed form of the same species. My specimen of this, identified as Ipomoea sp., is but about 10 cm high and has unusually narrow leaves. In floral characters, however, it closely approximates the Luzon form. VERBENACEAE CALLICARPA Linnaeus CALLICARPA OBTUSIFOLIA sp. nov. Frutex, ramulis et petiolis et inflorescentiis et subtus foliis uniformiter denseque cinereo-stellato-tomentosus; foliis ellipticis ad oblongo-ellipticis, supra glabris, subcoriaceis, usque ad 8 cm longis, apice rotundatis, obtusis, vel subacutis, margine denti- culatis, nervis utrinque 5 ad 7, perspicuis; cymis pedunculatis, dichotomis, usque ad 2.5 cm latis; floribus numerosis, circiter 2.5 mm longis, calycis extus dense albido-stellato-tomentosis. A shrub, the branchlets, petioles, inflorescences, and lower sur- face of the leaves densely and uniformly cinereous-stellate-pu- bescent, the indumentum covering the entire surface. Branches terete, pale-brownish, glabrous. Leaves elliptic to oblong-elliptic, subcoriaceous, 5 to 8 cm long, 2.5 to 4 em wide, the apex rounded, obtuse, or sometimes subacute, base usually obtuse, margins entire below, in the upper part distinctly denticulate, the upper surface brownish-olivaceous, glabrous or when young stellate- pubescent along the midrib; lateral nerves 5 to 7 on each side of © the midrib, curved, distinct as are the primary reticulations; petioles 5 to 10 mm long. Cymes axillary, peduncled, dichot- omous, up to 2.5 cm wide, the peduncles about 1.5 cm long; bracts linear-lanceolate, acuminate, 2 to 2.5 mm long; pedicels 0.5 mm long or less. Flowers rather crowded, pink. Calyx cup-shaped to obconic, about 1.6 mm long, densely stellate-pu- bescent, the teeth 4, short. Corolla glabrous, 2.5 mm long, the lobes equal, orbicular-ovate, rounded, nearly 1 mm in diameter. Filaments and style 5 to 6 mm long. Fruit globose, dark-brown and rugose when dry, about 2 mm in diameter. 452 The Philippine Journal of Science 1919 LUZON, Ilocos Norte Province, Burgos, Bur. Sci. 32921 Ramos, July 27, 1918, in dry thickets at low altitudes, Ilocano anayop. The alliance of this species is manifestly with Callicarpa blan- cot Rolfe, from which it is especially distinguished by its elliptic to oblong-elliptic, usually rounded or obtuse, never acuminate leaves. GESNERIACEAE CYRTANDRA Forster CYRTANDRA ILOCANA sp. nov. Frutex glaberrimus; foliis chartaceis, integris, oblongis ad oblongo-obovatis vel oblongo-oblanceolatis, 3 ad 7 cm longis, in siccitate castaneis, basi acutis, apice acuminatis, nervis utrinque 5 ad 7, tenuibus; inflorescentiis tenuibus 1- vel paucifloris, pe- dunculis capillaris, 3 cm longis; calycibus cupulatis, 3 ad 4 mm longis, lobis late ovatis, 1.5 mm longis, acuminatis. An entirely glabrous shrub, the branches grayish, 3 to 4 mm in diameter, marked with large petiolar scars; the branchlets slender, smooth, dark reddish-brown. Leaves subequal or one of each pair slightly smaller than the other, chartaceous, entire, oblong to oblong-obovate, or oblong-oblanceolate, 3 to 7 cm long, 1.5 to 2.5 em wide, the upper surface dark-castaneous when dry, the lower much paler, base usually acute, often slightly inequi- lateral, apex distinctly acuminate; lateral nerves slender, 5 to 7 on each side of the midrib, curved, anastomosing, close to the margins, the reticulations very lax; petioles 5 to 9 mm long. Inflorescences very slender, axillary, one- or at most few-flowered, the peduncles capillary, about 3 em long, each bearing one fruit and also presenting one or two scars at the apex as if there might have been two or three umbellately arranged flowers; pedicels slender, up to 1 cm long. Calyx cut-shaped, 3 to 4 mm long, the lobes broadly ovate, 1.5 mm long, acuminate. Luzon, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 33370 Ramos, August 21, 1918, on forested slopes near the summit, altitude about 1,050 meters. ; This species is allied to Cyrtandra glabrifolia Merr. (C. gla- bra Kranzl., non Jack) but differs in its smaller entire leaves and shorter petioles. CYRTANDRA PANAYENSIS sp. nov. Frutex, ut videtur scandens, ramulis junioribus ferrugineo- villosis, ceteroquin glaber; foliis oppositis, aequalibus, coriaceis vel subcoriaceis, ellipticis ad oblongo-ellipticis, 7 ad 10 cm longis, breviter acuminatis, longe petiolatis, nervis utrinque 7 vel 8, XIV, 4 Merrill: Noteworthy Philippine Plants, XV 453 perspicuis, margine distanter serratis; floribus axillaribus, fas- ciculatis, 2 cm longis, corollae tubo infra angustato, aequaliter 5-lobato, lobis circiter 2 mm longis; ovario glabro. A shrub, apparently scandent, the young branchlets usually densely villous with brown or ferruginous hairs, the older parts entirely glabrous. Leaves opposite, equal, coriaceous or subco- riaceous, elliptic to oblong-elliptic, 7 to 10 em long, 3.5 to 5 cm wide, dark-olivaceous and smooth above when dry, the lower surface paler, glabrous, apex shortly acuminate, base acute to decurrent-acuminate, the margins distantly serrate; lateral nerves 7 or 8 on each side of the midrib, very prominent on the lower surface, the reticulations lax, distinct; petioles at first ferruginous-villous, soon becoming entirely glabrous, 3 to 4 cm long. Flowers axillary, pink, about 2 cm long, fascicled, their ferruginous-villous pedicels about 1 cm long or less, subtended by a whorl of small, lanceolate bracts. Calyx about 10 mm long, 7 mm in diameter, ovoid or ellipsoid, black when dry, subcoria- ceous, glabrous, the lobes 5, oblong-ovate, acute or slightly acu- minate, 3 to 3.5 mm long. Corolla 2 cm long, the lower 12 mm slender, about 2 mm in diameter, then somewhat enlarged and 6 to 7 mm wide when flattened out, slightly pubescent externally or glabrous, the lobes 5, equal, broadly ovate, obtuse, about 2 mm long. Disk glabrous, truncate, about 1.5 mm high. Ovary and style glabrous. Fruit soft, fleshy, ellipsoid, about 12 mm long. PANAY, Antique Province, Culasi, Bur. Sci. $2241 (type), 82411 McGregor, May 24, June 20, 1918, in the mossy forest, altitude about 1,000 meters. This species is closely allied to Cyrtandra tayabensis Elm., from which it is distinguished by its smaller, fewer-nerved leaves and its entirely glabrous ovaries. CYRTANDRA SANTOSII sp. nov. Frutex 1 ad 2 m altus, partibus junioribus et inflorescentiis et subtus foliis dense et longe ferruginoso-villosis; foliis rigidis, coriaceis, oblongo-ovatis, usque ad 10 cm longis, longe petiolatis, acutis ad acuminatis, basi acutis ad subrotundatis, margine dis- tanter et obscure serratis, supra laevis, nigrescentibus, vetus- toribus glabris vel subglabris, subtus ferrugineis, ad costa et nervis et reticulis dense et longiter ferrugineo-villosis; inflores- centiis cymosis, paucifloris, pedunculatis, usque ad 6 cm longis, bracteis oblongis, circiter 2 cm longis; floribus circiter 18 mm longis, extus densissime villosis, calycis lobis longe caudato- acuminatis. 454 The Philippine Journal of Science 1919 A shrub, 1 to 2 m high, the younger parts, inflorescences, and lower surface of the leaves, especially on the midrib, nerves, and reticulations, densely villous with long, soft, silky, ferruginous hairs. Leaves coriaceous, rigid, oblong-ovate, 7 to 10 cm long, 3 to 4.5 cm wide, often somewhat inequilateral, the apex acute to acuminate, base obtuse to acute, margins obscurely serrate, the upper surface smooth, blackish when dry, when young more or less villous, especially along the midrib, becoming glabrous, the lower surface ferruginous; lateral nerves about 8 on each side of the midrib, curved-ascending, prominent and projecting as are the rather dense reticulations; petioles up to 3.5 cm in length. Inflorescences axillary, solitary, peduncled, cymose, rather lax, densely ferruginous-villous, the peduncles up to 3 cm in length, the bracts membranaceous, oblong, villous, about 2 cm long. Pedicels, in anthesis, about 5 mm long, very densely fer- ruginous-villous, in fruit up to 1.5 cm in length. Flowers about 1.8 cm long. Calyx glabrous inside, densely villous outside, about 13 mm long, the tube 7 mm long, the lobes with a 4 mm long, stout, caudate acumen from a broad base. Corolla 1.8 cm long, densely villous externally, inside glabrous, the throat and lobes granular, the throat somewhat oblique, the lobes unequal, orbicular-ovate, about 4 mm in diameter. Disk truncate, gla- brous; style 4 mm long, prominently capitate-glandular. Fruit white when fresh, brown when dry, oblong-elliptic, 1 to 1.5 em long. LUZON, Benguet Subprovince, Pauai, Bur. Sci. 32071 Santos, May, 1918, on slopes, altitude about 2,000 meters. This very characteristic species is perhaps as closely allied to Cyrtandra hypochrysoides Krinzl. as to any other described form, but differs remarkably from that species in numerous characters, especially in its entirely different inflorescence, longer petioles, coriaceous rigid leaves, its much denser and longer indumentum, and its very different flowers. CYRTANDRA LANCIFOLIA sp. nov. Suffrutex circiter 40 cm altus, e basi ramosus, ramis tenuibus; foliis lanceolatis, membranaceis, in paribus inaequalibus, 5 ad 13 cm longis, utrinque subaequaliter angustatis, acuminatis, margine leviter undulatis, nervis utrinque 4 vel 5, utrinque gla- bris vel junioribus subtus leviter pubescentibus; floribus axilla- ribus, subsessilibus, circiter 5 em longis, plerumque solitariis, bracteis foliaceis, spathaceis, usque ad 1.5 cm longis, deorsum connatis, perspicue acuminatis; calycis oblongis, 7 mm longis, subaequaliter 5-lobatis. XIV, 4 Merrill: Noteworthy Philippine Plants, XV 455 An undershrub, up to 40 cm high, branched from the base, the stems terete, about 4 mm in diameter, the branchlets brown- ish, slender, somewhat pubescent with very short hairs. Leaves lanceolate, membranaceous, those of each pair similar in shape but unequal in size, one usually one-third to one-half smaller than the other, 5 to 13 cm long, 1 to 2.7 cm wide, subequally narrowed to the acuminate apex and base, the margins some- what undulate, the upper surface glabrous, brownish-olivaceous, the lower pale, glabrous or when young very slightly pubescent on the midrib and nerves; lateral nerves 4 or 5 on each side of the midrib, curved-ascending, anastomosing, the reticulations very lax; petioles 3 to 10 mm long. Flowers axillary, solitary or in pairs, white, sessile or very shortly pedicelled, about 5 cm long, each subtended by a pair of foliaceous spathelike bracts which inclose the calyx, being united by their margins in the lower 4 mm, these bracts 1.5 to 1.7 cm long, glabrous or nearly so, prominently acuminate, up to 9 mm wide below. Calyx ob- long, about 7 mm long, slightly pubescent, subequally 5-lobed, the lobes lanceolate, acuminate, 2mm long. Corolla 4.5 to 5 cm long, pilose with long scattered hairs externally, the lower 1 cm of the tube very slender, not exceeding 2 mm in diameter, then - broadened, the throat, when flattened, about 1.5 cm wide, sub- equally 5-lobed, the lobes ovate, rounded, about 1 cm long. Disk cylindric, glabrous, 2 mm long, somewhat crenate. Ovary gla- brous. Stamens about 3 cm long. Luzon, Ilocos Norte Province, between Bangui and Claveria, Bur. Sct. 33078 Ramos, August 29, 1918, along small streams in forests at low altitudes. This species belongs in the group with, and is closely allied to, Cyrtandra livida Krinzl. and C. agusanensis Elm., from both of which it differs notably in its very much larger flowers, while the leaves are merely undulate, not toothed. ACANTHACEAE HEMIGRAPHIS Linnaeus ' HEMIGRAPHIS NUMMULARIFOLIA sp. nov. Herba glabra, tenuis, ramosis, prostratis vel subprostratis; foliis orbicularis ad orbiculari-ovatis, integris, 5 ad 10 mm longis; floribus axillaribus, solitariis, ebracteolatis, circiter 1.38 cm longis, calycis segmentis lineari-lanceolatis, 6.5 mm longis; capsulis anguste oblongis, 9 mm longis. A slender, much-branched herb, the main branches spreading or prostrate, up to 20 em long, the secondary branches more or 456 The Philippine Journal of Science 119 less ascending, glabrous or nearly so, none of the branches more than 1 mm in diameter. Leaves orbicular to orbicular-ovate, 5 to 10 mm long, rounded, entire, olivaceous, the upper surface with rather numerous, irregularly scattered cystoliths, the lower sur- face slightly hispid on the midrib and nerves; lateral nerves 2 or 3 on each side of the midrib, not prominent. Flowers axillary, solitary, ebracteolate, about 1.3 cm long, sessile. Calyx segments nearly free, about 6.5 mm long, 1.5 mm wide or less, linear-lan- ceolate, acuminate. Corolla white or faintly tinged with blue, the lower 6 mm of the tube slender, cylindric, then broadened; lobes 5, subequal, orbicular, about 3 mm in diameter. Filaments bearded; anthers 1 mm long. Capsules narrowly oblong, acute or slightly acuminate, glabrous, about 9 mm long, 1.5 mm in diameter. Seeds about 12, orbicular-elliptic. Luzon, Ilocos Norte Province, between Bangui and Claveria, Bur. Sci, 33031 (type), 32974 Ramos, August 30, 1918, on dry open banks, the leaves when fresh reddish-purple beneath. This species belongs in the group with Hemigraphis reptans Nees and is most closely allied to H. fruticulosa C. B. Clarke. It is well characterized by its unusually small, orbicular, entire leaves ; by its axillary, solitary, and ebracteolate flowers; and by being glabrous or nearly so throughout. COMPOSITAE VERNONIA Schreber VERNONIA GLANDULIFOLIA sp. nov. Frutex scandens, ramulis junioribus bracteisque exceptis gla- ber; foliis lanceolatis ad oblongo-lanceolatis vel oblongo-ellip- ticis, subcoriaceis, 4 ad 7 cm longis, basi acutis, apice acutis vel leviter acuminatis, subtus minute et dense glanduloso-punctatis, nervis utrinque 6 ad 8, distinctis; capitulis paucis, solitariis vel binis, circiter 2 cm longis, multifloris, bracteis pubescentibus, in- terioribus circiter 1 cm longis. A scandent shrub, glabrous except the very young branchlets and the involucral bracts, the branches grayish-brown when dry, not lenticellate, the slender branchlets dark reddish-brown. Leaves lanceolate to oblong-lanceolate or oblong-elliptic, subcoria- ceous, olivaceous when dry, entire, 4 to 7 cm long, 1 to 2 cm wide, subequally narrowed to the acute base and the acute or slightly acuminate apex, the lower surface rather densely and minutely glandular-punctate, the numerous glands sunk in minute pits; lateral nerves 6 to 8 on each side of the midrib, distinct, anas- tomosing; petioles 5 mm long or less. Heads few, solitary or in XIV, 4 Merrill: Noteworthy Philippine Plants, XV 457 pairs at the tips of the branchlets, in fruit about 2 cm long, their peduncles about 1 cm long; involucral bracts more or less pubescent, the outer ones 2 mm long or less, the inner ones nar- rowly oblong-lanceolate, about 1 cm long, their margins ciliate. Achenes about 4 mm long; pappus hairs numerous, about 1 cm long, pale reddish-brown. Luzon, Ilocos Norte Province, Mount Palimlim, Bur. Sci. 23374 Ramos, August 21, 1918, on forested slopes, altitude about 1,000 meters. This species is well characterized among the woody scandent forms by its small, densely glandular-punctate, glabrous leaves and its few heads. Its alliance is with Vernonia pyrrhodasys Schulz-Bip. THE PHILIPPINE JOURNAL OF SCIENCE VoL. XIV MAY, 1919 No. 5 SOME BACTERIOLOGIC PHASES OF THE CHOLERA- CARRIER PROBLEM By JoHN A. JOHNSTON (From the Biological Laboratory, Bureau of Science, Manila) It has been tersely stated by Munson(3) that the dangerous nature of frank cases of cholera with diarrhea, vomiting, and collapse is well recognized by the people and consequently would be avoided, reported, isolated, and followed by disinfection of their environment; in short, cases of this nature would not oc- casion a health department great anxiety. The cholera carrier, however, is to be regarded with concern by health officials as a dangerous source of infection; for, having no symptoms, he is not considered a menace, and may go about infecting privies, food, drinks, and those who come in contact with him. The Sanitary Commissioner of India says: (6) The part played by the Kumbh fair at Hardwar in the dissemination of cholera infection in the United Provinces has already been referred to; in the Punjab this fair seems to have played an even more definite part in the spread of this disease. It is very significant that the year 1903 was the date of the previous Kumbh fair. Cholera was practically absent from the Punjab during the first three months of the year but in April pilgrims returning from Hardwar spread the infection far and wide and during the next six months most districts suffered more or less severely. A few figures here may not be amiss: From July 1 to No- vember 1, 1914, some 30,000 specimens of feeces from prisoners in Bilibid, not cholera suspects, were examined and 1.75 per cent found to be positive. The fact that an individual is a cholera carrier does not give immunity. Prisoners 8617, 12765, and 30351 were found posi- tive on September 10 and 11, 1914. They were isolated, and 165559 459 460 Philippine Journal of Science 1919 their feeces were examined every other day and found con- tinuously positive. No. 8617, after having been a carrier for seventeen days, became ill with frank cholera; No. 12765 was a carrier for sixteen days before he came down with the disease, and No. 30351, for eighteen days; the last-mentioned individual died of the disease. The question naturally arises: What is the cause of develop- ment of cholera in carriers? In so far as our present knowl- edge of the subject goes this is an unanswerable question. For want of anything more definite, it may be said in a general ~ way that a lowering of vital resistance permits invasion. If this hypothesis be true, then we can easily understand how a carrier may become a victim to the disease at this time. But not everything is explained even by the theory of lowered vital resistance; it is still an open question. Some workers are in- clined to believe that the organism itself, as occurring in persons of carriers, fluctuates in virulence; but this belief is scarcely tenable, and may be dismissed with scant discussion. However, it may be that the cholera vibrio requires passage through an intermediate human host before causing symptoms. My own observations are to the effect that the vibrios isolated from a carrier, either while living or after death, behave in exactly the same manner as do those obtained from a frank cholera case; there is absolutely no difference. Once a cholera carrier, how long does this condition exist? This is a question of the greatest importance for the health officer. The answer can only be obtained from the laboratory, and unfortunately we are unable to answer definitely. We know that in the typhoid carrier the period of infection lasts for years, very likely for the life of the individual. Some obser- vations made on prisoners in Bilibid would seem to indicate that very likely the cholera-carrier state may be of indefinite duration. Prisoner 8841 was positive on September 12, 1914, had seven negative examinations at intervals of four days, was released from quarantine, and was not reéxamined until July 15, 1916, when he was positive again; negative until September 10, when he was once more positive; negative until December 31; posi- tive again January 13 and 18, 1917; negative on January 24, but a nonagglutinable vibrio was isolated; January 25, 27, 30, and February 3, respectively, a nonagglutinable vibrio was isolated; February 7 he was once more positive; February 10, 15, 17, 20, 22, 27, and March 1, respectively, a nonagglutinating XIV, 5 Johnston: The Cholera-carrier Problem 461 vibrio was isolated; March 3 he was again positive; March 6 and 8, respectively, a vibrio was found; March 15 he was nega- tive and remained so to December, 1917. Prisoner 8472 was positive September 28, 1914, and then had eight negative examinations at intervals of four days; he was found positive again July 13, 1916. No. 9638 was found positive January 17, 1915, followed by eight negative examinations at intervals of four days; he was again positive July 18, September 138, and December 1, 1916, and has not been positive again. Is it not reasonable to suppose that these men harbored the cholera vibrio for periods of one and two years? If this be true, why not for longer periods of time? No. 11040 was positive for the first time August 2, 1916; negative until September 22; again negative until October 19; negative again until November 25; negative, November 29; positive, December 3; negative, December 15 and 16; December 20 a nonagglutinable vibrio was isolated; negative, December 23 and 29; on January 8 and 11, 1917, respectively, a non- agglutinable vibrio; January 18, positive; January 20, 24, 25, and 27, respectively, a nonagglutinable vibrio; January 30, negative; February 3, a nonagglutinable vibrio; February 7, positive; February 10, 15, 17, and 27, respectively, a nonagglu- tinable vibrio; March 1, negative; March 3, positive; March 6 and 8, negative; April 3 and 4, positive; and since then, negative to December. I have records of many other cases that I could cite, but these are sufficient to show the irregularity with which positive re- sults occur. I believe this irregularity is more apparent than real; for I believe that, if a cholera carrier were kept under exactly the same conditions of temperature, humidity, and sur- roundings; given the same food and the same work; and if the specimens were taken at the same time of day, in the same way, and transmitted to the laboratory under exactly the same conditions, the number of positive results would be more than doubled. The researches of Schébl(7) and of Schébl and Paf%aniban (8) on cholera carriers in guinea pigs have shown that food plays an important réle in the appearance and the disappearance of the vibrios in the feces. The normal period for recovery of the cholera vibrios from artificially infected guinea pigs has been definitely established to be approximately fourteen days, after > 462 Philippine Journal of Science 1919 which time they can no longer be found. If bile is adminis- tered, however, these same animals continue to excrete the vibrios for indefinite periods, the exact length of time not having been determined definitely. In 1916, in Bilibid Prison,(4) 600 carriers (8.2 per cent of the population) were found; a total of 112,577 stool examina- tions were made. A word as to the technic: A sterile swab was plunged into each man’s feeces and thus inoculated; this swab was placed in a test tube containing about 2 cubic centimeters of plain agar of a reaction of —1.0 to phenolphthalein. The purpose of the agar is simply to prevent the drying out of the inoculum. Upon receipt at the laboratory, 10 cubic centimeters of a 2 per cent peptone solution were added, and the tubes in- cubated over night. In the morning hanging drops from these peptone tubes were examined for suspicious motility, and the tubes showing it were set aside for further culturing. Dieu- donné plates were used and, after streaking, were incubated for a further twenty-four hours. All suspicious colonies were fished out, emulsified in a sterile salt solution, and tested for agglutina- tion. For the latter test an immune serum having a titer of 1 to 4,000 was used in a dilution of 1 to 500. The cholera vibrio agglutinates promptly with this, and nothing but the true cholera vibrio will give a positive reaction. There are vibrios, however, that correspond with the true organism in every way except agglutination; these may or may not be true cholera vibrios. The majority of these vibrios are hemolytic; that is, in blood- agar plates a clear zone of hemolyzed blood is produced about the colony. Some authors have reported strains of cholera vib- rios producing hemolysis also. In the light of our present knowl- edge, however, it may be safely accepted that any vibrio that does not agglutinate promptly on being brought into contact with a drop of a 1 to 500 dilution of a known cholera-immune serum of high titer is not a true cholera vibrio. That these nonagglutinating vibrios bear some more or less intimate relation to cholera seems to be borne out by the fol- lowing, from O’Connell: (1) These choleroid organisms are an extremely interesting phenomenon. Their relationship to true cholera may be said to have been established but not defined. Bacteriologists in service in the Far East have noted their appearance in specimens subjected to microscopical examination immediately before and during cholera epidemics. They are not known where cholera is not or has not been recently. Their morphology and biological characteristics are such that it is impossible to distinguish them from true cholera organisms by microscopic examination. XIV, 5 Johnston: The Cholera-carrier Problem 463 Greig,(2) working in India, has made an intensive study of these choleralike vibrios, and he classifies them according to their serological reactions with each other. He makes arbitrary division into six groups. During the past year thirty strains of nonagglutinating vibrios have been studied at the Bureau of Science. These strains were obtained from cholera cases, from carriers, and from presumably normal individuals who were contacts of cases of cholera. After obtaining a strain in pure cul- ture, 0.1 cubic centimeter of a suspension of a twenty-four hour culture was injected into the gall bladders of a number of guinea pigs under strict asepsis. The animals were sacrificed from ten to twelve days subsequent to injection, and the contents of the gall bladder were used to inoculate peptone tubes; direct plates on Dieudonné were made also. The supply of animals being short it was decided to use bile, and transfers were made every three days. After forty such transfers eight of these pre- viously nonagglutinable strains gave prompt agglutination in 1 to 500 dilution. Of these eight strains five held the acquired prop- erty for approximately three months. Three lost it after two months, and none held it for longer than four months. ‘These facts were reported at the time as showing that the property of agglutination might be acquired, but my belief now is that these strains were true cholera that had lost the ageglutinability. TREATMENT In Bilibid the carriers were given urotropin and, following Schobl’s experiments with ox bile in guinea pigs, 0.65 cubic centi- meter of ox bile three times a day for two days; after an interval of five days the treatment was repeated. Those cases negative after the treatment were released from quarantine. The prison authorities at first, I believe, used sodium tauro cholate; but, the supply of this soon becoming exhausted, I suggested the use of inspissated beef bile. It was not supposed that the bile would cure the carrier condition, but that it would cause more vibrios to enter the intestinal tract from the gall bladder. The following is quoted from the Report of the Philippine Health Service for 1916: (5) Results—From 185 cases prior to the use of ox gall, 15 or 8 per cent became repeaters within 2 months. The first ox-gall period produced the following result: Of 155 treated September 8, 9, 14 and 15, 5 were reported positive September 9, and 6 September 17; 84 of these 155 were discharged to brigades September 20, of whom 10 were returned positive on September 22; thus producing a total of 21 positives or 13.5 per cent of the 155 treated. 464 Philippine Journal of Science Of 529 prisoners receiving ox-gall treatment in the cholera brigade, 120, or 23 per cent, became repeaters. Of this number, 68, or 13 per cent, were returned as repeaters from the cholera brigade as follows: Following first ox-gall treatment, 21 or 31 per cent. Following second ox-gall treatment, 42 or 62 per cent. Following third ox-gall treatment, 5 or 7 per cent. It will be seen from this that the administration of bile would seem to be helpful in the detection of chronic carriers. REFERENCES (1) Annual Report of the Health Officer of the Port of New York. Albany (1915) 14. (2) Greic, E. D. W. Bacteriological studies of cholera like vibrios isolated from stools of cholera cases in Calcutta. Indian Journ. Med. Re- search 4 (1916-1917) 658-671. ae (3) Munson, E. L. Cholera carriers in relation to cholera control. Philip. Journ. Sci. § B 10 (1915) 1. (4) Report of the Philippine Health Service for the year 1916. Manila (1917) 25. (5) Report of the Philippine Health Service for the year 1916. Manila (1917) 26. (6) Report Sanitary Commissioner, India (1915) 62. (7) ScH6sL, Otto. Experimental cholera-carriers. Journ. Infect. Dis. 18 (1916) 307-314. Ipem. The relation between the amount of cholera culture injected into the gall bladder and the state of cholera carriers in experi- mental animals. Philip. Journ. Sci. § B 11 (1916) 153-155. Ipem. The influence of bile upon the duration of the state of cholera carriers in experimental animals. Philip. Journ. Sci. § B 11 (1916) 157 and 158. IprM. The influence of bile upon the distribution of cholera vibrios in the digestive system of experimental cholera carriers. Philip. Journ. Sci. § B 12 (1917) 23 and 24. Ipgm. A survey of certain chemicals with regard to their bactericidal action on cholera vibrios within the body of experimental cholera carriers. Philip. Journ. Sci. § B 12 (1917) 215-281. (8) ScHOBL, Orro, and PANGANIBAN, C. S. Experimental cholera carriers and immunity. Philip. Journ. Sci. § B 12 (1917) 43-49. PHILIPPINE RAW MATERIALS FOR GLASS MAKING By T. Dar JUAN and V. ELiIcaNo (From the Chemical Laboratory and the Division of Mines, Bureau of Science, Manila) THREE TEXT FIGURES Some years ago experiments were made by private concerns in the manufacture of glass from materials found in Pampanga and Tarlac, but the attempts were fruitless. At present there is a bottle factory at Santa Ana, Manila, that operates entirely with broken glass, or cullet; however, the operation of this factory is very intermittent, and its production does not meet the demand. The imports of glass bottles and other ordinary glassware into the Philippine Islands have shown an almost constant increase since 1915, and have now reached a point where the establish- ment locally of a modern glass factory would appear to be justi- fied. Information obtained by the Bureau of Science indicates that 15,000,000 split and pint bottles are used annually in the Philippines, 10,000,000 of which are returned empty to the soda- water and beer factories, leaving a probable market for not less than 5,000,000 bottles per year. ‘Table I gives the value of Philippine imports from Japan of glass bottles, drinking glasses, and lamp chimneys for the last four years. TABLE I.—Philippine imports from Japan. [Numbers give values in. pesos.] 1918 1917 1916 191s IGW wmnbte hottlew: 222 gee 440,090 | 198,098 | 158,419 | 129, 197 DRA PMR oo eee 95,057 | 142,076 | 60,268] 52,859 Lamp chimneys Shes Sabre Bc een Se oe eee 87,210 | 16,445} 20,887 | 17,023 gu | SAR cm RYE G Re Nw Und Re Pee re ee site OU ce 572,357 | 351,614 | 239,024 | 199,079 Prices of bottles and other ordinary glassware have increased considerably during the past two years, and at present it is very difficult to secure them reasonably. In the Philippine Islands there are 12 soda-water factories (in Manila), 2 breweries, 440 465 i. 466 Philippine Journal of Science 1919 drug stores, 81 alcohol distilleries, 17 repacking establishments, and a few other industries the operation of which depends en- tirely on the solution of the bottle problem. In view of the above considerations, we undertook this investigation in order to determine the suitability of Philippine materials for glass making. Technically speaking, there are two general classes of glass recognized ; namely, lime glass and lead glass. Lime glass is the material most commonly used for making the cheaper articles, such as ordinary bottles, demijohns, glass jars, drinking glasses, lamp chimneys, etc.; while lead glass is a more expensive product and is used chiefly in the manufacture of cut glassware and for optical purposes. The principal raw materials used in the man- ufacture of lime glass are lime, silica, and alkali. LIME Coralline and crystalline limestones of good quality occur throughout the Philippine Islands,’ but the deposits most avail- able to Manila are those of Montalban, Binangonan, and Taya- bas. Both the Montalban and the Binangonan deposits are within the boundaries of Rizal Province; the limestone is of Mio- cene formation and is hard, crystalline, and of uniform chemical composition. Montalban is located about 30 kilometers from Manila, on a line of the Manila Railroad. Binangonan is situated about 35 kilometers from Manila, with water transportation facilities via Pasig River and Laguna de Bay. The Tayabas deposit is a metamorphosed coralline limestone and is located at Palsabangan, a barrio on the railroad and be- tween the towns of Pagbilao and Malicboy. It is estimated that limestone may be obtained from any of these places at from 5 to 8 pesos? per metric ton, including cost of quarrying and transportation to Manila. The Bureau of Science in its experimental kiln has manufac- e tured from Montalban, Binangonan, Cebu, and Tayabas lime- stones over 100 tons of excellent lime, suitable for use in sugar centrals and in other industries: and there is no reason why these limestones cannot be used for glass making. In this connection it would be of interest to note the composi- *Cox, A. J., Philip. Journ. Sci. § A 4 (1909) 211; Cox, Reibling, and Reyes, Philip. Journ. Sci. § A 7 (1912) 332. ?One peso Philippine currency equals 50 cents United States currency. xiv,s Dar Juan and Elicaio: Philippine Glass Materials 467 tion of some Philippine limestones. The analyses given in Table II show that they are almost pure. TABLE II.—Analyses of some Philippine limestones. (Numbers give percentages.] Binango-| Binango-| Montal- | Montal- | Palsaba- Constituent. Cebu. jnan, Ri-|nan, Ri-| ban, Ri-| ban, Ri- Ingan, Ta- zal, 19124) zal, 1918. |zal, 1912 | zal, 1918.| yabas.> eek en ipnition: ee 43.45 43.31 43.49 43.04 43. 98 43.48 Silica (SiOz) ieee ve 0. 86 1.12 0.88 0. 94 0.26 0.69 Iron and aluminium oxides (R2Os) __ 0. 23 0.15 0. 48 1.14 0.09 0.85 dame (CaO) 54. 67 63.78 54.39 54.61 54. 97 54.68 Magnesia (MgO) ____._-.__-._._...__ 0. 46 1,19 0.68 0. 22 0.72 0. 48 Alkalies (K20+-Na20) - seve BT acs ue & Beate: Cerne * Analyzed by Forest B. Beyer. » Analyzed by F. D. Reyes, chemist, Bureau of Science. SILICA The principal source of silica is white sand or quartz. Sand deposits of good quality occur at Paracale, Camarines Province, and in the Mountain Province at Baguio; but the deposits of commercial importance are the extensive beaches of quartzose sand and the vein-quartz gravel and bowlders found in Lubang Island, Mindoro. The quartzose sand is the result of the disintegration of the intrusive granite that outcrops on the narrow isthmus at the south end of the island, between Looc and Tabahin Bays. This sand is composed mainly of quartz, orthoclase, and mica, with variable proportions of clay. The sand deposits are found along the southern shore of Looc Bay, from the barrio of the same name to Balacbalac Point, and along the shores of Tabahin Bay, which lies between Caybanac and Natulo Points. Each of these extends both above and be- low the present beach and together make an area of about 2.5 square kilometers, Between the high- and the low-water marks the sand is subjected to the constant action of the waves, with the result that a gradual concentration takes place, leaving a product of almost pure quartz. The width of the area containing the clean quartz sand averages 2.5 meters, extending throughout the length of both the Looe and Tabahin shores. From the high-water mark to the foot of the hills, which apparently at one time was a water level, the sand is mixed with greater or less quantities of clay. Below the low-water mark the sand ex- tends into the sea about 200 meters. The probable thickness 468 Philippine Journal of Science 1919 and average width of the underwater deposit is not known, as no soundings were made. The depth of the concentrated sand between the high- and the low-water marks is over a meter, and there are wells above the shore line in the vicinity of Looc that have clean sand bottoms at a depth of from 2 to 3 meters. In- sufficient data are at hand to make even an approximate estimate of the total amount of sand available. The deposits of vein-quartz gravel and bowlders are found along the coast bordering the schist and gneiss formation at the north and the south sides of the intrusive granite mentioned above. The gravel has accumulated in great quantities in all the inlets, and the supply is fairly large. The principal areas of accumulation are the northern shore of Looc Bay toward Tumbaga Point, and from Antipolo Point to Pula Point near the town of Agcauayan. It is also found on the east shore of Lubang Island from Caybanac Point to Quebrada Point, and around the southern shore from Natulo Point to Baiacbalac Point. Patches of accumulated quartz gravel are also to be found on the northern shores of Golo Island from Bulacan toward Caypandan Bay. This gravel ranges in size from that of a pea to more than 75 milli- meters in diameter and is mixed with quartz bowlders weighing from a few kilograms to a ton or more. The gravel and the bowlders come from the fracturing of the segregated lenticular veins of quartz found in the schist and the gneiss. Some of the quartz lenses outcropping in Dilau Point measure approximately 6 meters in thickness. However, the main supply of these quartz veins is found around Agcauayan, where several lenticular veins outcrop, and floats of quartz breccias and bowlders cover the tops — of Agcauayan, Puti, and Pula Hills (see fig. 1). _ Besides the materials above described, the discharge slimes from the cyanide treatment of auriferous quartz of the Colorado Mining Company cyanide mill located at Aroroy, Masbate, and other such mills, may prove to be of value as a possible source of Silica in glass manufacture. The utilization of this pulp has the advantage that, being already finely pulverized, it can be mixed directly with lime and alkali without preliminary treatment. Al- though no actual test has been made of this material, we are of the opinion, judging from its chemical] composition, that it will make a good bottle glass. Sand of lower purity, but useful as a fluxing material in glass making, is found in large quantities at Pasay Beach, Rizal Prov- 469 Philippine Glass Materials Dar Juan and Elicano XIV,5 the gives ts Table III imen ince. . Tarlac River, Tarlac Prov analyses of silicious materials used in our exper in , andi ince *‘JAeyo APAING Oepoey pu ys0D & wHOAz peotiy, “spusjsy Jueqny ey} jo dew ‘lL ‘Sly 03003! ’ MM eGil __ B02! 208! :O¥ ef! EN Avg 200T q YW Cnt porns - ‘yg oeueghe) SCNVISI DNVAN'T eel 0S ofl == ‘l uanyeary | /2049 Duangy ee eet Poon § $2021 O2e0Zi Sloodl -D1 02! 470 Philippine Journal of Science 1919 TABLE III.—Analyses of silicious materials. {Numbers give percentages.] Constituent Gees | abe | Zee, | See Fae, Loss on ignition.___________ 0.34 0.80 5.40 0. 82 0.55 Silica (SiOz) 97, 49 86. 60 56.00 57.54 84, 85 Ferric cxide (raise) trace 0. 48 6.24 7.89 3.51 Aluminium oxide (Al2Os) 1.58 8.12 19,16 20. 47 6.29 Lime (CaO) trace 1,20 6. 18 7.31 1.65 Magnesia (MgO) trace 0.22 4,84 3.27 2.18 Manganese oxide (MnO2) ---| trace trace trace trace: foo 232: Alkalies (K20+Naz20)_.....---...._....... 0.50 ©2. 58 2.66 ©2.70 €1.02 * Analyzed by F. Pefia, chemist, Bureau of Science. > Analyzed by A. S. Argiielles, chemist, Bureau of Science. © Determined by difference. ALKALI This material is derived from sodium carbonate or sulphate, which for glass making should be as nearly free from iron as pos- sible. Sodium carbonate, or soda ash, fuses more readily with Silicious material and lime than does sulphate; but, since the e latter is a cheaper product, it is more generally used in glass making, especially in the production of cheap articles. If sodium sulphate is used, some form of carbon should be admixed to assist in its reduction. Sodium carbonate and sodium sulphate, which are intermediate products in the preparation of caustic soda, de- pending upon the process used, may be imported from the United States or may be prepared locally from common salt. During 1917 the Philippine Islands imported 1,423,532 kilograms of caus- tic soda, valued at the port of origin at 326,813 pesos. The es- tablishment in the Philippines of a caustic soda factory, in con- nection with a glass factory or a paper-pulp plant, will save to the Islands the value of their imports of this commodity and will encourage the exploitation in the Philippines of such industries as the manufacture of paper pulp, glass, soap, etc., all of which use caustic soda or one or more of its intermediate products as raw material. In a soda factory the largest item is fuel, and the next, common salt. It is believed that, by establishing a fac- tory in a place where there is an ample supply of firewood, and where at the same time climatic conditions are favorable for the solar evaporation of sea water, sodium carbonate can be produced locally at approximately 102 pesos per metric ton. In addition to the materials already described cullet ig in- variably used in commercial glass factories. The object of mix- xiv,s Dar Juan and Elicano: Philippine Glass Materials 471 ing cullet with the batch is to utilize the fragments of glass left over during the process of manufacture, and to serve as a flux and to lower the temperature of reaction of the materials employed. Mixing the batch with cullet also considerably re- duces the cost of production. However, since the main object of this investigation was to ascertain the suitability of Philippine materials for glass making, no cullet was used in our experiments. PROPORTIONS OF CONSTITUENTS The proportions in which the different ingredients of glass are mixed vary considerably, depending upon the quality and composition of the raw materials available, the quantity of broken glass mixed with the batch, the temperature, and the quality and color which it is desired to give the finished product. Table IV gives the proportions of the materials used for the different batches, the percentage composition of glass materials in the mixture, and the percentage composition of the glass obtained. The mixtures used in the first and second batches were rather difficult to melt and gave a glass that at white heat was not sufficiently fluid to expel all the gas bubbles. If to mixture 1 a small quantity of arsenious or manganese oxide is added, as shown in the table, a practically colorless glass is obtained. Tubing and bottles made from these mixtures are hard and strong, but they become cloudy after a few weeks. Mixture 3 is not difficult to melt, the gas bubbles are easily expelled, and the glass remains perfectly transparent. Mixture 4 gives a glass practically free from bubbles, and bottles made from it have an excellent appearance. Also in this case the addition to the mix- ture of a small amount of arsenious or manganese oxide decol- orizes the product almost completely. Mixture 5 gives a greenish glass excellent for soda-water or beer bottles. The addition of small quantities of arsenious oxide (batch 6), or of manganese oxide (batch 7), produces a lighter-colored glass. Mixture 8 gives a green glass suitable for soda-water or beer bottles, and mixtures 9 and 10, a glass rather too dark for soda-water, but good for beer and wine bottles. The furnace used in these experiments was built after the ordinary Japanese glass pot furnace and has over-all dimen- sions of 188 by 79 centimeters at the base and a height of 132 centimeters. Text figs. 2 and 8 give a general idea of the fur- nace construction. Its most important feature is the air pas- sage, represented in fig. 2 by the letters la, 2a, and 3a, which = 1919 rence Journal of Se ippine al Ph 472 cr dewacece cl | eoeay nidgne oe CRAMER AI RRO Ararat a >: Wa ieee Gelato G ok) de tas pos a oo ereeeeeoccrsss""" (Uy) eseuesuey wenn eeteen anne ne vaipiaintaininia Wicket ache (sagt Sat aban pater eee cette Pres es ee Mee ES eA a . voneo"="-="" (a) STURT 98 OL GT ‘ot ESL: Bao oie errr? PILI 8 “6T oP ST ere ae oT See ea 58 ~~ (O%8N+O02H) Senexry Le °T 88 °T DEW: are er Ss, “gakertentors ay 110 eT eae eee ee en ree Te Sho eet oe ee ce AO Oe 96 ‘OT 90°6 19° RIS eRe aan: peered r8'8 00°6 OF er a ee Re Be sere a te eee ae ee eee ee eS eee, TS Zt 90 “FT LOO: -apritrea ts Dor a aD 9L'T Ste ier rao Port Toons eomees ===" (8024) Seprxo unyuyun]s pue UOIE Te 59 18°89 GOGO: erent ee 09 °L9 29 OL VU Us: ) Saex 2 (eas ia BR ae a eae gern tee ee ee YGUNLXIN AHL NI STVINDLVW SSV'TD 10 FVINGOUdd | | | > eg Sears sakes Cres ee est Toco" "=""== (QUIN) OPIxo esouBsuRy mee ee ene Teomrs=nosee="s-(8Qasy) @pfxo snojusery oe mee 8°88 9°88 ance eee nee PR eat “--="=-- aug] peyeapaH ee ae * ee, er wi: ap “"-"equuoqawo umnpog Bicone ay gests “pues OBETY, eee eee ene see manne en enn 0°00T pues yoveq Auseg snenmnntanden ge aes ibs mek; cehlabemeteal et a wenn eee RE he et a ea ee ne en ae ee Suvqn’] woe-- ‘T ‘ON Pe ee onset e eae ae a caren ra he uaa ne Namemecaietts meet ge 0°8I 0°8T ee ote Baie ae ape Bie er Po ett e pi ep OMS 0°9% 8°88 0'OF 00P 0 0b Org 0°98 9°28 O°00L | 0°00T {ot |----------|-------- ofa sak Sue ae tr Sep reas 0°00L | 0°00T | 000T 7 “ON "9 "ON "g “ON "Pp ‘ON "3 ‘ON F 4 “ON wwe | wiwg | goeg | yg | yoreg | yoreg yoqeg ‘ *quenz1}8U0p ‘LHDINA AM SLUVd NI SUNLXIN AVY ‘peuyn7go 9805 pun saunjayw fo uorpsodwos burmoyg— Ay #14aVvJ, Glass Materials 473 icano: Philippine XIV, 5 Dar Juan and El *UdeT “g “g “ON UBY} JazYySYT *), ‘souslr9yIp Ag o ‘souslog Jo n¥veing “ystulayo ‘eueg *q Aq pezsfeuy q ‘ueeI8 YTB “OT “usaIs yIuq “6 “GON UBY} LozYBYT “9 “YstUoeIH *¢ “eBuy YSIMolex “y “BU YSLMoyex “g “T “ON UBY} JezyBrY “2 ‘a3ury ystmorax *T *IN}X{UL ABI OY} JO SJUTperZut ey} Jo Sasf[eue oy} WIZ pozelNo[VD » wren ween 9081} qa 08 “OT a0 ors i a 00°19 www ewe ens wenn een e- ieee ay Bayne Oy) eee ge Ue areas Tr a ssounsueyy SR mmr aan i lore mat Sy sa Act Si caps ig a a cle rig a ame hes eta (SV) dues PORE Ne oe ee gen em eee oe gee Seg (O28N+O2%M) soyexry eel ie OR Oe ie te Cet ee (O34) Biseuseyy WOME ST Cae oe ae ae Me etm ai Gone (08D) ou 99°T , Eke: (O24) eprxo wnruruinye pue uozy PE Oe Gil hes eam neg ht Mameae ams a” Lge ree (O!S) BIg ‘CUNIVIGO SSVID JO NOLLISOUWOO FOVLINEOURA 474 Philippine Journal of Science 1919 is built in a zigzag form below the hearth and extends across the width of the furnace. From the plan it may be seen that this furnace belongs to the gas-producer type, for the combus- tion in the fire chamber is incomplete. The heat radiating from the hearth preheats the air passing through the zigzag passage before it mixes with the hot gases in the combustion chamber. The result is that a more complete combustion of the gases takes place, and a higher temperature is obtained from the flame around the crucible. The peep hole c, which is covered with a sheet of mica, permits the operator to observe the temperature around the crucible and also enables the removal of any soot that may have accumulated along the passage from the crucible chamber to the chimney. When slot b is obstructed, the door d permits the cleaning of the passage from the fire grate to the combustion chamber. The operation of this furnace is very simple. After the fire has been started, sufficient coal is piled in the fuel chamber to fill it to the door level. The height of the fuel bed is kept at a constant level by feeding the fire chamber with 1 to 2 kilograms of coal every fifteen to twenty minutes, and occasionally cleaning the grate to avoid accumulation of cinders. If the temperature around the crucible decreases, it is an indication that the quan- tity of coal burning in the fire grate is insufficient; that too much or too little air is being admitted through the air passage a or in the fire grate; or that the passage from the crucible chamber to the chimney is obstructed with soot. This furnace gives a temperature sufficiently high to expel the gas bubbles from the molten glass and, besides being economical in fuel, has the additional advantages of simple construction and easy operation. If a glass factory is to be established in the Philippine Islands, the following facts should be taken into consideration: The quality and price of fuel available are as important fac- tors as are the raw materials to be melted. At present the market price of coal in Manila varies from 40 to 50 pesos per metric ton, depending upon the quality and the market condi- tions. With such a high price for coal, it may prove advanta- geous to use coconut shells when practicable. This fuel can be obtained in commercial quantities at from 8 to 10 pesos per metric ton. : Labor is also an important factor. Glass making is a new, or practically unknown, industry in the Philippines, and naturally there are few glass blowers in the Islands. If a factory with xiv,s Dar Juan and Elicato: Philippine Glass Materials 475 a capacity of from five to ten thousand bottles is to be operated, it will be unwise to depend entirely upon unskilled labor. ea-< | | | | | 8 QS i ! ae ae \\ t+ --- s6- - - In this connection it would be advisable to in- vestigate the capacity limits to which ma- chine blowers can be operated economically, and to draw conclusions from the results of this investigation as to whether it would be more economical for the present to contract expert glass blowers from Japan to train Filipino laborers or to operate an automatic machine blower. As shown by the composition and the qual- ity of the glass, there is not the slightest doubt that the raw materials are suitable for glass making. Furthermore, in order to de- termine whether or not they would stand the pressure and the sudden changes of temper- ature to which glass bottles are subjected in a factory under ordinary working conditions, a number of glass bottles made by the Bu- reau of Science were tested in local , soda-water factories and breweries; the Wf \\ results were very satisfactory. CHUM, AY we /2 a ’ fs et —-—--@ -- - + / Fire Brick Red Brick Oe ee Pace a ee 1 WN oo BOE | df Ld v fi he, 3 : fife 1 Vefifet, Japs he) OI hd fA afifi LA, da & « £ -+ ' ay > ’ ‘ ’ ed aViésr : ’ 4 : Via I a? ‘ Yafrfr Oo fei I fy) v4 - a vf 3 ee 4 ‘ 4 ! PSS) } f fx ys kt i tf 1 VLA aff : Zhi * +fa sft 2h: es 7 ae wear YY oO 4 Hey ‘ i> iP ‘ ‘he —- 38 4- —E NAZI +h, 3 oo 4th, 1 oe | FT TIS RRS ASS NIT IO ITNT NUTS pS, . 43 ames Serie stent St ei peg tee r i Re a BER ee en) a ee te a ee Fic. 2. Elevation of the experimental glass furnace. Section through ABCD. a, chimney, made of galvanized or plain sheet iron 1/16 inch thick. 165559 —2 1919 Philippine Journal of Science 476 “SI9jIWUTZUSS UT USAIS oe SUOTSUSUTIG] ‘“soVUINZ SSe[B [ejUsUILIedxe ey} Jo ue “g “DI NY My : . Cas pe aie ~ 4G Perey MWg Bg > WH 24,4 xiv,5 Dar Juan and Elicaio: Philippine Glass Materials 477 In experimental tests made in the Bureau of Science, the quartzose beach sand and the vein quartz gave glass very sim- ilar in quality, except that that made from the vein quartz is of a lighter color due to a lower iron content, as shown in Table IV. The materials being of equal manufacturing suitability, the - quartzose beach sand is preferable for the following reasons: 1. The quartzose sand already contains some feldspar, which lowers the melting point and supplies to the charge a certain amount of lime, alumina, and alkalies; while, if the vein quartz is used, all the fluxes have to be added. 2. The individual grains of the quartzose sand do not exceed 3 millimeters in diameter and thus, as a natural product, this sand can be readily charged into the furnaces. The vein-quartz gravel or bowlders, or the materials quarried from the veins, will necessarily have to be subjected to quenching, pulverizing, and sieving, which will increase the cost of manufacture. 3. The cost of mining sand is very much cheaper than vein quartz. The former can be readily dug from the beach and loaded into boats, while in the case of the vein quartz blasting and mucking will have to be done. The vein-quartz gravel of small size already mentioned as accumulated along the beach is not inexhaustible. It has been stated above that a sand concentrate, averaging 2.5 meters wide and composed mainly of quartz grains, is found along the beach between high- and low-tide levels. It is con- sidered that this supply of concentrate is inexhaustible on account of the continuous concentrating action of the waves on the adjacent deposit of impure sand, which readily and con- stantly supplies pure sand. With an average width of 2.5 meters and a depth of 1 meter, extending 5 kilometers in length on both Looc and Tabahin Bays, there is ready on hand.a supply of sand concentrate meas- uring 12,500 cubic meters. By actual test, a cubic meter of the sun-dried sand weighs approximately 1.5 metric tons. This indicates that there is available a supply of about 18,750 metric tons of sand, which will last a factory, with a daily capacity of 5 tons, twelve and one-half years of three hundred working days a year. It is presumed that after the twelfth year digging may again be done on the original place, and a concentrate be found of similar quality as before. From data now on hand it is not possible to estimate accurately the cost of equipment for a glass factory of a given capacity, but any technical information or suggestions as to manufacture can be furnished to prospective investors. : .2 ILLUSTRATIONS TEXT FIGURES Fig. 1. Map of the Lubang Islands. Traced from a Coast and Geodetic Survey chart. ; 2. Elevation of the experimental glass furnace. Section through ABCD; a, chimney, made of galvanized or plain sheet iron Ye inch thick. Dimensions are given in centimeters. 8. Plan of the experimental glass furnace. Dimensions are given in centimeters. 479 CYSTOLITHIASIS AMONG FILIPINOS IN ASSOCIATION WITH DIETETIC DEFICIENCY By Recino G. Papua (From the Departments of Surgery and Physiology, College of Medicine and Surgery, University of the Philippines) ONE TEXT FIGURE INTRODUCTION Osborne and Mendel(19) in their series of dietetic experiments on lower animals found an incidence. of phosphatic calculi in the urinary passages of rats that had been subjected to a ration of inadequate nutritional value. In their conclusion theSe au- thors say: We can not offer any rigid proof of the etiologic relationship between the ration deficient in fat soluble vitamine and the formation of phosphatic calculi. The impressive statistical fact that in every case where the calculi were observed the animals had existed for some time on a diet of the sort indicated, *together with our failure to discover any other pathogenic factor common to all the affected animals, makes the hypothesis suggested above worthy of serious consideration. Following this valuable suggestion, I attempted to carry on an investigation that has practically become twofold: (1) an attempt to show as far as lay in my power the possible appli- cation of the hypothesis to a people collectively noted for an unbalanced, essentially avitamine diet such as exists in the Philippines; (2) an investigation into the chemical composition of Filipino vesical calculi, which is interesting in itself and may prove valuable, considering the relative scarcity of similar in- vestigations in foreign countries. The consensus among various writers who have studied the nutritional condition of the Filipino people is that a large ma- jority of them are not receiving sufficient nourishment for the maintenance of a stable physiological activity. The recent ex- periments of Concepcion(8) on the nutrition of normal Filipino subjects seem to confirm this opinion. The ordinary daily ration of the Filipino masses, consisting principally of rice which is at times so improperly prepared that most of the pericarp where the vitamines are said to exist has been removed, vegetables of moderate amount, scarcely any meat . 481 482 Philippine Journal of Science 1919 or fish, and some other dietary articles of minor importance, is as a whole deficient from a nutritional standpoint. An almost exclusively vegetarian existence, low in phosphorus and protein (Aron and Hocson(2)), or a protracted consumption of a one- sided diet may result in diseases of metabolic or nutritional deficiency, exemplified in these Islands by the existence of beri- beri, the theory of dietetic deficiency as the causation of which is apparently well established by the work of Andrews,(1) Cham- berlain, Vedder, and Williams,(7) Fraser and Stanton,(11) de Haan,(9) Highet,(14) Gibson,(12) Williams and Saleeby,(26) and many others. MATERIALS AND METHODS In the museum of the surgical department of the Philippine General Hospital there were found, at the beginning of this work, forty-eight bladder stones from Filipino subjects, some of which were intact, while others were half-cut. These had been col- lected from the latter half of 1914 to almost the end of 1917. The incompleteness of the clinical histories of these forty-eight cases caused some difficulty in correlating the facts in the attempt to establish a relation between the nutritional state of the patient on admission and the character of the stone surgically removed. However, during the chemical analyses of these calculi, ten cysto- lithiasis cases were admitted to the surgical ward during the last two months of 1917, and these were thoroughly examined from a nutritional standpoint. Therefore, the present investigation is based on fifty-eight cases of vesical calculi with a view to cor- relating as far as practicable the clinical data with the chemical composition of the different layers of the individual stones. To present a picture of the nutritional condition of the patients, it was planned to collect and condense all the essential points in the clinical-history record of the cases. All of the patients were Filipinos, of different age, sex, social condition, occupation, etc.—all of which factors must be considered in determining the relative frequency, and in tracing the nutritional etiology, of the disorder in each group. It must be emphasized that, except for the last ten cases, the history of beriberi as a whole was un- reliable. Fortunately, however, most of the hospital records show, from objective examination of the patients, the degree of their body development and nutrition. The importance of con- sidering the associated diseases, including intestinal parasitic in- festation, lies in the fact that they tend to decrease the vital resistance of the body to the development of superimposing mal- XIV, 5 Padua: Cystolithiasis with Dietetic Deficiency 4838 adies. Diseases of the genito-urinary tract have been included in the case records to aid in explaining the formation of calculi. Whenever autopsy was performed on any of the cases, attempt was made to note the anatomical diagnosis for possible evidences of beriberi. Urine examinations have been made in most of the patients. After weighing the stones, they were cut with a fine saw into two approximately equal parts, one of which was utilized in the analysis, while the other remained in the museum for further reference or subsequent study. Those that were cut were again divided into two, thus making the analyzed portion about one- fourth of the original stone. The layers of the portion analyzed were peeled off and weighed individually. Having determined by proportion the weight of the individual layers in toto, based on the weight of the whole or of half of the stone and that of the analyzed portion of the layer, the latter was subjected to a qualitative chemical analysis according to Heller’s scheme. Experience showed that the chemical phenomena were some- times so misleading that the interpretation of results did not in some instances approach the desired accuracy. To check this error, I adopted the following modification in the analysis of each layer: The layer was ground to a powder and a small amount of the powder was dissolved by the aid of gentle heat in dilute hydrochloric acid; after filtering the solution, two drops of the filtrate were put on a slide and subjected to the action of ammonia vapor beneath a shallow glass container; about fifteen or twenty minutes were allowed for the reaction, and the crystals or sediments formed were examined and identified under the microscope. In case of multiple calculi, the largest was weighed and ana- lyzed in the same manner. The smaller ones were also cut and analyzed, but not recorded. The weight of the largest stone and that of the individual layers were calculated, as it was intended to determine or compute as far as practicable the rate of growth. As a whole, the patients on admission were in a state of rela- tively low vitality, which is shown by the prolonged recovery after operation in the majority of the cases, sometimes three months as in case 25. Eleven of the cases gave a positive history or signs (or both) of beriberi. Eighteen were undernourished individuals with an unreliable history of beriberi, all of which belong to the first group of my clinical cases where the data were taken with very little or no stress on the nutritional bearing of the disease. Twenty-three of the series were determined as 484 Philippine Journal of Science 1919 well-developed and well-nourished individuals, all of which except one pertain to the first group above mentioned. The rest of the cases (six in number) received no mention as to their nutri- tional condition. For the sake of clearness, the series may be divided into four groups as follows: I, cases which gave positive signs of beriberi; II, cases which indicated no reliable history of beriberi, but which were undernourished and belonged to the poorer class; III, cases which were well developed and nourished; and IV, cases regarding whose nutritional condition we have no data. FINDINGS AND DISCUSSION The fifty-eight cases studied, with their short clinical histories and the qualitative chemical analyses of their respective calculi, are too long for publication in full in this paper. Examples of the condensed case histories and chemical findings in only four cases, each representing a group, are given to show how the data were arranged for presentation in tabular form. CASE 53 (GROUP I) E. M. (55787), 25 years old, married, male, Filipino, fisherman; born and living in Taliptip, Bulacan, Bulacan; admitted to the hospital November 26, 1917, complaining of painful, frequent, and at times bloody urination of about one year’s standing. Had measles while young, cholera at 10 years of age, and several attacks of rheumatic fever. Denies venereal diseases. He is well developed but poorly nourished. Signs of beriberi present and history of having had the disease is positive. Clinical diagnosis.—Cystolithiasis, pulmonary tuberculosis, and trichu- riasis. Urine examination—Reaction, slightly alkaline; sugar, negative; al- bumin, marked trace; microscopic, abundant red cells, occasionally casts, pus cells, and some squamous epithelial cells. Cystolithotomy was performed November 30, 1917, and the patient died a few days afterward. Calculus analysis (one yellowish sténe). Grams. Whole stone 26.5 Portion analyzed exe 14.0 First and second layers (inseparable) 6.0 Third and fourth layers (inseparable) 5.0 Fifth layer 2.0 Nucleus 1.0 Weight by proportion. Grams. Cortex and second layer (inseparable) 11.4 Third and fourth layers (inseparable) 9.4 Fifth layer 3.8 Nucleus 1.9 XIV, 5 Padua: Cystolithiasis with Dietetic Deficiency 485 Composition of layers. ——. es Pc ala te ee Constituent. Layer. Chief. Minor. Cortex and second layer_______ Amorphous and triple phos- | Calcium carbonate and trace phate. of calcium oxalate. Third and fourth layers ______- Amorphous phosphate _____._. Calcium oxalate, calcium car- bonate, and ammonium ; urate. WUGRIBVOR co os ccc Amorphous phosphate -_______ Do. PUUGIOUS «255 sons ec se Calcium oxalate ......-._.._... Amorphous phosphate and trace of uric acid. CASE 9 (GROUP II) E. M. (37388), 18 years old, single, female, Filipino, student; born and living in Santo Tomas, Batangas; admitted to hospital July 26, 1915, complaining of painful, frequent, difficult, and scanty micturition, and pus discharge occasionally. The condition is of about six years standing. No history of past diseases nor of beriberi. She is moderately developed but poorly nourished. Clinical diagnosis——Cystolithiasis, chronic cystitis, and prolapse of the rectum. : Urine analysis.—Reaction, alkaline; sugar, negative; albumin, a decided trace. Microscopic, abundant pus cells, some blood and mucus. Cystolithotomy was performed August 8, 1915; the patient was dis- charged September 2, 1915. Calculus analysis (one white wregular stone). : Grams. Whole calculus 31.5 Portion analyzed 15.1 Cortex, analyzed portion 5.5 Second layer 5.2 Third layer 4.0 Nucleus 0.4 — Weight by calculation. Grams. Whole cortex 11.5 Whole second layer 10.9 Whole third layer 8.3 Whole nucleus 0.8 : Composition of layers. “2 Constituent. 1 Layer. : Chief. Minor, Cortex. Triple and amorphous phos- | Calcium carbonate and am- phate. monium urate. “ Second layer (brownish | Amorphous phosphate and uric | Ammonium urate and cystine. white). acid sete ie Third layer (brown) .....__.__-| Calcium oxalate ___._._...__. .. | Amorphous and triple phos- phate. Nucleus -------+...-.----------| Triple and amorphous phos- | Uric acid, ammonium urate, phate. and calcium sulphate, 486 Philippine Journal of Science 1919 CASE 23 (GROUP III) E. G. (44226), 26 years old, married, male, Filipino, laborer; born and living in Quingua, Bulacan; admitted to the hospital June 19, 1916, com- plaining of frequent and bloody urination of about one year’s standing. He had fever of short duration during childhood and frequent attacks of malaria at the age of 15 years. He is well developed and nourished. No history of beriberi. Clinical diagnosis——Cystolithiasis, trichuriasis, and ankylostomiasis. Urine examination.—Reaction, acid; sugar, negative; albumin, abundant; microscopic, very abundant pus. Cystolithotomy was performed June 20, 1916. The patient was dis- charged fourteen days after the operation. Calculus analysis (one yellow stone). Grams. Whole calculus 97.0 Portion analyzed 43.6 Cortex, or first layer, analyzed solve Second, third, fourth, and fifth layers (inseparable) 5.3 Sixth layer, analyzed 0.9 Seventh layer, or nucleus ; 0.2 Weight by proportion. Grams. Whole cortex 82.8 Whole second, third, fourth, and fifth layers 11.8 Whole sixth layer ; 2.00 Whole seventh layer, or nucleus 0.4 ° Composition of layers. Constituent. Layer. Chief. Minor. Cortex of first layer (brown- | Uric acid .__..........---------- Calcium oxalate. ish). Second layer (reddish) -._.____|_____ a eat aa ah casio eee re Pee rea Do. Third layer (dark red) -.--..._. Fourth layer (whitish) -.-..... scam oo tt ROE SNS aes se Uric acid. Fifth layer (dark red) --...__.. Sixth layer (brownish) --...... REPO WOU ac oe Sa cousscckesce Calcium oxalate. Nucleus (dark red) .- gterenenatnate Some calcium oxalate. CASE 31 (GROUP IV) B. D. (47613), 4 years old, Filipino boy; born and living in Tondo, Manila; admitted to the hospital December 27, 1916, complaining of dif- ficult and painful urination of about one month’s standing. History of past illness and of beriberi unreliable. Nutritional condition undetermined. Clinical diagnosis —Cystolithiasis and acute abscess of abdominal wall. Urine examination.—Reaction, alkaline; sugar, negative; albumin, an appreciable trace. Microscopic, numerous erythrocytes and a few degen- erated epithelial cells, Cystolithotomy was performed December 29, 1916, and the patient was discharged February 11, 1917. XIV, 5 Padua: Cystolithiasis with Dietetic Deficiency 487 Calculus analysis (one dull whitish stone). Grams. Whole stone 5.5 Portion analyzed 2.0 Cortex analyzed 0.9 Medullary layer analyzed 0.8 Nucleus analyzed 0.3 Weight by proportion. Grams. Whole cortex 2.5 Whole medullary layer 2.2 Whole nucleus 0.8 Composition of layers. | Constituent. . Layer. Chief. Minor. OMELET Amorphous phosphate and cal- | Calcium oxalate and trace of cium carbonate. ammonium urate. Medullary layer _._..._.__.____ Calcium carbonate __..---..--. Amorphous phosphate, cal- cium oxalate, and ammo- nium urate. UIGUS 5.005 S a ee ce ae Gece ee ec Calcium oxalate, ammonium urate, and amorphous phos- phate. The analysis of each layer of an individual calculus is given as each may be regarded as a distinct stone formation. It should be recalled that a stone is formed, in the presence of a binding substance, around a nucleus, which may be a mass re- sulting from an inflammatory process of the vesical mucosa (pus, blood, or bits of necrotic tissue), or foreign bodies such as frag- ments of bougies, pins, silk sutures, or a previously formed stone which may have come from the kidney, or one that has been formed earlier i in the bl ad idder around which other layers of stone _ are formed with the same or different chemical composition. The disposition of the bladder calculus in layers of different chemical substances may be the result of a change in the com- position of urine secondary to a modified general metabolic pro- cess. Sondern,(22) in speaking of the increase of the calculus in size, says that— The abnormal condition during which the nucleus is found may be tem- porary and that the stone can continue to grow in size even if normal urine is excreted. Once the nucleus of stone is formed around a mass of “foreign body” coated by the binding substance or albuminoid framework, the tendency is toward a more or less continuous deposit of prac- tically the same chemical substances in almost the same propor- 2 + 488 Philippine Journal of Science 1919 tions as in the original primary stone. It is possible that with the decline of health, usually secondary to faulty metabolism and nu- trition, unfavorable fermentative changes may take place inside the bladder, which may subsequently give rise to the formation of a second layer of stone of distinct chemical composition, thereby making the primary calculus the nucleus of the second stone which Wells(25) considers a secondary calculus. Because of the injury produced by the resulting calculus, the condition may aggravate or the disease and general nutrition may improve as a result of treatment. In either case, a third layer of stone is formed with different chemical composition around the two- layered already existing calculus as a nucleus. This last layer may be considered as a third stone formation. Incidentally the analysis of several cases of multiple calculi removed (as in case 25) show that the medullary layer—that is, the layer outside of the nucleus—of the larger one is chemically almost indentical with the nucleus of the two-layered smaller one and the cortices, or the outside layers, of both are similar, as indicated in fig. 1. It is apparent that the first stone formation (primary stone) of b (o} --—-+-+---(®) took place simultaneously with . b the second layer of a and that : a _ the two-layered calculus of a Tone we eeewen meu othd of F Hucletie of the third layer, thus playing the réle of the primary stone of 6. Conse- quently, for a given stone there are as many stone formations as there are layers, which may or may not be physically and chemically distinct from one another. From the fifty-eight cases examined there have been made out one hundred eighty-nine different layers or stone formations, the compositions of which are given in the following tables. In instances of multiple calculi the chemical composition of the largest stone is the only one considered. In the beriberi cases, Group I, thirty-six layers, 66.67 per cent, were chiefly phosphatic; and in those cases, Group II, having histories of poor nutritional condition fifty-nine layers, 72.88 per cent, were phosphatic. In the group of well-nourished in- dividuals, Group III, seventy-six layers, there were 40.79 per cent of uric acid and urates, against 36.84 per cent of chiefly phos- phatic layers. In Group IV, where the nutritional condition was not noted in the clinical record, the incidence of phosphatic composition is also higher (nine layers or 50 per cent phosphate) against five layers or 27.77 per cent of uratic concretions. I 489 ciency Cystolithiasis w Padua XIV,5 ith Dietetic Defi Oot 89 oot 9 oot 8% 00T SI oor je a, Behe si 8 Ie ag ‘ BRAS tk 9°8 9 LST I L°8 z a9°9 I 16 Bc ae SO arate “""" geqeuoqaED 8 "SI 8 Lot I wut b age I 3'8t ae nS a hea... he OTs 8T 8°88 z 9°8P or L°9T g SL hears aaa etek ee besa Cheer a ys aa Moe a he ae 8°88 z P08 L 22h st Poh Rees Bes et ee eRe kt er "4U90 Lag *4Uao Lag "9U90 dag “quan Lag *7U99 AAT aor -zequinyy bs ay “zoquin N poate g “z9quInN band ‘zequinN Beste *z9quIn NT “squenj1{sU09 JeryD “[810], “AI dnoay ‘HI dnory "II dnoray "I dnoayy ‘POPONU UA SJUaNnzySU0D foryo fo e2UepwUyT—]] WIAavI, OG OOR | ASE et aes ee LESTE oe tear cher, We a hee 5 Ee eer: BBR a 98 SEGR So ce gree pyar ee tees eee ee tee Spe ie 99°% g 93° T 96° § 69 °T T erctinres rire n,n ee ere (eyeu0gzeo pus oe dsoyd ‘oy8]exo puv oyeydsoyd ‘oywin pus oyeydsoyd) pexipy OL's L TT TL z £9°S td OL'T T s3°g z NES Gah i ~~ (UINIOTBD) e7eUOG.IeD 90°OT 61 9¢°9 T 6L “ST or 8L°9 ¥ 9s°9 rd Ie osananer aa a ce pamela coat 3 get Ca i. ce coeess* (uInyoyBo) O78[BXO 19°86 ¥g LL“LE g 6L OF T§ 96 °9T or ar 8 Tae eee wpe oe “"o>" (97810 UNnTUOUIUTIE puB plo¥ Oran) $93B1(:) 80°S9 | FOL 00°0¢ 6 v8 98 82 88 “SL SP 19°99 Ha VR Ree MeN 2. SEFC (CAnyo]¥o pu ‘e[di23 ‘snoydzoure)soyeydsoyg "9U99 Lag *2U99 49g *qUao Lag "2U92 4ag "9ua0 wag : — | ‘asouop | . ‘souop | , ‘asouep | , ‘aouep | , ‘eouop | . eon SI9kB] hone BIOke'] et s190407] an s10£8'} aor: si0hB"] “S]USRZYsUOe JOIN ‘T2107, “AI dnory ‘TIT dnorwy ‘II dnoay *T dnoiy ‘s4ahmy quasafip ay) fo siuenzysu0s fovya 947 {0 sousprous fo efnyuerseg—y] wav, : 490 Philippine Journal of Science 1919 The totals for one hundred eighty-nine layers show one hun- dred four layers or 55.03 per cent with phosphates as the chief constituents, and fifty-four layers or 28.57 per cent of chiefly uratic composition. Oxalates occurred as chief constituents in nineteen or 10.05 per cent of the layers, carbonates in seven or 3.7 per cent, and there were five layers or 2.65 per cent with mixed phosphate and urate, phosphate and oxalate, and phos- phate and carbonate as the chief constituents. Table II shows the chief constituents of the primary stones. The percentage of primary phosphatic calculi is greatest in the individuals belonging to Groups I and II, in whom the nu- tritional condition was below par. In Group III, where the percentage of urates predominates, the affected individuals belong to a class with adequate food supply. These findings are not in accord with the findings of foreign investigators. Wells(25) and others say that urate calculi ex- ceed the rest in frequency. Ultzmann(24) found that out of five hundred forty-five cases of primary calculi 80.9 per cent wer with nuclei consisting of uric acid (and urates), 5.6 per cent of calcium oxalate, 8.6 per cent of earthy phosphate, 1.4 per cent of cystin, and 3.3 per cent of some foreign body. Gordon’s(13) statistics on his investigation of one hundred calculi which give a predominance of uratic stone in England may here again be recalled. Reed(20) reports a case in which the calculus was composed of urate and concentric rings of calcium oxalate with a colloid nucleus. Kahn and Rosenbloom(15) have analyzed a number of urinary calculi and in their conclusion they state: 1. The large majority of renal stones are composed of oxalate of lime. 2. Uric acid and the urates are found in all renal concretions, but it is rare to find a renal -calculus that is mainly composed of the urates. These same authors present analyses of two vesical calculi and both were made up almost exclusively of uric acid. Lastly Spiegel,(23) in an analysis of a large number of urinary stones, both renal and vesical, finds that only twenty of fifty stone formations contain phosphates, even as a minor constituent. Oxalate stones which, according to Gordon and to Cabot, (6) may be derived from oxalate-containing vegetable foods, such as rhubarb and sorrel, are not common among Filipinos. While Spiegel and others believe that calcium carbonate calculi may originate from the excessive use of hard water (as is the common well water of the Islands), such stone formations are infrequent in this series. I have not found xanthine, urostealith, or fibrin calculi, which are occasionally mentioned in the literature. xiv,s Padua: Cystolithiasis with Dietetic Deficiency 491 An inference might be drawn from comparison of my findings among Filipinos with those among Caucasians. The association of phosphatic calculi in the cases here classified as undernourished (with an unbalanced diet), and the frequency of urate stones among a class of people with nourishment sufficient for normal physiological processes are points that strongly agree with the incidental findings of Osborne and Mendel(19) in their experi- ments upon animals. As a whole, the total phosphatic estimates give an incidence of more than half of the total number of calculi among Filipinos. It is possible that quite a number of the cases at hand, although suffering from the so-called latent form of beriberi or other nutritional-deficiency disease, did not give any history nor present any suspicious sign of the disease at the time of admission, and in many instances the malady may have been entirely overlooked (Manalang,(17) Saleeby(21)). All but one of my cases (cases 49 to 58, Group I) gave a positive his- tory of having had beriberi, and in this particular case the stone was essentially oxalate in composition. In only two of the beri- beri cases were the calculi made up mainly of urates, and there- fore the percentage of phosphatic stone formations in association with this disease is greatest. : The character of the stone deposit is partly dependent upon the influence of the urine reaction. Thus Ballenger and Elder, (3) in speaking of the cementing substance that holds the crystals together to form the calculus, say: This doubtless varies with different stones according to the acidity or alkalinity of the urine. Stones composed of uric acid, urates, calcium _ Oxalate, cystin and xanthin develop in acid urine while those consisting of calcium carbonate and acid phosphate of calcium arise in alkaline urine. Stones of ammonium or magnesium phosphate are precipitated from stagnant urine with local infection and inflammation. Spiegel, in his series of chemical analyses, has corroborated this fact. Kahn and Rosenbloom, however, slightly differ from this view. These authors state: While uric acid and acid salts are soluble in alkaline medium and _ insoluble in acids, the exact opposite holds good for calcium oxalate and calcium phosphate, which are deposited in alkaline medium and dissolved in acids. Blatherwick(4) says: High urinary acidity favors the formation of uric acid calculi, which comprise from 60 to 81 per cent of all urinary concrements. Wells states that: “Phosphate calculi are formed as a result of decomposition of the urine, with the formation of ammonia from the urea. In the ammoniacal solution 165559 —3 492 Philippine Journal of Science 1919 thus formed the magnesium is precipitated as NH:MgPO,, the calcium as Ca; (PO,)2 and calcium oxalate and ammonium urate are also thrown down, so that the concretions consist of a mixture of these substances, the magne- sium salts being the most abundant.’’ He then goes on to say that “the formation of phosphatic concretions is always a matter of urinary reaction and not of diet.” Due to lack of data ag to the reaction of the urines before the onset of the disease, the present urine examinations are to be correlated with the chemical composition of the cortex or outside layer of the stones. In my series there are cases of alkaline urine where the stone is covered with uric acid or urate cortex. The latter phenomenon may be explained by the presence of inorganic phosphates, or by the fact that the alkaline fermenta- tion has not yet been brought to completeness. The opposite condition may be true with the phosphate or carbonate stone in an acid medium. The neutral urine, on the other hand, may be a transitional stage in the process of ammoniacal fermen- tation in the urine brought about by infection, or may result from the so-called “alkaline tide.” That food influences the reaction of the urine is shown by Blatherwick, who states: In general, foods yielding an alkaline ash were found to decrease urinary _ acidity, while those yielding an acid residue increased it. The urine of Filipinos is normally less acid than that of Amer- icans or Europeans, owing chiefly to the low protein content of their diet, which consists chiefly of rice and a little fish among the poorer classes. The reaction of the urine from the cases on admission to the hospital may be compared with the composition of the cortices of the calculi in. forty-nine cases of the series. Thus there were found in acid urines nine phos- phate and nine urate or uric-acid cortices and one oxalate cortex; in alkaline or neutral urines, there were twenty-three phosphate and four uric-acid or urate cortices, one oxalate cortex, and one oxalate and phosphate cortex. The results show a tendency toward the formation of phosphate concretions in neutral or alkaline urines in accord with the literature. Taking the composition of the nuclei of the stones for com- parative reference in regard to the age when the first symptoms of cystolithiasis developed (which may not coincide with the exact date of the beginning of the formation of the urinary stone), we have Table IV. We notice that among these fifty-eight cases there was a greater incidence of stone formation in children and young adults, contrary to the opinion of Bugbee,(5) who found a greater fre- quency of vesical calculi among patients past middle life. It xiv,5 Padua: Cystolithiasis with Dietetic Deficiency 493. is also to be noted that in the first four decades, and even among’ the undetermined cases, except one who is more than 50 years of age, the phosphatic calculi take the lead, while in the fifth and sixth decades of life, the urates predominate. This finding is contrary to that of Cabot who states that uric acid and cal- cium oxalate are often found in the urine of poorly nourished children due to imperfect oxidation of the tissues, and of old individuals who take food beyond their power of assimilation. Also Ellinger (10) considers stone in children to be almost always uratic in composition. TABLE 1V.—Age tnetdenge, with reference to the composition of the nuclei. Age in years. Chief constituents, 1to 14. | 14 to 20. | 21 to 30. | $1 to 40. | 41 to 50. 51 to 60. | 61 to 70. Per cent. | Per cent.| Per cent. | Per cent. | Per cent. | Per cent. | Per cent. Phosphates 2:2. 2... 225. 52.9 100 85.3 50 66.7 OO Ve aS RITORON oo ee Oe ye Beyer aie. 23.5 16.7 16.7 80 100 Oxalates 22 ee 35.3 16.7 ee lee seal ee Carbonates_........... Veet: eae oe 5.9 OO ee ee ah on Total (cases) -___. ...__- 17 1 17 6 6 5 2 Chief constituents. Pre SP Total. 1 | Per cent. | Per cent. PS DOU ccel ewtacon eee cae ee 75 46.6 WINNS Sasa ee Soe ee RE Et STE 25 31.0 ORO GS A a ee ee eae og et he 13.8 om spies ges dos ee DE POORER DED Ora Sense Cam ee ee 8.6 PH NON oe conch en a ke 4 58 The occupation of the patient at the time when the stone was removed may have a bearing on the composition of the latest stone formation, which is the cortical layer. Table V gives an idea of the chief constituents of cortex with reference to the latest occupation of the patient. It is noticed that stone formation is of more frequent occur- rence among ordinary laborers and farmers; and in all branches of occupation the phosphatic calculi are most frequent in inci- dence. Concepcion has pointed out the insufficiency of the earn- ings of the average unskilled laborer for the proper maintenance of a family under sanitary conditions. The general poverty of the laboring class naturally predisposes them to nutritional dis- orders. As a rule farmers are vegetarians, and their circum- stances of living are such that they consume very little or no 494 Philippine Journal of Science 1919 TABLE V.—Incidence of chief constituents of cortex of calculus as related to present occupation. | Boys Chief constituents. Laborers. Farmers. be ng Students. poco Total. | | tion PRGA css sicnme reer eo eee: - 8 10 6 | 4 10 38 UPObes Ss Soria, casa ae caus q _ ty ERS Se soos 1 3 15 Oamle tes ooss pesos. sak hee ee 1 - 3 ae ee jg Se eS 3 Phosphates and oxalates _-___.--.--_- S EI RARE 30 ARON Baheae BAS estaekcre si 1 Phosphates and carbonates -_.._.._.|_-..-.---|--------.. bdledecccwesuledanassuse 1 RetehGs. cds aad 17 15 7 6 13 58 ® Carpenters, sellers, plumbers, fisherm@ny ete. meat. Boys without occupation would naturally be dependent on the earnings of their parents, whose income is, in many in- stances, not sufficient for the maintenance of a normal physio- logical existence. Students in the public schools who come from the average family do not as a rule receive enough allowance to provide them with a wholesome and nutritious food on account of the high cost of living in the city. The remaining cases (in- cluded under miscellaneous) belong to a class of occupations with relatively low earnings and consequent low standard of living and probable deficiency in nutrition. The distribution of cystolithiasis may be of some value in tracing the nutritional relations, particularly of the phosphatic calculi. Foreign authors attribute the irregularity of distribu- tion to the habits of the people, their mode of living, selection of food and drink, and climatic conditions (Morton(18)). This investigator in his clinical lecture delivered at the Long Island College Hospital pointed out the prevalence in India of stones that are primarily uratic, infiltrated with oxalates and triple phosphates, and only found among the legumen eaters. Gordon also found a predominance of uratic stones in England, and says: “Tt has been attributed, we think erroneously to eating meats.” The same author attributes the prevalence of stone in Holland to the hard-water supply, contrary to the opinion of Lewis(16) who maintains that: The problem of oxalurics is not to prevent the introduction of calcium salts but the absorption of exogenous oxalic acid in the foodstuffs. In the Philippines the phosphate stones predominate in almost all of the provinces concerned except Pampanga. Table VI gives this incidence, and the character of the stone here is determined by the chemical composition of the cortex, as this may be affected _XIV,5 Padua: Cystolithiasis with Dietetic Deficiency 495 by the regional conditions under which the patient was living at the time immediately previous to the removal of the stone. TABLE VI.—Number of cases of cystolithiasis by provinces and by con- stituents of the stones. | Province Phos- | urate. | Oxalate eo ae oT | Total : phate. es | oxalate, | Carbon ate. | SSeS RA UAAR PMP ES OE WET aie ce ee Wee, mg has Sh ee ee are poe ee 2 beara Ace ee te eee 1 OAM ee ea 7 Biden SE eee cases 9 fui Tao» BROPRE ante hot on dice ie aa eR 6 3 a UE sey a ete emer 10 CONE sesh et ee z Brera eee oes he co 4 (FT i | ene an cae ae ote eee oe Ag Bee ates ere ee LC PN LoS ae eam: 2 * EOS 0 1111 Sagopa ge arora a Wide Bisse Cbg ee re, eee, 1 MOMs ice ee 4 pe ES aS Rate 1 6 Nweva- Meija i236 05 es oo ig OS oes bs BaP ERS tne BS ee ay ce 5 Pampingeie. 2219.00 tice os a 3 Cy ea OED BE nas eR ae 8 Pangasinan <2 Se » A, ae intact at | SAE BT ABE tes 2 MiG os ee 4 Dy feeb oe ee ee 6 AGROB oo sg oe ig 1 Bee A ee ee 2 4 Mambalet: oc siete cl eae Rati Ae os ee eS oe ee 2 The social condition and sex may here be disregarded because they would not have any influence in changing the dietetic con- dition of the patient. SUMMARY AND CONCLUSIONS The observations of Osborne and Mendel on rats indicate a possible association of dietetic deficiency with the formation of phosphatic urinary calculi. The Filipino diet is essentially of an insufficient and limited character, particularly in its vitamine content. Beriberi is still frequently encountered in these Islands, and Saleeby and others have indicated the frequency of the dis- ease in a latent form which, with further limitation in diet, may become active and widespread. My personal experience in the Philippine General Hospital confirms this view. The results of the present investigation show that a relation apparently exists between the general dietetic inadequacy and deficiency among Filipinos and the incidence of phosphatic calculi, in con- trast with the reported predominance of uric acid and urate calculi in Europe and the United States, ' Of ten cases of cystolithiasis which I have particularly studied since this investigation was undertaken, nine not only gave a history of having had beriberi, but exhibited actual signs refer- able to the disease. Although the clinical data of all but two 496 Philippine Journal of Science 1919 of the other forty-eight cases are, as a whole, obscure with respect to the history of nutritional disease, still they indicate that the majority of the patients were undernourished and un- derdeveloped. It is probable that among these forty-eight cases, there were many, besides the two, who were beriberic either in active or latent form, as the symptoms of beriberi were not especially looked for in preparation of their clinical histories from a purely surgical standpoint. The cases studied have been classified into four groups; (1) eleven cases, nine of which were my own, giving positive his- tory or signs of beriberi; (II) eighteen cases with an unreliable history of beriberi, but undernourished and from the poorer class; (III) twenty-three well-nourished and well-developed in- dividuals; and (IV) six cases of uncertain nutritional condition. The percentage of primary phosphatic calculi (nuclei), as de- termined by qualitative tests, is greatest in the individuals belonging to groups I and II, being 45.4 and 72.2, respectively. Primary urate calculi were most frequently found in the well- nourished cases (group III), occurring in 43.5 per cent of these cases. Of the one hundred eighty-nine layers from fifty-eight stones examined, one hundred four layers or 55.03 per cent were phos- phatic in composition; fifty-four or 28.57 per cent, uratic; nineteen or 10.05 per cent, oxalates; seven or 3.70 per cent, carbonates; five or 2.65 per cent, mixed phosphate and urate, phosphate and oxalate, and phosphate and carbonate. A majority of the cases occurred in children and young adults. Uric acid or urate nuclei were found in six out of eight cases over 50 years of age, while phosphates predominated in the earlier decades. The classification of the cases as regards occupation indicates the prevalence of stone formation chiefly among laborers and farmers. The provincial distribution of the cases is as follows: Bulacan, 10; Batangas, 9; Pampanga, 8; Rizal, 6; Manila, 6; Nueva Ecija, 5; Cavite, 4; other provinces, 10. Phosphate cor- tices predominated in all occupations and in all the provinces except Pampanga. : Among ,forty-nine cases the chief constituent of the cortices was phosphate in twenty-four cases, with a neutral or alkaline urine, and in nine cases with an acid urine. The reaction of Filipino urine is normally acid, but to a less degree than in the United States and Europe because of the low protein intake. It is apparent, therefore, that the inadequate dietetic conditions xIv,5 Padua: Cystolithiasis with Dietetic Deficiency 497 among Filipinos and concomitant nutritional disorders such as beriberi favor the formation of phosphatic stones, owing to the low daily protein intake and the lessened acidity of the urine resulting therefrom. I wish to extend my profound gratitude to Dr. Potenciano Guazon, chief of the department of surgery, College of Medicine and Surgery, University of the Philippines, for the exceptional opportunities he has given me to utilize the stones in the museum of the operating room of the Philippine General Hospital for chemical analysis. To Prof. Bowman Corning Crowell, chief of the department of pathology and bacteriology, University of the Philippines, I express my sincere appreciation for his strong encouragement in the present study. I acknowledge the valuable codperation of my colleague, Doctor Aguilar, who has many times taken my place when on duty in the hospital. But above all, I wish to express my due acknowledgment and most hearty ap- preciation to Prof. Robert Banks Gibson, chief of the department of physiology, University of the Philippines, for his suggestions and priceless help in the preparation of this paper. BIBLIOGRAPHY (1) ANpREws, V. L. Infantile beriberi. Philip. Journ. Sci. § B 7 (1912) 67. (2) ARoN, Hans, and Hocson, FELIx. Phosphorus starvation with special reference to beriberi: IH. Philip. Journ. Sci. § B 5 (1910) 98. (3) BALLENGER, E. G., and ELprer, O. F. Stones in the urinary tract. New York Med. Journ. 105 (1917) 203-205. (4) BLATHERWICK, N. R. Diet in the treatment of nephritis and urinary caleuli. New York Med. Journ. 102 (1915) 362. (5) Bucper, H. G. A clinical study of lithiasis based on a series of one hundred and ninety-eight cases. Journ. Am. Med. Ass. 69 (1917) 1492-1498. (6) Casot. Keen’s Surgery. New York 4 (1914) 335. (7) CHAMBERLAIN, W. P., VeppER, E. B., and Wriuiams, R. R. Etiology of beriberi. Philip. Journ. Sci. § B 7 (1912) 39. (8) Concepcion, I. A study of the nutrition of the Filipinos. Read before the IV Asamblea Regional de Médicos y Farmacéuticos de Filipinas, February 4-8, 1918. (9) De Haan, J. On the etiology of beriberi. Philip. Journ. Sci. § B 5 (1910) 65. (10) ELiincer, ALEXANDER. Harnsteine. Oppenheimer’s Handbuch der Biochemie 3' (1910) 681. (11) Fraser, Henry, and STANTON, A. T. Etiology of beriberi. Philip. Journ. Sci. § B 5 (1910) 55. (12) Gipson, R. B. Influence of compensated salt mixtures on the develop- _ment of polyneuritis gallinarum and beriberi. Philip. Journ. Sci. § B 8 (1913) 351. 498 (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) Philippine Journal of Science GoRDON, G. S. Origin of the various constituents found in urinary stone. Northwest Med. n. s. 3 (1911) 110. HicHet, H.C. Beriberiin Siam. Philip. Journ. Sci. § B 5 (1910) 73. KAHN, MAx, and RosENBLOoM, JAcos. A report of some new chemical analyses of urinary calculi. Journ. Am. Med. Ass. 59 (1912) 2252. Lewis, B. Multiple vesical calculi: over 1700 present. Trans. Am. Urol. Ass. 8 (1914) 310, MANALANG, C. Degeneration of peripheral nerves. Philip. Journ. Sci. § B 12 (1917) 169. Morton, H.H. Stone in the bladder. Medical Times 45 (1917) 37-389 OsBoRNE, T. B., and MENpEL, L. B. The incidence of phosphatic uri- nary calculi in rats fed on experimental rations. Journ. Am. Med. Ass. 69 (1917) 82. Reep, T. W. Vesical calculus: Report of a case. U.S. Naval Med. Bull. 10 (1916) 680. SALEEBY, N. M. The most important medico-economic problem of the Philippine Islands. Philip. Journ. Sci. § B 10 (1915) 88. SONDERN, F. E. Analysis of differentiated urine. Post Graduate 26 (1911) 64. SPIEGEL, L. Uther die Zusammensetzung von Nierensteinen. Ber. Deutsch. Pharm. Gesellsch. 9 (1899) 318. ULTZMANN. Cited by Mandel, T. A., in Hammarsten, Olaf. Text Book of Physiological Chemistry 51 (1911) 616. WELLS, H. G. Chemical Pathology. W. B. Saunders Co. (1907) 380. Wittiams, R. R., and SALeEBy, N. M. Experimental treatment of human beriberi with constituents of rice polishings. Philip. Journ. Sei. § B 10 (1915) 99. THE GROWTH OF HEVEA BRASILIENSIS IN THE PHILIPPINE ISLANDS By Harry S. Yates (From the Botanical Section of the Biological Laboratory, Bureau of Science, Manila) ONE TEXT FIGURE Probably no agricultural industry within the Tropics has un- dergone so remarkable a development during the past twenty years as has the plantation-rubber industry. Hevea brasiliensis, the principal source of plantation rubber, was first grown under cultivation when it was introduced into Ceylon about 1877.1 Very little further planting was done until the experimental stage was passed and seed became availahle in quantity. The beginning of the cultivation of Hevea in Ceylon may be said to date from about 1890 to 1895 though for the first ten years development was comparatively slow. In Malaya Hevea attracted very little attention until about 1897 , when, coffee becoming less profitable, planters saw the need of some product to supplement or replace it. However, when the Ma- layan planters did turn their attention to rubber, the develop- ment of the industry, stimulated by the high price prevailing for the product, took place very rapidly; and at present Malaya produces a very large share of the plantation rubber of the world. _ The cultivation of Hevea having proved successful and profitable in Ceylon and Malaya, many plantations have been started in other countries; more especially in Sumatra, Java, Borneo, India, Burma, Cochin China, New Guinea, and other places in the Asiatic Tropics, the Gold Coast and elsewhere in Africa, and in various parts of tropical America. The area planted to rubber in Malaya alone up to and including 1916 was 385,372 hectares (951,870 acres),? while the total area planted to rubber in the * Those interested in an account of the early development of the industry are referred to Petch, Notes on the history of the plantation rubber industry of the East, Ann, Roy. Bot. Gard. Peradeniya 5 (1914) 433-520; and Burkill, The treatment to which the Para rubber trees of the botanic gardens, Sin- gapore, have been subjected, Gardens’ Bull. 1 (1915) 247-295. *{ndia Rubber Journ. No. 18, 55 (1918) 10. | ne 501 502 Philippine Journal of Science 1919 Dutch East Indies in 1916 has been estimated at 209,312 hectares (517,000 acres) .* Although the United States takes considerably more than one- half * of the world’s supply of crude rubber, very little is pro- duced in American territory. Unmanufactured India rubber imported into the United States during the year ending December 31, 1917, amounted to 184,381,035 kilograms (405,638,278 pounds),° valued at 466,441,808 pesos. It is true that a part of this was wild rubber from South America, Africa, and else- where; but the imports of crude rubber in 1917 from the British and Dutch East Indies alone amounted to 115,915,868 kilograms (255,014,910 pounds), valued at 300,303,192 pesos. In view of this very large importation of rubber into the United States and the importance of the industries there dependent upon it, it would seem very desirable to develop a rubber-planting in- dustry within American territory should land and climatic con- ditions suitable for the successful growth of Hevea brasiliensis be found. There has long been a question as to the suitability of certain sections of the Philippines for the growth of Hevea brasiliensis, but notwithstanding its remarkable development in nearby coun- tries the industry has received very little attention here. How- ever, a small area of Hevea brasiliensis has been planted and is now producing rubber in profitable quantities; and, while it must be admitted that the industry has hardly passed the ex- perimental stage, it has been demonstrated that rubber will grow and yield satisfactory returns in the Philippine Islands. During the latter part of 1917 I had the opportunity of visiting the Basilan Plantation Company’s estate on Basilan’ Island and while there made a number of girth measurments of trees of various ages. The results of these measurements are given in the present paper together with data regarding the climatic and soil conditions in the southern part of the Philippine Archipelago. The rate of growth of Hevea brasiliensis in the Philippines is compared with that recorded in other countries, and a comparison is made of the climatic conditions and soils of this part of the Philippine Archipelago with those in other countries, more especially in the Orient, where Para rubber is now being successfully cultivated. *India Rubber World 54 (1916) 382. * The India Rubber Journal No. 19, 55 (1918) 10 quotes a statement that in 1917 the United States took 71 per cent of the entire crude-rubber production of the world. "India Rubber World 57 (1918) 384. XIV, 5 Yates: Growth of Hevea Brasiliensis 503 CLIMATIC CONDITIONS SUITABLE FOR THE GROWTH OF HEVEA The original home of Hevea brasiliensis is in tropical South America along the Amazon and its tributaries, a region having a moist warm climate, and a fairly uniform distribution of rainfall. Meteorological data from the observatory at Belem (Para), located near the coast and slightly south of the mouth of the Amazon, indicate a climate with a rainfall averaging about 2,500 millimeters fairly evenly distributed throughout the year, though with a dryer season during September, October, and November. The temperature appears to be fairly uniform during the year; it averages about 25° C., and rarely falls below 20° C. or rises above 87° C. The distribution of rainfall by months and the number of rainy days in each month at Para are shown in Table I. TABLE I.—Rainfall at Belem (Para), State of Para, Brazil." [Rainfall is given in millimeters.] 1907 1909 1910 Month. G58 Rainfall.| Days. | Rainfall.| Days. | Rainfall. Days. SaMesy oc. a 208. 9 27| 295.8 24| 262.4 30 | Sd daeg foe ee a ee 361.4 27; 387.2 25} 253.0} 25 } WR TREE 284.3 30| 219.8 27} 404.5/ 31 | BORE oe 174.6 27} 349.4 25 | 246.5 27 MAS FSS ea se 283.7 28} 262.5 27| 275.8 23 WON ele pie ce ee 242.7 26 132.5 22! 333.0 23 IO 5 Be Se ee 178.8 24 188.4 14} 252.5 19 Ansett. oo Se 188.8/ 17 117.5 10} 153.4 22 Seperate. ek a ee ee 81.8 | 23 89.9 21 101.1 22 NRO OR 55 gS ed ete 67.3 15 93.1 21 76.6 22° NaVeMber ce as 110.6) 22 89.5 18 82.5| 22 | Metiabher oe ae TAN 260.1; 22 168.2 16| 282.5 21 WON cs ee 2, 388.0 288 | 2,828.3; 250] 2,793.8! 287 "From Bol. do Museu Goeldi 7 (1910) 48. ‘When a plant is introduced into new regions, it will usually be found to succeed best where the climate approaches that of its original habitat; and this rule has been found to apply to Para rubber. The successful growth of Hevea is not strictly limited to regions where conditions are precisely similar to those prevailing in the Amazon Valley; yet it certainly finds its optimum habitat in regions at comparatively low elevations, possessing a climate with a rainfall not much below 2,000 millimeters, evenly distributed throughout the year, and with a temperature rarely if ever falling much below 19° or 20° C., at least for any extended period. The 1919 Philippine Journal of Science 504 *9z «(PIGL) NBaaNg doyiwaA ‘e[IUBL ‘seuIddyyg oy Ul [peyuret Jo uoINgLysip pus JuNowe [enuUYy ‘Ose, BiIepeg Woy ev 0 082 % < 6 880% vPL T‘T18 ‘T at | oveor't | emt 6°988‘T 99T 6°99 'T a tala s[BIOL, 3 6L6 v6 0*83aT 9°S88 ¥008 STS OvTLT 9°66 361 $°S9T o StL 9°@9 618 9 SET $002 088T TL 0vE 9 “OLE 6 '28T b Tar 6 ‘08T 2°18 ‘~sABp Auyey “Treyayey “sep Aurey “Treyurey 6 9°00T b 88s 9°aT8 §°99T 918s P°SbT 0°98 $°38 $00T td 218% 3 g°LT b 2°F% TI 616 L § CIT 6088 8 °SbL P'OtL 2 OFT L 0°¢S b COT Z 96 0 00 I g"t 3g @s1 “sep cares | UPFUIE “sAep Au 1ey “TTeyusey 9 STs 8°88 aa 8 “OTS 0°68T SLL 8" b'88 3908 S°9L GSLs 6°68 8% 9°86T 8686 6 “L8T een nnn n nen nena nen n eterna er cnet a ns 9"="""79qUIaAON cpiGacckek vous bees bnsagseeenennue- + i~- i seeuee adequieydag ee 6 Oe ooo Hee SE RR RS TRE RE RE eH eee See qsnsny onnnee haiti legsean viele oar Ae =i a Rr ane e NI |. * a sowen é Ue Wibkesqs= nnen chin eeaes aune we ee ee en os een ee meee eee anne AUR prea ane st + veh vghacs i atenes\ dear } Passes) * Group Il]. CASES OF HIGH SOCIAL STANDING OR THOSE WHO SHOULD HAVE ADEQUATE DIET. 1 2) M SO) Mi = |-Laboret si. - sas Calumpit, Bulacan_-_..-_--_- 6 25.0 84. Pace] 6 Ce 56 se iL | Pe a. 60, Sek 66:7) Ces 2 2 4|/ M 68:j Mees 0335 SSeS Bacsor; Cavite. SS 1 21.0 13.6] U. Pa, 628 5-0 Pa si $00 1- Us 3 6| M EB «Se Se vi ese Secs eure Baliwag, Bulacan.-._..._---- 2 62.5 1:54 Vase. Pt, Ce, Uas_..-- tke ee Ge WS 7.5) Vai.225 PS Oc. 4 7|M SRB hice AG eo Apalit, Pampanga -._---.---- 12 184.0 8861+ 02S Pa, O65 od 2.4 | Oc 5} 10| M St). ME | Pater. 2c Lipa, Batangas -__--.---_---- 1 Sg eee 2 = cea 0.5 | Pa 2:6 | Passe. 6 14| M 10+.:8 | Student 27a cS Britta; Manila. sce 6 38.0 18.9 | Pt, Pa-j}-Ua, Oe; Ce. 2.22) Ces 7 15; M 5-4 8 i Note co ec c Agoo, Ga Union: 2s: =. 33. 2 6 6.0 4.5| Pa Ua, Ce Sst 0.5") Us: 8 146; M $2 + Laborer os eg Salinas, Cavite. ..........._. 3 98.0 60.6 | Pt, Pa__| Ua, Ge. 2.0° 1. Rt Pa. 9| 17; M 291 S 10 18.1 9:44 Pec.) Ue eek oe 0:6} Ph wesw toes = 10} 20| M 53| M 28.5 10.6 | Us. [Pa Ra $92) Usa t Pe, Bu 11} 23] M {| 2] mM 97.0 U.S 1Oe'S ; o 04 1c. | 12} 24|M | 40| M 5 27.0 Spee Pe, Ua, Oc ----- baat ee Pe, Ua, Oc------ BE: ee Pe, Ua; O62 2.-<.)c segue -~= 652555: 13| 28| M 42; M 1 28.5 eee) Oc, Ce..:.c..2.|-- &8 | Picco Uae, Oe, Ce ae a ae Ua, Oc, Ce 14} 29; M 6} Ss 5 Beg ee ee. 4 Pa... nso i.) . 82 LOC LR. Pe Use f 15} 36) M |} 25| M 2. Ot Oe ay Oe. Oc. sc ecec sl 27 | Pa Ob cia a a ee PIG} SSE 60) M % CTL Sees le eh Sat 17| 33; M | 2| M ee a 25.0 Raves 18; 39|M | 451M get B40 pies Sela . 19 48 M : 85 E M ~ 2 He ‘ 93.5 bh = 8 ee eee 1 : Vee = 20; 44; M/ 4; 8S a ns See Saath Se ari he) Wen J REA. a —---- 21, 47|M | 16| S Tiere SED) 3 Be eee i ae ee, 22| 48; M | 68|M oe es ee woe Sioa 23| 50; M | 2/| S ey 14.5 12| M s 21| M | | -s 380] M M | | 42) M a 4) M a: XIV, 5 Yates: Growth of Hevea Brasiliensis 507 distributed rainfall of any of the stations listed. —— is, however, a very close second. Rainfall data from various localities in the East ide Para rubber is being successfully grown are compiled in Table IV. TABLE IV.—Average rainfall in rubber-producing districts of the Orient. [Rainfall is given in millimeters.] Ceylon, 1905-1908.= Malaya. — Borneo, 1910-1918.4 | Month. Kuala | Paya, | godas | FoM B: (Ine Toi coin toite Sandakan. | Jesselton. 1915-1916.b 2 |< January ....-....- 67.0 47.8 74.9 822.3 50.0 801.2 106.9 February _..______- 66.5 95.8 53.3 88.3 80.8 249.7 12.9 eS See ae ai 100.0 125.5 201.6 228.1 237.5 187.2 45.7 Aor oe 137.4 220.7 297.4 253.7 212.3 117.5 138.1 May oe 133.1 304.8 136.7 132.5 316.5 126.5 209.8 ee 241.8 234.6 170.9 163.3 251.2 114.8 807.1 Silvie 211.8 111.2 228.2 211.8 802. 8 154.4 209.0 Avdgast 23522253 122.9 74.4 237.9 157.5 224.5 212.8 249. 4 September _________ 168, 1 123.9 183.7 148.0 354.1 364.5 853.1 Ontobarc2055 | 874.4 449.6 188.3 149.9 458.2 227.8 436.4 November ________- 173.7 361.7 193.2 252.2 805.6 292.6 280.2 ! December ___._____. 131.0 65.5 167.3 237.0 152.1 364.5 139.2 Lote 1,927.2} 2,215.5} 2,083.4; 2,844.6] 2,945.6] 2,663.0] 2,542.8 * Report of the director, Royal Bot. Gardens of Ceylon (1906-1909). > Supplement to the Official Gazette, F. M. S. Selangor Administrative report (1916-1917). © The Gardens’ Bulletin, Straits Settlements (1914-1916). 4 Manual of Statistics of British North Borneo (1915). TEMPERATURE Hevea brasiliensis appears to reach its best development in regions having a comparatively high and uniform temperature throughout the year. At Belem (Para) near the mouth of the Amazon and about 1° south latitude the minimum temperature recorded during a period of three years was 20.2° C., and the | maximum, 36.6° C.; the average yearly maximum for this period was 32.04° C., and the average yearly minimum, 22.23° C. In the Philippines the lowest temperature recorded during a period of three years (19138-1915) at Davao was 19.0° C.; at Cotabato, 19.2° C.; and at Jolo, 20.2° C. During this period the maximum temperature at Davao was 36.0° C.; at Cotabato, 36.0° C.; and at Jolo, 33.4° C. In general the temperature in the Mindanao region appears to range between about 20° and 34° C, and is fairly uniform throughout the year, except for a somewhat warmer period during February, March, and April, when the temperature occasionally reaches 36° or 37° C. 165559—4 * 508 Philippine Journal of Science 1919 SOIL Hevea brasiliensis does not appear to be exacting in the quality or kind of soil necessary for its successful growth. Generally speaking, Para rubber may be said to do well on any soil that will support a heavy forest growth, provided drainage is good. Hevea brasiliensis is rather intolerant of excess of water in the soil, and very wet soils should be drained in order to obtain the best results. Heavy clay soil is not so suitable for the growth of rubber as are loamy or even sandy soils. Cox * has shown that, in general, Philippine soils are good and contain ample quantities of the inorganic compounds necessary for plant growth. There are undoubtedly very large areas in the Mindanao region where the soils are suitable for the suc- cessful growth of Hevea brasiliensis. Table V gives the results of an analysis of five samples of soil taken on the Basilan Plantation Company’s property on Basilan. TABLE V.—Soils from Basilan plantation. * Sample No.— 1b Qe Be | 44 be ittroment UNS) ce oo Gea eee 0, 093 0.119 0. 108 0.127 0.134 Phosphoric anhydride (P2Os5) -....._____-___.__ 0. 063 0. 078 0. 086 0. 067 0. 075 Rates (OW) ook on eee ado ek es 0.10 0.27 0.11 0. 22 0,14 Poteel: CRO se 0.077 0.083 0.064 0. 089 0.119 PEWINOB oeeeetare cus tens ae 1.18 1.40 1.16 1,29 1.22 Loss on ignition _____ Scbckoae aceon 13.80 13. 92 14.16 14. 60 13. 72 Detritus, not passing 1 mm sieve _____________. 40.4 61.1 32 Ba | 89.6 45.8 Coarse sand, 1.0 to 0.56 mm _..-_.-..._..- 111... 4.8 2.8 2% 4.1 4 Medium sand, 0.5 to 0.25 mm ________-.-22.2... 4.7 2.9 2.5 3.9 14 Fine sand, 0.25 to.0.10 mm-_-___.._-.....2.-.-..- 4.0 3.6 4.2 4.4 7.2 Very fine sand, 0.10 to 0.05 mm________.-._.._. 6.1 6.9 14 8.0 6.7 Silt, 0.05 to 0.01 mm ____._. ae 4.6 4.6 y 8 Be § 2.4 0.7 Fine silt, 0. 01 to 0.002 mm ..-.. 2-2 6..02.52 2... 18.9 28.6 24.8 22.5 22.0 Clay, less than 0. 002 miti.: oc 56.9 50.6 53.5 54.7 60.3 Total (except detritus) __-_.-_. 22.28, 100.0 100.0 100.0 100.0 | 100.0 2 Analyzed by A. S. Argiielles, Bureau of Science, Manila, P. I. » Rolling land; trees (Hevea) poor. © Sandy loam; coconuts. 4 Gently rolling land; trees (Hevea) good. © Swampy land; Hevea. * Philip. Journ. Sci. § A 6 (1911) 279-830. XIV, 5 Yates: Growth of Hevea Brasiliensis 509 _ The samples were taken at a depth of about 30 to 40 cen- timeters. Sample 1 was taken from a block planted to Para rubber in which the growth of the trees was very poor. The trees presented an unhealthy, ill-nourished appearance. It is interesting that analysis of the soil from this block showed it to be low in all inorganic substances necessary for plant growth. In Table VI are given analyses of other soils from the Min- danao region of the Philippines, in Table VII analyses of Ceylon soils are tabulated for comparison, and in Table VIII analyses of soils from the Federated Malay States are given. The Ma- layan soils are characterized by their richness in nitrogen. TABLE VI.—Philippine soils. * Lake Lake i | Fort Lanao, | Lanao, | Saran- | cota) BIR, | naeh | auth | eae | cm. 10to 40 | 10to40 | em. cm. em. MMU TG ING gc ey a eee ee 161 159 150 153 165 Motetute <2 ee Se eee ee 8.35 18, 47 16.61 3.56 4.48 Toes on lgnition 2.255 6 ee 10.16 11.39 19.47 7.80 5.58 Nitvoweri CN} oo. Se ees 0.207 | 0.184 0. 216 0.173 | 0.115 Phosphoric anhydride (P205) _____..______-__- 0.142 0.196 . 0.180 0. 097 0. 188 Die CORON so ee i ee 0. 87 192) FT 2.24 8.18 Magnesia. tig) 2 ee ae 9 oes Geren. Meet Dancer aerate 0.71 Datel CRW oo 0.29 | 0.55 | 0.240 | 0.173 | 0.24 RGR SINREOD 6s oa Se ee He ta Be Sag ek, SPR ONCTi, IRE Re eee ee WRONG s tescs Sareea mis ics Ce ec nk 3.25 2.12 2.55 2,17 Soil acidity (per cent CaCOs) 22.2 o0o2 obo eee 0. 0036 0. 0022 0.0072 | 0.0000 Detritus—not passing Imm sieve ____.___.___.|-__._-___- small small 2.29 small Coarse sand; 1.0 4600.5 tii ed ae 0.99 2.61 3.81 5. 76 Mediom sand, 0.5 to 0; 26 mm 43. 4.33 13.20 27. 68 19.37 Fine sand, 0:25 te 0.10 min 26 a ee ke 7.11 18. 80 24.81 25.84 Very fine sand, 0.10 to 0.05 mm_____-.-..-.----|.--.--.--- 13.81 | 13.82 18.09 18.08 Silt; 0; 06 to 0.008 tm cos eh 59. 30 | 28. 48 20, 45 23. 20 Clay, less than 0.005 mm ___.--_- PERRY Eos pal Meese 14.11 | 28.61 5.85 7.89 leases 99. 65 100. 52 100.19 100.14 u * Cox, Philip. Journ. Sci. § A 6 (1911) 279-830, from Table VIII, p. 309, and Table IX, pp. 314 and 815. e 510 Philippine Journal of Science 1919 TABLE VII.—Ceylon soils. * Alluvial capoors| Bee | gama | TeSa™4 | Swampy Sky | nent inter- planted land, soil. station, planted with black Pera- with rubber. seit, deniya: rubber. Moisture SS ee 3.300} 3.000; 3.600| 3.000} 5.600 Organic matter and combined water------ ---- 8.000 11. 000 4, 600 6.000 | 20.400 Oxide of iron and manganese --.-.----.-.------- 7. 400 8.000 7.200 5. 200 1. 200 Gite ie Be ooo ene nn n= - 5 8. 200 9.717 6. 786 13. 049 5. 232 Tames Siseonches 0. 060 0. 130 0. 160 0. 160 0. 050 Sane Ss ee 0. 054 0.259 0. 216 0. 490 0.115 Pubenh sso oa cnc; 0.085 0. 162 0.077 0. 401 0. 061 Phosphorie acid... =... os. ce eee 0.010 0. 076 0. 064 0. 089 0. 064 BOOS ceeded wae science ldan ns BR esa el eid 0.074 0. 188 0. 233 0.187 0. 182 Sulphtrie 6018 ooo soca Sr ooucces 0.008 0.054 0. 048 0. 068 0. 048 CMIORING 6S Soc is see atest ee 0.003 0.014 0, 016 0. 006 0. 048 Sands-and silichtes 5. i600. 52st 72. 806 67. 400 77. 000 71.400 | 67.000 100.000 | 100.000 | 100.000 | 100.000 ; 100.000 Containing nitrogeii...2020- 53.2 aS 0. 128 0. 230 0. 100 0. 162 0, 448 Ruel te ammonia 26552525 co Se eke 0. 156 0. 280 0. 122 0. 195 0. 554 Lower oxide of iron_-....-----_- Scat ae eee trace much trace trace much ROA oe ee much | neutral fair much much Citeie soluble potaslis. o.oo: 2c oe 0. 006 0,018 0. 008 0. 025 0. 009 Citric soluble phosphoric acid --_....-.-.------ nil trace nil trace nil Fine soil passing 90 mesh __.__----...---------- 11.50 53. 90 48, 00 34.00 | 59.00 Fine soil passing 60 mesh -...----.-..--..------ 9.50 43. 00 42, 00 25.00 | 36.00 Medium soil passing 30 mesh ____-_-___-_____-- 4.00 3.00 8.00 10. 00 1.00 Coarse sand and small stones -__-_.-__-__-____. 75.00 0.10 2.00 81.00 4.00 100. 00 100. 00 100. 00 100.00 | 100.00 « From Circulars and Agr. Journ. Roy. Bot. Gard. Ceylon 3 (1905) 67-72. 511 Yates: Growth of Hevea Brasiliensis XIV, 5 "2 ‘9061 20F “S “W ‘a JO enyMosy jo 10j0a1Iq jo yaodey worg y 000 °00TH} 000°00T | 000°00T | 000°00T | 000°00T | 000°00T | 000°00T | 000°00T | 000°00T | 000°00T - 3 00°9% | 00° | 00°%% |00°8T | 00°OE | 00% Ao A PORT REG earmNR wtoatal am) Big ATLe Taph ie rags hye Gods, “ree-"" S@U0}S [[BUS PUB PUES eSB) | 00°FE | 00°82 | 00°% | 00°8 OO FOP RL OEE a ar eee as MME Gia gi: os yy. oy ey eS ysour 0g Suyssed umnipayy 000 =| 00% | 00°08 ‘00°88 | 00% | 00°98 | o0'rs | 00's | os OO ere ee ee ysou 09 Fujsevd [jos oupy 00°0S | 00°9T | 00°92 | 00°96 | 00° | 00°F | 00% | 00°89 [0996 | 00°96 |--- fa £19.18 Seas Yseour 06 Bujssed [jos oupg poxteul | poxyieu | poyleul | payieul | payivul ; peyxreu peyreu | payaeu ee: Vapi Fe ee eee er eet oe We a ma Cet ete ean mare AVIpDy poos poos poos poo poos poos poo poos arey nee ree ee oe Oe ek ae UOd! JO @pIxo JeMO'T 592°0, | SFPD: | CSP'O |) (6000 SO TNO SR MRRP PRIS «Oty area racers rec erent ByUoUTUTE 07 penboy War'0.” | CORD. «|| SOO. | SEED | SFO SRO. RD. FOOD LO | | LOR fn ee ee ees ueSoaqu Suyurezu0y On mel em mee en | ee ee en mm wee | eww nn mons mmm enn ww en ewww en eons oor 610°0 =| 210°0 een eee eee et Dee ne ee 009°6L | 008°L8 | 009°88 | 0092 | 008°S6 | 000°68 | 002°6L | 000°T8 | 008 "ZL CA Midi Reomaiarape ess oto igs errant: dak Soh iad epee iaiai ig lec oyBOTTIS pues puss T90°0 TS0 “0 ¥90°0 #900 Ts0°0 910°0 7900 9L0°0 2100 GE0 0 ee ee oe eee oo ee ne eee pre o1oydsoy g F100 120 "0 0&0 “0 $900 F100 6200 970 “0 PIO'O 69T ‘0 Gi. Gas baie theta pecunte ie apes hc lk nee ae Mt aan ao no PRT aE yseyog PPL 'O O&T “0 o&t "0 00T'O OsT “O 980°0 00T "0 082 °0 188 °0 WES Oa ea ee rn Beene mC ett on ne te sen Bissuseyy OFE"O | OFT" = | ODT" | OBE"O | OPED | OBE"O | OOS | OME" | OB. | PRB fmrn tan nntna ane ea tennant nent ania nn emn nn ann nnn nanan eury TS6 °P 896% 91S °% E81 'h SOL 'T gg8 “T 069° 069°S 610°S Oo Be Tea PBRaiarisaer clade etiicd ats ale rong ot xh Naame aa BUlUIN[B JO epIxO 000% | 00870 00L "0 0FZ‘8 008 “0 00P°Z 009 "T 000°S 002 'T (4 Wes Ma tabi ohana raaprr yt 1 3cru aes eg 7 es0UBSULUI PUB OI! Jo epIxO 002 °6 000°9 009 "g 009 6 000 8 00% *F 000 °6 000°8 0F9 *9T 1 gee MEE Re ciety aia duhere ne 4 aa, 410}8M poulquios pus J043euUI d1UeSIQ 008 ‘T 009° 008 °% 000 'F OOF *T 000 “2 006 °F 000 °S 099 °¢ OD a eer ae eer een ree ear ie nn eee tae ee eanisloW “posqug “sulBo] Apuug “SABO [VANITY 897015 Anjo payosapay ay7 fo spog—y]I]A Fav 512 Philippine Journal of Science 1919 ELEVATION Hevea brasiliensis appears to make its most rapid growth at low elevations. Though it has been grown in Ceylon and else- where at elevations up to 600 meters or more, the rate of growth ~ is apparently much less rapid. It is probable that the retard- ative effect of altitude upon rubber is related to the lower temperatures prevailing at higher elevations. The extreme retarding effect of high elevations on the growth of trees in the Tropics is shown by the dwarfed condition of vegetation on mountain tops in these regions. Table IX gives the results of an experiment in Malaya to show the effect of altitude upon the growth of Hevea. While the small number of trees observed, especially at the higher altitudes, renders the results somewhat inconclusive, yet the experiment indicates a considerable reduction in the rate of growth as elevation is increased. The trees in clearing 8 were, however, said to be as healthy as those in clearing 1. TABLE IX.—Experiment on Gunong Angsi.* s | Clearing. Elevation. | realy to Bela . —— et Feet. | Meters. | {. sith eee eS Sater 300 91.44 | 365 | 529 | 69 te 600 | 192.88 | oe ee es ERED SAGE set Shy San 1,000 | 304. 80 | $21 | 665 | 48 Poe = --| 1,200] 365.76! (b) (b) | (b) $.: : _-| 1,600; 487.68 | 24 | 90 | -ar | | Paitnenentonre oe en SN ee a 1,800} 574.84 | 17 | 91 | 18 | Tice cccn $s ani edcanaie apemeigaperet on ce 2,100} 640.08 ...._.___. ch ee 6 iin dbo Ena hg ee ea naan ee 2,400; 781.52 | 2 | 90 | 2 * From Agr. Bull. Straits & F. M. S. 10 (1911) 257. > Not planted. GROWTH IN THE PHILIPPINES It appears from the preceding discussion that climatic and soil conditions in the Mindanao region of the Philippine Islands compare well with those in regions in the eastern Tropics where plantation rubber is now being successfully produced. It re- mains to compare actual measurements of growth and yield of rubber of trees growing in the Philippines with those in other rubber-producing districts. In Table X a series of measure- ments is given of trees of various ages growing in the Philip- pines, and an attempt has been made to determine from these measurements the rate of growth of Hevea up to the age of about 10 years. XIV, 5 Yates: Growth of Hevea Brasiliensis 513 TABLE X.—Growth of Hevea rubber trees in the Philippine Islands. Girth of trees in block No.— Order No. 2 | 2 | ge | sa | a | of | 106 | 10m |. 18: | 265 | SLE mone eng Rae a 100.0} 64.0} 57.0} 28.5 | 108.5/ 33.0] 57.0] 34.0! 44.0] 15.0 os ee eos 91.5; 60.0} 58.0} 40.0] 70.0]- 30.0] 70.0] 55.0! 36.5 | 11.0 Bc acscee Sie 118.0 | 40.0/ 40.0| 47.0| 65.0| 44.0| 61.0] 49.0! 42.0! 21.0 Pe a as ae 91.0| 56.5| 61.0} 82.0/ 66.5] 44.0] 585]! 51.5/| 42.5) 12.0 a ne ne eee 86.5; 64.0] 39.0] 25.0] 61.6/ 50.0] 58.0] 87.0! 66.0| 18.0 ea ae ae 76.0} 57.6 | 45.0| 80.0) 63.5] 381.5] 66.0] 58.0! 46.5 | 16.0 , Cea in ge Weer are 86.0} 51.0] 36.5] 35.0) 57.0] 28.5| 52.0] 49.0] 53.0/ 18.0 Ss ase 73.5} 48.5| 40.0] 30.0] 58.0] 87.0] 75.0] 59.5| 41.5 | 15.6 Ba a a 78.0 | 51.0] 34.0] 27.0] 65.0] 23.0! 63.0} 64.5) 58.5! 14.0 eos ei 86.0 | 54.0] 89.0] 82.0) 61.0} 81.5) 67.0! 47.0! 48.0! 110 yy een ES pe 91.0! 61.0! 46.0} 37.0] 77.0) 45.0} 58.0! 68.0! 50.0| 18.5 Mc i cS egnelcs 142.0 | 67.0] 37.0); 33.0} 55.5) 49.0] 785} 48.0/| 50.0| 110 SAE eee Rr eaee aor 84.0} 56.0} 49.0) 35.5| 76.0] 38.0] 67.0} 59.01 45.0/ 12.0 VS cite cae ae 69.0} 67.0; 38.0!) 37.0] 52.0} 52.5] 64.0] 67.0! 50.0| 15.0 Os oi a ea 73.0| 55.0| 58.5] 85.5/ 65.0] 30.0] 57.0; 63.0! 41.0] 18.5 Wino eon en 81.0] 51.0) 69.0] 36.5] 53.0) 38.0] 66.0} 64.5! 48.5/ 13.0 bf SR eR Ee 76.0| 64.0! 65.0) 27.5] 90.0] 22.5] 45.0! 68.0) 48.0) 18.0 Wen acc wnee se 87.0} 55.0} 54.0) 25.5/| 68.5] 25.0] 59.0] 47.0] 42.0/ 15.0 UD on ca emiena 126.0] 48.0| 55.0) 87.0} 57.5] 33.0] 45.0| 54.0) 48.0/ 15.0 20 74.0] 52.0| 53.0| 27.0) 54.0] 55.0; 69.0] 52.5] 52.0] 13.5 Planted in June, 1910; 2.44 by 2.44 meters (8 by 8 feet) on a flat; age, 88 months. © Planted in July, 1912; 5.49 noi eacas acaiiy (18 by 18 feet) on gently rolling ground; age, 63 months. * Planted in May, 1913; ia by 6. ts (18 by 18 feet); land slightly sloping; age, 53 months. °* Planted in August, 1910; originally interplanted with hemp at somewhat irregular distances, average about 4.57 by 5.49 meters (15 by 18 feet) ; land slightly sloping ; age, 86 months. ' Planted in June, 1914; 5.49 by 5.49 meters (18 by 18 feet), on a steep hillside; age, 41_ months. ® Planted in June, 1911, on a moderate slope, 4.88 by 4.88 meters (16 by 16 feet) ; age, 76 months. * Pianied bo Suse 811; oo a mentle slope, 500 by 6.0 wabets ith bo 10 font) age 0 ue ‘Planted in July, 1918, on praetically level land, 5.49 by 5.49 meters (18 by 18 feet); age, 51 months. ' Planted in June,,1916, on slightly rolling land, 6.49 by 549 meters (18 by 18 feet) ; age 17 months, 514 Philippine Journal of Science 1919 TABLE X.—Growth of Hevea rubber trees in the Philippine Islands—Ctd. Girth of trees in block No.— Order No. Qa | Qb Be 5a Je 9f | 10z | 10b | 181 | 263 | Bois SS 77.0} 56.5 | 46.0} 39.0 /_.....- 84.0) 69.0} 51.5] 44.0 |_--.... 36 cee 89.0} 68.0] 59.0} 38.0 |______. 24.0} 58.0] 61.0] 47.0 |___-_. WS A A 77.5 | 56.0] 52.0} 41.0 ]..._.. 89.0 | 53.0) 62.0} 58.0 |_..--.. S25 7.0} 68.0| 41.5] 88.0 }.......| 34.0] 76.0] 58.5] 45.6 |______. Fee el EO G50 SO) 00-85. ek 90..6.|:.49.0.1- 41,0 |. s: 40 a OO | OB 68:0}. 48 0) So ssc 56.0} 61.5] 39.5 |_._-._. EWES ae Ae 89,0: 69,04 38:0. 460.0 2i5 ok 50.0} 45.5 | 58.0 j_.._._. 42 S601 O80) GEO 0.6 56.0 | 68.0} 41.5 |_---.- ee ee ames te ee ee 98.6. 65:0} 60,57 87.0 jo. 023 -3.. SBA 80.0) BS ee en nena ae Ree REP 10K OS BOS 20 Cea as 63.0 | 6L.01. 49.0 1. cz. Oe ee 840} 66.01 48.01 98.0)..2-sccf. 2c 63.5 | 55.0| 44.0 |____. O35 S56 80) B64 GOL se 65.0.1 -- 62-0.) 47.0.1. aes. 47.=- = 1016} 68.0} 60.01 $8.04.5..cc0 22550 68.0] 65.5) 87.0 |... ABs So See ae SEO O04 46.0) SiG Fh KS. 59.5 | 66.0} 35.0 |....... by saree septa ea 89:0 60.04 34.5} $46.04 69.0-}. 66.5.1. 47.0: |... Wee eS 98.6%: 60.0) PROT BO Lace os 60.0 | 68.0} 48.0 |...-... Shs 92. 04- 6050+: 99:0}. 88.04. hes 61.0 | 60.5 | 45.5 |-.....- Bivins ae 16,04: $8.64):-48:-0 |: SRB Rakes 61.61-62.0 jc SOsiis Sk ee 83.5] 74.5] 60.0] 40.5 |_ 68.6.4. 51,8 |252 ised SS ea SEO BO a as] 68.0 86.01 ee 55 ms ewe} BOGS 40503: SE G645 SX Se ey SO = 46.0 98.0 52.5 | $2.5 Pe 56.0 eer ane eee ee SG iat BGT. 800 1 ak 8 $9.0: 58 ----| 124.0 85.0 |. Nate, oe aan Sires 49.6.1... 5) Ss ek ORO 61.0 |. 46,0 |i 60._.. 73.0 i gk Spent debe pees 46.0 61 WOO fo cic (U8 Seren Been Neher tene Mevoee on Serene M4, 0. esaoe Esa ee he Sey Le 62.01 63 77.0 ee Sats SESS eee 65 5 64____ Ce 2 Se ered rs stra cae a acct 60.0 65... oS eerie ese | meee je ad 48.0 66. ~~ by Sa si ised ae Ol... 0s ee SE 01 | a as 105.6 |... ¢ Shc a Ob * Planted in June, 1907; land flat; planting rather irregular but probably originally planted : about 2.44 by 4.88 meters (8 by 16 feet) ; age, 124 months. Planted in June, 1910; 2.44 by 2.44 meters (8 by 8 feet) on a flat; age, 88 months. * Planted in July, 1912; 5.49 by 5.49 meters (18 by 18 feet) on gently rolling ground; age, 63 months. * Planted in May, 1913; 5.49 by 5.49 meters (18 by 18 feet) ; land slightly sloping; age, 53 months. e Planted in August, 1910; originally interplanted with hemp at scmewhat irregular ier ay » average about 4.57 by 5.49 meters (15 by 18 feet) ; land slightly sloping; age, 86 mon ebsaslg June, 1914; 5.49 by 5.49 meters (18 by 18 feet), on a steep hillside; age, 41 mon . pene June, 1911, on a moderate slope, 4.88 by 4.88 meters (16 by 16 feet) ; age, 76 mon : » Planted in June, 1911, on a gentle slope, 5.49 by 5.49 meters (18 by 18 feet) ; age, 76 months. gecagen —— 1918, on practically level land, 5.49 by 5.49 meters (18 by 18 feet); age, months. 3 soon ~ June, 1916, on slightly rolling land, 5.49 by 5.49 meters (18 by 18 feet); age, months. : ; XIV, 5 Yates: Growth of Hevea Brasiliensis 515 TABLE X.—Growth of Hevea rubber trees in the Philippine Islands—Ctd. j Girth of trees in block No.— Order No. sp mes saree : > 8 tee «= eee Qs | a | Se | 5a | Ze | Of | 10e | 10m | 18: | 255 ei ees i Se ee Crees Pres ferent sere Uns ements Mui hil Beers yA ae Na a a aa or ta BED lop [ eae tae eee cele ees 2g 2 ee Seared bE a Ee pe Pk ar ide ore eens Pater: Meee Rerroees Reemnrecs Winger soc CSE. ME as sat Ce EN Ee Oliaities og _ dea 50.0 |... DP ceciicaiccnacs a 2ecnet MOG Aes ed cla fe ees uaieso| social Aas eed olens. MA. 0 scicge 2 I ER ea pee BBO fo ncn eso) sen tae pa anal enamen| ann cay -fascnns a pbeioaie WOogogereccn eaccesTeLeade EB Sm imseeice Macao Bivtie iar s Maier peegaaies 7 Saeade oh ae Peale GSR os a (a FB ee Meine Ueeee ceed Meee Sapwaees boiaea] Marrone 8.05). 0: rer nt TOD. | Sao ons nasa in lard comeenesmarhn onenabiech il cooly 49.5 |... SE a ee Sade appe Paapeeei acens Are ss sien cs erate aCe 49,5 |...-... ph Se oe dl cas eS Ragas Fentee el BE a moan ate exp fi Sees ee a ere nr eres pero Nees Sane Oe I eet CED oo. Se Sins vanes 9B On warned = peu nd mew Sohunwbhllanen Sede sew owe 9060-4.£%,.. Sap tS «os 5h as Bea Corin Pe mcs BO en aeciale oh wae Sis an ee WL, Gf ott fale el eeatey. ie eel ee oe ack eto oo a ee nied Cee Cr er ee Por Pei veal Bree 48.0 |... ce PF, Ss eer em ee Ce Seo Dea Rene ils Gaara og ne ere e ne Oey OG ae oe Te A Ae «aes Be a ace ine ee Ok i Et pa Ae Me re ea eee 48.0 |... 6s. ii... ee a ea es ee 49.0} oo Re ie Ee ee a er (eee ns an, poner ime: SEB oda od es is a OE Bee Be, apc anes Be eet ag ee Dis ee eee ee eee ee ettemreser heel mt ca oes oA a cee ys de sear tei ie ee eae a a ce Ee 62.0 |---.... eeepc eT Ie ak EE ps Faia ee ee, ee ee ty eee coe etna Nese Lae Nee, a ts acaio ReaineS: Waianae Reseda Meets avn Beas 6.25. i gama ie rea bate eto ee de rb gti 54.0 |... Wea ecuee oe Pra netalee boars lene Mion| cae aaee leccadlnseacsfanuseen BUD Acc eas Re En ee ore oa eee eee 0.0 [ux Average =o. 11. 89.4; 58.4) 46.3 | 35.4/ 64.9/ 38.9| 61.7| 55.5| 47.8 | 14.2 LL, 142.0 | 74.5 | 65.0 | 650.0 | 108.0} 55.0] 80.0} 68.0| 66.0| 21.0 Minimum... __._.. 69.0} 40.0} 83.0} 23.5] 50.5| 22.5/ 45.0] 34.0| 36.5 [19.0 | * Planted in June, 1907; land flat; planting rather irregular but probably originally planted : about 2.44 by 4.88 meters (8 by 16 feet) ; age, 124 months. Ba Ea ee - » Planted in June, 1910; 2.44 by 2.44 meters (8 by 8 feet) on a flat; age, 88 months, | * Planted in July, 1912; 5.49 by 5.49 meters (18 by 18 feet) on gently rolling ground; age, 63 months. ee ese Or Es eee “Planted in May, 1913; 5.49 by 5.49 meters (18 by 18 feet): land slightly sloping; age, 53 months. * Planted in August, 1910; originally interplanted with hemp at somewhat irregular _ distances, average about 4.57 by 5.49 meters (15 by 18 feet) ; land slightly sloping ; age, 86 months. ‘ Planted in June, 1914; 5.49 by 5.49 meters (18 by 18 feet), on a steep hillside; age, 41 months. : ; ® Planted in June, 1911, on a moderate slope, 4.88 by 4.88 meters (16 by 16 feet); age, 76 _" Planted in June, 1911, on a gentle slope, 5.49 by 5.49 meters (18 by 18 feet) ; age, 76 months. ' Planted in July, 1913, on practically level land, 5.49 by 5.49 meters (18 by 18 feet); age, tr eee 2 pes SOUS) Beat } Planted in June, 1916, on slightly rolling land, 5.49 by 5.49 meters (18 by 18 feet); age, 516 Philippine Journal of Science 1919 It is recognized that the results are subject to criticism from a scientific viewpoint, but it is believed definite data in regard to the girth of Hevea brasiliensis in the Philippine Islands are of sufficient practical importance to warrant their publication. It would undoubtedly have been desirable to remeasure the same trees through several successive years; but, since this has not been practicable, the results of the present series are given in the table to show the growth attained by trees at certain known ages. An attempt has been made, by averaging a number of measurements, to neutralize individual variation among trees of the same age as such variation in Hevea is well known to be considerable. Under apparently identical conditions certain in- dividuals may attain as great a girth in five years as others do in seven or eight years. Conditions of cultivation are approximately equal for all the trees considered. A strip about 3 meters wide is kept clean- weeded along the rows, and the remainder plowed about three times each year. The older trees were planted at various dis- tances, but later plantings are all 5.49 by 5.49 meters (18 by 18 feet). In each case the planting distance is indicated. In the older blocks the retarding effect of close planting on the growth of the trees is very evident. The trees have practically all been tapped since they attained a diameter of 50.8 centimeters (20 inches) at 45.7 centimeters (18 inches) above the ground. MEASUREMENTS OF TREES ON BASILAN ISLAND In Table X the measurements of five hundred fifty-one trees of various ages are given. The age in months of each group is recorded, together with the planting distance and general char- acter of the land. Measurements were made with a tape to the nearest 5 millimeters at 1 meter above the ground. For each group the maximum, minimum, and average girths are indicated. A certain error is caused by the fact that the measurements were necessarily made over the tapping surface, but this error is prob- ably never very great and is in the direction of a reduction rather than an increase in the recorded girth of trees at the various ages. Most of the measurements of trees in other coun- tries with which the Philippine trees are compared have been measured at 1 yard from the ground; and, since the trunks of Hevea brasiliensis taper rather abruptly, this difference in height of measurement has the result of reducing the girth of the Philippine trees when compared with others. This comparative reduction ranges from a few millimeters for the younger trees up to approximately 1 centimeter for trees 10 years of age. It XIV, 5 Yates: Grpwth of Hevea Brasiliensis 517 has not been found practicable to apply any correction for this difference due to height of measurement; but when comparing the girth of trees measured at different heights, the fact should be borne in mind that such difference exists and that it increases with the size of the tree. In fig. 1 is plotted the average circumference of trees of the various ages measured, and from this curve the average circum- ference of trees from 1 to 10 years old has been calculated. The results obtained in this manner are given in Table XI. TABLE XI.—Diameters attained by average trees in Basilan. From 1 to 10 years old with increment. Aevinyean, | Greene | Tee | cm, cm, 12 ee eee | 10.3 10.3 | BIS os ol Behe 18.7 8.4 ee eee oe ee Paes 5 | 8.5 oe es ae ies 35.6 8.4 SSS ase | 44.0 8.4 Cree Se ee bo §a4 8.4 Ue oo ies ae es | - 60.9 8.5 Bet SL SEO Aa OS | 69.4 8.5 Os ee st ae po Te 8.5 $0 ar eh pe a ee | 86.5 8.6 Age in months. 10 20 30 40 50 60 70 80 $0 00 “a 120 ee of \ a re Circumference in centimeters. (\ . ae a aig Or “|.-“Fre ‘ "a Fie. 1. Comparison of the growth of Hevea brasiliensis in the Philippine Islands with that in Ceylon and Malaya. 518 Philippine Journal of Science 1919 In the same figure are also plotted the figures given in Table XV for the growth of trees at Edangoda, Ceylon, and also the second of the two series of measurements of trees 5 to 9 years old given in Table XII. From Table XI it appears that the annual increase in cir- - egumference of trees up to 10 years of age averages 8.6 centi- meters. Ridley* gives measurements of trees in Malaya which approach very closely those recorded for trees in the Philippines. Table XII is copied from Ridley’s paper, with the measurements reduced to centimeters. TABLE XII.—Growth of trees in Malaya.* Increment, six Age in years. 1904 1905 1906 1909 years. ROOT Sou cm. |Inches.| em. |Inches.| em. |Inches.| cm. |Inches.| cm. | Inches. ORB Fede -| 44.77 178 | 54.61 214 | 65.72 253 | 89.85 35g | 45.09; 172 So ee 45. 88 188 | 58.55 234 | 68.80 27% | 91.68 363 | 45.75 | 188 Ey | Pesce 115. 57 45% | 120.97 47% | 129.54 51 | 149.28 583 | 33.02 18 | Nee 278.18 | 109% | 283.84 | 1113 | 288.92) 118%] 806.07| 1204| 27.94 ul * From Agr. Bull. Straits & F. M. S. 9 (1910) 257. His measurements of trees from 5 to 9 years of age indicate a somewhat faster growth than is shown by trees in Basilan, but this may be due to the effect of close planting on the latter. In the same paper Ridley shows that in a block of closely planted trees the outside trees grew 20.3 centimeters in six years, while those inside grew only 10.3 centimeters during the same period. He also states that he considers a growth under fair conditions of 7.6 to 10.2 centimeters (3 to 4 inches) per year, for trees from 5 to 15 years of age, to be satisfactory. The average yearly increment of all trees measured in the Philippines appears to be about 8.6 centimeters (3.4 inches). An average of ninety- seven trees in block 18 indicates a growth of 47.8 centimeters (18.0 inches) in fifty-one months, which amounts to about 11.2 centimeters (4.4 inches) per year. Ridley * gives the measurements of some trees growing under jungle conditions in the Singapore Botanic Gardens. A copy of his table appears in Table XIII with the measurements reduced to centimeters. The trees were planted about ten years before the first measurements were made. He states that they are fully twice as large as trees grown under cultivation. "Agr. Bull. Straits and F. M. S. 9 (1910) 25. * Agr. Bull. Straits & F. M.S. 7 (1908) 254. XIV, 5 Yates: Growth of Hevea Brasiliensis 519 TABLE XIII.—Growth of trees in the Singapore Botanic Gardens.” [Age of trees, 14 years.] Girth in— Registered No. of tree. Remarks. 1904 | 1905 | 1906 | 1907 | 1908 ob» PASS Re ee Seapine mt re 97.8 | 102.9} 108.5) 118.6! 121.9 | Slope. Desig Si eee ca teceri too ol. te 81.3 88.3 93.3 96.5 Do, sty | PORES Esteem tin ican eager oe 60.9 62. 2 66.0 68. 6 71.1 Do. Booming egy et ecw a 136.5} 147.0} 147.8} 151.1} 152.4 | Top of hill. gs eee Pen geaget eR oa 158.7} 159.7 | 166.1] 174.0] 182.9 | Top of hill, 100 feet tall. a 5 a as aes 86.4 88.9 90.5 93.7 96.5 | Slope. boy < Sens AE ae ae aeeeabet ast eS 87,0 92.7 96.5 | 101.6! 104.1 Do. POO sccneicu ret bad oc ae 85.1 98.7} 105.1} 106.1} 108.0 Do. 1 Res Sepa ae ee 158.4} 1738.0) 181.3] 191.8{ 200.74 Top of hill. 2 From Agr. Bull. Straits & F. M. S. 7 (1908) 254. The record extends over a period of four years and indicates an average growth during this period of 5.38 centimeters per year. The most rapidly growing tree, No. 1281, shows a growth of 10.5 centimeters per year, almost double that of the average. In Table XIV are given measurements of trees up to 3 years old. The best growth indicated for that period is 23.9 centi- meters, which is slightly less than that attained by trees of the same age at Basilan. TABLE XIV.—Open-planting experiments. Planted 7.5 by 3.75 meters (25 by 12.5 feet).* Stumps 2 months from Stumps 6 months from F nursery. nursery. Seeds sown at stake, 1907. Seeds sown in nursery, Octo- || Seeds sown in nursery, Octo- ber, 1907. ber, 1907. > ° om. in. em, in. cm. in. October, 1907 ___ saat plant-|| December, 1907_- on ef — April, 1908_...- 2.41 #8 October, 1908...’ 38.1 3; || December, 1908-.| 6.5 2%5 |; April, 1909.__.--| 7.9 34 October, 1809. 16.1 6% || December, 1909__| 14.2 58 || April, 1910_-.--- 23.0} 9d October, 1910_._! 23.6 9,3; || December, 1910..| 23.9 92 * From Report of Director of Agriculture F. M. S. in Agr. Bull. Straits & F. M. S. 10 (1911) 255-256. Other records from Malaya indicate in some cases more rapid growth than those cited. In measurements of certain trees on the Bukit Raja Estate in Selangor ® in which five trees (Nos. 1, 3, 6, 7, 8), all approximately 4 years of age, attained an * Agr. Bull. Straits & F. M. S. 1 (1902) 333. 520 Philippine Journal of Science 1919 average circumference of 47.8 centimeters at 1.36 meters (4.5 feet) above the ground. However, even this rate of growth is exceeded by certain selected trees in Basilan. The best group of three consecutive trees in block 18 (Nos. 79, 80, 81) shows an average girth of 57.7 centimeters (22.7 inches) at the age of 51 months, which is equivalent to about 54.24 centimeters (21.3 inches) at the age of 4 years. Willis *° presents the measurements of some trees in Ceylon. The original trees, planted at Henaratgoda in 1876 when prob- ably about 1 year old, attained a girth of 35.6 centimeters (14 inches) two years later, and in 1882 the largest tree was 63.5 centimeters (25 inches) in circumference 1 yard from the ground. The girth of this tree was measured annually there- after and in 1886, when 10 years of age, its girth was 124.5 centimeters (49 inches). Even this growth appears to be in- ferior to that attained by the best tree measured at Basilan, No. 12 in block 2, which attained a circumference of 142 centi- meters (55.9 inches) at an age of 124 months. In the same paper Willis gives the average measurements of trees at Edangoda and Yattipowa, Ceylon. Willis’ figures are given in Table XV. TABLE XV.—Growth of trees in Ceylon. Locality. Age. Trees. Circumference. Years. cm. tn. | Edangoda _._...___. ae 4 100 32.9 12.96 Do ...___ 3 50- eae D6 oss esse oi See Se ee ee 2 20 12.6 4.96 Sathipe was 26 os ee eC via 3 108 23.8 9.37 ppc ews MIN SR nO a a ae el eA 3 108 23.2 9.13 The trees were measured 3 feet above the ground and so are closely comparable with measurements in the Philippines made at a height of 1 meter. These measurements indicate a de- cidedly slower growth than is shown by trees of the same age in Basilan. At an age of 4 years the circumference attained by trees at Edangoda was 2.7 centimeters (1.1 inches) less than that of trees of the same age in the Philippines. Other measurements of Para rubber trees recorded from Ceylon in- * Cireular and Bull. Roy. Bot. Gard. Ceylon 5 (1910) 17 and 18. * XIV, 5 Yates: Growth of Hevea Brasiliensis 521 dicate that the rate of growth in that country, while rather variable, is in general considerably less rapid than in Malaya. Johnson “ gives girth measurements of trees in the Tarkwa Botanic Station, Gold Coast. Data from his table are presented in Table XVI. The rate of growth does not appear to differ greatly from that of trees of the same ages in the Philippines. TABLE XVI.—Growth of Para rubber at the Tarkwa Botanic Station, Gold Coast. Girth at 3 feet from ground. | 2 Date of planting. Distance planted. ree December, 1905.|December, 1906. 1904, Feet. Meters. cm. in. em. in. gt prio ace Rai atin IS by 15| 4.6by 4.5] 17.78 7} 30.48/ 12 5 Re bP er eAaRENY ANE 12by 12] 3.6by 3.6} 15.24 6} 25.40) 10 | eS oo paces eee Acconte on P 1b by 15| 45 by 4.5] 16.24 6} 25.40; 10 | 1S fos dp sce, Sn ete 20 by 20] 6.0by 6.0| 15.24 6| 27.94; 1 | sie ree pee ee AA 80 by 30] 9.0by 9.0} 10.16 4/2286) 9 | Vii: 6S 40 by 40 | 12.0 by 12.0} 10.16 4] 2s) 9 | Wit tos OT aap is Coady ye 12by 12] 8.6by 3.6} 10.16 4; 2286, 9 VIII | August and September -_---- Why 2 | S6by 3.6) 10.16 4) 25:40) 10 | While much additional information in regard to the growth of Hevea is available, it is believed that sufficient material has been presented to indicate that the rate of growth of Para rub- ber in the Philippines compares favorably with that recorded elsewhere. YIELDS OF HEVEA BRASILIENSIS The yield obtained from Para rubber trees now being tapped in the Philippines is very satisfactory and seems to compare well with that obtained in other countries from trees of the same ages. The rubber obtained in 1916 from 20,510 trees in tapping by the end of the year was 14,991 kilograms (32,982 pounds), or an average of 0.73 kilograms (1.6 pounds) per tree. Almost 6,000 of these trees came into tapping during 1916 and 11,000 in 1915. The oldest trees were between 8 and 9 years of age at that time. Unfortunately no separate record has been kept of the yield from trees of different ages, and this renders difficult a comparison with the yields obtained in other countries. “Johnson, The Cultivation and Preparation of Para Rubber, 2d ed. (1909) 25. 522 Philippine Journal of Science 1919 Records of yields from an estate in the Federated Malay States ** indicate that trees 6 years old produced 215.5 kilo- grams per hectare (192 pounds per acre) ; trees 8 years old, a yield of 288.5 kilograms per hectare (257 pounds per acre) ; and trees 9 years old, 401.9 kilograms per hectare (358 pounds per acre). This yield may be compared with that at Basilan for trees (most of them only 4 to 6 years old) which, assuming there were 331 trees per hectare (134 per acre), amounts to about 241 kilograms per hectare (215 pounds per acre). Carruthers ** reports that in 1907 the average yield of dry rubber over the entire Malay Peninsula amounted to 0.79 kilo- gram (1 pound, 12 ounces) per tapped tree, and in Selangor 0.65 kilogram (1 pound, 7 ounces) per tapped tree. The aver- age yield obtained in Malaya in 1907 compares perhaps most closely with the yields obtained in Basilan of any records avail- able, since in both cases the trees were young and a very high percentage of the total number was tapped for the first time. Johnson * gives records of the yield obtained on a number of Ceylon and Malayan estates. The average of yields from fourteen estates in 1905 was 0.61 kilogram (1.35 pounds) per tree. The total number of trees considered was 80,272. The greatest yield recorded was 1.47 kilograms (3.25 pounds) per tree but that was for only 100 trees. The greatest yield for a large number of trees was 1.32 kilograms (3 pounds) per tree for 10,000 trees. The average of the yields from twenty-two estates in 1906 was 0.67 kilogram (1.47 pounds). The greatest yield recorded was 3.23 kilograms (7.11 pounds) per tree for 807 trees. The average of yields from eighteen estates in 1907 was 0.83 kilogram (1.83 pounds) per tree. The highest yield was 1.49 kilograms (3.28 pounds) per tree for 68,236 _ trees. A discussion of the records of yields obtained from Para rubber could be continued indefinitely, but it is believed that sufficient data have been presented to show that the yield ob-— tained from Hevea on Basilan compare favorably with that ob- tained from young trees in Malaya and Ceylon. “India Rubber Journ. No. 22, 53 (1917) 14, “ Agr. Bull. Straits & F. M. S. 7 (1908) 582. “The Cultivation and Preparation of Para Rubber. London (1909) 149-152. XIV, 5 Yates: Growth of Hevea Brasiliensis 523 CONCLUSIONS 1. A comparison of rainfall, temperature, and soil conditions obtaining in the southern part of the Philippine Archipelago with those in countries where Para rubber is successfully cul- tivated indicates the suitability of the former region for Hevea brasiliensis. 2. Measurements of Hevea brasiliensis at present growing in the Philippines show a rate of growth comparing favorably with records of the growth of Hevea in Ceylon and Malaya. 3. Yields obtained from Para rubber trees now being tapped are satisfactory and compare well with those obtained in Ceylon and Malaya. 1655596 ILLUSTRATION TEXT FIGURE Fic. 1. Chart, showing a comparison of the growth of Hevea brasiliensis in the Philippine Islands and in Ceylon and Malaya. 525 A BIOLOGICAL AND SYSTEMATIC STUDY OF PHILIPPINE PLANT GALLS? By LEopoLtpo B. UICHANCO (From the College of Agriculture, University of the Philippines, Los Banos) FIFTEEN PLATES INTRODUCTION Galls are abnormal growths on the stems, leaves, roots, or other parts of plants, caused by the action of insects, arachnids, or fungi, or by unknown agencies. Just how these peculiar struc- tural developments are brought about is still open to discussion and speculation, experimental proofs being, up to the present, too deficient to warrant our drawing any definite conclusion. These malformations have been ascribed to various causes, the more commonly accepted, in the absence of more reasonable, ex- planations being the following:* 1, a severe mechanical injury to certain parts of the plant; 2, a continuous mechanical irrita- tion; 3, secretion of chemical stimulus by the causal animal or fungus. One, or a combination of two or all, of these causes may give rise to the production of galls. In the formation of zodce- cidia, the third factor—that is, the action of the virus secreted at the time of oviposition or during the development of the para- site—is probably the most important, the first two being of minor or absolutely no use whatever. This fact was shown by Molliard in connection with his experiments on Aulax papaveris Perris, a cynipid gall maker on the pistil of Papaver rhoeas Pall. From time to time he removed a small quantity of the virus secreted by the developing Awlax larva, and injected the fluid into the growing pistils of Papaver. This artificial treat- ment resulted in the formation of galls which resembled in all respects those formed in the presence of the larva itself. He was thus enabled to draw the conclusion that the virus alone, without the influence of mechanical irritation from the presence of the animal within, is sufficient to produce the characteristic Papaver gall. The importance of chemical stimuli, as related to gall formation, was recognized as early as 1686 by Malpighi, ‘Thesis presented in partial fulfillment of the requirements for the degree of Master of Science, University of the Philippines, 1918. * Cook, M. T., Insect Galls of Indiana. Dept. Geol. and Nat. Resources, Indiana, 29th Ann. Rep. (1904) 801. *Compt. Rend. Acad. Sci. 165 (1917) 160. 527 528 Philippine Journal of Science 1919 a physician to Innocent XII, and professor of medicine at Bo- logna and, later, at Messina. In his De Gallis, which is the ear- liest systematic treatise on galls, he maintained that, at least in the case of Cynips, the galls formed on the plant were caused by a certain acid secreted by the insects. A more recent paper by A. Cosens has the following to say in connection with these chemical secretions: ° The larva secretes an enzyme capable of changing starch to sugar [and] which acts on the starchy constituents of the nutritive zone [of the gall] and accelerates the rate of their change to sugar. The material thus prepared supplies nourishment to both the larva and the gall. Galls can be produced only “when the tissue of the plant is interfered with during, or prior to, the actual development of the tissue.”° After the plant tissue has become fully matured, no amount of stimulus, whether physical or chemical, will lead to the formation of galls. In the present work only galls caused by the action of animals, known as zodcecidia, are taken into consideration. A zodceci- dium may be caused by the action of a member of either of two zodlogical classes—Insecta and Arachnida. Galls caused by the former vary in structure from a simple convolution in the leaf lamina or a swelling in the stem to a more or less complex for- mation on different parts of the plant. Insects that have the power of producing plant galls are confined to the sacagiaes or- ders and families: Order. Families. 1. Rhynchota, or Hemiptera. Psyllide, Aphidide, and Coccide. 2. Diptera. Itonidide, or Cecidomyiide, and Trypetide. 3. Hymenoptera. Cynipide and Tenthredinide. 4. Lepidoptera. Gelechidex. 5. Coleoptera. Buprestide.” 6. Thysanoptera. Arachnid gall-makers are all mites, which are members of the family Eriophyide (formerly called Phytoptide). Galls caused by this family are called erinea* and are generally simple con- volutions in the leaf famina, with hairy outgrowth on the concave surfaces. When the gall is young, the gall-making eriophyids are found among these hair tufts. “Cook, M. T., op. cit. 802. * Trans. Canad, Inst. 9 (1912) 297-387; [Ent. News 24 (1913) 187-189.] ‘Butler, F. H., Galls. Encyel. Brit. 11 (1910) 425. "Cook, M. T., op. cit. 802. . Banks, Nathan, Acarina or Mites, Rep. U. S. Dept. Agr. 108 (1915) 135. XIV, 5 Uichanco: Philippine Plant Galls 529 ~ OBJECT AND METHODS Very little is known about Oriental galls, the only noteworthy treatise on the subject being the series of articles on Javanese, Sumatran, and Celebes galls by W. and J. Docters van Leeuwen- Reijnvaan. These publications are almost purely botanical in their nature, the causal animal being merely mentioned in pass- ing in practically all cases. On account of the very close re- semblance existing between the fauna and the flora of the Philippines and those of the Dutch East Indies, I found in the works of the authors just mentioned much valuable assistance in the preparation of the present paper. In the Philippines the subject of plant galls is one which covers an absolutely virgin field, practically no attention having been paid to these curious abnormalities in this country. Al- though including many new and interesting species, the results of the present investigation do not pretend to be anything more than a mere breaking of ground for a future more thorough and comprehensive series of investigations along this line. This branch of research bids fair to be highly productive of results in the way of elucidating obscure facts relative to these interesting formations. In gathering the materials for the study of galls, excursions were made to points in the vicinity of Los Bafios, Laguna Prov- ince, Luzon, and especially to the thickets adjoining the college farm, the forests of Mount Maquiling, Balong Bulo Hill (near the college farm), the adjoining lowlands and valleys, and along the banks of Molauin Creek. The forests around Los Bafios Falls and the thickets at the outskirts and in the barrios of Bay, Los Bafios, Calamba, Cabuyao, Santa Rosa, and Bifian were vis- ited several times and search made for galls. The materials were brought fresh into the laboratory and, as soon as possible, before the galls wrinkled up or otherwise became discolored or distorted, photographs were taken (to show the natural appear- ance of the galls on the parts of the plant attacked), preliminary descriptions made, and specimens saved in a medium of which the following is the formula: Parts. Water 48 95 per cent alcohol 48 40 per cent formaldehyde (formalin) 4 Large-sized homeopathic vials or wide-mouthed dispensing bottles of convenient dimensions and with tight-fitting corks served as preserving vessels. Each bottle or vial had the acces- sion number written on the cork and on the labels that accompa- 530 Philippine Journal of Science 1919 nied the specimens in the preserving fluid, the galls and insects being given the same number in the accession record. Longitudinal and cross sections were made of the mature galls, and these were drawn to show the details in structure and the mechanism peculiar to each case. No hard and fast rule can be laid down as to the methods to be followed in breeding the insects from their galls. Condi- tions were different in each case, and laboratory methods had to be modified accordingly. In general, battery jars with fine muslin held on by elastic bands at the top proved to be the best form of breeding receptacle, a decided advantage in favor of these being that the green parts of the plants were kept ade- quately exposed to the light and that inspection and cleaning were easy. During dry months the materials had to be mois- tened at least once a day by spraying them with water from a small hand atomizer. Care was taken not to wet the plants too much, for then decay would set in, or the adult insects on emer- gence would be entrapped by the film of water on the plant sur- face or the inner wall of the jar and spoiled. The jars were inspected every morning and the adults caught by means of a short test tube. The removal of the adults as soon as they emerged was a necessary measure, because when allowed to stay in the vessel for a longer time many of them would be lost or badly mutilated among the moist plant materials. Throughout the progress of the work precautions were taken against certain small spiders which, when accidentally intro- duced into the jar with the plant materials, would prey upon the gall insects as fast as the latter emerged. As a measure against this pest, the pieces of plants were thoroughly shaken, one by one, before being placed into the jars, and a constant watch was kept for the appearance of any of the spiders that might have hatched from eggs accidentally introduced. Another piece of apparatus used consisted of a light-tight cardboard box with the mouth ends of test tubes or vials inserted into one side. This device was supposed to work on the prin- ciple that most insects are attracted to the light ; and the interior of the box being dark, except at the insertion of the glass tubes, the insects on emerging were supposed to enter the latter. The apparatus had been used with good results and had been indorsed by the California Board of Horticulture and by the cotton-boll weevil investigators.? The results with this device in * Banks, Nathan, Directions for Collecting and Preserving Insects, Bull. U. S. Nat. Mus. 67 (1909) 115, with figure. XIV, 5 Uichanco: Philippine Plant Galls 531 the present experiments were most disappointing; in almost all trials the adult insects would not readily enter the test tubes, barely 5 per cent of those that emerged in the box having been captured in this manner. With certain plants artificial breeding in the laboratory was found to be impracticable. The leaves of many species of Ficus, for example, dry up in less than an hour after being removed from the tree and placed under ordinary laboratory conditions; when confined in a vessel with a supersaturated atmosphere, decay readily sets in, or the material soon becomes moldy. The same is true with Astronia and certain other plants that grow at high altitudes. When the plant was conveniently near the laboratory, the galls were encased in muslin on the tree, so as to prevent the escape of the adults when they emerged; and daily inspection was made of these. However, in the case of the psyllid galls on the leaves of Ficus ulmifolia Lam., a badly infested plant of which happened to grow in close proximity to the entomological laboratory, the specimen was left without any cover. On examining the plant, it was found on January 25, 1917, after about two months’ daily visits, that some of the galls were newly opened. On that day the plant was visited at frequent intervals, and it was discovered that the newly emerged adults, on growing stronger, had crawled over to the younger and softer tips of the branchlets, where they were easily caught in test tubes. For several days in succession captures were made in this manner. The thysanopterous gall makers were easily secured from their galls. Provided the galls were not too old, the insects could always be found within in all stages of growth. They were scared out of the galls by gently tapping the latter, and could then be brushed off into a vial of 70 per cent alcohol. The eriophyids were not so easy to secure on account of their habit of clinging fast to the hair tufts of the galls and also because of their exceedingly minute size. Several methods were tried, but the simplest process was to brush the animals onto the surface of a mirror by means of a small camel’s-hair brush, and then remove and mount them. In the attempts to breed the gall insects from their galls, some other specimens—that is, parasites, inquilines, ete.—were often found. At times two or more species of a gall-making family were bred out, and in this case it was not easy to decide which was the gall maker and which the inquiline. There oc- curred a number of other instances where it was exceedingly 532 Philippine Journal of Science 1919 difficult to arrive at a definite conclusion as to the real gall maker. The adult gall insects were kept alive for from six to twelve hours in a cotton-plugged vial in order to give them time to dry and assume their normal color before they were killed. With the exception of the thrips and the eriophyids, two series of preserved specimens were prepared, one on small pins and the other on microscopical slides in Canada balsam or turpen- tine solution of colophony. If the specimens were scarce, only the former series was made of the adults. The thrips in all instars were always mounted on slides. When plenty of speci- mens of fresh insect galls could be secured, a number of these were carefully dissected, and the insects in their earlier stages removed and mounted on slides. In the case of transparent specimens, such as the earlier instars of psyllid nymphs and the larve of Itonidide, which would be almost invisible when mounted on slides, resort was had to a previous staining in 0.5 per cent aqueous solution of magenta red for about twenty-four hours. Drawings of anatomical parts of the gall insects were made with the aid of a camera lucida. Whenever practicable cor- responding parts of different species of the same family were drawn to the same scale and in the same position, in order to facilitate comparison. Most of the gall sections were too large to be drawn conveniently by the aid of the microscope and camera lucida, so that practically all the atch acts of these ma- terials were made freehand. RESULTS Fifty-seven species of galls have been worked with. These are distributed among twenty-six plant families, as follows: Apocynacez, 1. Lecythidacex, 1. Aracem, 1. Leguminosex, 2. Araliaces, 2. ; Loganiacex, 1. Boraginacez, 1. Melastomatacesx, 2. Celastracex, 1. Menispermacee, 3. Combretacez, 2. Moracez, 7. Compositz, 1. : Piperacez, 1. Dilleniacezs, 1. Rubiacee, 1. Dipterocarpacez, 4. Sapindacee, 1. Euphorbiacex, 10. Tiliacexw, 2. Guttiferm, 2. Urticacem, 1. Hernandiacez, 1. Verbenacem, 3. Lauraceex, 2. Vitacem, 3. XIV, 5 Uichanco: Philippine Plant Galls 533 The present work has not been comprehensive enough to warrant our drawing very definite conclusions; but an examina- tion of the above list tends to show that the Euphorbiacez have the greatest number of gall-making species, with the Moracez ranking second. Classified as to causes, the following numbers of galls were obtained: Species of insects. Itonidid galls 19 Psyllid galls 7 Thysanopterous galls 7 Eriophyid galls 7 Gelechid galls 1 Miscellaneous galls * 16 2 “Miscellaneous galls” includes all the galls of which the causative agent is not definitely known. Galls of the insects mentioned in the introductory paragraphs of this paper, other than those of the families enumerated above, have not been met with in the present investigation. Species of Cynipide and Tenthredinide ** have been reported from the Philippines, and the chances are that further work will lead to the discovery of the galls of the former, at least. The latter, Selandria (Paraselandria) imitatrix Ashm., according to Prof. C. S. Banks, the collector of the type, is an exophagous species and not a gall maker. The Aphidide and the Coccidz are well represented in the Philippine fauna; but none of their galls have been found in connection with my work. It is not unlikely that their galls can be found here; for in Australia, a country that has many insects closely allied to Philippine species, among the largest and most remarkable galls are produced by some members of these families.'* The families Trypetidee (Diptera) and Buprestide (Coleop- tera) include many important pests of Philippine economic plants. Their work, however, has not been known to result here in the formation of galls. fe In the various excursions made, it was noted that altitude influences the number of species of plant galls to a considerable ” See Kieffer, Nouveaux cynipides des Philippines, Philip. Journ. Sci. § D 9 (1914) 188; Neuer Beitrag zur Kenntniss der philippinischen Cyni- piden, Philip. Journ. Sci. § D 11 (1916) 279. ™ See Ashmead, Proc. U. S. Nat. Mus, 28 (1905) 971. ™ Frogatt, Australian Insects (1907) 369, 370, and 380-383. 5384 Philippine Journal of Science 1919 extent—the galls occurring most abundantly within a belt ex- tending from sea level to an altitude of about 600 meters. Be- yond that limit the galls are either very scarce or entirely absent. It was further noted that from sea level to an altitude of about 600 meters the greatest number of species occurred nearer sea level, their abundance diminishing with each 100- meter zone upward. This fact may be ascribed to at least two causes; namely, the prevalence or scarcity of the plant hosts in a given altitudinal zone; and the influence of temperature, moisture, pressure, absence or presence of natural enemies, and possibly other factors on the distribution of gall-making insects in different altitudinal zones. The following is the distribu- tion of gall species, found by me, with respect to altitudes: Species. Sea level to 100 meters 36 100 meters to 200 meters 31 200 meters to 300 meters 25 300 meters to 400 meters 23 400 meters to 500 meters 15 500 meters to 600 meters 9 600 meters to 900 meters 1 Above 900 meters 0 In the following pages the galls are divided into groups, ac- cording to their causative agents, which are indicated by the center heads; and each species is treated separately, with the specific name of the plant host as a side heading. Accounts of the causal insects, including descriptions of new species, will be given in later publications. All of the specimen numbers cited in this paper refer to the collection of the department of entomology, College of Agri- culture, Los Bafios, Laguna, Philippine Islands. GALLS CAUSED BY ITONIDIDA (CECIDOMYIIDA) Acalypha stipulacea Klotz. Euphorbiacez. Leaf galls caused by Schizomyia acalyphe Felt. Monothalamous; subcylindrical; red; basally yellowish or ‘concolorous with leaf; covered all over with long, stiff, bristle- like hairs. Walls thin, fleshy; interior smooth. Opening api- cal; covered with a circular flap with edges confluent with the rest of the gall’s surface, not visible until a few minutes pre- vious to emergence of midges. Pupal exuvie often found pro- jecting half way out of opening. Length of galls, 3.5 to 5.5 millimeters; diameter at base, 1.5 to 2. XIV, 5 Uichanco: Philippine Plant Galls 535 On the nether surface of leaf, along principal veins or at points where two small nervules meet. LuZzoN, Laguna, College of Agriculture, near lower nursery, Los Bafios; at an altitude of about 45 meters. March 15, 1917. Type gall No. 18313, College of Agriculture collection. Numerous specimens of these galls were found on several young leaves of a small Acalypha plant. 20. . Mallotus philippensis (Lam.) Muell.-Arg. Leaf gall caused by Megatrioza pallida sp. nov. (MS). Longitudinal median section. xe. PLATE VI . Ficus ulmifolia Lam. Leaf gail caused by Pauropsylla deflexa sp. nov. (MS). Longitudinal median section. X 5. . Ficus variegata Blume. Leaf gall caused by Pauropsylla montana sp. nov. (MS). Open, after emergence of adult. Distal aspect. x 6. . Schefflera odorata (Blanco) Merr. and Rolfe. Leaf gall caused by Gynaikothrips chavice var. heptapleuri Karny. Longitudinal median section. X 5. . The gall shown in fig. 2. Spine from exterior wall. x 130. . The gall shown in fig. 2. Longitudinal median section. X 7.5. . Dillenia reifferscheidia F.-Vill. Leaf gall. Section across leaf. x 7.5. PLATE VII . Old galls of the species illustrated in fig. 3, showing openings through which the adults escaped. View from nether surface of leaf. x 1.5. . Grewia stylocarpa Warb. Leaf galls caused by Asphondylia gre- wiz Felt. View from upper surface of the leaf. x 1. . Barringtonia luzonensis Rolfe. Young or nearly mature, unopened leaf galls, caused by Kronodiplosis uichancoi Felt. View from nether surface of leaf. x 1.5. . Parashorea plicata Brandis. Leaf galls, probably caused by 7'rv- contarinia luzonensis Felt. View from nether surface of leaf. * 1.5. PLATE VIII y . Diplodiscus paniculatus Turez. Terminal stem gall caused by Schizomyia diplodisci Felt. Nearly mature specimen, without crevice, X 1.5. . The gall shown in fig. 1. Old specimen, showing crevice through which adults escaped. 1.5. . Acalypha stipulacea Klotz. Leaf galls caused by Schizomyia acaly- phe Felt. View from nether surface of leaf. X 1.5. Cissus trifolia (L.) K. Sch. Stem gall caused by Asphondylia vitea Felt. Type 1. x 1.5. Pexderia tomentosa Blume. Leaf gall caused by Itonida pxderix Felt. Lateral view. x 1.5. The gall shown in fig. 4. Type 2. x 1.5. Cissus adnata Wall. var. Leaf galls caused by Hyperdiplosis banksi Felt. View from nether surface of leaf. x 1. XIV, 5 FIG. Fic. Fig. 3. 4, Uichanco: Philippine Plant Galls 553 PLATE IX Leea manillensis Walp. Leaf galls caused by Lasioptera manilensis Felt. Superior lobes on upper surface of leaf. x 1.5. . The gall shown in fig. 1. Inferior lobes on nether surface of leaf. x 1.5. . The gall shown in Plate X, fig. 1. View from nether surface of leaf. - «°1.5. . Antidesma leptocladum Tul. Leaf galls caused by Ctenodactylomyia antidesmez Felt. View from upper surface of leaf. x 1.5. . Symphorema luzonicum F.-Vill. Leaf galls caused by Luzonomyia symphoremz Felt. Young or nearly mature, unopened specimens. View from nether surface of leaf. x 1.5. , . The gall shown in fig. 5. View from upper surface of leaf, show- ing depressions formed by galls at opposite surface. X 1.5. . The gall shown in fig. 5. Old specimens, showing circular openings, distad, through which the adults escaped. View from nether surface of leaf. x 1.5. PLATE X . Spatholobus gyrocarpus (Wall.) Benth. Leaf galls caused by Heliodiplosis spatholobi Felt. View from upper surface of leaf. re . Callicarpa erioclona Schauer. Leaf gall caused by Asphondylia callicarpe Felt. Superior aspect. x 1.5. . Vernonia lancifolia Merr. Leaf galls caused by Diceromyia verno- nie Felt. 1.5. . The gall shown in fig. 3. Longitudinal median section, showing larval chambers. X 1. PLATE XI Mallotus moluccana (L.) Muell.-Arg. Leaf galls. Type 1, with acute apices. View from nether surface of leaf. x 1. . The gall shown in fig. 1. Type 2, with obtuse apices. x 1.5. . The gall shown in fig. 1. View from upper surface of leaf, showing long, slender projections at bottom of galls. x 1.5. . Siphonodon celastrineus Griff. Leaf galls caused by Kamptodiplosis reducta Felt. View from upper surface of leaf. x 1.5. PLATE XII Ficus ulmifolia Lam. Leaf galls, No. 1, caused by Pauropsylla de- flexa sp. nov. (MS). Old specimens, showing openings through which the adults escaped. View from nether surface of leaf. x 1.5. : . Alstonia scholaris R. Br. Leaf galls caused by Pauropsylla tuber- culata Crawford. View from upper. surface of leaf, showing bottoms of galls. xX 1.5. - The galls shown in fig. 2. View from nether surface of leaf. « 2S. : Ficus ulmifolia Lam. Leaf galls No. 2. Young or nearly mature galls. View from upper surface of leaf. X 1.5. 5d4 FIG. FIG. Fic. p Philippine Journal of Science . The galls shown in fig. 1. Superior lobes on upper surface of leaf. x 16. . The galls shown in fig. 1. Inferior lobes on nether surface of leaf. x 1.5. PLATE XIII . Ficus variegata Blume. Leaf galls caused by Pauropsylla montana sp. nov. (MS). Closed, before emergence of adults; open, after emergence. View from upper surface of leaf. -x< 1. . Mallotus philippensis (Lam.) Muell.-Arg. Leaf galls caused by Megatrioza pallida sp. nov. (MS). View from upper surface of leaf. x 1.5. . The galls shown in fig. 2. View from nether surface of nafs 4G. . Calophyllum inophyllum Linn. Marginal leaf gall. View from nether surface of leaf. x 1.5. PLATE XIV Dillenia reifferscheidia F.-Vill. Leaf galls. View from nether surface of leaf. The seven subhemispherical, white bodies along the midrib of the leaf are coccids, which have been accidentally introduced in the picture. They have absolutely nothing to do with the formation of the galls. 1.5. . Mallotus philippensis (Lam.) Muell.-Arg. Leaf galls caused by Neoheegeria mendax Karny. X 1.5. . Ficus ulmifolia Lam. Leaf galls caused by Gigantothrips elegans Zimmerman. Lateral view. xX 1.5. . Garcinia venulosa Lam. Leaf galls. View from upper surface of leaf. x 1.5. PLATE XV . Glochidion album (Blanco) Merr. Leaf galls caused by Gelechide. View from nether surface of leaf. x 1. The gall shown in fig. 1. View from upper surface of leaf. x 1. Raphidophora perkinsie Engl. Leaf galls. Proximal portion of leaf. x 1.5. The gall shown in fig. 38. Distal portion of leaf. x 1.5. Schefflera odorata (Blanco) Merr. and Rolfe. Leaf galls caused by Gynaikothrips chavice var. heptapleuri Karny. View from upper surface of leaf. x 1.5. . The galls shown in fig. 5. View from nether surface of leaf, show- ing openings of the galls. x 1.5. . Piper loheri C. DC. Leaf galls caused by Gynaikothrips chavice Zimmerman. View from upper surface of leaf; nether surface “of leaf exposed at rolled-up margin. x 1.5. s UICHANCO: PHILIPPINE PLANT GALLS, ] [PuHmip. JourN. Sci., XIV, No. 5. Logg Wh// bMababinbadss, ie A 4 xO” Hq O PLATE |. PLANT GALLS. : UICHANCO: PHILIPPINE PLANT GALLS. ] (Pup. Journ. Sci., XIV, No. 5 PLATE Il. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS. ] [PHILIP. JouRN. Sct., XIV, No. 5. ( WANS AY | | \ | We \ f ye, Wi ZZ PLATE Ill. PLANT GALLS. UIcHANCO: PHILIPPINE PLANT GALLS.] [Pururp. Journ. Sct., XIV, No. 5 6 PLATE IV. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS.] [PuHiuip. Journ, Sct., XIV, No. 5 PLATE V. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS. ] (PHILIP. JouRN, Sci., XIV, No. 5 PLATE VI. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS. ] [PuHiurp. JourN. Sct., XIV, No. 5 a PLATE VII. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS. ] (Pumie. Journ. Scr., XIV, No. 5 PLATE Vill. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS. ] {PuHitre. Journ. Scr, XIV, No. 5 ‘ : fe 5 NO. 6, ¢ PLATE IX. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS. ] [Puiuie. Journ. Scr., XIV, No. 5 PLATE X. PLANT GALLS. [PHmip. Journ. Scr., XIV, No. 5. UICHANCO: PHILIPPINE PLANT GALLS.] PLANT GALLS. PLATE XI. UICHANCO: TF HILIPPINE PLANT GALLS. ] [Purre. Journ. Sci., XIV, No. 5 PLATE XII. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS, ] [PuH1Iuie. JourRN. Sct., XIV, No. 5. PLATE XIll. PLANT GALLS. UICHANCO: PHILIPPINE PLANT GALLS.] [PuHILie. Journ. Scr., XIV, No. 5 - ° PS oO. oO. 4 PLATE XIV. PLANT GALLS. UIcHANCO: PHILIPPINE PLANT GALLS. ] (Puiuie. Journ. Scr., XIV, No. 5 PLATE XV. PLANT GALLS. THE SPHECODINE BEES OF THE PHILIPPINE ISLANDS By T. D. A. CockERELL (University of Colorado, Boulder) The genus Sphecodes Latreille consists of small or medium- sized bees, usually with red abdomen, widely distributed over both hemispheres. The venation, with three submargina] cells, well-developed stigma, and arched basal nervure, is much like - that of Halictus, but the end of the female abdomen is quite without the characteristic rima of that genus. The second sub- marginal cell is narrow, higher than broad. Sixteen species are known from India, three from Japan, one from Formosa, and three from Java. There are also a species in Celebes, one in New Guinea, and one in Australia. Up to the present time only one Philippine species has been known, but four others may now be added. They may be separated by the following table: Females eo cldeeaptrgetelniea ebb cinigs Rope made at Sappaac, Abra Province, based on assumed true area. € Only sufficient rope to make one test. “No test made. 624 Philippine Journal of Science 1919 FICUS BENJAMINA Linn. Balete.™ Local names: Aufvgu (Isabela) ; baléte (Ilocano) ; baletéon, salisi (Nueva Vizcaya); balete-puld (Laguna, Tayabas); baléti (Abra, Bataan, Cama- rines, Cavite, Ilocos Norte, Laguna, Manila, Nueva Ecija, Pangasinan, Pampanga); baliti (Pampanga); baliting-ibon (Batangas, Rizal); gisi (Ibanak, Apayao Subprovince); kolis (Bataan); kuliamat (Negrito in Bataan); sirisin (Cagayan). A strangling fig with coriaceous, glabrous, densely nerved, smooth, entire leaves and small, axillary, globose fruits; through- out the Philippines at low altitudes. Ficus benjamina bast strips are salmon-buff and show con- siderable variation; some are soft and pliable, others are hard and stiff. They vary in width from 2 to 5 millimeters, their average thickness is 0.60 millimeter, and their mean length is 1,227 millimeters. The strands are about ten strips thick. When dry, rope made from this bast has a medium tensile strength and breaking length. Wetting the fiber reduces the tenacity by 2 per cent, which for practical purposes is negligible. All of the five wet test specimens ruptured in the unspliced portion of the rope, whereas two of the five dry ones failed in eye-splices. The maximum variations from the mean ten- sile strength in the dry and the wet series were 19 and 22 per cent, respectively. Immersion in water gives rise to a markedly higher mean elongation. In the Philippines bast from Ficus benjamina is not commonly used for rope making owing to its comparative scarcity. A summary of the tests of this species made in the Bureau of Science is given in Table XXX. TABLE XXX.—Physical tests of rope made from the bast of Ficus benjamina. [Rope made at Langiden, Abra Province.] ‘Mean diameter: : Millimeters G8 Inches 0.29 Mean perimeter, or girth: Millimeters = 23 Inches 0.91 True mean sectional area: Square millimeters 32.5 Square inches 0.050 ““Balete” or “baliti,’” in the majority of Philippine dialects, is @ broadly generic term used to designate all the “strangling” figs (Ficus spp.), and is rarely or never applied to any epiphytic or climbing plant of any other family. XIV, 6 King: Philippine Bast-fiber Ropes 625 TABLE XXX.—Physical tests of rope made from the bast of Ficus benjamina—Continued. Ultimate tensile strength (dry): Mean in kilograms 156 Maximum in kilograms 176 Minimum in kilograms 126 Mean in pounds 343 Maximum in pounds 388 Minimum in pounds 277 Ultimate tensile strength (wet): Mean in kilograms 153 Maximum in kilograms 164 Minimum in kilograms 119 Mean in pounds 338 Maximum in pounds 362 Minimum in pounds 262 Mean ultimate tensile strength per unit area (dry): Kilograms per square centimeter 480 Pounds per square inch 6,830 Mean ultimate tensile strength per unit area (wet): Kilograms per square centimeter A471 Pounds per square inch 6,700 Mean elongation at instant of rupture: Dry (per cent) 9 Wet (per cent) 15 Mean weight per unit length: Grams per meter 31.0 Pounds per foot 0.0208 Average breaking length: Meters 5,030 Feet 16,500 Moisture (per cent) 8.41 FICUS FORSTENI! Mig. Balete. Local names: Baléte, balite (Bataan, Mindoro, Rizal, Zambales) ; basakla (Abra, Ilocano); dalékit (Negros); latgaban (Moro in Cotabato); puos or puspus (Ilocano, Itneg). A strangling fig with oblong, distantly nerved, coriaceous leaves and ellipsoid, sessile, axillary fruits; throughout the Phil- ippines at low altitudes. Ficus forstenii bast is ochraceous salmon, and numerous fine fibrils protruding from the strips give the rope a ragged ap- pearance. Plate IV, fig. 11, shows this hairy character and would seem to indicate that the fiber has been beaten to make it more pliable. The average length of the bast strips in the rope obtained for test is 738 millimeters. The strands are four strips thick. When dry, rope made of Ficus forstenii is very low in tensile strength and breaking length. However, im- mersion for twenty-four hours in water improves the tenacity 166157——-5 626 Philippine Journal of Science 1919 44 per cent. All of the wet specimens failed in the unspliced portion of the rope, and the maximum variation from the mean | tensile strength in the five specimens tested was 21 per cent. Two of the dry specimens ruptured in splices, the greatest va- riation from the mean being 23 per cent. Wetting appreciably increases the mean ultimate elongation of the rope. A summary of the tests of this species made in the Bureau of Science is given in Table XXXI. TABLE XXXI.—Physical tests of rope made from the bast of Ficus forsteni. [Rope made at Dolores, Abra Province.] Mean diameter: Millimeters 7.0 Inches 0.28 Mean perimeter, or girth: Millimeters 22 Inches 0.87 True mean sectional area: Square millimeters 38.8 Square inches 0.060 Ultimate tensile strength (dry): Mean in kilograms 59.8 Maximum in kilograms 70.3 Minimum in kilograms 46.3 Mean in pounds 182 Maximum in pounds 155 Minimum in pounds 102 Ultimate tensile strength (wet): Mean in kilograms : 86.2 Maximum in kilograms 104 Minimum in kilograms 71.2 Mean in pounds 190 Maximum in pounds 230 Minimum in pounds 157 Mean ultimate tensile strength per unit area (dry): Kilograms per square centimeter 154 Pounds per square inch : 2,200 Mean ultimate tensile strength per unit area (wet) : Kilograms per square centimeter 222 Pounds per square inch 3,160 Mean elongation at instant of rupture: Dry (per cent) 11 Wet (per cent) 18 Mean weight per unit length: Grams per meter 37.3 Pounds per foot 0.0251 Average breaking length: Meters 1,600 Feet 5,260 Moisture (per cent) 9.71 _XIV, 6 King: Philippine Bast-fiber Ropes 627 FICUS PACHYPHYLLA Merr. Balete. Local names: Baliti (Laguna); lunug (Occidental Negros); pasakia (Abra, Itneg). A strangling fig with very coriaceous, glabrous, somewhat el- liptic, rather densely nerved leaves and nearly globose axillary fruits; of local distribution at low altitudes and widely distri- buted in the Philippines. Like Ficus forstenii bast, that from Ficus pachyphylla also is characterized by finely frayed filaments that cover the surface. The strips of bast are uniformly pecan brown; they average 2 millimeters wide, 0.28 millimeter thick, and 838 millimeters long. Each strand averages eight strips thick. Rope made of Ficus pachyphylla has good strength, particularly when wetted. When dry it has a medium tensile strength and breaking length, but after immersion in water for twenty-four hours its tensile strength is increased 17 per cent. However, the results are discordant, the maximum variation from the mean of the five specimens being 25 per cent. The dry tensile- strength values showed better agreement, the maximum varia- tion from the mean being 16 per cent. Four of the five dry, and three of the five wet, test specimens ruptured in eye-splices. Filipinos say that rope made of this bast is very durable. A summary of the tests of this species made in the Bureau of Science is given in Table XXXII. TABLE XXXII.—Physical tests of rope made from the bast of Ficus pachyphylla. [Rope made at Dolores, Abra Province.] Mean diameter: Millimeters 7.0 Inches 0.28 Mean perimeter, or girth: Millimeters 22 Inches 0.87 True mean sectional area: Square millimeters 32.3 Square inches 0.050 Ultimate tensile strength (dry): Mean in kilograms 150 Maximum in kilograms : 161 Minimum in ‘kilograms 126 Mean in pounds 330 Maximum in pounds 355 Minimum in pounds : , 277 628 Philippine Journal of Science 1919 TABLE XXXII.—Physical tests of rope made from the bast of Ficus pachyphylla—Continued. Ultimate tensile strength (wet): Mean in kilograms 176 Maximum in kilograms 209 Minimum in kilograms 132 Mean in pounds 388 Maximum in pounds 460 Minimum in pounds 290 Mean ultimate tensile strength per unit area (dry) : Kilograms per square centimeter 464 Pounds per square inch 6,600 Mean ultimate tensile strength per unit area (wet) : Kilograms per square centimeter 544 Pounds per square inch 7,760 Mean elongation at instant of rupture: Dry (per cent) 12 Wet (per cent) 17 Mean weight per unit length: Grams per meter 30.3 Pounds per foot 0.0204 Average breaking length: Meters 4,930 Feet 16,200 Moisture (per cent) 9.09 FICUS PALAWANENSIS Merr. Balete. Local names: Agamid, agamit (Abra, Itneg); baléte (Cavite, Laguna, Lanao, Palawan, Tayabas) ; tibig (Misamis). A large strangling fig with coriaceous, oblong, glabrous, acu- minate leaves and axillary ellipsoid fruits; found throughout the Philippines at low altitudes. : Like the other Ficus basts tested, the strips of Ficus palawa- nensis also are characterized by numerous fibrils that cover the surface. The bast strips are colored bittersweet pink; they average 0.41 millimeter thick, 3 millimeters wide, and 821 milli- meters long. The strands average eight strips thick. Of all the Ficus basts tested, Ficus palawanensis is the strongest, and it is one of the highest in tensile strength and breaking length. An examination of Table XLII will show that it is stronger than machine-laid rope made of maguey Govern- ment grade Cebu No. 2, and is nearly twice as strong as hand-laid maguey rope made of unclassified fiber extracted by the salt-water retting process in Ilocos Sur. In fact, Ficus palawanensis bast rope compares very favorably in strength with machine-made “G” grade abaca rope, and in one case exceeds the mean tensile XIV, 6 King: Philippine Bast-fiber Ropes 629 strength of abacad rope 26 millimeters in circumference made of “G” grade fiber. When wetted for twenty-four hours, rope made of Ficus palawanensis bast is increased, and maguey is de- creased, in tensile strength, and the former is 18 per cent more tenacious than the latter. Four dry, and four wet, specimens ruptured in the unspliced portion of the rope, but the results obtained from the dry series of tests were much more uniform than those from the wet test pieces. The latter behaved erratically, and the maximum varia- tion from the mean was 30 per cent, while the greatest variation from the average value in the dry series was 12 per cent. The mean elongation of the wetted specimens was considerably higher than that given by the dry test pieces. Filipinos say that rope made of Ficus palawanensis bast is both stronger and more durable than that made of maguey. On account of its great strength, toughness, and durability the fiber is used for making wild-hog traps. A summary of the tests of this species made in the Bureau of Science is given in Table XXXIII. TABLE XXXIII.—Physical tests of rope made from the bast of Ficus palawanensis. [Rope made at Dolores, Abra Province.] Mean diameter: Millimeters 6.1 Inches 0.24 Mean perimeter, or girth: Millimeters 19 Inches 0.75 True mean sectional area: Square millimeters 20.5 Square inches 0.032 Ultimate tensile strength (dry): Mean in kilograms 154 Maximum in kilograms 167 Minimum in kilograms 136 Mean in pounds «©3839 Maximum in pounds 868 Minimum in pounds 300 Ultimate tensile strength (wet): Mean in kilograms 157 Maximum in kilograms : 195 Minimum in kilograms 110 Mean in pounds 846 Maximum in pounds 429 Minimum in pounds 243 630 Philippine Journal of Science 1919 TABLE XXXIII.—Physical tests of rope made from the bast of Ficus palawanensis—Continued. Mean ultimate tensile strength per unit area (dry): Kilograms per square centimeter 752 Pounds per square inch 10,700 Mean ultimate tensile strength per unit area (wet): Kilograms per square centimeter 766 Pounds per square inch ~ 10,900 Mean elongation at instant of rupture: Dry (per cent) 2 Wet (per cent) 19 Mean weight per unit length: Grams per meter 24.7 Pounds per foot 0.0166 Average breaking length: Meters 6,230 Feet 20,400 Moisture (per cent) 9.08 BORAGINACEAE CORDIA CUMINGIANA Vid. Anonang-lalaki. 3 Local names: Anénang-laléki (Mindoro) ; maratarong (Ilocano, Benguet). A small tree with broadly ovate, cordate, somewhat hairy leaves, many-flowered inflorescences, and small, ovoid, fleshy fruits; of local occurrence at low altitudes in Luzon. The strips of this bast that constitute the strands of the rope procured for testing vary considerably in size and color; some strips are 15 millimeters wide and 1 millimeter thick, and are stiff and amber brown. Others are only 0.2 millimeter thick, 6 millimeters wide, soft, and buff. Plate V, fig. 1, shows that most of the Cordia cumingiana bast strips are broad and very coarse. The strands average seven strips thick. _ When dry, rope made of this bast is low in tensile strength and breaking length. Wetting decreases the mean tensile strength of the fiber 6 per cent, but for all practical purposes does not affect the mean elongation. All of the wet specimens broke in the unspliced portion of the rope, whereas three of the five dry specimens ruptured in eye-splices. The latter tests. showed more discordant results than the former, the maximum variations from the mean in the five dry, and the five wet, spec- imens being 20 per cent and 13 per cent, respectively. A summary of the tests of this species made in the Bureau of Science is given in Table XXXIV. XIV, 6 King: Philippine Bast-fiber Ropes 631 TABLE XXXIV.—Physical tests of rope made from the bast of Cordia cumingiana. [Rope made at Disdis, Benguet Subprovince.] Mean diameter: Millimeters 13.1 Inches 0.48 Mean perimeter, or girth: Millimeters 38 Inches 1.50 True mean sectional area: : Square millimeters 86.8 Square inches 0.135 Ultimate tensile strength (dry): Mean in kilograms 337 Maximum in kilograms 383 Minimum in kilograms 270 Mean in pounds 743 Maximum in pounds 845 Minimum in pounds 595 Ultimate tensile strength (wet): Mean in kilograms 316 Maximum in kilograms 356 Minimum in kilograms 275 Mean in pounds 696 Maximum in pounds 785 Minimum in pounds 605 Mean ultimate tensile strength per unit area (dry): Kilograms per square centimeter 388 Pounds per square inch 5,500 Mean ultimate tensile strength per unit area (wet): Kilograms per square centimeter : 364 Pounds per square inch 5,160 Mean elongation at instant of rupture: Dry (per cent) 16 Wet (per cent) 16 Mean weight per unit length: Grams per meter 75.6 Pounds per foot 0.0508 Average breaking length: Meters : 4,450 Feet 14,600 Moisture (per cent) 12.61 CORDIA MYXA Linn. Anonang. Local names: Anénang (Albay, Bataan, Batangas, Benguet, Camarines, Ilocos Sur, Laguna, Mindoro, Pampanga, Pangasinan, Rizal, Sorsogon, Tayabas) ; bibili (Visayan) ; giima (Balabac); saléyong (Tagalog); sina- ligan (llocos Sur). 632 Philippine Journal of Science 1919 A tree, usually about 10 meters high, glabrous or nearly so; leaves ovate to elliptic-ovate, margins usually repand; inflores- cences lax, axillary and terminal, 5 to 10 centimeters long; flowers white or yellowish-white, small; drupe fieshy, ovoid, about 1 centimeter long, yellowish white, the pulp scanty; com- mon and widely distributed in the Philippines at low altitudes. Cordia myxa bast is wood brown, and the strands of the rope used in the test are from eleven to fifteen strips thick. The strips average 3 millimeters wide and 0.10 millimeter thick. When dry, rope made of Cordia myxa bast is very low in tensile strength and breaking length. Wetting decreases the mean tensile strength 19 per cent. Filipinos say that the fiber is not suited for use in the wetted condition. Four of the five wet, and three of the five dry, specimens failed outside of eye- splices. The latter series of tests showed better agreement than the former, the maximum variation from the mean tensile’ strength being 15 and 34 per cent, respectively. A summary of the tests of this species made in the Bureau of Science is given in Table XXXV. TABLE XXXV.—Physical tests of rope made from the bast of Cordia myx. [Rope made at Langiden, Abra Province.] Mean diameter: Millimeters 7.96 Inches 0.31 Mean perimeter, or girth: Millimeters 25 Inches 0.99 True mean sectional area: Square millimeters 44.5 Square inches 0.069 Ultimate tensile strength (dry): Mean in kilograms 144 Maximum in kilograms 162 Minimum in kilograms 122 Mean in pounds 318 Maximum in pounds 357 Minimum in pounds 269 Ultimate tensile strength (wet): Mean in kilograms : 117 Maximum in kilograms 157 Minimum in kilograms 101 Mean in pounds _ 257 Maximum in pounds 845 Minimum in pounds 222 XIV, 6 King: Philippine Bast-fiber Ropes 633 TABLE XXXV.—Physical tests of rope made from the bast of Cordia myxa—Continued. Mean ultimate tensile strength per unit area (dry): Kilograms per square centimeter 324 Pounds per square inch 4,610 Mean ultimate tensile strength per unit area (wet): Kilograms per square centimeter 263 Pounds per square inch 3,730 Mean elongation at instant of rupture: Dry (per cent) 11 Wet (per cent) 14 Mean weight per unit length: Grams per meter 34.9 Pounds per foot 0.0235 Average breaking length: Meters 4,125 Feet 13,500 Moisture (per cent) 8.54 GNETACEAE GNETUM sp. Kuliat. Local names: Kalidt (Ilocano); kalidt or kandidt (Benguet); kandidt (Itneg) ; kulidt (Bataan). A small tree with oblong, glossy, entire, acuminate, thin leaves; flowers minute, crowded in the points of the inflores- cences; fruit ellipsoid, one-seeded, about 2 centimeters long, the seed edible; widely distributed in the Philippines at low alti- tudes; usually sylvan. Gnetum sp. bast strips are rich brown, free from irregu- larities, and have a decidedly waxy touch and appearance. The waxy surface imparts a glossy sheen and, combined with the smoothness of the strips, assists them to slip over each other when strained. The strands of rope average ten strips thick. The strips vary in width from 1 to 3 millimeters, average 0.25 millimeter thick, and have a mean length of 503 millimeters. It will be noted that the strips are not only narrow and thin, but that they are shorter than those of any other bast measured. Of all the bast ropes tested, that made of Gnetum sp. fiber is the most tenacious and has the highest breaking length; moreover, it is exceptionally pliable. Gnetum sp. rope is stronger than machine-laid maguey rope made of Government grade Cebu No. 2 fiber and in tenacity approaches closely cordage made of the most superior grade of abaca fiber. When wetted for twenty-four hours this bast rope increases 31 per cent in 634 Philippine Journal of Science 1919 strength and is actually stronger than machine-laid abaca rope made of “F”’ grade fiber. All test specimens of both the five dry, and the five wet, specimens, excepting one in the latter, ruptured outside of eye- splices. The maximum variation from the mean tensile strength of the dry specimen values was 29 per cent, and that of the wet specimens was only 11 per cent. Wetting not only greatly increases, but improves, their uniformity. Rope made of Gnetum sp. bast is held in high esteem by Filipinos because of its great strength, pliability, and lightness, and is considered by Igorots and Ilocanos in Benguet, * Abra, and Union superior to that made of any other local fiber. It is preferred for making wild-hog traps and for mounting seines and for trawl fishing. Fiber for these purposes is subjected to severe exposure and treatment and must be durable and re- liable. When used for mounting nets and for fishing lines the bast meets with extraordinarily hard usage. It is customary to stretch a rope from one bank of a river to the other, and at regular intervals to suspend short lines to which hooks are attached. The rope of course is submerged, and frequently con- siderable débris becomes lodged against it. More than one instance is on record in which a man was saved from being washed away by grasping such a rope, of about the size of a lead pencil. A summary of the tests of this species made in the Bureau of Science is given in Table XXXVI. TABLE XXXVI.—Physical tests of rope made from the bast of Gnetum sp. [Rope made at Disdis, Benguet Subprovince.] Mean diameter: Millimeters 5.7 Inches 0.23 Mean perimeter, or girth: Millimeters 18 Inches 0.71 True mean sectional area: Square millimeters 21.5 Square inches 0.033 Ultimate tensile strength (dry): Mean in kilograms 166 Maximum in kilograms 214 Minimum in kilograms 142 Mean in pounds 366 Maximum in pounds 471 Minimum in pounds 312 XIV, 6 King: Philippine Bast-fiber Ropes 635 TABLE XXXVI.—Physical tests of rope made from the bast of Gnetum sp—Continued. Ultimate tensile strength (wet): Mean in kilograms 217 Maximum in kilograms 240 Minimum in kilograms 205 Mean in pounds 478 Maximum in pounds 529 Minimum in pounds 452 Mean ultimate tensile strength per unit area (dry): Kilograms per square centimeter 773 Pounds per square inch 11,100 Mean ultimate tensile strength per unit area (wet): Kilograms per square centimeter 1,000 Pounds per square inch 14,500 Mean elongation at instant of rupture: Dry (per cent) 10 Wet (per cent) 12 Mean weight per unit length: Grams per meter _ 19.7 Pounds per foot 0.0132 Average breaking length: Meters 8,450 Feet 27,700 Moisture (per cent) 11.45 BOMBACACEAE BOMBAX CEIBA Linn. Malabulak. Local names: Boboi (Tagalog); bobor or tarokték (Abra, L[locano) ; bubui-gubat (Mindoro, Rizal); ddldol (Ilocano); kdpas, kapdi (Jolo) ; malabilak (Laguna, Nueva Ecija). A very large deciduous tree, the trunk with large pyramidal spines; leaves palmately compound, leaflets oblong to lanceolate, 10 to 20 centimeters long; flowers very large, red, appearing before the leaves; capsules about 15 centimeters long; seeds hairy; throughout the Philippines at low altitudes. Bombax ceiba bast is orange-buff. The strips of rope procured for the test average 6 millimeters wide, 0.388 millimeter thick, and 967 millimeters long. Each strand averages ten strips thick. When dry, rope made of this fiber is medium in tensile strength and breaking length. Wetting diminishes its mean tensile - strength 13 per cent. Three of the four dry, and one of the five wet, specimens broke in eye-splices. The maximum varia- tion from the mean tensile strength of the dry test pieces was 9 per cent, and that shown by the wet series was 37 per cent. 636 Philippine Journal of Science 1919 This bast is rare and, therefore, is not commonly used for rope making. Filipinos say that it is suitable for wet-weather use. A summary of the tests of this species made in the Bureau of Science is given in Table XX XVII. TABLE XXXVII.—Physical tests of rope made from the bast of Bombax ceiba. {Rope made at Langiden, Abra Province.] Mean diameter: Millimeters 8.3 Inches 0.33 Mean perimeter, or girth: Millimeters 26 Inches 1.02 True mean sectional area: Square millimeters 43.0 Square inches 0.067 Ultimate tensile strength (dry): Mean in kilograms 174 Maximum in kilograms . 184 Minimum in kilograms 158 Mean in pounds 383 Maximum in pounds 405 Minimum in pounds 348 ‘Ultimate tensile strength (wet): Mean in kilograms 151 Maximum in kilograms 176 Minimum in kilograms | 94.8 Mean in pounds 332 Maximum in pounds 387 Minimum in pounds 209 Mean ultimate tensile strength per unit area (dry) : Kilograms per square centimeter 405 Pounds per square inch 5,720 Mean ultimate tensile strength per unit area (wet): Kilograms per square centimeter 351 Pounds per square inch 4,960 Mean elongation at instant of rupture: Dry (per cent) 8 Wet (per cent) 10 Mean weight per unit length: Grams per meter 37.7 Pounds per foot 0.0253 Average breaking length: Meters 4,620 Feet : 15,150 Moisture (per cent) 15.78 XIV, 6 King: Philippine Bast-fiber Ropes 637 ULMACEAE TREMA ORIENTALIS Blume. Anabiong. Local names: Agandung ( Cagayan) ; alindagon (Balabac, Moro); ana- bion (Visayan and Tagalog) ; anabiong (Mindoro, Negros, Rizal); anagum (Bicol) ; anariong (Batanes); anarong (Zambales) ; andaluyong (Bukid- non); annadung, hubulos (Bontoc) ; dalunot maladurung (Bataan); hagod (Laguna, Tayabas) ; hanag-ding, balibégo (Rizal) ; hanadiéng, hinagding, inangdén, kungdén (Mindoro); hanagdéng (Guimaras, Samar, Surigao, Tayabas) ; hinagding (Visayan) ; imangdén (Mindoro) ; indai-luging (La- nao); lagod, balibégo, dalunot, hanadiéng (Tagalog); lamai, arandong (Abra); malarurung (Igorot, Tagalog); malasikongdiron, dalunit, hinla- laong (Pampanga) ; nagdén (Occidental Negros) ; pargarandéngen (Panga- sinan) ; pangarandingin (Benguet) ; tatagtég (Guimaras). A small tree with oblong, ovate, somewhat harsh, oblique, acuminate leaves; small axillary inflorescences of very numerous minute flowers; and small, fleshy, ovoid drupes; very common and widely distributed in settled areas of the Philippines, Trema orientalis bast strips are cinnamon in color. They average 3 millimeters wide and 0.38 millimeter thick. The strands average nine strips thick. Plate V, fig. 5, shows that the strips are very irregular and ragged looking. When dry, rope made of Trema orientalis has the lowest ten- Sile strength and breaking length of all ropes tested. However, when wetted, its resistance is nearly doubled. Even so, the tensile strength islow. Filipinos say that this bast is very weak, and that it makes a poor rope. For this reason it is not com- monly used for the fabrication of cordage. A summary of the tests of this species made in the Bureau of Science is given in Table XXXVIII. TABLE XXXVIII.—Physical tests of rope made from the bast of Trema orientalis. [Rope made at Dolores, Abra Province.] ' Mean diameter: Millimeters 8.3 Inches : 0.33 Mean perimeter, or girth: Millimeters 26 Inches 1.02 True mean sectional area: Square millimeters 46.6 Square inches 0.072 638 Philippine Journal of Science : 1919 TABLE XXXVIII.—Physical tests of rope made from the bast of Trema orientalis—Continued. Ultimate tensile strength (dry): Mean in kilograms 62.6 Maximum in kilograms 79 Minimum in kilograms 46.7 Mean in pounds 138 Maximum in pounds 174 Minimum in pounds 103 Ultimate tensile strength (wet): Mean in kilograms 122 Maximum in kilograms 122 Minimum in kilograms 121 Mean in pounds 268 Maximum in pounds 269 Minimum in pounds 267 Mean ultimate tensile strength per unit area (dry) : Kilograms per square centimeter 134 Pounds per square inch 1,920 Mean ultimate tensile strength per unit area (wet): Kilograms per square centimeter 262 Pounds per square inch 3,720 Mean elongation at instant of rupture: Dry (per cent) 7 Wet (per cent) 21 Mean weight per unit length: Grams per meter . 44.1 Pounds per foot 0.0296 Average breaking length: Meters 1,420 Feet 4,660 Moisture (per cent) 8.90 ANNONACEAE GONIOTHALAMUS AMUYON Blanco. Amuyong. Local names: Amiiyong (Tayabas); lantitan (Occidental Negros); sa- gidt (Ilocano, Abra). A glabrous tree with oblong coriaceous leaves, rather large flowers, and cylindric, elongated aromatic fruits; of local occur- rence, but widely distributed at low altitudes in the Philippines. The most conspicuous characteristic of Goniothalamus amuyon bast is its attractive apricot-buff color. Like most of the basts tested, the fiber of this rope is in the form of strips, which average 3 millimeters wide, 0.33 millimeter thick, and 794 milli- meters long. : When dry, rope made of this bast is very low in tensile strength and breaking length. Wetting reduces the mean ten- sile strength 15 per cent. The wet specimens gave very uniform XIV, 6 King: Philippine Bast-fiber Ropes 639 breaks, the maximum variation from the mean being 9 per cent, while that shown by the five dry specimens was 21 per cent. Three of the dry, and two of the wet, specimens failed in eye-splices. A summary of the tests of this species made in the Bureau of Science is given in Table XXXIX. TABLE XXXIX.—Physical tests of rope made from the bast of Goniothalamus amuyon. [Rope made at Dolores, Abra Province.] Mean diameter: Millimeters 9.2 Inches 0.36 Mean perimeter, or girth: Millimeters 29 Inches 1.14 True mean sectional area: Square millimeters 40.9 Square inches 0.063 Ultimate tensile strength (dry): Mean in kilograms 141 Maximum in kilograms 157 Minimum in kilograms 112 Mean in pounds 311 Maximum in pounds 347 Minimum in pounds 247 Ultimate tensile strength (wet) : Mean in kilograms 120 Maximum in kilograms ' 130 Minimum in kilograms 109 ’ Mean in pounds 264 Maximum in pounds 287 Minimum in pounds 241 Mean ultimate tensile strength per unit area (dry): Kilograms per square centimeter 345 Pounds per square inch 4,940 Mean ultimate tensile strength per unit area (wet): Kilograms per square centimeter 293 Pounds per square inch 4,180 Mean elongation at instant of rupture: : Dry (per cent) 8 Wet (per cent) 13 Mean weight per unit length: Grams per meter 38.4 ; Pounds per foot 0.0258 Average breaking length: : Meters 3,680 Feet * 12,050 Moisture (per cent) 8.65 * 640 Philippine Journal of Science 1919 . VARIATIONS IN TESTS The handmade bast-fiber rope, with one or two exceptions, gave very erratic results. Taken as a whole the wet specimens gave somewhat more erratic results than the dry ones. For the dry specimens the maximum average variation from the mean was 17 per cent, and for the wet, 20 per cent. Table XL records the maximum variations from the mean in tensile- strength tests of five samples, each, of dry and wet rope specimens. TaBLE XL.—Mazimum variation from the mean tensile strength of rope in the dry and the wet series of five tests each. HANDMADE BAST-FIBER ROPES. | Maximum variation from mean tensile strength. Fiber. | : Dry Wet series. series, ; : Per cent. | Percent. Abroma fastuosa; unretted bast...-.-. 25-2 ooo. on ee es 21 22 A bron sastuoed: vetted pant. 2.5 oe ee ca ee a6 (b) Commerionic bast ee as ee ee 20 26 Fleliaterds Waving a aw ons os 17 il Tilaialiowits Omar oo a ek es Ss ne nsen mee epheeaae 17 30 Pterocymbium tinctoriwm -_....-----------------------------------------0-7--- 26 18 Plevoaperimum dvterniyonaie 2 nS 2 no a ee cn wee 14 8 Stevetilid Criramnbs oe ee 5 16 Sieweulia foalide os beeen ete 4 28 Steve Wolo a SS A aN ea eae 9 17 Stercislea stipalaae 66 ea 29 22 Bombycidendron vidalianwm: .. ..... 5. 2... 5 on oc ct etwas nse 18 30 Malashed: fiscal a ey Sin wow omen nena 20 6 Saeleh CEO ra a ee 12 e7 Thadpated Taide oak ee eae eee ewe 24 21 trend lobala ss ee ee ee eee 34 11 Colunibia blancok 3 ee ee eee 16 10 Corchorice olitaries (22 ooo oe re eae eanse 7 18 Grewia erinéarpa ...5 oS ee 19 34 Growin bilimalliia 25.050 oc ee ee ee 11 29° Grew multiiore.... Se a 13 29 Annee as Glauber eS a ee ee eee 15 19 Artocarpus communis; young bast ..__.--.------------------------------------ 16 Pheu Gena oo i a a ea enw eae 19 22 Ficus forstenii -..... eNO EM LES earn arongin emer rath 21 Wicks oaehyphella. 2 ee ee 16 25 Ficus palawanensis __..........------- ect A eteewh, a Oe 12 30 CuidtaGedaleka.. oS ec St 20 18 Conde Guess Se as ii needs ee a 15 34 Paaleiin pa e eS oa ee cs anen nonwnege 295 it Bombaz ceiba a eee 9 37 Tass prtentalie-o5 os i ee ee 25 (4) Glonsdabiaiais Wied oo a a cc cas cacupanw ens ncasnocsanes 21 9 Riba es a ee 17 aa e| XIV, 6 King: Philippine Bast-fiber Ropes 641 TABLE XL.—Maximum variation from the mean tensile strength of rope in the dry and the wet series of five tests each—Continued. Maximum variation! from mean tensile strength, Fiber. ass oie Dry Wet series. series. Per cent.| Per cent. Musa textilis; grade “F’’ abacd rope, $1 millimeters in cireumference _______. 5 8 Musa textilis; grade “F”’ abac4 rope, 15 millimeters in circumference ________ 10 13 Musa textilis; grade ‘‘G’’ abac4 rope, 26 millimeters in circumference ________ 7 6 Musa textilis; grade “‘G" abacé rope, 16 millimeters in circumference ________ 3 15 Agave cantala; maguey, grade Cebu No. 2, rope; 26 millimeters in circum- ROE ia ion sign tonreamcen otis pees Pe Ba 8 ll silamianetch as neh oe ee oe A ae 11 HAND-LAID ROPES MADE OF MISCELLANEOUS FIBERS, Cocos nucifera; rope 50 millimeters in-clrenmterenes 26 8 412 Cocos nucifera; rope 44 millimeters in tircumierenee 2 8 a2 Cocos nucifera; rope 24 millimeters in eireumilerente 22005 os a 25 20 Coregiie Ralus betta 76 oe ees 11 10 PN OTINNE ID sao sg co oa wniapaecat ig ae 15 18 PITT DINAN i 2 ie see oe ole ee es 14 e10 Dendrocdlamus merrilliamus.._ ooo. occ ence een 29 19 PROMI eset inkgret en sioe Sag ocakcct cece ge ee 16 13 * Three tests only. © Four tests only. > No test made. 4 Only two tests made and these are practically identical. In Table XLI are tabulated the number of rope test speci- mens ruptured, together with the manner of their failure. Two hundred forty-eight dry specimens were broken, of which 122, or 49 per cent, failed in eye-splices. Of the 217 wet test pieces 78, or 36 per cent, ruptured in eye-splices, Fifty-seven per cent of the dry, and 69 per cent of the wet, specimens that gave maximum tensile-strength values failed in the unspliced portion of the test pieces. Breaks in eye- splices do not necessarily imply that the various test specimens are faulty, abnormal, or low in value; for frequently the minimum values in both wet and dry specimens were obtained from those that failed in the unspliced portion of the rope between the eye-splices, whereas the maximum values were often obtained for pieces that ruptured in the eye-splices, In the dry specimens the number of breaks and of maximum values from test pieces breaking in the eye-splices is about equal to the number of those obtained from test pieces that 166157——6 642 Philippine Journal of Science - 1919 failed in the unspliced sections of the test specimens. How- ever, wetting reduced considerably the number of specimens breaking in eye-splices and seemed to cause failure in the un- TABLE XLI.—Number of rope test specimens ruptured and the manner of their failure. {U =in the unspliced section; E=in an eye-splice.] HANDMADE BAST-FIBER ROPES. poe S| ee tare 4 On ee plices. Fiber. Maximum Minimum value. value. Dry. | Wet. | Dry. | Wet. Dry | Wet | Dry | Wet = series. | series, | series. | series, Abroma fastuosa (unretted) ___ 5 5 2 1 U U U U Commersonia bartramia_..-.-- 5 5 4 1 U U E U Helicteres hirsuta_..........__- 5 5 0 eee 8 U U EB Kleinhovia hospita._.......-_.. 5 5 0 2 ee 33 U U U Pterocymbium tinctorium ____. 5 5 3 0 E U E U Pterospermum diversifolium__- 5 5 5 1 E U E U Sterculia crassiramea__________ 5 5 1 8 U U U E Sterculia foetida ...-._......... 5 4 0 se ee E U U Sterculia oblongata ___. 5 5 2 $s). 0 U U U Stereulia stipularis __._.__._._. 5 5 2 5| U E U E Bombycidendron vidalianum 5 5 1 7 eae 9 U U U Malachra fasciata_.__.....__.__ 5 5 2 s Bees ob U E E Mdseusila 2 5 4 3 $1.0 U E E Thespesia lampas -_.._________. 5 5 3 1 E U U U Crone lobate 5 5 5 4 1-28 U E U Columbia bla bcos 5 5 2 2 U U E U Corchorus olitorius.._....--.... 5 5 8 Cd pee SR U E Grewia eriocarpa __.__.....__.- 5 5 2 $1.28 EB. E E Grewia bilamellata_______ ae 5 5 1 0; U U U U Grewia multiflora... -. 5 5 3 8| B U U E Allaeanthus glaber ____..._.__.. 5 5 4 2 E E U U Artocarpus communis; (young Mays a a 5 5 2 ae U U U Ficus benjamina...........-..- 5 5 2 SO}. AE U E U Ficus forstenti --.... 2... 22... 5 5 2 o| U U E U Ficus pachyphylla...___._._____ 5 5 1 7 ee 8 E U E Fieus palawanensis______-_____ 5 5 1 1 E Ui U E Cordia cumingiana _........--- 5 5 3 0| £E U U U Cordia myzxa ....-.....-..-.-... 5 6 2 +i 2 E Ret Sas | EE custghdg ene IE 5 5 0 a By U U U Bombaz ceiba ___._...-....--._. 47°°- Ss 3 co Bees E js dae 2 Trema orientalis ____.......... 5 2 ‘ ipce E U U Goniothalamus amuyon__._.__- 5 5 4k 8 U BR. (28 ae, oe Na 19} 165/168) ss] @ | ® | © | @ *19U, 59 per cent; 13E, 41 per cent. 422U, 69 per cent; 10E, 31 per cent. >23U, 72 per cent; 9B, 28 per cent. ! 42.7 per cent. © 20U, 63 per cent; 12E, 87 per cent. 182.9 per cent. © : XIV, 6 King: Philippine Bast-fiber Ropes 643 TABLE XLI.—Number of rope test specimens ruptured and the manner of their failure—Continued. MACHINE-LAID ROPES MADE OF Smee ar seca CORDAGE FIBERS. | Sedebusaes bro-| kenineye- | splices. | Occurrence of failure in specimen giving— Number of | tests averaged. | Fiber. Maximum Minimum value. _ value, Dry- | Wet. | Dry. | Wet. Dry Wet Dry Wet series. | series. | series. | series. Musa textilis\ grade “F”’ abacé rope, 81 millimeters in cireumference______-_.____ 14 10 14 6 E E E E Musa textilis; grade “GQ” abacaé rope, 26 millimeters in circumference ______._.___ 15 10 12 6 U U E E Agave cantala; grade Cebu i ING. Bits ae ee 15 10 | il 2 E E E U Total ics “| 30] ¥s7}/ 14| @ | © | © | & HAND-LAID ROPES MADE OF MISCELLANEOUS PHILIPPINE FIBERS. 1 | Cocos nucifera; coir rope, 50 | millimeters in circumference. 5 4 - 0 E U U U Cocos nucifera; coir rope, 44 millimeters in circumference. 8 4 i ego | U U E E Cocos nucifera; coir rope, 24 3 millimeters in circumference. 12 6 2 3 U U U U Corypha elata; buntal rope, 22 millimeters in circumference. 5 5 8 2 U U E E Annum U8. ce ies 5 5 1 2 E U U E Anamirta coceulus___.__.._.__. 5 3 0 U E U U Dendrocalamus merrillianus__- 5 5 4 4 E E E U WOU iol 45 82{ m7 n13 | (e) (b) {e) (=) Grand total 248 217! 0122 P73! (a) (r) (s) () °1U, 33 per cent; 2E, 67 per cent. 240.6 per cent. t 8E, 100 per cent. ° 49.3 per cent. ® 4U, 57 per cent; 8E, 43 per cent. P86 per cent. »65U, 71 per cent; 2E, 29 per cent. 424U, 57 per cent; 18E, 43 per cent. ® 84 per cent. _ *29U, 69 per cent; 13E, 31 per cent. 146.7 per cent. *24U, 57 per cent; 18E, 43 per cent. ™ 37.8 per cent. ‘t 27TU, 64 per cent; 15E, 36 per cent. spliced sections of a larger proportion of the specimens giving maximum values. On the other hand, the preponderance of minimum values is not found in those that broke in eye-splices. About half of the dry test pieces giving minimum values failed in the unspliced section of the test piece. As in the case of the maximum values, wetting had the effect of causing: a still less Beaten of breaks in eye-splices. 644 Philippine Journal of Science 1919 Machine-laid rope made of commercial cordage fibers gave very concordant results, the average maximum variations from the mean tensile strength in the dry and the wet specimens being 7 and 11 per cent, respectively. Handmade coir rope (coir is a filamentous fiber like abacé) also gives excellent agreement; this indicates that machine fabrication is not nec- essarily the main factor in determining whether the tensile- strength values will be concordant or not. Therefore, the failure of a test specimen in an eye-splice is normal and there can be no question as to the acceptability of such values. Apparently uniformity in tenacity values depends largely upon the nature of the fiber, and the irregular tensile-strength values must be attributed largely to the crude, nonuniform nature of the bast itself. The summary of averages of the tensile-strength tests, break- ing length, weight per unit length, elongation, girth, diameter, etc., of the test specimens is arranged by species in Table XLIl. SUMMARY The averages of two series of tests on the tenacity of Philip- pine bast-fiber ropes are given; the one on air-dry specimens, and the other on specimens that had been immersed in fresh tap water for twenty-four hours. In order to show the relative strength of bast ropes, as compared with standard fiber cordage, a few tests of machine-laid abacé and maguey rope are given. A few results of tests of miscellaneous ropes made of material not coming under the classification of either bast or standard cordage fiber have been included also. A primitive wooden rope-laying apparatus is described and illustrated. In Table III are recorded the mean tensile strengths of the dry bast ropes tested. Dry rope made of Gnetwm sp. bast, having a mean circumference of 18 millimeters, heads the list with a maximum mean tensile strength of 773 kilograms per square centimeter (11,100 pounds per square inch). This value com- pares favorably with the values for “F” and “G” grades abaca rope. Rope made of coir fiber, which ranges in strength from 170 to 185 kilograms per square centimeter (2,420 to 2,640 pounds per square inch), ranks with the poorest basts. In Table IV are recorded the mean tensile strengths of ropes that had been immersed for twenty-four hours in fresh tap water. Rope made of Gnetum sp. fiber still heads the series with even augmented tenacity. It will be noted that wetting XIV, 6 King: Philippine Bast-fiber Ropes 645 has a marked influence upon, and generally causes a slight decrease in, the tensile strength of most bast ropes. In Table V are recorded, in the order of their magnitude, the breaking lengths of the ropes tested, and the general order follows that of the relative tenacity. Of the bast ropes, the rope made of Gnetum sp. is the highest, with a breaking length of 8,450 meters (27,700 feet), which closely approaches the values of the various abaca ropes tested. Despite its light weight, rope made of coconut fiber, which is sometimes considered a standard cord- age material, has a breaking length inferior to most of the bast-fiber ropes tested. The elongation of bast-fiber ropes is generally less than that of most standard cordage fibers. Wet- ting has the effect of increasing the mean elongation of nearly all of them. The minimum and maximum averages for dry and wet specimens are 6 and 16 per cent, and 9 and 25 per cent, respectively. 1919 Philippine Journal of Science 646 069‘8 | 892 Lag 5 608 Mote ee WEN ee sags enyoenty osL‘b | 288 ¥29 #82 | o98’s =| OLe 80L v3 derie) svar ae i era as neary Pe Ore gee a DOIN DIME ong ‘S| OST pat 869 os‘b =| 088 ¥LZ a Sara Pear NaS a eee eqerewnng PianaD oars | 188 itd 8°86 0899 | 68 9% Soaked Wo anh TE OR ee WEAUOOS DEAE) oor*s | 098 092 Sit | Ost‘L | 809 p98 Se Poe eee WM ee snit0z1j0 sn4oYyodoD ore’b | 908 OLL eas - | OL =| B08 G9L Ps Seo Mea ee ATs ag 29 PRT ee joounsg wquinjoD 00z‘9 | 998 PPS TIT | os8'9 | 28h 38 ig Sener) Misuse h SPR poate WUE ee phnoen.1qnoe “ABA DINGO] DUBL/) ost*> | 162 OLE sot | 0088 =| 89% 18 Seale, SIE COSti a! TOON PE Te ae spduny msedsoy.], 06t'L | 209 oF 908 | 09L'9 | SLP oF a Suet amines ai WONTNL Ee ee eee Dynan VPLS O0L‘L | SF9 £92 SIt | 0806 | 189 863 Mee ae soba rts tn. SE a AiG kh DynosDs D4yoDIO PL 019°9 | 896 Lz8 SI =| Fe ‘BS | (089 88h OL Peete ge ON a ee wnUDIOpIA wospuspronquiog 002‘ | 998 682 80r | o08‘e =| 89% gtT PO ee ee MOET BULIOJ SLAIN YS DYND49IS ost*b | 162 9T9 612 | os9'9 =| 868 abs Ace ping 5 Men Hai erostinkan) Es mY D70U0]90 DY)NI.1995 0r8'S | 002 Sat 1°99 0rs‘s | 008 gt Ni See ie PIGS ii SBI Ti gS: MOOT Dprisos wymos0ss oseF | 808 828 88 | 0999 | 868 #90 ‘T Pe ee WHCN DaUDsSsD.L9 DYNI497S 069° | 192 991 8 "9h OrL‘s | 898 L9T BB PO ee eee RON ee wmniyofisioarp wnuseds01a¢g ost‘9 | SEF 860 ‘T 96h =| Oer‘d =| TSB 696 nite RSME HELMS NN ae” wenrLozoUIy wnIGquhoo.Lorg 00°F | 982 LEL res =| «OLE "P| «608 862 O08 RT ee ee SO se ee wydsoy piacyurary oz9"s | 968 989 11g | 0829 | 88 T9L a ee MMRE eee MAMI eet. 27 Dynsiry 89.1.9}99]0 7 08L°8 | 99% ons 6s oss" | 268 88 Bee Wee aea satis attra WN ee DIUWD.AGLDG DIUOS.LOULUWOD @) () (i) (1) | 0016 =| BF9 ss We ee ik See qs¥q poqze1 Jo opwut odor ‘neongsn/ Dmo.gy osa‘b | GIs 129 8s} O9L‘L =| gs £90 T ee wecubaehargte 0 iahyie “--- oqeuy |-~-388q Jo Sdyys epnad Jo epeul edor ‘neonjzens DuoiQV *YOUr “ua “Your “mo "bs 4ad | “bs sad | ‘spunog SOP "bs dad | “be ued | spunog | ‘so71y spunog | sony spunog | sony o'IOM *£1q i % poon sjuourtios 3e0U SUEN “oureu oy Rue;eS *Y}SueI}s [Isus} UBOPT ‘SHdO0U UAALI-LSVA © ‘emoads Aq pabunten ‘sadot vagy auddyyq fo 33894 4ayjo pun yj2buewys-epsua, fo Nhinumung— TTX Fav], 647 Bast-fiber Ropes : Philippine King XIV, 6 “BUj}89} B10Jaq BINOY aINOJ-AyUaM AeyeA de} ysery ur peStauiqns ado > *@pBUr 4893 ON ; 901} Bunok wory ysvg y “0t} Plo WOAZ yseq ‘SuoHAOdord [Enda ynoqe Uy S\8¥q #110740 S1oYyo10D PUB DINQ0) DUEL; JO oINIXIU B JO ope sem ayduTES adox SIHL p O8T ‘F 03L ‘8 096 ‘“F 008 ‘FT 08L ‘8 O9T ‘g 006 ‘OT O9L ‘L O9T ‘ 00L “9 () 088 ‘F ¥9Z 892 G88 8LP LSé O@T oer TST LIZ Lit OF6 "Pb 026 ‘T 021 ‘8 OOT ‘IT O19 ‘F 008 ‘g 00L “OT 009 ‘9 003 *% 088 “9 026 "@ 0L0 ‘9 TPT 9°29 PLT 99T PPL 1&8 PSI OST 8°69 9ST 626 TOL saehin shal ail hace) cetacean as ~oo"""" “oAnUD snucDnYyj01U0s) we eee ene ited hehe ttt ee 9199 xpquog . : . TT ie ee Piitog Soe ee A" 2 ee OD sndino0z4y Ct Tone ere ee eee 4 StUNUWULOD End.DI0LLP 1919 Journal of Science ippine al Ph 648 *BUT}S9} BtOJaq sInoy ANOJ-AjUeM} 193BVM d¥} YseTy Ul pax1eurqns edoy o “A1OD pd}jOL JayVM-qVS Worl ‘ang sodo[] ‘uBABORD ye apeU odoy uy ‘ado PUVI}S-9914} B OJUT PazySIA} UIA OATUG w “OpBUL 489} ON | “ING SOd0T] ‘saA[Ng Fe epBUI sdoy ,; “ANG SOdo[] ‘osurwog OjUeg je ape adoy y ‘yea, wTed-1INq ey} JO sajorjed ay} UL pUNoyZ sroqy ABMosBA ay] JO epeu adoy y ‘wyed 1inq ay} JO Jeet ey} Worl apeur odoy » ‘aqualog JO Nvaing ey} Je pouBalo oUIYOeUT Jeqy UloIy opeu edoy > ‘epeupuey ere Sedor Jayjo [Be | edor opeul-euryoeyy q ‘shep ou0-AjuaM} Jajem dv JUBUSEyS Ul pesIeuU, Useq Suravy Ioqye YUSuer}g ¢ 009‘F | 928 Pat G"§8 096 “F 6FS 861 Cy Pexemnana eats adi taeda {0 ORME: ts ea aie ae ds wnwowpy OLS‘T | OTT 8Té Kaa id a 6FT bar 1 Gag 5 ba peachy eee bi Pune ee sponges wm 82)N9000 DILIUDUY ore‘e =| GLT 829 982 088 ‘8 L&Z Fg8 elas aca ead oe OR ash ROM TR er SNUDIP LIU ENULDMI0ApUaCT OOT ‘Ze | ShLe Late 9LGe | 06F'S =| OAT ast ee op, EGR, VER SSR eee the puedes co mmemaeta mis ch nah aie | 0L0°S | OFT Oar T61 OF ‘2 OLT O06 rs MES ae ler ci Ache hw na ob 7 as he (accra Sale and cl tercny bah ay oz ype --""" 9 2109 ‘nuafionu 80909 Or6‘T | 98T 6LP LIZ 0F9 ‘2 gst 199 any Web air ea Nee 0 asin, 11 REE Ramen Ss fag" 02st are tae hl aah u d109 ‘Duafionu 80009 os9‘s =| 192 L6T 8°68 OST ‘8 S36 OLT Va ore err ne Pee ee ae ee ee eg ee ne () (1) () W 008s — | a8 188 80? Ie Nie ges PONS ee a o DzDIa DYAN.L0D 1) (i) (1) () 080 ‘*F 883 OLZ SUE Mame arias lsc win oeiiny a3 Airs AY MOP, Race Ri i maa 1 DIDJUDE aavdy () @ () (@) =| 098°S | 80F 228 ME OPO a Oe ee re ““"y DyDpUDs savby 022'6 | TS9 929 $83 OOF “OT =| 68h 60L BOR ce ea Pi tee oe MORNE ee 4% ‘ON NgaQ ‘nypzuvo sandy OOL‘9T | OST T 998 99T OOL‘ST =| OTL T Shs A+ Sie ian a Soe NRE i ni goats Pa ren oe tts re re od 008 ‘OT | 69 968 90% 009 “OT 4 818 ara AGAR RIN UR NaC DE 1 CE, TRE ial Cen Ri ts sige qpouqe ,,5),, opeis fe1pi7x%a7 venpy OOLSE | 826 918 ST 008 “8 | FL6 888 Bade eae aes WP tart er ee eee og 009‘8T | 9F6 09 ‘T LoL OOF ‘ST | 8h6 963 ‘T io Teun RGR NT RTE Se Wa ead PW oe qPoUqE JT ,, opuas ‘srryx07 venpy “Your "mo “your “Wo ‘bs 4ad; | “bs sad | *spunog | “sony “bs 4ad | “bs sad | "spunog | ‘sojty | spunog | sopiy spunog | “so7ty oe Sis *SODUTAOI OUBDOTT 9Y} U , | pesn Krmkanios pa i waver “ouleU OYyHUSIDg “YBu9I4s o[Isue} UBOY | | ‘Suadld ANIddITIHd SQOANVTITAOSIN JO AGVW SAdOU ' ‘ponurju0g—sarveds fig pabunstn ‘sados woqy amddyryg fo 83803 wayj,0 pun yQbuasys-ansuaz fo hamuumng—]]{TX Wav, . King: Philippine Bast-fiber Ropes 649 XIV, 6 *suo}}20do01d penba ynoq ‘S19}0UN1} UBD O¢ JO YISua] esnvs wv pey saad 48a} 10430 [Ie !S19}OUIZUID OT B UY SySUq 8NYL0}0 SNLOYIL0D PUB Y}nqG0) Dues; JO aANyx}UI B JO apuU sea odures ador sys, » ‘apeur 489} ON ; “‘yjBus] aBner » $30 9 = 6L0 08 6 9 3°08 We (OOK. Me Te Us-mjaeg )----~ oe Ddano014e DuMets) 0 | on luo lg or 8 9°08 | 9080°0 | 00B'ZT | Ogp'g. [-=n==--=-== anos vo. 5 PRR 8n3.10}170 en.40yo.10-) | ¥9°0 Let | 69°T 8 sts sia 1°86 ORS (Or. (8 Ce Winky 9, Sma tae soounng DrquenzoD | 98°0 49 880 1 Ir 8 L°8% 100 | ORR OD ee WAN | -“p Dnoen..qvos “awa MyngQ0) DUIL—) | 8870 96 =| SIT 08 &I 6 L’8p O00 | COPE! | Care | roster: estate ox. eae sodun] vysodsoy,y 080 92 | 96°0 ¥% 8I as 9°88 SN: Oe ea BAL O oe, ARR a coin DyNoD DPI 1z°0 ye iy LI or 8 6°LT SS OM TE ees ee DyDosD/ DAyon]D, 880 o2 =| 18°0 a as ar 9°82 SOD SO TORS Pree ueqasay j----- wnuDyppLa LoLpUopronquog | 88°0 O'L 13°0 3 OL L 9°92 TOS Oe ees ae BULIOS seinendiye wYyns1975 , 8P°0 2 88 vLs 3 1°98 S00 | CONV OE [ooh eenr rennet ere DIDOUOIGO DYNIL975 _ 88°0 OL 180 &% or 9 8°82 doened A Paced TREE oi 2, Seek PRIMI com Naess ccc ee To DpHoos pyns49}5 | 09"0 Lt | gat Ov Es 81s $°96 SO, TOO [Oboe oer wepeyerueg Po"-------~ DoUDsesDso DYNILAS 90 L’9 £80 13 aI II 6°82 bindisdind tts Sieandl c FRY Esse. dh RE WIRED beaten hata eta Aoreg |----~ wnyofssoarp wnutedso.torg £20 rst | 99°T ap 4 OL s »°96 SID OORT PRN ore ee sudeyereyy |-------- weniwoj}QUy WnLQUhZ01g 6r0 es a 68 gla SIs 6°26 $290°0 | OOL“@E {| OLB*g [->tnn t= See Avian. 2 ee Dydsoy maoyuropy 0 or | 88'T 98 8a g 2 rls 1890°0 | 096°@I | 096'g = [-~-~>>--- seta vglhsds Etye-20.) 30°, aa DInsary Se4az0N OFT 80 Ter | ost 88 sls Pla 6°86 E900 |. | OOERE | GEO) [rer concer oteeae nicocnn?, DIUWD424Dg DYUOSLOULUMLOD ¥e"0 T'9 $L°0 6r () () | 3°02 BELO" OOP TORR [reir ieriteetcenerenth ao. SANS, URSiaiaimmaiaepsrcer cs 38Bq pe3301 JO epeul odor ‘nsonzenf nmougy a0 OTT | OT 98 8a als L"88 SOND": QOSST. POURS. [mettre rere ee ste SE oan qeuq Jo sdjx4s apna “ue “usu “ua ‘mu | *9«99 49g | *2W90 wag ecteag — aod "yaaq | “suojayy a a supa | PUNT 7 odox Jo y3.a13 03 PA | Aa [enbe esuereyurns é ‘ ‘ . uo pore edna jo | JOGA BWV | osmans zo | auatomofwsony | supyeeeaeangy | pone sSges out ous ut Neat oie: A9JoUIUIP esRIVAW UB}SU! 843 7B UOT “Bsu0le esBIIAY ‘SUdOU UUalI-LSva 1919 Journal of Science ppine v Phil 650 “siaj}eWIJUIe Og JO YISus, oBnvs v *opeul 4893 ON ; ‘aaa, Bunod woIz ysugq y “aer} plo Wory ysBq f¢ pey Saoaid 4s90} Joyjo [jw !s1ojeuUIyUaD ONT ‘yIBuUs] aBney 0 | 36. | mt. | 6 st 6 pee | saz0°0 | ono'st: | oag'g: [rennet oro eaynieg fo ~--- uohnun snuppoyzoruoy 88°0 &°8 20° 9% TZ L TPP 9620 "0 099 “F Lo (Ui aca Ni Re oc: Sct e “- Suopuery |-----~ ene ee S2]DZUITLLO DULALT 880 £8 30 °T 96 or 8 L°L8 €920 "0 OST “ST OSS Fe ee 40} 7O1B} IO tOgog |--"" "~~ ~~ asec ““““pq1ao xpquiog &@'0 L°9 TL"0 8T rat or L°6T Z8T0'O 00L ‘LZ OSF ‘8 oF sage as akin apcban sn i qeyey ee eam IE. ce “ds wnjauy a a at It eve = | saan | DONDE” T OOE Drees Suvuouy |----~- sd gees 3. “7 pace myptog sO || Tat lost) | ge 913 ors 9°92 = | 8090°0 «| OOD PE. | OSPR ftnomete mnt atenoe Buoreyereyy |-------------- DuDouruns DYpLog ¥o'0 T’9 gL°0 61 61 rai Lvs 99100 00F “03 1 iy REA BG Ba CoN ee SISUOUDMDIDE SNOT ae i ee eae ut at 8708 | p0z0"O | OORAT «| O86" ->---->- >> shat haath RR nme vy hydnyond snang - 830 0k 18°0 rad 8T It £°l8 Tg20"0 092 ‘8 DO ee sndsnd to song |~--""~->-~ pad Thea ig? Spmopesos Snot 62°0 gL T6 “0 &@ ST 6 aa 2 8020 “0 00$ ‘OT OS ae ee ot Ee ~ puguencuag snory 0 | 96) =| SIT tf o8 (0) @) | 8's — | szsor0. | 006‘*T | 099% === =~ ==> Be chi, eaens aR eee fF sYUNUALOD sNd90,.L ¥3°0 19 gL°0 6T ST Tr 8°SS €ST0 0 009 ‘FT 1 a acme MRR Ra Mie he hpi erate | 1) be Sanaa neaeag “yStunUULoD sndin20jq4y a 9a gt 3 pre = | g890°0 | 0808 | OFP'Z [>= reneh in ete taqnps snygwnany]¥ v0 | ST jot fos Its ors 618 | T6900 | O00@E «| 099g >>> Lr a 4 Ah erat DLoyfynut DLMeL 08°0 9°L 96°0 xd 43 It 9°@& 61200 009 ‘Zt TLE Be Sei Ope el datas aoe apc ol sechae? «Ses HOS anaaaa auetas DIDPAUDN DINILE) ““aqou |, “ud ray ous mus | uaa Lag |"7U92 Lag nt. har Soe yoag | *s0a2077 ‘adoa Jo wai on ee) jenbe aoussayumno ‘ ; : 5 ede Aetiiaes tienes “ectsie aalaaaie Win count eae ie yo EA anes0Ay -eanydn Jo neo soeave Pa ine iiocs Igaoomans | pst picttiee Ja}OUIEIP aFvIOAV gees abkay 9 : penujuey—STAOU UAAIA-LSVA ‘penuryu0g—sarveds fiq pabuvwan ‘sados saqy oumddyryg fo 8180} 10410 pun y76ua.138-apsua) fo AsDuUwUngS— TTX I1avy, 651 Bast-fiber Ropes : Philippine King XIV, 6 “A109 peyjal Je}ea-7[ es Wolry “Ing soso ‘uBABovD ye epeur edoxy , ‘edor puvijs-sa1y4} B ojUt PO}SIMY SUIA OIQUG wu. “OpBUl 489} ON ; “ANS SOdol]T ‘daaing 4e spew adoy , “ANG soso] “STOJOUTTFUGD 0g JO YISua] eBnes v psy saoeid 489} aay}0 ye ‘Feo Wyed-jing ayy Jo sajoyed ey, ur punoz ‘oSujwog OjUNg 4 opeu odoy y 4 *S19}9UIZUI2 ONT ‘Yue, osney » Staqy tvjnosea ay} Jo apeur adoy , “ued Ling ey} Jo Fea, ay} woaz opeu edoy » ‘eoUaldg Jo NveIng ay} 4B pauLajo suTysVUT Jaqy Wory epeur adoy , ‘opeulpuBy e148 sedor t9y40 |e ‘ ador Spsul-ouIyIVW q ——, 92°0 sg °0 830 08 "0 s3°0 £9°0 82 °0 62°0 T8°0 62 °0 £8 °0 020 880 61 “0 68 °0 OL 6 2°91 tT3 613 T ‘09 912 sla 9°66 es 0g P'6I oP 68 "18 83 9 3 2°@IL ST las 8°06 () Sl 3 3°TIT () (° | 37s () () 6°88 Lis rae) 8°rs ST &I ST 6T 2 9T2 GLP cat 8 Lat Sl 2 i 8°8g 60T0 “0 ¥0F0 "0 6990 °0 08100 8hS0 0 9S10°0 OFTO “0 9F10°0 9TZ0 "0 2920 "0 T820 “0 bOTO “0 61€0 "0 $800 °0 9680 "0 OST ‘ST 009 ‘OT OS “ET 039 “TT 096 “8 019 ‘8 OOT ‘aT 006 ‘TT 008 ‘er 008 ‘ZI 002 “08 008 “$8 008 “LZ 000 “68 006 ‘Ob 08g *g 008 “8 008 “8 oss “g 08L ‘3 029 ‘2 001 ‘8 089 ‘¢ Is ‘s 092 ‘8 098 ‘6 002 ‘OT 068 ‘8 006 “IT 008 ‘aT Reig idiicdoa essed “"""" BIg Wee teen nee wee se cece op----- ee ew ewe ewe eee Tere g een wee eee ween ane aor rere weet eee wwe we eww ewes | eee ed teannne wn o====""--=- dg summouy aeeiegeiateaanaets wsnjnov00 DaUDULy sense SNUDYILLLIUW snUBMooLpUE eves. aot cealuas = awed ga ‘heme aan aaa “"" yu Tod ‘DAB LLINU 80909 srrocesosoess = 5 ge pdfs So i ane aes Eatee es “""""""=") pmumo eanby ““"""" 4% ‘ON NgeQ ‘nypzun0 savoy “"q BOBQE ,,2),, PPwss ‘s1y1979) Deny eee cess ee me oe ee a s eiaeieeted od "4 FORE ,,7,, PPVss ‘sryxeg nenyy ‘SUAGId ANIddITIHd SNHOANVTISOSIN JO AAV Sadou FIG. Fic. Fic. 8. 9. 10. ai; 12. 18. ILLUSTRATIONS PLATE I . Laying a three-strand rope having a mean girth of 30 millimeters, made of crude strips of Artocarpus communis bast at Sappaac, Abra, obtained from the trunk of an old tree. - Heavy hardwood rope-laying apparatus (pagtalian, Ilocano) used in Caoayan, Ilocos Sur, for making large-sized mooring ropes, together with two coils of rope made therewith. One coil of rope is made of maguey (Agave cantala) and the other of a species of bamboo (Dendrocalamus merrillianus). PLATE II . Laying a three-strand rope having a mean girth of 19 millimeters, made of retted Abroma fastuosa bast dt Sappaac, Abra. The view shows the triple-crank end of the crude rope-laying ap- paratus with strands still separate. In the foreground are two men holding spun and unspun bast. . Form of specimen used for making tensile-strength tests of Philip- pine bast-fiber ropes. PLATE III. PHILIPPINE BAST-FIBER ROPES . Rope having a mean girth of 19 millimeters, made of retted Abroma fastuosa bast. . Rope having a mean girth of 86 millimeters, made of crude (un- retted) strips of Abroma fastuosa bast. . Rope having a mean girth of 38 millimeters, made of crude strips of Commersonia bartramia bast. . Rope having a mean girth of 35 millimeters, made of crude strips of Helicteres hirsuta bast. . Rope having a mean girth of 39 millimeters, made of crude ribbons of Kleinhovia hospita bast. . Rope having a mean girth of 42 millimeters, made of crude strips of Pterocymbium tinctorium bast. . Rope having a mean girth of 21 millimeters, made of crude strips of Pterospermum diversifolium bast. Rope having a mean girth of 40 millimeters, made of crude ribbons of Sterculia crassiramea bast. Rope having a mean girth of 22 millimeters, made of Sterculia foetida bast. Rope having a mean girth of 38 millimeters, made of crude ribbons of Sterculia oblongata bast. Rope having a mean girth of 22 millimeters, made of crude strips of Sterculia stipularis bast. Rope having a mean girth of 22 millimeters, made of crude strips of Bombycidendron vidalianum bast. Rope having a mean girth of 17 millimeters, made of retted Malachra lineariloba bast. 653 654 Philippine Journal of Science 1919 14. Rope having a mean girth of 24 millimeters, made of retted Sida acuta bast. 15. Rope having a mean girth of 30 millimeters, made of crude Thes- pesia lampas bast. PLATE IV. PHILIPPINE BAST-FIBER ROPES Fig. 1. Rope having a mean girth of 21 millimeters, made of a mixture, in about equal proportions, of crude strips of Urena lobata and Corchorus olitorius bast. 2. Rope having a mean girth of 48 millimeters, made of crude strips of Columbia blancoi bast. 3. Rope having a mean girth of 22 millimeters, made of crude strips of Corchorus olitorius bast. 4. Rope having a mean girth of 24 millimeters, made of crude strips of Grewia bilamellata bast. 5. Rope having a mean girth of 20 millimeters, made of crude strips of Grewia eriocarpa bast. 6. Rope having a@ mean girth of 36 millimeters, made of crude strips of Grewia multiflora bast. 7. Rope having a mean girth of 39 millimeters, made of crude strips of Allaeanthus glaber bast. 8. Rope having a mean girth of 19 millimeters, made of crude strips of young Artocarpus communis bast. 9. Rope having a mean girth of 30 millimeters, made of crude strips of old Artocarpus communis bast. 10. Rope having a mean girth of 23 millimeters, made of crude strips of Ficus benjamina bast. 11. Rope having a mean girth of 22 millimeters, made of crude strips of Ficus forstenii bast. 12. Rope having a mean girth of 22 millimeters, made of crude strips of Ficus pachyphylla bast. 13. Rope having a mean girth of 19 millimeters, made of crude strips of Ficus palawanensis bast. PLATE V. PHILIPPINE BAST- AND MISCELLANEOUS-FIBER ROPES Fig. 1. Rope having a mean girth of 38 es serge made of crude strips of Cordia cumingiana bast. 2. Rope having a mean girth of 25 emcee. made of crude strips of Cordia myzxa bast. 3. Rope having a mean girth of 18 millimeters, made of crude strips . of Gnetum sp. bast. 4. Rope having a mean girth of 26 millimeters, made of sulle strips of Bombaz ceiba bast. 5. Rope having a mean girth of 26 millimeters, made of crude strips of Trema orientalis bast. : 6. Rope having a mean girth of 29 millimeters, made of crude strips of Goniothalamus amuyon bast. 7. Rope (two strands only) having a mean girth of 46 millimeters, made of a species of bamboo (Dendrocalamus merrillianus). 8. Rope having a mean girth of 44 millimeters, made of the entire stem of a vine Laois cocculus) . XIV, ¢ Ads 12. King: Philippine Bast-fiber Ropes - 655 - Rope having a mean girth of 50 ‘millimeters, made of coir (fiber extracted from the husk of the fruit of the coco palm, Cocos nucifera). - Rope having a mean girth of 47 millimeters, made of the leaf of the buri palm (Corypha elata). Rope having a mean girth of 22 millimeters, made of the vascular fibers (buntal) found in the petioles of the buri palm (Corypha elata). Rope having a mean girth of 20 millimeters, made of strips of Amomum sp. bast. _ TEXT FIGURES - Typical cross-sectional profiles of ropes tested. Actual size. a, rope made of Cordia cumingiana bast. 6, rope made of Kleinhovia hospita bast. ¢, rope made of split bamboo (Dendrocalamus merrillianus) . d, rope made of “F” grade abac4 (machine laid). . Showing the twist of two-, three-, and four-strand ropes, respectively. (Pum. Journ. Sci., XIV, No. 6. KING: PHILIPPINE BAST-FIBER ROPES. ] -Strand rope. Laying a three Fig: 1. Rope-laying apparatus. 2. Fig. PLATE I. KING: PHILIPPINE BAST-FIBER RoPEs.] (Pum. Journ. Sct., XIV, No. 6 Fig. 1. The triple-crank end of the crude rope-laying apparatus. a, ‘ yom aK ’ a ax Gare a ee Qe Se AEA Fig. 2. Form of specimen used for making tensile-strength tests of Philippine bast-fiber ropes. PLATE Il. KING: PHILIPPINE BAST-FIBER ROPES. |] [Puit. Journ. Scr., XIV, No. 6. PLATE Ill. SAMPLES OF PHILIPPINE ROPES. KING: PHILIPPINE BAST-FIBER ROPES. ] [Purm. Journ. Sct., XIV, No. 6 PLATE IV. SAMPLES OF PHILIPPINE ROPES. 6. XIV, No. [Pui JourRN. SCt., w wl o o or wl z a a. A = a re ° ” ul ad o = < 7) > Ww K < a a oe NRA S! 9 i > mas 3 ° 2 ce as fr eee " & ee at >, eS ! 3 POD r if fa KING: PHILIPPINE BAST-FIBER ROPES. ] PINK DISEASE OF CITRUS By H. ATHERTON LEE and Harry S. YArTss* (From the Bureau of Plant Industry, U. S. Department of Agriculture, Washington, D. C., and the Botanical Section of the Biological Laboratory, Bureau of Science, Manila, P. I.) SEVEN PLATES AND TWO TEXT FIGURES During field work in 1917 an apparently serious stem and branch disease of citrus trees was discovered. Subsequent and more extended observation has indicated that the disease may cause the death of the affected branches or even of the whole tree. The spread of the disease and subsequent destruction of the trees during the rainy season was found to be very rapid. Specimens of diseased material in various stages were collected, and examination showed the organism responsible to be the pre- viously well-described Cortictum salmonicolor B. & Br.2 Though this species of Corticitum is known to cause very serious diseases of Hevea brasiliensis (HBK) Muell., Theobroma cacao Linn., Coffea sp., Thea sp., and other plants in the Orient, and has previously been reported upon Citrus, it has apparently not been known to cause an important disease of the latter host. Since the disease at present appears to be localized in its occurrence and as its spread over the Islands may involve serious damage not only to Citrus but also to cacao, coffee, rubber, and possibly *The writers wish to express their appreciation to Mr. E. Bateson, director of agriculture and mycologist for the Government of British North Borneo, for assistance and many courtesies extended to the junior writer, while on a recent visit to that country. The senior writer is indebted to Col. Adriano Hernandez, director, and Mr. S. Apostol, chief, division of plant industry, Philippine Bureau of Agriculture, for their kindness in making available to him opportunities for field work. He wishes to thank also Dr. Alvin J. Cox, former director, and Mr. E. D. Merrill, botanist, of the Philippine Bureau of Science, for many courtesies and the facilities of the botanical laboratories of that bureau. 2The identification of Corticium salmonicolor B. & Br. is based, not only upon a careful study of the organism and a comparison with the published descriptions and figures, but also upon comparison of material of Corticium salmonicolor B. & Br. collected upon Hevea brasiliensis (HBK) Muell. and on Annona sp. in British North Borneo where an opportunity was found to observe pink disease as it occurs in the field on Hevea brasiliensis (HBK) Muell., Hibiscus sp., and Annona sp. 166157—1 657 658 Philippine Journal of Science 1919 other plants, it has seemed desirable to present certain data regarding the occurrence, appearance, and means of control of Corticium. Photographs of the various stages on Citrus are also presented, so that this disease may be recognized and the necessary measures applied for its control or eradication before it becomes widely distributed over the Philippine Islands. The name pink disease is commonly accepted by rubber planters throughout British Malaya and elsewhere as denoting the disease caused by Corticium salmonicolor B. & Br. This term is de- scriptive of the commonest form taken by the fungus and, as its use seems acceptable, it is continued here. Although Corti- cium has been thoroughly studied by us, no pretension is made that this paper contributes anything to the excellent studies on the morphology of the fungus already published by Rant, Brooks and Sharples, and others. aS HISTORY Corticium salmonicolor B. & Br. apparently first attracted attention about 1897 when it appeared as a disease of coffee in Malaya. Specimens sent to Kew were described by Massee * as a new species and genus of Fungi imperfecti which he named Necator decretus Massee. In 1901 Zimmermann,‘ working on the disease on coffee in Java, described the organism as Corticium javanicum Zimm. In 1904 Ridley * mentioned a specimen of a fungus occurring upon Hevea that had been sent from Sandakan, apparently to be referred to this species, and in 1905° he re- ported the disease on two estates in the Malay Peninsula. Ma- terial sent to Kew was determined by Massee, this time as Corticiwm calceum Fr., a harmless and widely distributed species of Europe and America. In 1906 it was also found by Petch’ to attack Hevea in Ceylon and southern India, and was recorded as Corticium javanicum Zimm. In 1909 Gallagher *® reported Corticium zimmermanni Sacc. & Syd., which is a synonym of Corticium salmonicolor B. & Br., to be associated with a disease of rubber trees in Malaya. Petch® showed that material of this fungus was collected by Thwaites in Ceylon and was named *Kew Bull. (1898) 119. “Centralbl. f. Bakt. Abth. 2, 7 (1901) 146. * Agr. Bull. Straits & Fed. Malay States 3 (1904) 174. “Agr. Bull. Straits & Fed. Malay States 4 (1905) 423. "Report of Gov. Mycologist Ceylon (1906). *Bull. Dept. Agr. F. M. S. 6 (1909). ) *Petch, T., Physiology and Diseases of Hevea brasiliensis.. London (1911) 209. : XIV, 6 Lee and Yates: Pink Disease of Citrus 659 Corticium salmonicolor by Berkeley and Broome in 1873; as this name has undoubted priority over all other names proposed, it must be used for the species. Rant" has described the disease known as ‘‘djamoer oepas” (poison fungus) in Java, caused by Corticium javanicum Zimm., and has listed one hundred forty-one wild and cultivated plant species attacked by the fungus; among these are Citrus sp. and Murraya exotica Linn. Inoculation experiments made by Rant proved for the first time the identity of Necator decretus Massee as a form of Corticium javanicum Zimm. He also published a number of very good figures. In 1914 Brooks and Sharples‘? gave the history of the disease in the Malay Peninsula together with a description of the life history of the fungus and sug- gestions for treatment. In 1908 von Faber * reported its oc- currence on coffee in the Cameroons. It has also been reported to our knowledge from Burma, Sumatra, Borneo, southern India, the Caucasus, and Formosa. From a study of the history of the disease we find it to be very widely distributed throughout the eastern Tropics, and to occur upon a very large number of economic and wild plants. In 1900 the disease was practically unknown. Sharples stated that in 1912 the fungus causing pink disease was known in only three small centers in Malaya, but by 1914 it had spread over practically the whole peninsula. One of us on a recent visit to British North Borneo found ‘the disease to be very common on rubber in that country, where it would undoubtedly cause serious financial losses if control measures were not applied. CORTICIUM IN THE WEST INDIES A disease of cacao, lime, and other economic plants has been reported from the West Indies. It appears to have been long known as pink disease and was supposed to be due to a Corti- cium; however, it was believed to be a different species from that known in the East and was identified as Cortictwm lilacino- fuscum B. & C., a species described from temperate North America. Stockdale ** calls attention to the existence of a sim- ilar disease in the East, but does not suggest the possibility ® Journ. Linn. Soc. Bot. 14 (1878) 71. * Bull. Jardin Bot. Buitenzorg 4 (1912) 1-50, f. 1-14. % Bull. Dept. Agr. F. M, S. 21 ( 1914). 8 Centralbl. f. Bakt. Abth. 2, 21 (1908) 112. _™ Kew Bull. (1917) 227. , % West Indian Bull. 9 (1908) 178. eo 660 Philippine Journal of Science 1919 that the organisms may be identical. Petch** appears to have been the first to suggest the identity of the organism causing pink disease in the two regions. Though not all of the literature relating to the disease occurring in the West Indies is available here, we note that recently Stevenson ** has reported Corticium salmonicolor B. & Br. as the cause of a disease of Citrus in Porto Rico. The determination of the organism is credited to Doctor Burt, of the Missouri Botanic Garden. OCCURRENCE OF CORTICIUM SALMONICOLOR B. AND BR. IN THE PHILIPPINES Corticium salmonicolor B. & Br. has been found only recently in the Philippines, and the supposition seems justified that it is of recent introduction here. It was first collected in the fall of 1917 at Novaliches, near Manila, on Annona sp. Since that time it has been found at Los Bafios, Laguna Province, and abundantly at Lamao, Bataan Province. The Lamao experi- ment station is the center of plant introduction and distribution in the Philippines, and plant materials are known to have been shipped from Lamao to both Novaliches and Los Bafios. It seems very likely that Corticium salmonicolor B. & Br. was brought to Lamao on some plant introduced from another part of the Orient; for, while the fungus flora of the Philippines is far from being fully known, it seems very unlikely that botan- ical collectors and agricultural inspectors would have overlooked so conspicuous and destructive a fungus as Corticitwm—one likely to occur upon a very large number of hosts—had it existed in the Philippines for any considerable period. As Corticium is very destructive to Hevea brasiliensis in other parts of the Orient, it is interesting to note that ap- parently it has not made its appearance on this host in the Phil- ippines. The junior writer, while visiting the Basilan Rubber Plantation in November, 1917, made a careful survey of the estate for pink disease, but none was discovered. While Para - rubber is cultivated in other parts of the Philippines, the Ba- silan plantation is the oldest and most accessible, and probably the one most likely to be infected through the introduction of young plants from Malaya. In the Philippines, Corticiwm sal- monicolor B. & Br. has been collected only upon cultivated species, never upon naturally occurring plants. ” Circulars. & Agr. Journ. Roy. Bot. Gard. Ceylon 4 (1909) 189. * Stevenson, John A., Fifth Report of the Board of Commissioners of Agriculture of Porto Rico (1915-1916) 43, 44. XIV, 6 Lee and Yates: Pink Disease of Citrus 661 DESCRIPTION OF THE DISEASE ON CITRUS The first stage of the disease appears to be a slight gumming (Plate II, fig. 1, Plate IV, fig. 2), similar in many ways to the gumming which is produced coincident with many other injuries to the bark of Citrus. At this stage the bark seems dry and very hard, and it adheres closely to the wood. Often longi- tudinal cracks form in the bark at the same time with the gumming. Usually the first appearance of the fungus is in the form of pustules, which push through the hardened bark (Plate I, fig. 1, Plate II, fig. 1). These pustules are dirty white to pinkish in color, usually 1 millimeter or less in diameter, and rather irregular in shape when observed under the hand lens; so far as known, they are sterile. These pustules appear to be the “Hockerchen” form described by Rant. Following the pustule stage, or sometimes occurring without any appearance of it, there is a formation of the mycelium of the fungus over the surface of the branch. This mycelium spreads and extends over the entire lesion (Plate IV, fig. 1, Plate III), forming a pink incrustation, which varies somewhat in shade but perhaps typi- cally approaches most closely the “orange-pink” of Ridgway’s ** nomenclature. This color is the most striking characteristic of the fungus (Plate I). The incrustation is at first almost entirely smooth and of a velvety appearance, but with age it becomes roughened and broken up into patches, passing through tran- sitional stages of color until it finally fades out to a dirty white. The mycelium advances up and down the branch as a char- acteristically delicate, smooth, fan-shaped white mycelium (Plate li; fe 2): In some cases, upon cutting through the bark beneath the actively advancing mycelium, no visible effect of the fungus is) found even though the mycelium on the surface has advanced | considerably beyond the cut; in others the bark is shown to— have a watery, darkened, greenish appearance; while in still other cases the affected area extends beyond the limits of the surface mycelium. This variation “may perhaps be explained by different seasonal conditions. As the disease advances, the watery appearance of the bark passes off, and the tissue becomes still darker in color and presents a dried-out appearance. The bark at this stage is a dry mass of shredded fibers, running * Ridgway, Robert, Color Standards and Color Nomenclature. Wash- ington (1912) pl. 2. 662 Philippine Journal of Science 4919 longitudinally with the branch; the last condition seems typical of the effect of Corticiwm. 3 On Citrus the outermost layers of the wood become discolored to a light brown. The extent of this brown tissue, however, is small in comparison with the area covered by the surface ‘mycelium of the fungus, and the discoloration of tissue has never been observed to extend deep into the wood. Brooks _ and Sharples, in their study of the disease on rubber, found \ that the fungus hyphae followed the medullary rays well into ' the tissue. : The ultimate effect of the growth of the fungus is the stop-. page of the water conduction, followed by the wilting of the leaves and the ultimate death of the affected limb. This last stage takes place very rapidly; frequently the leaves wilt, yellow, and drop within a period of five days. Small twigs, main branches, or even trunks of trees are attacked (Plate VI, fig. 1, Plate VII, fig. 2). It has never been observed on roots. The susceptibility of young and old trees seems to be equally great. The accompanying photographs probably show these stages much more clearly than can a written description. Pink disease has been found upon the following horticultural varieties of Citrus species: Citrus nobilis Lour.: Tizon, Kishiu, Suntara Nagpur. Citrus sinensis Osbeck: Native cajel, Pineapple orange, Washington navel, Mediterranean, Valencia, White Siletta, Jaffa, St. Michael, Sawyer’s navel, Navelencia. ; Citrus aurantifolia Swingle: Tahiti, West Indian, Dayap. Citrus limonia Osbeck: Valencia, Clark, Villa Franca, Messina, Belair, Lisbon. — Citrus aurantium Linn.: Seville. Citrus maxima Merr. (C. decumana Linn.): Marsh, Triumph. Citrus medica Linn, var. sarcodactylis. Citrus hystrix DC. ; Citrus micrantha Wester, Citrus webberi Wester var. montana Wester. ‘Citrus species: Natsumikan. Citrus hybrid: Tangelo, Sampson. Feronia elephantum Corr, and Murraya exotica Linn., the genera of which are closely related to Citrus, have also been observed as hosts of Corticiwm salmonicolor B. & Br. in the Philippines. The omnivorous character of the fungus and the diversity of Citrus varieties attacked by Corticiwm seem to in- dicate that the occurrence of species or varieties of Citrus re- sistant to the disease is highly improbable. XIV, 6 Lee and Yates: Pink Disease of Citrus 603 DISSEMINATION OF THE FUNGUS Very little is known about the means of dissemination of the fungus. Our observations upon the pustule, or “Héckerchen,”’ stage have always shown it to be sterile. We have always ob- served the pink incrustation, which is the most noticeable stage of the fungus, to be sterile upon Citrus. Brooks and Sharples report and figure basidiospores upon the pink incrustation formed on rubber;;but we have not observed them on the in- crustation as it occurs upon Citrus. Plate II, figs. 3 and 4, show the gross appearance of the stage first described by Massee as Necator decretus. This stage is found very abundantly on Citrus during the early months of the rainy season, usually upon wood that has been previously killed by the fungus. It is conspicuous for its bright orange color (bitter-sweet orange of Ridgway), very distinct from the pink color of the sterile mycelium of the fungus. Irregularly shaped, thin-walled spores are formed abundantly in this stage; and these spores would appear to be the principle means of dissemination of the fungus, at least over any considerable distance. This fact assumes a bearing on the methods of control, since it indicates the value of the removal of all twigs and branches affected with the disease before the appearance of the Necator stage. The Necator stage as it occurs on rubber has been well figured and described by Brooks and Sharples; and, in general, our observations of this stage on Citrus confirm their conclusions. CONTROL EXPERIMENTS In the fall of 1917, in connection with experiments in the control of citrus canker conducted at Lamao experiment station ‘ by the senior author, control methods against Corticitum were ‘also taken up. The orchards selected at Lamao for the citrus- canker experiments were so badly affected with pink disease that control measures were desirable in order that they might be maintained in good condition. In the past, control measures against pink disease have been for the most part based upon the removal of affected tree parts, followed by disinfection. A number of disinfection methods have been practiced, probably the most commonly employed being the application of tar. Van Hall” suggests carbolineum ” Van Hall, C. J. J., Cocoa. Macmillan & Co. Ltd., London (1914) 252. - 664 Philippine Journal of Science 1919 as a disinfectant, and experiments in Malaya with other similar compounds are reported as being successful. Painting with a Bordeaux paste has also been recommended by a number of writers. Butler 2° writes that, in India, the application of Bordeaux mixture, 6—4—45, to the parts of Para rubber trees most sus- ceptible to attack by pink disease has resulted in a reduction of the disease of from 50 to 75 per cent. The mixture was applied with a brush around the forks, for a foot or two down the stem, and up the branches. Two or three applications during the year appear to have produced the most satisfactory results. Butler states that over 200,000 trees were treated at a cost varying from 1 to 2 rupees an acre (about 1.65 to 3.30 pesos per hectare). With reference to the practice of spraying as a preventive and remedial measure for pink disease, Petch ** states that ap- plications of Bordeaux, 5-5-45, against pink disease on cacao resulted favorably. His statement is but a brief mention of fact in connection with another subject, and he does not give his spraying data. Bancroft 22 has also recommended the prac- tice of spraying without, however, giving his experimental data. So far as we know, the results discussed in this paper constitute the first experimental evidence of the value of spraying practices against pink disease. Orchard A at the Lamao station, which consists of various horticultural varieties and species of Citrus, was the most badly affected by pink disease and the control-experiment data and results for this orchard are presented in the following pages. The diagram (fig. 1) shows the position of the trees and the rows in the orchard, and also indicates the location of the cases of pink disease existing in October, 1917. The orchard was divided into plats as follows: Rows 1, 2, 83, 4,5, 6, in Plat I; rows 7, 8, 9, in Plat II; rows 10, 11, 12, in Plat III; rows 13, 14, 15, in Plat IV; rows 16, 17, 18, 19, 20, in Plat V. These plats were treated as follows: * Butler, E. J., Fungi and Disease in Plants. Thacker, Spink & Co., Caleutta (1918) 505. *Petch, T., Physiology and Diseases of Hevea brasiliensis. London (1911) 212. * Bull. Dept. Agr. F. M. S. 1 (1912) 218. * XIV, 6 Lee and Yates: Pink Disease of Citrus 665 TREATMENT OF ORCHARD A, LAMAO EXPERIMENT STATION Plat I, rows 1, 2, 3, 4, 5, and 6. October, 1917: Limbs affected with pink disease pruned out. November 28: Lime sulphur. . January 10: Neutral Bordeaux. May 31, 1918: Lime sulphur (change necessitated by increase of seale insects). June 25, 1918: Neutral Bordeaux, It was the original intention to continue the application of neutral Bordeaux through the entire period of the control experiments, but the increase of scale insects made changes to lime sulphur necessary. Plat II, rows 7, 8, and 9. Neither pruning nor spraying was employed in this plat, the trees consisting of a check on the treated plats. Plat III, rows 10, 11, and 12. October, 1917: Limbs affected with pink disease pruned out. November 28: Formalin 1-100. December 1: Formalin 1-75, December 6: Lime sulphur plus formalin 1-80. January 9, 1918: Neutral Bordeaux plus formalin 1-100. May 31: Lime sulphur. June 24: Lime sulphur. July 29: Lime sulphur. It was originally the intention to spray this plat entirely with lime sulphur; the applications of formalin, however, were employed be- cause of the citrus canker experiments, while the application of neutral Bordeaux was made because no lime sulphur was available. Plat IV, rows 13, 14, and 15. October, 1917: Pruned for Corticium. November 28: Formalin 1-100. December 6: Lime sulphur plus formalin 1-80. January 9, 1918: Neutral Bordeaux. May 31: Lime sulphur. June 26: Bordeaux 4-4—50. The changes from Bordeaux to lime sulphur and formalin were neces- sitated by the increase in scale insects and developments in the citrus canker work. Plat V, rows 16, 17, 18, 19, and 20. The trees in this plat were neither sprayed nor pruned. When pruning is mentioned it means that the trees of the plat indicated were gone over thoroughly to locate each case of the disease. Every affected limb was cut back far enough beyond the extent of the fungus to avoid all possibility of the fungus remaining in the unpruned portions. After each cutting the wound and the pruning implements were disinfected with a 2 per cent formalin solution. Subsequently each wound was painted with white lead. In August, 1918, notes were made of the cases of pink disease 666 Philippine Journal of Science 1919 in the orchard. The diagram, fig. 2, shows the number and lo- cation of diseased trees at that date. DISCUSSION OF RESULTS The text figures (figs. 1 and 2), showing the orchard with distribution of the spray plats, indicate perhaps most clearly the control obtained. To summarize, however: In October, 1917, there were upon Plats I, III, and IV twenty- six trees affected with pink disease. At the same time in the unsprayed plats, II and V, there were fifteen cases of pink disease. ) In August, 1918, the middle of the warm rainy season, no case was found in pruned and sprayed Plats I, III, and IV, the results amounting to effective and complete control. In untreated Plats II and V the fifteen cases of pink disease had increased to eighteen cases, in addition to which two trees, re- corded in 1917 as having pink disease, had been killed as a result of the attacks by the disease. _ The junior writer has observed control attempts on rubber in British North Borneo by pruning and disinfection methods alone. Such attempts appeared to give fairly satisfactory results with- out the accompanying spraying. However, the value of spraying _is demonstrated by several cases observed at the Lamao orchard. In September tree 6, row 10, of this orchard was badly affected and nearly girdled with pink disease on the trunk. Because it was a large, valuable tree the station superintendent did not wish to have it entirely destroyed; and in deference to his wishes the tree was left, although its recovery appeared very improbable. Nevertheless, at every spraying period this lesion was given a strong application of the fungicide. In the last observation, August, 1918, the tree was in fine growing condition, and al- though in the middle of the rainy season the mycelium had not reappeared. A similar experience was had with tree 8, row 11, exhibiting a very bad case of pink disease on the trunk and forks, where spraying resulted in the complete recovery of the tree. Trees in the unsprayed rows that showed similar attacks were killed within a short time. Very similar results on rubber are reported by Rickatds 3 as having been obtained in the Federated Malay States by painting - * Richards, R. M., Diseases of the leaves and stem of Hevea brasiliensis in the Malay Peninsula, Proc. First Agr. Conference Malaya, Kuala Lumpur (April 25, 1917) 45-47. XIV, 6 Lee and Yates: Pink Disease of Citrus 667 Row No.— 1 2 3 4 5 6 7 aes 10.28 $2 ts OAC 16 oot AB tte -e8 = iis. «| ° SOO ee oe ° 62 eS 6 Oi OF er ee ® x ° perlee - a wee, : Paice © ei * x ° x OBS 8 x ° x ee ee Oo) 2-38 ° ° O58 ° os * on ee x S Besee < eer" mein *) ° x x ° 6 x ° ° ° PGS ak « Ss» Ro A ee Meee, See + x x ee. eS A ee ee ee ree ee, ea: tee x ° $256 Glas Ge. Wee ‘‘3ey ooo Mh i EO be, % ° Po gt Bede ee ° x 0 x ° x oe} O. 6e 8 eS eae * oo ° x aie aay ve BAtrag «acs « aeeraae, cates“) } 2 3 ere 6 78 B51 fee bd AS I ME de AF a0 = — cea Ss SS —— wr neat 4 Piat I Plat il Plat iil Platlv Plat V Fie. 1. Orchard A, at Lamao, Bataan Province, Luzon, showing location of Corticium cases in October, 1917. Cireles show location of unaffected trees, Crosses show location of affected trees. ‘ Row No. (foe Be, Bee Oe i Nee ee . O- 6.5 O36 ~ Bee ee 8S 0 2 22 On SB i 65-82 2 OS Re | OO 98 68 ae pee Se O86 so e776 * iT) oe 6s ew <6" Coe Ss ee ° C26 gee! AS EE OOS x : 66S es. geo Oo ore Oe ee ee pee» 6.240 oe Oe > 08 28 39 ad 0°’ o Ooo 16 Oe 8 4 4g. 6. 8 o oO Sa ee 2S se 6. 8 Cote &. te 8 weer S66 be 8 Oe. Se eee Ce oe e6- 6 ee eae eee a 8 Oe ee ee ee ° Beer a ee Ne Ae ee Doe - & x eX , z eee ca ni Plat il Prat lil abit ew Slav "Fic. 1, Orchard A, at Lamao, Bataan Province, Luzon, showing location of Corticium cases in August, 1918. Circles show location of unaffected trees. Crosses show location of affected trees. 668 Philippine Journal of Science 1919 the affected places with tar. Although it does not seem rea- sonable to assume that such applications penetrate and kill the mycelium within the tissues, the report states that control is obtained in most cases. These results are in agreement with the observations made in the Philippines on the superficial ap- plication of fungicides; there, strangely enough, such applica- tions have resulted in apparently complete control. Although spraying is usually not practicable against pink dis- ease on rubber, because of the size of the trees, it would seem to be the simplest method for the control of this disease on Citrus, since the application of fungicides may be combined with control methods against citrus canker, citrus scab, and, when lime sul- phur is used, against scale insects. It seems probable that spray applications in connection with pruning would also prove an effective and simple control of the disease upon cacao, coffee, tea, and similar plants. RECOMMENDATIONS It is apparent that a control approaching almost complete eradication has been attained at Lamao. It seems probable that the fundamental procedure is the careful removal of affected limbs together with precautionary disinfection methods. How- ever, it is also apparent that much can be accomplished by spraying as a preventive measure. Although copper sprays, such as Bordeaux and Burgundy mixtures, have proved successful against pink disease, they have the disadvantage of killing the fungus parasites of scale insects, thus bringing about a great increase in the latter. The use of copper sprays, therefore, is not recommended in the treatment of pink disease on Citrus, because of the great increase of scale insects following their repeated application. For this reason lime sulphur is suggested, not only because it is apparently as efficacious as are the copper sprays against pink disease, but also because it has been found of value against scale insects. Lime sulphur 1 to 40 is suffi- ciently strong, although it has been found that lime sulphur 1 to 35 will cause no injury to the foliage of citrus trees. A dis- cussion of the preparation of lime sulphur is given by Waite.* The time for spray applications varies with the season. For a normal season in central Luzon an application about May 15, another June 15, and another July 15, should be sufficient; that * Waite, M. B., Fungicides, and their use in preventing disease of plants, Bull. U. S. Dept. Agr. 243 (1916). XIV, 6 Lee and Yates: Pink Disease of Citrus 669 - is, at the beginning of the wet season, in the middle of the wet season, and toward its close. Should heavy rains and high hu- midity continue into August, another application in August would be advisable. It should be emphasized that moisture conditions vary greatly in different parts of the Philippines and that the time for spray- ing in different islands would also vary with the periods and degree of rainfall in the various localities. It seems probable that similar control methods would prove of value against pink disease of tea, coffee, cacao, and other crop plants, although we have no experimental data on which to make a definite statement. DISCUSSION OF THE SERIOUSNESS OF PINK DISEASE AND ITS OCCURRENCE IN THE PHILIPPINES Pink disease when uncontrolled causes serious injuries and often results in the death of entire trees and consequent econom- ic losses. Plates V, VI, and VII show some of the effects of the disease. At present it is apparently not widely distributed in the Philippine Islands; but, should control measures not be adopted, conditions in regard to this disease may be expected to develop similar to those existing in the Federated Malay States, where the disease had gained a foothold before its serious- ness was appreciated. Pest gangs are now maintained upon many rubber estates there for the purpose of disinfecting Corti- cium wounds and treating other diseases. It is to be hoped that Filipino planters will profit by experience in other countries, especially as the activities of the disease on Citrus trees at Lamao indicate what may take place if no control methods are attempted. In an area of not more than 12 hectares (30 acres) at least three hundred cases of Corticiwm have been observed. It should be re- membered also that pink disease is a killing disease rather than one that merely takes its percentage from the financial returns by lowering the efficiency of the tree. Although control by spraying and pruning is simple and comparatively inexpensive on Citrus, because of its localized occurrence at present, it seems possible to go further and en- tirely eradicate this disease from the Philippine Islands. Such eradication being effected, enforcement of adequate quarantine regulations should prevent the reintroduction of Corticium sal- monicolor B. & Br., and the continual fight against the disease and the consequent expenditure of money would not need to be 670 Philippine Journal of Science 1919 repeated from year to year. That eradication is possible is shown by the results being obtained in the United States against citrus canker, a disease which is much more infectious and dif- ficult to control. It is particularly desirable to take steps to eradicate this disease before it has spread into privately owned Filipino holdings, where eradication would be extremely difficult. Another question perhaps worthy of consideration is the pos- sibility of the introduction of the disease into the southern United States. Von Speschnew * has reported Corticiwm salmo- nicolor B. & Br. on tea in the Caucasus. More recently Sawada has reported the fungus in Formosa at a latitude much the same as that of lower Florida, and the climate of southern Florida would seem quite as tropical as is that of Formosa. Rant, moreover, has reported Corticiwm on cinchona in Java, at an elevation of 2,000 meters, at which altitude in the Tropics the temperature is very considerably lowered. Therefore, there seems to be the possibility that this fungus would develop readily in southern Florida or even in the other Gulf States, where the summer months are warm, wet, and humid. The necessity of preventing the introduction of this disease into the southern United States by quarantine measures is emphasized even more strongly by the report of Corticiwm salmonicolor B. & Br. in Porto Rico. SUMMARY 1. Pink disease caused by Corticium salmonicolor B. & Br. _ has been observed in the Philippines upon various species and varieties of Citrus. It has been found in only three localities and is apparently a recent introduction. 2. Review of the literature shows it to be widely distributed throughout the oriental Tropics and that it causes serious eco- — nomic losses in various countries on a number of hosts. 3. The appearance of the disease on Citrus is described. Con- trol experiments are discussed which show that removal of affected limbs, accompanied by spraying, results in an effective control approaching entire eradication. The methods are sim- ple and comparatively inexpensive. The results indicate that the commonly used fungicides are equally effective; however, lime sulphur is recommended since it is of value against scale * Yon Speschnew, N. N., Die Pilzparasiten des Theestrauches, Berlin (1907). XIV, 6 Lee and Yates: Pink Disease of Citrus 671 insects also. The control methods employed against pink disease _ combine well in a general spraying plan against other citrus disease. 4. It is probable that these control methods may be equally effective against pink disease on cacao, coffee, tea, and similar plants. 5. Since pink disease is as yet localized, we recommend that strong measures be taken for its complete eradication. Effective measures taken now would probably save the Philippine Islands from considerable financial loss in the future. 6. The possibility of this disease becoming established in the Gulf States of the United States is indicated. ILLUSTRATIONS [From photographs by Lee and by Cortes.] PLATE I Fic. 1. The sterile mycelium stage and pustule stage of Corticium salmoni- color B. & Br. The white advancing mycelium is also shown. Natural size. 2. Corticium salmonicolor in culture upon potato plug; in culture and occurring naturally the color is the most characteristic feature of this fungus in its sterile stages. Natural size. PLATE II Fic. 1. Early stage of pink disease on Citrus, showing gumming and first appearance of pustules. Natural size. 2. Showing the color of the sterile mycelium and advancing hyphae of Corticium salmonicolor. Natural size. 3. Necator stage of the fungus on Citrus twigs. Natural size. 4. Another specimen, showing Necator stage of the fungus on Citrus twigs. Natural size. PLATE IIT Showing early appearance of Corticium salmonicolor on Citrus; the pustule stage. Natural size. PLATE IV Fic. 1. Early stages of surface mycelium of Corticium salmonicolor on Citrus sp. Natural size. 2. A specimen similar to that of fig. 1; also showing gumming. Natural size. PLATE V Effect of pink disease on tree of Citrus maxima in nursery row. PLATE VI Fic. 1. Showing effect of pink disease on tree of Washington navel (Citrus sinensis). 2. The pink disease as it spreads up and down branches; in this case on Washington navel tree. PLATE VII Fic. 1. Tree of Citrus sinensis killed by pink disease. Lamao, Bataan, P. I. August 16, 1918. 2. Another view of orange tree killed by pink disease. Lamao, Ba- taan, P. I. TEXT FIGURES Fic. 1. Orchard A, Lamao, showing location of Corticium cases in October, 1917. 2. Orchard A, Lamao, showing location of Corticium cases in August, 1918. 166157-——8 673 LEE AND YATES: PINK DISEASE.] [Puit. Journ. Scr., XIV, No. 6. PLATE |. CORTICIUM SALMONICOLOR B. & BR. LEE AND YATES: PINK DISEASE. ] {Puiv. Journ. Scr., XIV, No. 6. PLATE II. PINK DISEASE ON CITRUS. LEE AND YATES: PINK DISEASE, | {Puiu. Journ. Sct, XIV, No. 6. PLATE Ill. CORTICIUM SALMONICOLOR ON CITRUS; THE PUSTULE STAGE. NATURAL SIZE. LEE AND YATES: PiNK DISEASE. ] [Pui. Journ. Scr., XIV, No. 6. 9g p z Fig. 1. Early stages of surface mycelium of Corticium salmonicolor on Citrus sp. 2. Similar to fig. 1; also showing gumming. Both figures natural size. PLATE IV. LEE AND YATES: PINK DISEASE.] (Pum. Journ. Scr, XIV, No. 6. t : = / as ql Bees | } PLATE V. EFFECT OF PINK DISEASE ON TREE OF CITRUS MAXIMA IN NURSERY ROW. LEE AND YATES: PINK DISEASE,] [Puit. Journ. Scr., XIV, No. 6. Fig. 1. Pink disease. Fig. 2. Pink disease. PLATE Vi. LEE AND YATES: PINK DISEASE. ] {Puit. Journ. SCI, KEV, No, 6. ja Bas ‘mary, eS Fig. 1. A citrus tree killed by pink disease. Fig. 2. An orange tree killed by pink disease. PLATE VII. ERRATA Page 1, line 21, for Xanthorriza read Xanthorrhiza, Page 3, line 11 from the bottom, for SIMBA read SIAMEA, Page 50, column 2, line 1, Table XXIII, for Intramuscular read Intraven- ous. Page 299, column 2, line 15 from the bottom for Meyerianus read meyen- ianus. Page 300, column 1, lines 25, 26, and 28, for siphonosphathus read sipho- nospathus. Page 300, column 2, lines 1, 3, and 5, for siphonosphathus read siphonos- pathus. Page 326, last line, for cagayensis read cagayanensis. Page 328, lines 5 and 6 from bottom, for cagayensis read cagayanensis. Page 338, line 7 from bottom, for FRUCTICANS read FRUTICANS. Page 347, line 10 from the bottom, for siphonosphathus read siphonospathus. Page 354, line 25, for SIPHONOSPHATHUS read SIPHONOSPATHUS. Page 396, line 7 from the bottom, for zamboanguensis read zamboangensis. Page 487, line 2 from the bottom, for machranta read macrantha. $ be 675 INDEX {New generic and specific names and new combinations are printed in clarendon; synonyms and names of species incidentally mentioned in the text are printed in italic.] A Abaca, 561. Abig6n, 592, Abra River, ipon fisheries of, 127. Abroma augusta Linn., 581. fastuosa Jack., 561, 581-584, 606. Abrome, 581. Abuta, 2. Abutra, 1, 2. Aeab-acab, 351. Acacia, 3. Acal, 353. Acalypha stipulacea Klotz., 534. Acanthaceae, 455. Acarina, 183, 184. Acid, uric, 494. ACOSTA-SISON, HONORIA, and CALDE- RON, FERNANDO, Pelvimetry and cephalometry among Filipino women and newborn babies, 253. Actinobacillus Lign. & Spitz, 60, 64. Actinocladothrix Affan. & Schultz, 58, 63, 64, 67. nocardi, 58. Actinomyce Meyen, 63, 65. horkelii Meyen, 65. Actinomyces Harz, 55, 57-61, 63-66. bovis Harz, 55-59, 64, 66. Actinorhytis Wendl. & Drude, 324. calapparia Wendl. & Drude, 295, 299, 300, 324. Adisca albicans Blume, 244. Adonidia Becc., 295, 298, 329. -merrillii Bece., 299, 329. Adupong, 598. Afulut, 607. Agamid, 628. Agamit, 628. Agandung, 637. Aglaia Lour., 410. | ; diffusifiora Merr., 410. pallida Merr., 411. AGUILAR, R. H., The lumbang-oil in the Philippine Islands, 275. Al-alinau, 617. Albotra, 1, 2. Albustra, 1, 2. Aleurites moluccana, 275, 277. trisperma 275, 282. ; Alibabai, 619. Alibakai, 619. Alimoran, 350. Alinau, 617. industry Alindagon, 637. Alitagtdg, 619. Alkali, 470. Allaeanthus glaber Warb., 619, 620, 621. Allodape, 191. Allophylus Linn., 417. dimorphus Radlk., 417. quinatus Radlk., 417. stenophyllus Merr., 417. Alokon, 619. Alsine media Linn., 382. Alstonia, 544, scholaris R. Br., 544. Alyxia Banks, 448. retusa Merr., 448. Amagéng, 605. Ambong, 336, 581. Ambylanthopsis Mez, 445. crassifolia Merr., 445. Amiyong, 638. Anabion, 637. Anabiong, 637. Anabé, 581. ' Anabong, 581. Anabi, 581. Anacardiaceae, 411. Anamirta cocculus W. & A., 2. Anariong, 637. Anarong, 637. Anayop, 452. Andaluyong, 637, Andropogon Linn., 357. gryllus Linn., 368. gryllus Linn. subsp. genuinus Hack. var. pallidus (R. Br.) Benth., 369. gryllus Linn. var. philippinensis Merr., 368, longipes Hack., 368. philippinensis Merr., 367. pseudograya Steud., 369. sanguineus (Retz.) Merr., 369. sorghum Linn., 290. Anibung, 338. Anilau, 610, 617. Anisosis phalaris 0. S., 186. Anitap, 585. Ankylostomiasis, 486. Annabé, 581, 601. Annadung, 637. Annonaceae, 383, 638. 677 678 Index Areca Linn—Continued. Annona sp., 657, 660. Anoek anoek, 245, Anénang, 631. Anénang-lalaki, 630. Anonongkot, 607. Anophelinae, 174, Anthophoridae, 195. Anthophora Fabr., 195. cengulata Fabr., 195. korotonensis Ckll., 195. luzonica CkIl., 195. whiteheadi Ckll., 195. zonata Linn., 195. zonata stantoni Ckll., 195. zonata whiteheadi Cklil., 195. Antidesma leptocladum Tul., 288, 535. Antipélo, 621. | Antiponong lalaki, 621. Apiean, 338. ipot Becc. var. polillensis Bece., 298, 308, 309. laosensis, 302, macrocalyx, 301. macrocarpa Bece., 296, 298, 301, 302, 306, 308. mammillata Bece., 312. mammillata Beee. var. méindanaoensis Bece., 312. (Mischophloeus) paniculata Scheff., 301. oxycarpa, 301. parens Becc., 296, 298, 301, 303, 307, 308. triandra, 301, 302, 306, 312. vidaliana Becc., 298, 301, 302, 304, 312, 813. whitfordii Becc., 296, 298, 301, 302, 306. whitfordii Bece., var. luzonensis Becc., 298, 302, 306, Apidae, 77, 79. Apis Linn., 79, 80. binghami CkIl., 80. dorsata Fabr., 80. Arecella, 303. Arecinaceae, 296. Arecopsis, 303. Arenga Labill, 295, 297, 335. florea Fabr., 80. indica Fabr., 80. mellifera Linn., 80, nigrocincta Smith, 80. unicolor Latreille, 80. Aplit, 617. Aplonyx, 287, Apnit, 393. Apocynaceae, 248, 448, 544. Aponogeton, 371. Aquilicia sambucina Linn., 245. Araceae, 870, 549. Arachnida, 184, _ Araliaceae, 435, 549. Aramus vociferans, 88. Arandéng, 637. ambong Becc., 299, 335. mindorensis Becce., 299, 335. saccharifera Labill., 295, 297, 299, 300, 335. tremula Becc., 335. tremula Blanco, 299. Argemone mexicana, 1, Aristolochiaceae, 381. Aristolochia Tournef., 381. imbricata Mast., 382. membranacea Merr., 381. sericea Blanco, 382. Arit, 349. Artabotrys R. Br., 383. cagayanensis Merr., 384. monogynus Merr., 383. Areangelisia flava (Linn.) Merr., 1, 3, 5. lemniscata Beec., 1, 5. Ardisia Swartz, 443. ilocana Merr., 443, longipetiolata Merr., 444, perrottetiana A. alee 445. racemoso i a Mez, 444, Areca Linn., 296, 296, 300. caliso Beec., 298, 301, 308, 310. nee Bece., 296, 298, 301, 303, catechu, 296, 300-302, 304-810. eatechu (non Linn.) Bece., 304. eatechu Linn. forma communis, 302, 304. catechu Linn. var, batanensis Bece., 298, 802, 304. eatechu Linn. var. communis, 298. eatechu Linn. var. humilis Blanco, 308. eatechu Linn. var. longicarpa Bece., 296, 298, 301, 302, 305, 306. eatechu Linn. var. silvatica Beec., 296, 298, 301, 302, 304. costulata Bece., 298, 301, 308, 310. glandiformis, 301. aye agen Bece., 298, 301, 302, 804, ivot B Bece., 298, 301, 308, 808-310. suaveolens Blume, 384, ‘Arthrocnodax aphiphila Felt, 292. coprae Felt, 291. Artocarpus communis Forst., 621, 622, 623. Asearis lumbricoides, 207. Asilidae, 182, 186. Asphondylia callicarpae Felt, 536. grewiae Felt, 289, 538. vitea Felt, 587. Astronia Noronha, 434, 531. brachybotrys Merr., 434. pachyphylla Merr., 435. pauciflora Merr., 435. verruculosa Merr., 435. Attai-na-baka, 603. Aulax papaveris Perris, 527. Aungu, 624, Babayan, 619. — Bacille de farcin, 59. Bacillus coli (Esch.) Mig., 132, 136, 137, 148. dysenteriae, 158, 208, 218, 235, 237. leprae Hansen, 13. Index 679 Bacteriaceae, 58. Bacterium Ehrenb., 63. actinocladothrix Affan., 58, 63. Baeg, 619. Bafe nga bunsung, 588. Bago, 581. Bagokon, 617. Bagoong, 129, 612. Baguilumbang, 282. Bakan, 597. Bakembakes, 601. Balala, 352. Balaniug, 325. Balanocarpus, 801, 308, 310. Balanog, 351. | 5 Baléte, 624, 625, 628. Baletéon, 624, Balete-pula, 624. Baléti, 624. Balibago, 616, 637. bulbusin, 587. Balinad, 593. Balitagtag, 619. Balite, 625. Baliti, 624, 627. Baliting-ibon, 624. Balitnong, 616. Balla, 341. Ballang, 340. Baloi, 590. Baluealad, 282. Palucanag, 282. Balulau, 592. Banaba, 597. ‘Banga, 332, 333. Bangar, 595. Bangat, 592. Baniakalaw, 593. Banikad, 593. Banilad, 593, 597. BANKS, CHARLES S., Iontha ida, a new Philippine noctuid, 71; Phlebotomus nicnic, a new species, the first Philip- pine record for this genus, — 163; The bloodsucking insects of the Philippines, 169, ~ Bannakalav,. 593: Bantana, 588. Banucalag, 282. - Baria-an, 616. Barit, 349. Bariw-an, 616. Baroi, 590. Barringtonia luzonensis Rolfe, 536. Basakla, 625. Basbasot, 603. Basilan, Hevea in, 516. Bayau, 592. Bayoég, 590. Payog-bayég, 590. Baydég, 590. Bayong, 590. BECCARI, O., The palms of the Philippine Islands, 295. Bedbugs, 169. Bees, nomadid, 83. obtuse-tongued, 191. Philippine, 77, 83, 191, 195, 555. social, 77. sphecodine, 555. Begoniaceae, 424. Begonia Linn., 424. alvarezii Merr., 427. collisiae Merr., 424. copelandii Merr., 427. lancilimba Merr., 424, longistipula Merr., 428. nigritarum Steud., 427. obtusifolia Merr., 425. panayensis Merr., 428. rubrifolia Merr., 426. serpens Merr., 427. Beilschmiedia, 397. Belostomidae, 185. Belostoma indica St. F. & Serv., 182, 185. BRengalhanf, 611. Benglaleng, 614. Benglaling, 617. Benglareng, 614. Bentinckia, 296. Berberis aetnensis, 1. aquifolium, 1, aristata, 1. asiatica, 1. lycium, 1. vulgaris, 1. Beriberi, 483, 496. human, treatment of, 11. in children, 158. infantile, pseud ingitic, 158. Biau, 277. Bibili, 631. Biknong, 588. © Bilis, 325. Bilisan, 328. Biluan, 588. Binong, 588. Binunga, 588. Bison americanus, 178. Bisong, 598. RBitanag, 588. Bitnong, 588. Bitondg, 588. Bitonog, 588. Biyug, 590. Bleodsucking insects, list of, 184. Blumea DC., 250. chinensis DC., 250. pubigera (Linn.) Merr., 250, riparia DC., 250. Bobo, 595. Bébog, 595. Béboi, 635. Bobor, 595, 635. Bodobodo, 581. Boea sahoepang, 248. Boehmeria Jacq. 381. zollingeriana Wedd., 381. Boeloeng boeloeng pagar, 245. 680 Index Boerlagiodendron Harms, 440. catanduanense Merr., 441. caudatum Merr., 440. Bogtong, 354, Bombacaceae, 635. Bombax ceiba Linn., 635. Bombidae, 77, 80. Bombus Latr., 80. irisanensis Ckll., 81. mearnsi Ashm., 80, 81. rufipes Lep., 81. rufipes melanopoda Ckil., 81 senex Snell., 81. sumatrensis Ckll., 31. ” Bombycidendron vidalianum Merr. & Rolfe, 600. Bonigog, 595. Bongon, 619. Bonotan, 698. Boophilus, 183. Boraginaceae, 620. Botrytis, 56. Boymia glabrifclia Champ., 409. Brachycera, 185. Breaking length of rope, 575. Bremia macrofilum Felt, 290. Buacat, 5. Bubog, 595. Bubui-gubat, 635. Bubur, 595. Bud rot, coconut, cause of, 131. Bug, assassin, 181. electric-light, 182. : giant water, 182. Bugs, water, 182. “ Bulakbuldkan, 605. Bulalog, 414. Bungon, 619. Buntot-usa, 587. Buprestidae, 533. Bute, 1, 2. Butnong, 588. Cc Calamus Linn., 295, 296, 297, 848, 349, 8651, 354. : arugda Becc., 300, 345, 352. bicolor Beee., 348, 856. blancoi Kunth., 299, 343, 349. brevifrons Mart., 350. cumingianus Bece., 299, 344, 351. diepenhorstii, 297. diepenhorstii var. exulans Becc., 299, 300, 344, 350. dimorphacanthus Beec., 300, 355. dimorphucanthus forma typica, 348. dimorphacanthus var. montalbanicus Becc. _ 300, 848, 355. thus var. zambalensis Becc., ~ 300, 348, 355. : Calamus Linn—Continued. discolor Bece., 300. diseolor Mart, 299, 350. discolor Rod. var. negrosensis Becc., 299, 344, 351. elmerianus Becc., 300, 346, 353. filispadix Bece., 299, 344, 350. foxworthyi Becc., 300, 344, 352. gracilis Blanco, 350. grandifolius Bece., 300, 345, 352. haenkeanus Mart., 349. halconensis Bece., 300, 348, 356. hookerianus Becc., 350. horrens ¢ Vidal, 354. inflatus Warb., 355, jenningsianus Becc., 300, 345, 353. lindenti Rod., 351. manillensis H. Wendl., 300, 345, 352. maximus Beec., 351. maximus Blanco, 350. megaphyllus Bece., 300, 346, 353, melanorhynchus Bece., 299, 343, 350. merrillii Beee., 351. merrillii Bece., forma typica, 299, 344. merrillii var. merrittianus Bece., 300, 344, 351. merrillii Bece. var. nanga Bece., 300, 344, merriitianus Bece., 351. meyenianus Schauer, 299, 343, 349. microcarpus Bece., 300, 356. microcarpus forma typica, 348, microcarpus Bece. var. diminutus Becc., 300, 356, microsphaerion Bece., 300, 347, 353. microsphaerion forma typica, 347. microsphaerion Bece. var. spinosior Becc., 300, 347, 354. mindorensis Becc., 300, 345, 352. mitis Bece., 300, 346, 358. mollis Blanco, 299, 348, 348. mollis forma typica, 343, mollis Blanco var. major Bece., 299, 343, 349. mollis Blanco var. palawanicus Becc., 299, 348, 349. moseleyanus Bece., 800, 345, 852. multinervis Bece., 300, 245, 352. ornatus, 297. ernatus Blume var. philippinensis Becc., 299, 800, 344, 850. parvifolius Vidal, 360. ramulosus Becc., 300, 347, 354. reyesianus Becce., 300, 346, 353. samian Beec., 300, 346, 353. simphysipus Mart., 299, 300, 344, 351. siphonospathus Mart., 300, 354. siphonospathus Mart. forma typica Bece., 347, 354. siphonospathus Mart. var. batanensis Bece., 300, 348, 355. : siphonospathus Mart. var. farinosus Bece., 355. siphonospathus Mart. var. oligolepis major Bece., 300, 347, 355. siphonospathus Mart, var. oligolepis minor Bece., 300, 347, 355. Index Calamus Linn—Continued. siphonospathus Mart. var. polylepis Becc., 300, 347, siphonospathus Mart. var. sublaevis Becc., 800, 347, 355. spinifolius Becc., 300, 346, 353. symphysipus, 297. trispermus Becc., 300, 345, 352. usitatus Mart., 349. vidalianus Becc., 300, 347, 354. vinosus Becc., 300, 345, 352. viridissimus Becc., 300, 346, 353. Calcium oxalate, 494. sulphate, effect of, on cement, 221. Caleuli, phosphatic, 491, 494, 496. vesical, 481. Calcutta flax, 611. CALDERON, FERNANDO, see Acosta-SIson and CALDERON. Caliso, 310. Callicarpa Linn., 249, 451. blancot Rolfe, 452. previpetiolata Merr., 249, erioclona Schauer, 536. obtusifolia Merr., 451. rubella Lindl., 249. Calophyllum inophyllum Linn., 544. Calospatha, 296. Camaog, 4. Camarin, 451. CARREON, MARCIANO, The absence of both hind legs below the femur in a full-term pig, 201. Carteria (Carterii, sic !) Musgrave & Clegg, 61, 64, Caryophyllaceae, 382. Caryota Linn., 295, 297, 336. cumingii Lodd., 299, 336, 337, 338. majestica Lind., 299, 336, 337, 338. merrillii Becc., 299, 336, 337, 338. mitis Lour., 295, 299, 300, 336, 337, 338. palindan Blanco, 332. rumphiana Mart., 295, 300, 336. rumphiana (vix Mart.) Beec., 337. rumphiana Mart. var. oxyodonta Becce., 299, 336, 237. rumphiana Mart. var. Bece., 297, 299, 336, 337. tremula Blanco, 335. urens (non Linn.) Blanco, 337, 338. urens (non Linn.) foliis angustioribus Blanco, 338. Casopanguil, 249. Cassia florida, 3. siamea Lam., 1, 3, 4, 6. Cataract, operation for, 159. Caulophyllum thalictroides, 1. Caustic soda, 470. Cecidomyia philippinensis Felt, 294. Celastraceae, 244, 542. Celastrus alpestris Blume, 244. Cement, calcium sulphate on, effeet of, 221, 227. exposure to the air, effect of, 227. hydrated lime in, effect of, 227. philippinensis 681 Cephalometry among Filipino women and newborn babies, 253. Ceratina humilior CkIl., 557. philip pinensis nigrolateralis humilior Ckil., 557. Ceratolobus, 296. Chanvre de Calcutta, 611. Chaoborinae, 174. Chariessa Miq., 415, 416. Chaudhuri, C., notice of his Modern Chemis- try and Chemical Industry of Starch and Cellulose, 153. Chelidonium majus, 1. Chinchorros para ipon, 128. Chironomidae, 174, 178, 186. Cholera, bacteriologic phases of carriers of, 459. effect of bile on vibrio of, 462. faeces from carriers of, 460. nonagglutinable vibrio of, 461. treatment of carriers of, 463. Chrysopogon gryllus Trin., 368. Chrysops, 175. signifer Walk., 185. Cimicidae, 184. Cimex lectularius Linn., 169, 184. Cipadessa Blume, 244. baeccifera (Roth) Miq., 244. Cissus adnata Wall. var., 587. - trifolia (L.) K. Sch., 537. Citriobatus A. Cunningham, 365, 401. javanicus Boerl. & Koord., 401. Citrus aurantifolia Swingle, 662. aurantium Linn., 662. decumana Linn., 662. hybrid, 662. hystrix DC., 662. limonia Osbeck, 662. maxima Merr., 662, 673. medica Linn. var. sarcodactylis, 662. micrantha Wester, 662. nobilis Lour., 662. sinensis Osbeck, 662, 673. webberi Wester var. montana Wester, 662. Citrus, pink disease of, 657. Cladothrix Cohn, 56-59, 62-64. bovis (Harz), 59. eanis Rabe., 57. dichotoma, 56, 58. foersteri (Cohn) Schroter, 59. (Streptothrix) foersteri Cohn, 64._ Clerodendron Linn., 249. intermedium Cham., 249. paniculatum Linn., 249. Clethraceae, 248. Clethra Linn., 248. sumatrana J. J. Sm., 248. Cloézia Brongn. & Gris, 365, 429. urdanetensis (Elm.) Merr., 429. Coeculus palmatus, 1. COCKERELL, T. D. A., The social bees of the Philippine Islands, 77; The Philippine bees of the family Nomadidae, 83; The Prosopidae, or obtuse-tongued bees, of 682 COCKERELL, T. D. A.—Continued. the Philippine Islands, 191; The Philip- pine bees of the families Anthophoridae and Melectidae, 195; The sphecodine bees of the Philippine Islands, 555. Coconut, bud rot of, 131. Cocos Linn., 339. nucifera Linn., 299, 300, 339. Coeloeline polycarpa, 1. Coffea sp., 657. Cohnistreptothrix Pinoy, 61, 62. Cola, 4. Coleoptera, 533. Columbia blancoi Rolfe, 610, 611. Colymbus cornutus, 99, 100. Commersonia bartramia (Linn.) Merr., 586, 587. Compositae, 250, 456, 543. Cone-nose, red-banded, 180. Conjugate, diagonal, determination of, 254. Connaraceae, 403, Connarus Linn., 403. castaneus Merr., 403. Conorhinus megistus Burm., 180. rubrofasciatus de Geer, 180, 184. sanguisugus Le Conte, 180. Contarinia saltata Felt, 289. sorghicola Coq., 290. Convolvulaceae, 450. Convolvulus brownii Spreng., 450. robertianus Spreng., 450. . Conyza chinensis Linn., 250. pubigera Linn., 250. riparia Blume, 250. Coptis teeta, 1. trifolia, 1. . Corchorus capsularis Linn., 612. olitorius Linn., 604, 608, 609, 611, 612, 613, 614. Cordia eumingiana Vid., 566, 581, 630, 631. myxa Linn., 681, 632, 633. Coriaria intermedia Mats., 1, 5, 6. Corticium caleeum Fr., 6538. javanicum Zimm., 658, 659. lilacinofuscum B. & C., 659. salmonicolor Berk. & Br., 657-660, 662, 669, 670, 673. zimmermanni Sacc. & Syd., 658. Corypha Linn., 341. elata Roxb., 295, 299, 300, 341, 342. gebanga BI., 341. umbraculifera (non Linn.) Vidal, 341. umbraculifera var. cubang F.-Vill., 341. umbraculifera var. sylvestris F.-Vill., 341. Coscinium fenestratum, 1. ; Crataegus glabra Thunb., 403. Crests, measurement of, 254. Crocisa Jurine, 196. caelestina Ckll., 196, caeruleifrons Ckll., 198. eaeruleifrons Kirby, 198. callura CkIl., 198, 199. crucifera Ckil., 196, 198, decora Smith, 198, 199. emarginata Lepel., 196. 585, Index Crocisa Jurine—Continued. irisana Ckll., 196. kalidupana Ckil., 196, 197. lamprosoma Boisd., 196. luzonensis Ckll., 196. nitidula Fabr., 196, 197. omissa CkIl., 198, quartinae Ckll., 198. quartinae Gribodo, 197, 198. reducta Ckil., 199. ridleyi Ckll., 199. rostrata Friese, 197. Cryptocarya R. Br., 395. densiflora Blume, 397. elliptifolia Merr., 396, euphlebia Merr., 396. intermedia Elm., 396. - laevigata Elm., 397. oblongata Merr., 396. ramosii Merr., 395. vidalii Merr., 396. ‘ zgamboangensis Merr., 396. Cryptocoryne Fisch., 365, 370. aponogetifolia Merr., 370. auriculata Engl., 371. ciliata Fisch., 371. pygmaea Merr., 371. Ctenocephalus canis Curtis, 185. felis Bouché, 179, 185. Ctenodactylomyia Felt, 287. antidesmae Felt, 287, 535. Culex fatigans Wied., 174. ludlowi Blanch., 174. microannulatus Theob., 174. Culicidae, 174, 186. Culicinae, 174. Culicoides judicandus Bezzi, 178, 186. Cullet, 465, 471. Curacling, 353. Cyanophyceae, 65. Cylicodaphne leefeana F. Muell., 242. Cymbospatha, 356. Cynips, 528. Cyperaceae, 369. Cyperus kyllingiaeoides Vahl, 370. Cyrtandra Forster, 452. agusanensis Elm., 455. glabra Krinzl., 452. glabrifolia Merr., 452. hypochrysoides Kriinzl., 454. . ilocana Merr., 452, lancifolia Merr., 454. livida Krinzl., 455. panayensis Merr., 452. santosii Merr., 453. tayabensis Elm., 453. Cyrtostachys, 296. Cystitis, chronic, 485. Cystolithiasis among Filipinos in association with dietetic deficiency, 481. Cystolithotomy, 484, 485, 486. — D Daanan, 343. Daclis de ipon, 128. = Index Daemonorops Blume, 356. affinis Bece., 298, 300, 358, 359. clemensianus Becc., 300, 357, 359. curranii Beec., 300 358, 359. gaudichaudii Mart., 298, 300, 357, 359. gracilis Becc., 298, 300, 358, 359. hystrix, 298. loherianus Becce., 300, 357, 358. longipes, 298. manillensis Mart., 352. margaritae, 298. margaritae Beee. var. palawanicus Becc., 800, 356, 358. ochrolepis Becce., 298, 300, 357-359. ochrolepis Bece. var. radulosus Bece., 358. oligolepis Bece., 300, 357, 359. pannosus Becce., 300, 357, 359. pedicellaris Becc., 300, 357, 358. rumphii, 298. urdanetanus. Becc., 300, 357, 358. virescens Becc., 298, 800, 356, 358. Dalakit, 625. Dal-dal-upang, 605. Dallag, 617. Dalunit, 637. Dalunot, 637. maladurung, 637. Dalupang, 607. Damalina semperi O. S., 186. Dangli, 617. Danglin, 617. aso, 587. kalabau, 587. Dangling, 617. Danglog, 618. DAR JUAN, T., and ELICANO, V., Philippine raw materials for glass making, 465. Debregeasia Gaud., 241. longifolia (Burm. f.) Wedd., 241. Deer flies, 175. Deficiency, dietetic, 481. metabolic, 482. DE LEON, WALFRIDO, see HAUGHWoOUT and DE LEON. Dendrocalamus merrillianus, 566. DENNEY, OSWALD E., A_ photographic study of leprosy, 13. Dermacentor variabilis Say, 183, 184, Desmodium Desv., 243, 405, 409. dependens Blum®, 409. lasiocarpum (Beauv.) DC., 243. latifolium DC., 243. ormocarpoides DC., 248, 409. ormocarpoides Desv., 409. virgatum Zoll., 243. zonatum Mia., 243, 409. ( Diameter, Baudelocque’s, measurement of, 254, intertuberal, relation to second stage of labor, 263. ee . i : : ‘Dibodl, 600.0. ae ae - Diceromyia vernoniae Felt, sis. ; Dichapetalum obovatum Elm., 418. — Didinium nasutum, 214. — Didymosperma tremulum Wendl., 886. 683 Dillenia Linn., 421. megalophylia Merr., 421. papyracea Merr., 422. reifferscheidia F.-Vill., 547. suffruticosa Grieff., 422. Dinopsylla cornuta Crawf., 545. Diospyros Linn., 248. wallichii King & Gamble, 248. Diplodiscus paniculatus Turez., 538. Diplycosia Blume, 442, calelanensis Elm., 443. glabra Merr., 442, heterophylla Blume, 443. Diptera, 185, 533. Dipterocarpaceae, 541. Discomycetaceae, 62, 67. Discomyces Rivolta, 57-60, 62, 64, 66, 67. bovis (Harz) Rivolta, 57, 60, 64, 66, 67. equi Rivolta & Miscellone, 57. pleuriticus canis familiaris Rivolta, 57. (Streptothrix) indica, 60. Diver, 182. Djaring, 243. Djaroem djaroem, 250. Djering, 243. Dédol, 635. Dongrareng, 614, Dugui-is, 443. Duidui, 592. Dumayaca, 328, 335 Durarong, 618. Duyduy, 592. Dysentery, clinical observation on cases of, 167. ingestion of erythrocytes by Pentatricho- monas sp., in a case of, 207, E Ebenaceae, 248. Ectomocoris atrox Stal, 181, 184. biguttulus Stal, 181, 184, flavomaculatus Stal, 181, 184. Elaeocarpaceae, 418. ; Elaeocarpus Linn., 418. cumingii Turez., 419. forbesti Merr., 419. -ilocanus Merr., 418, Elatostema Forster, 372. acumatissimum Merr., 375, angustatum C. B. Rob., 375. appendiculatum Merr., 379, benguetense C. B. Rob., 376. carinoi W. R. Shaw, 374, 379, 380. contiguum C. B. Rob., 375. diversilimbum Merr., 378. elmeri Merr., 373. _ filicaule C. B. Rob., 374. _ haleonense C. B. Rob., 377. macgregorii Merr., 373. obovatifolia Wedd., 373. — obovatum Wedd., 378. panayense Merr., 372, _ pictum Elm., non Hallier f., 373. pilosum Merr., 376, robinsonii Merr., 376, _ suborbiculare Merr., 377. zamboangense Merr., 374, 684 Elatostematoides C. B. Rob., 380. manillense C. B. Rob., 381. samarense Merr., 380. Eleocharis R. Br., 370. spiralis (Rottb.) Steud., 370. Elevation, relation of, to Hevea, 512. ELICANO, V., see Dag JUAN and ELIcANo. Elongation of rope, 574. Entamoeba coli, 212, 215, 216. histolytica, 158, 209, 210, 213-215. Epholchiolaphria aurifascies Macq., 186. leucoprocta Wied., 186. partialis Bezzi, 186, vuleanus Wied., 186. Ericaceae, 248, 441. Eriophyidae, 528. Erodium L’Hérit., 365, 409. cicutarium (Linn.) L’Hérit., 409. Erythrocytes, ingestion of, by Pentatricho- monas, 207, Erythropalum Blume, 242. scandens Blume, 242, Eskobilla, 603. Eskuba, 603. Eskubilla, 608. Euareca, 302. Eugeissonia, 296. Eulyes illustris Stal, 181, 184. Euphorbiaceae, 244, 534, 535, 540, 546, 548, 549. Eurycoma dubia Elm., 409. Evodia Forster, 409. glauea, 1. lunur-ankenda (Gaertn.) Merr., 410. meliaefolia (Hance) Benth., 409. semecarpifolia Merr., 410. sessilifoliola Merr., 410, F Fabrication of rope, 565, Farcin du boeuf, 60. Feldspar, 477. FELT, E. P., New Philippine gall midges, 287. Feronia elephantum Corr., 662. Festuca Linn., 365, 369, myuros Linn., 369. Ficus benjamina Linn., 624. forstenii Miq., 625, 627. nervosa Hey., 545. pachyphylla Merr., 627. palawanensis Merr., 628. ulmifolia. Lam., 531, 545, 546, 547. variegata Blume, 546. Fil de jute, 611. Firmiana colorata (Roxb.) R. Br., - -marsigli, 246. Fisheries, ipon, 127, Flacourtiaceae, 246, 423. Flea, human, 180, plague, 180. Fleas, 179, cat, 179, dog, 179. 246. Index Flies, bat, 173. black, 174. deer, 175. forest, 175. moth, 174, 175. robber, 182, true, 174, 176. Fly, cattle, 177. tsetse, 177. Fractures, demonstration of important cases of, 156. Fulica, 88-92, 94-96, 98. americana, 88, 96. Fungi, a genus of, validity of name of, 55. Fungus, dissemination of, 663. G Galiuyguiuan, 409. Gallinula, 94-96, 98. chloropus, 87. Gallinule, Philippine giant, osteology of, 87. Gall midges, 287. Galls, methods of collecting, 529. plant, Philippine, 527. Gambir gambir, 243. Garcinia venulosa (Blanco) Choisy, 547. Gekko kikuchii, 118. mindorensis Taylor, 105, 115, 118. monarchus, 117, 118. Gelechidae, 549. Gemiisepappel, 611. Geniostoma Forster, 448. pachyphyllum Merr., 448. Geodorum nutans Ames, 1, 4, 6. Geoffrayra inermis, 1. Geraniaceae, 409. Geranium cicutarium Linn., 409. Gesneriaceae, 452. GIFFEN, A. M., review of Keen’s The Treat- ment of War Wounds, 363. Gigantothrips elegans Zimmerman, 547. Gigliolia, 296. Girth of rope, 569. Gisi, 624. Glass, furnace for, 471. lead, 466. lime, 466. : raw materials for, 465. Glochidion album (Blanco) Boerl., 549. Glossina sp., 177, Se Gnats, buffalo, 178. Gneiss, 468. Gnetaceae, 633. Gnetum sp., 582, 633. . Gold, colloidal, in treatment of typhoid fever, 42, Goniothalamus Hook. f. & Th., 385. amuyon Blanco., 638. lancifolius Merr., 385. longistylus Merr., 386. malayanus Hook. f. & Th,,. 386. Gramineae, 357. Grewia bilamellata Gagnep., 614, 615. eriocarpa Juss., 616, 617. multiflora Juss., 614, 617-619. stylocarpa Warb., 289, 538, Index Guioa Cav., 417. obtusa Merr., 418. parvifoliola Merr., 417. Gulick, Walter Vose, notice of his Mental Diseases, a Handbook Dealing with Diagnosis and Classification, 251. Guima, 631. Gunny, 611. Guttiferae, 544, 54 Gynaikothrips chavicae Zimmerman, 548, chavicae var. heptapleuri Karny, 549. H Haematocarpus Miers, 365, 383. comptus Miers, 383. subpeltatus Merr., 383. Haematopinus eurygaster Nitzsch, 184. eurysternus Nitzsch, 172. spinulosus Denny, 172. spinulosus Nitzsch, 184. stenopsis Burm., 172. stenopsis Nitzsch, 184. tubereulatus Nitzsch, 171, 184. urius Nitzsch, 172, 184. Hagod, 637. Halictus, 88, 555. Hamitanago, 588. Hamlis, 351. Hanadiéng, 637. Hanagdéng, 637. Hanag-ding, 637. Hantak, 597. HAUGHWOUT, FRANK G., and DE LEON, WALFRIDO, On the ingestion of ery- throcytes by Pentatrichomonas sp., found in a ease of dysentery, 207. Heart, beriberi, in children, 158. Hedysarum adhaerens Poir., 409. lasiocarpum Beauv., 243. Helicteres hirsuta Lour., 578, 587, 588. Heliodiplosis spatholobi Felt, 542. Hemidactylus depressus, 113. luzonensis Taylor, 112. Hemigraphis Linn., 455. fruticulosa C. B. Clarke, 456. nummularifolia Merr., 455. reptans Nees, 456. Hemiptera-Aptera, 184, Hemiptera—Heteroptera, 184. Hemiptera—Homoptera, 185. Hernandiaceae, 539. Heterocera, 71. Heterospathe Scheff., 295, 298, 325, 329. cagayensis Becc., 328. elata, 295, 299, 300, $25, $28, 329. elmerii Becc., 328. negrosensis Bece., 299, 325, 826. philippinensis Bece., 299, 325, 826. sibuyanensis Becc., 299, 325. Hevea, 141, 142, 148, 149, 658. brasiliensis (HBK) Muell., 501, 5238, 657, 660. Hexamastix ardindelteili Derrieu & Raynaud, ba bag psi see Hexanthus umbellatus Lour., 242. 512, 516, | 685 Hibiscus Linn., 245. macrophyllus Roxb., 240. Hinagding, 637. Hinla-laong, 6387. Hipon, 127, Hippobosca equina Linn,, 185. maculata Leach, 185. Homalium Jacq., 423. multiflorum Merr., 423. ramosii Merr., 423, samarense Merr., 423. Homo sapiens Linn., 172. Horseflies, 174, 175. Howell, William H., notice of his A Text-book of Physiology for Medical Students and Physicians, 559. Hubulos, 637. Hulecoeteomyia pseudotaeniata Giles, 174. Huligano, 592. Hunung, 588. Hydrastis canadensis, 1. Hydrophasianus, 98, 101. echirurgus (Scop.), 98, 101, 103, 104. Hyperdiplosis banksi Felt, 537. Hyphomycetes, 59. I Ieacinaceae, 414. Iguanura, 295. Iliigera luzonensis (Presl) Merr., 539. Imbubuykon, 618, Imkabau, 619. Inangdén, 637. Indai-luging, 637. Indian grass, 611. Indigofera Linn., 405. endecaphylla Baker, 405. hendecaphylla Jacq., 405. pusilla Lam., 405. | Indischer Flachs, 611. Inga jiringa Jack, 243. Insecta, 184. Insects, bloodsucking, 169. Jontha Doubleday, 71. ida Banks, 71. ‘ umbrina Doubleday, 72, 73. Ipomoea Linn., 450, heterophylla R. Br., 450. polymorpha Roem. & Schultes, 366, 450. Ipon, 127. : : Itangan, 403. Itapan, 410. Itonida paederiae Felt, 293, 540. | Ixodidae, 184. Ixodoidea, 184. a Jacana spinosa (Linn.), 101. Jateorhiza palmata, 1. Jeffersonia diphylla, 1. Jew’s mallow, 611. JOHNSTON, JOHN A., Some bacteriologic phases of the cholera-carrier problem, 459. Jordan, Edwin O., review of his A Text-book of General Bacteriology, 363. Judenhanf, 611. 686 Judenpappel, 611. Jusi, 354. Jute, 611. Kabag, 619. Kakaab, 58T. Kakaag, 585, 587. Kakaomalve, 581. Kaliat, 633. Kalitkalit, 617. Kaikuttanhanf, 611. Kalukalumpangan, 593. Kalumpang, 595. . Kamansi, 621. Kamptodiplosis reducta Felt, 542. Kanaroset, 618. Kandiat, 633. Kapas, 635. Kapas-kapas, 605. Kap6i, 635. Karud, 619. Kastule, 603, 605. ™ Kastuli, 605. Kaupp, B. F., notice of his The Anatomy of the Domestie Fowl, 251. Keddéng, 610. Keen, W. W., review of his The Treatment of War Wounds, 363. Kerbaka, 603. KING, ALBERT E. W., The mechanical prop- erties of Philippine bast-fiber ropes, 561. Kleinhovia hospita Linn., 566, 581, 588-590. Koendoelen pamal, 244, Kolis, 624. Kol-lokoliét, 607. Kol-lokol-lot ti bao, 587. Kollolét, 607. Kolot-kolétan, 607. Korthalsia Blume, 295, 297, 342. laciniosa Mart., 295, 298-300, 342, 343. merrillii Bece., 298, 299, 342. scaphigeroides Beec., 298, 299, 342. squarrosa Becc., 299, 342, 343. Kronodiplosis uichancoi Felt, 536. Kubal, 587. Kulat-kulat, 607. Kuliamat, 624. Kuliat, 6338. Kullu-kullik, 607. Kuluk, 607. Kulu-kulét, 607. Kulut-kulitan, 607. Kungdén, 687. ‘ Labnai, 598. Labon, 581. Lacaubi, 342. Laccotrephes robustus Stal, 182, 185. Laemopsylla cheopis Rothsch., 180, 185, Laginlaginan, 587. Lagod, 687. Lagtal, 1. — Lagtan, 1. Lagtang, 1. : Lailaiginan, 587. : Sera agy s -Lamai, 687. Sa Index Langaban, 625. | Langosig, 618. | Lanit, 616. Lanétan, 600. LANTIN, PEDRO T., A comparative study of different methods of treatment of ty- phoid fever, 19. Lanut, 618. Lanitan, 600, 638. Laoe gappa gappa, 249. Laphria dimidiata Macq., 185. Lapi, 616. Lapni, 616. Lapnis, 588, 618. Lapnit, 597, 616. Lasianthus Jack, 250. oblongus King & Gamble, 250. rhinocerotis Blume, 250. Lasioptera falcata Felt, 288. manilensis Felt, 539. Lassee, 356. Lauraceae, 242, 395. Lead glass, 466. Leaf hoppers, 181. Lecythidaceae, 536. LEE, H. ATHERTON, and YATES, HARRY S., Pink disease of citrus, 657. Leea Royen, 245. ‘indica (Burm. f.) Merr., 245. manillensis Walp., 539. sambucina Willd., 245. sambucina (Linn,) Willd., 245. Leguminosae, 243, 405. Leiolepisma, 119. Leontice thalictroides, 1. Lepidocaryeae, 296, 297. Lepidodactylus christiani Taylor, 115. naujanensis Taylor, 105, 113. Lepidoptera, 71. Leprosy, study of, 13. Leptogaster princeps O. S., 186. Leptothrix buccalis, 56. Leptus akamushi Brumpt, 183. Liba, 619. Libtik, 592. Lice, 170. Licuala Rumph., 295, 339. spinosa Wurmb, 295, 296, 299, 300, 339. Ligtang, 1. Lime, 466, — glass, 466. Limestone, coralline, 466. crystalline, 466. Limuran, 350. Litsea Lam., 242. amara Blume, 242. ferruginea Blume, 242. ferruginea (R. Br.) Benth. &*Hook., 242. hexentha Juss., 242. leefeana (F. Muell.) Merr., 242. perakensis Gamble, 243. ae umbellata (Lour.) Merr., = ® Livistona R. Br,, 295, 339. cochinchinensis Mart., 295, 907: 299, 200, ——- 889, 840. : inaequisecta Beee., 340. Index Livistona—Continued. merrillii Bece., 297, 299, 839, 340. microcarpa Bece., 341. mindorensis Becce., 341. robinsoniana Becc., 297, 299, 340, 341. rotundifolia Mart., 295,.297, 299, 300, 339, 340. rotundifolia Mart. var. luzonensis Becc., 299, 340. rotundifolia Mart. var. microcarpa Bece., 340, 341. rotundifolia Mart. var. mindorensis Becc., 299, 340, 341. spectabilis Griff., 340. vidalit Bece., 341. whitfordii Beec., 340. Lizards, Philippine, 105, 112. Loganiaceae, 448. Losdéban, 600. Louse, body, 171. carabao, 171. crab, 170. goat, 172. hog, 172. monkey, 172. rat, 172. Lubang Island, glass materials in, 465. Lubi lubi, 4 Lumbang, 277, 282. banucalag, 282. bato, 277. gubat, 282. oil in the Philippines, 275. Lumbia, 342. Lunug, 627. Luséban, 600. Lustiban, 600. Luzonomyia symphoremae Felt, 543. Lydston, G. Frank, notice of his Impotence and Sterility with Aberrations of the Sexual Function and Sex-gland Implan- tation, 153. Lyperosia exigua de Meij., 177, 186. minuta Bezzi, 178. M Madura disease, 59. Magindanau, 588. Makakapas, 695. Malaampipi, 416. _ Malabanilad, 597. Malabitispapa, 601. Malabého, 597. Malabilak, 635. Malabinga, 597. Malachra fasciata Jack, 581, eer-08. Malagaséha, 598. . : . Malakadiés, G19. 687 Malasikongdiron, 637. Malatakon, 587, Malibago, 588. Mallory, Frank Burr, and Wright, James Homer, notice of their Pathological Technique, A Practical Manual for Workers in Pathological Histology and Bacteriology including Directions for the Performance of Autopsies and for Clinical Diagnosis by Laboratory Meth- ods, 559. Mallotus moluccana (Linn.) Muell.-Arg., 540. philippensis (Lam.) Muell.-Arg., 546, 548. Malobago, 588. Malvaceae, 245, 561, 600. Mamadling, 610, Mamauéd, 610. Mangkit, 607. Manila Medical Society, Proceedings of the, 155, 157, 159. Mansonioides annulifera Theob., 174. uniformis Theob., 174. Marakapas, 681, 588, 592, 605, Maratakkimbaka, 603. Maratarong, 581, 605, 630. Margaropus australis Fuller, 183, 194. Marighoy, 328. Mariscus Gaert., 369. - dregeanus Kunth, 370. glomeratus Barton, 369. niveus (Murr.) Merr., 369, Mauve de juif, 611, 612. Measurements of Hevea trees in Basilan, 516. Medinilla Gaud., 431. hasseltii Blume, 432. oblanceolata Merr., 431, Megabotrya meliaefolia Hance, 409. Megapis Ashm., 79, 80. Megarhininae, 174. Megatrioza pallida Uichanco, 546. Melastomataceae, 247, 431, 540. Melectidae, 195. Meliaceae, 244, 410. Melia baccifera Roth, 244. ‘Meliponidae, 77. Melochia, 611. Meluchia, 611. Memecylon Linn., 247, 432. appendiculatum Blume, 248. costatum Miq,, 248. : laruei Merr., 247. paniculatum Jack, 248, 540, ramosii Merr., 432, stenophyllum Merr., 432. Menispermaceae, 388, 416. Menispermum canadense, 1. MERRILL, E. D., Notes on the flora of Su- -matra, 239; New or noteworthy Phil- -ippine plants, XV, 365. | MERRILL, E. D., and WADE, H. W., The validity of the name Discomyces for the * genus of fungi variously called Actin- omyces, Streptothrix, and Nocardia, 55. | Methylene blue-eosin lactose agar, use of, in searching for Bacillus planer af in stools, 235. 688 Metroxylon Rottb., 342. rumphii Mart., 295, 297, 299, 300, 342. Mica, 467, 474, Micrapis Ashm., 79, 80. is Micrococcus ascoformans Johne, 57. botryogenes Rabe., 57. Micromyces Griiber, 63. hoffmanni Griiber, 58, 63. Micronecta quadristrigata Bredd., 182, 185. Midges, 174, 178. d gall, 287. Milk in treatment of typhoid fever, 39. Mimosa Linn., 243. djiringa Roxb., 244. invisa Mart., 243. jiringa Jack, 243. keoringa Roxb., 244. Miquelia Meissner, 416. cumingii, 417. philippinensis Merr., 416. Mitrephora Hook. f. & Th., 387. cagayanensis Merr., 387. Mono, 310. Monoesporium, 56. Moraceae, 545-547, 561, 619. Morton, H. H., notice of his Genitourinary Diseases and Syphilis, 153. Mosquitoes, 174. Mosquito, elephant, 175 rock, 174, small woolly, 174, white-capped, 174. Murraya exotica Linn., 659, 662. Muscidae, 174, 176, 186. Musea domestica Linn., 176. -Muskraut, 611. Mycobacterium, 59. leprae, 13. Mydaidae, 182. : Mydas fruhstoferi v. d. Wulp, 186. Myiophthiria reduvioides Rond., 186. Myrialepis, 296. Myrsinaceae, 443. Myrtaceae, 429. Myzomyia febrifera Banks, 174. rossi Giles, 174. N Nabé, 581, Nabu, 581, Nagdén, 637, Nagling, 618. Nardurus filiformis Steud. Franch., 369. Necator decretus Massee, 658, 659, 663. Necrobia rufipes de Geer, 292. Negegan, 581. Nenga, 296. Neoheegeria mendax Karny, 548. Neolitsea Merr., 397. cassiaefolia Merr., 399. lanceolata Merr., 397. paucinervia Merr., 398. villosa Merr., 399. zeylanica Merr., 399. var. chinensis Index Nephotettix, 182. apicalis Motsch, 181. bipunetatus Fabr., 181. Nepidae, 185. Neutral urine, 492. Nienic, 175. Niog-niogan, 3825. Nipa Rumphius, 338. fruti¢ans Wurmb, 78, 295, 299, 300, 338. Nitidula, 197. Nocardia Trev., 55, 58-61, 63-67. Noctuidae, 71. : Noctuid, new species of, 71. Nomada Scop., 83. adusta Smith, 86. attrita CkIl, 83, 84, bakeri Ckll., 83, 85, 86. banahaonis Ckll., 83, 85, concessa Ckll., 83, 85. exheredans Ckll., 83, 85. lusea Smith, 85, 86. makilingensis Ckll., 88, 84. mindanaonis CkIl., 83-85. palavanica CkIll., 88, 84. pervasor Ckll., 88, 84. subpetiolata Smith, 86. Normanbyae sp. Beece., 329. Normanbya, 329. merrillii Becce., 329. muellerii, 329. Notonectidae, 185. . Notornis, 87, 88. Oo Cbstruction, ureteral, 156. Oeboeng oeboeng, 245, Oil, lumbang, 275. Olacacease, 242. Olfersia nigrita Speis., 185. Glus judaicum, 612. Ommatius chinensis Fabr., 185. Oncocarpus A, Gray, 413. densiflora Merr., 413. obovata (Elm.) Merr., 413. Oncosperma Blume, 295, 330. filamentosa Blume, 295, 297, 299, 300, 330. gracilipes Bece., 297, 299, 330. horrida Scheff., 295, 297, 299, 300, 330. platyphylla Beee., 297, 299, 330. Oospora Wallroth, 59, 61, 63. indica Kanthack, 59. Operations, mastoid, 160. Orania Blume, 295, 297, 331. decipiens Bece., 299, 311-333. decipiens Becc. var, mindanaoensis Bece., 299, 332, 333. decipiens Bece. var. montana Becc., 299, 382, $83. : macrocladus, 297, 331, 333. molueeana, 297. palindan Merr., 382, paraguaensis Becc., 299, 331, 332. philippinensis Scheff., 299, 231-234. philippinensis Scheff. var. sibuyanensis Becce., 299, 331, 332. es ie Index Orania—Continued. regalis, 297, 331. rubiginosa Becc., 299, 333, 334. Oreocnide Mig., 241. nivea Merr., 241. Oring-oring, 329. Ormittroctona nigricans Leach, 185. Ormosia Jackson, 405. basilanensis Merr., 407, 408. clementis Merr., 406. grandifolia Merr., 407. macrodisca Baker, 406, 408. orbiculata Merr., 405, 407, 408. paniculata Merr., 406. Orthoclase, 467. Orthogonis scapularis Wied., 186. Osmelia Thwaites, 246, bartlettii Merr., 246. maingayi King, 247. Osteology of Philippine birds, 87. Otanthera Blume, 433, strigosa Merr., 433, Outlet, pelvic, 263, Oxymitra Hook. f. & Th., 388. longiflora Merr., 388, 889. multinervia Merr., 388. obtusifolia Elm., 388, 389. platyphylla Merr., 388, bg Paang-baliwis, 601. Paathanf, 611. PADUA, REGINO G., Cystolithiasis among Filipinos in association with dietetic deficiency, 481. Paederia tomentosa Blume, 294, 540. Pakak, 621. Pakalkal, 581. Pakin bakit, 587. Palak-palak, 593. Palaquium Blanco, 446. dubardii Elm., 446. glabrifolium Merr., 446 Palasan, 351. Palindan Blanco, 3382. Palms, Philippine, 295, Pampar, 588. Panabao, 341. Panampat, 588. Panapotien, 406. Pandanaceae, 366. Pandanus Linn., 366. camarinensis Merr., 357. esculentus Martelli, 357. panayensis Merr., 366. philippinensis Merr., 366. PANGANIBAN, C. S., and SCHOBL, O., Ex- perience with methylene blue-eosin lac- tose agar in searching for Bacillus dysenteriae in stools, 235. - Pangarandéngen, 637. Pangarandingin, 637. Pangardisen, 600. 689 Papaver rhoeas Pall., 527. Paper pulp, 470. . Papualthia Diels, 389. heteropetala Merr., 389, lanceolata Merr., 390, Parameria Benth., 248, : barbata (Blume) K, Schum., 248. glandulifera Benth., 248. Paramoecium caudatum, 214. Parashorea plicata Brand., 541. Parishia Hook. f., 418. malabog Merr., 414. oblongifolia Merr., 413, Parsonsia barbata Blume, 248. Pasak, 621. Pasakla, 627. Pasau, 611. Pauropsylla deflexa Uichanco, 545, montana Uichanco, 546. tuberculata Crawf., 544. Pavetta, 250. Pedicinus eurygaster Gerv., 172, 184. Pediculidae, 184. Pediculus, 173. corporis de Geer, 171. humanus Linn., 184. Pelvic outlet, 263. Pelvimetry among Filipino women, 253. Pelvis, measurement of, 254, Pentatrichomonas, 207, 210, 212, 218, 215, 217. bengalensis Chatterjee, 212, 216. Peptone, in treatment of typhoid fever, 38. Perrottetia HBK, 244. alpestris (Blume) Loesen., 244. Phaeanthus Hook. f. & Th., 390. ebracteolatus Merr., 391. pubescens Merr., 390, 391. villosus Merr., 391. Phaleria Jack, 429. cumingii F.-Vill., 429. perrottetiana F.-Vill., 429. platyphylla Merr., 429, Philaematomyia crassirostris Stein, 177, 186. inferior Stein, 177, 186. insignis, 177. Philippine Islands, Anthophoridae of, 195. bast-fiber ropes of, 561. bloodsucking insects of, 169. cause of coconut bud rot in, 131. cephalometry in, 258. gall midges of, 287. glass materials in, 465. Hevea brasiliensis in, growth of, 501. ° lumbang oil in, 275. medicinal plants of, 1. Melectidae of, 195. new noctuid from, 71. new or rare reptiles of, 105. new plants of, 365. Nomadidae of, 83. obtuse-tongued bees of, 191. palms of, 295. pelvimetry in, 253. pink disease in, 567. 690 Philippine Islands—Continued. plant galls of, 527. Prosopidae of, 191. social bees of, 77. sphecodine bees of, 555. Philodicus longipes Schiner, 186. Phiebotomus, 175. nicnic Banks, 163, 164, 186. Phoenicospermum, 420. javanicum Jungh., 421. javanicum Migq., 419, 420. Phoenix Linn., 339. hanceana Naud. var. philippinensis Becce., 839. hanceana Planch. var. philippinensis Bece., 299. Pholidocarpus, 296. Photinia Lindl., 408. serrulata Lindl., 403. urdanetensis Elm., 429. Phthirus pubis Linn., 170, 184. Phytophthora, 140. faberi Maubl., 138, 137, 140, 141, 148, 145- 149. jatrophae, 147. Phytoptidae, 528. Pig, absence of both hind legs in, 201. Pigafetta, 296. Pilig, 341. Pinanga Blume, 295, 296, 313. barnesii Beec., 299, 314, 320. barnesii Bece. var. macrocarpa Becc., 320. basilanensis Becc., 299, 317, 322. batanensis Bece., 297, 299, 317, 322. copelandi Bece., 299, 314, 320. eurranii Bece., 299, 314, 320. disticha, 297, 817, 318. elmerii Becc.,.299, 315, 320, 321. geonomaeformis Beec., 299, 313, 318. _ heterophylla Becce., 299, 314, 319. insignis Bece., 297, 316, 322-824. insignis Bece. forma typica, 299. insignis Bece. ener: loheriana Beee., 316, 322. insignis Beec. var. gasterocarpa Becc., 2 816, 322. insignis Bece. var. leptocarpa Beec., 299, 316, 822. insignis Becc. var. icheriaan Becc., 299. isabelensis Bece., 299, 314, 318. maculata Porte, 297, 298, 313, 317, 318. modesta Bece., 299,314, 318. * negrosensis Bece. 299, 316, 322. philippinensis Bece., 299, 315, 320. rigida Beec., 299, 316, 321. Samarana Becc., 299, 315, 321, sclerophylla Beecc., 299, 316, 322. sibuyanensis Beec., 297, 299, 317, 324. speciosa Becc., 297, 299, 317, 323. urdanetana Becc., 299, 315, $21. urosperma Beec., 299 315, 321. woodiana Becc., 299, 316, 322. | Pink disease of citrus in the Philippines, 657, 669. Index Piperaceae, 548. Piper loheri C. DC., 548. Piptospatha, 356. Pithecolobium Martius, 243. jiringa (Jack) Prain, 243. lobatum Benth., 243, Pittosporaceae, 401. Pittosporum Banks, 401. acuminatissimum Merr., 402. glaberrimum Merr., 402. odoratum Merr., 403. pseudostipitatum Merr., 401, resiniferum Hemsl., 402. Planodiplosis Kieff., 292. Plants, medicinal, constituents of, 1. _ Philippine, new or noteworthy, 365. Plectocomia Mart. & Blume, 295, 297, 342. eimerii Becc., 299, 342. muellerii, 298. Plectocomiopsis, 296. Pleuropetalon Blume, 415, 416. Pleuropetalum Hook., 416. Podilymbus, 99, 100. podiceps, 98, 99. Pedophyllum peltatum, 1. Pogahan, 337. Pogonosoma cyanogaster Bezzi, 186. Pola, 337. Polyactis, 56. Polyalthia ®lume, 391. dolichophylla Merr., 391. Polyosma Blume, 399. longipetiolata Merr., 400. verticillata Merr., 400. villosa Merr., 399. Porphyrio, 89, 91, 92, 94-98. madagascariensis, 88. pulverulentus Temm., 87-89, 98, 103. Premna Linn., 248. pyramidata Wall., 248. Proceedings of the Manila Medical Society, see Manila Medical Society. Promachus bifasciatus Macq., 186. forcipatus Schiner, 186. Prosopidae, 191. Prosopis Fabr., 191. benguetensis CklIl., 191, 193. contradicta CkIl., 192, 194. euneifera Ckll., 191, 192, feai Vachal, 193. jacobsoni Friese, 192. luzonica Ckll., 192, 194. mindanensis CkIil., 191, 193, 194. mustela Vachal, 194. opacissima Ckll., 191, 193, palavanica Ckll., 191, 193. philippinensis Ashm., 191. taclobana CkIl., 191, 192. tagala Ashm., 191, 192. worcesteri Ckll., 191, 192. Protein, foreign, in treatment of typhoid fever, ages Protozoa, endoparasitic, 211. Pryor, J. C., notice of his Naval Hygiene, 153. Index Pseuduvaria Miq., 392. grandifiora Merr., 392. philippinensis Merr., 393. Psychodidae, 174, 175, 186. Psychoda, 175. Psyllidae, 544. Ptelea trifoliata, 1. tinctorium (Blanco) Merr., 581, 592, 593. Pterospermum Schreber, 413, 420. blumeanum Korth., 421. diversifolium, 590-592. megalanthum Merr., 420. Ptychoraphis Bece., 295, 326, 329. angusta, 328. cagayanensis Becce., 299, 326, 328. elmerii Becc., 299, 326, 328. intermedia Becc., 299, 326, 328. microcarpa Becc., 299, 326, 327, 328. philippinensis Becce., 326. singaporensis, 328. Pubentjil, 245. Pulex irritans Linn., 180, 185. Pulicidae, 174, 185. Pulicinae, 185. Puos, 625. Puot-sintang, 607. “Pupipara, 185. Puriket, 607. Puspus, 625. Pyrgus, 444. Pythium palmivorum Butler, 132, 137. Q Quadrifinae, 71. Quartz, 467. R Ranatra parmata Mayr, 182, 185. Raphidophora perkinsiae Engl., 549. Rapok, 598. Rassee, 356, Rauwolfia Linn., 449. membranacea Merr., 449. Reduviidae, 181, 184. ‘ REINKING, OTTO A., Phytophthora faberi Maubl.: The cause of coconut bud rot in the Philippines, 131. Reptiles, Philippine, 105. REVIEWS: Abstract of Surgery, An abstract of the war literature of general surgery that has been published since the declaration of war in 1914 prepared by the Division of Surgery, Surgeon-General’s Office, St. Louis, 153. Chaudhuri, Tarini Charan, Modern Chem-— istry and Chemical Industry of Stareh and Cellulose, 153. Gulick, Water Vose, Mental Diseases, A Handbook Dealing with Diagnosis and Classification, 251. Howell, William H., A Text-book of Phys- iology. for Medical Students and Physi- cians, 559. Jordan, Edwin O., A Text-book of Gen- eral Bacteriology, 363. 691 REVIEWS—Continued. Kaupp, B. F., The Anatomy of the Do- mestie Fowl, 251. Keen, W. W., The Treatment of War Wounds, 363. Lydston, G. Frank, Impotence and Steril- ity with Aberrations of the Sexual Function and Sex-gland Implantation, 153. Mallory, Frank Burr, and Wright, James Homer, Pathological Technique, A Prac- tical Manual for Workers in Pathological Histology and Bacteriology including Directions for the Performance of Au- topsies and for Clinical Diagnosis by Laboratory Methods, 559. Morton, Henry H., Genitourinary Diseases and Syphilis, 153. Pryor, James Chambers, Naval Hygiene, 153. Sluder, Greenfield, Concerning some Head- aches and Bye Disorders of Nasal Ori- gin, 153. Whetzel, Herbert Hice, An Outline of the History of Phytopathology, 251. Wittich, F. W., Information for the Tu- berculous, 251. Rhamnaceae, 244. Ricini, 183. Ricinus communis L., 183. Robber flies, 182. Rope, breaking length of, 575. elongation of, 574. fabrication of, 565. mechanical tests of, 566. tensile strength of, 572. twist and girth of, 569. Rosaceae, 403. Rottboellia sanguinea Retz., 369. Rourea Aublet, 404. erecta (Blanco) Merr., 405. luzoniensis Merr., 404. Rubiaceae, 250, 540. Rutaceae, 409. Ss Sabethinae, 174. Sadawag, 323. Sagiat, 638. Saginsaginan, 587. Sagisé, 325. Saguerus saccharifer Wurmb, 335. Saguisi, 325. Salacon, 323. Salaniog, 325. SALEEBY, N. M., The treatment of human beriberi with autolyzed yeast extract, 11. Salike, 603. Salisi, 624. Saléyong, 631. Saléyot, 611. Saliyot, 611. Saliyut, 611. Sambonotan, 356. Samian, 353. Sapindaceae, 417. Sapotaceae, 446. 692 Saraway, 323. Sarep, 128. Saropogon rubricosus Bezzi, specularis Bezzi, 186. Sarungas 4 dadakkel, 587. Saurauia Willd., 422. elegans (Choisy) F.-Vill., santosii Merr., 422. Saxifragaceae, 399. Sayapo, 581. Sayapu, 581. Schefflera Forster, 435. alvarezii Merr., 436. brevipes Merr., 436, 439. catanduanensis, 440. eaudatifolia Merr., 437. clementis Merr., 438. digitata (Blanco) Merr., 437. globosa Merr., 437, 438. macrantha Merr., 487. obtusifolia Merr., 435, odorata (Blanco) Merr. et Rolfe, 549. panayensis Merr., 439, platyphylia Merr., 438, santosii Merr., 438, Schist, 468. Schistosoma, a case of, 156. Sehizomycetes, 58, 59, 61. Schizomyia acalyphae Felt, 534. diplodisei Felt, 538. Schizotrypanum cruzi Chagas, 180. SCHGBL, O., see PANGANIBAN and ScHést. Seheenus coloratus Linn. var., 369. niveus Murr., 369. Scirpus glomeratus Linn., 869. spiralis Rottb., 870. Scorpions, water, 182. Selandria (Paraselandria) Semecarpus Linn. f., 411. ferruginea Merr., 412, gigantifolia F,-Vill., 412. philippinensis Engl., 413. subsessilifolia Merr., 411, Sempaling, 244, Serrar, $41. Shorea guiso Blume, 541. ' SHUFELDT, R. W., The osteology of the giant gallinule of the Philippines, Por- phyrio pulverulentus Temminek: With notes on the osteolozgy of Tachybaptus philippensis (Bonnaterre) and Hydro- phasianus chirurgus (Seopoli), 87. Gaston 119. auriculatum Taylor, 120. infralineolatum Peters, 120. kempi Taylor, 105, 115, 118. quadrivitatum Peters, 120. Siapo, 618. Sida Linn., 245. acuta Rurm., 603-606. corylifolia Wall., 245. periplocifolia Linn., 245. Silica, 467.5... Silvestrina Kieff., 292. 186, 423. imitetrix Ashm., | Streptothrix Colin, 55-61, 63-65, Index Si madoeridoeri, 243. Simuliidae, 174. Simulium, 178. Sinaligan, 597, 6381. Si panggil, 249. Siphonaptera, 179, 185. Siphonodon celastrineus Griff., 542. Sirisin, 624. Si silan niboet, 244, Skiagrams of cases showing eoins, etc., im the cesophagus and lower tract, 161. Skusea diurna Theob., 174. Sloanea Linn., 365, 419, 420. javanica (Miq.) Szyszyl., 419. Sluder, G., notice of his Concerning some Headaches and Eye Disorders of. Nasal Origin, 153. Smeringolaphria alternans Wied., Snakes, Philippine, 105. Soap, 470. Sodium carbonate, 470. sulphate, 470. Soil, relation of, to Hevea, 508. Sonchus volubilis Rumph., 250. Spatholobus gyrocarpus (Wall.) Benth., 542. philippinensis Merr., 294. Sphaerodema rustica Fabr., 185. rusticum Fabr., 182. Sphaerotilus Kuetz., 58, 60, 64. (Actinomyces) bovis Engl., 60. Sphecodes Latr., 555. bakeri Ckil., 555. biroi Friese, 556. fumipennis Smith, 556 javanicus Friese, 557. latifrons Ckll., 555, 556, 557. rotundiceps CkIl., 555, transversus Ckll., 555, 556, tristellus Ckll, 555, 557. Sphenomorphus curtirostris, 121. jagori, 122, 123. lednickyi Taylor, 105, 120. Hanosi Taylor, 105, 121. Sphodronyttus erythropterus Burm., 181, Spines, measurement of, 254. Splenoctomy, a series of cases of, 155. Staphylea indica Burm, f., 245, Stegomyia persistans Banks, 174. scutellaris Walk., 174. 186. ‘| Stellaria Linn., 382. Sterculiaceae, 246, 420, 561, 581. Sterculia colorata Roxb., 246. crassiramea Merr., 589, 593-595. foetida Linn., 595-597. > media (Linn.) Cyr., 382. oblongata R. Br., 597. stipularis R. Br., 598-600. Stomoxinae, 186, Stomoxys, 176. ealcitrans Linn., 176, 186. nigra Maca., 276, 186. Stone, 487, 495. _actinomyces Rossi-Doria, 58. - foersteri Cohn, 55-59, $i, 65. israeli Kruse, 61. Index : Stylaphorum diphyllum, 1. Stylocoryna mollis Wall., 250. Suma, 1. Sumatra, flora of, 239. Sumbaviopsis J. J. Sm., 244. albicans (Blume) J. J. Sm., 244. Sycanus fulvicornis Dohrn, 184. siali Dohrn, 181, 185. Symphorema luzonicum F.-Vill., Symplocaceae, 447. Symplocos Jacquin, 447. brachybotrys Merr., 447. purpurascens Brand, 447. 543. _ e Tabanidae, 174, 185. Tabanus rubidus Wied., 185. striatus Fabr., 185. striatus Linn., 175. Tabernaemontana Linn., 450. ecarinata Merr., 450. pandacaqui Poir., 450. Tachybaptus, 98-100. . philippensis (Bonn.), 98. Tacohtoh, 308. Tagabang, 611. | Tagisi, 325. Taguiti, 355. Tagung-tingan, 592. Taka magindanau, 611. Takkinbaka, 603. Takong, 592. Takilau-blanco, 600. Takung, 592. Talakau, 587. Talingaan, 590. Taloktok, 588. Talosan, 587. Taldto, 592. Talito, 592. Tamanag, 588. Tamba loea, 245. Tamesis, 351. Tan-ag, 588, 589. Tanak, 589. Tangalo, 324. Tadto, 592. Tapinag, 5938. Tarenna Gaertn., 250. — mollis (Wall.) Val., 250. winkleri Val., 250. Taroi, 618. Tarokték, 635. Tatagtag, 637. Tattu, 592. TAYLOR, Abra River, 127. Telsis, 340. Temperature, relation of, to Hevea, 507. Tensile strength of rope, 572, Teratoma of the maxillary antrum, 161. Tetranthera ferrugines Re 2a “‘Tetratrich , 215, 216. Thalictrum flavum, 1. Thea sp., 667. EDWARD H., New or rare Phil- ippine reptiles, 105; Ipon fisheries of | T'rombidiodea, 184. 693 Thelepogon sanguineus Spreng., 369. Theobaldiomyia gelida Theob., 174. Theobroma cacao Linn., 657. Thespesia lampas (Cav.) Daiz. & Gib., 581, 605-607. Thrips, 182.. Thymelaeaceae, 429. Thysanoptera, 182, 547. Tibig, 628. Tick, cattle, 183. dog, 183. Ticks, 182, Tiliaceae, 588, 561, 610. Tipélo, 621. Tjipa tjipa, 246. Toddalia asiatica, 1. Tolosan, 587. Tong-tongking, 587. Tonsillitis, 160. Toot-plant, 5. Trachoma, grattage of the lids for, unusual complication after, 160. Trema orientalis Blume, 637, 633. Trichomonas, 208, 218, 215, 216. prowazeki, 215. Trichomycetes, 60. Trichuriasis, 484, 486. Trichuris trichiura, 207. Tricontarinia luzonensis Felt, Trifidacanthus Merr., 405. unifoliolatus Merr., 405. Trigona Jur., 77. ambusta CkIl., 78. atripes Smith, 78, 79. bakeri Ckll., 78, 79 biroi Friese, 77, 79. busara Ckil., 78, 79. fimbriata Smith, 79. fulvomargineata Ckll., 78, 79. tridipennis 79, itama Ckil., 78, 79. lacteifasciata Cameron, 78. laeviceps Smith, 77, 79. Juteiventris Friese, 77. palavanica Ckll., 77. penangensis Ckil., 78, 79. thoracica Smith, 78, 79. valdezi Ckll., 78, 79. ventralis Smith, 79. 541. } Trimeresurus flavomaculatus, 1, _gramineus Shaw, 1 halieus, 111. : megregori Taylor, 105, 110. Tripogon Roth, 365, 369. chinensis (Franch.) Hack., 369. ‘Tristania R. Br., 430. oblongifolia Merr., 430. Trombidiidae, 184. ‘Trombidium, 183. spp., 184. Trypanosome, human, 180. -Trypetidae, 533. ‘Tuberculosis, pulmonary, 484, Tumalin, 352. — Tungao, 183. Twist of rope, 569. 694 Typhlops ater, 108. inornatus, 108. kraalii, 106. longicauda Taylor, 105, 108. luzonensis Taylor, 105, manilae Taylor, 105, 106. mindanensis, 110. reginae, 110. rossti, 110. ruber, 106. ® ruficauda, 106. rugosa Taylor, 105, 109. Typhoid fever, treatment of, 19. U Ualis-ualisan, 603. Uaualisin, 603. Uay 349. Ubbli, 352. UICHANCO, LEOPOLDO B., A biological and systematic study of Philippine plant galls, 527. Ulmaceae, 6387. Umbubuyukan, 618. Unapong, 589. Unopong, 589, Ués, 597. Urena lobata Linn., 607-610. Urticaceae, 241, 372, Urtica longifolia Burm., 241. Uvaria Linn., 393. leytensis (Elm.) Merr., 394. macgregorii Merr., 394, ovalifolia Blume, 395. panayensis Merr., 393, Vv Vaccine, nonsensitized, in treatment of ty- phoid fever, 35. sensitized, in treatment of typhoid fever, 22, 32, Vaccinium Linn., 248, 441. hasseltii Miq., 248. ilocanum Merr., 441, platyphyllum Merr., 442. Verbenaceae, 248, 451; 536, 543. Vernonia Schreb., 456. glandulifolia Merr., 456. lancifolia Merr., 543. . phyrrhodasys Schulz-Bip., 457. Index Villaresia Ruiz & Pavon, 365, 414, 416. latifolia Merr., 415. philippinensis Merr., 414, 416. suaveolens (Blume) Val., 415. Villebrunea Gaud., 241. Vitaceae, 245, 587, 539. w WADE, H. W., see MerRILL and WADE. Wallichia oblongifolia (non Griff.) Veec., 335. — Water bugs, 182. Webera mollis Hook., 250. sumatrana Boerl., 250. WELLS, A. H., The physiological active con- stituents of certain Philippine medicinal plants: III, 1. West Indies, Corticium in, 659. Whetzel, Herbert Hice, notice of his, An Out- line of the History of Phytopathology, 251. aos Wissadula Med., 245. periplocifolia (Linn.) Thwaites, 245. WITT, J. C., The effect of calcium sulphate on cement, 221. Wittich, F. W., notice of his Information for the Tuberculous, 251. \ Woreesteria grata Banks, 175. x Xanthorrhiza apiifolia, 1. Xanthoxylum americanum, 1. clava-herculis, 1. senegalense, 1. Xenomyza vitripennis O. S., 186. Xylopia polycarpa, 1. bg Yaka, 611. YATES, HARRY S., The growth of Hevea brasiliensis in the Philippine Islands, 501; see also Leg and YATES. Yeast, autolyzed extract of, 11. Yields of Hevea brasiliensis, 521. Yute, 611. Z Zalacca Reinwardt, 295, 297, 298, 342. clemensiana Becc., 298, 2°, 342. edulis, 298. Zalaccella, 296. Zizyphus Juss., 244. ealophylla Wall., 244. | ornata Mia., 244. - oO