‘ , ‘ st ! ‘ By nih , , ae) ver , Pn ee Ca fe et at a ae t are ee ft woke he ber iwe ad ai weve b i bea Hae we sree oa BN ES ee, ee ee ee ee ee kb WO OF Oe ae ee Ie ee ee ALT HE HOI TD Nhe mye te the dg WE end ME: pA Cok eee Bed HE bate oR ta at oe Cer ke Ce oe ere sri ryt war ier weer ie wot ee Mn ee ere ee ee Ce a Pere) re eee ee Se eee i eee ee ie ee ee rn ee ete rarer Newt ae vie ire err Brit ty ety A er ek eee ie Sr Cr er Te oO Oe ewer 6 er Moet ory Aan ias een le erik ie 9 Ce Le RARE Ae Ek OQ Ok Rh a Gee ae MPL ge ew Bik hrbesi) otee: Berg irh va meas te ce ewe eae ee eee woe Noe ene ere ie eC eee ene * HER ee we ge Op ren a) reer. See eke ie ee ee eee ea see eae bese tg Re get ME a ARs OL oln Oana, Hs Hye Anh errr rere mre era ee Pe ee ere Me Wer ee er We ee eh Pe ee nie hi rere ee ern Si 0 ee earner Ser tir te ee ee ee oe ee ae re) Pee Wet ee eee oe SC ee ee ee eee ee eC | Aer eee Cre rs oe ee pees Parent eee er ree ae ie Mer are Tee oe re Te ee ee ee edd eee sonny Briere tein ir | eee NOL be Ae We Cee EO OO Od, AM. © Gb here 44. Oc AUR MELO eh we prereset er ict) See ee re arate ‘ we Seen ar freee tere ene Se eer eee Wee re CEC end Ce ree a 144 et ee on + , Loh bok mee age awe ee ms Cree eer AEE Ag Reka ep nthe ee Pe Meee Sn eT ere BC oreo tae de a earn arate evr iin erm ken mim CW te here Se Rr ee Pe i re err ee oy Maren ty teste ee Rew t Fak Mir ideks Weel rarer me eC Me SC me ee Ce re eed ee nee te ty teeth nw @UNetLee NUMA. We ws Me kb MB yah Mele De A. wmatew a) ra i ir ah a a re cern ir ck he ik che ee eee wee ee Ce ee CTE OPO e ee ea ee er a ee eC Cas owe Piha Preys Serer: wire es he te ee eee ee en ete Pere ae ree te See eC es eee ee a few een oh SNA ter ies ee 8 gy ih GB M0 Ade. OL ei yo ow Ts be Cee SC ie See ere eee ee cee CeCe oe eee Ne A A te Reig eg tg erm ty meet Ce ee ee ee ee a sees Ce. ee eee ee ee yee ees ree Pee een wane Vn At owe Pt Ee nn P Memte tet . pee oA Eas gif AO sey tanteine de gk Menesipt MA HM Aw, Farad ot mtn teen (earn ere: fe Teeter ieee Metre Tne Sc ae ne acer re Were her) Bet ae mM he tM eo eT eer Mr Ce eRe ec wre TO Mrer RCM Ct eke el alas THA en eee ee er Sone he eo ater aku 40,4 ANOS large ik hy dee Buk’ NOW A aed eres asia tats Bee Mee et tree Cee ns ee te Oper win aint Marre Tome art hy er a rere Cr wee oe OO ee ee Femi GA ny MMe at avags . ee ae rete metbi tt Fw hee ee toon bot 4 sae re eC ec ee ee) eee rrr reenter a - oa How bee Lae WE RR a a ue me eit Pet bathe ot ” ‘ vee Ppa ge See er Cn Mee Cees poreey Carrera rer in ey ean a CEE EH Fe Rhee ey Ca eee) rere eee ob LAE Sie MAC wR a ke Qui heese, ALAS 4 bee peed TANT OR ae ee ey nee ee ce ee Coen ee a CE EE Be Or abe yt eyes re meen ete eee eit bee ++ rire remeewe oC nee, Wir irs Meee ete LS tre ea teow Pes S SOe oe ee ee vi * are Vem de ara hohe my re ere eee ae emerge reine fete k te ae He ene eo oe Fk ele et rk mie eee We ork & te Msenet ahome phe ae me ee ee Yt he ages WR Amato tee tem ap ati de (ener x tien) pepe ween a kot ie we Se Oe eo Cee or it fier qete oe ees ee ted betel Yet ge > Meds Aeyeorer NEE Re Meme mn Be te ape et ah ube at ae 2 EEE Wee wt ited PN tte tag he Cerny aie are SU yo er ‘ Sonn ee SS eb ete eons ” Lee ey ee ep ee geen rere eee mer eee econ Pee ere ee rc Wr oe a) SS ehh ted em ST ie oir yh bytes An Dem ees DUPE MERE ee ee Rode SN Rm eee he PE Mike tote eer Cnr Macon tet Yet A le ny teow Pagan ree ve eM aor hee RR ATE thet ete Pevrer Manni irr: nce Se i oo veer ei a 2 SNM aN IN A aT ante taety MIM Gee gta tbe tery Sr Site Tie Pett en ee ee ash te Saad eas SP ey ey th Motes & Bertheibee! Hee 4. oh Vite betig eae mem gr eer git ets St eb. bed” vo ee tothe tetewe eet tele a te S Sere Roce ikelh Br itaie le te tetagrova fle tha fee Rye be Lp St eae e Meme ttersigrt Wie Pea hyge ty Boite bette Be foley serait be eg Patino atria eee Ort ie eee ere Seve ee rr ee A oaeagte ee ote bets FO Sete te de he te te eros raed: FAs es he eS tes Vv one Caras one Pe eee mat em Sy eos sete 8 ergrey irk) ee ere ene seme Sd 4g eet Sth elBry % aoe # vy atsteee Wats dees VAR FRB Was D3 ee) osaay a wee VHB | bd ee ph deat It ae es a Fas a inh Them Peery ee rn en ae a . aft Re wee Od sheet atest sabe tgler ie ab hin we Oe ai Dae aw Odd eae ede Rb 38 Becks dg ded aie & Pa ee ed sare hain Fo Ryd ase ns rte ’ error etme t | He ae ee anh den G0 al weed Foot bet ia eae deed ee Lr een ee ee ee ae daw We me eet & Dodie dO pbs 49 Mtns AG Leyes o deat ; ee ee weed ate ee oie Bab abe ae mean tnt ary aes * we er ad mie ge ee Pierre ee ec Ce ed i te oe Brera eee a ort etd Pareene trae iar rot td Sale Scare ee Ad wd ween tay Gata tpl ee ate Biro bern i ways ‘ we - re) oy eat wea vein Wee ne Aeurrrcers werrr writ incr Tie ers . a Ase BAAR a? . ¥ ee ee ee coc Wieoarc rere retiree liar ir ne ere re ie ae ar a Vague eenwrrrerarer es nv tr me rieaers Caen oie lie ey fb soc sbaee ol otal wh AL oe ei e ey Piet en Wom ae memati Me a Tene) ee ema et ott ae tek eg bates wet ae STERN Ms OETA ee Aah AOR A Havas Laem res ee BB) Be hen i tn aie ar Gie ae Rene eee Pereror mr mrer rit matt er te ad i Meth CR ae erie ee ne ee) ro ee Ae hee Oe Tees eine ges ee wre eee wee Ane t pera we eae ry Sees cL teaches: aha ea de Bae Prone semen ttre or arr ee wn 2 Le jeoee x Perera SNe Cnet en fe ee hi St pA haw ay felipe ede Preston ron ae cee eomrae oe ent @ whee bea arom em ak ek ek ea Per ee ne Peer Re Se a ee eee be Bb Da A dae anne ie! Bd etAle nt at aca cect Perr Pe pd *e or Pernt ret So ee Pee Se Pe ee a aoe ODD) Le pene eee se mens roe rete Ly ee Rarer tener ewer een pert yea Gee cea a TA EAs guna ado ataetieal aimed acuri sad @vihlr amine @ uti im @. hbo Oo ari Pera ee ee oe Se Anbrere wren y rie 0 ere ert eto ee “4 Seer Bak aa anon oe mead abn at aba at) Ree eae a ire 2 ts Nee ee ar Sd ot hh moras ane ae trie b yivede Ra bleamGecdd whsye Sie yi Qarelzre mat Lod aitante Gas ete dt Gate Fine Aas Gaara tea ae hs Sts Latin tt telethon : x aid ND ts aia aA Pre see rh | POULT n et LOM tC ik CCB hal ae aaah We ae ree PYRE RRC n i ie ari | Rarer er fy bt fr) Ere IL’ Hae ent prnbaeiamrnenniien LYE ete, ER ee ee ge LAV PEO a8 8 Og wae meas ds arvetecowile menssshgees road oh wh RG Us 1 prs Oa Ser Se Or en ee 1a art Gre eye Wehr ar gere ws ‘ ka eb nal be eae Bera Perera eet foes Bateres trace we bed 24 Pay ey i ar ee eS sada fer erer Steen ee err wert ht ke sad Vea eek aren rice ee ieee Se ee eC Cerri wr i rk neni s ‘i sev bet ew ete tae ae Ss a2 ch Ay di oh cee Bat we seca we gtr ee Saree erin PNT ar ee car ret ie ates he . oF : + aa ded me te ee ee tes ten ie Ek aed RH tt aed eeAhl ahi pI IE AS oe awed 48 he Pome 0 9 fe ie Perce eer ea ore aa er ceak eae S08 GT Gat taal Uh eee LU SST MA GU nL ame oar og ee ME A A ot 4 awe e ae) Oyen erin ir eo Me eels Ue ERE. A Oe ig WAP LL Ha he Tent Wes A ft Ba hy yd fe SNS og pete ee re ee Awe eee a As bee boy me Aas Fad Ut erted or uw creda nya g Medan nee Ne eC Ferd eras cg BAe Aide Ald, ulep ame Hr waa ed a Be hc BEN Oe ne ote Oren ror Pinar rae we ee ee ee ey ee é ‘bes oy Ay . wt * are Va wad aah oe eu eee 6 oh ee ipierlete- te Lae wpe ty a ere rer eee Oe ere eR a) PPI reo) eet wma aed | uh weg ede dhe My! bate ah ge Gelahe WAM Wb) Hy she tis Utag She @ eet eae § Sor Pee eOn Ay rier OL Bee a me Re te ae asta a de Wor lt Gas 6 thet Pe woth a Onno eran LaG sah EER Pi A alee tor Whi as aesds ema or ops pee dono say wed 2b OW: aH Tne ¥8/ 8 Oy) ‘ ee a pagreraaaeeronrter inet ar Ment Wel Teer We COS Ded ae teed EES SA Dla bali car arent ane a oat Aight eee eee ala bam 0's C050D seu ths a 5b AG Wife. presses ees U6 08h eth Bs ihe $20) Os He Olde ESOT OES He OR ae : a Pe ene nar here or tewene eee Mrer ny tee Nor oe kA leer Karke: P aM ate a tank hae fee & Peery mm ec ee ne ee ee Ue Padi Seriya sealer Ad iret ini te Se a tae at aati athe wiety odie. gal hie Rees yy Pree tn anew vam a ‘‘ i ith Gh tah i arta © ve: aici Gwe ei eee sain eM tar less oe date Ow alg etnew aa bene® Nome i Lr oe ean ae PROCEEDINGS OF THE GENERAL MEETINGS FOR SCIENTIFIC BUSINESS OF THE ZOOLOGICAL SOCIETY OF LONDON. 1922, pp. 1-481, WITH 21 PLATES AND 214 Trxr-FrictREs. pei ASONTARSS of a ee SS LIBRARIES PRINTED FOR THE SOCIETY, SOLD AT ITS HOUSE IN REGENT’S PARK. LONDON: MESSRS. LONGMANS, GREEN, AND CO. PATERNOSTER ROW 1 Meyish Je OF THE COUNCIL Ohya AND OFFICERS ' OF THE ZOOLOGICAL SOCIETY OF LONDON. 1922. Patron. His Masusry Tue Kine. COUNCIL. His Grace Tan Duk or Beprorp, K.G., F.R.S., Presedent. Tar Hon. Ceci Barine, M.A. | Lr.-Con. 8. Monerron Copr- MAN, M.D., F.R.S. CHARLES DRUMMOND, Treasurer. AuFRED Ezra, Esq., O.B.E., Vice-President. HiucH SS. GLADSTONE, M.A., F.R.S.E. Tne Rr. Hon. 'Tun Viscount Grey, K.G., B.C. Sim Sipnry F, Harmer, K.B.E., WigsNe, IDSs, Ilatsses Ge President. Masor Tur Lorp ALASTAIR 2L0BERY [nnES-KeEr, D.S.O. Pror. Ernest W. MacBripn, IDisies,, IbAb Day Wotton, Wace President. Ksq., Ksq., Cou. Sir A. Henry McManon, G.C.M.G., K.C.1.E., Vice- President. E.G. B. Mreapr-Watpo, Hsq. P. Guaumers MircHery, Esq., GRIER, Wiles IDES, Jbl, 1D). F.R.S., Secretary. Tue Hari or Onstow, O.B.E. Masor Artzverr Pam, O.B.E. Lue Lorp QUEENBOROUGH. THe MarquEss oF SLIGO, F.S.A., Vice-President. Pror. G. Euiior Suire, M.A., MED baie Ss. Ricwarp 8. Taytor, Esq. Anvuony H. WINGFIELD, HsqQ. A. Smrra Woopwarb, Esq., LL.D., F.R.S:, Vice-Pre- sident. PRINCIPAL OFFICERS. P. Caatmers Mircueny, C.B.M., M.A., D.Sc., LL.D., F.B.S., Secretary. R. I. Pococr, F.R.S., F.L8., Curator of Mammals and Resident Superintendent of the Gardens. D. Sers-Surru, Curator of Birds and Inspector of Works. Epwarb G. BouLuncer, Curator of Reptiles. Miss L, HK. Curusman, F.E.S., Curator of Insects. C. F. Sonnrac, M.D., Anatomist. N.S. Lucas, M.B., Ch.B., Pathologist. R. J. Orrterp, M.A., Hon. Parasitologist. F, Martin Duncan, F.R.M.S., Librarian. F. W. Bonn, Accountant. W. H. Conn, Chief Clerk. LIST OF CONTENTS. 1922, pp. 1-481. EXHIBITIONS AND NOTICKS. The Srcrerary. Report on Additions to the Society’s Menagerie during the months of November and Mecem berimlODil ppaknAn ep Ai wanes 6 Lie aes cake Mr. A. H. Evans. Exhibition of Cuckoos’ eggs taken near Wanmilorialcerye pe. 3y8 . Sneha ee Nae hk Oy oe a eaie Lord Ciirrorp or Cuupieien, F.Z.S. Exhibition of photo- graphs of Wotothervum mvitchellt 2.2.0.6. .00.cccceeesetee es Miss L. E. Cuznsmay, F.E.S. On the position and function of the Siphon in the Amphibious Molluse, Ampullaria BOM UORTUAS ps Serciascrcu ch otha seeuraa eee at hia! ater eh tid: Mr. G. C. Rogson, F.Z.8. Exhibition of Models demon- strating the respiratory mechanism of Ampullaria QEIIE ONMUUSI: as SUAVE A thi ke he Saha doh ite hath Ae Sahin tas bees The Sucrerary. Report on Additions to the Society’s Menagerie during the month of January, 1922 ...... The Secrerary. Exhibition of a photograph of the Society’s Gardens taken from an aeroplane ............ Miss L. KE. Cuzrsman, F.E.S. Exhibition of living speci- mens of the Amphibious Molluse, Aimpeullaria vermi- “OUTS, BIOS as SORE MATE SOLO a os See aE ee Mr. G. C. Rosson, F.Z.8. On the respiratory mechanism OLE (lave) Ja Uc OU VENICE en bake Aaah Or hares pedal aah seer Cees cee Mr. F. Marvin Duncan, F.R.M.S., F.Z.S. Exhibition of cinematograph films taken in the Society’s Gardens.. 202 1V Miss L. KE. Curzsman, F.E.S. On the habits, in captivity, of the Fresh-water Crab, Cardisoma armatum ......... Mr. C. Tare Rucan, M.A., F.R.S. Exhibition of living specimens of the Indian Tortoise-Beetle, Aspido- WO NO SEHOCR AG FONENS © soc shnooens@anoegnads Wedeednaonscoads The Secretary. Report on the Additions to the Society’s Menagerie during the month of February, 1922 ...... Mr. C. Tare Reean, M.A., F.R.S. Exhibition of a series of lantern-slides illustrating specimens of various blimdsiresh=waiterpMishesy)..0 eken.-cecere see ert aseeeeee Mr. R. H. Burne, M.A., F.Z.S. Exhibition of specimens demonstrating the recessus orbitalis in Flat Fishes... Dr. L. Hocpen, M.A., F.Z.S. Exhibition of, and remarks upon, some small examples of metamorphosed Mexican Salamanders (Amblystonva tigrinwm) .......60esece eee e es The Secrerary. Report on Additions to the Society’s Menagerie during the month of March, 1922 ......... The Secrerary. Exhibition of some dressed skins of a Tree-Hyrax (Procavia valida) from Tanganyika Reh eI 01 Bea toe GP RIO a RPR MAUR HUbcy AaaNododnddencic Prot. P. T. Fuynn. Exhibition of, and remarks upon, a cast of the skull of a Squalodont Whale ...................4. The Secretary. An account of Mr. Loveridge’s experiences while watching “ Lions at their Kill” eee ter ee cee wee ees Page 203 203 479 479 A79 479 480 481 481 bo ite 12. PAPERS. . The Fauna of East Africa and its Future. By C. W. ROBT EW. (OVI G OHM Zins. biarhs,srascelectne nae atete a aitretes . Onthe Temporal Arches of the Reptilia. By R. Broom, D.Se., F.R.S., C.M.Z.8. (Text-figures 1-4.) ......... . Animal Communities in the Southern North Sea. By Tab; [sieitenyiw, Wye edailk, (uesxteabletingey Wy Goncoosacuocobe . Evolution within the Genus Dendronephthya (Spon- godes) (Aleyonaria), with descriptions of a number of Species. By W. Rat Suerrirrs, M.A., D.Sc., F.L.S. (PlatesMcS il) Nexto ures pl 30))i (ae seenuesseses. ee . The Life-history of the Water-Beetle, Pelobius tardus Herbst. By Frank Bartrour- Browne, M.A., IDS eehoel Dayle Ags loan (Leena) ta Le LIL) Wade ere ganwetann sous . On the Vagus and Sympathetic Nerves of the Kdentata. By Cuarves F. Sonnrac, M.D., F.Z.S. (Text- JBSUORSYS ALEETS i ca La NP etna CE an cs . Contributions to the Morphology, Classification, and Zoogeography of Indian Oligocheta. By J. SrspHen- son, M.B., D.Sc., F.Z.8. (Plate I.; Text-figures 1-19.) . On the Vagus and Sympathetic Nerves of Hyrax capensis. By CuHartes F. Sonnrac, M.D., F.Z.S. (Hest eM MBO SOS) ea a anise tenes) « Maps tad aeatn ey . The Cichlid Fishes of Lake Victovia. By C. Tare Regan, M.A., F.R.S., F.Z.S. (Plates I.-IV.; Text- ICSU Sl lho) ane APE PORE aR a one aici ana naran re . Direct Development in a Dromiid Crab. By SrepHEen K. Monveomery, B.A., B.Sc. (Text-figures 1-3.) ... Notes on the Life-history of Cweulus canorus, with exhibition of Eggs. By A. H. Evans, F.Z.8. ......... A Revision of the Nematode Parasites of Elephants, with a description of four new species. By M. Kuauit, M.D, D.P.H., D.T.M. & H. (Text- FICHUIMOS Mel lene aise a eae es UNUht nid sulliitCid hte Page 79 wy) 109 149 157 193 197 205 13. 14, 15. 16. Wie 18. AOR bo bo 23. V1 Report on the Deaths which occurred in the Society’s Gardens during 1921. By N.S. Lucas, M.B., F.Z.8. Some Spiders from South Annam. By H. R. Hoee, Wh Nog Ate ) UNepaserthernnes dkI\0)5) Saassgecstotoanss0853 New Reptiles from Tanganyika Territory. By Arruur MovERTDGE, ©. MNZiSe (ee aeeey wen. od ee uence eee Notes on the Zebras and some Antelopes of Angola. By Gineert Buarse, ¥.Z.8. (Plates 1—VIIL.) ...... Notes on the Respiratory Mechanism of the Ampulla- mde, 9 ByiGuy C. ROBSON, MAC REA /Z2>) anne ee eee Monkeys and the Fear of Snakes. By C. P. Cuaumrrs Whinerenoin. [DyStese DIL ED ES Walesiee, C)8)J91, ccoccosancoosse Mendelian Experiments on Fowls. II. Production of Dominant Pile Colour. By J. T. Cunnincuam, MA B28; \(Blates 1-11.) oe. ayence en momar eres . Observations on the Land Crab, Cardisoma armatum, with especial regard to the Sense Organs. By inser TORO (Oinaimee IMG pes gyoeeanocosoacaccccosesacaa04 3 . On the External Characters of some Hystricomorph Rodents. By R. ft. Pococw, F.B.S., H:Z.S. (Wext- ITV UU Gs lca) Eee Xo) Mae Re eR ee ARE PUBL AR PR Dis Noanondacdace . On the Anatomy of the Drill (Zandrillus lewcopheus). By Cuartes F. Sonntac, M.D., F.Z.S. (Text- foes, OPA rss easiest MeN opt oo aac a eee ee On the Persistence of the Mesopterygoid in certain Reptilian Skulls. By R. Broom, F.R.S., C.M.ZS. (Text-figures 5-7.) Cee eee wer eee ew neers reece ne sere res ene eseree . A New Species of Gisophagostomum (Wsophagostomwum wert, sp. n.) from a Rodent (Xerws sefosus). By R. J. Ortuepp, M.A. (Text-figures 1-6.) eer eee ne ee . Note on the Bat-Parasite, Cyclopodia greeffi Karsch, and on a new Species of Hymenopterous (Chalcid) Parasite bred from it. By F. W. Unica, Hueu Scott, M.A., D.Se., and J. Warsrston, D.Sce., F.Z.S. Wextetioumes 1). i860). PON Oe Ae EA ni a Page 281 285 313 D7 341 B47 349 361 365 461 471 ALPHABETICAL LIST CONTRIBUTORS, With References to the several Articles contributed by each. (1922, pp. 1-481.) BLAINE, Gineerr, F.Z8. oe Notes on the Zebras and some Antelopes of Angola. (GR reS ils VAIPELS iret et ec ES eee tL) ENE aE Ce 317 BurevaD, H., Dr. Phil. Animal Communities in the Southern North Sea. (GV ce IST NY etsy cro. rera wr accls snpacse.c od maya Mea eee eae e ose rte 27 Broom, R., D.Sc., F.R.S., C.M.Z.S8. On the Temporal Arches of the Reptilia. (Text- Ue A rate taal Gacsta SU ter ene pare tame we bh 17 On the Persistence of the Mesopterygoid in certain Reptilian Skulls. (Text-figures 5-7.)..................00008. ABD Browne, FRANK Batrour-, M.A., F.R.S.E., F.Z.8. The Life-history of the Water-Beetle, Pelobius tardus Hest Ueeubese ll WI co 5 Sota quan gy seinen. arieiactate salen eee 79 Burne, R. H., M.A., F.Z.S. Exhibition of specimens demonstrating the recessus OUD Mal I Mend NVSINES Gry eda rinsuaunoncnnnn damon eemonasa a aseet 479 vill Page CuereEsman, Miss L. E., F.E.S. On the position and function of the Siphon in the Amphibious Molluse, Ampullaria vermiformis .........+4. 202 Exhibition of living specimens of the Amphibious Mollusc, Ampullaria vermiforniis.............00ceesnene sas ore 203 On the habits, in captivity, of the Fresh-water Crab, OLA RUOSOTOCMOTALOLLIGHIO: spticso aa shoe Roo npEpaoac OA sbobbmenasAgn nboooC 203 Observations on the Land Crab, Cardisoma armatum, with especial regard to the Sense Organs .................. 361 CrupieIeH, The Lord Ciirrorp oF, F.Z.8. Exhibition of photographs of Vototheriwm mitchelli ... 202 CunnineHamM, J. T., M.A., F.Z.8. Mendelian Experiments on Fowls. III. Production of Dominant PileiColour:) (Blabes) 1h) ete ee seaadee 349 Duncan, F. Martin, F.R.M.S., F.Z.S. Exhibition of Cinematograph Films taken in the Socretys Gardens | i..6 5.0 ae eee eee eee 203 Evans, A. H., F.Z.S. Notes on the Life-history of Cweulus canorus, with exiiloitonvotwhiogs:-.... cin tae teak. .a6b ke eee eee ee eae ach SOM Exhibition of Cuckoos’ eges taken near Cambridge ... 202 Bova. Prot Pet. Exhibition of, and remarks upon, a cast of the skull Olansqualodonth Wihaleseiser cs. coss sce. 4s enna eee eee eeeee Gee Leolt Hostey, C. W., C.M.G., C.M.Z.S. The Fauna of Wast Africa and its Future ............... i Hoesen, Dr. L., M.A., F.Z.8. Exhibition of, and remarks upon, some small examples of metamorphosed Mexican Salamanders ( dinblystoma IGE UO CHT) SUSUR ep SEN eR PEN a RRO an dee DMEM SRG 5 479 Hoe, H. R., M.A., F.Z.8. Some Spiders from South Annam. (Lext-figures 1-10.) Jeet Niitivg, UG, WED Da ebiS ins DEOMI Leaccg dal, A Revision of the Nematode Parasites of Elephants, with a description of four new species. (Text- HERTNE SUES abana ccc c tence ee con mDM ERE na oc pon SU MOeoer LoveripGcE, Arruur, C.M.Z.S. New Reptiles from Tanganyika Territory ............... Lucas, N.8., M.B., F.Z.S. (Pathologist to the Society). Report on the Deaths which occurred in the Society’s Gardens durrimon lO 2 1.) iacccmam cep mcactyaae ites? as sacle Hone salsa Mrroueny, Dr. P. Cuaumers, C.B.H., F.R.S8. (Secretary to the Society). Report on the Additions to the Society’s Menagerie during the months of November and December, 1921 ... Report on the Additions to the Society’s Menagerie dlimime thepmonthvom January, LO Aeenasees essence nent ser Exhibition of a photograph of the Society’s Gardens falkcermatrccrmmelmae Op larlel ee eee ner nee eee. ae aeee = Monkeys amd thie Mearof Snakes senses ee serere : Report on the Additions to the Society’s Menagerie during the month of February, 1922) .....5......5...cee00 +0 Report on the Additions to the Society’s Menagerie dunimesthemmonth of Maxehy 1922) ies ceeseaeennene cca Exhibition of some dressed skins of a Tree-Hyrax (Procavia valida) from Tanganyika Verritory............... An account of Mr. Loveridge’s experiences while Watchmaceabionsrat ther Keil (Mh eae ian en lt cculye en: MontcoMErRY, SrtepHEN K., B.A., B.Sc. Direct Development in a Dromiid Crab. (Text- BAUCUS LON) ug. cect tcncd oeyatetn ance eet gangaana rn va sla Proc. Zoo. Soc.—1922. b Page 205 315 281 480 481 x Ortierp, R. J., M.A. (Hon. Pathologist to the Society). A New Species of Gisophagostomum (Hsophagostomum xeri, sp. n.) from a Rodent (Xerus setosus). (Text- \ Ao uMes L265) 0 eco 5 5). Hees eee kt tele Sates oe See Ee ee Pocock, R. I., F.R.S., F.Z.S. (Curator of Mammals). On the External Characters of some Hystricomorph Rodents... (lext-teures) Ie28.) sce." ...cchue waeseeee ee ane Reean, C. Tarn, M.A., F.R.S., F.Z.S. The Cichlid Fishes of Lake Victoria. (Plates I.-IV.; Mextstioures WA) ee cites «ce cil oae eee ieee eRe eee Exhibition of living specimens of the Indian Tortoise- Beetle, Aspidomorpha sanctoe-Cructs ..........0..ecersseee econ Exhibition of a series of lantern-slides illustrating specimens of various blind fresh-water Fishes ............ Rogson, G. C., M.A., F.Z.8. Exhibition of models demonstrating the respiratory mechanism of Ampullaria vermiformis ........224- v.10. On the respiratory mechanism of the Ainpullaride ... Notes on the Respiratory Mechanism of the rtp willl a eas". 8 i'. arinvaigaaln ounce ate semen cree Re ee aR eee Scorr, Hucu, M.A., D.Sc. See Warerston, J., D.Sc., F.Z.S. SHERRIFFS, W. Rag, M.A., D.Sc., F.L.S., Evolution within the Genus Dendronephthya (Spongodes ) (Alcyonaria), with descriptions of a number of Species. (Blates) IID. Pext—-figureswl3.0);) 22a eee eee ease Sonnrac, Cuaries F., M.D., F.Z.8. (Anatomist to the Society). On the Vagus and Sympathetic Nerves of the Bidembatae u(dlext-fieuires 10.) aleasaopeere: bea er pees On the Vagus and Sympathetic Nerves of Hyrax Capensrs, exit eUres)O—O4) sneak tener are eee eee eeaeEe ee On the Anatomy of the Drill (J/andrillus leucopheus). (Ressbafionies O22 4) octane a eee eRe ee ea Page 461 369 479 33 99 ‘ x1 Page STEPHENSON, J.. M.B., D.Sc., F.Z.S. Contributions to the Morphology, Classification, and Zoogeography of Indian Oligocheta.—IV. (Plate I.; PEG TT OMT Ss Me DE oy ea 8 cara cie ware eaaneeeramEM Meenas ook 109 Unico, F. W. See Warersron, J., D.Sc., F.Z.8. Wavrerston, J., D.Sc., F.Z.S., and others. Note on the Bat-Parasite, Cyclopodia greeffi Karsch, and on a new Species of Hymenopterous (Chalcid) Earagiver bred iron ib. \(WMext=tieure ls): 1. .9st-eees sence: ATi INDEX OF ILLUSTRATIONS, Acanthion longicauda, Figs. 20, 26, pp. 404, 414. Adrastis lashbrooki, Fig. 5, p. 294. Agalena doris, Fig. 8, p. 301. tenuis, Hig. 8, p. 301. Agama, Fig. 6, p. 457. hispida var. aculeata, Fig. 5, p. 456. Amira pileata, Figs. 65-68, pp. 267- 269. —— sameera, Figs. 69-71, pp. 270-272. Antelope, Angolan Sable, Pl. V., p- 317. S Antidorcas angolensis, Pls. VI1., VIIL.. | p. 317, —— euchore, Pls. VII., VIII, p. 317. Argyroelos micans, Fig. 2, p. 288. Atherura africana, Figs. 1, 8, 26, pp. 368, 384, 414. Oy Capromys pilorides, Figs, 2, 9, 21, 27, pp. 370, 386, 406, 416. Captorhinus agutt, Fig. 1, p. 18. Cavia aperea, Fig. 6, p. 378. —— porcellus, Figs. 4, 407, 418. Chetogaster, Figs. 2-6, pp. 118, 114. —— orientalis, Fig. 1, p. 111. Chart showing Stations worked by the 23; 28, pp. 373, ‘George Bligh,’ 6th-Sth July, 1921, | Fig. 1, p. 30. Chinchilla lanigera, Figs. 4, 7, 14, 25, pp. 378, 382, 594, 411. Proc. Zoon, Soc,—1922. Celogenys paca, Figs. 3, 5,7, 13, 23, 28, pp. 872, 375, 382, 398, 409, 418. Coendu nove-hispanie, Fig, 20, p. 404. —— prehensilis, Figs, 2, 4, 10, 11, 26, pp. 370, 378, 387, 388, 414. Conodectes favosus, Fig. 7, p. 459. Ctenodactylus gundi, Figs. 4, 6, 18, 28, pp- 9738, 378, 400, 418. Ctenomys imendocinus, Figs. 4, 6, 18, 22, pp. 878, 378, 400, 407. Cienus flavidus, Fig. 7, p. 299. Dasyprocta pp. 378, 393, 409. spa’, Bigs. 16, | Dendronephthya, Fig. 5, p. 40. -——- annectens, Pl. IIT., p. 33, Figs. 27, 28, pp. 73, 74. brevirama, Fig. 26, p.71. cervicornis, Fig. 14, p. 55. clavata, Pl, I., p. 38, Fig. 7, p. 45. collaris, Fig. 20, p. 68. coronata, Fig, 25, p. 69. — disciformis, Fig. 22, p. 65. —— ehrenbergi, Fig. 11, p. 51. jlammea, Fig. 16, p. 57. —— gigantea, Pl. I1., p. 33, Figs. 8, 9, pp. 47, 48. habereri, Fig. 23, p. 67. —— hyalina, Fig. 30, p. 77. —— hklunzingeri, Fig. 17, p. 59. { (eset longicaulis, Fig. 21, p. 64. —— microspiculata, Fig. 18, p. 60. c X1V Dendronephthya mirabilis, Fig. 15, p. 96. —— mollis, Hig. 18, p. 53. —— orientalis, Fig. 19, p. 61. pumilio, Big. 24, p. 68. —— punicea, Wig. 10, p, 50. — simplex, Fig. 29, p. 79. —— suensoni, Fig. 12, p. 52. Desmogaster, Figs. 11, 14,15, pp. 186, _ 140, 142. Diea shirleyi, Fig. 3, p. 291. Dinomys branickii, Figs. 6, 19, pp. 378, 402. Divaricate, a typical, Figs. 2, 6, pp. 56, 41. Dolichotis patagonica, Figs. 5, 7, 16, 23, 28, pp. 372, 382, 398, 409, 418. Drawida, Figs, 10, 13, 14, pp. 135, 187, 140. = enjoomniaga’,, VPN Voy Fos WO), Hchimys cajennensis, Fig. 6, p. 378. Krethizon dorsatum, Figs. 2, 6, 10,11, | 26, pp. 370, 378, 887, 388, 414. Hupelinus urichi, Fig. 1, p. 475. Huphractus vitlosus, Sigs. 3, 4, pp. 104, | 106. Hupotygaster, Figs. 12, 14, pp. 187, 140. Husthenopteron foordi, Fig. 7, p. 459. Galea littoralis, Figs. 6,17, 28, pp. 378, | 399, 409. Glomerate, a typical, Figs. 1, 6, pp. 36, | 41. Haplochromis acutirostris, Fig.7,p.181. altigenis, Pl. 1., p. 157. annectens, Vigs. 2, 190. —~ argenteus, Fig. 12, p. 187. —— dentex, Pl. 1ED., p. 157. —— dichrourus, Fig. 6, p. 179. — gracilicauda, Pl. IV., p. 157. — longirostris, Pl. LV., p. 157. — macrodon, Fig. 4, p. 176. —-- macrognathus, Pl. IIT., p. 157. 14, pp. 168, | INDEX OF ILLUSTRATIONS. Haplochromis imaculipinna, Fig. 5 p. 178. melanopus, Fig. 1, p. 165. orthostoma, Fig. 9, p. 184. pellegrini, Fig. 11, p. 186. plagiostoma, Fig. 8, p. 181. —-— sauvage, Fig. 13, p. 189. serranoides, Pl. II,, p. 157. —— teniatus, Fig. 3, p. 171. xenostoma, Fig. 10, p. 185. Hippotragus equinus, Pls. I11., IV., p. 17. niger, Pls. I¥., 1V., p. 317. variant, Pls. 1.,1V., p. 317. Hoplotilapia retrodens, Fig. 18, p. 189. Hydrocherus capybara, Figs. 7, 17, D4. 28, pp: 382, 399, 410, 418. Hyrax capensis, Figs. 6-8, pp. 150, 152, 154. Hystrix africe-australis, Figs. 1, 8, 20, 26, pp. 368, 384, 404, 414. Ichthyosaurus communis, Fig. 5, p. 22. Kerodon rupestris, Fig. 15, p. 396. Lagidium pernanum, Pig. 15, p. 396. Lagostomus trichodactylus, Figs. 7, 14, 25, pp. 382, 394, 411. Langliana klossi, Big. 1, p. 286. Leiperenia galebi, igs. 5-7, pp. 211— 213. --—— leiperi, Pigs. 1-4, pp. 207-210. Loncheres armatus, Wig. 6, p. 378. Tumbricus, Fig. 8, p. 158. Macropleurodus bicolor, Fig. 13, p. 189. Mandrillus leucopheus, Wigs. 9-22, pp- 480, 431, 433-435, 437, 458-442, 444,445, 447, 449-451. Memphisia aziza, Figs. 59-64, pp. 261— 264. memphisia, Figs, 52-58, pp. 256- 260. Mendelian Experiments on Pls We, py S49: Menemerus dabanis, Fig. 10, p. 308, —— felia, Fig. 9, ». 306, Fowls, INDEX OF ILLUSTRATIONS. Mizosairus atavus, Fig. 3, p. 22. Moniligaster, Bigs. 13, 14, pp. 187, 140. Murshidia hadia, Bigs, 15-21, pp. 224— 227. —— linstow?, Figs. 13, 14, p. Myocastor coypus, Bigs. 3, 6, 12, 21, 27, | pp. 372, 378, 390, 406, 416. 999 aa. Ocnerodrilus occidentalis, Fig. 9, p. 134. Octodon degus, Figs. 4, 6, 9, 22, 27, pp. 373, 378, 586, 407, 416. (Hsophagostomum wert, Figs. pp. 462-466. Oligocheta, Pig. 7, p. 138. Oiigochzete, ancestral terrestrial, Fig. 19, p. 144. 1-6, Palystes ledleyi, Fig. 6, p. 297. Pandercetes ochrea, Fig. 4, p. 292. Parabronema smithii, Figs. 8-12, pp. 216-219. Pardosa dranensis, Fig. 7, p. 299. Pareiasaurus sp., Fig. 1, p. 18. Pelomedusa galeata, Fie. 2, p. 20. Pelobius tardus, Pls. 1—I1., p.79. Petalomera lateralis, Figs. 1-38, pp. 195, | 196. | Placochelys placodonta, Fig. 2, p. 20. Placodus sp., Fig. 2, p. 20. Platyteniodus degeni, Rig. 14, p. 190. Flesiosaurus macrocephalus, Fig. 2, p. 20. XV Polyp, Armature of lateral point of, Fig. 4, p. 38. Pteridopharyne pp. 251-254. anisa, Figs. 47-51, Quilonia africana, Fig. 46, p. 248. apiensis, Figs. 22-28, pp. 251- 234. — brevicauda, Figs. 35-39, pp. 240- 243. ethiopica, Bigs. 40-45, pp. 244— 247. uganda, Figs, 29-34, pp. 236-289. Springbuck, Angolan, Pl. V., p. 317. Syngenodrilus, Figs. 16-18, pp. 142, 143. Tamandua tetradactyla, Figs. 1, 2, pp- 100, 102. Tatusia novemcincta, Fig. 5, p. 107. Thalattosaurus alexandre, big, 4, p. 23. Thrynomys swinderianus, Rigs. 1, 7, 19, pp. 368, 382, 402. Umbellate, a typical, Figs. 5, 6, pp. 37, ails Youngina capensis, Fig. 4, p. 23. Zebra, Hartmaun’s, Pl. VI., p. 317, iat i i y ; c 1) OS Bib, aa igs ies =) a | } : a fs 4 ~ ea 1 oy ONO nee SIE Ba pipe PND EX: 1922.—Pages 1-481. [New names in clarendon type. Systematic references in italics. (z.8.1..) indicates additions to the Society’s Menagerie. | Abra prismatica, 81. Acanthion, 371, 383, 405, 412, 413. Acanthobdella, 117. Adrastis lashbrooki, | Th n a M oS 5 | + mo» 2 iS) S iz 2 s = gb fo) ° = a5 S = - = S oO =] a . = re = oa s © © 2 WoLLog {oO UTLOVAVAH rie = D Oo = i= wa L oe 5 nD | ue ise) ee n z ree | 4 > | iS) & 2 . 4 Te | 55 | “ ©, | a s ss a = a 2 ea = “ s — | S sy i= S Mie ie ee | a = a S S | 3 Be © © | | | | Sao, lac | ® 3 3 3 2° S) 32S | d S om Sy Fs | | Ss ESS | | SS J SCS. wel co a ae ie ae a IHOIAM TYLOY, | i. sas He. ees ie | eH |) Gy 9% ‘ds wniaayy yy pee ee Peg cof et ee eee gy Pg | gn | ao tt fucteettinetrtese ads maungngetog | | | z anl snpsnyuiag snunbodngr § Gin Se Sle a ES a aaa aa as GaP 2 aug Seay 3] RCL Seog signujsot snyauhysouazy | I 2. eS eer Rot see | tre eee Jens eee ens ant snunpjaaisspa sagsh.op ; a3 eer eis ‘antl sogn.undap snung10g | ““w“afiurds snoosougsv5 | se | 06 BIIVUUN,) | 1.0 Om [eco se Le Ws] KO) I izes) Eb ey als Hy Ge ake ee i | oo eeprremTey) | ee 0 ; srusofyyl vunryduy F ! eects er £.0 | de leu. oo eyo. 2 ee ey ee ee 1.0 I Wagon ears saoboone snyprsnd snuphoouyor aie ie (Wpaone ee ZOE ett eG ee ee ee ee ge ers He tay forseeteresens can fleds “ T I wee see on 2.0 WSL sae woe wee wae tee wae WS UL. see eae wae wee eee . Ung psod UUNIPADIOULY I HT Simi ~ Tro | S| | O “wSrq) PLT | | a aes eee eee | eee Jose. see eee OO B.0 G ee ove T I wee wee z I. tae . . ee vee eee COL? tee pee [eee ee ween ers ere CEE : ee le ge mee Ot ne MR PR a es ow ee eee Sit : . F.0 it I oe ae see tise | on see one dies Ser wee eee see eve ose aoe or Ser see bee aYsie tee oe [egeegmngani the opinpetaeeg3 ‘ds DIADUD) T | oxoreca: a =) ce re 30 DR. H. BLEGVAD ON ANIMAL COMMUNITIES to the “ Venus ”-community, if anything; so it is possible that the hard sand and gravel bottom here in the strong current off Lowestoft is a transition area between these two communities ; at all events it is certainly a very poor area, exception made of the Epi-fauna (to this also must be reckoned Stronglocentrotus drobachiensis, the Hydroids, and Sabellaria alveolata, which last- named species is not found in Danish waters inside the Skaw). Stations 3-1] and 16-19 are typical ‘“‘ Venus ”-stations, with the characteristic species Venus gallina, Tellina fabula, Mactra Text-figure 1. 3 Chart showing Stations worked by the ‘ George Bligh,’ 6th-8th July, 1921. Stations 1-2. Venus (?) Community. aN 3-11, 16-19. Venws Community. ‘ 12-15. Echinocardium-filijormis Community. Station 20. Abra Community. subtruncata, Donax vittatus,and Hchinocardium cordatum. Among the ‘‘ attendant species,” ¢.e. species frequently occurring in the community in question, but which may also be found in others, IN THE SOUTHERN NORTH SEA. 31 I may particularise Vatiea alderi, Ophelia limacina, Pectinaria korent, and Hehinocyamus pusillus. In stations with no character- istic animals, I have attached importance to these species, which in conjunction with the depth, condition of bottom, and scarcity of animals, all point to the ‘‘ Venws”-community. The occurrence of Abra prismatica and Mactra elliptica in some stations is in- dicative of the “ deep Venus” community, which has been found in the deeper parts of the Danish ‘“ Venus”-community ; but as the characteristic species Hchinocardium flavescens and Spatangus purpureus, so typical of this community, are wanting, I can for the present only designate these stations as ‘“‘ Venus ”-stations in general. Stations 12-15 are characterised by such animals as Amphiura jiliformis, Turritella terebru, Auvinus flecwosus, and Hehinocardiwm cordatum. Stations 12 and 13, however, are transitional from the “ Venus ”-community, containing several of the characteristic animals of this community but mostly as small specimens, and only a few Amphiura. But as a whole these stations must be reckoned as belonging to the “ Hchinocardiwm-filiformis” com- munity. The bottom, consisting previously of hard sand or eravel, is here softer, mixed with detritus and clay. ‘This community is not represented in Dr. Petersen’s hypothetical chart in Rept, X XIT. (on this point see remarks in Rept. XXIII., 1915, p. 10, at the top) ; it was charted in Rept. XXI. as the & Hehinocardiua- Pur ritella ’-community, which 1s widely distri- buted in the Kattegat, Skagerrak, and Christiania Fjord. Without doubt it will later become apparent that this community covers big areas in the deeper parts of the North Sea. The last station, No. 20, in the neighbourhood of Graa-deep light-vessel, is a typical “Abra ”-station, with Abra alba, Nucula nitida, and fragments of Hchinocardium cordatum. The existence here of an “Abra”-community was already proved in 1912 by some few samples taken by the ‘ Michael Sars’ in the neighbour- hood of Graa-deep and Horns Reef. We have thus met with four different communities :—a ‘“J/acoma baltica”’-community along the shore in Lowestoft Harbour; a “« Venus ”-community in the greater part of the investigated area of the North Sea, with depths most frequently between 9 and 38 m. and a pure sand and gravel bottom; an Kchinocardiwm jiliformis community in depths of 38-41 m. with a dark, soft sandy bottom ; and finally, an Abra-community near the shores of Jutland. Possibly this last-named community also will be discovered near Lowestoft, as ina single sample taken with the grab in Lowestoft Harbour, in a depth of 3 m.at ebb tide, I found one adult Abra alba. Butit is beyond doubt that the greater part of the bottom fauna of the North Sea below 38 m. will turn out to belong to ‘“ Venws”-community, as indicated in Dr. Petersen’s chart. ‘his is confirmed also by the few samples taken in former years with the bottom sampler off Thyboron (Report XXI., Appendix pp. 63-66), and by the numerous samples which J . 32 ON ANIMAL COMMUNITIES IN THE SOUTHERN NORTH SEA. according to the friendly communication of Dr. Russell, have been taken with the bottom sampler this year on the Dogger Bank. As to the number of animals per square mile, it will be seen from the accompanying list (pp. 28-29) that, setting aside the Epi- Fauna, we have to do with a comparatively poor fauna. For instance, the “Venus”’-stations together give only 26°3 gr. per 2°6 sq. m. or 10-1 gr. per sq.m. But on the other hand, most of the animals are useful as being excellent fish-food. Compared with the ‘ Venus”-community in the Kattegat (Rept. XXI., Appendix, pp- 9-18), it will be seen that the apparent richness of the latter is due mainly to big echinoderms such as Hehinocardium and Ophiurids, or big lamellibranchs such as Cyprina or Aiacoma calcarea, useless as fish-food. On the other hand, we may find in the North Sea as well as in the Kattegat small areas within the “ Venuws’”-community, in- habited by large numbers of Mactra subtruncata (see, for instance, St. 8 in Rept. XXI., Appendix, p. 65, with over 2 kg. per sq. m., also among the samples taken by the ‘George Bligh’ this year on the Dogger Bank, several of which contain some hundreds of specimens per 0°2 sq. m.). It would be of the greatest interest to know something further about the distribution and abundance of this most important food species, at different times of the year and at different places in the North Sea. At the same time continued investigation of the stomach-content of the food-fishes is called for, following up the excellent work of Todd (Second Report on the Food of Fishes, North Sea Investigation Committee, Southern Area, Second Report, Part I., 1904-5). From his paper it would appear that the food of the fish in the south- western North Sea is composed mainly of “ Venuws’’-animals. But, as mentioned above, we have in the deeper central parts of the North Sea an extended area, which is still a terra incogita and still awaits charting. ny ath vies Waitt Lee AY. 5 >, Z S, Il, SSVERRUEIES, Jel DENDRONEPHTHYA CLAVATA, Kuklenthal. PZ S, Il, SVERIRWEES, Pl, Ul, DENDRONEPHTHYA GIGANTEA, Verrill. Ri Zo So MI, SVR WMS, Il. IO! DENDRONEPHTHYA ANNECTENS, sp. n. EVOLUTION WITHIN THE GENUS DENDRONEPHTHYA. 33 4, Evolution within the Genus Dendronephthya (Spongodes) (Aleyonaria), with descriptions of a number of species. By W. Rae Suerrirrs, M.A., D.Sc., F.LS., Lecturer in Zoology, University College, Southampton *. [Received August 10, 1921: Read November 8, 1921.1] (Plates I.-III.f ; Text-figures 1-30.) Part [.—GeENERAL. § 1. All who have worked at Aleyonarians will agree as to the difficulties presented by the genus Dendronephthya or Spongodes. For here we have to deal with a multitude of species within a relatively narrow range. Thus Kiikenthal, in his * Versuch emer Revision der Alcyonarien: II. Die Familie der Neph- thyiden, 2. Teil’ 1905, deals with no fewer than eighty-seven species ; and Henderson, in the ‘ Aleyonarians of the Indian Ocean,’ Part II. 1909, with another series of sixty different from the former, And yet in both cases the species are described in minute detail. The continual experience in investigating a collection of repre- sentatives of this genus is that, in spite of an initial determi- nation to refrain from adding to the already large number of described species, one is forced to do so. And there is no denying that each of these new forms has a distinct individuality. This experience inevitably raises a number of ztiological questions which it may be useful to state, although they cannot be more than partially answered. Some of them at least could be replied to by an investigator in a good locality, having at his command a large number of specimens of any given form. § 2. The outstanding phenomena which present themselves are, apart from the multitudinous species, the following :— A. That the specific distinctions are all of a relatively trivial sort, such as mode of branching, grouping of polyps, length of polyp stalk, strength of supporting bundle, presence or absence of a definite ‘ crown,” the number and the arrangement of the spicules in the anthocodial points, the distinction between the cortex of the polyparium and that of the “sterile stalk,” and the nature of the spiculation of the canal-walls. B. That the species differ one from the other to a large extent in the congeries or collocation of such characters as we have mentioned ; that is to say, two species with similar anthocodial armature may differ in the mode of branching, and vice versa. C. That within the limits of a colony there is, in most cases, * Communicated by Prof. J. ARTHUR THomson, M.A., LL.D. + For explanation of the Plates see p. 77. Proc. Zoou. Soc,— 1922, No. ITI. 3 34 DR. W. RAH SHERRIFFS ON EVOLUTION relatively little variability of architecture, though there are, of course, quantitative differences in size of spicules, strength of supporting bundle, and the like, which are merely growth characters. I must emphasize the saving clause in: most cases, in view of the occurrence of a number of very variable species, such as D. gigantea, where there is great variability of antho- codial armature in one and the same colony. In the second part of this paper particular attention is paid to the interesting form referred to; and it does not stand alone. § 3. The multiplicity of form detail within a narrow range admits of various interpretations :— I.(a). It may be urged that the observed differences are individual, not specific, characters. If the observed differences are purely individual, they may be either extrinsic modifications or of the nature of intrinsic variations. And before considering these alternatives separately, I may note the general answer that in many cases a considerable number of specimens from the same or different localities agree together. ‘Thus Kikenthal speaks of twenty specimens of D. savignyi, twelve of D. gigantea, ten of D. rubra, seven of D. pumilio; while I have myself got twelve of D. ehrenbergi, ten of D. gigantea, ten of D. clavata, eight of D. habereri, and the same number of D. hyalina. Therefore the general suggestion that observed differences are purely individual does not apply in every case. At the same time, the unsatisfactoriness of basing a new species on a single specimen when the specific characters are of such a tenuous description is plain. (6) It may be said that the observed’ differences are of the nature of exogenous modifications correlated with environmental peculiarities. Against a too ready acceptance of this easy solution is the occurrence of the same form in widely separated localities. For example, D. robusta is reported from Ternate, Borneo, Zanzibar, and Madagascar; D. florida from Hong Kong, the Philippines, and Port Jackson (Australia). It may also be pointed out that certain factors in the immediate environment in different parts of a large colony must show considerable diversity, cf. shelter; and yet, as already pointed out, there is rarely diversity in a colony except in colour and the like. (c) It may be, however, that the observed differences are expressions of individual intrinsic variability, not correlated with any peculiarities of environment. There is no way of testing this interpretation until numerous similar specimens from the same locality and of the same age and size are examined by some investigator with abundance of living material at his disposal. II. It may be that the multiplicity of distinct forms within a narrow range is due to cross-fertilization between allied species. a | WITHIN THE GENUS DENDRONEPHTHYA. 3E Tt is quite conceivable that species which have attained individu- ality and fixity in the course of isolation and inbreeding may be brought into contiguity by subsequent spreading or removal of barriers. In such a ease, analogy points to the likelihood of numerous new patterns arising by permutations and combinations of the previously segregated species. This hypothesis is certainly suggested by the fact that two 0 species may agree in anthocodial armature and yet differ in mode of branching, or may agree in having a foliaceous collar and yet differ in the nature of the supporting bundle. In short, in the distinctions between species there is a distinct suggestion of the shuffling of unit characters. III. There seems to be a third possible interpretation—namely, a mutation of species apart from any hybridizing influence. It may be that certain widespread and strongly established species such as D. gigantea, D. chrenbergi, and D. brevirama have been the stocks from which mutations have been thrown off after the fashion of Qnothera lamarckiana. An indirect argument in favour of this interpretation may perhaps be found in the possibility of discriminating similar radiations of evolution within the three great groups : Glomerates, Divaricates, and Umbellates. A move direct argument may be found in the variability of certain species such as D. gigantea, to which reference has already been made. As Kiukenthal remarks, “ Aus diesen Beschreibungen geht hervor, dass D. gigantea eine in ihrem Aufbau sehr variable Art ist.” (‘ Revision,’ p. 553.) § 4. Kukenthal has divided this difficult genus into the three main groups: (1.) Glomeratie ; (II.) Divaricatee ; (III. ) Umbellatee, giving precision to similar suggestions by previous workers such as Holm. I. The Glomerate are characterized by :— (a) the comparatively slight branching of the polyparium ; (6) the grouping of numerous bundles of polyps into roundish bunches which make the surface of the polyparium entirely irregular. There is a marked definiteness about the Glomerate division which suggests “naturalness,” and makes it easy to refer a species to the group. Text-fig. 1 is a diagrammatic illustration of what is meant by the Glomerate habit of growth. If. The Divaricate are characterized by :— (a) the profuse branching of the polyparium ; (6) the length and slenderness of the twigs ; (c) the divergent separateness of the polyp bundles ; (d) the absence of anything that can be called een of the Glomerate type or umbelg of the Umbellate type. It should be noted that a Divaricate polyparium may have a continuous contour like that of a well-pruned tree (see diagram). 2% Text-figure |. A typical Glomerate. Text-figure 2. A typical Divaricate. EVOLUTION WILHIN THE GENUS DENDRONEPHTHYA., 37 IIT, The Umbellate are characterized by :— (a) the umbel-like or sometimes corymb-like aggregates formed by the terminal twigs, the heads of the umbels being bundles of polyps; (b) the disposition of all or most of the polyp heads on the surface of the colony. Text-figure 3. A typical Umbellate. Thus the Umbellate differ from the Glomeratze, and agree with the Divaricate in showing much minor branchiug. They differ from the Divaricatze in the presence of umbels or corymbs of polyp bundles and in the entirely superficial arrangement of the polyp heads. As this continuous superficial disposition is obviously advantageous—giving all the polyps equal expo- sure for nutritive and respiratory purposes—we may regard the Umbellate as the latest expression of the evolution of the Dendronephthya polypavium. Moreover, some Umbellatie pass through a Divaricate stage. Of minor importance is the question whether or not the umbels combine into large secondary bunches rising like hillocks on the surface of the polyparium. Nor can I attach more than slight inportance to the shape of the polyparium whether flattened or quite symmetrical, whether spherical or disc-like, and so on, for these features probably depend to a large extent on local environ- mental conditions. It may be noted that a large and very handsome colony of D. annectens was, as a whole, strikingly divaricate at first sight, but a closer inspection soon showed that it was an Umbellate having the terminal twigs massed into distinct umbels. (See Plate ITT.) 38 DR. W. RAE SHERRIFFS ON EVOLUTION $5. Before proceeding to consider the specific characters selected for examination and criticism, it will be well to recognize clearly the typical architecture of a Dendronephthya polyp as Text-figure 4. Spicules or tentacle Anthacodia/ Intermediate spreuses spoicule pear) 3) } Crown (©) Accessory spicule -- ——- —- = Supporting b6uUNnd/e ages (main) spicules Armature of lateral point of Polyp. regards its external spiculation (text-fig. 4). When every por- tion is represented, the ascending order above the polyp stalk is as follows: (1) The spicules of the supporting bundle (Stutzbiindel), several of which are usually very large (often up to 3 or 4mm. in length) and often projecting beyond the polyp for a distance of 0°5-1-5mm. It is useful to distinguish in the supporting bundle the main spicules and the accessory spicules, the latter often mingling gradually with those of the polyp stalk. (2) The “crown,” consisting of a few rows, very commonly three, of spicules arranged horizontally round the base of the anthocodial part of the polyp. (3) The anthocodial armature, in the strict sense, which consists mainly of the spicules entering into the com- position of each of the eight points. Between each pair of points tiny spicules are frequently found. These I term intermediates and discount their value because they are so variable. (4) Very minute spicules on the tentacles, the nature and number of which seem to be quite unimportant. WITHIN THE GENUS DENDRONEPHTHYA. a9 § 6. If, as I think probable, the Glomerates, Divaricates, and Umbellates represent three distinct stocks or lines of evolution, the particular problem is to discover whether similar morpho- logical radiations can be discerned in each of these main divisions. A prior question is—What characters can be re- garded as having real morphological importance ? 1. Negatively, it is quite plain that little emphasis can be laid on the size of the colony, looseness or density of branching, amount of sterile stalk or colour. 2. Positively, the features which seem most indicative of relationship are in order of importance :— (1) the architecture of the anthocodial armature—e. g., whether crown and points are distinct, the number of spicules in each point, their degree of projection and their differenti- ation within the point ; (2) the differentiation of the supporting bundle ; (3) whether the polyps are marked by long stalks or are very short-stalked ; (4) characteristic features in the spiculation, such as the occur- rence of particular forms in special regions of the colony e.g., the “sterile stalk,” canal-walls. § 7. In studying the possible relationships of different species, I make the following provisional assumptions :— (a) That the presence of very numerous small spicules in the anthocodial points is more primitive than a reduced number of large spicules. (6) That the absence of a definite “‘ crown ” is more primitive than its differentiation. (c) That a slightly differentiated supporting bundle with a large number of spindles tending to form a sheath rather than a buttress is a more primitive condition than the differentiation of a bundle out of a few large spicules. And here it may be noticed that in a few species—e. g., D. clavata, the supporting bundle is so slight that the polyps may at first sight be mistaken for those of a Hunephthya, where there is no supporting bundle whatever. (d) That the more primitive forms show :— (1) less variety of spicule: i.e., are predominantly provided with spindies, the characteristic spicule, and less variety of spicule in the anthocodial point ; (2) less localization of particular forms of spicule in parti- cular areas. Thus there can be no doubt that the presence of special interlocking spicules in the ‘sterile’ stalk” is a later differentiation ; (3) the presence of few spicules in the canal-wall. Thus D. piitéeri with a multitude of eanal-wall spicules is in other respects highly specialized. Conversely, it is interesting that 40 DR. W. RAE SHERRIFFS ON EVOLUTION D. clavata with no canal-wall spicules belongs to what I would call the more primitive types. So careful a worker as Kikenthal attaches great importance to the canal-wall system of spicules, and it is obviously important to inquire what correlation there is between the primitiveness of a species and the number and nature of the canal-wall spicules. The unsatisfactory feature here is that the examination of the canal-wall spiculation has not been made uniformly by the various investigators. Minute spicules which escape attention in a fragment of canal - wall when Text-fig. 5. boiled down may be found by the other method of examining a piece intact. But one is not uniformly Ih successful with the sample examined. §8. Bearing in mind the first of the above assumptions, it is possible to ex- pand it so as to reduce all the different 7 KK Species to one or other of the follow- Laer gw, ing six grades, ranging from the most aaa primitive (1.) to the most specialized a (VI.) :— VI. Only one pair of spicules, or even but a single spicule re- placing all the others in each Semmes point above the crown ; ‘Sas VY. reduction of the point that one pair (or, it may ae one of the 4 ? ? uppermost pair) predominates ees Y N over the others, which are not aE above three pairs in total number above the crown ; UN GN IV. one pair or one spicule predo- on IN minating over 3—4 others, no OT crown ; a” IIT. about 4-6 pairs, the uppermost slightly specialized, no crown ; ‘ II. about 6-8 pairs, strong and aN uniform, no crown ; ZX ° . ° WW t 4 ‘I. about 8-12 pairs in each poin t 4S numerous and small, rising \\ from indefiniteness to definite - we ness, no crown. — ==») Ui §9. The suggestion has been made by critics of Dendroneph- ihya species that the differences between the Glomerate, Divari- cate, and Umbellate groups are not of taxonomic value, but 4 WITHIN THE GENUS DENDRONEPHTHYA, 4) depend on conditions of growth. On this view, the occurrence of a species with the same anthocodial armature in each of the three groups would not mean that a similar specialization had occurred three times on different lines of evolution; it would mean that one and the same species occurred in three different vegetative forms—Glomerate, Divaricate, and Umbellate, which depended on growth conditions. For various reasons this view is improbable:—(1) In many cases the Glomerate, Divaricate, or Umbellate mode of growth is clearly expressed in young forms 1-3 em. high. (2) When exactly the same anthocodial armature occurs in two or all of the three groups, the two or three species in question differ in other details, such as the length of the polyp stalk, the nature of the supporting bundle, and the spiculation. This does not look like the occurrence of the same species in three vegetative guises. There are Umbellate colonies, however, which seem to pass through a Divaricate stage when they are young and small, but these are in my experience fewer than forms which are Umbel- late from near the start. Text-figure 6. Glomereate UmbeHate — Divaric32te In support of the view that a colony usually shows very early whether it is going to be Glomerate, Divaricate, or Umbellate, I have given a series of diagrammatic representations of a dozen small colonies drawn .to actual size. It goes without saying that the definite architecture cannot be expressed until there are quite a number of polyps in the colony. But the limit is uncertain. The central figure of the group depicts the primitive, quite undifferentiated colony from which scon the distinctive form embodies itself. Above it to the left are Glomerates, to the right Umbellates, while below is a row of Divaricates. 10. Starting with the Glomerate, I suggest that the fol- lowing should be regarded as illustrations of primitive types :— 1. In the savignyt group :— D. argentea with irregular arrangement—most primitive of all, 49, DR. W. RAK SHERRIFFS ON EVOLUTION with ventrally and laterally nothing more than minute cylindrical corpuscles. D. fusca with indefinite double rows of about 6 pairs— on a slightly higher level, but also showing primitiveness in numerous quite small elements. D. savignyi with indefinite steep double rows reduced to 6—somewhat higher, but also showing, like the preceding, very numerous small cylindrical bodies. 2. In the hemprichi group :— D. clavata, where occur 10-12 pairs in each double row and a very weak supporting bundle. It may be noted that in this group we have a gradual reduction in the number of spicules composing each point 5-7, 5-6, 4, 3, 2 and 1, reaching a climax in forms like D. punicea, where are found a couple of slightly hockey-club like spicules occupying the whole area which in D. clavata, for instance, shows 10-12 pairs. 3. In the studeri group :— Here we have, on the whole, a more differentiated group with forms like D. mayi with 6-8 pairs in each point to be ranked as relatively primitive compared with the highly-specialized D, kollikert. § 11. So with the Divaricates, on my view the series begins with forms like D. japonica with 8 pairs in each point, none protruding—a very primitive arrangement—and ends with D. suensoni, which has but one pair. As intermediate between grade IT., which I define as having numerous pairs regularly arranged in each point (grade I. possessing very numerous irregular pairs), and grade III., with 5-6 equal pairs, we have cases like D. armata, where the lateral points go up to 9 pairs. There must also be noticed D. cerulea inclining to be primitive in minuteness but not in number of spicules, while D. Alunzingert may be considered as rising out of grade II. because one of the uppermost pairs is very prominent. It is interesting to see that in so many of Kiikenthal’s minor groups his arrangement of the species ends in one pair, so that the anthocodial armature, which I hold to be the evolutionary key, is, as used by Kiikenthal, very probably just diagnostic. Again, in the rigida group, D. microspiculata is undoubtedly primitive with 6-8 pairs, though the uppermost again is slightly larger and projecting. § 12. In the third great division—the Umbellates—grade I. is represented by D. australis with its very numerous, small polyp spicules, grade IT. perhaps by D. wmbellulifera with 6-8 pairs and the practical absence of a supporting bundle. In the dendrophyta group, D. villosa belongs to grade I. with many spicules in each double row, while grade II. in the florida group may be illustrated by D. brevirama (with 6-8) and D. florida (with 5-7). WITHIN THE GENUS DENDRONEPHTHYA. 43 D. pectinata must be regarded as the climax in the line of evolution. Again in the rubra group, grade I. is well represented by D. repens, while in the spinulosa group, grade II. is probably represented by D. spinulosa itself with 6-8 pairs very small and not projecting, even better by this than by D. flabellifera where we have 8-9, but in which I recognize what I regard as the beginning of differentiation: viz., a specialization of the highest pair. In tabular form my conclusions may be represented as follows :— I. 10. JON GRADE. GLOMERATES, DIVARICATES. UMBELLATES. c ph Ge Az) he Ul Skee Cae | eid: (| D. punicea (Stud.)*. ( -D. piitteri (Kiik.). ( D. lutea (Kik.). | D. carnea (Wr. &Stud.). | D. swensoni (Holm). || D. longicaulis (Kiuk.). VL4 D. deederleini (Ktik.). | D. orientalis (Hend.). } | D. simplex (sp. n.). : < | D. flammea (sp. u.). | | | | D. cervicornis(Wr. &Stud.). | U L| D. laxa (Wr. & Stud.). Ne v.{ ' D. eburnea (Kiik.). ; | D. coronata(Wr.&Stud.). (| D. spinifera (Holm), ( D. mollis (Holm). ‘a D. disciformis (Kiuk.). | | D. kéllikeri (Ixik.). | _ D. cirsium (Kiik.). | D. rubra (May). IV J ) D. involuta (Kiik.). } D. pumilio (Studer). a} | D. mirabilis (Hend.). D. nigrescens (Kiik.). | | | D. collaris (Wr. & Stud.). lL L |. D. sinensis (Piit.). (| D. studeri (Ridley). ( j ~ DD. dendrophyta (Stud.). lll D. gigantea (Verr.). \ D. hyalina (Kiik.). j l / D. mexicana (Kiik.). . ; D. florida (Esper). ( D. mayi (ik.). ( D. klunzingeri (Stud.). ( D. brevirama (Burch.). D. hemprichi (Klunz.). | D. ehrenbergi (Kik.). | D. umbellulifera (Kiik.). IL.< ~ D. japonica (Kuk.). <~ D.spinulosa (Gray). | j | D. microspiculata (Put.). D. habereri (Kiik.). Uy D. erinacea (Kuk.). D. annectens (sp. n.). | D. argentea (Kik.). ( D. villosa (Kiik.). 1) D. clavata (Kiik.). —— / D. australis (Kiuk.). (l _ D. repens (Kiik.). * The species printed in italics are those worked out from the Dendronephthya collection of the ‘Siboga ’ Expedition. As a convenient means of expressing the composition of the anthocodial point, I have drawn up an anthocodial formula on the following lines:—The spicules of the point are denoted by “P” if big and strong, and by ‘‘p” if small and weak; those of the crown by “Cr”, and the supporting bundle by the letters ““S.B.” preceded by a qualifying adjective, such as weak, medium, ~ very strong, etc. In this way the anthocodial grade and formula of D. annectens, for example, can be set forth briefly as : IJ.=8 p+0Cr+very weak S.B. 44. DR. W. RAE SHERRIFFS ON EVOLUTION Part I1.—DrscrIPTION OF SPECIES TAKEN BY THE ‘S1poGaA’ EXPEDITION. TABLE OF SPECIES. Genus DenpronepuTeya (Kik.). | Number | examined. | \ Locality. I. GLOMERATA. | Station 302. Stations 165, 213, 240 261, 303. 5 | Station not marked. . DIVARICAT A. | Stations 164, 258. | Station 164. ‘ | Stations AmboinaReef, 310 B, 366. ; 5 BOR ETO: | es 99, 282. | Station 91,Macaras Reef. per 5 VRe, 19R % 315. ; bs 125, East of Polloe Weh. § . UMBELLAT A. 5 3D), Cllawwennay (IRE) ga of Soe ono 10 ID, eae) (WEN) oon oss vss anavo- 10 Bs ID), joe) (SiC) soeces che sad sue 1 I A. I). ehrenbergi (Kuk.):...........| 12 5. D. suensoni (Holm) ........ ...... 2 6. D. mollis (Holm).............00-.1 4 7. D. cervicornis (Wr. & Stud.)... 2 8. D. mirabilis (Hend.) ............| 2 @), JD), iilerravone, (js We) aco vennse soo 00: 2 10. D. klunzingeri (Stud.)............ 2 11. D. microspiculata (Piitter)......) 2 12. D. orientalis (Hend.) ............) 4, Ill 13. D. collaris (Wr. & Stud.) ...... 2 14. D. longicaulis (Kik.) ..,......... 1 15. D. disciformis (Kwk.) ........... 6 UGS JD), Ineoeswer (UGH) Sone. Gencnde: 8 17. D. pumilic (Studer) ........ 4, 18. D. coronata (Wr. & Stud.)...... 3 19. D. brevirama (Burch.) 6 20. D. anmectens (sp: 1.) ... ....... il PAL, 1B), smajolas (Gos Ws)) sonsst ceo ooses 1 22. D. hyalina (Kiuk.) 8 400) 925) oo eee a 93 Station 99. 310. Stations 99, 164. » 258, 282. |, 99, 258, 315. Station not marked. Stations 164, 258. Station not marked. - not marked. Stations 257, 318, 321. J. GLOMERATAL. ‘1. Dendronephthya clavata (Kiikenthal). Previously reported from | S.W. of Japan. ) | ) Japan, Hong Kong Japan. Red Sea. Japan. Japan. Lifu, Funafuti, Kei. Andamans. Red Sea. | Philippines, Hong Kong, Amboina. | Andamans. Kei. Japan. China Sea. Japan. Japan. Torres Straits. Torres Straits, China Sea. Pescadores. (Plate I.) Described by Kiikenthal, ‘Revision der Alcyonarien,’ pp. 549-6 (1905). Draenosis :—Glomerate; with great bare gap between the collar and the polyp-bearing area ; polyps in dense groups at the ends of the terminal branches: polyp stalks very long ; support- ing bundle weak; point spicules 7-8 pairs, short and uniform ; no distinct crown; grade I.; tentacles with tiny, flat, Jagged spicules; the foliaceous collar with large curved spindles ang WITHIN THE GENUS DENDRONEPHTHYA, AD small, quaint, flattened forms; colour: general cortex and sterile stalk yellow, collar vivid red, anthocodial armature whitish, twigs and polyp stalks with red spicules. * ANTHOCODIAL GRADE AND ForMULA:— L=aP+0Cr 4+ weak S.B. Descriptive Noves :— 1. Colony as a whole. Ten specimens of a peculiar form, the smallest 15 mm. in height, the largest 17 em. ‘They are remark- able in showing, especially in the larger forms, a great bare gap between the collar and the loosely glomerate polyp-bearing area. In the largest specimen this interval extends for about 6 em. Here and elsewhere it is extremely limp, and it looks as if the polyp-bearing area had rested on the mud. In the upper part of the colony there are numerous galls due to and containing small Balanide, and the polyps are very muddy. It should be noted as a feature that the upper part of the polyparium is distinctly top- heavy; in the largest specimen it has a breadth of 10 em., whereas the stalk bearing it is only i em.across. The colony broadens again at the very foliaceous collar and measures 6 em. from side to side. Below that the sterile stalk, which is markedly flattened, measures 3°5 cm. in breadth. There are numerous stolons. 2. Branching. The three largest spe- cimens show two lateral short branches esate Glee ae arising immediately above the collar, fe but not affecting the bareness of the gap pballag alluded to. 6 ulin 3. Colouring. The twigs and polyp 7) stalks bear red spicules; the anthocodial J armature is whitish or transparent; the 4, general cortex is yellowish; the collar is 4 ‘of a very conspicuous red; the sterile Y stalk has the same colour as the general cortex. 4. Polyp stalks and their spicules. The length of the polyp stalks is very ereat—often reaching 3 mm. in the collar and about 1:5 mm. in the upper polyparium. The spicules are red and D. clavata Kiik. very thorny. 5. Polyps. Found in dense groups at the tips of the end branches. 6. Polyp spicules. The anthocodial architecture is primitive, having 7 or 8 pairs of very uniform short rod-like spindles in each point and no distinct crown. The supporting bundle is * In the anthocodial formule the point spicules are denoted by “P” if big and strong, and by “ p” if small and weak; the crown spicules by “Cr,” and the supporting bundle by “S.B.” preceded by a qualifying adjective (ef. page 43). 46 DR. W. RAE SHERRIFFS ON EVOLUTION very weakly developed and shows only occasionally a projecting needle. 7. Other spicules. Those of the sterile stalk are almost en- tirely small, star-shaped forms. In the foliaceous collar we find (1) large, curved, almost smooth spindles up to 5 mm. in length; (2) quaint, flattened forms, like biscuit-fingers in shape, not exceeding 1 mm.; (3) the small red, rough spicules of the antho- codia ; and (4) the tiny, flat, jagged ones of the tentacles. SALIENT FEATURES :—The peculiar mode of growth (which seems to be unique), the limpness of the whole colony, the length of the polyp stalks. I refer my specimens to D. clavata, although Kikenthal’s description does not mention any growth peculiarity such as has been noted here. The important points of agreement are the following :— (1) a glomerate mode of branching (Glomerate), with the polyp- bearing twigs predominating over the stem and main branches in building up the polyparium (hemprichi group); (2) a primitive anthocodial armature with 7-8 (D. clavata 10-12) pairs of very uniform, short, rod-like spindles in each point; no distinct crown ; (3) a very weakly developed supporting bundle which only occasionally shows a projecting needle ; (4) a long polyp stalk of from 1-2 mm, (about 1 mm. in D. clavata) ; (5) a well-developed foliaceous collar ; (6) an almost exclusive occurrence of small, plump stars and double stars in the cortex of the sterile stalk ; (7) a practically complete absence of spicules from the canal- walls. With agreements so numerous it seems entirely justifiable to disregard the extraordinary growth pecu- liarities. Locauiry :—Station 302. 2. Dendronephthya gigantea (Verr.). (Plate IT.) Described by Verrill, Bull. Mus. Comp. Zool. Cambridge, p. 40 (1864). DiaGnosts :—Glomerate ; with numerous rounded, polyp- bearing masses, the lowest branches slightly foliaceous; polyps densely arranged in characteristic hemispherical masses ; polyp stalks short, under 1 mm.; supporting bundle very strong ; point spicules very variable, 1— 6 pairs; no crown; grade IV.; . spicules of the upper cortex are large (4 mm.), bovered with blunt warts and lie transversely, lower cortex and canal-walls have stouter and shorter spindles, triradiates and multiradiates; colour : cortex white, polyps deep orange. WITHIN THE GENUS DENDRONEPHTHYA. 47 ANTHOCODIAL GRADE AND FoRMULA :— III. = (1—6) P + 0 Cr + very strong 8.B. Descriptive Noves :— 1. Colony as a whole. ‘Two handsome orange-coloured colonies, the largest 28 cm. wide and 20 em. high, eight others smaller in size. 2. Branching. Markedly glomerate. ‘The numerous rounded, polyp-bearing, boss-like masses predominate over the stem and main branches. Therefore the colonies must be included in Kiikenthal’s hempricht group. The lowest branches of the colony show slight foliation. 3. Colouring. Generally deep orange-coloured polyps and white cortex. The colony from station 240, however, showed quite a different colour-scheme, viz. bright red spindles in the supporting bundle and anthocodiz, but the general cortex greyish with opaque, white, almost porcellanous spindles. 4, Polyp stalks and their spicules. Short, under | mm. 5. Polyps. Densely arranged in very characteristic plump hem1i- spherical masses of 7 or so. 6. Polyp spicules. The anthocodial architecture without enter- ing into details, may be noted as having the 8 points often elon- gated with the tips meeting above the polyp. Another feature is the practical absence of any true crown. Kiikenthal notes in his definition of D. gigantea that each of the 8 points has 5-6 pairs of spicules, the uppermost much larger than the others and markedly projecting. It is easy enough to find on the specimens anthocodize which exactly correspond with this description. Text-fig. 8. 3 Text-fig. 9. Y AAA D. gigantea Verr. D. gigantea Very. Yet on the same colony may be found points with 5, 4, 3, or 2 pairs, and finally even a single pair, abutting against a few transitional small spindles leading on to the supporting bundle. Great care was taken to avoid damaged or disturbed antho- codiz, and the annexed diagrams show a series of frequently recurring types of architecture. A general feature may be noted that in the great majority of eases the two uppermost predominate, and that when the number 48 DR. W. RAE SHERRIFFS ON EVOLUTION is reduced to 2 pairs or to 1 pair, these occupy the same space as 3-6 rows. The larger numbers 5-6 were certainly less frequent than the rest. There is not in this case any confusion between what might be counted to points and what to crown, for in no case did I see what in other species is called a crown. It is also striking that after an examination of scores of anthocodiz hardly any case of intermediate spicules between the points was found. The supporting bundle is enormously developed. Its largest spicules may attain to a length of 6 mm. and project for 1-5 mm. These large spindles are densely covered with minute, blunt, thorns, except for 0°6 mm. at the projecting tip, which is smooth, as is also the case in LD. giganteé. Very marked in the large supporting-bundle spicules of the colony (Station 240) is the disappearance of the red colour from the proximal third of most of the largest. The same tendency is to be recognized in the yellow colonies. 7. Other spicules. The spindles of the upper cortex are often up to 4 mm. in length. They are covered with large, blunt, truncate and sometimes compound warts, and they lie trans- versely. Many show an abrupt narrowing at one end. In the lower cortex and canal-walls there are numerous stouter and shorter spindles exceeding 1 mm. in length and bearing crowded, rough tubercles often compound. Large triradiates and multi- radiates also occur, besides bracket-like and quite irregular forms approaching the stellate type. A few curiously starfish-like forms occur, and triradiates with one arm exuberantly branched. Many much smaller counterparts of the large types occur. There are also occasional long, narrow spindles. Within the hemprichi group the specimens agree best with D. gigantea, though this may not be obvious at first sight. Many of the anthocodize show but one pair of spicules in each point, which suggests approximation to D. carnea, D. dederleimi, and D. punicea. From the first they are at once separated by the massive spicules, which are visible from a distance and lie in very regular transverse rows. This conflicts too seriously with the original description by Wright and Studer (p. 196), where it is said: “The stem and branches ..... are leathery and thickly packed with fine spicules. The latter are not recognisable by the unassisted eye and lie scattered confusedly in several layers.” Both from D. dederleint and from D. punicea they are sufficiently separated by the enormous strength of the supporting bundle. Depvuctrion :—The specimens agree more closely with D. gigantea than with any other Glomerate known to me, and not least in showing quite an unusual degree of variability. Among the - important features of agreement the following may be noted :— (1) the very characteristic plump hemispherical masses closely beset with polyps ; (2) the supporting bundle of great strength and showing many WITHIN THE GENUS DPNDRONEPHTHYA. AQ spindles of over 4 mm. (the maximum size mentioned for D. gigantea) and projecting for 1:5 mm. (3) the large spindles of the supporting nual densely covered with minute, blunt thorns, except for the last half milli- metre at the projecting tip, which is smooth as in D. gigantea ; (4) the spindles of the upper cortex frequently reach 4 mm. in length, are covered with large, blunt warts, and lie trans- versely ; (5) the spicules of the lower cortex and of the canal-walls; (6) the eight anthocodial points are often elongated and ‘have the tips meeting over the polyp; (7) the lowest Michies of the colony show slight foliation ; (8) the colour-scheme of deep orange polyps and white cortex was found in some of the described species of D. gigantea. LOCALITIES :— Station 165 (3 specimens). Station 240 (3 specimens). » 213 (1 specimen). », 261 (2 specimens). ., 03 (3 specimens), 3. Dendronephthya punicea (Studer). Described by Studer, Ann. Mag. Nat. Hist. (6) vol. i. p.70 (1888). Diaenosis :—Glomerate; bundles of polyps in rounded clusters ; main stem and main branches little developed; polyps closely crowded together, polyp stalks medium ; ; Supporting bundle strong ; point spicules ive only, strong and converging ; crown of alot three rows of spicules irr eoularly disposed ; grade VII.; spicules: upper cortex has big, strong, thick spindles (2 mm.), lower cortex smaller and more fborny forms, including roundish and stellate types; colour: branches and all spicules scarlet; polyps pale yellow. ANTHOCODIAL GRADE AND ForMULA :— VI. =2P+4 3Cr + strong8.B. Descriptive Noves :-— 1. Colony as a whole. A handsome Glomerate colony, with the bundles of polyps forming rounded clusters, often well separated from one another. 2. Branching. ‘There is relatively little development of the main stem and main branches as compared with the stalks of the bundles of polyps composing the clusters. It is therefore one of the hempricht group. 3. Colouring. Branches and all spicules scarlet while the polyps themselves are pale yellow. A. Polyp stalks and their spicules. The polyp stalks measure almost 1 mm. in length. 5. Polyps closely crowded together. 6. Polyp spicules. The anthocodial armature has each point Proc. Zoou. Soc.— 1922, No. LV. 4 50 DR. W. RAE SHERRIFFS ON EVOLUTION consisting of two strong, very thorny, converging spicules bent at thé base in hockey-club-like fashion, one slightly larger than the other. Between adjacent points lie a pair of much smaller, straight spindles Text-fig. 10. longitudinally disposed, but in some eases there seem to be two pairs (thus connecting with D. dederlein). Below (7 AX) the points there is a crown of horizontal spindles in two or three rows, rather Zo irregularly disposed. The supporting bundle is strongly developed, especially as regards the three uppermost spindles, the lower median one of which projects for almost D. punicea Stud. 1 mm. Depuction :—Among the members of the hempricht group there are several with only one pair of spicules in each of the anthocodial points, and beside these this specimen must be ranked. It agrees very closely with D. punicea in the following features :— (1) each anthocodial point composed of two spicules, one slightly larger than the other ; (2) between adjacent points usually a single pair of small intermediate spicules ; ; (3) below the points the crown of horizontal spindles is present with two or three rows ; (4) the strong supporting bundle. Locaurty :—Station not marked. Il. DIVARICATA. 4. Dendronephthya ehrenbergi (Kiik.). Described by Kiikenthal, ‘ Korallentiere des Rothen Meeres,’ p- 56 (1904). Draenosts :—Divaricate; contours irregular; polyps in little groups of 5-8; polyp stalks short; supporting bundle medium ; point spicules 6 pairs; crown absent; grade ITI.; spicules: canal- walls show few bent rough spindles, the stalk cortex has very distinctive long curved red spindles and irregular clubs, discs, and spheres ; colour: rich red in the coonenchyma, anthocodial spicules grey-yellow, polyp spicules of the points white. ANTHOCODIAL GRADE AND KorMULA :— Il. =6p + 0Cr + medium 8.B. Descriptive Norres :— 1. Colony as a whole. Ten specimens in all, including two fine specimens agreeing closely with Kiikenthal’s definition of this species, the largest 40 cm. in length. WITHIN THE GENUS DENDRONEPHTHYA, Hil 2. Branching. Markedly divaricate but not markedly flat- tened. The contour of the colony is irregular. These features point to the swensoni section, and in that section the only species to which the specimens could be referred is D. ehrenbergi. The mode of branching is the same, and the resemblance of these specimens to Kiikenthal’s coloured figures is striking except that they have much less of a sterile stalk. 3. Colouring. Rich red in the ecenenchyma and with greyish- yellow anthocodial spicules. 4. Polyp stalks short. Text-fig. 11. 5. Polyps arranged in little divaricate bat groups of about 5-8. 6. Polyp spicules. There are in most ‘i 6 pairs of white spicules arranged en A y Xe gh chevron in each of the anthocodial lines : in Kukenthal’s forms there were 4—5. There is no very distinct crown. The supporting bundle consists of thick, rough red spindles (up to 1:5 mm.), projecting freely for about 0°5 mm. 7. Other spicules. The canal-walls show a few bent, rough spindles. The spicules of the stalk cortex are mostly very distinctive, long (up to 3°5 mm.), eurved, red spindles, also ovals and D. ehrenbergi Kik. irregular clubs covered with massive tubercles mainly with very characteristic rough blunt heads. There are also very rough discs, approaching spheres, covered with the same massive tubercles. mm. The small “stellate” forms at the base are very irregular in shape. Many are almost globular; many are minute quadriradiates; many show one predominant boss among the radiating knobs ; while many again correspond exactly with those figured by Kiikenthal for D. japonica (fig. Lb) *. Very few canal-wall spicules could be found. Those that occurred very sparingly were narrow rodlets with a few distinct tubercles. Depuction :— Although this splendid specimen has no uniform contour it presents many resemblances to D. brevirama (Burch.), such as : (1) the anthocodial armature of 6-8 spicules in each point, the uppermost not projecting ; (2) the short polyp stalk ; 3) the numerous 2 mm. spindles of the general eortex ; (4) the crowded stellate bodies of the lower cortex ; (5) in the polyp-bearing twigs a corymbose - dichotomous branching, almost identical with that of specimens which agree in detail with D. brevirama. * Kiikenthal, “Versuch einer Revision der Alcynarien.” Zool. Jahrbch. Jena. System. vol. xxi. p. 576. WITHIN THE GENUS DENDRONEPHTHYA. 75 _ Yet this specimen differs deeply from D. brevirama in the following respects : (a) it has a different type of supporting bundle, namely a sheath, instead of showing one markedly predominating needle ; (6) it has practically no spicules in the canal-walls ; (c) it has not more than ‘a hint of triradiates. It therefore seems best to establish a new species, closely like D. brevirama (an Umbellate) in some respects and very like D, japonica (a Divaricate) in others. Locaurry :—Station not marked. 21. Dendronephthya simplex (sp. n.). Draenosis :—Umbellate; compact; slightly flattened umbels not combined into large hemispherical masses; lower branches foliaceous ; contour uniform ; polyps compact; polyp stalks short ; supporting bundle strong; point spicules, 2, long converging only; crown of some 3 rows of curved spindles; grade VI.; spicules: cortex has stout tuberculate spindles but no stars; colour: generally drab-yellow. ANTHOCODIAL GRADE AND KoRMULA :— VI. = 1 or2 P+ 3 Cr + strong S.B. Descriptive NOTES :—- 1. Colony asa whole. The entire colony is 9 em. high, of which 2'5 em. go to the sterile stalk. The breadth of the polyparium is6mm. The polyp-bearing twigs form small umbels, but these do not combine in larger hemispherical masses. The specimen should therefore be referred to [Ciken- thal’s second group of the Umbellatz exif. 29: and to the florida group, though the S shape of the polyparium is rather cir- cular than long oval. There is a uniform, very slightly interrupted contour to the whole colony, which has a somewhat flattened circular shape. The general appearance is compact. Practically the whole surface is covered by the polyps on their umbels, and no polyps occur except on the surface. 2. Branching. Some of the lower branches are foliaceous. 3. Colouring. On the whole the colour is drab-yellow. D. simplex, sp. 0. 4. Polyp stalks. Short. 5. Polyps in compact bundles of 10 or so, dense corymbs. 6. Polyp spicules. The anthocodial armature consists of 8 points of 2 long converging spindles each, and it may be said at 76 DR. W. RAE SHERRIFFS OF EVOLUTION once that it is almost precisely that of D. piitteri (Divaricate) and not far from D. lutea (Umbellate, but with interrupted contour). The length of the point spicules is about 0°75 mm., but one is often slightly smaller than the other. Between the points there lies a pair of much smaller intermediates. At the base of the points there is a crown of about 3 rows of curved spindles, but some of these are often displaced so as to appear like the lower point spicules in D. lutea, in regard to which Kikenthal notes that an approach to a horizontal position sometimes occurs. This dislocation of what is really a horizontal band suggests the need of examining numerous polyps. There is a strongly developed supporting bundle, the largest spicule of which, about 2 mm. in length, projects beyond the polyp for about 0°75 mm. It is abutted by two strong spindles on each side and by a number of smaller ones transitional to the polyp. The supporting-bundle spicules are slightly curved spindles, densely covered with relatively minute blunt and narrow cones. The spindles of the 8 points are similar, but show the characteristic hockey-club-like bend at the base. 7. Other spicules. All over the cortex there is a felting of more irregular and stouter spindles, often “‘f” shaped, with larger and often compound blunt tubercles. No stellate forms occur, but very numerous, quite irregular types are found which might be derived from spindles and club-like ends or from boomerangs. Their compound warts are often very strong and stronger on one side. Locaity :—Station not marked. 22. Dendronephthya hyalina (Kiik.). Described by Kiikenthal in ‘Versuch einer Revision der Aleyonarien,’ ii. pp. 688-689 (1905). D1acnosis:— Umbellate; flaccid; hyaline; umbels not in large hemispherical groups; outline not uniform; distinct foli- aceous collar; polyps loosely aggregated ; polyp stalks medium; supporting bundle strong; point spicules 4—5 pairs, without pro- jection of the top pair; crown absent; grade III.; spicules: sterile stalk has numerous stellate and multiradiate forms; colour: cortex very hyaline, with loosely disposed white spicules. ANTHOCODIAL GRADE AND ForMULA :— III. = 4-5 p+ 0 Cr 4 strong 8.B. Descriptive Noves :— 1. Colony as a whole. Eight young colonies, the largest of which measures 6°5 cm. high. They are very flaccid in texture and hyaline in appearance. ‘The youngest forms show little hint of umbellate branching, but this is well marked in the biggest. As the umbels do not compose large hemispherical groups, the WITHIN THE GENUS DENDRONEPHTHYA. ve specimens should be referred to section B of Kiikenthal’s Umbel- late, and the presence of unequally long branches resulting in no very uniform outline points to the spinulosa group. 2. Branching. There is a ‘distinct foliaceous collar. Text-fig. 30. 3. Colouring. Cortex very hyaline spicules. 4, Polyp stalks. Slightly over 1 mm. long. d. Polyps. Few are in good con- \S dition, owing to the colonies having ZX been much compressed in most case 6. Polyp spicules. The anthocodial ——— points consist of 4-5 pairs en chevron, with almost no projection of the upper- most pair and almost no crown. D. hyalina Kik. developed, and the strongest spicule may project for half a millimetre. 7. Other spicules. The sterile stalk shows numerous stellate and multiradiate spicules. and bearing loosely disposed white small groups of 3 or 4. The supporting bundle is fairly well Locaniries :—Stations 257, 318,321. Saleyer, 10-25 fms. EXPLANATION OF THE PLATES. Prare I. Dendronephthya clavata Kiikenthal. Prate II. Dendronephthya gigantea Verrill. PuateE III. Dendronephthya annectens, sp. n. Obes Meth MeMiy As Wao . cilia edrr lated ore . a ta ee i dopeonayg me aus vier a 2) Ra, en ih a awh i M aes ee aur Peed! bik ie oye a ru es Diy ty bned An ey hs ee Ath hee aa + Nae a) ahs EBs \ se eis 1922. Balfour-Browne. PI. [. ll I2 Life history of the WATER BEETLE (PELOBIUS TARDUS) HERBST. VITTY & SEABORNE, LTD., LONDON. P.Z.S. 1922. Balfour-Browne. PI. Il. | milinelic. Life history of the WATER BEETLE (PELOBIUS TARDUS) HERBST. VITTY & SEABORNE, LTD., LONDON, ‘ d r a Se em P.Z.S. 1922. Balfour-Browne. PI. Ill. Life history of the WATER BEETLE (PELOBIUS TARDUS) HERBST VITTY & SEABORNE, LTD., LONDON, THE LIFE-HISTORY OF THE WATER-BEETLE. 74) The Life-History of the Water-Beetle Pelobius tardus Herbst. By Frank Batrour-Browns, M.A. (Oxon. et Cantab.), F.R.S.H., F.Z.8., F.E.S., Lecturer in Zoology (Entomology) in the University of Cambridge. Or Received January 10, 1922: Read March 7, 1922.] (Plates I.-ITT.) Sunumary. The paper has been paragraphed under the following headings :— 1. The family Pelobiide. p. 79. 2. Britannic distribution of P. tardus. p. 79. 3. The Imago. (a) Habitat and Habits. p. 82. (6) Longevity in artificial environment. p. 83. (c) Stridulation. p. 83. (d) Sexual differences. p. 84. 4, The Life-history. (a) Oviposition. p. 84. (6) Incubation. p. 85. (c) Vital staining of Embryo. p. 85. (d) The Larva. p. 87. (e) Food of the Larva. p. 89. (f) Stomodeum of the Larva. p. 90. (g) Habits of the full-grown Larva. p. 92. . The Life-cycle. p. 95. On (1) The Family Pelobiide. The family Pelobiidz contains the single genus Pelobiuws and only four species, one of which is European, one Chinese, and two Australian, The family holds a somewhat isolated position, being related to the Carabide and showing similarities of struc- ture with Dytiscide and other aquatic beetles of the series Adephaga, these being due, according to Sharp (10)*, not to close relationship but to adaptation to life in the water. (2) Britannic Distribution. The single Huropean species is not uncommon in Britain, but belongs to the Southern or “ English” type, as defined by Watson (16), being chiefly found in the Southern and Eastern counties of England. It has been recorded from one Scottish, * The numbers refer to the bibliography at the end of the paper. 80 MR. F. BALFOUR-BROWNE ON THE LIFE-HISTORY three Irish, and thirty, out of the seventy, English county and vice-county divisions. The only Scottish record, for Midlothian, occurs in Duncan’s Catalogue (3), where the species is described as “ very scarce, Pond at Coates” on the authority of Sir P. Walker. ‘The same record is repeated in the ‘ Entomologia Edinensis’ (17), but neither Andrew Murray (9) nor Sharp (11) refers to it. For the North of England there are two isolated records for the species, one for Northumberland South and the other for Cum- berland. The former is given by Stephens (15) on the authority of G. Wailes: ‘‘ Once near Newcastle by Mr. Hewitson.” Bold’s comment that this is ‘‘ probably erroneous as it has never since been met with” (4) is quite valueless. The Cumberland record is bare, being a note by James McDougall (8), who mentions having taken the species at Wigton, and although I have made enquiries T can find out nothing further about it. In the case of many species, records for localities far beyond the normal range are frequently regarded as due to errors in identification, but Pelobius is not a beetle likely to be confounded with any other, so that, unless a mistake has been made in labelling, records for this species may be regarded as correct, and are probably to be accounted for on the assumption that individuals have, from time to time, wandered northwards. In such a manner might the species extend its range, but in this case the attempts to extend northward seem to have failed up to the present. The extreme northern range limit seems to be Yorks S8.E., where the species is ‘‘ fairly common in the stagnant pools near Withernsea,” according to H. W. Baker of Hull(4). The records for Yorks S.W., Lanes 8., and Derby may or may not be for single specimens in each case. The Yorks $.W. record appears in the Victoria County History (2) as “ Doncaster (J. Wilcock).” The Lanes. record is an old one given by W. EH. Sharp (12) as “ Rufford (Gregson),” though, for some unexplained reason, this record is omitted in Sharp’s ievisseltists of 1906 (18) and 1908 (14). The fact that Jahn (6) mentions “ two specimens in a pool near Cheadle ” seems to suggest that here again we are within the normal range of the species; and although records even south of this are scarcee—in many cases only one in a county or vice- county,—there is little doubt but that the species is fairly common in the eastern and southern parts of the country. The accompanying ‘“typo-map” will give a general idea of the distribution of the species, and the letters representing the county and vice-county names will be sufficiently intelligible, as they are more or less in correct geographical position. Pelobius is scarcely to be described as established in Ireland, since, in the case of both Cork Mid. (7) and Clare (7), only one specimen was found, and possibly the Wicklow record, ‘near Lara,” in the Haliday MS. (5) also refers to a single specimen. I have worked many of the Irish counties fairly thoroughly, and OF THE WATER-BEETLE PELOBIUS TARDUS. 81 ok ! TYPOMAP SHOWING BRITANNIC DISTRIBUTION OF PELOBIUS TARDUS Herbst. SI or oe NS CA HB Ss RW RE EL BF AN WI El PN AS AM PM FF KI DN SG PC KF CT RE LE ED HD AY LA PE BW NN WD ED LD AN WT KB DF SK RX SN FE TY AR DO CU WL NY DM WM SL LE MO Me ML MY EY EM RO CV LH SE MVS WONG LE WH WE ° GR DB FT CH DY NM LS SG KC KD DU MN MG SP ST LR CB WN EN CL NT Qc OW WI CD RA HF WO WW NO HU WSES NK LK ST KK WX PB CM BR GE OX BX BD HT NE SK MC EC WA GM MMGWNW BK MX SE wc NS SW NH SR WK EK EG. SD IW SC WC . ; Seaha F : Underlined letters indicate County and vice-County from which Pelobius has been recorded. Bi.=Berks. CB.=Cambridge. CL.=Clare. CU.=Cumberland. DT.=Dorset. DY.=Derby. ED. (Scot.)=Edinburgh. EK.=EHast Kent. EN.=East Norfolk. ES.=BHast Suffoll. EX.=East Sussex. GM.=Glamorgan. HF .= Hereford. IW.=1. of Wight. LN.=Lincoln, North. LR.=Leicester + Rutland. MC.=Mid Cork. MX.=Middlesex. NE.=North Essex. NH.=North Hants. NS. (Eng.)=North Somerset. SE.=South Essex, SH.=South Hants. SL. (Eng.)=South Lanes. SN.=South Northumberland. S.=Surrey. ST. (Eng.)=Stafford. SY.=South-Hast Yorks. ‘WI. (Ive.)= Wicklow. WK.=West Reo WN.=West Norfolk. WO.=Worcester. WW.=Warwick. WY.=South-West Yorks. Proc. Zoou. Soc.—1922, No. VI. 6 82 MR. F. BALFOUR-BROWNE ON THE LIFE-HISTORY although in some of these, e.g. Wexford, there were many ideal ponds, and although I was specially looking for this species, I never found a specimen. I therefore regard the Irish individuals as having strayed from the English breeding centres, the species so far having failed to establish itself west of the Irish Sea. (8) The Imago. (a) Habitat and Habits. Pelobius is a pond species, never, in my experience, occurring in drains, lakes, or running water, but it seems to be very par- ticular as to the type of pond it frequents. It is apparently only found in ponds where the bottom is covered with a fine ‘ ooze,” so that gravel- or marl-pits are ideal habitats. In some cases the water of such ponds is clear and bright, while in others the pond is a favourite haunt of cattle during hot weather, the water then becoming thick and turbid and containing a high percentage of ammonia. The cattle pond is quite as thickly populated as the clean pond, so that it is the “ooze” which seems to be the im- portant characteristic of the habitat. In this ‘‘ ooze” the beetle frequently buries itself, so that only the apex of the body, often with a protruding air-bubble, is visible, and, during the summer, it may remain thus buried for even more than thirty minutes at a time, coming up rapidly to the surface to renew its air-supply, which takes from five to ten seconds, and then once more disappearing in the mud. By keeping the beetles in tumblers of water half filled with pond mud, I have found that at irregular intervals the air-bubble protruding from the apex of the body is released, and another then appears in its place and gradually increases in size. At first I was satisfied to explain this on the supposition that these were bubbles of exhausted air, but I can find no statement as to the mechanism of respiration in the Water-Beetle, or as to how the sub-elytral air is utilized; and I am not now inclined to speculate on the subject beyond saying that possibly the released bubbles ave merely due to the expansion of the air in the sub-elytral air- space, due either to the body temperature of the insect or to the temperature or pressure of the mud. The food of the imago consists of insect larva and worms, and while I was working at the life-history I kept my specimens in tumblers, feeding them mainly upon chopped-up earthworm, which they took readily. Agrionid dragonfly nymphs, Chironomus larve, Sialis larvee, and various other forms, were readily devoured, but more active types, such as Chloeon, could usually keep out of the way. The mouth-parts are typically mandibulate, and in nature the insect hunts for its food, swimming over the surface of the ground and poking into hollows and under stones. It seems to have a good sense of smell, since, in swimming over the surface of the mud, it will suddenly check and dive downwards; and, in such circumstances, if the beetle is dug up, it usuaily has some food OF THE WATER-BEETLE PELOBIUS TARDUS. 83 between its jaws. Also, in my tumblers, the beetles immediately became very active when a piece of worm was quietly dropped in, even when they could not see it, and they would at once begin to hunt round until they found it. 3 (b) Longevity of imago in artificial environment. In some of my tumblers I kept pairs, in. others single indi- viduals, each tumbler being supplied with a smali piece of water-weed, usually ZVodea, and also with a few pebbles to give foothold. The water was changed daily or on alternate days, according to circumstances, and I had no difficulty in keeping the beetles alive for months and even for more than a year in this way. Deaths that occurred were almost always among the pairs, and the male succumbed more frequently than the female. Presumably the cause of death in these cases was overcrowding, six or seven ounces of water, even when changed daily, not being sufticient for two individuals. [ had previously experienced this difficulty in keeping several individuals of a species in a confined space. For instance, Hydrobius fuscipes L., a smaller beetle than Pelobius, lived much. longer when one was kept in each tumbler than when two or three were sokept. Of twelve individuals of such a small species as Bidessus ninutissimus Germ. only six survived after a month, two after three months, and one shortly after that, the last survivor living for many months. At the same time six indi- viduals of this same species, each in a separate tumbler, survived for periods varying from one to three years, the water in these tumblers never being changed, only a little bemg added from time to time to make up loss due to evaporation. T have mentioned that, in the case of Pelobiws, when one pair was confined in a tumbler the male usually died first, which suggests that the male is not so hardy as thefemale. Experiments showed that the female lives longer than the male. As I was unable to find the details of about a dozen experiments carried out in 1913—before the work described in this paper was com- menced—I set aside four tumblers in May 1920, each containing much decaying leaf material rich in Chironomus larvee and various other forms of life. In each tumbler I placed one Pelobiws, using for the experiment two males and two females. A small piece of Elodea was placed in each tumbler to keep the water fresh, and each tumbler was covered with a piece of glass to reduce evapo- ration. Nothing was done to these tumblers until May 1921, by which time, out of the six ounces of water originally placed in each of them, only about three remained. The two females were, alive but both the males were dead. 3 (c) Stridulation. Pelobius used to be sold in St. Martin’s Lane, London, under the name of the ‘“Squeak beetle,” owing to its being able to §* 84 MR. F, BALFOUR-BROWNE ON THE LIFE-HISTORY make a loud, strident noise, which it does by rubbing the apex of the abdomen. against the underside of the apices of Sue elytra, in which position is a file-like structure (vide Pl. FIT. fig. 6). Both sexes are alike inthis. The beetle ‘‘ squeaks ” when it is alarmed or annoyed; for instance, when it is caught in the water-net or held between finger and thumb. In my tumblers, if one individual tried to seize a piece of worm upon which another was feeding, the latter “squeaked,” the squeak in this case presumably being equivalent to the growl of a dog with a bone. 3 (d) Sexual differences. There is very little difference in outward appearance between the sexes, the larger tarsal pads of the male being about the only one. The female is slightly larger than the male on the average and perhaps slightly broader in ‘proportion to length. (4) Life-history. The beetle is to be found in its habitat from March until October. In the latter month it descends into the mud in the bottom of the pond to hibernate, apparently remaining there without moving, its legs and antenne folded in against the body, until the following spring. Early in March, or possibly some- times late in February, it reappears and towards the end of March oviposition commences. 4A (a) Oviposition. In the labovatory my earliest date for the appearance of eggs is March 23rd, but under natural conditions eggs are very scarce until about the second week of April, when they become common. Whereas Dytiseid eggs are buried in the tissues of plants, the females possessing piercing ovipositors, Pelobius eggs are laid upon the surface of sub-aquatic vegetation. In my tumblers and tubs the eggs were laid in rows, end to end, sometimes as many as eight to ten in a row, though odd eggs were also to be found. When about to oviposit the female lies along the plant stem, clasping it with her legs so that the apex of her abdomen touches the stem. The egg, mon is blunt oval in form, 15 mm. long by -87 mm. broad and equally curved at both ends is very clear at the moment of extrusion and enclosed in a thin gelatinous envelope which at once adheres to the plant. This envelope quickly absorbs water and swells up into a thick ‘protective covering. Having laid an ege the beetle moves slowly forward, waving her antenne in the water and usually touching the stem with her palpi. She moves just far enough to permit the next egg to emerge so that it will touch the end of the previous egg, ane in this way she works along the stem, producing a string of eggs. OF THE WATER-BEETLE PELOBIUS TARDUS. 85 I thought at first that the beetle judged her distance by means of the palpi feeling their way, but, as I have mentioned, these organs do not invariably keep contact with the stem, while the gonapophyses do, so that. it appears as if these latter decide how far forward the beetle is to move after depositing an egg. I have not been able to determine the total number cf eggs laid by a female, the greatest number of which I have a record being twenty ; neither do I know whether an individual lays all her eggs during one short period or whether she lays a number of batches. The period of oviposition for the species is a long one, lasting from March until July. I have not actually found eggs later than June 15th, but I have found very young larve in August under rather interesting circumstances. On August 5th, 1915, in a disused chalk pit at Beckham in East Norfolk, where there were several shallow clear-water ponds, I found various stages of the beetle. There were old males and females, newly-emerged soft males and females, full- grown larve and a few larvee not more than ten days or a fortnight old. As incubation in the warmest period of the year occupies only nine days, the eggs from which these larvee hatched must have been laid in July. 4 (b) Incubation. The earliest laid eggs took twenty-five days to hatch, but the incubation period gradually became shorter until, in June, nine or ten days became normal. Eggs laid early in April and placed in an incubator at summer temperature hatched in nine or ten days, so that temperature is evidently the controlling factor in the length of the incubation period. J shall return to this point later on. So far as 1 can find, there is no special hatching apparatus such as thatin the larva of Dytiscus (1), the shell ripping from end to end along an irregular line and apparently always along the ventral side of the embryo, however it may happen to be lying in the egg. A small pair of spines are certainly visible on the head between the eyes, near the position of the hatching spines of Dytiscus lapponicus, but they are not functional, as is evident from the manner in which the shell rips, and I imagine that, if anything weakens the latter along the breaking line, it must be the tarsal claws. 4 (c) Vital Staining of the Embryo. In order to observe the developing embryo I removed many of the eggs from the water-plants upon which they had been laid and kept them upon cotton-wool saturated with water. I have used this method with the eggs of several other water insects— e.g., Agvionid dragonflies and others, Ranatra linearis, and 86 MR. F. BALFOUR-BROWNE ON THE LIFE-HISTORY various water-beetles both Hydradephaga and Hydrophilide ; and as a rule they do very well, but occasionally an egg shows signs of being unhealthy, that is, the surface becomes the home of sitter hens of minute roneran and Alege. Such an egg if left to itself invariably dies, but I discovered a way of treating it which usually saved it, and this was by saturating the cotton- eae with a solution of methylene blue. Any moderate strength seemed to be suitable, and the methylene blue killed off the Protozoa or Alge without injuring the egg. But the effect of the methylene blue was not confined to destroying the Protozoa and Algz on the ege-shell. It did not in any way injuriously affect the developing embryo, but certain parts of the living tissues took up the stain and stood out bright blue. The staining varied with different individuals, with the same individual at different times and possibly also with different strengths of methylene blue, but in all eases the embryonic appendages of the first abdominal segment, which are well developed in this species just as they ave in Dytiscus lapponicus, took up the stain (v. Pl. I. figs. 9,10, & 11). Otherwise the staining was very irregular. In the case of one larva just ready to hatch, the dorsal pharyngeal muscles and the - malpighian tubules were stained, and on the left side of each thoracic segment and of the first two abdominal segments a short line, curved in the thoracic, straight in the abdominal segments, was also stained. At one end this line came to the surface of the embryo and suggested a spiracle, although there is no pro-thoracic spiracle in the larva and the spots were above the real meso- and meta-thoracic spiracles which were also stained, Incidentally it may be mentioned here that none of the spiracles are functional in the larva. In the abdomen of this specimen parts of the dorsal longitudinal muscle system under the terga were stained, but not evenly, certain strands having taken the stain, others not having taken it. In the alimentary canal was some greenish fluid which I took to be methylene blue changed by the action of the secretions. I give this example merely to show the strange effect of methylene blue, and without making any attempt to explain it, T assume that the parts which stain are in some way active at, the time they stain. For instance, possibly the one-sided staining of the dorsai longitudinal muscle system in the specimen referred to may have indicated that those muscle-strands were doing work while the corresponding strands upon the other side were doing none. If this suggestion is sound it mee perhaps be possible, by watching any embryo develop in an egg kept in methylene blue, to recognize various parts functioning a different times. The staining of the first abdominal appen \dages, which disappear after hatching, is also of some interest if this suggestion is correct. These appendages first appear when the embryo is about three days old and after the three pairs of thoracic appendages, but they are more distinct than any of the following abdominal appendages which rapidly disappear, and they persist, as more OF THE WATER-BEETLE PELOBIUS TARDUS, 87 or less globular structures with a deep cup-like pit in them, right up to the time of the escape of the larva from the egg, and can be seen, especially after the staining referred to, after the larva has hatched, disappearing as it expands. I had somewhat similar experiences as to the staining powers of methylene blue in the case of the embryos of Dytiscus lapponicus, and I used to run a kind of hospital of methylene-blue tumblers for the eggs. But in the case of that species [ found another use for methylene blue. When rearing the larve 1 found that I could not keep them alive in tumblers, one in each, unless the water was so charged with the stain that it looked like ink, in which case the larvee grew up quite successfully. This solution was changed every two or three days, but in no case was I suc- cessful with the larvee in tumblers of plain water, even when I changed the water daily. Possibly the explanation in this case is that the methylene blue oxidized the fecal products and remnants of food of the larve and made them innocuous. 4(d) The Larva. The larva, on hatching, is colourless and possesses a long medio-dorsal spine ou the last abdominal segment and two lateral cerei. Within the egg these structures are doubled beneath the embryo, the last abdominal segments occupying the end of the egg. On escaping from the egg, the larva at once straightens out, and these appendages are at first about equal to the body-length, though before the end of the first stage the body has increased in length while the appendages remain the same. Within an hour the chitin hardens and the pigment has appeared and darkened. .The main colour of the larva is brownish green, but there is a pattern of pale yellow, and this varies in different individuals, some of which might be described as almost black with a few light markings, sale others are pale yellowish with a few dark markings. The colouring apparently does not vary much during the life of the individual—that is, one which is dark in the first stage will be dark in its final stage. An infinite number of diagrams could be given to show the range from dark to light, but I have selected five specimens, two at the end of their first stage (Nos. 1 and 2) and three in their third stage (Nos. 3, 4, and 5) to show the range (vide Pl. I. figs. 1 to 5). A study of these diagrams will make it clear that, although variations occur, the disappearance of the dark pigment in the lighter individuals is not a phenomenon which may happen at any spot on the surface of a segment, but occurs in more or less definite areas; and this is seen even better by comparing a number of individuals of about the same degree of darkness or lightness when a rather surprising uniformity of pattern is recognizable. Where i is, I think, some very interesting work to be done upon 88 MR. Fe BALFOUR-BROWNE ON THE LIFE-HISTORY the variations in colour-pattern within a species*. I have only paid slight attention to the subject, and only in the case of water- beetles. In many of the Hydradephaga one finds a dark ground- colour marked with yellow lines or spots, and there is usually considerable variation to be found within a species, some indi- viduals showing more yellow and others being almost without it. Such species as dgabus abbreviatus ¥., A. didymus Ol., and Platambus maculatus L. are most interesting studies, while in such a genus as Veronectes four out of the five Britannic species give excellent material for research. To return to Pelobius. In the newly-hatched larva the head is more or less triangular in shape and the prothorax is but little narrower than the head, the body thereafter tapering away to the Jast abdominal segment, which is only about one-fifth the width of the head, The larva is heavier than water and breathes subaquatically by means of a series of filamentous gills attached at the sides of the sterna of the thoracic and first three abdominal segments. At the inside of the base of each of the front legs is a tuft of three gill-filaments. Four rather larger filaments lie inside the bases of the middle legs, and the two posterior legs have each a tuft of three gills situated as in the case of the anterior legs. The abdominal gills consist each of a pair of filaments arising from a common base. and one of these is attached on each side to the posterior edge of the sternite, those of the first abdominal segment lying posterior to the pair of embryonic appendages already referred to. The gills are tracheate, with a single unbranched tube running the length of each filament. The legs are long and delicate. The coxa is large, the trochanter consists of two small segments, and the femur, tibia, and tarsus are each of one segment, the tarsus bearing a pair of claws almost as long as the segment. The tarsus in each case, but more especially of the second and third pairs of legs, is strongly feathered, and therefore forms an efficient swimming organ; but it has also another function. Gilled insect nymphs and larve, resting in stagnant water, usually adopt some means of cireulating the water about the respiratory apparatus. Agrionid dragonfly nymphs will frequently lash the abdomen from side to side; the nymphs of the Ephemerid, Chlocon, flap the lamellate gills ranged along the sides of the abdomen, and the larvee of eertain Perlids “dance” up and down, raising and lowering the body between the legs. The larvee of Pelobius use their feathered legs as fans, the middle leg on one side and the anterior and posterior legs on the other being raised off the * Vide Brown, Annetta F., “ Evolution of Colour-pattern in Lithocolletis.” Journ. Acad. Nat. Sci. Philadelphia, xvi. (1914) pp. 105-165 ; and Palmer, Miriam 4., “Some Notes on Heredity in the Coccinellid genus Adalia Mulsant.” Ann. Ent. Soc. America, iv. (1911) pp. 283-808. OF THE WATER-BEETLE PELOBIUS TARDUS. 89 support and rapidly vibrated so that a current of water passes under the body and bathes the gills. The first stage of larval life occupies from six to eleven days, seven or eight days being apparently the usual period, and the larva then moults. There is little difference in general appearance between first and second stage larvee except that the head is more rounded and less triangular in shape, and the three “ tail” processes are somewhat shorter in proportion to the body-length, but some change has taken place in the number of gill-filaments. Those of the prothorax remain as in the first stage, but there are now six filaments round the base of each of the middle legs and five round the base of each of the hind legs. The only change in the abdominal gills is the appearance of an additional pair of very small ones on the first segment, one gill towards either side of the sternite. This stage lasts from nine to thirteen days, when the larva once more moults, reaching its third and final stage. In the third stage again the body-length increases more than that of the “ tail’’ processes, which are now only about one-third the body-length, and the gills are also more complex in that each filament shows several transverse constrictions, as 1f an attempt had been made to produce segmented filaments such as are seen in the larva of Sialis. As to numbers of filaments, no change takes place on the pro- or meso-thoracic segments, but an additional pair appears on the metathorax, one usually very ‘short gill appearing anterior to and outside each. of the posterior legs. The abdominal gills do not again increase in number, but remain as in the second stage. A diagram showing the arrange- ment of the gills will be found on PI. IIT. fig. 4. In all stages of the larva nine pairs of spiracles are to be found, two thoracic and seven abdominal, but, as has been mentioned previously, these all remain closed and functionless throughout the larval period. As a rule it was quite easy to determine when a larva had moulted, because the cast skin would be visible in the tumbler or, ‘if the moult had just taken place, the larva would be white, though the pigment appears very quickly after the ecdysis; but individual larve possessed idiosyncrasies, some always retiring into the mud just before moulting and not appearing again until their colours had been restored. JI have records of quite a number of larvee, amongst my earlier batches, which apparently erew up without moulting at all, and it was not until I discovered skins amongst the mud at the bottom of the tumblers that I realized what had happened. 4 (e) The Food of the Larva. I had considerable difficulty at first in keeping the larve alive. I adopted the same method as that used in the case of the 90 MR. F. BALFOUR-BROWNE ON THE LIFE-HISTORY Agrionid dragonfly, Hydrobius fuscipes and Dytiscus, keeping records of individuals, each in a separate tumbler, but I could find no food which they would eat. The newly-hatched Pelobius larva is only about 2 zm. long in body-length, so that minute food was obviously necessary, but Paramecium, upon which I fed the newly-hateched dragonfly nymphs, had no attraction for them, and minute alge, Cladocera, and Copepoda were equally useless. I noticed, when I placed a larva in a tumbler with a layer of mud at the bottom, that this mud was continually examined and even burrowed into, which suggested that the necessary food was some mud-inhabiting species. As, however, these larvee also died, I made a journey to one of the ponds where the beetle occurred and brought home a supply of mud from there, and, on using this in the tumblers, it was at once obvious that the food-problem was solved. This mud was swarming with Tubifex, and it was most interesting to watch the larva hunting for and capturing its prey. It would move slowly over the surface, touching it with its antenne and palpi, and, on dis- covering a Zubifex burrow, the larva would either at once leap into it, sometimes disappearing except for the apices of its “tails,” or it would sit and watch, like a cat at a mouse-hole, its action presumably depending upon the position of the worm in the burrow. Sometimes the snatch at the prey would fail and the larva would move elsewhere, but with these first-stage larvee it was difficult to decide whether the pounce had been successful, as it was rare for. the larva to appear from the burrow with the worm in its mouth. With the larger larve one could see the struggle, the worm trying to withdraw to the bottom of the burrow, and the larva, as it were, sitting back upon its haunches and pulling with all its might to get the worm out. In the case of these larger larvee, however, the feeding was sometimes done in the burrow. The feeding process is rather peculiar; the larva sits motionless with most of the worm projecting from its mouth and wriggling furiously, the larva to all appearance doing nothing; suddenly, however, the latter makes a gulp, and a little of the projecting part of the worm disappears into the mouth. After a short interval another gulp takes place, and thus by a series of gulps the worm disappears down the “throat.” The larva apparently always seizes the worm by the end, never by the middle. So far as I could find, Zubifex is the only food of the larval Pelobius, as, although I tried various other materials and could get the larve to eat Chironomus larve, I never succeeded in rearing one on any diet other than the little red worm. 4 (f) The Stomodeum of the Larva. The guping down of the worm struck me as rather extra- ordinary. There was no chewing or biting, and yet an examination of the contents of the alimentary canal always showed chewed-up OF THE WATER-BEETLE PELOBIUS TARDUS, 91 feod material. An examination of the stomodeum, however, explained the mystery. The mouth-parts consist of a pair of sharply-pointed mandibles which overlap when closed, the right overlying the left. The maxille are palp-like in appearance, the maxilla itself forming the basal part of the palp, and, when at rest, the whole structure is contracted by being telescoped within the base (v. Pl. IL. figs. 1 to 4). The labium has an apical piece, which, from its appearance, | will call the “scoop ” (s), and which, when at rest, is contracted within the basal piece which bears the two short labial palpi (v, Pl. II. figs. 1 to 4). When the floor of the mouth is examined (v. Pl. II. fig. 1), it will be seen that at the base of the extensible part of the labium is a small, projecting piece in the median line. This piece, which I will call the “ flap” (/), rises up in the “scoop,” and when the mouth is closed this flap stands up, and the lobes of the epi- pharynx—this structure being split like a hair-lip—descend from above and thus block the entrance. The anterior part of the floor of the mouth is covered with minute projections arranged more or less in rows radiating from a deep groove, while ‘the anterior part of the scoop is similarly studded. In the posterior region of the floor of the mouth is a dark brown chitinous plate (hyp) with a raised median piece posterior to it, and J regard these structures as representing the hypopharynx. On either side of the chitinous plate there projects inwards a membranous lobe (mal) which is extremely difficult to make out; L examined a large number of specimens before I could satisfy myself that it really existed. Possibly these lobes represent the maxillule. Immediately behind the hypopharynx the mouth contracts into the gullet, which is very short and opens out at once into the cesophagus (v. Pl. I. fig.6). This again is very short, and there is no pro-ventriculus clearly marked off at its posterior end, but this region is armed with eight dark-coloured spines (v. Pl. I. figs. 5 & 6), which project forward in the walls and posteriorly are enlarged. The inner faces of these spines are armed with strong, backwardly turned teeth which are specially numerous on the enlarged bases, and evidently form the grinding-mill by means of which the food is broken up. With the exception of a pair on each side whose bases are fused together, each spine is independent of the others. The median ventral one is the longest, and extends forwards for about two-thirds the length of the cesophagus, none of the other spines extending more than half the length of the ceesophagus. To the esophageal walls, between the pro-ventricular spines, are attached muscles which run to the head capsule, so that the cesophagus is capable of considerable expansion and contraction. In this case, therefore, almost the whole stomodeum acts as a sucking-pump, there being no pharynx marked off for this purpose, as is usually the case. 5/3} MR. F. BALFOUR-BROWNE ON THE LIFE-HISTORY The mechanism of all this complex apparatus seems to be as follows :—The Pelobius larva pounces upon the Tubifex and, by expanding the csophagus, sucks the end of the worm into the stomodeum, and the mandibles then close so that the prey is held either between the two, or more usually, between the upper and the epipharynx. I think, therefore, that it usually passes into the mouth between the two lobes of the epipharynx (v. Pl. I. figs. 11 & 12). At the moment of seizing upon the worm the maxille and labium are shot forward and the prey is received into the “scoop, ’ and these parts then retract, the flap in the “scoop” pressing the worm upwards into the epipharyngeal slit, and the closing of the jaws bending the body of the worm upwards so that it is completely jammed and cannot move backwards or forwards (v. Pl. I. fig. 12). At the first suck, the end of the worm pre- sumably reaches well down into the pro-ventricular mill, where it is well bruised, and, when this portion has been sufficiently crushed, the hold upon the worm is momentarily relaxed by the opening of the jaws and the shooting forward of the maxille and labium ; the sucking action is repeated, and once again the mouth- parts jam the body until another section has been crushed by the pro-ventricular mill. 4 (g) Habits of the full-grown Larva. The larva is full-fed about twenty-two days after the second moult, by which time it measures about 15 mm. in length, the median “ tail” adding another 5 mm., and it is now ready to leave the water. In my tumblers it was always evident when this stage was reached, because the larva would swim round and round at the surface and endeavour to climb up the glass. As I wished to watch the work of the larva in forming its pupal cell in the earth, I tried, with this species, the same method as I had used successfully with Dytiscus lapponizus—that is, | placed the larve in an earth-bottomed vivarium with a glass side, made an artificial burrow in the earth against the glass, and carefully directed the steps of the larva to the entrance of the burrow, closing the mouth of the latter when the larva had entered. With this species, however, the method did not work so successfully, as the iarva, having reached the end of the artificial burrow, usually continued it on its own account and as a rule in a direction away from the glass. Digging up such larvee and starting them again seemed to discourage them, and they were apt to sulk and make no cell at all. I therefore devised a special glass cell consisting of two Jantern- slide cover-glasses. Between these two, at adjacent corners, two small pieces of cork, half an inch thick, were placed so that, along one side, these two plates could not come into contact. Along the opposite side the two plates were brought into contact and held OF THE WATER-BEETLE PELOBIUS TARDUS. 93 together by string or narrow adhesive tape. The cell was formed between these two glasses, a strip of cotton-wool filling up the open ends between the two plates and a similar strip being pushed down into the narrow angle where the two plates came together, and this cell was then filled up with sand (v, Pl. III. fig. 9). By keeping the cotton-wool moist by occasional wetting, the sand was also kept moist and in a suitable condition for the require- ments of the larva or pupa of Pelobius. An artificial burrow was then made perpendicularly downwards in the sand with a pencil, which, of course, after penetrating the sand for about 2 inches, came into contact on either side with the glass plate ; and into this burrow a larva was pushed uncere- moniously, the entrance to the burrow being then closed by covering the surface of the sand with a narrow strip of glass. The larva was thus in such a position that whichever way it worked it was compelled to form its cell in contact with the glass. These glass cells were kept in the dark so that the larva should not be worried by the light, but I found that when once the larva had accepted the situation, the light did not affect it unless it was very intense. Tn starting its burrow, the larva digs itself in head first, and having reached the depth at which it is going to make its cell, a depth which varies from 2 inches to at least 5 under different soil conditions, it moves its head from side to side and, with its jaws, chews up the pellets of earth, causing the grains to pack more closely, and in this way it forms a space around itself. Practically the whole space of the cell is produced by this re- arrangement of the minute particles of soil, though the entrance from the tunnel becomes closed during the operation. At first the larva, in a straight tunnel, is rather confined in its movements, but, as soon as space permits, it changes its position and sits up, alternately chewing up the pellets of soil and pressing them into the walls, using its head asa 1am. The three “tail ? processes, which one would expect to be very much in the way in the confined space, become very useful. They are bent over the back, like a Squirrel’s tail, and wedged against the wall of the cell, and on them the body turns about freely in all directions {v. Pl. HI. figs. 1 & 2). The work of constructing the cell usually took from twelve to fifteen hours, though sometimes two or three days seemed to be necessary; but in these cases the larva seemed to work inter- mittently and without enthusiasm, whereas in most cases great energy was shown and no rest taken until the cell was completed. After the completion of the cell the larva rests for from six to eleven days (v. Pl. Ill. fig. 3) before it changes into a white pupa of the usual Coleopterous type. The pupal period lasts” about sixteen days, and the imago, white at first, takes twelve hours or longer to attain its colours, and remains in the cellfor a week or more before breaking out. 94 MR. F. BALFOUR-BROWNE ON THE LIFE-HISTORY (5) The Life-Cycle. Thus from the hatching of the egg to the appearance of the imago occupies from nine to about fifteen weeks. On this basis, the earliest imagines should appear about the middle of June, and this raises the question whether eggs laid in July and August may not be the commencement of a second annual generation. From an examination of the ovaries of a few beetles in August which emerged in the previous June, I am inclined to think that there is normally only one generation a year. Perhaps also the longevity of the beetle, at any rate of the female, which will live for three years and perhaps longer, suggests a one-year cycle. On the other hand, the possibility of a few precocious indivi- duals breeding soon after emergence must not be lost sight of (and also the possibility of old females laying a second batch of eges must be kept in mind), though it does not follow that the resulting larvee would complete their metamorphosis. I make this suggestion that a second generation may be commenced, even if not completed, because of a number of experiments I made some years ago with a few species of Hydrophilid beetles. By keeping these in tumblers in an incubator at a high summer temperature, 19° to 21°C. (66° to 70° F.), I found that it was possible to get egg-cocoons from some of them (e. g¢., Hydrobius fuscipes L. and Philhydrus mari- timus Thoms.) in November, after only a month of treatment— ege-cocoons which, under normal conditions, would not have been produced until the following April or May. A hot June and July might therefore cause the ovaries of Pelobius to mature rapidly, and this would account for the young larve which I found in August at Beckham. The life-history of Pelobius is interesting because of the eill-bearing larva, which does not require to come to the surface to renew its air-supply. In this respect it differs entirely from the larve of the Dytiscids and Hydrophilids, and resembles those of Gyrinids and Haliplids, but like, apparently, all other aquatic beetles, the pupa requires free air, and therefore, like the majority of them, the full-grown larva leaves the water and burrows into the soil, where it forms its pupal cell. The change on the part of the larva from an aquatic to a terrestrial habit does not involve a change in the respiratory apparatus, although I had expected one or more pairs of spiracles to open during the last larval stage, as happens in the last stage: of the nymph of the dragonfly. But, whereas the dragonfly nymph comes from the water into the air with the object of drying itself, the Pelobiws larva leaves the water and quickly burrows into the earth, and therefore, presumably, is under sufficiently damp conditions to allow the gills to function until the larval skin is thrown off. OF THE WATER-BEETLE PELOBIUS TARDUS, 95 This work on the life-history of Pelobius was begun in 1915 as the commencement of a study of subaquatie beetle larve, my intention then being to work out the life-histories of a Gyrinid and an Haliplid. The work on Pelobius was almost completed by the autumn of that year, when military duties took me away from Cambridge. Since my return to entomological work, time has not permitted me to continue on this line of research, and I have therefore merely completed a number of observations on Pelobius and written up the life-history of the beetle. Bibliography. . Batrour-Browne, F.—‘*‘ The Life-history of a Water-beetle.” Journal Royal Institution, 1913. Bayrorp, E. G., and Lawson, THompson M.-—Yorkshire: Victoria County History. Coleoptera. 1907. . Duncan, J.—‘* Cat. of the Coieopterous Insects found in the neighbourhood of Kdinburgh.” Mem. Wernerian Soe. vi. pp. 443-538. 1826-31. . Fowxer, W. W.—‘ Pelobins tardus Hbst.in Yorks.” Natura- list, xvil. p. 234. 1892. . Hauipay, A. H.—‘‘ MS. Notes on Irish Coleoptera.’’ Publ. in Proc. Belfast Nat. Field Club, Appendix viii, 1885. JAHN, L. H.—‘‘ Coleoptera occurring in North Staffordshire. Prelim. List.” North Staffs. Field Club Ann. Rep. and Trans, 1904-5, xxxix. p. 73 et seq. Jounson, F. W., and Hatsert, J. N.—‘A List of the Beetles of Ireland.” Proc. Roy. Irish Acad. ser. 3, vi. 1901. . McDoveatt.—Nat. Hist. Journal. York and School Register, vi. 1880-82. . Murray, A.—Coleoptera of Scotland. 1853. . SHarp, D.—‘‘ Dytiscide.” Sci. Trans. Roy. Dublin Soe. ser. 2,11. 1880-82. ‘“‘ The Coleoptera of Scotland.” Scottish Natura- list, 1. to iv. 1871-78. 29 . Saarp, W. E.—“ The Hydradephaga of Lancashire and Cheshire” (Ent. Soc. of Lancs. and Cheshire), 1892. Victoria County History: Lancashire. 1906. “ Coleoptera of Lancashire and Cheshire ” (Lanes. and Cheshire Ent. Soc. 31st Ann. Rep. and Proce. for 1907). 1908. . SrepHens, J. F.—IIl. Brit. Ent., Mand. ii. App. 1829. . Watson, H. C.—‘‘ Cybele Britannica.” 1847-59. 4 vols. and Supp. 1. 1860. . Witson,J,,and Duncan, J.—‘‘ Entomologia Edinensis.” 1844. 96 MR. F. BALFOUR-BROWNE ON THE LIFE-HISTORY EXPLANATION OF THE PLATES. Puate I. Figs. 1-5. Colour-patterns of Pelobius larve. Figs. 1 and 2 are taken from larvae Fig. Pig. Wig. near the end of their first stage, and Fe. 3, 4, and 5 from larvee in their third stage. The specimens were merely selected tu show the range of pattern, there being no noticeable difference in different stages of the same larva. Diagram of alimentary canal of a Pelobius larva to show proportionate lengths of Stomodzeum (8), Proctodeum (P), and mid-gut (MG). The egg of Pelobius 24 to 36 hours old. The embryo has begun to appear. Note the large head-lobes, the first signs of the neural canal and the post. end submer ged in the ‘yolk, which, : at this stage, is not segmented. . The embryo about 3 days old (in June when incubation lasts only 9 or 10 days), showing the appendages appearing, the antennz between the head- lobes and the first abdominal appendages, the last of the series at this particular stage. The yolk is now segmented and the amnion is completely formed. . The embryo 4 to 5 days old (in June); the amnion still intact. The first abdominal appendages very large and distinct. The depressions at the sides of the thorax and first abdominal segments are, presumably, the commencements of tracheal-tube formatiun, though the fact that there is no prothoracic spiracle perhaps makes this doubtful. The embryo 7 or 8 days old (in June). The eyes have just developed, the membranes have ruptured, rolled back, and disappeared within the dorsal region of the embryo. ‘he first abdominal appendages are now smaller. . Front view of head of a third-stage larva to show the relative positions of parts. CL=Clypeus. LBR=the labral ridge, beneath which are the two epipharyngeal lobes, EP. F is the “flap” standing up in the “scoop” of the labium, LAB. LAB.P=labial palp. Mx=mazxilla, sunk in its base. Mx.P=Maxillary palp. M=Mentum. GU=Gular sclerite of head capsule. ANT=Antenna. O=Ocelli. MDB=Mandibles which are in a position for seizing the prey. Front view of head as in fig. 11, except that a worm is in the mouth and the mandibles have closed upon it. This diagram shows how the worm is bent upwards by the mandibles and thus jammed so that it cannot move backwards or forwards. Prate II. Mouth-parts of a third-stage larva with labrum, epipharynx, and “roof” of mouth removed to show the relative positions of labium (LAB), maxilla (Mx), fully extended, and mandibles (Mdb). S=the “scoop” fully extended and F, the “flap” in the “scoop.” Mxl possibly represents the maxillule. Hyp, a dark chitinous plate in the floor of the mouth just where 1t narrows to the “ gullet,” presumably represents the hypopharynx. Note the four holes in this plate which seem to be always present. Labium and maxille as in fig. 1, but with the “scoup” fully retracted and the maxillee almost fully retracted within their bases. Ventral view of labium and maxilla, showing the “scoop ” fully extended and one maxilla extended and the other fully retracted within its base. M=mentum, not clearly marked off from the bases of the maxillz, but recognizable. G=the gular sclerite. Ventral view of labium as before, but with the scoop fully retracted. It will be seen therefore that the scoop telescopes into the rest of the labium just as the maxille do into their bases. 4 Vigs. 5 & 6. Views of the pro-ventricular spines of a third-stage larva, in situ, showing the backwardly directed ‘“‘ teeth” and hairs on the inner faces. Fig. 5 is a view from above and fig. 6 is a lateral view. These two figures are from: camera lucida drawings, and the millimetie scale indicates their magnification. OF THE WATER-BEETLE PELOBIUS TARDUS. 97 Prats III, Figs. 1 & 2 are from sketches of the positions adopted by the larva at work when constructing its cell. ‘ig. 3 shows the position of the larva resting in the completed cell, waiting to change into a pupa, i?) 19 Fic. 4. Ventral view of a third-stage larva, showing the gills on the left side of the thoracic and right side of the abdominal segments, the others having beer omitted for the sake of clearness. This diagram is based upon a camera lucida drawing, Cc=Coxal cavity. Sp=the closed meso- and meta-thoracic spiracles. Fig. 5. A diagram of one of the glass cells used to enable the work of cell con- struction by the larva to be watched. Fig. 6. Under side of the right elytron of a male specimen of the beetle to show the file, upon which the apex of the abdomen rubs to produce the “ squeak.’’ There is a file on each elytron, and the files are alike in the two sexes. Proc. Zoou. Soc.—1922, No VII i . Brie as i on a er « ’ i sae WEL oS isc Fo cenig ia & Efe hi iba ON THE VAGUS NERVES OF THE EDENTATA G9 6. On the Vagus and Sympathetic Nerves of the Hdentata. By Caarues F. Sonntag, M.D., F.Z.S., Anatomist to the Society. [Received December 28, 1921: Read February 21, 1922. | (Text-figures 1-5.) In a former paper (2) I showed that the cervical parts of the vagus and sympathetic nerves are fused to form vago-sympathetic cords (type 1), or are simply linked together by communicating branches (type 2). And these forms are distributed in the Edentata as follows :— Type 1 :—Bradypus tridactylus. Type 2 :—Huphractus villosus and Tatusia novemeincta. In Tanandua tetradactyla both forms are present in the neck of the same animal, the first occurring on the right side and the second on the left. In all species the course of the vagi from the foramen lacerum posticum to the point where they reach the sides of the cesophagus in the posterior part of the thorax is similar to that in the Marsupialia (1,2). The relations differ, however, for the branches of the aortic arch are more frequently similar to those in Homo. Consequently the relations resemble those described in Anthro- potomy. The branches also differ in many respects from those in the Marsupialia. Tamandua tetradactyla (text-figs. 1 and 2). The vagus nerves communicate with the glosso-pharyngeal, spinal accessory and hypoglossal nerves by very fine filaments, but no branches run to the sympathetic in the anterior part of the neck. And neither vagus has any trace of the ganglion nodosum in the neck, but they are flatter and wider in the foramen lacerum posticum. The usual branches of distribution are given off, but these have a longer course, and are more easily studied than in most Mammals. They form thick bands, with the spinal accessory _ nerves in the lacerate foramina. The pharyngeal nerve (a) is the largest cervical branch. It runs along the side of the pharynx and wsophagus and supplies both. It forms a plexus with branches of the glosso-pharyngeal nerve (ix), but no branches of the sympathetic are seen running into it. On the right side a branch (/.6.v) of the hypoglossal nerve (xii) communicates with the pharyngeal branch ; and this nerve may correspond to what is termed the “ lingual branch of the vagus” in human anatomy (8). The superior laryngeal nerve (6) divides into three branches. One gains the interior of the larynx through the thyro-hyoid 100 DR. C. F. SONNTAG ON THE VAGUS AND interval, and the others end in the extrinsic laryngeal muscles. It does not communicate with the sympathetic. Text-figure 1. The cervical and anterior thoracic parts of the vagus and sympathetic nerves in Tamandua tetradactyla. Sg.b.: branch of hypoglossal nerve to sterno- glossus; V-S: vago-sympathetic cord. Other letters in text. The nerves have been separated for demonstration purposes so that the super- ficial and deep cardiac plexuses appear in line. The right recurrent nerve (d) has the usual origin, course, and relations. It communicates freely with the sympathetic ( S) and the superficial cardiac plexus, and is brought into communication SYMPATHELIC NERVES OF HH BDENTATA. 101 with the le/t recurrent nerve (e) thereby. The latter also com- municates wth the lower cardiac branch of the left vagus. But no direct branch runs between the two recurrent nerves. The Cardiac Nerves (jf) :—The left vagus gives two fine nerves to the superficial cardiac plexus (S.C.P), wherein they communi- cate with the cardiac branch from the inferior cervical ganglion of the left sympathetic. But no ganglia are found in connection with them. The right vagus gives two large and two fine filaments to the deep cardiac plexus (D.C.P), and fine filaments to the superficial one. Reddish-brown ganglia are connected to the nerves. ‘There are no upper or middle cervical cardiac nerves. The pulmonary nerves (g) come from the vagus ou the left side, and from the vagus and deep cardiac plexus on the right. In the posterior part of the thorax branches are given off to the esophagus and aorta, and the esophageal plexus (O.P) is a link between the vagi, for there are no direct branches connecting them. And the main cords run through the diaphragm on the sides of the cesophagus. Abdominal Parts of the Vagi:—The left vagus runs along the ventral surface of the stomach close to the lesser curvature. The right one crosses behind the cesophagus and breaks up into a number of filaments. These reunite and the vagus ends in the right part of the cceliac plexus. Several gastric nerves (j) run to the dorsal aspeet of the stomach, and communicating twigs run to the hepatic (7) and splenic (£) plexuses. f did not observe any communications between the vagus and phrenic nerves; but fibres may get from the vagus into the phrenic sympathetic plexus and be carried up to the diaphragm to anastomose with the phrenic nerve. The Cervical Sympathetic :—The superior and middle cervical ganglia are absent, but the inferior cervical ganglia (1.C.G) are large and round. They give off the cardiac nerves (c.b.s), thymic branches (¢.0), subclavian (S.V.B) and phrenic commu- nicating (¢.p.n) nerves. The left cardiac nerve ends in the superficial cardiac plexus. The right cardiac branch ends in the deep plexus. The right sympathetic communicates with the superficial cardiac plexus and left recurrent nerve. The Cardiac Plexus surrounds the aorta and its branches, supplies them and the heart, and gives twigs to the right pulmonary plexuses. It consists of superficial (S.C.P) and deep (D.C.P) parts. The former lies on the ventral surface of the aortic arch. It receives branches from both inferior cervical ganglia, two branches from the left vagus, and twigs from both recurrent nerves. The latter lies between the aortic arch and innominate artery ventrally, and the trachea dorsally, and it contains three reddish-brown ganglia (C.P.G). It receives fibres from the right inferior cervical ganglion, and four branches from the right vagus. Branches are given off to the trachea (¢.p), heart, and right lung. Both plexuses communicate by fine nerves. 102 DR. C. F. SONNTAG ON THE VAGUS AND The thoracic cords have few ganglia. The first left ganglion (G.S) is long, and connected by the Annulus of Vieussens (A.V) to the inferior cervical ganglion. When ganglia are present they lie on the necks of the ribs. The left cord is tortuous and diminishes in thickness from before backwards; it united Text-figure 2. The posterior thoracic and abdominal parts of the vagus and sympathetic nerves in Tamandua tetradactyla (A), and the fused vagus and spinal accessory nerves in the lacerate foramina (B). The vagi have been drawn to the right so as to extend the cceliac plexus. Letters in text. posteriorly by two filaments with a branch running up to the aorta at the upper end of the descending part. The right sympathetic is larger than the left one, and is not connected to SYMPATHETIC NERVES OF THE EDENTATA. 103 any ascending aortic branch. In addition to the communicating nerves to the intercostals, and filaments to the esophagus, sorta and mediastinum, splanchnic nerves are given off. . The Solar or Celiac Plexus (text- tig. 2) consists of several ganglia and communicating bundles of fibres, and is divisible into upper and lower parts. It is connected to the sympathetic cords by splanchnic nerves (4); these are three in number, but the lowest is really a bundle of fibres. It lies between the layers of the mesentery, along the course of the cceliac axis, and is not set across the vertebral “column, The following are the branches of distribution :— A. From the upper part of the plexus. 1. Hepatic plexus (i) which communicates with the gastric branches of the right vagus (7) by nerves meeting in a small ganglion. 2. Splenic plexus (k) communicating with the right vagus 2a supplying spleen and pancreas. 3. Phrenic plexus (/) communicating with the phrenic nerve in the diaphragm. B. From the lower part of the plexus. 1. Right renal plexus (mn). 2. Communicating branches to the splanchnic nerves and aortic plexus (1). The Aortic Plexus (p) accompanies the aorta and its branches. It communicates with the splanchnic nerves, sympathetic cords (RS and LS), and their common ganglion. And it gives off the left renal plexus (0). The abdominal parts of the sympathetic have few ganglia. They gradually diminish in size and meet in a common ganglion in front of the sacrum. From the latter nerve plexuses run to the pelvic viscera. Huphractus villosus (text-figs. 3 and 4). The vagus and sympathetic nerves differ in many ways from those in Tamandua. The Vagus Nerves:—The ganglion nodosum (G.N) is present on both sides. And the branches of communication to the other cranial nerves and the superior cervical ganglion of the sympa- thetic are well marked. In the middle of the neck a thick communicating branch connects the right vagus and sympathetic. The pharyngeal branch (a) arises above he ganglion nodosum, and no nerve connects it to the hypoglossal nerve. The superior laryngeal nerve (6) is a thick branch arising orn the centre of the ganglion nodosum. It divides into a thick internal laryngeal nerve (i./.2) and a thin external laryngeal nerve (e./.2) which receives a loop from the superior cervical 104 DR. C. F. SONNTAG ON THE VAGUS AND ganglion of the sympathetic. Two small ascending branches pass to ganglia whence fibres run up to, and along the course of the arteries. At the level of these ganglia the common carotid artery divides into thyroid, ectocarotid, entocarotid,and occipital arteries, and fibres from the ganglia accompany them. The ganglia also communicate with the superior cervical ganglion of the sympa- thetic. Text-figure 3. DY ocr The cervical aud anterior thoracic parts of the vagus and sympathetic nerves in Euphractus villosus. The superior cervical ganglion (S.C.G) has been cut down as it is very large and round and masks other structures. Letters in text. The Cardiac Nerves (f):—The two cardiac branches of the left vagus end in a well-marked ganglion lying in front of the lower SYMPATHETIC NERVES OF THE EDENTATA. 105 border of the aortic arch. Fibres radiate from it over the arch and its branches, and constitute the superficial cardiac plexus. A branch of the left recurrent nerve (¢) ends in the plexus, but the recurrent and lower left cardiac nerves do not communicate as in Tamandua. The right vagus gives two branches to the deep cardiac plexus, but none to the superficial one. The right recurrent nerve (d) has the usual origin, course, and relations. It communicates with the sympathetic, and a fine branch crossing the front of the trachea, connects it to the left nerve. The pulmonary nerves (g) are given off as the vagi reach and pass behind the roots of the lungs. And they are connected by fine nerves to the deep cardiac plexus. The posterior thoracic parts of the vagi differ from those in Tamandua. The vight vagus runs along the ventral aspect in the csophagus. The left nerve runs along the dorsal surface of the esophagus and communicates with the right one. It gives ascending, transverse, and descending branches to the ceesophagus and aortic plexus. Jn the abdomen the right vagus breaks up into a number of branches which run along the ventral surface of the stomach close to the lesser curvature. The left vagus gives several twigs to the celiac plexus and its offshoots, and ends in the dorsal wall of the stomach. The Cervical Sympathetic:—The superior cervical ganglion (S.C.G) is large and round. It communicates with the cranial nerves in its vicinity, and sends a loop to the external laryngeal nerve (el.n). Twigs of communication run to the ganglia connected to the laryngeal nerve. The internal carotid nerve (i.c.n) is given off as usual to accompany the internal carotid artery into the skull. The mddle cervical ganglia are absent, and the inferior cervical ganglia (1.C.G) are smaller and narrower than those in Zamandua. And the cardiac nerves are similar to those in the latter. The Cardiac Plewuses (S.C.P and D.C.P) differ from those in Tamandua. The superficial one is formed entirely by the left vagus and sympathetic, and has a well-marked ganglion of Wrisberg (G.W). The deep plexus is’ composed of branches of the right vagus and sympathetic, and its solitary ganglion is applied to the wall of the right auricular appendix. The two plexuses are united by fine communicating branches. The thoracic cords contain few ganglia as in Vamandua. They increase in thickness from before backwards, and the splanchnic nerves (/) are given off from an enlargement. There is no branch comparable to the ascending aortic branch in Tamandua. The abdominal part has few ganglia, and its rami communicantes to the lumbar nerves (¢./.p) are well marked. The Celiac Plewus (text-fig. 4) consists of right and left halves connected by bundles of fibres, and united to the sympathetic cords by the splanchnic nerves (A). It lies between the layers of the common mesentery, along the course of the cceliac axis. 106 DR. C. F. SONNTAG ON THE VAGUS AND It differs from that in Zamandua in that it receives branches of the left vagus, whereas that of the latter receives the right vagus. Text-figure 4. f IV, L.Ve a./). 0.7). boo a, XQ LX | MSO a UW Fie S./77.p. cp. cat : : : : , ; I'he posterior thoracic and abdominal parts of the vagus and left sympathetic nerves in Huphractus villosus. «a.n.: aortic nerves; 0.2.: cesophageal nerves. Other letters in text. There is no division into upper and lower parts. The following are the branches connected to it :— A. To the right half of the plexus :— . Communicating to left vagus (c./.v). . Coronary plexus (q). Hepatic plexus (7). Splenic plexus (4). Right renal plexus (m). Ort Co be SYMPATHETIC NERVES OF 'THE EDENTATA. 107 B. To the left half of the plexus :— 1. Superior mesenteric plexus (s.m.p). 2. Aortic plexus (p). 3. Left renal plexus (0). 4, Phrenic plexus (/). 5. Communicating to left vagus (c.1.v). The arrangement differs, therefore, from that in Tamandua. Tatusia novemeincta (text-fig. 5). The Vagus Nerves:—Both ganglia nodosa are present, and communicating branches run to the superior cervical ganglion of the sympathetic and other cranial nerves in its vicinity ; and the branch to the spinal accessory is very prominent (¢.s.a). The pharyngeal branch (a) resembles that in Huphractus. No branches run to the sympathetic trunk. Text-figure 5. A. The vagus and sympathetic in Tatusia novemcineta. A. Upper cervical part; B: posterior thoracic and abdominal parts of the vagi. Description in text. The superior laryngeal nerve (b) forms a loop which commu- nicates with the sympathetic cord distal to the superior cervical ganglion; but no vascular nerves run upwards from it, as in Euphractus. No lingual branch of the vagus is present as in Tamandua. The Cardiac Plewuses have the usual mode of formation, but the superficial one has no Ganglion of Wrisberg. In the posterior part of the thorax the left vagus divides into two. One branch runs along the dorsal aspect of the cesophagus, communicates with the celiac plexus and ends on the dorsal gastric wall. The other branch runs along the ventral aspect of the esophagus, communicates with the right vagus and ends on 108 ON THE VAGUS NERVES OF THE EDENTATA. the ventral wall of the stomach. The right vagus runs along the lesser curvature of the stomach to the pyloric region. Both vagi send branches to the cesophagus. The Caliae Plewus resembles that in Huphractus in most points. Summary and Conclusions. 1. The course of the vagus nerves resembles that in the Marsupialia, but their thoracic relations are similar to those in Man. 2. The superior cervical sympathetic ganglia are present in the neck in Muphractus and Tatusia, but are represented by wide expansions within the foramen lacerum posticum in Zumandua. 3. In Tamandua the pharyngeal plexus is more complicated than that in Huphractus or Tatusia, and it is the only species in which the lingual branch of the vagus is present. 4. In Huphractus the nerves accompanying the branches of the common carotid artery are larger and more numerous than in the other genera. 5. No Kdentate examined by me has a middle cervical sympa- thetic ganglion, and the thoracic ganglia are few in number. 6. The cardiac plexuses are built on the same plan, but there are variations in detail in their communications and ganglia in the various species. 7. Hach species exhibits a different arrangement of the vagi in the posterior part of the thorax, and their intercon:munications are variable. 8. The solar plexus in Zamandua differs in its arrangements and distribution from that in Huphractus and Tatusia. It does not lie across the vertebral column, but is situated within the mesentery along the course of the celiac axis. 9. The splanchnic nerves are well-developed in all species, but do not all arise from ganglia. 10. There is no separate depressor nerve, and no external communication between the superior and recurrent laryngeal nerves as in Hyrax capensis and Man. Bibliography. 1. Sonnac, C. F.—Proc. Zool. Soc. London, 1921, pp. 572-575. 2. Sonnrac, C. F.—Proc. Zool. Soc. London, 1921, pp. 873-876. 3. CunnincHam, D. J.—Text-book of Anatomy, p. 655. ts ~) — = ae Stephenson. 1922. P.ZS. heeta. igoc Morphology of Indian Ol VITTY & SEABORNE, LTD., LONDON. ON INDIAN OLIGOCH ATA. 109 7. Contributions to the Morphology, Classification, and Zoogeography of Indian Oligocheta. By J. StepHEN- peagrapnay $ ) son, M.B., D.Se., F.Z.S., Lecturer in Zoology in the University of Edinburgh. | Received October 29, 1921: Read February 7, 1922.] (Plate I.* Text-figures 1-19.) ContTENTs. Page IV. On the diffuse production of Sexual Cells in a species of Chetogaster (fam. Naididz)..............0..0... . 109 V. On Drawida japonica (Mchlsn.), 2 Contribution to the Anatomy of the Moniligastride ....0....0.....ccccccceeeeeeeeee 119 VI. On the Relationships of the Genera of Moniligastride ; with some Considerations on the Origin of Terrestrial Oliwoahastape cents. ticscce es cc ke dentens saaonas meee eae ee eS 1V.—ON THE DIFFUSE PRODUCTION OF SEXUAL CELLS IN A SPECIES OF CH #£TOGASTER (FAM. NAIDID#). The Naidide are a family of fresh-water Oligocheta, of small size, in which the usual mode of reproduction is asexual, by fission. Sexual reproduction probably occurs, at times, in all; but in many, perhaps the majority, of species, and in some whole genera, it has not as yet been observed. Where it has been described (except in the genus Pristina, where the organs are two segments further back) the testes and male funnels are in segment v, the atria and male apertures, and also the ovaries, in segment vi. Except in Chetogaster, a single sperm-sac is constituted by a posterior bulging of septum 5/6, and an ovisac, which envelopes the sperm-sac, by a similar protrusion of 6/7; the sexual cells given off from the gonads enter the sacs, where they ripen, and ultimately again leave the sacs to be discharged. Chetogaster is a somewhat aberrant genus of the Naidide, which was separated as a distinct family (Chetogastride) by Vejdovsky. It agrees, however, with the other genera of the Naididz in the prevalence of asexual multiplication, and (so far as hitherto known) with the majority (all except Pristina) in the position of the gonads in segments v and vi. Thus Vejdovsky (6) figures and describes the testes and ovaries of C. diaphanus in these segments; while Beddard (1) says that “the sperma- thece, testes, &c., are in the same segments in both the‘ families’ Naidomorpha and Chetogastride.” There appear to be no detailed descriptions of the sexual organs in any species of later date than these monographs, except my own of C. orientalis (4). The species of Chetogaster which is commonest in Lahore in N. India is that just mentioned, which I described first (3) as C. pellucidus, but later (the specific name being preoccupied) called orientalis (4). In this second paper I gave a description * For explanation of Plate I., see page 132, 110 DR, J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, of the genital organs, but although I had a number of specimens, including one in a very early stage, I did not discover the testes (‘the testes seem to disappear early, and I could not distinguish them in a specimen which showed sperm-morule in all stages of development, but no other genital organs, male or female, except the ovaries in an early stage”). Since that date I have again met with numbers of sexual specimens of this species, and have always looked for the testes, but have never found them, even in the earliest stages of sexuality. On the contrary, careful examination of the living animal (which is very transparent), of stained specimens, and of sections, have united to convince me that the male cells are produced in various parts of the body; and not only so—their production is not limited to the parent animal of the chain, but is spread over the whole series of individuals, and occurs in the youngest members of the chain, which are scarcely recognizable as separate entities, as well as in the most advanced. The situations where the male cells arise are mainly as follows :— (1) On strands passing between the alimentary canal and parietes. The genus as a whole, and certainly this species, is remarkable for the incompleteness of the septa; in this species there is nowhere a definite diaphragm across the coelomic space ; and many of the strands, in various parts of the animal, from which some of the sexual cells arise, should no doubt be regarded as representing septa. (2) On the inner surface of the body-wall, often in the angle where one of these strands joins the parietes. (3) In the new cellular-tissue which is formed in abundance as a ring round the body at the site of the future divisions of the animal. This cellular tissue gives rise to a number of new segments; the region is hence called the budding zone, and the production of this embryonic tissue is the first sign of subsequent fission. I propose to select as an example for description one out of numerous observations, and then to add some account of the appearances in sections, concluding with a few general remarks. I may say that I have been particularly alive to one (I think the only) source of possible error—namely, that the testes might be quite inconspicuous masses in the normal position, that the sexual cells might be detached very early in their history, and the developing morule, wandering widely through the body, become attached to the body-wall or to strands in various regions, so simulating the production of morule in these situations. This would, @ priori, not be impossible, for tbe following reason :—There are, in this species, no sperm-sacs or ovisacs, and this seems to depend on the incompleteness of the septa, of certain of which the sacs, in other genera, are backward protrusions; the sexual products, mstead of being received into AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ATA. 111 sacs, must therefore ripen in the general body-cavity ; and the incompleteness of the septa allows them to become widely distributed, both through the parent animal and the posterior components of the chain. I believe, however, that the obserya- tions I have made, only a part of which I record below, show that this is not the case, and that the cells are really produced in all parts of the chain. Whether the ovaries also are similarly non-localized I cannot say. I have described the ovaries as definite organs—as cellular aggregates or as hyaline masses of nucleated protoplasm in which cell outlines were not visible, suspended in the ventral part of the body-cavity on fine strands in front of the level of the sete of segment vi (7. e., about the position where septum 5/6 would normally be). The ovaries of C. diaphanus are several times figured by Vejdovsky as definite spherical or pear-shaped masses: they had previously been described by Lankester, and were seen by Vejdovsky also, in C. limnei. Ova are, however, sometimes met with in the hinder animals of a chain of C. orien- talis; in their early stages, moreover, male and female cells are indistinguishable; and it is possible that in some of the examples which I assume below to be the early stages of the male products, the cells may be young ova. Text-figure 1. To illustrate sites of production of genital cells in a specimen of Chetogaster orientalis. The above figure represents in outline a specimen of Chetogaster orientalis, in which the process of fission has demarcated eight future individuals, including the anterior one which we may regard as the parent. The first constriction probably appeared between the four anterior and the four posterior components of the chain, and the first break in the chain would subsequently take place here; the fifth individual was at the time of observa- tion the most completely differentiated (after the most anterior or parent). The figures and their reference lines indicate the places where the sexual cells were visible. None were seen at any stage of development in the living animal in the front part of the parent, nor were gonads distin- euishable; subsequently, after staining, examination in cedar oil showed a small morula, of about four cells, ventrally by the side 112 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, of the nerve cord, almost at the level of the anterior border of the crop*. At 1, two small morula were visible, attached to strands in the dorsal region of the body-cavity, and moving with the strands. At 2, ventrally situated at the level of the hinder part of the cromaehy were a number of morule of large size, of which one or more were quite free. At 3 were a considerable number, apparently fixed to strands. At 4, in the next indi- vidual, were a large number of morule behind the developing pharynx and oesophagus, all seeming fixed to numerous strands in this region; this attachment seemed also gene beyond doubt on subsequent examination in cedar oil. At 5 were seen three morule on a level with the anterior part of the crop, on strands, to which they remained attached even on moderately violent movements of the animal under the coverslip; one or two others in this component of the chain were quite free. At 6 were two morules, also apparently attached to strands; and this was con- firmed by subsequent examination in cedar oil under the immer- sion lens. At 7 were a number, one showing commencing differentiation into central (granular) and peripheral (cellular) parts ; two at least of these, well seen, were obviously, one would say, fixed, and not merely entangled in the strands. In the above example it happens that all the morulz in course of development were situated on strands in the body-cavity; and such strands appear to be the commonest site of their pro- duction. JI have repeatedly seen the morulz on the strands keep their attachment in spite of the animal’s movements and conse- quent swaying of the strands to and fro—indeed, I have no note of ever seeing them shaken off. These strands are frequently situated in the anterior region of young individuals, towards the front end or about the middle of what will be the crop, sometimes behind its middle, or behind the stomach; as exemplified above, they may be either dorsal or ventral to the alimentary tube; not infrequently they are in the angle between the strand and the body- wall, and might be described with equal correctness as arising from the parietes. The greatest production of morule may not be in the parent animal—perhaps seldom is so. I have, for example, noted the production as being most active in the individual behind the primary fission zone, and in the one succeeding this, and as being small in the anterior half of the chain. It Will be Hondereeaod! that besides the fixed and not fully-developed morule, there are other larger ones which float freely in the body-cavity. A curious and perhaps pathological condition was met with in some specimens which had been in a tube in the laboratory for a * The alimentary canal in Chetogaster shows a considerable degree of differentia- tion ; the pharyux may be said, in this species, to occupy the first three segments, the esophagus the fourth, the crop the fifth to part of the seyenth, the stomach most of the seventh and the eighth, and the intestine (very short compared with the part of the tube so called in other Naididz) the few remaining segments. AND ZOOGEOGRAPHY OF INDIAN OLIGOCHAETA. US long time, and many of which must have died. Very numerous small morule were budding from the body-wall, to which they were all, at the stage examined, adherent. Some were also budding from the surface of the alimentary canal; and here and there some were seen in the angle between strands and parietes. There was no special production in the proper genital segments of the parent animal. Sections 7 thick, stained in Delafield’s hematoxylin and differentiated by a short immersion in iron-alum, dispel any doubts as to the actual formation of morule on the strands; they show that sexual cells may also be produced from the lining of the body-wall, as well as—which is interesting—from the newly- forming ring of cellular proliferation around the site of each future fission. They also allow the history of the developing morule to be traced. Text-figs. 2-6, all taken from longitudinal sections, will illus- trate the various situations of origin, and will supply the place of further description. Text-figure 2. Proliferation of sexual cells from body-wall, near a zone of budding in a posterior individual of a chain. X 1000. The early stages in the production of morule may be shortly summarized as follows :— (1) The sexual cells when first recognizable as such (text-fig. 2 possess large, round nuclei, close together and with a diameter of 7m or vather less; a protoplasmic covering is scarcely dis- tinguishable; in the periphery of the cell the chromatin is scattered, but (as seen in sections) it forms a ring near the middle, the centre of the ring being clear (text-fig. 2), or more Proc. Zoou. Soc.— 1922, No. VIII. , 8 114 DR, J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, Text-figure 3. Text-figure 4. e. A group of sexual cells forming on Formation of sexual cells from the cell- a strand just infront of a zone ‘proliferation of a budding zone. of budding in the hinder part The subsequent separation of the ofachain. X 900. individuals would have taken place between the two groups of pro- liferating cells. 700. Text-figure 5. Text-figure 6. + SOL. Young morula on a strand in the Older morula, still attached to a strand, body-cavity. > 850. traversing the ccelom, in neighbourhood of a zone of budding. X 4380. Reference letters for all the above text-figs (2-6) :— e. surface epithelium: g. sexual cells; m. muscle strands of body-wall; mor., morula; p. peritoneal cells; s. strands traversing ccelom ; 2. cell- proliferation of budding zone. AND ZOOGEOGRAPHY OF INDIAN OLIGOCH AIA. 115 usually (perhaps at a later stage) shows a smaller ring-like aggre- gate in the centre (text-figs. 3 & 4). ‘hese characters s distinguish the young sexual cells from the ordinary peritoneal cells. and from the cells of the proliferations in the budding zones; ‘in these the nuclei are on an average-4°3 to 5m in diameter, anid the chromatin has no character “istic arrangement. (2) The next stage, or young monte: ‘consists of an ovoid ageregate of cells, en appear to be Tittle else than naked nuclei (text-fig. 5); these are still of about the same” size, with the clear ore and central aggreeate of chromatin. (3) The cells become peripheral, leaving a granular mass in the middle of the morula; cell boundaries are not distinguishable, but the peripheral vegion of nuclei is marked off from the central eranular mass (text- fis. 6). The characteristic enone or the chromatin disappear S. (4) The morulw, attached up. to this stage, may now become free. General hemarks. Chetogaster orientalis appears to be the only Oligochete in which a diffuse production of sexual. cells, similar-to that of the Polycheeta, has hitherto been deseribed. It is possible, however, that the phenomenon may occur in other species of the genus, though the statements of previous authors (in particular Vej- dovsky, cf. ant.) seem at first sight to negative this. The testes ef C. diaphanus are figured by Vejdovsky (pl. v. fig. 4) as rather loose masses of cells some little distanee behind the septum, and are said in the text to arise on the first pair of nerves given off from the ventral nerve cord. ‘The cells ave loosely aggregated, and not bounded by a peritoneal membrane ; they grow rapidly to four times their original size, become free in the body-cavity, and then soon aitach themselves to the var 10us organs and go through their further development—for example, on the ventral nerve cord, or on the alimentary canal. The later stages again become free. The original testes quickly disappear. Tt will be noted that.the mass of cells which Vejdovsky calls the testes are not in the normal position of these organs; and it seems possible that: they are only two among the numerous groups of sexual cells which are, according to my observations on CO. orientalis, scattered about the body. Probably also, in view of the results of my sections of C. orientalis, the deyeloping g male. cells which Vejdovsky describes as having become attached to various organs really originated from their peritoneal covering, in the same way that they originate, according to the above descriptions, from strands, from the body- -wall, and from the proliferating tissue of the budding. ZONES. In the Polycheta the genital organs are as a rule not: strictly localized ; the gonads are proliferations of peritoneal cells which originate on the parietes, on the blood-vessels, or on the alimentary 8 ** 116 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, canal throughout a considerable region of the body. In the Oligocheta, on the contrary, this proliferation is restricted, and the production of sexual cells is confined, in general, to one or two testicular segments and a single ovarian segment in the anterior part of the body; the gonads, moreover, arise from a narrowly circumscribed area on the posterior face of the septa. Tt is probable that the Oligochzta have originated from some group of Polycheta; the evolution of the gonads has therefore been from a non - localized to a more definitely localized condition. in Chetogaster orientalis the gonads (the male, at least) have again lost their localized character; they are no longer definitely circumscribed proliferations of peritoneal cells in a constant position. That this condition, which is not unlike that of the Polycheta, represents a regression, and not the persistence of a primitive state, may be taken as certain ; Chetogaster is the most modified genus of the Naididee—of the numerous characters in which it differs from the rest of the family, not one can be thought of as primitive; and I believe the Naidide to be themselves far from primitive among the Oligocheta. We may for the sake of clearness express the facts in a somewhat different way. In certain animals the first visible differentiation in development is that of the germ-cells; in Ascaris, for example, as is well known, the peculiarities of their chromatin constitution enable them to be distinguished from the tirst. And it is possible that a similar ‘‘germ-track” exists in all animals, but that we are unable, with our present means of investigation, to distinguish it in all—at least in its earlier stages. It is possible, in other words, that in all animals the germinal material (germ-plasm) is set apart from the first in certain cells and their descendants, and that in the shaping of the organism these find their way, sometimes by fairly extensive wanderings, to their station in the gonads, or rather to their particular station in the ccelomic lining membrane (in the Triploblastica) where, by their proliferation, they give rise to the gonads. In the Polycheeta their distribution is wide; in the Oligocheta extremely circumscribed ; in Chetogaster orien talis it has again become wide. With what pec elay of structure or habit, or with what feature of the past history of Cheetogaster can this change be correlated ? I can think of only one that is at all likely, and even here I do not think it is easy to suggest how the change has been brought about. I have pr eviously suggested (5) that Chetogaster is descended from parasitic ancestors. I do not wish to repeat the argument at length; I will only say that Chetogaster, alone of the Naidide, never shows the phenomenon of an ascending current of water beginning at the anus and impelled up the. intestine by anti- per Tistalsis and ascending ciliary action ; and that the loss of cilia and of the ciliary function of the intestine is presumably due AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ETA. ARG (using the principle which has in recent years been illustrated by Dollo and Abel) to a different habit of life, which has left its mark on the race; that this different habit must have been a life in some other medium than that of water—i.e., it must have been either terrestrial ov parasitic; that the first of these is entirely inadmissible, and we are thus thrown back on the second. Also that a number of the other peculiar features of the genus, of which there are many, are explicable in a similar way; and finally that Chetogaster at the present day is still in soine cases an ecto- or even an endoparasite (in the pulmonary chamber and liver of fresh-water Gastropods), while in others it is a commensal; and in still others, though free-living, it (alone among the Naidide, so far as I know) feeds predominantly on animal food *. Exactly how or why parasitism should have produced this change in the distribution of the germ-cells, it is not easy to say. Parasitism, of course, places a larger supply of material at the- disposal of the reproductive activities; but this would not necessarily lead to a different distribution—to a widening or scattering of the germ-track. J can only suggest that there is possibly some connection between past and present habits of nutrition of Chetogaster orientalis and a change—a scattering— of the germ-track. The affinities of the Hirudinea with the Oligocheta have been recognized for some time (compare, for example, the opening paragraph of the chapter on Leeches, by Beddard, in the ‘Cambridge Natural History,’ of date 1896), though text-books still often separate the two groups widely. Thus in the arrangement adopted by Parker and Haswell the Gephyrea and Archiannelida intervene between the Chetopoda and Hirudinea. Michaelsen (2) has recently set forth the evidences for a close relationship between the Oligocheta and the Leeches in detail. The argument rests on the interpretation of certain structures, in Sudanese and Sumatran species of leeches, as spermatheca which communicate with the alimentary canal,—comparable there- fore with those of many Enchytreids; on a consideration of the Branchiobdellide and of Acanthobdella as intermediate forms (though not necessarily in the direct line of descent of the Leeches); and on the demonstration that there are no pecu- liarities of the organization of Leeches that are not found, or at least foreshadowed, in certain Oligocheta. Detaching the Oligochta from their association with the Polycheta in the * T may mention an additional observation, made since the date of my former paper. I met with a specimen of Chetogaster orientalis endeavouring to consume a whole Nais—a worm about as big as itself ; the end of the Nais had: entered the pharynx of the Chetogaster, and I saw it drawn on into the region of the crop; the | part of the Nais outside the Chetogaster was wriggling. On transferring the two animals to a slide they unfortunately became separated—the Nais without its head. The Chetogaster would not take in the Nais again, though I put them in contact several times. } Rotifers also are a very common food of Chetogaster orientalis. 118 =r. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, Class Ghetopeds, Michaelsen unites them with the Hirudinea in a new class, Cliteilata. ‘I consider the Hirudinea,” says he, “trankly as Lumbriculide which have undergone special modifi-. cations in adaptatton to a carnivorous habit of life.” It might be justifiable, He thinks, to include the Leeches as a family, of Oligocheta, though as we have seen he does not actually do this. The charatter of the Leeches which is of interest from the: point of view of the present paper is the large number of testes. Michaelsen has no difficulty in showing that, as in the case of other peculiarities, this too is foreshadowed in the Oligocheeta ; for example, Lamprodrilus satyriscus, a Lumbriculid, has either three or four pairs of testes. The position of the testes behind the ovaries, however, offers greater difficulties; Michaelsen does’ not wish to assume a sudden dislocation backwards of the testes, in such a manter that they “hop over” the ovaries. A trans- lation of the.testes backwards can, he thinks, only take place through the intermediation of a stage in which the series of gonads are hermaphrodite. Hermaphrodite gonads have often been observed; for example, in specimens which show an abnor- mally large number of genital glands (“supernumerary gonads ” —such series may extend as far back as segm. xviii), the anterior are often testes;the middle ones: hermaphrodite, and the posterior Ovaries; though Beddard in a Uvocheta (= Pontoscolea corethrurus) found the hindmost gonads to be testes. It seenis to me, honmere, that such a condition as that described above in Cheetogaster orientalis furnishes, perhaps, a more likely origin for the multiple testes of the Hirudinea. On such an_ assumption there would be no necessity to invoke a series of hermaphrodite. supernumerary gonads—a double abnormality where it occurs to-day—as a transitional stage. In Cheetogaster orientalis male cells are produced abundantly, in every cau individual, in the segments behind the ovaries; a stricter locali-_ ation aranelie result in segmentally arranged Hegban. And the, ae cause Which we can surmise to have been active in the case. of Chetogaster, orientalis—the adoption of carnivorous or para-' sitic habits—must also have existed in the ancestors of the’ Leeches. References to Literature. 1. Bepparp, F. E.—A Monograph of the Order of Oligochxta. Oxford, 1895. 2. Micwartsen, W.—Uber die Beziehungen der Hirudineen zu den Oligochiten.- Mitth. Zool. Mus. Hamburg, vol. xxxvi. 1919, p. 131. 3. STEPHENSON, J.—Descriptions of two fresh-water Oligochete worms from the Punjab.’ Rec. Ind. Mus.,'vol. i. 1907. 4. , ‘Studies on the aquatic Onvecneel of the eons Rec. Ind. ° i Mus., vol. v. 1910. 5. % On Intestinal Respiration in Annelids : with Considerations on the Origin and Evolution of the Vascular System in that Group. Trans. Roy. Soc. Edin., vol. xlix. pt. iii. 1913. 6. Vrspovsxy, ¥.—System und Morphologie der Oligochiaten. Prag, 1884. AND ZOOGEOGRAPHY OF INDIAN OLIGOCHATA. 119 V.—On DrawipdA s4Ponrca (Mchlsn.); A ContRIBUTION TO THE ANATOMY OF THE MoniLicasTRip&. The Moniligastride are a family of Oligocheta with a rather restricted range. Hxcept for the recently discovered Syngeno- drilus, which differs very considerably from all other members of the family, they are endemic only in India’(mainly the South), Bur ma, and ee islands of the Malay Archipelago; a few species of the largest and best known genus, Drawida, however, have been able to spread more widely, : and have been found in Japan, China, and certain islands cf the Pacific. The writings of Beddard, Benham, Bourne, Rosa, and Michaelsen have sitet dated the anatomy of the several genera, and in a number of cases, by means of sections, the finer structure also. Since, however, the worms occur for the most part in localities remote from centres of population, writers on the group have as a rule been dependent for their material on collections made by others, without, it is to be supposed, any special methods of fixation. While such material is adequate for systematic work (for which it has generally been used), it is often quite hopeless as a basis for more minute investigations. The only students of the group who have had the opportunity of fixing living material for themselves appear to be Beddard (7), who received specimens of Drawida bahamensis alive from Kew Gardens, and Bourne (11), who colleeted his own material on the Nilgiris in Southern India. These authors do not, however, indicate the methods they employed for fixation. A few years ago one of my pupils, Mr. G. 8. Thapar, M.8c., now Professor at the Canning College,. Lucknow, brought back with him to Lahore from Murree in the W. Himalayas a considerable number of living specimens of Drawida japonicus, one of the per egrine species of the genus. These were kept for some days 1n the laboratory in scraps of moist blotting paper, and the alimentary canal having been freed in this way from earth, they were, after Harcotization: fixed in Zenker’s solution. Some specimens were dissected, . and others sectioned, both transversely and longitudinally, and the sections stained with Delafield’s hematoxylin followed by eosin. The chief attention has been devoted to the nephridia and the genital organs. — Of the nephridia‘in the genus Drawida, or in the family, we possess nothing that can be called.a detailed deser iption. The general : anatomy of the reproductive organs in the genus is, however, well known, and may be summarized as follows : — The testes are one pair, enclosed each with its funnel in paired testis sacs, which have a peculiar position ; they are suspended by septum 9/10, fairly high up in the body- cavity, projecting sometimes more on the poster ior, sometimes more on the anterior, side, sometimes equally on both sides of the septum. ‘The vas deferens leaves the sac, and descends, with many twists and coils, 120 DR. J. StEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, on one or other or both sides of the septum; on reaching the ventral body-wall it runs backwards for a short distance and joins the prostate, a glandular mass in segm. x of cylindrical or rounded shape, with a central lumen, which discharges to the exterior in the groove between segms. x and Xi. The ovaries are on the posterior face oe septum 10/ ‘11, the funnels on the anterior face of septum 11/12; the short oviducts reach the surface in groove 11/12. A large pair of ovisacs— backward pouchings of septum 11/12—extend backwards through, as a rule, several segments. The spermathece are ovoid sacs attached to the hinder face of septum 7/8; a winding duct leads down the septum to the body-wall, and ends in the corresponding groove; just before it enters the body-wail there is often a muscular sac, the spermathecal atrium, of moderate size, or small, or it may be even minute, connected with it. In the account which follows of the minuter structure of the various organs, references will be made where necessary to the previous work of Beddard (2, 3, 6, 7), Bourne (11), and Benham (10), who used serial sections in their investigations, and to that of Michaelsen, whose systematic papers (13, as, 16) contain the record of important anatomical details. (1) Zhe Nephridia. The nephridia (Pl. I. fig. 1) are attached to the posterior faces of the septa; in the middle of the body each has a total length of about 1 mm. Each is divisible on inspection into four portions :—(i.) The anteseptal portion with the nephrostome; (i.) the twisted nephridial tube; (iii.) larger than the last portion and more dorsally situated, an oval and vertically elongated sac, occupying the mid-lateral region as it lies on the body-wall in the dissection ; (iv.) projecting from the lower end of the sac is a stiff, straight terminal duct. The nephrostome (7) is 45—50 pw in Hinmiciee and has a marginal rosette of about 20 ciliated cells. From the nephrostome leads the nephrostomial canal (7.c.), 8-10 » in diameter, which pierces the septum, and after a straight course of about 170 « becomes the spiral coil. The portion of the organ which was roughly distinguished above as the twisted nephridial tube can be divided on closer inspection into three parts, which may be named the spiral coil, the twisted loop, and the spindle-shaped portion. The diameter of the tube throughout this part of its course is 40-45 p ; it is of a slightly brownish colour. The lumen is wide, but intracellular —bored through the middle of a single series of cells (drain- pipe arrangement). ‘The appearance of both longitudinal, transverse, and optical sections is at first sight misleading; the drain-pipe series of cells appears to be free within the lumen of a larger AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ETA‘ 121 tube. This larger tube, however, is easily recognized as the peri- toneal covering ; it consists of separate cells, shortly columnar in shape, with the nuclei peripherally situated. The spiral coil (coil) lies more ventrally than the rest of the organ ; it is altogether ventral to the lower end of the oval sac. 'The spiral consists of about two turns; its apex is below, so that the nephridial tube followed from the nephrostome first descends to the apex, and then ascends as the returning limb of the spiral. On quitting the spiral coil it becomes the twisted loop, a dorsal continuation of the coil, which lies by the side of the lower por- tion of the oval sac. Thus roughly half the vertical extent of the twisted nephridial tube consists of the spiral coil which is below the level of the oval sac, and half of the twisted loop, situated alongside the lower part of the sac; the total extent of the twisted nephridial tube is.about ‘55 mm. The twisted loop is retained by a mesentery which joins that of the oval sac. The twisted loop (Joop) contracts to 12 » in diameter at its end, and joins the oval sac by the intervention of a spindle-shaped portion (spin.). This part of the organ is 80-100 » in length and 40 w in greatest thickness; it is perfectly circular in transverse section, has thick (7) walls, and a lumen which may be widely patent, or may be almost blocked by faintly pink-staining nucleated cells. The oval sac (sac) is much elongated dorso-ventrally, gently curved round the alimentary canal, and somewhat flattened ; it is the largest and most conspicuous part of the whole apparatus, being °65-"73 mm. in length and -17 mm. or more in width (in- cluding its peritoneal covering). It has a very bulky coat of high, clear peritoneal cells (p.), 40 ~, 60 » or more in height; in one case the peritoneal coat formed a layer of cells thicker (-18 mm.) than the whole of the included sac. ‘The nuclei of the peritoneal cells are at various levels. The muscular coat (m.) of the sac is remarkably thick—12 p~— and consists principally of circular fibres. Inside the muscular coat is a thin but very obvious connective-tissue coat, staining an intense blue with hematoxylin, about 1» thick. The epithelial lining (ep.) consists of lightly staining low columnar or cubical cells; its height is about 12 ~—-about as thick as the muscular coat. A mesentery, about °32 mm. in width, passes from one border of the sac to the septum; it consists of a layer of blue-staining connective tissue, continuous with that of the wall of the sac and with that of the septum, covered on both faces by peritoneal cells. The spindle-shaped portion is similar in its structure and its mesenteric relations to the sae. The terminal duct (d.) has a length of *3-—33 mm. in its course from the ventral end of the oval sac to the body-wall, and its diameter is 40 u. It comes off just to one side of the ventral end 222 DR J, STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, of the oval sac, is straight and stiff, and- makes a right angle or rather less with the general direction of the sac (except where the duct has the lower opening, v. if.). In whole preparations of the nephridium the ectal end, where it is broken off from the body-wall, always appears trumpet- shaped, the end of the trumpet being oblique and the dorsal lip the more prominent (é., fig. 1). The duct is covered by a layer of cubical peritoneal cells; the remaining layers are the same as those of the oval sac, but all are thinner; the lining epithelium consists of approximately cubical slightly staining cells. The duct is attached by a narrow mesentery to the posterior face of the septum. Reaching the angle between the septum and the parietes 16 cilates somewhat, the dilatation corresponding to the trumpet-shaped ending of the duct as seen in the isolated nephridium (v. sup.). The duct may reach the hady- wall in one: e two situations— either at the level of the lateral or of the ventral sete; in the first case it makes an angle rather less than a right angle with the axis of the oval sae, Sanita in the second it appears as a direct downward continuation of the sac. At both these levels the longitudinal muscular coat shows an interruption in transverse sections ;- the . trumpet- shaped dilatation rests ther efore on the circular muséular coat. -If the duct has the lower position, it penetrates the parietes forthwith; the blue-staining layer becomes continuous with the similar layer of the body-wall, and the epithelium of the duct with the superficial epithelium of the body. The same may happen if the duct reaches the body-wall at the level of the lateral setee—the duct may penetrate the parietes immediately, in a slightly upward direction. Often, however, the duct continues in the body-wall in an upward direction, and reaches a level not very far from the mid-dorsal line—a level corresponding approximately on the dorsal to that of the ventral sete on the ventral side—before it comes to the surface. In the latter case the muscular layer of the duct ceases to be distinguishable; the blue-staining layer becomes continuous with a very marked similarly staining layer between the circular and iovnetntanoel muscular coats of the body-wall; and the epithelial layer of the duct is only distinct at the beginning and end of its course within the wall. For the- greater. part of this course the duct appears rather like a split in the_body-wall immediately internal to the circular. muscular layer; the epithelial layer is difficult to make out (except at the tyo ends), and the lumen of the duct appears in most parts to be ‘bounded by a very definite wall of the blue-staining connective tissue. 1 give below the positions of the endings of Ab aerial ducts in a consecutive series of eleven segments, both on the right and left sides. . It will be noted that there are no apertures in the lowest position (on the level of the ventral sete) on the left side and none in the dorsal position on the right side. AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ATA. 123 (Upp.=uppermost position, not far from mid-dorsal line; mid.= middle position, at level of lateral setee; low.=lowest position, at level of ventral sete.) Right. | Left. maids, | | upp. low.7 mid, mid, upp. low. upp. mid, upp. mid. mid. , low. upp. mid. mid. mid. ‘mid. mid. upp. mid, upp. (2) The Testis Sacs (Sperm-sacs). Beddard describes the sperm-sacs of D. barwelli (6). He. makes the interesting suggestion contained in the following extract (postscript to 6):—‘‘ The remarkable partial obliteration of a segment (the xilith) which Michaelsen has recently described in Nemertodrilus griseus, suggests that something of the same kind may have occurred in the Moniligastride, the supposed. sperm-sacs may be all that 1s left of the ccelom belonging to the segment which contains the testes. This is, of course, no more: than a suggestion; but the varying position of the essential organs in the Oligocheta requires, as I point out in a forthcoming number of the ‘ Quarterly Journal of Microscopical Science,’ some possibility of the intercalation or excalation of segments at the head end.” The sacs are also described by Benham in DP. indi- cus (10), where the eavity is traversed ky muscular fibres passing in different directions ; and by Bourne in D. grandis (11), As in all species of the genus Drawida, the testis sacs, containing the testes, male funnels, and ripening spermatozoa, are in the present species a single pair, suspended by septum 910. In some species of the genus they, project about equally’ into segms. ix and x, in some they project forwards into ix more than backwards into'x; often there are differences on the two sides— individual differences, not specific. In the present species the sacs project backwards more than forwards. The appearance is somewhat as if the septum had split, the sacs being cavities within it (fig. 2). | aaa ‘Both anterior and posterior walis of the sac have the usual peritoneal (p.), connective-tissue, and muscular (m.) components. The inner lining of the sac is in places a cubical or flattened epithelium (ep.) ; over the greater part of the wail, however, the lining is not distinguishable from the mass of cells which fills the cavity. This mass stains of a pinkish colour with eosin (g.), cell 124 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, outlines are not visible in it, and there are numerous lightly staining nuclei (n.); there are also many yellow granules resembling chloragogen, often aggregated into considerable masses (a’.). Blood-vessels (0.v.) traverse the cavity; in some eases the chloragogen-like granules are in—and indeed block— the blood-vessels. The male funnel is a region of high columnar epithelium (/-) surrounding the exit of the vas deferens from the sac; the cells are exceptionally high near the opening of the vas, and become lower by degrees at the periphery of the patch; their nuclei are much elongated, and are placed at or above the middle of the height of the cells. The vas deferens (v. def.) after emerging is attached in numerous closely-set short loops on the posterior side of the septum. The testis (¢.) is a cellular proliferation just anterior to the opening of the vas deferens; it is thus implanted actually on the funnel. Its attachment is narrow, but the cell-mass spreads out considerably inside the sac. Near its attachment the nuclei are small, and stain deeply ; in the middle portion of the organ the nuclei are closely packed, with the chromatin in discrete rounded particles ; more peripherally, cell outlines, which are wanting in the more central (or proximal) parts of the testis, appear ; and lastly, discrete cells are detached from the surface of the organ. There were no ripe, or even ripening, spermatozoa in the sacs (the Moniligastride are apparently characterized by a seasonal sexual maturity, like the Limicole, and unlike most earthworms; the clitellum, for example, is often absent or very poorly marked) ; moreover, it is pretty evident that the pinkly staining mass, with small lightly staining nuclei, which fills out the greater part of the sac, is not directly concerned in the formation of spermatozoa. (3) The Prostates. There already exist a number of observations on the structure of the prostates in the genus Drawida, but these are not concor- dant throughout. The lining epithelium, the layer of muscular - fibres, and the “ gland cells” are distinguished in all descriptions; but authors are not agreed as to whether the gland cells are to be referred to the peritoneal or to the epithelial layer. Beddard, in a first communication on D. barwelli (2), states that outside the layer of muscular fibres there is a layer of pecu- liarly modified peritoneal cells. In a later account (3) this glandular tissue is described as consisting of large granular cells separated into groups by partitions; each cell is prolonged into a fine process, which extends at least as far as the muscular wall; indeed, says the author, it is difficult to believe that the cells do not in some way or other reach the lumen of the atrium and there discharge their secretion. Nevertheless, this layer of gland cells is the modified peritoneal layer, and there is no peritoneal coat outside it. AND ZOOGEOGRAPAY OF INDIAN OLIGOCH ILA. 125 In a third paper (6) Beddard refers toa recent paper of Rosa’s on Desmogaster doric, ~ member of another genus of Monili- gastride (19); there is here the same annular layer of muscular fibres outside the lining epithelium, and in addition muscular fibres are interspersed between the bundles of glandular cells ; a delicate peritoneal investment surrounds the whole. Rosa, however, states that the gland cells empty into the lumen of the prostate by long ducts, and the figure also indicates that these ducts—apparently prolongations of the cells—pierce the layer of circular muscular fibres ; nevertheless, Beddard thinks it 1s still possible to refer all that lies outside of the lining epithelium to the peritoneum (using “ peritoneum” in a wide sense, Beddard speaks of the muscular layer also as having been developed from the peritoneum). Beddard also gives an account of the prostate of D. bahamensis (7). The gland-cells le in groups in the interstices of a some- what loosely arranged muscular mass; the ducts of the cells— simply filiform prolongations of their substance—pierce the muscular layer in bundles, and evidently pour their secretion into the Jumen. ‘There is no peritoneal layer outside. Benham examined MJoniligaster indicus (=Drawida robusta) (10); the specimens were badly preserved. The necks of the glandular cells pass through the muscular coat nearly up to the epithelium, but not quite. The epithelium is traversed by narrow, clear tubes resembling ducts, which differ from the epithelial cells in having no inner boundary. There is a ccelomic epithelium outside; hence the gland cells are not peritoneal, and must belong to the layer.of epithelial cells. Wise Bourne, in D. grandis (11), found the gland cells arranged in groups, each cell, nowever, sending its duct to take its place among the columnar cells of the epithelial lining. There are two layers of muscle—one overlying the epithelial layer and within the glandular layer, and another outside the glandular layer; a layer. of coelomic epithelium invests the whole. There is thus an extraordinary variety in the accounts of this organ. The layer of gland cells is said to be separate from the epithelial layer, and to send prolongations as ducts through the epithelium to the lumen; to have outside it a peritoneal invest- ment, and to have no such layer; and to belong to the epithelial layer, or to the peritoneal layer. In the present species there seems to be no doubt that there is a peritoneal layer outside the layer of glandular cells; but equally definitely these latter seem to be altogether cut off from the epithelial layer and the lumen of the organ. JI am disposed to believe that the cells belong to the peritoneum rather than to the lining epithelium; and to agree with Beddard, who would hold that the presence of a flattened peritoneal layer outside the glandular cells does not preclude the peritoneal origin of these latter. In the present species the prostates (fig. 3) are situated in 126 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, seem. x. Hach is a short, thick cylinder, with a soft, white lobulated surface and a central canal opening on the surface in groove 10/11. The outermost coat is’ by far the most bulky; it is composed of large cells (prost.) with distinct outlines, the cytoplasm of which stains moderately with hematoxylin. The cell substance isin general granular, the nuclei (mwe.) are small, homogeneous, of various shapes, often peripherally situated, and do not stain much more deeply than the cytoplasm. In one specimen there are numerous vacuoles (#.) in the cells, each with a small, round, homogeneous, lightly staining mass in the centre. In thickness this ‘layer is 120 » or more. The cells do not ierce the muscular and connective-tissue layers to join the epithelial lining of the prostatic lumen; some cells are sessile on the muscular layer, others have their rounded inner ends slightly separated from it (ef. fig. 3). There can, I think, be no real doubt that the layer of large cells is not a bursting of the lining epithelium through the muscular coat. Outside this layer there is often to be seen a peritoneal coat as a very thin blue-staining membrane—a mere line—with here and there more or less flattened nuclei, often ‘apparently degenerate. ‘In some cases the membrane spans the angle between the outer ends of neighbouring cells. Sometimes a few muscular fibres may be seen outside the layer of large cells. The peritoneal layer may also carry blood-vessels of considerable size, some branches of which penetrate the layer of large cells. From what are the cells of this layer derived? The presence of a peritoneal layer surrounding them would seem to preclude the possibility of their being themselves peritoneal in origin, as I at first supposed; while the fact that they are totally cut off from the lining epithelium—separated from it by the museular layer—seems to shut out the possibility of referring them to the epithelial layer. I believe that they are peritoneal in origin, however; the fact that they are covered in places by a flattened peritoneal epi- thelium does not necessarily negative this view. A peritoneal membrane persists over the layer of large cells in the same way that such a membrane persists over the lobes of the testes in species of Lumbricillus: the testes are proliferations of the celomic epithelium, which in this genus (as in Ocnerodrilus) ave covered in at first by a delicate peritoneal membrane. (4) The Vas Deferens. The vas deferens descends from the testis sac to the ventral body-wall, and in doing so forms a number of coils on both sides of septum 9/10. In this part of its course its diameter is about 40»; its lining epithelium consists of cells staining pink with eosin, which almost occlude the lumen and whose nuclei are situated below the middle of their height; it possesses also a AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ETA. 127 well-marked connective-tissue (blue-staining) coat, 2°5 yu thick and rather loose in texture. Having reached the parietes it runs backwards for a short distance, and then, sinking into the body-wall, lies in the blue- staining connective tissue between the two muscular layers ; here its diameter is 30 ,, and its connective-tissue coat is still moderately thick. Arrived at the base of the prostate, it enters this organ at its lower end, and runs upward outside the muscular coat of the prostate and within the layer of large cells, in a direction parallel to the prostatic lumen; its diameter is now 20, and it still possesses a deeply staining connective- tissue layer outside its. epithelium. Sinking “through the muscular layer ot the prostate, the rest of its course lies. within that coat, in the lining epithelium (fig. 3, v. def.); it finally opens into the. prostatic lumen at the ental extremity of the latter. In this‘last part of its course it loses its blue- -staining layer and becomes a small, round tube of cubical cells, 18 u in diameter. The actual relations ‘of vas and prostate, are thus disguised, inasmuch as the vas, entering the prostate at the base ‘of the latter, and ascending ath its wall to the apex, is hidden from view during this part of its course. In a species of Drawida (D. raui), eich! I have recently described (26), and which is in respect of its male apparatus the most primitive of the genus (it possesses two pairs of prostates, an ancestral character), “ the vas deferens, joining the prostate below, can be seen running up its surface towards the free upper end. Dann Moniligaster deshian yest the prostates are very large and sausage-shaped, and extend back through several segments ;° the vas deferens passes back along it to fuse with ‘it some little distance from its ental end; and presumably sections would show that here, too, the mae is a separate tube as far as the ental end of the prostate, where the one presumably passes into the other; in two varieties of this species the vas can be seen in an ordinary dissection to enter the ental end of the prostate. In certain species of the genus Drawida, therefore, and more obviously in the genus Monzlig iguster, the prostate i is to be regarded as the thickenéd « continuation of the vas deferens surroundéd: by a. bulky mass of peritoneal cells. These cells give the organ thé soft and papillose surface which frequently characterizes Goseoln other species of Drawida, however, the surface of the prostates is firm and shining—an appearance which is associated with a firm muscular investment; cither there is here no soft covering of swollen peritoneal cells, or the layer is so intermixed with and covered by muscular fibres as to give a firm and resistant surface. (5) The Ovarian Chamber and its Contents. The first indication that there is anything peculiar in the septa which come into relation with the female organs in the 128 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, genus Drawida occurs, perhaps, in Benham’s description of Moniligaster indicus (= Drawida robusta Bourne), where it is said (10) that septum 10/11 is thrown back a whole somite, and is confluent with 11/12 along a certain part of its course laterally, but becomes separate towards the middle of the body (7. e., pro- bably, towards the mid-dorsal and mid-ventral lines). This is, however, not the condition of the ‘ovarian chamber” which I wish to describe. Bourne (11) describes an arrangement of the septa which results in the cavity of segment x1 being reduced to two sacs ; one of these contains the ovaries and nephridia of the segment, and into it the oviducts open internally ; the other sac contains the portions of the dorsal vessel, alimentary canal, and ventral vessel which belong properly to the segment. ‘Ihe description is not clear, nor does the figure solve the difficulties. Michaelsen (18), describing D. willsi and D, nepalensis, speaks (in the latter case with doubt) of an ovarian chamber formed by septa 10/11 and 11/12, and a fine connecting membrane; the chamber is thus a separated-off portion of segment xi. In a subsequent paper (14) he seems inclined to recognize the ovarian chamber as itself the xith segment in D. travancorensis, mattha, and ghatensis. Cognetti states (11 a) that D. fakir has the ovaries enclosed in a thin-walled peri-cesophageal capsule. I have myself recorded the relations of the female organs in a number of species (21, 22, 23, 24, 25), and have found that the condition is not the same in all. Some species have nothing note- worthy in the arrangement of the septa in this region (D. pellu- cidus var. stewarti, D. kanarensis). The complete ovarian chamber, found in D. rotungana, kempi, robusta, rangamatiana, nepalensis, ghatensis, decourcyt, barwellt var. impertusus, and in Moniligaster deshayesi, is produced by the two septa which bound the somite containing the ovaries and funnels coming together and fusing, except over a horseshoe-shaped space which spans the alimentary canal above ; the canal passes through an opening in a membrane which represents the fused septa 10/11 and 11/12, and a needle can be passed above it and below the ovarian cham- ber without doing any injury to the chamber or any organ. In a few species (D. hodgarti, afinis, papillifer, chalakudiana) the septa concerned in the formation of the chamber are not fused in the dorsal part of their extent; the consequence is that while the chamber is formed below in the manner just explained, and spans the alimentary canal as’ described, it reaches and is bounded by the dorsal parietes above. In two species (D. brunnea and parambikulamana) the septa meet above exactly at the dorsal parietes, so that the chamber is just and only just excluded from the dorsal body-wall. As shown by sections, the condition (in this worm at least) is somewhat more complicated than was imagined, on the basis of dissections, by either Michaelsen or myself. In the present species (fig..4) there is dorsally in place of AND ZOOGEOGRAPHY OF INDIAN OLIGOCHAETA, . 129 septa 10/11 and 11/12 only a single septum, which is to be regarded as a product of the fusion of the two. More ventrally these separate, so that in a longitudinal section taken to one side of the alimentary canal, segm. xi appears as a wedge-like space with its apex upwards and base at the pavietes. This wedge- like space is continuous above the alimentary canal with the similar space on the other side. There are, however, other complications in the form of this ehamber which are difficult to describe in a few words. If the above were a complete description, it is obvious that segm. xi would be an ordinary segment, except that its bounding septa would be fused dorsally; and it would contain a section of the alimentary tube, dorsal and ventral vessels, and ventral nerve cord. What actually happens may be described by saying that (a) throughout the whole of the gizzard and reproductive regions of the worm the septa are incomplete in their mid-ventral por- tions—around the ventral vessel and ventral nerve cord—as if a series of arches were cut out of them; (6) that all round the alimentary canal, between septa 10/11 and 11/12, a ring-shaped partition isolates an annular ccelomic space (figs. 4 & 6, ann.) round the tube; this annular cavity contains also the dorsal vessel (d.v.), but the ventral vessel is below it; (c) longitudinal and almost vertical partitions are also established, extending in the ventro-lateral region between septa 10/11 and 11/12, and thus isolating the mid-ventral portion of the coelomic cavity here (fig. 6, v.ch.). The cavity of segm. xi is thus altogether absent dorsally ; and what exists of 1t is divided up as follows:—(i.) An annular space immediately surrounding the alimentary canal and containing also the dorsal vessel ; (i1.) a larger space, the ovarian chamber (figs. 4 & 6, o.c.), surrounding this in turn, of which (111.) the mid-ventral portion is isolated by longitudinal partitions on each side of the ventral vessel and ventral nerve cord, but is continuous in front and behind with the ccelomic cavities of neighbouring segments *. The ovarian chamber contains the ovaries with their funnels, and the nephridia. The ovisae (fig. 4, sac) is an evagination of its posterior wall. The ovaries are proliferations of the epithelium covering the anterior wall of the chamber. The funnels (/) are constituted by modifications of the epithelium of the posterior wall. The walls of the chamber have the usual constitution of septa—peritoneum on the outside, muscular and connective tissue composing their substance. The inside of the chamber is * There are other peculiarities of the septa in the neighbouring regions which are difficult to make out. Thus the floor of the annular chamber round the gut in segm. xi is continued forward and backward as a sheet of tissue which extends from behind the gizzard region to the spermatiecal segment. This sheet may have free lateral margins, or the lateral margins may be bent upwards to join the ventro-lateral wail of the cesophagus. Proc. Zoou. Soc.—1922, No. 1X. 9 130 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, lined by a layer of cubical epithelium, modified as follows in certain regions : (1) The funnel—a long, rather curved groove on the posterior wall, extending from the roof of the chamber downwards to the beginning of the oviduct. The cells composing it are high and narrow, with much elongated nuclei situated above the middle of the height of the cells; the cytoplasm is granular; the free surface is bordered by a layer of rodlets (‘ Stiibchen ”’). (2) The ovary, on the anterior wall, consisting of a number of finger-like lobes attached along an oblique area on each side of the alimentary canal; the upper portion of this area is quite narrow, but it evadually widens below. (3) Various “proliferations of small cells, with deeply staining nuclei; these constitute somewhat indefinite heaps which do not seem to be on the way to develop into ova, since the characters of the cells are the same at the periphery of the mass as deeply in its substance. These masses occur on both anterior and pos- terior walls (one is cut dorsal to the gut on the posterior wall of the chamber in the figure), and also at various places within the ovisacs. In another specimen the ovisac was ial filled by pinkish- yellow (yolk) granules, amongst which nuclei were occasionally seen, and a few of the small cells mentioned above. A thick layer of the grains replaces the coat of proliferating cells on the inner Sumtaces of the wall of the ovisac. In the chamber and ovisacs there may be numerous small yellow masses like those in the testis sacs ; these frequently have a number of dark granules within them. (6) The Spermatiece. The spermathecal apparatus in the genus consists of a spherical or ovoid ampulla, attached to the hinder surface of septum 7/8, and communicating with the exterior by a long, coiling duct; the duct leads downwards on the septum, and ends in an atrial chamber which debouches to the exterior in the intersegmental fissure. The atrium is the variable part of the apparatus; it may be of considerable size, and either simple in form, projecting to one side or other of the septum where this joins the ventral parietes; or bifid, one horn projecting on one side and one on the other side of the septum; or it may be small—even invisible in dissection, owing to its being embedded in the body-wall; or possibly, in some cases, 1t may be altogether absent. In the present species the ampulla consists of the usual coats. The lining epithelium is shortly columnar over the vault of the sac and away from the origin of the duct, and the nuclei are round; it is high, and may be very high, on the lower part of the sac near the origin of the duct, and on the anterior wall, and the nuclei are correspondingly elongated. All the cells are provided with rather short cilia. Embedded in the much elongated epithelium at one part of the wall was a quantity of bright yellow AND ZOOGEOGRAPHY OF INDIAN OLIGOCHATA, W331 matter, lighter in tint than that in the testis sac, in granules, and also in larger spherical aggregations up to 20 4 in diameter. The ampulla is attached by a connective-tissue mesentery to the septum. The first part of the duct is wide—125 ~»—and has a consider- able lumen ; it is lined by moderately high columnar epithelium, with cilia which spring in bunches from each cell, ‘The neck by which it is joined to the ampulla is narrower, 60 « in diameter and similarly ciliated. The duct is also suspended by a mesentery. The coiled part of the duct is narrower, with a diameter of D0; 161s remarkable for the great thickness of its muscular coat—15 »—much more marked than in the previous portions of the apparatus ; the fibres are mostly circular, The epithelium is cubical and non-ciliated ; a lumen is not visible. The mesentery, with marked muscular fibrils, branches to its attachment to the various coils. The atrium in this species is small, and is contained within the lower part of the septum; it narrows downwards to pierce the body-wall. ‘The duct joins it near the apex; but it is included for some distance within a common connective and muscular coat before the lumina coaiesce. The epithelium of the atrium is high and irregular, and hence the lumen is small, and irregular also; the nuclei are situated towards the base of the cells; there are no cilia. The muscular coat is independent of the musculature of tlie septum. In the body-wall the epithelium is lower, columnar, non-ciliated ; the muscular coat is here apparently derived from the circular fibres of the body-wall; the diameter of the tube, including its muscular coat, is 85 pw. (7) Accessory (Copulatory) Glands. The sexual animal is characterized by a number of markings, each a transversely oval patch, slightly raised above the general surface, with a circular groove in its centre; these are two. three or four in number, may ‘be paired or unpaired, and may occur on the ventral surface of segms. Vil, vill, 1x, and xii, or any of these. Corresponding to each of these markings there j is found inter- nally a small, almost spherical mass, attached to the body-wall by a short stalk-like portion. On investigation by sections they pre- sent the following characters (fig. 5):— Each is about -42 mm. in diameter, and is solid, its centre con- sisting of a mass of cells. The outer thin coat of peritoneum is in places perceptible only with difficulty; it appears as an ex- tremely fine membrane with slight thickenings here and there which represent much flattened nuclei; in places it carries blood- vessels. The muscular coat (m.) is very well marked—9 p thick ; its fibres are derived from both the muscular layers of the body- wall. There is no special (blue-staining) connective-tissue layer, though there are traces of this tissue in between the muscular fibres, as in the body-wall. The cells which All up the interior of the organ (c.) are large i ho} if fS) 5) 132 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, and stain a light pink; the nuclei, 3-4 u in diameter, do not stain deeply; the cell contents are uniformly granular. Cell outlines are present; the cells in the depth of the. organ appear polyg gonal in section, while those nearer the mouth are elongated in the direction of the neck, and at the mouth these are continuous with the surface epithelium. It may be that all the cells of the interior reach the surface. The figure of one of these organs shows a remarkable similarity toa simple type of spermatheca ; and it does not seem impossible that spermathece may have arisen from such structures. (8) Rudimentary (1.e., Incipient) Gizzards. Invisible in dissection, but revealed in the microscopic examina- tion of sections, is an incipient gizzard belonging to segm. X1 (fig. 4, rad.), and, more rudimentary still, one in x. Notable is the thickening of the connective-tissue layer in the incipient gizzards, outside the layer of the strengthened circular muscular fibres. (For references to literature see end of next article.) Heplanation of Plate J. .1. Nephridium of Drawida japonica, X70. coil, spiral coil; d., duct; ep., epithelial lining of sac; Joop, twisted loop; m., muscular coat of sac ; n., nephrostome; 2. c., nephrostomial canal; p., peritoneal coat of sac; sac, oval sac; spin., spindle-shaped portion ; ¢7., trumpet-shaped expan- sion of duct at junction with body-wall. . 2. Section through portion of testis sac, X100. 6.v., blood-vessels ; e., cells in sac; ep., epithelial lining of sac; f£, elongated epithelium of funnel ; g., pink-staining ground-substance of sac contents; m., muscular fibres ; m., nuclei in ground-substance ; sept., septum 9/10; p., peritoneal coat of sac; ¢., testis; v. def., vas deferens; 2., yellow or brown chloragogen-like erains and masses. r, 8. Section of prostate, somewhat oblique, X50. The covering layer of flattened peritoneal cells, if and where present, is not to be seen at this magnifica- tion. ats., atrium in centre of prostatic cells, with epithelial and muscular coats; wuc., nuclei of prostatic cells; prost., prostatic cells; wv. def., vas deferens in wall of atrium ; w., vacuoles with central particle. Fic. 4. Longitudinal section, rather oblique, through ovarian chamber and neigh- bouring segments (semi-diagrammatic), 20. ann., annular chamber round gut; d.v., dorsal vessel; ep., epithelium of cwsophagus; ~, female funnel ; giz., muscular thickening of cesophageal wall, constituting the gizzard of segm. xii; par., parietes, with epithelial, circular, and longi- tudinal muscular layers; o.¢., ovarian chamber; rzd., rudimentary vizzard of segm. xi; sac, ovisacs; 10/11, 11/12, 12/13, the "corresponding septa. Fig. 5. Section of accessory (copulatory) gland, X50. ¢., cells filling up interior of gland; e.m., circular muscular Rabe ot body- ‘vals ce.m.’, two strands of circular muscle traver sing plug of cells at mouth of gland ; ep., surface epithelium ; m., muscular coat of gland (the peritoneal coat not visible at » this magnification); sept., septum 9/10; ¢., trabecular (blue-staining) connective tissue, the longitudinal muscular layer of the body-wall being absent here. Fig. 6. Transverse section through ovarian chamber (semi-diagrammatic), x12. alim., alimentary canal; ann., annular chamner round gut; d.v., dorsal vessel; o.¢., ovarian chamber; ova, masses of ova, cut tangentially to attachment of ovary to anterior wall of chamber; paz., parietes; x., upper limit of ovarian chamber; v.ch., ventral chamber; v.7.¢c., ventral nerve cord; v.v., ventral yessel; y., yolk masses. es gg Fi gg a go AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ATA. WAS VI.—On rue RELATIONSHIPS OF THE GENERA OF MONILIGASTRID#; WITH SOME CONSIDERATIONS ON ‘tHE ORIGIN oF TERRESTRIAL OLIGOCH ETA. — (1) The Testis Sacs of the Oligocheta. . The primitive condition of the male organs of the Oligocheta is that seen in most of the lower families as well as in many Megascolecide and Lumbricide. The testes are attached to the anterior wall of their segment, the funnels to the posterior; the sperm cells are shed into the ccelomie cavity at an early stage of their development, and ripen in the general cavity of the segment, or more usually in cxeal sacs, the seminal vesicles, formed by bulgings of the septa (text-fig. 7). In this case there are no testis sacs. Text-figure 7. as aes ; SV t f ud. Primitive condition; no testis sacs. f, male funnel; s.v., seminal vesicle (shown in-seems. ix and xii); ¢., testis; v.d., vas deferens. Not infrequently, however, it happens that the lower part of the segment, where the testes and funnels lie, is cut off by a partition from the rest of the cavity of the segment. ‘Ihe Vext-figure 8. The ordinary form of testis sacs, e.g. as in Lwmbricus. References as before ; in addition, ¢.sac., testis sac (median seminal vesicle). organs are then contained in special com partments of the ccelom, the testis sacs—either a single one in each testicular segment, 134 DR. J, STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, containing the organs of both sides, or a pair in each segment, one for the or ans vot the right and one for those of the left side. Thus in the common European genus Lumbsicus there is a single testis sac (often called the median seminal vesicle) in each of the testis segments ; while in many species of the common Indian genera, Pheretima and Hutypheus, the sacs are paired (text- fig. 8). In the genus Ocnerodrilus, and in a few species of Enchytreus, testis sacs of a different type are found. The sacs enclose the testis, and are attached like the latter to the anterior septum of the segment, projecting freely into its cavity; the funnels, however, are outside the sac, and, as usual, on the posterior septum of the segment; it is evident that the sac must rupture before the spermatozoa can enter the funnel. In this case the sac can hardly be anything else than the superficial layer of the proliferating peritoneal cells which give rise to the gonad; this layer separates off as a connected membrane, beneath which the genital cells are shed and ripen (text-fig. 9)*. The lobes of the Text-figure 9. Testis sacs of Ocnerodrilus occidentalis. References as before. testes of some species of Luwmbricillus are also (at any rate in their young state) surrounded by a thin membrane, within which the genital cells are cast off and undergo some of their development. Lastly, in all the Moniligastride the testis sacs are paired vesicles, suspended by the septa in such a way that they project into the segment in front or into the segment behind, or (more usually) into both; the degree of projection forwards or backwards varies in different individuals of the same species, and sometimes on the two sides of the same individual. Within each sac—usually perhaps on its floor towards the anterior wall— is the testis, a proliferation of the epithelial lining of the sac; the funnel is also included in the sac, and is situated on the * T donot know of any observations on migration of genital cells in the Oligo- cheeta. If such took place, it might be possible to interpret the sac wall in these, cases as the peritoneum, raised up and distended by the ripening germ cells which had taken up their position beneath it. AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ATA. 135 floor near the !testis, sometimes actually surrounding the attachment of the testis as in Drawida japonica. The vas deferens leads off from the funnel, and descends on the septum to the ventral body-wall (text-fig. 10). Text-figure 10. Testis sac of Drawida. References as before; in addition, p7’., prostate.+ | (2) The Testis Sacs of the Moniligastride considered as Segments. Leaving aside for the moment the Moniligastride, it will be seen from the above that throughout the Oligocheta the testis is morphologically on the posterior face of the anterior septum of its segment, of the peritoneal covering of which it is a proli- feration; and the funnel faces it on the anterior face of the posterior septum, of the peritoneal covering of which it is a modification. On now comparing the figure which represents the condition in the Moniligastride (text-fig. 10) it is evident that the most feasible explanation of the nature of the testis sacs of this family is that which would regard them as the equivalent of segments, of which the anterior and posterior walls have fused together for the greater part of their extent. The cavity of the sacs cannot be a mere split in the septum, since it is lined throughout with epithelium. Following the usual rule, the cavity must be ccelomic, and the epithelium peritoneal epithelium. In Syngenodrilus, to which reference will again be made in the sequel, there is given off from the posterior testis sac a seminal vesicle (text-fig. 16) in precisely the same way as the seminal vesicle is given off from the testis segment in the more primitive forms (cf. text-fig. 7). Again, we have seen in the foregoing article how, in the genera Moniligaster and Drawida, there is a tendency to shut off the ovaries and their funnels in special chambers. These chambers represent segment xi, and various stages in the gradual limitation of the extent of this segment can be traced in actually existing forms. The ovarian chamber contains the ovaries and 126 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, their funnels. and the alimentary canal is excluded. The testis sacs have similar relations; and if the ovarian chamber were to lose its nephridia, and to become divided into two, one for the organs of each side, we should have an identical condition reproduced. It is difficult to resist the conclusion that the testis saes owe their origin to the same process which we can see and follow in the case of the ovarian segment. Beddard, so long ago as 1891 (6), suggested this origin for the testis sacs ; he thinks they ‘“ may be all that is left of the ccelom belonging to the segment which contains the testes.” He adds: ‘“« This is of course no more than a suggestion”; and, so far as I know, the suggestion has never since been considered. But I believe that the above considerations show with some degree of probability that this is really the true nature of the saes. (3) The Relationships of the Genera of Moniligastride. In the present section I omit the consideration of the recently discovered genus Syngenodrilus, to which reference will be made later, and confine myself to the previously known genera of the family. The most primitive of the four genera is Desmogaster, in which the sexual organs are arranged as shown in the accompanying figure (text-fig.11). I use letters to indicate the several segments, since It is easier thus to make comparisons between this and other genera. Text-figure 11, 7 CHR.) Lee eS on Si ovls Sptn. L OSA MILE VEO. Of. Desmogaster. 1.h., last heart; m.f.. male funnel; o., ovary; o.f., ovarian funnel ; ovis., ovisac; spth., spermatheca; ¢., testis; ¢.sac., testis sac; w.d., vas deferens. There may be either one or two pairs of spermathece ; two pairs are shown in the figure, but if only one pair is present it may be either the anterior or the posterior. Hupolygaster is derived from Desmogaster by the disappearance of the posterior set of male organs and the anterior sperma- thece (the anterior spermathecee are the complement of the posterior male organs; these face each other when the worms AND ZOOGEOGRAPHY OF INDIAN OLIGOCHAETA. 137 are apposed in copulation); otherwise the figures are identical (text-fig. 12) :— Text-figure 12. Eupolygaster. In Drawida and Moniligaster the figure is as below (text- fig. 13). Text-figure 13. Drawida and Moniligaster. The ovarian chamber corresponds to the ovarian segment in the other genera; as in Hupolygaster, there is a single pair of spermathece and a single set of male organs. But there is a further reduction still—another segment has disappeared. Assuming for the moment that the similarly lettered segments in text-figs. 11 and 13 correspond, ¢ with cand g with g, it will be seen that the single testis sac of Drawida is left to correspond with the two testis sacs and the intervening segment e of Desmo- gaster ; in other words, there has been a squeezing together of the segments, and the two testis sacs have fused. This fusion probably gave at first a single sac with a vertical diaphragm ; later the diaphragm disappeared (cf. text-fig. 14, c). In thus fusing, the testis sacs have annihilated the space e—in other words, the cavity of the segment between them; the most posterior heart, which was in this space, has also disappeared, and hence in these two genera, Drawida and Moniligaster, the last heart is in the segment in front of the testis sac. The actually existing testis, funnel and duct of these genera may not improbably be the testis of the original anterior sac and the funnel and duct of the posterior. Michaelsen (12) imagines the genus Drawida to have arisen in a rather different way—by the disappearance of the anterior set 138 DR, J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, of male organs of Desmogaster and the dislocation forwards of all the other sexual organs. Though the difference between the two views is not great, | prefer the above wording, for the following reasons :-—— First, the approximation of the septa in the region of the testes is observed in other genera ; it has been noted by Michael- sen in some species of Acanthodrilus (15); it occurs in Hutypheus, where septa 8/9-10/11 are usually very close together (testes and funnels have, however, in this genus disappear ed from segment x); it 1s seen also in Hoplocheetela' (Stephenson, 24), where septa 9/10, 10/11, and 11/12 may be so closely approximated as to appear at first like one hugely thickened septum. Next (and this ‘I am inclined to regard as important), the process that I have assumed for Drawida is apparently seen in progress in Syngenodrilus (cf. text-fig. 16), where the two testis sacs are contiguous, indeed fused, and the intervening segment has been squeezed out of existence ventrally, though a portion persists dorsally. Lastly, the trabeculee which Benham describes as crossing the testis sac in Moniligaster indicus (= Drawida robusta) (10) may possibly represent the remains of the septum which, on my view, at first separated the cavities of the two sacs after they had come together. It may be added that the existence of a rudimentary second prostate, in front of that belonging to the normal male apparatus, in Drawida willst (Michaelsen, 13), and of a similarly situated well-developed second prostate in D). raui (Stephenson, 26), shows only that the actual funnel and vas deferens of Drawida are the equivalent of the posterior pair of Desmogaster ; the testis may quite well be the anterior testis, as I have supposed, and the testis sac the product of fusion of the original two.* (4) The Numbering of the Segments in Desmogaster. The segments 7, g, h, f, e (text-figs. 11 & 12) correspond in Des- mogaster and Hupolygaster; and on the view here advocated, 2 and f# correspond in Desmogaster and Drawida (text-figs. 11 & 13). Now 7i-e are segments vili-xi, with the anterior testis sac, In Desmogaster, and vii-x, with the testis sac, in Hupolygaster ; i and f# are vill and ix in Desmogaster, vii and vili in Hupoly- gaster and Drawida. That is to say, the whole of the sexual organs are one segment further forward than they ought to be in Hupolygaster and Drawida. It would be easy arbitrarily to suppose that a segment in the anterior part of the body had somehow been lost in “the ancestors of these two genera, or, with Michaelsen, to assume a “ dislocation forwards” of the sexual organs (which comes to much the same * It may be said that if it is the posterior vas deferens which persists in Drawida, the persistent: spermatheca should be the anterior and not the posterior. This is not necessary, however; as already stated, there are species of Desmogaster with two pairs or with only one pair of spermathece ; in the latter case, the single pair of spermathecee may be the anterior or the posterior. In D. schildi it is the posterior ; if such a form were to undergo the assumed squeezing together and fusion of the testis sacs, the exact condition of Drawida and Moniligaster would be arrived at. AND ZOOGHOGRAPHY OF INDIAN OLIGOCH ATA. 139 thing). But there is no reason to suppose that Hupolygaster and Drawida had a common origin—the reverse in fact; and if they originated separately, this curious dislocation must have occurred twice independently. I believe the explanation to be that the accepted numbering of the segments in Desmogaster is incorrect, and that the numbers given to all the segments should be diminished by one. This view requires some justification. There are three means of establishing the enumeration of the segments at the anterior end of the body—the sete, which commonly begin in segment 11; the septa, which are, however, invariably wanting in the anterior segments ; and the annulation of the body. Of the four species of Desmogaster, the sete are invisible in the anterior segments in 2). dorie ; they do not begin till segment vili (laterally not till ix) in D. schildi; they do not begin till segment 1x in J). giardi ; and we have no information concerning them in D. horsti, except that they are minute. The sete are thus useless for the determination of segments. As to the septa, in all species the extreme displacement of the septa in the anterior part of the body is remarked by the authors. In D. horsti the tenth septum (presumably the meaning is septum 10/11) is either absent, or displaced backwards to the level of 11/12; if we suppose it to be absent, the numbering of the segments is brought up to that which is generally accepted for the genus, while if we make no such supposition, the numbering of the segments is apparently that for which I am contending. Concerning the annulation, we only know that in D. horsti “the two anterior segments have their longitudinal diameter smaller than that of the succeeding ones”—which appears to admit of the supposition that they really constitute a single segment; and that in D. schildi segments i-iii are simple while iv-xll are biannular. But since there are no sete near the anterior end of this worm, the distinction of primary end secon- dary grooves is conceivably somewhat arbitrary, and a slight difference in estimating the value of the grooves would reduce the number of segments by one. It is worth noticing that in a species of Hupolygaster (HL. browni, 13) Michaelsen is in doubt regarding the numbering of the segments because of the secondary annulation, sete being absent here also from the anterior - segments. Naturally, when the numbering of the segments had once been determined in the first species of Desmoyaster (rightly or wrongly), subsequent writers describing new species would, if in any doubt, bring their numbering to correspond, as Michaelsen does in the example of the Hupolygaster just quoted *. * The first species of Desmogaster to be described was D. dorie (Rosa, 19). There seem to be inconsistencies in the data there given. ‘he ovaries are said to be im segm. xiil; but the funnels to be on anterior face of septum 14/15—7.e., in sezment xiv. The spermatheceze (there is only a single pair in this species) are first stated to be in viii, but further on to be on the posterior face of septum 6/7—1.e., in vil. This would seem to indicate that the estimation of the segments was not easy, and that the author himself came to different results at different times. 140 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, Since, therefore, the criteria for estimating the numbering of the segments in the anterior part of the body of Desmogaster allow of results of only doubtful value, I believe we are justified in taking a numbering which corresponds—allowing for the evolution described above—with that of the other genera of the family. In any case, if subsequent investigations, directed specially towards this point, should show that the accepted numbering is correct, I should prefer to account for it by assuming an increase in the number of segments in the anterior part of Desmogaster after the stocks of Hupolygaster and Drawida had been given off, rather than by an independently produced elimination of a segment, or a dislocation forwards of all the organs, in these two genera. The following figures show the relations of the three genera ; Text-figure 14. A. Desmogaster; B. Eupolygaster ; C. Drawida and Moniligaster. The actual segments ave numbered, and the relations indicated, AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ATA. 141 the letters show the morphological correspondences, and the roman numerals show the actual segments (revised as above im the case of Desmogaster). (5) The Relationship of Moniligaster to Drawida. The genera Drawida and Moniligaster differ only in the fact that Moniligaster has, and Drawida has not, connected with the ectal end of the spermathecal duct, a bilobed glandular appendage, of which each lobe consists of a number of closely-packed branches. In Drawida there may be a swelling on the end of the duct, or at most a relatively small sac-like atrial diverticulum. Michaelsen (18) takes Drawida for the ancestor, Moniligaster for the descendant; he traces the history of the at first simple duct in Drawida. In some species it “widens and is transformed into a muscular coat. In other species of this genus this widened muscular distal (2. ¢., ectal) end grows into a real muscular atrial chamber, which further on enlarges at one side to form a separate blind sae, depending into the second segment, or, in the species D. robusta and its subspecies, at two sides, forming two outgrowths, one depending into the seventh segment, the other into the eighth, the two outgrowths being separated by the septum 7/8,-and the duct of the pouch entering the atrial chamber at the angle between these two outgrowths. The structure found in the species of Moniligaster may be compared with this structure in D. robusta. The two outgrowths of the atrial cavity in the latter species are continued in Moniligaster each into a large gland, consisting of a large, much-branched tube, the branches and twigs of which are packed together and enveloped in a peritoneal membrane.” This derivation does not seem easy tome; I should prefer to read the series (if indeed it is a series) in the reverse direction. T cannot see in Moniligaster any atrial chamber at all—only two glands, with ducts which unite, like the limbs of a Y, to form a single canal. This is well shown in Perrier’s original plate (18) ; my own description of the condition in Perrier’s species 1s similar (22):—‘ The stalk (atrium) is relatively narrow, and pifurcates, one branch going to each half of the mass. Neither the stalk nor its two branches are in any sense sacs; the appearance is that of two ducts proceeding one from each half of the mamillated glandular mass, which unite to form a common duct; this common duct is less than half as long as the glandular part and nowhere dilated.” Itseems difficult to imagine a slight swelling of the end of the duct dilating, becoming a sac, bifureating, and then sprouting glands from each horn, itself concomitantly narrowing to a Y-shaped duct. If the series really is a natural one, it would be easier to read it the other way — as a gradual reduction, of the glands toa non-functional vestige and their final disappearance. It is possible that the glands ave entirely independent of the spermathece in origin; compare 142 Dh. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, the accessory spermathecal glands in Drawida schunkarai (Michaelsen, 16); similarly there are associated yet independent glands in the genera Hoplochetella and Erythreodrilus, belonging to another family, the Megascolecide. (6) Lhe Origin of the Moniligastride. If the testis sacs represent segments, they must have existed as such in the ancestors of the present genera. Let us take Desmogaster, as the most archaic of existing genera of Monili- gastride, and expand the testis sacs into segments; we get a form which may be represented as in the accompanying figure (text-fig. 15). Text-figure 15. Desmogaster, with the testis sacs expanded to appear as segments. Now let us take the remarkable genus Syngenodrilus, described by Smith and Green (20), and separated by them as a distinct subfamily. The actual arrangement of the sacs and septa is approximately represented in the subjoined figure (text-fig. 16), which is a diagrammatic copy of the section illustrated in the original paper. Text-figure 16. Vil VIM Syngenodrilus. The external segmentation is indicated. Pv., prostate. Note the three pairs of prostate glands, which, however, are none of them in close association with the male ducts or pores; it is quite possible, therefore, that these prostates of Syngenodrilus have nothing to do with the structures that go under the same name in the other genera of the family. If we expand this figure, so as to separate the testis sacs (which are close up against AND ZOOGEOGRAPHY OF INDIAN OLIGOCH2TA. 143 one another) and correct the extreme displacements of the septa, we get the following (the seminal vesicle may be neglected, as being only a diverticulum of the posterior testis sac) (text-fig, 17). Text-figure 17. Syngenodrilus, with the testis sacs separated from one another. The condition is similar to that in Desmogaster, except that the spermathece are a segment further back, and that there are three pairs of prostates. If now we expand the testis sacs into segments, we get a form with the following characters (text-fig. 18) :— Text-figure 18. Syngenodrilus, with the testis sacs expanded so as to appear as segments. The figure is identical, except for the characters already mentioned, with that of the ‘“‘ expanded ” Desmogaster. Probably neither text-fig. 15 nor 18 represents a form that actually existed. In the first place, there is a gap of a segment between the two pairs of testes; and in the second place, the apertures of the spermathecze are separated by only one segment, while those of the male apparatus are separated by two segments, and hence the spermathecal apertures would not coincide with the male pores of another worm in copulation. It has been pointed out by Michaelsen (12) that in all the lower families of Oligocheta the gonads follow each other in successive segments; the ovarian segment immediately follows the testicular segment or segments; and the normal arrangement in the higher families (testes in segments x and xi, ovaries in xiii) is to be explained by the dropping- -out of a pair of ovaries in xii. The original condition—two pairs of testes followed by two pairs of . ovaries—still exists in the genus Haplotuxis; and hence Michael- sen (12, 17) and Beddard (9) look on this genus as the ancestor of the higher Oligocheta. 144 DR. J, STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, To get a complete series of gonads in the ancestors of the Moniligastride, however, we must supply a third pair of testes in the vacant segment between the two pairs already present, as well as a pair of ovaries between the last testicular segment and the actual ovarian segment, ‘That the insertion of the third pair of testes is Justified is seen from the presence of three pairs of prostates in Syngenodrilus. One further addition is to be made— three pairs of male pores imply three pairs of spermathecal pores apposed tothem in copulation. The diagram of the ancestor thus becomes that shown in text-fig. 19. Text-figure 19. The ancestral terrestrial Oligochzte, with complete series of gonads and spermathece. The above diagram probably represents an animal in which as yet the moniliform series of gizzards characteristic of the family had not been evolved. The gizzards are in front of the testis segments in Syngenodrilus, behind them in the other genera: that is, the muscular wall of the esophagus has been strengthened in one place in one subfamily, in another place in the other. Syngenodrilus (text-fig. 16) is derived from the above form by a contraction, or huddiing together, of the testis segments in such a way that the middle one is almost squeezed out of existence and the other two become testis sacs, by a corresponding reduction of the number of spermathece to two pairs opening in 7/8 and 8/9, and by the disappearance of the anterior pair of ovaries with their funnels and ducts. Desmogaster (text-fig. 11) is derived from the common ancestor by the contraction of the segments of the first and third pairs of testes to form testis sacs, and the disappearance of the middle pairs of male organs; by a corresponding diminution in the number of spermathece, the remaining pairs cpening in 6/7 and 7/8; and by the disappearance, as in Syngenodrilus, of the anterior ovary with its funnel and duct. Eupolygaster (text-fig. 12) came into existence by the further disappearance of the posterior pair of male organs and the anterior pair of spermathece of Desmogaster. Drawida and Moniligaster (text-fig. 13) originated from a form similar to Desmogaster by a contraction or huddling together of the testis segments like that in Syngenodrilus, but carried further, AND ZOOGEOGRAPHY OF INDIAN OLIGOCH ATA. 145 so that the testis sacs fuse, extinguishing the intervening segment; the spermathece are again reduced to one pair. (7) Lhe Origin of Terrestrial Oligocheta. From an ancestor having the characters illustrated above (text-fig. 19), then, we can derive the various genera of Monili- gastride without, as it seems to me, forcing the facts in any way, and relying only on processes which can be illustrated from actual examples of the genera themselves or of other families of Oligocheeta. I was, however, surprised, and I must confess to feeling pleased, to find that the diagram arrived at as above gives in a very natural way a starting-point for the other families of terrestrial Oligocheta as well. These all agree in having, primarily, two pairs of testes in segments x and xi, and one pair of ovaries in xiii; the primary number of spermathece in the Megascolecide is two pairs, opening in grooves 7/8 and 8/9. For the Glossoscolecide and Lumbricide it is not so easy to give the primitive position of the spermathece ; perhaps all that can be said is that the Glossoscolecine, probably the ancestral group of these two families, typically have spermathece which open in front of the testis segments (Michaelsen, 17). The arrangements of the gonads and their funnels im the higher families is thus obtained simply by the disappearance of the posterior pairs of testes and ovaries respectively ; and the arrangement of the spermathecee—in the Megascolecide at any rate—is arrived at by the disappearance of the anterior of the three pairs of the ancestral form (the anterior pair of sperma- thece would correspond, as already explained, to the posterior pair of male openings). The primary position of the spermathece in the Glossoscolecides and Lumbricide is, as has just been indicated, scarcely ascertained with precision. The occurrenee, in existing earthworms, of gonads or of traces of gonads, in the segments in which, on the present theory, they are supposed to have existed regularly in the ancestor, confirms the view here taken. Woodward (28) has shown that in the common English species the presence of additional pairs of ovaries in segment xiv is muca commoner than has been supposed, and that additional testes are not rare. There are also cases where the supernumerary ovaries are not limited to one pair; Woodward (27) examined a specimen with seven pairs. Confining ourselves, however, to cases where the condition is definitely that of the presumed ancestor—one extra pair of ovaries in segment xlv—we have the following records :—Bergh (quoted by Woodward) described three cases in species of Laum- bricus. Woodward's own investigation shows nine cases of this condition out of fifty worms examined. Beddard, in an examination of a number of specimens of Perronya excavatus GUE Proc. Zoou. Soc.—1922, No. X. 10 146 DR. J.STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, found a number of cases where there were additional female pores one segment behind the normal, and where therefore there were probably additional ovaries and oviducts internally to correspond. The testes which we suppose to have disappeared from segment xii also sometimes crop up again. Thus Woodward (28) found testes here in three cases out of fifty examined. Beddard (4) found a pair of testes in segment xii in Pontoscolex corethrurus (in this case the gonads of segment xiii also appeared to be testes). On the other hand, the gonads of segment xii, which ought, when they reappear (if the common ancestor of earthworms was constituted as above supposed), to be testes, are when present sometimes ovaries. Beddard (/.c.) found this to be so in some examples of Pontoscolex; it is so (if we may draw conclusions from the position of the female pores) in some specimens of Perionyx excavatus (Beddard, J. c.); and Woodward had six cases of this condition out of the fifty specimens of British earthworms which he examined. In a number of cases the nature of the gonad in segment xii is indeterminable, or the evidence is inconclusive. Beddard (l.c.) believes that the same gland in Pontoscolex may produce either ova or spermatozoa. In one specimen examined micro- scopically by Woodward the gland was actually hermaphrodite. In three very young specimens (? of Lumbricus rubellus) a gonad was present on each side in segment xii, but at that stage could not be identified as definitely testis or ovary (Woodward suggests that this pair is always developed in the embryo). Beddard (5, 8) found this pair of glands present in embryos of Octochetus multiporus, but of course could not identify it, at that stage, as either testis or ovary. Beddard also notes (9) that two pairs of ovisacs have been found both by Horst and himself in certain. species of Perichetide, in segments xiii and xiv; these would correspond to ovaries in xii and xi; Beddard remarks on certain resemblances of these organs to sperm-sacs, and, since neither gonads were found in segment xli nor ova in the corre- sponding sacs, it seems not improbable that the organs in segment xiii were vestigial sperm-sacs, and the gonads which formerly existed in segment xii were testes. There is thus no doubt that ovaries occur not infrequently at the present day in segment xiv, and that segment xii also frequently contains a pair of gonads, which in some cases are ovaries, in some cases testes, and in some cases their nature is equivocal. According to our view they were, in the ancestral earthworms, regularly testes. The more usual view of the origin of the terrestrial Oligocheta (Beddard, 9; Michaelsen, 12, 17) is rather different. The Phreoryctide, with two pairs of testes and two pairs of ovaries, in AND ZOOGEOGRAPHY OF INDIAN OLIGOCHATA. 147 segments x—xi and xii-xiii respectively. and with an oesophageal gizzard, are held to be the present-day representatives of the common ancestor. The Phreoryetide are probably in turn to be derived from the Lumbriculidee, where the number of gonads is variable, and also their position; Lamprodrilus satyriscus has three or four pairs of testes, and other forms—e. g., Lumbriculus variegatus—have more than one pair of ovaries. But on the above hypothesis of the origin of the terrestrial families of Oligocheta we require as our starting-point a form with a series of five pairs of gonads—three pairs of testes and two pairs of ovaries; while the discovery of Syngenodrilus makes it probable that the ancestor of the Moniligastride had no special thickening of either cesophagus or anterior intestine—the gizzards developed independently in the two subfamilies, in the Syn- genodriline in front of and in the Moniligastride behind the testicular segments. It is obvious, therefore, that the Phreoryctids are already too specialized for our purpose; the ancestral form as here conceived has a larger number of gonads, and is without an oesophageal gizzard. The ancestral group more nearly resembled the Lumbriculide than the Phreoryctide; the latter are not in the direct line of descent of the earthworms. The Moniligastride, which, as Beddard has repeatedly insisted, vetain many of the characters of the Limicoline section (clitellum one cell only in thickness, non-union of the vasa deferentia of the same side, vasa deferentia passing over only one segment), arose independently of the other earthworms by the disappearance of testes from segment xi and of ovaries from xili, and the con- traction of the original segments x and xii to form testis sacs; while the common stem of the Megascolecide, Glossoscolecide, and Lumbricide originated by the disappearance of testes from xii and ovaries from xiv, the lengthening of the vasa deferentia, aud the union of those of the same side. References to Literature. 1. Bepparp, F. E.—Descriptions cf some new or little-known Farthworms, together with an account of the Variations in Structure exhibited by Perionyx excavatus K.P. Proc. Zool. Soc., 1886, p. 298. 2. a Note on the Reproductive Organs of Moniligaster. Zool. Anz., ix. 1887, p. 679. 3. 44 On the Structure of three new species of Harthworms... . Quart. Journ. Mic. Sci., xxix. 1888, p. 119. 4. bs On certain points in the Structure of Urocheta .P., and Dichogaster, nov. gen., with further remarks on the Nephridia of Earthworms. Quart. Journ. Mic. Sci., xxix. 1888, p. 235. ; 5, as On the Homology between Genital Ducts and Nephridia in the Oligocheta. Proc. Roy. Soc., xlviii. 1890, p. 452. 6. ' Observations upon the Structure of a Genus of Oligocheta belonging to the Limicoline Section. Trans. Roy. Soc. Wdin., xxxvi. 1891, p. 7. 10# 148 ON INDIAN OLIGOCH@TA, 7. Bepparp, Ff. H.—On some new species of HKarthworms from various parts of the World. Proc. Zool. Soc., 1892, p. 693. 8. 5 Researches into the Embryology of the Oligocheta. I. On certain points in the Development of Acanthodrilus multiporus. Quart. Journ. Mic. Sci., xxxiii. 1892, p. 497. 9. A Monograph of the Order of Oligocheta. Oxford, 1895. 10. Bile W. B.—Description of a new species of Moniligaster from India. Quart. Journ. Mie. Sci., 1893, p. 363. 11. Bourne, A. G.—On Moniligaster grandis A. G. B., from the Nilgiris, S. India.... Quart. Journ. Mic. Sci., xxxvi. 1894, p. 316. lla. Cognerrt pE Marrttts, L.—A Contribution to our knowledge of the Oligo- cheta of Travancore. Ann. Mag. Nat. Hist., ser. 8, vii. 1911, p. 494. 12. Micwartsen, W.—Die Geographische Verbreitung der Oligochaten. Berlin, 1908. 13. k The Oligochseta of India, Nepal, Ceylon, Burma, and the Andaman Islands. Mem. Ind. Mus., i. 1909, p. 101. 14. nd Die Oligochitenfauna der vorderindisch-ceylonischen Region. Abh. Naturw. Verein Hamburg, xix. 1910, p. 1. 15. Die Oligochaten von Neu Caledonien und den benachbarten Inselgruppen. In: F. Sarasin und ‘I’. Roux, Nova Cale- donia, A, Zoologie, Bd. 1. Wiesbaden, 1913, p. 171. 16. sy Oligochaten von Travancore und Borneo. Mt. Mus. Ham- burg, xxx. 1913. 17. 6 Die Lumbriciden, mit besonderer Berticksichtigung der bisher als Familie Glossoscolecide zusammengefassten Unterfamilie. Zool. Jahrb., Abt. f. Syst., xli. 1917, p. 1. 18. Perrier, E.—Recherches pour servir 2 |’ Histoire des Lombriciens terrestres. Nouv. Arch. Mus. Paris, viii. 1872. 19. Rosa, D.—Moniligastridi, Geoscolecidi ed Hudrilidi (Viaggio di Leonardo Fea in Birmania e Regioni vicine). Ann. Mus. Genova, ser. 2 a, ix. (xxix.), 1890, p. 368. 20. Suita, F., and Brsstz R. Green.—Descriptions of New African Earthworms, including a new genus of Moniligastride. Proce. U.S. Nat. Mus., lv. 1919, p. 145. 21. SrrpHEnson, J.—Oligocheta, in: Zoological Results of the Abor Expedition, 1911-12. Ree. Ind. Mus., viii. 1914, p. 365. 22. fs On some Indian Oligocheta, mainly from Southern India and Ceylon. Mem. Ind. Mus., vi. 1915, p. 35. 23. er On a Collection of Oligocheta belonging to the Indian Museum. Ree. Ind. Mus., xii. 1916, p. 299. 24. ra On a Collection of Oligochzeta from various parts of India and Further India. Ree. ind. Mus., xiii. 1917, p. 353. 25. 53 On a Collection of Oligocheta from the lesser-known parts of India and from Eastern Persia. Mem. Ind. Mus,, vil. 1920. 26. 7p Oligochzta from Manipur, the Laccadive Islands, Mysore, and other parts of India. Rec. Ind. Mus., xvii. 1921. 27. Woopwarp, M. F.—Description of an abnormal Earthworm possessing Seven Pairs of Ovaries. Proc. Zool. Soc., 1892, p. 184. 28. 3 Further Observations on Variations in the Genitalia of British Karthworms. Proc. Zool. Soc., 1913, p. 319. ON THE VAGUS NERVES OF HYRAX CAPENSIS. 149 8. On the Vagus and Sympathetic Nerves of Hyrawx capensis. By Caaries F. Sonnvrac, M.D., F.Z.8., Anatomist to the Society. [Received January 23, 1921; Read February 21, 1922.] (Text-figures 6-8.) The present paper is based on the examination of several speci- mens of Hyrax capensis, both immature and fully adult, which died in the Society’s Gardens. Variations were observed in the branches of their vagus and sympathetic nerves; and the con- ditions present in some of the animals are more complex than those already described in my papers on the Marsupialia (8, 4) and Kdentata (5). ‘They are also more complex than those in many animals belonging to other orders. The anterior cervical parts of the nerves are placed deeply, and it is necessary to remove the wide ascending rami of the mandible to gain access to them and their branches. In all examples the cervical parts of the vagus and sympathetic nerves are fused on both sides to form vago-sympathetic cords, the union taking place at the level of the middle of the thyroid cartilage. The cords are resolved again into their component elements at the root of the neck, but the separation is usually higher on the left side. After they have parted, however, branches of communication may run between them. Tar Vagus NERVES. Course:—As it emerges from the foramen lacerum posticum each vagus has the usual relations to the glosso-pharyngeal, spinal - accessory and hypoglossal nerves, and it communicates with them and the superior cervical ganglion by well-marked branches (text-fig. 6. C1x, Cx1, Cxir and S.0.G). No ganglion nodosum is present in the neck, as in Zamandua tetradactyla (4), or within the foramen lacerum posticum. I+ then courses postero-mesially, gives off communicating branches to the cervical plexus (C.C.P) and unites with the cervical sympathetic to form the vago- sympathetic cord (V—S) at the level of the middle of the thyroid cartilage. And the nerve up to this point is internal to the mandible. The vago-sympathetic cords break up again in the posterior third of the neck; the vagus half runs still postero- mesially and enters the thorax. Owing to the high level reached by the heart and aorta the first part of the intra-thoracic course is short, but the calibre of the left nerve remains thicker than that of the right one. Behind the roots of the lungs the vagi are united by a sinuous cord which is as thick as the left vagus (text-fig. 7, br), and the nerves emerge again from its extremities, so the arrangement is H-shaped. The left vagus is replaced by vight and left branches. 150 DR. C. F. SONNTAG ON THE VAGUS AND The former runs along the ventral border of the cesophagus, passes through the cesophageal opening in the diaphragm and ramifies over the anterior and ventral aspects of the body and fundus of the stomach. The latter runs along the dorsal border of the cesophagus and unites with the right vagus which deseribes a wide curve to meet it; and the combined trunk ends on the dorsal aspect of the stomach. Numerous communicating branches link the vagi to one another and to the cceliac plexus, and some go to the duodenum. Text-figure 6. The Cervical and Upper Thoracic Parts of the Vagus and Sympathetic Nerves of Hyrax capensis. V: vagus; S: sympathetic; Mn: line of mandible. Other letters in text. Branches :—Numerous branches of communication and distri- bution are again given off, many of which are very complex, ~ The pharyngeal nerves (text-fig. 6,a) are given off from the trunk of the vagus and the loop (C) connecting its superior and recurrent laryngeal branches. They form a plexus with branches of the glosso-pharyngeal nerve and sympathetic. In some speci- mens the laryngeal nerves do not form a loop which gives off pharyngeal branches. SYMPATHEDIC NERVES OF HYRAX CAPENSIS. 151 The superior laryngeal nerve (6) is replaced by a strong cord which unites with one of the divisions of the recurrent nerve. This cord gives off internal laryngeal (7./.n), two external laryn- geal (eli), two thyroid (¢.y.n), and three external pharyngeal (a) nerves. Small branches (v.n) accompany the common carotid artery. The internal laryngeal branch enters the larynx through the thyro-hyoid interval. The left recurrent nerve (e) is given off in the common position. It encircles the aortic arch and runs antero-laterally through the neck. In the middle of the neck it shows a fusiform expansion (1) and divides into a mesial nerve (R.L.N), which is the true inferior laryngeal nerve, and a lateral cord (C) which unites with the superior laryngeal nerve to form a loop, whose upper part has been described above. Numerous tracheal (TN) and cesopha- geal nerves (ON) are given off. They form plexuses of which the tracheal one is the more superficial. And the esophageal plexus anastomoses with the cesophageal branches of the right recurrent nerve (d). The latter also forms a loop with the right superior laryngeal nerve, but there is no fusiform expansion. And both recurrent nerves give off cardiac branches (C.N) which are thicker and more numerous on the left side. In some specimens these loops are absent, and there is no external communication between the superior and recurrent laryngeal nerves. The existence of anastomoses between the laryngeal nerves 1 Man has long been known. And Landois and Stirling (2) sum- marised our knowledge as follows :—‘“ A connecting branch runs from the superior laryngeal to the inferior (the anastomosis of Galen), which occasionally gives off sensory branches to the upper half of the trachea (sometimes to the iarynx?%); perhaps also to the cesophagus (ZLonget), and sensory fibres (?) for the muscles of the larynx supplied by the recurrent laryngeal. According to Francois Frank, sensory fibres pass by this anastomosis from the recurrent into the superior laryngeal. According to Waller and Burekhard, the motor fibres of both laryngeal nerves are all de- rived from the accessorius, while Chauveau maintains that the crico-thyroid is an exception.” In a recent paper Dilworth (1) described and figured a well-marked branch running between the recurrent and internal laryngeal nerves, and says: “*.... the laryngeal nerves are really a plexus of nerves. Just as the vagus breaks up into its various plexuses in the body, it does the same in the larynx. It isa highly modified plexus. I would further suggest that it arose by the larynx separating a strand of fibres from the vagus—that this strand is represented by the continuous nerve joining the internal and recurrent laryngeal, and that the separation from this strand of further fibres forms the various nerves of the larnyx.” The conditions shown in text-fig. 6 sup- ort Dilworth’s views, and I would extend his views of the origin ot the laryngeal nerves from the continuous strand to account for the origin of the tracheal, cesophageal, and thyroid nerves. 152 DR. C. F. SONNTAG ON THE VAGUS AND The Cardiac Nerves (text-fig. 6,7) arise on the left side from the vagus distal to the origin of the left recurrent nerve, but the right ones arise from the vagus and.its recurrent branch. They run to the cardiac plexus wherein they become associated with the sympathetic (c.B.s). . The wsophageal nerves (text-figs. 6, on, 7,0N, and 8, vB) arise in the neck from the cords uniting the laryngeal nerves. In the upper part of the thorax they are branches of the right vagus, but they come from the two divisions of the left vagus distal to the roots of the lungs. The tracheal nerves (text-figs. 6,1.N and 7,1.N) arise from the laryngeal cords in the neck, but its terminal part and the main bronchi receive their nerve-supply from the thick bridge between the vagi (67). Text-figure 7. The Lower Thoracic and Abdominal Parts of the Vagus Nerves of Hyraa capensis. L.V: parts of left vagus; R.V: right vagus. Other letters in text. The pulmonary plecuses (text-fig. ight one is formed by two branches from the ee vagus and communi- cating twigs from the deep cardiac plexus. But the left one consists of many branches from the bridge between the two vagi. The latter sends a branch to the aortic plewus (A.P), which receives many twigs from the left vagus. And no ganglia were found in the pulmonary and aortic plexuses. The Gastric Nerves (text-fig. 7):—The left vagus reaches the stomach along the ventral border of the cesophagus and breaks up into two branches. The first gives off twigs which run along SYMPATHETIC NERVES OF HYRAX CAPENSIS. 153 the lesser curvature, and the latter supplies the fundus. Both groups anastomose with branches from the right vagus, and with the celiac plexus. The cord formed by the right vagus and part of the left one supplies the dorsal aspect of the stomach almost to the pylorus. Its branches anastomose with those of the lett vagus, and twigs from the celiac plexus. A well-marked branch (splanchnic?) runs to the left sympathetic cord (A). Many of the cesophageal nerves run posteriorly and anastomose with both vagi, so there is a close network round the lower end of the cesophagus and over the left part of the stomach. Tur SyMPATHETIC NERVES. The Superior Cervical Ganglion (text-fig. 6,8.c.¢) is round or oval, and flat. It gives off an internal carotid nerve of con- siderable length (1.c.N) which accompanies the internal carotid artery into the skull. The nerve communicates with the glosso- pharyngeal and hypoglossal nerves, and the ganglion is connected to the vagus, cervical plexus, and loop between the superior and recurrent laryngeal nerves. But no connection exists between either and the spinal accessory nerve. From the posterior pole of the ganglion the sympathetic cord runs laterally and joins the vagus at the level of the middle of the thyroid cartilage to form the vago-sympathetic cord. In Ayraaz, as in all animals possessing the vago-sympathetic cord, no middle cervical ganglion is present, and no direct rami communicantes run to the middle cervical nerves. The sympathetic separates again in the posterior part of the neck, but communicates with the vagus after the partition. It exhibits a well-marked inferior cervical ganglion (I.C.G) on the left side, but none on the right. From the ganglion branches of communication run to the brachial plexus, but I was unable to detect any communication between the right sympathetic and brachial nerves, or between either cord and phrenic nerves. The right sympathetic gives off a branch which accompanies the vertebral artery (V.A.N), and a loop connects it to the right recurrent nerve. And as no branch runs directly to the cardiac plexus, the right recurrent and right vagus nerves conduct all the right sympathetic filaments to the heart. The left inferior cervical ganglion gives off a medium-sized cardiac branch (c.B.8) which runs almost parallel to the left recurrent nerve, passes to the dorsal aspect of the aortic arch and ends in the deep cardiac plexus. The Annulus of Viewssens (A.V) is present only on the left side, and the right sympathetic cord passes in front of the right subclavian artery. The left sympathetic thoracic cord (text-fig. 8, u.s) has few ganglia in its anterior part, and it forms loops. At the level of the middle of the root of the left lung it divides into lateral and mesial divisions. The former possesses a loop anda small ganglion, 154 DR. C. F. SONNTAG ON THE VAGUS AND and is continued as a thick nerve which ends in the celiac gang- lion (text-fig. 8, c.¢). The latter ends in a large ganglion whence two nerves emerge. One joins the lateral division and the other becomes the gangliated cord of the sympathetic. In the thorax it gives branches to the aortic plexus(A.P) and a long branch runs to the ceeliac ganglion. The lateral division of the cord, and the eceliac branch of its mesial half constitute splanchnic nerves (Sp.N). Text-figure 8. The Thoracic and Abdominal Sympathetic of Hyrax capensis. V.B: aortic and cesophageal branches of the vagi; L.D: line of diaphragm. Other letters in text. The abdominal part of the left sympathetic (text-figs. 7 and 8) possesses four ganglia. It gives off two groups of lateral branches to the semilunar ganglion, rami communicantes to the lumbar nerves, and branches to the aortic plexus. It has a great ten- dency to subdivision. The thoracic part of the right sympathetic (text-fig. 8, R.s) has few ganglia in the anterior part, and its posterior part is very complicated, but not so much as the left cord. It divides into mesial and lateral parts and these are united at intervals by common ganglia or communicating branches. The lateral division ultimately continues the cord back to the sacrum. The mesial division is thick and strong. Branches of the cord run to SYMPATHETIC NERVES OF HYRAX CAPENSIS. 155 the aortic plexus (A.P), vena azygos major (Az.N), thoracic nerves, and left sympathetic. Jn the abdomen the mesial division divides into two, and the following are the fibres of distribution :— . To the eceliac ganglion (A). . Hepatic plexus (contained in A). . Phrenic plexus (contained in A). . Fibres to the right renal plexus (R.R.P),. . Fibres to Meckel’s Tract and Duodenum (M.B). Fibres to the colon (C.B). The abdominal part lies close to the left sympathetic and fibres connect them. It gives off the right renal plexus (R.R.P), right spermatic plexus (R.S.P), and filaments to the aortic plexus. The Cardiac Plexus consists almost entirely of the deep part, the superficial plexus consisting only of a few filaments derived from it. ‘The nerves entering into it are :— 1. Three branches of the left vagus. 2. A twig from the left recurrent nerve. 3. A branch from the inferior cervical ganglion of the left sympathetic. 4, Two branches from the right reeurrent nerve. 5. Two branches from the right vagus. No ganglia are present, nor are there separate depressor nerves ; but these may be contained in the loops uniting the superior and recurrent laryngeal nerves. Oftshoots of the plexus can be traced into the pulmonary and tr: acheal plexuses. The Aortic Plexws (A.P) derives its fibres from the two sympa- thetic cords, the bridge between the vagi in the thorax, and the solar plexus. The Solar Plexus (text-fig.7):—In the upper part of the abdomen dorsad of the stomach there is a broad band of fibres with a large reddish-brown semilunar ganglion at its left extremity. The fibres are closely packed and have wide con- nections as foilows :— 1. The splanchnic nerves from the left sympathetic cord entering the anterior pole of the ganglion (A). 2. Two bundles of fibres from the left sympathetic cord entering the mesial border of the ganglion (B). 3. Fibres from the right vagus to the large left splanchnic nerve (C). 4. Fibres from the right vagus to the semilunar ganglion (D). 5. Fibres from the left vagus to the semilunar ganglion (E). 6. Fibres from the right vagus running into the plexus and turning down to the splenic plexus (F). 7. Fibres from the semilunar ganglion to the splenic plexus (G). 8. Branches from the solar to the splenic plexus (1H). 9. Branches from the right sympathetic to the semilunar ganglion (1). 10. Hepatic plexus (J) giving off pyloro-duodenal nerves (K). 11. Diaphragmatic plexus (L). HOU Co bo 156 ON 'THE VAGUS NERVES OF HYRAX CAPENSIS, SuMMARY AND CONCLUSIONS. 1. In all examples of Hyrax capensis the cervical parts of the vagus and sympathetic nerves are fused. And the ganglion nodosum is frequently absent in the neck. 2. The recurrent and superior laryngeal nerves are frequently, but not always, connected by a loop. 3. There is no separate depressor nerve. 4, The internal carotid nerve has a very long cervical course. 5. The right recurrent nerve has well-marked cardiac branches. 6. The left sympathetic alone has an inferior cervical ganglion and an Annulus of Vieussens. 7. The posterior thoracic parts of the vagi have a complicated arrangement. 8. The thoracic sympathetic cords have few ganglia, and there is only one semilunar ganglion. 9, The right sympathetic is distributed to the colon, and the left one and cceliac plexus supply the small intestines. BIBLIOGRAPHY. 1. Ditwortu, T. F. M.—Journal of Anatomy, 1921, pp. 48-52. 2. Lanpots and Srrrtinc.—A Text-book of Human Physio- logy, vol. ii. 3. Sonntac, C. F.— Proc. Zool. Soc. London, 1921, pp. 572-575. 4. 3 Proc. Zool. Soc. London, 1921, pp. 873-876, 5. - Proc. Zool. Soc. London, 1922, pp. 99-180. # ‘SINZOILIV SINOUHODOIdVH ‘dur yany ‘UTL F8 Tap weet 'P AL Wel IN oka Gia Gillis) “arta Nad ae m4 SS Gh GUN etal se a IN Ose Gta ve ‘dit yyny ‘WMT ZO Tap users Pp NS \ \ SUI UciON Vane ttie:G: Se SE) at Chatien Srone: Seer ES es ie eo ae vate e ie) - ‘OQNHIVNDSOWOVN HS Xd LNaAd SINOYUHOOTdVH T -durr w4ny : ‘USTT 49 TSP User 'C Title NV Daa oGcels 42d ae ST LTeOeroONOT Hw VOUlvol love SstNOadn oO rary hee : ‘dart HFT FO TSP Wear © Atic NVOae 6cetS 4c ON THE CICHLID FISHES OF LAKE VICTORIA, Lae 9, The Cichlid Fishes of Lake Victoria. By C. Tarn Rraan, M.A., F.RS., F.Z.8., Keeper of Zoology in the British Museum (Natural History). [Received January 6, 1922: Read February 21, 1922. | (Plates I-IV.*; Text-figures 1-14.) A revision of the Cichlide of Lake Victoria is a much more difficult task than that of revising the Cichlide of Nyassa, inasmuch as the types of all the described species from Nyassa are in the collection of the British Museum (Natural History), whereas a number of the types from Lake Victoria are in Paris, Berlin, and Genoa. Dr. Pellegrin has kindly sent me notes on two or three species in the collection of the Paris Museum about which I had some doubts, and I am indebted to Mr. Boulenger for photographs of the types of Haplochromis nuchisquamulatus, H. obliquidens, H, longirostris, and H. sauwvagei, which were sent to him by the late Professor Hilgendorf in 1898, and to Dr. Pappenheim for information about the first two of these species. In Boulenger’s ‘Catalogue of African Freshwater Fishes’ 47 species of Cichlids are described from Lake Victoria (including the Victoria Nile, Lakes Salisbury, Kioga, etc.). The number in the present revision is practically the same (50), but, as no less than 18 species are described below as new and some others formerly regarded as synonyms have been re-established, it will be evident that a considerable number of species recognized by Boulenger have been eliminated. A summary of what is known of the Cichlide of the great African lakes may now be given ‘. From Lake Albert only three species have been recorded (Tilapia wilotica, Haplochromis wingatu, and H, multicolor), all of which occur also in the Bahr-el-Gebel. Tilapia nilotica is found in Lakes Edward and Kivu; in addition, Lake Edward has a species of Tilapia (1'. eduardiana) which is closely related to 7. variabilis of Lake Victoria, 5 endemic species of Huplochromis, 4 apparently related to H. cinerews and the fifth to H. spekii, of Lake Victoria, and an endemic monotypic genus, Schubotzia, distinguished from Haplochromis by the dentition. Kivu has five species of Haplochromis peculiar to the lake, of which two seem to be related to H. cinereus. From Tanganyika 89 species belonging to 37 genera are known. Except 7%lapia nilotica, all these are endemic and the majority of them belong to endemic genera. Nearly all the Tanganyika Cichlid belong to genera which may have evolved — * For explanation of the Plates see p. 191. + For Cichlide of Lakes Edward and Kivu, see Ann. & Mag. Nat. Hist. (9) viii. 1921, p. 632; of Tanganyika, Ann. & Mag. Nat. Hist. (9) v. 1920, p.30; of Nyassa, P. Z.S. 1921, p. 678. 158 MR. C. TATE REGAN ON THE in the lake from two ancestral types, Limnotilapia and Haplo- chromis; but, although there are a number of genera which appear to be derived from Haplochromis, that genus itself is represented by two species only. From Nyassa 84 Cichlide belonging to 14 genera* have been described, nearly all endemic. The most striking feature is the presence of 53 endemic species of Haplochromis, which appear to form a natural group and have evidently evolved in the lake from one or a few ancestral forms. Of the endemic genera four (with 12 species) are evidently derived from forms closely related to species of Haplochromis now living in the lake, and two more (Cynotilapia, Pseudotropheus) are related to Haplochromis. Ot the others Hemitilapia and Otopharynx are near Tilapia, Chilo- tilapia is Otopharynx specialized, and Corematodus is mainly distinguished from the endemic Tilapia sqwamipinnis by its mouth and dentition. The indications are that the endemic Nyassa Cichlids have originated in the lake from about haltf-a- dozen ancestral forms. In Lake Victoria there are 50 species of Cichlid, all peculiar to the lake except Tilapia zillii, a Nilotic species. The endemic species area Tilapia, 44 Haplochromis, and 4 monotypic genera distinguished from Haplochromis by peculiarities of the dentition, or in one by an increased number of anal spines. The species of Haplochromis exhibit almost as great a diversity as in Nyassa, yet there are certain features which enabie one to say almost at a glance to which lake a species belongs. In most of the Nyassa Haplochromis the caudal fin is more or less distinctly emarginate, and is covered with small scales in the adult fish, whereas in most of the Victoria species it 1s rounded or truncate and is scaly only in the basal half. In the Nyassa Haplochromis a few distinctive types of coloration are prevalent— for example, 15 species have a dark band on each side from nape to caudal fin, and several others have four dark spots on each side; in most of the Victoria Haplochromis, when markings are present, they take the form of regular dark cross-bars with a dark band from head to caudal fin and another above the lateral line. There can be little doubt that Haplochromis cinereus, a small species, with rather short decurved snout and moderately small and slightly obliqne mouth, is a generalized type; species with these characters are widely distributed in African rivers. A number of species in Lake Victoria are closely related to H. cinereus, but differ in the dentition of the jaws or of the pharyngeals, in haying thicker lips, more gill-rakers, a larger eye, etc. Haplochromis serranus and its allies are somewhat larger * In my paper I recognized 15, but I now find that Astatotilapia cannot be maintained as distinct from Haplochromis, several of the Lake Victoria species having enlarged teeth at the ends of the premaxillaries more or Jess developed and not at all constant. CICHLID FISHES OF LAKE VICTORIA. 159 species, with a larger mouth and more prominent lower jaw; in some species of this group the mouth is oblique, and in some the outer teeth are strong and spaced. Finally, there are species with a strongly projecting lower jaw, in some of which the cleft of the mouth is nearly vertical. One species, H. ishmaeli, is of peculiar interest; it is so like H. cinereus that the types (12 specimens) include 6 examples of the latter, but it differs in having much larger and more massive pharyngeals, with the teeth large armel blunt, ihn eas in H, cinereus they are small and slender. ie have described a similar pair of species (/7. tetrastigma and H. placodon) from Nyassa. A remark- able group of three species includes Haplochromis sauvagei and the monotypic genera Macropleurodus and Hoplotilapia, which scarcely differ from each other except for the considerable differ- ences in dentition, and seem to illustrate Cope’s paradoxical view that a species may persist through several genera. It is especially interesting to note that the normal markings are those described above as characteristic of many Victoria species, but that in addition all three are known to possess a ‘ bicolor” form, a sort of piebald, in which the dark pigment is concentrated into a number of blotches and irregular cross-bars which extend on to the vertical fins. Another interesting example of ahs same species in different genera is provided by Haplochromis annectens and Platyteniodus degent ; ; here the former shows a slght departure from the normal Haplochromis dentition towards the Platyteniodus type. The conclusion is that in Victoria most of the endemic Cichlide have evolved in the lake from a type very similar to Haplochromis cinereus, and that the primary evolution has been in the size and shape of the mouth, and the number, structure, and arrangement of the teeth, doubtless in adaptation to different kinds of food and different methods of feeding. The facts seem more in harmony with Gulick’s theory of habitudinal segregation than with the modern idea of evolution by accidental mutations. The Cichlide of Tanganyika and Nyassa lead to the same conclusion. In my paper on the Nyassa Cichlide I called attention to two striking examples of convergent evolution, Pseudotropheus and Aulonocara of Nyassa, respectively showing a great resemblance to Tropheus and Trematocara of Tanganyika. Lake Victoria provides some more examples : Haplochromés chilotes has the lips thick and produced into lobes as in Lobochilotes of Tanganyika ; ff, ishmaeli has the pharyngeals massive and provided with large blunt teeth as in H. placodon of Nyassa. Further, lomaginiies obliquidens and Macropleurodus bear a resemblance to Nyassa genera in their dentition, the former to Hemitilapia, the latter £0 Chilotilapia. If the degree of differentiation be taken as a guide, one may form the conclusion that the Cichlid have inhabited Tanganyika, longer than Nyassa, and Nyassa longer than Victoria. 160 MR. C. TATE REGAN ON THE From what has been said above as to the evolution and relationships of the Cichlidz of Victoria, it will be evident that I do not regard the classification here proposed as entirely satisfactory. A number of divergent species are placed in Haplochromis and a few extreme types are regarded as generically distinct, although the close relationship of each to a species of Haplochronis is obvious. At present I am not in a position to improve this arrangement. Synopsis of the Genera. ES SEMICE COOOL |: cooooobopscdbnonooude-iancdoto seqounssceadaueaan | Ibon tLe INA. Il. Scales ctenoid. A. Teeth in 2 or more series anteriorly ; a single series on each side of upper jaw. SHAVUOT sassso svavcacongoanneued esseggooncdenapsacauaasasonocana sc, Bn JelayNIOGorOaS: 4to 6 anal spines ........ Levesseeseee. 8. Astatoreochromis. B. Several series of teeth on each side of upper jaw. 1. Upper jaw with an outer series of enlarged teeth and several inner series of small teeth anteriorly, and 3 or 4 series of enlarged teeth laterally. 4. Macropleurodus. 2. Teeth small, conical, in bands. Bands rather broad, narrower at the sides than in front ...... 5. Hoplotilapia. Bands very broad, upper broader at the sides than in front... 6. Platyteniodus. 1, Trnapra A. Smith, 1840. Pharyngeal apophysis formed by parasphenoid only. Scales eycloid. An outer series of bicuspid teeth and several inner series of tricuspid teeth. Africa and Syria. 1. TILAPIA VARIABILIS Bouleng., 1906. Tilapia variabilis Bouleng. Cat. Afr. Fish. 11. p. 167, fig. 108. Teeth in 4 to 8 series, 50 to 120 in outer series of upper jaw. 17 to 19 gill-rakers on lower part of anterior arch. Dorsal XVI-XVIIT 10-13. . Anal III (IV) 10-11. Pectoral longer than head. Total length 320 mm. L. Victoria and Victoria Nile. 2. TILAPIA ZILLIT Gerv., 1848. Tilapia zillii Bouleng. Cat. Afr. Fish. iu. p. 197, fig. 126. Teeth in 3 to 6 series, 20 to 60 in outer series of upper jaw. 8 to 11 gill-rakers on lower part of anterior arch. Dorsal XIV— XVI 10-13. Anal IIT 7-10. Pectoral usually not longer than head. Total length 290 mm. Syria to the Niger and the Victoria Nile. 2. Haprocuromis Hilgendorf, 1888. Pharyngeal apophysis formed by parasphenoid in middle and basioccipital at sides. Scales ctenoid. Teeth in 2 or more series CICHLID FISHES OF LAKE VICTORIA. 161 anteriorly, becoming a single series laterally, conical or com- pressed, unicuspid, or outer mostly bicuspid and inner tricuspid. Africa and Syria. All the species described below are peculiar to L. Victoria and the Victoria Nile. Haplochromis obliquidens, the type species, has a dentition unlike that of most of the species which have been placed in this genus, but the discovery of a species in L, Kivu (2. astatodon) some individuals of which have a typical Ctenochromis dentition, whilst others approximate to H. obli- quidens, makes 1t possible to regard this difference as only sub- generic and to still include species with conical or cuspidate teeth in Haplochromis. Synopsis of the L. Victoria Species. 1. Teeth slender, cuspidate, in 5 to 8 series, the inner well developed and not separated by a distinct interspace from the outermost. (Weochromis Regan.) Snout usually a little projecting beyond lower jaw, which is short and broad, about + lengthof head ©....... . L. nigricans. Jaws equal or lower slightly projecting, relatively slender, avout --slenathwommead. oa. chsscreck ec section cose stone ecnceeer 2. nuchisquamulatus. If. Outermost series of teeth conical or bicuspid, enlarged, separated by an inter- space from the smaller inner teeth. (Ctenochromis Ptefter.) A. Jaws equal anteriorly, or lower shorter than upper, or rarely slightly pro- jecting. 1. Lips not produced into lobes. a. Interorbital width not more than 4 length of head. a. Pharyngeal teeth slender. %* Lower jaw not shorter than upper; lips normal. Caudal rounded; caudal peduncle as long as deep; maxillary extending to below anterior part of eye ............0....00. Caudal subtruncate; caudal peduncle longer than deep ; maxillary extending to below anterior edge of eye ...... Caudal truncate; 11 or 12 gill-rakers on lower part of ANGELIOTRELG lyases gen ieee va Nae SAUL at craet 0 atuaveincye ane aes Caudal truncate ; 9 or 10 gill-rakers; eye 3 in head, twice preorbital depth Caudal tr teats 7 to 10 gill- rakers ; g ey e - more e than 311 in 1 head * Lower jaw not shorter than upper; lips thick. Co . gestrt. . nubilus. . melanopus. . Macrops. . cinereus. NOD OO Teeth in 4 to 8 on 30 to 46 in outer series of upper jaw... 8. sawvagei. Teeth in 3 to 5 series, 20 to 24 in outer series of upper jaw... 9. erassilabris. *k* Lower jaw shorter than upper............... 10. annectens. 8. Middle pharyngeal teeth stout, subconical... 11. humilior. y. Pharyngeal teeth large, obtuse .........:........ 12. ishmaeli. b. Interorbital width more than + length of head. 13. obesus. 2. Lips thick, each produced into a lobe anteriorly .. 14. chilotes. B. Lower jaw distinctly, but usually not strongly, ree A. Outer teeth numerous, close together. 1. Maxillary not extending to below eye ............ 15. teniatus. 2. Maxillary extending to below eye, or nearly. a. Hye + length of head or more (in ee of 100 to 110 mm.).. .. 16. martini. b. Hye less than 4 leng th of head, ‘except i in the very young. Gan Pr; emaxillary pedicels extending to be- tween anterior edges of orbits ............... 17. nigrescens. . Premaxillary pedicels not reaching orbits. * Caudal peduncle 15 to 2 as long as deep. D. XIV-XV 9-10. 8 or 9 gill-rakers on lower part of anterior arch ......... mabhedeasaenunecane! Ish (MeCN anmpas. Proc. Zoou. Soc. 1922, D No. xc il 162 MR. C. TATE REGAN ON THE D. XIV-XV 8-10. 10 to 12 gill-rakers on lower part of AMPETION ATC cele. sau ete each ete EE EERE ee ee LOM anieno don: D. XV-XVII 9-10. 9 to 11 gill-rakers on lower part of BIMAEVKOT CANO 5.5 snsnn500 90° ; Nes 20. guiarti. ** Caudal padaneiens once Sis re as done as deep; 5 series of scales on cheek. Depth of prorbital not greater than diameter of eye (in a specimen of 200 mm. ie .. 21. serranus. Depth of preorbital considerably. ereater than diameter of eye (in a specimen of 220 mm. aie ane Aed Suanbonadesouacans ie CNC HIS *** Caudal peduncle ig as =p loins as deep; 6 or 7 series of scales on (B)INGIS) Fearn npalsoacoumdand bn ab Bando Boe cacaotaatd esate eMC K DONATI B. Outer teeth rather strong, set well apart. Pectoral not reaching anal; caudal subtruncate; mouth little oblique ...... . 24, bayoni. weet reaching anal ; “caudal tr ‘ancate ; F “mouth ‘oblique .. 25. macrodon. . Lower jaw strongly projecting. j gs 2} $ i Pectoral reaching anal; caudal truncate; mouth moderately oblique; maxillary not extending to below eye. Eye 5 in head (in a specimen of 170 mm.) ..................... 26. prognathus. Hye 4 in head (in a specimen of 180 mm.) ..................... 27. maculipinna. B, Pectoral reaching anal, or nearly; caudal rounded or subtruncate ; maxillary reaching vertical-from anterior edge of eye. 1. Mouth ‘moderately oblique ; maxillary extending to below eye. a. Premaxillary pedicels ending above nostrils ... 28. dichrourus. 6. Premaxillary pedicels ending well behind nostrils. Maxillary extending to below anterior Re of eye; caudal peduncle longer than deep .......... 29. spekit. Maxillary barely | reaching vertical from anterior ‘edge of eye; caudal peduncle : asilonetasideep) weresaeee cere eens 30. serranoides. . Mouth very oblique. Maxillax y reaching vertical from anterior edge of eye.......... 31. acutirostris. Maxillary not reaching vertical from anterior edge of eye . 382. plagiostoma. C. Pectoral not reaching anal; mouth not very ppiGnee lower jaw very prominent, with the anterior teeth exposed. Outer teeth forming a close-set series; maxillary reaching vertical from anterior edge of eye ............... ec ee cess eee 33. macrognathus. Outer teeth strong, spaced; maxillary not quite reaching to below eye ...... 34. dentex. Outer teeth strong, spaced ; “maxillary not nearly reaching tovbelogrueyien 2) phe ee Aiea ed elo Muar Ulan, Une We ee 35. mento. D. Pectoral not reaching anal; mouth very oblique. 1. 8-10 scales from origin of dorsal to lateral line... 36. cavifrons. 2. 5 to 7 scales from origin of dorsal to lateral line. : a. Depth of body 22 in the length ..................... 87. orthostoma. 6. Depth of body 3 to 44 in length, a, Caudal pedunele 14 to 13 as long as deep. Lower jaw projecting upwards above end of snout ............ 38. xenostoma. Lower jaw not projecting above end of snout; maxillary not quite reaching vertical from anterior edge of eye ...... 39. pellegrini. Lower jaw not projecting above end of snout; seek ending midway between nostril and eye ned ane .. 40. argenteus. (3. Caudal peduncle 13 to 21 as lea as JED. Snout 13 to twice diameter of eye (in specimens of 100 to USO son Tae)) Were saben acon haat aanre nates accoce aba melnchucscsatemdere 2 llat ory. URa arn Snout 1; to 13 diameter of eye a ieee of 115 to 160k) AB igdishSihik cohrseiia nds HABE REL REALE LEE. ILO CLCULECA LOC ILI. Outer teeth few and ee with ce anterior cusp and mdistinct posterior cusp (Bayonia Bouleng.) Je Sek ee Mase ee BA’ aren OwOr. IV. Teeth slender, distally expanded and compressed, outer obliquely truncated (Haplochromis ELil orercl) is VEE aga meme natn OL obliquidens. CICHLID FISHES OF LAKE VICTORIA. 163 1. HAPLocHROMIS NigRIcANS Bouleng., 1906. Tilapia wigricans (part.) Bouleng. Cat. Afr. Fish. iii. p. 240, fig. 160. Tilapia simotes Bouleng. t.c. p. 242, fig. 161. 9 Depth of body 23 to 3 in length, length of head 3 to 33. Snout decurved, about as long as diameter of eye, which is 31 to 33 in length of head, greater than preorbital depth, equal to interorbital width. Jaws equal or lower jaw a little the shorter, short and broad, about 4 length of head; maxillary reaching vertical from anterior edge of eye or a little beyond ; teeth slender, cuspidate, in 6 to 8 series, outer but little enlarged and not separated by a distinct interspace from inner; 40 to 70 in outer series of upper jaw. 3 or 4 series of scales on cheek. 7 to 10 gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 31 to 33 scales in a longitudinal series, 5 or 7 from origin of dorsal to lateral line. Dorsal XV-XVI 9-10; last spine 3 length of head. Anal IIIT 8-9; third spine nearly as long as last dorsal. Pectoral a little shorter than head, not reaching anal. Caudal truncate or subtruncate. Caudal peduncle as long as or a little longer than deep. Brownish or greyish, with traces of blackish cross-bars; males with 2 to 4 ocelli on anal. Nine specimens, including types of 7. nigricans and 7’. simotes, 70 to 145 mm. in total length, and a skeleton. Boulenger’s figures show 7. wigricans as having quite a different physiognomy from 7’. simotes, but this is because the head is directed upwards in the former. 2. HAPLOCHROMIS NUCHISQUAMULATUS Hilgendorf, 1888. Chromis nuchisquamulatus Hilgend. Sitzb. Ges. naturf. Fr. Berlin, 1888, p. 76. 9 Ctenochromis nuchisquamulatus Pfeffer, Thierw. O.-Afr. Fische, p. 14 (1896). Tilapia ngricans (part.) Bonleng. Cat. Afr. Fish. iii. p. 240. Haplochromis nuchisquamulatus (part.) Bouleng. t.c. p. 290. Depth of body 23 in length, length of head 3 to 31. Snout with straight profile, about as long as diameter of eye, which is 33 to 33 in length of head, greater than preorbital depth, equal to or greater than depth of cheek; interorbital width 31 to 34 in length of head. Jaws equal or lower slightly projecting ; rami of lower jaw longer and more slender than in H. nigricans, about 2 length of head. Teeth as in the preceding species, in 5 to 8 series, 50 to 80 in outer series of upper jaw. 2 or 3 series of seales on cheek. 9 to 11 gill-rakers on lower part of anterior arch.. Pharyngeal teeth small. 31 to 33 scales in a longitudinal series, 5 or 6 from origin of dorsal to lateral line. Dorsal XV— XVI 9-10; last spine 4 length of head. Anal III 9-10; third spine 2 to 4 head. Pectoral as long as or a little shorter than Tales 164 MR. C. TATE REGAN ON THE head, nearly or quite reaching anal. Caudal truncate. Caudal peduncle 14 to 14 as long as “deep. Silvery or greyish, with or without 8 dark cross- bars an opercular spot; sometimes a bar below eye. Lake Victoria. Two specimens, 100 and 145 mm. jong, from Kakindu (Bayon) and Buddu Coast (Stmon), and two Sivelecone. Very near 1. nigricans ; the difference in the structure of the lower jaw is well seen on comparing the skeletons. 3. Hapiocnromis Gestri Bouleng., 1911. Haplochromis desjontainesti (part.) Bouleng. Cat. Afr, Fish. 11. p. 302. : Paratilapia ae (part.) Bouleng. t. c. p. 318, Hea Abid Depth of body 22 to 3 in length, length of head 22 to 3, Snout with straight or convex profile, ‘from as ; long as to i diameter of eye, which is 34 to 43 in length of head, greater than preorbital depth, in adults less then depth of cheek ; interorbital with 33 to 44 in length of head. Jaws equal anteriorly or lower slightly projecting ; premaxillary pedicels not quite reaching aD between orbits; maxillary extending to below anterior 4 or 3 of eye; teeth in 2 to 4 series, conical in adult, 34 to 60 in gue series of upper jaw.. 3 to 5 series of scales on cheek. 8 gill-rakers on lower part of anterior arch. Pharyngeal teeth caval 30 to 33 scales: in a longitudinal series, 5 or 6 from origin or dorsal to lateral line. Dorsal XV-X VI 8-10; last spine 2 to 4 oe of head. Anal TIT 8-10; third spime 2 ‘or 2 head. Pectoral § to 4 head, reaching vent. Caudal rounded. Caudal medtomalle as ions as deep. Often a dark band from opercular spot to base of coal. sometimes a second above lateral line; a bar below eye usually present ; soft dorsal and caudal with or without spots; males with blackish pelvic fins and oeelli on the anal. Several specimens, 70 to 160 min. long. Orit 4, HapLOCcHROMIS NUBILUS Bouleng., 1906. 2 Paratilapia victoriana Pellegrin, Bull. Soc. Zool. France, 1903, p. 185, and Mem. xvii. 1905, p. 182, pl. xvii. fig. 3. Pilapia nubila Bouleng. Cie Air. Wish. ps 2307 nes l55. Haplochromis nuchisquamulatus (part.) Bouleng. t. c. p. 290, mes IEE Haplochromis desfontaineswt (part.) Bouleng. t. ¢. i 302. Depth of body 24 to 3 in length, length of head 25 to 3. Snout as long as or a Tittle longer than diameter of eye, which is 3 to 4 in length of head, greater than preorbital depth, in adult equal to depth of chesle® : ee Suonlbitall width 354 to 4 in seme i of head. Jaws equal anteriorly, or lower feebly projecting; maxillary extending to below anterior edge of eye; teeth es or some conical, in 2 to 5 series, 40 to 603 in cuter series of upper jaw. 3 or 4 series of scales on ‘cheek. 8 to 10 gill-rakers on lower part CICHLID FISHES OF LAKE VICTORIA. 165 of anterior arch. Pharyngeal teeth small. 30 to 32 scales in a longitudinal series, 4 to 6 from origin of dorsal to. lateral line. Dorsal XV—X VI 8-10; last spine } to 4 length of head. Anal IIL 8-10; third spine as long as or a little shorter than last dorsal. Pectoral ? to as long as head, reaching vent or anal fin. Caudal subtruncate. Caudal peduncle longer than deep. Oliva- ceous or greyish to blackish, uniform or with dark cross-bars ; often an opercular spot and a dark bar below eye; anai fin with ocelli in males. Numerous examples up to 125 mm. in total length. Paratilapia victoriana Pellegrin appears to be very closely related to, if not identical with, 7. nabilus, but if the figure be accurate the pectoral fin is longer than in any examples of H. nubilus that I have examined, reaching the middle of the anal. 5. HAPLOCHROMIS MELANOPUS, Sp. n. (Text-fig. 1.) Tilapia lacrimosa (part.) Boulenz. Cat. Afr. Fish. iii. p. 234. Haplochromis stanleyt (part.) Bouleng. t. ec. p. 295. Depth of body 2% to 3 in length, length of head 3 to 34. Snout decurved, as longas or a little shorter than diameter of eye, which Text-figure 1. Sf SEK Haplochromis melanopus, sp. 1. is 3 to 32 in length of head, greater than depth of cheek, not twice preorbital depth; interorbital width 33 to 37 in length of head. Jaws equal anteriorly; maxillary extending to below anterior edge of eye; teeth cuspidate or some conical, in 2 to 4 series, 50 to 65 in outer series of upper jaw. 3 to 4 series of scales on cheek. 11 or 12 gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 33 scales in a jongitudinal series, 6 or 7 from origin of dorsal to lateral line. Dorsal XVI 8-9; last spine 2 to L length of head. Anal III 8-9; third spine } to nearly 4 length of head. Pectoral as long as or a little shorter than head, 166 MR. C. TATE REGAN ON THE reaching vent or origin of anal. Caudal peduncle 1+ to 13 as long as ‘deep. antl 6 more or less distinet dark cross- bars ; sometinies an interrupted lateral band ; an opercular spot ; a dark bar. below eye; soft dorsal and cane spotted ; pelvics blackish ; 1 to 3 ocelli on anal fin. Seven specimens, 80 to 100 mm. long, from Entebbe and Bunjako. 6. HApLocnromis MAcRors Bouleng., 1911. ? Astatotilapia jeannelli Pellegrin, Bull. Soc. Zool. France, xxxvil. 1913, p. 313. Tilapia macrops Bouleng. Cat. Afr. Fish. 141. p. 258, fig. 157. Haplochromis stanleyi (part.) Bouleng. t. c. p. 295. Depth of body 22 to 3 in length, length of head 3. Snout shorter than diameter of eye, which is 3 in Tength of head, slightly greater than iterorbital width, twice preorbital depth. Jaws equal anteriorly; maxillary extending to vertical from anterior edge of eye ; teeth in 4 or 5 series in upper jaw, 3 or 4 in lower, outer bicuspid or some conical, 50 to 60 in outer series of upper jaw. 9 series of scales on cheek. 9 or 10 gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 32 scales ina longitudinal series, 5 or 6 from origin of dorsal to lateral line. Dorsal XV 9; lasb spine 2 length of head. Anal III 10; third spine as long as last dorsal. aanor al shorter than head, reaching vent or origin of anal. Caudal truncate. Caudal peduncle longer than deep. Silvery, with or without cross-bars and two dark lateral bands, or uniformly blackish. Three specimens, 85 to 105 mm. long. 7. HAPLOCHROMIS CINEREUS Bouleng., 1906. Tilapia pallida (part.) Bouleng. Cat. Afr. Fish. 11. p. 231, fig. 152. Tilapia lacrimasa (part.) Bouleng. t. ¢. p. 234, fig. 154. Haplochromis ishmaeli (part.} Bouleng. is (G5 Je ASS, Paratiiapia victoriana (pent Boulene. t. ec. p. 341. Paratilapia cinerea Bouleng. t. c. p. 344, fig. 232. Pelmatochromis riponianus Bouleng. t. c. p. 411, fig. 281. Pelmatochromis obesus (part.) Bouleng. t. ¢. p. 414. Depth of body 2} to 34 in length, length of head 25 to 3} Snout ou ved, nearly as long as or lon ger than diameter of eye, which is 33 to Aa in length of head and greater than depth of preeor mel aiterorbie a width 34 to 4 in length of head. Mouth shehtly oblique ; jaws equal anteriorly ; maxillary extending to below anterior margin of eye, or nearly; teeth in 3 to 6 series in upper jaw, in 2 to 5 in lower, 36 to 70 in outer series of upper jaw. Pharyngeal teeth small. 7 to 10 gill-rakers on lower part of anterior arch. 30 to 34 scales in a longitudinal series, 6 or 7 from ogue of dorsal to lateral line. Dorsal XIV-XVI 8-10; last spine 3 to 3 length of head. Anal ITI 8-10; third spine as CICHLID FISHES OF LAKE VICTORIA. 167 long as or shorter than last dorsal. Pectoral as long as head, or a little shorter, extending to origin or anterior part of anal. Caudal truncate. Caudal peduncle 1} to 1? as long as deep. Silvery or greyish, with or without dark cross-bars and a con- tinuous or interrupted lateral hand; often bars across the snout and a vertical one below the eye; males with pelvic fins dusky and ocellar spots on anal fin. Lake Victoria. Numerous examples up to 140 mm. in total length. 8. HAPLOCHROMIS SAUVAGEL Pfeffer, 1896. Ctenochromis sauvaget Pfeffer, Thierw. O.-Afr. Fische, p. 14. Paratilapia granti (Bouleng., 1906) Bouleng. Cat. Afr. Fish. i. p. 342, fig. 231. Paratilapia crassilabris (part.) Bouleng. t. c. p. 345. Paratilapia bicolor (part.) Bouleng. t. c. p. 346. Paratilapia retrodens (part.) Bouleng. t. ¢. p. 347, fig. 235. Depth of body 2% to 34 in length, length of head 22 to 33. Snout as long as or longer than diameter of eye, which is 33 to Ai in length of head, greater than przorbital depth, from a little greater to a little less than depth of cheek; interorbital width 33 to 4 in length of head. Mouth as broadas long; lips thick; Jaws equal anteriorly ; maxillary extending to below anterior edge of eye, or not quite so far; teeth in 4 to 8 series, cuspidate or conical, 30 to 46 in outer series of upper jaw. 3 to 5 series of scales on cheek. 7 to 9 gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 32 to 34 scales in a longitudinal series, 6 or 7 from origin of dorsal to lateral line. Dorsal XTV—XVI 9— 10; last spine 4 to 4 length of head. Anal [iL 8-9; third spine as long as or a little shorter than last dorsal. Pectoral as long as ov a little shorter than head, reaching vent or origin of anal. Caudal truncate. Caudal peduncle 17 to 13 as long as deep. Body without markings or with regular dark cross-bars, or with irregular blackish blotches and cross-bars extending on to vertical fins; usually a dark lateral band and sometimes another above lateral line; an opercular spot; often 2 bars across snout, another between posterior margins of eyes, and another from eye to end of maxillary; soft dorsal and caudal sometimes spotted ; males with 1 to 3 ocelli on anal fin. Numerous specimens 80 to 150 mm. in total length. 9. HAPLOCHROMIS CRASSILABRIS Bouleng., 1906. Paratilapia crassilabris (part.) Bouleng. Cat. Afr. Fish. 11. p. 345, fig. 233. Paratilapia retrodens (part.) Bouleng. t. c. p. 347. Lepth of body 22 to 3 in length, length of head 3. Snout from as long as to 14 diameter of eye, which is 32 to 44 in length of head, greater than preorbital depth, about eyual to depth of cheek; interorbital width 3) in length of head. Mouth broader 168 MR. C. TATE REGAN ON THE than long; lips thick; jaws equal anteriorly; maxillary not ex- tending to below eye; teeth in 3 to 5 series, outer strong, conical in adult, 20 to 24 in outer series of upper jaw. 3 series of scales on cheek. 7 to 9 gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 381 to 33 scales in a longitudinal series ; 6 or 7 from origin of dorsal to lateral line. Dorsal XV—XVI 8— 10; last spine from a little less to a little more than # length of head. Anal III 8-9; third spine as long as or a little shorter than last dorsal. Pectoral a little shorter than head, reaching origin of anal. Caudal truncate or subtruncate. Caudal peduncle 11 to 12 as long as deep. An opercular spot; traces of dark cross-bars and a dark lateral band. Four specimens, 100 to 150 mm. long, from Entebbe; several smaller examples are not included in the description. 10. HaPLocHROMIS ANNECTENS, sp.n. (Text-fig. 2.) Haplochromis ishmaeli (paxt.) Bouleng. Cat. Afr. Fish, 111. p. 293. Depth of body 2? in length, length of head 3. Snout decurved, a little longer than diameter of eye, which is 4 in length of head, Text-figure 2. Haplochromis annectens, sp.n. nearly twice preorbital depth, equal to depth of cheek; inter- orbital width 33 in length of head. Lips thick; lower jaw shorter than upper ; maxillary extending to below anterior edge of eye. Teeth conical, in 4 or 5 series; outermost series enlarged, 36 in upper jaw; inner series of upper jaw forming a band which narrows slightly at the sides and then slightly increases in width at each end. 4 series of scales on cheek. & gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 32 scales in a longitu- dinal series, 7 from origin of dorsal to lateral line. Dorsal XVI9; last spine 2 length of head. Anal III 9; third spine 2 head. Pectoral as long as head, reaching anal. Caudal truncate. CICHLID FISHES OF LAKE VICTORIA. 169 Caudal peduncle longer than deep. A: dark lateral band from opercular spot to base of caudal; a dark bar below anterior part of eye; pelvics blackish ; 3 ocelli on anal. A single specimen, 180 mm. in total length, from Buddu Coast. 11. Hapiocuromis HumiLIoR Bouleng., 1909. Tilapia humilior Bouleng. Cat. Afr. Fish. iii. p. 280, fig. 151. Tilapia bayoni Bouleng. t. c. p. 240, fig. 159. Depth of body 3 to 34 in length, length of head 3. Snout decurved, as long as or a little shorter than diameter of eye, which is 34 in length of head, nearly twice depth of preorbital, greater than depth of cheek ; interorbital width 45 in length of head. Jaws equal anteriorly; maxillary extending to below anterior margin of eye or a little beyond; teeth cuspidate, in 4 to 6 series, 54 to 70 in outer series of upper jaw. 3 to 4 series of scales on cheek. 7 or 8 gill-rakers on lower part of anterior arch. Middle pharyngeal teeth rather stout, subconical. 33 or 34 scales in a longitudinal series, 7 from origin of dorsal to lateral line; pectoral scales very small. Dorsal XV—XVI 9-10, last spine 4 or nearly 3 length of head. Anal IIT 8-10; third spine + to a little more than 2 length of head. Pectoral shorter than head, not reaching anal. Caudal subtruncate. Caudal peduncle 14 to 14 as long as deep. Silvery, brassy, or coppery ; back darker; a dark bar below eye; males with 2 or 3 ocelli on anal, Three specimens, 100 to 130 mm. long, one of the types of the species and two types of 7’. bayoni. The second specimen of 1’. humilior isin very poor condition and is not included in the description. 12. HapnLocHroMis IsHMAELI Bouleng., 1906. Tilapia pallida (part.) Bouleng. Cat. Afr. Fish. ii. p. 231. Haplochromis ishmaeli (part.) Bouleng. t. c. p. 293, fig. 199. Paratilapia victoriana (part.) Bouleng. t. c. p. 341, fig. 230. Depth of body 23 to 3 in length, length of head 23 to 3. Snout decurved, about as long as diameter of eye, which is 37 to 33 in length of head, considerably greater than preorbital depth, equal to or greater than depth of cheek; interorbital width 33 to 4in head. Jaws equal anteriorly or lower very slightly pro- jecting; maxillary extending to below anterior edge or anterior 2 of eye; teeth cuspidate or conical, in 3 or 4 series, 34 to 60 in outer series of upper jaw. 3 or 4 series of scales on cheek. 8 or 9 gill-rakers on lower part of anterior arch. Lower pharyngeal broad and massive, with strong blunt teeth. 31 to 33 scales in a longitudinal series, 6 to 8 from origin of dorsal to lateral line. Dorsal XV-XVI 9-10; last spine 4 to 2 length of head. Anal ILL 8-9; third spine as long as or a little shorter than last dorsal. Pectoral as long as head, extending to above anterior part of anal. Caudal truncate. Caudal peduncle 13 to 13 as long 170 MR. C. TATE REGAN ON THE as deep. Silvery, with or without dark cross-bars; sometimes an interrupted lateral band ; an opercular spot ; usually a bar below eye; soft dorsal and caudal sometimes spotted ; males with ocelli on anal fin. Several specimens, 90 to 135 mm. in total length. The remarkable pharyngeal dentition might well be held to justify the genus Labrochromis (Regan, 1920), were it not that in all other characters the species is nearly identical with H. cinereus. 13. HarLocyromis opEsus Bouleng., 1906. Pelmatochromis obesus (part.) Bouleng. Cat. Afr. Fish. ii. p. 414, fig. 283. Depth of body 2 in length, length of head 3. Snout longer than diameter of eye, which is 4 in length of head, greater than depth of preeorbital, less than depth of cheek ; interorbital width 25 in length of head. Mouth wide, oblique; lower jaw not pro- jecting ; maxillary exposed, extending to below eye; teeth rather stout, conical, biserial, about 50 in outer series of upper jaw. 4 series of scales on cheek. 10 gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 32 seales in a longitu- dinal series, 7 from origin of dorsal to lateral line. Dorsal XV 9; last spine 2 length of head. Anal III 8; third spine 3 head. Pectoral as long as head, reaching anai. Caudal truncate. Caudal penduncle as long as deep. Traces of dark cross-bars ; an opercular spot and a bar beiow eye; soft dorsal spotted; anal fin with 3 ocelli (3). One of the types, 150 mm. long. An isolated species, for which I proposed the generic name Lipochromis (‘ Annals,’ 1920). 14. HapLocHromis cutnores Bouleng., 1911. Paratilapia chilotes Bouleng. Cat. Afr. Fish, ii. p. 338, fig. 228. Depth of body 3 in length, length of head 22. Diameter of eye 4 in Jength of head. Jaws equal anteriorly; lips very thick, each produced anteriorly into a lobe; maxillary not extending to below eye ; teeth small, conical. 3 series of scaleson cheek. 8 or 9 gill-rakers on lower part of anterior arch. 31 or 32 scales in a longitudinal series. Dorsal XVI-XVIT 9. Anal III 8-9. Pectoral 2 head. Caudal truncate. Dark cross-bars and a lateral band. Total length 98 mm. 15, HAPLocHROMIS THNIATUS, sp. n. (Text-fig. 3.) Paratilapia prognatha (part.) Bouleng. Cat. Afr. Fish. i. Depth of body 32 in length, length of head 23. Head 22 to 22 as long as broad; upper profile slightly concave. Snout 13 to 1? diameter of eye, which is 4 to 44 in length of head, greater CICHLID FISHES OF LAKE VICTORIA. 7p than preorbital depth, equal to or greater than depth of cheek ; interorbital width 5 in length of head. Maxillary not extending to below eye; lower jaw moderately projecting ; chin obtuse ; teeth in 3 series in upper jaw, 2 or 3 in lower, some of the inner tricuspid, outer series conical, 40 in upper jaw, anterior mode- rately strong. 3 series of scales on cheek. 8 or 9 gill-rakers on lower part of anterior arch. Dorsal XV-XVI 9; last spine Text-figure 3. Haplochromis teniatus, sp. n. longest, 4 length of head; longest soft rays less than half length of head. Anal ITI 9-10; third spine stronger than and as long as last dorsal. Pectoral 2 length of head. Caudal truneate. Caudal peduncle 12 as long as deep. 31 or 32 scales in a longi- tudinal series, 6 from origin of dorsal to lateral line. Silvery, back darker; an opercular spot; a dark band along middle of side, another above lateral line; dorsal and caudal with small spots. Lake Victoria. Two specimens, 95 and 115 mm. long, from Entebbe (Degen) and Kavirondo Bay (Alluaud). | 16. Haritocuromis MARTINI Bouleng., 1906. Tilapia martini (part.) Bouleng. Cat. Afr. Fish. iii. p. 239, ‘fig. 158. Depth of body 22 to 3 in length, length of head 22 to 3. Upper profile of head convex; snout shorter than diameter of eye, which is 22 to 3 in length of head, greater than interorbital width, twice depth of preorbital. Lower jaw a little projecting; maxillary extending to below anterior 4 of eye; teeth in 3 or 4 series, cuspidate, 50 to 70 in outer series of upper jaw. 4A or 5 series of scales on cheek. 8 or 9 gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 33 seales in a longitu- dinal series, 6 from origin of dorsal to lateral line. Dorsal XV— XVI 8-9; last spine } length of head. Anal Hi 8-9; third spine as long as last dorsal. Pectoral as long as head, extending 1 We MR. C. TATE REGAN ON THE to above anterior part of anal. Caudal truncate. Caudal peduncle 13 to 13 as long as deep. A blackish opercular spot ; a blackish stripe from head to caudal fin, a second above lateral line. Three of the types, 100-110 mm. long, from Bunjako. 17. HAPLOCHROMIS NIGRESCENS Pellegrin, 1909. Astatotilapia nigrescens Pellegrin, Bull. Soc. Zool. France, XxXxiv. p. 157, and Meém. xxii. 1910, p. 292, pl. xiv. fig. 3 ? Astatotilapia roberti Pellegrin, Bull. Soc. Zool. France, xxxvul. 1913, p. 312. Haplochromis percoides (part.) Bouleng. Cat. Afr. Fish. i. p. 296, Paratilapia parvidens Bouleng. t. c. p. 323, fig. 215. Paratilapia serranus ea ) Bouleng. (ty (6) 105 33 4, Depth of body 22% to 3 in length, length of head 25 to 3. Snout 13 to 13 diameter of eye, which is 4 to 42 in length of head, greater than depth of preorbital, equal to or ereater than depth ‘of cheek ; interorbital width 4 in length of head. Upper profile of head straight or slightly concave; mouth moderately oblique; premaxillary pedicels extending to between anterior edges of orbits; maxillary extending to below anterior edge or anterior + of eye; lower jaw usually a little projecting. Teeth conical, or outer bicuspid and inner tricuspid, in 3 to 5 series in upper jaw and 3 or 4 in lower, 40 to 55 in outer series of upper jaw. Cheek with 4 series of scales. 8 or 9 gill-rakers on lower part of anterior arch, seer aymgiee teeth slender. 52 or 33 scales in a longitudinal series, 5 or 6 from first dorsal spine to lateral line. Dorsal XV-XVI 9-10; last spine from 3 to nearly § length of head. Anal III 8-9; third spime stronger than and nearly as long as last dorsal. Pectoral 2 ? length of he: ad, reaching vent or origin of anal. Caudal ee or subtr uncate. Caudal peduncle 1 to 13 as long as deep. Coloration uniform, or a dark lateral hand, or traces of 8 to 10 dark cross-bars; soft dorsal and caudal sometimes with series of spots. Five specimens, 95 to 140 mm. in total length, from the Victoria Nile and Jinja (Bayon). 18. HapLocHROMIS FLAVIPINNIS Bouleng., 1906. Haplochromis percoides (part.) Bouleng. Cat. Afr. Fish, 11 p. 296, fig. 201. Pelmatochromis flavipinnis Bouleng. t. c. p. 418, fig. 286. Depth of body 3 to 31 in length, length of head 27 GON oe Snout 14 to 14 diameter of eye, which is 4 to 5 in fone) of head, equal to or greater than przorbital depth, from #% to a little greater than depth of cheek ; interorbital width 5 ia length of head. Head twice as long as broad; upper profile concave y above the eyes; mouth oblique; maxillary extending to below anterior edge of eye, or nearly; lower jaw projecting; teeth in 3 or 4 CICHLID FISHES OF LAKE VICTORIA. Uys series, outer conical or bicuspid, inner conical or tricuspid ; 45 to 60 in outer series of upper jaw. Cheek with 4 to 6 series of scales. 8 or 9 gill-rakers on lower part of anterior arch. Pharyn- geal teeth slender. 31 or 32 scales in a longitudinal series, 6 to 8 from first dorsal spine to lateral line. Dorsal XIV-XV 9-10; last spine 4 length of head. Anal III 8-9; third spine as long as or shorter than last dorsal. Pectoral 3 to # length of head, not extending to above anal. Caudal subtruncate. Caudal peduncle 14 as long as deep. Four broad dark eross- bars on body and a dark spot at base of caudal fin ; sometimes a longitudinal band connecting first two bars below lateral line and another running forward from caudal spot; a bar between anterior edges of eyes and sometimes another in front of it across snout; a vertical bar below eye and another running upwards and back- wards from posterior edge of eye; an opercular spot; fins yellow, or dorsal and caudal dusky; one or two orange ocelli on anal fin in males. Three specimens, types of the species and of H. percoides, 85 to 145 mm. long. 19. HapLocHROMIS MICRODON Bouleng., 1906. Tilapia lacrimosa (part.) Bouleng. Cat. Afr. Fish. 11. p. 234. Haplochromis stanley: (part.) Bouleng. t. ec. p. 296, fig. 200. Paratilapia serranus (part.) Bouleng. t. c. p. 334. Pelmatochromis microdon Bouleng. t. ¢. p. 412, fig. 282. Depth of body 3 to 3} in length, length of head about 3. Upper profile of head straight or slightly concave; snout from a little shorter than to !4 as long as diameter of eye, which is 3 to 4 in length of head, greater than depth of preorbital, equal to or greater than depth of cheek ; interorbital width 32 to 4 in length of head. Mouth oblique; lower jaw projecting; maxillary reaching vertical from anterior edge of eye; teeth in 3 or 4 series in upper jaw, 2 to 4 in lower, 40 to 70 in outer series of upper jaw. 3 or 4 series of scales on cheek. 11 or 12 gill-rakers on lower part of anterior arch, the posterior much expanded. 30 to 33 scales in a longitudinal series, 44 to 6 between first dorsal spine and lateral line. Dorsal XTV—XV 8-10; Jast spine 1 to 2 length of head; longest soft rays § to 2? length of head. Anal 111 8-9; last spine as long as or a httle shorter and stronger than last of dorsal. Pectoral as long as or a little shorter than head, reaching anal. Caudal truncate or slightly emareinate. Caudal peduncle 14 to 13 as long as deep. Silvery or greyish, back olivaceous or brownish ; sides with or without dark cross-bars and a dark lateral band; males with a dark bar below eye, blackish pelvic fins and ocellar spots on anal. Lake Victoria. The specific name refers to the fact that the type, a specimen of 175 mm., has very small teeth ; this I believe to be due to mal- formation of the lower jaw, which does not bite against the 174 MR. C. TATE REGAN ON THE upper ; in all other characters it agrees with seven specimens of 80 to 140 mm., including the type of H. stanleyi figured by Boulenger. 20, Harnocuromis GuiArti Pellegrin, 1905. Tilapia guiarti Pellegr. Mém. Soc. Zool. France, xvii. p. 184, jOilaea iil waifee le Tilapia perriert Pellegr. ib. xxii. 1910, p. 295, pl. xiv. fig. 4. Tilapia pallida (part.) Bouleng. Cat. Afr. Fish. 111. p. 232. Paratilapia longirostris (part.) Bouleng. t. c. p. 332. Paratilapia serranus (part.) Bouleng. t.¢. p. 334, fig. 225 Par atilapia guiarti Bouleng. t. ¢. p. 236, fig. 226. Depth of body 3 to 4 in length, length of head 3 to 33. Snout from a little shorter than to twice diameter of eye, soit is 3 (young) to 54 in length of head, in adult equal to or less than depth of cheek and equal to or nob much greater than depth of preorbital ; interorbital width 3 3 to 41 in Jength of head. Lower jaw more or less distinctly projecting ; sane pail ap nearly or quite reaching vertical from anterior edge of eye, sometimes a little beyond ; teeth cuspidate in young, conical in adult, in 3 to 5 series, 36 to 70 in outer series of upper jaw. 39 to 5 series of seales on cheek. 9 to 11 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. 32 to 34 scales in a longitu- dinal series, 6 or 7 from origin of dorsal to lateral line. Dorsal XV-XVII 8-10; last spine from + to more thau 2 length of head. Anal III aL 10; third spine stronger and as long as or a little shorter than last dorsal. Pectoral fr om # to nearly as long as head, reaching vent or origin of anal. Caudal truncate or very slightly emar einate, sometimes rounded below. Candal peduncle 12 to 2 as long as deep. Silvery or golden on sides, back darker ; faint dark erogs-bars sometimes present ; often a blackish band from opercular spot to base of caudal and a second above lateral line; soft dorsal and caudal usually spotted; anal with ocelli in males. Numerous examples up to 225 mm. in total length. 21. HAPLOCHROMIS SERRANUS Pfeffer, 1896. Hemichromis serranus Pfeffer, Thierw. O.-Afr. Fische, p. 23. Pelmatochromis spehit (part.) Bouleng. Cat. Afr. Fish. 111. p. 416. Depth of body 22 to 31 in length, length of head 22 to 23. Head 2 to 24 as long as broad ; upper profile straight. Snout 13 to nearly twice diameter of eye, which is 4 to 5 in length of head! equal to or greater than preorbital depth, equal to or less than ‘depth of cheek; interorbital width 4 to 42 in length of head. Mouth moderately oblique; maxillary reaching vertical from anterior margin of ey ; lower jaw projecting ; teeth conical, or inner ti icuspid, in 3 to 5 series in upper jaw and 2 to 4 in eee 40 to 80 in outer sevies of upper jaw. 5 series of scales on cheek. 8 or 9 gill-rakers on lower part of anterior arch. Pharyngeal CICHLID FISHES OF LAKE VICTORIA, 175 teeth slender. 31 to 33 scales in a longitudinal series, 7 or 8 from origin of dorsal to lateral line; pectoral scales very small. Dorsal XV—XVI 9-10; last spine longest, less (adult) or more (young) than 4 length of head ; longest soft rays 3 to 2 length of head. Anal IIT 8-9; third spine stronger and a little shorter than last dorsal. Pectoral 3 to # length of head, nearly or quite reaching anal; pelvics reaching vent or origin of anal. Caudal truncate or subtruncate. Caudal peduncle a little longer than deep. A dark opercular spot, usually a more or less distinct dark band from eye to caudal fin, another above lateral line, and a dark stripe at base of dorsal; traces of dark cross-bars, some- times a dark bar below eye; dorsal and caudal sometimes with series of small dark spots. Pelvies and anal pale or dusky, some- times with ocelli on posterior part of anal. Four specimens, 125 to 200 mm. in total length. 22. HAPLOCHROMIS ALTIGENIS, sp. n. (PI. I.) Paratilapia longirostris (part.) Bouleng. Cat. Afr. Fish. iii. p- 332. Pelmatochromis spekii (part.) Bouleng. t. ¢. p. 417. Depth of body 3 in length, length of head 22 to 23. Head 2 to 27 as long as broad ; upper profileconvex. Snout morethan twice as long as diameter of eye, which is 52 to 6 in length of head, less than preeorbital depth and little more than 3 depth of cheek ; interorbital width 43 to 43 in length of head. Maxillary extending to below anterior margin of eye; lower jaw projecting ; teeth conical, in 4 or 5 series in upper jaw and 3 or 4 in lower, 60 to 70 in outer series of upper jaw. 5 series of scales on cheek. 9 or 10 gill-rakers on lower part of anterior arch. Pharyngeal teeth all slender. 32 or 33 scales in a longitudinal series, 7 or 8 from origin of dorsal to lateral line. Pectoral scales very small. Dorsal XV 9-10; last spine longest, + to 2 length of head; longest soft rays less than 3 length of head. Anal III 8-10; third spine stronger than last dorsal, 2 to 7 length of head. Pectoral length of head or less, reaching origin of anal or not. Caudal subtruncate. Caudal peduncle longer than deep. Silvery; back darker; a bar below eye; an opercular spot and a lateral band more or less distinct. Vertical fins dusky, soft dorsal and caudal with or without series of spots; pelvics blackish; 4 or 5 ocelli on posterior part of anal. Two specimens, 220 and 235 mm. in total length, from Bunjako. A specimen of 100 mm. from Bulolo, L. Kioga, is not included in the description. 23. HAPLOCHROMIS SQUAMULATUS, nom. n. Paratilapia pectoralis (not Ctenochromis pectoralis Pfeff.). Bouleng. Ann. Mus. Genov. 1911, p. 66, pl. i. fig, 2, and Cat. Afr. Fish. ui. p. 339, fig. 229. Closely related to H. serranus and H. altigenis, with the snout 176 MR. C. TATE REGAN ON THE more convex than the former and the preorbital narrower than the latter. 6 or 7 series of scales on cheek. Total length 185 mm. Ripon Falls. HAPLocHROMIS BAYONI Bouleng., 1909. Paratilapia bayoni Bouleng. Cat. Afr, Fish. 111. p. 337, fig. 2 Depth of body 34 in length, length of head 24, Snout as curved, a little more than kos diameter of eye, which i is 53 in length of head, equal to depth of preorbital, less than depth of aihsel 3 interor one al width 44 in length of head. Mouth mode- rately oblique ; lower jaw pr ojecting ; maxillary not extending to below eye; teeth conical, in 4 series in ae jaw, 3 in lower: outer teeth rather strong, set well apart. 4 series of scales on cheek. 9 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. 32 scales in a longitudinal series, 6 from oatsie of dorsal to lateral line. Dorsal XVI 10; last spine 2 length of head. Anal [11 8; third spine + 1head. Pectoral 2 length of head, not reaching anal; pelvics reaching origin of anal. Caudal subtruncate. Caudal peduncle longer than deep. An opercular spot. Here described from one of the types, probably a female, 160 mm. long. ‘The figured specimen, a male of 180 mm., differs in having the first pelvic ray produced into a long flament and in the. presence of two ocelli on the anal fin. 25. HAPLOCHROMIS MACRODON, sp. n. (Text-fig. 4.) Pelmatochromis spekit (part.) Bouleng. Cat. Afr. Fish. iii. p. 417. Depth of body 3 in length, length of head 25 to 23. Upper profile of head somewhat concave. Snout 14 io 12 Ceamene: of Text-figure 4, Haplochromis macrodon, sp. 0. eye, which is 43 to 43 in length of head, a little greater than preorbital depth, equal to or a little less than depth of cheek ; CICHLID FISHES OF LAKE VICTORIA. LG, interorbital width 4 to 43 in length of head. Mouth oblique; maxillary not extending to below eye; lower jaw projecting ; teeth in 3 or 4 series in upper jaw, 2 or 3 in lower, outer conical, strong, and set well apart anteriorly. Cheek with 3 or 4 series of scales. 10 gill-rakers on lower part of anteriorarch. Pharyngeal teeth slender. 31 to 33 scales in a longitudinal series, 5 or 6 from origin of dorsal to lateral line. Dorsal XV 9-10; last spine 3 length of head or a little more. Anal ITI 8-10; third spine as long as or a httle shorter than last dorsal. Pectoral ? head, reaching origin of anal. Caudal truncate. Caudal peduncle 14 to 1§ as long as deep. Silvery; back darker; male with blackish pelvic fins and 3 ocelli on anal. Three specimens, 135 to 155 mm. jong, from Entebbe and Munyonya. 26. HAPLOCHROMIS PROGNATHUS Pellegrin, 1905. Paratilapia prognatha (part.) Bouleng. Cat. Afr. Fish. iii. p. 333, fig. 224. Depth of body 24 in length, length of head 22. Head 22 as long as broad; upper profile slightly concave. Snout twice diameter of eye, which is 5 in length of head, equal to preorbital depth or interorbital width, ? depth of cheek. Mouth moderately oblique; maxillary not extending to below eye; lower jaw strongly projecting ; chin acute; teeth in 4 series in upper jaw, 3 in lower, some of the inner tricuspid, outer series conical; 60 in upper jaw, anterior moderately strong. 3 series of scales on cheek. 10 gill-rakers on lower part of anterior arch. 32 scales in a longitudinal series, 5 from origin of dorsal to lateral line. Dorsal XIV 10; last spine longest, 2 length of head; longest soft rays 3 length of head. Anal III 9; third spine stronger than and as long as last dorsal. Pectoral ? length of head, extending toanal spines; pelvicsreaching anal. Caudal truncate. Caudal peduncle 1} as long as deep. Silvery; back darker; an opercular spot ; a dark bar below anterior part of eye. One of the types, 170 mm. long, from Kavirondo Bay. A specimen of 80 mm. from Hntebbe (Degen) seems to belong to this species. 27. HAPLOCHROMIS MACULIPINNA Pellegr., 1913. (Text-fig. 5.) Paratilapia maculipinna Pellegr. Bull. Soc. Zool: France, xxxvii. p. dll. Paratilapia prognatha (part.) Bouleng. Cat. Afr, Fish. iti. p. 333. Depth of body equal to length of head, nearly 3 in length of fish. Head 2 as long as broad ; profile slightly concave. Snout 1} diameter of eye, which is 4 in length of head, greater than preorbital depth, equal to depth of cheek ; interorbital width 42 in length of head. Mouth obhique; lower jaw strongly projecting; maxillary not quite extending to below eye: teeth triserial, some Proc. Zoou. Soc.—1922, No. XII. 12 178 MR. C. TATE REGAN ON THE of the inner tricuspid, outer series conical, 56 in upper jaw. 3 or 4 series of scales on cheek. 11 gill-rakers on lower part of anterior arch. 33 scales in a longitudinal series, 7 from origin of dorsal to lateral line. Dorsal XV 9; last spine longest, 3 length of head; longest soft rays 3 length of head. Anal IIT 8; third spine stronger than last dorsal, 4 length of head. Pectoral 2 length of head, extending to origin of anal; pelvies reaching Text-figure 5. Haplochromis maculipinna. anal. Caudal truncate. Caudal peduncle 13 as long as deep. Silvery ; back darker; an opercular spot anda lateral bana ; dark spots on dorsal and caudal. A specimen of 180 mm. from Bunjako. The type, 156 mm. long, is described as having the eye 3+ in length of head. : 28. HAPLOCHROMIS DICHROURUS, sp. n. (Text-fig. 6.) Paratilapia serranus (part.) Bouleng. Cat. Afr. Fish. ii. p. 334. Depth of body 33 in length, length of head 2%. Snout 12 diameter of eye, which is 5 inlength of head, equal to Saiecgrpiicl width, greater than depth of przeorbital, less than depth of cheek. Mouth moderately oblique ; premaxillary pedicels ending above nostril ; maxillary extending to below anterior edge of eye; lower jaw strongly projecting ; teeth in 3 series, outer Conical, some inner tricuspid, 50 in outer series of upper jaw. 4 or 5 series of scales on cheek. 9 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. Dorsal XVI 9; last spine longest, less than 4 length of head; longest soft rays less than 3 length of head. Anal II] 8; third spine ? length of head. Pectoral a little less than ? length of head, not quite reaching anal. Caudal subtruncate. Caudal peduncle a little longer than deep. 33 scales in a longitudinal series, 6 from first dorsal spine to lateral line, 5 or 6 between pectoral and pelvic fins. Silvery ; CICHLID FISHES OF LAKE VICTORIA. 179 back darker ; dark spots on snout and above and below eye; an opercular spot; belly dusky; dorsal and upper half of caudal Text-figure 6. Haplochromis dichrourus, sp. n. brown, anal and lower half of caudal bright red; an ocellus on posterior part of anal; pelvics blackish. Buganga, lL. Victoria (Degen). A single specimen, 135 mm. in length. 29. HapLocHromis speKkit Bouleng., 1906. Pelmatochromis svekii (part.) Bouleng. Cat. Afr. Fish. iii. p. 417, fig. 285. Depth of body 23 to 3 in length, length of head 22 to 23. Head 27 to 22 as long as broad; upper profile straight or slightly concave. Snout 14 to twice diameter of eye, which is 4 to 54 in length of head, greater or less than preorbital depth, 1 to 12 in depth of cheek ; interorbital width 4 (adult) to 5 (youn g)in length of head. Maxillary extending to below anterior q of eye ; lower jaw strongly projecting; teeth conical (adult) or outer bicuspid and inner tricuspid (young), in 3 to 5 series in upper jaw, 2 to 4 in lower; 40 to 60 in outer series of upper jaw. Cheek with 4 or 5 series of scales. 9 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. 31 or 32 scales in a longitudinal series, 5 or 6 from origin of dorsal to lateral line. Dorsal XV 9-10; last spine longest, 3 (adult) to 2 (young) length of head; longest soft rays a little more than 3 (@) or 2 (d) length of head. Anal IIT 8-10; third spine 2 to 4 length of head, stronger than 7 7 last dorsal. Pectoral ? to 3 length of head, reaching origin of anal or beyond ; pelvies reaching origin or anterior part of anal. Caudal rounded or subtruncate. Caudal peduncle 11 to 13 aslong as deep. Silvery; back darker; a dark opercular spot; usually a dark lateral band from eye backwards, extending on caudal fin ; a dark bar below anterior part of eye (¢); soft dorsal and caudal 12* 180 MR. C. TATE REGAN ON THE sometimes spotted; pelvics and anal yellow (Q) or pelvics blackish and anal greyish, with several ocelli posteriorly ( ¢). A male of 235 mm. from Bunjako (specimen figured) ; a female of 170 mm. from Entebbe, and four young (100-120 mm.) from Entebbe and Jinja. 30. HAPLOCHROMIS SERRANOIDES, sp.n. (PI. IT.) Paratilapia serranus (part.) Bouleng. Cat. Afr, Fish. iil. p. 354. Pelmatochromis spekii (part.) Bouleng. t. c. 2 AIT. Depth of body equal to length of Menlo ? in length of fish. Snout with straight upper profile, 13 to twice diameter of eye, which is 32 to Bd in leneth of head, cn adult less than depth of prsorbital; interorbital “width 4 to 4% inlength of head. Mouth moderately oblique; maxillary banal reaching vertical from anterior edge of eye ; lower jaw projecting, moderately i in young strongly 1 im ‘adult : teeth conical,in 3 or 4 series in upper jaw aa 2 or 3 in Jower, 40 to 60 in puter series of upper jaw. Cheek with 3 or 4 series of scales, once to 13 diameter of eye. 8 gill-rakers and 2 rudiments on lower part of anterior arch. Pharyngeal teeth slender. 382 scales in a longitudinal series, 5 or 6 from origin of dorsal to lateral line. Dorsal XVI 9-10; last spine 4 to 2 3 length of head. Anal T{I 10-11; third spine 7 to more than = head. Pectoral 2 to 2 = length of inewdl, reaching mene or origin of anal. Caudal subtruncate. Caudal peduncle as long as deep. Silvery or greyish; a dark opercular spot; spinous dorsal dusky ; soft dorsal and anal dusky at the base, pale distally, the dark colour with a well-defined undulating margin; caudal dusky at base. Adult male with a blackish bar below the eye, blackish pelvic fins, and ocelli on the anal fin. Three specimens, 95 to 220 mm. in total length, from Lake Victoria (Delmé Radcliffe) and between L. Kioja and Murchison Falls (J/elland). 31. HAPLOCHROMIS ACUTIROSTRIS, sp. n. (Text-fig. 7.) Paratilapia prognatha (part.) Bouleng. Cat. Afr. Fish. 111. p. 333. Depth of body 3 to 3} in length, length iu head 22 to 24. Snout 14 to 2 diameter of eye, which is 4 to 5Lin length of head, in adult less than depth of preorbital or cheek; nterorbital width 44 to 5in length of head. Mouth oblique, anteriorly above level of eye; lower jaw strongly projecting; maxillary extending to vertical from anterior edge of eye; teeth conical in adult, some cuspidate in young, in 3 or 4 series in upper jaw and 2 or 3 in lower, 40 to 50 in outer series of upper jaw. Cheek with 4 to 6 series of scales. 8 to 10 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. 81 to 33 scales in a longitu- dinal series, 5 or 6 from origin of dorsal to lateral line. Dorsal XV-XVI 8-10; last spine longest, 2 or a little less than ? length of head, Anal IIT 8-10; third spinestronger and as long as and CICHLID FISHES OF LAKE VICTORIA. 181 a little shorter than last dorsal. Pectoral # length of head, nearly or quite reaching anal. Caudal rounded or subtruncate. Caudal peduncle longer than deep. An opercular spot and a Text-figure 7. Haplochromis acutirostris, sp.u. lateral band; vertical fing dusky, the soft dorsal and caudal sometimes with clear spots; adult male with blackish bar below eye, blackish pelvic fins and ocelli on anal fin. Four specimens, 90 to 180 mm. in total length, from Bunjako (Degen) and L. Salisbury (Jackson). 32. HaAPLOCHROMIS PLAGIOSTOMA, sp. n. (Text-fig. 8.) Paratilapia longirostris (part.) Bouleng. Cat. Afr, Fish. 111. p. 332. Depth of body 3 in length, length of head 27. Head a little more than twice as long as broad; upper profile straight. Snout Text-figure 8. Haplochromis plagiostoma, sp. n. 14 diameter of eye, which is 43 in length of head, slightly greater than preorbital depth, # depth of cheek; interorbital width 43 182 MR, C. TATE REGAN ON THE in length of head. Mouth very oblique; lower jaw strongly projecting ; maxillary not extending to below eye ; teeth conical, 4 series in upper jaw, 3 in lower; outer series regular, about 50 in upper jaw. 4 series of scales on cheek. 8 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. Dorsal XV 9; last spine longest, } length of head; longest soft rays 3 length of head. Anal IIT 9; third spine stronger and a little shorter than last dorsal. Pectoral ? length of head, reaching origin of anal ; pelvies reaching vent. Caudal subtruncate, Caudal peduncle a little longer than deep. 390 scales in a longitudinal series, 5 from origin of dorsal to lateral line, 7 or 8 from base of pectoral to middle of chest. Silvery; back darker, an opercular spot and a dark lateral band. A single specimen, 140 mm. long, from Bunjako. 33. HAPLOCHROMIS MACROGNATHUS, sp. n. (PI. ITI. fig. 2.) Paratilapia longirostris (part.) Bouleng. Cat. Afr. Fish. ii. p.332. Depth of body 32 in length, length of head 23. Head 3 times as long as broad; upper profile nearly straight. Snout as long as postorbital part of head, 24 diameter of eye, which is 6 in length of head, a little less than preorbital depth, ? depth of cheek; interorbital width 53} in Jength of head. Maxillary extending to vertical from anterior margin of eye ; lower jaw very strongly projecting, the anterior teeth exposed to the innermost series; teeth conical, in 5 series in upper jaw and 4 in lower, about 80 in cuter series of upper jaw. 5 series of scales on cheek. 9 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. 32 scales in a longitudinal series, 6 from origin of dorsal to lateral line. Dorsal XV 10; lastspine longest, # length of head; longest soft rays 2 length of head. Anal III 9; third spine stronger than last dorsal, } length of head. Pectoral 3 length of head, not reaching anal; peivies reaching vent. Caudal subtruncate. Caudal peduncle 14 as long as deep. A dark band connecting opercular spot with a spot on basal part of caudal, another above lateral line and a dark stripe at base of dorsal; dorsal and caudal with series of dark spots; pelvics blackish ; two ocelli on posterior part of anal. A single specimen, 195 mm. in total length, from Bunjako. This species resembles H. mento in the strongly projecting lower jaw, but differs in the much larger mouth, longer head, shorter caudal peduncle, ete. 34, HapPLocHRoMIS DENTEX, sp.n. (PI. III. fig. 1.) Paratilapia longirostris (part.) Bouleng. Cat. Afr. Fish. 11. p.332. Depth of body 32 in length, length of head 23. Head 23 as long as broad; upper profile slightly convex. Snout nearly twice diameter of eye, which is 5 in length of head, equal to preorbital depth, less than depth of cheek; interorbital width 4 CICHLID FISHES OF LAKE VICTORIA. 183 in length of head. Snout decurved; mouth little oblique; maxillary not far short of vertical from anterior margin of eye; lower jaw strongly projecting, with the anterior teeth exposed. Teeth conical, triserial, outer strong and set well apart anteriorly. 4 or 5 series of scales on cheek. 10 gill-rakers on lower part of anterior arch. Pharyngeal-teeth slender. Dorsal XV 10; last spine longest, 4 length of head; longest soft rays 3 length of head. Anal IIT 9; third spine stronger than and as long as last dorsal. Pectoral more than % length of head, ending above vent; pelvics reaching origin of anal. Caudal slightly emarginate. Caudal peduncle 12 as long as deep. 34 scales in a longitudinal series, 5 from origin of dorsal to lateral line, 10 from base of pectoral to middle of chest. Silvery; back darker; an opercular spot and an interrupted lateral band; dorsal and caudal greyish; pelvics and anal yellow. A specimen of 155 mm. from Sesse Isds. (Bayon). Two young (60 and 80 mm.) from Entebbe may belong to this species. 30). HAPLOCHROMIS MENTO, Sp. n. Paratilapia longirostris (part.) Bouleng. Cat. Afr. Fish. i. p- 332, fig. 223. Depth of body 34 in length, length of head 232. Head 2? as long as broad ; upper profile slightly convex. Snout 23 diameter of eye, which is 6 in length of head, equal to preorbital depth, less than depth of cheek; interorbital width 44 in length of head. Maxillary not nearly reaching vertical from anterior margin of eye; lower jaw strongly projecting, with the anterior teeth exposed; teeth conical, 5 series in upper jaw, 3 in lower, outer strong and set well apart anteriorly. 4 series of scales on cheek. 10 or 11 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. 34 scales in a longitudinal series, 6 or 7 from origin of dorsal to lateral line. Dorsal XVI 10; last spine longest, nearly 4 length of head ; longest soft rays 2 length of head. Anal III 10; third spine stronger than last dorsal, 7 length of head. Pectoral 2 length of head, ending above vent ; pelvics reaching origin of anal. Caudal rounded (?). Caudal peduncle 1? as long as deep. Silvery ; back darker; an opercular spot; pelvics dusky; 3 ocelli on posterior part of anal. A specimen of 210 mm., from Bunjako. 36. HapLocHRoMIs CAvIFRONS Hilgendorf, 1888. “Pelmatochromis cavifrons Bouleng. Cat. Afr. Fish. 1. p. 419, fig. 287. Depth of body about 3 in length, length of head 23 to 23. Snout twice diameter cf eye, which is 5 to 53 in length of head, nearly equal to preorbital depth, | depth of cheek ; interorbital width 4 in length of head. Mouth very oblique; lower jaw strongly projecting ; maxillary not quite reaching to below eye; 184 MR. CG. TATE REGAN ON THE teeth mostly conical, in 4 or 5 series in upper jaw, 3 or 4 in lower, 60 to 70 in outer series of upper jaw. 5 or 6 series of scales on cheek. 7 to 9 gill-rakers on lower part of anterior arch. 33 to 36 scales in a longitudinal series, 8 to 10 from origin of dorsal to lateral line. Dorsal XV-XVI 8-10; last spine 3 head. Anal ITI 8-9; third spine ? head. Pectoral 2 head, not reaching anal. Caudal rounded or subtruncate. Caudal peduncle 1; to 14 as long as deep. Body with numerous small irregular dark spots. Seven specimens, up to 200 mm. long. 37. HAPLOCHROMIS ORTHOSTOMA, sp. n. (Text-fig. 9.) Pelmatochronis spekii (part.) Bouleng. Cat. Afr. Fish. 111. p. 417 (1915). Depth of body nearly equal to length of head, 22 in length of fish. Head 25 as long as broad; upper profile convex to above anterior part of eye, thence straight. Snout longer than diameter of eye, which is 5 in length of head, nearly equal to interorbital width or preorbital depth. Mouth very oblique; maxillary not quite reaching vertical from anterior edge of eye; lower jaw strongly projecting, more than 4 length of head; teeth conical, triserial, about 60 in outer series of upper Jaw. Cheek as deep Text-figure 9. Haplochromis orthostoma, sp. 1. as long, with 4 or 5 series of scales. 9 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. 33 scales in a longi- tudinal series, 6 or 7 from origin of dorsal to lateral line. Dorsal XV 9; last spine longest, 2 length of head. Anal IIE 8; third spine stronger than and nearly as long as last dorsal. Pectoral shorter than head, not reaching anal. Caudal rounded. Caudal peduncle 13 as long as deep. Greyish; a dark bar below anterior part of eye; pelvic fins blackish ; an ocellus on posterior part of anal. A single specimen, 115 mm. in total length, from Lake Salisbury. a CICHLID FISHES OF LAKE VICTORIA. 185 38. HAPLOCHROMIS XENOSTOMA, sp.n. (‘Text-fig. 10.) Paratilapia prognatha (part.) Bouleng. Cat. Afr. Fish. 111. p. 333. Depth of body 34 to 32 in length, length of head 2 to 3. Head 2% as long as broad; upper profile slightly concave. Snout 14 to 13 diameter of eye, which is 43 to 4% in length of head. greater than prworbital depth, nearly equal to depth of cheek ; interorbital width 42 in length of head. Mouth oblique; maxil- lary not extending to below eye; lower jaw very prominent, projecting upwards above end of snout; teeth in 3 series, inner tricuspid, outer conical or some bicuspid, 50 to 56 in upper jaw, Text-figure 10. Haplochromis xenostoma, sp. i. 4 series of scales on cheek. 9 gill-rakers on lower part of anterior arch. 33 scales in a longitudinal series, 6 from origin of dorsal to lateral line. Dorsal XVI 8-9; last spine longest, ? length of head; longest soft rays 2 length of head. Anal III 8; third spine stronger than and as long as last dorsal. Pectoral 3 to 7 length of head, not reaching anal; pelvics nearly reaching vent. Caudal subtruncate. Caudal peduncle 14 as long as deep. An opercular spot and a lateral band ; small dark spots on soft dorsal and caudal. Two specimens, 105 and 125 mm. long, one collected by Sir H. H. Johnston, the other from Entebbe (Degen). 39, HAPLOCHROMIS PELLEGRINI, sp. n. (Text-fig. 11.) Paratilapia prognatha (part.) Bouleng. Cat. Afr. Fish. in. p. 333, Depth of body 32 in length, length of head 2? to 2%. Head 25 to 23 as long as broad ; upper profile slightly concave. Snout 12 diameter of eye, which is 43 in length of head, greater than preorbital depth, equal to or a little less than depth of cheek, equal to interorbital width. Mouth oblique; maxillary not far short of vertical from anterior edge of eye; lower jaw strongly projecting, but not above end of snout ; teeth conical, 4 series in 186 MR. C. TATE REGAN ON THE upper jaw, 3 in lower, 60 in outer series of upper jaw. 3 or 4 series of scales on cheek. 8 or 9 gill-rakers on lower part of anterior arch. 32 scales in a longitudinal series, 6 or 7 from origin of dorsal to lateral line. Dorsal XIV 10; last spine longest, 3 or a little more than 3 length of head: longest soft rays 4 length of head. Anal III 9; third spine 2 or a little more than 2 length of head. Pectoral 2 length of head, extending to i Text-figure 11. Haplochromis pellegrini, sp. nu. above vent; pelvics reaching origin of anal. Caudal truncate. Caudal peduncle 14 as long as deep. An opercular spot; small dark spots on dorsal and caudal; one specimen with 3 ocelli on posterior part of anal. Two specimens, 125 and 130 mm. long, from Entebbe (Degen). Two others (75 mm.), from Entebbe, seem to belong to this species ; they have XV 9 dorsal rays. 40. HAPLOCHROMIS ARGENTEUS, sp. n. (Text-fig. 12.) Paratilapia longirostris (part.) Bouleng. Cat. Afr. Fish. iii. p. 332. Depth of body 3 to 33 in length, length of head 24 to 23. Head 3 times as long as broad; upper profile slightly concave. Snout 12 to 1? diameter of eye, which is about 44 in length of head, slightly greater than preorbital depth, equal to depth of cheek ; interorbital width about 5 in length of head. Mouth oblique; lower jaw strongly projecting; maxillary not nearly reaching vertical from anterior margin of eye ; 3 series of teeth in upper jaw, 2 in lower, anterior inner teeth tricuspid, outer mostly conical, some lateral ones bicuspid ; 46 to 60 in outer series of upper jaw, the anterior rather strong. 3 or 4 series of scales on cheek. 8 or 9 gill-rakers on lower part of anterior arch. Pharyn- geal teeth slender. 32 or 33 scales in a longitudinal series, 5 or 6 from origin of dorsal to lateral line. Dorsal XV 9-10; last CICHLID FISHES OF LAKE VICTORIA. 187 spine longest, a little less than + length of head. Anal III 9; third spine stronger than and about as long as last dorsal. Q » . = Pectoral 2 to $ length of head, reaching vent. Caudal truncate. Text-figure 12. Haplochromis argenteus, sp. 1. Caudal peduncle 14 as long as deep. Silvery; back darker ; soft dorsal and caudal with or without series of small dark spots. Two specimens, 115 and 140 mm. in total length, from Bunjako. 4 fie. to) 1. Hartocuromis Lonerrostris Hilgend., 1888. (Pl. IV. i) 7) Paratilapia longirostris (part.) Bouleng. Cat. Afr. Fish. 11. p. 332. Depth of body 34 to 33 in length, length of head 25 to 3. Head 23 to 24 as long as broad; upper profile straight or shghtly concave. Snout 12 to twice diameter of eye, which is 43 to 54 in length of head, equal to or a little more than preorbital depth, equal to or a little less than depth of cheek ; interorbital width 4! to42 in length of head. Mouth oblique; lower jaw projecting ; maxillary not extending to below eye; teeth conical (inner tricuspid in young), 3 or 4 series in upper jaw, 2 or 3 in lower, 50 to 60 in outer series of upper jaw. 3 or 4 series of scales on cheek. 10 or 11 gill-rakers on lower part of anterior arch. Pharyngeal teeth slender. 33 scales in a longitudinal series, 6 or 7 from origin of dorsal to lateral line. Dorsal XV-XVI 9-10; last spine longest, + to 3 length of head. Anal LIT 9; third spine stronger than and as long as last dorsal. Pectoral 2to 7 length of head, not reaching anal. Caudal truncate. Caudal peduncle nearly twice as long as deep. Silvery; back darker; an oper- cular spot ; soft dorsal and caudal sometimes spotted. Three specimens, 100 to 160 mm. long. 188 MR. C. TATE REGAN ON THE 42, HAPLOCHROMIS GRACILICAUDA, sp. n. (PI. IV. fig. 1.) Paratiiapia longirostris (part.) Bouleng. Cat. Afr, Fish. in. p. 332. Depth of body 4 to 4+ in length, length of head 3 to 33. Snout 1! to 12 diameter of eye, which is 4 to 4% in length of head, greater than preorbital depth ; interorbital width 47 in length of head. Mouth oblique; lower jaw projecting; maxillary not extending to below eye; teeth conical (cuspidate in young), tri- serial, 50 to 60 in outer series of upper jaw. 3 or 4 series of scales on cheek. 10 or 11 gill-rakers on lower part of anterior arch, Pharyngeal teeth slender. 33 scales in a longitudinal series, 6 from origin of dorsal to lateral line. Dorsal XVI 9-10; last spine 2 to + length of head. Anal III 8-9; third spine as long as or a little longer than last dorsal. Pectoral ? to + head, not reaching anal. Caudal slightly emarginate. Caudal peduncle 2 to 23 as long as deep. Silvery; back darker; an opercular spot; dorsal and caudal spotted. Two specimens, 105 and 150 mm. long, from Bunjako and Entebbe. 43. HAPLOCHROMIS XENODON Bouleng., 1911. Bayonia xenodonta Bouleng. Cat. Afr. Fish. iii. p. 488, fig. 338. Very near A. cinereus, distinguished principally by the dentition. Outer teeth few and large, compressed, with long anterior cusp directed inwards and posterior cusp very short or indistinct. Total length 90 mm. 44, HAPLOCHROMIS OBLIQUIDENS Hilgend., 1888. Hemitilapia bayoni Bouleng. Cat, Afr. Fish. i. p. 491, fig, 840. ? Hemitilapia materfamilias (Pellegrin, 1913), Bouleng. t. c. p. 341. Very near H. nuchisquamulatus. Teeth in 4 to 6 series, slender, distally expanded and compressed ; teeth of outer series enlarged, obliquely truncated. Total length 138 mm. 3. AsTATOREOCHROMIS Pellegrin, 1904. As Haplochronvis, but with 4 to 6 anal spines. L. Victoria. ASTATOREOCHROMIS ALLUAUDI Pellegrin, 1904. Haplochromis allwaudi Bouleng. Cat. Afr. Fish. ii. p. 305, fig. 206. Near H, gestri, especially distinguished by the increased number CICHLID FISHES OF LAKE VICTORIA. 189 of dorsal and anal spines and the large blunt pharyngeal teeth. Dorsal XVII-XIX 6-9. Anal IV—VI 6-9. Total length 155 mm. 4, MACROPLEURODUS, gen. n. Differs from Haplochromis in the dentition of the upper jaw, which has an outer series of enlarged teeth and several inner series of small teeth anteriorly and 3 or 4 series of enlarged teeth laterally, which are exposed when the mouth is shut. L. Victoria. Text-figure 13. Dentition of 1. Haplochromis sawvagii; 2. Macropleurodus bicolor ; 3. Hoplotilapia retrodens. MAcropLEevRoDUS BIcoLoR Bouleng., 1906. Paratilapia bicolor (part.) Bouleng. Cat. Afr. Fish. 11. p. 346, fig. 234. Paratilapia retrodens (part.) Bouleng. t. ¢. p. 347. Depth of body 22 to 32 in length, length of head 3 to 33. Snout decurved, as long as or a little longer than diameter of eye, which is greater than preorbital depth and 4 in length of head ; interorbital width 3 in head. Mouth wide; lower jaw a little shorter than upper; maxillary extending to below anterior edge of eye; 5 or 6 series of teeth in upper jaw, 4 or 5in lower. 3 or 4 series of scales on cheek. Gill-rakers short and stout, 7 or 8 on lower part ofanterior arch. Pharyngeal teeth small. 32 scales in a longitudinal series, 6 or 7 from origin of dorsal to lateral line. Dorsal XV 8-9; last spine ? length of head. Anal III 8-9. Pectoral as long as head, reaching anal. Caudal truncate. Caudal peduncle longer than deep. Olivaceous, with a faint dark lateral band or with irregular dark cross-bars extending on to vertical fins. Lake Victoria. Two specimens, 135 and 150 mm. long, from Bunjako. I take the figured specimen of P. bicolor as the type and so restrict the name to the species described above. 190 MR. C. TATE REGAN ON THE 5. Hoprormaria Hilgend., 1888. Cnestrostoma Regan, 1920. Differs from Haplochromis in the dentition. ‘Teeth small, conical, in rather broad bands, which are well-developed on the sides of the jaws, being formed of 3 or 4 series of teeth posteriorly. L. Victoria. HLoPLOrILAPIA RETRODENS Hilgend., 1888. Paratilapia retrodens Hilgend. Sitzb. Ges. Nat. Fr. Berlin, 1888, p. 76. Hemichromis retrodens Pfeff. Nhierw. O.-Afr. Fische, p. 19 (1896). Paratilapia polyodon Bouleng. Ann. Mus. Genova, (3) iv. 1909, p. 306, fig., and Cat. Afr. Fish. i11. p. 849, fig. 236. Paratilapia bicolor (part.) Bouleng. Cat. Afr. Fish. i. p. 346. Lisp of body 24 to 3 in length, length of head 3. Snout once to 14 diameter of eye, which is 34 to 4 in length of head, greater sheen preorbital depth, about equal to depth of ‘cheek: inter orbital width 8 to 34 in length of head. Jaws equal anteriorly ; maxillary extending to below anterior edge of eye; teeth in 5 to 8 series in upper jaw, 6 to 10 in lower. 4 or 5 series of scales on cheek. 8 gill-rakers on lower part of anterior arch. Pharyngeal teeth small. 33 scales in a longitudinal series, 7 or 8 from orig of dorsal to lateral line. Dorsal XV—X VI 9-11; last spine 4 to 2 length of head. Anal IIT 8-9. Pectoral as Jong as head, reaching anal. Caudal truncate. Caudal peduncle longer than deep. Traces of regular dark cross-bars and of a band along middle of side and another above lateral line, or irregular blackish cross-bars extending on to vertical fins. Four specimens, 110 to 170 mm. in total length. 6. PLatyraniopus Bouleng., 1906. Near Haplochromis, but jaws with very broad bands of small Text-figure 14. i 2 Dentition of 1. Haplochromis annectens ; 2. Platyteniodus degeni. CICHLID FISHES OF LAKE VICTORIA. i91 conical teeth, that of the upper jaw broader at the sides than in front. L. Victoria. PLATYTENIODUS DEGENI Bouleng., 1906. Platyteniodus degeni Bouleng. Cat. Afr, Fish. iii. p. 426, fig. 292. Very near /aplochromis annectens, differing especially in the dentition. Total length 140 mm, EXPLANATION OF THE PLATES. Prater |. Haplochromis altigenis. is Jif 3 serranoides. 5” WOT, ines, al, 53 dentex. fig. 2. 3 macrognathus. 5 We alee, » gracilicauda. fig. 2. _ longirostris. , ‘a ett Bali why a hich, el ee, Reshegs sit Hi Preaek: Tratgn sa grchcpa - mai Pe Peas: EN a | : i ete iyi t ¥ ie : Adis has edits Wie Sas ee : ie 1 Sa aasateee Basia. ge Mes: ‘n: Raee me ae i tla ue bs : PF i ie Ha suai ON DIRECT DEVELOPMENT IN A DROMIID CRAB. 193 10. Direct Development in a Dromiid Crab. By Srepaen K. Monteommry, B.A., B.Sc.* [Received February 10, 1922; Read March 7, 1922.] (Text-figures 1-3.) The specimen here described is in the collection of the British Museum (Natural History), and was submitted to me for examination by the the kindness of Dr. W. 'T. Calman, to whom thanks are also due for much assistance in studying it. The specimen was presented to the Museum many years ago by the late Dr. Henry Woodward. It is dried, and the only information regarding its origin is the locality, “‘ Bass’s Strait,” given on the label. The specimen is an adult female Petalomera lateralis (Gray), a species referred to by Haswell as “ very common ” in Australian seas. Under the abdomen are carried about 20 young in a post-larval stage, which, although possessing the same general form as the adult, differs from it in many details. A. similar occurrence is recorded by Miss Rathbun § in a paper to this Society in the case of the Oxyrhynch Vawioides serpulifera (Guérin); this is the only other available record of such a case among marine Brachyura. The adult agrees entirely with the descriptions, and the figure of Stimpson ||. It is figured here for comparison with the young. Only one stage of development has been observed in this ease, not two as in the case of WV. serpulifera. The carapace of the young crab is longer than broad, the measurements being about 175mm. long and 1:2 mm. broad, is flatly convex, much flatter than in the adult, and in the centre shows, under a high magnification, a minute reticulation. The regions are well marked, again in contrast to the adult, and there are numerous very fine hair-like spines on the upper surface. The front consists of three very prominent forwardly directed spines, each of which bears 5 to 8 pointed spinelets. The middle spine is directed forward and downward and, with the two lateral spines, forms a deep, almost V-shaped gutter between the supra-orbital borders, the opening of the V pointing forwards ‘(text-fig. 1). me : The supra-orbital border bears a similar prominent spine with accessory spinelets, which is directed forward and outward (text-fig. 1,@). Behind this spine the border runs almost directly backwards; there is a slight protuberance bearing a few spinelets * Communicated by Dr. W. T. Catman, F.R.S., F.Z.S. + For the transference of this species from Cryptodromia to Petalomera, cf. li. A. Borradaile, Ann. Mag. Nat. Hist. (7) x1. 1903, p. 3v0._ ~ W. A. Haswell, Catalogue of Aust. Stalk- and Sessile-eyed Crustacea, 1882, p. 139. § Mary J. Rathbun, Proc. Zool. Soc. 1914, p. 653, pl. u. figs. 9, 10. || W. Stimpson, Smithsonian Mise. Coll. 1907, xlix. pl. xx. fig. 3. » Proc. Zoot. Soc.—1922, No. XIII. 13 194 MR. S. K. MONTGOMERY ON DIRECT about the middle of its length (text-fig. 1, b), opposite which is the eyestalk. The border then turns almost at right angles iaterally and ends ina prominent tooth, covered with spinelets, forming the post-ocular tooth of the young specimen (text-fig. 1, c). On the lateral border, behind this, another strong tooth bearing spinelets occurs before the cervical groove is reached. Directly posterior to the groove is another similar spine, which is not so strongly marked. No subhepatie tooth can be discerned. In the adult there are two teeth on the antero-lateral border ; of these, the most anterior (text-fig. 2, c) must represent the post-ocular tooth of the young. The adult post-ocular tooth (text-fig. 2,6) may represent either the protuberance on the supra-orbital border of the young, opposite the eyestalk; in this case, the sub-hepatic tooth is a later development, and the prom- inent anterior supra-orbital tooth of the young (text-fig. 1, a) must become reduced to whe weakly developed supra-orbital tooth of the adult (text-fig. 2, d); or, alternatively, the eye in the adult may have passed even more anterior ly, so that the post-ocular tooth of the adult would represent the anterior supra-orbital tooth of the young, In this case, the adult supra-orbital tooth would be a new structure, developed from the spines in the region, and the sub-hepatic tooth may arise from the protuberance on the supra-orbital border opposite the eyestalk. The chele in the adult bear on the carpus a strongly marked tooth (text-fig. 2, e) on the external angle, a small bluntly rounded tooth (f) forming the extremity of the upper border, and between them a very strong forwardly directed tooth (9) forming the most anterior part of the carpus. Posteriorly, on the outer border of the car pus are two ridgelike teeth (/) standing close together. In the young (text-fig. 3), the first three teeth are represented by weakly developed protuberances, but no sign can be distinguished of the two ridgelike teeth. The hand of the young is smooth, except for a few small spines and hairs. The movable finger on its outer margin bears five sharp teeth, excluding that at the tip (text-fig. 3); there 1s an indication of a sixth tooth behind. The fixed finger bears four sharp teeth, with a fifth poorly developed, again excluding the apical tooth. In the adult, the teeth of the fingers are exactly similar, with the exception that they are slightly blunter, and that the posterior poorly developed teeth are slightly more prominent. The hand of the adult is grooved on both sides towards each finger to receive the thick tomentum which in life covers the carapace and the greater part of the limbs; this grooving, though typical in the Dr Beds as a whole, is not pr axon in NS young. The only description of the development of a Dromiid which is available is that of Dromia vulgaris by Cano* in 1894. According to him, the crab is hatched from the egg in a zoea stage, from which a metazoea develops; this is followed by a * G. Cano, Atti del Accad. Sci. Soc. Reale Napoli, ser, ii, vol. vi. 1894, No. 2 pls. i. & ii. DEVELOPMEN'’ IN A DROMIID CRAB. 19 megalopa stage 5mm. in length. Cano then deseribes a post-larval stage, hardly differing from the adult, except for the supra-orbital spines and the middle spine of the rostrum which are not yet developed. Text-figure J. S. HPA. dod. Petalomera lateralis Gray. Post-larval stage. X 25. Text-figure 2. P. lateralis; «dult female. x 3. The specimens here described look most: like the megalopa of Cano, but are considerably more advanced. The antennules (internal antenne of Cano) are exactly similar to those of the adult, consisting of a peduncle of three joints anda 196 ON DIRECT DEVELOPMENT IN A DROMIID CRAB. flagellum, which appears to be double, but owing to the brittleness of the dried specimen this could not be determined with certainty. The antenne consist of four joints, so far as can be made out, with a squame and a flagellum, the squame being rudimentary and similar to that in Cano’s figure. Text-figure 3. y = = ay ~ i ios he E.° la \ at I” S.KMSeL. Cheliped of post-Jarval stage. woe The abdomen is not extended and folded at the tip as in the megalopa figured by Cano, but is tucked beneath the body in typically Brachyurous manner. It bears four pairs of pleopods similar to those of the adult; the telson is also like that of the adult, there being simply a trace of the sixth pleopod visible between the 6th and 7th segments, and not, as in Cano’s megalopa, a distinct ramus on either side. The last two pairs of legs are distinctly cheliform, and not weakly subchelate, as in the megalopa. Sternal grooves cannot be distinguished in the young, nor can it be seen whether there is an epipodite on the cheliped. The young in this case are definitely post-larval, and the growth-changes which occur before reaching maturity are (@) the broadening of the carapace until it is broader than long, (6) the forward movement of the eye, (c) the effacing of the grooves between the regions of the carapace, (d) the thickening of the legs and chelipeds, and (e) the appearance of ridges on them. The thorns and spinules of the young are generally absent in the adult, which, on the other hand, in life was covered with a short tomentum, which is not seen in the young. It is very extraordinary that in a well-known species such as this, no record should hitherto have been made of the direct development of a post-larval stage from the egg. The possibility that the larvee have hatched from the egg, and passedthrough zoea, metazoea and megalopa stages clinging to the pleopods under the abdomen of the female, may be ruled out for the reason that it would be impossible for an adequate food-supply to be maintained. And, in fact, no record has been found of P. lateralis or NV. serpulifera having been captured with eggs, though this is a common occurrence with other Oxyrhynchs and Dromiids, numerous examples being recorded in the collections both of the ‘Sealark’* and the Ceylon Pearl Oyster Commission 7. * Mary J. Rathbun, Trans. Linn. Soc. London, (2) Zool. xiv. pt. 2, 1911, p. 191 sqq. + R. Douglas Laurie, Herdman’s Rep. Ceylon Pearl Oyster Fish. (Royal Soc., London) y. 1906, p. 349 sqq. ON THE LIFE-HISTORY OF CUCULUS CANORUS. 197 11. Notes on the Life-History of Cuculus canorus, with exhibition of eggs. By A. H. Evans, F.Z.8. | Received February 7, 1922: Read February 7, 1922. ] A further paper on the Common Cuckoo might well be con- sidered superfluous, if it were not for the fact that an examination of the literature on the subject shews that this is by no means the case; many of the details of the bird’s life-history are found still to need proof, and are merely reiterated from the pages of previous authors, who in place of such proof make assertions, which though highly probable, require confirmation. The eggs now exhibited have been collected during a period of more than twenty years, and give sure proof of the facts pointed out below. s Those of us who saw Mr. Edgar Chance’s admirable film, which was shown in this room some weeks ago, and redounded equally to his credit and that of his photographer, were initiated into the ways of an individual hen Cuckoo when depositing her eggs and were given a clue to the probable number of them; while a meeting of the British Ornithologists’ Club next month is to be devoted to the Cuculide as a Family: but the present paper deals neither with an individual and its idiosyneracies, nor with the whole of this world-wide Family, but with a species—our British Cuckoo—and its regular habits. [t is by no means easy to examine into the bird’s life-history, but the writer has had unusual good luck. In the comparatively small grounds of Histon Manor, near Cambridge, he and the owner, Mr. W. A. Harding, found that no fewer than five hen Cuckoos would lay their eggs in a single year, and that in one case the egg was so remarkable that recognition was instantaneous. Hardly less remarkable were specimens from the Cam, most of which are now exhibited by kind permission of Dr. Ticehurst and Mr. Bonhote, and are used to strengthen the evidence for the conclusions arrived at. Incidentally the abundance of hens at Histon tends to discredit the theory of polyandry. The first series of clutches passed round shews the same hen laying in the nests of the Greenfinch, Spotted Fiycatcher, and Pied Wagtail, and proves that she does not always choose the same foster parent. The egg, in this case peculiar and unmistakable, bears no resemblance to the others inthe nest. Such resemblance is the exception rather than the rule, except where, as in the ease of the Meadow-Pipit, the typical Cuckoo’s egg is more or less similar to that of its host, whose nests are plentiful and easily found. Several examples are exhibited. The second series affords strong corroboration ; the Cuckoo’s egg, though less peculiar, is in each case undoubtedly that of 198 MR. A. H. EVANS ON THE the same bird, which shews a partiality for Robins’ nests (3 out of 4). The third and fourth series are exhibited for the same purpose ; in the third the egg is again remarkable, in the fourth Hedge- Sparrows are the favourite hosts. The fifth series includes reddish or greyish eggs, of a type not uncommon along four or five miles of the Cam. ‘The preference here is for the nests of Sedge- or Reed-Warblers, and the simi- larity of coloration may shew the range in their case of kindred birds. The specimens were taken by Messrs. Richmond, Mills and myself: where two occurred in the same nest, they were clearly the produce of different hens. The single egg next to these was laid in a Pied-Wagtail’s nest after the young had flown, the sole instance in my experience. With regard to the choice ot foster parents, it should be noted that in the latter part of May and in June there is often little real choice, especially on the moors. The Cuckoo must take what she can get, and individual hens seem to have a restricted range, if we are to judge by the eggs. ‘They appear to be somewhat lazy birds, which follow the “line of least resistance,’ choosing Reed- and Sedge-Warblers for hosts on rivers and lakes, Hedge- Sparrows, Robins, Wagtails and the like in lanes and gardens, Pipits on moors and commons, with a decided preference for a small, comfortably lined nest. No one would look for Cuckoo’s eggs in a dense wood, or low down ina thick hedge or pile of sticks, where Hedge-Sparrows often breed. An examination of the dates on which the specimens exhibited were taken shews a still more important fact, namely, that the identical hen returns for several years in succession to the same grounds or even to the same copse. Here we have definite proof, as opposed to assumption. ‘This fact is not only made evident by the eggs from Histon Manor, but by the two very remarkable series from the collections of Dr. Norman Ticehurst and Mr. J. L. Bonhote. In the first series, where the Cuckoo’s egg resembles that of 1 Spotted Flycatcher, the foster parent was invariably a Pied- Wace The nests were in pollard Willows between Grant- chester and Waterbeach, a range of some eight miles. An egg, less richly coloured, also taken near Waterbeach, about ten years earlier, is sufficiently near the types to justify the conclusion that kindred birds occurred there before the dates on the rest of the series. The second series of large grey eggs with uniform small stipples came from a clay-pit near Wianerbeach. and, as in the last case, were found annually for several years in Sedge- or Reed-Warblers’ nests. Here, again, I exhibit a similar, though greener, egg taken many years later, which apparently shews kinship in the bird which laid it. Be this as it may, the dates of the main series prove the annual return of a hen bird to the very same spot. The number of eggs of the foster parent allowed to remain LIFE-HISTORY OF CUCULUS CANORUS. ; 199 in the nest varies greatly. As shown by Hancock (N.H. Trans. Northumb. and Durham, viii.) almost the whole elutch may be undisturbed. Although this article has been written to accompany an exhi- bition of eggs, and to bring out certain points which have received insufficient attention, without the intention of criticizing the work of others, it is impossible to end without mentioning the standard paper on Cuculus canorus by Dr. Hugene Rey, published at Leipzig in 1892 as No. 11 of Marshall’s ‘ Zoologische Vortrage. This laborious and elaborate piece of work may be taken to comprise all that was known up to that date of the bird’s nesting-habits, and contains long dissertations on the number of eggs laid in a year, the date of deposition, their size, weight, texture, and the composition of the shell, with lists of the foster parents, the number of their eegs suffered to remain in the nest during inenbation, and so forth. Tables and even graphs are given to elucidate the text, and, finally, full details of the immense number of specimens accumulated. Admirable, however, though this paper is, it fails in the points upon which I am now insisting. Sixteen out of Rey’s seventeen conclusions may be taken as correct, but when he states that most hen Cuckoos lay in the nests of definite species of foster-parents, except on rare occasions, his theory runs counter to experience. Of course the objection raised is only to his use of the word “most,” and would have failed if he had said “ not uncommonly.” Again, he tells us that the bird generally lays her eggs in the same locality or even spot, that is, that she does not wander far in the nesting-season. If he had only been able to add the expression ‘‘in successive years,” he would have anticipated my observations, made over a longer period than his. | i i a mi . Hie a) yeti) V iA ' ( ine +l i vee a 3 eh Ny Ha awe Sari Aste <4 Tiara ede: By, Gy i} 4 hh | nh i mips aye ay Ae SAAN TEES Pa ila mis ne THE SECRETARY ON ADDITIONS TO THE MENAGERIE. 201 EXHIBITIONS AND NOTICES. February 7th, 1922. Dr. A. Smiru Woopwarp, F.R.S., Vice-President, in the Chair. The Secretary read the following Report on the Additions to the Society’s Menagerie during the months ,of November and December, 1921 :— NovemMBeErR 1921. The registered additions to the Society’s Menagerie during the month of November were 221 in number. Of these 104 were acquired by presentation, 66 were deposited, 5 were received in exchange, 44 were purchased, and 2 were born in the Menagerie. The following may be specially mentioned :— 2 Brown Capuchins (Cebus fatuellus), 1 Squirrel-Monkey (Saimiris sciureus), 1 Hairy-rumped Agouti (Dasyprocta prym- nolopha), and 27 Rough-eyed Cayman (Caiman sclerops), from British Guiana, presented by Dr. G. M. Vevers on November 12th. 4 Lions (Felis leo), born in India, presented by the Jam Sahib of Nawanagar on November 7th. 1 Puma (Félis concolor), from Chaco in the Argentine, pre- sented by Arthur R. T. Woods, on November 11th. 1 Yellow-bellied Phalanger (Petaurus australis), from New South Wales, presented by William Huntsman, F.ZS., on November 21st. 1 Squirrel-like Phalanger (Petaurus sciureus), from New South Wales, presented by Mrs. Edward Goldsmith on November 28th. 1 Great Kangaroo (Jucropus giganteus), 2 Wombats (Phasco- lomys mitchelli), 2 Long-nosed Bandicoots (Perameles nasuta), from Australia, 1 Tasmanian Devil (Sarcophilus harrisi), from Tasmania, 1 Tawny Frogmouth (Podargus strigoides), from Australia, 1 Six-plumed Bird-of-Paradise (Parotia lawesi), 2 Magnificent Birds-of-Paradise (Diphyllodes hunsteini), from S.E. New Guinea; also 2 Magnificent Fruit-Pigeons (Megaloprepia magnifica), and 2 White-fronted Bronze-winged Pigeons (Henz- cophaps albifrons), from New Guinea, new to the Collection. Purchased on November 15th. ) 1 Long-tailed Roller (Coracias caudatus), from South Africa, presented by the Marquis ef Tavistock, F.Z.8., on November 16th. New to the Collection. 2 White-bellied Sea-Eagles (Haliaétus leucogaster), from Australia, presented by Alfred Ezra, O.B.E., F.ZS., on November 15th. 2 Australian Lung-Fish (Ceratodus forsteri), from Queensland, purchased on November 15th. Proc. Zoou. Soc.—1922, No. XIV. 14 202 THE SECRETARY ON ADDITIONS TO THE MENAGERIE. Decemper 1921. The registered additions to the Society’s Menagerie during the month of December were 79 in number. Of these 30 were acquired by presentation, 39 were deposited, and 10 were pur- chased, The following may be specially mentioned :— A pair of Lions (felis leo), from Kathiawar, India, presented by H.M. The King, on December 20th, 2 Northern Lynxes (Felis lyne isabellinus), from Dharmo Valley, Almora District, 10,000 ft., presented by P. Wyndham, CHC Bab ales: 2 very large Grooved Tortoises (Vestudo calcarata), from N. Nigeria, presented by The Emir of Katsina. Mr. A. H. Evans, F.Z.8., exhibited, and made remarks upon, a series of Cuckoos’ eggs taken near Cambridge. Lord Currrorp or CHaupieren, F.Z.S8., exhibited, and made remarks upon, a series of photographs of Vototheriam mitchelli. Miss L. E. Cuzpsman, F.E.S., gave an account of the position and funetion of the Siphon in the Amphibious Molluse, Ampul- laria vermiformis. Mr. G. C. Rosson, F.Z.8., exhibited, and made remarks upon, a series of models demonstrating the respiratory mechanism of Ampullaria vermiformis. February 21st, 1922. Prof. EK. W. MacBripz, D.Sc., LL.D., F.B.S., Vice-President, in the Chair. The Secrerary read the following Report on the Additions to the Society’s Menagerie during the month of January, 1922 :— The registered additions to the Society’s Menagerie during the month of January were 151 in number. Of these 52 were acquired by presentation, 76 were deposited, and 23 were purchased. The following may be specially mentioned :— 1 Macedonian Wolf (Canis lupus), from Maeedonia, presented on January 14th by Lt.-Col. F. L. Giles, R.E., British Repre- sentative Serbo-Bulgarian Boundary Commission, to whom it was given by King Boris of Bulgaria. ON THE INDIAN TORTOISE-BEETLE. 203 1 Dybowski’s Deer (Cervus hortulorum), from Manchuria, bred at Horsham, purchased on January 2nd. 11 Plumbeous Quails (Synacus plumbeus), from S.E. New Guinea, new to the Collection, deposited on J anuary 21st. 2 Angel-Fish (Petrophyllum scalare), new to the Collection, from Brazil, purchased on January 30th. The Srcrerary exhibited, and made remarks upon, a photo- graph of the Society’s Gardens taken from an aeroplane. Miss L. EK. Curzsman, F.E.S., exhibited, and made remarks upon, living specimens of the Amphibious Mollusc, Ampullaria vermiformis. Mr. G. ©. Rozson, F.Z.8., described in fuller detail the series of models demonstrating the respiratory mechanism of Ampul- laria vermiformis exhibited at the previous Scientific Meeting, and made remarks upon the respiratory mechanism of the Ampullariide. Mr. F. Martin Duncan, F.R.M.S., F.Z.S8., exhibited a series of cinematograph films he had recently taken in the Society's Gardens, showing the movements of Ampullaria vermiformis ; of the Freshwater Crab, Cardisoma armatum; and of the Hawk's bill Turtle, Chelonia imbricata. Miss L. KE. Currsman, F.E.S., described the habits, in captivity, of the Freshwater Crab, Cardisoma armatum. March 7th, 1922. Sir Stipvey F. Harmer, K.B.E., F.R.S., Vice-President, in the Chair. Morr Catan RnGan, FUR.S. RUZ.Ge exhibited, and made remarks upon, some living specimens of the Indian Tortoise- Beetle. Aspidomorpha sancte-crucis. ROLES ei A ae a aut OF THE - PROCEEDINGS | OF THE OF LONDON. Cvrk : WiNV 4 YU 7 wy Oy Pay Ee 1385 Texr-FIGURES. ‘JUNE 1922. PRINTED FOR THE SOCIETY, SOLD AT ITS HOUSE IN REGENT’S PARK, ; LONDON: PATERNOSTER ROW. GENERAL MEETINGS FOR SCIENTIFIC BUSINESS ZOOLOGICAL SOCIBTY Va ES: ASOnian D } ie = CONTAINING Paces 205 to 481, wivH 10 PLatTEs aNnp MESSRS. LONGMANS, GREEN, AND CO. 741922 ? ee [Price Twelve Shillings. } =. LIST OF. CONTENTS, 1922, Part II. (pp. 205-481). EXHIBITIONS AND NOTICES. The Secrerary. Report on Additions to the Society’s Menagerie during the month of ebriany l922 arrestee. My, ©. Tarr Reeay, M.A., F.R.S. Exhibition of a series of lantern-slides illustrating specimens of various blind fresh-water Fishes Ce er ee Mr. R, H. Burns, M.A., F.Z.S. Exhibition of specimens demonstrating the recessus orbitalis in Flat Fishes ee eC ec Ca Dr, L. Hocsny, M.A., F.Z.S. Exhibition of, and remarks upon, some small examples of metamorphosed Mexican Salamanders (Amblystoma tigrinum) ee ee eee ee tw we ce ae The Szcrerary. Report on Additions to the Society’s Menagerie during the month of March, 1922 a i eC er rr i er we The Secretary. Exhibition of some dressed skins of a Tree-Hyrax (Procavia valida) from Tanganyika Territory eC ee eer ry Prof. P. T. Fuyyy, Jxhibition of, and remarks upon, a cast of the skull of a Squalodont Whale i i The Sucrerary. An account of Mr. Loveridge’s experiences while watching “ Lions at their Kill” ee te ree mee te PE OH ew Dee ew ee wwe mH eee OL er BOB eesereesoreseceenrene Page 479 479 466 481 48] Contents continued on page 3 of Wrapper. THE NEMATODE PARASITES OF ELEPHANTS, 205 12. A Revision of the Nematode Parasites of Hlephants, with a description of four new species*. By M. Kwatit, SIDI, IO IP Lae SO OUEIES ie ISL) (From the Helminthological Department of the London School of Tropical Medicine.) [Received March 2, 1922: Read April 4, 1922. (Text-figures 1-71.) Since Cobbold’s adinirable work on the parasites of the Indian elephant in 1882, our knowledge of the Helminths harboured by this species of elephant has been considerably enlarged by Lieut.-Col. C. Lane, Railliet & Henry, and others. Little advance has been made regarding the parasites of the African elephant. ‘The new species recorded here, with one exception, are parasites of the latter animal. The material upon which this study is based was collected by Prof. Leiper in Uganda as far back as 1909 from two elephants shot by him. Im addition, specimens received from Dr. J. J. Simpson, West Africa, the Raymond Laboratories in India, and from Lieut.-Col. C. Lane, the lattex’s representing most of his type- species from the Indian elephant, were available for study and comparison. Leiperenia galebi was collected by Prof. Leiper from an Indian elephant that died in the London Zoological Society’s Gardens. Besides all this unique material, Prof. Leiper placed at my disposal his extremely valuable notes on the type-species in the Vienna and Berlin Natural History Museums. These proved to be of great assistance with regard to Belascaris lonchoptera Diesing, and Murshidia linstowt, sp. n. I feel deeply indebted to Prof. Leiper for his unrestricted generosity and ever-ready help both as regards material and information, without either of which this work could not have been done. The nematode parasites of the elephants are peculiar to these hosts, and it is an interesting fact that the parasites found in the African and Indian elephants never belong to the same species. Grammocephalus clathratus, which was formerly supposed to occur in botn species of elephants, has lately been shown by Lane to be represented in the Indian elephant by a separate species. Murshidia faleifera is another example. In this paper it is shown that the African elephant harbours two species of Mir- shidia, one of which apparently was mistaken for JMurshidia falcifera. * Communicated by Prof. R. T. Lerppr, D.Sc., M.D., F.Z.S, 7 In a preliminary note published in the Ann. & Mag, Nat. Hist. for Feb. 1922, brief diagnoses were given of two new genera and seven new species which are here fully described and illustrated. Proc. Zoou. Soc.—1922, No. XV. 15 206 DR. M. KHALIL ON THE NEMATODE There is some confusion regarding the correct zoological terminology of the species of elephants. For this reason the terms African and Indian are used throughout this work. NEMATHELMINTHES. Class NEMATODA. Super-family Ra4BDIASOIDEA Railliet, 1916. Family ATRACTID# Travassos, 1919. This family name was given by Travassos to include the following genera :—Atractis, Ozolaimus, Rodonia, Labiduris, Crossocephalus, Macracis, Cobboldina, and Cyrtosomum, to which IT add the genus Leiperenia. Travassos’ diagnosis of the family runs as follows :—Cisophagus with an anterior and a posterior bulb; viviparous, the female genital system is generally single; vulva is placed posteriorly or in common with the anus; parasites of vertebrates. Genus LereErEeNtA Khalil, 1922. Small nematodes just visible to the naked eye. The males are slightly smaller than the femalés. Mouth is surrounded by more than six lips. Cisophagus is divided into an anterior, short, and strongly muscular portion, to which the name pharynx is given. The posterior portion of the esophagus is about three times as long as the pharynx. The anterior end of the body is provided with a cuticular membranous expansion on either side. There is a large excretory vesicle. The excretory pore is raised on a apilla which is striated radially. Male: The caudal extremity of the male is curved. There are two unequal spicules and an accessory piece. There are four papillz on the tail. Female: Vivipareus, and the embryos reach an advanced stage of development im wtero. The vulva is placed in the posterior part of the body, separate, and a short distance in front of the anus. Type-species, Leiperenia leiper: (from the African elephant). LEIPERENIA LEIPERI Khalil, 1922. (Text-figs. 1-4.) Material.—The material consists of two males and three females, taken from Prof. Leiper’s collection of Elephant nematode parasites. Shape of body.—These nematodes are very small and just visible to the naked eye. The males are 3°8 mm. long. The females are slightly longer, being 3°9 mm. in length. The maximum diameter of the body is about the middle: it is*2 mm. in the male and-:21 mm. in the female. The body narrows slightly as if is traced towards the head end, which is truncated PARASITES OF ELEPHANTS. 207 in appearance. Posterior to the head a cuticular expansion is present on either side of the body. The female is practically straight ; its tail is very long and is gradually attenuated to a fine point. The caudal end of the male is bent towards the ventral surface, and its end is more rounded than in the female. Skin.—The cuticle is finely striated at intervals of -007 mm. throughout the whole length of the body. The cuticular expansions near the cephalic end of the body are placed on both lateral sides, and extend from the level of the pharynx to prac- tically the middle of the esophagus. They are *3 mm. in length and ‘037 mm. maximum breadth, and are striated. Text-figure 1. Leiperenia leiperi Khalil. Cephalic end. Mouth collar—Yhe mouth collar is very short and has a rounded shape. It is ‘012 mm. long and -06 mm. in diameter. The mouth-opening is practically circular and is surrounded by ten small lips. The lip in the mid-ventral line and that in the mid-dorsal line are the largest and broadest. The lips placed on either side of these lines are smaller and rounded. There is no mouth capsule. Pharynx.—This is a short muscular canal. Its musculature is more closely set and of a finer texture than that in the succeeding esophagus. It is also distinctly demarcated from it by a deep groove. For these reasons | named that specialized part of the cesophagus the pharynx. The shape of the pharynx is cylindrical, with a small bulging in the middle. It is ‘083 mm. long and "06 mm. maximum diameter. From its cephalie end project conical processes ; their apices are directed antero-laterally. The number of these processes and their structure can only be ascertained under the oil immersion lens. They are eight in number surrounding the mouth-opening, and they are muscular La 208 DR. M. KHALIL ON THE NEMATODE processes from the pharynx. The lumen of the pharynx is continuous with that of the esophagus. (Hsophagus.—Vhe csophagus is long and thin. In some specimens it pursues a wavy course. Its posterior end is a little swollen. It is ‘38 mm. long and ‘073 mm. maximum diameter. Text-figure 2. Leiperenia leiperi Khalil. Anterior part of the body. Chyle intestine.—‘vhe intestine is irreguiarly bent in its course. Near its commencement its wall is surrounded by a clear refractile band, the nature of which is unknown. The cellular wall is slightly pigmented. The rectum is a short canal; in most cases its structure is obscured by the greatly distended genital organs of the female. Hecretory system.—The large excretory vesicle receives two excretory ducts, one from the lower part of the body and one from the head end. The excretory pore is wide and is placed "92 mm. from the cephalic end. It is placed on a raised papilla, striated radially round the pore. Nerve collar.—TVhe nerve collar surrounds the thinnest part of the cesophagus *22 mm. from the head end. Genital organs.—Male: The irregularly convoluted testis reaches within -8 mm. from the head. The cement gland is long, and the ejaculatory duct opens into the ventral aspect of the cloaca. Female: In the mature female the genital tract is so distended with embryos that the details of this system are practically impossible to make out. By the help of an immature specimen the general features of the organs can be recognized. There is only one ovary and one uterus, placed in the axis of the body. The vagina is short, and opens in the posterior part of the body +1 mm. cephalad of the anus. Spicules.—There ave two unequal spicules and an accessory PARASITES OF ELEPHANTS, 209 piece. The longer spicule is-3 mm.in length. It begins close to the dorsal surface, and lies parallel to 1t for about two-thirds of its course. Then it bends ventrally to pass into the cloacal canal anterior to the accessory piece. he shorter spicule is*19 mim. in length. - It lies by the side of the terminal portion of the longer spicule. Its thick cephalic end is bent ventrally, and is thus seen in the concavity of the longer spicule. The accessory piece is ‘09 mm. long. Its cephalic extremity is large and rounded; its end is attenuated to a sharp point. Text-figure 3. 100 Leiperenia leiperi Khalil. Caudal end of male. The tail.—The male tail is curved, forming a semicircle. It is *38 mm. long. Its extremity is rounded; -16 mm. from the tip of the tail there are four papille. Two of these are placed near the mid-ventral line, one slightly caudal to the other. The other two are placed one on either side of the tail. These are all simple papille and project very little above the surface. The female tail is very much longer and thinner than that of the male, being *7 mm. in length; its termination is very fine. There are four papille on either side of the anus. No other papille could be detected on the female tail, The Embryos.—The females are viviparous ; the embryos reach an advanced stage of development while they are still in the uterus. Six embryos can be seen in the uterus at one time. They distend that organ enormously, and make identification of the neighbouring structures difficult, 1f not impossible. Some of these embryos measured °52 mm. in length, and -06 mm. in 210 DR. M. KHALIL ON THE NEMATODE diameter. The pharynx, cesophagus, and a straight gut can be easily recognized in them. Habitat. The intestine of the African elephant (Uganda). Text-figure 4. Leiperenia leiperi Khalil. Tail end of female. LEIPERENTA GALEBI Khalil, 1922. (Text-figs. 5-7.) Material.—The material consisted of ten specimens collected by Prof. Leiper from an Indian elephant that died in the London Zoological Society’s Gardens. They were the only Helminth parasites found in that elephant. Shape of body.—These nematodes are just visible to the naked eye. The male is smaller than the female. It is 3°25 mm. long, while the female is 3°8 mm. The maximum diameter is attained near the middle of the body. It is *17 mm. in the male and ‘18 mm. in the female. The head end is truncated, while the caudal is attenuated to a fine end, more delicate in the female than in the male. The caudal end of the male is curved ventrally. The head end of the body has a lateral cuticular expansion on either side. Skin.—The cuticle is striated throughout its length, except for a short distance at the cephalic extremity. The striations are at regular distances of °008 mm. apart. The cuticle is expanded in a membranous fashion on either side of the cephalic end of the body fora distance of -6 mm. It is striated in its greater extent, the part corresponding in level with the pharynx is not striated. The cuticular expansion is ‘023 mm. in breadth. Mouth collar—The mouth collar is distinctly separated from PARASITES OF ELEPHANTS. ANAL the body bya groove. It is 0:18 mm. long and ‘08 mm. in diameter. It has a rounded margin. The oral opening is circular and is surrounded by ten rounded lips. The one in the mid-ventral line and also that in the mid-dorsal line are broader and divided into two by a superficial depression. There is no oral cavity. Text-figure 5. Leiperenia galebi Khalil. Cephalic end. Pharynx.—The short cylindrical muscular pharynx is "14 mm. in length and-] mm. maximum diameter. It is thus much longer and broader than in J. leiperi. From its cephalic end spring eight cone-shaped processes that surround the mouth- opening. The fine apices of these processes correspond more or less to the interlabial depressions. (sophagus.—The thin muscular cesophagus is irregularly bent. It is °39 mm. long in the male and -4 mm. in the female. Its maximum diameter is at posterior end, and measures in the male -06 mm. and in the female ‘07 mm. Chyle intestine —The intestine pursues an almost straight course. The cells of the intestinal wall are slightly pigmented. The short rectum is indistinctly marked from the intestine. Exeretory system.—There is a large excretory vesicle similar to that described inthe type-species. The excretory pore is placed on a large raised papilla 1-2 mm. from the head end. The papilla is radially striated round the pore. 212 DR. M. KHALIL ON THE NEMATODE Nerve collar.—The small nerve collar surrounds the thinnest part of the cesophagus ‘25 mm. from the head end. Genital organs.—Male: The convoluted testis gives rise to the ejaculatory duct. This pierces the enormously elongated cement gland to end in the cloaca. Female: The female is viviparous; the embryos reach an advanced stage of maturity, and so distend the uterus as to make the examination of other structures difficult. The convoluted ovary 1s single, and leads to the enormously dilated uterus. The short vagina opens on the surface of the body :06 mm. in front of the anus. The convolutions of the ovary reach within 52 mm. of the head end. Text-figure 6. Leiperenia galebi Khalil. Tail end of male. Spicules.—There are two unequal spicules and an accessory piece. The longer spicule is -25 mm. long; thus it is shorter than that of the type-species. It lies in the axis of the body, midway between the ventral aud dorsal surfaces. Near its lower end it bends boldly ventrally to pass into the cloaca. The shorter spicule is °13 mm. jong. It is very slightly curved, lying by the side of the terminal portion of the longer spicule. The accessory piece is ‘13 mm. long. It has a thick bulbous end, and is gradually attenuated to end in a sharp point buried in the posterior lip of the cloacal opening. It lies practically across the axis of the body. The tail.—The male tail is gracefully curved and is *43 mm. in length. itsendis biuntly rounded. ‘There area group of four papille similar to those in ZL. lewperi. They are situated -17 mm. from the tip of the tail. Two papille are in the middle line, one PARASIVES OF ELEPHANTS. 213 slightly anterior to the other. Of the other two, one is placed on either side of the body at the same level. The female tail is short and conical, unlike that of Z. leipert. Its tip is pointed. It is 53 mm. long. The anal opening is not vaised or surrounded by prominent lips. On either side there are four papille. Text-figure 7. Leiperenia galebi Khalil. Tail end of female. Embryos.—The uterus is crowded with these giant embryos. They attain -48 mm. in length. The various parts of the digestive tract can be easily made out in them. Habitat. Intestine of Indian elephant (India; died in the London Zoological Gardens). Discussion. This genus has some points in common with the genus dA tractis. The latter was made by Dujardin to include Ascaris dactyluris of Rudolphi, 1819, but was not defined. The genus was defined by yon Linstow in 1902. From von Linstow’s description, together with a detailed account of the type-species by Railliet and Henry, Leiperenia can be easily separated by the following characters :— In Atractis there are six lips surrounding the mouth. There are no cuticular membranous expansions in the anterior part of the body. The esophagus is divided into two portions, of which the anterior is the longer. ‘The number of the conical processes is limited to six. There isa cuticular expansion along the posterior two-thirds of the body. he species of Atractis are parasitic in Land and Freshwater Tortoises and Reptilia. 214 DR. M. KHALIL ON THE NEMATODE The two species of Leiperenia differ mainly in the length of the spicules and accessory piece, and particularly in the shape of the female tail, which is very long and narrow in JZ. leiperi and stumpy and broad in Z. galebi. Some of the genera included in the family Atractide are credited with the ability of reproduction and maturation in the same host. They are found in tremendous numbers in the host. Super-family Ascarorpe4 Raill. & Henry, 1915. Family AsScARID# Cobbold, 1864. Sub-family AscArina Travassos, 1913. Genus Brxascaris Leiper, 1907. BELASCARIS LONCHOPTERA (Diesing, 1851) Leiper, 1911. Strongylus elephanti Rud., 1819. Ascaris lonchoptera Diesing, 1851. Belascaris lonchoptera Leiper, 1911 (unpublished notes). This was the first Helminth parasite recorded from the elephant. It was listed by Rudolphi in his ‘Synopsis Entozoorum’ under doubtful strong yles. The specimens were collected from the bile- ducts of an Indian elephant and deposited in the Museum of Vienna. In 1847, Jackson recorded in the ‘ Descriptive Catalogue of Boston Medical Improvement Society’ the presence of Ascaridz and Flukes in the Indian elephant. In 1851, Diesing described the doubtful specimens of Rudolphi, and found them fa be female Ascaride and not Strongyles. He named them Ascaris lonchoptera. He also referred to the Ascaride of the Boston Catalogue as the same species. In 1882, Cobbold expressed a doubt if such an illustrious man as Rudolphi would have mistaken an Ascaris for a Strongyle. In the same year Drasche redescribed and figured A. lonchoptera from the Vienna collection, and left no doubt as to its Ascarid nature, thus confirming Diesing’s observations. Leiper, in unpublished notes to which I have had access, records the results of his examination of the same specimens in the Vienna Museum, and coneluded that they belonged to the genus Belascaris (Leiper, 1907). Although a good deal of attention has been paid lately to the elephant parasites, especially in India, no Ascaridze have been reported. So far, the male of Lelascaris lonchoptera remains undescribed. Habitat. Bile-ducts of the Indian elephant. PARASITES OF ELEPHANTS. 215 Super-family Sprrororpea Raill. & Henry, 1915 (emend. Hall, 1916). Family Sprrurip& Oerley, 1885. Sub-family Sprrurin# Railliet, 1915, Genus PArABroneMA Baylis, 1921. Spiruride: Polymyarian worms, having the mouth bordered by paired lateral lips, external to which there are a dorsal and a ventral shield of cuticle. Each lip has one large median lateral papilla and a pair of small sublateral papille. Of the dorsal and ventral shields, each carries a pair of larger papille, situated some distance behind the extremity of the head. The cuticle of the head is thick, and folded in a complicated manner so as to form a circlet of six horseshoe-shaped auricular appendages, of which two are lateral, two subventral, and two subdorsal. The body is rather slender, tapering rather more in front than behind. The cuticle is transversely striated. The buccal cavity a short distance from the oral aperture is transversely elongated, then passes into a long cylindrical cuticular tube. The cesophagus consists of a short, narrow anterior portion and a long, somewhat wider posterior portion. The anterior portion is surrounded by the nerve-ring. The tail of the male is coiled ventrally into a spiral. The spicules are markedly unequal. A somewhat triangular accessory piece is present. ‘here are six pairs of papillee, four pre-anal and two post-anal. There is also a large median double papilla immediately in front of the cloacal opening. The female is considerably larger than the male. The tail is short and curved towards the dims sal side. The vulva is in the region of the end of the cesophagus. The vagina is long and narrow. It has a curious U-shaped bend in “Es course, a “short distance from the vulva. The two uteri are parallel, running backwards at first, one of them returning towards the anterior end of the body. The female is viviparous. The uterus contains immense numbers of embryos. Type-species, Parabronema indicum. PARABRONEMA INDICUM Baylis, 1921. The male‘is 7°9 mm. and the female is 13 mm. in length. The vulva is situated a little caudad of the end of the esophagus. The longer spicule of the male is 0°93 mm. long and the shorter is 0°39 mm. The tail of the male is 0°17 mm. and that of the female is 0°32 mm. in length. Habitat. Stomach-wall of the Indian elephant (India). 216 DR. M. KHALIL ON THE NEMATODE PARABRONEMA AFRICANA Baylis, 1921. This species was apparently mistaken by Baird for female Grammocephalus clathratus. It was isolated and described from the original material by Baylis in 1921, but Leiper in his notes in 1911 recorded and drew this species from material in the Berlin Museum, where it was kept under the name Ascaris lonchoptera. He recognized this mistake in diagnosis, and added that the same species is included wrongly by Baird in his type of G. clathratus in the Natural History Museum of London. The male is 40 mm. and the female is 57 mm. in length. The lips have a median inward projection on their opposed edges. Vulva is placed just caudad or cephalad of the end of the esophagus. The left spicule is 3°15 mm. long and the right is 0°68 mm. The male tail is 0°85 mm. and the femalé tail is 0°52 mm. in length. Habitat. Stomach-wall of the African elephant (died in London). PARABRONEMA SMIrHit Cobbeld, 1882. (Text-figs. 8-12.) Filaria smithii Cobbold, 1882 (mot Filaria smitht Sambon, 1907, from the Grouse Lagopus scoticus). Spiroptera smithi Raill., Henry & Bauche, 1914. Parabronema smithi Baylis, 1921. Source of Material—The material consisted of one male and three females sent from the Raymond Laboratories, Calcutta. They were collected from an Indian elephant. Shape of body.—Very small nematodes. The females much larger than the males The male is 4:1 mm. long and the female 8 mm. The maximum diameter of the body is about the middle of the worm, being 0-2 mm. in the male and -26 mm. in the female. The female is always curved, the tail directed dorsally. The tail of the male is spirally coiled ventrally. Skin.—The cuticle is very finely striated throughout the length of the body. The head.—The cuticle surrounding the head end is raised from the surface of the body, forming a complicated design. Its caudal margin surrounds the circumference of the body 0-083 mm. from the head end. It forms six auricular appendages, two lateral, two subventral, and two subdorsal. Hach auricle is open towards the head. ‘Their posterior margin is nearly flat, unlike the rounded appearance in P. indicum or the angular appearance in P. africana. Hach of the two lateral lips bears three papille, a large one near the centre and two smaller ones near the margin on either side. ‘The cuticular shields lower down bear four papille. The terminal extremity of the head is 0°03 mm, in diameter. Cervical papilie.—The cervical papille are placed 0°209 mm. from the head end. Digestive system.—There is a buccal cavity just behind the oral PARASITES OF ELEPHANTS. PATE aperture. This leads into a long cuticular tube leading into the esophagus. The length of that part of the digestive system is ‘limm. The wsophagus consists of two portions. The cephalic part is thin and practically straight. It is 0-14 mm. long in the male and 0:16 mm, in the female. This portion is surrounded by the nerve collar. The caudal part of the esophagus is very long and broad in diameter. It pursues a crooked course, being acutely bent from side to side. It is 1-06 mm. long in the male and 1-09 in the female. Its caudal end projects into the intestine and is protected by a trilobed valve. Text-figure 8. Text-figure 9. OF 50p Parabronema smithii Cobbold. Parabronema smithii Cobbold. Ventral view of head. Lateral view of head, Chyle intestine.—The intestine pursues a straight course and is lightly pigmented The rectum is narrow and is constricted from the rest of the intestine. Exeretory pore.—This could only be made out in the female, where it is 0°25 mm. from the head end. Nerve collar.—Vhe thick nerve collar surrounds the cephalic portion of the esophagus. It is placed 0°22 mm. from the head end in the male and 0°23 mm. in the female. Genital organs.—Male: The coils of the convoluted testes lie more in the longitudinal axis of the body. The long cement gland curves in the caudal end of the body to end in the ventral aspect of the cloaca. Female: The vulva lies a little cephalad of the end of the cesophagus in the mid-ventral line. At a distance of 0-1 mm, from the vulval opening, the caudally directed vagina produces a distinct U-shaped kink, and then resumes its course caudally, where it is directly joined by the two uteri. The uteri proceed 218 DR. M. KHALIL ON THE NEMATODE caudally. They are packed with embryos and no signs of eggs. The ovaries are confined mostly to the caudal half of the body. Spicules.—There are two unequal spicules. The long spicule is 0°56 mm. long. It lies to the left of the short spicule, which is 0:21 mm. long. The latter lies parallel with the lower end of the long spicule, and curves ventrally towards the cloacal opening. There is a small triangular accessory piece 0-03 mm. in length. Text-figure 10. Text-figure 11. Parabronema smithii Cobbold. Female tail. Parabronema smithii Cobboid, Anterior part of the body. Caudal papille.—The male tail is provided with four pairs of pre-anal papille lying on either side of the mid-ventral line. Each is provided with a long core and a single termination. Two pairs of post-anal papille could be made out, one close to the cloaca and the other midway between the cloaca and the tip of the tail. Tail.—The male tail is spirally coiled ventrally. It.is shorter than that of the female and its tip is rounded. It is 0:18 mm. in length. PARASITES OF ELEPHANTS. 219 The female tail is slightly bent dorsally. It narrows gradually to its tip, which is slightly rounded. There is apparently no papillz on the female tail. The length of the tail is 0-36 mm. Text-figure 12. Parabronema smithii Cobbold. Male taii- Embryos.—The female is viviparous. The embryos are so closely packed that it was impossible to arrive at an approximate estimate of their length. Habitat. Coats of the stomach of the Indian elephant (India). Discussion. Baylis described fully P. indiewm and P. africana. He was not able to find material containing P. smithii, which has only been alluded to once since it was described by Cobbold. There has been some doubt as to the identity of P. indicum and P. smithii, as Cobbold’s description was not full. For this reason the latter species is described here in full. It is certainly a different species from VP. indicum, although found in the same host. It isa much smaller worm, and the measurements of its various organs are likewise shorter than the corresponding figures of P. indicum. The shape of the auricular appendages is different from that alluded to above. The African species is of very great length, and its auricular appendages of the cuticular thickening of the head are quite distinctive. All its other dimensions are correspondingly longer than in the other species. MIcROFILARIA. Evans and Rennie record the presence of a microfilaria in blood taken from an elephant during the daytime. It measured 1804 in length and 6 in breadth. It is possible that it was really a microfilaria, but I suggest that it is more probable that the 220 DR. M. KHALIL ON THE NEMATODE viviparous species of Parabronema living in the stomach-wall void their embryos into the lymphatics or blood and not into the intestinal canal. Super-family SrroncyLorpea Wienland, 1858. Family StRONGYLID# Cobbold, 1864. Sub-family StroneyvLin# Railliet, 1893. Genus DecrusiA Lane, 1914, Fairly stout worms with a cup-shaped mouth capsule. There is a marked duct of the dorsal cesophageal gland running along the mid-dorsal line of the mouth capsule. Male: The dorsal ray is undivided except at its extreme tip. The spicules are equal and similar, fine-pointed, and without accessory piece. Female: The tail is blunt, the vulva is in the caudal third, the uteri are divergent, the caudad-running uterus turning imme- diately cephalad. Type-species, Decrusia additiciia, Decrusia ADDITICTIA Raill., Henry & Bauche, 1914. Strongylus additictus Raill., Henry & Bauche, 1914. Decrusia decrusi Lane, 1914. Decrusia additicia Raill., Henry & Bauche, 1915. Decrusia additicia Lane, 1915. Strongylus additicius Ihle, 1919. This species has the characteristics of the genus. The mouth capsule is tilted dorsally. The external leaf-crown consists of 140-150 rays. Male is 14 mm. long and 1:0 mm. maximum diameter. The spicules are 2'4mm. long. Female is 15mm. long and 1-4mm. maximum diameter. The tail is 0:3 mm, long, blunt at the tip. The vulva is 73mm. from the tip of the tail. The ova are 75 long by 40 broad. Thle thinks that this genus is an unjustifiable division of the genus Strongylus. Although there are points of similarity between the two genera, the dorsal tilt of the mouth-opening, the long dorsal ray of the bursa, and the absence of an accessory piece make this genus justifiable. In the genus Strongylus itself, it was suggested by Looss that S. vulgaris, which is more allied to S. equinus the type-species than Decrusia additictia, might eventually prove to belong to another genus. Habitat. Warge intestines of the Indian elephant (India). Genus Mursuipra Lane, 1914. Lane’s diagnosis of the genus runs as follows :—“ Fairly slender worms tapering towards the head end. The male is widest just PARASITES OF ELEPHANTS, 22h cephalad of the bursa, the female tapers caudad to a fine point. The head is discoid in shape. “The oral aperture is hounded by two slightly marked lateral lips, each carrying a sessile lateral and two prominent submedian papilla. The oral capsule is roughly cylindrical, and circular in cross-section, while the wall varies in shape in different parts. Since this is thicker caudad and thinner cephalad on its dorsal and ventral than on its lateral aspects, its cavity is, atits cephalad end, wider dorso-ventirad than latervo-laterad. The cuticle bounding the mouth cavity is applied to the mouth capsule only over its caudad portion. Along a cir- cumferential line, which is further cephalad on the dorsal and ventral than on the lateral walls, the cuticle turns axo-cephalad and splits into rays having a similar direction ; and since further these rays are shorter dorsally and ventrally than laterally, the result is that their points which form the boundary of the oral aperture, do not describe a circle but produce a dorso-ventral slit. The cesophagus is fairly short, widening caudad of the nerve- collar. Its caudal end is guarded by three intestinal valves. The boundaries of the cells of the chyle intestine are more marked than usual. The cephalic and the cervical glands are large and well developed. The lateral cervical papille are long, slender, and project somewhat cephalad. “The male expands ventrad just cephalad of the bursa. The number of bursal rays is as usual. The externo-dorsal rays lie close to the lateral rays. The dorsal ray is deeply cleft in the mid-line, and each half is composed of three subdivisions. There appear to be no pre-bursal papillae. The spicules are equal and similar, and the head of each is bent like the head of a golf- driver, while the shaft may be straight or may be bent near the oint. There is a colourless thickening of the cuticle in the position of the accessory piece, which is §-shaped in lateral optical section. The other male organs are as usual. “The female has a long and pointed tail, the vulva lying shortly cephalad of theanus. The vagina runs cephalad and divides into two uteri, which have the same direction, are provided with ovejectors, run parallel with one another, and end abruptly in the two cephalad-running ovaries.” I may add to this that an accessory piece is present, and although the male is broad just above the bursa, this is not always the greatest diameter of the body. Type-species, Murshidia murshida. MursHIDIA MURSHIDA Lane, 1914. This nematode has the general characteristics of the genus. The external leaf-crown consists of sixty leaflets. The vulva has a cuticular prominence both cephalad and caudad of it. (For measurements vide Table I., p. 228.) The lateral and dorsal rays of the bursa have a rugged outline and a bulbous origin. Habitat. Ceecum of Indian elephant (India). Proc. Zoou. Soc.—1922, No. XVI. 16 Die, DR. M. KHALIL ON THE NEMATODE MURSHIDIA FALCIFERA Cobbold, 1882. Strongylus falcifer Cobbold, 1882. Nematode No. 3. Evans & Rennie, 1910. Strongylus falcifer Mitter, 1912. Cylicostomum falciferum Raillet, Henry & Bauche, 1914. Murshidia faleifera Lane, 1914. The oral aperture is bounded by eighty rays. The bursal rays have not the rugged outline seen in d/, murshida. (For measure- ments see Table I.) Mitter in 1912 gave a short description of this nematode with figures. He represented the bursa with one ray missing, which is either the externo-dorsal or one of the terminal branches of the dorsal ray. In either case this may be attributed to inaccuracy. Habitat. Intestine of Indian elephant (India). MURSHIDIA LINSTOWI, sp. n. (Text-figs. 13 & 14.) Sclerostomum rectum von Linstow, 1907, from the African elephant (not Strongylus rectus von Linstow, 1906, from Dolichotis patagonica). Cylicostomum rectum Gedoelst, 1916. Murshidia recta Thle, 1919. This species was first described by von Linstow in 1907 as Sclerostomum rectum. The genus Sclerostomum Rudolphi, 1809, Text-figure 13, Text-figure 14. Murshidia linstowit, sp. n. Lateral view of male bursa, Murshidia linstowi, sp. 0. Cephalic end of body. By kind permission of Prof. Leiper. is a synonym of Strongylus Goeze, 1782, both having as their type-species S. equinus Mueller, 1780, Thus Strongylus has PARASITES OF ELEPHANTS. J23 precedence. Von Linstow described in 1906 Strongylus rectus from Dolichotis patagonica. According to Article 36 of the International Law of Zoological Nomenciature, the specific name rectum, 1907, must be rejected and can never be used again. This is in complete agreement with the views expressed by Lane. He rightly was unable to arrive at the conclusion that the nematode described by von Linstow belongs to the genus Murshidia, because the original description and figures are very unsatisfactory. Only the figure of the male bursa.shows any marked resemblance to Murshidia. I have been fortunate in having access to a series of camera- lucida drawings of von Linstow’s original material in Berlin Natural History Museum, made by Prof. Leiper and very generously placed at my disposal. Unfortunately, the exact magnification was not noted, and therefore von Linstow’s measure- ments are relied upon, as they correspond in proportion to the camera-lucida drawings. The mouth capsule is wide, surrounded by a chitinous ring of the shape characteristic of the genus. The cesophagus is short and thick. The excretory pore is placed a considerable distance eaudad of the cesophagus. The vulva is surrounded by raised cuticle, specially cephalad. The male bursa is short and the lateral rays are not rugged in outline. The spicules are straight and sharply bent near their terminations, which are bulbous in shape. (For details of measurements see Table I.) Habitat. From the African elephant (Cameroon). Mursuipia HADIA Khalil, 1922. (Text-figs. 15-21.) Material —The material consisted of one male and four females, selected from Prof. Leiper’s collection of Elephant nematodes. Shape of body.—The body of the male is straight, the female tail is slightly bent ventrally. The male is 18-5 mm. long and the female is 24mm. long. The maximum diameter of both sexes is about the middle, being *67 mm. in the male and °82 mm. in the female. The antero-posterior diameter of the male body just cephalad of the bursa is 55mm. This is a little smaller than the diameter at the middle of the body. Skin.—The cuticle is finely striated at intervals of 009 mm. Mouth collar.—This is well developed and rounded in out- line. It is°05mm. long and ‘2mm. in diameter. It is distinctly separated from the rest of the body by a deep groove. External leaf-crown.—Arising from the inner surface of the mouth collar is the external leaf-crown, consisting of forty rays. These rays are longer laterally than ventrally or dorsally. Each ray is narrow and gradually tapers to a point. Head papille.—There are the usual four submedian head papille. They project freely above the head. Each is sur- mounted by a small knob which, as already mentioned, represents 16* DR. M. KHALIL ON THE NEMATODE bo i) = Text-figure 15, a aS sete == —— SO Murshidia hadia Khalil. Head. Text-figure 16. \\ /] Ta. 3 v5, Murshidia hadia Khalil. Cephalic end of body. — _—— PARASITES OF ELEPHANTS. 225 the outlet of the cephalic glands. The two lateral head papille do not project above the surface, and are marked by a slight depression. They are conical in shape with a wide base. Cervical papille.—The cervical papille are very thin, long, and point cephalad. They are placed 1-4 mm. in the female and 1:14 mm. in the male from the head end. Mouth capsule.—The mouth cavity is very large and almost globular in shape. It is ‘12mm. long and -15 mm. maximum diameter. The oral opening is ‘07 in diameter. The mouth capsule is a massive chitinous structure ‘(07 mm. in length and ‘025mm. in thickness. It is slightly curved inwards. A fine chitinous process of the mouth capsule bounds the oral cavity in the region of the mouth collar. The broad floor of the oral cavity is free from any teeth. Text-figure 17. Text-figure 18. Gan Murshidia hadia Khalil. Ovum. Murshidia hadia Khalil. Tail end of female. (Hsophagus.-—There is a cone-shaped cesophageal funnel. It is ‘06 mm. in length and -1 mm. at its junction with the oral cavity. The cesophagus is broad for its length. It is °85mm. long in the male and ‘9mm. in the female. Its maximum diameter is -26 mm. in the male and ‘34 mm. in the female. The caudal end of the esophagus projects into the beginning of the chyle intestine, forming three small lobules. 226 DR. M. KHALIL ON THE NEMATODE Chyle intestine.—The intestine takes a practically straight course in the axis of the body. The wall of the intestine is annulated on the external surface by shallow depressions between the cells. Except for a short distance at the beginning and end the whole length of the gut-wall is deeply pigmented. 'The rectum is an elongated and narrow canal indefinitely demarcated from the intestine. Exeretory pore.—The excretory pore is placed just caudad of the nerve-ring in the mid-ventral line. It is ‘068 mm. from the head end. Nerve collar.—This is very thin in comparison with the size of the worm. It is placed -6 mm. from the cephalic end. Genital organs.—Male: The testis 1s irregularly coiled, reach- ing within 5 mm. of the head. There is no distinct seminal vesicle. ‘he cement gland is of enormous length. Female: The wide opening of the vulva is surrounded by a raised cuticular margin. The strongly muscular vagina runs straight towards the head end. The two uteri are convergent, ending in strong ovejectors lying side by side and joining the cephalic end of the vagina. The ovaries are irregularly coiled, reaching within 5 mm. of the head end. Bursa.—The bursa is very short, and indistinctly divided into three lobes of approximately the same length. It measures ‘22 min. in length and ‘7 mm. in breadth. Text-figure 19. Text-figure 20. woe, Murshidia hadia Khalil. Murshidia hadia Khalil. Lateral view of bursa. Dorsal ray of bursa. The ventral ray is bifid throughout its length, and is placed a considerable distance from the ventral edge of the bursa. The lateral rays arise by a common origin. At the site of their bifurcation each ray has a bulb-like swelling. The externo-lateral ray lies separate from the other two and close and parallel to the ventral ray. It ends in a papilla on the external surface a short distance from the edge of the bursa. The medio-lateral and dorso-lateral vays lie close together. The externo-dorsal ray has PARASITES OF ELEPHANTS. 227 a rugged outline. It arises in common with the dorsal ray, and lies close to and parallel with the postero-lateral ray, ending a short distance from the edge of the bursa on the external surface. The dorsal ray arises by a very broad origin and bifurcates into two branches, each ending in three rays. The outer two rays arise by a common trunk and lie close together. The medial ray is stouter and longer, and ends some distance from the corre- sponding ray of the opposite side. The dorsal ray is ‘25 mm. long. Genital cone.—The genital cone ends in a sharp point. The cloacal canal opens a little distance posterior to the apex. Spicules.— The two similar spicules are very stout and practi- cally straight. The club-shaped tip is bent sharply forwards. They are 1-1 mm. in length. There is a small, curved accessory piece ‘1mm. in length. The cephalic portion of the spicular sheath is reticulated. Text-figure 21. Wurshidia hadia Khalil. Spicules and accessory piece. Termination of the female.—The female tail gradually tapers to a sharp point. It is 2°25mm. long. The vulva is placed ‘6 mm. cephalad of the anus, Ova.—The ova in the vagina are small: 40 long and 23 pu broad. They are then in the morula stage of development. Habitat. Intestine of the African elephant (Uganda). 228 DR. M. KHALIL ON THE NEMATODE Discussion. The main specific differences between the members of the genus Murshidia ave to be found in the shape of the rays of the male bursa, the length of the spicules, the length of the female tail, and the distance between the vagina and the anus. These and other differences are for convenience tabulated below. Upon these data the species Murshidia linstowi was differen- tiated. Railliet, Henry and Bauche have recorded the presence of Murshidia falcifera in the African elephant. This was probably either Jf. hadia or linstowi, which bear a superficial resemblance to MW. falcifera. In the following table all measurements given are in millimetres. The measurements of J. murshida and MW. falcifera are based partly on Lane’s record and partly on material of both species sent by him to Prof. Leiper. ‘These specimens are labelled by Lane as type material of the respective species. The measure- ments of I. linstowi are based on the original record of this species by von Linstow. Where the sex is not stated, the measurements are put in the centre of the column. Taste [. M.murshida. | M. faleifera. | M.linstowi,sp.n.| M. hadia. 3. India. aul a. India. Og. Africa. Q. | 3. Africa. Q. min. mm. | mm. mm. | mm. mm. | mm, mm. Total length ............... | 18-20 22-281/22-95 28-30] 25 29 185 24 Maximum diameter...... 6 “9 “74. TO |) Oe Te | By “82 CBOFOINEVAUE) sosdansaasaacen | 2 uss 5) || 497 8 | 85 ‘9 Nervelcollaneeeereeay see ale 31 | “48 BOT ecny ita aes | 6 Diameter of head ......... 17 BOW Say iil ee ie eceeaes | 2 Rays of leaf-crown ...... 60 80 Pore Hae abe | 40 Diam. of mouth capsule. ‘08 | “il |e eed toy “15 Length of spicules ...... | 1:28 155 | “96 relia VEIN OH TEVAEMIE) dobadacdare 1G | 22 31 | 2°25 Vulva from tail end .... 23 | 29 | 4°3 | 2°85 ORES» Ghendaend heeael Nat 05 X03 052X024. | 04 023 Cuticular striation ...... 007 ‘007 ‘01 | ‘009 Cervical papille ......... "85 1:2 see 14 1-4 Hxcretory pore ............ 8 LOSE REN Rite 68 Genus Cuonrancium Raill., Henry & Bauche, 1914. Asifia Lane, 1914. Fairly stout, straight worms; the head is obliquely truncated dorsad. The oral opening is directed towards the dorsal surtace. PARASITES OF ELEPHANTS. 529 The oral aperture is surrounded by a circle of converging rays. The mouth capsule is large and like a funnel in shape. Rounded cuticular prominences protrude into the oral cavity. Male: The most striking feature of the bursa is an accessory ray projecting from the dorsal aspect of the main stem of the lateral ray. The two spicules are similar and equal. There is an accessory piece. Mailliet wrongly recorded that the dorsal ray divides into four branches instead of three. Female: The vulva is close to the anus, where the worm suddenly narrows. The tail is bluntly conical. Type-species, Choniangium epistomum. CHONIANGIUM EPISTOMUM Piana & Stazzi, 1900. Sclerostomum epistonum Piana & Stazzi, 1900. Choniangium epistomum Raill., Henry & Bauche, 1914. Asifia vasifa Lane, 1914. The male measures 14 mm.in length, with a maximum diameter near the head of 0°75 mm. ‘The spicules ave 20 mm. in length. The female is 19 mm. in length, with a maximum diameter of 1-'Omm. The vagina runs cephalad, joining the two parallel uteri. The ova measure 50 w by 25 p. Habitat. Cecum of the Indian elephant (India). Genus QurtontA Lane, 1914. Hvoansia Railhet, Henry & Joyeux, 1913 (not Hvansia Scott, 1906, for a copepod). @uilonia Lane, 1914. Nematevansia Thle, 1919. Quilonia Ihle, 1919. Fairly slender worms; the head discoid, the mouth terminal and surrounded by two sessile lateral and four prominent sub- median papille. The rays of the external leaf-crown are few and characteristically curved. The cuticular lining of the oral cavity does not lie directly in contact with the chitinous mouth capsule. The mouth-opening is narrower than the diameter of the oral cavity. The esophagus is nearly cylindrical in shape. Two or more teeth project into the oral cavity. Male: The bursa is more or less distinctly divided into three lobes. The dorsal lobe is longer than the lateral lobes. The bursal rays are the usual number. The dorsal ray is bifurcate, each branch having three subdivisions. The spicules are equal and similar, and each has a sickle-shaped point. The accessory piece 1s curved from side to side, the concavity being ventrad. Seen from the dorsum it is wider cephalad than caudad. Female: The vulva is in the caudal third of the body. The uteri are opposed, divergent and furnished with marked ovejectors. The caudad uterus turns cephalad immediately beyond its ovejector. ‘The two uteri run cephalad side by side. Brown a ‘ 230 DR. M. KHALIL ON THE NEMATODE cement on the vulva is fairly marked. The ova are colourless and thin-shelled and have a granular yolk. Type-species, Quilonia renniei. QUILONIA RENNIEI Raill., Henry & Joy., 1913. Nematode No. 2 from Indian elephant. Evans & Rennie, 1910. Hvansia renniet Ruill., Henry & Joy., 1913. Vuilonia quilong Lane, 1914. Kvansia rennier Lane, 1915. Nematevansia rennier hile, 1919. Vulonia renniet Ihle, 1919. The main features of this species, as described by Evans and Rennie and also by Lane, are: («) the leaf-crown projects freely above the head, (6) the leaflets are thim and long, (c) the dorsal vay bifurcates low down in the bursa, and (d) its tri-radiate final branches are very thin and arise at different levels. (For measurements see Table IT.) Habitat. Cecum of the Indian elephant (India). (UILONIA TRANVACRA Lane, 1914. Quilonia tranvacra Lane, 1914. Hvuansia tranvacra Raill., Henry & Bauche, 1915 Nematevansia tranvacra Ihle, 1919. Quilonia tranvacra Lane, 1921. The leaf-crown does not project above the head. It consists of ten rays. The branches of the dorsal ray are stouter and longer than in Q. rennieit. (For measurements see Table II., p. 249.) Habitat. Cecum ? of the Indian elephant (India). QUILONIA APIENSIS Gedoelst, 1916. (Text-figs. 22-28.) Hvansia apiensis Gedoelst, 1916. QYuilonia apiensis Lane, 1921. This parasite was recorded by Gedoelst unaccompanied by figures. In the genus Quilonia it is extremely difficult to differentiate species on description only. Lane, in describing YY. africana, expressed his regret that Q. apiensis was recorded without figures. The same difficulty was experienced in examining the material at my disposal without a full knowledge of @. apiensis. Later, Prof. Gedoelst very generously sent co-type material of his species to be added to Prof. Leiper’s collection. This made possible a complete description of the species with camera-lucida drawings. Material—The material consisted of 25 specimens labelled Quilonia apiensis (co-types) from Prof. Gedoelst. Of these, eight specimens are kept in the Helminthological collection of the London School of Tropical Medicine. In addition, three speci- mens were identified by me in Prof. Leiper’s s material, Shape of body.—The body is straight, slender, and tapering towards either end. The male is 16 mm. long and the female is PARASITES OF ELEPHANTS. sit 19-21 mm. long. The maximum diameter of the body is about the middle. It is -62 mm. in the male and -78 in the female. The average length of this species given here is greater than that origine ally recorded. Gedoelst stated that the length of the male 1s 12: 8mm.and the female 17°83mm. That shes: measure- nents were based on a single individual male and one female is evident from his remarks: “ LD’ unique exernplaive que nous avons étudié mesurait, etc.” The measurements recorded here are the average of all the specimens available. Shin. —The cuticle is striated at regular intervals of ‘02 mm. Mouth collar.—Vhe mouth collar is distinctly separated from the rest of the body by a groove. There is a curious cuticular prolongation in the form bi a spine at the site of this groove Text-figure 22. Quilonia apiensis Gedoelst. Lateral view of head. The mouth collar is irregularly rounded in outline, being -04 mm. in length in the male and -08 in the female. It is -16 in diameter. Kxternal leaf-crown.—This consists of twelve leaflets. These are thin and curved at the top, ending in a fine point. They project freely above the head. Head papille.—The four submedian head papille are thick. Hach is surmounted by a knob-like structure. The two lateral papille are thin and do not project above the surface. Their termination 1s marked by a depression. Cervical papille.—The two cervical papille are thin and point slightly cephalad. | They are placed on the lateral lines ‘98 mm. from the head end in the male and 1:04 mm. in the female. Mouth capsule.—The chitinous mouth capsule is a short ring "04 mm. in length and -16 mm. in diameter. It is not in 932 DR. M. KHALIL ON THE NEMATODE contact with the wall of the oral cavity. The oral cavity is wide laterally and contracted from before backwards. It is cone-shaped. Internal leaf-crown.---This consists of twelve leaflets corre- sponding to the external Jeaf-crown. Each is broad, with a rounded blunt end and does not project above the mouth collar. (Hsophagus.—There is a small but distinct cesophageal funnel having a chitinous wall. From the top of the cesophageal columns two spinous processes project into the mouth capsule. These are not so prominent as in YY. africana. The csophagus is ‘73 mm. long in the male and ‘8 mm. in the female. [t is °25 mm. in maximum diameter in the male and :28 mm. in the female. Text-figure 23. SOW Quilonia apiensis Gedoelst. Dorsal view of head. The cesophagus is slightly narrowed at the site of the nerve collar. The caudal end of the cesophagus projects into the beginning of the intestine, forming a tri-lobed valve. Chyle intestine.—The intestine is straight and thick-walled. It is deeply pigmented in its middle two-thirds. In most specimens there is a characteristic constriction in the outline of the intestine 2 mm. from its caudal end. Rectum.—A marked groove indicates the beginning of the rectum. Its walls are thinand chitinous. The lumen is spindle- shaped, -42 mm. in length. Excretory system.—The excretory vesicle is small and has a delicate wall. The excretory pore is placed ‘6 mm. from the head end. Nervous systenu.—The nerve collar surrounds the narrowest PARASITES OF ELEPHANTS. 933 part of the cesophagus and is ‘24 mm. from the head end. ‘Iwo long bands stretch from the nerve collar caudally. Geiital organs.—Male: The fine tubules of the testis are bent irregularly, reaching to within 4mm. of the head end, There is no distinct vesicula seminalis. The cement gland is of enormous length, ending in the anterior part of the genital cone. Female: The two convoluted ovaries reach within 5 mm, of the head end. The two uteri lie side by side in the longitudinal axis of the body. Each ends in a strong ovejector—that of the anterior uterus is cephalad and that of the posterior uterus is caudad of the vagina. The two ducts from the muscular part of Text-figure 25. Text-figure 24. Quilonia apiensis Godoelst. Ventral view of anterior portion of body. Quilonia apiensis Gedoelst. Female tail. the ovejector join a very short vagina placed transversely to the longitudinal axis of the body. The vulva is not surrounded by raised lips. It is situated 6°2 mm. from the tip of the tail. brownish cement surrounds the vulval opening. Bursa.—The bursa is more or less divided into three lobes. The posterior lobe is slightly longer than the lateral lobes. Seen from the lateral aspect, it resembles somewhat the bursa of Q. uganda in shape, although the dorsal ray is totally different. The bifid ventral ray lies close to the lateral rays. ‘he fissure 234 DR. M. KHALIL ON THE NEMATODE between the medio-lateral and the postero-lateral rays is deeper than that between the medio-lateral and the ventro-lateral rays. The externo-dorsal ray arises from the stem of the dorsal ray. he dorsal ray is ‘7 mm. long. It bifurecates in its lower third, Each branch after a short course divides Text-figure 26. Quilonia apiensis Gedoelst. Ovum. into three rays at the same level. In many specimens, however, the lateral two branches tend to lie close together, and the fissure between them is not so deep as that which separates the medial branch, Text-figure 27. Text-figure 28. /00p Quilonia apiensis Gedoelst. Dorsal ray of bursa. Quilonia apiensis Gedoelst. Lateral view of bursa. Genital cone.—The genital cone is blunt at the apex, which lies in line with the ventral surface of the body. The cloacal canal opens posterior to the apex and is surrounded by four rounded lips. PARASITES OF ELEPHANTS. 935 Spicules.—The two similar spicules are gracefully curved ventrally. Their tips are fine and bent dorsally. ‘hey are ‘91 mm. in length. The accessory piece is*19 mm. in length. It is concave ventrally, providing a canal for the spicules. It is thicker at its cephalic than at its caudal end. Termination of female.—The tail of the female is rather broad. — The tip in most specimens is diseased and deeply pigmented, but a few retain the tip, which is fine. The tail is 2°66 mim. in length. The anal opening is slightly raised above the surface. Ova.—The ova in the vagina ave thin-shelled, measuring 88 ju by 48. They contain a morula. Habitat. Intestine of African elephant (Belgian Congo and Uganda). (JUILONIA AFRICANA Lane, 1921. This species is represented by more specimens than any other in the Uganda material. It agrees in all characters with Lane’s description, except in the length of the female tail and the distance of the vulva from the anus. These were given by Lane as 4 mm. the length of the tail, and 4 mm. the distance of the vulva from the anus. On re-examining Lane’s type-species, kept in the Natural History Museum, I could not find a single specimen with these measurements. All corre- spond to my previously taken measurements, being 3°6 mm. length of tail, and 3:25 mm. distance of vulva from the anus. Lieut.-Col. Lane ascribes this to the fact that the proofs of his paper were not corrected by him. (For detailed measurements based on my observation, see ‘lable IT.) Habitat. Stomach of African elephant (South Africa and Uganda). “e QuiLontA UGANDA Khalil, 1922. (Text-figs. 29-34.) Material.—The material consisted of fifteen specimens sorted from the Uganda collection. Shape of the body.—The body is short and straight. The female is slightly longer and thicker than the male. The male is 11°5-12 mm. long and the female is 16 mm. long. The maximum diameter of the body is about the middle. It is .OD mm. in the male and ‘64 mm. in the female. Skin.—The cuticle is striated more closely near the head end than in the middle of the body. Near the head the intervals between each two striations is ‘01 mm., near the middle of the body it is °02 mm. Mouth collar.—The mouth collar is distinctly separated from the body by a groove. It is rounded in outline, -04 mm. in length and *19 mm. in diameter. Anterior leaf-crown.__From the inner surface of the mouth collar projects the anterier leaf-crown, It consists of twelve 236 DR. M. KHALIL ON THE NEMATODE leaflets, fairly broad and bent characteristically at their tips. They project a little distance above the head. Head papille.——The four submedian head papille project a little above the mouth collar. They point straight forwards. Their bases are broad and each is surmounted by a long knob, the function of which is pointed out elsewhere. The two lateral head papille do not project, and their termination is marked by a depression. Cervical papille.—The two delicate cervical papille are placed a little posterior to the level of the caudal end of the esophagus. They are °78 mm. from the head in the female and -6]1 mm. in the male. Text-figure 29. Quilonia uganda Khalil. Head. Mouth capsule.—The chitinous mouth capsule is -15 mm. in diameter. It is oval in optical cross-section and is not in contact with the wall of the oral cavity. The oral cavity is shallow and funnel-shaped. It is 06 mm. deep and *11 mm. diameter at its floor. The mouth-opening is 07 mm. in diameter, From the floor of the oral cavity project two large chitinous oval teeth, each surmounted by a delicate sharp spine. These are placed on the cephalic end of the two sub-ventral cesophageal columns. Internal leaf-crown.——The short stumpy leaflets of the internal leaf-crown correspond in number to the external crown. They have a rounded end and do not project above the surface. (Hsophagus.— There is a shallow cesophageal funnel surrounded by a chitinous rim. The cesophagus is short and thick, being -52 mm. in length in the male and 62 mm. in the female. Its maximum diameter is °2 mm. and :21 mm. in the respective sexes. The end of the cesophagus projects into the chyle intestine, its opening being guarded by a trilobed valve. PARASITES OF ELEPHANTS. 237 Chyle intestine.—Vhe course of the chyle intestine is straight in The intestinal wall is thick and deeply the axis of the body. The rectum is a narrow, long canal pigmented in its caudal half. "28 mm. in length, demarcated from the intestine by a groove round the circumference of the gut. Text-figure 31. Text-figure 30. = ao — ARTA RAS See H i —— saci din on a on er LL rel ene ala ile 100 7 Quilonia-uganda Khalil. Cephalic end of body. Quilonia uganda Khalil. Female tail. Text-figure 32. Quilonia uganda Khalil. Ovum, Nerve collar.—The thick nerve collar is placed °25 mm. from It surrounds the narrowest part of the the cephalic end. cesophagus. Proc, Zoou. Soc.-—1922, No. XVII. ‘lis 238 DR. M. KHALIL ON THE NEMATODE Excretory pore.—The excretory pore is ‘47 mm. from the head end. Genital organs.—Male: The convoluted testis reaches within 3mm. of the head end and is irregularly coiled. The vesicula seminalis is spindle-shaped, lying in the long axis of the body. The cement gland is very long, measuring 3:5 mm. Female: The irregularly convoluted ovaries reach within 3 mm. of the head. The divergent two uteri lie side by side for most ef their course. Hach ends in a strong muscular ovejector which lies, in the case of the anterior uterus, cephalad and, in the case of the posterior uterus, caudad of the vagina. The latter is a short horizontal duct joining both uteri to the vulva. The opening of the vulva is not surrounded by raised lips, and is placed 2:17 mm. cephalad of the anus. Text-figure 33. Quilonia uganda Khalil. Lateral view of bursa. Bursa.—Seen from the lateral aspect the bursa has a spindle shape, lying obliquely to the longitudinal axis of the body. The dorsal lobe is slightly longer than the lateral lobes and has a rounded end. The ventral, lateral and externo-dorsal rays con- form to the general design in the genus, but are all more slender. The dorsal ray is long and thin, being °5 mm. in length. It bifurcates in its lower third into two long and slender branches. These in their turn branch nearer their ends. The most lateral ray arises first, and the two medial rays arise by a common trunk branching lower still. This arrangement is exactly the opposite of that in Y. apiensis. ‘; The pre-bursal papille are placed °52 mm. from the tip of the bursa. Genital cone-—This cone is placed nearer the ventral than the dorsal surface. Its tip is rounded. The cloacal opening is PARASITES OF ELEPHANTS. 239 marked by a depression in its outline a little distance dorsal to the apex. Spicules.—The two similar spicules are ‘72 mm. long, the shortest in all the members of the genus. The tips of the spicules are sharp and bent ventrally. The elongated accessory piece isl mm.in length. It is curved ventrally to lodge the spicules, Text-figure 34, Quilonia uganda Khalil. Wentyral view of bursa and spicules. Termination of female.—The female tail is 2°43 mim. in length. It ends in a fine point. The anal opening is marked by a round depression from the surface. Ova.—The ova seen in the terminal part of the uterus are thin-shelled. Their contents are in the morula stage. The dimensions of the ova are 60 p in length by 30, in breadth. Haoitat. Intestine of African elephant (Uganda). QUILONIA BREVICAUDA, sp. n. (Text-figs. 35-39.) - Material.—The material consisted of two specimens sorted out from the Uganda collection of nematode parasites of the elephant. Shape of the body.—The female is much stouter and longer than the male. The outstanding feature of this species, as its name implies, is the short tail of the female ending bluntly, as can be seen with the naked eye. The male bursa is very much smaller than in the other species. The male is 13mm. long. The female is 21 mm. long. Tbe maximum diameter Was 240 DR. M. KHALIL ON SHE NEMATODE of the body in either sex is about the middle. It is ‘45 mm. in the male and ‘85mm. in the female. Skin.—The cuticle is striated at intervals of ‘02 mm. through- out the length of the body. Mouth collar.—The mouth collar is rounded in outline. It is ‘06 mm. long in the male and ‘08 mm. in the female. Its diameter is -18 mm. in the male and -22 mm. in the female. Eeternal leaf-crown.—This consists of ten very slender leaflets, bent near them termination. They do not project freely above the head. Text-figure 35. Quilonia brevicauda, sp.n. Head. Head papille.—Vhere are the usual four submedian head papillae. These are stumpy and surmounted by a knob each. The lateral head papille do not project above the surface and are marked with a depression. Cervical papille—The two cervical papille are very thin and directed slightly cephalad. They are *95 mm. from the head end in the male and 1:1 mm. in the female. Mouth capsule.—The chitinous mouth capsule is very short. It has a diameter of -13 mm. in the male and *14 mm. in the female. It lies closer to the oral cavity than in the other species. The oral cavity is funnel-shaped, narrowing tcwards the mouth-opening. Into the mouth capsule project three formid- able Jong teeth; each is placed on the top of one of the three cesophageal columns. This peculiarity is apparently limited to this species. Internal leaf-crown.—This consists of biunt and slightly pro- jecting processes placed deep in the oral capsule. PARASITES OF ELEPHANTS. 241 Hsophagus.—There is a small esophageal funnel. The ceso- phagus is elongated, being -52mm.in length in the male and ‘7 mm.in the female. It is slightly constricted in the region of the nerve collar. From its posterior end project the three lobes of the cesophageal valves. Chyle intestiie.—The chyle intestine pursues a straight course. It is pigmented in its middle. The rectum is sharply demar- cated from the end of theintestine. It is narrow and thin-walled, and measures ‘24 mm. in length. Text-figure 36. Quilonia brevicauda, sp.n. Cephalic end of body. Execretory pore.—This ig situated in the mid-ventral line -73 mm. from the head end in the male and °88 in the female. Nerve collar—The nerve collar surrounds the cesophagus -29 mm. from the head end in the male and °39 mm. in the female. Genital organs.—Male: The convoluted testis reaches within 3 mm. of the head end. There is no distinct seminal vesicle. The cement gland is very long. Female: The convoluted ovaries reach within 6 mm. of the head end. The two divergent uteri lie side by side for the most of their course. Each ends ina strong muscular ovejector. That 242 DR. M. KHALIL ON THE NEMATODE of the posterior uterus makes a loop caudad of the vulval cpening before joining the ovejector of the anterior uterus to form the very short horizontal vagina. The vulva is covered with dark brown cement. It is 5°l mm. from the tail end and 3:35 mm. from the anus. Text-figure 37. Text-figure 38. SO Quilonia brevicauda, sp. n. Ovum. Quilonia brevicauda, sp, n. Female tail. Bursa.—Vhe male bursa is short and broad. It is 63mm. in length and 52 mm.in breadth. The ventral, lateral and externo- dorsal rays are similarly arranged as in other members of the genus. The dorsal ray is shortand broad. It is°5 mm. in length. It divides in its lower third into two thick branches. Each in turn ends in three small and thick rays. The most lateral arises separately and at higher level. The middle ray is the shortest. Genital cone—The blunt genital cone hardly protrudes into the bursa. ‘The cloacal opening lies a little dorsal to its blunt apex. Spicules—The two similar spicules are ‘81mm. in length. Their termination is very fine and directed ventrally. The accessory piece is "19 mm. in length. Its lateral margins curve forwards, forming the dorsal wall of the cloacal duct. PARASITES OF ELEPHANTS, 243 Termination of the female.—This forms the most. charac- teristic feature of this species. The female tail is broad and short. It is 1:75 mm. in length. It is of practically uniform diameter till near the tip, where it is cone-shaped. The tip is blunt. Text-figure 39, 1 Quilonia brevicauda, sp. n. Bursa and spicules. Ova.—The ova seen in the terminal portion of the uterus contain a morula, They measure 56 in length and 33, in breadth. Habitat. Intestine of African elephant (Uganda). QUILONIA ETHIOPICA, sp.n. (Text-figs. 40-45.) Material.—The material consisted of eight specimens sorted from the Uganda collection. Shape of body.—The body is straight, broader near the middle, and gradually tapering towards either end. Male is 13°5 mm. long and °55 mm. maximum diameter The female is 19 mm. in length and -7 mm. maximum diameter. Skin.—The cuticle is finely striated through the whole body except for a short distance caudad of the head. The striations are at intervals of :(029 mm. Mouth collar—The mouth collar is rounded in outline. It is 944 DR. -M. KHALIL ON THE NEMATODE ‘06mm. in length in the male, ‘08mm. in the female. The diameter is ‘14 mm. in both sexes. Keternal leaf-crown.—The external leaf-crown consists of twelve leaflets. These are the most prominent characteristic of the species in their appearance and arrangement. They are broad, leaf-like, and slightly bent at the tip. They do not protrude above the head. Head papille.—The four submedian head papille are displaced laterally by the external leaf-crown. They project directly forwards, and each is surmounted by a small papilla, the nature of which is referred to elsewhere. The lateral papille do not project, and their termination is marked by a depression. Cervical papille.—The two laterally placed cervical papille are directed slightly cephalad. They lie ‘7mm. from the head end in the male and ‘86 mm. in the female. mele 40, 50 Quilonia ethiopica, sp.n. Head. Mouth capsule—The chitinous mouth capsule is extremely shallow. In optical cross-section it lies oblique to the axis of the body. Itis -018 mm. in depthand-12 mm. in diameter. The chitinous mouth capsule does not come in contact with the wall of the oral cavity. The latter is practically cylindrical in shape and slightly contracted in the middle. There are apparently no teeth projecting into the mouth cavity. Internal leaf-crown.—The twelve leaflets of the internal leaf- crown can be traced from the floor of the mouth cavity to about its middle. Hach leaflet is broad at its base and tapers toa point. It has a distinct cleft in the middle, and thus each is really two leaflets side by side. @sophagus.—There is a shallow cesophageal funnel surrounded by a chitinous rim lying on the three cesophageal columns. The cesophagus is shortand thick. It is ‘55 mm. in length in the male and ‘07 mm. in the female. It is slightly narrowed at the PARASITES OF ELEPHANTS. QA5 site of the nerve collar. Its maximum diameter is ‘21 mm. 1n the male and ‘26mm. in the female. The posterior end of the cesophagus projects into the intestine, and its opening is guarded by a trilobed valve. - Chyle intestine. —The straight chyle intestine is pigmented in Text-figure 41, 420 Ms — Quilonia ethiopica, sp.n. Cephalic portion of body. its middle third. The narrow rectum is -23 mm. in length and sharply demarcated from the rest of the intestine. Excretory pore.—The excretory pore lies in the mid-ventral line °6 mm. from the head end. ‘Text-figure 42. Quilonia ethiopica, sp. ii. Ovum. Nerve collar.—The nerve collar is °25 mm. from the head end. Genital organs. Male: The testis is convoluted irregularly, reaching within 3°5 mm. of the head. There is a peculiarly 246 DR. M. KHALIL ON THE NEMATODE shaped seminal vesicle, having the form of two spindles con- nected with a narrow duct. The cement gland is long and pierced by the long ejaculatory duct, which opens into the ventral aspect of the cloaca. ° Female: The convoluted ovaries reach within 4 mm. of the head end. The divergent uteri lie side by side for most of their course. Each is provided with a well-developed muscular ovejector. The posterior ovejector loops caudad of the vulvar opening before joining the anterior ovejector to form the short horizontally placed vagina. The vulva is covered with brown cement and lies 4°97 mm. from the tail end. Text-figure 43. Quilonia ethiopica, sp.n. Male bursa and spicules. Bursa. The male bursa conforms to the usual type. The dorsal lobe is slightly longer than the lateral lobes. The bursa is ‘8mm. inlengthand*6mm.in breadth. The ventral, lateral and externo-dorsal rays are arranged in a similar fashion, as seen in the other species. The dorsal ray is ‘67mm. in length. It bifureates in its lower third, The terminal branches are slender and long. ‘The lateral branch separates first. Genital cone.—The genital cone is fairly long and pointed. The cloacal opening lies near its apex. Pre-bursal papille.—These lie on the lateral lines ‘75 mm. from the end of the bursa. PARASIVES OF ELEPHANTS. 247 Spicules.—The two similar spicules are ‘83 mm. in length. They end in a fine termination, bent forwards. ‘The accessory Text-figure 44. fl 100 p Quilonia ethiopica, sp.n. Ventral view of bursa. Text-figure 45. tone Quilonia ethiopica, sp.n. Female tail. piece is *l1 mm. long. It has a bulbous cephalic extremity. grooved ventrally to lodge the spicules. Itis 248 THE NEMATODE PARASITES OF ELEPHANTS. Termination of the female.—The female tail is 2°3 mm. long. It tapers gradually. Its tip is rounded. Ova.—The ova in the terminal part of the uterus are oval in shape, thin-shelled, and contain a morula. ‘They measure 63 pu long and 35 pw broad. Habitat. Intestine of African elephant (Uganda). Text-figure 46. Quilonia africana Lane. Abnormality in the rays of the bursa. In a male specimen of Q. africana which was otherwise normal, the externo-dorsal ray was completely missing on one side, as is shown in the accompanying camera-lucida drawing. This abnormality must be very rare, and has apparently not been recorded before. Discussion. As was mentioned before, the identification of any particular species of the genus Qwéilonta from descriptions only is very difficult. Diagrams representing the head and tail ends are necessary. ©. brevicauda is easily distinguished on account of its short stumpy tail. For the rest, careful comparison is neces- sary for diagnosis. ‘he important features in this connection are the shape of the head, the external leaf-crown, the dorsal ray of the male bursa, the length of the spicules, the length of female tail, and the distance between the vulva and anus. The fact that the external leaf-crown may or may not project above the head divides the genus into two groups. In Q. renniei, i ce