I?'' O J rfi'ix. t?or- THE PHILIPPINE JOURNAL OF SCIENCE Paul C. Freer, M. D., Ph. D. SUCCEEDED BY Alvin J. Cox, M. A., Ph. D. GENERAL EDITOR Section C. botany E. D. MERRILL, M. S. EDITOR WITH THE COOPERATION OF P. W. GRAFF, B. S.; W. H. BROWN, Ph. D. H. N. WHITFORD, Ph. D. VOL. VII 1912 With 22 Plates and 2 Text Figures MANILA bureau of PRINTING 1912 Z2^8S(] DATES OF ISSUE. No. 1, pages 1 to 46, May 14, 1912. No. 2, pages 47 to 124, July 1, 1912. No. 3, pages 125 to 208, September 2, 1912. No. 4, pages 209 to 258, September 30, 1912. No. 5, pages 259 to 362, November 15, 1912. No. 6, pages ^2^ to 419, January 15, 1913. 3^3 ^*oT~yrri~ CONTENTS. No. 1, April (May), 1912. Pa«e. Ames, 0. Notes on Philippine Orchids with Descriptions of New Species, IV 1 Brand, A. Additional Philippine Symplocaceae, II 29 Brown, W. H. The Mechanism of Curvature in the Pulvini of Mimosa pudica 37 Copeland, E. B. The Genus Thayeria 41 No. 2, June (July), 1912. Copeland, E. B. The Origin and Relationships of Taenitis 47 Copeland, E. B. New or Interesting Philippine Ferns, VI 53 Copeland, E. B. New Sarawak Ferns 59 Copeland, E. B. New Papuan Ferns 67 Groves, H. & J. Characeae from the Philippine Islands 69 Merrill, E. D. Sertulum Bontocense: New or Interesting Plants Collected in Bontoc Subprovince, Luzon, by Father Morice Vanoverbergh 71 Wester, P. J. A Contribution to the Nomenclature of the Culti- vated Anonas 109 No. 3, August (September), 1912. Ames, 0. Notes on Philippine Orchids with Descriptions of New Species, V : The Genus Bulbophyllum in the Philippine Islands.. 125 Merrill, E. D. Notes on the Flora of Manila with Special Reference to the Introduced Element 145 No. 4, September, 1912. Brown, W. H. The Relation of Rafflesia manillana to its Host 209 Merrill, E. D. Nomenclatural and Systematic Notes on the Flora of Manila 227 Warnstorf, C. Die Sphagna der Philippinen 253 No. 5, November, 1912. Merrill, E. D. New or Noteworthy Philippine Plants, IX 259 Review 358 No. 6, December, 1912. Merrill, E. D. The Pineda Monument and the Probable Site of the First Botanic Garden in the Philippines 363 Merrill, E. D. On the Identity of Evodia triphylla 373 Merrill, E. D. Notes on Philippine Euphorbiaceae 379 Robinson, C. B. Roxburgh’s Hortus Bengalensis 411 Errata 421 Index 423 III April, 19 THE PHILIPPINE JOURNAL OF SCIENCE PAUL C. FREER, M. D., Ph.D. GENERAL EDITOR Section C. Botany E. D. MERRILL, M.S. EDITOR WITH THE COOPERATION OP P. W. GRAFF, B. S. ; W- H. BROWN, Ph.D. ^ N. WHITPORD, Ph. D. ' • L, MANILA ■ BURfiAUi OP PRINTING . 1912; ^ PtrsnCATIONS foe sale bt the bheeait of science, MANHA, PmtlPPIHE ISLAHDS/ BUPORT pp THE itNTEBNATIQNAIi PliAGUE COSPPEBEIfCE. - Held at Mukden, April, 1911, under tke auspices of ' ^ the Chinese Go-yiernment, ; , ^ ' Edited by Richabd P. Stbong. . , - ' 485 page*, 18 plate* (2 colored, 4 halMonea, 12 chart* and maps). Order; No. 4l6b Cloth, f3.50; Paper, A2iS0;Uhited State* currency, postpaid. The proceeding* of thl» International Conferende and .Information gained' therefromi together with the reeulte of certain baoterlolonlcaJ Investigations, constitute the present report. . , Nothing hitherto ha* been, published Which gives such a complete and oomprehentive account of the. entire subjeol of pneumonlo plagpe. • ■ . < : ■ ' Delegates from .America (United Stated of),:' Austria,Hungary,,Franoe, Germany, Great Britain, Italy, Japan,. Mexico,, the Netherlands, .Russia, and China attended the /Conference. . < The Bureau ef Scietioe of the Government'.of the .Philippine Islande has .h.een appointed toie agent for tha distribution cf the 'printed proceedings of the International Plague Conference. - .< THE Sp^GAB INDTjrSTBY JX THE ISliAND OF NEGBOg. ' ' ■ By' HeBbeet S. WAtKEB. 145 pages, 10 plates, 1 map. No. 412. . Paper,' $1.25 United States currency, postpaid.. Considered from the viewpoint of' praotloal utility; Mr. Welker’e Sugar; industry In the Island, of Negros Is one of the most Important papers published by .the. Bureau of Soienoe. This volume' is a real contribution'. to the 'subjeotj. It Is; not. :a mere compilation, -for. the author was in .lhe .field / and underst^inds the oon'ditions of Which writes. The following. . Is a brief synopsl* of the opntents; ^ ' , , , ^ , : Tables of soil analyses, bptli otemloal and, Physical analyses of the cane, jploe and bagasse; OalimateSi based on aotijal Informatipn as to the’ costs of production and'. of cultivation; and esti-.; mates of the cost and location, of possible central factories. The Island is' considered bycugHr. producing' districts; the 'area of . cultivation ''and the production per hectare are given, and the.. Possibility for future ercnansion discussed. . The plates Illustrate various phases of sugar Industry from the oultivation of the field; to the traneporiation of sugarr-in native .sailboats. . A MANUAIi OP PHlIilPPINE SILK CULI^UBE- By ClHAELES'S. Banks. 55 pages, 20 plate*. Order No. 415. , Paper, $0.75, United Stale* currency, postpaid. . The Silk Industry I* particularly adapted to be undertaken by persons with' small capital, and like the making of'hats in the Philippine, Islands it should thrive with a little encouragement, ! In A Manual of PhlllopUie SilkiCullure we have presented the-result*. of several yaars' .actual werkiWlth sllk-produclng. larva .together with a description of thyi new Philippine rape, . HalMono plaW iHuslrele in- natural . sl?«ySllkworms In - different stages of deyelopmertl, pupse, adult mpths,; samples of cloth made from ' erl silk, hand reel, and *llk house. .Other, platee illustrate the, various appliances, used. In 'ital'slhg ‘Silkworms and in spinning silk; hand' ai>d power .feels efo Illustrated; working dfawlngs are given for a silk house and for , a hand reel. , , . , , THE PHILIPPINE Journal of Science C. Botany VoL. VII APRIL, 1912 No. 1 NOTES ON PHILIPPINE ORCHIDS WITH DESCRIPTIONS OF NEW SPECIES, IV. By Oakes Ames. {From the Ames Botanical Laboratory, North Easton, Mass., U. S. A.) The following paper contains descriptions of twenty-seven new species from the Philippine Islands, one genus, heretofore unrecorded as a native of the Philippines, and one species, namely Dendrobium angusti folium Blume, a native of Java, Sumatra, and the Malay Peninsula, which is an addition to the list of Philippine plants. Cestichis Cumingii (Ridley) Ames is here included as a distinct species. In Orchidaceae 1:13, doubts were expressed as to its distinctness from C. compressa. From careful studies of C. com- pressa, as represented by Javan material, and of C. Cumingii, as represented by the specimens collected by Weber on Mount Mariveles, I am convinced that two species are before us. What is true of C. Cumingii is true of numerous other Philippine or- chids which have close allies in Java and the Malay Peninsula. A general similarity undoubtedly exists, but comparative studies of each individual case show differences, the sum of which in- dicate clearly that there are constant differentiating traits of specific value. This fact becomes more and more apparent with each case studied and makes it seem highly probable that many Philippine plants which have been referred to extra-limital species will be found, after critical examination, to be distinct 108801 2 AMES. and endemic. An interesting example of this is offered by the genus Dendrochilum in the Philippines, which is composed en- tirely of endemic species, if we exclude one or two which have been ascribed to the Philippines conjecturally on evidence too slight for recognition. HABENARIA Willd. Habenaria Curranii sp. nov. Planta 6 dm alta, foliis paulum infra medium partem caulis congestis, lanceolatis, acuminatis. Sepala lateralia lanceolata, vel lineari-lanceolata, aristata, sepalum superius simile, lanceo- latum, acuminatum, cucullatum. Petala bipartita, lacinia pos- terior linearia, lacinia anterior posteriori similis sed longior, 4.5 cm longa. Labellum usque ad basim tripartitum; laciniae laterales media longiores, filiformes ; lacinia media 1.7 cm longa. Calcar prope apicem subdilatatum, processus stigmatis promi- nentes. Plant tall, slender throughout, leaves five or more, narrowly lanceolate, acuminate, acute, 7.5 to 11 cm long, 1 to 2 cm wide, clustered near the middle of the stem or lower, about 2 dm above the roots. Bracts of the stem sheathing, those above the leaves acuminate, aristate, about 2 cm long. Raceme slender, elongated, smooth, many-flowered (about 20), flowers greenish. Bracts of the inflorescence aristate, about equaling the flowers, ovary and pedicel about 2 cm long, smooth. Lateral sepals spreading, narrowly lanceolate, 3-nerved, acuminate, aristate, about 1 cm long, 2 mm wide. Upper sepal similar to but shorter than the laterals, 7.5 mm long. Petals bipartite, posterior divi- sion (free from the dorsal sepal?) linear, 1 cm long, anterior division filiform, 4.5 cm long. Labellum tripartite, lateral divi- sions much longer than the middle one, filiform, about U cm long (!), middle division broader than the laterals, about 0.5 mm wide, 1.7 cm long. Stigmatic processes elongated, tapering, longer than the anther-canals. Spur stout, somewhat inflated near the tip, about 2.4 cm long. Luzon, Province of Cagayan, Lalloc, H. M. Curran, For. Bur. 171SS, February 22, 1909: Benguet Subprovince, Sablan, E. Fenix, Bur. Sci. 12578, December 6, 1910. Similar in habit to H. ponerostachys Reichb. f., but unlike that species in the size and structure of the flowers. It is also like H. aristulifera Reichb. f., but the floral divisions are conspicuously longer in H. Curranii Habenaria Delessertiana Kranzl. Gen. et Sp. Orch. 1 : 233. Kranzlin cites Cuming’s 2086 as the type of this species, but under Habenaria muricata Vidal (H. hystrix Ames) he also cites this number of NOTES ON PHILIPPINE ORCHIDS, IV. 3 the same collection without any allusion to a mixture having been made in the distribution of Cuming’s specimens. In my studies of the material collected for the Bureau of Science I have examined two series of specimens which are clearly referable to H. Delessertiana, being chiefly characterized by the linear petals, protuberant in part at the base, by the aristate sepals, and by the middle lobe of the labellum being about one-fourth longer than the lateral lobes. Another characteristic is the crowding of the oblong-lanceo- late leaves near the middle of the stem. All of the material distributed as 2086 of Cuming’s collections which I have examined is referable to Habenaria hystrix, and as H. muricata Vidal this number is listed in “Phanerogamae Cumingianae Philippinarum.” {H. muricata is a synonym of H. hystrix). Kranzlin cites “Calananz?” as the habitat of H. Delessertiana. The material from Luzon which I have examined was collected at Calauan (written on the field label as Calauang). May not the localities in question be identical, Kranzlin’s “Calananz?” being the same as Calauang?’ Luzon, Province of Laguna, Calauan, R. C. McGregor, Bur. Sci. 12407, November 30, 1910. Mindanao, Lake Lanao, Camp Keithley, Mary Strong Clemens 865, November, 1906. Habenaria Mearnsii sp. nov. Habitu H. clavellatae (Michx.) Spreng. baud dissimilis. Folia oblongi-elliptica et lineari-lanceolata. Bracteae inflorescentiae ovariis longiores, excedentes flores infimos. Sepala lateralia lanceolata vel ovato-lanceolata, subfalcata. Sepalum superius ellipticum, obtusum. Petala simplicia, oblonga, obtusa. Ldbel- lum simplex, linearis; prope basim labelli tuberculum. Calcar ovario brevius. Tuberoids fleshy (3 ?) . Plants 1.4 to 3.6 dm tall, strict, few- leaved, in leafage somewhat resembling H. clavellata (Michx.) Spreng. Loivermost leaf elliptic-oblong, obtuse, 4 to 7 cm long, 1.3 to 2 cm wide, much longer than those above it, which are bract-like and about 4 in number. Upper leaves linear, acute, the uppermost one about 1 cm long, linear, acute. Floral bracts linear-lanceolate, acute, lower ones exceeding the flowers. Raceme 4.7 cm long, many-flowered, flowers pale-green. Lateral sepals lanceolate or ovate-lanceolate, subfalcate, obtuse, 3.5 mm long, about 2 mm wide near the base. Upper sepal elliptic, obtuse, 3 mm long. Petals simple, oblong, obtuse, 3 mm long, 1.25 mm wide. Labellum 4 mm long, 1 mm wide, simple, fleshy, linear, tapering toward the tip, with a fleshy retrorse callus or ’ “Calananz” is unquestionably a misprint for Calauan, or Calauang as it is frequently spelled. Cuming collected in Calauan from the end of Sep- tember to the 15th of December, 1836, according to a letter written by him from Manila Dec. 24, 1836, to Sir W. J. Hooker, now preserved in Hooker’s correspondence at Kew. There is no town by the name of Calananz in the Philippines. E. D. M. 4 AMES. protuberance near the base. Spur 5 to 6 mm long, slender, not at all inflated. Stigmatic processes wanting or very much reduced. Luzon, Subprovince of Benguet, Pauai, Major E. A. Mearns, Bur. Sci. US 13, July, 1907; Mount Pulog, R. C. McGregor, Bur. Sci. 8815, 8835, in grass lands near the summit, July 3 and 4, 1909. H. Mearnsii is similar in its leafage and general appearance to H. clavellata (Michx.) Spreng., although totally dissimilar in the details of the inflorescence. The gyno- stemium of the flowers examined was so pressed out of shape that it was impossible to describe it with fullness. Habenaria ponerostachys Reichb. f. in Bonplandia 3: 213; Kranzlin in Engler’s Jahrb. 16: 179; Gen. et Sp. Orch. 1: 385. This species was described by Reichenbach in 1855 from material collected in the Philippines by Hugh Cuming (no. 2095). The next reference to the species was made by Kranzlin in Engler’s “Jahrbiicher” and subsequently in “Genera et Species Orchidacearum.” In the Herbarium of the British Museum of Natural History there is one of Cuming’s specimens from the Philippines (no. 2095) which it is reasonable to suppose represents H. ponerostachys, but in several details it is not in agreement with Reichen- bach’s description. Kranzlin’s description, on the other hand, is quite satisfactory. In Engler’s “Jahrbiicher,” Kranzlin states that Reichenbach drew up his description from a depauperate specimen preserved in the Boissier Herbarium. In April, 1910, Mr. Merrill secured specimens from the Island of Negros which are almost a perfect match for the specimen of Cuming 2095, preserved in the British Museum of Natural Histoi'y. These, I believe, are clearly referable to H. ponerostachys. Description: Plant 2 to 5 dm tall, very slender, with several sheathing, obtuse or acute bracts below the leaves. Leaves clus- tered 7 to 14 cm above the stout fleshy roots, lanceolate to ovate- lanceolate, acute, shortly aristate, 3 to 9 cm long, 1.6 to 2.7 cm wide. Spicate raceme 6 to 25 cm long, very slender, many- flowered. Bracts lanceolate, erect, acute, 5 to 10 mm long, about equaling the matured capsules. Lateral sepals elliptic, 2.5 to 3 mm long, about 1.5 mm wide. Upper sepal ovate, about equal to the laterals. Petals asymmetrically ovate, blunt, 1-nerved, 3 mm long, 2 mm wide, simple, lahellum 3-lobed, similar to that of Peristylus tentaculatus J. J. Smith, 3 mm long to tip of middle lobe. Lateral lobes linear, longer thSn the middle one, abruptly curved at the tip, 2 mm long. Middle lobe gradually tapering to the obtuse tip, 1.5 mm long, about 1 mm wide. Spur longer than the lip, slender at base, dilated toward the tip, about 5 mm long. Negros, Canlaon Volcano, Elmer D. Merrill 7022, April, 1910. Terres- trial, ,on forested ridges, about 1500 m above sea level, flowers greenish- white. NOTES ON PHILIPPINE ORCHIDS, IV. 5 Habenaria Robinsonii sp. nov. Folia linearia, acuta. Bracteae aristatae. Pedunculus glab- ratus. Flores albi, 1 ad 4. Bracteae inflorescentiae aristatae. Sepala lateralia oblongi-ovata, subacuta, 3-nervia. Sepalum su- perius cucullatum, obtusum. Petala spathulata, obtusa, 1-nervia. Labellum 4-Iobatum; lobi laterales cuneati, truncati, lobi termi- nales rotundati. Calcar 4.7 cm longum. Whole plant from 10 to 27 cm high (including the inflo- rescence). Bulbs elongated, roots fleshy. Leaves linear, acu- minate, 5 to 13 cm long, 4 to 9 mm wide (average about 6 mm), slightly contracted at the base, sheathing, passing rather abruptly into the sheathing, aristate bracts. Braets 3 or more, 2 to 3.5 cm long, long-aristate. Raceme glabrous, slender, flowers 1 to 4 (rarely 4). Floral bracts 2 to 3 cm long, aristate, sheathing. Pedicel, together with the ovary, about 2 cm long, glabrous. Floivers large, white (in general aspect recalling H. militaris). Lateral sepals deflexed, oblong-ovate, subacute, 3-nerved, about 7 mm long, 4 mm wide. Upper sepal cucullate, obtuse, about 8 mm long. Petals simple, spathulate, rounded at the apex, 1-nerved, 8 mm long, 2 mm wide near tip. Labellum very large in comparison with the other perianth-organs, about 2 cm long, 2.5 cm wide, four-lobed, basal lobes cuneate, the free end obliquely truncated, 12 mm long, 9 mm wide at the tip, many-nerved. F7’07it lobes separated from the basal ones by a narrow isthmus 3 mm long, 2.5 mm wide, semi-rotund, margin slightly crenulate. (The front lobes might be more properly designated as the bifid or cleft middle lobe of the lip.) Stigmatic processes prominent, tapering, somewhat decurved, 2.5 mm long. Spur slender, longer than the ovary, 4.7 cm long. Luzon, Province of Laguna, Molauin River, C. B. Robinson, Bur. Sci. 9666, Feb. 2, 1910, on rocks in the river, 50 to 150 m above sea level; Elmer D. Merrill 6S06, Feb. 9, on mossy boulders in stream bed. This very beautiful species bears a striking similarity to H. militaris from which it is readily distinguished by its very different leaves, long- aristate bracts, and white flowers. Habenaria rosulata sp. nov. Folia rosulata, linearia, acuta. Bracteae aristatae. Floi'es 3 vel 4, albi. Bracteae inflorescentiae acutae, ovariis breviores. Sepala lateralia ovata vel ovato-lanceolata, subacuta. Sepalum superius ovatum. Petala linearia, 1-nervia, obtusa, simplicia. Labellum tripartitum, laciniae laterales media longiores. Calcar inflatum, processus stigmatis hippocrepiformes. Plant 1 to 2 dm tall, slender. Roots fleshy. Leaves rosulate, about 5, linear, acute, suberect or spreading, 2 to 4.5 cm long, 6 AMES. 2 to 5 mm wide, acuminate, acute. Bracts sheathing, 3 or 4, aristate, 5 to 10 mm long. Flowers few, 2 to 4, white. Floral bracts about 5 mm long, about one-third as long as the pedicel and ovary, aristate. Lateral sepals narrowly ovate, subacute, larger than the dorsal one, about 7 mm long, by 3 mm wide. Upper sepal ovate, cucullate, about 5 mm long. Petals simple, linear, 1-nerved, obtuse, 6 mm long, 1 mm wide. Labellum tripartite nearly to the base, divisions filiform, about equally wide, the laterals longer than the middle one, laterals about 1.3 cm long, middle division 7 mm long, 0.5 mm wide. Spur about 13 mm long, slender near the opening, dilated from about the middle, resembling the abdomen of an ichneumon fly. Anther canals about equaling the stigmatic processes. Stigmatic processes cylindric, hippocrepiform. Luzon, Province of Tayabas, Quinatacutan, Foxworthy & Ramos, Bur. Sci. 13208, on rocks at the edge of the stream, 75 m above sea level. Described in collector’s note as “tuber-bearing or with fleshy roots.” This species, which I have been unable to refer to any described Habe- naria, belongs, in my opinion, to the § Diphyllae. The flowers resemble those of H. falcigera, H. diphylla, etc., but the foliage is quite characteristic. GASTRODIA R. Br. Gastrodia javanica (Bl.) Lindl. Palawan, Napsahan, on the west coast, Elmer D. Merrill 7233, Sptember 19, 1910. “About rotten stumps along trail in dense forest. Petals yellowish, whole plant brownish, with purplish tinge; a very brittle, suc- culent, leafless saprophyte.” The genus Gastrodia has not, heretofore, been recorded as a native of the Philippines. KUHLHASSELTIA J. J. Smith. Kuhihasseltia Merrillii Schlechter in Fedde Report. 9 (1911) 437. Haemaria Merrillii Ames in Philip. Journ. Sci. 2 (1907) Bot. 316; Orchidaceae 3 (1908) 21, pi. 30. Doctor Schlechter, loc. cit., refers to J. J. Smith’s recently established genus Kuhihasseltia (1910) the species which, with hesitation, I described under the genus Haemaria, in my paper on the orchids collected on Mount Halcon. As stated in “Orchidaceae”* the genera which constitute the group to which Haemaria belongs are dilferentiated by means of characters which are not only of questionable value, in my opinion, but which are extremely perplexing when it is attempted to place in its proper position a new species of the Neottiinae-Physureae. As defined by J. J. Smith the genus Kuhihasseltia includes Haemaria Merrillii and Doctor Schlechter is undoubtedly correct in his views expressed in Fedde’s “Repertorium.” But what do we gain at the present time by multiplying the genera of this puzzling section of the Orchidaceae ? It would seem that an exhaus- tive monograph of the Neottiinae-Physureae should be produced before the number of genera is inordinately increased. 3: 21. NOTES ON PHILIPPINE ORCHIDS, IV. 7 TROPIDIA Lindl. Tropidia calcarata sp. nov. Folia 1 (-2), ovato-Ianceolata, acuminata, acuta. Bracteae inflorescentiae lineares. Flores albi in racemo laxo. Sepala lateralia connata, scrotum conspicuum efficientia, basim labelli includens, partes terminales liberae, 3 mm longae, acutae. Sep- alum superius anguste lineari-lanceolatum. Petala lanceolata, subacuta vel obtusa. Labellum integerrimum, lanceolatum, carin- atum, calcar breve ad apicem obtusum. Related to T. angulosa Bl, Terrestrial, 2 to 3 dm high. Stems slender, about 2 mm thick, sheathed with obtuse, tubular bracts, 1.5 to 3.4 cm long. Leaves 1 or 2 (only one bifoliate specimen among the three examined) sheathing at the base, ovate-lanceo- late, acuminate, acute, many-nerved, when mature about 1 dm long, about 5 cm wide. Peduncles terminal, shorter than the leaves, erect, 5 to 9 cm long, slender ; floral bracts linear about 1 cm long. Racemes 3 to 5 cm long. Flowers white, 1.3 to 1.8 cm long. Lateral sepals connate beyond the middle, at the base forming a conical spur 3 mm long which encloses the spur of the labellum, free portion about 3 mm long, acute. Upper sepal linear, acute, about 1 cm long. Petals lanceolate, subacute or obtuse, 1 cm long, 2.5 mm wide near the base, mid-nerve prom- inent. Labellum slightly sigmoid, calcarate, lamina lanceolate, the apex very slightly cucullate. Just below the middle, near the margin, two inconspicuous longitudinally extended keels are situated. Main nerves three in number, slightly prominent. Spur obtuse, about 2 mm long, curving slightly toward the ovary, blunt. Gynostemium 7 mm long, rostellum bifid, exceeding the acuminate anther. Luzon, Province of Laguna, Mount Maquiling, Elmer D. Merrill 71hh, September 2, 1910. “Rare, only three plants seen.” Tropidia calcarata is very near T. angulosa Blume, but distinguishable from it by the leafage and by the carinae of the lip. Unfortunately I have seen very little material of T. angulosa and my knowledge of it is chiefly based on descriptions and plates. The material from which the above description was taken consists of three specimens. The unifoliate character appears to be normal. In the single case where two leaves terminate the stem, one of them is very small and bract-like. CESTICHIS Pfitzer. Cestichis Cumingii (Ridley) Ames Orchidaceae 1: 13. Liparis Cumingii Ridley Journ. Linn. Soc. Bot. 22: 292. What I blieve to be this species is well represented by specimens from Mount Mariveles, recently collected by C. M. Weber. The suborbicular lip with a minute tooth at the obscurely retuse tip, the very characteristic inflorescence, and the broad, somewhat oblanceolate leaves, agree very well 8 AMES. with the specimen collected by Cuming (no. 2H1) preserved in the Her- barium of the British Museum of Natural History. This species should not be confused with Cestichis compressa which is a much larger plant with a larger lip. That C. compressa is very closely allied to C. Cumingii can not be denied, but for the present it seems wise to regard them specifically distinct. Luzon, Province of Bataan, Mount Mariveles, C. M. Weber 26, Feb- ruary 9, 1911. On trees in mossy forest, about 800 m above sea level. OBERONIA Lindl. Oberonia benguetensis sp. nov. Folia ensiformia, acuta, ad basim imbricata. Pedunculus elongatus, multo folia excedens, multibracteatus. Bracteae cir- citer 2 mm longae, setaceae. Racemus gracilis. Flores flavidi, minuti, circiter 1 mm longi. Sepala lateralia ovato-lanceolata. Sepaliim swperius triangulum. Petala linearia, attenuata. Lab- ellum ad basim rotundatum, oblongum, ad apicem bilobum; lobi minuti, valde separati. Roots fibrous. Plants diminutive, about 4 cm high to the tip of the tallest leaf, stem very short. Leaves ensiform or some- what falcate, acute, the largest about 3 cm long and 4 mm across from edge to edge, fleshy. Peduncle up to 1 dm long, much longer than the leaves, beset with numerous, spreading, setaceous bracts 2 mm long. Raceme somewhat drooping, about 5 cm long, 5 mm in diameter at the base when in flower. Flowers yellowish, irregularly verticellate, minute, exceeding the bracts. Floral bracts linear-acute, about 2 mm long, margin minutely denticulate. Pedicel and ovary 2 mm long, smooth. Lateral sepals ovate-lanceolate, concave, about 1 mm long. Upper sepal triangular, subacute (half lanceolate). Petals linear-lanceolate, tapering gradually from the base to the subacute tip. Labellum rounded at the base, 1 mm long, from about the middle to the tip oblong, each distal angle produced into a diminutive blunt lobe, recalling, although vaguely, Oberonia similis Lindl. (which is, however, lobed at the base), lobes about 0.5 mm apart. Column minute. Luzon, Benguet Subprovince, Suyoc to Pauai, Elmer D. Merrill Jt855, November 7, 1905. Epiphytic in mossy forest about 2,200 m above sea level. Oberonia hispidula sp. nov. Aff. 0. mindorensi speciei quam habitu et structura haec species conspicus simulat. Folia imbricata, ensiformia. Peduncidus elongatus, gracilis. Racemus densiflorus. Bracteae aristatae. Sepala lateralia ovata. Sepalum superius simile. Petala ovata. NOTES ON PHILIPPINE ORCHIDS, IV. 9 integerrima. Labellum pandurata, ad apicem bilobum, lobi ro- tundi. Ovarium hispidulum. Plants caulescent, similar to Oberonia mindorensis Ames, but different in the lobing of the lip, different petals and shorter pedicels. From base to tip of uppermost leaf, between 1 and 2 dm tall. Leaves imbricating, ensiform, about 4 cm long, free portion spreading, somewhat recurved, 1 to 4 cm long, about 1 cm deep where it forms an angle with the stem. Peduncle long, slender, about 1 dm long, about 4 mm through when the flowers are expanding. Bracts linear-lanceolate, aristate, equaling or slightly exceeding the flowers. Flowers greenish, approximate. Lateral sepals ovate, 0.75 mm long. Upper sepal similar to the laterals and about equal to them in size. Petals entire, ovate, obtuse, about 0.5 mm long. Labellum pandurate, 1 mm long, nearly 1 mm wide across the tip, margin entire, deeply retuse at the apex, or cleft into two, rounded, terminal, slightly divaricate lobes. Pedicel and ovary 1.5 mm long. Ovary hispidulous. Luzon, Benguet Subprovince, H. M. Curran, For. Bur. 512U, August 20, 1906. Oberonia Merrillii sp. nov. Planta parvula. Folia ensiformia, acuta. Peduncidus gra- cilis. Racemus elongatus. Flores verticellati. Bracteae an- gustatae, acutae. Bracteae inflorescentiae lanceolatae prope apicem, 2-dentatae. Sepala lateralia rotundata. Sepalum siip- erius ovato-oblongum. Petala orbiculata, denticulata. Label- lum 3-lobatum. Lobi laterales minuti, rotundati, lobus medius subquadratus. Columna minuta. Plant up to 9 cm tall from copious fibrous roots, acaulescent or with the stem very short, less than one quarter as long as the leaves. Leaves fleshy, equitant, ensiform, acute, 2.5 to 7 cm long, about 5 mm wide seen from the side, shorter than the inflorescence. Peduncle slender, 7 to 12 cm long, spike 4 to 9 cm long, about 4 mm in diameter when the flowers are fully expanded, below the spike beset with numerous, linear, spreading bracts 2 mm long; floral bracts lanceolate, acute, near the tip angled or 1-toothed on each side. Flmvers minute, reddish, verticellate, or subverticellate, verticels 2 to 3 mm apart. Pedicel and ovary about 1.5 mm long, slender. Lateral sepals rotundate or round-ovate, obtuse, 0.75 mm long. Upper sepal ovate-oblong, obtuse, about 1 mm long. Petals rotund, 0.75 mm long, 0.75 mm wide, margin minutely denticulate. Labellum 3-lobed, 0.75 mm long. Lateral lobes minute, smaller than the middle lobe. 10 AMES. roundish, margin irregular; middle lobe subquadrate, 0.75 mm long, 0.75 mm wide. Column minute, clinandrium entire (?). Luzon, Manila and vicinity (Masambong) , Elmer D. Merrill 73U8, No- vember 6, 1910, on trees, 6 to 8 m above sea level: Province of Nueva Ecija, Cabanatuan, R. C. McGregor, Bur. Sci. 5298, September, 1908. Oberonia setigera sp. nov. Planta parvula. Folia ensiformia vel subfalcata, obtusa vel acuta. Pedunculus elongatus. Bracteae setiformes. Sepala lat- eralia ovato-lanceolata. Sepalum superius lanceolatum, acumi- natum. Petala serrulata. Labellum 3-lobatum; lobi laterales majores valde dentati; lobus medius subquadratus, integer. Roots fibrous. Plants caulescent, 6 cm tall to tip of uppermost leaf. Leaves equitant, ensiform or somewhat falcate, obtuse or acute, 2 to 4.5 cm long, 5 to 7 mm wide. Peduncle 12 cm long, thickly beset below the raceme with setiform bracts nearly 1 cm long. Raceme densely many-flowered, about 7 mm in diameter when in fruit. Floral bracts very long, setiform, longer than the flowers, clustered at intervals. Flowers laven- der, minute, clustered at intervals, but the clusters approximate. Lateral sepals ovate-lanceolate, about 1 mm long, 0.5 nim wide. Upper sepal lanceolate, acuminate, slightly exceeding 1 mm in width. Petals fimbriate, oblanceolate, obtuse, 1 mm long. Lab- ellum about 1 mm long, 3-lobed, lateral lobes much larger than the middle one, each with the margin broken up into about 7 finger-like divisions. Middle lobe minute, subquadrate. Column minute, clinandrium entire. Luzon, Province of Bataan, Lamao River, Mount Mariveles, H. N. Whitford 1122, March 3, 1905. The flowers of Oberonia setigera are similar to those of O. Griffithiana Lindl., but the middle lobe of the labellum is quite distinct and the setiform bracts constitute an unmistakable differentiating character. PHAIUS Lour. Phaius linearifolius sp. nov. Folia 5, linearia, elongata, acuta. Scapus erectus, bracteis tubularibus vestitus. Bracteae inflorescentiae persistentes. Flores flavi. Sepala lateralia oblonga, falcata. Sepalum su- perius oblongi-lanceolatum, obtusum. Petala ovato-lanceolata. Labellum 3-lobatum ; lobi laterales rotundati, obtusi, lobus medius crenulatus, retusus. Discus tricarinatus. Gynostemium arcua- to-ascendens, clavatum, antice canaliculatum, dorso gibbum, facie anteriore concaviusculum et dense villosum. Stem thickened at the base, somewhat rounded, tapering upward. Leaves about 5 in number, the lowermost one about 2 NOTES ON PHILIPPINE ORCHIDS, IV. 11 dm above the ground, sheathing at the base, about 55 cm long, 1.5 to 3 cm wide, linear, acuminate, acute, prominently 5 -nerved. Scape erect, smooth, .about 6 dm long, with four, closely ap- pressed, sheathing, acute bracts about 4 cm long. Floral bracts persistent, smooth, lanceolate, acute, 1.5 to 2.5 cm long. Flowers 5.5 cm in diameter when spread out, 6 to 15 in a loose raceme, the lowermost 4 cm apart. Pedicel and ovary 3 cm long, slender. Lateral sepals oblong, subacute, broadly falcate, 3 cm long, 1 to 2 cm wide. Upper sepal lanceolate-oblong, obtuse, 3.5 cm long, 11 mm wide. Petals ovate-lanceolate, subfalcate, obtuse, 3.2 cm long, 1.4 cm wide. Lahellum obscurely three-lobed, exclud- ing the short spur, about 2.5 cm long. Lateral lobes slightly shorter than the middle lobe, rounded, obtuse, obscurely cren- ulate, smooth. Middle lobe irregularly dentate, the margin crumpled or ruffled, deeply retuse at the rounded tip. Disk smooth, the three central nerves carinate above the middle, somewhat wavy, diminishing as they approach the tip. Spur cylindric, round-pointed, 7 mm long. Column comparatively stout, 1.5 cm long, hairy on the anterior surface. Pollinia 8. Flowers yellow, the ruffled margin of the middle lobe brownish. Luzon, Bontoc Subprovince, Father M. Vanoverbergh 1288, June 28, 1911, growing in brooks, 1,650 m above sea level. Phaius linearifolius belongs to the groups of which Phaius fiavus is the best known species. I have refrained from identifying it with P. flavus as I find many points of dissimilarity between my material and the specimens of P. flavus examined, and also as none of the numerous diag- noses in literature describe my plants adequately, and as the illustrations of P. flavus fall far short of depicting them. CALANTHE R. Br. Calanthe lacerata sp. nov. Radices lanatae. Caules abbreviati. Folia anguste lanceo- lata, acuminata, acuta, plicata. Petiolus ad basim vaginans. Scapus foliis longior, gracilis, erectus, elongatus, tomentosus, bracteis bene appressis, vaginantibus. Bracteae inflorescentiae lanceolatae, pedicellis breviores. Sepala lateralia elliptico-lance- olata vel elliptica, acuta. Sepalum superius simile. Petala oblonga ad medium satis dilatata, ad apicem truncatum, apicu- latum. Labellum 4-lobatum, lobi inferiores oblongi, subfalcati, obtusi. Lobi anteriores divaricati, cuneati, flabellati, lacerati, papillosi. In disco ultra columnam utroque calli complanati. Calcar gracilis. Roots stout, lanate. Stem very short. Leaves about 4, about 2 dm long, up to 2 cm wide, narrowly lanceolate, acute, plicate. 12 AMES. shorter than the scape, sheathing at base. Petiole narrow, sul- cate. Scape slender, tomentose, up to 4 dm long. Bracts closely appressed, sheathing, 1 to 2 cm long, acute, the lower ones subacute or rounded at the tip, sparsely pubescent. Raceme 4 to 9 cm long, densely many-flowered, bracts of the raceme lance- olate, 6 to 10 mm long. Pedicel and ovary about 1.5 cm long, slender, exceeding the bracts. Floivers 1 to 1.5 cm across, odorless, white, labellum purplish. Lateral sepals elliptic-lan- ceolate, acute, 6 to 8 mm long, 3 to 5 mm wide. Petals oblong, slightly dilated near the middle, truncate, apiculate, 3-nerved, 5 to 8 mm long, 2.5 to 3.5 mm wide. Lahellum 1 cm long, 4-lobed. Basal lobes oblong, subfalcate, rounded at the tip, 4 mm long, about 2 mm wide, toothed in front, or almost entire, apical lobes 3 mm long, 3 mm wide across the tip, divaricate, cuneate, fla- bellate, margin lacerate-dentate, papillate. In front of the col- umn on the disk two rows of flattened, elongated calli are sit- uated. Column about 3 mm long, fleshy; ovary pubescent. Spur cylindric, or subscrotiform, 4 to 5 mm long. Pollinia 8 (4 large, 4 small). Luzon, Benguet Subprovince, Pauai, Elmer D. Merrill ^756, November 8, 1905, terrestrial in mossy forest along the trail, about 2,050 m above sea level: Province of Tayabas, Mount Banahao, H. N. Whitford 960, at 1,700 m above sea level: Lepanto Subprovince, Mount Data, Elmer D. Merrill iS82, November 4, 1905, in damp shaded ravine, mossy forest, about 2,100 m above sea level. The specimen collected on Mount Banahao, by Whitford, has broader leaves than the type (5 cm), but appears to be similar in the details of the flower. • Calanthe Ramosii sp. nov. Planta foliosa. Folia oblongi-lanceolata, acuminata, acuta, nervosa, in petiolum vaginantem angustata. Scapus tomentosus, elongatus, bracteis bene appressis, obtusis, vaginantibus, tubula- ribus. Bracteae inflorescentiae deciduae. Racemus laxus, pedi- celliis cum ovario pubescens. Sepala lateralia extus pubescen- tia, anguste lanceolata, acuminata. Sepalum dorsale simile. Petala lineari-lanceolata, acuta. Labellum 3-lobatum. Lobi laterales acuti. Lobiis medius valde emarginatus, apiculatus, crenulatus. In disco ultra columna utroque callus carinatus. Calcar o. Plant leafy, from stout elongated roots. Stem thickened at base, attenuated, about 15 cm long, terminated by 3 to 5 leaves. Leaves oblong-lanceolate, acuminate, acute, about 2 dm long, up to 3 cm wide, plicate, sheathing, conspicuously 5-nerved, with numerous less conspicuous nerves. Scape longer than the leaves. NOTES ON PHILIPPINE ORCHIDS, IV. 13 finely pubescent above, with several, tubular, obtuse, sheathing, closely appressed bracts about 2 cm long. Raceme loosely flow- ered, bracts deciduous. Pedicels slender, finely pubescent, 1 cm long. Ovarij pubescent, about 8 mm long. Color of the flowers not noted. Lateral sepals 11 mm long, 3 mm wide, narrowly lanceolate, acuminate, pubescent. Upper sepal similar to the laterals and about equal to them. Petals 8 mm long, 2 mm wide, linear-lanceolate, acute, smooth. Labellum 7 mm long, 3-lobed. Lateral lobes directed forward, 3 mm long, acuminate, acute, margin slightly crenulate. Middle lobe flabellate, deeply emar- ginate, apiculate, margin crenulate or blunt-toothed. Disk smooth with two semi-elliptic carinae one on each lateral nerve in front of the column, fleshy, pubescent. Spur none. Luzon, Province of Zambales, Maximo Ramos, Bur. Sci. A987, December 14, 1907. EULOPHIA R. Br. Eulophia Vanoverberghii sp. nov. Planta gracilis. Folia ? Racemus laxus, bracteae lineares. Pedicelli graciles. Sepala lateralia ligulato-oblonga, obtusa. Sepalum superius simile. Petala anguste elliptica. Labellum 3-lobatum; lobi laterales baud conspicui, minuti, obtusi, lobus medius oblongus, obtusus. Discus papillosus, ultra columnam utroque callus carinatus. Plants slender, 2 to 4 dm tall from a large corm. Scape sheathed at the base by several, nervose, tubular, imbricating bracts, near the middle 1- or 2-bracteate. {Leaves absent at flowering time, not seen.) Raceme loosely flowered, 10 to 15 cm long. Bracts of the raceme linear, about 1 cm long, about equaling the pedicellate ovary, erect or slightly spreading. Pedicels slender, smooth, about 7 mm long ; ovary smooth, about 5 mm long at the period of anthesis. Floivers about 1 cm long, erect. Lateral sepals 1.4 cm long, 3.5 mm wide, ligulate- oblong, obtuse, green. Upper sepal similar to the laterals, slightly shorter. Petals 1 cm long, about 5 mm wide, narrowly elliptic, rounded at the tip, green. Labellum 1.3 cm long, 7 mm wide at the tip, oblong, very obscurely 3-lobed, the lateral lobes minute and inconspicuous, margin wavy or fluted, nerves papillose, disk bilamellate at base. Sac obtuse, 1 mm long. Column about 11 mm long. Pollinia 2. The labellum is de- scribed by the collector as being yellow and all the perianth organs as being marked^jwith red lines. Luzon, Bontoc Subprovince, Father M. Vanoverbergh 336, April 2, 1910. “Root edible;” found 1,350 meters above sea level on hillocks. 14 AMES. DENDROBIUM Sw. Dendrobium albayense sp. nov. (§ Aporum, Hemiphylla) . Cum D. cuneato Schltr. comparandum videtur, a quo tamen foliis differt. Caulis ad basim teres, foliosus, compressus, anceps. Folia disticha, lanceolata a latere visa. Sepala late- ralia triangulari-ovata, obtusa. Sepalum superius oblongum obtusum, subquadratum. Petala ligulata, obtusa. Labellum cuneatum ad apicem 3-lobatum. Lohi laterales divaricati, obtusi. Lobus medius obtusus. Prope apicem labelli callus bilobus. Mentum brevem, obtusum, rotundatum. Plants up to 3 dm in length, simple or -with a few branches, leafy to about the middle, floriferous portion of the stem leafless. Leaves distichous, lanceolate, acute, when viewed from the side, 1.5 to 3 cm long from tip to point of insertion; at the middle of the leafy part of the stem 1.5 cm long from tip to inner angle formed with the stem; free portion up to 1 cm wide. Lateral sepals fleshy, broad, triangular-ovate, subacute or obtuse, about 3.5 mm long, 2.5 mm wide. Upper sepal fleshy, 3 mm long, 2.5 mm wide, rounded at the tip. Petals ligulate, obtuse, 3 mm long, 1.5 mm wide. Labellum equaling the lateral sepals, 3.5 mm wide at the tip, obscurely 3-lobed in front, lateral lobes divaricate, obtuse, about 1 mm long to the angle formed with middle lobe; middle lobe shorter than the laterals, obtuse, forming with the cuneate-bilobed callus above it a shallow pocket. Mentum rounded, about 1.5 mm long. Column 1 mm long. Luzon, Province of Albay, A. Loher 6018, June 26, 1905, plant flowering in Manila. “Flowers odorless, petals with purple lines.” Dendrobium angustifolium (Blume) Lindl. Gen. et Sp. Orch. 76. Desmotrichum angustifolium Blume Bijdr. 330; Kranzlin in Engl. Pflanzenreich 45 (1910) 350. The material which I refer to this species exhibits Several slight variations from authentic specimens of D. angustifolium. These slight variations seem to me unimportant. I have exam- ined living material collected by Elmer D. Merrill on Mount Mariveles in 1905. Luzon, Bontoc Subprovince, Father M. Vanoverbergh 708, 709, July 23, 1910: Province of Bataan, Lamao River, Mount Mariveles, R. S. Williams 67U'. Province of Rizal, San Isidro, M. Ramos, Bur. Sci. 12123, June 24, 1910. Dendrobium basilanense sp. nov. Caulis compressus, anceps. Folia disticha, lanceolata, acuta. Flores albi, terminales. Sepala lateralia, oblonga, acuta. Sep- NOTES ON PHILIPPINE ORCHIDS, IV. 15 alum swperius lanceolatum. Petala lineari-lanceolata. Label- lum lineari-oblongum, ad apicem bilobum. In leafage and in the bilobed tip of the labellum similar to Dendrobium aloifolium Reichb. f. Stems simple or branching, 1 to 4 dm tall or even taller. Leaves distichous, extending along the stem from the base to the summit, gradually diminishing in size, from the point of insertion of the sheathing base to the tip 1 to 2 cm long (average somewhat exceeding 1.5 cm), free portion of the leaf lanceolate, acute, about 1 cm long from inner angle to tip. Inflorescence terminal, floriferous part of the stem up to 2 cm long. Pedicels arising from small heads of minute bracts. Pedicel and ovary 3 mm long. Lateral sepals 3 mm long, oblong, acute. Upper sepal 2.75 mm long, lanceo- late. Petals linear-lanceolate, 3-nerved, obtuse or subacute, 2.25 mm long, slightly less than 1 mm wide. Labellum 3 mm long, 1 mm wide at base, linear-oblong, slightly dilated below the middle, above the middle contracted, the tip deeply retuse, or cleft, the lobes 1 mm long, obtuse. At the point where the lip is contracted, a transversely situated fleshy callus occurs. Mentum obtuse, about 2 mm long. Column 1 mm long. Basilan, W. I. Hutchinson, For. Bur. 3968, January 23, 1906. Dendrobium basilanense is nearly allied to D. aloifolium, from which it is to be distinguished by the linear-oblong labellum, scarcely, if at all, broader near the middle than at the tip. In his monograph of the Dendrobiinae ® Doctor Kranzlin cites D. Merrillii Ames Orch. 2: 181, as a synonym of D. aloifolium. I believe this is clearly the result of an error in judgment. D. Merrillii is unlike D. aloifolium in leafage and has an entire or obscurely retuse middle lobe of the labellum. With the original description of D. Merrillii mention was made of two collections, one Merrill 3357, and one Loher 6017. On the sheet of Loher’s specimen in my herbarium I find in A. A. Eaton’s handwriting “Not aloifolium." This note was made when Eaton and I were comparing the Philippine species of § Aporum with the specimens preserved at the British Museum of Natural History and at Kew. Doctor Kranzlin refers to Merrill 3357 in the geographical notes under D. aloifolium and from the exclamation mark after the number it would seem that he had examined the type. Yet, he makes no revision in the description of D. aloifolium which accounts for the inclusion of material which has a three-lobed labellum with the apical lobe oblong, and inconspicuously, if at all, retuse at the tip. D. merrillii Ames is also given by Kranzlin ^ under his “Species Dendrobii imperfecte notae.” Dendrobium Brongniartii Kranzl. in Engl. Pflanzenreich 45 (1910) 210. This is a robust species of § Aporum. The flowers are de- scribed by collectors some as white, some as yellow, others as ^ Engl. Pflanzenreich 45 (1910) 206. ‘ 1. c. 299. 16 AMES. reddish, but these differences are very likely the result of difference in age of the flower; the color turning from white to brown as it matures. In all cases I have been unable to identify the material at hand by an analysis of a flower, but where an examination has been possible the perianth organs agree with Kranzlin’s description. In a specimen collected by W. S. Lyon I was able to And perfect pollen masses. These are four in number. Luzon, Province of Zambales, Mount Pinatubo, F. W. Foxworthy, Bur. Sci. 3029, June 13, 1907, on stunted trees in small rocky desert, flowers white and fragrant, bracts yellow; Bur. Sci. 2617, April 26, 1907, on plain of coarse sand on trees: Province of Rizal, Ahern’s collector. For. Bur. 330^, September, 1905; Bosoboso, M. Ramos, Bur. Sci. 5639, November 30, 1909, fls. pure white, bracts yellow; Montalban, W. Schultze, Bur. Sci. 5620, May 25, 1908, flower white, fragrant, basal bracts yellow: Province of Cagayan, H. M. Curran, For. Bur. 168^i, March 9, 1909: Province of Tayabas, H. M. Curran, For. Bur. 9651, March 22, 1908: Province of Camarines, H. M. Curran, For. Bur. 11345, May 10, 1908. Mindanao, Province of Surigao, F. H. Bolster 378, August 15, 1906, epiphyte on trees, flowers yellow: District of Davao, Malalag, E. B. Copeland 692, March 28, 1904, ephiphyte, and on rocks over the sea: District of Zamboanga, H. N. Whitford and W. I. Hutchinson, For. Bur. 9207, January, 1908, altitude above the sea 150 meters, flowers white. Basilan, W. I. Hutchinson, For. Bur. 4018, February 10, 1906, growing on tree near beach, flower brownish- yellow. Dendrobium Clemensiae sp. nov. Caules uniarticulati, monophylli. Folium oblongum. Sepala petalaque ad basim triangularia in caudas longas producta. La- bellum 3-lobatum costis 2 usque ad basim lobi medii percurren- tibus. Lobi laterales obtusi, parvi. Lobus medius linearis ad apicem dilatatus, acutus, medio papillosus. Stem uniarticulate, somewhat stouter at the base than at the tip, rugose when dry, 7 to 14 cm long, monophyllous. Leaf narrowly oblong, fleshy, rigid, unequally bilobed at the tip, 11 to 15 cm long, 11 to 14 mm wide. Peduncle slender, weak, about 5 cm long, 1-flowered, from the summit of the stem, ter- minal, sheathed by an elongated spathe 3.5 to 4 cm long. / ate- ral sepals about 5 cm long, triangular at the base, prolonged into a long, slender. Aliform tail. Upper sepal similar to the laterals, about 5 cm long. Petals similar to the sepals, but shorter and more slender, about 4.5 cm long. Labellum up to 1.8 cm long, 3-lobed, suborbicular at base, abruptly continued from between the small, obtuse, lateral lobes into a linear, papil- lose limb which terminates in a lanceolate, membranaceous NOTES ON PHILIPPINE ORCHIDS, IV. 17 lamina; orbicular base about 5 mm long, 6 to 7 mm wide, with a membranaceous, sulcate lamina on each side of the middle nerve which extends beyond the lateral lobes, the linear papil- lose portion about 1 cm long, lamina about 4 mm long, 1.5 mm wide. Foot of the column 4 mm long. Mindanao, Lake Lanao, Camp Keithley, Mary Strong Clemens 508, April, 1906: Negros, W. S. Lyon UO, December, 1907, “epiphyte, pedicel green, base of perianth pure white, prolongations of sepals and petals lemon-yellow.” This species, which is closely allied to Dendrobium filiforme J. J. Smith, belongs to the section which, in his recent monograph. Doctor Kranzlin has established as the genus Diplocaulobium, characterized by a uniarticulate stem, 3-lobed crested labellum, and ephemeral flowers. Dendrobium longi- colle Lindl., another species of this section, has been reported from the Philippines, but without sufficient authority. Dendrobium iunatum Lindl. in Journ. Linn. Soc. Bot. 3 (1859) 4. In my preceding paper,' I referred Dendrobiu^n Iunatum Lindl. to the synonymy of D. indivisum Miquel. Further studies have convinced me that this proceeding was a matter of mistaken judgment. From the material referred by me to D. indivisum, D. Iunatum differs slightly in foliage and also slightly in the labellum. The leaves are not so spreading as in D. indivisum, their tips pointing more exactly in the same direction as the stem. The labellum of D. Iunatum is not 3-lobed, although in my specimens it is hardly so reniform or lunate as shown in the sketch preserved with the Cuming specimen from the Philippines in Lindley’s Herbarium. The flowers of both species are whitish or yellowish with purple streaks. According to Kranzlin’s recent monograph Dendrobium indivisum Miquel (not Bl.) and D. distichum (Presl) Reichb. f., are conspecific. The specimens previously referred by me to D. indivisum Miquel belong to D. distichum (Presl) Reichb. f. The specimens of D. Iunatum which I have examined come from Palawan, F. W. Foxworthy, Bur. Sci. 907, May 6, 1906. Dendrobium MacGregorii sp. nov. (§ Aporum, Hemiphylla) . Caulis ad basim teres, foliosus, compressus, anceps. Folia disticha, lanceolata, a latere visa lanceolata. Sepala lateralia triangulari-lanceolata, subacuta. Sepalum superius oblongi- ovatum, obtusum. Petala lineari-oblonga, obtusa. Labellum cuneato-spathulatum, ad apicem trilobatum, lohi rotundati, ob- tusi, laterales obscuri, medius obtusus. Mentum obtusum. Plants (somewhat resembling D. teloense J. J. Smith, but with very diiferent leaves) leafy to about the middle, slender and nearly terete at the base, about 5 dm tall. Leaves about 3.5 cm long from the tip to point of insertion, 1.5 to 2 cm long from tip to angle formed by the inner edge of the leaf and “ This Journal 6 (1911) Bot. 52. 108801 2 18 AMES. the stem, free portion lanceolate, acute, about 1 cm wide meas- ured perpendicularly. At about the middle of the stem the leaves suddenly decrease in size and give place to bract-like leaves 5 to 10 mm long, where they are free, which are closely appressed, their tips erect and nearly touching the stem. From the bases of these bracts, at intervals, the flowers break through. Flowers pale-yellowish-white with two lines of dark-carmine on the lip. Lateral sepals 3 mm long, triangular-lanceolate, subacute. Upper sepal oblong-ovate, obtuse, 3 mm long. Petals linear-oblong, obtuse, about 3 mm long, slightly less than 1 mm wide. Labellum cuneate-spathulate, 3-lobed in front, the side lobes rounded, the middle lobe half-elliptic, obtuse, rounded (in dried specimens the margin of each lateral lobe appears to have been stained with carmine). In general outline the lip resem- bles that of D. Merrillii, and appears to be minutely papillose on the disk near the base of the middle lobe. Mentum obtuse, 2 mm long. Column 2 mm long. POLILLO, R. C. McGregor, Bw. Sci. 10U57, epiphytic in mangrove swamp, October 12, 1909. Dendrobium parciflorum Reichb. f. ex Lindl. in Journ. Linn. Soc. Bot. 3 (1859) 4. D. Jenkensii Griff, in Calcutta Journ. Nat. Hist. (1854) 367, t S5. D. kcntrophyllum Hook. f. FI. Brit. Ind. 5 (1890) 725; Icon. Plant, t. 2021. D. marivelense Ames Orchidaceae 2 (1908) 180. The above is the bibliographical history of the species which in my second volume of “Orchidaceae” I described as a novelty under the name D. marivelense. In his monograph of the “Dendrobiinae” Doctor Kranzlin concluded that D. parciflorum and D. marivelense should be regarded as conspecific. In addition to the locality given in Orchidaceae 2: 180, the following is of interest: Mindanao, Lake Lanao, Camp Keithley, Mary Strong Clemens 611, June, 1906. Dendrobium Ramosii sp. nov. Caules gracilis, gerentes folia circiter 13. Folia linearia, acuta in anthesi decidua. Flores pallide rosei vel purpurei, in pedun- culis brevibus. Sepala lateralia oblonga, obtusa, ad apicem cucul- lata. Sepalum superius oblongum, obtusum. Petala ovato- oblonga, denticulata. Labellum cuneato-obovatum, unguicu- latum, prope basin callus hippocrepiformis. Caules foliosi et florigeri simul exstant. Plant very slender, about 3 dm tall, leafy, about 2 mm thick near the base. Leaves distichous, linear, acute, about 7 cm NOTES ON PHILIPPINE ORCHIDS, IV. 19 long, about 6 mm wide, base sheathing. Sheathing bases of the leaves tubular, in dried specimens tinged with purple, about 1 cm long. Flowers pale-purplish, in short lateral racemes. Peduncle 1 cm long. Bracts lanceolate, purplish, 2 mm long. Pedicellate ovary about 1 cm long, pedicel very slender. Flowers about 5, about 1 cm long, with a sharply curved mentum. Late- ral sepals oblong, obtuse, somewhat cucullate at the tip, 3 mm wide. Upper sepal cucullate, lanceolate, 6 mm long. Petals ovate-oblong, rounded at the tip, 6 mm long, 3 mm wide, margin minutely, denticulate. Labellum 12 mm long, 6 mm wide near the tip, simple, cuneate-obovate, sharply bent at the base to conform to the curved (genuflected) mentum, finely denticulate along the margin of the rounded, broad tip. In front of the column-foot the lip is dilated (nearly orbicular), then contracted at the point where a hippocrepiform membranaceous callus is situated, beyond the callus the limb is cuneate obovate, with three prominent longitudinal central nerves. In some flowers the mentum is so much curved that the blunt apex points in the same direction as the sepals and petals. Luzon, Lepanto Subprovince, Maximo Ramos, Bur. Sci. 70^9, January, 1909. Dendrobium Ramosii belongs in the same section with D. O’Brienianum Kranzl., and D. Epidendropsis Kranzl., two Philippine species which are characterized by the sharply bent mentum. From both it is very distinct in foliage and in the color and size of the flowers. I have seen two other collections from the Philippines which appear to be referable to this species. The leaves and stems are longer than in the specimens gathered by Ramos, but the flowers are very similar, the lip varying slightly in outline. Luzon, Bontoc Subprovince, Banco, Father M. Vanoverbergh 9, 1,300 meters above sea level, January 8, 1910: Benguet Subprovince, Bugias, Elmer D. Merrill U682, October 28, 1905. Dendrobium ventricosum Kranzl. in Engl. Pflanzenreich 45 (1910) 209. Luzon, Province of Zambales, A. Loher 600i, June 11, 1905, flowers greenish-yellow, with purple lines. Mindanao, Butuan Subprovince, Agusan River, Talacogon, C. M. Weber 176, epiphyte, flower light-green, July 20, 1911: District of Davao, Rev. R. F. Black, August, 1906, “flowers white, blue-veined inside;” from same locality. Black, 1909 Apparently a very variable species. The material which I have examined would seem to indicate that the labellum varies considerably in the condition of the anterior margin which may be minutely and inconspicuously denticu- late or almost fringed. In the specimen collected by Rev. R. F. Black the margin of the lateral and middle lobes is lacerate-denticulate. My understanding of Doctor Kranzlin’s D. ventricosum is based on his original description; I have not seen the type. 20 AMES. ERIA Lindl. Eria ( Mycaranthes) Vanoverberghii sp. nov. Planta robusta, erecta, foliosa. Folia ligulata, coriacea, acuta, disticha. Pedunculi terminale, circiter 6 bracteae imbricatae. Bracteae inflorescentiae lineari-lanceolatae, acutae, circiter 1.5 cm longae. Pedunculus rhachisque sericeo-pubescens. Pedicel- Im et ovarium pubescentia. Sepala externe pubescentia. Sep- ala lateralia oblonga, 5-nervia. Sepalum dorsale oblongi-lan- ceolatum. Petala oblonga, obtusa. Labellum 3-lobatum. Lobi laterales obtusi ; lobus medius quadratus, apiculatus, callus parvus ad basim labelli ; prope apicem labelli tuberculum permag- num ; utroque in disco callus carinatus lobis lateralibus parallelus. Plant robust, erect, probably 10 dm high (only upper part seen), leafy. Leaves ligulate, coriaceous, acute, sheathing, dis- tichous, about 20 cm long, up to 22 mm wide. Inflorescences several (as many as 6) sheathed by large, imbricating bracts. Peduncles rather stout, about 2 dm long, whitish-pubescent. Floral bracts linear-lanceolate, acute, crowded, reflexed, the lower- most exceeding 2 cm in length, about 5 mm wide, longer than the densely pubescent flowers. Lateral sepals oblong, acute, 5- nerved, densely pubescent on the exterior surface, 7 mm long, 3.5 mm wide. Upper sepal oblong-lanceolate, 7 mm long, 3 mm wide, 5-nerved. Petals oblong, or nearly elliptical, obtuse, 5 mm long, about 2 mm wide. Labellum 3-lobed, 6 mm long. Lateral lobes triangular, obtuse, from the tip to base of sinus formed with middle lobe 1.5 mm long, middle lobe subquadrate, apiculate, 3 mm long, 4.5 mm wide. Calli 4, one in the middle of the base, one on each side near the sinus formed by the middle and lateral lobes, each continued from the free forward protruding apex to near the base of the lip, passing into a gradually diminishing keel, one at the apex, a large triangular, obtuse, erect tooth-like callus, 2 mm high, laterally flattened with the apex slightly re- flexed. Column very short, free portion almost none. Luzon, Bontoc Subprovince, Father M. Vanoverbergh 789, August 27, 1910, 1,550 meters above sea level. Eria Vanoverberghii is near Eria longibracteata Leavitt, but is unlike it in the dense inflorescence, large flowers, and very different calli. PHREATIA Lindl. Phreatia (§ Euphreatia) infundibuliformis sp. nov. Folia linearia. Pedunculus valde elongatus, prope basim fractiflexus, bracteae infundibuliformes, magnae. Racemus elongatus, gracilis. Bracteae inflorescentiae lineari-lanceolatae, acutae, ovarium pediceUum aequantes. Sepala lateralia ovato- NOTES ON PHILIPPINE ORCHIDS, IV. 21 lanceolata. Sepalum superiiis oblongi-Ianceolatum. Petala oblonga, angustata, obtusa. Labellum breviter unguiculatum ; lamina cordatum, breviter apiculata. Columna minuta. Leaves linear, much shorter than the peduncle, about 9 cm long, 5 to 7 mm wide, asymmetrical at the tip. Peduncles elongated, somewhat zig-zag below the raceme, the angles marked by the large infundibuliform bracts, 4 or 5 in number. Bracts about 1 cm long. Peduncle 14 to 27 cm long. Raceme 8 to 14 cm long, 5 to 7 mm through in fruiting specimens. Bracts of the raceme linear-lanceolate, 3 to 7 mm long, the lowermost ones equaling or exceeding the flowers on fruiting specimens. Flowers diminutive, white. Lateral sepals ovate-lanceolate, 1 mm long. Upper sepal oblong-lanceolate, about 1 mm long Pet- als narrowly oblong, obtuse, 1 mm long, 0.5 mm wide. Labellum 0.75 mm long, cordate, shortly apiculate, 3-nerved, very shortly and inconspicuously unguiculate. Mentum minute. Column diminutive. Mindanao, Province of Misamis Mount Malindang, Major E. A. Meams 6 W. I. Hutchinson, For. Bur. 4611, May, 1906. Plants abundant at an altitude of 1,700 meters above sea level. From nearly related Philippine species readily distinguished by the large infundibuliform bracts below the raceme. In habit closely resembling P. myosurus (Forst.) Ames. Phreatia (§ Euphreatia) Mearnsii sp. nov. Habitu P. densiflorae baud dissimilis. Folia ligulata. Race- nius densiflorus. Braeteae inflorescentiae lanceolatae, acumina- tae. Sepala lateralia triangulari-ovata. Sepalum superius simile. Petala oblongi-ovata, obtusa. Labellum trapezoideum. In habit similar to P. Vanoverberghii and P. densiflora. Leaves ligulate, asymmetrical at the tip, up to 25 cm long, 1.6 to 2.5 cm wide, coriaceous. Peduncle up to 4 dm long, exceeding the leaves, with several lanceolate sheathing bracts below the raceme. Bracts about 2 cm long, the upper ones about 1 cm long. Raceme about 1 dm long (1 cm through when the cap- sules are nearly ripe). Floral bracts lanceolate, long-acuminate, 3.5 cm long, spreading, subrigid. Flowers yellow, numerous, crowded. Lateral sepals triangular-ovate, 1.5 mm long, sub- acute. Mentum 0.5 mm long. Upper sepal 1 mm long. Petals oblong-ovate, obtuse, 1.5 to 2 mm long, 1.5 mm wide at about the middle. Capsule ellipsoid, 3 mm long. Pedicel and ovary in fruiting specimens about 5 mm long. Mindanao, Province of Misamis, Mount Malindang, E. A. Mearns & W. I. Hutchinson, For. Bur. 4592, May, 1906. Luzon, Province of Pam- panga, H. M. Curran, For. Bur. 19313, March, 1910. 22 Phreatia Meamsii is nearly related to P. aristulifera Ames and P. Vanoverberghii Ames, but differs from them in details of the flowers. The petals of P. Meamsii are not unguiculate nor in any sense rhombic in outline. Phreatia (§ Euphreatia) Ramosii sp. nov. Planta parvula. Folia linearia. Pedunculus folio longitudine fere aequalis vel longior. Flores albi. Sepala lateralia trian- gulari-lanceolata, obtusa. Sepalum superius ovatum, obtusum. Petala ovata, ad apicem rotundata. Lahellum oblanceolatum vel obovatum. Mentum obtusum. In habit similar to Phreatia myosurus (Forst.) Ames. Plants about 1 dm tall. Leaves linear, asymmetrical at the tip, 4 to 9 cm long, up to 5 mm wide, spreading, equaling the peduncle or slightly shorter. Peduncle slender, 5 to 10 cm long, bracts sheathing, about 5 mm long, acuminate, aristate. Raceme 2 to 4 cm long, about 7 mm through when the flowers are open. Flowers white, diminutive, about 1 mm apart. Floral bracts rounded, prolonged at the tip, aristate, 3 mm long, aristate tip 1.5 mm long. Pedicel and ovary 3 mm long. Lateral sepals about 2 mm long (forming a blunt mentum 0.5 mm long), triangular-lanceolate, obtuse. Upper sepal 1.75 mm long, broadly ovate, obtuse. Petals ovate, rounded at the tip, 1.5 mm long, about 1 mm wide. Lahellum 2.5 mm long, oblanceolate or obovate, about 1 mm wide at the middle. Column simple, Luzon, Province of Abra, Maximo Ramos, Bur. Sci. 7140, February 8, 1909. Epiphyte, found on small trees in forests. Phreatia sulcata J. J. Smith. In a preceding paper ° I described a new species from the Philippines as P. bracteata. An examination of additional material from Luzon convinces me that the specimens cited under P. bracteata are conspecific with P. sulcata J. J. Smith. A series of specimens from Bontoc Subprovince, Luzon, collected in June, 1911, by Father M. Vanoverbergh, No. 1289, are most certainly referable to Phreatia sulcata and compare well with specimens from Sumatra. Phreatia (§ Euphreatia) Vanoverberghii sp. nov. Habitu P. densi florae (Bl.) Lindl. baud dissimilis sed in foliis angustior et in racemo gracilior. Folia linearia, Racemus densiflorus. Bracteae inflorescentiae aristatae. Sepala late- ralia ovata. Sepalum superius ovatum, obtusum, concavum. Petala unguiculata, rhombico-spathulata. Lahellum unguicu- latum, ad basim cuneatum, super medium dilatatum, ad apicem in laminam subrhombicum dilatatum. This Journal 4 (1909) Bot. 674. NOTES ON PHILIPPINE ORCHIDS, IV. 23 Leaves linear, asymmetrical at the tip, 20 to 26 cm long, 1,5 to 1.7 mm wide. Peduncle up to 24 cm long, with several closely appressed bracts below the inflorescence. Raceme slen- der, about 1 cm long, very densely flowered, about 6 mm through prior to development of the capsules. Floral bracts equaling the flowers, 2.5 mm long, cuneate at the base, rounded, prolonged into a conspicuous aristate tip (cf. P. aristulifera Ames). Pedicel and ovary of expanding flowers 2 mm long. Lateral sepals about 1 mm long, ovate, forming a mentum 0.5 mm long. U^iper sepal ovate from a broad base, 1 mm long, obtuse, about 1 mm wide. Petals cuneate at the base, roundish, obtuse, or rhombic above the base, 1.5 mm long, 0.75 mm wide near the tip. Labellum slightly less than 2 mm long, unguicu- late, cuneate, dilated, above the middle elliptical or subrhombic, rounded, margin entire, about 1 mm wide near the tip. Column 1 mm long. Pollinia 8. Luzon, Bontoc Subprovince, Father M. Vanoverbergh 1115, on trees, altitude 1,650 meters, February 21, 1911. Here also belongs a series of specimens with leaves 3 dm long by 13 mm wide collected in Luzon, Province of Abra, by Maximo Ramos, Bur. ScL 714-5, altitude about 600 m, February 8, 1909. In habit similar, in a very general way, to P. densiflora (BL) Lindl., from which it differs in the aristate floral bracts and very slender raceme. BULBOPHYLLUM Thou. Bulbophyllum lancipetaium sp. nov. Pseudobulbi minuti. Folia elliptico-lanceolata, coriacea, acuta, prominente 1-nervia. Scapus foliis brevior. Racemus strictus. Bracteae inflorescentiae ovariis longiores lanceolatae, acumi- natae, acutae. Sepala lateralia triangulari-lanceolata, ad apicem aristata. Sepalum dorsale simile, cymbiforme, acuminatum, acutum. Petala lineari-lanceolata, obtusa, pubescentia in mar- gine. Labellum lingulatum ciliolatum. Columna stelidiis 2. Rhizome stout, about 5 mm in diameter. Pseudobulbs small, about 1 cm long, pyriform. Leaf elliptic-lanceolate, 12 cm long, about 2.5 cm wide, coriaceous, acute, prominently nerved along the middle, tapering at base into an elongated petiole. Petiole 5 to 10 cm long, sulcate, rigid. Scape shorter than the leaves, up to 17 cm tall, sheathed by about five, loose, acute bracts 9 to 13 mm long. Raceme strict, spicate, 6 to 7 cm long, about 20-flowered. Bracts of the raceme surpassing the ovaries, lan- ceolate, acuminate, acute, the largest one about 5 mm long. 24 AMES. Lateral sepals triangular-lanceolate from a broad base, 7 mm long, tapering into a slender bristle-like tip, at base, in front protuberant, forming a rounded mentum. Upper sepal shorter than the laterals, about 6 mm long, cymbiform, acuminate, acute. Petals linear-lanceolate, obtuse, 4 mm long, 1-nerved, somewhat thickened at the tip, very finely pubescent. Lahellum Ungulate, about 4 mm long, margin ciliolate, strongly deflexed near the middle, base rotund when spread flat. Column minute with two erect stelidia. Mindanao, Province of Misamis, Mount Malindang, Major E. A. Mearns & W. I. Hutchinson, For. Bur. i608, flower yellow-white, 2,400 m above sea level, June 4, 1906. In general habit similar to B. dasypetalum, but very distinct from that species in the details of the inflorescence. The specimen from Mindanao, Mearns & Hutchinson, For. Bur. JtGOl, referred tentatively to B. adenopetalum Lindl.' appears to belong here. From B. adenopetalum it differs in its larger leaves and in the absence of caudate tips to the lateral sepals. According to the brief description of B. braccatum given by Reichenbach f. in “Linnaea,” B. lancipetalum appears to be distinct from that species. In the Philippines there is a puzzling group of Bulbophyllums, all so closely allied that for proper interpretation a careful study of many specimens will be necessary. To this group belong B. adenopetalum and B. Ramosii characterized by caudate-tipped lateral sepals, B. dasypetalum and B. lancipetalum characterized by triangular- lanceolate lateral sepals, shortly if at all caudate, and B. braccatum which is said to have apiculate petals. At the present time, from material at hand, a thorough study of this group is being made which will be published in the near future. Bulbophyllum nutans (Lindl.) Reichb. f. in Walper’s Ann. 6: 260; Lindley in Bot. Reg. (1839) Misc. 118; Bot. Mag. t. ^^18. This interesting species of the § Cirrhopetalum has come to hand from the Island of Polillo. The leaves are very thick, elliptic or nearly orbicular, 3 cm long by 17 mm wide or there- abouts, very shortly petioled, arising from diminutive, rugose, pyriform pseudobulbs, 1 cm long. Scape 1.5 to 2 dm tall, slender, with several minute, closely appressed, acute bracts. Umbel many-flowered, nodding. Bracts of the umbel linear, about 3 mm long, crowded. Pedicel slender, together with the ovary 5 mm long. Lateral sepals ligulate, 2 cm long. Upper sepal 5 mm long, lanceolate, prolonged into a slender, bristle- like tip, 3-nerved. Petals lanceolate, acuminate, acute, 3-nerved, about 3.5 mm long, margin of the upper half entire, margin of ' Orchidaceae 2 : 206. NOTES ON PHILIPPINE ORCHIDS, IV. 25 the low^er half ciliated or serrulated. Labellum diminutive, Ungu- late, about 2 mm long, obtuse, fleshy, bicristate. POLILLO, C. B. Robinson, Bur. Sci. 9077, August 9, 1909, growing on Heritwra littoralis, buds deep-yellow, flowers somewhat paler. The plants flgured in the “Botanical Magazine” have longer leaves, and larger pseudobulbs than the specimens from Polillo. Bulbophyllum Ramosii sp. nov. Folia oblonga, ad apicem rotundata. Scapus gracilis, brac- teis tubularibus. Flores albi. Bracteae inflorescentiae lineares. Sepala lateralia triangularia, caudata. Sepalum superius conca- vum, lanceolatum, caudatum. Petala lineari-lanceolata, acuta, 1-nervia. Labellum lingulatum, ciliatum. Columna bifurcata. Plant small, to the tip of the leaf not exceeding 1 dm. Rhizome comparatively slender, rooting freely. Pseudobulbs v^anting, the leaves arising directly from the creeping rhizome. Leaves rigid, leathery, oblong, rounded at the tip, tapering gradually at the base, lamina up to 7 cm long, 11 mm wide, with a promi- nent mid-nerve. Scape weak, slender, longer than the leaves, up to 8 cm long, with several tubular bracts about 5 mm long below the raceme. Raceme lax, about 7-flowered. Flowers white. Bracts of the raceme linear, exceeding the pedicellate ovary, about 4 mm long. Lateral sepals triangular, prolonged into a slender caudate tip, triangular base about 3 mm long, caudate tip 5 mm long. Upper sepal shorter than the laterals, concave, lanceolate, caudate-tipped, 7 mm long. Petals linear- lanceolate, acute, 1-nerved, about 3 mm long. Labellum Ungu- late, strongly bent at the middle, apical half nearly 2 mm long, oblong, rounded at the tip, margin coarsely hairy, basal half sulcate, when spread out orbicular. Column at the tip bifurcate. Luzon, Province of Laguna, San Antonio, Maximo Ramos, Bur. Sci. 12091, August 26, 1910. This is a smaller species than Bulbophyllum adenopetalum Lindl., to which it is closely allied. The flowers resemble the sketch of B. adenopet- alum, by Bindley, preserved at Kew. The leaves are much shorter than in Lindley’s material and much broader in relation to their length. Both species are different from B. dasypetahirn Rolfe, in their caudate-tipped lateral sepals, and ciliolate lips. B. braccatum, a Philippine species described by Reichenbach f., belongs in the same group. Of this species, unfortunately, I have seen no material; no type is designated by its author. The descrip- tion is vagnie, no details regarding the foliage being given, and, in view of the fact that there is a group of species in the Philippines closely allied with B. adenopetalum and B. dasypetalum, wholly inadequate for purpose of identification and useless for comparative studies. 26 AMES. DENDROCHILUM Blume. Dendrochilum (§ Acoridium) longibulbum sp. nov. Pseudohulhi graciles, fusiformes, elongati. Folia linearia, obtusa. Peduncidus gracilis, filiformis, folio longior. Racemus 4 ad 5 cm longus. Bracteae inflorescentiae distichae. Sepala lateralia lanceolata, acuta. Sepalum superius lateralibus simile. Petala cuneato-ovata. Labellum crassum, 3-lobatum ; lobi latera- les semicrescentiformes ; lobus medius subquadratus, apiculatus, prope basim bicallosus, et callo medio vel mammilla instructus. Pseudobulbs densely clustered, elongated, cylindric, rugose, 2.5 to 4 cm long, 2 to 4 mm in diameter when dry. Leaf shortly petioled, linear, obtuse, 7 to 14 cm long, 5 to 7 mm wide, midnerve prominent. Peduncle filiform, slightly exceeding the leaf, up to 1 dm long. Raceme 4 to 5 cm long, loosely many-flowered (fls. about 15), subtended by 2 or 3, glumaceous, scarious, im- bricating, empty bracts; flowers white, 2 to 3 mm apart, dis- tichous, floral bracts exceeding and concealing the ovary. Lat- eral sepals lanceolate, acute, about 3.5 mm long, 1.5 mm wide, 3-nerved. Upper sepal similar, 3.5 mm long. Petals cuneate- ovate, acute, 3 mm long, about 1.5 mm wide, 3-nerved. Labellum fleshy, 3-lobed, 1.5 mm long; lateral lobes half crescent form, obtuse, shorter than the middle lobe; middle lobe 0.75 mm long, subquadrate with a triangular tooth in front. On the disk of the lip there are three calli, one in the middle at the bottom of the depressed or subsaccate base and one on each side near the base of each lateral lobe. Column typical of the section. Luzon, Bontoc Subprovince, Father M. Vanoverbergh 782, altitude 1,550 m, August 23, 1910. Dendrochilum longibulbum might readily be mistaken for a narrow-leaved form of D. pumilum, but its smaller flowers and very different petals are differentiating characters. Of the sixty or more species of this genus which have been described from Philippine material all are endemic ; I have been unable to discover a single exception. Of the described Philippine species I have examined every one except D. convallariaeforme Schauer (which is supposed to include D. Copelandii Ames) and D. maleolens Kranzl., the latter a recently described species which is said to be close to D. oliganthum (Ames) Pfitzer. Dendrochilum microchilum (Schlechter) Ames Orchidaceae 2 (1908) 87. Platyclinis microchila Schlechter in Bull. Herb. Boiss. II 6 (March, 1906) 302. Acoridium venustulum Ames in Proc. Biol. Soc. Wash. 19 (Sept., 1906) 147. Dendrochilum venustulum (Ames) Pfitzer in Engl. Pflanzenreich 32 (1907) 116. NOTES ON PHILIPPINE ORCHIDS, IV. 27 In my second volume of “Orchidaceae,” having no material and judging wholly from the original description, I referred to Platyclinis microchila as a near affinity of Dendrochilum tenue Pfitzer. In a letter dated Novem- ber 7, 1910, Doctor Schlechter states that Dendrochilum microchilum and D. venustulum Pfitzer are identical. Dendrochilum (§ Acoridium) Vanoverberghii sp. nov. Pseudobulbi pyriformes in sicco rugosi. Folia linearia, acuta. Pedunculus folio longior. Inflorescentia laxiflora. Bracteae glumaceae, ovario longiores. Sepala lateralia ovato-Ianceolata, acuta, Sepalum superius oblongi-lanceolatum, acutum. Petala lanceolata, acuta. Labellum crassum, 3-lobatum, lobi laterales obtusi, lobus medius apiculatus. Whole plant, including the peduncle, up to 24.5 cm high, as shown by the type. Pseudobulbs elongated pyriform, 1.5 cm long, about 5 mm in diameter at the base, rugose in dried speci- mens, clothed with the fibrous remains of sheathing bracts. Leaves linear, acute, 4.5 to 7,5 cm long, about 2.5 mm wide, grass-like when dry. Peduncle free, from the summit of the pseudobulb, slender, graceful, about twice longer than the leaf, about 1 dm high, not exceeding 13.5 cm in the type. Inflorescence loosely flowered, subtended by two imbricating, scarious bracts ; flowers about 3 mm apart, distichously arranged. Pedicel and ovary 2 mm long, concealed by a glumaceous bract. When fresh the flowers are white with a “red” lip, in dried specimens they are brownish. Lateral sepals ovate-lanceolate, acute, 3- nerved, 2.5 mm long, 2 mm wide near the base. Upper sepal oblong-lanceolate, acute, 2.5 mm long, 1 mm wide, middle nerve conspicuous (in dried specimens). Petals lanceolate, acute, 3- nerved, 2 mm long, 1 mm wide. Labellum fleshy, 1.5 mm long, somewhat saccate near the base, 3-lobed ; lobes subequal, laterals rounded, obtuse in front; middle lobe longer than the laterals, subquadrate, apiculate, broader than long. Column minute, characteristic of the Section Acoridium. Luzon, Bontoc Subprovince, Father M. Vanoverbergh, November-De- cember, 1910. Dendrochilum Vanoverberghii appears to be a near relative of D. exile' but differs from it in the details of the flower, and in general habit, the leaves and peduncle of D. exile being about subequal, while in D. Vanover- berghii the peduncle exceeds the leaf to a conspicuous extent. D. Mac- Gregorii is also a near relative but the lip-characters of that species and the dense inflorescence are conspicuous differentiating peculiarities. ‘Orchidaceae 2: 86 fig. The Philippine Journal of Science, C. Botany. Vol. VII, No. 1, April, 1912. ADDITIONAL PHILIPPINE SYMPLOCACEAE, II. By A. Brand. (Sorau, Germany.) Since the publication of my first paper under the above title ^ a considerable number of specimens of Symplocos have been collected in the Philippines, and in addition to the material secured through the medium of recent exploration, there is now preserved in the herbarium of the Bureau of Science, a set of specimens collected by Mr. A. Loher in Luzon in the year 1906, distributed from the Kew Herbarium. This accumulated mate- rial has been submitted to me by Mr. Merrill for study, and I have fortunately been able to examine the Philippine specimens distributed by Mr. A. D. E. Elmer, preserved in the Delessert Herbarium. Three new species of Symplocos described by Mr. Elmer ^ which were previously known to me only by description, are represented in the Delessert Herbarium by cotypes. I consider that two of his proposed species are valid, but the third, S. angnlaris Elm., I must consider to be a synonym of S. Cu- mingiana. It is not surprising that Mr. Elmer considered Sym- plocos angularis to be a distinct species, for Symplocos Cumingiana has exceedingly variable leaves, and specimens with small, somewhat coriaceous leaves look quite different from those with large chartaceous ones. Having now a large series of specimens of S. Cumingiana, a species previously rather imper- fectly known to me, I must confess that I erred in referring to it For. Bur. 825U Curran & Merritt ® ; I now consider that this number represents a distinct species, readily distinguished from S. Cumingiana by its bright leaves and purplish midribs. In the material examined two additional new species were discovered, so that the total number now known from the Archipelago has been increased to twenty-six. The new species ‘ This Journal 4 (1909) Botany 107-110. ^Leafl. Philip. Bot. 2 (1908) 508-610. 'L. c. 29 30 BRAND. all belong to the section Bobua, and accordingly the key to the species given in my previous paper must be changed as follows : 1. Inflorescentiae compositae. 2. Folia ramos adpressa, imbricata 3. S. imbricata 2. Folia patentia baud imbricata. 3. Corolla extus sericea. 4. Stamina ca. 100; folia basi valde angustata 4. S. patens 4. Stamina ca. 60; folia basi plerumque rotundata.... 5. S. floridissima 3. Corolla extus glabra. 4. Stamina ca. 100 6. S. polyandra 4. Stamina 25 ad 50. 5. Inflorescentiae glabrae 7. S. Hutchinsonii 5. Inflorescentiae pilosae. 6. Fructus globosus. 7. Inflorescentiae axillares 8. S. ferruginea 7. Inflorescentiae terminales - 9. S. Ahernii 6. Fructus ellipsoideus 10. S. adenophylla 1. Inflorescentiae simplices. 2. Flores sub foliis prominentes 11. S. oblongifolia 2. Flores in axillis foliorum. 3. Inflorescentiae terminales 12. S. imperialis 3. Inflorescentiae axillares. 4. Ramuli glabri. 5. Flores in axillis foliorum sessiles, fasciculati 13. S. Loheri 5. Flores spicati vel racemosi. 6. Flores racemosi. 7. Folia 1.5 ad 2.5 cm. lata 14. S. fragrans 7. Folia 3 ad 3.5 cm lata 15. S. peninsularis 6. Flores spicati. 7. Folia minus quam 3 cm (vel raro 3 cm) lata. 8. Inflorescentiae multo longiores quam petioli. 9. Folia chartacea 16. S. betula 9. Folia coriacea. 10. Nervus intermedius in facie inferiore foliorum flavidus 17. S. Whitfordii 10. Nervus intermedius in facie inferiore foliorum pur- purascens 18. S. purpurascens 8. Inflorescentiae vix longiores quam petioli. 19. S. curtiflora 7. Folia plus quam 3 cm lata 20. S. Cumingiana 4. Ramuli ferruginei vel pilosi. 5. Fructus inconspicuus, vix 2.5 mm in diametro. 21. S. inconspicua 5. Fructus 4 ad 5 mm longus. 6. Folia pleraque plus quam 4 cm longa. 7. Fructus cylindricus, pilosus 29. S. cagayanensis 7. Fructus ovoideus vel ovoideo-globosus, glaber. 8. Folia subtus ad costam dense rufo-setosa. 23. S. luzoniensis 8. Folia subtus ad costam glabra vel parce ciliata. 24. S. Merrilliana ADDITIONAL PHILIPPINE SYMPLOCACEAE, II. 31 6. Folia pleraque minus quam 4 cm longa. 7, Folia utrinque glaberrima 25. S. palawanensis 7. Folia plus minus pilosa 26. S. depauperata 1. Symplocos confusa Brand in Pflanzenreich 6 (1901) 88. Additional material: Negros, Province of Negros Oriental, Dumaguete, Cuernos Mountains, Elmer 9532, May, 1908; Canlaon Volcano, mossy forest on ridges, Merrill 6990, April, 1910. In these specimen the corolla is somewhat more hairy, than is usually the case. The species is now known from the Islands of Luzon, Negros, and Mindanao. 3. S. imbricata Brand in Philip. Journ. Sci. 4 (1909) Bot. 108. Descriptio aucta: Frutex vel arbor 3 ad 9 m alta, ramulis purpureis vel atro- purpureis, glabris. Folia crasse coriacea, ramulis adpressa et imbricata, 5 ad 6 cqji longa, 2.5 ad 4 cm lata, ovata vel late ovata, serrato-dentata, utrinque glaberrima, in apicem bTevem subito producta, basi leviter cuneata vel truncata, costa supra impressa; petiolus 5 ad 10 mm longus. Spicae terminales et subterminales, compositae, glabriusculae, juniores densae, vetus- tiores laxiflores, petiolo multo longiores, fructiferae incrassatae et elongatae; bractae 3, ovato-rotundatae, sericae, calycem oc- cultantes ; calycis tubus brevissimus, glaber, lobi rotundati, obtusi, sericei, tubo multo longiores, post deflorationem supra ovarium glabrum convergentes ; corolla alba vel lutea, glabra; stamina 50 ad 60; stylus glaber, calyce duplo longior. Fructus niger, ovato-ampulliformis, 10 ad 12 mm longus, valde rugosus, glaber, trilocularis, sed loculis binis plus minus abortientibus ; lobi calycini discum comose superantes. Additional material: Luzon, Benguet Subprovince, Mount Pulog, altitude 2500 m above sea level, Merrill 658i, May, 1909, with mature fruit. For. Bur. 18091 Curran, Merritt, & Zschokke, January, 1909; Mount Ugo, Bur. Sci. 5715 Maximo Ramos, December, 1908, in flower: Bontoc Subprovince, Banco, in forests, 1650 m above sea level Father M. Vanoverbergh lOOU, Feb.-Mar., 1911. 4. S. patens C. Presl Rel. Haenk. 2 (1831) 61, forma 1 eupatens Brand in Philip. Journ. Sci. 3 (1908) Bot. 5. Additional material: Luzon, Province of Abra, For. Bur. 1U57U Darling, February, 1909; Baco, 10 m above sea level. For. Bur. 14668a Darling, February, 1909. The specimen first cited has flowers less silky than is usual, and is a connecting form between S. patens and S. polyandra. Forma 2 ciliata (C. Presl) Brand in Philip. Journ. Sci. 3 (1908) Bot. 5 (S. ciliata C. Pres! 1. c.) 32 BRAND, Ramuli glabri. Petiolus ca. 3 cm longus. Luzon, Bontoc Subprovince, Bauco, in forests, altitude 1650 ra above sea level. Father M. Vanoverbergh 1300, June-July, 1911. A tree, 6 m high, now discovered by the American botanists for the first time. 5. 8. florid issima Brand in Pflanzenreich 6 (1901) 35, var. serrata Brand in Philip. Journ. Sci. 4 (1909) Bot. 108. Descriptio aucta : Arbor 10 m altus, ramulis atropurpureis glabris. Folia ten- uiter coriacea, 10 ad 13 cm longa, 4 ad 7 cm lata, ovalia vel elliptica, grosse serrata, utrinque glabra, in apicem brevem subito producta, basi nunc rotundata, nunc cuneata; petiolis ca. 2 cm longus. Paniculae puberulae, petiolo 4- ad 6-plo longiores, pedicellis calycem acquantibus vel superantibus; bracteae mini- mae ;s- calyx dense ferrugineus, lobis rorimdato-triangularibus tubum aequantibus ; corolla calyce duplo longior, 5-partita, extus sericea ; stamina ca. 60 ; stylus glaber, calyce triplo longior ; ova- rium dense sericeum. Fructus nigro-brunneus ca. 10 mm longus, ampulliformis. Additional material: Luzon, Province of Laguna, San Antonio, Bur. Sci. 12015 Ramos, August, 1910. The leaves are less serrate than in the type; a form connecting typical S. floridissima with the variety. 5. S. polyandra Brand in Pflanzenreich 6 (1901) 36. Additional material: Luzon, Province of Pangasinan, For. Bur. ISiSO Medina, February, 1909, the timber used in house construction. 8. S. ferruginea Roxb. var. philippinensis Brand in Philip. Journ. Sci. 3 (1908) Bot. 6. Additional material: Luzon, Benguet Subprovince, Baguio, Elmer 8761, March, 1907 : Province of Laguna, Cavinti, Loher 6231 ; San Antonio, Bur. Sci. 10995 Ramos, August, 1910 (a tree 12 m high, with white flowers) : Province of Nueva Ecija, For. Bur. 221 9 i Alvarez, December, 1910. Mar- INDUQUE, Mount Tayubao, altitude 600 m above sea level. For. Bur. 12168 Rosenbluth, April, 1908, tree 8 m high, diameter 15 cm, common name bocboc. This is the first Syrnplocos to be found on Marinduque Island. Known otherwise from Luzon and from Dinagat. 9. S. Ahernii Brand in Philip. Journ. Sci. 3 (1908) Bot. 6. Additional material: Luzon, Benguet Subprovince, Sablan, Bur. Sci. 12717 F&nix, November-December, 1910, in flower, common name chaniusiu: Prov- ince of Tayabas, Lucban, Elmer 7908, May, 1907. 13. S. Loher I Brand sp. nov. Arbor (?) glaberrima. Folia ad apices ramulorum congesta, subcoriacea, oblonga vel oblongo-lanceolata, 6 ad 7 cm longa, 1.5 ADDITIONAL PHILIPPINE SYMPLOCACEAE, II. 33 ad 2 cm lata, integerrima, breviter et sensim apiculata, basi in petiolum vix 1 cm longum, flavidum, subalatum, sensim et valde attenuata, costa supra impressa et caniculata, subtus prominula, flavida. Flores ignoti, in axillis foliorum sessiles, ad 5 fasci- culati (ex fructibus junioribus). Fructus oblongo-cylindricus, laete brunneus, ca. 10 mm longus, lobis calycinis flavescentibus discum occulantibus. Luzon, Province of Rizal, Angilog, Loher 6192, March, 1906. This is the first Syniplocos with fasciculate flowers to be found in the Philippines. Its closest ally seems to be Symplocos congesta Benth., of southern China. The number of Symplocos species with fasciculate flowers is not great. 14. S. frag pans Elmer, Leafl. Philip. Bot. 2 (1908) 508. Ramuli glabri purpurascentes. Folia juniora chartacea, vet- ustiora subcoriacea, elliptica vel oblonga, 4.5 ad 6.5 cm longa, 1.5 ad 2.5 cm lata, integerrima, utrinque glaberrima, in apicem longiusculum subito producta, basi cuneata, costa supra impressa, subtus prominula, flavida, petiolus 5 ad 8 mm longus, purpuras- cens, baud alatus. Inflorescentiae minutissime et adpresse hir- tellae, petiolo multo longiores, laxiflorae; pedicelli calyce multo longiores; bracteae lanceolatae, hirtellae; calyx minimus hirtel- lus, lobis ovatis tubo subbrevioribus ; corolla glabra calyce duplo longior; stamina ca. 50; stylus claviformis, crassiusculis, cum ovario hirtellus. Fructus mihi non visus. Negros, Dumaguete, Cuernos Mountains, Elmer 9873, April, 1908. The description is taken from the specimen preserved in the Delessert Herbarium. 17. S. Whitfordii Brand in Philip. Journ. Sci. 3 (1908) Bot. 8. Additional material: Luzon, without locality, Loher 6199: Province of Laguna, Mount Banajao, Loher 6201, 6207: Province of Tayabas, Lucban, Elmer 7U87, May, 1906, distributed as S. laneifolia; same locality, Elmer 7850, 9103: Benguet Subprovince, Mount Tonglon, altitude 1800 m above sea level, Merrill 8005, May, 1911, a tree 6 to 8 m high. Negros, Canlaon Volcano, Merrill 7031, Phil. PI. 239 Merrill, April, 1910, growing at an altitude of 2100 m above sea level. 18. S. purpurascens Brand sp. nov. Symplos Cumingiana Brand in Philip. Journ. Sci. 4 (1909) Bot. 110, non Brand in Pflanzenreich 6 (1901) 58. Glaberrima. Ramuli teretes purpurascentes. Folia coriacea, elliptica vel oblonga, 4.5 ad 6 cm longa, 2 ad 2.5 cm lata, breviter et obtuse apiculata, basi sensim in petiolum subalatam, purpuras- centem, 5 ad 10 mm longum attenuata, integerrima, utrinque 108801 8 34 BRAND. nitida, costa utrinque purpurascens, supra impressa, subtus pro- minula, nervi secundarii in foliis junioribus purpurascentes, in vetustioribus flavidi. Inflorescentia spicatae, petiolo 2- ad 4-plo longiores, pauci- et laxiflorae; bracteae minimae, ovato-triangu- lares ; calyx glaber, lobis rotundatis tubo subbrevioribus ; corolla glabra, calyce triplo longior; stamina (ex unico flore) 45; stylus crassiusculus, claviformis ; ovarium glabrum. Fructus mihi non visus. Luzon, Province of Zambales, Mount Tapulao, For. Bur. 825i Curran & Merritt, December, 1907, altitude 1500 ni. 19. S. curtiflora Elmer Leafl. Philip. Bot. 2 (1908) 509. Frutex glaber, ramulis purpurascentibus, apice viridescenti- bus. Folia chartacea, 5 ad 7 cm longa, 2 ad 3 cm lata, oblonga vel elliptica, integerrima, in apicem longiusculum subito pro- ducta, basi cuneata, costa subtus prominula, flavida, petiolus ca. 1 cm longus. Inflorescentiae spicatae, breves, petiolo baud vel vix longiores, densiflorae, sericeae ; calyx glaber, lobis rotundatis tubum aequantibus; corolla (alabastra tantum vidi) videtur cal- yce triplo longior, glabra; stamina ca. 35 (?); stylus glaber. Fructus viridis globosus, glaber, ca. 4 mm in diametro; lobis calycinis brevissimis coronatus. Negros, Dumaguete, Cuernos Mountains, Elmer 9802, April, 1908 ; Mount Marapara, altitude 1200 m above sea level. For. Bur. 13624 Curran & Fox- worthy, September, 1909. 20. S. Cumingiana Brand in Pflanzenreich 6 (1901) 58. S. angularis Elmer Leafl. Philip. Bot. 2 (1908) 510. Frutex 2 m altus, ramulis glabris. Folia 6 ad 17 cm longa, 2.5 ad ^.5 cm lata, valde variabilia, elliptica vel oblonga, leviter undu- lata vel serrata, utrinque glaberrima, breviter apiculata in petio- lum nunc brevissimum nunc longiorem flavidum sensim attenuata. Spicae puberulae, 5- ad 10-florae; bracteae sericeae, lanceolato- acutae; calycis tubus subglaber, lobi rotundati sericei; corolla alba, calyce duplo longior; stamina ca. 25; ovarium glabrum. Fructus 9 ad 11 mm longus, dilute brunneus vel atroviridis, glaber, rugosus, cylindricus, trilocularis, exocarpio tenui. Additional material: Luzon, Benguet Subprovince, Baguio, Williams 1308, August, 1904, Elmer 8736, 8554, 8695, March, 1907, For. Bur. 13497 Ramos, May, 1911 ; Bur. Sci. 2521 Meams, April, 1907. Negros, Cuernos Mountains, Elmer 10324, June, 1908, distributed as S. ferruginea; Canlaon Volcano, Phil. PI. 244 Merrill, April, 1910, distributed as S. curtiflora. ADDITIONAL PHILIPPINE SYMPLOCACEAE, IL 35 22. S. cagayanensis Brand sp. nov. Frutex vel arbor parva ca. 4 m alta, ramulis sordide villosis. Folia chartacea vel subcoriacea, elliptica vel oblonga, 4 ad 6.5 cm longa, 1.5 ad 3 cm lata, repanda vel subtiliter denticulata, juniora dense pilosa, vetustiora praeter costam glabrata, breviter apiculata, basi cuneata costa subtus valde prominula, petiolus 4 ad 8 mm longus, sordide hirtus. Spicae villosae, laxi- et pauci- florae, petiolo ca. triple longiores; fructus juniores et flores vetustiores brevissime pedicellati, flores juniores sessiles; brac- teae ovato-triangulares, bracteolae lanceolatae, pilosae; calyx pilosus, lobis lanceolatis tubum aequantibus ; corolla lutea, glabra, lobis calycinis duplo longior ; stamina ca. 25, corolla sublongiora ; ovarium glabrum, stylus glaber, elongatus, in fructu juniore persistens, Fructus viridis, pilosus, oblongis, 9 ad 10 mm longus, lobis calycinis brevissimis coronatus. Luzon, Province of Cagayan, Mission River, For. Bur. 16725 Curran, March, 1909 (type) ; Mount Ababaca, altitude 300 m, For. Bur. 18U69 Alvarez, Feb.-March, 1909: Province of Rizal, Angilog, Loher 6190, March, 1906; without locality, ex Herb. Ateneo de Manila, collector unknown (fruiting specimen). With some hesitation I refer also to this species a specimen collected by Maximo Ramos in 1909 on Mount Bagagan, Province of Abra {Bur. Sci. 7207). It differs from the species as described in its larger leaves (up to 8 cm long) and denser pubescence. The specimen Vidal 2H1, Province of Nueva Ecija, referred previously to S. luzoniensis “ may be S. cagayan- ensis, judging from the pilose fruits. 23. S. luzoniensis Rolfe in Journ. Bot. 24 (1886) 348. Frutex ramulis rufo-setosis. Folia chartacea vel fere sub- coriacea, 4 ad 7 cm longa, 1.5 ad 2.5 cm lata, lanceolata vel oblongo-lanceolata, crenato-serrata vel serrata, praeter nervOm medium setosum glabra; spicae villosae, 4- ad 12-florae, petiolo 2- vel 3-plo longiores ; bracteae lanceolatae, villosae ; calycis lobi lanceolati; corolla calyce duplo longior; stamina ca. 25 corolla sublongiora; stylus glaber; ovarium breviter pilosum. Fructus junior ovoideus, calyce coronatus, glaber, rugosus. Luzon, Province of Rizal, Angilog, Loher 619U, March, 1906. This specimen agrees fully with the original description, the type specimen from Mount Banajao, being unknown to me, except that the ovary is called glabrous, while in my specimen it is slightly hairy. . I think, however, the pubescence of the ovary may be easily overlooked, if only superficially examined. ‘ This Journal 3 (1908) Bot. 9. 36 BRAND. 24. S. Merrilliana Brand in Philip. Journ. Sci. 3 (1908) Bot. 9. Additional material: Luzon, Province of Laguna, Mount Banajao, Bur. Sci. 6081, 654.7, 9803 Robinson, Loher, 5696, 6179, February, 1906: Benguet Subprovince, For. Bur. 10839 Curran, December, 1908; Baguio and vicinity. Bur. Sci. 11978 Robinson, May, 1911; Mount Tonglon, Bur. Sci. 5387 Ramos, December, 1908. One of the specimens collected by Dr. Robinson {Bur. Sci. 6547) grows at an altitude of only 1250 m above sea level and has broader leaves than the others. The specimen collected by Mr. Curran is from a larger tree than is usual ; it is 7 m high. 26. S. depauperata Merrill in Govt. Lab. Publ. (Philip) 29 (1905) 45. Additional material: Luzon, Province of Benguet, Baguio, Elmer 8546, 8594, 8805, March, 1907, For. Bur. 14149 Merritt, December, 1908, For. Bur. 18301 Alvarez, December, 1908; Mount Pulog, For. Bur. 18059, 18075, 18107^ Curran, Merritt, & Zschokke, January, 1909; Mount Tonglon, For. Bur. 14166 Merritt, December, 1908; Mount Ugo, Bur. Sci. 5833 Ramos, December, 1908: Province of Rizal, Mountains back of San Mateo, L. Guerrero, 1910. The specimens collected on Mount Pulog are of peculiar interest. No. 18107 has inflorescences longer than the leaves. No. 18075 is a shrub 4 m high, and No. 18059 grows at the extreme altitude of 2900 m above sea level. Var. angustissima Brand var. nov. Folia lanceolata, 9 ad 11 mm lata (in speciminibus ceteris 13 ad 24 mm lata) . Luzon, Bontoc Subprovince, Bauco, Father M. Vanoverbergh 1095, Jan- uary, 1911. The Philippine Journal of Science, C. Botany. Vol. VII, No. 1, April, 1912. THE MECHANISM OF CURVATURE IN THE PULVINI OF MIMOSA PUDICA. By William H. Brown. {From the Botanical Section of the Biological Laboratory, Bureau of Science, Manila, P. I.) According to PfefFer ^ the curvature resulting from the stimu- lation of the pulvini of the petioles of Mimosa pudiea L., is due to a fall in the turgor of the cells of the lower or reacting half of the pulvini, which cells are compressed by the expansion of those of the upper turgid half. The decrease in turgor is shown by the stimulated pulvini becoming more flaccid and less rigid than the unstimulated ones, and by the passage of water from the cells, of the lower half, to the intercellular spaces. Pfeifer states that from the load required to prevent curvature it can be calculated that the energy necessary for movement is from two to five atmospheres and that hence the curvature can not be due to an active contraction of the protoplasm as was claimed by Vines * and by Gardener.® A reduction in turgor might be produced by a change in the permeability of the plasma membranes, as claimed by Lepesch- kin,* although as pointed out by Pfeifer this would neces- sitate a passage of dissolved substances from the cells. It seems more likely, as is thought by Pfeifer, that the reduction in the turgor of the cells of the pulvini is due to a decrease in the osmotic pressure of the cell-sap, which would allow an exosmosis of water without the dissolved substances. In either case, however, there should be a decrease in the number of osmotically active molecules in the cells, and if this were the cause of the curvature in the living pulvini, it seemed likely ' Pfeffer, W., Physiology of Plants, translated by A. J. Ewart, Oxford, 1906. ’Vines, Arb. Bot. Inst. Wurzburg 2 (1878) 146. ' Gardener, Annals of Botany 1 (1887-8) 366. * Lepeschkin, W. W. Ber. Deutsch. Bot. Gesell. 26 (1908) 724. 37 38 BROWN. that a similar curvature might be produced in dead pulvini if the proteids of the cells could be coagulated around the osmot- ically active substances thus forming osmotic cells. A reduc- tion in the pressure within the cells could then be obtained by replacing the water in them with some liquid in which the osmotically active substances would not dissolve. In order to test this hypothesis a large number of leaves were killed in boiling water. The pulvini used were those of the leaflets as these could be killed more quickly than the larger ones of the petiole. When leaves were killed after the production of curvature in the pulvini the pulvini remained curved; but at- tempts to kill the pulvini, before bending had taken place, were only partially successful as they were always stimulated, to some extent, by the treatment before losing the power of responding. However, thirty-eight leaves were obtained in which the curv- ature of the pulvini was only about half as much as it would have been if completed. In order to remove the water from the cells of the pulvini the leaves were run up through several grades of alcohol to absolute alcohol after which they were transferred to xylene. Since sugars are practically insoluble in this it would seem probable that there could be little or no osmotic pressure in the cells of the pulvini after they had been placed in it. Those pulvini which had been killed after complete curvature showed no change in shape, while in those in which the curvature was only partial, it had been completed. Since in these later cases the effect of replacing the water with xylene was to reduce the pressure in all the cells of the pulvini, it would seem that the completion of curvature could have been caused only by a reduc- tion in the osmotic pressure in the cells of the concave or reacting half of the pulvini, without the aid of pressure due to the expan- sion of the cells of the convex half. Concave is used here to designate the half of the pulvinus toward which bending takes place, and convex the opposite half. These terms are substituted for upper and lower as applied to the pulvini of the petioles, since physiologically the upper half of the pulvini of the leaflets, with which we are dealing, corresponds to the lower half of the pulvini of the petioles. The production of curvature in the pulvini placed in xylene would indicate that the cell walls of the convex half were less elastic or more rigidly placed than those of the concave half, for an equal contraction of both halves would not produce curva- ture; and the osmotic pressure must have been largely removed from both. In this connection it is interesting to note that a longitudinal section of a pulvinus shows that on the concave half CURVATURE IN THE PULVINI OF MIMOSA PUDICA. 39 there are several deep wrinkles running perpendicular to the longitudinal axis of the pulvinus. Any decrease in the pressure in the cells should cause a deepening of the wrinkles and conse- quently a curvature of the pulvinus. In living leaves the turgor of the cells of the concave half is probably not entirely removed when curvature is produced so that the pressure exerted on these by the turgid ones of the opposite side may at least increase the rapidity of movement. After taking observations on the leaves in xylene they were transferred back to alcohol and run down through several grades of water. Those pulvini which had been killed after curvature was complete remained as before while those with only a partial curvature, which had been completed in xylene, straightened out to the same extent as when first killed. It would seem that this could have been due only to the restoration of the osmotic pressure in the cells of the concave or reacting half of the pulvini by the redissolving in the water of the osmotically active sub- stances in the cells. In no case did the pulvini straighten out to a greater extent than when killed. The phenomena exhibited by these dead pulvini would seem to show that stimulation pro- duces a relatively permanent reduction in the number of osmot- ically active molecules in the cells and that the extent of the movement which can be produced in the dead pulvini stands in some inverse ratio to the extent of this reduction. The reduction in the number of osmotically active molecules in the cells might be brought about by a chemical change in the contents of the cells or by a change in the permeability of the plasma membranes which would allow an exosmosis of some of the osmotically active substances dissolved in the water of the cells. While, in the case of Mimosa, the last possibility is not excluded the first would seem more probable, for if the loss in turgor is due to a change in the permeability of the membranes which allows the passage into the intercellular spaces of dissolved substances, a reverse change in permeability would not restore these lost substances to the cells, and turgidity could be restored only after the manu- facture of additional osmotically active substances. If, how- ever, the fall in turgor is due to a chemical change in the con- tents of the cells, a reverse change would restore turgidity. The closure of the leaf of Dionaea, which will be discussed more fully in a later paper, also appears to be due to a change in osmotic pressure. When the leaves close there is a passage of water from the cells of the inner or concave surface to those of the outer or convex, which causes the latter to become greatly 40 BROWN. stretched. If the leaves are killed soon after closure and trans- ferred from water to xylene, so that the osmotic pressure in the cells of the convex side is removed, the leaves open. A change in the permeability of the membranes might cause the passage of water from the cells of the inner surface but could hardly explain its entrance into those of the opposite side. In both Mimosa and Dionaea there appears to be a reduction in the os- motic pressure in the cells of the side toward which bending takes place. In Mimosa the water which passes from the cells goes largely into the intercellular spaces, while in Dionaea it is taken up by the cells of the opposite surface. In Mimosa move- ment is due to a fall in the turgor of the cells of the concave half of the pulvinus while in Dionaea it is due to the stretching of the cells of the convex surface. The experiments with the dead pulvini of Mimosa show that changes in the osmotic pressure in the cells of the reacting half can cause movement without the aid of any vital phenomena and may therefore be taken as a confirmation of Pfeffer’a conclusion that ’the movement of the living pulvini is due to changes in the osmotic pressure of the cells of the reacting half. Since the experiments with the pulvini of the leaflets and Pfeifer’s with those of the petioles lead to similar conclusions it would seem that the reactions in the two cases are similar. The Philippine Journal of Science, C. Botany. Vol. VII, No. 1, April, 1912. THE GENUS THAYERIA. By Edwin Bingham Copeland. (From the College of Agriculture, University of the Philippines, Los Banos, P. I.) The genus Thayeria was described by myself about six years ago,^ from sterile specimens collected in the mountains north of Zamboanga. The type species was given the name T. Cornu- copia. It was found on a ridge above the source of the Sax River, a hurried visit to the place requiring three days of hard travel. I made the trip twice in 1905, the second time for the special purpose of finding fruiting fronds of this fern, but without success. In the same year I collected sterile specimens in Lepanto-Bontoc, but could find none fertile. Baker had de- scribed a New Guinea plant collected by Beccari, with essentially identical vegetative structures, as Polypodium nectariferump and these structures are so peculiar that it seemed probable that the fertile fronds were also alike. As far as my specimens showed, the identity was so perfect that I ascribed my Luzon plant to' Baker’s species, as Thayeria nectarifera. Thayeria is a fern of the Drynaria group, as shown unmis- takably by the very stout, fleshy rhizome, with a dense coat of brown scales, the structure and venation of the leaf, and very characteristic splitting off of the segments from the midrib, the humus-gathering habit, and various minor details. Its es- sential peculiarity is the specialization of branches of the rhi- zome, as phyllopodia. Each of these branches bears a single large leaf, the lower part of which is very broad, with the sides rolled together so as to form a broad cup like a cornucopia. The end of the branch is in the bottom of this cup, where it bears a dense cluster of roots. In the cup falling leaves and twigs collect and decay. Each branch makes therefore a sort of com- * This Journal 1 (1906) Supplement 165, Plate 28. ’ Malesia 2 : 247, Plate 65. 41 42 COPELAND. plete physiological unit, the leaf and roots working together — so close together that the stem is hardly in evidence. The main rhizome produces these successive units and serves as their point of attachment, and also as a place of storage, at least of water. Except presumably in every young plants, which have not been seen, the rhizome bears no roots except numerous short ones confined to the side against the tree on which the fern grows. Their main function is certainly clinging; but they doubtless absorb some water, and very little food. In each sinus between the segments of the leaf is an evident gland, but I have seen no ants around these. For its mineral food the plant probably depends practically altogether on the detritus collected by the leaves. In the perfection of the humus-gathering apparatus, and in the extension of the specialization to all parts of the plant, Thayeria has gone so far beyond any other ferns, or any other plants whatever, that I believed, and believe, that its recogni- tion as a distinct genus is justified and advisable, though on vegetative characters alone. Diels * reduced Baker’s Polypodium nectariferum to Drynaria, and van Alderwerelt * did the same with Thayeria Cornucopia. Both authors seem to have failed to grasp the characteristics of the plants of which they were treating; and their placing these plants in Drynaria is unjustified even with such light as they had, while both of them place Aglaomorpha meyeniana, which they knew to agree in fructification with Polypodium nectari- ferum, in Polypodium, and maintain Dryostachyum as a distinct genus. Aglaomorpha is decidedly nearer Dynaria than is Thayeria. Van Alderwerelt has even included in his Section Thayeria a real Drynaria. Thayeria was first collected fertile in the Philippines by Ramos, Bureau of Science 7192, in Abra, in northern Luzon. In 1910, I collected copious fruiting specimens on Mount Santo Tomas, in Benguet. These showed that the Luzon plant was distinct from that of New Guinea, with which it had been identified.® As the Zamboanga plant had been made the type of the genus, I immediately made another attempt to find its fruit but was again unsuccessful. Accompanied by Mr. Merrill, I went for it a fourth time in November, 1910. After we had hunted along the entire ridge where it is common sterile, and had 'Naturlichen Pflanzenfamilien P: 330. ‘Bull. Dept. Agr. Indes Neerland. 21 (1908) 8. ® This Journal 6 (1911) Botany 140. THE GENUS THAYERIA.. 43 given it up, fertile fronds were detected in a single tree at a greater altitude. Fifteen fertile fronds were obtained by feeing the tree. With the ample material now in hand from Mindanao and Luzon, I conclude that the differences formerly found between them are not constant, but that all are one species. This differs from Thayeria nectarifera in not being dimorphous. As the ac- companying plate shows, the sori are born, as in Aglaomorpha, on the constricted upper part of otherwise normal fronds ; while the sterile lower part of the fertile frond of Thayeria nectarifera is pinnate, this frond being stipitate and not humus-collecting. The fertile pinnae of Thayeria nectarifera are also more con- stricted. The figure in “Malesia” shows no lamina between the sori, as is typical of Aglaomorpha § Psygmium; while Thayeria Cornucopia has more or less lamina everywhere, as the plate shows, but the amount of development of this wing is variable. The known species of Thayeria, then, are two: T. Cornucopia Copel., of Mindanao and Luzon. T. nectarifera (Baker) CopeL, of New Guinea. Copeland: Gen [Phil. Joubn. Sci.. I. THAYERIA CORNUCOPIA Copel. J PLATE ILLUSTRATION, Pljvte I. Thayerfa Cornucopia. Herbarium specimen from the type locality. Photograph by Martin. 46 V v,rji^iat!ATio;»s'i'(m saije by theI bttbba^ of sciebce, ^ » , ' mJSJL&j FHIUBPINE l$IASm qpg;E subanuns of sinbangan bay. , ^ ‘By EitEBSoJ? B. 'Chbistie. Bil p«g«3, 1 map, 2^ plate*. ' ' < ^ ' ' •'’ Pfder Njff. 416. Paper, $1.25 Uo«et( States eurrenoy, postpaid. 1 ^ -Sfndapgarv 8?iy I* situated oit the northern coast of^Zamboanga Penlnsplp. The Suhanuns of ''t'this region ^ere studied by ^r. Christie during two periods of five and six weeks, respectively. ' The following Is an abstraot from the contents of Mr, Christie’s 'report on the Sobanuns: ^ Hahl^t arid' history; relations with the Moros; material culture; houses; Industries; trade; »i^!?agrieulture;' family, life;: social customs; •administration of justloe; retlgjon; th^. medioinO: man; I ' ceremonies; tales; word-lists; physical measurements. The plates Illustrate the 'Subanuns at work and at play; the|r Industries^ houses, altars, and Implements; and the 'People themselves. , ‘ ' ' ' / TBtE HISTpBY OF S1[lliU. „ > < By Wajeeb M. Saxebby. ^ 275 pages, 4 maps, 2 diagrams. Order No. 406; ’ Paper, $0.75 United Stales currency, postpaid. « tn the preparation of his manuscript for The History of Sulu Doctor Saleeby spent much tim^ and effort in gaining access to documents in the possession of the Sultan of Sulu, It 15 fortunate that these records have now been translated and preserved In permanent form. This book is ft. history ef the Moros in the Philippines from the earliest times to the Amerioari occupation. ^ ' STtlBIES Ilf MOBO HISTORY, LAW, AND RELIGION.' ^ ' V ", By Naje^ iL Sauseby. < - \ 107 pages, 16 plates, 5 diagrams. Order No. 405. r Paper, $0.25; half morocco, $0.75 United States currency, postpaid. This Volume deads with the earliest written records of the Moros in Mindanao. Doctor Saleeby was fortunately able to obtain exao^ copies of carefully preserved early records written tb the Magindanao dialect with Arabic characters. The author presents translations of these OS well as 16 half-tone Illustrations ot certain pages from thu originpls, >The names of the rulers of / Maglhdanao are recorded In five folding diagrams. ^ , ’ NEGRITOS OF" 2AMBALES. ' ' ' . " By WmCiiAir Atlas Beep, , " ^ < S3 pages, 62 plates. 'OrderfNoi:402i.' . :v Paper,, $0.25; half-moroocp, $0;75 tJntfed States currency,; postpaid. The introductory chapter 'deals with ipe'general distribution of Negritos pad with the distribu- iitlon;:ofi the Pl^iitppine branoh of the, race. ! ' The succeed ing , chapters deal with .the various Indus', tries, amusements and lioCial relations of th^se little men. ,, ’ Plates from photographs, the greater ^part of which' was talcen for this publication, show Urna'menls, houses, men making Ore with bamboo, bows and arrows, dances, and various types of tbe\Peopte themselfes. , ' ' ' CONTENTS Pige. AMES,, OAKES. Notes on Philippine Orchids with De- scriptions qf New Species, IV i BRAND, A. Additional Philippine Symplocaceae, H 29 BROWN, W. H. The Mechanism of Curvature in the Pulvini of Mimosa pudipa ' 37 COPELAND, E. B. The Genus Thayeria 41 ' The “Philippiine Journal of Sciejice” is Issued as follows: 1 : ourrenoy. . Section A. Cbenilcal and Geological Sciences and the Industries.. $2.00 Section B. The Philippine Journal of Tropical Idedicine 1. 3.00 Section C. Botany ....... j 2,Q0 Section D. General Bfojogy, Ethnology . and Antliropology ( Sec-''0:L f ", tion D-hegan with Volume V) j...' Ehtire Journal, Volume II, III, IV, or V SlOO' Entire Journal, beginning with Volume VI ‘ Single numbers 1 150'; Volume I, 1906 (not divided into sections) ■ ' '’to-oA- Supplement to Volume I (botany) : Volume I (without supplement) ' 6.$^ Each section is separately paged and indexed. Publications sent in exchange for the Philippine Journal of Science should be addressed: Library, Bureau of Science, Manila, P. I. Subscriptions may be sent to the Business Manager, Philippine Jour- nal of Science, Bureau of Science, Manila, P. L, or to any of the agents listed below : The Macmillan Company, 64-06 Elfth Avenue, New York City; U. S. A: Wm. Wesley & Son, 28 Essex Street, Strand, London, W. G., England. Martinus NijholT, Noheistraht 18, ThO Hague, tlolland. Mayer & Muller, Prinz Louis Perdinandstrasse 2, Berlin, N.W., Germany. Kelley & Walsh, Xiimlted, A2 Raffles Place, Singapore, Straits Settlements. A. M. & J.'Ferguson, 19 Baillie Street, Colombo, Ceylon. Thacker, Spink & Co., P. ,0. Box 54, Calcutta, India. ■ r Entered «t thevpoet-olWoe at Manila, P. I., a« second-olase matter, f-' .jK'' ,n, ' ''Joidi I9ia‘ m&ik' ;i?;/<-r r|i^ '■ / ', 'i ,*• 1*' /^l’*-' ^ 1 ''■^ BJ&'i W. pf-BBSWN, ?S.D. ■“ .., Cl' PTOIICATIONS FOR SAIE BY THE BITREAIT OF SCIEHCE» MAUILA, PHILIFPINE ISLANDS REPORT OP THE INTERN ATIONAIj PLAGtrE CONPERENCE. the Chinese Government. ' , Edited by Ebioh Martint, (t. ^’. Petrie, Arthur Stanley^ and Kichabo ' P. -yTRONd. ■ „■ ■■■..,•,, 483 pages< 18 plates (2 colored, 4 half-tones, 12 charts and maps). Order No. 416. Cloth, $3.50; paper, -$2,50 United States currency, postpaid. The proceedings of this International Conferenoe and information gained therefrom, together with the results of certain bacteriological investigations, constitute the present report. - Nothing -hitherto has been published which 'gives suoh a complete and comprehensive account of the entire subject of, pneumonic plague. Delegates from America (United Slates of), Austria-Hungary, France; Germany, Great Britain, Italy, Japan, Mexico, the Netherlands, Russia, and China attended the Conference. The Bureau of Golence of the Government of the Philippine Islands has beep appointed sole agent for the distribution of the printed proceedings of the International Plague Conference; THE STJGAR INDUSTRY IN THE ISUAND OP NEGROS. ^ ^ . Q 145 pages, 10 plates, 1 map. , - Order No. 412. Paper, $1.23 United States currency, pofrtpaid. Considered from the viewpoint of praotioal utility, Mr. Watker’s’ Sugar Industry in the Island of Negros Is one of the most important papers published by the Bureau of Science. This volume is a real contribution to the subject; it is not a mere oompilatlon, for the author was In the field and understands the conditions of which he writes. The foticwing is a brief synopsis of the contents: Tables of soil analyses, both chemical and physical; analyses 6f the cane, juice and bagasse; estimates based on actual information as to the costs of production and of cultivation; and esti- mates of -the cost and location of possible central factories. The island Is considered by sugar- producing districts; the area of cultivation and the production per hectare are given, and the possibility for future, expansion discussed. - , - r The plates illustrate various phases of sugar industry from the cultivation of the field- to the transportation of sugar in native sailboats. A MANUATi OF PHIUIPPINB SILK CUIiTURE. ■ , ' V, ' .By ChaeuesS. Banks. ■ 53 pages, 20 plates. - - Order N6; 413; Paper, $0.75 United Slates currency, postpaid. The silk industry is particularly adapted to be undertaken by persons with small capital, and lixe the making of hats in the Philippine Islands It should thrive-with a little encouragement. In' A Manual of Philippine Silk Culture we have presented the results of several years’ actual work with silk-producing larvsB together with a description- of the new Philippine race. Half-tone plates illustrate in natural size silkworms in different stages of development, pupa;, adult moths, .samples of cloth made from eri silk, hand reel, and silk house. Other plates [llustrate the various appliances used in raising silkworms .and in spinning silk; h^nd and, Power reels, are illustrated; working drawings are given for a silk house and for a hand reel- OBITUARY ^aul Casipar Jfreer DIRECTOR OF THE BUREAU OF SCIENCE OF TH E GOV ER N M E NT OF TH E PH I LI PPI N E ISLAN DS DEAN OF THE COLLEGE OF MEDICINE AND SURGERY AND PROFESSOR OF CHEMISTRY OF THE UNIVERSITY OF THE PHILIPPINES, AND FOUNDER AND EDITOR-IN-CHIEF OF THIS JOURNAL We are deeply grieved to announce the death of Doctor Freer at Baguio, Philippine Islands, on April the seventeenth, in his fifty-first year, from arterio-sclerosis and acute nephritis. In an effort formally to express our sorrow and to honor his memory a memorial meeting of the members of the Staff of the Bureau of Science, the Council of the University of the Philippines, and the members of the Philippine Islands Medical Association will be held on July 1, 1912. The proceedings of this memorial meeting will be published in a future number of this Journal. At a meeting of the members of the Staff of the Bureau of Science, held on the eighteenth day of April, the following resolutions were adopted: OTHtereasI it has pleased Almighty God in His Wise and Inscrutable Providence to remove from our midst Paul Caspar Freer, M. D., Ph. D., Director of the Bureau of Science of the Government of the Philippine Islands, since the time of its organiza- tion as the Bureau of Government Laboratories in the year 1901, Dean of the College of Medicine and Surgery, and Professor of Chemistry, University of the Phil- ippines, and Founder and Editor-in-Chief of the “Philippine Journal of Science,” who, for many years, has been our Leader, Counselor, and Friend; and WB^ertai at best we can do little to indicate at this time our real appreciation of him as a man and as a worker for the general good: Therefore be it 3&f£(olbEb, That we, tbe Members of the Staff of the Bureau of Science in Manila, Philippine Islands, do hereby express our deepest sorrow and keen feeling of personal loss in the death of Doctor Freer; and be it further J^ESolbeb, That he holds a place of highest respect, admiration and appreciation both officially and personally in the hearts of all of us, and especially of those who were most intimately associated with him in scientific work; and be it further J^ESolbeb, That it is the sense of the Members of this Institution that the Bureau of Science has suffered a very great loss and that the cause of Science in these Islands has been deprived of one of its most zealous and conscientious advocates; and be it further iREeolbEb, That we extend our sincere sympathy and condolence to his Widow in her overwhelming grief, to his Sister, Brother and other Relatives; and be it further 3&ESolbEb, That copies of these resolutions be engrossed and sent to the bereaved Widow and Brother of Doctor Freer, and that they be filed in the Archives of the Bureau of Science, transmitted to the Bureau of Civil Service, published in the forth- coming Number of each Section of the “Philippine Journal of Science,” in the newspapers of Manila, in a paper in the City of Chicago, Doctor Freer's birth-place, and in “Science,” the Official Organ of the American Association for the Advancement of Science, of which Doctor Freer was a Fellow. For the Staff of the Bureau of Science; RICHARD P. STRONG, CHARLES S. BANKS, E. D. MERRILL, ['-• S.] ALVIN J. COX, OSCAR TEAGUE, A. E. SOUTHARD, Committee. At Manila, Philippine Islands, this eighteenth day of April, in the year of our Lord one thousand nine hundred and twelve. THE PHILIPPINE Journal of Science C. Botany VOL. VII JUNE, 1912 No. 2 THE ORIGIN AND RELATIONSHIPS OF TAENITIS. By Edwin Bingham Copeland. (From the College of Agriculture, University of the Philippines, Los Banos. ) Taenitis is one of the ferns which, up to this time, has eluded very successfully all attempts at natural classification. It has been in such a tribe as the Grammiticleae of Hooker and Baker’s “Synopsis,” in company with Notholaena, Brainea, Meniscium, Vittaria, Hemionitis and Drymoglossum, that it has seemed least out of place, as this tribe has no semblance of naturalness to be disturbed by it. A tribe named for this genus was proposed by Presl,^ and is maintained with changed composition, as a subfamily of Polypodieae, by Diels ^ and Christensen.® Some of the other genera included in the group, Drymoglossum for instance, are obviously related to the real Polypodieae, and the group must have been given its position on their account, for Taenitis itself offers hardly a suggestion of such an affinity, except in the naked sorus, which it shares with the various genera named above, and with many others. Taenitis has a hairy rather than scaly rhizome, of very characteristic reddish color, non-articulate stipe, and altogether non-polypodioid venation. What slight superficial resemblance it has to any plants called Polypodium is to certain species of Selliguea (Phymatodes) , but its remoteness from these is attested by the absence of their very characteristic foliar endodermis. In ’Tent. Pterid. (1836) 222. ’ Engler und Prantl, Nat. Pflanzenfam. 1 302. ’Index Filicum (1906) 46. 110692 4Y 48 COPELAND. my work on the Polypodiaceae of San Ramon,* in which I carried the natural arrangement of the family farther than had before been attempted, I was confronted by the unnaturalness of the supposed near relationship of this fern to Polypodium, and, not knowing what its real affinities might be, did not include it at all in the “family tree” of the Polypodiaceae. It has since become clear that Taenitis belongs in the only large and nearly natural tribe not represented among its com- panions in the standard works, the Davallieae. In the family tree just cited, Microlepia represents the central group in this tribe. By removing Davallodes, I have since made it possible to constitute a really natural genus, including Microlepia in Dennstaedtia. Dennstaedtia and its nearer relatives are un- failingly characterized by hard, creeping, reddish rhizomes clothed with small, harsh, reddish hairs which are pluricellular at the base but narrowed above to a single row of cells. The color and form of a negro’s hair do not testify to blood relation- ship more certainly than do the color and form of the pubescence of Dennstaedtia, Saccoloma, Leptolepia, Odontosoria, Tape- inidium, Saccoloma, and Taenitis. Trichomes enough like these to betray relationships are found not only in various Davallieae, but also, as we shall presently see, in various ferns which have apparently not been suspected of such affinity. Of ferns known to me, the most like Taenitis blechnoides (Willd.) Sw. in superficial aspect is Schizoloma ensifoUum (Sw.) J. Sm. This fern is extremely variable in its venation, the veins being sometimes almost free, sometimes anastomosing very similarly to those of Taenitis, whose venation is likewise far from uniform. Taenitis sometimes, though not usually, has the peculiar form of leaflet, broadest near the base, characteristic of Schizoloma ensifoUum. The rhizomes and the bases of the stipes are so alike in the two ferns that these parts are often practically indistinguishable. A young plant of either is likely to be mistaken for the other. The difference in the position and protection of the sorus is hardly greater than that found in Vittaria, between V. scolopendrina and V. elongata, where it is not usually regarded as constituting a generic distinction. Taenitis itself is variable in the position of the sorus, as the accompanying photograph shows (Plate II). And there are species of Schizoloma whose sori are by no means marginal. However, internal structure affords in general a better clue to affinity than does external; and the steles in the rhizome of ^ This Journal 2 (1907) Bot. 74. ORIGIN AND RELATIONSHIPS OF TAENITIS. 49 Taenitis and Schizoloma are unlike. Schizoloma ensifolium, and other species of the genus, have the peculiar solid steles of the Lindsayae.^ But Dennstaedtia seems to have in all species a solenostele with very short foliar gaps. As Gwynne-Vaughan states, T'aenitis has a dictyostele very near in nature to a solen- ostele; it is so near in fact that in a rhizome with internodes 16 mm long the foliar gaps may overlap by less than 1 mm. The only plant usually recognized as a very near relative of Demistaedtia, in which I have found a dictyostele, is Sac- coloma moluccanum, in which this structure is derived from a solenostele by the shoi’tening of the internodes. Davallodes, which contemporary pteridologists have treated as a Microlepia, has a complicated dictyostele; but my appreciation of its dis- tinctness from Microlepia has strengthened since I raised it to generic rank, and I now group it with the other epiphytic Davalliae, — Leucostegia, Davallia, Humata, Oleandra, etc., — in which open dictyosteles are the I’ule. Drymoglossum has a real dictyostele, as have its polypodioid relatives, Goniophlehium, and the numerous related groups, and Hymenolepis, Eschatogramme, and Paltonium lanceolatum. While technically bearing the same name, the stele of Taenitis is very much less like that of Drymoglossum than like any solen- ostele of the Dennstaedtia group. There are other characters of more or less interest which might be considered; but I believe that enough has been said so that nobody, having the plants in hand, will question the conclusion that Taenitis is a reasonably near derivative of Dennstaedtia, and not at all a near relative of Polypodium. We will now see that Taenitis is one of a very natural group of ferns, but a group as different as possible in composition from those which have borne its name. 'Gwynne-Vaughan, Annals of Botany, 17 (1903) 689, Plate XXXIV, fig. 23, representing the stele of “Davallia” repens, might have been drawn from Schizoloma ensifolium. The data in this paper are of great taxo- nomic value, in spite of the statement that the “Anatomical characteristics do not, for the most part run parallel to ... . systematic position.” The trouble is ■with the systematic position; thus, the author finds true solen- osteles in Dicksonia, Davallia, Lindsay a, Pteris and Poly podium; but the situation clears up when it develops that the Davalliae are all Microlepia or Leptolepia; the Dicksoniae, all Dennstaedtia; the Lindsay a, Odontosoria; the three species of Pteris, severally, Paesia, Histiopteris and Doryopteris ; and the Polypodium, probably Hypolepis. When one recognizes genera which in a proper sense are genera, all of the apparently very diverse plants just named present themselves as near relatives. Contradicting the statement quoted above, Gwynne-Vaughan later suggests such a con- clusion. 50 COPELAND. Piaty taenia requiniara (Gaudich.) Kuhn is treated by Diels as the nearest relative of Taenitis. This fern is accredited to the Philippines, but is known to me only by leaf fragments ex herb. Kunth, collected in Waighiou in 1825, and kindly sent me from the Berlin Botanic Garden. So far as can be judged from these fragments, it is altogether like Taenitis except for the acrostichoid fructification and a correlated narrowing of the fertile pinnae; the close affinity of the two is not doubtful. Lomagramma is like Platytaenia in fructification, and fairly similar in venation, but does not seem to be a member of this group. Genera with the Gym.no gramme type of fructification are usually found in the groups with acrostichoid genera, and may be regarded as intermediate between the latter and the more primitive forms with definite sori. In the Taenitis group, this position is occupied very exactly by Syngramma pinnata J. Sm. The stele and trichome characters are perfectly typical of the group; and I have fronds from New Guinea which would be determined without hesitation as Taenitis, if they were sterile. Sy7igramma is a natural genus, whether or not its separation into several genera can be justified. The simple-leaved sections have modified types of venation, but are alike in the important structural characters, including the natural type of stele. Ex- amining a small number of individuals, I have found soleno- steles in Syngiamma alismifolia J. Sm., S. ca^'tilagidens (Baker) Diels, and S. horneeyisis (Hooker) J. Sm. ; overlapping foliar gaps in S. Wallichii Hooker and S. angusta Copel. ; and in S. Hookeri C. Chr., gaps usually but not always overlapping, this character varying along a single rhizome. Because of its affinity to Taenitis and Platytaeiiia, and espe- cially to the more primitive genera, ScJiizoloma and Dennstaed- tia, Syngramma may with reason be regarded as the most primi- tive member of one or perhaps two other groups, embracing a considerable part of the genera included in the Pterddeae of Diels. These include first the genera with fructification of the same type, as Craspedodictyum.,^ Coniogramme and Hemionitis; and second such genera as Doi-yopteris, with the sori of Pteris. These all preserve the ancestral stelar characters, and the fronds of most of the species have the same structural peculiarities. The paleae of all except the first-named are less abruptly nar- rowed, and in some cases less pigmented and with thinner walls. ‘This Journal 6 (1911) Bot. 84. ILLUSTRATION, Plate II. Taenitis blechnoides (Willd.) Sw. Two pinnae showing the difference in arrangement of the sori. 51 I Rrlationsiiips of Taenitis.] [Phil. Journ. Sci., VII, PLATE II. TAENITIS BLECHNOIDES (WiUd.) Sw. The Phojppine Journal of Science, C. Botany. Vol. VII, No. 2. June, 1912. NEW OR INTERESTING PHILIPPINE FERNS, VI. By Edwin Bingham Copeland. (From the College of Agriculture, University of the Philippines, Los Bahos.) Pleurcgramme minor (Fee) Copel. comb. nov. Vittaria minor Fee 3rd. Mem. (1851-52) 23, pi. IV, f. 2. Pleurogramme loheriana Christ in Bull. Herb. Boiss. II 6 (1906) 1006. Christ^ distinguishes his Pleurogramme loheriana from P. pusilla (Blume) Christ by its “having the soriferous line sunk in an exactly marginal grove.” Fee states of his species “sporotheciis .... exacte marginalibus." Comparison of absolutely authentic specimens leaves no doubt as to the identity of the two species. Trichomanes craspedoneurum Copel. sp. nov. Rhizomate filiforme repente; stipitibus brevibus vel subnullis, minute pilosis; fronde 1 ad 1.5 cm longa, oblanceolata, integra vel saepius paucilobata, deorsum angustata et pilifera, sursum glabra; venatione pinnata, venis spuriis obliquis sat conspicuis, cum vena submarginale anastomosantibus ; soro solitario apicale, tubo cylindrico omnino immerse, limbo dilatato, cum margine anastomosante et deinde bilabiate. Luzon, Province of Tayabas, Infanta, alt. 100 m, J. B. Leiberg, sheet No. 593183, U. S. Nat. Herb. Distinguished from T. sublimbatum K. Mull, and T. henzaianum Hooker by the evident submarginal vein. Near T. Petersii A. Gray of Alabama. Pteris Taenitis Copel. sp. nov. (Plate III.) Stipitibus densissime confertis, 3 ad 10 cm altis, fulvo-casta- neis, deorsum paleis fulvis crinitis vestitis, sursum rhachique praecipue in sulco puberulis glabrescentibus ; fronde 25 ad 35 cm alta, pinnata; pinnis 20 ad 30 cm longis, 4 ad 6 mm latis, 1 ad 3 paribus, fere erectis, utrinque sensim angustatis, decurrentibus et plerumque ala angustissima connexis, integris, coriaceis, “ This Journal 2 (1907) Bot. 175. 53 54 COPELAND. opacis, mox glabrescentibus, fertilibus baud diversis; margine reflexo sat lato. Mindanao, Agusan Subprovince, Mount Hilonghilong, alt. 115 m, C. M. Weber 1H3, March, 1911. A relative of Pteris opaca J. Sm. but much smaller; easily distinguished from all forms of Pteris cretica by the simple and entire, decurrent, coria- ceous pinnae. Dryopteris dichrotricha Copel. sp. nov. Thelypteris, D. (Nephrodio) adenophorae affinis; rhizomate erecto, paleis angustis castaneis vestito ; stipite 30 cm alto, brevis- sime hirsute, sursum pinnis in auriculas minutas reductis donate ; fronde 35 cm alta, 12 ad 15 cm lata, acuminata, ubique densis- sime minute albido-setosa ; pinnis utroque latere ca. 20, sessilibus, acuminatis, 10 ad 15 mm latis, | ad costam pinnatifidis; utroque facie ad costa venasque setis majoribus vestitis; venis simplicibus, utroque latere 6 ad 9 ; soris medialibus, indusio persistente, setose. Mindanao, Agusan Subprovince, Mount Hilonghilong, alt. 120 m, C. M. Weber 1173, March 1911. Nearer to D. adenophora than to any species with free veins. Dryopteris mesodon Copel. sp. nov. Species gregis D. dissectae, stipite ad basin paleis linearibus castaneis 15 mm longis vestito, supra sparse spinuloso; fronde 35 ad 40 cm alta et lata, rhachibus castaneis, supra costisque velu- tinis ; pinnis infimis oppositis multo maximis, stipitatis, deltoideis, valde acuminatis; earum pinnulisi infimis stipitatis, acuminatis; harum pinnulis^^ brevistipitatis, cordatis, acutis, ad basin pin- natis; pinnulis”’^ adnatis, obtusis, grosse serratis, utroque latere una ; pinnulis ordinum omnium superioribus decurrenti-connexis, et ala decursa saepe dente uno ornata ; segmentis penultimis acute grosse serratis, acutis, oblanceolatis ; soris a costa remotis, parvis ; indusio persistente, minute praecipue ad insertionem puberulo. Mindanao, Agusan Subprovince, Mount Hilonghilong, alt. 120 m, C. M. Weber 1176, March, 1911. This bears the same relation to Dryopteris dissecta that D. pnrpurascens does to D. sparsa. Its nearest relative is D. balabacensis Christ. Tectaria Weberi Copel. sp. nov. Sagenia, rhizomate ca. 1 cm crasso, suberecto, apice paleis angustis atrocastaneis 4 mm longis dense vestito; stipite 20 ad 25 cm alta, 15 ad 20 cm lata, breviter decurrente, ad alam 2 ad 5 mm latam pinnatifida; segmentis plerumque 5, sat remotis, late lanceolatis, sinuatis, obtusis, versus costam angustatis, her- baceis, lamina glabra; venis minute glanduloso-piliferis, venis primariis remotis, fere ad marginem protensis, areolis primariis PHILIPPINE FERNS, VI. 55 unisoriferis ; soris utroque latere venarum uniseriatis; indusio orbiculare cum sinu minute, 2 mm late, insertione lineare glan- duloso-pilifero, aliter glabram, tenue sed persistente. Mindanao, Ag-usan Subprovince, Mount Hilonghilong, alt. 42.5 m, C. M. Weber 1H8, March, 1911. Nearest to Tectaria decurrens, but in appearance approaching T. Men- yantkidis; distinguished from the former by the few segments, slightly decurrent base, and conspicuously remote main veins. Humata microsora Copel. sp. nov. (Plate IV.) Rhizomate 2 mm crasso, paleis castaneis apice acicularibus vestito, ad truncos arborum late repente; stipitibus remotis, 4 ad 7 cm altis, sparse paleaceis; fronde usque ad 18 cm alta, a basi truncata 15 ad 18 mm lata, sensim sursum angustata, deor- sum fere ad costam latam pinnatifida, segmentis 1 cm latis, trun- catis vel rotundatis, integris, rigide coriacea, glabra; venis conspicuis, furcatis; soris in lobe quoque usque ad 12 submar- ginalibus 0.6 ad 0.8 mm latis. Mindanao, Agusan Subprovince, Mount Hilonghilong, alt. 180 m, C. M. Weber 1H6, March, 1911. A very distinct species in the group of Humata angustata J. Sm. Adiantum scabripes Copel. sp. nov. (Plate V.) Adiantellum rhizomate setis parvis saturate atropurpureis dense vestito; stipite nitido, atrocastaneo, 20 ad 40 cm alto, deorsum paleis angustis 3 mm vestito vel ob baseos illarum scabro, sursum rhachibusque ubique glabris; fronde deltoidea, 15 ad 20 cm alta et lata, bi- vel tripinnata, pinnis paucis, acutis ; pinnulis stipitatis, dimidiatis, oblongis, apice rotundatis, 10 ad 13 mm longis, marginibus superioribus et exterioribus ubi steril- ibus cartilagineo-denticulatis, coriaceis, non opacis, supra nitidis- simis, infra luce directa pallidis luce oblique incidente subcupreis sed oblique versus obscuritatem visi fere albis; soro plerumque uno, usque ad 5 mm longo, interdum interrupto ; indusio brunneo, coriaceo, 0.7 ad 1 mm lato. Mindanao, Agusan Subprovince, Mount Hilonghilong, alt. 175 m, C. M. Weber 117U, March, 1911. A relative of Adiantum opacum and A. cujyreum, and like these a very promising fern for cultivation. Adiantum flabellulatum L. Sp. PI. (1753) 1095. Luzon, Province of Ilocos Sur, Mount Dagat, Bur. Sci. 7 760 Ramos. New to the Philippines. Tropical Asia, China, and Japan. Athyrium lanceum (Thunb.) Milcle Bot. Zeit. (1870) 354. Asplenium lanceum Thunb. FI. Jap. (1784) 333. Luzon, Province of Ilocos Norte, Bur. Sci. 7761 Ramos. New to the Philippines. Ceylon and India through China to Japan. 4 4 ILLUSTRATIONS. (Photographs by Charles Martin.) Plate III. Pteris taenitis Copel. IV. Humata microsora Copel. V. Adiantum scabripes Copel. 57 Copeland: Philippine Ferns, VI.] [Phil. Journ. Sci, VII. C. PLATE III. PTERIS TAENITIS Copel. CopELANn: PHILIPPINE Ferns. VI.] [Phil. Journ. Sci, VII, C. No. 2. 20 cm Copeland: Philippine Ferns, VI/ [Phil. Journ. Sci, VII, C, No. 2. PLATE V. ADIANTUM SCABRIPES Copel. The Philippine Journal of Science, C. Botany. Vol. VII, No. 2. June, 1912. NEW SARAWAK FERNS. By Edwin Bingham Copeland. {From the College of Agriculture, University of the Philippines, Los Banos.) Marattia Brooksi Copel. sp. nov. Stipulis 25 mm long-is latisque; stipite sicco 7 mm crasso, deorsum paleis parvis castaneis vestito, sursum rhachique ob emergencias carnosas insigniter transversaliter pseudorugosis ; fronde tripinnata; pinnis longistipitatis ; pinnis” brevistipitatis, ca. 3-paribus ; pinnulis utroque latere ca. 5, infimis solumodo sti- pitatis, late cuneatis, argute serratis, acutis vel saepius acumin- atis, subcoriaceis, pallidis, superioribus majoribus 5 cm longis, 12 mm latis; venis simplicibus, sub lente hyalinis; soris a mar- gine remotis, 3- ad 7-locularibus ; indusio conspicuo. Sarawak, Mt. Poe, alt. 900 m, in damp valley district. Brooks 138. Remarkable for the contorted transverse thickenings on the main axis. Dryopteris aquatiloides Copel. sp. nov. Nephrodium D. salicifoliae subsimile et cum ea confuso; rhi- zomate 2 mm crasso, breviter repente; stipite frondis sterilis 10 ad 15 fertilis 20 ad 25 cm alto, stramineo, sursum minutis- sime pubescente, valido, rhachi costique densius pubescentibus ; fronde 15 ad 25 cm alta, 12 ad 17 cm lata, pinna apicale aliis simile; pinnis utroque latere ca 6, infimis baud diminutis, stipi- tatis, praecipue superioribus unilateraliter subauriculatis, usque ad 9 cm longis et 1 cm latis, obtusis, inferioribus vel omnibus utrinque angustatis, integris vel crenulatis, papyraceo-coriaceis, lamina glabrescente ; venulis utroque latere 2 vel 3; soris parvis, medialibus, interdum coalescentibus, indusio rudimentario. Sarawak, Bungo Range, Brooks 9. This is the plant referred to as Dryopteris salicifolia in my description of the Papuan D. aquatilis.^ Mr. Brooks has corrected my mistake, which was not a verj’^ reasonable one. ^ This Journal 6 (1911) Bot. 75. 59 60 COPELAND. Oryopteris porphyricola Copel. sp. nov. Nephrodium, rhizomate erecto; stipite ca. 10 cm alto sparse ferrugineo-paleato ; fronde 80 cm alta, 27 cm lata; pinnis in- fimis abrupte in auriculas auriculatas integras reductis, parte deinde angusta 25 cm alta; pinnis sequentibus deflexis non re- ductis, basi truncatis, 14 cm longis, 22 mm latis, 4 ad ad costam pinnatifidis, supra appresso-sericeis, infra minute pilosis et dense glandulosis, membranaceo-papyraceis ; lobis ca. 3.5 mm latis, falcatis; venis utroque latere ca. 7 quarum 2 vel 3 anas- tomosantibus ; soris parvis, medialibus; indusio glanduloso et breviter piloso, persistente. Sarawak, Bau, Jebong Valley, on porphyry dyke. Brooks 112. In the same general group as the following. More like Dryopteris jacu~ losa (Christ) C. Chr. than any other species known to me. Dryopteris angustipes Copel. sp. nov. Nephrodium, rhizomate breve et crasso, paleis brunneis 5 mm longis vestito; stipite 1 ad 2 cm alto, valido, paleaceo; fronde 80 cm alta, 20 cm lata, rhachi straminea, nisi in sulco glabrescente ; pinnis inferioribus subabrupte in auriculas multas pinnatifidas utroque latere ca. 15 reductis, parte indeque an- gusta 15 ad 20 cm alta; pinnis medialibus horizontalibus, proxi- mis, sessilibus, acuminatis, supra basin acroscopice auriculatum 13 ad 15 mm latis, }j ad costam pinnatifidis, costa costulis et venis infra minute appresso-pilosis, costa glabrescente, lamina infra minute glandulosa supra sparse albo-pilosa, membranacea ; lobis 2 ad 2.5 mm latis, subfalcatis; venis utroque latere ca. 7 quarum 1 vel 2 anastomosantibus ; soris medialibus, parvis; indusio in centre glanduloso, alibi sparse piloso vel glabro, persistente. Sarawak, Single, in swamp near foot of mountain. Brooks 110. This keys out with van Alderwerelt’s “Malayan Ferns” as his Dyropteris sumatrana {Nephrodium violle var. major Bedd.) which he seems not to have seen and which has never had an adequate specific diagnosis, and may be suspected of being D. stipellata (Bl.) 0. Ktze. J. Smith, in a marginal note in his copy of Blume’s “Enumeratio,” construed Aspidium stipellatum as A. molle. Dryopteris angustipes is not merely exceedingly distinct in appearance from D. parasitica but differs in the character of the pubescence from that species to such an extent that Beddome could not have placed them together. Lomagramma Brooksii Copel. Mr. Brooks has sent me a very complete specimen, collected at Bau. It has the fertile and sterile fronds distinct, as in other species of the genus. The frond of a juvenile plant “which germinates on the limestone and may grow in large masses before reaching a trunk, and becoming scandent”, has membranaceous, serrate, and very oblique pinnae. Fronds of adult NEW SARAWAK FERNS. 61 plants are as described, except in size. The sterile fronds are a meter or more long and 30 cm wide. The fertile frond is still wider, its pinnae 2 to 4 mm wide, and straight or curved. The species is nearest to Loviagramma perakensis Bedd., differing con- spicuously in the very short stipes and persistent scaliness. Athyrium sorsogonense (Presl) Milde, var. poetise Copel. var. nov. Forma laxa et grandis, pinnis remotis infimis usque ad 25 cm longis, 2.5 ad 3.5 cm latis, rhachi mox glabrescente. Sarawak, Mount Poe, alt. 900 m, Brooks US. Athyrium carnosum Copel. sp. nov. Species gregis A. maximi vel A. pinnati, fronde pinnata, 45 cm alta, fere 30 cm lata, rhachi furco excepto glabra vel mox glabrescente; pinnis liberis utroque latere ca. 10, subsessilibus vel sursum adnatis, horizontalibus, falcato-acuminatis, basi late cuneatis, 3.5 cm latis, ^ ad I- ad costam pinnatifidis, glabris, carnoso-subcoriaceis, viridibus; lobis 9 mm latis, rotundatis; venis ca. 9, angulo acuto distantibus, simplicibus; soris lineari- bus, usque ad 10 mm longis, paginam totam complentibus, in- dusio angusto mox sporangiis occulto. Sarawak, Matang, collected by native collector of the Sarawak Museum. The pinnae look like pinnules of Athyrium maximum, but are remark- ably large and coarsely cut and peculiar also in the clear but deep-green color and almost fleshy texture. Athyrium (Diplazium) polycarpum Copel. sp. nov. Stipite teste Brooks 60 cm alto et paleis paucis brunneis ovatis vestito; fronde 75 cm alta, 45 cm lata, sulcis axium exceptis glabra, deorsum hand angustata, sub apice pinnatifido bulbifera, rhachi brunnea, deorsum sparsissime argute spinosa; pinnis stipitatis, subacuminatis, basi inaequaliter truncatis, infimis 25 cm longis, 45 mm latis, praecipue deorsum fere ad costa pinna- tifidis cum segmentis ibidem solumodo remotis; segmentis ca. 2 cm longis, 1 cm vel ultra latis, rotundatis vel truncatis, ob- scure serrulatis, supra punctis albis marginalibus donatis, coria- ceis; venis utroque latere ca. 11, simplicibus vel furcatis, omni- bus soriferis; indusio lineare, atro. Sarawak, Mount Penrissen, alt. 1,050 m. Brooks 151. Also Bidi, Brooks, Aug., 1908. Intermediate between the groups of Athyrium sorsogonense and A. maximum. Athyrium muricatum Copel. sp. nov. Species gregis polymorphae A. maximi (Don) Copel., rhizo- mate 12 mm crasso; stipite 30 cm alto, 4 mm crasso, spinoso, basin versus paleis lanceolato-ovatis incurvatis sordide brunneis 5 mm longis vestito; pinnis infimis 20 ad 25 cm longis, 5 cm 62 COPELAND. latis, stipitatis, acuminatis, ad basin vix ad costam pinnatis, sursum sensim leviter pinnatifidis, segmentis oblongis, 10 mm latis, obtusis vel subacutis, plerisque apices versus serratis, glabris, coriaceis, infra pallidioribus ; venulis utroque latere ca. 12, furcatis, segmentis infimis reductis; parte apicale frondis sensim per pinnas sessiles lobatas et adnatas serratas ad apicem grosse pinnatifidum angustata; soris a costula remotis, 1 ad 2 mm longis, indusio mox evanescente. Sarawak, No. 35 of the Sai'awak Museum collection. Distinguished from most others of its group by the scattered spines on the stipe, and from all by the short sori near the apices of the veinlets, and the transient indusia. Athyrium Hewitti Copel. sp. nov. Species gregis A. cyatheifolii Milde, fronde 75 cm alta, 30 cm vel paullo ultra lata, rhachi nisi in sulco glabra, castanea ; pinnis pinnatis utroque latere ca 8, pinnatifidis, ca. 13, brevistipitatis, infimis 23 cm longis, 5 cm latis, oppositis, acuminatis, pinnulis horizontalibus et rectis, subsessilibus, basi truncatis plerisque sursum vel utroque latere subhastatis, subacutis, supra basin 1 cm latis, ^ ad costam incisis, sursum integrioribus, glabris, papy- raceis; lobis 2 mm latis; venis furcatis, obliquis; soris 2 ad 4 mm longis, linearibus, indusio laete brunneo. Sarawak, Matang, collected by J. Hewitt, the former curator of the Sarawak Museum, to whom the species is dedicated. Athyrium sarawakense Copel. sp. nov. Species gregis A. cyatheifolii Milde, stipite 3 mm crasso, sparse spinuloso, glabro; fronde 45 cm lata, ut videtur 75 cm alta, bipin- nata, rhachi, sulco et alls exceptis, glabra, laete castanea ; pinnis brevistipitatis, 25 cm longis in caudam argute serratum protrac- tis, pinnis liberis paucis, sessilibus, acutis, usque ad 5 cm longis, 1 cm latis, serratis, basi truncatis, subhastatis, glabris, subcoria- ceis; venis furcatis, obliquis; soris 2 ad 3 cm longis, costalibus, indusio laete brunneo. Sarawak, Mount Matang, J. Hewitt, .April, 1908. Athyrium confertum (Baker) Copel. Asplenium confertum Baker. Ann. Bot. 8 (1898) 125. Sarawak, Mount Bongo Brooks The Philippine Journal of Science, C. Botany. Vol. VII, No. 2. June, 1912. NEW PAPUAN FERNS. By Edwin Bingham Copeland. (Fro«i the College of Agriculture, University of the Philippines, Los Banos.) Stenochlaena intermedia Copel. sp. nov. Rhizomate ca. 1 cm crasso, basique stipitis paleis atrocas- taneis 15 mm longis, 1.5 mm latis sat dense vestito et deinde ob baseos duras spiniformes earum horrida; stipite 40 cm vel alto, valido, sursum rhachique paleis latioribus vestitis et baud spinescentibus ; fronde 60 ad 70 cm alta, pinnata, pinnis ad rhachin articulatis; pinnis utroque latere ca. 8, frondis sterilis subsessilibus, basi plus minus obliquis, rotundato-truncatis, apice abrupte breviter caudatis, 20 ad 22 cm longis, 45 ad 55 mm latis, integris, papyraceo-coriaceis, glabris, venis liberis; frondis fer- tilis pinnulis stipitatis (stipitibus 4 ad 8 mm longis), 15 ad 20 cm longis, 2 mm latis, sursum sensim angustatis. Papua, Ambasi, King 370. This fern has the deciduous pinnae of S. aculeata (Bl.) Kunze, but otherwise is rather suggestive of Lomariopsis. In general appearance it suggests Lomagramma rather than any Stenochlaena. Humata tenuis Copel. sp. nov. Rhizomate 2 ad 3 mm crasso, paleis brunneis dense vestito; stipitibus frondium sterilium ca. 3 fertilium ca. 7 cm altis, paleaceis; fronde sterile 5 ad 8 cm alta, deltoidea, tripinnatifida, pinnulis oblongo-linearibus, coriaceis, glabris, rhachibus palea- ceis; fronde fertile 10 ad 13 cm alta, deltoidea, acuminata, quadripinnatifida ; pinnulis infimis oblongo-linearibus, in seg- mentis angustissimis pinnatifidis; soris apud baseos segmenta- rum, indusio grande. Papua, Gira Track, King 367. Near H. vestita, but much more finely cut than is this species or H. introrsa Christ. This is possibly Davallia Chaerophylla Ces. but that name was used earlier. 67 68 COPELAND. Humata dimorpha Copel. sp. nov. Rhizomate 2 mm crasso, paleis castaneis lanceolatis 6 mm longis vestito; stipitibus glabris, gracilibus, frondis sterilis 5 cm fer- tilis 7 cm altis; fronde sterile 5 cm alta, 4 cm lata rhachi sor- dide squamulosa, pinnis utroque latere sub apice late crenulata ca. 4, fere 2 cm longis, infimis 1 cm latis, subintegris, basi an- guste adnatis, truncatis, obtusis, sequentibus paullo minoribus glabris rigide coriaceis; fronde fertile 13 cm alta, 8 cm lata, valde acuminata; pinnis inferioribus 3-paribus aequalibus, 4 ad 4.5 cm longis, usque ad 12 mm latis, stipitatis, rigide coriaceis; pinnulis remotis, infimis solummodo liberis, vix quam sequen- tibus majoribus, 2 ad 3 mm latis, utroque latere in dentes 2 ad 4 triangulares pinnatifidis, soro quoque dentem complente, indusio lato, interdum quam alto duplo latiore, centro nigro. Papua, Lakekamu, King 326. A very distinct species. Davallodes viscidulum (Mett.) v. A. v. R. Davallia viscidula Mett. (1869) This species has the indusial characters of D. gymnocarpum, but the frond is more lax and finer cut, v^ith the pinnules next to the rhachis dilated. Lindsaya papuana Copel. sp. nov. Synaphlebium gregis L. decompositae Willd. pinnis vel pin- nulis integris soro continuo; fronde pinnata pinnis usque ad 3 cm longis vel tripartite et pinnulis ca. 2 cm longis; pinnulis brevistipitatis, apice obtusis, integris, maximis rarius incon- spicue incisis ; soro fere marginale, lamina ultra venam sori- feram bullata. Papua, Lakekamu, King 358. Lindsaya azurea Christ is a plant vrith much smaller and truncate pinnules. L. nitida Copel., of Borneo, has pinnules of quite different shape, the base hardly truncate and apex broadly rounded, with the sorus at the end. The pinnules of L. papuana are broadest at the base and abruptly truncate, and the apex is obtuse rather than rounded. The sorus extends from the point next to the rhachis to the apex, almost always without a break. Polypodium linguaeforme Mett. King 368, 190, and 21 in part, are P. linguaeforme, judging by diagnosis only: I have not seen what I could feel sure was this species. These specimens fall well within the range of variation of Polypodium musi- folium Bl. The Philippine Journal of Science, C. Botany. Vol. VII, No. 2. June, 1912. CHARACEAE FROM THE PHILIPPINE ISLANDS. By H. and J. Groves. {London, England.) 1. NITELLA Agardh. 1. Nitella acuminata Braun in Hook. Journ. Bot. 1 (1849) 292. Luzon, Benguet Subprovince, Trinidad, in a roadside ditch, altitude about 1,200 m. Bur. Sci. 14-108 Robinson, May, 1911. Mindanao, District of Zamboanga, “im Graben bei Sambanga” (Zamboanga), Wichura 2005, 1862, fide Braun Fragmente 37. This species is widely distributed especially in the Tropics. It occurs in southern Asia from India eastwards, in Borneo and in several of the other Islands of the Malay Archipelago, also in Africa and in North and South America. Like the other widely distributed species it is variable. Braun has described two varieties from this region, Belangeri from India, char- acterized by its large size, stout stem, and short end-segments to the branchlets, and indica from Java and the Philippine Islands, being more slender and having longer end-segments. The Trinidad plant is a stout form of the latter. 2. Nitella Roxburghii Braun in Hook. Journ. Bot. 1 (1849) 292. N. polyglochin Braun Monatsb. Berk Akad. (1867) 816. Luzon, Benguet Subprovince, in a streamlet, Daklen to Kabayan, Merrill 4867 (young), October, 1905; Baguio, Merrill 4868, November, 1905, altitude about 1,500 m, Merrill 7928, May, 1911, in small still streams: Province of Bataan, Lamao River, Mount Mariveles, Elmer 6760, November, 1904, Merrill 7625, March, 1911, Limay, For. Bur. 19160 Curran, December, 1909. Mindoro, Binabay River, in stagnant pools, border of forests, altitude about 10 m, Merrill 5686, November, 1906. Fairly normal forms of this species, which occurs also in Ceylon, southern India, lower Burma, Nicobar Islands, Borneo, Java, Celebes, Sumbawa, new Guinea, and Australia (Endeavour River). The closely allied Nitella Mauritiana is found in Mauritius. Braun in his later writings applied to this species the name of Nitella polyglochin, which he had previously used in a wider sense, but there is no valid reason for discarding the earlier name. 2. CHARA Linn. 3. Chara corallina Willd. in Mem. Acad. Berol. 1803 (1805) 89, t. II, f. 2. Philippines, Llanos, 1853, in Herb. DeCandolle, fide Braun Fragmente 108. Braun described this plant as var.? basilaris on account of the fruits being produced at the base only, and not also at the nodes of the branchlets, 69 70 H. AND J. GROVES. as in the type, and quotes for it Chara congesta Spreng. var. P. Fr. Ant. Llanos, Fragm. de algunas Plantas de Filipinas (1851) 112. 4. Chara Braunii Gmelin Flora Badens. Alsat. 4 (1826) Suppl. 646. {Chara coronata Auct.) Luzon, Bontoc Subprovince, Lessep, Father M. V anoverbergh 590, June, 1910, in still water, altitude 1,100 m. A specimen from Benguet Subprovince, collected by Major E. A. Mearns, Bur. Set. 2825, probably also belongs to this species, but may possibly be a very small form of Chara corallina. Being old and without fruit, it cannot be determined -with certainty. Chara Braunii is almost world-wide in its distribution, occurring in all the continents. The specimen from Bontoc may be characterized as /. ^neiocarpa microptila unilateralis. 5. Chara Benthami Braun in Monatsb. Berl. Akad. (1867) 799. A specimen from stagnant fresh water from near sea-level, Bucas Island, a small island northeast of Mindanao, Merrill 5265, October, 1906, is, we think, best referred to this species. Braun separated Chara Benthami from C. gymnopitys, to which it is closely allied, by its having one stipulode to each branchlet instead of two as in the latter species, but in the original specimen from Hongkong in the Kew Herbarium, this character is not constant. 6. Chara gymnopitys Braun in Linnaea 25 (1852) 708. SiBUYAN, Magallanes, Elmer 12382, April, 1910. This species has been found in Socotra, India, Ceylon, Straits Settlements, Tonkin, China, New Guinea, Australia, New Zealand, and Tasmania, also in the United States of America. The plant from the Philippines may be characterized as /. tylacantha macracantha meiocarpa. 7. C. flaccida Braun in Hook. Journ. Bot. 1 (1849) 296. Luzon, Manila, in shallow pools (about 20 cm), IF. R. Shaw 1173, February, 1912. In describing Chara flaccida, Braun separated it from C. Hydropitys by the yellow oospores and the always uncoated branchlets. C. gymnopitys was subsequently described with black or nearly black oospores, but it is so very similar in other respects to C. flaccida that we think a more extended knowledge of its various forms may prove that it should be regarded as a subspecies. 8. Chara brachypus Braun in Hook. Journ. Bot. 1 (1849) 298. Luzon, Province of Rizal, in rather swift water of the Maraquina River, near the Manila waterworks dam at the Montalban Gorge, Merrill 5098, March, 1906. A fairly typical form of this rare species which occurs in India and New Guinea, also in Africa and in Australia. 9. Chara zeylanica Willd. in Mem. Acad. Berol. 1803 (1805) 86, t. II, f. 1. C. gymnopus Braun, sens. lat. Luzon, Province of Rizal, Maricaban, near Manila, Merrill 7^62, Feb- ruary, 1911, in shallow muddy pools of fresh water at sea level; Pasay, Merrill 7586, April, 1911, in stagnant pools of fresh water at sea level. The commonest of the tropical species of Characeae, occurring in India, Ceylon, Burma, and Java, as well as in Africa, North and South America, and Australia. The plants from Luzon come under Braun’s variety ceylonica. TfiE Philippine Journal of Science, C. Botany. Vol. VII, No. 2. June, 1912. SERTULUm BONTOCENSE. NEW OR INTERESTING PLANTS COLLECTED IN BONTOC SUBPROVINCE, LUZON, BY FATHER MORICE VANOVERBERGH. By E. D. Merrill. {From the Botanical Section of the Biological Laboratory, Bureau of Science, Manila, P. I.) In the year 1909 Father Morice Vanoverbergh, who had re- cently come to the Philippines from Belgium, was assigned to the mission station at Bauco, Bontoc Subprovince, Luzon, and shortly after his arrival there entered into correspondence with the writer in regard to botanical matters. Father Vanover- bergh having .expressed his deep interest in the study of botany, and a desire to become acquainted with the flora of the region in which he was located, an offer was made to determine, so far as possible, whatever he might submit for identification ; and a small collecting outfit was sent to him. He commenced his botanical exploration of Bauco and vicinity in January, 1910, and continued it during his leisure hours until August, 1911. In this time he collected and transmitted to the Bureau of Science for determination about 1,400 numbers of plants, representing approximately 1,300 different species. The material was identi- fied from time to time as it was received, and lists of names were sent to the collector, with special reference to those forms of which additional or more complete material was desired. In general collections received here, there is always a certain per- centage of species represented which for lack of complete material can not be accurately determined, especially in those families having dioicious flowers, and in those groups where our classification is based on a combination of flower and fruit characters. Father Vanoverbergh has assiduously searched for the additional material required and accordingly I have been able to determine with a reasonable degree of accuracy a high percentage of the material collected by him. 71 72 MERRILL. The present paper is based primarily on specimens collected by Father Vanoverbergh ; but in a number of cases I have been able, with aid of his material, to complete the determinations of specimens previously secured by other collectors and which for the most part had been determined only to the genus or family, or in some cases erroneously identified. From a study of his material I have been able to record, in the present paper, three genera, Microcarpaea, Trachelospermum, and Teucrium, previously unknown from the Archipelago, and two, Microchloa and Blyxa, which were included by Fathers Villar and Naves in the “Novissima Appendix” to the third edition of Blanco’s “Flora de Filipinas,” and whose records have not previously been veri- fied. An apparently undescribed genus, Vanoverberghia, has been discovered in the collection, and I have been able to reduce one genus, Cleistoloranthus, previously proposed by me, so that the number of genera definitely known from the Archipelago has been increased by five. As to species originally described from extra-Philippine material but previously not definitely known from the Archipelago, I have been able to record the following eleven ; Blyxa octandra Planch., Microchloa setacea (L.) Merr., Scleria per gracilis Kunth, Cocculus laurifolius DC., Rhynchosia volubilis Lour., Lysimachia capillipes Hemsl., L. decurrens Forst., Microcarpaea minima (Koenig) Merr., Acro- cephalus indicus O. Ktze, Utricularia exoleta R. Br., and Cam- panumoea truncata Endl. Thirty-one species are described as new. I have previously described one new species of Begonia from Father Vanoverbergh’s material, B. vanoverberghii Merr. Dr. C. B. Robinson has indicated two new species of Elatostema, E. scapigerum C. B. Rob., and E. variabile C. B. Rob., in the same collection. Several new species of orchids have been described by Mr. Ames,^ while additional ones have been indicated for future publication. A study of Father Vanoverbergh’s material has therefore increased our knowledge of the Philippine flora by six genera and over sixty species. I have dedicated the new genus Vanoverberghia, described below, to its discoverer and collector, and as its specific name I have associated with it the appellation sepulchrei in honor of the late Father Jules Sepul- chre, Father Vanoverbergh’s colleague and companion at Banco, and the founder of the mission station at that place. Father Vanoverbergh informs me that Father Sepulchre took a lively interest in the botanical work and was of much assistance to Supra 1-27. SERTULUM BONTOCENSE. 73 him in his botanical labors. Father Sepulchre died very sud- denly in Bontoc on February 13th, 1912, and was buried in the church erected by himself. It is but a very slight acknowl- edgement of Father Vanoverbergh’s services to Philippine botany to fulfill his request and to associate with the genus dedicated to him the name of his regretted friend, colleague, and companion. In the year 1909 I described a new genus of the Loranthaceae, Cleistoloranthus,^ which on reexamination and comparison with excellent material representing the same species, secured by Father Vanoverbergh, I am now able to reduce to LorantJms. The type of the genus and species proves to have been a spec- imen with immature flowers. Cleistoloranthus may still be worthy of sectional rank on account of the peculiar appendages to the petals, but it can hardly be retained as a distinct genus if we follow the limits of genera generally accepted in this family. The types of the new species indicated in the present paper are preserved in the Herbarium of the Bureau of Science, which also contains a complete or nearly complete set of all the botan- ical material collected in Bontoc by Father Vanoverbergh. He has also sent some material to the University of Louvain, Bel- gium, and many duplicate orchids have been supplied to Mr. Ames. Some duplicate material still remains to be distributed by the Bureau of Science. Father Vanoverbergh’s botanical work in Bontoc was inter- rupted in August, 1911 by his transfer to a new station in Union Province, but on the death of Father Sepulchre he was reas- signed to the Banco station, where he plans to continue his botanical explorations. HYDROCHARITACEAE. BLYXA Thouars. Blyxa octandra (Roxb.) Planch, ex Thwiates Enum. PI. Zeyl. (1859-64) 332. Vallisneria octandra Roxb. PI. Corom. 2 (1798) t. 165. Blyxa roxburghii Rich, in Mem. Inst. Fr. (1811) 77, t. 5; Hook. f. FI. Brit. Ind. 5 (1888) 660; Naves Novis. App. (1880) 214. Luzon, Bontoc Subprovince, Bauco, Vanoverbergh 66^, July, 1910, in rice paddies, altitude about 1,250 m. Naves’ record for this species as a Philippine one has not previously been verified. Tropical Asia through Malaya to Australia, and the Caro- line Islands. ’This Journal 4 (1909) Bot. 150. 74 MERRILL. GRAMINEAE. MICROCHLOA R. Br. Microchloa indica (Linn, f.) comb. nov. Nardus indica Linn. f. Suppl. (1781) 105. Rottboellia setacea Roxb. PI. Coromandel 2 (1798) 18, t. 132. Microchloa setacea R. Br. Prodr. (1810) 208; Hook. f. FL Brit. Ind. 7 (1897) 283; F.-Vill. Novis. App. (1883) 319. Luzon, Bontoc Subprovince, Bauco, Vanoverhergh 764, August, 1910, altitude about 1,300 m: Rizal Province, Novaliches, Loher 7179, June, 1904. A species widely distributed in the Tropics, previously reported from the Philippines by Fernandez-Villar, but his record not before verified. The earliest valid specific name is here adopted. CYPERACEAE. CLADIUM P. Br. Cladium cyperoides sp. nov. Planta tenuis, perennis, plus minusve pilosa, foliis omnibus radicalibus, tenuibus, subrigidis, 4 ad 8 cm longis, circiter 1 mm latis; inflorescentiis longe et tenuiter pedunculatis, umbellatis, simplicibus vel compositis more Fimbristylis diphyllae; spiculis compressis, 5 ad 8 cm longis, oblongis, glumis puberulis, om- nibus vel inferioribus exceptis distichis. A slender, tufted, erect, perennial plant, the leaves and culms slightly ciliate-pubescent with white hairs, the leaves slender, all basal, flat or involute, rather rigid, 4 to 8 cm long, usually less than 1 mm wide, acute, the sheaths inflated, purple, enclosing the base of the stems. Inflorescence long-exserted, the slender leafless peduncle 10 to 30 cm long, deeply sulcate, about 0.5 mm thick. Inflorescence a simple or compound umbel, similar to that of Fimbristylis diphylla and allied species, rather dense, or lax, the spikelets fascicled or solitary, some sessile, others pedi- celled, 5 to 20 in each inflorescence, the inflorescence subtended by 1 to several narrow, rigid, 0.5 to 2 cm long, leaf-like bracts which are inflated below and pubescent; rays few, up to 5 cm in length, usually much shoi’ter. Spikelets brown, oblong, com- pressed, 5 to 8 mm long. Glumes all distichous, or the lower one or two spiral, puberulent, acute or acuminate; first glume empty, broadly ovate, 2.5 to 3 mm long, apiculate-acuminate, the apiculus 0.6 mm long; second glume 4 mm long, about 2 mm wide, acuminate, somewhat keeled, apiculate, 1-nerved, contain- ing a perfect flower. Anthers 3, 1.8 mm long; style thickened at the base, jointed with the nut, about 3 mm long, with three 2 mm long arms; hypogynous bristles none. Nut white, obovoid, 3-angled, minutely verrucose, about 1 mm long; style deciduous. SERTULUM BONTOCENSE. 75 Succeeding glumes similar to the second one, all subtending per- fect flowers, usually 6 or 7 flowering glumes in each spikelet. Luzon, Bontoc Subprovince Bauco, Vanoverbergh 273, hillsides, altitude about 1,320 m, first collected in March, 1910, again in May or June, 1910, and more abundant and representative material in May, 1911. A curious species which may possibly be referable to some other genus than Cladium. It has very much the habit of various species of Fimbri- stylis of the F. diphylla group, but is at once distinguished from that and allied genera by its glumes being distichous, not spiral, or only the lower one or two spiral, very strongly resembling the spikelets of Cyperus in this character. It is anomalous in Cladium in its habit, its very slender, slightly pilose leaves, which are all basal, in its Fimbristylis-\ike inflores- cence, its spikelets with mostly perfect flowers, and its glumes mostly dis- tichously arranged. When first received I could not refer it to its proper genus, and speci- mens were accordingly sent to Rev. G. Kiikenthal, who reported it as probably referable to Cladium; in the meantime the second collection having been received, I had also tentatively placed the material in Cladium. The description has been drawn up with reference to all three collections. SCLERIA Berg. Scleria pergracilis (Nees) Kunth Enum. 2 (1837) 354; Clarke in Hook, f. FI. Brit. Ind. 6 (1884) 685. Hypophoriim pergracile Nees in Edinb. Phil. Journ. 17 (1834) 267. Luzon, Bontoc Subprovince, Bauco, Vanoverbergh 889, October 5, 1910, on hillocks, altitude about 1,400 m. Mindanao, District of Cotabato, near Makar, Copeland, December, 1911. Not previously reported from the Philippines; Tropical Africa, through India and Ceylon to Yunnan. Remarkable for its lemon-scented leaves. ZINGIBERACEAE. ALPINIA L. Alpinia vanoverberghii sp. nov. § Probolocalyx. Herba erecta circiter 2 m alta, subtus foliis minute puberulis, inflorescentiis terminalibus pubescentibus ; foliis oblongis, brev- iter petiolatis, usque ad 30 cm longis, 11 cm latis; racemis simplicibus quam folia longioribus, multifloris, bracteis brac- teolisque nullis, pedicellis inter se remotis, solitariis vel binis, vix 2 cm longis; floribus pro genere magnis, albis, 5 ad 6 cm longis, connectivo vix appendiculato. An erect herb about 2 m high. Leaves alternate, distichous, the sheaths of the uppermost ones frequently 30 cm long, rather loose, glabrous, the ligule chartaceous or subcoriaceous, 1.5 cm long or less, usually inequilaterally 2-lobed, the petioles 1.5 cm long or less ; leaf-blades oblong, subcoriaceous, up to 30 cm long, 11 cm wide, glabrous and somewhat shining above, beneath slightly paler, shining, more or less puberulent, the base broadly 76 MERRILL. rounded or somewhat acute, the apex with a slender acumen. Racemes simple, terminal, as long as or longer than the leaves, the rachis and pedicels rather densely cinereous-pubescent, the pedicels 1.8 cm long or less, solitary, rarely in pairs, mostly 1.5 to 2 cm apart. Flowers white, 5 to 6 cm long. Calyx in bud cylindric, in anthesis spathaceous, split to the base down one side, 3 to 3.5 cm long, when spread 2.2 cm wide at the base, about 1 cm wide near the apex, the apex with three subequal, oblong-ovate teeth 3.5 to 5 mm long, densely pubescent on the back, the remainder of the calyx slightly pubescent on the back, obscurely about 17-nerved. Corolla exserted from the base of the calyx and at right-angles to it, the tube about 1.5 cm long, somewhat enlarged upwards, and nearly 1.5 cm wide at the apex. Corolla-lobes subequal, narrowly oblong, about 4 cm long, 1 cm wide, blunt, wrinkled when dry. Lip as long as the petals, about 2 cm wide above, apparently more or less cucullate, very much wrinkled when dry. Stamen-filament stout, broad at the base, when spread nearly 1 cm wide, narrowed upwards, 3 cm long; anther oblong, 2.2 cm long, nearly 5 mm in diameter, the con- nective not crested; staminodes linear, 1.5 cm long. Ovary pubescent, ovoid or obovoid, 3-celled, 5 to 8 mm long. Luzon, Bontoc Subprovince, Bauco, on hillsides, altitude about 1,250 m, Vanoverbergh 573, June 11, 1910, locally known to the Igorot as calawhi. A species manifestly belonging in the subgenus Probolocalyx, the first one of this group to be found in the Philippines. It is well characterized by its large flowers, crestless anthers, rather lax, racemose inflorescence, and in the entire absence of bracts and bracteoles. VANOVERBERGHIA gen. nov. Calyx in alabastro cylindraceus, demum spathaceo-fissus, apice latus, planus, 3-dentatus. Corollae tubus calyce brevior, lobi anguste oblongi, subaequales, 2 antici alte ad connati. Staminodia lateralia, linearia vel filiformia, elongata. Labellum petalis anterioribus alte adnatum, bifidum, lobi anguste lineares. Filamentum elongatum, concavum ; antherae loculi discreti, connectivo sublatiusculo, concavo, non producto. Ovarium 3- loculare, loculis oo-ovulatis; stylus filiformis, stigmate sub- ovoideo, ciliolato ; nectaria bina, crassa, compressa, erecta. Herba altissima, aromatica, e rhizomate crasso. Inflorescentia terminalis, racemosa, suberecta vel nutans. Flores in quaque bractea solitarii, bracteolis nullis. Vanoverberghia sepulchre! sp. nov. Planta erecta, robusta, usque ad 4 m alta, glabra, aromatica; foliis oblongis, usque ad 40 cm longis, longe caudato-acuminatis. SERTULUM BONTOCENSE. 77 breviter petiolatis, vaginis superioribus inflatis ; racemis termin- alibus, suberectis vel nutantibus, bracteis calycibus subaequi- longis, persistentibus, bracteolis nullis; floribus roseis, 4 ad 4.5 cm longis, petalis anterioribus alte connatis; labelliim bifidum, membranaceum, lobis albidis, linearibus. A tall, erect, glabrous, coarse plant, with a strong anise-like odor, reaching a height of 4 m, the rhizome 5 cm in diameter, red outside, the stem green, 5 cm in diameter below, higher up 2 cm in diameter, and at the apex 1 to 1.5 cm thick. Leaves alternate, distichous, oblong, subcoriaceous, the lower ones rather short, gradually increasing in length upwards, the longest ones at the top of the stem, up to 40 cm long and 10 cm wide, the apex slenderly caudate-acuminate, the acumen about 5 cm long, the base acute or somewhat rounded and shortly decurrent-acu- minate, the petiole short, 1 cm long or less ; nerves very numerous, slender, ascending, densely arranged; upper sheaths inflated, coriaceous, up to 20 cm long, when spread up to 6 cm wide, longitudinally striate, the upper ones tipped with much reduced leaves, or the uppermost one subtending the inflores- cence acuminate and with no vestige of a leaf. Inflorescence terminal, racemose, peduncled, at length nodding, the flower- bearing part up to 20 cm long. Flowers very numerous, about 4.5 cm long, pink nutside, white within, densely disposed, one to each bract, the bracts brown, oblong, persistent, somewhat boat- shaped, 2.5 to 3 cm long, 8 to 10 mm wide, much wrinkled when old; bracteoles wanting. Pedicels stout, 5 to 10 mm long, often twisted. Calyx in bud cylindric, in flower spathaceous, split down one side nearly to the base, above flattened and about 8 ’mm wide, about 10-nerved, the apex with 3 oblong-ovate, obtuse, 3 mm long teeth. Corolla-tube about 1.7 cm long, the posterior lobe narrowly oblong, 3 cm long, 6 to 7 mm wide, about 7-nerved, obtuse, slightly cucullate at the apex, flat below, the base boat- shaped, inclosing the stamen, the upper one-half ultimately reflexed or spreading; anterior lobes about as long as the pos- terior one and similar to it, united for one-third to one-half their length, keeled below the union, the free parts flat, 5- to 7-nerved, apex obtuse and slightly cucullate. Lip sessile, membranaceous, adnate to the anterior petals as high as the cleft along the middle, the margins free, narrow, cleft above the part united to the petals into two linear, thin, flexuous lobes about 1.8 cm long, 2 mm wide below, narrowed upwards, the lobes about as long as the free parts of the anterior petals. Stamen somewhat curved, the filament stout, cylindric, about 2 cm long and 2.5 mm thick, 78 MERRILL. channeled on the inside and inclosing the style; anther about 12 mm long, 3 mm thick, somewhat curved, the connective rather broad, but semi-cylindric, so that the lateral anther-cells are contiguous, not produced, truncate or retuse at the apex. Stam- inodes linear, 1 to 1.5 cm long, thin, flexuous, 1 mm wide or less from a somewhat broader base, attached at the angle between the lip and the posterior petal. Ovary 3 to 4 mm long, 3-celIed ; ovules many, spreading, attached along the inner angles of the cells; style slender, surrounded by the hollowed filament below, emerging above, nearly as long as the stamen; stigma obovate, about 2 mm long, somewhat oblique, truncate or hollowed, minutely ciliate. Nectaries 2, suborbicular, compressed, about 2 mm in diameter, one on each side of the style at the base but free from it. Fruit edible, subellipsoid, 2 to 2.5 cm long, the pericarp coriaceous, very tardily dehiscent. Seeds numerous, about 3 mm long. Luzon, Bontoc Subprovince, Malamey (Pinggat), Vanoverbergh 956, October 19, 1910, and mature fruits August 17, 1911, along streams at an altitude of about 1,500 meters, locally known to the Igorots as agbab. This genus is closely allied to Riedelia Oliver, and appears to be some- what intermediate between that genus and Alpinia. It differs from Riedelia in its rather large floral bracts, and especially in its two anterior petals being united for one-third to one-half their length, its narrow and mem- branaceous lip being adherent to the two anterior petals as high as the cleft between them, the lip there split into two linear lobes as long as the free parts of the anterior petals. The genus is dedicated to Father Morice Vanoverbergh, by whom it was discovered, in appreciation of his intensive and enthusiastic botanical exploration of the region in and about Bauco where his mission station is located. The species is dedicated to his colleague. Father Jules Sepulchre who came to the Philippines in November, 1907, and whose untimely death occurred in Bontoc February 13, 1912. Father Sepulchre established the Bauco mission, and rendered much assistance to Father Vanoverbergh, who was associated with him in the missonary work, in the latter’s botan- ical explorations. LORANTHACEAE. LORANTHUS L. Loranthus appendicuEatus nom. nov. Cleistoloranthus verticillatus Merr. in Philip. Journ. Sci. 4 (1909) Bot. 150, 5 (1910) Bot. 345, non Loranthus verticillatus Ruiz & Pav. This was originally described by me as a new genus and species, on the basis of a single collection from Mount Pulog, Bengmet Province, Luzon. The original material was taken to represent mature or nearly mature specimens, but a recent collection of the same species shows that the type was quite immature, the flowers being in young bud only. In the original specimens the corolla was considered to be composed of four almost entirely SERTULUM BONTOCENSE. 79 united petals, forming a cylindric tube quite closed at the apex by an appendag'e from each corolla-lobe uniting to form a conical body at the mouth, the four free spreading lobes being broad and very short. This description is true of the original specimens, but quite mature flowers on recently collected material shows them not to be cleistogamous, and the corolla to be composed of four entirely free petals, and hence the species being referable to Loranthus, § Heteranthus. Specimens from Malamey, Bontoc Subprovince, Luzon, collected by Father Vanoverbergh, in June, 1911, (no. 1293) match the type of Cleis- toloranthus verticillatus in all respects except that some of the flowers are quite mature and open. The original description of the species must be emended as follows: Inflorescence solitary or fascicled at the nodes, the flowers in umbellately disposed triads, or by reduction, in diads. Calyx 2.5 to 2.8 mm long. Petals 4, free, valvate, 1.6 cm long, about 1.7 mm wide, the reflexed part above the insertion of the stamen narrowly oblong, 5 mm long, about 2 mm wide, each petal with a somewhat triangular, thick appendage on the inner surface near the apex, the four appendages forming a conical body that closes the corolla in bud, the petals in bud with a spreading, 2 to 2.3 mm broad, short, truncate tip. Stamens 4 mm long, erect; filaments 2.5 mm long; anthers 1.5 mm long, continuous with the filaments and of the same diameter. In spite of the reduction of Cleistoloranthus to Loranthus the species is a very distinct one in the latter genus, characterized by the peculiarly appendaged petals. In the generally accepted interpretations of the genera of Loranthaceae, Cleistoloranthus cannot be considered to be generically distinct from Loranthus, but with any treatment of genera approaching that proposed by Van Tieghem, it would probably rank as distinct as most of the genera indicated by that author. Unfortunately the specific name originally used by me is not valid in Loranthus, hence the necessity of coining a new specific name under the latter genus. Loranthus tetranthus sp. nov. § Lepiostegeres. Glaber; foliis oppositis, coriaceis, oblongo-ellipticis ad lanceo- latis, usque ad 6 cm longis, obtusis vel acutis, rariter leviter acuminatis, basi acutis, nervis obsoletis; capitulis axillaribus, solitariis, 4-floris, in alabastro oblongo-lanceolatis, acuminatis, bracteis numerosis, ovatis ad lanceolatis, interioribus gradatim longioribus; floribus 4-meris, sessilibus, corolla circiter 2 cm longa. Glabrous throughout, 0.5 m high, the branches terete, dark- colored when dry, the internodes rather short. Leaves opposite, oblong-elliptic to lanceolate, 4 to 6 cm long, 1 to 2.5 cm wide, coriaceous, dull, of about the same color on both surfaces, the base acute, the apex blunt, acute, or rarely somewhat acuminate ; 110592 3 80 MERRILL. petioles 4 to 8 mm long; midrib prominent, the lateral nerves obsolete. Flowers in axillary, solitary, 4-flowered, sessile heads, surrounded by numerous, grayish-brown, chartaceous, imbricate bracts, the outer bracts broadly ovate, acute or acuminate, 3 to 4 mm long, the inner ones gradually longer, ovate-lanceolate, prominently acuminate, the innermost up to 1 cm long, the ul- timate two or three strongly imbricate forming a beaked 1 cm long calyptra that falls as a whole. Flowers pink and greenish, sessile. Calyx 3.5 to 4 mm long, somewhat angled by contact with those of the other flowers in the same receptacle, slightly and irregularly 3- or 4-toothed, the limb slightly produced. Corolla about 2 cm long, 4-lobed, the lobes united for the lower 3 to 4 mm forming a short tube about 2.5 mm in diameter, the part above the insertion of the stamens linear-oblong, 6 mm long, about 0.6 mm wide. Filaments 1.5 mm long; anthers continuous, linear-lanceolate, acuminate, 3 mm long. Style slender, 2 cm long. Luzon, Bontoc Subprovince, Malamey, Vanoverbergh 129Jf, June 28, 1911, parasitic on trees, altitude about 1,890 m. A very distinct species of the section Lepiostegeres, entirely different from the few other species of the section known from the Philippines. Loranthus pentagonus sp. nov. § Dendrophthoe. Frutex glaber, ramulis crassis, acute pentagonis; foliis alter- nis, coriaceis, oblongis vel oblongo-ellipticis, usque ad 12 cm longis, petiolatis, obtusis vel acutis, basi acutis vel leviter acu- minatis, nervis utrinque vix distinctis, 5 vel 6 ; inflorescentiis in axillis superioribus ; floribus 6-meris, circiter 5 cm longis, ses- silibus, solitariis, in diadibus, vel in triadibus dispositis. A glabrous parasitic shrub, the branchlets stout, 5 to 8 mm in diameter, brown when dry, distinctly and sharply 5-angled. Leaves alternate, thickly coriaceous, oblong to oblong-elliptic, 7 to 12 cm long, 3 to 5.5 cm wide, brownish-olivaceous when dry, the lower surface slightly paler than the upper one, slightly shining, apex acute or obtuse, base more or less narrowed, acute or somewhat decurrent-acuminate, rarely subrounded; nerves 5 or 6 on each side of the midrib, not prominent, the reticulations obsolete or nearly so; petioles 1 to 1.5 cm long. Inflorescence in the uppermost axils, the rachis stout, 4 to 6 cm long, bearing alternate stout branchlets 1 cm long or less, each branchlet bearing at its apex from 1 to 3 sessile flowers, the bracteole subtending the flowers broadly triangular-ovate, concave, obtuse, about 4 mm long. Flowers 6-merous, yellowish, stout. Calyx cylindric, about 8 mm long, 4.5 mm in diameter, truncate, or very slightly and obscurely toothed. Corolla 4.5 cm long, the SERTULUM BONTOCENSE, 81 tube slightly gibbous at the base and 5 mm in diameter, the lobes 3 mm wide below, narrowed above, the reflexed part above the insertion of the stamens about 1.5 cm long, thickened, 2 mm wide, the margins incurved. Filaments 3 to 4 mm long; anthers continuous, erect, linear-lanceolate, about 1 cm long. Luzon, Bontoc Subprovince, Bauco, in forests, altitude about 1,650 m, Vanoverbergh 125U, June 13, and August 17, 1911. A characteristic species, its distinctive features being its stout, 5-angled branchlets, and its comparatively large flowers, the corolla being somewhat gibbous at the base. MENISPERMACEAE. COCCULUS DC. Cocculus laurifolius DC. Syst. 1 (1818) 530, Prodr. 1 (1824) 100; Diels in Engl. Pflanzenreich 46 (1910) 239. Luzon, Bontoc Subprovince, Daoait, Vanoverbergh 1237, June, 1911, along trails, altitude about 1,200 m. Not previously reported from the Philippines, the third species of the genus to be found in the Archipelago. India to Japan, southern China, Formosa, and Java. The specimen is a fairly close match for Wallich 1^965, a specimen of which is in our herbarium, as well as several specimens from plants cultivated in the Botanic Garden at Buitenzorg, Java. MAGNOLIACEAE. ILLICIUM Linn. Illicium montanum sp. nov. Arbor parva, glabra, aromatica ; foliis oblongo-ellipticis, utrin- que acutis vel leviter acuminatis, coriaceis, usque ad 8 cm longis, nervis utrinque circiter 7, obscuris vel subobsoletis ; floribus axillaribus, fasciculatis, circiter 3 cm diametro ; folliculis circiter 12, radiatis, 1.5 ad 2 cm longis, acuminatis. A small, glabrous, aromatic tree about 4 m high. Branches olivaceous when dry. Leaves coriaceous, dark-brown when dry, slightly shining, oblong-elliptic, narrowed about equally at both ends and acute or slightly acuminate, 5 to 8 cm long, 2 to 3.3 cm wide, the midrib very stout and prominent; lateral nerves about 7 on each side of the midrib, very faint, often nearly obsolete, the reticulations obsolete; petioles about 1 cm long. Flowers white, axillary, somewhat fascicled, the buds ovoid, ses- sile, soon becoming short-pedicelled. Sepals and petals about 20, the outer ones in bud ovate, apiculate, their margins ciliate, deciduous or persisting until anthesis and elongated, the outer sepals in flower oblong, about 15 mm long, 6 mm wide, obtuse, the inner ones gradually shorter and narrower, the innermost petals lanceolate, long-acuminate, about 10 mm long. Stamens 82 MERRILL. about 20, in two rows, 3 to 4 mm long. Carpels about 12, acu- minate, 5 mm long, in fruit radiately arranged, laterally compres- sed, lanceolate, 1.5 to 2 cm long, about 5 mm deep. Luzon, Bontoc Subprovince, Bauco, Vanoverbergh 10^8, January, 1911, in forests, altitude about 1,650 m. The second species of the genus to be found in the Philippines, differing from Illicimn philippinense Merr., to which it is closely allied, in its larger flowers, more numerous perianth-segments, and larger and more numerous mature carpels. LEGUMINOSAE. RHYNCHOSIA Lour. Rhynchosia volubilis Lour. FI. Cochinch. (1790) 460; DC. Prodr. 2 (1825) 385; Hemsley in Journ. Linn. Soc. Bot. 23 (1887) 196. Luzon, Bontoc Subprovince, Bauco, Vanoverbergh 65^, July, 1910, altitude about 1,400 m. Not previously reported from the Philippines; China to Formosa and Japan. The specimen seems to be referable to Loureiro’s species, as that is at present interpreted; it differs from it, as described, in having the leaflets shortly acuminate, not obtuse. The Luzon plant is matched in all respects by one from Formosa, Nakahara 195, July, 1905. DERRIS Lour. Derris oblongifolia sp. nov. § Paraderris. Frutex erectus vel scandens, partibus junioribus plus minusve ferrugineo-pubescentibus vetustioribus glaber vel subglaber; foliis pinnatis, circiter 20 cm longis, foliolis 13, oblongis vel anguste oblongo-obovatis, acuminatis, 3 ad 7 cm longis; racemis axillaribus, ferrugineis, quam folia multo brevioribus; floribus fasciculatis, albidis vel roseis, circiter 1.7 cm longis, vexillo piloso, basi subcalloso; leguminibus oblongis, 5 ad 8 cm longis, 1.5 ad 2 cm latis, subglabris. A scandent woody vine, or suberect and about 1 m high, the younger parts more or less ferruginous-pubescent, in age becom- ing glabrous or nearly so. Branches dark-brown, lenticellate, slightly pubescent, the younger parts very densely ferruginous- pubescent. Leaves alternate, about 20 cm long, pubescent with scattered, short hairs, becoming nearly glabrous; leaflets 13, oblong to narrowly oblong-obovate, when young membranaceous, becoming firmly chartaceous, 3 to 7 cm long, 1.5 to 2 cm wide, base rounded or subacute, the apex shortly and rather abruptly acuminate, the lower surface slightly paler than the upper, in age becoming quite glabrous except for scattered hairs along the midrib; lateral nerves about 10 pairs, slender but distinct; petiolules 2 to 3 mm long. Racemes in the upper axils, solitary. SERTULUM BONTOCENSE. 83 appearing with the leaves, forming a terminal leafy inflorescence, rather densely ferruginous-pubescent. Flowers fascicled at the nodes, white or pink, their pedicels pubescent, 6 to 7 mm long. Calyx ferruginous-pubescent, cup-shaped, subtruncate, 5 to 6 mm deep and wide, faintly and broadly toothed. Standard about 17 mm long, the claw 3 to 4 mm long, the lamina suborbicular, retuse, pubescent on the back, reflexed in anthesis, the base with thickened but not projecting callosities. Wings adherent to the keel, their claws slender, about 6 mm long, their blades narrowly oblong, about 4 mm wide, similar to those forming the keel. Vexillary filament free at the base, united with the others above. Ovary somewhat pubescent. Fruit very thin, oblong, 5 to 8 cm long, 1.5 to 2 cm wide, acuminate, when young with very few scattered short hairs, apparently when mature glabrous or nearly so, one wing about 2.5 mm wide, the other less than one-half as wide. Luzon, Bontoc Subprovince, Bauco, on hillsides, altitude about 1,325 m, Vanoverbergh 280, in flower April, 1910, in fruit May, 1911. A species apparently undescribed, and quite different from any of the Philippine forms previously known. Although the callosities of the stand- ard do not project as in Derris elliptica Benth., still they are sufficiently evident as thickenings of the tissue, and apparently refer the present species to the section Paraderris. TERAMNUS Sw. Teramnus angustifolius sp. nov. Herba scandens omnibus partibus plus minusve ferrugineo- pubescens; foliolis lanceolatis usque ad 7 cm longis, vix 1 cm latis; calycis dentibus lanceolatis, acuminatis, aequalibus, 2 mm longis. A scandent, slender, herbaceous vine reaching 1 m in length. Stems 4-angled, 1.5 mm in diameter or less, rather densely pubescent with somewhat spreading ferruginous hairs. Leaves trifoliolate, the petioles, petiolules and lower surfaces of the leaflets more or less appressed-pubescent with ferruginous hairs, the stipels pubescent, acicular, about 1 mm long; petioles 1.5 to 3 cm long; petiolules of the lateral leaflets usually about 2 mm long, that of the terminal one longer. Leaflets lanceolate, 4 to 7 cm long, 5 to 9 mm wide, chartaceous, base rounded or acute, narrowed upwards to the apiculate apex. Racemes pubescent, axillary, solitary, few-flowered, 2 to 9 cm long, slender. Calyx 4.5 mm long, appressed-pubescent, the 5 lobes subequal, lanceo- late, acuminate, 2 mm long. Corolla almost included in the calyx, blue and white; standard 5.5 mm long, the limb orbicular- elliptic, retuse, not auricled at the base, 4 mm long, 3 mm wide. 84 MERRILL. the claw about 1.5 mm long; wings and keel nearly as long as the limb, clawed, narrow. Stamens monadelphous, the vexillary one united with the others at the base, then free for about 0.5 mm, then united with the others again, forming a closed tube, five of the filaments bearing fertile anthers about 0.4 mm long, the other five without anthers. Ovary cylindric, somewhat pu- bescent, bearded at the apex, the style short, curved ; ovules about 10. Young pods somewhat flattened, slightly pubescent, 4 cm long, 3 mm wide. Luzon, Bontoc Subprovince, Bauco, Vanoverbergh 716, July 28, 1910, in cultivated lands, altitude 1,300 m. A very distinct species, at once recognizable by its lanceolate, compara- tively very narrow leaflets. RUTACEAE. EVODIA Forst. Evodia villosa sp. nov. Frutex circiter 1.5 m altus, ramulis, subtus foliis inflorescenti- isque dense f ulvo-villosus ; foliis unifoliolatis, foliolis ellipticis vel oblongo-ellipticis, breviter acuminatis vel acutis, 10 ad 15 cm longis, nervis prominentibus, utrinque 7 ad 10; inflores- centiis axillaribus, brevibus, in alabastro densissimis; floribus 4-meris, calycis villosis. A shrub about 1.5 m high, the branchlets, petioles, lower surfaces of the leaves and inflorescence densely fulvous-villous, the indumentum rather pale. Leaves opposite, 1-foliolate, the petiole stout, about 2 cm long. Leaflets elliptic to oblong- elliptic, coriaceous, 10 to 15 cm long, 3 to 6.5 cm wide, rather pale when dry, shining, the apex shortly acuminate or merely acute, the base acute, rarely somewhat rounded, the upper surface fulvous-villous, densely so on the midrib and nerves; lateral nerves 7 to 10 on each side of the midrib, prominent, arched-anastomosing, the reticulations distinct. Inflorescence in the upper axils, in bud dense, not exceeding the petioles, many- flowered. Flowers 4-merous, yellowish-green. Sepals obtuse, oblong, densely villous outside, 2 mm long. Petals (in nearly mature bud) oblong-elliptic, obtuse, punctate, 2 mm long. Ovary ovoid, densely villous, about 1.3 mm long; style subquadrangular, subsessile, about 0.6 mm in diameter. Stamens 4, short. Fruit unknown. Luzon, Bontoc Subprovince, Bauco, Vanoverbergh 1002, November 15, 1910, in forests, altitude about 1,650 m. A species not closely allied to any other known Philippine form, well characterized by its villous, pale-fulvous indumentum. SERTULUM BONTOCENSE. 85 VITACEAE. TETRASTIGMA Planch. Tetrastigma brunneum sp. nov. Frutex scandens, glaber, ramulis teretibus; foliis palmato 3- ad 5-foliolatis, foliolis coriaceis, in siccitate brunneis, ovatis, oblongo-ovatis, vel subellipticis, basi acutis vel rotundatis, apice abrupte subtenuiter acute acuminatis apiculatisque, margine distanter crenato-serratis ; inflorescentiis axillaribus, laxis, corymboso-paniculatis ; floribus 4-meris, subumbellato dispositis, circiter 1.5 mm longis, numerosis; fructibus obovoideis vel oblongo-obovoideis, 8 mm longis; seminibus 2, plano-convexis, rugosis, baud sulcatis. A woody vine 5 to 6 m in length, glabrous throughout. Stems and branches terete, light-brown, striate, the branchlets usually smooth, black or dark-colored when dry. Tendrils leaf-opposed, forked, up to 18 cm in length. Leaves palmately 3- to 5-folio- late, their petioles 1.5 to 5 cm in length. Leaflets ovate, oblong- ovate, or subelliptic, firmly coriaceous, always brown when dry, the lower surface a little paler than the upper one, slightly shining, 5 to 10 cm long, 2 to 5 cm wide, the middle leaflet larger than the lateral ones, abruptly and rather slenderly acutely acuminate, the acumen apiculate, the base rounded or acute, the margin distantly crenate-serrate, the teeth usually rather prominent, often apiculate; nerves 5 or 6 on each side of the midrib, not prominent, anastomosing, the reticulations lax, obscure; petiolules of the lateral leaflets 4 to 8 mm long, that of the middle one about 1.5 cm in length. Inflorescence axillary, rather lax, corymbose-paniculate, many-flowered, peduncled, about 8 cm long, enlarged in fruit and often forming terminal or subterminal more or less leafy panicles. Male flowers : Calyx shallowly cup-shaped, truncate, about 1 mm in diameter. Petals 4, free, oblong, 1.8 to 2 mm long, obtuse, cucullate at the apex inside. Filaments about 1 mm long, inserted under the broad disk; anthers 0.5 mm long, broader than long. Disk prominent, obscurely 4-angled or lobed. Female flowers: Calyx somewhat funnel-shaped, truncate, 1.2 mm in diameter. Petals as in the males. Staminodes slender, 1 mm long, slightly dilated at the apex. Ovary cylindric-ovoid, truncate, glabrous, 1.2 mm long; stigma sessile, disk-like, not at all or very obscurely lobed, 0.7 mm in diameter. Fruit obovoid to oblong-obovoid, about 8 mm long, apparently some- what fleshy, each containing two plano-convex seeds, or the 86 MERRILL. inner face slightly concave, not sulcate, somewhat rugose, brown, the albumen ruminate. Luzon, Bontoc Subprovince, Malamey, Vanoverbergh 795 (type), August, 1910, in forests, altitude about 1,550 m, with flowers, a second collection in December, 1910, with immature fruits: Benguet Subprovince, Pauai, Bur. Sci. 8410 McGregor, June, 1909, with very immature flowers, altitude about 2250 m; Mount Tonglong, Bur. Sci. 5125 Ramos, December, 1908, with immature fruits, altitude about 2,100 m, Williams 1336, October, 1904: Province of Tayabas, Mount Binuang, Bur. Sci. 9434 Robinson, August, 1909, with ? flowers; Lucban, Mount Banajao, Elmer 7796, May, 1907, with mature fruits, distributed as Tetrastigma lanceolarium Planch.: Province of Bataan, Mount Mariveles, Bur. Sci. 1560 Foxworthy. Probably also should be referred here Elmer 10867 from Mount Apo, Mindanao, distributed as Tetrastigma pergamaceum Blume. Apparently allied to Tetrastigma pergamaceum Blume, but differing from that species, as described, in its terete stems, its 3- to 5-foliolate leaves, never 1-foliolate as is usually the case in Blume’s species, and in its axillary, not leaf-opposed cymes. The species is not typical of Tetras- tigma, as the stigmas are perfectly circular in outline, not at all lobed, but has been placed in the genus as it agrees in other details, especially in its 4-merous flowers, and in its fruit and seed characters. Tetrastigma philippinense sp. nov. Species T. loheri similimis et ut videtur valde affinis differt petalis extus distincte pubescentibus. A tendril-bearing dioecious vine reaching a height of 10 m, and the trunk a diameter of 2 cm, glabrous except the inflores- cence. Branches terete, brownish, smooth, slender. Leaves 3-foliolate, their petioles 1 to 2.5 cm long; leaflets lanceolate, submembranaceous to coriaceous, dull or slightly shining, of about the same color on both surfaces, the middle one longer than the lateral ones, equilateral, obtuse, rounded, or subacute at the base, the lateral ones somewhat falcate, inequilateral at the usually rounded base, all acuminate at the apex, margins often slightly undulate, distantly serrulate-crenate, the teeth usually mucronulate, 6 to 16 cm long, 1.5 to 4.5 cm wide; lateral nerves 7 to 10 on each side of the midrib, not prominent, the reticula- tions lax, obsolete or indistinct; petiolules of the middle leaflets 1 to 2 cm long, of the lateral ones 5 mm or less. Tendrils simple, 10 to 15 cm long or more. Inflorescence axillary, pubescent, about as long as the petioles, compound, umbellately corymbose, the flowers subumbellately arranged on the ultimate branchlets. Female flowers: Calyx obscurely 4-toothed, about 1.5 mm in diameter. Petals 4, oblong, pubescent externally, obtuse, some- what cucullate at the apex, 2 to 2.2 mm long. Ovary ovoid, glabrous, 1.5 mm long, slightly contracted upwards; stigma SERTULUM BONTOCENSE. 87 sessile, distinctly but shallowly 4-lobed. Staminodes slender, spatulate, 1 to 1.2 mm long. Male flowers similar to the female ones in size. Filaments 1.5 mm long; anthers broadly ovoid or ellipsoid, 0.6 mm long. Disk prominent. Fruit oblong-obovoid, red, turning dark-purple or black, the pericarp more or less fleshy when fresh, apparently thin. Seeds solitary, with a shallow longitudinal groove along each side, obscurely sulcate, the albumen ruminate. Type specimen R. S. Williams 1025, Antamok, Benguet Subprovince, Luzon, October 22, 1904, -with $ flowers, altitude about 1,000 m. This species is common and widely distributed in the Philippines, and most of the material below referred to it has been determined here as Tetrastigma lanceolarium Planch., a few specimens as T. angustifolium Planch., and very recently a number have been referred to Tetrastigma loheri Gagnep., the latter determinations being based on a specimen of Whitford 19, referred by Gagnepain to his species.® However, Whitford’s specimen is with fruit only, and its determination as Tetrastigma loheri Gagnep., must have been based on strong vegetative resemblances between it and the type of the species, Loher 371. According to Gagnepain’s de- scription, and according to the characters used by him in his key to the Asiatic species,* all the material cited below must belong to a different species than T. loheri, for that species has glabrous petals, while all our material of T. philippinense has pubescent petals. As to Whitford 19 it is impossible to say whether or not the petals were glabrous or pubescent, but abundant material from the same locality in flower, which matches it in all respects, so far as the material in different stages is comparable, has pubescent petals. I have no hesitation in referring to Tetrastigma philippinense the following material : Luzon, Province of Ilocos Norte, For. Bur. 12Jt89 Merritt & Darling, November, 1908: Bontoc Subprovince, Banco, Vanoverbergh 6^6, June, October, 1910, ? flowers: Benguet Subprovince, lower slopes of Mount Pulog, Merrill 639Jt, May, 1909, ? flowers; Sablang, Bur. Sci. 12730 Fenix. November, 1910, with young fruits: Province of Bataan, Lamao River, Mount Mariveles, Williams 3A9, 792, December, 1903, January, 1904, the former with c? Flowers, the latter with immature fruits. For. Bur. 2310 Meyer, December, 1904, ? flowers. Bur. Sci. 1872 Foxworthy, December, 1906, c? flowers, Whitford 19, April, 1904, with mature fruits; Dinalupijan, Merrill 1590, December, 1902, with c? flowers: Province of Pampanga, Arayat, Merrill H19, March, 1903, with mature fruits: Province of feaguna, Cavinti, Loher 5837, February, 1906, with immature fruits: Province of Rizal, Montalban, Loher 582U, January, 1905, with $ flowers; Caytimon, Bur. Sci. 10897 Ramos, October, 1909: Province of Tayabas, Bur. Sci. 13277 Ramos, March, 1911, in fruit. Bur. Sci. 12325 Foxworthy, with very young fruits. Mindoro, Baco River, McGregor 261, April, 1905, in fruit, Merrill 12^0, January, 1903, in fruit. Palawan, Merrill 8JiM, February, 1903, with A]VGA?f BAY. B7 EmekSon B. Christie, 121 pages, 1 map, 29 plates. , Order No. 410. , ■ Paper; $1.^ United States fcurredcy, postpaid. Sindangan Bay is situated on the northern coast of Zamhoanga Peninsula. The Subanuns of this region were studied by Mr. Christie during two periods of five and six weeks, respectively. The following is an abstract from the contents of Mr. Christie's report on the Subanuns; Habitat and history; relations with the; Moros; material culture; houses; industries; trade; agriculture; family life; social customs; administration of justice; religion; the medicine man; ceremonies; tales; .Word^lists; physical measurements. The 29 plates illustrate the Subanuns at work and at piay; their. industries, bouses, altars, and implements; and the people themselves. THE PISTOKY OF SUBU, . ; By^NAJEEB M. .Saleebt. 275 pages, 4 maps, 2 diagrams. Order No. 40fi. Paper, $0.75 United States currency, postpaid. In the preparation of his manuscript for The History of Sulu Doctor Saieeby spent much time and effort in gaining access to documents In the possession of the Sultan of Sulu. U is fortunate that these records have now been translated and preserved in permanent form. This book is a history of the Moros in the Philippines from the earliest times to the American occMpation. STinOIES Ilf MOBO HISTOUY, BAW, AND BEBIOION. .. .By:NAJEEB M. .Salebbt. • 107 pages, 16 plates, 5, diagrams. Order No. 405. Paper, $0.25; half morocco, $0.75 United States currency, postpaid. This volume' deals with the eartiset written records of the Moros in Mindanao; DootoT Saieeby was fortuaatply able to obtain exact copies of carefully preserved early records written in the Magindanao dialect with' Arabic characters. The author presents translations of these ax well as 16 half-tone illustrations of certain pages from the originals. The names of the rulers of Magindanao are recorded in five folding diagrams. NEGBITOS OF ZAMBAIiES. By WiEEiAM .Aixan Reed. . 85 pages, 62 plates. Order No. 402; Paper, $0,25: half morofico,. $0.75 United States currency, Postpaid. The introductory chapter deals with the general distribution of Negritos and with the distribu- tion of the Philippine branch of the race. The suaoeeding chapters deal with the various indus- tries, amusements, and social relations of these little men. Plates from photographs, the greater part of which was taken for this Publication, show ornaments, houses, men making fire with bamboo, bdws and arrows, dances, and vatioua types of the people themselves. . i . , CONTENTS, COPELAND, EDWIN BINGHAM. The Origin, dnd Relationships of Taenitis ^ .. 47 COPELAND, EDWIN BINGHAM. New or ^nterest^ ing Philippine Ferns, VI .. — . 53 Copeland; EDWIN BINGHAM. New Sarawak “ ; Ferns 59 COPELAND, EDWIN BINGHAM. Ne^y Papuan Ferns , . . - . , 67 GROVES, H. and J. ChWaceae from the Philippine Islands , -- - - ... . 69 MERRILL, E. D. Sertulum Bontocense: New or In- teresting Plants Collected in Bontoc Subprov^nce, - ' ' Luzon, by Father Morice Vanoverbergh. . . _ . 71 WESTER,' P, J. A Contribution to the Nomenclature of the Cultivated Anonas . . 109 " ' u. s. 'Fhe “Philippine Journal of Science” is issued as follows: ourwnoy. Section A. Qhemica! and Geolog'ioal Sciences and the Industries.. $2.00 Section B. The Philippine Journal of Tropical, Medicine 3.00'^ Section C. Botanj^ . . . 3.00 Sebtion 1). General Biology, f^thnology and Anthropology {Sec- tion B began with Volume V)... . 3,00 Entire Journal, Volume II, III, IV, or V ... . ' 5. 00 Entire Journal, beginning wltli Volume A'l '7.00 Single numbers ^ , . , , . , .50 Votome I, 1906 ( not divided into sections > -.......'.-.v. 10.00 Supplement to Volume I (botany) .".i'.".. -3.50 Volume f (without supplemenf) ^ . . , . , . JSaeh section is separately paged and indexed,' ' ' j"' ^ Ppblications sent in exchange for the Philippine Journal of Science* should be addressed: Library, Bureau of Science, Manila, P. I. Subscriptions may be sent to the Business Manager, Philippine Jour- > nal of Science, Bureau of Science, Manila, P. I., or to any of the agents listed below: The Macmillan Company, 64^66 jPiftli Avenue, New York City, tJ. A. \Vm. Wesley &'Son, 3« E^sex Street, Strand, liondon, W. .C,. England. Martmps Nijhoff, Nohelstraat IS, The Hg.gue, Holland. Mayer Sk Miiller, Prina honis Perdinandstrassc 2, Berlin, N.W., Germany. Kelley & WalSh, Limited, 32 Baffles Place, Singapore, Straits Settlbmenig. A. M. ^ J. Eerguson. 1 0 Baillie Street, Colombo, Ceylon. ■ ' = ; ' Thacker, Spink & Co., P. O. Box 64, Calcutta, India. August, 1912 THE PHILIPPINE JOURNAL OF SCIENCE ALVIN J, COX, M. A., Ph.D. GENERAL EDITOR Section C. Botany E. D. MEERILL, M. S. EDITOR WITH THE COOPERATION OP P. ,W. GRAFF, B. S. ; W. H. BROWN, Ph, D. ,1. H. N. WHITFORD, Ph. D. 'sf.VX' i J ■-J/ (»s>f '•iVMi.M/ MANILA BUREAU OP PRINTING 1912 PtTBIICATIONS FOE SAIE BY THE BTTEEAH OF SCIEITCE. MANILA, PHILIPPINE ISLANDS REPORT OF THE rNTERSTATIOJfAIj PLAGTTE CONFERENCE. Held at Mukden, April, 1911, under the auspices of the Chinese Governmout. - ; . Edited by Ebich Maetini, G. F. Peteie, Akthdb Stanley, and Richaed , , : P/ Strong., 4JSJ pagte, IS plates (2 colored, 4 haif.tonss, 12 charts and maps). Order No. 416. Cloth, $350; Paper, $2.50 United States currency, postpaid. The proceedings of this Intemationar Conference and information gained therefrom, together with the results of Certain bacteriological Investigations, constitute the present report. Nothing hitherto has been published which gives such A complete and comprehensive account of the entire subject of pneumonic plague. • Delegates from America (United States of), Austria-Hungary, France, Germapy, Great Britain, Italy, Japan, MeKico, the Netherlands, Russia, and China attended the Conference. The Bureau of Science Of the Government of the Philippine Islands has been appointed sole agent for the distribution of the printed proceedings of the International, Plague Conference. THE SUGAR INDUSTRY IN THE ISLAND OF NEGROS. By Herbert S. Walkes. 145 pages, 10 plates, 1 map. Order No. 412. Paper, $1.25 United States currency, postpaid. Considered from the viewpoint pf praDtlCal utility, Mr. Walkeris Sugar Industry in . the Island of Negros is one of the most important papers published by the Bureau of Soienoe. This volume is a real contribution to the subject; it is not a mere compUation; for the author was in the field and understands the conditions of which he writes, , The fcllowing is a brief synopsis of the. contents: . ' j u Tables of soil analyses, both Chemical and physical; analyses of the cane, juice and bagasse; estimates based on actual information as to the costs of production and of cultivation; and esti- mates of the cost and location of possible Central factories. The Island Is considered by sugar- producing districts; the area of Cultlvatiort and the production per hectare are given, and the possibility for future expansion discussed. The plates illustrate various phases of sugar Industry from the cultivation of the field to the transporlation.of sugar in native sailboats. A MANUAL OF PHILIPPINE SILK CULTURE. 53 pages, 20 plates. Order No, 413. Paper, $0.75 United States currency, postpaid. The silk industry is particularly adapted to be undertaken by persons with small capital, and like the making of hats in the Philippine Islands it should thrive with a liUlo encouragement. In A Manual of PhiliDpine Silk Culture we have presented the results of several years’ actual work with silk-producing larvae together With a description of the now Philippine race. Half-tone plates illustrate in natural size silkworms in different stages of development, pupae, adult moths, samples of cloth made from er! silk, hand reel, and silk house. Other plates iljustrato Iha various appliarices used In raising silkworms and in epinnlng silk; hand and power reels are Illustrated; working drawings are given for a silk house and for a hand reel. THE PHILIPPINE Journal of Science C. Botany VoL. VII AUGUST, 1912 No. 3 NOTES ON PHILIPPINE ORCHIDS WITH DESCRIPTIONS OF NEW SPECIES, V: THE GENUS BULBOPHYLLUM IN THE PHILIPPINE ISLANDS." By Oakes Ames. {From the Ames Botanical Laboratory, North Easton, Mass., U. S. A.) The following paper contains descriptions of nineteen new species of the genus Bulhophyllum together with a list of the species heretofore ascribed to the Philippine Islands. Those which belong to the § Racemosae are closely related to one an- other, although distinguishable by characters which I have set forth in a provisional key. I am convinced that we have in the Philippines a puzzling group of species, closely allied to Bulbophyllum adenopetalum Lindl., and to B. dasypetalum Rolfe, which cannot be rightly understood until a great deal more material than I have seen has been assembled and thoroughly studied. In this group we find plants which bear flowers so similar in detail that they seem to belong to the same species, although they are easily distinguishable and specifically unlike in their foliage. Around Bulbophyllum dasypetalum there is grouped a number of species which have triangular-lanceolate lateral sepals, and around B. adenopetalum there is grouped ‘ Proof read by E. D. Merrill. 111022 1 25 126 AMES. a number of species which have caudate-tipped lateral sepals. Aside from this conspicuous floral distinction, one easily recog- nized, the species of each group are perplexingly similar. Often the points of dissimilarity are difficult to make clear in a description and stand out distinctly only when actual com- parisons are made. The general habit and the characteristic flowers of this group may be seen in Bulbophyllum dasypetalum Rolfe, which is figured in Orchidaceae 1 : 99. The flowers of the species with caudate- tipped sepals are for the most part similar to those of Bulbo- phyllum flavescens Lindl. A figure of this species is given by J. J. Smith in his “Orchideen von Java,” Atlas, fig. CCCIV. A LIST OF THE PHILIPPINE SPECIES OF BULBOPHYLLUM. The following species of Bulbophyllum have been reported as natives of the Philippine Islands. Those species marked with an asterisk have been verified by an examination of material from the Archipelago; those marked with a double asterisk are doubtfully natives of the Philippines. *1. Bulbophyllum adenopetalum Lindl. Bot. Reg. 28 (1842) Misc. 85. *2. B. alagense Ames in Philip. Journ. Sci. 2 (1907) Bot. 233, Orchidaceae 3 (1908) 54. ♦3. B. antenniferum Reichb. f. in Walp. Ann. Bot. Syst. 6 (1860) 248. Cirrhopetalum antenniferum Lindl. Bot. Reg. 29 (1843) sub t. U9. In Lindley’s herbarium at Kew this species is represented by a single flower collected in the Philippines by Cuming. *4. B. auratum Reichb. f. in Walp. Ann. Bot. Syst. 6 (1860) 261. Cirrhopetalum auratum Lindl. Bot. Reg. 26 (1840) Misc. 50. **5. B. balaeniceps Reichb. f. in Hamb. Gartenz. 19 (1863) 280. In “Index Kewensis” this species is given as an equivalent of B. Napelli Lindl., a native of Brazil. The original description of B. balaeniceps appears to have been drawn entirely from a flower. It was not stated positively by Reichenbach that B. balaeniceps came from the Philippines; its origin was obscure. Pfitzer “ gives Balaenoidea as a section of the genus and refers to it a single species, "B. balaeniceps Rchb. f. von den Philippinen.” •6. B. bataanense Ames Orchidaceae 1 (1905) 96. Probably conspecific with B. Cheiri Lindl. *7. B. Bolster! Ames. See below. *8. B. bontocense Ames. See below. 9. B. braccatum Reichb. f. in Linnaea 41 (1877) 117. *10. B. canlaonense Ames. See below. *♦11. B. careyanum Spreng. Syst. Veg. 3 (1826) 732. *12. B. carinatum Ames. See below. ’Engl. & Prantl, Nat. Pflanzenfam. 2* (1889) 179. PHILIPPINE ORCHIDS. 127 •13. B. Cheiri Lindl. Bot. Reg. 30 (1844) Misc. 44. 14. B. chryseum (Kranzl.) Ames in Philip. Journ. Sci. 6 (1911) Bot. 54. Cirrhopetalum chryseum Kranzl. in Fedde Repert. 8 (1910) 97. •15. B. Clemensiae Ames. See below. ♦16. B. Copelandii Ames Orchidaceae 1 (1905) 97. *17. B. Cumingii Reichb. f. in Walp. Ann. Bot. Syst. 6 (1860) 261. Cirrhopetalum Cumingii Lindl. Bot. Reg. 29 (1843) sub. t. U9. For a colored plate see Bot. Mag. t. ^996. *18. B. cuneatum Rolfe in Ames Orchidaceae 1 (1905) 98. *19. B. Curranii Ames. See below. •*20. B. Blumei J. J. Smith Orch. Jav. (1905) 459. B. cuspidilingue Reichb. f. in Walp. Ann. Bot. Syst. 6 (1860) 264. Philippines given as a habitat in Index Kewensis. Java, Singapore, New Guinea. *21. B. dasypetalum Rolfe in Ames Orchidaceae 1 (1905) 98, fig. *22. B. Dearei Reichb. f. in Flora 71 (1888) 156. Sarcopodium Dearei Reichb. f. in Gard. Chron. (1883) 108, fig. 17. *23. B. Elmeri Ames. See below. *24. B. exile Ames Orchidaceae 2 (1908) 208, fig. *25. B. gimagaanense Ames. See below. *26. B. gnomon if erum Ames Orchidaceae 2 (1908) 209, fig. *27. B. halconense Ames in Philip. Journ. Sci. 2 (1907) Bot. 334. *28. B. lancipetaium Ames supra 23. **29. B. lasianthum Lindl. in Gard. Chron. (1855) 53. “The first knowledge we had of this remarkable plant was from a coloured drawing, from Sumatra, shown us by our learned friend Prof, de Vriese. It has now been examined in a living state, a specimen said to have come from Manila having flowered with Messrs. Loddiges.” etc., Lindley 1. c. *30. B. lasioglossum Rolfe in Ames Orchidaceae 1 (1905) 100, 2 (1908) 3, pi. 18. *31. B. lepantense Ames. See below. 32. B. longiflorum Thouars Orch. Afr. (1822) t. 97. Cirrhopetalum Thouarsii Lindl. Gen. and Sp. Orch. (1830) 58. This species has been referred to the Philippines by Lindley and by other authors (cf. Bot. Mag. sub. t. 72H; Gard. Chron. 2 (1891) 69). I have not seen B. longiflorum from the Philippines and probably the species referred by various authors to the Archipelago under this name is B. Weberi Ames, described below. *33. B. luzonense Ames. See below. *34. B. Makoyanum Reichb. f. in Gard. Chron. 1 (1879) 234, in note; Ames in Philip. Journ. Sci. 6 (1911) Bot. 55. *35. B. marivelense Ames. See below. *36. B. maxi Hare Reichb. f. in Walp. Ann. Bot. Syst. 6 (1860) 248. Cirrhopetalum maxillare Lindl. Bot. Reg. 29 (1843) sub. t. 4-9. In Lindley’s herbarium at Kew this species is represented by a flower and leaf collected in the Philippines by Cuming. *37. B. MacGregor! i Ames. See below. *38. B. Merrittii Ames in Philip. Journ. Sci. 2 (1907) Bot. 334. *39. B. mindorense Ames 1. c. 334, Orchidaceae 3 (1908) 56, pi. 45. 128 AMES. *40. B. nutans Reichb. f. in Walp. Ann. Bot. Syst. 6 (1860) 260; Ames Supra 24. Cirrhopetalum nutans Lindl. Bot. Reg. 25 (1839) Misc. 71; Bot. Mag. t. U18. **41. B. orthoglossum Wendl. & Kranzl. in Gard. Chron. (1896) 326. This species was collected by W. Micholitz in “Sarangui insula,” accord- ing to its authors. If by “Sarangui insula” Sarangani Island is intended, then B. orthoglossum is a native of the Philippines.’ •42. B. paucifiorum Ames. See below. 43. B. penicillium Par. & Reichb. f. in Trans. Linn. Soc. 30 (1874) 151; King & Pantl. Orch. Sikkim Himalaya t. 108. Rolfe in both Vidal’s Phanerogamae Cumingianae Philippinarum (1885) 80, 149, and in the Journal of Botany 23: 215 cites Cuming 2076, from the Philippines, as Bulbophyllum penicillium Par. «& Reichb. f. *44. B. pleurothallioides Ames in Philip. Journ. Sci. 2 (1907) Bot. 335, Orchidaceae 3 (1908) 55, pi. UU. *45. B, profusum Ames. See below. *46. B. Ramosii Ames supra 25. 47. B. saurocephalum Reichb. f. in Gard. Chron. 2 (1886) 262. *48. B. stellatum Ames. See below. *49. B. trisetum Ames. See below. *50. B, vagans Ames & Rolfe in Orchidaceae 2 (1908) 210, iig. *Var. angustum Ames in Philip. Journ. Sci. 2 (1907) Bot. 335. *51. B. Vanoverberghii Ames. See below. *52. B. Weber i Ames. See below. *53. B. Whitfordii Rolfe in Ames Orchidaceae 1 (1905) 100. *54. B. zambalense Ames. See below. The following key represents a provisional effort to differen- tiate the group of the racemose Bulbophyllnms which are closely allied to B. adenopetalum and B. dasypetalum. In order to make it more serviceable several other racemose species, clearly sep- arable from all other Philippine species of the genus, are included. The key is based on characters which may exhibit a wide range of variation when more material is available for comparison. Unfortunately very few duplicates of the new species described in this paper have been received so that it has been impossible to arrive at a clear conception of those characters which will, in a key, make positive identification possible. || Bulbophyllum lasianthum Lindl. has been ascribed to the Phil- ^ ippines from whence it was supposed to have been imported | into England by Loddiges, but I have seen no material referable | ’ Sarangani Island is known to several professional orchid collectors, , | Micholitz and others, as the home of certain species of commercial value, ; | and has been visited at various times by different collectors. It is a small ‘ j i' island off the south coast of Mindanao. j BULBOPHYLLUM Thouars. E. D. M. PHILIPPINE ORCHIDS. 129 to B. lasianthum among the Philippine specimens I have ex- amined. A. Raceme sharply decurved B. cuneatum B. Raceme erect or simply flexuose. a. Labellum or sepals hairy. b. Labellum hairy B. lasioglossum (6. Sepals hairy B. lasianthum) a. Labellum ciliolate; sepals smooth. b. Lateral sepals caudate-tipped. c. Leaves 7.5 to 15 cm long B. adenopetalum c. Leaves 5 to 6 cm long B. Ramosii b. Lateral sepals triangular-lanceolate, not caudate-tipped. c. Leaves about 15 cm long B. lancipetalum c. Leaves 3.5 to 5.5 cm long B, gimagaanense c. Leaves 8 to 10 cm long B. Clemensiae a. Labellum smooth on the margin. b. Plants caespitose. c. Pseudobulbs 1 cm long; flowers several B. Merrittii c. Pseudobulbs wanting or minute. d. Flowers several B. lepantense d. Flowers two B. pauciflorum b. Rhizome creeping; leaves distant. c. Flowers 7 mm long B. vagans c. Flowers less than 7 mm long. d. Raceme erect, many-flrfwered, longer than the leaves, e. Petioles 3 to 10 cm long. /. Leaves elliptic-oblong B. Vanoverberghii f. Leaves narrowly lanceolate B. dasypetalum e. Petioles about 1 cm long B. zambalense d. Raceme weak, few-flowered, shorter than the leaves; petals spathulate B. marivelense d. Raceme 2-flowered B. Curranii (Insufficiently known) B. braccatum § RACEMOSAE. Bulbophyilum Clemensiae sp. nov. Aff. B. adenopetalo et B. clasypetalo. Rhizoma crasswm. Pseudobulbi 0. Folia oblanceolata, coriacea, petiolata. Scapus gracilis flexuosus, bracteis tubularibus prope basim. Racemus elongatus, bracteis inflorescentiae glabris, lanceolatis, ovarium aequantibus. Sepala lateralia triangulari-lanceolata, acuminata. Sepalum superius lineari-lanceolatum, acutum, concavum. Pet- ala linearia, acuta, 1-nervia, margine cellulari-ciliato. Labellum ad basim rotundatum, lingulatum ab medio, vel lingulatum, pars inferior explanata suborbicularis ; ad basim bicallosum. Colum- na brevis, ad apicem bifurcata. Rhizome comparatively stout, copiously rooted. Pseudobulbs 0. Leaves long petiolate, oblanceolate, coriaceous, tapering abruptly 130 AMES. at the tip, lamina 8 to 10 cm long, 1.6 to 2.3 cm wide. Petiole 2 to 4 cm long. Scape weak, flexuouse, sheathed at the base with several tubular bracts, shorter than the leaves or equal- ing them. Raceme loose, 10- to 20-flowered. Bracts subtending the flowers about equaling the pedicellate ovary, scarious, lan- ceolate, concave, acute. Flowers in dried specimens appear to have been yellowish or whitish when fresh. Lateral sepals triangular-lanceolate, acute, broadly caudate-tipped, 5 mm long, 2 mm wide. Upper sepal linear-lanceolate, concave, acute, 4.5 mm long. Labellum 3.75 mm long, Ungulate, margin of the apical half minutely dilate, somewhat deflexed, margin of the basal half erect ; basal half rotundate when spread out ; bicallose. Column bifurcate at the tip. Mindanao, Lake Lanao, Camp Keithley, March, 1906, Mary Strong Clemens iiO. Bulbophyllum Curranil sp. nov. Rhizoma gracile, ad nodos radicans. Pseudobulbi nulli. Folia oblongi-lanceolata vel oblonga, coriacea, acuta, in sicco rugosa, erecta, distantes, vix petiolata. Scapus brevis, biflorus, bracteis tubularibus obtectus, prope apicem bractea tubular! adornatus. Bracteae inflorescentiae flores sub.tendentes, pedicellis longiores, lanceolatae, acutae, in sicco scariosae. Flores flavidi. Sepala later alia triangulari-lanceolata, acuminata, acuta, 3-nervia. Sep- alum superius simile, lanceolatum, acutum, 3-nervium. Petala lineari-oblonga, 1-nervia, acuta. Labellum subhastatum vel sub- quadratum, ad apicem rotundatum, obtusum, 3-nervium. CoU umna breviter bicuspidata. Rhizome elongated, very slender, in dried specimens less than 1 mm thick. Pseudobulbs, scarcely if at all developed, mere thickenings of the rhizome at the base of each leaf. Leaves 2 to 3 cm distant, elliptic-oblong, or oblong-lanceolate, acute, cori- aceous, 1 to 2 cm long, up to 5 mm wide. Scape very short, scarcely exceeding 6 mm in length, concealed by sheathing bracts. Flowers in pairs, yellow, small, subtended by lanceolate bracts 2 to 3 mm long. Lateral sepals triangular-lanceolate, 5 mm long, 2 mm wide. Upper sepal 4.5 mm long. Petals much shorter than the sepals, about 2 mm long, less than 1 mm wide, 1-nerved. Labellum 2 mm long, about 1 mm wide, smooth. Luzon, Benguet Subprovince, summit of Mount Tonglon, August 8, 1906, H. M. Curran, For. Bur. Jt953. “Epiphytic vine.” Bulbophyllum gimagaanense sp. nov. Aff. B. adenopetalo sed foliis minor et sepalis lateralibus non caudatis. Rhizoma elongatum, repens, gracile. Pseudobulbi 0. PHILIPPINE ORCHIDS. 131 Folia lineari-oblonga, obtusa, coriacea, in petalos graciles con- tracta. Scapus flexuosus, folio longitudine fere aequalis, vel brevior, bracteis 2, laxe vaginantibus, acutis. Racemus laxi- florus, pauciflorus (flores ± 3). Bracteae lanceolatae. Sepala lateralia triangulari-lanceolata, acuta. Sepalum superim ob- longi-lanceolatum, acutum, concavum. Petala spathulata, 1-ner- via, margine papillose vel cellulari-ciliato. Labellum lingulatum, ad basim bituberculatum, 3-nervium, margine ciliato. Columna bifurcata. Rhizome slender, elongated, creeping, copiously rooted. Pseud- obulbs 0. Leaves coriaceous, linear-oblong, 3.5 to 5 cm long, 5 to 8 mm wide, contracted into slender petioles. Scape very slender, sheathed at the base by several, tubular, acute bracts. Raceme loose, few-flowered, flowers 5 to 10 mm apart. Bracts about 2 mm long, lanceolate, acute, exceeding the ovaries. From dried specimens the flowers appear to have been yellowish or whitish when fresh. Lateral sepals triangular or triangular- lanceolate, acute, hardly caudate-tipped, 5 mm long, about 2.5 mm wide. Upper sepal oblong-lanceolate, acute, about 4.5 mm long, concave. Petals spathulate, 3 mm long, ciliolate on the margin, 1-nerved. Labellum 3.5 mm long, lingulate, obtuse, margin ciliate, base bituberculate. Column minute. Negros, Province of Occidental Negros, May 26, 1906, H. N. Whitford 1610. On tall trees, 65 m above sea-level. B. gimagaanense differs from B. adenopetalum Lindl. and B. Ramosii Ames in the form of the lateral sepals. In habit it resembles B. Ramosii rather closely but is readily distinguishable from that species by means of its elongated rhizome and widely separated leaves. It is also similar to B. vermiculare Hook. f. but in the details of the flowers is quite distinct. Bulbophyllum lepantense sp. nov. Pseudobulbi minuti, caespitosi, globosi, rugosi, 3.5 mm alti. Folia crassa, in sicco rugosa, oblongi-elliptica, subacuta, breviter petiolata. Scapus foliis longior, bracteis carinatis tubularibus infra medium. Racemus sublaxiflorus. Bracteae ovato-lanceo- latae, acutae. Sepala lateralia triangulari-lanceolata, 3-nervia, acuta. Sepalum superius ligulatum, acutum, concavum, 3-ner- vium. Petala rhombica vel rhombico-spathulata, obtusa, 1-ner- via. Labellum lingulatum, obtusum, ad basim bituberculatum. Pseudobulbs minute, approximate, rugose, globose, about 3.5 mm in diameter. Leaves fleshy, 3.5 to 5.5 cm long, 1 to 1.6 cm wide, deeply wrinkled in the dried state. Petals compara- tively short, stout, 3.5 mm long. Scape at least twice longer than the leaves, ±10 cm tall, sheathed below the raceme with 132 AMES. about three, carinate, tubular bracts 1 to 1.5 cm long. Raceme loosely flowered. Bracts subtending the flowers ovate-lanceo- late, about 3 mm long. Perianth about 5 mm long, the divisions scarcely spreading, yellowish-white. Lateral sepals triangular- lanceolate, acute, not caudate-tipped, 3-nerved, about 5 mm long. Upper sepal 5 mm long, ligulate, acute, 3-nerved, concave. Pet- als rhombic or rhombic-spathulate, obtuse, 1-nerved, about 3 mm long, the margin minutely cellular-ciliate. Labellum Un- gulate, about 3 mm long, fleshy, bituberculate at the base. Column minute, bifurcate at the tip. Luzon, Lepanto Subprovince, October 29, 1905, Elmer D. Merrill U586. Epiphyte in mossy forests, 2,200 m above sea-level. Bulbophyllum marivelense sp. nov. Aff. B dasypetalo, a quo tamen foliis differt. Rhizoma repens. Pseudobulbi nulli. Folia oblongi-lanceolata, obtusa, coriacea, petiolata. Scapus folio brevior, flexuosus, pauciflorus. Flores flavidi. Sepala lateralia triangulari-lanceolata, 3-nervia, acuta. Sepalum superius oblongum, acutum. Petala spathulata, 1-ner- via. Labellum lingulatum, ad basim bicarinatum. Columna breviter bicuspidata. Rhizome creeping, 2 mm thick. Pseudobulbs wanting. Leaves 1 cm apart, petiolate, coriaceous, oblong-lanceolate, obtuse, lam- ina 10 to 12 cm long, 12 to 13 mm wide, in general outline similar to that of B. adenopetalum. Petioles 3 to 4 cm long, slender, shorter in relation to the length of the lamina than those of B. dasypetalum to which species this one bears a close resemblance in the details of the flowers. Scape erect, bracteate, shorter than the leaves, 8.5 to 10 cm tall. Bracts subtending the flowers ovate-lanceolate, acute, about 1 mm long, equaling the ovaries, hyaline in dried specimens. Raceme loosely flow- ered, flowers about 10, yellow. Lateral sepals triangular-lan- ceolate, acute, 3-nerved, 4.5 mm long. Upper sepal oblong or oblong-lanceolate, acute, concave, 4 mm long. Petals spathulate, 1-nerved, obtuse, 2.5 mm long. Labellum Ungulate, margin smooth, about 2.5 mm long, bicarinate at the base. Column minute, bicuspidate. Luzon, Province of Bataan, Mount Mariveles, January 1, 1904, Elmer D. Merrill 3738. On exposed ridges, epiphytic, at 1,100 meters above the sea. Bulbophyllum pauciflorum sp. nov. Planta parvula, caespitosa, pseudobulbi minuti vel nulli. Folia conferta, coriacea, elliptica vel oblanceolata, breviter petiolata. Scapus foliis breviores, flores duos gerentes. Flores flavidi. PHILIPPINE ORCHIDS. 133 Sepala lateralia, triangulari-lanceolata, 3-nervia. Sepalum su- perius concavum, oblongi-lanceolatum, 3-nervium. Petala spath- ulata, 1-nervia. Labellum ovato-lanceolatum, 3-nervium, acutum. Columna breviter bicuspidata. Plants small, copiously rooted, caespitose, leaves crowded. Pseudobulbs practically wanting. Leaves fleshy, coriaceous, rugose when dry, elliptic or oblanceolate, rounded at the retuse tip, 1 to 2 cm long, 6 to 8 mm wide. Petiole very short, causing the leaves to appear sessile. Scape shorter than the leaves or barely equaling them, up to 2 cm long, bearing, in the specimens studied, only two yellowish flowers. Bracts of the inflorescence about half as long as the shortly pedicellate ovary. Lateral sepals triangular-lanceolate, acute, 4.5 mm long. Upper sepal about 4 mm long, concave, oblong-lanceolate, acute. Petals much smaller than the sepals, about 2 mm long, spathulate, 1-nerved. Labellum comparatively large, ovate-lanceolate, acute, 3-nerved, 3 mm long, 2 mm wide. Column minute, bicuspidate. Luzon, Bontoc Subprovince, August 17, 1911, Father M. Vanoverbergh 1381. Epiphytic, altitude 1,650 meters above sea level. Bulbophyllum Vanoverberghii sp. nov. Pseudobulbi 0. Folia petiolata, coriacea, oblongi-elliptica. Scapus gracilis, erectus, folio longior, bracteis tubularibus prope basim. Racemus gracilis, multiflorus. Bracteae ovariis longio- res, anguste lanceolatae. Sepala lateralia triangulari-lanceolata, acuminata, acuta, subcaudata. Sepalum superius lanceolatum, acutum ad basim concavum. Petala lineari-oblonga, obtusa, 1- nervia. Labellum crassum, 3-nervium, lingulatum, obtusum, ad basim bituberculatum, bicarinatum prope medium. Columna bifurcata. Rhizome creeping, about 5 mm in diameter, copiously rooted. Pseudobulbs 0. Leaves petiolate, very thick, coriaceous, yel- lowish when dry, rugose. Petiole 2.5 to 5 cm long, sulcate, about 3 mm thick, rigid. Lamina elliptic-oblong, obtuse, bilobed at the tip, abruptly or gradually contracted at the base, margin revolute, average length about 15 cm, average width 4 cm. Scape erect, sheathed up to the lowermost flower by about four, loose, acute, vaginate bracts. Raceme slender, many-flowered, 12 to 22 cm long. Bracts exceeding the ovaries, linear-lanceo- late, acuminate, acute, diaphanous in dried specimens, about 7 mm long. Flowers yellowish or whitish, up to 8 mm long, segments of the perianth hardly spreading. Lateral sepals tri- angular-lanceolate, acuminate, subcaudate-tipped, acute, smooth. 134 AMES. 6 to 8 mm long. Upper sepal lanceolate, acute, about 5 mm long. Petals linear-oblong, obtuse, 1-nerved, 3 mm long, about 1 mm wide. Labellum 4.5 mm long, Ungulate, obtuse, fleshy, 3-nerved, bituberculate at the base, bicarinate or furnished with two fleshy marginal ridges on the apical half below the middle, strongly deflexed at about the middle, obscurely if at all 3-lobed, margin of the basal half erect, margin of the apical half deflexed, slightly revolute. Column bifurcate. Luzon, Bontoc Subprovince, July-August, 1911, Father M. Vanoverhergh 1367. To this species I refer the following collections: Luzon, Province of Nueva Vizcaya, December 22, 1908, M. L. Merritt, For. Bur. 15861, altitude above sea-level 1,600 meters: Benguet Sub- province, August 8, 1906, H. M. Curran, For. Bur. 5062; Subaan trail, January, 1909, W. Schultze, Bur. Sci. 5629: Bontoc Subprovince, July 23, 1910, Father M. Vanoverhergh 712: Province of Tayabas, November 3, 1907, Curran and Merritt, For. Bur. 7835: Province of Zambales, Decem- ber 10, 1907, Maximo Ramos, Bur. Sci. i.995: Province of Rizal, Bosoboso (San Isidro), January, 1909, Maximo Ramos, Bur. Sci. 2025: Province of Laguna, Mount Banajao, November, 1907, Curran and Merritt, For. Bur. 8025. Bulbophyllum Vanoverberghii belongs to a small group of Philippine species of which B. dasypetalum Rolfe and B. adenopetalum Lindley are representatives. It appears to be closely allied to B. braccatum Reichb. f. a species which is known to me only through the brief and incomplete description published by Reichenbach f.^ Up to the present time I have been unable to assign with certainty any Philippine specimens to B. braccatum. B. Vanoverberghii is distinguishable from B. dasypetalum Rolfe chiefly by the very dissimilar foliage. The flowers of these species are very similar in detail. Bulbophyllum zambalense sp. nov. Rhizoma elongatum. Pseudobulbi nulli vel minuti. Folia oblanceolata, coriacea, petiolata. Scapus folio longitudine fere aequalis vel longior, ad basim bracteis imbricatis. Racemus laxus. Flores flavidi. Sepala lateralia triangulari-lanceolata, subcaudata, acuta. Sepalum superius lanceolatum, acutum, con- cavum. Petala spathulata, obtusa, 1-nervia. Labellum lingu- latum, 3-nervium. Columna breviter bicuspidata. Rhizome creeping, elongated, bearing at intervals of 3 to 5 cm, coriaceous, oblanceolate leaves. Pseudobulbs reduced to small thickenings on the rhizome. Leaves petiolate, lamina about 5 cm long, 6 mm wide. Petiole about 1 cm long. Scape equaling or somewhat longer than the leaves, clothed at the base by several, scarious, imbricating bracts. Raceme loosely few-flowered, bracts 2 mm long, linear-lanceolate, acute. Flowers *Linnaea41: 117. PHILIPPINE ORCHIDS. 135 yellow. Lateral sepals triangular-lanceolate, subcaudate, 7 mm long. Upper sepal lanceolate, about 5 mm long, concave. Petals spathulate, obtuse, 1-nerved, 2.5 mm long. Labellum lingulate, 3-nerved, 2.5 mm long. Column minute, shortly bicuspidate. Luzon, Province of Zambales, December 13, 1907, Maximo Ramos, Bur. Sci. i996. The type of this species is composed of a single specimen in the Herbarium of the Bureau of Science and of a fragment preserved in my own herbarium. The elongated rhizome is very characteristic. The label- lum of the flowers examined lacked the conspicuous tubercles or callosities at the base, which are usual in flowers of this group. § MONANTHAPARVA. Bulbophyilum bontocense sp. nov. Rhizoma elongatum, repens, gracile, ad nodos radicans. Rad- ices crassae. Pseudobulbi pyriformi, rugosi, paiwi. Folia brevissime petiolata, coriacea, elliptica vel ovata, ad apicem aristata. Scapus foliis longior ortus a basi pseudobulbi vel a rhizomate; ad basim vaginis tubularibus bracteiformibus ; sub florem solitarium bractea tubularis. Sepala lateralia triangulari- lanceolata, longe caudata, 3-nervia. Sepalum superius simile. Petala variabilis, 1-nervia, oblonga, ad apicem irregulariter retusa, mucronata vel marginibus inaequaliter dentatis. Label- lum 3-lobatum, unguiculatum, 3-nervium, lobis lateralibus erectis, lanceolatum ab medio, obtusum vel subacutum, pars inferior explanata orbicularis, pars anterior crassa. Rhizome creeping, elongated, slender. Pseudobulbs 2 to 3 cm apart, pyriform, rugose, monophyllous, 4 to 6 mm long. Leaves coriaceous, elliptic or ovate, setigerous at the tip, 7 to 12 mm long, 5 to 7 mm wide. Scape slender, longer than the leaves, 1 to 3 cm long, arising from the base of the pseudobulbs and from the rhizome about midway between the pseudobulbs, bracteate at base, unibracteate at the base of the solitary white flower, persistent. Lateral sepals triangular-lanceolate, acumi- nate, caudate-tipped, 3-nerved, up to 8 mm long. Upper sepal similar to the laterals. Petals variable, 1-nerved, 3 mm long, 1 mm wide, oblong, somewhat cuneate at the base, apex mucro- nate or irregularly dentate, the apex exhibiting both conditions in the same flower. Labellum 3-Iobed, 2 mm long. Lateral lobes semiorbicular, erect, when spread out giving to the basal portion of the labellum an orbicular aspect, middle lobe lanceo- late, fleshy, acute, 1.5 mm long. Column 1.5 mm long. Luzon, Bontoc Subprovince, August 17, 1911, Father M. Vanoverhergh 1382. Growing on trees, altitude 1,650 meters above sea-level. 136 AMES. Bulbophyllum canlaonense sp. nov. Cum B. halconensi comparandum videtur, a quo tamen sepalis differt. Rhizoma longe repens, gracile, elongatum, ad nodos radicans. Pseudobulbi pyriformi, 3 mm alti, in sicco rugosi, distantes. Folia 1 ad 3 cm longa, 4 ad 9 mm lata, elliptico- oblonga, acuta, coriacea, petiolata. Scapus usque ad 4.5 cm longus, foliis longior, 1-florus, a basi pseudobulbi vel a rhizomate in medio inter pseudobulbos ortus. Flores atropurpurei, usque ad 12 mm longi. Sepala lateralis ovato-lanceolata, acuta. Sep- alum superius simile. Petala 6 mm longa, 3 mm lata, elliptica, ad apicem rotundata, acuta. Labellum 4.5 mm longum, cordato- lanceolatum, acutum. Columna brevis, in medio columnae dens vel papilla. Negros, Canlaon Volcano, April 10, 1910, Elmer D. Merrill 7015. Epiphyte, rather abundant, altitude above sea-level 1,200 to 1,700 m, flower dull-purple. A near ally of Bulbophyllum halconense from which it differs mainly in the entire absence of caudate tips to the sepals (cf. Orchidaceae 3: 54, pi. U, fig. 3 C. c.) Bulbophyllum Elmer! sp. nov. Rhizoma repens, gracile, elongatum. Pseudobulbi pyriformes, rugosi, distantes, monophylli. Folia vix petiolata, lanceolata, acuta, breviter apiculata. Scapi filiformes, ad basim bractea tubular! vaginati, a basi pseudobulbi vel a rhizomate in medio inter pseudobulbos ortus, interdum scapi orti a pseudobulbo uno. Sepala lateralia triangulari-linearia, acuminata, valde acuta, 3-nervia. Sepalum si(,perius simile. Petala elongata sepalis vix breviora, linearia, acuminata, longe aristata, 1-nervia. Labellum 3-lobum, lobi laterales parvi, rotundati, lobus medius subcordatus, crassus, apex erectus vel recurvatus, incrassatus, ad basim lab- ellum Carina brevis, gnomon! solari similis. Columna brevis, antice utroque ad apicem processus setiformis. Rhizome about 1 mm thick. Pseudobulbs about 1 mm apart, 5 mm tall. Leaves 1 to 2.5 cm long, up to 5 mm wide. Scape equaling the leaves in length or shorter. Lateral sepals about 4 mm long, about 1 mm wide ; upper sepal similar to the laterals. Petals very slender, prolonged into a thread-like or aristate tip, 4 mm long. Labellum 1.5 cm long, 3-lobed, cuneate at the base, middle lobe 0.5 mm long. iLuzon, Beng-uet Subprovince, Baguio, March 8, 1904, A. D. E. Elmer 6618. PHILIPPINE ORCHIDS. 137 Bulbophyllum MacGregorii sp. nov. Aff. B. mindorensi, specei quam habitu et structora haec species conspicue simulat. In cortice arborum depressum, decumbens. Rhizoma elongatum omnino pseudobulbis obtectum. Pseudo- bulbi cylindracei, depress!, rugosi, 1-foliati. Folia elliptica vel oblanceolata, coriacea, breviter petiolata, erecta. Scapus 1-florus, a basi pseudobulbi ortus. Ovarium hystricosum vel papillosum. Flos flavidus in sicco ® a bractea tubular! subtensus. Sepala lateralia oblonga, acuta, -3-nervia, margine prope apicem cellulari- dentato. Sepalum superius oblongi-lanceolatum, 3-nervium. Petala abbreviata, 1-nervia, spathulata, ad apicem incrassata. Labellum crassum, oblongum, obtusum, integerrimum, ad apice valde incrassatum, ad basim concaviusculum ; apex saepe deflexus. Columna gracilis, bicuspidata. Petalis brevior. Pseudobulbs 6 mm long. Leaves shortly petiolate, 8 to 13 mm long, 2 to 3.5 mm wide. Scape slender, filiform, about 1 cm tall. Bract subtending the fiower up to 2 mm long. Lateral sepals 4 to 7 mm long, up to 2 mm wide, conspicuously 3-nerved when dry. Upper sepal similar to the laterals. Petals 2 to 2.5 mm long, about 0.75 mm wide, thickened at the tip. Labellum 3 mm long, 1 mm wide. Column 2 mm long. Luzon, Benguet Subprovince, Pauai, June, 1909, R. C. McGregor, Bur. Sci. 8322, altitude above sea-level about 2,100 meters; same locality, Elmer D. Merrill 6624, May, 1909, epiphyte in mossy forest, “flowers dull-red," altitude 2,250 m above sea-level. B. MacGregorii is nearly allied to B. mindorense and to the Bornean species, B. corticicola, recently described by Doctor Schlechter. From the former it differs in the sepals, petals and lip, and from the latter it is readily distinguishable by its 3-nerved sepals, B. corticicola having 5-nerved sepals in the specimens I have examined. Bulbophyllum profusum sp. nov. Rhizoma ramosum, elongatum, diffusum, bracteis imbricatis, arete vaginantibus, dense vestitum; scapis e vaginis rhizomatis a pseudobulbis longe remotis erumpentibus. Pseudobulbi par- vuli, cylindracei, in sicco rugosi. Folium coriaceum, elliptico- oblongum, breviter petiolatum. Scapus abbreviatus, uniflorus. Flores flavidi, sub florum solitarium bractea infundibuliformis. Sepala lateralia elongata, caudata, lineari-triangularia, 3-nervia, glabra. Sepalum dorsale simile. Petala elliptica vel ovata, ‘ In dried specimens the flowers are yellowish or orange-yellow. McGregor made no remarks regarding the flowers of his specimens. Mer- rill describes the flowers of his specimen as dull-red. 138 obtusa, 1-nervia. Labellum lanceolatum, obtusum, 3-nervium. Columna minuta. Rhizome much branched, clothed with imbricating, scarious bracts from among which the numerous 1-flowered scapes arise, crowding the space between the widely separated pseudobulbs. Pseudobulbs about 2 cm apart, 5 mm long, 2 mm thick, closely appressed to the rhizome, rugose and angled when dry, cylin- draceous, truncate, bearing at the summit a solitary, fleshy, elliptic-oblong leaf. Leaves shortly petiolate, 2.2 to 5 cm long, about 1 cm wide, yellowish-green in dried specimens, retuse at the rounded tip. Flowers about 6 mm long, yellowish, borne singly on very short, slender scapes, the scapes almost entirely concealed by the bracts of the rhizome. Subtending each flower there is an infundibuliform, acuminate-tipped, hyaline bract which partly conceals the ovary. Lateral sepals linear-trian- gular, caudate-tipped, 6 mm long. Upper sepal similar to the laterals. Petals elliptic or ovate, obtuse, much shorter than the sepals, 1.75 to 2 mm long, 0.75 mm wide. Labellum lanceolate, more fleshy than the other perianth divisions, about 1.5 mm long, 3-nerved, smooth. Luzon, Province of Rizal, Mount Canumay, October 15, 1911, Maximo Ramos, Bur. Sci. 13785, epiphytic; Province of Cagayan, April 20, 1909, Ramos, Bur. Sci. 7971. B. profusum resembles the Celebesian B. myrianthum Schltr. so closely that the slight differences between them, observable when specimens are laid side by side, may be considered too trivial for recognition as valid specific characters. In B. myrianthum the leaves are ovate or suborbicular and, in the specimens I have examined, do not exceed 2.5 cm in length. In B. profusum, however, the leaves average more than 2.5 cm in length and are rather oblong-elliptic than ovate or suborbicular. In B. profusum the labellum is broader than in B. myrianthum, very obtuse, and broader in proportion to its length. The sepals of B. profusum are narrower in proportion to their length than those of B. myrianthum. Both species are near relatives of the Javan B. sessile J. J. Smith {B. clandestinum Lindl.) which is rather widely distributed, having been reported from Borneo, the Malay Peninsula, Singapore, and Tenasserim. An examination of additional material may show that B. myrianthum and B. profusum are referable to B. sessile and that the differences now relied on for differentia- tion are of slight importance. § CIRRHOPETALUM. Bulbophyllum antenniferum (Lindl.) Reichb. f. in Walp. Ann. Bot. Syst. 6 (1860) 245. Cirrhopetalum antenniferum Lindl. Bot. Reg. 29 (1843) sub. t. 1^9. The type specimen of Cirrhopetalum antenniferum in Lindley’s Her- barium at Kew consists of a solitary flower. There are no leaves and no pseudobulbs. Lindley’s description of the flower is incomplete, no PHILIPPINE ORCHIDS. 139 reference being made to the labellum. The nearest approach to Lindley’s type, among the Philippine specimens I have examined, is a Bulbophyllum from Mindanao collected at Camp Keithley on Lake Lanao by Mary Strong Clemens, no. 445. The leaves of this specimen are about 10 cm long by 2.5 cm wide, from pseudobulbs 2 cm long. The lateral sepals are 5 cm long, about equaling the upper sepal. The petals are minute, prolonged into setiform tips. The lip is missing. Bulbophyllum auratum (Lindl.) Reichb. f. in Walp. Ann. Bot. Syst. 6 (1860) 261. Cirrhopetalum auratum Lindl. Bot. Reg. 26 (1840) Misc. 107; 29 (1843) t. 61. To this species I refer two series of specimens from Mindanao. The flowers suggest those of B. Cumingii, but are considerably smaller with a much broader upper sepal, abruptly terminating in a filiform appendage, and with the yellow-fringed petals broader in relation to their length. The foliage of the two species is very similar but in B. auratum the pseudobulbs are much smaller than in B. Cumingii. Plate 61 in the “Botanical Register” represents, with reasonable closeness to details, the specimens here referred to B. auratum. In B. Cumingii the lateral sepals are about 2.6 cm long. In B. auratum they are 1.6 cm long. B. Cumingii the upper sepal is 7 mm long including the filiform appendage, and in B. auratum it is about 4 mm long. Mindanao, Butuan Subprovince, San Vicente, Agusan River, C. M. Weber 137, June 2, 1911, epiphyte, flowers pink; near Butuan, A. Celestino, Bur. Sci. 12128, October, 1910. Bulbophyllum Bolster! sp. nov. Aff. B. chryseo sed in floribus major. Rhizoma repens. In- ternodia 2.3 cm longa. Pseudobulbi 1.5 ad 2 cm longi, cylin- dracei, ad apicem fastigiati, in sicco rugosi, monophylli. Folia breviter petiolata, elliptico-lanceolata, acuta, coriacea, 3.3 ad 5.2 cm longa, usque ad 1.7 cm lata. Petiolus 2 mm longus, sulcatus. Scapus filiformis, 8 ad 12 cm longus, uniflorus, bracteis tubu- laribus 2. Bractea 1 in basi pedicelli unici. Pedicellus cum ovariis 1.5 cm longus, filiformis. Sepala lateralia leviter con- glutinata, lineari-lanceolata, acuta, 1.6 cm longa, 3.5 mm lata prope basim. Sepalum superius simile valde concavum. Petala lanceolata, 1-nervia, 6.5 mm longa, 2.5 lata, diaphana, glabra. Labellum petalis longior, lineari-lanceolatum, obtusum, 9 mm longum, 1.5 mm latum, ad basim biauriculatum, auriculi rotun- dati, erecti. Columna minuta, stelidia subulata, gracilia, 1 mm longa, glabra. Mindanao, Province of Surigao, F. H. Bolster 349, May, June, 1906. Epiphyte, flowers yellowish with purple markings. Here should be re- ferred Bur. Sci. 3076 Ramos, collected in Rizal Province, Luzon, a plant previously confused with B. chryseum!' This Journal 6 (1911) Bot. 54. 140 B. Bolateri and B. chryseum (Kranzl.) Ames are very near allies, differing mainly in the more rambling habit of B. Bolsteri and in the proportions of the flowers, B. Bolsteri being a larger flowered species with smooth petals, the median nerve unbranched. Unfortunately, I have not seen the type of B. chryseum, but among the Philippine plants in my herbarium there is one collected by W. S. Lyon 152, which answers so closely to Kranzlin’s description in Fedde’s “Repertorium” that I feel sure it must be that species. B. Bolsteri is suffused with purple in dried specimens, conspicuously so at the base, while B. chryseum is light-yellow. Bulbophyllum carinatum sp. nov. Pseudobulbi cylindracei, rugosi, 3 cm longi, juniores vaginis mox in fibras solutis inclusi, monophylli. Folia permagna, petio- lata, oblongi-elliptica, coriacea, in petiolum attenuata, in sicco rugosa. Scapus elongatus, bracteis tubularibus vestitus, uni- vel pauciflorus. Flores aurantiaci. Pedicellus gracilis. Sepala lateralia elongata, subfalcata, sulcata, valdi carinata, ad basim connata et pedi columnae adnata, mentum breve formantia, supeme libera. Sepalum superius sepalis lateralibus multo brevier, ovato-lanceolatum, acuminatum, valde acutum. Petala ovato-lanceolata, acuta, Labellum crassum, interne sulcatum, marge partis dimidiae basilaris erectus; pars dimidia superier, cenduplicata, margine cepiesi pilis instructis. Celumna bipar- tita, laciniae bipartitae. Leaves, including the sulcate petiele, 21 te 30 cm leng, 3.5 te 5.5 cm wide. Petiele abeut 5 cm leng. Scape 17 te 26 cm leng. Bracts abeut 1 cm leng, infundibuliferm. Pedicel and evary 2 cm leng. Flewers 1 er 2, mettled, yellew-red. Lateral sepals 4 cm leng, 5 mm wide, censpicueusly carinate, sulcate abeve, triquetreus. Upper sepal 3 cm leng, abeut 1 cm wide. Petals strengly reflexed, 1.5 cm leng, 6 te 7 mm wide. Labellum 1.4 cm leng, cenduplicate fer the mest part, the basal pertien cerdate when spread eut, with setiferm, marginal teeth near the peint ef insertien, the margin etherwise, te abeut the middle ef the lip, irregularly and shertly dentate. The lip, just beyend the middle is cenduplicate, the upper margin leng-papillese, papillae fleshy; en the under side it is lengitudinally sulcate. Celumn abeut 1 cm leng, divided at abeut the middle inte twe spreading lateral arms er wings, each wing deeply unequally cleft, the anterier divisien terminating in a setiferm tip. Mindanao, Province of Misamis, Mount Malindang, Mearns and Hut- chinson, For. Bur. U593, May, 1906, found at an altitude of about 900 m above sea-level. B. carinatum suggests B. Leysianum Burbidge Journ. Roy. Hort. Soc. PHILIPPINE ORCHIDS. 141 134 (1894) fig. 19, from which, in detail, it is clearly distinct. The carinate lateral sepals, the fleshy labellum, sulcate beneath, with its papillose apex, are marked peculiarities which distinguish B. carinatum from all other Philippine species of the genus heretofore described. Bulbophyllum luzonense sp. nov. Rhizoma repens. Pseudobulbi approximati, cylindracei. Folia oblonga, coriacea, ad apicem attenuata, breviter petiolata. Petiolus sulcata. Scapus folia brevior, bracteatus, suberectus. Bracteae inflorescentiae numerosae, confertae. Pedicelli gra- ciles, divaricata. Flores albidi, racemo in umbellam constricto. Sepala lateralia longe triangulari-lanceolata, acuta, 3-nervia, libera, patentia. Sepalum superius simile. Petala triangulari- lanceolata, angustata, acuminata, acuta, glabra, 1-nervia. Label- lum minutum, lingulatum, glabrum, obtusum. Columna parvula. Pseudobulbs 2 to 2.7 cm long. Leaves about 8 cm long, 1.2 to 1.8 cm wide. Petiole 3 mm long. Scape about 8 cm long, in fruiting specimens 10 cm long. Bracts subtending the umbel linear, about 5 mm long. Pedicel and ovary about 1.5 cm long. Flowers about 8 in each umbel. Lateral sepals 7 to 9 mm long, about 1.5 mm wide near the base. Petals about 4 mm long, less than 1 mm wide near the base. Labellum about 2 mm long, obtuse. Luzon, Province of Cagayan, H. M. Curran, For. Bur. 16787, March, 1^09. This is an interesting species of § Cirrhopetalum easily distinguished from all other Philippine species of the same section by its loose umbels and long, narrowly triangular petals which are smooth and similar in form to the sepals. The upper sepal is remarkable because of its similarity to the lateral sepals. Bulbophyllum stellatum sp. nov. Rhizoma repens, crassum. Pseudobulbi cylindracei, ad basim tumidi, prope apicem paulatim fastigati, rugosi, 2.5 ad 3 cm longi in sicco, ad basim 8 mm in diametro, monophylli. Folia coriacea, oblongi-elliptica, breviter petolata, 8.5 ad 9 cm longa, usque ad 3 cm lata. Petiolus 5 mm longus, sulcatus, rigidus. Scapus folio brevior, usque ad 11 cm longus, bracteis 3 vaginan- tibus. Bracteae inflorescentiae confertae, circiter 2 mm longae, lineari-lanceolatae. Pedicellus cum ovario 1 cm longus, gracilis. Flores albidi et rubidi, racemo in umbellam constricto. Sepala lateralia subpatentia, 5 mm longa, prope basim 2 mm lata, 3- nervia, lanceolata, acuta. Sepalum superius valde concavum, 6 mm longum. Petala lanceolata, acuminata, acuta, prope basim 111022 2 142 AMES. 3-nervia, 5 mm longa, 1.5 mm lata prope basim. Labellum parvulum circiter 2 mm longum, anguste lingulatum, obtusum. Columna brevis, stelidia longe aristata, filiformia. Luzon, Bontoc Subprovince, Father M. Vanoverbergh 1360, August 3, 1911. Found on trees, altitude 1,650 m above sea-level. The flowers are described by the collector as yellowish-white and red. Among the Philippine species of § Cirrhopetalum, B. stellatum is most closely allied to B. luzonense Ames, from which it differs in its larger leaves, shorter pedicels and smaller flowers. Bulbophyllum trisetum sp. nov. Aff. B. longifloro Thou. Rhizoma crassum, repens. Pseudo- bulbi pyriformes, plus minus tetragoni, remoti. Folia coriacea, petiolata. Lamina elliptico-oblonga. Petiolus sulcatus. Scapus elongatus, erectus, bracteatus, gracilis. Flores albi et purpurei, racemo in umbellam contracto. Bracteae inflorescentiae lan- ceolatae, acutae. Pedicelli graciles. Sepala lateralia supra ba- sim fere ad apicem cohaerentia, glabra, oblonga vel ligulata. Sepalum superius cucullatum, in caudam filiformem, constrictum, ad apicem dilatatum, margo dentatus. Petala triangularia, pur- purea, in caudam elongatam, integerrimam, abrupte attenuata, ciliata. Labellum crassum, lingulatum, glabrum, obtusum, con- vexum. Columna erecta, basi in pedem producta, superne 2- alata, alis utrinque in brachium erectum productis. Pseudobulbs 3 to 4 cm long, about 1.5 cm thick. Leaves about 12 cm long, up to 3.5 cm wide. Petiole 1.6 cm long. Scape about 15 cm long. Umbels few-flowered. Lateral sepals 2.7 cm long, each one about 3 mm wide. Upper sepal from base to end of setiform or filiform tip 2.3 cm long, filiform tip 1.5 cm long, slightly dilatated at the end. Petals 1.2 cm long to end of the setiform tip, setiform tip 5 to 6 mm long. Labellum 7 mm long. Column wings 4-toothed at the tip; anterior tooth prolonged into a setiform appendage 2 mm long. Luzon, Benguet Suprovince, Itogan to Baguio, E. S. Bacani, For. Bur. 15899, December 24, 1908, found on trees. Bulbophyllum trisetum. is closely allied to B. longiflorum Thou., from which it differs in the color of the flowers, the longer appendages of the sepals, and in the different toothing of the wings of the column. It is also near Cirrhopetalum Koordersii Rolfe, a Celebes species. The collector described the flowers as being white, but in dried specimens the petals ajid upper sepal are of a rich deep-madder color. Bulbophyllum Weberi sp. nov. Pseudobulbi plus minus tetragoni, 2 ad 3 cm longi, paulatim fastigiati ad apicem, juniores vaginis mox in fibras solutis inclusi. Folium late oblongi-ellipticum, coriaceum, petiolatum, 9.5 ad 13.5 PHILIPPINE ORCHIDS. 143 cm longum, 2.5 ad 4.4 cm latum. Petiolus circiter 2 cm longus, sulcatus, rigidus. Scapus folio longior, gracilis, bracteis circa 3 vaginantibus. Bracteae inflorescentiae lineari-lanceolatae, 3 mm longae, valde acutae. Flores circiter 8, flavidi et rubidi, 2.5 cm longi, speciosi, racemo in umbellam contracto. Pedicelli graciles, bracteas excedentes. Pedicellus cum ovario 1.5 cm longus. Sepala lateralia 2.7 cm longa, supra basim fere ad apicem connata in lamina cohaerentia; lamina oblongi-elliptica, 11 mm lata, flavida. Sepalum superius, 6 mm longum, valde concavum, suborbiculare, non ciliatum, ad apicem in caudam fili- formem constrictum; cauda 4 mm longa. Petala 5 mm longa, ciliata, 3-nervia, in caudam filiformem, 1.5 mm longam, con- stricta. Labellum valde arcuatum, circiter 5 mm longum, lin- gulatum, obtusum, glabrum. Columna superne 2-alata, alls in brachium erectum producta. Anthera denticulata. Luzon, Province of Rizal, Mount Canumay, C. M. Weber 8, found on mossy trees, 700 m above sea-level. The affinity of this species seems to be with B. Cumingii Reichb. f. and with B. longiflorum Thouars. From the former it is readily Separated by the color of the flowers, the smooth margin of the dorsal sepal, and by the arms of the column. The column-arms are similar to those of B. longiflorum as shown in the figure of Cirrhopetalum Thouarsii in the “Botanical Magazine” plate 7214. The denticulate margin of the anthers is another distinguishing character in the comparison of B. Weberi and B. Cumingii. From B. longiflorum it is not so easy to distinguish B. Weberi. Here we have a striking similarity in the structure of the column, the erect wings and the denticulate anther-cap in both species. The plate formed by the union of the lateral sepals is rather more elliptic in B. Weberi than in B. longiflorum, and the leaves are broader in relation to their length. From the Cirrhopetalum Thouarsii of the Botanical Register (1838) t. 11 and of the Botanical Magazine, t. 4239, a plant which appears to be different from the C. Thouarsii of Bindley, B. Weberi is hardly distinct. i ! The PHH.IFPINE Journal of Science, C. Botany. Vol. VII, No. 3, August, 1912. NOTES ON THE FLORA OF MANILA WITH SPECIAL REF- ERENCE TO THE INTRODUCED ELEMENT. By E. D. Merrill. (From the Botanical Section of the Biological Laboratory, Bureau of Science, and the Department of Botayiy, University of the Philippines, Mar.ila, P. I.) In compiling data and in making explorations of Manila and its vicinity with the ultimate object of the preparation of a flora of the region, which is now in press,^ much information other than that of a purely systematic nature has been secured. It has not been considered advisable to include this, however, in the forthcoming publication. An opportunity has been taken in the present paper for some discussion of the climatic con- ditions, the different types of vegetation, and the geologic and geographic features of the region under consideration. In con- nection with the above there are also included some facts and conclusions regarding the vegetation of Manila and its vicinity, the introduced element and its origin, the pantropic species and their dissemination, and other matters that cannot prop-' erly be considered in a short preface or introduction. Most of the present paper is devoted to a consideration of the intro- duced element, which is a very large one in the vicinity of Manila, and, on the whole, the most interesting botanical feature of the area covered. The area under consideration comprises approximately 100 square kilometers, roughly a triangle with its base on Manila Bay, and its apex at the town of Pasig. Its limits are as follows : From the town of Malabon on Manila Bay, north of the City of Manila, in a southeasterly direction through Masambong, San Francisco del Monte, San Juan del Monte, etc., to Pasig, thence ’ Merrill, E. D. A Flora of Manila. Bureau of Science Publication (1912). 145 146 MERRILL. in a southwesterly direction to the town of Paranaque on Manila Bay south of Manila. Practically the entire area occupied by the city of Manila consists of an alluvial plain, for the most part located but little above the limits of high tide. This plain is divided by the Pasig River and some of its tributaries, the smaller tidal streams being locally known as esteros. Undoubtedly the greater part of the area now occupied by the city was originally a mangrove swamp; but with the extension of the town the low places have been graded, and the original mangrove vegetation destroyed. Vestiges of this original vegetation still remain along some of the esteros; and at Malabon, just north of the city, the mangrove vegetation is rather conspicuous in places. The word Manila itself is generally assumed to have been derived from nilad or nilar, one of the Tagalog names for a mangrove tree, Scyphyphora hydrophyllacea Gaertn. (Rubia- ceae) . The prefix ma means “where is,” so that the word Manila signifies the place where nilar grows. Whether or not this theory is correct the species is no longer to be found in or near Manila, although it is sufficiently abundant in other parts of the Archipelago. Within the area covered by the “Flora of Manila,” and the region on which the present paper is based, there are a number of low hills, none of which exceed an altitude of 50 meters. They consist of water-laid tuff, with usually a thin covering of poor clayey or gravelly soil. These hills occur in limited areas within the city limits, but are much more numerous inland. There are, then, only two geological formations, the alluvial plain and the low hills of water-laid tuff with their thin cover- ing of poor soil. By far the greater part of the area under consideration consists of waste places, cultivated and fallow fields, and thickets. In restricted areas we have the strand formation occurring along the sandy shores of Manila Bay ; the vestiges of mangrove swamps along tidal streams and muddy shores; and swampy areas with fresh water, brackish water, and salt water. The entire area under consideration, so far as its vegetation is concerned, has been profoundly altered by the presence of man, and at no place near the city do conditions at all approach- ing primeval ones exist. The original vegetation has been almost entirely destroyed, for presumably the present location of the City of Manila was for the most part covered with mangrove forests, as noted above, while the dry hills undoubtedly sup- THE FLORA OF MANILA. 147 ported a low and more or less scrubby forest. Comparatively few of the original forest types exist in the area at the present time. From a consideration of the above notes it will readily be seen that the soil and topographical conditions in the vicinity of Manila are not favorable to a diversified flora. In connec- tion with these matters the question of rainfall must be con- sidered, and here again the local conditions militate against a multiplicity of species. From June to November, the period of the southwest monsoon, there are heavy rains and typhoons. This season is followed by several months of comparatively dry weather, with only scattered showers. February to May, the period of the north-east monsoon, is very dry, with almost no rain, making the conditions on the dry hills almost desert ones during these months. In those parts of the Philippines, where the rains are fairly continuous throughout the year, one finds at low altitudes a great number of species that are not found near Manila. These species, moreover, are not encountered at low altitudes in those parts of the provinces contiguous to the city and having a rainfall similar to it. Epiphytes, always abundant in continuously humid regions in the tropics, are represented by very few species in the vicinity of Manila. From the standpoint of the collector or the systematist, the vegetation in the vicinity of Manila is a disappointing one, especially in consideration of the tropical habitat. Most of the species found are common and well known ones of very wide distribution. Including all the more commonly cultivated forms, there are 1,007 species, of which about 225 are never found outside of cultivation, leaving 782 that are indigenous or naturalized. Undoubtedly the original flora of the country in and about Manila was a comparatively poor one, so far as the number of species was concerned, and at the present time it is certain that in species the flora is much richer that it was in primeval times. With the destruction of the original forests, and subsequent cultivation of the soil, the proper habitats were provided for species adapted to conditions of growth in the open. Since the advent of man in the Philippines, some thousands of years ago, the process of introduction of exotic, light-loving species, pur- posely and accidentally, has been going on, and is still being continued at the present day. Then, too, the proper habitats were provided for the rice-paddy, and other small-seeded weeds of cultivation, which were thus enabled to persist and to extend 148 MERRILL, their range after once being introduced. Some of these were undoubtedly introduced in dirty grain, but for the most part probably by natural causes; by winds, and especially by the transmission of their small seeds in mud on the feet or feathers of migratory birds. Thus directly and indirectly, through the agency of man, many species have been added to the flora of the Archipelago. Likewise through the agency of man, espe- cially in the destruction of the original vegetation over vast areas, it is very probable that many local indigenous species have been exterminated. When, then, we consider the nature of the geological forma- tion in and about Manila, the alluvial plain, the dry hills of water-laid tuff covered with a thin poor soil, the lack of alti- tude, the uneven distribution of rainfall through the year, causing a protracted very dry season, the entire lack of forests, and the fact that primeval conditions as to vegetation are not to be found in the entire area, a flora of approximately 1,000 different species, or, excluding the cultivated ones, 780 indige- nous or naturalized ones in an area not exceeding 100 square kilometers, cannot be considered a poor one so far as individual species are concerned. In fact, as to the number of species, the flora of Manila com- pares very favorably with that of like areas in the Archipelago supporting a primeval vegetation. Mount Mariveles, just across the bay from Manila, is densely covered with unbroken forests, and its vegetation, except for a narrow coastal strip, has not been disturbed by man. In contrast to the 1,007 species known from Manila and vicinity, or excluding the cultivated forms, 782 native or naturalized ones, only about 1,114 are knovTi from Mount Mariveles, although the latter has the climatic advan- tage of 1,400 meters of altitude, is densely forested, and supports several types of vegetation that are quite absent in and about Manila. Properly to understand the local conditions as to vegetation, the fundamental differences between the flora of the settled and cultivated areas in contrast to that of the undisturbed regions supporting a primeval vegetation, and the enormous influence man has had on the present condition of the vegetation of the Archipelago, it is necessary to take into consideration the prob- able nature of the vegetation before the advent of man. Considering the Philippine Archipelago as a whole, at the present time approximately thirty-three and one-third per cent of the entire land area is covered with virgin forest, sixteen and THE FLORA OF MANILA. 149 two-thirds per cent by second-growth forest, forty per cent is grass land, and ten per cent cultivated land.^ It is practically certain that before the advent of man in the Philippines, the entire country was covered with unbroken forest, of one kind or another, from sea-level to the tops of the highest mountains, except, perhaps, where the vegetation had been temporarily destroyed by natural causes, such as volcanic eruptions. Such types of vegetation as the extensive grass-covered hills, moun- tain sides, and plains, and the open cultivated areas, now such prominent features in the landscape, did not originally exist, so that the whole aspect of many localities must have been quite different from what it is to-day and from what has been its condition within historic times. When we consider that about two-thirds of the entire land surface of the Archipelago consists of cultivated areas, open grass lands, thickets, and second-growth forests, and that all these types of vegetation are due directly or indirectly to the presence of man, some idea can be obtained of the profound changes that have been wrought in the vegeta- tion of the country in past centuries. The first agricultural method employed in the Archipelago, whether by the supposedly aboriginal Negritos, or by later Malay invaders, was certainly the “caingin” system, still very exten- sively practiced. This primitive system of agriculture consists simply in clearing a selected area by felling the trees, burning the debris, and raising some simple crop on the area thus cleared. Such a clearing is, in general practice, utilized but one or two years, rarely longer ; and is abandoned as coarse grasses, weeds, shrubs, etc., commence to gain an ascendency over the cultivated plants. With only most primitive agricultural instruments, the average native to-day, as in centuries past, finds it much easier to fell and burn the trees, thus clearing a new area, than to combat the encroachment of weeds, and especially grasses such as the cogon (Imperata ctjlindrica Beauv.) with deep-seated perennial rhizomes. Moreover in each new clearing that he makes he has the advantage of virgin soil, although this last factor is in most cases doubtless a secondary consideration. A clearing once abandoned, may, depending on local conditions, revert to a forested condition through a succession of coarse annual weeds, shrubs, and various quick-growing trees, or it may become covered with coarse grasses, especially the cogon {Imperata cylindrica Beauv.), or sometimes other species, such ^Whitford, H. N., Bull. Bureau Forest. (Philip.) 10' (1911) 12. 150 MERRILL. as Saccharum, Rottboellia, Themeda, etc. The primitive agricul- turist selects another forested area, usually contiguous to the one just abandoned or about to be abandoned, fells the trees, burns the debris, and plants his crop. If the abandoned clearing becomes reforested, then it is only a question of time when the land will again be cleared for cultivation by the “caingin” system. If, however, as is most often the case, it becomes covered with coarse grasses, the chances are that the area will not become reforested. Annually recurring grass fires are purposely started either for improving grazing conditions, or to make the country more easily traversed. These prevent the reproduction of forest conditions by killing the young trees, seedlings, etc., which may have become established among the grasses, and, in fact, all the plants except those which, like the coarse grasses, are pro- vided with deep-seated perennial rhizomes. As the grass-covered areas become more extended, the fires often gradually push back the edges of the forest by the destruction of the young and the mature trees along the edges of the clearing, even when not aided by man. As a result in many hilly areas in the Phil- ippines one finds the continuous slopes covered with grasses, the forest persisting only in the ravines, where, perhaps, the original clearings made by man were of comparatively limited extent. Forest fires, as such, are practically unknown in the Philip- pines, except perhaps in some very limited areas. The average high humidity, the very rapid decay of fallen leaves, branches, trees, etc., and the nature of the undergrowth militate against them. To permit the ravages of forest fires in the Philippines, climatic conditions in ages past must have been very diiferent from what they are to-day, and any great difference in this respect, especially in decreased humidity, is exceedingly unlikely. Forest fires as a factor in the establishment of the open grass- covered areas can hence be ignored. In grass-covered areas, protected against fire, it is however, only a question of time when forest conditions will again prevail, the original vegetation of the country claiming its owm. It is, therefore, due to the presence of man that the open grass country exists here, and to his continued presence that this type of vege- tation persists and still continues to extend itself at the expense of the forested area. Man is directly responsible for nearly all of the annually recurring grass fires in all parts of the Archipelago, very few being started from natural causes. THE FLORA OF MANILA. 151 Assuming, then, that the original vegetation of the Philippines was practically unbroken forest, it becomes manifest that a very high percentage of the species now dominant in the settled areas, that is, in waste places in and about towns, fallow fields, cultivated areas and thickets, and in the open grass country, must have been introduced into the Archipelago after the advent of man and after sufficient time had elapsed for man to have provided the habitats to which these species are adapted. The great bulk of these plants are those requiring abundant light, adapted to growth in the open, and which cannot persist under forest conditions. They are for a large part species of very wide distribution that certainly have not originated in the Archi- pelago, but which must have developed their characteristics as to habitat in non-forested, or at least very thinly forested countries. On the assumption that the great majority of the species now found at low altitudes in the settled areas of the Philippines have reached the Archipelago by one means or another since the advent of man, it is manifest that should any large area become depopulated for a long term of years, and the vegetation thus become protected against fires and other disturbing factors due to the presence of man, the original vegetation of the Archi- pelago would again occupy the entire area. The absolute or nearly absolute extermination of the introduced forms which depend for their existence on an open habitat, and which cannot thrive in forests, would follow. This would mean that the great majority of the species now found in the settled areas would become either extinct, or of rare and very local occurrence in the limited areas where conditions for their growth and re- production might persist. In addition to the herbs, undershrubs, and other plants of the country, now so common and widely distributed, which would in all probability be exterminated with the reversion of the country to primeval conditions, it is exceed- ingly probable that very many of the introduced trees, including the naturalized ones, would eventually become extinct, from their inability to withstand forest conditions, that is, their inability to compete with the native trees and vines in the struggle for light. In Table I, following, is given a summary of the species con- sidered in the “Flora of Manila,” a total of 1,007 distributed into 595 genera and 136 families. Of these 1,007 species it is estimated that about 550 are indigenous, that is, true natives of the Archipelago or those that have reached the Islands through 152 MERRILL. natural agencies, and that 457 have purposely or inadvertently been introduced by man. Of the introduced species somewhat over one-half, or 232, are spontaneous, while about 225 are never or but very rarely met with outside of cultivation. One hundred and seventy-eight of these introduced forms are con- sidered to have originated in tropical America, the remainder for the most part in the tropical parts of the Old World. Four hundred and twenty-five species are found in the tropics of both hemispheres, but less than 90 of these are considered as being possible or probable natives of both; 334 are considered as having been manifestly distributed by man, and chiefly within historical times. Of the total of 1,007 species found in and about Manila 812 extend to the Malay Archipelago, 789 to tropical Asia, 425 to tropical Australia, 355 to Polynesia, and 402 to tropical Africa. Table I. — Summary of the families, genera, and species, found in and about Manila, with indications of the origin and range of the various forms. THE FLORA OF MANILA. 153 Table I. — Summary of the families, genera, species, etc. — Continued. r Intro- duced. Ameri- can origin. Pantro- pic: Dis- tributed- 1 1 Families and genera. Total species. 1 Indigenous. Spontaneous. I Cultivated only. | j Endemic. Spontaneous. 1 j Cultivated only, j By natural agen- 1 cies. i Through agency of man. | To Malaya. To tropical Asia. To Australia. To Polynesia. To Africa. Lycopodiaceae; 4 4 4 4 1 i Selaginellaceae: 3 2 1 1 Cycadaceae; 1 Typhaceae: 1 1 1 1 Pandanaceae: 1 Potamogetonaceae: 1 1 1 1 1 \ 1 1 Najadaceae; Najaa 1 1 Hydrocharitaceae: Hydrilla 1 i Vallisneria \ \ J \ \ 1 Ottelia 1 1 1 J Halophila i 2 2 1 1 j Thalassia - 1 1 1 Gramineae; Zea 1 1 1 1 Coix.. j J 1 \ J Dimeria 1 1 J 1 J Imperata 1 J 1 1 Pogonatherum ... | 1 1 J 1 1 Polytrias 1 1 1 1 Arthraxon ■ J Pollinia... [ 1 1 1 1 Saccharum 2 1 1 1 2 2 1 J Rottboellia ... 1 1 1 J Manisuris | 1 1 J 1 J Ophiurus 1 1 1 J Apluda 1 1 1 J 1 t Ischaemum [ 3 2 2 Themeda ! 1 li ! \ Andropogon \ 11 6 3 2 3 9 10 5 3 3 Zoisia.. . _ 1 1 1 1 1 1 Eriochloa \ 1? \ J 1 1 Arundinella ... I 1 1 Paspalum | 4 1 2 1 4 4 3 3 3 Digitaria _ I 6 4 2 2? 5 4 4 1 2 ■ “ I Isachne... . [ 1 1 1 1 Panicum j 17 11 5 1 4? 1 15 2 16 6 6 6 Oplismenus .. 2 2 1? 2 2 2 2 Cenchrus i 1 1 I 1 1 154 MERRILL. Table I. — Summary of the families, genera, species, etc. — Continued. 1 Intro- duced. Ameri- can origin. Pantro- pic: Dis- tributed- 1 j Families and genera. Total species. 1 Indigenous. Spontaneous. j Cultivated only, j j Endemic. Spontaneous. | ! Cultivated only. | By natural agen- 1 cies. 1 Through agency of man. | To Malaya. To tropical Asia. To Australia. To Polynesia. To Africa. Gramineae— Contd. j 1 1? 1 2 3 2 2 1 i 1 1 1 1 : 1 1 1 J i 1 1 i 1 J 1 J 1 1 . 1 1 i J 1 1 2? 2 ! 2 2 2 2 1 J 1 !i 1 1 1 1 . i! 1 1 1 1 I ! i 1 1 I 1 i 9 5 4 ! 2 2 J 6 3 1 6 2 2 j 1 1? 1 1 i 1 1 1 1 Chloris J 1 1 ! ^ ' 1 ( Eleusine 1 J 1 1 1 J 1 1 J ' Dactyloctenium 1 1 1 1 1 1 1 1 ' Leptochloa - . - 1 i.--_ 1 1 1 3 3 1 3 J j Bambusa ! Schizostachyum 1 i 1 i j Cyperaceae: t Kyllinga 2 2 2 2 1 2 2 Pycreus 3 ! 3 ! 3 3 3 3 } Juncellus 1 ! 1 1 1 1 Cyperus 12 11 J 7? 1 12 12 7 12 i Mariscus 4 4 4 1 4 \ 2 j Torulinium - 1 1 1? 1 1 1 1 1 1 1 1 Eleocharis 3 3 1 1 3 1 2 3 1 1 j Fimbristylis - - 11 11 1 5? 10 10 9 7 B 1 5 5 2? 6 5 1 i 4 2 1 4 j Scirpus - - Fuirena 1 1 1 1 1 Lipocarpha 1 1 1 1 1 1 Rhynchospora 1 1 I 1 1 4 4 4 2 ' Diplacrum . 1 J 1 Palmae: j Livistona ~ 1 1 1 -- 1 1 1 Corypha 1 1 1- ( 1 1 Caryota 2 2 J 1.... 1 J ..._i 1 Nipa 1 1 1 j. . 1 1 1 J L.. 1 1 1 1 1 1 1 1 1 ] 1 I 1 1 1 1 1 i 1 1 1 1 Oreodoxa J 1 1 1 1 1 1 ! J \ 1 1 1 Areca 1 1 1 ! 1 i 1 L.. 1 1 ! THE FLORA OF MANILA. 155 Table I. — Summary of the families, genera, species, etc. — Continued. Families and genera. Total species. Indigenous. Intro- duced. 1 Endemic. Ameri- can origin. Pantro- pic: Dis- tributed- 1 To tropical Asia. 1 To Australia. To Polynesia. To Africa. Spontaneous. j Cultivated only, j Spontaneous. i Cultivated only, j By natural agen- 1 cies. 1 Through agency of man. | Araceae: 1 1 1 1 J J 1 1 1 J J 1 1 1 Amorphaphallus 1 1 2 1 1 1 2 1 1 1 5 4 1 1 , J J 1 1 1 1 1 Lemnaceae: i 1 1 2 1 1 1 1 1 1 Xyridaceae: Xyris j J 2 2 2 Eriocaulaceae: I EriocRulon i 3 3 1 2 2 2 ■ 1 2 Bromeliaceae: j Ananas 1 1 J 1 1 2 2 1 2 1 Commelinaceae: j Commelina 2 2 1? 2 2 2 2 Aneilema 2 2 2 1 1 Rhoeo - 1 1 J J Cyanotis.. , 3 3 2 1 Zebrina J J 1 1 Pontederiaceae; Monochoria 2 2 2 2 Liliaceae; Smilax 1 1 1 Asparagus 3 3 3 3 3 Allium 2 2 2 2 2 2 2 Sansevieria 1 J J 1 1 1 Cordyline 1 J J 1 1 1 Yucca J 1 1 1 Amaryllidaceae: | Agave 1 ; 1 2 1 1 1 2 Polianthes 1 ..... 2 J 1 1 J 1 1 Zephyranthes 1 1 1 J 1 Crinum 3 1 2 1 3 1 Pancratium 1 1 1 Hymenocallis 1 1 1 1 ! 1 1 Eurycles _ J I 1 Eucharis \ 1 1 J Taccaceae: Tacca.. . . 2 2 2 2 2 1 1 Dioscoreaceae: Dioscorea ... 6 4 1 1 3 i 4 2 156 MERRILL. Table I. — Summary of the families, genera, species, etc. — Continued. Intro- duced. Ameri- can origin. Pan tro- pic: Dis- tributed- Families and genera. ! Total species. 1 j Indigenous. Spontaneous. Cultivated only. | 1 Endemic. Spontaneous. j Cultivated only. | By natural agen- I cies. 1 Through agency of man. | To Malaya. To tropical Asia. To Australia. j To Polynesia. To Africa. Iridaceae: 1 1 1 1 1 1 1 1 Musaceae: (?) 1 1? ! 1 1 1 1 \ ' 1 1 Zingiberaceae: 1 1 J 1 1 1 1 1 1 1 1 1 .....1 1 1 1 ! 2 2 2 1 1 1 1 1 1 Costus 1 1 1 ■ 1 Cannaceae: i Canna Marantaceae: Maranta.. i 6 2 4 2 4 1 1 ^ 1 1 > 1 1 2 Orchidaceae: i Didvmonlexia | I 1 1 1 1 Geodorum 1 Habenaria 3 3 1 2 2 Cleisostoma 1 1 1 Dendrohiiim 1 Oberonia - 1 \ 1 Casuarinaceae: Casuarina 1 1 1 1 1 1 Piperaceae: Peneromifl 2 1 1 1? Piper 2 1 2 2 2 1 Ulmaceae: Trema .. ... ! 1 1 1 1 Moraceae: Ficus . i 10 8 1 3 3 2 Artocarpus 1 1 1 2 1 2 Streblus i ^ 1 1 Castilloa _ 1 1 1 1 1 Moms . 1 1 1 J 1 Malaisia.. 1 1 1 J Cudrania 1 1 J Fatoua 1 1 1 Urticaceae: Pintnrns 1 Laportea Boehmeria.. 1 1 1 1 1 1 1 1 Fleurya . .. 1 1 1 1 1 THE FLORA OF MANILA. 157 Table I. — Summary of the families, genera, species, etc. — Continued. Families and genera. Urticaceae— Contd. Pilea 1 Elatostema 1 Pouzolzia 1 Loranthaceae: Loranthus.- 2 Opiliaceae: Champereia 1 Olacaceae: 01ax___ 1 Aristolochiaceae: Aristolochia 2 Polygonaceae: I Polygonum 2 Antigonon ] 1 Muehelenbeckia 1 1 Chenopodiaceae: Chenopodium j 1 Amaranthaceae; Deeringia 1 Celosia 1 Gomphrena 1 Alternanthera 3 Amaranthua 4 Pupalia 1 Cyathula 1 Aerua 1 Achyranthes 1 Nyctaginaceae: Pisonia 2 Bougainvillea 1 Mirabilis 1 Boerhaavia 1 Aizoaceae; Sesuvium 1 Trianthema 1 Mollugo 2 Portulacaceae: Portulaca 3 Basellaceae: Basella 1 Anredera 1 Nymph.®aceab: Castalia 1 111022 8 To Polynesia. 158 MERRILL. Table I. — Summary of the families, genera, species, etc. — Continued. Families and genera. Total species. Indigenous. | Intro- duced. Endemic. Ameri- can origin. Pantro- pic: Dis- tributed- To Malaya. To tropical Asia. To Australia. To Polynesia. To Africa. Spontaneous. Cultivated only. Spontaneous. Cultivated only. | By natural agen- cies. 1 Through agency of man. | Ceratophyllaceae; 1 1 1 i 1 2 1 1 Ranunculaceae; 1 1 Menispermaceae: 1 1 1? j J 1 1 Magnoliaceae: 2 2 1 Anonaceae: 1 Anona 3 g 3 3 3 3 1 3 Artabotrys 1 1 Canangium 1 1 1 1 Unona 1 1 1 Polyalthia 1 1 1 Lauraceae: Cinnamomum 1 1 2 2 2 1 1 1 1 2 1 J Cassytha 1 1 1 j 1 1 1 Hernandiaceae: Tllifypra 1 1 Papaveraceae: j Argemone.. 1 1 1 1 2 1 1 Cruciferae: 1 NflQtnrt.inm 1 1 1 Brassica 1 1 1 1 1 1 1 1 Raphanus 1 1 1 1 2 1 1 Capparidaceae: 1 r^apparift 4 g 1 4 1 Cleome 1 1? 1 1 1 1 Gynandropsis 1 1? 1 1 1 Moringaceae: Moringa 1 1 1 1 j 1 1 1 Crassulaceae: Bryophyllum 1 1 1 1 1 1 1 1 Tfalanphop 1 1 1 J Pittosporaceae: Pi ff r»«pr>r 11 m 1 j 1 Connaraceae: j 1 1 Connarus 1 1 THE FLORA OF MANILA. 159 Table I. — Sumviary of the families, genera, species, etc. — Continued. Families and .erenera. Intro- duced. Endemic. j Ameri- can origin. Pantro- pic: Dis- tributed- To Malaya. To tropical Asia. To Australia. j To Polynesia. 1 1 To Africa. 1 Total species. Indigenous. Spontaneous. Cultivated only, j Spontaneous. j Cultivated only. | By natural agen- I cies. o ba p; it o'® Leguminosae; J 1 , 1 1 1 1 1 1 3 3 2 1 1 1 2 2 J 1 1 J J 1 1 \ 1 J J 1 1 1 1 J 1 1 3 3 2 J 1 1 1 1 Caesalpinia . 4 2 2 1 2 1 4 4 2 2 2 Peltophorum 1 1 1 1 1 5 4 j 4 5 5 5 3 3 3 Tamarindus 1 i 1 1 1 1 1 1 1 J 1 Pahudia _ ... 1 1 ( Cynometra \ 1 1 Aeschynomene - _ 1 1 1 1 1 Arachis 1 1 J J 1 ^ ! Zornia _ - 1 1 1? J J J 1 1 1 Desmodium 11 3 8 2 1? 3 g g 3 4 5 Alyscicarpus 3 3 3 1 3 \ 1 ■"1 1 1 1 1 1 Lourea - . _ 1 1 1 1 Indigofera 3 1 3 3 3 2 1 3 j Tephrosia .. j Millettia 1 J Gliricidia _ 1 1 Clitorea 1 1 1? 1 1 1 1 1 Parosela . 1 Pterocarpus 2 2 2 , J Dalbergia Pongamia _ 1 1 1 1 3 3 2 2 2 1 Pionm 1 1 1 1 Ahrno 2 2 1 2 2 1 1 2 Sesbania 3 3 1 3 2 2 3 Crotalaria 6 2 4 1 2 5 5 5 2 4 Flemingia^. 2 1 1 1 1 1 Medicago 1 1 1 \ 1 Pueraria 2 1 1 2 2 Dunharift 1 1 1 1 1 CantharoBpermum . 2 2 2 3 1 160 MERRILL. Table I. — Summary of the families, genera, species, etc. — Continued. THE FLORA OF MANILA. 161 Table I. — Summary of the families, genera, species, eic.— Continued. 162 MERRILL. Table I. — Summary of the farnilies, genera, species, etc. — Continued. 1 Families and genera. s 1 1 Indigenous. j Intro- duced. 1 Endemic. | Ameri- can origin. Pantro- pic: Dis- tributed- To Malaya. | To tropical Asia. j To Australia. To Polynesia. 1 To Africa. Spontaneous. 1 Cultivated only. Spontaneous. Cultivated only. By natural agen- cies. Through agency of man. Rhamnaceae: 2 1 2 1 1 J X 1 1 1 1 1 X X Vitaceae: 1 1 1 3 2 1 3 3 2 1 T J 1 J Tiliaceae: 1 1 1 1 1 X J 1 1 1 1 1 2 2 2 2 2 X 2 3 3 3 3 j Malvaceae: ^ Malvastrum 1 1 J 1 i Sida - 6 6 4 g g 4 3 5 1 Abutilon .. « 1 1 1 1 1 1 ! Malachra 2 2 2 1 1 \ 1 1 1 1 1 X J ( Malvaviscus 1 1 1 1 Kosteletzkya 1 1? 1? j Abelmoschus 1 1 2 2 2 2 2 2 Hibiscus - - 6 1 4 1 4 5 5 2 3 i Thespesia - 2 2 2 2 1 Gossypium 1 2 2 2 2 2 2 X Bombacaceae: Bombax _ j 1 1 1 X Ceiba t 1 1 1 1 1? J 1 J J Sterculiaceae; Stprcnlia 4 1 4 3 1 1 1 1 Heritiera 1 1 X 1 1 Theobroma [ ! 1 1 J 1 [ X I Kleinhofia J 1 J 1 J 1 TTplifttfirps 1 1 ^ 1 J 1 1 Pterospermum.. 1 1 1 , I Melochia » 1 1 1 1 1 J Waltheria 1 ! 1? 1 J 1 X J Dilleniaceae: Tetracera, . 1 1 1 1 1 Dillenia J 1 1 1 Guttiferae: 1 1 Calophyllum 1 1 X 1 1 Elatinaceae: "Rorcrio 1 i 1 Bixaceae: j 1 1 Bixa . 1 ' I 1 1 1 1 1 1 1 THE FLORA OF MANILA. 163 Table I. — Summary of the families, genera, species, etc.— Continued. Total species. Indigenous. Intro- duced. Endemic. Ameri- can origin. Pantro- pic: Dis- tributed- To Malaya. j 1 To tropical Asia. j To Australia. To Polynesia. j To Africa. Families and genera. Spontaneous. j Cultivated only. Spontaneous. [ Cultivated only. | 1 By natural agen- 1 cies. Through agency 1 of man. ViOLACEAE: 1 1 1 1 2 1 2 2 3 2 2 1 1 6 2 5 > ■ 1 1 1 Flacourtiaceae; 1 1 Passifloraceae: 1 2 1 2 3 3 3 1 1 1 I 1 Caricaceae: 1 1 ■ ■ ■ - 1 1 Begoniaceae: ■ ■ Cactaceae: Cereus 2 2 2 1 2 2 2 2 2 2 1 1 1 2 1 2 2 2 1 1 1 Nopalea Lythraceae; Tifiwsnnia 1 1 Lagers troemia 2 2 2 2 Rntala I? Animannia 1 Sonneratiaceae: Sonneratia 1 Punicaceae: Punica . _ . ■ ■ Lecythidaceae: Barringtonia 2 2 1 1 1 2 1 1 1 1 Rhizophoraceae: Rhi^ophora Bruguiera Combretaceae; Tenninalia "T Quisqualis Lumnitzera Combretum. . Myrtaceae: Eucalyptus . . Eugenia 3 1 1 1 2 2 (?) 2 2 1 1 I 1 4 1 1 2 5 4 1 2 2 5 Psidinm Melastomataceae: Memecylon ■ Oenotheraceae; Ludwigia 2 1 2 3 (?) Jussiaea 2 1? 2? 1? 2 1 (?) Araliaceae: Nothopanax 5 1 Schefflera 1 L.... 1 1 - . 164 MERRILL. Table I. — Summary of the families, genera, species, etc. — Continued. Families and genera. j Total species. I Indigenous. i Intro- duced. I Endemic. Ameri- can origin. Pantro- pic: Dis- tributed- To Malaya. To tropical Asia. To Australia. To Polynesia. | To Africa. I Spontaneous. I Cultivated only. ! Spontaneous. I 1 Cultivated only. By natural agen- j Through agency j of man. I Umbelliferab: 1 1 1 2 2 J 1? 1 J 1 1 Myrsinacbae: 1 Plumbaginaceae: 1 2 3 2 Sapotaceae; 1 2 1 1 1 1 1 1 J Ebenaceae: 2 1 1 2 1 Oleaceae: Toeminiim 4 1 3 1 2 3 1 2 Salvadoraceae: A '/imo 1 1 2 2 Loganiaceae: Buddleia 1 1 1 1 2 Gentianaceae: Canscora . 1 1 1 2 1 1 Apocynaceae: 1 Plumeria 1 1 2 1 1 1 1 1 2 Cerbera 1 1 1 1 Thevetia . . 1 1 1 1 1 1 2 Lochnera 1 1 1 1 1 1 J Tabernaemontana _ _ 4 3 1 3 1 1 1 Alston! a. 1 2 1 1 1 Allamanrlff 2 2 2 2 2 1 1 Wrightia 1 "Morin m 1 2 , 2 2 1 2 St.r opb p n t.b n s 1 1 2 ! Pprsonsia 1 1 1 1 1 1 1 Trbnorarpns 1 1 1 Asclepiadaceae; r!rvpt.nstog*ia 1 1 1 1 1 2 Streptocaulon 1 j A solopias 1 1 J 1 1 1 J J Tsohnnstomma 1 1 1 1 rjymnoma 1 1 1 1 Tylophora 1 1 1 , Conchophyllum 1 1 Telosma 2 1 I- 1 1 1 1 1 1 Hoya 1 1 1 Dischidia 1 ll. 1 1 . 1 THE FLORA OF MANILA. 165 Table I. — Summary of the families, genera, species, etc. — Continued. Intro- duced. Ameri- origin. Pantro- pic: Dis- tributed- Families and genera. Total species. Indigenous. Spontaneous. j Cultivated only. | Endemic. Spontaneous. | Cultivated only. | By natural agen- j des. 1 Through agency of man. ! To Malaya. To tropical Asia. 1 ! To Africa. Convolvulaceae: 1? 1? 1 1 1 1 1 2 2 2 2 2 2 2 Calonictyon Operculina 2 2 2 2 1 1 1 1 ■Riven 1 1 1 2 1 1? 1 1 g 5? 1? 6 6 3 2 Ipomoea Hydrophyllaceae: Hydrolea Boraginaceae: Cordia 10 1 4 4 2 L-... 2 2 2? 4 10 9 1 6 1 5 1 4 1 Ehretia 2 2 1 1 Tournefortia 1 1 ______ Heliotropium Trichodesma 1 1 1 1 1 2 2 1 2 Coldenia 1 1 J ■Verbenaceae: Avicennia 1 1 ! 1 Lantana 1 1 j 1 Tiippia 1 j ^ 1 j j Stachytarpheta Duranta 1 j 1 1 1 1 1 1 1 Petraea 1 1 j 1 1 1 1 Callicarpa Clerodendron 2 2 1 7 4 3 2 1 5 4 1 1 Vitex 3 3 3 2 1 PremTia 4 4 3 Gmelina 1 1 1 Labiatae: Rosmarinus 1 Ocimum 2 2 2 2 1 2 2 Moschosma . 1 1 1 4 4 4 4 ) 1 Coleus 2 1 1 1 1 Pogostemon Mentha 1 1 1 1 1 Anisomeles 1 1 Leucas 2 2 2 1 1 Leonurus 1 1 1 1 1 1 166 MERRILL. Table I. — Summary of the families, genera, species, etc. — Continued. Families and genera. Solanacbae: Datura Brunfelsia Ceatrum Nicotiana Solanum Physalis Lycopersicum ' Capsicum i Scrophulariaceae: i Scoparia Centranthera Russelia Bacopa '• Lindenbergia Limnophila ^ Mazus ! Vandellia Torenia i Bonnaya ! Dopatrium j Bignoniaceae; j Jacaranda Oroxylura ^ Tecoma i Dolichandrone ] Spathodea - , Kigelia . I Crescentia,- _ I Pedaliaceae; ! Sesamum ! Gesneriaceae: Rhynchoglossum I Lentibulariaceae; I Utricularia ' Acanthacbae: ^ Dicliptera Elytraria I Justicia I Barleria j Odontonema Pseuderanthemum _ Rhinacanthus Graptophyllum Thunbergia Ameri- can origin. Pantro- pic: Dis- tributed- I 1 To tropical Asia. THE FLORA OF MANILA. 167 Table I. — Summary of the families, genera, species, etc. — Continued. Intro- duced. Ameri- can origin. 1 Pantro- pic: Dis- j tributed-j Families and grenera. Total species. Indigenous. Spontaneous. Cultivated only. Endemic. Spontaneous. 'c 1 3 o By natural agen- cies. Through agency of man. To Malaya. To tropical Asia. j To Australia. To Polynesia. To Africa. Acanthaceae— Contd. 1 1 2 2 1 1 5 1 2 4 1 2 1 1 2 i 1 i 2 1 1 2 2 1 2 1 2 1 J 1 ! 1 1 1 1 1 ^ ! 1 ! 1 1 1 1 1 1 2 1 1 1 1 1 Plantaginaceae: . i i 1 1 i i| 4 1 ■ 1 ! 1 Rubiaceae: 1 6 Mitragyna ' Dentella 1 i 2 1 -1 Oldenlandia i 1 1 Mussaenda ) Gardenia 1 ..... J ....... i Randia J 1 1 ' Hypobatherum * Coffea J 1 1 3 1 2 3 2 1 ' Ixora 1 2 2 1 Pavpf.ta 1 2 1 ' Morinda i 2 ... .I P.qvrhotria 1 1 1 Pleotrnnia 1 2 1 1 1 1 2 1 2 2 1 2 2 2 1 1 1 1 ■ Paaderia 1 - .i 1 Spermacoce 1 Cucurbitaceae: 1 Melothria 1 ! Cucurbita 1 1? i 1 i 1 1 1 2 1 1 1 j ! Trichosanthes 2 1 1 Lagrenaria 1 1 2 1 1 1 1 2 ‘ 1 2 1 1 1 1 ......j 1 2 I ! Luffa } Benincasa I Citrullus - t Cucumis j Momordica Campanulaceae: Sphenoclea 1 ! Isotoma r-.. ! ‘ 1 Goodenoviaceae: Calogyne 1 1 1 1 Stylidiaceae; Stylidium i I i...... 168 MERRILL. Table I. — Summary of the families, genera, species, etc. — Continued. Families and genera. 1 Total species. 1 Indigenous. I Intro- duced. Endemic. Ameri- can origin. Pan tro- pic: Dis- tributed- j To Australia. To Polynesia. | < H Spontaneous. Cultivated only. Spontaneous. Cultivated only. By natural agen- cies. j Through agency of man. 1 To Malaya. To tropical Asia. Compositae; 1 1 2 2 2 2 J J J 1 1 1 J 1 1 1 Emilia 1 1 1 1 Elephantopus 3 3 3 3 1 1 2 1 I Sphaeranthus . 1 1 1 1 1 Blumea 4 2 2 3 1 1 1 Pluchea . 1 1 1 1 Grangrea 1 1 1 1 1 Crossostephium 1 1 Synedrella 1 1 1 1 1 1 1 1 1 Artemisia __ _ 1 1 1 1 1 1 Epaltes 1 1 1 1 Centipeda j 1 Erigeron 1 1 J 1 J Tagetes 1 1 1 1 1 1 Chrysanthemum 3 3 3 2 2 Enhydra 1 1 1 Eclipta 2 2 2 \ J Helianthus 1 1 Wedelia 2 2 J J Cosmos j 1 \ 1 Bidens 1 1 1 j 1 1 1 Lactuca 1 1 1 j \ 1 J 1 Sonchus 1 j 1 1 1 1 1 1 J Totals 1,007 550 232 225 124 83 95 89 ! 223 812 789 1 425 355 402 The introduced element in a flora is not frequently treated as such, although the subject is a fascinating one in many respects. Investigation of the subject involves not only an extensive knowledge of the flora of the region under discussion, but also detailed information as to the nature of the plants, their relative abundance or rarity, their habitats, their native names, special methods by which their seeds may have been disseminated, ancient and modern trade routes, and other factors. In the case of a species suspected of having been introduced it is desir- able that its original home be determined as nearly as possible. This involves a study not only of the plant itself, but of its allied forms and their recorded ranges, and especially a study THE FLORA OF MANILA. 169 of botanical literature, and floras of different regions. A knowl- edge of the properties of the various species is essential, for very many plants, some of which in the case of the Philippines are so thoroughly naturalized as to appear truly indigenous, have purposely been introduced for deflnite reasons, for food, for medicinal purposes, for dyes, for fibers, and for ornamental uses. In many cases it can be determined with a reasonable degree of certainty whether a certain species has been purposely or inadvertently introduced. Finally there must be taken into consideration the communication with various other countries both in ancient and in modern times, the origins of peoples in archipelagos like the Philippines, and their intercourse with other peoples. Manifestly the aboriginal inhabitants of the Philippines must have reached the Archipelago from some other country, and it is certain that in the thousands of years that have elapsed since man first reached the Islands, very many plants have been introduced, some purposely, some inadvertently, by the early invaders and their later successors from Malaya. If we exclude the abaca plant {Musa textilis Nee) and the various trees yielding timbers, gums, and resins, a few palms, some bamboos, the rattans, etc., it will be found that practically all the species now found in the Archipelago that are of the greatest impor- tance in the economy of the native, whether for food, for con- diments, for clothing, for dyes, for ornamental purposes, and very many for medicinal purposes, have originated outside of the Philippines, and have purposely been introduced at one time or another. Not a single important food plant or fruit tree has originated in the Archipelago, but all have been intro- duced. While prehistoric trade routes by which plants have been brought to the Philippines are not definitely known, still from the geographic location of the Archipelago, it would seem that communication must have been mostly with other Islands of the Malay Archipelago, and to a lesser degree with the Asiatic continent. A great many plants of economic importance were introduced in prehistoric times, and certainly at the same time a considerable number of weeds of cultivation. It is possible to determine, with at least a reasonable degree of certainty, most of the introduced species that were established or cultivated in the Archipelago before the advent of the Europeans. With the advent of the Europeans and the consequent in- creased communication between the Philippines and neighboring 170 MERRILL. countries, and especially with tropical America, a great impetus was given to the dissemination of plants, and the flora of this Archipelago, like that of many other regions has been greatly enriched by species introduced within historical times. While many of the introduced forms have proved to be of the greatest economic importance, still many undesirable citizens were repre- sented among the numerous aliens. With the advent of the Spaniards came a considerable number of European species, both weeds and plants of economic value, and very many forms from other parts of the Orient. A very large number were introduced from tropical America through the medium of the galleons on the old Acapulco-Manila trade route, which will later be given a more detailed consideration. The Philippine Archipelago, as geologists reckon time, is of comparatively recent formation, and the logical conclusion re- garding the origin of its flora is that the original vegetation of the group, like its human inhabitants, reached the Archi- pelago from older neighboring countries of the Indo-Malayan region. Much of this original vegetation may have reached the group through previously existing land connections, but ocean currents, winds, migratory birds, and lastly, and perhaps not least, man must be considered, for species have been intro- duced by all these factors, and all have hence had their influence on the present character of the rich and varied flora. It is not my intention at this time to discuss the probable origins of the flora as a whole, it being primarily the same as that of the Malay Archipelago. We have, however, in the high lands of northern Luzon a very pronounced continental, that is, Asiatic element, where the Himalayan types are more numerous than in any other part of the Malay Archipelago; there is also a decided Polynesian, and a decided Australian element in the flora, and I have previously advanced the theory^ that our dom- inant Malayan element has been introduced chiefly from south- eastern Malaya, rather than from western Malaya, that is, the Malay Peninsula and the Sunda Islands. Whatever may have been the geological history of the Archi- pelago, and whatever the source of its flora, it is manifest from a study of the present flora that the Islands as such have been separated from neighboring countries for a very long period of time, sufficient, in fact, to develop a very marked endemism. While the percentage of endemism as to genera is comparatively “Ann. Jard. Bot. Buitenz. Suppl. 3 (1909) 277-306. THE FLORA OF MANILA. 171 small, for species it averages approximately 40 per cent, and the indications are that as botanical exploration progresses this percentage will be increased rather than diminished. As noted above the average percentage of endemism for the entire Archipelago is about 40. In many families it runs to 90 per cent, or even higher, but in others it is much lower than the average. Wherever primeval conditions exist, that is, wher- ever the original forests have been undisturbed by man, the percentage of endemism is always high but wherever the coun- try is thickly settled, the destruction of the original forests and the large introduced element, cultivated plants, weeds, etc., cor- respondingly decrease the percentage of endemism, so that we find in Manila a percentage of endemism of only 8.7, or including cultivated endemic species, only 12 per cent. This great differ- ence in endemism emphasizes the radical differences between the vegetation of the settled areas and that of the forested regions, and is one strong reason for considering a high percentage of the species found in the settled areas as introduced. In support of the theory that very many of the species now common and widely distributed in the settled areas in the Phil- ippines have been introduced into the Archipelago since the advent of man, and in many cases have been spread in the Archipelago by his agency, the present vegetation of many of the smaller islands can be cited. On very many of such islands which are not inhabited, and only visited at irregular intervals by fishermen, or by collectors of turtles, turtle’s eggs, or bird’s eggs, it is exceedingly rare that one finds a single species that has presumably been introduced by man, although the conditions for their growth and reproduction are frequently present. I have personally visited a number of small and isolated islands in the Sulu Sea, and while many of these support a fairly rich flora, the species, without exception, are those that have been distributed by natural causes, that is, by water, winds, or mi- gratory birds. The characteristic species of the open country on the larger islands are practically wanting, although as noted above, favorable conditions for their growth on these small islands are sometimes present. However, as small islands in- variably support a flora poor in species as compared with larger land masses, the evidence from this source alone cannot be of great value. The case of the Island of Polillo is of interest in this connection, for it ranks about sixteenth in size among the islands of the 172 MERRILL. Archipelago, having an area of somewhat over 850 square kilo- meters. This island is situated off the eastern coast of Luzon, in about the same latitude as Manila, the channel separating it from Luzon being about 32 kilometers wide in its narrowest part; its highest point. Mount Malulud, is about 340 meters above sea level. This island was brought under control by the Spaniards at an early date, for the town of Polillo was fortified for protection against Moro raids in early colonial days. While much of its area is covered with virgin forest, still a strip along the coast has been cleared and settled, and the island supports a population of about 3,000 people. It was presumably inhabited before the Spanish conquest. In the botanical collections made on this Island by Doctor C. B. Robinson, and Mr. R. C. McGregor, in four months, August to November, 1909, about 515 species of flowering plants and vascular cryptogams are represented,* most of them of wide distribution in the Philippines and in the Indo-Malayan region generally; but 15 of them being known only from Polillo. In studying the Leguminosx of this collection my attention was called to the paucity of the unquestionably introduced forms in the Polillo flora,'^ and an examination of the entire list of species collected or observed shows that what was true of the Legumi- nosse applies to the other families as well. For instance, of the 178 species of plants of American origin that are now found in and about Manila, and have reached the Philippines since the advent of the Spaniards in 1521, but about 25 or 30 have reached Polillo, so far as our material shows. Of the 136 in- troduced species of presumably oriental origin, apparently not more than 35 or 40 are found on the island; and of the 109 species of cosmopolitan distribution, but whose original homes are uncertain, more than one-half of which in all probability have been distributed by man, only about 15 are found on Polillo. Now the conditions for the growth and reproduction of our numerous introduced species are excellent in Polillo, but from our present data regarding the flora of the island, it is estimated that less than 100 of the total 457 undoubtedly introduced species that are found in and about Manila, have as yet reached that island. This is due to the somewhat isolated position of the island, and to the comparatively slight communication between it and the more densely populated parts of Luzon. The very ‘Robinson, C. B. This Journal 6 (1911) Botany 185-228. ‘ This Journal 5 (1910) Botany 5. THE FLORA OF MANILA. 173 fact that so few species of American origin have reached Polillo in the past 375 years, leads us to infer that very many of the species of oriental or of unknown origin that for the most part are now cosmopolitan in the tropics, which like those of known American origin are comparatively few in Polillo, have reached the Philippines since the advent of the Europeans. Not only does the vegetation of the small and medium-sized islands bear out the theory that very many of our characteristic open country species are of recent introduction, but the condi- tions in some parts of the larger islands are practically the same. In the Province of Bataan, just across the bay from Manila, and only about 35 kilometers distant, a fairly intensive study has been made of a strip of land extending from the seashore to the summit of Mount Mariveles, attaining an altitude of about 1,400 meters.® While most of the area is covered with primeval forest, still a strip along the coast, several kilometers in width, consists of open lands, low thickets, etc., with climatic conditions closely approxi- mating those in Manila. This coastal strip has at one time or another been entirely deforested, but is now, except for small cultivated areas and open grass lands, covered with second growth forest, thickets, etc. This region yielded 1,114 species of flower- ing plants and vascular cryptogams, and of this number only 90 are supposed to have been introduced by man.’^ Practically all of these introduced forms are found only in the coastal region, in the open lands and thickets, almost none of them being found in the forests. In contrast to these figures, Manila, with its vegeta- tive conditions entirely changed by the destruction of the original vegetation, presents a total of 1,007 species, including the culti- vated plants, of which 457 have presumably been introduced by man, either purposely or accidentally. The percentage of endemism for the Lamao flora is about 40, the general average for that of the Archipelago as a whole, while for Manila it is only about 12. If we exclude 36 endemic species which are included in the “Flora of Manila,” but which are not true natives of the area covered, then the endemism for Manila is only about 8.7 per cent. Much has been published on the origin of cultivated species, and the subject has been admirably summarized by A. DeCandolle in his work on the “Origin of Cultivated Plants.” While this * Merrill, E. D. The Flora of the Lamao Forest Reserve. This Journal 1 (1906) Suppl. 1-141. ’ The original estimate was fifty-four, Merrill 1. c. 9. 111022 — 4 174 MERRILL. phase of the subject has been given considerable attention, com- paratively little has been written regarding the origin of various pantropic weedy plants that unquestionably have been distributed by man. Manifestly the question is a very complicated one, and in the case of very many species it is exceedingly doubtful if definite proof as to whether certain forms have originated in the Tropics of the Eastern or the Western Hemisphere can be com- piled. In many cases we can at most indicate that a certain species is a native of tropical America, and another a native of tropical Asia. To illustrate some of the reasons for considering certain species as having been introduced, and as natives of the Eastern or of the Western Hemisphere, it is well, perhaps, to cite and discuss some specific cases, for in a limited space it is impossible to give for each species the various reasons for considering it as having been introduced, and the reasons why, with a reasonable degree of certainty, it may be assumed to be a native of this or that region. The evidence from botanical literature is important, but is by no means always to be taken as definite proof that a species is a native of a certain country. It has not infrequently happened that species that are natives of tropical America have first been described in botanical literature from material collected in the Eastern Hemisphere, this simply showing that such species were introduced into India and other countries at an early date in colonial history. For example, Waltheria indica L. { = W. ameri- cana L.) was first described, in part, from specimens collected in Ceylon by Hermann before the year 1679. The species is of world-wide distribution in tropical countries, but in looking up the generic range it is found that there are approximately 30 species in the genus, all confined to tropical America, with the exception of W. americana L. (W. indica L.). This species in the Philippines, and in other countries in the Orient, occurs as a weed in the settled areas, and is a manifestly introduced one. All evidence points to its American origin, and it can safely be assumed to be a native of some part of tropical America. It was probably among the first of the numerous species inadvertently introduced into the Orient by the early European explorers or colonists. Sometimes it has happened that a species originating in Amer- ica has been described from specimens grown in Asia, but with indications as to its origin. For instance Roxburgh® published • Plants of the Coast of Coromandel 1 (1795) 67, t. 99. THE FLORA OF MANILA. 175 the first description of Pithecolobium dulce Benth. (as Mimosa dulcis) , from specimens grown in India, but definitely states that it was introduced from the Philippine Islands, his inference being that it was a native of this Archipelago. However, the species is of wide distribution in tropical America, and although no definite record as to the date of its introduction into the Philip- pines has been found, still we know from its usual habitat here and from its native names that it is not an indigenous plant. It is locally kown as camanchile, camonsil, camatchile, and variations of these names, manifestly corruptions of its old Mexican name, quamochitl, the name having been introduced by the Spaniards with the plant. The genus Rotala of the Lythraceae contains about 38 species, and has representatives in most warm countries, some confined to one hemisphere, some to the other. Rotala imnosior (L.) Koehne extends from the northern United States southward through Mexico and Central America to Brazil, and is also of wide distribution in the Philippines, occurring here as a rice- paddy weed. It is unknown from any other part of the Orient. It is certainly not an indigenous species with us, and undoubtedly was accidentally introduced at an early date through the medium of the Acapulco-Manila galleons, and finding favorable local con- ditions, became established. Elephantopiis, a genus of the Compositae, has about 16 species, mostly confined to tropical America, but we have in the Philip- pines three manifestly introduced ones. One of these, E. scaber L., is now distributed to most tropical countries, but E. spicatus Aubl. and E. mollis HBK. are unknown outside of tropical Amer- ica, except in the Philippines and Guam; in the Philippines all three are common and widely distributed. Unquestionably they have all been introduced from tropical America, and through the medium of the Acapulco-Manila galleons. Acacia farnesiana Willd. is a species of very wide distribution in tropical and subtropical countries, and most authors are con- tent with indicating its range as cosmopolitan in the tropics without attempting to determine its origin. In some countries it is thoroughly naturalized and is very abundant, as in some parts of the Philippines, while in others it is cultivated only. It was originally described from American material and there is every reason to believe that it is of American origin, and that it has been purposely distributed for the sake of its fragrant flowers. In the Philippines it is now never cultivated, and is so very 176 MERRILL. common in open places, thickets, in and about towns, and on dry- hillsides, etc., that the casual observer would certainly consider it to be an indigenous species. It is, however, primarilly a species of the open country, and one that cannot exist in forests, unless possibly such as are composed of small and scrubby trees. Hence it would be excluded from the Philippines as an indigenous species, assuming the theory that the country was originally covered with forest. The genus is of world-wide distribution, so that there is no evidence to be secured from the generic distribu- tion, except that its closest allies are all American and African ; of the 14 given by Bentham in the series Gummifei'eae, 10 being American, including the cosmopolitan A. farnesiana Willd., and 4 African. Bentham ® considered it probable that it was indig- enous in western America from Chili northward to Texas, also in tropical Australia, and possibly in South Africa. As to the Philippines, it is certainly introduced, and, so far as this Archi- pelago is concerned, is undoubtedly of American origin, for it is universally known among the natives by its Spanish name aroma, no other name being recorded for it in the Philippines, and no variations of its Spanish name. It was undoubtedly introduced by the Spaniards from Mexico at an early date, and presumably is a native of that country. Cassia alata L., given by Baker'® as “Cosmopolitan in the Tropics,” with no indication as to its probable origin; other authors more definitely state it to be a native of tropical America. From its habitat in the Philippines it is manifestly not an indig- enous species, although it is widely distributed here, sponta- neous, and thoroughly naturalized. While it has several native names that are purely Malayan in origin, still it is not uncom- monly known here as acapulco, or capurco, a name at once point- ing to its origin so far as the Philippines are concerned ; that is, it was brought to the Philippines in early colonial days on the galleons sailing from Acapulco. It must be considered a pur- posely introduced plant on account of its medicinal properties. All evidence points to its American origin and it can be safely assumed to have originated there. The genus Hyptis has about 350 known species, practically all of which are confined to tropical America. We have in the Philippines as common and widely distributed plants of the settled regions four species, H. caqntata Jacq., H. brevipes Poir., 'Trans. Linn. Soc. 30 (1875) 502. “Hook. f. FI. Brit. Ind. 2 (1878) 264. THE FLORA OF MANILA. 177 H. suaveolens Poir., and H. svicigera Lam., all of wide American distribution. Of these four species two, H. suaveolens Poir. and H. brevi'pes Poir., are now of wide distribution in the tropics of both hemispheres, but H. spicigera Lam. is known in the Eastern Hemisphere only from the Philippines and Guam, and H. capitata Jacq. only from the Philippines, Guam, and Java, although in the latter place it presumably is of comparatively recent introduction, as it is apparently not of wide distribution there, although com- mon in places.^ In confirmation of the American origin of these four species, if such were necessary, all are known from Guam, a small island in the Marianne group, at which the Spanish galleons stopped en route from Acapulco to Manila ; unquestion- ably all reached the Philippines through this trade route. None of them, although all are common in the Plulippmes, have any well-defined native names. Gomphrena globosa L. is a commonly cultivated plant in the Philippines, and is not infrequently found subspontaneous about dwellings, etc. It is by some authors given as of American origin, and by others Asiatic. It is certainly not a native of the Philip- pines, and it is equally certain that it was purpo.sely introduced by the Spaniards, for it is usually known only by names of Spanish origin, or corruptions of such names, boton (button), botoncillo, etc. In looking up the generic range it is found that of the 90 known species, nearly all are American and confined to tropical America, hence it is only reasonable to suppose that Gomphi'etia globosa L. originated in the same part of the world. In Table II, following, is given a summary of the species of pantropic distribution that are found in and about Manila, a total of 425, distributed into 296 genera and 86 families. Of these at least 334 owe their present wide distribution directly or indi- rectly to man, and have, for the most part at least, been distrib- uted from the Eastern Hemisphere to the Western or vice versa within historical times, that is, since the first circumnavigation of the globe. About 90 are assumed to have been of natural distrib- ution, but in at least one-half of the cases so considered, there is some doubt as to whether they may not have been distributed by man. In very many of the cases man has undoubtedly been instrumental in their dissemination whether or not he was instru- mental in distributing them from one hemisphere to the other. Purposely introduced species, including those of prehistoric dis- tribution, aggregate 242, and inadvertently distributed ones, 92. ” Backer in Ann. Jard. Bot. Buitenz. Suppl. 3 (1909) 404. 178 MERRILL. Again, so far as the Philippines are concerned, 197 species are found only in cultivation, while 227 are spontaneous and more or less thoroughly naturalized. As to their origins, 177 have, for most part at least, originated in tropical America, and 138 in the tropical or subtropical parts of the Old World. In 109 cases it has been impossible to determine, with any degree of certainty, where they may have originated. In my consideration of the Manila flora I have assumed that about 550 species, considerably more than one-half of the total, are indigenous, that is, true natives of the Philippines, or plants that have reached the Archipelago by natural causes and without the direct aid of man. It must be admitted, however, if we assume as correct the theory that the original vegetation of the Archipelago was a practically unbroken forest, that very many of these species must have reached the Philippines since the advent of man. In contrast to the 550 species that are presum- ably indigenous, about 457 have certainly been introduced by man, either purposely or accidentally, some in prehistoric times, and others at more recent dates. A very considerable percentage of these introduced species must have I'eached the Archipelago in recent times, that is, since the Spanish conquest or after the year 1521. Of the manifestly introduced species, about 225 are cul- tivated only, or at least spontaneous to but a small degree, and could not persist without the aid of man. Taking into consideration only those species included in my “Flora of Manila” I am of the opinion that my estimate of 457 species introduced by man directly or indirectly, in the thousands of years that have elapsed since the group was first inhabited, is a very conservative one. For many centuries before the Euro- peans reached the East the commerce between the Philippines and surrounding countries was by no means inconsiderable. The early Spaniards found here Chinese, Japanese, and Indian mer- chants and traders, and there must from very early times have existed intercommunication between the Philippines and Malaya. Since the Spanish conquest, that is, in the 890 years that have elapsed since the Spaniards first came to the Archipelago, no less than 178 species, manifestly of American origin, and over one- third of my estimated total of introduced forms, have been intro- duced into the Philippines, purposely or inadvertently, by man. More than 100 of these were brought to the Philippines between the years 1521 and 1815, for at the latter date regular communi- cation between the Philippines and the west coast of Mexico ceased. As the communication between the Philippines and THE FLORA OF MANILA. 179 Mexico was chiefly by annual galleons, it is certainly safe to assume that there was more communication between the Philip- pines and the neighboring countries in the Orient than between the Philippines and America, so that if in that time more than 100 species were introduced from America, we might reasonably expect that as many more of Oriental origin would be introduced during the same period. In this connection it is curious to note that many of our common weeds, both of American and of Oriental origin, have no standard names in the local dialects, but that many of them are known by Spanish names, or by corrup- tions of Spanish names. The great majority of the species found in and about Manila are the characteristic ones of the open country, weeds of culti- vation, etc., that are dominant in the settled areas at low alti- tudes throughout the Archipelago, few being rare or of local occurrence. Very many of these plants, considerably over one- half, must have originated in treeless countries, or at least in regions where only scattered trees existed, for they cannot persist in a thickly forested region. Many have doubtless originated in regions that have been more or less under cultivation from very ancient times, and have thus become adapted to the environment of cultivation. Table II. — Summary of species represented in the Manila flora that are of pantropic distribution, ivith indications of their condition of growth here, and their origin. Families, genera, and species. Natural distri- bution. Purposely intro- duced. Accidentally in- ' troduced. ! Cultivated. | Spontaneous. American origin. Oriental or Euro- i pean origin. j Origin doubtful or unknown. Polypodiaceae: Dryopteris parasitica 0. Ktze. X X X Nephrolepis cordifolia Presl X X X MprvhmlprjiQ flnppicrprn IWnorp X X X Nephrolepis hirsutula Presl X X X Ceropteris calomelanos Und. X X X Adiantum macrophyllum Sw X X X Adiantum tenerum Sw X X X Adiantum trapeziforme L X X X Adiantum philippense L. X X X 1 Adiantum caudatum L. X X X Acrostichum aureum L. X X X Pteris longifolia L. X X X Pteris quadriaurita L X X X Pteris biaurita L. _ X X X Parkeriacbae: Ceratopteris thalictroides Brongn X X X 180 MERRILL. Table II. — Summary of species, etc. — Continued. 1 Families, genera, and species. Natural distri- bution. Purposely intro- duced. Accidentally in- troduced. Cultivated. 1 Spontaneous. American origin. Oriental or Euro- ' pean origin. Typhaceae: ! X X Potamogetonaceae: X X X X Hydrocharitaceae: X X Gramineae: X X X X X X X X X X X X? X X X X X X X X X Andropogon sorghum L. X X X Eriochloa ramosa OK. X? X Paspalum scrobiculatum L. X? X Paspalum distichum L, X X Paspalum conjugatum Berg. X X X Digitaria sanguinalis Scop. X? X Digitaria violascens Link.. X? X Panicum colonum L. X? X Panicum crus-galli L. X? X Panicum amplexicaule Rudge X X Panicum maximum Jacq. X X X Panicum repens L. . X X Oplismenus compositus Beauv. X? X Cenchrus echinatus L. X X X Setaria viridis Beauv. X X X Rptnrip flpvpi TCimth X? X Oryza sativa L . . X X X Leersia hexandra Sw. _ X X Sporobolus indicus R. Br X? X Sporobolus virginicus Kth. X Phragmites vulgaris Trin. X X Eragrostis interrupta Doell. . X? X Eragrostis unioloides Nees . X X X Eragrostis pilosa Beauv. ... X? X Cynodon dactylon Pers. X? X Chloris barbata Sw. X X X Eleusine indica Gaertn. X X X Dactyloctenium aegyptium Willd. X X X Ramhusa vnlp’arin Rehrad. X? X Cyperaceae: Kvllinva brevifolia Rottb. X? X Kvllinc’a monocephala Rottb. X? X Pycreus odoratus Urb. X X X X X X X X X X X X X X xxixxxxxixxxiixixixii X XX X ^ doubtful THE FLORA OF MANILA. 181 Table II. — Summary of species, etc. — Continued. Families, genera, and species. Cyperaceab— Continued. Juncellus pygmaeus Clarke Cyperus difformis L Cyperus haspan L. Cyperus flabelliformis Rottb Cyperus diffusus Vahl Cyperus compressus L Cyperus distans L Cyperus rotundus L Cyperus radiatus Vahl Mariscus flabelliformis HBK Torulinium ferax Ham Eleocharis eapitata R. Br Fimbristylis diphylla Vahl Fimbristylis ferruginea Vahl Fimbristylis barbata Benth Fimbristylis miliacea Vahl Fimbristylis monostachya Hassk. Scirpus supinus L. Scirpus erectus Poir Scleria lithosperma Sw Palmae: Cocos nucifera L. Elaeis guineensis Jacq Oreodoxa regia HBK. Areca catechu L. Araceae; Pistia stratiotes L. Acorus calamus L Caladium bicolor Vent Colocasia esculentum Schott Lemnaceae: Lemna paucicostata Hegelm. Spirodela polyrrhiza Schl Bromeliaceae: Ananas sativus Schultes Commelinaceae: Commelina nudiflora L Rhoeo discolor Hance Zebrina pendula Schnizl Liliacbae; Asparagus sprengeri Regel. Asparagus officinalis L. Asparagus plumosus Baker Allium cepa L. Allium porrum L. Sansevieria zeylanica Yucca aloifolia L Cordyline terminalis Kth i \ 2 Origin doubtful 182 MERRILL. Table II. — Su7timary of species, etc. — Continued. Families, genera, and species. [ Natural distri- bution. Purposely intro- duced. Accidentally in- troduced. j Cultivated. Amaryllidaceae; X X X X X X X X X X Iridaceae: X X X X Musaceae: X? X X Zingiberaceae: X X X X X X Cannaceae: Canna indica L. __ .. X Canna edulis Sims . . _ X X Canna coccinea Mill. _ _ X X Canna warscewiczii Dietr. X X Canna g:lauca L. X X Canna flaccida X X Marantaceae: Maranta arundinacea L X X Piperaceae: Peppromi;! pelliirida HRtC X Peperomia argyreia Morr. - . .. X X Piper betle L X X Piper retrofractum Vahl X X Moraceae: Ficus elastica L. X X Artocarpus communis Forst. X X Castilloa ellastica Cerv. X X Morus alba L>. X X Urticaceae: Boehmeria nivea Gaudich. . X X Pilea microphylla _ X Aristolochiaceae^ Aristolochia elegans Mast. X X Polygonaceae: Antigonon leptopus H. & A. X X Muehelenbeckia platyclada Meissn X X Chenopodiaceae: r’K^^nnpnHinm T. X Amaranthaceae; Celosia argentea L. X Gomphrena globosa L. X X Alternanthera sessilis R. Br X 8 c 5 c X X X X X X X Origin doubtful THE FLORA OF MANILA. 183 Table II. — Summary of species, etc. — Continued. Families, genera and species. ! Natural distri- j bution. i Purposely intro- [ duced. j Accidentally in- 1 troduced. Cultivated. Spontaneous. American origin, j Amaranthaceae— Continued. X X X X X X X X X X X X X X X X X X Nyctaginaceae: X X X X X X X X X X X? Aizoaceae: X X Triant.hema portnla.caRt.nim Ti X X Mollugo lotoides 0. Ktze. X X Mollugo oppositifolia L. _ , X X Portulacaceae: Portulaca pilosa L. X X X Portulaca oleracea L .. X X Basellaceae: Basella rubra Ti. X X Anredera scandens Moq. X X X Ceratophyllaceae: Ceratophyllum demersum L. X X Menispermaceae: Cissampelos pareira L X? X Magnoliaceae: Michelia cbamnaca Ti. X X Anonaceae: Anona muricata L. X X Anona reticulata L. _ X X X Anona squamosa L X X Lauraceae: Cinnamomum zpivlamVnm Ti. X X Persea americana Mill. X X X CasRvtba filiformia Ti. X X Papaveraceae: Argemone mexicana L. - _ . _ X X X Cruciferae: Brassica juncea L X X Ranbanns sativna Ti. X X Capparidaceae: Cleome viscosa L. X? X Gynandropsis pentaphylla DC. X? X Moringaceae: Moringa oleifera Lam. X X 1 i X X X X X X X X X X X? X X X X I j Origin doubtful 184 MERRILL. Table II. — Summary of species, etc. — Continued. Families, g-enera, and species. Natural distri- bution. Purposely intro- duced. Accidentally in- troduced. Cultivated. Spontaneous. American origin. Oriental or Euro- pean origin. Origin doubtful | or unknown. j Crassulaceae: X X X Leguminosae; X X X X X X X X X X X X X X X X X X X? X X X X X X X X X X X X X X X X X X X X Cassia sia.mf»a Tiam. X X X Cassia tora L. X X X Cassia occidentalis L. X X X X X X Cassia alata L, X X X Tamarindus indica L. - - X X X Arachis hypogaea L. X X X Zornia diphylla Pers. X X X? DesTTiodium scorpiurus Desv. X X * X Desmodium procumbens Hitchc. X X X Desmodium triflorum DC. . X? X X Desmodium lasiocarpum DC. X X X Alyscicarpus vaginalis DC. X X X Indigofera hirsuta L. X X X Indigofera suffruticosa Mill. X X X Indigofera tinctoria L. X X X Gliricidia sepium Steud. X X X Clitorea ternatea L. X X X? Pnmcpln crla-nrlnloRa Mprr. X X X Pienm Qativnm T, i X X X Abrus precatorius L. X X X SpshaTiia prandifloi’a. Pers. X X X Crntnlaria T.. X X X Crotalaria incana L. 1 X X X Flemingia strobilifera R. Br. 1 X X X dent.irnlata Willd. 1 X X X Cajanus cajan Merr. - X X X X X >< r’.r.o,,,!.-, r>n ^ X X X Canavalia ensiformis DC. -- X X X? X X X Centrosema plumieri Benth. X X X Phaseolus lunatus L — . X X X 1 THE FLORA OF MANILA. 185 Table II. — Summary of species, etc. — Continued. Families, genera, and species. Leguminosae— Continued. Phaseolus semiereetus L Phaseolus vulgaris DC Phaseolus adenanthus Mey Phaseolus radiatus L. Vigna sesquipedalis L Vigna lutea A. Gray Pachyrrhizus erosus Urb. Dolichos lablab L Psophocarpus tetragonolobus DC. Oxalidaceae: Oxalis repens Thunb. Oxalis martiana Zucc. Biophytum sensitivum DC Averrhoa carair.bola L. Averrhoa bilimbi L. Erythroxylacbae: Erythroxylum coca L Zygophyllaceae; Tribulus cistoides L. Rutaceae: Murraya exotica L Triphasia trifoliata DC. Citrus lima Lun. Citrus decumana Murr Simarubaceae: Quassia amara L. Meliaceae: Aglaia odorata Lour. Melia azedarach L. Malpighiaceae; Galphimia glauca Cav Euphorbiacbae: Euphorbia splendens Boj. Euphorbia heterophylla L Euphorbia pulcherrima Willd. Euphorbia hypericifolia L Euphorbia hirta L. Euphorbia thymifolia Burm. Euphorbia prostrata Ait Phyllanthus urinaria L Phyllanthus niruri L Jatropha podagrica Hook Jatropha multifida L Jatropha curcas L. Jatropha gossypifolia L Aleurites moluccana Willd Codiaeum variegatum Blume Acalypha hispida Burm XXjXl IXXjlllXX I American origin. 186 MERRILL. Table II. — Summary of species, etc. — Continued. Families, genera, and species. Natural distri- bution. Purposely intro- duced. Accidentally in- troduced. Cultivated. Spontaneous. American origin. ' Oriental or Euro- 1 pean origin. j Origin doubtful | or unknown. j Euphorbiaceae— Continued. X X X X X X X X X X X X X X X Anacardiaceae; X X X X X X X X X Sapindaceae: X X X? Balsaminaceae: X X X Tiliaceae: X X X Triumfetta bartramia L. X X X Triumfetta semitriloba Jacq. X X X Coreborns ransnlaris Tj X X X CorchnriiR olit.orins Ti X X X Corchorus acutangrulus Lam. X X X 1 Malvaceae: I Mfllvastriim ri'tromanHplirmm X X X ^ SiHa c.ordifolia Tj X X X j Sida acuta Burm. . _ X X X Rida rhomhifolia Tj X X X Rida rf't.iifia Tj. X X X Ahiit.ilon indienm Don X X X Malaehra canitata T». X X X X X X Tirana lobata Tj. X X X Malaviaoiia pilosna DC. X X X X? X X X Abelmoschus esculentus Moench X X 1 T X X X X X 1 X W,h;Q...,» T X X X H;n, a;r,onQ,'« T X X X X X X X X X r^ooypLiTYi Ki.OfliBoT.ao Moof X X X Bombacaceae: Coibn pontoTidrQ Caortn X X X? Stbrculicaeae: X X X X X X X X X? Bixaceae: Bixa Orellana L X X X THE FLORA OF MANILA, 187 Table II. — Summary of species, etc. — Continued. r i Families, genera, and species. Natural distri- bution. Purposely intro- duced. Accidentally in- troduced. j Cultivated. I Violaceae: X X Passiploraceae: X X X X X Caricaceae: X X Cactaceae: X X X X X X Lythraceae: X X X X Rotala mexicana C. &S. - X? Rotala ramosior Koehne __ X Punicaceae: Punica granatum L. X X Myrtaceae: Eucalyptus tereticornis Sm - X X Psirliiim o'uaiava Ti. X Oenotheraceae: 1 Jussiaea repens Ti. X 1 Jussiaea suffrut.icosa li X Jussiaea linifolia Vahl. _ X Araliaceae: Nothopanax cochleatum Miq. X X Nothopanax fruticosum Miq. X X Umbelliferae: Foeniculum vulgare Gaertn. X X Hydrocotyle rotundifolia Roxb. X Centella asiatica Urb. _ X? X Sapotaceae: Achras sapota L X X Ebenaceae: Diospyros ebenaster Retz X X Oleaceae: Jasminum sambac Ait. X X Apocynaceae: Plumeria acutifolia Poir. _ ... X X Thevetia neriifolia Juss. X X Lochnera rosea Reichb X Tabernaemontana divaricata R. Br X X Allamanda cathartica L X X Allamanda hendersonii Bull X X Nerium indieum Mill X X Origin doubtful 188 MERRILL. Table II. — Summary of species, etc. — Continued. Families, genera, and species. Natural distri- bution. Purposely intro- duced. Accidentally in- troduced. Cultivated. Spontaneous. American origin. Oriental or Euro- | pean origin. i Origin doubtful or unknown. Asclepiadaceae: X X X X X X Convolvulaceae: X? X X X X X X X X X X X X X X X X X X X X X X X X X X ^ Capsicum anuum L. - - X X X \ Physalis ang'ulata L. X X X 1 Physalis minima L. _ ' X X X j SCROPHULARIACEAE: 1 Scoparia dulcis L. X X X i Russclia juncea Zucc. _ X X X ' Bacopa monniera Wettst. X X X X X X. j Vandellia Crustacea Benth. j Bignoniaceae: ' Jacaranda ovalifolia R. Br. X X X 1 Tecoma stans Juss. X X X Crescentia cajute L. X X X Crescentia alata HBK. X X X Pedaliaceae: Sesamum orientale L. X X X Acanthaceae: Elytraria tridentata Vahl X X X Barleria lupulina Lindl. X X X Odontonema nitidum 0. Ktze. X X X Thunbergria alata Boj. X X X Thunbergia grrandiflora Roxb. X X X Graptophyllum pictum Griff. X X X Blechnum brownei Juss. X X X Plantaginaceae: Plantago major L. . X X X Rubiaceae: Coffea arabica L, . X X X Oldenlandia corymbosa L. X X X? Ixora finlaysoniana Wall X X X Gardenia florida L X X X THE FLORA OF MANILA. 189 Table II. — Summary of species, etc. — Continued. Families, genera, and species. Natural distribu- tion. Purposely intro- duced. Accidentally in- troduce. Cultivated. Spontaneous. American origin. Oriental or Euro- pean origin. Origin doubtful or unknown. Cucurbitaceae: X X X X X X X X X X X X X X X X X X X X X X X X? X X X? X X X X X X X X X? X X X X X Ipomoea triloba L. • X X Ipomoea batatas Poir. X X X Boraginaceae: Heliotropium indicum L. X X X? Verbenaceae; Lantana camara L. X X X Lippia nodi flora L. X X X Stachytarpheta jamaicensis Vahl X X X Duranta repens L. . _ __ __ X X X Petraea volubilis JacQ. X X X Clerodendron thompsonae Half. X X X Labiatae: Hyptis capitata Jacq. X X X Hyptis brevipes Poir. X X X Hyptis suaveolens Poir. X X X Hyptis spicigera Lam. __ X X X Coleus blumei Benth X X X Leonurus sibiricus L. X X X Mentha arvensis L, X X X Solanaceae; Solanum nigrum L. X X X Solanum melongena L. X X X? X X X? Solanum wendlandii Hook. f. X X X X X X CitrnlliiQ viiltrariq Srhrad X X X Cucumis melo L. X X X Cnonmis sat.ivna Ti. X X - X Momordica rharant.ia Tj. X X X Campanulaceae: Sphenoclea zeylanica Gaertn. X X X? Isotoma longiflora Presl. X X X 111022 5 190 MERRILL. Table II. — Summary of species, etc. — Continued. Families, genera, and species. Natural distribu- tion. Purposely intro- duced. I Accidentally in- troduced. Cultivated. j Spontaneous. American origin. 1 Oriental or Euro- | 1 pean origin. | Origin doubtful or unknown. COMPOSITAE: Eupatorium triplinerve Vahl. _ X X X Ageratum conyzoides L. X X X Emilia sonchifolia DC. X X X? Elephantopus scaber L. . X X X Elephantopus mollis HBK. - _ X X X Elephantopus spicatus Aubl. __ _ ^ X X X SynedrellanodifloraGaertn. X X X Artemisia vulgaris L. X X X Rrityeron linifnlins Willd X X X Taffetea eroota Ti. X X X Rr»lirifn aIHs) X X X Helianthna annnna T.. X X X PlnQTTiAQ pniiHatna TTPtTT X X X Bidens pilosa L. _ X X X Lactuca sativa L. _ X X X Sonchus oleraceus L. -- X X X Chrysanthemum indicum L. X X X Totals 90 242 92 197 227 177 138 109 Very many of our species are to be found in practically all tropical countries in both hemispheres; in fact, of the 1007 included in the Manila flora, about 425 are now found in tropical America and in the tropics of the Orient, although not all of these are of general distribution. Some of these are manifestly of natural distribution, but the greater number have unquestion- ably been distributed by man, either purposely or accidentally, and within the past 390 years. I consider it very probable that not more than 92 of the 425 pantropic species under considera- tion are of natural distribution, that is, those species that have extended to the tropics of both hemispheres, and sometimes also to more temperate regions, without the direct or indirect aid of man. In at least one-half of these 92 cases which I have considered as presumably of natural distribution, there is a reasonable doubt as to whether they may not have been distri- buted from one hemisphere to the other by man. Among the species that are unquestionably of natural distri- bution in the tropics of both hemispheres may be mentioned the following: Dryopteris parasitica 0. Ktze., Nephrolepis cordi- folia Presl, N. hirsutnla Presl, Adiantum philippense L., A. cauda- tum L., Pteris longifolia L., P. quadriaurita Retz., P. biaurita L., THE FLORA OF MANILA. 191 Acrostichum aureum L., Ceratopteris thalictroides Brongn., Val- lisneria spiralis L., Ruppia maritima L., Typha angustifolia L., Imperata cylindrica Beauv., and doubtless other grasses and sedges, Pistia stratiotes L., Lemna paucicostata Hegelm., Spiro- dela polyrrhiza SchL, Sesiivium portulacastrum L., Portulaca oleracea L., Ceratophyllum demersum L., Cassytha filiformis L., Caesalpinia nuga L., C. crista L., Canavalia lineata L., Vigna lu- tea A. Gray, Jussiaea suffruticosa L., J. repens L., Hisbiscus tiliaceus L., Bacopa moyiniera Wettst., and Ipomoea pes-caprae Roth. It will be noted that a considerable number of the forms enu- merated above are aquatic species or those that grow in swampy places, having minute seeds that might readily be transported by adhering in mud to the feet or feathers of migratory wading or swimming birds. Such plants are Ceratopteris, Acrostichum, Lemma, Spirodela, Pistia, Ceratophyllum, Vallisneria, Bacopa, Jussiaea, etc., although the wind may have been the agency of dispersal in some cases. Others are characteristic species of the strand, such as Hisbiscus tiliaceus L., Canavalia lineata DC., Cassytha filiformis L., Caesalpinia nuga Ait., C. crista L., Vigna lutea A. Gray, and Ipomoea pes-caprae Roth, and are undoubtedly ocean-distributed plants. While it is comparatively easy to explain the probable methods by which species like the above have been distributed, it is, excluding the theory of their distribution by man, impossible to explain the wide distribution of many grasses and sedges and other weedy plants which in most tropical countries are found in waste places in and about towns, in fallow lands, and as weeds in cultivated fields. While many of these might be distributed by winds, by water, or by their comparatively small seeds adhering in mud to the feet or feathers of migratory birds, still it is more probable that most of them have been disseminated by man, for most part inadvertently, in packing material, in earth trans- ported for one purpose or another, in ballast, and by other means. From the habitats and distribution of such species as the follow- ing, it is only reasonable to suppose that most or all of them have been distributed by man, although their original homes are unknown or uncertain; wherever else they may have originated they are certainly not natives of the Philippines: Various grasses and sedges, Commelina nud.iflora L., Alternanthera sessilis R. Br., Amaranthus spinosus L., A. viridis L., Trianthema portulacas- trum L., Mollugo lotoides O. Ktze., M. oppositifolia L., Cissam- pelos pareira L.. Cleome viscosa L., Gynandropsis pentaphylla 192 MERRILL. DC., Desmodium trifloruvi DC., CrotalaHa verrucosa L., Teram- nus labialis Spr., Oxalis repens Thunb., Biophytum sensitivum DC., Euphorbia hirta L., E. thymifolia Burm., E. hypericifolia L., Phyllanthus niruri L., P. urinaria L., Sida cordifolia L., S. acuta Burm., S. rhombifolia L., S. retusa L., Melochia corchorifolia L., Centella asiatica Urb., Evolvulus alsinoides L., Sokmum nigrum L., Emilia sonchifolia DC., Eclipta alba Hassk., and Bidens pilosa L. In most treatments of tropical floras, whether of the eastern or of the western hemisphere, such species as the above are usually listed and discussed as native ones. It is far more prob- able that some have originated in one hemisphere and some in another, and that, for the most part at least, they have been accidentally distributed by man within the past 400 years. Cer- tainly many of them were widely disseminated at an early date in colonial history, and in most cases it is probable that their original homes will never definitely be known. Purposely introduced species comprise those of various other tropical countries that, for reason of their economic impor- tance, have been introduced either in prehistoric or within his- toric times. Naturally the first plants introduced were those of the Malayan region that were familiar to the original in- vaders or their successors in western Malaya, and these include such species as Job’s tears {Coix lachryma-jobi L.), the more common form of use for ornamental purposes, another form cultivated for food; sugar cane {Saccharum officinarum L.), as a source of sugar; lemon grass (Andropogon citratus DC.), used as a condiment; vetiver {Andropogon zizanioides Urb.), for its aromatic root; sorghum {Andropogon sorghum L.), for food; Italian millet {Setaria italica Beauv.), for food; rice {Oryza saliva L.) , for food ; bamboos {Bambusa vulgaris Schrad., and B. blumeana Schultes), for purposes of construction; coco- nut {Cocos nucifera L.^D» foi" food; betel-nut palm {Areca catechu L.), for its stimulating properties; sweet flag {Acorus calamus L.), medicinal; taro {Colocasia esculentum Schott), food; yam, “ubi” {Dioscorea alata L.), for food; garlic {Allium sativum L.), for food; various varieties of the banana {Musa paradisiaca L.), for food; various zingiberaceous plants {Kaemp- feria galanga L., Curcuma zedoaria L., C. longa L., Zingiber zerumbet Sm., and Z. officinale Rose.), for condiments, etc.; betel-pepper {Piper betel L.) for use with the betel-nut for ” This species is unquestionably of American origin, but reached the Orient long before the advent of Europeans. THE FLORA OF MANILA. 193 chewing; bread fruit (Artocarpus communis Forst.), and the jak fruit (A. integrifolia L. f,), for food; amaranths (Ama- ranthus gangeticus L., A. caudatus L.), for ornamental purposes and food; “libato” (Basella rubra L.), for food; champaca (Mi- chelia champaca for its fragrant flowers; siempre viva {Bryophyllum pin7iatum Kurz), for medical purposes; horse radish tree (Moringa oleifera Lam.), for food and medicine; sappan {Caesalpinia sappan L.), for dyeing; the tamarind {Ta- marmd'us indica L.), for food; indigo {Indigofera tmctoria L.), for dye; “caturay” {Sesbania gra7idiflora Pers.), for its edible flowers and its resinous exudation; the pigeon pea {Cajanus cajan Merr.), for food; the cow pea (Vigna sesquipedalis L.), for food; the asparagus pea {Psopho carpus tetragonolobus DC.), for food; “batao” (Dolichos lablab L.), for food; the mungo {Phaseolus radiatus L.), for food; various citrus fruits, such as the pomelo {Citrus decumana Murr.), the lime (C. lima Liman.) , and varieties of the orange (C. am-antium L.) , for food ; the santol (Sa^idoricum koetjape Merr.), for food; the lansone {Lansi'um domesticum Jack), for food; some euphorbias (E. tirucalli L.), for medicine; “iba” {Cicca disticha L.), for food; crotons {Codiaeum variegatum Blume), for ornamental pur- poses; castor oil plant (Ricmus communis L.), for medicine; croton oil plant {Croton tiglium L.) , for medicine and for poison- ing fish; balsam {Impatiens balsamina L.), for medicine and for ornamental purposes; cotton {Gossypium sp.), for textile purposes; silk cotton tree {Ceiba pentandra Gaertn.) for its fibrous floss; various Eugenias {E. jambolana Lam., E. malac- censis L., E. jambos L., and E. javanica L.), for food; “papua” {Nothopanax fi-uticosum Miq.), for medicine and for ornamental purposes; jasmine {Sasmmum sambac Ait.), for its fragrant flowers; “solasi” {Ocimum basilicum L., and O. sanchim L.), for condiments; sesame {Sesamum orientate L.) for its oily seed; the bottle gourd {Lagenaria leucantha Rusby), for food; the sponge gourd {Luija cylindrica Roem. and L. acutangula Roxb.), for food; the “condol” {Benincasa hispida Cogn.), for food; and the “ampalaya” {Momordica charantia L.) for food. From an examination of the above list it will readily be seen that a great number of species were purposely introduced in prehistoric times from various parts of the East, chiefly through Malaya, for one reason or another, usually for their food value or for other reasons of economic importance. It is quite certain that none of the species enumerated above are natives of the This may have been introduced later by the Spaniards. 194 MERRILL. Philippines, and it is equally certain that none of them reached the Archipelago without the aid of man. Again it is equally certain that, with possibly very few exceptions, all these species were introduced by the early Malay invaders, by their succes- sors, or by peoples of various other nationalities with whom they came in contact, long before the advent of the Europeans in the Orient. In addition to the species above mentioned, for the most part undoubtedly of prehistoric introduction, there is still another source to be considered, and that is those species introduced by the Spaniards from Europe, or from various neighboring coun- tries in the East. A brief summary of such plants follows: Elaeis guineensis Jacq., the oil palm ; Asparagus spp. ; Sanser- vier’ia zeylanica Willd., for its fiber and for ornamental pur- poses; Cordyline terminalis Kunth, for ornamental purposes, although this may have been of prehistoric introduction; the onion {Allium cepa L.), for food; Crinum zeylanicum L. and C. latifolium L., for ornamental purposes; Pancration zeylani- cum L., for ornamental purposes; Belamcanda chinensis Lem., for ornamental purposes; Ravenala madagascariensis Sonn., for ornamental purposes; Ficus elastica L., a shade tree; Morus alba L., for feeding silk-worms; Muehelenbeckia platyclada Meissn., for ornamental purposes; Boehmeria nivea Gaudich, for its fiber; Artabotrys uncinatus Merr., for its fragrant flow- ers ; Unona chinensis DC., for its fragant flowers ; Cinnamomum zeylanicum L., one of the sources of cinnamon; Brassica juncea Goss., for food; Raphanus sativus L., for food; Capparis cordi- folia Lam., from the Marianne Islands, used as a condiment; Albizzia lebbeck Benth., a shade tree; Bauhinia tomentosa L., an ornamental shrub; Cassia siamea Lam., a shade tree; Lourea vespertilionis Desv., for ornamental purposes; Pisum sativum L., for food; Phaseolus vulgaris L., for food; Mangifera indica L., for its edible fruit; Aglaia odorata Lour., a shade tree with fragant flowers; Melia azedarach L., for ornamental purposes; Euphorbia splendens Boj., for ornamental purposes; various forms of Codiaeum variegatum Blume, for ornamental purposes ; Acapypha hispida Burm., and A. wilkesiana Muell-Arg., orna- mentals; Euphoria longana Lam., for its edible fruit; Zyzyphus jujuba Mill., for its edible fruit; Hibiscus sabdarijfa L. H. mu- tabilis L., H. rosa-sinensis L., H. schizopetalus Hook, f., all orna- mentals ; Viola odorata L., for its fragrant flowers ; Lagerstroe- mia indica L., an ornamental; Lawsonia inermis L., for its fragrant flowers ; Punica granatum L., for its edible fruit ; THE FLORA OF MANILA. 195 Eucalyptus tereticormis Sm., a shade tree; Fceniculum vulgare Gaertn., a condiment; Plumbago auriculata Lam., P. zeylanica L., and P. rosea L., ornamentals; Mimusops elengi L., for its fragrant flowers; Jasminum grandiflorum L., and J. multiflorum Roth, for their fragrant flowers ; Tabernaemontana divaricata R. Br., an ornamental shrub ; Nerium indicum Mill., for its flowers ; Cryptostegia grandiflora R. Br., for its flowers ; Telosma odoratis- sima Coville, for its fragrant flowers; Argyreia nervosa Eoj., Ipomoea cairica Sweet, vines ; Clerodendron thomsonae Balf ., C. fragrans Vent., and C. calamatosum L., ornamentals; Rosmarinus officinalis L., for medicinal purposes; Mentha arvensis L., mint; Spathodea campanulata Beauv., an ornamental tree ; Kigelia o.fri- cana Benth., a tree ; Graptophyllum pictum Griff., an ornamental shrub; Thunbergia glandiftora Roxb., T. alata Boj., and T. erecta And., ornamentals; Plantago major L., for medicinal purposes; Gardenia florida L., for its fragrant flowers; Coffea arabica L., the source of coffee; Ixoi'a finlaysoniana Wall., I. coccinea L., and /. chinensis Lam., ornamental shrubs; Cucumis sativus L., for its edible fruit; Artemisia vulgaris L., for medicinal purposes ; and Lactuca saliva L. for food. Comparatively few of the above species have become natural- ized in the Philippines, and for the most part their continued existence here is dependent directly on the presence of man. Sanseviera zeylanica Willd. is at least subspontaneous, and also, in some localities, Belamcanda chinensis Lam., and Muehelen- beckia platyclada Meissn. Brassica juncea Goss., is an occa- sional weed, and some species of Plumbago are naturalized, as well as Thunbergia alata Boj., Plantago major L., and Artemisia mdgaris L. More than 50 of the species above enumerated have almost certainly been introduced within the past fifty or sixty years, if one can judge by their non-inclusion in Blanco’s “Flora de Filipinas” (1845). Blanco included in his work chiefly cul- tivated plants and the species found in the settled areas, giving comparatively little attention to the forest flora ; and, from lack of material, entirely ignoring the rich mountain flora of the Archipelago. He seems, however, to have included most of the species found in and about towns, except the grasses and sedges, and especially the cultivated and economic plants. The species purposely introduced within historical times from various parts of the Old World aggregate a considerable num- ber, as will be noted from an examination of the above list. In addition to those enumerated above, there are still found in 196 MERRILL. the Archipelago a great number of plants of wide distribution in the Indo-Malayan region, which from their distribution and occurrence in the Philippines are certainly not indigenous spe- cies. Manifestly they must have originated in some other part of the Orient, and have been introduced into the Philippines, either within historic times, or at a still earlier date. It is only reasonable to assume that many of them have reached the Archi- pelago within the past 400 years. It is, however, practically impossible definitely to determine whether many of them have been introduced inadvertently by man, or by natural causes. Excluding the purposely introduced species enumerated above, and for the present ignoring those species of oriental origin having minute seeds, which in all probability have reached the Philippines through natural causes, there still remains a con- siderable list of weedy plants that presumably have been acci- dentally introduced, such as various grasses, Rottboellia exaltata L. f., Apluda nutica L., Andropogon aciculatus Retz., Panicum flavidum Retz., Panicum distachyum L., Centotheca latifolia Trin., Eragrostis spp., certain sedges, and representatives of other families, such as Fleurya interrupta Gaudich., Cyathula prostrata Bl., Aerua lanata Juss., Achyranthes aspera L., Por- tulaca quadrifida L., Nasturtium indicum DC., Aeschynomene indica L., Desmodium pulchellum Benth., D. . triquetrum DC., D. gangeticum DC., D. lasiocarpum DC., Sesbania cannbina Pers., Crotala7'ia quinquefolia L., C. linifolia L. f., Flemingia strobilife7'a R. Br., Pueraida phaseoloides Benth., Euphorbia spp., Phyllanthus spp., Sida spp., Abutilon indicum Don, Urena lobata L., Abelmoschus moschatus Moench, Merremia spp., Coldenia procumbens L., Moschosma tenuiflorum Heynh., Anisomeles indica O. Ktze., Leucas spp., Leonurus sibirica L., Datura alba Nees, Vernonia patula Merr., Emilia sonchifolia DC., Sphaei'an- thus africanus L., Grangea madei'aspatana Poir., Erigeron lini- folius Willd., Wedelia prosD'ata Hemsl,, Sonchus ole^’aceus L., and Crepis japonica Benth. While some of the above species may have been introduced by natural causes, by birds, or by winds, it is only reasonable to suspect that most of them, especially those species in genera having special adaptations for the dispersal of seeds by sticking to clothing, to the hair of animals, etc., have been introduced by man. In contrast to this, a large number of genera undoubt- edly owe their presence in the Archipelago to natural causes, their small seeds having been transported in one method or another by migratory birds, by winds, etc., as, for example, cer- THE FLORA OF MANILA. 197 tain grasses and sedges, especially the rice paddy forms, and such genera as Eriocaulon, Commelina, Aneilema, Cyanotis, Monochoria, Mollugo, Sesuvium, Trianthema, Polygala, Salo- mania, Rotala, Ammannia, Ludivigia, Jussiaea, Canscora, Hyd- rolea, Centranthera, Lindenbergia, Bacopa, Mazus, Vandellia, Torenia, Bonnaya, Dopatrium, Utricularia, Sphenoclea, Calo- gyne, Stylidium, and doubtless, so far as the flora of the whole Archipelago is concerned, many other genera. In this con- nection the distribution of two characteristic Australian types, both found near Manila, is of interest, Calogyne pilosa R. Br., of the Goodenoviaceae, and Stylidium alsinoides R. Br., of the Stylidiaceae. Both of these are rice paddy weeds with us, and there is every reason to believe that they have been present in the Philippines for a very long time. There appears to be no doubt whatever as to the Australian origin of both. The former species is also found in southeastern China, but the latter is unknown north of Australia except in Luzon. It is believed that the most reasonable explanation of their occurrence in such widely-separated regions is that they have been disseminated by their minute seeds adhering in mud to the feet or feathers of migratory wading birds, such as the snipe, and it is con- fidently expected that eventually both will be found in inter- mediate localities in Gilolo, Celebes, and New Guinea, when those islands are botanically better known. It is possible, however, that they may have existed in one or all of these islands in the past, and that they may have become exterminated by adverse conditions. It is certainly unreasonable to suppose that their seeds have been transported directly from tropical Australia to Luzon, a distance of approximately 4,000 kilometers. Having considered the species of presumably oriental and those of unknown origin and the agencies by which they have reached the Philippines, there still remains to be discussed the great number of forms that have originated in tropical Amer- ica, and which have been purposely or accidentally introduced here within the past 390 years; that is, since the discovery of the Archipelago by Magellan in 1521, and its subsequent con- quest by the Spaniards. To a greater or lesser extent in the early colonial days, the Philippines were a distributing center of American species to the neighboring countries, not only of those purposely distributed for their economic value, but also many weeds and other useless plants inadvertently disseminated. Previous to the arrival of the Spaniards in the Orient the Por- tuguese had doubtless introduced various American species into 198 MERRILL. their possessions in tropical Asia and Malaya, and later the Dutch continued the work, so that it is now quite impossible, in many cases, to determine by whom the various economic plants and many weeds were first brought to the East; doubt- less many were distributed by representatives of all three na- tionalities. The Spaniards and the Portuguese were, however, unquestionably the pioneers in this respect, and the interchange of species between the tropics of the two hemispheres initiated by these pioneer explorers and colonists still continues. 'The American element in the Philippine flora is of peculiar interest as showing the effect 'of commerce on the vegetation of a country. Even with the limited communication between the Philippines, and Mexico, it is surprising to consider the num- ber of American forms introduced here through the medium of the galleons in the years when all communication between Spain and the Philippines was via Mexico. From the time of the Spanish conquest up to the year 1815, a period of nearly 300 years, the government galleons sailed annually for Manila, first from Navidad and later from Acapulco, on the western coast of Mexico. These galleons carried not only the civil, military, and ecclesiastical authorities between Spain and the Philippines via Mexico, but also other travellers, merchants, etc., as well as large quantities of merchandise. At an early date various Spanish officials, but, apparently, chiefly the priests, introduced here the various species of eco- nomic value, food plants, medicinal plants, fruits, etc., that were familiar to their countrymen in tropical America, most often bringing seeds, but in some cases most certainly living plants. Undoubtedly many species were introduced at that time that failed to become established here. Among the American species purposely introduced from Mexico may be mentioned the follow- ing: Maize (Zea mays L.), introduced for food; the pineapple {Ananas sativus Schultes), for its edible fruit and its fiber; maguey {Agave cantula Roxb.), for its fiber; the tuberose {Po- lianthes tuberosa L.), for its fragrant flowers; the spider lily {Hymenocallis tenuiflora Herb.), for ornamental purposes; the canna {Canna indica L.), for ornamental purposes; arrowroot {Maranta arundinacea L.), for food; “aposotis” {Chenopodium ambrosioides L.), for medical purposes; four o’clock {Mirabilis jalapa L.), for ornamental purposes; “libato” {Anredera scan- dens Moq.) ; various species of Anona {A. muricata L., A. reti- culata L., and A. squamosa L.), for their edible fruits; the avo- cado {Persea americana Mill.), for its edible fruit; the Mexican THE FLORA OF MANILA. 199 poppy (Argemone mexicana L.), for ornamental purposes or for medicine; “camanchile” {Pithecolobium dulce Benth,), for its tanbark and its edible fruit; “aroma” {Acacia farnesiana Willd.) for its fragrant flowers; “ipel-ipel” {Leucaena glauca Benth.), as a hedge plant; the sensitive plant {Mimosa vudica L.), for ornamental purposes; “cabellero” {Caesalpinia pulcherrima Sw.), for ornamental purposes; “acapulco” {Cassia alata L.), for medicinal purposes; the peanut {Arachis hypogaea L.), for food; indigo {Indigofera suffruticosa Mill.), for dye; “madre cacao” {Gliricidia sepium Steud.), for hedges and for ornamental purposes ; the lima bean {Phaseohis lunatus L.) , for food ; the yam bean {Pachyrrhizus erosus Urb.), for its edible root; the bilimbi and carambola {Averrhoa hilimbi h., and A. carambola L.), for their edible fruits; physic nut {Jatropha curcas L.), for medi- cine, as well as J. multifida L. for ornamental purposes ; cassava {ManiJwt utilissima Pohl), for food; the cashew nut {Anacar- dium occidentale L.) for its edible fruit; “ciruelas” {Spondias purpurea L.) for its edible fruit; cotton {Gossypium braziliense Macf.), for textile purposes; cacao {Theobroma cacao L.), the source of chocolate and cacao; acheute {Bixa orellana L.), for dye; the papaya {Carica papaya L.), for its edible fruit; various species of cacti {Nopalea and Cereus), for ornamental purposes; the guava {Psidium guajava L.) , for its edible fruit; the “chico” {Achras sapota L.), for its edible fruit; the “chico mamey” {Lucuma mamviosa L.), for its edible fruit; the “sapote negro” {Diospyros ebenaster Retz.), for its edible fruit; the temple flower {Plumeria acutifolia Poir.), for its fragrant flowers; the periwinkle {Lochnera rosea Reichb.), for its ornamental flowers; “campanello” {Thevetia neriifolia Juss. and Allamanda cathar- tica L.), for ornamental and medicinal purposes; some species of convolvuli {Ipomoea nil Roth, Quamoclit pinnata Boj., Colo- nictyon aculeatum House), for ornamental purposes, and the sweet potato {Ipomoea batatas Poir.), for food; lantana {Lan- tana camara L.), for ornamental purposes; “dama de noche” {Cestrum nocturnum L.), for its fragrant flowers; tobacco {Ni- cotiana tabacum L.) ; the tomato {Lycopersicum esculentum Mill.) ; the peppers {Capsicum frutescens L. and C. annuum L.), for condiments and for food; some ornamental Bignoniaceae {Crescentia alata HBK., and Tecoma stans Juss.); (?) the squash {Curcubita maxima Duch.), for food; some ornamental Compositae {Tagetes erecta L., Helianthus annuus L., Cosmos caudatus HBK., and C sulphureus Cav.), and “ayapana” {Eupa- torium triplinerve Vahl), for medicine. 200 MERRILL. The majority of the above species are found only in cultivation, but some, such as Leucaena glauca Benth., Mimosa pudica L., Argemone mexicana L., Lantana camara L., Cosmos caudatus HBK., Acacia farnesiana Willd., Gliricidia sepium Steud., Psid- ium guajava L,, Jatropha cureas L., Pithecolobium dulce Benth., Ipomoea nil Roth, and Pachyrrhizus erosus Urb., are so thor- oughly naturalized and so common and widely distributed in the Philippines that the casual observer would consider them to be indigenous species. In addition to the species above enumerated that were orig- inally purposely introduced and which have become thoroughly naturalized, we must now consider a very large number of trop- ical weeds that have originated in America, which were inad- vertently introduced here and which have now extended their range, certainly through the agency of man, to most tropical countries. Among these American aliens may be mentioned the following: Paspalum conjugatum Berg., Cenchrus echinatus L., Chloris barbata Sw., Peperomia pellucida HBK., Pilea micro- phylla Leibm., Celosia argentea L. (possibly introduced for or- namental purposes), Alternanthera frutescens R. Br., Cassia tora L., C. occidentalis L., C. hirsuta L., Zornia diphylla Pers., Des- modium scorpiurus Desv., D. procumbens Hitchc., Parosela glan- dulosa Merr., Crotalaria incana L., Phaseolus semierectus L., Euphorbia prostrata Ait., Malvastrum coromandelinum Garcke, Malachra capitata L., M. fasciata Jacq., Triumfetta bartramia L,, (T. rhomboidea Jacq.), T. semitriloba Jacq., Waltheria ameri- cana L., Passiflora foetida L., Rotala ramsoir Koehne, Jussiaea linifolia Vahl, Asclepias curassavica L., Ipomoea triloba L., Lip- pia nodiflora Rich., Stachytarpheta jamaicensis Vahl, Hyptis suaveolens Poir., H. brevipcs Poir., H. capitata Jacq., H. spicigera Lam., Physalis angulata L., P. mmima L., Scoparia dulcis L., Elytraria tridentata Vahl, Blechum broivnei Juss., Age7'atum conyzoides L., Elephantopus scaber L., E. mollis HBK., E. spi- catus Aubl., and Synedrella nodiflora Gaertn. Nearly all the species in this last list of American weeds are very common and widely distributed in the Philippines. Most of them are confined to the settled areas, and are so thoroughly naturalized that they might well be considered indigenous species if nothing were known of their origin. Many are individually numerous, that is more dominant, than are native species, or those that have been introduced from the tropics of the Orient. While most of these American weeds are now found in most tropical countries in the East having proper climatic conditions THE FLORA OF MANILA. 201 for their reproduction and growth, it is interesting to note that a number of forms are found in none of the other countries of the Eastern Hemisphere, while others are just commencing to estab- lish themselves in neighboring regions. Many of these American aliens in the Philippines that have not reached neighboring coun- tries are among our most common and widely distributed species, while others are of local occurrence only. In connection with this matter the following species may be mentioned; Alternan- thera frutescens R. Br., common in waste places ; Portulaca pilosa L., occasional in dry localities; Anredera scandens Moq., very abundant in hedges and thickets along streets; Pithecolobium didce Benth., very common in the Philippines, and thoroughly naturalized, thriving especially on gravel bars along many streams, now also scattered in Java and in India; Prosopis vidal- iana Naves, described from Philippine material, and by some authors reduced to P. juliflora Sw., certainly of Mexican origin, and now very abundant about the shores of Manila Bay; Desmo- dium scorpiuris Desf., abundant here, also in Polynesia; Gliri- cidia sepium Steud., very abundant in the Philippines, and now just beginning to appear in neighboring countries in the Orient; Parosela (Dalea) glandidosa Merr., very abundant on dry grassy slopes and in waste places, first described by Blanco; Schrankia quadrivalvis (L.) Merr., rare, or at least very local; Muntingia calabura L., also in Siam, introduced there from Manila; Mal- achra fasciata Jacq., abundant in waste places, now occasional in Java, but still rare there ; Ipomoea trUoba L., very common in the Philippines and in Guam, now making its appearance in Singa- pore, Borneo, Java, and Mauritius; Hyptis capitata Jacq. and H. spicigera Lam., both also in Guam, the former also in Java, but still rare there; Blechum brownei Juss., abundant in Guam and in the Philippines, now also in Formosa; Elytraria tridentata Vahl, abundant locally; Elephantopus mollis HBK., also in Guam, and E. spicatus Pers., among our most common and widely distributed weeds ; Rotala ramosoir Koehne, now a common rice paddy weed with us ; Cenchrus echinatus L., very common, now also occasional in neighboring countries, Borneo, etc., and Eleutherine americana Merr., occasional in some parts of the Philippines, locally abundant. The presence of these American weeds in the Philippines is directly due to the old Acapulco-Manila trade route, and indicates the necessity of taking into consideration the communication between different countries in working out problems regarding the introduced element in any particular region. 202 MERRILL. The direct introduction of various species from tropical Amer- ica apparently ceased with the withdrawal of the galleons be- tween Manila and Acapulco, but at a later date other plants of American origin, mostly ornamentals, have come in indirectly, chiefly from neighboring countries in the Orient, Among the comparatively recently introduced species of American origin may be mentioned Ceropteris calomelanos Und., Adiantum macrophyl- lum Sw., A. tenerum Sw. and A. trapeziforme L., Panicum maximum Jacq,, Oixodoxa regia HBK., Cladium bicolor Vent., Rhoeo discolor Hance, Zebrina pendula Schinzl., Yucca aloifolia L., Eucharis grandiflora Planch., Ganna spp,, Peperomia argyreia Morr., Castilloa elastica Carv., Aristolochia elegans Mast., Anti- gonon leptopus H. & A., Alternanthera versicolor Regel, Bougain- villea spectabilis Willd., Enterolobium saman Prain, Bauhinia monandra Kurz, Centrosema plumieri Benth., Oxalis martiana Zucc., Euphorbia heterophylla L., E. pulcherrima Willd,, Jatropha podagrica Hook, f., J. gossypifolia L., Hevea brasiliensis HBK., Manihot glaziovii Muell.-Arg., Malvaviscus pilosus DC., Abelmos- chus esculentus Moench., Gossypium hirsutum L., Passiflom quadrangularis L., P. edulis Sims, Allamanda hendersonii Bull., Ipomoea purpurea Lam., Operculma tuberosa Choisy, Duranta repens L., Petraea volubilis Jacq., Brunfelsia americana L., B. acuminata Benth., Solanum wendlandii Hook, f., Russelia juncea Zucc., Jacaranda ovalifolia R. Br., Crescentia cujete L., Odonto- nema nitidum 0. Ktze., Isotoma longiflora Presl as well as various ©rnamental Compositae, such as Dahlia, Corepsis, Cosmos, Zinnia, Gaillardia, etc. Very few of the above species are found outside of cultivation, but some are now becoming established, such as Euphorbia hetey^ophylla L., Enterolobium saman Benth,, Rus- selia juncea Zucc,, Ceropteris calomelanos Und., Cladium bicolor Vent., and Zebrina pendula Schinzl. If we can trust to the evidence afforded by local botanical literature, all of the above species have been introduced within the last 70 years, or since the publication of the second edition of Blanco’s “Flora de Filipinas” in 1845. None of them are included in his work, although about one-half are mentioned by F. Villar and Naves in the “Novissima Appendix” to the third edition of the “Flora de Filipinas,” 1880-83, as having been seen by them in the Archipelago. Taking into consideration only those species found in and about Manila, we have seen that about 178 or over one-sixth of the total number have originated in tropical America, some of which have been introduced here purposely for one reason or THE FLORA OF MANILA. 203 another, and others inadvertently. In other parts of the Archi- pelago additional species of American origin occur which are not now found in or near Manila, while in other neighboring countries of the Indo-Malayan region there are a certain number of other species, also of American origin, which have not as yet reached the Philippines. In the entire Indo-Malayan region it is probable that there are to-day, either in general cultivation or as weeds, more than 200 species that have originated in tropical America and which have found their way into the Eastern Hemisphere since the advent of the Europeans, that is, within the past 400 years. In the interchange of species between the Old World and the New, those originating in tropical America and now occurring in the Indo-Malayan region are apparently much more numerous than those of oriental origin which have been introduced into tropical America. This seems to apply not only to the ubiquitous weeds of cultivation, but also the economic plants, especially those that are cultivated for food. Without access to extensive and representative collections of tropical American plants, it is impossible properly to deal with the subject of introduced plants in the countries extending from Mexico southward to Brazil, either of oriental of other origin. That many species from the Indo-Malayan region have been introduced into tropical America is a fact, but it is impossible to form a clear conception of the number, and especially of their abundance, without some personal knowledge of the floras of the latter region. Species that are merely cultivated are as a general rule ignored by botanical collectors, and for this or for other reasons are frequently excluded from botanical works treating of restricted areas. From an examination of the liter- ature available here I have been able to compile a list of only about 80 species of Indo-Malayan origin that have purposely been introduced into tropical America, and which are for the most part only cultivated there, although naturally some have become established where favorable conditions exist. As to how general these species are in the great area included in tropical America, I have no information, but many are doubtless of very local occur-, rence. As to weeds of cultivation I have found records of only about twenty species that are to a greater or less degree distrib- uted in tropical America, and which have undoubtedly originated in the Old World, and are of inadvertent introduction into the New. It is apparent that by far the greater part of our ubiquitous 204 MERRILL. cosmopolitan tropical weeds have originated in tropical America, and at the same time that region seems to lead in the number of plants of economic value, especially those cultivated for food. These facts may indicate a very ancient civilization in some parts of tropical America, by which the various weeds have become adapted to the environment of cultivation, and by which many of the economic plants have become domesticated. It is more probable, however, that at least the weed-flora has been devel- oped in a region with thin forests, or with very scattered trees, or perhaps in nonforested areas. Such types of vegetation as the above have probably existed in some parts of tropical Amer- ica, especially in the drier regions, from very ancient times, but in the eastern tropics with their high and continuous humidity, such as is found in the entire Malay Archipelago, and in a large part of tropical Asia, extensive areas of open country could hardly exist under primeval conditions. In tropical countries like the Philippines in respect to anthesis there is a constant succession during all months of the year, and hence, in order thoroughly to investigate the flora of any region, explorations and observations must be continuous throughout the year. Very many species flower only during the dry season, some at its beginning, others as it progresses, and still others toward its end, while very many others flower only during some part of the rainy season. While most of the species found in and about Manila, and in other parts of the Archipelago as well, have well defined and restricted periods of anthesis, a comparatively large number flower throughout the year, some continuously, and some at intervals of greater or less duration. With the exception of those species that flower continuously, the time of anthesis is found to be closely correlated with the distribution of the rainfall. In the Philippines where the time and duration of the rainy season is very largely modified by local conditions, the influence of the monsoons, geological and geographical features, etc., it is a well-known fact that the period of greatest precipitation in one part of the Archipelago may be and often is in the period of greatest drought in other parts. Generally speaking the western part of the Archipelago is dry while the regions bordering on the Pacific coast are wet, and vice versa, but the local distribution of mountain masses frequently modifies this general rule, and in the vicinity of some mountains the precipitation is often fairly continuous through- out the year. Manifestly then, observations as to the periods THE FLORA OF MANILA. 205 of anthesis of any species or group of species have only a local value. That some species growing in and about Manila flower during certain months is no indication that the same species will be found in flower in other parts of the Archipelago during the same period of time. The dates of flowering for individual species are approximately the same only in those regions where the dry season and the rainy season cover similar periods. Most of our indigenous species have well-defined and restricted periods of anthesis, especially the perennial herbs and the woody plants, sometimes lasting a few days or a few weeks, more rarely extending over several months in a homogeneous area. Judging from herbarium records alone it is not infrequent to find that certain woody species have been collected in flower in all months of the year, taking into consideration the material from all parts of the Archipelago. If, however, the material is considered by provinces it is almost always found that the specimens from one region all flower during a certain period, those from another region at another period, etc., more or less closely correlated with the distribution of the rainfall. Annual species, being more dependent on rainfall than the perennial ones, are for the most part found in flower only during the rainy season or shortly after its close, and on the whole their period of anthesis appears to average somewhat longer than for the perennial ones. Very many of the annuals disap- pear with the advance of the dry season, only to make their appearance as a new generation after the rains have commenced. Among the indigenous species the constituents of the strand flora and of the mangrove vegetation, including most of the perennials and many of the annuals, flower more or less freely throughout the year, in this respect being sharply contrasted with the inland flora. The explanation of this phenomenon is probably to be found in the soil moisture, which in the man- groves is practically the same at all seasons, and along sandy beaches and tidal streams more or less uniform throughout the year. An examination of the indigenous and introduced elements in relation to their periods of anthesis brings out a striking and rather interesting fact. Only about 8 per cent of the indigenous species in and about Manila flower all the year, while among the introduced ones the corresponding percentage is more than 70. Very many of the species making up the 8 per cent for the indigenous element will be found in the representatives of 111022 6 206 MERRILL. the strand flora and mangrove vegetation, mentioned above. Of the remaining 30 per cent of the introduced element, very- many of the species, although not flowering all the year, have decidedly prolonged periods of anthesis, such as Enterolobium saman Prain, 7 months, from March to September, Delonix regia Raf., 5 months, from March to July, etc. There are various reasons that will in part account for this enormous difference in continuous anthesis between the indi- genous and the introduced elements. For instance, many of the cultivated plants are comprised in the latter group, and these are always under more or less abnormal conditions, while very many of them have been greatly altered by the process of selec- tion in the course of domestication. In those species that are cultivated for their ornamental flowers, unquestionably the very fact of continuous anthesis has in many cases been one of the reasons for their original selection and domestication, their popularity as ornamentals, and hence their wide distribution. A considerable number of these plants locally never produce fruits, such as several species of Hibiscus, Ixora finlaysoniana Wall., I. chinensis Lam., 7. coccinea L., Odontonema nitidum 0. Ktz., Graytophyllam pictum Griff., Thunbergia grandiflora Roxb., T. erecta And., Ipomoea cairica Sweet, Barleria cristata L., Allamanda cathartica L., A. hendersonii Bull., and many others, and this very fact unquestionably has its influence on the question of continuous anthesis. The above factors, how- ever, can explain this striking difference in anthesis only in part. It is impracticable, if not quite impossible, to determine the nature of these plants in their native homes as to restricted or continuous anthesis. Doubtless some of them in their original habitats flower all the year, but it is most probable that the great majority of them were originally species with restricted anthesis, a character which has undergone extension or modi- fication due to changed environment and climatic conditions in their new homes. Whatever may have been the causes leading to the develop- ment of this character the fact remains that most of our ubiqui- tous tropical weeds owe their success largely to the fact that for twelve months in the year they are continuously developing and maturing seeds, giving such species an enormous advantage in the struggle for existence over those having a restricted period of anthesis. Naturally also this continuous seed produc- tion gives such plants a great advantage over native species. THE FLORA OF MANILA. 207 which lack this character, and helps to account for the wide and rapid distribution of tropical weeds. SUMMARY. 1. The original vegetation of the Philippine Archipelago was undoubtedly a continuous forest of one type or another. The present enormous deforested areas, cultivated and fallow lands, thickets, second growth forests, and open grass lands are directly or indirectly due to the presence of man. The original forests have been destroyed by crude methods of agriculture and recur- ring grass fires in abandoned clearings. 2. Most of the species found in the settled areas, that is, light- loving species adapted to growth in the open, and which cannot withstand forest conditions, must have been introduced into the Archipelago after the advent of man, and after the nature of the vegetation had been altered over greater or less areas. These species cannot persist and reproduce themselves in forests. 3. A high percentage of the species in the settled areas have been introduced into the Philippines some through natural agen- cies and very many purposely or inadvertently by man. Many of these species reached the Archipelago in prehistoric times, while many others have been introduced since the advent of the Europeans. 4. Practically all the useful plants in the Philippines, exclud- ing the forest trees, the rattans, some bamboos, a few palms, and the abaca plant {Musa textilis Nee) have been purposely intro- duced from other countries. These introduced economic plants include all the important species yielding food, the cereals, nearly all the fruit trees, condiments, many medicinal plants, and most of the commonly cultivated ornamentals. Many were introduced in prehistoric times, chiefly from the Malayan region, but a high percentage have been introduced within the past 400 years, including a great number of American origin. 5. The percentage of endemism in regions where the original forests have not been disturbed by man is over 40 ; in the settled areas and in those regions where the forests have been destroyed it is but about 12. 6. Of the 1,007 species found in the vicinity of Manila, a region where the original vegetation has been largely destroyed by man, about 550 are indigenous, and 475 introduced. Of the introduced species over one-half, or 232, are spontaneous; 225 are found in cultivation only, or but sparingly spontaneous. 7. As to the ranges of these 1,007 species, 812 extend to the 208 MERRILL. Malay Archipelago, 789 to tropical Asia, 425 to tropical Aus- tralia, 424 to tropical America, 402 to tropical Africa, and 355 to Polynesia. Only 124 are endemic, including the locally intro- duced and cultivated species. 8. Species that are cosmopolitan in the tropics, at least those that are found in some parts of both hemispheres, aggregate 424, of the total of 1,007 considered. About 90 of these are considered to have spread from one hemisphere to the other through natural causes, although in about one-half of these cases the species may have been transmitted by man. As to the Phil- ippines, 334 of these pantropic species have been introduced by one means or another, about 242 purposely, and about 92 inadvertently. 9. As to the origin of these pantropic species, 177 are consid- ered to have originated in tropical America, and 138 in the tropics of the Eastern Hemisphere. In the case of 109 species it has been impossible definitely to determine where they may have originated. 10. Pantropic weeds of American origin are more dominant and more numerous in species and in individuals in the Phil- ippines, than are those of oriental origin. This is probably due to fundamental differences in aboriginal floristic conditions. In some parts of tropical America, at least in the drier regions, open treeless or partly treeless regions have probably existed from the earliest times, thus permitting the development of a flora consisting of annual, or of small, perennial, light-loving plants. In the Philippines, the Malayan region generally, and perhaps in large parts of tropical continental Asia, the high and continuous humidity has tended to the development of con- tinuous forests, and hence, as to the herbs and small shrubs, the development of a shade-loving flora. The destruction of forests over large areas has provided the habitats for light- loving species, an element lacking or but poorly developed in the original flora, thus permitting the growth and rapid disse- mination of plants adapted to these conditions when once intro- duced. 11. Among the introduced element in and about Manila, more than 70 per cent of the species flower continuously, or practically continuously, during all months of the year. Of the indigenous species but about 8 per cent flower continuously, over 90 per cent having restricted periods of anthesis. Vol. VII, No. 2, including pages h? to 124-, was issued July 1, 1912. PTJBiaCATIONS FOE SALE BY THE BHEEAU OF SCIE5ICE, MANILA, PHIUPPmE ISLANDS THE SUBAXUNS OF SlNDAKGAJf BAY. , By Emebson B. Cheistie. 121 page», 1 tnapi 29 plates. Ordet No. 410. Paper, $1.25 United States currenoy, postpaid. Sindangan Bay is situated on the- nori^iern coast qf Zamboan^B Peninsula. The Subanuhs of this region were studied by- Mr. Christie during two perisids of five and six weeks',; respectively. The foliovvino is an abstract froin the contents of Mr. Christie’s report on the Subariuris: Habitat and history; relatibris with the Mofos; matefial culture; houses;, industries; trade; agrioulture; family life; soojat customsf administration of justice; religion; -the medicine- man; ceremonies; tales; word-lists; physical measiirdihehts. The 29 plates illustrate the Subanuns at work and at play; their industries, houses, altars, and implements; and the people themselves. THE HISTORY OF SUIiU. By Najeeb M. Saieeey. 275 pages, 4 maps', 2 diagrams; Order No. 406.- ' Paper, $0.75. United States currency, postpaid. In the preparation of . his menusoript for The History of Sulu CoOtor-Saieeby spent much time and effort in gaining access to documents in the possession of the Sultan of Sul p- It is fortunate that these records' have now been translated and preserved in permanent form. This book is a history of the Moros in the Philippines from the earliest times to the American occupation. STUDIES IN MORO HISTORY, UAW, AND RELIGION. . By Najbee M.,Saeeebt. • 107 pages, 16 plates, 5 diagrams. Order No. 405. Paper, $0.25; half morocco, $0.75 United States currency, postpaid. This volume- deals with the earliest written records of thev Moros in Mindanao. Doctor Saleeby was fortunately able to obtain exact oopiss; of carefully preserved early records written, in the Magindanao dialect with Arabic characters; The author presents translations of these- as: welt as 16 half-tone illustrations of certain pages from the originals;. The names of the rulers of Magindanao are recorded in five folding diagrams. NEGRITOS OF ZAMBALES. By WnxiAM Ai/tAN Eeetd. ' ' S3 pagesy 62 plates; Order No. 402. Paper, $0.25; half morocco, $0.75 United States currency, postpaid. The introductory chapter; deals with the general distribution of NegrUos and with the distribur tion jcf the Philippine branch of the race. The. succeeding chapters' deal with the various indus- tries, amusements,, and social relations of these little men. - ’ Plates from photographs, the greater part of which was taken for this publioaticn, show ornaments, houses, men making- fire with bamboo, bows and arrows, dances, end various types of the people themselves. CONTIiNIS AMES, OAKESr. Notes on Philippine XXrchids with Descriptions of New" Species, V: The ),/ ,V , tion D began, with Volume V) J.'— ‘ i4S*0»+ Entire Journal," Volume II, III, IV,’ or V ir.*’'.., 5.00 , Entire Journal, beginiung with Volume Vl , ^ i ^.00 ^ Single numbers ^ .v--------'--'---*-------------- ^ / ’> *50' Volume 1, 1906’ (not divided into sections) ■ 10.^00 i Supplement to Volume I (botany) v'-t - a.50'. * Voluoix^ 1 (tVithout supplement) ,..^—^,.4, v6;60' JBaph section is separately paged and indexed. ‘ t ' ^ Publications bent in exchange for the Philippifte Joitfn&l of Sdeiice ^ should be addressed:, Library, Bureau of Science, Manila, P: I. , ? ' Subscriptions may be sent to the Business MANAGERj Philippine jour- ■; ‘ nal of Science, Bui:ea,u of Science, Manila, P. I., or to any hf^dte- aunts' listed below:’ ' , ^ ■ ?'/■ ‘ 'i'Vf V V'r'-;-'’ ■' The Macmillan Company, 64-66 Eifth Avenue, Jfevjr. '^oVk City, p/s. A. • Wm. Wesley & Son, '28 Essex Street, Strgnd, Lobdon,.W, C., England.- Martinus Nijhoff,, Nobel^traat 18, The Hagne^ Holland. - Ms^er & Miiller,' Prinz Louis Perdlnandstrasse 3, ]^erlip» N.W.,* Germany* ’ Kelley & Walsh, Limited, ?2 Baffles Plae^ Singapore, Straita Settlemmats. • , At M. & J. Ferguson, 19 Baillie Street, Colombo, Ceylon. ‘ >, Thacker, Spink &',Co., P. O. Box 54, Calcutta, India. / ' ^ Eniered al thfi sb%Uo%se at Manila P.. I., ^5 secend-BlaUfV matter. )L. vni SEC., c, So. 4 I ; , TQE PHILIPPINE JOURNAL OF SCMCE ALVIN J, COX, M. A., Ph.D. , GENERAL EDITOR / ' SEctioN C. Botany E; D. MERRILL, M- S. ), EDITOR WITH THE C05peRATI0N OF p/w. ORAEF, B.S.; W. H. BROWN, Ph.D. a. N. WHITFORD, Ph.D. ' V ' MANILA , ’ BU^Ap OP PRINTING PUBLICATIONS FOR SALE BY THE BUREAU OF SCIENCE, MANILA, PHILIPPINE ISLANDS K^JPORT OP THE INTEBNATIOIfAIi PIiACiUE CONFERENCE. Held at Mukden, April, 1911, yiider the auspices of the Chinese Government. Edited by Emcn Martini, G. F. BferRiE, iUTHtm. Stant,et, and EioSabd ■ P. Strono. 4S3 page,, IS plate* (2 colored, 4 half-tones, 12 charts and map,). Order No. 416. Cloth, $3.50: paper, $2.50 Onlted States currency, Postpaid. The proceedings of this internationar Conference and. inforrhation’ plained therefrom, together with the results of certain bacteriological: Investigations, constitute the .present report. Nothing hitherto has been published which gives such a complete and pomprehenslve account of the entire subject of pneumonic piagup. . ^ ^ Delegate from America (United States of), Austria-Hungary, France, Germany, Great Britain, Italy, Japan, Nlexloo, the Netherlands, Russia, and China attended the Conference. The Bureau of Scierioe of the.Government of the l^hilipplne island* has been appointed sole agent for the distribution of the printed prooeedings of the Internallbnal Plague Conference. THE SUGAR INHUSTBY IN THE ISLAND OP NEGROS. By Herbert S. Walker. Order No. 412. 145 pages, 10 plates, 1 map. Paper, $1.25 United States currency, postpaid. Considered from the Viewpoint of practical utility, Mr. Walker’s Sugar Industry in the Island of Negros is one of the most imporiant papers published by the Bureau of SCIenoei This volume is a real contribution to thC subject; it is not a mere oompllatjpn, for the aulhoi- was in the field and understands the oonditibns . of Which he writes. The following, is a brief synopsis of the contents: l Tables of soil analyses, both chemical and physical; analyses of the cane, juice and bagasse; estimates based ort actual Iriformatiori as to the ooste of production and of cultivation; and esti- mates of the cost and location of possible central factories. T^je Island is considered by sugar- producing dislriots; the area of cultivation and the produption per hectare are given, and th* possibility for future expansion discussed. . « u . .i. The plates iiltistrale various phases of sugar industry from the ouUivalion of the field to the transportation of .sugar in native sailboats. > A MANUAL OP PHILIPPINE SILE CULTURE. By CKABLES S. Banks. ' 53 pages, 20 plates. Order No. 413. Paper, $0.75 United States currency, postpaid. The silk industry Is parlioularfy adapted to be undertaken by persons- with small capital, and like the making of hats in the Philippine Islands it should thrive with a little encouragement. in A Manual of . Philiopine; Silk Culture yie have presented the results ipf several years actual work With silk-producing lanrae together with a description, of the new Philippine race. Halt-tone plates Illustrate in natural size silkworms in different stages of development, pupse; adult mplh*, samples of cloth made from eri silk, hand reel, and silk house. Other plates illustrate the various appliances used in raising silkyvOrms and in spinning silk; hand and power reels are Illustrated; working drawings are given for a silk house and for a hand reel. THE PHILIPPINE Journal of Science C. Botany VOL. VII. SEPTEMBER, 1912 No. 4 THE RELATION OF RAFFLESIA MANILLANA TO ITS HOST. By William H. Brown. {From the Botanical Section of the Biological Laboratory, Bureau of Science, Manila, P. I.) The genus Rafflesia is particularly interesting on account of the large size of its flowers and the great reduction of its veg- etative parts. A flower of Rafflesia arnoldi R. Br., according to a description by Arnold in a letter to Robert Brown, measured a yard across and sprang from a root about the size of two fingers. Brown described the flower as parasitic on a species of Cissus and thought that part of the base was derived from the host. This opinion he later confirmed ^ and showed that the shoot of Rafflesia consisted of a short thick stalk terminating in a single large flower, around which, in its early stages, there were a number of bracts ; and that the base of the shoot was em- bedded in a cup-shaped mass of tissue formed by the host. The general relation of the parts of the flower and host is shown in figure 3 which represents an expanded flower bud of Rafflesia manillana. Robert Brown’s drawings show the young buds covered by the bark of the host. ‘ Brown, R. An Account of a New Genus of Plants, named Rafflesia. Trans. Linn. Soc. 13 (1842) 201-234. ^ Brown, R. Description of the Female Flower and Fruit of Rafflesia Arnoldi, with Remarks on its Afflnities ; and an Illustration of the Structure of Hydnora Africana. Trans. Linn. Soc. 19 (1845) 221-239. 111784 209 210 BROWN. According to Schaar,^ Unger/ in a paper which is not avail- able to the writer, says in reference to Rafflesia patma, “Wir sehen den Parasiten mit der Nahrpflanze in der Art vereinigt, dass derselbe unmittelbar iiber dem Holzkorper seines Tragers entspringt und durch sein Gefasssysten mit dem Gefasssystem der Nahrpflanze anastomosirt.” Solms-Laubach ® showed that the vegetative portion of Pilo- styles haussknechtii, one of the Rafflesiaceae, consisted of simple cell strands in the pith, medullary rays, vascular elements and bark of species of Astragalus. These cell-strands Solms-Laubach designated as mycelium on account of their resemblance to the mycelium of a fungus. They extend into the bases of the leaves where they form cushions from which the flower buds are de- veloped. These buds break through the tissues of the host and appear upon the basal portion of the leaves. A similar structure of the vegetative portion of the plant has been shown by Solms-Laubach ® and Peirce ^ for Brugmansia zippelii, by Solms-Laubach ® and Pierce ^ for Rafflesia patma and by Schaar ° for Rafflesia rochussenii. The present study is based upon material of Rafflesia manil- lana Teschem.^° collected during January and February, 1912, on a forested ridge of Mount Maquiling, Province of Laguna, Luzon, at an elevation of about 400 meters. The original description of Rafflesia manillana by Tesche- macher was based on small unopened male buds, preserved in alcohol. The only description of an expanded flower is a short and unsatisfactory one by Blanco of a male flower of R. phil- ’ Schaar, F. Uber den Ban des Thallus von Rafflesia Rochussenii Teysm. [und] Binn. Siztber. Acad. Wien 107 (1898) Abt. 1: 1039-1056. * Unger, F. Beitrage zur Kenntniss der parasitischen Pflanzen. Annalen des Wiener Museums der Naturgeschichte. 2 (1840). Solms-Laubach, H. Graf zu, Uber den Thallus von Pilostyles haus- sknechtii. Bot. Zeit. 32 (1874) 49-59, 65-74. ® Solms-Laubach, H. Graf zu. Die Entwicklung der Bluten bei Brug- mansia zippelii und Aristolochia clematitis. Bot. Zeit. 34 (1876) 449. ’ Peirce, G. J. On the Structure of the Haustoria of some Phanerogamic Parasites. Ann. Bot. 7 (1893) 291-327. * Solms-Laubach, H. Graf zu, Rafflesiaceae. Engler & Prantl Nat. Pflanzenfam. 3^ (1889). “Schaar, F. Sitzber. Acad. Wien 107 (1898) Abt. 1: 1039-1056. “ Teschemacher, J. E. On a new species of Rafflesia from Manilla. Boston Journ. Nat. Hist. 4 (1844) 63-65, t. 6; Ann. & Mag. Nat. Hist. 9: 381-384. “Blanco, M. Flora de Filipinas, ed. 2 (1845) 565. RAFFLESIA MANILLANA. 211 ippinensis, which is evidently the same species as R. manillana Teschem. The female flower has not been previously described. The material of Rafflesia manillana from Mount Maquiling was parasitic on a species of Cissus. The male and female flow- ers have the same general appearance. (A male flower is shown in flg. 1 and a female one in fig. 2.) The differences seen in the photographs are individual variations and are not character- istics of the sexes. When fully expanded the flowers are from 15 to 20 cm in breadth and from 7 to 9 cm in height. The base of the flower is embedded in a cup-shaped mass of tissue formed from the root of the host (figs. 3, 9, 10). The outer surface of this tissue is rough but not reticulate as in R. arnoldi E. Br. The young buds (fig. 9) are enveloped in several series of sessile, imbricate, coriaceous, brown, broadly ovate, rounded bracts with prominent veins, the inner of which become gradually larger. When the flower is expanded the innermost bracts are about 7 cm long and 5.5 cm wide. As the bud expands these bracts become broken so that usually only the inner ones and the basal portion of the outer ones remain attached to the fully expanded flower. When the flowers first open they have a light reddish-brown color. This color gradually grows darker until it becomes black in old flowers. The perianth (figs. 1-3) consists of a broad, some- what turbinate tube 7 to 9 cm in diameter, bearing 5 suborbicular to subreniform lobes. The lobes are about 7 cm broad and 5 cm long, at first spreading, in age becoming recurved. The per- ianth-tube is crowned inside by a prominent, continuous, in- curved annulus, the tube, inside, being about 2.5 cm deep. The inner surface of the tube and annulus is thickly spotted with white warty patches. These increase in length from a fraction of a mm at the base of the tube to nearly a cm at the top (figs. 1, 3). At the base of the tube these consist of rounded masses on slender stalks, but as they increase in size they become sessile The inner surface of the lobes is spotted with smooth, white, raised patches (figs. 1, 2). In the center of the flower (figs. 1-4) there is a large column which rises from a circular, basal disk, thickly beset with stout brown hairs. The column terminates in an expanded, circular, convex disk surrounded by a raised margin. On the top of the disk there are from 10 to 30 large raised processes. The tips of these processes and the margin of the disk are densely brown-tomentose. The column is about 2 cm in height and the disk at its summit 4.5 cm broad. 212 BROWN. All parts of the flower are very thick, fleshy and brittle. The flower has a distinctly putrid odor which is, however, not very strong. The anthers (fig. 4) are suspended in a circular arrangement from the lower surface of the disk at the summit of the central column and are partitioned off from each other by plates of tissue. In the material examined there were from 12 to 14 anthers. Ac- cording to Teschemacher there are 10. The anthers are round and open by an apical pore which connects with a number of cavities running through the anther longitudinally. The pollen, when discharged, is suspended in a liquid. This liquid, contain- ing the pollen, hangs as a yellow drop from the lower end of the anther mass. Corresponding to each anther there is in the central column a deep groove which runs down into the basal disk. In the female flower these grooves are developed to a much less extent or may be entirely wanting. Without an examination of the lower surface of the disk at the top of the column, which necessitates the mutilation of the flower, the relative develop- ment of the grooves in the column and basal disk is the only superficial character by which the male and female flowers can be distinguished. The only parts of these grooves, which are visible in an entire flower, are the lower ends in the basal disk. The stigmatic surfaces of the female flower correspond in posi- tion to the anthers of the male (fig. 3). The ovarial cavity con- sists of labarynthine passages occupying a disk-shaped region below the central column (fig. 3) . In the male flower this region is occupied by parenchymatous tissue (fig. 4). The ovules are small and exceedingly numerous. Although both the condition of the pollen and the odor of the flower suggest that pollination is performed by insects, no fly- ing insects have been seen around the plants. Old flowers are, however, always infested with insect larvae, prominent among which are those of a species of Coleoptera. A large number of anthers were examined but in no case did the pollen appear to have been disturbed. Moreover no seed or developing embryos have been discovered. This is probably due to a scarcity of the pollinating insect as both the pollen and embryo-sacs appear to be perfectly normal. The ovule is anatropus and has one integument. The arche- sporium differentiates in the apex of the nucellus as a single hypodermal cell surrounded by a single epidermal layer. The archesporial cell without further division functions as a mega- spore mother cell, its nucleus undergoing the usual heterotypic RAFFLESIA MANILLANA. 213 prophases. After becoming considerably enlarged the mega- spore mother cell divides into a small micropylar and a larger chalazal daughter cell. The micropylar daughter cell usually divides in a plane parallel with the longitudinal axis of the nucellus. The chalazal daughter cell divides unequally into two megaspores, the chalazal or innermost one of which gives rise to the embroyo-sac. The three micropylar megaspores degenerate. The embroy-sac is formed from the megaspore by three succes- sive divisions and is of the usual 8-nucleate type. It contains an egg, two synergids, three antipodals and two polar nuclei. At first it is somewhat elongated but as growth proceeds it becomes more nearly round. The antipodals are ephemeral and degen- erate early. The polar nuclei fuse without the occurrence of fertilization. No sign of fertilization has been observed in the material examined, and in old ovules the embryo-sacs had degen- erated. The rarity of Rafflesia manillana may well be due, in part, to the fact that its flowers are frequently not pollinated. In the cases observed the open flowers did not increase in size. The chief changes observed were that they turned black and shriveled to some extent while the tissues gradually died. The ovules remained about the same size. There was a tendency for the tissue around them to break down into a semi-fluid mass, in which they degenerated in a few days. The flowers became infested with a number of insects and larvae some of which bored from the outside into the cavity containing the ovules. This allowed other insects to enter the cavity, and several kinds were usually found within it. It is possible that the seeds, if formed, are disseminated by some of these insects, as the ovules are enclosed in the fleshy tissue of the flower and there is no evident means of seed dispersal. VEGETATIVE STRUCTURE. The vegetative structure of Rafflesia manillana is quite similar to that of the other species of the Rafflesiaceae, previously men- tioned. It consists, for the most part, of rows of single cells running through the bark, cambium, medullary rays and xylem elements. In the material examined these cell-rows occurred more frequently in the tissues of the woody cylinder than in those of the bark or in the cambium. In the woody cylinder they were about equally frequent in the medullary rays and xylem elements. They were, perhaps, somewhat more numerous in the phloem and sclerenchyma than in the parenchyma of the bark. The cells composing the rows are usually fairly short, rather densely filled with protoplasm, and each contains one prominant opaque nu- 214 BROWN. cleus. The arrangement in rows, rather than in tissues, produces an appearance not unlike the mycelium of a fungus. The char- acter of the individual cells is, however, quite distinct from that of fungus cells, which are usually long, vacuolated, and have small inconspicuous nuclei. From the features, mentioned above, it will be seen that the vegetative cells of Rafflesia manillana are quite similar to embryonic cells in higher plants, and except for shape, they have an appearance which is quite like that of the embryonic cells in the flower. A row of Rafflesia cells in a medullary ray is shown in figure 14. In the upper and lower parts of the figure are sections of xylem rays between which there is a medullary ray. The black spots are medullary ray cells with densely staining contents. The cells of Rafflesia are seen as a row running transversely across the center of the figure. They have deep-gray colored contents and prominent nuclei. The cells of Rafflesia manillana, which form the rows in the roots of Cissus, are so intimately connected with those of the host that they appear to unite with them in forming a continuous tissue. The cells of the parasite, when in the cambium, medul- lary rays, or the parenchyma of the bark, are frequently dis- tinguishable from those of the host only by a difference in size and shape, and by the denser contents and more prominent nuclei of those of the parasite (fig. 14) . The cells of Rafflesia manillana, in the medullary rays of the woody cylinder, are usually wider, more densely filled with pro- toplasm, and relatively shorter than those in the xylem. Those in the region of the sieve-tubes have, in general, a greater diameter and denser contents than those in any other portion of the root. Those in the sclerenchyma usually contain large vac- uoles and little protoplasm. This vacuolation of the cells, when among the dead and empty ones of the sclerenchyma, the more vigorous appearance of those in the region of the sieve-tubes, and the larger size of the ones in the medullary rays as compared with those in the xylem would indicate that there is little conduction of food material by the vegetative cells of the parasite. The cell-rows of Rafflesia manillana run through the tissues of the host in all directions, but are more prominent in the longitudinal and radial planes of the root than in the tangential. Since they run in all directions in the various tissues, and are certainly continuous for very considerable distances, it is evident that the cells must grow and multiply in all parts of the root. Peirce states that the cells of Brugmansia zippelii “grow and Ann. Bot. 7 (1893) 291-327. RAFFLESIA MANILLANA. 215 multiply in the cambium, they grow only slowly, and do not mul- tiply in the wood and bast.” That the cells of Rafflesia manil- lana do not grow and multiply only in the cambium is shown by their presence in the primary tissues of the root. The greater development of the rows in the longitudinal and radial planes of the root than in the tangential or other planes may be due, in part, to their having been formed in embryonic tissue and then carried away from the meristematic region by the development of the tissues of the host, but it is also in the longitudinal and radial planes that the cells of the parasite could most easily push in between those of the host for it is in these planes that the latter are arranged in most regular rows. If the cells of the parasite did not grow and multiply in other than embryonic regions they would become separated by the growth of the host as in Brugmansia zippelii.^^ In Rafflesia manillana, however, the cells show little if any tendency to become broken by the growth of the host. In fairly large roots rows of cells several millimeters long may be found running radially in the wood. The appearance of the cells of the parasite would, moreover, frequently indicate that they were actively growing through the tissues of the woody cylinder and bark, and in the material examined they were cer- tainly relatively more numerous in the woody cylinder than in the cambium region. In other cases the arrangement of the cells of the parasite in the tissues of the host would seem to be due to their having pushed in between the cells of the root after the latter had ceased growing. The cells of Rafflesia manillana not only occur in the tissues of the host as rows of single cells but also, at times, as strands of tissue in the outer part of the bark of the root. Since these strands of tissue have not been seen in meristematic regions it would seem that they must have been developed in the outer part of the bark. Smaller strands (fig. 15), plates and irregular masses (fig. 16) of tissue are also frequently found in the woody cylinder. The strand of tissue shown in figure 15 is in the xylem of the root and can be distinguished from the latter by the con- tents and prominant nuclei in its cells. The plates of tissue usually lie in radial planes parallel to the longitudinal axis of the root. It would appear that the cells of Rafflesia manillana grow and multiply in all of the tissues of the host, but that growth is more active in the tissues of the woody cylinder than in the bark. In- dividual sections, however, show great variations in this respect. Peirce, Ann. Bot. 7 (1893) 291-327. 216 BROWN. In some cases the rows of cells of the parasite appear to be wedged in between the cells of the root, and in others they apparently replace them as by a dissolving action. In either case there is very little distortion of the cells of the root, which would indicate that the growth of the parasite through the tissues of the host is made possible by a dissolving action of the parasite rather than by pressure exerted by its cells. In the first case the dissolving action would be exerted on the middle lamella and in the second on the whole cells. The expansion of the cells of the parasite between those of the host must, however, be due in part to pressure exerted by the turgescence of the cells of the parasite. The bulk of the vegetative cells of Rafflesia manillana is small compared with that of the root of Cissus, and since they do not cause a distortion of the root or its tissues and probably take only enough nourishment to carry on their growth, it is likely that these vegetative cells, in themselves, do comparatively little harm to the host. As will be shown later, it is not probable that the fiowers are nourished by these vegetative cells to any great extent. The flowers of Rafflesia manillana are not produced progress- ively in the direction of growth of the root of the host, but young buds are frequently found between the scars of old flowers. This would indicate that the cells in one region retain their activity for a considerable length of time, although some of the latter buds may, of course, be due to more recent infections of the root. DEVELOPMENT OF THE FLOWER. The flower shoots of Rafflesia manillana have their origin in rows of cells which usually cross the cambium of the host. When such a row is to give rise to a shoot, the cells in the region of the cambium proliferate and form a globular mass of meristematic cells (fig. 11) , which is embedded partly in the woody cylinder and partly in the bark. The first proliferation may take place equally in the two regions or almost entirely in either. The cambium is destroyed by the developing parasite. These rudi- ments of flower shoots have been found in roots less than 2 mm in diameter. The method of the origin of the flowers of Rafflesia manillana is quite similar to that of Brugmansia zippelii.^^ The formation of the flower shoot usually begins in a medullary ray, although it would appear that they sometimes originate between Solms-Laubach, H. Graf zu, Bot. Zeit. 34 (1876) 449. RAFFLESIA MANILLANA. 217 the xylem and phloem. The growth of the embryonic flower shoot does not, at first, cause any distortion of the tissues of the host, which is probably due in part to the absorption of the lat- ter by the parasite. Soon, however, the embryonic shoot begins to grow more actively in the region toward the phloem, so that the rounded mass of cells becomes top-shaped (fig. 18). This is accompanied by a tendency of the xylem rays to be spread apart around the developing shoot, which process may take place to some extent even before the parasite begins to be top-shaped. The change in the shape of the parasite may take place simply by a more rapid enlargement of the portion in the phloem of the host. The top-shaped appearance is, however, frequently preceded by an irregular outgrowth of the parasite into the bark. This is usually the case when the first proliferation takes place almost entirely in the woody cylinder. At the same time that the xylem rays are becoming spread apart, the bark is pushed up in a dome-shaped projection (figs. 5, 8-11-13). The presence of the developing parasite in the tissues of the host causes an increased activity of the cambium of the host in the region around the parasite. This, together with the tend- ency of the parasite to cause the spreading out of the xylem rays, results in the production of a vase-shaped mass of xylem which projects from the side of the root, and in which the base of the pai;asite is embedded (figs. 21, 22). This vase-shaped mass be- comes more prominent as development proceeds. The increased production of xylem is accompanied by a similar increase in the phloem. The presence of the developing shoot not only causes a greater activity in the cambium, but there is also frequently a tendency for the host to develop a phellogen-like layer around the parasite. When this layer is produced it is usually formed first in the bark and then later may extend completely around the embryonic shoot. The phellogen-like layer, when once formed, gives rise to cells which soon become empty and have an appearance entirely similar to cork. Whenever these cells completely surround the parasite the latter is cut off from its food supply and con- sequently dies. The time at which the phellogen-like layer is produced, as well as the extent to which it is developed, varies very greatly in different specimens. It would seem that in some cases it does not make its appearance at all, but that in the major- ity of cases the embryonic shoots are killed by its growth. Usu- ally the shoots are surrounded and killed before they are large enough to cause any appreciable distortion in the tissues of the 218 BROWN. host, but in other cases death does not occur until the swelling on the side of the root is a centimeter or more in diameter. To the host the advantage of the process just described seems to be evident, for it must cause a great reduction in the number of flowers produced by the parasite. It may also be of advantage, indirectly, to the parasite for it probably saves the host from be- coming exhausted and allows the flowers of the parasite to be produced over an extended period. Why the phellogen-like layer should be produced in some cases and not in others is not evident, but this is probably connected with the relative vigor of host and parasite. As the shoot of Rafflesia develops it would soon burst through the bark of the host if considerable growth did not take place in the latter. The shoot, however, does not break through until a swelling, about 3 cm in diameter, has been formed on the side of the root (flg. 6). If the shoot is formed in a small root, the diameter of the swelling may be much greater than that of the root (flgs. 5, 6). The growth of the bark is accomplished by the division of cells throughout that part pushed up by the parasite. These cell divisions sometimes take the form of cam- bium-like layers. The result is that the cells of the bark are arranged in rows running toward the apex of the swelling (figs. 19, 21, 22). There is, however, considerable irregularity in this arrangement, as the divisions take place in a number of places, and the direction in which the phloem is cut off from the true cambium is constantly changing due to the continual spreading of the xylem. When the parasite is small, the phloem cells, formed from the cambium around it, are cut off in a plane parallel to the longitudinal axis of the root (fig. 19) ; but as the excessive growth and spreading apart of the xylem rays con- tinues, the phloem cells come to be cut off in a plane which approaches more and more toward being perpendicular to this (fig. 3). As mentioned before, the most rapidly developing part of the embryonic shoot is the rounded portion in the bark. The en- largement of this soon carries the region of most active growth of the parasite beyond the cambium of the host (figs. 18, 19). As the parasite continues to grow it becomes oval in longitudi- nal sections, with a pointed base (fig. 19). At this stage the vascular bundles are differentiated in a circular arrangement in the outer part of the parasite. These are shown as dark lines in figure 19. A cambium layer is formed on the outer side of each bundle. During the early stages there is also a RAFFLESIA MANILLANA. 219 tendency for the cells between the bundles to divide and produce a cambium. This process, if continued, would produce a con- tinuous ring of cambium. Soon, however, the cambium instead of forming such a ring grows around each bundle, the result being that each bundle comes to have a concentric arrangement (fig. 24) , as in Brugmansia zippelii and Rafflesia rochussenii.^^ While the vascular bundles and cambium are making their appearance, an apical growing region is also in the process of differentiation. This latter is preceded by the formation of a cap over the apex (fig. 20) . Since the whole surface of the parasite is joined to the host it would seem that the apex of the shoot could not become a growing region, which would cut off the bracts and floral parts, unless there were some process whereby the growing region would be separated from the host. Apparently the function of the cap is to cause this separation. The cap is formed from a layer, several cells thick, which ex- tends over the upper surface of the parasite (fig. 20) . Growth apparently ceases in these cells, after which they become stretched by the enlargement of the part of the parasite below them. A split then comes in between the cap and the under- lying tissue, while the cap remains firmly attached to the bark of the host. The apex of the shoot, beneath the cap, is then organized into a growing region from which the bracts and floral parts are cut off successively (figs. 21, 22, 12, 13). The cap continues to be attached to the bark of the host for a con- siderable period. In figure 21 this shows as a dark streak just under the bark of the host. A large part of the short stalk of Rafflesia manillana is formed from the apical growing region and the parasite again becomes top-shaped (figs. 21, 22). The rapid enlargement of the bracts and floral parts causes the cells of the shoot to be arranged in rows more or less perpen- dicular fx) the axis of the shoot (figs. 21, 22). The average bundles of Rafflesia manillana, which show as dark streaks in figures 19 and 22, connect with the host in the general region of the cambium. The exact point of connection varies in the same and different specimens. What is perhaps the most usual arrangement is for the cambium around the outer part of the bundles of the parasite to end in approximately the region of the cambium of the host. In other cases, however, various parts of the cambium of the parasite may be in con- nection with that of the host or the cambium of the parasite ” Peirce, C. J., Ann. Bot. 7 (1893) 291-327. Schaar, F., Sitzber. Acad. Wien 107 (1898) Abt. 1: 1039-1056. 220 BROWN. may be entirely within that of the host. After the separation of the apical region from the host the parasite grows con- siderably before it breaks through the bark. This is accom- panied by a considerable stretching of both host and parasite. The cambium of the parasite has comparatively little to do with the growth of the shoot. It would appear that all of the tissues, except the vascular elements, remain more or less meristematic until a very late stage and that growth and division take place in all parts of the shoot. The result is that there is a considerable enlargement of all parts, except perhaps the tip of the base. Figure 23, which is from a cross section of the central part of a young bud, shows divisions taking place in all directions. While the above-mentioned processes are going on, a phel- logen layer is differentiated in the outer layers of that part of the bark of the host which surrounds the parasite. As the parasite develops, this phellogen layer continues to produce new cells while the older ones on the outside become stretched and tend to be separated by cracks. This results in giving the bark a rough appearance, which becomes more prominent as devel- opment proceeds. As the parasite continues to grow, the flower bud, which has become separated from the bark of the host. Anally breaks through it (flgs. 6-8) . The bark of the host, however, remains attached to the stalk of the parasite until a very late stage when it may become separated in places by the activity of the phel- logen-like layers to be described later. The first sign of the breaking through of the flower bud is the appearance of a crack in the bark of the host across the top of the swelling (fig. 6). This gradually increased in length and other intersecting ones make their appearance. The flower bud, which is continually enlarging, then grows out through the opening made by these cracks (figs. 7, 8). The formation of the cracks is evidently due to pressure exerted by the turgid, growing bud. MacDougal,^^ while experimenting with artificially produced parasites, found that one plant could not be made to grow upon another unless the juices of the graft had a higher osmotic pres- sure than those of the stock. Judging from the great turgid- ity of the cells of Rafflesia manillana, it would seem that their contents must have a higher osmotic pressure than the contents of those of the root. The ability of the parasite to grow in the ” MacDougal, D. J., and Cannon, W. A., The Conditions of Parasitism in Plants. Carnegie Institution of Washington. Publ. 129 (1910). RAFFLESIA MANILLANA. 221 root of the host and of the bud, to break through the bark is prob- ably due, in part, to the turgor produced by this high osmotic pressure. The enlargement of the tissues of the host around the bud is, however, not a stretching caused by an increase in the size of the parasite, but an active growth which would seem to be due to a stimulus connected with its presence. After the bud breaks through the bark of the host, that part of the bark which surrounded the flower bud becomes recurved and gradually disappears (fig. 7). The stalk, however, conti- nues to be surrounded by the tissues of the host, which project from the side of the root as a vase-shaped mass. The base of this mass is composed of both wood and bark, while the upper part consists entirely of bark (fig. 3) . The wood and bark are separated by the cambium layer. The base of the stalk of the parasite is thus in contact with xylem of the host, and the upper portion with the phloem. After the parasite breaks through the bark all parts of the shoot continue to enlarge. This is accomplished by a prolifera- tion of cells from the cambium of the host and also in that part of the bark which surrounds the shoot. These processes are similar to those described in the earlier stages. When the flower is mature the stalk is still surrounded by the tissues of the host. There is, however, a tendency for the tissues of the parasite and host to become separated in places. This is due to phellogen-like layers in both host and parasite. In the late stages of the growth of the shoot there is a considerable develop- ment of a phellogen-like layer in, the inner layers of the bark of the host next to the parasite. This layer has an appearance which is almost identical with the one frequently produced around young shoots, and the development of the two is prob- ably due to similar stimuli. When this phellogen-like layer is formed in the bark of the host another is sometimes developed opposite it in the outer layers of the parasite. The extent of these layers varies in different specimens. In some they are produced over the greater part of the region around the parasite, while in others they extend over only a small portion. In none of the old stages examined did they occupy the whole region and they were never entirely absent. Whenever these two layers were produced opposite each other and were developed to any considerable extent there was a tendency for the host and par- asite to become separated. When the shoot of Rafflesia manillana first breaks through the bark of the host only a few of the bracts which surround 222 BROWN. the flower are visible (fig. 8). Owing, however, to the larger size of the inner bracts more and more of these become visible as growth proceeds (fig, 9). Finally the corolla appears through the bracts and the bud looses its rounded shape and becomes tur- binate (fig. 10), After this the flower soon opens. The growth of the bud is very slow, a bud about the size of one shown in figure 9 taking six weeks to develop into a fully expanded flower. Peirce states that both the sieve tubes and xylem of Brug- mansia zippelii are connected with the corresponding elements of the host. Schaar describes the xylem of Rafflesia rochus- senii as connected with that of the host but was unable to find any union between the sieve tubes of the two. The medullary rays comprise a large part of the roots of the Cissus on which Rafflesia manillana is parasitic. In consequence of this a con- siderable proportion of the vascular elements of the parasite end in connection with the medullary rays of the host. Both the xylem and sieve tubes of Rafflesia are, however, frequently in direct union with the corresponding elements of Cissus. Ow- ing to the concentric arrangement of the bundles of the parasite, this is true of a larger proportion of the xylem than of the sieve tubes and in one specimen in which the bundles of the parasite appeared to end completely within the cambium of the host, there seemed to be little if any chance for a connection between the sieve tubes of host and parasite. In the latter case, how- ever, the parasite appeared to be as vigorous as usual. An examination of figures 3 and 4 shows that a large part of the base of the shoot of Rafflesia manillana is surrounded by the phloem of the host. This would appear to obviate the necessity for a very close connection between the sieve tubes of host and parasite, and probably explains the vigorous condition of the parasite in the case just mentioned. The flowers of Rafflesia manillana, observed, decayed soon after opening and gradually disappeared. The part of the bark of the host in contact with the shoot also disappeared. The vase-shaped mass of xylem around the base of the shoot, how- ever, remained as a prominent scar on the root. A recent scar is shown in figure 25 and an older one in figure 26, As the flower disappears a callous is formed over the scar. An examination of figures 27 and 28, representing a cross and longitudinal section of a root with a scar, appears to show “Peirce, G. Z.,Ann. Bot. 7 (1893) 291-327. “Schaar, F., Sitzher. Acad. Wien 107 (1898) Abt. 1 1039-1056. RAFFLESIA MANILLANA. 223 that the growth of the flowers of Rafflesia manillana interfers with the conducting power of the xylem of the host to only a very limited extent. This is due to the excessive development of the xylem which gives the parasite a firm anchorage without its penetrating deeply into the main tissues of the root. The great reduction in the vegetative parts of Rafflesia and the excessive growth of the tissues of the root around the shoot would seem to be a fine adjustment between host and parasite which is of great advantage to both. The developing flowers are furnished with an abundant food supply without destroying the tissues of the host to any considerable extent. It would seem, indeed, that the taking of food from the cells of the root is the chief damage which the parasite does to the host. SUMMARY. Rafflesia manillana is parasitic on the roots of a species of Cissus. The male and female flowers are similar in shape and color, and from 15 to 20 cm in diameter. The base of the flower is embedded in a vase-shaped mass of tissue formed from the root of the host. Pollination was not observed but is probably performed by insects. The ovules are small and exceedingly numerous. The embryo sac is of the usual 8-nucleate type. The vegetative portion of Rafflesia manillana consists for the most part of rows of cells but also of strands, plates and irregular masses of tissue. The rows of cells occur in the xylem, medullary rays, cambium, phloem and schlerenchyma of the host and apparently grow and multiply in all of these tissues, except perhaps the schlerenchyma. They seem to have little power of conduction and probably do but slight damage to the tissues of the host. The flowers originate from rows of cells which usually cross the cambium. The row of cells proliferates and forms a rounded mass of meristematic cells which becomes top-shaped by the rapid growth of the part in the bark. Layers of cork-like cells are frequently produced in the tissues of the host around the parasite. These may cut off the food supply of the latter and cause its death. The presence of the parasite causes an excessive growth of both the xylem and bark of the host in the region around the parasite and also a spreading apart of the xylem rays. This results in the formation of a vase-shaped mass of tissue in which 224 BROWN. the base of the shoot is embedded, and in the early stages a dome- shaped projection of the bark which surrounds the apex of the developing shoot. The differentiation of the growing point takes place long before the shoot breaks through the bark of the host and is preceded by the formation of a cap of tissues which becomes separated from the apex. The vascular bundles are concentric and end in the general region of the cambium of the root. Some of the xylem and sieve tubes are connected directly with the corresponding elements of the host. The enlargement of the parasite finally produces cracks in the bark of the host through which the parasite grows. The bark around the flower gradually disappears but the base of the shoot remains embedded in a mass of xylem and bark formed by the root of the host. The excessive production of xylem and bark by the root gives the parasite a Arm anchorage and abundant food supply while the parasite does not destroy the xylem of the host to any con- siderable extent. The taking of food from the cells of the root would appear to be the chief damage which the parasite does to the host. DESCRIPTION OF THE PLATES. Figures 1 to 13 and 25 to 28 are from fresh material and are photo- graphs by Cortes. Wherever a scale is shown it is divided into cm. Figures 14 to 22 are microphotographs by Martin and 23, 24 by Cortes. Plate XII. Fig. 1. Male flower of Rafflesia manillana Teschem. 2. Female flower of Rafflesia manillana Teschem. The differences in the shape of the flowers shown in figures 1 and 2 are individual valuations and are not characteristic of the sexes. Plate XIII. Fig. 3. Longitudinal section through female flower and root of host. The base of the flower is in contact with a vase-shaped mass of xylem formed by the host. The bark of the host extends around the base of the flower from the xylem half way to the corolla. The cambium between the wood and bark of the host shows plainly as a dark line. The ovarial cavity of Rafflesia is shown in the white spongy tissue in the center of the base. 4. Section of unopened male bud of Rafflesia manillana. The root of the host is shown in cross section. The bark of the host extends around the base of the flower and ends just under the bracts. Two pollen sacs are seen under the disk which terminates the central column. Plate XIV. Fig. 5. A root of Cissus with two small swellings caused by Rafflesia manillana. 6. A bud of Rafflesia manillana breaking through the bark of Cissus. 7. A slightly older bud of Rafflesia manillana. 8. A root of Cissus with a bud of Rafflesia manillana which has, and two which have not, broken through the bark. Plate XV. Fig. 9. A root of Cissus with a half grown bud of Rafflesia manillana and two swellings. 10. Bud of Rafflesia manillana about ready to open. 11. Cross section of root of Cissus containing a bud of Rafflesia manillana. X 1.5 12. Older stage than flgure eleven. X 1-5 13. Older stage than figure twelve. X 1.5 111784 2 225 226 BROWN. Plate XVI. Fig. 14. Root of Cissus containing cells of Rafflesia manillana. In the upper and lower parts of the figure are sections of xylem rays between which there is a medullary ray. The black spots are medullary ray cells with densely staining contents. The cells of Rafflesia are seen as a row running transversely across the center of the figure. They have deep-gray contents and prominent nuclei. X 155 15. Strand of tissue of Rafflesia manillana in xylem of Cissus. The cells of Rafflesia can be distinguished by their deep-gray color and prominent nuclei. X 155 16. Large mass of tissues of Rafflesia manillana in central cylinder of Cissus root. X 115 Plate XVII. Fig. 17. Embryonic shoot of Rafflesia manillana in root of Cissus. X 155 18. Older shoot of Rafflesia manillana. Tissues still undifferentiated. X 60 Plate XVIII. Fig. 19. Shoot of Rafflesia manillana at the stage when the growing point and vascular bundles are being differentiated. The vascular bundles show as dark streaks in the light colored tissue of the parasite. X 8.5 20. Enlarged view of apex of shoot of Rafflesia manillana shown in figure 19. The dark tissue at the top of the figure is the bark of the host. The layer, a few cells in thickness, which is just below this, is the cap which is formed over the growing point of Rafflesia. X 100 Plate XIX. Fig. 21. Cross section of Cissus root containing a young bud of Rafflesia manillana. The cap over the apex of Rafflesia shows as a dark streak under the bark of the host. X 10 22. Longitudinal section of Cissus root containing older bud of Rafflesia manillana. X 7.5 Plate XX. Fig. 23. Cross section of parenchymatous tissue of Rafflesia manillana showing irregular division of cells. X 100 24. Cross section of young vascular bundle of Rafflesia manillana. X 100 Plate XXL Fig. 25. Scar left on root of Cissus after the disappearance of the flower of Rafflesia manillana. X 1-3 26. Scar left on root of Cissus after the disappearance of the flower of Rafflesia manillana. X 1-3 27. Cross section through a scar on root of Cissus. X 1.3 28. Longitudinal section through a scar on root of Cissus. Xl.3 Fig. 1. Male flower of Ragia*ia manillana PLATE brown : Rafflesia Manillana Teschem.1 [Phil. Journ. Sci.. VII. C. No. 4 Fig. 4. Section of unopened male bud of Rafflesia manillana. PLATE brown: Rafflesia Manillana Teschem.] 5. A root of Cissus with two small swellings caused by Ra^esia maniilana. Fig. 6. A bud of Rafflesia ntanillana breaking through the bark of Cissus. Fig. 7. A slightly older bud of Rafflesia vianillana. Fig. 8. A root of Cissus with a bud of Rafflesia manillana which has, and two which have not, broken through the bark. PLATE l! brown: Rafflesia 'II. C. No. Manillana Teschem.] [Phil. Journ. Set.. V PLATE PLATE PLATE XVII. |; [Phil. Journ. Sci.. VII. C. No. PLATE XXL The Philippine Journal op Science. C. Botany. Vol. VII, No. 4, September, 1912. NOMENCLATURAL AND SYSTEMATIC NOTES ON THE FLORA OF MANILA. By E. D. Merrill. (From the Botanical Section of the Biological Laboratory, Bureau of Science, and the Department of Botany, University of the Philippines, Manila, P. I.) In connection with the proximate publication of my “Flora of Manila” ^ it has been found necessary to make certain changes in nomenclature in order to bring the work into conformity with the at present generally accepted code of botanical nomencla- ture, that of the Vienna Botanical Congress, as modified by the additional list of nomina conservanda adopted by the Brussels Botanical Congress. Then too a rather intensive exploration of the region covered by the “Flora of Manila” has brought to light a few apparently undescribed species, and a number of forms, previously described from extra-Philippine material, which have never been definitely recorded as growing in the Archipelago. For the former it has become necessary to prepare diagnoses and descriptions, and for the latter it has been considered worth while to record them as Philippine species. The author does not consider that a publication of the nature of the forthcoming “Flora of Manila” is the proper place in which to describe new species, to make new combinations, to enter into extended dis- cussions of disputed points in nomenclature, or to discuss the validity or nonvalidity of species. Accordingly such matters are considered in the present paper, so that the “Flora of Manila” will contain no descriptions of new species, no new names, and no new combinations. The following paper contains the descriptions of five presum- ably new species, and the names of about 13 additional species now definitely recorded from the Archipelago for the first time. ‘Bureau of Science Publication (1912). 227 228 MERRILL. One family, Stylidiaceae, with the genus Stylidiiim, although known as Philippine for several years, is here definitely added to the Philippine list. In nomenclatural studies it has been found necessary to publish twenty-three new combinations or new names, in strict accordance with the provisions of the Vienna Code. POTAMOGETONACEAE. RUPPIA L. RUPPIA MARITIMA L. Sp. PI. (1753) 127, subsp. ROSTELLATA Koch in Reichb. Ic. PI. Crit. (1824) 66, t. 17 U; Graebner in Engl. Pflanzenreich 31 (1907) 144. Luzon, Province of Rizal, Malabon, Phil. PI. 799 Merrill, January, 1911, in brackish water of fish ponds. Widely distributed in temperate and tropical regions of both hemis- pheres. Ruppia maritima L. has previously been reported from the Phil- ippines by Naves,“ but his record has not previously been verified. HYDROCHARITACEAE. HALOPHILA Thou. HALOPHILA BECCARII Aschers. in Nuovo Giorn. Bot. Ital. 3 (1871) 302. Luzon, Province of Rizal, Paranaque, Merrill 80 U5 (det. C. H. Osten- feld), July, 1911. Abundant in some of the fish ponds, in salt water, in certain ones quite covering the muddy bottom in water 1 m deep or less. A species previously known only from Ceylon and from Sarawak, Borneo, the fourth one of the genus to be found in the Philippines. . GRAMINEAE. ANDROPOGON L. ANDROPOGON ZI2ANI0IDES (L.) Urb. Symb. Ant. 4 (1903) 79. Phalaris zizanioides L. Mant. 2 (1771) 183. Andropogon squarrosus L. f. Suppl. (1781) 433; Hack, in DC Monog. Phan. 6 (1889) 542. Vetiveria zizanioides Nash in Small FI. S. E. U. S. (1903) 67. If Andropogon is to be considered in the broad sense, following Hackel’s conception of the genus, then the oldest specific name for this well known species, which according to the Vienna Code must be taken up, is that indicated above. More recently the same transfer has also been made by Hochreutiner.^ The species is widely distributed in the Philippines, but has probably been introduced within historical times. "Novis. App. (1880) 297. “Bull. N. Y. Bot. Card. 6 (1910) 262. THE FLORA OF MANILA. 229 ARTHRAXON Beauv. ARTHRAXON QUARTINIANUS (A. Rich.) comb. nov. Alectridia quartiniana A. Rich. Tent. FI. Abyss. 2 (1851) 448, t. 99. Arthraxon ciliaris Beauv. subsp. quartinianus Hack, in DC. Monog. Phan. 6 (1889) 356. For the purposes of a local flora it is deemed expedient to consider this form as speciflcally distinct; it is not uncommon in northern Luzon, and has recently been found in the vicinity of Manila. In working over the Philippine material of this genus it was noted that our two forms, placed by Hackel as subspecies of Arthraxon ciliaris Beauv., were really much more distinct from each other than are some generally recognized species of allied genera. It is believed that the subspecies langsdorffii Hack., and the subspecies quartinianus Hack., should be held to be distinct species, and in accordance with the Vienna Code, the following new combination is necessary: ARTHRAXON HISPIDUS (Thunb.) comb. nov. Phalaris hispida Thunb. FI. Jap. (1784) 44. Arthraxon ciliaris Beauv. Agrost. (1812) 111, t. 11, f. 6. Arthraxon ciliaris Beauv. subsp. langsdorffii (Trin.) Hack, in DC. Monog. Phan. 6 (1889) 254. Not uncommon in northern Luzon, but not known from the vicinity of Manila. ERAGROSTIS Host. ERAGROSTIS IVi ANGALORICA Hochst. ex Steud. Syn. Gram. (1854) 265. Eragrostis tenella, R. & S., var. viscosa Stapf, subvar. contracta Stapf in Hook. f. FI. Brit. Ind. 7 (1897) 316. There are two specimens in our herbarium that appear to agree sufficiently with this form, and as our Philippine specimens show no intergradations between the form with the dense, spike-like contracted panicle (E. mangalorica Hochst.), and the lax-panicled one {E. viscosa Trin.), it may, at least for local purposes, be best considered as having speciflc rank. The specimens are Merrill 871, collected in Manila, August, 1902, and Bur. Sci. 7U52 Ramos, from Aparri, Province of Cagayan, Luzon. The former has been determined by Pilger ^ as Eragrostis tenella var. viscosa Stapf. SPINIFEX L. SPINIFEX LITTOREUS (Burm. f.) comb. nov. Stipa littorea Burm. f. FI. Ind. (1768) 29. Stipa spinifex L. Mant. 1 (1767) '"84. Spinifex squarrosus L. Mant. 2 (1771) 300. According to the provisions of the Vienna Code, this new combination is necessary. The oldest speciflc name is that supplied by Stipa spinifex L. (1767), but as the use of duplicate binomials is inadmissible under the Vienna Code, the next oldest specific name must be adopted, that supplied by Stipa littorea Burm. f. (1768). ‘Perk. Frag. FI. Philip. (1904) 148. 230 MERRILL, Zoisia Willd. {Zoysia Auct.) ZOISIA MATRELLA (L.) comb. nov. Agrostis matrella L. Mant. (1767) 185. Zoisia pungens Willd. in Ges. Naturf. Fr. Neue Schrift. 3 (1801) 441; Hook. f. FI. Brit. Ind. 7 (1897) 99 (Zoysia)-, Merr. in Philip. Journ. Sci. 1 (1906) Suppl. 342. Osterdamia inatrella O. Kuntze Rev. Gen. PI. (1891) 781. This species is common and widely distributed in the Philippines especially near the seashore. The new combination is necessary according to the Vienna Code, as Zoisia Willd. (1801) is included in the list of nomina conservanda in preference to Osterdammia Neck. (1791), while Agrostis matrella L. supplies the oldest specific name. BAM B USA Schreb. BAMBUSA GLAUCESCENS (Wifid.) Sieb. ex Munro in Trans. Linn. Soc. 26 (1868) 89, in syn. Ludolphia glaucescens Willd. in Ges. Naturf. Fr. Berl. Mag. 2 (1808) 320, Enum. Hort. Berol. (1809) 1035. Panicum glaucescens “Lam.” ex Roem. & Schultes Syst. 2 (1817) 846. Arundinaria glaucescens Beauv. Agrost. (1812) 144, 152. Bamhusa nana Roxb. Hort. Beng. (1814) 25, nomen, FI. Ind. 2 (1832) 199; Gamble in Ann. Bot. Card. Calcutta 7 (1896) 40, pi. 38, Hook, f. FI. Brit. Ind. 7 (1897) 390, Philip. Journ. Sci. 5 (1910) Bot. 268. This species is only cultivated in Manila and is a native of China or Japan. Its synonymy is quite complicated, but what is apparently the oldest valid specific name is here adopted. Panicum arborescens Linn. Sp. PI. (1753) 59 may in part be referable here, as “Index Kewensis” gives this species as in part referable to Arundinaria glaucescens and in part to Panicum sparsicomun Nees. I am unable to determine the type of the Linnean species, the first reference being to “FI. Zeyl. 43,” and the description given by Linnaeus in that work does not appear to me to be applicable to Bambusa glaucescens; the second reference is to “Hort. Cliff. 27,” which work is not available here. The reference to “Panicum glau- cescens Lam. Encycl. 4 (1798) 749,” quoted by many authors and given in “Index Kewensis,” does not appear in Lamarck’s work, and Roemer and Schultes appear to have been the first authors to use the name. I am informed by the Director of the Royal Gardens at Kew that the original manuscript of “Index Kewensis” gives no additional information regarding this name, and that it was probably taken up from Roemer & Schultes “Syst. Veg.”, or from Kunth’s “Enumeratio” without checking the reference. Panicum arborescens Lam. is published on page 749 of the “Encyclopedie,” and “Panicum glaucescens” of Roemer and Schultes was probably an error in transcribing the name. Panicum arborescens Lam. (non Linn.) is undoubtedly the same as Bambusa glaucescens (Willd.) Sieb. The specific name “glaucescens” undoubtedly dates from Willdenow as given in the synonymy above. THE FLORA OF MANILA. 231 CYPERACEAE. MARISCUS Gaertn. MARISCUS DILUTUS (Vahl) Nees in Wight Contrib. (1834) 90. Cyperus dilutus Vahl Enum. 2 (1806) 357. Mariscus microcephalus Presl Rel. Haenk. 1 (1830) 182; Clarke in Hook. f. FI. Brit. Ind. 6 (1894) 624, Philip. Journ. Sci. 2 (1907) Bot. 88. Sphaeromariscus microcephalus Camus Not Syst. 1 (1910) 239. Very recently E. G. Camus has proposed the new genus Sphaeromariscus, based on Mariscus microcephalus Presl. Should this proposal meet with general acceptance, Vahl’s specific name will later have to be taken up under the new generic name, in accordance with the principles of the Vienna Code of Botanical Nomenclature. FIMBRISTYLIS Vahl. FIMBRISTYLIS CORN I CU LATA sp. nov. § Trichelostylis. Annua, caespitosa, erecta, glabra, circiter 20 cm alta; culmis tenuibus, efoliatis ; spiculis paucis, 3 ad 5, simpliciter umbellatis, pedicellatis vel centrale sessile, brunneis, 3 ad 8 mm longis; squamis oblongis; brunneo-puncticulatis, margine ciliatis, apice rotundatis vel retusis, carinatis, carina prominente, excurrente; nux obovata, verruculosa, 0.5 mm longa; stylo 3-partito. A tufted, glabrous, apparently annual plant, the culms slender, about 20 cm high, terete, slightly striate. Basal leaves few, linear, 5 to 8 cm long, about 1 mm wide; culm-sheaths all near the base, short, leafless. Spikelets few, 3 to 5, arranged in a simple umbel, the middle one sessile, the others pedicelled, ulti- mately usually reflexed, brown, 3 to 8 mm long, the glumes 15 or less. Involucre at the base of the umbel of few, lanceolate, acuminate bracts 5 mm long or less. Empty glumes about 2.5 mm long, the keel very prominent, excurrent. Flowering glumes oblong, about 2 mm long, brown-puncticulate, margins ciliate- pubescent, apex rounded and cucullate, or retuse, the keel very prominent, excurrent as a distinct awn, the glumes deciduous from the base upward, leaving prominent scars on the rachis. Stamens 2 or 3. Nut obovate, black, verruculose, obscurely triquetrous, about 0.5 mm long; style slightly pubescent, 3-partite. Luzon, La Loma, near Manila, Merrill 7359, December, 1910, in open grasslands, altitude about 10 m. A species apparently well characterized by its few spikelets and especially by its glumes being rounded and cucullate or retuse, the prominent keel excurrent as a distinct awn. It seems to be allied to F. tenera R. & S. 232 MERRILL. ERIOCAULACEAE. ERIOCAULON L. ERIOCAULON ALATUM H. Lecomte in Journ. de Bot. 21 (1908) 104, fig. 2. Luzon, Province of Rizal, Bosoboso, Bur. Sci. 1831 Ramos, January, 1907; Manila, La Loma, in old rice paddies, Merrill 7362, December, 1910. The specimens agree perfectly with Lecomte’s description and figure, and with a specimen in our herbarium from Cochinchina cited by Lecomte in his original description, ex herb. Pierre. Previously known only from Indo-China. COMMELINACEAE. ANEILEMA R. Br. ANEILEMA M ALABARICUM (L.) comb. nov. Tradescantia malabarica L. Sp. PI. ed. 2 (1763) 412. Commelina nudicaulis Burm. FI. Ind. (1768) 17, t. 8, f. 1. Aneilema nudiflorum R. Br. Prodr. (1810) 271, in nota; C. B. Clarke in DC. Monog. Phan. 3 (1881) 210. This species is very common and widely distributed in the Philippines, the above change of name being in accordance with the provisions of the Vienna Code of Botanical Nomenclature. Robert Brown simply mentions the species in a note, basing his Aneilema nudiflorum on Commelina nudiflora Vahl, who in turn based his species on Commelina nudifiora L. Mant. 177. Unfortunately the original Commelina nudiflora L. Sp. PI. (1753) 61, is a valid species of the genus in which it was placed by Linnaeus, but Commelina nudifiora L. Mant. (1767) 177 is a quite different species and is Aneilema nudiflorum R. Br. Even if Commelina nudiflora L. Mant. (1767) 177 were valid, it is antedated by the publication of Tradescantia malabarica L. Sp. PI. ed. 2 (1763) 412. ANEILEMA VERSICOLOR Dalz. in Hook. Journ. Bot. & Kew Miscel. 3 (1851) 136; Clarke in DC. Monog. Phan. 3 (1881) 208; Hook. f. FI. Brit. Ind. 6 (1892) 378. Luzon, Province of Rizal, Antipolo, Bur. Sci. 10877 Ramos, October, 1909; Caloocan, Merrill 36UU, November, 1903; Masambong, Phil. PI. 763 Merrill, November, 1910. This species has not previously been reported from the Philippines, although it is abundant in old rice lands, etc., in the vicinity of Manila, from Caloocan to La Loma and Masambong. The flowers are russet-brown or brownish-yellow when fresh, but the petals turn dark-purple in drying. India. ZEBRINA Schnizl. ZEBRINA PENDULA Schnizl. in Bot. Zeit. 7 (1849) 870; C. B. Clarke in DC. Monog. Phan. 3 (1881) 318. Luzon, Manila, Merrill s. n., September, 1909, from cultivated specimens, Baja 203, August, 1907; Province of Laguna, Nagcarlan, Bur. Sci. 2U69 Foxworthy, March, 1907. This widely cultivated species is not uncommon in cultivation in Manila, THE FLORA OF MANH.A. 233 and is occasionally found as an escape; in Laguna Province it appears to be thoroughly naturalized, being found about the bases of coconut trees, along roadsides, etc. A native of Mexico. IRIDACEAE. ELEUTHERINE Herb. ELEUTHERINE PALMIFOLIA (L.) comb. nov. Sisyrinchium palmifolium L. Mant. 1 (1767) 122, saltern pro maxima parte; Naves Noviss. App. (1880) 252. Sisyrinchium bulbosum Mill. Gard. Diet. ed. 8 (1768) no. 3. Ixia americana Aubl. PI. Guian. 1 (1775) 33. Moraea plicata Sw. FI. Ind. Occ. 1 (1797) 82. Eleutherine plicata Herb, in Hot. Reg. 29 (1843) sub t. 57; Baker in Journ. Linn. Soc. Bot. 16 (1877) 100. Antholyza meriana Blanco FI. Filip. (1837) 24, ed. 2 (1845) 18, non L. This American species must have been introduced into the Philippines at an early date, and at the present time is of local occurrence, occasionally cultivated, and in some places thoroughly naturalized. It was collected by Cuming (herb. Kew.), and has later been collected by Loher in Benguet Subprovince, Luzon (no. 1609), by Mr. Elmer in southern Negros, and by myself in eastern Mindanao. In accordance with the Vienna Code of Botanical Nomenclature, the above new combination seems to be necessary. The oldest generic name is Galatea Salisb. (1812) but Eleutherine is retained in the list of noniina conservanda of the Vienna Code. OPILIACEAE. CHAMPEREIA Griff. CHAMPEREIA MANILLANA (Blume) comb. nov. Cansjera manillana Blume Mus. Bot. Lugd.-Bat. 1 (1850) 246; Hallier f. Meded.’s Rijks Herb. 1910 (1911) 14. Opilia cumingiana Baill. Adansonia 3 (1862) 124. Opilia manillana Baill. 1. c. Champereia griffithiana Planch, ex Kurz in Journ. As. Soc. Beng. 44 ^ (1875) 154; Hook. f. FI. Brit. Ind. 5 (1886) 236; Forbes & Hemsl. in Journ. Linn. Soc. Bot. 26 (1894) 409. Champereia griffithii Kurz For. FI. Brit. Burma 2 (1877) 330; F.-Vill. Novis. App. (1880) 185; Vid. Sinopsis Atlas (1883) f. 81, f. D., Phan. Cuming. Philip. (1885) 141. Champereia cumingiana Merr. in Philip. Journ. Sci. 1 (1906) Suppl. 50. This common species is represented in the Herbarium of the Bureau of Science by about 50 specimens, from all parts of the Philippines, from northern Luzon to southern Mindanao. It is known from Tenasserim to the Andaman Islands, Penang, and the Malay Peninsula, Formosa, and the Philippines. So far as I have been able to determine, Blume’s Cansjera manillana supplies the earliest specific name, which is here adopted. Hallier ° f., who has seen the type in the Leiden Herbarium, states that '’Meded.’s Rijks Herbarium 1910 (1911) 14. 234 MERRILL. Cansjera manillana Blume is the same as Champereia griffithiana Planch.; I had previously proposed to take up the name Champereia cumingiana," at least for the Philippine form, as it antedates Planchon’s specific name. I was not sure, however, that the Philippine plant was the same as the Indian one, although at Kew the Philippine material had been referred to Planchon’s species. The type of Cansjera manillana Blume was collected in the Philippines by Perrottet, and the type of Opilia manillana Baill., was also a Perrottet specimen, undoubtedly the same collection. Baillon does not cite Blume’s Cansjera manillana, and the use of the same specific name by both authors may have been merely a coincidence, or Baillon may have neglected to cite Blume’s previous publication (as Cansjera) . The first actual description of the plant seems to have been Blanco’s Malulucban,^ but this name has no standing, not being a binomial. Many authors have considered that Malulucban, and species similarly described by Blanco, to be of generic rank, but I believe this to be unwarranted. Blanco certainly did not intend them for genera, but simply as descriptions of species that he was unable to place in the Linnean system to his satisfaction; this is made manifest by the fact that the names both in the text and in the index are included in parentheses. The numerous Philippine specimens have been distributed in part as Champereia griffithii Planch., and in part as C. cumingiana (Baill.) Merr. While there is some variation in vegetative characters, I am of the opinion but that a single species is represented, and that its proper name is Champereia manillana (Blume) Merr. PORTULACACEAE. PORTULACA L. PORTULACA PILOSA L. Sp. PI. (1753) 445. Luzon, Province of Cavite, Mendez Nunez, Bur. Sei. H53 Mangubat; Province of Rizal, Mariquina, For. Bur. 5207 Curran; San Juan del Monte, Guerrero: Province of Pangasinan, Tayug, Bur. Sci. 8297 Ramos. Not previously reported from the Philippines. Widely distributed in tropical America, introduced in the Philippines and naturalized. ANONACEAE. ARTABOTRYS R. Br. ARTABOTRYS UNCINATUS (Lam.) comb. nov. Anona uncinata 'Lam. Encycl. 2 (1786) 127. Unona uncinata Dun. Monog. Anon. (1817) 105, t. 25; DC. Prodr. 1 (1824) 90. Artabotrys odoratissumus R. Br. Bot. Reg. 423; Hook. f. FI. Brit. Ind. 1 (1872) 54; King in Ann. Bot. Card. Calcutta 4 (1893) 44, pi. 55. This species occurs in the Philippines only as an introduced and cultivated plant. Sir George King expresses the opinion that it is truly wild only in southern India and in Ceylon; it is, however, much cultivated in India, and in other tropical countries. ° This Journal 1 (1906) Suppl. 50. ' FI. Filip. (1837) 186. THE FLORA OF MANILA. 235 CAPPARIDACEAE. CAPPARIS L. CAPPARIS CORDIFOLIA Lam. Encycl. 1 (1785) 609. Capparis mariana Jacq. Hort. Schoenbr. 1 (1797) 57, t. 109; Blanco FI. Filip, ed. 2 (184-5) 305; F.-Vill. Noviss. App. (1880) 11; Safford in Contr. U. S. Nat. Herb. 9 (1905) 212. Capparis baducca Blanco FI. Filip. (1837) 438, non Linn. Capparis spinosa var. mariana K. Scb. in Engl. Jabrb. 9 (1887) 201; K. Scb. & Lauterb. Deutscb Scbutzgeb. Siidsee (1901) 335. Blanco knew tbis species only from Paranaque, where it was cultivated in bis time, tbe seeds having been brought from the Marianne Islands; it is still cultivated in Paranaque (Guerrero, Sept., 1911), although not in great quantities, but I have seen no specimens from any other place. It is reported from the Marianne Islands, Guam, from the Caroline Islands, Kuschai and Ualan, from the Marshalls Islands, Nawodo, and from Timor. It may be only a variety of the European Capparis spinosa L., as considered by K. Schumann, but whatever its status, Lamarck’s specific name is unquestionably the oldest one, and the type of Capparis cordifolia was from the Marianne Islands. LEGUMINOSAE. DUNBARIA W. & A. DUNBARIA PUNCTATA (W. & A.) Benth. PI. Jungh. (1852) 242. Dolichos punctatus W. & A. Prodr. (1834) 247. Dolichos conspersus Grab, in Wall. Cat. (1831-32) no. 5542, nomen nudum. Dunbaria conspersa Benth. 1. c. 241; Baker in Hook. f. FI. Brit. Ind. 2 (1876) 218. Luzon, Province of Rizal, Malapad na bato, near Manila, Phil. PI. iS3 Ramos, September, 1910. What is the oldest valid name for this species is here adopted, for Dolichos conspersus Grab., as originally used, is only a nomen nudum. Not previously reported from the Philippines. India to China, south- ward to northern Australia. SESBANIA Pers. SESBANIA SESBAN (L.) comb. nov. Aeschynomene sesban L. Sp. PI. (1753) 714. Coronila sesban Willd. Sp. PI. 3 (1806) 1147. Sesbania aegyptiaca Poir. in Lam. Encycl. 7 (1806) 128; Pers. Syn. 2 (1807) 316; Baker in Hook. f. FI. Brit. Ind. 2 (1876) 114; Prain in Journ. As. Soc. Beng. 66 ^ (1897) 367. Emerus sesban O. Ktze Rev. Gen. PI. (1891) 180. Manila, Shaw, September, 1911. Not previously reported from the Philippines; widely distributed in the tropics of the Old World, and according to Prain 1. c., the var picta (Sesbania picta Cav.), a native of tropical America, introduced in India. The Philippine specimen is without doubt the variety typica of Prain, its flowers uniformly yellow. The above new combination is necessary if the oldest specific name is to be used, for it scarcely forms a “duplicate binomial” excluded by the Vienna Code. 236 MERRILL. SIMARUBACEAE. HARRISONIA R. Br. HARRISON I A PERFORATA (Blanco) comb. nov. Paliurus perforatus Blanco FI. Filip. (1837) 174, ed. 2 (1845) 122. Paliurus dubius Blanco 1. c. 175, 122. Lasiolepis paucijuga Bonn. PI. Jav. Ear. (1838) 202, t. i2. Lasiolepis multijuga Benn. 1. c. Lasiolepis bennetii Planch, in Hook. Lond. Journ. Bot. 5 (1846) 570. Harrisonia bennetii Hook. f. ex A. W. Benn. in Hook. f. FI. Brit. Ind. 1 (1875) 519. This species is widely distributed in the Philippines at low altitudes, extending from northern Luzon to southern Mindanao. Blanco’s specific name appears to be the oldest valid one for the species, both Paliurus perforatus and P. dubius being manifestly the same species. Burma to southern China, southward to Java. MALPIGHIACEAE. TRISTELLATEIA Thouars. Tristellateia australasiae A. Rich, presents an interesting case of synonymy on account of the approximately simultaneous publication of Platynema laurifolium Wight & Arn., which all authorities agree to be an exact synonym of Richard’s species. The synonymy is as follows: TRISTELLATEIA AUSTRALASIAE A. Rich. Sert. Astrolab. Atlas (1833) pi. 15. {T. australasica Auct.) Gaertneria laurifolia Wall. Cat. (1832) no. 7265, nomen. Tristellateia australis A. Rich. Sert. Astrolab. (text) 2 (1834) 38; Sprague in Curt. Bot. Mag. IV 6 (1910) t. 833U. Platynema laurifolium Wight. & Arn. in Edinb. New Philosoph. Journ. (Apr.-July, 1833) 179; Prodr. (1834) 107. Hiraea reclinata Blanco FI. Filip. (1837) 378, non Jacq. Tristellateia malintana Blanco 1. c. ed. 2 (1845) 267. As there appears to be no means of determining the relative status of the publication of Richard’s and Wight & Arnott’s names as to priority, the former, being in general use (as T. australasica) is here accepted. Authorities generally agree that the volume of plates accompanying Richard’s “Sertum Astrolabianum” was published in 1833, and the text, volume one in 1832, volume two, 1834; the plate of Tristellateia australasiae shows a complete dissection of the flower, and is hence a valid publication. As to Platynema laurifolium Wight & Arn., this was based on Gaertneria laurifolia Wall. (1832), and hence this name is the oldest, although un- fortunately the “publication” in Wallich’s “List” is only a nomen nudum and has no standing; in the “Edinburgh New Philosophical Journal” the genus Platynema is described, but the species, P. laurifolium, is enumerated without description; the specific description was published in the following year. THE FLORA OF MANILA. 237 I am indebted to Prof. Isaac Bayley Balfour of the Royal Botanic Gardens at Edinburgh, and to Mr. W. Craib of the Kew Gardens for abstracts of certain articles and notes bearing on the case. MELIACEAE. SANDORICUM Cav. SANDORICUM KOETJAPE (Burm. f.) comb. nov. Melia koetjape Burm. f. FI. Ind. (1768) 101. TricMlia nervosa Vahl Symb. 1 (1790) 31. Sandoricum indicum Cav. Diss. 4 (1787) 359, t. 202, 203. In accordance with the rules governing priority of specific names, the above combination is necessary in the case of this common and well known species; it is unfortunate that the specific name indicum, long in use, must be replaced by such a barbarous one as koetjape. POLYGALACEAE. SALOMONIA Lour. SALOMONIA CILIATA (L.) DC. Prodr. 1 (1824) 334. Poly gala ciliata L. Sp. PI. (1753) 705. Salomonia ohlongifolia DC. 1. c.; Benn. in Hook. f. FI. Brit. Ind. 1 (1872) 207. This species is widely distributed in the Philippines and is not uncommon. Previous authors have identified the local form with Salomonia ohlongifolia DC., but Trimen," who has examined the specimen in Hermann’s herbarium on which Linnaeus based his FI. zeyl. 270, and later his Polygala ciliata, is authority for the statement that Linnaeus’ type quite corresponds with Salomonia ohlongifolia DC. Manifestly the specific name ciliata should be applied to the present form, and if the species described under this name in Hooker’s “Flora of British India” is really distinct, then it should receive a new name. EUPHORBIACEAE. EUPHORBIA L. EUPHORBIA PROSTRATA Ait. Hort. Kew. 2 (1789) 139; Boissier in DC. Prodr. 15" (1862) 47. Specimens of this species were collected in waste places about Manila by Doctor C. F. Millspaugh of the Field Museum of Natural History, in November, 1911. In habitat it is very similar to the common and widely distributed Euphorbia thymifolia Burm., at once distinguishible by its capsules being glabrous except for the ciliate-hispid keels of the cocci. The species is otherwise represented in our herbarium by Wilson 228 from Cuba, and by Ridley 123, 126 from Christmas Island (south of Java). The species is undoubtedly of American origin, and has previously not been reported from the Philippines. 'Journ. Linn. Soc. Bot. 24 (1888) 146. 238 MERRILL. MACARANGA Thou. MACARANGA PORTEANA E. Andre in Rev. Hort. 60 (1888) 176, /. 36; Hook. f. in Curtiss’ Bot. Mag. (1895) t. 7U07. Croton grandifolium Blanco FI. Filip. (1837) 753, ed. 2 (1845) 518, non Macaranga grandifolia Muell.-Arg. Macaranga mappa F.-Vill. Noviss. App. (1880) 195; Vid. Sinopsis Atlas (1883) t. 8^, f D, Rev. PI. Vase. Filip. (1886) 246; Merr. in Philip. Journ. Sci. 1 (1906) Suppl. 80, non Muell.-Arg. This species is fairly well distributed in Luzon, and seems to be suffi- ciently distinct from Macaranga mappa Muell.-Arg., to which Philippine material has been referred by various authors. It was first described by Blanco, but his specific name is invalid in Macaranga. Endemic. MALLOTUS Lour. MALLOTUS PAPILLARIS (Blanco) comb. nov. Adelia papillaris Blanco FI. Filip, ed. 2 (1845) 562 (pappilaris) , ed. 3, 3: 225. Mallotus zollingeri F.-Vill. Nov. App. (1880) 195, non Muell.-Arg. Frutex erectus 2 ad 3 m altus, omnibus partibus densissime molliter stellato-puberulis vel stellato-pubescentibus ; foliis op- positis, late ovatis vel orbiculari-ovatis, basi late subtruncatis vel leviter cordatis, apice acuminatis, margine subintegris vel obscure repandis, usque ad 16 cm longis, basi triplinerviis ; racemis $ axillaribus, solitariis, usque ad 18 cm longis; stami- nibus circiter 60; capsulis dense stellato-pubescentibus, proces- sibus paucis, mollibus, brevibus instructis. An erect shrub 2 to 3 m high, all parts more or less densely covered with a pale or somewhat yellowish indumentum com- posed of short, stellate hairs. Older branches terete, dark- colored, nearly glabrous. Leaves opposite, broadly ovate to orbicular-ovate, chartaceous, 5 to 16 cm long, nearly as wide, the base broad, subtruncate, sometimes slightly cordate, the apex rather abruptly and often sharply acuminate, the margins sub- entire or slightly repand, the lower surface with numerous, pale, waxy glands scattered through the indumentum, the indumentum on the upper surface much less dense than on the lower surface ; basal nerves three, the lateral pair prominent, reaching to the middle of the leaf or above, the primary ones above the basal pair about 4 on each side of the midrib, prominent, straight or slightly curved, the reticulations distinct, subparallel; petioles 2.5 to 6 cm long. Staminate inflorescence in the upper axils, racemose, solitary, 6 to 18 cm long, densely stellate-pubescent, the flowers 5 to 10 in each fascicle, ebracteolate, their pedicels THE FLORA OF MANH.A. 239 about 5 mm long, densely stellate-pubescent. Calyx-segments four, oblong-elliptic, acute or slightly acuminate, outside densely stellate-puberulent. Stamens about 60 ; filaments glabrous, 2 mm long. Disk none. Capsules about 1.2 cm in diameter, of three dehiscent cocci, depressed, sulcate between the cocci, outside densely pale-stellate-puberulent, and with scattered, soft, pubes- cent, 2 mm long or less, spine-like processes or papillae which do not form a continuous covering. Luzon, Province of Rizal, Malapadnabato, Merrill 2720, Phil. PI. 434 Ramos; Morong, Bur. Sci. 1417, 3324 Ramos; San Mateo, For. Bur. 1110 Ahern’s collector: Province of Pampanga, Arayat, Bolster 56: Manila, Didrichen 3148 (Galathea Expedition). The original description of Adelia papillaris (pappilaris) Blanco, is entirely inadequate, consisting only of the following “Adelia pappilaris, Adelia de pezones. Arbolitos dioicos con las cagillas cubiertas no de barbas sino de pezones cortos. Guadalupe.” His material was from near Manila (Guadalupe, Province of Rizal), and I know of but two species of the genus to which the short description applies, the form above described and Mallotus playfairii Hemsl; although the latter species is widely distributed in Luzon, we have no material of it from Rizal Province. The above form needing a name, I have assumed it to be the species so in- adequately described by Blanco, transferring his specific name to Mallotus, and written up a fairly complete description of the species. The Malapad- na-bato specimens are from a locality in the vicinity of Guadalupe, the type locality of Adelia papillaris Blanco, and it also occurs in thickets just across the Pasig River from Guadalupe. The species is well characterized by its dense, stellate-puberulent in- dumentum and by its capsules being supplied with scattered, soft, spine-like processes, similar to those of Mallotus playfairii Hemsl. TILIACEAE. TRIUMFETTA L. TRIUMFETTA BARTRAMIA L. Syst. Nat. ed. 10 (1759) 1044, Sp. PI ed. 2 (1763) 638. Bartramia indica L. Sp, PI. (1753) 389, non Triumfetta indica Lam. Triumfetta rhomboidea Jacq. Enum. PI. Carib. (1760) 22, Stirp. Amer. Hist. (1763) 147. The Linnean name for this common and widely distributed species is here retained, as it has manifest priority over the commonly used T. rhom- boidea Jacq. Linnaeus may have included in Triumfetta bartramia more than Triumfetta rhomboidea Jacq., as that species in now understood, as suggested by Trimen,® but it seems that in most part the Linnean species is the same as Jacquin’s. 'FI. Ceyl. 1 (1893) 178. 240 MERRILL. MALVACEAE. MALVASTRUM A. Gray. MALVASTRUM COROMANDELINUM (L.) Garcke in Bonplandia 5 (1857) 297. Malvastrum tricuspidatum A. Gray PI. Wright. (1852) 16. Trimen “ states that Malva coromandelina L. is not, as Garcke supposed, Malvastrum, but is Sida acuta, basing this statement on an examination of Plukenet’s original specimens in the Herbarium of the British Museum. From an examination of Linnaeus’ description, it seems to be evident that two different species were included in his Malva coromandelina, and there is a chance for disagreement as to what constitutes the type of the species. The first reference, “Malva foliis ovato-oblongis acute serratis, capsulis tricuspidatis. Hort. cliff. 346. *Hort. ups. 201”, manifestly cannot apply to any species of Sida, and is undoubtedly the species named by many botanists Malvastrum tricuspidatum A. Gray. If the first reference be taken as the type of the species, then Garcke is correct, but if the second reference be taken as the type, and this is based on on a figure, Pluk. Mant. 10, t. 334, /• 2, then Trimen would be correct; unfortunately, should the second reference be selected as the type of the species, this course would necessitate a new name for the very common and widely distributed Sida acuta Burm., which it antedates by fifteen years. I consider it best to retain the Linnean name for the Malvastrum. OENOTHERACEAE. JUSSIAEA L. JUSSIAEA LINIFOLIA Vahl Eclog. Amer. 2 (1798) 32; DC. Prodr. 3 (1828) 55. Jussiaea acuminata Sw. FI. Ind. Occ. 2 (1800) 745; DC. 1. c. 54; Rolfe in Journ Bot. 23 (1885) 312; Vid. Phan. Cuming. Philip. (1885) 115, Rev. PI. Vase. Filip. (1886) 140. Jussiaea blumeana Presl Epim. Bot. (1851) 217, non ? DC. Jussiaea costata Presl 1. c. This species is very common in the Philippines, extending from northern Luzon to southern Mindanao, a weed in rice-paddies, waste places, etc. It has undoubtedly been introduced from tropical America, but is now so abundant and widely distributed as to appear like an indigenous species. All the recently collected material has been distributed as Ludwigia prostrata Roxb., to which it has indeed a strong resemblance in superficial characters. It can readily be distinguished, however, by its 8 stamens and by its pods being cylindric, not prominently 4-angled, and the seeds scarcely or not at all visible through the walls. As to Jussiaea blumeana Presl, and J. costata Presl, I think there can be no doubt as to their identity not only with each other but also with J. linifolia Vahl {J. acuminata Sw.). Presl cites the same number of Cuming’s collection (665) under both, two specimens of which are in our herbarium; both sheets undoubtedly represent a single species, whatever FI. Ceyl. 1 (1893) 141. THE FLORA OF MANILA. 241 may be the case in Presl’s herbarium. His descriptions appear to me to apply to a single form. It is suspected also that Jussiaea blumeana DC., reported by Miquel from Java, Sumatra, and Borneo, may prove to be identical. Ludwigia prostrata Roxb., w^hich has locally been confused with the above, appears to be a comparatively rare plant in the Philippines, being represented in our herbarium by four specimens only: Luzon, Province of Cagayan, Tuguegarao, Bur. Sci. 7936 Ramos, in part: Province of Rizal, Masambong, Marave JO; Caloocan, Merrill 3670, November, 1903. Min- danao, District of Davao, Williams 2633. ARALIACEAE. NOTHOPANAX Miq. The type of this genus is unmistakably Panax fruticosum L., which species most authors agree is generically distinct from Panax quinque- folium L., the type of the Linnean genus. However, there is great difference of opinion as to the proper place of the species and the manifestly allied forms, and the question is greatly confused by the numerous horti- cultural forms that have received specific names in various allied genera, and by the great variation of many of the species. There are in Manila a number of cultivated forms, all of them introduced, which are difficult to classify not only on account of their variation, but also because most of them rarely or never produce flowers. It seems to me that the logical course to follow is to recognize the genus Nothopanax Miq., as valid and typified by N. fruticosum (L.) Miq., and to refer to it certain forms that have been described in various genera. NOTHOPANAX FRUTICOSUM (L.) Miq. FI. Ind. Bat. F (1856) 765. Panax fruticosum L. Sp. PI. ed. 2 (1763) 1513. Polyscias fruticosa Harms in Engl. & Prantl. Nat. Pflanzenfam. 3’ (1894) 45. Commonly cultivated in and about the larger towns throughout the Philippines, probably of prehistoric introduction. Generally known as papua, literally “curled,” from the crisped leaves. An exceedingly variable plant, with several distinct varieties in cultivation here, which have received various horticultural names. NOTHOPANAX COCHLEATUM (Lam.) Miq. 1. c. 766. Aralia cochleata Lam. Encycl. 1 (1783) 224. Panax cochleatum DC. Prodr. 4 (1830) 255. Commonly cultivated, rarely or never flowering, locally known by its Spanish name, “platitos.” NOTHOPANAX ORNATUM (Bull.) comb. nov. Panax ornatum Bull. Cat. (1888) 9. Commonly cultivated, exceedingly variable in foliage. Of comparatively recent introduction here, occasionally flowering. NOTHOPANAX CRISPATUM (Bull.) comb. nov. Panax crispatum Bull. 1. c. Commonly cultivated, rarely or never producing flowers here. 111784 — 3 242 MERRILL. NOTHOPANAX GUILFOYLEl (Cogn. & March.) comb. nov. Aralia guilfoylei Cogn. & March. PI. Ornam. 2 (1874) t. 58. Commonly cultivated, but of comparatively recent introduction; probably a native of Polynesia. Rarely or never producing flowers here. APOCYNACEAE. TABERNAEMONTANA L. TABERNAEMONTANA SUBGLOBOSA sp. nov. Arbuscula glabra usque ad 5 m alta ; foliis chartaceis, oblongis, usque ad 18 cm longis, utrinque acuminatis, in siccitate plus minusve nitidis, subtus pallidioribus, nervis utrinque 13 ad 15, distinctis; floribus albis, corollae tube circiter 1.4 cm longo, calycis lobis orbiculari-ovatis, late rotundatis; folliculis in vivo aurantiacis, subglobosis vel ovoideis, vix carinatis, usque ad 4 cm longis, seminibus circiter 20. A glabrous shrub 5 m high or less. Branches terete, light- gray, rather slender, usually somewhat striate when dry, lenti- cellate. Leaves oblong, chartaceous, 9 to 18 cm long, 2.5 to 6.5 cm wide, the apex rather shortly blunt-acuminate, the base sharply decurrent-acuminate, the upper surface dark-colored and somewhat shining when dry, the lower surface much paler, usually somewhat brownish; nerves 13 to 15 on each side of the midrib, distinct beneath, curved upwards near the margins but not anastomosing, the reticulations rather indistinct ; petioles 0.8 to 2 cm long. Cymes in the upper axils, few-flowered, 5 cm long or less, the branches few, spreading, the pedicels 12 mm long or less. Calyx about 4 mm long, narrowed below, the lobes broadly orbicular-ovate, 2 to 2.5 mm long, the apex broadly rounded. Corolla white, the tube cylindric, about 1.4 cm long, the lobes about 1 cm in length. Anthers included, 2 mm long. Follicles usually paired, when fresh subglobose to ovoid, smooth, orange-red or red, somewhat fleshy, not keeled, when dry more or less wrinkled, up to 4 cm in length, each containing about 20 seeds, the seeds irregular, oblong, about 8 mm long. Luzon, Province of Bataan, Lamao River, Merrill 2511 (type), 3137, June, October, 1903; For. Bur. 629 Borden, with smaller flowers than in the type, April, 1904, Whitford 370, June, 1904, Williams 6, Elmer 6783, November, 1904, Leiberg 6H8, July, 1904: Province of Rizal, Phil. PI. 333, 3JtO Ramos; Merrill 1628, with smaller leaves than in the type: Province of Bulacan, For. Bur. 7186, 7193 Curran: Province of Tarlac, For. Bur. 9607 Zschokke: Province of Zambales, Merrill 2076, For. Bur. 916 Maule. Mindoro, Merrill 9Ji2, 1190, 223 If, For. Bur. 67 k3 Merritt. Masbate, For. Bur. 12608 Rosenbluth, For. Bur. 1695 Clark. THE FLORA OF MANILA. 243 A species well distinguished from other Philippine forms by its sub- globose fruits, but in other characters rather closely similating forms of Tabernaemontana pandacaqui Poir. It is probably most closely allied to T. sphaerocarpa Bl., but differs from that species in its smaller flowers and leaves. ASCLEPIADACEAE. TELOSMA Coville {Pergularia Auct., non L.) The genus Pergularia L. has been misinterpreted by most later authors. It is the same as the genus Daemia {Doemia) R. Br. The new name Telosma was proposed by Coville,” for Pergularia Auct., non L., with the transfer of a single species, Telosma odoratissima (Lour.) Coville. Later Mr. N. E. Brown ” proposed the new name Prageluria, but without the transfer of any of the species. Prageluria N. E. Br. is an exact synonym of Telosma Coville. TELOSMA PROCUMBENS (Blanco) comb. nov. Pergularia procumbens Blanco FI. Filip. (1837) 201; F.-Vill. Noviss. App. (1880) 134; Vid. Sinopsis Atlas (1883) t. 68, f. D. Pergularia glabra Blanco 1. c. ed. 2 (1845) 141, ed. 3, 1: 254; Naves 1. c. pi. 397. Pergularia filipes Schltr. in Perk. Frag. FI. Philip. (1904) 135; Merr. in Philip. Journ. Sci. 3 (1908) Bot. 329. Babuyanes Islands, Camiguin, Bur. Sci. U097 Fenix. Luzon, Province of Pangasinan, Bautista, Merrill, July, 1903: Province of Bulacan, Malinta, Bur. Sci. 6121 Robinson & Merritt, July, 1908: Province of Rizal, Bosoboso, For. Bur. 3196 Ahern’s collector, July, 1905, Bur. Sci. H73 Ramos, September, 1906; Masambong, Merrill, August, 1910; Montalban, Merrill, July, 1910: Province of Bataan, Lamao River, For. Bur. HU3 Ahern’s collector, August, 1904, For. Bur. 2105 Borden, November, 1904: Province of Cavite, Mendez Nunez, Bur. Sci. 1332 Mangubat, August, 1906. Min- danao, District of Zamboanga, Merrill 5^74, October, 1906. A widely distributed endemic species found at low altitudes. The oldest specific name is here adopted for there can be no doubt but that Pergularia filipes Schltr., described in 1904, is identical with Pergularia procumbens Blanco, described in 1837, and well figured by Naves in the third edition of the “Flora de Filipinas.” The species is not uncommon in thickets in the vicinity of Manila. TELOSMA ANGUSTILOBA (Warb.) comb. nov. Pergularia angustiloba Warb. in Perk. Frag. FI. Philip. (1904) 134. Pergularia accedens Vid. Phan. Cuming Philip. (1885) 127, Rev. PI. Vase. Filip. (1886) 189; F.-Vill. Noviss. App. (1880) 134, non Blume. The type of this species is Cuming 133i, in the Berlin Herbarium, and the specimen was collected in Cagayan Province, Luzon, according to Cuming’s own list of localities. In the original description the type is cited as “(Cuming) Warburg no. 1334)”; this is an error as the number refers to Cuming’s specimen, not to Warburg’s. I could find no specimen of this species in the Berlin Herbarium collected by Warburg. ”Contr. U. S. Nat. Herb. 9 (1905) 384. ” Kew Bull. (1907) 325. 244 MERRILL. CONVOLVULACEAE. IPOMOEA Linn. IPOMOEA REPTANS (Linn.) Poir. in Lam. Encycl. Suppl. 3 (1814) 460. Convolvulus reptans Linn. Sp. PI. (1753) 158, pro parte. This species is very common and widely distributed in the Philippines, but there is some doubt as to its proper specific name. The question is one of an interpretation of types only. Hallier f.“ states that the specimen of Convolvulus reptans in the Linnean Herbarium is Merremia caespitosa Hallier f. Dr. C. B. Robinson has kindly verified this for me, and confirms Hallier’s opinion; he also writes that the species is listed by Linnaeus in a manuscript catalogue of his herbarium (1755), as being then in his possession, so that there is every reason to believe that the specimen was in Linnaeus’ hands when he wrote the “Species Plantarum.” However, he assumed that a certain plate in Rheede’s “Hortus Malabaricus” was the same as the specimen in his herbarium, and in his descriptions cites the reference to Rheede first. Now Ballel Rheede Hort. Malabar. 1 1 : 107, t. 52 is unquestionably the plant usually known as Ipomea reptans Poir. (7. aquatica Forsk.), and this reference has been accepted by all botanists up to the present date as typifying the Linnean species. The sole question is, whether the specimen in the Linnean herbarium is the type of his Con- volvulus reptans, or is the species typified by the reference to Rheede’ work. The case seems to be very nearly balanced, but if an arbitrary rule be followed and the first reference to a plate be taken as typifying the species, then the reference to Rheede typifies the species; but on the other hand the specimen in the Linnean herbarium may be interpreted as the type, for Linnaeus undoubtedly possessed the specimen at the time the “Species Plantarum” was written. If the reference to Rheede be taken as the type, then no change of name is necessary, but if the specimen in the Linnean herbarium be taken as the type, then two changes are necessary, first the adoption of the binomial Ipomea aquatica Forsk., to designate the species now commonly known as I. reptans Poir., and secondly the transfer to Convolvulus reptans Linn, to Merremia, as the oldest specific name of Merremia caespitosa Hallier f. In order to avoid any change of name, the reference to Rheede has been accepted by me as typifying Ipomea reptans (Linn.) Poir. (7. aquatica Forsk.). MERREMIA Dennst. MERREMIA HIRTA (L.) comb. nov. Convolvulus hirtus L. Sp. PI. (1753) 159; Hallier f. Meded.s’ Rijks Herb. (1910) 21. Convolvulus caespitosus Roxb. Hort. Beng. (1814) 14, nomen, FI. Ind. 2 (1824) 70. Ipomoea linifolia Blume Bijdr. (1825) 721. Skinneria caespitosa Choisy Conv. Or. (1825) 105, t. 6, DC. Prodr. 9 (1845) 435. Merremia caespitosa Hallier f. in Engl. Bot. Jahrb. 16 (1893) 552. Ipomoea philippinensis Choisy 1. cc. 93, 367. This widely distributed and variable species is common in the Philippines, what is assumed to be the earliest valid specific name being here adopted. Meded.’s Rijks Herbarium (1910) 21. THE FLORA OF MANILA. 245 Hallier f.“ calls attention to the fact that Convolvulus hirtus L. is the same as Merremia caespitosa (Roxb.) Hallier f., after an examination of the specimen in the Linnean herbarium. Dr. C. B. Robinson has kindly examined the Linnean material, and writes that there are two specimens in the Linnean Herbarium under Convolvulus hirtus, the first named by Linnaeus, the second not named, but under it Linnaeus has written “Ind.”; this second specimen is the slender-leaved, glabrous form which has been described as Ipomoea philippinensis Choisy, and Dr. Robinson states that it might well have been a part of the specimen which Linnaeus named Con- volvulus reptans (see above under Ipomoea reptans Poir.). However, it is the first specimen that Linnaeus described, and as he wrote the name on it only, it must be the type. Dr. Robinson states that Linnaeus’ description is excellent, the specimen having much broader leaves than the second one, ovate to suborbicular, the inflorescence more developed, and the stems hirsute with spreading hairs. It is possible that more than one species is represented in what Hallier would call Merremia caespitosa and that the narrow leaved, glabrous form should be separated from the broad leaved form with hirsute stems. The evidence from specimens in this herbarium, however, seems to indicate that intergrades occur; for instance, some specimens are glabrous or nearly so, with linear leaves and 1- or 2-flowered peduncles; some have linear leaves, hirsute stems, and several-flowered peduncles; some have leaves varying from linear to oblong or oblong-ovate, with the base varying from acute to hastate; and some have mostly ovate leaves. None of the Philippine specimens agree with Linnaeus’ description as to leaves “cordato-subrotunda,” and none of our broader leaved forms have hirsute stems. VERBENACEAE. CLERODENDRON L. CLERODENDRON COM IVI ERSON 1 1 (Poir.) Spreng. Syst. Veg. 2 (1825) 758; Schauer in DC. Prodr. 11 (1847) 673; Miq. PI. Ind. Bat. 2 (1856) 882. Volkameria commersonii Poir. in Lam. Encycl. Bot. 8 (1808) 688 Volkameria nereifolia Roxb. FI. Ind. 3 (1832) 64. Clerodendron neriifolium Wall. Cat. (1829) no. 1789; Schauer 1. c. 660; Clarke in Hook. f. FI. Brit. Ind. 4 (1885) 589; Gamble in Journ. As. Soc. Beng. 74^ (1909) Extra Number 827. Clerodendron capsulare Blanco FI. Filip. (1837) 509, ed. 2 (1845) 355. Clerodendron inerme Auct. Philip., non Gaertn. This species is common and widely distributed in the Philippines along the seashore, and is enumerated here simply to call attention to the oldest specific name, provided the species is distinct from Clerodendron inerme Gaertn., as Clarke and Gamble have considered it. Poiret’s description most certainly applies to this form, although Clerodendron commersonii was considered by Schauer among the “species dubiae.” Gamble gives the range of Clerodendron neriifolium Wall. (C. com- mersonii Spreng.), as from the Malay Peninsula and Burma to the Phil- ippines, China, Australia, and Polynesia. In addition to the very extensive series of Philippine specimens in this herbarium, we have also the following: L. c. 246 MERRILL. Macao, Gaudichaud; Formosa, Henry, Kawakami & Nakahara 822; Celebes, Foxworthy 552; Caroline Islands, Yap, Volkens 2H. The type of Clerodendron commersonii (Poir.) Spreng. was from the Philippines, Poiret stating that it was collected by Commerson; like the other Philippine plants ascribed to Commerson as collector, it was in all probability collected by Sonnerat."" SCROPHULARIACEAE. VANDELLIA L. VANDELLIA VISCOSA (Willd.) comb. nov. Hornemannia viscosa Willd. Enum. (1809) 654. Vandellia hirsuta Ham. ex Benth. Scroph. Ind. (1835) 36, DC. Prodr. 10 (1844) 414; Hook. f. FI. Brit. Ind. 4 (1884) 280. This widely distributed species is not uncommon in the Philippines. It extends from India to southern China southward to Malaya. What is apparently its oldest valid specific name is here adopted. VANDELLIA PUSILLA (Willd.) comb. nov. Gratiola pusilla Willd. Sp. PI. 1 (1797) 105. Torenia hirta Cham. & Schlecht. in Linnaea 2 (1827) 571. Vandellia scabra Benth. Scroph. Ind. (1835) 36, DC. Prodr. 10 (1844) 414; Hook. f. FI. Brit. Ind. 4 (1884) 281. This very common and widely distributed species has several specific names older than the one adopted by Bentham and generally used by most authors. Selago pusilla Thunb. Prodr. PI. Cap. (1794-1800) 99, may be even earlier than Willdenow’s use of the same specific name, but this I am unable to determine from the literature available here. LIMNOPHILA R. Br. LIMNOPHILA MANILENSIS sp. nov. Herba annua, aromatica, erecta vel suberecta, 10 ad 50 cm alta, simplex vel paullo ramosa, leviter viscoso-hirsuta ; foliis sessilibus, oppositis, oblongo-lanceolatis ad lanceolatis, acutis vel obtusis, 1.5 ad 3 cm longis, margine serratis; floribus axillaribus, pedicellatis, solitariis vel in racemis axillaribus dispositis, calycis hirsutis, 5 ad 6 mm longis, profunde 5-fidi, laciniis lanceolatis, acuminatis; corolla circiter 9 mm longa; capsulis ovoideis, 3 ad 4 mm longis. An erect or ascending, annual, very aromatic herb 50 cm high or less, simple or sparingly branched, the branches, when present, short, all parts slightly viscid-hirsute with short scattered hairs. Stems rather weak, pale when dry, striate. Leaves opposite, sessile, oblong-lanceolate to lanceolate, 1.5 to 3 cm long, 4 to 8 mm wide, acute or obtuse, the margins distinctly serrate, the base usually more or less narrowed, both surfaces very sparingly Lamarck Encycl. 4 (1797) 387. THE FLORA OF MANILA. 247 hirsute with short scattered hairs, the lower one prominently glandular. Flowers solitary, axillary, pedicelled, or the lower ones usually in short, few-flowered, open, axillary, leafy racemes, the floral leaves similar to the others but much smaller, the racemes 1 to 5 cm long; pedicels hirsute, shorter than the calyx. Calyx hirsute, 5 to 6 mm long, deeply 5-fld, the lobes lanceo- late, acuminate. Corolla purplish, about 9 mm long. Capsule ovoid, 3 to 4 mm long. Luzon, near Manila, Merrill 7US2 (type), January, 1911, Phil. PL Merrill, December, 1910: Province of Rizal, Mariquina, Marave 150, January, 1895: Province of Bulacan, Norzagaray, Yoder 16, December, 1906. I suspect also that Loher 4339, of which a mere fragment is before me is referable here, as well as Vidal 3383 in herb. Kew., both of which have been named Limnophila diffusa G. Don. While in a broad conception of that species the Philippine specimens might be referred to it, still the differences are apparently sufficiently great to warrant distinguishing our local form from the Indian one. Limnophila manilensis is unquestionably closely allied to L. diffusa G. Don, but differs in its sessile, not petioled leaves, and very distinctly pedicelled flowers. LENTIBULARIACEAE. UTRICULARIA L. UTRICULARIA NIVEA Vahl Enum. 1 (1805) 203; Trimen FI. Ceyl. 3 (1895) 270. Utricularia racemosa Wall., var. filicaulis (Wall.) C. B. Clarke in Hook. f. FI. Brit. Ind. 4 (1884) 333. In old rice lands, La Loma, near Manila, Phil. PI. 761 Merrill, November, 1910. Not previously reported from the Philippines, and here recorded under its oldest specific name. The Philippine plant, however, has pale-purplish flowers rather than white, as in the type of Vahl’s species. India and Ceylon to southern China and Malaya. UTRICULARIA TENERRIMA nom. nov. Utricidaria scandens Oliver in Journ. Linn. Soc. Bot. 3 (1859) 181; C. B. Clarke in Hook. f. FI. Brit. Ind. 4 (1884) 332, non Benj. Luzon, La Loma, near Manila, Merrill 8041, September, 1911, in open wet grass lands, scattered, twining on the stems of grasses and other plants. Benjamin in the original diagnosis of Utricularia scandens described the slender twining form of U. wallichiana Wight, and attached the name to the corresponding specimens in Hooker’s herbarium, and also so named a distinct species mounted on the same sheet, as described by Oliver, 1. c. Oliver in his consideration of the Indian species of Utricularia discovered this fact and redescribed Utricularia scandens Benj., this time basing the description on the other plant, there thus being two different descriptions of Utricularia scandens Benj., based on two quite distinct plants. It is believed that Benjamin’s diagnosis should stand for the species as described by him, not as described by Oliver, and hence the necessity for a new name for Utricularia scandens Oliver, non Benj. 248 MERRILL. Utricularia tenerrima Merr. (C7. scandens Oliver, non Benj.) has previously been reported only from Madras, but the Philippine specimens appear to me to agree perfectly with the description as given by Oliver. ACANTHACEAE. PSEUDERANTHEMUM Radik. PSEUDERANTHEMUM PULCHELLUM (Hort.) comb. nov. Eranthemum pulchellum Hort. Gartenmag. (1810) 176, t. 17. Eranthemum hicolor Schrank. Hort. Monac. (1819) t. 8; Nees in DC. Prodr. 11 (1847) 456. Pseuderanthumum bicolor Radik, ex Lindau in Engl. & Prantl Nat. Pflanzenfam. (1895) 330, Perk. Frag. FI. Philip. (1904) 39. Common and widely distributed in the Philippines; Java, Timor, etc. STAUROGYNE Wall. STAUROGYNE RIVULARIS sp. nov. Herba erecta, plus minusve ramosa, usque ad 20 cm alta, partibus junioribus inflorescentiisque glanduloso-pubescentibus ; foliis breviter petiolatis, basi angustatis, anguste oblongo-obovatis vel oblanceolatis, 2 ad 6 cm longis, integris vel obscure undulato- crenatis: floribus brevissime pedicellatis ; corolla 7 ad 9 mm longa ; capsula oblonga, 5 mm longa. An erect herb, apparently from a perennial root, usually branched from the base, 10 to 30 cm high, the stems terete, slender, brownish-purple. Younger parts and inflorescence glandular-pubescent. Leaves narrowly oblong-obovate to oblan- ceolate, 2 to 6 cm long, 5 to 15 mm wide, base narrowed to the short petiole, apex acute or obtuse, entire or obscurely undulate- crenate, thin. Flowers solitary in the axils of the uppermost reduced leaves, forming a terminal, somewhat leafy, spike or spike-like raceme 1 to 5 cm long. Bracteoles glandular-hairy, green, oblanceolate, acuminate or acute, about 7 mm long, 2 mm wide. Calyx green, about 7 mm long, glandular-hairy, the upper lobe lanceolate, 1.6 mm wide, the other 4 linear, about 0.5 mm wide, free nearly to the base, subequal and a little shorter than the upper one. Corolla 7 to 9 mm long, tubular, sparingly pale-purplish or nearly white, the lower lip with distinct deep- purple stripes. Lips subequal, the upper one with 2 orbicular or somewhat reniform, rounded, 1.6 mm long lobes, the lower lip with 3 similar but narrower lobes, glandular at the tip. Stamens unequal, included, the filaments white-hirsute; con- nective very broad, the cells diverging downward, the anther when spread about 1.5 mm wide. Ovary ellipsoid, 1.5 mm long; style 4 to 5 mm long. Capsule oblong, cylindric or slightly THE FLORA OF MANILA. 249 compressed, sessile, 5 mm long, 2-celled, many seeded. Seeds very numerous, 0.2 mm long or less. Luzon, Province of Rizal, Masambong, along small streams on ledges subject to overflow in high water, Merrill 7396, December, 1910 (type), Phil. PI. 755 Merrill, January, 1911. A rather characteristic species, quite different from the few other forms of the genus at present known from the Philippines. STYLIDIACEAE. In Milbraed’s recent monograph of the family^® six genera and one hundred and twenty species are recognized, with the following range: Donatia with two species, one in Antarctic South America to Chile and one in New Zealand and Tasmania; Phyllachne with four species, one in Antarctic South America and three in New Zealand; Forsteria with four species in New Zealand and Tasmania; Oreostylidium with a single species in New Zealand; Levenhookia with 6 species in Australia, chiefly in the western part; and Stylidium with 103 species, nearly all confined to Australia, a few extending to Tasmania, and three extra-Australian species. These three belong in the section Andersonia, which has twelve species, of which nine are confined to Australia, one, S. uliginosum Sw., Australia (Queensland), southern China, and Ceylon, and two do not occur in Australia, S. tenellum Sw., India to Tonkin and Malacca, and S. kunthii Wall., confined to India. For some years a species has been known from the Philippines, but its occurrence here has not before defi- nitely been published. This species is of special interest as it adds another distinct Australian type to our knowledge of the Philippine flora, S. alsinoides R. Br., previously known only from Australia, forming with iS. tenerrhnum F. MuelL, the section Alsinoides. STYLIDIUM Sw. STYLIDIUM ALSINOIDES R. Br. Prodr. (1810) 572; DC. Prodr. 7 (1839) 337; F. Muell. Fragm. 1 (1858) 151; Benth. FI. Austral. 4 (1869) 24; Milbraed in Engl. Pflanzenreich 35 (1908) 40; Ewart, White and Wood in Proc. Royal Soc. Viet. N. S. 23 (1911) 299 (var. cordifolium) pi. 56. Luzon, Province of Nueva Vizcaya, Merrill 107, June, 1902: Province of Zambales, Hallier s. n.: Province of Rizal, Bur. Sci. 10898 Ramos: Ma- nila, Guerrero, s. n.; Loher 6U78: without definite locality, Loher 372i, 5188 (herb. Kew.). Growing in open wet grassy places, old rice paddies, etc. The determination of the Philippine material was, I believe, first made by Mr. Rolfe, but one specimen that I sent to Kew was so determined by Mr. Hemsley. Later I examined the material in the Kew herbarium and came to the conclusion that the determination was correct. Still more recently, through the kindness of Mr. F. Manson Bailey, of Brisbane, Dr. A. J. Ewart of Melbourne, and Mr. J. H. Maiden of Sydney, I have received a fine series of Australian specimens of Stylidium alsinoides R. Br., and also representative material, including a fragment of the type, of S. tenerrimum F. Muell. This material enables me to make a direct Pflanzenreich 35 (1908) 1-98. 250 MERRILL. comparison between the Philippine and Australian forms, and although the plants are not identical in all details, no characters have been noted that would warrant distingxiishing the Philippine form from the Australian. In our material the length of the capsule varies from 1.5 to 2 cm in length, while in the Australian specimens it is usually 2 cm long or a little longer. The vegetative and floral characters appear to be identical. It is confidently expected that eventually this species will also be found in Celebes and in New Guinea. The Australian material I have had for comparison is the following: Stylidium alsinoides R. Br. Mulgrave River, F. M. Bailey; Cairns, E. Betche, August, 1901; Rockingham Bay, Dallachy; Kimberley District, C. W. Nyulasy; Port Darwin, Holtze (var. cordifolia Ewart, White, and Wood). Stylidium tenerrimum F. Muell. North Australia, between Providence Hill and McAdams Range, F. Mueller, October, 1855; without definite locality. Rev. J. E. Tenison Woods. COMPOSITAE. BLUMEA DC. BLUMEA TENERA sp. nov. Herba erecta, vix vel pauce ramosa, tenera, usque ad 50 cm alta, parce pilosa vel subglabra; foliis oblongo-oblanceolatis, chartaceis vel subrigidis, subtus parce papilloso-pilosis, 2.5 ad 8 cm longis, acutis, basi angustatis, subsessilibus, margine dis- tanter dentatis, dentibus subrigidis, patulis, spiniformibus ; capitulis paucis, paniculatis, pedunculatis, 6 ad 7 mm longis, bracteis linearibus, parce pilosis ; pedunculis dense pilosis ; recep- taculis nudis; floribus hermaphroditis paucis, corolla 4-dentata. A slender, erect, simple or slightly branched herb 20 to 50 cm high, the stems and branches glabrous or nearly so, less than 1.5 mm in diameter. Leaves oblong-oblanceolate, rather rigid when dry, sessile or subsessile, 2.5 to 8 cm long, 5 to 17 mm wide, acute, base gradually narrowed, margins with rather distant, spreading, rigid, short, spine-like teeth, not lobed, the upper surface slightly hispid when young, scabrid in age, the lower surface with few, soft, white, spreading hairs from papil- late bases. Upper part of the stem leafless or nearly so. Heads few, scattered, paniculate, up to 10 in each panicle, peduncled, 6 to 7 mm long, the peduncles slender, densely pilose. In- volucral bracts green, linear, acute, the outer ones 1 to 2 mm long, the inner gradually longer, and the innermost 6 to 7 mm long, scarious, slightly hairy or in age nearly glabrous, less than 0.5 mm wide. Flowers yellow, the outer ones numerous, very slender, the corollas about 4 mm long, the styles slightly ex- serted; interior perfect flowers few, as long as the female ones. THE FLORA OF MANILA, 251 the corolla stouter, somewhat enlarged upwards, 4-toothed, the style not exserted. Achenes 1.2 mm long, linear-oblong or oblong, angled, slightly hairy; pappus hairs about 15, soft, slender, white, about 4 mm long. Disk glabrous. Luzon, Province of Rizal, between La Loma and Maypajo, near Manila, Merrill 7363 (type), December 31, 1910, in open grass lands a few meters above sea level, not common. Apparently referable here are Merrill 679 from Culion, and For. Bur. 587J^ Curran, from Zambales Province, Luzon, both small forms. A species apparently related to Blumea oxyodonta DC., differing in its more erect habit, the stems not dichotomous, only slightly pubescent and quite differently shaped leaves. ENHYDRA Lour. EN HYDRA FLUCTUANS Lour. FI. Cochinch. (1790) 511; Hook, f, FI. Brit. Ind. 3 (1881) 304; F.-Vill. Novis. App. (1880) 117. Luzon, Manila, in open wet grass lands, F. Espinosa, March 26, 1911. India to China southward to Malaya. This species was reported from the Philippines by F.-Villar, but the specimen cited above is the only one of the species to be recorded from the Archipelago since F.-Villar’s previously unverified record was made. ECLIPTA L. ECLIPTA 2IPPELIANA Bl. Bijdr. (1826) 914; DC. Prodr. 5 (1836) 490. Luzon, Manila, Merrill hh. May, 1902. Mindanao, District of Zam- boanga, Bur. Sci. 11739 Robinson, June, 1910; San Ramon, Hallier, Feb- ruary, 1904. This form, which is very much stouter and coarser than Eclipta alba (L.) Hassk., I have referred to Blume’s species, although with the short published descriptions of that species, it is impossible to determine with certainty whether or not the Philippine material is really referable here. Our material differs from the common Eclipta alba not only in its size and habit, but in being prominently hirsute with spreading hairs on all parts, notably on the younger stems, leaves, and petioles, in its broader, differently shaped leaves which are more prominently toothed, and in its much shorter peduncles, which rarely or never exceed 1 cm in length. Eclipta zippeliana Blume has been reported only from Java, and has been reduced by various authorities to Eclipta alba (L.) Hassk. WEDELIA Jacq. WEDELIA PROSTRATA (Hook. & Arn.) Hemsl. in Journ. Linn. Soc. Bot. 23 (1888) 434. Verbesina prostrata Hook. & Arn. Bot. Beech. Voy. (1841) 195. Wollastonia prostrata Hook. & Arn. 1. c. 265. Luzon, Manila, Merrill 751^9, April 12, 1911. In open grassy places about the walled city, apparently an introduced plant here. Japan to southern China and Formosa; not previously reported from the Philippines. The Philippine Journal of Science. C, Botany. Vol. VII, No. 4, September, 1912. DIE SPHAGNA DER PHILIPPINEN. Von C. Warnstorf. (Berlin-Schoneberg.) Obwohl die Torfmoose der zum Monsungebiet gehbrenden Provinz der Philippine!! in meinem 1911 im Pflanzenreich von Engler erschienenen Werke: “Sphagnales”^ nach Moglichkeit beriicksichtigt worden sind, so ist dennoch unsere Kenntniss dieser iiberaus schwierigen Pflanzengruppe schon jetzt nach kurzer Zeit nicht unerheblich erweitert worden, indem der Botaniker Herr Merrill in Manila die Giite hatte, mir aus dem dortigen Herbarium des “Bureau of Science” samtliche dort vorhandenen, von den Philippinen stammenden Torfmoosformen zur Untersuchung zuzusenden, wofiir ich ihm an dieser Stelle verbindlichst danke. Nach brieflicher Mitteilung dieses Herm kommen auf dieser Inselgruppe Sphagna nicht unter 1,400 m Meereshohe vor, sondern sie steigen auf der Nordinsel Luzon bis 2,700, auf der Siidinsel Mindanao bis etwa 2,250 m iiber dem Meere empor. Vertreten sind nach unserer gegenwartigen Kenntniss aus dem Gebiet nur 5 Gruppen : 1 Acutifolia, 2 Sericea, 3 Cuspidata, und 4 CymbifoUa mit je einer Species, wahrend 5 die Subsecunda durch zwei endemische Arten bekannt gewor- den sind. Es fehlen demnach Vertreter von folgenden Sek- tionen: 1 Truncata, 2 Polyclada, 3 Rigida, 4 Squarrosa, und 5 Mucronata. Am griindlichsten scheint Luzon durchforscht zu sein, da von dort die meisten Torfmoose bekannt sind. Die weiteste Ver- breitung findet dort das zu den Acutifoliis zahlende Sphagnum Junghuhnianum, von dem auf Mindoro nur zwei Fundorte zu unserer Kenntniss gelangt sind. Das zu der Sericeum Gruppe gehbrige, mir bis jetzt nur aus der Westmalayischen Provinz ^ Warnstorf, C. Sphagmales-Sphagnaceae. Engl. Pflanzenreich 51 (1911) 1-546, fig. 85. 253 254 WARNSTORF. (Sumatra u. Java) bekannte S. sericeum ist gegenwartig nur von Mindoro bekannt, wahrend das schbne charakteristische S. cuspidatulum, das ich bisher nur aus der Nordwest-, Siidwest- und Zentralmalayischen Provinz sah, an zwei Punkten der Sub- provinz von Benguet auf Luzon nachgewiesen werden konnte. Von den zwei endemischen Species der Subsecundumgruppe. S. luzonense und S. Robinsonii ist die letztere Art fiir die Wis- senschaft neu und soli unten ausfiihrlich beschrieben werden; beide kommen auf Luzon vor. S. japonicum var. philippinense endlich, der einzige Vertreter von den Cymbifoliis, bewohnt ebenfalls nur die Nordinsel. Es sind demnach im Ganzen in der Provinz der Philippinen nur sechs Artentypen von Torf- moosen aufgefunden worden, Dennoch zweifele ich nicht, dass es Botanikern, die die alpine Region der Nord- und Sudinsel besuchen, bei grosserer Aufmerksamkeit, die sie den in Rede stehenden Bryophyten zuwenden, gelingen wird, dort auch noch andere interessante Species dieser merkwiirdigen Moosgruppe aufzufinden, zumal, wenn sie sich zum Gesetz machen, von jedem einzelnen Sphagnumrasen, der ihnen bei ihren Explorationen aufstosst, Proben aufzunehmen. Sectio I. LITHOPHLOEA Russ. Subsectio 3. Cuspidata Schlieph. 1. SPHAGNUM JUNGHUHNIANUM Doz. et Mlkb. in Verhandel. Kon. Akad. Wetensch. Amsterd. (1854); Warnst. in Engl. Pflanzenreich 51 (1911) 114. Diese aus der Provinz der extratropischen Himalaya und der Provinz Yunnan, sowie aus dem Monsungebiet und dem temperierten Ostasien bekannte Species besitzt darnach eine sehr weite Verbreitung und ist, wie bereits hervorgehoben, auch auf den Philippinen, besonders auf Luzon, eine haufige Erscheinung. Fast alle mir aus dieser Monsunprovinz zugekom- menen Proben gehoren zu Var. a TYPICUM Warnst. in Engl. Pflanzenreich 51 (1911) 116. Luzon: Province of Abra, Mount Paraga, Bur. Sci. 7315 Ramos: Benguet Subprovince, Bur. Sci. 3^55 Mearns; Baguio, Elmer 8528; Mount Pulog, 2500 m. For. Bur. 16Jfl6 Curran, Merritt, & Zschokke; Pauai, circ. 2075 m, Merrill, Bur. Sci. -8678 McGregor: Lepanto Subprovince, Mount Data, 2200 m, Merrill ^919: Province of Laguna, Mount Banajao, Copeland. Mindanao: District of Davao, Mount Apo, Elmer 11369. f. DASYCLADUM Warnst. 1. c. Luzon : Province of Zambales, Mount Pinatubo Bur. S'ci. 25U0 Foxworthy : Benguet Subprovince, Pauai, 2135 m. Bur. Sci. 45i0 Mearns, Merrill i920, U869; Baguio, 1400 m. Bur. Sci. H052 Robinson. Mindoro: Mount Halcon, 2240 m, Merrill 5707. DIE SPHAGNA DER PHILIPPINEN. 255 f. GR AGILE Warnst. forma nova. Planta fere submersa, gracilior, ramorum fasciculis remotis; rami patentes tenuiores, recurvati, attenuati, laxe foliosi. Folia ramulina minora, 1.4-1. 5 mm longa, 0.7 mm lata, plerumque subito acuminata. Folia caulina permagna, salpe subito acu- Fig. 1. — Sphagnum Junghuhnianum var. typicum f. gracile Warnst. (a) Stamm- (6) Astblatt. minata, circum anguste limbata, 1.7-1. 8 mm longa, 0.9 mm lata, usque ad basim multifibrosa et utrinque multiporosa. (Fig. 1.) Luzon: Benguet Subprovince, Baguio, Elmer 8528; Mount Pulog 2700 m. For. Bur. 164.21 Curran, Merritt, & Zschokke. Var. PSEUDOMOLLE Warnst. 1. c. 117. Luzon : “Luzon central,” Loher. Subsectio 2. Sericea (C. Mull.) Warnst. 2. SPHAGNUM SERICEUM C. Miill. in Bot. Zeit. (1847) 481; Warnst. 1. c. 169. Mindoro: Mount Halcon, auf Felswanden, 2000 m, Merrill 6161. Diese schbne, trocken seidenglanzende, zarte Pflanze ist fiir das Gebiet neu und wird sich gewiss auch noch an anderen Standorten der alpinen Region nachweisen lassen. Subsectio 3. Cuspidata Schleph. 3. SPHAGNUM CUSPIDATULUM C. Miill. in Linnaea 38 (1874) 549; Warnst. 1. c. 186. Auch die vorliegende Species war bis jetzt von den Philippinen unbe- kannt, obgleich ihr Verbreitungskreis innerhalb des Monsungebiets dort ihr Vorkommen vermuten Hess. Dieser Verbreitungskreis ist, wie derjenige von S. Junghuhnianum, ein sehr ausgedehnter ; denn er erstreckt sich iiber die Nordwest-, Siidwest- und Zentralmalayische Provinz. Aus der Provinz der Philippinen ist mir nur folgende Form bekannt geworden: 256 WARNSTORF. Var. a MALACCENSE Warnst. 1. c. 187. Luzon: Benguet Subprovince, Mount Pulog, 2600 m, For. Bur. 16395, 16410 Curran, Merritt & Zschokke, Merrill 6402; Pauai, 2250 m, Merrill 4972. Subsectio 4. Subsecunda Schlieph. 4. SPHAGNUM LU20NENSE Warnst. in Bot. Centralbl. 76 (1898) 388; Engl. Pflanzenreich 51 (1911) 397. Luzon : Loher in Herb. Brotherus. Var. MACROPHYLLUM Warnst. 1. c. 398. Luzon: Lepanto Subprovince, Mounta Data, 2250 m, Merrill 4911. Var. SORDIDUM Warnst. 1. c. Luzon : Benguet Subprovince, 2300 m, Merrill 6678. 6. SPHAGNUM ROBINSONII Warnst. sp. nov. Planta cineracea, habitu et magnitudine fere S. compacto similis. Hyalodermis caulina stratis 2 composita. Cylindrus Fig. 2. — Sphagnum Rohinsonii Warnst. (a) Zwei Stammblatter ; (b) Zwei Astblatter ; (9) Astblattquerschnitt ; (st) Teil eines Querschnitts durch das Stammchen. lignosus aetate obscure fuscescens. Folia caulina magnitudine variabilia, lingulata, auguste limbata, apice rotundata, denticu- lata, 1-1.3 mm longa, 0.6-0.8 mm lata, sursum plus minusve fibrosa; cellulae hyalinae saepius septatae, interiore superficie DIE SPHAGNA DER PHILIPPINEN. 257 foliorum poris magnis multis in cellularum angulis, dorso fere tantum poro uno in angulis superioribus instructa. Ramorum fasciculi ramis 3-4; rami patentes baud parum dense foliosi. Folia ramulina late ovata, 1.3-1. 8 mm longa, 1-1.14 mm lata, con- cava, angustissime limbata, apice rotunda vix truncata denti- culataque, utrinque aequaliter multiporosa; pori veri plerumque bini ternive callularum angulis conjunctis, ei prope margines laterales majores; praeterea saepe pseudoporis in series breves ad commissuras dispositis instructa. Cellulae chlorophylliferae sectione transversali anguste rectangulae, ab utroque latere foliorum liberae (Fig. 2). Pflanzen sowohl nach Grosse als auch Habitus noch am mois- ten an nicht squarrose Formen von S. compactum erinnernd. Stammchen stark, ihre Hyalodermis rings 2-schichtig und aus weitlichtigen, diinnwandigen Parenchymzellen zusammengesetzt. Holzkbrper im Alter dunkelbraun, Stammblatter desselben Stengels in der Grosse veriinderlich, zungenformig, 1-1.3 mm lang und 0. 6-0.8 mm breit, an der abgerandeten Spitze meist gezahnelt und an den Seitenrandern schmal- oberwarts hyalin- gesaumt; im oberen Teile, zuweilen bis zur Mitte herab, fibres, die Hyalinzellen, z. T. septiert, auf der Blattinnenfiache mit vielen grossen ringlosen Ldchern in den Zellecken, die Riickseite der Blatter aber fast nur mit einzelnen kleinen Poren in den oberen Zellecken. Astbuschel 3- bis 4-astig und 2 starkere, kurz zugespitzte, mehr oder minder dicht- und rundbeblatterte Aestchen absteliend oder teilweis aufstrebend. Ihre Blatter breit-oval, hohl, mit abgerundeter, kaum oder undeutlich ge- stutzter, klein gezahnelter Spitze und sehr schmalem Randsaum, 1.5-1. 8 mm lang und 1-1.14 mm breit. Hyalinzellen reich- faserig und auf beiden Blattfiachen mit ziemlich zahlreichen, gleichliegenden Poren ; die wahren Locher allermeist zu 2 und 3 an den zusammenstossenden Zellecken, die gegen die Seiten- rander des Blattes allmahlich grosser werden. Ausser diesen wirklichen Poren finden sich hier und da noch zu kurzen Reihen verbundene Pseudoporen an den Commissuren. Chlorophyllzel- len im Querschnitt schmal rechteckig, centriert, und beiderseits freiliegend. Luzon: Benguet Subprovince, Baguio, 1440 m, Bur. Sci. 11996 Robinson. Diese Pflanze gehort in die nachste Verwandtschaft von Sphagnum sub- rufescens Warnst. in Engl. Pflanzenreich 51 (1911) 423. 111784 — 4 258 WARNSTORF. Sectio II. INOPHOLOEA Russ. Subsectio 5. Cymbifolia Lindb. 6. SPHAGNUM JAPONICUM Warnst. in Allgem. Bot. Zeitschr. (1895) 230; Warnst. in Engl. Pflanzenreich 51 (1911) 459. Var. PHILIPPINENSE Warnst 1. c. (1911) 521. Luzon: Province of Isabela, For. Bur. 16996, Alvarez. Es wiirde mich ungemein freuen, wenn vorstehender Artikel liber die Torfmoose der Philippinen dazu beitragen wiirde, dass die dortigen Bota- niker auf ihren Reisen diesen nach Bau und Lebensweise gleich interes- santen Brophyten ein erhohtes Interesse entgegen brachten und sie reichlich einsanimelten, wo sich aueh immer dazu Gelegenhet bote. Vol. VII, No. 3, including gages 125 to 208, was issued September 2, 1912. pubucahons fob sale by the bxtbeau of science, MANIIA, PHmPPINE ISLANDS THE SUBANUNS OP SINDANGAN BAY. By Emebson B. Chbistie. 121 pag««, 1 map, 29 plate*. Order No. 410, Paper, $1.25 United States eurrenoy, postpaid. Sindangan Bay is situated on the northern coast of Zamboanga Peninsula. The Subanuns of this region were studied by Mr. Christie during two periods of five and six weeks, respectively. The following is an abstract from the contents of Mr. Christie’s report on the Subanuns: Habitat and history; relations with the Moros; material culture; houses; - industries; trade; agriculture; family life; social customs; administration of justice; religion; the medicine man; ceremonies; tales; word-lists; physical measurements. ^ The 29 plates Illustrate the Subanuns at work and at play; their industries, houses, altars, and implements; and the people thenwelves. THE HISTORY OP StTLU. By Najeeb M. Sauseby. 275 pages, 4 maps, 2 diagrams. Order No. 406. Paper, $0.75 United States currency, postpaid. ; : In the preparation of his manuscript for The History of Sul u Doctor Saleeby spent much time and effort In gaining access to documents In the possession of the Sultan of Sulu. It is , fortunate that these records have now been translated and preserved in permanent form. This book 1* a history of the Moros in the Philippines from the earliest times to the American occupation. STUDIES PSr MOBO HISTORY, DAW, AND BEDIGION. / By Najeeb M. Sai.eeby. ■ , 107 pages, 16 plates, 5 diagrams. Order No. 405. Paper, $0.25; half morocco, $0.75 United States currency, postpaid. This Volume , deals with the earliest written records of the ' Moros in Mindanao. Doctor Saleeby was fortunately able to obtain exact copies of carefully preserved early records written in the Magindanao dialect with Arabic characters. The author presents translations of these as well as 16 half-tone illustrations of certain pages from the originals. Jhe names of the rulers of Magindanao are recorded in five folding diagrams. NEGRITOS OP ZAMBADES. By WIEEIAM Aeian Eeed. 83 pages, 62 plates. Order No., 402. Paper, $0.25; half morocco, $0;75 United States: currency, postpaid. The introduotory chapter deals with the general distribution of Negritos -and with the distribu- tion of the Philippine branch of the race. The succeeding chapters deal with the various indus- tries, amusements, and social relations-of these little men. - Plates from photographs, the greater part of which was taken for this publication, show ornaments, houses, men making fire with bamboo, bpws and arrows, danoei; and various types of the people' themselves. CONTENTS, BROWN, W. H. The Relation of Rafflesia manillana ^ to its Host ao^ MERRILL, E. D. Nomenclatural and Systetnatic Notes on the Flora of Manila ' - \*a7 WARNSTORF, C. Die Sphagna der PhiHppinen 253 The “iPhillppine Joumal of Science” is issued. as follows: eurrpitcy. Section A. Chemical and Geological Sciences and the Ihdastries;i < $2;0'Q Section B. The Philippine Journal of Tropical MedicSne ....i.vJui.': ' 3;00 Section €. Botany . 2.00 Section B. General Biology, Ethnology and Anthropology ’ (Sec^ i " tion D began with Volume V)^ - 2.00 Entire Journal, Volume II, III, IV, or V ... 5.00 Entire Journal, beginning with nVolume VI .....4.1'.:.-...,;. .........I.... > . 7.00 Single nximbers vSO Volume I, 1906 (not divided into, sections) : 10.00 Supplement to Volume ,I (botany) 8.50: Volume I (withont supplement) —..—i.,...;-—.:.. -4 ; O.OO Each section is separately paged a/nd indexed. , Publications sent in exchange for the Philippine Joumal of Science should be addlress^ : Library, Bureau of Science, Manila, P. I. Subscriptions may be sent to the Business MANAdER, Philippine .Jour^ naj of Science, Bureau of Science, Manilay F. 1., or to any of the agents' listed below ; The Macmillan Company, 64-66 Fifth Avenue, New Ydrk lJily, II, Sl: A, Wm. Wesley :& Son, 28 Essex Street, Strand, London, W. C., England;' Martlnus Nijhoff, Nobelstraat 18, The Hague, Holland. : , » . . v ' Mayer & Muller, Prinz Louis Fer din andstrasse 2, Berlin, N.W., Germany. Kelley & Walsh, Limited, 32 Baffles Place, Singapore, Straits Settlements. A. M. & J. Ferguson,. 19 Bailhe Street, Colombo, Ceylon. ; Thacker, Spink &: Co., P. O. Box 54, Calcutta, India, EntWdd art th* pod^offiM at Manila, P. I., as s«cond.oi)i*y.niiatt«r. November, 1912 ; THE PHII2PPINB JOURNAI. OF SCIENCE AtV^I J. COX, M. A., PH.D. ' ^ ‘ ’ GENERAL EDITOR ^ Section C. Botany E. D. MERRILL, H. S. , EDITOR / WITH^THB COOPERATION OP E. ’W. pRAFF, B.S.-, W. H. BROWN, Ph.D. H. N. WHITFOED, Ph. D. f TTBUCATIONS' ron $AIE Bt BtTEEAV Scicii” ' fl . ■|:|{: s ^: :: '-• BEPORT OF THE INTERNATIONAL ELAEUE CONFERENOE.! Held at Mukden, Apnl» 191-1, und^i* ±Ve au$pices of ' '"’ ' %' *- , the Chinese Government! ' ^ ^ Edited by Eeich MA»tim,’G-, F. fi^ETfilc, A^TSttia SxAiqU?rr,''^Airo,Rxo«Aja> *' 485 PM«$, 18 prl^tes (2 colored, 4 half-tones, 12 cb&rts and ?naps>.' j. j'' ' Order No. 416. Cloth, 85 50; Paper, $2 50 United ^ta^ea "turrehoy, postpaid, ; , The- proce^dinos of this: Internalional.' Gpnferenoe and Information aained' therefrom,' tod(l|lfar i ■ with the'resuUs of oeAain' baeteriotoglcal ' invesUoations, oonstitute tho'present report..' . : ’ . Nothing hitherto hhs.heeit, published which dives puoh a oomplete and ‘ffomPrebenptve'aocouWt;- ' of the entire sutyfepl of pbelimonio plague. , , ^ ' n - i i * * . ,: OeTegates from'- America (United^ Scales- of),, AustrIa>Hungary,>France,' G:erraaov,i Gredt'-jBrttatn,. v 1taly> .Japan, Mexico, the- Netherlands,. Russia,' and China. attended the Conference. ; .The Bureau of .Science of the. Goyernme'nt of the .Philippine .Islands has been -..appointed, tolh.-^' .adept fon the dlstnbution bf ,the .prJn.ted:.ptope;edlng» wf ,the International Flhgue: Conferohee- ■ TRE SUGAR industry, IN THE ISLAND OF NEGROS* " . 145 pages, lQ.p1atos, 1 mal>. ■ - Order Nd. 4^.' '' > , ^ ». Paper, $1.25 United States ourrenoy, postpaid. ;! 'V-.- -Considered from the .viewpoint practical -utility, ';.Mri, 'WaIKeTlsi Su.par -lndusti^dn the'tslahd of tlBgrds Is one of tha ihost IhFiportant papers published by the Bureau of -Splenoe* . This Volumb- ^; vis a 1:631 obntr.ibutlon to the subject; It.ls not.a.mere.DompIiationj.ior ths'authof was lit, the field y and understands ihe conditionS-.of .whiph he writes.;.: The follovytng; is^.a'.brief;.syn|Op8ls -df, tjje contents: - . - / ' . ' b » ' ' , " Tables of soli analyses, both chemical and physical; dnalyses'pf theeane, iuies ahd bagasse; > estimates based on ’adtual ' lnformatlon as to. the costs .of production apd of' oMltlvaliWir^hd esll- .; . mates -of the cost'-and tooatron. of possible central factories.; The island vis oonsideled- byisugar-:: S - producing dlsiri'ots; ..the area of cultivation and tha^productidiv Per. 'hec,tara;-.are> givens «pd 'tha r^ possibility for future'oxpansjon d scussed. ' ^ ' - 'The plates- illustrate' various phases. of sugar Industry'.from the o^ltiV^iioniof -we ^ ' a iS -4^ .A MANUAL OF PttlLIFFIiiiE ^ILK' CULTURE^ r By ChablesS. Batiks. -• ’ 53 pages, 20 plates. , . ^ ^ ^ ^ ^ ~ ^ , Order No. 413. ^ ' Paper, $0.75 Ohfted States ourrenoy^ postpaid. . The Silk Industry Is ^jarllcularly adapted to be underfak'en by persons with, small -Oapltal, and like the making of tfals In the Philipplno Islands- it should thrive;.with a little enoouraOemept . In A Manual of PhilloplneiSilk Culture we have presented th.^ results of .eeveml years?, actual, • work with silk-producing larva together with'.a description of-.Mie new .Philippine. race.: Half-tone .plates illuslrale, In :natural :eize;silkwor{ns In. different stages of development. Pup®;, adpit ' moths*;, samples of cloth .ipade from-. Art silk, -hand reeV and. silk- house. -: Other plates 'illpstrate tho^^' various •appllanoes' used In raising, silkworms and Tn spinning silk; hand and power reels ire': Hlustraled: working dcawlnge are given for a silk house and for a hand reel.: • - •'dfe m m THE PHILIPPINE Journal of Science C. Botany VOL. VII NOVEMBER, 1912 No. 5 NEW OR noteworthy PHILIPPINE PLANTS, IX. By E. D. Merrill. {From the Botanical Section of the Biological Laboratory, Bureau of Science, and the Department of Botany, University of the Philippines, Manila, P. I.) The last paper published under this title appeared in the year 1910,^ and the present contribution, like the preceding numbers of the series, is composed of descriptions of presumably new species, records of species previously described which have recently been discovered in the Philippines, notes on nomencla- ture, synonymy, etc- Two new genera are proposed, Freeria, of the Icacinaceae, and Macgregorianthus, of the Thymelaeaceae. Four genera, previously not reported from the Archipelago, are included, Parishia, Suriana, Cansjera, and Vallaris. One hun- dred presumably new species are described, all dicotyledons, while about 10 species, previously described from extra-Phil- ippine material, are recorded from the Archipelago for the first time. As is the case with the preceding numbers of the series, all the material on which the present paper is based is preserved in the Herbarium of the Bureau of Science. CHLORANTHACEAE. CHLORANTHUS L. CHLORANTHUS PH I LI PPI N ENSIS sp. nov. Species C. henryi Hemsl. ut videtur valde afRnis, differt flor- ibus multo minoribus, circiter 2.5 mm longis. An erect, simple, glabrous herb about 40 cm high, from aro- 113111 ’ This Journal, 5 (1910) Botany 167-257. 259 260 MERRILL. matic, woody rootstocks, the stems brown, angled and sulcate when dry, with about 5 nodes. Leaves 4, whorled at the apex of the stem, membranaceous, broadly obovate, 12 to 15 cm long, 8 to 9 cm wide, slightly shining, the lower surface paler than the upper one, the apex prominently and sharply acuminate, the base acute, margins sharply and rather finely serrate, the teeth more distant below, the basal margins quite entire ; nerves about 9 on each side of the midrib; petioles about 1 cm long. Inflorescence terminal, about 15 cm long, slender, the peduncle about 7 cm long, the branches opposite, usually 4 in two pairs, the lower ones about 6 cm long, the internode 2 cm long or less. Anther-scale about 2.5 mm long, divided nearly to the base into three lobes, the lateral lobes a little shorter than the middle one, oblong, obtuse, the middle one with a 4-locellate anther, the laterial ones with 2-locellate anthers ; anthers about 1 mm long, half as long as the lobes or less. Fruit unknown. Luzon, Province of Cagayan, Abulug River, Weber 1582, January, 1912, growing in forests, altitude about 250 m, the roots said to be fragrant, and the fruits white, although none of the latter were preserved. A species manifestly closely allied to Chloranthus henryi HemsL, of China, apparently differing chiefly in its smaller flowers, which are about one-half as large as in Hemsley’s species. In habit, general appearance, vegetative characters, etc., the Philippine plant very closely matches a specimen in the Herbarium of the Bureau of Science, representing Chloranthus henryi Herns!., collected by Farges in Su-tchuen; the only dif- ference appears to be in the flowers. I had at first determined the Luzon plant to be the same as Chloranthus oldhami Solms-Laub., a species known only from Formosa, and of which the flowers are unknown. That species, however, is described as having its leaves subsessile, which does not apply to our plant, nor are the leaves of our species crenate-dentate throughout, the lower 2 to 4 cm of the margins being quite entire. ULMACEAE. TREMA Lour. TREMA VULCANICA sp. nov. Frutex circiter 2 m altus subtus foliis ramulisque villosis; foliis numerosis, confertis, oblongo-ovatis vel oblongis, usque ad 5 cm longis, acutis vel acuminatis, basi leviter cordatis, supra scabridis, petiole circiter 2 mm longo; cymis brevissimis, vix 7 mm longis, floribus 5-meris. A shrub about 2 m high, the branchlets, petioles, and the lower surface of the leaves, especially on the nerves, rather promi- nently villous with rather short, pale, spreading or somewhat appressed hairs, the older branches terete, glabrous or nearly PHILIPPINE PLANTS, IX. 261 so, reddish-brown ; branchlets somewhat crowded, leafless in their lower parts, the leaves rather crowded above the middle. Leaves subcoriaceous, brittle when dry, brownish, oblong-ovate or oblong, 2.5 to 5 cm long, 1 to 2 cm wide, acute or somewhat acuminate, base somewhat cordate, equilateral or subequilateral, the upper surface scabrid, usually with few hairs along the midrib, the lower surface of about the same color as the upper, prominently pubescent with pale hairs on the midrib, lateral nerves, and primary reticulations, the ultimate reticulations close, evident on both surfaces, margins closely and finely coriaceous- serrulate ; lateral nerves about 5 on each side of the midrib, prom- inent, ascending, impressed on the upper surface ; petioles densely pale-villous, about 2 mm long. Cymes axillary, few- flowered, pubescent, 4 to 7 mm long. Flowers 5-merous. Male flowers: Sepals 5, elliptic to elliptic-ovate, concave, about 1.8 mm long, imbricate, glabrous except the slightly ciliate margins. Stamens 5. Pistillode cylindric, truncate, 1 mm long, glabrous, except the distinctly ciliate base. Female flowers similar to the males, the style-arms 1 mm long. Fruit reddish, 2.5 mm long, ovoid, fleshy, glabrous, more or less wrinkled when dry, the seed subglobose, about 1.2 mm in diameter. Camiguin de Mindanao, in thickets on slopes of the old volcano. Bur. Sci. H600 Ramos, April 27, 1912. A species well characterized by its depauperate size, small crowded leaves, short petioles and cymes, and other characters. It resembles somewhat Trema amboinensis Blume, and is undoubtedly allied to that species, but is quite distinct in many essential characters. LORANTHACEAE. LORANTHUS L. LORANTHUS CONFUSUS sp. nov. § Cichlanthus. Frutex parasiticus, ramis elongatis, tenuibus, ramulis subtus foliis inflorescentiisque indumento stellato-leproso pallido dense obtecto; foliis oppositis, breviter petiolatis, oblongis ad anguste elliptico-oblongis, rotundatis vel acutis, usque ad 5.5 cm longis; racemis brevibus, paucifloris; floribus 4-meris, corolla 12 mm longa. A slender parasitic shrub, the branches elongated, the branch- lets terete, about 1 mm thick, these, the lower surfaces of the leaves and the inflorescence densely covered with a pale, stellate- leprose indumentum, the tips of the branchlets sometimes fer- rugineous. Leaves opposite, coriaceous, oblong to narrowly elliptic-oblong, 2 to 5.5 cm long, 8 to 20 mm wide, apex rounded 262 MERRILL. or acute, base narrowed, acute, the upper surface, when young, stellate-leprose, soon becoming quite glabrous; nerves 3 or 4 on each side of the midrib, slender, not prominent; petioles 2 mm long. Racemes axillary, short, solitary or fascicled, 2- to 5- flowered, the rachis about as long as the petiole. Flowers 4- merous, somewhat curved. Pedicels about 3 mm long, the brac- teole ovate, concave, 0.5 mm long. Calyx 3.5 to 4 mm long, narrowed below into a pseudo-stalk, narrowly obovoid, truncate. Corolla 12 mm long, tube slightly inflated. Reflexed parts of the lobes above the insertion of the stamens 4 mm long, narrowly oblong to linear-oblong. Anther sessile, erect, 1 mm long. Fruit (immature) narrowly obovoid, 5 mm long, shortly stipitate densely stellate-leprose with a pale indumentum. Luzon, without definite locality, Cuming 1959 (type) : Province of Union, Elmer 5711, February, 1904, Bur. Sci. 12951 Fenix, December, 1910: Province of Pangasinan, Bur. Sci. U977 Ramos, December, 1907, Alberto 35, May, 1904 : Province of Tarlac, Merrill s. n., July, 1903. This form was previously considered by me ^ to be referable to Loranthus sphenoideus Blume, but I am now of the opinion that it is specifically dis- tinct. It differs from the Philippine material that has been referred to Loranthus sphenoideus in its pale, not ferruginous indumentum, its rela- tively much narrower, differently shaped leaves, and its somewhat smaller flowers. It is manifestly very closely allied to Blume’s species, but less closely allied to Loranthus estipitatus Stapf. LORANTHUS SIMILIS sp. nov. § Cichlanthus. Species praecedente affinis, differt foliis majoribus, petiole longioribus, indumento ferrugineo, floribusque longioribus, 1.8 cm longis. A slender parasitic shrub, the branches terete, brown, lenti- cellate, the branchlets densely brown-stellate-leprose, as are the petioles, lower surfaces of the leaves, and the inflorescence. Leaves elliptic to oblong-elliptic, coriaceous, 5 to 7 cm long, 2 to 3 cm wide, rounded, base somewhat narrowed, rounded or sub- acute, brown when dry, the upper surface glabrous, somewhat shining, the lower surface densely brown-stellate-leprose ; nerves 5 or 6 on each side of the midrib, rather distinct beneath, anas- tomosing; petioles 8 to 10 mm long. Racemes axillary, solitary or several at each node, few-flowered, the rachis about as long as the petioles, the pedicels 2 to 3 mm long, the bracteole concave, ovate, 1.3 mm long, all parts, including the outside of the flowers, densely stellate-leprose. Calyx subcylindric, 3 mm long, slightly or not narrowed toward the base. Corolla 18 mm long, slender. ’This Journal 4 (1909) Bot. 139. PHILIPPINE PLANTS, IX. 263 slightly curved, very slightly inflated in the lower one-half, the lobes 4. Luzon, Benguet Subprovince, Baguio, Williams 98^, July, 1904. A species allied to Loranthus confusus Merr., and to L. sphenoideus, differing from both in its longer flowers, and from the former also in its ferruginous indumentum, larger leaves, and longer petioles. Like the preceding species, the type was previously referred by me to L. sphenoideus Blume. LORANTHUS SUBSESSILIS sp. nov. § Dendrophthoe. Glaber, ramulis teretibus ; foliis oppositis, oblongo-ovatis, acuminatis, petiolatis, coriaceis, usque ad 15 cm longis; floribus sessilibus vel subsessilibus, e axillis defoliatis, solitariis vel binis, 6-meris, magnis, miniatis, circiter 7 cm longis, corolla falcata. A glabrous parasitic shrub, the branches up to 1 m in length, the branchlets terete, slender, grayish or reddish-brown, the ultimate ones 1 to 1.5 mm in diameter. Leaves opposite, the internodes 3 to 9 cm long, coriaceous, oblong-lanceolate, 8 to 15 cm long, 3 to 6 cm wide, acuminate, base narrowed, acute, rarely somewhat rounded, rather pale and somewhat shining when dry ; nerves 6 to 8 on each side of the midrib, obscure, slender, often nearly or quite obsolete; petioles about 5 mm long. Flowers in the axils of fallen leaves, large, red, sessile or subsessile, solitary or in pairs, each subtended by two bracteoles, the outer one broadly ovate, slightly apiculate, about 5 mm long, the inner one similar but truncate. Calyx about 7 mm long, 4 mm thick, slightly enlarged upward, truncate. Corolla 6.5 to 7 cm long, distinctly curved, the tube 4 to 4.5 cm long, about 3 mm in di- ameter at the base, gradually enlarged upward and 7 mm in diameter at the throat, the lobes 6, 4 mm wide at the base, soon narrowed, the reflexed part above the insertion of the stamens 2 cm long, lanceolate, acuminate, about 3 mm wide, much thick- ened. Filaments 10 to 11 mm long, erect; anthers linear-lan- ceolate, continuous, about 6 mm long. Mindanao, District of Zamboanga, Sax River, Merrill 8315, (type), November 28, 1911, altitude about 900 m; Williams 2^25, March 5, 1905, altitude about 600 m. A species characterized by its large flowers which are sessile or subsessile in the axils of fallen leaves, solitary or in pairs. When in full flower the plant is a very showy and striking one in the forests where it occurs. Its flowers are larger than in any other known Philippine species. LORANTHUS PUBIFLORUS sp. nov. § Dendrophthoe. Frutex inflorescentiis exceptis glaber, ramis ramulisque tere- tibus, crassis ; foliis crasse coriaceis, oblongis vel anguste oblongo- ovatis, usque ad 15 cm longis, verticillatis, petiolatis, acutis vel 264 MERRILL. acuminatis; floribus in triadibus fasciculatis in radicibus vel in ramulis specialibus, inflorescentiis brevibus, puberulis; floribus 4-meris, circiter 3 cm longis, tenuibus, extus puberulis. A parasitic shrub glabrous except the inflorescence, the branches and branchlets smooth, grayish or brownish, terete, stout, smooth, the apparently specialized flower-bearing branches (roots?) with a brown scaly bark. Leaves in whorls of three or four, the internodes 10 to 12 cm long, petioled, usually brown when dry, scarcely shining, thickly coriaceous, brittle, oblong to narrowly oblong-ovate, acute or somewhat acuminate, 12 to 15 cm long, 4 to 8 cm wide; petioles stout, 1 to 1.5 cm long; lateral nerves 4 to 6 on each side of the midrib, not prominent, the reticulations obsolete. Flowers in fascicled triads, the fascicles scattered, brown-puberulent, the inflorescences, excluding the corollas, less than 1 cm long. Bracteoles oblong-ovate, obtuse, puberulent, about 1.5 mm long. Calyx urceolate-campanulate, brown-puberulent, about 3 mm long, the limb somewhat spread- ing, about 1 mm long, truncate. Corolla slender, about 28 mm long, outside slightly puberulent, the lobes united for the lower 2 to 3 mm into a short tube, the free parts slender, about 1.5 mm wide, the reflexed part above the insertion of the stamen narrowly oblong, about 5 mm long, 1.3 mm wide, acute or obtuse. Anthers erect, continuous, narrowly oblong, about 3 mm long, the filaments 2 mm long. Luzon, Province of Laguna, near Paete, For. Bur. 13086 Curran, De- cember 17, 1911, in forests. A very characteristic species, distinguishable by its whorled, thickly coriaceous, petioled leaves, and its flowers puberulent externally and borne in fascicled short triads on the roots or on special leafless branches. In the scattered character of its inflorescences it resembles Loranthus mira- bilis Muell.-Arg. & Van Huerck, but is not, however, at all allied to that species. LORANTHUS CAPITULI FERUS sp. nov. § Lepiostegeres. Ramulis brunneo-furfuraceis exceptis glaber; foliis oppositis vel suboppositis, oblongo-ellipticis ad oblongo-ovatis, coriaceis, utrinque angustatis, petiolatis, usque ad 12 cm longis; capitulis axillaribus, solitariis, sessilibus, paucifloris; floribus circiter 1.8 cm longis, 6-meris. A glabrous parasitic shrub, the branches and branchlets stout, brown, the younger parts furfuraceous, more or less rugose when dry. Leaves opposite or subopposite, coriaceous, oblong- elliptic to oblong-ovate, olivaceous and slightly shining when dry,, 7 to 12 cm long, 2.5 to 5 cm wide, obtuse, base narrowed, PHILIPPINE PLANTS, IX. 265 acute or cuneate; lateral nerves 4 or 5 on each side of the mid- rib, indistinct, slender; petioles about 1 cm long. Heads axil- lary, solitary, sessile, but the flowers quite enclosed by imbricated bracts, forming a globose head. Flowers 6-merous, about 5 in each head, sessile. Calyx 2 mm long, truncate. Corolla 1.6 cm long, the lobes free nearly to the base, the reflexed part above the insertion of the stamens 4 mm long. Anther erect, sessile, 4 mm long. Mindanao, District of Zamboanga, Sax River Mountains back of San Ramon, Merrill 8270, November 28, 1911, parasitic on Englehardtia, altitude about 1100 m. Allied to Loranthus williamsii Merr., but with much shorter flowers. OLACACEAE. CANSJERA Juss. CANSJERA RHEEDII Gmel. Syst. Nat. 1 (1791) 280; Mast, in Hook. f. FI. Brit. Ind. 1 (1875) 582. Cansjera scandens Roxb. PI. Coromandel 1 (1795) 582. Cansjera malabarica Lam. Encycl. 3 (1791) 433. SiBUTU Island, Sulu Archipelago, Merrill 5291, October 13, 1906, in thickets along the seashore. India to southern China, through Malaya to northern Australia; the genus new to the Philippines. Gmelin’s speciflc name is here retained as being the one in common use. Lamarck’s Cansjera malabarica was, however, published during the same year, and it will be difficult if not impossible to determine which has priority. The species has previously been reported from the Philippines by Meisner,® but solely on Blanco’s error in interpreting the species. The species Blanco referred here is an Antidesma. MENISPERMACEAE. STEPHAN I A Lour. STEPHANIA RAMOSII Diels sp. nov. Caulis scandens glaber. Foliorum petiolus 4.5 ad 5 cm longus, lamina peltata, papyracea, supra fere lucida, subtus paulo pal- lidior, triangulari-ovata, apice acuta, acuminato-mucronulata, 8 ad 9 cm longa, 4 ad 5.5 cm lata. Inflorescentiae $ pseudo- racemosae, 5 ad 7 cm longae; pedicelli 1.5 ad 2 mm longi. Sep- ala 6, 3 exteriora anguste spathulato-oblonga, 1.5 mm longa, 0.4 mm lata, 3 interiora latiora, elliptico-ovata, concava, cir- citer 1.2 mm longa, 0.8 mm lata. Petala 5 conchiformia vel ®DC. Prodr. 14 (1857) 519. 266 MERRILL. fere orbicularia basi intus glandulosa, 0.7 ad 1 mm longa et lata. Synandrium 1 mm diametro. Luzon, Benguet Subprovince, near Baguio, Bur. Sci. 13U87 Ramos, May, 1911. Sect. Thamnothyrsa Diels. Species nova foliis triangulari-ovatis S. cato- sepalae Diels eiusdem provinciae indigenae baud absimilis sepalis petalisque duplo minoribus baud deflexis conspicue differt. PYCNARRHENA Miers. PYCNARRHENA CELEBICA (Boerl.) Diels in Engl. Pflanzenreicb 46 (1910) 53. Cocculus celebicus Boerl. Cat. PI. Bogor. (1899) 40. Luzon, Province of Nueva Ecija, For. Bur. 22157 Alvarez, December, 1910. Mindanao, District of Davao, Mount Apo, Elmer 12000, September, 1909. Previously known only from Celebes, and from plants cultivated in tbe Botanical Garden at Buitenzorg; tbe Philippine specimens agree closely with specimens in our herbarium from Buitenzorg, and have, morever, been determined by Doctor Diels. LIMACIA Lour. LIMACIA BLUMEI (Boerl.) Diels in Engl. Pflanzenreicb 46 (1910) 215. Cocculus blumei Boerl. Cat. PI. Bogor. (1899) 40. Mindanao, District of Davao, Mati, C. V. Piper U27 , May 15, 1911. The specimen closely matches typical material in our herbarium taken from cultivated plants in the Botanical Garden at Buitenzorg, and has, moreover, be.^n determined by Doctor Diels. Previously known only from specimens cultivated at Buitenzorg which originated in some part of the Malay Archipelago; new to the Philippines. ANONACEAE. MITREPHORA Hook. f. & Thoms. MITREPHORA WEBERI sp. nov. Arbor circiter 8 m alta partibus junioribus inflorescentiisque exceptis glabra; foliis ovatis vel oblongo-ovatis, subcoriaceis, nitidis, usque ad 18 cm longis, acuminatis, basi late rotundatis, nervis utrinque 8 ad 10; floribus hermaphroditis, circiter 4 cm diametro, petalis interioribus oblongis, acutis, extus pubescen- tibus, intus glabris, interioribus brevioribus, arcuatis, basi longe angustatis, glabris, laminibus densissime ciliato-villosis ; ovario glabro, ovulis circiter 12. A tree about 8 m high, glabrous except for the younger parts and the inflorescence. Branches slender, terete, grayish- or red- dish-brown, lenticellate, glabrous, the buds and very young leaves densely ferruginous-pubescent. Leaves ovate to oblong-ovate. PHILIPPINE PLANTS, IX. 267 subcoriaceous, shining, green and of about the same color on both surfaces, 10 to 18 cm long, 4 to 8 cm wide ; lateral nerves 8 to 10 on each side of the midrib, prominent, curved upward, anastomising, the reticulations lax; petioles 7 to 10 mm long. Flowers perfect, yellow and purple, extra-axillary or leaf-opposed, solitary, or at least but one flower opening at a time, the short rachis, the bracts, and the calyx externally densely ferruginous- pubescent. Sepals broadly ovate, acute or slightly acuminate, about 6 mm long. Outer three petals oblong, acute or somewhat obtuse, about 2 cm long, 8 mm wide, externally somewhat ap- pressed-pubescent with brown hairs, inside glabrous, the inner three petals about 14 mm long, vaulted, conniving, their claws slender, glabrous, about 9 mm long, their limbs about 5 mm long, subrhombic, but when spread much wider than long, the inner surface very densely ciliate-villous with long, dark-brown hairs. Stamens indefinite, narrowly oblong, 1 to 1.2 mm long, their connectives truncate, overlapping. Ovaries 12 to 15, glabrous, 1 to 1.2 mm long, oblong; ovules about 12, 2-seriate; styles clavate-oblong. Fruits unknown. Busuanga (Calamianes Islands), C. M. Weber 1550bis, February 25, 1911, in forests, limestone region, altitude about 30 meters. Apparently a very characteristic species, distinguishable by the inner faces of the limbs of the interior petals being very densely ciliate-villous with long, dark-brown hairs, and by its outer petals being quite glabrous inside. OXYMITRA Hook. f. & Thoms. OXYMITRA PUBESCENS sp. nov. Frutex scandens omnibus partibus plus minusve ferrugineo- pilosis, ramulis densissime pilosis; foliis oblongis vel anguste oblongo-obovatis, abrupte brevissime acuminatis, basi leviter cordatis, subcoriaceis vel chartaceis, usque ad 24 cm longis, nervis utrinque circiter 10, distantibus, distinctis; petiolo crasso, vix 6 mm longo; floribus ignotis; fructibus ovoideis vel ellipsoideis, circiter 8 cm longis, aurantiacis, ferrugineo-pilosis. A scandent shrub, all parts more or less ferruginous-pilose, the branchlets very densely so, the older branches bearly black, lenticellate, becoming glabrous. Leaves subcoriaceous or firmly chartaceous, oblong or narrowly oblong-ovate, 14 to 24 cm long, 5.5 to 8.5 cm wide, usually widest above the middle, the apex shortly and abruptly acuminate, below usually more or less nar- rowed, the base rounded, somewhat cordate, both surfaces more or less pilose with ferrugenous hairs, the upper surface in age becoming more or less glabrous, the lower surface slightly paler. 268 MERRILL. the nerves and reticulations especially pilose; lateral nerves 10 on each side of the midrib, prominent, somewhat ascending, anastomosing, the reticulations subparallel, distinct ; petioles densely ferruginous-pilose, thickened, 6 mm long, or less. Flow- ers axillary, sessile or very shortly pedicelled, apparently solitary (not seen). Fruits umbellately arranged on stout, axillary, pubescent tubercles, 6 or more in each umbel, their pedicels pubescent, 5 to 8 mm long, the fruits yellow when fresh, dark- brown when dry, ovoid or ellipsoid, about 8 mm long, more or less ferruginous-pilose, the apex minutely apiculate. Luzon, Province of Cagayan, Abulug River, Bur. Sci. 13870'' Ramos, January 26, 1912, on trees in forests. A species -well characterized by its pilose, ferruginous indumentum and comparatively few, distant, lateral nerves. POLYALTHIA Blume. POLYALTHIA LOHERI sp. nov. Species P. lanceolatae Vid. similis et affinis, differt foliis ma- joribus, tenuiter acuminatis, floribus majoribus, circiter 2 cm longis. A tree 5 to 10 m high, nearly glabrous, the axils bearded with short, stiff, ferruginous hairs, the branchlets slender, terete, dark-colored when dry, usually lenticellate. Leaves chartaceous or subcoriaceous, glabrous, rather pale when dry, shining and of the same color on both surfaces, lanceolate to oblong-lanceo- late, 15 to 25 cm long, 2 to 5 cm wide, gradually narrowed into a long and rather slender acumen, base rounded or narrowed, minutely cordate, often very slightly inequilateral ; primary nerves about 19 on each side of the midrib, spreading, anas- tomosing, the secondary ones often nearly as prominent, the reticulations rathei\ lax, evident; petioles 3 mm long or less. Flowers axillary, solitary, perfect, white, cream-colored, or yellow, the pedicels slender, somewhat thickened upward, 3 to 4 cm long. Sepals triangular-ovate, acute or slightly acu- minate, 3 to 4 mm long. Petals 6, subequal, oblong to oblong- obovate, thick, coriaceous, somewhat pubescent especially along the margins toward their tips, obtuse or acute, somewhat united below, 2 to 2.5 cm long, 4 to 7 mm wide. Anthers numerous, the connectives truncate. Ovaries villous. Fruit ellipsoid, apparently red, about 2 cm long, glabrous or nearly so. Luzon, Province of Cagayan, Cuming , 13 Culasing Mountain, For. Bur. 18Jt37 Alvarez (type), February 22, 1909, in dipterocarp forests; Pamplona, For. Bur. 1U780 Darling, March, 1909: Province of Rizal, San PHILIPPINE PLANTS, IX. 269 Mateo, For Bur. 1103 Ahern’s collector, May, 1904, in fruit; Montalban, Loher 5531, January, 1906, with immature flowers: Province of Tayabas, Mauban, For. Bur. 10191 Curran, March, 1908. The alliance of this species is manifestly with Polyalthia lanceolata Vid., and some of the material cited above has been distributed under that name. P. loheri, however, has much larger, differently shaped leaves, and much larger flowers, and is abundantly distinct. Var. CAGAYAN ENSIS var. nov. A typo differt ramulis pallidis, foliis majoribus, usque, ad 35 cm longis, venis magis distinctis. Luzon, Province of Cagayan, Abulug River, near Dabba, Bur. Sci. 139^8 Ramos, February 2, 1912, in forests, a tree about 5 m high with yellow flowers. Additional material may show that this form is specifically distinct, or may merge it with Polyalthia loheri. POLYALTHIA TENUIPES sp. nov. Ut videtur arbor parva, ramulis junioribus exceptis glabra; foliis lanceolatis, sensim angustatis, acuminatis, basi angustatis, obtusis, minute cordatis, usque ad 12 cm longis, nervis utrinque 10 ad 12; floribus in axillis superioribus, solitariis, tenuiter pedicellatis, circiter 8 mm longis, pedicellis 2 ad 3 cm longis; petalis exterioribus triangulari-ovatis, interioribus lanceolatis quam exterioribus duplo longioribus. Apparently a small tree, nearly glabrous, the growing shoots densely appressed ferruginous-pubescent, the branches slender, glabrous, dark-colored, terete. Leaves lanceolate, chartaceous, 9 to 12 cm long, 1.5 to 2.5 cm wide, gradually narrowed above to the slenderly acuminate apex, the base somewhat narrowed, obtuse, slightly cordate, of nearly the same color and shining on both surfaces when dry; primary nerves 10 to 12 on each side of the midrib, spreading, anastomosing, irregular, the secondary ones often nearly as prominent as the primary ones, the reticulations lax; petioles 2 mm long or less. Flowers soli- tary, perfect, axillary, in the upper axils, about 8 mm long, their pedicels very slender, 2 to 3 cm long, with a small oblong- lanceolate bracteole near the base. Sepals slightly pubescent, triangular-ovate, acute, about 2 mm long, united. Petals 6, in two series, dissimilar, the outer ones triangular-ovate, acute or somewhat acuminate, 5 mm long or less, the inner three lanceo- late, somewhat acuminate, about 8 mm long. Luzon, Province of Tayabas, Kabibihan, Bur. Sci. 13U72 Foxworthy & Ramos, March, 1911. A species with vegetative characters quite similar to those of Polyalthia lanceolata Vid., but sharply differentiated by its dissimilar petals. 270 MERRILL. CAPPARIDACEAE. CAPPARIS L. CAPPARIS LOHERI sp. nov. Frutex scandens, armatus, subtus foliis ramulisque pubescen- tibus; foliis papyraceis, concoloribus, lanceolatis ad late lanceo- latis, brevissime petiolatis, basi rotundatis, apice sensim angus- tatis, acuminatis, nervis utrinque circiter 11, distinctis, anas- tomosantibus ; floribus longe pedicellatis, 4 ad 6 in umbellis axillaribus. A scandent shrub, the nodes armed with short, sharp, straight spines 2 mm long or less, the branchlets terete, brownish, pubescent. Leaves lanceolate to broadly lanceolate, rarely ovate- lanceolate, chartaceous, 9 to 19 cm long, 2 to 4.5 cm wide, green and of about the same color on both surfaces when dry, slightly shining, the upper surface quite glabrous or the midrib slightly pubescent, the lower surface softly pubescent at least on the midrib and nerves, base rounded, gradually narrowed up- ward from about the middle to the long and slender acumen, sometimes slightly falcate ; lateral nerves about 11 on each side of the midrib, spreading, prominent, looped-anastomosing, the recticulations lax, distinct; petioles pubescent, about 2 mm long. Umbels axillary, solitary, the peduncles pubescent, slender, 1 to 1.5 cm long, 4- to 6-flowered, the pedicels 2 to 3 cm in length, slender. Sepals 5 to 6 mm long, concave, orbicular-elliptic, rounded. Petals hyaline, thin, oblong, rounded, about 3.5 mm long, base truncate, margins ciliate. Stamens numerous, their filaments about 2 cm long; anthers narrowly oblong, 2 mm long. Gynophore slender, 3 cm in length in anthesis, elongated and 3.5 to 5 cm in length in fruit; ovary narrowly ovoid. Fruit (immature) ellipsoid, fleshy, 1.8 cm long. Luzon, Province of Rizal, Montalban, Loher 2016 August, 1890, in herb. Kew., 6787 (type), November, 1905. A species well characterized by its long and slenderly acuminate pubescent leaves, its axillary, solitary, few-flowered umbels, its long- pedicelled flowers, and greatly elongated gynophore. It is allied to Capparis lobbiana Turcz., but is very different from that species in vegetative and many other characters. CAPPARIS LITTORALIS sp. nov. Species C. lobbianae Turcz., similis et affinis, differ! omnibus partibus glabris, vix molliter pubescentibus. An erect (or somewhat scandent?) shrub about 2 m high. PHILIPPINE PLANTS, IX. 271 the branchlets terete, slightly pubescent, the nodes with very short sharp spines 1.5 mm long or less. Leaves ovate to oblong- ovate, 1.5 to 5 cm long, 1 to 2.5 cm wide, coriaceous, shining, glabrous, acute or very slightly apiculate-acuminate, base broad, cordate ; nerves about 8 on each side of the midrib, very slender, not at all prominent; petioles 2 mm long or less. Flowers (not seen) in axillary, solitary, apparently few-flowered umbels, the peduncle in fruit 1.5 to 2 cm long, slender, glabrous, the pedicels (gynophore) 1.5 to 2.5 cm in length. Fruit ovoid or subglobose, about 8 mm in diameter, white when fresh, brown when dry, apparently fleshy, 2- or 3- seeded, the seeds 3 to 4 mm in length. Luzon, Province of Tayabas, For. Bur. 11111 Curran (type), April, 1908, on sand dunes along the seashore: Province of Ilocos Norte, Bur. Sci. 7602 Ramos, March, 1909, in sandy places near the sea. A species manifestly very closely allied to Capparis lobbiana Turcz., which it closely resembles, differing in having entirely glabrous, shining, coriaceous or subcoriaceous, not softly pubescent leaves. SAXIFRAGACEAE. POLYOSMA Blume. POLYOSMA PIPERI sp. nov. Frutex vel arbor parva ; foliis anguste oblongo-obovatis vel late oblanceolatis, usque ad 16 cm longis, integris, apice acuminatis, basi angustatis, acutis, oppositis, chartaceis, subtus ad costa ner- visque plus minusve hirsutis, nervis utrinque circiter 12, promi- nentibus; racemis terminalibus, pubescentibus, quam folia lon- gioribus; floribus 4-meris, circiter 12 mm longis. A shrub or a small tree, the branches terete, glabrous, the younger ones nearly black when dry and more or less pubescent with short, spreading, often fulvous hairs. Leaves opposite, narrowly oblong-obovate to broadly oblong-oblanceolate, charta- ceous, 12 to 16 cm long, 3.5 to 5 cm wide, entire, the apex rather prominently acuminate, gradually narrowed from about the middle to the acute base, the upper surface glabrous or nearly so, the lower of about the same color, more or less fulvous or pale-pubescent along the midrib and lateral nerves; lateral nerves about 12 on each side of the midrib, prominent, curved, anastomosing, the reticulations rather lax; petioles 1.5 to 2.5 cm long, pubescent. Racemes terminal, more or less hirsute- pubescent with short, pale or grayish hairs, a little longer than the leaves, erect. Flowers numerous, about 12 mm long, their 272 MERRILL. pedicels 2 to 2.5 mm long, the bracteoles lanceolate, acuminate, pubescent, about 3 mm long. Calyx somewhat urceolate, about 2.5 mm long, slightly constricted at the apex, the 4 teeth spread- ing, triangular-ovate, acuminate, short. Petals 4, linear, obtuse, 10 to 11 mm long, to 1.5 mm wide, longitudinally 5- or 6-nerved, the nerves distinct, stout, dark-colored, outside more or less appressed-pubescent with short hairs, inside ciliate-pilose with white hairs. Filaments 6 mm long, ciliate-pilose ; anther linear, 4 mm long. Fruit (rather immature) ovoid-ellipsoid, glabrous, somewhat longitudinally sulcate when dry, about 12 mm long, 6 to 7 mm wide, somewhat narrowed at both ends. Mindanao, Province of Surigao, Hinatuan, C. V. Piper 519, May 16, 1911. A species distinguishable by its leaves being somewhat pubescent beneath, its long racemes, and relatively short flowers. POLYOSMA VERTICILLATA sp. nov. Arbor parva 4 ad 5 m alta, partibus junioribus inflorescen- tiisque parce pubescentibus ; foliis oblongis, coriaceis, ternis, verticillatis vel subverticillatis, nitidis, acuminatis, 4 ad 10 cm longis, irregulariter glanduloso-dentatis ; racemis axillaribus, multifloris; floribus purpureis, vix 1 cm longis. A small tree 4 to 5 m high, nearly or quite glabrous except the sparingly pubescent inflorescence and younger parts. Leaves in whorls of threes, coriaceous, when young somewhat pubescent, when mature quite glabrous, shining, usually oblong, 4 to 10 cm long, 1.5 to 3.5 cm wide, rather coarsely and irregularly glandular-dentate, sometimes subentire, acuminate, base acute; nerves 15 to 20 on each side of the midrib, spreading at nearly right angles, slender, anastomosing; petioles 1 to 2.5 cm long. Racemes axillary, slightly pubescent, peduncled, 4 to 5 cm long. Flowers numerous, crowded in the upper one-half of the ra- ceme, their pedicels 2 mm long, pubescent, with 3 oblong, acute, 1 mm long bracts at the apex subtending the calyx. Calyx pubescent, somewhat urceolate, 2.5 mm long, the lobes 4, trian- gular-ovate, acute, 1 mm long. Petals 4, narrowly oblong, 8 mm long, 1.5 mm wide, slightly appressed-pubescent with very short, scattered hairs. Fruit subglobose to ovoid, black when mature, about 8 mm long. Luzon, Benguet Subprovince, Mount Tonglon (Santo Tomas), Phil. PI. 772 Merrill, May, 1911 (type), Williams 1526, November, 1904, Bur. Sei. 51^11 Ramos, December, 1908; Mount Ugo, Bur. Sei. 5593 Ramos; Mount Pulog, For. Bur. 18H8 Curran, Merritt, & Zschokke; Baguio, Elmer 8798. A very characteristic species, readily recognizable by its leaves being PHILIPPINE PLANTS, IX. 273 in whorls of threes, not opposite as in most species of the genus; it differs from Polyos7na philippinensis also in its considerably shorter and less pubescent flowers, as well as in its usually more prominently dentate leaves. LEGUMINOSAE. DESMODIUM Desv. DESMODIUM TRIFOLIASTRUM Miq. FI. Ind. Bat. F (1855) 248. Mindanao, Province of Surigao, Hinatuan, C. V. Piper 549, May 16, 1911. This species has not been reported from the Philippines previously, and, in fact, is definitely recorded only from Java. The species has been deter- mined through the kindness of Sir D. Prain, director of the Royal Gardens, Kew, to whom a specimen was sent for comparison with the material in the Kew Herbarium. DERRIS Lour. DERRIS CEBUENSIS sp. nov. § Dipteroderris. Frutex scandens, glaber; foliis 13 ad 20 cm longis, foliolis 7, nitidis, oblongis vel oblongo-ellipticis, in siccitate pallidis, utrin- que concoloribus, 4 ad 8 cm longis, breviter obtuse acuminatis; floribus ignotis; leguminibus oblongis vel oblongo-lanceolatis, 6 ad 10 cm longis, planis, 2.5 ad 3 cm latis; seminibus 2 vel 3. A scandent vine, glabrous throughout (flowers unknown), the stems brownish-olivaceous, prominently lenticellate, terete, the young branchlets smooth. Leaves odd-pinnate, 13 to 20 cm long, the common petiole and rachis 7 to 11 cm long. Leaflets 7, coriaceous or subcoriaceous, pale, shining and of the same color on both surfaces when dry, oblong to oblong-elliptic, 4 to 8 cm long, 2.5 to 3.5 cm wide, the base obtuse or rounded, apex shortly and obscurely blunt-acuminate, or sometimes obtuse; nerves slender, obscure, anastomosing, 7 to 10 on each side of the midrib; petiolules 3 to 8 mm long. Flowers unknown. In- frutescence axillary, paniculate, up to 25 cm long, entirely glabrous, the lower branches up to 8 cm long, the upper ones shorter. Persistent calyces shallowly cup-shaped, truncate, 5 mm in diameter. Pods thin, flat, oblong to oblong-lanceolate, pale when dry, 6 to 10 cm long, 2.5 to 3 cm wide, the base acute or acuminate, apex shortly and slightly falcately apiculate- acuminate, the wings thin, 3 to 7 mm wide, of the same width on both sides of the pod. Seeds 2 or 3, rarely only 1, thin, flat, 2 mm long. Cebu, Bucacao, Bur. Sci. 11014 Ramos, March 19, 1912, on dry hills. A species apparently most closely allied to Derris mindorensis Perk., and to D. micans Perk., but distinguishable by being entirely glabrous throughout. 274 MERRILL. MEDICAGO L. MEDICAGO LUPULINA L. Sp. PI. (1753) 779; Baker in Hook. f. FI. Brit. Ind. 2 (1876) 90. Luzon, Benguet Subprovince, Baguio, Merrill 7762, May, 1911, in waste places, open grassy slopes, altitude about 1500 m. Introduced and appar- ently established, as is Medicago denticulata Willd. Widely distributed in Europe and Asia, introduced in North America. SIMARUBACEAE. SURIANA L. SURIANA MARITIMA L. Sp. PI. (1753) 284; A. Gray Bot. Wilkes U. S. Explor. Exped. (1854) 307. Lumbucan Island, Sulu Sea, Phil. PI. ^03 Merrill, September, 1910. Mangsee Island, Wilkes Expedition. This species has already been reported from the Philippines by A. Gray, specimens having been collected by the Wilkes Expedition on Mangsee Island, a short distance south of Lumbucan this record having been overlooked by me in my paper on the Philippine plants collected by the Wilkes Expedition.'' It is widely distributed in the tropics of the World, but its occurrence in the Old World is rather peculiar and apparently restricted. I have found no published record of its occurrence on the Asiatic continent, nor in Formosa. I have specimens collected on Pratas Island, in the China Sea between Hongkong and Luzon, sent to me by Mr. Kawakami of Taihoku, Formosa. Both Miquel and Boerlage record it as Malayan, but without citing any locality or localities for it. It is found in the Laccadive Islands, in German New Guinea, and on the islands off the north-east coast of Australia, while it seems to be widely distributed in southern Polynesia. MELIACEAE. AGLAIA Lour. AGLAIA BRACHYBOTRYS sp. nov. § Euaglaia. Arbor circiter 8 m alta, ramilis subtus foliis inflorescentiisque minute subferrugineo-ciliato-lepidotis ; foliis alternis, circiter 40 cm longis, foliolis membranaceis, alternis, circiter 7, oblongo- ovatis vel oblongo-ellipticis, usque ad 17 cm longis; paniculis lateralibus, circiter 9 cm longis; floribus racemose dispositis, brevissime pedicellatis, 4-meris, tubo stamineo cum petalis di- stincte coalito. A tree about 8 m high. Branches pale-brown, terete, densely covered with minute ciliate-lepidote scales. Leaves alternate, unequally pinnate, about 40 cm long, the petiole, rachis, and ‘This Journal 3 (1908) Bot. 73-84. PHILIPPINE PLANTS, IX. 275 petiolules densely covered with pale-brown, lepidote, ciliate scales. Leaflets alternate, usually 7, membranaceous, ovate-oblong to oblong-elliptic, 12 to 17 cm long, 5.5 to 8 cm wide, slightly shining when dry, glabrous above, beneath of the same color and ciliate-lepidote on the nerves and midrib, and with scattered, small scales on the surface, the base of the lateral ones inequi- lateral, somewhat acuminate, of the terminal one equilateral, the apices somewhat acuminate; lateral nerves 8 to 12 on each side of the midrib, prominent; petiolules of the lateral leaflets 1 cm long, of the terminal one about twice as long. Panicles lateral, about 9 cm long, the peduncle less than 1 cm long, somewhat dichotomously branched, the branches spreading, all parts covered with small, pale-brown, ciliate-lepidote scales. Flowers about 3 mm long, racemosely rather densely arranged on the ultimate branchlets, their pedicels 0.5 mm long or less. Calyx ciliate- lepidote, when spread nearly 2 mm square, the teeth 4, rarely 5, broadly triangular, acute, less than 0.5 mm long. Petals 4, glabrous, elliptic-oblong or oblong-obovate, obtuse, about 3 mm long, 2 mm wide, adherent to the staminal-tube in the lower one-third. Staminal-tube 2.5 mm long, truncate, glabrous, bear- ing 7 or 8 stamens attached at or just above the middle, the anthers about 1 mm long, included. Ovary sessile, ovoid, minute, pubescent. Fruit unknown. Luzon, Province of Cagayan, San Vicente, in dense forests at an altitude of about 15 m, For. Bur. 15U97 Bernardo, April 15, 1910. A species with somewhat the general appearance of Aglaia hexandra Turcz., but with fewer leaflets, much shorter panicles, and quite different flowers. According to DeCandolle’s scheme of classification it is much more closely allied to Aglaia turczaninowii A. DC., but is entirely different from that species. AGLAIA CAGAYAN ENSIS sp. nov. § Euaglaia. Arbor circiter 7 m alta, partibus junioribus inflorescentiisque lepidotis, vetustioribus glabris; foliis usque ad 14 cm longis, 3- ad 5-foliolatis, foliolis anguste oblongo-ellipticis vel oblongo- ovatis, caudato-acuminatis ; paniculis axillaribus, quam folia brevioribus, paucifloris; floribus racemose dispositis, longe pedi- cellatis, tubo stamineo liber. A tree about 7 m high, the branchlets and inflorescence rather densely lepidote, the older parts entirely glabrous. Branches slender, terete, dark-colored when dry, the branchlets pale. Leaves 14 cm long or less, alternate, the rachis and petiole 3 to 5 cm long; leaflets subcoriaceous, brittle when dry, 3 to 5, 113111 2 276 MERRILL. narrowly oblong-elliptic to narrowly oblong-ovate, 6 to 10 cm long, 1.5 to 3 cm wide, narrowed at both ends, the apex somewhat caudate-acuminate, base acute, both surfaces shining, the lower a little paler than the upper; nerves 6 to 8 on each side of the midrib, not prominent ; petiolules 2 to 3 mm long. Panicles in the upper axils, 10 cm long or less, peduncled, with few branches, each branch with from 1 to 3 racemosely disposed flowers, the pedicels 3 to 10 mm long. Calyx cup-shaped, about 2 mm in diameter, shallowly 5-toothed. Petals glabrous, orbi- cular, concave, 2.5 to 3 mm in diameter. Staminal-tube globose, free from the petals, about 2 mm long. Anthers 5, 1.5 mm long, included. Luzon, Province of Cagayan, Abulug River, Bur. Sci. 13801 Ramos, January, 1912. A species well characterized by its lax, very few-flowered panicles, the flowers long-pedicelled. AGLAIA CURRANII sp. nov. § Euaglaia. Arbor circiter 8 m alta, ramulis petiolis inflorescentiisque plus minusve ciliato-lepidotis ; foliis alternis, circiter 30 cm longis, imparipinnatis, foliolis utrinque 5, lanceolatis, subcoriaceis, us- que ad 11 cm longis, acuminatis, basi valde inaequilateralibus, subtus ad costa plus minusve ciliato-lepidotis; paniculis axilla- ribus, circiter 12 cm longis; floribus 5-meris, sessilibus vel sub- sessilibus, spicatim dispositis; petalis liberis. A tree about 8 m high. Branches terete, glabrous, lenticellate, brownish, the younger parts more or less covered with minute, ciliate, lepidote scales, pale-brownish in color. Leaves alternate, about 30 cm long, the petioles, rachis, and petiolules densely covered with small, brownish, lepidote-ciliate scales. Leaflets usually 11, lanceolate, subcoriaceous, rather pale and of about the same color on both surfaces when dry, slightly shining above, 9 to 11 cm long, 2 to 3 cm wide, the apex prominently acuminate, the acumen blunt, the base strongly inequilateral, acuminate; nerves about 12 on each side of the midrib, slender, not prom- inent; petiolules about 1 cm long, that of the terminal leaflet about 1.5 cm long. Panicles axillary, ciliate-lepidote, about 12 cm long, branched from just above the base, the primary branches up to 7 cm in length. Flowers spicately disposed on the ultimate branchlets, sessile or subsessile, not at all crowded. Calyx densely ciliate-lepidote, the teeth 5, ovate, obtuse, ciliate, less than 0.5 mm long. Petals 5, glabrous, free from the staminal tube, oblong, obtuse, 1.5 to 1.8 mm long (just before anthesis). Staminal tube very short, glabrous, 1 mm long or less, truncate. PHILIPPINE PLANTS, IX. 277 Stamens 5, inserted on the base of the tube, included, the anthers 0.5 mm long. Fruit unknown. Luzon, Province of Bataan, Limay Peak, Mariveles Mountain, For. Bur. 17580 Curran, November, 1909, on dry forested ridges at an altitude of about 700 m. A species well characterized by its lanceolate leaflets, its small ciliate- lepidote scales on the inflorescence, petioles, etc., its spicately arranged, sessile or subsessile flowers, and very short staminal-tube. According to De- Candolle’s scheme of classiflcation it seems to be allied to Aglaia llanosiana C. DC., but in all respects is quite different from that species. AGLAIA DIFFUSA sp. nov. § Hearnia. Arbor circiter 25 m alta, inflorescentiis exceptis glabra; foliis alternis, imparipinnatis, 3-jugatis, vix 30 cm longis; foliolis lanceolatis vel elliptico-lanceolatis, utrinque acuminatis aequi- lateralibusque, usque ad 14 cm longis, nervis utrinque circiter 10; paniculis axillaribus, folia aequantibus, diffusis, multifloris, ramis primariis patulis, usque ad 15 cm longis, plus minusve lepidotis vel puberulis; floribus 5-meris, pedicellatis, racemose dispositis. A tree about 25 m high, glabrous except the somewhat lepidote and puberulent inflorescence. Branches terete, brownish, gla- brous, the ultimate ones 4 mm in diameter or less. Leaves alternate, less than 30 cm long, the petiole and rachis 10 to 13 cm long. Leaflets 7, the lateral ones opposite, lanceolate or elliptic-lanceolate, 9 to 14 cm long, 2 to 4 cm wide, narrowed and acuminate at both ends, rather pale and somewhat shining when dry, of about the same color on both surfaces, glabrous, or the midrib beneath with few, small scales; nerves about 10 on each side of the midrib, slender, not very prominent, the reticulations lax, subobsolete; petiolules 4 to 6 mm long. Pani- cles in the upper axils, usually but one on a branchlet, diffuse, about 30 cm long, the lower branches spreading, 15 cm long, very many flowered, the rachis somewhat brown-lepidote, the ultimate branchlets somewhat brown-puberulent. Flowers yel- low, racemosely arranged on the ultimate branchlets, scattered, their pedicels 1 to 1.5 mm long. Calyx-teeth 6, broadly ovate, obtuse, 0.3 mm long, ciliate. Petals 6, glabrous, oblong-obovate to obovate, obtuse, 1.6 mm long, free from the staminal-tube. Staminal-tube broadly obovoid, truncate, about 1 mm long and 1.2 mm in diameter. Stamens 5, inserted on the edge of the staminal-tube, slightly indexed, 0.25 mm long. Luzon, Province of Tayabas, Guinayangan, For. Bur. I8684. Darling, November 21, 1909. In forests at an altitude of about 60 m, locally known as malasaguin. A species of the section Hearnia, well characterized by being nearly 278 MERRILL. glabrous throughout, and by its very diffuse, ample, many-flowered panicles which are more or less lepidote and puberulent. It does not seem to me to be closely allied to any of the previously described species of the genus, although in many respects it resembles Aglaia harmsiana Perk. DYSOXYLUM Blume. DYSOXYLUM LAXUM sp. nov. § Eudysoxylum. Arbor circiter 10 m alta, partibus junioribus puberulis ex- ceptis glabra; foliis alternis, abrupte pinnatis, circiter 30 cm longis, 4- vel 5-jugis, foliolis oblongo-ovatis ad late oblongo- lanceolatis, valde acuminatis, in siccitate pallidis; paniculis axil- laribus, folia aequantibus, parce ramosis, paucifloribus ; floribus 4-meris, circiter 1 cm longis, petalis supra puberulis. A tree about 10 m high, glabrous except the somewhat pu- berulent younger parts. Branches terete, brownish, lenticellate, the growing parts grayish and somewhat puberulent. Leaves alternate, evenly pinnate, about 30 cm long, the rachis and petiole slender, more or less puberulent, about 15 cm long ; leaflets 4 or 5 pairs, the uppermost ones opposite, the others subopposite or alternate, thinly chartaceous or membranaceous, pale when dry, concolorous, slightly shining, oblong-ovate to broadly oblong- lanceolate or somewhat elliptic-oblanceolate, the lower ones in- equilateral at the base, 4 to 5 cm long, the upper ones equilateral, gradually larger and up to 15 cm long and 4.5 cm wide, base acute, apex rather slenderly acuminate; lateral nerves about 10 pairs, slender; petiolules 2 to 5 mm long. Panicles axillary, slender, as long as the leaves, lax, each with very few spreading branches, the lower branches 6 cm long or less and each bearing 2 or 3 flowers, the slender pedicels up to 1.5 cm long. Flowers pinkish, 4-merous. Calyx narrowly funnel-shaped, narrowed below, about 5 mm long, puberulent, obscurely toothed. Petals narrowly oblong, obtuse, about 8 mm long, glabrous on the back except at the apex where they are somewhat puberulent. Sta- minal tube cylindric, free or nearly so, 6 mm long, somewhat toothed at the apex. Stamens 8, inserted at the top of the tube, included. Disk cylindric, 4-toothed, about 2.5 mm long, glabrous. Ovary hirsute; style 5 mm long, slender, glabrous. Young fruits obovoid or somewhat ellipsoid, 1.5 cm long, gla- brous, apiculate at the apex. Camiguin de Mindanao, Bur. Sci. H69i Ramos, March 28, 1912, in forests at Panaytayum. A very distinct and characteristic species recognizable by its lax, very few-flowered panicles, long-pedicelled flowers, etc. It is probably most closely allied to Dysoxylum pauciflorum Merr., but is entirely distinct from that species. PHILIPPINE PLANTS, IX, 279 TOONA Roemer. TOON A PAUCIJUGA sp. nov. Species ut videtur T. calantas Merr. & Rolfe valde affinis, diifert foliolis paucioribus, 7 vel 9, glabris, subtus in axillis venarum prominente glandulosis. A tree reaching a height of about 10 m, glabrous throughout (inflorescence unknown). Branches brownish, glabrous, some- what striate when dry. Leaves alternate, odd-pinnate, 30 to 35 cm long. Leaflets 7 or 9, oblong-ovate to elliptic-ovate, firmly chartaceous, brownish and of nearly the same color on both surfaces when dry, shining, the median ones 10 to 13 cm long, 5 to 6 cm wide, the lower ones somewhat smaller, opposite or subopposite, entire, the base inequilateral, rounded, or on one side of the lamina sometimes acute or acuminate, the apex dis- tinctly acuminate, the acumen usually somewhat falcate, blunt; lateral nerves 10 to 12 on each side of the midrib, distinct, their axils on the lower surface with distinct glands; petiolules 1 cm long or less. Flowers unknown. Panicles terminal, in fruit about 12 cm long. Mature fruits quite similar to those of Toona calantas, about 3,5 cm long, Leyte, between Dolores and Ormoc, in forests at an altitude of about 100 m. For. Bur. 12618 Rosenbluth, February, 109, locally known as calantas. A species well characterized by its few leaflets which are prominently glandular in the axils of the veins on the lower surface. DICHAPETALACEAE, DICHAPETALUM Thou. DICHAPETALUM CILIATUM sp. nov. Frutex scandens, ramulis petiolis foliisque utrinque ad costa nervisque insigniter ciliatis ; foliis oblongis vel oblongo-ellipticis, usque ad 14 cm longis, chartaceis, apice acuminatis, basi acutis vel subrotundatis, nervis utrinque circiter 8, prominentibus ; cymis axillaribus, brevibus, petiolo subaequantibus ; fructibus dense griseo-villosis, 14 mm longis, 2-locellatis, A scandent shrub, prominently ciliate with long, spreading hairs. Branches slender, terete, reddish-brown when dry, the older ones nearly glabrous, the younger ones prominently ciliate with rather pale hairs. Leaves oblong to elliptic-oblong, 8 to 14 cm long, 3 to 6 cm wide, chartaceous, somewhat narrowed to the acute or subrounded base, the apex rather slenderly and sharply acuminate, the acumen 0.5 to 1.5 cm long, the margins and the midrib and nerves on both surfaces prominently ciliate with spreading, pale or somewhat brownish hairs, otherwise 280 MERRILL. glabrous, both surfaces shining, the lower of about the same color as, or a little paler than, the upper; lateral nerves about 8, very prominent on the lower surface, curved-ascending, anas- tomosing, the primary reticulations rather lax, prominent; pe- tioles 5 to 10 mm long, prominently ciliate. Cymes axillary, small, sessile or subsessile, dense, not longer than the petioles, ciliate, many-flowered, the bracteoles lanceolate, 2 mm long or less. Sepals oblong, obtuse, 2 mm long, outside densely pubes- cent. Petals narrowly oblong-spatulate, the apex broad, retuse or slightly cleft, the cleft less than 0.5 mm deep, the base nar- rowed. Filaments about 2 mm long ; anthers 1 mm long. Ovary pubescent. Fruit broadly ellipsoid, more or less compressed, densely gray-pubescent, about 1.5 cm long, divided longitudinally by deep lateral grooves and slightly retuse at the apex, 2-celled. Mindanao, District of Zamboanga, Port Banga, For. Bur. 9293 (type), 929^., 923U Whitford & Hutchinson, December, 1907 and January, 1908, Bur. Sci. 11819 Robinson, July, 1910, in forests at low altitudes. A species well caracterized by its prominently ciliate branchlets, in- florescence, and leaves, the latter glabrous except for the ciliate midrib and lateral nerves on both surfaces and usually also the margins. DICHAPETALUM ROBINSONII sp. nov. Arbor parva, circiter 8 m alta, subtus foliis ramulis inflores- centiisque dense sericeo-villosis ; foliis subcoriaceis, late ellipticis vel ovato-ellipticis, usque ad 15 cm longis, supra nitidis, parce pubescens, basi rotundatis, apice acutis, nervis utrinque circiter 9, prominentibus ; petiolo crasso, circiter 4 mm longo; inflores- centiis in axillis superioribus vel subterminalibus, breviter pe- dunculatis, floribus congestis ; fructibus obovoideis, dense breviter fulvo-villosis. A small tree, reaching a height of 8 m, the branches terete, reddish-brown, somewhat pubescent, the younger branchlets densely covered with short, villous, fulvous hairs, as are the petioles, inflorescence, and lower surfaces of the leaves. Leaves subcoriaceous, broadly elliptic to ovate-elliptic, 10 to 14 cm long, 6 to 8 cm wide, the upper surface brownish-olivaceous when dry, shining, sparingly pubescent, the hairs more numerous on the midrib and nerves, the lower surface densely fulvous-pubes- cent with short hairs; petioles about 4 mm long; lateral nerves about 9, prominent, the primary reticulations rather lax, distinct. Inflorescence in the upper axils and subterminal, shortly pe- duncled, up to 4 cm long, all parts fulvous-villous, the bracteoles linear, 1 to 1.5 mm long. Sepals ovate-lanceolate, acute or somewhat acuminate, about 2 mm long, externally densely pubes- PHILIPPINE PLANTS, IX. 281 cent. Petals glabrous, elliptic to obovate, cleft to the middle. Filaments as long as the sepals; anthers about 0.5 mm long. Fruit apparently obovoid (immature) , 1 to 1.5 cm long, externally densely fulvous-pubescent, apparently 1-seeded. Mindanao, District of Zamboanga, Port Banga, Bur. Sci. 11771 Robinson, July 5, 1910, growing at the edge of the forest. A species very similar in appearance Dichapetalum luzoniense Merr. & Rolfe, and distinguishable only by some minor characters. The leaves are quite different in shape, not narrowed to the base, acute, not at all acuminate, with a few more nerves and more densely pubescent on the lower surface than in the Luzon species. D. luzoniense Merr. & Rolfe is manifestly a woody vine, while D. rohinsonii is described as a tree, reclining, and reaching a height of 8 m. The inflorescence is also subterminal, not strictly axillary, and not long-peduncled. ANACARDIACEAE. PAR I SHI A Hook. f. PARISHIA MALABOG sp. nov. Arbor alta, glabra, vel partibus junioribus inflorescentiisque minute puberulis ; foliis 30 ad 40 cm longis, imparipinnatis ; foliolis 11 ad 15, nitidis, crasse coriaceis, oblongis vel oblongo- ovatis, acuminatis, basi obliquis, valde inaequilateralibus, sub- acutis vel rotundatis, usque ad 14 cm longis; paniculis folia aequantibus, glabris vel obscure puberulis ; floribus breviter pedicellatis, petalis quam sepalis vix longioribus, segmentis calycinis auctis, anguste oblongis, obtusis, usque ad 10 cm longis. A tall tree, reaching a height of 25 m, glabrous or nearly so. Ultimate branches thickened, 1 to 1.5 cm thick, grayish or brownish, marked with large petiolar scars, glabrous, or the growing tip somewhat puberulent. Leaves somewhat crowded at the ends of the branchlets, 30 to 40 cm long, their petioles stout. Leaflets 11 to 15, oblong to oblong-ovate, thickly coria- ceous, 8 to 15 cm long, 4 to 6 cm wide, shining, inequilateral, entire, the apex acuminate, the base oblique, one side of the lamina much broader than the other, usually broadly rounded, the narrower side usually acute; nerves about 14 on each side of the midrib, prominent, anastomosing, the reticulations distinct ; petiolules about 8 mm long. Panicles in the upper axils,, as long as the leaves, glabrous or very slightly puberulent, their rachises stout, straight, the branches rather few, scattered, the lower ones 10 cm long or less, the upper ones shorter. Flowers 4-merous, short-pedicelled. Calyx at anthesis slightly puberu- lent, 4-lobed, the lobes about 6 mm long, obtuse, somewhat spreading. Petals 4, triangular-ovate, obtuse, glabrous, reti- 282 MERRILL. culate, imbricate, about 4 mm long. Ovary ovoid, pubescent, 1-celled, with a single ovule pendulous from near the apex, tapering into the short style ; style trifid, sometimes bifid ; stigmas capitate; rudimentary stamens 4, inserted outside of the disk, the filaments 1 mm long, the anthers as long as the filaments. Male fiowers not seen. Fruit densely villous, about 2 cm long, the accrescent calyx-tube about 2.5 cm in diameter, 1.5 cm long, the lobes when young reddish or reddish-brown, when mature brown, narrowly oblong, rounded, up to 10 cm long, 1 to 1.2 cm wide. Cebu, Buacao, Bur. Set. 11109 Ramos, March 19, 1912 (type), with leaves, immature fruits, and female flowers. TiCAO, For. Bur. 12530, 125^0 Rosenbluth, December, 1908, sterile specimens. Masbate, For. Bur. 21030 Darling, May, 1910, sterile, Foxworthy. Negros, For. Bur. 17356 Curran, September, 1909, seedlings and fallen fruits. Tablas, For. Bur. 19526 Jose. Mindoro, For. Bur. 9831 Merritt. This very characteristic species is most closely allied to Parishia insignis Hook. f. of Tenasserim and the Andaman Islands, and Penang, although quite different from that species. It is the first representative of the genus to be found in the Philippines. Rosenbluth notes that it is the most common tree species on the Island of Ticao, especially abundant on dry rocky hills, and that the trunk has a clear length seldom exceeding 15 meters. He notes also that it has abundant white juice in the bark and leaves. In Ticao the lumber is not used by the natives, but in Masbate it is used for making bancas, that is, dugout canoes. Mr. Elmer has noted on two of the sheets of sterile specimens “Anacardiaceae fide Radlkofer, see Elmer 1216U.” The inference is that Elmer 1216i, a number apparently undistributed as yet, represents the same species. Our material shows conclusively that the plant is refer- able to Parishia, and the specific name is taken from its current Visayan name, maldbog. SEMECARPUS L. SEMECARPUS ACU IVI I N ATISSI IVl A sp. nov. Arbor parva (vel frutex?) infiorescentis exceptis glabra vel subglabra; foliis anguste oblongo-ellipticis ad oblongo-oblanceo- latis, chartaceis vel subcoriaceis, usque ad 20 cm longis, basi angustatis, acutis, apice subcaudato-acuminatis, subtus pallidio- ribus baud glaucescentibus, nervis prominentibus, utrinque circiter 16; paniculis terminalibus, brevibus, angustis, quam folia multo brevioribus, parce pubescentibus vel puberulis, fioribus masculinis parvis, 5-meris. A small tree or a shrub, except the inflorescence nearly gla- brous. Branches terete, slender, grayish. Leaves oblong- elliptic to oblong-oblanceolate or narrowly oblong-ovate, 13 to 20 cm long, 4.5 to 6 cm wide, chartaceous or subcoriaceous, narrowed from about the middle or below to the acute base, PHILIPPINE PLANTS, IX. 283 the apex prominently and rather slenderly subcaudate-acumi- nate, the acumen straight or somewhat curved, 1 to 2 cm long, the upper surface glabrous, when dry somewhat olivaceous, shining, the lower surface pale but not glaucous, slightly shin- ing, very minutely puberulent or nearly glabrous ; lateral nerves about 16 on each side of the midrib, prominent, spreading, anas- tomosing, the reticulations distinct, the primary ones slender, leaving the nerves at about right angles and anastomosing with the secondary lateral nerves, the ultimate free ends of the nervules rather distinct; petioles 1 to 2 cm long. Panicles ter- minal, slender, narrowly pyramidal, about 7 cm long, the lower branches about 2 cm in length, puberulent or somewhat pubes- cent with grayish hairs, the branchlets more or less ferruginous. Male flowers somewhat crowded on the ultimate branchlets, their pedicels 1 to 1.5 mm long, the bracteoles lanceolate, acuminate, pubescent, 1 to 1.4 mm long. Calyx pubescent, 2.5 mm in diam- eter, 5-lobed, the lobes broadly ovate, obtuse, puberulent. Pet- als prominently punctate-striate, oblong-ovate, obtuse, 2 mm long. Disk 1 mm in diameter, ferruginous-pubescent. Anthers 1 mm long. Immature fruits with puberulent pedicels 5 to 8 mm long, the hypocarpium puberulent, when dry about 4 mm long and wide, the drupe obliquely ovoid, somewhat compressed, 5 to 6 mm long, very slightly pubescent, apparently becoming quite glabrous. Luzon, Province of Tayabas, Kabibihan, Bur. Sci. 132^1 Ramos, February 27, 1911, in forests along streams, /' PtSHCAIlONS .POA SAIE B7 THE BTHlEAir OB BCIEHCE, ; ' ' MAHIIA, PHUJPPDIE ISIAKBS THte SUBANXrSrs OP SINBANGAN BAY. ' ’ - By EitcBSON B, CHBi;sTtE. V ..'121'Page», 1. map, -29. plates;, . Ord*r Noi 410.- ■ ' ' -Paper, ^.25 United Stales ourrertoy; ppstpald* ■ -Sindangan Bay Is. aliyated ;>on . the no^^^Iern coast of Zatnbo^inBa. Peninsula, The Subanuns of this region were, studied by Mf. Christie' during, tWo-perlods of five and six weeks, respeotivety. - - The: folipwing -is an. abstract 'from the contents ’of Mr; Chrislie’s report on the Subandns; Habitat and- bIstory;,' relations with, the' Moros;. material- culture; houses; industries; trade; agriouiture; family ' life;' social customs; administration of justice; religion; the medicine man; ceremDnles;tale$;^word,Usts;.Rhysica1measureinents.' ' The 29. plates illustrate', the Bubanuns at work and at play; their Industries; houses, altars, and implements; and the people themeelves; ' TdE BlSTOBY OP SBIiTJ. - ' ‘ ^ By NajeeS M. Saueebt. . ' . r 275 pages,: ..4 .maps, '2 diagrams.' ■ r,- Order No. 406. < -■ - Paper, $0;75 United Stales ourreney, postpaid. , -' . in the preparation of his manuscript for The History of Sulu Doctor Saleeby spent much time >ah:d effprt In' gaining' access to: documents in the possession of the Sultan , of Sulu. ;lt Js fortunate - that these' reoords have now been translated and -preserved in permanent form. This book Is a history of the Moros In the Philippin^es from: the earliesl tlmes to the American oooUpation. STUDIES IN Mono HISTORY, DAW, AND RELIGION. ' ' . By Najeeb M. Saleebt. - ^ : , . ' - , : V : ..JL07 pages, 16 plates, 5 diagrams, Order No. 405., . -Paper, $0.i?S; half morocco, $0.75 United States currency, postpaid. TbiS' volume, deals ^wlth the earliest written' recards'..of the Moros in . Mindanao. Doctor , Saleeby was fortunately able tg ’Obtain' exact copies of. carefully preserved early records writteh V in the Maglndanao dialect with Arabic characters;. The,. author presents translatipns pf these as , iwell as 16 halMone illustrations of. certain pagps from the originals-. The names of the rulers of ; Maglndanao, are. recorded in five folding diagrams. ' , " ; . , NEGBITOS op za^ales. ^ ‘ , ' By Wiixiam’ Ax'i^Aisr Keed. . , 83 pages, 62 platOs. , 'Ordar.No. 402. - ; , ' Paper;^0.25; Half morocop, $0.75 United Stales ourrenoyvjpostpaid. V The introductory chapter deals with the general distribution, of Negritos and: with the dlslrlbu-; tion of the Philippine branoh .of the raoe.'r ;The succeeding chapters deal with’ the varlpus indus.- tries, amusemenia; and sooial-retaltons of these little men. ; Plates from photographs, the greater part, of which was taken for this pubHoation, show ornaments, houses, men making fire with bamboo, botys and arrows, ;danoes, and various types of the people themselves. ’ CONTENTS. Pi^ MERRILL, E. D. New or Noteworthy Philippine 259^ REVIEW ;___ 358 ■, ■ ' ’1[| 'S' " '' The ^‘Philippine JournnJ ot Science’* is Issued as follows: currency. ' Section A. Ohemlcal and fSeological Sciences' and the InduSiarles.. $2.0 0 *■ Section B. The Philippine Journal of Tropical Medicine L S.OP Section C, Botany 2.00 Section B. General Biology, JRithndlOgy and Anthropology (Seci ' tion 1) began with Volume V).. . 2.00 Entire Journal, Volume It, III, IV, Or V 5.00 Entire Journal, beginning with Volume ^VI 7.0O Single numbers .'. .50 Volume I, 1906 (not divided into sections) 10.00 Supplement to Volume I or to toy of the agents listed below: AGEIfTS. The Macmillan Companyf 64-66 Fifth Avenue, New Tork. City, tJ; S. ‘A* Wm. Wesley & Son,- 28 Essei Street^ Strand, Iiondon, W. C:y Engla^; Martinus Nijhoff, Nobelstraat 48, The Hague, Holland. , Mayer & Miiller, Prin* Louis PeMlpandsfrasjSe 2,iBerlin, N.W,, Germany* Helfe# & Walsh, Limited, S2 Haffles Flace, Singapore, StraitO Settlements. A, Mi & J. Ferguson, 19 Batllic Street, Colombo, Ceylon. ' v TbaOker, Spink & Co., P. O. Box 54, Calcutta, India. Entered at (he pott-^ffioe at Manila, P. I., :at.>sa^aon(l-olatt miitter.' gWiWiisi Kiiliitf avp^W. QOX, M. A;\ ,Ph.D. lliisi®®®iS i^SEd^jok; Ct, ’Bo^TAfe' *> . D. d: MERRILL, M. S/ ' \ ■' ^ ' TBE COQPE|6Aa'IOjSi; OP '-, •''’ ' BROWN, Ph.O. _N, 'wiiltoRD/REuD. '' wans sas infeiiTOLt- ' .?: JMlANILA PITBIICATIONS rOR SALE BY THE BUREAU OF SCIENCE, MANILA, PHILIPPINE ISLANDS REPORT OP THE IKTERKATIONAE PLAGUE CONFEREJfCE. ' Held at Mukden April, 1911, under tlie auspices of ‘ ’ the Chinese Government. Edited hy Ekicn M4.hti>’i, G. F. Peteie, Aexhur Stanley, anp Hiciiabd . P. Steono. 483 pages, IS plates (2-colored, 4 half-tones, 12 charts and maps); Order No. 416. Cloth, $3.50; paper, $2.50 United States currency, postpaid. The proceedings of this International Conference and Information gained therefrom, together with the results of certain baotenological investigatiems, constitute .the present report, - Nothing hitherto has been published which gives such a complete and comprehensive account : of the entire subject of pneumonic plague. : ■ Delegates from America (United States of), Austria-Hungary, France, Germany, Great Britain, Italy, Japan, Mexico, the Netherlands, Russia, and China attended the Conference. The Bureau of Science of the Government of the Philippine Islands has bean appointed sola agent for the distribution of the printed proceedings of the International Plague Conference. * THE SUGAR INDUSTRY IN THE ISLAND OP NEGROS. By Heebert S. Walkee. » ^ 145 pages, 10 plates, 1 map. Order No. 43,2. .■!,■■ ■■■■■. Paper, $1.25 United States currency, postpaid. Considered from the Viewpoint of practical utility, Mr. Walker’s Sugar Industry in the Island of Negros is one of the most important papers published by the Bureau of Science. This volume is a real contribution to the subject; it is not a mere compilation, for the author was in the field and understands the conditions of which he writes. The following is a brief synopsis of the contents: Yables of soil analyses,, both chemical and ^physical; analyses of the cane, Juice and bagasse; estimates based on actual information as to the costs of production and of cultivation; and esti- mates of the cost and location of possible central factories. The island is considered by sugar- producing, districts; the area of cultivation and the production pec hectare are given, and the possibility for future expansion discussed. The plates illustrate various phases of sugar industry from the oultivation of the field to the transportation of sugar in native sailboats. , ^ A MANUAL OP PHILIPPINE SILK CULTURE. ' - By Charles S. Ba^*ks. 53 pages, 20 plates. Order No. 413. Paper, $0.75 United States currency, postpaid. The silk industry is particularly adapted to be undertaken by persons with small capital, and like the making of hats in the Philippine Islands it should thrive with a little encouragement. .In A Manual of Philiopine Slik Culture we have presented the results of several years’ actual work with slik-producing larvae together with a description of the new Philippine race. Half-tone Plates, illustrate in natural size silkworms in different stages of development, pupae, adult moths, samples of cloth made from erl silk, hand reel, and silk house. Other plates illustrate the various appliances used in raising silkworms and in spinning silk; hand and power reels are Illustrated; working drawings are given for a silk house and for a hand reel. THE PHILIPPINE Journal of Science C. Botany VOL. VII DECEMBER, 1912 No. 6 THE PINEDA MONUMENT AND THE PROBABLE SITE OF THE FIRST BOTANIC GARDEN IN THE PHILIPPINES. By E. D. Merrill.* {From the Botanical Section of the Biological Laboratory, Bureau of Science, Manila, P. /. The Pineda monument probably localized the first botanic garden to be established in the Philippines, and one of the first in the Orient, although an institution long since forgotten. The monument was erected in 1792 to the memory of Antonio Pi- neda who, up to his death, was in charge of the natural history investigations of the Malaspina Expedition. As the monument has now been destroyed it has been considered advisable to reproduce what is probably the only picture of it extant, to com- pile what is definitely known regarding Pineda, and at the same time to correct some misconceptions regarding the nature of his work. The Malaspina Expedition, so called from its commander. Captain Alejandro Malaspina, was dispatched by the Spanish Government, and was one of the first great scientific and explor- ing expeditions sent out by the various European nations. Montero ^ states that the frigates Descuhierta and Atrevida, which made up the squadron, were thoroughly equipped with instruments and with scientific literature, for astronomical, mathematical, and physical research, as well as for investiga- tions in natural history. The ships left Cadiz, Spain, July 30, * Associate Professor of Botany, University of the Philippines. ‘Hist. Filip. 2 (1894) 321. 114346 363 364 MERRILL. 1789, and the expedition was undoubtedly the most thoroughly equipped one of its kind up to that date. In fact it is doubt- ful if any similar expedition had previously been dispatched by any nation, with the same objects in view. Very many voyages had previously been made for geographic and com- mercial purposes by the navigators of Spain, Portugal, England, Holland, and France, but the more purely scientific aspects of exploration had been ignored or largely subordinated to more practical matters. An account of the expedition has been given by Novo,^ whose work I have not seen, but it is said by Safford ® to be greatly condensed and is hence of comparatively little value. In spite of the thorough equipment of the expedition, the time occupied, nearly five years, and the known industry and effec- tiveness of some of its staff, notably the botanists Thaddeus Haenke and Luis Nee, the expedition proved to be peculiarly barren of published results. Enormous collections of botanic material were made by Haenke and Nee, much of which is still extant. Haenke’s collection was sent to Prague, Haenke himself having died in South America in 1817. In 1825 Presl undertook the publication of his pretentious work entitled “Reliquiae Haenkeanae,” which extended over the years 1825 to 1836, but which was never finished, only one volume and a portion of the second being printed. The most complete set of Haenke’s botanic material is still preserved at Prague, in part in the “Museum Kralovstvi Ceskeho,” and in part in the University Herbarium in that city, while more or less duplicate material has been dis- tributed to other institutions, the K. K. Naturhistorisches Hof- museum, Vienna, the Kgl. Botanisches Museum, Munich, the Kgl. Botanisches Museum, Berlin, and the Bernhardi Her- barium, now the property of the Missouri Botanical Garden, St. Louis, Missouri. Scattered specimens are to be found in the DeCandolle Herbarium, and the Delessert Herbarium, both in Geneva, the Rijks Herbarium, Leiden, and doubtless in other European botanic institutions. Haenke’s botanic material, to a greater or less degree, has been accessible to various European specialists, and his specimens are constantly cited in monographs. Nee, however, was less fortunate in the disposition of his material, which was deposited in the herbarium of the Jardin * La vuelta al mundo por las corbetas Descubierta y Atrevida al man do del Capitan de Navio Don Alejandro Malaspina, desde 1789 a 1794. Publi- cado con una introduccion en 1885 por el Teniente de Navio Don Pedro de Novo y Colson. •Contr. U. S. Nat. Herb. 9 (1905) 28. THE PINEDA MONUMENT. 365 Botanico at Madrid, where, presumably, it is still preserved, for Vidal * who examined certain types in 1886, states that the material was then in an excellent state of preservation. Both Colmeiro “ and Jordana ® state that Nee’s collections comprised 10,000 species, of which J^,000 were new. Only a few hundred, however, were ever described, a very few by Nee himself, the others by Cavanilles. No duplicates seem to have been distrib- uted to other institutions and the material has not been avail- able to other botanists. Nee is almost never quoted as a collector, and DeCandolle,^ in his extensive enumeration of botanic authors and collectors, does not even mention Nee in spite of the latter’s enormous collections. If Colmeiro is correct in his estimate of the number of species collected by Nee, the record has perhaps been equalled or exceeded by but few botanists or collectors. Antonio Pineda was the naturalist in charge of the natural history investigations of the Malaspina Expedition.® Most Spanish authors consider him as a botanist, but I can find no evidence to this effect. Sternberg ® definitely states that he was a zoologist, which statement is verified by the inscription on the Pineda monument, quoted below, which contains a direct reference to the fauna, but no mention of the flora. Pineda was born in Guatemala in 1759, and was an official in the Spanish military service. Zaragoza states that he was a colonel, but this needs verification. He died at Badoc, Prov- ince of Ilocos Norte, Luzon, in June, 1792. His monument, now entirely destroyed, was erected in Manila by his friends and colleagues. The monument is figured in the plate accom- panying this article. Very little is definitely known regarding Pineda and his life, and I have been able to find but short, meager, and at least in some respects, mostly inaccurate accounts of his work. Stern- berg’s few notes comprise the earliest published record of ‘Rev. PI. Vase. Filip. (1886) 13. 'La botanica y los botanicos de la peninsula Hispano-Lusitana (1858) 182. * Revista de Filipinas 2 (1877) 174; Bosquejo geografico e historico natural del Archipielago Filipino (1883) 356. 'La Phytographie (1880) 391-462. •Jordana 1. cc. (1885) 173, 356. •Presl Rel. Haenk. 1 (1825) XII. ” Colmeiro 1. c. "La Ilustracion Filipina 2 (1892) 271. "Presl Rel. Haenk. 1 (1825) XII. 366 MERRILL. Pineda which I have seen. Colmeiro^® gives a short and ap- parently mostly accurate account from which later authors have largely drawn. Jordana, in an article entitled “Don Antonio Pineda y la flora de Pilipinas” gives the longest account known to the writer. In this he amplifies Colmeiro’s notes, and gives some additional data. He quotes the inscription on the Pineda monument, which is, however, printed wrongly, in 18 lines, instead of in 12, as it appears on the monument. He later gives an abbreviated account of Pineda. Zaragoza’s statements are manifestly very incorrect. Montero draws his data largely from Jordana’s earlier account and quotes the inscription from him. Blanco speaks of him as a botanist,^® but is in error as to the date of his death. Both Colmeiro and Zaragoza consider him as a botanist, while Montero is authority for the statement that he was commissioned by the Spanish Government to study the flora of the Philippines and the state of its agriculture. Whether Pineda was a botanist or a zoologist is perhaps of little importance. Aside from the references given above no allusions to him have been found in botanic literature, although the genus Pineda Ruiz and Pavon {^Banara Aubl.) seems to have been dedicated to him. Pineda apparently published noth- ing, either in botany or in zoology, and there is no record that he collected any material of either a botanic or zoologic nature; at least no such material is preserved. Zaragoza states that Pineda’s chief work was the establish- ment of a mulberry plantation at a place since called Pineda, situated between Pasay and Malate, but this is certainly in- correct. Pineda arrived in Manila on March 27, 1792, and died about three months after his arrival. Sternberg states that he selected the region about Bulacan for his investigations, so that it is probable that he left Manila shortly after the arrival of the expedition. In the three months intervening between his arrival and death, especially considering the slow methods of transportation, and the fact that he died at Badoc, more than 200 miles from Manila, he could have had little time for the “Revista de Pilipinas 2 (1877) 173, 174. " Bosquejo geografico e historico-natural del Archipielago Filipino (1885) 356. “La Ilustracion Filipina 2 (1892) 271. “ Hist. Filip. 2 (1894) 321. “Flora de Pilipinas (1837) IV. “Igualmente y a principios de este siglo, habia fallecido en estas Islas el sabio botanico D. Antonio Pineda.” THE PINEDA MONUMENT. 367 work indicated by Zaragoza, nor could he have visited many islands of the Archipelago as stated by him. As to the place called Pineda, this was until recently the official name of the present town of Pasay, immediately south of Manila. In fact on one Spanish map, drawn as late as 1896, Pineda is given as the only name of the town now known as Pasay. There is evidence that a botanic garden was in existence in the city of Manila at the time of the arrival of the Malaspina Expedition, but the only definite reference to it that I have been able to find is the following quotation from an article written by James Britten,^® the data for which were taken from Des- champs’ manuscripts in the library of the British Museum, and from other sources. “Norona died at the Isle de France of an obstruction of the liver, contracted in Madagascar * * *. The Spaniards have erected a monument to the memory of their indefatigable countryman in the island of Luzon, near Manilla, on ground belonging to the royal botanic garden, which, during his residence there. Dr. Norona had done everything in his power to bring into order, and to stock with many valuable plants.” Noronha died in 1787, according to Colmeiro,^® and according to corroborative evidence given by Britten, who quotes from Dupetit-Thouars “Genera Nova Madagascariensia,” which is direct evidence that a botanic garden was in existence near Manila before the arrival of the Malaspina Expedition. It has been impossible, however, to find any traces of the Noronha monument, nor have any other published references to it been noted. Mr. A. E. W. Salt, of the University of the Philippines, who has made an exhaustive examination of the archives of the city of Manila, has found no references to a botanic garden earlier than the one established in the zone of fortification of the Walled City in 1858. This, however, applies only to such an institution as might have been established or supported by the city of Manila. No references, naturally, would be found in the city archives to an institution belonging to the central government. If the Pineda monument localized this long forgotten botanic garden, as it probably did, the institution was certainly located outside of the limits of the city of Manila, as the city was constituted from 1780 to 1800. “ Bibliographical Notes XXX. — L. A. Deschamps and F. Noronha. Joum. Bot. 41 (1903) 282-285. ”L. c. 214. 368 MERRILL. There is little doubt but that the Pineda monument was erected in what was at that time, 1792, the Botanic Garden, and the Noronha monument, mentioned above, of which no traces have been found, was probably located on the same tract of ground. There could be no logical reason, otherwise, for erecting the monument to Pineda in such an isolated place, distant from the city, and near no street. Zaragoza’s picture, reproduced below, gives an excellent idea of the isolation of the monument, and up to the beginning of the present century, when its destruction was completed, it had quite the appearance of being located in the midst of abandoned fields. The abandon- ment of the Botanic Garden, when and why not evident, was the probable reason that the Pineda monument was allowed by the authorities to fall into a ruinous condition. We have Are- nas’ evidence, cited below, that the monument had long been uncared for, as early as 1850. Whether or not the Pineda monument localizes this long forgotten Botanic Garden, the institution was the first of its kind to be established in the Phil- ippines, if not in the entire Orient. The tract of land on which the Pineda monument was erected is at the present time occupied by the Philippine Bureau of Agriculture as an experimental station. The ruins of the monument, consisting of the foundation only, remained in situ until the early part of the year 1904, when they too were removed in the construction of what is now known as Wright Street. The location of this monument was approximately in the middle of what is now Wright Street, immediately north of the new Malate school building. Arenas,^^ writing in 1850, considered it strange that the Pineda monument still stood at that date after the many years during which no one had cared for it : “.Extrano es tambien que se conserve en pie despues de tantos afios no habiendo ninguno que lo cuide.” Its abandonment, then, long antedates 1850, and nothing seems to have been done after that date to keep it in repair, for Zaragoza, writing in 1892, speaks of the abandoned monument in the midst of a sementera (small plantation) hav- ing excited his curiosity as a youth, and closes his article as follows: “De sentir es que nuestro ilustre Ayuntamiento tenga en complete abandono aquel monumento, eregido por nuestros antepasados a la memoria de un sabio, que sacrifice su vida en aras de la ciencia y por el bien de estas Islas.” Montero, writing Memorias historicas y estadisticas de Filipinas (1850) no. 10. THE PINEDA MONUMENT. 369 in the same year, states that the monument was then almost completely destroyed. According to all verbal accounts that I have been able to secure, the monument still stood at the time of the American occupation in 1898, but was then in a very ruinous condition. On my first visit to it in the early part of the year 1904, nothing remained but the square foundation, all parts above the surface of the ground having been removed. The granite slab bearing the inscription was left near the foundation, but was later removed by the authorities of the City of Manila and has been placed in the center of a small triangular grass plot at the intersection of three streets in the parked portion of the Cementerio del Norte. The stone of which the monument was constructed was soft volcanic tuff, locally known as Guadalupe stone, and is said to have been removed by the person or persons who rented the land between the years 1898 and 1902. The tract of land, on which the monument was erected, was purchased by the Phil- ippine Government in the year 1903, for the site of an agricul- tural experiment station, for which, in part, it is still used. The inscription on the granite slab, said by Colmeiro, Jordana, Montero, and others, to have been written by Haenke, is as follows : ANTONIO PINEDA. TRIBUNO. MILITUM. VIRTUTE. IN. PATRIAM. BELLO. ARMISQUE. INSIGNI. NATURAE. DEMUM. INDEFESSO. SCRUTATORI. TRIENNII. ARDUO. ITINERE. ORBIS. EXTREMA. ADIIT. TELLURIS. VISCERA. PELAGI. ABYSSOS. ANDIUMQUE. CACUMINA. LUSTRANS. VITAE. SIMUL. ET LABORUM. GRAVIUM. DIEM. SUPREMUM. ORIIT.“ IN. LUCONIA. PHILIPPINARUM. . VI. CALENDAS. lULII. MDCCXCII. PRAEMATURAM. OPTIMI. MORTEM. LUGET. PATRIA. LUGET. FAUNA. LUGENT. AMICI. QUI. HOCCE. POSUERE. MONUMENTUM. The accompanying plate, representing the Pineda monument, is reproduced from the figure given by Zaragoza in a periodical formerly published in Manila, entitled “La Ilustracion Fili- pina.” It is, in all probability, the only picture of the monu- ment extant. According to Montero,®^ who quotes Jordana, the total height of the monument was three and one-half meters. “ Apparently a sculptor’s error for OBIIT. ^'2 (1892) 271. ^"Hist. Filip. 2 (1894) 322. ILLUSTRATION. Plate XXII. The Pineda monument reproduced from a figure given by Zaragoza in “La Ilustracion Filipina” 2 (1892) 271. 371 Merrill; The Pineda Monument.] [Phil. Journ. Set., VII. C, PLATE I. THE PINE MONUMEN' The Philippine Journal op Science, C. Botany. Vol. VII, No. 6, December, 1912. ON THE IDENTITY OF EVODIA TRIPHYLLA. By E. D. Merrill.* {From the Botanical Section of the Biological Laboratory, Bureau of Science, Manila, P. I.) It not infrequently happens in taxonomy that, due to im- perfect original descriptions, nonaccessibility of type or typical material to later botanists, or other reasons, the current con- ception of a particular species is quite different from that origin- ally intended by its author. An excellent illustration of this is to be found in the case of Evoclia triphylla DC., which, on examination of the type, proves not to be an Evodia at all, but a species of the allied genus Melicope. As the species has passed in botanic literature for about ninety years as a true Evodia, and has been credited with the extended range of Tenasserim and Burma to China, Japan, the Philippines, and Malaya, due to misinterpretation of Fagara triphylla Lam., it is only reason- able to suppose that the error will persist in some quarters for some time in the future. In the year 1788, Lamarck ^ described as a new species Fagara triphylla, which DeCandolle later transferred to the genus Evodia, in which it has been retained by all botanists up to the present time, and to it many species described by other authors have been reduced as synonyms, mostly erro- neously. Evodia triphylla DC. has been confused with E. rox- burghiana Benth., by many authors, a species which it certainly very closely resembles in gross characters, but which is really generically distinct. Hooker f.^ retains Evodia roxhurghiana Benth. and E. triphylla DC. as distinct species, but notes that there is some doubt as to which name the former should bear because of the obscurity of Evodia triphylla DC. Guillaumin ® * Associate Professor of Botany, University of the Philippines. 'Encycl. 2 (1788) 447. » FI. Brit. Ind. 1 (1875) 488. ® Lecomte FI. Gen. Indo-chine 1 (1911) 632. 373 374 MERRILL. reduces E. roxburghiana Benth. to E. triphylla DC., including most of the synonyms of the two as given by Hooker f. Lamarck’s description is based on a fruiting specimen col- lected in the Philippines by Sonnerat, which is preserved in the Herbarium of the Paris Museum of Natural History. Nat- urally his original description is incomplete, as the flowers were lacking. To the species he also referred Ampacus angiistifolius Rumph. Herb. Amboin. 2: 188, t. 62, and Rumphius’ plate has undoubtedly been the basis of interpretation of Lamarck’s species for most if not all succeeding botanists. The figure very closely resembles our Philippine “Evodia triphylla,” and may represent the same species, but this can be definitely determined only by an exhaustive botanical exploration of Amboina, and a critical study of the material secured. Sonnerat’s specimen must be the type of the species, because it is the plant described. The reference to Rumphius, which may or may not represent the same form as the Philippine plant, cannot logically be inter- preted as the type. In 1824 DeCandolle ^ transferred Lamarck’s species to Evodia as Evodia triphylla DC., giving only a very short diagnosis. This name has been retained until the present time, although there is nothing in the original description definitely to indicate that the species really belongs in Evodia. Later authors referred to it Philippine material, Cuming 1819, which was accepted by all botanists considering the species, until very recently K. Schumann,® examining the specimen in the Berlin herbarium, noted that it had 8 stamens, and was therefore a Melicope and not an Evodia. The Cuming plant, with other Philippine mate- rial was designated by Engler as a new species, Melicope luzon- ensis Engl., and published under this name by Doctor Perkins 1. c. Shortly after the publication of Melicope luzonensis Engl., I examined our Philippine material, which had been referred to Evodia triphylla Lam., and found that, so far as flowering specimens were concerned, all of the collections had 8 stamens, and that all the material was hence Melicope and not Evodia.^ Since that time I have examined many Philippine specimens in the hope of finding one having flowers with 4 stamens that would agree with Lamarck’s description of Fagara triphylla, but without success. In November, 1911, Doctor C. B. Robin- ‘ Prodr. 1 : 724. • Perk. Frag. FI. Philip. (1905) 162. ‘Bur. Govt. Lab. Publ. (Philip.) 35 (1905) 24. ON THE IDENTITY OF EVODIA TRIPHYLLA, 375 son had an opportunity of examining Lamarck’s type, the Phil- ippine plant collected by Sonnerat, in the Herbarium of the Paris Museum of Natural History. He has supplied me with a carbon rubbing showing form and venation of characteristic leaves of the type, and with the following note “All Philippine numbers here are determined as Evodia triphylla Lam. : there is no doubt about the correctness of the determinations, Gaudichaud 255, Barthe s. n., Cuming 1819, Lohei' 199, and the type.” The leaf- form and venation of the type, as shown by the carbon rubbing made by Doctor Robinson, is quite the same as in our abundant material of Melicope luzonensis Engl. There is, hence, no doubt but that Fagara triphylla Lam.=Evodia triphylla DC., is really a species of Melicope, that, so far as is known, is confined to the Philippine Islands. The synonymy of the Philippine plant, as well as of the Asiatic forms referred to DeCandolle’s species by various authors, through error, needs adjustment. Chamisso ^ considered the species under Zanthoxylum refer- ring to it a Philippine plant collected by himself, but arbitrarily changed the name to Zanthoxylum lamarkianum. Blanco ® a little later described the same form as Fagara octandra (non Linn.). Champion ® referred Cuming 1819 to his Zanthoxylum pteleaefolium, that species being based in part on Cuming’s specimen. Bentham later reduced this, together with Fagara triphylla Lam., to his Evodia lamarckiana, the latter name being a transfer of Chamisso’s Zanthoxylum lamarkianum. Vidal and Fernandez-Villar referred the Philippine material to the correct genus, but to the wrong species, Melicope ternata Forst. Recently Engler has redescribed the species as Meli- cope luzonensis, while the most recent description is that given by Mr. ElmeF^ as Melicope odorata. I can see no valid reason for distinguishing the latter species. The full synonymy of the Philippine form is given below: MELICOPE TRIPHYLLA (Lam.) comb. nov. Fagara triphylla Lam. Encycl. 2 (1788) 447. Evodia triphylla DC. Prodr. 1 (1824) 724; Hook. f. FI. Brit. Ind. 1 (1875) 448 p.p., quoad Philip.; F.-Vill. Novis. App. (1880) 34; Vidal Phan. Cuming. Philip. (1885) 100, Rev. PI. Vase. Filip. (1886) 74. ' Linnaea 5 (1830) 58. “FI. Filip. (1837) 67, ed. 2 (1845) 48. “Bentham in Hook. Kew Journ. Bot. 3 (1851) 330. “FI. Hongk. (1861) 59. “Perk. Frag. FI. Philip. (1905) 161. “Lead. Philip. Bot. 2 (1908) 476. 376 MERRILL. Zanthoxylum lamarkianum Cham, in Linnaea 5 (1830) 58. Fagara octandra Blanco FI. Filip. ((1837) 67, ed. 2 (1845) 48, ed. 3, 1 : 90. Zanthoxylum pteleaefoUum Champ, ex Benth. in Hook. Kew Joum. Bot. 3 (1851) 330, p.p., quoad Cuming 1819. Evodia lamarckiana Benth. FI. Hongk. (1861) 59, p.p. Melicope ternata Vid. Sinopsis Atlas (1883) t. 2i, f. A; F.-Vill. Novis. App. (1880) 34, non Forst. Melicope luzonensis Engl, in Engl. & Prantl Nat. Pflanzenfam. 3* (1897) 122, nomen; Perk. Frag. FI. Philip. (1905) 161; Merr. in Govt. Lab. Publ. (Philip.) 35 (1906) 24, Philip. Journ. Sci. 3 (1908) Bot. 411, 5 (1910) Bot. 355. Melicope odorata Elm. Leaf!. Philip. Bot. 2 (1908) 476. The genesis of this rather formidable list of synonyms is given above. In some cases they are due to arbitrary changing of specific names, in others manifestly to careless work, and in all cases more or less to the imperfect original description. The species is common and widely distributed in the northern and central parts of the Philippines, being uniform in essential characters, but like most other species showing some variation in vegetative characters and in the denseness or laxness of its inflorescence. It is dioecious, but the male flowers constantly have 8 stamens and the female ones 8 staminodes, so that with flowering material the generic determination is evident if the flowers are examined. In gross characters, habit, leaves, leaf- lets, inflorescence, and fruits it strongly resembles Asiatic speci- mens that have variously been referred to Evodia triphylla DC., E. roxhurghiana Benth., etc., and which are true Evodias, having four stamens. The resemblances are so great that it is not surprising that representatives of two distinct genera have been referred by several prominent botanists to a single species. So far as can be determined at present Melicope triphylla is confined to the Philippines, for all the Asiatic material that I have had an opportunity of examining, determined as Evodia triphylla, has four stamens, or four staminodes. I have exam- ined the following Philippine material, preserved in the Her- barium of the Bureau of Science: Luzon, without definite locality, Jagor 732: Cagayan Province, Bolster 118: Benguet Subprovince, For. Bur. 15867 Bacani, Williams 1070, Elmer 6282, For. Bur. 18180, 18211 Curran, Merritt, & Zschokke: Province of Nueva Vizcaya, For. Bur. H858 Darling: Province of Bataan, For. Bur. 3056 Borden: Province of Cavite, For. Bur. 7691 Curran: Province of Rizal, Bur. Sci. i671, 13562 Ramos, For. Bur. U39 Ahern’s collector, Merrill 2787, Phil. PI. 67 Ramos, For. Bur. 10021 Curran: Province of Laguna, Bur. Sci. 12333 McGregor, Bur. Sci. 12025 Ramos, Holman 73: Province of Sorsogon, For. Bur. 10520 Curran. Mindoro, For. Bur. 67^9, 9999 Merritt, Merrill ON THE IDENTITY OF EVODIA TRIPHYLLA. 377 2397. Palawan, Merrill 736, Bur. Sci. 860 Foxworthy. Romblon, Hallier. Ticao, For. Bur. 1069 Clark. Masbate, Merrill 2289. Negros, Elmer 9529, 9628, 10432. CEBU, For. Bur. 6422 Espinosa. Bohol, Cuming 1819. We have seen above that Fagara triphylla ham.= Evodia triphylla DC., is really not an Evodia at all, but must be clas- sified under the genus Melicope as M. triphylla (Lam.) Merr. The numerous Philippine synonyms have been disposed of, but there remain to be considered the still more numerous names, for the most part based on material from the Asiatic continent, that have erroneously been referred to Evodia triphylla DC. From an examination of the specimens available here I am personally of the opinion that two distinct species are repre- sented in the Asiatic material, or at least among the synonyms, that have been referred to Evodia triphylla DC. by Guillaumin,^® in this being in agreement with the expressed opinion of Ben- tham and Hooker f.^® Much of the material from China and apparently from Indo- China in superficial characters very strongly resembles the Phil- ippine Melicope triphylla; in fact the resemblances are so great that it is difficult if not impossible to separate them unless the specimens are in anthesis. The earliest valid name for this form appears to me to be that supplied by Xanthoxylum, pteleae- folium Champ., and the following new combination is accordingly made: EVODIA PTELEAEFOLIA (Champ.) comb. nov. Xanthoxylum pteleaefolium Champ, ex Benth. in Hook. Kew Journ. Bot. 3 (1851) 330, p.p., excl. Cuming 1819. Evodia lamarckiana Benth. FI. Hongk. (1861) 59, p.p., excl. syn. Lamarck and Chamisso. Evodia gracilis Kurz in Journ. As. Soc. Beng. 2 (1871) 48. I refer here the following specimens in the Herbarium of the Bureau of Science: Yunnan, Henry 10351A, 11658A; Tonkin, Bon 6187. Two specimens from Formosa, Nakahara 294, with fruit, and a sterile specimen. No. 1268, may be referable here, or they may be the same as the Philippine Melicope triphylla Merr. The Indian and Malayan material, for the most part deter- mined as Evodia roxburghiana Benth., appears to me to rep- resent a species distinct from Evodia pteleaefolia. Its leaves are apparently always pale when dry, are quite different in shape and venation, while the infforescence is decidedly different. The Philippine form described by me as Evodia semecarpifolia “ Lecomte FI. Gen. Indo-chine 1 (1911) 632. ”F1. Hongk. (1861) 59. ” FI. Brit. Ind. 1 (1875) 488. ”Govt. Lab. Publ. (Philip.) 35 (1906) 23. 378 MERRILL. may ultimately prove to be the same species. The following is an attempt to straighten out the rather complicated synonymy. EVODIA LUNUR-ANKENDA (Gaertn.) comb. nov. Fagara lunur-ankenda Gaertn. Fruct. 1 (1788) 334, t. 68, f. 9 Lepta triphylla Lour.” FI. Cochinch. (1790) 82. Zanthoxylum zeylanicum DC. Prodr. 1 (1824) 728. Fagara triphylla Roxb. FI. Ind. 1 (1832) 416; Wight Ic. t. 20U, non Lam. Zanthoxylum roxburghianum Chamisso in Linnaea 5 (1830) 58. Evodia roxburghiana Benth. FI. Hongk. (1861) 59, in obs.; Hook. f. FI. Brit. Ind. 1 (1875) 487; King in Journ. As. Soc. Beng. 62’ (1893) 208. Zanthoxylum marambong Miq. FI. Ind. Bat. Suppl. (1861) 533. Evodia marambong Miq. Ann. Mus. Bot. Lugd, Bat. 3 (1867) 244. Evodia triphylla Guillaumin in Lecomte FI. Gen. Indo-chine 1 (1911) 632 p.p. This species appears to be widely distributed, and is repre- sented in the Herbarium of the Bureau of Science by the follow- ing material: Southern India, Fischer 2751. Singapore, Rid- ley, four specimens. Borneo, Sarawak, Hewitt. Its reported range is from the Khasia Mountains south to Ceylon, Tenas- serim, the Malay Peninsula, Sumatra, and Java. As noted above, the Philippine Evodia semecarpifolia Merr., may even- tually have to be reduced to this species; it is, at least, very closely allied. ” From the very short description this may be the same as the preceding form, E. pteleaefolia. Loureiro’s specific name is, however, invalid in Evodia. The Philippine Journal of Science, C. Botany. Vol. VII, No. 6, December, 1912. NOTES ON PHILIPPINE EUPHORBIACEAE. By E. D. Merrill.* {From the Botanical Section of the Biological Laboratory, Bureau of Science, Manila, P. I.) This family is one of the largest in the Philippines, both in number of genera and in species; in the Spermatophyta it is probably exceeded in the Archipelago only by the Rubiaceae and the Orchidaceae. While much work has been done on the group in the past few years, a great deal still remains to be accomp- lished, and in order thoroughly to understand the relationships of the various forms, a great deal of field work is necessary. In the majority of species the plants are dioecious, and hence usually at least three collections are necessary before adequate material is secured for study, for it is always desirable to have not only male and female flowers, but also mature fruits. In the case of our Philippine forms, and this is also true of extra-Philippine ones as well, a high percentage of the species are inadequately known for the reason that material in all stages has never been collected. In the present paper about twenty-six new species are pro- posed, in twelve genera. One genus, Alcinaeanthus, of the Eu- phorhiaceae-Gelonieae is proposed as new, manifestly allied to the Malayan genus Cheilosa Blume. Two genera, Cladogynos Zipp., and Blumeodendron Kurz, are recorded from the Arch- ipelago for the first time, the first by the discovery of a rather widely distributed Malayan species in two localities in the Phil- ippines, and the second by transfer of an endemic species from Sapium where it was originally described. A few new combina- tions are made, and some nomenclatural notes are included. ALCINAEANTHUS gen. nov. {Euphorbiaceae-Gelonieae.) Flores dioici, apetali. Floris $ sepala 4, aequalia, valde im- bricata. Disci glandulae 6, minutissimae. Stamina 6; filamen- ta libera; antherae late ovoidae, breves. Ovarii rudimentum * Associate Professor of Botany, University of the Philippines. 114346—2 379 380 MERRILL. villosum, oblongo-ovoideum, subtruncatum, interdum 2-fidum. Floris 2 sepala 4, leviter inaequalia. Ovarium 2-loculare; styli 2, breves, basi breviter connati, apice 2-fidi, lobis brevissimis, crassiusculis ; ovulis in loculis solitariis. Capsula ellipsoidea, vix sulcata, 1-sperma, in valvas 4 dehiscens vix in coccos 2- valves dissiliens. Semen ecarunculatum, magnum. Arbor parva, fere glabra. Folia alterna, petiolata, penninervia, plus minusve glanduloso-denticulata, basi supra 2-glandulosa. Pani- culae $ axillares, laxae, multiflorae ; inflorescentiae 2 ut videtur racemosae. ALCINAEANTHUS PHILIPPINENSIS sp. nov. Arbor parva, 4 ad 10 m alta, inflorescentiis puberulis exceptis glabra; foliis chartaceis vel subcoriaceis, obovatis vel elliptico- obovatis, breviter acuminatis, 8 ad 20 cm longis, basi acutis, supra 2-glandulosis, margine distanter glanduloso-denticulatis, nervis utrinque 6 vel 7 ; paniculis s diffusis, usque ad 10 cm longis; floribus minutis, 4-meris; staminibus 6; fructibus ellip- soideis, circiter 2.5 cm longis, extus dense pallide puberulis. A small tree 4 to 10 m high, glabrous except the puberulent inflorescence, the branches slender, terete, grayish. Leaves chartaceous to subcoriaceous, obovate to elliptic-obovate, usually brownish and somewhat shining when dry, of the same color on both surfaces or somewhat olivaceous above and brownish beneath, 8 to 20 cm long, 3.5 to 12 cm wide, the base acute and with two small but distinct glands on the upper surface at the junction with the petiole, the apex shortly acuminate, the mar- gins distantly glandular-denticulate, or sometimes subentire, sometimes the teeth rather prominent in the upper part of the leaf ; lateral nerves 6 or 7 on each side of the midrib, prominent, anastomosing, the reticulations distinct; petioles 1.5 cm long. Male inflorescence paniculate, the panicles puberulent, axillary, usually solitary, diffusely branched, up to 10 cm long, many- flowered. Flowers small, 4-merous, the bracts ovate, acute or acuminate, about 1.3 mm long, the bracteoles similar but smaller, each subtending a single flower; pedicels 1 to 1.5 mm long, the buds globose, glabrous. Sepals 4, strongly imbricate, orbi- cular, 1.5 to 2 mm in diameter, subequal. Stamens 6; filaments about 1 mm long; anthers broadly ovoid, 0.5 mm long. Rudi- mentary ovary villous, about 1 mm long, oblong-ovoid, obtuse or truncate, sometimes 2-fld; disk-glands 6, alternating with the filaments, very minute. Female inflorescence apparently race- mose, the flowers not seen, but the persistent sepals in very young fruits four, unequal, ovate, acute, up to 2 mm long; styles NOTES ON PHILIPPINE EUPHORBIACEAE. 381 2, shortly united below, 2-fid, the four resulting lobes thick, 1 mm long or less; ovary 2-celled, cells 1-ovuled, puberulent. Fruits when nearly mature ellipsoid, 1-celled, 1-seeded, 2.5 cm long, not at all sulcate and not formed of cocci, but of four valves which are subequal, oblong-elliptic, acute at both ends, crustaceous, very densely and softly puberulent externally with a pale indumentum. Seed (immature) 2 cm long. Mindoro, in hill forests south of Lake Naujan, For. Bur. 6851 Mer- ritt, April 6, 1907, with very young fruits, altitude 80 meters. Leyte, near Ormoc, For. Bur. 12759 Rosenbluth, March, 1909, with male flowers, in forests, altitude about 20 meters. Mindanao, District of Davao, Williams 288i, June 6, 1905, with nearly mature fruits: District of Cotabato, Lebak, For. Bur. 11815 (type), 11767 Whitford, March, 1912, both with male flowers, in river bottoms at low altitudes. This proposed new genus is manifestly closely allied to the monotypic Javan Cheilosa Blume, and belongs in the Euphorbiaceae-Gelonieae. It is, in fact, so similar to Blume’s genus that I at first considered it to be refer- able to Cheilosa. Cheilosa javanica Blume is represented in the Herbarium of the Bureau of Science by two specimens collected in Java by Blume himself, and presumably cotypes; both of these have female flowers. Our Philippine material resembles these specimens in vegetative characters, except that Blume’s species lacks the basal leaf-glands so characteristic of Alcinaeanthus. In the arrangement of both the male and female flowers it is also similar to Cheilosa, except that the male flowers are solitary in each bracteole, while the entire or 2-fid rudimentary ovary is similar in both genera. The differences, however, are found in the leaves, in the male and female flowers, and in the fruits, and are, it is considered, of sufficient importance to warrant the separation of the Philippine form as a distinct genus. As noted above, in Alcinaeanthus the leaves are 2-glandular at the base, although in other characters very similar to those of Cheilosa. The flowers of both sexes are 4-merous, not 5-merous, while the male flowers are solitary in the axils of the bracteoles, and the sepals are very strongly imbricate. In the female flowers the ovary is 2-celled, not 3-celled as in Cheilosa, and the styles are two, very short, each with 2, equal, short, thick lobes. The fruits are not at all sulcate, and are not composed of cocci, but split into four equal valves; they are also 1-celled and 1-seeded, not composed of three 1-seeded cocci as in Cheilosa. The genus is dedicated to Father Francisco Ignacio Alcina of the Jesuit Order, in whose honor Cavanilles has already proposed the genus Alcina' (=iAlcinia HBK.), which is a synonym of the earlier Melampodium of Linnaeus. Father Alcina was born in Gandia, Spain, in the year 1610, entered the Jesuit Order in 1624, and came to the Philippines in 1632. Here he was rector of various colleges, but his chief labors were of an evangelical nature in the Visayan Islands. He died July 30, 1674. He is the author of an interesting and still unpublished work on the natural history of the Philippines, a copy of the first part of which is preserved in the Library of the Ateneo de Manila, and which I have examined. The full title of the ’leones 1 (1791) 10, t. 15. 382 MERRILL. work is “Historia natural del sitio, fertilidad y calidad de las islas e indios de Visayas, compuesto por el Padre Francisco Ignacio Alzina, de la com- pania de Jesiis, despues de mas de treinta y tres anos en ellas, y entre ellos de Ministerio. Ano de 1668.” Added to the copy in the Ateneo de Manila, is the following note “1802 — A uso de Fr. Agn. Ob. el. de N. Sega.,” which may be interpreted: “For the use of Fr. Augustin, Bishop-elect of New Segovia”, indicating that the copy was made, presumably from the original, in the year 1802. Facing the title-page is a note pasted in which gives the following information regarding the origin of the copy: “Esta 1.® parte de la obra del P. Alcina (la 2.“ no se encuentra) pertenecio al Prelado D. Fr. Augustin Pedro Blaguier, que tomo posesion del Obispado de Nueva Segovia en 1779, y ha sido comprada en una almoneda en Paris este ano de 1875 por 52 francos.” The above data supplement the very meager account of Alcina given by Colmeiro," while Pritzel ’ simply mentions Alcina’s name, giving a ref- erence to Colmeiro. F.-Villar ^ quotes the title of Alcina’s work correctly. I am indebted to Father Sola, S. J., of the Ateneo de Manila, for assistance In securing data regarding Father Alcina and his work. The species described by Mr. Elmer as Elateriospermum paucinervium,^ of which Pax repeats the description ® expressing some doubts as to whether or not it is referred to the proper genus, cannot possibly be an Elateriospermum. It is, I think, possibly referable to Alcinaeanthus or Cheilosa; at any rate it is undoubtedly referable to the Euphorbiaceae- Gelonieae, and apparently represents a generic type near those above- mentioned if not one of them. As fruits alone are known, it will be im- possible properly to refer the species to a genus until either the female or the male flowers, or both, are known. It is represented by the follow- ing specimens, all with fruits: Luzon, Province of Bataan, Lamao River, For. Bur. 758 Borden, May, 1904: Province of Tayabas, Lucban, Elmer 7^16 (type number). May, 1906. SiBUYAN, Elmer 12367, April, 1910. ALCHORNEA Swartz. ALCHORNEA PARVI FLORA (Benth.) Muell.-Arg. in Linnaea 34 (1865) 168, DC. Prodr. 15^ (1866) 902. Stipellaria parviflora Benth. in Hook. Journ. Bot. & Kew Miscel. 6 (1854) 4. In a previous paper I reduced this species to Alchornea sicca (Blanco) Merr.’, but I am now of the opinion that two distinct species are rep- resented in the material then referred by me to Blanco’s species. Alchornea parviflora Muell.-Arg. seems to be represented by Bur. Sci. 5220 Ramos, Rizal Province, Luzon, and by Loher 681.3, from Union Province, Luzon, in addition to Cuming 1800, the type number. It is distinguished from A. sicca (Blanco) Merr. by its somewhat larger, distinctly pubescent leaves, ’ La Botanica y los Botanicos de la Peninsula Hispano-Lusitiana (1858) 158. ’Thesaurus Lit. Bot. (1872) 4. ‘Blanco FI. Filip, ed. 3, 4’ (1880) IV. ’ Leafl. Philip. Bot. 2 (1908) 484. * Pflanzenreich 44 (1910) 102. 'This Journal 5 (1910) Bot. 192. NOTES ON PHILIPPINE EUPHORBIACEAE. 383 its elongate, slender, staminate inflorescence, which is 10 to 15 cm in length, and its somewhat rugose capsules. Alchomea sicca (Blanco) Merr. is glabrous or nearly so, and its dense male spikes are usually less than 2 cm long, rarely 4 to 5 cm in length. In addition to the material previously referred here, it is represented by Merrill ^100, Phil. PL 988 Merrill, and Didrichsen 3146, 3170, all from Manila. ALCHORNEA RUGOSA (Lour.) Muell.-Arg. in Linnaea 34 (1865) 170, DC. Prodr. 15^ (1866) 905; J. J. Sm. in Koord. & Val. Bijdr. Boomsoort. Java 12 (1910) 466. Cladodes rugosa Lour. FI. Cochinch. (1790) 704. Conceveibum javanense Blume Bijdr. (1825) 614. Adelia glandulosa Blanco FI. Filip. (1837) 814. Alchomea javensis Muell.-Arg. in DC. Prodr. 15 ‘ (1866) 905. This widely distributed species is represented in our herbarium by about 30 specimens from the Philippines, for the most part determined and dis- tributed as Alchomea javensis Muell.-Arg. It extends from northern Luzon to southern Mindanao. It is represented also by Ridley 107, from Christmas Island (south of Java), and by specimens from Java collected by Boerlage. India, through Malaya to the Admiralty Islands. ANTIDESMA Burmann. ANTIDESMA CRASSIFOLIUM (Elm.) comb. nov. Sapium crassifolium Elm. Leafl. Philip. Bot. 2 (1908) 485. Negros, Cuernos Mountains, Elmer 10312, June, 1908, type number. This species is an Antidesma in all respects, and is apparently very closely allied to Antidesma bunius Spreng. It differs from that species in its more pubescent spikes, and thicker, fewer-nerved leaves. Mature leaves are very thickly coriaceous, but immature ones, distributed under the type number, are membranaceous or chartaceous. Pax and K. Hoff- mann * state that the species does not belong in the Euphorbiaceae, as the specimen examined by them appeared to have perfect flowers. All the flowers on my specimens are staminate, and are the characteristic ones of Antidesma; in poor material the rather prominent pistillode might have been mistaken for a fertile ovary. ANTIDESMA MINDANAENSE sp. nov. Frutex 3 ad 4 m altus ramulis inflorescentiisque exceptis glaber; foliis oblongis vel anguste oblongo-ovatis, acuminatis, usque ad 20 cm longis, breviter petiolatis, nervis utrinque circiter 7, prominentibus, anastomosantibus ; stipulis oblongo- lanceolatis, acuminatis, 8 ad 10 mm longis; racemis termina- libus, pubescentibus, 5 ad 15 cm longis, solitariis; floribus 5 4-meris, sepalis ovatis, acutis, circiter 4 mm longis; fructibus ovoideis, purpureis, in siccitate rugosis, 4 ad 5 mm longis. A shrub or small tree 3 to 4 m in height. Branches slender, terete, grayish, glabrous, the younger ones somewhat pubescent. * Pflanzenreich 52 (1912) 257. 384 MERRILL. Leaves alternate, oblong to narrowly oblong-ovate, chartaceous, 10 to 20 cm long, 4 to 9 tm wide, when dry shining and of about the same color on both surfaces, glabrous, the apex promi- nently and sharply acuminate, the base rather broadly rounded or subacute; nerves about 7 on each side of the midrib, some- what ascending, prominent, anastomosing near the margin, the reticulations lax, distinct; petioles 3 to 5 mm long, slightly pubescent; stipules oblong-lanceolate, acuminate, 8 to 10 mm long, glabrous or nearly so, often slightly falcate. Male flowers unknown. Female racemes terminal, solitary, 5 to 15 cm long, pubescent. Flowers 4-merous. Fruit ovoid, slightly com- pressed, rugose when dry, purple, 4 to 5 mm long, glabrous, tipped by the remains of the style; pedicels pubescent, about 2 mm long. Persistent calyx pubescent, the lobes 4, ovate, acute, about 0.5 mm long. Mindanao, District of Zamboanga, Sax River, San Ramon, Williams 2117 (type), February 5, 1905, Hallier s. n., February, 1904; near Zam- boanga, Merrill 827U, December 6, 1911. Apparently distinct from all other Philippine forms, and possibly most closely allied to Antidesma cumingianum Muell.-Arg., but very different from that species. It appears to be well characterized by its shortly petioled leaves, its pubescent simple racemes, 4-merous flowers, and oblong- lanceolate stipules. BLUMEODENDRON Kurz. BLUMEODENDRON SU BROTUN Dl FOLI U M (Elm.) comb. nov. Sapimn subrotundifolium Elm. Lead. Philip. Bot. 3 (1910) 930. SiBUYAN, Elmer 12349, April, 1910, type number. Although the flowers of this species are unknown, and the fruits in the type collection very immature, I have no hesitation in transferring the species from Sapium, to which genus it cannot possibly be referred, to Blumeodendron. Kurz’s genus has been sunk in Mallotus as a subgenus, but is so very different from the other plants usually grouped under Mal- lotus that I consider J. J. Smith ® entirely justified, in his recent treat- ment of Blumeodendron, in again restoring it to generic rank. We have in the Philippines a second species of the genus, as yet undescribed, repre- sented by abundant material, with vegetative characters very similar indeed to Blumeodendron subrotundifolium, but with entirely different fruits. BR I DELI A Willdenow. BR I DELIA PLATYPHYLLA sp. nov. § Monospermae. Species B. moonii Thw., ut videtur similis et affinis, differt foliis distincte acuminatis, utrinque glabris. A tree 8 to 20 m high, quite glabrous except for the slightly pubescent axillary fascicles. Branches terete, slender, lenti- " Meded. Dept. Landbouw 10 (1910) 458 (Reorders & Valeton Bijdr. Boomsoort. Java 12). NOTES ON PHILIPPINE EUPHORBIACEAE. 385 cellate, usually light-brown or somewhat reddish, rarely dark- colored. Leaves firmly membranaceous to chartaceous, elliptic to oblong-elliptic, up to 20 cm in length and 10 cm wide, smaller leaves half as large, shining when dry, of about the same color on both surfaces, the base usually broadly rounded, apex prom- inently acuminate; nerves 13 to 16 on each side of the mid- rib, prominent, somewhat ascending, curved near the margins, scarcely or very slightly anastomosing, the reticulations sub- parallel, slender; petioles 5 to 10 mm long. Flowers 5-merous, small, in dense, axillary, many-flowered fascicles, male and female flowers in the same fascicle, sessile. Male buds globose, the anthers broadly elliptic, about 0.5 mm long. Petals about 1 mm long. Female flowers 2 mm long, the calyx-lobes oblong- ovate, acuminate, 1.5 mm long. Petals narrowly oblong, thin, 0.5 mm long. Disk slightly hairy, surrounding the ovary, the narrowly ovoid ovary and the style together 2 mm long. Bracts thin, more or less hirsute, about 2 mm long. Fruit reddish-purple when mature, ellipsoid, 8 to 10 mm long, shortly stipitate, apex rounded and minutely apiculate, indehiscent, 1-celled, 1-seeded, the seed hard, very deeply sulcate longitudinally. Luzon, Province of Cagayan, Lalloc, For. Bur. 18504- Alvarez, in fruit, March, 1909; Abulug River, Bur. Sci. 14557 Ramos, in fruit, January, 1912: Province of Laguna, Siniloan, For. Bur. 17634 Curran, February, 1910, in fruit; Santa Maria, For. Bur. 22229 Mariano, sterile; Mount Maquiling, For. Bur. 11725 Whitford, sterile: Province of Bulacan, Angat, For. Bur. 11167 Aguilar (type), in flower, April, 1908: Province of Ba- tangas, Santo Tomas, For. Bur. 21528 Tamesis, May, 1910, in flower: Province of Tayabas, Guinayangan, For. Bur. 18676 Darling, sterile. Ta- BLAS, For. Bur. 19530 Jose, sterile. Mindanao, District of Cotabato, For. Bur. 15421 Pray, March, 1911, in flower: Province of Misamis, Dapitan, Piper 87, April, 1911, in flower. All of the above specimens have some or all of the leaves large with from 12 to 16 pairs of lateral nerves; some of them in addition have smaller leaves, about one-half the normal size, with from 8 to 10 pairs of nerves. The following series of specimens have only the smaller, few-nerved leaves, but in all essential characters appear to be the same as the species described above: Luzon, Province of Camarines, For. Bur. 10431 Curran, May, 1908, in fruit: Province of Tayabas, Guinayangan, Merrill 2025, in flower. Min- danao, District of Zamboanga, Ahern 372, 641, in flower. Basilan, Hallier s. n., in flower. Most of the material cited above has been identifled as Bridelia ovata Decne., a species belonging in the section Eubridelia, which does not appear to occur in the Philippines. At least one specimen of the small- leaved form, Merrill 2025, has been named by Gehrmann in the Herbarium of the New York Botanical Garden as Bridelia subnuda K. Schum. & Lauterb., a species described from New Guinea material, which is also placed by Gehrmann in the section Eubridelia.^" Fruiting specimens of 386 MERRILL. what I take to be exactly the same form as Merrill 2025 have 1-celled, 1-seeded fruits, with deeply longitudinally sulcate seeds. Native names: Tamarina, Maragata (Negrito, Cagayan) ; Subiang (Tag., Laguna) ; Pahobo (Tag., Bulacan) ; Cubi tahiran (Tag., Tayabas) ; Bago- tadhau (Vis., Tablas) ; Magantumit (Cotabato) ; Taran (Zamboanga). CLADOGYNOS Zippel. CLADOGYNOS ORIENTALIS Zipp. ex Span, in Linnaea 15 (1841) 349; Muell.-Arg. in DC. Prodr. 15“ (1866) 384; J. J. Sm. in Koord. & Val. Bijdr. Boomsoort. Java 12 (1910) 384. Conceveiba tomentosa Span. 1. c. Rottlera albicans Hassk. Hort. Bogor. (1858) 238. Adenogynum discolor Reichb. f. & Zoll. in Verb. Nat. Ver. Ned. Ind. 1 (1856) 23, Linnaea 28 (1856) 325. Chloradenia discolor Baill. Etud. Euph. (1858) 472, t. 19, f. 2i, 25. Cephalocroton albicans var. genuinus Muell.-Arg. in Linnaea 34 (1865-66) 155. Cephalocroton discolor Muell.-Arg. in DC. Prodr. 15“ (1866) 761. Cephalocroton orientalis Schelf. in Ann. Mus. Bot. Lugd. Bat. 4 (1868) 120. SiQUiJOR, Merrill 7208, September, 1910, in thickets along roads near the seashore. Cebu, Bur. Sci. 11055 Ramos, April, 1912. I have followed J. J. Smith in the synonymy as given above, and in the reduction of the genus Chloradenia to Cladogynos. The genus has not previously been reported from the Philippines, and I am now able to extend its range still farther, as C. orientalis is undoubtedly represented by Robinson 1321 from Nha-trang, Annam, Indo-China, March 16, 1911. Previously known from Java, Timor, Celebes, and Kisser. CLEISTANTHUS Hooker f. CLEISTANTHUS ANGUSTIFOLIUS sp. nov. Frutex 2 ad 3 m altus, glaber; foliis lanceolatis vel anguste lanceolatis, subcoriaceis, sensim tenuiter acuminatis, usque ad 12 cm longis, basi obtusis vel rotundatis, brevissime petiolatis, nervis utrinque circiter 10, obscuris; floribus ut videtur fasci- culatis, 4-meris, sessilibus, calycis lobis lanceolatis, 2.5 mm longis, villosis; fructibus ovoideis vel subglobosis, leviter 3-sul- catis, crustaceis, circiter 1 cm longis, extus parce pilosis. A shrub 2 to 3 m high, glabrous, monoecious. Branches terete, slender, reddish or somewhat grayish. Leaves lanceo- late to narrowly lanceolate, 3 to 12 cm long, 8 to 10 mm wide, subcoriaceous, shining, rather pale and of about the same color on both surfaces when dry, the base abruptly rounded or obtuse, gradually narrowed from the lower one-fourth to the long and slenderly acuminate apex, often slightly falcate; nerves about 10 on each side of the midrib, very obscure, anastomosing, the reticulations close but not elevated; petioles 2 mm long or less. “Engl. Bot. Jahrb. 41 (1908) Beibl. 95:30. NOTES ON PHILIPPINE EUPHORBIACEAE. 387 Flowers apparently fascicled, possibly solitary, axillary, sessile or subsessile, 4-merous. Female flowers: Calyx about 3 mm long, the lobes somewhat appressed-hirsute, lanceolate or oblong- lanceolate, 2 to 2.5 mm long. Petals 4, narrowly oblong-obo- vate, truncate, about one-half as long as the calyx lobes. Ovary ovoid, not sulcate, densely hirsute; styles 3.1 mm long, cleft. Capsules ovoid or subglobose, crustaceous, slightly longitudinally 3-sulcate, sparingly pilose, 8 to 10 mm long. Luzon, Province of Cagayan, Abulug River, Weber 1562 (type). For. Bur. 17796 Curran, Bur. Sci. 13927 Ramos, January, 1912, in forests, alti- tude 80 to 180 m, said by Mr. Curran to be one of the commonest shrubs in the ridge forests. A very characteristic species, distinguished by its narrow, lanceolate, short-petioled, long-acuminate, obscurely nerved leaves. In leaf-form it is quite different from all the other known Philippine species. CYCLOSTEMON Blume. CYCLOSTEMON MEGACARPUS sp. nov. Species ut videtur C. bordenii Merr. afflnis, differ! fructibus multo majoribus, usque ad 3 cm diametro. A tree, nearly glabrous (flowers unknown). Branches terete, the older ones light-gray, shining, the growing tips more or less olivaceous. Leaves oblong, coriaceous, of the same color and strongly shining on both surfaces, smooth, glabrous, 11 to 15 cm long, 3 to 7.5 cm wide, entire, the apex acute or ob- scurely acuminate, the base inequilateral, acute on one side of the midrib, subacute or somewhat rounded on the other; lateral nerves about 7 on each side of the midrib, rather prominent, curved-ascending, anastomosing near the margins; petioles 8 to 10 mm long. Flowers unknown. Fruits globose, 3 cm in di- ameter, somewhat gray-pubescent, the indumentum deciduous, the pedicels 1.5 cm long, solitary or in fascicles apparently from the larger branches or from the trunk, the persistent calyx about 8 mm in diameter. Seeds 2 in each fruit. Luzon, Province of Laguna, San Antonio, Bur. Sci. H953 Ramos (type), June, 1912, Bur. Sci. 13521 Ramos, August, 1910, in forests. In vegetative characters the species rather closely resembles Cyclos- temon bordenii Merr., but its fruits are very much larger, and are appar- ently cauline or borne on the larger branches. CYCLOSTEMON RAMIFLORUS sp. nov. A praecedente differ! fructibus paullo majoribus, foliis basi acutis, aequilateralibus. A species manifestly closely allied to Cyclostemon mega- carpus, differing in the above characters. Glabrous through- out. Branches stout, light-gray, terete. Leaves oblong, coria- 388 MERRILL. ceous, shining, 15 to 20 cm long, 5 to 7 cm wide, entire, apex apparently obtuse, base acute, equilateral or subequilateral, never rounded on one side; lateral nerves 6 or 7, slender, as- cending; petioles about 1 cm long. Flowers unknown. Fruit from the old branches below the leaves, solitary or fascicled, from tubercles, globose, fleshy, glabrous, apparently about 4 cm in diameter, red, with an agreeable flavor {teste Foxworthy), each with two seeds. Negros, Mount Marapara, For. Bur. 136^0 Curran & Foxworthy, Sep- tember, 1909, on forested ridges, altitude about 700 meters. Very similar to Cyclostemon megacarpus Merr., and perhaps not specif- ically distinct. The material, however, shows no intergradations in the differential character presented by the bases of the leaves. CYCLOSTEMON SUBCRENATUS sp. nov. Arbor circiter 12 m alta; foliis oblongo-lanceolatis, chartaceis vel subcoriaceis, usque ad 8 cm longis, rectis vel leviter falcatis, plus minusve inaequilateralibus, apice obtuse acuminatis, basi acutis vel leviter acuminatis, margine distincte irregulariter crenatis ; f ructibus axillaribus, oblongo-ellipsoideis, dense griseo- puberulis, circiter 1 cm longis, 6 mm diametro, 1-locularibus, 1-spermis. A tree about 12 m high, glabrous except the obscurely pu- berulent petioles, younger branches and the infrutescence. Branches terete, grayish, lenticellate, rather slender, glabrous, younger ones somewhat olivaceous, more or less angular and slightly puberulent. Leaves alternate, oblong-lanceolate, char- taceous or subcoriaceous, of the same color and shining on both surfaces when dry, 5 to 8 cm long, 1.5 to 2.5 cm wide, straight or somewhat falcate, the base somewhat inequilateral, acute or somewhat acuminate, the apex obscurely and bluntly acuminate, the margins distinctly and irregularly crenate; primary nerves about 10 on each side of the midrib, not prominent, anastomos- ing, scarcely more distinct than are the secondary nerves and the reticulations; petioles 2.5 to 5 mm long. Flowers unknown. Fruits axillary, solitary, oblong-ellipsoid, densely gray-puberu- lent outside, about 1 cm long and 6 mm in diameter, crowned by the styles, 1-celled, 1-seeded, the pedicels very slightly pubescent, about 4 mm long. Luzon, Province of Batangas, near Santo Tomas, For. Bur. 21567 Ta- mesis, May 22, 1910, in forests at an altitude of about 200 m, locally known as balichican. A species distinguished from the other Philippine representatives of the genus by its comparatively narrow leaves which are distinctly crenate. In some respects it resembles C. cumingii Benth., but has much smaller, differently shaped leaves. NOTES ON PHILIPPINE EUPHORBIACEAE. 389 EXCOECARIA Linnaeus. EXCOECARIA PH I LI PPI N ENSIS Merr. var. EUPHLEBIA var. nov. Very similar to the type in all essential characters, differing chiefly in its much more prominent lateral nerves. In Excoecaria philippinensis the primary nerves are scarcely more prominent than are the secondary ones, while in the present form on the lower surface they are raised and very much more prominent than the secondary ones. In the shape of the leaves, their variation in form and size, texture, and in the flowers and fruit so far as our material shows, the variety appears to be quite the same as the species. Luzon, Province of Cagayan, For. Bur. 16591, 16853, 17808, 19629 Curran, Bur. Sci. 13973 (type), 7^78 Ramos. Two specimens from Benguet Prov- ince may be referable here, but they have fewer nerves than the Cagayan material. Bur. Sci. 25 lU Meams and Bur. Sci. 13502 Ramos. Similarly two specimens from Tayabas Province differ, Whitford 773 and Bur. Sci, 133 lf5 Ramos. EXCOECARIA OBTUSA sp. nov. Frutex glaber; foliis oblongo-obovatis, subcoriaceis, usque ad 12 cm longis, apice late rotundatis, basi acutis, nervis utrinque 10 ad 12, patulis, anastomosantibus, distinctis, margine recur- vatis; inflorescentiis $ in axillis superioribus, densis, quam petioli vix longioribus, floribus brevissime pedicellatis, brac- teolis reniformibus, rotundatis. A glabrous shrub. Branches terete, brownish or olivaceous when dry. Leaves subcoriaceous, oblong-obovate, 8 to 12 cm long, 3,5 to 6.5 cm wide, shining, the lower surface considerably paler than the upper, the apex broadly rounded, more or less narrowed below the middle to the acute base, margins recurved ; primary nerves 10 to 12 on each side of the midrib, spreading at nearly right angles, distinct, anastomosing and forming a looped marginal nerve, the secondary nerves about as prominent as the reticulations; petioles 1.5 to 2.5 cm long. Male racemes in the upper axils, solitary, shorter than the petioles, densely many-flowered, the rachis stout, the pedicels 1 to 1.5 mm long, the axillary glands prominent; bracteoles reniform, rounded, about 1.5 mm wide, 1 mm long, obscurely glandular on the margins. Sepals broadly ovate, acute, slightly glandular on the margins, 1 mm long. Filaments 1 to 2 mm long. Female flowers and fruits not seen. Luzon, Province of Cagayan, Abulug River, in forests near Dabba, Bur. Sci. 13917 Ramos, January, 1912, flowers yellow. A species manifestly allied to Excoecaria philippinensis Merr., but dis- tinguished at once by its very differently shaped, broadly rounded, not acute or acuminate leaves. 390 MERRILL. HOMALANTHUS H. Jussieu. HOMALANTHUS POPULNEUS (Geisel.) Pax, var. MINOR (Muell.-Arg.) comb. nov. Carumbimn populneum Muell.-Arg. var. minus Muell.-Arg. in DC. Prodr. 15" (1866) 1145. Homalanthus populneus Pax, var. siccus Pax in Engl. Pflanzenreich 52 (1912) 46 (non Excoecaria sicca Blanco!). Mueller’s name for this variety must be maintained, for Pax and K. Holfmann are in error in the adoption of the varietal name siccus, based on Excoecaria sicca Blanco. Mueller first suggested that Excoecaria sicca Blanco was a synonym of Carumbium populneum^'^ but was not sure, “Excaecaria sicca Blanco etiam hie pertinere videtur.” An exam- ination of Blanco’s description shows that this reduction is impossible, as he described a very different plant, with leaves very similar to those of tuba camaisa {Croton tiglium L.), 3-nerved at the base, said by him to be common along the margins of estuaries and rivers. Excoecaria sicca Blanco is common in some localities in and about Manila, in the habitats ascribed to it by Blanco, habitats in which Homalanthus never grows; moreover no Homalanthus is known from the immediate vicinity of Manila. I am con- fident that F.-Villar was correct in his reduction of Blanco’s Excoecaria sicca to the genus Alchornea; it is not, however, the same as Alchomea mollis Muell.-Arg. I have already “ transferred Blanco’s name to Alchomea, as A. sicca (Blanco) Merr., and reduced to it the more recently described Alchornea parviflora Muell.-Arg. While there is absolutely no doubt as to the correctness of the transfer of Blanco’s species to Alchornea, I am now of the opinion that I was in error in the reduction of Alchornea parvi- flora to A. sicca Merr., and that two distinct species are represented. See page 382. Homalanthus populneus var. minor is common and widely distributed in the Philippines, and is represented by the following material: Luzon, Province of Laguna, Elmer: Province of Zambales, Hallier: Province of Bulacan, Bur. Sci. 1303 Jf Ramos: Province of Bataan, Williams 289, For. Bur. 20008 Topacio, For. Bur. 1209 Borden, Whitford U78, 1281: Province of Rizal, Merrill 1879 (not “Elmer 1879,” as quoted by Pax), Phil. PL 309 Merrill, Loher ^801 : Province of Batangas, For. Bur. 21559 Tamesis, Elmer 9226, Merrill 2893 (not “Elmer 2893,” as quoted by Pax), Ritchie U5: Province of Albay, Bur. Sci. 2889, 2891 Meams, Bur. Sci. 6i09 Robinson: Province of Sorsogon, For. Bur. 10525 Curran. Marinduque, collector unknown. Negros, For. Bur. 11209 Everett, Whitford 1626. Leyte, Elmer 7338. Palawan, Merrill 702, 843 (not “Elmer 702, 843,” as quoted by Pax). Mindanao, District of Zamboanga, Williams 2077, For. Bur. 9245 Whitford & Hutchinson: District of Davao, Elmer 12499. Native names: Balanti, bayanti (Negros, Marinduque) ; bota-bota (Ta- yabas) ; buta (Rizal) ; malabinunga, binuang-susong (Bataan) ; banti (Tavn-tawi). Prodr. 15": 1144. This Journal 5 (1910) Bot. 192. NOTES ON PHILIPPINE EUPHORBIACEAE. 391 HOMONOIA Loureiro. HOMONOIA JAVENSIS (Blume) Muell.-Arg. in Linnaea 34 (1865) 200, DC. Prodr. 15^ (1866) 1022; J. J. Sm. in Koord. & Val. Bijdr. Boomsoort. Java 12 (1910) 544. Spathiostemon javense Blume Bijdr. (1826) 622. Adelia javanica Miq. FI. Ind. Bat. F (1858) 388. Mallotus eglandulosus Elm. Lead. Philip. Bot. 1 (1908) 313. Leyte, Palo, Elmer 7152 (type of Mallotus eglandulosus Elm.), 7318, January, 1906, in fruit; Ormoc, For. Bur. 11631 Whitford, March, 1909, with staminate flowers. Cebu, Danao, For. Bur. 6U62, 6UU, 6k67 Everett, February, 1907, with staminate flowers. Mindanao, District of Davao, Williams 2808, May, 1905, with staminate flowers. Mallotus eglandulosus Elm. was described from fruiting specimens, and was reduced to Homonoia javensis Muell.-Arg. by J. J. Smith 1. c., a species not previously recorded from the Philippines under its proper name. An examination of our material with staminate flowers confirms the reduction, for the stamens are arranged in branched phalanges. Java, Timor, Amboina, Key, and New Guinea. Var. CILIATA var. nov. A typo dilfert foliis subtus ad costa petiolisque ciliato-hirsutis. Negros, Cauayan, For. Bur. 72i6 Everett (type), March, 1907; Tanjay River, For. Bur. 12325 Everett, April, 1908. MACARANGA Thouars. MACARANGA GIGANTIFOLIA sp. nov. Arbor parva, foliis glabris vel subglabris, late ovatis, circiter 60 cm longis, distanter denticulatis vel integerrimis, basi late rotundatis, alte peltatis, apice acuminatis, subtus glandulis minu- tis ceraceis nitidis instructis ; inflorescentiis $ ut videtur axilla- ribus, pubescentibus, circiter 50 cm longis, anguste paniculatis; bracteis orbicularibus, concavis, 3 ad 4 mm diametro, minute denticulatis ; staminibus 2 ad 4. A small tree, the leaves nearly glabrous, the inflorescence rusty- pubescent. Leaves firmly chartaceous or subcoriaceous, broadly ovate, about 60 cm long, 40 cm wide, acuminate, base broadly rounded, deeply peltate, the petiole inserted about 13 cm from the basal margin, the margins entire or with few, scattered, minute teeth, the upper surface quite glabrous, rather pale when dry, shining, the lower one brownish, slightly pubescent along the nerves and reticulations, and with numerous, minute, shin- ing, pale-yellow, waxy glands ; nerves 9, radiate, very prominent, the lateral ones above the base about 10 on each side of the midrib; primary reticulations very prominent, parallel; petioles 392 MERRILL. stout, at least 50 cm in length, about 1 cm in diameter, glabrous, or near the apex more or less pubescent. Staminate inflores- cence apparently axillary, about 50 cm long, rather densely brown-pubescent with very short hairs, the lower branches up to 14 cm long, the upper gradually shorter, very numerous. Bracts very numerous, orbicular, concave, pubescent, 3 to 4 mm in diameter, minutely and irregularly toothed. Male flowers 6 to 10 in the axil of each bract, sessile. Calyx 0.6 to 0.8 mm long. Stamens 2 to 4. Female flowers and fruit unknown. Camiguin de Mindanao, Bur. Sci. H702 Ramos, March, April, 1912. A species in the same group with and allied to Macaranga noblei Elm., but very different from it. A single excellent distinguishing character is its nearly glabrous leaves. MACARANGA AMPLIFOLIA sp. nov. Arbor circiter 10 m alta, partibus junioribus dense ferru- gineo-tomentosis glabrescentibus, ramulis crassis, 1 ad 1.5 cm diametro, stipulis lanceolatis, acuminatis, 3 ad 4 cm longis, sub- persistentibus ; foliis coriaceis, alternis, usque ad 40 cm longis, integris, ovatis ad oblongo-ovatis, acuminatis, basi acutis vel subtruncatis, prominente 3-plinerviis, subtus dense pallide glan- duliferis; inflorescentiis axillaribus, paniculatis, diffusis, amplis, quam folia brevioribus; capsulis glandulosis, parvis, 2-coccibus, inermibus. A tree about 10 m high. Branchlets stout, 1 to 1.5 cm in diameter, brown, glabrous, the tips densely brown-tomentose, the stipules lanceolate, acuminate, 3 to 4 cm long, subpersistent, appressed-ciliate on the back. Very young leaves densely fer- ruginous-tomentose, soon becoming glabrous or nearly so. Leaves alternate, ample, ovate to oblong-ovate, coriaceous, 20 to 40 cm long, 9 to 20 cm wide, long and rather slenderly acu- minate, margins entire, base subtruncate or acute, prominently 3-plinerved, with one or two additional pairs of very short basal nerves, the upper surface shining, quite glabrous, brownish, the lower surface somewhat paler, rather densely covered with minute, pale, waxy glands, the nerves often slightly hairy; lateral nerves above the basal pair 9 or 10 on each side of the midrib, ascending, parallel, very prominent, anastomosing close to the margin, the reticulations subparallel, prominent; petioles about 20 cm long, at first pubescent, becoming glabrous or nearly so. Panicles axillary, very densely ferruginous-pubescent, ample, diffuse, many-flowered, usually exceeding the petioles in length, the bracts obovate, 3 mm long, subdigitately toothed. NOTES ON PHILIPPINE EUPHORBIACEAE. 393 Capsules with two cocci, each coccus globose, about 4 mm in diameter, glandular, unarmed. Luzon, Province of Laguna, near San Antonio, Bur. Sci. 13513 Ramos (type) August, 1910, in fruit; between Paete and Piapi, For. Bur. 10153 Curran, March, 1908, with very young flowers. A species characterized by its large, entire, acuminate, long-petioled leaves which are not at all peltate or cordate, its paniculate, densely pubescent inflorescence, and its unarmed capsules, each consisting of two small, globose, glandular cocci. MACARANGA LEYTENSIS sp. nov. Species M. gigantifoliae affinis differt foliis vix peltatis, basi latissime subtruncato-cordatis subauriculatisque, subtus eglan- dulosis, bracteolis distincte palmato-lobatis, lobis oblongis, brevibus, obtusis. A tree about 12 m high. Leaves broadly ovate, about 35 cm long, 25 to 30 cm wide, chartaceous, entire, the upper surface olivaceous when dry, shining, glabrous, or slightly pubescent on the nerves, the lower surface paler, shining, not at all glandular, ciliate with scattered, long, white hairs on the nerves and reticulations, apex subcaudate-acuminate, the acumen about 1.5 cm long, the base very broadly subtruncate-cordate, at the in- sertion of the petiole subauriculate, prominently 3-nerved, with 4 or 5 supplementary pairs of progressively shorter basal nerves, the longest about 9 cm long, the others very much shorter; lateral nerves above the basal pair about 12 on each side of the midrib, prominent, anastomosing close to the margin, the reticulations prominent, subparallel ; petioles more or less ciliate, about 30 cm long. Male panicles axillary, they or their branches 20 to 30 cm long, ferruginous-tomentose, the younger parts densely so, the older parts glabrescent and dark-colored when dry. Flowers very numerous, the bracts obovate, densely pubescent, 2.5 to 3 mm long, 4 to 6 fascicled at each node, subpalmately lobed, the lobes 10 to 12, oblong, obtuse, about 1 mm long, each bract subtending about 8 sessile flowers. Calyx about 0.8 mm long, narrowly funnel-shaped, obscurely 3-lobed, somewhat pubescent above. Stamens 2 or 3, the filaments about as long as the perianth. Female flowers and fruits not seen. Leyte, Mount Ibuni, near Dagami, Bur. Sci. 15253 Ramos, August 21, 1912, on forested slopes near streams. A species manifestly allied to Macaranga gigantifolia Merr., M. noblei Elm., and M. amplifolia Merr., distinguished from all by its eglandular leaves. It is, pei'haps, most closely allied to the first, from which it is also distinguished by its very differently shaped leaves, truncate-cordate and subauriculate at the base, not deeply peltate, and its different bracts. 394 MERRILL. MACARANGA GRANDIFOLIA (Blanco) comb. nov. Croton grandifolius Blanco FI. Filip. (1837) 753, ed. 2 (1845) 518 {grandifolium) . Macaranga porteana Andre in Rev. Hort. 60 (1888) 176, /. 36; Hook. f. in Curtiss’ Bot. Mag. (1895) t. 7407; Merr. in Philip. Journ. Sci. 7 (1912) Bot. 238. Macaranga mappa F.-Vill. Novis. App. (1880) 185; Vid. Sinopsis Atlas (1883) t. 84, f. D, Rev. PI. Vase. Filip. (1886) 246; Merr. in Philip. Journ. Sci. 1 (1906) Suppl. 80, non Muell.-Arg. This species is apparently distinct from the Malayan Macaranga mappa Muell.-Arg., the reference of Philippine material to which has been largely due to Mueller’s erroneous reduction of Blanco’s Croton grandifolius. In considering the species previously,” I erroneously considered Blanco’s specific name invalid in Macaranga, but there is no “Macaranga grandifolia Muell.-Arg.” to invalidate it, and the name is quite valid in Croton where originally described. I accordingly, transfer Blanco’s specific name to Macaranga as the oldest valid one for the species. It is represented by the following specimens : Luzon, Province of Cagayan, For. Bur. 16717 Curran: Benguet Subprov- ince, Sablan, Phil. PI. 473 Fenix: Province of Bataan, Lamao River, For. Bur. 2059 Borden, Whitford 12, 1052: Province of Rizal, Merrill 1288: Province of Tayabas, Bur. Sci. 13109 Foxworthy & Ramos. Mindoro, Ritchie s. n. Nearly all the material was distributed as Macaranga mappa Muell.-Arg. MACARANGA MONTANA sp. nov. Arbor circiter 12 m alta; foliis alternis, longe petiolatis, vix peltatis, oblongo-ovatis, usque ad 20 cm longis, acuminatis, in- tegris, basi late subrotundatis, subtus parce glandulosis, costa nervisque leviter pubescentibus, nervis utrinque circiter 11 ; paniculis ? axillaribus, quam petioli brevioribus, paucifloris, leviter pubescentibus; ovario dense glanduloso, 1-loculare; stig- mate laterali, lineari, papillose. A tree about 12 m high, subglabrous. Branches dark-colored when dry, terete, glabrous, the growing parts slightly lepidote with deciduous, brownish, minute scales. Leaves alternate, long-petioled, oblong-ovate, subcoriaceous, shining, 15 to 20 cm long, 8 to 10 cm wide, entire, apex shortly and rather sharply acuminate, base broad, subrounded, the upper surface quite glabrous, the lower surface of nearly the same color when dry, with scattered, minute, pale, waxy glands, the midrib and lateral nerves slightly pubescent ; nerves pinnately arranged, prominent, somewhat curved, anastomosing near the margin, about 11 on each side of the midrib, the reticulations subparallel; petioles 7 to 9 cm long, usually a little thickened and curved at the apex; stipules narrowly oblong, subacute, about 6 mm long, deciduous. Supra 238. NOTES ON PHILIPPINE EUPHORBIACEAE. 395 Male flowers and fruits unknown. Female inflorescence axillary, solitary, rather lax, paniculate, about 6 cm long, slightly pubes- cent; bracts oblong-lanceolate, acuminate, 3 mm long or less. Calyx about 1.5 mm long, irregularly splitting into 3 or 4, ovate- lanceolate, acuminate lobes. Ovary globose, densely covered with pale waxy glands, 1-celled, 1-ovuled ; stigma lateral, simple, papillose throughout, somewhat curved, tapering, 5 to 6 mm long. Mindanao, District of Zamboanga, Sax River Mountains back of San Ramon, Merrill 8076, November 28, 1911. In forests on ridges, altitude about 1,100 m. A species probably belonging in Macaranga, although the male flowers are unknown, well characterized by its 1-celled ovaries, lateral simple styles which are stigmatose throughout, and pinnately veined, non-peltate leaves. MAL LOTUS Loureiro. MALLOTUS ALTERNIFOLIUS sp. nov. § Eumallotus. Arbuscula vel arbor parva circiter 4 m alta, plus minusve glandulosa, novellis inflorescentisque stellato-puberulis ; foliis membranaceis, oblongo-ovatis vel ovatis, basi cordatis, tripli- nerviis, apice acuminatis; inflorescentiis S terminalibus, race- mosis, floribus numerosis, staminibus circiter 20 ; capsulis dense molliter echinatis. A shrub or small tree reaching a height of about 4 m. Branches terete, slender, glabrous, smooth, reddish or reddish-brown, the branchlets slightly puberulent. Leaves alternate, membrana- ceous, oblong-ovate or ovate, 8 to 15 cm long, 5 to 8.5 cm wide, the base rather broad, cordate, the basal lobes rounded, the sinus narrow, about 0.5 cm deep, the apex prominently and rather slenderly acuminate, the margins slightly subrepand-denticulate or nearly entire, the upper surface glabrous, somewhat shining, with usually two glands on either side of the midrid at the base, the lower surface slightly paler than the upper, with few, scattered glands, the midrib and nerves often slightly ciliate; basal pair of nerves reaching from one-third to one-half the length of the leaf, with usually two additional short, supplement- ary pairs of basal ones, the lateral nerves above the basal pair 5 or 6 on each side of the midrib, somewhat ascending, anasto- mosing, the reticulations subparallel, lax; petioles 2.5 to 5.5 cm long, usually puberulent. Flowers usually appearing with or just before the new leaves, the racemes terminal, erect, stellate-puberulent, the staminate ones 8 cm long or less. Flowers in fascicles of 10 or less, their pedicels 1 to 3 mm long, puberulent, the bracts none or very early deciduous. Calyx- 114346 — 3 396 MERRILL. segments 3, narrowly ovate, acute, about 2 mm long, 1-nerved, slightly stellate-puberulent. Stamens about 20, the filaments 2 mm long or less. Disk none. Pistillate flowers unknown. Young fruits glandular, puberulent, densely covered with nu- merous, soft, spine-like processes. Palawan, near Puerto Princesa, For. Bur. 3905, 4124. (type) Curran, April, 1906, Bur. Sci. 250 Bermejos, December, 1905. This species has been confused in the herbarium with Mallotus leuco- calyx Muell.-Arg., but differs from that species not only in its distinctly alternate leaves, not in unequal pairs, but also in many other characters; it is probably most closely allied to Mallotus helferi Muell.-Arg. MALLOTUS AURICULATUS sp. nov. § Eumallotus. Arbor parva, partibus junioribus plus minusve ciliata; foliis longe petiolatis, usque ad 15 cm longis, oblongo-ellipticis vel late ovato-lanceolatis, chartaceis, subtus parce glandulosis, pen- ninerviis, acuminatis, basi angustatis, biauriculatis, auriculis brevibus, obtusis, supra glandulis 1 ad 3 instructs; nervis utrinque circiter 8, subtus prominentibus ; stipulis setaceis; in- florescentiis $ axillaribus, solitariis, spicatis; fructibus pubes- centibus, valde muricatis. A small tree reaching a height of about 5 m, the branchlets, inflorescences and under surfaces of the leaves with scattered, rounded, colored or pale, waxy glands. Branches terete, brown- ish, glabrous, the branchlets, petioles, and midrib and nerves on the under surface of the leaves with scattered, long, white, ciliate hairs, ultimately becoming nearly glabrous. Leaves al- ternate, oblong-elliptic to broadly ovate-lanceolate, 8 to 15 cm long, 2 to 6 cm wide, brownish or somewhat olivaceous when dry, paler beneath, slightly shining, firmly chartaceous, the margins obscurely and distantly repand-denticulate or subentire, the apex subcaudate-acuminate, the acumen blunt, often 2 cm long, the base narrowed, biauriculate, the auricles about 2 mm long, rounded, the sinus very narrow, each auricle with from one to three, prominent, dark-colored glands on the upper surface, the lower surface, of the leaves with scattered, round, pale or colored, waxy glands; lateral nerves about 8 on each side of the midrib, curved-ascending, distant, prominent beneath, looped-anasto- mosing, the reticulations lax, prominent; petioles 2.5 to 5 cm long, when young ciliate, ultimately nearly glabrous; stipules setaceous, brown, acuminate, about 6 mm long. Staminate in- florescence axillary, solitary, spicate, the spikes about 5 cm long, pubescent, glandular, the bracteoles ovate to ovate-lanceo- late, acuminate, about 2 mm long, each subtending from one to several sessile or subsessile flowers. Calyx-segments obovate. NOTES ON PHILIPPINE EUPHORBIACEAE. 397 about 2 mm long, 1.3 mm wide, acute or obtuse, with scattered glands outside. Disk none. Stamens 20 to 25; filaments about 1.8 mm long, glabrous. Pistillate inflorescence axillary, ap- parently paniculate, the flowers unknown. Fruits brown, con- sisting usually of two dehiscent cocci, each coccus subglobose, about 8 mm in diameter, brown-puberulent, glandular, covered with numerous, stout, stiff, spreading spines 1.5 to 2 mm long; seeds brown, globose, about 5 mm in diameter. Mindanao, Lake Lanao, Camp Keithley, Mrs. Clemens 1013, April, 1907, and without number, June, 1907. Negros, Himugaan River, For. Bur. 7271 Everett, May 20, 1907 (type) ; Sicaba River, For. Bur. 7259 Everett, May, 1907; Faraon, For. Bur. 13563 Meyer & Foxworthy, August, 1909; For. Bur. 13696, 19393 Curran, Sept., 1909, June, 1910. Leyte, Dagami, Bur. Sci. 15298 Ramos, August, 1912. A species well characterized by its distinctly auriculate leaves. MALLOTUS BARBATUS (Wall.) Muell.-Arg. in Linnaea 34 (1865) 184; Hook. f. FI. Brit. Ind. 5 (1887) 428; J. J. Sm. in Koord. & Val. Bijdr. Boomsoort. Java 12 (1910) 407. Rottlera barbata Wall. Cat. (1847) No. 7822. Luzon, Province of Tayabas, Atimonan, Whitford 711, August, 1904. I refer this specimen here with some hesitation, as it has pistillate flowers only. It agrees, however, with the description in essential char- acters, differing chiefly in its leaves not being at all 3-lobed. In Mallotus barbatus, however, the leaves appear not to be always 3-lobed, as in Henry 9525 from Yunnan. Not previously reported from the Philippines; Burma to southern China, southward to Java. MALLOTUS CAMIGUINENSIS sp. nov. § Eumallotus. Arbuscula vel arbor parva, plus minusve glandulosa pubescens- que; foliis suboppositis vix vel paulo inaequalibus, apice longe acuminatis, basi cordatis, trinerviis ; calycis segmentis 4, reflexis ; staminibus circiter 35; capsulis 3-coccis, glandulosis, glabris, aculeis paucis circiter 2 mm longis vix rigidis instructis. A shrub or small tree, the younger branchlets, petioles, in- florescences, and nerves on the lower surfaces of the leaves more or less pubescent with simple hairs, and including the fruits and lower surfaces of the leaves with scattered, yellowish, waxy glands. Branches terete, reddish-brown, glabrous. Leaves oblong-ovate or narrowly ovate, membranaceous, 11 to 17 cm long, 5 to 7.5 cm wide, the base rather broad, rounded, cordate, the basal lobes short, rounded, the sinus narrow, the apex gradually long and slenderly acuminate, the margins distantly denticulate or subentire, the upper surface glabrous, or when young with few hairs on the nerves, the lower surface with numerous, scattered, yellowish, waxy glands, glabrous except for the pubescent midrib and primary nerves, the base distinctly 398 MERRILL. 3-nerved, the lateral nerves reaching one-third the length of the lamina, there anastomosing with the upper lateral nerves, the base also with two or three pairs of supplementary, very short, nerves; primary nerves above the basal ones 4 or 5 on each side of the midrib, ascending, anastomosing, distinct beneath, the reticulations rather lax; petioles 2.5 to 5.5 cm long, pubes- cent; stipules lanceolate, acuminate, about 3 mm long, early deciduous. Staminate inflorescence in the upper axils, solitary, racemose, slightly pubescent, about 10 cm long, the flowers arranged in fascicles of from 4 to 9 flowers each, their pedicels 1.5 to 2.5 mm long. Calyx-segments usually 4, reflexed, glan- dular, oblong to oblong-ovate, subequal, about 2.2 mm long, 1- or 2-nerved. Stamens about 35 ; filaments 2.5 mm long, glabrous, the connective not produced. Fruits about 1.5 cm in diameter, consisting of three dehiscent cocci, depressed, deeply sulcate between the cocci, glabrous, glandular, with few, scattered, rather soft, glabrous, spine-like processes about 2 mm in length. Babuyanes Islands, Camiguin, Bur. Sci. iOi? Fenix, June 27, 1907, in thickets in old clearings. This species has previously been reported by me as Mallotus leuco- calyx Muell.-Arg.,“ to which it is manifestly allied, but from which it differs strikingly in its fruit characters, M. leucocaiyx having capsules covered with numerous, elongate, softly pubescent spine-like processes. MALLOTUS CARDIOPHYLLUS sp. nov. § Eumallotus. Ut videtur arbor parva, partibus junioribus inflorescentiisque dense fulvo-stellato-tomentosis ; foliis cordato-ovatis, acuminatis, usque ad 10 cm longis, basi late cordatis, supra glabris, subtus eglandulosis, leviter hirsutis glabrescentibus ; racemis erectis, circiter 10 cm longis ; floribus masculinis pedicellatis ; staminibus circiter 50, locellis oblongis, connectivo baud producto. Apparently a shrub or small tree, the branches glabrous, dark-reddish-brown, terete, the growing parts, the young leaves, and the rachis of the inflorescence very densely covered with matted, fulvous, stellately arranged, tomentose hairs, the leaves when mature nearly glabrous. Leaves opposite, cordate-ovate, subcoriaceous, 6 to 10 cm long, 5 to 7 cm wide, base broadly rounded, usually prominently cordate and with two large, black glands near the insertion of the petiole, not at all peltate, the apex acuminate, pale and somewhat shining when dry, the upper surface glabrous, minutely papillose, the lower surface eglandular, more or less stellate-ciliate with pale-brown hairs, the indumentum more or less deciduous; basal nerves a single This Journal 3 (1908) Bot. 416. NOTES ON PHILIPPINE EUPHORBIACEAE. 399 pair reaching the middle of the leaf or above, the lateral nerves above the basal ones 3 or 4 on each side of the midrib, prominent, the reticulations subparallel; petioles 2 to 5 cm long, those of each pair of leaves unequal in length, when young densely tomentose, becoming glabrous or nearly so. Racemes terminal and in the upper axils, erect, many-flowered, 10 cm long or less, the rachis densely tomentose, the bracteoles lanceolate, acuminate, 1.5 mm long, tomentose; pedicels ciliate-tomentose, about 4 mm long. Buds globose, slightly tomentose, the lobes 3 or 4, in flower reflexed, elliptic to oblong-elliptic, acute or obtuse, often concave, 4 to 5 mm long. Stamens about 50; filaments up to 4 mm in length; anthers 1 to 1.3 mm long, the cells oblong, one often situated a little higher than the other, the connective very short, merely uniting the two cells. Pistil- late flowers and fruits unknown. Luzon, Province of Cagayan, Cuming 1267. This form, collected over 70 years ago, has not appeared in any recent collections. It is apparently undescribed, and does not appear to be closely allied to any other known Philippine form. Unfortunately the fruits are unknown, so that it is difficult to determine its proper alliance. M ALLOT US CAULIFLORUS sp. nov. § Eumallotus. Arbor parva, ramulis petiolis inflorescentiisque plus minusve pubescentibus ; foliis oblongis vel oblongo-ovatis, chartaceis vel subcoriaceis, usque ad 33 cm longis, glabris, penninerviis, nervis utrinque 11 ad 15, prominentibus, apice acuminatis, basi acutis, obtusis, vel subauriculato-cordatis, supra 2-glandulosis, subtus vix glandulosis; inflorescentiis truncifloris, fasciculatis et in axillis foliorum solitariis, racemosis, elongatis; coccis pellucido- glandulosis, echinatis. A small tree, 2 to 3 m high fide Curran, 10 m high fide Ramos. Branches, petioles, and inflorescence more or less pubescent, otherwise glabrous. Leaves opposite, penninerved, oblong to oblong-ovate, chartaceous or subcoriaceous, opposite, unequal, one of each pair considerably smaller and with a shorter petiole than the other but of about the same shape, prominently acu- minate, base more or less narrowed, acute, obtuse, or subau- riculate-cordate, 2-glandular on the upper surface at the base, the lower surface not at all waxy-glandular, quite smooth and glabrous; nerves 11 to 15 on each side of the midrib, prominent, anastomosing; petioles of the larger leaves about 5 cm long, of the smaller ones about one-half as long; stipules lanceolate, somewhat acuminate, about 2 cm long, striate. Inflorescence racemose, the racemes slender, up to 15 cm long, solitary in the axils of the leaves but more abundant on the trunk, springing 400 MERRILL. from elongate, rugose tubercles, solitary or several on each tubercle. Male flowers numerous, yellow, their pedicels pubes- cent, about 5 mm long, the subtending bracteoles ovate, obtuse or acute, about 1.5 mm long, stellate-pubescent. Sepals usually 3, reflexed, elliptic-ovate, acute, about 3.5 mm long. Stamens about 30, 2 to 2.5 mm long, the connectives broad, truncate, the anther-cells about 0.5 mm long. Fruit of two or three cocci, about 1 cm wide, depressed-globose, 3-sulcate, the individual cocci subglobose, externally prominently waxy-glandular with small yellow glands, and with scattered, short, spine-like processes 2 mm long or less. Luzon, Province of Cagayan, Abulug River, Bur. Sci. 13909 Ramos (type), January, 1912, For. Bur. 17794- Curran, January, 1912. On hill- sides, limestone region, altitude about 80 meters. A species well characterized by its cauline inflorescence, but frequently also with axillary racemes on the same specimen, its penninerved, glabrous, non-glandular leaves, etc. It somewhat resembles in vegetative characters Mallotus camiguinensis Merr., but is not closely allied to that species. MALLOTUS PANICULATUS (Lam.) Muell.-Arg. in Linnaea 34 (1865) 189, DC. Prodr. 15’ (1866) 965. Croton paniculatus Lam. Encycl. 2 (1786) 207. Mallotus cochinchinensis Lour. FI. Cochinch. (1790) 635; Hook. f. FI. Brit. Ind. 5 (1887) 430. Croton appendiculatus Elm. Leaf!. Philip. Bot. 1 (1908) 312. Luzon, Province of Tayabas, Lucban, Elmer 9215, May, 1907, in fruit, cotype of Croton appendiculatus Elm.: Province of Laguna, San Antonio, Bur. Sci. 10990 Ramos, August, 1910, with staminate flowers. Negros, Faraon, For. Bur. 17465 Curran, October, 1909, with pistillate flowers. Mindanao, District of Davao, Elmer 11926, October, 1909, with staminate flowers and fruits. This species has previously been reported from the Philippines by F.-Villar”, but the record was not before verifled. The specimen re- ported by me as Mallotus paniculatus “ is M. playfairii Hemsl. Croton appendiculatus Elm., was described from fruiting specimens only, and has been reduced to Mallotus cochinchinensis Lour. (=M. paniculatus Muell.- Arg.) by J. J. Smith,” and by Hallier f.” There is no doubt as to the correctness of the reduction, for the Philippine specimens agree both with the descriptions of Mallotus paniculatus, and with extra-Philippine material representing the species in the Herbarium of the Bureau of Science: India, Wallich 7818, Hongkong, Curran, Singapore, Ridley, Penang, Ridley, and Borneo, Sarawak, Foxworthy 87, 146. “Novis. App. (1880) 195. “For. Bur. Bull. (Philip.) 1 (1903) 31. ” Koord. & Val. Bijdr. Boomsoort. Java 12 (1910) 413. “ Meded. ’s Rijks Herb. 1910 (1911) 7. NOTES ON PHILIPPINE EUPHORBIACEAE. 401 MALLOTUS MOLUCCANUS (L.) Muell.-Arg. in Linnaea 34 (1865) 958, DC. Prodr. 15^ (1866) 958. Croton moluccanus L. Sp. PI. (1753) 1005. This species is very common in the Philippines, extending from northern Luzon to southern Mindanao, being one of the characteristic trees of the settled areas, second growth woodlands, etc. Its known extra-Philippine range is from Formosa southward through the Malay Archipelago fo Sumatra, and New Guinea, with a variety in the Marianne Islands. There is some doubt as to the type of the species, for Linnaeus in the original description of the species* cites first FI. zeyl. 3U6, and second Burm. zeyl. 170, giving the range of the species as Ceylon and the Mo- luccas. Now the species does not occur in Ceylon, and Trimen states that the specimens in Hermann’s herbarium to which Linnaeus had access are only leaves and almost certainly belong to Givotia rottleriformis Griff. Muell.-Arg.,*" however, indicates that he has seen the specimen in the Linnean herbarium and cites it with certainty as being Mallotus moluc- canus; there is no indication, however, that the specimen in Linnaeus’ her- barium is the type. Unless the reference to Burmann is in part Mallotus moluccanus, then it would seem that the specific name would have to be transferred to Givotia, and a new one adopted to designate the plant that is usually known as Mallotus moluccanus Muell.-Arg. Our Philippine material is somewhat variable, and most of the speci- mens have very densely tomentose capsules; a form from Mindanao has capsules glabrous or nearly so, and is indicated as follows: Var. PENDULUS var. nov. Inflorescentiis quam in typo longioribus, pendulis, usque ad 60 cm longis, capsulis glabris vel subglabris, baud dense to- mentosis. Mindanao, Butuan Subprovince, Veruela, Weber llOi, July 3, 1911: District of Zamboanga, Basagan River, Merrill 8305, November, 1911: District of Cotabato, Bur. Sci. 11552 Robinson, June, 1910. MALLOTUS RAMOSII sp. nov. Arbor parva, circiter 4 m alta, inflorescentiis parcissime hirsutis exceptis glabra; foliis altemis, chartaceis, ellipticis vel ovato-ellipticis, usque ad 14 cm longis, eglandulosis, penninerviis, basi rotundatis, apice tenuiter acuminatis, margine crenato-ser- ratis, nervis utrinque 8 ad 10, distinctis; inflorescentiis $ axil- laribus, spicatis, 12 ad 20 cm longis, floribus numerosis, ad nodos fasciculatis. A small tree, glabrous except the very slightly hirsute in- florescence, about 4 m high. Branches grayish, terete, glabrous. Leaves alternate, chartaceous, elliptic to elliptic-ovate, sometimes FI. Ceyl. 4 (1898) 51. DC. Prodr. 15* (1862) 958. 402 MERRILL. oblong-ovate, 8 to 14 cm long, 3 to 8 cm wide, eglandular, of the same color and slightly shining on both surfaces when dry, the base rounded, the apex rather slenderly acuminate, margins dis- tinctly crenate-serrate ; lateral nerves 8 to 10 on each side of the midrib, distinct, anastomosing near the margins, the reticula- tions slender, the primary ones subparallel; petioles 1 cm long. Male inflorescence spicate, very slender, 12 to 20 cm long, axil- lary, solitary, slightly hirsute at the nodes, the flowers numerous, fascicled or subsolitary at the nodes, the internodes 1 to 4 mm long. Sepals 3, broadly ovate, acuminate, 2 mm long. Sta- mens many; filaments short. Female flowers and fruits not seen. Camiguin de Mindanao, Bur. Sci. H602 Ramos, April 27, 1912, in forests near the summit of Mount Catarman. A species difficult to refer to its proper subgenus in the absence of female flowers and fruits, but well characterized by its short-petioled, eglandular, pinnately nerved, crenate-serrate leaves, and by its slender, elongate staminate spikes. MALLOTUS SANCHEZII sp. nov. § Eumallotus. Arbor parva vel frutex, plus minusve stellato-villosus ; foliis anguste ovatis ad oblongo-ovatis, usque ad 20 cm longis, tenuiter caudato-acuminatis, basi late rotundatis, peltatis, utrinque glan- dulosis, margine repando-dentatis ; capsulis depressis, glandu- losis, processibus mollibus ciliatis instructis. A shrub or small tree, the branchlets, inflorescence, and both surfaces of the leaves with scattered, stellate, pale or brownish, ciliate hairs, the tips of the branches very densely pubescent. Leaves alternate, narrowly ovate to oblong-ovate, chartaceous, 12 to 20 cm long, 6 to 11 cm wide, brownish and somewhat shining when dry, of about the same color on both surfaces, the apex long and slenderly caudate-acuminate, the acumen 2.5 to 3 cm long, the base broadly rounded, peltate, the petiole inserted at from 1 to 1.5 cm from the margin, both surfaces with scat- tered, stellate, pale, ciliate hairs, and with scattered, shining, yellowish, waxy glands, the margins repand-dentate, the teeth short, less than 1 cm apart, the basal margin quite entire; basal nerves 7 to 10, radiate, the lower ones short; lateral nerves above the basal ones 6 to 8 on each side of the midrib, ascending, prom- inent, the reticulations subparallel, prominent; petioles 5 to 10 cm in length, more or less ciliate-pubescent ; stipules linear, densely pubescent, 1 cm long or less, deciduous. Inflorescence NOTES ON PHILIPPINE EUPHORBIACEAE. 403. racemose, terminal, up to 20 cm in length. Flowers unknown, the bracteoles densely pubescent, linear, about 5 mm long. Capsules depressed, of three cocci, about 1 cm wide, waxy- glandular with yellowish dots, and covered with numerous, scat- tered, spreading, ciliate, rather soft processes 2 to 3 mm in length. Mindanao, Province of Misamis, Talisayan, Q. F. Quadras. The type is from the herbarium of the Ateneo de Manila, communicated to me by Father Francisco Sanchez, to -whom this most distinct species is dedicated. A strongly marked species, readily recognizable by its peltate, oblong- ovate to narrowly ovate, long and slenderly caudate-acuminate, repand- dentate leaves. ' OSTODES Blume. OSTODES ANGUSTIFOLIA sp. nov. Frutex glaber circiter 1 m altus; foliis plus minusve con- fertis, alternis, sessilibus vel subsessilibus, lineari-lanceolatis, acuminatis, basi cordatis, usque ad 27 cm longis, 1 ad 1.5 cm latis ; peduneulis tenuibus, circiter 8 cm longis, ut videtur paucifloris. A glabrous shrub or undershrub about 1 m high. Branches terete, slender, grayish. Leaves somewhat crowded toward the apices of the branchlets, sessile or subsessile, firmly char- taceous or subcoriaceous, linear-lanceolate, 18 to 27 cm long, 1 to 1.5 cm wide, gradually narrowed into the slender acumen, base abruptly cordate, often somewhat surrounding but not clasping the stem, shining when dry; nerves 0.8 to 1 cm apart, spreading, anastomosing, rather distinct beneath but slender, the reticulations lax. Peduncles axillary, slender, about 8 cm long, apparently few-flowered. Pedicels about 1.5 cm long, slightly thickened upward, and somewhat sulcate or angled at the apex. Sepals persistent, oblong-ovate, acute or obtuse. Ovary glabrous, 3-celled, cells 1-ovuled ; styles 3, cleft nearly to the base. Mindanao, District of Zamboanga, Port Banga, For. Bur. 93^3 Whit- ford & Hutchinson, January 14, 1908, in dipterocarp forests, altitude about 30 m. This species has been described from imperfect material, no flowers having been seen, and only one immature fruit; it belongs, with very little doubt, to the genus Ostodes, and is well characterized by its sessile, very long and narrow leaves which are abruptly cordate at the rather broad base. 404 MERRILL. OSTODES MACROPHYLLA (Muell.-Arg.) Benth. ex Pax in Engl. Pflan- zenreich 47 (1911) 18. Tritaxis macrophylla Muell.-Arg. in Flora 47 (1864) 482. Trigonostemon macrophyllus Muell.-Arg. in Linnaea 34 (1865) 213, DC. Prodr. 15=^ (1866) 1106. Ostodes serrato-crenata Merr. in Philip. Journ. Sci. 4 (1910) Bot. 283. Pax is perfectly correct in his recent reduction of Ostodes serrato- crenata Merr. to O. maerophylla (Muell.-Arg.) Benth., a conclusion I had already reached from an examination of the abundant material now in the Herbarium of the Bureau of Science. It is represented by the following material : Luzon, Province of Cagayan, For. Bur. 7078, 11271 Klemme, For. Bur. 1524.7 Bernardo: Province of Laguna, For. Bur. 22305 Mariano, Bur. Sci- 10051, 10966 Ramos, For. Bur. 17637 Curran. Sibuyan, Elmer 12366, 12480, April, May, 1910. It is also represented by a specimen from Sarawak, Borneo, Foxworthy 141, so that its known range is now from the Malay Peninsula, through northern Borneo to Sibuyan, central and northern Luzon. PHYLLANTHUS Linnaeus. PHYLLANTHUS LU20NIENSIS sp. nov. § Cicca, Prosorus. Arbor parva, 6 ad 9 m alta, dioica, decidua, partibus junioribus plus minusve ciliatis, vetustioribus glabris; foliis subellipticis vel oblongo-obovatis, rotundatis, usque ad 5 cm longis; floribus 4-meris, S fasciculatis, tenuiter pedicellatis, biseriatim 4-par- titis, disco annulari, disciformi; fructibus crustaceis, in valvis 4 dehiscentibus. A small, dioecious, deciduous tree 6 to 9 m high, in age quite glabrous, but the young branchlets, young leaves, and flowers more or less ciliate with weak, scattered, brownish hairs. Branches terete, grayish, the branchlets lenticellate, slender, sometimes reddish-brown. Leaves when young, at time of an- thesis, membranaceous, in age becoming quite glabrous, scattered, subcoriaceous, subelliptic to oblong-obovate or obovate, usually rounded at the apex or obtuse, the base acute, 2.5 to 5 cm long, 1 to 2.5 cm wide, shining, the lower surface a little paler than the upper one; nerves slender, 5 or 6 on each side of the midrib, anastomosing, the reticulations not prominent; petioles slender, 2 to 4 mm long ; stipules lanceolate, acuminate, ciliate, deciduous, about 1.5 mm long. Male flowers fascicled, appearing with the leaves, their pedicels slender, 5 to 8 mm long, ciliate. Perianth- lobes 4, in two pairs, 1.5 to 1.8 mm long, slightly ciliate, two broadly ovate to obovate, rounded, about 1.5 mm wide, the other 4wo oblong, about 0.8 mm wide. Disk annular, disciform, about Pflanzenreich 52 (1912) 286. NOTES ON PHILIPPINE EUPHORBIACEAE. 405 0.8 mm in diameter. Stamens 4, the filaments 1.5 mm long, free; anthers 0.8 mm long, opening by vertical slits. Female flowers not seen, apparently solitary, their pedicels, in fruit, up to 1.5 cm long, the persistent calyx of 4, reflexed, broadly ovate, 2 mm long sepals, the disk annular, disciform, about 1.3 mm in diameter. Fruit globose, glabrous, crustaceous, about 8 mm in diameter, splitting into 4 valves, containing 3 or 4 seeds. Seeds 4 mm long, their outer surface convex, the inner two plane, usually of a brilliant blue color. Luzon, Province of Pampanga, Mount Arayat, For. Bur. 17698 Curran (type), March, 1910, on dry slopes, altitude about 50 m, male flowers: Province of Nueva Ecija, For. Bur. 8JfS5 Curran, with very old fruits. For. Bur. H320 Saraca, sterile: Province of Bataan, Mount Mariveles, on dry slopes, altitude 10 to 200 m. For. Bur. 63^8, 6391 Curran, sterile. CORREGIDOR, For. Bur. 13225 Curran, September, 1911, with mature fruits. This species apparently belongs in the subsection Prosorus of the sec- tion Cicca, following Mueller’s classification. I am of the opinion, how- ever, that the subsection Eucicca should be maintained as a distinct genus, Cicca, as originally proposed by Linnaeus, and as recently considered by Doctor Robinson. The present species very strongly resembles Fluggea virosa Bail!., in its vegetative characters and in its inflorescence, espe- cially those specimens collected at time of anthesis. Several sterile speci- mens were collected by Mr. Curran, with the hope that the species could be determined from such material, as the species is abundant locally on dry hills, and is one of the dominant if not the dominant species over considerable areas in the Provinces of Bataan, Pampanga, and Nueva Ecija. Although the male flowers have been known for some time, our material did not show any of the details of the female flowers until it was recently collected on Corregidor Island by Mr. Curran. PHYLLANTHUS ROBINSON 1 1 sp. nov. § Emblicastrum? Frutex circiter 2 m altus, glaber; foliis distichis, imbricatis, coriaceis, nitidis, 4 ad 7 mm longis, anguste obovatis, leviter obliquis, acutis, nervis obsoletis; floribus 9 solitariis, 5-meris, sessilibus, 2.5 ad 3 mm longis, disco angusto, ovario anguste ovoideo, stylis brevissime 3-lobatis. A shrub about 2 m high, glabrous, the branches slender, terete, rather stiff, grayish or brownish, the branchlets 4 to 6 cm long, slender, distinctly angled, solitary, in pairs, or somewhat fascicled at the nodes. Leaves numerous, small, shining, dis- tichous, imbricating, narrowly oblong-obovate, somewhat oblique, 4 to 7 mm long, 2 to 3 mm wide, acute, subsessile, the midrib very obscure, the lateral nerves obsolete; stipules about 1 mm long, deciduous. Female flowers solitary, sessile, axillary, in a single row on the lower side of the branchlets, the bracteoles lanceolate, acuminate, brown, 1.5 mm long. Perianth 2 to 2.5 mm long, the lobes 5, unequal, oblong, obtuse, coriaceous, 1.5 406 MERRILL. to 1.8 mm long, or the inner one shorter. Disk annular, narrow, distinct. Ovary narrowly ovoid, 1.2 mm long, glabrous; style about 1 mm long, columnar, exserted about 0.5 mm, brown, very slightly 3-lobed at the apex. Male flowers unknown. Fruit globose, about 3.5 mm in diameter, slightly villous, splitting into cocci. Cebu, Buacao, Bur. Sci. 11059 Ramos, March 19, 1912, on dry hills. This species is similar to, and probably closely allied to Phyllanthus curranii C. B. Rob., although its male flowers being unknown, its proper section is uncertain. It is distinguishable by its nerveless leaves. An- other form, apparently undescribed, from Bataan Province, Luzon, is apparently still more closely allied than is P. curranii, but its leaves have distinct nerves. SAPIUM P. Browne. SAPIUM SANCHEZ 1 1 sp. nov. § Parasapium. Arbor parva circiter 5 m alta, glabra, vel partibus junioribus parce pubescentibus ; foliis oblongis vel oblongo-ellipticis, char- taceis vel subcoriaceis, usque ad 10 cm longis, acute acuminatis, basi acutis vel rotundatis, margine leviter crenato-serratis, nervis utrinque 15 ad 20, baud prominentibus ; capsulis ovoideis, 2 cm longis, inermibus, abrupte acuminatis; seminibus ellipsoideis, glabris, maculatis, 8 mm longis. A small tree, about 5 m high fide Bolster, glabrous, or the young twigs sparingly pubescent (inflorescence not seen). Branches terete, gray, lenticellate, the twigs reddish-brown. Leaves alternate, oblong to oblong-elliptic, chartaceous or sub- coriaceous, 5 to 10 cm long, 2 to 3 cm wide, the upper surface usually shining, the lower dull, often somewhat glaucescent, the apex usually sharply acuminate, base acute or somewhat rounded, 2-glandular or glandless, margins finely crenate-serrate ; nerves 15 to 20 on each side of the midrib, slender, not prominent, spreading, somewhat curved, anastomosing, the reticulations not distinct; petioles slender, about 8 mm long. Inflorescence ter- minal, flowers not seen. Pedicels in fruit 1.5 to 2 cm long, the young fruits apparently fleshy, their style-arms terete, re- curved. Mature fruits ovoid, black and shining when dry, 2 cm long, nearly as wide, unarmed, obscurely reticulate, the base broadly rounded or somewhat heart-shaped, the apex abruptly sharply acuminate, composed of 3 dehiscent cocci, the mesocarp hard, woody, the columella short. Seeds ellipsoid, 8 mm long, equally rounded at both ends, not at all keeled, smooth and shining, one side nearly black, but the most of the surface blotched and spotted with grayish-brown. Mindanao, Province of Surigao, Surigao, F. H. Bolster 373, August 15, 1906 (type), along tidal streams; Mainit, J. F. Quadras 209, ex herb. NOTES ON PHILIPPINE EUPHORBIACEAE. 407 Ateneo de Manila, locally known as bantiyano: District of Cotabato, For. Bur. 187 8 Jf Tarrosa, Miranda, & Rafael, For. Bur. 18289 Miranda, June, 1912. A species manifestly closely allied to Sapium virgatum 'Hook. f. and to 5. indicum Willd., differing from the former in its smaller fruits and differently shaped seeds. Prom the latter it differs in its ovoid, promi- nently acuminate, not depressed-globose fruits and in its smaller, differ- ently shaped, not at all keeled seeds. Dedicated to Father Francisco Sanchez of the Ateneo de Manila. SAUROPUS Blume. SAUROPUS ROBINSON 1 1 sp. nov. § Ceratogynum. Frutex erectus, glaber, 1 ad 1.5 m altus, ramis teretibus, ramulis plus minusve angulatis vel angustissime alatis; foliis alternis, oblongo-ovatis ad anguste ovatis, membranaceis vel chartaceis, usque ad 18 cm longis, acuminatis, basi rotundatis vel subacutis, breviter petiolatis, nervis utrinque 7 vel 8 ; floribus 5 3 ad 4.5 mm, $ 6 ad 8 mm diametro. An erect, monoecious, slightly branched, glabrous shrub 1 to 1.5 m high, the stems naked, the leaves mostly at the top; branches terete, the branchlets more or less compressed, often narrowly winged, frequently somewhat zig-zag. Leaves oblong- ovate to narrowly ovate, membranaceous or chartaceous, 8 to 18 cm long, 3 to 8 cm wide, olivaceous when dry, somewhat shining on both surfaces, the lower surface a little paler than the upper one, apex acuminate, base rounded or subacute; nerves 7 or 8 on each side of the midrib, prominent beneath, curved, anas- tomosing, the reticulations rather lax, distinct; petioles about 3 mm long; stipules lanceolate, acuminate, 4 to 5 mm long. Male flowers axillary, fascicled, few in each axil, their pedicels 4 to 9 mm long. Calyx 3 to 4.5 mm in diameter, 6-lobed, the lobes broadly ovate, rounded, their appendages surrounding the an- thers, thick, erect, suborbicular, 0.5 mm long or less. Female flowers axillary, solitary, subpendulous, their pedicels 2 to 2.5 cm long, thickened upward, in fruit elongated, and up to 6 cm in length. Calyx dull-red, 6 to 8 mm in diameter, the lobes 6, broadly ovate, obtuse or acute, imbricate, 3 to 3.5 mm long, persistent, somewhat accrescent in fruit and 5 to 6 mm long. Ovary 3-celled; cells with two ovules; stigma broad, flat, about 3 mm in diameter, the branches stout, recurved, borne on the margin of the truncate top of the ovary. Fruit pink or red, somewhat fleshy, subglobose, 1 to 1.5 cm in diameter, ultimately dehiscent, 3-celled, 6-seeded. Luzon, Benguet Subprovince, Baguio, Elmer 6i^l (type), June, 1904, 6308, May, 1904; Twin Peaks, Elmer 5907, March, 1904. Mindanao, Lake Lanao, Camp Keithley, Mrs. Clemens 939, February, 1907. 408 MERRILL. A distinct species, entirely different from the two previously known from the Archipelago. Named in honor of Dr. C. B. Robinson, formerly of the Bureau of Science, who monographed the Philippine Phyllanthinae.^" The material cited above had been distributed into the herbarium without having been studied, as Solanum sp., and hence was overlooked by him. TRIGONOSTEMON Blume. TRIGONOSTEMON EVERETTII sp. nov. Arbor parva, inflorescentiis gemmisque exceptis glabra; foliis chartaceis, oblongis, nitidis, acuminatis, basi acutis, integris, usque ad 18 cm longis, nervis utrinque 6 ad 8, petiolo 1 ad 2 cm longo ; inflorescentiis brevibus, oppositifoliis, cymosis, paucifloris ; floribus 5-meris, s circiter 1,5 cm diametro, petalis obcordatis; antheris 3, fllamentis supra liberis; fructibus circiter 8 mm diametro, ovoideis vel subglobosis. A small tree or shrub about 6 m high, glabrous except the buds and the slightly pubescent inflorescence. Branches slender, terete, reddish-brown or grayish, the ultimate ones about 1 mm in diameter. Leaves alternate, chartaceous, when dry brownish, somewhat shining, of about the same color on both surfaces, oblong, 7 to 18 cm long, 2 to 6.5 cm wide, apex rather prominently acuminate, the base acute, with usually two small glands on the upper surface at the junction with the petiole, shortly 3-nerved; nerves 6 to 8 on each side of the midrib, rather prominent, curved-ascending, anastomosing near the margin, the reticula- tions rather lax, distinct; petiole 1 to 2 cm long. Cymes leaf- opposed, as long as the petiole or somewhat shorter, slightly appressed-pubescent, each with 3 to 5 branches, the flowers racemosely disposed, but one or two developing at a time, each pedicel subtended by 2, oblong, 1.5 mm long bracteoles, the pedicel and the two bracteoles forming a whorl. Flowers red, mon- oecious, 5-merous. Male flowers: Pedicels slender, about 1 cm long. Sepals 5, free, elliptic-oblong, imbricate, obtuse, concave, about 4 mm long, 2 mm wide. Petals 5, free, imbricate, spread- ing, about 8 mm long, 7 mm wide, broadly obovate and the apex deeply retuse, or obcordate, apex broad, the sinus narrow, rounded, about 3 mm deep, the lobes broadly rounded, base narrowed, acute. Disk flat, entire, about 1.5 mm in diameter, the glands none. Staminal column 1.5 to 2 mm long ; anthers 3, 0.8 to 1 mm long, basiflxed, the cells close, parallel, dehiscing longitudinally; filaments 0.5 mm long, spreading. Rudimentary ovary none. Capsules ovoid or subglobose, about 8 mm in diameter, of three dehiscent 1-seeded cocci, the pedicels about This Journal 4 (1909) Bot. 71-105. NOTES ON PHILIPPINE EUPHORBIACEAE. 409' 3 cm long, thickened upwards, the persistent sepals 5, reflexed, oblong, about 4 mm long. Negros, Mount Silay, in forests at an altitude of about 500 m, For. Bur. 7257 Everett, (type) September, 1907; Sicaba River, For. Bur. 558U Eve^ rett, November, 1906. This very characteristic species is not referable to any recognized sec- tion of the genus Trigonostemon, although it apparently belongs in that genus. In essential characters it agrees with Trigonostemon, differing in its leaf -opposed cymose inflorescence, and in its filaments being united below into a column but free and spreading above. It is possible that the examination of a full series of specimens and direct comparison with the known species of Trigonostemon may show the characteristic features of the present species to be sufficiently distinct to warrant its separation as a distinct genus. TRIGONOSTEMON OBLONGI FOLI US sp. nov. § Eutrigonostemon. Arbor parva usque ad 6 m alta, subglabra; foliis oblongis, usque ad 18 cm longis, integris, obtusis, basi acutis, coriaceis, nervis utrinque circiter 9, distantibus, reticulis subobsoletis ; petiolo 1.5 ad 2.5 cm longo; paniculis angustis, axillaribus, folia subaequantibus, longe pedunculatis ; floribus s numerosis, sub- fasciculato-glomeratis, 3 mm longis; antheris 3, apice baud bifidis; capsulis depresso-globosis, pubescentibus, circiter 1 cm diametro. A small tree or a shrub 6 m high or less, nearly glabrous. Branches terete, brownish, striate when dry, glabrous, the growing parts more or less fulvous-hirsute with appressed hairs, these hairs often persisting in the axils of the petioles. Leaves alternate, entire, oblong, coriaceous, 9 to 18 cm long, 2.5 to 5 cm wide, coriaceous or subcoriaceous, narrowed at both ends, the apex obtuse, the base acute, glabrous or nearly so, rather pale when dry, somewhat shining, the lower surface often brownish; nerves about 9 on each side of the midrib, distinct on the lower surface, curved-anastomosing, the reticulations obsolete or nearly so; petioles 1.5 to 2.5 cm long, sparingly appressed-pubescent, ultimately glabrous or nearly so. Panicles axillary, solitary, narrow, long- and slenderly peduncled, slightly pubescent, the peduncles 7 to 10 cm long, less than 1 mm in diameter. Male flowers very numerous, more or less glomerate- fascicled, pedicelled, the bracts lanceolate, 2 to 3 mm long, the bracteoles much smaller. Sepals imbricate, the outer two about 1.6 mm long, the inner three larger, about 2.5 mm long, orbic-- ular-ovate, concave, imbricate. Petals 5, dark-purple when dry, free, obovate, about 3 mm long, 2.5 mm wide, apex rounded- truncate, base much narrowed. Disk cupular, truncate, about 0.4 mm high. Filaments entirely united, nearly 1 mm long; 410 MERRILL. anthers 3, at the top of the staminal column, about 0.7 mm long, 2-celled, extrorse, apex scarcely bifid or divaricate. Female flowers not seen, 5-merous. Pedicels in fruit about 1.5 cm long, thickened upwards, the persistent calyx about 6 mm in diameter, the lobes ovate, 1.5 to 2 mm long. Ovary densely villous; stig- mas 3, cleft, the lobes 0.5 mm long. Mature capsules depressed- globose, about 8 mm long, 10 mm wide, appressed-hirsute ex- ternally, pale, splitting into 3 dehiscent cocci. Seeds ovoid, smooth, mottled, somewhat pointed at one end, about 6 mm long. Luzon, Province of Cagayan, Abulug River, Bur. Sci. 13965 Ramos (type), February, 1912; trail between Viola and Mount Narig, For. Bur. 1960Jf Curran, January, 1912, both with male flowers; Province of Ilocos Norte, Bur. Sci. 77i5 Ramos, March, 1909, with mature fruits. A species probably most closely allied to Trigonostemon sumatranus Pax & K. Hoffm., but with differently shaped, obtuse, not caudate-acuminate leaves, narrow panicles, etc. An allied species is represented by Trigonostemon longipedunculatus (Elm.) Leafl. Philip. Bot. 4 (1911) 1306 (Croton longipedunculatus Elm. 1. c. 1 (1908) 311, represented by Elmer 726k from Leyte, and Piper 325 from Butuan Subprovince, Mindanao. T. longipedunculatus Elm., differs, however, from the species above described, in its larger, sparingly appressed- hirsute leaves which are acute or acuminate, not obtuse, and which have from 15 to 18 pairs of lateral nerves. The Philippine Journal of Science, C. Botany. Vol. VII, No. 6, December, 1912. ROXBURGH’S HORTUS BENGALENSIS. By C. B. Robinson. {From the New York Botanical Garden, New York, U. S. A.) Carey’s introduction to this work is dated June 4, 1814, and shows that the text had already been printed;^ the title-page is dated 1814. Consecutively paged after the Hortus Bengalensis, is “A Catalogue of Plants, described by Dr. Roxburgh, in his Mss. Flora Indica, but not yet introduced into the Botanical Garden.” Although its title-page is dated 1813, it must have been printed after the Hortus Bengalensis, of which it prac- tically forms a part, and it appears safe to keep the accepted date, 1814, as that of publication of both. Together, these contain nearly 4,000 binomial names, and it has been customary to consider either, that all such as appear for the first time are therein published, or that none of them are deserving of such recognition. A brief consideration of four cases, each typical of many, will be sufficient to show that the binomial names of the Hortus Bengalensis fall into at least four classes. Systematic names. Synonyma & Remarks. Native place. Donors and time, &c. Dura- tion & habit. Time of flowering. Time of CANNA. America.- W. Hamilton. Esq. 1799. 1795 H. E. R. S. C. S. PHRYNIUM. dichotomum, R. H. SeetMl-patee. 3. India. &c . b E. H. S. apicatum, R. Pegu Mr. F. Carey, Tf E. R. S. GRATIOLA. amara, R. 6. Moluccas - 1809. H. C. 1800 c. s. O. D. C. S. ^ “In examining the catalogue, several errors in the terminations of the specific names have been detected The greatest part of it was printed off before this was discovered.” 114346 4 411 412 ROBINSON. There is clearly no sufficient description for publication of the name Canna glauca, but this had already been accomplished by Linnaeus.'^ Neither is Phrynium spicatum published, but this is its first appearance, and Roxburgh himself subsequently ® printed for it a sufficient diagnosis. The “3” after Phrynium dichotomum refers to a footnote, which is “S. Seetwla-pz^trika. H. A. 4. t. 7.” This enables the insufficient information given in the text to be supplemented by the figure cited in Rumphius’ Herbarium Amboinense, which is accompanied by text, and is an ample description of the species. But Roxburgh had already ^ published this name. Similarly, the “5” after Gratiola amara, points to the footnote “Karanga amara. Vahl. H. A. 5. t. 170. 1 1.” This is the first appearance of Gratiola amara, and it is here published. Vahl actually wrote the generic name as Caranga,^ and Jussieu as Curanga;^ both cited the figure of Herbarium Amboinense. There are ninety-one cases, approximately of the same nature as the last of these, where a new binomial is published, by cita- tion either of a previous binomial, or of a sufficient description with or without a plate. Admittedly, some of those referring to a previous binomial do so merely by inference, the generic name only being given. However, even these would be con- sidered to be published, by almost all botanists, when the specific name was used by the author cited under the given generic name. As it happens, this is not always true of the new com- binations made in such fashion by Roxburgh; some, therefore, must be rejected as unpublished; and no other of the names in the Hortus Bengalensis based on a binomial of earlier authors has stood the test of time. Oldenlandia herbacea may prove to be an exception. It may fairly be questioned whether there was any intention of publishing species in the Hortus Bengalensis, but Carey, at least, thought that this was being done. “The number of de- scribed species now in the garden amounts to about Three Thou- sand Five Hundred; for the knowledge of no fewer than One Thousand Five Hundred and Ten of which, as named and de- scribed in this catalogue, we are indebted to his (Roxburgh’s) indefatigable and discriminating researches.” There is much reason to believe, if from nothing else than the title of the second “ Sp. PI. (1753) 1. 'FI. Ind. 1 (1820) 5. 'As. Res. 11 (1810) 324. 'Enum. PI. 1 (1805) 100. 'Ann. Mus. Hist. Nat. 9 (1807) 319. ROXBURGH’S HORTUS BENGALENSIS. 413 part of the catalogue, that all of these had been written up by Roxburgh for his Flora Indica from living plants; and that he would have considered the plates cited in the Hortus Bengalensis mere synonyms, rather than the foundation for his species. We are nevertheless confronted with the fact, that they can not be considered as published in the earlier work except by the citation of these plates, and that substantially similar publication by very many other authors has obtained universal recognition. For various reasons, it seems desirable to list the species that appear to have been published in the Hortus Bengalensis. Species Published On Page By Citation of Gratiola amara 4 Gratiola integrifolia 4 Anthistiria cymbaria 6 Andropogon bicolor 7 Andropogon saccharatum 7 Panicum spicatum 7 Ixora fulgens 10 Ixora stricta var. 1. 10 Ixora stricta var. 2. 10 Ixora barbata 10 Pladera virgata 10 Pladera pusilla 10 Oldenlandia alata 11 Oldenlandia crystallina 11 Oldenlandia diffusa 11 Oldenlandia herbacea 11 Brucea sumatrana 12 Ipomoea sepiaria K.’ 14 Morinda bracteata 15 Scaevola Taccada 15 Cordia campanulata 17 Vitis latifolia 18 Heliconia buccinata 19 Achyranthes triandra 19 Echites coryophyllata 20 Tabernaemontana crispa 20 Asclepias annularia 20 Asclepias pseudosarsa 20 Asclepias odoratissima 20 Panax conchifolium 21 Semecarpus Cassuvium 22 Crinum Toxicarum 23 Curculigo sumatrana 24 Karanga amara Vahl; H. A. 5. f. 170. f. 1. H. M. 9. t. 57. Andropogon cymbarius Linn. Mant. 303. Holcus Linn. Holcus Linn. Holcus spicatus Linn. H. A. 4. t. US. H. A. 4. t. U7. B. M. 169. H. M. 2. t. 13. Exacum diffusum Willd. Hopea dichotoma Vahl. Exacum sessile Willd. Hedyotes racemosa Willd. Hedyotes pumila Willd. Hedyotes diffusa Willd. Hedyotes herbacea Willd. H. A. 7. t. 15. H. M. 1 1. f. 53. H. M. 1. t. 52. H. M. 4. t. 59. H. A. 2. t. 75. H. M. 7. t. 7. H. A. 5. t. 62. f. 2. H. M. 10. t. 11. H. M. 9. t. lU. H. M. 1. t. U6. H. M. 9. t. 7. H. M. 10. t. 34. H. M. 7. t. 26. f. 1. H. A. 4. t. 31. H. A. 1. t. 70. H. A. 6. t. 69. H. A. 6. t. 53. Koenig. 414 ROBINSON. Species Published On Dracaena angustifolia Yucca superba Bambusa baccifera Bambusa spinosa Leersia aristata Polygonum rivulare Laurus malabathrica Laurus porrecta Caesalpinia paniculata Garuga pinnata Jussiaea exaltata Bergia ammannioides Metrosideros vera Diospyros Sapota Callophyllum Bintagor AJuga disticha Ajuga fruticosa Plectranthus aromaticus Plectranthus strobiliferus Ruellia zeylanica Ruellia cernua Volkameria infortunata Gmelina arborea Columnea balsamica Modecca tuberosa Sterculia guttata Hibiscus tortuosus Carpopogon giganteum Dolichos virosus Aeschynomene spinulosa Hedysarum Neli-tali Melaleuca Cajuputi Serratula anthelmentica Serratula cinerea Arum viviparium Arum gracile Calla oblongifolia Ficus infectoria Ficus Tsiela Ficus racemifera Gnetum scandens Urtica tenacissima Morus paniculata Sagus spincsus Sagus inermis Saguerus Rumphii Phyllanthus Leucopyrus Sapium sebiferum Bradleia multilocularis By Citation of H. A. 4. t. 25. Agave tuberosa Hort. Kew. H. M. 1. t. 60. H. A. 4. t. 3. H. M. 10. t. 12. H. M. 12. t. 76. H. M. 5. t. 53. Marsd. Sumat. p. 129. H. M. 6. t. 19. H. M. 4. t. 33. H. M. 2. t. 50. Lechea verticillata Willd. H. A. 3. t. 7. Sapota-nigro Sonnerat. H. A. 2. t. 71. Ballota disticha. Nepeta malabarica. •H. A. 5. t. 101. H. M. 10. t. 90. H. M. 9. t. Jf5. H. A. 6. t. 23. f. 1. Clerodendrum. H. M. 1. t. 41. H. M. 9. t. 85 & 12. t. 36. Modecca. H. M. 8. t. 20. H. M. 4. t. 61. H. M. 1. t. 30. H. M. 8. t. 36. H. M. 8. t. 45. Coronilla aculeata Willd. H. M. 9. t. 18. Aeschynomene L. H. A. 2. t. 17. f. 1. Vemonia Willd. Conyza L. H. M. 12. t. 9. H. M. 11. i. 21. H. A. 5. t. 108. H. M. 3. t. 64. H. M. 3. t. 63. H. A. 3. t. 93. H. M. 7. t. 22; H. A. 5. t. 7, 8. Marsd. 75. H. A. 4. t. 56. Sagus laevis H. A. 1. p. 76; Willd. 4. 404-5. Sagus genuina H. A. 1. p. 75. t. 17 & 18. H. A. 1. t. 13. Arenga Labillard. Fluggea Willd. Croton Linn. Stillingia Willd. Agyneja Willd. Page 24 25 25 25 26 29 30 30 32 33 33 34 37 40 41 44 44 45 45 46 46 46 46 47 49 50 51 54 55 56 57 59 60 60 65 65 65 66 66 66 66 67 67 68 68 68 69 69 69 ROXBURGH’S HORTUS BENGALENSIS. 415 Species Published On Page By Citation of Eradleia pinnata 69 Phyllanthus obscurus Willd. Pandanus furcatus 71 H. M. 2. t. 8. Bails spinosa 71 H. A. 5. t. 15. f. 2. Calamus latifollus 73 H. M. 12. t. 65. Lycopodium aristatum 75 Dill. Muse. t. 66. f. 7. Ophloglossum flllforme 75 H. M. 12. t. 3U. Polypodium excavatum 75 H. A. 6. t. 35. f. 1. Pterls scandens 75 H. M. 12. t. 35. Eranthemum montanum 80 Justicia montana Cor. PI. Piper arborescens 80 H. A. 5. t. 28. f. 1. The following are at first sight in the same class, but must be held unpublished, except possibly Pandanus Millore. Name. Page. Remarks. Eleusine strlcta 8 “Cynosurus Linn.” There is no Cynosurus strictus Linn. Cerbera quaternifolla 19 “H. A. 2. t. 363.” There is no such plate, and no other means of identifying the species. Crlnum brevifoilum 23 “See asiaticum. Curt. Mag. 1073.” Hedysarum lagenarla 57 “Aeschynomene L.” There is no Aeschynomene lagenaria Linn., though there is one of Loureiro. Roxburgh’s name, at the utmost, is a synonym. Leucacephala graminlfolla 68 “Eriocaulon Linn.” There is no Eriocaulon graminifolium Linn., nor was the generic name Leuca- cephala ever published. Leucacephala spathacea 68 Same as preceding. Pandanus Millore 71 “Nicobar bread fruit tree, As. Res. 3. 161. seems only a variety of odoratissima.” A short supplementary list might be made of species where Roxburgh’s spelling differs somewhat strikingly from that of preceding authors, but where he apparently had no intention of forming new names. Perhaps the best cases of this are Jasmi- num Zambac, PettospermMni Tobira, Laurus Culitlaban, Vitex trifoliata, and Piper Malmaris. This list is based on the assumption that a published binomial can never be used for any species other than that to which it was originally applied, whether or not the name -be valid in the genus to which it was attributed or in any other to which it may be transferred. There are about 185 additional names in the Hortus Bengalensis which would be considered as published therein, were it not that they had already been used. In nearly all of these cases, Roxburgh was correctly using names which 416 ROBINSON. had previously been published by other authors; occasionally, his identification is considered wrong. Thus Cype7’us procerus Roxb. is distributed by Clarke * between Cyperus malaccensis Lam., Pycreus puncticulatus Nees, and (herbarium only) Cypei'us tuberosus Rottb. But if Cyperus procerus Roxb. be considered as described in the Hortus Bengalensis, it is by the citation of Rheede’s Hortus Malabaricus 12: pi. 50, which according to Clarke would identify it with C. malaccensis Lam. However, there is an older C. p^'ocei'us.^ Returning to the ninety-one names published in the Hortus Bengalensis, investigation shows that the priority of singularly few is changed by the throwing back of the date of publication. On the other hand, the new means of typifying them raises or solves many intricate problems. These will briefly be stated. The following remain unchanged as good species : Oldenlandia diffusa, Bnicea sumatrana, Dracaena angustifolia, Garuga pin- nata, Bergia ammannioides, Sterculia guttata, Ficus Tsiela, Sapium sebiferum, and Calamus latifolius. Morinda b^'acteata is also unaffected, but there is grave doubt as to its value as a species. But Ficus I'acemifei'a Roxb. replaces F. variegata Bl. Metrosidei'os vera and Gmelina ai’borea are to be credited to Roxburgh, instead of to Lindley and Linnaeus, respectively. Half of the names have no importance except as synonyms, except that a few, with unaltered status, form the basis for generally accepted binomials. The synonyms are : Gratiola amai'a, Plade^^a virgata, Pladera pusilla, Scaevola Taccada, Cordia campanulata, Vitis latifolia, Heliconia buccinata, Achy- ranthes triandra, Asclepias pseudosarsa, Asclepias odoratissima, Panax conchifolium, Cidnum Toxica^'um, Curculigo sumaDana, Yucca superba, Leersia aristata, Polygonum, rivulare, Caesalpinia paniculata, Jussiaea exaltata, Diospy7-os Sapota, Calophyllum Bintagor, Ajuga disticha, Ajuga fTruticosa, Plect7'anthus strobili- ferus, Ruellia zeylanica, Volkamc7da info7dunata, Columnea balsamica, Modecca tuberosa, Hibiscus tortuosus, Ca7'popogon giganteum, Dolichos vh^osus, Aeschynomene spinulosa, Hedy- sarum Neli-tali, Melaleuca Cajuputi, Serratula anthelmentica, Serratula cinerea. Arum viviparium. Arum, gracile, Urtica ten- acissima, Sague7ms Rumphii, Phyllanthus Leucopyrus, B7'adleia multilocularis, B7'adleia pinnata, Ophioglossum filiforme. Poly- podium excavatum, Ptc7ds scandens, and Eranthemum montanum. The remaining thirty-two cases all require comment, some of * FI. Brit. Ind. 6 (1893) 608, 593, 616. ” Rottb. Descr. Nov. PI. (1773). ROXBURGH'S HORTUS BENGALENSIS. 417 the issues raised being of considerable importance. In several cases, new specific combinations seem to be necessary, but as it is not possible here to form definite conclusions on the actual merits of the problems, except in the case of a few species, none are herein proposed. Gratiola integrifolia was placed by Hooker under Vandellia erecta Benth., but he appears to have been in doubt as to the identity of Roxburgh’s species. Rheede’s figure and description both show 2-stamened flowers, whereas in Vandellia, the number of stamens is always 4. Andropogon bicolor, Andropogon sacckaratus, and Anthistiria cymbaria are treated by Hackel “ as different from Holcus bicolor Linn., Holcus sacckaratus Linn., and Andropogon cym- barius Linn., respectively, but they must be typified by those species. Panicum spicatum Roxb. is given by Hooker as a synonym of Pennisetum typhoideum Rich., but the oldest name cited by him is Panicum americanum Linn., still valid for transfer to Pennisetum. Ixora fulgens, I. stricta, and I. borrbata are all maintained by Hooker,^® but as the plates on which they are based are not cited by him, no opinion can be formed as to whether they rep- resent the original types. Oldenlandia alata Koenig is also retained by Hooker as a good species, but O. alata Roxb. is based on Hedyotis 7'acemosa Willd., which is Lamarck’s species of that name, and this is said by Hooker generally to be cited under 0. paniculata Linn., but probably to belong to a different section of the genus. Oldenlandia crystallina Roxb. is also maintained by Hooker,^* with Hedyotis pumila Linn. f. as a synonym, and H. pumila Willd. must be held typified by the Linnaean species. There is no apparent reason for calling the species other than Oldenlandia pumila (Linn, f.) DC. Oldenlandia herbacea Roxb. is based on Hedyotis he^'bacea Willd., which is typified by H. hei'bacea Linn., although the two are treated as doubtfully identical.^® If, then, Hedyotis hei'bacea “FI. Brit. Ind. 4 (1884) 281. “DC. Monog. Phan. 6 (1889) 519, 510, 629, 668. “FI. Brit. Ind. 7 (1896) 82. “FI. Brit. Ind. 3 (1880) 145-148. ” FI. Brit. Ind. 3 (1880) 70. ”F1. Brit. Ind. 3 (1880) 65. “ FI. Brit. Ind. 3 (1880) 65. 418 ROBINSON. Linn, is the same as Oldenlandia Heynei R. Br., the latter name must be replaced by 0. herbacea (Linn.) Roxb. Ipomoea sepiaria is maintained, but the oldest name for the species would appear to be Convolvulus marginatus Desr. ex Lam., a name still available for transfer to Ipomoea. In the case of Echites caryophyllata (to use Carey’s permis- sion to correct obvious typographical errors) , it is evident from Hooker’s discussion that this can not be taken as the basis for the species now known as Aganosma caryophyllata (Roxb.) Don ; and it also seems that the species known under the name of Tabernaemontana crispa Roxb. is not that of the Hortus Ben- galensis. Asclepias annularia Roxb. is identified as Holos- temma Rheedii Wall. Roxburgh’s name is the oldest, and is available for transfer. Semecarpus Cassuvium is cited by Engler as of Sprengel, who also published the name, basing it on Anacardium longi- folium Lam., a Philippine species, said by Lamarck also to grow in India. Rumphius’ plant, the basis of Semecarpus Cassuvium Roxb., appears to differ from any known Philippine species ; and, curiously enough, the apparent position of the matter is that Semecarpus Cassuvium Roxb., but not of Sprengel, is the species treated by Engler under that name; while S. Cassuvium Spr. is a homonym, its exact identity to be determined by that of Anacardium longifolium Lam. Bambusa baccife?u Roxb. appears to be the oldest name for Melocanna bambusoides Trin., and to require transfer.^^ Bambusa spinosa Roxb., as typified by Hort. Amb. 4: pi. 3, is probably not B. arundinacea Willd., and there are various points in Rumphius’ description, which oppose its identification as B. Blumeana Schult. f. Laurus malabathrica Roxb., is considered a doubtful species.-® L. porrecta, by the changed date of publication, becomes the oldest name for Cinnamomum Parthenoxylon (Jack) Meissn., and is available for transfer. PlecU'anthus aromaticus Roxb. is the basis for Coleus aro- maticus Benth., but Coleus amboinicus Lour, is said to be the oldest name. Roxburgh’s species is based solely on Herb. Amb. 5: pi. 101, which is also the basis of Ocimum scutellarioides Linn.=Coleus scutellarioides (Linn.) Benth. If Loureiro’s spe- cies is the same as the C. aromaticus of the Flora of British ”F1. Frit. Ind. 4 (1883) 209. ” DC. Monog. Phan. 4 (1883) 487. ”F1. Brit. Ind. 3 (1882) 664. ” FI. Brit. Ind. 7 (1896) 417. “FI. Brit. Ind. 3 (1882) 648. "FI. Brit. Ind. 5 (1886) 136. "FI. Brit. Ind. 4 (1883) 21. ROXBURGH’S HORTUS BENGALENSIS. 419 India, there is double reason for taking it up. It is based primarily on “Marrubium Amboinicum album Rumph. Amb. 1. 8. c. 75. tab. 72.” It has especial importance as the type of the genus Coleus Lour. Ruellia cernua Roxb. is maintained for a species of very local distribution in India, which can hardly possibly be typified by a description founded on a plant of Amboina. Calla oblongifolia is the basis for Aglaonema oblongifolium Schott, but following the synonymy cited, would not have been the oldest name, but for the corrected date of publication. Ficus infectoria Roxb. is not the species of that name in recent works,^® but being based on Hort. Malab. 3: pi. 64-, is Ficus Tjakela Burm. Gnetum scandens Roxb. is maintained in the Flora of British India, but its oldest name is G. edule (Willd.) Bl., based on the same plate of Rheede. Morus paniculata Roxb. appears to be the oldest name for Pipturus velutinus (Decne.) Wedd. The treatment of the names Sagus spinosus and S. inermis, as compared with that of the plates on which they are based, is different in the Flora of British India and in the Hortus Bengalensis. Pandanus furcatus Roxb. is maintained,^® but Rheede’s plate on which it was based is cited under P. fascicularis Lam. Batis spinosa Roxb. is treated as the original publication of the species known as Plecospermum spinosum Trecul, but with the exclusion of the synonym of Rumphius, which is its sole basis. Lycopodium aristatum is doubtfully placed by Baker®® under Selaginella exigua Spring. Piper arborescens Roxb. ex Wall. Cat. is identified as Thottea dependens Kl., but there can be no doubt that the true Piper arborescens, typified by Herb. Amb. 5: pi. 28. f. 1, is a genuine Piper. The plate number is wrong, and a better citation would be Herb. Amb. b\ 2M. vl. 102, f. 2. ==F1. Brit. Ind. 4 (1884) 413. ^“Fl. Brit. Ind. 5 (1888) 514, 515. ” 6 (1893) 481. FI. Brit. Ind. 6 (1893) 484. ”F1. Brit. Ind. 5 (1888) 491. •"Fern Allies (1887W67. “FI. Brit. Ind. 5 (1886) 74. Vol. VII, No. 5, including pages 259 to 362, was issued November 15, 1912. ERRATA Page 43, line 8, for born read borne. Page 63, line 2 from the bottom, for stiplacea read stipulaced. Page 102, line 11 from the bottom, for H. palustris read M. palustris. Page 196, line 24, for cannbina read cannabina. Page 285, line 9, for MAGABOTRYS read MEGABOTRYS. Page 290, line 19, for SEMICARPUS read SEMECARPUS. Page 307, line 16, for Guttiforae read Guttiferae. Page 342, line 3, for mindorensis read mindorense. Page 343, lines 7 and 16, for farimosa read farinosa. Page 359, line 18, for comprenhensive read comprehensive. 421 INDEX (New genera and species and combinations published for the first time are in black-faced type ; synonyms and species mentioned incidentally in the text are indicated by the page references being in italics.) A Page. Acanthaceae lOS, 248 Achyranthes triandra Roxb 413, il6 Acoridium venustulum Ames 26 Acrocephalus capitatus Benth 101 indicus (Burm.) O. K 101 Adelia glandulosa Blanco SSS javanica Miq SOI papillaris Blanco 2S8 Adenogynum discolor Reichb. f. & Zoll. .. 396 Adiantum cupreum Copel 55 flabellulatum L 55 opacum Copel 55 scabripes Copel 55 Aeschynomene sesban L 235 spinulosa Roxb 414, il6 Aganosma acuminata G. Don SSI caryophyllata (Roxb.) Don .... US velutlna A. DC SSi Aglaeonema oblongifolium Schott ilO Aglaia Lour 274 brachybotrys Merr 274 cagayanensis Merr 275 curranii Merr. 276 diffusa Merr 277 harmsiana Perk. 278 hexandra Turcz. 275 llanosiana C. DC. _ 277 turczaninowii A. DC 275 Aglaomorpha meyeniana Schott 42 Agrostis matrella L 280 Ajuga disticha Roxb 414, 116 fruticosa Roxb 414, il6 Alangiaceae 319 Alangium Lam 319 brachyanthum Merr ’319 lamarckii Thwaites S20 longiflorum Merr 319 meyeri Merr 321 octopetalum Llanos 320 salviifolium (L. f.) Wanger- in 319, 320 salviifolium hexapetalum (Lam.) Wangerin 320, 321 Alchornea Sw 382 javensis Muell.-Arg SSS mollis Muell.-Arg 390 parviflora (Benth.) Muell.- Arg 382, 320 rugosa (Lour.) Muell.-Arg 383 sicca (Blanco) Merr 382, 383, 390 Page. Alcinaeanthus Merr 379 philippinensis Merr 380 Alectridia quartiniana A. Rich 229 Alpinia L 75, 78 vanoverberghii Merr 75 Alsomitra Roem 352 clavigera F.-Vill 353 integrifoliola (Cogn.) Hayata 353 sarcophylla (Wall.) Roem 352 Alyxia Banks 330 blancoi Merr 330 luzoniensis Merr 331 monilifera Vid 330 parvifolia Merr 331 stellata F.-Vill 330 Anacardiaceae 281 Anacardium longiColium Lam 118 Andropogon L 228 bicolor Roxb 413, 417 cymbarius L 113, 117 saccharatus Roxb. 413, 417 squari'osus L. f. 228 zizanioides (L.) Urb 228 Aneilema R. Br. 232 malabaricum (L.) Merr 232 nudiflorum R. Br. 232 versicolor Dalz 232 Anodendron A. DC. . 331 axillare Merr. 331 candolleanum Wight 382 loheri Merr 332 manubriatum (Wall.) Merr. 333 paniculatum A. DC 332, 333 pauciflorum Hook, f 331 Anonaceae 109, 2.34, 266 Anona cherimolia Mill 112, 113 glabra L 113, 119 laurifolia Dun. 119 muricata L 113, 118 palustris L 119 reticulata L 113, 117 squamosa L. 113, 115 uncinata Lam 231 Anthistiria cymbaria Roxb 413, 417 Antholyza meriana Blanco 233 Antidesma Burm 383 bunius Spreng 383 crassifolium (Elmer) Merr. .. 383 cumingianum Muell.-Arg 381 mindanaense Merr 383 423 424 INDEX. Page. Apocynaceae 242, 330 Aquilaria Lam 314 Araliaceae 95, 241, 318 Aralia cochleata Lam 241 guilfoylei Cogn. & Marche 242 Ardisia Sw. 323 cagayanensis Merr 323 jagori Mez 325 ramosii Merr 324 scabrida Mez 324 Artabotrys R. Br 234 odoratissimus R. Br 234 uncinatus (Lam.) Merr 234 Arthraxon Beauv 229 ciliaris Beauv. ... 229 ciliaris langsdorfhi Hack 229 ciliaris quartinianus Hack 229 hispidus (Thunb.) Merr 229 quartinianus Merr 229 Arum gracile Roxb 414, 416 viviparium Roxb 414, 416 Arundinaria glaucescens Beauv 239 Asclepias annularia Roxb 413, 418 odoratissima Roxb 413, 416 pseudosarsa Roxb 413, 416 ambiguum Schkuhr 62, 63 confertum Bak. ... 62 lanceum Thunb. . 55 Aspidium stipellatum Bl. . 60 Athyrium carnosum Copel. 61 christii (C. Chr. ) Copel 62 confertum (Baker) Copel 62 cumingii (Presl) Milde 63 elatum (Fee) Copel 63 hewitti Copel 62 lanceum Milde 65 maximum Milde 61 muricatum Copel 61 pinnatum (Blanco) Copel 63 polycarpum Copel 61 sarawakense Copel 62 sorsogonense poense Copel 61 subserratum (Bl.) Milde 63 Ballota disticha L 414 Bambusa Schreb 230 arundinacea Willd 413 baccifera Roxb 414, 418 blumeana Schult. f 413 glaucescens (Willd.) Sieb 230 nana Roxb 230 spinosa Roxb 414, 418 Bartramia indica L 239 Batis spinosa Roxb 415, 419 Begoniaceae 308 Begonia L. 308 affiiiis Men 308 agusanensis Merr 399 ciliifera Merr 399 contracta Warb 310, 312 crispipila Elmer 311 cumingu A. Gray 311 elatostematoides Merr 309 Page. Begonia leytensis Elmer 311 macgregorii Merr 310 oblongata Merr 310 quercifolia A. DC 311 Bergia ammannioides Roxb 414, 416 Bignoniaceae 352 Blumea DC 60, 106, 366 bicolor Merr 366 oxyodonta DC 251 tenera Merr 250 vanoverberghii Merr 105 Blumeodendron Kurz. .. 384 subrotundifolium (Elmer) Merr. 384 Blyxa Thou 73 octandra (Roxb.) Planch 73 roxburghii Rich. 73 Brabejum lucidum Blanco 330 Bradleia multilocularis Roxb 414, 416 pinnata Roxb .,... 415, 416 Brainea J. Sm . 47 Bridelia Willd 384 ovata Decne. .. 335 platyphylla Merr 384 subnuda K. Schum. & Lauterb. .. 385 Brucea sumatrana Roxb 413, 416 Bulbophyllum Thou 23, 126 adenopetalum Lindl. .. 24, 25, 126, 128, 131, 134 alagense Ames 126 antenniferum Reichb. f. .. 126, 138 auratum Reichb. f 126, 139 balaeniceps Reichb. f 126 bataanense Ames 126 blumei J. J. Sm 127 bolsteri Ames 126, 139 bontocense Ames 126, 136 braccatura Reichb. f. .. 24, 25, 126, 134 canlaonense Ames 126, 136 careyanum Spreng 126 carinatum Ames 126, 140 eheiri Lindl 126, 127 chryseum (Kranzl. ) Ames 127, 140 clandestlnum Lindl 138 clemenslae Ames 127, 129 copelandii Ames 127 corticola Schltr. 137 cumingii Reichb. f. .. 127, 139, 143 cuneatum Rolfe 127 curranii Ames 127, 130 cuspidilingue Reichb. f 127 dasypetalum Rolfe .... 24, 25, 127, 128, 134 dearei Reichb. f 127 elmeri Ames 127, 136 exile Ames 127 gimagaanense Ames 127, 130 gnomoniferum Ames 127 halconense Ames 127, 136 lancipetalum Ames 23, 127 lasianthum Lindl 127, 128 lasioglossum Rolfe 127 INDEX. 425 Page. Bulbophyllum lepantense Ames 127, 131 leysianum Burbidge UO longiflorum Thouars 127, U2, US luzonense Ames 127, 141, 142 macgregorii Ames 127, 137 makoyanum Reichb. f 127 marivelense Ames 127, 132 maxillare Reichb. f 127 merrittii Ames mindorense Ames 127 . 127, 137 myrianthum Schltr 138 nutans Reichb. f ... 24, 128 orthoglossum Wendl. & TCranzl. 128 pauciflorum Ames . 128, 132 penicilliura Par. & Reichb. f 128 pleurothallioides Ames 128 profusum Ames 128, 137 ramosii Ames 2i, 25, 128, 131 saurocephalum Reichb. f. .. 128 sessile J. J. Sm 138 stellatum Ames 128, 141 trisetum Ames 128, 142 vagans Ames & Rolfe 128 vanoverberghii Ames .... 128, 133 vermiculare Hook, f 131 weberi Ames 127, 128, U2 whitfordii Rolfe 128 zambalense Ames 128, 134 C Caesalpinia paniculata Roxb 414, il6 Calamus latifolius Roxb 416, 416 Calanthe R. Br 11 lacerata Ames 11 ramosii Ames 12 Calla oblongifolia Roxb 414, 419 Callicarpa L 338 angusta Schauer 3il cauliflora Meir 338 dolichophylla Men- 339 elegans Hayek 338 longifolia subglabra Schauer . SiO pentandra Schauer Si2 ramiflora Merr SS9 rivularis Mei r 340 Calophyllum L 92, 307 bintagor Roxb 414, il6 blancoanum PI. & Tr 92 glabrum Merr 92 kunstleri King 307 Campanumoea B1 104 axillaris Oliv 101, truncata Endl 104 Canna glauca Linn 411, 412 Cansjera Juss 265 malabarica Lam 265 manillana Bl. . 233 rheedii Gmel. . 265 scandens Roxb — 265 Page. Capparidaceae 235, 270 Capparis L 235 baducca Blanco 235 cordifolia Lam 235 littoralis Merr _... 270 lobbiana Turcz . 270, 271 loheri Merr. 270 mariana Jacq 235 spinosa mariana K. Sch. 235 Caranga amara Vahl il3 Carpopogon giganteum Roxb . 414, . iU Carumbium populneum Muell.-Arg 390 populneum minus Muell.- Arg 390 Cephalocroton albicans genuinus Muell.- Arg 386 discolor Muell.-Arg. 386 orientalis Scheff 386 Cerbera quaternifolia Roxb. 416 Cestichis Pfitz _ 7 compressa Ames .. 8 cumingii (Ridl. ) Ames .. 7 Champereia Griff £33 cumingiana Merr 233 griffithiana Planch. 233 griffithii Kurz 233 manillana (Blume) Merr. 233 Characeae 69 Chara L _... 69 benthami A. Br 70 brachypus A. Br 70 braunii Gmel. 70 congesta Spreng 70 corallina Willd 69 corallina basilaris A. Br 69 flaccida A. Br 70 gymnopitys A. Br 70 gymnopus A. Br 70 hydropitys A. Br 70 zeylanica Willd 70 Cheilosa javanica Bl 881 Chloradenia discolor B.aill 386 Chloranthaceae 259 Chloranthus L - — 259 henryi Hemsl 260 oldhami Solms-Laub. 260 philippinensis Merr, 269 Cinnamomum parthenoxylon (Jack) Meissn 418 Cirrhopetalum antenniferum Lindl. ... ,. 126 , 138 auratum Reichb. f. . . 126, 139 chryseum Kranzl. .. 127 cumingii Lindl 127 koordersii Rolfe ... 14% maxillare Lindl 127 nutans Lindl — . 123 thouarsu Lindl ,. 127, U3 Cissus L 209 Cladium P. Br 74 cyperoides Merr 74 Cladodes rugosa Lour _ S8S Cladogynos Zippel. .. 388 orientalis Zipp 388 426 INDEX. Page. Cleistanthus Hook, f 38(5 angustifolius Merr 386 Cleistoloranthus verticillatus Merr 7S Clerodendron L 98, 245, 341 Clerodendron brachyanthum Schauer 98 capsulare Blanco 245 commersonii ( Poir. ) Spreng 245 elliptifolium Merr 341 inerme Auci 245 infortunatum L 98 klemmei Elmer S42 macrostegium Schauer 98 mindorense Merr 342 multibracteatum Merr 98 neriifolium Wall 245 quadriloculare Merr Si2 simile Merr Si2 vanoverberghii Merr 98 villosum B1 98, 342 Cocculus DC 81 blumei Boerl 266 celebicus Boerl 266 laurifolius DC 81 Codonopsis truncata Wall 204 Coleus Lour . 344 amboinicus Lour 344, 418 aromatlcus Benth 544. 418 scutellarloides (L.) Benth 418 suganda Blanco 544 Columbia Pers 297 macgregorii Meir 297 megacarpa Merr 298 serratifolia (Cav. ) DC 298 subaequalis Planch 297 Columnea balsamlca Roxb 414, 425 Commelinaceae 232 Commelina nudicaulis Burm 252 nudiflora L 252 Compositae 105, 250, 356 Conceveiba tomentosa Span 555 javanense Bl. 555 Coniogramme Fee 50 Convolvulaceae 244 Convolvulus caespitosus Roxb 244 hirtus L 244, 245 marginatus Desr 425 reptans L 244. 245 Cordia campanulata Roxb 413, 425 Coronllla aculeata Willd . 424 sesban Willd 255 Craspedodictyon Copel . 50 Crinum brevifolium Roxb 416 toxicarum Roxb 413, 425 Croton appendiculatus Elmer 400 grandifolius Blanco 255, 554 longipedunculatus Elmer 410 moluccanus L . 401 paniculatus Lam. . 400 Cucurbitaceae 104, 352 Curculigo sumatrana Roxb 413, 425 Cyclodon distans Griff. ... 105 Cyclostemon Bl . 387 Page. Cyclostemon bordenil Merr 557 cumingii Benth. 555 megacarpus Merr 387, S8S ramiflorus Merr. 387 subcrenatus Merr 388 Cyperaceae 74, 231 Cyperus dilutus Vahl. _ 252 malaccensis Lam 416 procerus Rojb 425 tuberosus Rottb. 416 D Davallia chaerophylla Ces 57 viscidula Mett 55 Davallodes Copel 45 viscidulum (Mett.) v. A. v. R. 68 Dendrobium Sw 14 albayense Ames 14 aloifolium Reichb. f 25 angustifolium (Bl.) Lindl. .. 14 basilanense Ames 14 brongniartii Kranzl 15 clemensiae Ames 16 distichum Reichb. f 27 epidendropsis Kranzl 19 filiforme J. J. Sm 27 indivisum Miq. ... 27 jenkensii Griff. .. 25 kentrophyllum Hook, f 25 lunatum Lindl. ... 17 macgregorii Ames 17 merrillii Ames ... 25 o’brienianum Kranzl 19 parciflorum Reichb. f 18 ramosii Ames 18 ventricosum Kranzl. 19 Dendrochilum Blume 26 convallariaeforme Schauer 25 copelandii Ames 25 exile 2lmes 27 longibulbum Ames 26 macgregorii Ames 27 maleolens Kranzl 25 marivelense Ames 25 microcheilum (Schltr.) Ames 26 oliganthum Pfitz 25 pumilum Reichb. f 25 vanoverberghii Ames 27 venustulum Pfitz 25 Dennstaedtla Bernh _.. 4*, 49 cuneata obtusa Copel 63 Derris Lour 82, 273 cebuensis Merr 273 elliptica Benth 55 micans Perk. .. 275 mindorensis Perk 275 oblongifolia Merr 82 Desmodlum Desv 273 trifoliastrum Miq 273 Desmotrichum angustifolium Blume 24 Dichapetalaceae 279 INDEX. 427 Page. Dichapetalum Thou 279 ciliatum Merr 279 luzoniense Merr. & Rolfe.. 281 rohinsonii Merr 280 Dilleniaceae 91, 306 Dillenia L 305 bolster! Merr 305 philippinensis Rolfe soe Dionaea Ellis 40 Diospyros sapota Roxb 116, 414 Diplazium acuminatum B1 62 christii C. Chj 62 elatum Fee .... _ 6S malaccense Presl 6S petiolare Presl 63 schkuhrii J. Sm 62, 68 Discocalyx Mez 325 insignis Men 326 longifolia Mcir 325 macrophylla Merr 326 raaculata Men. 326 montana Elmer 326 Dolichos conspersus Grab 233 punctatus W. & A 235 virosus Roxb. ... 414, il6 Doryopteris J. Sm 50 Dracaena angustifolia Roxb 414, 416 Drymoglossum Presl iT, i9 Drynaria Adans 42 Dryopteris adenophora C. Chr 54 angustipes Copel 60 aquatilis Copel 59 aquatiloides Copel 69 balabacensis Christ 54 dichrotricha Copel 64 dissecta O. Ktz 54 jaculosa C. Chr. 60 mesodon Copel 64 parasitica O. Ktz 60 porphyricola Copel 60 purpurascens Copel 54 salicifolia C. Chr 59 sparsa O. Ktz 54 stipellata O. Ktz 60 sumatrana v. A. v. R 60 Dryostachyum J. Sm _ 42 DunbariaW. & A 236 conspersa Benth 235 punctata (W. & A.) Benth 236 Dysoxylum B1 ... 278 laxum Merr. ... 278 pauciflorum Merr 278 E Echites afBnis R. & S 334 caryophyllata Roxb 413, 418 caryophyllata Roth 334 caudata Blanco . 333 coriacea Wall. ... 333 manubriata Wall 333 paniculata Roxb 333 Eclipta L 251 alba Hassk 251 114346 6 Page. Eclipta zippeliana B1 251 Elaeocarpaceae 295 Elaeocarpus L 295 grandiflorus Sm. 295 lanceolatus BI 295 pustulatus Merr 295 verruculosus Aug. DC 296 Elateriospermum paucinervium Elmer .... 382 Eleusine stricta Roxb 416 Eleutherine Herb 233 americana (L.) Merr 233 plicata Herb 233 Embelia Burm. f . 326 nigro-punctata Merr 326 viridiflora (A. DC.) Scheff 327 Emerus sesban O. Ktz 235 Enhydra Lour 251 fluctuans Loui 251 Eragrostis Host .. 229 mangalorica Hochst 229 tenella viscosa Stapf 229 viscosa Trin _ 229 Eranthemum bicolor Schrank 248 montanum Roxb 416, 416 pulchellum Hort 248 Eria Lindl 20 longibracteata Leavitt 20 vanoverberghii Ames 20 Ericaceae 96, 321 Eriocaulaceae 232 Eriocaulon L 232 alatum H. Lecomte 232 Eschatogramme Trev 49 Eugenia L 314 bataanensis Merr 316, 317 camiguinensis Merr 314 ciliato-setosa Merr 315 paucivenia C. B. Rob 316 propinqua Merr 315 tenuipes Merr 316 Eulophia R. Br .. 13 vanoverberghii Ames 13 Euphorbiaceae 237, 379 Euphorbia prostrata Ait 237 thymifolia Burm 237 Eupatorium L 355 benguetense C. B. Rob 356 camiguinense Merr 365 japonicum Thunb 355 , sambucifolium Elmer 356 toppingianum Elmer 356 Evodia Forst 84, 375 gracilis Kurz 377 lamarckiana Benth 375, 376, 377 lunur-ankenda (Gaertn.) Merr... 378 marambong Miq 378 pteleaefolia (Champ.) Merr. .. 377, 378 roxburghiana Benth 373, 376, 377, 378 semecarpifolia Merr 377, 378 triphylla DC 373, 374. 375, 377 triphylla Guillaumin 378 villosa Merr. 84 Exacum L 95 428 INDEX. Page. Exacum diffusum Vt illd ilS parviflorum Merr 96 sessile Willd. US tetragonum Roxb 96 Excoecaria Linn. 389 obtusa Merr 389 sicca Blanco 390 philippinensis Merr S89 philippinensis euphlebia Merr 889 F Fagara lunur-ankenda Gaertn 378 octandra Blanco 375, 376 triphylla Lam 373. 37 i, 375 triphylla Roxb 378 Ficus infectoria Roxb 414, U9 racemifera Roxb 414, 416 tjakela Burm 419 tsiela Roxb 414, 416 variegata B1 U6 Fimbristylis Vahl 231 corniculata Merr 231 diphylla Vahl 75 tenera R. & S 231 Fluggea virosa Baill i05 Freeria Merr . 292 j repanda Merr 292 Gaertneria laurifolia Wall 236 Garcinia L 307 ramosii Merr 307 G Garuga pinnata Roxb 414, 416 Gastrodia R. Br 6 javanica (Bl.) Lindl 6 Geniostoma Forst 329 cumingianum Benth 330 stenophyllum Merr. 329 Gentianaceae 96 Geunsia Bl . 342 cumingiana (Schauer) Rolfe .... 343 farinosa Bl 343 hookeri Merr 342 Givotia rottleriformis Griff 401 Gmelina arborea Roxb 414, 418 Gnetum edule (Willd. ) Bl 419 scandens Roxb. 414, 419 Goniophlebium Presl 49 Goodenoviaceae ... 353 Gramineae 74, 228 Gratiola amara Roxb 411, 412, 413, 416 integrifolia Roxb 413, 417 pusilla Willd 246 Grewia L 89, 298 acuminata Juss 300, 301 edulis Men . 298 laevigata Vahl 90, 302 multiflora Juss 90, 302 ovata Merr 299 palawanensis Merr 300 parva Merr 301 Page. Grewia rizalensis Men 303 rolfei Mei-r. 302, 303 setacea Merr. 89 stylocarpa Warb 299 tiliaefolia Rolfe 302 umbellata Roxb 300 Guttiferae 92, 307 Gynostemma elongatum Merr 353 integrifoliolum Cogn 353 Gynura Cass _ 366 piperi Merr 365 sarmentosa DC 355 Gyrinopsis Decne 313 brachyantha Merr 313 cumingiana Decne 814 H Habenaria Willd 2 curranii Ames 2 delessertiana Kranzl 2 diphylla Dalz 6 falcigera Reichb. f 6 hystrix Ames 2, 3 mearnsii Ames 3 muricata Vid 2, 3 ponerostachys Reichb. f 4 robinsonii Ames 6 rosulata Ames 6 Haemaria merrillii Ames 6 Halophila Thou 228 beccarii Aschers 228 Harrisonia R. Br 236 bennetii Hook, f 236 perforata (Blanco) Merr 236 Hedyotis diffusa Willd 413 herbacea Willd 413, 417 pumila Willd 413, 417 racemosa Willd 413, 417 Hedysarum lagenaria Roxb 416 nelitali Roxb 414, 416 Heliconia buccinata Roxb 413, 416 Hemionitis L. 47, 50 Hibiscus tortuosus Roxb 414, 416 Hippocrateaceae 291 Hiraea reclinata Blanco 236 Histiopteris integrifolia Copel 63 stipulacea (Hook.) Copel 63 Holcus bicolor L. 417 saccharatus L 417 spicatus L 413 Holostemma rheedii Wall 413 Homalanthus H. Juss 390 populneus minor (Muell.- Arg.) Merr 390 populneus siccus Pax 390 Homonoia Lour 391 javensis (Bl.) Muell.-Arg 391 javensis ciliata Merr 391 Hopea dichotoma Vahl 413 Hornemannia viscosa Willd 246 Humata Cav 49 angustata J. Sm SB brooksii Copel 64 INDEX. 429 Page. Humata dimorpha Copel 68 67 microsora Copel 55 puberula Copel 64 tenuis Copel 67 67 Hydrocharitaceae 73, 228 49 Hypophorium pergracile Nees : 75 I I /* 5> 1 rm /• o 9.09. 333 acuminatus F.-Vill. ... 334 frutescens F.-Vill 333, 334 macrocarpus F.-Vill. . 334 333, 334 ovatifolius A. DC 333 velutinus F.-Vill 334 81 montanum Merr. 81 philippinense Merr 82 244 aquatica Forsk 244 linifolia Bl 244 philippinensis Choisy 244 reptans Poir 244, 245 sepiaria Koenig 413, 418 233 Ixia americana Aubl 233 /tin ,117 1 Q ,417 i17 J Jussiaea L . 240 Jussiaea acuminata Sw 240 blumeana Presl 240 costata Presl 240 linifolia Vahl 240 Justicia montana Roxb 415 K Kuhlhasseltia J. J. Sm 6 merrillli Schltr 6 h Labiatae - 100, 344 Lasiolepis bennetii Planch 2SG multijugra Benn. 2S6 paucijuga Benn- 226 Laurus malabathrica Roxb 414, 418 414, 418 Lechea verticillata Willd- 414 Leersia aristata Roxb . 414, 416 95K 9.7a T +->v 1 • in5> 947 Lepidagathis Willd 103 dispar Clarke 103 psilantha Nees 103 subglabra Merr 103 Page. Lepta triphylla Lour S7S Leptolepia Mett iS Leucacephala graminifolia Roxb 415 spathacea Roxb 415 Leucostegia Presl 49 Ligustrum L 329 cumingianum Decne S29 pubescens DC S29 pubinerve 111 329 Limacia Lour 266 blumei (Boerl.) Diels 266 Limnophila R. Br 246 diffusa G. Don 2J,7 manilensis Merr 246 Lindsaya azurea Christ 68 nitida Copel 68 papuana C opel 68 Liparis cumingii Ridl 7 Litobrochia incisa integrifolia Bedd. : 6S Lobelia L 105 bialata Merr 105 filiformis Lam 105 frutescens Mill S5S radicans Thunb 105 Loganiaceae 329 Lomagramma J. Sm. 67 brooksu Copel 60 perakensis Bedd 61 Lomariopsis Fee 67 Loranthaceae 78, 261 Loranthus L 261 appendiculatus Merr 78 capituliferus Merr 264 confusus Merr. .. 261, 263 estipitatus Stapf 262 mirabilis Muell.-Arg. & Van Huerck 26i pentagonus Merr 80 pubiflorus Merr 263 similis Merr 262 sphenoideus Bl. 262, 268 subsessilis Merr 263 tetranthus Merr. 79 verticillatus R. & P 78 williamsii Merr. 265 Ludolphia glaucescens Willd 230 Ludwigia prostrata Roxb. 2il Lycopodium aristatum Roxb 415, 419 Lysimachia L 93 capillipes Hemsl 93 decurrens Forst. f 93 M Macaranga Thouars 238, 391 amplifolia Merr 392, 393 glgantifolia Merr 391, 393 grandifolia (Blanco) Merr... 394 leytensis Merr 395 mappa F.-Vill 288, 394 montana Merr 394 noblei Elmer 392, 393 porteana Andre 238, 394 Macgregorianthus Merr 312 430 INDEX. Page. Macgregorianthus paniculatus Merr 312 Maesa Forsk 327 ferruginea Merr 327 manillensis Mez S28 pachyphylla Merr 328 ramentacea Wall S28 Magnoliaceae 81 Mallotus Loui 238, 395 alternifohus Merr 395 aunculatus Merr 396 barbatus (Wall.) Muell.-Arg. .. 397 camiguinensis Merr 397, iOO cardiophyllus Merr 398 cauliflorus Merr. 399 cochinchinensis Lour iOO eglandulosus Elmer 391 helferi Muell.-Arg 396 leucocalyx Muell.-Arg 396, 398 moluccanus (L.) Muell.-Arg. .. 401 moluccanus pendulus Merr 401 paniculatus (Lam.) Muell.-Arg. 400 papillaris (Blanco) Merr 238 playfairii Hemsl 239, iOO ramosii Men 401 sanchezii Merr 402 zollingeri F.-V ill 238 Malpighiaceae 236 Malvaceae 90, 240 Malva coromandelina L 2i0 Malvastrum A. Gray 240 coromandelinum ( L. ) Garcke.. 240 tricuspidatum A. Gray 240 Marattia brooksi Copel 59 Mariscus Gaertn 231 dilutus (Vahl) Nees 231 microcephalus Presl 231 Marrubium album amboinicum Rumph. .. 3H Medicago L 274 denticulata Willd 274 lupulina L 274 Medinilla Gaudich. 94 confusa Men’ 9i intermedia BI 9i myriantha Merr. . 9i subsessilis Merr. 94 vanoverberghii Merr 94 Melaleuca cajuputi Roxb. .. 414, 416 Melastomataceae 94 Meliaceae . 237, 274 Melia koetjape Burm. f 237 Melicope Forst 375 luzonensis Engl 374, 375, 376 odorata Elmer 375, 376 ternata Vid 375, 376 tripylla (Lam.) Merr 376 Meliosma B! 294 macrophylla Merr 294 Melocanna bambusoides Trin. : 418 Melothria L 104 lobata Merr. 104 Meniscium Schreb. . 4T Menispermaceae 81, 265 Mentha auricularia Blanco 346 cablin Blanco 345 Page. Merremia Dennst 244 caespitosa Hallier f 244 hirta (L.) Merr 244 Merrittia Merr. 106 Mesona Bl . 101, 346 chinensis Benth. 102 clausa Mei r 345 palustris Bl 102 philippmensis Merr 102 procumbens Hemsl 102 prunelloides Hemsl 102 wallichiana Benth 102 Metrosideros vera Roxb 414, 416 Micrechites Miq _ 334 schreickii Rolfe 97, 334 Microcarpaea R. Br 100 alterniflora Bl. .. 100 minima (Koenig) Merr 100 muscosa R. Br. . 100 Microchloa R. B 74 indica (L. f.) Merr 74 setacea R. Br 74 Microlepia Presl 48 Mimosa pudica L 37 Mitrephora Hook. f. & Thoms. - 266 weberi Merr. 266 Modecca tuberosa Roxb 414, 416 Moraea plicata Sw 233 Morinda bracteata Roxb 413, 416 Morus paniculata Roxb 414, 419 Myrsinaceae . 323 Myrtaceae 314 N Nardus indica L. f 74 Nepeta malabarica L. . _ . 414 Nephrodium molle major Bedd 60 Nitella Agardh 69 acuminata A. Br. 69 mauritiana A. Br 69 polyglochin A. Br 69 roxburghii A. Br 69 Notholaena R. Br . 47 Nothopanax Miq 241 cochleatum (Lam.) Miq 241 crispatum (Bull.) Merr 241 fruticosum (L.) Miq 241 guilfoylei (Cogn. & Marche) Merr 241 ornatum (Bull.) Merr 241 O Oberonia Lindl 8 benguetensis Ames 8 griffithiana Lmdl 10 hispidula Ames 8 merrillii Ames 9 setigera Ames 10 Ocimum capitellatum L. f 101 scutellarioides L 418 Odontosoria Fee .. 48 Oenotheraceae . 240 Olacaceae 265 INDEX, 431 Page. Page. fRl ^ T.inHl crystallina Roxb 413, 417 infundibuliformis Ames 20 mearnsii Ames .. 21 herbacea Roxb 412, 413, 417, 418 heynei R. Br US paniculata L U7 pumila (L. f. ) DC U7 myosurus Ames 21 ramosii Ames 22 sulcata J. J. Sm 22 vanoverberghii Ames 22 Oleandra Cav 49 oblanceolata Copel 64 Ophioglossum filiforme Roxb 416, US spicatum Roxb 411, 412 Phyllanthus L 404 curranii C. B. Rob 406 Opilia cumingiana Baill 233 manillana Baill ZSS Orchidaceae 125 Osterdamia matrella O. Ktz 2S0 Ostodes Blume .. 403 leucopyrus Roxb 414, 416 luzoniensis Merr. 404 obscurus Wind. U5 robinsonii Men 406 Phymatodes J. Sm 47 W^nll angustifolia Men- 403 macrophylla (Muell.-Arg. ) Benth. 404 serrato-crenata Merr 404 loheri Merr 293 Piper arborescens Roxb 416, 419 Pipturus velutinus (Decne. ) Wedd 419 Pladera pusilla Roxb 413, US P Paederota minima Koenig 100 Paliurus dubius Blanco 2S6 Platyclinis microcheila Schltr 26 Platynema laurifolium W. & A 236 Platytaenia requinaria Kuhn 50 Plecospermum spinosum Tree 419 Paltonium Presl 49 Panax cochleatum DC 2U conchifolium Roxb 413, US crispatum Bull ZU fruticosum L 2U ornatum Bull 241 Pandanus fascicularis Lam 419 furcatus Roxb 416, 419 millore Roxb 415 Panicum americanum L U7 arborescens L 230 strobiliferus Roxb 414, US Pleurogramme loheriana Chi ist 53 minor (Fee) Copel 63 pusilla Christ 53 Pogostemon Desf .345 cablin (Blanco) Benth. .. 346, 347, 349 heyneanus Merr 346 membranaceus Merr 347 nepetoides Stapf 101, 347 nepetoides glandulosus Merr 347 sparsicomum Nees 230 spicatum Roxb 413, 417 Parishia Hook. 1 281 insignia Hook, f 282 malabog Merr 281 Parsonsia R. Br. 335 confusa Merr 335 cumingiana A. DC 335 oblancifolia Merr 336 spiralis Wall 335 Pennisetum typhoideum Rich. U7 Pergularia Auct. 243 patchouli Hook, f 346 patchouli suavis Hook, f 346 patchouly Pellet 346 patchowli Dalz. & Gibs. 346 reticulatus Merr 348 suavis Ten 346 Polyalthia B1 268 lanceolata Vid 269 loheri Men 268 loheri cagayanensis Merr 269 tenuipes Merr 269 Polygalaceae 237 accedens Vid 243 angustiloba Warb. 243 filipes Schltr 243 glabra Blanco 243 procumbens Blanco 243 Phaius Lour 10 flavus Lindl 11 linearifolius Ames 10 Phalaris hispida I hunb 229 zizanioides L 228 Phreatia Lindl. 20 Polygala ciliata L. .. 237 Polygonum rivulare Roxb 414, 416 Polyosma B1 271 philippinensis Merr 273 piperi Merr 271 verticillata Merr 272 Polypodiaceae 41, 47 Polypodium L 48 excavatum Roxb 415, 416 lineare Thunb. . 65 linguaeforme Mett 68 aristulifera Ames 22 bracteata Ames 22 loxogrammoides Copel 66 macrophyllum Reinw 65 432 INDEX. Page. Polypodium merrittii poense Copel 66 musaefolium B1 68 nectariferum Bak i2 taeniophyllum Copel 66 wrayi Bak 65 Polyscias fruticosa Harms 2il Portulacaceae _ 234 Portulaca L 234 pilosa L. .. 234 Potamogetonaceae 228 Primulaceae 93 Prunella indica Burra 101 Pseuderanthemum Radik 248 bicolor Radik 2iS pulchellum (Hort.) Merr 248 Pteris L SO cretica L. ... - Si opaca J. Sm Si scandens Roxb 416, il6 taenitis Copel 63 Pterospermum Bl. 304 elmeri Merr 304 longipes Merr 304 niveum Vid SOS obllquum Blanco SOi subpeltatum Merr SOi Pycnarrhena Miers 266 celebica (Boerl. ) Diels 266 Pycreus puncticulatus Nees il6 R Radermachera Hassk 362 elliptica Merr SS2 ■whitfordii Merr 362 Rafflesiaceae 209 Rafflesia R. Br 209 manillana Teschem 209 philippinensis Blanco 210 Rhus toxicodendron L. 290 venenata DC 290 Rhynchosia Lour. . 82 volubilis Lour. 82 Riedelia Oliv 78 Rottboellia setacea Roxb. 7i Rottlera albicans Hassk. S86 barbata Wall 397 Ruellia cernua Roxb 414, 419 zeylanica Roxb 414, US Ruppia L 228 maritima rostellata Koch 228 Rutaceae 373 S Sabiaceae 294 Saccoloma Kaulf iS Saguerus rumphii Ro.xb 414, ilS Sagus inermis Roxb 414, 419 laevis Willd. ili spinosus Roxb 414, 419 Salacia L 291 griffithil Laws 292 perakensis King 292 philippinensis Merr 291 Page. Salomonia Lour. 237 ciliata (L.) DC 237 oblongifolia DC 2S7 Sandoricum Cav 237 indicum Cav .— 2S7 koetjape (Burm. f.) Merr. .. 237 Sapium P. Br 406 crassifolium Elmer S8S indicum Willd, i07 sanchezii Men. 406 sebiferum Roxb 414, 416 subrotundifolium Elmer 38i virgatum Hook, f i07 Sarcopodium dearei Reichb. f 127 Sarcostigma W. & A. 29S Saurauia Willd 91, 306 macgregorii Merr. ..._ 306 vanoverberghn Merr 91 Sauropus Bl 407 robinsonii Meir 407 Saxifragaceae _ 271 Scaevola L 363 frutescens (Mill.) Krause 363 frutescens sericea (Forst.) Merr 354 koenigii Vahl S53, SSi plumieri (L.) Vahl SSi sericea Forst. SSi taccada Roxb 413, il6 Schefflei-a Forst 95, 318 albida Merr. 95 stellulata Merr. 318 Schizoloma ensifolium J. Sm iS Scleria Berg 76 pergracilis (Nees) Kth 75 Scrophulariaceae 100, 246 Scutellaria L 349 copelandii Merr 349 indica L S50 Scyphularia pentaphylla ( Bl. ) Fee 6S simplicifolia Copel 64 triphylla (Hook.) Fee 65 Selaginella exigua Spring il9 Selago pusilla Thunb. .. 2iS Selliguea Bory i7 Semecarpus L 282 acuminatissima Merr 282 albescens Kurz 290 anacardium Blanco 289 cassuvium Roxb 413, 418 cassuvlum Spr US cuneiformis Blanco 288, 289 euphlebia Merr. .. 283 lanceolata Merr. 284 megabotrys Merr. 285 micrantha Perk. . 290 microcarpa F.-Vill 289 obtusifolla Merr. 286 paucinervia Merr. 286 perrottetii March 286, 289 perrottetii glabra March 290 philippinensis Engl 28S, 290 pilosa Merr 287 pubescens F.-Vill 289 INDEX. 433 Page. Semecarpus sideroxyloides Perk 289 taftiana Perk 290 whitfordii Merr 288 Serratula anthelmentica Roxb 414, US cinerea Roxb. ... 414, US Sesbania Pers 236 aegyptiaca Poir 235 picta Cav. 235 sesban (L.) Merr . 235 Sida L . 90 acuta Burm. . 240 corylifolia Wall . 91 puberula Merr. 90 Simarubaceae ... 236, 274 Sisyrinchium bulbosum Mill 233 palmifolium L 233 Skinneria caespitosa Choisy . 244 Solanaceae 350 Solanum L 350 blumei Nees 351, 352 epiphyticum Meir 350 parasiticum Bl. 351 phillppinense Merr 351 Spathiostemon javense Bl 391 Sphaeromariscus microcephalus E. G. Camus 231 Sphagnaceae 263 Sphagnum L 264 cuspidatulum C. Mull 265 japonicum Warnst 268 junghuhnianum D. & M 264 luzonense Warnst 266 robinsonii Warnst 256 sericeum C. Miill 265 subrufescens Warnst 257 Spinifex L 229 littoreus (Burm. f. ) Merr 229 squarrosus L 229 Springia indica Muell.-Arg 333 Staurogyne Wall 248 rivularis Merr 248 Stenochlaena aculeata Kunze 67 intermedia Copel 67 Stephania Lour 265 catosepala Diels 266 ramosii Diels . 266 Sterculiaceae 304 Sterculia guttata Roxb 414, 416 Stipa littorea Burm. f 229 spinifex L. 229 Stipellaria parviflora Benth 382 Stylidiaceae 249 Stylidium Sw 249 alsinoides R. Br 249. 260 tenerrimum F. Muell 240, 260 Suriana L 274 maritima I. 274 Symplocaceae 29 Symplocos ahernii Brand 32 angulans Elmer 33, Si cagayanensis Brand 35 ciliata Presl SI confusa Brand 31 congesta Brand 33 Page. Symplocos cumingiana Brand 34 curtiflora Elmer ... 34 depauperata Merr. 36 depauperata angustissima Brand 36 ferruginea philippinensis Brand 32 floridissima Brand 32 fragrans Elmer S3 imbricata Brand ..._ 31 loheri Brand 32 luzoniensis Rolfe 35 merrilliana Brand 36 patens Presl 31 polyandra Brand . 32 purpurascens Brand 33 whitfordii Brand . _ 33 Syngramma alismifolia J. Sm 50 angusta Copel. ... 50 bomeensis J. Sm. 50 cartilagidens Diels 50 hookeri C. Chr. . 50 pinnata J. Sm. .. 50 wallichii Hook 50 T Tabernaemontana L 242, 337 cordata Merr 337 crispa Roxb 413, 418 pandacaqui Poir 243 sphaerocarpa Bl 243 subglobosa Merr 242 Taenitis Willd 47 blechnoides Sw. 48 Tapeinidium C. Chr. .. 48 Tectaria decurrens Copel. 54 menyanthidis Copel 55 weberi Copel 64 Telosma Coville 243 angustiloba (Warb.) Merr 243 procumbens (Blanco) Merr 243 Teramnus Sw 83 angustifolius Merr .-. 83 Tetrastigma Planch. . 86 angustifolium Planch 87 brunneum Merr 86 crassipes Planch 89 lanceolarium Planch 86, 87 loheri Gagnep. .. _ 87 pedunculare Planch 89 pergamaceum Bl 86 phillppinense Merr 86 sepulchre! Merr 88 Teucrium L .; 100 phillppinense Merr 100 tomentosum Heyne _ 101 Thayeria Copel 41 cornucopia Copel. U, 43 nectarifera (Bak.) Copel 41, 43 Thottea dependens K1 il9 Thymelaeaceae 92, 312 Tiliaceae 89, 239, 296 Toon a Roemer 279 paucijuga Merr. ..._ 279 434 INDEX. Page. Torenla hirta Cham. & Sehlecht 2i6 Trachelospermum L 97 Trachelospermum fragrans Hook, f 97 jasminoides Lem 97 philippinense Elmer .. 97, SSi vanoverherghii Merr. .. 97 Tradescantia malabarica L 232 Trema Lour 260 amboinensis Bl. 261 vulcanica Merr 260 Trichilia nervosa Vahl 237 Trichomanes craspedoneurum Copel 63 henzaianum Hook 53 sublimbatum C. Mull 53 Trigonostemon Bl 408 everettii Merr 408 longipedunculatus Elmer.. 410 maerophyllus Muell.-Arg. 404 oblongifolius Merr 409 sumatranus Pax & K. Hoffm 410 Tristania E. Br. 317 littoralis Merr 317 Tristellateia Thouars 236 australasiae A. Rich 236 australasica Auct. 236 australis A. Rich. 236 malintana Blanco 236 Tritaxis macrophylla Muell.-Arg 404 Triumfetta L 296 bartramia L „ 239 fabreana Gaudich 296 indica Lam. 239 procumbens Forst 296 radicans Bojer 296 repens (Bl. ) Merr. & Rolfe.. 296 i-homboidea Jacq 239 Tropidia Lindl 7 angulosa Bl. 7 calcarata Ames 7 II ' Ulmaceae 260 Unona uncinata Dun 234 Urtica tenacissima Roxb 414, 416 Utricularia L 102, 247 exoleta R. Br 102 nivea Vahl 247 racemosa lilicauhs C. B. Clarke .... 247 scandens Benj 247 scandens Oli- 247 tenerrima Merr 247 wallichiana Wight 247 V Vaccinium L 16, 321 bancanum Miq 322 barandanum Vid. 323 camiguinense Merr 321 epiphyticum Merr 322 hutchinsonii Merr 323 Vaccinium indutum Vid jagori Warb loheri Merr microphyllum Bl mindorense Rendlc Vallaris Burm Page. 96 322 323 323 323 335 angustifolia Merr. 335 lancifolia Hook, f 336 Vallisneria octandra Roxb 73 Vandellia L 246 erecta Benth 417 hirsuta Ham. 246 pusilla (Willd.) Merr 246 seabra Benth 246 viscosa (Willd.) Merr 246 Vanoverberghia Merr 76 sepulchre! Merr 76 Verbenaceae 98, 246, 338 Verbesina prostrata H. & A 251 Vernonia Schreb 106 acrophila Merr 106 phanerophlebia Merr 106 Vetiveria zizanioides Nash 228 Visiania pubinervis Miq 329 Vitex L 343 nitida Merr. ... 343 pentaphylla Men 344 Vitis latifolia Roxb. . 413, 416 Vittaria Sm 47 elongata Sw 48 minor Fee . 53 scolopendrina Thw. 48 Voacanga Thouars ... 336 megacarpa Merr. 336 Volkameria commersonii Poir 245 infortunata Roxb 414, 416 nereifolia Roxb 245 W Wedelia Jacq. 251 prostrata Hemsl. 251 Wikstroemia Endl .... 92, SIS angustissima Merr 92 lanceolate Merr 93 Wollastonia prostrata H. & A .. 251 Y Yucca superba Roxb. 414, 416 z Zanonia sarcophylla Wall 352 Zanthoxylum lamarkianum Cham ... 375, 376 marambong Miq 378 pteleaefolium Champ 376 376, 377 roxburghianum Cham. ... 378 zeylanicum DC 378 Zebrina Schnizl. 232 pendula Schnizl Zingiberaceae 232 76 Zoisia Willd. 230 matrella (L.) Merr 230 pungens Willd 230 o PUBLICATIONS FOE SALE BY THE BUREAU OF SCIENCE, MANILA, PHILIPPINE ISLANDS A LIST OF THE MAMMALS OP THE PHILIPPINE ISLANDS, EXCLUSIVE OF THE CETACEA. By Ned Hollister. Order No. 418. Paper, $0.50 United States ourrenoy, postpaid. This is the only recent attempt to enumerate the mammals of the Philippine Islands. The distribution of each species Is fliven and the original descriptions are olted. PRICE-LIST OP PHOTOGRAPHS. For sale by the Bureau of Science. Order No. 417. Por free distribution. This Is a list of selected photographs from the splendidly complete collection of the Bureau of Science. A MANUAL OP PHILIPPINE BIRDS. By Richard C. McGregor. 2 parts, 769 pages. Order No. 103. Paper, $4 United States ourrenoy, postpaid. Mr. McGregor spent some eight years in active field work, visiting many parts of the Archipel- ago, before beginning work on this book. Therefore, he was well prepared to undertake the preparation of the manual. A Manual of Philippine Birds contains in compact form descriptions of all the known species of Philippine birds. The usual keys and diagnoses of orders, families, and genera help the novice In Identifioation. Under each species are found native, English, and scientific names, distribution by islands, descriptions of the birds and In many instances notes on nesting, migrations, and other habits. A CHECK-LIST OF PHILIPPINE FISHES. By David Starr Jordan and Robert Earle Richardson. 78 pages. Order No. 102. Paper, $0.75 United States currency, postpaid. This list will be found a convenient guide to the synonymy of Philippine Ichthyology. The nomenclature is thoroughly revised and the distribution of each species within the Philippine Islands is given. This check-list is uniform in size and style with McGregor and Worcester’s Hand-list of Philippine Birds. INDO-MALAYAN WOODS. By Fred W. Foxworthy. 182 pages, 9 photographic plates. Order No. 411. Paper, $0.50 United States currency, postpaid. In Indo-Malayan Woods, Doctor Foxworthy has brought together a large amount of accurate Information concerning trees yielding woods of economic value. The work is based largely upon the author’s own experience in the Philippine and neighboring regions, but previous publications and Information generously given by other dendrologists have been used to correlate commercial and native names of useful Indo-Malayan trees. PHIIilPPINE HATS. By C. B. Robinson. Order No. 415. Paper, $0.50 United States currency, postpaid. This paper is a concise record of the history and present condition of hat making in the Philippine islands. The various materials used and the different kinds of hats made in each center of production are fully described. Not of the least importance are the botanical identifica- tions of the plants from which the hat materials are obtained. The plates illustrate the hat materials and various kinds and grades of hats. A map of central Luzon shows the towns chiefly concerned in this industry. * THE COCONUT PALM IN THE PHILIPPINE ISLANDS. 149 pages, 30 plates. Order No. 37. Paper, $1 United States currency, postpaid. The reprint contains the following articles: On the Water Relations of the Coconut Palm (Cocos nucifera), by Edwin Bingham Copeland; The Coconut and its Relation to Coconut Oil, and The Keeping Qualities of Coconut Oil and the Causes of its Rancidity, by Herbert S. Walker; The Principal Insects Attacking the Coconut Palm (Parts I and II), by Charles S. Banks; with an introduction by Paul C. Freer. A VOCABULARY OP THE IGOROT LANGUAGE AS SPOKEN BY THE BONTOK IGOROTS. By Walter Clayton Clapp. S9 pages. Order No. 40S. Paper, $0.75 United States currency, postpaid. The introduction to this vocabulary contains notes on pronunciation, vowels, diphthongs, con- sonants, verbs, conjugations, syllabication and reduplication. The vocabulary is given in Igorot- English and English-lgorot. THE NABALOI DIALECT. , By Otto Scheeeer. 65 pages, 29 plates. AND THE BATAKS OF PALAWAN. By Edward Y. Miller. 7 pages, 6 plates. Order No. 403. Paper, $0.25; half morocco, $0.75 United States currency, postpaid. The Nabalol Dialect and the Bataks of Palawan are bound under one cover. THE BATAN DIALECT AS A MEMBER OP THE PHILIPPINE GROUP OF LANGUAGES. By Otto Scheeeer. AND “F” AND “V” IN PHILIPPINE LANGUAGES. By Carlos Everett Conant. These two papers are Issued under one cover, 141 pages. Order No. 407. Paper, $0.80 United States currency, postpaid. Orders for these publications may be sent to the Business Manager, Philip- pine Journal of Science, Bureau of Science, Manila, P. I., or to any of the agents listed below. Please give order number. The Macmillan Company, 64-6 6 Fifth Avenue, New York City, U. S. A. Wm. Wesley & Son, 28 Essex Street, Strand, London, W. C., England. Martinus Nijhoff, Nobelstraat 18, The Hague, Holland. Mayer & Miiller, Prinz Louis Perdinandstrasse 2, Berlin, N.W., Germany. Kelley & Walsh, Limited, 32 Raffles Place, Singapore, Straits Settlements. A. M. & J. Ferguson, 19 Baillie Street, Colombo, Ceylon. Thacker, Spink & Co., P. O. Box 54, Calcutta, India. PUBLICATIONS FOB SALE BY THE BUBEAU OF SCIENCE, mNILA, PHILIPPINE ISLANDS I'HE SUB^VXtJ?fS OP SINDANGAN BAY. ^ , By Emessojt B. Ciikistie, 121 pages, 1 map, 29 plates. - Order No. 410. ' Pajler, $1.25 Uniled Stales currency, postpaid. Sindanaan Bay is situated on the northern coast of Zamboanga Peninsula. The Subanuns of this region were studied by Mr. Christfe during two periods of five and six weeks,. respectively. , The following is an abstract from the contents of .Mr. Ghristleis report on the Subanuns: Habitat and history; relations with the Moros; material culture; houses; industries; trade; agriculture; family life; social customs; administration of luslioe; refigion; the medicine manr ceremonies; tales; word-lists; physical measurements. ’ • - ' The 29 plates illustrate the Subanuns at work and at play; their industries, houses, altars, and Implements; and the people themselves. THE HISTORY OP SUBtJ. By ISTajeeb M. Saleeby. 275 pages, 4 maps, 2 diagrams. , Order No. 406. - Paper, $0.75 United States currency, postpaid. In the preparation of his manuscript for The History of Swiu Doctor Saleeby spent much, lima and effort in gaining access to documents in. the possession of the Sultan of Sulo. . It is fortunate that these records have now been translated and preserved in permanent form. This book Is a history of theJVIoros in the Philippines from the earliest limes lo the American oooupatiom ' STUDIES lY MORO HISTORY, LAW, AXD RELIGION. By Najeeb M, Saleeby. 107 pages, 16 plates, 5 diagrams. Order No. 405. ^ ; Paper, $0.25; half moroeoo, $0.75 United Statesjcurrsncy; postpaid: This volume deals with the earliest written records of the Moros in Mindanao. Doctor Saleeby was fortunately able to obtain exact copies of carefully preserved early records written in the Maglndanao dialect with Arabic characters. The author presents translations of these as: well, as 16 half-tone illustrations of certain pages from the originals, Jhe names of the rulers of Magindanao are recorded in five folding diagrams. NEGRITOS OF ZA3VIBALES. ‘ By William Allan Reed. - .. ■' 83 pages, 62 plates. v.. v' Order No. 402. Paper, $0.25; half morocco, $0.75 United States currency, poslpaidv The introductory chapter deal's with the general distribution of Negritos and with the distribu- tion of the Philippine branch of the race. The succeeding chapters deal with Hhe various Indus-: tries, amusements, and social relations of these little men. ' V • Plates from photographs, the greater part of which was taken for this publication, show; ornaments, houses, men making fire with bambooi bows and’arrows, dances, and various types of the people themselves. »v=>r’-nr' iVlf, CO^TMTf> 'r :f ' '■ MERRILL, E. D. The Pineda Monument and the"' ,' Probable Site of the First Botanic Garden imthe ,^;, ; , y';V; ‘-^\;'^hiiipi>infes MERRILL, E> D. On the Identity of Evbdia triphyjla- ‘ ' / ,^ MERRILL} P.* Notes on Philippine Euphorbiaceae _ 3^9 ^ RQBINSON, C. B. Roxburgh's Hortus Bengalensis :%rata;- , cui The ^‘Philippine' Journal of Sciehep** is issued as follows: Section A. Chemical and Geological' Sciences and the Industries.. Section B. The Philippine Journal of Tropical Medicine "Section . C, Botany ,v.i,,iA...„.:,j..:.:....:..'.'.:.'......:,..Jj ^Section I>. General Biology, ■EthnOlog'y and Anthropology (SOc* tion 1) began with Volume V) Entire Journal, Volume IV, or V .1 .J '....“V. Entire Journal, beginning with yoludie YJ Single numbers I ] ’* Volume 1, 1906 (not divided intp' sections) Supplement to Voltjnie ll (botany) 1 Volume 1 (without- supplement). ^..:..;......i,i ...' , Publications sent in exchange for the Philippine Journal of Soiekee should be addressed: Library, Bureau of Science, Manila, k I.' . / Subscrjptions may be sent io the Business M,anager, Philippine Jour- nal of Science, Bureau of Science, Manila, P.J., or to any of, the, agents listed b&low: ' , 'y. i?ie Macmillan Company, 64-66 Eiftb Avenue, New Yotk City, S. A. Win. .Wesley & Son, 28 Essex Street, Strand, London, W, C., England. Martinns Nijhpff, Nobelfelraat IS, The Hague, Holland. Mayor &MiUler, Prinz Louis Eerdiuandstrasse 2, Berliii, TiT.W., Germapy. Kelley. $! Walsh, Tumited, 3,2 liases Place, Singapore, Staraits Settlements. A, M. & J. Ferguson, l9 Baillie' Street, Colombo,' Ceylou.'> Thaeker, Spink & Cp., P. O. Box 54, Calcutta, India. : .7 -'v ,• ‘Ji