‘ae ' ' . : ~ ke \ ‘ ’ . ae cal als z * 4 ‘ \} > s , Me '| : 3 \ ‘ O\ ¢ 1.2 : A | ea” : 7 VS THE PHILIPPIN EK JOURNAL OF SCIENCE VOLUME XV JULY TO DECEMBER, 1919 WITH 64 PLATES AND 32 TEXT FIGURES MANILA BUREAU OF PRINTING er, 1920 K/ \w 168743 ” VU an ment it > EDITORIAL BOARD ELMER D. MERRILL, M.S., Editor R. C. McGrecor, A.B., Associate Editor ALBERT H. WELLS, A.B.; J. R. WRIGHT, PH.D.; A. P. West, PH.D. T. Dark JUAN, A.B., PHaAR.D.; F. AGcAorLi, A.B.; F. D. Reygs, B.S. A. S. ARGUELLES, B.S.; VicToRIANO ELICANo, B.S. Chemistry, Physics, and Geology H. W. Wands, M.D.; Otro ScHOsL, M.D. F. G. HAUGHWoUT; STANTON YOUNGBERG, D.V.M. Experimental Medicine Lisorio GomMEZ, M.D., PH.D.; F. CALDERON, B.A., L.M. VICENTE DE JESUS, M.D. Clinical Medicine W. H. Brown, PH.D.; C. F. BAKER, M.A.; H. S. Yates, M.S., PH.D. O. A. REINKING, B.S.A., M.S.; L. M. Guerrero, PHAR.D. Botany C. F. BAKerr, ‘M.A.; S. F. Ligut, M.A.; C. S. BANKS, M.A. L. D. WHarTON, M.A.; W. ScHULTZE; H. O. Bryer, M.A. Zoblogy and Ethnology A. B. BANYEA, Copy Editor CONTENTS No. 1, July, 1919 [Issued December 23, 1919.] COCKERELL, T. D. A. Philippine bees of the genus Nomia................ COCKERELL, T. D. A. The metallic-colored halictine bees of the eo Bea |! 8 Seb est eal is Se Sa toa DR, Ree OUR orc ET SA FUNKHOUSER, W. D. New records and species of Philippine Mem- ee Saar Rae aun SSN = eee RAR rN pene ee MU Maka One plate. SHUFELDT, R. W. Osteological and other notes on the monkey-eating eagle of the Philippines, Pithecophaga jefferyi Grant.................... Eleven plates. UICHANCO, LEOPOLDO B. General facts in the biology of Philippine MOUs ichirrw SET oss ee ac Four plates. BAKER, C. F. The Malayan Macherotine (Cercopide) ........0........ Three plates and one text figure. Cowias, KR. 2. Habits of tropical Crastacen: U7) One plate. LEVINE, C. O. Milk produced in southern China.............022200........-- One plate. Witt, J. C. The analysis of Portland cement raw mixture................ Two text figures. No. 2, August, 1919 {Issued December 29, 1919.] CRAWFoRD, Davip L. The jumping plant lice of the Palxotrepics and the South Pacific islands. Family Psyllide, or Chermida, Homoptera ................- Three plates and three text figures. DAMeE, Coy. The gents Erisoa (iassidee) oo Five plates. Page. 15 31 59 67 81 91 107 123 125 139 209 iv Contents No. 3, September. 1919 [Issued January 7, 1920.] DEL ROSARIO, MARIANO V., and MARANON, JOAQUIN. The physico- chemical evaluation of tikitiki extract............00000000000000000.. Sees MERRILL, E. D. Additional notes on the Kwangtung flora............... Koorpers, S. H. Notiz tiber Hoya imbricata Callery ex Decaisne und Hoya pseudomaxima Kds. in den Filipinen auf Grund von einigen Herbar-Exemplaren des Bureau of Science in Manila... Four plates. COCKERELL, T. D. A. The black halictine bees of the Philippine CTE RECS BWR gap apuntee ey BEAT pe a ar tr ee IRE plc eee EO BANKS, CHARLES S. The swarming of anopheline mosquitoes............ BANKS, CHARLES S. The Philippine leaf-mining buprestids, one DOME COW ait le Ria Sie ie eas Three plates. BAKER, C. F. Notices of certain Fulgoroidea, II: The genus Tro- ONDINE: 5625) a tae ai hccpcist Reienanctbtncgega gk Six text figures. GROFF, ELIZABETH H. Soy-sauce manufacturing in Kwangtung, RSPR ole sie Sag ee aa ccd: ashok Cons a eee eee aes No. 4, October, 1919 {February 27, 1920.] OSHIMA, MASAMITSU. Formosan termites and methods of prevent- Ing SRO: Gambro se a ee sd ee Thirteen plates and five text figures. COCKERELL, T. D. A. A new scale insect on Rhizophora.................... One text figure. DE LEON, WALFREDO, Balantidium haughwouti, new species, parasitic in the intestinal tract of Ampullaria species, a morphological ses See en ane aca py nei Mier he See erence aeRO One plate and five text figures. No. 5, November, 1919 {March 1, 1920.] Bezzi, M. Fruit flies of the genus Dacus sensu latiore (Diptera) fromthe Philippine lands... 355 ce ‘ Two plates. FLEUTIAUX, Ep. Melaside nouveaux (Coléoptéres) recoltés par C. “UP RB a eee pe ee re CANIZARES, MIGUEL. Some abnormalities of the vertebral artery........ One plate and five text figures. . PERKINS, GRANVILLE A. The rancidity of Philippine coconut oil........ BuRKILL, I. H. The genus Gordonia in the Philippine Islands.......... REINKING, O. A. Higher Basidiomycetes from the Philippines and their: hidwte, foe a ee 55 cis cutoubsa nes eed eae Page. 221 225 263 301 307 317 319 385 389 Contents No. 6, December, 1919 {March 30, 1920.] SHAW, WALTER R. Campbellosphaera, a genus of the Volvocaceae.. Two plates and one text figure. Muir, FREDERICK. Some Malayan Delphacide (Homopitera)............ One plate. WORCESTER, DEAN C. A nesting place of Micropus subfurcatus in be ge SAP LecaE REO REE AEP COT PT OO NEON IOS CANE NESS CELE MRRP EIS NE REE PREP ICRC Ce HAUGHWOUT, FRANK G. A method for labeling slides used in routine TIE Re RSG ALOT GENET, SSE SED OT HOES OE ARE NT One text figure. Merritt, E. D. Additions-to the flora of Guami:...:...........00c0008 SCHULTZE, W. Seventh contribution to the Coleoptera fauna of the Re PRRRR INI asics care ic gg cles es aC Sa eces lb acces ous escdind Pict ovicits One text figure. HauGHWOUT, FRANK G., LANTIN, PEpRO T., and FERNANDEZ, RICARDO. A case of acute mania associated with Plasmodium vivax infec- gay Magee ts oa eas cae! Na RUA aE nga SU ato SEEN eS ER ecu s 545 THE PHILIPPINE JOURNAL OF SCIENCE VoL. XV JULY, 1919 No. 1 PHILIPPINE BEES OF THE GENUS NOMIA By T. D. A. CocKERELL Of the University of Colorado The andrenid genus Nomia is widely distributed over the world, but is much more numerous in species in the Eastern than in the Western hemisphere. The species show such diverse characters that authors have at different times proposed to divide the genus into several genera and subgenera. It is doubt- fully expedient, however, to recognize more than one genus, though most of the names proposed may properly be used in a subgeneric sense. The type of the genus, as Meade-Waldo has shown, is the Asiatic Nomia curvipes of Fabricius. This has colored tegumentary bands on abdomen, postscutellum unarmed, and the hind femora of the male greatly swollen and sharply toothed beneath. It may be considered typical of a group which in the Philippine fauna includes N. thoracica, N. longitarsis, N. lusoria, N. strigata, and N. iridescens; but none of these has the hind femora of the male as in curvipes. In N. thoracica these femora are swollen, but not toothed beneath. Hoplonomia Ashmead resembles true Nomia, but has a bi- dentate process, or two teeth, on the postscutellum. The type is N. quadrifasciata, a Philippine species. The following key will facilitate the separation of the Philip- pine species: Key to the Philippine species of Nomia. Abdomen pale red (Sulu Island) dimidiata Vachal. Abdomen not red bi 1. Antenne and legs entirely dull ferruginous; female, 12.5 millimeters ME RIG WET) Fog. sis apres icc philippina Vachal. _Antennz and legs otherwise colored; or, if reddish, smaller species... 2. 2. Postscutellum bidentate or bispinose..........—.............--.----e.---cceee-ee--ne 3. Postscutellum unarmed 4, 3. Male with hind tibie, except a large black spot behind, and hind tarsi, yellowish : quadrifasciata (Ashmead). Male with hind tibiz, except apical process, and hind basitarsi, black or brown quadrifasciata notha Cockerell. 166316 2 Philippine Journal of Science 1918 4, Abdomen with tegumentary bands, not covered with hair.................... 5. Abdomen without such bands, though the hind margins of segments may be reddish or testaceous............... 7. 5. Bands white ... thoracica stantoni (Ashmead). Bands yellowish or red longitarsis Cockerell. Bands green, sometimes suffused with red................ Sus Os 6. First abdominal segment without a band...................... iridescens Smith. First abdominal segment with a band.................... strigata (Fabricius). Like strigata, but hind legs modified (see description). lusoria sp. nov. SF Fe ey UU RO) ro a ea ee 8. Fe NIE Gat yc SN Re cas C0 a neg eee en ee RE ETON 10. 8. Fifth ventral segment of abdomen of male ith two red spots bearing DINE POMOC she GRE elongatula Cockerell. ee a ING IO ga canines an Raps eweteeree ea 9. 9. First abdominal segment of male narrow and little punctured. lautula sp. nov. First abdominal segment broader and more distinctly punctured; IREQOT: RPOCI Gs. FSS SES A - Ssd levicauda Cockerell. BG IO iis Sco ciestanr cas lore fe dare sttenain no aac nr Mice ta eae LeN Eta uhin con sip mtn 11. Big cr Nee em CE SEAR SI TARY SEN Seno ieee Op ats SpE SSS see SRC Epes avg Tee 12. 11. Hind basitarsi pallid: hind femora enormously swollen. goniognatha sp. nov. FAI DARIN ENS GAS on ck nitascadivsiencics aa philippinensis (Friese). 12. Rather large species, quite robust, with coarse punctures on mesothorax and excessively fine and dense punctures on abdomen; hind margin of first abdominal segment black recessa sp, nov. Smaller species, with other characters 13. 18. Stigma small; area of metathorax almost or quite linear. philippinensis (Friese). Stigma large; abdomen of male clavate pont 14. Second abdominal segment smooth and shining............ levicauda sp. nov. Second abdominal segment rough and punctured..... palavanica Cockerell. Nomia quadrifasciata (Ashmead). Hoplonomia quadrifasciata ASHMEAD, ¢ (as “?”), Journ. N. Y. Ent. Soc. 12 (1904) 4. Nomia quadrifasciata (ASHMEAD) COCKERELL, Entomologist 48 (1915) iB ie LuZzON, Manila (W. A. Stanton). MINDANAO, Dapitan (Baker). The Dapitan specimen agrees with Ashmead’s account, but is I think certainly only a variation of the commoner form, Nomia notha. Unless additional material from Manila shows better distinctive characters, I must conclude that there is only one distinct species involved. Nomia quadrifasciata notha (Cockerell ined.). Nomia notha COocKERELL, Entomologist 51 (1918) 112. Hoplonomia quadrifasciata ASHMEAD, ¢ (not “?”’), Journ. N, Y. Ent. Soc. 12 (1904) 4. Se OM, XV, 1 Cockerell: Philippine Bees of the Genus Nomia 3 Nomia incerta “GrRIBODO” COCKERELL, Entomologist 48 (1915) 177 (not incerta Gribodo). The receipt of genuine Nomia incerta from Java (Mount Salak, 2,500 feet, 3. 7. 09, Bryant and Palmer, United States National Museum), shows that this species (female) has the mesothorax with punctures of two sizes and the lobes of the postscutellar process very obtuse. The Philippine insect is clearly distinct, having the mesothorax with strong practically uniform punc- tures, and the lobes of the postscutellar process long and acute. The type, from Los Bajios, is a female having the three abdo- minal bands apple green suffused with vermilion. All my ma- terial is from Baker. LuzON, Manila (Stanton): Laguna Province, Los Bafios and Mount Maquiling: Tayabas Province, Mount Banahao. MIN- DANAO, Dapitan and Davao. LEYTE, Tacloban. PALAWAN, Puerto Princesa. Related to N. simplicipes Friese.’ The following varieties look very distinct, but do not repre- sent true races: | Nomia quadrifasciata var. aurantia var. nov. Female.—Like Nomia notha, but abdominal bands rich orange, slightly tinged with apple green. LuzoN, Laguna Province, Mount Maquiling (Baker). Two specimens; one is Baker 7469. Nomia quadrifasciata var. viridans var. nov. Female.—Like Nomia notha, but abdominal bands very bright emerald green. Luzon, Laguna Province, Los Bafios (Baker). Nomia thoracica stantoni (Ashmead). Paranomia stantoni ASHMEAD, Journ. N: Y. Ent. Soc. 12 (1904) 4. Nomia thoracica stantoni (ASHMEAD) COCKERELL, agar bat 48 (1915) 177. LUZON, Manila (Stanton): Laguna Province, Los Bafios and Mount Maquiling (Baker). An insular race, smaller than typ- ~ jeal N. thoracica Smith. -*Nomia simplicipes was described from Kaulun, near Hongkong. The male has the first segment of the abdomen green banded, and may not be conspecific with the female. As the specific name is derived from a character of the male, I herewith designate that as the type. This male is readily separated from N. notha; but a female simplicipes (agreeing with Friese’s description) from Trong, Lower Siam, January-February, 1899 (W. L. Abbott; United States National Museum), is essentially like N. notha, except that the depressed part of the second abdominal segment is much broader (anteroposteriorly) in the middle. rae Philippine Journal of Science 1919 Nomia longitarsis Cockerell. ; Nomia longitarsis COCKERELL, Entomologist 49 (1916) 158. The type is a male. The female (Mount Maquiling) is sim- ilar, with the usual sexual differences, the mesothorax more finely punctured, and the abdomen with much shallower and more feeble punctures. The stigma in both sexes is ferruginous. Luzon, Laguna Province, Mount Maquiling (Baker). Nomia longitarsis eboris var. nov. Male and female.—Stigma black or reddish black; abdominal bands (except the first in female) clear opaque ivory color. Luzon, Nueva Vizcaya Province, Imugan, male (type): Ben- guet Subprovince, Baguio, female. Both from Baker. Probably a subspecies, but without strong characters. Nomia strigata (Fabricius). Andrena strigata FaBricius, Ent. Syst. 2 (1793) 311. Nomia strigata LEPELETIER, Hist. Nat. Ins. Hymén. 2 (1841) 291. Nomia iridescens “SMITH,” COCKERELL, Entomologist 48 (1915) 177 (not iridescens Smith). I had formerly supposed ? that the male of Nomia iridescens differed from the female in having the first five abdominal seg- ments with colored bands, but I now find that there are males and females with (strigata) and without (iridescens) the band on first segment. Some specimens of N. strigata, as is usual in this group of Nomia, have bands which fail to assume the proper color, but then these bands are pallid or reddish, very different from the opaque black hind margin of the first segment in iri- descens. It must be stated, however, that strigata and iridescens are extremely closely allied, and it is by no means certain that they are not forms of a single species. Nomia quadridentata Smith is based on a female that is evidently strigata and on a male of the N. ellioti group. The first and fullest description is that of the female, which is herewith designated as the type. PALAWAN, Puerto Princesa. MINDANAO, Davao. NEGROS, Cuernos Mountains. All from Baker. The Philippine insect agrees with that from Java (Depok, Pelaboeam, Ratoe, Buiten- zorg), collected by Bryant and Palmer (United States National Museum). Nomia strigata var. ridleyi (Cockerell). Nomia iridescens var. ridleyi COCKERELL, Ann. & Mag. Nat. Hist. VIII 5 (1910) 503. *Ann. & Mag. Nat. Hist. VIII 5 (1910) 503. XV, 1 Cockerell: Philippine Bees of the Genus Nomia 5 MINDANAO, Davao (Baker 7468). Described from Singapore and also collected by Bryant and Palmer in Java (Buitenzorg, March, 1909; Depok). It is a color variety, doubtless alter- nating in Mendelian fashion with the typical form. Nomia iridescens Smith. Nomia iridescens SMITH, Journ. Proc, Linn. Soc. London, Zool. 2 (1857) 43. LuzON, Laguna Province, Los Bafios and Mount Maquiling. MINDANAO, Butuan and Davao. All from Baker. I have a fe- male from F. Smith’s collection, and the Philippine form agrees with it. Nomia iridescens var. rhodochlora var. nov. Male and female—Abdominal bands bright yellowish green, suffused with vermilion. MINDANAO, Dapitan (Baker 7466), male. NEGROS, Cuernos Mountains (Baker), female. The type locality is Dapitan. Nomia lusoria sp. nov. Male.—Like N. strigata ridleyi, but with third (as well as following) antennal joints clear red beneath; hind femora swollen, the upper side strongly convex; hind tibie greatly en- _larged apically, the end reddish yellow, the inner side flattened and expanded apically, its margin broadly rounded and ex- panded; hind basitarsi pale reddish dusky apically. PALAWAN, Puerto Princesa (from Baker). Nomia elongatula Cockerell. Nomia elongatula CocKERELL, Entomologist 48 (1915) 178. LuZON, Laguna Province, Los Bafios (type locality) and Mount Maquiling: Benguet Subprovince, Baguio. LEYTE, Tacloban. MINDANAO, Iligan. PALAWAN, Puerto Princesa. All from Baker. Nomia levicauda sp. nov. Nomia elongata “FRIESE,” COCKERELL, Entomologist 48 (1915) 178 (not N. elongata Friese). Finding that this group of Nomia included several distinct species, recognizable by the characters of the male abdomen, I was led to reéxamine the species from Luzon which Friese had identified with his N. elongata, described from Java. Evidently the Philippine species is distinct, for although Friese’s descrip- tion of the male is brief, he states that the fifth ventral abdom- inal segment is bare, whereas in our insect it is hairy right 6 Philippine Journal of Science 1919 across. The female is very similar to that of N. philippinensis, but the area of metathorax is rather broad (anteroposteriorly) and irregularly plicate, and the abdomen is shiny. The stigma is very much larger than that of philippinensis and is of the same general type as that of N. palavanica, but dusky or with a dusky margin. The second abdominal segment is smooth and shining, whereas in palavanica it is rough and punctured. LUZON, Laguna Province, Los Bafios (type locality) : Benguet Subprovince, Baguio. MINDANAO, Zamboanga. The females come from the following places: LUZON, Laguna Province, Mount Maquiling: Benguet Subprovince, Baguio. MINDANAO, Davao and Iligan. Ali from Baker. The series of females may in- clude examples of the next species, which is perhaps not clearly separable in that sex. Nomia lautula sp. nov. Maie.—Length, about 6 millimeters; slender, with the basi- tarsi cream color, their apices and the small joints of tarsi fer- ruginous; abdomen clavate, slender basally; the hind margins of segments with pale ocherous-tinted hair bands; face with pale golden hair; stigma large, dusky reddish. This species is so similar to N. levicauda that I thought it might be only a race, but it is certainly a distinct species, differing by the narrow and little-punctured first abdominal segment, the very densely punc- tured mesothorax, and especially the fifth ventral segment of abdomen, which has a pair of separate rounded elevations, cov- ered with short, dark-tipped, more or less bifid bristles, instead of being hairy right across as in N. levicauda. MINDANAO, Davao (Baker; two bear the numbers 7470, 7471). Nomia palavanica Cockerell. Nomia palavanica COCKERELL, Entomologist 48 (1915) 178; 49 (1916) 158. PALAWAN, Puerto Princesa (Baker 3848). Nomia philippinensis (Friese). Nomia takauensis philippinensis FRIESE, COCKERELL, Entomologist 48 (1915) 178. Luzon, Laguna Province, Los Bafios (type locality) and Mount Maquiling: Benguet Subprovince, Baguio: Tayabas Province, Malinao and Mount Banahao. MINDANAO, Davao. All from Baker. XV, 1 Cockerell: Philippine Bees of the Genus Nomia 7 Nomia goniognatha sp, nov. Male.—Length, 7 to 8 millimeters; black, with the knees, tibiz at apex, and tarsi except apex, yellowish ferruginous; pu- bescence pale ocherous, the mesothorax and scutellum with short, fulvous, mosslike hair; tongue very slender; eyes brown, strongly converging below; mandibles broad, pale testaceous, with the apex piceous and the swollen inner side of base broadly black, while the lower margin, beyond the middle, is conspicu- ously angular; vertex minutely rugose-punctate; area of meta- thorax very small, with five oblique strie; tegule reddish testaceous, with piceous base; wings dusky, stigma testaceous, not large; hind femora enormously swollen, helmet-shaped; hind tibiz short and broad, curved basally; abdomen finely punctured, the segments with pale fulvous hair bands; fourth ventral seg- ment emarginate, with a median keel on a ferruginous band; fifth ventral with a strong dark keel produced to a claviform apex. The flagellum is bright ferruginous beneath, except at base and apex. MINDANAO, Davao (from Baker). In Bingham’s table of Indian species this falls nearest to N. aurifrons Smith, to which are allied the later-described N. aure- obalteata Cameron and N. kangrz Nurse. Nomia aureobalteata has a large metathoracic area, different legs, etc. Nomia auri- frons has testaceous legs. Nomia kangrex also differs in the legs and in other ways. The hair on the thorax in N. goni- ognatha recalls the much larger N. thoracica or the Indian N. ardjuna Cockerell. Nomia recessa sp. nov. Female.—Length, about 9 millimeters or rather more; robust, black, with fine pale cinereous to white (on cheeks) hair, on mesothorax and scutellum thin and mainly fuscous, on vertex with a pale golden tint, on inner side of tarsi orange-ferrugi- nous; hind tibiz with hair mainly yellowish white, but on outer face dark fuscous; hind margin of first abdominal segment black, of the others broadly testaceous; no hair bands on apical ‘margins of segments, but third segment with an even white basal fringe beneath the testaceous margin of second. Eyes dark reddish; mandibles dark reddish, black at base; antenne black, the flagellum red apically, especially beneath; face broad, with thin cinereous hair; mesothorax densely and strongly punc- tured, but shining between the punctures; scutellum shining, 8 Philippine Journal of Science with large well-separated punctures; area of metathorax very small, but with strong ridges; tegule piceous with pallid margins; wings somewhat dusky, especially the apical marginal area; stigma dark castaneous; second submarginal cell receiv- ing first recurrent nervure much beyond middle; legs entirely dark; abdomen dull, with excessively fine punctures; venter with reddish fulvous hair. Luzon, Laguna Province, Mount Maquiling (Baker). Some- what related to the Indian N. basalis Smith, but differing in the pubescence and the color of the wings. Nomia dimidiata Vachal. Nomia dimidiata VACHAL, Miscellanea Entomologica 5 (1897) 8 of separate. ‘ SULU ISLAND. One female in collection of Gribodo. Length, 6.5 millimeters. Nomia philippina Vachal. Nomia philippina VACHAL, Miscellanea Entomologica 5 (1897) 7 of separate. PALAWAN. One female in collection of Vachal. The author suggested that it was possibly a variety of N. buddha Westwood. The latter he had from Perak. THE METALLIC-COLORED HALICTINE BEES OF THE PHILIPPINE ISLANDS By T. D. A. COCKERELL Of the University of Colorado The subfamily Halictinz, numerously represented in almost every part of the world, contains groups of species in which the surface of the body is black, brown, or yellowish, and others in which portions, at least, are metallic green or blue. The large genus Halictus, as generally understood, contains both metallic and nonmetallic species. In the Holarctic Region, Robertson has separated the metallic species as genera Chloralictus Robert- son and Seladonia Robertson. He also recognizes, in North America, genera Dialictus Robertson (with only two submar- ginal cells) and Paralictus Robertson. How many such genera should be segregated from the old Halictus, it is hard to decide; and certainly the Robertsonian groups are less distinct than the genera commonly recognized among bees. The French entomol- ogist Vachal went as far from current usage on the other side and not only refused to accept Robertson’s genera, but threw into Halictus such genera as Augochlora and Agapostemon, almost universally considered distinct. The genus Nomioides Schenck, placed as a subgenus of Halictus by Dalla Torre, must be con- sidered a distinct genus. Its members are small, usually with conspicuous yellow markings, almost perfectly imitating the American panurgine genus Perdita. Yet the venation is like that of Halictus, except that the marginal cell is narrowly trun- cate at the end. These bees are so unlike the species of Halictus that Cameron described several of them in Ceratina. Thus my Nomioides comberi from India must be called N. punjabensis (Cam.) ; for Meade-Waldo, on examining Cameron’s types, found that his Ceratina punjabensis, published about four years ear- lier, was in fact my insect. It also appears that Ceratina cerea Nurse is to be called Nomioides pulchella Schenck or, according to the synonymy indicated by Alfken, N. parvula (Fabricius). Sere Sa ss Mesothorax steel blue, highly polished 2. 1. Larger; abdomen with four entire yellow bands; mesothorax duller. dapitanellus Cockerell. Smaller; abdomen with four (varying to two) lateral yellow marks, representing two widely interrupted bands; mesothorax more «shin- ing. valdezi Cockerell. 9 © 10 Philippine Journal of Science ~ 1919 2. Larger (length, about 6 millimeters); abdomen black, without light markings cies ccelavened lac meatier serum .. melanogaster Cockerell. Smaller; abdomen marked with yellow........................... valdezi Cockerell. Nomioides dapitanellus Cockerell, Nomioides dapitanellus CoCKERELL, Entomologist 48 (1915) 107. MINDANAO, Dapitan (Baker 3161). Nomioides valdezi Cockerell. Nomioides valdezi COCKERELL, Ann. & Mag. Nat. Hist. VIII 16 (1915) 487, MINDANAO, Cagayan (Baker 3670), type locality. Also PALA- WAN, Puerto Princesa (Baker 8065). Inthe Palawan males the mesothorax is usually steel blue, as in the type, but occasionally it ig shining green. There are two females from Palawan. They are about 5.2 millimeters long; mesothorax green; scu- tellum with two large yellow spots; postscutellum with a yellow bar; abdomen with large oblique cuneiform yellow marks at sides of second and third segments. Nomioides melanogaster Cockerell. Nomioides melanogaster COCKERELL, Ann. & Mag. Nat. Hist. VIII 16 (1915) 487. MINDANAO, Dapitan (Baker 3162). Genus HALICTUS Latreille Females Ry ge A Males 4, 1. Mesothorax shining green; abdomen more or less purplish, green at base ...... = subpurpureus sp. nov. Mesothorax dullish or grandiar or abdomen not purplish... ae 2. Abdomen bright green 3. Abdomen bluish green to olive green, but knees red; scape red at base. adonidie sp. nov. Abdomen dark, at most slightly greenish.............. pervarians sp. nov. 3. All the tibie red; scape red at base................ taclobanensis Cockerell. Tibiz black; scape all black.......... imuganensis sp. nov. 4. Abdomen black, slender. oa pervarians sp. nov. Abdomen metallic, not slender............. 5. 5. Abdomen blue-green; legs mainly dark cyanescens sp. nov. Abdomen yellowish green; tibiz and tarsi clear ferruginous; base of CRE FOETUS a a taclobanensis Cockerell. Halictus taclobanensis Cockerell. Halictus taclobanensis CocKERELL, Ann. & Mag. Nat. Hist. VIII 16 (1915) 488. LEYTE, Tacloban (Baker 3672), type locality. Luzon, Mount Maquiling (Baker), female. MINDANAO, Davao (Baker 8067), XV, 1 Cockerell: Metallic-colored Halictine Bees 11 male; Cagayan (Baker), male. The male has the form of a female, as in certain Australian species, and the only conspicuous differences are in the longer antennz and the caudal end of the abdomen. Halictus subpurpureus sp. nov. Female.—Length, about 6.5 millimeters; head very dark greenish, almost black; mandibles rufous at apex; clypeus feebly and sparsely punctured; supraclypeal area purplish; front with a sericeous luster; antenne entirely black; mesothorax yellowish green, shining but not polished, without evident punctures; scu- tellum and hind margin of mesothorax glaucous green; area of metathorax large, with a very fine irregular reticulation; pos- terior truncation not sharply margined at sides; tegule dark brown; wings dusky, stigma black; first recurrent nervure meeting second transverse cubital; third submarginal cell narrow, quadrate, shaped nearly like the second; legs black; abdomen dorsally shining without bands and without evident punctures, the general color black, but the base greenish and the middle segments conspicuously suffused with purple; venter with large tufts of pale ocherous hair at the sides of the segments. LUZON, Baguio, Benguet (Baker). Related to H. pervarians, but larger, with mesothorax more shining, and abdomen suf- fused with purple. Halictus pervarians sp. nov. Female.—Length, about 5.5 millimeters; head rather small, glaucous green; antennze black; mesothorax and scutellum yellowish green, shining but not polished, without evident punc- tures; mesothorax with a median longitudinal depression; area of metathorax appearing dull and granular under a lens, the broadly rounded margin shining; tegule pale brown, dark at base; pleura bluish green; wings faintly dusky; stigma black; first recurrent nervure meeting second transverse cubital; third submarginal cell short; outer nervures of third submarginal and outer discoidal cells weakened; legs black; abdomen without bands, black with a greenish or brassy tint. Ventral segments of abdomen with lateral ocherous tufts as in H. subpurpureus. Variety a. Front peacock green and supraclypeal area purple. Variety 6b. Length, about 5 millimeters; tegul# pale testaceous; tarsi ferruginous; stigma brown. Variety c. Flagellum reddish beneath; mesothorax and scutellum very dark green; area of metathorax dark bluish. Stigma black as in the typical form. 12 Philippine Journal of Science 1919 Male.—Antennz very long, black; face green; abdomen slen- der, black; tarsi, and anterior tibie in front, ferruginous. A male from Mount Banahao is considerably larger than one from Mount Maquiling. LUZON, Mount Maquiling (Baker), type and varieties a and b; Imugan (Baker), variety ¢; Mount Banahao (Baker), male. Very variable, but apparently all one species. I have nine specimens before me. Halictus imuganensis sp. nov. Female.—Length, about 7 millimeters; bright golden green, the clypeus, mesothorax, and scutellum variably suffused with coppery red, or entirely green, or the clypeus peacock green; mandibles obscure reddish; face, front, mesothorax, and scu- tellum with a roughened granular surface; antenne black; area of metathorax shining, with strong, irregular plice, forming a coarse irregular network; posterior truncation sharply defined at sides; thorax above with thin ocherous hair; tegule rufo- piceous; wings very faintly dusky; stigma black; second submar- ginal cell large and broad, receiving first recurrent nervure at its apical corner; outer transverse cubital and recurrent ner- vures weakened; legs black; abdomen shining, without bands; a well-developed ventral scopa. LUZON, Imugan (Baker), type locality. Mount Banahao (Baker): Baguio, Benguet (Baker). Five specimens are be- fore me. Allied to H. taclobanensis. Halictus adonidie sp. nov. Female.—Length, about 6.5 millimeters; shining green, with dark tarsi, tegulze rufotestaceous, and basal half of scape red. Closely allied to H. tmuganensis Cockerell, differing in the color of scape, the bluish green (not brassy green) abdomen, the broader head, the scutellum and posterior half of mesothorax brilliantly shining, and the paler (brownish) stigma. LUZON, Manila, Malate (R. C. McGregor), October 27, 1918, at flowers of Adonidia merrillii Beccari. Halictus cyanescens sp. nov. Male.—Length, a little over 5 millimeters; head, thorax, and abdomen blue-green; face and front, including clypeus, dull and granular; antennz black; mesothorax granular; area of meta- thorax irregularly reticulate; posterior truncation sharply de- fined at sides; tegule*testaceous; wings very faintly dusky, XV, 1 Cockerell: Metallic-colored Halictine Bees 13 stigma piceous; legs black; tarsi reddened apically; abdomen shining, without bands. LuZON, Mount Banahao (Baker), two specimens. Very close to H. imuganensis and H. taclobanensis, having the robust form of a female in the male sex. The sexes of H. taclobanensis are colored alike, so I suppose the present insect to be distinct from H. imuganensis, from which it principally differs by being much smaller and quite differently colored. The green Halicti of the Philippines fall into two very dif- ferent groups or subgenera; one group consists of H. subpur- pureus and H. pervarians, and the second consists of the other three species. The first is Chioralictus of Robertson; the second, with males resembling the females, may form a new subgenus Homalictus; type, H. taclobanensis. NEW RECORDS AND SPECIES OF PHILIPPINE MEMBRACID A By W. D. FUNKHOUSER Of the Zoédlogical Laboratory, University of Kentucky ONE PLATE Since the publication of a revised checklist of the Membra- cide described from the Philippine Islands’ some very inter- esting new species and records have been received from Mr. R. C. McGregor, of the Bureau of Science, Manila, and from Prof. C. F. Baker, of the College of Agriculture, Los Bafios, ‘P. I., the study of which throws considerable light on the dis- tribution and the range of certain species that are extremely rare. Most of the material on which this paper is based was col- lected by Mr. McGregor in Antique Province, Panay Island, and on the small island of Batbatan, just off the west coast of Panay. Other valuable records received from Mr. McGregor were secured in Laguna and Rizal Provinces, Luzon Island, a few of them from the immediate environs of Manila. Mr. Mc- Gregor has secured the first data to be published on the food plants of a number of the species and has also collected for several species the nymphal forms, which have never been re- corded. I am greatly indebted to Mr. McGregor and again to Professor Baker for the opportunity of studying this very in- teresting material. Centrochares horrificus Westw. Plate I, fig. 1. Add: Habitat.—PANAy, Antique Province, Culasi (McGregor). Host.—Acalypha stipulacea Klotz. A good series of nineteen specimens collected at Culasi by Mr. McGregor in 1918. Six males and six females were taken May 18, in the forest at an altitude of about 700 meters; one male and four females, on May 27, on shoots from a stub of Acalypha stipulacea; one female, on May 29, and one female, on June 15. In spite of the long series of specimens of this species from the Islands which I have seen at various times, I am unable to * Funkhouser, W. D., Notes on the Philippine Membracidez, Philip. Journ. Sci. § D 13 (1918) 21-88. 15 16 Philippine Journal of Science 1919 recognize specific differences sufficient to admit C. posticus Buck- ton and C. bucktoni Distant as distinct species. It has already been noted in previous papers” that C. horrificus shows con- siderable variation in size and in coldr, some of which conforms to Buckton’s descriptions and figures for his supposed new species. I am still unconvinced that they are distinct. In addition to the adults recorded above, Mr. McGregor sends a nymph of this species, which is apparently in the third or fourth instar, taken with the adults at Culasi on May 27 and sent to me on a section of the leaf on which it was captured. Since the immature form (Plate I, fig. 1) of this insect has never been noted in literature, it seems advisable to include a brief description, as follows: Entirely yellow (specimen preserved in alcohol); length, 4 millimeters; width between humeral tubercles, 1.5; height from apex of clypeus to top of metopidium, 2.8; roughly sculptured; densely pubescent; pronotum showing a suprahumeral tubercle on each side and a median dorsal posterior process which ex- tends backward over the mesothorax but does not reach meta- thorax. Head longer than broad, roughly pubescent, without tubercles, ocelli not visible, eyes large. Thorax distinctly divided into segments; prothorax rough, densely pubescent, metopidium perpendicular, pronotum showing two strong tubercles representing the suprahumeral horns, posterior process extending backward over mesonotum but not reaching metanotum; mesonotum with a small tubercle on each side of median line; wing pads well developed, extending to first abdominal segment; entire thorax yellowish and densely pubescent with grayish hairs. Abdomen showing nine segments, each segment slightly nod- ulate above with tufts of bristly hairs and extended ventro- caudad at lateral margin with edges of segment pilose; anal tube short, blunt, slightly upraised. Legs flattened, first two pairs somewhat foliaceous, very densely pubescent; tarsi well developed, three joints distinct; claws swollen and heavy. ! Judging from the development of the pronotal tubercles, the length of the wing pads, and the invisible ocelli, this nymph is in its third instar. It was attached by the first pair of legs to thg midrib of the leaf in a position for molting. * Funkhouser, Philip. Journ. Sci. § D 10 (1915) 870 and 13 (1918) 23. XV, 1 Funkhouser: Philippine Membracidz 17 Pyrgonota bifoliata Westw. Add: Habitat.—PANAY, Antique Province, Culasi (McGregor). Eighteen specimens collected at Culasi by Mr. McGregor in 1918. Two specimens were taken in the forest at an altitude of about 700 meters on May 18, one on May 24, three on May 29, one on May 30, ten on June 15, and one on June 20. This species has often been figured to represent the grotesque and unusual development of the pronotum in the Membracide and usually stands as a “‘show species” in museums. It is un- usual to find so large a series as the present material offers. Pyrgonota bifurca Stal. Add: Habitat.—Luzon, Baguio, Benguet Subprovince (Baker). This is the first record of this species since its original de- scription by Stal in 1870.° Stal’s description is very detailed and closely fits the single specimen from Baguio. The lateral carina on the posterior process of the pronotum is not as dis- tinct as might be expected from the original description but is rather irregular throughout and obsolete at the extremity. In all other respects, however, the characters seem identical. The specimen before me bears Professor Baker’s duplicate No. 7657. Pyrgonota semperi Stal. Plate I, figs. 2 and 3. Add: Habitat—PANAY, Antique Province, Culasi (McGregor). Host.—Acalypha stipulacea Klotz. It has been a great pleasure to be able at last to recognize this species from three specimens collected by Mr. McGregor at Culasi—one on May 28, one on May 29, and one on June 30, 1918. The species has never been recognized since its original de- scription; * this description is very meager, being merely four lines of Latin comparing the type with P. bifurca. However, I feel confident that the Culasi material represents the insect which Stal described. It has the pale marking on the posterior process which is the chief distinctive character according to Stal, and the measurements exactly agree with those of the type. On account of the fact that the original description is so brief and the fact that the species has never been figured, I am including a more complete list of its characters as follows: Very dark brown, almost black, coarsely and evenly punctate, sparingly pubescent with grayish hairs; pronotal horn long, *Oef. Kong]. Vet.-Akad. Forhandl. No. 7 (1870) 731. * Loe. cit. 1668162 18 Philippine Journal of Science » 1919 erect, subcylindrical, narrowing toward the apex, apex forked, the prongs of the fork twisted and broadly foliaceous; poste- rior process long, slender, slightly elevated behind the middle, marked with a pale median line for basal two-thirds; tegmina entirely very dark brown or black. (Cf. Stal: “processu postico thoracis ante medium macula pallescente notato, pone medium quam anterius altiore, tegminibusque totis piceis.”’)* Head longer than broad, foliaceous, roughly sculptured, dark brown, coarsely punctate, finely pubescent; base sinuate, highest on either side of median suture; sutures distinct, elevated; eyes prominent, brown; ocelli very prominent, opalescent, shining, much farther from each other than from the eyes and situated well above a line drawn through centers of eyes; inferior mar- gins of genz sinuate, foliaceous, sloping downward abruptly to clypeus; clypeus foliaceous, distinctly trilobed, two smooth longi- tudinal strie down middle, finely punctate, sparingly pubescent, tip pilose. Pronotum uniform dark brown or black except narrow yellow- ish line on basal two-thirds of posterior process, roughly punc- tate, sparingly pubescent with grayish hairs on upper parts, very densely pubescent and tomentose just before humeral an- gles; humeral angles not prominent, obtuse; anterior prono- tal horn erect, long, slender, turriculate, heavy at base, gradually becoming narrower toward apex, bearing a distinct median carina in front, two or three irregular lateral ridges and a sharp posterior ridge which is strongly toothed, the teeth be- coming smaller and closer together as they approach the poste- rior process, apex of horn forked, forks bent backward, with a slight extension of the median column between the two prongs of the forks; prongs of the fork as seen from the front widely separated at their bases, then approaching each other and nearly touching at their tips, much compressed laterally below, some- what twisted and triquerate at apex, apical ends obliquely and sinuately truncate; posterior process long, slender, sinuate, high- est behind middle, basal two-thirds marked with narrow yellow- ish vitta, sharply carinate and denticulate above, tip depressed and extending somewhat beyond lateral angles of tegmina. Tegmina opaque, dark brown or black, strongly punctate and coriaceous in basal and costal regions, a very small pale spot at internal angles, tips long and acute, five apical and three discoidal areas. * Loc. cit. ee ee eT ae ee ee a ee J . XV, 1 Funkhouser: Philippine Membracidx 19 Sides of mesothorax and metathorax toothed, densely tomentose with white hairs; undersurface of body and abdomen brown. Legs entirely yellow; tibie flattened and minutely spined; claws ferruginous. Length, including tegmina, 7 millimeters; width between hu- meral angles, 2; height of pronotal horn from humeral angles to tips, 10. Leptocentrus reponens Walk. Add: Habitat.—Luzon, Manila: Rizal Province, Montalban: Laguna Province, Paete. PANAY, Antique Province, Tibiao, Culasi, and Lipata. (McGregor.) Host.—Blumea balsamifera (L.) DC.; Semecarpus cu- neiformis Blanco; Acalypha wilkesiana Muell.-Arg.; Mallotus moluccanus (L.) Muell.-Arg.; Hibiscus rosa- sinensis L. An excellent series with records of the host plants, collected as follows: Two from Montalban taken February 22, 1918, on Blumea balsamifera and Semecarpus cuneiformis; one from Ma- nila, February 1, 1917; three from Manila, February 5, 1917, on Mallotus moluccanus; one from Manila, February 5, 1917, on Acalypha wilkesiana; one from Manila, February 10, 1917, on Hibiscus rosa-sinensis; one from Lipata, July 9, 1918, on “a spiny shrub;” two-from Paete, March 6 and 13, 1917; five from Tibiao, May 14, 1918; four from Culasi taken in the forest at about 700 meters’ elevation on May 18, 1918; two from Culasi, May 20, 1918; one from Culasi, June 3, 1918; one from Culasi, June 19, 1918, and two from Culasi, July 24, 1918. The above data for hosts are the first ever recorded for this species, and all of the localities are new. Mr. McGregor’s field notes record the interesting fact that this species was not attended by ants. Leptocentrus leucaspis Walk. Add: Habitat.—Antique Province, Batbatan Island (Me- Gregor). = One male taken on Batbatan Island on June 30, 1918. Tricentrus robustus Funkh, Add: Habitat—Panay, Antique Province, Tibiao, Culasi, and Flores (McGregor). Host.—Acalypha stipulacea Klotz. A series of nine specimens, all females, collected in Panay in 1918 as follows: One taken in the forest at Culasi at 700 meters’ 20 Philippine Journal of Science 1919 elevation on May 18; one from Culasi collected on shoots from a stub of Acalypha stipulacea on May 27; three from Culasi, June 1; one from Culasi, June 20; one from Culasi, July 14; one from Tibiao, May 11; one from Flores, June 7. e Tricentrus convergens Walk. Add: Habitat.—Luzon, Manila. PANAyY, Antique Province, Culasi. (McGregor.) Host.—Mallotus moluccanus (L.) Muell.-Arg.; Acalypha wilkesiana Muell.-Arg. Six specimens collected as follows: Two females and one male taken at Manila, February 14, 1917; one female collected at Manila, February 5, 1917, on Mallotus moluccanus; one male from Manila, April 5, 1917, on Acalypha wilkesiana; one female from Culasi taken in the forest at an altitude of 700 meters on May 18, 1918. Tricentrus attenuatus Funkh. Add: Habitat.—PANAY, Antique Province, Culasi (Mc- Gregor). Two males from Culasi, one taken in the forest on May 18, 1918, and one at.a lower elevation on July 14, 1918. Tricentrus attenuatus was described from a male, and all of the specimens which I have seen of this species have been of the same sex. This fact makes me suspicious that perhaps we have here the opposite sex of some other species described from female material only. It will be of interest to discover an undoubted pair of this species. Tricentrus capreolus Walk. Add: Habitat—Luzon, Laguna tae Paete, Sarai bar- rio (McGregor). One female collected on March 19, 1917. Tricentrus fairmairei Stal. Plate I, fig. 4. Add: Habitat—Luzon, Rizal Province, Montalban: Laguna Province, Paete. PANAyY, Antique Province, Culasi. (McGregor.) Host.—Solanum verbascifolium L.; Eugenia calubcob C. B. Rob. Attendant ants.—Polyrhachis (Myrmhopla) dives Smith. A good series of specimens of this species has been received together with nymphs and attendant ants. Five specimens were collected at Paete on March 18, 1917; eighteen were taken XV,1 Funkhouser: Philippine Membracidz 21 at Montalban, February 22, 1918, on fruit heads of Solanwm ver- bascifolium and on young shoots of Eugenia calubcob; and one specimen was collected in the forest of Culasi near a 700-meter elevation. The nymphs (Plate I, fig. 4) were all collected at Montalban on February 22, 1918, on Eugenia calubcob and are all in the last instar. Since the immature form of this insect has not been previously recorded, a brief description is given as follows: Brilliant yellow, marked with black; length, 5 millimeters; width, 3; smooth, shining, not punctate, very sparingly pubes- cent, head tuberculate above; pronotum with three strong spines, one on median dorsal line and one on each side; mesonotum with one strong median dorsal spine; abdomen with a strong lateral spine on each segment except the last; wing pads reaching the first abdominal spine. Head much broader than long, bright yellow, smooth, shining, impunctate, only lightly pubescent at margins; base strongly elevated at middle and bearing a strong tubercle on each side of median line; eyes very prominent, produced laterally, jet black; ocelli not visible; sutures distinct; inferior margins of genz almost transverse; clypeus longer than broad, extending for more than half its length below the inferior margins of the genx, tip rounded and somewhat pilose. Thorax distinctly divided into prothorax, mesothorax, and me- tathorax, not punctate, sparingly pubescent, shining, jet black in front, bright yellow on sides and behind spines; humeral angles very prominent, auriculate; pronotum having three strong spines, one on median dorsal line pointing upward and slightly backward, one on each side above humeral angle projecting upward and outward, an extension of dorsum behind median spine extending to the spine of the mesonotum; mesonotum with strong central spine at posterior margin; metanotum without spine, bright yellow above, jet black at sides; wing pads very well developed, reaching backward as far as the first lateral spine of the abdomen. _ Abdomen broad, flattened dorsoventrally, bearing seven dis- tinct segments, of which the first six have strong lateral spines extending directly outward, these spines gradually becoming longer from before backward; entire abdomen shining, impunc- tate, not pubescent, black above, yellow on sides, apex strongly curved upward; anal tube long, upraised, wider at base than at tip. Legs and undersurface of body shining luteus. 30 Philippine Journal of Science 1919 The form as above described is very apparently in the fifth instar and about ready for the final molt. Mr. McGregor’s field notes on Tricentrus fairmairei state that both adults and nymphs are attended by black ants of which he sends specimens. These ants have been determined by Prof. W. M. Wheeler, of Harvard University, as Polyrhachis (Myrm- hopla) dives Smith. I am again indebted to Professor Wheeler for his kindness in making these determinations. Tricentrus acuticornis sp. nov. Plate I, figs. 5 and 6. Near Tricentrus convergens Walker but differing in size and particularly in the formation of the suprahumeral horns. On superficial examination I had placed these specimens with the males of 7. econvergens, which they closely resemble in general appearance; but further study shows them to belong to a dis- tinct species. Small, dark brown with large white tomentose patches behind each eye, closely punctate, densely pubescent; suprahumeral horns acute, slightly diverging, longer than the distance between their bases, not at all flattened; posterior process straight, ex- tending just beyond internal angles of tegmina; tegmina smoky hyaline; hind trochanters armed with strong spines. Head a little broader than long, subquadrangular, black, the fine punctures almost entirely hidden by the dense silvery pu- bescence; base weakly sinuate, rounded, highest above ocelli; eyes very prominent, gray, extending laterad half as far as the humeral angles; ocelli large, pearly, about equidistant from each other and from the eyes and situated above a line drawn through centers of eyes; inferior margins of gene weakly sin- uate, extending slightly downward from eyes to about middle of clypeus; clypeus longer than wide, densely pubescent, ex- tending for somewhat more than half its length below inferior margins of gene. Pronotum black, closely punctate, densely pubescent; meto- pidium perpendicular, broader than high, slightly convex; su- prahumeral horns straight, triquerate, acute, longer than the distance between their bases, extending upward, forward, and slightly outward, tips somewhat recurved; humeral angles prom- inent, auriculate; dorsum almost straight, sloping downward from suprahumeral horns to posterior process; posterior pro- cess short, acute, sharply carinate above, tip triquerate, sharp, extending just beyond internal angles of tegmina; median ca- rina percurrent, faint on metopidium, strong on posterior process. XV,1 Funkhouser: Philippine M embracids 23 Tegmina much wrinkled, smoky hyaline; base coriaceous, black, punctate and pubescent; tip marked with narrow brown spot at costoapical angle; five apical cells. Sides of thorax densely white tomentose; undersurface of tho- rax black; abdomen dark brown; legs ferruginous with bases of femora darker; tarsi ferruginous; claws brown; hind tro- chanters very prominently armed with strong teeth. Length from front of head to tips of tegmina, 5.3 millimeters; width between tips of suprahumeral horns, 3. PANAY, Antique Province, Culasi (McGregor), 3 females. Type, a female, in my collection. Described from three females from Culasi, one taken May 18, 1918, and two taken May 27, 1918. Type and two paratypes in my collection. Sipylus dilatatus Walk. Add: Habitat—Luzon, Manila: Rizal Province, Pasig. Pa- NAY, Antique Province, Culasi. (McGregor.) Hosts.—Phyllanthus reticulatus Poir.; Acalypha sti- pulacea Klotz. , Seven specimens collected as follows: One male and one fe- male taken at Manila in April, 1917; one male and two females taken at Pasig, February 17, 1918, on Phyllanthus reticulatus; two females taken at-Culasi, May 27, 1918, on shoots from a stub of Acalypha stipulacea. Mr. McGregor’s field notes on this species state that the insects were solitary and were not attended by ants. With the specimens of this species now available for study, it appears that the males average considerably smaller and darker than the females but show the same characteristic broadened pronotum and the nodulate veins on the tegmina. Gargara varicolor Stal. Add: Habitat.—LuzoNn, Manila: Rizal Province, Montalban (McGregor). Hosts.—Acalypha wilkesiana Muell.-Arg.; Solanum verbascifolium L. Attendant ants.—Polyrhachis (Myrmhopla) dives A fine series of this species, of which three specimens were collected at Manila, February 14, 1917; twenty-three at Manila, April 5, 1917, on Acalypha, wilkesiana; and ten at Montalban, February 22, 1918, on fruit heads of Solanum verbascifolium Linn. * PA Philippine Journal of Science 1919 This species, according to Mr. McGregor’s notes, is gregar- ious and is attended by black ants of which a number have been received. Professor Wheeler determines these ants as the same species [Polyrhachis (Myrmhopla) dives Smith] as those found with Tricentrus fairmairei Stal. Gargara nitidipennis Funkh. Add: Habitat.—Luzon, Laguna Province, Paete. PANAY, Antique Province, Culasi. (McGregor.) One male and eight females taken at Paete on March 29, 1917, and one female taken at Culasi on May 29, 1918. This species shows more variation in size than any other spe- cies from the Islands, but I can find no specific differences be- tween the large and the small forms. Gargara nigrofasciata Stal. Add: Habitat—Luzon, Laguna Province, Paete (McGregor). Three males and one female from Paete. It has previously been recorded * that this species also shows considerable varia- tion. The specimens from Paete are larger and darker than the type described by Stal, but agree with specimens from Minda- nao collected by Professor Baker.’ Gargara tuberculata Funkh. Add: Habitat—LuZzon, Manila (McGregor). One female collected at Manila on April 5, 1918. Gargara pygmaea Walk. Add: Habitat—Panay, Antique Province, Culasi (Mce- Gregor). Host.—Acalypha stipulacea Klotz. . One female taken at Culasi, May 27, 1918, on shoots from a stub of Acalypha stipulacea. Gargara rugonervosa Funkh. Add: Habitat—PANAy, Antique Province, Culasi (Mce- Gregor). Host.—Croton leiophyllus Muell.; Acalypha stipula- cea Klotz. One male taken May 18, 1918, on Croton leiophyllus; and an- other male taken May 27, 1918, on Acalypha stipulacea. Both specimens are from Culasi. * Philip. Journ. Sci. § D 10 (1915) 398. "Ibid. 13 (1918) 82. XV,1 Funkhouser: Philippine Membracidz 25 Gargara pulchripennis Stal. Add: Habitat—PANAY, Antique Province, Culasi (Mc- Gregor). Host.—Acalypha stipulacea Klotz. One male collected May 27, 1918, on shoots from a stub of Acalypha stipulacea. Gargara grisea sp. nov. Very near G. tuberculata and may be only a variety of that species. It may be at once recognized by the very dense gray pubescence which is quite distinctive. Large, heavy-bodied, light brown, but so covered with dense grayish white pubescence that the body color and the punctation are almost entirely hidden; metopidium convex; posterior process just reaching internal angles of tegmina; tegmina translucent, base and veins densely pubescent with grayish white hairs, veins strongly tuberculate. Head broader than long, densely tomentose with white pubes- cence; base regularly sinuate, highest above ocelli; eyes large, prominent, grayish brown; ocelli large but not prominent on ac- count of the dense pubescence, glassy, about equidistant from each other and from the eyes and situated slightly above a line drawn through centers of eyes; inferior margins of genz almost straight; clypeus longer than wide, densely tomentose with white pubescence, extending for more than half its length below in- ferior margins of genz. Pronotum densely pubescent with grayish white hairs which are thicker near base of head and around humeral angles; me- topidium strongly convex; humeral angles large and auriculate; dorsum arcuate; median carina strongly percurrent; posterior process short, heavy, blunt, sharply carinate above, tip just reaching internal angles of tegmina. Tegmina translucent, not clearly hyaline; base coriaceous and densely pubescent; veins pubescent and strongly tuberculate; teg- mina with five apical areas; hind wings with three apical areas. Sides and undersurface of thorax densely white tomentose; abdomen brown, densely pubescent with white below, segments bordered with white above. Femora dark brown, smooth; tibie ferruginous and spined; tarsi ferruginous; claws brown; hind trochanters unarmed. Length, 4 millimeters; width between humeral angles, 2.2. LUZON, Manila (McGregor), a pair. Type, a female. 26 | Philippine Journal of Science 1919 Described from a pair collected at Manila, the female on Feb- ruary 14, 1917, and the male on April 5, 1917. The male is slightly smaller than the female but shows the same characteristic grayish white pubescence. Type and allotype in my collection. Cryptaspidia tagalica Stal. Add: Habitat.—PANAY, Antique Province, Culasi eats ced One female taken June 1, 1918. Cryptaspidia elevata sp. nov. Plate I, fig. 7. Recognized by the thick, elevated posterior process and the auriculate humeral angles. Black, punctate, hairy ; no suprahumeral horns; humeral angles projecting as broad rounded lobes; scutellum entirely concealed; _ posterior process suddenly turned upward at tip; tegmina marked with black. Head wider than long, black, densely covered with long yel- lowish hairs which entirely conceal the punctation; base weakly sinuate; eyes large, brown; ocelli prominent, pearly, a little farther from each other than from the eyes and situated slightly above a line passing through centers of eyes; clypeus wider than long, oval, extending only a little below the rounded outline of the inferior margins of the genx, tip broadly rounded and hairy. Pronotum black, punctate, thickly covered with long, yellow- ish hairs; metopidium very convex, rounded, wider than high, slightly flanged at inferior margin; humeral angles produced into broad earlike lobes rounded at tips; suprahumeral horns absent; posterior process short, heavy, much depressed at base, suddenly elevated at apex to form a heavy flattened spine, tip suddenly acute, flattened laterally and having carinz above, below, and in the center of each side and reaching just beyond the internal angles of tegmina. Tegmina closely impinging on lateral margins of pronotum; smoky hyaline, base and basal fourth of costal margin black and punctate; faint brown band across center of tegmen below in- ternal angle, most distinct at costal edge. Undersurface of body black and very hairy. -Posterior tro- chanters unarmed. Legs uniform brown; tibize armed with long, thin, spinous hairs, distal ends with black spines. Length, 7 millimeters; width between tips of humeral angles, 3.8. Luzon, Benguet Subprovince, Baguio (Baker). Type, a female. . V2 Funkhouser: Philippine Membracidz 27 Cryptaspidia longa sp. nov. Plate I, fig. 8. Very long-bodied, narrow, black, closely and coarsely punctate, sparingly pubescent; posterior process not reaching internal angles of tegmina; tegmina smoky hyaline with very large black veins; scutellum entirely hidden; trochanters unarmed. Head wider than long, subtriangular, black, weakly punctate, pilose, convex, median suture deeply depressed; base almost straight; eyes large, prominent, black with brown inner borders; ocelli prominent, translucent, yellowish, situated on broad eleva- tions, about equidistant from each other and from the eyes and situated about on a line drawn through centers of eyes; inferior margins of gene arcuate; clypeus slightly longer than broad, much deflexed, black, punctate, pilose with long golden hairs, tip rounded. Pronotum entirely black, coarsely punctate, sparingly pilose with long golden hairs; median carina percurrent but weak; humeral angles not prominent, triangular; metopidium strongly convex; dorsum nearly straight; scutellum entirely concealed but present; posterior process short, straight, somewhat flat- . tened, acute, tip sharp and not reaching the internal angles of : the tegmina. Tegmina very long, smoky hyaline, wrinkled; base coriaceous, black, punctate and pubescent; veins very strong, heavy, black, sparingly pilose; tips of tegmina extending well beyond apex of abdomen. Sides and undersurface of thorax black, densely pilose with “long golden hairs; abdomen very long, entirely black, densely pilose below; ovipositor smooth, black, shining, sparingly pubes- cent; legs ferruginous, tibisee tinged with darker, tarsi and claws brown, hind trochanters without teeth. Length, 8.2 millimeters; width, 3. LUZON, Benguet Subprovince, Baguio (Baker). Type, a female. Fa. DAAMNRwONr ILLUSTRATIONS — PLATE I . Centrochares horrificus Westwood, nymph. . Pyrgonota semperi Stal, lateral outline. . Pyrgonota semperi Stal, front view of apex of pronotal horn. Tricentrus fairmairei Stal, nymph. . Tricentrus acuticornis sp. nov., lateral outline. . Tricentrus acuticornis sp. nov., frontal outline. Cryptaspidia elevata sp. nov., lateral outline. . Cryptaspidia longa sp. nov., lateral outline. FUNKHOUSER: PHILIPPINE MEMBRACID. ] [PuHurp. Journ. Scr., XV, No. $ WS 7 . . a 2 PLATE I. PHILIPPINE MEMBRACIDA. OSTEOLOGICAL AND OTHER NOTES ON THE MONKEY- EATING EAGLE OF THE PHILIPPINES, PITHECO- PHAGA JEFFERYI GRANT By R. W. SHUFELDT Major, Medical Corps, United States Army, Washington, D. C. ELEVEN PLATES Some time during the latter part of March, 1918, the captive specimen of the monkey-eating eagle of the Philippines (Pithe- cophaga jefferyit Grant), of the botanic garden of Manila, died, and the dead bird was immediately sent to the Bureau of Science, Manila. In his letter to me of April 1, 1918, Mr. McGregor stated that from this material he— got a good skin—the first! in our collection, although we have a poor mounted specimen. The bones are being cleaned. They will lack the feet, the skull, and the ends of the wings. However, I shall supply a head and a foot from another specimen, so that the thing is nearly complete. The description of this should make a good paper, especially if you can lay hands on bones of the harpy eagle and other related genera. All this material was received by me in the summer of 1918; and it was in excellent condition. I prepared and photographed the specimens myself. Later, the division of birds of the United States National Museum kindly loaned me certain skeletons, with which I intended to compare the bones of Pithecophaga jefferyi. These consisted of an imperfect skull of Morphnus guianensis (Daudin), adult (No. 18468) ; a complete skeleton of an adult Thallasoaétus pelagicus (Pallas), collected by Paul L. Jouy in Korea (No. 18222) ; an Aquila chrysaétos canadensis (Gmelin), female,>adult, complete and disarticulated (No. 18802); a com- plete, disarticulated skeleton of Gypaétus barbatus (Linnzus) (No. 17834) ; a complete skeleton, disarticulated, of Halixetus leucocephalus (Linnzeus), adult (No. 19278); and, finally, an incomplete skeleton of the harpy eagle, Thrasaétos harpyja (Lin- nzeus), adult (No. 225806). The last-named specimen lacked the skull. This bird was, during life, for several years one of the ornaments of the bird collection of the National Zodlogical Park at Washington. The skull was probably left in the skin. Sub- sequently I endeavored to obtain a skull of this species from the 31 32 Philippine Journal of Science 1919 Academy of Natural Sciences of Philadelphia, but was unable to do so. It is very much to be regretted that a skull of the harpy eagle is not at hand for comparison with that part of the skeleton of Pithecophaga. Probably there is not a cleaned skull of the harpy eagle in the United States—adult, subadult, or young. This is invariably the case with respect to any bird before a glut takes place in museums and private collections of birds; the skulls are left in the skins! THE CRANIUM, MANDIBLE, AND ASSOCIATED BONES OF PITHECOPHAGA It matters but little from what point of view we may regard the cranium of this great eagle, the fact will at once be ap- preciated that at least two of its characters are so pronounced and so characteristic of the species that to confuse this part of the skeleton with the cranium of any other known eagle would be quite out of the question. In the first place, the lacrymal bones and their accessory outer pieces are of great size, while a still more conspicuous character is to be seen in the size and form of the osseous superior mandible; this is powerfully hooked, with cultrate tomia. It is transversely compressed almost to the last degree, the transverse diameter at the base beneath being slightly less than 2.5 centimeters, while the vertical height at the same point measures nearly 4 centimeters. Measured me- sially below, it has a length of about 5.02 centimeters. Hither narial aperture is very large, and these openings do not com- municate through and through; anteriorly either one merges into the general surface of the side of the beak, while posteriorly greater depth is present, and the surrounding margin is sharp and triangular in outline; the apex of the triangle is posterior and above (Plate I, fig. 1). Turning to the large lacrymals mentioned above, their form and proportions are well shown in the illustration (Plate I, fig. 2), while the descending limb is portrayed on the same plate. As is the case with so many large raptorial birds, these lacrymal bones, in adult life, do not fuse with the frontal and nasal upon either side nor with the large, subtriangular accessory piece supported at the outer end of each (fig. 2). The descending orbital limb of a lacrymal is constricted superiorly, anteropos- teriorly flattened below, while it is broader and concaved in front from above downward. This part of a lacrymal does not come in contact either with the small, thin pars plana within or with the zygoma below. iste ss ielet ait ee eT ee ee ee ee os Fh XV, 1 Shufeldt: The Monkey-eating Eagle 33 Passing again to the dorsal aspect of the cranium, it will be observed, in the craniofacial region, that a considerable depres- sion or concavity is present in the median area; posterior to this the frontal region is broad and flat, with the superior orbital margin cultrate. As in most eagles, the broad cranial vault is rounded and smoeth, being marked only by the wonderfully fine, rambling, anastomosing lines that in life harbor the vessels of that region. I find these very well marked in the cranium of a white-headed eagle (Halixetus leucocephalus) (Plate II, fig. 1) where, likewise, the depression in the craniofrontal region is pronounced. Throughout the diurnal Raptores, this character varies to some degree for the species. It is fairly well marked in the cranium of Morphnus guianensis; and several of the cra- nial characters thus far described in the cranium in Pitheco- phaga are more or less in agreement with the corresponding ones in this bird (Plate HI, fig. 3). This in no way takes size into consideration, for Morphnus is not more than half the size of the monkey-eating eagle while, in a way, it does refer to the marked craniofacial depression, the lacrymals, and the general form of the osseous superior mandible. No other parts whatever have been seen or examined by me, and even the remainder of this cranium is very imperfect. When the material was selected for me at the United States National Museum, in addition to those designated by myself, this imperfect cranium of Morphnus was included. In Gypaétus barbatus a lacrymal does not support a free distal piece as in some eagles, while the superior outstanding portion is to a degree elongate and to some extent twisted upon itself. This bird has considerably more eagle in its skull, and in the rest of its skeleton for that matter, than it has vulture. How- ever this may be, it has no special affinity with Pithecophaga, in so far as its skeleton would seem to indicate; and it is very much to be doubted that any of the rest of its morphology would point to any such relationship. Nothing would be gained, there- fore, by a comparison of the skeletal characters of these two birds, beyond a demonstration of the fact that they possess no near relationship; and this much is evident to any expert avian osteologist at a glance. This in no way applies, however, to such a form as the big eagle of Korea, Thallasoaétus pelagicus, a complete disarticulated skeleton of which (No. 18222, United States National Museum) I have before me. Paul L. Jouy col- lected it in that country many years ago. In its cranium it exhibits all the aquiline characters of an average eagle, including 166316——3 84 Philippine Journal of Science 1919 the accessor pieces on the lacrymals. Upon summing up its cranial characters, however, it is very evident that it is not as nearly related to Pithecophaga as is the golden eagle (No. 18802, United States National Museum) (Plate II, fig. 2). All eagles thus far examined by me possess, upon lateral aspect of the cranium, a deep, rounded, separated crotaphyte fossa; a large, squarish postfrontal process; and a squamosal projection. This does not show very well in fig. 1 of the present memoir, due to faulty lighting of the skull; nor do we, in the same figure, gain any idea of the enormous capacity and depth of the orbit that exists in the cranium of our subject. Indeed, it measures as much as 3 centimeters from the lower edge of the postfrontal process to the edge of the subcircular vacuity in the interorbital septum in Pithecophaga. This vacuity is also to be seen in the cranium of the golden eagle, but not in the white- headed species of North America. The anterior wall of the brain case in our subject is markedly concaved on its anterior aspect, and nonperforated beyond the usual foramina for vessels and nerves, all of which latter are comparatively small and individualized. So far as I have examined their crania, this is more or less true of all eagles. As will be observed in Plate I, fig. 1, the quadratojugal bar in this monkey-eating eagle is quite straight, and of nearly uniform caliber from one end to the othér. It is so in all true aquiline species, and in some the maxillojugal suture persists throughout the life of the individual. A quadrate is a big, stout bone, with a well-developed orbital process, and a broad, smooth surface on its outer aspect. A broad, longitudinal valley lies between the two thoroughly se- parated articular facets that occur on its lower mandibular por- tion. It presents two heads for articulation with the skull, and they are separated by a wide, smooth notch. Nearly all parts of this cranium are pneumatic, and to this the quadrate forms no exception. Hither pterygoid bone is small, straight, short, and slender. When duly articulated, it does not come in contact with its fellow of the opposite side. In the golden eagle the outer margin of a pterygoid is very sharp and thin. This aspect in the white- headed eagle is longitudinally grooved, and the posterior extrem- ity of the bone possesses an upturned process for additional articulation with the quadrate. All true eagles possess a vomer, which is a long, thin plate XV,1 Shufeldt: The Monkey-eating Eagle 35 of bone compressed from side to side, with a finely pointed, free anterior apex; and, further, it is ununited posteriorly with the palatines. Pithecophaga possesses large, spongy maxillopalatines that fill up a good part of the rhinal chamber, being carried far to the front above the prepalatines, where they appear to merge to a Slight degree. They are separated for their entire length in Aquila, but extensively fuse in Halixetus leucocephalus, where they form a nearly solid roof in the mouth in connection with prepalatines and maxillaries. Pithecophaga has a very solid bony nasal septum, as has the white-headed eagle; while in the golden eagle a good-sized va- cuity may occur at the posterior superior angle. The hinder part of a palatine in any true eagle has the postero- external angle rounded off, the bone in this locality being rather broad, with its inner margin produced and turned down, and its outer one to some degree thickened. In this part of their extent the palatines are in contact with each other, clasping the - . vomer between them anteriorly and the sphenoidal rostrum , above. : The basipterygoidal processes are entirely absent in the skulls 4 of all the eagles at hand at this writing, save in the case of the . golden eagles, where they are represented by small triangular prickles, flattened from above downward. A lip of bone usual- ly underlaps the entrances to the carotid arteries. Pithecophaga—and other species of eagles depart but little from it—presents a triangular area for the basitemporal space of the cranium, with a marked depression just anterior to the hemispherical occipital condyle. The foramen magnum is large and nearly circular in outline. A superoccipital prominence is fairly well developed in all these eagles, including the Korean species. Although the “occipital area” is well defined, the ridge that _ exists to mark its limitations is not much raised. This is also true of the “crotaphyte fosse,” which are extremely shallow, subcircular in outline, and widely separated posteriorly. At the side of the cranium the crotaphyte fossa extends over onto the external surface of the postfrontal apophysis. The cranial capacity is only of moderate proportions, when taken in conection with the balance of that part of the skull; and this is especially true of the cranium of Pithecophaga. Laterally, the osseous aural apertures are much exposed; in life, however, the soft structures of the internal ear are more or ~ 36 Philippine Journal of Science 1919 less protected by the conformation of the quadrate when that element, upon either side, is placed in articulation. There are some twelve to fifteen sclerotal plates to the circlet in either eyeball; the smaller ones are situated anteriorly and the larger ones behind (Plate III, fig. 9). Each platelet is an irregular quadrilateral in outline, the whole arrangement being fitted to the requirements of the form of the external moiety of the globe, assisting very materially in protecting it from certain injuries, and maintaining the sphericity of the organ. Addi- tional strength is afforded through the generous overlapping of the plates. We find the characters presented on the part of the mandible among the eagles pretty much the same, and it is V-shaped in all of the species at hand. In the long-faced species the V is more elongate, and the angle formed by the rami more acute. The bone is more or less pneumatic, which is also true of the cranium as a whole. In Pithecophaga a side of the mandible . has a length of 9.7 centimeters, and the symphysis 2.5 centi- meters, the latter being concave transversely above and corres- pondingly convex on its ventral aspect. All the elements are very solidly fused together, with the superior and inferior borders rounded. There are no perforating vacuities anywhere, the entire bone being very strong and smooth. A low coracoid pro- cess is present on either side, and from either of them the superior border slopes rather abruptly downward to the articular enlargement of the same side (Plate I, fig. 1; Plate III, fig. 4). Posteriorly, an articular extremity is considerably compressed from above downward, being truncate posteriorly. Its mesial, inturned process is somewhat elongate and well developed, with the usual pneumatic foramina near its apex superiorly. These processes are directed upward and inward, the apex in either case being blunt. Each presents two surfaces ventrally, and these are smooth. One surface is continuous with that of the jaw, the other pertaining to the underside of the inturned process. Superiorly either one of these articular ends is concaved cen- trally, and that surface does not come in contact with the quadrate of the same side in articulation. Upon either side of this con- cavity, however, internally and externally situated, there is to be noted a conspicuous articular facet for articulation with the corresponding ones on the quadrate of the same side. In many publications, here and abroad, I have described the . skeletons of various species of eagles, falcons, hawks, and their ~ XV,1 Shufeldt: The Monkey-eating Eagle 37 near congeners, both fossil and existing forms, and it may be said that we meet with few marked differences in any of them with respect to the characters presented on the part of the hyoidean arches, or skeleton of the lingual apparatus (Plate III, fig. 10). The anterior soft part of the glossohyal is long and narrow, being covered for its anterior two-thirds with the usual horny sheath. Posterior to this, the glossohyal is narrow and bifur- cated, articulating, as usual, with the basihyal posteriorly. This part is likewise thinly overlain with a horny sheath. All this part is very feebly developed, extremely elongate, and narrow. : Triangular in outline, the basihyal supports posteriorly an elongate urohyal, which is of small and nearly uniform caliber; it is tipped off with a very small bit of cartilage behind. This urohyal is perpendicular to the transverse line of the base of the basihyal, and in the right angle upon either side of the former articulates the head of a hypobranchial. Each hypo- branchial is long and curved upward for its entire length. Both the anterior and the distal ends are somewhat enlarged, the posterior enlargements being continued in cartilage for a few millimeters, when, upon either side, its place is taken by a cera- tobranchial. Either one of these is about 1.5 centimeters long, slender, and to some extent curved in line with the upcurved hypobranchial of the same side. Each ceratobranchial termi- nates behind in a fine, needle-pointed, cartilaginous tip. Our white-headed eagle (Halixetus leucocephalus) possesses a hyoid very similar to this; but the angle made by the urohyal and the basihyal is an obtuse one instead of a right angle, as it is in Pithecophoga, and the cartilaginous part of the glossohyal is extremely short, while the broad, osseous part has the form of a capital letter H, with the sides converging toward each other from behind forward. Eagles, in so far as I have examined them, possess a very simple form of larynx and trachea. The former presents the ‘usual osseous elements, but they form no special articulations with each other, the contour of the structure being sustained in their membrane, with the aforesaid elements simply maintained in their several positions by it, and nowhere in contact with. each other. All the tracheal rings—and they are very numerous —appear to be, to some extent, performed in cartilage (Plate III, figs. 6,10). Each is of the usual ornithic type, the broader ones being above and below. There are upward of seventy of 38 Philippine Journal of Science 1919 them, the lower end of the tube not having more than half the caliber of the upper, where it makes a simple union with the larynx. Distally, the bronchial branches are small, and but half closed in, the mesial aspect of either being a thin, simple * . membrane stretched across. There is a single pessalus present. THE SKELETON OF THE TRUNK The vertebral column.—aAll the skeletons of eagles that I have examined possess fourteen vertebre in the cervical region of the spine, between the cranium and the first true dorsal vertebra. Each of the last two cervicals—the thirteen and the fourteenth —supports a free pair of ribs. On the thirteenth the pair is rudimentary, to the extent that the body of the bone on either side is lacking to some considerable degree, while the articulation is perfect for the head and angle. With respect to the pair of ribs on the fourteenth cervical vertebra, it has a length of some 5 centimeters, terminating in a pointed, free extremity well above the sternum. As this is the case with respect to these last two vertebree in Pithecophaga, Thrasaétos harpyja, Aquila, and our white-headed eagle, it is probably what we will find in the skeleton of any true aquiline species, irrespective of the part of the world it inhabits. As a matter of fact, the vertebral column of this monkey-eating eagle of the Philippines is, character for character, almost the counterpart of that series of bones in the harpy eagle, and departs but very little from what we find in other species. In our present subject the atlas lacks both neural and hzemal spines, while the articular cup for the occipital condyle is notched above. Laterally, there may be a notch or a foramen for the passage of the vertebral artery upon either side. Its neural arch is rather broad, and the facet for the centrum of the axis is quadrilateral in outline, the width being twice that of the vertical height. At the middle point below, it is pierced longitudinally by a minute foramen; a broad notch being found at the same place in the atlas of the white-headed eagle. In this species, too, similar notches allow, in life, the passage of the vertebral arteries to the cranial cavity. Similar ones, though larger, are to be seen in the axis of Pithecophaga, in which vertebra are developed a stumpy hzmal and a neural spine, as well as a more or Jess insignificant odontoid process. The neural canal throughout the cervical series of vertebra is eylindrical in form and quite uniform in caliber. From the third to the twelfth vertebra, inclusive, the lateral foramina for Tee * XV, 1 Shufeldt: The Monkey-eating Eagle 89 the vertebral arteries are, in all instances, entire—that is, with respect to their osseous walls; on the other hand, the carotid canal is an open passage for the arteries of that name in the fifth to the ninth cervical vertebra, inclusive (Plate III, figs. 3, 5, 7; and 8). All of these cervical vertebrz are very large and strong— indeed, quite massive in character. When present, the neural spine is situated just within the posterior margin of the bone; it is directed backward and upward in the third vertebra; is vertical and peglike in the next following one, with broader and triangular base in the next three (fifth to seventh inclusive), wherein it moves forward to the middle of the neural arch, the angle being situated anteriorly, and either side being directed backward and outward. In the third cervical the prezygapophyses and postzygapophy- ses are joined by a plate of bone, in which appears, on either side, a small elliptical foramen; this foramen in the next following vertebra becomes an extensive subelliptical notch, being reduced to a minute spine on either side in the fifth vertebra. Returning to the matter of the neural spines, we find that the broad, triangular form they assume—described in a previous paragraph—persists in the eighth to the twelfth vertebra, in- clusive. Here they are more massive and occupy an extreme posterior position on* the several remaining vertebre of this series. On the twelfth the spine begins to assume the form of the neural spine as we find it in the leading dorsals, while in the thirteenth and the fourteenth not only are the neural spines in agreement with those processes in the thoracic vertebre, but they present almost all the other characters of that series. In the midcervical series, the pleurapophyses are short and stumpy. The rather massive prezygapophysial processes in the fifth cervical face directly upward; in the sixth they look inward and backward, and they maintain this position to include the ninth. : In the rest of the series they face inward again. The “carotid canal” is present and open in the fourth to the eighth cervical, inclusive, being most nearly closed in the last one named. Coming to the dorsal vertebre (Plate IV), we find them very closely interlocked in articulation, with long, spinelike me- tapophyses on the last four of the five which occur in this section of the spinal column. The hemal spines are stumpy and short, being entirely absent on the last two dorsals. The neural canal is cylindrical in form; the facets for each pair of ribs are entire ‘ 40 Philippine Journal of Science 1919 and are situated well back from the anterior limitation of the centrum on any particular vertebra; that is to say, there are no demifacets (Plate VI). There are no epipleural appendages on either the last pair of the cervical ribs or the last pair of the pelvic ones (Plate IV) ; and, while of no great size on the first pair of dorsal ribs and the leading pair of pelvic ribs, they are conspicuously long and massive on the ribs of all pairs constituting the midseries. Any of these, in articulation, overlaps the body of the next rib behind; and all of these processes, at least in adult life, are very extensively and firmly codssified to the rib to which in any case they belong. With respect to direction, they all point upward and backward. In the harpy eagle the basic portion of any one of these epi- pleural appendages is notably extensive, and occupies, in the case of the second pair to include the fifth, a large part of the posterior border of the rib—fully a third at least. There is a smaller pair of these apophyses on the last pair of pelvic ribs (Plate V). All the ribs in Pithecophaga, save the cervical ones, articulate with the sternum by means of costal, or sternal, ribs; and there are no floating ones on the last pair as there are in the case of the harpy eagle (fig. 17). In our subject these sternal ribs increase in due proportion, in size and length, as we proceed from the first to the last pair; they are more or less massive, in keeping with the rest of the skeleton of this ponderous bird, and the last two pairs exhibit more or less upward curvature (Plate IV). Our white-headed eagle has the last pair of pelvic ribs more or less feebly developed. The thoracic pair, upon one side or the other, may be more or less aborted with respect to its length and not descend to meet the usually well-developed corresponding pair of sternal ribs in this bird. The pelvis and the caudal vertebre.—When we come to ex- amine and compare the pelves of various species of eagles from different parts of the world, we are struck by the marked sim- ilarity of form and of characters among them. This applies with special significance to the pelvis as we find it in our pres- ent subject and in the harpy eagle. Here the different char- acters are of the most trivial description possible—so much so, indeed, that a detailed account of the pelvis of an adult Pithe- cophaga jefferyi would answer admirably for the same bone of the skeleton in the harpy. Upon comparing Plates IV and V of the present memoir, it would appear that the disposition of XV, 1 Shufeldt: The Monkey-eating Eagle 4] the hinder part of the pubic style was very different in the two birds; but this is by no means the case, for in the harpy those elements are held in their normal position, as in life, by a lig- ament stretching between their distal ends; while in Pithe- cophaga, when this ligament has been cut, it allows the pubic styles to spring away from each other and hang down as shown on Plate IV. Moreover, the point of view from which I pho- tographed those two trunk skeletons was not quite the same, and this causes the reproduction of the rest of the bone to pre- sent slight differences, which do not really exist. Viewed from above, it is to be noted that the ilia project considerably beyond the sacral crest; and on the upper surface of their anterior border there is a raised emargination which is produced backward and finally runs out as a bounding line to the postacetabular area of the superior surface of the sacrum upon either side. As the ilia pass the ‘“‘sacral crest,” their margins thoroughly coédssify with it and in the same plane an- teriorly—that is, up to the point where these bones begin to diverge and are raised above the general surface on this dorsal aspect of the bone. At the angle where this divergence com- mences, the sacrum and the ilia are completely fused, and every semblance of posterior openings of the “ilioneural canals’’ is completely obliterated. So, too, with the rather abruptly down- ward-sloping “postacetabular area;’’ here, likewise, every sem- blance of sutural traces—the intersacroiliac ones—has been absorbed; while the intervertebral foramina, so’ conspicuous in this area in the pelves of some birds, are reduced to mere little pits in the general surface of the bone (Plate VI). Seen upon lateral aspect (Plate IV), the anterior two-thirds of bone—or all that part anterior to and above the acetabulum and antitrochanter—is supplied by the ilium of that side. Its surface is generally concaved and faces upward and outward. Anteriorly, its outer border is emarginated, and below this may be seen the forepart of the pelvic sacrum, the continuation of which, posteriorly, may also be observed through the large, cir- cular acetabulum and immense ischiadic foramen; the former is entirely lacking in any osseous base, while the latter occupies fully one-half of the lateral area posterior to the rather large, subtriangular antitrochanter and elliptical obturator foramen. Posterior to this great ischiadic foramen, the lateral surface of the ischium is triangular in outline, concave above, convex below, and smooth throughout. The rounded posterior ischiac border is nearly straight and presents no semblance of any indent that might suggest the presence of an “ilioischiadic.” » 42. Philippine Journal of Science 1919 The pubic style closes the obturator foramen; and, after pass- ing it along the lower margin of the ischium, it becomes much attenuated and terminates as a fine point in a line below the middle of the ischiadic foramen above it. A small interval is then present, when the bone begins again in a fine point, to increase gradually in size as it passes below the inferior ischiac border, to be produced posteriorly in a much thickened condition in the direction of its fellow of the opposite side. Ventrally, the pelvis of this eagle presents many interesting features. As well as I am able to judge from the pelvis of an adult bird of this species, the sacrum ‘would seem to contain fifteen vertebrxe, or the same number as we find within the grasp of the iliac bones of the harpy’s pelvis. The leading seven possess very large centra, and they have their lateral pro- cesses extending upward and outward, to codssify with the ventral surfaces of the ilia. Posterior to this arrangement, we arrive at the very deep pelvic basin, where the three vertebre opposite the cotyloid rings fail to throw out lateral processes. These are followed by three others that have the lateral processes for their external moieties fused into a common plate of bone upon either side, which fuses outwardly with the inner surface of the pelvic wali. A row of three elliptical foramina is left upon either side of the centra here, and osseous trabecule are thrown up into the deep space above. Finally, the two terminal sacral vertebrz occupy a much lower plane than the preceding— that is, their transverse processes do—and these are thrown directly outward, to fuse distally with the inner wall of the ischium upon either side. Below them the surface—and it is an extensive one—on either hand is smooth, being furnished entirely by an ischium. The conformation I have attempted to describe here gives rise to four fairly well-defined cavities, each imperfectly walled in by the surrounding parts of the pelvis as a whole. Through these cavities, mesially, passes the big, codsified pelvic “sacrum.” The first cavity occupies the anterior half of the pelvis back- ward to a point where the vertebre cease to send their lateral processes directly outward to the iliac walls upon either hand. Then follows the well-marked, deep cavity opposite the aceta- bule; posterior to this we define cavity number three, which lies between the big ischiadic foramina; and, finally, ventrad to the last two, there is the general concavity of the pelvic basin, having cavities two and three above it, and its lateral walls formed by the ischium, descending deeply upon either side. XV, 1 Shufeldt: The Monkey-eating Eagle 43 The skeleton of the tail of this eagle is composed of eight free vertebre, plus a large, quadrilateral pygostyle. The first three vertebre are pneumatic, while the rest of this caudal series does not enjoy this condition. Passing to the harpy eagle, we find that the tail skeleton contains but seven free vertebree and a pygostyle; the latter is similar to that of the monkey- eating species, but here the leading four vertebre are pneu- matic—the rest are nonpneumatic (Plate III, fig. 1, and Plate V). These vertebre, including the pygostyle, are massive in struc- ture, and present the usual ornithic characters of this pas of the vertebral skeleton in the aquiline types. Pithecophaga possesses a very large quadrilateral pygostyle, the lower portion of which seems to be a codssified terminal caudal vertebra, though now an inherent part of the bone and included in giving to it its peculiar shape. Superiorly and an- teriorly the pygostyle possesses sharp edges, while the posterior one is thickened and rounded. Above its articulation in front, there is a small opening for the terminal of the spinal cord to enter, and below this, a double-faced articulation for the last free caudal vertebra. Just posterior to this the pygostyle is perforated from side to side by a smooth, subcircular foramen of some size (Plate III, fig. 2). The lower half, posteriorly, has the form of an isosceles triangle, with the apex above. Its sides are rounded, and its base is much thickened. Interiorly it is excavated, and the above-described foramen passes through the excavation, while the floor of the excavation exhibits a larger foraminal opening of an elliptical outline, with the major axis in the anteroposterior line. In the harpy eagle a stumpy apophysis projects from the superoposterior angle of the pygostyle, while the foraminal open- ings are generally smaller. Throughout the falconine types of the world the pygostyle, while it may vary somewhat in form, possesses the same general characters, is always conspicuously large, and is commonly of a quadrilateral form. _ Bones of the shoulder girdle, or pectoral arch.—Taking into consideration the marvelous power of flight possessed by this big eagle, we need experience no surprise at finding the bones composing this distinctive arch as in all birds of great strength, massive in structure (Plate IV). This is preéminently true, and probably no existing bird of its size possesses bigger and stronger bones in its shoulder girdle than an average eagle; to this statement our aquiline giant of the Philippines forms no ex- ception. As is the case with all the bones of the trunk skeleton, 44 Philippine Journal of Science 1919 they are highly pneumatic in character, and consequently very light in weight for their size. The os furculum, or fourchette, has the form of a very wide- spreading capital U; the lower part of the arch is not especially strong, while it rapidly increases in size from below upward, to terminate in immense, free clavicular extremities. Hither of these is much compressed from side to side, with its great superior end accurately molded, externally, to the mesial head of the scapula of the same side. When duly articulated, its apex barely comes in contact with the anteromesial angle of the head ‘of the scapula of the same side. The clavicular limbs of the furculum are greatly compressed from side to side, and the edges thus formed are sharp, espe- cially the mesial ones. ‘There is but small evidence of a hypo- cleidium, and the entire bone is powerfully curved backward and upward. Its pneumatic foramina are principally situated in the concavities of the upper clavicular extremities, upon their moderately shallow outer concavities, on that part of the bone, upon either side, which goes to complete the foraminal passage among the heads of the three bones of this girdle. The os furculum in the skeleton of our subject agrees very well with the corresponding bone in the harpy eagle; though in the latter species the free clavicular ends are more extensively rounded off than they are in the case of the Philippine bird. Our white-headed eagle possesses a more delicately fashioned furculum, while in such a form as Thallasoaétus pelagicus, of Korea, the clavicular free ends are enormously developed; but, owing to the high degree of pneumaticity, the bone, as a whole, is extremely light in weight. Doubtless other eagles present still other differences, not only in the furculum, but also in the remaining bones of the girdle. When articulated as in life, the coracoids in Pithecophaga do not meet in the median line, being separated by an interval of several millimeters. Either bone is very massive in form; but is light in weight, owing to the high degree of pneumaticity it enjoys. All of its parts are conspicuously developed, the broad scapular process being perforated by a foramen piercing it from before backward. There may be pneumatic foramina of some size on the mesial aspect of the big head of the bone and still others on its outer aspect. A sternal extremity is considerably expanded, the bone in articulation extending beyond the sternum laterally. Here it is much compressed from before backward, and thickened at its mesial angle. XV,1 Shufeldt: The Monkey-eating Eagle 45 The harpy eagle possesses coracoids very similar to those of Pithecophaga, while in the American white-headed species the head of the coracoid is much compressed from side to side. In agreement with the other bones of its pectoral arch, Pithe- cophaga possesses a scapula of great size and unusual strength. As in the case of the coracoid and the furculum, it is highly pneumatic, the foramina for the admission of air into its interior being found on the ventral aspect close to the head of the bone. The head is broad from side to side and somewhat compressed from above downward; at its outer angle there is an elliptical articular surface that constitutes about one-third of the glenoid cavity, while at the mesial angle a big quadrilateral process pro- jects forward for the attachment of the coracoscapular ligament. The neck of the bone is broad and thick; the borders are rounded off- This part constitutes about one-third of the blade; pos- terior to it we have the curved scimitarlike blade of the bone. This is carried backward to a blunt point, the outer border being rounded, and the mesial one having a thickened edge; the latter commences abruptly near the middle point of the bone as a whole. From head to apex this scapula presents a uniform and gentle curvature in the plane of the bone’s blade, the con- cavity of the curve being along the outer border. The harpy eagle has the distal moiety of a scapula very broad, with the outer margin sharp, but the inner one rithmed pretty much as we find it in the monkey-eating species. In the golden eagle the terminal part of the blade is distinctly truncated, and the entire bone is much compressed from above downward. This is also true of the scapula in our white-headed bird; here the blade is extremely narrow and is drawn out posteriorly to a truncated apex. In this species it does not seem to be espe- cially pneumatic. The Korean eagle has a scapula resembling that bone in the golden eagle. These birds—Pithecophaga forms no exception to the rule— possess a big sternum of extraordinary capacity. Ventrally it is _ one deep basin from side to side as well as lengthwise. Down the middle line there is a row of scattered foraminal openings to admit air during the life of the individual. Its xiphoidal border is at right angles to the long axis of the bone and presents but a shallow concavity for its middle third. Well within this border, upon either hand, there is an elliptical foramen of no great size. Seven facets occur upon each costal border, with a great many pneumatic foraminal openings among them. Their interarticular cavities are very shallow. On the outer aspect of 46 Philippine Journal of Science 1919 the body of this sternum, at the anterosuperior angle and in- cluding the distance occupied by the first four hemapophysial articular facets, there is a marked concavity which is the con- tinuation of the costal groove of the same side. Its lower bound- ary is curved, with the convexity toward the keel (Plate IV). The keel is not deep by any means and is continued only two- thirds the length of the body, its lower border being uniformly convex. Anteriorly, the carinal angle is rounded off, the con- cave, anterior border above it being sharp for its lower half and flat for its upper, terminating superiorly in a stumpy manubrium, which is truncated anteriorly, leaving, as it were, a small, triangular surface, with the angle below in the middle line. The deep coracoidal grooves are remarkably well defined in front, in the middle line, at which point they decussate to some extent. The external surface of the sternal body, upon either side of the carina, is smooth, and the line of the pectoral muscle is but faintly defined. Occasionally we find in the body of the sternum of this and other eagles one or two small, sporadic foraminal openings, just as though a bird shot had made a perforation and the wound subsequently healed. I find a rather large, nearly cir- cular one of these in the sternum of the harpy eagle, it being situated on the right side of the body of the bone, near its middle. In this species the sternum is somewhat smaller than the one just described; but, ventrally, it presents the same profound con- cavity, with fewer evidences of pneumaticity, however. The xi- phoidal elliptical foramina, one upon either side, are very much larger; and this extremity of the bone has an increased width © due to the lateral extension of the xiphoidal portion, which is well seen upon lateral view (Plate V). There is the same num- ber of articular facets upon either costal border as we described for Pithecophaga jefferyi—indeed, all eagles seem to have the Same number. The Korean eagle lacks any foraminal openings in the xiphoidal end of the bone, and the sharp border there is entirely lacking in notches, being wholly at right angles to the carina, which fails to run out to it by at least 2 centimeters. Within the thoracic cavity of the body of the bone foraminal per- forations occur not only down the middle line of the bone to a point opposite the last pair of facets on the costal borders, but likewise in a great area in front and along both sides to a similar point. The coracoidal grooves decussate in this eagle, as they do in the white-headed species, and they are invariably of some considerable depth. In the latter bird there is also an absence Ce a ee es ee XV, 1 Shufeldt: The Monkey-eating Eagle AT of the xiphoidal foramina, and that border presents a squarish prolongation of no great size in the middle line; it possesses all the other aquiline characters of this bone. In the bearded vulture, of Europe, the body of the sternum is square in outline, and not a parallelogram as in eagles; its coracoidal grooves do not decussate, and it has but six pairs of costal ribs. In fact, the bone is that of a big vulture, and in no way suggests that of an aquiline species of any sort. The pectoral limb.—Plates VII and VIII. It has long been a well-known fact that all big raptorial birds possess limbs of great size, power, and proportions. This is what we would usually look for when we come to consider their habits and the character of their prey. It also explains the fact that most of the bones of their limbs enjoy a very perfect state of pneumaticity; the skeleton of the foot, however, often forms an exception to this condition. Possibly, in some of our eagles, even the foot bones may be pneumatic; they appear to be so in some degree in our white-headed species, but surely not in the golden eagle. True vultures, almost without exception, possess a skeleton presenting a lightness and an extremely perfect pneumaticity unequaled by any other family of birds. Gypaétus forms a partial excep- tion to this rule, in so far as the skeleton of its feet is concerned; but this bearded species approaches the eagles, while our Amer- ican vultures are birds that practice long-sustained and steady ‘flight, and do not capture their prey—hence a more perfect aéra- tion has evolved in them. In the different species of eagles, in so far as I have examined them, the humerus varies but little in form or in general char- acters. It is invariably a large and thoroughly pneumatic bone, and to this statement Pithecophaga presents no exception. It has a length of about 20 centimeters; and, taken as a whole, its shaft presents the “sigmoid curve” in a nearly perfect degree. Rather less than its middle third is very smooth and quite cy- lindrical in form. Its radial crest is short and triangular in outline, while the ulnar tuberosity is very conspicuously de- veloped and arches over—to some considerable extent prox- imally—the deep pneumatic fossa, in which may be seen the pneumatic foramina of very large size, but generally few in number. There is also a row of these foramina along the base of the smooth, elliptical head of the bone on the anconal side; they are of no great size in this locality. A very distinct elon- gate elliptical area—raised above the general surface—with its major axis parallel to the bone’s shaft, may be seen at the distal 4S Philippine Journal of Science 1919 base of the radial crest on the palmar aspect of the expanded proximal end of this humerus; it denotes the place of insertion, in life, of the pectoralis muscle, and is found in the same location in all true eagles. All of the ornithic characters seen in the avian humerus, in so far as falconine species go, are to be found at the distal end of this bone of our present subject, and each and all of them are unusually prominent. This applies especially to the two articular tubercles, the trachial fossa, and the tendinal grooves on the anconal aspect. In the aforesaid fossa a few, small, scat- tered, pneumatic foramina may be observed, especially just be- yond the radial and ulnar tubercies, The radial crest of the left humerus of this individual exhibits the results of some previous disease, and it has manifested itself in the form of quite an extensive exostosis. The ulnar crest is carried down onto the shaft for a distance of about a centimeter as a sharp and distinct border; while at its middle we note a small foraminal perforation, with a groove leading into or out of it, on the palmar surface of the proximal, expanded extremity of this bone. . This foraminal perforation is absent in the radial crest of the humerus of the harpy eagle, and the crest itself is of a triangular outline (Plate VII, fig. 2) ; while otherwise, in all other matters, the two bones are notably similar in these two eagles. How- ever, in the harpy it may be noted that the caliber of the shaft is greater and presents less sigmoidal curvature. : In the Korean eagle (Thallasoaétus pelagicus) the humerus is fully 1.5 centimeters longer than’it is in Pithecophaga jefferyt; it is also straighter and somewhat slenderer. Its large, trian- gular, radial crest extends farther down the shaft, while in all other respects the two bones are very similar. Curiously enough —the above fact notwithstanding—the skull of the Korean bird is neither as large nor as massive as is the skull of the eagle of the Philippines. This difference is possibly due to a longer wing in the former species; but the material is not at hand at this writing either to prove or to refute any such statement. In September, 1918, I published an account of Pithecophaga jefreryi, illustrated by natural-sized figures of its head and foot.’ In that article I made the statement that the species was the largest of all existing raptorial birds. Possibly this may be so; while, upon the other hand, the big eagle of the Orient (Thalla- * Am. Forestry 24 (1918) 555-557, 2 figs. XV, 1 Shufeldt: The Monkey-eating Eagle 49 soaétus pelagicus), found in Kamchatka and Japan as well as in Korea, may be a bigger bird in some respects. Sharpe included the Old World vultures in the true raptorial group. In the genera Vultur and Serpentarius there are some big species the comparative weights and proportions of which have never been taken for a series of living specimens or compared with the cor- responding data on Pithecophaga jefferyi. I have never compared the wedge-tailed eagle, Uroaétus audax Lath., of Australia and Tasmania, with our present subject; but I am inclined to believe that it is not so large a bird. The radial crest of the humerus of the white-headed eagle of the United States is also triangular in outline; while in Aquila chrysaétos canadensis this feature of the bone under considera- tion is not so lofty and, while triangular in general outline, it extends very much farther down the shaft of the bone. Here, too, the osseous emargination of the pneumatic fossa is broader and more extensive, thus closing in upon the true cavity, though in no other way diminishing its capacity. The bearded vulture of Europe has a humerus fully one-third larger than that bone in Pithecophaga, and its characters are very similar, the most striking departure being the shallow pneu- matic fossa in the former species, with all of its foramina merged into one subcircular foramen. In the antibrachium of Pithecophaga both the ulna and the radius exhibit some degree of curvature between proximal and distal extremities. Air gains access to their interiors through minute foramina at the proximal and the distal end of each; at the latter situation they articulate in the usual manner with the radiale and the ulnare of the carpus, bones that here present the ‘avian characters usually seen among the eagles. The radius has an extreme length of some 20.5 centimeters, and the ulna is about 2 centimeters longer than this. The latter bone has a double row of osseous papille down its cylindrical shaft; these, as in other birds, are for the attachment of the quill butts of the secondary feathers of the wing. There are ten in each row, and all, to the last pair at either end, are opposite each other. The anterior third of the radius is subcylindrical in form, while the remainder of the shaft is trihedral on section. Its “radial tuberosity” is concaved in the center, with the inner margin sharp. Our harpy eagle skeleton lacks the bones of the forearm and manus. Thallasoaétus pelagicus has the radius and the ulna much longer than the monkey-eating eagle, and each is markedly 166316——4 50 Philippine Journal of Science 1919 slenderer. We are also to notice that the papille on the shaft of the latter are closer together. Ten pairs of them appear to be present, and the ulna in this bird has an extreme length of 24.3 centimeters. America’s golden eagle departs from others examined, in that the humerus and the-emur are the only pneumatic bones of the pelvic and pectoral limbs. Unfortunately, I have not at hand the skeleton of manus either in the case of the harpy or of the species here being osteologically considered; so that a description of these parts will have to be undertaken by someone else, when such material falls into the hands of science. Very likely, in other published papers of mine, descriptions of these parts of the skeleton in other species of eagles will be found, either of fossil or existing species. The pelvic limb.—Plate VII, fig. 1; and Plates IX, X, and XI. Pithecophaga jefferyi possesses a big femur, which is permeated by air in all of its parts. There is an extensive group of pneu- matic foramina in the deep popliteal concavity, and another large opening on the proximal end of the bone anteriorly, at the termination of the trochanter major; it is bounded internally by the linea aspera. Caput femoris is sessile, with an extensive pit for the ligamentum teres. The summit is smooth and convex from before backward, and concave from head to the trochanter, which latter rises above the summit of the bone. It is very broad and rough on its external surface. Passing to the shaft, we find that to be very stout and uniformly arched, the convexity being in front. For its middle third it is subcylindrical in form and very slightly roughened throughout. Distally the condyles of this femur are enormous with very prominent articular surfaces. The side of the inner condyle is, for the most part, flat and smooth, while the outer one is slightly roughened and presents a conspicuous tuberosity. As usual, this condyle is vertically divided posteriorly, in that it may articulate with the head of the fibula of the leg. This femur has an extreme length of about 13.2 centimeters, or 2 millimeters more than the length of the femur of a harpy eagle; but the little that the harpy lacks in length in this bone it amply makes up in stout- ness; and, apart from this general stoutness, there is no specific difference worthy of notice or description. The femur of the golden eagle is of about the same length as the two just noticed; but it is a far slenderer bone, with the ellip- - tical pneumatic foramen at its proximal end very conspicuous, and the muscular lines on the shaft are likewise so. The poplit- XV, 1 Shufeldt: The Monkey-eating Eagle 51 eal cavity is shallow in the femur in this species, as it is in some other eagles. Halixetus leucocephalus also has a slenderer femur, while Thallasoaétus pelagicus has one of moderate pro- portions—that is, not as long or as stout as in Pithecophaga, and the shaft is slightly more curved. In the femur of this species the noteworthy characters consist in the great prominence of the tubercle on the outer condyle and the unusual depth of the pit on the caput femoris for the ligamentum teres. Air does not enter so extensively into the bones of the leg in Pithecophaga as it does in the case of the femur. Still, they are to some considerable degree pneumatic; for, upon ordinary mac- eration, the bones largely whiten, though not nearly as much as does the femur, while in no part do they turn fatty yellow, as happens to the metatarsus and toe bones after similar treatment. The tibiotarsus in Pithecophaga measures some 20.4 centi- meters in length, and the fibula measures 15.8 centimeters (com- pare Plate VII, fig. 1, and Plate IX, figs. 2 and 8). The first- named bone is nearly straight from end to end, being but very slightly arched to the front. For its distal third it is markedly flattened in the anteroposterior direction—less so above, where it is flat for its anterior surface and sides, the shaft being here trihedral on section. Distally the condyles are very much sup- pressed posteriorly and correspondingly pronounced in front, where the valley between them is deep. Above them, to the in- ner sides, the tendinal groove is of great depth; it is very dis- tinctly defined as to its boundaries, and the oblique “tendinal bridge” is strong and similarly characterized. Proximally the raised. ridge for articulation with the fibula has a length of 3.5 centimeters and lies wholly within the upper third of the shaft, which here is very flat and smooth. Above this we note the low and short cnemial processes, separated by a wide and rather deep valley. The ectocnemial process termi- nates in a rather blunt, down-turned hook, extending but very slightly above the summit of the bone. The latter slopes to the ‘outer side, and presents a prominent, rounded, articular tubercle near the head of the femur, which may or may not be especially noticeable in this bone in other eagles. At the sides, as well as posteriorly, the summit bulges out over the shaft, being most thickened along its inner course, beyond which point a concavity occurs, bounded in front by the entocnemial process. Turning to the fibula, it is to be observed that its shaft below the fibular ridge of the tibiotarsus is long and slender, being much compressed anteroposteriorly, opposite the middle third 52 Philippine Journal of Science 1919 of the companion bone of the leg. Below this, it runs onto the shaft, and here the contact almost amounts to a direct fusion with it. At its lowest point it is not over 3.5 centimeters above the lower margin of the external tibial condyle. All diurnal Raptores possess fibule of greater or less length; especially do we find this to be the case in Pandion. Superiorly, in our sub- ject, the fibula is much compressed. from side to side, with its head produced posteriorly. Its entire summit, convex from be- fore backward, is an articular facet for the outer condyle of the femur. Pithecophaga possesses a rather large patella, which is elon- gate from side to side, in which direction the surface is convex, while posteriorly it is concave above and doubly concave poste- riorly for articulation with the condyles of the femur. Inferiorly, the surface is limited and flat. Sometimes, as in the case of our present subject, there is a small, circular ossification in the tendon of the muscle, several millimeters below the true sesamoid of the knee joint; it occurs in both limbs. Eagles exhibit very considerable variation in their leg bones, but more in the tibiotarsus than in the fibula. For instance, in the harpy the latter bone is more than a centimeter shorter than it is in the Philippine bird now being described, while at the same time the shaft is very conspicuously flattened from before backward; at the same time, this flattening is associated with a marked increase in the shaft’s width. With respect to the fibula of the harpy, it broadens below the articular ridge, and is grooved for the passage of the muscle that passes over it (Plate VII, fig. 1). In the white-headed eagle the cnemial processes of the tibiotarsus are greatly reduced in all particulars, and to a lesser extent this is also true of the Korean eagle; the fibula is very long in the species. ; With respect to the golden eagle (Aquila chrysaétos canaden- sis), the bones of the leg are nonpneumatic for their entire length below the proximal extremities, the shaft of the tibiotarsus in this species being curved from end to end, the concavity being along the inner or mesial aspect. Skeleton of the foot.—Plates X and XI. As already noted, the tarsometatarsus and other bones of the foot in this eagle are entirely nonpneumatic, while in some other forms the re- verse of this is the case (Halixetus leucocephalus). Irrespective of species the tarsometatarsus possesses the same general char- acters throughout the group, and its form is very characteristic. All this part of the skeleton in the harpy was left in the skin of XV, 1 Shufeldt: The Monkey-eating Eagle 58 the mounted specimen at the United States National Museum, so that no comparison can be made here along such lines with the corresponding bones in our subject. In the latter the bone has a length of about 12 centimeters and an average breadth of 2 centimeters. Its summit exhibits two shallow articular facets for the condyles of the tibiotarsus. An intercondylar tubercle stands between them on the anterior margin, the width of the surface being double its depth, while posteriorly is to be noted a hypotarsus of the usual aquiline type. This consists of a quadrilateral process standing out at right angles to the bone—its hinder border being considerably thickened—and its base opposite the inner articular concavity at the summit. Op- posite the outer one is another process, it being stumpy and much aborted; between the two is a wide, shallow valley. Posteriorly the shaft of this tarsometatarsus is smooth, broad above, and somewhat narrower distally. It is concaved through- out its entire extent, its margins being more or less sharpened, except where the accessory metatarsal is swung by ligament about a centimeter above the inner trochlea. The outer aspect of the shaft is flat, being broad at the middle third, and taper- ing somewhat to the ends. It is the posterior margin of this surface that forms the inner sharpened border of the posterior aspect alluded to above. Anteriorly the surface is much twisted upon itself, the upper half being concaved longitudinally and convexed for the rest of the extent, the two merging into each other. At the upper third, anteriorly, we note the usual twin perforations a short distance below the summit; and below them, to the outer side, is the elongate tubercle for muscular insertion. Distally there are three very substantial trochlez for the basal phalanges of the toes, the inner one being the largest and at the same time the lowest on the shaft. Slightly up on the shaft, - between the outer and middle condyle, is to be noted the usual foraminal perforation present in nearly all birds; it transmits the flexor tendon of the outer toe and, it is said, the anterior tibial artery. The accessory metatarsal is of considerable size, is flattened and somewhat twisted upon itself, and supports distally a very large, transversely disposed, trochlear facet; the latter is devoted to the articulation of the basal joint of the hallux, which is a digit of enormous power and strength, as will be appreciated through a glance at Plate X. Equally powerful are the joints and the talons of the anterior toes, the arrangement and pro- 5A Philippine Journal of Science 1919 portions of which are upon the usual ornithic plan of the aqui- line pes, with the osteological characters I have often described before, of existing forms as well as of fossil eagles. CONCLUSIONS Coming to the relationships of Pithecophaga jefferyi, with respect to other representatives of the same group and in the light of the osteological material above described and compared with its skeleton, I am of the opinion that its nearest ally, in so far as we are at present acquainted with the morphology -° of other eagles, is the harpy eagle (Thrasaétos harpyja). Sharpe? makes a very different disposition of this species; he places it between Circaétus and Spilornis, of the Aquiline (Subfamily V); while the harpy eagle we find arrayed with the Buteonine (Subfamily III), following Harpyopsis, the next following Subfamily (IV) being the Gypaétine (Gypaétus bar- batus). The osteology of Pithecophaga surely does not support this arrangement. PREVIOUS PAPERS BY THE AUTHOR ON THE OSTEOLOGY OF THE EAGLES AND THEIR ALLIES [Papers on the owls and the vultures are not ineluded.} 1. On the ossicle of the antibrachium as found in some of the North American Faleconide. Bull. Nutt. Orn. Club 6 (1881) 197-203. 2. On the free post-pubis in certain of the Falconide. Auk 3 (1886) 133, 134. 1 text figure. 3. Osteology of Circus hudsonius. Journ. Comp. Med. and Surg. 10 (1889) 126-159. 17 text figures. 4. A peculiar character referable to the base of the skull in Pandion. Auk 8 (1891) 236, 237. 5. Some comparative osteological notes on the American kites. Ibis VI 3 (1891) 228-232. 6. On a collection of fossil birds from the Equus beds of Oregon. Am. Nat. 25 (1896) 303-306. Plates. 7. Fossil birds from the Equus beds of Oregon. Am. Nat. 25 (1891) 818-821. 8. Tertiary fossils of North American birds. Auk 8 (1891) 365-368. ' 9. A study of the fossil avifauna of the Equus beds of the Oregon Desert. Journ. Acad. Nat. Sci. Philadelphia 9 (1892) 389-425, pls. 15-17. 10. ~ cases ~ complete fibule in existing birds. Ibis VI 6 (1894) 361-366, ge. 1s 11. Some of the “outliers” among birds. Pop. Sci. Monthly 44 (1895) 760-780. 10 text figures (not osteological). 12. On the affinities of Harpagornis. Trans. New Zealand Inst, 28: 665. 13. Observations on the classification of birds. Proc. Acad. Nat. Sci. Philadelphia (1898) 489-499. Illustrated. (Osteology largely used.) * A Hand-list of the Genera and Species of Birds. London 1 (1899) 265. XV, 1 Shufeldt: The Monkey-eating Eagle 55 14. 15. 16. a a & 18. An arrangement of the families and higher groups of birds. Am. Nat. 38 (1904) 8338-856, text figs. 1-6. (Osteology largely used.) Where the skeletons of American birds may be studied. Auk 26 (1909) 217-218. Osteology of birds. Bull. N. Y. State Mus. 130 (1909) 367 pp. Educa- tion Department Bulletin No. 447. (This book contains many cuts and plates devoted to the osteology of the Falconide and has been particularly useful in studying the skeleton of the eagles.) Review of the fossil fauna of the desert region of Oregon, with a de- scription of additional materia] collected there. Bull. Am. Mus. Nat. Hist. 32 (1918) 123-178, pls. 9-48, figs. 1-578. Further studies of fossil birds, with descriptions of new and extinct species. Bull. Am. Mus. Nat. Hist. 32 (1913) 285-306, pls. 51-59, figs. 1-94. ILLUSTRATIONS [All the figures are reproductions of photographs by the author made direct from the specimens. ] PLATE I Fig. 1. Pithecophaga jefferyi, adult; right lateral view of the skull, natural size. All the bones of this bird figured on the plates are of two individuals. The specimens were received from Mr. R. C. MeGregor, of the Bureau of Science, Manila, P. I. 2. The skull shown in fig. 1, seen directly from above, natural size. PLATE II Fie. 1. Halixetus leucocephalus, white-headed eagle, sex?, adult; cranium, direct view from above, very slightly reduced. Median lon- gitudinal diameter measures 11.1 centimeters in this specimen. No. 19278, United States National Museum collection. This skull belongs to the skeleton of the specimen used throughout this paper. 2. Aquila chrysaétos, female, adult; cranium, direct view from above, very slightly reduced. Median longitudinal diameter measures 11.9 centimeters in this specimen. No. 18802, United States National Museum collection. 3. Morphnus guianensis, sex?, adult; cranium, direct view from above, very slightly reduced. Median longitudinal diameter measures 9 centimeters in this specimen. No. 18468, United States National Museum collection. Puate [il [All figures of Plate III natura] size and from the same species, Pithecophaga jefferyi.] Fig. 1. The six caudal vertebre (a to 6) arranged in normal sequence and viewed from above; a, anterior one of chain. . Pygostyle, right lateral aspect. Lower part pierced by a foramen. . Fifth cervical vertebra, seen from above. Mandible, seen from above. Anterior aspect of ninth cervical vertebra. . Trachea, twisted and distorted through drying; fragments of bronchial tubes at inner end. . The atlas, posterior aspect. . Anterior view of axis; neural spine directed downward. . Sclerotal circlets of the eyes; elements as they occur or are arranged in life. 10. Hyoid arches viewed from above, with larynx in situ; the latter removed at a point seen at the outer end of the trachea in fig. 6. PLATE ay Pithecophaga jefferyi; right lateral view of the trunk skeleton, including bones of the shoulder girdle, reduced about one-third. The pygostyle and the six caudal vertebre preceding it have been removed. aoe wh © 00 ~] 57 58 Philippine Journal of Science PLATE V Thrasaétos harpyja, harpy eagle; right lateral view of the trunk skeleton, including bones of the shoulder girdle, reduced less than one- half. Length of sternum in the specimen equals 13 centimeters. No. 225806, United States National Museum collection; a specimen from Brazil. PLATE VI Pithecophaga jefferyi; trunk skeleton, seen upon direct dorsal view, with the shoulder girdle in situ, reduced about one-third. The pygo- style and the six caudal vertebre preceding it have been removed. PLATE VII Fie. 1. Thrasaétos harpyja; anterior aspect of the right tibiotarsus and fibula, slightly reduced. From the same specimen as Plate V. 2. Thrasaétos harpyja; anconal aspect of left humerus, slightly reduced. From the same specimen as Plate V. PLATE VIII Fig. 1. Aquila chrysaétos, golden eagle; anconal aspect of right humerus. No, 18802, United States National Museum collection. Length of bone, about 18.2 centimeters. 2. Thrasaétos harpyja; anconal aspect of right humerus. From the same specimen as Plate V. 3. Pithecophaga jefferyi; anconal aspect of right humerus, reduced about one-third. Bone in life measures about 19.7 centimeters in length. PLATE IX Fie. 1. Pithecophaga jefferyi; left femur, posterior aspect. Extreme length of bone in specimen, about 13 centimeters. Bone entirely pneumatic. — 2. Pithecophaga jefferyi; right tarsometatarsus and fibula, viewed on direct anterior aspect. Length of tarsometatarsus in speci- men, about 23 centimeters. 3. Pithecophaga jefferyi; left tarsometatarsus and fibula, viewed on direct outer aspect. PLATE X Pithecophaga jefferyi; skeleton of the right foot, mesial aspect, natural c size. PLATE XI Fic, 1. Halixetus leucocephalus; skeleton of right foot, inner aspect, natural size. No. 19278, United States National Museum collection. 2. Halizetus leucocephalus; skeleton of left foot, outer aspect, natural size. The same bird as in fig. 1. SHUFELDT: OSTEOLOGY OF PITHECOPHAGA, ] [Putuie. Journ. Sct., XV, No. 1. PLATE I. PITHECOPHAGA JEFFERY]. SHUFELDT: OSTEOLOGY OF PITHECOPHAGA.] (PHILIP. Journ. Scr., XV, No. 1 PLATE I]. CRANIA OF THREE EAGLES ] I > ae uJ im Le uJ => i. 8 Oo 4 = oO. Oo oO ul rs i oO PLATE Ill, PITHECOPHAGA. ] OSTEOLOGY OF SHUFELDT: ‘\AY¥3a445f VOVHdOOSHLId JO NOLATISMS ANNYL ‘Al 3ALVId "TON SAX “IOS “NUNOG “dITIHG] [°VOVHdOOdHLIgG AO ADOTIOSNISO :LaTIaNHs “WPAdHVH SOLAVSVHYHL JO NOLSTSYS MNNYL “A 3LV 1d “TT “ON “AX “IOS ‘NUNOL “dITIHG] ['VOVHd00aHIIg 40 ADO TIOSISO : LAUTAANHS WAdasasdal VOVHdOOAHLId JO NOLSTANS MNNYL "IA 3ALVId ‘LON SAX “IOS “NUNOL ‘aITIHG] ['VOVHd00aHLIg dO ADOIONLSO : LaTANHS “VPAdYVH SOLAVSVYHL ‘NA ALVId VOVHdO ) iHLid ‘SA1T9V3 S3SYHL JO IYSWNH CHIA ALWId ‘LT ‘ON “AX “IOS ‘NUNOF ‘dITIHG] [‘VWOVHdOOUHLIg JO ADOTOSLSO : LATAINHS “\Adga4dd35r VOVHdOOSHLId JO SANOA DAT *X!I ALV Id "IT “ON SAX “IOS “NUNOF ‘dITIHd] |" VOVHdO0OOFHLIG JO ADOTIONLSO : LaTaANHS ‘AWS 443f VOVHdOOFHLId JO L004 *X ALVId niceties = ‘TON ‘AX “IDS ‘NUNOL “AITIHG] ['VDVHd00SHLIG JO ADOIOSNISO : LATIANHS “SN1IVHd390090N31 SNLAVIIWH 4O Laa4d ‘IX ALWId ‘TON ‘AX “IOS “NENOR ‘aITIHG] ['VOVHdOOaHLIg 40 ADOTONLISO :? LaTadOHYg GENERAL FACTS IN THE BIOLOGY OF PHILIPPINE MOUND-BUILDING TERMITES By LEopotpo B. UICHANCO Of the Department of Entomology, College of Agriculture, University of the Philippines, Los Battos FOUR PLATES The mound-building species of Philippine termites so far known are all members of the genus Termes. They are common in rural and suburban places; and in caifigin fields they are left undisturbed, although undoubtedly a menace to farming operations, both because they are always in the way and because they are prolific sources of severe anay' infestation. The tao tolerates the nests, not because he is too lazy to dig them out, but because a tradition current among Filipino peasants makes the nuno? the regular occupant of the mound. The nuno is a sort of wood-inhabiting spirit, believed to be normally invisi- ble; he, however, occasionally appears to us mortals in the form of a small, unsightly, evil-looking old man, wearing a large salacot.2. His every desire is a wicked one. He takes pleasure in causing human suffering; and to injure him, even though accidentally, means sickness, and often death, to the offender. The termite mounds found in the Philippines do not reach the proportions of those met with in some other countries. In Australia, for example, termite mounds grow so large that they often appear to the traveler to be huge earthen towers. Old termite mounds in the Philippines very rarely reach a height of two meters. They generally approach a round-topped, conical shape, and are usually free from green vegetation, although they are occasionally found covered all over with a thick mat of cogon or low shrubbery. Sometimes a mound is built around or against a large tree trunk, in which case its shape becomes very irregular. The outer wall of the nest is remarkably com- pact, considerable force being necessary to break it open with a mattock. The main constituent of this wall is very fine clay, * Local name for the termite, or white ant. * Tagalog for grandfather. * A native helmet, usually made of bamboo or palm leaves, and rattan. 59 60 Philippine Journal of Science 1919 which has probably been made firm and sticky by admixture with certain substances from the bodies of the termites. Mounds over which a large heap of grass or sticks had been burnt were, on opening, observed to have all their inhabitants, includ- ing the delicate nymphs and the eggs, absolutely unaffected by the heat, : Beneath the thick outer crust of clay are the “fungus gardens,” composed entirely of a mixture of fine pellets of woody excreta and plasterlike material, which have passed out of the alimen- tary canals of the workers. In a typical anay mound these gardens occupy the greater part of the inner cavity; compact slabs of moist clay, with numerous irregular passageways run- ing through them, form the partitions and, at the same time, the supports of the gardens. Each garden is composed of nu- merous vertical galleries formed by the intersection of the walls; the general appearance is that of a sea coral, with the upper convex margin of each partition neatly finished and the lower part either unfinished or made rugged by being bitten off. As the name indicates, the termites utilize the gardens in the cultivation of certain species of fungi, or mushrooms, to be more specific. They regularly harvest the “buttons” of the fungi, as soon as they appear, for use as food. At certain seasons of the year, during the rainy months, the mushrooms break through the outer wall of the nest and develop into full- grown plants. These mushrooms are a common sight in the regions about Los Bafios on the mounds, or on the surface over underground anay nests; and they are much relished as an article of food by everybody who eats them. The local farmers know them under the general name of mamunso.* Several theories have been proposed in connection with the periodical appearances of this mushroom. One explanation is that on the advent of a more favorable season, the fungi in the termites’ gardens become so numerous that a certain number find a chance to escape and develop into perfect mushrooms before the workers have a chance to harvest the “buttons.” Another is that the termites purposely allow the mushrooms to develop into mature plants in order to furnish seed for the following season. The galleries of the fungus gardens and the network of tunnels in the intervening slabs of clay also serve as nurseries, where the eggs and the young nymphs are taken care of. An upright, irregularly shaped pillar of clay, which forms. the core of the nest, serves the double purpose of sup- * Of or pertaining to the mound (Tagalog) . arias wig cat ae ek ee rote — ieee na Se eBLiN eS xv,1 Utchanco: Philippine Mound-building Termites 61 porting the series of fungus gardens and of being the main prop that prevents the outer shell of the nest from collapsing. At least six different forms of adult individuals are present in the nest; namely, the king, the queen, the large and the small workers, and the large and the small soldiers. -Each caste has a distinct function in the nest; and, due to the similarity of the termites to the true ants in community economy, they are often termed ‘“‘white ants.” The various castes present in each nest are the offspring of a single pair, which has settled down and mated after swarming, The swarming of winged termites occurs after sunset or late in the evening, usually during the months between May and September. The swarms are composed of individuals of both sexes, the males for some reason far outnumbering the females. Several counts made on certain swarms of Termes (Macro- termes) philippinensis Oshima, in 1915, gave an average of less than five females to a hundred males. These swarms come from well-established anay nests where, at certain seasons of the year, winged individuals are produced. On certain nights, prob- ably as a result of favorable climatic conditions, large swarms of winged termites are observed to occur at the same time in widely separated localities; for example, the campus of the College of Agriculture and the municipalities of Los Bafios and Calamba. These insects are exceedingly feeble fliers and, for this reason, cannot travel far on their wings; they could not possibly have reached these very distant places from a nest located at any one spot. The more logical inference is that there occurs a simultaneous swarming from nests over a large area. An enormous number of individuals composes a swarm; but a great many of them perish in their conjugal flight by being preyed upon by their numerous enemies—principally bats, birds, lizards, and ants. After shedding their wings, the couples run about, the male following the female by clinging fast with his mouth parts to the posterior extremity of her body, until they finally succeed in locating a fit spot in which to build their nest. This is usually under a stone or a piece of wood on moist ground, where they dig a small hollow in which they establish themselves. Some observers maintain that a group of wandering workers of the same species must fall in the way of each newly mated couple and help it establish the colony; but more sound evidence is necessary definitely to prove this assertion. The paired in- dividuals coming down from the swarm and building a nest go 62 Philippine Journal of Science 1919 for a considerable length of time without food. About five days after starting the new home, the female begins to lay from one to five eggs,-at intervals of two or three days. It takes the eggs about a month to hatch, and the adults that de- velop therefrom are mostly workers, only a very few soldiers being produced. Sexual individuals are not produced until later, when the colony is already well established. The nymphs of soldiers do not exhibit the characteristic appearance of their respective mature forms until about the last instar, or imme- diately before the insect reaches the adult stage. Nymphs of the sexual caste are larger than those of either soldier or worker and are easily distinguished by the presence of wing pads on those areas of the body where the full-grown wings are to develop later. The nymphal stage of the soldiers and work- ers lasts about a month; no definite observation has as yet been made in connection with that of the winged form. In about three months after the establishment of the nest, a sufficient number of small workers has been developed to start the work on a fungus garden and on the preliminary construction of the new mound which is to become the permanent home of the colony. In a well-established nest the royal chamber is located a little below the level of the surface of the ground. It is a hollow, plano-convex chamber, with thick, irregular, clay walls and nu- merous passageways connecting with the adjoining parts of the nest. Within the chamber are the king and the queen, attended by a large bodyguard of soldiers and workers. Occasionally, two queens are found occupying the same chamber in a nest; but in all cases not more than one male is present. The king has not changed his appearance since he first came down from the nuptial flight; he is still very lively and is likely to slip away and escape detection unless some care is exercised in re- moving the royal chamber from the nest. On the other hand, the queen has changed considerably. Her abdomen is much engorged with eggs; and the abdominal tergites and sternites, once closely connected, are now situated far apart, with their connecting membranes greatly distended. The insect at this time presents a characteristic sausagelike appearance. Her ac- tivity is restricted in so far as locomotion is concerned; and she is now entirely helpless—actually a prisoner in her own cell. The workers feed her, clean her body, and look after her personal comfort. Her abdomen exhibits a succession of peristaltic move- ments, attended by a continuous discharge of more or less ellip- aaa ee SE eieneamapnesiitandeiridt scares caremtaaaaiaeeememeet ee XV, 1 Uichanco: Philippine Mound-building Termites 63 tical eggs. A full-grown queen lays eggs at the rate of thirty to sixty a minute. The workers remove the eggs immediately after they are extruded and ¢arry them away to be taken care of in the nurseries. The original king does not live long, being probably replaced many times during the existence of the colony. The queen’s life has been estimated to last about twenty years. Certain species of termites insure the perpetuation of their colony by providing for a substitute queen, sometimes designated ne- oteinic. The latter is distinguishable from the true queen by the absence of any indication of her having ever attained the power of flight. She is fully as capable of reproduction as the true queen, but does not live quite so long. The soldiers and the workers are asexual individuals. They have no external indication of the organs of sight and never develop wings. They avoid the light and build protective covers wherever they go, though there is a species of black termites ° which travels readily in the daylight through forests and in open places when necessary. Numerous passageways radiate in all directions, under or upon the surface of the ground, from their nests to the objects of their attack—old logs, living tree trunks, house posts, fences; in fact, all kinds of ligneous ma- terials that may come within their reach. Accompanied by a comparatively small number of soldiers, a number of workers set out upon their work in regular processions, usually under cover of dirt tunnels. The others that remain at home are on duty attending the queen, the eggs, and the young; enlarging the mound; harvesting the crops; protecting the nest against intruding enemies (of which the worst is a species of red ant, Solenopsis geminata Fabr., very frequently found living in large ‘ colonies within easy reach of the mounds); and keeping the mound in a healthy and sanitary condition. To keep their nest clean, the termites maintain, or otherwise encourage, in their colony a force of scavengers, often termed “guests,” consisting of beetles, earwigs, cockroaches, spring-tails, and myriapods, which continually rid the premises of waste materials. Communication among the termites is undoubtedly carried on by means of their senses of touch and of smell. They have a remarkable ability to locate the different places in the perfectly dark nest, and can start on long journeys and search out their various necessities with an admirable degree of precision. A * Probably not a mound-building species. 64 Philippine Journal of Science hole made in a tunnel or in the outer crust of the nest is sure to bring in response the soldiers’ powerful mandibles protruding from within, around the edges of the break, while the workers come in succession, each carrying a small pellet of materials for repair. The workers, architects by nature, know just when a portion of their structure is not strong enough to prevent col- lapse, for in tunnels where the walls span out very wide to the two sides, they reénforce the weaker spots with supporting pillars of prepared earth. In a termite colony each worker is apparently his own boss; yet no confusion or trouble arises as a result, for they all seem to think, plan, and work together in perfect unison, as though they were interrelated units of one single organism—a marvelous example of the true communistic society where every effort of every individual is directed toward the general welfare, the good of the race. ll Rg Fig. 1. . Vertical section of the mound shown in fig. 1. Note the thickness Fic. Fic. Fig. CO NO et PON ILLUSTRATIONS PLATE I A typical termite mound, or punso, in the Philippines. of the outer wall and the central upright pillar of clay. The fungus gardens are arranged between this pillar and the outer crust. PLATE IT . A layer of fungus gardens, showing arrangement in the nest. . Royal chamber, top view. . Royal chamber, vertical section through the middle portion. The flat floor and concave roof are characteristic features of the chamber. PLATE III . Fungus garden, top view, with large fungus “buttons” probably preparing to break through the outer wall of the nest and develop into full-grown mushrooms. . A portion of the slab of compact clay from the central upright pillar of the nest, showing openings to the passageways of soldiers and workers. . A typical fungus garden. Top view, on the right, showing the characteristic even smoothness of the edges of the walls; nether aspect, on the left. The rugged condition of the edges of the walls is natural and very characteristic. PLATE IV . Tunnel of prepared earth built by workers on dead wood to protect them from light in their expeditions. . Work of termites on structural bamboo. . Work of termites on Oregon pine. The royal couple and retinue. The big, sausagelike individual is the queen, with a batch of freshly laid eggs near the tip of the abdomen. The individuals with large shiny heads and long man- dibles are soldiers. All the rest are workers. 65 166316——-5 ‘| LV 1d ‘punow jO U013z99S |BOIPIeA = *zy *BIY *punow 93/W49} ;eoldAz VY ‘LT ‘Big ‘T ‘ON ‘AX “IOS ‘N¥nOP ‘arTIHg] [‘SLMNYaT, DNIGTING-GNOOW® : OONVHOIQ UicHANCO: MoUND-BUILDING TERMITES. ]} [Puiuir. Journ. Sct., XV, No. 1. | veer nate ire ma cttit eates Fig. 3. Royal chamber, vertical section. PLATE Il. UICHANCO: MOUND-BUILDING TERMITES. ] [PHILIP. JoURN. Scr., XV, No Fig. 1. A fungus garden. Fig. 2. The royal chamber. Fig. 3. A typical fungus garden. PLATE Ill. UICHANCO: MOUND-BUILDING TERMITES. ] [PHiuip. Journ. Scr., XV, No. 1. al ~, : Fig. 4. The royal couple and retinue. PLATE IV, THE MALAYAN MACHA#ROTINAD (CERCOPIDAL) By C. F. BAKER Of the Philippine College of Agriculture, Los Banos THREE PLATES AND ONE TEXT FIGURE The insects composing this group, remarkable in both struc- ture and habit, have been latterly considered to form a sub- family in the Cercopide. When only Machxrota was known, this seemed doubtful, but through other recently described genera the relationship is much more evident. The habits of members of this group distinguish them from all other Cerco- pide, and from most other insects as well; since the nymphs live in calcareous tubes, of curious form characteristic of the various species, and possess remarkable structural adaptations to life in these tubes, which remain attached to their special food plants. Had the Enderleiniini not been included in the group, the Macherotinze would have been easy to define, the remarkable scutellar appendage and characteristic structure of the tegmina being diagnostic. Schmidt defined the Enderleiniini as having the scutellum without spinous appendage, but elongate caudad and reaching or exceeding the apex of the abdomen. However, Macheropsis, as illustrated by Lallemand,' has a scutellum not reaching half the length of the abdomen, though in structure otherwise it appears to be somewhat macherotine; and one of Schmidt’s most recent genera, Modiglianella, is very close to cercopid genera like Quinquatrus Distant and Hindola Kirkaldy (—Carystus Stal). It seems, arguing from structure alone, that the tribe Enderleiniini should be transferred to the Aphro- phorine, which already includes far more diverse elements, the latter subfamily to be divided into a number of codrdinate tribes.2. This would not preclude the possibility of the true Macherotine being offshoots from Enderleinia-like forms. The *Cercopide of the Genera Insectorum (1912). ?JIn Fauna Brit. India, Rhynch. 6 (1916) 184, Distant describes a new genus, Conditor, of which he says, “Allied to Machexropsis.” It does not, however, belong in the same tribe with Machexropsis (Enderleiniini), but is apparently closely related to Pachymachxrota in the Machzrotini. 67 68 Philippine Journal of Science 1919 present paper will deal only with the true Macherotine known from Malaya. The first species known in this group was Machxrota ensifera Burmeister, described in 1835, from Manila. In 1907 * Schmidt presented a monograph of the group, separating it into four tribes and recognizing eight genera. In 1912 the same‘ author added two genera, Modiglianella from Sumatra and Neuroma- cherota from German East Africa. Two Australian genera, Polychztophyes and Pectinarophyes, were described by Kir- kaldy® in 1906, though these were not included in Schmidt’s monograph. Machzropsis Melichar is known only from Ceylon, Hnder- leinia Schmidt only from Togo, and Pachymachxrota Schmidt from Australia and Sumba. Therefore, half of the known ge- nera and the larger proportion of the known species come from the Malayan Region. Since this region, as well as Australasia, is but little explored as to the Homoptera, the known forms of this group probably represent but a small part of the genera and species in existence. So far, Machxrota appears to be the only genus of wide distribution, species of it being found from India (with Ceylon and Andamans) and China, to the Philippines, Malaya, and Flores. It will undoubtedly be found still farther to the southeast, through the Moluccas. New Guinea will al- most certainly yield interesting members of the group. In the present paper, two new Philippine species are added to Macherota; Grypomachxroia is shown to be represented by two new species in Borneo and to occur in Penang; and Maxudea is for the first time recorded from the Malay Peninsula. The Machzrotine present a most interesting field for anatom- ical study. The sclerites of the thorax are deeply and sharply marked, and present most remarkable forms. A sounder class- ification will undoubtedly be possible when these have been ex- haustively studied. The hind femora (? in Maxudea) usually have a strong tubercle beneath at base. The membranous ap- pendix of tegmina is usually transversely fluted. The impressed horizontal suture in front of ocelli is the actual anterior margin of vertex, the sutures passing forward from its extremities being the lateral margins of frons. The upper portions of the cheeks are thus distinctly superior. It thus results that the projection of the head in Grypomachxrota and Sigmasoma is * Ent. Zeitg. Stettin 68 (1907) 165-200. *Ent. Zeitg. Stettin 73 (1912) 173-178. * Bull. Haw. Sugar’ Plant. Exp. Sta. 1 (1906) 384-386. See ey” helen XV, 1 Baker: The Malayan Macheroting 69 entirely referable to the frons. For purposes of description in this paper, however, the entire superior surface of the head between the eyes is termed vertex, Subfamily MACHAZROTIN4ZE: sensu stricto Key to the tribes. a*. Body slender; scutellum arched and with a distinct dorsal furrow; pronotum without produced lateral angles. 6’. Frons not produced above; hind tibie without a tooth... Machzrotini. b*. Frons acutely produced above; hind tibie with one tooth..Sigmasomini. a’. Body very thick and stout; scutellum nearly flat and with dorsal furrow subobsolete; pronotum with lateral angles produced into high thin WDA oo ee a ee Sos Maxudeini. Tribe MACH Z2ROTINI This tribe includes but one Malayan genus, so far as known. Genus MACHZROTA Burmeister Key to the species. a’. Claval vein apically forked. b*. Body of scutellum not high arched posteriorly, its greatest depth much less than length; vertex anteriorly broadly rounded; spatulate scu- tellar furrow about half length of body of scutellum (Plate I, figs. tend. 9) «aoe a M. notoceras Schmidt. 6°. Body of scutellum high arched posteriorly, its greatest depth as seen from side about equaling length; vertex anteriorly subangulate; spatulate scutellar furrow much more than half length of body of scutellum (Plate I, figs. 2, 3, and 10; Plate II, fig. 1). M. philippinensis sp. nov. a’, Claval vein simple. b*. Median cell short and broad; length of vertex far less than width between eyes posteriorly; spatulate scutellar furrow less than half length of body of scutellum. c*. Length of body of scutellum distinctly greater than that of pronotum; length of scutellar spine less than once and a half the length of body of scutellum; ocelli nearer to median line than to eyes (Plate I, figs. 4, 5, 11, and 14; Plate II, figs. 3 and 4).. M. ensifera Burm. c’, Length of body of scutellum subequal to that of pronotum; length of scutellar spine a little less than twice length of body of scu- tellum; ocelli as near median line as to eyes (Plate I, figs. 6 and 12; Plate 17, figs. 5,6, and-18).....0.— 2... M. luzonensis Schmidt. b*. Median cell long and narrow; length of vertex subequal to width between eyes posteriorly; spatulate scutellar furrow much more than half length of body of scutellum (Plate I, figs. 7, 8, and 13; Plate II, figs. 2, 7, and 10)... . M. fusca sp, noy. In Macherota notoceras, M. philippinensis, and M. ensifera the sharp upper border of the scutellar spine is extended into the scutellar furrow as a sharp median carina to half its length, whereas in M. luzonensis and fusca, it scarcely enters this 70 Philippine Journal of Science 1919 furrow. All of the species possess a small, thin, semitranslu- cent, decolored spot on the upper border of the spine near its base. The depth of coloring is widely variable in all of the species, and the males are always smaller and darker than the females. Structural characters only should be depended upon for classification. What appear to be the normal color forms are described in the following pages.° Macherota notoceras Schmidt. Machzxroia notoceras SCHMIDT, Ent. Zeitg. Stettin 68 (1907) 192. This species is common on Penang Island, Wellesley Province, Straits Settlements. Macherota philippinensis sp, nov. Female.—Length to end of abdomen, 4.5 millimeters; to ends of tegmina, 6; to end of spine, 7. Color chocolate; frontal ridge and apex of frons piceous; sides of frons with ten horizontal yellowish stripes, these with numerous short, coarse, appressed white hairs; second antennal joint rufous with white distal border; sides of body of scutellum with curved ivory-white ares, reaching from anterior lateral angles to anterior extremity of scutellar groove; the thin mar- gins of scutellar groove alternately black and yellow spotted; a few minute black dots on posterior lateral walls of scutellar body; posterior border of scutellar body below spine ivory- white, the stramineous spine separated from brown scutellar body by an irregular piceous border. Veins of tegmina stra- mineous with scattering piceous dots, the tegmina suffused with very pale stramineous, the costa brown near base. Abdom- inal dorsum black, first tergite yellow; venter largely piceous; tibi# and tarsi paler brown than femora. Frons very minutely punctate-rugose, near clypeus and with clypeus, coarsely punctate-rugose ; median ridge broad and strong on upper half of frons; vertex sculptured like the frons. Pro- notum deeply, coarsely reticulate-punctate, this becoming weaker near anterior border; the low weak median ridge is continuous throughout; on either side anteriorly are two short, broad, oblique, shallowly depressed ares, which do not reach anterior margin, and are thickly, minutely punctate or shagreened within. Body of scutellum coarsely reticulate-punctate; scutellar furrow * Characters given in the keys are not repeated in the descriptions, so that for complete diagnoses of the species, the two should be combined. XV, 1 Baker: The Malayan Machzroting 71 reticulate within, the lateral margins high and thin; spine with its point about 1 millimeter above apex of tegmina. Male.—Length to end of abdomen, 3 millimeters; to end of tegmina, 4.5; to end of spine, 5.5. Colors deeper than in the female, the pronotum almost piceous. The spine is more or less minutely dotted, and the lateral yellow marks on body of scutellum are abbreviated posteriorly. This is a common species at Los Bafios and on Mount Ma- quiling, both localities in Laguna Province, Luzon Island, Phil- ippine Islands. s Macherota luzonensis Schmidt. Machzxrota luzonensis ScHMIpT, Ent. Zeitg. Stettin 68 (1907) 189. Female.—Length to end of abdomen, 4 millimeters; to ends of tegmina, 6; to end of spine, 7. Color stramineous; spine, and scutellum caudad, pale reddish brown; legs yellowish. Frons yellowish, with eight horizontal piceous bands on either side, and with no median dark color except above; scutellum with an indistinct, median piceous stripe; posterior yellow border of body of scutellum broader below, the intramarginal, deeply depressed line piceous; the curved yellowish marks on sides of body of scutellum very in- distinct; a few dark dots on borders of scutellar groove. Abdomen stramineous, ovipositor piceous. Frons nearly smooth, indistinctly shagreened, the lower bor- der, with clypeus, more coarsely sculptured; face almost entire- ly without appressed white pubescence. Pronotum coarsely reticulate-punctate except near anterior margin, but the retic- ulations are not sharp; the four slightly depressed oblique aree near fore margin short, minutely roughened, and piceous; the median ridge sharp and high. Reticulate character of punc- turation less evident on body of scutellum; spine strongly curved at base, its point a millimeter above the apex of tegmina. In all of the Philippine material before me, there appear to be but two specimens that approximately fit the original de- scription of Schmidt’s species. These are from Davao, Min- danao Island, Philippine Islands. Macherota ensifera Burm. Macherota ensifera BURMEISTER, Handb. Ent. 2* (1835) 128. The disposition of colors on pronotum in longitudinal stripes clearly marks this species in all of its various light and dark phases. The common form in Luzon is pale in color, but very J Philippine Journal of Science 1919 dark-colored forms come from Palawan. The palest forms in my collection were taken in Imugan, Nueva Vizcaya Province, Luzon. The very broad, shining, black, median portion of face, broader above, is also characteristic in all color phases; in the Luzon forms this dark color is lacking on the lower frons and on clypeus, whereas in the dark Palawan forms it is broad throughout; only rudiments of horizontal, lateral, frontal stripes are present in any of the color phases. The four slightly de- pressed and finely sculptured oblique areze near anterior border of pronotum are concolorous. The spine is scarcely curved where “it joins body of scutellum, and its point is about 1.5 millimeters above apex of tegmina. Abundant throughout the Philippine Islands. Macherota fusca sp. nov. Female.—Length to end of abdomen, 5 millimeters; to ends of tegmina, 7.25; to end of spine, 8.5. _ Color stramineous to chocolate, to piceous, and strongly con- trasted. Entire face black; vertex, pronotum, and scutellum chocolate; posterolateral arez of pronotal disk stramineous; lateral angles of pronotum shining piceous; lateral areze of vertex and short longitudinal stripes inclosing the ocelli pos- teriorly, stramineous; lateral yellow stripes on body of scutellum long, straight, more than two-thirds length of scutellar body and somewhat broader caudad; posterior border of scutellar body very narrowly and shortly yellowed; scutellar spine piceous. Pleure stramineous; venter piceous tc brown; legs brown, femora stramineous at apex, fore and middle tibize stramineous at base, hind tibie stramineous on basal three-fourths. Ab- domen stramineous above, with broad longitudinal lateral are brownish, the narrow venter brownish. Tegmina suffused with stramineous in the region of apical cells, where the veins are pale brown; the remainder of corium subhyaline with piceous veins. Frons shining and nearly smooth except apically, laterally with subobsolete indications of a few horizontal lateral grooves; clypeus subrugose. Supra-antennal arem of vertex foveate; frontal area of vertex depressed and with a delicate median carina, the true vertex rugose between the ocelli. Pronotum very coarsely reticulate-punctate except near fore margin, the reticula not sharp; the four finely sculptured depressed ares near anterior margin very small; median ridge of pronotum weak, becoming obsolete posteriorly. Body of scutellum as coarsely sculptured as pronotum but more irregularly; scutellar furrow XV, 1 Baker: The Malayan Machexrotine 73 cross wrinkled within, and with borders but little raised; lower posterior angle of posterior border of scutellum unusually prom- inent. Spine strongly curved, its point 1 millimeter above the apex of tegmina. Two females of this fine species were taken at Baguio, Moun- tain Province, Luzon, Philippine Islands. Tribe SIGMASOMINI Key to the genera. a’. Vertical projection’of head very long and sword-shaped, the head thus longer thar pronetim:3. ae ee Sigmasoma Schmidt. a’. Vertical projection of head shortly acute, the head much shorter than DIONOQUIN oc ee ee ee ee, Grypomacherota Schmidt. Genus SIGMASOMA Schmidt Sigmasoma bifaleata Schmidt. . Sigmasoma bifaleata Scumipt, Ent. Zeitg. Stettin 68 (1907) 181. This remarkable insect was described from Java, but similar forms will probably be discovered in other parts of Malaya also. Genus GRYPOMACHZEROTA Schmidt Schmidt 7 describes but one species of this genus, G. turbinata, crediting it to Java, Sumatra, and Borneo. Due to the very imperfect figures presented by him, and his very insufficient de- scription, it may be fairly questioned if all of his material belongs to one species. As shown hereinafter, some species which present an extraordinary similarity in color, are com- pletely distinct structurally. Schmidt’s failure to give full structural details makes it difficult to compare G. turbinata with the three species recorded herein from Borneo. Key to the species. a’. Medial cell about five times as long as wide and far longer than stem vein; anterolateral and posterolateral margins of pronotum deeply ineurved; length of frons above eyes greater than length below antenne; posteroinferior angle of scutellum high above lower margin and minutely acute; color pale, darker punctured; face pale with dark crossbars (Plate II, figs. 8, 9, and 11; III, figs. 1, 4, 5, and 8). G. borneensis sp. nov. a’. Medial cell little more than twice as long as wide or less, and shorter than stem vein; anterolateral and posterolateral margins of pronotum not deeply incurved; length of frons above eyes less than length below antennz; posteroinferior angle of scutellum in line with lower margin, and large, obtuse; color dark, head black. b*. Seutellar spine not strongly decurved, its apex far above apex of tegmina. ™ Ent. Zeitg. Stettin 68 (1907) 183. 74 Philippine Journal of Science 1919 c. Seutellar spine but slightly longer than scutellum; height of head above eye equal to depth of eye.................... G. turbinata Schmidt. c. Scutellar spine a half or nearly a half longer than scutellum; height of head above eye greater than depth of eye (Plate II, fig. 14; Pinte Thi fed. 2 ant 6} 2c G. tricolor sp. nov. b?. Scutellar spine strongly decurved, its apex near apex of tegmina; height of head above eye much less than depth of eye (Plate I, figs. 12 and 15; Plate Ill, figs. 8, 7, 9).............. G. breviceps sp. nov. Grypomacherota borneensis sp. nov. Female.—Length to end of abdomen, 5 millimeters; to ends of tegmina, 6.5; to end of spine, 7. Stramineous; lower part of face, sternum, and venter piceous, the latter sometimes paler; frons piceous basally and apically, the middle two-fourths stramineous, and each side with eight narrow oblique piceous stripes; vertex dark brown except the stramineous superior cheek arez, an area connecting ocelli, and the postocular aree. Punctures of pronotum brown, this ren- dering conspicuous the stramineous reticula, median carina,* and posterolateral borders. Scutellum and base and lower bor- der of spine pale brown; lateral pale stripes on body of scu- tellum straight, oblique, passing backward and upward to beyond middle; median dorsal line on body of scutellum and entire upper portion of spine except at base (where it is piceous), strami- neous; posterior margin of body of scutellum shortly and broadly yellowish and immediately above this piceous; the inferolateral border also narrowly yellowish. Tegmina slightly suffused with stramineous apically, the veins brown to piceous. Legs brown to piceous, apices of femora and nearly all of tibize stramineous. Frons finely, transversely, rugosely wrinkled and shining, more coarsely sculptured below and on clypeus; frons evenly convex on lower half, but with a strong median carina on upper half, this reaching the upper angle of appendage, where the lateral margins are also sharply carinate. Vertex horizontally rugose, with a median carina which apically becomes sharply raised and double; lateral aree of frontal portion of vertex concave. Pro- notum coarsely reticulate-punctate except near anterior border, the reticula not sharp; four small, depressed, oblique, finely sculptured aree near fore margin as in Macherota. Scutellum sculptured like pronotum, the furrow cross wrinkled and with the upper carina of spine scarcely entering it; spine at base with lateral disks sharply longitudinally depressed, its suiau 1.5 millimeters above apex of tegmina. This is a common species at Sandakan, British North Si cnate XV, 1 Baker: The Malayan Macherotinz 75 Grypomacherota tricolor sp. nov. Female.—Length to end of abdomen, 4 millimeters; to end of tegmina, 5.5; to end of spine, 6.5. Color black, with red-brown spine, short yellow lateral marks and narrow yellow posterior border of scutellum; tegmina with piceous veins, these apically smoky bordered. Frons shining, minutely, irregularly, transversely rugose, a sharp median carina only on upper fourth where the lateral margins of appendage are very shortly carinate; clypeus coarsely: rugose. Vertex strongly transversely wrinkled, with a short double median carina near apex; lateral aree of frontal portion of vertex convexly rounded. Pronotum thickly, very coarsely reticulate-punctate, more weakly near fore margin, the retic- ula sharp; a weak median carina on apical third only; the four small depressed arez near fore margin rather narrow and elongate. Scutellum sculptured like pronotum; the furrow short, broad apically, and very shallow; lateral surfaces of scutellum below furrow strongly depressed; the lateral depressions on base of spine very deep, foveate; apex of spine 2 millimeters above apex of tegmina. This very distinct species occurs at Sandakan, British North Borneo. Grypomacherota breviceps sp. nov. Length to end of abdomen, 3.5 millimeters; to ends of tegmina and of spine, 5. Black, with red-brown spine, short lateral yellow marks on seutellum and short yellowish posterior border; legs brownish; tegmina apically suffused with stramineous and with brownish veins; corium transparent, with piceous veins. Frons shining, finely transversely rugose, entirely without median carina, and with upper lateral carine of appendage very weak; clypeus more coarsely sculptured. Vertex finely trans- versely wrinkled and without median carina. Pronotum and scutellum sculptured asin G. tricolor; sides of scutellum below furrow not strongly or broadly depressed, the posterior sub- marginal impressed line in this species being a long, broad fovea; base of spine laterally not deeply foveate; the very strongly curved spine approaches ends of tegmina within a half millimeter, and does not pass the ends of tegmina, thus differ- ing from all other Malayan Machzrotini. Specimens of this species were collected on Penang Island, 76 Philippine Journal of Science 1919 Wellesley Province, Straits Settlements. One might be inclined to place this with G. tricolor on color alone, but its structure is very different. Tribe MAXUDEINI Genus MAXUDEA Schmidt From the viewpoint of species, the same difficulty pertains to this remarkably distinct genus as to Grypomachexrota. The fig- é d Fic. 1. Maxudea schmidtii sp. nov.; a, lateral view of head, pronotum, and scutellum; b, dor- sal view of head and pronotum; e, face; d, tegmen. ure and the description given by Schmidt are exceedingly crude. His lateral view of the insect is a quartering one with the head turned slightly away, so that it is impossible to secure exact XV, 1 Baker: The Malayan Macherotine 77 proportions of scutellum or of pronota!l lamine. The structure at the base of the scutellar spine as shown for the Sumatran species described by Schmidt,’ Maxudea crassiventris, is so en- tirely different from that of a species obtained from Penang, that it is necessary to maintain the latter as distinct, at least until better structural data as to M. crassiventris can be secured. Key to the species. a’. Lower margin of scutellar spine, above posteroinferior angle, strongly CONCAVGQ225 a8 eh ee ae eee mews, M. crassiventris Schmidt. a*. Lower margin of scutellar spine, above posteroinferior angle, straight. M. schmidtii sp. nov. Maxudea schmidtii sp. nov. Female.—Length to end of abdomen, 6.5 millimeters; to ends of tegmina, 7.5; to end of spine, 8; breadth from tip to tip of pronotal lamine, 5. Color largely piceous, the pronotal lamine shining. Face stramineous, a broad median stripe and lower third of frons and clypeus black; lateral portions of frons with about eight dark horizontal lines; vertex stramineous with dark sutures; the frontal horizontal lines continued to ocellar (true frontal) transverse suture. Anterior portion of pronotum stramineous, with four dark spots; disk of pronotum piceous, laminz black. Scutellar body stramineous with dark punctures, without yellow lateral marks or yellow posterior margin, the spine and infe- rolateral margins of scutellar body chocolate. Abdomen pic- eous, the segments with paler margins. Legs reddish brown, basal two-thirds of femora black. Tegmina suffused with stra- mineous, veins brownish, darker basally. Frons shining, irregularly shallowly roughened, the horizontal lines being distinct shallow furrows; clypeus coarsely trans- versely wrinkled. Frontal portion of vertex with an acutely triangular median depression just in front of ocellar suture. Pronotum thickly, coarsely reticulated punctate except near an- terior margin, the reticule and median carina sharp (stronger anteriorly) ; the four small depressed aree near fore margin are irregularly subcircular; anteriorly and posteriorly the flattened laminz are sharp margined; subparallel with anterior margin of lamina, and near to this margin, a sharp carina passes from disk of pronotum to tip of lamina, forming a suleus between * Ent. Zeitg. Stettin 68 (1907) 176. 78 Philippine Journal of Science this and margin. Scutellar body sculptured like pronotum, the area of scutellar furrow slightly depressed, without raised margins, the strong reticula here greatly lengthened horizontally to form a series of horizontal carinz; inferoposterior margin practically absent, the normal inferoposterior tooth of the other Macherotine adjoining the base of the spine. All pleur strongly rugose; seven tergites thickly, finely rugose-punctate; lower surface of femora bisulcate. Apex of spine 3 millimeters above ends of tegmina. A specimen of this unique insect was taken on Penang Island, Wellesley Province, Straits Settlements; this is the first record, belonging to the Asian mainland, for this genus. FIG, FIG. Ww bo or 10. 11. 12. 13. 14. 15. ILLUSTRATIONS - PLATE I . Machxrota notoceras Schmidt, lateral view of head, pronotum, and scutellum. . Macherota philippinensis sp. nov., face. . Macherota philippinensis sp. nov., lateral view of head, pronotum, and scutellum. . Machzrota ensifera Burm., lateral view of head, pronotum, and scutellum. . Machzxrota ensifera Burm., face. - Machxrota luzonensis Schmidt, lateral view of head, pronotum, and scutellum. . Machexrota fusca sp. noy., lateral view of head, pronotum, and scutellum. . Macherota fusca sp. nov., face. . Machxrota notoceras Schmidt, dorsal view of head and pronotum. . Macherota philippinensis sp. nov., dorsal view of head and pro- notum. Machexrota ensifera Burm., dorsal view of head and pronotum. . Machzrota luzonensis Schmidt, dorsal view of head and pronotum. . Machzxrota fusca sp. nov., dorsal view of head and pronotum. . Macherota ensifera Burm., posterior legs, inner and outer view. PLATE II . Macherota philippinensis sp. nov., tegmen. . Machexrota fusca sp. nov., hind wing. . Machzrota ensifera Burm., tegmen. Macherota ensifera Burm., hind wing. . Machzxrota luzonensis Schmidt, tegmen. . Machzxrota luzonensis Schmidt, tegmen, showing more normal venation than in fig. 3. Machzrota fusca sp. nov., tegmen. . Grypomacherota borneensis sp. nov., hind wing. . Grypomachezrota borneensis sp. nov., tegmen. Machzrota fusea sp. nov., upper surface of scutellum, with furrow. Grypomacherota borneensis sp. nov., upper surface of scutellum, with furrow. Grypomacherota breviceps sp. nov., tegmen. Machezrota luzonensis Schmidt, upper surface of scutellum, with furrow. : Grypomachexrota tricolor sp. nov., upper surface of scutellum, with furrow. Grypomachxrota breviceps sp. nov., upper surface of scutellum, with furrow. 79 80 Philippine Journal of Science Puate II Fic. 1. Grypomacheroia borneensis sp. nov., lateral view of head, pronotum, and scutellum. 2. Grypomacherota tricolor sp. nov., lateral view of head, pronotum, and scutellum. 8. Grypomacherota breviceps sp. nov., lateral view of head, pronotum, and scutellum: 4. Grypomacherota borneensis sp. nov., as seen when dorsum of pro- notum is horizontal. 5. Grypomacherota borneensis sp. nov., as seen when surface of vertex is nearly horizontal. 6. Grypomacherota tricolor sp. nov., as seen when surface of vertex is nearly horizontal. 7. Grypomachezrota breviceps sp. nov., as seen when surface of vertex is nearly horizontal. 8. Grypomachezrota borneensis sp. nov., face. 9. Grypomachzxrota breviceps sp. nov., face. TEXT FIGURE Fic. 1. Maxudea schmidtii sp. nov.; a, lateral view of head, pronotum, and scutellum; 6, dorsal view of head and pronotum; ¢, face; d, teg- men. BAKER: MALAYAN MACHEROTIN. ] (Puiute, Journ. Scr., XV, No. 1. PLATE |. MALAYAN MACHAROTINA. BAKER: MALAYAN MACHAROTIN, | [PuHiuip. Journ. Sci., XV, No. 1. : 8 PLATE Ill. MALAYAN MACHAROTINA. HABITS OF TROPICAL CRUSTACEA: III By R. P. CowLss Of the Department of Zodlogy, College of Liberal Arts, University of the Philippines, Manila ONE PLATE HABITS AND REACTIONS OF HERMIT CRABS ASSOCIATED WITH SEA ANEMONES The hermit crabs of the Philippine Islands form one of the most interesting groups of the tropical Crustacea on account of their peculiarities in structure and habits. Some of these hermits have an asymmetrical abdomen, of such a shape that the crab can insert it into the empty, spirally coiled shell of a mollusk and carry the shell about as a protection against ene- mies. Some hermit crabs of this kind spend practically all their lives in the sea among the rocks and the corals of rather deep water; some live close to the shore in shallow water; and others live far inland, visiting the sea only during the breeding period, at which time they usually exchange their old shells for new ones. Other hermit crabs have a straight, symme- trical or almost symmetrical abdomen; one of these, Pylocheles miersit Alcock and Anderson, lives in the sea and inhabits a joint of bamboo. Another kind, Birgus latro Linnzus, the coco- nut, or robber, crab, lives on land, carries no mollusk shell, and like the ordinary land hermits visits the sea only occasionally and possibly only at the breeding time. While the peculiar anatomy and fitness for the environment of the hermit crabs are of interest, certain habits of these Crustacea are even more interesting. The hermit begins life in the sea; and, if it is of the ordinary kind that carries a spirally coiled mollusk shell, it seeks and enters one of a size suitable to its body. As the crab grows older it increases in size, thus necessitating a change now and then to a larger shell. This habit is not a new one to zodlogists, for undoubtedly it has been observed many times in many parts of the world. While, ordinarily, a hermit occupies a shell of the right size for its body, it sometimes hap- pens that the shell is too small, so that a large part of the anterior portion of the body is exposed; or that the shell is so large that the crab can hardly be seen when it withdraws into the shell. 1663166 81 82 Philippine Journal of Science 1919 It seems probable, as Bouvier,! Thompson,? and Bohn* have held, that sometimes the change of shell is brought about spon- taneously and at other times because the crab is uncomfortable. An even more surprising habit which has been reported has to do with the association of hermit crabs with sea anemones. Eupagurus bernhardus Linneus ‘ carries on its shell the anemone Sagartia parasitica Gosse, while Hupargus prideauxit Leach,’ Pagurus arrosar Herbst,* and Paguristes oculatus Fabricius’ carry Adamsia palliata Bohadsch, Adamsia vondeletu Della Chiaje, and Actinia equina Linneus, respectively. Recently * Pagurus deformis H. Milne-Edwards from the Loyalty Islands and British New Guinea has been described as bearing sea anem- ones on its shell, and a search through the literature shows that other hermit crabs in a similar condition have been report- ed; but I shall not list these since their habits do not seem to have been investigated. The mere occurrence of sea anemones attached to the mollusk shells inhabited by hermit crabs does not close the list of the habits of these interesting animals. It has been reported ° that when the. hermit changes its shell it removes the sea anemones, * Bouvier, E. L., Observations sur les moeurs des pagures faites au laboratoire maritime de Saint-Vaast La-Houghe pendant le mois d’Aout 1891, Bull. Soc. Philomath. Paris 4 (1891-1892) 7. *Thompson, M. T., The metamorphosis of the hermit crab, Proc. Boston Soc. Nat. Hist. 31 (19038) 1938. * Bohn, G., Les animaux marines littoraux, Bull. l’Instit. Gén. Psychol. 6 (1903) 625-626. “Smith, Geoffrey, and Weldon, W. F. R., Cambridge Natural History, Crustacea and Arachnids 4 (1909) 172. * Wortley, Stuart, On the habits of Pagurus Prideauxii and Adamsia palliata, Ann. & Mag. Nat. Hist. Il] 12 (1863) 888-390; Brehm’s Thier- leben 10 (1900) 41; Bronn’s Thierreich (Ortmann) 5 (1901) 1254; Keller C., Das Leben des Meers (1895) 71; Calman, W. T., The Life of Crustacea (1911) 213; Agassiz, A., Instinct (?) in hermit crabs, Ann. & Mag. Nat. Hist. 17 (1875-1876) 100. * Brunelli, G., Atti. Rend. R. Accad. Lincei. Roma 19 (1910) 177-82. ‘Brunelli, G., loc. cit. * Borradaille, L. A., On Stomatopoda and Macrura brought by Dr. Willey from the South Sea Islands, Willey Zodlogical Results pt. 4 (1900) 424. * Wortley, Stuart, op. cit.; Gosse, P. H., On the transfer of Adamsia palliata from shell to shell, The Zoologist 18 (1859) 6582; Sluiter, C. P., Der Commensalismus zwischen Hinsiedlerkrebsen und Actinien, Kosmos (1882) 391; Chevreux, M. E., Le Pagurus Prideauxii et ses Commensaur, Compt. Rend. Assoc. Ffane. Avancem. Sci. (1884); Brehm’s Thierleben 10 (1900) 41; Brunelli, G., op. cit. se XV, 1 Cowles: Habits of Tropical Crustacea — 83 placing them on the new shell, and that the sea anemones aid the hermit in this procedure.’ Some zodlogical text books mention the fact that this trans- planting of anemones has been reported; but the subject, if dis- cussed, is generally treated, whether intentionally or not, as though there might be an element of doubt as to the occurrence of this habit. For the reasons mentioned, and because five of my coworkers and I have seen the transfer for the first time in the case of two species of hermits living in Philippine waters, I shall now record my observations. Along the coast of Mindoro Island, in Port Galera Bay, there occur hermit crabs belonging to two species, Pagurus deformis and Pagurus asper de Haan, which inhabit mollusk shells of Do- lide, Strombide, Cassis, etc. Both of these hermits almost in- variably carry two different kinds of sea anemones on their shells; one, a large grayish brown form usually on the sides and another, much smaller, almost colorless form usually on the underside of the mollusk shell below the protruding head of the crab (Plate I). As many as eight of the large anemones and three or four of the smaller kind have been found on a single Dolium shell, but so great a number is rather unusual. While in the majority of cases the small anemones are attached on the underside of the mollusk shell below the head of the hermit and the large ones on the upper and lateral surface of the shell, occasionally specimens are found in which the order is reversed. Either the anemones when in the larval condition attach themselves to the mollusk shell carried by the hermit, or they are placed there by the hermit when they are still very small. On one occasion a beautiful, clean shell which was not beach worn was collected containing a young specimen of Pagurus deformis. On the upper surface of this shell were three of the brown anemones, 2, 4, and 8 millimeters across, and just inside the mouth of the shell was one of the colorless species, 3 milli- meters in breadth. The extreme smallness of the anemones in this case might indicate that they became attached to the mollusk shells in the larval stage, and this is borne out by the fact that in all our collections at Port Galera we have never found these anemones except on shells inhabited by hermits; but since ” Risig, H., Studien iiber tiergeographische und verwandte Erschein- ungen, I. Zum Verstiindniss des Commensalismus der Hinsiedlerkrebse (Pa- guriden) und Seeanemonen (Aktinen), Das Aussland (1882) 55, 681; Gosse, P. H., op. cit. RA Philippine Journal of Science 1919 the hermits show considerable facility in transferring the anem- ones, it may be that they detach very young ones from rocks in deep water unexplored by us. On one occasion three young hermits (Pagurus deformis) were brought in by collectors, and the shells that they occupied were devoid of anemones. Wortley “ has reported that the hermit crab when feeding may transfer pieces of food with its pincers to the sea anemone, but this behavior has not been observed by me nor, so far as I know, by other workers. In fact the accuracy of Wortley’s observation has been doubted.'? Hermits of both of the species of which I collected specimens at Port Galera usually frequent shallow water where there is a rather heavy growth of eel grass which harbors many small organisms. While it does not seem probable that the hermits feed the sea anemones, yet it must be admitted that the anemones are benefited by associa- tion with them in that the active, wandering life of the hermit crab presents an ever-changing feeding ground. The small colorless anemones attached at the shell’s mouth and located directly under the head of the hermit crab undoubtedly obtain food from the hermit crab, but I have never seen them feeding (Plate I). It need hardly be pointed out that the sea anemones afiord protection for the hermit crab since, in addition to forming a covering which hides the latter, they discharge great numbers of stinging cells when disturbed by an enemy. The crab itself would undoubtedly be a choice morsel for large fishes such as sharks, but it has been pointed out that one mouthful of hermit crabs carrying sea anemones flavored with the piquant stinging . cells would satisfy a fish for some time. As soon as our collectors at Port Galera began to bring in hermits with anemones on their borrowed shells, I was anxious to see if the former would change their shells and transplant the anemones, as had been reported for a few other species in European waters. Accordingly I placed several individuals in an aquarium with a number of clean, empty Doliuwm shells. Nothing of interest happened for some time, but after about two hours one of the hermit crabs became active, took hold of an empty shell, withdrew its abdomen from the shell it was occupying, and inserted its abdomen, almost without examination, into the new shell. In order to have a better chance to observe what might happen, the old shell bearing the anemones and the hermit carrying its new, bare shell were placed in another aquarium, The hermit showed no interest in the anemones * Wortley, S., op. cit. " Hisig, H., op. cit. XV, 1 Cowles: Habits of Tropical Crustacea R5 even after three-quarters of an hour, so after giving the hermit every encouragement such as placing the old shell near it, etc., I came to the conclusion that the species under consideration differs in this respect from the European species. But further observation with other specimens led me to believe that the removal of the hermit from one aquarium to another just after it had vacated the anemone-covered shell disturbed the hermit so much that the usually instinctive behavior in which the anemones are transferred to the new habitation was inhibited. I shall now describe my observations of a case in which the behavior of Pagurus deformis during its change of habitation may be called normal although inaccurate. In this case a hermit, in a shell carrying two large, grayish brown anemones on the upper surface and a small white one in the mouth of the shell, was placed near a Doliwm shell somewhat larger than the one it was occupying. The hermit began to examine the new shell almost immediately ; it put its pincers and some of its legs inside of the shell and remained in this position for a consider- able length of time. Occasionally the pincers were moved about and were used apparently to examine the outside of the shell. Then, suddenly, the abdomen was inserted into the new shell, and the hermit without moving away grasped one of the large anemones, clawing, pinching, and pulling it. The anemone, instead of contracting as it would if disturbed by the observer, remained expanded even though the mauling it received was really very strenuous and although it was frequently jammed against the newly occupied shell. After these movements had continued for at least ten minutes the base of the anemone be- came loose in several places. Instead of completing its work at once, the hermit began to pinch and claw the other large anemone which withdrew its tentacles, soon however protruding them again. Then suddenly the hermit left the second one, and picked off the small white one which it applied at the mouth of the new shell almost directly under its own head, where the small anemone soon became firmly attached. Then the crab again attacked one of the large anemones, removed it easily, turned it round and round between its legs and pincers, and finally pushed the base against the side of the new shell. The anemone failed to become attached, slipped down onto the her- mit’s legs, stuck to them for a moment, dropped off, and finally attached itself to the glass bottom of the aquarium. Again the hermit crab seemed to lose interest in the anemone upon which it was working, and began tugging, pinching, and scratching the only one remaining attached to its old shell. After a long 86 Philippine Journal of Science 1919 period of handling and some little prying up of the edges of the base of the anemone, during which time the tentacles contracted completely, the anemone suddenly slipped off, apparently of its own accord. The hermit then caught the anemone which had just slipped off, rolled it around and around for a few sec- onds and then, with the aid of the hind legs, applied the tentacle side against the shell. Again the anemone slipped off and was disregarded for a moment while the hermit picked up the other anemone. Still holding the latter it lifted up the one that had just slipped off and rolled them both around for some little time. Finally one became attached to the side of the shell by one edge of the base and was left to shift for itself. The other was passed up on the opposite side of the shell with the aid of the posterior legs of the hermit and soon both anemones became firmly attached with their tentacles beautifully expanded. Strange as it may appear, the crab never seems to cause any injury, although the base of the anemone is very thin and when taken off by man is easily torn. While the actions of this hermit crab were not nearly so accu- rate as those of some others I have observed, yet like all of them it showed a constant and hurried activity which seemed to indi- cate that some inherited nervous condition existed which was directing its actions. Furthermore, during the time in which this remarkable behavior was taking place the hermit crab, although at other times reacting quickly to movements outside of the aquarium, showed almost no reaction to the stimuli ordi- narily produced by people passing or moving in front of the aquarium. After the hermit had transferred all of the anemones from its old shell it began to work on the anemones of a small shell in- habited by another hermit crab. The latter immediately with- drew into its shell but left the large pincer exposed and open, so that one of the legs of the intruder slipped into the trap and was caught. The hermit, thus held fast, discontinued its attack for a time; in fact, until the captor loosened its hold and allowed the leg to be withdrawn. Instead of leaving, however, the at- tacking hermit renewed the scratching, pinching, and pulling until it was caught again by the leg. The intruder was held in this way for about a*half hour during which it was inactive; but as soon as the leg was released the mauling of the anemones -began again. These activities were repeated several times, but as long as I watched the two hermit crabs no anemone was removed. et seh eR ig agai = ¥ & p i \ E , £¥, + Cowles: Habits of Tropical Crustacea 87 I have observed other hermit crabs of the species Pagurus de- formis which were even more inaccurate in their actions than the one we have just considered. In one case a hermit changed to an empty shell, worked on the anemones of the old shell for a few moments, and then took up its habitation again in the old Shell. Soon, however, it inserted its abdomen again into the new shell and after about an hour’s work succeeded in loosening an anemone. This it lifted upside down to the upper surface in such a manner that it became attached to the apex of the shell merely by one edge of the base where it hung with the tentacles hanging downward. Soon, however, partly as the result of the hermit moving about and turning its shell, the anemone became firmly attached in its normal position with the tentacles directed upward. About an hour later the hermit succeeded in removing another large anemone from its old shell, but instead of setting it on top the hermit placed it in the mouth of the new shell where the smaller species of anemone is usually put. The transfer of the latter (there were two of these) was not seen. Observations on this hermit were discontinued until the next morning when I found that both of the large anemones were in their usual po- sitions on the back of the new shell, the one formerly placed at the mouth either having been transplanted by the hermit during the night or having moved of its own accord to the outside of the shell. One of the small anemones had been removed from the old habitation and apparently an attempt had been made to attach it at the mouth of the shell; but the transfer had been inaccurate, so that the anemone had become attached at the base of the walking legs of the crab near the mouth. The other small anemone was still attached to the old shell. In the case of the species of hermit crab Pagurus asper the transfer of anemones from one shell to another was accomplished in practically the same manner as in the case of Pagurus defor- mis, except that the movements were slower. My observations on the active and speedy shore crab Ocypoda arenaria Catesby “ and the experiments of Yerkes and Huggins * with the crayfish show that, in the case of the former, past ex- periences leave an impress at least on the nervous system, and that the crayfish is able to solve very simple problems, such as * Cowles, R. P., Habits, reactions, and associations in Ocypoda arena- ria, Papers Tortugas Lab. Carnegie Inst. Washington 2 (1908) 34. * Yerkes, R. M., and Huggins, G. E., Habit formation in the crawfish Cambarus affinis, Harvard Psychological Studies 1 (1903) 565. 88 Philippine Journal of Science finding its way out of a very simple labyrinth with greater and greater accuracy as the number of trials is increased. Such a “mind” is probably possessed by the hermit crab, but it would require a great stretch of imagination to believe that the hermit crab, if it really feeds the anemone, does so with the knowledge that it is caring for an animal which protects it from enemies; or that, when the hermit crab removes the sea anemones from a shell which it has left and plants them on a new home, it knows that they will be of future use; or that, when the anemone allows itself to be pinched and pulled and pried away from a shell by a hermit, it knows ‘that it is in the hands of a friend. We cannot believe that this lowly crustacean, during its lifetime, has learned by experience that its care of the sea anemone is advantageous, although we know that crabs in general do profit by experience. Yet, assuming that the remarkable behavior of the hermit is due to instinct—that is, to an “inherited combination of reflexes’ which have been so brought together by the nervous system that the behavior has become fixed and adaptive in the species—it is extremely difficult to conceive how it has acquired these habits. ILLUSTRATIONS PLATE I Fig. 1. A hermit crab, Pagurus deformis Edw., carrying large and small sea anemones on its shell. Drawn from a preserved specimen from Port Galera, Mindoro. Slightly less than natural size. 2. Hermit crabs, showing large and small sea anemones attached to the shells. From a photograph of living specimens. 89 COWLES: TROPICAL CRUSTACEA: III.] [PHILip. Journ. Sci., XV, No. 1. Fig. 1. A hermit crab, carrying large and small sea anemones on its shell. Fig. 2. Hermit crabs with sea anemones attached to the shells. PLATE I. MILK PRODUCED IN SOUTHERN CHINA’? By C. O. LEVINE Of Canton Christian College ONE PLATE Through investigations begun in November, 1916, some inter- esting data have been secured in regard to the milking qualities of native and European cattle in southern China. Three classes of milk animals have been studied: European cattle, the native water buffalo (known in the Philippine Islands as carabao), and the native, humped cattle. The work has been done chiefly in the vicinities of Canton and Hongkong, the only regions in Kwangtung (which is the southernmost province in China) where dairying has as yet developed into an industry of any extent. European cows, or foreign cows as the Chinese call them, are the most popular for dairy purposes. In the dairies of Canton and Hongkong, which utilize approximately 1,600 cows, about 1,200 are European. The European cows have been imported from Australia, America, and England, or are the offspring of such imported cows. In the various dairies may be found Short- horns, Holsteins, Ayrshires, Guernseys, Jerseys, and crosses among these breeds and with the native, humped cows. There are a few pure-bred cows. The Shorthorns, which are the com- monest, are chiefly of the white, hornless variety that have been imported from Australia. European cows are never allowed to graze because of the presence everywhere in the grass of the fever tick. They readily contract fever when exposed to it, and the disease usually proves fatal. However, European cattle born in southern China readily become immune to the fever, as they do in Texas where fever is common. Im spite of the greater labor and risk experienced in keeping European cows, they are preferred to the native cows because of the larger amount of milk given and their efficiency, as compared with native cattle, in converting feed into milk. A good description of the water buffalo (Bubalus bubalis Lyd.), ‘ Some of the data contained in this paper have been published in bulletin 17, Canton Christian College, Canton, China. o 92 Philippine Journal of Science 1919 of southern China, is given by Bailey.?, The buffalo is princi- pally used for draft purposes. Old animals are slaughtered for beef. In Canton, Hongkong, and other cities of southern China an increasing number is kept for the rich milk produced. There are no special breeds of dairy buffalo in southern China as there are in India. Buffaloes kept for milk are usually mature cows that have been used as draft animals for some time. During the period a cow is milked she is not required to work in that capac- ity. When milked, and fed grain and good grass, the cows have a long lactation period. It is not uncommon for them to give milk for from eight to twelve or more months. While the amount of milk is very low—usually from 3 to 15 pounds a day (about 1.4 to 6.8 kilograms), with an average of about 4.5 pounds (about 2 kilograms) a day for the lactation period—it is very rich in fat, containing from three to four times as much fat as European cows’ milk. GHstruation in the female buffalo does not occur, as a rule, until the heifer is two years old. It occurs one month after par- turition, and reoccurs regularly every twenty-eight to thirty days until the animal again becomes pregnant. I have secured definite records on the exact length of the gesta- tion period with only two cows; in one case it was three hundred ten days, and in the other, three hundred fourteen. In order to secure a long lactation period as well as the max- imum amount of milk, buffalo cows kept for milk are usually not bred until three or four months after freshening. The cows are considered profitable for dairy purposes until they are about 15 years old. The Chinese in South China call the native cow wong ngan, “yellow cow.” This bovine is a variety of the humped species of cattle (Bos indicus) common in the Orient. The hump is much less pronounced than it is in most breeds of Indian cattle. In the males the hump is usually 6 to 8 inches high above the shoulders. It is much smaller in the females than in the males. The dewlap is large, but is not developed to the degree common in Indian cattle. In color, these native yellow cattle are similar to the Jerseys. They vary from yellow-red to brown-black. Many are brindle. There are no white, and very few spotted, individuals. The tongue, nostrils, and teats are black. The cream-colored ring above the nostrils in the Jersey is also a characteristic of these cows. Males weigh from 800 to 1,000 pounds (about 362 to 454 kilograms). Mature females weigh * Cyclopedia of American Agriculture 3 (1908-09) 292. Ty XV, 1 Levine: Milk Produced in Southern China 93 from 600 to 800 pounds (about 272 to 362 kilograms). Their milk is considerably richer in fat than is that of any European breed, though not so rich as is the buffalo milk. The amount of milk given is usually about the same as that given by the buffalo cows, or a little less. They have a full, deep quarter and a deep layer of meat on the loin and back. They are used chiefiy for draft and beef purposes. They are gentle, and much easier to handle than are the buffaloes. Very few are milked. MILK ANALYSES In making the fat analyses of milk a Babcock fat-testing outfit was used. The proteins were determined by the Kjeldahl method described by Hawk.* The total solids were determined by evaporating a weighed sample of milk on a steam bath until the weight became constant. The ash was determined by heat- ing the evaporated total solids over a gas flame until the weight became constant. The amount of sugar was found by sub- tracting the sum of the fat, ash, and proteids from the total solids. The percentage of each was found by dividing the weight of the final product by the weight of the sample of milk analyzed. TABLE I.—Showing analyses of Canton buffaloes’ milk, European cows’ milk in Canton, European cows’ milk in America, and native, yellow cows’ milk. [Numbers give percentages. ] ee oe —o | European European’ ; Canton Yellow | Constituent. buffalo. Cant Pic cow. Aes Dee aera ee ey Saree, Mies { | Boe Sine eater ie oe ae See eee 12.60! 3.80 3.69! 8.00 Po Pebteine cs. 65 ee Ne ee ce 6.04 3.23 8.53 | Meade mee | pS a ik Ae se ee en re eee hee 3.70 5.96 re geo oe A eee i ee 6) eT BM fe MNS Fan ssi er cass Tan sues ensure ee eainn .90| O20}. ST.37 1x oe | pS ESS EE SS mR ewe i oe aed in BA 23.20| 18.90) 12,25 |... All the samples of milk analyzed were taken by me directly from the barn as each cow was milked; not from bottled milk that is sold to the public and is frequently diluted with water. The analyses show that European cows’ milk is practicaily the same as when produced in the countries from which the cows have come. With more analyses the slight differences in some of the constituents will probably prove to be still less. Sixty duplicate analyses of fat, or one hundred twenty in all, from the ° Practical Physiological Chemistry. P. Blakiston’s Son & Co., Phila- delphia (1916) 483. 94 Philippine Journal of Science 1919 milk of twelve cows, extending over a period of eighteen months, were made. Ten analyses were made for each of the remaining constituents. The analyses of buffalo milk included four hundred duplicate analyses for fat in a herd of fifty buffalo cows kept in the Canton Christian College dairy, including both individual and herd tests, and herd tests of five dairies in Canton. Compound analyses of both morning and afternoon milk were also made. The average was found to be 12.60 per cent fat. The lowest test was 9.80 per cent. The lowest average for a lactation period of all the cows was 9.65 per cent. The highest average was 15.60 per cent. The percentages of constituents in the buf- falo milk other than fat represent averages of ten analyses. The protein varied from 5.60 to 6.10, with an average of 6.04 per cent. Sugar was low, varying from 3.51 to 3.75, with an average of 3.70 per cent. The average for ash was 0.86, varying from 0.71 to 0.99 per cent. The total solids averaged 23.20, varying from 21.00 to 25.20 per cent. Tables IT and III give the analyses of milk for fat, and Tables IV to XVII the production records of buffalo cows for which we have records extending over a period of several months, or for entire lactation periods. The analyses for butter fat were made twice a month. The milk for twenty-four hours was weighed twice a month. The average of the two analyses was taken as the average for the month, and the average amount of milk at the two weighings was taken as the average daily production for the month. The cows were all in the Canton TABLE II.—Complete analyses of buffalo milk. {The samples analyzed were composite samples taken from twelve cows in the college dairy in November, 1917. The milking in the morning was begun at 4 o’clock, and in the afternoon, at 2 o’clock. Numbers give percentages.] ae Se ee eee eee Sample No. Time. Fat. Ash. | Protein. | Sugar. — Water. 1 Eepeareine SRS Aen PERE “2 ee con 11.60 0.94 §, 04 4.00 22. 98 17. 02 _ RST NaS i. View. 12. 80 0.90 6.10 3.57 23.37} 76.63 | 3 ee be aecis: 08 0.71 5.71 3.32 23.02} 76.98 se ee ee See 0.74 5.94 3.57 22.88] 77.08 STE st SNe eee OAR OO Spano SM 30 0.90 6.14 $.83{ 22.87| 77.04 | peice Pas a BAS * ES 14. 00 0. 92 6.42 3.87 25.21 | 74.79 | Ysa a amt He 0.95}. 5.80 3.70 22.95 | 77.05 | Sterne tas eer ...do....| 12.00] 0.98] 6.00} 8.60; 22.54] 77.45 | , ERC nee mammeee ere! Bes (yee 12.34 1.04 6.28 3.71 23.871 76.63 | We ee Leet 12.90 0.77; 5.90 4.23} 23.02! 76.98 | nag tT ec er Nae 12. 46 0.88; 6.04 8.70 | 28.08 | ay a a XV, 1 Levine: Milk Produced in Southern China 95 Christian College dairy at the time that the tests and the records were made. The grain fed consisted of a ration made up of 2 pounds wheat bran and 1 pound rice chop. Each of the cows was fed about 1 pound of this mixed feed a day for each pound of milk given. The chop was fed cooked, and the bran was added to the rice after the rice was cooked. This feed was fed separately TABLE III.—Complete analyses of European cows’ milk. Analyses made during January and February, 1917. [Numbers give percentages.] | eat Sample No.4 Fat. Ash. | Protein. Sugar. feb a3 Water. ee A ek eee 3.20} 0.745 3.18 6.47; 13.70] 96.30 | NS Fae IAS Oe BERD eee We 3.50 0.840 3.20 5.28 13.82] 86.18 | 96 ee na 4.50] 0.736 8.24 5.18| 14.65] 85.35 | $e ere 4.60} 0.900 4.10 5.20} 13.80] 86.20 | $i Sk iade cee aes 3.80] 0.780} 3.25 5.64; 13.37] 86.63 | Gos isishinictarie inne Selgsawn case 4.44] 0.800 3.10 5.60] 18.94] 86.06 | RMR Se Dee Oe eT 3.501 0.880 3.80 6.00} 13.97] 86.03 | Cs ee ee 4.10} 0.888 3.40 5.63] 14.12! $5.88 | Ee EIN Cycle Oe ee 5.20! 0.740 3.20 5.30} 14.44] 85.56 i nS re 4.00} 0.800 3.05 6.10} 18.95} 86.05 | Avetiiges =.2052 snk 3.96| 0.820, 3.23| 5.96] 18.90; 86.21 | i 4 Samples 1 and 2 are of milk from the Hongkong Dairy Farm Co., in Victoria, Hongkong. The samples are from bottled milk from a herd of from 500 to 700 European cows, most of which are Holsteins, Samples 3, 4, 5, 6, and 7 are from the mixed milk of the Sinkee Dairy, in Canton. In this dairy 90 to 100 foreign cows are kept, Guernseys and Shorthorns pre- dominating. Samples 8, 9, and 10 are from the mixed milk of four cows in the Canton Christian College dairy. These four cows are of mixed breeding with Shorthorn blood predominating. TABLE IV.—Milk produced by buffalo cow 1. Average fat for six months, 12.23 per cent. [Freshened in August, 1916. Lactation period ended June 30, 1917. Records began January 1, 1917.] Morning. Afternoon. | For 24 hours. Total for month.} Aver- Month. Milk.| Fat. | Milk.| Fat. | Milk./ Fat. | ** | min. | Fat. lbs. | P.ct.| lbs. | P.ct.| lbs. | lbs. | P.et.| lbs. lbs. PRUNE eect aweos sei dei oo 2.20 | 10.24} 2.00 | 12.00; 4.20) 0.484} 11.5 | 180.2 | 14.97 ORK WNNe te 2.00} 10.20; 2.30 | 12.90} 4.30/ 0.503} 11.9 | 120.4 | 12.82 March SE Sani te 1.50 | 11.20; 1.80 | 13.80} 2.80 | 0.846} 13.8) 84.0 | 11.21 fo Sen cE ES Sek eee 2.00 | 12.50) 1.50 | 13.50} $.50/ 0.450} 13.0} 105.0 | 13.65 WIS cc ea eee 1.70 | 12.50} 1.20] 14.50| 2.90; 0.386) 12.6; 89.9 | 11.95 SUNG 25. es 0.50 | 12.00} 0.25 | 14.00] 0.75 | 0.100; 13.3 22.50] 2.99 Totals for six months___|_______ SRE, ceo ir) ELGG, Reine Beene us Danie Cee 552.0 | 67.59 96 Philippine Journal of Science 1919 to each cow twice a day. About 0.75 ounce of salt was mixed with the grain at each feeding. About 40 pounds of water were added to the grain at each feeding, making a very wet feed, the cows drinking it. This is the usual method of feeding grain to cattle in southern China. The cows were fed four times a day all they could consume of a mixture of green grasses, cut along the canal dikes or from the college campus. The amount of grass eaten daily by each cow varied from 40 to 60 pounds. The milk of the native yellow cow is rich in fat, averaging about 8 per cent. No analysis has been made of this milk for constituents other than fat. TABLE V.—Milk produced by buffalo cow 2. Average fat for lactation period, 11.80 per cent. {Freshened January 28, 1917. Lactation period ended September 30, 1917.] Morning. Afternoon. | For 24 hours. ebues Total for month. Month. age Milk. | Fat. | Milk.| Fat. | Milk.| Fat. | 4% | Mik. | Fat. lbs. | P.ct.| lbs. | Pct. | lbs. tha;--:|- Px ot; lbs. lbs. WOpruar yr 285026 ce eet 8.8 2.3 13.8; 5.0 | 0.541) 10.9 140 15. 26 MERE os ee 2:7 9.0 pes 13.0; 4.0 | 0.474 | 10.1 124 13. 64 © XS | SRST Sas Sanam ae a ene ees ITN 3.3 | 10.0 2.8 15.1; 6.1 | 0.758 | 12.2 183 22.57 MRSS ce. ot 2.0 9.0 19 18.1} 4.1 | 0.447 | 10.8 127 18.72 SUMO ss ee 2.75 |. 9.00 | 2.25 18.5 | 6.0 | 0.451 | 9.0 150 13.50 Palys a Se 2.5 | 10.5 2.2 12.6; 4.7 | 0.539 | 11.5 148 16.75 ROGUEE ee 2.6 5149 1.5 14.2) 3.5 | 0.482 | 12.2 109 18. 45 Bepsebiber eo gh eee eS 83 5 es 14.4; 3.0 | 0.872 | 12.4 90 11.16 Totals for eight months'_______ | Se | At Assis Sao ge Peano Raat | Ro eee 1, O71 120.05 j i TABLE VI—Mulk produced by buffalo cow 8. Average fat for six months, 12.95 per cent. {Freshened October 20, 1916. Lactation period ended June 28, 1917. Records began January 1; 1917.) = pales: | Morning. Afternoon. | For 24 hours. Peet Total for month. H er- Month. age Milk. | Fat. | Milk.| Fat. | Milk.| Fat. | 8% | min. | Fat. lbs. | P.ct.| lbs. | P.ct.| lbs. | lbs. | Pct.| Ibs. lbs. Senay i 2.0 | 11.8 2.7 | 183.8 4.7 | 0.606 | 12.38 145.7 | 17.45 MOREE Oo a 2.3 | 12.0 15 | 14.0 3.8 | 0.486 | 12.6 106.4 | 13,40 REC OS aie 2.3 | 13.2 1.1 | 16.4 3.4 | 0.478 | 14.0 105.4 | 14.76 Fy | Senos Ce io Aen ees 2.1 112.2 | L2 |14.2 | 8.3 | 0.436] 13.3 99.0 | 13.16 WN os ee 1.9 | 13.5 13 | 14.6 8.2 | 0.444 | 18.90 99.2 | 18.79 SORE Se Lo {12:2 0.75 | 18.0 2.65 | 0.322 | 12.2 79.5 | 9.70 Totals for six months __.|_______ ass eer careers eee ee eee | 635.2 | 82,26 —— =e i oe ee I es +4 at XV, 1 Levine: Milk Produced in Southern China 97 In all regions where buffalo milk has been analyzed it has been found to contain a great deal more fat than does European cows’ milk. However, in no region from which the analysis of milk has been reported does the buffalo milk contain as much fat as in southern China, A breed of buffaloes known as tee Delhi buffalo (known in the Philippine Islands as the ram’s horn buffalo) from North Central India has been introduced into Hongkong. This is an excellent breed for dairy purposes, because of the amount and quality of milk given. According to Hongkong authorities the milk contains about the same amount of fat as does the native Chinese buffalo milk. One of these Delhi buffalo dairies at TABLE VII.—Milk produced by buffalo cow 4. Average fat for lactation period, 12.63 per cent. {Freshened January 2, 1917. Lactation period ended November 30, 1917.] Morning. Afternoon. | For 24 hours. pom Total for month. Month. age Milk.| Fat. | Milk.| Fat. | Milk.} Fat.| f | min. | Fat. ‘Ibs. | P.ct.| lbs. | P. et.) lbs. lbs. | P. et. lbs, lbs. AORURET SG ie se a4: AL 2.2| 14.2 4.6 | 0.581 | 12.6 133.4 | 16.80 PODCUEES: dic ce cees— oe 2.8.} -12.0 2.0} 15.6 4.8 | 0.648 | 13.3 134.4 | 17.87 March ice ss eee $.2{ 11.8 2.4; 14.1 5.6 | 0.715 | 12,8 173.6 | 22.22 POPE oss siesta 2.6; 11.0 2.11 14.5 4.6 ; 0.579 | 12.6 138.0; 17.39 epee Ie Regen 2.4) 11.4 2.0; 14.8 4.4} 0.570) 12.4 182.0} 16.38 WONG os 68s oe eas Coe 2.2} 11.0 1.8] 16.0 4.0 | 0.514; 12.8 120.0 | 15.16 Do)» eer ener) aes es BCe oy ee 2.2; 12.5 1.3 2 16.0 3.5 | 0.438 | 13.8 108.0; 14.90 Mma tt a sxc ck 5 ch see teeta 1.9) 10.2 2451 2380 3.0 | 0.3385; 11.0 93.0} 10.23 September ee 23°) 110 1.5; 148 3.8 | 0.475 | 12.5 114.0; 14.25 WCTOUGE csc eet. - uae 2.61 11.6 12) 16.8 8.8 | 0.483 | 12.4 f 1144.0! 14.11 November. ---.--..-- aaa veo i L&{ 10.2 1.0 12.5 2.4: 0.278} I1L1 72.0 8.00 Totals for eleven | i mit eee egies ee fog aah aries lvosise Seas | 1,382.4 | 167.315 TABLE VIII.—Miulk produced by buffalo cow 5. Average fat for five months, 13.12 per cent. [Freshened August 10, 1916. Lactation period ended June 30, 1917. Records began January, 1917.) | Morning. Afternoon. | For 24 hours. Ai. Total for month. f: pee: mitk.| Fat. | Mik.| Fat. | Min.| Fat. | ** | wine. | Fat. ibs. | P.ct.| tbs. | P.ct.; lbs. Iba. | P.ct.| lbs. tba. LAS aie Bia ene 2.7| 11.25} 2.0 12.70) 4.7| 0.587 | 11.87| 145.7 | 17.29 TOURS 3 cnc tia 2.3/ 12.50; 2.0) 18.80) 4.3) 0.494) 11.54] 120.0) 13.68 Mate ee 2.3| 13.50| 1.1/ 15.50} 9.4| 0.481 | 14.15| 105.4 | 15.31 POS SUED TIES: 2.0/ 14.00] 1.5| 15.00) 3.5 0.505] 14.40; 105.0 | 18.12 i i a Rae | 19/1s.90/ 13/1510! 3.2/ 0.460| 1490) 99.2 | 14.18 | 576.8 | 75.58 Totals for five months -_/-_..--- eee ENS eee eee pcre grremts 166316——7 98 Philippine Journal of Science 1919 TABLE IX.—Milk produced by buffalo cow 6. Average fat for six months, 12.34 per cent. [Freshened August 2, 1916. Lactation wT June 24, 1917. Records began January 1917. Morning. Afternoon. | For 24 hours. ais Total for month. Month. | age << | Milk. | Fat. | Milk.| Fat. | Milk.| Fat. | f* | mim. | Fat. | —veteneni — —- eaneapatone = lbs. } P.ct.| lbs. | P. et. | lbs. lbs. | P. ct. | iba. 7 ea | danvery 5 2:2 |-9:0 1.7{| 12.9) 3.9| 0.414; 10.6 | 120.9} 12.81 Pebrisrys, oo es 2.4} 10.0 1.7] 141] 4.110.480] 11.7 | 114.8] 13.41 | Wereh. e 1.9 | 13.0 1.4] 16.3; 8.8 | 0.485 / 18.2 | 102.1 | 13.47 Wp ee 2.0 | 10.5 1.4] 18.9; 38.4] 0.425] 11.9 | 102.0 12.14 Wilts iste sages he 1.2 esd 0.9} 12.5) 2.1) 0,268 | 12.88) 68.1) 8.42 5 ao SRO OOS Re oR rae ueig 12/102 | 12] 125] 2.4] 0.282! 11.69] 72.0] 8.41 Totals for six months...|.......|___-._- pra MER Rees Bees ah cote | 574.9 | 68.66 { i | | ! TABLE X.—Milk produced by buffalo cow 7. Average fat for five months, 13.82 per cent. (Freshened July 4, 1916. Lactation period ended May $1, 1917. Records began January 1, 1917.] Morning. Afternoon. For 24 hours. eis Total for month. Month. age |——-——-———-- Milk.! Fat. | Milk.| Fat. | Milk. | Fat. fat. | wink. | Fat. | lbs. | P.ct.| lbs. | P. ct.| lbs. | lbs. | P.ct.| lbs. lbs. ee ei eae el aan 2.2} 122| 1.2] 186] 8.4/0.482| 12.84] 92.0] 11.81 | WODERENP os so oo £73250 1.5; 15.2 3.2 | 0.482 | 18.2 89.6 | 11.83 Mabe Se ee 1.3 | 12.0 2.0; 15.3 3.2 | 0.464 | 14.1 118.0 | 16.64 Fes) SRS, Senies anne eee See 1.8] 18.1 2.1; 16.0} 8.9] 0.441 | 14.7 107.0 | 15.73 Wine. SS ec 0.6) 160 1 a 0.6 | 0.096 | 16.0 18.6 | 2.87 Totals for five months __|_____-_ TEEN Sees | RRO ee amen eeuek | eee 425.2 | 58.88 TABLE XI.—Milk produced by buffalo cow 8. Average butter fat for five months, 15.48 per cent. {Freshened May 2, 1916. Lactation period ended May 31, 1917. Records began January 1, 1917.] ' Morning. Afternoon. | Por 24 hours. | Eee tietaitoninaieh: Month. [Rigen ae eS SS Milk. | Fat. | Milk.| Fat. | Milk.| Fat. | fat. Milk. | Fat. lbs. | P.ct.| lbs. | P.ct.| lbs. lbs. | P. ct. Ibs. lbs January _..........-.-....---| L&| 140| 17] 17.2] &2]| 0.481) 15.0} 99.2| 148 Welwoes 1.8} 146/ 1.0] 17.4| 2.8] 0.487| 15.6| 76.4} 11.92 fae 14] 144] 0.8] 17.6| 22/ 0.338] 15.1! 68.2) 10.30 | Bee ee 1.1! 15.0} 0.7| 16.7; 1.81 0.227| 15.4| 64.0| 832 | ee Se 11] 16.5] 1.0| 17.6] 21/0.347/ 16.5] 65.1] 10.74 | Totals for five months __|--___-- ee |-------|-------|------- Jel | 362.9 | 56.08 Kowloon, Hongkong, had thirty cows that were giving from 10 to 20 pounds of milk a day on dry rice straw and very little rice and bran, when visited by me in January, 1919. When o 4 he ise & xV,1 Levine; Milk Produced in Southern China 99 given succulent feed they give from 30 to 50 pounds of milk a day. These cows have a lactation period of from eight to twelve months. How well the flow of milk is kept up throughout the lactation period is not known. Buffalo milk is pure white, with absolutely no tint of yellow. Cream and butter made from it are rather oily and have but a faint tint of yellow. By mixing European cows’ cream with the buffalo cream and using vegetable butter color the color and the texture can be greatly improved. — TABLE XII.—Milk produced by buffalo cow 51. Average fat for lactation period, 9.65 per cent. [Freshened July 4, 1918. Lactation period ended May 30, 1919. Records began immediately after freshening.] Morning. Afternoon. | For 24hours. eae Total for month. i Month. oe ad Ee tee eee eee ees ae age LEAP alt 1GReE Mitk.| Fat. | mitk.! Fat. | Milk. Fat. | fat. | milk. | Fat. i] los. | P.ct.| lbs. | P.ct.| lbs. | lbs. | P.ct.| lbs. lbs, | | ide ee 6.2| 85 | 50| 95/ 11.2) 1.002) 89 | 202.4) 22.67 | Amun ee 6.8} 85 | 4.8) 10.0} 11.6] 1.058) 9.12) 9359.6] 32.77 } Sebtenibiiic. ) ao 3 & XV,1 Levine: Milk Produced in Southern China 101 modifying milk for infants requiring modified milk, the college physicians prefer buffalo milk to European cows’ milk because of the ease with which it lends itself to modification. The usual formula for modification is 100 grams of buffalo milk, 18 grams of sugar, and enough boiled or distilled water to make 300 grams. This makes the percentages of fat, proteins, sugar, and ash about right for infants, as when thus modified the milk compares well with human milk. For children of 3 years or more the milk is usually simply thinned by adding an equal amount of water. FEED FOR DAIRY COWS The commonest concentrates fed are equal proportions of a cheap grade of rice chop and wheat bran, fed twice a day. Na- tive cattle are grazed on the public grave and hill land during the daytime. Green grass is cut daily for the European cows, which are kept stabled all day. Most of the grass is cut from TABLE XVI.—Milk produced by buffalo cow 60. Average fat for siz months, 11.20 per cent. [Freshened December 1, 1918. Lactation period not erage ao Morning. Afternoon. | For 24hours. aie Total for month. Month. Sa ae Fe eS “| age Tah Milk. | Fat. | Milk.| Fat. | Milk.| Fat. | ** | Milk. | Fat. | iene eee en ae ee _ — | —_— —_-- ~ an eogmarectoemats Ibs. P.ct.| lbs. | P.ct.| tbs, | lbs. | Pet.| tbs. Ibs. | Dasember scene ie i 6.2| 9.50) 4.2) 11.75; 10.4] 1.881] 1135| 322.41 36.59 | ene. ga oe 6.0 10.20} 4.0, 12.00 | 10.0 | 1.092 | 10.92 | 810.0) 33.85 | Viiriity 3 5.8 10.50! 4.0, 12.40! 9.8 | 1.105 /1L27 | 274.4) 80.90 | Mai 2 at: | 4.8! 9.25} 30/1300} 7.8; 0.934! 11.97] 242.8] 28.74 | Diet iss ywanneth ty cpce ee 4.0, 9.50) 2.0 11.50 6.0] 0.610.) 20.26 | 190.0} 18.19 | _ ean eae 8.5 10.00| 38.0. 10.00 6.6 | 0-680 | 10.00 | 201.5 | 20.15 | Totals for six months.|.--_.-. ----.-- pene ie Ps cgasac ae eae | 1,631.1 168. 42 sea TABLE XVII.—Milk produced by buffalo cow 63. Average fat for six months, 10.88 per cent. {Freshened December 10, 1918. Lactation period not ee Secret LEE NS { Morning. | Afternoon. |Per aH asu ‘Total for Five aia Month. | ERIE PRE Sect | — i ae ruin at mink.| Fat. | Mitk.| Fat. | Mitk.! Fat.) % | yin, | Fat. | ie j j i } lbs. Pet. | Ibs, | P.ct.| tba. | Ibs. | P.ct.| lbs. | iba. ate mage 5.0 | 98 | 4.6] 112 9.6 | 0.965 | 10.05 | 201.6 , 19.35 JORUAES oo 5.8 | 96 | 4.8) 11.4 oe ete eee $13.0, 82,71. 4 Venrtaty 6.0 | 9.7 | 4.21 11.6 | 10.2 | 1.069 | 10.48) 283.6! 28.92 Merch 5 cite, eg 3 i 5.5 | 98 | 465/11.25| 100/104! 10.45! 310.0) 31.39 Ey Rae PA ee ' 4.8 | 10.0 | 4.0 | 11.5 8.9 | 0.940 | 10.56 | 267.0} 28.19 | May......-----.--..-----.---} 6.5 | 10.75; 3.0| 12.0 | 851.0591 11.26 263.5, 29.66 | Totals for six months... ieee Sea Pee eae Suet a et -oenee a 1, Gated] 190-28 | 102 Philippine Journal of Science 1919 TABLE XVIIIl.—Analyses of buffalo milk produced in southern China com- pared with that produced in other countries. {Numbers give percentages.] ree eee ee ee ; ~~ - - -- eevee =] | Philip- | | se ee | Paes a Sa 12. 60 7.95 6.84, 7.99 Pebbaiirsa ore ie et aati 6.04 4.00 4.97| 4.18 MA i a 8.70 6.18! 5.16{| 4.75 fe Se ee as ra plete - are aaeggpce oT NE aS airy ay came alee Setgs N° XV, 1 Witt: Portland Cement Raw Mixture aa | they have some effect on the accuracy of the results. One such modification is to precipitate and reprecipitate the iron and alu- minium from hydrochloric acid solution without first removing the silica. The calcium in the filtrate is precipitated once as the oxalate, and the precipitate is titrated with permanganate.** The method may be still further shortened by neutralizing the original hydrochloric acid solution with ammonia and adding some organic acid in which calcium oxalate is not appreciably soluble. The calcium may be precipitated as oxalate without the removal of iron and aluminium. The calcium oxalate is filtered out, and the determination is completed in the regular manner. Citric,** tartaric,** and oxalic®® acid have been used. Of the methods noted above, several are used extensively for controlling the calcium carbonate content of raw mixtures in cement plants. These include some modification of the determi- nation of calcium by precipitation with ammonium oxalate and titration with permanganate, the determination of carbon dioxide with the calcimeter, the determination of total alkalinity by titration with standard acid and alkali, and the determination of acid-insoluble matter. The last named is used much less fre- quently than the others. The procedure in each case that I have found best suited to the raw mixture discussed in this paper is as follows: In determining the calcium, the usual method of igniting the raw mixture before adding hydrochloric acid at the beginning of an analysis is not sufficient.*°° Results are the same either with or without ignition, but almost 2 per cent too low. It is necessary to fuse with sodium carbonate. To avoid one filtra- tion, the whole sample is fused instead of simply the acid- insoluble residue. The melt is dissolved in hydrochloric acid and the iron and the aluminium are removed by twice precipitating with ammonia. The calcium in the filtrate is precipitated once with ammonium oxalate, and the precipitate is dissolved in dilute sulphuric acid and titrated with permanganate. The titration method has been found more satisfactory than ignition, for use “ Enright, B., Journ. Am. Chem. Soc. 26 (1904) 1003. ® Passon, Max, Zeitschr. f. Ang. Chem. (1898) 776. “Blum, L., Zeitschr. f. Anal. Chem. 39 (1900) 152. * Meade, R. K., Chem. Eng. 1 (1904) 21. “ With reference to igniting the sample, W. F. Hillebrand, Journ. Soc. Chem. Ind. 21 (1902) 25, says: “Whether all cement mixtures can be thus rendered wholly soluble I am unable to say, but doubtless most of them can be, and if so, this method should always be resorted to in order to avoid the use of alkali carbonates.” 112 Philippine Journal of Science 1919 in the Tropics. The high humidity of the atmosphere renders it difficult to weigh calcium oxide in an ordinary platinum cruci- ble without absorption of moisture. Results obtained with the method just described have been found to compare favorably with those of the official method; consequently it was adopted as standard for the present work. In making a calcimeter determination, the sample is treated with hydrochloric acid (density, 1.125), and the calcium car- bonate content is calculated from the volume of carbon dioxide evolved. In some laboratories the amount of sample taken is not kept constant, but tables are used which indicate the weight that corresponds to the observed temperature and pres- sure. When this amount is placed in the appparatus, no reduc- tion of the observed volume to standard conditions is necessary. However, this arrangement has certain disadvantages when used in routine work. The continual changing of weights be- comes tiresome and the average routine analyst is likely to make mistakes, especially at night. Also, if for any reason it is de- sired to make a number of determinations at one time, the samples cannot all be weighed at once because the temperature or the pressure may change before the work has been com- pleted and thus render the results incorrect. Also, if the ap- paratus was standardized at a given temperature and pressure with Iceland spar, the result could not be applied to a determi- nation made after these conditions had changed without giving rise to error. The method used in connection with this work has proved more satisfactory. A standard raw mixture is prepared, having as nearly as possible the ideal composition of the regular mixture. This is prepared in quantities sufficient to last for a month or two and is analyzed in triplicate for calcium by the potassium permanganate titration method. At the beginning of his day’s work the analyst records the temperature and pressure, weighs out an equal amount of the standard and of his first sample to be tested, and places them successively in the calcimeter. As long as the temperature and the pressure remain the same, it is not necessary to use another portion of the standard; but, if either changes, a new standard must be weighed and analyzed with the next routine sample. If during the course of the day they both return to former values, the result of the standard at that temperature and pressure may be used. Since with each determination conditions are the same for both standard and sample, the calcium carbonate content of the oh ke XV, 1 Witt: Portland Cement Raw Mixture 118 sample may be directly calculated from the known content of the standard by proportion. A table for making these calculations covering the usual range of results for both the standard and the sample may be easily prepared. Errors, such as those due to absorption of carbon dioxide in the hydrochloric acid, lack of accuracy in either thermometer or barometer, and the like, are thus automatically corrected. Everything considered, the labor required by this procedure is no greater than that required by the other, and the results are more satisfactory. When the sample contains considerably more silicate than the standard there will be an error in the result, the reason for which is ex- plained in this paper. It has been proposed to correct this by preparing a series of standards ranging from the lowest to the highest calcium content likely to be found in a sample, and then, for a given determination, using the standard nearest that sample in composition. This would be an improvement, but it is doubtful if the increased labor would be justified. In the acid—alkali titration method, the sample is placed in a 300-cubic centimeter Erlenmeyer flask, fitted with a rubber stopper bearing a vertical glass tube as a condenser. A measured amount of 0.4 N hydrochloric acid is added, and brought just to a boil over a Bunsen burner. The solution is cooled, diluted to from 100 to 150 cubic centimeters, and titrated with standard sodium hydroxide of the same concentration, using methyl orange as the indicator. Of the several concentrations of acid tried, 0.4 N was the lowest that would dissolve nearly all the sample without the application of heat. However, that solution was not quite complete without warming. Methyl orange is pre- ferred to phenolphthalein, although the end point is more dif- ficult to see. It is not necessary to remove the carbonates in preparing the standard sodium hydroxide solution (as should be done if phenolphthalein is to be used), and it is not necessary to boil the acid after the sample is dissolved, to remove the carbon dioxide. The method for determining acid-insoluble matter is the same as the one given by Meade, except that hydrochloric acid having a density of 1.125 is used. The sample is treated with this acid, and the mixture is boiled several minutes. The insoluble matter is then collected, washed, ignited, and weighed. . The raw mixture with which this paper is concerned is made from three materials; namely, a hard limestone, and two sili- ceous materials that are designated silicate 1 and silicate 2. The analyses of these materials are shown in Table I. 166316——8 114 Philippine Journal of Science 1919 TABLE I.—Analyses of raw materials. {Numbers give percentages. ] Silicate. | Lime- | stone. | | i ee ae LAN Mit) SISTENT a cs Pbk | anes i admotedasus sshecl Leb lue ls ae 42.86 5.77 3.24 Poi Aaa a a ei os Skea Ssh ok we ese] 0. 68 53.76 61.35 ARN re os ee wes See 0.30 6.84 4,82 pie Onde See) ee es es oe 0,34 22.88 18. 46 Rieter eee MOBO) 2 cea Soc siee dS sok ee ee 54. 72 5.41 5. 22 Wiarreaitiaihs Wie GU) ooo pce ss Se os cab eeu en eee 0. 56 1.88 2.73 PRUE ORIG LINES coe le sic seo os cc seo ee 0.60 1.65 2.26 remeettin ansae CSO) Sota oc Se eke trace 1.84 1.95 Pipi an ivoriie (es) o.oo Sia i trace trace | trace i On analyzing a number of samples by the four methods men- tioned above and calculating all results to percentage calcium carbonate, the agreement was not satisfactory. The analyses of Most of these were prepared in the laboratory from samples of the raw materials, dried and ground to pass a 200-mesh sieve. comparative analyses that have been made from time to time. ten samples are shown in Table II. These are only a few of many TABLE II.—The analyses of raw mix by several methods. [Numbers give percentages. ] Sample No. | By titration with potas- sium perman- ganate. Total calcium is cal- culated to cal- cium carbon- ate. | | By calcimeter. | By acid-alkali | ; Total carbon titration. Total Matter insolu- ioxide i Ro ble in hydro- acer) a8 = | ee, “a chioric acid cium earbon- | calcium car- | With density ate. bonate. of 1.125 51.25 53. 08 22.30 59. 42 61,31 17.25 64.00 65.35" 16. 10 76.00 76.26 11.20 T1.25 76.30 21.68 77.00 76.30 21.82 78, 63 78. 88 11.04 83. 08 81.71 8.10 84.17 83. 22 7.00 86. 00 85.14 6.70 CAUSES OF ERROR In any analytical procedure certain errors of the process it- self—such as inexact calibration of apparatus, slight solubility of precipitates, difficulty of complete washing, and the like—can- 2 x XV, 1 Witt: Portland Cement Raw Mixture 115 not be eliminated. There is another class of errors due to the composition of the substances under investigation; errors of this class have been found to be largely responsible for the non- concordant results of Table II. In the first column, the calcium is determined, and the results are calculated to calcium carbonate. In the case of the calci- meter or with any other method in which the carbon dioxide is determined, the result in terms of calcium carbonate must be calculated from the total carbon dioxide found. This, of course, assumes that all calcium is present as carbonate and that all carbon dioxide is combined with calcium, which in this case is not true. Carbonates other than calcium are present in small amount in the limestone, causing the results to be slightly too high. However, the calcium in the two siliceous materials is not present as carbonate and consequently is not included when the determination by calcimeter is made. The net result is that the values obtained are low. In the acid—alkali titration method, the alkalinity or acid- consuming power is calculated to calcium carbonate. In this ° case the errors due to apparatus and to manipulation may be made very small. However, there are much greater errors due to the composition of the mixture. On being treated with standard acid, all carbonates go into solution and are calculated to calcium carbonate. Of the calcium in the siliceous materials, a portion goes into solution without neutralizing acid equivalent to that which would be required for calcium carbonate, and a portion is undissolved by the acid. Some of the iron and the aluminium in these materials go into solution and consequently neutralize the acid, thereby tending to produce a higher result.*” The results obtained by determining the acid-insoluble matter were so irregular that no further analyses by that method were made. Slight errors are caused by the presence of moisture in the raw mixture as it comes from the tube mill. It has not been found practicable to eliminate these in the routine work. The moisture content of the material as it enters the tube mill is determined at regular intervals and is kept as low as possible *'S. B. Newberry, Cem. and Eng. News, March (1903) 35, describes a double titration method by which correction for magnesium may be made. With the raw mixture under investigation, however, the error due to the effect of iron and aluminium cannot be corrected. The error due to mag- nesium is not corrected because it tends to compensate the other one. 116 Philippine Journal of Science 1919 under the prevailing conditions at the plant. It is not practical to dry each sample before analysis, because the results cannot be obtained as rapidly as they are needed. The limestone as it comes from the quarry is frequently contaminated to a greater or less extent with clay different in composition and in analyt- ical behavior from the two siliceous materials regularly used in making the mix. The results obtained by testing the three raw materials by the calcimeter and the alkali—acid titration method are shown in Table II. TABLE III.—Partial analyses of the three raw materials by different methods. Silicate. Calcium carbonate. —— sea a is 2. ‘ Per cent. | Per cent. | Per cent. By calcimeter (total CO2 calculated to CaCO3)___._---_---.-------- 94. 50 0.00 0.00 By titration (total acid neutralized calculated to CaCOg)_-_____-___ 95. 40 2.60 3.30 It will be noted that although the two silicates contain con- siderable calcium there is no evolution of gas when they are treated with acid. However, although they contain no car- bonates, they have acid-consuming values equivalent to 2.60 per cent and 3.30 per cent calcium carbonate, respectively. The portion of each, soluble in 0.4 N hydrochloric acid, gives the results shown in Table IV, calculated to the basis of the original sample. TABLE IV.—Constituents of the two silicates. Silicate. 1, 2. Per cent. | Per cent. Tron and aluminium oxides 2.57 6.65 :cuei onto dannwbaeebiee SSS pela Ilel sea etUS 2.26 3.08 The sources of error with the analytical methods employed and their effect on the results are illustrated in fig. 1. This is based on the typical analysis shown in Table V. Fs LF 117 ‘ta[Vus Aayeuoyjtodoid p Supyeul Aqaioy} ‘ssourvala Jo ayw"s ay} r0F 9 Aq poesia Sexe oie a pue q ynq ‘aruos 0} UMBIP 91B p puB o SB2aIY ‘“SpoyjoU [wAeAaS Aq S}[NSeI Jo JUsWeaTHeUOU JOJ sasnes oUIOS PUL VINFXIU MBI JO UOIsoduI0D Burmoyg *T "DIT me ee Yuu "0282 uayzdayi0 spunodwos ut 08) EY ee a Gy : | : y = : 2 o £022 8D paeUIqwoy 087 p S MS oi | L213 = : oS Aa 3 9 YG BY 3 S |&| AY GF |s ES| iS) Is ® Q 2 jj; Yy : ©0282 SO peauUlgui02 #499 Fe) S 2 D ay Aa Yy i m/e oe cag: §Q2©2 UPUT J2YT0 SpPURnOAUWO Ul 269 q : SZUANPZISUOD saYyIO HP (1) i A ee 118 Philippine Journal of Science ine TABLE V.—Typical analysis. Equiva- lent to Constituent. calcium carbon- ate. Percent. | Per cent. Calcium oxide combined as calcium carbonate --_-__.------------------------- 42.08 715. 05 Cates 0x80 i O8NOr COMDOUNOE. 655 oe oe ee een eles 1.23 2.19 Carbon dioxide combined as calcium carbonate __-___.-._--._--------.-------- 32. 99 75. 05 Corben dickide im other Compounds Clinker Fercentage Fig. 2. Analysis of raw mixture at various stages ef its conversion into elinker. 120 Philippine Journal of Science calcium present must be based on the determination of the cal- cium itself. The reason for this is shown in fig. 2, which is based on the analysis of a number of samples of raw mixture at various stages of conversion into clinker. Curve a shows total calcium; b, total alkalinity; and c, carbon dioxide—all calculated to percentage of calcium oxide. The raw mixture as it entered the kiln contained almost 44 per cent calcium oxide as indicated by the three methods of analysis, taking into account the errors, which have been explained. The first part of all three curves rises, because during the first period no chemical change in the constituents of the mixture is taking place, and the percentage of calcium oxide is increasing because of the loss in moisture, organic matter, and the like. However, the change at once becomes pronounced, when the tem- perature becomes sufficiently high to decompose the limestone. Due to the loss in weight caused by the evolution of carbon dioxide, the percentage of calcium oxide present continually in- creases and the percentage of carbon dioxide decreases. The curve representing the alkalinity, or the acid-consuming value, of the mixture does not change so much as the other two. The factors that tend to make the curve rise and those having the reverse effect largely neutralize each other.** SUMMARY In determining the calcium carbonate content of Portland ce- ment raw mixture the methods available may be divided into two classes; namely, those in which the calcium is determined and those in which some other constituent is determined. Although some methods of the first class have been greatly shortened by several investigators, generally some method of the second class is employed in control work because of simplicity and rapidity. While all methods are subject to errors arising from the appa- ratus and from the procedure employed, errors due to the nature of the constituents of the raw mixture may greatly influence the results. The causes of errors that may arise have been studied. ’ * An investigation of certain phases of the subject is now in progress. ae — ILLUSTRATIONS TEXT FIGURES Fic. 1. Diagram, showing composition of raw mixture and some causes for nonagreement of results by several methods. Areas c and d are drawn to scale, but 6 and e are exaggerated by 5 for the sake of clearness, thereby making a proportionally smaller. 2. Diagram of analyses of raw mixture at various stages of its con- version into clinker. 121 REVIEW The Elements | of | Animal Biology | by | S. J. Holmes, Ph. D. | professor of zodlogy, University of California | with 249 illustrations | Philadelphia | P. Blakiston’s Son & Co, | 1012 Walnut Street | Cloth, pp. i-vii + 1-402, including index. Professor Holmes’s book, Elements of Animal Biology, is a high-school text in animal biology. It is divided into three sec- tions: Part I, The Animal Kingdom; Part II, The Elements of Phisiology; and Part III, General Features and Adaptations. The part devoted to The Animal Kingdom gives the student an abbreviated but excellent idea of most of the important animals of the various groups, including their habits and in some cases their economic value and relation to disease. The second part is introduced by a chapter on the Chemical Basis of Life and one on Cells and Tissues. The treatment is brief but adequate. A valuable chapter on Bacteria and Disease completes the part. In the final part, General Features and Adaptations, evolution and heredity are considered, and the essential facts for a high- school student are presented. The book is well written and interesting. It should be a valuable text for high schools that offer courses in animal biology. BR. 2 ©. 123 THE PHILIPPINE JOURNAL OF SCIENCE VoL. XV AUGUST, 1919 No. 2 ON THE APPLICATION OF THE GENERIC NAME MELODORUM OF LOUREIRO By E. D. MERRILL Botanist, Bureau of Science, Manila The genus Melodorum was proposed by Loureiro’: in 1790 for two species from Cochinchina, Melodorum fruticosum Lour. and M. arboreum Lour., of which the first is manifestly the type of the genus. While the description is not very definite, and while manifestly the fruit characters are based on material from some plant different from the type and in all probability not congeneric with it, it seems to be distinctly advisable to in- vestigate the status of the genus in view of the misinterpre- tation of it in all modern botanical literature. Fortunately for the purpose of this investigation Loureiro’s type of Melodorum fruticosum is preserved in the herbarium. of the British Museum, and through the kindness of Dr. A. B. Rendle and Mr. J. G. Baker I have been supplied with a sketch of it with certain notes regarding the specimen. The material consists of a leafy branch with the broken pieces of a flower, from which it was possible to sketch an entire petal. After receiving the sketch mentioned above it was matched by material in the herbarium of the Bureau of Science and in order to verify my interpretation of Loureiro’s species frag- ments of several specimens were forwarded to the British Mu- seum for purposes of direct comparison with Loureiro’s type. In reference to this material Doctor Rendle notes under date of July 3, 1919: Loureiro’s plant is very much like the one labeled Polyalthia siamensis Boerl., cult. Hort. Bog.; the large petals seem identical in form but the *FL Cochinch. (1790) 351. 166470 125 126 Philippine Journal of Science 1919 prominent lateral veins of the leaf are rather less numerous (about 10 or 11 on each side) in Loureiro’s plant, and also slightly farther apart. The twigs are also thicker in Loureiro’s specimen. We find nothing that agrees better with it. Dunal? seems to be the first author to adopt Loureiro’s gen- eric name, making it a section of the genus Unona. He referred both of Loureiro’s species to Unona; Melodorum fruticosum Lour. == Unona dumetorum Dunal and Melodorum arboreum Lour. = Unona sylvatica Dunal. He saw no specimen but inter- preted both species from Loureiro’s descriptions. In proposing the section Melodorum, Dunal referred to it several other species; Unona latifolia Dunal — Melodorum latifoliwm Hook. f. & Th., U. lucida DC.—Xylopia longifolia A. DC., Unona acutiflora Dunal — Xylopia sp., Unona xylopoides Dunal, and Unona poly- carpa DC. == Xylopia polycarpa Oliv. In current botanical literature the authority for the generic name Melodorum is given as Hooker f. and Thomson,’ but these authors credit the authority for the genus to Dunal, citing as synonyms Unona, section Melodorum Dunal; Uvaria, section Mel- odorum Blume; and Polyalthia, section Kentia Blume. Hooker f. and Thomson apparently interpreted the genus largely from the first species cited by Dunal; namely, the Malayan one cur- rently known as Melodorum latifolium (Dunal) Hook f. & Th., described and figured by Blume as Uvaria latifolia Blume.* Hooker f. and Thomson examined Loureiro’s type of Melo- ’ dorum fruticosum in the herbarium of the British Museum with the following comment: Loureiro’s Melodorum is different, as we have determined by an inspec- tion of the materials in the British Museum. In that collection there is an authentic specimen of M. fruticosum Lour., which is an undescribed plant, of doubtful affinity, as we have not examined the flower, but cer- tainly not belonging to this genus. It has no fruit. There is no authentic specimen of M. arboreum Lour., but it is described as a large tree and is perhaps a Mitrephora, While they excluded both of Loureiro’s species from Melo- dorum, Hooker f. and Thomson considered it advisable to retain the generic name in the sense in which it was interpreted (in part) by Dunal and by Blume (as a section of Unona and of Uvaria). All subsequent authors have been content to follow Hooker f. and Thomson, and we hence have an entirely illogical * Monogr. Fam. Anon. (1817) 98, 115, 116. * Fl. Ind. (1855) 112. *Fl. Jav. Anon. (1828) 87, t. 15, 25A. \r m (> Joe XV,2 Merrill: Melodorum of Loureiro 127 case of a generic name proposed by Loureiro currently applied to a group of species quite different from Loureiro’s original conception of the genus. From an examination of the sketch of Loureiro’s type kindly prepared by Mr. J. G. Baker I am confident that Melodorum fruticosum Lour. is the species currently known as Polyalthia aberrans Maingay and which has been variously described by several authors as Polyalthia affinis Teysm. & Binn., P. siamensis Boerl., and Unona mesneyi Pierre, and which was eventually transferred to Popowia by Pierre as Popowia aberrans Pierre. From this I cannot distinguish Popowia diospyrifolia Pierre by any satisfactory characters. Pierre * has given a detailed account of the characters of this species, indicating that it approaches Popowia in certain respects, but concluding that it cannot be referred to Endlicher’s genus. He found that it is, in most re- spects, a Polyalthia but proposed for it the group name Mesneya, as a subgenus of Unona. Safford ® has clearly shown that the so-called Old World species of Unona have nothing to do with Unona as originally described by the younger Linnaeus, this being an American genus, the proper generic name for the Old World species being Desmos Lour. King’ notes that Polyalthia aber- rans Maing., as interpreted by him, “save, and except in the much smaller size of the flowers,’ much resembles the plant figured and described by Pierre under the name of Unona mesneyt, and to which Pierre reduced Polyalthia aberrans. An examination of Cochinchina material now available to me for study and comparison shows that the size of the flowers in Pierre’s drawing is greatly exaggerated, and that they are drawn all out of proportion to the leaves. In interpreting the true status of Melodorum as described by Loureiro, we find that the type of the genus is a species aberrant in all of the genera in which it has been placed. It is to be noted that Pierre proposed for it the group name Mes- neya (as a subgenus), and that Scheffer, quoted by Boerlage,* proposed for the same form the generic term Sphaerocoryne. For the present it seems best to retain Melodorum as a genus closely allied to Popowia, for which the names Mesneya Pierre and Sphaerocoryne Scheffer are exact synonyms. Polyalthia * Fl. Forest. Cochinch. 1 (1880) t. 17. * Bull. Torr. Bot. Club 39 (1912) 501-508. "Ann. Bot. Gard. Calcutta 4 (1893) 78. *Boerl. in Ie. Bogor. 1 (1899) 196, in nota sub Polyalthia siamensis Boerl. 128 Philippine Journal of Science 1919 Blume may then be retained for the numerous Indo-Malayan species currently so called, but the various species of Melodorum of all authors except Loureiro will need a new generic name, for which Fissistigma Griff. is available. MELODORUM Loureiro MELODORUM FRUTICOSUM Lour. FI. Cochinch. (1790) 351. Unona dumetorum Dunal Monog. Anon. (1817) 31. Polyalthia (?) aberrans Maingay in Hook. f. Fl. Brit. Ind. 1 (1874) 67; King in Ann. Bot. Gard. Calcutta 4 (1893) 78, t. 109A. Polyalthia affinis Teysm. & Binn, in Nat, Tijdschr. Nederl. Ind. 27 (1864) 37; Boerl. in Ic. Bogor. 1 (1899) 124, 183, ¢. 63. Polyalthia siamensis Boer]. in Ic, Bogor. 1 (1899) 124, 195, t. 69. Sphaerocoryne siamensis Scheff. ex Boerl. |. ¢. in syn. Popowia aberrans Pierre ex Finet & Gagnep. in Bull. Soc. Bot. France 53 (1906) Mem. 4: 109, et in Lecomte FI. Gén. Indo-Chine 1 (1907) 83. Unona mesneyi Pierre Fl. Forest. Cochinch. 1 (1880) #. 17. Popowia diospyrifolia Pierre ex Finet & Gagnep. in Bull. Soc. Bot. France 53 (1906) Mém. 4: 110, et in Lecomte op. cit. 84, t. 10, f.. 11-19. Melodorum glaucum Scortech. ex King in Ann, Bot. Gard. Calcutta 4 (1893) 78, in syn. Melodorum clavipes Hance in Journ. Bot. 15 (1877) 328. There may be more than one species represented here, but if so they are very closely allied. According to Loureiro (Melo- dorum fruticosum), Boerlage (Polyalthia affinis Teysm. & Binn., P. siamensis Boerl.,), Pierre (Unona mesneyi Pierre), and Gag- nepain (Popowia aberrans Pierre, P. diospyrifolia Pierre), it is an erect shrub or small tree, but according to Maingay and King it is a scandent shrub. The difference between Polyalthia siamensis Boerl. (Polyalthia aberrans Maing.) and P. affinis Teysm. & Binn. appears to me to be trivial. As between Popowia aberrans Pierre and P. diospyrifolia the same statement holds, the distinguishing character depended upon being only that of the stigma. It is to be noted that Finet and Gagnepain consider that Pierre’s drawing of Unona mesneyt, as to the habit sketch, is Popowia aberrans Pierre, and as to the details of the flower, P. diospyrifolia Pierre. The species extends from Indo-China to Siam and the Malay Peninsula and is in cultivation at Buitenzorg, Java. I have examined the following specimens: Indo-China, Lowreiro (sketch of Loureiro’s type as preserved in the herbarium of the British Museum, flowering specimen), Thorel 391 (det. by Gagnepain as Popowia diospyrifolia Pierre, flowering specimen): Malay Peninsula, Pungah, Curtis 2957; Goping, Kunstler 6136; Ma- | j | \r i XV,2 Merrill: Melodorum of Loureiro 129 lacca, Burkill 2510; Perak, Scortechini 1946 (all these distributed as Polyalthia aberrans Maing., flowering specimens) : Java, cult. Hort. Bogor XI-A-41-71 (four sheets, Polyalthia siamensis Boerl., flowering specimens from the type plant), XJ+A-63 (Po- lyalthia affinis Teysm. & Binn., flowering and fruiting specimens). As several detailed descriptions of this species, as well as no less than four illustrations of it, with details of the flowers and fruits, have been published, it would seem that a further de- scription is unnecessary. The genus Melodorum Lour., as I understand it, contains a single definitely known species, which while well defined and characteristic as a species presents a combination of characters which render it somewhat difficult to separate Melodorum from several not closely allied genera. It is clearly no Unona (that is, Desmos); and it is equally aberrant in Polyalthia and in Popowia, the two other genera in which it has been placed. Iam personally of the opinion that Melodorum as originally described by Loureiro and typified by Melodorum fruticosum Lour. (not of modern authors) is a valid genus, more closely allied to Popowia than to Polyalthia, and that it belongs in the tribe Mitrephorae. Pierre in his critical discussion of Unona mesneyi notes that Maingay described the inner petals as imbricate in bud, but that he found them to be perfectly valvate in Wallich’s specimen. Both series of petals touch by their thickened margins, and those of the inner series remain in this position long after anthesis. I have seen no fresh material of Melodorum fruticosum, but the figures given by Pierre, Boerlage, and King present the outer series of petals as more or less spreading in anthesis; in all the herbarium specimens examined by me none of the petals are spreading. The persistent valvate position of the inner petals is a character by which the genus can be readily distinguished from Polyalthia and all the other genera in the tribe Unonae; while in the Mitrephorae it is readily distinguished from Fissistigma Griff. (Melodorum auct., non Lour.) by its globose buds and more or less spreading (?) outer petals; from Popowia Endl., which seems to be its true alliance, it differs entirely not only in its facies, but also in its larger, long-pedicelled flowers; in its outer petals, which are much larger and entirely different from the sepals; and by the inner petals being valvate by their much thickened margins but not connivent. Melodorum arboreum Lour.® — Unona sylvatica Dunal,’® the * Fl. Cochinch, (1790) 351. * Monog. Anon. (1817) 91. 180 Philippine Journal of Science 1919 second species of the genus described by Loureiro, is one of doubt- ful status, but in all probability is not congeneric with Melo-- dorum fruticosum Lour. It is described as a large tree growing in forests, the leaves tomentose beneath, the calyx and corolla as in Melodorum fruticosum Lour. The indicated floral char- acters can be considered as no more than approximate. Hooker f. and Thomson have suggested that this may prove to be a Mitrephora and, from the description and the various species of Mitrephora now known from Indo-China, I strongly suspect that Loureiro’s species is the same as Mitrephora thorelii Pierre. It is perfectly evident that whatever disposition be made of Melodorum as originally described by Loureiro, whether it be considered as a valid genus allied to Popowia, or whether it be considered as synonymous with Popowia, or with Polyalthia, it can no longer be retained in the sense in which it is currently used; that is, as erroneously interpreted by Hooker f. and Thom- son. The genera proposed by other authors currently considered as synonyms of Melodorwm Hook. f. & Th., are Fissistigma Griff. (1854), Mitrella Mig. (1865), and Pyramidanthe Miq. (1865). Of these Fissistigma Griff. is an exact synonym of Melodorum as interpreted by Hooker f. and Thomson and ante- dates Melodorum as used by the latter authors by one year. It is typified by Fissistigma scandens Griff. I accordingly pro- pose to adopt Griffith’s generic name for the numerous species currently but erroneously known as Melodoruwm. Boerlage '* retains Kentia Mig. and Pyramidanthe Miq. as valid genera allied to Melodorum Hook. f. & Th., that is, Fissistigma Griff.; but in this consideration of the case I have followed current usage and have reduced both to Fissistiqma. ' FISSISTIGMA Griffith (Melodorum auct. plur., non Lour.) FISSISTIGMA AFRICANUM (Benth.). Melodorum africanum Benth. in Trans. Linn. Soc. 23 (1862) 477. Africa. 9X° “FissiSTIGMA BALANSAE (Aug. DC.). Melodorum balansae Aug. DC. in Bull. Herb. Boiss. II 4 (1904) 1070; Finet & Gagnep. in Lecomte Fl. Gén. Indo-Chine 1 (1907) 104. Indo-China. * Notul. 4 (1854) 706. "Tc. Bogor. 1 (1899) 129, 1380. qr ee \- XV,2 Merrill: Melodorum of Loureiro 181 FISSISTIGMA BECCARII (Scheff.). Melodorum beccarii Scheff. in Ann. Jard. Bot. Buitenz. 2 (1885) 24. Mitrella beccarii Diels in Engl. Bot. Jahrb. 49 (1912) 149, New Guinea. FISSISTIGMA BICOLOR (Roxb.). Uvaria bicolor Roxb. Fl. Ind. ed. 2, 2 (1832) 662. Melodorum bicolor Hook. f. & Th. Fl. Ind, (1855) 119; King in Ann. Bot. Gard. Calcutta 4 (1894) 133, t. 175B. Eastern Himalayan region to Assam and Burma. FISSISTIGMA BORNEENSE (Mia.) Melodorum borneense Miq. in Ann. Mus. Bot. Lugd.-Bat. 2 (1865) 36. Borneo. FISSISTIGMA CINERASCENS (Mia.). Melodorum cinerascens Mig. Ann. Mus. Bot. Lugd.-Bat. 2 (1865) 37. Sumatra. FISSISTIGMA CHRYSOSERICEUM (Finet & Gagnep.). Melodorum chrysosericeum Finet & Gagnep. in Bull. Soc. Bot. France 54 (1907) 88, et in Lecomte Fl. Gén. Indo-Chine 1 (1907) 98, f. 12. Indo-China, Thorel 2429! FISSISTIGMA CYLINDRICUM (Maingay). Melodorum cylindricum Maingay in Hook. f. Fl. Brit. Ind. 1 (1872) 80; King in Ann. Bot. Gard. Calcutta 4 (1893) 136, t. 177B. Melodorum cylindraceum Boerl. in Ic. Bogor. 1 (1899) 183. Malay Peninsula, Singapore, Borneo, Ridley 1823!, s. n.! FISSISTIGMA ELEGANS (Wall.). Uvaria elegans Wall. Cat. (1832) No. 6474A, nomen nudum. Melodorum elegans Hook. f. & Th. Fl. Ind. (1855) 122; King in Ann. Bot. Gard. Calcutta 4 (1893) 141, t. 184B. Malay Peninsula, Penang, Wray 1079!, King’s collector 6367!, Kunstler 5115!, Ridley 13516!, s. n.! FISSISTIGMA FAGIFOLIUM (Ridl.). Melodorum fagifolium Ridl. in Kew Bull. (1912) 286. Borneo, Hose 397! FISSISTIGMA FULGENS (Wall). Uvaria fulgens Wall. Cat. (1832) No. 6482, nomen nudum. Mitrephora fulgens Hook. f. & Th. Fl. Ind. (1855) 120; King in Ann. Bot. Gard. Calcutta 4 (1893) 132, t. 174. Malay Peninsula, Singapore, Borneo (not in the Philippines!), Kunstler 2654!, Haniff & McNur 2317!, Wray 2441! Ridley 4587! 132 Philippine Journal of Science 1919 ed FISSISTIGMA GLAUCESCENS (Hance). Melodorum glaucescens Hance in Journ. Bot. 19 (1881) 112. Hongkong. FISSISTIGMA HYPOGLAUCUM (Miq.). Melodorum hypoglaucum Mig. Ann. Mus. Bot, Lugd.-Bat. 2 (1865) 87; King in Ann. Bot. Gard. Calcutta 4 (1893) 136, t. 180A. Malay Peninsula, Borneo, Sumatra, King’s collector 5060!, 5806!, 1022! FISSISTIGMA KENTII (Blume). Unona kentii Blume Bijdr. (1825) 16. Polyalthia kentti Blume FI. Jav. Anon. (1828) 77, t. 38, 52A. Melodorum kentii Hook. f. & Th. Fl. Ind. (1855) 116, Mitrella kentii Miq. Ann. Mus. Bot. Lugd.-Bat. 2 (1865) 39; Boerl. in Ic. Bogor. 1 (1899) 130, t. 43, f. C. ‘ Java, cult. Hort. Bogor. XJ-A-29-64! Boerlage, l. ¢., retains this as a distinct genus. FISSISTIGMA KINABALUENSE (Stapf). Melodorum kinabaluense Stapf in Trans. Linn. Soc. Bot. 4 (1894) 130. Borneo. FISSISTIGMA KORTHALSII (Miq.). Melodorum korthalsti Miq. Ann. Mus. Bot. Lugd.-Bat. 2 (1865) 37. Borneo. FISSISTIGMA LANUGINOSUM (Hook. f. & Th.). Melodorum lanuginosum Hook. f. & Th. FI. Ind. (1855) 117; in Ann. Bot. Gard. Calcutta 4 (1893) 138, ¢. 182. Uvaria tomentosa Wall. Cat. (1832) No. 6454, nomen nudum, Malay Peninsula, Penang, Singapore, Indo-China, King’s col- lector 4664!, 7890!, 4581!, Hullett s. n.!, Ridley s. n.! FISSISTIGMA LATIFOLIUM (Dunal). Unona latifolia Dunal Monog. Anon. (1817) 115. Uvaria longifolia Blume Bijdr. (1825) 13. Uvaria latifolia Blume Fl. Jav. Anon. (1828) 37, t. 15. Melodorum mollissimum Miq. Fl. Ind. Bat. Suppl. (1861) 374. Melodorum latifolium Hook. f. & Th. Fl. Ind. (1855) 116; King in Ann. Bot. Gard. Caleutta 4 (1894) 1385, ¢. 178. Malay Peninsula (Perak), Java, Mt. Salak, ex herb. Bogor.! collector not indicated. Borneo, Indo-China, fide Gagnepain, Moluccas. QF ty XV,2 Merrill: Melodorum of Loureiro 133 The type of this species is Cananga sylvestris III latifolia Rumph." of Amboina. The species is currently interpreted from Blume’s figure and description which were based on Javan ma- terial. Boerlage expresses the opinion that Melodorum moilis- simum Miq. is specifically distinct from M. latifolium Hook. f. & Th. FISSISTIGMA LEICHHARDTI! (Benth.). Melodorum leichharditi Benth. Fl. Austral. 1 (1863) 52. Australia (Queensland, N. S. Wales), White! Boorman! FISSISTIGMA LITSAEFOLIUM (King). Melodorum litsaefolium King in Journ, As. Soc. Beng. 61’ (1889) 103, Ann. Bot. Gard. Calcutta 4 (1893) 182, t. 178. Malay Peninsula, Perak, King’s collector 4986!, 4063. FISSISTIGMA LONGIPETALUM (Ridl.). Meledorum longipetalum Ridl. in Kew Bull. (1912) 387. Borneo. FISSISTIGMA MABIFORME (Griff.). Uvaria mabiformis Griff. Notul. 4 (1854) 709. Melodorum pisocarpum Hook, f. & Th. Fl. Ind. (1855) 123; King in Ann. Bot. Gard. Calcutta 4 (1893) 142, t. 187A. Melodorum pyramidale Maingay ex King 1. c. in syn. Malay Peninsula, Singapore, Sumatra, King’s collector 6411!, 10602!, Forbes 2182!, Ridley s. n.! FISSISTIGMA MACCREAI (F. Muell.). Melodorum maccreai F. Muell. Fragm. 6 (1867) 176. Australia. FISSISTIGMA MACRANTHUM (Kurz). Melodorum macranthum Kurz in Journ, As. Soc. Beng. 41° (1872) 291; King in Ann. Bot. Gard. Caleutta 4 (1892) 140, ¢. 186. Unona macrantha Kurz in Andam. Rep. (1867) App. B. I. Pyramidanthe macrantha Kurz op. cit. ed. 2 (1870) 29. South Andaman, Kurz! FISSISTIGMA MAINGAYI (Hook. f. & Th.). Melodorum maingayi Hook. f. & Th. Fl. Ind. (1855) 189; King in Ann. Bot. Gard. Calcutta 4 (1893) 189, t. 184A. Malay Peninsula, Penang, Borneo, Wray 1112! * Herb. Amb. 2: 198. 134 Philippine Journal of Science 1919 FISSISTIGMA MANUBRIATUM (Wall). Uvaria manubriata Wall. Cat. (1832) No. 6456, nomen nudum. Melodorum manubriatum Hook. f. & Th. FIL. Ind. (1855) 118; oe in Ann. Bot. Gard. Caleutta 4 (1893) 134, t. 176. Melodorum bancanum Scheff. in Nat. Tijdschr. Nederl. Ind. 31 (1870) 348, Flora 53 (1870) 244. Malay Peninsula, Penang, Bangka, Borneo, Kunstler 7832!, King’s collector 4948!, 5786!, Ridley 4712!, Cult. Hort. Bogor. XVI-E-107!, XI-A-15-67! FISSISTIGMA OBLONGUM (Craib). Melodorum oblongum Craib in Kew Bull. (1914) 5. Siam, Kerr 1879! FISSISTIGMA OLDHAM! (Hemsl.). Melodorum oldhamit Hemsl. in Journ, Linn. Soc. Bot. 23 (1886) 27. Formosa, Southern China,-Dunn 6339!, Faurie 430!, Tutcher 1052!, Japanese collector 1292! FISSISTIGMA OVALIFOLIUM (Ridl.). Melodorum ovalifolium Ridl. in Kew Bull. (1912) 387. Borneo. FISSISTIGMA OVOIDEUM (King). Melodorum latifolium Hook. f. & Th. var. ovoidea King in Journ. As. Soe. Beng. 61* (1892) 106, Ann. Bot. Gard. Calcutta 4 (1898) 135, ta 178, Malay Peninsula, Singapore, King’s collector 8119!, 10418!, Burkill! This differs remarkably from the typical form of Melodorum latifolium as figured and described by Blume in its very much larger, entirely different fruits, = is certainly worthy of specific rank. FISSISTIGMA PALLENS (Finet & SGikass, Melodorum pallens Finet & Gagnep. in Bull. Soc. Bot. France 53 (1906) Mém. 4: 187, et in Lecomte Fl. Gén. Indo-Chine 1 (1907) 3100, &. 22, Ff. 7-H. Indo-China. FISSISTIGMA PANICULATUM (Ridl.). Melodorum paniculatum Ridl. in Kew Bull. (1912) 386. Borneo. ay ve aw XV,2 Merrill: Melodorum of Loureiro 185 FISSISTIGMA PARVIFLORUM (Scheff.). Melodorum parviflorum Scheff. in Nat. Tijdschr. Nederl. Ind. 31 (1870) 344, Flora 53 (1870) 244; King in Ann. Bot. Gard. Cal- cutta 4 (1893) 1387, ¢. 181. Malay Peninsula, Bangka, Borneo, King’s collector 7276! 8366!, 6498! FISSISTIGMA POLYANTHOIDES (Aug. DC.). Melodorum polyanthoides Aug. DC, in Bull. Herb. Boiss. II 4 (1904) 1070; Finet & Gagnep. in Lecomte Fl. Gén. Indo-Chine 1 (1907) 103. Indo-China, Thorel 2431! FISSISTIGMA POLYANTHUM (Wall.). Uvaria polyantha Wall. Cat. (1831) No. 6467, nomen nudum. Melodorum polyanthum Hook. f. & Th. Fl. Ind. (1855) 121; King in Ann. Bot. Gard. Calcutta 4 (1893) 131, t. 172A. Chittagong. FISSISTIGMA PRISMATICUM (Hook, f. & Th.). Melodorum polyanthum Hook. f. & Th. Fl. Ind. (1855) 121; King in Ann. Bot. Gard. Calcutta 4 (1894) 140, t. 185. Uvaria rufa Wall. Cat. (1832) No. 6455, nomen nudum. Pyramidanthe rufa Miq. Ann. Mus. Bot. Lugd.-Bat. 2 (1863) 39; Boerl. in Ic. Bogor. 1 (1899) 180, t. 48, f. D. Oxymitra bassiaefolia Teysm. & Binn. in Nat. Tijdschr. Nederl. Ind. 25 (1863) 419. Malay Peninsula, Singapore, Borneo, King’s collector 5737!, -$932!, Baker 5705! Boerlage retains Pyramidanthe Miquel as a genus distinct from Melodorum (Fissistigma). FISSISTIGMA PUNCTULATUM (Baill.). , Melodorum punctulatum Baill. in Adansonia 10 (1871) 107. New Caledonia. FISSISTIGMA RIGIDUM (Ridl.). Melodorum rigidum Ridl. in Kew Bull. (1912) 386. Borneo. FISSISTIGMA RUBIGINOSUM (A. DC.). Uvaria rubiginosa A. DC. in Mém. Soc. Phys. Genév. 5 (1832) 202. Uvaria nervosa Wall. Cat. (1832) No. 6479, nomen nudum. Uvaria fulva Wall. op. cit. No. 6427, nomen nudum. Melodorum rubiginosum Hook. f. & Th. Fl. Ind. (1855) 138; King in Ann. Bot. Gard. Calcutta 4 (1893) 188, t. 183. Sylhet to Chittagong, Tenasserim, Penang (var. oblongum King, King’s collector 5082!), Indo-China, and Borneo. 1386 Philippine Journal of Science 1919 FISSISTIGMA RUFINERVE (Hook. f. & Th.). Melodorum rufinerve Hook. f. Th. Fl. Ind. 1 (1855) 121; King in Ann. Bot. Gard. Calcutta 4 (1893) 130, t. 172B. India. FISSISTIGMA RUFUM (Presl). Anona rufa Presl Rel. Haenk. 2 (1830) 75. Melodorum clementis Merr. in Philip. Journ. Sci..3 (1908) Bot. 136. Melodorum rufum Merr. op. cit, 223. This species was originally described from Luzon material and it extends from central Luzon to Mindanao. It is closely allied to Fissistigma latifolium (Dunal) Merr. (Melodorum latifolium Hook. f. & Th.), and Philippine specimens of it have been referred to the latter species. FISSISTIGMA SCANDENS Griff. Notul. 4 (1854) 706. Melodorum griffithii Hook. f. & Th. FL Ind. (1855) 120; King in Ann. Bot. Gard. Calcutta 4 (1893) 131, t. 172B. Burma, Indo-China. FISSISTIGMA SCHLECHTER] (Diels). Mitrella schlechteri Diels in Engl. Bot. Jahrb. 49 (1912) 150. New Guinea. FISSISTIGMA SCHEFFERI (Pierre). Melodorum schefferi Pierre ex Finet & Gagnep. in Bull. Soc. Bot. France 53 (1906) Mém. 4: 134, et in Lecomte Fl. Gén. Indo-Chine 1 (1907) 99, ¢. 18, f. 1-8. Indo-China. MELODORUM SPHAEROCARPUM (Blume) Migq. Fl. Ind. Bat. 17 (1858) 35; King in Ann. Bot. Gard. Calcutta 4 (1893) 137, t. 180B. Unona sphaerocarpa Blume Bijdr. (1825) 12, Fl. Jav. Anon. (1828) 39, £. 26. Malay Peninsula, Java, King’s collector 4002!, Backer 25558! FISSISTIGMA THORELI! (Pierre). Melodorum thorelii Pierre ex Finet & Gagnep. in Bull. Soc. Bot. France 53 (1906) Mém. 4: 183, et in Lecomte Fl. Gén. Indo-Chine 4 (1907) 97, t. 14, f. 1-6. Indo-China. FISSISTIGMA TONKINENSE (Finet & Gagnep.). Melodorum tonkinense Finet & Gagnep. in Bull. Soc. Bot. France 53 (1906) Mém. 4: 135, et in Lecomte FI. Gén. Indo-Chine 1 (1907) 100, t. 12, f. 1-6. Indo-China. 7 | 4 » ma | | | | i i | oJ ~ XV,2 Merrill: Melodorum of Loureiro 137 FISSISTIGMA UHRII (F. Muell.). Melodorum uhrit F, Muell. Fragm. 6 (1867) 2. Australia. git? FISSISTIGMA UONICUM (Dunn). qr \ Melodorum uonicum Dunn in Journ. Bot, 48 (1910) 323. Hongkong. FISSISTIGMA VERRUCOSUM (Hook. f. Th.). Melodorum verrucosum Hook. f. & Th. Fl. Ind. (1855) 119; King in Ann. Bot. Gard. Caleutta 4 (1893) 134, t, 177A. India. The Hainan specimen referred here by Hance probably represents some other species. FISSISTIGMA WALLICHII (Hook. f. & Th.). Melodorum wallichii Hook. f. & Th. Fl. Ind, (1855) 118; King in Ann. Bot. Gard. Calcutta 4 (1894) 183, t. 175A. Uvaria bicolor Wall. Cat. (1832) No. 6466, nomen nudum, non Roxb. Assam, Khasia, and Sylhet. FISSISTIGMA ZIPPELII (Miq.). Melodorum zippelit Mig. Ann. Mus. Bot. Lugd.-Bat. 2 (1865) 38. Sumatra. DOUBTFUL AND EXCLUDED SPECIES MELODORUM MICRANTHUM Warb. in Engl. Bot. Jahrb. 18 (1893) 190 = Pap- ualthia longirostris (Scheff.) Diels in Engl. Bot. Jahrb. 49 (1912) 140, f. 1, G-H (Goniothalamus longirostris Scheff., G. inaequilaterus K. Schum. & Lauterb.). MELODORUM LEFEVRII Baill. in Adansonia 10 (1871) 108. The type was from Indo-China. I can find no mention of the species in Finet & Gagnepain’s treatment of the Annonaceae of Indo-China in Lecomte Fl. Gén. Indo-Chine 1 (1917) 42-123. THE JUMPING PLANT LICE OF THE PALASOTROPICS AND THE SOUTH PACIFIC ISLANDS FAMILY PSYLLID4, OR CHERMIDA!, HOMOPTERA By Davip L. CRAWFORD Of the College of Hawaii, Honolulu THREE PLATES AND THREE TEXT FIGURES Our knowledge of the psyllid fauna of the Old World Tropics is gradually being extended, though it is far from complete. It appears that the subfamilies Pauropsylline and Carsidarine are much more extensively represented in this part of the world than in either the North or the South Temperate Zone, and perhaps are the chief groups in numbers of species. _ This family is widely distributed and often numerously rep- resented, even on the Pacific islands. Some very interesting evolutional features are to be noted in the psyllid faunz of certain island groups. The most remarkable is that of the Hawaiian Archipelago, where thirteen of the fifteen known species are apparently derivatives of one species long ago established here. Some of these species have diverged so far from the ancestral type, a Trioza, that two other generic groups embrace them. In the Malay Archipelago there is another interesting group, Megatrioza. This genus of the Triozinze is most numerously represented in this Archipelago but extends into the Philippines and south into Australasia, and one species has found its way as far as the Hawaiian Islands. Several other genera, more fully treated on subsequent pages, appear to have sprung from this one. Economically, this family is of less importance than the Aphidide and much less than the Coccide. A few species in the Palzotropics cause considerable damage, and no doubt when the habits of certain other species are known the number of economically important species will be augmented. Buckton’s Psylla isitis (probably the same as Arytaina punc- tipennis Crawford ’12: 431) is a pest on indigo; and Euphalerus citri, on citrus trees. The habits of the majority of the species herein described are unknown. 139 140 Philippine Journal of Science 1919 A detailed description of the anatomy and the taxonomy of the family, as a whole, is given in my monograph and need not be repeated here. However, a key to the subfamilies rep- resented in the Old World Tropics is included here for con- venience. This taxonomic work is made possible chiefly through the extensive collecting activities of Prof. C. F. Baker and of Mr. Frederick Muir. Key to the subfamilies of the Psyllide. a’, Frons usually visible as a small sclerite, bearing front ocellus; genz seldom produced into processes; body short and robust; thorax strongly arched, head strongly deflexed; coxal spurs short. Pauropsylline Crawford. a. Frons covered by genz and not visible, except a very narrow border around front ocellus; body usually more elongate; coxal spurs usually not very short. b'. Genal processes (cones) usually wanting and head deeply cleft in front between the antenne or, if head is not cleft, the forewings have one or two pseudo (cross) veins; body typically long and slender; forewings elongate and often with pseudoveins; basal tarsus of hind legs usually with only one clawlike spine at apex. Carsidarine Crawford. b*. Genal processes (cones) usually present and head not cleft in front,’ except as the divergent genal cones may give the appearance of a cleft; forewings never with pseudoveins. c’. Forewings with cubital petiole (media and cubitus with a common stem) ; basal tarsus of hind legs with two clawlike spines at apex. Psylline Puton. c’. Forewings usually without cubital petiole (the radius, media, and cubitus all emerging from basal vein at the same point); basal tarsus of hind legs with no clawlike spines at apex. Triozine Puton. PAUROPSYLLIN 4 Body robust; thorax strongly arched, broad; head short, much defiexed, rounded forward and downward, with eyes prominent; genal cones usually wanting or small and not prominent; frons usually visible, bearing the anterior ocellus at upper end; antenne usually short, often with long sete at apex. Legs slender and usually not large. Forewings large, usually broadly rounded at the apex, venation various. Many of the species of this subfamily are gall makers. Key to the genera. a’. Genz not produced into distinct conical processes, but sometimes more or less swollen beneath (somewhat conical in Paurocephala magni- frons Crawford. Mei 9 é& ~~ XV,2 Crawford: Jumping Plant Lice 141 b*. Hind legs not larger than middle pair, apparently not saltatory, hind coxe small, and coxal spur nearly wanting.... Apsylla Crawford. b*. Hind legs longer than middle pair, apparently saltatory; metacoxze much larger than mesocoxe, and metacoxal spur well developed. c’. Forewings membranous, more or less transparent. ad’. Media, cubitus, and radius not forking at same point from basal vein; hind wings not nearly wanting. e*. Antenne short, seldom longer than head and thorax combined; ; head usually much deflexed and retracted into large pro- thorax. : jf’. First marginal cell of forewing narrow and long, parallel- ing posterior margin; media-cubital stem as long or nearly as long as radial stem; metacoxal spur usually long, four to eight times as long as thick; antenne usually longer than width of head; genz sometimes swollen conspicuously beneath antennal bases Paurocephala Crawford. f*. First marginal cell not narrow and long; media-cubital stem usually shorter than radial stem or wanting. Metacoxal spur usually very short, one to three times as long as thick; antennz typically very short, seldom longer than width of head; genz seldom swollen. Pauropsylla Riibsaamen. e*. Antenne longer than head and thorax combined; head less deflexed and retracted into prothorax. f. Antenne not longer than body to tip of abdomen; meta- scutum with two prominent hornlike epiphyses projecting WMOWOAIG 660i eta 1 LSE Eas Boats Diceraopsylla Crawford. f. Antenne very long and slender, longer than body to tip of abdomen; metascutum without horns; gene swollen, often spherical beneath antennal bases. Heteropsylla Crawford. @. Media, cubitus, and radius of forewing forking from basal vein at same point (as in Triozinz) ; cubital vein unforked at apex; hind wings nearly wanting..............-..-...--...-- Leptynoptera g. nov. c*. Forewings nearly or quite opaque, more or less coriaceous; veins greatly contorted; front tibiw foliaceous........ Heteroneura g. nov. a. Genz swollen beneath into two more or less conical processes, these usually quite separated at base. b*. First marginal cell of forewing long and narrow with Cu: more or less parallel with margin of wing.............-- Paurocephala Crawford. b*. First marginal cell of forewing not long and narrow, Cu: not parallel with margin. : ce’. Body large; radial sector and media connected by a cross vein; first marginal cell small, much smaller than second. Phacopteron Buckton. ¢. Body small; cross vein wanting; first marginal cell very large, much larger than second; genal cones in some species scarcely Gineerhible 3 _.. Calophya Loew. 1664702 142 Philippine Journal of Science 1919 Genus PAUROPSYLLA Riibsaamen Key to the species.* a’. Forewing with pterostigma. b*. Antennz with two or more very long setz at tip. c’. Antenne much shorter than width of vertex between eyes; vertex showing median suture distinctly. dad. Antenne with eight long setz, tassellike. (Malay Archipelago.) P. brevicornis sp. nov. @. Antenne with two long setx, not tassellike. (Bengal.) P. nigra sp. nov. ¢. Antenne about as long as width of vertex between eyes, with three long setz; vertex with median suture not visible. (Malay Archipelago.) P. verticis sp. nov. 6°. Antennz without long terminal sete. é. Forewing broadly rounded, marginal cells poderatels large, ptero- stigma narrow; insect of medium size. (Ceylon.) P. spondiase Crawford (’15: 260). ec’. Forewing acute at apex, broad, marginal cells very large, ptero- stigma elliptical, broad; insect large. Malay Archipelago; CCT 2 Bg eR Sao eine tas P. apsylloides sp. nov. a’. Forewings without pterostigma. b*. Second marginal cell much larger than first; fourth furcal (M.-+:) angulately arched and touching radius. ec’. Insect minute, about 2 millimeters long from head to tip of folded wings; forewings with black spots scattered along veins; first marine cell about one-third as large as second. (Ceylon.) P. floccosa Crawford (15: 259). ¢e*. Insect larger, 3 millimeters long or longer; forewing veins without black dots; first marginal cell about one-tenth as large as second. (Bengal.) . P, tuberculata Crawford (’12: 430). b*. Second marginal cell not larger than first, usually subequal; fourth furcal not touching radius. e. Media and cubitus without stem, the radius, media, and cubitus forking from basal vein at same point (as in Trioza). (Phil- ippines and Java.).................... P. triozoptera Crawford (713: 296). ¢. Media and cubitus with a short stem, not as above. ad’. Head and thorax black, reticulately marked; anterior ocellus above, near posterior ocelli; abdomen large, not flattened. (Philippines and Java.) ............ P. udei Riibsaamen (’99: 264). ad". Head and thorax reddish or light brown, reticulation indistinct; anterior ocellus in front, more distant from posterior ocelli. (Beigel) we: P. depressa Crawford (712: 429). Pauropsylla brevicornis sp. nov. Plate I, figs. 1 and 2. Length of body, 1.3 millimeters; length of forewing, 1.9. General color reddish to chocolate brown, with venter of * Because of lack of knowledge concerning them I have not included in this key Pauropsylla ficicola Kieffer and P. globuli Kieffer. ee “= + XV,2 Crawford: Jumping Plant Lice 143 abdomen and metathorax whitish or pale greenish yellow; legs and antennez yellowish or pale; wings membranous, clear. Body small, but robust. , Head small, much deflexed, appearing very short from dorsal aspect, not as broad as thorax; vertex roundly convex from frontal view, without prominent foveal impressions, slightly elevated at posterior ocelli; anterior ocellus visible from front, at lower margin of vertex; frons not visible; genze mostly concealed by retraction of head toward thorax; clypeus and beak small. Eyes small. Antenne very short, not half as long as width of head, with several very long, slender sete giving the antennz a tassellike appearance. Thorax short and well arched. Leg’s short, small and slender, spines very inconspicuous. Forewings broad and rounded at apex, veins slender and not prominent. Male: Genital segment smail, short; forceps nearly as long as anal valve, very slender, tapering to acute apex, somewhat bent forward in apical third; anal valve broadest at base, converging to narrow apex. Female: Genital segment not quite as long as rest of abdomen, dorsal valve a little larger than ventral, trunc- ate at.apex. TENIMBER ISLANDS, Larat (Muir), 1 male and 2 females. Pauropsylla nigra sp. nov. Length of body, 1.2 millimeters; forewing, 1.6. General color shining black; vertex and thorax dorsad glossy black, abdomen black above and below with connective membrane orange yellow. Genz orange ventrad, thoracic venter orange in part and the rest black; antennz yellowish; legs yellowish except hind fem- ora, which are mostly black. Structurally very similar to Pauropsylla brevicornis, dif- fering in the following characteristics: Antenne about one-third as long as width of head, with only two long sete, these both at the tip and as long as the antenne. Forewings shaped and veined much as in the related species but a little smaller. Female genitalia similar but with a pair of small projections from caudal end between dorsal and ventral valves. INDIA, Pusa, Bihar (C. S. Misra), sii peceace 5, 1914, 1 female, found in the laboratory. Pauropsylla verticis sp. nov. Plate I, figs. 3 and 4. Length of body, 2.3 millimeters, forewing 2.5. General color 144 Philippine Journal of Science 1919 light brown with darker streaks on dorsum; venter of thorax pale whitish with a few brown streaks; abdomen mostly dark brown; antennz and legs light yellowish brown; wings clear. Head strongly declivous, not as broad as thorax. Vertex somewhat quadrate in outline, nearly perpendicular to axis of body, nearly as long as broad, median suture not visible, sur- face without depressions, broadly convex and not flat, front por- tion extended forward between antennal bases. Frons small, visible from in front. Genz slightly swollen beneath antennal sockets. Antenne short, about as long as width of vertex between eyes, with three long setz at end of each. Eyes large, recessive. Thorax broad, well arched. Legs rather slender; hind tibiz without basal spur. Forewings hyaline, clear, narrowly rounded at apex, veins narrow, yellow; pterostigma narrow, radius short; first marginal cell (in both forewings of the only specimen at hand) divided transversely across the middle. Female genital segment nearly as long as abdomen, rather slender, dorsal valve a little longer than ventral, with a serrated carina along apical dorsal portion; both valves acutely pointed. TENIMBER ISLANDS, Larat (Muir), 1 female. This species is remarkable because of its vertex, which is with- out a visible median suture. The divided marginal cell of fore- wing is probably not a constant specific character but only an abnormality. Pauropsylla apsylloides sp. nov. Plate I, fig. 7. Length of body, 2.9 millimeters; forewing, 4.2. General color light to dark brown, sometimes with dirty brown blotches on thoracic dorsum; eyes dark; forewing clear, with four brown spots on posterior margin and one at apical end of pterostigma. Body large and robust; surface reticulately marked. Head strongly declivous, not quite as broad as thorax. Ver- tex nearly perpendicular to axis of body, about three times as broad as length down median suture, arcuately concave on posterior margin; frons visible between genx, with large front ocellus at apex; clypeus large, visible below frons. Genz swollen beneath antennal sockets. Antenne about as long as width of vertex between eyes, sometimes a little longer, relatively thick, segments short and terminal sete rather short. Eyes large, recessive over propleurites. Thorax very broad and robust, strongly arched, and finely reticulately marked. Legs short and thick; hind tibie short, without basal spur, four thick, black spines at apex within. ~ ~ XV,2 Crawford: Jumping Plant Lice 145 Forewings very large and broad, hyaline, clear, acutely pointed, veins slender; marginal cells very large; pterostigma elliptical, radius short. Abdomen thick and short. Male genital segment short; for- ceps as long as anal valve, slender, not much bowed outward, tip slightly enlarged and provided with a short, downwardly and inwardly directed point. Anal valve slender, apical third in- clined caudad and tubular in shape. Female genital segment longer than abdomen, much narrower and acuminate, dorsal valve a little longer than ventral and both acutely pointed. TENIMBER ISLANDS, Larat (Muir), 1 female. BORNEO, San- dakan (Baker), 1 male. SOUTHERN CHINA, Macao (Muir), 1 female. HAWAII, Oahu (A. Koebele), 1 imperfect specimen ap- parently of this species. The last was no doubt an immigrant, perhaps taken on some recently imported plant. It appears not to have become established on this island as it has not been again. taken in subsequent collecting during many years. The male specimen differs from the others in being darker colored and in having the antenne nearly as long as width of head including the eyes, the basal tarsus of hind legs a little longer, and the cubital petiole (M-—+ Cu) longer than the stem of the cubital vein, while in the females it is shorter. Whether or not these are specific differences can scarcely be determined until examples of both sexes from each locality are available for study. This insect is somewhat suggestive of Apsylla cistellata, an Indian species, but seems to be closer to Pauropsylla than to Apsylla, Pauropsylla udei Riibsaamen. Pauropsylla udei RUBSAAMEN, Ent. Nachrichten 25 (1899) 262-266. Pauropsylla bakeri CRAWFORD, Philip. Journ. Sci. § D 10 (1915) 258. Length of body, 2 millimeters; length of forewing, 2.6; length of antenne, 1. General color dark reddish to chocolate brown; antenne, legs, ventral portion of genz, and genital segment orangé yellow or paler; eyes pale; wings clear. Body robust. Head short, defiexed, not as broad as thorax; vertex reticu- lately marked, elevated at posterior ocelli, extending forward between antennal bases and covering frons, inclosing anterior ocellus so that latter appears to be on vertex; frons scarcely visible; genz swollen beneath antennal bases, without cones; clypeus small. Eyes rather large; front ocellus near posterior ocelli and visible from above. Antenne a little longer than width of head, slender. 146 Philippine Journal of Science 1919 Thorax very broad, arched, robust, surface finely reticulately marked. Legs slender; hind legs long. Forewings very broad distally, more than half as broad as long, clear, very thin and delicate; veins slender; marginal cells small. " Abdomen moderately large. Male genital segment small; for- ceps simple, nearly one-half as wide (lateral view) as long, roundly acute at apex, arched; anal valve about one-half longer than forceps, simple, truncate at apex. Female genital segment about one-third as long as abdomen; valves about equal in length, acutely pointed at apex. JAVA, Salatiga (Leewwen-Reijnvaan), April 5, 1912, on Ficus variegata Blume, 6 males and 10 females. These psyllids pro- duce galls resembling those of cecidomyid flies. Type locality—SUMATRA (Ude). Distribution.—LuzON, Laguna Province, Mount Maquiling (Baker). JAVA, Salatiga (Leeuwen-Reijnvaan), April 5, 1912. The type specimens were taken from galls on an undetermined plant said to belong to the family Rubiaceew. The galls were subspherical, flattened below at attachment to leaf, covered with thickly set, unbranched spines and opening after the maturity of the insect into a flowerlike shape by the splitting and spread- ing of sectors of the spherical gall. Pauropsylla triozoptera Crawford (’13: 296). Type locality—Los Bafios, Luzon, Philippines. Several specimens, apparently belonging to this species but with slightly larger and more broadly rounded wings and larger body with longer legs, were taken in JAVA, Sindaglaija (Muir), 1 male and 4 females. Luzon, Benguet, Baguio (Baker), 1 male, also larger than the type specimens. The differences between the Philippine and Javanese forms are slight and such as might naturally develop from isolation and separation. Pauropsylla triozoptera setifera Crawford (713: 297). By some error in the manuscript, this variety was incorrectly placed with Pauropsylla triozoptera instead of with Pauroce- phala psylloptera. The specimens bear the correct label, but somehow the transfer in the manuscript was made. This error should be corrected to read Paurocephala psylloptera setifera Crawford. Pauropsylla tuberculata Crawford (’12: 480). One mutilated specimen appears to be very close to this In- dian species. One forewing is intact and most of the other, but XV,2 Crawford: Jumping Plant Lice 147 the head and abdomen including the genitalia ar® mutilated beyond recognition of characters. The venation is very similar . but not quite identical. In the Indian species the medial vein forks just about midway between the cubitus and radius, while in the Pacific form the furcation of the media is nearer to the radius. It is possible that an unmutilated specimen would show other differences of specific value. TENIMBER ISLANDS, Larat (Muir), 1 specimen of somewhat doubtful identity, tentatively included here. Genus LEPTYNOPTERA novum Head defiexed, rather short, vertex broader than long, an- terior angles typically produced over base of antenne; genz scarcely produced into cones; frons not covered by gene. An- tenn rather short, second segment not barrel-shaped but pro- duced distally on lower side. Thorax well arched. Legs long. Forewings membranous, broad distad, cubital petiole and first marginal cell both wanting, the cubitus reaching margin un- forked. Hind wings nearly wanting, exceedingly short. Type of the genus, Leptynoptera sulfurea sp. nov. The position of this genus is a little doubtful, since it bears similarities to both Trioza and Pauropsylla. The shape of head, exposure of frons, shape of forewing, and some venational char- acters indicate a closer relationship to Pauropsylla than to Trioza. The absence of cubital petiole appears to be not ex- clusively a character of Triozinz, as several unmistakable examples indicate. The unforked cubitus is a characteristic approached more or less closely by several species of Pawr- opsylla, suggesting that that vein of the forewing is subject to considerable variation. On the other hand, the same vein in Triozine appears to be more constant than many other char- acters of the body. The aborted hind wings are characteristic not only of this genus but also of a few species of Trioza, and to a lesser degree of species of Pauwropsylla and several other specialized genera. There appears to be a more or less wide tendency in the entire family toward reduction of the hind wings in function, strength, and size. Leptynoptera sulfurea sp. nov. Plate I, figs. 5 and 6. Length of body, 1.7 millimeters; forewing, 2.6. General color bright sulphur yellow, with four narrow, inconspicuous, brown stripes on mesonotum; abdominal tergites with white transverse bands; eyes pale whitish; legs pale, femora whitish; antennal segments 1 and 2 yellow, remainder whitish with three black 148 Philippine Journal of Science 1919 bands, one at end of segment 4, another at tip, and another midway between; forewings clear except brown in anterior basal cell. Head not as brown as thorax, strongly declivous, with long, sparse pubescence; vertex a little more than half as long as broad, anterior portion bent downward, posterior ocelli elevated, with a rather deep foveal impression on each side of median line, anterior outer corners produced into -a short, upturned epiphysis over each antennal base. Genz a little swollen beneath antennal sockets, bulging slightly forward of vertex. Frons wholly visible, broadest at base; clypeus moderately small; ros- trum short. Antennz a little longer than width of head, banded black and white, second segment with one side of apex produced. Thorax broad, strongly arched, sparsely clothed with long hairs, mesonotum very short, transverse; pseudonotum (behind metanotum) large, long, with a pair of slender, upturned, whitish horns between this and basal segment of abdomen. Legs long, hind tibize long, with a very small basal spur and short distal spines. Forewings somewhat rectangular, hyaline, veins narrow, pterostigma wanting, radius not straight, cubitus unforked. Hind wings reaching about to primary fork of basal vein, veins practically wanting. Abdomen short (genitalia mutilated). Mo.Luccas, Amboina (Muir), 1 example. Genus PAUROCEPHALA Crawford Key to the species. a‘, Forewing with pterostigma; vertex rounded forward and down; head strongly deflexed. b*. Metascutellum with a conical, usually acutely pointed, epiphysis dor- sad; female genital segment usually defiexed at right angies to abdomen. ¢. Antenne distinctly longer than width of head; posterior ocelli prominently elevated; tip of female genitalia straight. d’. Body surface and wing veins not hairy or with only very in- conspicuous pubescence. (Philippines, Ceylon, and Malay Archipelago. } 02s oe P. psylloptera Crawford (13: 294). d’. Body surface and wing veins conspicously hairy. (Philippines, Fiji, and Singapore.).... P. psylloptera setifera Crawford (’13: 297). ¢. Antenne not longer than width of head, usually not as long, posterior ocelli not much elevated; tip of female genitalia curved outward. d’. Antenne about as long as width of head with eyes; forewing with three dark spots on costal margin; insect very small. {PRA INOE ct eae P, minuta sp. nov. i a sa 4 I~ ~ xVv,2 Crawford: Jumping Plant Lice 149 d@’, Antenne shorter, about as long as head between eyes; forewings with five or six marginal spots; insect a little larger. (Phil- TPDINOR.) sic. ee P. brevicephala Crawford (’17: 163). b*. Metascutellum without conical epiphysis; female genital segment typi- cally horizontal. c’, Frons large and prominent, not covered by gene; forewings con- verging in distal half to apex; gene swollen subconically beneath antennal bases. da‘. Forewings narrowly rounded at apex; hyaline; frons very large. (Southern Mexico.) ................ P. magnifrons Crawford (’14: 42). a’. Forewings subacute at apex, not hyaline and often nearly opaque; frons smaller. (Malay Archipelago.)........ P. conigera sp. nov. c’. Frons smaller, mostly covered by gene; forewings not converging, broadly rounded; gene scarcely swollen beneath antennal bases. (Southeastern United States.) P. ilicis (Ashmead) Crawford (’14: 43). a’, Forewing without pterostigma; vertex more flattened and head less deflexed. b*. Forewings conspicuously maculated; radius and medial veins sinuate. (Singapore. ) P. maculata sp. nov. b*. Forewings not maculated, clear; veins not sinuate. (Philippines.) P. orientalis Crawford (’15: 261). Paurocephala psylloptera Crawford. This appears to be a common species, widely distributed in the South Pacific regions. Since additional material has come to hand it seems necessary to supplement the original descrip- tion which was drawn up from six specimens. The color of the male is not uniformly darker than the female; both sexes are usually very dark chocolate brown to dull black; lighter forms seem to be less matured, the color deepening with age; vertex and thoracic dorsum sometimes with orange yellow streaks but more commonly uniformly dark, antenne usually as dark as vertex; legs often lighter in color or even yellowish. Body surface often not hairy; vertex and thoracic dorsum finely, reticulately marked. The beak is moderately long and the man- dibular setz sometimes exserted to a length of 1 millimeter or less. The hind tibize have a row of about seven black spines on outside, several spines at apex and are more or less hirsute. The metacoxal spurs are long and slender. Pterostigma of fore- Wings usually dark colored, veins black. In the Philippine Islands this species has been collected in Luzon, Los Bafios (Baker), on Ficus ulmifolia; and on Mount Maquiling (Baker). CEYLON, Peradeniya (Rutherford), on Ficus hispida and Ficus asperrima. Additional specimens are now before me from TENIMBER ISLANDS, Larat (Muir), Decem- _ ber, 1907, 18 males and 18 females. 150 Philippine Journal of Science 1919 Paurocephala psylloptera setifera Crawford. Pauropsylla triozoptera setifera CRAWFORD, Philip. Journ. Sci. § D 8 (19138) 297. Several specimens from Fiji Islands appear to belong to this variety. The color is uniformly a little darker than the average of the Philippine forms, but the markings on vertex and notum are similar and the antennz are similarly colored, darker at tips of segments 3 to 8. The body surface is a little less hirsute than in the Philippine forms. By a mechanical error this variety was originally placed with the wrong species. Fist ISLANDS (Muir), 2 males and 3 females, 1905. SINGA- PORE (Baker), 1 male. The specimen from Singapore differs from the type of the variety in being orange yellow with some darker streaks on notum. The body surface and wing veins are less hirsute, but the male genitalia are quite similar. Paurocephala brevicephala Crawford. Pauropsylla brevicephala CRAWFORD, Philip. Journ. Sci. § D 10 (1917) 163, pl. 1, fig. 11. This Philippine species was misplaced in the genus Paur- opsylla. The wing venation, form of vertex and presence of the metascutellum epiphysis point to its close affinity with species of Paurocephala. A few additions to the original description are here made. Vertex not prominently elevated at posterior ocelli, without foveal impressions. Surface of vertex and dorsum finely re- ticulately marked. Metascutellum produced upward into a thick epiphysis with a very small, acute point. Hind legs longer than others, but relatively shorter than in some other species of the genus. Female genital segment turned downward, with tip of dorsal valve curved outward, acute. The shortness of the an- tenn is an unusual character in this genus. MINDANAO, Davao (Baker), 3 females. Paurocephala minuta sp. nov. Plate I, fig. 8. Length of body, female, 0.9 millimeter; length of forewing, 1.4. General color brown to dark brown, with legs, venter, and antennsz lighter or pale; a whitish or pale spot on thorax at base of each forewing. Body small. Resembling P. psylloptera in many characteristics, and ap- parently closely related to that species. Vertex relatively broader and with foveal impressions and postocellar elevations less prominent. Antenne short, scarcely as long as width of i~ XV,2 Crawford: Jumping Plant Lice 151 head. Thorax well arched, broad; metascutellum with a large, blunt epiphysis; legs short; hind fibize relatively shorter than in related species, with fewer spines; forewings small, similar in shape to related species, but pterostigma relatively longer and narrower and less-dark colored; three dark spots on costa] mar- gin, one at base of pterostigma, another at tip of same, and a third at tip of radius. Female genital segment deflexed but shorter than in P. psyll- optera, tip of both valves very acute and flexed outward. Luzon, Laguna Province, Los Bafios (Baker), 1 female. Paurocephala maculata sp. nov. Plate I, fig. 9. Length of body, 1.2 millimeters; length of forewing, 1.4. Gen- eral color reddish brown, more or less glossy, smooth, venter of thorax and abdomen tawny, femora orange, tibiz darker; antennal segments 1 and 2 reddish, 3 to 8 white or pale, 9 and 10 black or brown; forewings beautifully maculated, mostly brownish with several large clear areas and numerous, minute, dark spots in the brown. Head short, nearly as broad as thorax, deflexed, vertex smooth and shining, without prominent depressions or elevations, rounded downward in front sharply, anterior ocellus visible from front only; frons visible from in front and below, small; gene swollen beneath antennal bases, making an emargination in which the clypeus is visible. Antenne a little longer than width of head, segment divisions not distinct. Thorax arched, broad, dorsal surface smooth and glossy ex- cept a large elevated callus on each side of dorsulum (meso- scutum) rough and lighter red in color; metascutellum with a long, low elevation along median line. Legs short. Forewings. a little longer than body, broadly rounded at apex, conspicuously maculated; radius and media sinuate; pterostigma wanting, first marginal cell long and narrow. Female genital segment moderately large, not deflexed, dorsal valve a little longer than ventral, acutely pointed and tip flexed outward. SINGAPORE (Baker), 1 female. This species is somewhat different from the type of Pauro- cephala (in the absence of pterostigma and the sinuate veins), but provisionally, at least, it is included in this genus. Paurocephala conigera sp. nov. Plate I, fig. 10. Length of body, 2.7 millimeters; forewing, 4. General color orange to yellowish brown or brown; venter and legs a little 152 Philippine Journal of Science 1919 lighter in color; antenne dark distad; forewings semitrans- parent, usually brownish or sometimes brown and opaque. Head strongly declivous, fully as broad as thorax. Vertex a little more than half as long as broad, subperpendicular to axis of body, with a shallow sulcus across posterior portion, surface somewhat convex. Frons visible in front, with large ocellus. Genz swollen beneath antennal bases into subconical processes very broad at base, subacutely pointed and about one-third as long as vertex. Antenne a little less than twice as long as width of head, slender. Thorax not strongly arched. Legs rather short, hind tibize with a small spur at base. Forewings long and narrow, sub- acute at apex, membrane thick, semitransparent or often opaque because of the brown color. Abdomen long. Male forceps slender, little more than half as long as anal valve, acutely pointed; anal valve long and rather slender, tapering from base to subacute apex. Female genital segment about half as long as abdomen, dorsal valve with a large, rounded convexity dorsad, acutely pointed, a little longer than ventral valve. Mo.uccas, Amboina (Muir), 1 male and 3 females. This resembles P. magnifrons, a tropical American species, more closely than other species of this genus. Quite possibly these two species should be referred to a distinct genus. Genus HETERONEURA novum Head somewhat as in Pauropsylla, much deflexed and short; vertex roundly convex; frons inferior, with anterior ocellus visible only from in front; genz small, not much swollen, cones wanting; eyes rather small; beak short; antenne as long as width of head or longer. Thorax well arched; legs short, front tibie flattened, thin and foliaceous; hind tibiz with a tooth near apex; metacoxal spurs short. Forewings coriaceous, opaque, broad; marginal cells long and narrow as in Paurocephala; second marginal cell flexed toward costal margin; radius contorted; pterostigma closed. Type of the genus, Heteroneura oceanica sp. nov. This genus appears to belong to the subfamily Pauropsylline, but differs from other genera of this subfamily in the coriaceous forewings, the much contorted radius, and the foliaceous front tibie. Heteroneura oceanica sp. nov. Plate I, fig. 11. Length of body, 1.8 millimeters; length of forewing, 2; length of antennz, 1.2. General color dark brown; meso- and meta- Sr ST XV,2 Crawford: Jumping Plant Lice : 158 scutellum dirty white or yellowish; antennal segments 3 to 8 dark at apex and light basally, 9 and 10 black; forewings opaque, brown, small area at base and another at tip whitish, veins tawny; between the veins rows of many, small, light-colored patches, giving a very striking and characteristic appearance to the wings. Body surface sparsely covered with rather long hairs. Head short, not as broad as thorax, much deflexed, hairy; vertex roundly convex, without foveal impressions, scarcely ele- vated at posterior ocelli; frons scarcely visible; front ocellus not visible from above, close to clypeus; genz# small, somewhat swollen beneath each antennal insertion, without cones. An- tennz twice as long as width of head, slender. Thorax broad, robust, hairy. Legs rather short; front and middle pairs of tibiz conspicuously flattened and somewhat fo- liaceous and front femora somewhat flattened in distal half; hind tibiz with a large, prominent tooth near apex. Forewings coriaceous, not flat but somewhat convex, broadly rounded at apex; venation very singular and characteristic; pterostigma shaped like a marginal cell; radius much contorted, second mar- ginal cell long and narrow and parallel with posterior and apical margins of wing. Abdomen about as long as thorax. Female genital segment short, about one-fourth as long as abdomen; dorsal valve a little longer than ventral, both slender and acute at tip, dorsal valve with an acute epiphysis at apex. SINGAPORE (Baker), 1 female. LUZON, Bataan Province, Mount Limay (Baker), 1 female. BASILAN (Baker), 1 female. Genus HETEROPSYLLA Crawford This American genus, comprising seven described species, is a well-marked group. One species apparently abundant in the South Pacific is closer to this genus than any other. Although it differs in some important characteristics from the American species, it is nevertheless included herein for the present. Heteropsylla longicornis sp. nov. Plate I, fig. 12. Length of body, 2 millimeters; length of forewing, 2.1; length of antenne, 4. General color light to dark reddish brown; tibie, tarsi, metapleurz, and a small area beneath each eye pale; sev- eral streaks on thoracic dorsum white or pale; two longitudinal rows of white spots on each side of abdomen. Body surface pubescent. Head nearly as broad as thorax, short, deflexed; vertex broad, 154 Philippine Journal of Science 1919 triangular on each side of median line with long side next to eye, prominently elevated at posterior ocelli, with median suture conspicuous and convex between antennal sockets and median suture, with an elongate sulcus distally on each side of median line; anterior ocellus scarcely visible from above; frons mostly covered by genze; gen# very expansive, but not swollen, contin- uing to clypeus in nearly the same plane with distal portion of vertex. Clypeus moderately large, beak very long and prom- inent. Eyes very large. Antennz one and a half to two times as long as body without wings, very slender and threadlike, darker apically. Thorax broad, weil arched; propleurites covered by recessive eyes, metascutellum elevated into a large, prominent epiphysis. Legs large, hairy; hind tibiz with spur at base and several spines at apex; basal tarsus of third leg with two black spines at apex. Forewings membranous, clear, with several (usually six) black or brown spots along posterior and apical margins; veins setiferous. Abdomen somewhat laterally compressed dorsad, and more or less triangular in transverse section, sharply angled above; each segment with a dense fringe of hairs on posterior margin. Male genital segment not large; forceps not quite as long as anal _ valve, nearly as broad; anal valve long and slender, inclined toward forceps, truncate at apex. Female genital segment about half as long as abdomen, both valves subequal in length, acute, hairy. TENIMBER ISLANDS, Larat (Muir), December, 1907, 57 males and females. Mo.tuccas, Amboina (Muir), January, 1908, 1 male and 12 females. This differs from the other species of the genus in the larger size, longer antennz, armed hind tibiz, and epiphysis on meta- scutellum. The other species are American. An imperfectly preserved specimen from the Philippines seems to belong to this genus. Phacopteron lentiginosum Buckton. Phacopteron lentiginosum CRAWForRD, Rec. Indian Mus. 7 (1912) 420, pl. 33, figs. A, B, F; pl. 35, fig. A. Nymph.—tThe nymph of this species is very large and robust, not flattened as in most psyllids but more cylindrical or saccate, the shape characteristic of most gall-forming species. Head very short, much narrower than thorax; eyes not bulging, small; antennze about as long as width of head, stout; clypeus large and prominent. Thorax nearly as broad as abdomen, legs short a XV, 2 Crawford: Jumping Plant Lice 155 and stout. Abdomen large, the broadest part of body broadly rounded caudad. Four adult specimens and three nymphs of this’ large and striking psyllid are before me, collected in Pollibetta, Coorg, India, on May 27, 1914, from galls on leaves of an unidentified tree. CARSIDARIN This subfamily has much larger representation in tropical countries, only a few species being found as far north as the southern United States. Previous to 1911 so few representa- tives of this subfamily had been discovered and examined that those few were assigned positions in genera variously related to the other Psyllidze; but when an increasingly large number of forms from many tropical localities began to come to hand, it became apparent that a comparatively large tropical fauna of this type existed. As our knowledge of this group advances we shall no doubt have to recast our taxonomic conceptions. One of the most striking peculiarities of the representatives first examined was the deeply cleft head, with the thickened antennz enhancing the cleft appearance, and the absence of genal cones. Exam- ination of more species of this subfamily, however, ,has shown that these characteristics are not to be found throughout the subfamily, though the genal cones usually are absent. The body always has a characteristic elongate form, and the wings have a peculiar venation; though this latter peculiarity: is one not easily explained nor easily appreciated by one unfamiliar with the group. As the tropical countries of the world are more carefully explored for their insect life, it is probable that many new and perhaps quite different types of this subfamily will be brought to light, necessitating changes in our taxonomic disposition of these very interesting little creatures. The following subfamily description is intended to supersede the previous definitions of this group: Body typically elongate, often slender. Head usually cleft in front, vertex more or less concave on front margin and genz swollen out forward beyond vertex at each antennal base, and the latter often swollen and increasing the birostrate appearance of the head, but in at least two genera, Nesiope and Tenapha- lara, head riot birostrate. Genal cones usually wanting or very small and inconspicuous. Beak very long. Eyes hemispherical, usually not at all recessive. Antenne usually long, often thick- 156 Philippine Journal of Science 1919 ened and sometimes very hairy. Thorax not strongly arched. Hind tibize usually with a spur at base and spines at apex; basal tarsus of hind legs usually with one black claw (two in Homotoma, but the relationships of this genus are not clear). Forewings always membranous and hyaline, usually more or less acutely pointed, often with a veinlike callus or two con- necting veins distad. There is a scarcely definable, characteristic manner of branching of medial and cubital veins. Although I have seen no representatives of certain Autralian genera, the illustrations and descriptions of these point to an affinity with Carsidarine so clearly that they have been included here, tentatively. Homotoma is an old genus and still of very doubtful relationship. The presence of two claws on the basal tarsus of the hind legs is distinctly a characteristic—and appa- rently an important one—of Psylline and not of Carsidarine. The genx, also, often are produced into distinct cones. It is possible that the genus should be referred to another subfamily, but for the present at least it may be retained here. Key to the genera. a’. Cubitus and media with common stalk (cubital petiole). b*. Antenne thick throughout, not longer than body to tip of abdomen. e. Antenne thickly hirsute and often carinate; gene usually with conical processes directed downward on each side of clypeus. Homotoma Guerin. ec’. Antenne not hirsute, nor carinate, but terete; gene not produced into cones . Epicarsa Crawford. b*. Antenne slender at least beyond third segment, usually distinctly longer than body (except in Macrohomotoma), not hirsute. c’. Head birostrate in appearance, by the projection forward of genz on each side of midline and attachment of antenne at end of projecting genz. ad’. Outer anterior angles of vertex elevated above genz, more or less hornlike; insect usually very large. e’. Female genitalia with ventral valve nearly as long as dorsal, latter without thick fringe of hairs; vertex produced into a large horn over each antennal socket; hind tibie without basal spur, with very large spines at apex; first antennal segment very long; forewing with marginal cells nearly TL 30h aia ceca Dynopsylla Crawford. e’?. Female genitalia with ventral valve very short, dorsal valve many times larger and very thickly fringed with stiff hairs along ventral margin, vertex sharply angled but not produced into horns in front; hind tibiz with basal spur; first marginal cell of forewing much smaller than second. Thysanogyna g. nov. @. Vertex not as above. e’. Forewings with pterostigma, hind tibie usually with basal spur; antenne very long and slender. ~ XV,2 Crawford: Jumping Plant Lice 157 f’. Transverse callus connecting radius and media; hind tibie with spur at base; pterostigma open or closed. Carsidara Walker. f°. Transverse callus in forewing wanting... Mycopsylla Froggatt. e*, Forewings without pterostigma; hind tibie with or without - basal spur. f’. Antenne long and very slender. g. Vertex produced in front into two long points (not genal cones); hind tibiz without basal spur. Geijerolyma Froggatt. g. Vertex not so produced, but more or less deeply cleft in front; hind tibie with basal spur. Tyora Walker (Mesohomotoma Kuwayama). f°. Antenne long but not slender, vertex deeply cleft in front. Freysuila Aleman. _ f*. Antenne very short and slender; vertex not deeply cleft in TONG. = Macrohomotoma Kuwayama. c. Head not birostrate, vertex more or less quadrate. d@. Margins of vertex turned up in a narrow rim; female genital segment long and acuminate.......00002220.....--..--- Nesiope Kirkaldy. @. Margins of vertex not turned up, but vertex usually inclined roundly forward and downward; female genital segment with dorsal valve constricted subapically.... Tenaphalara Kuwayama. a. Cubitus and media without stalk (triozine). : b*. Antenne thick and hairy; genz sometimes swollen beneath into more OT less CONICRL: PrOCOSSeS...c sec0u.,. 2k teaes ican Bactericera Puton. b*, Antenne slender and not hairy; genze without cones beneath. c’. Head more or less birostrate (as in Carsidara).... Rhinopsylla Riley. c’. Head not at all birostrate; vertex nearly quadrate, declivous. Tenaphalara triozipennis Crawford. Genus THYSANOGYNA novum Head cleft in front as in other genera of this subfamily; genal cones almost or quite wanting; vertex more or less quad- rate, concave, outer anterior corners sharply angled and ele- vated, with anterior ocellus in front at notch of frontal cleft; antennez slender except segments 1 and 2, not excessively long; beak moderately long. Thorax large; legs large; hind tibie with spur at base and large spines at apex; basal tarsus of hind legs with one black spine at apex. Forewings large, with closed pterostigma, short radius and indistinct pseudovein between radius and media; first marginal] cell much smaller than second. Abdomen large; female genitalia somewhat dissimilar from other genera; dorsal valve large, with a dense fringe of hairs on lower margin; ventral valve very small and often retracted into preceding abdominal] segment; ovipositor large, broadly truncate, edges fluted. Type of the genus, Thysanogyna minor Crawford. 166470——3 158 Philippine Journal of Science 1919 This species was first described from a single male as belong- ing to Dynopsylla, but some characters were noted in which the species did not conform to the generic type. Subsequently many other specimens of this species have come to hand and among them females. The genitalia of the female at once show a marked difference not only from Dynopsylla but from most other genera. This species, therefore, is designated the type of a new genus, which appears to be related to other genera of Car- sidarine. Since the first description did not include the female and was somewhat inadequate in a few other respects, a more detailed specific description is given herewith to replace the other. Thysanogyna minor Crawford. Plate II, figs. 2 and 4. Dynopsylla minor CRAWFORD, Philip. Journ. Sci. @ D 12 (1917) 263. Length of body, 2.7 millimeters; length of forewing, 5; greatest width, 2.3. General color light brown with darker brown streaks on thoracic dorsum or the color may be dark brown with lighter streaks; antennz with apical third or half of segments 3 to 8 and all of 9 and 10 black. Body surface more or less covered with white flocculent excretion. Insect large and robust. Head about as broad as prothorax, but mesothorax consider- ably broader, not much deflexed, cleft in front. Vertex a little more than half as long as broad, deeply concave, the median suture prominent, and the two halves nearly quadrate, the outer anterior corner acute and upstanding, giving the appearance of a horn (somewhat as in Dynopsylla); anterior margin of ver- tex somewhat emarginate, the protruding genz at base of an- tennz much emphasizing this emargination. Anterior ocellus in front at base of cleft; frons visible as a small and narrow sclerite between gene and below front ocellus. Genz very large, protruding in front to form antennal bases, beneath the head and just in front of labrum produced into a pair of very small wartlike cones. Labrum not large; beak about as long as third antennal segment. Antennz moderately long and slender, not as long as body but usually as long as head and thorax or a little longer. Thorax very large, well arched and broad, sparsely hairy. Pronotum large; mesonotum very large; metanotum with a pair of short blunt horns or epiphyses caudad. Legs long and large; hind tibize with spur at base and several long fingerlike spines at apex; tarsilong. Forewings large and long, narrowly rounded at apex, transparent, with four small brown spots along poste- le XV,2 Crawford: Jumping Plant Lice 159 rior margin and one in pterostigma; an indistinct pseudovein connecting radius and media, as in Carsidara. Abdomen large. Male anal valve a little longer than forceps, broadest at middle; forceps somewhat fusiform, thick at middle, apically tapering to a narrow and truncate end. Female genital segment large; dorsal valve very large, with a small, curved, acute epiphysis caudad, and a large anal opening dorsad; the lower margin densely fringed with slender hairs, as a brush. Ventral valve very small and retracted more or less completely into preceding abdominal sclerite with an acute epiphysis at apex. Ovipositor large, very broadly truncate at apex, appa- rently permanently exserted, margin fluted. LuzON, Laguna Province, Mount Maquiling (Baker), 1 female: Tayabas Province, Malinao (Baker), 2 females. TENIMBER Is- LANDS, Larat (Muir), December, 1917, 5 males and 13 females. Type locality —LuzoN, Laguna Province, Los Bajios. The Larat specimens very closely resemble those from the Philippines except in some very minor color characters and the following two wing characters: Forewing of Larat specimens a little less acute at apex, and radius shorter; in Philippine specimens the radius joins the margin at a point as far from tip of pterostigma as the length of latter; in Larat specimens the length of pterostigma is nearly twice the length from tip of radius to tip of pterostigma. This probably indicates the incipiency of another species. Genus TYORA Walker This genus is very close to Carsidara, differing chiefly in the absence of pterostigma in the forewing. Froggatt’s Tyora ster- culiz, however, does possess the pterostigma. I am inclined to believe that Tyora should be merged with Carsidara, but until more material is available for comparison I am unwilling to make this change. Tyora indica, described below, is evidently very closely related to T. hibisci Froggatt but has no claw nor spine at apex of hind femora said to be present in both the Australian species of this genus. Mesohomotoma Kuwayama is apparently very close to Tyora, differing in no essential character whatever, so far as I can discover, from Kuwayama’s description. However, until I have examined a representative of the Japanese genus, I am loath to make any alteration. Tyora indica sp. nov. Plate II, fig. 1. Length of body, male, 2.4 millimeters; female, 3.3; length of forewing, male, 3.5; female, 4.7; length of antenne, 1.9. Gen- 160 Philippine Journal of Science 1919 eral color orange yellow to yellowish green, dorsum of thorax sometimes with paler streaks or often white stripes in male; antennal segments 4 to 8 black at tip, 9 and 10 all black; wings clear with several small dark spots along margins. Body sur- face often covered with white flocculent material. Head small, little broader than prothorax, not defiexed, deeply cleft in front, vertex with edges and median line elevated into a narrow rim with two deep fossze extending from posterior margin slightly divergently toward antennal bases; in the male the lateral ridges and ridge on each side of median suture are often whitish; anterior outer angle of vertex acute and forming a small hornlike epiphysis over eye; anterior ocellus at front margin of vertex, a little above apex of frontal cleft and visible from above. Genz not swollen into cones beneath, but project- ing forward at antennal bases forming frontal cleft. Frons very narrow, scarcely visible between gens. Antenne a little more than half as long as body, slender. Beak moderately long. Thorax not much arched, not broad. Legs rather short and slender; hind tibize with a small spur at base and several black spines at apex; basal tarsus of hind legs with one black spine at apex. Forewings hyaline, long, about two and three-fourths times as long as broad, with pseudovein connecting short radius and media; pterostigma wanting. Male abdomen slender; genital segment small; forceps very slender, arched, acutely pointed; anal valve shorter than forceps, relatively very broad, expanded laterally. Female abdomen larger, broader; genital segment nearly as long as rest of ab- domen; dorsal valve bulging upward and hirsute midway and abruptly constricted in apical third to an acute point; ventral valve tapering to an acute point, nearly as long as dorsal. Mo.uccas, Amboina (Muir), January, 1908, 5 males and 6 females. INDIA, Mercara, Coorg (Y. R. Rao), May 24, 1917, 3 females. This species is very close to Tyora hibisci Froggatt, differing in some color characters and in the genitalia of both sexes. Me- sohomotoma camphorz Kuwayama is very similar to this species and apparently is not generically distinct. Genus NESIOPE Kirkaldy Nestope KIRKALDY, Proc. Linn. Soc. N. S. W. 33 (1908) 389. Carsidaroida CRAWFORD, Philip. Journ. Sci. § D 12 (1917), 165. Nesiope and Tenaphalara differ sharply from the remainder of the subfamily in the shape of the head, which in these two genera is not at all cleft as it is in the others, Notwitstanding te XV,2 Crawford: Jumping Plant. Lice 161 this disparity there is undoubtedly a close relationship between these two and the others, as indicated by similarity of venational and certain other body characters, as the presence of but one claw on the basal tarsus of hind legs. The close similarity between Kirkaldy’s Nesiope and my Car- sidaroida did not appear until the latter had been published, wherefore it becomes necessary to merge my genus with the other. The differences between the species are brought out in the following key: Key to the species. a. Thoracic dorsum conspicuously striped with black (or brown) and yellowish (or orange) longitudinal bands; length of insect more than 5 millimeters. b*. Forewings scarcely spotted, more or less fumate; insect large, usually over 7 millimeters long. (Philippines, Singapore, and Am- Dena & N. heterocephala Crawford (’17: 165). 0’, Forewings conspicuously spotted in apical half with many small brown and large black spots. Insect about 5 millimeters long. (Larat and: Java,) 2222. oac. N. heterocephala intermedia var. nov. a’. Thoracic dorsum not as above, but with “a central pale line down dorsulum and mesonotum very narrowly bordered with black; sides of mesonotum broadly ferruginous, with two pale, curved, very narrow lines;” forewings spotted with brown in apical half. (Fiji.) N. ornata Kirkaldy (’08: 389). Nesiope heterocephala Crawford (’17: 165). CELEBES and AMBOINA (Muir), 5 males and 3 females. SIN- GAPORE (Baker), 1 pair. Type locality—LUZON, Benguet, Baguio (Baker). Nesiope heterocephala intermedia var. nov. Similar to the species, except body of both sexes distinctly smaller and forewings with apical third or half dotted with numerous brown spots and several elongate, black spots. In some individuals of the species there is a slight spotting as in the variety, but this is not nearly so conspicuous. Genitalia of both sexes similar to the species but smaller. The maculation in this variety appears to be quite distinct from that of N. ornata, judging from Kirkaldy’s description and figure of the latter, TENIMBER ISLANDS, Larat (Muir), December, 1907, 4 males and 8 females. JAVA, Buitenzorg (Muir), 1 male. Genus HOMOTOMA Guerin Although this genus bears unmistakable resemblance to other genera of Carsidarine, its relationship is not wholly clear yet. 162 Philippine Journal of Science 1919 I attach a good deal of importance, taxonomically, to the clawlike spines on the basal tarsus of the hind legs. In Psylline there are two such “claws,” in Triozinz none, while in Carsidarine there is one “claw.” In Homotoma, however, there are usually two and in Mycopsylla there appear to be two. In some other respects, also, this is somewhat of an aberrant genus, but it is retained in this subfamily, for the present at least. Key to the species. a'. First marginal cell much smaller than second or wanting; radius and media quite or nearly contiguous for a greater or less length; ptero- stigma wanting. b'. Medial vein contiguous with radius for about one-half length of latter, leaving an open cell between them on costal margin. c’. First marginal cell present; a black band following radius to margin and another following cubitus. (Formosa.) H. radiatum Kuwayama (’08: 181). c’. First marginal cell wanting; the two black bands wanting, but radial vein prominently black. (Bengal.) H. distincta Crawford (’12: 433). b*. Medial vein contiguous with radius for nearly or quite the entire length of latter, leaving no cell or only a very small one between them; a black band following radius and media, and a second one following cubitus; first marginal cell present but small. c. Insect large, 6 millimeters long to tip of folded wings; antenn2 terete, without carine; costal margin of forewing roundly arched, wing nearly half as broad as long. (Philippines and Straits Set- eee es H. bakeri Crawford (’15: 263). ce’. Insect smaller, about 3 millimeters long to tip of folded wings; antenne carinate; costal margin of forewing much less arched, wing about one-third as broad as long. (Philippines.) H. bilineata Crawford (’17: 164). a’, First marginal cell nearly as large as second; radius and media not contiguous; pterostigma present; wing veins margined with black; antenne carinate. (Philippines.).... H. pacifica Crawford (’15: 262). Homotoma bakeri Crawford (715: 263). A single male has been taken by C. F. Baker on Penang Island, Straits Settlements. As the original description of this species was drawn up from a female example, the male genitalia are here described. Male.—Forceps as long as anal valve or longer, basal half with a carina on outer face, distal half enlarged, somewhat spearhead-shaped with tip bluntly rounded. Anal valve very slender in distal half, base broader. Hind wings in both sexes very small and almost nerveless. XV,2 Crawford: Jumping Plant Lice 163 Genus TENAPHALARA Kuwayama Tenaphalara KUWAYAMA, Trans. Sapporo Nat. Hist. Soe. 2 (1908) 154. Strogylocephala CRAWFORD, Philip. Journ. Sci. § D 12 (1917) 166. Body elongate, slender; head not much defiexed, nearly or quite as broad as thorax; vertex more or less rounded forward and downward, with anterior ocelius at front end of head and antennez inserted at end on each side of front ocellus; genz not produced into cones, slightly swollen beneath antennal sockets, usually not wholly covering frons, antenne slender; eyes hemi- i: spherical, not recessive. Thorax slender, not arched; legs slender, often rather long. Forewings long and narrow, more or less acutely pointed; vena- tion of Carsidarinze type, with two pseudoveins or calluses; pterostigma present. Abdomen long and slender. Type of the genus, Tenaphalara acutipennis Kuwayama. This is rather an anomalous genus, suggestive of Aphalara and also Pauwropsylla in the head shape, and of Carsidarine in wing venation, tarsi, and eyes. Kuwayama placed the genus at the end of his subfamily Aphalarine, but the majority of the characters point rather to an affinity with Carsidarine. Al- though the head is not cleft in front, as in Carsidara, never- theless this genus is now referred to the subfamily Carsidarine. This is a tropical subfamily and many forms doubtless remain undiscovered. Nesiope appears to be intermediate between Car- sidara with its deeply cleft head and Tenaphalara with its head not at all cleft in front. iw Key to the species. a’. Vertex smooth and uniformly rounded forward and down to gene, with antenne slender and well separated at base. 6°. Forewings with a longitudinal brown vitta; vertex very abruptly descending; insect dark colored, small, less than 3 millimeters @s long to tip of folded wings. (Philippines.) . T. fascipennis Crawford (’17: 167). b*. Forewings clear, without vitta; vertex less abruptly descending; insect greenish yellow, larger, at least 4 millimeters long to tip of folded wings. (Formosa, Philippines, and Bengal.) T. acutipennis Kuwayama (08: 155). a’. Vertex not smooth nor uniformly rounded forward and downward to genz. b*. Vertex with a prominent epiphysis just behind each antennal base; posterior ocelli elevated; antenne approximate at base. c’. Forewings with pterostigma; third antennal segment distinctly thicker than succeeding segments. (Malay Peninsula.) T. malayensis sp. nov. te 164 Philippine Journal of Science 1919 o. Forewings without pterostigma; third antennal segment not thickened. ©“ US69rn66: )20 i ee T. juliana sp. nov. b. Vertex without epiphyses, with longitudinal dark stripes; ocelli not elevated; antennz slender, not approximate at base. : ce’. Pseudo (cross) veins present in forewings; radius and medio- cubitus not forking from basal vein at same point; vertex flattish, with four longitudinal dark stripes. (Malay Peninsula.) T. striata sp. nov. c’. Pseudo (cross) veins wanting: radius, media, and cubitus all fork- ing from basal vein at the same point (as in Triozine) ; vertex more rounded, with three black stripes. (Malay Peninsula.) T. triozipennis sp. nov. Tenaphalara acutipennis Kuwayama. Tenaphalara acutipennis KUWAYAMA, Trans. Sapporo Nat. Hist. Soc. 2 (1908) 155. Tenaphalara elongata CRAWFORD, Rec. Indian Mus. 7 (1912) 482, pl. 34, figs. M, N, P, Q; pl. 35, fig. O. Length of body, 2.5 to 2.9 millimeters; forewing, 2.8 to 3.3. General color greenish yellow; dorsum sometimes with whitish longitudinal streaks; eyes brown; antennz black at tip. Body long and slender. Head not deflexed, not quite as broad as thorax, anterior end broadly rounded both horizontally and dorsoventrally. Vertex quadrate, a very little longer than broad, inclined downward somewhat roundly, without prominent foveal depressions. Genz slightly swollen beneath antennal bases but not meeting over frons; frons visible between genze; anterior ocellus at front end_ of head, at end of frons. Antennze about two and one-half times as long as width of head, slender. Thorax narrow, not arched, pronotum long; dorsulum short. Legs slender, rather long; hind tibize with spur at base and three large, black spines at apex. Forewings long and slender, three times as long as broad, hyaline, transparent, sometimes slightly flavous, acute at apex, with pseudovein between radius and radial sector and another between radial sector and media.. Abdomen long and slender, tapering to narrow apex. Male forceps nearly as long as anal valve, slender, arched, roundly acute at apex; anal valve broadest in basal third, tapering to apex. Female genital segment about one-fourth as long as ab- domen; dorsal valve longer than ventral, somewhat enlarged at tip, ventral valve acute. Luzon, Laguna Province, Los Bafios (Baker), 10 specimens, both sexes, taken on Sterculia foetida. Type locality—ForRMOSA (Matsumura). Several specimens from Bengal were described by me as a new species, T. elongata, very closely similar to the Formosan Cw xv.2 Crawford: Jumping Plant Lice 165 species. Further comparison of these, however, indicates their identity with 7. acutipennis. This indicates a very wide dis- tribution of the species. Tenaphalara fascipennis Crawford. Strogylocephala fascipennis CRAWFORD, Philip. Journ. Sci. § D 12 (1917) 166, pl. 1, fig. 12. The study of several newly discovered species of Malayan psyllids evidently related to Tenaphalara has led me to group together in this genus not only these new species but also the related and more or less intermediate Philippine form for the reception of which a new genus had been erected, Strogyloce- phala. As this was a monotypic genus, this name becomes syno- nymical with Tenaphalara. LuzoN, Laguna Province, Los Bajfios. Tenaphalara malayensis sp. nov. Plate II, fig. 3. Length of body, 3 millimeters; forewing, 3.3. General color flavous to fulvous; thoracic dorsum with slightly darker streaks; vertex brown; antennal segments 3 to 10 black or brown; front and middle tibia brown; eyes black. Head about as broad as thorax. Vertex a little broader than long, quadrate, with posterior ocelli elevated and a prominent, hornlike, flattened epiphysis on each side between antennal base and posterior ocellus. Genz swollen beneath antennal sockets and slightly protruding roundly in front of vertex, covering frons; clypeus long, narrowly rounded at apex; beak long. An- tenn about four times as long as width of head, closely approx- imate at base, segments 1 and 2 short and not very thick, but segment 3 distinctly thicker than succeeding segments. Eyes prominent. Thorax narrow. Legs long and slender; hind tibie with a large spur at base and four black spines at apex. Forewings hyaline, clear, venation as in congeners, but pseudovein between radius and radial sector often wanting. Abdomen long. Male forceps short, slender, subacutely pointed; anal valve shaped somewhat as in Aphalara, with a caudally directed lobe which is nearly as long as basal width. Female genital segment about two-thirds as long as abdomen; valves subequal in length but dorsal blunt at apex and ventral very acutely pointed; dorsal valve with a single horizontal row of twenty or more hairs around end; ovipositor stout, exserted. PENANG (Baker), 2 females. SINGAPORE (Bake7), 1 male and 2 females. 166 Philippine Journal of Science 1919 Tenaphalara juliana sp. nov. Closely related to T. malayensis, differing as follows: Antennze slender throughout, third segment not thicker than succeeding ones; legs slenderer and shorter; forewings without pterostigma and the marginal cells relatively different in size. BORNEO, Sandakan (Baker), 1 example (genitalia broken off). Tenaphalara striata sp. nov. Plate II, fig. 5. Length of body, 2.7 to 3.1 millimeters; forewing, 3 to 3.4. General color orange yellow or fulvous, with narrow black mark- ings on vertex and thoracic dorsum; vertex with five black stripes, one on median line and two on each side; abdomen brownish; femora brown apically; antenne dark at tip of each segment; forewings hyaline, with several small brown fiecks in distal third. Head not quite as broad as thorax, vertex broader than long, rectangular, nearly straight across front margin, more nearly flat than in congeners, with a slight depression on each side of median line; gene bulging roundly forward in front of vertex and beneath antennal sockets, covering frons, with a pair of very small, approximate, acutely pointed, conical processes far back just in front of clypeus. Frons not visible except at front ocellus. Antennse about twice as long as width of head, slender. Beak long. Thorax moderately broad and robust; metascutellum with two small, blunt epiphyses dorsad. Legs stout; hind tibise with a large spur at base and four black spines at apex, one longer than the others and bifid. Forewings long and narrow, hyaline, with a whitish tinge, several black flecks on blisterlike elevations on membrane and a few on veins in distal third; marginal cells subequal. Abdomen moderately long. Male forceps about as long as anal valve, slender, arched, slightly enlarged at tip; anal valve constricted at base and subspherically swollen above the narrow base, outline in lateral view subcircular. Female genital seg- ment long, nearly as long as abdomen; dorsal valve with a large prominent dorsal projection caudad of large anus, rounded broadly at its apex and bearing ‘a single horizontal row of about ten long sete; apex of dorsal valve acute and, with acute point of ventral valve, closely embracing the stout ovipositor. SINGAPORE (Baker), 1 pair. This is obviously close to the typical Tenaphalara but some- what different in the shape of head. XV,2 Crawford: Jumping Plant Lice 167 Teuaphalara triozipennis sp. nov. Plate II, fig. 6. Length of body, 2.1 millimeters; forewing, 2.4. General color reddish brown; thoracic dorsum with several lighter streaks; vertex with four black stripes; venter and legs paler; antenne light brown; biack at tip. Head scarcely deflexed, nearly or quite as broad as thorax; vertex about as long as broad, somewhat quadrate in outline but anterior portion sharply bent downward, median suture. im- pressed and black, and a parallel black sulcus on each side of it; posterior ocelli smali, not elevated; front ocellus not visible from above. Genz not quite covering frons, scarcely swollen beneath antennal bases, cones wanting. Antenne a little longer than width of head, slender. Eyes large. Rostrum long. Thorax scarcely arched, narrow; pronotum moderately long. Legs slender; hind tibiz: with a small spur at base and several short black spines at apex. Forewings narrow, acutely pointed, with several marginal brown spots and apical third maculated irregularly with brown, pterostigma wanting; venation trio-— zine, but resembling that of Tenaphalara also. Abdomen slender, long. Female genital segment long, nearly as long as abdomen, constricted midway; dorsal valve more con- stricted than ventral, with distal half bluntly rounded at apex. SINGAPORE (Baker), 1 female. This is a very interesting species, for it is unmistakably closely allied to Tenaphalara, resembling especially T. sulcata in characters of the head, genitalia, and thorax, but differing in one important venational feature in which the species is dis- tinectly triozine. The absence of the cubital petiole has always been used as the basis for segregating the subfamily Triozine, but there are several species which seem to weaken this view. Pauropsylla triozoptera, several species of Rhinopsylla, and now this new species—all point to a possible fallacy in the current classification. PSY LLIN 42 In the North Temperate Zone the Psyllinz is the most nume- rously represented of the subfamilies, the Triozine ranking a close second. In-the Tropics, however, there are far fewer species of this subfamily, the Pauropsylline and the Carsidarinze being the largest. Australia has a good many species of Psylla, but their habitat is scarcely tropical. Of the ten or more Tem- perate Zone genera only three (Psylla, Arytaina, and Euphalerus) are known thus far to have representatives in the Palzotropics, 168 Philippine Journal of Science 1919 while two others (EH pipsylla and Metapsylla appear to be limited to these Tropics. These five genera are included in the following key, a more nearly complete synopsis of the subfamily being available in my monograph. Subfamily characteristics are not numerous. The gene are always more or less produced into either conical or flattish, spat- ulate processes, vertex not cleft in front, usually more or less flat; antenne always slender, either short or long. Basal tarsus of hind legs always with two clawlike spines at apex; hind tibiz with or without basal spur. Forewings oval or rhomboidal, very seldom acutely pointed at apex, always with cubital petiole (media and cubitus with common stem). The most closely related subfamily appears to be Pauropsyl- linge, and this differs in several characteristics, chiefly the almost humpbacked thorax, the usually absent genal processes, usually visible frons, short metacoxal spurs, and several others of less constancy. Although not strictly a part of the paleotropical fauna, Frog- gatt’s Australian genera of this subfamily are included in the following key: Key to the genera. «’. Genal processes conical, acutely or bluntly pointed; pterostigma of forewing closed (occasionally wanting); antenne at least longer than width of head. b*. Genal cones as long as or longer than vertex, slender and acute, directed forward; antenne very long, usually longer than body, forewing with or without pterostigma............ Epipsylla Kuwayama. b*. Genal cones shorter than vertex, usually directed more or less down- ward. ec’. Pleural suture of prothorax extending to middle of lateral extrem- ity of pronotum; genal cones broad and bluntly rounded, usually extending more or less parallel with plane of vertex; forewings often maculated Arytaina Foerster. ec’. Pleural suture of prothorax extending obliquely to posterior corner of lateral extremity of pronotum; genal cones usually declivous, more or less acutely pointed; forewings usually clear. Psylla Geoffroy. a’. “Genal processes flat, broad, not conical in outline; antenne seldom long, usually very short. b*. Genal processes large, broad, more or less quadrate, scarcely divergent; antenne seldom very short, though usually moderately so. oe. Genal processes in same plane with vertex; flat; eyes more or less recessive and flattened; forewing often rhomboidal, with closed pterostigma or none. : Euphalerus Schwarz. ce’. Genal processes depressed somewhat from plane of vertex; scarcely flat; eyes usually hemispherical; forewings not rhomboidal, with pterostigma................ Arytaina Frst. also Eucalyptolyma Froggatt. b’. Genal processes short and rather broad; antenne very short. XV,2 Crawford: Jumping Plant Lice : 169 c’. Pterostigma wanting; veins not sinuous........ Brachypsylla Froggatt. ce. Pterostigma gf forewing present and closed; veins sinuous; antennz not longer than width of head..._................. Metapsylla Kuwayama. ce. Pterostigma present but open. ad’. Wings membranous; eyes flattened........ Syncarpiolyma Froggait. d@’, Wings coriaceous; eyes reniform.................... Eriopsylla Froggatt. Genus EUPHALERUS Schwarz Head nearly or quite as broad as thorax, strongly deflexed; vertex flat, often rugulose, with posterior ocelli scarcely elevated and anterior ocellus visible from above; genal cones large, flat, = continuing plane of vertex almost without interruption, usually i quadrate; eyes large, somewhat recessive; antenne variable in length. Thorax strongly arched and broad; propleural suture ex- tending to middle of end of pronotum, mesonotum large; legs robust, hind tibie usually with basal spur; forewings large, broad, rounded or somewhat angulate at apex; membrane thick, subhyaline or opaque, maculated or unicolorous; pterostigma present. Type of the genus, Huphalerus nidifex Schwarz. Four species are recorded from North America. Five are now known from the Eastern Hemisphere. Key to the species. a’. Head and thorax with a prominent black stripe from tip of gene to base of forewings and continued along posterior margin of forewing. b*. Forewing with distinct pterostigma; first marginal cell larger than second; genal cones quadrate, not at all divergent; vertex deeply rugose or vermiculate-punctate. (South India and Bengal.) E. vittatus Crawford (’12: 423). b*. Forewing with pterostigma small or wanting; first marginal cell a little smaller than second; genal cones slightly divergent; vertex only slightly rugulose. (Philippines.) E. nigrivittatus Crawford (’13:298). a’, Head and thorax without lateral black stripes; forewings maculated. b*. Length to tip of folded wings not over 3 millimeters; antennz not 2 more than twice as long as width of head; genal cones somewhat divergent, not quadrate. : ce’. Forewing much more than twice as long as broad, narrow in basal half; thorax not mottled with brown; inhabiting Citrus. (Phil- ippines, Formosa, Java, Malay Archipelago, Bengal, southern India, and southern China.)................------- E. citri (Kuwayama). ce. Forewing scarcely twice as long as broad, broad in basal half, thorax mottled with brown. (Fiji.)............ E. maculosus sp. nov. b*. Length to tip of folded wings over 4 millimeters; antennz three times as long as width of head; genal cones quadrate, not divergent; fore- wings slightly rhomboidal. (Malay Archipelago.) E. grandis sp. nov. 170 Philippine Journal of Science 1919 Euphalerus grandis sp. nov. Plate II, fig. 7. Length of body, 3.1 millimeters; forewing, 3.2. General color yellowish or greenish brown, mottled on thoracic dorsum, and often on vertex also, with brown; tarsi brown; antennz black at tip of each segment and apical two segments brown; fore- wings mottled and banded with brown. Head not as broad as thorax, strongly defiexed. Vertex quite flat, rugulose, with a very shallow concavity on each side of median line, a small tubercle on each outer angle behind an- tennal base, a little more than half as long as broad, anterior ocellus on top. Genal cones large, broad, flat, quadrate, two- thirds as long as vertex, on same plane with vertex, nearly con- tiguous along inner margin, with thick, white pubescence distad. Eyes large, black, somewhat recessive. Antenne more than three times as long as width of head, very slender. Thorax very broad, robust, strongly arched. Pronotum long, mesonotum very large. Legs thick; hind tibize with a very small spur at base and several black spines at apex. Forewings large, broad, subrhomboidal, veins thick, alternately brown and fla- vous, membrane thick, flavous, subhyaline, maculated with nu- merous small brown spots which are aggregated along apical margin into a broad band; pterostigma very large and broad. Hind wings smaller, thin, hyaline, delicate. Abdomen rather short, thick, humped upward. Male forceps about two-thirds as long as anal valve, relatively thick, bluntly rounded at apex, sides subparallel; anal valve simple, broadest at base and narrowing slightly toward apex. Female genital segment very short, dorsal valve a little longer than ventral, apex subacute and sometimes flexed upward, with long hairs near middle and short bristles at apex. Mouuccas, Toeal (Muir), 3 females. TENIMBER ISLANDS, Larat (Muir), December, 1907, 3 females and 1 male. Euphalerus maculosus sp. nov. Plate II, fig. 8. Length of body, 1.9 millimeters; forewing, 2. General color mottled brown and greenish or dirty white; femora brown, tib- iz pale; antenne brown at apex of each segment; forewings conspicuously mottled and maculated with brown. Head small, deflexed, about as broad as prothorax, much nar- rower than mesothorax across base of forewings. Vertex more than half as long as broad, flat, rugulose, anterior ocellus above; genal cones short, less than half as long as vertex, flat, in same XV,2 Crawford: Jumping Plant Lice 171 plane with vertex, divergent, subacutely pointed. Antenne less than twice as long as width of head, slender. Thorax broad, arched, legs rather short; hind tibize without basal spur, and apical spines short. Forewings subcoriaceous, not transparent, somewhat rhomboidal, veins thick, pterostigma incomplete, membrane and veins spotted and ‘mottled. Hind wings clear, delicate. Abdomen short. Male forceps about two-thirds as long as anal valve, narrow, subacute at apex, sides parallel; anal valve longer, slender, broadest at base, upper third narrowed. Figi ISLANDS, Rewa (Muir), March, 1906, 3 males. Euphalerus vittatus Crawford (’12: 423). One slightly damaged specimen (female) probably of this species is at hand, collected at Madhavgiri, Bombay, India (H. H, Mann), May, 1916, on Cassia fistula. The lateral vitta of thorax is wanting and the wing is more extensively mottled with brown. Structurally this specimen resembles the species, but superficially there are some differences. Additional mate- rial will show whether or not there are two species in India on the same food plant. Euphalerus nigrivittatus Crawford (’13: 298). This species described from one female. Subsequently three additional examples, 1 male and 2 females, of the species have been collected by C. F. Baker at Davao, Mindanao. The descrip- tion of the male genitalia is now added to the original description. Male.—Forceps moderately broad at base, narrowing to mid- dle, then abruptly broadened and deeply emarginate at apex, terminating in two points; anal valve longer than forceps, apical half much narrower than base. The presence of a distinct basal spur on each hind tibia was not mentioned in the first description and is now added. Distribution —LuzoN, Laguna Province, Los Bafios (Baker). “MINDANAO, Davao (Baker). Euphalerus citri (Kuwayama). Diaphorina citri KuwAYAMA, Trans. Sapporo Nat. Hist. Soc. 2 (1908) 160. Euphalerus citri CRAWFORD, Rec. Indian Mus. 7 (1912) 424, pl. 33, figs. N, O, P; pl. 35, fig. D; Philip. Journ. Sci. § D 8 (1913) 299; 12 (1917) 168. In Mr. Muir’s collection there are six specimens of this pretty citrus psyllid, taken at Macao, near Hongkong, China, Novem- ber, 1906; also, one from Amboina, Moluccas. No food plant 172 Philippine Journal of Science 1919 data accompany these, but elsewhere the species inhabits foliage of Citrus. The species appears to be very widely distributed throughout the southern portion of the Eastern Hemisphere. Genus ARYTAINA Foerster Psyllopa CRAWFORD. Head nearly or quite as broad as thorax; vertex usually nearly or quite flat, with anterior ocellus visible from above; gene produced into conical or subquadrate processes extending for- ward nearly in same plane with vertex but separated from ver- tex by impressed line. Eyes large, more or less recessive over propleurites. Antenne slender, moderately long. Thorax usually strongly arched and broad; propleural suture extending to middle of more or less swollen end of pronotum; hind tibie with or without basal spur. Forewings usually broadly rounded, hyaline or colored or rarely subcoriaceous, pterostigma present. Hind wings smaller, hyaline and clear. Key to the species. a’. Veins of forewing alternately black and yellow or white. b*. Membrane as well as veins spotted with black or brown; antenn# less than twice as long as width of head; female genital segment not flexed upward at apex. c’. Antennz about as long as width of head; forewing with a broad, irregular band of brown at apex, and remainder irregularly spotted. (Malay Archipelago.) -.............-..0....... A, pulchra sp. nov. ec’. Antenne nearly twice as long as width of head; forewing without apical band, but more uniformly spotted with small maculz. (Bengal and Ceylon.) ............ A. punctipennis Crawford (’12: 431). b*. Membrane clear, only veins spotted with black; female” genital seg- ment flexed upward at apex. (Singapore.) A. punctinervis sp. nov. a. Veins of forewings not conspicuously spotted as above. Antenn#® at least twice as long as width of head. b*. Forewings hyaline, clear; marginal cells subequal. ce’, Genal cones subacute; first marginal cell of forewing not quadrate; pterostigma not short and broad. (Bengal.) - A. obscura Crawford (712: 432). c*, Genal cones very broadly rounded at apex; first marginal cell some- what quadrate; pterostigma short and broad. (Philippines.) A. iolani sp. nov. b*. Forewings colored or maculated, not clear; vertex with a tubercle at each outer angle behind antennal base. ec’. Genal cones broad, broadly rounded or quadrate; forewings not much over twice as long as broad. ad. Forewings unicolorous, without marginal bands. e*. Forewings flavous; insect about 3 millimeters long to tip of folded wings. (Singapore.) -..000000000 ul A, flava sp. nov. XV,2 Crawford: Jumping Plant Lice 173 e*. Forewings brown; tubercles on vertex large; insect 4 millimeters long to tip of folded wings. (Philippines.) A, tuberculata Crawford (’17: 169). @. Forewings not unicolorous, margins darker, banded. e’. Forewings with an apical band of solid color or made up of small spots, and costal margin often darker; male forceps spatulate; very broad at apex. (Philippines and Borneo.) A, variabilis Crawford (’17: 168). e*. Forewings with an apical and costal band of many small brown spots; male forceps not spatulate, with a prominent con- vexity on anterior margin (in lateral view). (Malay Archi- DOING. iiss A. meridionalis sp. nov. c. Genal cones acutely pointed; short, strongly divergent; forewings about three times as long as broad, mostly brown, with five small clear spots on apical margin. (Singapore.) A. brevigena sp. nov. Arytaina meridionalis sp. nov. Plate II, fig. 9. Length of body, 2 millimeters; length of forewing, 2.1. Gen- eral color pale straw yellow to very light brown; abdomen with brown markings; antennz brown on apical third; wings brown except in center pale. Body surface finely pubescent. Head as broad as thorax, deflexed, nearly transversely straight on hind margin. Vertex flat, large, a little more than half as long as broad, with a shallow depression on each side of median line, a small tubercle on each outer angle just behind antennal base, pubescent; genal cones broad and large, broadly rounded at apex, about half as long as vertex, a little broader than long, not contiguous on inner margin, only slightly depressed from plane of vertex. Eyes large; anterior ocellus on top. Antenne very long and slender, about two and one-half times as long as width of head. Thorax broad, well arched. Legs short and thick; hind tibie with a black spur at base, and five black spines at apex. Fore- wings brown on anterior and apical margins and more or less spotted and maculated on rest of membrane, subcoriaceous, more hyaline in central portion, veins with double row of fine sete; pterostigma pubescent. Male forceps nearly as long as anal valve, bluntly rounded and somewhat toothed at apex, straight or slightly concave on posterior margin (profile) and with a prominent convexity on anterior margin midway between base and apex; anal valve simple, converging a little toward apex. Female genital segment about half as long as abdomen, both valves acute, subequal in length. TENIMBER ISLANDS, Larat (Muir), December, 1907, 1 pair. 1664704 174 Philippine Journal of Science 1919 This appears to be closely related to the Philippine species, A. variabilis. In wing colors and venation some of the variable Philippine specimens closely resemble the southern form, but in male genitalia they are distinct though suggestively similar, also. Arytaina flava sp. nov. Plate I, fig. 10. Length of body, 2.2 millimeters; length of forewing, 2.3. Gen- eral color flavous yellow, with several duil-colored streaks on thoracic dorsum; eyes black; antennal segments black at tip of each. Body surface sparsely pubescent. Head not quite as broad as thorax; vertex flattish, with a small foveal impression on each side of median line; with a small tubercle on each outer angle behind antennal base, front ocellus above. Genal cones a little broader than long, about half as long as vertex, broad at apex, scarcely divergent. An- tenn about twice as long as width of head or less, very slender. Thorax broad, well arched. Legs short, thick; hind tibize with spur at base and five or six black spines at apex. Fore- wings flavous, broadly rounded, veins setigerous, venation some- what similar to that of A. meridionalis. Male genital segment short; forceps about two-thirds as long as anal valve; moderately stout, not tapering much toward apex, bluntly rounded at tip; anal valve inclined caudad in apical third, a little broader at base than apex, narrow apically. Female genital segment half as long as abdomen or more, both valves acute or subacute, subequal in length. SINGAPORE (Baker), 2 females. BORNEO (Baker), 1 male and 1 female. The latter differ slightly from the Singapore speci- mens in having shorter antennz, and the second marginal cell of forewing not quadrate. All appear, however, to be repre- sentatives of the same species. Arytaina iolani sp. nov. Plate II, fig. 11. Length of body, 2 to 2.3 millimeters; forewing, 2 to 2.3. Gen- eral color flavous to orange red, eyes black; antenne black at tip of each segment; forewings slightly flavous. Body surface glabrous. Head nearly as broad as thorax, deflexed. Vertex fiat, about half as long as broad, with a very shallow depression on each side of median line, with a small tubercle on each outer angle behind antennal base. Genal cones as broad as long, broadly rounded at apex, somewhat divergent. Antenne a little less than twice as long as width of head. Thorax arched, broad. Legs moderately long and slender; XV,2 Crawford: Jumping Plant Lice 175 hind tibiz with spur at base. Forewings hyaline, veins yellow, - with three small brown spots on apical and posterior margins; pterostigma short. Male forceps a little more than half as long as anal valve, shaped irregularly somewhat as in A. meridionalis; anal valve plain, narrowing toward apex. Female genital segment as long as abdomen, slender; dorsal valve a little longer than ventral, both acute. Luzon, Laguna Province, Los Bafios (Baker), 1 pair. The male genitalia, shape of head, and venation of forewing point to a close relationship between this species and A. meri- dionalis, of the Malay Archipelago; and it appears, also, that A. flava, of Singapore, is related to both. Arytaina brevigena sp. nov. Plate III, fig. 4. Length of body, 2.2 millimeters; forewing, 2.3. General color chocolate brown with fine reddish spots and markings on thorax and larger reddish areas on vertex; legs variegated chocolate brown and light brown; antenne black on apical half of each segment, rest flavous; forewings brown, lighter in central por- tion; with five round clear spots on apical margin. Head nearly as broad as thorax, not much deflexed. Vertex about half as long as broad, with a foveal depression on each side of median line and a blunt, thick tubercle on each outer angle behind antennal base. Genal cones short and strongly divergent, acutely pointed, extending only a very little beyond antennal socket; antenne three times as long as width of head, slender. Thorax not strongly arched, narrower than in many congeners, pubescent with very short, fine hairs. Femora short and thick; hind tibize longer, slender, with a small spur at base. Fore- wings uniformly broadening from base to apex, latter rounded, margins nearly straight, veins setigerous, dark; membrane black or brown except in central portion of wing and five small spots on apical margin. Female genital segment about two-thirds as long as abdomen; acutely pointed, dorsal valve longer than ventral, with a prom- inent constriction on dorsal valve beyond end of anal opening, terminal half acuminate and slender. SINGAPORE (Baker), 4 females (2 in poor state of preserva- tion). Arytaina pulchra sp. nov. Plate III, fig. 3. Length of body, 2.2 millimeters; forewing, 2.3. General color brown; vertex dirty brown, thorax reddish or chestnut brown; 176 Philippine Journal of Science 1919 antennz black at tip of each segment; femora dark, tibie light brown; forewings maculated with brown on veins and membrane, and apical portion solid brown, membrane whitish between macule. Head not as broad as thorax, well defiexed; vertex nearly flat, slightly convex at median line in anterior portion, a little more than half as long as broad, slightly rugulose; genal cones nearly in same plane with vertex, quadrate, broad on apical margin, nearly contiguous, a little less than half as long as vertex, slightly pubescent. Antenne scarcely longer than width of head. Thorax broad and strongly arched, pronotum moderately long. Legs thick; hind tibize without:basal spur. Forewings subhya- line, slightly rhomboidal, veins thick and alternately brown and whitish; membrane whitish between brown maculz which are scattered over membrane and at apex are concentrated into a broad, brown band; pterostigma short and broad. Female genital segment very short, dorsal valve a little longer than ventral, both acute. TENIMBER ISLANDS, Larat (Muir), 1 female. Arytaina punctinervis sp. nov. Plate ITI, fig. 1. Length of body, 1.5 millimeters; length of forewing, 1.5. Gen- eral color black variegated with orange red; dorsum of head and thorax mostly orange red with many black markings and figures; abdomen and venter mostly black with red markings; genal cones red above and below, and black laterad; antenne with alternating black and orange bands; forewings hyaline, veins and margins alternately black and yellow; legs orange, with black dots and bands. Head not as broad as thorax. Vertex flat or slightly convex with a narrow, black, foveal depression on each side of median suture, a little more than half as long as broad. Genal cones about two-thirds as long as vertex, subacute, somewhat diver- gent, nearly in same plane with vertex. (Antenne mutilated.) Thorax well arched, robust. Legs short, small; hind tibie without basal spur. Forewings broadly rounded, hyaline, con- spicuously marked somewhat as in A. punctipennis, but mem- brane between veins not maculated as in that species. Female genital segment about half as long as abdomen, dorsal valve longer than ventral, flexed upward and very acutely pointed at apex; ventral valve acute. SINGAPORE (Baker), 1 female. This species is apparently closely related to A. punctipennis, ee ~ te XV, 2 Crawford: Jumping Plant Lice WT a species found in Bengal and Ceylon. The differences between the two species are sharp but do not conceal their close relation- ship. Arytaina punctipennis Crawford. Psyliopa punctipennis CRAWFORD, Rec. Indian Mus. 7 (1912) 431, pl. 34, figs. K, O; pl. 35, fig. U. Arytaina punctipennis CRAWFORD, Philip. Journ. Sci. § D 12 (1917) 170. Distribution.—BENGAL, Pusa. CEYLON, Peradeniya. Food plant.—Indigo (Indigofera). Arytaina obscura Crawford. Psyllopa obscura CRAWFORD, Rec. Indian Mus. 7 (1912) 482, pl. 34, fig. S; pl. 35, fig. M. Distribution. BENGAL, Pusa. Food plant.—Mango (Mangifera). Arytaina variabilis Crawford (‘17: 168). Distribution.— MINDANAO, Butuan, Davao. LUZON, Tayabas Province, Malinao: Laguna Province, Mount Banahao. BoRNEO, Sandakan (Baker), 1 male. Arytaina tuberculata Crawford (’17: 169). Distribution.—MINDANAO, Davao. Genus EPIPSYLLA Kuwayama Head long, not strongly deflexed, vertex rather flat, genal cones very long, longer than vertex, conical and more or less divergent, porrect; eyes large; antennz long, slender. Thorax not strongly arched; pronotum long. Forewings hyaline, rounded at apex, pterostigma distinct or nearly wanting. The general aspect of the body is similar to that of Psylla, but the very long genal cones easily distinguish this genus. Type of the genus, Epipsylla albolineata Kuwayama. Key to the species. a’. Head and thorax with a pair of conspicuous white or yellow, black- margined stripes from tip of genal cones to base of abdomen, where they meet. b‘. Head with a second pair of white, black-margined stripes, one from each antennal socket to posterior ocellus; pterostigma of forewing narrow; male forceps with a long tooth on inner face; genal cones about as long as vertex. (Philippines and Malay Archipelago.) E, pulehra Crawford (’13: 297). 6*. Head without second pair of white stripes; pterostigma wanting; male forceps without inner tooth; genal cones longer than vertex. (For- mosa.) ... __.. E, albolineata Kuwayama (’08: 178). 178 Philippine Journal of Science . 1919 a’, Head and thorax without white stripes. hb’. Pterostigma of forewing wanting; antennz one and ‘one-half times as long as body (without wings); forceps distinctly shorter than anal valve; much of thoracic dorsum red. (Formosa.) E. rubrofasciata Kuwayama (’08: 179). b*. Pterostigma large; antenne about as long as body (without wings) ; forceps longer than anal valve; thoracic dorsum orange yellow. (Philippines and Malay Archipelago.) E. forcipata Crawford (717: 167). Epipsylla pulchra Crawford (713: 297). This species has been taken by C. F. Baker at three stations in the Philippine group; namely, Los Bafos and Baguio, Luzon; and Davao, Mindanao. Frederick Muir has taken one male on Amboina, in the Moluccas. This specimen has all the charac- teristics of the Philippine specimens except that the forceps of the male genitalia have a much smaller tooth on the inner face. Epipsylla forcipata Crawford (’17: 167). The type locality of this species is Puerto Princesa, Palawan, Philippine Islands. The distribution seems to be wider, how- ever, for a single female has been taken by C. F. Baker on Penang Island. Genus PSYLLA Geoffroy This genus is a large one and is widely distributed. Its tax- onomic definition is not wholly clear, as it embraces at present species which probably should properly not be included. Several Australian species, for example, and a few Asiatic species are probably generically distinct but at present are not separated. A full description and a discussion of the genus are given in my monograph of this family. In the following key an attempt has been made to include all described species of the South Pacific, Australia, southern Asia and its islands, including Formosa, and the Philippine and Malay Archipelagoes. Many of these species are known to me only by their descriptions, so that the key must be more limited in value than it might be otherwise. Key to the species. a’, Forewings hyaline, clear, not colored, veins usually brown or black. b*. Body bright red; antenne about two and one-half times as long as width of head. c’. Genal cones longer than vertex; male forceps spatulate, broad at apex; antennz and legs reddish. (Philippines.) P. colorada Crawford (’17: 170). c. Genal cones shorter than vertex; forceps acutely pointed; antennz and legs whitish. (Formosa.).... P. coccinea Kuwayama (’08: 171). Fi ee oe XV, 2 Crawford: Jumping Plant Lice 179 b*. Body not bright red. ec’. Genal cones longer than vertex. d@’, Antennz three-fifths as long as costa; forewings always clear; body yellowish red, with notal streaks. (Formosa.) P. arisana Kuwayama (’08: 168). @, Antenne one-half as long as costa; forewings sometimes slightly yellowish; body yellowish brown with notal streaks. (For- peste] ta) Cec ea pe sehen pe aeaieeee ey P. kiushuensis Kuwayama (’08: 174). ce. Genal cones not longer than vertex. d’. Antenne about twice as long as width of head. e’, Genal cones subacutely pointed, distinctly conical in shape. f*. Body brown; forewings with a black spot at tip of claws. (THORS ee ES P. simle Crawford (’12: 426). f’. Body orange yellow, with conspicuous black and white mark- Me ings on thoracic dorsum and vertex; wing clear, without black spot. (Philippines.) ........-..0002....... P. bakeri sp. nov. e*. Genal cones broadly rounded; forewings without black spot. ( Australia.) f*. Body pubescent, light brown; genal cones small; male forceps fingerlike; on Acacia pendula. (Australia.) P. acacie-pendule Froggatt (’01: 247). f’. Body not pubescent, gray, genal cones large; on Acacia pendula. (Australia.).... P. gracilis Froggatt (’03: 327). da’. Antennz four or five times as long as width of head; forewing without pterostigma. e’. Genal cones well separated and divergent, about as long as vertex and of similar color; costa of forewing not hairy. 2 Bee OS Sots ae aaa oe a a P. leprosa sp. nov. e?. Genal cones not divergent, contiguous distad, not as long as vertex, white, costa of forewing hairy. (Fiji Islands.) P. compta sp. nov. a. Forewings not clear, more or less flavous, fumate or colored. b*. Forewings not opaque, more or less transparent. c’. Forewings not mottled nor maculated. d’, Forewings about three times as long as broad. e’. Antenne about twice as long as width of head or more; body reddish brown. f'. Thorax mottled with black; legs long; genal cones not very small; body about 1.6 millimeters long. On Acacia pycnantha. (Australia.) P. acacie-pyenanthe Froggatt (’01: 243). f. Thorax not mottled; legs short; genal cones very short, not visible from above; body about 1 millimeter long. On Aca- cia decurrens. (Australia.) P. candida Froggatt (’01: 252). e?, Antenne searcely longer than width of head; body ocherous; male forceps slender. On Sterculia. ( Australia.) P. sterculie Fraggatt (01: 255). @. Forewings about two and one-half times as long as broad or less, body color dull yellowish or brown. 180 Philippine Journal of Science 1919 e’. Insect small, body 0.9 millimeter long; genal cones short, quadrate; male forceps thick, short. On Acacia dealbata. (Australia. soos. P. acacix-dealbate Froggatt (’03: 326). e?. Insect larger, body nearly or quite 2 millimeters long. jf’. Genal cones nearly as long as vertex, subacute; body color dull yellow. (Formosa.) P. toroensis Kuwayama (’08: 172). f. Genal cones scarcely half as long as vertex, rounded at apex; body color brown with white markings. (Malay Archipelago.) : P, muiri sp. nov. @. Forewings mottled, maculated or banded. d'. Forewings about twice as long as broad, covered with numerous small dark spots; body color dull yellowish; antennz rather short. e*. Genal cones about as long as vertex, truncate at apex; wings. yellowish, body not pubescent. (Formosa.) P. tripunctata Kuwayama (’08: 174). e’. Genal cones very short, broad, rounded; wings gray, semi- opaque; dorsum of body gray, pubescent. On Acacia bailey- ana. (Australia.) P. acacie-baileyane Froggatt (’01: 257). d@*. Forewings not covered with numerous small spots, but irregularly clouded or banded. e’, Thorax and head yellowish; wings about three times as long as broad, semiopagque, mottled with brown; antennz long, genal cones rather large, broad. On Acacia sp. (Australia.) P. frenchi Froggatt (’01: 245). e*, Thorax and head brown or black. f’. Forewings subhyaline, more or less fumate or dark in apical portion; body moderately large. g’. Antenne as long as forewings or longer; forewings irreg- ularly fumate throughout, darker apically; insect over 4 millimeters long. h*. Genal cones as long as vertex; female genital segment as long as abdomen. (Philippines.) P. crenata Crawford (’17: 171). h?. Genal cones scarcely two-thirds as long as vertex; female genital segment one-third as long as abdomen. (Bor- neo.) P. fumosa sp. nov. g. Antenne about half as long as forewings; wings glassy, with . a brown band from tip of clavus to tip of radius; insect about 3 millimeters long. (Formosa.) P. spadica Kuwayama (’08: 165). f. Forewings semiopaque, broadly and irregularly clouded with brown. g’. Thorax brown with yellowish patches and markings. On Acacia implexa. (Victoria.) P. lidgetti Maskell. Froggatt (’01: 245). g*. Thorax black or dark brown; genal cones very small; antenne short; wings about twice as long as broad, with a large whitish area near middle and rest brown or mottled. On Acacia juniperina. (Australia.) P. acacie-juniperine Froggatt (’03: 328). et ee XV,2 Crawford: Jumping Plant Lice 181 b*. Forewings opaque, mottled with brown; genal cones acutely conical; wings about twice as long as broad. c’. Thorax ocherous, mottled with brown; forceps long and slender. d’. Genal cones very long, acutely pointed; antenne very long; legs long; wings twice as long as broad. On Acacia decurrentis. CAUSUPalie) P. acacie-decurrentis Froggatt (’01: 248). d*. Genal cones short, broad at base, acutely pointed; antenne moderately short; legs short; wings less than twice as long as broad. On Capparis mitchelli. (Australia.) P. capparis Froggatt (01: 250). e¢. Thorax dark brown, motiled with dull yellow; forceps short; genal cones large, moderately long, pointed; antennz very short; legs short. On Apophyllum anomalum. (Australia.) P. schizoneuroides Froggatt (’01: 253). Psylla fumosa sp. nov. Length of body, 2.8 millimeters; forewing, 3. General color brown, with small lighter areas about base of wings and on pro- notum; antennz brown except basal one-fourth yellowish; fore- wings smoky, a little more densely browned irregularly in apical portion. Body surface pubescent with short, fine hairs. Structurally similar to P. crenata Crawford (’17: 171), to which species this is apparently closely related, differing as fol- lows: Genal cones about two-thirds as long as vertex, bluntly rounded at apex, somewhat divergent; antenne a little longer than forewings but not quite as long as body and wings com- bined; thorax more slender; wing venation similar, except cu- bital petiole (M+Cu) only half as long as stem of cubitus (Cu), while in P. crenata it is nearly three-fourths as long; forewing narrower relatively. Male forceps stout, as long as anal valve, somewhat falcate, broadest midway and arched forward, subacutely pointed; anal valve moderately long, tapering from base to apex, apical third slender and inclined backward. Female genital segment short, about one-third as long as abdomen. BORNEO, Sandakan (Baker), 1 male and 1 female. This is a close relative of Psylla crenata but is surely distinct. It approaches to a confusing degree certain species of Arytaina. Psylla leprosa sp. nov. Plate III, fig. 8. Length of body, 2 millimeters; forewing, 2.9; antenne, 4.2. General color lemon yellow with leprous-white areas; vertex in center of pronotum, posterior edge of dorsulum, paraptera, and metascutum white; legs pale; wings clear or slightly yellow, antennal segments 1 and 2 pale, 3 black at extreme tip, 4 black on distal third, 5 black on distal half, 6 black on distal two-thirds, 7 pale at base, 8 to 10 black. ©. 182 Philippine Journal of Science 1919 Head broad and short, about as broad as thorax, deflexed; vertex scarcely half as long as broad between eyes, roundly con- vex on each side of median line in front, posterior ocelli well elevated; genal cones as long as vertex, subacute, divergent, scarcely contiguous even at base, sparsely hirsute; eyes large; antennz very long and slender, longer than body to tip of folded wings, more than four times as long as width of head. Thorax broad, arched, dorsal surface sparsely hairy. Legs rather long and slender; hind tibiz without spur at base, and apical spines short. Forewings ‘clear or slightly flavous, veins yellow, narrow; pterostigma wanting or very narrow. Abdomen short. Female genital segment very short, about one-third as long as abdomen, dorsal valve a little longer than ventral, both subacute. BASILAN (Baker), 1 female. Psylla bakeri sp. nov. Plate III, fig. 9. Length of body, 1.9 millimeters; forewing, 2.3. General color orange yellow or paler, with conspicuous white and black mark- ings as follows: Pronotum black cephalad and white caudad; dorsulum black on anterior margin, white on posterior margin and center orange colored; mesoscutum black on part of anterior margin; mesoscutellum and metascutum white; vertex with a small black or brown discal area on each side of median suture, and a white area caudad of these and extending to median suture; eyes dark; antenne orange with apex of each segment brown; wings clear. Head strongly deflexed, about as broad as thorax, short; vertex not half as long as broad between eyes; roundly convex on each side of median suture in front; with a small, shallow, foveal depression on each side, elevated at posterior ocelli. Genal cones as long as vertex, well separated, and inner margins nearly parallel to near base, subacute. Antenne a little more than twice as long as width of head, slender. Thorax broad, arched, not pubescent. .Legs rather short, but hind legs long; hind tibiz without basal spur and distal spines short. Forewings clear or very slightly fumate, veins yellowish, pterostigma large. Abdomen short. Female genital segment about as long as abdomen, thick at base and narrowing abruptly to acute tip, dorsal valve longer than ventral, both acutely pointed. Luzon, Laguna Province, Los Bafios (Baker), 1 female: Ben- guet Subprovince, Baguio (Baker), 1 female. ‘| XV, 2 Crawford: Jumping Plant Lice 183 Psylla compta sp. nov. Plate III, fig. 11. Length of body, 1.9 millimeters; forewing, 2.3; antennez, 3.5. General color orange; genal cones entirely white, pronotum pos- teriorly margined broadly with white, dorsulum on posterior margin narrowly margined with white and two white areas on mesoscutellum; venter of thorax and dorsum of abdomen dark brown; abdomen ventrad pale whitish; legs flavous, except front tarsi brown; antennz brown except three basal segments orange; forewings clear. Head strongly deflexed, nearly or quite as broad as thorax; vertex half as long as broad between eyes, elevated at posterior ocelli, with a very small foveal depression on each side of median suture, sparsely hairy, median suture scarcely impressed and narrowly black. Genal cones slender, conical, not quite as long as vertex, acutely pointed in apical half or third, closely appressed to each other, very briefly pubescent. Antennz very long, five times as long as width of head, segments 3 and 4 slightly thicker than in most species. Thorax slightly pubescent; legs long, slender, hind tibiz with basal spur, distal spines numerous. Forewings clear, veins yellow or brownish, costal vein conspicuously hairy; pterostigma narrow or wanting. Male genital segment moderately small, forceps small, about half as long as anal valve, slender, acutely pointed, simple; anal valve rather long, broadest at base and converging slightly to- ward subacute apex. Fiji ISLANDS (Muir), 1905, 1 male and 1 imperfect example. The most characteristic feature of this pretty species is the form of the genal cones which is suggestive of the genus Neotriozella. Psylla muiri sp. nov. Plate III, fig. 5. Length of body, 2 millimeters; forewing, 2.1. General color chocolate brown with irregular and narrow white markings and stripes on thorax, vertex, and abdomen; legs light brown or femora darker; antennz pale with apex of each segment darker ; forewings fumate or brown, with a dark spot at tip of clavus and three others on posterior margin. Head deflexed, not quite as broad as thorax, vertex roughly triangular in outline, half as long as broad, with a smal] pro- tuberance behind eyes on outer margin, a small foveal depression on each side of median suture and the two connected by a shallow transverse fossa. Genal cones scarcely half as long as vertex, narrowly rounded at apex, declivous from vertex, divergent; 184 Philippine Journal of Science 1919 antennze a little more than twice as long as width of head, slender. Eyes larger, somewhat recessive. Thorax arched, broad, robust. Legs rather short, stout; hind tibiee with a small spur at base. Forewings short, browned but hyaline, veins with short setz; pterostigma short and broad. Abdomen thick. Male forceps not quite as long as anal valve, broadly spatulate in apical half, broad at apex and part of apical margin finely toothed; anal valve simple, broadest at base, nar- rowing to subacute apex. Female genital segment nearly as long as abdomen, dorsal valve longer than ventral and both acutely pointed. TENIMBER ISLANDS, Larat, December, 1907 (Muir), 10 females. Mo.Luccas, Amboina (Muir), 2 males and 1 female, This species approaches Aryiaina in some respects, as in the character of genal cones and shape of forewing. TRIOZIN 4= The most easily visible, though not the most constant, char- acter of this subfamily is the point of furcation of the basal vein of the forewing—the cubitus, media, and radius leaving the basal vein at quite or nearly the same point. Ceropsylla and Hemitrioza, both American genera, are exceptions to this. On the other hand, several species belonging to other subfamilies, possess this characteristic, as follows: Rhinopsylla, and Tena- phalara triozipennis, of the subfamily Carsidarine; and Pawur- opsylla triozopiera and Leptynoptera, of the subfamily Paur- opsylline. However, the exceptions are comparatively few, and this character remains the best for subfamily distinction. The body is typically slender, eyes hemispheric, frons covered by the gene (except in Cerotrioza bivittata); basal tarsus of hind legs always lacks the clawlike spines which are present in most of the other subfamilies. A full description of the sub- family Triozine is given in my monograph, page 64. Key to the genera. a‘, Hind tibia without basal spur and without subapical spiniform tooth; no anterior processes of metacoxe. b*. Forewings with three marginal spots, made up of numerous minute punctations, on hind margin. c’. Genal cones present, but not always conical. d’. Genal cones more or less divergent, not closely appressed to each other. e’. Notum scarcely arched; pronotum relatively long; genal cones dieieted TorWard 3. Leuronota Crawford. declivous. XV,2 Crawford: Jumping Plant Lice 185 jf’. Usually not gall makers, though sometimes leaf curlers. A large varied assemblage of species................ Trioza Foerster. f’. Gall makers. A small and poorly defined assemblage of mostly Southern Hemisphere species................ Cecidotrioza Kieffer. d’. Genal cones not at all divergent, long and closely appressed to One CUNT Sa a gee Neotriozella Crawford. c. Genal cones wanting; gene sometimes swollen beneath antennal sockets; notum usually not much arched...... Kuwayama Crawford. b*. Forewings without marginal spots on hind margin; veins usually WOOT UR ore ne Se ee Hevaheva Kirkaldy. a’. Hind tibize with a basal spur (sometimes small). b*. Antenne slender, not densely hairy. c’. Forewings not opaque nor maculated, but transparent, with radius usually short; vertex not produced cephalad into horns; meta- coxe with a pair of anteriorly directed processes; hind tibiz usually with a conspicuous spiniform tooth near apex; genal cones present or wanting, seldom conical.... Megatrioza Crawford. c’. Forewings semiopaque or opaque, maculated, with radius extending nearly to tip; vertex produced cephalad into a pair of horns. Cerotrioza g. nov. b*, Antennz thick and very densely hairy; genal cones present; ver- tex very broad; forewings transparent, radius long. Stenopsylla Kuwayama. Trichochermes Kirkaldy (Kuwayama ’10: 54), as represented by two Japanese species, is not recognized in this work as dis- tinct from Trioza. It was separated chiefly on the pubescent or hirsute dorsum, but this makes a very unnatural and wholly un- satisfactory division. At least one species of Trioza, named by Kuwayama himself, has the pubescent dorsum characteristic of the other group but was placed by him in Trioza rather than in Trichochermes. Trichochermes bicolor Kuwayama appears to be very close to my Trioza divisa. Both of Kuwayama’s species of Trichochermes are included in the synoptic key to the species of Trioza. Epitrioza Kuwayama (’10: 55), as originally characterized, appears to be not a well-defined genus. Its chief distinguishing character was said to be the position of the forewing apex in relation to the apical veins, but this characteristic is so variable in the genus Trioza that confusion and quite unnatural segrega- tion of species would surely result if this criterion were adhered to closely. Kuwayama’s characterization of both the genus and its one species is rather incomplete, but certain features suggest a rather close relationship of this to my Megatrioza. However, until an opportunity is had to examine material of this Japanese species, I do not wish to make any disposition of it nor to merge Megatrioza into it. 186 Philippine Journal of Science 1919 Genus TRIOZA Foerster Trioza ig a very large genus of which many of the species are difficult to distinguish. It appears to be a group in which very active evolution of new species is in progress, especially in the North Temperate Zone. In the Tropics and the South Tempe- rate Zone it is much less numerousiy represented, most of the triozine species there being in other genera, chiefly Megatrioza with its armed hind tibie. With Cecidotrioza Kieffer I am un- familiar. Whether or not it is truly distinct from Jrioza or Megatrioza is not yet clear. Key to the species. a’. Hind wings normally developed, at least half as long as forewings. b*, Forewings more or less colored, not wholly clear. ce’. Dorsum and vertex hairy; forewings thickened and darkened on basal third, remainder glassy. da’. Forewings nearly three times as long as broad; media and cubitus each forking beyond its midpoint. (Japan.) T. (Trichochermes) bicolor Kuwayama (710: 54). d*. Forewings two and one-half times as long as broad; media and cubitus each forking at or near its midpoint, making the mar- ginal cells relatively longer. (Philippines.) T. divisa Crawford (717: 172). c*. Dorsum and vertex not hairy. d’. Forewings somewhat coriaceous, dull opaline in color, about two and one-half times as long as broad. (Tasmania.) T. tasmaniensis Froggatt (’03: 329). d’. Forewings semitransparent, somewhat browned, about four times as long as broad. (Tasmania.) T. dobsoni Froggatt (03: 331). b*. Forewings not colored, glass clear or sometimes very slightly yellowed or smoky. c’. Forewings broadly rounded at apex, not angular nor acutely Eo cgrecng less than three times as long as broad. d‘. Genal cones distinctly shorter than vertex. e’. Forewings nearly three times as long as broad; antenne long; forming galls on Hucalyptus. f*. Legs short, stout; genal cones with long hairs. (Australia.) T. orbiculata Froggatt (’01: 274). f*. Legs long and slender; genal cones less hairy. (Australia.) T. eucalypti Froggatt (’01: 277). e’. Forewings not more than two and one-half times as long as broad; antennz rather short. f'. Body light green; genal cones very short; curling foliage of Olearus. (Tasmania.)...... T. olearie Froggatt (’03: 332). f. Brown to almost black; genal cones one-half to two-thirds as long as vertex; forming galls on Metrosideros. (Hawaii.) T. ohiacola Crawford (’18: 442). d@. Genal cones as long as vertex or sometimes longer. XV,2 Crawford: Jumping Plant Lice 187 e’. Antenne two to three times as long as width of head; dorsum searcely hirsute; insect small to medium in size. f’. Genal cones large, broadly rounded at apex, scarcely conical, with long pubescence; forming galls on Eucalyptus. (Aus- tralia and Tasmania.) .......... T. earnosa Froggatt (’01: 275). f’. Genal cones more or less acutely pointed, distinctly conical. g'. Cubitus forked at or very near its midpoint; inhabiting galls on Metrosideros. h*. Costa of forewing with setwz; male forceps notched be- hind near apex; color usually orange. (Hawaii.) T. iolani Kirkaldy. Crawford (’18: 441). h?, Costa of forewing without easily visible setz2; male for- ceps not notched near apex. j'. Genal cones longer than vertex; antenne two and one- half to three times as long as width of head; thoracic dorsum usually striped with brown; male forceps abruptly narrowed near apex. (Hawaii.) T. lanaiensis Crawford (’18: 443). j*. Genal cones about as long as vertex. k*. Color of body black, dorsum reticulately marked; insect about 3.5 to 4 millimeters long. (Hawaii.) T. pullata Crawford (’18: 444). k?. Color orange or flavous, dorsum not so reticulated; insect 4 to 5 millimeters long. (Hawaii.) T. hawaiiensis Crawford (’18: 444). g’. Cubitus forked distinctly distad of its midpoint. h*. Body color dark brown; wing veins margined with brown; on Casuarina. (Australia.) T. casuarine Froggatt (’01: 284). h?. Color yellow to orange; wing veins not dark margined. j. Antenne three times as long as width of head, black at tan Cindia.).... T. hyalina Crawford (’12: 428). 7. Antenne twice as long as width of head, dista] half black. (India and Europe.) T. urtice Linn. Crawford (’12: 434). e’. Antennz four times as long as width of head; dorsum sparsely hirsute and wing veins setose; insect large. f’. Genal cones about as long as vertex; wing veins uniform in color from base to apex of wing. (India.) T. gigantea Crawford -(’12: 428). 7’. Genal cones nearly twice as long as vertex; wing veins black in distal half, remainder orange red. (Singapore.) T. tenuicona sp. nov. c’. Forewings more or less acutely angled or pointed at apex, not broadly rounded. d'. Forewings three times as long as broad or longer. e’. Gall makers. f'. General color yellowish; forewings very long and acutely pointed; forming galls on Eucalyptus. (South Australia.) T. multitudina (Tepper). Froggatt (’01: 289). 188 Philippine Journal of Science 1919 f?. Color reddish brown; forewings very long but very acutely pointed; forming galls on Tristania. (Queensland.) T. tristanee Froggatt (’03: 334). e?. Not gall makers; general color greenish, yellowish or light brown. f'. Radius of forewing very short, straight; antennz short; insect small, about 1 to 2 millimeters long. (Australia.) T. banksie Froggatt (’01: 281). f?. Radius very long, extending nearly to apex of wing, more or less sinuate; antennze not very short; insect larger, 8 to 4 millimeters long. g'. Head black, abdomen brown above and pale beneath. Jepan,) 24. T. nigriceps Kuwayama (710: 60). g°. Head yellowish green, abdomen green. (Japan and Phil- ippines.) .. T. magna Kuwayama (’10: 59). @. Forewings less than three times as long as broad. e’. Thoracic dorsum not conspicuously pubescent nor hirsute. f’. Thorax chiefly black or dark brown; antennz mostly brown to black, sometimes basal segments pale. g’. Genal cones black. h*. Antenne wholly black or dark brown. +. Antennz about three times as long as width of head; radial stem forking distad of its midpoint. (Japan.) T. nigra Kuwayama (710: 57). ?. Antenne two times as long as width of head; radial stem forking at, its midpoint. (India.) T, analis Crawford (’12: 429). h*, Antennz with third segment whitish. (Japan, Europe.) T. galii Foerster. Kuwayama (’10: 57). g*. Genal cones yellow to orange; antenne twice as long as width of head, black or basal third brown. (Formosa.) T. formosana Kuwayama (710: 58). f°. Thorax yellow, greenish or orange to reddish, rarely darker. g'. Forewings with first marginal cell larger than second; thoracic dorsum yellowish with brown streaks; genal cones short, half as long as vertex. (Formosa.) T. brevifrons Kuwayama (’10: 61). g*. Marginal! cells nearly equal in size. . re; fae oen! short, straight. . Genal cones as long as vertex, axtte at apex; insect about 4 millimeters long or more. j. Antennz about two and one-half times as long as width of head, black; abdomen green ventrad. (Japan and Europe.) T. remota Frst. Kuwayama (’10: 60). j. Antennz less than twice as long as width of head, light brown; abdomen brown. (Béngal.) T. jambolane Crawford (’17: 173). #. Genal cones much shorter than vertex, rounded at apex; insect less than 4 millimeters long. , 3 itt ic XV, 2 Crawford: Jumping Plant Lice 189 j. Antennz more than twice as long as width of head; insect small, about 2.5 millimeters long, forming galls on Eucalyptus. (Australia.) T. circularis Froggatt (’01: 279). j. Antenne very short, not much longer than width of head; insect larger, 3.5 millimeters long, not gall makers. (Bengal, Singapore.) T. fletcheri Crawford (’12: 434). h*. Radius not very short, reaching nearly to tip of wing, often curved or sinuate. 7. Abdomen and genal cones black, latter as long as vertex. (Japan.) T. striola Flor. Kuwayama (’10: 62). +. Abdomen and genal cones light brown or yellow; cones shorter than vertex. j. Radius of forewing nearly straight or bent inward near middle; antenne usually pale yellow, but sometimes brown. k’. Antenne rather short, not over twice as long as width of head. lr’. Body orange yellow to red, sometimes dark red; insect about 4 millimeters long. (Japan.) T. salicivora Reut. Kuwayama (’10: 59). [. Body green or yellowish, about 3 millimeters to tip of folded wings. (Japan.) T. viridula Zett. Kuwayama (’10: 61). k*. Antenng three times as long as width of head; body bright yellow, abdomen green. (Japan.) T, silacea M. D. Kuwayama (710: 58). j. Radius of forewing sinuate; antenne black, thorax yellowish. (Japan.) T. curvatinervis Foerster. Kuwayama (’10: 62). e*. Thoracic dorsum more or less conspicuously pubescent or sparsely hirsute; radius short. , f’. Genal cones nearly or quite as long as vertex; forewings very acutely pointed; pubescence moderately dense. g’. Fore tibiz with two stout spines at apex; body color yellow to reddish brown; apex of forewing at end of media. (AGsTANS} T. eugenie Froggatt (’01: 282). g’. Fore tibie without apical spines; body color dark brown; apex of forewing within second marginal cell. (For- mosa.) T. (Trichochermes) hyalina Kuwayama (10: 55). f?. Genal cones scarcely half as long as vertex; forewings less acutely pointed, veins setigerous; pubescence very sparse. (Philippines. ) -............... T. luzonensis Crawford (’17: 173). a’. Hind wings greatly aborted, reduced to very small stubs; forewings long and acutely pointed, marginal cells far separated. (Singapore.) T. diptera sp. nov. The foregoing key has been arranged in part on descriptions of species which in some instances are too meager and indefinite for 166470——_5 ra 190 Philippine Journal of Science 1919 this purpose. It is quite possible, therefore, that some errors may exist in the synopsis; notwithstanding this, the key will serve as an assistance in future work because of the scattered condition of the literature on this subfamily. Trioza magna Kuwayama (710: 59). Length of body, 2.2 to 2.6 millimeters; forewing, 4 to 4.5. General color green or yellowish green; forewings whitish or glass clear; antenne brown or black; dorsum and vertex briefly and sparsely hairy. Head nearly as broad as thorax, not strongly deflexed, vertex large, broadly concave, median line deeply impressed in anterior half; genal cones as long as vertex, slender and acute, strongly divergent; antennz about two and one-half times as long as width of head, moderately thick. Thorax long and narrow; legs slender, rather long. Fore- _ wings long and narrow, nearly four times as long as broad, acute at apex, veins with whitish sete, membrane whitish in color; first marginal cell larger than second, radius very long. Male (according to Kuwayama) with anal valve about as long as genital segment, much broadened midway in lateral aspect; forceps simple, about as long as anal valve. Female genital segment nearly as long as rest of abdomen, anus large, valves subequal in length, dorsal acutely pointed. Type locality.—Japan, Hekone, Honshu (Matsumura). Phil- ippines—LUZON, Baguio, Benguet (Baker), 1 female. Trioza fletcheri Crawford (12: 434). One male specimen taken at Singapore (Baker) appears to belong to this species, previous representatives of which were taken in India. Trioza tenuicona sp. nov. Plate ITI, fig. 2. Length of body, 2.2 millimeters; forewing, 4. General color pale brown to dark brown or nearly black; vertex in dark forms light brown; eyes black; antennz mostly black, basa] third some- times reddish brown; legs light brown or sometimes paler; fore- wings clear, veins in distal half conspicuously black, yellowish in basal half. Head broad, nearly as broad as thorax; vertex less than half as long as broad, median suture sulcate cephalad, bulging on each side; genal cones nearly twice as long as vertex, directed vertically downward, slender and acutely pointed, not widely divergent, sparsely hirsute. Antennz about four times as long as width of head, rather thick. XV,2 Crawford: Jumping Pla Lice 191 Thorax arched, broad, with scattering hairs. Legs long and stout; tibiz very long, hind tibize without basal spur, four small spines at apex. Forewings large, broadly rounded at apex, veins thick and heavy, not unicolorous. Hind wings about half as long as forewings or more, narrow. Abdomen thick. Male forceps half as long as anal valve, rec- tangular in outline (lateral), truncate at apex, with three teeth on apical margin; anal valve long, broadest at base, tapering to narrow apex. Female genital segment about half as long as rest of abdomen, dorsal valve a little longer than ventral, both subacute. SINGAPORE (Baker), 1 male and 2 females. Trioza divisa Crawford (’17: 172). The description of the female genitalia, having been omitted in the first account of this species, is given here. Abdomen very short; genital segment short, but nearly as long as abdomen, dorsal valve a little longer than ventral, both subacute and tawny in color. LUZON, Benguet Subprovince, Baguio (Baker), 1female. The first specimens, 2 males, were from the same locality. This species is very close to the Japanese Trichochermes bi- color Kuwayama, differing in only a few respects as indicated in the foregoing key. Trioza diptera sp. nov. Plate III, figs. 6 and 7. Length of body, 1.9 millimeters; forewing, 3.1. Thorax orange red, abdomen reddish brown on notum and mostly white on venter; vertex pale brown, with three darker stripes, posterior ocelli red; genz white; antennze white, with three broad black rings; legs white or pale; wings clear. Body surface more or less white-pulverulent. Head nearly as broad as thorax, declivous; vertex nearly as long as broad, median suture sulecate, and a long, slightly cres- centic sulcus on each side; genal cones broad at base, acutely pointed, strongly divergent, directed forward, pubescent with white hairs. Antenne about one and one-half times as long as width of head, slender, conspicuously black and white. Thorax narrow, not much arched. Legs rather short; hind tibiz without basal spur, with four black spines at apex, three together and one alone. Forewings long, acutely pointed, trans- parent, marginal cells far separated. Hind wings reduced to mere stubs, almost wanting. Abdomen long and slender. Male genital segment unusually 192 Philéppine Journal of Science 1919 long and slender, nearly as long as abdomen, reaching far caudad of forceps, slender and acuminate; forceps a little longer than anal valve, situated near base of genital segment, somewhat triangular in lateral outline, broadest near base, shortly petiolate, converging to blunt point at apex, scarcely arched toward each other; anal valve small, not as long as forceps, with a fingerlike projection caudad below midpoint. Female genital segment about as long as abdomen, slender, acuminate; ventral valve about half as long as dorsal, latter long and slightly constricted midway. SINGAPORE (Baker), 2 males and 2 females. This is a very striking species, remarkable for its genitalia and differing from most Trioze in wing characters. It should be perhaps referred to a distinct genus. In the next genus, Megatrioza, there is a marked tendency toward abortion of the hind wings. Genus MEGATRIOZA Crawford Head more or less declivous, usually not as broad as thorax; vertex broader than long; genal cones usually rather short but in a few species long, usually thick and bluntly rounded; frons concealed. Thorax broad or narrow, strongly arched or some- times nearly flat, usually sparsely clothed with long hairs; legs long, apparently strongly saltatory; metacoxee with large pos- terior spurs and also a pair of spiniform processes extending cephalad between bases of hind legs; hind tibize with a small or large spur at base, a more or less conspicuous, thick spine or tooth near apex and three smaller spines at extreme apex. Fore- wings hyaline, sometimes colored, usually acutely pointed, long and narrow, venation triozine with radius and clavus short. Hind wings usually much reduced, seldom more than half as long as forewings and sometimes very greatly reduced. Abdomen short or long; male anal valve broad, with lateral wings pro- duced caudad. Type of the genus, Megatrioza armata Crawford (’15: 264). This genus was first erected for a single large and striking species of the Philippines. The subsequent study, however, of additional South Pacific psyllids makes it appear that this genus has very many representatives throughout tropical Asia and the South Pacific Islands, and no doubt in other tropical areas as well. The most striking characteristics of this generic group are the armed hind tibie, unusual in Triozine; the secondary metacoxal Sn ema ean te ily ie ares le a stead eal eel Soa eal XV,2 Crawford: Jumping Plant Lice 193 spurs directed forward, not found in other genera of the family so far as known to me; and the peculiarly long and slender fore- wings. So many of the species are known to be gall makers or leaf curlers that it probably is a safe surmise that most or per- haps all of the species have similar habits. One species of southwestern United States, Trioza diospyri Ashm., is beyond doubt a representative of this genus and should be referred to it. It has stood heretofore as a species strikingly different from other American Triozx, but never- theless remaining in that genus. It is quite possible that certain other species of Trioza in other parts of the world must be referred to this genus, also. Key to the species. a. Forewing with media forking well beyond its midpoint; first marginal cell usually a little larger than second. b*. Genal cones rather slender, fully as long as vertex; head scarcely declivous; thorax narrow; hind wings about two-thirds as long as forewings. c’. Forewings glass clear; antennz about two and one-half times as long as width of head. (Philippines and Borneo.) M. armata Crawford (’15: 264). ec’, Forewings ocherous, transparent; antenne twice as long as width of head. (Malay Peninsula.) _......... M. armata ochreata var. nov. b’. Genal cones broad and shorter than vertex or wanting; head usually strongly declivous, thorax broad; hind wings about half as long as forewings or sometimes more. c’. Genal cones nearly as long as vertex, broad at base, subacute; head considerably narrower than thorax; female genital segment acutely pointed. (Philippines.) ...........2....2...... M. robusta sp. nov. c’. Genal cones about half as long as vertex, bluntly rounded; head nearly as broad as thorax; female genital segment short. ad. Length of insect to tip of folded wings about 5 millimeters or less; marginal cells of forewing subequal. e’. Dorsum dark brown to black, usually with reddish or orange longitudinal markings; vertex broadly concave between post- ocelli; antennz less than twice as long as width of head; insect about 5 millimeters long. (Philippines, Borneo, and bit «3 iy peemens namely SS Mapa M. eugenioides Crawford (’17: 171). e*. Dorsum uniformly yellowish to light brown; vertex scarcely concave between postocelli; insect about 4 millimeters long. (Fiji; Philippines?) -......... M. vanue (Kirkaldy) (’07: 104). @. Length of insect about 6 to 7 millimeters; first marginal cell of forewing larger than second. e’. Cubitus forked at or near its midpoint; antenne not quite twice as long as width of head; female genital segment large; general color brown to light brown. (Hawaii.) M. palmicola Crawford (718: 452). 194 Philippine Journal of Science . 1919 e?, Cubitus forked well distad of its midpoint; antenne more than twice as long as width of head; female genital segment smaller; general color dark brown to black, with reddish or yellowish stripes and markings. (Malay Peninsula.) M. gigantea sp. nov. ¢. Genal cones wanting, genx slightly or considerably swollen beneath antennal bases; vertex and dorsum very hairy; first marginal cell of forewing very large. (India.) M. hirsuta Crawford (’12: 427). a®?, Medial vein forking at or near its midpoint, upper fork (M:+:) joining costa at apex of forewing; marginal cells subequal; genal cones shorter than vertex. b'. Thoracic dorsum black, without lighter markings. c’. Body slender, nearly all glossy black; hind wings very much reduced; forewings very narrow, radius not margined with black. d. Hind wings reaching to tip of abdomen; male anal valve very large and expansive, reaching caudad of forceps. (Philippines.) M. magnicauda sp. nov. @. Hind wings reaching about half way to tip of abdomen; male anal valve smaller. (Philippines.) M. asiatica Crawford (’15: 266). c’. Body robust; thorax dull black; hind wings about half as long as forewings, reaching beyond tip of abdomen; forewings broader, basal vein and radius conspicuously bordered with black. (Phil- ippines.) . is M. melanoneura sp. nov. b*. Thoracic dorsum brown or. lighter, usually more or less striped or banded with orange. c’. Male forceps acuminate, uniformly tapering to acutely pointed tip; female ovipositor exserted, styliform; antennz# twice as long as width of head. (Ehitippines and Malay Peninsula.) M. stylata sp. nov. . Male forceps not as above, with a long fingerlike process distad; ovipositor not as above; antenne less than twice as long as width of head. (Philippines, Malay Peninsula and Archipelago, Fiji, Java, and Ceylon.)........ M. vitiensis (Kirkaldy) (’07: 108). Megatrioza armata Crawford (’15: 264). The body color of this species is not uniform as indicated in the first description, but the thoracic dorsum has longitudinal bands of brown with pale brownish or yellow-brown between; vertex brown along median suture; forewings clear except an irregular brown area in clavus. The hind wings in this species are long and narrow, about two-thirds as long as forewings, proportionately longer than in most other species of the genus. This is one of the large species of the genus, probably having a wide distribution as indicated by its being found in two separated portions of the South Pacific Archipelago. Type locality—Philippine Islands. Occurring also in Bor- neo—Manorg, West Borneo (Muir), 1 male; Sandakan (Baker), 1 male. (eee ee XV,2 Crawford: Jumping Plant Lice 195 Megatrioza armata ochreata var. nov. Body a little smaller than the species and slenderer; thorax brown, alternating with a brown or yellow-brown duller than in the species; forewings ochraceous throughout, not clear though nearly transparent; clavus brown. Head similar to the species, except genal cones not longer than vertex; antenne scarcely more than twice as long as width of head. Thorax similar but slenderer; legs shorter. Forewings narrower, vena- tion similar; hind wings relatively as long as in the species. Male genitalia similar to the species except forceps slenderer than in the species and a little longer than anal valve. Female genital segment not as long as abdomen. SINGAPORE (Baker), 1 pair. Megatrioza vitiensis (Kirkaldy). Trioza vitiensis KIRKALDY, Proc. Hawaiian Ent. Soc. 1 (1907) 108. Trioza eugenizw CRAWFORD, Philip. Journ. Sci. § D 10 (1915) 265, pl. 1, fig. E, nec Froggatt. Length of body, 2.6 millimeters; forewing, 5 to 6. General color brown, dorsulum with a pale or yellow central streak and often two lateral ones; mesoscutum with several yellowish or pale longitudinal bands; femora dark brown, tibie lighter; antenne pale yellow, black at tip; forewings clear with a small brown spot at anal angle of both front and hind wings. Color of newly emerged adults usually yellowish or pale brownish yellow, without the markings described above. Head declivous, not as broad as thorax; vertex about half as long as broad between eyes, sparsely pubescent, deeply impressed on each side of median line, with a transverse sulcus connecting the two fovezs, somewhat convex in anterior half; genal cones about half as long as vertex, broad, rounded, not much divergent, pubescent. Antenne about one and one-half times as long as width of head. Eyes large. Thorax strongly arched, rather broad, clothed sparsely with long hairs. Legs long and stout; hind tibize stout, with basal spur not very large and apical spines moderately small. Fore- wings long, about three times as long as broad, acutely pointed, marginal cells very long. Hind wings about half as long as forewings or a little more. Abdomen short and broad. Male forceps about as long as anal valve or a little shorter, slender, basal two-thirds about three times as long as thick, rounded apically, and distal third abruptly narrowed with a fingerlike process directed inward and acutely pointed; anal valve about as broad as long, lateral 196 Philippine Journal of Science 1919 wings broadly convex. Female genital segment very short, ‘ much shorter than abdomen, dorsal valve a little longer than ventral, both subacute. Type locality —Fi1JI ISLANDS, Rewa (Muir), March, 1906, 3 males and 2 females. SINGAPORE (Baker), 1 pair. AMBOINA (Muir), 1 female. Pemalonga (Leeuwen-Reijnvaan), 1 male and 1 female, on Eu- genia malaccensis, April 10, 1912. CEYLON, Peradeniya (A. Rutherford), 9 males and females from galls on leaves of Hugenia malaccensis, May 12, 1913 (described as Trioza eugeniz). This appears to be a widely distributed species in the South Pacific in tropical Asia, making galls on Eugenia malaccensis (known also as Jambosa domestica). The name Trioza euge- niz, applied to some newly emerged Ceylonese representatives of this species, was preoccupied by an Australian species named by Froggatt, but subsequent study shows the species to belong to Megatrioza and furthermore to be synonymous with Kirkaldy’s T. vitiensis. The Fiji specimens before me bear no identifica- tion mark indicating that they were examined by Kirkaldy, but they agree well with his description of T. vitiensis and probably are paratypes in as much as the date of collection and locality are identical in both lots. Megatrioza stylata sp. nov. Text fig. 1. Length of body, 3 millimeters; forewing, 5 or more. General color brownish yellow to brown; pronotum, medial portion of thoracic dorsum, and abdomen dark brown or black; legs and antenne yellowish, latter black distad; forewings clear. Head strongly des- cending, narrower than thorax; vertex nearly three times as_ broad - as long, median suture os impressed and a_ fo- Ne ie veal depression on each Fic. 1. Megatrioza stylata sp. nov. side, sparsely pubes- cent. Genal cones broad, rounded, more than half as long as vertex, hairy. An- tennz about twice as long as width of head, segments 3 and 4 a little thicker and more setose than following segments. Thorax very broad and large, sparsely hirsute. Legs stout, hairy; hind tibize long with basal spur small and apical spines moderately large. Forewings large, transparent, acutely pointed, ne ene Mo a oI XV,2 Crawford: Jumping Plant Lice 197 marginal cells long and large. Hind wings about half as long as forewings. Male forceps long, acuminate, curved, acutely pointed, hairy. Anal valve a little longer than forceps, broadest at base, some- what triangular in outline, about twice as long as basal width. Female genital segment about half as long as abdomen, dorsal valve a little longer than ventral, both subacute. Ovipositor very slender and acute, styliform, exserted. SINGAPORE (Baker), 1 pair. Luzon, Laguna Province, Mount Maquiling (Baker), 1 female. i In wing venation this species is apparently allied closely to M. vitiensis. Megatrioza asiatica Crawford. Trioza asiatica CRAWFORD, Philip, Journ. Sci. § D 10 (1915) 266. This species must be referred to Megatrioza, although the basal spur of the hind tibia is almost wanting and the subapical spine is very near the apex. The large anterior metacoxal spurs and the shape and venation of the forewings indicate a close alliance to Megatrioza species. The hind wings in this species are remarkable in being reduced to mere stubs that are not at all functional. This and the following species (M. magnicauda) bear a closer similarity to the American 7. diospyri Ashmead than any other known species, indicating a probable relationship. MINDANAO, Davao (Baker), 1 pair. Megatrioza magnicauda sp. nov. Text fig. 2. Length of body, 2 millimeters; forewing, 4. General color glossy black; antennz yellowish, black at tip; legs black, hind tibiz yellowish or whitish on basal two-thirds; forewings clear, a black spot in clavus. Head and thorax similar to M. asiatica, to which this species is closely related, but distinctly narrower; genal cones a little shorter and directed more down- ward than forward; antenne a little shorter. Forewings rela- / ~—< tively a little broader and mar- S— __ ginal cells shorter and broader; .... 2 sepetrisca magnicands ep. wiv. ‘ hind wings reaching to end of abdomen, while in the related species they scarcely reach half way there. Male forceps similar, but anal valve immense, reaching caudad beyond forceps, rectangular in outline with anal opening much 198 Philippine Journal of Science 1919 elevated. Female genital segment longer and more acutely pointed. Luzon, Mount Limay (Baker), 1 pair. This is obviously close to M. asiatica but appears to be a dis- tinct species. Megatrioza eugenioides Crawford. Trioza eugenioides CRAWrorRD, Philip. Journ. Sci. § D 12 (1917) 171. Length of body, 2 millimeters; forewing, 4. General color chocolate brown with orange yellow markings; vertex flavous; dorsulum brown with a flavous stripe down center and flavous borders; mesoscutum with several tawny stripes and bands; legs except hind femora and antennez except at tip yellowish; fore- wings clear with a black spot in clavus. Head nearly as broad as thorax, only a little declivous; vertex about half as long as broad, with a deep, brown, foveal depres- sion on each side of median line; anterior ocellus in front; genal cones directed downward, about half as long as vertex, narrowly rounded at apex. Antenne very slender, about one and one-half times as long as width of head. Thorax somewhat arched, reticulately marked, sparsely pubes- cent. Hind tibie with two small spurs at base close together, subapical spines moderately large. Forewings clear, very trans- parent; hind wings a little more than half as long as forewings. Female genital segment very short, bluntly pointed. MINDANAO, Butuan (Baker), 3 females. INpIA, Pusa, Bihar (Misra), 1 female taken on the wing, January 26, 1914. Bor- NEO, Sandakan (Baker), 1 female. The last specimen differs slightly from the others in being less pubescent on the dorsum. This species appears to be a widely distributed one and may prove to be identical with some already-known species of Trioza. Megatrioza vanue (Kirkaldy). Trioza vanuze KIRKALDY, Proc. Hawaiian Ent. Soc. 1 (1907) 104. Length of body, 2 millimeters; forewing, 3.2. General color yellowish to light brown, abdomen brown to black on basal half, reddish brown caudad; vertex light brown, with a narrow black line down center. Dorsum very sparsely hairy. Head nearly as broad as thorax, broader than prothorax, not strongly deflexed; vertex only slightly impressed on either side of median line, postocelli scarcely elevated; genal cones short, broadly rounded, not half as long as vertex. (Antennz broken off at base.) Thorax not broad nor much arched; pronotum somewhat de- ! EO Me Le ee a XV,2 Crawford: Jumping Plant Lice 199 pressed and with projecting episternum (dorsal aspect) appear- ing somewhat like a collar. Legs moderately long, femora stout; hind tibiz with an inconspicuous spur at base and a prominent spiniform tooth near apex. Forewings long and slender, three times as long as broad, acute at apex, wholly hyaline, marginal cells about equal in size, radius short. Hind wings about half as long as forewings, very narrow and delicate. Female genital segment short, not longer than preceding ter. gite, with an abruptly acute apex. FIJI ISLANDS, Rewa (Muir), March 3, 1916, 1 female. Luzon, Laguna Province, Mount Maquiling (Baker), 1 male doubtfully of this species. The latter specimen is so imperfectly preserved that its identity is a matter of doubt, but it appears to represent this species. The forceps are nearly as long as anal valve, trian- gular in lateral aspect, about three times as long as basal width, acutely pointed. Anal valve with caudal margin somewhat sin- uate, bulging caudad midway. The specimen from Fiji bears the same date of collection as that examined by Kirkaldy, but there is no indication that he had examined it. Megatrioza gigantea sp. nov. Length of body, 3.6 millimeters; forewing, 4.3. General color brown and orange yellow; vertex yellowish, brown on posterior portion; thoracic dorsum variegated with brown and dull yellow; pleurse yellowish; abdomen mostly brown; femora brown but remainder of legs yellowish; antennez reddish, black at tip; eyes brown; forewings clear with black spot on anal angle. -Head nearly as broad as thorax, about 1 millimeter wide, somewhat declivous; vertex hirsute, about half as long as broad, nearly rectangular in outline, with median suture depressed and a prominent sulcus on each side connected by a transverse, slightly impressed, black line near middle of vertex. Genal cones declivous, thick, about half as long as vertex, bluntly rounded distad. Antennz a little more than twice as long as width of head, slender. Thorax very large and broad, well arched, slightly hirsute. Legs stout, hirsute; hind tibize with two basal spurs, a large subapical spine and three large apical spines. Forewings very large, acute apically, veins thin, membrane delicately transparent. Abdomen short and thick. Female genital segment short, dorsal valve a little longer than ventral, bluntly rounded. Ovi- positor sheath heavy and conspicuously serrate. PENANG (Baker), 1 female. 200 Philippine Journal of Science 1919 Megatrioza robusta sp. Nov. Length of body, 3 millimeters; forewing, 5.1. General color brown and dull yellow; pronotum brown; meso- and metanotum alternating brown and dull yellow streaks; abdomen mostly brown; legs and antennz yellowish, latter brown at tip; fore- wings clear, with a small black spot on anal angle. Head considerably narrower than thorax, declivous; vertex sparsely hirsute, about half as long ag broad, front and hind margins subparallel, arcuate forward, with a foveal depression on each side of median line. Genal cones nearly or quite as long as vertex, slightly declivous, broad at base, subacutely pointed. Antenne scarcely twice as long as width of head. Thorax very broad and robust, sparsely hairy, finely impunc- tate; legs rather short and stout; hind tibize with small spurs at base, apical and subapical spines moderately large. Fore- wings large, clear, acutely pointed. Hind wings about half as long as forewings, narrow. Abdomen short and thick. Female genital segment short, dor- sal valve a little longer than ventral, both subacute. Luzon, Mount Banahao (Baker), 1 female. Megatrioza melanoneura sp. nov. Length of body, 3 millimeters; forewing, 5. General color dark brown to black; vertex, gen, thoracic pieurs, hind tibiz, and antenne all dull reddish yellow; antenne and hind tibie black distad; forewings clear, with a black spot on anal angle and a black band following basal vein along radius to costal margin, Head narrower than thorax, declivous. Vertex about half as long as broad, sparsely hairy, bulging roundly in front, with a transverse depression extending from in front of posterior ocelli arcuately toward occipital margin of head. Genal cones decli- vous, about half as long as vertex, bluntly rounded, hairy. An- tenne about one and one-half times as long as width of head. Thorax broad, well arched, sparsely hairy. Legs thick, very hairy; hind tibie with small basal spur, a moderately large subapical spine and two apical spines. Forewings large, trans- parent, veins narrow and black; basal and radial veins conspic- uous by the black border. Hind wings about half as long as forewings. Abdomen large and thick. Male forceps about as long as anal valve, slender, arched and subacutely pointed. Anal valve mod- erately long, about one-third as broad at base as long, nar- » Da mre SR Sal cce XV, 2 Crawford: Jumping Plant Lice 201 rowing toward apex. Female genital segment about three-fifths as long as abdomen, converging to an acute point, dorsal valve a very little longer than ventral. Luzon, Mount Banahao (Baker), 1 male: Benguet Subprov- ince, Baguio (Baker), 1 female. Megatrioza hirsuta Crawford. Kuwayama hirsuta CRAWFORD, Rec. Indian Mus. 7 (1912) 427, pl. 33, fips. V5; pl.-85, fig, 1. With additional material of this species available for compar- ison, it now appears that it must be referred to Megatrioza. The wing venation, the spurred hind tibie, the anterior meta- coxal projections, and the galil-forming habit, all relate this un- doubtedly to this genus rather than to Kuwayama, in which it was first placed with much hesitation. The gene are scarcely swollen beneath the vertex and not at all conical, but there is a strong tendency in Megatrioza toward the suppression of the genal cones. The large size and the hairy head and thorax make this a striking species. INDIA, Lonavla, Bombay (H. H. Mann), 1 male, in galls on Terminalia tomentosa, July, 1916. Genus CEROTRIOZA Crawford Head scarcely declivous, rather long; vertex produced in front into two hornlike epiphyses; genze produced into cones directed forward; antenne slender. Thorax not strongly arched, narrow; hind tibiz with small basal spur or callus and subapical spine (as in Megatrioza). Forewings narrow, opaque or semiopaque, maculated, first marginal cell usually larger than second; hind wings nearly as long as forewings. Type of the genus, Cerotrioza bivittata Crawford (’18: 454). This genus appears to be related to, probably sprung from, Megatrioza but differs sharply in the thickened and semiopaque forewings and peculiarly shaped vertex. The type species is one found in the Hawaiian Archipelago. Key to the species. a’. Genz produced into more or less conical processes directed forward, without dark vitta on each side of body. b*. Length of insect to tip of folded wings about 3 millimeters; color dark brown; vertex horns nearly half as long as genal cones. (Sinwanore:) csiacicicn, .... ©. eorniger sp. nov. b*. Length to tip of folded wings about 4 millimeters; vertex horns much SIIB NOT. =< -CROPNOO oe a C. microceras sp. nov. 202 Philippine Journal of Science 1919 a?. Genz not produced into conical processes, but somewhat spherically swollen beneath antennal bases and protruding vertex; with a prom- inent dark vitta on each side of head and thorax and extending to tip of each forewing. (Hawaii.) C. bivittata Crawford (718: 454). Cerotrioza corniger sp. nov. Plate III, fig. 12. Length of body, 1.9 millimeters; forewing, 2.6. General color brown, mottled and banded with yellowish white; vertex with margins and median suture bordered with dirty white; pronotum with three whitish points ; notum of thorax with irregular stripes and bands of dirty white; abdomen mostly dark brown; legs yellowish, femora with brown spots; antennz pale brown. Head slightly declivous, as broad as thorax; vertex about three-fifths as long as broad between eyes, with a relatively large foveal depression on each side of median line, anterior horns situated over genal cones and about half as long as latter, acute and upturned; front ocellus visible from above; genal cones about half as long as vertex, directed forward, divergent, bluntly pointed, scarcely hairy. Antenne about one and one-half times as long as width of head, slender. Thorax narrow. Legs long and slender, especially the tibiz; hind tibie with very small basal spur and a larger, black sub- apical spine. Forewings nearly opaque, membrane thickened and whitish or pale brownish with numerous small, dark brown spots all over; broadest subapically but the apex abruptly acutely pointed; first marginal cell about twice as large as second. Abdomen long. Female genital segment gency. as long as abdomen, slender, acutely pointed. SINGAPORE (Baker), 1 female. Cerotrioza microceras sp. nov. Plate III, fig. 10. Length of body, 2.2 millimeters; forewing, 3.1. General color grayish white with pale brown markings on mesonotum and pleure; legs pale; antennz yellowish with black tip; abdomen light brown. Head somewhat declivous, not quite as broad as thorax; ver- tex about three-fourths as long as broad, with an irregular star- shaped depression on each side of median line, a pair of small epiphyses in front adjoining median suture and a larger pair outside of these, the latter scarcely one-fourth as long as genal cones; front ocellus visible from above. Genal cones a little more than half as long as vertex, divergent, bluntly pointed, directed forward, somewhat pubescent. Antennz a little more than twice as long as width of head. AN, Crawford: Jumping Plant Lice 203 Thorax narrow, somewhat arched; legs not as long as in re- lated species; hind tibize with distinct basal spur and large sub- apical spine. Forewings nearly opaque, membrane thickened and whitish, with numerous small, light brown spots all over. Shaped as in related species, but apex more pointed. Abdomen long. Female genital segment as long as abdomen, acutely pointed, slender. WEST BoRNEO, Mowong (Muir), 1 female. Genus STENOPSYLLA Kuwayama Head declivous, not as broad as thorax; genze produced into conical processes, not covering frons completely but nearly so; clypeus subspherical; beak of median length. Antenne long, very thick and thickly pubescent. Eyes large. Thorax broad; legs stout; hind tibiz with basal spur and subapical spine and several apical spines. Forewings large, hyaline, triozine in ve- nation, radius very long. Type of the genus, Stenopsylla nigricornis Kuwayama. This genus bears some resemblance to Bactericera but differs in the very pubescent antennez and presence of genal cones. In the armature of the hind tibiz there is a close similarity to Megatrioza. Key to the species. a’. Body yellowish or light brown, dorsum flecked with brown; genal cones acutely pointed; antenne half as long as forewings; insect about 5 or 6 millimeters long. (Formosa.) S. nigricornis Kuwayama (’10: 54). a’, Body light brown, dorsulum and vertex black or dark brown; genal cones broadly rounded at apex; antenn# three-fourths as long as fore- wings; insect 6 to 7 millimeters long. (Philippines.) S. longicornis sp. nov. Stenopsylla longicornis sp. nov. Text fig. 3. Length of body, 4 millimeters; forewing, 6; antennez 4.5. Dorsulum and eyes dark brown or black; vertex, mesoscutum and abdominal notum brown; rest of body orange yellow to light brown; antenne brown to black; legs orange yellow. Head much narrower than thorax, strongly declivous; vertex smooth, broadly concave, about half as long as broad, median suture longer than lateral edges, without foveal depressions. Genal cones nearly as long as vertex, directed vertically down- ward, broad and bluntly rounded, divergent, sparsely hairy. Frons slightly visible at front ocellus. Antenne about five times as long as width of head. segments 1 and 2 short and thick, « 204 Philippine Journal of Science 1919 not hairy; 3 and 4 as thick as 2 and densely covered with stiff setee set on tiny tubercles; 5 to 10 gradually becoming more slender, about half as thick at tip as segment 3, all uniformly densely setose. Thorax very broad, well arched, surface smooth and glossy. Legs short and stout; hind tibie a little longer than femora, with a moderately large basal spur and large distal spines. poe: : Fia. 3. Stenopsylia longicornis sp. nov. Forewings large, acutely pointed, hyaline, veins heavy, costa se- tose; radius very long; first marginal cell much larger than second. Abdomen stout. Female genital segment about half as long as abdomen, dorsal valve a little longer than ventral, subacute, descending. LuZON, Laguna Province, Mount Maquiling (Baker), 1 female. This appears to be very close to the Formosan species of the same genus, but is a larger insect and differs structurally as well as in coloration. : BIBLIOGRAPHY The following bibliographical list is by no means a complete one. A more nearly complete list may be found in my mono- graph, 1914. Only the works referred to in the preceding pages are listed below: CRAWFORD, D. L.: 12 Indian Psyllide. Rec. Indian Mus. 7 (1912) 419-4387. 18 New genera and species of Psyllide from the Philippine Islands. Philip. Journ. Sci. § D 8 (1913) 293-301. ‘14 A Monograph of the Jumping Plant Lice (Psyllide) of the New World. Bull. U. S. Nat. Mus. 85 (1914) 1-186. ‘15 Ceylonese and Philippine Psyllide. Philip. Journ. Sci. § D 10 (1915) 257-269. 17 she be and Asiatic Psyllide. Philip. Jowrn. Sci. § D 12 (1917) 18 The jumping plant lice of the Hawaiian Islands. Proc. Hawaiian Ent. Soe. 3 (1918) 4380-457. XV,2 Crawford: Jumping Plant Lice 205 FROGGATT, W. W.: 700 Australian Psyllide. Proc. Linn. Soc. New South Wales 25 (1900) 250-302, pls. 9-14. ’01 Australian Psyllide. Part II. Proce. Linn. Soc. New South Wales 26 (1901) 242-298, pls. 14-16. 03 Australian Psyllide. Part III. Proc. Linn. Soc. New South Wales 28 (1903) 315-338, pls. 4 and 5. KIRKALDY, G. W.: 707 On two new Vitian Chermide. Proc. Hawaiian Ent. Soc. 1 (1907) 103 and 104. 08 A catalogue of the Hemiptera of Fiji. Proc. Linn. Soc. New South Wales 33 (1908) 345-390. KUWAYAMA, S.: 08 Die Psylliden Japans I. Trans. Sapporo Nat. Hist. Soc. 2 (1908) 149-189. 710 Die Psylliden Japans II. Trans. Sapporo Nat. Hist. Soc. 3 (1910) 53-66. RUBSAAMEN, E. H.: ’99 Gallen aus Europa, Asien, Afrika und Amerika. Ent. Nachricht. 25 (1899) 225-281. ; 1664706 ILLUSTRATIONS PLATE I . Pauropsylla brevicornis sp. nov., forewing. . Pauropsylla brevicornis sp. nov., antenna. . Pauropsylla verticis sp. nov., forewing. . Pauropsylla verticis sp. nov., female genital segment. . Leptynoptera sulfurea g. et sp. nov. Leptynoptera sulfurea g. et sp. nov., basal segments of antenna. . Pauropsylla apsylloides sp. nov. ' . Paurocephala minuta sp. nov. . Paurocephala maculata sp. nov. . Paurocephala conigera sp. nov. . Heteroneura oceanica g. et sp. nov. 12. Heteropsylla longicornis sp. nov. FIG. _ SHMAAMR WH He ry _ PLATE II Fic. 1. Tyora indica sp. nov. 2. Thysanogyna minor Crawford, caudal view of male forceps. 3. Tenaphalara malayensis sp. nov. 4. Thysanogyna minor Crawford, lateral view of female genital segment. 5. Tenaphalara striata sp. nov. 6. Tenaphalara triozipennis sp. nov. 7, Euphalerus grandis sp. nov. 8. Kuphalerus maculosus sp. nov. 9. Arytaina meridionalis sp. nov. 10. Arytaina flava sp. nov. 11. Arytaina iolani sp. nov. PLATE ITI Fig. 1. Arytaina punctinervis sp. nov. 2. Trioza tenuicona sp. nov. 3. Arytaina pulchra sp. nov. 4. Arytaina brevigena sp. nov. 5. Psylla muiri sp. nov. 6. Trioza diptera sp. nov., male genitalia. 7. Trioza diptera sp. nov., forewing. 8. Psylla leprosa sp. nov. 9. Psylla bakeri sp. nov. 10. Cerotrioza microceras sp. nov. 11. Psylla compta sp. nov. 12. Cerotrioza corniger sp. nov. TEXT FIGURES Fic. 1. Megatrioza stylata sp. nov. 2. Megatrioza magnicauda sp. nov. 8. Stenopsylla longicornis sp. nov. 207 CRAWFORD: JUMPING PLANT LICE.] [PHILIP. JOURN. Sci, XV, No. 2. PLATE |. NEW PSYLLIDA. CRAWFORD: JUMPING PLANT LICE.] [Puiuie. Journ. Scr, XV, No. 2. PLATE Il. NEW PSYLLIDA. CRAWFORD: JUMPING PLANT LICE.] [PuHIuie, JouRN. Sct., XV, No. 2. Se = Ge > SS PLATE Ill. NEW PSYLLIDA, THE GENUS KRISNA (JASSIDAS) By C. F. BAKER Of the College of Agriculture, University of the Philippines, Los Banos FIVE PLATES The Orient possesses an extensive fauna of large, fine jassoid insects related to Selenocephalus. These are all stout-bodied, short-headed jassids and commonly with the face, or the vertex, or the pronotum, or all three, coarsely sculptured, this sculpturing commonly consisting of transverse rugosities or wrinkling, and frequently with the clavus and the base of the corium strongly punctate or rugose-punctate. The anterior border of the head varies from sharp margined to very obtusely rounded between the vertex and the face. In general, the jassids in question are distinguished by two important structural characters. The antennal scrobes are large, and not confined to the cheek, but involve a portion of the lateral surface of the frons, and are overhung by a thick, sharp, mar- — gined ledge, which passes on to the frons, the carinate edge curving downward on the latter, in some cases nearly to the clypeus. The lateral margin of the frons thus passes upward through the scrobe and cuts the supra-antennal ledge. Neither previous descriptions nor drawings bring out this important structural feature, and previous drawings of the face are uni- formly incorrect as to this area. The ocelli are on the upper an- terior border of the head or near it, near to the eyes, and often visible from above. The actual superior frontal border is just below the ocelli; this suture is rarely distinct, though frequently rudimentary at the sides. The anterior border of the head may be margined by a more or less sharp transverse carina above the ocelli, and at the same time may have another similar carina just below the ocelli, in which case we say that the ocelli lie in a transverse furrow or sulcus. Either one or both of these carine may be absent or there may be several sharp transverse carine. Sometimes the sulcus is very narrow; at other times, broad with a few to many transverse wrinkles within. Sometimes both ca- rinz are lacking, and only the transverse band of wrinkles re- mains. But the wrinkles or the carine or the sulcus, in some form, are always present in the members of this group. The 209 210 Philippine Journal of Science 1919 superior frontal margin thus always lies below the apparent upper anterior border of the head, which may not be at all ex- tended above and cephalad of the superior frontal suture, or may be strongly so as in the Tartessusaria. Some of these insects possess a superficial resemblance to bythoscopids, and Walker wrongly described various species of Tartessus and other genera in Bythoscopus. But in all, including Tartessus, the ocelli are far higher and very close to the eyes, and this and other char- acters mark them as true jassids. In superficial habitus Krisna presents a very remarkable resemblance to Gypona, so much so that Kirkaldy described a new genus, Hogypona, in the Gypo- nine, for one of its species. It has, of course, not the remotest relationship to Gypona. In addition, the true Gyponinz (ex- cluding the Penthimiine) are exclusively American. A minor character common to this assemblage may be ob- served in the lore, the suture of which, above, usually does not reach the frontal suture, but commonly ends in a more or less radiate arrangement of minute wrinkles some distance from the frontal margin. In rare cases this suture is subobsoletely ex- tended and strongly recurved to the lateral frontal margin. making the lora ear-shaped. Previous drawings, even those of Signoret, are incorrect in this detail of structure. The jassids discussed above fall into three tribes as follows: Key to the tribes. a. Head as wide as or wider than pronotum; frontoclypeal suture usually distinct; lateral pronotal carina very short. b’, Pronotum strongly, narrowly rounded cephalad, and extending beyond anterior margin of eyes; ocellar area broad and usually transversely wrinkled; appendix of tegmina usually extending around apex, and very broad apically, rarely wanting as in Macroceps; ovipositor long, exserted; Pediopsis-like in form................... Tartessusaria. b*. Pronotum not extended cephalad of anterior margin of eyes; ocellar area narrow, usually once or several times sulecate; appendix of tegmina usually extending only to apex or less, and broadest sub- apically; ovipositor little exserted.... Selenocephalaria. a’. Head distinctly narrower than pronotum; frontoclypeal suture obsolete or subobsolete; ocellar area broad and transversely wrinkled and with one or more transverse carinew above it; lateral pronotal carina (and pronotum behind eyes in lateral view) long; ovipositor little or not exserted; Gypona-like in formicc2.i.00o.cccc ce o Stegelytraria. Distant names two of these tribes but gives no diagnostic char- acters. He places Krisna in the Selenocephalaria, but Gessius which might easily be called a subgenus of Kriena he puts in the Tartessusaria. Several genera formerly placed in the Sele- nocephalaria do not belong in this congeries at all, and a number XV,2 Baker: The Genus Krisna (Jassidx) 211 of genera formerly classified elsewhere must be removed to it. In fact, some of the species described in Selenocephalus do not even belong in the Selenocephalaria. The tribe Tartessusaria includes Tartessus Stal, (?) Borduria Dist., Macroceps Sign., Putoniessa Kirk., Sarpestus Spang., and Thymbris Kirk. Some of these may be synonymous with Tartessus. Drabescus Stal should be removed to the Selenocephalaria. The present paper is concerned especially with two genera of the Stegelytraria, the genera of which may be separated as follows: Tribe STEGELYTRARIA Key to the genera. a’. Tegmina without distinct appendix; length of tegmina beyond clavus about one-third length of clavus. b*. Vertex anteriorly horizontally rounded; anterior pronotal margin arched, posterior margin slightly emarginate; tegmina with two SUvaniCal GCOS oe a i ee as Stegelytra M. and R. b*. Vertex anteriorly subangulate; pronotum anteriorly subtruncate be- tween eyes, posteriorly subtruncate between scutellar angles; teg- ming with: four.subsdpical . cells.22cix:. see Iberia Kirk. a’. Tegmina with a well-developed appendix; length of tegmina beyond clavus more than half length of clavus. b'. Venose portion of tegmina beyond clavus strongly narrowed, the apical cells long and narrow, anterior margin of pronotum broadly rounded between eyes; frons medially never carinate and always much broader than width of eye in facial view. c’, Tegmina apically with numerous supernumerary veins (Plate I, MOM Los I Ne ee che ea Krisna Kirk. ec’. Tegmina apically normally without supernumerary cross veins Renee Yl a ee eee . Gessius Dist. b*. Venose portion of tegmina beyond clavus not narrowed, the apical cells broad in proportion to length; anterior margin of pronotum narrowly rounded between eyes; frons medially carinate; distance between eyes subequal to width of eye in facial view. Celidioides Sign. Genus KRISNA Kirkaldy A common evening visitor to one’s table lamp, almost any- where in the Malaisian countries, is a vividly virescent leaf- hopper of the genus Krisna, a half inch, more or less, in length, and with the transverse fore margin of the vertex black or a bright reddish color. This is likely to be Krisna strigicollis, but other species frequently occur. These insects were known for many years under the generic name Siva Spinola,' but that name being preoccupied, Kirkaldy * renamed the genus Krisna. *Mem. di Matem. o di Fis. Soc. Ital. Modena (1852) 167. *Entomologist 33 (1900) 243. Bie Philippine Journai of Science 1919 The type species of the genus as described by Spinola was Siva strigicollis, now supposed to be widely distributed through the Orient. In 1891 Kirby® described striata from Ceylon, though neither Kirby nor other authors have given diagnostic characters to distinguish it clearly from strigicollis. In 1908 Distant‘ described Krisna sherwilli from Sylhet, actually sep- arating it on color characters only, which are extremely un- stable in this genus, the structural characters mentioned being entirely generic in value or without diagnostic significance. In 1910 Bierman * described a Siva rosea from Java. Here again most of the structural characters mentioned are only generic in value. Long ago, Walker indicated three species now recognized as pertaining to Krisna; namely, Acocephalus stramineus,’ By- thoscopus indicatus,’ and Bythoscopus testaceus.’ Distant refers all of these to Krisna strigicollis, but since no author has pre- viously described the diagnostic structural characters of strigi- collis, and since it is entirely probable that under insufficient description several distinct things have been referred to strigi- collis, all of this synonymy will have to be reéxamined. In 1901 Breddin * attempted to define a AKrisna straminea WIk. var. in- dicata Wlk., from Celebes, thereby adding to the usual Walker- ian uncertainty, since practically his entire description is com- posed of generic characters. Several species of the genus have also been described from Africa. This genus includes a very homogeneous group of species, the form and markings of all being very similar. The vividly vi- rescent species commonly fade to stramineous in dried specimens, and carmine markings usually entirely disappear. When deep olive green or some shade of rufous the color may not change in drying, though in life there is wide variation in the shade of rufous. The black dot at the apex of the clavus is quite constant in most of the species, but the black markings on the vertex are very inconstant, sometimes varying in the same species from a narrow continuous black stripe near the fore margin to separated spots at the middle and sides, and even these may be absent. Specimens of green species may occur with a strong pinkish or reddish tinting. * Journ. Linn. Soe. Zool. 24 (1891) 171. “Fauna Brit. India, Rhynch. 4 (1908) 299. * Notes Leyden Mus. 33 (1910) 61. *List. Hom. 3 (1851) 847. "List. Hom. Suppl. (1858) 266. *Journ. Linn. Soc. Zool. I (1857) 173. *Abh. Naturf. Ges. Halle 24 (1901) 130. XV, 2 Baker: The Genus Krisna (Jassidx) 213 All of this means that no sound classification of the species (as in most jassids) can be based on colors, but that detailed study must be made of all important features of the external anatomy. Nor does it avail in the least to build up species de- scriptions by repeating characters common to all or even part of the species in the genus. All of these species are long oval in outline and strikingly gyponoid in form, with the vertex very bluntly angulated or rounded between the eyes (it may be quite different in the sexes) ; the face and the vertex variously rugose and wrinkled; the pronotum, except anteriorly, thickly and finely cross striate; the scutellum having a strongly curved discal furrow, before which the surface is rugose, behind it, cross wrinkled; and the tegmina more or less thickly punctate. The number and the position of the supernumerary veins vary in the same species. The pronotum always possesses a long, strong, complete lateral carina. The frons is thickly rugose of varying degrees of coarseness. The cheek is slightly roughened below the eye, obliquely wrinkled on the middle third, and rugose below. The clypeus and the lore are more or less rugosely roughened. The size is variable within narrow limits in all of the species. ; Key to the species of Krisna. a’. Lower margin of transversely wrinkled ocellar area with several strong, irregular, noncontinuous wrinkles of similar size, and with wrinkles on supra-antennal area; frons below ocellar area not or only moder- ately depressed; smaller species of slenderer form. b*, Frons evenly convex from margin of vertex to clypeus; length of male, 8.5 millimeters... we K. minima sp. nov. b?. Frons distinctly depressed below ocellar area; size usually larger. ¢. Vertex with but one irregular submarginal carina, distinct at least at sides, the entire disk uniformly coarsely rugose. d. Supra-antennal carina passing on to frons in an even curve, or for a short distance straight and then curved; tegmina trans- lucent, the puncturation rather fine. e', Lateral carina of pronotum about same length as eye in lateral view. f. Supernumerary veins numerous; wings milky subhyaline; face concolorous with remainder of undersurface. K. strigicollis Spin. f'. Supernumerary veins few; wings dark smoky. g'. Supra-antennal area as broad as long, face concolorous. K. simillima sp. nov. g. Supra-antennal area longer than broad; frons black. K, nigrifrons sp. nov. e’. Lateral carina of pronotum distinctly longer than eye in lateral K, penangensis sp. nov. 214 Philippine Journal of Science 1919 d@, Supra-antennal carina sharply angled where it crosses frontal suture, from this point passing straight cephalad for a short distance; tegmina thickly coriaceous, opaque, the puncturing very coarse. e’. Transverse wrinkles numerous on face above, and fine and regular; supra-antennal area with about ten small wrinkles between its lower point and the ocellus........ K. muirii sp. nov. e’. Transverse wrinkles few on face above, and coarse and irreg- ular; supra-antennal area with about five large wrinkles between its lower point and the ocellus. K, olivascens sp. nov. c’. Vertex with a series of about four submarginal carinz, the pos- terior incomplete, together occupying about one-half of vertex, the posterior portion of disk rugose; otherwise like the species. K. olivascens var. singaporensis var. nov. a’. Lower margin of transversely wrinkled ocellar area with a strong, uniform, transverse carina, without or with very few weak wrinkles below it, and with the wrinkles above it much weaker; supra-antennal area smoothly concave; frons below ocellar area strongly deeply depressed, and with three large smooth foveate are above; punctura- tion of tegmina minute, weak, sparse, most evident along veins; vertex without a submarginal carina near anterior marginal carina. b'. Disk of pronotum very strongly arched; color pale virescent; wings SAUKY BUDA UAIING oor a a K, magna sp. nov. b*. Disk of pronotum very weakly arched; teemina carmine basally; wings een Sneek K, colorata sp. nov. Krisna minima sp. nov.'® Plate I, fig. 2; Plate II, fig. 8; Plate Ill, figs. 3 and 6. Length, male, 8.5 millimeters. Pale stramineous dry, but doubtless virescent in life. No dark spot at apex of clavus and no dark markings on head. Vertex anteriorly margined with carmine. Similar to K. strigicollis but very much smaller. LAPPA ISLAND, Maca) (F. Muir). Krisna strigicollis Spinola. Plate I, figs. 1, 5, 6, and 8; Plate II, figs. 7 and 9; Plate III, fig. 10; Plate IV, figs. 6 and 12. Length, female, 12 millimeters; male, 10 millimeters. Pale stramineous dry, but vividly virescent in life, with vertex an- teriorly bordered more or less broadly with carmine, and with a submarginal black stripe, which is often obsolete on middle of sides, leaving two black dots near margins of eyes and two near median line, but these also may be obsolete. Wings usually milky subhyaline. Last ventral segment of female behind trun- cate to more or less broadly angularly emarginate. “ Characters given in the synopsis are not repeated in the descriptions. XV,2 Baker: The Genus Krisna (Jassidz) 215 A common species in the Malay Peninsula, Singapore, Borneo, and the Philippines. I also have a specimen taken in Java by Mr. F. Muir. Krisna simillima sp. nov. Plate II, fig. 2; Plate III, fig. 11; Plate IV, figs. 7 and 9. Length, female, 11 millimeters. Rufous-stramineous; a sub- marginal dark dot on either side of vertex near eye margins. Wings smoky. Last ventral segment subtruncate posteriorly. BORNEO, Sandakan (Baker). Larger series may prove this to be an extreme variety of K. strigicollis. Krisna nigrifrons sp. nov. Plate I, fig. 4; Plate II, fig. 10; Plate IIl, fig. 2; Plate IV, fig. 10. Length, male, 9 millimeters. Rufous-stramineous. Vertex marked as in K. strigicollis. Disks of frons, clypeus, and lore black. SINGAPORE (Baker). Krisna penangensis sp. nov. Plate II, fig. 4; Plate Ill, fig. 8; Plate IV, figs. 5 and 11. Length, female, 11 millimeters. Pale rufous above, stramin- eous below, vertex marked as in K. strigicollis. Three inner apical cells of tegmina transparent, remainder opaque. Wings smoky. Last ventral segment truncate posteriorly. PENANG (Baker). Krisna muirii sp. nov. Plate I, fig. 7; Plate III, figs. 13 and 15; Plate IV, figs. 2 and 8. Length, female, 12 millimeters. Clear rufous above, stramin- eous below. Vertex with a submarginal fine uniform dark line. Tegmina thickly coriaceous and strongly punctured throughout, the appendix smoky and the inner apical cells stained with smoky. Face with a dark arcuate stripe passing across middle of cellar area. Last ventral segment somewhat angularly emar- ginate posteriorly. WESTERN BORNEO, Telok Ayer (Muir). Krisna olivascens sp. nov. Plate I, fig. 3; Plate II, fig. 6; Plate III, figs. 5 and 9; Plate IV, fig. 4. Length, female, 13 to 14 millimeters; male, 11 millimeters. Deep olive green above, scutellum, anterior portion of prono- tum, and vertex paler; below stramineous, tibiz more or less carmine, the foretibie with angular edges black. Vertex an- 216 Philippine Journal of Science 1919 teriorly narrowly margined with black, this margined behind with carmine; face above with a submarginal narrow black line, this again margined below with carmine. Tegmina thickly co- riaceous, very strongly punctured, the appendix smoky, and some of the inner apical cells stained with smoky. Hind margin of last ventral segment incurved. This very fine species is not uncommon in Singapore and Penang; and I have specimens from Sandakan, British North Borneo. A remarkable variety of this species, from Singapore, has the vertex quite differently sculptured, and it is somewhat paler in color, but otherwise similar to the typical form. [ call it var. singaporensis var. nov. Krisna magna sp. nov. Plate II, figs. 3 and 5; Plate III, figs. 1, 4, 12, and 14; Plate IV, fig. 3. Length, female, 14 millimeters; male, 13 millimeters. Dis- tinct from all other oriental species in the genus by its very robust form, the body being very thick, and very strongly arched across base of pronotum. Very pale virescent above, stramin- eous below. Vertex anteriorly with two minute black dots next to eyes. No spot at apex of clavus. Wings milky subhyaline. Sexes strikingly different in form of head. Specimens come from Penang Island and from Sandakan, British North Borneo, though apparently the species is not com- mon. This is the largest known oriental species of the genus. Krisna colorata sp. nov. Plate II, fig. 1; Plate III, fig. 7; Plate IV, fig. 1. Length, female, 13 millimeters. Anterior portion of pronotum, vertex, and all below stramineous; pronotum posteriorly and scutellum pale rufous. Tegmina smoky subhyaline, veins and basal two-thirds washed with carmine. Wings deep smoky. Vertex with a marginal line of carmine, and two large black dots near eye margins. Last ventral segment with the margin broadly scarious. BRITISH NORTH BORNEO, Sandakan (Baker). Genus GESSIUS Distant There is no question of the very close relationship of this genus to Krisna; the venation of the tegmina furnishes the only real means of separating them and even this shows close rela- tionship, in as much as occasional abnormal supernumerary veins XV,2 Baker: The Genus Krisna (Jassidz) 217 occur in Gessius. Here are included some common insects of the Malay Peninsula, Borneo, and the Philippines, which I have no doubt were frequently collected previously ; and specimens will probably be found in collections placed with Krisna, especially Krisna strigicollis. Among these insects there is a most confusing variation of color shades from bright rufous to stramineous (virescent). All specimens in a large series of several species and varieties possess an unmarked vertex with at most a crimsoned anterior margin, a dot at the apex of the clavus, the wings more or less smoky, the dorsum more or less crimsoned, and all below stramin- eous. The three or four inner apical cells of the tegmina are subhyaline and impunctate, in strong contrast to the remainder of the tegminal surface. General sculpturing very similar to that of Krisna and simi- larly distributed. In all of my material I have no specimens with ‘the claval area fuscescent, this color longitudinally continued to apex of tegmina”’ as described by Distant for the type of the genus, G. verticalis, from Burma.’ Is it possible that this fuscescent area is due to the color of the smoky wings beneath? Key to the Malayan species. a’. Transverse wrinkles of ocellar area confined to a space twice the width of an ocellus, the frons suddenly depressed below this, and irregularly rugose; supra-antennal area smooth................... G. malayensis sp. nov. a*. Entire upper part of face, down to lower margin of supra-antennal ledges, including supra-antennal arex, with coarse uniform wrinkles; frons scarcely depressed G. pallidus sp. nov. Gessius malayensis sp. nov. Plate V, figs. 1, 2, 5, 7, 10, and 12. Length, female, 10.5 millimeters; male, 9 millimeters. Above clear shining pale rufous (variable), below stramineous. The form of the last ventral segment in the female is distinctive, it being deeply, broadly, angularly emarginated, the sides of the emargination sharply notched. Wings very dark veined. A common species in the Malay Peninsula and Borneo. A closely similar form is found in Butuan, Mindanao, and in Basilan, which differs only in that the lateral notches in the emargination of the female segment are uniformly shallow and reduced to a strong sinuation. This form may be known as var. mindanaensis var. nov. “ Fauna Brit. India, Rhynch. 4 (1908) 302. 218 Philippine Journal of Science Gessius pallidus sp. nov. Plate V, figs. 3, 4, 9, and 18. Length, female, 10 millimeters; male, 8.5 millimeters. Above stramineous (in life) to pale virescent; veins of wings pale. The truncate last ventral segment of female is distinctive for the typical form of this species. It is the analogue of Krisna strigicollis but smaller and slenderer. SINGAPORE (Baker), frequent. A form very closely similar to this in every way, except that the hind margin of the last ventral segment of the female is shallowly sinuate, occurs commonly at Sandakan, British North Borneo. I name this var. sinuatus var. nov. Another variety, collected at Telok Ayer, Western Borneo, by Mr. F. Muir, is pale rufous above, and has the hind margin of the last ventral segment of the female distinctly incurved, and this may be known as var. incurvatus var. nov. In my Singapore material I have one male specimen, only 8.5 millimeters long, which I cannot separate structurally from the typical form, but which is very dark sordid rufous above, with apical third of tegmina pale. I name this var. suffusus var. nov. (Plate V, figs. 6, 8, and 11). ILLUSTRATIONS PLATE I Fic, 1. Krisna strigicollis Spin., forma; tegmen. 2. Krisna minima sp. nov., head and pronotum. 3. Krisna olivascens sp. nov., tegmen. 4. Krisna nigrifrons sp. nov., head and pronotum. 5. Krisna strigicollis Spin., wing. 6. Krisna strigicollis Spin., male; head, pronotum, and scutellum. 7. Krisna mutrii sp. nov., head and pronotum. 8. Krisna strigicollis Spin., female; head and pronotum. PLATE II Fic. 1. Krisna colorata sp. nov., head and pronotum; head slightly separ- ated from pronotum in this specimen. 2. Krisna simillima sp. nov., head and pronotum. 3. Krisna magna sp. nov., female; head and pronotum. 4. Krisna penangensis sp. nov., head and pronotum. 5. Krisna magna sp. nov., male; head and pronotum; head slightly separated from pronotum in this specimen. 6. Krisna olivascens sp. nov., head and pronotum. 7. Krisna strigicollis Spin., lateral view of male genitalia. 8. Krisna minima sp. nov., male genitalia. 9. Krisna strigicollis Spin., male genitalia. 0. Krisna nigrifrons sp. nov., male genitalia. PLATE III Fic. 1. Krisna magna sp. nov., male; face. 2. Krisna nigrifrons sp. nov., face. 3. Krisna minima sp. nov., face. 4. Krisna magna sp. nov., female; face. 5. Krisna olivascens sp. nov., face. 6. Krisna minima sp. nov., head and pronotum, side view. 7. Krisna colorata sp. nov., face. 8. Krisna penangensis sp. nov., face. 9. Krisna olivascens sp. nov., head and pronotum, side view. 10. Krisna strigicollis Spin., face. 11. Krisna simillima sp. nov., face. 12. Krisna magna sp. nov., male; head and pronotum, side view. 13. Krisna muirii sp. nov., face. 14. Krisna magna sp. nov., female; head and pronotum, side view. 15. Krisna muirii sp. nov., head and pronotum, side view. PLATE IV 1. Krisna colorata sp. nov., female genitalia. 2. Krisna muirii sp. nov., head and pronotum, side view. 3. Krisna magna sp. nov., female genitalia. FIG. 219 220 Philippine Journal of Science 4. Krisna olivascens sp. nov., female genitalia. 5. Krisna penangensis sp. nov., female genitalia. 6. Krisna strigicollis Spin., female genitalia. 7. Krisna simillima sp. nov., female genitalia. 8. Krisna muirii sp. nov., female genitalia. 9. Krisna simillima sp. nov., head and pronotum, side view. 10. Krisna nigrifrons sp. nov., head and pronotum, side view. 11. Krisna penangensis sp. nov., head and pronotum, side view. 12. Krisna strigicollis Spin., head and pronotum, side view. PLATE V Fic. 1. Gessius malayensis sp. nov., tegmen. 2. Gessius malayensis sp. nov., head and pronotum. 8. Gessius pallidus sp. nov., head and pronotum. 4. Gessius pallidus sp. nov., female genitalia. 5. Gessius ‘malayensis sp. nov., female genitalia. 6. Gessius pallidus var. suffusus var. nov., head and pronotum, side view. 7. Gessius malayensis sp. nov., head and pronotum, side view. 8. Gessius pallidus var. suffusus var. nov., face. 9. Gessius pallidus sp. nov., head and pronotum, side view. 10. Gessius malayensis sp. nov., antenna. 11. Gessius pallidus var. suffusus var. nov., head and pronotum. 12. Gessius malayensis sp. nov., face. 13. Gessius pallidus sp. nov., face. BAKER: THE GENUS KRISNA.] [PuHuip. Journ. Sct., XV, No 6 AAA PLATE Il. SPECIES OF KRISNA. BAKER: THE GENUS KRISNA.] [Puiviep. Journ. Sci., XV, No. 2. PLATE lll. SPECIES OF KRISNA. BAKER: THE GENUS KRISNA.] (Purr, Journ. Scr, XV, No. 2. PLATE IV. SPECIES OF KRISNA. THE PHILIPPINE JOURNAL OF SCIENCE VoL. XV SEPTEMBER, 1919 No. 3 THE PHYSICO-CHEMICAL VALORATION OF TIKITIKI EXTRACT? By MARIANO V. DEL RoSARIO and JOAQUIN MARANON Of the University of the Philippines For a long time this country has been paying very dearly for exotic drugs. At present, however, there is the praiseworthy tendency to utilize our own resources in the preparation of the medicines used in combating the diseases peculiar to this part of the world. The extract of tikitiki,? although recognized and prescribed for a number of years, has not usually been prepared according to any definite, standard method. Hence, differences in composi- ' tion are found in the several brands that may be obtained in the market; that is, the products show different proportions as to their components. Consequently, we believe that our scientific authorities should adopt an official method for the preparation of this extract that would comply with established physical and chemical principles. The compound that we are considering as a standard is the one exemplified by the extract prepared by the Bureau of Science, Manila, according to the method of Chamber- lain and Vedder,*® which is the one used in this laboratory. No definite conclusions leading to the formulation of a standard extract of this drug can be drawn from the results obtained in * From the laboratory of pharmaceutical chemistry, School of Pharmacy, University of the Philippines. * Tikitiki is the Philippine word meaning rice polishings. * Chamberlain, Weston P., and Vedder, Edward B., The cure of infantile beriberi by the administration to the infant of an extract of rice polish- ings, and the bearing thereof on the etiology of beriberi, Bull. Manila Med. Soc. 4 (1912) 26. 167148 291 999 Philippine Journal of Science 1919 our analytical work on the different brands of tikitiki extract as prepared in local drug stores and private laboratories. However, inspection of the results shows that extracts 1, 2, and 3 are of approximately equal concentration, while extracts 4 and 5 occur as much lower concentrations, this being shown by their specific gravity and the proportions of total solids they contain. The results of our analyses are summarized in Table I. TABLE I.—Results of analyses of samples of tikitiki extract. E Bureau of Science, Extract 2. | Extract3. | Extract4. | Extract 5. : extract 1. Temperature -__.__-... oC..| 28 to25 26 to 25 30.3 to 30.1 | 28.5 to 30.1 | 29.3 to 29.6 Specific gravity.......----- 1. 2800 1, 2960 1. 2878 1, 2254 0. 9572 Total solids-__-... per cent-.- 53. 23 56. 20 50.33 25. 53 5.40 Reducing sugar (by direct reduction) -.._- per cent... 23.70 20. 80 24. 96 15, 27 1.39 Disaccharides (saccharose . by inversion) ._per cent. 2.56 0.30 0.65 0.57 0.17 Mineral residue_ -__. ro ee 4.40 3.32 4.87 0.76 0.19 Total nitrogen by Kjel- Gel. ste: per cent__ 2.008 2.300 1. 640 0. 856 0.140 Phosphorus as P205_do.____ 1.70 0.375 2.14 0. 24 0.165 Sulphur (total) _-._. Gos) 0.26 0. 109 pie § 0. 059 0.001 Amylolytic enzyme -__.-_.- negative negative negative negative | negative A striking fact in this table is that extract 3, which has a higher specific gravity than extract 1, has however a lower per- centage of total solids than that extract. This fact might be accounted for by the different methods used in removing the alcohol from the extract and by the extent of its dilution with water so as to obtain it at the desired concentration. Further- more, we have the anomalous condition shown in extract 5, in which a sample with 5.4 per cent of total solids shows, at 29.3° to 29.6° C., a specific gravity of 0.9572. This condition might be explained by the presence of an unremoved residuum of the alcohol that was used as a solvent, The determination of total solids in vacuum is undoubtedly the most accurate method for collecting such data; but, regret- tably, our electric pump was not in working order at the time we undertook this investigation and we were forced, therefore, to employ other methods to obtain our results. The amounts of reducing sugar and disaccharides are practi- cally proportional to the total solids. The proportion between them, however, is variable and depends upon the degree of their hydrolyzation. In extract 2, which was purposely hydrolyzed, the proportion of disaccharides is very small. ——oe Pee SF xv,8 Del Rosario and Maranon: Valoration of Tikitiki 223 It is very remarkable that in the majority of the samples the proportion of nitrogen is higher than that of phosphorus. Ex- tracts 3 and 5, however, show the reverse condition. The amount of phosphorus in extract 3 is one and one-fourth as great as the nitrogen found in the same sample, and compared with the phos- phorus of extract 2 it is six times as high, while the standard sample of the Bureau of Science is one-fourth lower in phospho- rus than extract 3. We have obtained uniformly negative results in our attempts ‘to demonstrate the amylolytic enzyme. We assume that the reason for this is that the treatment of the extract with alcohol renders the enzyme inactive. The figures we have obtained must be considered as being merely approximate, for it is quite apparent that analyses of the different samples of the same manufacturer, prepared at different times, will of necessity yield results that will vary to some extent. This is not surprising in view of the lack of a standard method for the preparation of this drug, to which we already have called attention. In fact, even though the same general procedure were followed, any qualitative variation in the raw material used, the concentration of the alcoholic solvent, the time during which the solvent and drug are in contact, the, temperature at which extraction takes place, and the time over which evaporation is allowed to proceed when the vacuum is not employed—all these are factors that must be taken into account in the determination of the percentage composition of the extract. The above discussion regarding tikitiki extract has been ad- vanced from the viewpoint of the chemist; but it should be well borne in mind that systematic clinical studies by the physician must be carried out and carefully interpreted before determina- tion can be made of the most desirable composition and the ullimate standardization of this drug, now so generally admin- istered in the treatment of endemic neuritis. ADDITIONAL NOTES ON THE KWANGTUNG FLORA By E. D. MERRILL Botanist, Bureau of Science, Manila This third paper on the flora of Kwangtung Province, China, ig essentially like its predecessors.t In it have been included the descriptions of fifteen presumably new species, and records of thirty-four previously described forms either new to China or new to Kwangtung Province. The material on which these data are based has been collected by Mr. C. O. Levine and Mr. G. W. Groff, of the Canton Christian College, ably assisted by their Chinese collector To Kang P’eng. In this paper have also been included some necessary changes in nomenclature for Chinese species occasioned by a recent pre- liminary study of Loureiro’s Flora Cochinchinensis. This work was published in 1790. It contains the descriptions of one thousand two hundred ninety-four species and a few varieties, of which about six hundred thirty were described as new, the others ascribed to binomials established by Linnaeus. A pre- liminary examination of Loureiro’s work shows that he fre- quently described some species twice, or sometimes even three or four times, under the same or under different generic names. In nearly one hundred cases reductions have been made, so that the total number of distinct species described by Loureiro is probably less than one thousand two hundred. In his interpre- tation of Linnean species he made numerous and frequently grave errors, and in about three hundred seventy cases, or about 56 per cent, he was wrong in the interpretation of such species. Loureiro described as new a total of one hundred eighty-five gen- era of which forty-two are generally adopted as valid ones or, if strict priority be followed, eliminating those names excluded by the lists of nomina conservanda adopted by the Vienna and the Brussels Botanical Congresses, fifty should be adopted. Fourteen of Loureiro’s new genera have never been satisfacto- rily reduced and remain as doubtful ones. * Merrill, E. D., Notes on the flora of Kwangtung Province, China, Philip. Journ. Sci. 12 (1917) Bot. 99-111; Notes on the flora of Loh Fau Mountain, Kwangtung Province, China, op. cit. 13 (1918) Bot. 123-161. 225 226 Philippine Journal of Science 1919 The title “Flora Cochinchinensis” is somewhat misleading, al- though more species were described from Cochin China than from any other single region. The geographic sources of his material are as follows: From Cochin China alone, about 697; from China alone, about 254; from both Cochin China and China, about 292; from tropical East Africa opposite Zanzibar, 29; from Mozam- bique, 9; from Zanzibar, 8; from India, 5; with 1 each from the Philippines, Sumatra, and the Malay Peninsula. Most of the Chinese material mentioned by Loureiro was from the immediate vicinity of Canton, where he resided for a period of three or four years. In 1774 Loureiro mentions having sent about sixty specimens with descriptions to Europe, and in 1779 another lot of two hundred thirty specimens. I have not as yet succeeded in lo- cating the first lot; the second shipment apparently consisted of those preserved in the herbarium of the British Museum, In the herbarium of the Paris Museum of Natural History ninety specimens from Loureiro are preserved, these having been se- cured by Geoffroy Saint-Hilaire in Lisbon in 1808. The bulk of Loureiro’s collection, however, was retained by him in Lisbon and has long since been destroyed.? Of the one thousand two hundred ninety-four species recog- nized by Loureiro not more than three hundred eighty are represented by known extant botanical materia] from his col- lections. In the much more numerous cases where Loureiro’s types are no longer extant, the species must be interpreted from the original descriptions and such other data as can be secured for the regions in which the specimens were collected. Considerable time was devoted to a preliminary study of Lou- reiro’s species, and a manuscript commentary on the Flora Co- chinchinensis was prepared by me and completed April 15, 1919. In this commentary Loureiro’s species, so far as possible, were reduced to a family arrangement following the Engler and Prantl system. An attempt was made to determine the oldest valid specific name for each species and the necessary synonymy was added to explain the acceptance of the specific name in each case; all local names cited by Loureiro were recorded; and a more or less critical discussion of each species was given, to- gether with the place of origin for each as cited by Loureiro. This manuscript was prepared in six copies, one of which is * De Candolle, A, La Phytographie (1880) 430. ‘> XV, 3 - Merrill: Notes on the Kwangtung Flora 227 retained in Manila; one was sent to Dr. A. Chevalier, director of the Institut Scientifique in Saigon, Indo-China; one to the Canton Christian College, Canton, China; one to the library of the United States Department of Agriculture, Washington; one to the British Museum (Natural History), London; and one to the Muséum d’Histoire Naturelle, Paris. The essential object in preparing this manuscript was to establish a basis for further investigations regarding the status of Loureiro’s species, as there still remains a high percentage of forms that are unintelligible from the descriptions alone. Eventually, when we shall have secured sufficient additional data to warrant doing so, it is hoped that a critical revision of Loureiro’s species may be prepared and published. The manuscript mentioned above summarizes in convenient form all the data at present available regarding the status of Loureiro’s species in relation to those described by other authors, and from it can be determined those that are definitely known and those that are of a more or less doubtful status. Fourteen genera have not definitely been placed, of which twelve have not been referred to their proper families, while about three hundred seventy-five species are still more or less doubtful, and cannot, from the description alone, be safely correlated with those of other authors. Of these doubtful species many cannot be re- ferred to their proper genera, and nearly fifty cannot be even referred to their proper families. Any great reduction in this rather high percentage of doubtful species cannot be expected until intensive field work shall have been prosecuted, with spe- cial reference to the problem, in southern China but more es- pecially in Cochin China, in the vicinity of Hue where Loureiro resided. GRAMINEAE AGROPYRON Gaertner AGROPYRON CILIARE (Trin.) Franchet in Nuov. Arch. Mus. Paris II 7 (1884) 151. Triticum ciliare Trin. in Bunge Enum. Pl. Chin. Bor. (1831) 72. Kwangtung Province, Shiuchow region, To Kang P’eng 2827, April, 1919. A species of wide distribution in northern and central China but no representative of the genus hitherto recorded from as far south as Kwangtung. n\5 XV \ i? 228 Philippine Journal of Science 1919 ARACEAE POTHOS Linnaeus POTHOS REPENS (Lour.) comb. nov. Flagellaria repens Lour. Fl. Cochinch. (1790) 212, ed. Willd. (1793) 263. Pothos loureirii Hook. & Arn. Bot. Beechy’s Voy, (1841) 220; Schott Aroid. 1 (1853) 28, t. 49, Prodr. (1860) 567; Engl. in DC. Monog. Phan. 2 (1879) 87; N. E. Br. in Journ. Linn. Soc. Bot. 36 (1903) 186, Curtis’s Bot. Mag. t. 7744; Engl. Pflanzenreich 21 (1905) 35, #.... ihe Pothos terminalis Hance in Ann. Sci. Nat. V 5 (1866) 247. Pothos microphyllus Schott Aroid. 1 (1853) 23, t. 40, f. B. This species is known from Kwangtung Province, Hainan, and Tonkin, and is represented by Levine 1989 from Teng Woo Mountain, Kwangtung Province, and by Hongkong Botanic Gar- den 2243 from Hainan. The oldest valid specific name is here adopted, as Loureiro’s description applies in all respects to the species as currently interpreted. ARISAEMA Martius ARISAEMA KWANGTUNGENSE sp. nov. § Pedatisecta, Herba circiter 40 em alta; foliis solitariis, pedatisectis, seg- mentis plerumque 11, omnibus sessilibus, lanceolatis, acuminatis, 6 ad 9 cm longis, 1 ad 1.8 em latis, basi angustatis; scapus 15 ad 18 cm longus; spathae tubus 5 cm longus, cylindricus, deorsum leviter angustatus; lamina ovata ad oblongo-ovata, 4 ad 5.5 cm longa, usque ad 3 cm lata, tenuiter acuminata, basi rotundata; spadicis appendix usque ad 10 cm longa. Corms unknown. Peduncular part of the plant 15 to 20 cm long, bearing one leaf and one inflorescence, the free petiolar part of the leaf 4 to 6 cm long. Leaves pedately lobed, the seg- ments mostly 11, membranaceous, lanceolate, slenderly acumi- nate, narrowed below, all sessile, 6 to 9 cm long, 1 to 1.8 cm wide. Scape 15 to 18 cm long, smooth, slender. Spathe pale green, its tube cylindric, slightly narrowed below, 5 cm long, the lamina membranaceous, ovate to oblong-ovate, 4 to 5.5 em long, up to 3 cm wide, base rounded, apex slenderly acuminate but not caudate. Spadices unisexual, the staminate ones about 3 mm in diameter, floriferous for a distance of about 3 cm, the appendage slender, smooth, up to 10 cm long, 1 to 1.3 mm in diameter. ene Cry i? “st” XV, 3 Merrill: Notes on the Kwangtung Flora 299 Kwangtung Province, Shiuchow region, Levine 3565, May 3, 1919, scattered along roads northeast of Nam Wa monastery. I am unable to refer this specimen to any previously described species. It is apparently allied to Arisaema japonicum Blume. LILIACEAE SCILLA Linnaeus SCILLA SINENSIS (Lour.) comb. nov. Ornithogallum sinense Lour. Fl. Cochinch. (1790) 206, Barnardia scilloides Lindl. in Bot. Reg. t. 1029. Scilla chinensis Benth. Fl]. Hongk. (1861) 373. Loureiro’s species is manifestly identical with the one cur- rently known as Scilla chinensis Benth., the latter having been published independently of Convallaria chinensis Osbeck * which is unquestionably a synonym although very imperfectly described. It is not uncommon in open grassy places in the vicinity of Can- ton whence Loureiro secured his material, and is represented by the following Kwangtung material: Merrill 10048, Levine 2270, 3421, the latter with the recorded local name shik sun tau. DISPORUM Salisbury DISPORUM CANTONIENSE (Lour.) comb. nov. Fritillaria cantoniensis Lour. Fl. Cochinch. (1790) 206. Disporum pullum Salisb. in Trans. Hort. Soc. 1 (1812) 331. Uvularia chinensis Ker in Curtis’s Bot. Mag. t. 916. Loureiro’s material was from plants cultivated in Canton, for which he cites the local name lin ni hoa. Wright‘ admits Fri- tillaria cantoniensis Lour. with the following comment: “A doubtful plant supposed by Gawler to be the same as Uvularia chinensis, which is now reduced to Disporum pullum Salisb.” Hooker f.,° under Disporum pullum Salish. states: “The type of _ this species is the Chinese Uvularia chinensis of the Botanical Magazine, a purple flowered plant hardly distinguishable from shortly spurred specimens of calcaratum.’”’ Loureiro’s descrip- tion is ample and applies unmistakably to Disporum; his specific name should be retained for the Chinese form currently referred to Disporum pullum Salisb. I am by no means certain that all the Indo-Malayan material currently referred to Disporum pullum Salish. is conspecific with the Chinese form. * Dagbok Ostind. Resa (1757) 220. * Journ. Linn. Soc. Bot. 26 (1903) 136. * Fl. Brit. Ind. 6 (1892) 260. 230 Philippine Journal of Science 1919 ASPARAGUS Tournefort ASPARAGUS COCHINCHINENSIS (Lour.) comb. nov. Melanthium cochinchinense Lour. Fl. Cochinch. (1790) 216. Asparagus lucidus Lindl. in Bot. Reg. (1844) Misc. 29. Loureiro observed this species in both China and Cochin China, and his description applies unmistakably to the well-known As- paragus lucidus Lindl., in spite of his description of the fruit as a capsule, this being a manifest error on the part of Loureiro. Kwangtung material representing the species, and for which the Cantonese name tin tung is recorded, corresponding to the form Loureiro cites, tien muen tum, is as follows: Merrill 10699, Le- wine 2174, Groff 2290, Dunn 6337. I have also examined the following specimens: Hongkong, Curran. Hainan, Miss Monin- ger 62. Formosa, Faurie 947, Bot. Inst. Tokyo 1570. ZINGIBERACEAE PHRYNIUM Willdenow PHRYNIUM PLACENTARIUM (Lour.) comb. nov. Phyllodes placentaria Lour. Fl. Cochinch. (1790) 138. Phrynium parviflorum Roxb. Fl. Ind. 1 (1820) 7; K. Schum, in Engl. Pflanzenreich 11 (1902) 54. Loureiro observed this species both in China and in Cochin China. The generic name Phyllodes antedates Phrynium, but the latter is retained in the list of nomina conservanda adopted by the Vienna Botanical Congress. K. Schumann cites Lourei- ro’s species as a doubtful synonym of Phrynium capitatum Willd., a purple-flowered species recorded from both China and Cochin China. From Loureiro’s description of the flowers of his species as white I am convinced that he had specimens of the species currently known as Phrynium parviflorum Roxb., of which I _ have excellent specimens from Cochin China, Pierre 626, and from Kwangtung Province, China, Levine 1873, Groff 2524, with the recorded Cantonese name chung ip, corresponding to Lou- reiro’s recorded Cantonese name toung iep. ORCHIDACEAE 4 SPIRANTHES L. C. Richard Ao SPIRANTHES ARISTOTELIA (Raeusch.) comb. nov. Epidendrum aristotelia Raeusch. Nomencl. ed. 3 (1797) 265. _Aristotelia spiralis Lour. Fl. Cochinch. (1790) 522. © Spiranthes australis Lindl. in Bot. Reg. (1824) sub t. 823, non Koch. Neottia sinensis Pers. Syn. 2 (1807) 511. Spiranthes sinensis Ames Orch. 2 (1908) 538. Aristotelia spiralis Lour., described by him as a new genus and species, was based on specimens from the vicinity of Canton. ayn 7 av XV,3 Merrill: Notes on the Kwangtung Flora 231 It is represented by the following Kwangtung material, Levine 1014, 2036, 2072, growing in open grasslands, flowering in April and May. The species is one of very wide distribution, extend- ing from India to Japan southward to New Zealand. Loureiro’s type is preserved in the herbarium of the Paris Museum of Natural History. MORACEAE ANTIARIS Leschenault ANTIARIS TOXICARIA (Pers.) Lesch. in Ann. Mus, Paris 17 (1810) 478. Ipo toxicaria Pers. Syn. 2 (1807) 566. Kwangtung Province, Kochow region, To Kang P’eng 2755, at Koon Shan temple, west of Kochow city, with the local name to yiuk. Widely distributed in the Indo-Malayan region, but no repre- sentative of the genus previously recorded from China. This is the “deadly upas tree,” its milky juice being widely used in the Indo-Malayan region for the purpose of poisoning spears and arrows. CUDRANIA Trécul CUDRANIA PUBESCENS Tréc. in Ann. Sci. Nat. Bot. ITI 8 (1847) 125. Kwangtung Province, Shiuchow region, Tan Ha Shan, To Kang P’eng 2847, April 25, 1919. Yunnan Province, China, Burma, Java; new to Kwangtung. FICUS Linnaeus FICUS VARIEGATA Blume Bijdr. (1825) 459; King in Ann. Bot. Gard. Calcutta 1 (1888) 169, t. 212. Kwangtung Province, Kochow region, Ngau Tsai Wan, To Kang P’eng 2743, February 25, 1919, with the local name tong kwo muk. This is the typical form of Blume’s species, with broad, undu- late-f¥ethed leaves and large fruits. It is certainly specifically distinct from Ficus chlorocarpa Benth. which King referred to Blume’s species as a variety. Ficus chlorocarpa Benth. is rep- resented by Merrill 10262 and Levine 1889 from Kwangtung Province. FICUS BENJAMINA Linn. Mant. 1 (1767) 129; King in Ann. Bot. Gard. Calcutta 1 (1887) 48, t. 52. Kwangtung Province, Kochow region, Tai Shek Ling, To Kang P’eng 2663, March 18, 1919, with the local name sai ip yung. Wild in various parts of the Malayan region and the Philip- qa 232 Philippine Journal of Science 1919 pines, frequently planted in other parts of tropical Asia. Not before recorded from China proper, although known from Hainan. LORANTHACEAE LORANTHUS Linnaeus LORANTHUS PENTANDRUS Linn. Mant. 1 (1767) 63; Blume FI. Jav. Loranth. (1828) 38, t. 10; Hook. f. Fl. Brit. Ind. 5 (1886) 216. Kwangtung Province, Sai Sha, Sz Ooi, Groff 2403, April 24, 1918, This species has not previously been reported from China; the specimens agree closely with the descriptions and with our rather full series of Malayan specimens. LORANTHUS PARASITICUS (Linn.) comb. nov. Scurrula parasitica Linn. Sp. Pl. (1753) 110, excl. syn. Camell et Petiver. Loranthus scurrula Linn. Sp. Pl. ed. 2 (1762) 472, non auct. plur. Loranthus estipitatus Stapf. (p. p.) in Trans. Linn. Soc. Bot. 4 (1894) 221; Forbes & Hemsl. in Journ. Linn. Soc. Bot. 26 (1844) 405; Dunn & Tutcher Fl. Hongk. Kwangtung (1912) 229 (as to the Chinese plant). Kwangtung Province, vicinity of Canton, Merrill 9987, Levine 1277, 1866, 1948, Groff 2314. Hongkong, Hongkong Herbarium 1282. The history of this species is as follows: The original binomial, Scurrula parasitica Linn., was manifestly based on a specimen, indicated by Linnaeus as originating in China, and in all prob- ability collected by Osbeck near Canton; to the species Linnaeus erroneously referred “Viscum vitict innascens Camell. luz. 3 n. 36. Pet. gaz. t. 23, f. 8.” Camell’s description was based on specimens from Naic, Cavite Province, Luzon, and is manifestly Loranthus philippensis Cham. & Schlecht., an endemic Philip- pine species. Petiver’s figure was in all probability based on Camell’s drawing or on specimens from him, and a copy of the figure kindly supplied by Mr. Oakes Ames shows that it also represents Loranthus philippensis Cham. & Schlecht. The Lin- nean description does not apply to Loranthus philippensis Cham. & Schlecht., but does apply word for word to the Chinese form currently referred to Loranthus estipitatus Stapf, which is the commonest species of Loranthus found in the vicinity of Canton, the region in which Osbeck botanized. It does not apply to Loranthus scurrula of modern authors. In the second edition of the Species Plantarum Linnaeus made Seurrula parasitica the XV, 8 Merrill: Notes on the Kwangtung Flora 233 basis of a new binomial, Loranthus scurrula Linn., but the earlier specific name should be adopted. Fragments of recently collected material, cited above, were sent to London and were critically compared by Doctor Stapf with the Linnean type and with the type of Loranthus estipitatus Stapf. He writes under date of May 31, 1918, that as a result of his comparison this Chinese form must be accepted as Loran- thus scurrula Linn. [= L. parasiticus (Linn.) Merr.], but that it is distinct from L. estipitatus Stapf. Loranthus chinensis DC is closely allied, if not identical. > LORANTHUS LEVINEI sp. nov. § Scurrula. Frutex parasiticus, ramis usque ad 60 cm longis, teretibus, glabris, minute lenticellatis, ramulis dense ferrugineo-puberulis ; foliis oblongis, coriaceis, obtusis, basi obtusis ad rotundatis, us- que ad 8 cm longis, supra glabris, nitidis, olivaceis, subtus den- sissime ferrugineo-puberulis vel tomentosis, nervis utrinque 4 ad 6, supra distinctis, subtus obscuris; floribus axillaribus, 4- meris, fasciculatis, circiter 2.5 cm longis, curvatis, densissime ferrugineo-tomentosis, pedunculis 1- ad 3-floris, 3 ad 4 mm longis. A parasitic shrub, the branches up to 60 cm in length, terete, glabrous, dark colored when dry, with scattered minute lenticels, the branchlets densely ferruginous-puberulent. Leaves opposite, subopposite, and alternate, oblong, coriaceous, 6 to 8 cm long, 2 to 3.5 em wide, obtuse, base obtuse to rounded, the upper surface glabrous, olivaceous, shining, the lower densely ferruginous- puberulent or tomentose; lateral nerves 4 to 6 on each side of the midrib, rather distinct on the upper surface, the reticulations very lax, on the lower surface obscure or even obsolete; petiole 4 to 10 mm long, densely ferruginous-puberulent. Flowers axillary, fas- cicled, 4-merous, curved, about 2.5 cm long, densely ferruginous- pubescent or tomentose, the peduncles 1- or 2-flowered, 3 to 4mm long, when 2-flowered the pedicels about 1 mm in length. Calyx oblong-ovoid, about 3 mm long, truncate, densely ferruginous- tomentose, subtended by a small, ovate bract. Corolla in bud curved, about 2.2 em long, in anthesis split down one side, ex- ternally densely ferruginous-tomentose, the tube about 1.6 cm long; lobes reflexed, about 6 mm long, somewhat spatulate. An- thers continuous with the filaments, about 2 mm long. Kwangtung Province, Lin District, Lo Chi Chui, Levine 3321, October 14, 1918, on trees along the river. This species belongs in the group with Loranthus parasiticus Merr. and is perhaps most closely allied to Loranthus yadoriki 934 Philippine Journal of Science 1919 Siebold. It may be the Kwangtung species recorded by Dunn & Tutcher as Loranthus scurrula Linn., but is certainly not the Linnean species. ELYTRANTHE Blume ELYTRANTHE FORDII (Hance) comb. nov. Loranthus fordii Hance in Journ. Bot. 23 (1885) 38. This species is apparently common in Kwangtung Province and is clearly an Elytranthe. It is represented by the following specimens: White Cloud hills, Levine 2076; Ting Woo Monastery, Levine 2025; Honam Island, Levine 1006, and North River, Tseng Uen, Levine 2390. The specimens have the following local names: shui chi kei shaang, koh muk kei shang, and wo ko. VISCUM Linnaeus VISCUM STIPITATUM Lecomte in Sargent Pl. Wils. 3 (1916) 319. Kwangtung Province, Lin District, Leung Kong Ngon, Levine 8465, October 28, 1918. Lecomte’s species was based on material from Yunnan Provy- ince, and Levine’s excellent specimen exactly matches the de- scription. In leaf characters the species is distinctly similar to Ginalloa. VISCUM ANGULATUM Heyne ex DC. Prodr. 4 (1830) 225; Hook. f. Fl. Brit. Ind. 5 (1886) 225. Kwangtung Province, Poon Yue District, Levine 3165, No- vember, 1918, with the local name kei shang. This species has not previously been recorded from China. The specimen is in fruit but agrees closely with Indian material representing Heyne’s species. India to Australia. RANUNCULACEAE RANUNCULUS Linnaeus RANUNCULUS DIFFUSUS DC. Prodr. 1 (1824) 38. Kwangtung Province, Shiuchow region, Tan Ha Shan, To Kang P’eng 2901, April, 1919, in grassy places. India to China, Java, and Sumatra; not previously recorded from Kwangtung Province. BERBERIDACEAE NANDINA Thunberg NANDINA DOMESTICA Thunb. FI. Jap. (1784) 9. Kwangtung Province, Shiuchow region, Fan Kwai No Shan, To Kang P’eng 2778, April 19, 1919, in forests. Widely distributed in Japan and China, but not previously found so far south as Kwangtung Province. XV,3 Merrill: Notes on the Kwangtung Flora 235 MENISPERMACEAE DIPLOCLISIA Miers DIPLOCLISIA CHINENSIS sp. nov. Frutex scandens, glaber; foliis late ovatis ad subreniformibus, 5 ad 10 cm longis, 7 ad 12 cm latis, apice acutis, basi 5-nerviis, late truncato-rotundatis ad leviter cordatis, petiolo 4 ad 7 cm longo; inflorescentiis axillaribus, pedunculatis, umbellato-cymo- sis, 1.5 ad 3 cm longis; floribus ¢ 6-meris, sepalis ellipticis ad obovatis, circiter 2.5 mm longis, lineolatis, petalis rnomboideis 1.5 mm longis, apice rotundatis, basi cuneatis, auriculis laterali- bus acutis, inflexis. Scandent, glabrous, the branches and branchlets usually red- dish brown, terete. Leaves chartaceous, olivaceous on both surfaces or somewhat glaucous beneath, broadly ovate to sub- reniform, 5 to 10 cm long, 7 to 12 cm wide, entire or the margin obscurely undulate, apex acute, base broadly truncate-rounded to shallowly cordate, 5-nerved; petioles 4 to 7 cm long. Inflores- cences axillary, solitary, umbellate-cymose, the peduncles 1 to 2 cm long, the flower-bearing portion less than 1 cm in diameter, the pedicels 2 to 4 mm long, lineolate. Petals 6, rhomboid, 1.5 mm long, apex rounded, base acute, the auricles lateral, acute, inflexed ; filaments 2 mm long. Kwangtung Province, Shiuchow region, Fan Kwai No Shan, To Kang P’eng 2764 (type), 2779 vp. p., April 19, 1919. This species is closely allied to Diploclisia affinis (Oliv.) Diels but differs in its larger leaves which are broadly ovate to sub- reniform, their bases truncate-rounded to shallowly cordate and not at all peltate. My specimen of No. 2779 cited above consists in part of this species and in part of Pericampylus glaucus (Lam.) Merr. LAURACEAE | LITSEA Lamarck LITSEA CUBEBA (Lour.) Pers. Syn. 2 (1807) 4. Laurus cubeba Lour. Fl. Cochinch. (1790) 252. Litsea piperita Juss. in Ann. Mus. Paris 6 (1805) 218. Persea cubeba Spreng. Syst. 2 (1825) 269. Daphnidium cubeba Nees Syst. Lour. (1836) 615. Tetranthera cubeba Meisn. in DC. Prodr. 15* (1864) 199. Litsea citrata Blume Bijdr. (1825) 595; Lecomte Fl. Gén. Indo-Chine 5 (1914) 138; Gamble in Journ. As. Soc. Beng. 75% (1912) 146. Tetranthera citrata Nees Syst. Lour. (1836) 560. Tetranthera polyantha Wall. Cat. (1830) No. 2538, nomen nudum, Nees in Wall. Pl. As. Rar. 2 (1831) 67, Syst. Lour. (1836) 545. Tetranthera floribunda Champ. in Hook. Kew Journ. Bot. 5 (1853) 199 Ai 236 Philippine Journal of Science 1919 Loureiro’s material was from Cochin China, but his species and the numerous synonyms based upon it have never been satis- factorily placed and it is not mentioned by Lecomte in his recent treatment of the Lauraceae of Indo China. Hemsley‘ states that he had seen only the fruit as it appears in commerce. It is evident from Loureiro’s other descriptions of species of Laurus that the statement that the leaves of Laurus cubeba were nerve- less was intended by him to imply that there were no longitudinal nerves as in Cinnamomum, for he placed all the species of Cinna- momum known to him under Laurus. The fruits are black, about the size, shape, and color of the fruits of black pep- per, as Loureiro notes, and like the leaves are very aromatic. The species extends from Central China to India southward to Java, and it is manifest that Loureiro’s specific name should be adopted for the species currently known as Litsea citrata Blume. I have examined the following material: China, Kwangtung Province, Merrill 10960, Levine & Groff 158, Levine 1398, 3063, 3248, with the local name tan shi keung, To Kang P’eng 2658, with the local name ts’ing tsz muk: India, Meebold 5532, Craib 255: Indo China, Bon 4278: Malay Penin- sula, Perak, Scortechini 270b, Haniff € McNur 2323: Java, Koor- ders 3173, 25602, 27798, 32911, 38125, Winckel 256. MACHILUS Nees MACHILUS LEVINEI sp. nov. Arbor parva, glaberrima; foliis oblongis, crasse coriaceis, us- que ad 18 cm longis, acuminatis, subtus glaucescentibus, nervis utrinque 15 ad 18, subtus distinctis; inflorescentiis terminalibus, ramis umbellato-fasciculatis, 2 ad 3.5 em longis; perianthii seg- mentis anguste oblongis, coriaceis, glabris, circiter 10 mm longis. A small, entirely glabrous tree, the branches and branchlets reddish brown, wrinkled when dry, not lenticellate. Leaves thickly coriaceous, oblong, 12 to 18 cm long, 3 to 4.5 cm wide, base obtuse to acute, apex shortly acuminate, the upper surface smooth, shining, rather pale when dry, the lower glaucous, the midrib impressed on the upper surface, very prominent and red- dish brown on the lower; lateral nerves 15 to 18 on each side of the midrib, slender, distinct beneath, curved-ascending at an angle of about 45 degrees; petioles reddish brown, about 2.5 em long. Inflorescences terminal, the primary branches about 8, reddish brown, umbellate-fascicled, 2 to 3.5 cm long, few-flowered. Perianth segments oblong or narrowly oblong, acute to obtuse, ‘ “Fl. Gén. Indo-Chine 5 (1914) 107-158. ‘Journ. Linn. Soc. Bot. 26 (1891) 380. not £10 XV,8 Merrill: Notes on the Kwangtung Flora 237 coriaceous, about 10 mm Jong and 3 mm wide, glabrous. Ovary glabrous. Immature fruit globose. Kwangtung Province, Teng Woo Mountain, Levine 2024, May 26, 1918, scattered on slopes, altitude about 300 meters. This species is well characterized by its oblong, thickly coria- ceous, rather numerously nerved leaves which are glaucous be- neath, and by its long, coriaceous, glabrous perianth segments. It is apparently as closely allied to Machilus phoenicis Dunn as to any other species, which, however, has smaller, differently shaped, fewer-nerved leaves and much shorter petioles. LINDERA Thunberg LINDERA SUBCAUDATA (Merr.) comb. nov. Neolitsea subcaudata Merr. in Philip. Journ. Sci. 13 (1918) Bot. 137. Additional material with staminate flowers representing this species shows it to be a Lindera, allied to L. strychnifolia (Meisn.) F.-Vill. It is now represented by Merrill 11016, Le- vine 1351, Groff 2463, To Kang P’eng 2707, the latter with stam- inate flowers bearing the local name heung kau shii. SAXIFRAGACEAE ANDROSACE Linnaeus ANDROSACE UMBELLATA (Lour.) comb. nov. Drosera umbellata Lour. Fl. Cochinch. (1790) 186. Androsace saxifragifolia Bunge in Mém. Acad. St. Pétersb. 2 (1836) 127; Pax & Knuth in Engl. Pflanzenreich 22 (1905) 179. Loureiro cites both Chinese and Cochinchinese names for this species, although he also states “Habitat in China.” It would seem that he observed the species in both regions. The species extends from India to Japan southward to Indo-China and north- ern Luzon and occurs at low altitudes both in Kwangtung Prov- ince, China, and in Indo China. The species is clearly no Drosera, but among all the species of plants known from southern China, Loureiro’s description applies only to Androsace saxifra- gifolia Bunge. There is, hence, no reason why Loureiro’s specific name should not be adopted for this well-known species. Plan- chon * has already indicated that Loureiro’s species might be an Androsace. ROSACEAE PYGEUM Gaertner PYGEUM TOPENGII sp. nov. § Sericophyllum. Arbor circiter 12 m alta, ramis glabris, ramulis ferrugineo- pubescentibus; foliis coriaceis, elliptico-ovatis, usque ad 9 cm *Ann. Sci. Nat. III 9 (1848) 304. 167148——-2 238 Philippine Journal of Science 1919 longis, breviter obtuse acuminatis, basi plerumque acutis, leviter inaequilateralibus, haud vel obscurissime glandulosis, supra in siccitate pallidis, nitidis, glabris, subtus ad costa nervisque lev- iter ciliatis, nervis utringue 6 vel 7, subtus perspicuis; infruc- tescentiis axillaribus, solitariis, depauperato-paniculatis, 4 ad 5: em longis; fructibus subreniformibus, glabris, minute apiculatis, circiter 8 mm longis, 10 ad 12 mm latis; seminibus solitariis, testa extus sericeis. A tree about 12 m high, the branches dark reddish brown, lenticellate, glabrous, the branchlets ferruginous-pubescent as are the petioles and apparently the inflorescences. Leaves co- riaceous, elliptic-ovate, 6 to 9 cm long, 3 to 5 cm wide, shortly and rather bluntly acuminate, base acute, usually somewhat in- equilateral, eglandular or the glands obscure and not at all pro- jecting, the upper surface, except in very young leaves, glabrous, pale and shining when dry, the lower somewhat ciliate on the midrib and nerves; lateral nerves 6 or 7 on each side of the midrib, prominent; petioles 5 to 7 mm long; stipules oblong, pubescent, deciduous, about 5 mm long. Infructescences axil- lary, solitary, from the branchlets below the leaves, 4 to 5 cm long, sparingly pubescent, usually with a single basal branch 1 to2 eminiength. Fruits subreniform, brown when dry, slightly apiculate, glabrous, about 8 mm long, 10 to 12 mm wide, their pedicels 3 mm long or less. Seeds solitary, the testa distinctly silky-villous. Kwangtung Province, Kochow region, Shek Kau Tong, To Kang P’eng 2750, March 5, 1919. Pygeum henryi Dunn is the only species of the genus definitely known from China, but the present one is entirely different. Pygeum latifolium Mig. is recorded from Hongkong by Hemsley but Miquel’s species is definitely known only from Java, having long been confused with a Philippine species. It is possible that the present species is identical with the Chinese form pre- viously referred to Pygeum latifolium Miq., but it is safely not Miquel’s species. SANGUISORBA Ruppius SANGUISORBA OFFICINALIS Linn. Sp. Pl. (1753) 116. Kwangtung Province, Lin District, Shan Mo Ling and Sing Tize Foo, Levine 3181, 3191, September 15, 1918, with the local name mar lin on. The genus is new to Kwangtung Province. There seems to be some difference of opinion among botanists as to the distri- bution of Chinese forms of this genus between Sanguisorba offi- ve re XV,3 Merrill: Notes on the Kwangtung Flora 239 cinalis Linn. and S. canadensis Linn., the present material matching specimens from various parts of China, some identified as S. officinalis Linn. and some as S. canadensis Linn. The Kwangtung material closely matches European material of S. officinalis Linn., and I assume this name to be correct for it. Sanguisorba formosana Hayata, as represented by Formosan material, Kawakami 93, is scarcely to be distinguished. POTENTILLA Linnaeus POTENTILLA DISCOLOR Bunge Enum. Pl. Chin. Bor. (1881) 25, Kwangtung Province, Shiuchow region, To Kang P’eng 2775, April 19, 1919. Widely distributed in northern and central China, extending to Formosa (a variety) ; not previously recorded from Kwang- tung Province. LEGUMINOSAE PITHECOLOBIUM Martius PITHECOLOBIUM TURGIDUM sp. nov. Arbor parva, partibus junioribus ferrugineo-pubescentibus ; foliis bipinnatis, pinnis 1-jugis, foliolis amplis, bijugis, membra- naceis vel chartaceis, glabris, nitidis, acuminatis, majoribus us- que ad 15 cm longis, oblongo-ovatis, basi acutis, nervis utrinque 6 ad 8, perspicuis; leguminis turgidis, rectis, dehiscentibus, oblongis, 7 ad 9 cm longis circiter 3 cm latis, et 1 ad 1.5 cm crassis, valvis coriaceis; seminibus ellipsoideis, 1.8 ad 2.5 cm longis, haud compressis. A small tree, 3 to 4 m high fide Levine, glabrous except the younger parts which are ferruginous- or castaneous-pubescent. Leaves bipinnate, pinnae 1-jugate, leaflets bijugate, the petiole 2 to 6 cm long, with a single large gland at the apex below the insertion of the single pair of terminal pinnae, the rachises of the pinnae 5 to 8 cm long, each bearing four large leaflets and usually with a terminal gland; leaflets membranaceous to char- taceous, oblong-ovate to elliptic-ovate, pale greenish and shining when dry, glabrous, base acute, equilateral, the apex acuminate, 9 to 15 cm long, 4 to 7 cm wide, the lateral nerves 6 to 8 on each side of the midrib prominent. Panicles pyramidal, in bud up to 14 cm long, the flowers in globose heads at the tips of the branchlets, pubescent. Pods oblong, turgid, dehiscent, 7 to 9 em long, about 3 cm wide, 1 to 1.5 em thick, each usually with about four seeds, the sutures not thickened, the valves coria- ceous, continuous, brown, smcoth. Seeds contiguous, ellipsoid, 240 Philippine Journal of Science 1919 not compressed, brown, smooth, shining, 1.8 to 2.5 cm long, persistent for a considerable period after the pod dehisces by the elongated, slender, more or less curved funiculus, the latter about 1 cm in length. Kwangtung Province, Teng Woo Mountain, Levine & Groff 86, November 18, 1916, with mature fruits, Levine 1976, April 26, 1918, from the same tree, with immature buds and fully mature seeds; along streams, altitude about 300 meters. This species, distributed as an Albizzia, like its congener, Pithecolobium balansae Oliv., and apparently P. attopenense Pierre (of which the fruits are unknown), is anomalous in Pi- thecolobium in its straight, turgid, not at all twisted or curved pods, and is equally anomalous in Albizzia. It is suspected that it may prove to represent a distinct generic type. PAHUDIA Miquel PAHUDIA XYLOCARPA Kurz Forest Fl. Brit. Burma 1 (1877) 413. Kwangtung Province, Kochow region, Sai Ngon, To Kang P’eng 2705, February 18, 1919, a single tree, said to have been grown from seeds secured in Burma. It is locally known as min ke, The specimen, which presents a mature pod and seeds, agrees closely with Kurz’s description, the species having been based on material originating near the Burmese border of Siam. CASSIA Linnaeus CASSIA FISTULA Linn. Sp. Pl. (1753) 877. Kwangtung Province, Tak Hing, Levine 3587, April, 1919, from an introduced and cultivated tree. A native of tropical Asia, pantropic in cultivation. ORMOSIA Jackson ORMOSIA HAINANENSIS Gagnep. in Not. Syst. 3 (1914) 81. Hainan, Hongkong herbarium 443! Miss Moninger 122!, in fruit: Kwangtung Province, Kochow region, Shek Kau T’ong, To Kang P’eng 2664, March 8, 1919, in forests. The pods, description from Miss Moninger’s Hainan specimen, contain from 1 to 4 seeds, and when more than 1-seeded are somewhat torulose. They are 2 to 4 cm long and 1 to 1.5 em wide, much thickened, glabrous, brown when dry, the valves thickened, somewhat woody, irregularly twisted after dehiscence. Seeds red, dangling from the pod after dehiscence, not arillate, 15 to18 mm long. When but 1-seeded the pods are usually con- s 4 “ XV, 3 Merrill: Notes on the Kwangtung Flora PAL spicuously stipitate, but the pseudostalk presents several un- developed ovules. The species seems to be allied to Ormosia fordiana Oliv.°® PTEROLOBIUM R. Brown PTEROLOBIUM ROSTHORNI! Harms in Engl. Bot. Jahrb. 29 (1900) 410. Kwangtung Province, Lin District, Pak hill, Levine 3208, Oc- tober 21, 1918, with the local name ye tau. This is the second species of the genus to be found in Kwang- tung Province. The material agrees very closely with the orig- inal description, which, however, is rather short and imperfect. It has not otherwise been reported except by the original col- lections in southern Szechuen. The Kwangtung material is in fruit, the wings being 1.2 to 1.5 em wide, brown and shining when dry, and apiculate-acuminate by the nearly straight upper suture which is slightly produced at the tip. DERRIS Loureiro DERRIS ELEGANS (Grah.) Benth. in Mig. Pl. Jungh. (1852) 252, Journ. Linn. Soc. Bot. 4 (1860) Suppl. 109; Baker in Hook. f. Fl. Brit. Ind. 2 (1878) 252. Pongamia elegans Grah. in Wall. Cat. (1832) No. 7540, nomen nudum. Kwangtung Province, Shai Chiu Mountain, Levine 2074, May 4, 1918, with the local name kau ngar fa. This species has not previously been reported from China, but the flowering specimen cited above is an excellent match for our large series of Philippine specimens representing it; fruiting specimens of the Chinese form are desirable to verify the correctness of the determination. Tenasserim, Andaman Islands, Malay Peninsula, Sumatra, and the Philippines. DERRIS TRIFOLIATA Lour. Fl. Cochinch. (1790) 433. The genus Derris was based by Loureiro on two species. The first, D. pinnata, the type of which is preserved in the herbarium of the British Museum, is Dalbergia pinnata (Lour.) Prain, a species of wide distribution in the Indo-Malayan region more commonly known as Dalbergia tamarindifolia Roxb. The second species described by Loureiro, D. trifoliata, was based on speci- mens from the vicinity of Canton, and the type is preserved in the herbarium of the Paris Museum of Natural History. I am of the opinion that this species should be interpreted as the type of the genus Derris. The species by many authors has been reduced to Derris uliginosa (Roxb.) Benth., and a recent critical °In Hook. Ic. IV 5 (1895) t. 2422. 242 Philippine Journal of Science 1919 examination of the type by Doctor Gagnepain shows that it is identical with Roxburgh’s species. Prain'? thought that Lou- reiro’s description did not apply sufficiently closely to Derris uliginosa to warrant reducing Loureiro’s species to the latter. The examination of the type by Doctor Gagnepain, however, definitely settles this matter; and Loureiro’s name, being the older, should be retained for this very common, characteristic, and widely distributed species. It occurs typically along the margins of tidal streams more or less subject to the influence of brackish or salt water, from tropical East Africa through India to southern China and Formosa, southward through Ma- laya to tropical Australia and Polynesia. DUNBARIA Wight and Arnott DUNBARIA ROTUNDIFOLIA (Lour.) comb. nov. Indigofera rotundifolia Lour. Fl. Cochinch. (1790) 458. Dolichos conspersus Grah. in Wall. Cat. (1831-32) No. 3342, nomen nudum. Dunbaria conspersa Benth..in Miq. Pl. Jungh. (1852) 242. Dunbaria punctata Benth. 1. c. Dolichos punctatus Wight & Arn. Prodr. (1834) 237. Loureiro’s type was from the vicinity of Canton, and his description applies closely to the species currently known as Dunbaria conspersa Benth. except that the pods have more than two seeds. No other leguminous species known from Kwang- tung agrees at all with Loureiro’s description. I have examined the following Kwangtung specimens: Merrill 10146, Levine 1111, 3345. Loureiro records the Cantonese name as o tam sin; that recorded on one of Levine’s specimens is chin tang, not very different from Loureiro’s name if the words be reversed. MUCUNA Adanson MUCUNA COCHINCHINENSIS (Lour.) A. Chev. in Bull. Agr. Inst. Sci. Saigon 1 (1919) 91. Marcanthus cochinchinensis Lour. Fl. Cochinch. (1790) 461. Carpopogon niveum Roxb. Fl. Ind. ed. 2, 3 (1832) 385. Mucuna nivea Wight & Arn. Prodr. (1834) 255. Stizolobium niveum O. Kuntze Rev. Gen. Pl. (1891) 207. Kwangtung Province, Lin District, Levine 3233, with the local name kaw chau tau tang. Loureiro’s material was from Cochin China, undoubtedly from the vicinity of Hue where he resided most of the time while “Journ. As. Soc. Beng. 66* (1898) 458. 3) XV,3 Merril: Notes on the Kwangtung Flora 243 in Cochin China. His description applies unmistakably to the widely distributed and cultivated species currently known as Mucuna nivea Wight & Arn., and his specific name will replace that based on Roxburgh’s binomial. Loureiro’s description of the pods was apparently based on fresh rather than on dried material. Loureiro resided at Hue, and a mature pod secured from this locality under the local name cited by Loureiro, sub- mitted to me by Dr. A. Chevalier, is identical with Mucuna nivea Wight & Arn. Prof. C. V. Piper informs me that he examined Loureiro’s type in the herbarium of the British Museum in 1912, a leaf specimen only, making the note at that time that it might be any of the species allied to Mucuna nivea, but that it probably represented the latter species. POLYGALACEAE POLYGALA Linnaeus POLYGALA TENUIFOLIA Willd. Sp. Pl. 3 (1800) 879. Kwangtung Province, Shiuchow region, To Kang P’eng 2809, 2770, 2903, in thickets and forests. In China previously recorded from Chihli, Shingking, and Shantung, but not previously reported from southern China. The specimens cited above agree closely with material from Chihli, differing chiefly in some of the leaves being broader than in the northern form, EUPHORBIACEAE EUPHORBIA Linnaeus EUPHORBIA ESULA Linn. Sp. Pl. (1753) 461; Boiss. in DC. Prodr. 15” (1862) 160; Forbes & Hemsl. in Journ. Linn. Soc. Bot. 26 (1894) 412. Kwangtung Province, North River, Fu Ok, Groff 2279, March, 1918. Widely distributed in Asia, but not previously reported from southern China. EXCOECARIA Linnaeus EXCOECARIA COCHINCHINENSIS Lour. Fl. Cochinch. (1790) 612; Muell.-Arg. in DC. Prodr. 15” (1866) 1215. Kwangtung Province, cultivated at the Canton Christian College, Groff 2963. This is the typical form with red leaves, cultivated for ornamental purposes; namely, =e Excoecaria bicolor Hassk. 244 Philippine Journal of Science 1919 EXCOECARIA COCHINCHINENSIS Lour. Fl. Cochinch. (1790) 612, var. VIRIDIS (Pax & K. Hoffm.). Excoecaria bicolor Hassk. Retzia 1 (1855) 158, var. viridis Pax & K. Hoffm. in Engl. Pflanzenreich 52 (1912) 159. Kwangtung Province, Kochow region, Kwong T’am, To Kang Peng, 2672, March 22, 1919. I believe the cultivated form with colored leaves described by Loureiro as Excoecaria cochinchinensis to be identical with Excoecaria bicolor Hassk., which being the case Loureiro’s name should be retained. The type of Excoecaria bicolor Hassk. var. viridis Pax & K. Hoffm. was from Cochin China, and the description agrees entirely with the specimen cited above. The species is new to China. ALCHORNEA Swartz ALCHORNEA RUGOSA (Lour.) Muell.-Arg. in Linnaea 34 (1865) 170. Alchornea hainanensis Pax & K. Hoffm. in Engl. Pflanzenreich 63 (1914) 242! Cladodes rugosa Lour. Fl. Cochinch. (1790) 574. Kwangtung Province, Kochow region, Kwanshan temple, To Kang P‘eng 2662, March 18, 1919. This is the first record of the species from China proper, al- though it had previously been recorded from Hainan Island. Unless Alchornea rugosa (Lour.) Muell.-Arg. is variable in the number of its stamens, it would seem that Pax and Hoffmann are wrong in their interpretation of Alchornea rugosa, and that they described as a new species the typical form as described by Loureiro under Cladodes rugosa. Loureiro describes his species as having eight stamens, the only character depended | upon by Pax and Hoffmann in separating the Hainan form from Alchornea rugosa. In Alchornea hainanensis the staminate flowers have eight stamens, as does the Kwangtung specimen cited above, thus agreeing with Loureiro’s original description of Cladodes rugosa. J. J. Smith states that in all staminate flowers of the Javan form referred by him to Alchornea rugosa the number of stamens was four; Pax and Hoffmann state stamens 4, rarely 5 or 6. The actual specimens so closely re- semble each other that I strongly suspect that the species has a variable number of stamens, 4 to 8. Should this not prove to be the case, then Loureiro’s specific name will have to be re- tained for the form characterized by Pax and Hoffmann as Alchornea hainanensis, while for the common Malayan form the name Alchornea javanensis (Blume) Mouell.-Arg. will have sd ee eee » XV,3 Merrill: Notes on the Kwangtung Flora PAD to be revived, or the still earlier one, Croton apetalum Blume, transferred to Alchornea. Itis to be noted that Pax and Hoff- mann saw no Cochin China material representing Loureiro’s species. ANACARDIACEAE POUPARTIA Commerson POUPARTIA CHINENSIS sp. nov. Arbor circiter 8 m alta, inflorescentiis exceptis glabra; foliis 20 ad 30 em longis, foliolis 11 ad 15, membranaceis ad char- taceis, oblongo-lanceolatis, 6 ad 9 cm longis, acuminatis, leviter inaequilateralibus, glabris vel junioribus subtus in axillis leviter barbatis, nervis utrinque 9 ad 12; petiolulis circiter 2 mm longis; inflorescentiis terminalis, amplis, circiter 30 cm longis, ramis inferioribus usque ad 15 em longis, cinereo-pubescentibus; flo- ribus ¢ 5-meris, calyces circiter 1.5 mm diametro; petalis oblongis, 2.5 mm longis, reflexis; staminibus 10, filamentis 2 ad 2.3 mm longis; ovario glabro, 4- ad 5-locellato. A tree about 8 m high, the inflorescences more or less cine- _reous-pubescent. Ultimate branches about 5 mm in diameter, smooth, glabrous, terete. Leaves 20 to 30 cm long, the rachis sparingly pubescent; leaflets 11 to 15, membranaceous to char- taceous, oblong-lanceolate, 6 to 9 cm long, 1.5 to 2.5 em wide, somewhat inequilateral at the base, apex acuminate, glabrous, or the younger ones sparingly bearded in the axils on the lower surface; lateral nerves 9 to 12 on each side of the midrib, slender; petiolules about 2 mm long. Inflorescence a terminal leafy panicle about 30 cm in length, the lower branches up to 15 cm long, subtended by normal but usually reduced leaves, the upper 15 to 20 cm of the panicle leafless; the branches, branchlets, and pedicels cinereous-pubescent. Flowers numerous, white, pistillate and staminate ones in the same inflorescences. Calyx of the staminate flowers about 1.5 mm in diameter, the lobes 5, ovate, acute or obtuse, about 0.5 mm long. Petals oblong, 2.55 mm long, their margins somewhat inflexed, reflexed in an- thesis, nerveless. Stamens 10, their filaments filiform, 2 to 2.3 mm long. Pistillate flowers similar to the staminate ones. Ovary glabrous, 4- or 5-celled; styles 4 or 5, about 1 mm long. Kwangtung Province, Honam Island, on the campus of Can- ton Christian College, Levine 3521, May 138, 1919. This is the second species of the genus to be found in China, differing radically from Poupartia fordit Hemsl. in its very much larger, terminal, distinctly pubescent inflorescences, the 246 Philippine Journal of Science 1919 staminate and pistillate flowers borne in the same infiorescences ; much smaller pistillate flowers; much more numerous leaflets; and shorter petiolules. CELASTRACEAE CELASTRUS Linnaeus CELASTRUS HOOKERI Prain in Journ. As. Soc. Beng. 72? (1904) 197; Rehd. & Wils. in Sargent Pl. Wils. 2 (1915) 352. Kwangtung Province, North River and Shiuchow regions, Groff 2300, 2294, To Kang P’eng 2875. India; previously recorded from China from Yunnan, Szech’- uan, and Fokien, but not before reported from Kwangtung Province. SAPINDACEAE KOELREUTERIA Lakman KOELREUTERIA BIPINNATA Franch. in Bull. Soc. Bot. France 33 (1886) 463, Pl. Delavay. (1889) 143, t. 29, 30. Kwangtung Province, Ying Tak District, Levine 3484, Decem- ber 9, 1918. This species is new to Kwangtung Province and this record represents a considerable southward extension of range for it. The specimen is in fruit and agrees closely with our rather full series of specimens from Yunnan Province. ELAEOCARPACEAE = ELAEOCARPUS Linnaeus ELAEOCARPUS DUBIUS A. DC. in Bull. Herb. Boiss. Il 2 (1903) 366; Gagnep. in Lecomte Fl. Gén. Indo-Chine 1 (1910) 572. Kwangtung Province, Kochow region, Shek Kau Tong, To Kang P’eng 2686, March 5, 1919, in forests. The specimen agrees in all respects with the descriptions of this species and with Bon 4298!, 2671! from Tonkin. Previously known only from Tonkin. MALVACEAE HIBISCUS Linnaeus HIBISCUS SURATTENSIS Linn. Sp. Pl. (1753) 696. Kwangtung Province, Kochow region, Fat Tsz Ling, To Kang P’eng 2730, February, 1919, along roadsides. This widely distributed Indo-Malayan species has been re- _ corded from Hainan, but I can find no record for it from China proper. wv wf » XV, 3 Merrill: Notes on the Kwangtung Flora 247 THEACEAE EURYA Thunberg EURYA GROFFIl sp. nov. Frutex vel arbor parva; ramis teretibus, glabris, ramulis pi- losis; foliis lanceolatis, chartaceis vel subcoriaceis, usque ad 6 cm longis et 1.2 cm latis, nitidis, supra glabris, subtus pilosis, apice tenuiter acuminatis, basi obtusis, plerumque leviter inae- quilateralibus, costa supra impressa, subtus cum venis promi- nulis; fructibus axillaribus, fasciculatis, globosis vel ovoideis, glabris, stylis connatis; sepalis elliptico-ovatis, exterioribus lev- iter pilosis, coriaceis, 1.5 ad 2 mm longis. A shrub or a small tree, the branchlets rather densely pilose. Branches terete, glabrous, dark reddish brown. Leaves numerous, lanceolate, chartaceous to subcoriaceous, greenish olivaceous and shining when dry, 3.5 to 6 cm long, 8 to 12 mm wide, the mar- gins denticulate, the upper surface glabrous, the lower pilose, the midrib above impressed, prominent beneath, the lateral nerves obsolete or subobsolete on the upper surface, distinct and some- what projecting on the lower surface, the apex slenderly acum- inate, the base obtuse and often minutely inequilaterally cordate, sessile or subsessile. Fruits axillary, glabrous, globose or ovoid, 3 to 3.5 mm in diameter, smooth, their pedicels 1 to 1.5 mm long; styles united for the lower 1 mm, the arms about 1 mm long. Sepals coriaceous, elliptic-ovate, 1.5 to 2 mm long, rounded, the outer ones somewhat pilose. Kwangtung Province, Tiu Kaan Shan, Tseng Uen, Groff 2378, March, 1918, on mountain sides. In vegetative characters and general appearance this species strongly resembles Hurya swinglei Merr., but differs radically in its entirely glabrous fruits and much shorter styles. From Eurya distichophylla Hemsl. it is readily distinguished by its slenderly acuminate leaves and the veins obsolete or nearly so on the upper surface and projecting on the lower surface. The fruits and staminate flowers of Hemsley’s species are as yet unknown. FLACOURTIACEAE XYLOSMA Forster f. XYLOSMA CONGESTUM (Lour.) comb. nov. Croton congestum Lour. Fl. Cochinch. (1790) 582, excl. deser. fruct. Xylosma racemosum Miq. Ann. Mus. Bot. Lugd.-Bat. 2 (1865-66) 155. Hisingera racemosa Sieb. & Zucc. Fl. Jap. Fam. Nat. 1 (1843) 169. Xylosma japonicum A. Gray in Mem. Amer. Acad. II 6* (1863) 381. Flacourtia chinensis Clos. in Ann. Sci. Nat. Bot. IV 8 (1857) 219. ee 248 Philippine Journal of Science 1919 This species is very common in thickets in the vicinity of Can- ton, Loureiro’s type having been from Canton. J. Mueller ‘ notes that Loureiro’s description of the flowers and of the infiorescen- ces does not conform to Croton; further it does not conform with the characters of any euphorbiaceous plant known from Kwangtung Province, but, with the exception of the fruit de- scription, agrees entirely with the species currently known as Xylosma racemosum Miq. It is clear that Loureiro either added the fruit description to make his species agree with the generic characters of Croton, or described the fruits from material originating from a species entirely unrelated to the flowering specimen described by him. J have examined the following specimens from Kwangtung Province, mostly from the imme- diate vicinity of Canton: Merrill 9850, 9993, Groff 2252, 2353, Levine 18, 171, 172, 177, 279, 365, 366, 371, 1749, 1809, 1829, 2084, 3261, 3341, To Kang P’eng 2727, 2737. The local names recorded are wu ying shue (vicinity of Canton), ch’ui tung ts’at (North River region), and chii nga lak shii (Kochow region). The name pa tau recorded by Loureiro should probably be ex- cluded as it is the same as the name recorded by him for Croton tigltum Linn. THYMELAEACEAE AQUILARIA Lamarck AQUILARIA SINENSIS (Lour.) comb. nov. Ophiospermum sinense Lour. Fl. Cochinch. (1790) 281. Aquilaria chinensis Spreng. Syst. 2 (1825) 356. Aquilaria grandiflora Benth. Fl. Hongk. (1861) 297. Loureiro cites no definite locality, but from the fact that he indicated the local name pa mou yong as Chinese it is clear that his material was from China, and in all probability from the vicinity of Canton. The description definitely applies to the species commonly known as Aquilaria grandiflora Benth., one that is not uncommon in the vicinity of Canton. Loureiro’s description of the perianth as 6-merous was apparently due to an error on his part, for he enumerates the stamens as five, thus indicating that he was describing a 5-merous flower. I have examined the following Kwangtung material: Groff 2487, Mer- rill 10962, Levine 996, 1400, 2070, from Tsangsheng, Wa Shau Toi, White Cloud Mountain, and Honam Island, the last two localities being in the immediate vicinity of Canton. Loureiro’s “DC. Prodr. 15* (1866) 696. \ ») %, q VW yo’ XV, 3 Merril: Notes on the Kwangtung Flora 249 type is preserved in the herbarium of the Paris Museum and is identical with the species as here interpreted. RHIZOPHORACEAE CARALLIA Roxburgh CARALLIA BRACHIATA (Lour.) comb. nov. Diatoma brachiata Lour. Fl. Cochinch. (1790) 296. 2'° Carallia lucida Roxb. Hort. Beng. (1814) 92, nomen nudum, Pi. Coro- mandel 3 (1819) 8, é. 211. Carallia integerrima DC. Prodr. 3 (1828) 33. Petalotoma brachiata DC. op. cit. 295. Kwangtung Province, White Cloud Mountain, near Canton, Levine 3386, local name nik nga tsai; Poon Yue District, Levine 3138; Kochow region, To Kang P’eng 2676, 2751, with the local name ngo shen muk. Loureiro’s generic name has priority over Carallia, but the latter is retained in the list of nomina conservanda adopted by the Vienna Botanical Congress. Petalotoma DC. is merely a new generic name for Diatoma. I believe Loureiro’s description applies to the common and widely distributed Indo-Malayan species commonly known as Carallia lucida Roxb. and as C. in- tegerrima DC., the last two certainly being conspecific. MYRTACEAE EUGENIA Linnaeus EUGENIA BULLOCKI! Hance in Journ. Bot. 16 (1878) 227. Kwangtung Province, Kochow region, Shek Tan Kong, To Kang P’eng 2995, March, 1919. This specimen, although in fruit, agrees closely with the orig- inal description and with Hainan material collected by Miss Moninger, which I have identified with Hance’s species. It was previously known only from Hainan. ARALIACEAE ARALIA Linnaeus ARALIA SPINIFOLIA sp. nov. § Arborescentes. Frutex erectus, circiter 3 m altus, foliis inflorescentiisque per- spicue spinosis atque setosis; foliis magnis, bipinnatis, pinnis circiter 30 cm longis, 5- ad 9-foliolatis; foliolis oblongo-ovatis, usque ad 12 cm longis, membranaceis, acuminatis, brevissime petiolulatis, basi rotundatis, serratis, utrinque ad costa nervis- que parce spinosis atque setosis; inflorescentiis magnis, laxis, 250 Philippine Journal of Science 1919 umbellulis longe pedunculatis; fructibus ovoideis, 5 mm longis, glabris, 5-carinatis, perspicue 5-sulcatis, pedicellis 1 ad 1.5 cm longis. | An erect shrub about 3 m high. Leaves large, bipinnate, the rachis, partial rachises, and leaflets on both surfaces with scattered, long, slender, nearly straight spines and with more numerous, slender, spreading setae, the spines 3 to 10 mm long, the setae 1.5 to 3 mm in length. Pinnae 5- to 9-foliolate, about 30 cm long; leaflets oblong-ovate, membranaceous, dark brown or olivaceous when dry, the lower surface somewhat paler than the upper, 9 to 12 cm long, 4 to 6 cm wide, subsessile, base rounded, often slightly inequilateral, apex acuminate, margins serrate, the teeth apiculate, the spines few, scattered, and mostly confined to the midrib with a few on the primary nerves, fewer on the lower surface than on the upper, the setae more numerous and scattered all over the epidermis on both surfaces. Inflores- cences ample, lax, the ultimate branchlets up to 10 cm in length. Umbels subglobose, about 30 fruits in each, the pedicels 10 to 15 mm long, setose. Fruits ovoid, glabrous, about 5 mm long, prominently 5-keeled and deeply 5-sulcate, the depressions broad, rounded, Kwangtung Province, Chan Tung hill, Levine 3242, October 30, 1918, with the local name lak cha tsui. This species is readily distinguished among its congeners by its spiny and setose rachises, partial rachises, leaflets, and in- fiorescences. PRIMULACEAE LYSIMACHIA Tournefort ' LYSIMACHIA CANDIDA Lindl. var. DEPAUPERATA var. nov. A typo differt planta multo minoribus 5 ad 8 cm altus, foliis oblanceolatis ad obovatis, 5 ad 15 cm latis, obtusis vel subacutis, pedicellis quam floribus haud longioribus. Kwangtung Province, Kochow region, Fung Mun, To Kang P’eng 2735, February 22, 1919, in fields. LYSIMACHIA ALFRED! Hance in Journ. Bot. 15 (1877) 356; Pax & Knuth in Engl. Pflanzenreich 22 (1905) 281. Kwangtung Province, Shiuchow region, To Kang P’eng 2858, Levine 3524, April, 1919, in woods. The identification has been made from the descriptions, the former number cited agreeing better with it than the latter, although differing in some details. The species has previously been known only from Fokien Province. XV, 3 Merrill: Notes on the Kwangtung Flora 951 * EBENACEAE DIOSPYROS Linnaeus DIOSPYROS SINENSIS Hemsl. in Journ. Linn. Soc. Bot. 26 (1889) 71. Kwangtung Province, Shiuchow region, To Kang P’eng 2785, 2788. Previously known from Szechuen and Hupeh Provinces, but not before reported from as far south as Kwangtung. SYMPLOCACEAE SYMPLOCOS Jacquin SYMPLOCOS FUSONII sp. nov. § Hopea, Lodhra. Frutex 2 ad 3 m altus, ramulis castaneo- vel purpureo-pubes- centibus, inflorescentiis cinereo-pubescentibus. Foliis coriaceis vel subcoriaceis, glabris, ellipticis ad oblongo-ellipticis, usque ad 5.5 em longis, nitidis, margine minute glanduloso-serrulatis vel crenulatis, deorsum integris, apice acutis vel brevissime acumi- natis, basi acutis, nervis utrinque 5 vel 6, anastomosantibus, costa supra haud impressa; racemis axillaribus, solitariis vel binis, paucifloris, 8 ad 10 mm longis, floribus omnibus pedicella- tis; fructibus oblongo-ovoideis, leviter adpresse pubescentibus, circiter 5 mm longis. Species S. anomalae affinis. A shrub 2 to 3 m high, the branches reddish brown, glabrous, the branchlets slender, somewhat angled, castaneous or purplish, appressed-pubescent with short hairs. Leaves coriaceous or sub- coriaceous, shining, glabrous, elliptic to oblong-elliptic, 3 to 5 cm long, 1.5 to 3 cm wide, the apex acute or shortly acuminate, base acute, margins minutely glandular-serrulate or crenulate, toward the base usually entire; lateral nerves 5 or 6 on each side of the midrib, slender, anastomosing, the reticulations lax, the midrib not impressed on the upper surface; petioles 2 to 3 mm long, glabrous or when young pubescent. Racemes axillary, solitary, or in pairs, short, few-flowered, appressed cinereous- pubescent, 8 to 10 mm long, simple, usually 5 to 10 flowers in a raceme. Flowers white, their pedicels 1.5 to 2.5 mm long, the apical bracteoles triangular-ovate, acute, about 1 mm long, some- what pubescent. Calyx pubescent, the tube short, the limb 2.5 mm in diameter, the lobes orbicular-reniform, spreading, rounded, slightly pubescent, about 1 mm long and 1.2 mm wide. Petals glabrous, oblong-elliptic, 3 mm long. Stamens indefinite, the filaments glabrous, 3 to 4 mm long, slightly united below, forming about 5 indistinct phalanges. Ovary 3-celled; style gla- ott yt 252 Philippine Journal of Science 1919 prous, 5mm long. Fruits sparingly appressed-pubescent, oblong- ovoid, terete, about 5 mm long. Kwangtung Province, Kochow region, Lai Tung and Shan Mi, To Kang P’eng 2714 (type), 2677, February and March, 1919, the former with the local name ye cha fa. This species, dedicated to Mr. C. G. Fuson, through whose interest it was possible to do some field work in the Kochow region, is distinctly allied to Symplocos anomala Brand of Yun- nan Province, from which it differs in its shorter, much less acuminate, fewer-nerved leaves, and somewhat longer, more numerously flowered racemes. SYMPLOCOS CHINENSIS (Lour.) Desvaux MS. in herb. Mus. Paris. comb. nov. Myrtus chinensis Lour. Fl. Cochinch. (1790) 313. Symplocos sinica Ker in Bot. Reg. 9 (1828) t. 710; Brand in Engl. Pflanzenreich 6 (1901) 34. Loureiro’s material was from the vicinity of Canton, and his description applies unmistakably to the species currently known as Symplocos sinica Ker. It is common on hills near Canton and is represented by the following Kwangtung material: Merrill 10725, Levine 15, 294, 687, 1787, 2372, 3198, 3403. The only recorded local name is hak tsz in the Lin District; Loureiro records the Cantonese name as tan quat xiong. Among the other new genera and species described by Loureiro Dicalyx cochin- chinensis Lour. is Symplocos cochinchinensis Moore; Decadia aluminosa Lour. is apparently identical with Symplocos spicata Roxb. ; Drupatris cochinchinensis Lour. is certainly a Symplocos; and Myrtus zeylanica Lour. (non Linn.) is probably a Symplocos. Loureiro’s type of Myrtus chinensis is preserved in the herba- rium of the Paris Museum of Natural History. LOGANIACEAE STRYCHNOS Linnaeus STRYCHNOS UMBELLATA (Lour.) comb. nov. Cissus umbellata Lour. Fl. Cochinch. (1790) 84. Strychnos paniculata Champ. in Hook. Kew Journ. Bot. 5 (1853) 56. Planchon,'? in excluding Loureiro’s species from the Vitaceae, suggests that it might be a Strychnos. I am of the opinion that this is the correct disposition of Cissus umbellata Lour., and * DC. Monog. Phan. 5 (1887) 626. “y XV,8 Merrill: Notes on the Kwangtung Flora 258 further that it is identical with Strychnos paniculata Champ., a species strongly characterized by its 4-merous flowers and the only representative of the genus known from the vicinity of Canton. LIGUSTRUM Linnaeus LIGUSTRUM GROFFIAE sp. nov. Frutex circiter 2 m altus, subtus foliis et ramulis et inflores- centiis perspicue molliterque ferrugineo-villosis; foliis oblongo- ovatis, usque ad 7 cm longis, acuminatis, basi acutis, supra olivaceis, nitidis, leviter pubescentibus, subtus pallidioribus, nervis utringue circiter 6, tenuibus; inflorescentiis axillaribus, paniculatis, 5 ad 6 cm lJongis, ramis inferioribus usque ad 3 cm longis; floribus numerosis, corolla 4 mm longa, glabra, tubo 2 mm longo. A shrub about 2 m high, the leaves on the lower surface and especially the branchlets and inflorescences densely and softly ferruginous-villous with spreading hairs. Branches terete, those up to 5 mm in diameter more or less pubescent, the ultimate branchlets about 2 mm in diameter. Leaves chartaceous, oblong- ovate, 5 to 7 cm long, 2.5 to 3 cm wide, apex acuminate, base acute, the upper surface olivaceous, shining, somewhat pubescent, the midrib and nerves impressed, the lower surface paler, softly villous; lateral nerves about 6 on each side of the midrib, slender; petioles 4 to 5 mm long, ferruginous-villous. Panicles axillary, 5 to 6 cm long, many-flowered, the branches usually spreading, the lower ones up to 3 cm in length. Flowers white, fragrant, their pedicels 1 to 2 mm long, glabrous; bracteoles oblong- lanceolate, somewhat acuminate, about 1 mm long. Calyx some- what cup-shaped, glabrous, 1 to 1.2 mm long, shallowly 4-toothed. Corolla 4 mm long, the tube 2 mm long, the lobes somewhat elliptic, obtuse. Stamens exserted. Kwangtung Province, Shiuchow region, Tan Ha Shan, To Kang P’eng 2820, April 24, 1919, with the local name mo ch’ung shii. Possibly referable to this species is No. 2795 of the same col- lection from the same locality with the local name sha yeuk shu; this specimen, however, has much smaller leaves and much-fewer flowered inflorescences than the type. This species is well characterized by its dense ferruginous- villous indumentum. It is dedicated to Miss Elizabeth H. Groff, through whose interest it was possible to have collections made in the Shiuchow region. 1671483 254 Philippine Journal of Science 1919 ASCLEPIADACEAE CRYPTOLEPIS R. Brown CRYPTOLEPIS SINENSIS (Lour.) comb. nov. Pergularia sinensis Lour, Fl. Cochinch. (1790) 169. Emericia sinensis Roem. & Schultes Syst. 4 (1819) 402. Pergularia chinensis Spreng. Syst. 1 (1825) 836. Vallaris sinensis G. Don Gen. Syst. 4 (1888) 79. Cryptolepis elegans Wall. Cat. (1829) No. 1639, nomen nudum, G. Don Gen. Syst. 4 (1838) 82. Aganosma edithiae Hance in Ann. Sci. Nat. Bot. V 5 (1866) 227. Loureiro’s material was from China, presumably from the vicinity of Canton, and has long been considered a species of doubtful status. All the synonyms cited above, except the last two, are based on Loureiro’s binomial. The description applies closely to the species currently known as Cryptolepis elegans Wall., a species not uncommon in Kwangtung Province and represented by the following specimens: Merrill 10806, Levine 356, 1850, 3201. Loureiro described the seeds as naked, perhaps because he saw only those from which the coma had fallen, or perhaps because he really saw no seeds. I am convinced, how- ever, that the species as here interpreted is the one Loureiro intended. APOCYNACEAE ALYXIA Banks ALYXIA LEVINEI sp. nov. Frutex scandens, glaber, ramis ramulisque tenuibus, olivaceis; foliis oppositis et ternatis, junioribus membranaceis, vetustiori- bus chartaceis ad subcoriaceis, ellipticis ad oblongis, usque ad 8 cm longis, utrinque subaequaliter angustatis, obtuse acuminatis; inflorescentiis axillaribus brevibus, breviter pedunculatis, ut vi- detur paucifloris; fructibus junioribus ellipsoideis, circiter 9 mm longis. A scandent glabrous shrub, the branches and branchlets slen- der, the internodes up to 25 cm in length, branchlets 1.5 mm in diameter or less, somewhat angled or striate, the older branches terete, smooth, somewhat reddish brown. Leaves op- posite and in whorls of three, elliptic to oblong, 5 to 8 cm long, 2 to 3 cm wide, the younger ones membranaceous, the older ones chartaceous to subcoriaceous, olivaceous, shining, subequally narrowed to the acute or somewhat acuminate base and to the bluntly acuminate apex, the nerves often obsolete, never promi- nent; petioles 4 to 7 mm long. Inflorescences axillary, solitary, short, apparently very few-flowered, their peduncles 5 mm long XV,3 Merrill: Notes on the Kwangtung Flora 255 or less, obscurely puberulent, the persistent sepals oblong, 1.5 mm long. Young fruit ellipsoid, about 9 mm long. Kwangtung Province, Ting Woo Mountain, Levine 1975, May 26, 1918, scattered along the banks of streams, altitude about 300 meters. This is the second species of the genus to be found in Kwang- tung Province and differs remarkably from Alyxia sinensis Champ. in its much larger, differently shaped, acuminate leaves. LABIATAE PRUNELLA Linnaeus PRUNELLA VULGARIS Linn. Sp. Pl. (1753) 600. Kwangtung Province, Shiuchow region, To Kang P’eng 2854, April 25, 1919, with the local name ha fi ts’o. A widely distributed species in the North and South Temperate Zones. Central and northern China and Formosa, but not pre- viously recorded from Kwangtung Province. BORAGINACEAE TRIGONOTIS Steven TRIGONOTIS PEDUNCULARIS (Trev.) Benth. ex Baker & Moore in Journ. Linn. Soc. Bot. 17 (1879) 384. Myosotis peduncularis Trev. in Schrift. Naturf. Ges. Berl. 7 (1813) 147. Kwangtung Province, Shiuchow region, Levine 2543, along roads, May, 1919: The genus is new to Kwangtung Province, the species being common in central and northern China, the present locality being far south of its previously known range. BIGNONIACEAE DOLICHANDRONE Seemann DOLICHANDRONE STIPULATA (Wall.) Benth. ex C. B. Clarke in Hook. f. Fl. Brit. Ind. 4 (1884) 379. Spathodea stipulata Wall. Cat. (1832) No. 6518, nomen nudum, Pl. As. Rar. 3 (1832) 20, t. 238. Kwangtung Province, Kochow region, Kwanshan temple, To Kang P’eng 2726, February 18, 1919, with the local name mau mi muk. The Kwangtung specimen is with mature fruits and agrees closely with the description. The species was previously known only from Burma. 256 Philippine Journal of Science 1919 ACANTHACEAE 4 HEMIGRAPHIS Nees A 4 uo HEMIGRAPHIS PROCUMBENS (Lour.) comb. nov. Barleria procumbens Lour. Fl. Cochinch. (1790) 377. Ruellia chinensis Nees in DC. Prodr. 11 (1847) 147. 3 Hemigraphis chinensis T. Anders. in Journ. Linn. Soc. Bot. 26 (1890) 238. Strobilanthes scaber Hance in Journ. Bot. 16 (1878) 231, non Nees. Loureiro’s material was from the vicinity of Canton, where the species is still common, and on account of its yellow flowers is rather conspicuous in dry thickets. His description applies unmistakably to the species currently known as Hemigraphis chinensis T. Anders. It is represented by Merriil 10135 and Levine 181. CAPRIFOLIACEAE d LONICERA Linnaeus | LONICERA DASYSTYLA Rehder in Rept. Mo. Bot. Gard. 14 (1903) 158, tb ode: Kwangtung Province, North River, Sai Sha, 8z Ooi, Groff = 2402, April 24, 1918, with the local name kam ngan fa. The specimen agrees closely with Rehder’s figure and descrip- y tion, differing in but few minor details. The leaves average a smaller than in the type and are glabrous, while the older branchlets are brownish rather than grayish. The pubescent { style is characteristic. Previously reported only from Tonkin. | CUCURBITACEAE GYMNOPETALUM Arnott S \O° avMNOPETALUM CHINENSE (Lour.) comb. nov. Euonymus chinensis Lour. Fl. Cochinch. (1790) 156. Bryonia cochinchinensis Lour. op. cit. 595. Gymnopetalum cochinchinense Kurz in Journ. As. Soc. Beng. 40? (1871) 57; Cogn. in DC. Monog. Phan. 3 (1881) 391. Both of Loureiro’s descriptions apply unmistakably to the species currently known as Gymnopetalum cochinchinense Kurz, and it is indeed curious that he should have placed a cucurbita- ceous plant in the celastraceous genus Huonymus. The type of Euonymus chinensis was from the vicinity of Canton, for which Loureiro records the Cantonese name kam qua. It is repre- sented by the following recently collected material from the vicinity of Canton, with the recorded names ka shui kwah and ye kwah, Levine 1108, 1705, 2188. » XV,3 ‘Merrill: Notes on the Kwangtung Flora 257 RUBIACEAE 2400 PLECTRONIA Linnaeus \ PLECTRONIA LEVINEI sp. nov. Frutex glaber, ramis spinis longis rigidis rectis armatis; fo- liis chartaceis, ellipticis ad oblongo-ellipticis, 1.5 ad 4 cm longis, obtusus ad acutis, basi angustatis, acutis, nervis utrinque 2 vel 3, obscuris, subtus in axillis subobsolete glandulosis, reticulis obsoletis; fructibus axillaribus, solitariis, tenuiter pedicellatis, ovoideis, in siccitate nigris vel pruinosis, rugosis, circiter 6 mm longis. A glabrous shrub, the branches terete or the ultimate branch- lets obscurely angled. Leaves usually in pairs on two very short opposite branchlets, appearing like four leaves at each node, chartaceous, elliptic to oblong-elliptic, or sometimes somewhat obovate, rather pale when dry, slightly shining, 1.5 to 4 cm long, 1 to 2 cm wide, obtuse to acute, base narrowed, acute; lateral nerves 2 or 3 on each side of the midrib, slender, indistinct, their axils obscurely glandular on the lower surface, the reticulations obsolete; petioles 1 to 2 mm long; stipules about 1 mm long. Fruits axillary, solitary, black or somewhat pruinose, rugose, ovoid, about 6 mm long, when young crowned by the cylindric, 5-toothed, about 4 mm long calyx-tube, this soon deciduous, the pedicels 10 to 12 mm long, slender. Seeds usually two. Spines straight or slightly curved, stiff, rather slender, sharp, 1 to 1.5 em long. Kwangtung Province, Heung Shan District, near Macao, Le- vine 3487, January 18, 1919. This species belongs in the group with Plectronia horrida Benth. & Hook. f., P. parvifolia Benth. & Hook. f., and P. parvi- flora Bedd., but is readily distinguished by being entirely gla- brous, and by its solitary, slenderly pedicelled fruits. WENDLANDIA Bartling apt WENDLANDIA CHINENSIS Sp. nov. Species W. paniculatae affinis, differt stipulis hirsutis, inflores- centiis densissime cinereo-villosis, floribus dense confertis, glo- meratim dispositis, more W. tinctoriae. A shrub or small tree up to 8 m high, the branches glabrous or nearly so, the branchlets more or less brownish- or cinereous- pubescent. Leaves chartaceous, oblong-elliptic to oblong-lanceo- late, olivaceous and shining when dry, 10 to 12 cm long, 3 to 5 em wide, subequally narrowed to the acute base and rather slenderly acuminate apex, the upper surface very slightly sub- 258 Philippine Journal of Science : 1919 strigose-hirsute with widely scattered, short hairs, the lower surface sparingly pubescent with widely scattered, short, cine- reous hairs on the midrib, nerves, reticulations, and epidermis; lateral nerves about 10 on each side of the midrib, distinct; petioles 5 to 10 mm long; stipules coriaceous, persistent, or- bicular-reniform, somewhat hirsute, 7 to 9 mm wide, rounded, sometimes contracted below and distinctly stipitate. Panicles terminal, ample, up to 20 cm long and wide, densely cinereous- villous. Flowers very numerous, densely crowded in glomerules on the ultimate branches; bracts 5 to 7 mm long. Calyx densely cinereous-villous, about 2.5 mm long, the lobes oblong. Corolla- tube 4 to 5 mm long, slender, glabrous externally, sparingly pubescent within. Kwangtung Province, Kochow region, Shek Kau Tong, To Kang P’eng 2691 (type), March 6, 1919. To this species I also refer the following specimens, both described as shrubs about 2 m high, both with somewhat smaller leaves than the type and with unopened flowers: To Kang P’eng 2754, 2702, the former from Sheung Ko Wan, with the local name fo shiu nap, the latter from Shek Ling, with the local name chii liit shii. This species is manifestly allied to Wendlandia paniculata (Roxb.) DC., the type of which was from the Molucca Islands. Comparison with Amboina material, Robinson 1731, represent- ing the typical form of Roxburgh’s species, shows that the Chinese form differs radically in its very densely cinereous- villous inflorescences and calyces, and in its very densely crowded flowers, in the disposition of the flowers strongly resembling Wendlandia tinctoria DC. The Chinese form described by Hance as Wendlandia uvariifolia has been reduced to Wendlandia pani- culata DC. It is represented by Levine 2338, from the North River region, and is distinctly different from Wendlandia pani- culata and I believe should be retained as of specific rank under Hance’s name. It differs from both Wendlandia paniculata DC. and W. chinensis Merr. in its leaves being rather densely ferru- ginous-pubescent beneath, while the disposition of the flowers and the indumentum of the inflorescences and calyces are quite different from the latter species. MUSSAENDA Linnaeus MUSSAENDA PARVIFLORA Miq. Ann. Mus. Bot. Lugd.-Bat. 3 (1867) 110. Kwangtung Province, Ting Woo Mountain and at Wan Lo Mountain, Kochow region, Levine 1979, To Kang P’eng 2692, April, 1918, and March, 1919. v $ \gaul XV,3 Merril: Notes on the Kwangtung Flora 259 The specimens agree closely with Formosan material and also conform to Miquel’s description. I believe this to be, at least in part, the Kwangtung form referred by Dunn and Tutcher to Mussaenda frondosa Linn., but I have seen no Chinese ma- terial at all approaching the typical Ceylon form of the Linnean species. RANDIA Linnaeus RANDIA ACUMINATISSIMA sp. nov. Arbor parva, usque ad 8 mm alta, ramulis et inflorescentiis et subtus foliis ferrugineo-pubescentibus; foliis chartaceis, oblongo-ellipticis ad oblongo-lanceolatis, usque ad 20 cm longis, utringue angustatis, basi acutis, apice tenuiter caudato-acumi- natis, supra in siccitate olivaceis, glabris, nitidis; nervis utrin- que 10 ad 12, subtus perspicuis; stipulis lineari-lanceolatis, acuminatis, usque ad 1 cm longis; ¢ymis oppositifoliis, 3 ad 4 cm longis, breviter pedunculatis; floribus ad apices ramulorum confertis, calycis segmentis lanceolatis, acuminatis, circiter 1.5 mm longis; fructibus globosis, glabris, 6 ad 8 mm diametro, in siccitate nigris, nitidis; seminibus numerosis, compressis, cir- citer 1.5 mm diametro. A small tree about 8 m high, the branchlets, inflorescences, and the lower surface of the leaves rather densely ferruginous- pubescent. Branches dark reddish brown, usually terete, gla- brous. Leaves oblong-elliptic to oblong-lanceolate, chartaceous, 11 to 20 cm long, 4 to 7 cm wide, subequally narrowed below to the acute, equilateral base, and above to the slenderly caudate- acuminate apex, the acumen sometimes falcate, the upper sur- © face glabrous, olivaceous and shining when dry, the lower sur- face rather softly pubescent; lateral nerves 10 to 12 on each side of the midrib, prominent on the lower surface, anasto- mosing, the reticulations rather lax, distinct; petioles usually pubescent, 5 to 8 mm long; stipules linear-lanceolate, acuminate, pubescent, up to 1 cm long. Cymes leaf-opposed, shortly pe- duncled, ferruginous-pubescent, 3 to 4 em long (corollas un- known), the flowers sessile or shortly pedicelled and somewhat crowded at the tips of the branchlets... Calyx ferruginous- pubescent, about 4 mm long, the lobes lanceolate, acuminate, about 1.5 mm long. Fruits globose, glabrous, 6 to 8 mm in diameter, black and shining when dry. Seeds many, flattened, - orbicular-ovate to ovate, about 1.5 mm long. Kwangtung Province, White Cloud Mountain, Levine 3130 (type) 3267, August 29, November, 1918; Kong Moon, Groff N\S a 260 Philippine Journal of Science 1919 2471, March, 1918; Ukantin, Hongkong Herbarium 10918, dis- tributed as Randia densiflora Benth. This species is manifestly allied to Randia racemosa (Cav.) F.-Vill. (R. densiflora Benth.), from which it is easily distin- guished by its indumentum. COMPOSITAE cust GYNURA Cassini GYNURA SEGETUM (Lour.) comb. nov. Cacalia segetum Lour. Fl. Cochinch. (1790) 486, in nota. Cacalia pinnatifida Lour. |. c. non Linn. Gynura pinaatifida DC. Prodr. 6 (1837) 301. Kwangtung Province, Kochow region, Kwong T’am Mountain, To Kang P’eng 2671, March 22, 1919, in a garden, with the local name tung fung twp. The type of Loureiro’s Species was from Canton, where he observed it growing in rice paddies. He records the Cantonese name as cien fan sat. His description applies closely to the specimen cited above. I consider his specific name pinnatifida to be invalidated by the earlier Cacalia pinnatifida Linn., an entirely different species, and hence adopt the casual name published by him: “unde vernaculum nomen Sinense Cacalia Segetum.” 4 CROSSOSTEPHIUM Lessing CROSSOSTEPHIUM CHINENSE (Linn.) comb. nov. Artemisia chinensis Linn. Sp. Pl. (1753) pao, excl. syn. Gmelin; Lour. Fl. Cochinch. (1790) 492. Artemisia judaica Lour. Fl. Cochinch. (4790) 489, non Linn. Crossostephium artemisioides Less. ex Cham. & Schlecht. in Linnaea 6 (1831) 220. The genus Crossostephium was based on cultivated specimens from Manila and from Canton, the species being widely cul- tivated in Japan, China, the Philippines, and Indo-China. I have seen no specimens from wild plants, although the species is manifestly a native of either China or Japan. It is currently known in Manila, where it is cultivated in pots, as ajenjo, a Spanish name properly belonging to Artemisia. The type of the Linnean species was a specimen collected in China by Lager- stroem, and the Linnean description based on this specimen clearly applies to the species currently known as Crossostephium artemisioides Less. The species is still common in cultivation in Canton. Both of Loureiro’s descriptions cited above apply to this species, ans * g KV, 3 Merrill: Notes on the Kwangtung Flora 261 EMILIA Cassini EMILIA PRENANTHOIDEA DC. Prodr. 6 (1837) 303. Kwangtung Province, Teng Woo Mountain, Levine 2041, 8221, May 26, 1918. India to the Philippines; this species has previously been tentatively recorded from China by Forbes and Hemsley,'® who state that in preparing their list they had recorded the species from China on the basis of a specimen collected by Fortune, but that the specimen was not to be found at the time their manuscript was written. The species is readily distinguished from Emilia sonchifolia DC. not only by its narrow leaves, but also by its involucral bracts being much shorter than the flowers and by its glabrous achenes. SENECIO Tournefort SENECIO OLDHAMIANUS Maxim in Bull. Acad. Pétersb. 16 (1871) 219. Kwangtung Province, Shiuchow region, To Kang P’eng 2852, April 25, 1919, in grassy places. Not previously recorded from Kwangtung Province; common in central China. * Journ. Linn. Soc. Bot. 23 (1888) 449. NOTIZ UEBER HOYA IMBRICATA CALLERY EX DE- CAISNE UND HOYA PSEUDOMAXIMA KDS, IN DEN FILIPINEN AUF GRUND VON EINIGEN HERBAR- EXEMPLAREN DES BUREAU OF SCIENCE IN MANILA Von S. H. Koorprers Buitenzorg, Java MIT VIER TAFELN 1. HOYA IMBRICATA Callery ex Decaisne, forma TYPICA Kds. Taf. I. Decaisne giebt in DC. Prodromus 8 (1844) 637 folgende von ihm in Delessert, Icones selectae plant. 5 (1846) 37 und spater von Miquel, Flora Indiae Batav. 2 (1857) 520, unverindert iibergenommene Spezies-Beschreibung von Hoya imbricata: Seandens radicans, foliis abortu alternis unilateralibus imbricatis or- bicularibus supra convexis aveniis subtus venosis marginibus reflexis, pedunculis multifloris, pedicellis gracilibus glabris, callyce brevi, corollae laciniis triangularibus acutis, coronae stam. fol. supra convexiusculis mar- ginibus revolutis angulo int. porrecto antherumque membrana attenuata stigmati apiculato incumbentibus, folliculis laevibus. Frutex scandens arbor. truncis conjunctissime affixus. (Decne. l. ¢.) Von dieser characteristischen, bisher im Buitenzorger Herbar fehlenden Art, erhielt ich leihweise aus dem Herbarium des Bureau of Science in Manila, von Herrn E. D. Merrill, folgendes von ihm in den Filipinen gesammeltes Herbar-Exemplar: LuZON, Rizal Province, Tanay, Merrill 2363, ausgebliiht im Mai, 1903. Dieses Exemplar ist von Merrill etikettiert: Hoya imbricata Dene., und von ihm mit folgender Bemerkung versehen: “This should be typical H. imbricata. Leaves mottled, green and purple.” Das Exemplar wurde nach seiner Angabe von ihm gesammelt ziemlich nahe bei dem Fundort von den authentischen Exemplar von Hoya imbricata des Pariser Herbar. Es besteht aus einem Bliitenzweig mit 4 sich dachziegelig deckenden Blattern und einem ausgebliihten Bliitenstand, und einer Papierkapsel mit einem einzelnen Laubblatt. Von allen mir aus dem Herbarium des Bureau of Science in Manila, unter dem Namen Hoya imbricata leihweise zuge- 263 264 Philippine Journal of Science 1919 schickten Herbarexemplaren, stimmt kein einziges so gut tiberein mit der Beschreibung und Abbildung von Hoya imbricata in DeCandolle Prodromus, und in Delessert, wie obengenanntes Spezimen (Merrill 2363), dass es als “Topotype” (im Merrill- schen Sinne) von Hoya imbricata betrachtet werden darf. tiber dieses bemerkenswerte Herbarexemplar (Merrill 2363) lasse ich folgende Beschreibung folgen: Blatter kreisrund, + ganzrandig oder eckig-ausgeschweift, am Grunde abgerundet-abgestutzt, oben abgerundet oder emar- ginat; oberseits convex, ohne deutliche Nerven, véllig kahl und glatt (auch ohne Cuticular-Hécker) ; unterseits concav (auch ohne Cuticular-Hécker) ; + in der Mitte dem sehr kurzen, fast fehlenden Blattstiel peltat inseriert und handnervig, mit un- deutlichen, gabelig verzweigten Nerven; + 9 cm lang und 10 em breit. Epidermis des Blatt-Oberseite ohne Haare, mit — glatter oder schwachwelliger etwa 20 » dicker Cuticula, fast ohne Cuticular-Hécker und dadurch aussen + glatt (nicht kérnigrauh) aussehend. Epidermis Zellen des Blattes Unter- seite je mit einem wagerecht abstehendem, einzelligem, unver- zweigtem, fadenférmigem oder schmalkegelférmigem, spitzem, glattem, inhaltlosem, farblosem, gerade oder + gekriimmtem, 30-50 p»p langem Haar versehen und mit nur 5 » dicker Cuticula. Stengel stielrund, + 4 mm Durchmesser, kahl (auch ohne Pa- pillen Cuticular-Hiécker), mit zahlreichen, dichtstehenden Haft- wurzeln, welche meistens ganz oder fast ganz von den dachig deckenden Laubblittern geschiitzt sind. Pedunculus axillar (in sicco abstehend), + 10 cm lang, stielrund, kahl; rhachis + 9 cm lang, kahl, dicht bedeckt mit den Narben der abgefallen- en Pedicelli, spindelférmig, in der Mitte fast 1 cm breit. Dieses Exemplar (Merrill 2363) ist von mir Hoya imbricata Callery ex Decaisne, forma typica Kds. (msc. 1918) etikettiert worden. 2. HOYA IMBRICATA Callery forma BAS!I-SUBCORDATA Kds. forma nov: Taf. Ti Ti: A typo recedit foliis basi subcordatis. Blatter kreisrund, + ganzrandig, oben abgerundet, am Grunde bis 1 cm (oft nur 0.5 cm) untief-herzférmig, + in der Mitte peltat, dachig deckend, bisweilen (Copeland 399) auseinander entfernt. Bliiten + wie der Typus. Friichte unbekannt. Die folgenden 3 Exemplare des Manila-herbars halte ich fiir diese neue Form (foliis basi-subcordatis) von Hoya imbricata: BILIRAN, Bur. Sci. 18893 McGregor, bliihend im Juni, 1914. => XV,3 Koorders: Notiz tiber Hoya Imbricata 265 MINDANAO, Davao District, bei Davao, Copeland 399, bliithend im Marz, 1904. SAMAR, Bur. Sci. 24910 Edano. Diese Exemplare stimmen Beide im Bliitenbau mit der Be- schreibung und Abbildung von Hoya imbricata in Decaisne gut iiberein, und besitzen obenso wie auch Merrill 2363, beiderseits vollig kahle Blatter (die auch keine Cuticular-Hécker haben), jedoch ist der Blattfuss deutlich “subcordat’’. Die Tiefe des herzférmigen Blattfuss-Einschnittes betrigt + 0.5-1 cm. Die Bliiten von Bur. Sci. 18893 McGregor sind noch nicht vollig ausgebildet und kleiner als von Copeland 399. Bei beiden Exemplaren liegen keine Friichte vor. Mit obiger forma basi-subcordata zeigt folgendes Filipinen- Exemplar einige Ahnlichkeit, jedoch ist die Korolla auf innen auffallend dicht behaart: SAMAR, Catubig River, Bur. Sci. 24910 Edano, Bliihend im Februar-Marz, 1916. Diesem Exemplar ist von Merrill folgende Bemerkung hinzugefiigt: “Flowers ap- pear to be different from Hoya imbricata.” Auch bei diesem Exemplar sind keine Friichte vorhanden. Vielleicht liegt hier eine von Hoya imbricata neue Spezie vor. 3. HOYA PSEUDOMAXIMA Kds. sp. nov. Taf. IV. A Hoya imbricata Callery foliis basi profunde cordatis et a Hoya maxima Kds. foliis supra glaberrimis et apice haud raro emarginatis differt. LuzON, Rizal Province, Bosoboso, Bur. Sci. 22089 Ramos, bliihend am 7ten December, 1913. Blatter kreisrund, am Grunde bis 3 cm tief-herzférmig, ober- seits véllig kahl (nicht nur unbehaart, sondern auch ohne Cuti- cular-Hécker) und glatt, unterseits kahl und auch ohne Cuti- cular-Hoécker, bis 8 ecm lang und 9 cm breit. Bliiten + wie. Hoya imbricata, Friichte unbekannt. Bemerkung: Wihrend bei allen (12) Einsammlungs-Num- mern meiner Hoya maxima aus Nord-Ost-Celebes (Kds. n. 16204 f, u.s.w.) die Blatt-Oberseite stets mit Cuticular-Héckern dicht bedeckt ist, welche dauernd oder seltener nur in der Jugend je ein Flaumhaar tragen, bei der nur aus den Filipinen (Mount Isarog) bekannten Hoya pseudomaxima ist die Blatt-Oberseite villig glatt (glaberrima) ohne Cuticular-Hécker und ohne Flaumhaare. ERKLARUNG DER ABBILDUNGEN {Originalzeichnung von Mangoendimedjo, direxit Koorders ; Nachzeichnung von J. K. Santos.] Fig. Fie. FI. TAFEL I Hoya imbricata Callery ex Decaisne, forma typica Kds. Blitenzweig mit ausgebliihtem Bliitenstand, wovon alle Bliiten abgefallen sind. . Blatt, Unterseite. Ausgebliihter Bliitenstand. - Querschnitt durch die Epidermis der Blatt-Oberseite, . Idem der Blatt-Unterseite. (Original, nach Merrill 2363 in Herb. Bureau of Science in Manila. Gesammelt in den Filipinen, auf Luzon, in der Provinz Rizal bei Tanay, im ausgebliihtem Zustand, im Mai, 1903.) In dieser Gegend wurde nach der brief- lichen Mittheilung von Merrill das Original von Hoya imbricata gesammelt. Damit stimmt Herb. Merrill 2363 nach der Abbildung und nach der Beschreibung von Decaisne vorziiglich itberein. Auf der Original-Einsammlungs-Etikette von Merrill 2363 steht: — “leaves mottled, green and purple.” Ferner wurde von ihm auf seinem mir geliehenen Original-Herbarbogen folgendes erwahnt: “This should be typical Hoya imbricata.” TAFEL II Hoya imbricata Callery forma basi-subcordata Kds. Bliitenzweig, Unterseite. . Blattzweig, Oberseite. . Blattzweig, Unterseite. . Bliitenstand. . Bliite. (Original, nach Bur. Sci. 18893 McGregor.) TAFEL III Hoya imbricata Callery forma basi-subcordata Kds. 1. Bliitenzweig, Unterseite. 2. Bliite. 8. Translatoren. (Original, nach Copeland 399.) aAaNantk wnNwr TAFEL IV Hoya pseudomaxima Kds. Blattzweig, Oberseite. Bliite. . Bliite. . Translatoren. . Blattzweig, Oberseite. . Blatt, Unterseite. Bliitenzweig. . Bliite. (Original, 1-4 nach Bur. Sci. 970 Ramos; 5-8 nach Bur. Sci. 22089 Ramos.) 267 Koorpers: Notiz tper Hoya.) [PHuip. Journ. Sci., XV, No. - BAN NG “ms KDA a} ff AX) tty Ay | TAFEL |. HOYA IMBRICATA CALLERY EX DECAISNE, FORMA TYPICA KOORD. Koorpers: Notiz tiper Hoya. ] [Puiuip. Journ. Scr, XV, No. 3. 4 TAFEL Il. HOYA IMBRICATA CALLERY, FORMA BASI-SUBCORDATA KOORD. KoorpDers: Notiz UBER Hoya.] ({Puiute, Journ. Scr, XV, No, 3. TAFEL Ill. HOYA IMBRICATA CALLERY, FORMA BASI-SUBCORDATA KOORD. [Puuire. Journ. Sci., XV, No. 38. Notiz tiper Hoya.] KOORDERS HOYA PSEUDOMAXIMA KOORD. TAFEL IV. THE BLACK HALICTINE BEES OF THE PHILIPPINE ISLANDS By T. D. A. CocKERELL Of the University of Colorado The black halictine bees of the Philippines have all been referred to the genus Halictus; but the subgenus Nesohalictus, of Crawford, remarkable for its very long tongue, may be ele- vated to generic rank. Genus NESOHALICTUS Crawford Halictus subgenus Nesohalictus CRAWFORD, Proc. U. S. Nat. Mus. 38 (1910) 120. “Closely resembles Halictus, but differs by the greatly elongate tongue, which is about as long as the antennw, by having the inner spur of the hind tibiz in the male armed with a few large teeth, and by having all the transverso-cubital veins, the cubitus beyond the second transverse cubital, and the upper two-thirds of the second recurrent vein, obsolescent.” (Crawford.) This diagnosis was based on the male. The female is larger (about 7.5 millimeters long), with the basitarsi piceous (they are ferruginous in the male); the hind spur has long spines. On account of the long linear tongue the genus is related to Thrinchostoma, but it differs in the struc- ture of the hind legs, the absence of the hair patch on the wings of the male, and in other ways. The clypeus of the male is en- tirely black, flattened and shining in the middle. Nesohalictus robbii (Crawford). Halictus (Nesohalictus) robbii CRAW¥FoRD, Proc. U. S. Nat. Mus. 38 (1910) 120. Luzon, Manila (M. L. Robb and R. E. Brown): Laguna Province, Mount Maquiling (C. F. Baker), Los Bafios (Baker). PANAY, Antique Province, Culasi (R. C. McGregor), June, 1918; Maralison Island (McGregor), June 18, 1918. MINDANAO, Da- pitan, Davao, and Cagayan (Baker). Another species of this genus, Nesohalictus lativentris (Halictus lativentris Friese), occurs in Java. Genus HALICTUS Latreille The metallic species have been previously described, but one described below (H. opisthochlorus) is partly metallic. The 1671484 269 270 Philippine Journal of Science 1919 species of Halictus are ground-nesting bees of relatively feeble flight, and the wide distribution of several species through the islands of the Archipelago is rather surprising. The smaller species are sometimes referred to as sweat bees, from their habit of settling on the skin, apparently attracted by the per- spiration. Is it perhaps possible that they occasionally attach themselves to birds, and thus get carried about? The follow- ing key should suffice for the separation of the species before me. I include Nesohalictus, as it is likely to be mixed with Halictus. ooo acta sesscvcssncuncesdsnndeecbbmiendiwebinsnsonatngss sheveeseomuntveds 2 Females 8. 1. Clypeus with a yellow or whitish transverse apical band 2, Clypeus entirely black = le 2. Tibie yellow, with dark markings; scape with a light stripe; area of metathorax not plicate or only very delicately so near base. eschscholtzi sp. nov. Tibiz mainly or wholly dark..... 3. 3. Flagellum ferruginous beneath; hind tarsi or, at least, basitarsi yellow. philippinensis Ashmead. Flagellum dark or dusky beneath; hind tarsi dark. philippinensis nigritarsellus var. nov. i ec at er i ee a ea ea 5. Mesothorax shining 5. Larger; stigma pale ferruginous; tarsi light red. Nesohalictus robbii Crawford. Smaller; stigma dark brown; tarsi dark.................... melanurops sp. nov. 6. Head very large; scape largely red; anterior knees red.... scapalis sp, nov. Head not unusually large 6a. 6a. Very small species; mesothorax highly polished, without conspicuous punctures = lionotulus sp. nov. Small species; mesothorax shining, but finely and distinctly punctured. itaminus sp. nov. Larger; mesothorax with evident punctures c. 7. Mesothorax coarsely punctured; first abdominal segment closely punc- tured all over..... baguionis Crawford. Mesothorax polished, more finely punctured; first abdominal segment without evident punctures... luzonicus Strand. 8. Tibiz mainly yellow or reddish yellow; tubercles yellow. eschscholtzi sp. nov. sue See tooercios not this: yellow. ..0 52 oo 9. 9. Mesothorax dull, without evident sculpture... 10. Mesothorax more or less shining or evidently sculptured... 12. 10. Metathorax dark green (Baguio) 000... opisthochlorus sp. nov. Metathorax black, the area granular, without evident sculpture........ 11. 11. Apical part of metathoracic area shining and with a median groove (Puerto Princesa) caroli sp. nov. Apical part of area entirely dull; larger species (Culasi and Mount Baquiling) (sno canter mre tedesseciceeeisigs megregori sp. nov. XV,3 Cockerell: The Black Halictine Bees 271 12. 13. 14. 15. 16 1%. 18. 19. 20. 21. 22. 23. 24. Area of metathorax entirely dull, very coarsely reticulate. thoracicus sublustrans subsp. nov. Area of metathorax shining or closely sculptured 13. Large species, with dark tegule and very coarsely punctured mesotho- rax; apical area of first abdominal segment closely punctured. baguionis Crawford. Mesothorax not thus coarsely punctured... Scutellum sparsely punctured Scutellum closely or finely punctured, often dullish or dull 22 Scutellum dullish, with large, very sparse punctures; mesothorax dull, with large sparse punctures (Mount Banahao).... oligostictus sp. nov. Scutellum shining; mesothorax with fine or close punctures................ 16. Mesothorax highly polished, little punctured; tegulze amber................ 17. Mesothorax dull or distinctly punctured. o.oo... elect ceeeeceeeeeeeeee- 18. Hind margins of abdominal segments testaceous; area of metathorax distinctly striate, except the shining apical part (Mount Maquiling). lionotulus sp. nov. Hind margins of abdominal segments not testaceous; area of meta- thorax not distinctly sculptured (Mount Banahao). scintillans sp. nov. Head very large; scape largely red; anterior knees red in male; female unknown (Mount Banahao) scintillans sp. nov. Head not large; scape dark; abdomen with distinct bands of tomentum at bases of at least some of the segments 19. Larger species; middle of hind basitarsi with ferruginous hair on inner side; mesothorax shining and with very distinct punctures. luzonicrs Strand. Smaller species; basitarsi without strongly red or orange hair............ 20. Stigma pale testaceous; area of metathorax shining, with very dis- tinct longitudinal plice; first abdominal segment with a tuft of White 7air On each fide. =. oa philippinensis Ashmead. PRLS ON ggasi scans do pao ale cpt ins 21 Disk of mesothorax very sparsely pimnetured: posterior truncation of metathorax brilliantly shining (Baguio) .................. postlucens sp. nov. Disk of mesothorax much more closely punctured; posterior truncation dull or dullish, except upper end itaminus sp. nov. Hind margins of at least some of the abdominal segments pale tes- taceous; tegulze rufotestaceous; stigma dull reddish, not dark........ 23. Hind margins of abdominal segments not testaceous 25. Testaceous band on second abdominal segment in middle occupying -about half of segment; first segment with broad testaceous margin; hind tibie and tarsi dull reddish (Mount Maquiling). fulvovittatus sp. nov. Testaceous band on second abdominal segment, if present, not occupy- ing a fourth of segment. 24, Larger; truncation of metathorax distinctly innrpinea above (Davao). davaonis sp. nov. Smaller; truncation not bounded above....................... nesiotus Crawford. Like the last, but more robust, with red knees and redder tegulz, nesiotus domitus var. nov. 272 Philippine Journal of Science 1919 25. Second abdominal segment wholly without basal band or patches of tomentum; mesothorax dull; punctures of scutellum very irregular; area of metathorax dullish, with strong plice (Mount Banahao and Imugan) melanurops sp. nov. Second abdominal segment with basal band or lateral patches of to- mentum ..26. 26. First recurrent nervure meeting second transverse cubital; tongue linear. Nesohalictus robbii (Crawford). First recurrent nervure joining second submarginal cell; mouth parts ordinary for the genus; hair of postscutellum light fulvous, banahaonis Cockerell. Halictus itaminus sp. nov. Female.—Lenegth, about 6 millimeters; black, with thin white hair, conspicuous on sides of thorax and forming a dense to- mentum on postscutellum; head normal, face broad but eyes strongly converging below, clypeus somewhat produced; man- dibles black, reddish at end; underside of head fiattened, finely and densely striate; antenne dark, the flagellum faintly brown- ish beneath; front dull, extremely densely punctured; clypeus irregularly punctured; mesothorax shining but not highly pol- ished, the punctures distinct and well separated, the median groove well developed; scutellum convex, shining, sparsely punctured, with no median groove (H. gedensis has it dullish, with a median groove) ; area of metathorax appearing roughened (but under microscope seen to be reticulated) at middle and base, the posterior part at side being smooth (microscopically tessellate), while the posterior middle is somewhat swollen, with a median depression; posterior truncation dull - except upper part, which is shining, its sides angular; sides of thorax dull, except a shining area below the wings; tegule piceous; wings dusky, stigma (which is large) and nervures piceous; first recurrent nervure meeting second transverse cubital; legs black, with pale hair; hind spur with long spines; abdomen shining, with white hair bands, broadest laterally (not always clearly visible) at bases of second and third segments; venter with white hair. Male.—Very much like the female, but smaller and slenderer; clypeus entirely black; legs entirely dark; mesothorax shining. Luzon, Laguna Province, Mount Maquiling, 10 females and 8 males (the type is a female); Mount Banahao, 1 female: Mountain Province, Baguio, 1 female. All from Baker. This is very close to certain species of Java, H. gedehensis Friese and especially H. gedensis Ckll., differing from the latter especially in the scutellum, as is indicated in the preceding description. . XV, 8 Cockerell: The Black Halictine Bees 273 Halictus nesiotus Crawford. Halictus nesiotus CRAWFORD, Proc. Ent. Soc. Washington 19 (1918) 169. LuzON, Mountain Province, Baguio (Wirt Robinson), type locality: Laguna Province, Mount Maquiling (Baker), 4 females; Mount Banahao (Baker), 1female. I have not seen Crawford’s type, but he has kindly confirmed my identification from a speci- men sent to him. Halictus nesiotus var. domitus var. nov. Another specimen from Mount Banahao is more robust, with red knees, bright ferruginous tegule, and third and fourth ab- dominal segments (but not first and second) rather broadly margined with testaceous. It may be a distinct species, but for the present it may stand as a variety. Halictus baguionis Crawford. Halictus baguionis CRAWFORD, Proc. Ent. Soc. Washington 19 (1918) 170. LuzZON, Mountain Province, Baguio (Robinson, Baker): Mount Banahao (Baker): Nueva Vizcaya Province, Imugan (Baker). This species appears to be abundant at Baguio, but a few specimens have been seen from other localities. I have one of Crawford’s cotypes. Halictus luzonicus Strand. Halictus luzonicus STRAND, Berl. Ent. Zeitschrift 54 (1909) 208. ? Halictus manile ASHMEAD, Canad. Ent. 36 (1904) 281. Halictus manilz COCKERELL, Proc. U. S. Nat. Mus. 36 (1909) 420; CRAWFORD, Proc. Ent. Soc. Washington 19 (1918) 169, 170. Strand’s type was from Luzon (Jagor); Ashmead’s, from Manila (Stanton). Ashmead’s type, preserved in the United States National Museum, is the same as H. luzonicus; but Ash- mead’s description states that the insect, a female, is only 6 millimeters long, and the head from the base of the antenne upward is greenish metallic. It appears probable that Ashmead began his description with one thing and finished with another, accidentally picking up the wrong specimen. At all events, I know of no species agreeing with his description and prefer to use the name luzonicus, which is supported by a full and accurate account of the characters. LuzoN, Mountain Province, Baguio (Robinson, Baker): La- guna Province, Los Bafios (Baker) ; Mount Maquiling (Baker) : 274 Philippine Journal of Science 1919 Nueva Viscaya Province, Imugan (Baker). It appears to be abundant at Baguio, but is. represented by a few specimens from the other localities. Halictus philippinensis Ashmead. Halictus philippinensis ASHMEaD, Proc. U. S. Nat. Mus, 28 (1904) 128; CoCKERELL, Proc. U. S. Nat. Mus. 36 (1909) 419; CrAWwForD, Proc. Ent. Soc. Washington 19 (1918) 169, 170. LuzON, Manila (Stanton): Laguna Province, Los Bafios (Baker); Mount Maquiling (Baker): Mountain Province, Baguio (Baker): Tayabas Province, Malinao (Baker). LEYTE, Tacloban (Baker). PANAyY, Antique Province, Culasi (Mc- Gregor), June, 1918. NrEGRos, Cuernos Mountains (Baker). MINDANAO, Davao (Baker); Cagayan (Baker); Zamboanga _ (Baker). -BASILAN (Baker). PALAWAN, Puerto Princesa (Baker). This is the commonest and most widespread Halictus in the Archipelago, extending from Luzon to Mindanao and Palawan, A specimen from Los Bafios carries a manuscript name given by Friese. Halictus philippinensis var. nigritarsellus var. nov. Male.—Hind tarsi dark; flagellum dark or dusky beneath. The tarsi are entirely without yellow, but the last joint is fer- ruginous apically. Type of the variety from Pagsanhan, La- guna Province; others from Mount Maquiling and Los Banos, Luzon. All from Baker. Halictus thoracicus sublustrans subsp. nov. Female.—Length, about 6 millimeters; similar to H. thoraci- cus Friese, from Java, but with the abdomen shining and tegule, stigma, and nervures piceous or black. The dense to- mentum on the postscutellum is white, often with a delicate ocherous tint. The species is very peculiar for the wholly dull area of the metathorax, with large reticulations, and especially for the sculpture of the mesothorax, which appears rough under a lens but under the microscope is seen to have a raised network or cancellation all over, entirely different from the fine tessella- tion often seen on the thorax of bees. 3 Luzon, Nueva Viscaya Province, Imugan (Baker), 5 (in- cluding the type): Mount Banahao (Baker), 2. PANAy, An- tique Province, Culasi (McGregor), 3, at flowers of Homalanthus populneus Pax, June 3, 1918. . fo XV,3 ° Cockerell: The Black Halictine Bees 275 Halictus thoracicus var. merescens var. nov. Female—Tegule rufous or rufotestaceous, sometimes darker; stigma reddish brown to dark brown. Luzon, Mount Banahao (Baker) 7; Los Bafios (type locality of variety) (Baker), 3; Mount Maquiling (Baker), 5. PANAY, Antique Province, Culasi (McGregor), 1, at flowers of Homalan- thus populneus Pax, June 3, 1918. LEYTE, Tacloban (Baker), 1. MINDANAO, Davao (Baker), 1. I have seen true H. thoracicus Friese from Buitenzorg, Java (Bryant and Palmer). Halictus banahaonis Cockerell. Halictus banahaonis CocKERELL, Ann. & Mag. Nat. Hist. VIII 14 (1914) 365; 15 (1915) 264; CRAwrorD, Proc. Ent. Soc. Washington 19 (1918) 169. LuzON, Mount Banahao (Baker), 4; Paete (Baker), 1. The hind margin of the first abdominal segment is lineolate, not punctured all over as in baguionis Crawford, contrary to the statement in Crawford’s table, loc. cit. Halictus banahaonis var. macerula var. nov. Female.—Smaller; anterior wing, about 5 millimeters long (about 6 millimeters in type) ; length of insect, 6 to 6.5 milli- meters. LuzoN, Mount Banahao (Baker), 18. The difference of size and bulk is very conspicuous, and the two lots are uniform. Presumably the two forms are adapted to different flowers. The type of the variety is Baker 2556. Halictus eschscholtzi sp. nov. Halictus manilz STRAND (not of Ashmead), Berl. Ent. Zeitschrift 54 (1909) 210; Crawrorp, Proc. Ent. Soc. Washington 19 (1918) 169. Female (type).—Length, about 7 to nearly 9 millimeters; black, with the tubercles yellow; mandibles bright chestnut red, except at base; knees, tibis, and tarsi reddish yellow, or the anterior tibie light red with a yellow stripe, the middle and hind tibize and hind basitarsi each with a large dusky area; flagellum red beneath, and scape with a red basal spot; hair of head and thorax ochraceous, abundant on thorax, forming a fine down on mesothorax, conspicuous in certain lights, not in others, but sides of mesothorax with dense bands of tomen- tum, posteriorly extending narrowly in front of scutellum; mesothorax dull, it and the scutellum very minutely rugosely 276 Philippine Journal of Science 1919 punctate; area of metathorax rather large, dull, microscopically rugose, with a slight tendency to plication basally; tegulee testa- ceous, with a yellow spot; wings hyaline, with pale ferruginous stigma and nervures; first recurrent nervure joining second submarginal cell some distance before its end; third submar- ginal cell much broader in middle than second; outer recurrent and transverse cubital weakened; posterior face of metathorax densely covered with hair; legs with pale ocherous hair; hind spur with four long slender spines and a fifth rounded rudi- ment; abdomen broad, moderately shining, with a sericeous surface; basal part of first segment with dense pale ocherous hair, and segments 2 to 4 with dense basal bands; ventral seg- ments polished and shining basally. Male.—Length, about 6.5 millimeters; clypeus with a broad yellow apical band; labrum yellow; mandibles broadly yellow in middle, but with a black spot at base; scape yellow in front; flagellum rather long; dark areas on tibiz smaller, and hind basitarsi all yellow; hind margins of abdominal segments very narrowly fulvous; apical plate very broad and rounded; third ventral segment with a dense stiff brush of whitish hair along the margin, not reaching the sides. LuzON, Laguna Province, Los Bafios (Baker), type locality; Mount Maquiling (Baker); Pagsanhan (Baker), the male de- scribed and 2 females: Manila (Eschschoitz), type of H. manilx Strand. MINDANAO, Cagayan (Baker); Dapitan (Baker). A male from Mount Maquiling is Baker 2560. A very dis- tinct species, on account of the color of the legs resembling H. roepkei Friese, from Java. Halictus opisthochlorus sp. nov. Female.—Length, nearly 7 millimeters; anterior ‘wing, 4.75 millimeters; black, with the upper part of the metathorax dark blue-green; pubescence scanty, face with inconspicuous dark fuscous hairs, mesothorax almost bare, sides of thorax with thin white hair, abdomen without hair bands or patches, but with a scopa of long, white, curled hairs on ventral surface; head broad and short; antennz black, scape reaching to level of ocelli; face and front dullish with a sericeous luster; meso- thorax and scutellum dullish, with a sericeous luster and with- out evident punctures; area of metathorax with very fine striz, _ appearing granular under a lens; posterior truncation not sharply defined; tegule brownish black; wings dusky hyaline, strongly iridescent; stigma and nervures piceous; second and XV, 3 Cockerell: The Black Halictine Bees 277 third submarginal cells about the same size, high and narrow, together smaller than first; first recurrent nervure meeting first transverse cubital; legs black, inner side of hind tarsi with bright ferruginous hair; hind spur pectinate; abdomen shining black, without evident punctures. The abdomen is like that of H. melanurops. Luzon, Mountain Province, Baguio (Baker). This species and the next two show a general relationship to some of the Australian forms. On the other hand, H. baguionis Crawf. is related to the Formosan H. perangulatus Ckll.; so we have in the Philippines a meeting place of northern and southern groups of the genus. Halictus caroli sp. nov. Female.—Length, 6 millimeters; black, with broad abdomen; hair of head and thorax white, abundant at sides of thorax and long on clypeus; antenne black, with basal half of scape red - in front; inner orbits rather strongly concave; face and front dullish; mesothorax and scutellum dull, without evident punc- tures; area of metathorax dull and granular, a little shining at apex, where there is a median sulcus; tegule pale testaceous; wings faintly dusky, nervures and stigma piceous; second and third sumarginal cells high and narrow, about equal; first re- current nervure joining second submarginal cell near end; legs black, small joints of tarsi ferruginous; abdomen black, shining, without bands or patches of tomentum; venter with a scopa of long curled hairs, tinged with ocherous at sides. PALAWAN, Puerto Princesa; sent by Prof, C. F. Baker, after whom it is named. Halictus megregori sp. nov. Female.—Length, 7 millimeters; black, with broad abdomen; hair of head and thorax scanty, but dense and dull white on tubercles, and sides of metathorax with a conspicuous white fringe; head broad, clypeus short; antennz black; supraclypeal area strongly convex; face and front moderately shining; me- sothorax and scutellum dull, without evident punctures; area of metathorax large, entirely opaque, faintly striate at sides toward base; posterior truncation hairy; tegule ferruginous; wings dusky, nervures and stigma piceous; second and third submarginal cells high and narrow; first recurrent nervure meeting second transverse cubital; legs black, hind tibie and tarsi with black hair on outer side and fulvous on inner; hind spur with a few large teeth; abdomen dullish, without band or 278 Philippine Journal of Science 1919 patches of tomentum; venter with a scopa of long curled hairs and large patches of pale fulvous hair at sides of segments. PANAY, Antique Province, Culasi (MeGregor), type, June 3, 1918. Luzon, Laguna Province, Mount Maquiling (Baker), 1. Halictus oligostictus sp. NOV. Female.—Length, nearly 8 millimeters; very robust; black, pubescence scanty, fringe of tubercles dense and tinged with ocherous; antenne black, the flagellum faintly reddish apically ; clypeus with large punctures; front dullish, finely punctured and substriate; mesothorax and scutellum dull, with large, . widely scattered punctures, very few on scutellum; area of me- tathorax dull, with strong longitudinal plice; posterior trun- cation finely tomentose, sharply defined above and at sides; prothorax with prominent tubercular and anterolateral angles; tegule black; wings slightly dusky, stigma and nervures pic- * eous; first recurrent nervure meeting second transverse cubital; third submarginal cell more than twice as long (broad) as second; legs black; hind tibiz and tarsi with black hair on outer side and pale yellowish hair on inner; hind spur with a few large spines; abdomen very broad, shining black, without evident punctures; second segment with a narrow patch of dull white tomentum on each side basally; venter with pale hair, but no long curled scopa or lateral patches. LuzoN, Mount Banahao (Baker). Halictus lionotulus sp. nov. Female.—Length, about 5.5 millimeters; black, with rather narrow thorax; hair of head and thorax dull white, not abund- ant; face narrowed below; clypeus and supraclypeal area shining, sides of face with dense grayish white hair; flagellum obscurely brown beneath; mesothorax and scutellum polished and shining, without conspicuous punctures; area of metathorax dull, with fine plicee, the rounded apical margin shining; posterior trun- cation very hairy; tegule bright fulvous; wings dusky, stigma and nervures dusky reddish brown, not dark; outer recurrent and transverse cubitals much weakened; type with only two submarginal cells, the second transverse cubital absent, but another female, evidently of the same species, has three sub- marginals, the third broader than the second, but not twice as broad (the male also has three submarginals) ; legs black or dark brown, with much white hair, that on hind tibiz en- tirely white; hind basitarsi with white hair, but with an orange brush at apex; mesopleura polished and shining; abdomen XV, 3 Cockerell: The Black Halictine Bees 279 shining, without evident punctures, hind margins of segments rather narrowly hyaline; the cotype (but not the type) shows some pale tomentum at extreme bases of second and third seg- ments; fifth segment covered with white hair; venter with stiff white hair, but no curled scopa or lateral patches. Male.—Length, about 5 millimeters; slenderer, flagellum not elongated, clypeus all black. LUZON, Mount Maquiling (Baker), 2 females and 1 male. The type is a female. Halictus scintillans sp. nov. Female.—Length, about 6 millimeters; like H. lionotulus, but larger, with the hind margins of the abdominal segments not hyaline, and the area of the metathorax without distinct strie, except at the extreme base, where they are very delicate. LuzoN, Mount Banahao (Baker), 1. It is quite likely that additional material will prove this to be only a local race or variety of H. lionotulus. Halictus seapalis sp. nov. Male.—Length, about 8 millimeters; black, robust, having the general form and superficial appearance of H. oligostictus; head very large, face extremely broad; cheeks very broad; man- dibles ferruginous, black at base and apex; face and front glistening, with short pale ocherous hair; labrum chestnut red, with a pair of prominent projecting tubercles; clypeus closely punctured, with a median smooth raised band; scape slender and curved, basal half red; a groove running between the pos- terior ocelli; mesothorax and scutellum polished and shining, the mesothorax with rather close delicate punctures, the scu- tellum very sparsely punctured; area of metathorax well defined, shining, its basal half with strong regular plicze; thorax with pale ocherous hair; mesopleura dullish, finely lineolate; tegule reddish; wings slightly dusky; stigma rather small, dark red- dish, nervures brown; first recurrent nervure joining second submarginal cell before its end; third submarginal cell very broad, much larger than second; outer nervures distinct; legs dark brown or nearly black, with pale ocherous hair, anterior knees red, tarsi ferruginous apically; abdomen broad, shining, first segment without evident punctures, second finely punc- tured; pale tomentum at sides of base of segments 2 to 4; apical plate reddish, very broad and rounded. Flagellum short and abdomen broad as in females. Luzon, Mount Banaho (Baker). A remarkable species. 280 Philippine Journal of Science 1919 Halictus postlucens sp. nov. Female.—Length, about 7 millimeters; black, with pale ocher- ous hair; head ordinary; face glistening, clypeus with a de- pression at apical middle; front dull, narrowly shining along orbits; antennz black; scutellum with fuscous hair, postscu- tellum with pale tomentum; mesothorax and scutellum shining, sparsely and finely punctured; area of metathorax poorly de- fined, glistening, with feeble, raised lines, and apically distinct traces of transverse striz; posterior truncation well defined, smooth and shining; tegulz piceous, with a reddish spot; wings dusky, stigma and nervures dull reddish, not dark; first recur- rent nervure joining basal corner of third submarginal cell; third submarginal cell considerably larger and broader than second, though second is fairly broad; legs black, with pale ocherous hair, tarsi reddened at apex; abdomen shining, with- out evident punctures; some pale tomentum at bases of second and following segments; venter with pale ocherous hair, not forming a long curled scopa or lateral patches. LuzON, Mountain Province, Baguio (Baker). Halictus fulvovittatus sp. nov. Female.—Length, 6 to 6.3 millimeters; black, the hind mar- gins of the abdominal segments broadly testaceous or pale golden, the bands shading into ferruginous along their anterior margins, the bands on first two segments broader in the middle _ than at the sides; hair of head and thorax white with a creamy tint, dense on upper border of prothorax laterally, on margins of tubercles, and on postscutellum, thin on face; head ordinary; mandibles dark red in middle; flagellum obscurely reddish beneath; middle of clypeus depressed and punctured; a shining line along anterior orbits; mesothorax and scutellum finely and rather closely punctured; area of metathorax finely but very dis- tinctly striate on basal half; posterior truncation with long hair and without sharp margins; tegule bright fulvous; wings faintly dusky, nervures and stigma dilute reddish sepia; first recur- rent nervure joining second submarginal cell near apex; third submarginal cell larger than second, but higher than broad; knees and small joints of tarsi ferruginous; hind tibize and tarsi dull reddish; hind spur with three stout teeth; abdomen broad, conspicuously thinly hairy, but without bands or patches of tomentum ; venter with short hair, but no curled scopa or lateral patches. Luzon, Laguna Province, Mount Maquiling (Baker), 2. XV, 8 Cockerell: The Black Halictine Bees 281 Halictus davaonis sp. nov. Female.—Length, about 6.5 millimeters; black, differing from H, fulvovitiatus thus: Head broader; area of metathorax with a double sculpture of exceedingly delicate striz all over and coarse rugez basally; legs black, with red knees; pale mar- ginal tegumentary bands on abdomen much narrower, those on first two segments not enlarged in middle. MINDANAO, Davao (Baker), 2. Closely related to the last, but surely a distinct species. Halictus melanurops sp. nov. Female.—Length, about 7 millimeters; black, with little hair; abdomen polished and shining, without bands or patches of tomentum; head broad; clypeus dull basally, apically shining and with large punctures; supraclypeal area dull; front dull, densely punctured; sides of vertex closely punctured; antennze black; cheeks striate, with a large polished shining space on lower part; mesothorax and scutellum dullish, finely punctured, the punctures very irregular on scutellum; area of metathorax longitudinally plicate; posterior truncation shining, not sharply bounded at sides; mesopleura dull; tegule black; wings dusky, stigma and nervures brown; first recurrent nervure meeting second transverse cubital; third submarginal cell not much larger than second, both rather broad; legs black; hind tibie with fuscous hair on outer side and white hair on inner, hair on inner side of hind basitarsi white; abdomen with stiff hair on venter, but no curled scopa or lateral patches; along the apical margin of the covered portion on dorsal segments 2 to 4 is a very fine fringe of plumose white hairs, forming a very delicate line. Male.—Very similar but smaller; clypeus all black; flagellum elongated and submoniliform, as in typical Halictus males; tarsi dark. Luzon, Mount Banahao (Baker), 1 female, type: Nueva Viz- caya Province, Imugan (Baker), 1 female: Mountain Province, Baguio (Baker), 1 male. THE SWARMING OF ANOPHELINE MOSQUITOES ' “By CHARLES S. BANKS Professor of Entomology, University of the Philippines The swarming of Culicidz has been reported from all parts of the world where these insects are found. Almost every rec- ord of such swarming has to do with members of the subfamily Culicine, as far as it is possible to ascertain. W. W. Smith says? that in New Zealand “a train passed through a wall of mosquitoes three quarters of a mile in length, twenty feet high and eighteen inches thick” and that this swarm was composed of Culex (Uranotenia) argyropus Walk. J. W. Douglas re- ports * a similar occurrence in London, where for ten days hordes of members of the genus Culex swarmed over the neigh- borhood and appeared like smoke when it issues from chimneys. They rose in the air for 10 meters and continually danced up and down in the twilight stillness. Holiday * records a like phenomenon in which Culicada nemorosa Meig. (synonym Culex detritus) was the species. F. V. Theobald® gives an account of Culex pipiens L. on the Downs, near Wye, England, “‘dancing in little clouds where they were sheltered from the wind.” Romolo Gessi Pascha® speaks of myriads of mosquitoes, which obscured the air at Meshra-el-Rek. Accounts similar to these have come from every Arctic ex- plorer; and persons who have returned from Juneau, Nome, and Skagway, as well as other parts of Alaska, and from northern Siberia including Kamchatka Peninsula, even north of the Arctic Circle, mention the immense swarms of mosquitoes that are to be encountered. It would seem from the reports that many of the swarms there were made up largely of females, which accords with my observations in northern New York in years gone by. The consensus seems to be that the swarms *From the entomological laboratory, College of Agriculture, University of the Philippines, Los Bajos. ? Ent. Mo. Mag. (1890) 321. * Ibid. (1895) 239. *The Entomologist 1 (1883) 151. * Monograph of the Culicide of the World 1 (1901) 73. *Seven Years in the Soudan (1892) 47. 283 284 Philippine Journal of Science 1919 seen in the warmer parts of the earth are largely composed of males. No mention seems to have been made, in accounts dealing with mosquito swarming, of Anophelines performing this in- teresting act. Two years ago this month (in March, 1917) having had occasion to descend Faculty Hill at the college cam- pus, at dusk, I encountered two or three distinct swarms of Myzomyia rossii Giles dancing in the air, at a height of about 25 meters above the roadway. Rapidly sweeping my hand through the swarm I caught a few specimens for the purpose of identification, as I did not observe at the time that they were Anophelines, not Culicines. A note of the occurrence was made, but no further observations were had and the matter was for- gotten until March 4 of this year (1919) when, upon returning home from my laboratory at 6.25 in the evening I encountered four distinct swarms within 60 meters of my house, which stands on a rise of ground at the foot of Faculty Hill. These swarms were hovering at about 1.5 meters above the roadway; their music was quite audible, and they were flying against a wind having a velocity of 12 to 15 kilometers an hour, maintaining themselves without the slightest difficulty at any point desired (as, for instance, directly overhead) and following me as I passed through the swarm. When my hand swept among them, in order to capture a few, the whole mass darted off to right or left, or up or down the road, with the greatest ease and with a decidedly concerted motion, returning when the disturbance ceased. Hastening to the house, I secured a net and returned to find that they were even closer to the ground than before, the base of the swarm being less than a meter from the roadway; but, as the entire swarm was dancing up and down, the individuals frequently rose to a height of 3 to 4 meters in the air. With a couple of sweeps of the net, I caught one hundred twenty-seven individuals out of two of the swarms and I es- timated that each swarm must have contained in the neighbor- hood of a thousand individuals. On the evening of March 5 the same phenomenon was wit- nessed about 100 meters farther up the same road, at precisely the same hour. The road at this point was well shaded on both sides by several high trees. The swarms in this case were very much larger than those seen the previous evening. They ex- tended higher into the air and seemed to dance with much more XV,3 Banks: Anopheline Mosquitoes 285 vigor, the wind at that time blowing at only about 9 to 10 kilo- meters an hour. Out of these swarms seven hundred ninety- four mosquitoes were captured, all males, though sweeps were made at the edges of the swarms in the hope that females might be lurking on the outskirts. As night fell very rapidly, it was impossible on either of these occasions to ascertain the presence of females by observing a different character of flight between the two sexes. It should be noted here that on the evening of the fifth a great number of females of this species was taken in the house at light, and they were also found in considerable abundance clustering and flying close to the nipa roof of the veranda, a favorite hiding place for Anophelines and a place where thou- sands of them must be captured nightly by a certain little spider, the habits of which are to be described elsewhere. Further observations, made on the evening of March 6 at the same hour, revealed other interesting and important fea- tures of this swarming of Anophelines. Having stationed myself at 6.15 in the evening, at the upper part of the road nearly opposite the Forestry Station, I noticed some half dozen males of M. rossii flying swiftly up and down the road, about 1.5 meters above it. Within five minutes several hundreds had assembled, and within five minutes more there were many thousands, grouped roughly into three distinct clusters about 8 to 10 meters apart, each cluster connected with the others by numerous stragglers. Selecting a favorable spot, which would place the swarms between me and the sky, which was then of a pinkish blue, I waited to see what would occur. At 6.30 the first female was seen to be caught by a male, and the pair flew slowly and obliquely upward and away from the swarm and was soon lost to sight. At intervals of about fifteen seconds other females were observed and this continued until 6.40; so that approximately fifty females were seen to enter the swarm and to be seized by males, each time the pair flying slowly but directly out of the swarm. Two pairs were seen to separate after about thirty seconds in copulation. At 6.45 the swarms had diminished more than half and at 6.50 only a few stragglers could be seen against the rapidly darken- ing sky. Whether the mosquitoes came from the thick undergrowth at both sides of the road, as is probable, or whether they flew up or down the road, it is impossible to say; but the rapidity 167148——5 226 Philippine Journal of Science 1919 of assembling and dispersing and the scant half hour during which the swarms were definite enough to be called such, might indicate a precise purpose and an exact time of the day for carrying out that purpose. On this afternoon a very mild breeze was blowing, but there were occasional gusts which seemed to stimulate the mosquitoes to greater activity and to cause them to bunch together with a very quick movement, as players in a foot-ball scrimmage. If I stood perfectly still beneath a swarm, it came close to my head; if, on the other hand, I even gently thrust my hand up- ward, the Anophelines rose en masse away from it. During the whole time of swarming, two dragon flies were darting in and out of the swarms, and each quite obviously caught a mosquito every time. The two remarkable features of these occurrences are that the insects are Anophelines, and that they were swarming during a very stiff breeze on each occasion. Many observers, entomologists as well as non-entomologists, have maintained that high winds are inimical to the welfare of the mosquito, and that the insects will not venture forth when strong winds are blowing. It has been repeatedly stated that when mosqui- toes are found at a distance from water, they have been wafted thither by gentle breezes. G. M. Giles, says:’ * * * as naturalists are generally agreed that gnats cannot travel to any considerable distance, it follows that food both for adult and larva must be obtained within a limited area, for mosquitoes cannot and do not fly far. It is impossible to fix any absolute limit to their powers in this respect, but it may be safely asserted that few individuals ever stray much more than a quarter of a mile from the pool in which their larval youth was passed, and the great majority never travel further than the nearest shady spot. Nor, in spite of popular beliefs to the contrary, can they be carried far by the wind. Mosquitoes, indeed, exhibit a well- founded, instinctive dread of boisterous weather, and will not leave shelter in a high wind. Those accidentally carried away are, I am inclined to think, rapidly disabled. Another reason that makes it impossible for Mosquitoes to be carried overseas any considerable distance by the wind is that, whatever may be the rate of travel that they can bear without injury, the entire journey must be made at night, for in tropical regions shelter from the sun during the day is a matter of life and death to a Mosquito * * *, For these reasons, we may, I think reject, as having no foundation in fact, such popular beliefs as that the swarms of Mosquitoes that sometimes appear on the Persian coast, have been carried by the wind 200 miles across the Gulf from the Arabian shore; albeit you must be prepared to hear this belief quoted as an established fact, even by European residents. ‘Gnats or Mosquitoes, ed. 2. London (1902) 112, 113. XV,8 Banks: Anopheline Mosquitoes 237 I feel very safe in asserting that, from the observations made upon these anopheline mosquitoes in their action of flying against a stiff wind, we. must certainly conclude that they have much greater power of flight than would be inferred from their apparently frail structure. If it be true that this species is so resistant to the wind—or rather, so capable of maintaining itself in safety in a high wind—what may not be expected of the more robust Culicines, such as Culex fatigans Wied., Manso- nia uniformis Theob., Stegomyia scutellaris Walk., and one or two other species, the occurrence of which in a given locality can only be explained by the fact that they must have “come upon the wind?” If Anophelines, naturally more delicate in structure and with narrower wings and slenderer legs than those mentioned above, can fly against a 12-kilometer wind and main- tain themselves at a given point, how much more easily could they or other mosquitoes be carried along by a wind, in which case they would simply be required to “give way” to its force and keep their balance! The ready ease with which these insects evade the captor’s hands and even his net would indicate that they possess in a very keen degree the percussion sense, that this would materially aid them in their windborne journeys in avoiding obstacles such as trees, and that they would find little or no incovenience in being wafted through wooded areas or around houses and other similar objects. The ability to fly against the wind, together with their sense of smell, would make it possible for them at any time in their journey to “drop out” of the wind current in which they might be carried, and attack any animal or enter any house in their path. This is always on the assumption that females as well as males are at times addicted to the swarming habit, and that they too would manifest the same activities as those of the opposite sex. There is certainly nothing to refute such a hypothesis. Many difficulties lie in the way of ascertaining precisely the minimum, or even optimum, distances which mosquitoes will fly; but it is certain that, at least in the Philippines, we must revise our ideas with respect to their being such frail creatures, not daring to venture forth except in a summer zephyr; and it is hoped that further observations along this line will enable us to say with greater certainty how far a mosquito can or will fly. 288 Philippine Journal of Science Molauin Creek, a rapid mountain stream, dividing the campus of the College of Agriculture, lies northwest by north of the point where the first swarms of Anopheline mosquitoes were seen. It is not more than 150 meters away, and is separated by a moderately thick grove containing many bamboos. The strong breeze mentioned is the daily tail end of the southeast monsoon of this time of the year in this locality; these facts might be adduced to show that neither the intervening woods nor the strong breezes would keep the mosquitoes from invading the college buildings as well as the houses higher up the hill, all of which lie to the windward of their breeding places. Malaria is most prevalent in this locality during the dry season; the creek furnishes an ideal breeding place for My- zomyia febrifera Banks during the entire dry period, and the range of activities of this pernicious little mosquito would, if it have the same powers of flight as M. rossii, appear to be limited only by two factors, gamete carriers and the number of individuals exposed. Ne eee. aia TWO PHILIPPINE LEAF-MINING BUPRESTIDS, ONE BEING NEW? By CHARLES 8. BANKS Professor of Entomology, University of the Philippines THREE PLATES Buprestide# are metallic beetles the larve, or grubs, of which are called flat-headed, or hammer-headed, borers. They are usually wood borers, but Kellogg ? states that the smaller species sometimes mine in leaves or live in galls. Comstock * says: In some of the smaller species the larvz are cylindrical, and are furnished with three pairs of legs. These are leaf miners; and in the adult state the body is much shorter than in the more typical species. It is true that the larve here under discussion are somewhat more cylindrical than is usual in this family, but there is ab- solutely no indication of legs, while the adults conform to the description given by Comstock. It may be that larve of Bu- prestide from other parts of the world are provided with legs, but Schiote‘* says of the buprestid larva “pedes nulli,” and his figures of the larva of Trachys minuta L. very strongly resem- ble those of this species, which is closely related. The occurrence of these larvee in leaf mines is so rare, how- ever, that a note concerning two species may be of interest, especially as the insects attack a plant of some economic im- portance because of its use as an ornamental. The two species under consideration are Endelus bakeri Kerrem.,® a small beetle not more than 4.5 millimeters in length and particularly noteworthy because of its prominent eyes, which are placed on conical, tubercular projections of the epi- cranium; and LH. calligraphus, a new species. From the department of entomology, College of Agriculture, University of the Philippines, Los Bajios. ? American Insects (American Nature Series) (1908) 266. * Manual for the Study of Insects (1896) 549. *De Metam. Eleuth. @bs., Naturh. Tidsskr. 6 (1876) 361, pl. 2, figs. 18-22. * Philip. Journ. Sci. § D 9 (1914) 88. 289 290 Philippine Journal of Science 1919 ENDELUS BAKERI KERREMANS The larve of the first species were found by one of my stu- dents, Mr. F. Q. Otanes, in blotch mines, singly, on the leaves of the bird’s-nest fern (Aspleniwm nidus L.) on January 27, 1918, and the adults emerged on February 25. Since they were first brought to my attention I have succeeded in discovering all stages of the insect as well as the larve, the pups, and the adults of the parasites that attack them. EGG The egg is a very flat, broadly elliptical object, with obtusely rounded ends. It is 1.75 millimeters long and 1.09 millimeters wide, dark chestnut brown over its discal area and pale to white around its margins, which are slightly crenulate, the submarginal area having a crinkled or undulated surface pre- ceded mediad by a series of broad, rectangular, radiate fascia more or less indistinct. Along the median line are four groups of black, imbricated, subspinose prominences and an area of brown striz which, being in subparallel series, may represent the area of the micropyle (Plate I, fig. 6). The egg is so firmly glued to the leaf that it remains long after the affected leaf has died and becomes completely desic- cated, and it serves as an excellent index to the beginning of a mine, where the battered condition of the leaf might make its discovery otherwise difficult. It is almost invariably laid on the lower surface of the leaf (Plate ITI, fig. 1) although it has been found on the upper surface of leaves near the center of the plant or where very little space intervenes between succes- sive leaves (Plate III, fig. 2). LARVA Length, 7.5 millimeters; greatest width, 1.8 millimeters; thick- ness, 0.9 to 1 millimeter. It is pale apple green, the head being ochraceous or pale buff and this color dilutes the pale green of prothorax and mesothorax. The anal segment is almost white, slightly tinged with buff. The body is shiny and very translu- cent, so that the body fats and internal organs are plainly visible. The head when at rest is retracted for four-fifths its length into the broad, lobate prothorax, and is regularly cordate in outline, with a very narrow, dark brown line around its anterior half. The frontal suture is dark, well defined, and bifurcate caudad. The trophi are ferruginous, the antennz short, stout and pale, with a stout, blunt terminal spine mediad and a slender seta, eight times as long, ectad (Plate J, fig. 3). XV, 3 Banks: Two Philippine Buprestids 991 The prothorax has the laterocephalic lobes strongly, roundly produced cephalad, their anterior margins being on a line with the tip of the mouth or a little beyond. Its integument is finely reticulate-punctate and its lateral and cephalic margins are sparsely white haired, especially the lobes. The mesothorax is two-thirds the length of the prothorax and is minutely punctate, these two segments being consequently dull glabrous. All the segments of the body caudad of the pro- thorax are strongly lobed laterad, except the twelfth, thirteenth, and fourteenth, the last being spherical; the other two, evenly rounded laterad; and the penultimate, five-eighths as wide as the antepenultimate. The caudal margins of the first to the sixth abdominal seg- ments are one-third to one-half longer than the cephalic mar- gins; hence the lateral lobes are obcordate in outline (Plate II, fig. 3) and the segmental incisions are very deep. At the base of each lateral lobe on the dorsum is a convex, lunate, longitudinal suleus which, with its neighbors, roughly outlines the lateral limit of the abdominal segmental articulations. On the ventrum the second to the eighth abdominal segments are provided with sublateral, circular pseudopodal papille, laterad of which appear sulci, as dorsad, but somewhat more profound (Plate II, fig. 4). In the last instar and previous to cessation of feeding, the larva measures 10 millimeters in length and is more buff or very dark cream, darker on the thoracic segments. The scu- tum and sternum of the prothorax are sharply outlined and darker and more chitinized than the surrounding integument. The anal segment is obtusely bilobed caudad. The abdominal pseudopodal papilla are more prominent and glabrously cor- rugated. The movements of the larva are very sluggish, both in the mine and when removed therefrom, and it has the peculiar habit of switching its tail from side to side in the gallery, thus leaving behind a fine string of excreta in a broad zigzag chain (Plate I, figs. 1, 2, and 8). PUPA The pupa just after formation is 6 millimeters long and 2.25 millimeters wide, and is greenish cream-white, glabrous, and subcutaneously opaque. It follows very closely the general form of the adult except that its abdomen is narrower, with the segmental articulations constricted somewhat as in the 292 Philippine Journal of Science 1919 full-grown larva. It is quite devoid of hairs, cilia, spines, or roughened or chitinized areas of any kind. Its skin is micro- scopically shagreened. ADULT The description of the adult is translated from Kerremans’s original. Endelus bakeri sp. nov. Of the group of EZ. weyersi Rits. and modiglianii Kerrem. related on account of the head and prolongation of the ocular tubes [cones] to HZ. dia- bolicus Kerrem., but differing from that species by the elytral impressions and by the general coloration. Short, pentagonal, entirely more or less bright bronze above; beneath almost black. Head wide, deeply excavated; eyes protruding very much ectad, emerging from short truncated tubes [cones]; the surface almost glossy and very lustrous. Pronotum very much wider than long, short and stout, truncated cephalad with the anterior angles sharp and projecting, the sides widely arched cephalad and sinuous caudad with the posterior angles obtuse; the middle of the disc with two wide transverse carinz. Scutellum small, triangular. Elytra short, stout, protruding at the shoulders, sinuous along their sides, attenuated caudad, individually widely rounded at their apices and minutely denticulated; the surface wavy and unequal, with wide lateral and discal impressions which are rounded, except those along the suture, on each side of the apex, which are elongated. Beneath smoother and more shiny than the elytra. Length 3.5-4; width 1.25-1.5 mm. Luzon, Los Bajios. These beautiful little purplish bronze beetles are especially noticeable by reason of the prominent, almost stalked eyes and the consequent excavation of the front of the head. The elytral undulations add to the brilliancy of the metallic color of which the legs, especially the femora, partake in a marked degree (Plate II, figs. 1, 5, 7, 10). They remain for a certain period within the pupal skin after having apparently acquired all adult characters, including hardening and brilliancy of the integument, and for a some- what shorter period in the mine. They fly rapidly, alighting after brief flights and running quickly over the leaf. Their copulation has not been observed nor has the egg-laying process been noted. MINES The eggshell can always be found at the beginning of the mine, which for a distance of 12 to 15 millimeters is narrow, usually not over 2 millimeters in width, and follows the leaf veins. It may then turn at an obtuse angle, continue for 15 ere XV, 3 Banks: Two Philippine Buprestids 293 to 25 millimeters more, and spread out as a blotch covering nearly the entire space between midrib and margin (Plate I, fig. 2). The tendency of the larva in mining is to proceed from the base to the apex of the leaf, but this is not the invariable course (Plate I, fig. 1). After it has made the main blotch mine, it invariably makes a second, very much smaller, chamber at a distance of from 10 to 15 millimeters, connected with the first by a gallery of an almost uniform width of 3 millimeters in nearly every instance. A third chamber may be made (Plate I, fig. 7), and even as many as four subsidiary chambers may be mined out by a single larva (Plate I, fig. 1) before it decides to pupate. It closes the entrance to this chamber by a lunate mass of compacted excrement (Plate I, figs. 2 and 3). Occa- sionally larve mining in opposite directions will cross each other’s connecting galleries and continue in their determined direction. The thin epidermis of the leaf soon dries and becomes torn, leaving very large, jagged, brown-bordered scars which detract from the beauty of this very ornamental plant. FEEDING HABITS OF THE ADULT The adults of Endelus bakeri feed on the upper sides of the leaves in a very peculiar manner. They first eat out an oval space, devouring all of the upper epidermis and the paren- chyma, leaving the lower epidermis entire. When this oval space is about the size of the body of the insect, instead of continuing to feed from outside the space, it deliberately set- tles down in the center of the denuded spot and proceeds with its meal from the bare white space. It is difficult to see why it should feed in this very strange manner, because its dark body is much more conspicuous against the white spot than against the darker green of the leaf. It will be seen by reference to Plate III, fig. 2, that the white spots are located near the apex of the leaf; but this is not always the case, although the majority of them are found on the apical half of it. When a beetle is feeding it is not easily frightened from its repast and the hottest sunshine does not seem to annoy the insect. PARASITES The larva of Endelus bakeri is parasitized by at least one species of chalcis fly, the female puncturing the epidermis of the leaf and laying her eggs through the dorsal skin of the 294. Philippine Journal of Science 1919 mesothoracic segment of the larva, leaving a black scar near its articulation with the prothorax. As many as eighteen punctures made by the ovipositor of the parasite were counted in the upper epidermis of a leaf over the fifth chamber of the larva. The larve of the parasite emerge from the pupa of the beetle and pupate around it, after the maner of many spe- cies of the Chalcidide. The adults emerge through numerous circular punctures which they chew in the leaf epidermis. As many as twelve exuvize of the parasite’s pupze have been counted around one beetle pupa carcass, and twelve adults were bred from one mine. Adults of the parasite are tiny, metallic, purplish blue-green insects, with brownish black eyes and snow-white, black-tipped tarsi. The body is regularly ornamented with snow-white hairs, which are also found on the antenne and legs. The length is about 1.25 millimeters. It is believed that the larve are also attacked by a fungus disease, the disintegrated remains of several larve giving evidence of some such attack. Numerous mites of an unde- termined species inhabit the leaf chamber, feeding on the débris left by the beetle larvz. INJURY BY THE BEETLE AND REMEDY FOR ITS ATTACK It is rare to find more than three larve in a single leaf, but each larva can do much toward injuring the leaves and thus spoiling the general appearance of the fern. The most ex- peditious method of combating the pest is to examine each leaf for the incipient mine and its occupant and, when located, simply to kill the larva by squeezing the spot between thumb and finger. The seasonal prevalence of the beetle is not known; but, from the appearance of a large number of ferns examined at the college, it is believed they do most of their damage in the period from December to March. One of my students states that the insect was abundant in July and August of 1917. In examining bird’s-nest ferns for further specimens of the blotch mine buprestid, Endelus bakeri, my attention was at- tracted to a short-leaved plant, perhaps another variety of Asplenium nidus, in which the mines were linear and very tor- tuous, as shown in Plate I, fig. 4. Specimens of these leaves, containing nearly full-grown larve, were kept in cages, and to my great delight two adults emerged which are entirely dif- ferent from EH. bakeri. The specimens, both of which are fe- XV, 3 Banks: Two Philippine Buprestids 295 males, prove to be a new species which I propose to call Endelus calligraphus from the great resemblance of the mines to hand- writing. Leaf specimens thus far examined show the eggs as being laid almost invariably near the tips, the larve working toward the petiole and then crossing and recrossing their older gal- leries, often returning as close to the leaf tip as possible, but always going back down the midrib for pupation in a very slightly enlarged chamber at the end of the gallery near the leaf base. This entire gallery may be made in that portion of the blade lying on one side of the midrib (Plate I, fig. 5). The larve succeed, with great skill, in keeping the gallery walls intact, as will be noticed, even when the turns lie very close together. But when they cross an old gallery it is usually at right angles and in a straight course. All the galleries, old and new, lie in the same plane, so there is no doubt that on crossing an old one they actually tap it and do not go above or below it as might be supposed. This is borne out by dissection of the leaf. The life history of this species has not yet been worked out. It will be noted in the figures in Plate II that the dark spots shown along the central line of the gallery indicate the distri- bution of the pellets of excreta, or frass. Their arrangement is much more regular than is possible in a blotch mine, the growth of which is far less regular. The total length of the gallery or mine shown in Plate I, fig. 5, is 89 centimeters, while the mine shown in Plate I, fig. 4, is 111 centimeters in length, the average being exactly 1 meter. Endelus and allied genera of Buprestide, for example, Apha- nisticus and Trachys, will probably be found to contain many species the larve of which are miners in the leaves of plants of the Asplenium nidus group and related genera of ferns. A NEW BUPRESTID Thus far but the two species of Endelus mentioned here have been recorded from the Philippines, but there are at least two other undescribed species in the college collection. Endelus calligraphus sp. nov. Entirely brown with uniform greenish bronze sheen over entire dorsum; head more brilliantly greenish bronze; ventrum a black bronze; legs similarly colored; eyes darker brown. Head and pronotum microscopically reticulate and with 296 Philippine Journal of Science 1919 coarse, sparse, shallow punctures; elytra coarsely undulate- punctate. Ventral surface and legs smooth. Head in dorsal view subglobose, front obtusely excavate (an- gle of 134°) (Plate II, figs. 2 and 12), frontal sulcus deep; eyes obtuse-ovate, occupying one-half the total width and three- fourths the length of head, in profile front strongly retreating so that trophi nearly touch prosternum; antenne with segments 5 to 11 progressively acutely serrate, finely setose, segments 1 and 2 ovate, glabrous, 3 slender, 1 with few setz, 2 and 3 bare; antennal groove somewhat sinuous (Plate II, fig. 6). Pronotum two and one-half times as wide as long, cephalic margin feebly sinuous, except laterad where it projects ce- phalad with an acutely rounded angle, lateral margins obliquely arched and somewhat convergent caudad to form a slightly ob- tuse angle with caudal margin; caudal margin nearly straight, but with a curved submedial emargination to fit elytra; disk with two large, shallow, transverse, submedian sulci, separated by an area one-half their width, caudolateral area shallowly excavate, cephalolateral area less so and separated from former by an oblique elevation which is continued mediad to form cau- dal border of the transverse sulci described above; scutellum small, subequilaterally triangular, depressed mediad (Plate II, fig. 2). Elytra very slightly wider than pronotum with sides parallel for a little more than five-eighths of their length, then evenly, roundly converging to their apices which are subsemicircular and with a few almost obsolete denticulations, humeral area with a prominent knob, apical area with a narrow, shallow obsolescent subsutural sulcus, each elytron with an obsolescent transverse depression at its first and second thirds; entire ely- tral area transversely confluently undulate, rugose or punctate- rugose. Legs bronze-brown, brilliant, glabrous, basal halves of pos- terior femora fit into well-defined transverse pits in the sub- lateral portions of the metasternum; femora lenticulately flat- tened, with shallow, wide sulcus on apical area, between which and the body the tibia and the tarsus are hidden when at rest; tarsi with golden to pale yellow ventral pilosity. Length, 3.65 millimeters; width at elytra, 1.50; length of head, 0.45; length of pronotum, 0.55; length of elytra, 2.65; length of scutellum, 0.25. Luzon, Laguna, Los Bafios, March 20, 1918 (Banks), two PO Fe OP ah ot ee OEP casey Ree * XV, 8 Banks: Two Philippine Buprestids 297 bred specimens. Type No. 18397 in College of Agriculture col- lection, Los Banos. This species approaches Endelus marseulii H. Deyr. in size; it differs from E., xthiops H. Deyr., which it closely resembles, in being bronzy brown, in the greater obliquity of the face, in the sides of the elytra being parallel for five-eighths of their length and lacking the strong impression and the large gibbosity at their bases. The specific name is suggested by the galleries made by the larvee in the leaves of Aspleniwm sp., which have the appearance of handwriting. A Us «f Remes 9 Fie. CONANT Rh wn Ee ILLUSTRATIONS [Drawings and photographs by the author.] PLATE I Five-chambered mine of Endelus bakeri Kerrem., in Asplenium leaf, showing crescentic mass of excreta used to close the entrance to chamber 5, and zigzag line of frass left by the larva. . Two-chambered mine of Endelus bakeri, showing two exit holes gnawed by an adult in emerging from the leaf. . Mine made in laboratory by Endelus bakeri, showing daily progress of a larva. Point 1 was reached two days after hatching; point 2, the next day; and so successively until one month thereafter point 12 was reached. . Leaf of Asplenium sp. with serpentine mine of Endelus calligraphus sp. nov., on both sides of the midrib. The total length of the mine is 111 centimeters. The dots represent excreta, and the enlarged chamber at the base of the leaf is the point of pupation, with the exit hole of the adult. . Leaf of Asplenium sp. with mine of Endelus calligraphus entirely on one side of the midrib. This mine was 89 centimeters long. Note the digression of the mine around a hole near the base of the leaf. Note also that the eggs were laid in these two leaves near the apex and the pupal chambers were made near the base. . Egg of Endelus bakeri, very greatly enlarged. . Unusual style of mine made by Endelus bakeri, in which the larva has turned back and excavated its pupal chamber near the point where the egg was laid. PLATE II . Endelus bakeri Kerrem., half outline of adult. Endelus calligraphus sp. nov., half outline of adult. . Endelus bakeri, larva, dorsal aspect. . Endelus bakeri, larva, ventral aspect of abdomen. . Endelus bakeri, adult, profile of head and thorax. . Endelus calligraphus, adult, profile of head and thorax. Endelus bakeri, adult, ventral aspect of head and thorax. Endelus calligraphus, adult, ventral aspect of head and thorax. Endelus calligraphus, adult, antenna. . Endelus bakeri, adult, antenna. . Diagram, showing facial angle of Endelus bakeri. . Diagram, showing facial angle of Endelus calligraphus. PLATE III . Leaf of Asplenium nidus L., showing eggs and blotch mines of Endelus bakeri. . Leaf of Asplenium nidus L., showing, in addition to mines, white spots caused by feeding of adult beetles of Endelus bakeri. 299 [Purir. Journ. Scr, XV, No, 3. BANKS: LEAF-MINING BUPRESTIDS.] EGG AND MINES OF ENDELUS SPECIES. PLATE I. BANKS: LEAF-MINING BUPRESTIDS, ] [Pumir, Journ. Scr, XV, No. 3. 9 / 12 PLATE Il. ENDELUS BAKERI AND E, CALLIGRAPHUS. BANKS: LEAF-MINING BUPRESTIDS.] [Purrr. Journ. Scr., XV, No. 38. PLATE IIl. LEAVES OF ASPLENIUM NIDUS L., SHOWING MINES OF ENDELUS BAKERI. NOTICES OF CERTAIN FULGOROIDEA, II: THE GENUS TROBOLOPHYA By C. F. BAKER Of the College of Agriculture, Los Baiios SIX TEXT FIGURES In 1913, Melichar' described a new genus, Trobolophya, of the family Dictyopharide, based upon a single specimen col- lected in Java by Jacobson. Just before this I had sent to Melichar a Philippine representative of the same genus, which he referred to as being the same as the Javan species.2 After 1913 I discovered that this genus is widely and commonly dis- tributed in the Philippines in several species, and that it occurs in western Malaysia as far north as Penang Island, Straits Settlements. In making a careful study of the accumulated material, it appears that the Javan species is quite distinct from any known Philippine form. It is hard to explain why so generally distributed a genus, with some of the species very common, remained unknown to science until 1913, unless it be due to the fact that the species are montane and arboreal; though this is only in so far as my ex- perience goes. mae External anatomical features in this genus are difficult to appraise b at their true comparative value ric. 1. «, Trobolophya philippinensis sp. since, as in many Derbide and std apa BF. AE BE. BOX, other fulgorids, high and thin : margins may incline one way or the other in drying, and partial collapse of certain areas may occur. It is thus very important to have ample series of specimens for study, and in most cases I have obtained these. The genitalia will, as in other Fulgoro- idea, undoubtedly furnish characters of great diagnostic value, but in this case would require dissection and special preparation * Notes Leyden Mus. 36 (1913) 92. * Loc. cit. 1671486 301 302 Philippine Journal of Science 1919 which would be a considerable undertaking in itself. The spe- cies so far recognized may be separated as follows: Synopsis of the species of Trobolophya. a‘, Intraocular portion of vertex, on median line, less than one-third length of anteocular; viewed from side, lateral margin of vertex highly arched at base; outer anteapical cell long and narrow, inner ante- apical three times as long as broad (fig. 1, b) tegmina transversely banded. ion ye EES “GER co oF eS CR Reese eR melichari sp. nov. a. Intraocular portion of vertex, on median line, about as long as ante- ocular; viewed from side, lateral margin of vertex not so highly arched at base; outer anteapical cell broad, inner little longer than broad (fig. 1, a); tegmina never transversely banded. b‘. Angle between lateral margins of vertex and front, as viewed from side, more than a right angle, and with the black spot roundish, not reaching eye. c’. Clypeus, in side view, strongly raised above the level of the front. penangensis sp. nov. c’. Clypeus, in side view, not or very little raised above level of front. d. Lateral spot of vertex large; veins of tegmina fuscopiceous, jacobsoni Mel. @. Lateral spot of vertex small; veins of tegmina piceous black. montana sp. nov. bd’. Angle between lateral margins of vertex and front, as viewed from side, a right angle or less, and with the black spot in the form of a large band, narrowing eyeward, and nearly or quite reaching eye. ce’. Vertex about as long as broad at base; smaller, paler species with Ny Eh, ay se a ey ee philippinensis sp. nov. ec’. Vertex distinctly longer than broad at base; larger, dark green species with piceous black veins...........0....... benguetensis sp. nov. Trobolophya melichari sp. nov.. Fig. 2, a, b.* Female.—Length, 5 millimeters. Pale greenish, legs stramin- eous, becoming entirely stramin- <—‘\ eous with a darker scutellum in @ drying. The lateral black marks on head show through on vertex O ag as two parallel lateral lines on \ apical half and as black spots at basal angles. Tegmina hyaline, a b veins fuscous; a narrow, irregu- Fic. 2. Trobolophya melichari sp. nov.: a, \2Y, curved, smoky band passes vertex, pronotum, and seutellum; 6, from stigma to near apex of. head, side view. clavus. : Lyzon, Laguna Province, Mount Maquiling: Tayabas Prov- ince, Malinao. MINDANAO, Davao (coll. Baker). Differs from *Characters given in the synopsis are not repeated in the following descriptions. XV, 3 Baker: The Genus Trobolophya 303 all other known species of the genus in form of head, venation, and coloration, and constitutes a distinct section of the genus. Named for Dr. L. Melichar, the founder of the genus. Trobolophya penangensis sp. nov. Fig. 3, a, b. Female.—Length, 6 millime- ters. Head and legs stramineous, thorax sanguineous, abdomen pale green; lateral spots of head scarcely showing through on ver- tex; anterior tibie with outer - nd border piceous. Tegmina hya- line, veins fuscous. Penang Island, Straits Settle- ments (coll. Baker). Near to b T. jacobsoni but distinct in form of clypeus and in coloration. se cedlee nekeeiberant cate &. Trobolophya jacobsoni Mel. head, side view. Melichar’s detail drawings (pl. 3, fig. 1, le) are very diagram- matic. For instance, the drawing shows the lateral margins of vertex parallel throughout, which is not true for any species of the genus. Also, the position of submedian keels of scutel- lum are never as shown. Melichar does not mention, or show in his drawing, that the narrow caudal appendage of scutellum is separated by a distinct transverse carina connecting the pos- terolateral margins of scutellum; the caudal moiety is more or less concave and possibly represents a postscutellum. The ver- tex is nearly as long as broad, if the length is measured on the median line only; but Melichar evidently measured it at the side, or length over all, since he says “vertex once and a half as long as wide at base.” Trobolophya montana sp. nov. Fig. 4, a, b. Female.—Length, 7 millime- ters. Body deep green, head and o> legs stramineous; frontal andscu- \— “| os tellar carine and outer margins \\¢~ _ \ 4 of fore tibie piceous; spots on oe Ee} > \>\ sides of head scarcely showing \\ = NS Mey through on vertex. Tegmina hyaline or slightly milky clouded, longitudinal veins piceous, costa frie. 4. Trobolophya montana sp. nov.; a, and cross veins pale fuscous. head, side'view; 6, vertex. Male.—Length, 6 millimeters. Coloring paler. Luzon, Nueva Vizcaya Province, Imugan (coll. Baker), 304 Philippine Journal of Science abundant. A slightly smaller, paler form is abundant on Mount Banahao, Laguna, Luzon; this form has also been encountered at Malinao, Tayabas, Luzon. Trobolophya philippinensis sp. nov. Fig. 5, a, b. Female.—Length, 5.5 millime- ters. Pale green, head and legs stramineous; frontal and scutellar carinz and outer margins of fore tibiz fuscous; spots on sides of head showing through as small spots on vertex at outer anterior angles. Tegmina hyaline; longi- Fic. 5. Trobolophya philippinensis sp. tudinal veins dark fuscous, darker nov. ; a, vertex, pronotum, and scutellum ; ze b, head, pronotum, and scutellum, side basally and on clavus, cross veins ver and costa paler. Male.—Length, 5 millimeters. Color paler. LUZON, Laguna Province, Mount Maquiling and Mount Ba- nahao (coll. Baker), common. A smaller, paler form occurs in the mountains of Benguet and Nueva Vizcaya. It is noteworthy that in the case of T. montana, a larger, deeper green form oc- curs in the northern mountains, and a smaller, paler form, on Maquiling and Banahao; while in the present species just the opposite is the case. This species is the one formerly referred to T. jacobsoni by Melichar. Trobolophya benguetensis sp. nov. Fig. 6, a, b. Female.—Length, 7 millime- ters. Body, including head, deep green, sometimes with sanguin- eous tinting between scutellar ca- rine; facial and scutellar carine concolorous; outer margin of fore tibiz piceous. Lateral spots of head showing through as narrow black marks on vertex at antero- a b lateral angles. Tegmina hyaline, ~~ 2 an pl ak Ge stigma stramineous ; veins, except heedl.« gide View: costa, all piceous-black, darker ’ basally. LUZON, Benguet Province, Baguio: Nueva Vizcaya Province, Imugan (coll. Baker). This species might be readily confused with T. montana on superficial examination, as the two are of similar size; but they can be readily distinguished ‘by the dif- ferent form of head and by the markings. CPTI gee ee eT ee 6, ae a4 ILLUSTRATIONS TEXT FIGURES Fic. 1. a, Trobolophya philippinensis sp. nov., tegmen; 6b, T. melichari sp. nov., tegmen. 2. Trobolophya melichari sp. nov.; a, vertex, pronotum, and scutellum; b, head, side view. 3. Trobolophya penangensis sp. nov.; a, vertex, pronotum, and scutel- lum; b, head, side view. 4. Trobolophya montana sp. nov.; a, head, side view; b, vertex. 5. Trobolophya philippinensis sp. nov.; a, vertex, pronotum, and scutel- lum; b, head, pronotum, and scutellum, side view. 6. Trobolophya benguetensis sp. nov.; a, vertex, pronotum, and scutel- lum; 6, head, side view. 305 SOY-SAUCE MANUFACTURING IN KWANGTUNG, CHINA! By EvizABeTH H. Grorr Of Canton, China SEVEN PLATES After a Chinese provides himself with rice, a little meat, and some vegetables, nothing is probably more important to him than the sauces which he eats with his food. In the book of Chau Lai Gi fi) .° the ceremonial rites of the Chau Dynasty (fl HY), written before 1000 B. C., we read that the king’s cook used one hundred twenty jars of sauces. Thus the writ- ten records of the Chinese show that they have been using these sauces for over three thousand years. Soy sauce, known among the Chinese as Ch‘au yau (Hf Wh), “drawing oil,” or pak yau ( yh), ‘white oil,” is without ques- tion the best liked and most widely used. Kwangtung Provy- ince is famous all over China for the soy sauce which it produces. Canton as its capital is naturally the center of all this trade. Wholesale shops can be found in great numbers, and prac- tically every food shop sells soy sauce. Each neighborhood also has its peddler who goes from door to door selling soy and other sauces. In Canton, jars of soy can always be seen in the making, as much of it is placed on the roofs to sun. Sainam (pq FB), “southwest,” a city of about 30,000 inhab- itants, 50 miles southwest of Canton on the Samshui (Qf¢ =), “Three Waters,” Railway, is famous for the excellent quality of soy that it produces. The first-class shops in Canton all have signs advertising Sainam Ch‘au yau (jG Hh yh), “Sainam soy sauce,” although most of this soy is made locally in their own establishments. The process of securing information on the making of soy sauce is lengthy and difficult, and accurate data can only be obtained after months of experiment carried on with the help of a workman who has grown up in one of these establishments. * All Chinese characters in the Cantonese are romanized according to the Eitel-Genaho Dictionary, but some of the diacritical marks are omitted. 2 Chau Lai is one of the thirteen classics of Confucius. 307 308 Philippine Journal of Science 1919 Frequently a soy-sauce manufacturer will smile at the ques- tions asked him and answer: “We pay workmen to tell us those things. Why don’t you do the same?” The methods of approaching Chinese with regard to the process of manufacture are intricate. Unlimited time, knowl- edge of Chinese customs, and courtesy are all important fac- tors to success. It is often necessary to work through three or four Chinese before the man can be secured who will properly introduce one to the manufacturer from whom accurate data are to be obtained. Many hours of friendly chatting over tea and cakes must be spent with these men before the business in hand can be approached. But when the final introduction takes place, one is treated as an old friend and every courtesy is tendered. The manager of a large sauce-manufacturing plant, On Shing Lung (# se RE), at “West Gate” (pf FY) Sai Mun, was ap- proached with this method. He has given full access to his books, allowed his head workmen to spend hours talking and answering questions, and has shown a keen interest in the in- vestigations. — THE EQUIPMENT OF A SOY BEAN MANUFACTURING PLANT Grounds and buildings.—The equipment in the largest estab- lishments is very meager and differs from the small establish- ments only in capacity. In the average factory about one-third of the ground is covered with one-story buildings, which are usually built of gray brick and roofed with Canton tile. They are divided into storerooms, boiling shed, workmen’s quarters, and mold room. The buildings usually surround the four sides of the plot, with a large court in the center for the sunning of the soy sauce. Boiling shed.—The shed in which the beans are boiled is large enough to store sufficient raw materials for one boiling and for the board on which the beans and flour are mixed. The iron pan tit wok (#4 i), in which the beans are boiled is semi-spherical, varies in size and pricegand is sold by weight at about 14 dollars local silver* for 100 catties* A pan large enough to boil 700 catties of beans can be purchased for about 100 dollars local silver. Its diameter is 52 inches, and its depth * Local silver, at the present rate (1918), is worth about 1.05 dollars for 1 dollar Hongkong currency; 1.30 dollars Hongkong currency are worth 1 dollar United States gold. “One catty is equal to one and one-third pounds. 0s aos utecith, & XV, 3 Groff: Soy Sauce in Kwangtung 809 is 82 inches. The brick oven constructed around this pan costs about 100 dollars local silver additional (Plate I). Mold room.—The mold room is usually placed to the north so that the doors which control the light and ventilation, very es- sential to good mold, can be opened to the south and plenty of sunlight allowed to enter. This room can be made almost com- pletely dark by the closing of these doors. It contains wooden racks on which the trays of beans and flour are placed to mold (Plate II, fig. 1). Jars.—Brown earthenware cylindrical jars known as Shiu- hing kong (4 J fiL)—-now no longer manufactured—are the ones preferably used to sun the beans, salt, and water (Plate V, fig. 2). The Tsinguen kong (¥ pf fil), a brown earthenware jar of inferior quality but with the same contour as the Shiu- hing jar, is now the only jar obtainable on the market. These jars are so named because they are manufactured in Shiuhing and Tsinguen, cities of Kwangtung. The Tsinguen jar leaks very easily, and the manufacturers prefer repaired Shiuhing jars to this inferior Tsinguen jar. The latter jar, 19.5 inches in diam- eter and 18.5 inches deep, with a capacity. of 180 catties, can be purchased for 1.70 dollars local silver. About one thousand of these jars are kept in the sunning yard at one time, although at times only about two-thirds of them are in use. The Chinese believe that the jars are greatly improved by long sunning, and when space permits the jars are allowed to sun in the court yard for months at a time. Racks.—The mold room is filled with crudely constructed wooden racks with horizontal partitions every 5 inches, on which are placed the trays of beans for molding (Plate II, fig. 2). Trays.—Two types of trays are used, the commonest being the circular bamboo tray, wo (#5), made in Canton (Plate VII, fig. 1). This tray can be purchased in any size, but the most commonly used for soy making is about 3 feet in diameter with a rim of 1.5 inches. This tray is in general use in Canton for the making of many different sauces and the drying of vege- tables. In order to save space many manufacturers construct light wooden frames with horizontal strips of bamboo placed close enough to support a very inferior quality of matting (Plate. IV, fig. 1). These trays are made so as to fit the racks snugly. They are usually about 5 feet 2 inches long and 3 feet 9 inches wide. The matting for this size of tray can be purchased for 60 cents local silver. This seems to be an excellent method, for the matting can be replaced when worn out. 310 Philippine Journal of Science 1919 Baskets.—The bamboo baskets, lo ( i), used to drain the beans after they are boiled are round at the top and narrow toward the bottom, which is flat and more square than round. Various sizes can be purchased, but the commonest are 15 inches deep. They cost about 65 cents local silver each (Plate IV, fig. 2). Covers.—Nothing is more important than a good cover with which to protect the sauce at night and when it rains (Plate V, fig. 1). A standard conical bamboo cover, 21 inches in diameter and 12 inches high, is used. These fit snugly over the top of the cylindrical jars and can be purchased for 35 cents.each. This cover is called tsim teng chuk lap (3z 3H 7 4), “pointed top bamboo.” Raw materials.—The raw materials used in the making of soy sauce are soy beans (Plate VI, fig. 1), flour, salt, and water. All of these are available in great quantity on the Canton market. The wong kam tau (iG 4 @), “yellow bean,” grown in Manchuria and known as coming from Ngau Chong (4 if), is recognized as the best variety of soy bean to be used. This bean is yellow and is slightly smaller than the American soup bean ; the outside coat is thick and tough and does not break apart easily after the bean is boiled. The Chinese consider this char- acteristic to be very important, for they wish to keep the bean as much intact as possible for the molding process. To fui min (+. JR Hi), local third wheat flour,” comes from Kwang- tung. The shaang im (+: fa), “raw salt,” comes from Tientsin. Method of mixing.—Soy sauce can be made in almost any quantity, but the beans mold much better and faster in large quantities. It is very difficult to obtain figures on the amounts of materials that are used. Each maker will tell you that he fills his iron sauce pan with beans and buys sufficient flour to mix with it and adds the salt solution at the proper time. In order to secure satisfactory data on the quantity of material used, it is necessary to be present when each process takes place. Boiling of beans.—One Canton manufacturer * purchases 1,400 catties of béans at one time, dividing them into two boilings. This amount of beans, together with 1,200 catties of flour, provides sufficient molded beans to stock thirty-six jars. The ‘beans are placed in a large iron pan (Plate I) and covered with about 1,100 catties of water. They are then boiled until soft. Care must be taken that the outside coat does not break. * The establishment referred to on page 308. ™ XV,3 Groff: Soy Sauce in Kwangtung 311 The length of time required to soften the beans depends en- tirely upon the amount:of heat applied. Another satisfactory process is to drop the beans into boiling water. The makers use both methods. If the beans are boiled constantly, three to four hours are sufficient for the softening process. However, in order to save time, many of the shops boil the beans at night, allowing the fire to die out and then removing the beans at about 4 o’clock in the morning. They are then placed in bam- boo baskets, allowed to drain, and become almost cold before being mixed with the flour (Plate IV, fig. 2). Mixing of beans and flowr.—The beans are then poured upon the mixing board, two baskets at a time. Two men stand, one on either side of the board, and thoroughly mix the beans and flour with their hands. Care must be taken that each bean is covered with flour.. This mixture of flour and beans is then placed on the trays to a thickness of about 1.5 inches, and the hand is used to furrow them so that they get proper ventilation (Plate II, fig. 2). The mixture remains in the mold room from one to two weeks, depending entirely upon the time of year and the weather. Much less time is needed in the hot rainy season. After the middle of November the manufacturing plants stop boiling beans and do not begin again until the spring season opens in February, as the weather during this period is unfavorable to produce the mold. After the beans are placed on the trays, they begin to mold in about three days. Mold.—The natural yellow mold is probably a species of Aspergillus, and the undesirable black is a Mucor. Care must be taken that only the yellow mold is used in the making of soy, and all black mold should be removed before placing in the jars of salt water (Plate VI, fig. 2). Method of sunning beans and flour—The mold from 1,400 catties of beans and 1,200 catties of flour is divided into thirty- six jars. A salt solution, of 150 catties of water to 40 catties of salt, is then poured into the jars until they are full. The salt solution is thoroughly cleaned before using by allowing the dirt to settle and then pouring off the solution. The jars of bean-flour mold and salt water are then placed in the yard to sun where they remain from two to six months (Plate V, fig. 2). The longer the period of time the better. Most of the shops, *These general determinations were made by Prof. C. W. Howard, biologist at the Canton Christian College, who is making cultures for further study. 312 Philippine Journal of Science 1919 however, make the first drawing in from three to four months. At night or when it rains, the jars must be covered with bam- boo covers. This sunning process results in an evaporation of the liquid in the jars; and three days before the drawing off of the soy, salt solution is used again to fill the jars. The first drawing is then made by siphon, kwo kong lung ( jh 4 We ) (Plate III, fig. 1). About 60 catties of the liquid are drawn off. This liquid is allowed to settle and is again drawn off, reducing the quantity to about 50 catties. It is then placed in clean jars and allowed to sun again for from one to six months. Some of this soy is at times allowed to sun for three years, but this is too expensive and is rarely done commercially. First drawing.—This soy is called teng ch‘au (1h Ff), “first drawing.” The material that remains in the jar is called teng shi (J 8%), “first salted,” and is sold as a separate sauce, used as the base of a number of different sauces or as the base for the “second drawing,” i ch‘au (— Hh). Second drawing.—A salt solution of 150 catties of water and 80 catties of salt is now poured on the teng shi, or the beans which remain in the jar from the first drawing. The jars are again placed in the sun for from one to two months. Salt water is again added three days before the drawing, after which the soy is drawn off, about 50 catties, after it is cleaned, and placed in the sun from one to two months. This is called i ch‘au (= Fh). “second drawing.” The material which remains in the jar is called tin shi (Ji #%), “beginning salted,” and is sold as a sauce for 4 cents per catty, used as the base of a number of other sauces and as the base for sdém ch‘au (= fh), “third drawing.” Third drawing.—The same method is used in the making of sém ch‘au as in i ch‘au and the material which remains is called chung shi (th 84), “middle salted.” This is sold as a sauce at 2 cents per catty, used as the base of a number of different sauces and as the base for sz ch‘au (PY Hi), fourth drawing.” Fourth drawing.—This is made the same as i ch‘au, sim ch‘au, and sz ch‘au, and the material left in the jars is also called chung shi (th $§) and is sold as a sauce at 1 cent per catty; it is used as the base of a number of very cheap sauces. Boiling method.—The sunning method takes so much time that many of the manufacturers boil the second, third, and fourth drawings instead of sunning them. This makes a decidedly XV,3 Grof: Soy Sauce in Kwangtung 313 inferior quality of soy, but it can be sold very cheaply. After the soy is drawn from the beans, it is placed in an iron pan and boiled from two to four hours. The longer the better, but it must be boiled at the least two hours or it will not keep. It is then taken off and allowed to cool and is ready for salt. The manufacturers never boil the first drawing. They always sun this and sell it for their finest grade. Mixing of soy.—It is interesting to note that of the four drawings of soy the only drawing that is sold as it is drawn is the teng ch‘au (iq fh), “first drawing.” The others are all mixed together and are sold under the names of the price they cost per catty. This mixing process is shown in the list of samples submitted. Prices of raw materials.—The prices of beans and flour vary considerably, but the soy beans, wong kam tau (i 4 DB), can be bought for about 6 cents per catty wholesale; the flour, t‘o fui min ( +. Jr Hi), for about 6.5 cents per catty; and the salt, shang im (/f Fai) , for about 4 cents per catty. The retail price is 8 cents per catty for the beans, 10 cents per catty for the flour, and 5 cents per catty for the salt. The manufacturers figure that the materials for one jar cost about 8.50 dollars. They expect to sell the different grades of soy and the beans left for about 14 dollars. This does not include labor, equipment, rent, etc. Candied molasses, kat shui (#4 7K), which is added to the very cheapest soy as a coloring and to sweeten it, costs about 8 cents per catty. Sainam soy.—Sainam, 50 miles distant from Canton on the Samshui Railway, is famous for its soy. The establishments there are much larger than in Canton, and a superior quality of soy is produced. There are eight factories, all of about the same capacity, doing a business of over 100,000 dollars a year. The methods used in Sainam appear to be the same as those used in Canton, the only noticeable difference being in the quality of materials used. The Sainam soy makers advise buying the very best quality of raw materials, while the Canton manu- facturers are content with materials of inferior grade. They believe that these inferior materials make just as good soy. The yards in Sainam also are very spacious, allowing the manu- facturers to sun their product to better advantage. Ground is much less costly, and hence the sunning process is not so ex- pensive. In one yard it was also noted that there were about five hundred jars not in use which were being sunned. Both 314 ' Philippine Journal of Science Canton and Sainam makers believe it is an excellent plan to sun jars. Limited space in Canton makes this impossible ex- cept on a small scale. Making soy from rice.—Many of the village people make their own soy from the rice that has stuck to the bottom of the ves- sel in which it is boiled. A handful of rice is pressed into a ball and sprinkled with hot water. These balls are placed in a covered jar and allowed to mold. In about two weeks these balls are placed in salt solution, using one part of salt to five of water. Two parts of rice are used to three parts of salt solu- tion. The more rice used the stronger the sauce will be. This is then sunned as the regular soy and produces a very inferior grade of rice soy, which is used by the poorer classes of Chinese as a substitute for the soy-bean sauce. Soy samples and prices gathered on the Canton market.— There is no soy standard, but the different grades of soy are known in most of the shops as follows: T* in teng ch‘au yau (K Wy Hh 4h), “best selected drawn oil.” Retailed at 40 cents local silver per catty. Made of teng ch‘au (ja fy), “first drawing,” which has been sunned four months after ‘drawing off.” Tsin pat ch‘au you ($$ A Hh wh). “fourteen cents per catty drawing oil.” This is made of the “first drawing” which has been sunned two months after “drawing off.” Kou luk ch‘au yau (AA Fh ih), “eleven cents per catty ch‘au yau.” Made of 50 per cent of the “first drawing” and 50 per cent of the “‘second drawing.” Sz pat ch‘au yau (PG A fii uh), “eight cents per catty te oil.” Made of 25 per cent each of “first drawing,” “second drawing,” “third drawing,” and “fourth drawing.” Sam luk chung ch‘au (= % Hh Hh). “six cents per catty middle drawing.” Made of 50 per cent salt solution, 50 per cent “third drawing” and “fourth drawing,” and colored with candied mo- lasses, kat shut. Sheung pak yau ( + | Wh), “upper white oil.” This is sold at 4 cents per catty and is made of 50 per cent salt and water and 50 per cent sz ch‘au, “fourth drawing,” and colored and sweetened with candied molasses, kat shui. Pat sin shing ch‘aw (A 4h 7E $i), “eight cents raw drawing.” This is made of equal parts of “first drawing” and “second draw- ing,” but after it has been drawn off the beans it is not boiled or sunned. This soy is used for soup and does not keep longer than about a week. ILLUSTRATIONS PLATE I The iron pan in which the soy beans for the making of soy sauce are boiled is used extensively in all sauce-manufacturing establish- ments; it varies greatly in size and is sold by weight. Note the bamboo basket fastened to a bamboo pole, by which the beans are removed from the pan. PLATE IT Fig. 1. The dark room in which the beans and flour used in the making of soy sauce are allowed to mold. The light and the ventilation, very essential to good mold, are controlled by doors, which open to the south. Trays of beans and flour now in the process of molding. The man in the picture is the head soy maker in this establishment. 2. The beans and flour, used in the making of soy, are thoroughly mixed and placed in rudely constructed wooden racks in the dark room for the molding process. PLATE III Fic. 1. Soy sauce being drawn from the beans by means of a siphon. Great care must be taken not to stir the beans. The material which remains in the jar is called teng shi, and is sold as a separate sauce or is used as the base of the second drawing of soy. 2. The first drawing of soy, t‘in teng chau yau, which is placed in clean jars and again sunned from two to six months. PLATE IV Fic. 1. This light wooden frame, with horizontal strips of bamboo placed close enough to support the matting on which the soy beans and fiour are placed for molding, is frequently used in place of the circular bamboo tray in order to conserve space. 2. The bamboo baskets, in which the boiled soy beans are placed to cool and drain, are round at the top and narrower toward the bottom, which is flat and almost square. PLATE V Fig. 1. The conical bamboo covers, which are used to protect the jars at night and when it rains. 2. Rows of the standard-sized soy-sauce jars filled with the molded beans and flour and salt water. These remain in this position from two to six months. Note the conical bamboo covers in position for the quick covering of the jars in case of rain. 315 316 Philippine Journal of Science PLATE VI Fic. 1. The soy beans shown here are, natural size; they are yellow and have a thick outer coat. Great care is taken that the beans do not become mashed. 2. Beans and flour, which have been in the mold room for five days. The natural yellow mold is probably a species of Aspergillus, and the undesirable black mold is Mucor. PLATE VII Fic. 1. The circular bamboo tray that is commonly used in the making of soy and many other sauces and in drying vegetables. The trays have been stacked and are ready for removal to a storage shed. 2. Soy sauce ready for shipment to northern China. It is placed in sealed earthenware jars, which are incased in bamboo holders. Oe I te ee — - eS — 2 GROFF: Soy SAUCE. ] [PHILIP. JouRN. Scr., XV, No. 3. ae ilies PLATE |. THE IRON PAN IN WHICH THE SOY BEANS ARE BOILED. GrRorr: Soy SAuCE.] [Puiuir. Journ. Sci., XV, No. 3. Fig. 1. The dark room, Fig. 2. Racks in the dark room. PLATE Il. > > No. 3. Sci., XV, JOURN. [ PHILIP. J Soy SAUCE ROFF = ~ x C Fig. 1. Soy sauce being drawn by means of a siphon. Fig. 2. The first drawing of soy. PLATE lll. Grorr: Soy SAUCE.] [PuHiLie. JouRN. Sci., XV, No. 3. Fig. 1. A light frame, used in making soy. fire DERE” Fig. 2. Bamboo baskets, in which boiled beans are cooled and drained. PLATE IV. GROFF: Soy SAUCE. ] [PHIuip. Journ. Scr., XV, No. 3. Fig. 1. Conical bamboo covers. Fig. 2. Soy-sauce jars. PLATE V. Fig. 1. Soy bean in the mold room Fig. 2. Soy beans an d flour, five days PLATE VI. Grorr: Soy SAUCE. ] [Puiutre. JourRN. Scr., XV, No. 3. ougere ho eat t t ¥ee< me |S 2, ~ Fig. 2. Soy sauce ready for shipment. PLATE VI. REVIEWS The | Medical Clinics | of | North America | November, 1918 | published bi-monthly by | W. B. Saunders Company | Philadelphia and London | Pp. 645-920. The Philadelphia Number, Volume II, No. 3, contains the following papers: The influenza epidemics of 1889 and 1918, by Dr. Alfrea Stengel. Influenza and some of its complications, by Dr. H. R. M. Landis. The surgical complications and sequele of influenza, by Dr. John B. Deaver. Bacteriologic study of sputum in the recent epidemic, by Dr. Charles W. Burr. Bacteriology of influenza, by Lieut. Eugene A. Case. Nose, throat, and ear affections complicating or following the recent epidemic of so-called influenza, with a ventured interpretation of their significance, by Dr. J. Leslie Davis. Influenza in children, by Dr. Maurice Ostheimer. Feeding babies during their second year, by Dr. Maurice Ostheimer. Sciatica, by Dr. Thomas McCrae. Intraspinal therapy in syphilis, by Dr. Jay Frank Schamberg and Dr. Albert Strickler. Chylothorax, by Elmer H. Funk. Aortic aneurysm with esophageal rupture, by Elmer H. Funk. Tuberculosis and pregnancy, by Elmer H. Funk. Medical treatment of biliary affections, by Dr. Martin E. Rehfuss. Dilatation of the colon in children, with especial reference to the idiopathic form, by Dr. J. P. Crozer Griffith. Cerebral palsies of children, by Dr. Charles 8S. Potts. Diabetes, by Dr. Leon Jonas. X-ray diagnosis of lung diseases, by Dr. David R. Bowen. Physiologic psychiatry, by Dr. S. D. W. Ludlum. Influenza—Remarks upon symptoms, prevention, and treatment, by Dr. David Riesman. Volume I September—December, 1918 Number 1 | Quarterly Medlcal Clinics | A Series of Consecutive Clinical Demonstrations and Lectures | by | Frank Smithies, M. D., F. A. C. P. | [four lines of titles] | August- ana Hospital | Chicago | published by | Medicine and Surgery Pub- lishing Company, Inc. | Metropolitan Building | St. Louis | Paper, pp. 1-188, $1.50. FROM THE PREFACE For some time, it has been the practice to have a clerk report my Clinics and Lectures given at Augustana Hospital to the Senior Students of the School of Medicine of the University of Illinois. The notes taken by the clerk have been edited, 1671487 317 318 Philippine Journal of Science mimeographed and given to the students at each succeeding Clinic. This procedure has proved more valuable than when students themselves made notes. It has also enabled me to keep a record of the subjects presented, has avoided repetition and furnished a definite material from which examination questions might be selected. Frequently, students and visiting physicians have suggested that the Clinics and Lectures be issued in consecutive order and in form more substantial than loose mimeographed sheets. Dif- ficulties attendant upon such arrangement finally have been overcome and this volume contains the bulk of the clinical matter presented during the past three months. It is hoped that each quarter similar collections of Clinics and Lectures | may be issued and that the venture may prove as useful to medical men, generally, as it has proved already to a limited group of students. Ultra Violet Rays In | Modern Dermatology | Including the evolution of artificial | light rays and therapeutic | technique | by | Ralph Bern- stein, M. D. | Philadelphia, Pa. | [eighteen lines of titles] | author of | “Elementary dermatology”—numerous | brochures on skin diseases, etc. | illustrated | Achey & Gorrecht | 5-9 North Queen Street | Lancaster, Pennsylvania | Cloth, pp. i-xiii + 1-162. FROM THE PREFACE The efficacy of ultra violet rays in the practice of dermatology is now thoroughly established, and its bactericidal action, its anti-pruritic and analgesic effects, as well as its reconstructive action upon epidermal cells and its constitutional effect upon the general economy, cannot now be denied. The author has been prompted to write this book because no American author has heretofore attempted it, the profession relying entirely upon the writings of their European colleagues. It is the author’s intention to present the subject as briefly as possible with clearness, and with no intention of going into histo-pathologic details. The mere thought in mind will be to give in a practical way to the profession the results of his clinical experience with the use of ultra violet radiations in the treatment of the various skin diseases for which he has found the ultra violet light a desirable agent. The author has thoroughly gleaned the literature of the day and hereby gives credit for the material which he has used therefrom to the foreign writings of Morris, Dore, Bach, Wag- ner, Kruger, and Blaschko, and to the following writers from our own continent: Martin, MacKee, Jordan, Plank, Collins, and Allen. THE PHILIPPINE JOURNAL OF SCIENCE VoL. XV OCTOBER, 1919 No. 4 FORMOSAN TERMITES AND METHODS OF PREVENTING THEIR DAMAGE By MASAMITSU OSHIMA Of the Institute of Science, Government of Formosa THIRTEEN PLATES AND FIVE TEXT FIGURES CONTENTS INTRODUCTION. DESCRIPTIONS OF THE FORMOSAN TERMITES INJURIOUS TO WOODEN STRUCTURES. CERTAIN HABITS OF COPTOTERMES FORMOSANUS SHIRAKI. Different castes in the colony. The foundation of a new colony. Situation of the nest. Damage to buildings and other materials. PRINCIPAL Foop oF COPTOTERMES FOR- MOSANUS SHIRAKI. TERMITE-PROOF BUILDING CONSTRUC- TION. DEFECTS OF THE TERMITE-PROOF BUILDING CONSTRUCTION. TESTS OF THE RELATIVE RESISTANCE or NATIVE AND Exotic Woops. RELATIONSHIP BETWEEN THE RESIST- ANCE AND THE PHYSICAL PROP- ERTIES OF TIMBER. RELATIONSHIP BETWEEN THE RESIST- ANCE AND THE CHEMICAL PROP- ERTIES OF TIMBER. THE VOLATILE CONSTITUENTS OF CYPRESS PINE. THE VOLATILE CONSTITUENTS OF TEAK. THE VOLATILE CONSTITUENTS OF FoocHow CEDAR AND’ RANDAI CEDAR. STuDIES ON CAMPHOR GREEN OIL. THE RELATIVE EFFECTIVENESS OF PREVENTIVES. SUMMARY, INTRODUCTION One of the most serious problems in the Tropics with regard sie lla elas to man’s industry is to discover a method of preventing the damage caused by termites. Because of their subterranean habits and insidious methods of attack, termites are very dif- ficult to destroy. Moreover, as stated by Dr. K. Escherich:' “‘Nichts ist vor ihren Kiefern und zerstérenden Sekreten sicher *Die Termiten (1909). 167553 : 319 320 Philippine Journal of Science 1919 ausser Eisen und Stein.” Such being the case, not only is the extermination of these insects almost impossible, but also pre- ventive measures against their damage are difficult to apply. Fortunately, Japan is located in a temperate region; and, therefore, her people have not been obliged to pay attention to these formidable pests. However, one species, Leucotermes speratus Kolbe, which very often causes somewhat serious dam- age to wooden structures, has been recorded in Japan since 1724. About twenty years ago, Japan occupied Formosa, which lies in a semitropical region; that is, in the western Pacific Ocean, between the southern and eastern China Seas. In this possession the people have been compelled to fight against the common pest of the Tropics, and the investigation of the biology of termites has become one of the most important prob- lems of the architect and the entomologist. In Formosa and in Japan there are fourteen species of ter- mites, four of which, namely Coptotermes formosanus, Leuco- termes speratus, Leucotermes flaviceps, and Odontotermes for- mosanus, are known as pests of wooden structures. Coptotermes formosanus, which is distributed in Formosa, Riu Kiu Islands, and in the southern parts of Japan proper, is especially formid- able to buildings. It is certain that the other three attack woodwork, and wooden structures as well, but their ravages are negligible in comparison with those of Coptotermes for- mosanus. During the last ten years, I have been investigating the Japanese termites, especially the habits of Coptotermes formo- sanus. I approached the problem of the method for preven- tion with the following ideas as a working basis: | 1. Some changes are necessary with regard to the construction of build- ings in the Tropics in order to prevent damage by termites. 2. The value of termite-proof building construction is not absolute, unless all sorts of nonresistant timbers are eliminated from the building materials. 3. If it shall be proved that the elimination of nonresistant timbers is practically impossible, it becomes necessary to treat them chemi- cally or physically in order to confer a special resistant property. 4. It is necessary to prove whether or not there are naturally resistant timbers in the Tropics. 5. If there are naturally resistant timbers, an investigation with the object of discovering the cause of resistance becomes important. 6. If the causes are definitely known, methods of artificially treating nonresistant timbers will be more easily discovered. By the kind and valuable assistance of Mr. Kinzo Kafuku, former expert chemist of the Government Institute of Science, fe Se XV,4 Oshima: Formosan Termites 321 Formosa; Mr. Tetsukichi Katayama, chief of the chemical de- partment of the same institute; and Mr. Saichi Tasaki, my assistant, I have carried on my investigations and very fortu- nately have gained some satisfactory results, which are recorded in the following pages. Here I wish to express my sincere thanks for the courtesy shown by those gentlemen and to Dr. Tomoe Takaki, former director of the above-mentioned in- stitute, through whose most courteous assistance I have been able to continue my work for so many years. DESCRIPTIONS OF THE FORMOSAN TERMITES INJURIOUS TO WOODEN STRUCTURES Family MESOTERMITIDA® Holmgren Subfamily COPTOTERMITIN4A Holmgren Genus COPTOTERMES Wasmann Coptotermes formosanus Shiraki. Plate I, figs. 1 to 3. Coptotermes formosanus SHIRAKI, Trans. Ent. Soc. Jap. 2 (1909) 239; OsHIMA, Rep. Term. 1 (1909) 33, pl. 1, figs. 1-3; pl. 2, figs. 11, 12; Rep. Term. 3 (1912) 75, pl. 1, figs. 8, 28; pl. 2, figs. 3, 21; Philip. Journ. Sci. § D 8 (1913) 276; Rep. Term. 4 (1914) 2; HOLMGREN, Termitenstudien 4 (1913) 76, pl. 2, fig. 10; Hozawa, Journ. Coll. Sci. Tokyo 35 (1915) 92, pl. 3, figs. 18-20, text figs. 22-25. Coptotermes gestroi OSHIMA, Zool. Mag. Tokyo 22 (1910) 376; Rep. Term. 2 (1911) 5; Nawa, Insect World 14 (1910) 597. Coptotermes formosae HOLMGREN, Termitenleben auf Ceylon (1911) 192; Termitenstudien 2 (1911) 74; Annot. Zool. Jap. 8 (1912) 121; YANo, Rep. Jap. For. Exp. Station 9 (1911) 62, pl. 4, figs. 12-20. Imago.—Head brown; anteclypeus whitish; postclypeus, an- tennez, labrum, and labial palpi brownish yellow; pronotum yellowish brown, with Y-shaped marking; mesonotum, meta- notum, and abdomen pale brown; legs yellow, tibia and tarsus darker; head, thoracic plates, and abdominal tergites densely pilose; wing stumps beset with hairs. Head round, fontanelle distinct; antennze 21-jointed, second joint cylindrical, nearly as long as third and fourth taken to- gether; third joint short, ring-shaped; eye round, markedly prominent; ocellus oval, separated from eye by a distance less than its shorter diameter; anteclypeus trapezoidal, narrower than postclypeus, the latter short and slightly swollen; pro- notum semilunar, anterior border concave, posterior border 322 Philippine Journal of Science 1919 bilobed, posterolateral corners rounded; mesonotum and meta- notum narrower than pronotum, their posterior borders slightly concave; wing stumps subequal in size, anterior ones overlapping base of posterior; wings hyaline, costal margin yellowish, veins yellowish at base; wing membrane densely beset with minute hairs, radius nerve of anterior wing parallel to costal margin; median nerve faint, nearer to cubitus than to radius, with no branch; cubitus below middle of wing, not reaching tip, with about seven branches, of which the proximal four are stronger; median nerve of posterior wing starts from proximal portion of radius. Measurements of the imago. mm. Length of body with wings 14.00 Length of body without wings 6.00 Length of head 1.40 Width of head 1.31 Width of pronotum 1.31 Length of pronotum : 0.87 Length of anterior wing 11.00 Queen.—Head and thorax similar to those of imago; meso- notum and metanotum with a pair of triangular wing stumps; abdomen much enlarged; abdominal tergites pale reddish brown, separated from each other by a distance about four times as great as its length; abdominal integument milk white. Measurements of the queen. mm. Length of body 21.00 Length of abdomen 19.00 Width of abdomen 5.50 King.—All the characters agree very well with those of the imago; abdomen not enlarged; thorax provided with two pairs of triangular wing stumps. Substitute queen.—Head yellow; postclypeus somewhat paler; anteclypeus whitish; mandibles pale yellow, with brown tips; antenne and legs pale yellow; thoracic plates yellow; abdomen milk white, with pale yellow abdominal tergites; head and thoracic plates very coarsely beset with minute hairs; abdominal tergites smooth. Head round, Y-suture distinct, whitish; fontanelle round, whitish, situated at the center of the suture; eye round, with no pigment, slightly prominent; ocellus round, whitish, separated from eye by a distance greater than its diameter; postclypeus trapezoidal, more than twice as broad as long; anteclypeus slightly swollen; labrum tongue-shaped, entirely overlapping Retanoe aia winter ane ee Ar NS cali = aS BL ee : ee Pa ime MA cst te (» Sa XV,4 Oshima: Formosan Termites 823 mandibles; antenne 17-jointed, second to fourth joints equal in length; pronotum semilunar, anterior border bilobed, posterior border clearly indented at middle, anterolateral corners rounded; mesonotum and metanotum much broader than pronotum, their posterior borders nearly straight, their posterolateral corners strongly projected posteriorly, forming rudimental wing pads; abdomen more or less enlarged, chitinous plates crescent-shaped. Measurements of the substitute queen. mm. Length of body 12.00 Length of abdomen 9.00 Width of abdomen 3.50 Length of head 1.25 Width of head 1.34 Width of pronotum 1.25 Length of pronotum 0.78 Substitute king—Unknown. Soldier.—Head yellow; mandibles dark brown; labrum brown- ish yellow; antennze somewhat paler; pronotum pale yellow; abdomen and legs pale straw color; head coarsely pilose; tho- racic plates hairy; abdominal tergites densely covered with short spiny hairs. Head suborbicular, posterior border rounded, sides arcuate, strongly converging anteriorly; fontanelle distinct, opening directed forward, situated just behind the base of postclypeus; antennz 15-jointed, second joint quadrate, longer than third, the latter narrowest and shorter than fourth; mandibles saber- shaped, slender, with incurved, piercing tip, cutting edge tooth- less, labrum lancet-shaped, tip hyaline, reaching middle of mandible; anterior border of anteclypeus slightly convex; post- clypeus very short, not separated from forehead; pronotum sub- reniform, narrower than head, anterior and posterior borders slightly bilobed, lateral borders rounded, converging posteriorly ; mesonotum oval, the former slightly narrower than pronotum, while the latter is broader. Measurements of the soldier. mm. Length of body 5.00 Length of head with mandibles 231 Length of head without mandibles 1.56 Width of head 1.18 Width of pronotum 0.84 Length of pronotum 0.50 Worker.—Head, thorax, and abdomen milk white, densely covered with short spiny hairs. 324 Philippine Journal of Science 1919 Head spherical; antennz 15- or 16-jointed, second joint longer than third, fourth joint ring-shaped; clypeus trapezoidal, an- terior border straight, boundary between anteclypeus and postelypeus indistinct; pronotum much narrower than head, semilunar. Measurements of the worker. mm. Length of body 4.50 Width of head 1.15 Width of pronotum 1.00 Distribution.—South China (Foochow and Amoy) ; Formosa; Kotosho (Botel Tobago Island) ; Riu Kiu Islands; Shikoku; Kiu- shiu; southern part of Honshu, chiefly the coasts along the Inland Sea. Subfamily LEUCOTERMITIN 4 Holmgren Genus LEUCOTERMES Silvestri Leucotermes (Reticulitermes) flaviceps Oshima. Termes (Leucotermes) flavipes OSHIMA, Rep. Term. 1 (1909) 30, pl. 1, figs. 4-8; Zool. Mag. Tokyo 22 (1909) 345. Leucotermes flavipes SHIRAKI, Trans. Ent. Soc. Jap. (1909) 231; OSHIMA, Rep. Term. 2 (1911) 3; Nawa, Insect World 15 (1911) 14. Leucotermes flaviceps OSHIMA, Rep. Term. 3 (1912) 74, pl. 1, fig. 10; pl. 2, figs. 15-17; Philip. Journ. Sci. § D 8 (1913) 277. Leucotermes (Reticulitermes) flaviceps OSHIMA, Rep. Term. 4 (1914) os Imago.—Head chestnut brown, antenne and abdomen some- what darker; pronotum yellow; mesonotum and metanotum dark brown; femur dark; tibia and tarsus yellow; head, thorax, and abdomen densely covered with hairs. Head quadrate, posterior border rounded; clypeus yellow, nearly twice as broad as long, slightly swollen, boundary between anteclypeus and postclypeus indistinct; labrum tongue-shaped, as long as broad; eyes rather small; ocellus separated from eye by a distance equal to its diameter; antennz very long, nearly twice as long as length of head, 17-jointed, second joint as long as fourth, third joint smallest, shorter than second; pronotum narrower than head, semilunar, anterior border nearly straight, posterior border rounded, weakly incurved at middle; posterior border of mesonotum and metanotum rounded; wing stumps subequal, the anterior scarcely covering base of posterior; wings fuscous, with brownish costal margin; radius nerve of anterior wings near and parallel to costal not branched; median nerve midway between radius and cubitus, connected with the former - i+ (~ 7 2 XV, 4 Oshima: Formosan Termites 825 by many short branches, cubitus with about ten branches, not reaching tip of wing, its tip anastomosing and connecting with median nerve. Measurements of the imago. mm. Length of body with wings 9.00 Length of body without wings 4.50 Length of head 1.03 Width of head 1.00 Width of pronotum 0.90 Length of pronotum 0.59 Length of anterior wing 8.00 Soldier.—Head yellow; mandibles reddish brown; thorax and abdomen yellowish white; antennz and legs straw-colored; head moderately pilose, thorax densely provided with short, spiny hairs; abdominal tergites densely pilose. Head cylindrical, posterior border rounded, anterior border nearly straight, sides straight and parallel; clypeus very short, indistinctly separated from forehead; labrum lancet-shaped, tip whitish, scarcely reaching middle of mandibles; mandible saber- shaped, tip incurved, cutting margin smooth; antenne 14- to 16-jointed; third joint smallest, nearly half as long as second; fourth joint slightly shorter than second; fontanelle indistinct; pronotum heart-shaped, anterior border bilobed, posterior border nearly straight, sides converging posteriorly, mesonotum and metanotum narrower than pronotum. Measurements of the soldier. mm. Length of head 5.00 Length of head with mandibles 2.50 Length of head without mandibles 1.75 Width of head 1.03 Width of pronotum 0.78 Length of pronotum 0.47 Worker.—Head yellowish white; thorax, abdomen, antennae, and legs whitish; head, thorax, and abdominal tergites densely pilose. Head round, slightly widening anteriorly; length of post- clypeus less than half its width; antennz 14-jointed, second joint nearly as long as third, fourth joint as long as third; pronotum narrower than head, anterior border bilobed, slightly raised, posterior border nearly straight, sides converging pos- teriorly; mesonotum nearly as broad as pronotum; metanotum much broader than mesonotum. 326 Philippine Journal of Science 1919 Measurements of the worker. mm. Length of head 4.00 Width of head 1.03 Width of pronotum 0.69 Distribution—Formosa (Taihoku, Horisha, and Koshun) ; Kotosho (Botel Tobago Island). Remarks.—The present species is very closely related to Leuco- termes speratus, from Japan proper. It is very hard to draw a fast line between the imagoes of the two species; however, the soldiers of the two species distinctly differ as follows: a’, Head narrow and long, 1.75 to 1.81 millimeters long, 1.03 to 1.09 milli- meters broad; lateral borders of head straight and parallel; posterior border of pronotum nearly straight.............-.-200222022..-.1-- L. flaviceps. a’, Head rather broad and short, 1.56 to 1.71 millimeters long, 1.06 to 1.09 millimeters broad; lateral borders of head convex, slightly con- verging anteriorly; posterior border of pronotum clearly indented OG ied oi ee a a a see L. speratus. In addition, the swarming seasons of the two species are quite different. The winged forms of L. flaviceps swarm in the beginning of January, while those of L. speratus swarm in April or May. Family METATERMITIDZ Holmgren Genus ODONTOTERMES Holmgren Odontotermes (Cyclotermes) formosanus (Shiraki). Plate I, figs. 4 to 6. Termes formosana SHIRAKI, Trans. Ent. Soc. Jap. 2 (1909) 234; OsHIMA, Rep. Term. 3 (1912) 81, pl. 1, figs. 7, 26, 27; pl. 2, figs. 11, 18. Termes vulgaris SHIRAKI, Trans. Ent. Soc. Jap. 2 (1909) 233; Osuma, Rep. Term. 1 (1909) 37, pl., figs. 9, 10; Zool. Mag, Tokyo 22 (1910) 379; Rep. Term. 2 (1911) 7; MarsumuraA, Schidl. u. Niitzl. Insekt. v. Zuckerrohr Formosa (1910) 2, pl. 1, figs. 2-4. Odontotermes (Cyclotermes) formosanus HOLMGREN, Termitenstudien 3 (1910) 38; Termitenstudien 4 (1913) 116, pl. 4, fig. 11; pl. 5, fig. 11; Hozawa, Journ. Coll. Sci. Tokyo 35 (1915) 105, pl. 3, figs. 21-23, text figs. 26-28. Odontotermes formosanus HOLMGREN, Annot. Zool. Jap. 8 (1912) 127. Odontotermes (Cyclotermes) formosana OSHIMA, Philip. Journ. Sci. § D8 (1913) 278. Imago.—Head, thorax, and abdomen chestnut; anteclypeus yellowish white; antennz, labial palpi, T-shaped marking and anterolateral corners of pronotum, and legs brownish yellow; Sieh asa > cs ~ XV,4 Oshima: Formosan Termites 327 wings fuscous, with yellowish subcostal band; head, thorax, and abdomen densely pilose. Head round; fontanelle somewhat elevated, minute; antenne 19-jointed, second joint twice as long as third, fourth joint as long as third; eye round and prominent; ocellus oval, separated from eye by a distance more than its longest diameter; post- clypeus swollen, width more than twice its length; pronotum narrower than head, subsemilunar, anterior border nearly straight, sides strongly converging posteriorly, posterior border rounded, slightly curved at middle; mesonotum and metanotum longer than pronotum, their posterior borders strongly indented at middle; anterior wing stumps somewhat larger than posterior, not covering the latter; radius nerve of anterior wing runs near and parallel to costal, with no branch; median nerve starts from proximal part of cubitus, giving off four bifurcating branches at the apical area, cubitus with about ten branches, not reach- ing tip of wing; median nerve of posterior wing starts from. basal part of radius. Measurements of the imago. mm. Length of body with wings 27.00 Length of body without win 9.00 Length of head , 2.20 Width of head 2.00 Width of pronotum 24.00 Length of pronotum 1.20 Length of anterior wing 24.00 Soldier—Head reddish yellow; mandibles reddish brown; labial palpi, antenne, and legs yellow; abdomen pale yellowish white; head, thoracic plates, and abdominal tergites sparingly pilose, long hairs mingling with subequal minute hairs. Head ovoid, longer than broad, sides slightly converging an- teriorly; antenne 16-jointed, third joint smallest, second joint about twice as long as third, fourth joint shorter than third; mandibles saber-shaped, with incurved and upcurved tip, each mandible provided with one tooth, of which the left one is stronger and distinct; labrum lancet-shaped, scarcely reaching middle of mandible, with acutely rounded tip; clypeus very short, not separated from forehead; pronotum slightly broader than long, anterior border elevated and distinctly bilobed, posterior border rounded, concave at middle; mesonotum oval, nearly as broad as pronotum, posterior border curved at middle; meta- notum much broader than pronotum, posterior border rounded. 328 Philippine Journal of Science 1919 . Measurements of the soldier. mm. Length of body 4.50 Length of head with mandibles mae Length of head without mandibles 1.40 Width of head 1.12 Width of pronotum 0.74 Length of pronotum 0.47 Worker.—Head yellow; clypeus somewhat paler; antenne and legs straw-colored, thorax and abdomen whitish; head moder- ately pilose; thorax and abdominal tergites densely provided with minute hairs. Head spherical; antennze 17-jointed, third joint smallest, second joint elongate, more than twice as long as third, fourth joint slightly longer than third; postclypeus swollen, more than twice as broad as long, anterior border concave; anteclypeus whitish, anterior border convex; pronotum saddle-shaped, much - narrower than head, anterior border strongly elevated, indented at middle, posterior berder rounded, curved at middle. Measurements of the worker. mm. Length of body 3.50 Width of head 1.81 Width of pronotum 0.62 Distribution.—Odontotermes formosanus is one of the com- monest species in Formosa. It has been recorded from Ishiga- kijima and South China (Amoy and Foochow). _ Remarks.—This species attacks living plants. It is a serious pest to sugar cane and young camphor trees. CERTAIN HABITS OF COPTOTERMES FORMOSANUS SHIRAKI DIFFERENT CASTES IN THE COLONY Generally the members of a termite colony differ greatly at different times of the year. Eggs and newly hatched larve of Coptotermes formosanus are most numerous in the summer; winged forms and nymphs are not present after the swarming season (from the end of May to the beginning of June) ; nymphs increase in number in the spring, becoming most abundant in April; at the end of May the nymph changes to an imago and usually swarms during the first ten days of June. A complete colony contains the following castes: 1. Newly hatched larve. The heads of all are alike in dimensions and are provided with 10-jointed antenne. 2. Larve of soldier, derived from 1. Distinctly differs from the other castes in having somewhat elongate, toothless mandibles and sub- orbicular head. ae ~ a u~ XV,4 Oshima: Formosan Termites 329 3. Larve of worker, derived from 1. Large-headed and provided with clearly denticulated mandibles, the tip and the inner margin of which are more or less brown. 4. Larve of royal form, derived from 1. Small-headed; other exter- nal characters are similar to those of worker larve. 5. Nymph of royal form. The members of this caste are provided with two pairs of wing pads and one pair of nonpigmented eyes and ocelli; body milk white. . Winged forms with pigmented eyes. Body reddish brown. At the beginning of May vast numbers of this caste are found in the nests; they emerge from the old nest early in, June. 7. A single queen, derived from a female of the winged form. Ab- domen greatly enlarged, with two pairs of triangular wing stumps on the thorax. 8. A single king, derived from a male of the winged form. Abdomen normal, with two pairs of triangular wing stumps on the thorax. . Substitute royal forms. It is certain that there are several] kings and queens belonging to this class; however, the substitute king is not known. The substitute queen which was collected in Kiushiu by Mr. Tatsuo Yoneyama, engineer of the Imperial Railway, is 12 millimeters long; head yellow, abdomen milk white, thorax with no wing stumps. This caste is very rare in the colony of Copto- termes formosanus. According to Yoneyama’s information, the nest which contained this queen was orphaned and some fifteen of the same form were captured at the same time. - The following are descriptions of young forms, given some- what fully: Newly hatched undifferentiated larve—Head, thorax, and abdomen milk white and densely provided with minute hairs. _ Head squarish, sides strongly converging anteriorly, posterior border nearly straight, posterolateral corners rounded; antennz 10-jointed, joints enlarging apically, first joint quadrilateral, second joint slightly shorter than first, broader anteriorly, third joint semidivided, slightly shorter and narrower than second, fourth to sixth joints ring-shaped, seventh to ninth joints spher- ical, apical joint elongate, oval, slightly narrowed anteriorly; anteclypeus short, its anterior border rounded; postclypeus not separated from forehead; labrum tongue-shaped, longer than broad, anterior border rounded; mandibles triangular, tip ob- tusely rounded, cutting margin of both sides with traces of minute teeth; pronotum narrower than head, crescent-shaped, posterior border rounded; mesonotum and metanotum broader than pronotum. or) = Measurements. mm. Length of body 1.50 Length of head 0.27 Width of head : 0.52 Width of pronotum 0.26 Length of pronotum 0.10 330 Philippine Journal of Science 1919 Larva of soldier, with 12-jointed antenne.—All parts of body milk white; tips of mandibles pale brown; head coarsely pilose; thoracic plates hairy; abdominal tergites densely covered with short hairs. Head broadly oval, slightly longer than broad, broader anteriorly; fontanelle indistinct; antennz 12-jointed, second joint nearly as long as third, third joint narrowest, fourth joint shorter than second, fifth to ninth joints spherical, apical joint elongate, narrower anteriorly; clypeus twice as broad as long, anterior bordef rounded, boundary between anteclypeus and postelypeus indistinct; labrum lancet-shaped, mandibles rather short and broad, with pointed and incurved tip; pronotum nar- rower than head, oval, anterior and posterior borders rounded; mesonotum and metanotum round, the former narrower than the latter and broader than pronotum. Measurements. mm. Length of body 4.00 Length of head with mandibles 1.50 Length of head without mandibles 0.93 Width of head 0.90 Width of pronotum 0.53 Length of pronotum 0.31 Larva of royal form, with 12-jointed antennx.—Head, thorax, and abdomen whitish and sparingly provided with minute hairs. Head round, nearly as long as broad, anterior corners of transversal band yellowish brown; postclypeus slightly swollen, longitudinally depressed along median line; anteclypeus as long as postclypeus, anterior border rounded; labrum tongue-shaped, overlapping the mandibles; mandible triangular, cutting edge dark brown; antenn# 12-jointed, second joint as long as fourth, third joint smallest, nearly half as long as second; pronotum ovoid, anterior border convex, slightly emarginate at middle, posterior border rounded; mesonotum and metanotum broader than pronotum, mesonotum narrower than metanotum. Measurements. mm. Length of body 3.50 Length of head 0.84 Width of head ' 0.87 Width of pronotum 0.50 Length of pronotum 0.21 Remarks.—in the present phase the external characters of the larva of the worker are similar to those of the larva of the royal form, except the size of the head. The former is provided ee Rete ee ~~ ~ XV, 4 Oshima: Formosan Termites 331 with a much larger head, measuring 0.91 to 0.94 millimeter Jong and 0.88 to 0.91 millimeter broad; while the latter has a smaller head, measuring 0.84 to 0.91 millimeter long and 0.84 millimeter broad. This distinction between the two forms is observable in the earlier stage when the larve are provided with 11-jointed antenne. Nymph of royal form.—Head, thorax, and abdomen milk white; eyes nonpigmented; head and pronotum sparingly pilose; wing pads with a few short hairs; abdominal tergites densely covered with minute hairs. Head round, slightly broader than long; fontanelle spotted, round and white, situated at center of head; eyes round and prominent; ocellus separated from eye by a distance equal to its diameter; postclypeus short, anterior border nearly straight, shorter than posterior; anteclypeus as long as postclypeus, with rounded anterior border; labrum tongue-shaped, sides slightly converging posteriorly, tip and cutting margin of mandibles dark brown; antennz 19-jointed, second joint longer than third, fourth joint nearly as long as second, semidivided; pronotum semicircular, anterior border concave, posterior border indented at middle, anterolateral corners rounded; lateral parts of meso- notum and metanotum much produced posteriorly, forming primi- tive wing cases, which are provided with rudimental veins. Measurements. ™m. Length of body 8.50 Length of head 1.15 Width of head 1.31 Width of pronotum 1.40 Length of pronotum 0.93 THE FOUNDATION OF A NEW COLONY In Formosa the swarming of Coptotermes formosanus takes place at the end of May or early in June, while in Japan proper the winged individuals swarm a little later; that is, at the end of June. TABLE I.—Records of the swarming season of Coptotermes formosanus. iC ear. Formosa. Japan proper. | ET Pt UN ae Ce canu ce wae June 12 to 19; at Marugame. ie) Sain la ss June 26; at Mitaijiri. | 1914! June 8; at Taihoku _____--___-- Le _...| June 25; at Gifu. | $9165) Je FT: at TAMOKG: wc0 oe ee No record. HIG. | Sune 4: at Tathoku ooo clic ee a cs Do. | Wt) Janet wk Talhah. A | De. 332 Philippine Journal of Science 1919 After the adults have rapidly emerged from the parent nest in a vast swarm and have flown a short distance in an irregular, wobbly manner, they fall to the ground. As soon as they reach the ground, they cast the wings. Then the male is attracted to the female and follows her tirelessly and closely, performing the so-called “Termiten Liebes-Spaziergang.” In this manner the imagoes separate into pairs and enter hiding places in order to establish new colonies, usually under pieces of decay- ®===) ing wood, in holes and crevices in wood, or directly in the earth. In order to observe the starting of a new colony, many new couples were captured in glass tubes, arranged as shown in fig. 1. The bottom of a test tube, 2 by 16 centimeters, is bored with a capillary pore, and a piece of absorbent cotton, a, is inserted. Then a proper quantity of crushed clay, b, is placed in the tube to serve as the abode of the fu- ture colony ; upon this layer pieces of soft wood or cotton, a, are laid for food. A male and a female that have recently flown from the nest are now placed in the tube, which is tightly corked and put in a dark place. In order to keep captive termites in a healthy condi- tion, it is absolutely necessary to give them proper hu- Fic. 1. midity. This is done by placing the end of the tube in water, thus allowing the absorbent cotton to draw a sufficient amount of water through the pore. As a result of these investigations, the following facts were determined: 1. During the first stage of colonization both the male and the female are active. They forage for themselves and are equally important in establishing the new colony and in rearing the first brood of young. 2. Egg laying in a newly established colony begins from five to thirteen days after the swarming. TABLE II.—Time between swarming and egg laying. RECORDS FOR 1915. Days after swarming. Cases. 5 3 6 8 2 10 8 5 9 6 10 5 i 3 2 3 an ea ei hi a i ane eR aoe Ms A ee y., See XV,4 Oshima: Formosan Termites 333 TABLE II.—Time between swarming and egg laying—Continued. RECORDS FOR 1916. Days after swarming. 5 8 9 10 12 3. The rate of egg laying is not rapid in the first batch, one to four eggs being laid on one day. 4. Eggs hatch out in from twenty-four to thirty-two days after they are laid. 5. After five months of captivity, the nests in two of the tubes were opened. In these there were no unhatched eggs; the king and the queen, which still retained the normal form of the imago, ran actively, in company with rather small individuals of the worker type and soldiers. The number of individuals in each nest agreed well; one contained twenty-two individuals of the worker type and two soldiers, the other, twenty individuals of the worker type and three soldiers. These facts clearly show that about twenty- five eggs compose the first batch in a newly established colony, and that no more eggs are laid until the eggs of the first brood are all hatched. 6. In the first brood the soldiers are few in comparison with individuals of the worker type, about 10 per cent of the number of the latter being soldiers. 7. Soldiers hatch from the eggs that are laid by true royal females.* for) neuen SITUATION OF THE NEST Coptotermes formosanus does not construct a large mound as do some species in the Tropics. Usually it makes its nest in the ground at a depth of from 6 to 10 feet (1.82 to 3.04 meters). Very often the nest is made at the junction of rafters of buildings or in the inner part of infested timbers, in wooden boxes or cabinets, or in the interspaces in walls. Generally the nest is round and honeycombed, and in the center there is a small, slitlike royal chamber (Plate II; Plate III, fig. 2). The nest consists of a mixture of abdominal excreta and clay or sand, pasted together with a special secretion of the salivary glands. Sometimes it is rigid and compact and seems like a piece of rock. However, it is inflammable and burns rapidly, leaving a small amount of ash. * As it is almost impossible from external characters to draw a fast line between larve of royal forms and of workers, nothing at present can be said with regard to the origin of these two forms. Philippine J ournal of Science 334 1919 » TABLE III.—The rate of egg laying, Coptotermes formosanus. | Eggs. Larvee Date. | Total. |Increase.! Total. | Increase. Bay DO sk a i ws Soe fon ean] ened oeecnencofeenensesee ay Rh eee ean cs anude seannneen 3 pig ERE ey Oe re eee Sige. Seer ge ee Sees ed cee rem ee 4 yay et x once bE Be ee eS a ea ba cake gun wkcedeless occas. 6 a ae Se, Poh Reeniae Meee rie he 7 , § ees eee | pS Ba SRE SiS SAT eo een mn Se OR ee 8 ee es Fo Skier ns ices eos nea bac te eee pe weet eee 10 2 ester Cy & BES RESE i RRs Fane Rae a copter Saenger eect argc 10 MEIN Wo ee es ie oe knw nao seacnaun 10s aaneekh cea et ace tee Bsngeece Sor soa tke ec assaxcber ets n eee saee ce s5 10 fctvonee 8 UTE BERD SSE Sy OSE, Se eine toy cen CUR CRE Sern Re ee Ne Sree: 10 = Rag? SRE PERS: SSCS OE S.No gee een aa mee epee 12 | ae ei: eieeier Ee OSS Rea Sse Se eS ea eer 16 Pg Bs Caran uae ae [aie Te os oes 16 {22.2 lee ek ee Ee Si eee rp ae cee 18 @ erete ae Pe fn Se Se Se RCAC a heer beep Tee ree po Oe Ey era ee, 18 jcc osc Nj sts ea is Wa ee ria uh Ae ees ens PP eeu (Cy rogt pec Ee ee ee RS eke cae eee pte, ect ene Repee y—ee rs Mbp ieremnanes June 12 -.-..-- aor 22 E bemmyedunenfoccncee ake fA ne eR ee aca ne Ck duces Sean seeewn Re i neu asaatecce wena | ee ee ae saves eas bo aekaea ace = ae RE ME ae Oe PN Sy Pe St RD nee wend pee poe So ee cies vacs Ses ckn Ooh Gh. cosnnleesencuapeloeeercees poe IG ee BEN sich ccalebaweweetuls awennabe June 25 = oo «Re Caer perenne June 26 py eee 3 3 |: SRT a Wie 5 2 June 23 fe 16 6 1 | UNE BD wae ca ee Bs ee ee ae fee es 9 3 | June 30-2 aoe ene i oa |: Ae. ie ein eee eta ee 13 gee eee } Way. SS eanel uc antes Geta interes oan ee 2 baw Soo es “ Ter foc ocancoee 10 1 Jmy € eee ans Set 4 tek cese. 2 a0: ashe Say Oly 6 os ee ee 14 4 12 s Pd deurenanee manent ston Sue 12 1 15 3 | SUNS Ss a 12 4 19 4 | Duby $0522 en Seok 13 1 19 oA UD oe ee es eee cas Whee tg BAS Ne a ae So we i eae. pig eee See Oe WME BG as bo cos ceo encesttadis ences goctesuuae sce Bg Bees pit eee WO foxcusaanuel Waly We ewbseCesacesacss Mosca cee se Btls 20 1 Sale We soo es ee ae eS be genre ome 20 disease ORGIES eas, iniea pas TS ei scscee TD Lic xccincuns pt gi | pee A Nee Reon eerie Se ata SRI woe scar gl Maine he 21 Fr “ Date of capture. DAMAGE TO BUILDINGS AND OTHER MATERIALS Coptotermes formosanus is essentially a wood destroyer and attacks very seriously all sorts of woodwork and wooden struc- tures. Because of its habit of attacking Japanese pine, which is a > d 2 4 XV,4 Oshima: Formosan Termites 335 an important building material, it is classed as a most formidable pest throughout Formosa. Moreover, as the method of at- tack of this insect is insidious, usually leaving the outer layer of wood intact as a protective covering from sunlight and its natural enemies, so that the damage is always hidden until beyond repair, it is a most dangerous enemy to buildings (Plate IV). TABLE IV.—The rate of egg laying, Coptotermes formosanus. Eggs. Larve. Date. Total. |Increase.| Total. | Increase. ay 0 eae a a a ee May 26.3 ee 2 Ree a, Fer ear MAY 265250 Ss eA ee ee 5 Bee eS. Eee Ree WAR OU. oS ee ee cease. A Reese Sean BPs Tea Dee Te od MAF Do scccasincsi ata | ee seid gy Og ena cre a ee aan Sree ones 6 11, Meher AER ERY Os agatha cece cess 8 Otis sie eee! WS BY a ee ee Ss 9 bE ROB ihla Rake eae ee es BUMS Acasa a ll _ of BERR REREr er sea PE Ba ec ope eee ocusdssecawenee 13 a ets NI ee oa ts a 14 Bas i es AI esate as ne A Se eS a dace Mees ode ee a, OMS Ge cadtae a teeta one as kere ne Caedeceseckeatandl 15 : 2 Sauer es Seeker wre I Ta ee oe oats coseee cetuce De eae ah pe keicheonecacane ae RR Bg oo Se SL eee eee 16 i Sie ON Ti os oe a ew a ceceseees 5G toe wh ce: MiG 185s hesaiens coed ar aes JSeeaseunngs eiag ieee es! 17 sR Aa PMN FIORE UNG Ae else a ie Case eines een dug eee BM Acvccccuos oud anaues woe eae Sane 16 cen a 18 pI eeonesiew te ieee based TIO se a ae pe a een St eke 19 Dl enctarcoicee ad ane 0 ee ee 20 “Etter ies Pe eee WUBO BES. cleo se este eel ekge cases aca eneee! ek pewaeuen i Baia mak ON ea ak ax aakee Laue ceS nb abe weea toe edad oeue OO Looe ee ee RN oak Soe he a alii og cb ae ewedca seen. 20 SDI nen, “yar OTE oe oa ka i eek eine pl een awn eenenese nnn. ps orneettnaemr aren | everest a Gene Se IO Rt a ae Se cea e ce cenceneun) OF tice nee 3 3 PUNE Mer sc. oe ee 16 4 1 ee pe air Sens epee Cag ea awe on ccenkn Seer pT RS Retin ie, 5 1 June 30. ep ea ap eee ee See aS eae ees 6 1 July 3. oS rash Ronnie > eagta aee pee em CE a WD oss oe 10 4 wy Goss eee ares iy Rkerw rs 14 4 BLE gle Boosh cantina Teer ioe eh gia Cee ace eee a Sie Mitsuki e TEN ee ee ee aoc i a hemes cet ke en Re 16 2 Fe a RE ee A ESSE SES A ee ee eae OS SRO ener cel bg Samir ata lly Tsao a a ee es ag Bare pean, 17 1 Pe eS es oe ga os Fa 3 EY Of PPD Ee Denaee & PR Te a a eee eaaae 0 20 3 * Date of capture. 336 Philippine Journal of Science 1919 TaBLE V.—Incubation period of eggs of Coptotermes formosanus. RECORDS FOR 1915. Days. Cases. 25 3 26 1 27 2 30 z RECORDS FOR 1915. 24 + 26 4 2T L 28 5 29 3 30 5 31 5 32 4 In infesting buildings it generally gains entrance from out- door colonies. By means of subterranean tunnels of consid- erable length, which originate from a nest in the ground, it reaches the foundation of a building. At first it attacks founda- tion timbers, flooring, or supports of porches or steps that are in contact with the ground, and gradually extends its tunnels and excavation into the first, the second, or even the third floor and into the roof, passing through the walls or the interior of timbers (Plate VI, fig. 1). In cases where the foundation is of stone, concrete, or other impermeable material, it constructs a covered tunnel of a mixture of earth and saliva over the surface and reaches the woodwork. Generally a cross section of this artificial tunnel is semicircular, the diameter being from 5 to 10 millimeters (Plate VII, fig. 1). Damage to buildings——Ordinary Japanese houses are chiefly constructed of wood and clay, the foundation timbers being laid in contact with the ground. Thus the construction itself is not fitted to prevent the attack of termites. Moreover, Japanese pine and cryptomeria, which are most liable to be attacked by Coptotermes formosanus, are the principal building materials. Such being the case, the Japanese people suffer a great deal from the damage when they erect buildings of their own style in Formosa. As the work of Coptotermes formosanus is hidden, it is dif- ficult to detect the presence of the insects or the damage that they are doing. After they have caused the fall of a building, the beams, the rafters, and other important parts of the wood- work are often found to be mere shells, the interior being en- tirely honeycombed (Plates V and VI). i= XV, 4 Oshima: Formosan Termites 337 Damage to brick walls.—There are several records in For- mosa of modern brick buildings having been infested by Coptotermes formosanus. It is absolutely impossible for it to attack brick itself; but, according to my actual observation, the mortar which is used as a cementing material for bricks is subject to attack. In building a thick brick wall the following method is adopted: The face bricks are laid upon a bed of mortar; the mortar, in a semifluid condition, is then poured into the space between the face bricks; the bricks are then pushed rapidly, horizontally for a short distance, into their position; a certain amount of the mortar is thus displaced; this rises in the side joints and completely fills all the interstices; should the mortar not rise to the top of the joints, the vacant spaces are filled up, when the next course is larried. All the interstices between the bricks would be completely filled, if the bedding could be carried on theoretically; but this is practically impossible, so that there are almost sure to be voids in brick walls. Under such conditions, termites prefer to pass through brick walls, if they have a chance to permeate them, rather than to excavate ordinary walls; because the voids are always in a favorable condition for their life, being pro- tected from the sunlight and containing a proper amount of moisture. There are two kinds of mortar used for constructional work; namely, cement mortar and lime mortar. The former is com- posed of sand and Portland cement (the ratio of Portland cement to sand is from one to two to one to four), while the latter is composed of moderately hydraulic lime and sand (the ratio of lime to sand is usually one to two). In Formosa until a comparatively recent date lime mortar was exclusively used for the bedding of bricks. It has been found, however, that Coptotermes formosanus easily penetrates brick buildings and causes serious damage (Plate VII). Since the Government has required the use of cement mortar instead of lime mortar there has been no record of damage to brick walls. Therefore, it seems reasonable to conclude that there exists some special relationship between lime and the destructive power of Coptotermes formosanus. The soldier of Coptotermes formosanus is provided with a special gland on the forehead, which secretes a milky, acidulous fluid; its excreta and saliva are also acidulous, while those of the worker are alkaline. In the passages perforating the lime 338 Philippine Journal of Science 1919 mortar of brick walls vast numbers of soldiers are found. Such being the case, it is said that the soldier of Coptotermes formo- sanus attacks lime mortar by dissolving the lime with the acidulous secretions. Damage to railway sleepers——Coptotermes formosanus attacks railway sleepers as well as buildings. According to a state- ment of the Bureau of the Formosan Government Railway, the life of an untreated, chestnut sleeper is only two years in For- mosa, while in Japan proper it is from eight to ten years. In the southern part of Formosa another species of termite, Odontotermes formosanus, also attacks the sleepers, as shown in Plate VIII, fig. 3. The damage it causes is rather more serious than that caused by Coptotermes formosanus. There- fore, in the case of sleepers, it is important to prevent the damage caused by these two species. Usually the length of the spike which is used to fix the rail to a sleeper is a little greater than the depth of the latter, so that its tip always penetrates beyond the bottom of the sleeper, causing some damage to that surface (Plate VIII, fig. 1). This point is most liable to be attacked. As the pests excavate the wooden tissue surrounding the spike (Plate VIII, fig. 2), the latter becomes ineffective. Suppose such damage occurs successively in several sleepers; the results are quite obvious—the rails spread and cause great -danger to passing trains. Damage to ships——One can hardly believe that steamships and launches in the water have been attacked by Coptotermes formosanus; but it is an undeniable fact. There are many records of such cases in the harbor of Keelung, Formosa. It happens in this way. In the swarming season, many thousands of winged males and females start from the nests in the vicinity of the shore. Some of them have a chance to fly into the ships moored in the harbor and there start new colonies. The interior of the ship is dark enough to favor the establishment of a nest. Usually the bottom of the ship is constructed of Japanese pine, which is most durable in water; all the timbers contain a favorable amount of water. Thus, all conditions being exceedingly favorable to Coptotermes formosanus, the ravages are extended step by step, until the vessel becomes unseaworthy, because of the unexpected destruction of important parts of the woodwork. In such a case extermination of the pests is very difficult. When the Government trawler Ryokai-maru was found to be infested, it was purposely sunk in the sea in order to destroy the insects. a # | i i = XV, 4 Oshima: Formosan Termites 339 Besides the above-mentioned damage, that done to bridges, telegraph poles, books, paper, wood pulp, cotton, and clothing ‘is sometimes very serious. PRINCIPAL FOOD OF COPTOTERMES FORMOSANUS The stem of an exogenous perennial is a complex of struc- tural elements of varied form and function. Of these we may distinguish three main groups: a, vessels; b, wood cells proper; ce, medullary tissue. The growing cell of plant tissue consists of cell wall and protoplasm, the living functions depending upon the activity of the latter. However, the above-named three main structural elements of the wood do not contain nitrogenous substance—that is, protoplasm—but mainly consist of the special constituent of the cell wall known as cellulose. There are, as might be expected, a great many varieties of cellulose, and the term must be taken as denoting a chemical group. Cellulose, taken as a group, presents the following characteristic: A colorless substance, insoluble in all simple solvents; generally but variously resistant to oxidation and hydrolysis; .nonnitrogenous, having the empirical constitution characteristic of the carbohydrates. The composition of pure cellulose is represented by the percentage numbers C 44.2, H 6.3, O 49.5, corresponding to the empirical formula (C,H,,0,).. It is flexible, slightly elastic, permeable, but only slightly absorbent, and does not readily undergo fermentation. When treated with acid it passes into a starchlike condition, as is evidenced by its turning blue with iodine; and under certain conditions in the living plant it would seem capable of being formed from sugar or of passing into it. It must be noted, however, that the typical cellulose is not separated from the plant in a pure state, but in admixture or in intimate chemical union with other compounds or groups of compounds. The latter are distinguished by greater reac- tivity; for example, they readily yield to alkaline hydrolysis, to oxidation, or to the action of the halogens. In the latter is included the very important group of lignified cellulose, or lignocellulose, distinguished by the presence of ketohexene groups in union with the cellulose, and therefore combining directly with the halogens. Generally, walls of cellulose, fibers, and vessels in the wood acquire mechanical strength or resistance by undergoing a change known as lignification. This consists in their impreg- nation with a substance known as lignin, forming a compound 340 Philippine Journal of Science 1919 cellulose, namely, lignocellulose. Lignin, like cellulose, consists of three elements—carbon, hydrogen, and oxygen—but in dif- ferent proportions, its percentage composition being C 49, H 6, O 44. Its chemical constitution is as yet unknown. It is harder and more elastic than cellulose, readily permeable by water, but not absorbent. It is more soluble in acids than is cellulose and is recognized by turning deep magenta when treated with phloroglucinol in hydrochloric acid. As shown in the preceding pages, Coptotermes formosanus seriously injures all sorts of woodwork and wood products. However, why it attacks such materials or, in other words, what was the principal food of Coptotermes formosanus contained in wood was quite unknown. In order to settle this question the following experiments were made: Experiment 1.—A living worker or soldier of Coptotermes formosanus was placed on a microscope slide, and the tip of its abdomen was pressed, the excrement being thus discharged. This was treated with phloroglucinol in hydrochloric acid under a cover glass. The color changed to deep magenta, showing the characteristic reaction of lignin. * Experiment 2.—A piece of the nest was treated with the same reagent. It also gave a deep magenta coloration, characteristic of lignin. : Experiment 3.—A piece of camphor wood and a nest of Copto- termes formosanus made from camphor wood were analyzed. TABLE VI.—Composition of camphor wood and of a nest of Coptotermes formosanus. Analyzed by T. Katayama, [Numbers give percentages.} for: oe ; Ss caer Aqueous Pen Noncel | - oncel- Water. Ash. éxtenet: (tesa; Cellulose, taloun. Oemmbor Wo0d 52s oc eS 11.51 1.29 4.53 13. 92 48.35 20. 40 WRONG ik sho eden owe eaes teen 11.39 17.86 4. 83 6. 02 12.73 47.17 | Nest (calculated as a substance } Wilt AGOMN) ks acd. Se, 13.87 0.00 5.88 7.38 15.50 | «57.42 It is quite obvious that the amount of cellulose is the main difference between the constituents of the camphor wood and those of the nest. As there occurs no decrease of noncellulose, it is clear that cellulose has been taken as the food when the camphor wood passed through the alimentary canal; and non- cellulose, that is, lignin, which is produced as a decomposed material of lignocellulose by the special function of the alimen- tary canal, is discharged as the building material of the nest. cottons =o |S Cf See ee XV, 6 Haughwout et als.: Mania with Plasmodium vivax 567 to believe that all of them fell under treatment during the course of active malarial infection, while it must be borne in mind that ours was a chronic case showing no symptoms of actiye malaria. Wurtz and van Malleghem say they could find no parasite save Plasmodium vivax; while Hesse, according to the abstract, con- tents himself with speaking of his cases as being “benign” ter- tian, without naming the parasite. We are inclined to suspect that he was dealing with Plasmodium falciparum infections. The possibility that the Roentgen rays in some manner precip- itated the attack of mania might be borne in mind. We think this extremely unlikely unless their application operated to ac- tivate the parasite in some way—a thing that might possibly come to pass. However, in the short series of cases of chronic malaria that we have treated with the Roentgen rays, this is the only one that developed cerebral symptoms. Skinner and Carson,(4) who first undertook the treatment of malaria with the Roentgen rays, report no such occurrence in their series; nor does Pais,(3) nor Deutsch,(1) who followed. Pais states his belief that new generations of the parasite appear to display exalted virulence under the influence of the rays. If this be true, we may have an explanation of the phenomena in our case, although it must be borne in mind that the parasites never were in the circulation in large numbers and that hyper- pyrexia never developed. Whatever the explanation of this strange case we can only repeat that painstaking search failed to disclose any evidence of infection with Plasmodium falciparum, while the parasites demonstrated in the circulating blood were unquestionably Plas- modium vivax. REFERENCES 1. DEuTscH, Feuix. Tiefelstrahlung der Milz bei Malaria. Wien. klin. Woch. 30 (1917) 207. 2. HESSE, WALTER. Malaria Comatosa und Malariameningitis bei Tertia- nafieber. Zentralbl. ~. Innere Med. 39 (1918) 385. Abstracted in Trop. Dis. Bull. 12 (1918) 351. 3. Pats, A. Influence of Roentgen rays on malaria. Gazz. degli Ospedali e delle Clin. Milan. 38 (1917) 1121. 4, SKINNER, BRUCE, and CARSON, H. W. Curative influence of Roentgen rays in malaria. Brit. Med. Journ. No. 2617 (1917) 451. 5. Wurtz, R., and VAN MALLEGHEM, R. Acces graves chez des paludeens atteints de tierce dite benigne. C. R. Acad. Sci, 164 (1917) 797. Abstracted in Trop. Dis. Bull. 11 (1918) 295. cd ILLUSTRATION TEXT FIGURE Fic. 1. Temperature chart of the patient infected with Plasmodium vivax. 569 REVIEWS Volume 2 New York Number Number 4 | The | Medical Clinics | of North America | January, 1919 | Published bi-monthly | by | W. B. Saunders Company | Philadelphia and London. CONTENTS Sterility in women, with especial reference to Endocrine treatment of same, by S. W. Bandler. A new pluriglandular compensatory syndrome, by Walter Timme. Pneumococcus endocarditis, by Walter W. Palmer. Mitral stenosis and auricular fibrillation. Digitalis—Its uses and dangers, by T. Stuart Hart. Non-hemolytic Streptococcus endocarditis, by Albert R. Lamb. Cystitis: A discussion concerning its diagnosis, by Leo Buerger. Certain aspects of the modern treatment of diabetes mellitus, by Henry Rawle Geyelin. Local evidence of tonsil involvement in the causation of distant or systemic _ disease, by Jesse G. M. Bullowa. Infiuenza of head and chest, by Jesse G. M. Bullowa. Cases illustrating diagnostic problems, by A. S. Blumgarten. I. Primary malignant tumor of lung. II. Cerebrospinal syphilis. III. Three cases illustrating problems in nephritis. IV. Two cases illustrating diagnosis of aortic syphilis. Auricular tachycardia in children: Two cases, by A. MclI. Strong. Renal disease, by Dana W. Atchley. The basal metabolism as a guide in the diagnosis and treatment of thyroid disease, by Eugene F. DuBois, Advanced pulmonary tuberculosis, a borderland disease, by Willy Meyer. Volume 2 Baltimore Number Number 6 | The | Medical Clinics | of | North America | May, 1919 | Index Number | Published bi-monthly | by | W. B. Saunders Company | Philadelphia and London. CONTENTS Funicular myelitis, or combined sclerosis of the spinal cord, by Lewellys F. Barker. Personal experience in the treatment of ulcer of the stomach, by Julius Friedenwald. Various types of achylia gastrica as revealed by the Rehfuss method of fractional analysis, by Julius Friedenwald. Some of the aspects of epidemic influenza in children, by John Ruhriah. Fundamentals in the treatment of pulmonary tuberculosis, by Gordon Wilson. Pneumococcus sepsis, by Paul W. Clough. The clinical diagnosis of epidemic influenza, by Arthur L. Bloomfield. 571 572 Philippine Journal of Science 1919 Notes on the gastric signs and symptoms in diseases other than those of the stomach, by Thomas R. Brown. Gastro-intestinal disturbances in metabolic diseases and diseases of the ductless glands, by John H. King. The réle of diet in treatment of digestive diseases, by E. H. Gaither. Esophagoscopy as an aid in the diagnosis and treatment of esophageal disease, by Elmer B. Freeman. The roentgenologie signs of joint lesions in children, by Frederick H. Baetjer. Introductory remarks to a discussion of diabetes, by Louis Hamman. Serous membrane tuberculosis, by Louis Hamman. Auricular fibrillation, by Louis Hamman. A case of multiple tuberculosis in childhood, by Allen K. Krause. Barbed Wire Disease: | A Psychological Study | of the Prisoner of War | by | A. L. Vischer, | M. D. Basle, M. R. C. S. Eng. | translated from the German, with additions by the author | with an introductory chapter by | S. A. Kinnier Wilson, | M. A., B. Sc. M. D. Ed, F. R. C. P. Lond. | and a frontispiece by | Miss E. Fortescue-Brickdale | London: | John Bale, Sons & Danielsson, Ltd. | Oxford House | 83-91, Great Titchfield Street, W. 1 | 1919 | pp. 1-84. Cerebrospinal Fluid | in Health and in Disease | by | Abraham Levinson, B. S., M. D. | [7 lines of titles] | with a foreword by | Ludvig Hektoen, M. D. | with fifty-six illustrations, including | five color plates | St. Louis | C. V. Mosby Company | 1919 | Cloth, pp. 1-231, including index. FOREWORD The author was kind enough to ask me if I would look over his manuscript and then tell him whether it seemed to me worthy of publication. Later, when I told him that in my opinion he had produced a valuable little book, he requested me to state the reasons for this favorable opinion in the form of a foreword. This I can do in a few brief statements. In the first place, on reading the manuscript, I soon became aware that the author had come to his task with not only a large experience behind him in the examination by various methods of the cerebrospinal fluid as an aid in diagnosis, but with a highly creditable record in the scientific study of this fluid as well. Evidently he had been drawn to his work on the cerebro- spinal fluid because of its attractiveness as a field of research, as well as on account of its importance in diagnosis. It is to this happy combination of true philosophic interest and first-hand practical knowledge on the part of the author that the book owes its chief merit, namely, thoroughness and freshness in the parts dealing with fundamental problems, as well as in those dealing with practical matters. In the second place there could be no doubt in regard to the timeliness of a work of this. XV,6 Reviews 573 kind. Indeed it seemed to me that a definite want would be supplied; for, in spite of an increasing importance in medicine, there was as yet no comprehensive book on all phases of the cerebrospinal fluid. FROM THE PREFACE Cerebrospinal fluid is of great physiologic importance for various reasons. It is the clearest and most transparent of all the fluids of the body. It is clearer than blood, than bile, and even clearer than urine, and under normal conditions experi- ments may be made on it without fear of clot formation or color change. Furthermore, cerebrospinal fluid, like blood and urine, can be removed from the living body without injury to the system. This gives one the opportunity of working with pro- cesses in the living body—a distinct advantage over the study of dead tissue. From the standpoint of pathology also, cerebrospinal fluid presents an exceptional opportunity for study. The slightest change in the color of the fluid, the smallest increase in the pro- tein content or in the cell count, all of which are easily discern- ible, indicate the presence of a pathologic process. One is able to follow the course of disease throughout all stages by noting the various changes the cerebrospinal fluid undergoes from time to time. These changes may be manifested not only by the presence of the causative organisms themselves, but just as frequently by specific physical, chemical, cytologic and physico- chemical processes. A close study of the changes in the cere- brospinal fluid under pathologic conditions throws light, not only on the specific diseases of the nervous system, but on the condition of other systems. One can readily see, therefore, how large is the scope for the study of cerebrospinal fluid. 1687483—_6 INDEX [New generic and specific names and new combinations are printed in elarendon; synonyms and names of species incidentally mentioned in the text are printed in ttalic.] A Acacia baileyana, 180. dealbata, 180. decurrens, 179. decurrentis, 181. farnesiana (L.) implexa, 180. juniperina, 180. pendula, 179. pycnantha, 179. Acalypha stipulacea Kl., 15, 17, 19, 20, 23-25. wilkesiana Muell.-Arg., 19, 20, 23. Acanthaceae, 256. Acineta, 399. Acocephalus stramineus Walk., 212. Acronia ? decimaculata Schultze, 548. Actinia equina Linn., 82. Acute mania associated with Plasmodium vivax infection, a case of, 563. Adamsia palliata Bohadsch, 82. rondeletti Della Chiaje, 82. Adonidia merrillii Bece., 12. Aganosma edithiae Hance, 254. Agapostemon, 9. Agaricaceae, 485. Agropyron Gaertn., 227. ciliare (Trin.) Franch., 227. Ajenjo, 260. Albizzia, 240. Alchornea Swartz, 244, 245. hainanensis Pax & K. Hoffm., 244. javanensis (Blume) Muell.-Arg., 244. rugosa (Lour.) Muell.-Arg., 244. Alecides adversarius Schultze, 558. arenatus Schultze, 559. catanduanensis Schultze, 558. figuratus Schultze, 558. plagiatus Schultze, 559. . schuetzet Schultze, 559. Aleurites moluccana (Linn.) Willd., 480, 482, 485, 487. Alsophila extensa (Forst.) R. Br., 540. haenkei Presl, 540. Alstonia macrophylla Wall., 485, 487. Alyxia Banks, 254. levinei Merr., 254, sinensis Champ., 255. Ampullaria, 389, 407. Anacardiaceae, 245. Analyses, Portland cement, 107. Ancylostoma duodenale, 564. Ancystropodium maupasi, 394. Willd., 483, 487. Andrena strigata Fabr., 4. Androsace saxifragifolia Bunge, 237. umbellata (Lour.) Merr., 237. Aneilema R. Br., 542. vitiense Seem., 542. vitiense Seem. var. petiolata C. B. Clarke, 542. Annonaceae, 137. Anona rufa Presl., 186. Anopheline mosquitoes, swarming of, 283. Antiaris Lesch., 231, 482, 487. toxicaria (Pers.) Lesch., 231. Antidesma bunius (L.) Spreng., 481, 487. Aphalara, 168, 165. Aphalarinae, 163. Aphanisticus, 295. Aphididae, 189. Aphrophorinae, 67. Apocynaceae, 254. Apophyllum anomalum, 181. Apsylla Crawf., 141, 145. cistellata, 145. Aquila, 35, 38. chrysaétos canadensis (Gm.), 31, 49, Aquilaria Lam., 248. chinensis Spreng., 248. grandiflora Benth., 248. sinensis (Lour.) Merr., 248. Araceae, 228. Araliaceae, 249. Aralia Linn., 249. : spinifolia Merr., 249, Arcofacies Muir, 525. fullawayi Muir, 526. insignis Muir, 526, penangensis Muir, 525, Arenga pinnata (Wurmb) Merr., 487. saccharifera Labill., 487. Arisaema Mart., 228. japonicum Blume, 229. kwangtungense Merr., 228. Aristotelia spiralis Lour., 230. Artemisia chinensis Linn., 260. judaica Lour., 260. Artocarpus integrifolia, 417. Arytaina Foerster, 167, 168, 172, 181, 184. brevigena Crawf., 173, 175. flava Crawf., 172, 174, 175. iolani Crawf., 172, 174. meridionalis Crawf., 173-175. obscura Crawf., 172, 177. pulchra Crawf., 172, 175. punctinervis Crawf., 172, 176. 575 52. 576 Arytaina Foerster—Continued. punctipennis Crawf., 189, 172, 176, 177, tuberculata Crawf., 173, 177. | variabilis Crawf., 178, 174, 17%: Asclepiadaceae, 254, Asiatic barn swallow, 533. Asparagus Tournef., 230, cochinchinensis (Lour.) lucidus Lindl., 230, Aspidosperma, 348, Asplenium, 297. nidus L., 290, 294, 295. Atlas preservative A, 875. Augochlora, 9, Auriculariaceze, 480. Auricularia Bull, 480, aurieula-judae (Linn, ) Schroet., 480, 488, mesenterica ( Dicks.) Fr., 480, 488-490, Polytricha (Mont. ) Sace., 430, 487-490, tenuis Lev., 480, 488-490, Avenarius carbolineum, 376. Merr., 230, B Bactericera Puton, 157, 203. Bactrocera Guér.-Mén., 415, 416, 617. albistrigata de Meij., 412, 416, fasciatipennis Dol., 417, longicornis Guer.—Men., 417. umbrosa Fabr., 412, 413, 417. BAKER, C. F., The Malayan Machaerotinae Cercopidae), 67; The genus Krisna (Jassidae) , 209; Notices of certain Ful- foroidea, II: the genus Trobolophya, 301. Balantidium, 340, 389, 390, 392, 395. 392, 894, 400, 402-404, coli var. hondurense Barlow, 392, duodeni Stein, 391, 394, elongatum Stein, 390. entozoén Ehrenb., 390, 391, 400, 402, faleifarum Walk., 392, 402, 403. haughwouti de Leon, 389-409, 497. italieum Sangiorgi & Ugdulena, 392, medusarum Meresch., 391, minutum Schaud., 391, 392. orchestis Wats., 392. Bambusa blumeana Schultes, 481-484, 487. sDinosa Roxb., 481-484, 486, 487, Bambusibatus Muir, 522. albolineata Muir, 522, BANKS, CHARLES S., nt Sis aaa ing new, 289. Barlerig procumbens Lour., 256. Barnardia scilloides Lindl. 229, i the Philippines and their hosts, I, 479, Bauhinia Sp., 485, 489. Bees, metallic-colored halictine, of the Philip- Dine Islands, 9. Philippine, of the genus Nomia, 1, Bernstein, Ralph, Rays in review of his Ultra Violet Modern Dermatology, 318, Campbellosphaera, Index Besseyosphaera powersi, 516, powersi Shaw, 513. BEZZI, M., Fruit flies of the genus Dacus sensu latiore (Diptera) from the Phil- ippine Islands, 411, Bignoniaceae, 255. Birgus latro Linn., 81. Biumea DC., 544. balsamifera (Linn.) DC., 19, laciniata (Roxb. ) DC., 544. Boophilus annulatus, 102. Boraginaceae, 250. Borduria Dist., 211, Bos indicus, 92. Bougainvillea, 391. Boveria labialis, 406, subcylindrica 406, Brachycraera Muir, 522. albolineata Muir, 522. Brachypsylla Froggatt, 169, Bryonia cochinchinensig Lour., 256, Bubalus bubalis Lyd., 91. Buffalo milk production, 95-101. Buprestidae, 289, 295. Buprestid, a new, 295. BURKILL, I. H., The genus Gordonia in the Philippine Islands, 475, Buteoninae, 54, Bythoscopus, 210. indicatus Walk., 212 testaceus Walk., 212. Cc Cacalia Pinndtifida Linn., 260, pinnatifida Lour., 260, segetum Lour., 260. Cajanus ecajan (Linn. ) Millsp., 481, 488. Cailantra Walk., 415, 416. Callistomyia Bezzi, 411, Caiiivis glauca R. Br., 348, 353. Calophya Loew., 141, Shaw, 493, 494, 497, 519- 513, 515. obversa Shaw, 494, 495, 505, 510-512, 515, 516, number of cells in, 519. Camphor green oil, studies on, 365. Cananga sylvestris IT latifolia Rumph., 133. CANIZARES, MIGUEL, Some abnormalities of the vertebral artery, 451, Capparis mitchelli, 181, Caprifoliaceae, 256. Carallia Roxb., 249, brachiata (Lour.) Merr., 249, integerrima DC., 249, lucida Roxb., 249, Carbolin, 376, Carbolineum atlas, 377, Carpopogon niveum Roxb., 242, Carsidara Walker, 157, 159, 163. Carsidarinae Crawf., 139, 140, 155, 156, 162, 168, 167, 184, Carsidaroida Crawf., Carystus Stal, 67, 160, 161. Index Cassia Linn., 240. fistula Linn., 171, 240. grandis L., 485, 488. Cassis, 83. Castilloa elastica Cerv., 480, Casuarina, 187. Cecidotrioza Kieff., 185, 186. Cedrus atlantica, 348. deodar, 348. Celastraceae, 246. Celastrus Linn., 246. hookeri Prain, 246. Celtis philippensis Blanco, 480, 483, 485, 488. Cement, analyses of raw materials, 114, 116, 118. analyses of raw mix by several methods, 114, determination of calcium by precipitation as calcium oxalate, 108. determination of calcium without the use of ammonium oxalate, 108. determination of constituents other than calcium, 109. Centrochares bucktoni Dist., 16. horrificus Westw., 15, 16. posticus Buckton, 16. Ceratina, 9. cerea Nurse, 9. punjabensis Cam., 9. Cercopidae, 67. Cerebrospinal fiuid in health and-in disease, 572. Ceropsylla, 184. Cerotrioza Crawf., 185, 201. bivittata Crawf., 184, 201, 202. corniger Crawf., 201, 202. microceras Crawf., 201, 202, Chaetodacus Bezzi, 415, 417. ablepharus Bezzi, 412, 414, 416, 417, 422, ablepharus mindanaus Bezzi, 412, 422. absolutus Walk., 414. aegualizs Coq., 4138. apicales de Meij., 412. atrichus Bezzi, 412, 414, 416, 417, 420, 422. atrichus davaoanus Bezzi, 412, 421, bakeri Bezzi, 413, 419, 426, 435. bezzi Miyake, 421. biguttatus Bezzi, 412. bipustulatus Bezzi, 412. caudatus Fabr., 411-418, 419, 428, 431. caudatus nubilus Hend., 412, 429. chrysotoxus Hend., 418, 414. , cilifer Hend., 412, 414, 416. continuus Bezzi, 418, 414, 419, 424, 431, 432, correctus Bezzi, 412. cucumis French, 413, cucurbitae Cog., 411-413, 419, 428. curvipennis Froggatt, 4138. davaoanus Bezzi, 417, diffusus Walk., 414. discipennis Walk., 414. diversus Coq., 412. 485, 488. o¢7 Chaetodacus Bezzi—Continued. dorsalis Hend., 411-413, 418, 423. duplicatus Bezzi, 412. emittens Walk., 414. expandens Walk., 414. facialis Coq., 413. ferrugineus Fabr., 412, 414, 423. ferrugineus O.-S., 4238. « ferrugineus dorsalis Hend., 412, 413, 423. ferrugineus incisus Walk., 412, 423. ferrugineus limbiferus Bezzi, 413, 414, 424, ferrugineus occcipitalis Bezzi, 412, 414, 423, ferrugineus pedestris Bezzi, 411, 412, 414, 428, 424, ferrugineus versicolor Bezzi, 412. frauenfeldi Schiner, 418, 416. froggatti Bezzi, 413. gareiniae Bezzi, 412. hageni de Meij., 412, 413. impunctatus de Meij., 412. kirki Froggatt, 413. limbiferus Bezzi, 411, 419, limbipennis Macq., 412. maculipennts Dol., 412-414. mecgregori Bezzi, 418, 414, 419, 426, melanotus Coa., 413. mindanaus Bezzi, 417, mundus Bezzi, 418, 414, 419, 429, 431. obscuratus de Meij., 412. occipitalis Bezzi, 411, 418, 424. ornatissimus Froggatt, 413. parvulus Hend., 412. passiflorae Froggatt, 413. pectoralis Walker, 414. pedestris Bezzi, 411, 418, 428, 424. peptsalae Froggatt, 413. pubescens Bezzi, 411, 413, 420, 434, retrachaetus Bezzi, 419. ritsemae Wey., 412. rorotangae Froggatt, 413. scutellatus Bezzi, 412, 484, seutellatus Hend., 412, 421, 434. scutellinus Bezzi, 411-413, 420, 432. sp. @ Hend., 412. sp. 6 Hend., 412. synnephes Hend., 412, 429-431. terminifer Walk., 414, 422. tetrachaetus Bezzi, 413, 414, 431, tongensis Froggatt, 413. tuberculatus Bezzi, 412. virgatus Coq., 413. xanthodes Brown, 413. zonatus W. W. Saund., 412, 413. Chermidae, 139. Chlamydomonas, 516. Chloralictus Robertson, 9, 13. Chrysomphalus quadriclavatus (Green), 886. rhizophorae Ckll., 385, 386. Cinnamomum, 236. camphora Nees & Ebermeyer, 366. Circaétus, 54. Cissus umbellata Lour., 252. 578 Citrus, 169, 172, 442. nobilis Lour., 487, 488. Cladoderris Pers., 481. dendritica Pers., 481, 488, 489. Cladodes rugosa Lour., 244. Coccidae, 139. COCKERELL, T. D. A., Philippine bees of the genus Nomia, 1; The metallic- colored halictine bees of the Philippine Islands, 9; The black halictine bees of the Philippine Islands, 269; A new seale insect on Rhizophora, 385. Coconut oil rancidity tests, 467. Coelidioides Sign., 211. Coleoptera fauna of the Philippines, seventh contribution to the, 545. Columbia serratifolia (Cav.) DC., 488. Commelinaceae, 542. Compositae, 260, 544. Convallaria chinensis Osbeck, 229. Copelandosphaera dissipatrix Shaw, 513. Coptotermitinae Holm., 321. Coptotermes Wasmann, 321. formosae Holm., 321. formosanus Shiraki, 333, 334, 336-342, 358, 361, 370, 371, gestrot Oshima, 321. Cordia, 484, 488. myxa L., 480, 488. . Coriolus Quel., 485. hirsutus (Fr.) Quel., 485, 488—490. COWLES, R. P., Habits of tropical Crus- tacea: III, 81; review of Holmes’s The Elements of Animal Biology, 123. Crab, coconut, 81. hermit, 81. robber, 81. Cranium, mandible, and associated bones of Pithecophaga, 32. Cratoxylon sp., 480, 485, 488. CRAWFORD, DAVID L., The jumping plant lice of the Paleotropies and the South . Pacific Islands, 139. Crepit, 377. Crinipellis Pat., 487. galeatus (B. & Curt.) Pat., 487, 488. stipitarius (Fr.) Pat., 487. Crossostephium Less., 260. artemisioides Less., 260. chinense (Linn.) 260, Croton, 248. apetalum Blume, 245. congestum Lour., 247. leiophyllus Muell., 24. tigliwm Linn., 248. Crustacea, tropical, habits of, 81. Cryptaspidia elevata Funkh., 26. longa Funkh., 27, tagalica Stal, 26. 480, 484, 820, 821, 328, 344, 346, 347, 373, 379, 331, 350, Index ! Cryptolepis R. Br., 254. elegans Wall., 254. sinensis (Lour,) Merr., 254. Ctenochiton rhizophorae Maskell, 385. Cucurbitaceae, 256. Cudrania Tréc., 231. pubescens Tréc., 231. Culicidae, 288. Culicinae, 283. Culex detritus, 283. fatigans Wied., 287. pipiens L., 283. (Uranotaenia) argyropus Walk., 283. Culicada nemorosa Meig., 283. Cunninghamia konishii Hayata, 364, 365. sinensis R. Br., 365. Curculionidae, 546, 549. Cyatheaceae, 540. Cyathea Sm., 540. haenkei (Presi) Merr., 540, marianna Gaudich., 540. Cyclomyces Kunze, 484, cichoriaceus (Berk.) Pat., 484, 490. tabacinus (Mont.) Pat., 484, 490. Cyperaceae, 541. Cypress pine, volatile constituents of, 358. D Dacinae, 412, 414. Dacryomycetaceae, 48i. Dacus, 411, 415, 416. annulatus, 416. asiaticus Silvestri, 412, bezzii Miyake, 434, blepharogaster, 416. brevistylus Bezzi, 412, 416. cucumis French, 416. erythraeus, 416. fascipennis Wied., 417. ferrugineus (Wied.) Maca., frenchi Froggatt, 417. hamatus Bezzi, 41€. icarus O.-S., 411. longistylus Wied., 412, 416. mochii, 416. oleae Gm., 416. trigonus Bezzi, 416. Dairy cows, feed for, 101. Dalbergia pinnata (Lour.) Prain, 241. tamarindifolia Roxb., 241. Daphnidium cubeba Nees, 235. Decadia aluminosa Lour., 252. DE LEON, WALFRIDO, Balantidium haugh- wouti, new species, parasitic in the in- testinal tract of Ampullaria species, a morphological study. With remarks on the relation between the Meganucleus and the Micronucleus, 389, Delhi buffalo, 97. Delphacidae, 521, DEL ROSARIO, MARIANO V., and MARA- NON, JOAQUIN, The physico-chemical evaluation of tikitiki extract, 221. 416, 411. Index Derris Lour., 241. elegans (Grah.) Benth., 241. pinnata, 241. trifoliata Lour., 241. uliginosa (Roxb.) Benth., 241, 242. Desmos Lour., 127, 129. Dialictus Robertson, 9. Diaphorina citri Kuwayama, 171. Diapodius griseus Bonvouloir, 450. Diatoma brachiata Lour., 249. Dicalyx cochinchinensis Lour., 262. Diceraopsylla Crawf., 141. Dicranotropis Fieb., 529. pseudomaides (Kirk.), 529. Digitaria Heister, 540. mariannensis Merr., 541. pacifica Stapf, 540, 541. platycarpha Stapf, 541. robinsonii Merr., 540, 541. stenotaphrodes Stapf, 541. stenotaphrodi Stapf, 540. Dinglas, 348. Diospyros Linn., 251, 480, 488. sinensis Hemsl., 251. Diplochorda O.-S., 416. Diploclisia Miers, 235. affinis (Oliv.) Diels, 235. chinensis Merr., 235. Diplodiseus paniculatus Turez., 480, 488. Diseases common among cattle, 102. Disporum Salisb., 229. cantoniense (Lour.) Merr., 229. pullum Salisb., 229. Dolichandrone Seem., 255. stipulata (Wall.) Benth., 255. Dolichos conspersus Grah., 242. punctatus W. & A., 242. Dolidae, 83. Doliops curculionoides Waterh., 546. geometrica Waterh., 546. siargacensis Schultze, 548. Drabescus Stal, 211. Dracontomelum dao (Blanco) Merr. & Rolfe, 482, 488. ¥ Dromaeolus bipartitus Fleut., 446, cylindricus Fleut., 446, 447. depressus Fleut., 447, opacus Bonvouloir, 447. semigriseus Bonvouloir, 446. Drosera umbellata Lour., 237. Drupatris cochinchinensis Lour., 252, Dunbaria W. & A., 242. conspersa Benth., 242. punctata Benth., 242. rotundifolia (Lour.) Merr., 242, Dynopsyila Crawf., 156, 158. minor Crawf., 158. Dyscolocerus bakeri Fleut., 450. subnitidus Bonvouloir, 450. Eagle, monkey-eating, osteology of, 31. Ebenaceae, 251. Elaeocarpaceae, 246. 579 Elaeocarpus Linn., 246, 482, 488. dubius A. DC., 246. Elytranthe Blume, 234. fordii (Hance) Merr., 234. Emericia sinensis Roem. & Schultes, 254. Emilia Cassini, 261. prenanthoidea DC., 261. sonchifolia DC., 261. Endelus aethiops H, Déyr., 297. bakeri Kerrem., 289, 290, 292-294. calligraphus Banks, 289, 295. diabolicus Kerrem., 292. merseulii H. Deyr., 297. modiglianii Kerrem., 292. weyersi Rits., 292. Endevleinia Schmidt, 68. Enderleiniini, 67. Eogypona, 210. Ephelota, 399. Epicarsa Crawf., 156. Epidendrum aristotelia Raeusch., 230. Epipsylla Kuwayama, 168, 177. albolineata Kuwayama, 177, forcipata Crawf., 178. pulechra Crawf., 177, 178. rubrofasciata Kuwayama, 178. Epitrioza Kuwayama, 185. Eriopsylla Froggatt, 169. Erythrophloeum lim, 348. Eucalyptolyma Froggatt, 168. Eucalyptus, 186, 187, 189. marginata, 348. Euchirus dupontianus Burm., 546. Euclea siargaona Schultze, 547, tagala Heller, 547. tagala subsp. rufofasciata Schultze, 547, Eucope, 391. Eudorina, 516. Eugenia Linn., 249. bracteata Roxb. var. roxburghii Duthie, 348. bullockii Hance, 249, calubeob C. B. Rob., 20, 21. malaccensis, 196. Eumetopina Breddin, 527. bakeri Muir, 528. flava Muir, 528. maculata Muir, 527, Euonymus chinensis Lour., 256. Eupagurus bernhardus Linn., 82. prideauxit Leach, 82, Euphalerus Sechw., 167-169. ceitri Buckt., 189. citri Crawf., 171. citri (Kuwayama), 169, 171. grandis Crawf., 169, 170, maculosus Crawf., 169, 170. nidifex Schw., 169. nigrivittatus Crawf., 169, 171. vittatus Crawf., 169, 171, Euphorbiaceae, 2438, 543. Euphorbia Linn., 243, 548. esula Linn., 243. macgillivrayi Boiss., 543. serrulata Reinw., 543, 580 Eurya Thunb., 247. distichophylla Hemsl., 247. groffii Merr., 247. swinglei Merr., 247. Excoecaria Linn., 248. bicolor Hassk., 248, 244. bicolor Hassk. var. viridis Pax & Hoffm., 244, cochinchinensis Lour., 243, 244. cochinchinensis Lour. var. viridis, 244. F Favolus Fries, 484. philippinensis Berk., 484, 489. spathulatus (Jungh.) Bres., 484, 487. tener Lev., 484, 489. Feed for dairy cows, 101. FERNANDEZ, RICARDO, see HaucHwovut, LANTIN, and FERNANDEZ. Ficus asperrima, 149. benjamina Linn., 231. chlorocarpa Benth., 231. hispida, 149. religiosa L., 482, 488, ulmifolia, 149, variegata Blume, 146, 231. sp., 480, 481-484, 488. Fissistigma Griff., 128-130. africanum (Benth.) Merr., 130. balansee (Aug. DC.) Merr., 180, beccarii (Scheff.) Merr., 181, bicolor (Roxb.) Merr., 131, borneense (Mig.) Merr., 181. chrysosericeum (Finet & Gagmep.) Merr., 131, cinerascens (Mia.) Merr., 1381. cylindricum (Maing.) Merr., 131. elegans (Wall.) Merr., 131, fagifolium (Ridl.) Merr., 131, fulgens (Wall.) Merr., 131. glaucescens (Hance) Merr., 132. hypoglaucum (Miq.) Merr., 132. kentii (Blume) Merr., 132, kinabaluense (Stapf) Merr., 132, korthalsii (Miq.) Merr., 132, —— (Hook. f. & Th.) Merr. latifolium (Dunal) Merr., 132, 136. leichhardtii (Benth.) Merr., 133, litsaefolium (King) Merr., 133. longipetalum (Ridl.) Merr., 133. mabiforme (Griff.) Merr., 133, maccreai (F. Muell.) Merr., 133, Maingayi (Hook. f. & Th.) Merr., 133, Manubriatum (Wall.) Merr., 134, oblongum (Craib) Merr., 134, oldhamii (Hemsl.) Merr,, 134, ovalifolium (Ridl.) Merr., 134, ovoideum (King) Merr., 134, pallens (Finet & Gagnep.) Merr., 134. paniculatum (Ridl.) Merr., 134, parviflorum (Scheff.) Merr., 135, polyanthoides (Aug. DC.) Merr., 135, Index Fissistigma Griff.—-Continued. polyanthum (Wall.) Merr., 135. prismaticum (Hook. f. & Th.,) Merr., 135, punctulatum (Baill.) Merr., 135. rigidum (Ridl.) Merr., 185. rubiginosum (A. DC.) Merr., 135, rufinerve (Hook. f. & Th.) Merr., 136. rufum (Presi) Merr., 196, seandens Griff., 180, 136. schefferi (Pierre) Merr., 136, schlechteri (Diels) Merr., 136. thorelii (Pierre) Merr., 136, tonkinense (Finet & Gagnep.) Merr., 136, uhrii (F. Muell.) Merr., 137, uonicum (Dunn) Merr., 137. verrucosum (Hook. f. & Th.) Merr., 137, wallichii (Hook. f. & Th.) Merr., 187. zippelii (Mig.) Merr., 137. Flacourtiaceae, 247, 543, Flacourtia chinensis Clos., 247. integrifolia Merr., 543. Flagellaria repens Lour., 228. FLEUTIAUX, ED., Melasidae nouveaux (Coléoptéres) récoltés par C. F. Baker, 445. Flora of Guam, additions to the, 539, Foochow cedar and Randai cedar, volatile constituents of, 362. Formosan termites, methods of preventing damage by, 319. Fornax diapodioides Fleut., 449. dorsalis Fleut., 447, fusiformis Fleut., 449, melanopterus Fleut., 449. scutellaris Fleut., 448. subacuminatus Bonvouloir, 449. tenuis, 449, umbilicatus Fleut., 448, venustus Bonvouloir, 449. Freysuila Aleman, 157. Fritillaria cantoniensis Lour., 229. Fulgoroidea, 3801. FUNKHOUSER, W. D., New records and species of Philippine Membracidae, 15. Fureraea gigantea Vent., 480, 488. G Ganoderma Karst., 482. lucidum (Leys.) Karst., 482, 487. lucidum (Leys.) Karst. var. laceatum Pat., 482, 488. Gargara grisea Funkh., 25. nigrofasciata Stal, 24. nitidipennis Funkh., 24, pulchripennis Stal, 25. Ppygmaea Walk., 24. rugonervosa Funkh., 24. tubereulata Funkh., 24, 25. | varicolor StaAl, 23. | Garnotia Brongn., 540. | strieta Brongn., 640. | Geaster Micheli, 487. | velutinus Lloyd, 487, | Geijerolyma Froggatt, 157. Index Gessius Dist., 210, 211, 216, 217. malayensis Baker, 217, pallidus Baker, 217, 218, e verticalis Dist., 217. Ginalloa, 234. Goniothalamus inaequilaterus K. Schum. & Lauterb., 1387. longirostris Scheff., 187. Gonium, 516. Gordonia benguetica Burkill, 475, 477, 478. fragrans Merr., 476, 478. luzoniea Vidal, 475-478. penangensis, 477. polisana Burkill, 475-477, 478. subclavata Burkill, 475, 477, 478. welborni Elm., 476, 478. Gramineae, 227, 540. GROFF, ELIZABETH H., Soy-sauce manu: facturing in Kwangtung, China, 307. Grypomachaerota Schmidt, 68, 73, 76. borneensis Baker, 73, 74. breviceps Baker, 74, 75. tricolor Baker, 74, 75, 76. turbinata Schmidt, 73, 74. Guam, additions to the flora of, 539. Guepiniopsis Pat., 481. spathularius (Schw.) Pat., 481, 487. Gymnopetalum Arnott, 256. chinense (Lour.) Merr., 256. cochinchinense Kurz, 256. Gynura Cassini, 260. pinnatifida DC., 260. * segetum (Lour.) Merr., 260. Gypeétinae, 54. Gypaétus barbatus (Linn.), 81," 33, 47, 54. Gypona, 210. Gyponinae, 210. H Halixetus leucocephalus (Linn.), 31, 33, 35, $7, 51, 52. Haticti, 13. Halictinae, 9. Halictine bees of the Philippine Islands, 9, 269. Halictus Latreille, 9, 10, 269. adonidiae Ckll., 10, 12. baguionis Crawf., 270, 271, 277. banahaonis Ckil., 272, 275. banahaonis var. macerula Ckll., 275. earoli CkIl., 270, 277. cyanescens CkIil., 10, 12, davaonis Ckll., 271, 281. eschscholtzi Ckll., 270, 275. fulvovittatus Ckil., 271, 280, 281. gedehensis Fr., 272. gedensis Cklil., 272. imuganensis Ckll., 10, 12, 13. itaminus Ckll., 270-272. lionotulus Ckll., 270, 271, 278, 279. luzonieus Strand, 270, 271, 273. manilae Ashm,, 2738. manilae Ckil., 278. meanilae Strand, 275, 276. 581 Halictus Latreille—Continued. megregori Ckll., 270, 277. melanurops Ckil., 270, 272, 277, 281. nesiotus Crawf., 271, 273. nesiotus domitus CkIl., 271, 273. (Nesohalictus) robbit Crawf., 269. oligostictus Ckll., 271, 278, 279. opisthochlorus Ckll., 269, 270, 276, perangulatus CkIl, 277. pervarians Cklil., 10, 11, 13. philippinensis Ashm., 270, 271, 274. philippinensis var. nigritarsellus CkIL, 270, 274. postiucens Ckll., 271, 280. roepkei Fr., 276. scapalis Ckll., 270, 271, 279. scintillans CkIL, 271, 279. subpurpureus Ckil., 10, 11, 13. taclobanensis Ckll., 10, 12, 138. thoracicus Friese, 274, 275. thoracicus var. merescens CkIl., 275, thoracicus sublustrans Ckil., 271, 274. Harpyopsis, 54. HAUGHWOUT, FRANK G., A method for labeling slides used in routine stool examinations, 535. HAUGHWOUT, FRANK G., LANTIN, PE- DRO T., and FERNANDEZ, RICARDO, A case of acute mania associated with Plasmodium vivax infection, 563. Hayun lago, 542. Yedyotis Linn., 544. . fruticulosa (Volk.) Merr., 544, Hemigraphis Nees, 256. chinensis T. Anders., 256. procumbens (Lour.) Merr., 256. Hemiitrioza, 184. é Heritiera littoralis Dry., 481, 489. Hermit crabs associated with sea anemones, habits of, 81. Heterochaete Pat., 481. tenuicula (Lev.) Pat., 481, 488-490. Heteroneura Crawf., 141, 152. oceanica Crawf., 152, Heteropsylla Crawf., 141, 153. longicornis Crawf., 153. Hevaheva Kirk., 185. Hevea brasiliensis (HBK) Muell.-Arg., 480- 484, 486, 489. Hexagona Fr., 483. thwaitesii Berk., 483, 489. thwaitesii Berk. var. retropicta Bres., 483, 490. Hibiscus Linn., 246, rosa-sinertsig Linn., 19. surattensis Linn., 246, Hindola Kirk., 67. Hirundo javanica Sparrman, 533. Hisingera racemosa Sieb. & Zuce., 247, Holmes, S. J., review of his The Elements of Animal Biology, 123. Homalanthus populneus Pax, 274, 275. Homalictus Ckli., 13. Homalocyrtus, 554. * 582 : Index Homoptera, 68, 139, 521, Homotoma Guerin, 156, 161, 162. bakeri Crawf., 162. bilineata Crawf., 162. distincta Crawf., 162. pacifica Crawf., 162, radiatum Kuwayama, 162. Hopea plagata Vidal, 348. Hoplonomia Ashm., 1. quadrifasciata Ashm., 2. Hova imbricata Callery, 265. imbricata Callery ex Deene. forma typica | Kds., 263, 264. imbricata Callery forma basi-subcordata Kds., 264, 265. imbricata Decne., 265. : imbricata DC., 264. maxima Kds., 265. | pseudomaxima Kds., 265, | Hymenochaete Lev., 481. adusta (Lev.) Bres., 481. attenuata Lev., 481, 489. | pavonia Pat., 481, 487. perpusilla Pat., 481, 489. I Iberia Kirk., 211. Indigofera, 177. rotundifolia Lour., 242. | Insia bijuga Gray, 348. Tpil, 348, ; Ipomoea batatas (L.) Poir., 486, 489, ‘Ipo toxicaria Pers., 231. Ischaemum involutum Forst., 540. Ixonanthes longipedunculata Merr., 557. J Jatropha curcas Linn., 414. Jumping plant lice, 139, Juniperus virginiana L., 864. K Kentia Mia., 130. Koelreuteria Lakman, 246. bipinnata Franch., 246. Koordersiodendron pinnatum Merr., 485, 489. KOORDERS, S. H., Notiz iiber Hoya imbri- cata Callery ex Decaisne und Hoya pseudomaxima Kds. in den Filipinen auf Grund von einigen Herbar-exempla- ren des Bureau of Science in Manila, 268. Krisna Kirk., 209, 210-213, colorata Baker, 214, 216., magna Baker, 214, 216. minima Baker, 213, 214, muirii Baker, 214, 215, nigrifrons Baker, 213, 215. olivascens Baker, 214, 215. olivascens var. singaporensis Baker, 214, 216, penangensis Baker, 213, 215. sherwilli Dist., 212. Krisna Kirk.—Continued. simillima Baker, 213, 215. straminea Wik. var. indicata Whlk., 212. strigicollis, 211, 214, 215, 217, 218. Kuwayama Crawf., 185. hirsuta Crawf., 201. Kwanegtung flora, additional notes on, 225. L Labiatae, 255. LANTIN, PEDRO T., see HaucHwout, LANTIN, and FERNANDEZ. Las-aga, 540. Lauraceae, 235. | Laurus cubeba Lour., 235, 236. | Leaf-mining buprestids, 289. Leguminosae, 239, 542. | Lentinus Fr., 486. connatus Berk., 486. dactyliophorus Lev., 486. dichrous Lev., 486. exilis Kl., 486, 487. Lenzites Fr., 483. applanata Fr., 488, 490. palisoti Fr. 483, 487-490. tenuis Berk., 488, 487, 489, 499. | Leptocentrus leucaspis Walk., 19. reponens Walk., 19. | Leptynoptera Crawf., 141, 147, 184. sulfurea Crawf., 147. Leucaena glauca (Linn.) Benth., 480, 481, 483, 485-487, 489. Leucoporus Quel, 484, gallo-pavonis (Berk,) Pat., 484, 488. grammocephalus (Berk.) Pat., 484, 488. Leucotermitinae Holm., 324. Leucotermes Silvestri, 324. flaviceps Oshima, 820, 324, 326, 341, 879. (Reticulitermes) flaviceps Oshima, 324. flavipes Shiraki, 324. speratus Kolbe, 320, 326. Leuronota Crawf., 184, LEVINE, C. O., Milk produced in southern China, 91. Levinson, Abraham, notice of his Cerebro- spinal Fluid in Health and in disease, 572, Ligustrum, Linn., 258. groffiae Merr., 258, Liliaceae, 229. . sein Thunb., 237, strychnifolia (Meisn.) F.-Vill., 287. subcaudata (Merr.), 237, Litsea Lam., 235, 485, 489, citrata Blume, 235, 236. ecubeba (Lour.), 235. piperita Juss., 235. Loganiaceae, 252, Longicornia, 546, 547, Lonicera Linn., 256. dasystyla Rehder, 266. Loranthaceae, 232. Index Loranthus, 232. chinensis DC., 233, estipitatus Stapf., 232, 233. fordti Hance, 234. ‘i levinei Merr., 233, pentandrus Linn., 232. parasiticus (Linn.) Merr., 232, 233. philippensis Cham. & Schlecht., 232. seurrula Linn., 232-234. yadoriki Sieb., 233. Luffa cylindrica (Linn.) Roem., 486, 489. Lycoperdaceae, 487. Lycoperdon Tournef., 487. polymorphum Vitt., 487. roseum Zoll., 487, 489. Lysimachia Tournef., 250. alfredi Hance, 250. eandida Lindl. var. depauperata Merr., 250. M Macaranga tanarius (Linn.) Muell,-Arg., 483, 489. Machaeropsis Mel., 67, 68. Machaerota Burm., 67-69, 74. ensifera Burm., 68, 69, 71. fusca Baker, 69, 72, luzonensis Schmidt., 69, 71. notoceras Schmidt, 69, 70. philippinensis Baker, 69, 70. Machaerotinae (Cercopidae), Malayan, 67. Machaerotini, 69, 75. Machilus Nees, 236. levinei Merr., 236. phoenicis Dunn, 287. Macroceps Sign., 210, 211. ‘Macrohomotoma Kuwayama, 157. Maholoc layu, 544. Malaxa Melichar, 523. bakeri Muir, 523. javanensis Muir, 524. nigra Muir, 524. obtusipennis Muir, 523, Malayan Delphacidae (Homoptera), 521. Mallotus moluccanus (Linn.) Muell.-Arg., 19, 20. sp., 488, 484, 486, 489, Malvaceae, 246. Mangifera, 177. eaesia Jack, 485, 489. indica Linn., 480, 482, 488, 485, 486, 4989. Mansonia uniformis Theob., 287. MARANON, JOAQUIN, see vet Rosario and Maranon. Marasmius Fr., 486. pilopus Kalchbr., 486, 489. Marcanthus cochinchinensis Lour., 242. Maxudea Schmidt, 68, 76. crassiventris, Schmidt, 77. schmidtii, 76, 77. Maxudeini, 69, 76. Measurements of buffalo cows, 103, 583 Megatrioza Crawf., 139, 185, 186, 192, 201, 203. armaia Crawf., 192-194. armata ochreata Crawf., 198, 195, asiatica Crawf., 194, 197, 198. eugenioides Crawf., 193, 198. gigantea Crawf., 194, 199. hirsuta Crawf., 194, 201. magnicauda Crawf., 194, 197. - mMelanoneura Crawf., 194, 200. palmicola Crawf., 193. robusta Crawf., 193, 200. stylata Crawf., 194, 196. vitiensis (Kirk.), 193-195, Melanopus Pat., 484. guilfoleyi (Berk.) Pat., 484. Melanthium cochinchinensis Lour., 230. Melasidae nouveaux, 445, Meliaceae indet., 480, 489. Mellesis Bezzi, 415, 428, 434, 485, 442. aequalis Coqg., 438. bioculata Bezzi, 413, 484, 435, 4387, brachycera Bezzi, 412, 428. conopoides de Meij., 412, 438, 440, 442. crabroniformis Bezzi, 412, 438. destillatoria Bezzi, 412, 488. eumenoides Bezzi, 412. longicornis Wied., 412, 438. nummularia Bezzi, 411, 418, 435, 441. pedunculata Bezzi, 411, 418, 485, 439, 442, sphaeroidalis Bezzi, 412. subsessilis Bezzi, 413, 484, 435. vespoides Dol., 438. Melochia Linn., 543. hirsutissima Merr., 543. villosissima (Presl) Merr., 543, Melodorum Hook. f. & Th., 130. africanum Benth., 130. 197, 198. * arboreum Lour., 125, 126, 129. auct. non Lour., 129, auct. plur. non Lour., 130. balansae Aug. DC., 130. - bancanum Scheff., 134. beccarii Scheff., 131. bicolor Hook. f. & Th., 181. borneense Miaq., 131. chrysosericeum Finet & Gagnep., 131. cinerascens Miq., 131. clavipes Hance, 128. clementis Merr., 136. ceylindraceum Boerl., 131. eylindricum Maing., 131. elegans Hook. f. & Th., 131. fagifolium Ridl., 131. fruticosum Lour., 125-130. glaucescens Hance, 132, glaucum Scortech., 128. grifithiti Hook. f. & Th., 136. hypoglaucum Miq., 132. kentii Hook. f. & Th., 132. kinabaluense Stapf, 132. korthalsii Miq., 132. lanuginosum Hook. f. & Th., 132, 584 Index Melodorum Hook. f. & Th.—Continued. latifolium Hook. f. & Th., 126, 132, 134, 136. latifolium Hook. f. & Th., var. ovoidea King, 134. latifolium (Dunal) Hook. f. and Th., 126. lefevrt Baill., 137. leichhardtii Benth., 133. litsaefolium King, 133. longipetalum Ridl., 133. maccreai F, Muell., 1383. o macranthum Kurz, 133. maingayi Hook. f. & Th., 133. manubriatum Hook. f. & Th., 134. micranthum Warb., 1387. mollissimum Mig., 132, 133. oblongum Craib, 134, oldhamii Hemsl., 134. ovalifolium RidL, 134. pallens Finet & Gagnep., 134, paniculatum Ridl., 134. parviflorum Scheff., 135. pisocarpum Hook. f. & Th., 133. polyanthoides Aug. DC., 185. polyanthum Hook. f. & Th., 135. punctulatum Baill., 135. pyramidale Maing., 133, rigidum Ridl., 185. rubiginosum Hook. f. & Th., 135. rufinerve Hook. f. & Th., 1386. rufum Merr., 136. schefferi Pierre, 136. sphaerocarpum (Blume) Mia., 186. thorelii Pierre, 136. tonkinense Finet & Gagnep., 186. uhrit F. Muell., 137. uonicum Dunn., 137. verrucosum Hook. f. & Th., 187. wallichti Hook. f. & Th., 137. zippelii Miq., 187. Membracidae, records and species of, 15. Memecylon sp., 487, 489. Menispermaceae, 235. MERRILL, E. D., On the application of the generic name Melodorum of Loureiro, 125; Additional notes on the Kwang- tung flora, 225; Additions to the flora of Guam, 539. Merrillosphaera, 512, africana (West) Shaw, 512, carteri (Stein) Shaw, 512. Mesneya Pierre, 127. Meschomotoma Kuwayama, 159. camphorae Kuwayama, 160. Mesolecanium rhizophorae Ckll., 385. Mesotermitidae Holm., 321. Metapocyrtus bucasanus Schultze, 556. (Orthocyrtus) consobrinus Schultze, 554, (Orthocyrtus) insulanus Schultze, 552. (Orthocyrtus) malayanus Schultze, 552, 553, 554. (Orthocyrtus) ornatus Schultze, 555, (Orthocyrtus) subsp. atratus Schultze, 553, : | Metapocyrtus bucasanus Schultze—Cont. (Trachycyrtus) multisquamosus Schultze, 556. vielaceus Schultze, 555, 556. Metapsylla Kuwayama, 168, 169. Metatermitidae Holm., 326. Metrosideros, 186, 187. Microporus Beauv., 484. affinis (Nees) Pat., 484, 490. 1 erenatus (Berk.) Pat., 484. microloma (Lev.) Pat., 484, 488. sanguineus (Lev.) Pat., 485, 487, 489. xanthopus (Fr.) Pat., 485, 487-490. Micropus subfurcatus, a nesting place of, 533. Milk analyses, 93-95, 102. produced in southern China, 91. Mitrella Migq., 180. becearii Diels, 181. kentii Mig., 132. schlechteri Diels, 186. Mitrephorae, 126, 129, 180. Mitrephora fulgens Hook. f. & Th., 131. thorelii Pierre, 130. Modiglianella Schmidt, 67, 68. Molave, 348. Monacrostichus Bezzi, 416, 442. citricola Bezzi, 411, 418, 442. ¥ Monkey-eating eagle, osteological and other notes on, 31. Moraceae, 231. Morphnits guianensis (Daudin), 31, 33. Morus alba Linn., 485, 489. Mound-building termites, biology of, 59. Mucuna Adanson, 242. cochinchinensis (Lour.) A. Chev., 242. nivea. W. & A., 242, 248, MUIR, FREDERICK, Some Malayan Delpha- cidae (Homoptera), 521. Murraya exotica, 386. Musa sapientum L., 483, 486, 490. Mussaenda Linn., 258. Mussaenda frondosa Linn., 259. parviflora Miq., 258. Mycopsylla Froggatt, 157, 162. Myosotis peduncularis Trev., 255. Myrtaceae, 249. Myrtus chinensis Lour., 252. zeylanica Lour. non Linn., 252. Myzomyia febrifera Banks, 288. rossii Giles, 284, 285. N Nandina Thunb., 234. domestica Thunb., 234, Nectandra rodiae, 348. Neolitsea subcaudata Merr., 237. Neotriozella Crawf., 188, 185, Neottia sinensis Pers., 230. Nepenthes alata Blanco, 545. merrilliana Macf., 545. truncata Maef., 545. | escenario rr aan eee Ne Index Nesiope Kirk., 155, 157, 160, 161, 163. heterocephala Crawf., 161. heterocephala intermedia Crawf., 161 ornata Kirk., 161. Nesohalictus Crawf., 269, 270. lativentris, 269. robbii Crawf., 269, 270, 272. Nesting place of Micropus subfurcatus, 533. Neuromachaerota Schmidt, 68. Nomia, 1. ardjuna Ckil., 7. aureobalteata Cam., 7. aurifrons Sm., 7. basalis Sm., 8. buddha Westw., 8. curvipes Fabr., 1. dimidiata Vachal, 1, 8. ellioti, 4. elongata Fr., 5. elongata “Friese,” Ckll., 5. elongatula CkIL, 2, 5. goniognatha Ckll., 2, 7. incerta, 3. incerta ‘“‘Gribodo” CkIL, 3. iridescens Smith, 1, 2, 5. tridescens ‘“‘Smith,” Ckll., 4. iridescens var. rhodochlora CkIll., 5. iridescens var. ridleyi CkIl., 4. kangrae Nurse, 7. lautula Ckil., 2, 6. levicauda Ckll., 2, 5, 6. longitarsis Ckll., 1, 2, 4. longitarsis eboris Ckll., 4. lusoria Ckll., 1, 2, 5. notha, 2, 3. palavanica Ckll., 2, 6. philippina Vachal, 1, 8. philippinensis (Fr.), 2, 6. quadridentata Sm., 4. quadrifasciata (Ashm.), 1, 2. quadrifasciata (Ashm.) Ckll., 2. quadrifasciata notha CkIil., 1. quadrifasciata notha (Ckil. ined.), 2. quadrifasciata var. aurantia Ckil., 3, quadrifasciata var. viridans CkIl., 3. recessa Ckil., 2, 7. simplicipes Fr., 3. strigata (Fabr.), 1, 2, 4. strigata Lepel., 4. strigata ridleyi, 5. strigata var. ridleyi (CkIl.), 4. takauensis philippinensis Friese, Ckll., 6. thoracica Sm., 1, 8, 7. thoracica stantoni (Ashm.), 2, 3. thoracica stantoni (Ashm.) CkIl, 3. Nomiocides Schenck, 9. comberi CkIl, 9. dapitanellus CkIl, 9, 10. melanogaster CklIl., 10. parvula (Fabr.), 9. pulchella Schenck, 9. punjabensis (Cam.), 9. » valdezi Ckil., 9, 10. 585 0 Ocypoda arenaria Catesby, 87. Odontotermes Holm., 326. (Cyclotermes) formosana Oshima, 826. (Cyclotermes) formosanus Holm., 326. formosanus Holm., 326. (Cyclotermes) formosanus $26. formosanus (Shiraki), 348, 379. formosanus, 320, 328, 338, 350, 353, 377, 378. (Shiraki), | Olearus, 186. Oldenlandia fruticulosa Volk., 544. Opalina, 399. Ophiospermum sinense Lour., 248. Orchestia agilis, 392. Orchidaceae, 230. Ormosia Jackson, 240. fordiana Oliv., 241. hainanensis Gagnep., 240. Ornithogallum sinense Lour., 229. Ortalidae, 414. Orthocyrtus schénherri Waterh., 552, 553. OSHIMA, MASAMITSU, Formosan termites and methods of preventing their dam- age, 319. Ostwald, Wolfgang, review of his A Hand- book of Colloid-Chemistry, 491. Ozymitra bassiaefolia Teysm. & Binn., 135, P Pachymachaerota Schmidt, 67, 68. Pachyrrhynchus absurdus Schultze, 550. ardentius Schultze, 550, erichsoni Waterh., 551. signatus Schultze, 551. venustus Waterh., 549. virgatus Schuitze, 549, 550. virgatus subsp. insulanus Schultze, 550, 558. Paguristes arrosar Herbst, 82. asper de Haan, 83, 87. deformis H. Milne-Edwards, 82-85, 87. oculatus Fabr., 82. Pahudia Miq., 240. © xylocarpa Kurz, 240. Pandion, 52. Pandorina, 52 . Papualthia longirostris (Scheff.) Poralictus Robertson, 9. Paramoecium caudatum, 395, coli Stein, 390, 404. Paranomia stantoni Ashm., 3. Parashorea plicata Brandis, 485, 490. Parinarium sp., 481, 490. Parkia javanica (Lam.) Merr., 480, 482, 483, 485, 490. roxburghii G. Don, 480, 482, 483, 485, 490. Diels, 187. 586 Paurocephala Crawf., 141, 148, 150, 152. brevicephala Crawf., 149, 150. brevicornis Crawf., 142, 143. conigera Crawf., 149, 151. ilicis (Ashm.) Crawf., 149. maculata Crawf., 149, 151, magnifrons Crawf., 140, 149, 152. minuta Crawf., 148, 150. orientalis Crawf., 149. psylloptera Crawf., 146, 148-151. psylloptera setifera Crawf., 146, 148, 150. Pauropsylla Riibs., 141, 142, 145, 147, 150, 152, 163. apsylloides Crawf., 142, 144, bakeri Crawf., 145. brevicephala Crawf., 150. depressa Crawf., 142. ficicola Kieff., 142. floccosa Crawf., 142. globuli Kieff., 142. nigra Crawf., 142, 143. spondiasae, 142. : triozoptera Crawf., 142, 146, 167, 184. triozoptera setifera Crawf., 146, 150. tuberculata Crawf., 142, 146. udei Riibs., 142, 145, verticis Crawf., 142, 1438, Pauropsyllinae Crawf., 67, 139, 140, 168. Pectinarophyes Kirk., 68. Pediopsis, 210. Penthimiinae, 210. Perdita, 9. Peregrinus Kirk., 529. maidis (Ashm.), 529. Pergularia chinensis Spreng., 254, sinensis Lour., 254. Pericampylus glaucus (Lam.) Merr., 235. PERKINS, GRANVILLE A., The rancidity of Philippine coconut oil, 463; review of Ostwald’s A Handbook of Colloid- Chemistry, 491, Persea cubeba Spreng., 235. Petalotoma brachiata DC., 249. Petesia nitida Bartl., 544. Phaeopteron Buckt., 141. lentigincsum Buckt., 154. lentiginosum Crawf., 154. Philippine Anthribidae, 560. bees of the genus Nomia, 1. coconut oil, rancidity of, 463. Membracidae, records and species of, 15. Phrynium Willd., 230. capitatum Willd., 230. parviflorum Roxb., 230. pPlacentarium (Lour.) Merr., 230. Phyllanthus reticulatus Poir., 23. Phyllodes placentaria Lour., 230. Piroplasma bigeninum, 102, Pithecolobium Mart., 239, 240, attopenense Pierre, 240. balansae Oliv., 240. turgidum Merr., 239, Pithecophoga, 32-38, 40, 41, 43-45, 47, 49, 51, 52, 54, | _defferyi Grant, 31, 40, 46, 48-50, 54. serine prtainttnins Index Plasmodium falciparum, 566, 567. vivax, 565-567. Platydorina, 516. Plectronia Linn., 257. horrida Benth. & Hook. f., 257. levinei Merr., 257, parviflora Bedd., 257. parvifolia Benth., & Hook. f., 257. Pleodorina Shaw, 513. californica, 513, 516. ilinotsensis Kofoid, 513, 516. Polyalthia Biume, 127-180. aberrans Maing., 127-129. (?) aberrans Maing., 128. affinis Teysm. & Binn., 127-129. kentii Blume, 132. siamensis Boerl., 125, 127~129. Polychaetophyes Kirk., 68. Polygala Linn., 243. tenuifolia Willd., 243. Polyporaceze, 243, 482. Polyporus Micheli, 482. lignosus (Kl.) Bres., 482, 488. rugulosus Lev., 482, 489, 490. Polyrhachis (Myrmhopla) dives 22-24, Pongamia elegans Grah., 241. Popowia Endl., 129, 130. aberrans Pierre, 127, 128. diospyrifolia Pierre, 127, 128. . Portiand cement raw mixture, analysis of, 107, Potentilla Linn., 2389. discolor Bunge, 239, Pothos Linn., 228. loureirii Hook. & Arn., 228. microphyllus Schott., 228. repens (Lour.), 228, terminalis Hance, 228. Poupartia Commerson, 245. chinensis Merr., 245. fordit Hemsl., 245. Premna odorata Blanco, 480, 490. Primulaceae, 250. Prosopis vidaliana Naves, 488, 486, 490. Protelater bakeri Fleut., 445. Prunella Linn., 255. vulgaris Linn., 255. Psidium guajava Linn., 481, 484, 485, 490. Psilotum nudum Griseb., 539. Psyllidae, 139, 155. Psyllinae, 140, 156, 162, 167. Psylla Geoffroy, 167, 168, 177, 178. acaciae-baileyanae Froggatt, 180. acaciae-dealbatae Froggatt, 180. acaciae-decurrentis Froggatt, 181. acaciae-juniperinae Froggatt, 180. acaciae-pendulae Froggatt, 179. acaciae-pycnanthae Froggatt, 179. arisana Kuwayama, 179. bakeri Crawf., 179, 182. eandida Froggatt, 179. ecapparis Froggatt, 181. coccinea Kuwayama, 178. colorada Crawf., 178. Sm., 20, Index Psylla Geoffroy—Continued. compta Crawf., 179, 183. crenata Crawf., 180, 181. frenchi Froggatt, 180. fumosa 180, 181. gracilis Froggatt, 179. isitis Buckt., 139. kiushuensis Kuwayama, 179. leprosa 179, 181, lidgetti Maskell, Froggatt, 180. muiri 180, 183. schizoneuroides Froggatt, 181. simlae Crawf., 179. spadica Kuwayama, 180. 571. sterculiae Froggatt, 179. The Medical Clinics of North America, torcensis Kuwayama, 180. Volume 2, Baltimore Number, No. 6, tripunctata Kuwayama, 180. 571 Psyllopa Crawf., 172. obscura Crawf.,. 177. punctipennis Crawf., 177. Pterocarpus sp., 481, 490. Pterocymbium ‘tinctorium Merr., 482, 490. Pterolobium R. Brown, 241. rosthornii Harms, 241. Purohita Distant, 521. nigripes Muir, 521. Putoniessa Kirk., 211. Pyenothrix, 399. monocystoides, 398. Pygeum Gaertn., 237. henryi Dunn, 288. latifolium Migq., 238. . topengii Merr., 237. Pylocheles miersii Alcock & Anderson, 81. Pyramidanthe Migq., 130, 135. macrantha Kurz, 133. rufa Miq., 1385. Pyrgonota bifoliata Westw., 17. bifurca Stal, 17. semperi Stal, 17. Q Quereus sp., 484, 485, 490. Quinguatrus Dist., 67. R Rana esculenta, 390, 391. palustris Leconte, 392. temporaria, 390. Randia Linn., 259. ; acuminatissima Merr., 259, densiflora Benth., 260. racemosa (Cav.) F.-Vill., 260. Ranunculaceae, 234, Ranuneulus Linn., 234. diffusus DC., 234. REV [EWS—Continued. 587 Levinson, Abraham, Cerebrospinal Fluid in Health and in Disease, 572. Oswald, Wolfgang, Colloid-Chemistry, 491. Quarterly Medical Clinics, A Handbook of A series of Consecutive Clinical Demonstrations and Lectures, 317. The Medical Clinics of North America, Volume II, Philadelphia Number, 8, 817. The Medical Clinics of North No. America, Volume 2, New York Number, No. 4, Vischer, A. L., Barbed Wire Disease: a Psychological Study of the prisorer of war, 572. Rhinopsylla Riley, 157, 167, 184. Rhinoscapha, 558. merrilli Schultze, 557. Rhizophoraceae, 249. Rhjzophora, 385. mucronata Lam., 386. Rice, making soy from, 314. Riedleia villosissima Pres], 548. Rosaceae, 237. Rubiaceae, 257, 544. Ruellia chinensis Nees, 256. 5 Saccharum officinarum Linn., 486, 499. Sagartia parasitica Gosse, 82. Sanguisorba Rupp., 238. canadensis Linn., 239. formosana Hayata, 239. officinalis Linn., 238, 289. Sapindaceae, 246. Sapium merrillianum Pax & K. Hoffm., 490. Sarpestus Spang., 211. Saxifragaceae, 237. Schizaeaceae, 539. Schizaea Sm., 539. dichotoma (Linn.) Sm., 539, Schizophyllum Fr., 485, 488. commune Fr., 485, 489, 490. Schizostachyum sp., 486, 490. SCHULTZE, W., Seventh contribution to the Coleoptera fauna of the Philippines, 545. Scilla Linn., 229. chinensis Benth., 229. sinensis (Lour.) Merr., 229. REINKING, O. A., Higher Basidiomycetes from the Philippines and their hosts, Scleria Linn., 541. lithosperma (Linn.) Sw., 541. : a; 479: Scurrula parasitica Linn., 232. Resistance of native and exotic woods, test | Sea anemones, 81, of, 347. Seladonia Robertson, 9. REVIEWS: Selenocephalaria, 210, 211. - Selenocephalus, 209, 211. Semecarpus cuneiformis Blanco, 19. Holmes, S. J., The Elements of Animal Biology, 123. 588 Senecio Tournef., 261. oldhamianus Maxim, 261. Serianthes Benth., 542. nelsonii Merr., 542, Serpentarius, 49. SHAW, WALTER R., Campbellosphaera, a new genus of the Volvocaceae, 493. Shik sun tau, 229. SHUFELDT, R. W., Osteological and other notes on the monkey-eating eagle of the Philippines, Pithecophaga jefferyi Grant, 81. Sideroxylon, 348. Sigmasoma Schmidt, 68, 73. bifaleata Schmidt, 73. Sigmasomini, 69, 73. Siphonodon celastrineus Griff., 481, 490. Sipylus dilatatus Walk., 23. Siva Spin., 211. rosea Bierman, 212. striata Kirby, 212. strigicollis Spin., 212. Slides used in routine stool examinaticns, method for labeling, 535. Smithies, Frank, notice of his Quarterly Med- ical Clinies, A Series of Consecutive Clinical Demonstrations and Lectures, 317. Sogata Dist., 526. dohertyi Dist., 527. 4-spinosa Muir, 526, Sogatopsis Muir, 523. pratti Muir, 523. Solanum verbascifolium Linn., 20, 21, 23. Solenopsis geminata Fabr., 63. Soy bean manufacturing plant, equipment of, 308. sauce, manufacture of, in Kwangtung, 307. Spathodea stipulata Wall., 255. ° Sphaerocoryne Scheff., 127. siamensis Scheff., 128. Spilornis, 54, Spiranthes L. C. Rich., 230, aristotelia (Raeusch.) Merr., 230, australis Lindl., 230. sinensis Ames, 230. Spirostemum, 396. Stegelytra M. & R., 211. Stegelytraria, 210, 211. Stegomyia scutellaris Walk., 287. Stenocranus Fieb., 529. (2) singaporensis Muir, 529. (2?) taiwanensis, 529. Stenopsylla Kuwayama, 185, 203. longicornis Crawf., 203. nigricornis Kuwayama, 203. Stentor, 396. Stephanoon, 516. Sterculiaceae, 543. Sterculia, 179. foetida, 164. Stizolobium niveum O. Ktze., 242. Strobilanthes scaber Hance, 256. Index Strogylocephala Crawf., 163, 165. fasctpennis Crawf., 165. Sirombidae, 83. Strumeta conformis Walk., 417. Strychnos Linn., 252. paniculata Champ., 252, 253. umbellata (Lour.) Merr., 252. Subprotelater guttatus Fleut., 445, Sumac lada, 544. Swietena mahogani, 248. Symplocaceae, 251. Symplocos Jacq., 251. anomala Brand, 252. amomalae, 251. chinensis (Lour.) Desvaux, 252, cochinchinensis Moore, 252. fusonii Merr., 251, siniea Ker, 252. spicata Roxb., 252. Synearpiolyma Froggatt, 169. x Talorchestia longicornis, 392. Tamarindus indica Linn., 486, 490. Targionia merrilli Ckll., 385, 386. mooert (Green), 386. Tartessus Stal, 210, 211. Tartessusaria, 210, 211. Teak, volatile constituents of, 362. Tectoma grandis Linn., 348, 353. Tenaphalara Kuwayama, 155, 157, 160, 166, 167. acutipennis Kuwayama, 163-165. elongata Crawf., 164. fascipennis Crawf., 163, 165. juliana Crawf., 164, 166. malayensis Crawf., 163, 165, 166. suleata, 167. striata Crawf., 164, 166. triozipennis Crawf., 157, 164, 187, Termes, 59. formosana Shiraki, 326. (Leucotermes) flavipes Oshima, 324. (Macrotermes) philippinensis Oshima, 61. vulgaris Shiraki, 326. 163, 184. _Terminalia comintana Merr., 482, 490, tomentosa, 201. Termite-proof building construction, 341, 343, 344, 346. Termites, fungus gardens of, 60. mound-building, biology of, 59. mound-building, swarming of, 61. test tube for rearing, 332. woods not subject to attack by, 348, Tetradacus Miyake, 415, tsuneonis Miyake, 416. Teiranthera citrata Nees, 235. cubeba Meisn., 235. floribunda Champ., 235. polyantha Wall, 235. Thallasoaétus pelagicus (Pallas), 31, 33, 44, 48, 49, 51. ; Theaceae, 247. Index 589 Thelephoraceae, 481. Theobroma cacao Linn., 480, 482, 490. Thrasaétos harpyja (Linn.), 31, 38, 54. Thrinchostoma, 269. Thuarea Pers., 540. involuta (Forst. f.) R. & S., 540. sarmentosa Pers., 540. Thymbris Kirk., 211. Thymelaeaceae, 248. Thysanogyna Crawf., 156, 157. minor Crawf., 157, 158. Tikitiki, analyses of extract, 222. physico-chemical evaluation of extract, 221. Timber, resistance and chemical properties of, 356. resistance and physical properties of, 354. Timonius (Rumph.) DC., 544. nitidus (Bartl.) F.-Vill., 544. Trachys, 295. minuta Linn., 289. Trametes Fr., 482. aspera (Jungh.) Bres. 482, 488. badia Berk., 482, 490. flava (Jungh.) Pat., 482, 490. persoonii Mont., 482, 487-490. scopulosa (Berk.) Bres., 488, 488. Tremellaceae, 481. Trema sp., 484, 490. Tricentrus acuticornis Funkh., 22, attenuatis Funkh., 20. eapreolus Walk., 20. convergens Walk., 20, 22. fairmairei Stal, 20, 22, 24. robustus Funkh., 19. Trichochermes Kirk., 185. bicolor Kuwayama, 185, 191. Trichomonas, 404, 405. Tridacus Bezzi, 415, 416. Trigonotis Steven, 255. peduncularis (Trev.) Benth., 255. Triozinae, 140, 147, 162, 167, 184. Trioza Foerster, 189, 147, 185, 186, 198, 198. analis Crawf., 188. asiatica Crawf., 197. banksiae Froggatt, 188. brevifrons Kuwayama, 188. earnosa Froggatt, 187. easuarinae Froggatt, 187. circularis Froggatt, 189. curvatinervis Foerster, 189. diospyri Ashm., 193, 197. diptera Crawf., 189, 191, divisa Crawf., 185 186, 191. dobsoni Froggatt, 186. eucalypti Froggatt, 186. eugeniae Crawf., 195, 196. eugeniae Froggatt, 189. eugenioides Crawf., 198. fletcheri Crawf., 189, 190. formosana Kuwayama, 188. galii Foerster, 188. gigantea Crawf., 187. hawaiiensis Crawf., 187. 168745, 7 Trioza Foerster—Continued. hyalina Crawf., 187. iolani Kirk., 187. jambolanae Crawf., 188. lanaiensis Crawf., 187. luzonensis Crawf., 189. magna Kuwayama, 188, 190. multitudina (Tepper), 187. nigra Kuwayama, 188. nigriceps Kuwayama, 188. ohiacola Crawf., 186. . Oleariae Froggatt, 186. orbiculata Froggatt, 186. pullata Crawf., 187. remota Frst., 188. salicivora Reut., 189. silacea M. D., 189. striola Flor., 189. tasmaniensis Froggatt, 186. tenuicona Crawf., 187, 190. (Trichochermes) bicolor Kuwayama, 186. (Trichochermes) hyalina Kuwayama, 189. tristaneae Froggatt, 188. urticae Linn., 187. vanuae Kirk., 198. viridula Zett., 189. vitiensis Kirk., 195, 196. Tristania, 188. Triticum ciliare Trin., 227. Triion taeniatus, 390. Trobolophya, 301, 302. benguetensis Baker, 302, 304, jacobsoni Mel., 302-304. melichari Baker, 301, 302, montana Baker, 302, 308, 304. penangensis Baker, 302, 303. philippinensis Baker, 301, 302, 304. Tropidocephala Stal, 522. dryas (Kirk.), 522. malayana Mats., 522. Trypaneidae, 414. Tyora Walk., 157, 159. hibisci Froggatt, 159, 160. indica Crawf., 159, stereculiae Froggatt, 159. U UICHANCO, LEOPOLDO B., General facts in the biology of Philippine mound- building termites, 59. Ultra violet rays in modern dermatology, 318. Unona, 126, 127, 129. acutifiora Dunal, 126. dumetorum Dunal, 126, 128. kentit Blume, 182. latifolia Dunal, 126, 132. lucida DC., 126. macrantha Kurz, 138. mesneyi Pierre, 127-129. polycarpa DC., 126. section Melodorum Dunal, 126. sphaerocarpa Blume, 136. sylvatica Dunal, 126, 129. xylopoides Dunal, 126. 590 Upachara Distant, 522. straminea Muir, 522. Uroaétus audax Lath., 49. Uvaria bicolor Roxb., 131. bicolor Wall., 137. elegans Wall, 131. fulgens Wall., 131. fulva Wall., 135. latifolia Blume, 126, 132. longifolia Blume, 132. mabiformis Griff., 133. manubriata Wall., 134. nervosa Wall., 135. polyantha Wall., 135. rubiginosa A. DC., 185. rufa Wall., 135. tomentosa Wall., 132. Uvularia chinensis Ker., 229. Vv Valiaris sinensis G. Don, 254. Vertebral artery, abnormalities of, 451. Vischer, A. L., notice of his Barbed Wire Disease: A Psychological Study of the Prisoner of War, 572. Viseum Linn., 234. angulatum Heyne, 234. stipitatum Lecomte, 234. vitici innascens Camell., 232. Vitex littoralis Dene., 348, Volvocaceae, 498, 495, 504, 505, 510. Volvoceae, phylogeny of, 516. Volvox, 497, 511, 513. africanus West, 504-506, 512, 515, 516. aureus Ehrenberg, 512-516. aureus Klein, non Ehrenberg, 512. earteri Stein, 493, 511-518, 515, 516. globator Carter non Ehrenberg, 511, 512. - globator Ehrenberg, 511, 512, 514-516. minor Cohn, 514, Index Voivox—Continued. perglobator Powers, 514-516. rousseleti West, 514-516. spermatosphaera Powers, 518, 515, 516. tertius Meyer, 494, 512-516. weismannia Powers, 511, 512, 515, 516. Vorticella, 405. Vultur, 49, Ww Wendlandia Bartl, 257. chinensis Merr., 257, 258. paniculata DC., 258. | paniculata (Roxb.) DC., 258. paniculatae, 257. tinctoria DC., 257, 258. uvariifolia, 258. WITT, J. C., The analysis of Portland cement raw mixture, 107. Wood-preservative A, 375, 377. WORCESTER, DEAN C., A nesting place of Micropus subfureatus in Mindoro, 533. ’ x Xylopia longifolia A. DC., 126. polycarpa Oliv., 126. Xylosma Forster f., 247, 543. congestum (Lour.), 247, integrifolium Clos., 548. japonicum A. Gray, 247, nelsonii Merr., 543, racemosum Migq., 247, 248. x Yaeal, 348. Zea mays Linn., 486, 490. Zingiberaceae, 230. OC