ben ee ne he ee ete Ne 8 f Fee PN LS | cit trim te pte eee | teen, al rr ey \ J 2 = - ae mb ay TG , IAN USE Samiti PONE Ie OP > Be ‘Fe ee ! RECORDS : of the INDIAN MUSEUM (A JOURNAL OF INDIAN ZOOLOGY) Vol. XXIV, 1922 EDITED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA. Vinee Zs Sa Leal Easel Sead Tn ia OM ae @alentta : PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA — 1922 Ses, ad tea VE: EAS fe ae Nea Lee is MAT, + pe et: 7 ‘we CONTENTS. Part I. Published 6th May, 1922. New and rare Odonata from the Nilgiri Hills Observations on the Invertebrate Fauna of the Kumaon Lakes. ¢ JIJ.—The Freshwater Mollusca A Review of the Indian Species of Amblycephalus ... A new Snake from the Northern Frontier of Assam Structural Modifications in the Fish of Mountain Torrents Notes on Fishes in the Indian Museum. III.—On Fishes belonging to the a Cobitidae from nes altitudes in Central Asia A note on Bees of the genera -Xyloconn and Brain in the Indian Museum A Revision of the Burmese Unionidae ... | ei Part II. Published 29th June, 1922. Notes on Crustacea Decapoda in the Indian Museum. XV.— Pontoniinae is a Part III. Published 31st August, 1922. The Fauna of an Island in the Chilka Lake— Heteromera : Free-living Thysanura .. Dragonflies : Notes on Fishes in the Indian Museum. IV.—On Fishes Rene: ing to the genus Botia (Cobitidae) oe New records and species of Membracidae from India Five new species of the Rhynchotan genus Coriza On some Indian Derbidae (Homoptera) Aho Pa ge. 85 91 118 289 299 303 313 323 331 335 ll Contents. New Indian Homoptera Materials for a generic Revision of the Freshwater Gastropod Molluses of the Indian ae No. 5.—The Indian Planorbidae re bes 300 On a new Alycaeus from the Khasi Hills A list of the Dragonflies recorded from the Indian Empire with special reference to the collection of the Indian Museum. Part V.—The subfamily Gomphinae Part IV. Published 19th October, 1922. Some Earthworms from Kashmir, Bombay, and other parts of India 6 S08 HOD : Indian Mysidacea Parallel Evolution in the Fish nail Meee of Mountain Tor- rents : one Some Oriental om Ni in the Indian Museum Hirudinea from the Inle Lake, S. Shan States Descriptions of some Indo-Malayan species of Capritermes oe mitidae) ... Ban cic ae a APPENDIX. A. List of Literature referring to Indian Zoology (excluding Insecta) received in Calcutta during the year 1922 pp. B. Report on the iii Stor of India for the years 1920 — to 1923 : vee pp. Page. 343 357 365 367 427 445 505 511 521 535 i—xiii i—lvi LIST OF PLATES. Plate I (Odonata) Plate II (Mollusca) Plates III—IX (Crustacea) Plate X (Membracidae) Plate XI (Odonata) Follow page. 10 12 288 330 426 7 ie * i ase ‘ eu , i ae hi ee . One ee Bovitse Ciaatlswenhing Foam n) Goa. c iy; 4 are ie i {ifr esis He 7 7 Y y LA oy mn ¢ ee : “ee 3 ‘ é fe i ms | ; dae Ds . , 1 “y ‘ | ae , { A h , ; ny LIST OF AUTHORS. ANNANDALE, N., D.Sc. Materials for a generic Revision of the Freshwater Gas- tropod Molluscs of the Indian ey No. 5.—The Indian Planorbidae ... Parallel Evolution in the Fish and Tadpoles of vata Torrents. (In collaboration with S. L. Hora) a Buatr, K. G., B.Se. The Heteromera of Barkuda Island Dover, C. A note on Bees of the gonna ie and Bombus in the Indian Museum The Free-Living hyacinths of Barks Island The Dragonflies of Barkuda Island. (In collaboration with F. C. Fraser) ; Five new species of the Bhgeibboten genus derien (Com- ptled frem notes left by the late C. A. Paiva) Fraser, F. C., 7.M.S. New and rare Odonata from the Nilgiri Hills .. The Dragonflies of Barkuda Island. (In entlaborubios with C. Dover) A list of the Diapoates tides from the Hidign ime pire—Appendix Bo Some Oriental Ascalaphidae in the fnavan Musetue -Funxnouser, W. D. New records and species of Membracidae from India Gopwin-AustEN, H. H., F.R.S. On a new Alycaeus from the Khasi Hills Hora, 8. L., M.Sc. Structural Modifications inthe Fish of Mountain Torrents Notes on Fishes in the Indian Museum—- III.—On Fishes belonging to the family Cobitidae from high altitudes in Central Asia IV.—On Fishes belonging to the genus Botia (Cobitidae) Parallel Evolution in the Fish and Tadpoles of Mountain Torrents. (In collaboration with N. Annandale) 323° vi List of Authors. Kenp, 8., Sc.D. Notes on Crustacea Decapoda in the Indian Museum. XV.—Pontoniinae 3 Latpuaw, I. F. A list of the Dragonflies recorded from the Indian Empire with special reference to the ‘collection of the Indian. Museum. Part V.—The subfamily. SOmnnaS: (With an Appendix by F. C. sO Sat Muir, F. : On some Indian Derbidae (Homoptera) New Indian Homoptera Oxa, A. Hirudinea from the Inle Lake, 8. Shan States Paiva, C. A. (the late). Five new species of the Rhynchotan- genus Cohen (Compiled by C. Dover) oe eee ae PRAsHAD, B., D.Sc. Oheneiek on the Invertebrate Fauna of the Kumaon Lakes. III.—The Freshwater Mollusca ... A Revision of the Burmese Unionidae SILVESTRI, F. | Descriptions of some Indo-Malayan species of fe termes (Termitidae) a : “t STEPHENSON, J.,- D.Sc. Some Earthworms from Kashmir, Bombay, and other parts of India ~ 7 sted , TATTERSALL, W. M., D.Sc. Indian Mysidacea Watt, F., C.M.G., 1.M.S. A Review of the Indian species of Amblycephalus “A new Snake from the Northern Frontier of Assam Page. 113 367 835 ° * 343 521 331 it 91 535 427 aes 19 29 INDEX. [V.B.—An asterisk (*)-preceding a lino denotes a new variety or hubipebians a dagger (f) indicates a new species; a double dagger ({) a new genus or sub- genus ; synonyms are printed in italics. ] A Page. Acanthomysis longicornis 483 tAcanthucus minutispinus 323 Acheron trux “3 ae OLG Achilidae ...— - , 1IGk) {diversipes poe dbl) ensifrons 209, 210 frater ... 137, 139, 166 gorgonidarum .. 138 gracilis 300 141, 150 grandis ... 138, 197, 209, 210 hermitensis 5 115, 185 hertwigi A 38 holmesi Soo Calls} timpar 117, 140, 147, 148, 149 incertus 140, 150 indicus 115, 140, 144, 145, 146 infraspinis ; 140, 143, 144 tinornatus fe 117, 119 finvestigatoris 117, 139, 141, 160, 161, 162 kolumadulensis .. 220, 222, 223, 224 laccadivensis 116, 135, 139, 141, 152, {lanipes flatipollex lifuensis longicaudatus Hovacizcalandiaes: 153, 156, 158 . 136, 139, 141, 156 +135, 139, 141, 150, 151, 153, 154, 158 .. 220 | 139, 140, 141, 142 doa. | HORE &vi Page. fPericlimenes noverca 137, 139, 141, 162, 163, 164, 165 fobscurus ~ 115, 140, 144, 145, 146 parasiticus 117, 134, 138 parvus 140, 149, 150 petilthouarsi 139, 195, 196 petitthouarsi denticululu 197 pelitthouarst spinifera .. 195 pottst ... 123, 126 pusillus «. 195 frex 119, 137, 141, 158, 159, 160 rotumanus .. 226 scriptus 135, 136, 137, 139, 140, 147 seychellensis S60 ios, A137 soror ... 139, 141, 165, 166 spiniferus si 117, 139 spinigerus 231, 234 tenuipes 139, 218, 220, 223, 224 vitiensis 210, 214 Periclimenes (Ancylooaris) affinis 171, 214 fagag 170, 173, 197-200, 203, "204, 207, 208, 221. amboinensis 168, 172, 173 americanus 169, 179 amymone 171, 219 fandamanensis tie 204-209, 213 nilandensis 168, 172 brevicarpalis 169, 185, 189, 191, 192, 194, 214 brevinaris 170, 195 brocketti 170, 194 brocki 172, 226 calmani dies 169, 176 ceratophthalmus ... 168, 172, 173 compressus 170, 194 cornutus 168, 172, 173 demani 171, 219 ‘denticulatus 170, 197 tdigitalis 172, 224, 225 suvadivensis 171, 209 tdiversipes 169, 170, 179-182, 184, 185, 194, 195 edwardsi 168, 172 elegans 171, 214, 215, 216, 217, 218, 219 elegans dubius «ait 218 ensifrons 171, 209, 210 frater .. neo grandis. 171, 210, 21)-217, 219 holmesi . 171, 218, 219, ae finornatus 170, 182, 191-194, 2 korni ... 169, “i tleptopus 169, 173-176 lifuensis 171, 220 longimanus 171, 224 longipes 168, 173 petitthouarsi 170, 196 potina - 169, 182, 184, 185 ;proximus . 171, 201-204, 208 psamathe 168, 173 pusillus 170, 195 rotumanus 172, 226 tenuipes 171, 173, 220 seychellensis 169, 176-179 spiniferus 170, 195, 196 tenellus 169, 179 vitiensis 171, 214 xvii Page. Periclimenes siiboad igi nai mus RR. ant cornutus tee VIZ Periclimenes (Cristiger) brocki 226 commensalis axe . =: 166 incertus nae we Lot scriptus wwe 147 Periclimenes (Falciger) apinis 214 amethysteus es seen, 147 borradatlet : wes '220 brocketti ies wn, «| LO4: compressus ae = 194 dubius 218 kolumadulensis mn 220 nilandensis a ey 172 petitthouarsi iat . 196 potlsi ... Rie ws «6-126 seychellensis ees com 176 i el se w. 195 suvadivensis "0 209 Periclimenes (Ham iger) novae-zea- landiae ... 167 Periclimenes (Periclimenaeus) fim- briatus ... eopel6i novae-zealandiae .. ws 167 robustus aa see 167 Perionyx . 429, 430, 435 excavatus we = 435 tmodestus ne 00435 tPerissogomphus oc 375, 383, 384 Tstevensi 371, 383, 384, 385, sity 418 Petalophthalmus armiger i. = 445 Phallusia ... ae anaes Phenelia Ae «. 3653 Phenice pallata ah seog4l Phidole rhombinoda -cit £293 Pheretima aA Rou. Add elongata ae i 433 hawayana 3 ee 1) heterochaeta 433 houlleti 434 posthuma? «. 434 suctoria a 434, 435 Phra amplificata Se -. 338 Phyllomacromia nilgiriensis See 9 Phylloneura westermanni iis 3 Physa oe or ws. - 359 elongata : PL ees Physidae ... 357, 358 Physunio . os 92, 96 ferrugineus oe aoe 97 micropteroides 97 Pinna 117, 118, 251, 256, 260, 279, 281 bicolor 252, 281 nigrina see 252 rugosa aa sw) 9 287 vexillum 5 ise» 2b2 Pinnotheres 117, 260 Placobdella emydae 526, 527 parasitica §21, 529, 530 Planorbidae 13, 14, 357, 358 Planorbinae = -357, 358, 359 Planorbis ... 358, ee 360, 362, 363 albus ... i VOL barrackporensis ... my, 15 caenosus at Sy, 361, 362 calathns Ee ‘7 862 Paye. Planorbis exustus 358, 360 hindu ate - 360 huttont aes Ae 15 hyptiocyclos oe scan COL nitidus 20 362 sindicus 500 «. . 361 trochoideus 362 umbilicalis 15, 361 Planorbis (‘Segmentina) caenosus oe 15 Platydema velutinum 293 Platygomphus wee 396, 308, 399 dolabratus 371, 398 feae ... oe 371, 399 Podogomphus wae o. 392 praetorius fos 390 Polycarpa .. uss ue Ll annandalei ay LG Polyodontophis sagittarius ah 23 Polyzoa_... 11, 12 116, 120, 247, 248, 259, 260, 279, 287 Pontonia ... tanachoreta 116, 118, 261, 264-266 armata tue 6248 ascidicola 118, 260, 261, 264, 266 biunguiculata a 280 brevirostris 118, 260, 261 californiensis 287, 288 flavomaculata ... vee LOS maldivensis 266, 267 margarita 118, 260, 287, 288 minuta a 260, 261 nipponensis 260, 279, 282 Tokai ... ... 116, 118, 261-266 pinnae 118, 249, 251, 252, 260, 287, 288 Se Sa aed 260, 261 tridacnae 280 tyrrhena 117, 118, "359, 260 Pontonia (Harpilius) dentata ». 231 Pontonides beaufortensis Pontoniinae . 120, 139, 266, 267 117 era 115, 117, 119, 136, 138, 148, 246, 267 Pontoniopsis 120, 239 comanthi - 118, 239 Pontoscolex corethrurus -«. 440 Porifera_... see oe (AT Potamarcha obscura . 303, 304 Potamogeton «. 11, 12, 13, 14, 15 Potamomysis Soc spo ESE assimilis ... 446, 448, 487, 488 pengoi sie « 448 447, 474, 475 {Prionomysis one .. 447, 474, 475, 476 fstenolepis Protosticta ae ow §=64, 5 gravelyi 4, 5,8 thearseyi Tstevensi 33 5, 7 tsanguinostigma ... 5, 6 Pseudagrion microcephalum "303, 310 Pseudanchialina inermis a 446 pusilla «» 446 Pseudecheneis 33, 38, 44 sulcatus 42, 45, 54 TPseudoblaps barkudensis 290 javanus . 290 Pseudodon 92, 97, 99, 110 Gta ts oc . 98, 99 XViil Page. Pseudodon thedoug 97 chaperi 60 97 crebristriatus 98 crebristriatus peguensis 98 cumingii oot 97, 99 ellipticus 97 inoscularis 97, 99 orbicularis 97 ovalis 97 peguensis 98 peguensis curvata . 98 ponderosus 97 salvenianus ee 97 salwenianus 97, 98, 99 vondembuschiana oi6 97 vondembuschiana inoscularis 97 zollingert N00 97 Pseudophaea dispar 9 fraseri 8,9 *fraseri wynaadensis noc 8 Psilorhynchus Sele 32, 33, 37, 60, 61, 66 balitora a ae 67 Pteroeides .-- L17, 119, 240, 243 elegans 008 ww. = 248 Ptoleria 349 R Radiatula . Se 95 Rana see 506 afghana 506-509 alticola 506 assamensis 507 latopalmata 507 livida .. 507, 508 malabarica 506 Ranae Formosae 3 507, 509 Rhodischnura nuri ... 310 Rhopalophthalminae .., 457 Rhopalophthalmus _.., 457 egregius 900 445, 457 TRhotala gravelyi x 353 Rhotana arinennts a1 340 Rhotanini .. ae 340 Rhynchota ba 331 Rhyothemis variegata ‘vasiegata .. 807 s Sadia rostrata, - 5cC 351 Saron aes eee 222 Sataspes hauxwelli_.., 86 Scaptobdella blanchardi 533 horsti are 533 Schizothoracinae 33 Segmentina 12, 15, 16, nage 309, 3 361, 362 calathus ~ 16 Selysiothemis 396 Setenis furva 293 Shorea robusta 291 Sieboldius ... 372 albardae 372 grandis con GUI 370, 372 aeyponions Siluroidea . : ; ; 33, 51, 507 Page. Siphlocerus minius .. 618 Siriella ‘ 448, 450, 466 aequiremis 446 affinis 446, 447, 453, “454, 455 brevicaudata ait 450, 451, 452, 453 brookii «. 453 clausii 447, 453 dubia ... 446, 455 gracilis 446, 453 qhanseni .. 446, 448, Wee 450 jaltensis Med 447, 453 longipes .. 453 nodosa ‘, 448, 450, 452, 453 norvegica a 447, 453 paulsoni 445, 466 quadrispinosa 446, 448, 450, 453, 454 thompsoni .. 453 vulgaris + 446, Falls a 454 watasel 20 453 Siriellinae ... 448 Sisor 44 Sisoridae 507 Sogata pusana 351 Spenceriella 429 Spirogyra ... 11, 13, 15 Sphaerium 12, indicum 13, 17 Spondylus 118, 256 Spongicola venusta 117 Stegodyphus sarasinorum 300 Stegopontonia Ab 121, 268 commensalis 117, 268 Stenosis kraatzi 289 Stilomysis .. 488 +Strongylium annandalei 295 . borchmanni 295 marseuli $00 295 Stylogomphus inglisi ... 415 Suhpalacsa 900 B16, 516 obscura . 615 Suhpalacsini 613 Sumangala delicatula . 340 Suphalasca cervinus 515 placida 515 Suphalomitus 513 tserratus 513 verbosus 514 Sybaris testaceus 297 Sympetrum 12 pallidinervis 5 305 Syncrossus berdmorei ... 317~ Syrphidae ... So 88 T Taia intha 527 shanensis .. 629 tTemnogomphus 391, 394 bivittatus 371, 394, 395 Tenebrionidae Bo PHY Termitidae 535 Thalassinidea Boo tI} {Thaumastocaris oc 119, 120, 244 streptopus nD 244, 245, 246 Tholymis tillarga 307 Thysanura 5 299, 300 Tillarga ... 900 303 XIx Page. Toxicum oppugnans 293 Tramea ... 303 basilaris burmeisteri 306 limbata similata ... 306 similata 306 Trameini ... se 00d Trapezoideus 92, 109 dallianus een 110 exolescens 109, 110 foliaceus hyd 109, 110 foliaceus comptus wine LOD foliaceus zaleymanensis steel LO misellus ee 109, 110, 111 subclathratus sinie « 110 Tricentrus albomaculatus 323 allabens eee 323 brevis 323 projectus 323 pronus 323 resectus * 323 Tricula montana 16 Tridacna 118, 256, 4288 279, 286, 286 gigas ... 262 squamosa .. 256 Trigonodon 92, 97, 98 peguensis co 98 peguensis crebristriatus =o 98 peguensis curvata a 98 Trithemis ... wee 303 aurora 303 festiva 303 pallidinervis 305 trivialis eae we = 804 Trocheta ... es 532, 533 horsti 533 tquadrioculata 522, 530, 531, 532, 533 Typhlocaridinae 113 Typhlocaris 506 ols Typton ... 58 121, 286 bouvieri awe 286 spongicola ‘117, 286 U Unio male ae Bo 91 andersoniana 102 andersonianus oe 94 bhamoensis 100, 101, 102 bonneaudt cee 94 burmanus 103 comptus oO corrianus 94, 106 corruvatus 100 dolichorhynchus 106 evitatus eee 500 94 exolescens 109 feae ... wean LOS feddeni 104, 105 foliacea 109, 110 foliaceus Bae So) foliaceus fragilis .. 3 109 fragilis aa 109, 110 generosus 50 sc. ©=—«dL07 generosus angustior 107 generosus delapsus : ws - LOT gerbidont ae oe «|S 94 Page. Unio gianelli 106 gianelli degener 108, 109 houngdaranicus 101 humilis 94 lamellatus 108 lamellatus var. ; 107 laosensis 93 layardi 108 leioma 94 mandelayensis 100, 101, 102 marginalis cylindrica : 106 margtnalis obesa ... 106, 107 marginalis subflabellatta ‘ 106 marginalis tricolor 107 marginalis zonata 106, 107 microsomus or . 109 parma sae » 104 peguensis Nee 109, 110 pilatus see 94 protensus ellipticus 106 protensus obtusatus 106 pugio ... 96 pulcher lamellatiformis 107 pulcher nonderosulus 108 rectangularis 93 savoyensis ver LO4 scutum 600 «. 108 scutum humilior 108 sella 93 smaragdites 102 tavoyensis 102, ‘103, 104 triembolus 102, 104 trirostris 94 vulcanus 103 Unionidae ... .. 91, 92 Urocaridella 119, 122 gracilis 122 Urocaris 115, "134, 135, 136, We aesopius demani : ve indica . 144 infraspinis . 143 longicaudata . 142 longicaudatus 135, 141 longipes . 178 psamathe 173 Uromysis armata 446 Urothemis sanguinea 30°. signata signata 308 Vv Vanderia ... 396 {Vekunta flavipes 336 Vivipara bengalensis .. 17 bengalensis mandiensis 12,17 Viviparidae 17 xX Xylocopa ... 85, 86 acutipennis 85 albofasciata 85 amethystina 87 attenuata 85, 86 auripennis aed 8 bombayensia are +» 86, 87 xX Page. Page Xylocopa bryorum ... 290 87 | Xylocopa rufescens ... so 87 caerulea nb 200 87 tenuiscapa 206 nopie ate carinata adn ».. 86, 87 tranquebarica wie Ane 87 collaris co hee 87 | Xylocopinac 500 508 85 convexa a6 00 87 dissimilis ana 50 86 dubiosa od Ae 87 Zz fenestrata 000 90% 86 ferruginea mee noc 87 | fZengma fuscinervis ... oe = 400 flavonigrescens ... n00 87 | fZoraida brunnipennis w. 34] latipes ar 300 85 | {Zoraida (Peggiopsis) ala ve «= BA hinata Be 0p 86 | Zoraidinac « «dt malayana ap . 87 | Zoraidini ... nn -» 34] nitidiventris oon 00 | 87, 88 | Zygoptera .. on 309 pictifrons one .. 85, 86 | Zyxomma petiolatum , ne "303, 307 i er CALCUTTA ; PRINTED BY SYPPT. GOYT. PRINTING, INDIA, 8, HASTINGS STREET. i | : \ we cs te : anete : rm 7” fe q I \ tee ok a wake fem ter Rusiiern Fi Linen tien! Moder wan te the Ff a ‘Metes om bt heh @ Tie letine 4 jen (oatiee bea tigt sure s a Pate ee Have gl Gey Peter8 e pieve ou an = fF | eer (ies OY ’ © Revules of tee? 7 ~~ rie ” q by vrs @ H wat os a 2. Mas , i > ; : | \epcinaen ¢ Te. one : te vata ee - Lg a eT: ve 4 oo) oR) FOUL. PINE ) Howe om fee force tate Pacne it a W wales. Kites eh. a ar vy | { 1A vi. ladian Keecis sat Mad RECORDS of the ’ INDIAN MUSEUM (A JOURNAL OF INDIAN ZOOLOGY) Vol. XXIV, Part I. MAY, 1922. PAGE \ New and rare Odonata from the Nilgiri Hills. F.C. Fraser .. te I Observations on the Invertebrate Fauna of the Kumaon Lakes. Ill. The 5 Freshwater Mollusca. B. Prashad .. ae Po) eee Se A Review of the Indian Species of Amblycephalus. F. Wall A OLIO A new Snake from the Northern Frontier of Assam. J°. Wall .. se a9 Structural Modifications in the Fish of Mountain Torrents. S.L. Hora 31 Notes on Fishes in the Indian Museum. ill. On Fishes belonging to fhe family Cobitidae from high altitudes in Central Asia. S. LZ. Hora .. 63 A Note on Bees of the genera a de and Bombus in the Indian rite C. Dover a = a o. « SS A Revision of the Burmese iteiiaitae. B. Prashad Yo, Ik oe APR 14 1923 eS Calcutta : \ Matin igual Mss?” o/ PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY ome PRINTED AT THE BAPTIST MISSION PRESS. 1922. Price Two I upees. we rained te " ry ey © i ‘ t 7 . A i ¥ ed | Wet ¥ * ¢ i hy , ‘ eas repeat wsitibd ; my i: ae eae oa i i tt eS AD tot aa Ay 2 Pare ee ah he al] hi fy pith hy ya . 5 i. * dd 4 " a i " 2 f au fica iy i a Hy ie. dae * a ‘ NS » : . f a Ki . NEW AND RARE ODONATA FROM THE NILGIRI HILLS. By Major F. C. Fraser, I.M.S. (Plate I.) The Odonate fauna of the Nilgiris is of more than ordinary interest in that these hills have furnished some of the earliest known types of Dragonflies. The Nilgiris were first explored by Europeans in 1822 and Rambur described several species of dragonflies in 1842 which had been collected from those hills during the two decades which had elapsed. Unfortunately the descriptions given by this entomologist, although good, are often not too exact and lead to doubt as to what particular insect was described, especially in a case where there are several insects closely similar. Amongst these latter is the description of Indoneura gomphoides (Argia gomphoides Ramb.) which was obviously made from an immature insect and which . does not give the exact measurements. The Baron de Selys in his Synopsis des Agrionines, p. 20, 1842, redescribed I. gomphoides and gave the measurements of the abdomen and hindwing. It is evident that he knew the type and made his description from it, for in 1886, in his Revision des Agrionines, p. 171, he described under the same name, two speci- mens which he had received from Mr. McLachlan and, concerning them, noted that they differed from the type by being larger, by having the wings more rounded and Cw, rather longer. This implies a close; personal comparison of two types and leaves no doubt as to which was the original of Indoneura gomphoides, viz. the smaller of the two species. Of the two other closely related species found in the Nilgiris Disparoneura canningt Fraser has no blue markings and Disparo- neura westermannt Hagen is much larger than either of the two species described by Selys. During the last decade I have had several opportunities of collecting and examining large numbers of specimens of Indoneura over a wide range of the Nilgiris, and I find that there are two very distinct types of I. gomphoides as described by Selys, the smaller of which is undoubtedly Rambutr’s type, and the larger a new species for which I propose the name Indoneura ramburt. Indoneura gomphoides (Ramb). Argia gomphoides, Ramb., /ns. Nevyop., p. 256 (1842); Alloneura et Disparoneura gomphoides, Selys, Syn. des Agrionines, p.448 (1860) 2 Records of the Indian Museunt. [VoL,. XXIV, id., Rev. des Agrionines, p. 171 (1886) ; /ndonewra gomphoides, Laid.,. Rec. Ind. Mus. X111, p. 347 (1917). Male. Abdomen 36 mm. Hindwing 26 mm. The markings in the adult stage are all turquoise blue, the thoracic ground colour is citron yellow only in teneral specimens and soon passes to pale blue whilst the humeral stripes take on a deeper shade. The abdomen is moderately robust and short, and built more on the lines of Calicnemis eximia. The wings are moderately sharply pointed at the apex; «cis situated midway between the two antenodal nervures and meets ab well away from the posterior margin of the wing; Cw, is II cells in length: the stigma is dark brown framed in black ner- vures ; the postnodal nervures in the forewing number 19-22. Female. The markings in this sex are a dirty grey, the humeral stripe often having an ochreous hue. The markings on the end segments of the abdomen are bluish. Habits. ‘This species frequents streams on the kundahs (open grassy country) around Ootacamund, where in restricted localities it may be seen swarming during the early months of the year be- fore the onset of the monsoon. The break of the rains in June leads in a few days to its total disappearance although on fine days or during temporary breaks in the monsoon, a few stragglers may be seen. Indoneura ramburi, sp. nov. Disparoneura gomphoides, Selys, |.c, (1886). Male. Abdomen 44 mm. ~ Hindwing 30 mm. The markings do not differ markedly from those of Indoneura gomphoides and are coloured turquoise blue in the adult, citron yellow in immature specimens. The abdomen is very long and slim when compared to that of I. gomphoides and is built more on the lines of that of Coeliccza renvfera. The wings are more blunt at the apex; ac is situated much nearer the basal antenodal nervure and meets ab almost or just at the posterior margin of the wing ; Cu, is 13 or more cells in length; the stigma is black; the postnodal nervures number in the forewing 22-32. The anal appendages do not differ from those of Indoneura gomphordes. Female. Abdomen 42 mm. Hindwing 30 mm. Except for its greater length, not differing from the female of I. gomphoides. Stigma pale brown. Postnodal nervures in forewing 22-23. Habits. This species is found at a lower level than the former and has a much more scattered and extended range between 3000 ft. to 6500 ft. It is never found in swarms as is I. gomphoides and it often travels far from the neighbourhood of its native streams, extending deep into the jungle. 1922.] F. C. Fraser: Odonata from the Nilgiris. 3 Unlike J. gomphoides it does not appear much before the onset of the monsoon in June and from that time onwards during the rains, gradually increases in nutnber. I have never found the two species in company and they are certainly quite distinct: The type described by Selys is presumably in the McLachlan collection but I cannot say for certain as I omitted to take the measurements when examining the collection in 1920. In males taken at an elevation of 3000 ft., abdominal segments 8 and g show an almost constant invasion of the blue by the black ground colour from the base. ‘This latter colour projects into the blue as a subdorsal streak on each side and limits it also laterally so that on each segment an inverted blue ‘‘T”’ is formed. In all other respects these specimens agree with Indoneura rambur so that they are probably not more than a local variety of it. Type in my own collection, paratypes in British and Indian Museums. Phylloneura westermanni (Selys). Alloneuva westermanni, Bull. Acad. Belg. (2) X, p. 447 (1860); Disparo- neura westermanni, Selys, Mem. Cour. XXXVIII, p. 171 (1886) ; Laidlaw, Rec. Ind. Mus. XIII, p. 347 (1917). Male. Abdomen 51 mm. Hindwing 38 mm. This insect has been lost sight of for many years and some doubt exists as to what genus the insect really belonged to. Re- cently I have secured twelve male specimens of a dragonfly from near Gudalur, Nilgiris, 4500 ft., 26°vi'1g21, which fit the description and measurements given for P. westeymannt so exactly that there can be no doubt but that they belong to that species. The colouring is closely similar to that of I. gomphoides so that the two insects are apt to be mistaken for one another when resting or on the wing and I fell into this error when I took the first specimen of P. westermanni and imagined that I had taken a particularly fine and large specimen of J. gomphoides. The former insect is, however, very much larger and the blue on the abdom- inal segments more extensive, covering the apical half of the 7th segment as well as the whole of the 8th to roth. The venation of the wing is irregular and is of interest in that it shows well-marked traces of a transitional reduction from a complex to a simple form of venation. Rudiments of intercalated sectors are found in the wings of many specimens and the straightening out of a zig-zagged M, is well illustrated. P. westermanni is even more primitive than Indoneura and is not congeneric with the latter as Dr. Laidlaw had surmised; I have therefore placed it in a genus of its own. On the contrary it is more closely allied to Disparoneura, as ab extends outwards as far as Cu,a which it joins, as in all species of the latter genus. It thus differs markedly from Indoneura in which ab curves down- ward to meet the posterior margin of the wing so as to enclose a marginal cell. The primitive nature of the venation, however, separates it from Disparoneura as sharply as the same feature separates Indoneura from typical Caconeura. 4 Records of the Indian Museum. [VoL. XXIV, Postnodal nervures to the forewing 27-29 ; stigma black; Cu, is 13 cells in length or less than half the wing length. Female. Abdomen 46mm. Hindwing 35 mm. Very similar to the male but of stouter build, differing as follows :— Wings uniformly enfumed. In the right hindwing ab is con- nected to the posterior border by 2 transverse nervures. In both forewings it is nearly confluent with the posterior border at its outer part and the space between the 2 nervures can only be detected with the aid of a strong magnifying glass, In 3 out of the four wings there are rudiments of intercalated sectors. No blue on segment 7. Segments 8 and 9 have blue dorsal markings shaped like a German helmet with the top spike directed basalwards but not quite reaching the base on the 8th. Segment ro has the whole of the dorsum blue. Ovipositor of great size and much more conspicuous than in Indoneura. Habits. This species haunts the neighbourhood of mountain streams but, unlike Inmdoneura, is rarely found on low herbage but keeps to the shelter of overhanging branches at a height of 8 to 10 feet from the ground. The female is described from a single specimen taken at the same place as the males, near Gudalur, 14‘vili'21. It was captured in copula whilst ovipositing in water trickling over the surface of a rock. Genus Protosticta Selys. In addition to Protosticta gravely: Laid. described in these Records from Cochin and Kanara, I have to record three new species from the Nilgiris, all closely allied but differing in mark- ings and in the case of one species, in morphology. Two of these were found in company and all have the same characteristic habits. They are found in the beds of rocky, mountain streams, where they keep to the cover of the banks or rocky boulders. In the dark shadows of the latter they may be detected sitting with the body and abdomen held horizontally out and almost invisible save for the chain of tiny white spots on the head, prothorax and abdomen. When disturbed they hover continually with the abdomen held rigidly out in spite of its enormous length and move forward with a series of short, jerky flights. The females are found in almost nocturnal darkness, small caverns amongst the rocks being especially favoured by them for purposes of concealment. They appear to breed in the patchy morass bordering the streams they frequent. Sufficient stress has not been laid on the extraordinary morpho- logy of these dragonflies. The enormous size ol the eyes and the length of the abdomen are outstanding features and with regard 1922.] F. C. FRASER: Odonata from the Nilgiris. 5 to the latter, if compared to the length of the thorax, it is probably the longest abdomen found in any known dragonfly not Panis Mecistogaster. Key to the Protosticta from Southern India. 1. Abdominal segments 8 and g of nearly equal length ; stigma blood red... . BP. sanguinostigma. Abdominal segment 8 more than twice the length of segment 9; stigma black a: 2. 2. Prothorax white marked. with a posterior, dorsal, trian- gular spot = cot g 0 Prothorax entirely bluish white .. ree ... EP. hearseyt. 3. Abdominal segment 8 entirely black P. gravelyi: Abdominal segment 8 with the basal third or half bluish white, the apical third or half black, the dorsal carina iy ‘ of this segment on its basal half finely black ae Py stevenst. — Protosticta hearseyi, sp. nov. (Plate I, figs. 3, 4.) Two males and 15 females, 26'vi'21, Gudalur, Nilgiris, 4500 ft. Male. Abdomen 35 mm. Hindwing 2rmm. | = Head black, labium ashy white; labrum and genae palest blue, the former margined finely with black, the two basal joints of antennae pale blue. . Eyes pale blue changing to olivaceous on the crown and paling beneath. Prothorax pale blue, unmarked. Thorax glossy black, almost metallic on the dorsum, pale blue at the sides. The mid-dorsal carina strongly defined in pale blue. Laterally a broad, black stripe on the 2nd suture and foreborder of metepimeron. Beneath a black spot between the two hind legs and a pair of elongated spots posteriorly which converge on one another as they approach the first spot. Legs bluish white, the two posterior pairs with a linear, black stripe on the femora. Wings hyaline, the apices rather elongate; stigma black, its costal border shorter than the posterior, the inner border oblique ; ; 14 postnodal nervures in the forewing. Abdomen blackish brown marked with pale blue, this colour most marked on the end segments. The sides of segments 1 and 2 whitish, as is also a diffuse streak on the mid-dorsum, incomplete on the apical half of segment 2; pale basal annules on segments 3 to 7 which broaden laterally and obliquely ; segment 8 turquoise blue marked narrowly and apically with black ; segments 9 and 10 black, the former having a bluish marking on its side shaped like a crescent and star, .Relative size of the abdominal segments as for P. gravelyi Laid. Anal appendages of subequal length, about as long as the two last abdominal segments. ‘The superior stout at the base with a spine on the inner side of smaller size and nearer the base than that 6 Records of the Indian Museum. (VoL. XXIV, - found in P. gravelyt, somewhat bayonet-shaped in profile and chelate at the apices, one arm however expanded and roughly quadrate (figs. 3 and 4). Inferior appendages stout at the base, simple, tapering to a fine point and curving up slightly. Female. Abdomen 32°5 mm. Hindwing 22 mm. Very similar to the male and differing as follows :— The labrum is more broadly bordered with black; the eyes are pale olivaceous green changing to pale brown on the crown; the mid-dorsal carina of the thorax is only obscurely whitish at its upper part. : The basal annule on segment 7 is much broader and its border crenate, segment 8 is brownish black with a lateral, quadrate spot of white, whilst 9 is paler brown and marked with a broad, lateral spot of dirty white connected to a smaller spot subdorsally. Segment Io very small, black. Habits. Found in marshy spots hiding in the shadow of rocks or amongst scrub at the sides of precipitous ravines. This species differs from all others by the two sexes being of nearly equal length, by the male having the mid-dorsal carina of the thorax strongly marked with white, and by the prothorax being quite unmarked. Protosticta sanguinostigma, sp. nov. (Plate I, figs. 5, 6.) ! Two adult males, 2 teneral males, 3-vii'192I and 23'vii'1921, Coonoor Rd., roth mile, 1500 ft., Nilgiris. Male. Abdomen 47 mm.” Hindwing 25°5 mm. Head. Eyes bottle green, pale greenish blue beneath and marked uniquely with a broad band of dark blackish brown which begins above and behind and passes obliquely forward and down- ward along the sides. Labrum turquoise blue, narrowly bordered with black : lower part of epistome blue, the rest of head jet black save for an obscure, transverse fascia of pale brown at the back of head. Anterior surface of the two basal joints of antennae pale. Prothorax black above with an oval spot of blue at each side in the middle, the sides whitish. Thorax jet black with a coppery, metallic sheen above, the sides pale blue marked with a lateral stripe of black on the 2nd suture which is bordered diffusely behind with brown. Legs white with a broad, diffuse, pale brown annule near the dorsal end of the femora which are striped in their length with black on the extensor surface. Wings hyaline, the stigma blood red, its costal side much shorter than the posterior, its inner side much shorter than the outer, the latter strongly convex ; postnodal nervures to forewing 16-18. Abdomen very long and slender, black on the dorsum, paler brown on the sides, marked with pale turquoise blue. Segment 1922.] F. C. FRASER : Odonata from the Nilviris 7 3 with a very narrow and obscure basal annule, segments 4 to 7 with broader annules, increasing in breadth from the 4th to the 7th; 8th segment turquoise blue with the apical border more or less narrowly bordered with black, this colour being continued very narrowly along the dorsal carina and tapering gradually towards the basal end of the segment; segments g and 1o all black. The relative size of the segments is much the same as in the former species but 8 is only very slightly longer than 9. Anal appendages much the same as in the last species but the spine at the base is more on the outer side and much stouter, the chelate ends are broader and the broader arm bifid at its extremity. Female. Abdomen 39 mm. Hindwing 26 mm. Very similar to the male but of much stouter build and with a shorter abdomen. Head. Eyes deep bottle-green above, paler green beneath, these two shades of green separated by a thick, equatorial line of black. Rest of head as for maie but the blue on labrum and lower epistome is of a deeper shade. Prothorax blackish above, dirty white at the sides; the posterior lobe with lateral prolongations shaped as two projecting points. Thorax as for male but blue markings of a deeper shade. Wings hyaline; postnodal nervures to forewing 16, in the hind 15; stigma a cherry red; arc distal to the 2nd antenodal nervure., Abdomen black with white or blue markings as follows:— — Ist segment with a blue lateral spot, segment 2 has a bluish lateral basal marking prolonged along the ventro-latera' border, segment 3 has the middle two-thirds or three-fifths of its ventro- lateral border a pale whitish brown, seginent 4 has very obscure basal and ventro-lateral markings, segment 5 has a well-marked basal white annule, segment 6 obscure basal and ventro-lateral markings, segment 7 has a broad basal annule bluish in colour, occupying about one-third of its length, the remaining segments entirely black. Habits. Found in similar situations to the last but more retiring and never coming out into the open. The four specimens taken were in the deepest jungle clinging to maiden-hair fern sprouting from crevices in the rocks The insect is readily distin- guished from others by its red stigma and by the equality in size of segments 8 and 9. Type in my own collection, paratypes in British and Indian Museums. Protosticta stevensi, sp. nov. (PlateI, figs..1,. 2,7.) Five females and a considerable number of males taken on the Coonoor-Metuppalayam Rd., roth mile, 1500 ft., 3°vii'1921 and 24° vil IQ2r. 8 Records of the Indian Museum. - [Vor XXIV, Male. Abdomen 49 mm. Hindwing 22 mm. This species is very closely related to P. gravelyi from which it differs by the greater length of the abdomen and by. the 8th abdominal segment having its basal third of half pale blue instead of all black as in P. gravelyi, ‘The dorsal carina of. this Sep ens is narrowly black in its basal half. ; Female. Abdomen 37 mm. Hindwing 23 mm. Differing from P. gravelyi, which it much resembles, by the difference in the relative lengths of hindwing and abdomen and the markings of the abdomen. ‘The basal annule on segment 7 occupies about the basal fourth and there is no basal annule to segment 8 but a lateral, irregular spot. ‘There is also a large diffuse white spot on the outer side of the eye which is not present in P. gravelyt. The ovipositor of this and other species has a large. promi- nent, stout, upward-turned spine on its dorsal apical surface. Habits. As for the genus, but bolder and to be seen frequently flying in mid-stream. Large numbers of males were seen. on the 3rd of July all with their heads facing up stream and travelling slowly in that direction. A few were seen paired, but the females as a rule kept to the shelter of the scrub lining the stream, where apparently the males sought them. By the 23rd of July the num- bers had greatly diminished and few were seen. _ Type in my own collection, paratypes in British and Indian Museums. Pseudophaea fraseri Laidlaw. Rec. Ind. Mus. X1X, p. 23 (1920). Two females and a large number of males, Gudalur, pens Nilgiri Wynaad, g'vii'1921. Thanks to the kindness of Mr. Laidlaw I have been able to compare the above specimens taken by myself with a paratype of P. frasert and I find that the differences are so marked as to constitute a very distinct race if not a new species. For the present and until I receive Mr. Laidlaw’s opinion I propose to call this race wynaadensis. The differences are as tabulated :— P. fraseri, race wynaadensis. Length of forewing 38 mm. Greatest breadth 7 mm. P. frasevi. _ Length of forewing 35 mm. Greatest breadth 6°5 mm. Apical third of hindwing Considerably more than the opaque, from 85 to 10 mm. | apical third opaque, from 12 to long. 13 mm. long. Length of. abdomen 38 mm. Length of abdomen 43 mm. In addition to the above the anterior pair of femora’ are brownish black on the flexor surface, bright yellow on the extensor. The dorsal wedge-shaped line is bright turquoise blue and is sosharp- ly contrasted with the black background as to give the impression 1922.] F. ©. Fraser: Odonata from the Nilgirts. 9 that it is phosphorescent and in fact it appears to glow like the lamp of a glowworm. The second line on the thorax is ochreous in colour, and the first 6 segments of the abdomen are blood-red. Female. Measurements of the two specimens :— Abdomen 34 mm. Forewing 35 mm. Hindwing 32 mm. ” 34°5 mim. a 36 mm. he 34 mm. The wings are slightly enfumed throughout and the apices of both are tipped with brown, especially the hindwing where this colour extends in as far as the proximal end of the stigma. For the rest, the female does not differ from typical P. frasert except that the antehumeral lines do not meet anteriorly on the thorax, but are parallel throughout. Habits. ‘This insect is found perched on plants and twigs overhanging the borders of the streams it, frequents. Unlike most if not all other Calopterygines it is frequently seen settled with the wings outspread and the abdomen raised at an angle like many Libellulines. The females are rarely found near water, but penetrate into the neighbouring jungle where they may be found paired with the males. : Type male and female in my own collection, paratypes in British and Indian Museums. The distribution of this insect and P. dispar is extraordinarily local. The two never apparently occur together, but may be found on the same stream at different altitudes. At Gudalur P. dispar is found at an elevation of 4000 to 4500 ft., often in considerable numbers, whilst two miles further down the valley at an elevation of 3500 ft., P. fraseri is quite common. On the opposite side of the Nilgiris P. dispar is met with at elevations varying from 3500 to 6000 ft. at Coonoor, P. fraseri being entirely absent. Phyllomacromia nilgiriensis Fraser. Female. A single female, 24'vii'1921, near Kalar. The type is in the British Museum and was takenin June, 1917, by myself at astream not far above Kalar. I have now secured another specimen, also a female, taken on the same stream at about 100 yards from where the type was caught. The wings of this specimen are enfumed throughout, the colouration forming a diffuse network corresponding to the nervures. The saffronation extends out nearly to the trigones; otherwise it does not differ from the type. Taken whilst ovipositing in wet sand which formed the floor of a small, dark cavern amongst rocks bordering a mountain stream. ee ewe ees ewe eer eee D - 4 AWA. dete ti ell wpibaeay ta, met a : 7 Dia & Me Sett Te) —— pegs , I) ' a! i ) ‘h 1% wot ee of A yo 7. a bry? ( ia Paar? 4 a in a. i! me OF sigs Te ed POL TAY 5 7 7 1 ; : rh ) 7 Abadi h Tane V? i) eis ii vod 2 no wae Tecra willy an mee i i one RT we Ma le ; = © > orgy - se 7 el aM vier Unie ou me ‘a vu + ara! Li a ty Pony ut. wf LA Fic. 3) EXPLANATION OF PLATE I. 1.—Terminal abdominal segments and anal appendages of Protosticta stevensi, dorsal view. 2.—The same seen from the side. 3.—Terminal abdominal segments and anal appendages of Protosticta hearseyi, dorsal view. 4.—The same seen from the side. 5.—Terminal abdominal segments and anal appendages of Protosticta sanguinostigma, dorsal view. 6.—The same seen from the side. ’ 7,—Head and thorax of Protosticta stevensi. PLATE. XXIV, 1922. REC. IND. MUS., VOL. STTIH IMIOTIN AHL WOU VLVNOGO Ss OBSERVATIONS ON THE INVERTEBRATE FAUNA OF THE KUMAON LAKES. III. THE FRESHWATER MOLLUSCA. By B. Prasuap, D.Sc., Assistant Superintendent, Zoological Survey of India. The present paper on the Mollusca of the Kumaon Lakes is a continuation of the two papers published by Dr. N. Annandale and Dr. S. W. Kemp inrg1t2! on the Invertebrate Fauna of the Kumaon Lakes. It is based on a small collection of molluscs made by Dr. Kemp in 1911, and the large series of specimens obtained by Mr. S. L. Hora and myself from the various lakes and streams in August, 1920. The lakes visited were: Naini Tal, Sariya Tal, Khurpa Tal, Sukha Tal, Bhim Tal, Naukuchia Tal, Sat Tal, Damianti Tal and Malwa Tal. Collections were also made in the hill-streams running in the vicinity of these lakes. Notes on the situation, etc., of most of these will be found in the paper cited already, and I only include here a few general observations on the physical conditions of the lakes during August, 1920, with more detailed notes on the areas not visited by Dr. Kemp in ro1rr. The water-level in all the lakes was much higher at the time of our visit than at that of his, owing to large quantities of water that had been brought during the rains from the extensive catch- ment-areas atound each of them; the area of the lakes also was much larger. The shallower regions of the lakes, which in May, Ig1It, had been found to harbour rich growths of sponges and Polyzoa, had five to six feet of water, and sponges and Polyzoa were practically non-existent. The water in most of the lakes was Clear and held very little mud in suspension. All the lakes with the exception of Malwa Tal had, along the margins and up to a depth of about ten feet, thick growths of aquatic plants such as Chara, Potamogeton, Hydrilla and Nelumbium and large quantities of algae, such as Spirogyra. Sponges and Polyzoa were in a few cases found growing on the stems and leaves of these aquatic plants. The fauna as a whole was vety poor. The Peridiniid, which was found to be very common in 1871” and rather scarce in 1911, was only found in very small numbers in Bhim Tal and in still smaller numbers in Naini Tal. Leeches were plentiful under ! Rec. Ind. Mus. VII, p. 129 (1912). 2 Ann. Mag. Nat. Hist. (4) VII, p. 229 (1871). 12 Records of the Indian Museum. [VoL. XXIV, stones near the margins and a species of Glossosiphonia was found parasitic on V. bengalensis mandiensis. Dragon-fly larvae, both Libellulids and Aeschnids, were fairly common, but the number of aquatic Hemiptera and Coleoptera was very small; in Malwa ‘lal the Hemiptera were a little more numerous. A special feature of Bhim Tal was the large numbers of Chironomid larvae which were living in tubes attached to submerged stones and tree-trunks. Molluscs of the genera Limnaea, Gyraulus, Segmentina, Hippeutis and Sphaerium were found in varying numbers in almost all the lakes, but V. bengalensis mandiensis was only found in Naini Tal and Khurpa Tal. The conditions as to Crustacea were identical with what was found to be the case by Dr. Kemp in trogrr. Fishes of the genera Oveinus, Barilius, Barbus aud Ophiocephalus were common in all the lakes, and a species of the genus Labeo was also found in Malwa Tal. => SARIVA TAL, This is a rather small lake, or rather a inarsh in the course of a rapid hill-stream. It is situated at a distance of about three miles to the west of and at a slightly lower level than Naini Tal. It is a depression in the course of the hill-stream with about 3 to 8 feet of water; the area is not very large and the current in the lake is much slower than in the hill-stream. The entire area at the time of our visit supported a very thick vegetation consisting mainly of Chava, Potamogeton and large quantities of algae. No Sponges or Polyzoa were seen. Dragon-fly larvae of Sympetrum sp.,' all too young to identify specifically, were fairly abundant. The Molluscan fauna was very poor; only a few Limnaeae and Planorbids were found after careful search. KHURPA TAL, Khurpa ‘lal is situated at a distance of about five miles from Naini Tal at an altitude of 5365 ft. It occupies a nearly circular depression surrounded on all sides by high hills. The area during the dry season is rather small, but the lake becomes much more extensive during the rains. The lake was stated to be over ten feet deep near the middle, though near the margins it is quite shallow. It is not fed by any streams and there is no regular outflow of water. At the time of our visit there was no real aquatic vegetation and the water was quite clear. The fauna, which was very poor, consisted of the same spe- cies of fish as are found in the other lakes, a few dragon-fly larvae of the species Anax parthenope Selys, some Limnaeae and large numbers of V. bengalensis mandiensis along the banks, feeding on algae growing on stones. No Planorbids were seen. 1 | am indebted to Major F. C. Fraser, I.M.S., for the identifications of the dragon-fly larvae. 1922. ] B. PRASHAD : Kumaon Lakes. 13 SUKHA TAL. In May, IgII, this area was found to be quite dry, but a few Cladocera and Ostracoda were reared out of some earth brought back to Calcutta. In September, 1920, practically the whole of this area had 2 or 3 feet of water. The vegetation was very scanty, consisting only of a few stray plants of Potamogeton, but algae like Hydrodictyon and Spirogyra were very abundant. A fair number of Cladocera and Ostracoda were collected and water-bugs were plentiful near the margins. Larvae of dragon- flies of the species Anax parthenope Selys, Lestes cyanea Selys and Orthetrum triangulare Selys were fairly abundant. No Limmnaea was seen, but Planorbids of the genus Gyraulus were common amongst the algal filaments. DAMIANTYI ‘LAL. Situated at about the same level as the Sat ‘I'al, but about a mile to the east of it, is a spring known as the Damianti Tal. A small stream, which has been greatly widened and deepened for irrigation purposes, leads down from the spring to the valley below. At the time of our visit both the spring and the mouth of the stream were fullof cow-dung with many submerged grasses growing in them. The only interesting animals collected here were a few Limnaeae, a few Gyraulus and some bivalves of the common Indian species, Sphaerium indicum. The hill-streams were very uninteresting from the molluscan point of view. In the upper regions, where they are fairly rapid, no molluscs were found, but lower down they had a few Mollusca of the families Melaniidae, Planorbidae and Hydrobiidae. As these Molluscs were collected outside the limits of the Tal area and as they belong to common Gangetic species, I do not propose to include them in the present paper. Family LIMNAEIDAE. Genus Limnaea Lamatck. Two species of this genus, L. acuminata and L. luteola, were collected in the Tal area. The former is the common species and is represented by a number of forms or phases, while the latter has a much restricted distribution and was found only once ina pond on the roadside near Naukuchia Tal. Limnaea acuminata Jain. 1881. Limnaea acuminata, von Martens, Conch. Mitth. 1, p.75, pl. xiv. 1921. Limnaea acuminata, Annandale and Prashad, Rec. Ind. Mus. XXII, p. 568. pl. vii, figs. 1-3, text-fig. 12. In the paper cited above Dr. Annandale and I have given reasons for considering most of the Indian species of the older 14 Records of the Indian Museum. [Vor,. XXIV, authors as being only forms, variations or phases of this highly plastic species. A few further remarks are necessary in view of the material now collected from the Tal Lakes. Iarge series of specimens collected in Naini Tal, Sariya Tal, Khurpa Tal and Malwa Tal are like the typical L. patula Troschel figured by von Martens. Some of the shells from Naini Tal are referrable to the form amygdalum Troschel, while quite a large series of specimens are intermediate between the two ferms, In view of these facts our conclusions as to the desirability of sup- pressing tuese names seem to be justified. We were, however, in doubt as tothe form chlamys Benson. With a large series of speci- mens from‘a'marshy area near Bhim Tal and from Naukuchia Tal, I believe this to be a well marked phase. Its clongate shape with a subcylindrical body-whorl, the comparatively short spire, rather narrow and elongate mouth with a nearly straight outer lip and the sulcate sculpture are quite characteristic of this phase. The form referred to as Sowerby’siventricularius by Annan- dale! and ventricularius Kuster in the paper cited above was included on the authority of some Indian Museum specimens iden- tified by Preston. ‘The three shells, as I now find on comparison with the large series of shells from Naini Tal, are all young speci- mens of the form amygdalum Troschel and have nothing to do with the species L. ventricularius Parreiss, from ‘‘ Ostindien,’’ Limnaea [uteola Lam. 1920. Limnaea luteola, Annandale, Jud. Fourn. Med. Res. VAUII, p. 10g. - This species, as was stated by Annandale in the paper cited above, is identical with Deshayes’ L. succinea, but as Lamarck’s name has priority, it should be known as L. luteola. It is not very abundant in the Gangetic Valley, but is the common species of Peninsular India. The occurrence of large numbers of speci- mens in a muddy pool near Naukuchia Tal at an altitude of over 4000 ft. is, therefore, of special interest. In this pool the speci- mens were found attached to the stems of Potamogeton and toa grass which were growing abundantly in the muddy waters of the pool. All the specimens are quite typical and are fully grown. Genus Gyraulus Agassiz. 1919. Gyraulus, Annandale and Prashad, Rec. nd, Mus. XVIII, p. 52. 1921. Gyraulus, id. ,1b., XXII, p. 582. This genus is represented in the Tal Area by three species G. convexiusculus (Hutton), G barrackporensis (Clessin) and what appears to be an undescribed species. I do not, however, propose to describe it till the collection of the Indian Museum Planorbidae, 1 Ind, Fourn. Med. Res. VAIL, p. 110 (1920). 1922.] B. PRASHAD: Kusmaon Lakes. I5 now with Monsieur L. Germain of the Paris Museo is returned to India. Gyraulus convexiusculus (Hutton). 1921. Gyraulus convexiusculus, Annandale and Prashad, of. ctt., p. 582. Large numbers of specimens of this species were collected in Naini Tal, Sariva Tal and a hill-stream opening into the north- western corner of Bhim Tal, attached to the stems of Potamogeton and entangled in the filaments of algae like Spivogyra. Gyraulus barrackporensis (Clessin). 1886. Planorbis Barrackporensis, Clessin, Limnaeiden. Mart. Chemn. Conch. Cab., p. 125, pl. xviii, fig. 7. 1886. Planorbis Huttoni, id., 1b., p. 139, pl. xviii, fig. 4 1909. Planorbis barrackporensis, Germain, Rec. ‘Ind. Mus. III, p. 120. 1915. Planorbis (G.) barvvackporensis and P. (G.) huttont, Preston, Faun. Brit. lad. Freshw.-Moll. pp. 121, 120. I agree with Germ: uo in considering P. barrackporensis and P. huttoni as being thesame species. The species is known to occur in such widely separated localities as Barrackpore, Calcutta, Benares and Tibet. In the Tal area we collected specimens of hee species in Naukuchia Tal along with those of Hippeutis caenosus (Benson), Genus Hippeutis Agassiz. 1921. ? Hippeutis, Annandale and Prashad, op. cit., p. 584. In the paper cited above Dr. Annandale and I recently sug- gested that Benson’s Planorbis caenosus, which had hitherto been assigned to the genus or sub-genus Segmentina, agrees with his other species P. umbilicalis in shell-characters and is probably congeneric with it. We further questioned theit being included in the genus Segmentina and suggested that they should probably be placed in the genus Hippeutis. An examination of the soft parts and radula of the European H. fontanus confirms this opinion. Hippeutis caenosus (Benson). 1850. Planorbis caenosus, Benson, Ann. Mag. Nat. Hist. (2) V, p. 349. 1876. Planorbis caenosus, Hanley and Theobald, Conch. /nd. pp. xviii and 18, pl. xxxix, figs. 7-9. 1878. Planorbis caenosus, Sowerby, Conch, Icon. XX, pl. x, figs. 78, a, b. 1878. Planorbis caenosus, Nevill, Hand List Moll. Ind. Mus, 1, p. 246. 1886. Planorbis caenosus, Clessin, op. cit., p. 165, pl. xxiv, fig. 4. 1915. Planorbis (Segmentina) caenosus, Preston, op. cit., p. 127. 1918. Planovbis caenosus, Annandale, Rec. Ind. Mus. XIV, p. 113. This species has been recorded from Jamalpur, Bengal; Man- bhum, Orissa; Bhim Tal, United Provinces; and Vawnghwe Pro- vince, Burma. A fair series of specimens was collected by us in 16 Records of the Indian Museum. [VoL. XXIV, Naukuchia Tal. Unfortunately most of them were dead shells and are not, therefore, available for anatomical study. Genus Segmentina Fleming. 1921. Segmentina, Annandale and Prashad, op. cit., p. 585. In the Tal area we found specimens of S. calathus (Benson), the only Indian species which Dr. Annandale and I were able to assign definitely to this genus. Segmentina calathus (Benson). — 1821, ‘Segmentina calathys, Annandale and Prashad, op. cit., p. 585. We found only a few specimens of this widely distributed species amongst the algae in Naini Tal. The specimens were found only near the shores. Family HYDROBIIDAE. In spite of careful search no representatives of this family were discovered by us in the Tal Lakes and I am very. doubtful as to whether any of them are really endemic in the lakes. While making this rather bold statement I am aware of the record of some specimens of Tricula montana and Bythinia pulchellafrom Naini Tal by Nevillin his Hand-List,! but that does not necessarily mean that the specimens referred to were collected in the lake itself. Ben- son’s type-series of the former species was collected in a small stream flowing into Bhim Tal and probably Stoliczka’s specimens referred to by Nevill were also obtained from some stream around Naini Tal. The only specimen of B. pulchella (also from Stoliczka’s collection, but not now to be traced in the Indian . Museum) must also have been collected outside the lake, as the species is not known to inhabit large areas of clear water. Genus Tricula Benson. 1921. Yyvicula, Prashad, Rec. Ind. Mus. XXII, p. 67. I have nothing further to add to my recent account of the genus and of the species, 7. montana, of the Tal area. Genus Digoniostoma Annandale. 1920. Digoniostoma, Annandale, Ind. fourn. Med. Res. VIII, p. 104. 1921. Digoniostoma, Annandale, Rec. Znd. Mus. XXII, p. 4. The only species which we actually foynd in the Tal area was Benson’s. Paludina pulchella. It has, on shell-characters alone, been recently assigned to the genus Digomostoma, but the radula and soft parts are certainly different from those of D. cerameopoma, the type-species of the genus. I do not, however, wpotese ! Aand-List Moll. Ind. Mus. U1, pp. 35 and 62 (1884). 1922.] B. PrasHAD: Kumaon Lakes. 17 discuss the generic position here, as, in view of Robson’s ' recent remarks, I propose revising all the Indian Hydrobiidae when more material is available. Digoniostoma (?) pulchella (Benson). 1836. Faludina pulchella, Benson, Fourn. As. Soc. Bengal V, p. 476. Large series of specimens of this species were collected by us in a pond along the roadside near Naukuchia Tal at an alti- tude of over 4000 ft., together with specimens of L. luteola. Family VIVIPARIDAE. Genus Vivipara Lam. . In the Tal area this genus is represented by a race of the common Indo-Gangetic species V. bengalensis. Even this race was found to have a restricted distribution, as specie were found only in Naini Tal and Khurpa Tal. Vivipara bengalensis race mandiensis Kobelt. 1921. Vivipara bengalensis race mandiensis, Annandale, Rec. Ind. ° Mus. XXII, p. 271. I have nothing to add to Annandale’s detailed account of this and the allied races, beyond recording the occurrence of this race at such high altitudes as that of Naini Tal and Khurpa Tal. Family CYRENIDAE. Genus Sphaerium Scopoli. 1921. Sphaertum, Prashad, Rec. Jnd. Mus. XX(\1, p. 614. I have nothing to add to my recent account of this genus and of the widely distributed Indian species, S. indicum Desh., speci- mens of which were found by us in Damianti Tal. 1 Ann. Mag. Nat. Hist.(9) VIII, pp. 401-413 (1921). spe he avery. at fe ty ae 1 ae (lied Nebel | a etek wil A! ea at * - Wyte AN Mud pd ie ite vy etal ih Bh morc 4 phat oat ei Me hae dine Sane nla ; act sa i Mul eth wd A REVIEW OF THE INDIAN SPECIES OF AMBLYCEPHALUS. By Coionel F. Watt, C.M.G., I.M.S. I have recently had an opportunity of studying all the re- presentative snakes of the genus Amblycephalus in the Indian Museum and in the Bombay Natural Historv Society’s collections. I propose to add to this material the information derived from specimens I have collected myself, and to review the genus so far as it concerns Indian species. ‘The Indian Museum contains types of what have been up to date accepted as three distinct species, viz. modestus (Theobald), macularius (Theobald) and anderson (Boulenger), but which, I hope to show, should be regarded as a single species. It is to ‘be noted that many of the head-shields in individuals of some of the species are subject to frequent variation owing to confluence. Further I notice that in many specimens the details of the periocular lepidosis are difficult to determine in spirit specimens. The praeocular, subocular, and postocular are difficult to differentiate owing to creases which simulate sutures, and it is sometimes impossible to be certain whether merely a crease is present or a genuine suture. CHARACTERS OF THE GENUS. General. Short snakes not exceeding about 610 mm. (2 feet) in length. Head bluntly rounded anteriorly, separated from the body by a much constricted neck. Snout short, feebly declivous, with no canthus rostralis, Nostril piercing about the middle of an entire shield. Eye large, with brilliant yellow iris, and a verti- cal pupil. Body strongly compressed. Tail short, about one- sixth to one-ninth the length of the body. Lepidosis. Rostral rather broader than deep; the portion visible above less than the suture between the internasals. Inter- nasals: a pair; broader than long; the suture between them less than half the internaso-praefrontal sutures. Praefrontals: a pait ; the suture between them shorter than the internaso-praefrontal sutures. Touching the eye (except in cavinatus). Frontal longer than the snout, longer than the supraoculars, shorter than the parietals. Nasal entire. Loreal: one; touching the internasal ; touching the eye in some species. Praeocular variable; one usually present. Absent in some species. Postoculay variable; usually one, sometimes none. Swboculars variable; one to four. Temporals variable ; one to three anterior. Szzpralabials: 7 or 8; the 1st and 2nd touching the nasal, usually none touching the eye, 20 Records of the Indian Museum. - [VOL ost G. the last longer than the two preceding shields. Mental variable ; sometimes touching the anterior sublinguals, sometimes not. Swb- hnguals: three large pairs, roughly symmetrical, with no groove between them. Injralabials very small. Costals: in 15 rows in the whole body-length ; smooth, or some of the median rows feebly keeled. No apical pits. Vertebrals usually enlarged; arising by a gradual development, not a confluence of rows. Ventrals well developed, broad ; the first the largest of the series. Anal entire. Supracaudals in even tows, vertebrals not enlarged. Swbcaudals in pairs, OSTEOLOGICAL CHARACTERS. Praemaxilla about as broad as high. Nasals forming an os- seous suture with the frontals. Fvontals contributing to the rim of the orbit ; not constricted at midorbit. Pvaefrontal suture extend- ing beyond the middle of the frontal. Postfrontal not touching the frontal. Parietal contributing to the rim of the orbit. Supratem- poral rudimentary; not projecting beyond the quadrate ante- riorly. Quadrate well developed ; oblique from above backwards, Columella auris extending from about the middle of the quadrate to the exoccipital. Mavxilla about half the length of the dentary ; expanded in depth anteriorly; expanded laterally posteriorly. Teeth 1 to 6; anododont, syncranterian, scaphiodont. An edentulous space anteriorly, also posteriorly in some species. Ec- topterygoid well developed ; expanded anteriorly to overlie the posterior expansion of the maxilla. Palatine short; expanded laterally anteriorly. Teeth I to 3; anododont, kumatodont or scaphiodont. An edentulous space anteriorly, and in some species posteriorly. Pterygoid long. Teeth 7 to 20; anododont, scaphio- dont. Mandible. Angular present. Splenial present. Coronoid absent. Dentary about twice its distance to the quadrate. Teeth 15 to 23; anodadont, scaphiodont. Occipitals. The condyle is horseshoe-shaped, and formed by processes from the basioccipital and exoccipitals. Vertebrae. Neural spines. Absent on the atlas. Well de- veloped and as long as the body on the axis. Short and obliquely set backwards on the 3rd and 4th vertebrae, nearly as long as the body in the succeeding corporeal, and the caudal vertebrae. Hypapophyses. Well developed and vertical on the atlas. Bifid on the axis, the anterior vertical, the posterior obliquely set back- wards, Disappearing in the vertebrae in the second-eighth of the body.! Absent on the first two caudal vertebrae. ‘Iwo, laterally placed, on the 3rd and succeeding caudal vertebrae. Costae. First as long as the second, articulated to the 3rd vertebra. Last bifid, the outer ramus about one-third as long as the inner. Pseudocostal processes. Bifid on the Ist, 2nd and 3rd caudal vertebrae, single on the succeeding vertebrae. 1 As I find is the case in over twenty five species of Colubridae where the hypapophyses are not continued to the last vertebra. ‘This site suggests some connection with the shoulder girdle of some ancestral form. 1922.] F. Wat: Indian Species of Amblycephalus. aE Amblycephalus monticola (Cantor). A. monticola, Annandale, 7.A.S. Beng. 1905, p. 176; Rec. Ind. Mus. 1912, pp- 37, 50 and 54; Boulenger, Caz. III, 1896, p. 443; Sclater, List. Sn. Ind. Mus. 1891, p. 66; Wall, F. Bomb. N. H.S. 1908, p. 354; id., ibid., 1909, p. 350; zd., 1bid., 1910, p. 843. _ Colour. Uniform brown of various shades dorsally, lighter in the flanks. A series of narrow, blackish, vertical bars laterally, most distinct in the anterior part of the body, and tending to dis- appear at mid-body or posteriorly. Belly uniform paler brown to sordid yellow, with darker spots or dots. Head brown above. A more or less distinct narrow black bar on the neck, sending forwards a branch to the supercilium, and often another between the parietal shields. A narrow blackish streak from the eye to the ape. , i Length. My latgest specimen, a female, measured 750 mm. (2 feet, 54 inches). > Disposttion. A live specimen that I acquired in Assam ap- parently unscathed, proved to be a very quiet inoffensive creature, that. allowed itself to be handled without betraying any malice. In spite of every provocation I could not induce it to assume an attitude of offence, or bite any object, but it emitted the tongue in a lazy fashion. Its movements were slow, which is not sur- prising in a snake that has so strongiy compressed a body. Food.’ The diet appears to consist exclusively of slugs and snails. I have on some occasions in Shillong removed one or two large black slugs from the stomach, which I was informed were a species of Austenta. Many other specimens contained small snails, some devoid of shell, others with broken shell attached, and once one with a perfect shell. I have known as many as five of these small snails in one specimen. ; Breeding. J have examined three gravid females, and found eggs of such a size and character as to make it fairly certain that this species is oviparous. As many as six eggs were found in one example. The smallest specimens I have seen, apparently hatch- lings, were 168 and 178 mm. (68 and 7 inches) in length, but no dates of capture were available. The anal glands in both sexes furnish a custard-ltke secretion. The genitalia are different frcm those of any other snake I nave examined. They are slender cylindrical organs, which are bifurcate about half the length of their maximum extrusion. Each limb is cylindrical, and from base to apex there is no sign of any of those cartilaginous processes, which are seen in snakes of the families Colubridae and Viperidae. Lepidosis. Praefrontal touching the eye. Frontal hexa- gonal in shape. Length much greater than the snout, greater than its breadth, two-thirds to four-fifths the parietals Swpraoculars length subequal to, or rather greater than the praefrontals, half. to three-fifths the frontal, two-fifths to half the parietals. Loreal touching the eye. Praeocular wanting; replaced by the contact of the loreal with the eye. Postoculay one. I have seen this 22 Records of the Indian Museum. a [MOL XOXGiV: confluent with the supraocular on one side in one specimen. Suboculars usually two, sometimes three. Temporals two an- terior, the lower about half the length of the last supralabials Usually two lying along the parietals. Swupralabials 7, sometime. VPexr-tic. 1.—Lepidosis of the Head in a. Amblycephalus carinatus, Boie. b. A. moellendorffi (Boettgcr). ¢ dA. monticola (Cantor). d. A. andersoni, Boulenger. Type. e. A. modestus, Boulenger. Yype. J. A. macularius (Theobald). Co-type. 8 (rarely 6). In all my fresh specimens I found none touched the eye. In spirit specimens, however, it is not unusual to see the 4th, or the 4th and 5th touching the eye; 7th as long as, or longer than the 6th and 5th taken together. Mental usually touching the ‘anterior sublinguals, rarely not. Costals in 15 rows in the 1922.] F. WALL: Indian Species of Amblycephalus. 23 whole body length; obscurely keeled in the median rows of the posterior part of the body. Vertebral enlarged. Ventrals 181 to 198. Subcaudals 69 to 87. Eye. Diameter subequal to the Supeasentar! three- serous to four-thirds its distance to the edge of the lip. Dentition. From three skulls in my collection. Manilige.: 5 to 7; syncranterian, anododont, kumatodont. An edentulous space anteriorly that would take ‘two teeth. Palatine: 2 ot 3; anododont, isodont. An edentulous space anteriorly that would take about two teeth, and another posteriorly that would take about three. Pterygord: 11 to 13; anododont, very evenly scaphiodont. Mandibular: 20 to 24; anododont, very evenly scaphiodont. Distribution. Eastern Himalayas: Sikkim. Assam: Abor Hills (Ind. Mus.); Naga Hills (Samaguting, Ind Mus.) ; Khasi Hills (J. W.); Sibsagar (Ind. Mus.); near Jaipur (F. W.); Dibru- garth (F.W.). Note.—I ieee dit the authenticity of the record from the Nicobars on the authority of de Roepstorff. The specimen (No. 8888) in the Indian Museum is indubitably this species. De Roepstorff’s name is, associated with two other records equally untrustworthy in my opinion, he being the only authority to record the Indian Polyodontophis ea and the Ceylon Oligodon sublineatus from the Nicobars.' Amblycephalus moellendorffi (Boettger). A. moellendorffi, Boulenger, Cat. II1, 1896, p. 443; Sclater, List Sn. Ind, Mus. 1891, p. 67 Colour. Dirty white or greyish, gedvity mottled with very fine purplish-brown specks on the dorsum. Many small round whitish spots outlined with purplish-brown, showing a decided tendency to form crossbars. A more or less conspicuous whitish collar. Belly irregularly spotted with blackish laterally. Beneath the tail densely mottled with fine blackish specks. Head uniform purplish-brown. Young marked exactly like adults. Length. 350 mm. (1 foot, 12 inches). The smallest specimen I have seen was 162 mim. (63 inches) in length. Habits. The many specimens I acquired on Hong Kong Island were captured in the low scrub jungle on the slopes of the Peak. Lepidosis. . Praefrontal touching the eye. Fvontal hexagonal in shape. Length subequal to or rather greater than the snout, subequal to its breadth, three-fifths to four-fifths the parietals. Supraoculay shorter than the praefrontal, about half the length of the frontal, one-third to two-fifths the parietals. Loreal not touching the eye. Pyraeoculay one. Postocular usually none - (confluent with the subocular). Swbocular a single crescentic shield from the supraocular to the praeocular (sometimes not united with the postocular). Temporals the upper usually as long as the parietals, sometimes divided into two. The lower 24 Records of the Indian Museum. [VoL. XXIV, subequal to the last labial. Swpralabials usually 7 (sometimes 8). None touching the eye; 7th as long as the three preceding shields. Mental not touching the anterior sublinguals. Costals in 15 rows in the whole body length, not keeled. Vertebrals not enlarged. ~ Ventrals 136 to 159. Subcaudals 31 to 50. Eye. Diameter subequal to the supraocular, equal to or rather less than distance to lip. Distribution. Burma, ‘Tenasserim (No. 4870, Ind. Mus). Siam, Cochin China, S. China and coastal Islands. Amblycephalus macularius (Theobald). A. macularius, Boulenger, Cat. II], p. 444, Sclater, List, Sn. Ind. Mus. eet p. 67; Wall, Rec. Ind. Mus. 1900, p. 149. . modestus, Boulenger, Cat. III, p. 444; Sclater, List. Su. Ind Mus. Tigo. p. 66. A. andersoni, Boulenger, Cat. Ill, p. 444: F. Bomb. N.H.S. XVI, p. 235; Walland Evans, F. Bomb. NHS XIII, p. 61: ; Wall, F. Bomb. N.S. XVIII, p. 783. Colour. Dorsally densely mottled with very fine specks of purplish-brown, with several small round whitish, or parti-coloured whitish and purplish spots interspersed. Ventrally beautifully dappled with purplish-black and white, especially laterally. Head uniform blackish- purple with speckling on the upper lip. A female specimen sent to me from the Southern Shan States is very dark, and has no small round white or parti-coloured spots. Another from the same locality in the Bombay collection (ven- trals 161, subcaudals 42) is uniform in colouration like the type of A. ai ES. % Length. The largest I have seen measured 483 mm. (1 foot, 7 inches) in length. Habits. Captain Venning wrote when sending me a specimen from Kalaw, that it was found at dusk clinging to the tops of some rank grass. Food. As far asl am aware no observations have been made. Breeding. Captain Venning’s specimen, just alluded to, was a gravid female. It was killed on the gth of June, 1913, and con- tained six large eggs. Lepidosis. Pyraefrontal touching the eye. Frontal hexagonal in shape. Length much greater than the snout, three-seconds to four-thirds its breadth, rather shorter than the parietals. Swupra- oculays three-fourths, to equal to, the praefrontals, half to three- fifths the frontal, about two-fifths the parietals. Loreal not touch- ing the eye. Pyraeoculay usually one. (In specimen No. 8024 in the Indian Museum it is confluent with the praefrontal). Peost- ocular usually one. (In the type of modestus it is confluent with the supraocular on the left side, normal on the right.) Swub- oculars usually one crescentic shield. (in the type of modestus, and in specimens Nos. 8025 and 8026 inthe Indian Museum it is divided into two.) Temporals very variable. One or two ante- riorly. (In the type of modestus the upper appears to be con- 1922.] F. Wat: Indian Species of Amblycephalus. 25 fluent with the parietal.) There are usually two subequal shields lying along the. parietals, but these may be confluent, as in the type of anderson. (In the types of anderson: and modestus there is one long inferior temporal, apparently due to a confluence of the two normal shields.) Supralabials 7 (8 on one side in one | example). None touching the eye. Mental sometimes touching the anterior sublinguals, sometimes not. Costals in I5 rows in the whole body length. Some of the median rows keeled. Vertebvals not enlarged. Ventrals 150 to 169. Subcaudals 37 to 52: ; Eye. Diameter subequal to the length of the supraocular, subequal to or rather greater than its distance to edge of the lip. Dentition. From one bad skull in my collection, nearly all the teeth being broken. Maxillary: 3 (4? in the type of modestus). An edentulous space anteriorly that would take three teeth, and one posteriorly that would take two. Palatine: 1?. An edentu- lous space anteriorly that would take three teeth, and one poste- tiorly that would take two. Pterygotd: 7? left, g? right; no edentulous space anteriorly. Mandibular: 23? on the right side, ? left ; no edentulous space anteriorly or posteriorly. Distribution. Eastern Himalayas: Sikkim (Gopaldhara, Dar- jeeling District, No. 18665, Ind. Mus., type of A. andersont). Burma: S. Shan States (Tounggyi, Wall and Evans, and Bombay collection; Kalaw, F. W., Mogok, Brit. Mus.); Rangoon (No. 8028, Ind. Mus., type of ‘A. modestus) ; Tenasserim (Martaban, Nos. 8024, 8025 and 8026, Ind. Mus., types of A. macularius) ; Sukli, Dawna Hills (No. 17034, Ind. Mus.) Indo-China (Mocquard, Rept. VIndo-Chine, 1907, p 48). Note.—I have examined most critically four times during the last sixteen years the monotypes of A. modestus and A. ander- sont, and the three types of 4. macularius in the Indian Museum, and can come to no other conclusion but that all represent a single species. A. macularius has page priority over A. modestus, and both antedate (1868) Boulenger’s A. andersom (1888). I have now examined sixteen specimens. Amblycephalus carinatus Boie. A. carinatus, Boulenger, Cat. III, 1896, p. 445; Sclater, List. Sn. Ind. Mus. 1891, p. 67. Colour. Dorsally brown of various shades, with numerous dark small spots arranged with a tendency to form cross bars. Ventrally yellowish or whitish with darker spots or mottling, which is often heaviest in the median line. An X-shaped dark mark on the nape, and a narrow dark streak behind the eye, sometimes connected with the X. A specimen in the Indian Museum (No. 8022) from Tenasserim is a uniform drab colour. Length. ‘The longest I have examined is 603 mm. (1 foot, 11? inches) long, the tail 120 mm. (43 inches). The smallest, apparently a hatchling, was 184 mm. (7} inches). . 26 Records of the Indian Museum. [VOK. XSXTVy Lepidosis. Praefrontal not touching the eye. Frontal penta- gonal in shape. Length much greater than the snout, three- seconds to four-thirds its breadth, subequal to the parietals. Supraoculars }onger than the snout, subequal to the frontal, sub- equal to the parietals. Loveal not touching the eye. Praeocular one. Postoculay usually one, sometimes absent being confluent with the subocular; rarely two. Swubocular variable. Some- times one crescentic shield, sometimes confluent with the post- ocular, sometimes divided into three or four. Temporals usuaily three anterior, the longest about three-fifths to two-thirds the last supralabial. Three or four lie along the pariectals. Swpra- labials usually 7 or 8 (6 on the right side in specimen No. 8022 in the Indian Museum, 9 on the left side in specimen No. 51434 in the Indian Museum), None touching the eye. ‘The last longer than the two preceding taken together. Coséals in 15 rows in the whole body length ; several of the median rows keeled. Vertebrals not enlarged. Mental not touching the anterior sublinguals. Ven- trals 161 to 199. Subcaudals 53 to 92. Specimen No. 12781 in the Indian Museum from the Burma-Siam Hills has 92 (ventrals 193). Another, No. 11434 from Deli, Sumatra, has 87 (ventrals 187). Eye. Diameter less than the supraocular, subequal to the length of the snout. Dentition. “From the figure in Boulenger’s Catalogue. Vol. III, p. 438. Maxillary: 5; anododont, syncranterian, scaphio- dont. An edentulous space anteriorly that would take two teeth. Palatine: 3; anododont, coryphodont. An edentulous space anteriorly that would take two teeth. Pterygotd: 15 ; anododont, scaphiodont. No edentulous space anteriorly. Mandibular: 18; anododont, strongly scaphiodont. Distribution. Burma: Tenasserim (Mergui; Tavoy ; Burma- Siam Hills; Ind. Mus.). Siam (Malcolm Smith). Cochin China : Lao Mountains (Brit. Mus.). Malay Archipelago: Sumatra (Ind. Mus.); Java (Brit. Mus). Note. I have examined nine examples in the Indian Museum. Amblycephalus hamptoni Boulenger. _ A. hamptoni, Boulenger, F. Bomb. N.H.S. 1905, p. 230. Colour. ‘Pale brown above with numerous blackish bars interrupted on the middle of the back ; two black longitudinal streaks on the back of the head and nape; sides of head and lower parts yellow; a few black dots on the belly, and under theitaill: Length. 555 mm. (1 foot, 9? inches) ; tail 150 mm. (5% inches). Lepidosis. Praefrontal touching the eye. Frontal of hexa- gonal in shape. Length greater than the snout, equal to its breadth, three-fifths the parietals. Supraoculars length equals the praefrontals, three-fifths the frontal; two-fifths the parietals. Loreal not touching the eye. One on the right side, two (+) on the 1922.] F. Wai: Indian Species of Amblycephalus. 27 left. Pyvacoculay one. Postoculay confluent with the subocular. Temporals one ; about as long as the last supralabial. Supralabzals 8 on the right side, 7 on the left; none touching the eye. Mental touching the anterior sublinguals. Costals in 15 rows in the whole body length. Median rows feebly keeled. Vertebrals feebly en- larged. Ventvals 202 (Boulenger), I count them 197. Swubcau- dals : 96. Eye. Diameter subequal to the supraocular, greater than its distance to the edge of the lip. Distribution. Burma: Mogok, S. Shan States (Brit. Mus.). Note.—Known from a single specimen in the British Museum. oN ee Ne Le eee te eee et ie inn Bs phen ‘A noun | Wtnckivg’ tiph etth > se tral 0 k AP ary ny { ns. hal Mt aD i : in af ' Pe : 4 ‘ é ¥ ’ f ri oi 5 pune rr rhe at A NEW SNAKE FROM THE NORTHERN FRONTIER OF ASSAM. By Colonel F. Wau, C.M.G., I.M.S. Oligodon melanozonatus, sp. nov. O. erythvorhachts, Annandale, Rec. Ind. Mus. VIII, 1912, p. 48. Type, No. 16799; co-type, No. 16798. Both in the Indian Museum. Length 513 mm. (1 foot, 8f inches). Tail 83 mm. (3% inches). Lepidosis. Rostral touching six shields, the rostro-nasal shorter than the rostro-internasal sutures. Portion visible above three-fifths to three-fourths its distance from the frontal, one-third to two-fifths the length of the frontal. Imnternasals: a pair. The suture between them equal to that between the prae:rontal fellows, about half the internaso-praefrontals. Frontal: length greater than the snout, equal to the parietals. Supraoculars: length about equal to its distance to mid-internasals, two-thirds the frontal, three-seconds the temporal ; breadth two-fifths the frontal. Nasal entire. .Loveal absent. Pyraeoculay: one. Postoculars: two. Temporal «+2. Supralabials 6.' The tst and 2nd_ touching the nasal, the 2nd the praefrontal, the 3rd and 4th the eye, and the 5th the temporal. Posterior sublinguals shorter than the anterior pair; touching the 4th infralabial. Infralabials 4; the 4th largest, longer and broader than the posterior sublinguals, and touching two scales behind. Costals two heads-lengths behind the head 17, midbody 17, two heads-lengths before the vent 15. Ventrals 171 to 173. Anal divided. Subcaudals 42 to 45, divided. Colour. Dorsally light brown, obscurely mottled with black- ish. ‘Twenty rather ill-defined black bars cross the body to end low in the flanks, and four such bars cross the tail. In the smaller and half-grown specimen these bars are as light centrally as the dorsal brown, and are edged anteriorly and posteriorly with black as in albocinctus. A whitish, black-edged sagitta on the nape, the point directed forwards. Just before this is a black-edged, light brown sagitta with the point on the middle of the frontal shield. An obscure, blackish, praefronto-frontal bar, reappearing below the eye. Belly white with transverse, black, irregularly- disposed cross-bars, many as broad as the ventral shields. Similar marks beneath the tail. | These shields in the larger specimen are as shown in figure 4 on both sides, but in the smaller specimen as shown in figure c on the left side only, a 6th labial being wedged between the 5th and 7th. 30 Records of the Indian Museum. [Vo1. XXIV, 1922.] Locality. Uppet Rotung Valley, Abor Hills, Assam frontier, at about 2000 feet elevation. Note. Dr. Annandale referred these specimens to O. ery- throrhachis Wall, but a revision of the species of this genus from TEext-F1G. 1.—Oligodon melanozonatus, sp. nov. Lepidosis of head: X2. the material available in the Indian Museum and Bombay collec- tions, convinces me that they represent a species hitherto not described.. In O. evythrorhachis the costals are 15 anteriorly, 13 two heads-lengths before the vent. The ventrals are 154 and subcaudals 64. The supralabials are 7. STRUCTURAL MODIFICATIONS IN THE FISH OF MOUNTAIN TORRENTS. By SunpER Lat Hora, M.Sc., Assistant Superintendent, Zoological Survey of India. ConTENTS. PAGE Introduction eas See ae ak 31 Conditions affecting fish in rapid waters. ai anf as 34 Modifications for life in hill-streams ie in & 35 The external form di se $e it ry 35 The scale-covering, etc. 36 The paired fins and the skeletal and muscular structures connected therewith .. He bre 0 on 36 The caudal fin and its peduncle af, 40 The mouth, its position and see the jaws; the barbels the lips and their muscles __... : 5 41 Theeyes ... Be mee 42 The gill-opening, branchiostegal rays and membranes _.... ay 43 The air-bladder £. Fe cists ro 43 Special modifications of the skin a ef ne ae 44 The minute structure of the adhesive Se Dota $e abe ms 47 Cyprinoidea ate Fe Ae a ia 47 Siluroidea .., ae nee wad oct eae 51 Conclusion oe ds ie rs 58 * Origin of the hill-stream fauna Re ay ae Bh 58 Means of dispersal es ce 2 acne eh lg Methods of propagation ns ie Se nt 61 INTRODUCTION. Although in recent years considerable advance has been made in the study of animal adaptations to different types of environ- ment, little attention seems to have been paid to the wonderful modifications exhibited by the fauna of mountain torrents. Ex- cept for a few casual remarks found in descriptions of hill-stream fishes, no detailed account, so far asI know, has been published of the subject. Nikolsky' in 1891 published a paper dealing with the correlation between the shape of the body of fishes and the strength of the current of streams and Annandale,” in two recent papers, has described some adaptive features in the fauna of hill-streams. Nikolsky’s paper is unfortunately in Russian and is not available in Calcutta. From Annandale’s papers I have received much help. Dr. Annandale has visited a large number of hill-streams in India and elsewhere, and was greatly impressed by the interesting adaptations exhibited by the various groups ! Nikolsky, Rev. Soc. Nat. S. Petersb., pp. 137-139 (1891). 2 Annandale, Rec. Ind. Mus. VIII, pp. 29-32 (1912), and ibid., XVI, pp. 113-117 (1919). 20) - Recards of the Indian Museum. [NOI xOXSIN of animals inhabiting these streams; it was at his suggestion that the work here published was undertaken. Apart from their natural position in the animal kingdom, hill- stream fishes may be divided bionomically into two primary groups. The first group comprises those forms that migrate upstream at certain periods of their lives for spawning, etc.; these may be called the temporary inhabitants of these streams. Fishes of this group travel against the current by muscular effort and do not show, to any great extent, special modifications for life in rapid waters. The members cf the second group are the perma- nent residents of the streams and of still smaller torrents and many exhibit extreme adaptations. It is with the latter sree that the present paper is concerned. The greatest handicap in dealing with the subject was the paucity of material available in the Museum or to be obtained from the streams. Species of many of the genera dealt with in this paper were not only poorly represented, but the specimens often consisted of old and badly preserved individuals quite unfit for detailed morphological investigation. In the hill-streams, on the other hand, there may be plenty of fish, but the readiness with which they seek shelter underneath stones or the swiftness with which they dart away makes it extremely difficult to obtain a good series of specimens. Most of the species are, therefore, known from very few individuals. ‘Through the kindness of the Director, Zoological Survey of India, I was allowed to make tours in the Naga Hills, the Manipur Valley, the Khasi Hills, the Kumaon Hills, the Kharagpur Hills and the Darjiling Himalayas. Good collections were made at ali these places. For histological investigation the material, wherever practicable, was fixed either in formol-alcohol or corrosive sublimate; haematoxylin and eosin have chiefly been used in staining sections of the adhesive apparatus, . The taxonomy of the Indian hill-stream genera has hitherto been involved in a state of great confusion and this factor greatly impeded the progress of my work. Ina series of papers,’ chiefly dealing with hill-stream forms, I have tried to elucidate the taxo- nomy of those genera of which sufficient material was present in the collection of the Indian Museum, and I have also worked out completely the collections made by myself in Manipur’ in order to find out the correct names of the fishes with which this paper is concerned. ‘Besides these I have published recently a paper on some rare and new forms kindly sent to us by Mr. G. E. Shaw from the base of the Darjiling Himalayas. In. interpreting the generic position and specific limits of the various species assigned by Day to Erethistes and Psilorhynchus, I have derived great help from this collection. | Hora, Rec. Ind. Mus. XX1X, pp. 195-215, pls. x, xi (1919); ibid. XXII Pp. 13-19, ¢bid., XXII, p. 633, pl. Axiv-xxvi (1921). 2 Hora. Rec. Ind. Mus. XX 11, pp. 105-214 (1921). ¥ 1g22.] S. L. Hora: Fish of Mountain Torrents. 33 The figures illustrating this paper were drawn by me with the help of a camera lucida. The types selected for the study of hill-stream adaptations belong to the two chief orders of Indian freshwater fishes, the Cyprinoidea and the Siluroidea. The genera on which observations have been made are the following :-— Cyprinoidea. Siluroidea. Balitora Gray. Evethistes Mull. & Trosch. Rhavania Hora. Glyptosternum' McClelland. Psilorhynchus McClelland. Pseudecheneis Blyth. Parapsilorhynchus Hora. Glyptothorax' Blyth. Garra Ham. Buch. Laguvia® Hora. All these genera are found only in small mountain torrents, with the exception of certain species of the genus Gavva which descend into streams of fair size. All show special adaptations to this environment. ; The Schizothoracinae and some of the species of Nemachtlus which live in rapid-running rivers show similar but less well- marked adaptations. Some remarks on the nature of the adhesive apparatus of these forms are also included in this paper. . ' While recently attempting to revise the species of the genus Exastoma Blyth, I have found that this generic name canot be employed for the forms to which it is usually assigned. [ propose the following changes in view of the facts given below :— Exostoma Blyth = Glyptosternum McClelland. Glyptosternum McClelland =Glyptothorax Blyth. McClelland (Calcutta Fourn, Nat. Hist. Il, pp. 584-585, and 587-588, 1842) “described five species, Glyptosternon reticulatus, G. sulcatus, G. striatus, G. pectinopterus and G. labiatus under the new géneric designation Glyptosternon. Blyth (Fourn. As. Soc. Bengal XXIX, pp. 153-155, 1860) split up these five forms into four distinct genera, Glyptosternon, Psexdecheners, Glyptothorax and Exostoma. He regarded G. veticulatus from Atghanistan as the type-species of McClelland's Glyptosternon. According to McClelland this species is stated to be “‘ without spines; the first ray of the pectoral and ventral fins soft and pinnate, giving off soft pointed cartilaginous rays along the anterior margin; which are enveloped in the membrane of the fin. The under surface and anterior portion of the body form a flat corrugated surface.'’ Of the several species of Exostoma in the collection of the Indian Museum, all except &. berdmorvet, possess the outer ray of the pectoral and the ventral fins similar to that of McClelland’s Glyptosternon veticulatus; they ought, therefore, to be included in the same genus. Lxostoma berdmorei, Blyth, which is known from a single specimen from Tenasserim, now in a very bad condition, is the type-species of the genus Zxostoma. ‘The pectoral spine of this species is totally different from that of the others and corresponds to those forms which were included by Blyth under‘his genus Glyptothorax. The absence of a“ pectoral disk,’’ which led Blyth to separate the genus Exostoma from Glyptothorax, is not a valid generic distinction, because the thoracic adhesive apparatus of almost all the species included in the genus Glyptothorax may be- come indistinct in specimens which are old or have been badly preserved. The generic name Glyptosternon, McClelland was latinised into Glyptoster- num by Giinther (Cat. Fish. Brit. Mus. V, p. 185, 1864). 2 In a paper published recently (Rec. Ind. Mus. XXII, p. 739, 1921) .I have given reasons for separating Erethistes asperus (McClelland) (Calcutta Fourn. Nat. Hist. 1V, p. 404, pl. xxiv, fig. 2) along with the two new species from the base of the Darjiling Himalayas from the genus Erethistes and have placed them all in a new genus Laguvia. This genus is intermediate in certain respects between Evethistes and Glyptothorax. 34 Records of the Indian Museum. [Vol XXIV, CONDITIONS AFFECTING FISH IN RAPID WATERS. The conditions that influence the fauna of hill-streams are the - following :— (i) The chief factor is the strength of the current, and all the remaining conditions are due to it. The adaptations which are dealt with further on are all due primarily or secondarily to this one cause. The rate of flow of water varies considerably according to the season, but throughout the year its average flow is much higher than that of any stream in level country. This rapid flow of water would render life impossible to many animals if they did not possess special organs of adhesion or other appliances to counteract its influence. In places like Cherrapunji (Khasi Hills), where 458 inches of rain falls in a comparatively short time, the rate of flow of the water must at times be ex- tremely rapid, and at: such times some even of the most powerful fish cannot withstand it for more than a few minutes. It is unfor- tunate that I have not been able to collect any precise data to compare the rate of flow of water and the fauna inhabiting it. This is a case in which co-operation between a zoologist and a physicist is called ior. (ii) Next in importance are two factors on which the very existence of the animals deperds—food and sheiter. In a_hill- streani there is always a sufficient quantity of food, but the only type usually available consists of algal slime covering stones and rocks. There is no opportunity for any other type of vegeta- tion to grow, as it is liable to be uprooted and carried away by the strength of the current. In pools and ditches that are sometimes formed on the bank of the streams, there is generally a growth of water-weeds, but these cannot be referred to as rapid streams. | Certain fishes such as the species of Nemachilus feed on May-fly and Dragon-fly larvae, but this type of food is usually scarce. As regards shelter, there is plenty of it in a hill-stream for little fish. The species of Nemachilus, on the slightest provoca- tion, hice themselves underneath stones. ‘Those who have made collections in the hiil-streams know how advantageous it is to run the net aniong small stones and sometimes to pick uw stones in the net, because in this way all those forms which rest underneath stones are netted. (iii) Hill-streams are never very deep, and their water is usually very clear. Consequently during the day-time the ani- mals have to withstand intense iight. (iv) The water is well aerated as it is constantly in motion. These conditions do not apply to pools that occur in the course of hill-streams, and the fish-iauna of these pools is very different from that of the rapid current. It is possible that those forms which live in rapid waters are sometimes carried into these pools, but I have never come across any instance in which the typical sluggish-water forms have been found in rapid waters. Species of Danio, Lepidocephalichthys, Barbus, Barilius are generally 1922.] S. L. Hora: Fish of Mountain Torrents. 35 met with in the peels. Of these genera only smaller forms like Danio rerio are usualty iound, for they alone are able to find during the flood season sufficient shelter underneath rocks and stones. The pools are, however, sometimes inhabited by large species of Barbus and Bariliuvs which are sufficiently powerful to withstand floeds. MODIFICATIONS FOR LIFE IN HILL-STREAMS. The modifications for life in hill-streams may be considered under the following heading :— ° 1. The external form of the fish and its size. 2. The scale-covering, etc. 3. The paired fins and the skeletal and muscttlar struc- tures connected therewith. 4. ‘The caudal fin and its peduncle. 5. The mouth, its position and shape; the jaws, the barbels, the lips and their muscles. 6. The eyes. 7. The gill-cpenings, branchiostegal ravs and mem- branes. 8. The air-bladder. 9g. Special modifications of the skin. 1. The external form.—Nikolsky (ef. ci7.) has dealt with this - subject but as the text of his paper, which I have not seen is in Russian, I give my own cbservations in full. The fish with which this paper is concerned all live on the bottom, and the form is so modified as to offer the least resistance to the rapid current. The head and bedy are greatly flattened and in Balitora, Givpto- sternum aud in the most specialized hill-stream species of Garra and Glyptothorax the form is almost leaf-like. The ventral profile becomes straight and horizontal throughout and the dorsal profile is but slightly arched. The head is usually smali and semicircular and the snout is trenchant. The Bornean genus Gastromvzon is in shape a typicai hill-stream form. The shape of the body depends ttpon the strength of the current and any deviation from the characteristic form of the fish © is directly preportional to the rate of the flow of water. Thus the form of those fishes that live in places where the intensity of the flow is intermediate between that of a sluggish stream and of a hill-torrent is almost cylindrical, as in Crossochilus latia. Confining our attention to the members of the genus Garra, one can find all possible gradations in shape between such forms as Crossochilus latia and the most specialized hill-stream form such as Balitora. Garra mullya, one of the most widely distributed forms in the genus, iives in ponds, tanks and sometimes in rapid waters. The specimens collected from ponds and tanks are evlindrical, while those collected from rapid waters are sometimes flattened. Great modification in form is exhibited by G. lissorhyn- chus, G. kempi and G. nasvtus, ali of which are known from rapids 36 J Records of the Indian Museum. [VorL. XXIV, in the Fastern Himalayas. In. dealing with the fishes of the Manipur Valley I have shown how the fauna of a stream changes within very short limits according to whether the bed is rocky or muddy. Small size is a distinct advantage in hill streams, firstly because the streams are small and secondly because small forms can find more shelter under pieces of rocks and stones during floods. ‘2. The scale-covering, etc.—In those Cyprinid fishes that take to hill-stream life, the lepidosis undergoes considerable modification. In the Schizothoracinae the scales are small and partly buried in the skin or are totally absent except in the anal and scapular regions. If in a normal Cyprinid genus in which the scales are large and imbricate, the hill-stream forms be compared with those from other types of environment, it wili usually be found that the scales aré greatly reduced on the under surface, and in some cases they disappear altogether. The region of the chest, which» is to some extent employed in the process of adhesion, is the first to be modified, and then, with the increased rapidity of the flow of water, more and more of the under surface becomes naked. In two species of Garra, G. abhovai and G. rossicus, the dorsai surface in front of the dorsal fin is also naked. The reduction of scales on the lower parts is necessitated by the fact that a plain and smooth surface is necessary in order to allow adherence to rocks, I have not been able to understand why the scales should be reduced on the dorsal surface in Garva abhoya: and G. rossicus. Both possess a subcylindrical shape and are not among the most specialized hill-stream forms. E 3. The parred fins and the skeletal and muscular structures con- nected therewith.—The fins are very plastic structures in the anatomy and they have been employed for various functions by diverse groups of fishes. The modifications of the pectoral fins in Flving- - fishes, of the first dorsal fin in Sucking-fishes and of the ventral fins in Gobiidae and Gobiesocidae are a few instances among many. In hill-stream fishes the paired fins are used as organs of adhesion or of locomotion and for both these functions powerful muscles are required. In certaif cases they are probably used also for respiration. . The outer rays of the paired fins are employed for the func- tion of adhesion and the number of the inner rays is consequently increased. In Gastromyzon borneensis there are as many as 26-28 rays in the pectoral and 20-21 in the ventral fins. In an allied Indian genus, Balitora, there are 21 rays in the pectoral and II in the ventral fins. The outer rays of.these fins are greatly thickened and much flattened. Besides an increase in the number of fin-rays of the paired fins, their position and shape undergoes considerable change. ‘The fins, instead of being situated on the under surface of the fish, are pushed outwards and ultimately are placed horizontally on the sides of the body. This change is brought about tor two reasons, firstly to allow the ventral surface to be firmly applied to rocks, and secondly to enabje the fins to act as organs of adhe- ’ ‘ 1922.] S. L. Hora: Fish of Mountain Torrents. 37 sion. As regards the shape of the fins, some of the inner rays are directed upwards against the sides of the body, so that when the outer rays are used for the purpose of adhesion, the inner rays can be kept constantly in motion, probably for the purpose of respiration. I have embodied my observations on this point in an immature specimen of Psilorhynchus in a former paper.' In the genus Glyptosternum (fig. 8) only a few rays of the paired fins are visible from the under surface, while the remainder are reflected up- wards. I have not observed these fishes in nature, but on a recent tour to the base of the Darjiling Himalayas, I was able to verify the observation that I had previously made on the immature speci- mens of Psilorhynchus from the Naga Hills, by keeping a half- grown specimen of Garra annandalet in ar. artificial pond of water in the course of the Mahanadi River. In Parakomaloptera micro- stoma® the shape of the fins is somewhat less modified than is s the case in Glyptosternum. The greatest specialization as regards firetaueeice is found in Gasirvomyzon borneensis. The peaieeale begin with a long base, vertically below the eyes: the ventrals possess long curved bases, which are united posteriorly. Between the bases of the ventral - and the pectoral fins there is a lateral extension of the abdominal skin. “‘ By this arrangement the whole flattened abdominal surface, together with the fins and the flattened lower surface of the head forms an enormous suctorial disc.” 3 - I have already pointed out that the outer rays or the spines, as the case may be, of the paired fins are greatly flattened. Interesting, modifications Y take place in the outer rav of these Z fins in the genus Glyptosternum, ‘‘ soft Z) pointed cartilaginous rays’’ are given Z off along the anterior margin (fig. 1a) 4 to support the striated skin which f, forms the adhesive apparatus. This 3 is described in detail below when deal- ing with the modification of the skin in the formation of the adhesive appa- b ratus. 1 The pectoral and the pelvic girdles a are modified in certain hill-stream = -pxr-yig 1.—Pectoral spine fishes, owing to the acquisition of new of Glyptothorax and Glypto- functions by the paired fins. It is sternum. unfortunate that I have not been able 4. Outer pectoral ray of to study these structures in Psilorhyn- Ch/Ptosternum labratum. h Bi - e : 4. Pectoral spine of Glypto- chus, Bhavania, Balitora and Homal- thoyax berdmores. optera, on account of the paucity of ! tora, Rec. Ind. Mus. X1X, p. 212 (1920). 2 Weber and Beaufort, Fishes Fai Austr. Arch. Ill, p. 20. fig. 5 (tg16). 3 Weber and Beaufort, fbic., p. 2 (1916). 38 Records of the Indian Museum. [VoL. XXIV, matetial; but even in those Cyprinid genera of which material was available in sufficient quantity, I am unable to find any striking modifications. In Garra, for instance, the whole structure is more or less similar to that found in Labeo, except that the adductor and abductor systems of muscles are better developed. At the same time it must be remembered that the fins do not form the chief organs of adhesion in this genus. In the Siluroids, Glyptothorax and Pseudechencis, in which the chief adhesive organ is situated on the chest, the only modifica- tions consist in the fusion of the various bony elements for strength- ening of the girdle. On account of the horizontal position of the fin, the shape of the girdle is considerably changed (fig. 2). TExtT-¥riG. 2.—Dorsal view of the pectoral girdle in Glyptothorax madraspata- nus. ‘ a = interclavicle; 3, 4 = muscles of the pectoral spine. Great difficulty has been experienced in adopting suitable terms for the description of the various structures. I have fol- lowed Parker! in preference to McMurrich? in drawing up my des- criptions. Besides the modifications enumerated above, other charac- teristic specializations are also found in Glyptothorax and Pseude- cheneis. On the ventral aspect of the interclavicular bones (fig. 3) there are keel-like ridges. (fig. 3b) for the attachment of the muscles. These ridges are greatly elevated posteriorly and end in spine-like processes; but they slope down anteriorly and meet each other in the mid-ventral line close to the union of the clavicles and the interclavicles. 1 Parker, A Monograph of the structure and development of the Shoulder girdle and Sternum in the Vertebrata (1868) * McMurrich, Proc. Canadian Just. (n.s-) 11, pp. 301-306 (1884). 1922. | S. L. Hora: Fish of Mountain Torrents. 39 The muscles controlling the movement of the pectoral fin (figs. 2, 3) in Glyptothorax are also interesting. Besides the abductor — 7S _7 s s . =a Ss. 4) ‘ A“ CPs \ b 2 alee s . . ISK Xe TE ‘ >, : 5 : * . 3 ‘ 4 * t of yy é 4 ie ii fle) \s fon ih C., TEXT-FIG. 3.—Ventral view of the pectoral girdle in Glyptothorax madvaspata- nus. a = interclavicle: b = interclavicular ridge ; c = clavicle ; d= pectoral spine; e = pectoral finrays; f = groove for the attachment of abductor muscles; g= groove in the cubito-humeral precess ; / = abductor muscles; 1, 2 and q'refer to the muscles of the pectoral spine. Ae and the adductor systems, there are four special muscles to move the spine. Muscle 1 arises from the anterior grooved and thickened border of the clavicle and also from along its posterior border near the base of the interclavicular ridge against the sides. Its action is to pull the spine towards the body and fold the rays. Muscle 2 takes its origin from the anterior border of the clavicle, further forward than muscle r. Inits course, it passes underneath muscle I and its function is that of expanding the fin. Muscle 3 is very extensive and fan-shaped. It arises along the whole of the surface of the clavicle and the interclavicle on the dorsal side and in its course passes through a bony canal. Its action is the same as that of muscle 2. Muscle 4 is very strong and passes through a passage in the bone. The muscle takes a curved course and its action is somewhat like that of a rope passing over a pully. Its function is that of folding the fin. It arises from the grooved and thickened posterior border of the clavicle. The actions of the various muscles were studied by moving the muscles and by watching their effect upon the fin. It is clear that muscles rt and 4 are stronger than 2 and 3, because it is in the action of folding of the fin that the adhesive function of the outer rays is involved. In those species of Glyptothorax in which the adhesive apparatus is present on the under surface of the pec- toral spine, muscle 1 is the better developed. HO) 7 Records of the Indian Museum. [VoL. XXIV, In the genus Glyptosternum, where the fins act as organs of adhesion, the moidfications in the musculature are more marked (fig. 4) and the arrangement is different. The muscle labelled 2 in Glyptothorax correspond to 5 in Glyptosternum; muscle 4 is the same in both cases. Muscle 1 arises close to the mid-ventral suture of the clavicle and is inserted in the form of a glistening tendon on the anterior border of the pectoral spine. Its action is to ex- pand the fin. Muscle 2 arises near the mid-ventral line and is inserted on the bases of the spine and the first few rays. Its func- tion is, in all probability, to keep the spine and the few outer rays closely pressed against the substance on which the fish may be resting. This muscle is large and is not found in any other genus that I have studied; .it has no other muscles to counteract its action. 5 , Zy Text-FiG. 4.—Muscles of the pectoral fin in Glyptosternum labiatum. Numbers 1, 2, 3, 4 and 5 indicate the muscles referred to in the text. As regards the skeleton of the pectoral girdle in Glyplosternum, there are no bony ridges for the insertion of muscles. Otherwise it is very similar to that of GlyPtothorax. The pelvic fins also possess a special muscle (fig. 5) beside the abductor and abductor systems. This keeps the fins closely pressed against rocks, when the fish is resting, thus enabling it to adhere to rocks by means of striated skin on the under surface of some of the outer rays of the pelvic fin. 4. The caudal fin and its peduncle.—There is a general ten- dency amongst hill-stream fishes to possess a long, narrow, band- shaped caudal peduncle. For example in Nemachilus tenms and N.lhasae the caudal peduncle is more narrow and elongated than in any other species of the genus that I have seen. These two species resemble the Central Asiatic forms figured by Herzen- stein’ and it is possible that these features are correlated with high | Herzenstein, Wiss. Res. Przewalski Central. As Reis., Theil III (2), pls. i-vii (1888). 1922. ] S. L. Hora: Fish of Mountain Torrents. 41 altitudes and rapid running streams. Similarly in the genus Glyptothorax, two species, G. striatus and G. saisit, from the Khasi and the Parasnath Hills respectively, have a dif- ferent form of caudal peduncle from the re- maining species. It is long and narrow. In almost all species of Homaloptera, Bhavania and Balitora and in the most specialized species of the genus Garra the caudal peduncle is simi- larly modified. , As regards the fin, the chief modification consists in the inequal- ity of its lobes. In most cases the lower lobe is somewhat long- er than the upper, as Text-ric. 5.—Muscles of the pelvic fins in in Balitora brucei, Bha- Glyptosternum labiatus. vania australis, Glypto- The right side of the figure shows the ventral view thorax -striatus, Glypto- and the left side the dorsal view of the pelvic girdle. 5 add= adductor muscles ; add = abductor muscles ; sternum labratum and 5 = special muscle. Garra nasutus. In Gas- tromyzon borneensis, though the caudal fin is not deeply forked, the lower portion is longer and stronger than the upper. I was not able to follow the true significance of these modifica- tions, because the moveinents were too rapid for detailed analysis. It may be pointed out in this connection that in Elasmobranch fishes, where the mouth is on the under surface considerably be- hind the tip of the snout, the lower lobe of the caudal fin is much shorter than the upper. I hope to make further observations on this point on another occasion. . 5. The mouth, its position and shape ; the jaws ; the barbels ; the ips and theiy muscles.—The mode of life and the nature of food in mountain-rapids necessitates a change in the position of the mouth and the structure of the jaws. The mouth, instead of being a transverse cleft at the anterior end of the fish, is situated on the under surface considerably behind the tip of the snout. It is usually crescentic or semicircular in outline. The jaws are greatly strengthened and their edges become sharp and cutting. In most cases, Oreinus for example, the jaws are covered with a strong horny covering. This is due to the fact that hill- stream fishes have to strip algal slime from stones for their food. _ Barbels in rapid-waters would be a source of great encumbrance and, therefore, they are much reduced. In most of the hill-stream species they can only be made out after a careful examination. 42 Records of the Indian Museum. [VoOL. XXIV, In Balitora they are short and stumpy and liable to be overlooked. In the remaining hill-stream genera discussed in this paper they are short and thread-like. In Parapsilorhynchus, however, they are short and cylindrical. In Nemachilus and the Homalopterid genera the lips are so modified as to form a sucker with the help of the mouth, and consequently they exhibit diverse modifications and specializations. In the genus Nemachilus the lips are divided in the middle and are greatly swollen, so that when they are pulled outwards away from the mouth, their divided parts form a continuous ring-like sucker. In most cases the skin of the swollen region is plicated, but I have not been able to find any trace of definite spines such as will be described later in the structure of the adhesive apparatus of other genera. I have already described in a previous paper! the way in which, by the action of certain muscles, the lips of Bhavania annandalet are converted into a sucker. In Balitora the thick lips are cut up into several tentaculate processes and when pulled apart they form an effective sucker. In most of the species of the genus Glyptosternum the lips are ‘‘reflected and spread continuously round the mouth, so as to form a broad flat sucker.’’ Similar modifications occur in certain of the most specialized forms of the genus Glyptothorax. ; 6. The eyes.—With the flattening of the form in hill-stream fishes the eyes are more and more pushed towards the upper surface. In forms like Balitora brucei, B. maculata, Glyptotharax saisit, G. striatus., Pseudechenets sulcatus and in almost all species of the genus Glyptosternum the eyes are situated on the dorsal surface and are placed close together. Besides this change in position, they are much reduced in size. To what cause this re- duction is due, I do not know; but itis quite probable that the intensity of the light in the clear shallow waters of the hill-streams may have something to do with it. 7. The gill-openings, branchiostegal rays and membranes.— With the employment of the under surface for the purpose of adhe- sion to rocks and stones, the gill-openings are generally restricted to the sides. Except in the genera Glyptothorax and Laguvia, the gill-openings, in almost all the genera dealt with in this paper, do not extend beyond the base of the pectoral fin on the under sur- face. Incertain species of Garra the openings are somewhat wider, but even in them they are separated from each other by a con- siderable distance. The greatest modification as regards this character has taken place in two species of Glyptosternum. In these the gill-openings are situated above the base of the pectoral fin and there is a short. narrow passage from the interior of the gill-chamber to the exterior. With the restriction of the gill-openings to the sides, it is natural to suppose that respiration will suffer to some extent. Moreover, when a fish is feeding on the algal slime, the under ! Hora, Ree. Ind, Mus, XIX, p, 203, pl. x, fig. 2. 1922.] S. L. Hora: Fish of Mountain Torrents. ae surface of the head and body are firmly and closely applied to the rock to which it may be clinging at the time, and this also will make respiration difficult. In all probability the following factors help hill-stream fishes in respiration :— (i) The water in the hill-stream is better oxygenated and is purer than that of a sluggish stream in a flat country. (ii) By reducing the gill-openings, the fishes are enabled to retain water in their gill-chambers for a comparatively longer time. (iii) The inner rays of the pectoral fins in fishes of rapid streams are held in constant motion when the fish rests against a piece of rock. The movements of these rays may help respira- tion in two ways :— ; (a) The blood may be oxygenated in the rays themselves, or (b) they may force water in and out of the gill-opening. The following quotation from Mr. Chapin’s notes given by Nichols and Griscom! on the mechanism of respiration in Enchi- lichthys dybowskii (Vaillant) when clinging to rocks is very interest- ing :—‘‘ Two examples were brought alive in a basin where they stuck fast to the smooth enamel surface. When thus attached, the water for respiration enters by the back of the mouth, and the movement of the gills often makes the whole fish quiver or move slightly back and forth. Natives say they cling to rocks and eat algae. They can swim rapidly. The mouth is here drawn as though slightly extended ; while sucking, it of course contracts.” The above observations were made on fishes in a state of captivity and require confirmation. The sucker by means of which the fish adheres appears from the figure to surround the mouth completely, and it is probable that the fish uses both lips for adhesion as in the Indian hill-stream forms. The posterior jaw is in almost all cases more highly specialized for rasping the algal slime from the rocks than the anterior jaw and under the circumstances detailed above, it seems highly improbable that water can enter the gill cavities from the back of the mouth when the fish is either feeding on algal slime or clinging to a rock. With the reduction of the gill-openings and the backward shifting of the mouth on the under surface considerably behind the tip of the snout, the branchiostegal rays and membranes are greatly reduced. Usually these structures on the two sides of a fish such as Labeo rohita meet and overlap on the under surface, but in hill-stream fishes, with the exception of those belonging to the genera Glyptothorax and Laguvia, they form an obtuse angle on the under surface, if they meet at all. 8. The air-bladdey.—The bladder in the hill-stream forms shows considerable degeneration and in 1893 Bridge and Haddon* ! Nichols and Griscom, Bull. Amer. Mus. Nat. Hist. XXXVII, p. 720, pl, Ixxvi, fig. 3 (1917). 2 Bridge and Haddon, Zyans. Phil. Soc. London, vol. 184, part I (B), p. 305 (1893). 44. Records of the Indian Museum. [VOL XXIV ; attributed the reduction to the following causes, which bear repeti- tion even to-day. They say :—‘‘ The causes that have led to the degeneracy of the air-bladder in so many forms are in many in- stances not difficult to trace, and, as in so many Physoclist - Teleostei, the assumption of a purely ground habit of life is prob- ably the most important one. Not a few of the genera of Siluridae abnormales inhabit the comparatively shallow waters of rapidly flowing mountain streams and torrents often living at a considerable altitude, and in general habit are not unlike our common English Loaches. Many are provided with an adhesive, apparatus on the ventral surface of the body between the pectoral fins for attachment to stones, so that they may be enabled to, withstand the force of mountain torrents. Such fishes when not in motion by the exercise of their fins probably rest upon, or attach themselves to, the river bottom, and the utter uselessness and probable harmfulness of an air-bladder as a hydrostatic organ under such conditions is no doubt the cause of its degenerate and rudimentary conditions in such Siluroids as Sisov, Pseudecheneis, Glyptosternum, Euclypiosternum, L:xostoma, Amblyceps, etc.” - I have dealt with this interesting organ at some length in my pre- vious paper and have shown in the case of the genus Garra that the reduction in the organ is directly proportional to the strength of the current of the streams in which the fish live. In almost all the highly specialized hill-stream forms such as certain species of Loaches, Homalopterid fishes and the forms included under the Silurid genera, Glyptothorax, Glyptosternum and Pseu- decheneis the bladder is, divided into two lateral chambers (fig. 6) which are more or less connect- ed with each other by a short, marrow transverse tube. Moreover the blad- TEexr-r1G. 6.—Air-bladder of Glyptothorax der is wholly or partially madraspatanus. - encapsuled by a_ bony Stith case in almost all cases. 9. Special modifications of the skin.—Under this heading I include the diverse forms of modifications exhibited by the skin in the formation of adhesive organs. ‘The simplest form of special- ization occurs in Cyprinid fishes, where the skin covering the under surface of the few outer rays of the paired fins is greatly thickened and becomes cushion-like in places. By these cushion-like pads the fishes are enabled to cling to rocks and hold their own against a rapid flow of water. In the Silurid genera the skin instead of being plain is thrown into grooves and ridges. Such striated portions of skin may occur anywhere on the under surface of the fish but are generally found in the anterior third of the body. I have found such striated surfaces on the barbels, on the sides of the mouth, 1922.] S. L. Hora: Fish 0} Mountain Torrents. 45 on the chest between the bases of the pectoral fins and lastly on the under surface of the pectoral and pelvic spines, and I have been able to make out a series showing the gradual specialization of the adhesive apparatus in the Silurid genera. | The genus Erethistes comprises small hill-stream forms in which the under surface of the body is smooth and greatly flat- tened. In one member of the genus, F. elongata, the structure is, however, somewhat different. The whole of the chest and the belly (fig. 7, 6) is rugose and shows low, bnt well-marked striations. In the forms which I have assigned to my new genus Laguvia, these corrugations are restricted to the chest and the belly is quite smooth (fig. 7, a). This feature is still further marked in the Text-riG. 7.—Under surface of head and chest of Laguvia sp. and Ervethistes elongata. a. Laguvta sp. b.. Evethistes elongata. members of the genus Glyptothorax, where a definite U-shaped or V-shaped adhesive apparatus consisting of folds of skin is present on the chest hetween the bases of the pectoral fins. In certain species of the last genus from very rapid waters an adhesive ap- paratus of a similar nature is also present on the under surface of the pectoral and pelvic spines. In Pseudecheneis sulcatus the skin is somewhat differently modified on the chest, but the stria- tions on the spines of the paired fins are of a similar nature to those of the preceding genus. | Specialization has proceeded along another direction in the genus Glyptosternum. Here the skin on the under surface of the spines (figs. 8a and c) is striated and each ridge is supported by a short, pointed, cartilaginous ray given off from the outer side 46 Records of the Indian Museum. [Vor XXIV, of the first pectoral and pelvic rays (fig. 1, a). Besides this the striated region is supported by a definite, highly specialized tissue (fig. 16, s.¢.). The chest is absolutely devoid of any adhesive ap- paratus. In certain species of the same genus the under surface of the barbels and the skin near their bases is striated (fig. 8a). TextT-F1G. 8.—Structure and form of the pectoral fins and the position of the adhesive apparatus in Glyptosternum labiatum. a. Under surface of head and chest showing striated skin on pectoral spines and on maxillary barbels. 6. Pectoral fin showing reflected inner rays. c. Under surface of pectoral spine highly magnified. The disc of Garra (fig. 9a) with its associated structures is an efficient type of adhesive organ. The disc consists of a central callous portion (h) and of free tuberculated lateral and posterior borders (g). Its anterior border is formed by the poste- rior labial fold (f) which in its development has replaced the 1922.] S. L. Hora: Fish of Mountain Torrents. 47 posterior lip. The anterior labial fold (a) is fringed and tubercu- lated and helps the fish in adhering to rocks. A rudimentary form of disc has recently been described by me! in Parapsilo- rhynchus discophorus. The disc of Garra works on the suction principle. In the middle of the under surface of the callous portion, a strong tendon (fig. 9b, 2) is inserted and attached to the urohyal (7), so that when the urohyal is elevated, the callous portion of the disc is drawn in and thus a cavity is produced which is surrounded by fringed borders. ‘These fringed and tuberculated borders are pro- vided with efficient organs of adhesion as will be seen later when dealing with the minute structure of the adhesive organs. Text-ric. 9g—The disc of Garra and its associated structures. a. Disc as seen from the under surface. b. Dissection of disc from the dorsal surface to show the mechanism for suction. a =anterior labial fold ; 6, 6! = anterior jaw ; ¢ = mouth opening j d = poste- rior jaw ; e= connectives ; f f = posterior labial fold; g, g’ =free tuberculated border of the disc; h = callous portion of the disc ; 7 = tendon joining the centre of the callous portion of disc with urohyal; 7 = urohyal ; k = muscle joining the two sides of the anterior jaw. THE MINUTE STRUCTURE OF THE ADHESIVE APPARATUS. The simplest form of adhesive apparatus is found among hill-stream fishes of the order Cyprinoidea. It consists of the thickened skin covering the under surface of the few outer rays of the paired fins. In a transverse section of such a structure in Bhavania annandalei (fig. 10) the following arrangement may be seen :-—The epidermis consists of several tiers of cells, varying in shape and size with their depth and resting on a loose ccnnec- tive tissue (c.t.), which constitutes the dermis. The outer epider- mal layer is modified into stiff and strong spine-like processes (s), which are somewhat curved near their extremities. The inner limit of these processes is not well-defined and they appear to rest | Hora, Rec. Ind. Mus. XXII, pp. 13-19, figs. (1921). 48 _ Records of the Indian Museum. [Vor. XXIV, upon a homogeneous layer of protoplasm; they occupy as much as one-third of the total thickness of the epidermis and their inner ends are broad. ‘The nuclei of the spines are somewhat oval in outline and are placed in the proximal half; each is surrounded hy a whitish halo. The pro- toplasm of the spine and of the basal homogeneous layer stains lightly with hema- toxylin, eosin and borax car- mine. The deeper tissue takes up the stain readily. Below the homogeneous pro- toplasmic layer, the tissue consists of several layers of almost. rectangular cells (¢. c.) each with a distinct nu- cleus in the centre. The cells diminish in size with the depth of the tissue and become more and more ir- regular in form and arrange- ment. ‘The interspaces be- tween them become broader and in certain cells two nuclei are present. Below these and immediately TextT-r1G. 10.—Transverse section through above the basal epidermal a portion of the adhesive pad on the outer 5 ; rays of the pectoral fin in Bhavania annan- layer (0.c.) there is a tier of dalei. small, more or less regularly s= spine; e.c.= epithelial cells; b.c.= arranged cells, the nuclei basal epithelial cell ; c.¢. = connective tissue; of which are solid, deeply c = cavities for blood vessels. staining ovoidal bodies. The basal layer of epidermal cells is made up of columnar tissue; the nuclei are oval and lie almost in the middle of the cell, or nearer its upper than its lower ends. The upper as well as the lower limits of the basal cells are hardly distinguishable and both these ends stain lightly. The nuclei appear to be in a state of mitotic division as the chromatin substance in them is greatly diffused. The connective tissue (c.t.) below the basal layer of the epi- dermis is very loose and is richly interspersed with cavities (c) of the nature of blood-spaces. The nuclei are greatly elongated and stain deeply. The cell-limits in this tissue are not marked and the whole of the tissue is not so deeply stained as the middle layer of epidermal cells. In the genus Garra this form of adhesive apparatus is sup- plemented by the presence of the characteristic disc behind the posterior jaw on the under surface. Before dealing with the struc- ture of the adhesive disc, I propose to give a short account of that of the integument in this genus. 1922.]} S. L. Hora: Fish of Mountain Torrents. 49 In a vertical section of the skin (fig. rr) covering the tip of the snout, where scales are of course absent, the epithelial region (ep.d.) is made up of a homogeneous mass of protoplasm with a large number of nuclei scattered in it. The nuclei are aggregated either near the base or near the apex of the epithelial region ; some of them are surrounded by a white zone. Near the upper surface are present a number of large ampulliform gland-cells (g.c.). The protoplasm of the gland-cells is restricted to the periphery or to the base and the nucleus generally occupies the centre of the basal protoplasm. In the middle of the homogeneous epithelial mass ‘TExT-F1G. 11.—Transverse section of the integument of Garva annandalei from the tip of the snout, x 435. ep.d. = epidermis; d =dermis; g.c. = gland-cell ; c.c. = clavate cell: b.0.= blood vessel. are found big “ clavate cells” (c.c.) forming as it were a distinct row by themselves. Below the epithelium are blood vessels (b.v.) whose walls stain rather deeply with eosin. Underneath the blood vessels are a number of big cavities, which probably represent the adipose tissue. The adipose tissue is said to be present below the skin of most of the fishes, but its presence in this position between the epithelial and dermal regions, is interesting. Below the fat- cells is the connective tissue, which is marked by a large number of nuclei which are not surrounded by a white zone. ‘The cell- boundaries in this region are not distinguishable. 50 Records of the Indian Museum. [VOL. XXIV, The structure of the mental disc in Gavra may be treated under two headings, (i) the structure of the central callous portion of the disc and (ii) the structure of its tuberculated borders and of the fringed tuberculated anterior labial fold (fig. 12). The structure of the former is very similar to that of the integument TEXT-FIG. 12.—Structure of tubercles on the disc and the anterior labial fold of Gavra annandalei. a. Tubercle on anterior labial fold. 6. Tubercle on lateral free borders of the disc. described above. ‘The following are, however, some salient points of difference :— (i) The epithelial cells are better defined and in certain places the ampulliform gland-cells are more numerous. (ii) The ‘‘ clavate cells” are fewer in number and are situat- TrxtT-FIG. 13.-—Transverse section through a portion of the tuberculated anterior labial fold of Gavva annandalet, x 435. = spine; ep.d.= epidermis; @ = dermis. ed at great intervals. The protoplasm of these cells has receded inwards from near their upper cell-limits. (iii) The dermis is chiefly composed of an adipose tissue, which is bounded both above and below by a thin layer of fibrous con- nective tissue. 1922. ] S. L. Hora: Fish of Mountain Torrents. 51 The structure of the tuberculated region, on the other hand, is totally different. Ina vertical section (fig. 13) of the fringed portion of the anterior labial fold, the superficial epidermal layer covering a tubercle is modified into spines (s). It is perhaps significant that I have not been able to find any nuclei in the spines; the spines are, otherwise, shorter, thicker and stouter than those described in Bhavania annandalet. On examining a large number of sections of this region and of the free tuberculated borders of the disc, I have observed that the spine is formed as a prolongation of the outer cell-wall of the superficial epidermal layer. Below the spinous layer, there are several tiers of polygonal cells (ef.d.) which are vacuolated. In the basal region the cell- limits are not well marked and the nuclei stand out prominently with haematoxylin stain. In some cells the nuclei are surrounded by a whitish halo. The dermis (d) consists of a compact connec- tive tissue, with a large number of nuclei scattered just below the epidermis. The cell-boundaries in this region are not well-defined, In the region of the postetior labial fold and also in that of the free border of the disc, the tubercles are provided with a dermal plug. The dermis consists of a large number of branched irregular cells, forming a primitive type of connective tissue. The structure of the tuberculated region in Garra differs from that of the integument and of the central callous portion of the disc in the following points :-— (i) The gland-cells are absent. (ii) The ‘‘ clavate cells’? are absent. (iii) The superficial epithelial layer of cells is modified into spines which do not possess any nuclei. (iv) The adipose tissue is totally absent, aad the dermis, therefore, presents a compact, solid structure. In the order Siluroidea the modification of the skin to form an adhesive apparatus is very different. It is thrown into folds and ridges, which are characterized by a special structure. Any por- tion of the skin may thus be modified to serve the purpose of adhesion. To illustrate the structure of the adhesive apparatus in Silurid fishes, I will first describe that found in Glyptosternum labtatum. The structure of the integument found in this genus may in the first place be considered. In a vertical section (fig. 14) of the non-striated skin covering the dorso-lateral surface of the pectoral spine, the epithelium consists of several layers of small, more or less flattened and rectangular cells which possess well-marked cell-walls and relatively large oval nuclei. The cells near the sur- face are smaller than those immediately below them. Their out- line varies considerably according to the extent to which they may be packed together in a particular place. Black pigment (p) is present at certain places in the tissue. Most of the central space of the epithelium is occupied by a number of big “‘ clavate cells ’’ (c), which are distributed at regular intervals and form a distinct layer of their own. ‘They are fairly well developed and 52 . Records of the Indian Museum. (VOT. SOshy. sometimes their nuclei may be as big as an ordinary epithelial cell. The ‘‘clavate cell’’ possesses a distinct cell-wall and in most cases it contains more than one nucleus. The contents of the “ clavate cell’’ are very different from those of the surround- ing epithelial cells, as it is only lightly stained with haematoxy- lin and eosin. The nucleus is surrounded by a whitish area. The most interesting point is the degree of vacuolation that is generally met with in these cells. ‘The process of vacuolation sets in from the outer wall and the protoplasm gradually recedes towards the inner side. In extreme cases more than half of the cell is emptied of protoplasm. I have not been able to make out the structure of the contents of a vacuole. The nucleus is generally vesicular with a distinct membrane and in those cells, which contain only Tr Ot? ——_-z — ZI RI RE LEE aN ee SF de Bon oe Ree 2 MoS“ ‘P. fs ae 2% eee Caer me CG a Text-Fic. 14.—Transverse section through the integument of Glyptosternium labiatum xX 435. e.c. = small epithelial cells; p.= pigtaent ; c= clavate cell; g.c. = gland- cell; ¢.¢. = connective tissue; b.v, = blood vessel. one nucleus, its chromatin matter is diffused as if preparing for a mitotic division. I have not been able to observe distinct nuclear figures in any of these cells. If more than one nucleus is present in a cell, the nucleolus can also be readily made out. Ramsay Wright! described these cells in the integument of Amiuris catus. He observed, “‘there can hardly be any doubt that the clavate cells have an important physiological role to play. What that is . remains still obscure.’’ I have found these cells in the integument covering the tip of the snout in Garvra and in the position described above in Glyplosternum labiatum ; but am unable to understand their exact significance. It may, however, be pointed out that they are always absent in an adhesive tissue in fishes. The gland-cells in Glyptosternum are not scattered as has al- 1 Ramsay Wright, Proc. Canadian Inst. (n. s.) Il, pp. 254-255 (1884). 1922. ] S. L. Hora: Fish of Mountain Torrents. 53 ready been described in the integument of Garvva; but are aggre- gated to form definite structures. Each one of these structures is flask-shaped with the neck almost half as long as its total length. The mouth of the flask along with the adjoining tissue projects slightly above the surface of the integument. The body of the flask is occupied by a number of characteristic cells. The cells are elongated and are drawn out into long, fine processes which travel through the neck of the flask and open on the surface of the skin. The nuclei are very big and occupy almost the whole of the cell. The cells forming this characteristic structure do not begin at the same level and thus present an irregular bunch of cells hanging in the cavity of the flask by means of fine threads. Sometimes one or more epithelial cells make their way inside the flask and when seen they are usually found in the neck region. . It is after long hesitation that I have assigned to these cells the function of secretion. The following are the main reasons for holding this view :— (i) The unicellular glands, usually present in the integument of fishes, are absent. (ii) The mouth of the flask projects beyond the surface of the integument. (iii) The cells have fine canals which open on the surface of the skin. (iv) The cells are provi- ded with big nuclei. i Such glandular structures are found at a considerable distance from each other, and I have not been able to find more than three in any one section. In the structure of the adhesive apparatus (fig. 15) formed by the striation of the skin on the under sutr- face of the pectoral and pelvic spines, a distinct ad- vance is made upon that observed in Bhavama and TEXtT-F1G. 15.—Transverse section through Garra. The upper layer of the striated skin on the under surface of the epithelial cells is modified outer ray of the pectoral fin in Glyptosternum into curved spines (sp.), /atatum, x 435. whose inner limits are not S. p-= spine; n. = nuclear bodies ; 2.c.7. = Space te Ale Poppies eet tale 5 Gacaaley 2) os cannes. are provided with definite _ tive tissue. nuclei, which are situated in their lower swollen portions. In the non-cellular region imme- diately below the spinous layer are a number of deeply staining bodies forming a definite row. What these bodies are, I have 54 Records of the Indian Museum. [VoL. XXIV, not been able to determine definitely. Below the spines is a deep non-cellular layer (v.c. 7.) in which are scattered a number of spaces (c) having a definite shape. ‘They are almost crescentic in outline and possess a short spine-like process along their convex borders. Thelower limit of the non-cellular region is formed by a regular wavy line. In the curves of this wavy line are situated a number of characteristic columnar cells. Each is provided with a nucleus, a distinct. nucleolus and a small well-marked vacuole (v). Between two neighbouring cells there is generally a small chink-like cavity. Below these columnar cells, are a number of small epithelial cells which are irregularly arranged in four to five tiers. Usually they possess small, solid, deeply staining nuclei, but in certain cells the nuclei are altogether absent. Beneath the epithelial tissue is a loose connective tissue (c.t.) forming the dermis. In the genus G Iyioaierniing the adhesive apparatus is sup- ported by a definite tissue as has already been remarked. The supporting tissue (fig. 16) consists of a mass of polyhedral cells (fig. 17) which are very turgid and are closely packed together. The nuclei of ' these cells are fairly large and in certain cases there may be more than one nucleus in a cell. The cells vary in form and size to a considerable extent and towards the base they are so much _ pressed together that the cell- limits become almost obliterated and the nuclei become spindle- shaped. ‘The support- ing tissue (S.é.c.) is sur- TExt-FIG. 16.—Transverse section through the rounded by a loose con- thickened skin covering the pectoral spine in nective tissue (ac): Glyptosternum labiatum, xX 42. The structure of the s.£. = supporting tissue ; c.y.= clavate cell re- t}oracic adhesive appa- gion ; p.= pigment ; s.7. = striated region ; 6.v. = . Bont eae Pie its = i ratus of Psewdecheneis blood vessel ; c. = clavate cells. é hi sulcatus is different from all the three types described above. ‘The nuclei of the spines are situated in the non-cellular region and the cell-walls in the epithelial region are not well defined. In the cells of the first epithelial layer, immediately below the non-celluiar region, there are ill defined vacuoles. ‘The connective tissue is not so loose and is richly dotted with a large number of nuclei of various forms. 1922.] Within the genus Glyptotho- vax, the structure of the adhe- sive apparatus shows consider- able variation. In G. dorsalis, a large number of specimens of which were collected in the sluggish and muddy streams of the Manipur Valley, the spines are small and the spin- ous layer as a whole is not well developed. The epithelium is composed of several tiers of small, squarish cells; of these the uppermost and the basal layers are highly vacuolated. The cell-walls are quite distinct and the basal layer is somewhat columnar. S. L. Hora: Fish of Mountain Torrents. 55 TEXT-FIG. section 17.— Transverse through the supporting tissue of the adhesive apparatus on the fins of Glypto- sternum labiatum, X 435. c.t.c. = connective tissue cells; s.tc. = supporting tissue cells Below the epithe- lium is a dense sheet of connective tissue, in which are scattered 18.—Transverse tion through a portion of the thoracic adhesive apparatus of Glyptothorax madvasapatants, X 435. TEXT-FIG. sec- S.=spine; 2.s.= nucleus of the spine ; 2.c./. = noncellular layer ; c. = first row of cavities ; c’. = second row of cavities; c.e.== columnar epithe- lium ; b.c.= basal cells of epithelium ; c.t. = connective tissue. kind of structure. big cavities full of blood cor- puscles. Underneath the con- nective tissue are fat-cells with eccentric nuclei. In G. madras- patanus the structure (fig. 18) of the adhesive tissue is more advanced in so far as the cell- limits are not distinguishable. Immediately below the non-cellu- lar region there are two layers of cavities (c’.c’.); the upper is in the form of elongated spaces with intervening columns of pro- toplasm, while the second layer consists of rounded cavities in the substance of the protoplasm. Below this the epithelium con- sists of a row of columnar cells which are followed by a number of small, rounded cells. In the basal epithelial layer I have not been-.able to find any nuclear structures. The greatest specialization in the structure of the adhesive tissue within the genus is reached in those forms that possess an adliesive surface on the chest as well as on the under surface of the pectoral and pelvic spines. I take G. sp. from Madras as an example of this 56 _ Records of the Indian Museum. [VoL. XXIV, In a vertical section of the striated skin covering the under sutface of the pectoral spine of G.sazszi (fig. 19, a) the structure in all essential points corresponds to that of the previous examples. The spines (s) are hooked and more regular; their nuclei (7.s.) are situated near the base and each of them is surrounded by a whitish halo. The nuclei project into the non-cellular region (v.c.7.). Underneath the non-cellular region there are two rows of cavities or open spaces. The cavities of the first row (c’) are greatly elongated; they are broader near the upper than near the lower end. On focussing, a part of the cavity is found.to contain a lightly stained protoplasmic substance. The second row of spaces (c’) is similar to the first, but here the lower portion of the cavity is filled with protoplasm ; the upper margin of each cavity is deeply stained and in the section there appears an interrupted band. Then follow large columnar cells (c.e.) which are provided with big nuclei. The nuclei are oval and possess a well-defined nu- cleolus; they are surrounded by a whitish halo. Below these there are several rows of small, rounded epithelial cells, some of : which are devoid of any nuclear substance. The basal layer (0.c.) is represented by finger-like processes of protoplasm which do not possess any nuclei. It may be pointed out for the sake of clear understanding that the whole of the structure is one continuous mass and that the cell-boundaries are nowhere marked, but in drawing up the description it has been convenient to treat the structure as if it were composed of a number of distinct layers. It is intéresting to note the changes in the structure as we pass from the ridge to the grooved portion of the striated skin. Attention may be drawn to the following points of difference in the grooved area (fig. 19b):-— (i) The spines become smaller and smaller till they are repre- sented by small knob-like projections on the surface. In the middle of the groove the surface becomes entirely smooth. (ii) ‘he nuclei of the spines recede inwards and ultimately form a continuous layer just below the surface of the skin. (iii) The first layer of cavities is represented by small, oval or rounded spaces in the grooved region. In some cases this layer may be totally absent. (iv) The second row of spaces is represented by small cavi- ties; they are provided with a deeply staining upper margin. The structure, on the whole, appears as a mass of proto- plasm in which the nuclei are scattered either near the base or near the apex and a few cavities are present in the middle, The basal epithelium is represented by finger-like processes as described ~ above. In a horizontal section (fig. 19c) the structure of the adhe- sive tissue does not differ greatly from that seen in a vertical sec- tion. The chief difference liesin the form and extent of the various elements noted above. ‘The non-cellular region is separated from the underlying tissue by a regular wavy line of demarcation. The 1922.} S. lL. Hora: Fish of Mountain Torrents. 57 TEXT-FIG. 1y.— Minute structure of the adhesive apparatus on the under surface of the pectoral spine of Glyptothorax sp. a, Transverse section through a portion of the ridge, x 650. b. Transverse section through a portion of the groove, x 650. _ c. Horizontal section through a portion of the ridge, x 650. s==spine; .s.= nucleus of the spine; .c.y.=non-cellular region; ¢.= first row of cavities; c’ second row of cavities; c.e.=columnar epithelium c.e. = small, rounded epithelial cells; b.e. = basal epithelial cells. 58 Records of the Indian Museum. [VoL. XXIV, cavities are considerably smaller and do not show any proto- plasmic elements within their limits. The so-called columnar epithelium is not well-marked. The underlying structure corre- sponds to that described for the vertical section. Having described the structure met with in the different forms selected above, it will be advantageous to consider the lines along which evolution has taken place from the simplest to the most complicated structure. In the simplest type of adhesive appa- ratus the skin is thickened and its outer epithelial layer of cells is modified into curved spines; the gland cells and the ‘‘ clavate cells’? of normal skin-tissue totally disappear. Specialization proceeds along two lines, firstly there is an increasing tendency at every step towards vacuolation of the superficial epithelial cells and this ultimately results in the formation of definite cavities, and secondly the cell-walls of epidermal. cells become indistin- guishable. The final stage is reached in Glyptothovax sp. where the whole of the epidermal tissue appears as a syncytium. ‘The adhesive apparatus of the Silurid fishes is distinguished from that of the Cyprinid forms by the fact that in the former the skin is thickened and striated, whereas in the latter it is only thickened and forms a plain cushion-like pad. I have not been able to understand the true significance of increased vacuolation or of the syncytium formation in the tissue of the adhesive apparatus. There is, however, no doubt as to the function of the epidermal spines. All of them are curved in the same direction and probably they point posteriorly. As the fish sticks to a stone with the head pointed up stream, and the current of water tends to move it backwards, the spines when closely applied to a stone, fix it securely by taking hold of the unevennesses of the rock. In Silurid fishes the ridges are pulled outwards and a sort of vacuum is produced in the grooves which helps the fishes in adhering to ‘rocks. So far as I know, the type of structure described above is not met with ia tissues of adhesion in any other group of animals. Dahlgren and Kepner,' who have given a summary of the subject, have not anywhere referred to a spinous structure. Even in a longitudinal, vertical section through a small region of the grasp- ing organ on the head of Remora, the characteristic structure that I have described above apparently is not found. CONCLUSION. In conclusion I wish to refer briefly to the origin of the hill- stream fauna and to the means of dispersal and propagation adopted by it. In the following discussion I take up these points one by one. Origin of the Hill-stream Fauna.—There are two possibilities, firstly , that the forms now living in the mountain-rapids were once ! Dahlgren and Kepner, Principals of Animal Histology, pp. 409-417 (1908). 1922. ] S. L. Hora: Fish of Mountain Torrents. 59 accidentally carried into them, and secondly; that there has been a step by step colonisation of the hill-torrents from the sluggish- streams that flow in level country. As regards the first possibility it may safely be inferred that forms which have not previously acquired adaptive characters, cannot live in rapid waters because at every move they are liable to be swept away down stream. ‘The second view is more probable and the following are some of the main points in its favour :— (1) As a hill-stream changes into a sluggish stream in almost level country its fauna changes accordingly and intermediate forms, like Crossochilus latia, between the typical hill-stream fishes and the fishes of the slow streams are always met within the intermediate regions where the water is neither flowing very fast nor very slowly. , (2) The very fact that the members of certain genera such as Glyptothorax and Garra can be arranged in a series according to the degree of modification they exhibit in response to the strength of the current of stream, shows that there has been a gradual colonisation of rapid streams. (3) The hill-stream fishes in the course of their development pass through many different stages which clearly show, at any rate in the case of Garra,' that the evolution of such forms is from ~ those that live in sluggish streams. Not only is this shown by the form of the body but also in the modifications of such organs as the air-bladder, the position and form of the mouth and the eyes, and in the reduction of the branchiostegal membrane and rays. In short, the developmental series of Garva as given in a previous paper recapitulates the history of the evolution of the genus. Having subscribed to the view that the hill-stream fauna has originated by the process of gradual colonisation from the slow streams, it will not be out of place to discuss the causes that might have led to the migration of these forms. There are two chief factors which might compel such a migration:—food and safety. In the hill-streams there is always plenty of food in the form of algal slime on the exposed surface of rocks; but only those animals can make use of it which have their jaws specially adapted for rasping it off the stones on which it grows. As regards safety, it may be said that there is very little competition in the hill-streams among the fishes themselves. Moreover, they ate practically safe from the ravages wrought by birds and large predaceous fishes, crocodiles, etc. Certain fishes like Nemachilus can find shelter underneath stones and the readiness with which they hide themselves is marvellous. Probably the fishes inhabit- ing slow-running waters originally ascended the hill-streams step by step in search of food and gradually acquired certain char- acters which made them specially suitable for living in the newly chosen environment. ! Hora, Rec. Ind. Mus. XXII, p. 639 (1921). 60 | Records of the Indian Museum. [Vo,. XXIV, Annandale,! when discussing the evolution of the adhesive apparatus in hill-stream fishes, made the following remark about the genus Garrva :—‘‘ Whereas the chief factor in the case of Psilorhynchus was rapid-running water in a rocky stream-bed, in Discognathus the primary factor was a peculiar mode of feeding.’’ Quite recently I? also subscribed to this view, but a more detailed study of the adhesive apparatus has led me to modify my previous ideas. I believe that the mental disc of Garra has not primarily been evolved for the ‘‘ peculiar mode of feeding,’’ which is practi- cally similar in all the genera of hill-stream fishes, but for securing adhesion to rocks in rapid running waters. The adhesive appar- atus on the under surface of the paired fins is an additional organ of adhesion in species that live in very rapid waters. In certain highly evolved species of Garra, which have secondarily taken to live in lakes and pools, the peculiar pad-like structure on the under surface of the paired fins have disppeared though the charac- teristic mental disc is still present, as it is probably of use to the fish in its “‘ peculiar mode of feeding ’’ which was acquired as a direct response to a life in hill-streams. I am, therefore, led, to conclude that none of the hill-stream forms are ancestral forms; but that all of them are descended from migrants from the slow-running streams. The modifications that some of these forms exhibit are due to the physical condi- tions prevailing in mountain-rapids; and it is to this cause that we must ascribe the similarity in form and structure exhibited by the more advanced members of the genera dealt with in this paper. Means of dispersal.—When dealing with the fish of Manipur it was pointed out by me that most of the new species from the hill-streams had a localised distribution. This is the case with almost all the hill-stream fishes and naturally it is difficult to imagine a wide range of distribution of these forms. ‘The most highly modified forms are not capable of living for a long time in muddy channels, on account of the form of their bodies and the structure of their jaws. In cases where a very wide range has been attributed to a hill-stream species, it has always been found on comparison of material from different localities that several allied forms had been grouped together under the same name and that most of them are capable of specific separa- tion. For example Glyptosternum labiatum which was described from the Mishmi Hills in Upper Assam, was recorded by Vinci- guerra® from the Kachin Hills, Upper Burma. Regan* in 1905 ‘separated the Burmese specimens from those collected in the Mishmi Hills under the new name G. vinciguerrae. A remarkably wide range of distribution was attributed to Garra lamta, but as 1 Annandale, Rec. Ind. Mus. XIV, p. 117 (1919). 2 Hora, Rec. Ind. Mus. X1X, p. 213 (1920). 8 Vinciguerra, Ann. Mus. Civ. Stor. Nat. Genova, (2), 1X, p. 252 (1889). + Regan, Ann. Mag. Nat. Hist. (7) XV, p. 184 (1905). 1922.] S. L. Hora: Fish of Mountain Torrents. 61 has been shown elsewherc,' the forms from various localities are not specifically identical. Methods of propagation.—All the hill-stream fishes with which I am acquainted are oviparous. The rapid cutrent in these streams makes it impossible for them to lay their eggs loose in the rocky beds, as they are liable to be carried down stream and destroyed. In my tours to the hill-streams I have always found that. the pools in the courses of these streams were full of young specimens of the genera Nemachilus and Bartiius and of a. few other small fish that are usually found in pools. In no case was I able to find young of Garra or of any other highly modified hill-stream forms. Dr. Annandale and Major Sewell, however, found Garra and Psilorhynchus in pools of hill-torrents in the Western Ghats. There appear to be two possibilities, (i) that the hill-stream fishes migrate to slow running waters to lay their eggs and that every generation has later on to ascend up-stream to its natural habitat or, (ii) that the eggs are tightly fixed to stones. I am unable to say at present which hypothesis can be accepted as correct. | Hora, Rec. Ind. Mus. XXII, pp. 933-687 (1921). aN eee se ae oes NOTES ON FISHES IN THE INDIAN MUSEUM. III. ON FISHES BELONGING TO THE FAMILY COBITIDAE FROM HIGH ALTITUDES IN CENTRAL ASIA. By SuNDER LAL Hora, M.Sc., Assistant Superintendent, Zoological Survey of India. The Cobitid fishes! from the high altitudes of Central Asia are generally characterized by the absence of a suborbital spine, by the elongate form of their body, especially of the caudal peduncle, and by the total absence of any scales. ‘I'he belly is generally rounded and not depressed. The Indian Museum possesses a large number of specimens of this family from Tibet, Northern Kashmir, Western Turkestan and Seistan. When dealing with the fish of Seistan it was pointed out by Annandale and myself? that among those specimens which have been referred by several ich- thyologists to Nemachilus stoliczkae Steind. there were several forms capable of specific separation. In this note an attempt is made to elucidate these points and to discuss the specific validity of the various species represented in our collection. Here are also incorporated the results of an examination of the loaches recently collected in Kashmir by officers of the Zoolog- ical Survey of India. At the end I have added a short note on the sexual dimorphism exhibited by some of these species. The Central Asiatic forms belonging to the family Cobitidae dealt with in this note may be grouped into three distinct genera, which can be distinguished in the following manner:— A. Two bladders ; one lying free in abdominal cavity and second divided into two lateral chambers enclosed in bone ... Diplophysa. B. One bladder, consisting of two lateral chambers enclosed in bone. I. Soft dorsal fin between Gpny dorsal and caudal fins present ip ... Adtposia. Il. Soft dorsal fin absent . og ... Nemachilus. Genus Diplophysa Kessler. 1874. Diplophysa, Kessler, Bull. Soc. Sci. Moscow X\, pp. 1-63. 1888. Lefwa, Herzenstcin, Wiss. Res. Presewalski Central Asia. Rets., Zool. \II (2), p. gt, Unfortunately the paper in which Kessler proposed the generic ! A large number of specimens of the Cobitid genus Botia. which possesses. spines below the eye, have recently been collected in Kashmir. The genus extends to China as well, but I have not dealt with it here, © Annandale and Hora, Rec. Ind. Mus. XVIII, p. 179 (1920). 04 Records of the Indian Museum. j VOL. XXIV, name Diplophysa, is not available in Calcutta, so I take from Day ' the characters on which this genus was erected. The genus Diplo- piysa comprises those fishes in which the body is greatly elongated ' and strongly compressed posteriorly ; the eyes are surrounded with a fold of skin forming a lid; the lips are fleshy, the upper more or less denticulated, the inferior bilobed and more or less papillated and the air-vessel is divided into two parts, the anterior enclosed in a bony capsule and the posterior elongated and free in the abdominal cavity. I agree with Day (op. czt.) that the first three characters do not possess any generic value, but the last feature, that of the air-vessel, is quite sufficient to distinguish the genus Diplophysa from Nemachilus, to which it is closely allied. Day did not dispute the validity of the last character but suggested a re-examination of the Western Turkestan specimens and remarked that, ‘it would be very remarkable were the Nemacheili found in Europe, in fact throughout Asia, even in the Oxus,; to have their air-vessel enclosed in bone, whereas in the river Ili going to Lake Balkash, and the river Urdjar falling into Lake Ala (Ala-Kul), to have the same organ partially free in the abdomen, as is seen in the genus Botia.’’ Day did not think himself justified in recognis- ing Diplophysa as a distinct genus from Nemachilus even on the character of the air-bladder, which is so remarkable. Kessler * in 1879, when dealing with the Central Asiatic fishes, upheld his genus Diflophysa and described two new species under this generic designation. In reviewing Day’s criticism of the genus he pointed out that in all probability Dr. Stoliczka’s collec- tion was made in the area south west of the Tarim river-system, while Przewalski’s collection, which contained several representa- tives of the genus, was madejmuch further to the east. Moreover he considered the air-bladder to be as important for taxonomic purposes as the pharyngeal teeth, on which the two families Cobitidae and Cyprinidae are distinguished. Herzenstein * in his valuable monograph of Central Asiatic fishes agreed with Day and considered Diplophysa synonymous with Nemachilus. But at the same time he instituted a new genus Lefua to accommodate Diplophysa costata Kessler and Octonema plesket Herz. He characterized the genus Lefwa as follows .— “Caput valde depressum. Os fere terminale. Spina suborbitalis nulla. Nares anteriores cirro sat longo instructae. Cirri rostrales 4, supra-maxillares2. Vesicae natatoriae pars posterior in cavitate abdominali libere suspensa.’’ No notice seems to have been taken of this genus till 1907 when Berg * recognised it and considered the Japanese genus Eljxis Jordan and Fowler® as a synonym of ! Day, Proc. Zool. Soc. London, p. 793 (1876); Sci. Res. 2nd. Lavkand Mission, [chthyol., p. 12 (1878). 2 Kessler, Bull. Acad. Sci. St. Pétersbourg XXV, p. 302 (1879). 8 Herzenstein, Wiss. Res. Przewalski Central As. Reis., Zool. \\1 (2), p. 1 1888). 4 Berg, Proc. U.S. Nat. Mus. XXXII, p. 437 (1907). ® Jordan and Fowler, Proc U.S. Nat. Mus. XXV1, p. 768 (1903). 1922.] S. L. Hora: Central Asiatic Cobitidae. 65 Lefua, the definition of which he modified as follows :—“‘ Cirri 8, four rostral, two maxillary and two at the anterior nostrils. Scales present. No erectile spine below the eye. Dorsal fin about over the ventral, with few rays; caudal rounded. Air-bladder with a posterior part free in the abdominal cavity.’’ The genus Elixis was established to comprise those species of Nemachilus which possessed a pair of nasal barbels in addition to six others that surround the mouth. In £. ntkkonts, the genotype of the genus, and in F. coreanus subsequently described by Jordan and Starks ' no mention is made as to the nature of the air-bladder in them. I have examined some Indian species of the genus Nemta- chilus, such as N. evezardt Day* which possess a pair of well- developed nasal barbels but have not found in them a free bladder in the abdominal cavity. It is quite probable that the Japanese species with eight barbels may not possess a free bladder as is said to be present in the Chinese species with eight barbels assigned to the genus Le/ua by Herzenstein. I am led, therefore, to believe that Berg united the two genera merely on the consi- deration of the nasal barbels and paid little attention to the character of the air-bladder. He, moreover, considered Nemachi- lus dixont Fowler,’ Elixis coreanus Jordan and Starks and the two forms included by Herzenstein under his genus Lejfua as representing only one species, having examined a large number of specimens from wideiy different localities in China and Korea. I doubt the validity of this statement and suggest a re-examination of these specimens. There seems to me nothing at preseut in the definition of Lefua and Elixts, except the presence of nasal barbels, which could justify their separation from Dipflophysa and Nemachi- lus respectively, but Annandale and I (op. cit., p. 185) have already pointed out that we do not consider it a character of generic value. I conclude, therefore, that Lefua is a synonym of Diplophysa and Elixts of Nemachilus. Quite recently Weber and Beaufort* have recognised the genus Flixis and have referred Nemachilus obesus Vaill,’ to it only on the character of the nasal barbels. In a recent contribution to the ichthyology of Central Asia, Zugmayer ® has recognised the genus Diplophysa as distinct from Nemachilus, though closely allied to it. The chief distinction between the two genera lies in the fact that according to Zug- mayer a part of the air-vesse] lies free in tlhe abdominal cavity in Diplophysa, whereas in Nemachilus it is wholly enclosed in a bony capsule. Having examined the air-bladder in Diplophysa papiilloso-labiata Kessler, Zugmayer states that the two parts are distinct from each other. ! Jordan and Starks, Proc. U.S. Nat. Mus. XXVIII, p. 201, fig. 7 (1905). 2 Day, Fish. /ndia II, p. 613, pl. cliii, fg. 11 41878). ® Fowler, Proc. Acad. Nat. Sci. Philadelphia, p. 181 (1899, 1900). + Weber and Beaufort, Fishes Indo-Austral. Archipel. U1, p. 35, fig. 16 (1016). : 5 Vaillant, Notes Leyden Mus. XXIV, p. 134 (1902). 5’ Zugmayer, Zool. Fahvb. Syst. XXX, p. 294 (1910). 66 , Records of the Indian Museum. |Vot.. XXIV, I have myself examined specimens of the same species in our collection and agree with Zugmayer’s statement. Both Kessler and Zugmayer believe that there is only one bladder in Dip- lophysa and that the anterior part is enclosed in bone while the posterior lies free in the abdominal cavity. Zugmayer found the two bladders to be quite distinct from each other but regarded them as parts of the same bladder. On examining the bladder in young specimens of a new species from Eastern Tibet (Rham-tso) I find that the two bladders are totally distinct from each other and that they are not the two parts of a single structure. The posterior bladder, that lies free in the abdominal cavity, is con- nected with the oesophagus by a short pneumatic duct given off from its anterior end. ‘This duct is only distinct in young speci- mens and atrophies in the adult. In order to understand the true significance of the posterior bladder and its relation to the anterior, it is necessary to examine the various types of bladder commonly met with among the different genera of Cyprinoi- dea. The swim-bladder of a typical Cyprinid fish such as Labeo vohita is large and lies free in the abdominal cavity. It is con- stricted in the middle to form an anterior and a posterior chamber (fig. 1a). The pneumatic duct from the oesophagus opens into the constricted region.’ In those genera that live in rapid running waters the bladder undergoes considerable degeneration; this consists firstly in the gradual reduction of the two chambers and the ultimate disappearance of the posterior, and secondly, in the thickening of their walls. In extreme cases the bladder becomes completely enclosed in a bony. capsule derived from the transverse processes of the adjacent vertebrae. In the genus Psilorhynchus the posterior chamber is greatly reduced and the anterior is covered by a thick fibrous coat (figs. 10, Ic). In Nemachilus vittatus from the Kashmir Valley the anterior chamber is laterally flattened and covered by a bony capsule while the posterior chamber is small and thick walled (fig. if). ‘The pneumatic duct still opens into the constricted region between the two chambers. In other species of this genus the anterior part is divided into two lateral chambers which are enclosed in a bony capsule and all remains of the posterior chamber are wanting. In Adiposia rhadinaea there is still a short bulb-like structure re- presenting the posterior chamber (fig. 1g) otherwise it is very similar to that found in most species of Nemachilus. In extreme cases such as Balitora brucei the two lateral halves of the anterior chamber are much reduced and are somewhat separated from each other (fig. rh). Among the members of the genus Diflophysa the anterior bladder (fig. 17), which is enclosed in a bony capsule, is in all probability similar to that found in the genus Nemachilus and, thus, it may represent the primary or the true original bladder of the fish. The posterior bladder, that lies free in the abdominal cavity, is a secondary structure and in its origin and position is 1922 ] S. L. Hora: Central Asiatic Cobitidae. 67 Textreric. 1.—Types of air-bladder found in Cyprinoid fishes. a, Labeo rohita. b, c, Psilorhynchus balttora. d, Botia hymenophysa. e, Botia almorhae. F, Nemachilus vittatus. g£, Adiposia rhadinaea. h, Balitora brucei. J, Diplophysa stewartt. ae The dotted line in the figure indicates that the portion thus outlined is enclosed in bone, 68 Records of the Indian Museum. [VoL. XXIV, quite different from the normal Cyprinoid bladder. The following are the chief points of difference :— (i) There is a short pneumatic duct from the anterior end of the bladder to the oesophagus, while in the normal Cyprinoid type the pneumatic duct is long and opens into the middle of the bladder in the constricted region. (ii) The bladder may or may not be constricted in the middle. (iii) The pneumatic duct is present only in young specimens, while it is lost in the adult. In the genus Botta, the structure of the air-bladder differs considerably. Though in many respects of a typical Cyprinoid form, the anterior chamber is partially (fig. 1d) or wholly (fig. 1é) enclosed in a bony capsule formed by the transverse pro- cesses of the neighbouring vertebrae. On comparing drawings d and 7 in figure 1, it will be seen that Day was in error in suggest- ing that the bladder of Dipiophysa would have to be similar in structure to that in Bodza. The reduction of the swim-bladder in fishes that live in rapid- running waters is in all probability due to the fact that they live on the bottom and do not require to make vertical movements. The enclosure of the bladder in a bony capsule presumably has some special biological significance, but of this nothing is yet known. Zugmayer believed that the free bladder in Dzplophysa is to be explained by the assumption that the members of the genus have not yet acquired a true ground habit and that conse- quently the posterior half has not yet been affected. This, how- ever, does not appear to be a-correct interpretation of the fact as there are two distinct air-bladders in Diplophysa, the one enclosed in a bony capsule being possibly the original Cyprinoid bladder, while the other that lies free in the abdominal cavity is either a secondary acquisition or, as Dr. Annandale suggests to me, repre- sents the modified posterior chamber of the normal Cyprinoid bladder. In the latter case the anterior chamber has become en- closed in bone, as in Nemachilus vittatus, and the posterior cham- ber was nipped off, retaining its connection with the oesophagus through the primary pneumatic duct. ‘The members of the genus Diplophysa have in all probability come to live secondarily in the deep muddy waters of the lake-basins of Central Asia in which situation they require a hydrostatic organ for vertical movements ; I believe that they have originated from forms like Nemachilus in which the air-bladder is reduced and enclosed in a bony capsule. When the primary air-bladder became enclosed in bone it probably could not again be modified for the performance of a hydrostatic function to suit the new environment. T thus believe that Dzf/o- physa is a more specialized genus than Nemachilus, whereas Zugma- yer regards it as more primitive. Of the species of Diplophysa at present known, all described by Kessler, the following have been recorded either from lakes or from deep muddy waters at great altitudes in Central Asia :— 1922.] S. L. Hora: Central Asiatic Cobitidae. 69 Se Se strauchtt, [li river owing into Lake Balkash. labiatus, Urdjar river flowing into Ala-kul. 5 intermedius, Take Dalai-nor. Pr nasalts, ; ” "y costata, - -s dalaicus, : i kungessana, Kinges River. fA papilloso-labiata, Juldus. ss microphthalmus, Chami. Of the nine species enumerated above the first two are from Eastern Turkestan, the next four from Lake Dalai-nor, which is situated in the lake basin of Mongolia, and the remaining three from the Tarim river-system which ultimately drains into lake Lob-nor. It is significant that all the Cobitid fishes known from Lake Dalai-nor belong to the genus Diplophysa. A new species of this genus described here was obtained by Capt. Kennedy and Capt. Stewart in a small stream flowing into Rham-tso, a lake of considerable dimensions at an altitude of 14,700 ft. in Eastern Tibet. It will thus be seen that the genus is known from Eastern Turkestan and Mongolia on the one hand and from Rham-tso in Eastern Tibet on the other. This apparently discontinuous dis- tribution may be accounted for by the fact that very little is at present known of the ichthyology of the intermediate region. There is yet another possibility which may explain the distri- bution of this genus. It is possible that the genus is polyphyletic in origin, betause the character of the bladder on which it is solely based, may have originated on more than one occasion in response to life in deep waters which necessitated some hydrostatic mechanism. The genus Diplophysa is represented by two species in the collection of the Indian Museum, one of which is new to science. Diplophysa papilloso-labiata Kessler. 1878. Diplophysa eg UE notre Kessler, Bull. Acad. St. Péters- bourg XXV, p. 290.. 1878. Diplophysa papilloso- dateatae IXessler, Ml. biol. X, p. 257. 1888. Nemachilus strauchtt var. papilloso- -labiatus, Herzenstein, Wiss, Res. Przewalski Central As. Reis., Zool. U1 (2), p. 50, pl. vi, fig 1910. Dplaphien: (Nemach ilus) strauchti papilloso-labiata, Zug- mayer, Zool. Zahrb. Syst. XXX, p. 297. This is the only species of the genus Diplophysa collected by Dr. Stoliczka. There are seven specimens in our collection and they are labelled as having come from Yarkand, probably from the Yarkand river which forms a part of the Tarim river-system. Of the seven specimens four are males and the rest females. The species exhibits a well-marked sexual dimorphism. The free air-bladder is not constricted in the middle and lies almost in the middle of the abdominal cavity. In my dissections I have not been able to find any pneutnatic duct. 70 Records of the Indian Museum. [VoL. XXIV, The eggs of this species are very small and almost fill the whole of the abdominal cavity. . Diplophysa papilloso-labiata is known only foal fhe Tarim tiver-system (Eastern Turkestan). The longest specimen in our collection is 105 mm. in length without including the length ot the caudal fin. Diplophysa stewarti, sp. nov. (Text-figs. 2c, 2d.) 1908. Nemachilus stoliczkae, \V.loyd (in part), Rec. Jad. Afus. II, p. IQII. cena stoliczkae, Stewart (in part), Rec. Ind. Mus. V1, . 70. 1920. ome lhasae, Annandale and Hora (in part), Rec. Sad. Mus. XVIUII, p. 179. This species is represented in our collection by several voung and half-grown specimens. It closely resembles Nemachilus lhasae Regan with which Annandale and myself! confused it when dealing with the fish of Seistan. Both Lloyd and Stewart referred these specimens along with Nemachilus lhasae to N. stoliczkae. Diplophysa stewarti is, however, readily distinguished by the pre- sence of a second air-bladder and also by the nature of its skin, which is tuberculate all over. The dorsal profile is highest near the nape, in front of which it slopes considerably to the tip of the snout. The body is thickest anteriorly and gradually and regularly slopes to the base of the caudal fin. The head is round, narrow and pointed; its length is contained 4°2 times in the length of the fish without the caudal fin. The body is deepest at its commencement and the greatest depth of the hody is contained 1°6 times in the length of the head. The eyes are placed in the middle of the head and are scarcely visible from below; the diameter is contained 3°4 to 3°7 times in the length of the head. ‘There are six barbels, 4 rostral and 2 maxillary. The maxillary barbels are the longest; they are slightly longer than the diameter of the eye. The upper lip is fringed and the lower is interrupted in the middle and is strongly papillated. The lateral line is complete; anteriorly it is continued as a series of open pores below the eyes. There are a few open pores on the dorsal surface of the head near its posterior border extending downwards on each side to join the lateral line. The dorsal fin commences considerably in advance of the ventrals and its origin is equidistant from the tip of the snout and the base of the caudal fin. It is higher than the depth of the body immed- iately below it. The ventrals extend beyond the anal opening and almost reach the base of the anal fin. The caudal peduncle is long and narrow; its least height is contained 6:1 to 66 times in its length. The caudal fin is deeply concave with the upper lobe considerably longer than the lower. ! Annandale and Hora, Rec. Jud. Mus. XVIII, p. 179 (1920). 1922.] S. L. Hora: Central Asiatic Cobitidae. yi This species exhibits sexual dimorphism and the males can readily be distinguished by a tuberculate pad below the eye. The secondary bladder is large and constricted in the middle. It differs from the normal Cyprinoid type in the fact that the pneumatic duct here opens at the anterior end instead of in the constricted region. Textr-riGc. 2.—Cobitid fishes from Eastern Tibet. . Lateral view of Nemachilus tibetanus Regan. . Under surface of head and chest of same. . Lateral view of Diplophysa stewarti, sp. nov. . Under surface of head and chest of same. a b c d The colour in spirit is characteristic of the species. There are short black bars along the lateral line and on the back. The belly and under surface of the head and also the genéral colour of the body is pale olivaceous. The dorsal and the caudal fins are marked with black. Type-specimen.—F. 2894/1, Zoological Survey of India (Ind. Mus.), Locality:—The specimens were collected by Capt. R. S. Kennedy, I.M.S., and Capt. F. H. Stewart, I.MS.. in a small Records of the Indian Museum. [VoLr. XXIV, 72 stream flowing into Rham-tso (Eastern Tibet). There are two young specimens from Se-Chen iu Tibet, which I also refer to this species. Genus Adiposia, Annandale & Hora. 1920. Adiposia, Annandale and Hora, Rec. Ind. Mus. XVIII, p. 182. The genus was recently proposed by Annandale and myself for two species of Cobitid fishes from Seistan with a long soft dorsal fin between the bases of the dorsal and the caudal fins. We also referred a species from Turkestan, Adiposia longicauda (Kessler) ,' to this new genus. Genus Nemachilus v. Hass. The thirteen species of the genus Nemachilus from Central Asia in the collection differ from the numerous forms known from the Indian Empire in their large size and almost subcylindri- cal form, None oi them, moreover, possesses the vertical bands of pigment which characterize those from lower altitudes. The following is an artificial key to the Central Asiatic species of Nemachilus in the collection of the Indian Museum :— 1. Ventrals terminating a considerable distance in front of anal opening. A. Kyes wholly in anterior half of head, dorsal commenc- ing in advance of ventrals. B. Eyes in middle of head ; of dorsal 11. Ventrals just rez ching ¢ or r extending beyond anal opening. A. Anterior origin of dorsal almost equidistant between tip of snout and base of caudal. 1. lateral line incomplete, ending shortly atter its com- mencement 2. Lateral line complete or becoming somewhat obscure behind anal fin. a. \tye almost in middle of head. 1. Pectorals as long as liead il. Veceg as shorter than head. . Least height of caudal peduncle almost equal ae diameter of eye . Least height of caudal. ‘peduncle considerably orere than diameter of eye. 6. Snout longer than postorbital part of head 8B. Anterior origin of dorsal not equidistant between tip of pe and base of caudal. Anterior origin of dorsal nearer tip of snout than base “of caudal Anterior origin of dorsal nearer base of caudal than tip wee snout, a. Eye in middle of head. IE acopals distinctly extending beyond anal opening. ventrals commencing in advance . Least height of caudal peduncle 4 times in its_ length, lower lip alinost continuous .., B. Least height of caudal peduncle 5 times in its length, lower lip widely interrupted i. Ventrals just reaching anal opening N. yarkandensis. N, gracilis. N. vittatus. N. yasinensisé . NV. lhasae. N. kashmirensis. N. tenuis. N. ladacensts, NV. stoliczkae. N, tenuicauda. N. marmoratus. L Kessler ‘! Pisces’ figs. 22, 23 (1874). "in Iedtschenko's '' Refse in Turkestan,’ p. 38, pl. vi 1922] S. L. Hora: Central Asiatic Cobitidae. 73 b. Eye not in middle of head. i. Snout shorter than postorbital part of head. a, Anal fin separated from caudal by a distance almost equal to its own length "4 : B. Anal fin separated from caudal by a distance considerably less than its own length .. WN. yasinensis?. ii. Snout longer than postorbital part of head. .. WV, tibetanus. N, microps. Nemachilus yarkandensis Day. 1876. Nemachilus yarkandensis, Day, Proc. Zool. Soc. London, p.79. 1878. Nemachilus yarkandensis, Day, Sci. Res. 2nd Varkand Mission, . Ichthyol., p. 14, pl. v, fig. 3. 1889. Nemachilus yarkandensis, Herzenstein (in part), Wiss. Res. Pyzewalski Central As. Reis., Zool. 111 (2), p. 74. 1910. Nemachilus yarkandensis, Zugmayer, Zool. Fahrb. Syst. XXIX, p. 295. The Indian Museum possesses a large number of specimens of this species from Yarkand, Pas Robat, Yaukihissar and Kashgar. Besides these there is one specimen about 132 mm. in length labelled as having come from Kashmir which I am convinced also belongs to this species. So far N. yarkandensis. has been re- corded only from the Tarim River system and its extension into Kashmir is very doubtful and requires confirmation. My specimens correspond in every detail to the typical form. The various varieties described and figured by Herzenstein are not represented in our collection ; probably they are all far Eastern or Chinese forms. Nemachilus tayvimensis Kessler! has been considered to be synonymous with N. yarkandensis by Herzenstein, who figures Kessler’s original specimen of N. tarimensis as N. yarkandensis {s. st.). After a careful comparison I am lead to believe that the two species are different and that N. ¢avimensis as figured by Herzenstein differs from N. yarkandensis in the following points:— N. tarimensis. N. yarkandensts. The commencement of the dorsal fin is | ‘The commencement of the dorsal fin is almost equidistant from the tip of the distinctly nearer to the base of the snout and the base of the caudal caudal fin than to the tip of the fin. snout. The eyes are large and are not situated | The eyes are small and are situated entirely in the anterior half of the head. in the anterior half of the head. As regards the three varieties of this species, it is difficult to discuss their true relationships without examining Herzenstein’s specimens. N. yarkandensis longibarbus differs from the typical form in the commencement of the dorsal fin, which is situated in the middle of the body, the longer barbels and the curve of its dorsal profile. Probably it represents a new species. The other two varieties, brevibarbus and macropterus somewhat resemble our specimens. ! Kessler, Bull. Acad. St. Petérshourg XXV, p. 300 (1878). 74 Records of the Indian Museum. [ VOL. XXIV Nemachilus gracilis Day. 1876. Nemachilus gracilis, Day, Proc. Zool, Soc. London, p. 798. 1878. Nemachilus gracilis, Day, Sci. Res. znd Varkand Mission, Ichthyol., p. 16, pl. iv. fig. 5. : 1878. Nemachilus gracilis, ay, Fish. India \\, p. G21. 1889. Nemachilus gracilis, Day, Faun. Brit. Ind. Fish, 1, p. 237. 1898. Nemachilus stoliczkae, Alcock, Rep. Nat. Hist. Res. Pamir Bound, Comm.,, p. 35. ‘his species 1s readily distinguished by the nature of its lower lip which is widely interrupted in the middle and is thrown into a longitudinal fold on either side. I have examined Day’s type- specimen ‘‘ from Basgo, on the head waters of Indus.’’ oe I also refer to this species a specimen from Lukong River and several others from the affluents of the Yasin River near Darkot. The latter were collected by Col. Alcock. The waters of both these streams flow directly cr indirectly into the Indus River. ‘These specimens were previously recorded as N. stolicekae. Several young, half-grown and adult specimens have recently been collected in the Kashmir Valley. ‘The specimens were ob- tained from a lake about four miles from Sonmarg. ‘The species exhibits marked sexual dimorphism. The eggs are minute. The adult individuals possess 6 to 7 broad black bands across the back. In young specimens there is a series of black dots along the lateral line and the dorsal surface is mottled with black and brown. Nemachilus vittatus (Heckel). 1838. Cobitis vittata, Heckel, Fische Kaschm., p. 80, pl. xii, figs. 3 and 4. 1844. Cobitis vittata, Heckel, in Hiigel’s Kaschmiry 1V, p. 382, fig. Gunther ' combined Heckel’s two species of this genus from Kashmir and adopted for them the specific name marmoratus. Day? followed Gitinther and recognised only one form from the Kashmir: lakes. Zugmayer® perhaps doubted Giinther’s identifi- cation and in recording Nemachilus marmoratus from ‘‘ Wular Lake’”’ gave Giinther as the author of the species. On examin- ing the old collection of the Indian Museum, I find that out of 17 specimens from the Kashmir lakes, 16 belong to N. viltatus and one to another species, N. vittatus can be readily recognised by the nature of its lateral line which ends shortly after its com- mencement. I have not included references by Giinther, Day and Zugmayer under the title of this species as it is impossible to be sure of the identity of the species they recorded. They do not make any | Gunther, Cat. Brit. Mus. Fish. VII, p. 356 (1868). 2 Day, Proc. Zool. Soc. London, p. 798 (1876); Fish. India 11, p. 620 (1878). ® Zugmayer, Zool. Fahrb, Syst. XNAX, p. 296 (1910). S. L. Hora: Central Asiatic Cobttidae. 75 1922. ] mention of the lateral line in which, as explained above, the speci- fic character is to be found. Quite a number of specimens have recently been brought back from several places in the Kashmir Valley by the members of the Zoological Survey of India. Nemachilus yasinensi Alcock. Nemachilus yasinensis, Alcock, Rep. Nat. Hist. Res. Pamir Bound. Comm., p. 38, pl. ii, figs. 2, 2a. . This species has hitherto been known from a single male speci- men procured by Col. Alcock in the Yasin River. The specimen is now preserved in our collection. A large number of specimens have recently been obtained from a small stream flowing into the Sind River, a tributary of the Jhelum River. Among these there are three female specimens which differ considerably from the males. The following are some of the chief points of difference :— Male. Female. 1598. The dorsal fin commences midway bet- ween the tip of the snout and the base of the caudal fin. The snout is slightly longer than the pnstorbital part of the head. The lateral line is continued to the base of the caudal fin. ‘The caudal fin is forked. The dorsal fin commences somewhat nearer to the base of the caudal than to the tip of the snout. The snqut is shorter than the post- orbital part of head, The lateral line ends in front of the base of the ventral fins. The caudal fin is either rounded or truncate. Besides these the female specimens possess short paired fins, small eyes and a deep caudal peduncle as compared with the males. The males possess well-marked secondary sexual characters such as are described towards the end of this paper. The eggs are small. The species is now known from the head-waters of the Indus and Jhelum Rivets. Nemachilus lhasae Regan (Text-figs. 3a-c.) 1905. Nemachilus lhasae, Regan, Ann. Mag. Nat. Hist. (7) XV, Pp, 301. 1908. Nemachilus stoliczkae, \.loyd (in part), Rec. Ind. Mus., Il, p. 341. 1911. Nemachilus stolicekae, Stewart (in part), Rec. /nd. Mus. V1, Pp. 70 1920. Nemachilus lhasae, Annandale and Hora (in part), Rec. /nd. Mus. XVIII, p. 179. There are several young and half-grown specimens of this species before me, which have been referred to Nemachilus stoliczkae by Lloyd. They were collected by Capt. Kennedy and Capt. Stewart in Rhamtso, Nyang-chu, Langma-thang-chu, Phari and to the S.W. of Dochen, all in Eastern Tibet. My specimens agree with Regan’s description of the species. The young individuals, however, possess black blotches along the lateral line besides short cross-bars on the back. The species exhibits well-marked sexual dimorphism. 76 Records of the Indian Museum. [VoL. XXIV, Nemachilus kashmirensis, sp. nov. 1876. Nemachilus rupicola, Day, Proc. Zool. Soc. London, p. 799. 1878. Nemachilus rupicola, Day, Sct. Res. 2nd Yarkand Mission- Ichthyol., p. 17. To this new species I assign several specimens recently collected in Verinag, Kukarnag and in a small stream flowing from the Kashmir waterworks reservoir to the trout farm at Harwan. The species is characterized by an emarginate caudal fin and by the presence of broad, black bands across the back. Probably these characters led Day to refer some of his Kashmir examples to Nemachilus vupicola (McClell.)' I have recently visited the TEX?-FIG. 3.—WVemachilus lhasae, Regan. a. Lateral view of adult specimen. b. Same of young specimen. c. Under surface of head and chest of adult specimen. Simla Hills and have obtained some specimens of N. rupicola, which differ from N. kashmirensis in the following points :— N. kashmyrensis. The ventrals extend beyond the anal opening and almost reach to the base of the anal fin. The pectorals are shorter than the head. There are no definite black bands on the sides of the body. The body is absolutely devoid of a scaly covering. N. rupicola. The ventrals do not reach the anal opening and are separated from the anal fin by a considerable distance. The pectorals are longer than the head. There are several black bands on the sides of the body. There are minute scales covering at least the posterior three-fourths of the body. I propose to give a detailed description with figures of this species in my paper on the Indian species of the genus to be pub- lished in this journal at some future date. ! McClelland, Fourn. As. Soc. Bengal V11, p. 948, pl. 55, fig. 3. 1922.] S. L. Hora: Central Asiatic Cobitidae. ag Nemachilus tenuis Day. 1876. Nemachrlus tenuis, Day. Proc. Zool. Soc. London, p. 796. 1878. Nemachilus tenuis, Day, Sci. Res. 2nd Yarkand Misston, Ich- thyol., p. 15, pl. v, fig. 4. 1898. Nemachilus tenuis, Alcock, Rep. Nat. Hist. Res. Pamiv Bound. Comm., p. 14. 1906. Nemachilus stenurus, editorial note to Regan, Fourn. As. Soc. Bengal, II, p. 8. 1920. Nemachilus stolicekae, Annandale and Hora (in part), Rec. [nid. Mus. XVIII, p. 178. The specimens from which Day drew up his description of Nemachilus tenuis, came from two sources, from “ Aktash, where the waters of the Ak-su Pass to the Oxus’’ and from ‘‘ Yankihissar, where the rivers go to the Varkand River.’’ Day’s specimens from the latter locality are not to be found in our collection, though there are several specimens of other species of the genus from the same place. The record, therefore, requires confirmation as I think it improbable that the species really extends to the Tarim river-system, Several specimens from the Great Pamir, whence the waters pass to the Oxus system, were correctly referred to this species by Alcock. Quite recently Annandale and I (op. cit.) identified the Seistan examples as N. stoliczkae, but I now believe that they represent N. tenuis. ‘The Seistan specimens were collected in the Helmand River which may once have formed a part of the once- extensive Oxus system. Regan (op. cit.) referred these specimens to N. stenurus on account of their long and narrow caudal peduncle, but this is also a character of N. tenuis. The two species differ in the following points :— N. stenurus Herz. N. tenuis Day. The commencement of the dorsal fin is | The commencement of dorsal fin is nearer to the tip of the snout than to either equidistant from the base of the base of the caudal fin. the caudal fin and the tip of the snout or it is slightly nearer to the former than to the latter. The lower lip is continuous and entire. | The lower lip is widely interrupted in the middle and is greatly pliated. N. stenurus was described from Dy-tschu, the sources of the Yang-tse-kiang River; while N. fenuis is known from the Oxus systein. Vinciguerra’ on the authority of Regan referred his examples from Skardu in the Indus System to N. stenurus for he writes, “nel riferire questi individui al N. sfenuvus sono confortato dall’ avviso di Tate Regan, al quale li hc comunicati.”’ The specimens require re-examination. It is evident from the above discussion that the character of a long and narrow caudal peduncle is shared by a number of species. Such species are N. stenurus from the sources of the Yang-tse-kiang, N. tenuis from the Oxus system and N. lhasae ! Vinciguera, Ann. Mus. Stor. Nat. Genova XI.VII, p. 148 (1916). 7 Records oj the Indian Museum. [VoL. XXIV, from Eastern Tibet. The Skardu specimens probably represent another form in the same series. Nemachilus ladacensis Gunther, 1808. Memachilus ladacensis, Giinther, Cat. Brit. Mus. Fish. VII, Pp. 350. The Indian Museum possesses only one specimen of this species six and a half inches in length, from Kashmir ; it agrees closely with Giinther’s description but had been referred to N. stoliczkae by Day. Day’s!' N. ladacensis differs from Giinther’s account in several respects and probably represents a different species. ‘The descriptions differ in the following points :— N. ladacensis Giinther. N. ladacensis Day. ‘The origin of the dorsal fin is nearer | ‘‘ Dorsal commences midway between to end of snout than to the root of the front edge of the eye and the the caudal.”’ base of the caudal fin. ”’ : ‘The free portion of tail is very low, its | ‘‘ free portion of the tail twice as high depth being nearly one fourth of its as long at its base.” length. ” ‘he caudal fin is rounded ? The caudal fin is emarginate. The proportions are also different in the two species. Unfortunately the specimen from which Day drew up his description and which he “‘ deposited in the Indian Museum ”’ is not now to be found. Day’s specimen was said to have been collected by von Schlagintweit at Gnari Khorsum, Tibet. N. ladacensis is known from T,adak and Kashmir. x Nemachilus stoliczkae (Steind.). 1866. Cobitis stoliczkae, Steindachner, Verh. Zool-bot. Ges. Wien, p. 793, pl. xiv, fig. 2. 1868. Nemachilus stoliczkae, Gtinther, Cat. Brit. Mus. Fish. VAI, p- 360. A very wide interpretation has been given to this species by Day,’ Herzenstein,*’ Giinther,* and several otlier ichthyologists, I have examined a large number of specimens in our collection from Lukong stream, Chagra, Yarkand, Sirikol and Aktash which were referred to this species by Day but find that several distinct forms are represented among them, There are only six specimens which I can definitely refer to this species, one from Rupshu, the type-locality, three from Lukong Stream, one from Chagra and one from Kashmir. ‘The waters from these places pass to the Indus River. ! Day, Proc. Zool. Soc. London, p. 797 (1876); Sci. Res. 2nd Varkand Mission, Ichthyol., p. 15, pl. iv, fig. 4. 2 Day, Proc. Zool. Soc. London, p. 795 (1876); Sci. Res. 2nd Varkand Mission, Ichthyol., p. 14, pl. v, fig. 2 (1878). 3 Herzenstein, Wiss. Res. Prgewalski Central As. Rets., Zool. III (2), p. 14 1888). 4 Gunther, in Pratt’s “ Snows of Tibet’’, p. 249 (1892). 1922. ] S. L. Hora: Central Asiatic Cobttidae. 79 Herzenstein (of. cit.) has recognised several varieties of this species. most of which so far as I can judge from the figures represent different species. My specimens agree with the typical form, of which I have examined one specimen from Rupshu. Vinciguerra! has recorded the species from Skardu, but it appears from his description that he has grouped several distinct forms under one name. In identifying his specimens as N. stoliczkae he has followed Day and Herzenstein for he observes that, ‘‘ questa determinazione é basata non tanto sulla des- crizione e figura originale, quanto su quelle di Day, di Giinther e specialmente di Herzenstein.’ Lloyd® referred some specimens trom Eastern Tibet to this species. I have heen able to recognize at least three different forms among the material he examined! two of them belonging to the genus Nemachilus, viz. N.lhasae Regan and N. tibetanus Regan,® while the third belongs to the genus Diplophysa and is described here as new. Day * in his later works regarded Nemachilus Dine Giinther * as Synonymous with this species. Thave not examined any speci- men of Giinther’s species, but it appears from the description that the two are different. In N. griffithi ‘‘ the origin of the dorsal fin is midway between the root of the caudal and the end of the snout,” while in N. stoliczkae ‘‘the origin ot the dorsal fin is conspicuously nearer to the root of the caudal than to the end of the snout.” Nemachilus tenuicauda (Steind,). 1866. Cobitis tenutcauda, Steindachner, Verh. Zool.-bot. Ges. Wien XVI. p. 792, pl. 17, fig. 3. 1868. Nemachilus tenuicauda, Giinther, Cat. Brit. Mus. Fish. VII, P- 357: This species closely resembles N. stoliczkae, from which it is distinguished by its elongate and narrow caudal peduncle and by its colouration. Nemachilus tenutcauda is represented in our collection by two specimens from Leh. Nemachilus marmoratus (Heckel). 1838. Cobitis marmorata, Heckel, Fish. Kaschm., p.-76, pl. xii, figs. Rand 2, °4 >> 1844. Cobitis marmici@..a, Heckel, in Hiigel’s Kaschmiy IV, p. 380, fig. This is apparently a rare species and is represented in our collection by a few specimens recently obtained in Kashmir. The specimens were obtained from Kukarnag Spring and from ponds on the road between Martand and Ichabal. ! Vinciguerra, Ann. Mus. Stor. Nat..Genova XLVII, p. 146 (1916). 2 Lloyd, Rec. Ind. Mus. 11, p. 341 (1908). ® Regan, Ann. Mag. Nat. Hist. (7) XV, pp. 187 and 301 (1905) + Day, Fish. India Il, p. 620 (1878). 6 Giinther, Cat. Brit. Mus. Fish. VII, p. 360 (1868). 80 Records of the Indian Museum. [VoL. XXIV, I am unable to say whether Zugmayer’s (op. cit.) specimens represent this species or N. vittatus, but the latter is undoubt- edly more common in tlie Kashmir lakes. For the reasons already given under N. vittatus I have not included references by Giinther and Day under the title of this species. Nemachilus microps (Steind,). 1866. Cobitis microps, Steindachner, Verh. Zool.-bot. Ges. Wien XVI, p- 794, pl. 13, fig. 3. : ; 1868. Nemachilus micrvops, Giinther, Cat. Brit. Mus. Fish. V1, p. 357. This species is readily distinguished by its small eyes. We have two specimens in our collection from Mecma! (Yarkand Mission, Dr. Stoliczka’s collection) which agree with Stein- dachner’s description of the species. These specimens had previ- ously been referred to Nemachilus stoliczkae. ‘The male possess- es well-marked secondary sexual characters below the eyes. The species was originally described from Leh and in all probability the waters from Mecma pass to the Indus River. Nemachilus tibetanus Regan. (Text-figs. 2a, 2b.) 1905. Nemachilus tibetanus, Regan, Ann. Mag. Nat. Hist. (7) KV, p. 187. 1908. Nemachilus stolicskae, \.loyd (in part), Rec. Ind. Mus. II, p. 341. 1911. Nemachilus stoliczkae, Stewart (in part), Rec. Ind. Mus. V1, p. 70. The specimens before me ot this species were collected by Capt. Kennedy in Nyang-chu at Kangmar and by Capt. Stewart in Gyang-tse. These had been referred to Nemachilus stoliczkae by Lloyd and Stewart. The species differs from N. lhasae, which is known from the adjacent region, by the greater depth of its caudal peduncle, by the position of the dorsal fin, whose commence- ment is situated nearer to the root of the caudal than to the tip of the snout, and by the position of the eye, which is nearer to the tip of the snout than to the posterior margin of the head. Nemachilus tibetanus exhibits a well-marked sexual dimor- phism. Nemachilus sp. There are several specimens in our collection from Sirikol, which have been referred to Nemachilus stoliczkae by Day with the following remark, “in specimens from Sirikol the snout is rather more pointed.” I am unable to refer these specimens to any of the known species of the genus, but on account of their bad state of preservation I do not propose to describe them as a new species. ! The Yarkand Mission made collections in several places outside Yarkand. I have not been able to determine the exact locality of Mecma, but I suppose waters from this place flow into the Indus river-system. 1922.] S. L. Hora: Central Asiatic Cobitidae. 81 Among the specimens from this locality two forms are repre- sented, one in which the caudal peduncle is very low and the com- mencement of the dorsal fin is situated at an equal distance from the tip of the snout and the base of the caudal fin; while in the other the caudal peduncle is fairly deep and the origin of the dorsal fin is distinctly nearer to the base of the caudal than to the tip of the snout. Most of the females, which are about 54 mm, in length, are full of eggs. The snout is long and pointed and the eyes are situated in the middle of the head. NOTE ON THE SECONDARY SEXUAL CHARACTERS OF CERTAIN SPECIES OF COBITID FISHES FROM HIGH ALTITUDES IN CENTRAL, ASIA. In the Indian species of the genus Nemachilus which ex- hibit sexual dimorphism the male is provided with ‘‘a slit-like e eal Ss SA es SS Texr-Fic. 4.—Lateral view of head and upper surface of pectoral fin in a male specimen of Nemachilus tibetanus Regan, showing secondary sexual characters. deep groove in front of the eye which bends round a small knob- like rounded flap of skin protruding below the anterior one-third of the orbit, the ridge above the groove appearing slightly swollen and cushion-like.’’ The pectoral fins are also modified where. “there is a kind of padding and thickening on the upper surface ”’ and ‘‘on the padding, minute hooked denticular outgrowths are noticed.’’ These secondary sexual characters were described by Chaudhuri in Nemachilus mackenziei' and N. manipurensis? and ! Chaudhuri, Rec. Jnd. Mus. V, p. 183. 2? Chaudhuri, Rec. Jad. Mus. VII, p. 443, pl. xl, figs. 4, 4a, 46; pl. xli, figs. I, 1a, 1b. 82 . Records of the Indian Museum. [VoL XXIV, I have noticed similar modifications in mature males of several other Indian species of the genus. In the Central Asiatic forms, however, the secondary characters of the male are more marked and somewhat complicated. Of the thirteen species of the genus _ Nemachilus referred to in this paper, seven show marked sexual dimorphism. The male in these is usually provided with a raised tuberculate area below the nares, separated ventrally by a groove from the adjacent parts of the skin. The area is almost rectan- gular, commencing at the corner of the lips and extending poste- riorly below the anterior third of the orbit. In certain species such as N. tenuis, N. yasinensis and N. tubetanus (fig. 4), there is another tuberculate area immediately behind the first one. Sometimes the tubercles are irregularly scattered on the operculum and the sides of the head behind the eyes. In ali species that exhibit Vext-ric. 5.—Tubercles covering secondary ‘sexual pads of male of J. tibetanus (highly magnified). sexual dimorphism, the pectoral fin-rays are provided with thickened tuberculate pads on their dorsal aspect. ‘These tubercles on the fin-rays are not to be confused with encysted glochidia, which are sometimes found in this position, though they resemble them closely. A few scattered tubercles are sometimes found on the under surface of the pectoral fin-rays. In both the species of the genus Diplophysa in our collection, the male is modified on exactly the same lines as has already -been described for the genus Nemachilus. The structure of the tuberculate areas is somewhat interesting. Each of the tubercles is provided with a short, stout spine-like outgrowth (fig. 5) which is sharp and slightly curved towards the end. The spine rests on a broad cushion-like rounded base. Recently I have collected an interesting specimen of Nema- chtlus from the Simla Hills. It possesses a groove and a small 1922.] S. L. Hora: Central Asiatic Cobttidae. 83 knob-like rounded flap of skin below the eye, but on dissection was found to be full of eggs. There is no padding on the dorsal aspect of the pectoral fin-rays and even the sexual character below the eye is not so well marked. I have examined a large number of specimens of the species to which I think this example probably belongs, N. rupicola (McClelland), and have not been able to find any other specimen with secondary sexual characters. Ne eae eee eee Coa Daa lee cee acne heed reas A ae: 4 - Xe nh er ep A NOTE ON BEES OF THE GENERA XYLOCOPA AND BOMBUS IN THE INDIAN MUSEUM. By Cepric Dover, f.E.S. The accumulation of a large amount of unnamed carpenter- and bumble-bees in the collection of the Zoological Survey of India has led me to attempt their identification; and, as, though no novelties were among them, my results are not quite without interest, I have drawn up the following short note in the hope that it will prove useful. Subfamily XYLOCOPINAE.' Xylocopa tenuiscapa Westw. 1921. Nylocopa tenutscapa, Dover; Rec. Ind. Mus. XXII, p. 389. In the paper quoted I have given references to works which give characters by which the female of X. latipes and X. tenuiscapa may be readily separated. and I have also quoted Maidl, who re- gards Sichel’s albofasciata as the female of this species. ‘‘ In Bing- ham’s description of the male for tibiae read lasitarsi’’ (Cockerell). The Indian Museum possesses a specimen from Tenasserim which is probably a Jatipes 9 but differs from it in being much larger (about 37 mm.) and there are two deep dents on either side of the disc of the mesonotum, which is smooth and brightly polished. Xylocopa acutipennis Smith. The Indian Museum has examples from the following locali- ties unnoticed by Bingham: Darjiling District, above Tura in the Garo Hills of Assam, 3500 [t., and the Dawna Hills in Lower Burma, 2000-3000 ft. Xylocopa attenuata Perez. 1852. NXylocopa pictifrons. Smith, Tvans. Ent. Soc.(2) XI, p.42, Q necd. 1297. Xylocopa pictifrons, Bingham, Faun, Brit. Ind. Hym.\, p. 538, 2. 1901. Xylocopa attenuata, Perez, Act. Soc. Linn. Bordeaux ILXXV (6) Vi, pad, 9. 1912. Xylocopa attenuata, Maidl, Ann. Nat. Hofmus. Wien XXVI, p- 287. 1921. Xylocopaattennata, Dover, Fourn. Bomb. Nat. Hist. Soc. X XVII, p- 961, The species is found sparingly in the Indian plains, and commonly in Sikkim and Kumaon. It is found also in China, ' | have thought it best in this note not to recognise the genus Mesotrichia. 86 Records of the Indian Museum. [VoL. XXIV, Java, Formosa and the Malay Peninsula. Some confusion has existed as to the sexes of the species described by Smith and Bingham as picizfrons, but this has been admirably cleared up by Maidl in his paper. He regards the male described by Smith as typical pictifrons, while Bingham’s male and Smith’s female are a separate species for which he has adopted the naine attenuata of Perez. Xylocopa auripennis Lepel. In addition to the localities noticed by Bingham, the Indian Museum also possesses specimens from the Darjiling District, the Naga Hills and Sibsagar in Assam, South India and Nepal. The species is supposed to be mimicked by a Sphingid moth (Sataspes hauxwelli), which according to De Niceville (Journ. Bomb. Nat. Hist. Soc. XIII, p. 174) was ‘‘a beautiful mimic of the very common large blue carpenter-bee Xylocepa auripenms, Lepeletier.”’ The wings of the moth are a deep indigo-blue with bronze markings, which scarcely resembles the wings of the bee, and in the cabinet the whole insect seems entirely different. De Niceville does not say that the bee and the moth were taken together, and in the absence of definite field-observations, the moth has little claims to being a mimic of the Xylocopa. Xylocopa dissimilis Lepel. The Museum has specimens from Bangalore, Bandra in the Bombay Presidency, Mong-Wan in Yunnan (W. China) and South- ern China. : Xylocopa fenestrata Fabr. ¢ Xylocopa bombayensis, Cam. 2.58. 1921. Xylocopa fenestrata, Dover, Rec. Ind. Mus. XXII, p. 390. In the paper cited, I have noticed wiat appears to be an aber- ration of X. fenestvata from Barkuda Island in the Chilka Lake, with a comparatively large, and a small, almost reniform, hyaline marking on each of the hindwings. The Indian Museum possesses another specimen from Hamirpur Road in the United Provinces (Caunter , x'11), which has the lower halves of the wings semi-hyaline. The fact that King described an example with semi-lunate, hyaline markings on the hindwings under the name /unata, makes me now think that aberrations of this species, with hyaline markings of some sort on the wings are perhaps not uncommon. It might be of interest to note here that I remember to have seen a specimen of the closely allied African X. cavinata with an irregular hyaline patch on the right forewing. Can it be that these markings are caused by injuries sustained in the earlystages? X. fenestrata is a common Indian species, extending as far Celebes on the south-east and probably into Australia, and Madagascar on the south-west. It does not penetrate into South Africa, but is replaced there by 2922.] C. DovER: Notes on Bees. 87 X.carinata. Meade-Waldo has shown! that Cameron’s X. bomba- yensis is asynonym of this species. Xylocopa amethystina Fabr. Bengal, Chota Nagpur, Bihar and Sind are not recorded by Bingham. This is an apparently widely. distributed species in India. Xlyocopa bryorum Fabr. The Museum has specimens from Assam and the Andamans. Xylocopa collaris Lepel, This species is found in most parts of India, Burma and Ceylon and is known from Borneo, Sumatra, Java, the Philippines, Celebes and Malacca and from the Palaearctic Region. “Xylocopa tranquebarica Fabr. © 1804. Bombus tranquebaricus, Fabr., Syst. Piez., p. 343. 1917. Xylocopa tranquebarica, Cockerell, Philipp. Fourn. Sct. XII, p. 6 1921. ee vufescens, Dover, Rec. Ind. Mus. XXII, p. 390. Found in Sikkim, Bengal, South India, Burma, Andamans, Java, Sumatra, Borneo and the Philippines. There are three ex- amples in the Indian Museum under the name ferruginea which I think really belong to this species. Prof. Cockerell has shown that the more generally used name vufescens will have to be sunk in favour of tranquebarica. He has also noted its crepuscular habits. Xylocopa caerulea Fabr. This beautiful species has been found in Sikkim, Burma, Ceylon, Annam, Sumatra, Borneo, Java and New Caledonia. Xylocopa flavonigrescens Smith. 1918. Xylocopa flavonigrescens, Cockerell, Entomologist, LI, p. 104. The Zoological Survey possesses examples from Sikkim, Sylhet, Tenasserim, Ton-Kin and Malacca. Meade-Waldo, basing his opinion on the male, thought Cameron’s malayana (Proc. Zool. Soc. Lond. 1901, p. 32) to be the same as this species, but Prof. Cockerell notes that a female from the island of Penang is the same as X. malayana. Xylocopa nitidiventris, X. dubiosa aud X. convexa Smith. 1878. Smith, Scien‘. Res. 2nd. Vark. Miss. (Hym.) pp. 7 and 8. The types of these species, described from the neighbourhood of Yangihissar in Yarkand, are in the collection of the Indian ! Ann. Mag. Nat. Hist. X1V, p. 404, 1914. 88 Records of the Indian Museum. [VoL. XXIV, Museum. ‘The U.S. National Museum has specimens of X niti- diventris from Kukier, Eastern Turkestan. _ Subfamily BOM BINAE.! Bombus montivagus Smith. ‘The Indian Museum has examples from Upper ‘Tenasserim and Take-pum Mt. on the Chinese Frontier in N KE. Burma. A form has also been taken in Onari in British Garwhal, 11,000 ft., which has the colour of the pubescence on the apical three segments of the abdomen almost snow-white and not fulvous red. Bombus lapidarius var. tunicatus Smith. 1897. Bombus tunicatus, Bingham, Faun. Brit. Ind. Hym. |, p. 549. 1910. Bombus tunicatus, Cockerell, Ann. Mag. Nat. Hist. V, p. 417. 1916. Bombus lapidarius var. tunicatus, Meade-Waldo, Ann. Mag. Nat. Hist. XVI, p. 467. Following Meade-Waldo 1 consider tunicatws and Cockerell’5 gilgitensis to be varieties of the Ruropean B. /apidarius Linn. The Museum possesses specimens of the former variety from Garwhal, Simla Hills, Mussoorie, Nepal, and two examples from Calcutta. In Nature for May Igth, 1921, I recorded the capture of the two Calcutta examples and mentioned having seen what was probably a species of Bombus at the base of the Eastern Himalayas, as ‘“bumble bees”? are supposed never to descend below 3,000 ft. Burkill (Journ. As. Soc. Beng. (nm. s.) II, p. 521, 1906) found B. haemorrhoidalis common in the N.W. Himalayas at 1,600 ft., but the capture of Bombus actually in the plains is astonishing, and it is probable that such an incident may never occur again, _ though I originally mentioned that these bees probably occur, very rarely, in the plains. ‘There is an old record of B. orientalis in Calcutta which J think must be authentic; but as to how these strictly hill-species have been found here I can offer no explanation other than that these species of Bombus probably nest in the ground and have been conveyed here through the agency of man. A fly, Criorhina imitaltor cf the family Syrphidae, closely resembles this species and the case appears to be one of real mimicry. Brunetti in his original description (Rec. Ind. Mus. XI, Pp. 237, 1915) stated that it was a mimic of the bee Bombus trifas- ciatus (as understood by Bingham), but I think it will be admitted that it resembles ¢wnicatus more closely in the light pubescence on the anterior parts of the thorax, on the scutellum, on the basal abdominal segments, and in the colour of its wings and legs. The pubescence on the apical abdominal segments is also reddish, but unfortunately, it is not quite so dense as in the bee it resembles. Hingston in A Naturalist in Himalaya (Witherby : 1920, p. 184) . notices the resemblance of Bombylius to Bombus and of a species 1 As the use of the term Bremus for Bombus is dependent on the validity of the ‘ Erlangen ’’ list, and this is still a debatable point I have preferred to use the more generally known name, c+ C. DovER: Notes on Bees, 89 of Bombvlius to B. lapidarius var. tunicatus. I have never noticed this myself and judging from his description I think it just pro- bable that his Bombylius is really a Criorhina. Bombus eximius Smith. The Darjiling District, Khasi Hills, Shillong, and Mong-wan in W. China may be added to the localities given by Bingham. Bombus flavescens Smith, The Darjiling District, Nepal and Kumaon may be added to the localities given by Bingham. Bombus funerarius Smith. There are specimens in the Indian Museum from the Western Himalayas. Col. Bingham remarks that individuals with the pubescence on the apical three segments bright orange-red instead of greyish have only been found in Sikkim, but I have seen an example from the W. Himalayas. Bombus alienus Smith. 1897. Bombus? vallestvis, Bingham, Faun. Brit. Ind. Hym. I, p. 553. 1916. ere es alienus, Meade-Waldo, Ann. Mag. Nat. Hist. XVII, p. 467. This species (omitted from the ‘‘ Fauna’’) was taken in October, 1903, by Mr. R. E. Turner in Shillong. B. vallesiris agtees fairly well with the description of alienus, but as Smith’s type of the latter species is not available in Calcutta, and his type and? cotypes of vallestyis in the Indian Museum are almost unrecognisable, I can cffer no definite opinion. Meade-Waldo says that it is probable that vallestris is synonymous with alienus, Bombus Bhaemorrhoidalis Smith. _ The Museum possesses specimens from several localities in the Eastern and Western Himalayas. Bombus orientalis Smith. To Mr. Paiva’s list of the specimens in the Indian Museum (Rec. Ind. Mus. VIII, p. 80, 1912) I may add Yokohama and? Calcutta. Bombus longiceps Smith. igto. Bombus longiceps, Cockerell, Ann. Mag. Nat. Hist. V, p. 505. 1916. sae as ia Meade- Waldo, Ann. Mag. Nat. Hist. XV\I, p. 468. The Museum possesses a worn specimen which is, I think, the type of this species from Leh in Ladak. It has also been taken by Captain Hingston in Kashmir. I agree with the authors cited that longiceps cannot be a variety of hortorum. ay? ® a = Ree q | Ye yk “tt veh a. éopenm. of a peat Nepat, ws abe | May AGE Bu) 00s ay athe: oon eget pe ‘the y ee at A REVISION OF THE BURMESE UNIONIDAE. By B. PrasHaD, D.Sc., Assistant Superintendent, Zoological Survey of India. (Plate IT.) In spite of various eminent malacologists having paid con- siderable attention to the Burmese Unionidae from early times, our knowledge of these forms was hitherto in a very confused state. Of the earlier authors, references to whose works are etubodied in this paper, Benson Gould, Blanford, Theobald and Nevill deserve special mention ; all of them with the exception of Blanford, who in 1866! tried to summarize all that was known to date, described numbers of species at various times from collections made in various parts of Burma. The most compre- hensive collection of Burmese Unionidae was made by Leonardo Fea in the years 1885-1887 for the Genoa Museum and a detailed paper? on these collections was published by Tapparone-Canefri. In the part dealing with the Unionidae thirteen new species and varieties were described and notes were included on twenty-two of the already known species and varieties. A small part of the collection, however, which was probably received after the report was written, was not included in it. Apparently Tapparone- Canefri had to base his work to a very great extent, if not entirely, on the incomplete published descriptions of the earlier authors and on the illustrations in the Conchologia Indica of Hanley and Theobald, for most of his identifications are incorrect, this would not have happened if he had had authentically named material for comparison. He referred all his new species to the composite genus Unio, and gave elaborate descriptions but did not publish any figures ; his work, therefore, has been a great stumbling block in the way of all later work. Simpson ® tried to remedy this by an examination of the named duplicates of some of these species which the United States National Museum had received by ex- change, but did not succeed in many cases owing probably to the small amount of material available. Haas‘ also has tried to deal with some of the species, but the results of his work on the pre- dominent Indo-Burmese genera have not been published as yet. | Blanford, Fourn. As. Soc. Bengal, XX XV, pt. i, pp. 134-155 (1866). 8 Tapparone-Canefri, Ann. Mus. Civ. Stor. Nat.. Genova, XXVII, pp. 339- 355 (1889). 8 Simpson, Desc. Cat. Naiades (Detroit, 1914). 4 Haas, in Martini and Chemn. Conch.-Cab. Unio (in the course of publi- cation). 92 Records of the Indian Museum. [VoL XXTV The author whose work is most open to criticism, however, is Preston, who in two of his works! dealt with the Indo-Burmese Unionidae. He had for the basis of these works the entire collec- tion belonging to the Indian Museum, which besides being very rich in specimens of various species, is specially valuable because of the many type-specimens or of specimens from type-localities, in many cases named or seen by the authors of the species. Another feature of the collection is the existence of labels in the hand-writings of the various specialists, of whom Blanford, Theobald and Nevill deserve special mention. Nevill in particular had rearranged the whole collection and given provisional names to species and varieties which he considered as new. Preston without any further work accepted Nevill’s identifications and under his manuscript names described these species or varieties as new. He did not even attempt to sort out the specimens of different species where Nevill had left large series mixed up, but labelled all the specimens in one lot according to Nevill’s label which he found with it. In attempting to revise Preston’s work I found that it was guite impossible adequately to work out the Burmese forms without an examination of Tapparone-Canefri’s type-specimens, and I applied to Dr. R. Gestro of the Genoa Museum. He was not only kind enough sent me the whole of Fea’s Burmese collection on loan, but also generously presented to the Indian Museum specimens of a number of the species, duplicates of which were still available. This kindness on Dr. Gestro’s part, for which I am greatly indebted to him, has made it possi- ble for me to assign T.-Canefri’s species to their proper generic and specific position. I have besides carefully gone through the large collections of Indo-Burmese Unionidae already in the Indian Museum. The results of the work may be briefly summarized here. Most of the forms described by T.-Canefri and Preston were found to be referrable to already known species and I have not-come across any new forms. Notes are given on the generic position, relationships, structure and geographical distribution of the twenty-six species and varieties (excluding M. woodthorpi, Godwin- Austen) which I am now able to recognize as being endemic in Burma. They belong to the following genera, Margaritanopsis, Haas ; Indonaia, Prashad; Oxynaia, Haas; Physumio, Simpson; Pseudodon, Gould ; Tvigonodon, Conrad ; Indopseudodon, Prashad ; Parreyssia, Conrad; Lamellidens, Simpson and Tvrapezoideus, Simpson. Genus Margaritanopsis Haas. 1913. JMJargaritanopsis, Haas, Nachr. Deutsch. Malakozool. Ges. V, Pp. 33- 1913. Margaritanopsis, Haas, in-Martini and Chemnitz Conch. Cab. Unio, p. 121. 1914. Margaritana (in part) Simpson, Descr. Cat. Naiades, p. 511. | Preston, Rec. Ind. Mus. VII, pp. 279-308, pl. viii (1912) and Faun. Brit. Ind. Freshw.-Moll. pp. 134-195 (1915). 1922 ] B. PrRASHAD: Burmese Unionidac. 93 Haas erected this genus for Unio laosensis Lea in 1913, but Simpson considers that the species is an undoubted Margarttana and that the new genus is not justified. The genus, however, appears to be well characterized and I agree with Haas in separ- ating M. Jaosensis, with its peculiar distribution in Cambodia, Siam and Burma, from the other species of the genus Margart- tana. Godwin-Austen has recently described another species from the Shan States under the name AZ. woodthorpi', but of this I have seen no specimens. Margaritanopsis laosensis (Lea). PIU, figs. 1-4. 1863. Lnio laosenis, ea, Proc. Acad. Nat. Sct. Philadelphia VII, 1913. Uctaurabngsis laosensis, Haas, op. cits, Pp. 33: 1913. Margaritanopsis laosensis, Haas, op. ctt., pp. 122, 123, pl. vil, figs. I, 2. 1914. Margaritanopsis laosensis, Simpson, op. cit., pp. 520, 521. My reasons for agreeing with Haas in keeping this species in his new genus Margaritanopsis are based on an examination of four specimens collected by Fea in the Karin Hills, Burma, at an altitude of rooo-1z200 feet and labelled Unio sella T. Canefri, a manuscript name only as the species was never described as such. These specimens, as was rightly considered by Haas, are referrable to this species and are of special interest because they beautifully illustrate the changes that take place in the structure of the hinge during the growth of the young into the adult shell, changes which appear to be characteristic of the genus. The young shells are somewhat rhomboidal and only show a beginning of the arcuate outline of the ventral margin of the adult shells. They are thin and not at all solid. ‘The pseudo- cardinals in the right valves of the young shells are lamellar, thin, and lie one above the other; in the adult shell the upper or anterior becomes very thick, somewhat knob-like and lies just next to the scar of the anterior adductor muscle, the lower (or now the posterior) comes to be more or less in line with the anterior and is separated from it by a fairly deep groove, it now takes the form of an elongated ridge with its anterior edge raised into a trigonal tooth-like structure. In the left valve there is a single lamellar pseudocardinal in the young shells, but in the adult it becomes very thick and divided into two patts—an anterior smaller and somewhat trigonal and a posterior much larger and conical, for interlocking with the teeth of the other valve. I have nothing further to add to Lea’s original description of the species and to Haas’ elaborate notes on it. The species described as Unio rectangularis by Tapparone, Canefri (loc. cit., pp. 354, 355) is based on a single very young- shell. It is undoubtedly to be referred to the genus Margaritanop- ! Godwin-Austen, Rec. Jnd. Mus. XVI, pp. 202-204, pl. xv. (1919). 94 Records of the Indian Museum. [VoL. XXIV, sis and probably represents another species of the genus. Owing, however, to a single young shell being available I do not feel disposed to consider it as a distinct species but a figure of the unique specimen (pl. II, fig. 5) is published for future reference. Genus Indonaia Prashad. 1918. /ndonata, Prashad, Rec. Ind. Mus. XV, pp. 148-148, fig. 2. 1921. Jndonaia, id., 16., XXII, p. 602. _ Six species of this genus are known to occur in Burma. Of these I. caerulea has a wide distribution throughout India and Burma, I. bonneaudi and I. pachysoma occur in Assam and Burma, I. crispisulcata and I. chaudhuri are only known from Burma, while I. crispata has a wide range in Burma, Siam and Cambodia. Indonaia caerulea (Lea). 1889. Unio leioma, Vapparone-Canefri, op. cit., p..344. 1914. Nodularia caerulea, Simpson, op. cii., pp. 978-980. 1915. Nodulavia caeruleus, Preston, op. cit., pp. 136, 137. As a result of my examination of the large series of specimens of this species in the Indian Museum, I am able to confirm Simpson’s conclusion that Unio gerbidoni Eydoux, Unio humilis Lea, Unio corrianus Kister, Unto leioma Benson, Umo pilatus Lea, Unio evitatus Lea, Umo trirosiris Sowerby and Unio ander- sontanus Nevill (part only) are synonyms of this species. This is the commonest species of the genus throughout India and Burma and it is represented by a large series of specimens in the Indian Museum. Indonaia bonneaudi (EKydoux). 1889. Unio Bonneaudi, Tapparone-Canefri, op. czt., p. 343. 1914. Nodularia bonneaud:, Simpson, op. cit , pp. 988, 989. git. Nodularia bonneaudt, Preston, op. cit., pp. 140, 141. I have not seen the specimens referred to this species by Tapparone-Canefri, but have no doubt as to his identification. The species is widely distributed in Assam and Burma and is represented by a large series of shells in the Indian Museum. The specimens show great variation both as regards shape and colour. Normally they are oval or ovate but some are distinctly rostrate posteriorly ; in colour they vary from yellowish green to dull brown or even black, Indonaia chaudhurii (Preston). 1912. Nodularia chaudhurii, Preston, Rec. Ind. Mus. VII, p. 290. 1914. Nodularia chaudhurii, Simpson, op. cit., p. 988. 1915. Nodularia chaudhurii, Preston, op. cit., p. 140, fig. 7 (1, 2). I am not quite certain as to the validity of this species. The only specimens I have seen are the type-series of Preston. They come very near I. bonneaudi, but the shells are shorter, more 1922.] B. PrasHap: Burmese Uniontdae. 95. ovate, less inflated and have the sculpture more pronounced. For the present I propose considering this species as distinct, but believe that it will only turn out to be a form of I. bonneaudt when more material is collected. Indonaia pachysoma (Benson). 1914. Nodulayia pachysoma, Simpson, op. cit., p. 987. 1915. Nodularia pachysoma, Preston, op. cit., pp. 139, 140. I. pachysoma is nearly related to I. bonneaudi and I. caerulea. From the former it is distinguished by its more elongate, more inflated, but less deep shells, more pronounced umbones and much stronger hinge, while from the latter it differs in having much brighter and more inflated shells and in the entire absence of the radial sculpture on the sides. The species has practically the same distribution as I. bonne- audi and is represented in the Indian Museum by a large series of specimens from the Brahmaputra River, Assam, and the Irra- wadi River, Burma. Indonaia crispata (Gould). 1914. Nodularia crispata, Simpson, op. cit., pp. 994, 995. 1915. Nodelaria crispata, Preston, op. ctt., p. 142. Gould’s original description is very short but Simpson has recently given an elaborate description. It is a very character- istic form and is easily distinguished from all other Burmese species of the genus by its sculpture, which consists of green zigzag radial lines interspersed here and there with thicker nodules on a yellowish to brownish ground ; the ridges run transversely in the anterior region and vertically in the posterior part of the shell. In the Indian Museum collection the species is represented by specimens from Bhamo (Burma), Siam and Cambodia. Indonaia crispisulcata (Benson). 1914. Nodulavia crispisulcata, Simpson, op. cit., p. 1017. 1915. Nodularia (Radiatula) crispisulcata, Preston, of. cit., pp. 146, 147. Simpson in 1900 separated this species along with his N. lima to form a new section. Radtatula, of the genus Nodularia; but as I have recently '! shown there is no justification for separating I. lima from species like J. caerulea and I. bonneaudt. Nothing is known about the anatomy of J. crisfisulcata and I do not consider the shell characters alone as being sufficient for the separation of this species into a distinct section. The species, as represented by a large series of shells from Bongong River, Burma, in the Indian Museum, is remarkably constant in the sculpture of the shell. ! Prashad, Rec. Ind. Mus. XXII, p. 604 (1921). 96 Records of the Indian Museum. [Vor. XXIV, Genus Oxynaia Haas. 1913. Oxynata, Haas, op. cit., p. 34. 1913. Oxynaia, Haas, op. cit., p. 152. 1914. Nodularia (in part), Simpson, of. ert., p. 115. Haas established this genus for the species N. jourdyi, N. diespiter, N. micheloti and N. pugio of Simpson’s composite genus Nodularia. Of these I have: only seen specimens of Oxynaia pugio, but the descriptions of the other species and my examination of the specimens of O. pugio justifies Haas’ separa- tion of these species into a distinct genus. Oxynaia pugio (Benson). 1862. Unio pugio, Benson, Ann. Mag. Nat. Hist. (3) X, p- 193. 1889. Unio pugio, ''apparone-Canefri, op. cit. p. 344. 1913. Oxynaia pugio, Haas, op. cit., pp. 158, 159, pl. xiv, figs. 6, 7. 1914. Nodularia pugio, Simpson, op, cit., p. 990. 1915. Nodularia pugio, Preston, op. cit., p. 141. This species has a strongly marked and angled posterior ridge running to the cuneate posterior margin; the shell region lying internal to the ridge between the two valves is nearly flat, but is divided in some specimens by the line of union of the two valves rising in the middle; both the anterior and posterior margins are very short, the posterior being much the shorter of the two and distinctly cuneate owing to the ventral margin sharply rising up to meet the point of union of the posterior ridge; the beaks are elevated but not very full. The hinge is characteristic in that the pseudocardinals in the right valve are double, but the anterior is reduced to a thin, lamellar structure only, while the posterior is thickened into a triangular, conical and more or less canine-shaped tooth ; in the left valve also there are two pseudocardinals placed in line with one another, the anterior is small and somewhat coni- cal, the posterior is elongate, ridge-like or triangular and the two are separated from one another by a fairly deep concavity in which the tooth of the corresponding valve fits. Nothing is known about the anatomy of any of the species of the genus Oxynaza. In the Indian Museum the species is represented by a large series of shells from Tenasserim, Pegu, Sawaddy River and from Myadongin Burma. A single specimen from Arrakan appears to belong to a distinct variety, but with this scanty material I do not feel justified in describing it as such. Genus Physunio Simpson. 1918. Physwnio, Annandale, Rec. Ind. Mus. XIV, p. 138. In the paper cited above Annandale described two interesting species of this genus from the Inlé Basin. The soft-parts of these were described by Ghosh! and further notes on the anatomy were 1 Ghosh, Rec. Ind. Mus. XV, pp. 109-122, pl. xvi (1918). . 1922.] B. PRASHAD: Burmese Unionidae. 97 added by me! later. I have nothing further to add regarding these two species (P. micropteroides Annandale and P. ferrugineus Annandale). Genus Pseudodon Gould. 1844. Pseudodon, Gould, Proc. Boston Soc. Nat. Hist. 1, p. 161. 1853. hil Conrad, Proc. Acad. Nat. Sci. Philadelphia, V1, p. 209. 1914. Binasidon (in part), Simpson, op. cit., p. 1079. 1915. Pseudodon s. s. (in part), Preston, op. crt., p. 152. 1919. Monodontina, Prashad, Rec. Ind. Mus. XVI, pp. 403-408. 1920. Pseudodon (subgen, Monodontina), Haas, op cit., p. 318. I am afraid I am responsible for introducing some confusion in the already confused state of affairs regarding this genus. In the paper cited above I revived the generic name Monodontina for species like P. vondembuschiana, since the animal of P. chapert, which I consider as one of the varieties of this species, was very different from that of P. salwentanus (wrongly spelt salventanus) described by me in a previous paper.” in the genus Monodontina I also included the species P. inoscularis asa variety of P. vondembus- chianus, having through oversight considered P. salwenianus, in- stead of P. inoscularis, as the type of the genus Pseudodon. Since the genus Monodontina, with P. vondembuschiana as its type-species, is Synonymous with Pseudodon with P. inoscularis as its type, the former name must give way to the latter, it having been described about nine years after Pseudodon. ‘The genus Pseudodon as now restricted will include the species or varieties orbicularis, cambod- jensts, ovalis, ellipticus, zollingeri, vondembuschianus, chapert, ponderosus and inoscularts. The specimen which I doubtfully assigned to cumingit (loc. cit., p. 408) is not the true cumingit and cannot be included here. Pseudodon vondembuschiana var. inoscularis (Gould). 1919. Monodontina vondembuschiana var. inoscularis, Prashad, op. cit., p. 408. 1921. Pseudodon (Pseudodon) inoscularis, Haas, op. cit., p. 341, pl. xlii, fig. 7. In the paper cited above I have given reasons for considering this species as a variety of Lea’s vondembuschiana, but as I have stated above I made a mistake in adopting the generic name Monodontina. In the Indian Museum collection this variety is represented by two specimens from Tenasserim. Genus Trigonodon Conrad. 1865. Tvrigoncdon, Conrad, Amer. Fourn. Conch. |, p. 233 In view of the differences in hinge and other shell characters of the species that now have to be assigned to the genus Pseudodon, 1 Prashad, Rec. Ind. Mus. XIV, pp. 183-185, pl. xxii (1918) and XVI, p. 294, fig. 5 (1919). 2 Prashad, Rec. Ind. Mus XVI, p. 205, fig. 6 (1919). 98 Records of the Indian Museum. [VoL. XXIV, the species pegwensis with its two varieties must now be separated from it. The arrangement, however, is only provisional till the soft parts of these forms are investigated. Trigonodon peguensis (Anthony). 1900. Pseudodon crebristriatus var. peguensis, Simpson, op. cit. p., 835. 1QI4. Penueden peguensis, Simpson, of. cit., pp. 1083, 1084. 1915. Pseudodon peguensis, Preston, op. cit., p. 150. As stated in the notes on the genus above, I have been obliged to revive Conrad’s generic name Trigonodon for this species and its varieties. The type-species of the genus is Monocondylaea crebri- striatus Anthony, which I think is no more than a variety of T. pegu- ensis. Simpson in his first work treated this species as a variety of M. crebristriatus, but in his Descriptive Catalogue he was doubtful whether the two were distinct. His first course was not correct since, if the two forms are varieties of the same species, the name of the species should be T. eguwensis, this being the first of the two species described by Anthony. As a result of my examination of a fair series of specimens of this species and of the form cvebristriatus from Pegu I am unable to consider the two as distinct species, The latter, how- ever, owing to the shells being more compressed and the sculpture more strongly marked, with the umbones a little more inflated, may be regarded as a distinct variety. var. crebristriatus (Anthony). 1914. Pseudodon crebristriatys, Simpson, op. cit., pp, 1082, 1083. 1915. Pseudodon crebristriatus, Preston, op. cit., pp. 150, 151. There are only two specimens of this form in the Indian Muse- um collection, from Pegu, the type-locality. They resemble the original description very closely and only differ from typical 7. peguensis in the points already noted. var. curvata (Preston). 1915. ee ae lal var. cuvvata, Preston, op. cit., p. 152, fig. 9 (I, 2, 3). This form, of which I have seen a large series from Pegu, differs from the forma typica and the var. crebristriatus in having a less ovate shape, distinctly curved ventral margin, hardly projecting umbones and in having only very faint sculpture on the posterior wing. Genus Indopseudodon, nov. I have very reluctantly adopted the course of introducing a new generic name for the species P. sa/wenianus and P. ava, as the anatomy of the related forms is not known and as so many new 1922 ] B. PrasHap: Burmese Untonidae, 99 subgeneric names have recently been introduced by Haas. Prob- ably my new name may have to be dropped when the exact generic positions of the various subgenera of Haas can be decided by examination of the animals of these species. The soft parts of this genus were described by me in 1919! as those of Psewdodon, s.s., based on an examination of the animal of P. salwenianus. Indopseudodon salwenianus (Gould). 1844. Anodon salwenianus, Gould, op cit., p. 160. 1914. Pseudodon salwenianus, Simpson, op. cit., pp. 1093, 1094. 1915. Pseudodon salwenianus, Preston, op. cit., p. 152. 1919. Pseudodon salwenianus, Prashad, op. cit., pp. 295, 296, fiz. 6 (animal). 1920. Pseudodon salwenianus, Haas, of. cit., pp. 341, 342, pl. xliti, fig. 4. I. salwenianus, as Simpson pointed out in his description of the species, is distinguished from the allied species by its consider- able length and by the strong plicated sculpture on the posterior wing. ) In the Indian Museum collection it is represented by a fair series of specimens from the Tenasserim River and a shell with the label “ Burma ,’’ exact locality not stated. No specimens of this species were obtained by Fea. Indopseudodon ava (Theobald). 1873. Afonocondylaea avae, Theobald, Fourn. As. Soc. Bengal, X1.11, pt. ii, p- 209, pl. xvii, fig. 15. 1900. Pseudodon ava, Simpson, op. cit., p. 839. 1914. Pseudodon ava, Simpson, op, cit., p. 1098. 1915. Pseudodon ava, Preston, op. cit. pp. 153, 154- 1920. Psendodon avae, Haas, op. cit., p. 343, pl. xlili, figs. 5, 6. Simpson in his first work included this species in his section Binereus of the genus Pseudodon, but in his recent ‘ Catalogue’ was doubtful as to its exact position though he still retained it in this section. Haas, however, from an examination of an authentic Burmese specimen was able to assign the species to its exact position near J. salwenianus. I have before me one of Theo- bald’s specimens from Mandalay and can confirm Haas’ conclu- sions. ‘Theobald’s comparison of this species with cumingi and tnoscularis in the remarks at the end of his description is rather unfortunate as the species is not related to either of them. Genus Parreyssia Conrad, 1914. Parveysta, Simpson, op. cit., pp. 1103, 1104. 1919. Parreysia, Prashad, op. cit., p. 292, fig. 3. Eight species of this genus are now known from Burma. Of these only P. smaragdites occurs in Assam as well, all the others ! Prashad, Rec. Ind. Mus. XVI, pp. 295, 296, fig. 6 (1919). 100 Records of the Indian Museum. [Von. XXIV, being confined to Burma. Most of these Burinese species, though they show near relationships with the other Indian species of the genus, form a definite group among themselves. Parreyssia bham oensis (Theobald). 1873. Unio bhamoensis, Theobald, Fourn. As. Soc. Bengal XI.11, pp: 207, 208, pl. xvii, fig. 1. : 1876. Unio bhamoensis, Hanley and Theobald, Conch. Ind. p. 62, pl. cly, fig. 2. 1878. Unio bhamoensis (in part), Nevill, in Anderson’s Zool. Res. Yunnan Exped, p. goo. : 1890. Unio bhamoensis, Paetel, Conch Sam. Il, p. 146. 1899. Unio bhamoensis, von Martens, Arch. Naturgesch. LXV, pp. 38, 39, pl. v, figs. 2, 4. 1900. Parveysia bhamoensis (in part), Simpson, Proc. U.S. Nat. Mus. XXII, p. 483. IOI 4. Parveysia bhamoensis, Simpson, Descr. Cat. Naiades, pp. 1111, T1112. 1915. Parreysia (Parreysia) bhamoensis, Preston, Faun. Brit. Ind. Freshw. Moll. p. 163. The type-specimen of this species from Bhamo, with the tabel “ U. Bhamoensis n. sp.” written in Theobald’s hand, is pre- served in the Indian Museum collection. The species was stated to be arare one, and Theobald considered it and U. mandelayen- sis, the species descriked next to it, to form ‘‘a natural little subgroup of osculent species,’’ which, however, he did not feel “ justified in separating from the great Indian corrugatus group.” Nevill, while working out the Yunnan collections, did not agree with Theobald’s conclusions and united the two species U. bha- moensis and U. mandelayensis under the former name. ‘Tappa- rone-Canefri,! in his paper on the Burmese molluscs collected by Fea, agreed with Nevill in his interpretation of Theobald’s two forms, but wrongly selected the name U. mandelayensis for the snecies. I have examined one of Tapparone-Canefri’s specimens and find that it is a true mandelayensis. Von Martens, who published good figures of this species, considered the species U. bhamoensis as distinct from U. mandelayensis. Simpson in his first work united the two species under the name P. bhamoensis, and in this was followed by Preston ; in his later work, however, having examined more specimens, he rightly regarded the two species as distinct. I have examined the types of the two species besides a large series of specimens in the Indian Museum and find the following differences between the two species:—(i) ‘The shell of P. bha- moensts is only sub-triangular as opposed to the distinctly tri- angular shell of P. mandelayensis, (ii) the beaks in P. bhamoensis are high but not placed well forwards, (ili) in young shells of P. bhamoensis the beaks and the umbonal region have only a faintly marked zigzag radial sculpture which extends over the posterior wing and a little on the anterior side, but no tubercles t Tapparone-Canefri, Ann. Mus. Civ. Stor. Nat. Genova, XXVII, p. 342 (1889). 1922. | B. PrasHap: Burmese Unionidae. tor are ever developed; in P. mandelayensis on the other hand the tubercles are always distinctly developed and the sculpture is much coarser, and (iv) the hinge of P. bhamoensis has lamellar pseudocardiuals which are not very thick, not at like tooth-like and only slightly ragged. In the Indian Museum collection the species is represented by the type-specimen from Bhamo and a fair series of specimens of all ages from Sagaing, Zayleyman and ‘l'avoy in Burma. Parreyssia mandelayensis (Theobald). 1873. Unio mandelayensis, Theobald, op. cit., p. 208, pl. xvii, fig. 2. 1876. Unio mandelayanus, Hanley and Theobald, of. cit., p. 62, pl. cliv, fig. 4. 1878. Unio bhamoensts (in part), Nevill, op. crt., p. goo. 1889. Unio mandelayensis, Tapparone-Canefri, op. cit., p. 342. 1890. Unto mandelayensts, Pactel, op cit., ». 158. 1899. Unio mandelayensis, von Martens, of. c/t., p. 38. 1g00. Parreysia bhamoensts (in part), Simpson, op. cit., p. 843. 1914. Parreysia mandelayensts, Simpson, op. cit., pp. 1112, I113. 1c15. Parreysia (Parreysia) bhamoensis (in part), Preston, op. ctt., pp. 163, 164. The question of the validity of this species as distinct from P. bhamoensts has been discussed already in the account of the latter species, and I would only note here the distinguishing features of the species. The shell is triangular with a very high and forwardly placed beak, the beak and the umbonal region are both very strongly sculptured and often have distinct tubercles or even spines developed in this region. The hinge is very strong, with compact, thick and distinctly tooth-like pseudocardinals. I have examined a specimen from Theobald’s collection, which is the one figured by him in the paper cited above and is probably the type of the species. Besides I have examined one of the specimens named by Tapparone-Canefri, and other speci- mens in the Indian Museum collection from Bhamo, Sheinmagah, Maydong and Pegu in Burma. Parreyssia houngdaranicus (Tapparone-Canefri). BhEL, figs: 1889. Unio houngdaranicus, Vapparone-Canefri, op. cit., p. 341. 1900. Parveysia tavoyensis var. triembolus (in part), Simpson, of. cit., p- 344. ; 1914. Parveysia tavoyensis var. triembolus (in part), Simpson, op. cit., pp- 1115, 1116. 1915. Parveysia tavoyensis var. triembolus (in part), Preston, of. cit., p- 167. Simpson, on the basis of a specimen labelled U. houngdarani- cus from Fea’s collection in the U.S. National Museum, placed U. houngdaranicus in the synonymy of what he called Parreysia tavoyensis var. triembolus, but from his remarks it appears that he was not quite certain as to the correctness of his conclusions. 102 Records of the Indian Museum. [Vor. XXIV, As a result of a careful examination of Tapperone-Canefri’s type-specimen of the species and the forms with which Simpson in- cluded it, I am of opinion that Simpson’s conclusions are quite untenable. Not only is the species quite distinct from Benson’s Unio triembolus, but it also has no relationship whatsoever with Gould’s Unio tavoyensis. It is on the other hand to be grouped with species like U. bhamoensis and U. mandelayensis, forms in which the anterior margin is greatly shortened, the beak placed far forwards and the posterior side drawn out into a cuneate or elliptical lobe. The species may be redescribed as follows :— Shell subrhomboidal to subovate, moderately inflated , subsolid, inequilateral; beaks high and full, very forwardly placed and recurved outwards and downwards, with a fairly deep cavity, sculptured irregularly with low zigzag transverse bars extending over a little more than the depth of the shell; posterior ridge only feebly marked; dorsal margin slightly arched, somewhat truncate ; anterior margin very short, rapidly curving inwards between the umbones in the lunule region and regularly curving below over the podium to meet the nearly straight or slightly arcuate ventral margin; posterior margin longer than the anterior, sharply trun- cate and rather slanting; epidermis dark brown to black, some- what shining; ligament prominent, of an amber to chocolate brown colour; hinge-teeth moderately strong; pseudocardinals slightly ragged, three in the right valve, of which the middle is the largest, and three in the left valve, of which the posteriormost is the best developed; laterals slightly arched, single in the right and two in the left valve; anterior muscle scar deeply impressed, posterior quite shallow; nacre shining white in the umbonal region but with a light bluish tinge below. The type-series was collected by Fea in the Houngdaran River, Meetan, Tenasserim, ower Burma. Parreyssia smaragdites (Benson). 1862. Unio smaragdites, Benson, Ann. Mag. Nat. Hist., (3) X, p. 190. 1866. Unio smavagdites, Blanford, Fourn. As. Soc. Bengal XXXV, p. 147. 1876. Unio smaragdites, Hanley and ‘Theobald, op. cit., p. 5, pl. x, fig. ee 1877. Unio andersoniana (in part), Nevill, Fourn. As. Soc. Bengal, XLVI, p. 40. 1878. Unio andersoniana (in part), Nevill, op. cit., pp. 901, 902, pl. Ixxx, figs. 9, Qa, 9b. 1889. Unio smavagdites, Vapparone-Canelvi. op. cit., p. 3-43. 1890. Unio smaragdites, Paetel, op, cit., p. 107. 1899. Unio smaragdites, von Martens, op. cit., p. 39. 1900. Parveysia smaragdites, Simpson, op. cit., p. 843. 1914. Parreysia favidens (in part), Simpson, op. cit., pp. 1109, 1110. 1915. Parreysia (Parreysia) smaragdites, Preston, op. cit., p. 163. Simpson recently regarded P. smaragdites as only a synonym of P. favidens, but the former species, as is clear from the large series of specimens in the Indian Museum, is quite distinct from the latter. Nevill’s large series of Unio andersoniana from Burma 1922.] B. PRASHAD: Burmese Unionidae. 103 mostly consists of this species, the remainder being young shells of Indonaia caerulea. P. smaragdites, as was noted by Benson, is characterized by the shells being of a beautiful green colour interspersed with lemon-yellow in the middle, the beaks being submedian and greatly deflected forwards, with deep cavities and a well-marked lunule. Benson ’s specimens were taken in the Berhampooter (Brahma- putra) River, Assam, but the species is now known to have a wide range in Burma and Assam. Parreyssia burmanus (Blanford). 1869. Unio burmanus, Blanford, Proc. Zool. Soc. London, p. 449. 1875. Unio vulcanss, Hanley, Proc. Zool. Soc. London, ». 600. 1876. Unio burmanes and Unio vulcanus, Hanley and Theobald, op. cit. p. 19, pl. xlit, fig. 7 and p. 62, pl. ely, fig. 3. 1878. Unio burmanus, Nevill, op. cit., p. 9oo. 1879. Unio burmainus, Vapparone-Canefri, op. cit., p. 343. 1890. Unio burmanus and Unto vulcanus, Paetel, op. cit., pp. 146 and 172. 21899. Unio burmanus, von Martens, of. cit., p. 38, pl. v, fig. 5. 1900. Parreysia burmanus and P. ziulcanus, Simpson, of. cit., p. 845 and p. 844. 1912. Parreysta pernodulosa, Preston, Rec. Ind. Mus. VXI, p. 300. 1914. Parreysta burmanus, Simpson, op. cit., p. 1120. 1915. Parveysia burmanus, P. pernodulosa, and P. vulcanus, Preston, op. cit., pp. 170, 164, 168. The only specimen of this species which I have seen from Fea’s collection is a half-grown individual. It is decidedly longer in proportion to the height and is abnormal so far as the sculpture is concerned. The nodular sculpture which is a characteristic of the umbones of the young and half-grown shells of this species is quite obsolete and the radial sculpture over the rest of the beak is also feebly developed. The specimens figured by von Martens (Joc. cit.) are, in my opinion, not referrable to this species and I have therefore included a reference to his notes on this species with a reservation only. Hanley’s Unio vulcanus, which was described from a single speci- men and later figured by Hanley and Theobald in the Conchologia Indica, is undoubtedly based on a young specimen of this species. Some of the half-grown shells from Bhamo in the Indian Museum collection answer to Hanley’s description and are quite like the figure of the type-shell in the Conchologia Indica. According to von Martens (loc. cit., p. 38), however, the young shells of Unio tavoyensis resemble the figure of the type of U. vulcanus. Preston’s P. pernodulosa is based on very young shells of this species. The types of this species along with a large series of speci- mens from the type-locality, the Irrawadi River near Bhamo, are preserved in the Indian Museum. The types of Preston’s P. pernodulosa were collected by Dr. Anderson at Zaleyman in Upper Burma; Fea’s specimens were taken at Teinzo in the Mule Stream, north-east of Bhamo. 104 Records of the Indian Museum. [VoL. XXIV, Parreyssia tavoyensis (Gould). 1843. Unio tavoyensis, Gould, Proc. Boston Soc. Nat. Aist. 1, pp. 140, I4I. 1856. Unio tavoyensis, Ktister, in Martini and Chemnitz, Conch.-Cab., Unio, p. 1606, pl, xlviii, fig. 2. 1862. Unio tavoyensis, Gould, Otia Conch. p. 190. 1864. Unio tavoyensis, Reeve, Conch. Icon. XVI, pl. xiil, fig. 49. 1866. Unio tavoyensis, Blanford, Fourn. As. Soc. Bengal XXXV, p. 148. ; 1868. Unio parma, Benson, Sowerby in Conch. Icon. XVI, pl. xclviii, fig. 514, 1870. Margaron (Unio) tavoyensis, Lea, Synonyms, p. 31. 1876. Unio payma and U. tavoyensis, Hanley and Theobald, op. cit., p- 61, pl. cliv, fig. 1 and p. 62, pl cliv, figs. 6, 7. 1889. Unio parma, Tapparone-Canefri, 9p. cit., p. 239. 1890. Unio parma, U. savoyensis and U. tavoyensss, Pactel, op. cit., pp-164, 166, 169. 1899. Unio tavoyensis, von Martens, op. cit., pp. 37, 38. 1900. Parveysia tavoyensis, Simpson, op. cit., p. 843. 1914. Parreysia tavoyensis, Simpson, op. cit., pp. 1114, 1115. 1915. Parreysia tavoyensis, Preston, op. cit., pp. 167, 168. Unio parma Benson, was doubtfully included by Simpson and Preston in the synonymy of this species; having before me, however, one of Benson’s original specimens, probably a cotype of the species, I am now able to confirm Simpson’s conclusions. Tapparone-Canefri’s specimen from Bhamo, referred to in his paper cited above as U. parma and another from Tenasserim labelled Unio sp. also belong to this species. Simpson considered Benson’s Unio triembolus as a variety of P. tavoyensis, but an examination of one of Benson’s type-series of specimens shows that U. trtembolus is quite a distinct species. The umbones and a considerable part of the valves in the young shells are covered with a beautiful zigzag sculpture; this however, becomes obsolete with age and hardly a trace of it is left in full-grown individuals. P. tavoyensis is represented in the collection of the Indian Museum by a large series of shells from Pegu, Tenasserim, Tavoy and Arrakan. Parreyssia feddeni (Theobald). 1874. Unio feddeni, Vheobald, Fourn. As. Soc. Bengal XLII, pt. i, p. 208, pl. xvii, fig. 3. 1877. Unio feddeni, Nevill, Fourn. As. Soc. Bengal XI.VI, pt. il, p. ec: 1878. Unio feddeni, Nevill, op. cit., p. goo. 1900. Parreysia feddent, Simpson, p. cit.. p. 165. 1914. Pavreysia feddent, Simpson, p. cit., pp. 1113, T1114. 1915. Parreysia (Parreysia) feddeni, Preston, op. cit., p. 105. This species was described by Theobald from shells collected by Mr. F. Fedden and said to have been obtained from the Peem- gunga River in Centra) India. Later Nevill, when reporting on Dr. Anderson’s Yunnan collections, stated that the species is tol- erably abundant in the rice-fields at Pegu and also at Yaylay- maw in Burma. He also doubted Central India as the proven- ance of this species from the fact that in the ‘‘carefully kept 1922. ] B. Prasuap: Burmese Unionidae. 105 collections of Mr. H. F. Blanford”’ specimens of U. feddeni obtained from Fedden were labelled as from Burma. Since Fedden had col- lected in both localities the probabilities were that Theobald had mixed up the labels of his specimens. The only specimen of this species now in the Indian Museum collection is from Burma and none of the Central Indian specimens in the collection are referrable to this species It is probable, therefore, that Nevill was correct in considering this species as a true Burmese form. Theobald’s description of the shell of this species, except for the inaccuracy in his description of the hinge pointed out by me in a recent paper ', is quite complete and needs no amplification. The species is not represented in Fea’s Burmese collections. Parreyssia feae (‘lapparone-Canefri). Ply ties. 7. Oo: 1889. Unio feae, Vapparone-Canefri, of. cit., p. 340. 1900. Parreysia feae, Simpson, op. cit., p. 844. 1914. Pavreysia feae, Simpson, op. cit., pp. 1116, 1117. 1915. Parreysia (Parreysia) feae, Preston, op. cit., p. 168. This species, which was described from specimens collected at Meetan in the Houngdaran River, Burma, has never been figured and was hitherto known only from the author’s original description and the short notes added recently by Simpson from an examination of some of Fea’s specimens. ‘The following addi- tional notes are based on three specimens one labelled “‘ Type” and the other two “‘ Co-types,’’ which have been presented to the Indian Museum by Dr. R. Gestro of the Genoa Museum. The shells of this species vary in outline. In the young they are subrhomboidal but become more elongate as growth proceeds, The zigzag radial sculpture of the young shells becomes obsolete with age and in fully grown shells is just faintly indicated. The umbones are high, recurved forwards and inwards but not meeting in the middle line ; they are often weathered even in half-grown individuals. The young shells are dirty yellow, interspersed with green in the region with raised zigzag sculpture, older shells are yellowish-brown, while the full-grown type is dark chocolate- brown. The nacre is bluish white. Genus Lammellidens Simpson. 1914. Lamellidens, Simpson, op. cit., p. 1165. 1919. Lamellidens, Prashad, op. cit., p. 293, fig. 4. A large number of specific and varietal names have been given by previous authors to ordinary variations of the commoner Indo-Burmese forms of this genus, and it has been found neces- sary on examination of the large collections of Unionids now available, to drop most of these names. I am now able to recog- nize only six definite species and varieties as occurring within the ! Prashad, Rec. Ind. Mus. X1X, p. 713 (1920). 106 Records of the Indian Museum. [Voy,. XXIV, limits of Burma. Three of these L. generosus, L. lamellatus and L. scutum are confined to Burma, while the other three have a much wider distribution. Lamellidens marginalis (Lamarck). 1876. ? Unio marginalis var. zonata, Hanley and Theobald, op. cit., p. 20, pl. xliv, fig. 2. 1889. Unio marginalis with vars. subflabeilatta, cylindrica (nec Hanley and Theobald) and obesa (nec Hanley and Theobald), U. protensus var. obtusatus, Tapparone-Canefri, op. cit., pp. B15 0310) B50" Ler Whe Vet 1914. Lamellidens marginalis, Simpson, op. cit., pp. 1106-1168. 1915. Lamellidens marginalts, Preston, op. cit., pp. 175, 176. 1921. Lamellidens marginalis, Prashad, Rec. Jud. Mus. X X11, p. 606, fig OA. In the paper cited above I have recently given the distinc- tive characters of the species and have figured the hinge of a typical specimen. Tapparone-Canefri was apparently unaware of the great vatiation in the shape and form of this species and gave specific and varietal names to shells which are quite typical. As a result of my examination of Tapparone-Canefri’s named specimens I find that five of his names, including his true marginalis, must be treated as synonyms. Simpson doubtiully included Unto dolichor- hynchus and U. gianelli in the synonymy of L. marginalis, but the former on examination of the type was found to be an elongate specimen of L. corvrranus and the latter a half-grown L. scutum. ‘The various Burmese forms included by Preston as varieties and subspecies of this species, are discussed in the notes on the several species. L. marginalis has a very wide range of distribution, throughout India, Burma and Ceylon. Lamellidens corrianus (lea). Pl. I, figs. 9-11. 1889, Unio corrianus, U. dolichorhynchus, U. protensus and var. ellipticus, Tapparone-Canefri, op. cit., pp. 347-350. 1900. Lamellidens canefrinus, Simpson, op. cit., p. 857. 1914. Lamellidens canefyinus, Simpson, op. cit., p. 1170. 1915. Lamellidens canefrinus, Preston, op. cit., p. 187. ; 1921. Lamellidens corrianus, Prashad, op. cit., p. 609, fig. 29C. In the paper cited above I have given reasons for considering L. corrianus as a species distinct from L. marginalis. Asa result of my examination of ‘lapparone-Canefri’s types of U. protensus and its var. ellypiicus 1 find, that both of them should be referred to this species, Simpson’s new name canefricus must, therefore, be sunk in the synonymy of L. corrianus. The type specimen of L. dolichorhynchus differs from typical shells of L. corrianus in being a little more elongate and the cuneation of the posterior margin is therefore more pronounced, but these differ- ences in the shape of the shell in the case of a variable species 1922.] B. PrasHap: Burmese Unionidae. 107 such as L. corrianus are not enough to warrant the erection of a distinct variety, much less a separate species. L. corrianus like L. marginalis, is widely distributed through- out India and Burma. Lamellidens jenkinsianus subsp. obesa (Hanley and Theobald). 1920. Lamellidens jenkinsianus subsp. obesa, Prashad, Rec. Jnd. Mus, XIX, pp. 170-172, pl. ix, figs. 1, 2. In the paper cited above I have recently discussed the question of the various forms of L. jenkinsianus. In Fea’s collection from Burma there is a young specimen of the form obesa from Tonghoo. This specimen is one of the few unnamed specimens of Fea’s collection and only had the name ‘ Unio’ on the label. The specimen is from the same locality from which Theobald’s specimens, now in the Indian Muscum, were collected. It may also be noted here that the specimens referred to as Unio marginalis var. obesus by Tapparone-Canefri (/oc. cit., p. 346) are, as has been pointed out already, only typical specimens of L. marginalts. Lamellidens generosus (Gould). Pl. II, figs. 12-17. 1847. Unio generosus, Gould, of. cit., p. 220. 1870. Margavon (Unio) generosus, Lea, Synonyms, p. 20. 1876. Unto generosus and var. angustior, td., ib., p. 22, pl. xlvi, figs. 4; 7. 1876. Unio lamellatus, var. (nec I.ea), Hanley and Theobald, of. cit., p- 5, pl. ix, fig. 6. 1889. Unio marginalis var. zonata (nec Hanley and Theobald),! var. tricolor (nec MKiister), U. pulcher and var. lamellatiformis, U. genervosus and var. delapsus, Tapparone-Canefri, of. cit., PP: 346, 347, 35% 351, 352. 1899. Unio generosus, von Mattens, of. cit., p. 46. 1900. Lamellidens genevosus, Simpson, op. ctt., p. 857- 1912. Lamellidens marginalis subsp. sawaddyensis, Preston, Rec. Ind. Mus. VII, p. 305. 1914. Lamellidens marginalis var. tricolor and subsp. sawaddyensis, L. Burmanus? and L. generosus, Simpson, op. cit., pp. 1168, 116g. 1170, 1175. 1915. Lamellidens marginalis var. tricolor and subsp. sawaddyensis, and ZL. pulcher with var. lamellatiformis, Preston, op. ctt., pp. 176, 177, 185. The above elaborate synonymy is based ona careful examina- tion of the type-specimens of Tapparone-Canefri’s and Preston’s new species and also of authentic specimens of others in the Indian Museum collection. The specimens identified by both Tapparone-Canefri and Preston as belonging to the var. ticolor Kiister are undoubtedly ' Hanley and Theobald, Conch. Jnd. p. 20, pl. xliv, fig. 2. The shell figured is apparently a young specimen of L. marginalts. 2 The second name, Lamellidens Burmanus, on the same page (1170) certainly a dapsus calami for J. thwattest. 108 Records of the Indian Museum. [VoL, XXIV, the young of this species and I have little doubt that Kiister’s types also belonged to it. Simpson has recently described this species very fully and I have nothing more to add to his description beyond noting the changes that take place in the colour of the shell during growth. The young shells are fulvous or chocolate-brown in the umbonal region and the greater part of the shell is bordered by a broad band of deep yellow on the inner side, while the dorsal slope to- © gether with the posterior wing and the rest of the shell are shining green. As the shell grows the green and yellow gradu- ally disappear and the shells as a whole become dark chestnut to black ; the umbonal region, however, is always much lighter. A certain amount of variation is also exhibited by the posterior wing and the posterior margin; in young shells the wing is usually much broader proportionately and more marked, but as the shells increase in size it becomes much narrower ; the posterior margin shows much greater variation, it may be only somewhat narrowed or may even take on a distinct cuneate appearance. In the Indian Museum collection this species is represented by a large series of specimens of all ages from various localities in Burma. Lamellidens lamellatus (Lea). 1838. Unio lamellatus, \.ea, Trans. Amer. Phil. Soc. V1, p. 19, pl. vi, fig. 16. 1889. Unio pulcher var. ponderosulus, Vappaione-Canefri, op. cit., PP. 351, 352. 1914. Lamellidens lamellatus, Simpson, op. cit., pp. 1172, 1173. Both Simpson and Preston have wrongly included Lea’s Unio layardi in the synonymy of L. lamellatus. Lamellidens layardi has no relationship whatsoever with such Burmese species as L. generosus, L. lamellatus and L. scutum, but is closely allied to L. marginalis. L. lamellatus, as has been noted above, is allied to L. genero- sus, but is distinguished by its general shape, thinner shell, less well developed post-dorsal wing and more delicate hinge-teeth. I have examined a large series of this species from various localities in Burma, in the Indian Museum collection. Lamellidens scutum (Sowerby). 1868. Unio scutum, Sowerby, Conch. Icon. XVI, pl. xciv, fig. 510. 1876. Unio scutwm, Hanley and Theobald, Conch. Ind. p. 22, pl. xlvi, fig. 1. 1889. Unio gianelli with var. degener, Tapparone-Canefri, op. cit., - pp. 353) 354. BY 1899. Unio scutum, with var. humzlior, von Martens, op. cit., pp. 45, 46. 1912. Lamellidens marginalis var. sublamellata, Preston, op. cit., p. 305. 1914. Lamellidens scutum, Sinipson, Op. cit., pp. 1173, 1174. 1915. Lamellidens marginalis vars. sonata (nec Hanley and Theo- bald), sublamellata, scutum and humilzor, Preston, op. cit., pp. 177, 181, fig. 19 (1-3). 1922.] B. PRASHAD: Burmese Unionidae. 109 An examination of the types of Unio giannelli, its var. degener, L. marginalis var. sublamellata and shells identified as var. zonata, has shown that they are all referrable to this species. ‘The shells of var. humiltor von Martens also gradually fade into those of the typical form and it is impossible, therefore, to distinguish this variety. L. scutum has a comparatively less broad, less tumid, but more elongate type of shell than that of either iE. gencrosus or L. lamellatus—the other two species of this interesting group. The group, so far as is known at present, is confined to Burma. Genus Trapezoideus Simpson. 1921. Trapezoideus, Prashad, Rec. Ind. Mus. XXII, p. 609. In the paper cited above I have described the anatomy of this interesting genus. In Burma it is represeuted by five species, all of which, with the exception of T. foliaceus, are endemic in Burma, Trapezoideus exolescens (Gould). 1843. Unto exolescens, Gould, op. cit., p. 141. 1852. Margaron (Unio) exolescens, Lea, Synoyms, p. 32. 1862. Unio exolescens, Gould, Otia Conch., p. 191. 1866. Unio Sele Blanford, Fourn. As. Soc. Bengal, XXXV, pt. i, Ps 1876. Wie exlescens Hanley and Theobald, of. cit., p. 43, pl. cvii, BS 1877. Unto fragilis, Nevill, Fourvn. As. Soc. Bengal, XLVI, p. 39. 1878. Unio foliaceus var. ‘fragilis, Nevill, op. cit., p. 400, pl. Ixxx, fig. 8. ; 1889. Unio exolescens, Tapparone-Canefri, op, cit., p. 349. 1899. Unio exolescens, von Martens, of. ctt., p. 42. 1900. Tvapezoideus exolescens, Simpson, op. cit., p. 859. 19!14. Tvapesoideus exolescens, Simpson, op. cit., p. 1185. 1915: TZvapezotdeus exolescens, Preston, op. cit, p. 195. 1920. Tvapezoideus exolescens, Haas, op. cit., p. 272. An examination of Nevill’s type of Unio fragilis has shown that the species should be referred to T. exolescens rather than to var. compius of T. foliaceus as Simpson thought (loc. cit., p. 1182) or to T. misellus as Haas has done. One of the specimens in Fea’s collection is labelled ‘Unio microsomus, T. Canefri, n. sp.,’ but this is not referred to in his paper; the specimen is only a young example of 7. exolescens ; the other shells referred by T. Canefri to T. exolescens are correctly identified. The locality of the type-specimens was not certain, but von Martens’ specimens were obtained at Mandalay and the Indian Museum specimens are from Bhamo. Trapezoideus foliaceus (Gould). 1843. Unto foliacea, Gould, op. cit., p. 141. 1852. Margaron (Unio) foliacea, Wea, op. cit., p. 39. 1862. Unio foliacea, Gould, of. cit., p. 191. 1865. Unio Peguensis, Anthony, Amer. Fourn. Conch. J, p. 351, pl Xxv, fig. 2. ITO Records of the Indian Museum. [VoL. XXIV, 1866. Unio foliaceus and Unio peguensis, op. cit., pp. 148, 154. 1868. Unio peguensis, Reeve, Conch. Icon. XVI, pl. xcv, fig. 519. 1876. Unio foliaceus, Hanley and Theobald, op. cit.. p. 19, pl. xlii, fig. 3. 1889. Unio foliaceus, Tapparone-Canefri, op. ctt., p. 345. 1900. Tvapezoideus foliaceus, Simpson, op. cit., p. 858. 1912. Tvrapezoideus foliaceus, Preston, op. cit., p. 307. 1914. Tvrapesoideus foliaceus, Simpson, op. cit., pp. 1181, 1182. 1915. Trapezoideus foliaceus, Preston, op. cit., p. 193. 1919. Tvapezoideus foliaceus, Haas, op. cit., pp. 261, 262, pl. xxxil, fig. 3. Anthony’s Unio peguensis is synonymous with this species and is not referrable to the genus Psewdodon as Simpson believed. Deshayes and Julien’s Unio comptus is T. misellus, and the type of Unio fragilis, Nevill, is a specimen of T. exolescens (vide p. 109). Preston’s new variety zaleymanensis is not a variety of this spe- cies, but is based on young and half-grown shells of T. misellus. T. foliaceus is represented in the Indian Musuem collection by specimens from Bhamo and Zaleyman, Burma. Trapezoideus dallianus (I*rierson). 1899. Unio foliaceus, von Martens, op. cit., p. 42. 1913. Parreysia dalliana, Frierson, Nautilus, XXVI, p. 142. 1919. Tvapezoideus dallianus, Haas, op. cit., p. 263, pl. xxxil, fig. 4. ) Frierson’s species, as Haas has pointed out, is a Trapezoideus and not a Parvreyssia as the author of it thought. Haas has also, _I think, rightly referred von Martens’ specimens of T. foliaceus to this species. In the Indian Museum-there is a single right valve from Burma, exact locality not stated, which belongs to this species. Trapezoideus misellus (Morelet). 1912. Tvapezoideus foliaceus var. zaleymanensis, Preston, op. cit., De Pe ; 1915. YLvapezoideus foliaceus var. saleymanensis, Preston, op, cit., Dy 1 . 1919. Tyapeeaens misellus, Haas, op. cit., pp. 266-270, pl. xxxii, figs, 6-9, pl. xxxill, figs. 1-5. Haas has given the complete synonymy of this species quite recently. However, he wrongly included in it Nevill’s Unio fragi- lis, the type of which, as I have stated already, is T. misellus, and he did not include in the synonymy Preston’s var. Zaleymanensis of IT. foliaceus, which was described from young and _ half-grown shells of this species. ‘The species is represented in the Indian Museum collection by a fair series of specimens of all ages from Tenasserim, the Irrawadi River, Zaleyman and Bhamo. Trapezoideus subclathratus (v. Martens). 1919. Zvapezoideus subclathrvatus, Haas, op. ctt., pp. 270-272, pl. xxili, fig. 6. 1922.] B. PrAsHAD: Burmese Unionidae. TEE Careful examination of a single specimen of this form taken at Sheinpagali, Burma, leaves no doubt in my mind that, as Haas correctly states, this species is distinct from T. misellus, of which von Martens considered it to be a variety. ~ n e es ESeaSe ae SO | aid, BAN grows advent ee i Teale has, ‘ven ] pn va A. ‘ai ae i, bag ial a ae ae f iy % (ie: \ yy * ise, a LPO Bt bas fig. d BK ae ie ai * a’ - ‘xiviegi =Ois 7 Mag se 9 Thy . 4 “* t ~~ . Y } tl: — ate 4 i | ; Py ‘ _ i j iT ; is 4 c a. boat ae oF ere, ery Pe j rung ry Ly 2 hal -_ a ue alan )'¥s é a} ~~ s dy Fader ihona.sn 1aq 4ai4 Re | ad « q i 4 7 : Vo Redan) AR Mei cides D 7 : . 4 a A ; ; rT | - "= : << _—s - 2 eS Be at aprons hal ae haruion jamweh wie, Deion fe tatcded :tleocl sama 90) mgt! ma, | , | : Rare ae oe % i , -_ wy bag nia’ oY wren hla ir beieieelen em dsoved nye (oe , i 4 aw Ya we agli LOA ORID Ba4 eyiktan [feud raw 17 [ aL y r en ti = pata bon th oc) aac —. Preeniny if mr mi hing wh 7 a Ww EXPLANATION OF PI,ATE II. All figures are direct photographs of dry shells. Margaritanopsis laosensis (Lea). Fics. 1-4.—Shells of various ages, forming the type-series of Unio sella, Tapparone-Canefri: x}. Margaritanopsis sp. Fic. 5.—Type-shell of Unio rectangularis, Tapparone-Canefri : x Parreyssia houngdaramicus (Tapparone-Canefri). Fic. 6.—Type-shell from Houngdaran River, Burma: natural size. Parreyssia feae (Tapparone-Canefri). Fic. 7.—Adult shell from Meetan, Burma: natural size. ,, 8.—Type shell from the same locality: natural size. Lamellidens corrianus (Lea). Fic. 9.—Type-shell of oe protensus, Tapparone-Canefti : Xe: 10.—Type-shell of Unio protensus vat. elli plicus, Tap- parone-Canefri: x4. 11r.—Type-shell of Unio dolichorhynchus, Tapparone- Canefri: x4. ? a) Lamellidens generosus (Gould). Fic. 12,—Young shell of L. marginalis subsp. sawaddyensis , Preston: X#$. 13.—Specimen identified as Unio marginalis var. tricolor by Tapparone-Canefri: x4. 14.—Half-grown specimen of L. marginals subsp. sawaddyensis, Preston: X#. 15.—Type-shell of Unio pulcher, Tapparone-Canefri: XK 5. 16.—A ae identified by Preston as L. marginalis vav. sublamellata: X4. 17.—Type-shell of L. marginalis subsp. sawaddayensis, Preston: x4. If. 7 4 PLATI 1922. XXIV Mus., Von. IND. Rinc. eu wip gy AO Ou gr POU Oud ae pow a pone waodd “HCINOIN] ASAIO Se ‘ojoUd "[NpUOT, 0's q ries Soi : a “Bort . ae Aatht ae : | or , f] pea ‘; RECORDS of the INDIAN MUSEUM (A JOURNAL OF INDIAN ZOOLOGY ) Vol. XXIV, Part II. JUNE, 1922. PAGE Notes on Crustacea Decapoda in the Indian Museum, XV. Pontoniinae. Stanley Kemp x ae a ie He i sprigs 5 we) ee teal can Wig nial WNstitge Foy a Oy As . 4 Calcutta : PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA, PRINTED AT THE BAPTIST MISSION PRESS. 1922. Pr ice Jwe Rupees. 4 as a J -@ fet ete * NOTES ON CRUSTACEA DECAPODA IN THE INDIAN MUSEUM. XV. PONTONIINAE. By STANLEY Kemp, Sc.D., Superintendent, Zoological Survey of India. The Pontoniinae form one of the four subfamilies into which the Caridean family Palaemonidae is divided; the other three are the Palaemoninae, the Desmocaridinae and the Typhlocaridinae. Of the very numerous species known in the family all except three belong to the Palaemoninae and Pontoniinae. The Desmocaridinae comprise only a single species, Desmocaris trispinosus (Aurivillius), found in freshwater streams in West Africa, and Sollaud' who first drew attention to its peculiar characters regards it as the most primitive known Palaemonid. The Typhlocaridinae include two remarkable blind species, both belonging to the genus Typhlocaris Calman,’ which inhabit waters of subterranean origin in Palestine and Cyrenaica. Typhlocaris differs from all other Palaémonidae in the presence of a longitudinal suture in the carapace, resembling that found in certain Penaeidae and in the Thalassinidea. The Palaemoninae and Pontoniinae are closely related sub- families. distinguished from the other two by a number of impot- tant characters.’ They differ from one another in two respects. The pleurobranch found in the Palaemoninae above the base of the third maxilliped is invariably absent in the Pontoniinae, with the result that six large branchiae are found in the former sub- family as against five in the latter. The telson-tip in the Palae- inoninae is usually armed with two pairs of spines and a varying number of plumose setae, whereas in the Pontoniinae there are always three pairs of spines.* This character is not an invaria- ble one. There appears to be no real morphological distinction between spines and setae as found at the apex of the telson, in the Pontoniinae the median spines are frequently plumose and I have seen one species of Palaemoninae® in which there are three pairs of spines, almost precisely as in the related subfamily. ! Sollaud, Comptes rendus Acad. Sci. Paris CUA, p. 913 (4Gtt). 2 Calman, Trans. Linn. Soc. (2) Zool. XI, p. 93 (1909) ; Annandale and Kemp, Fourn. Asiat. Soc. Bengal (n.s.) UX, p. 245 (1913) ; Parisi, At.¢ Soc. /tal. Sci. nat. Milano \.1X, p. 241 (1920). 8 The characters of the four subfamilies are summarized by Borradaile, Tvans. Linn. Soc. (2) Zool. XVII, p. 326 (1917). : + Coutierea is said to possess merely asingle pair, but the genus is only known from one specimen. It may prove not to belong to the Pontoniinae. 5 A remarkable species from South India, allied to Palaemonetes and hitherto undescribed. 114 Records of the Indian Museum. [VorL. XXIV, In working through the large collection of Pontoniinae in the Indian Museum I have derived much assistance from the memoir which Borradaile has recently published '; his full lists of references to the species have been most useful to me. On a large number of taxonomic questions, however, I have formed conclusions which differ very widely from those which he has expressed, particularly in regard to the generic subdivision of the group. ‘The latter question, as Borradaile has pointed out, is one of great difficulty. In the course of my work I have repeatedly been struck by the very homogeneous nature of the subfamily as a whole, and it is to this fact that we must turn for an explanation of the ap- parently trivial characters on which many of the genera have been founded. The characters used for the generic subdivision of the Ponto- niinae contrast very strongly with those employed for the same purpose in certain other families and subfamilies of Caridea. In the Hippolytidae, for example, we find that the genera can be separated by trenchant morphological characters based for the most part on the branchial formula, on the structure of the mandible and on the carpal segmentation of the second peraeopods. We are thus able, in this family, to devise a scheme of classification which should satisfy even the most earnest seeker after phylo- genetic truth ; we have confidence that our genera form natural groups and that they can be arranged in a manner which will de- monstrate their true affinities. The Pontoniinae present a far more difficult problem. We search almost in vain for important morphological features which will serve to separate the large assemblage of species into natural groups. We are obliged to define our genera on characters of a much inferior order of magnitude and we are often far from certain that they are phylogenetically valid. This radical difference between two not distantly related groups of Caridea is perhaps to be explained by supposing that the Pon- toniinae have succeeded in evolving a structural type that can be adapted without any deep-seated modifications to all needful kinds of environment; whereas the Hippolytidae, with a less useful stock-pattern, must needs undergo drastic change, some- times assuming the most bizarre forms, in order to equip themselves for particular conditions of life. In this connection it is to be remarked that the Pontoniinae have proved themselves far superior to the Hippolytidae in their ability to accommodate themselves to unusual surroundings. In subdividing such a homogeneous group as the Pontoniinae it is, I believe, of first importance that the genera should be est- ablished on a broad basis and that the characters used in separat- ing them should so far as possible be unequivocal. That the classifi- cation of the family has hitherto been greatly lacking in this respect is clear from a study of the literature. As evidence of the confu- 1 Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 323 (1917): 1922.| S. Kemp: Notes on Crustacea Decapoda. 115 sion that has prevailed, it may be mentioned that Balss has recently redescribed the type species of Periclimenes as a new form of Urocaris and that a single species has been described by Schenkel, Nobili, Lenz and Miss Rathbun—all writers of experience—under the names Ancylocarts brevicarpalis, Palaemonella aberrans, Harpi- lius lativostris and Periclimenes hermitensis respectively. Borradaile’s recent system of -classification does little to re- move the sources of error. ‘The primary divisions in his synoptic key to the genera depend almost wholly upon habit of body. This character appears to me to possess little generic importance and, inasmuch as the subfamily comprises species with every imaginable gradation of form, between the most slender and the stoutest, it is frequently quite impossible to decide on the section to which any particular form should be allocated. I have attempted in this paper to devise a more workable arrangement. In so doing I have been led to discard Urocaris, Ancylocaris and Periclimenaeus as distinct genera and to merge all the species belonging to them, together with those of Borra- daile’s subgenera Falciger, Cristiger, Corniger and Hamiger under the single name Periclimenes. The large assemblage of species thus constituted is divided into three subgenera, Pertclimenes, Pertcli- menaeus and Ancylocaris, which together comprise the majority of known species of the subfamily. Except for Harpihopsis, which is no doubt identical with Harpilius, the remaining genera retain theit rank; several, however, are inadequately described and one or two may even prove not to belong to the subfamily. Whether the new grouping in the Periclimenes section demon- strates the real affinities of the species better than the old one is a question on which it is difficult to express a decided opinion. It is clear from the manner in which they are combined that many of the characters which are used in the distinction of species must necessarily be convergent in origin and it is impossible to be certain that this is not also the case with some of those to which I have at- tached generic or subgeneric significance. The new grouping, how- ever, removes some of the obvious anomalies that have hitherto existed and will, I believe, be found convenient in practice. In proposing this new scheme of classification it will be understood that I disagree with much that Borradaile has said regarding the phylogeny of the group and that my views on the way in which the different genera have originated differ very widely from those which he has illustrated in the form of a phylogenetic tree. The Pontoniinae are for the most part Indo-Pacific in distri- bution and the subfamily is almost exclusively marine. The only exceptions to the latter statement are Peviclimenes indicus, P. demant and P. obscurus, which frequent lagoons of variable salinity on the eastern side of the Indian Peninsula. The two former species are capable of enduring extreme alterationsin salinity and both have been found in water that is quite fresh as well as in pure sea-water. Peviclimenes obscurus has been found both in the sea and in brackish water. The members of the subfamily 116 Records of the Indian Museum. [VoL. XXIV, occur for the most part in sheltered portions of the littoral zone and are especially abundant in the vicinity of coral-reefs. A small proportion occur in moderate depths, up to 50 fathoms, and a few live in deeper water. The greatest depth from which any Pontoniine is known is 703 fathoms.! The most remarkable feature of the subfamily is the ability its members have shown in forming associations with other animals. In the variety of these associations they excel all other Caridea. Some are found on Sponges, others on Actinians, Alcyonaria and Madrepore corals, a few are to be met with on Asteroids and Echinoids and many live on Crinoids. A considerable number of species occur in the mantle-cavity of Lamellibranch molluscs and some are known from the branchial sac of Ascidians. Many species are, of course, free-living, but the association between a prawn and some other animal can usually be detected only by the collector and unless the facts are carefully noted on the label they are liable to escape notice. I have little doubt that many more species possess these associations than we now realize. As to the nature of the association we are at present very ignorant. The species that live in Lamellibranchs and in Ascidians find a safe retreat from the perils they would meet outside and through the activities of their hosts are, no doubt, well supplied with food. They are commensals in the strict application of the term and, in so far as they deprive their hosts of a portion of their nutriment, may also be regarded as parasites. In the absence of any evidence that their presence is of advantage to the host, they cannot be called symbiotic in the sense in which the word is gener- ally applied. The species that live on the giant sea-anemone, Dzscosoma, are probably protected by their host and those that live on Sponges, Alcyonaria, Madreporaria and Echinoderms doubtless obtain the benefit of shelter. The species on Discosoma perhaps share the food of their host, but it is not unlikely that those on Alcyonaria feed directly on the polyps and are thus true parasites. Dr. Asajiro Oka found two remarkable species of Pontonia when examining the Indian Museum collection of Tunicates and has pointed out that the size of the prawns indicates that they must have entered the Ascidian in the larval state and grown up to maturity in the branchial sac. In a specimen of Polycarpa annandalet Oka, in which the external measurements of the test were 33mm. X 23mm. X 19 mm., a male and female of Pontona anachoreta, sp. nov., were found, the prawns being 6°5 and 10°5 mm. in length. From Asczdia willeyt Oka, with test 35 mm. Xx 20 mm. a pair of Pontonia okai, sp. nov., 8 and 85 mm. in length, was obtained. When it is considered that these Pontoniids are heavily built forms, with one of the chelate legs of the second pair extremely large, it is evident that they could not possibly ! A specimen of Periclimenes laccadivensis collected by the R.I.M.S. ‘ Investigator.’ 1922.] S. Kempe: Notes on Crustacea Decapoda. I17 pass through the small apertures in the test of the Ascidian. They are thus, like Spongicola venusta in Euplectella, perpetual ptisoners. In the course of an extremely interesting note on sex-pheno- mena in Pinnotheres, Orton! has pointed out that female crabs are frequently found alone in a mollusc and that males are scarce. This corresponds with my own observations on this and other gen- era of Pinnotherid crabs in India: single specimens, usually females, are of common occurrence and it is quite exceptional to find two crabs in one mollusc. It is probable, as Orton has pointed out, that the male crabs wander freely and visit the molluscs from time to time in search of feniales. Conditions are different with the Pontoniiids that live in Lamellibranchs, for in practically every instance a male and female prawn are found together in the same mollusc. From this fact itis perhaps legitimate to infer that, as with the species in Ascidians, the prawns after they are once established in their host never leave it throughout the whole course of their existence. The animal associations recorded in the Pontoniinae are the following :— On PORIrERA. ? Periclimenes tmpar, sp. nov. Pontonia tyrrhena (Petagna).? Typton spongicola Costa. On COELENTERATA. On Actiniatia. Periclimenes brevicarpalis (Schenkel), on Discosoma. S tnornatus, sp. nov., on Discosoma. On Madreporaria. Pertclimenes spiniferus de Man. diversipes, sp. nov. Harpilius, probably all species. Coralliocaris, probably all species. On Alcyonaria. Periclimenes tnvestigatoris, sp. nov. - diversipes, sp. nov. Dasycaris symbtiotes, gen. et sp. nov., on Pteroeides Pontontides beaufortensis (Borr.), on a Gorgonian. Balssia gastt (Balss), on Covallium rubrum. On ECHINODERMATA. On Asteroidea. Periclimenes parasiticus Borr., on Linckia. On Echinoidea. Pertclimenes brocki de Man. Stegopontonia commensalis Nobili, on Echinothrix. ! Orton, Nature CVI, p. 533 (1920). 2 Fide Heller. The species usually lives in Pinna and it seems to me a little unlikely that it should also occur on sponges. 118 Records of the Indian Museum. (Vou. XXIV, On Crinoidea. _Palaemonella pottsi (Bort.), on Comanthus. afinis Zehntuer, on Actinometra. “ Palaemonella orientalis Dana,” de Man. Periclhimenes brockettt Bort. ceratophihalmus Borr. cornutus Borr. i commensalis Borr., on Comanthus. Pontoniopsis comanthi Bort., on Comanthus ) x) In Moriusca LAMELLIBRANCHIATA. In Pinna. Anchistus inermis (Miers). >, mersi (de Man). Pontoma tyrrhena (Petagna). » pinnae Lockington. Conchodytes biunguiculatus (Paxson). a domestica (Gibbes). In Tridacna. Anchistus miersi (de Man). biunguiculatus Borr. spinuliferus (Miers). mirabilis (Pesta). sh demant, sp. nov. Conchodytes tridacnae Peters. A meleagrinae Peters. In Meleagnina. Anchistus mierst (de Man). Conchodytes meleagrinae Peters. In Margaritophora. Pontonia margarita Smith. In Pecten,. Conchodytes domestica (Gibbes). In Spondylus. Anchistus miersi (de Man). In “ clamp-shells.”’ Pontonia brevirostris Miers. In ASCIDIACEA. Pontonia flavomaculata Heller, in Phallusia, Diazo- na and Ascidia. ascidicola Borr. “a okat, sp. nov., in Ascidia. 7s anachoreta, sp. nov., in Polycarpa. ” I have been able to include in this paper brief colour descrip- tions of a number of species which I have observed in the living state. Most of these are based on notes made at Port Blair in the Andaman Is., where the Pontoniid fauna is one of unparalleled richness. Though the colour pattern cannot as a rule be used in taxonomic work, there is no doubt that it is often of specific value 1922.] S. Kemp: Notes on Crustacea Decapoda. 11g and even when the actual tints are variable the distribution of the pigmentis frequently constant. A colour description of Coralliocaris superba made at Port Blair agrees in a wonderfully exact manner with the coloured figure published by Dana in 1852; had there been any doubts as to the identity of the species the evidence of colour would have been most helpful. The colourationof many species of Pontoniinae is very strik- ing and there can be little doubt that in some cases it is pro- tective. Potts' has observed that the rather strikingly coloured species which live on Crinoids usually harmonize well with their hosts anda remarkable correspondence with the host in both pigment and pattern was noticed by Col. Alcock? ina Pontoniid associated with Pteroeides. But protection will not always supply an explanation. Of the two Pontoniids associated with Discosoma, one, P. inornatus, is protectively coloured ; itis semitransparent, without any pigment- ation whatever, and can only be detected with difficulty as it crawls among the short tentacles of the Actinian. The other species, P. brevicarpalis, though very closely allied, is pigmented ina most remarkable manuer and is probably one of the most gorgeous prawns in existence. By reason of its colour it is always excessively conspicuous. Periclimenes vex, another species with very ‘brilliant colouration, is perhaps associated with a red and white sponge and it is possible that the colour, though very bright, is protective. In addition to the rich collection of the Zoological Survey of India, I have been able, thanks to the courtesy of Prof. Ch. Gra- vier, to examine a number of undetermined specimens belonging to the Paris Museum. Among other interesting species this collec- tion contains a very remarkable prawn for which I have proposed the new genus Thaumastocaris. ‘to Dr. W. T. Calman I am in- debted for much assistance while working at the British Museum and to Dr. C. Forster Cooper for the opportunity of examining some of the species described by Borradaile. The types of the new species, unless otherwise noted, are in the collection of the Zoological Survey of India. Key to the genera of Pontoniinae. A. Mandibular palp present, usually composcd of two seg- ments [rostrum laterally compressed with conspicuous teeth ; dactylus of last three legs (? always) simple]. B. Second maxilliped with podobranch ; first pleopod of male with appendix interna [free-living] Urocaridella, p. B'. Second maxilliped without podobranch ; first pleo- AP pod without appendix interna. [free-living or asso- ciated with crinoids] es se ... Palaemonella, p. A'. Mandibular palp absent. 122. B. Antennal scale well developed. ‘ Potts, Public. Carnegie Inst. Washington, no. 212, p. 81 (1915). _ 2, Alcock, A Naturalist in Indian Seas, p. 14 (1902). The species on which this observation was made is Dasycaris symbiotes, gen. et sp. nov. 120 Records of the Indian Musewm. [VOL XXIV, C. Dactylus of last three legs simple or biunguicu- late,! but without basal protuberance. D. All three maxillipeds with exopods. &. Inner lacinia of maxillula narrow ; free-liv- ing or epizootic on coelenterates or echino- derms. F. Carpus of first leg not segmented. G. Carapace not areolated ; basal antennu- lar segment normal in form ; abdominal pleura usually rounded inferiorly.? H. Rostrum laterally compressed, with conspicuous teeth. F. Carapace not depressed | free-liv- ing or associated with coelenterates or echinoder ms | ... Periclimenes, p. F'. Carapace depressed, often very 134. strongly [associated with corals) ... Harpilius, p. 226 Rostrum depressed and __ toothless secs aa with crinoids] Pontoniopsis, ). G’. Carapace areolated; basal antennular 239. segment greatly attenuated anteriorly ; third to fifth abdominal pleura sharply pointed inferiarly (rostrum laterally com- pressed, with dorsal teeth ; associated with alcyonaria | . Dasycaris, p. 240. F, Carpus of first leg segmented [rostrum laterally compressed, with teeth; cara- pace not areolated ; ? free-living] ... Lhaumastocaris, p. E', Inner lacinia of maxillula very broad ; 244. endozootic in lamellibranchs or ascidians. F. Rostrum laterally compressed in distal half, toothless or with small teeth at apex only; dorsal spines of telson very small [living in lamellibranchs ] Sa ... Anchistus, p. 247. fF’. Rostrum depressed, toothless; dorsal spines of telson usually large [living in lamellibranchs or ascidians] _... Pontonia, p. 259. D', Exopods absent from some or all maxillipeds. £, Rostrum toothless ; carapace not sculptured, without supra-orbital crest; no tooth on first abdominal somite; free-living (?), or asso- ciated with gorgonians his ... Pontonides, p. 266. E', Rostrum with teeth ; carapace deeply sculp- tured, with supra-orbital crest on either side armed with teeth; a mid-dorsal tooth on first abdominal somite; associated with red coral .... Balssia, p. 207. Ce Dactylus of last three legs simple o or - biunguicu- late ® and with a large basal protuberance. D. Rosirum very long; carapace areolated, with huge antennal and supraorbital spines and with pterygostomian spine; abdominal pleura sharp- ly pointed inferiorly {2 free-living ] ... Coutierea, p. 267. D', Rostrum little if at all longer than scale; carapace not areolated, without supraorbital or pterygostomian spines; antennal spine when present short; abdominal pleura inferiorly rounded 1 Biunguiculate in Peyiclimenes s. str., in Thaumastocaris and in some spe- cies of Anchistus and Pontonia. 2 The only exceptions are found in the genus Har pilius 8 Biunguiculate only in Conchodytes. 1922.] S. Kemp: Noles on Crustacea Decapoda. 12 E, Dactylus of last three legs with basal pro- tuberance double [rostrum toothless, concave above ; associated with echinoids | . Stegopontonia, p. E'. Dactylus of last three legs with basal pro- 268. tuberance single. F. Rostrum laterally compressed, frequently with teeth ; inner lacinia of maxillula nar- row; dactylus of last three legs with a single claw and a hoof-shaped basal pro- tuberance ; living on corals . .. Coralliocaris,! p. F', Rostrum depressed, toothless ; inner 268 lacinia of maxillula very broad; dactylus of last three legs with two claws and flat basal ae a living in Jamelli- branchs ce ... Conchudytes, p. 279. i _Antennal scale rudimentary. . Rostrum present, with or without teeth; distal Matis of maxilla well developed ; all maxillipeds with cxopods; dactylus of last three legs biun- guiculate [associated (? always) with sponges]... Typton, p. 280, C’. Rostrum absent; distal lacinia of maxilla rudi- mentary; second and third maxillipeds without exopods ; dactylus of last three Jegs simple .. Paratypton, p. 286. In this key Nobili’s Onycocaris, originally proposed as a sub- genus of Coralliocaris, is not included (see p. 278). I am not con- vinced that the two species for which it was founded are related to Coralliocarvis, and as I have not seen either I prefer to leave their position undetermined for the present. The generic position of a number of other species is doubtful*; when they are better known it is probable that some modification will be necessary in the generic arrangement here adopted. Balss’ Bathypalaemonella* evidently does not belong to the subfamily, as it possesses a series of arthrobranchs in addition to five pleurobranchs. Of the seventeen genera which I recognise Periclimenes com- prises by far the largest number of species. No less than eight genera are monotypic and the majority of these are known from single specimens only. _In the keys to the species I have followed Borradaile’s ex- ample and have in each instance inserted the rostral formula. An expression such as R. II-I4 : 2-3 indicates that the teeth on the upper border of the rostrum vary from 11 to 4 and that there are 2 or 3 teeth on the lower border. The length of a specimen, as given in the descriptive parts, represents the distance between the tip of the rostrum and the tip of the telson with the animal extended as nearly as possible in a straight line. The figures in the text, even when forming part of a single text-block, are not necessarily drawn to the same scale. ! Not including Onycocaris Renin 2 In my attempts to readjust the gencric classification of the subfamily I have found myself greatly handicapped by our inadequate knowledge of a number of species. It is important that we should have fuller knowledge of Onycocaris, of the two species from Japan which Bualss referred to Peviclimenes (see p. 138) and of the three forms attributed to BGO by Miss Rathbun (see p. 268). 3 Balss, Zool. Anz. XLIV, p. 598 (1914) 122 Records of the Indian Museum. [VoL. XXIV, Genus Urocaridella Borradaile. 1915. Uvocaridella, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 2v7. 1917. Urocaridella, Borradaile, Tyans. Linn. Soc. (2) Zool. XVII, P. 352. The presence of the appendix interna on the first pair of pleo- pods is a very remarkable character of this genus and one in which it differs, I believe, from all other known Caridea. It should be noted, however, that the appendix is to be found on the first pleopods in males only, not in both sexes as implied by Borradaile. Urocaridella gracilis Borradaile. 1915. Uvocaridella gracilis, Borradaile, Ann. Mag. Nat. Hist.(8) XV, . 210, 1QI7. Trocapiaewe gracilis, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 352, pl. li, fig. 2. This species was described by Borradaile frcm Suvadiva, Kolumadulu and Haddumati Atolls in the Maldives. It is here recorded from the Orissa Coast, the Andaman Is. and Mergui Archipelago. Specimens from the Andamans were transparent nen alive with brown speckling and with narrow transverse brown bands at the end of the carapace and on the second and third abdominal somites. There were brown patches in the middle and at the tip of the rostrum, on each side of the first abdominal somite, at the tips of the telson and uropods and at the base of the uropods. The antennules, antennae and all ane legs were broadly banded with red. The largest specimens in iM collection are ovigerous females about 30 mm. in length. 2183/7. Off Chilka J.ake, Orissa ‘Investigator,’ Jan., One. Coast, 11 fms. 1890. C 342/1. Port Blair, Andamans, 2-8 S. Kemp, _ Feb., Many. fms. 1915; Feb., March, 1921. C 343/1. Mergui Archipele ago, 1ofms., ‘Investigator,’ Oct., Many. 12°40! N., 98°26'30"E. 1913. C 344/1. Mergui Archipelago, 6 fms., ‘ Investigator,’ March, Three. 11°17/20" N., 98°29/40" E. 1914. The specimens from Port Blair were caught in bottom nets hauled in Ross Channel and at the mouth of Brigade Creek ; those from the Mergui Archipelago, none of which are fully adult, were obtained at night in surface nets. Genus Palaemonella Dana. 1852. Palaemonella, Dana, U. S. Explor. Exped., Crust. I, p. 582. 1917. Sit eae Borradaile, Trans. Linn. Soc. (2) Zool. XVII, Pp. 350. Borradaile includes twelve species in this genus, but except for the two originally described by Dana and the three that Borradaile himself named, all require re-examination. In general appearance 1922.} S. Kemp: Notes on Crustacea Decapoda. 123 the species of Palaemonella bear an exceedingly close resemblance to those of Peviclimenes. ‘The only valid distinction between the two lies in the presence of a mar dibular palp in the former genus and its absence in the latter. Unfortunately this character is one to which attention is seldom paid, with the result that the generic position of a number of species is doubtful. Palaemonella laccadivensis Alcock and Anderson does not possess a mandibular palp and is transferred to the genus Pertcli- menes ; in Pertclimenes pottst on the other hand this appendage is present and the species is in consequence removed to Palaemonella. Borradaile’s Palaemonella tridentata is in my opinion a synonym of Dana’s P. tenuipes and Zehntnet’s Palaemonella ambotinensis is perhaps synonymous with Periclimenes brevicarpalis (Schenkel). Several species with the dactyli of the last three legs biungui- culate have been referred to Palaemonclla, but the position of all is uncertain.! The five species that I have myself examined may be distin- guished thus :~— A. Hepatic spine present. B. Distal margin, of carpus of second leg toothed or angulate on its inner aspect, but without a large subtey minal spine. C. Antennal scale strongly narrowed distally, with spine extending far beyond apex ; aspine at distal end of merus of second leg. D. A vestigial supra-orbital spine ; propodus of third leg at most 4'5 times length of dactylus ; R. 6-8: 1-3 vestigialis, sp. NOV. D'. No vestige of supra- -orbital spine ; ; pr npodus of third leg more than 5 times length of dactylus ; R. Tao ee pottst (Borr.). C’. Antennal scale not narrowed distally, with spine scarcely extending beyond apex ;_ no spine at distal end of merus of second leg; R. 8:2... /ata, sp. nov. B’. A large subterminal spine on carpus of second leg [antennal scale narrowed distally, with spine extending much beyond apex; a spine at distal end of merus of second leg]; R. 6-8: 1-3 tenuipes Wana, ’. Hepatic spine absent [no Spur at distal end of merus of second leg]; R. 6-7: ae: ... ortentalis Dana. Palaemonella vestigialis, sp. nov. (Plate III, fig. 2.) The rostrum extends beyond the end of the antennular pe- duncle and reaches about to the apex of the antennal scale. It vaties somewhat in depth and is straight for the greater part of its length with the terminal portion sometimes turned a little upwards. On the pase bordet it bears from 6 to 8 teeth,’ aes 7; the pos- ! A specimen from Australia which Balss (K. Svenska Vet.-Akad. Handl. LXI, no. 10, p. 13, 1921) has doubtfully attributed to Nobili’s Palaemonella biunguiculata bears only four spines at the apex of the telson and probably be- longs to the subfamily Palaemoninae. 2 Of thirteen specimens four have 6 dorsal teeth, six have 7 and three have 8. 124 Records of the Indian Museum. [VoL. XXIV, - terior tooth is placed in front of the middle of the carapace, the second is behind the orbit, while the foremost is small and is not far removed from the apex. On the lower border there are from 1 to 3 teeth,' usually 2, which are large and placed in the anterior half of the rostral length. In the position usually occupied in other genera by the supra- orbital spine a small angular prominence or tubercle may be detected and extending downwards from this tubercle to the base of the antennal spine there is a well-defined curved ridge parallel with the orbit. From this ridge the carapace slopes obliquely inwards to the orbital margin, the orbit thus having a broadly bevelled edge. ‘The antennal spine is strong; the hepatic spine is placed behind it, but on a lower level. The eyes are stout with short, thick stalks. The cornea is a little wider than the stalk and frequently, as in some species of Periclimenes, shows two concentric bands of dark pigment. The ocular spot touches the cornea. The basal segment of the antennular peduncle (text-fig. 2a) is broad; the lateral process does not reach the middle of the seg- ment ; the terminal spine is rather short and the margin between this spine and the articulation of the second segment is nearly straight. The two distal segments are stout. The free portion of the shorter ramus of the outer flagellum is half or rather less than half the length of om) the fused basal part, the latter consisting of 8 to 10 segments. ‘The total length of the shorter ramus is equal to or rather less than that of the peduncle. The antennal scale (text-fig. 1) is from 3°3 to 4 times as long as wide, proportionately longest in males, and is strongly narrowed apically. The outer margin is straight or very slightly concave and terminates in a spine which reaches far beyond the end of the lamella. | There is a minute arthrobranch at the base of the third maxilliped. The exopod almost reaches the end of the antepenul- Texr-ric. 1.-—Pal = : i : one” timate segment and the ultimate segment, nella vestigialis, sp. nov. excluding the terminal spine, is about Antennal scale of female. three quarters the length of the antepenul- timate. The first peraeopods reach beyond the apex of the antennal scale by considerably more than the length of the chela. The carpus is about equal in length with the merus.and is from 1’o to 1°25 times as long as the chela. The fingers are longer than the palm and are unarmed. The second peraeopods in adults of both sexes reach beyond the antennal scale by the whole of the chela and carpus. The 1 Of thirteen specimens two have 1 ventral tooth, ten have 2 teeth and one hv: 3: 1922.] S. Kemp: Notes on Crustacea Decapoda. 125 merus bears a strong spine close behind the distal end of the lower margin and is from 5°5 to 6 times as long as wide and from 1°25 to 1°4 times! as long as the carpus. The carpus is conical, from 2°8 to 3°2 times as long as its distal breadth, most slender in females. From the distal margin on the inner side there project two small acute processes or teeth, the upper the most conspicuous ; the strong subterminal spine found in Palaemonella tenuipes is com- Cc. b. Text-riGc. 2.—Palaemonella vestigialis, sp. nov. a. Antennule. c. Telson. b. Last two segments of third peraeopod. piectely absent (c/. text-figs. 7a and 7b). Behind the distal edge, especially on the upper side, the carpus exhibits a transverse furrow, while the distal edge itself is somewhat dilated. The chela is from 2°3 to 2°65 times as long as the carpus and is proportion- -ately longest in males. The palm isa little swollen, wider than the distal end of the carpus, 3 times as long as its greatest breadth and from 1°3 to 1°5 times as long as the fingers. The fingers have inturned tips, their cutting edges are unarmed distally, but in the proximal half each bears two teeth, those on the dactylus in advance of those on the fixed finger. The last three pairs of peraeopods are slender; the fifth reach a little beyond the end of the antennal scale. In the third pair ' About equal to the carpus in a female from Mahé. 126 Records of the Indian Museum. [VoL. XXIV, the merus is from 9 to Io times as long as broad. ‘The propodite bears spinules on its posterior border (text-fig. 2b) and is from 3°5 to 4°5 times as long as the dactylus. The sixth abdominal somite is about 1°5 times the length of the fifth. The spinules on the dorsum of the telson (text-fig. 2c) are so arranged as to divide its length into three equal parts. Large specimens are about 18 mm. in length. C 394-5/1. Port Blair, Andamans. S. Kemp, March, Four, including 1915; Feb., 1921. Types. C 3096/1. Cheval Paar, Ceylon. T. Southwell, Nov., Five. 1910. 7717/6. Kabusa I., Mergui. ‘Investigator,’ One. March, 1887. 3980/1. Tor and Ain Musa, R. B. S. Sewell, Three. Gulf of Suez. 1916. I have also seen three specimens from Mahé, Seychelles, belonging to the Paris Museum (Alluaud coll.). The specimens from Port Blair were found at low water in rock-pools at Aberdeen and in North Bay. ‘The type-specimens are from the former locality. . A male and female from Port Blair, found on a muddy shore near the mouth of Brigade Creek, differ from the specimens described above in the absence of the vestige of the supra-orbital spine and in the longer dactyli of the last three legs. In the third pair the propodite is only from 2°6 to 3 times as long as the dactylus. In the male the merus of the second peraeopod is about 4°5 times as long as wide and the carpus about 2°5 times as long as its distal width. The male possesses three pairs of spines on the back of the telson; but this is no doubt an abnormality as the teeth are not arranged symmetrically. C 400/1. Port Blair, Andamans. S. Kemp, March, Two. 1921. The specimens were found among lumps of dead coral on ‘muddy ground. Palaemonella pottsi (Borradaile). 1915. Periclimenes (Falciger) pottst, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 213. 1915. Peviclimenes potts:, Potts, Publ. Carnegie Inst. Washington, no. 212, p. 82. 1Q17. Periclimenes (Falciger) pottsi, Borradaile, Trans. Linn. Soc. fe 2) Zool. XVII, p. 374. I have examined two specimens, both unfortunately in poor condition, brought by Mr. F. A. Potts from the Torres Straits and find that Borradaile was mistaken in referring the species to the genus Periclimenes. The mandibular palp is present and is composed of two segments. The species is very closely allied io ye vestigialis, differing as far as I am able to discover only in the following characters :— £922.] S. Kemp: Noles on Crustacea Decapoda. 127 (i) There is no vestige of the supra-orbital spine, though the orbit has a bevelled edge as in the allied species. (ii) The spine at the end of the merus of the second peraeopod is quite terminal in position. (iii) The dactylus of the last three peraeopods is much shorter, the propodite being from 5°3 to 5°5 times its length. ‘hese characters are not very convincing. It is possible that ot. distinctive features will be found in the second peraeopods of the male, for I have only seen one detached leg of the second pair in P. potisi and this appears to belong to a female. Palaemonella pottsi is purple in colour when alive and is asso- ciated with crinoids, whereas P. vesligialis is not conspicuously coloured in life and is free-living. There were no crinoids in the localities where the latter species was collected at Port Blair. The species is known only from the Murray Is. in the Torres Straits. Palaemonella lata, sp. nov. This species, which is represented by a single adult male, is closely allied to P. vestigialis and P. potts: but differs in the ~ following chatacters :— (i) There is no vestige of the supra-orbital spine (text-fig. 3). TExT-F1G. 3.—Palaemonella lata, sp. nov. Anterior part of carapace. (ii) The lateral process of the antennular peduncle is longer, extending beyond the middle of the segment and the terminal spine of the basal segment is also longer, reaching much beyond the middle of the second segment (text-fig. 4a). (iii) The outer antennular flagellum is more deeply cleft. The free portion of the stouter ramus is as long as the fused basal part, the latter comprising only 5 segments (text-fig. 4a). (iv) The distal end of the antennal scale is very much broad- er and the terminal spine reaches scarcely at all beyond the apex of the lamella (text-fig. 4b). ‘(v) The fingers of the first peraeopod are equal in length with the palm. (vi) There is no spine at the distal end of the merus of the second peraeopods (tert-fig. 53). In other respects there is little difference. The rostrum teaches beyond the end of the antennular peduncle and is rather 128 Records of the Indian Museum. [VoL. XXIV, TEXT-FIG. 5.—Palaemonella lata, sp. nov. Second peraeopod. b. a, a. In dorsal view. TERS 6. Mero-carpal articulation in la- 4 alaemonella lata, sp. nov. taral wiew, a. Antennule. 8. Antennal scale. c. Fingers. deep in lateral view. Vext-bic. 6.—Palaemo- nella lata, sp. nov. I.ast three segments of third peracopod. It bears 8 teeth above and 3 below, two of the former being situated on the carapace. The antennal scale is a little more than 3 times as long ‘as wide. The carpus of ' the first peraeopod is about 1°2 times the length of the chela. In the second peraeo- pods the merus is a little more than 5 times as long as wide. The carpus bears two conspicuous teeth on the inner side of its distal margin and is slightly less than 4 times as long as its distal breadth. The chela is about 2°5 times as long as the carpus and the palm is nearly 4 times as long as broad. There are two teeth in the proximal half of each finger as in P. vestigialis. The last three peraeopods are slender, the fifth reaching well beyond the antennal scale. In the third pair the merus is ro times as long as wide and the propo- dus, which bears spinules on its posterior edge, is 3°3 times as long as the dactylus. The telson spines are arranged as in the preceding species, 1922. ] S. Kemp: Notes on Crustacea Decapoda. 129 As in the two preceding species the mandibular palp is com- posed of two segments, but it differs in that the distal segment is very much shorter than the proximal. This is perhaps merely an abnormality and only one mandible was examined. The single specimen is about 15 mm. in length. In life it was perfectly transparent except for a few small red chromato- phores on the carpus and chela of the second legs. P. lata is readily distinguished from related species by the broad apex and short terminal spine of the antennal scale and by the absence of the spine at the distal end of the merus of the second peraeopods. € 401 1. Port Blair, Andamans. S. Kemp, Feb., 1921. One, Tyre. The specimen was found in a rock-pool at Aberdeen at low water. Palaemonella tenuipes Dana. 1852. Palaemonella tenuipes, Dana, U.S, Explor. Exped., Crust. |, p. 582. pl. xxxviii, figs. 3a-d. 1898. Palaemonella tridentata, Borradaile, Proc. Zool. Soc. London, p. 1007, pl. Ixiv, figs. 8a—c. 1899. Palaemonella tridentata, Nobili, Ann. Mus. civ. Genova (2) XX, p. 235. 1906. Palaemonella tenuipes var. (ann. sp. ?), Nobili, Ann. Sci. nat., Zool. (9) LV, p. 70. 1917. Palaemonella tenuifes and tridentata, Borradaile, Tvans. Linn. Soc. (2) Zool. XVII, pp. 323, 358- 1921. Palaemonella tenuipes, Vattersall, Fourn. Linn. Soc., Zool. XXXIV, p. 383. ? 1921. Palaemonella tenuipes, Balss, K. Svenska Vet.-Akad. Handl. ILXI, no, 10, p. 14. I have examined a single example of this species obtained at Peros Banhos in the Chagos Archipelago. It differs conspicuously form all other species of the genus that I have seen in the posses- sion of a large subterminal spine on the upper and inner aspect of the carpus of the second peraeopod in addition to one or two small angular projections on the actual distal margin of the segment. The subterminal spine is clearly shown in Dana’s fizure. In determining the specimen in the collection I have derived much assistance from the notes which Tattersall has recently pub- lished. I have no doubt that my specimen is specifically iden- tical with those that he examined and I accept his view that they should be referred to Dana’s P. tenuipes. The identification pre- supposes a considerable amount of error in Dana’s figures, but we have ample evidence that these are not to be trusted in the finer detail now necessary for systematic work on the Macrura. Tattersall remarks that Borradaile’s P. tridentata is closely allied to P. tenuipes and is doubtfully distinct. I go further and regard the former as a synonym of the latter. The specimen examined was obtained by Prof. Stanley Gardiner’s expedition and was determined by Borradaile as P. 130 tridentata. Records of the Indian Museum. [Vo., XXIV, Apart from the fact that it possesses only a single tooth on the lower border of the rostrum, it differs conspicuously from Borradaile’s figure in the proportions of the segments of the As ‘TExT F1G. 7.—Merus and carpus of second peraeopod viewed laterally from inner Side aaa a. Palaemonella vestigialis, sp. nov. 6. Palaemonella tenuipes Dana. to the less well-developed condition second peraeopods. The merus is longer than the carpus and much _ longer than the fingers and the carpus is stouter, only about 3'5 times as long as ils dis- tal breadth. Tattersall has given a tabular statement of the proportionate lengths oi the segments of the second peraeopod, the figures being derived from his own speci- mens, from Nobili’s measure- ments and from the illus- {rations by Dana ard Borra- daile. ‘The corresponding values for my specimen are merus 1['2, carpus I'0, palm 1°6 and fingers 0°8. In these proportions the specimen agrees fairly well with those that Tattersall and Nobili examined. The shorter palm in Dana’s figure may be due of his specimen; the very short merus in Borradaile’s figure is, I believe, an error in draw- ing. merus is sinuous, conspicuously con- vex in the middle; this character is shown in Borradaile’s figure and is probably found only in males, The dentition of the fingers is shown in text-fig. 8, Tattersall’s notes and the evidence of my specimen, identified as P. tridentata by Borradaile himself, all point to the conclusion that only one species of Palaemonella with subterminal carpal spine is at pre- sent known. : Dana’s specimen came from the Sooloo Sea. Borradaile’s original example of P. tvidentata was obtain- ed at Funafuti in the Ellice Is. and In my specimen, which is a male, the lower border of the . VTexr-2iG. 8.—Palaemonella tenutpes Dana. lingers of second peraeopod. he has since recorded the species under the same name from various localities in the Maldives and the Chagos Archipelago. Nobili has recorded a specimen under the name P. tridentata 1922.] S. Kemp: Noles on Crustacea Decapoda. 131 from Beagle Bay in British New Guinea. Nobili and Tattersall have examined specimens from the Red Sea, the former from Djibouti and other undetermined localities, the latter from Khor Dongonab and Suakin Harbour. I lock on most other records of P. tenutpes! with suspicion, but those of Stimpson from Ousima in the Loo-Choo Is., of Miss Rathbun from the Hawaiian Is. and of Balss from N. W. Australia are perhaps trustworthy. No reliance can be placed on de Man’s record from Amboina as his specimen did not possess either of the second legs and the identity of Ortmann’s specimens from Japan and the Maldives appears to me to be extremely doubtful. Zehnt- ner in recording a specimen from Amboina remarks that the colour is entirely black, a fact not noted elsewhere and possibly not true of real P. tenuipes. Heilprin’s record from the Bermudas cannot he accepted without corroboration. Palaemonella orientalis Dana. 1852. Palaemonella orientalis, Dana, U.S. Explor. Exped., Crust. }, > p.583, pl. xxxviil, figs. 4a-d. 21887. Palaemonella orientalis, de Man, Arch. Naturgesch. VII, i, p. 552. The single specimen which I refer to this species exhibits the following characters :— The rostrum (text-fig. 9) is slender, straight at the base and a little upturned at the tip; it reaches almost to the end of the antennal scale. On the upper border it bears 7 equidistant teeth, the hindmost placed on the carapace, the next a little in advance of the posterior limit of the orbit, and the foremost small and Text-FiG. 9.—Palaemonella orientalis \)ana. Anterior part of carapace, etc., in lateral view. situated close tothe apex. On the lower border there is a single tooth, placed beneath the fifth of those on the upper edge. The antennal spine is present, but both the supra-orbital and the hepatic are missing. The eyestalks are swollen and, in the middle, are distinctly wider than the hemispherical cornea. ‘The ocular spot is not visible. ! For references see Borradaile, loc. cit., 1917, p. 358: 132 Records of the Indian Museum. [VoL. XXIV, The lateral process of the antennular peduncle (text-fig. roa) reaches about to the middle of the basal segment. ‘The spine at the outer distal angle of the same segment is short and the margin between this spine and the arti- culation of the second segment is gently convex. The free portion of the shorter ramus of the outer antennular flagellum is only about one quarter the length of the fused basal part, the latter comprising 6 elongate segments. The antennal scale (text-fig. 10b) is narrow at the distal end and widest in the middle; its greatest breadth isa little less than one-third the total length. The outer margin is very slightly concave and ter- minates in a strong spine b. which reaches a little beyond the end of the a. Antennule. 6. Antennal scale. lamella. c. Mandible. The mandible (text-fig. Toc) resembles Dana’s figure, but the palp consists only of a single segment, bearing a seta near the apex. The exopod of the third maxilliped reaches only a little beyond the end of the antepenultimate segment. The ter- minal segment is two-thirds the lengtlt of the penultimate. The first peraeopods reach about to the end of the antennal scale. The merus is equal in length with the carpus and about 12 times as long as the chela; the palm is a little swollen, and is fully 1°5 times as long as the fingers. The second peraeopods (text-fig. 11a) are equal and reach beyond the scale by rather more than the entire length of the chela. The merus is stout, not more than 3°5 times as long as broad, about one-fifth longer than the carpus; it does not possess a spine at its distal end. The carpus is conical, less than 2°5 times as long as its distal breadth. Anteriorly, on the dorsal side, the car- pus is feebly furrowed transversely and the distal margin is re- flected outwards. ‘The carpus is a little longer than the fingers and is rather less than two-thirds the length of the palm. The chela is massive; the palm is about 2°5 times as long as broad and is 1°75 times the length of the fingers. The tips of the latter are inturned and their inner margins have blade-like cutting edges; on the dactylus there are two small and obscure teeth. Texr-ric, 10.—Palaemonella ortentalis Dana. 1922.] S. Kemp: Notes on Crustacea Decapoda. 133 The last three peraeopods are rather stout. The propodites (text-fig. 11) are unarmed except for a spinule at the distal end of the posterior margin; they are from 4'5 to 5 times the length of the dactyli. The dactylus is broad at the base, simple, strongly curved and is partially concealed by long setae springing from the end of the propodus. The appendix masculina on the endopod of the second pair of pleopods is fully formed; the specimen thus appears to be an adult male. The sixth abdominal somite is less than 1°5 times the length of the fifth. The telson has the usual three pairs of apical spines, but is unarmed on the dorsal surface except fora single spine on the right hand side placed quite close to the apex (text-fig. 11c). Ss Z mz 4. Textr—-ric. 11.—Palaemonella orientalis Dana. a, Second peraeopod. c. Tip of telson. 6. Last two segments of third peraeopod. The single specimen is about 9 mm. in total length. In life it was completely transparent. The specimen agrees almost exactly with Dana’s description and differs but slightly from his figures. The principal discrepan- cies are that in the Indian specimen the mandibular palp is one- segmented, that the exonod of the third maxilliped does not reach so far beyond the end of the antepenultimate segment and that the second peraeopods are rather longer and a little more slender. The specimens recorded by de Man differ more considerably. According to his description the first legs are much longer, with the carpus longer in relation to the chela. The second peraeopods are also much longer and the fingers bear teeth and are only half the length of the palm. The dactyli of the last three peraeopods are one-third the length of the propodus. 134 Records of the Indian Museum. [VoL. XXIV, Dana gives the length of the adult female as 8 lines, while a male examined by de Man was 13 mm. in length; the Indian speci- men is thus much ‘smaller than any previously referred to the species. C 353/1. Port Blair, Andamans. S. Kemp, March, 1915. One. The specimen was obtained at low water on the reef at the northern end of Ross Island and was not associated with a crinoid. Dana described P. orientalis from the Sooloo Sea. The speci- mens described by de Man were obtained on a crinoid at Amboina. Genus Periclimenes Costa. 1831. Pelias, Roux, Mém. sur les Salicoques, p. 25 (nom. praeocc.). 1846. Periclimenes, Costa, Cat. Crost. Napoli (unpaged). 1852. Anchistia, Dana, U. S. Explor. Exped., Crust. 1, p. 577. 1860. Uvocaris, Stimpson, Proc. Acad. Sci. Philadelphia, p. 39. 1861. Dennisia, Norman, Ann. Mag. Nat. Hist, (3), VIII, p. 278. 1902. Azcylocaris, Schenkel, Verhandl.naturf. Ges. Basel XIII, p. 565. 1915. Periclimenaeus and Periclimenes with subgenera Corniger, Cristigey and Falcigey, Borradaile, Ann. Mag. Nat. Hist. (8) f XV, p. 207. 1916. Periclimenes subgenus Hamiger, Borradaile, Brit. Antarct.” Exped. 1910, Zool. III, p. 87. 1917. Uvocaris, Ancyclocaris, Periclimenes and subgenera, Pertcli- menaeus, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, pp. 353 et seq. : {9Q19Q. Botinoles subgenera Laomenes and Cuapetes, Clark, Proc. Biol. Soc. Washington, XXXII, p. 199. In working through the collection of Pontoniinae in the Indian Museum I have reached conclusions regarding the limits of this genus which, as the above references show, differ widely from those expressed by Borradaile in his recent memoir. That Anchistia and Dennisia are synonymous with Perviclimenes has long been well established, but the inclusion of other names in the same category requires explanation. Almost at the beginning of my work 1 found the greatest difficulty in distinguishing the three genera Urocaris, Ancylocaris and Periclimenes, and it is evident from the literature that others have found themselves in the same position. In Borradaile’s key (loc. cit., 1917, p. 346) the three are placed under primary headings, distinguished for the most part by habit of body. Thus in Urocaris : ‘‘ Body very slender and compressed. Sixth abdominal segment much elongate’’; in Ancylocaris: ‘‘ Body moderately stout, not compressed. Sixth abdominal segment short”’ ; in Perz- climenes: ‘‘ Body never very slender, or much compressed. Sixth abdominal segment never much elongate.’’ The large assemblage of species which these three genera comprise exhibits a very great range of variation in the form of the body and between the most slender and the stoutest every degree of transition can be found. On these grounds it is quite impossible to distinguish separate gen- era with any certainty, Borradaile himself is inconsistent, for in P. parasiticus, which he retains in the genus Periclimenes, the habit of body is extremely slender and the sixth abdominal somite 1922.] S. Kemp: Notes on Crustacea Decapoda. 135 is decidedly longer than in nearly all the species referred to Urocarts. It remains to be seen whether there are any other characters which will justify the retention of Uvocaris and Ancylocarts as separate genera. ‘The type of Stimpson’s Urocaris is U. longicaud- atus from the West Indies. In this species,asinP scriptus (Risso), the type of Costa’s Periclimenes, the last three peraeopods have biunguiculate dactyli. Several Indo-Pacific species are closely allied to U. longicaudatus, but the latter does not possess the anten- nal spine of the carapace which is present without exception in all other species hitherto referred to Urocaris, Ancvlocaris and Pericli- menes. On the closest examination and comparison it does not seem possible to separate a group of species to which the name Urocaris cau be applied and, if the genus is to be retained, it must be monotypic and characterized solely by the absence of the an- tennal spine of the carapace. It is very difficult to assess the value of a unique character of this kind; but in view of the clear affiuity which exists between U. longicaudatus and various other species I am of the opinion that Urocarts should be regarded as a synonym of Periclimenes. An illustration of the impossibility of distinguishing between Uvocaris and Periclimenes, as usually ap- plied, is to be found in a recent paper by Balss, in which the type of the latter genus is redescribed as a new species of the former. Ancylocaris was erected by Schenkel for a species, A. brevicar- pais, which is now known to be commensal with giant anemones of the genus Discosoma. The same species has since been described under a variety of specific hames; it was referred to the genus Palaemonella by Nobili, to Harpilius by Lenz, and to Periclemenes by Miss Rathbun. - It will be seen from Borradaile’s key that Ancylocarts in reality differs from Periclimenes in only one character —-that the carapace of the female is strongly swollen dorsally. This feature is well developed only in large females and a slight swelling of the carapace is not infrequently seen in normal Periclimenes. Moreover, in a species described in this paper which is also com- mensal with Discosoma, the carapace is not at all swollen, though in all other respects it shows an extremely close affinity wih A. brevicar- palis. ‘“here is thus clear proof that the swollen carapace of the feinale iu A. brevicarpalis is not a character of generic value. As will be seen further on, the name Ancylocaris may be employed in a new sense for a subgenus of Periclimenes. . It may here be pointed out that the extent to which the outer antennular flagellum is cleft—a character to which Borradaile attrti- butes importance—cannot be used, at any rate in the Pertclimenes group, for the separation of genera. In Periclimenes there is a small and rather clearly defined group of species inhabiting. water of moderate or great depth and the four known representatives of this group agree among themselves even in a peculiar disposition of teeth in the second pair of chelae. In two of them (P. lattpollex and P. laccadivensis) the outer antennular flagellum is 136 Records of the Indian Museum. ; VOL. XXIV deeply cleft, with the free portion of the shorter ramus longer than the fused basal part: in a third (P. lamipes) the free portion is slightly shorter than the fused part: in the fourth (P. alcocki) the flagellum is scarcely cleft at all, the free portion of the shorter ramus being less than one-third the length of the fused basal part. Urocaris and Ancylocaris are thus, in my opinion, to be re- garded as synonyms of Periclimenes. As regards the subdivision of the large assemblage of forms included in the genus, it will be observed that Borradaile in 1915 proposed four subgenera, Ensiger, Corniger, Cristiger and Falciger and in 1916 added a fifth, Hamigey. Two of these terms are preoccupied as genera, and Mr. Austin H. Clark, who does not seem hitherto to have interested himself in carcinology, has felt it necessary to substitute others. The subgenus Ensiger includes only Dana’s Anchistia aurdun- tiaca, a species of doubtful affinity which has not been examined since 1852. From the original account it is not even certain that the species belongs to the subfamily Pontoniinae, for the telson is described as “a little hairy at tip, with two short spines.’ Any decision as to the proper position of Ensigey must therefore be postponed until the type-species has been rediscovered. Borradaile refers the great majority of the species which he includes in Periclimenes to the subgenera Cristiger and Falczger. He separates the two(loc cit., 1917, p. 360) by a number of features, but it will be seen that the only absolute criterion for their discri- mination lies in the form of the rostrum, which is stated to be convex in the former and straight.or concave in the latter. This character is one of very little values In determining the speci- mens in the Indian Museum I have made every endeavour to separate the species on the lines which Borradaile advocates, but have been forced to the conclusion that the division he recom- mends, even if it were possible in practice, tends only to obscure the real affinities of the species. The two Mediterranean species, P. amethysteus and P, scriptus, are so far as IT am aware distin- guished from one another only by colour, yet Borradaile refers the former to the subgenus Falcigey and the latter to Crvisizger. The subgenus Hamiger is without doubt synonymous with Periclimenaeus, the position of which is discussed below. To the curious little group of species in which the cornea is conoidal and pointed anteriorly Borradaile has applied the sub- generic name Corniger; but the character, though an interesting one, does not in my opinion, possess the importance that he attri- butes to it. In the collection on which this paper is based I have found one specimen with a conoidal cornea; but though in this respect it resembles the forms that Borradaile refers to Corniger, it is otherwise very different, for it possesses neither hepatic no1 supra-orbital spines. It is unfortunately impossible to draw up a specific description from this individual, as it is without locality and is much damaged, possessing only the first pair of legs. The existence of such a form seems, however, to indicate that the 1922.] S. Kemp: Notes on Crustacea Decapoda. 137 species with a conoidal cornea do not necessarily form a natural group. Elsewhere in the genus Periclimenes other modifications of the eye are sometimes found. In P. seychellensts there is a papilla on the eyestalk and in two of the three species of Periclimenaeus the cornea has a circular cup-shaped depression. The evidence we possess at present tends to show that the structure of the eye, when unsupported by other characters, does not afford a valid ° basis for subgeneric division. For these reasons I am unable to accept the subgenera pro- posed by Borradaile. I recommend instead an arrangement in which the primary division is based on the structure of the dactyli of the last three pairs of peraeopods, whether simple or with an accessory lobe or claw. The structure of the dactyli in these limbs is of generic importance in the more highly specialized Pontoniinae and the character is of established value in other Caridea. : Whether the arrangement leads to a natural grouping of the species on a phylogenetic basis, is a qttestion that cannot he answered in the present state of our knowledge. I incline to the view that it does. Insome species, however, the additional dacty- lar claw is reduced to a mere process or lobe,' and there is thus a possibility that certain specialized species in which the dactylus is simple may have been derived from forms in which it was once biunguiculate.* In P. scriptus, the type-species of Periclimenes, the dactyli are biunguiculate, and the subgenus to which this species belongs may thus be termed Periclimenes s.s. For the more primitive forms with simple dactylus Schenkel’s Ancylocaris may be em- ployed, though in a different sense to that in which it has hitherto been used. Borradaile’s Periclimenaeus, of which his Periclimenes subgen. Hanuger is a synonym, is at most a subgenus of Periclimenes. In the three known species the dactyli of the hinder peraeopods are biunguiculate, thus resembling Periclimenes s.s., but the hepatic spine of the carapace, which is invariably present in the latter, is here absent. The chelae of the second peraeopods are more mas- sive in Periclimenacus than in Periclimenes s.s., though the species of the latter subgenus exhibit a very great range of variation in this respect. The characters of the ‘three subgenera that it propose may be summarized thus :— Dactyli of last three peraeopods biunguiculate or with an accessory process or lobe behind terminal claw. (139. Hepatic spine present “s ee ... Periclimenes s.s., p. Hepatic spine absent ... Periclimenaeus, p. 166. Dactyli of last three peracopods simple [Hepatic spine usually present]... : ... Ancylocaris, p. 167. ! e.g., Peviclimenes vex and P. noverca. ? Of this P. frater is perhaps an example. 138 Records of the Indian Museum. [VoL XXIV, Under the subgeneric headings synoptic tables to the majority of the known species will be found. In Periclimenes s.s. 20 species are recognised, in Periclimenaeus 3 species and in Ancylocaris 44 species. The following are omitted from these tables :— Anchistia aurantiaca Dana, U.S. Explor. Exped., Crust. I, p. 581, pl. xxxviil, figs. 2a-d (1852). The generic position of this species is very doubtful and it is not certain that it belongs to the Pontoniinae. The mouth-parts have apparently not been examined and the telson is described as ‘“‘a little hairy at tip, with two short eee Dana’s specimens were found at the Fiji Is. Anchistia danae Stimpson, Proc. Acad. Sci. Philadelphia, 1860, p. 108. This species, from Tahiti, will probably never be recognized with certainty. There is no description of the second peraeopods and it is uncertain whether the posterior dactyli are simple or biunguiculate. The specimens doubtfuily referred to this species by Borradaile' perhaps belong to the P. gvandis section of Ancy- locaris, but the description is insufficient. Anchistia brachiata Stimpson, loc. cit. supra, p. 108. Found at Port Lloyd in the Bonin Is. There is no descrip- tion of the last three peraeopods. Anchistia notata Heller, Crust. ‘Novara’ Exted., p. 109, pl. x, fig. 3 (1865). ; Described from a specimen without the second peraeopods obtained at the Nicobars. Perichimenes parasiticus Borradaile, Ann. Mag. Nat. Hist. (7) II, p. 384 (1898) and in Willey’s Zool Results, p. 407, pl. xxxvi, fig. 4 (1899); Nobili, Ann. Mus. civ. Genova (2) POX ars. 235 (1899). The description of this species is most iindeeniaee: IT examined the type-specimens in the Cambridge Museum, but found that all the legs were missing except those of the first pair. The species was found at New Britain on a black starfish belonging to the genus Linckia. Periclimenes hertwigi and gorgonidarum Balss, Abhandl. math.-phys. Kt. K. bayer. Akad. Wiss. Suppl. Bd. II, pp. 49-52, text-figs. 28-32 (1914). Further particulars of these two remarkable species are required before their position can be determined. It is possible, as Borradaile has remarked, that they do not belong to the Pontonii- nae, | Borradaile, Proc. Zool. Soc. London, 1898, p. 1oo4, pl. Ixiti, figs. 4, 4a, 0. 1922. ] 5S. Kempe: Notes on Crustacea Decapoda. 139 Perviclimenes beaufortensis Borradaile, Ann. Mag. Nat. Hist. (9) V, p. 132 (1920). According to the description this species does not possess exopods on the second and third maxillipeds. It cannot therefore be retained in the genus Peviclimenes, but belongs in all probability to Pontonides (see p. 266). Periclimenes tenuipes Leach. Nobili’s statement that Leach described a Mediterranean species under this name is erroneous (v. infra, p. 223). Subgenus Periclimenes, sensu stricto. ‘The accessory claw or process found on the dactyli of the last three peraeopods in this subgenus is, J presume, to be regarded as a sign of specialization ; Periclimenes s.s. is thus less primitive than Ancylocaris. The species included in the subgenus exhibit great variation in habit of body. Some, such as. P. longicaudatus are extremely slender in build, while others, such as P. lanipes, are remarkably stout. P. scriptus, the type of the subgenus, is intermediate in form, without any strongly marked characters, anid it appears to me probable that it is from some such species as this that the remainder have evolved. P. latipollex, P. laccadivensis, P. alcocki and P. lanipes form a rather distinct section of the subgenus, distinguished by the tooth and socket arrangement in the dentition of the fingers of the second leg. P. soror and P. noverca ditfer from all other species of the subgenus in the possession of a series of fine teeth on the edges of the fingers of the first leg. In this they resemble P. spiniferus, P. petitthouarsi and P. denticulatus,’ which belong to the subgenus Ancylocaris. I think it most improbable that there is any real affinity between these two groups of species and regard the similarity in structure of the fingers of the first leg as an instance of convergence. Certain species possess characters which are unique in the genus: P. longicaudatus has no antennal spine, P. aesopius has a large compressed tooth on the third abdominal somite and in P. investigatoris the lateral process of the antennule is of abnormal length. ! The same character is also found in P. frater. Borradaile considers this species to be a close ally of P. sovor, but the dactylus is said to be simple and I have consequently included it in the subgenus Ancylocaris. InP. noverca the accessory claw of the dactylus is reduced to a mere lobe and it is easy to understand how this lobe might disappear altogether by further modification along the same lines. If Borradaile’s views on the relationships of P. frater are correct, the species has presumably been evolved from one with biunguiculate dactyli and has no affinity with the more primitive forms included in the subgenus Ancylocaris. The position of the species thus. requires further consideration. 140 Records of the Indian Museum. (Vou. Xeiy, Key to the species of the subgenus Periclimenes. A. Supra-orbital spine absent. B. Antennal spine absent ; R. 7-8 : 1-2 B’. Antennal spine present. C. Third abdominal somite produced backwards over fourth in the form of a large compressed aie R. g-11 : 2 Third abdominal somite little produced pos- gas PD, Fingers of chela of first ley unarmed. &. \.ateral process of antennular peduncle of normal length, not reaching beyond middle of basal segment. /. Second leg with carpus more than one- third length of palm. G. One or more upper rostral teeth situ- ated on carapace behind posterior limit of orbit. H. \actylus of last three legs slender, at least 4 times as long as “broad. F. Posterior dorsal tooth of rostrum separated from next by a wide in- terval; carpus of second leg much more than half as long as palm. K. Upper border of rostrum very strongly arched, with ventral teeth placed close to apex below or in advance of foremost dorsal tooth; fingers of second leg as long as a R. 6-8 : R, oF Il:1-3 Kk". Upper bowler of rostrum only a little convex, with ventral teeth placed behind toremost of dor- sal series; fingers of second leg usually shorter than palm; R. 7-10: {=2 ‘’. Posterior dorsal tuoth of rostrum not separated from second by a wider interval than that between second and third, carpus of second leg about half as lony as palm. &. Abdomen transversely banded and blotched with red ; R. 8-10: 2-4 Kk! Facts longitudinally str iped with violet; R, 8: 4 ' actylus of last. three legs stout, ee than 4 times as long as broad [posterior dorsal tooth of rostrum not separated trom second by a wider interval than met between second and third]; R.9g: No teeth of ice rostral series situ- oe on carapace behind orbit. H, Rostrum deep, downcurved ; apex of Se scale broadly rounded ; R. 6: rai H'. Resutin shallow, straight ; apex of ie scale wale rounded ; R.6: longicaudatus (Stimp- son). aesopius (Bate). infraspinis (Rathbun). indicus (IXemp). obscurus, Sp. Nov. scriptus (Risso). amethysteus (Risso). tmpar, Sp. Nov. parvus Borr. tncertus Borr. 1922. |] S. Kemp: Noles on Crustacea Decapoda. | I4I F’. Second leg with carpus one-third or less than one-third tength of palm. G. Rostrum with at most ro dorsal teeth. H. Fingers of second leg more than half as long as palm, ? without teeth on inner margins [merus of second leg with jo: at end of lower bor- der]; R.6: H'. Firgers of Pee leg half or less than half as long as palm. dactylus with a tooth fitting into a cavity in fixed finger, as Rostrum straight or upturned ; merus of second leg unarmed; last three legs slender with merus unarmed and without thick hair. K. Fused portion of outer anten- nular flagellum short; second legs smooth; two pairs of spines on back of telson. L. Hepatic spine on a level with antennal ; dactylus of second leg erik externally; R. Pane aerate spine below level of Seteeeae dactylus of second ‘leg not flanged externally ; R. 10 : 2-3 af an, | K'. Fused portion of outer anten- nular flagellum very long; se- cond legs minutely tuberculate ; four pairs of spines on back of telson eho 2a. i F'. Rostrum downcurved ; merus of second leg with tooth at end of lower border ; last three legs stout, inferior margin of merus with spinules and distal tooth, propodus densely clothed with hair; R. 8-9 : o-1 .. Resta with 23 dorsal ‘teeth, lower aedee unarmed E’. Lateral process of antenoular peduncle abnormally long, reaching distal end of basal segment; R.g: 1 D'. Each finger of chela of first leg with inner margin finely pectinate. E. Second leg with merus unarmed and fingers one-third length of palm; no tooth = distal end of merus of last three Oe WL I=13) 510 2 he leg with merus armed with a Pete at distal end of lower border and with fingers more than half as long as palm; a tooth at distal end of merus of last three legs; R. 7:0 A’. Supra-orbital spine present; R. 5: 2 gracilis (Dana). latipollex, sp. nov. laccadivensts (Alc. and And.). alcocki, sp. nov. lanipes, sp. nov. rex; Sp. nov. investigatovts, Sp. nov. sovor Nobili. noverca, Sp. Nov. commensalis Borr. Periclimenes (Periclimenes) longicaudatus (Stimpson). 1860. Urocaris longicaudatus, Stimpson, Proc. Acad. Sci. Philadelphia, P- 39- 142 Records of the Indian Museum. [VoL. XXIV, 1900. rd sags longicaudata, Rathbun, Proc. Washington Acad. Sci. , Pp. 155: 1902. Urocaris tngicaudate Rathbun, Bull. U.S. Fish Comm. XX, ii, » I . 1918. ate. longicaudata, Hay and Shere, Bull. U.S. Bur. Fisher- tes XXXV, p. 304. This species, which is the type of Stimpson’s genus Urocaris, inhabits the West Indies and the adjacent coasts of America as far south as Brazil. The specimens I have examined are from Punta Rassa in Florida. The anterior margin of the carapace, immediately below the orbit, projects in the form of a long strap-shaped process with rounded apex. This projection is hornologous with the less prominent infra-orbital lobe found in many related species and is imperfectly described by Miss Rathbun (loc. cit., 1902) as a ‘rounded extra-orbital tooth.’ The antennal spine which usually arises from the vicinity of the lower limit of the infra-orbital lobe is completely absent in P. longicaudatus, though it appears to be present in all other known representatives of the subgenus Pevi- climenes. Periclimenes aesopius (Spence Bate). 1864. Anchistia aesopia, Spence Bate, Proc. Zool. Soc. London, 1863, p. 502, pl. xli, fig. 5. 1917. Uvocaris aesopius, Borradaile, Tyans. Linn. Soc.(2) Zool. XVII, P: 354- Through the kindness of the authorities of the British Museum I have been able to examine the types of this remarkable species which has apparently not been rediscovered during the past fifty years. ‘There are two specimens, one complete and one which has been dissected and is in a fragmentary condition. ‘The rostrum is slender and straight, with the ventral portion below the midrib greatly reduced. On the upper margin there are g or Ir teeth, the three: hindmost placed on the carapace behind the orbit. On the lower margin there are two small teeth near the apex and behind these teeth a fringe of very long plumose setae. The carapace is prominently angled below the orbit. There are antennal and hepatic spines, the latter on a lower level than the former.' The eyes are slender, with stalk fully twice the length of the cornea. The lateral process of the antennule is short, not reaching the middle of the basal peduncular segment. The anterior margin of this segment external to the insertion of the second segment is greatly produced, as shown in Bate’s figure, reaching the end of the second segment and extending far beyond the spine that terminates the outer margin. The antennal scale is unusually broad distally ; it is about two and a half times as long as wide, with the terminal spine not reaching the end of the lamella. ! The position given to these spines in Spence Bate’s figure is erroneous. 1922. ] S. Kemp: Notes on Crustacea Decapoda. 143 The first peraeopods reach the end of the scale. The carpus is shorter than the merus and only three-quarters the length of the chela. The fingers are unarmed and are longer than the palm. On the outer edge of the fixed finger there are some tufts of hairs. The second per- aeopods reach beyond the scale by almost the entire — length of the chela. The Text-ric. 12.—Periclimenes aesopius (Sp. merus is ttnarmed and 1°'5 Bate). times as long as the carpus. Dorsal parts of third and fourth abdominal The carpus is conical and somites in lateral view. about 3 times as long as its distal width. ‘The chela is 2'5 times as long as the carpus and rather more than 4 times as long as wide ; the fingers are unarmed, a little shorter than the palm. The third peraeopods reach the end of the scale. The propo- dus in all three pairs is provided with spinules on its posterior border and is about 4 times the length of the dactylus. The dactylus is biunguiculate, with a deep and narrow cleft between the two claws, The form of the remarkable compressed tooth which projects backwards from the third abdominal somite is shown in text-fig. 12. I know nothing resembling it in any other species of the genus. The sixth abdominal somite is 2°3 times the length of the fifth. The anterior pair of dorsal spinules of the telson are placed in the middle of its length. As usual there are two spines, one of which is movable, at the end of the external margin of the outer uropod. If the complete specimen were straightened out it would prob- ably be about 24 mm. in length. The structure of the apex of the telson and of the mandibu- lar valp (found loose in the tube containing the specimens) afford proof that the species belongs to the Pontoniinae. It must cer- tainly be referred to the genus Periclimenes in which, however, by reason of the characters of the basal segment of the antennular peduncle and third abdominal somite, it occupies a very isolated position. The two specimens were found in the Gulf of St. Vincent, S. Australia (Angas coll.). Periclimenes (Periclimenes) infraspinis (Rathbun). 1902. Uvocaris infraspinis, Rathbun, Proc. U. S. Nat. Mus. XXIV, P- 903. 1904. Urocaris infraspinis, Rathbun, Harriman Alaska Exped. X, p- 31, text-figs. toa, b. 1921. Uvocaris infraspinis, Schmitt, Univ. Calif. Publ., Zool. XXIII, Pp: 37, fig. 22. California, Pacific coast of Mexico. 144 Records of the Indian Museum. [VoL. XXIV, Periclimenes (Periclimenes) indicus (Kemp). 1915. Uvocaris indica, Kemp, Mem. Ind. Mus. V, p. 275, pl. xiii, fig. a, text-fig. 26, A comparative statement of the principal differential charac- ters of P. indicus and P. infraspinis will be found under the above reference (p. 278). So far as is known at present P. indicus is restricted to the coasts of the Indian Peninsula. It is known from the Chilka Lake iu Orissa, from Ennur backwater and the Adyar River near Madras and from Pamban and Kilakarai at the upper end of the Gulf of Manaar. The species is estuarine as well as marine and in places like the Chilka Lake, where there are great seasonal changes in salinity, has been found in fresh water. TEXT-PIG. 13,—Periclimenes indicus (Kemp). a. Anterior part of carapace and rostrum. b. Dactylus of fifth peraeopod. I have no additional records of this species, but give further figures of the rostrum and dactylus of the last leg for comparison with P. obscurus. Periclimenes (Periclimenes) obscurus, sp. nov. The rostrum is longer in females than in males. In the former sex (text-fig. 14d) it extends beyond the end of the antennular peduncle, usually reaching the end of the antennal scale, while in the latter (text-fig. 14a) it reaches only to the middle or end of the second antennular segment. The upper portion of the blade is convex, but does not take the form of the strongly arched lamella found in P. indicus. On the upper border there are from 7 to Jo teeth, usually 8 or g!; the hindmost of these is separated by a considerable interval from the next of. the series, but is always situated further forwards than in P. indicus. The remaining dorsal teeth are more or less evenly spaced and 1 Of thirty-three speciméns five have 7 dorsal teeth, twelve have 8, twelve have 9 and four have to. 1922.] S. Kemp: Notes on Crustacea Decapoda. 145 extend to the tip, the second being above or slightly behind the posterior limit of the orbit. The lower border of the rostrum bears I, rarely 2 teeth' which are rather larger than those of P. indicus and occupy a different position. In P. indicus there are asa rule 2 very small teeth, the hindmost of which is placed below or in advance of the foremost tooth of the dorsal series, whereas the single tooth usually found in P. obscurvus is placed much further back, with at least one, often with two or three dorsal teeth in advance of it. In the antennules and antennae there is little difference between the two species, but in P. obscurvus the antennal scale (text- fig. 14c) is rather less parallel-sided than in P. indicus and the C. TExT-FIG. 14.—Periclimenes obscurus, sp. nov. a. Anterior part of carapace of male. b. The same parts of female. c. Antennal scale. fused portion of the outer antennular flagellum is shorter and composed of only 4 or 5 seginents. The motith-parts, maxillipeds and first peraeopods do not exhibit any distinctions worthy of note. The second peraeopods are often a little unequal and show much variation in the propor- tionate lengths of thesegtnents. Asin P. indicus they are unarmed. In ovigerous females (text-fig. 150) the carpus is slightly shorter than, as long as, or rather longer than the palm. In males (text- fig. 15a) it is sometimes longer than the palm, rarely shorter than it, while young individuals not infrequently resemble P. indicus in having the carpus as long as the chela. The fingers are as a tule clearly shorter than the palm, thus differing from those of P. indicus which are always fully as long as the palm. In young specimens, however, and rarely in full-grown females the dactylus ! Of thirty-three specimens thirty-one have a single ventral tooth and two have 2 ventral teeth. 146 Records of the Indian Museum. [Vo,. XXIV, is equal in length with the palm. The fingers are usually un- armed, but sometimes an obscure tooth is found on each, that on the fixed finger in advance of that on the dactylus. The last three peraeopods are for the most part similar to those of the allied species. The dactylus, however, is shorter ; it is from 4 to 4°5 times as long as its basal breadth, whereas in P. indicus it is from 5°5 to rather more than 6 times (cf. text-figs. 136 and 15c). Noclear distinctions are to be found in the abdo- men, telson or uropods. YL C. TEXT-FIG. 15.—Periclimenes obscurus, Sp. nov. a. Second peraeopod of a male. 6. Second peraeopod of a female. c. Dactylus of fifth peraeopod. Large specimens reach a length of about 17 mm. C 345-6/1. Springhaven, Madras S. Kemp, May, 1918. Twenty-four, Harbour, including TYPEs. C 347-1/1, Ennur backwater, near N. Annandale, Sept., Nine. Madras. 1915. The specimens from Springhaven were taken swimming round buoys and piles encrusted with sponges, hydroids and other matine organisms. ‘Those from Ennur backwater were found in company with P. indicus, from which they were easily distin- guished by the well-marked rostral characters. 1922.] ' §, Kemp: Notes on Crustacea Decapoda. 147 Periclimenes (Periclimenes) scriptus (Risso). 1916. Uvocarts de Mant, Balss, in Michaelsen’s Beitr. Kennt. Meeres- ] faun. West-afrikas II, p. 29, text-fig. to. 1917. Periclimenes (Cristiger) scyiptus, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 362 (sy7o7.). I am unable to find in Balss’ description any character which will distinguish his Uvocaris de Mant from P. scriptus, the type of the genus Periclimenes. - P. scriptus is common in the Mediterranean and has been found at the Channel Is.; if I am right regarding the identity of the specimen described by Balss its distribution extends southwards along the West African coast to French Congo. Periclimenes (Periclimenes) aimethysteus (Risso). 1826. Alpheus amethystea, Risso, Hist. nat. Euvope Mérid. V, p. 77, pl. iv, fig. 16. 1863. Anchistia amethystea, Heller, Crust. ztidlich. Europa, p. 258. 1917. Peviclimenes (Falciger) amethysteus, Borradaile, Tvans. Linn. Soc. (2) Zool. XVII, p. 370. Other references are given by Borradaile. The original des- cription is based entirely on colour and I know of no other cha- racter by which the species can be separated from P. scriptus. Heller’s account of the colouration differs considerably from that given by Risso and the only definite points of distinction appear to be those which I have noted above in the key to the species of the subgenus. At the Oceanographical Museum at Monaco I have examined specimens from Béne, in Algeria, which bore the name amethysteus, but was unabJe to find any difference in structure from P. scripius. ‘The validity of the species must remain uncertain until fresh information based on living material is forthcoming, P. amethysteus is known only from the Mediterranean. Periclimenes (Periclimenes) impar, sp. nov. (Plate III, fig. 1.) This species is allied to the four preceding forms but differs in the much broader dactylus of the last three pairs of legs. The rostrum is a little longer than the antennular peduncle, but does not reach the end of the antennal scale. The upper margin is convex and in the single specimen examined bears 9 more or less evenly spaced teeth. The hindmost tooth is placed on the carapace behind the orbit but is not separated from the second by a greater distance than that between the second and third. The middle teeth of the dorsal series are the largest. On the lower border there are 2 teeth, placed near the tip, and the margin from the pos- terior tooth to the base is nearly straight. There is no supra-orbital spine. The antennal spine is sharp with the hepatic behind it but on a lower level. The eyes are rather stout and the ocular spot touches the cornea. 148 Records of the Indian Museum. [VoL. XXIV, The lateral process of the antennule (text-fig. 16a) reaches about to the middle of the basal segment; the spine at the end of the external margin extends beyond the middle of the second segment. The second and third segments are to- gether less than half as long as the basal seg- ment. The free portion of the shorter ramus of the outer flagellum is longer than the fused part, the latter compri- sing only 3 segments. The antennal scale (text- fig. 165) isabout 3°2 times as long as broad; the outer margin is slightly concave and terminates in a spine which reaches almost to the end of the lamella. b | The first peraeopods 42 i (text-fig. 17a) reach Text-riG. 16.—Peviclimenes impar, sp. nov. about to the end of the a. Antennule. scale. The coxopodite 5. Antennal scale. has the usual ventral process and a similar process, much better developed than in allied species, is present on the basipodite. The carpus is equal in length with the chela and is a little shorter than the merus; the fingers are unarmed and are nearly three-quarters the length of the palm. The second peraeopods are unarmed and are unequal and dis- similar, both reaching considerably beyond the end of the scale. In the larger limb (text-fig. 17b) the merus aud ischium are sub- equal. The carpus is conical, about 2°3 times as long as its distal breadth, and is half the length of the palm and rather more than half the length of the merus, ~The chela is somewhat swollen, with fingers about two-thirds as long as the palm. In the middle of the cutting edge of each finger there is a shallow excavation bounded at either end by a small tooth ; the fingers in consequence gape a little when they are closed. In the smaller limb the carpus is much longer, only a little shorter than the palm and three-quarters the length of the merus; it is at least four times as long as its distal breadth. The third peraeopods reach about to the end of the basal segment of the antennule. The propodus in all the last three pairs (text-fig. 17c) bears spinules on its posterior margin and is from 5'5 to 6 times as long as the dactylus. The dactylus itself (text-fig. 17d) is biunguiculate with a rather wide excavation between the two claws. It is considerably broader than in any of a 1922.] S. Kemp: Notes on Crustacea Decapoda. 149 the allied species, the length from the base to the bottom of the cleft being only twice the basal breadth. The sixth abdominal somite is less than twice the length of the fifth. The anterior pair of dorsal spinules of the telson are placed in the middle of its length, the posterior pair midway between the anterior and the apex. The external margin of the outer uropod is ciliated. TEXT-FIG. 17.—Periclimenes impar, sp. nov. a. First peraeopod. c. Third peraeopod. 4, Larger second peraeopod. d. Dactylus of third peraeopod. The species is described from a single ovigerous female about 10 mm. in length. C 348/i. Port Blair, Andamans, S. Kemp, March, One, Tyre. 5 fms. 1915. : The specimen was found on a sponge of a pinkish colour and was transparent when alive with reddish patches on the abdominal pleura. . Periclimenes (Periclimenes) parvus Borradaile. 1898. Peviclimenes parvus, Borradaile, Ann. Mag. Nat. Hist. (7) Il, Pp. 384. 150 Records of the Indian Museum. [Vor. XXIV, ° 1899. Periclimenes parvus, Borradaile, in Willey’s Zool. Results, p. 407, pl. xxxvi, fig. 3. New Britain. Periclimenes (Periclimenes) incertus, Borradaile. 1915. Periclimenes (Cristiger) incertus, Borradaile, Ann. Mag. Nat. Hist. (8), XV, p. 210. 1917. Periclimenes (Cristiger) incertus Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 364, pl. liii, fig. 7. I have examined the types of this species and of P. parvus and agree with Borradaile that they are specifically distinct. In addition to the characters which he has mentioned, the carpus of the second peraeopod is proportionately longer in P. imcertus and the apex of the antennal scale more sharply rounded. In both species the foremost pair of spines on the dorsum of the telson is placed at about the middle of its length. P. incertus was found at the Maldive Is. Periclimenes (Periclimenes) gracilis (Dana). 1852. Anchistia gracilis, Dana, U. S. Explor. Exped., Crust. 1, p.578, pl. xxxvii, figs. 5a—m. Judging from Dana’s figures this species, the type of the genus Anchistia, will fall in the subgenus Periclimenes, but its position is a little’doubtful, for the accéssory tooth on the dactylus of the posterior legs is not mentioned in the description and according to figure 5/ it is articulated at the base. The lamella of the antennal scale is shown to be acutely pointed anteriorly in fig. 5a, but this is probably an error. P. grawlis is recorded by Dana from the Sooloo Sea. Periclimenes (Periclimenes) latipoflex, sp. nov. (Plate IV, fig. 3.) The rostrum is very slender, straight in its proximal part and trending very slightly upwards at its distal end. It reaches a little beyond the apex of the antennal scale and is armed above with 7 or % teeth, of which the posterior 2 or 3 are situated on the carapace behind the orbit. The posterior tooth is not widely separated from the second. Towards the apex the teeth are more distantly spaced than at the base, but in both the specimens with complete rostra the distribution is somewhat irregular. On the lower edge of the rostrum in its distal half there are 3 teeth. The supra-orbital spine is wanting. The hepatic spine is placed on a level with the antennal. The lobe on the frontal edge form- ing the lower limit of the orbit is acute. The ocular spot is merged in the cornea and can only be distinguished with difficulty. The spine at the outer distal end of the basal segment of the antennular peduncle (text-fig. 184) is long; the lateral process reaches about to the middle of the segment. The fused portion 1922.] S. Kemp: Notes on Crustacea Decapoda. I51 of the outer antennular flagellum is composed of three or four segments and is about two-thirds the length of the free portion of the shorter ramus. The antennal scale (text-fig. 180) is rather more than 3 times as long aS wide (in an adult female); its outer margin is slightly concave and terminates in a spine which reaches as far forwards as the lamella. The third maxilliped bears an arthrobranch; the ultimate segment is considerably shorter than the antepenultimate. The first peraeopod reaches beyond the scale by the length of the fingers. The carpus is much shorter than the merus and slightly shorter than the chela. The fingers are unarmed and about two- thirds the length of the palm. The second peraeopods ate equal or dice and reach beyond Text-Fric. 18.—Periclimenes latipollex, sp. nov. a, Antennule. c. Last two segments of third peraeopod. b. Antennal scale. d. Dactylus of third peraeopod. the scale by the whole length of the chela. Both merus and car- pus are unarmed. The merus is nearly twice the length of the ischium ; the carpus is conical, about 1°5 times as long as broad and scarcely one-fifth the length of the chela. The chela is as long as the three preceding segments combined and much exceeds the carapace-length; the palm is from 2:2 to 27 times the length of the dactylus and is from 4°5 to 5 times as long as broad. The fixed finger has a cutting edge armed in its proximal half with three small teeth and on the dactylus there is a cutting edge with a single basal tooth. When the claw is closed the cutting edges do not coincide but slide past each other like the blades of a pair of scissors, the single tooth on the dactylus fitting into a recess in the fixed finger. The tip of each finger is provided with an inturned 152 Records of the Indian Museum. ~— [VoL. XXIV, claw. On the external side of the dactylus there isa thin blade or flange which runs the whole length of the segment and is some- what reflected outwards ; from certain points of view the dactylus thus appears very broad. The last three peraeopods are comparatively stout; the third reach beyond the scale by about twice the length of the dactylus. The merus is about g times as long as broad. The propodus (text-fig. 18c) bears some setae and a few fine spinules on its poste- tior border and is from about 6°5 to 7 times as long as the dactylus. The latter segment (text ve 18d) is rather broad and the accessory tooth is small. The sixth abdominal somite is half as long again as the fifth. The foremost of the two pairs of spinules on the upper surface of the telson is placed a little in front of the middle of the telson- length, the second midway between it and the apex. The largest specimen, an ovigerous female, is about 16 mm. in length. C 340/t. Mergui Archipelago, 62 ‘Investigator,’ Three (two ovig. fms., 12°15'20"N., April, 1913. @). Tyres. 97°10'10"E. Periclimenes (Periclimenes) laccadivensis (Alcock and Anderson). : 1894. Palaemonella laccadivensis, Alcock and Anderson, Fourn. Asiat. Soc. Bengal \.XI1II, p. 157. 1896. Palaemonella laccadivensis, Alcock and Anderson, J/lust. Zool. ‘ Investigator,’ Crust. pl. xxvi, fig. 4. 1901. Palaemon (Brachycarpus) laccadivensis, Alcock, Cat. Ind. Deep- Sea Crust. Decap. Macrura and Anomala, p: 138 (in part). ? 1906. Palaemonella laccadivensis, Rathbun, Bull. U.S. Fish Comm. g XXIII, iii, p. 925. 1Q17. Palaemonelia laccadivensis, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 358. This species, originally described as a Palaemonella and sub- sequently transferred by Alcock to Brachycarpus, belongs in reality TEXT-FIG. oh Peviclamertes laccadivensis (Alc. and And.). Anterior part of carapace, rostrum, etc. to Perichimenes. The fact that the telson has six terminal spines and that there is no pleurobranch above the base of the third 1922.] S. Kemp: Noles on Crustacea Decapoda. 153 maxilliped indicates that the species must be referred to the Ponto- niinae and, as the mandible does not possess a palp, it cannot be placed in Dana’s Palaemonella. ‘The dactyli of the last three legs are biunguiculate and the species in all other characters agrees with Periclimenes s.s., as defined in this paper. Alcock in Igor recorded four specimens of this species, all of -which I have examined. The Jargest of the four is in ny opinion specifically distinct -from the other three, and I have described it below under the name of P. alcockt. Periclimenes laccadivensts is very closely related to P. latipollex, but is distinguished by the following characters :— TExT-FIG. 20.—Periclimenes laccadivensts (Alc. and And.). a. Second peraeopod. b. Fingers of second peraeopod. c. Dactylus of third peraeopod (setae at distal end of propodus omitted). The rostrum (text-fig. 19) is less slender and is shorter, not quite reaching the end of the antennal scale; it is armed with 10 teeth above and 2 or 3 below. The hepatic spine is situated on a lower level than the antennal. The antennal scale is rather broader; about 2°75 times as long as wide in an ovigerous female, and the distal spine does not reach quite as far forwards as the apex of the lamella. The carpus of the first peraeopod is a little longer than the chela. The peraeopods of the second pair (text-fig. 20a) are distinctly unequal, but otherwise resemble those of the related species ; the dactylus, however, is not flanged along its outer edge. The armature of the cutting edges of the fingers (text-fig. 20)) 154 Records of the Indian Museum. [VoL. XXIV, is similar and in minor details is variable. There are one or two teeth on the dactylus which fit into a recess in the fixed finger, while on the fixed finger itself there are only two teeth, both rather large, in place of the three found in P. latipollex. The last three peraeopods are rather more slender. In the third pair the merus is about 11°5 times as long as wide and the propodus is 9 times as long as the dactylus. The accessory claw of the dactylus (text-fig. 20c) is small and slender; it is sometimes © missing, having apparently been broken off. The three specimens are all ovigerous females. The largest, from which the figure in the Idlustrations of the Zoology of the ‘ Inves- tigator’ is drawn, is about 27 mm. in length. The specimens are from-deep water and have a soft membranous integument. 9221/4, Laccadive Sea, 703 ‘ Investigator,’ One, Tyre. fms., 10°47'45" N., Nov., 1891. 72°40'20"E., ; 2129-30/10. Laccadive Sea, 430 ‘Investigator,’ Two, Types. finSs eal GOMNe, Oct., 1897. 76°54'30" EB. The identity of the two specimens recorded by Miss Rathbun (loc. cit.) from the Hawaiian Is. appears to me to be doubtful. Periclimenes (Periclimenes) alcocki, sp. nov. 1001. Palaemon (Brachycarpus) laccadivensis, Alcock, Cat. Ind. deep- sea Crust. Decap. Macrura and Anomala, p. 138 (in part). This species is represented in the collection by a single large specimen obtained by the ‘ Investigator ’ and referred by Alcock to Palaemon (Brachycarpus) laccadivensis. It differs from the types of the latter species and from Periclimenes latipollex in TEXT-FIG. 21.—Periclimenes alcocki, sp. nov. Anterior part of carapace, rostrum, etc. a number of particulaits which appear to entitle it to specific distinction. It may be separated from the related forms by the following characters :— The rostrum (text-fig. 21) is deep and reaches just beyond the end of the antennular peduncle. On its upper margin it bears 9 1922.] S. Kemp: Notes on Crustacea Decapoda. 155 teeth, of which the foremost and hindmost are rather remote from the rest; three posterior teeth stand on the carapace behind the orbit. On the lower margin there are 3 teeth, the foremost small and placed close to theapex. The hepatic spine is placed on a lower level than the afitennal. The two rami composing the outer antennular flagellum (text-fig. 22a) are fused basally for a much longer distance than in the related species ; the fused por- tion is 3°5 times as long as the free part of the shorter ramus and consists of 12 segments. The antennal scale (text-fig. 220) is au. TEXT-FIG. 22.—Periclimenes alcocki, TEXT-FIG. 23.—Periclimenes alcocki, sp. nov. Sp. nov. : a. Antennule. a. Larger second peraeopod. 6. Antennal scale. 6. Fingers of same leg. broader, scarcely more than twice as long as wide; the outer margin is convex and terminates in a spine which reaches nearly to the end of the lamella. The carpus of the first peraeopods is proportionately longer than in either of the related species and is 1°5 times the length of the chela. The second peraeopods (text-fig. 23a) are unequal and are closely covered throughout with small tubercles, a remarkable character also found in certain species of the subgenus Ancyloca- vis. The fingers in both limbs are almost exactly half the length 156 ; Records of the Indian Museum. [Vot. XXIV, of the palm and are thus proportionately longer than in the mitied forms, In the longer limb the dactylus is conspicuously spatulate (text-fig. 23b) and has a single large and sharp tooth in its basal third which fits into a cavity in the fixed finger when the claw is closed. There are two teeth on the fixed finger, one a little be- hind the middle, which is accommodated in a socket placed in advance of the tooth on the dactylus, and another which is blunt and molariform nearer the base. The smaller chela is similar, but there are two teeth on the dactylus -—the posterior blunt and ' inconspicuous—and one, which is small, on the fixed finger. The merus of the third peraeo- pod is about 8 times as long as broad; the propodus is rather less than 7 times the length of the dactylus. The accessory claw. of the latter is small, as in P. laccadivensis. The teison (text-fig. 24) differs from that _ of all other Pontoniinae in the possession of four pairs of dor- sal spines in addition to the six which occur at the apex. It is possible that this is merely an abnormality, but the spines are ‘TEXT- FIG, 24.—Perviclimenes alcocki, arranged symmetrically on the sp. nov. two sides. Telson. ; The single specimen, an oviger- ous female, is 50 mm. in length. 4789/7. baccadive Sea, {06 fms. ‘Investigator,’ Jan., One, Typr. 9°34'57" N., 75°36'30" E. 1895. As in the preceding species the integutaent is soft aud mein- branous. Periclimenes (Periclimenes) lanipes, sp. nov. (Plate IV, fig. 4.) The rostrum is strongly curved downwards, with the tip a little upturned. It reaches just beyond the apex of the antennal scale and in lateral view is shallow. On the strongly convex upper border it bears 8 or 9! evenly spaced teeth, decreasing in size from behind forwards and with the hindmost situated above or a little behind the posterior limit of the orbit. The lower margin is un- armed, or with a single small tooth! placed beneath the seventh or eighth of those on the upper side. In dorsal view the rostrum is broad at the base, with a carina on either side forming a sort of superciliary ridge over the upper portion of the orbit. The lower limit of the orbit is defined by a ! Of three specimens two have 8 dorsal teeth and one has g; in two specimens the lower margin bears a single tooth and in one it is unarmed. 1922.] S. Kemp: Notes on Crustacea Decapoda. 157 sharp angle, beneath which there is a strong antennal spine; the hepatic spine is behind the antennal and on a level with it. There is no supra-orbital. Immediately behind the eye the orbital margin is conspicuously depressed, forming a hollow which ap- parently serves to accommodate the eyestalk when it is directed backwards. The eyes are short and stout, with the cornea hemi- spherical and not wider than the stalk. The ocular spot touches the cornea. The lateral process of the antennular peduncle reaches about to the middle of the basal segment ; the distal spine of this segment is very long, reaching the articulation of the second and third segments. The free portion of the shorter ramus of the outer flagellum is a little shorter than the fused part, the latter compris- ing 4 or 5 segments. The antennal scale is very broad, only twice as long as wide. The outer margin is slightly convex and ter- minates in a large tooth which reaches almost or quite as far forwards as the apex of the lamella. The third maxilliped bears a small arthrobranch. The exopod reaches the end of the antepenultimate segment and the last seg- ment is three quarters the length of the penultimate. The first peraeopods reach beyond the scale by more than the length of the chela. The carpus is a little longer than the merus and considerably longer than the chela. The fingers are a little shorter than the palm and are spatulate, without teeth or spines on their inner edges. _ The second peraeopods are stout and reach beyond the an- tennal scale by fully half the length of the chela. The merus is scarcely more than 2'5 times as long as wide and bears a strong spine at the distal end of its lower border. The-carpus is conical and very short, about as long as broad and half as long as the metus; it bears no spines but is fringed with setae anteriorly and is deeply notched on the inner side of its distal margin. The heavy chela is also clothed with setae, sparsely at the proximal end, but densely in the vicinity of the fingers. The palm is 2°5 times as long as broad and is rather more than twice the length of the fingers. The fingers have inturned tips and on the inner edge of the dactylus in its proximal half there is a large acute tooth. The fixed finger is sometimes unarmed, sometimes with a small tooth in advance of that on the dactylus and with three or four serrations at the proximal end. When the claw is closed the fingers slide past one another like the blades of a pair of scissors and the large dactylar tooth is received into a socket in the fixed finger. The last three peraeopods are stout; the third pair reaches beyond the scale by more than the length of the dactylus, the fifth reach the middle of the scale. In each pair the inferior edges of the ischium of merus are thickly set with soft hairs. The lower border of the merus ends in a strong tooth, behind which there are a few spinules. The propodus is stout and is densely clothed with long woolly hairs. which, at the distal end, are so thick 158 Records of the Indian Musewm. [Vor. XXIV, as to conceal the dactylus. ‘The dactylus itself has a small tooth on the posterior margin and is strongly curved and only about one- sixth the length of the propodus. The sixth abdominal somite is very little longer than the fifth. The anterior of the two pairs of dorsal spines on the telson is situ- ated inthe middle of its length; the posterior pair is a little nearer to the apex than to the anterior pair. The external margin of the outer uropod is ciliated. The largest of the three specimens, an ovigerous female, is about 13 mm. in length. This species is clearly allied to P. latipollex, P. laccadivensis and P. aicocki, but is easily distinguished by numerous well-marked characters. C 405/1. Mergui Archipelago, 12°48! ‘ Investigator.’ One, Tyre. N., 98°16'10" E., 24 fms. The other two specimens belong to the Paris Museum and were obtained by M. Heartel at Mozambique in water 20-25 m. deep. Periclimenes (Periclimenes) rex, sp. nov. (Plate V, fig. 5.) The rostrum extends beyond the end of the antennular pedun- cle but does not reach the tip of the antennal scale. It is extreme- ly deep in lateral view and is very strongly curved downwards. The convex upper border is‘serrated like a saw and in the single specimen examined, bears 22 small equidistant teeth, with one additional tooth placed far back on the carapace and widely separated from the rest. The lower border is unarmed and is strongly convex in its distal half. There is no supra-orbital spine. The lower limit of the orbit is drawn out into a narrow pointed process, beneath which is the antennal spine. The hepatic spine is large and placed on a lower level than the antennal. In dorsal view the eyestalk is widest at the base; the cornea is rounded and scarcely wider than the stalk, on which it is set obliquely. The ocular spot is distinct and touches the cornea. The antennular peduncle reaches only to about two-thirds the length of the antennal scale. The basal segment is very broad with a short lateral process. The distal margin external to the insertion of the second segment is produced anteriorly as a round- ed lobe (text-fig. 250); this lobe bears the customary terminal spine on the outer side of its apex and extends almost as far for- wards as the articulation between the second and third segments. The external margin of the second segment is similarly produced beyond the insertion of the third segment. The free portion of the shorter of the two rami composing the outer antennular flagellum is about half the length of the fused part, the latter comprising 7 segments. The antennal scale (text-fig. 25a) is 1922. ] S. Kemp: Notes on Crustacea Decapoda. 159 broad, about twice as long as wide; its outer margin is convex and terminates in a spine which fails to reach the end of the lamella. The third maxillipeds are stout and reach nearly to the end of the basal antennular segment. ‘They possess a small arthro- branch and the ultimate segment is about two-thirds the length of the penultimate. The first peraeopods are unusually heavy and reach beyond the scale by rather more than the length of the fingets. ‘The merus is a little longer than the carpus and is about 5 times as long as broad. The carpus is 4 times as long as its distal breadth and is a little shorter than the chela. The fingers bear tufts of setae and are broadly spatulate, rather shorter than the palm. The second legs are markedly unequal in the single specimen examined. ‘The left leg, which is the larger; reaches beyond the TEXT-FIG. 25.—Periclimenes vex, sp. nov. a. Antennal scale. c. Fingers of larger second peraeopods. b. Last two segments of antennule. d. Dactylus of third peraeopod. scale by the whole length of the carpus and chela, the smaller leg by the chela only. The merus of the larger limb is rather less than 4 times as long as wide and bears a blunt tooth at the distal end on the lower side ; it is about 2°2 times the length of the carpus. The carpus is conical, scarcely longer than its greatest breadth and has a deep and narrow excavation on the upper side of its distal margin. ‘The chela is fully 1°5 times the length of the carapace and is 2°25 timesaslong asthe merus. The palmis rather less than 4 times as long as wide and is 2°5 times the length of the fingers. The fingers (text-fig. 25c) have yellow inturned claws at their tips and are beset with a multitude of fine hairs.' On the inner edge of the dactylus at the base there is a large tooth which bears against a grinding surface at the proximal end of the dactylus, and in front of this, a little behind the middle point, there is a sharp conical tooth with a rounded excavation on either side. On the fixed finger there is a sharp tooth near the middle ! Not shown in text-fig. 25c. 160 Records of the Indian Museum. (VOU; point and behind it a semicircular excavation followed by a broad lobe with small denticulations on its summit. The smaller limb is closely similar, but the teeth on the fingers are less well developed. The three posterior pairs of peraeopods are short and stout. The third teach the tip of the rostrum, the fifth the end of the merus of the first pair, The propodites bear some fine hairs but are without spinules on their posterior margins. In the third pair the merus is about 6 times and the propodus about 7 times as long as wide. The dactylus (text-fig. 25d) is broad and is less than a quarter the length of the propodus. The accessory dactylar spine is greatly reduced. The sixth abdominal somite is about 1°5 times the length of the fifth. The anterior pair of spinules on the dorsum of the telson is placed at about the middle of the telson length, and the posterior pair midway between it and the apex. The terminal Spines are short. The single individual in the collection is an adult male about 21 mm. in length. When living, the specimen was most gorgeously pigmented. The general colour was bright red; on the carapace there was a very large transverse diamond-shaped patch of pale fawn with closely aggregated cream spots, the whole patch circumscribed by deep red. The rostrum was red with minute spots of white and of white ringed with black. On each abdominal somite there was a transverse pale dorsal patch similar to that on the carapace, the patches on adjacent somites being confluent with one another. The last abdominal somite and telson were entirely pale fawn with cream-coloured spots. ‘The ‘cornea was red and the eyestalks red with whitish spots. ‘The antennal scale was pale red, similarly spotted, and with the tip broadly margined with deep purple. The first two pairs of legs were red with the distal ends of the merus and carpus and the whole of the fingers purple. The last three legs were entirely rich purple, while the pleopods were red. Periclimenes vex seems to hold an isolated position in the sub- genus, but is perhaps distantly: related to the P. laccadivensis ' section. By the form and armature of the rostrum it is readily distinguished from all other known forms. C 402/1. Port Blair, Andamans, 8 fms. S. Kemp, March, One male, 1921. TYPE, The specimen, together with a single chela of a second indivi- dual, was found in Ross Channel, near the southern end. In the same haul of the net fragments of a red sponge with white tips were taken, the similarity in colouration suggesting that the prawn and the sponge were possibly associated with one another. Periclimenes (Periclimenes) investigatoris, sp. nov. (Plate V, fig. 6.) A species of rather stout build. The rostrum is deep; it extends a little beyond the end of the antennular peduncle but 1922. } S. Kemp: Notes on Crustacea Decapoda. 161 does not reach the apex of the scale. It is quite straight, with slightly convex upper border, and bears 9g dorsal teeth in the single specimen examined. The posterior tooth is placed on the carapace behind the orbit, but is not separated from the second by a greater interval than that between the second and third; the second tooth is placed immediately above the posterior limit of the orbit. The sixth, seventh and eighth teeth are larger than the rest; the foremost is extremely small and placed: close to the apex. The lower margin is strongly convex and bears a single tooth situated below the penultimate of those forming the dorsal series. There is no supra-orbital spine. The antennal spine is sharp, with the hepatic placed behind it on a lower level. The eye is stout, with the ocular spot touching the cornea. The basal seginent of the antennular peduncle (text-fig. 26a) is broad; the spine form- ing the laterAl process is of exceptional length, reach- ing as far forwards as the articulation of the second ~ segment. The terminal spine of the outer margin is also very long, reaching the base of the third seg- ment. The second and third segments are short | and broad and the fused portion of the outer flagel- lum is composed of only four .segments. The an- tennal scale (text-fig. 26d) is not quite 2°5 times as loug as wide; the outer _ margin is straight and ter- minates in a strong tooth which does not reach the ————— end of the lamella. a b. The autepenultimate VTEXT-11G. 26.—Periclimenes investigatoris, segment of the third sp. nov. maxilliped is somewhat a. Antennule. twisted and the ultimate 6. Antennal scale. segment is shorter than the antepenultimate. The first peraeopods reach beyond the end of the scale by the length of the chela. The carpus is shorter than the merus and about equal in length with the chela; the fingers are unarmed and shorter than the palm. The second peraeopods are unequal, the left much larger than the right and reaching beyond the scale by more than the length of the chela; the two are, however, similar in structure. In the larger the merus is 1°75 times the length of the ischium 162 Records of the Indian Museum. AYO, POIs and is less than 4etimes as long as wide.’ There are no spines on either merus or carpus. The carpus is conical, more than 1'5 times as long as wide. ‘The palm is 3 times as long as broad and the fingers are about two- thirds its length. The fingers (text-fig. 27a) have large apical claws which - cross one another when the claw is shut. The cutting edge of the fixed finger bears a series of small teeth in the proxi- mal half of its length and there is one rather larger . tooth in the basal third of the dactylus. The smaller limb of the* same pair is similar, but the carpus is twice as long as wide and the fingers almost as long as the palm and without a&. b. teeth. Text-F1G. 27.—Periclimenes investigatoris, The last three pairs of sp. nov. peraeopods are rather a. Fingers of larger second peracopod. _stout.; those of the third b. Dactylus of third peraeopod. pair scarcely reach the tip $ of the scale. The propo- dus bears setae on its posterior margin; in the third and fourth pairs it is about 4°5 times the length of the dactylus and in the fifth pair about 6 times. The dactylus (text fig. 270) is curved, rather slender and with a small accessory tooth. The sixth abdominal somite is fully one and a half times the length of the fifth. The telson bears the usual two pairs of dorsal spinules, the first a little in advance of the middle, the second nearer to the first than to the apex. The external margin of the uropod is ciliated. The above description is based ona single ovigerous female 15 mm. in length. ' P. investigatoris is easily distinguished from any other species in the same subgenus by the great length of the spine forming the lateral process of the antennule. C 350/1. Persian Gulf, 13 fms., “‘TInvestigator,”’ One, Tyre. 29°20’ N., 48°47’ BE. Oct., 1905. : The specimen is labelled ‘‘ found on an Alcyonarian.’’ Periclimenes (Periclimenes) noverca, sp. nov. The rostrum (text-fig. 28) reaches a little beyond the end of the antennuiar peduncle. It is straight, but directed downwards ! ‘The merus is too slender in the figure. 1922. ] S. Kemp: Notes on Crustacea Decapoda, 163 and is rather shallow in lateral view. On the upper border there are, in the single specimen examined, 7 equidistant teeth, the hindmost well in front of the posterior limit of the orbit. The lower border is slightly convex and is unarmed. TEXT-FIG. 28.—Peviclimenes noverca, sp. nov. Anterior part of carapace, rostrum, etc. There is no supra-orbital spine. The lower limit of the orbit is acute. The antennal spine is strong, with the hepatic placed behind it ona slightly lower level. The eyes are rather slender. . b, a. TExt-F1G, 29.—Periclimenes noverca, Sp. nov. a. Antennule. c. Third peraeopod. b, Antennal scale. d. Dactylus of third peraeopod (setae omitted). The ocular spot is confluent with the cornea, which is hemi- spherical and a little wider than the stalk. The lateral process of the antennular peduncle (text-fig. 29a) reaches beyond the middle of the basal segment; the anterior margiu of this segment is greatly produced externally, the spine 164 Records of the Indian Museum. [VoL. XXIV, reaching beyond the middle of the second segment. ‘The free por- tion of the stouter of the two rami composing the outer flagellum is rather more than half the length of the fused portion, the latter comprising 5 segments. The total length of the stouter ramus is less than the length of the peduncle. The antennai scale (text- fig, 29b) is about 2°5 times as long as wide. The outer margin is _ Trext-ric. 30.—Periclimenes noverca, sp. nov. a. First peraeopod. c. Second peracopod. b. Fingers of first peraeopod. d. Fingers of second peracopod. straight and ends in a spine which does not reach as far forwards as the very broadly rounded apex of the lamella. The first peraeopod (text-fig. 30a) is unusually stout and reaches a little beyond the end of the antennal scale. The carpus is .conspicuously shorter than the merus and is only 3 times as long as its distal breadth. The chela is very nearly as long as the catpus. ‘The fingers (text-fig. 30b) are equal in length with the 1922.] S. Kempe: Notes on Crustacea Decapoda. 165 palm; each is broadly spatulate with the inner margin finely pectinate throughout. The left second leg is missing in the single specimen examined, The right (text-fig. 30c) reaches beyond the antennal scale by less than half the length of the chela. The merus is about 3 times as long as wide and only two-thirds the length of the ischium ; it bears a strong tooth at the distal end of the lower border. The carpus is short and conical, two-thirds the length of the. me- rus and about 1°6 times as long as its distal breadth. The chela is about 3°6 times the length of the carpus; the palm is a little more than 3 times as long as wide. ‘The fingers (text-fig. 30d) are rather more than half the length of the palm and have inturned tips ; the dactylus is unarmed, but there are four small teeth on the fixed finger. ‘There are long sparse hairs on all the segments. The three posterior legs (text-fig. 29c) are short and stout; the third reach about to the eud of the antennalscale. The merus in this pair is nearly 3°5 times as tong as wide and bears a strong tooth at the distal end of its lower border; the propodus is 4°5 times as long as wide and from 5°5 to 6 times as long as the dactylus. The propodus bears spinules on its posterior margin and at the distal end is thickly clad with hairs that partially conceal the dactylus. The dactylus (text-fig. 29d) is small and curved, with the accessory claw found in most species of the subgenus re- - placed by a conspicuous rounded lobe. The sixth abdotninal somite is scarcely longer than the fifth. The anterior of the two pairs of spines on the dorsum of the telson is placed at about the middle of its length, the second pair midway between the first and the apex. The single SSCs is an ovigerous female about 16 mm. in length, P. noverca is closely Panton to Nobili’s P. sovorv, but is distin- guished, as shown below, by a number of well-marked characters. The type and only known example of this species was found at New Caledonia and is the property of the Paris Museum. Periclimenes (Periclimenes) soror Nobili. 1904. Periclimenes soror, Nobili, Bull. Mus. Paris, X, p. 232. 1906. Periclimenes soror, Nobili, Ann. Sct. nat., Foul io) IV, p. 50, pl. ii. fig. 6. This species, which I have not seen, agrees with P. noverca and differs from all other members of the subgenus Periclimenes in possessing a comb of fine teeth on each finger of the first peraeo- pod. According to Nobili’s description it differs from the allied species in the following points :— (i) There are 11-13 teeth on the upper margin of the rostrum. (ii) The tooth at the outer distal angle of the basal anten- nular segment is short. (iii) The first peraeopods are more slender, with carpus 4 times as long as its distal breadth. 166 Records of the Indian Museum. [Vor. XXIV, (iv) The merus of the second leg is equal to or slightly longer than the ischium and does not bear a spine at the distal end of its lower border. (v) The fingers of the second leg are only one-third the length of the palm. (vi) The lower border of the merus of the last three legs does not end in a tooth. (vii) The dactylus of the last three legs i is piavided with a small accessory spine and is only one-ninth the length of the propodite. P. soror was described from Djibouti i in the Red Sea. Periclimenes (Periclimenes) commensalis Borradaile. 1915. Perclimenes (Cristiger) commensalis, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 211. 1915. Periclimenes commensalis, Potts, Publ. Carnegie Inst. Washing- ton, nd. 212, p. 82. 1917. Pe clinenes (Cristiger) commensalis, Borradaile, Tvans. Linn. Soc. (2) Zool. XVII, p. 364. : I have examined the type of this species and think that Borradaile is mistaken in stating that there are two spines at the _ distal end of the basal antennular segment. The margin between the outer spine and the articulation of the second segment is some- what more produced than usual, but is rounded and does not end in a spine. P. frater, Borradaile, which I refer to the subgenus Ancylocaris, appears to be the only species of the genus in which two spines occur in this position. The accessory tooth on the dactyli of the last three peraeo- pods is small and inconspicuous in this species. P. commensalis was found by Mr. Potts on Comanthus annu- latus at the Murray Is., Torres Straits. Subgenus Periclimenaeus Borradaile. 1915. Periclimenaeus, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 20 1916. ar ieitanenek subgen. Hamiger, Borradaile, Brit. Antarct. Exped. 1910, Zool. III, p. 87. 1917. Periclimenaeus, ages Trans. Linn. Soc. (2) Zool. XVII. P- 377: The species of this subgenus resemble those of Periclimenes s.s, in having the dactyli of the last three peraeopods biunguiculate, but differ in the absence of the hepatic spine of the carapace. The second peraeopods are unequal and dissimilar and the chela of the larger limb is always very massive. The status of the subgenus is precarious. ‘The three species referred to it appear to form a natural group, but the only un- equivocal point of distinction from Periclimenes s.s. is the absence of the hepatic spine. If, as is not improbable, a species is dis- covered which lacks this spine, but possesses affinities with Per:- climenes s.s. tather than with Periclimenaeus, the latter subgenus 1922.] S. Kemp: Notes on Crustacea Decapoda. 167 will have to be abandoned. It will not be possible to distinguish the subgenus in a satisfactory manner by the form of the second peraeopods, as these limbs exhibit a very wide range of variation in Pertclimenes s.s. Borradaile in proposing Hamiger, a new subgenus ot Pert- climenes, for his P. novae-zealandiae, seems to have overlooked the fact that the species is closely related to the members of his Per7- climenaeus. P. novae-zealandiae differs from the two species re- ferred to the latter genus only in minor details of rostrum and chela which are clearly no more than specific. Key to the species of the subgenus Periclimenaeus. No teeth on lower border of rostrum; inner edges of fin- gers of larger chela provided with a knob fitting into a socket. ; Two posterior teeth of upper rostral series situated on carapaee; larger chela with knob on dactylus and socket on fixed finger; fringes of setae on legs not temarkably long; R.g:o0 .., we No teeth of upper rostral series situated on carapace ; larger chela with knob on fixed finger and socket on dactylus; fringes of setae on legs remarkably long; R. 4-7:0... a er ... jimbyiatus (Borradaile). Two teeth on lower border of rostrum; larger chela with a huge bifid tooth at base, overlapping dactylus ; R. 8:2 He ia oe robustus Borradaile). novae-gealandiae ( Bor- radaile). Periclimenes (Periclimenaeus) robustus (Borradaile). 1915. Periclimenaeus robust:ts, Borradaile, Ann, Mag. Nat, Hist. (8) XV, p. 213. : 1917. Peviclimenaeus robustus, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 378, pl. lv, fig. 20. Amirante I., 29-39 fms. Periclimenes (Periclimenaeus) fimbriatus (Borradaile). 1915. Periclimenaeus fimbriatus, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 213. 1917. Periclimenaeus fimbriatus, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 379. pl. lv, fig. 19. Mulaku Atoll, Maldives. Providence I., 39-50 fms. Periclimenes (Periclimenaeus) novae-zealandiae (Borradaile). 1916. FPertclimenes (Hamiger) novae-zealandiae, Borradaile, Brit. Antarct. Exped. 1910, Zool. III, p. 87, text-fig. 4. 7 mi. E. of N. Cape, New Zealand, 70 fms. Subgenus Ancylocaris Schenkel. IT include under this subgeneric name all those species of Peri- climenes in which the dactylus of the last three legs is simple, with- out the additional claw or process found in Periclimenes s,s. and in Periclimenaeus. 168 Records of the Indian Museum. [VoL. XXIV, As a primary character in dividing the large number of species which the subgenus contains I have employed the presence or absence of a spine or tooth at the distal end of the merus of the second peraeopod. De Man has found that a similar character in the third peraeopod is of great value in the genus Alpheus. I think it probable that a primary separation on these lines is at least as likely to demonstrate the true relationships of the species as any other, but the principal specific characters are combined in so many different ways that it is impossible in the present state of our knowledge to determine which indicate affinity and which are examples of convergence. The key which follows must therefore be regarded as artificial. Key to the species of the subgenus Ancylocars. Section I. Merus of second leg without a spine or tooth at distal end of lower border. A, Supra-orbital spine present [hepatic spine pre- sent]. B. Cornea hemispherical. C. Rostrum shallow; merus of second leg longer ie carpus; R.g: 4 ... nilandensis Borr. Rostrum deep; merus and carpus of second me subequal; R. 7; 3. .. edwards (Paulson). B',. Cornea conoidal, more or less pointed distally. : C’. Rostrum not reaching end of antennular peduncle. D. Eye with conspicuous terminal papilla; R. 4:0 Bs 248 “at . ceratophthalmus Borr. D', Eye without conspicuous terminal papilla ; Se ace cornutus! Borr. C’. Rostrum reaching beyond end of antennular peduncle [eye Acs conspicuous terminal papilla]; R.6: ah ah ... ambotnensis,' de Man. ', Supraorbital spine absent a Hepatic spine present. C. Rostrum reaching far beyond end of scale ; carpus of second leg twice as long as chela; R. 6:0 WM ScD 0 . psamathe (de Man). C'. Rostrum not reaching beyond end of scale ; carpus of second leg little if at all longer than chee D. Second legs excessively long, ischium almost reaching end of scale (? in males only) ; fingers of second leg scarcely one quarter Jength of palm; R.7:0 longipes (Stimpson). D'. Second legs rarely long, ischium not near ly reaching end of scale; fingers of second leg at least one-third as long as palm. £. Carpus of second leg more than half as long as palm. F. Distal spine of antennal scale reaching to or beyond end of lamella. G. Carpus of second leg conspicuously longer than palm; dactylus of last three legs nearly one half as long as propodus. | In this species the second Tees are unknown; it is assumed from its struc- tural resemblance to P. ceratophthalmus that it falls in this section of the genus. 1922.] S. Kemp: Notes en Crustacea Decapoda, 169 H. Second leg with carpus slightly longer than chela, fingers unarmed or with one minute tooth and much longer than palm; R. 8-9: 2 ia H'. Second leg with carpus shorter than chela, fingers shorter than palm with large teeth; R. 8-9 : 4-5 ... calmani Vattersall. G’. Carpus of second leg equal to or shorter than palm; dactylus of last three legs less than one third length of propodus. H. A small papilla on eyestalk; car- pus and chela of first leg subequal ; second leg with carpus as long as palm and palm about as long as fingers; R. 7-9: 2-5... ane H'. No papilla on eyestalk; carpus of first leg longer than chela; second leg with carpus shorter than palm and palm fully twice as long as fin- 7 gers; R. 7-9: 2-3 Ws - a (Kings- ey). leptopus, sp. nov. seychellensis Borr. F'. Distal spine of antennal scale not nearly reaching end of lamella. : G. Rostrum very shallow, downcurved, with 3 posterior dorsal teeth placed on carapace; last three legs extending far beyond scale; R.g:3 ... * G'. Rostrum deep or moderately deep, straight, with at most 1 posterior dor- sal tooth placed on carapace ; last three legs not extending beyond scale, H. Upper border of rostrum very strongly convex, ventral tooth placed behind foremost dorsal tooth; R. Rf One vee tee ve H'. Upper border of rostrum straight, ventral tooth in advance of foremost dorsal tooth; R. 5:1 ... axe E'. Carpus of second leg less than half as long as palm (distal spine of antennal scalé not nearly reaching end of lamella). F. No conspicuous! comb of spines on fingers of first leg. G. Rostrum with at least ten dorsal teeth ; sixth abdominal somite more than twice length of fifth; R. 10-13 : 3 i G’. Rostrum with at most eight dorsal teeth; sixth abdominal somite less than twice as long as fifth. H. Carpus of first leg longer than chela. F. Rostrum deep in lateral view; a single spine at distal end of basal antennular segment. K. Carapace of female greatl swollen dorsally; telson wit dorsal spines very small, both pairs situated in distal half of its length; R. 5-7 : 0-2 .. brevicarpalis Schenk. tenellus (Smith). diverstpes, Sp. lov. potina Nobili. korni (1.0 Bianco). ! Under the microscope fine incisions may sometimes be detected in the cut- ting edges of the fingers of the first leg in P. diversipes (text-fig. 39), p. 182). 170 Records of the Indian Musewm. kK’, Carapace of female not swol- len ; telson with dorsal spines well developed, anterior pair sit- uated in middle or in proximal half of its length. L. Form stout; rostrum bent downwards, upper border al- most straight with foremost tooth placed very close to apex; R. 6-8 : 0-2 a L’. Form © slender; rostrum straight, upper border strong- ly convex with foremost tooth not placed close to apex, R. Sig One F'. Rostrum very shallow in lateral view ; two spines at distal end of basal antennular segment ; R. 6:1 tae Carpus of first leg ‘about half ed of chela; R. 6:3 . Each finger of first leg with a conspi- cuous comb of spines {two spines at dis- . tal end of basal antennular segment ] ; R. 12:0 Hepatic spine absent. ie Second legs shorter than first; R. 5:0 C’. Second legs longer than first; R 6: 2 [Vor XXIV, tnornatis, Sp. nov. diversipes. sp. nov. brocketti Borr. compressus Bort. Frater Borr. brevinaris Nobili. pusillus Rathbun. Section II. Merus of second leg with a spine or tooth at distal end of lower border. A, Each finger of first leg with a conspicuous comb of spines [hepatic spine present]. &. Supra-orbital spine present fother characters as in P. petitthouarsi]; R. 6-9 : 2-5 5’, Supra-orbital spine absent. C. Merus of second leg with one spine below, car- pus withtwo terminal spines, inner margin of each finger with a large oval pit; R. 6-9 : 3-5 C’. Merus of second leg with four’ spines below, carpus with three terminal spines, inner margin of each finger with a series of small denticles ; So 7 CN Bere ‘, Fingers of first leg without a comb of ‘spines. es ae -orbital spine present. . Hepatic spine present, a Distal spine of antennal scale projecting far beyond end of lamella. E. Rostrum shallow ; last three legs long and slender, third pair with merus at least 11 times as long as broad, fifth pair reaching beyond scale. Ff. No conspicuous terminal spine on inner side of carpus of second legs. G. Carpus of first leg at least 1°75 times as long as chela; chela of second leg in males not more than 1°25 times, in fe- males equal to or a little shorter than carpus; R. 7-9: G', Carpus of first ie less than 1° 5 times as long as chela; chela of second leg spiniferus de Man. petitthowars: (Audou- in), denticulatus Nobili. agag, Sp. nov. 1922. ] S. Kempe: Notes on Crustacea Decapoda. 175 more than 1°3 times as long as carpus in both sexes; R. 6-7 : 2-3 F'. A conspicuous terminal spine on inner side of carpus of second leg. G. Rostrum usually with 8 or more dor- sal teeth ; carpus of second leg of male about equal to or shorter than merus ; R. 7-9: 2-4 ... wick Se G’. Rostrum with 6 or 7 dorsal teeth ; carpus of second leg of male conspi- cuously longer than merus; R. 6-7 : 2 E’, Rostrum moderately deep; last three legs stouter, third pair with merus at most Io times as long as broad, fifth pair not reaching end of scale. F. No spine at distal end of carpus of sec- ond leg; R. 7-8: 3 tee AE F', Atleast one spine at distal end of car- pus of second Icg. G. Only one spine at distal end of car- pus of second leg, situated on inner side. H. Foremost pair of dorsal spines of telson situated in anterior half of tel- son-length; R. 6-10 : 2-5 2B H'. Both pairs of dorsal spines of tel- son situated in posterior half of tel- son-length; R.6: 4. ... at G'. Twospines (at least in males) at dis- tal end of carpus of second leg, one on inner side and one above. H. Carpus of second leg 3 to 6 times as long as distal breadth ; propodus of last three legs with 2 irioles on posterior border. F. Carpus of second leg of male about 6 times as long as wide, slightly longer than merus; R. :2 ead Carpus of second leg of. male not more than 4'5 times as long as wide, shorter than merus. { R. 6-8 : 3-5 R. 6-7 : 3-4 nS H'. Carpus of second leg (in female) scarcely 2°5 times as long as distal breadth ; propodus of last three legs without spinules on posterior border ; R.8:3 Sie Py nA D'. Distal spine of antennal scale not project- ing beyond end of lamella (a terminal spine on inner side of carpus of second leg]; R. 7-9: 13 ‘ ene see C’. Hepatic spine absent ... na B’. Supra-orbital spine absent. C. Hepatic spine present. D. Rostrum reaching far beyond antennal scale with at least 6 ventral teeth ; R. 9-12 : 6-9 D’. Rostrum reaching little if at all beyond an- tennal scale with at most 2 ventral teeth. E. Antennular peduncle reaching beyond an- tennal scale, its last two segments extreme- ly long and slender; no ventral teeth on rostrum; R.6:0... , proximus, Sp. nov. andamanensis, Sp. nov. suvadivensis Borr. ensifrons (Dana). grandis (Stimpson). witrensts Borr. affinis Borr. . elegans (Paulson). holmesi Nobili. amymone de Man. demani Kemp. lifuensis Borr. tenuipes Borr. longimanus (Dana). 172 Records of the Indian Museum. [VoL. XXIV, &’, Antennular peduncle not reaching end of antennal scale, its last two segments of normal proportions; at least one ventral tooth on rostrum. fF, Rostrum with at least 9 dorsal teeth ; carpus of second leg unarmed. G. Carpus and merus of second leg equal in length and longer than palm ; propodus of third leg little more than twice as long as dactylus; R. 11:2... digitalis, sp. nov. G'. Carpus of second leg very much shorter than either merus or palm ;- propodus of third leg fully 4 times as long as dactylus; R. g-10: 1 ... byocki (de Man). fF’, Rostrum with only 6 dorsal teeth; car- pus of second leg with distal spine [car- pus and palm of second leg subequal] ; R.6:2 i ee ... votumanus Borr. C’. Hepatic spine absent; R. 3-4: 1 ee gevlachei Nobili.! Periclimenes (Ancylocaris) nilandensis Borradaile. 1915. Periclimenes (Falciger) nilandensis, Borradaile, Ann. Mag. Nat’ Hist. (8) XV, p. 211. 1917. Peviclimenes (Falciger) nilandensis, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 372, pl. liv, fig. 13. S. Nilandu Atoll, Maldives. Periclimenes (Ancylocaris) edwardsi Paulson. 1875. Anchistia edwardsi, Paulson, Crust. Red Sea, p. 114, pl. XVil, hg. 2. 1906. Anchistia edwardsi,-Nobili, Ann. Sct. nat., Zool. (9) 1V, p. 53: Red Sea. *Periclimenes (Ancylocaris) ceratophthalmus Borradaile. 1915. Periclimenes (Corniger) ceratophthalmus, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 211. © 1917. Periclimenes (Corniger) cevatophthalmus, Borradaile, Trans. Linn, Soc. (2) Zool. XVII, p. 365, pl. liv, figs. 9a, 0. Male Atoll, Maldives, on crinoid. Periclimenes (Ancylocaris) cornutus Borradaile. 1915. Periclimenes (Corniger) cornutus, Borradaile. Ann. Mag. Nat. Hist. (8) XV, p. 211. 1917. Periclimenes (Corniger) cornutus, Borradaile. Trans. Linn. Soc. (2) Zool. XVII, p. 365, pl. liv, figs. toa, b. Male Atoll, Maldives, on red and brown crinoid. Periclimenes (Ancylocaris) amboinensis (de Man). 1887. Anchistia amboinensis, de Man, Arch, Naturgesch. LIM, i, p. 546, pl. xxiia, figs, 2, 2a,d. : 1 This species belongs to the genus Harpilius, but is included here as it is very likely, to be confused with members of the subgenus Ancylocavis. 1922. ] S. Kemp: Notes on Crustacea Decapoda. 173 Both this species and P. cornutus were described from speci- mens in which the second peraeopods were missing; it is thus not altogether certain that they are properly referred to the subgenus Ancylocaris. They appear, however, to be closely related to P. ceratophthalmus, in which the merus of the second peraeopod is unarmed. P. amboinensis was describe from Amboina,. Periclimenes (Ancylocaris) psamathe (de Man). 1902. Urocaris psamathe, de Man, Abhandl. Senck. ALINE Ges. XXV, p: 816, pl. xxv, figs. 51, 51a-7. 1917. Uvocaris psamathe, Borradaile, Tvans. Linn. Soc. (2) Zool. XVII, Pp. 323. I have examined a specimen of this species in the Cambridge Museum and am able to state that it does not possess a mandibular palp. P. psamathe must thus be referred to the subgenus Ancylo- carts, in which, however, by reason of the remarkable character of the rostrum and second peraeopod, it occupies a very isolated position. The species was desctibed from Ternate and has since been recorded by Borradaile from N. Male Atoll in the Maldives and from Diego Garcia in the Chagos Archipelago. Periclimenes (Ancylocaris) longipes (Stimpson). 1860. Urocaris longipes, Stimpson, Proc. Acad. Sci. Philadelphia, P- 39. Stimpson remarks that the end of the ischium of the second leg in this species reaches almost to the apex of the antennal scale; it is thus probable that this leg is proportionately even longer than in such extreme forms as P. agag and P. tenutfes. Stimpson’s specimen was no doubt a male and, on analogy with other long-limbed species, it may be expected that the female does not possess such an inordinate length of leg. The species was found near Ousima I. at a depth of 20 fathoms. Periclimenes (Ancylocaris), leptopus sp. nov. A species of slender habit with long legs. The rostrum (text- fig. 31) is straight and reaches to the end of the second or middle of the third segment of the antennular peduncle. It is armed above with 8 or 9 teeth,' the hindmost of which is separated by a considerable interval from the next of the series and is situated on the carapace behind the posterior limit of the orbit. On the lower margin there are 2 teeth, smaller than those on the upper margin and situated in the anterior third of the rostral length, beneath the “wo foremost of those comprising the dorsal series. There is no supra-orbital spine. The hepatic and antennal ! In two specimens there are 8 teeth and in one g. 174 Records of the Indian Museum. [VoL. XXIV, are both sharp and are situated nearly on a level with one another. The eyes are large and stout, slightly flattened dorso-ventrally, and the ocular spot is confluent with the cornea. Text-Fic. 31.—Periclimenes leptopus, sp. nov. Anterior part of carapace, rostruin, etc. The basal segment of the antennular peduncle is broad; the terminal spine of the outer margin is short and the lateral process reaches about to the middle of the seg- ment. The tworami composing the outer antennular flagellum are fused for a dis- tance almost equal to the total length of the peduncle, the fused portion consisting of 8 to 10 elongate segments. The free portions of both rami are extremely short. The antennal scale (text-fig. 32) is narrow, nearly 4 times as long as wide. The outer margin is slightly concave and terminates in a spine which reaches a little beyond the apex of the lamella. The ultimate segment of the third maxil- liped is scarcely more than three-quarters _., the length of the penultimate. All the sa eae es peraeopods are very slender. ‘The first pair Fe ’ “teaches beyond the scale by the length of Amieanalseale the chela. The. carpus is a little shorter than the merus and a little longer than the chela ; the fingers have simple cutting edges and are about one-fifth longer than the palm. The second peraeopods (text-fig. 33a) extend beyond the scale by the whole length of the chela and carpus. All the segments are unarmed. The carpus considerably exceeds the length of the carapace (rostrum excluded) ; it is longer than the merus in the proportion of 4 to 3 and is fully 2°5 times as long as the palm. Its breadth at the distal end is about one-tenth its length. The chela is intermediate in length between the merus and carpus and the breadth of the palm is rather more than one-third its length. The fingers are straight with terminal claws that cross 1922. ] S. Kemp: Notes on Crustacea Decapoda. 175 one another when the chela is shut; they are without teeth on their cutting edges and are nearly 1°5 times as long as the palm. The last three peraeopods are extremely slender. The third pair (text-fig. 33b) reaches beyond the apex of the scale by two-thirds the length of the propodus, the fifth by about half the length of the propodus. There are some setae at the distal ends of the propodites, but no spinules on their posterior margins. The dactylus in each pair is very slender, simple, slightly curved and almost half the length of the propodus. TExt-ric. 33.—Periclimenes leptopus, sp. nov. a. Second peraeopod. b. Third peraeopod. The sixth abdominal somite is about 1°5 times the length of the fifth. The anterior pair of dorsal spinules of. the telson is placed a little in front of the middle point. The intermediate pair of apical spines is very long, about one-third the length of the telson. The outer uropod is nearly three times as long as wide. The largest specimen is an ovigerous female about 12} mm. in total length. P. leptopus is easily distinguished by the proportions of the segments of the second peraeopods and by the comparatively great length of the dactyli in the last three peraeopods. C. 354/1. Port Blair, Andamans. S. Kemp, Feb., Three (two ovig.), TQI5- Types. 176 Records of the Indian Museum. [VoL. XXIV, The specimens were caught in Brigade Creek in a net hauled over a bottom composed of decaying vegetation at a depth of 2-5 fathoms. Periclimenes (Ancylocaris) calmani ‘Tattersall. 1921. Periclimenes calmani, Tattersall, Fuurn. Linn. Soc., Zool. XXXIV, p. 385, pl. xxvil, fig. 11; pl. xxviii, figs. 14-15. The characters given for this species in the key on p. 169 are not all included in the description referred to above. Dr. Tatter- sall has, however, kindly informed me that the spine at the distal end of the antennal scale reaches to or very slightly beyond the apex of the lamella and that the dactylus of the last three peraeo- pods is simple. Asin P. leptopus the dactylus of these limbs is very long, about two-fifths the length of the propodus. The species was described by Tattersall from the Sudan coast. Periclimenes (Ancylocaris) seychellensis, Borradaile. (Plate VI, fig. 7.) 1915. Periclimenes (Falciger) seychellensis, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 212. 1917. Periclimenes (Falciger) seychellensis, Borradaile. Trans. Linn. Soc. (2) Zool. p. 375, pls. liv, lv, figs. 14 a-7. The rostrum reaches to, or a little beyond the apex of the antennal scale and is deep in lateral view with a concave upper border. Dorsally it bears from 7 to g teeth,' usually 8. ‘The two hindmost teeth are situated on the carapace behind the orbit and are separated by a rather wide interval, the first being only a little in advance of the middle of the carapace. On the lower border there are from 2 to 5 teeth,” usually 3 or 4. The foremost teeth, both dorsally and ventrally, are placed close to the tip.. The supra-orbital spine is absent; the hepatic is present and is situated on a lower level than the antennal. The eyes are rather slender, with hemispherical cornea. On the upper and anterior aspect of the stalk there is a small conical papilla, situated close to the cornea but separated from it by a shallow excavation. The development of the papilla is a little variable; as a rule it is quite conspicuous (text-fig. 34a), occasion- ally it is small and rarely it is almost indistinguishable, though the excavation is always distinctly seen when the eye is viewed from the proper angle. The ocular spot touches the cornea and is large. The cornea itself is traversed by two parallel wavy bands of dark pigment which are conspicuous in life and can often be detected in well preserved specimens. ! OF sixty specimens seven have 7 dorsal teeth, forty have 8 and thirteen have 9. 2 Of sixty specimens one has 2 ventral teeth, nineteen have 3, thirty-four have 4 and six have 5. : 1922. ] S. Kempe: Notes on Crustacea Decapoda. 177 The lateral process of the antennule reaches to the middle of the basal segment and the terminal spine of this segment is‘ well developed (text-fig..34b). The second and third segments are rather slender and subequal. The free portion of the shor- ter ramus of the outer an- tennular flagellum is about two-thirds the length of the fused basal part, the latter comprising from 5 to 7 seg- ments. The antennal scale (text-fig. 34c) is 3, or rather more than 3 times as long as wide. The outer margin is usually a little concave and ends in a spine which reaches almost or quite to the end of the lamella. The third maxilliped bears on arthrobranch: the ulti- mate segment is about two- thirds the length of the penultimate. The first per- aeopods (text-fig. 35a) reach about to the end of the antennular peduncle. The merus, carpus and chela , Sb. Cc. are subequal in length Texrt-ric. 34.—Periclimenes seychellensis and the fingers, which are Borradaile. unarmed, are about 1°25 a. Eye. b. Antennule. times the length of the c. Antennal scale. palm. The second peraeopods (text-fig. 350) are shorter than usual, extending beyond the scale by not more than half the length of the chela. There are no distal spines on the merus or carpus. I1n large specimens the carpus is a little shorter than the merus and equal to or slightly shorter than the palm; in smaller individuals the carpus is proportionately rather longer. The palm is a little’ inflated and is as long as or a trifle longer than the fingers. The fingers have inturned tips and straight cutting edges, with one or more small teeth at the proximal end. The last three pairs of peraeopods (text-fig. 35c) are slender and short, the third reaching by only a dactylus-length beyond the eye. The propodus bears a series of slender spines, frequently arranged in pairs, on its inferior margin and is from 3°5 to 4 times the length of the dactvlus. The dactylus is moderately curved and is simple. The sixth abdominal somite is 1°5 times the length of the fifth. The two pairs of dorsal spines on the telson are large and are placed so as to‘divide its length into three equal parts. The 178 Records of the Indian Museum. [Vor. XXIV, apex of the telson is sharply pointed and the intermediate pair of terminal spines is long. Large females from the Gulf of Manaar reach a length of about 19 mm. ‘ThoseI have seen from other localities are smaller, none exceeding 14 mm. : Specimens from the Andaman Is., when alive, were closely mottled with pale buff, lichen-green and brown. Borradaile does. not mention the curious papilla on the eye- . stalk, but I have examined his type-specimens and find that it is present. TEXxtT-riG. 35.—Periclimenes seychellensis Borradaile. a. First peraeopod. 6. Second peraeopod. c. Third peraeopod. The specimens in the collection are from the following locali- ties :— _ C€ 355/1. Ain Musa, G. of Suez. R. B. S. Sewell, One. . 1916. C 3560/1. Tor, G. of Suez. ' ditto Ten. C 357/1. Kilakarai, G. of Manaar. S. Kemp, Feb., Nineteen. 1913. C 358/1. Pamban, G. of Manaar, ditto F fteen. C 359/1. Port Blair, Andamans. S. Kemp, Feb., Many. 1915; Feb., March, 1921. The species was described by Borradaile from Praslin, Sey- chelles. The great majority of the specimens in the collection are ovi- gerous females. In the localities where I myself have found it, the species was taken among weeds in shallow water. 1922.] S. Kemp: Notes on Crustacea Decapoda. 179 Periclimenes (Ancylocaris) americanus (Kingsley). 1878. Anchistia americana, Kingsley, Proc. Acad. Sci. Philadelphia, 1882. Anchista americana, Kingsley, Bull. Essex Inst. X1V, p. 109 . ii, fig. 10. 1901. Fete oe opt. the Rathbun, Bull. U. S,. Fish Comm. XX, il, p. 121. This species is related to P. seychellensis, but differs in the following points :—(i) the rostrum is shallow and its upper border is nearly straight ; (ii) there is no papilla on the eyestalk ; (iii) the antennal scale is proportionately narrower ; (iv) the first perae- opods are much longer, extending beyond the scale by the length of the chela; (v) the carpus of these legs is conspicuously longer than the chela ; (vi) the second petaeopods are much longer, ex- tending beyond the scale by the whole of the carpus and chela ; (vii) the carpus in these legs is shorter than the palm and the. fingers are less than half as long as the palm; (viii) the last three peraeopods are much longer, the third reaching beyond the scale by nearly half the length of the propodus. J The species is known from the West Indies, Yucatan, Florida and the Bermudas. The specimens I have seen are from the last “named locality. Periclimenes (Ancylocaris) tenellus (Smith). 1882, Anchistia tenella, Smith, Bull. Mus. Comp. Zool. Harvard X, Pp. 55, pl. ix, fig. 1. N. W. Atlantic, 32°7’ N., 78°37’30” W., 229 fathoms. Periclimenes (Ancylocaris) diversipes, sp. nov. The rustrum (text-fig. 36) varies considerably in length. Usually it reaches to the end of the second segment of the anten- TExT-F1G. 36.—Periclimenes diversipes, sp. nov. Anterior part of carapace, rostrum, etc. nular peduncle; sometimes it is shorter, reaching only to the end of the first segment, sometimes longer, reaching the end of the e 180 Records of the Indian Museum. — [Vor XXIV, . peduncle. The upper portion of the blade is strongly arched and the rostrum is consequently deep in lateral view ; the lower margin is straight at the base and slightly convex near the tip. On the convex upper border there are from 5 to 7 teeth,’ most commonly 6. The hindmost of these is usually situated on the carapace behind the orbit and is not separated by any con- siderable interval from the next of the series. On the lower bor- der there are from 0 to 2 teeth,’ usually 1. The precise position of the ventral teeth is variable; the single tooth which the major- ity of specimens possess is situated below the ultimate or penul- timate member of the dorsal series. There is no supra-orbital spine. The hepatic spine is placed some distance behind the antennal and is on a level with it. The eye is moderately stout, with the stalk wider than the cornea. The . ocular spot is placed close to the cornea but is separate from it. The lateral process of the antennule (text-fig. 37a) reaches to the middle of the basal segment ; the distal tooth of the outer margin is slender. The free portion of the shorter ramus of the outer flagellum is much less’ than half the length of the fused portion, the latter comprising 7 to 9 segments. The antennal scale (text-fig. 376) is. from 2°5 to 2°75 times as long as wide; the outer margin is straight, ter- minating in a spine which is far exceeded by the narrowly pointed apex of the lamella. There is a small arthro-° branch on the third maxil- liped. ‘The first peraeopods (text- fig. 39a) reach about to the end of the antennal scale. The carpus is about equal in length with the merus and is from 1°3 to 1'6 times as long as the chela. The fingers bear some tufts of setae TEXT-FIG. 37.—Periclimenes diverstpes, Sp. nov. and are almost or quite as a. Antennule, long asthepalm. They are b. Antennal scale. somewhat spatulate; under . ; all ordinary magnifications their cutting edges appear to be entire, but when viewed under a high power of the microscope the edge is sometimes seen to be | Of ninety-six specimens twenty-seven have 5 dorsal teeth, forty-eight have 6 and twenty-one have 7. 2 Of ninety-six specimens nineteen have no ventral teeth, seventy-five have 1 ventral tooth and two have 2 teeth, © i 1922.] S. Kemp: Notes on Crustacea Decapoda. 181 divided by fine incisions into series of blunt-tipped teeth (text- fig. 39b). 5 The second peraeopods are usually unequal and are remarkable in that they exhibit four distinct types of structure within the limits of the species. The segments are always unarmed. In type a, the most highly developed form (text-fig. 38a), the limb may reach beyond the scale by more than half the length of the chela. The carpus is conical, little longer than broad and not much more than half the length of the merus. The chela is from 2°7 a. a. b. oe TEXT-FIG. 38.—Periclimenes diversipes, Sp. NOV. The four types of second peraeopod. to 3°3 times as long as the merus and the fingers are less than half the length of the palm. The dactylus is normal in form with straight or nearly straight cutting edge and an inturned tooth at ‘ the tip. The fixed finger is very strongly curved and is provided at the apex with a short chisel edge with a blunt tooth at either end, opposed to the distal part of the dactylus. Owing to the strong curvature of the dactylus the fingers gape considerably when the claw is closed. The second peraeopod of type b (text-fig. 38b)' may be nearly as large as type a. Thecarpus is more slender and the | The chela is viewed a little obliquely and its full breadth is consequently not shown. 182 Records of the Indian Museum. [VoL. XXIV, chela is at most twice the length of the merus; the fingers are both normal in form and are about two-thirds the length of the palm. Type c (text-fig. 38c)' is closely similar to type 0, but the fingers are equal in length with the palm. The whole limb is smaller and is frequently not solong as the first peraeopod. Type d (text-fig. 38d) differs widely from any of the others ; it is shorter than the first peraeopod and just as slender. The carpus is very slender, more than 2'5 times as long as wide and the fingers are twice or rather more than twice as long as the palm. On the inner face of the chela in types b,c and d the fingers tend to be hollowed out or spooned and this feature is particularly noticeable ind. I give below, at the end of this description, some notes on the different ways in which these four types of second peraeopod are combined to form a pair. The last three pairs of peraeopods are rather slender; the third pair (text-fig. 39c) reaches to or a little beyond the end of the antennular peduncle. The propodus bears some fine setae, but ex- cept for one, rarely two, at its distal end the posterior edge is devoid of spinules. The dacty- lus is moderately stout, simple and curved; itis from one-third to one- fourth the length of the propodus. The sixth abdominal somite is about I°7 times the length of the fifth. The anterior of a Birst peraeopad: the two pairs of dorsat b. Cutting edge of finger of first peraeopod, spinules of the telson is very greatly enlarged. situated a little behind Gayl hixdsperacapod. the middle of its length ; the posterior pair is midway between the anterior pair and the apex. Adult specimens do not exceed 11 mm. in length; those from the Andamans are decidedly smaller than those from the Gulf of Manaar. In life the species is transparent, sometimes with short streaks of red pigment on the eyestalk, carapace, sides of abdomen and Sapers P. diversipes is es related to Nobilis P. piotina and to P. inornatus, sp. nov.; the differences are explained below (pp. 184 and 194). Cc. Text-Fic. 39.—Periclimenes CELE RS) sp. nov. 1 The carpus is sometimes rather more slender than in this figure. ? 1922.] S. Kemp: Notes on Crustacea Decapoda. 183 C 364-5/1. Kilakarai, Gulf of Manaar. S. Kemp, Feb., Forty-four, includ- 1913. ing TYPES. C 366/1. Port Blair, Andamans, S. Kemp, March, About one hun- 1915; March, dred. 1921. In the Gulf of Manaar the species was caught at low water by working a hand-net among corals belonging to the genus Montipora, At Port Blair it was obtained by precisely similar methods from a large Alcyonarian belonging to the family Alcyon- idae. When the net was worked elsewhere no prawns were cap- tured and this fact leads me to believe that there is a real asso- ciation between the Carids and the Coelenterates on which they were found. The diversity of form in the second peraeopods is a very remarkable feature of this species. The largest specimens, as noted above, are from the Gulf of Manaar and of these the great ‘majority are ovigerous females. The collection from this locality has unfortunately suffered damage and only a comparatively small number of individuals possess both the second peraeopods. The collection from Port Blair contains very few ovigerous speci- mens; the majority are young and it is possible that the characters of the second legs would undergo modification with further growth. In specimens in which both the second legs are present the combination of structural types which go to form a pair is as follows :— | Types of NUMBER OF SPECIMENS. structure found in a pair.! x > ~ ~ VP i wn: w G.of Manaar. eae Andamans. Legs of types 6 and ¢ show a certain amount of variation and it is sometimes a little dificult to distinguish between them, Those of types a and d, on the other hand, appear to be very constant ; they show little variation and can always be recognised at a glance. Legs of type a are invariably 1ssociated with those of type d and the specimens which possess this combination are all ovigerous ‘ ! The letters in this column refer to the description on pp. 181, 182 and to the figures in text-fig. 38. 184 Records of the Indian Museum. [VoL. XXIV, females. The numbers are unfortunately low but there are numer- ous detached legs of type a in the Manaar collection. Type 0 is most commonly associated with type c; the com- bination occurs in both collections and the specimens are often ovigerous.: From the Andamans there are two examples of 0b, both females, and from the Gulf of Manaar a few 6d, all males. Legs of type c, when not combined with b, are associated with d or with another limb of type c. The combination cc ap- pears only in the Andaman collection, where it is very abundant in males and young females ; cd is found in the Manaar collection in one male and three ovigerous females. The combination dd is found only in two males from the Gulf of Manaar. Although the specimens on which these observations are based are numerous, any speculations on the significance of this remarkable diversity in the form of the legs would at present be unprofitable. Further large collections of adults are necessary to. provide more accurate data and valuable clues may be expected from field observations and from a knowledge of the relations that exist between the prawn and its hosts I will only remark here that I regard it as almost certain that legs of type c in course of growth reach type 0 and that it is not improbable that type b may develop into type a. One point remains to be meutioned—the very curious differ- ences between the two collections of specimens. The combina- tion cc, to which the majority of the Andaman specimens belong, is not represented in the Manaar series, while type d, found in a large proportion of specimens from the latter locality, occurs only in one individual (in the combination ad) from the Andamans. Had # not been for this last specimen I should have been doubt- ‘ ful whether the Andaman form did not belong to a separate race or subspecies. On’ the information available I am satisfied that all are properly attributed to a single species. The only difference between the two sets of specimens lies in the types of second peraeopod which are combined to form a pair. This may be concerned with the different hosts on which the two series were found and it wil be noticed that all four types of leg occur in both collections. Periclimenes (Ancylocaris) potina Nobili. 1905. Periclimenes potina, Nobili, Bull. Mus. Paris X1, p. 159. 1900. Periclimenes potina, Nobili, Bull. sci. France Belgique XL, p. 44, pl. iii, fig. 8. This species appears to be related to P. diversipes but, ac- cording to Nobili’s description, is distinguished by the form of the rostrum aud the proportions of the segments of the second peraeopod. The upper portion of the rostrum is not strongly arched, the posterior dorsal tooth is not situated on the carapace and the single tooth on the lower border is in advance of the fore- most on the upper border. The carpus of the second peraeopod 1922.] S. Kempe: Notes on Crustacea Decapoda. 185 is scarcely more than a quarter the length of the chela and the fingers are longer than the palm. In specimens of P. diversipes in which the carpus is very short, the paim is always longer than the fingers. P. potina was described by Nobili i three specimens ob- tained in the Persian Gulf, 16°35’ N., 54°26’ E., on floating x brown seaweed. Periclimenes (Ancylocaris) korni (Lo Bianco). 1903. Anchistia kornii, l.o Bianco, Mitt. sool..Stat. Neapel XVI, p. 250, pl. vii, fig. 13. ? 1910. Periclimenes Rorni, Kemp, Fourn. Marine Biol. Assoc. VALU, p. 4it. Near Capri, Mediterrannean, about 600 fathoms. ? Bay of Biscay, 412 fathoms. Periclimenes (Ancylocaris) brevicarpalis (Schenkel). (Plate VI, fig. 8.) ? 1880. Nicht bestimmte Palaemonide, Richters, in Mobius’ Meeresfauna Mauritius, pl. xviii, fig. 10. ? 1893. Palaemon sp., Saville- Kent, Barrier Reef of SEE SGS Pp: 145, col. pl. ii. 2? 1894. ealgarianeile amboinensis, Zehntner, Rev. szisse Zool. Il, p. 206, pl. ix, figs. 27, 27a. 1898. Aithynis sp., Cc outiére, Bull. Afus. Paris \V, p.108. 1902. Ancylocaris brevicarpalis, Schenkel, Verh. naturf. Ges. Basel XIII, p. 563, pl. xiii, figs. 21a—m. 1904. Palaemonellaabervans, Nobili, Bull. Mus. Paris X, p. 233. 1905. Harpilius lativostris, \enz, Abhandl. Senck. naturf. Ges. XXVII, p. 380, pl. xlvii, figs. 14, 14a-c. 1906. Ancylocaris aberrans, Nobili, Bull. sci. France Belgique XL, p: 52. pl. iv, figs. 9, 9a, b. 1906. Ancylocaris aberrans, Nobili, Ann. sei. nat., Zool. (9) IV, p. 1914. Periclimenes heymitensis, Rathbun, Proc. Zool. Soc. London, p. 655, pl. i, figs. 1-3. 1916. Ancylocarts aberrans, Kemp, Rec. Ind. Mus. XII, p. 389. 1917. Ancyclocaris abervans, lativostris, hermiitensis, Bronstagalis Borradaile, 7vans. Linn. Soc. (2) Zool. XVII, pp. 355, 356. Four specific names have been applied to brilliantly coloured Pontoniine prawns which are found living in association with giant anemones belonging to the genus Dtscosoma, but it appears to me improbable that more than one such species is at present known. Borradaile, who has summarized the characters by which the four described forms are distinguished, remarks on the difficulty of separating them and suggests that some will eventually have to be united. This is the more probable since the species, being assigned to four different genera, were originally described without any thought of comparison with one another. A series of specimens from Indian waters shows that the differential characters employed by Borradaile do not possess specific value. Though the normal variation is not great, it is 186 Records of the Indian Museum. [VoL Kachy, sufficient to account for all or nearly all the differences he has noted. The descriptions themselves do not indicate other fea- tures-of any importance and it is clear that if Nobili’s aberrans, Lenz’s lativostris and Miss Rathbun’s hermitensis are to be re- tained as distinct, it must be by reason of fresh and hitherto undis- closed characters. While in Paris in 1920 I was unable to examine the type of Nobili’s Palaemonella aberrans from Djibouti, as the specimen had unfortunately been mislaid; but, through the courtesy of Prof. Ch, Gravier, I was able to see the female from Bahrein I. in the Persian Gulf which Nobili subsequently referred to the same species. In the figure of this specimen (doc. ci¢., 1906, pl. iv, fig. 9) ‘the dorsal swelling of the carapace is very greatly exaggerated, and the statement that a podobranch occurs on the second maxil- liped is erroneous. P. brevicarpalis.in my estimation is a species of very wide distribution, extending from the Red Sea and east coast of Africa to the Santa Cruz Is. in Oceania. I have examined a specimen from the last named locality and have compared examples from the Torres Straits with the series in the Indian Museum collection. I am convinced that all belong to a single species. -The rostrum varies considerably in length. As a rule it reaches to or a little beyond the end of the antennular peduncle ; rarely it is shorter, sometimes extending only to the middle of the second peduncular segment. In lateral view it is deep, with convex upper and lower margins aind at the apex it is sometimes a trifle upturned. On the upper margin there are from 5 to 7 teeth! usually 6, which are for the most part evenly spaced. In about half the specimens the posterior dorsal tooth is situated a little behind the back of the orbit; in most of the others it is immedi- ately above this point, while very rarely it is placed further forward. The distal upper tooth is not so near to the tip as to give it a bifid appearance. On the lower margin there are 1 or 2 teeth,* nearly always 1; these are scarcely smaller than those on the upper mar- gin and are situated in the distal third of the rostrallength. Very rarely specimens are found with the lower border unarmed. The strong curvature or swelling of the dorsum of the cara- pace is only seen in large females; in males, and in females that are small or of moderate size, there is scarcely an indication of it.* 1 Of fifty-one specimens fifteen have 5 dorsal teeth, thirty-two have 6 and four have 7. 2 Of fifty-one specimens one has the ventral margin unarmed , forty-six have I ventral tooth and four have 2 teeth. * Only fourteen of the sixty-two specimens in the collection possess this swol- len carapace. That the feature is not shown in Schenkel’s figure is sufficiently explained by his statement,—*‘ der Cephalothorax war wie es scheint etwas aufge- ; trieben, namentlich auf der Oberseite ; leider hat er sich, der Weichheit des Tegu- mentes halber, nicht gut conserviert.’’ In Nobili's figure, as | have remarked above the character is greatly exaggerated. 1922. | S. Kemp: Notes on Crustacea Decapoda. 187 The antennal and hepatic spines are well developed, the latter being on.a much lower level than the former.! The eye is small and slender. In dorsal view the stalk is swollen at the base and broader than the hemispherical cornea.* There is a small ocular spot, placed close to the cornea but isolated frawivite: ©. The lateral process af the basal segment of the antennule (text-fig. 40a) reaches a little beyond the middle of the segment. Distally the basal segment projects beyond the articulation with the second and bears a small spine ex- ternally. The free portion of the shorter ramus of the outer flagellum is rather longer than the fused part, the latter consisting of 5 to 9 seginents, most commonly 5 or 6. The outer margin of the antennal scale (text- fig. 40b) is slightly convex, - terminating in a small spine which does not reach nearly as far forwards as the some- what pointed apex of the lamella. In largespecimens a b. the scale is rather less than 2'5 times as long as wide. Text-ric. 40.—Periclimenes brevicarpalis (Schenkel). The second maxilliped a. Antennule. does not possess a podo- 6. Antennal scale. “ branch. The third maxil- liped is short and slender ; it bears an arthrobranch and the exopod reaches to the middle of the penultimate segment. The ultimate segment is a little shorter than the penultimate. The first peraeopods (text-fig. 41a) reach beyond the end of the scale by fully half the length of the chela. The merus is slightly longer than the carpus, the carpus distinctly longer than the chela ; the fingers are unarmed and are about equal in length with the palm. ‘The second peraeopods (text-fig. 415) may reach beyond the scale by the whole of the chela and carpus in adult males; in females and young males they are slightly shorter. The legs of a pair are similar both in structure and size and except that they are longer in the male, there is no difference between the sexes. ! Nobili’s remark that the hepatic spine is placed further forwards in his specimens than in Schenkel’s is not borne out by his figures or by his specimen from the.Persian Gulf. 2 In the figure the eye is greatly fore-shortened with the result that the cornea appears broader than the stalk. 138 Records of the Indian Museum. [VoL. XXIV, There are no teeth om ischium, merus or carpus. The merus is neatly twice the lenzth of the ischium and is equal to gr a little longer thanthe palm. ‘The carpus is short and conical, about 1°5 times as long as wide, with a deep notch on the inner side of the distal margin. ‘The chela is large and in adults is about 5 times as long as broad. ‘The fingers are at least two-thirds the length of the palm and have incurved tips and a cutting edge extending through- out their length. In adult males there are teeth in the proximal TEXxT-FIG. 41.—Perviclimenes brevicarpalis (Schenkel). a. First peraeopod, 6, Second peraeopod.! c. Third peraeopod. third of the opposed margins, two on the dactylus and three or four on the fixed finger. These teeth are absent or inconspicuous in females. The last three peraeopods (text-fig. 41c) are similar, the third reaching to or a little beyond the apex of the antennal scale. All are comparatively stout, with some setae but no spinules on the propodus ; the latter segment is from 4°5 to 5 times as long as the dactylus. The dactylus itself is simple, broad at the base and curved. ! J.ess magnified than the other figures. 1922.] S. Kempe: Notes on Crustacea Decapoda. 189 The sixth segment of the abdomen is about 1°5 times the length of the fifth. ‘The telson (text- fig. 42) is_rounded above with two pairs of very small and inconspicuous. dorsal spines. These spines are placed further back than usual, the foremost pair being situated much behind the middle of the telson The terminal spines are unusually thort. The outer uropod is scarcely more than twice as long as broad. At the distal end of its outer margin there is, as usual, a p 4 movable spine separating the ciliated and non-ciliated portions of the mar- gin; but the fixed spine commonly found immediately in front of this movable spine is absent. The largest specimen I have seen is Lexr-ric. 42.—Periclimenes a female 31 mm. in length. brevicarpalis (Schenkel). The literature contains a number of Telson. references to the marvellous colouration of this species when alive and to its association with anemones of the genus Discosoma. Saville-Kent (loc. cit., 1893) has given a coloured drawing of a prawn found on Discosoma haddoni which is perhaps intended for this species, but the figure is extreme- ly poor. Coutiére (/oc. ctt., 1898), who refers to the prawn as Bithynis sp., says of specimens subsequently described by Nobili as Palaemonella aberrans,—‘‘Un Palémonidé du genre Bithynts Dana mérite une mention spéciale par son habitat et sa coloration. Il est absolument transparent, mais se signale par quelques ann- eaux d’un violet pale sur les appendices et l’abdomen, et surtout par des taches d’un blanc nacré éclatant, occupant la région stom- acale tout entiére, le coude de l’abdomen, l’extrémité des rames cau- dales et les épiméres du deuxiéme segment. Ce magnifique Crustacé se tient obstinément dans la zone de protection que circonscrit une grande Actinie assez commune dans les flaques profondes qui sépar- ent les madrépores. Etalé sur le sable, le disque oral de 1’Actinie, de couleur blanchatre, armé d’un trés grand nombre de courts tentacules urticants, atteint souvent o m. 30 de diamétre. Buthy- nis se tient dans ce circle, nageant a peu de distance au-dessus, souvent par couples, et se laisse assez aisément capturer a l’aide d’une éprouvette pleine d’eau que l’on descend doucement sur Vanimal.”” Lenz (loc. cit., 1905) describes the colour thus,—“ Voeltzkow gibt die Farbe im Leben als wasserhell an. Beine an den Gelenk stellen dunkelblau, K6rper dunkel und hell, mit rotbraunen und dunkelgelben Flecken ; Scheren an den Seiten mit weissem Langs- streifen. Augenstiele weiss.’’ This description does not agree well with my own observations. Miss Rathbun (loc. cit., 1914) has described the colour of a 190 cords of the Indian Museum. [VoL XXIlV, specimen preserved in formalin and I have myself (Joc. cit., 1916) given a brief account of the colouration of the species when des- cribing a Hippolytid which is also associated with Discosoma. The following colour description was drawn up from an oviger- ous female obtained in the Gulf of Manaar :— The entire prawn, except for certain pigmented areas noted below, was almost completely transparent. The colouration of the ventral side could clearly be seen in dorsal: view and the nerve- cord was distinctly visible. On the upper side of each eyestalk there was a white stripe which was continuous from side to side beneath the base of the rostrum. There was a large pure white patch on either side of the carapace and the gonad and associated organs were invested with a membrane covered with large closely- set white spots, clearly visible in dorsal view. The hepatic regions and lower muscular portions of the cara- pace were dull venetian red. On the sides of each of the first three abdominal somites there was a large oval patch of glistening white, heavily outlined in black, which extended on to the sternum, and there was a broad band of the same colour on the posterior edge of the last abdominal somite and oa the anterior half of the telson and uropods. In the latter half of the telson and of each uropod there was a brilliant eyespot ; that on the telson was light orange bordered with black, while that on each uropod was similar, but with the orange centre shading distally to dark purple. Ali the other appendages, except the pleopods, were strongly suffused with blue, which was specially dark at the distal ends of the merus, carpus and palm of the second legs and formed a transverse band across the fingers. The cornea was grey and the eggs sage green. It is evident from other notes, made by Col. Alcock on Gt. Coco I. and by myself at Port Blair, that there is very considerable variation in colour. The white patches on the abdomen were outlined in black in the specimen described above; but, just as frequently, they are bordered with orange, deep blue, or, according to Coutiére, pale violet, while the eyespots in the tail-fan may be yellow in the centre, verging to red at the periphery and circums- cribed with deep red-brown. In the distribution of the pigment, however, there appears to be little variation in specimens of the same sex and age. Males lack the two white spots on the carapace and the mem- brane which invests some of the internal organs is without pig- ment. In young specimens the pigmentation is less well developed than in adults. The appearance of this magnificently coloured prawn crawling and swimming in the immediate vicinity of the anemone is a sight. not readily forgotten. That the colouration is in no degree protec- tive is evident from the above description. The large white patches render it very conspicuous and I have already (loc. cit., 1916) drawn attention to the remarkable fact that similar white patches or bands are a characteristic feature of the colour pattern of a Hippo- lytid and two species of fish which are also associated with Disco- 1922.] S. Kemp: Notes on Crustacea Decahoda. - Riot soma at Port Blair. Much careful observation in the field is neces- sary before we can come to any conclusions regarding the significance of the colour pattern or the exact nature of the relations that exist between the prawn and the anemone. At Port Blair P. brevicarpalis was found at low water, either beneath the fringe of tentacles of the anemone, crawling on the disc or swimming in the immediate vicinity. I have not myself seen it enter the mouth of the anemone, though it is not improb- able that it may do so. On several occasions, both at Port Blair and in the Gulf of Manaar the species has been taken in nets hauled in shallow water. I think that its occurrence under these condi- tions is to be explained by the assumption that the net passed over an anemone in its passage along the bottom. I have fre- quently seen the anemone at Port Blair in ro feet of water. The following specimens are in the collection of the Zoological Survey of India :— No: - C 360/1. Kilakarai, G.of Manaar. §. Kemp, Feb., 1913. Two. 9299/6. Spike I., Andamans. ‘Investigator,’ Nov., 1888. One. eet Gt. Coco I:, Andaman ‘Investigator,’ Nov., 1890. Twelve. 2984-90/7. group. C 361-2/r. Port Blair, Andamans. S. Kemp, Feb., March, Forty-five. : 1915; Febr., 1921. In the British Museum I have examined specimens from Murray J., Torres Straits (Potts coll.) and a much damaged indivi- dual from Singapore (Bedford and Lanchester coll.). In the Paris Museum I have seen the specimen from Bahrein I. in the Persian Gulf, recorded by Nobili as Ancylocaris aberrans; also one from Pulo Condore (Germain coll.) labelled ‘‘ corps gélatineuse, taches jaunes,’’ and one from Vanikoro in the Santa Cruz BtOuD, Oceania, labelled ‘‘ sur actinie.’ The species is recorded from Zanzibar, okotont and Bawi in E. Africa (Lenz), from Djibouti in the Red Sea (Nobili), from Bahr- ein I., Persian Gulf (Nobili), from Macassar (Schenkel) from Hermite I., N.W. Australia (Rathbun) and from the Torres Straits (Borra- daile). The specimens figured by Richters from Mauritius and by Saville-Kent from the great Barrier Reef of Australia probably also belong to this species. Periclimenes (Ancylocaris): inornatus, sp. nov.- This species is closely allied to P. brevicarpalis and is also found in association with anemones of the genus Discosoma. The two species differ in the following particulars :— P. brevicarpalis Sch. P. inornatus, sp. nov. Rostrum with 5 to 7 dorsal teeth, the Rostrum with 7 or 8 dorsal teeth, the foremost not placed close to apex. foremost placed close to apex and often giving it a bifid appearance. Carapace greatly swollen dorsally in Carapace not swollen dorsally. adult females. L Hepatic spine of carapace situated on Hepatic spine of carapace situated a much lower level than antennal. nearly on a level with antennal. 12 - Records of the Indian Musewm. [Von. XXIV, P. brevicarpalis Sch. | P. inornatus, sp. nov. Dactylus of second peraeopod at Dactylus of second peraeopod not least two-thirds as long as palm. more than half as long as palm. Dorsal spines of telson very small, Dorsal spines of telson large, anterior both pairs situated in distal half of its | pair situated in proximal half of its length. : length. Brilliantly coloured when alive. ' Without colour when alive. TEXT-FIG. 43.—Periclimenes tnornatus, sp. nov. Anterior part of carapace, rostrum, ‘etc. The rostrum is bent downwards and reaches to or alittle beyond the end of the antennular . : peduncle (text-fig. 43). The upper margin is very slightly convex; the dorsal teeth! are evenly spaced, with the posterior tooth, as in the allied species, be- hind, above or a little in front of the hinder limit of the orbit. The single ventral tooth ‘ usually found? is small and situated beneath the fifth or sixth of those on the upper margin. Db The eye is less slen- : der than in P. brevicar- palis, but possesses an ocular spot as in that species. The fused part aD. TEXT-FIG. 44.—Peviclimenes inornatus, sp. nov. a. Antennule. 6. Antennal scale. ! Of twenty specimens one has 6 dorsal teeth, eleven have 7 and eight have 8. 2 Of twenty specimens one has no ventral teeth, eighteen have 1 tooth and one has 2 teeth. 1922.] S. Kemp: Notes on Crustacea Decapoda. 193 of the outer antennular flagellum (text-fig. 44a) is almost or quite as long as the free portion of the shorter ramus and comprises 3 segments only. The antennal scale (text-fig. 445) is about 2°25 times as long as broad. . There is no arthrobranch on the third maxilliped. The pro- portionate lengths of the segments of the peraeopods are much OM. TEXT-FIG. 45.—Periclimenes inornatus, sp. nov. a. First peraeopod. b. Second peraeopod. c. Third peraeopod. the same as in the related species but the first and last three pairs are stouter (text-figs. 45a, c) and the fingers of the second pair are never more than half the length of the palm (text-fig. 456). In large specimens the fingers of the second pair are provided with teeth on the proximal half of their inner margins; on the dactylus there are two or three recurved teeth and on the fixed 194 Records of the Indian Museum. [VoOL, XXIV, finger four or five of irregular disposition. ‘The dactyli of the last three peraeopods are simple and rather stout. The dorsal spines of the telson (text-fig. 46), by their size and position, afford a ready means of distinguishing the species from P. brevicarpalis. The terminal spines also are longer. The outer uropod is nearly 2°5 times as long as broad, with a single movable spine near the end of its outer margin as.in the allied species. The largest specimen is an ovigerous female about 17 mm. in length. In life the species is almost com- | ) pletely transparent with a faint brown- | ish tinge and with transparent eggs. It can only be detected with difficulty as it crawls among the short tentacles of the anemone. In many of its characters P. ¢norna- TEXT-FIG. 46.—Periclimenes tus resembles P. diversipes. The latter, inoynatus, sp. nov. however, is a much more slender spe- Telson. cies, with highly arched rostrum and with the foremost dorsal tooth not _ placed so near the apex. ‘The fused part of the outer antennular flagellum is much longer than the free part of the shorter ramus and is composed of 7 to g segments, and the antennal scale is narrower and more sharply pointed distally. The second peraeo- pods of type b in P. diversipes are not unlike those of P. inornatus, but the fingers in this type are always more than half the length of the palm. : C 363/1. Port Blair, Andamans. S. Kemp. March, Eighteen, includ- 1915. ing TYPEs. 2991-2/7. Gt. Coco I., Andaman ‘Investigator,’ Nov., Two. group. 1890. On both occasions the species was found on anemones of the genus Discosoma in company with P. brevicarpats. Periclimenes (Ancylocaris) brocketti Borradaile. 1915. Periclimenes (Falciger) brocketti, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 212. 1917. Periclimenes (Falciger) brocketti, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 374, pl. lv, fig. 15. Male Atoll, Maldives, on brown crinoid. Periclimenes (Ancylocaris) compressus Borradaile. 1915. Peviclimenes (Falciger) compressus, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 212. 1917. Periclimenes (ane) compressus, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 373, pl. lv, fig. 18. Saya de Malha. 1922. ] S. Kemer: Notes on Crustacea Decapoda. 195 ‘Periclimenes (Ancylocaris) brevinaris Nobili. 1905. Periclimenes borradailei, Nobili, Bull. Mus. Paris XI, p. 159. 1906. Peviclimenes brevinaris, Nobili, Bull. sei. France Belgique X1., Pp. 42, pl. iii, figs. 7, 7a. Nobili in his description of this species speaks of the spines on the carapace as the ‘‘antennale’’ and “ branchiostégale,”’ but judging from his figure the former is merely the acute lower angle of the orbit, while the latter is the antennal spine. If I have inter- preted the description accurately P. brevinaris lacks a hepatic spine and is related to Miss Rathhbun’s P. pusillus. In P. brevinaris the second peraeopods are shorter than the first, a character also found in some forms of P. diversipes. f The species is known only from a single specimen, obtained on the pearl-oyster banks in the Persian Gulf in 10-12 fathoms of water. Periclimenes (Ancylocaris) pusillus Rathbun. 1906. Periclimenes pusillus, Rathbun, Bull. U, S. Fish Comm. XXIUII, ili, p. 921, fig. 71. Oahu, Hawaiian Is. Periclimenes (Ancylocaris) spiniferus de Man. 1902. Perviclimenes petitthouarsi var. spinifeva, de Man, Abhandl. Senck. naturf. Ges. XXV, iii, p. 824. 1917. Peviclimenes (Falciger) spiniferus, Borradaile, Trans, Linn. Soc. (2) Zool. XVII, p. 369, pl. LII. Other references are given by Borradaile. In the series of specimens that I have examined there are from 6 to 9 teeth on the upper border,! usually 6 or 7, and from 2 to 5 on the lower border,” usually 3 or 4. ; This species and P. petitthouarsi differ from all other Pontoniids in the curious armature of the fingers of the second chela. A pit or socket in one finger, for the reception of a tooth borne on the other finger, is not an uncommon arrangement in the subfamily ; but in P. spiniferus and the related species each finger bears a large oval cup-shaped depression, the two cups being opposed to each other when the claw is shut. Tattersall remarks that a similar artrangenient is found in P. calmani,’ but judging from his figure he has misunderstood the structure in P. spiniferus. The cutting edges of the fingers in P. calmant appear to be quite normal and to bear teeth separated by rather deep notches, just as in P. demani and many other species of the genus. In all well-preserved specimens a ring of black pigment may be seen on the upper side of the cornea. Adult males, when living, are for the most part semi-transparent with minute red and white 1 Of sixty-eight specimens twenty-one have 6 dorsal teeth, forty-five have 7, one has 8 and oné has g. 2 OF sixty-eight specimens three (young) have 2 ventral teeth, thirty-eight have 3, twenty-five have 4 and two have 5. 3 Tattersall, Fourn. Linn. Soc., Zool. XXXIV, p. 386(1921). 196 Records of the Indian Museum. [Vor XXIV, dots. On the anterior part of the carapace (sometimes on the pos- terior parts also) there are oblique or transverse bands of white dots, broadly outlined with deep carmine or black and the eyestalks are striped with the same colour. The distal ends of the merus, carpus and palm of the second peraeopods are suffused with orange or orange red and beyond this suffusion a white patch is frequently found. ‘The fingers are spotted with black and often have a blue _ tinge. The other legs are finely dotted with red or reddish brown and with white. At the distal ends of the telson and each uropod there is a white spot and the setae of the uropods are sometimes dark blue at the base. C 367-S/1. Pamban, G. gf Manaar. S. Kemp, _ Feb., Sixty -three. 1913. g184,6. Ort Sentinel I., Anda- ‘Investigator,’ Jan., Two. mans, 20 fms. 1888. C 369 1. Port Blair, Andamans. S. Kemp, March, Five. 1915. (Geol Samoa. Purchased. One. The species has been recorded from Tamative Reef, Mada- gascar (Lenz), from Chagos Archipelago, Coetivy, the Seychelles and the Maldives (Borradaile), from Pulo Edam in the B. of Batavia, Amboina and Ternate (de Man) and from Tahiti (Heller). The species is usually, if not always, found on madrepore corals. Periclimenes (Ancylocaris) petitthouarsi (Audouin). 1825. Palaemon petitthouarsi, Audouin, Explic, somm. des planches de Crust., p. 91, in Savigny’s Descr. Egypte, pl. x, fig. See IQI5. Periclimenes Petitthouarsi1, Balss, Denk. math.-naturw. Kl. Akad. Wien XCI, p. 25. 1917. Pevriclimenes (Falciger) petitthouarsi, Borradaile, Trans. Linn. : Soc. (2) Zool. XVII, p. 369. 1921. Perviclimenes petitthouarsi, Tattersall, Fourn. Linn. Soc., Zool. XXXIV, p. 385. For other references see Borradaile. In the series of speci- mens I have examined there are from 6 to g teeth on the upper border ' of the rostrum, usually 7 or 8, and from 3 to 5 on the lower border,” usually 4. The ring of black pigment noticed in P, spiniferus on the upper side of the cornea is also present in this species. (Ci371/2.pamlion;, Gafoty Suez: R. B.S. Sewell, 1916. Twenty-three. The species is abundant in the Red Sea and has been recorded by Nobili ® from the vicinity of Arzana I. in the Persian Gulf. I have examined specimens “ea this last locality, belonging to the Paris Museum, | Of twenty-two specimens one has 6 dorsal teeth, fourteen have 7, six have 8 and one has g. 2 Of twenty-two specimens four have 3 ventral teeth, seventeen have 4 and one has 5 De 3 Nobili, Bull. sci. hrance Belgique XL, p. 42 (1906). 1922.] S. Kemp: Notes on Crustacea Decapoda. 197 Periclimenes (Ancylocaris) denticulatus Nobili. 1906. Periclimenes Petitthouarsi var. denticulata, Nobili, Bull, Mus. Paris XII, p. 257. 1907. Peviclimenes Petitthouarsi var. denticulata, Nobili, fem. Accad. Set Torino (2) 1.VII, p. 358. Gatavake, Polynesia. The species of the P. grandis group. The following eleven species are very closely allied and belong to what may be termed the P. grandis group. They agree (i) in the possession of supra-orbital and hepatic spines, (ii) in the. narrow antennal scale with distal spine far outreaching the end of the lamella, (iii) in the unarmed fingers of the first peraeopod, and (iv) in the presence of aspine at the distal end of the merus of the second peraeopod. The species of this group are more difficult to identify than any others of the subfamily. Many of the characters depend on the proportions of various segments of the legs, which are never very easy to determine and there is considerable variation within the limits of a species. The proportions of the segments of the second legs undergo remarkable alteration with growth, especially in males, and are usually very different in adults of the two sexes. In view of these difficulties I have thought it best to avoid the comparative method of description and, at the risk of becom- ing tedious, to give a detailed account of each of the species I have examined. Periclimenes (Ancylocaris) agag, sp. nov. (Plate VII, fig. 9.) The rostrum (text-fig. 47) is slender; it reaches to or a little beyond the end of the antennal scale and is longer than the Text-F1G. 47.—Periclimenes agag, sp. nov. Anterior part of carapace, rostrum, etc. carapace. At the base it is straight, but it is a little upturned in its distal half. On the slightly concave upper border it bears from 7 to g teeth, nearly always 8 or g.' The posterior tooth is rather -- ST a ES ES EE 7 ' Of twenty-seven specimens two have 7 dorsal teeth, twelve have 8 and thirteen have 9. 198 Records of the Indian Museum. [VoL. XXIV, widely separated from the others and is placed on the carapace, the second being immediately above the orbit, the foremost near the tip and the rest more or lessevenly spaced. On the lower border there are from 1 to 3 teeth, nearly always 2,' placed at about the middle of the rostral length. The supra-orbital spine is conspicuous. ‘The lower orbital angle is rounded, with the antennal spine below it.; the hepatic is placed behind the antennal but on a lower level. The eyes are large and somewhat depressed. The cornea is wider than the stalk and usually shows traces of two concentric bands of dark pigment. The ocular spot touches the cornea. The basal segment of the antennular peduncle has a short lateral process and the terminal spine (text-fig. 48a) is short, reaching little beyond the articulation of the second segment; the margin be- tween the spine and the articulation is nearly straight. ‘The second and third segments are slen- der. ‘The free portions of the two rami composing the outer antennular flagellum are extremely short; the fused portion comprises Some g to II . i segments. In the male the total length of the stouter ramus is not much less than that of the OV: abies peduncle, in the female it b. is proportionately rather TEXxT-FIG. 48.—Peviclimenes agag, sp. nov. shorter. The antennal a, Part of antennular peduncle. scale (text-fig. 480) is b. Antennal scale. neatly 5 times as long as wide; the outer margin is slightly concave, ending in a spine which extends far beyond the lamella. The apex of the lamella is broader than in most of the related species. The third maxilliped bears a small arthrobranch. The exopod reaches the end of the antepenultimate segment, the latter bearing a few short spines on its outer margin. Excluding the terminal spine the ultimate segment is about three-quarters the length of the penultimate. The first peraeopods (text-figs. 49a, 6) are long and slender ; in adult males the mero-carpal articulation reaches at least to the end of the second antennular segment. The carpus in adult males | Of twenty-seven specimens one has 1 ventral tooth, twenty-four have 2 teeth and two have 3. 1922.] S. Kempe: Noles on Crustacea Decapoda. 199 is fully 14 times as long as its distal breadth and may be nearly 1'5 times as long asthe merus. In females the carpus is about II times as long as broad and 1°3 times the length of the merus. In large males the carpus is twice or rather more than twice the length of the chela, in small males and females about 1°75 times. The fingers are about equal in length with the palm and are un- armed. The second peraeopods in large males reach beyond the scale by the chela, carpus and a considerable portion of the merus and are from 7 to 8'5 times the length of the carapace. The whole limb is d. e: Tzxt-ric. 49.—Periclimenes agag, sp. nov. a. First peraeopod of male. d. Carpo-propodal articulation of right b. First peraeopod of female. second peraeopod of male, viewed c. Second peraeopod of female (less from above. highly magnified than a, 6 orf). —e. The same, viewed from the inner side. jf. Third peraeopod. covered with minute asperities, visible only under the microscope. The ischium bears a small tubercle at the distal end of the lower margin. The merus is from Io to 14 times as long as wide,! with a strong distal spine on the lower side; it is from 1°3 to 2 times as long as the ischium and about three-quarters the length of the carpus. The carpus is from 12 to 15 times as long as its distal breadth in large males,” from I0 to12 times in small males. There is no conspicuous terminal spine on the inner side of the carpus; the distal end of the segment, viewed from the inner side shows two angular projections (text-fig. 49¢) and one of these when seen from ! In measuring the breadth the spine at the distal end is not reckoned. 2 The carpus is too slender at the distal end in plate vii, fig. 9. 200 Records of the Indian. Museum. [VoL. XXIV, above has the appearance of a short blunt tooth (text-fig. 49d). In this respect there is a marked difference between P. agag and certain related species such as P. andamanensis (cf. text-figs. 574, b, p. 207) in which there is asharp and prominent spine in this posi- tion. ‘The chela is from I'rt to 1'25 times the length of the carpus; ’ the palm in the largest males is 2°5 times, in medium-sized speci- mens Ig times and in the smallest 1°6 times the length of the fingers. The fingers show great variation in form; frequently the cutting edges are straight and meet throughout their length when the claw is closed, bearing a series of small teeth in the proximal half or two-thirds of their length (text-fig. 500). Often, however, there is a TExt-riG. 50.—Periclimenes agag, sp. nov. Fingers of second peraeopod of adult males. a. Excavate type. 6. Non-excayate type. rounded excavation in each cutting edge a little behind the middle, with the result that a gap, sometimes almost circular in outline, is seen when the claw is shut (text-fig. 5oa). The excavation in each finger is limited at either end by a tooth and a series of 3 to 6 teeth is found between the gap and the base of the fingers Males are not dimorphic in the structure of the fingers for specimens occur in an intermediate stage of development, with the notches in the fingers shallow and inconspicuous, In all large males which possess both the second legs the chelae of a pair are closcly similar in structure. In adult females the second peracopods (text-fig. 49c) are much shorter than in large males. ‘The carpus is from 1°3 to 1°5 1922. ] S. Kemp: Notes on Crustacea Decapoda. 201 times the length of the merus and is equal to, or a little longer than the chela. The palm is about 1°4 times the length of the fingers, which are unarmed or with very small and inconspicuous teeth near the base. The last three pairs of peraeopods are long and very slender (text-fig. 49f). In adult males the mero-carpal articulation of the third. pair reaches almost or quite to the end of the basal anten- nular segment. In both sexes the fifth pair reaches beyond the end of the antennal scale by fully the length of the dactylus. The merus of the third pair is from 16 to 18 times as long as broad in adult males, 13 to 14 times in females. The propodites bear a few slender spinules on their posterior edges: in the male they from 3 to 3°3 times thetength of the dactylus, from 2°7 to 3 times ii the female. The dactylus is simple and curved, with a few setae in the middle of the anterior margin; it is very slender, from 7 to 8 times as long as its basal breadth. The sixth abdominal somite is about 1 7 times the length of the fifth.' The foremost pair of dorsal spinules of the telson are placed in the anterior third of its length, the second pair midway between the first and the apex. The intermediate apical spines are very long. Large specimens are about 16°5 mm. in length. C 374-61. Port Blair, Andamans, S. Kemp, Feb., 1915; Thirty-five, in- 4-8 fms. Feb., March, 1921. cluding Types. All the specimens were found in Ross Channel on a bottom composed mainly of small corals and sponge-encrusted stones. Periclimenes (Ancylocaris) proximus, sp. nov. The rostruin (text-fig. 51) is slender and reaches almost to or a little beyond the apex of the antennal scale. It is a little up- TEXT-©1G. 51.—Fericlimenes proximus, sp. nov. Anterior part of carapace, rostrum, ete. turned in its distal third and bears on the slightly concave upper margin 6 or 7 large teeth, usually 7.*° The posterior tooth is placed on the carapace and is rather more distant from the second thau the second is from the third ; the remainder are more or less eveniy ' It is too short in plate vii, fig. 9. * Of twenty-one specimens four have 6 dorsal teeth and seventeen have 7. 202 Records of the Indian Museum. [VoL. XXIV, spaced and the seventh tooth, when present, is much smaller than the rest and placed quite close to the apex. On the lower border there are 2 or 3, nearly always 2 teeth,' which are large and placed in advance of the middle of the rostrum, Supra-orbital, antennal and hepatic spines are present as in P, agag. The eyes are large and depressed; the cornea is wider than the stalk and in freshly preserved specimens usually shows two concentric rings of dark pigment. The ocular spot touches the cornea. The basal segment of the antennular peduncle has a short lateral process, not reaching the middle of the segment; its ter- minal spine is long, reaching the middle of the second segment, and the margin between the base of this spine and the articulation is gently convex (text-fig. 52a). The two distal segments are slender. The free portion of the stouter ramus of the outer antennular flagel- lum is very short; the fused portion consists of 7 to II segments. The total length of the stouter ramus is con- siderably less than that of the peduncle. The antennal scale (text-fig. 52b) is from 4°5 to 5°8 times as long as wide and is proportionately longest in large males. aX ae The scale is very narrow at the apex. The outer a. Part of antennular peduncle. Hegel 2 strongly ses By Nntenncleeale: cave and terminates in a spine which far out- TENXT-P1G. 52.—Periclimenes proximus, sp. nov. reaches the end of the lamella. The third maxilliped bears a small arthrobranch. The exopod reaches the end of the antepenultimate segment, the latter bearing one or two spines on its outer edge. The ultimate segment, ex- cluding the terminal spine is about two-thirds as long as the penultimate. The mero-carpal articulation of the first peraeopods reaches the end of the second segment of the antennular peduncle in adult males, not quite so far in females. ‘The carpus in adults of both sexes is from 1I’2 to 1°3 times the length of the merus and about I°4 times the length of the chela (text-fig. 53a). The limb is stouter than in P. agag ; in an adult male the carpus is 10 times as long as £ Of twenty-one specimens twenty have 2 ventral teeth and ane has 3. 1922.] S. Kemp: Notes on Crustacea Decapoda. 203 its distal breadth, in an adult female 8°5 times. ‘The fingers are longer than the palm and are unarmed. The second peraeopods bear a conspicuous subterminal spine on the lower side of the merus. In large males they may be as much as 6 times the length of the carapace, extending beyond the _ scale by more than the length of the carpus and chela. ‘The legs of a pair are equal or subequal and similar in structure. As in P. agag the second legs of males are closely covered with minmte as- perities only visible under a microscope. In large males (text-fig. 53b) there is a conspicuous tubercle TExtT-FIG. 53.—Periclimenes proximus, sp. nov. a. First peraeopod of male. d. he same, viewed from the b. Second peraeopod of male. inner side. c. Carpo-propodal articulation of right second e. Second peraeopod of female. peraeopod of male, viewed from above. f. Third peraeopod. at the distal end of the lower border of the ischium. The merus is from 7'0 to 8'o times as long as broad. ‘The carpus is from I‘o to 1'2 times as long as the merus and from 70 to 8’o times as long as its distal breadth. The distal end of the carpus is similar in structure to that of P. agag and does not bear a conspicuous spine on the inner side (text-figs. 53c,d). The chela is from 1°4 to 1°7 times the length of the carpus; the palm is about 5 times as long as wide and from 1°95 to 2'2 times the length of the fingers. In all the males examined, the fingers meet throughout their length when the claw is closed.' Each is armed in the proximal half ! The number of large specimens in the collection is small; it is very prob- able that more highly developed males with gaping fingers remain to be dis. covered, 204 Records of the Indian Museum. [VoL. XXIV, with a series of 4 to 8 small teeth, very irregular in their size and disposition. In a small male the carpus is only about 6 times as long as its distal breadth, while the chela is 1°7 times its length. The palm is rather more ‘swollen than in large males, about 4°5 times as long as broad. The female differs conspicuously from that of P. agag in that the chela is always definitely longer than the carpus. In an ovigerous specimen (text-fig. 53e) the carpus is 7°5 times as long as its distal breadth and 1°25 times as long as the merus. The chela is I'4 times as long as the carpus, with palm 1°6 times as long as the fingers. The last three pairs of peraeopods are slender ; the fifth reach to or a little beyond the end of the antennal scale. In the third pair (text-fig. 53/) the merus is from 11°5 to 12°5 times as long as wide. ‘The propodite bears conspicuous spinules on its posterior border and is from 3°5 to 4’0 times as long as the dactylus, ‘he . dactylus is slender and curved, with a few setae in the middle of its anterior margin, and is from 5 to 6 times as long as its basal breadth. The sixth abdominal somite is rather less than 1°5 times the length of the fifth. The telson resembles that of P. agag. The largest specimen is a male, about 17°5 mm. in length. I have no notes on the colouration of living specimens as the differences between this and other closely related forms were not noticed in the field. In specimens, however, which have only been a few months in alcohol a bright red spot is to be seen at the end of the carpus of the second leg and a narrow red band across the fingers of the same appendage. This colouration is not found in any of the allied species. The principal differences between P. proximus and P. agag may be summarized thus :— P. proximus, sp. nov. | Rostrum with 6 or 7 dorsal teeth, Carpus of first peracopods less than 1°5 times length of chela. Chela of second peraeopods in males from 1°4 to 1°7 times length of carpus. Chela of second peraeopods in females conspicuously longer than carpus. Port Blair, Andamans, 4-8 fms. C 377-9/1. 1915; Feb., March, 1921. P. agag, sp. nov. Rostrum usually with 8 or 9 dorsal teeth, Carpus of first peraeopods in Iemales and young males 1°75 times, in adult males twice the length of chela. Chela of second peraeopods in males I'I to 1°25 times length of carpus. Chela of second peraeopods in females equal to or a little shorter than carpus March, Twenty-two, in- - cluding Typrs. IXemp, The specimens were found in Ross Channel in company with P. agag and P. andamanensis. Periclimenes (Ancylocaris) andamanensis, sp. nov. This species differs conspicuously from the two preceding in the presence of a conspicuous distal spine on the inner side of the carpus of the second peraeopods, 1922.] S. Kemp: Notes on Crustacea Decapoda.’ 205 The rostrum (text-fig. 54) is slender and reaches to or a little beyond the apex of the antennal scale. It is straight in its proxt- mal half but trends upwards distally. On the slightly concave upper border it bears from 7 to g teeth, nearly always 8 or 9; ' Text-FIG, 51.—Periclimenes andamaiensis, sp. nov. Anterior part of carapace, rostrum, etc. the hindmost of these is situated on the carapace and is separated by a rather wide interval from the second, which is placed above the hinder limit of the orbit. The rest of the teeth are more or less evenly spaced, extend- ing to the tip. On the lower border there are 2 or 3 teeth, most commonly 2,” placed a little in advance of the middle of the rost- rum. The supra-orbital, anten- nal and hepatic spines re- semble those of the prece- ding species. The eyes also are similar but usually show only one faint band of dark pigment on the cornea. The basal segment of the antennular peduncle has a a. b. short lateral process; the distal spine of the outer margin is very short (text- Be esa), ik Yemeride Wattle 5 Antennal seat beyond the articulation bet- ween the first and second segments and the margin between the spine and the articulation is almost straight. The two distal seg- TEX?T-¥IG. 55.—Periclimenes andamanensts, sp. nov. ! Of fifty specimens four have 7 dorsal teeth, twenty-nine have 8 and seven- teen have g. 2 Of fifty specimens thirty-seven have 2 ventral teeth and thirteen have 3. 206 Records of the Indian Musewm. [VoL,. XXIV, ments are slender. The free portions of the rami composing the outer antennular flagellum are very short. The fused portion comprises some 8 to II segments and the total length of the stouter ramus, in males, is a little longer than the peduncle. ‘The antennal scale is from 5 to 5'5 times as long as wide and is very TExT-FIG. 56.—Periclimenes andamanensis, sp. nov. a. First peraeopod. ce. Second peracopod of female. 6. Second peraeopod of male. d. Third peracopod. narrow at the apex. ‘The outer margin is concave with the termi- nal spine far outreaching the end of the lamella. . The third maxilliped resembles that of the preceding species. The mero-carpal articulation of the first peraeopods (text-fig. 56a) reaches nearly to the end of the basal antennular segment. The merus is 8 to 9 times as long as broad. The carpus is. about 1922. ] S.. Kemp: Notes on Crustacea Decapoda. 207 I'r times as long as the merus, from 7 to ro times as Jong as its distal breadth and from 1°4 to 1'°7 times as long as the chela; it is proportionately longest and most slender in adult males. The fingers are about equal in length with the palm and are unarmed. The second peraeopods of adult males (text-fig. 56)-extend beyond the antennal scale by the chela, carpus and a portion of fhe merus and may be as much as 6 times the length of the carapace. ‘They do not show the minute asperities with which the second legs of the two preceding species are covered and there is no tubercle at the distal end of the ischium. The merus bears the usual strong spine at the distal end of the lower border and, in adults, is from 8 to 9 times as long as broad. In all well grown males the merus is very slightly longer than the carpus, from 1°05 to I'r times its length; in small males the merus and carpus are equal or the latter is a shade the longer. The carpus is from 6 to 7 times as long as its distal breadth in adults, but in young males is more slender, sometimes as much as g times as long as wide. The carpus always bears a conspicuous spine on the inner side of its distal margin and in large males there is in addition a small acute projection or tooth on the upper and inner aspect (text-figs. 57a, b). The chela is from 1°8 to 2’2 times the length of the carpus in adults, in young specimens 1°5 times or even less. The palm in large specimens is about 6 times as long as wide; in adults it is from 1°8 to 2°1 times as long as the fingers, in young males pro- portionately shorter, from 1°5 to 1°7 times. The fingers resemble those of P. agag; in some specimens they are excavate on their inner margins, in others they meet throughout their length when shut and bear a series of small teeth in their proximal two-thirds. In females (text-fig. 56c) the second perae- opods are more slender and proportionately shorter than in adult males. The carpus is equal to or a little longer than the merus and is 8 to 9 times as long as its distal aie b. a ; breadth. As in males TExT-FIG. 57.—Periclimenes aindamanensis, the carpal spine is con- ' sp. nov. spicuous. The chela is a. Carpo-propodal articulation of right second from 1°35 to 1°6 times peraeopod of male, viewed from above. as long as the carpus, b. The same, viewed from inner side. with the palm about 1°4 times the length of the fingers. The fingers have some incons- picuous teeth in the proximal half. 208 | Records of the Indian Museum, + [Vou. XXIV, The last three pairs of peraeopods (textfig. 56d) are long and slender ; the fifth pair reaches to or a little beyond the end of the antennal scale. ‘The merus of the third pair is about 15 to 16 times as long as broad in adults. The propodus bears some slen- der spinules on its posterior edge and is from 2°7 to 3°6 times as long as the dactylus, proportionately longest in large males. ‘The dactylus is simple and curved, with a few setae on the middle of its anterior margin; it is from 7°5 to 8 times as long as its basal breadth. The sixth abdominal somite is about 1'7 times the length of the fifth. In the arrangement of the spines the telson resembles that of P. agag. A large male is about 19 mm. in length. Periclimenes andamanensis agrees with P. Bonn: and differs from P, agag in the comparatively stout first and second legs and in the greater length of the chela of the second legs, as compared with the carpus, in adults of both sexes. From both species it is distinguished by the presence of the carpal spine. It also differs from P. proximus in the greater number of upper rostral teeth, in the proportionate lengths of merus, carpus and chela in the second leg of the adult male and in the rather more slender legs of the last three pairs. Other minor differences will be found on com- parison of the two descriptions given above. C 380-1/1. “Port Blair, Andamans, S. Kemp, Feb., 1915 ; Many. 4-8 fms. Feb., March, 1921. The specimens were found in Ross Channel in company with P. agag and P. proximus. ‘The types bear the number C 380/T. Certain additional specimens obtained on muddy ground at the inner end of Port Blair are tentatively referred to P. andaman- ensis, but differ in certain characters which will perhaps prove to possess at least varietal value. Of the nine specimens eight are females and one a young male. The only points in which these specimens differ from the above description are as follows :— The rostrum is less shallow and bears as a rule g dorsal teeth and 3 ventral.! The ovigerous females are larger than any.typical -P, andamanensis that I have seen, with the carpus of the second peraeopods decidedly stouter, from 5°5 to 6 times as long as its distal breadth. The chela also is longer in relation to the carpus, about 1°8 times its length. In the last three legs the dactylus is considerably longer than in typical specimens. In large females the propodus of the third pair is only 2:25 times and in the young male only twice the length of the dactylus. The dactylus is also rather more slender from 8 to 9 times as long as its basal breadth in females, 11 times in the young male, ! Of eight specimens one has 8 dorsal teeth, six have 9 and one has 11; seven specimens have 3 ventral teeth and one has 4. ; 1922.] S. Kemp: Notes on Crustacea Decapoda. 209 In other respects there is practically no difference between the two sets of specimens. The young male resembles typical specimens of the same size and sex in the proportions of the seg-. ments of the second peraeopods. The carpus is about 7 times as long as its distal breadth and is a trifle longer than the merus ; the chela is 1°5 times as long as the carpus.: In females the car- pus is a little shorter than the merus and, in all the specimens, the carpal spine is conspicuous. The merus of the third leg is 12 to 13 times as long as broad in females, 17 times in the young male. In the absence of fully developed males it is not possible to identify the specimens with certainty, but the material examined seems to point to the conclusion that the muddy ground at the inner end of Port Blair is inhabited by a special variety of P. anda- manensts. The largest female is about 18 mm. in length. C 382-3/1. Port Blair, Andamans, S. Kemp, leb., Nine. 3-5 fms. March, 1921. ; The specimens were obtained off Viper I. and at the mouth of Brigade Creek on a bottom composed of mud and decaying vegetation. Periclimenes (Ancylocaris) suvadivensis Borradaile. 1915. Peviclimenes (Falcigey) suvadivensis, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 212. 1917. .Pertclimenes (Falciger) suvadivensis, Borradaile, Tyans. Linn. Soc. (2) Zool. XVII, p. 375, pl. lv, fig. 16. I have examined the types of this species and find that Bor- radaile was mistaken in supposing that they do not possess a supra- orbital spine. The species thus finds a place in the P. grandts section and is extremely closely allied to P. andamanensis. Un- fortunately I was not able to make a critical comparison of the two forms and the only characters that I can now give for the separation of the two species are that (i) the upper rostral teeth (6 or 7) are less numerous in P. suvadivensis than is customary in P. andamanensis and (ii) that the carpus of the second peraeopods in adult males is conspicuously longer than the merus in the for- mer species, whereas in the same sex of the latter the merus is longer than the carpus. The species was described from specimens taken at Suvadiva Atoll in the Maldives. Periclimenes (Ancylocaris) ensifrons (Dana). 1852. Anchistia ensifrons, Dana, U. S. Explor. Exped., Crust. 1, p- 580, pl. xxxviii, figs. 1a-g. 1899. Periclimenes ensifrons, Nobili, Ann. Mus. civ. Genova (2) XX, Pp. 234. 1907. Pertclimenes ensifrons, Nobili, Mem. Accad. Sci. Torino (2) LVII, p. 350. Nobili (1907) has examined two young specimens of this species from Polynesia and has pointed out that they differ from 210 _ Records of the I ndian Museum. [VoL. XXIV, those to which de Man and other authors have applied the name in the absence of the spine at the distal end of the carpus of the .second peraeopods. In this respect his specimens agree with Dana’s description ‘‘ carpus long, not armed or acute at apex.” According to Nobili de Man’s specimens probably represent a _ variety of Dana’s species, but with this I am unable to agree. For P. ensifrons, as applied by de Man, Stimpson’s name grandis may be employed. ‘This species differs from true P. enst- frons in possessing the carpal spine on the second legs and also in the proportions of the merus and carpus of the same limb. In P. ensifrons the carpus is decidedly longer than the merus (see Dana’s figure and Nobili’s measurements), whereas in P. grandis the merus in males is longer than, and in females equal to or a little longer than the carpus. P. ensifrons was described from the Straits of Balabac, North of Borneo and is recorded by Nobili from Beagle Bay, New Guinea and from the lagoons of Amanu and Fakahina in Polynesia. There does not appear to be any evidence that it occurs in the western part of the Indo-Pacific region. Periclimenes (Ancylocaris) grandis (Stimpson). (Plate VII, fig. 10.) 1860. Anchistia grandis, Stimpson, Proc. Acad. Sci. Philadelphia, p. 39. 1887. Anchistia ensifrons, de Man, Arch. Naturgesch. LIKI, i, p. 545. 1887. Anchistia ensifrons, Miller, Verh. naturf. Ges. Basel VIII, p. 471. 1894. Anchistia ensifrons, Ortmann, Fenaisch. Denkschr. VIII, p. 16. 1902, Peviclimenes ensifrons, de Man, Abhandl. Senck. naturf. Ges. XXV, p. 826. ; 1905. Periclimenes vitiensis, Pearson, Ceylon Pearl Oyster Rep. \V, p. 78. 1905. Periclimenes ensifrons, \.enz, Abhandl, Senck. naturf. Ges. XXVIII, p. 80. - 1906. Peviclimenes ensifrons, Nobili, Aun. Sct. nat., Zool. (9) 1V, p. 49. . 1915. Periclimenes ensifrons, Balss, Denk. math.-naturw. Kl. K. Akad. Wien XCI, p. 26. The rostrum reaches to or a little beyond the end of the an- tennal scale. In lateral view it is deep, more so in females than in males; it is straight at the base but in its distal half is directed upwards, the upper margin being thus slightly concave. The dorsal teeth are from 6 to 10 in number,! nearly always 7 or 8. The posterior tooth stands on the carapace.and is separated from the next by a rather wide interval ; the second is placed above or a little behind the posterior limit of the orbit; the foremost is very close to the apex and often gives it a bifid appearance. In the precise distribution of the teeth there is, as usual, some variation ; frequently, and especially in males, four teeth are placed rather close together above the eye, one or two near the apex and one mid- way between the two groups. On the lower border there are from ! Of eighty-four specimens one has 6 dorsal teeth, forty have 7, thirty-nine have 8, three have g and one has 10. 1922.] S. Kempe: Notes on Crustacea Decapoda. SE 2 to 5 teeth,’ usually 2 or 3; these teeth are large and the foremost. is alivays placed behind the most anterior of those on the upper border. Supra-orbital, antennal and hepatic spines are present; the hepatic is placed rather close behind the antennal, but on a lower level. The lower limit of the orbit is defined by a blunt angulation of the frontal margin. The eyes are stout. The cornea is hemispherical and wider than the stalk; in life it is traversed by two concentric bands of dark pigment and these may frequently be seen in preserved mate- tial. The ocular spot is distinct and confluent with the cornea. The lateral process of the antennule is short, not reaching the middle of the basal segment. The spine at the distal end of the’ outer border of thisseg- ment is stout, but does not reach to the middle of the second segment ; the margin between the spine and the arti- culation is strongly sinuous (text-fig. 58a). The two distal segments are slender. The outer antennular flagellum is cleft for only a very. short distance, the fus- ed basal portion com- prising Io to 13 seg- ments. In both sexes the total length of the outer ramus is con- siderably less than that of the peduncle. The Texr-ric. 58.—Periclimenes grandis (Stimpson). antennal scale (text-fig. a. Part of antennular peduncle. 58b) is narrow at the 6. Antennal scale. apex and is from 39 to 4'3-times as long as broad in adults ; in young specimens it is more slender, sometimes as much as 5 times as long as broad. The cuter margin is concave and ends in a spine which projects far beyond the end of the lamella. The third maxilliped bears a small arthrobranch and reaches about to the end of the basal antennular segment. The ante- penultimate segment is somewhat curved and bears from I to 3 slender spines on the outer margin; the exopod reaches a little beyond its distal end. The ultimate segment is three-quarters the length of the penultimate. The first peraeopods (text-fig. 59a) reach beyond the antennal ! OF eighty-four apecimens thirty-four have 2 ventral teeth, thirty-seven have 3, eleven have 4 and two have 5. 212 Records of the Indian Museum. [VoL. XXIV, scale by the length of the fingers. The carpus is a little longer than the merus, from 7 to 8 times as long as its distal breadth and from 1°35 to 1'5 times the length of the chela. The fingers are unarmed and are about as long as the palm. The second peraeopods are equal ; as in the forms already des- cribed they are much longer in males than in females and the pro- portions of the segments differ widely in the two sexes. In adult males (text-fig. 59>) they reach beyond the scale by the entire length of the chela and carpus. The merus is from 1°25 to 1°35 times the length of the carpus and is from 6 to 6°5 times as long as’ wide; it bears a conspicuous spine at the distal end of the~- lower margin, ‘The carpus bears a curved, forwardly directed spine on TExT-FIG, 59.—Periclimenes grandis (Stimpson). a. First peraeopod. c. Second peraeopod of female. 6. Second peraeopod of male. -d. Third peraeopod. the inner side of the distal border and is from 4 to 5 times as long as its greatest breadth, excluding the spine. ‘The chela in well- grown specimens is from 2 to 2°5 times the length of the carpus. The palm is about 4°5 times as long as wide and is from 1°6 to twice the length of the fingers. The cutting margins of the fingers are excavate in the middle so that an oval gap is left when the claw is closed. In front of this excavation there is a single tooth on each finger, that on the dactylus being in advance of the other. Behind the excavation there is a large tooth on the fixed finger, succeeded by a variable number of smaller teeth and there is a seties of medium-sized teeth, usually 4 or 5, on the proximal part of the dactylus. The tips of the fingers are inturned. ‘The excavation in the fingers is to be seen only in very large males; usually both 1922. | S. Kempe: Notes on Crustacea Decapoda. 213 chelae show a similar development, but I have s2en one specimen in which the gap was present in one chela only. In the female (text-fig. 59c) the merus is rather more slender, about 6°5 to 7°0 times as long as wide and equal to or a little longer than the carpus. ‘The carpus is from 5 to 5‘25 times as long as.its distal breadth and, as in the male, bears a conspicuous distal tooth on theinnerside. The chela is from 1°6 to 1°8 times the length of the carpus. The palm is equal to, a little longer or little shorter than the carpus and is from 1°3 to 1°6 times as long as the fingers. The fingers have inturned tips and may be provided with small in- conspicuous teeth on the proximal half or third of their cutting edges. The last three peraeopods are moderately slender; the fifth do not reach the apex of the antennalscale. In the third pair (text- fig. 59¢) the merus is from 9 to 10 times as long as broad. The propodus bears long spinules on its posterior margin and is from 2°8 to 3°3 times as long as the dactylus. The dactylus is simple and slightly curved with a few setae onits anterior margin; its length is from 6 to 6°5 times its basal breadth. ‘The sixth abdominal somite is about 15 times the length of the fifth. The dorsal spines of the telson are so arranged as to divide its length into three equal parts. The largest specimen, a male, is about 23 mm. in length. Specimens from the Gulf of Manaar were almost completely transparent when alive, minutely speckled with red and blue. In some individuals a blue patch was visible at the ends of the merus and carpus of the second legs and a brownish red patch on the outer side of the propodus. Stimpson’s description agrees in every particular with the large males that I have examined, except that the chela of the second legs is said to be nearly three times the length of the carpus with fingers less than half the length of the palm. The assumption that Stimpson described a more fully developed male than any I have seen will fully account for these discrepan- cies, In many respects P. grandis agrees with P. andamanensis. It differs, however, in its deeper rostrum, in the stronger spine at the distal end of the first antennular segment, in the much stouter merus and carpus of the second legs of the male and in the shorter and stouter legs of the last three pairs. The merus of the third legs is only 10 to 11 times as long as wide in P. grandis, from 15 to 16 times in P. andamanensis. C 384/1. Kilakarai, G. of Manaar, S. Kemp, Feb., 1913. Many. 0-2 fms. : C 385/1. Pamban, G. of Manaar. S. Kemp, Feb., 1913. One. . C 386/1. Cochin backwater, near F. H. Gravely, Sept., Thirteen. Ernakulam, S. India. 1914. C 3837/1. N. Cheval Paar, Ceylon. TL. Southwell, Nov., Six. 1910. C 338/1. Paway I., Mergui Ar- ‘Investigator,’ leb., Two. chipelago. 1914. 214 Records of the Indian Musewm., [VoL. XXIV, The species was very common in the Gulf of Manaar, among weeds in shallow water and also on the coral reefs, The species was described from Ousima I. (Stimpson) and has been recorded from Ternate and Pulo Edam (de Man), Trincomalee (Miiller), Cheval Paar (Pearson), Zanzibar (Lenz), Dar-es-Salaam (Ortmann) and the Red Sea (Nobili, Balss). Periclimenes (Ancylocaris) vitiensis Borradaile. 1898. Periclimenes vitiensis, Borradaile, Ann. Mag. Nat. Hist. (7) Il, Pp. 383. 1899. Peviclimenes vitiensis, Borradaile, Proc. Zool. Soc. London, p. 1005, pl. Ixiv, figs 6a, b. 1899. Periclimenes vitiensis, Nobili, Ann. Afus. civ. Genova (2), XX, » 234, 1917. Pay ieihes vitiensis, Borradaile, Trans. Linn. Soc. (2) Zool. XXII, p. 371 (part). This species is very closely related to P. grandis. I have examined the type, an ovigerous female in the Cambridge Museum, but my work at that time was not sufficiently advanced to enable me to make full use of the opportunity. I noted, however, that in the telson of the type specimen both pairs of dorsal spines afe situated in the posterior half, whereas in the specimens I have referred to P. grandis the foremost pair is situated in the anterior half. The position of these spines affords a valuable specific character in some species of Periclimenes (cf. P. brevicarpalis and P. inornatus) and I conclude, therefore, that P. vitiensis is possibly a distinct species. The specimens from Coetivy in the Seychelles subsequently referred by Borradaile to P. vitiensis should be re-examined, for it is not improbable that they belong in reality to P. grandis. P. vitiensis was described by Borradaile from Viti Levu, Fiji. Periclimenes (Ancylocaris) affinis Borradaile. 1915. Periclimenes (Falciger) affinis, Borradaile, Ann. Mug. Nat. Hist. (8) XV, p. 211. 1917. Periclimenes (Falciger) affinis, Borradaile, Trans. Linn. Soc. (2) Zool, XVII, p. 372, pl. liv, fig. 11. This species appears to be closely related to Paulson’s P. ele- gans, which is described below. According to Borradaile’s descrip- tion and figures it differs (i) in its straighter rostrum, armed with only 2 teeth below, (ii) in the greater proportionate length of the first peraeopods which outreach the antennal scale by the chela and half the length of the carpus, and (iii) in the much more slender and proportionately longer carpus of the second peraeopods, about 6 times as long as wide according to the figure and a little longer than the merus. P. affinis is recorded from Salomon I. in the Western Indian Ocean. 1922.] _ §. Kemp: Notes on Crustacea Decapoda. 215 Periclimenes (Ancylocaris) elegans (Paulson). 1875. Anchistia elegans, Paulson, Crust. Red Sea, p. 113, pl. xvii, figs. I, ta-h. 1906. Anchistia elegans, Nobili, Ann. Sct. nat., Zool. (9) 1V, p. 52. The rostrum (text-fig. 60) is rather deep, especially in females, and reaches to or a little beyond the end of the antennal scale. It is straight at the base, but in its distal half is directed upwards. Text-riG. 60.—Periclimenes elegans (Paulson). Anterior part of carapace, rostrum, etc. On the concave upper border there are from 6 to 8 teeth,! nearly always 7or8. ‘The teeth are arranged much as in P. grandis, but the distinction into two groups is sometimes even more ciearly marked than in that species. On the lower border there are from 3 to 5 teeth,’ nearly always 3 or 4. In the spines of the carapace and the eyes the species resembles P. grandis. A single band of dark pigment is frequently seen on the cornea. The antennules also resemble those of P. grandis, but the termi- nal spine of the basal segment is rather longer, reaching almost wv. 6. or quite tothe middle yyxrpig 61.—Periclimenes elegans (Paulson). of the second segment, a. Part of antennular peduncle. and the margin bet- 6. Antennal scale. ween this spine and | Of fifty specimens two have 6 dorsal teeth, thirty-six have 7 and twelve have 8, 2 Of fifty specimens thirty-one have 3 ventral teeth, eighteen have 4 and one has 5. 216 Records of the Indian Museum. [VoL. XXIV, the articulation is less clearly sinuous (text-fig. 61a). In both sexes the stouter ramus of the outer flagellum is shorter than the peduncle. In its distal third the antennal scale is rather more natrowed than in P. grandis and its outer margin more strongly concave ; it is from 4°5 to nearly 5°5 times as long as wide (text- fig. 610), The third maxilliped possess a small arthrobranch; the an- tepenultimate segment bears from I to 4 spines on its outer edge and the exopod reaches about to its end. ‘The ultimate segment is about three-quarters the length of the antepenultimate. TextT-FiG. 62.—Periclimenes elegans (Paulson), a. First peraeopod. c. Fingers of same. b. Second peraeopod of male. d. Second peraeopod of female. e. Third peraeopod. . The first peraeopods (text-fig. 62a) reach beyond the antennal scale by about half the length of the chela. The carpus is from one-ninth to one-twelfth longer than the merus and is from 7 to 7°5 times as long as its distal breadth and from 1°2 to 1°4 times as long as the chela. The fingers are equal to or a little shorter than the palm and are unarmed. The second peraeopods are equal. In males they extend beyond the antennal scale by the length of the carpus and chela. In none of the specimens I have seen do the legs present the rugose appearance described by Paulson. In males (text-fig. 626) the merus is from 6 to 7 times as long as broad, with the usual spine 1922.] S. Kemp: Notes on Crustacea Decapoda. Any, at the distal end of the lower margin; it is from 1°2 to 1°3 times as long as the carpus. The carpus is from 4 to 4°5 times as long as its distal breadth and bears two stout spines at the distal end, one on the inner side and one on the upper and inner aspect; inferiorly the distal end projects a little, producing the appearance of a tooth when seen in lateral view. ‘The chela is from 2 to 2°5 tines the length of the carpus and the palm is from 1'g to 2°4 tines as long as the fingers. The fingers (text-fig. 62c) frequently but not always show an excavation in the middle of their cutting edges as in P. grandis. Some specimens have comparatively laige teeth on the fingers while others have only a few very small teeth. In the female (text-fig. 62d) the merus of the second peraeo- pod is from 1’o to 1°3 times as long as the carpus and is from 710 7°5 times as long as wide. The carpus varies from 4 to 5'5 times as long as its distal breadth; the spine on the inner side is well developed and frequently an acute process or short spine can be seen on the upper and inner aspect, corresponding to the second spine found in the male. The chela is from 1°4 to 2°'I times the length of the carpus, with palm from ‘13 times to twice! as long as the fingers. The fingers bear small teeth in the proximal half of their inner margins. The last three peraeopods are stout; the fifth, when extended forwards, fall far short of the apex of the antennal scale. In the third pair (text-fig. 62¢) the merus is from 7'5 to 8°5 times as long as broad; the propodite bears a series of spinules on its posterior edge and is from 3'5 to 4 times as long as the dactylus. The dactylus itself is simple, very slightly curved, and generally with one or two long setae in the middle of its anterior margin; its © length is only from 4 to 4°5 times its basal breadth. The last abdominal somite and telson agree with those of P. grandis. The largest specimen, a male, is about 24 mm. in length. Although the above account differs in some respects from Paulson’s description (as translated by Nobili) and from his figures, I have little doubt that the identification is correct and that the discrepancies are mainly due to errors in the original account. A single adult male collected by Major R. B. Seymour Sewell in the Red Sea, belongs almost without question to Paulson’s species and this: individual is indistinguishable from specimens obtained in the Andamans. ‘The specimen which Balss has recorded as P. elegans from St. John’s I. in the Red Sea’? apparently does not belong to this species as the spine on the carpus of the second leg is said to be absent. C. 389-90/1. Port Blair, Andamans. S. Kemp, - March, Many. 1915; Feb., March, 1921. ' In exceptionally large females only. 2 Balss, Denk. math.-naturw. Kl. K. Akad. Wien, XCI, p. 26 (1915). 218 _ Records of the Indian Museum. [VoL XXIV C:391/1, East I., Andamans, A. R. S. Anderson, Two. ; 1808. C 392/1. Koweit Harbour, Per- R. B. Lloyd, Oct., Two. sian Gulf. 1905. C 393/1. Tor, Sinaitic Penin- R: AB: S:) Sewell, One. sula, Red Sea. 1916. At Port Blair the species is abundant at low water in pools on the coral beach; it was found on Ross I., and on the shores of Aberdeen and North Bay, and was never obtained by dredging. The species has hitherto been known only from the Red Sea. ar. dubius, Borradaile. 1915. Periclimenes (Falciger) dubius, Borradaile, Ann. Mag. Nat. ; Hist, (8) XV, p. 21t. 1917. Periclimenes (Falciger). dubius, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 373, aes liv, fig. 12. VEXxT-FIG. 63,—Periclimenes elegans, var. dubius, Borradaile. Second peraeopod of male. Certain specimens from Madras Har- bour differ from typical fF’. elegans only in the propor- tionately stouter carpus of the second peraeopods in males (text-fig. 63). In adults of this sex the carpus is only 3 times as long as its distal breadth and in females barely 4 times. This is the only difference I can find and I do not regard the two forms as specifically dis- tinct. So far as I can understand from the published ac- count Borradaile’s name dubius is cor- rectly applied to these specimens. ( 46/1. Madras Harbour. S. Kemp, May, 1918. Six. Periclimenes (Ancylocaris) holmesi Nobili. 1900. Anchistia tenuipes, Holmes (nec Borradaile), Occas. Papers California Acad. Sci. VII, p. 216. 1904. Peviclimenes tenutpes, Rathbun, Harriman Alaska ea X, Crust., p. 34, text-figs. 12a, b. 1907. Periclimenes holmesi, Nobili, Ann. Mus. Univ. Napoli (2) xT! 1921. Periclimenes tenuipes, Schmitt, Univ. California Publ., Zool. XXIII, p. 39, figs. 24a, b. 1922.] S. Kemp: Notes on Crustacea Decapoda. 219 This, the only species of Periclimenes known from the Pacific Coast of America, is very closely related to P. elegans; in the description which Holmes has given and in the additional notes and figures supplied by Miss Rathbun I am unable to find any differences worthy of note. In view of the widely distant localities in which the two forms have been found, it seems unlikely that they are specifically identical, but this can only be determined by actual comparison of specimens. P. holmesi is known only from the Californian Coast, extending from Santa Catalina I. to the Gulf of California. Periclimenes (Ancylocaris) amymone de Man. 1902. Periclimenes amymone, de Man, Abhandl. Senckenb. naturf. Ges. XXV, p. 829, pl. xxv, figs. 53a-g. In this species, which I have not seen, the legs are conspi- cuously stouter than in any of the related species. In the female on which de Man’s detailed description is based the carpus of the second peraeopod is only about 2°4 times its distal breadth with the merus 1°6 times its length. In the third pair the merus is only 6 times as long as broad and the propodite is five times as long as the dactylus. The dactylus is short and stout, scarcely more than 3 times as long as its basal breadth. P. amymone also differs from all related species in the absence of spinules on the posterior border of the propodite of the last three pairs of legs. The species is recorded from Ternate. Periclimenes (Ancylocaris) demani Kemp. 1915. Periclimenes demani, Kemp, Alem. Ind. Mus. V, p. 279, pl. xili. fig. 10, text-figs. 27 a-7. This species is related to P. grandis, but differs from it and from all other species in the same section of the genus in the structure of. the apex of the antennal scale: the spine which terminates the outer margin reaches only to, or toa very small extent beyond the apex of the lamella (text-fig. 64). It also differs from P. grandis in having the carpus of the second leg of the male as long as the merus and in the proportionately shorter chela, always less than 1'5 times the length of the carpus and in the longer and more slender legs of the last three pairs. As in P. grandis the anterior of the two pairs of spines on the dorsum of the telson is.placed = Text-ric. 64.—Periclimenes in the proximal half of the telson- demani Kemp. length. Antennal scale. C 514/1. Jack and Una Is., Mergui ‘Investigator,’ Dec., One. ‘ Archipelago. 1913. 220 Records of the Indian Museum. [Vo XXIV, P. demani was hitherto known only from the Chilka Lake in Orissa and from the Adyar R. and Ennur backwater near Madras. In the Chilka Lake it has been found living in water ranging in specific gravity from I:000 to 1:0265. Periclimenes (Ancylocaris) lifuensis Borradaile. 1898. Periclimenes lifuensis, Borradaile, Ann. Mag. Nat. iPS (7) I, 84 1899. Peni lifuensis, Borradaile, in Willey’s Zool. Results, p. 405, pl. xxxvi, figs. 1a—c. I have seen the type of this species in the Cambridge Museum, Lut have not dissected it to examine the mandible. It is the only known species of the genus in which the supra-orbital spine is present and the hepatic absent. The antennal scale is not much narrowed apically ; the outer margin is concave and the terminal spine extends beyond the apex of the lamella. The merus of the second peraeopods bears a spine at the distal end of the lower margin. Only one of the posterior legs is in existence; it is very stout, with the propodus only about 4 times as long as broad and with the dactylus simple, strongly curved and partially concealed by thick hairs. The telson bears two pairs of dorsal spines, placed rather close together near the middle of its length. P. lifwensis is known only from Lifu in the Loyalty Is. Periclimenes (Ancylocaris) tenuipes Borradaile. (Plate VIII, fig. 11.) 1898. Periclimenes tenucpes, Borradaile, Ann. Mag. Nat. Hist. (7) Il, p. 384. 1899. Periclimenes tenuipes, Borradaile, in Willey’s Zool. Results, p. 406, pl. xxxvi, figs. 2a-f. 1899. Periclimenes tenuipes, Nobili, Ann. Mus. civ. Genova (2) XX, P» 235. 1904. Periclimenes borradailet, Rathbun, Harriman Alaska Exped. X, Crust., Pp. 34. 1907. Pevielumencs borradailei, Nobili, Ann. Mus. Unga. Napoli (n.s.) II, no. 21, p. 5. 1915. Periclimenes (Falciger) kolumadulensis, Borradaile, Ann. Mag. Nat. Hist, (8) XV, p. 213. 1917. Periclimenes (Falciger) borradailei and kolumadulensis, Borra- dalle, Trans. Linn. Soc. (2) Zool. XVII, pp. 372, 376, pl. lv, g. 17. The rostrum varies greatly in length; in the specimens I have examined it is from 1°6 to twice the length of the carapace, while in the large male described by Borradaile as P. kolumadulensis it is said to be 2'5 times the length of the carapace. The rostrum is very slender, straight at the base, but with the distal half bent obliquely upwards. It bears from*9 to 12 teeth (nearly always 10 or II)! on the HBBSL border ; the posterior tooth is placed on the t Of thirty-two specimens two have g dorsal teeth, Wank have 10, seventeen have 11 and one has 12. 1922.] S. Kemp: Notes on Crustacea Decapoda. 221 carapace, but is not widely separated from the next, which is either above or a little behind the posterior limit of the orbit. In most specimens the upper teeth are arranged in two groups, the five proximal teeth being separated by a marked interval from the four or five distal, with or without a single isolated tooth between the two. On the lower border in the anterior two-thirds there are from 6 to 9 teeth (usually 7 or 8),! extending close up to the apex. The carapace is obtusely angled at the lower limit of the orbit. The antennal spine is strong and is flanked by a short carina; the hepatic is behind it, but on a lower level. There is no supra-orbital spine. The eyes are large and somewhat depressed, with the cornea wider than the stalk. The ocular spot touches the cornea. ' The basal segment of the antennular peduncle bears a short lateral process; the outer margin terminates in a sharp spine which do2s not reach the middle of the next segment. The second and third segments are slender, but the whole peduncle is not long, scarcely reaching beyond the middle of the antennal scale. The free portion of the shorter ramus of the outer antennular flagel- lum is extremely short ; the fused part is longer than the peduncle and is composed of some 12 to 15 segments. The antennal scale in full grown specimets is from 6°5 to 7 times as long as wide and is very narrow distally. The outer margin is strongly concave and the terminal spine projects far beyond the apex of the lamella. The third maxillipeds do not reach the end of the antennular segment ; the ultimate segment is about two-thirds the length of the antepenultimate. The mero-carpal articulation of the first peraeopods reaches the end of the antennular peduncle. The carpus is about 1°5 times the length of the merus, and is from 2 to 2°75 times as long as the chela. The fingers are a little longer than the palm and are un- armed. The second peraeopods in males may outreach the rostrum by the whole of the chela and carpus and a portion of the merus, they are from 5'5 to 7'5 times as long as the carapace. The legs forming a pair are, as a rule, equal and similar instructure. There is a strong spine at the distal end of the merus on the lower side. In the largest male the merus is about 6°5 times as long as wide. The carpus is 1°4 times as long as the merus; it is slender at the base but is suddenly dilated in its distal third, the length being about 7°5 times the distal breadth. On the inner side of the car- pus at the distal end there is a small obscure tooth, much as in Periclimenes agag. The chela is about 1°25 times the length of the carpus ; the palm is 5 times as long as broad and about 1'9 times the length of the fingers. In smaller males the limbs are more slender, with the carpus much less dilated at the distal end. In one stich male the merus is 8 times as long as broad and the carpus ! Of thirty-two specimens four have 6 ventral teeth, sixteen have 7, eleven have 8 and one has 9. 222 Records of the Indian Museuin, ‘[Vot. XEN S 13 times as long as broad, 1°5 times as long as the merus and a little longer than the chela, The palm in this specimen is 1°3 times the length of the fingers. In the smallest male in the collec- tion the carpus is as much as I°4 times as long as the chela. The series of specimens in the collection comprises a number of individuals which, in the proportions of the segments of the second peraeopods, are intermediate between those described above, indicating quite clearly that the differences are due to progressive growth. The second legs of very large males appear to develop in - a phenomenal manner, as in the Hippolytid genus Savon and in Palaemon. ; In ovigerous females the second peraeopods are from 4°5 to 5°7 times as long as the carapace. The carpus is from 1°5 to 1°8 times as long as the merus and from 1:2 to 1°4 times as long as the chela. The palm is about 1°3 times the length of the fingers. In the second peraeopods of some large males each finger is conspicuously excavate in its proximal half. In other males no trace of this excavation is visible; the fingers meet throughout their length when the claw is closed and are armed only with a series of very small teeth, most conspicuous at the proximal end. Specimens in intermediate stages, with the gape in the fingers poorly developed, are not uncommon. As a rule the fingers in both legs of a pair are similatly formed, but I have seen a specimen in which one chela only possessed gaping fingers, as in the type of Borradaile’s P. kolumadulensis. In large females the fingers sometimes exhibit a small excavation, similar to that seen in some large males but less well developed. The last three legs are extremely long and slender, the fifth reaching to or a little beyond therostrum. The merus of the third pair is from 20 to 26 times as long as wide. The propodus is from 4'°5 to 5'5 times the length of the dactylus; it bears some short spinules on its posterior edge and shows traces of subdivision into 5 to 7 subsegments. The dactylus is simple, curved, and with a few setae in the middle of its anterior margin; it is from 6°5 to 7°5 times as long as its basal breadth. The sixth abdominal somite is about one-third longer than the fifth. The foremost pair of dorsal spinules on the telson are situated in the anterior half of the telson, the second pair rather further from the foremost than from the apex. The intermediate apical spines are very long. The largest specimen, a male, is about 22 mm. in length. The species is characteristically coloured when alive. The carapace and abdomen are semitransparent, with a few narrow oblique streaks of white and red on the former and mid-dorsal and lateral red stripes on the latter. On the rostrum, at the junction of the middle and distal thirds, there is a band of dark red pig- ment; in front of this the rostrum is entirely sulphur yellow, while behind it on the inferior half there is a streak of the same colour. The tip of the telson and the basal portions of the uro- podial setae are bright red. The eyestalk has two white longitu- 1922.] S. Kemp: Notes on Crustacea Decapoda. 223 dinal streaks and some red speckling. On the first legs there is a sharply defined red spot at the distal ends of the ischium, merus and carpus. Between the bases of the first legs there is a bright red sternal spot. On the second legs there is a similar spot at the distal end of the ischium and a large red patch at the end of the merus. The carpus is sulphur yellow throughout, the colour ex- tending on to the base of the chela which is otherwise dull red. . The eggs are pale grey, when eyed with a bright blue eyespot. ‘Borradaile’s descriptions of P. tenuipes and P. kolumadulen- sis are both inadequate and I suspect that the figures of the for- mer are erroneous in several particulars. Re-examination of the types is necessary before the synonymy given above can be te- garded as beyond doubt. From the description I have given it will be seen that the range of variation is very great and that the characters which Borradaile gives in his account of P. kolumadul- ensts are insufficient for the distinction of two species. Seeing that the type-specimen of P. tenuipes was damaged it is unfortu- nate that Borradaile contented himself with a mere record of the additional examples obtained by Prof. Gardiner at Haddumati Atoll. Two misconceptions appear to have arisen regarding the pro- per name of this species. Miss Rathbun (/.c., 1904) proposed P. borradatlei under the impression that the name ftenuzpes was pre- occupied by Holmes. Holmes’ species was, however, not described until 1900. Nobili (/.c., 1907) has stated that Leach described a species from the Mediterranean under the name Pertclimenes ten- uipes and that Heller erroneously regarded Brachyvcarpus biungut- culatus as synonymous with this form. These statements apparent- ly led Borradaile in 1917 to abandon his P. tenutpes in favour of P. borradailet. The paper by Nobili was, I believe, written during the dis- tinguished author’s last illness. It is most unfortunate that it should even have been published, for it is evident from internal evidence that it is the product of a disordered mind. The Palae- monid gill-formulae which are given in the paper obviously have no relation to the real facts and the illustrations of the mouth- patts of Brachycarpus can only be regarded as mythical. Teach does not seem ever to have described Periclimenes tenuipes and the species is not referred to by Heller, nor is it a fact, as stated by Nobili, that in his work on the Red Sea Decapoda he himself proposed the name P. borradailet for Borradaile’s P. tenuipes. P. tenuipes may therefore stand as the name of this species, while for the form described by Holmes Nobili’s P. holmesi may be employed. C 461-5/1. Port Blair, Andamans, S. Kemp, Feb., Thirty-five. 4-8 fms. 1915, Feb., Mch., 1g21. 5525/9. Off Ao 34 fms., 6°o1' ‘Investigator.’ One. , 81°16’ E. 22A Records of the Indian Museum. [Vor. XXIV, T have also seen two specimens belonging to the Paris Museum from Mahé in the Seychelles (Alluaud coll.). The specimens from Port Blair were all obtained in Ross Channel on a bottom composed mainly of small corals and sponges. P. tenuipes was originally described from New Britain and has since been recorded by Nobili from Beagle Bay in New Guinea and by Borradaile from Haddumati Atoll in the Maldives and, as P. kol- umadulensis, from Kolumadulw Atoll in the same group. Periclimenes (Ancylocaris) longimanus (Dana). 1852. Anchistia longimana, Dana, U. S. Explor. Exped., Crust. I, p. 579, pl. xxxvii, figs. 6a, b. This species, of unknown locality, is easily distinguished from all other known members of the genus by the extraordinary length of the. antennular peduncle. Jt reaches well beyond the antennal scale and the ultimate segment, according to Dana’s figure, is 6 times as long as wide. Periclimenes (Ancylocaris) digitalis, sp. nov. (Plate VIII, fig. 12.) The rostrum reaches slightly beyond the end of the antennal scale. It is straight at the base, but a little upturned in its distal third. On the upper border, in the single specimen examined, there are II teeth; of these the two hindmost are situated on the cara- pace behind the orbit and the posterior tooth is separated from the next by a rather considerable interval. The remaining teeth are large and evenly spaced except for the foremost, which is small, placed near the tip, and rather remote from the next of the series. On the lower border there are 2 teeth, placed just in front of the middle of the rostral length. The carapace bears sharp hepatic and antennal spines, the former on a lower level than the latter. The lower limit of the orbit is defined by an acute process and there is a conspicuous ridge close behind the orbital margin and parallel with it. Superiorly this ridge ends in a minute tubercle which is probably a vestige of the supra-orbital spine, inferiorly it ends inthe antennal spine. The ridge is almost exactly similar to that found in Palaemonella vesti- gialis but is rather more sharply defined. The eye is large with the cornea spherical and wider than the stalk. The ocular spot is visible, but is partly conflyent with the cornea. The lateral process of the basal segment of the antennular peduncle (text-fig. 65a) reaches barely to the middle of the segment ; the terminal spine of the outer margin is short and the margin between this spine and the articulation of the second segment is convex. The outer flagellum is cleft for only a very short dis- tance; the fused basal part comprises 16 segments and is longer than the peduncle. The antennal scale (text-fig. 650) is a little more than 3 times as long as broad; the outer margin is straight 1922.] S. Kemp: Noles on Crustacea Decapoda. 225 or very slightly concave and ends in a strong spine which projects a trifle beyond the end of the lamella, c b. The exopod of the third maxilliped reach- TEextT-F1G. 65.—Periclimenes digitalis, sp. nov. es nearly to the end , of the antepenultimate segment, the latter _—— oF: oS bearing a series of 8 \ short spines on _ its x\ outer edge. The ulti- \\ mate segment is two- thirds the length of the penultimate. The first peraeopods reach beyond the an- tennal scale by the chela and fully half the length of the car- pus. The carpus is a little longer than the merus and fully 1°4 times the length of the chela. The fingers are longer than the palm and are unarmed. The second peraeo- pods in the single female examined are equal and very slen- der, reaching beyond the scale by the chela, carpus and one-third the length of the me- tus. The merus bears a. Antennule. 4. Antennal scale. a spine at the distal c. Mandible. end of its lower border ; it is rather more than rz times as long as broad and is exactly equal in length with the carpus. ‘The carpus is unarmed and is nearly 9 times as long as its distal breadth. The chela is almost 1°25 times the length of the carpus or merus. The palm is 4'5 times as long as wide and 1°3 times as long as the fingers. The fingers have inturned tips; their cutting edges are entire distally, but in the proximal third are provided with a few small teeth. The last three peraeopods are all very slender. The fifth reach beyond the scale by the dactylus and more than half the propodus. In the third pair the merus is about 18 times as long as wide. The propodus is entirely devoid of spinules on its poste- rior margin and is scarcely more than twice the length of the dactylus. The dactylus itself is simple, slightly curved and extremely slender, about 14 times as long as its basal breadth. TZ (ieee a .cseeoe== — w. A . 226 Records of the Indian Museum. [VoL. XXIV, The sixth abdominal somite is about 1°5 times as long as the fifth. The telson bears two pairs of dorsal spines, so arranged as to divide its length into three more or less equal parts. The outer margin of the external uropod is ciliated. The single specimen is an ovigerous female about 22 mm. in length. In the possession of a post-orbital ridge this species, as already noted, bears a close resemblance to Palaemonella vestigialis; the mandible, however, is devoid of a palp (text-fig. 65c). In the genus Periclimenes it does not appear to have any close allies. C 404/1. Port Blair, Andamans, S. Kemp, Feb., One, Type. 3-5 fms. 1g2t. The specimen was caught off Viper I. on a bottom composed of mud and decaying vegetation. Periclimenes (Ancylocaris) brocki (de Man). 1887. Anchtstia Brockit, de Man, Arch. Naturgesch. III, i, p. 548, pl. xxiia, figs. 3, 3a-d. 1917. Peviclimenes (Cristiger) brocki, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 324. I have examined a specimen from Suvadiva Atoll in the Maldives, determined by Borradaile and have nothing to add to de Man’s detailed description. The species was described from Amboina. Periclimenes (Ancylocaris) rotumanus Borradaile. 1898. Peviclimenes rotumanus, Borradaile, Proc. Zool. Soc. London, p- 1005, pl. Ixiv, figs. 5, Sa, b. 1899. Periclimenes rotumanus, Nobili, Ann. Mus. civ. Genova (2) XX, P: 235. I have seen the type of this species in the Cambridge Mu- seum ; the second peraeopods are now missing. ‘The species is recorded from Rotuma in the S. Pacific (Borradaile) and Beagle Bay, New Guinea (Nobili). Genus Harpilius Dana. -1852. Harpilius, Dana, U. S. Explor. Exped., Crust. 1, p. 575. 1917. Harpiliopsis and Harpilius, Borradaile, Tvans. Linn. Soc. (2) Zool. XVI], pp. 379, 380. 1921. Harpilius, Tattersall, Fourn. Linn. Soc., Zool. XXXIV, p. 338. This genus is very closely related to Periclimenes, agreeing _ with it in all important-struetural characters and differing only in its more clumsy and depressed form. In habit of body there is, moreover, considerable variation; of the species I have myself examined H. beaupresi and H. depressus are very strongly depressed, while in H, lutescens and H. gerlachei this feature is much less pro- nounced, In Harpilius the distal spine of the basal antennular segment is usually very long, the antepenultimate segment of the third 1922.] S. Kreme: Notes on Crustacea Decapoda. 227 maxilliped is often broadened, the second peraeopods are heavily built with the distal end of the merus flattened or hollowed beneath to accommodate the carpus when the limb is folded, and the last three peraeopods are stout, without spinules on the propodus and with a simple strongly hooked dactylus. The combination of these characters gives the species a very distinct facies, though a parallel to each may be found in the genus Periclimenes. Most if not all the species of the genus are found in associa- tion with corals and there can be little doubt that they are speci- ally adapted to life in this environment: the depressed form and © stout legs with hooked dactyli are obviously well suited to an existence among the branching stems of a madrepore colony. In general appearance Harpilius bears a close resemblance to Corallio- carts, the species of which are found in similar situations. , Text-FiG. 66.—Second maxilliped of Harpilius Mate SCCHSH Dana. a. As shown by Dana. an 46. With some of the errors portected: Tattersall has already questioned the validity of the genus Harpiliopsis and I endorse all that he has said. Borradaile’s reasons for establishing the new genus are indeed remarkable. Apart from the supposed absence of the arthrobranch on the third maxilliped in Harpilius, the difference between this genus and Harpiliopsis lies in the form of the second maxilliped. Of Harpilius Borradaile has seen no specimens and his description of the appendage is derived from Dana’s fig. 4f of H. lutescens. In his generic description of Harpilius he says “‘ second maxilliped.... with last joint posterior to preceding joint’’ and adds that ‘ the second maxilliped of the type Harpilius is so remarkable that no species which does not share this peculiarity can be retained in the genus.” It is, of course, evident at first sight that Dana’s figure is erroneous and that the narrowly triangular terminal segment, 228 Records of the Indian Museum. [Vor. XXIV, instead of being attached only by its apex (obviously an impos- sible arrangement), is joinedin normal. fashion to the propodus, the free edge of the latter being almost entirely concealed by the overly- ing ischium and merus. The erroneous division of the propodus into two segments is also seen in the figure of Oedipus superbus on the same plate. I give here (text-fig. 66) a copy of Dana’s figure, together with another in which the more important errors have been eliminated. ‘The latter does not differ in any noteworthy feature from the normal type. One of the specimens I have seen I doubtfully refer to H. lu- tescens. ‘Thisindividual has a normal second maxilliped and, as in H. beaupresi and H. depressus, possesses an arthrobranch on the third maxilliped. In AH. gerlachei, as Tattersall has pointed out, this gill is suppressed and the species is otherwise peculiar in the absence of the hepatic spine.. Tattersall has suggested that a new genus may be required for the species, but with this I am unable to agree and think that if any change is to be made it should be in the direction of merging Harpilius in Periclimenes. Owing to inadequate original description the recognition of Dana’s H. lutescens, Stimpson’s H. depressus and of the form which Ortmann called Anchistia spinigera is attended with much difficulty, and the possibilities of erroneous identification in this paper are enhanced by the fact that the specimens I have seen are all from the western part of the Indo-Pacific region while the des- criptions are based on material found much further to the east. The species of Harpilius, as I understand them, may be separated by the following characters :— A. _ Hepatic spine present. 8. Antero-lateral angles of carapace rounded ; ischium of second leg with at least one spine situated at distal , end of lower border, merus with spine at distal end of upper border, fingers with t to 3 large teeth. C. Antennal spine remote from lower orbital angle and flanked by a carina; hepatic spine on same level as antennal; antepenullimate segment of third maxilliped 3 times aslong as broad; ischium of second leg with 3 distal spines, 1 above and 2 below, carpus with dorsal spine, 1 tooth on dactylus and 2 on fixed finger; R. 4-7: 2-4 . beaupresi (Audouin). Cc’. Antennal spine close to lower orbital angle, with- out carina; hepatic spine on much lower level than antennal ; ‘antepenultimate segment of third maxil- liped 6 times as long as broad; ischium of second leg with 1 distal spine placed inferiorly, carpus without dorsal spine, 2 teeth on dactylus and 3 on fixed finger. D. Merus and palm of second leg each 3 times as long as broad ; posterior pair of dorsal spines of telson placed much nearer to anterior pair than to apex; R. 5-7: 2-5 .. depressus Stimpson. D'. Merus and palm of second leg each 5 times as long as broad; posterior pair of dorsal spines of telson placed midway between anterior pair and apex; R.7: 4 .. =6Var. gracilis, nov. B’. Antero-lateral angles of carapace rectangular ; iS- chium of second leg unarmed, merus without spine at 1922.] S. Kemp: Notes on Crustacea Decapoda. 229 distal end of upper border, fingers with 5 or more small teeth [antepenultimate segment of third maxilliped about 3 times as long as broad], C. Hepatic spine remote from frontal margin of carapace ; last three legs stout, propodus of third pair 4 times as long as broad, at distal end nearly twice as broad as dactylus; R. 7: 1-2 ‘ C’. Hepatic spine situated on frontal margin of carapace; last three legs more slender, propodus of third pair 7 times as long as broad, at distal end scarcely broader than dactylus; R. 7-9: 1-2 ..._ consobvinus de Man. . Hepatic spine absent [ischium of second leg unarmed, merus without spine at distal end of upper border; antepenultimate segment of third maxilliped 3 times as long as broad]; R. 3-5: 1... wif .. gervlachei Nobili. lutescens Dana. A! Harpilius beaupresi (Audouin). 1825. Palaemon beauprestt, Audouin, Explic. somm. des planches de Crust., p. 91, in Savigny’s Descr. d’ Egypte, pl. x, fig. 4 (1800). 2? 1891. Anchistia spinigera, Ortmann, Zool. Fahrb., Syst. V, p. 511, pl. xxxvi, fig. 23. ? 1901. Anchistia spinigera, Lenz, Zool. Fahrb., Syst. XIV, p. 434. 1915. Harpilius Beaupresii, Balss, Denk. math.-naturw. Kl. K. Akad. Wien XCI, p. 26. 1917. Harpiliopsis beauprest, Borradaile, Trans. Linn Soc. (2) Zool. XVII, pp. 324, 379, pl. lv, fig. 21. ' 1921. Harpilius beauprest, Vattersall, fourn. Linn. Soc., Zool. XXXIV, p- 389, pl. xxviii, fig. 8. Borradaile, who gives numerous other references, separates this species from H. depressus merely by the proportions of the antepenultimate segment of the third maxilliped. If, however, I have identified Stimpson’s species correctly, the two differ in a number of important characters. Text-Fic. 67.—Harpilius beaupresi (Audouin). Anterior part of carapace, rostrum, etc. The principal characters of A. beauprest are the following :— (i) The rostrum is rather shallow with from 4 to 7 dorsal teeth (usually 4 or 5) and 2 to 4 (usually 2 or 3) ventral. The posterior dorsal tooth is placed on the base of the rostrum in advance of the hinder limit of the orbit. The midrib of the rostrum is continuous with the orbital margin (text-fig. 67). (ii) The antennal spine is remote from the lower orbital angle 230 Records of the Indian Museum. [VoL. XXIV, and is supported by a carina which extends backwards to a point immediately above the base of the hepatic spine. The hepatic and antennal spines are about on a level with one another and the antero-latera! angle of the carapace is rounded. (iii) The spine on the outer side of the second segment of the antenna is very long. The terminal spine of the antennal scale reaches almost as far forwards as the apex of the lamella. : (iv) The antepenultimate segment of the third ‘maxilliped is broad, scarcely more than 3 times as long as wide. (v) The first peraeopod is slender, with carpus about 8 times as long as its distal breadth and with fingers more than half as long as the palm. (vi) In the second peraeopod (text-fig. 68) the ischium bears three distal spines, one above and two, pwc are smaller, below. The merus has a strong spine at the distal end of its upper border; the lower border ends in a sharp spine on the outer side and in a rounded lobe on the - inner side. The carpus has a sharp spine on the upper and outer aspect of the distal margin and an acute process, sometimes spiniform, on its lower side. The outer margin of the dactylus is straight or slightly concave and on the _ lower surface of the segment there is a sharp longi- tudinal carina. There is a large triangular Text-ric. 68.——Harpilius beaupresi (Audouin). tooth on the inner mar- a. Second peraeopod. gin of the dactylus a b. Fingers of same. little behind its middle point and at the base a rounded protuberance. The tooth fits between two teeth on the fixed finger, the hindmost of which is broad and frequently exhibits a serrated edge. ‘The palm is about 2°5 times as long as the fingers. (vii) In the third pair of peraeopods the merus is about 3°2 times as long as wide. The propodus is much narrower than the merus, about 6°5 times as long as wide, and at the distal end very little broader than the dactylus. (viii) ‘he pleura of the fourth and fifth abdominal somites are acutely pointed infero-posteriorly. b. 1922.} S. Kemp: Noles on Crustacea Decapoda. 231 (ix) The anterior of the two pairs of dorsal spines on the telson is placed a little behind the middle. The posterior pair is midway between the anterior pair and the apex. The largest specimen examined is about 16 mm. in length. Thanks to the excellence of Savigny’s figures the identity of- this species is beyond all doubt. Richters’ Pontonia (Harpilius) dentata, as de Man and Borradaile have suggested, is no doubt synonymous. Borradaile regards Ortmann’s A nchistia spinigera as a synonym of H. depressus, but while it may be true that the specimens he himself recorded under the former name in 1898 and 1899 belong t» Stimpson’s species, it does not seem probable that this is also true of those which Ortmann and Lenz have described. Both these authors refer to the presence of three spines at the distal end of the merus of the second Jeg and this character, so far as I am aware, occurs only .in H. beaupresi. On the other hand Ortmann states that the dactylus of the second leg bears two teeth and the fixed finger three and this applies to H. depressus rather than to H. beaupresi. Further information is necessary before the position of Ortmann’s species can be decided. The specimens of H. beaupresi in the Zoological Survey of India are from the following localities :— 7240/10. Aden. Brit. Mus. One. C 407/1. Tor, Gulf of Suez. R. B. S. Sewell, Eight. 1916. C 408/1. Port Blair, Andamans. J. Wood- Mason. Five. C 4590/1. Port Blair, Andamans. S. Kemp, March, One. 1915: The specimen from Aden had been determined by Miers as Anchistia petitthouarsi (Audouin). I have also seen specimens belonging to the Paris Museum from Mahé in the Seychelles (Alluaud coll.) and from Massouah, Red Sea (Raffray coll.). The species has been recorded from numerous localities in the Red Sea (Audouin, Heller, Paulson, Nobili, Balss) from the Chagos Archipelago and the Maldives (Borradaile) and from Pulo Edam near Batavia (de Man). If Ortmann’s Anchistia spinigera is syn- Onymous the species extends further east ue Samoa (Ortmann) and Laysan (Lenz). Harpilius depressus Stimpson. 1860. Harpilius depressus, Stimpson, Proc. Acad. Sci. Philadelphia, Pp: 38: 1898. Perislimene spinigerts, Borradaile, va Mag. Nat. Hist. (7) Il, , 1899. Perielimene spinigerus, Borradaile, in Willey’s Zool. Results 1903. Hogpilius depressus, Rathbun, Bull. U. S. Fish Comm. XXIII, iil, p. 920, text-fig. 68. 1915. Harpilius depressus, Balss, Denk. math.-naturw. Kl. K. Akad. Wien, XCI, p. 27. 1917. Harpiliopsis depressus, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 380, pl. Ivi, fig. 22. 232 Records of the Indian Muscum. [VoL. XXIV, Harpilius depressus was described by Stimpson from the Hawaiian Is. and I am not altogether certain that the form which occurs on the Indian coast is correctly referred to the same species. The specimens examined differ from the original description in two particulars: there is no difference between the sexes in the form of the third maxilliped and the fingers of the second peraeopod are always more than half the length of the palm. Stimpson’s descrip- tion is very brief and his account of the spines on the segments of . the second leg is inadequate. Further information on the form occurring in the Hawaiian Is. is necessary before the name of the Indian form can be regarded as beyond doubt. The principal characters of the specimens to which I apply the name are the following :— (i) The rostrum is deeper than in H. beaupresi and bears 5 to 7 teeth above (usually 6 or 7) and 2 to 5 below (usually 3 or 4). The posterior dorsal tooth is placed at the base of the rostrum in Text-¥iG. 69.—Harpilius depressus Stimpson. Anterior part of carapace, rostrum, etc. advance of the hinder limit of the orbit. ‘The midrib of the rostrum is continuous with the orbital margin (text-fig. 69). (ii) The antennal spine is placed close to the lower orbital angle and is not supported by a carina. The hepatic spine is placed on a much lower level than the antennal and the postero- lateral angle of the carapace is rounded. (iii) The spine on the outer side of the second segment of the antenna is very long. The terminal spine of the antennal scale does not reach as far forwards as the distal end of the lamella. (iv) The antepenultimate segment of the third maxilliped is broad, about 6 times as long as wide. (v) The first peraeopod is rather stouter tran in H, beawpresv. The carpus is less than 6 times as long as its distal breadth and the fingers are less than half as long as the palm. (vi) In the second peraeopod (text-fig. 70) the ischium bears a single spine, which is large and placed at the distal end of the lower border. The merus is closely similar to that of H. beau- presi. The carpus has one spine only placed on the lower side. 1922.] S. Kemp: Notes on Crustacea Decapoda. 233 The outer margin of the dactylus is convex and the segment does not possess the longitudinal carina seen in H. beaupresi. There are two large teeth on the dactylus fitting between three on the fixed finger. The teeth on the latter occttpy the whole length of the inner margin and the foremost is often broadly rounded. The palm is rather less than twice as long as the fingers. (vii) In the third pair of peraeopods the merus is rather more than 3°5 times as long as wide. The propodus is much narrower than the merus and is from 5’5 to 6 times as long as wide; at the distal end it is not broader than the dactylus. (viii) The pleura of the fourth and fifth abdominal somites are acutely pointed infero-posteriorly. (ix) The anterior of the two pairs of dorsal spines on the a. arash TEXxT-F1G. 70.—Harpilius depressus Stimpson. a. Second peraeopod. b. Fingers of another specimen. telson is placed in the middle of its length. ‘The posterior pair is placed very much closer to the anterior pair than to the apex. The largest specimen examined is about 24 mm. in length. In life the species was closely and elegantly striped with deep blue on a pale grey ground. There was a narrow mid-dorsal stripe of bright yellow on the third abdominal somite and a similar stripe close to the inferior margins of the first three pleura. The tail-fan was transparent olive-green, the uropods were blotched with blue and with milk-white tips. The chelae of the second legs were finely dotted and suffused with green, with yellowish fingers ; the basal segments and the other legs were spotted with blue, the dactyli of the last three pairs being reddish. The eggs were pale brown. C 410/1. Madras Harbour, 4-5 fms. S. Kemp, May, 1918. Five. 234 Reccrds of the Indian Museum. (VOL. xealy, H. depressus was described by Stimpson from the Hawaiian Is. and has since been recorded from that locality by Miss Rathbun. It has also been recorded by Borradaile from Rotuma and the Ioyalty Is. (as P. shinigerus) and from the Chagos Archipelago, the Maldives, Minikoi and the Seychelles, and by Balss from numerous localities in the Red Sea. var. gracilis, nov. A single specimen in the collection differs conspicuously from the remainder in its much moreslender form. It differs from typi- cal H, depressus of the same sex in the following particulars :— H. depressus, typical form. H. depressus var. gracilis. Antennal scale less than 3 times as Antennal scale 3°5 times as long as long as wide and not longer than cara- | wide and considerably longer than cara- pace. pace. Second peraeopod (text-fig. 70) with Second peraeopod (text-fig.71) with both merus and palm about 3 times as | merus 5 times and palm 5'5 times as long long as wide. Palm rather less than | as wide. Palm 2°5 ‘times as long as twice as long as fingers. fingers. Third peracopod with merus about Third peraeopod with merus fully 4°5 3°5 times and propodus 5'5 to 6 times | times and propodus 7 times as long as as long as wide. wide. ‘ Anterior dorsal spines of telson placed Anterior dorsal spines of telson placed about in the middle of its length; pos- | much behind the middle of its length ; terior pair much closer to anterior pair | posterior pair almost equidistant between than to apex. anterior pair and apex. In all other respects the variety closely resembles the typical form. The ros- trum is deep in lateral view and reaches nearly to the end of the antennal scale; it bears 7 teeth above and 4 below. The hepatic spine is present and situated on a lower level than the an- tennal, precisely as in typi- cal H. depressus. The differences in the pro- portions of the chela are very striking and it is pos- sible that the specimem deserves full specific recog- nition; of this, however, I find it difficult to be certain with the small number of specimens which are avail- able. It will be noticed that, apart from the attenuated TEXY-¥IG. 71.—Harpilius depressus var. form of certain appendages, of oe bie § ' the only character by which P P the variety can be distin- 1922.] S. Kemp: Notes on Crustacea Decapoda. 235 guished is the position of the spines on the back of the tel- son. The specimen is 16 mm. in length. 3252/10. Andamans. ‘Investigator.’ One, TYPE. 2? Harpilius Iutescens Dana. 21852. Harpilius lutescens, Dana, U. S, Explor. Exped., Crust. I, p. 570, pl. xxxvii, figs. 4a-h. 21901. Harpilius lutescens, Nobili, Ann. Mus. Univ. Napoli (n.s.) I, ? 1900. Hatpilius lutescens, Nobili, Ann. Sct. nat., Zool. (9) IV, p. 3 21915. Harpilius consobrinus, Balss, Denk. raat -naturw. Kl. K. Akad. Wien, XCI, p. 27. 1921. Harpilius depressus, ‘Vattersall, Fourn. Linn. Soc., Zool. XXXIV, p. 389, pl. xxviii, fig. 7. Dr. Tattersall has very kindly allowed me to examine the specimen from the Red Sea which he recently recorded under the name of Harpilius depressus. I find that this specimen is speci- fically distinct from those which I refer to H. depres- sus and agrees jess closely with Stimpson’s descrip- tion. The second leg has one spine at the distal end of the merus on its lower side, but none on the ischium and carpus, and on the inner margin of each of the fingers there is a series of five Yext-ric. 72.—? Harpilius lutescens Dana. small teeth. Of H. de- Anterior part of carapace, rostrum, cte. pressus Stimpson says, ‘“ Pedes secundi grandes, laeves ; ischii, meri, carpique apicibus dentibus spiniformibus armatis ; manu carapace duplo longiore, digitis palma dimidia brevioribus, intus forte 2—-3-dentatis.”’ Dr. Tattersall’s specimen bears.a very close resemblance to H. consobrinus, but differs from de Man’s exhaustive description in a few points which appear to have specific value. I attribute it with considerable doubt to H. lutescens, the identification presupposing a large amount of error in Dana’s figutes. The principal characters of the specimen are as follows :— (i) The rostrum is deep and bears 7 teeth above atid ‘2 below. The posterior dorsal tooth is situated on the carapace behind the orbit. The midrib of the rostrum is not continuous with the orbital margin, but curves round the orbit in the form of a sharp carina some distance behind the margin proper (text-fig. 72). (ii) The antennal spine is placed close to the lower orbital angle and is not supported bya carina. The hepatic spine is situated below the level of the antennal and the antero-lateral angle of the carapace is sharply rectangular. 236 Records of the Indian Museum. [VoL. XXIV, (iii) The spine on the outer side of the second segment of the antenna is short. The terminal spine of the antennal scale projects well beyond the distal end of the lamel]a. (iv) The antepenultimate segment of the third maxilliped is slightly more than 3 times as long as broad. (v) The carpus of the first peraeopod is about 7°5 times as long as its distal breadth and the fingers are very little shorter than the palm. (vi) In the second peraeopod (text-fig. 73) the ischium is unarmed, ‘The merus has no spine at the distal end of the upper border ; the lower border ends in a spine on the outer side and in a rounded lobe or process on the inner side. ‘The carpus is un- armed. The fingers are bent slightly inwards in relation to the b. TExtT-riG. 73.—? Harpilius lutescens Dana. a. Second peraeopod. 4. Vingers of second peraeopod.. palm and each bears in the proximal two-thirds of its inner margin a series of 5 small teeth. The palm is less than twice the length of the fingers. (vii) The last three peraeopods are stout. In the third pair (see Tattersall’s fig. 7) the merus is 4 times as long as wide. The propodus is as broad as the merus and is barely 4 times as long as wide.' At the distal end the propodus is nearly twice as broad as the dactylus. (viii) The pleura of the fourth and fifth abdominal somites are not acutely pointed infero-posteriorly. ! It is a little too broad in ‘Tattersall’s figure. 1922.] S. Kemp: Notes on Crustacea Decapoda. 237 (ix) The anterior of the two pairs of dorsal spines on the telson is placed a little behind the middle of its length. The posterior pair is midway between the anterior pair and the apex. The specimen bears a very close resemblance to H. conso- brinus. The following are the only points of any significance in which it differs from de Man’s fully detailed description :— (i) The carina behind the orbital margin is not mentioned by de Man. (ii) The hepatic spine is set far back from the frontal margin of the carapace. (iii) The fused portion of.the outer antennular flagellum is composed of It segments. (iv) The carpus of the second peraeopod does not exhibit on its upper side the ‘‘ scharfe kante”’ referred to by de Man; this, however, is not shown in his figures. The palm is slightly more than 1°5 times the length of the fingers, whereas in H. consobrinus it is less than t'2 times. Except that there are only 5 teeth on each finger, the second leg agrees closely in all other respects with de Man’s descriptions and figures. (v) The last three peraeopods are much stouter. In Z. consobrinus the merus of the third leg is 5 times and the propodus 7 times as long as wide. The breadth of the dactylus is scarcely more than half the distal breadth of the propodus, whereas ac- cording to de Man’s figure the two are almost equally broad in Hi. consobrinus. (vi) De Man speaks of three pairs of dorsal spines on the telson in H. consobrinus, but this is perhaps merely an abnormality. The specimen differs from Dana’s figures in a number of points, particularly in the deeper rostrum and in the much stouter carpus and shorter fingers of the second leg. ‘The figures, as de Man has pointed out, are doubtless erroneous in many respects, but the specimen agrees with them and differs from H. consobrinus in the position of the hepatic spine. The specimen from the Red Sea, which Nobili records without comment as H. lutescens, presumably belongs to the same species as that which I have examined. Nobili, however, when writing in 1906, appears not to have been aware that de Man had given the name H. consobrinus to the specimens he formerly described as H. lutescens. The specimens which Balss has recorded from the Red Sea as H. consobrinus also probably belong to this species. Harpilius lutescens was described by Dana from a specimen obtained at Tongatabu in Polynesia. If my identification is correct its distribution extends westwards to the Red Sea. Harpilius consobrinus de Man. 1887. Harpilius lutescens, de Man, Arch. Naturgesch. 1.111, i, p. 536, pl. xxiia, fig. 1. : 1902. Harpilius lutescens, de Man, Abhandl. Senck. naturf. Ges. XXV, p- 836, pl. xxvi, fig. 54. Ternate and Noordwachter Is. 238 Records of the Indian Museum. {VOE. DOxGhy,, Harpilius gerlachei Nobili. 1905. Harpilius Gerlachei, Nobili, Bull. Mus. Paris XI, p. 160. 1907. Harpilius Gerlachei, Nobili, Bull. sci. France Belgique X\., p. 45, pl. iv, figs. 10, 10a. : 1915. Harpilius Gerlachei, Balss, Denk. math.-naturw. Kl. K. Akad. Wien XCI, p. 27. 1921. Harpilius gerlachei, ‘Vattersall, Fourn. Linn. Soc., Zool. XXXIV, p. 390, pl. xxviii, fig. 9. ~ This species is readily distinguished from all other members of the genus by the absence of the hepatic spine of the carapace. It also differs from all, with the possible exception of H. conso- byinus, in the absence of an arthrobranch on the third maxilliped. The principal characters of the species are as follows :— (i) The rostrum is rather shallow and bears from 3 to 5 teeth above, usually 4, and 1 below. The posterior dorsal tooth is placed near the base of the rostrum in advance of the hinder Yext-ric. 74.—Hurpilius gerlachei Nobili. Anterior part of carapace, rostrum, etc. limit of the orbit. The midrib of the rostrum is not continuous with the orbital margin, but curves round the orbit in the form of an ill-defined crest some distance behind the margin proper (text- fig. 74). “Gi The antennal spine is placed close to the lower orbital angle and is not supported by a carina. The hepatic is absent. The antero-lateral angles of the carapace are a little produced, but rounded. (iii) The spine on the outer side of the second segment of the antenna is short. The terminal spine of the antennal scale projects well beyond the distal end of the lamella. (iv) The antepenultimate segment of the third maxilliped is a little more than 3 times as long as broad. (v) The carpus of the first peraeopod is from 5 to 5'5 times as long as its’ distal breadth and the fingers are little more than half the length of the palm. (vi) In the second peraeopod (text-fig. 75) the ischium is unarmed. ‘The merus has no spine at the distal end of the upper border ; the lower border ends acutely on the puter side and in a rounded lobe or process on the inner side. ‘The carpus is unarmed. 1922.] S. Kemp: Notes on Crustacea Decapoda. 239 The fingers are armed in the proximal three quarters of their length with from 3 to 7 teeth, very irregular in their size and distribution. The palm is less than twice the length of the fingers. (vii) In the third pair of peraeopods the merus is about 3°5 times as long as wide. The propodus is as broad as the merus and is about 4°5 times as long as wide; at the distal end it is very little broader than the dactylus. (viii) The pleura of the fourth and fifth abdominal somites are not acutely pointed infero-posteriorly. (ix) The anterior of the two pairs of dorsal Spities Of Text-¥iG. 75.—/arpilius gerlachei Nobili. the telson is placed behind 4A Seeccd aseaeee the middle of its length. b. Fingers of second peraeopod. The posterior pair is midway between the anterior pair and the apex. The largest specimen examined is a female about 18 mm. in length. ; The telson of one of the specimens is abnormal, bearing 5 teeth on one of the lateral margins and 3 on the other. C 412-3/1. Pamban and Kilakarai, S. Kemp, Feb., l‘our. Gulf of Manaar. . 1913: The specimens were all obtained on madrepore coral. Those examined by Nobili were found to the north-east of Arzana I. in the Persian Gulf, ‘‘ parmi les polypiers.” Tattersall’s specimens are from a coral reef at Khor Dongonab in the Red Sea and those recorded by Balss are from the Gulf of Suez, the Red Sea and the S. Coast of Arabia. ZZ Genus Pontoniopsis Borradaile. 1915. Pontoniopsis, Borradaile, Ann. Mag. Nat. Hist. (8) XV, p. 207. 1917. Pontoniopsis, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 377- This genus, of which I have seen no specimens, was erected by Borradaile for a single species, P. comanthi, found on crinoids in the Torres Straits. It appears to be very closely rélated to Periclimenes and Harpilius, but differs in its depressed and tooth- less rostrum, which is lanceolate in dorsal view. Supra-orbital and hepatic spines are wanting and the dactyli of the last three legs are simple. ’ 240 Records of the Indian Museum. [VoL. XXIV, Genus Dasycaris, nov. Rostrum long, laterally compressed, with teeth. Carapace lat- erally compressed, sculptured, with regions well-defined ; antennal and hepatic spines present, each flanked by a strong carina. Anten- nular peduncle with basal segment greatly narrowed distally ; anten- nal scale well developed. Mandible without palp ; inner lacinia of maxillula narrow; all maxillipeds with exopods, the second with- out podobranch, the third slender. Carpus of first peraeopod not divided into subsegments. Last three pairs of peraeopods with strongly hooked dactylus, without basal protuberance and with- out accessory claw. Pleura of third, fourth and fifth abdominal somites drawn out inferiorly into long acute processes. Type and only known species,— Dasycaris symbiotes, sp. nov. This genus is proposed for a remarkable Pontoniine prawn found on Alcyonaria belonging to the genus Pleroeides. In most of its characters the genus resemble Perichmenes, but the carapace is sculp- tured,*the basal segment of the antennular peduncle is strongly narrowed distally and some of the abdominal pleura are produced inferiorly and end in very sharp spinous processes. ‘The dactylus of the posterior legs appears simple under low magnifications, but when stained and examined under a high power it is seen to possess a pit on the posterior margin, through which a fleshy process can apparcntly be protruded. In certain species of Harpilius (H. beauprest and H, depressus) the pleura of the fourth and fifth abdominal somites are acutely produced infero-posteriorly, though not to the same extent as in Dasycaris. In Harpilius, however, the carapace is depressed and not sculptured and the basal antennular segment is very broad. In some respects Dasycaris resembles Nobili’s little known genus Coutierea. The latter, however, is a much more extreme form, with a pterygostomian spine on the carapace and with abnormally developed antennal and supra-orbital spines. In Coutierea, moreover, the dactylus of the posterior legs bears a basal protuberance, indicating affinity with Coralliccaris and Conchodytes rather than with the Pertclimenes group of genera. Dasycaris symbiotes, sp. nov. (Plate IX.) The rostrum reaches to the end of the second segment of the antennular peduncle in the female, to the end of the third segment in the male. It is straight, very slightly upturned at the tip and is extremely shallow in lateral view. It bearsabove 5 sharp teeth; of these the three posterior are placed close together, with two situated behind the posterior limit of the orbit, while the foremost is little, if at all, in front of the middle of the rostral length. ‘Ihe lower borderis unarmed. Behind the rostrum in the middle of the carapace there is another sharp tooth, widely separated from the posterior of those forming the rostral series; this tooth forms the 1922.] S. Kemp: Notes on Crustacea Decapoda. 241 termination of a sharp carina which commences in the posterior quarter of the carapace. The lower limit of the orbit is defined by an acute angula- tion of the frontal margin. The supra-orbital spine is absent. The antennal spine is large, with the hepatic placed behind it on the same level; both spines are supported by strong carinae. The surface of the carapace is uneven; a blunt ridge runs backwards from the lower orbital angle and is separated from the antennal and hepatic spines by a well-marked furrow. There is a similar furrow above this ridge and a large shallow depression on the gastric a. &. TEXxT-FIG. 76.—Dasycaris symbiotes, sp. nov. a. Antennule. c. Third maxilliped. b. Antennal scale. d, Fingers of second peraeopod. region. The upper limit of the branchial cavity is defined externally by a groove and an irregular fold. The eyes are rather slender. The cornea is hemispherical and searcely wider than the stalk and there is no trace of the ocular spot. The basal segment of the antennular peduncle (text-fig. 76a) is externally concave and is remarkably narrow in its distal third; its least breadth is only one quarter its length excluding the terminal spine. The lateral process does not reach the middle of the basal segment and consists of a comparatively broad plate with an acute termination; it thus differs considerably from that of Periclimenes in which the whole process has the form of a simple spine. The 242 Records of the Indian Museum. [VoL. XXIV, terminal spine of the outer margin is very sharp and long, extend- ing beyond the end of the second segment. The second and third segments are broad ‘and the length of the two combined is scarcely more than half that of the basal segment. The free part of the stouter of the two rami composing the outer flagellum is about one-third the length of the fused basal portion, the latter com- prising 6 segments. ‘The total length of the shorter ramus is less than that of the peduncle. The antennal scale (text-fig. 76b) scarcely reaches beyond the end of the antennular peduncle. It is only about 2:2 times as long as broad and the outer margin, which is very slightly concave, ends in a spine which reaches almost as far forwards as the broadly rounded apex of the lamella. The third maxilliped (text-fig. 76c) bears a foliaceous epipod. The exopod does not reach the end of the slightly curved ante- penultimate segment. The ultimate segment is as long as the penultimate. The first peraeopods are slender and reach beyond the anten- nal scale by the chela and a portion of the carpus. The chela is a little longer than the carpus and the merus a little longer than the chela. The carpus is about 6 times as long as wide. The palm is 4 times as long as wide and is twice as long as the fingers. The fingers bear some short hairs, but their inner margins are unarmed. In the male specimen the second pair of peraeopods is very unequal; in the female one leg only, apparently the larger of the two, is present. The larger limb extend beyond the antennal scale by the whole length of the chela and carpus and is covered with minute tuber- cles. The merus is longer than the ischium and is broadest distally, the lower border ending in a strongtooth. The carpus is very short, scarcely longer than broad; it is little more than one-third the length of the merus and is unarmed. ‘The chela is about 2°75 times the length of the merus; the palm is about 3'5 times as long as wide and is from 2:2 to 2'5 times as long as the dactylus. The dactylus is heavy, with strongly convex outer border (text-fig. 76) ; at the base of its inner margin it is provided with a large acute tooth which fits into a cavity in the fixed finger. In front of this cavity the fixed finger bears a small tooth. In the distal two- thirds of their length the inner margin of each finger is entire, the margin is, however, a little concave with the result that a small gap is left when the claw is closed. The tips are inturned and cross one another. In the smaller second leg the tooth at the distal end of the merus appears to be absent and the carpus is nearly twice as long as wide and rather less than half the length of the merus. The chela is 1°65 times the length of the merus, with fingers unarmed and slightly less than half the length of the palm. The three posterior peraeopods are stout ; the third reach beyond the antennal scale by the length of the dactylus. The merus is about 4°5 times as long as broad and is 2°3 times the length of the car- 1922.] S. Kemp: Notes on Crustacea Decapoda. 243 pus. The propodus is conspicuously curved, about 6 times as long as broad and 3 times the length of the dactylus; at the distal end of the lower border there are two pairs of spinules. The dactyli have the form of strong hooks and are about 3 times as long as their basal breadth. The dactyli appear simple under low magnifications, but when stained and mounted and viewed un- der a high power a pit or pore can be detected on the interior side near the base (text-fig. 77). In this pit a fleshy process is lodged and this process is continuous with striated ‘Texr-ric. 77-—Dasycaris symbiotes, sp. nov muscle tissue at the base Dactylus of third peraeopod, from a stained of the dactylus. From preparation. the structure of the parts it seems probable that the process can be protruded through the pit. Examination of living material is necessary before the func- tion of the process can be determined accurately ; it is possible that it acts as a pad and helps the prawn to retain a grip on the host. The abdominal somites are smooth. In both sexes the pleura of the third, fourth and fifth somites are produced inferiorly to long sharply pointed processes. In the male the pleura of the first two somites are pointed at their posterior angles, while in the female the pleura of these somites are rounded, with a small pointed projection in the middle of the lower margin of the second. The sixth somite is rather more than 1°5 times the length of the fifth ; posteriorly it bears a sharp spine on either side of the base of the telson. The telson is shorter than the uropods and possesses two pairs of dorsal spines; the foremost of these is placed a little in front of the middle point of the telson, while the second pair is rather nearer to the first than to the apex. The terminal, telson spines are short. The female specimen is 13 mm. in length, the male about 9°5 mm, With the female there is a note by Col. Alcock which reads ,— ““Transparent grey with dark points on a Pteroeid of exactly similar colour.’’ In A Naturalist in Indian Seas, p. 113, Col. Alcock further says,—‘‘ Another zoophyte that we often dredged was Pteroetdes elegans (ot a species intimately close to it), one of the sea-pens, of a grey colour profusely marked with little, blackish tings. In its leaves three small species of crustaceans are accus- tomed to hide, all of whom are coloured and spotted exactly like the living citadel in which they dwell.’ One of the other crustace- ans associated with the Plerocides is an Alpheid, but what the third is I do not know. 244 Records of the Indian Museum. (VOL. SOSDy - 1729/7. 24 miles E.S.E. of Santa- ‘Investigator,’ Feb., One @, pilli Lt., near Vizagapa- 1890. TYPE. tam, Madras Coast, 15- 17 fms. _ C 406/1. 3 miles E.S.E. of Kabusa ‘Investigator,’ Oct., One G, Is, Mergui, 12°44’30" 1913. Tyee. N., 97°55’30” E., 35 fms. Alcock’s notes refer to the female obtained at the first of these localities. The labels of the male do not indicate that it was found in any particular association. Genus Thaumastocaris, nov. Rostrum well developed, laterally compressed, with large teeth. Carapace laterally compressed, not sculptured. Basal seg- ment of antennule broad; antennal scale well developed. Man- dible without palp; inner lacinia of maxillula narrow; all maxilli- peds with exopods, the second without podobranch, the third slen- der. Carpus of first, peraeopods divided into a number of subseg- ments. Last three peraeopods with dactylus biunguiculate, but without basal process. Pleura of abilominal somites rounded in- feriorly. Type and only known species,—Thaumastocaris strepiopus, sp. nov. This genus is proposed for a Pontoniine prawn from New Cale- donia belonging to the Paris Museum which is remarkable for the fact that the carpus of the first pair of peraeopods is divided into a number of subsegments. In this curious feature it differs, I believe, from all Macrura hitherto known. The carpus of the second peraeopod is frequently segmented in Caridea and the character is of value in distinguishing certain of the families into which the tribe is divided. Much less signi- ficance is, however, to be attributed to the occurrence of the same feature in the first peraeopod of Thaumastocaris, for it is by this feature alone that it can be distinguished from Periclimenes. In Thaumastocaris the hepatic spine is absent and the dactylus of the last three legs biunguiculate. In these points it resembles Perz- climenaeus and I have no doubt that it is in this subgenus or in the closely related Periclimenes s.s. that it finds its nearest allies. It is not easy to decide how much importance should be attributed to a unique character such as that on which this genus is founded ; it is possible that its affinities would be more clearly shown by regarding it merely as a subgenus of Periclimenes. Thaumastocaris streptopus, sp. nov. The rostrum (text-fig. 78) reaches to the end of the antennal scale and is deep in lateral view. ‘The upper border is straight and in the single specimen examined bears a series of 10 closely set teeth which increase in size from behind forwards and are all very large ; the three posterior teeth are situated on the carapace behind the 1922.] S. Kemp: Noles on Crustacea Decapoda. 245 orbit. . The lower border is convex and bears three smaller teeth in its distal half. Text-Fic. 78.—Thaumastecaris stveptopus, sp. nov. Anterior part of carapace, rostrum, ete. The carapace is smooth, without trace of areolation or sculp- ture. The orbital angle is acute; below it there is a sharp antennal spine, but both supra-orbital and hepatic are missing. The eyes are large; the ocular spot is merged in the cornea and the breadth of the cornea is greater than that of the stalk. The antennular pedun- cle (text-fig. 79a) extends nearly to the end of the antennal scale. The later- al process does not quite reach the middle of the basal segment ; the spine at the outer distal angle . is long and the margin between this spine and the articulation of the second segment is a little convex. ‘The free portion of the shorter of the two rami composing the outer antennular flagellum is much shorter than the fused part, the latter com- prising g segments. The antennal scale (text-fig. TEXT-FIG. 79.—Thaumastocaris streptopus, sp. nov. a. Antennule. 6. Antennal scale. 79) is not quite 3 times as long as wide; the outer margin is slightly concave and terminates in a spine which reaches almost to the end of the lamella. 3 246 Records of the Indian Museum. [Vor. XXIV, The distal endite of the maxilla, as in most Pontoniinae, is. divided into two lobes. The third maxilliped extends to the middle of the second antennular segment and is slender. It possesses a small arthrobranch and the exopod does not reach the distal end of the antepenultimate segment. The ultimate segment is less than two-thirds as tong as the penultimate. The first peraeopods (text-fig. 80a) are very long and slender: the mero-carpal articulation reaches to the end of the basal anten- nular segment. The merus is about 14 times as long as wide and is divided by a rather obscure articulation into two subsegments, the distal about two-thirds the length of the proximal. The car- Cc. a. TExt-riG. 80.— Thaumastocaris streptopus, sp. nov. a. First peraeopod. ~~ c. Fingers of second peraeopod. 5. Second peraeopod. - d. Third peraeopod. e. Dactylus cf same. pus is very slender, about 1:35 times as long as the merus and 3°6 times as long as the chela. It is divided by transverse or oblique articulations into six subsegments, the order of which, when arranged according to length is 1, 6, 3, 2, 4,5. The first subseg- ment is twice as long as the sixth, the second, third and fourth are subequal and the fifth, which is the shortest, is about 2°5 times as long as wide. The chela is slender, with fingers unarmed and little more than half the length of the palm. Judging from the size of the basal segments the second peraeo- pods do not differ greatly in size, but only the left limb (text-fig. 80b) is present in the unique specimen. It extends beyond the 1922. ] S. Kemp: Notes on Crustacea Decapoda. 247 antennal scale by the greater part of the chela. The merus is rather less than 3 times as long as wide; it is conspicuously tuber- culate along its lower border, but does not bear a distal tooth. The carpus is a little longer than broad and is about half the length of the merus. Its surface is somewhat uneven and it bears one obscure tubercle on its upper surface and two beneath. There is an excavation in the anterior margin on the inner side and the border above this excavation is obscurely crenulate. The chela is about 3 times as long as the merus and the fingers are a little less than half the length of the palm. The palm is nearly 3 times as long as wide and is rather closely covered with conspicuous tubercles except on the middle of its inner face. The fixed finger is bent at an obtuse angle to the palm. There is a large triangular tooth at the base of the dactylus which fits into a socket in the fixed finger (text-fig. 80c). In the proximal third of the fixed finger there are two teeth separated by a shallow excavation ; the anterior of these is blunt and little developed, the posterior is broad and crenulate on the summit. The tips of the fingers are inturned and cross one another when the claw is closed. The last three peraeopods are stout; the third (text-fig. 80d) reach a little beyond the antennal scale, the fifth to the end of the basal antennular segment. In the third pair the merus is 5°5 times as long as wide and the propodus is 5'5 times as long as the dactylus. In the third and fourth pairs the posterior margin of the propodus is thickly furnished with spinules along: its entire length; in the fifth pair the spinules are restricted to the distal end. The dactylus (text-fig. 80¢) is broad and biunguiculate, with. the accessory claw large. The pleurobranchs as in other Pontoniinae are ive in number, one being situated above the base of each peraeopod. ~The sixth abdominal somite is short; it bears a strong spine on either side of the base of the telson and one at each postero- lateral angle. The telson is flattened above, with two pairs of large dorsal spines. The anterior pair is situated well in advance of the middle, while the posterior pair is midway between the anterior pair and the apex. At the tip of the telson there are as usual 6 spines, the intermediate pair the longest. The median pair is - unusually slender. The species is described from a single male about 24 mm. in length. The specimen is the property of the Paris Museum. It was obtained in August 1890 at Noumea in New Caledonia by Abbé Cullieret. Genus Anchistus Borradaile. ‘ 1898. Anchistus, Borradaile, Ann. Mag. Nat. Hist. (7) I, p. 387. 1917. Anchistus, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 387. The genera Anchistus and Pontonia comprise species which have adopted a more secluded mode of life than any of those contained in the preceding genera. Thespecies of Anchistus live in the mantle - 248 Records of the Indian Museum. [VOL. XXIV, cavity of lamellibranch molluscs, those of Pontonia in a similar situation or in the branchial sac of ascidians. In both genera the prawns probably enter their hosts when larvae and never leave them throughout the whole period of their lives (vide p. 117): The structural changes which they have undergone in response to this remarkable environment are not great. The species are more or less depressed in habit of body and, except for the occa- sional presence of the antennal, all the spines of the carapace have disappeared ; the second legs are very heavy, frequently unequal, and without spines on the ischium, merus or carpus To eacl of these characters a parallel can be found in other genera of the family. The only structural feature of unequivocal value is afford- ed by the inner lacinia of the maxillula, which is very broad and densely covered with hair.' In this respect Anchistus and Pontonia agree with Conchodytes—which also lives in lamellibranchs— and differ from all other genera of Pontoniinae in which the maxillula has been described. The characters available for separating Anchistus from Pon- tonta are very slight, though there can be little doubt that the genera constitute two natural groups of species. In Anchistus the rostrum is laterally compressed in its distal half and frequently bears small teeth at or near the apex. The two distal segments of the third maxilliped * are always slender and are not twisted as in the related genus. The dactylus of the last three legs is either sim- ple and strongly hooked, or is scoop-shaped with the distal part of the anterior border bent inwards, and with an accessory tooth. _ Minor distinctions are to be found in the last abdominal somite and telson. The former is bluntly produced on either side of the telson and with the postero-lateral corners more or less. rounded, whereas these four angles are sharply acute or spinous in Pontonza. In Anchistus the dorsal spines of the telson are very small and in- conspicuous, in Pontonia they are usually large. The distal endite of the maxilla, as in some species of Pon- tonia and most Periclimenes is divided into two lobes. Borradaile recognises five species of this genus and also in- cludes, though with some doubt, Milne-Edwards’ Pontonta armata. ‘This species can never be identified with certainty from the briet description which has been published, and the same remark also applies to A. spinuliferus (Miers). Pesta’s Marygrande mirabilis is no doubt an Anchistus; but the author seems to have confused two distinct species in drawing up his specific description. I have myself seen four species of Anchistus, two of which appear to beundescribed. They are distinguished by the following characters :— A. Rostrum toothless ; antepenultimate segment of third maxilliped very broad, contrasting strongly in width with two distal segments; chela of first leg with its lateral edges 1 SeeBorradaile’s figs. 25e and 26¢ loc. cit., 1917. 2 Borradaile distinguishes Anchistus from Pontonta by the slenderness of these two segments; they are, however, equally slender in some species of Pontonta. 1922.] S. Kemp: Notes on Crustacea Decapoda. 249 produced and bent downwards, the lower surface thus being deeply channelled; dactyli of last three legs simple, less than half as broad at base as distal end of propodus ..,._— terms (Miers). A'. Rostrum with teeth at or near apex ; antepenultimate segment of third maxilliped rather slender, not contrasting strongly in width with two distal segments.; chela of first leg normal in form; dactyli of last three. legs little nar- rower at base than distal end of propodus. B. Wactyli of last three legs normal in form, simple and consisting of a broad basal portion and a slender curved apical claw ; basal segment of antennular peduncle with a short tooth at distal end of outer margin [antennal spine present | a ape be ne B'. Dactyli of last three legs scoop-shaped with distal part of upper border reflected inwards, biunguiculate ; basal segment of antennular peduncle without terminal tooth. ©. Rostrum more or less pointed with teeth on upper border near apex; antennal spine present; dactyli of last three legs with sharp accessory claw and very minute and inconspicuous spinules ts ~ C'. Rostrum squarely truncate with teeth only at the apex; antennal spine absent; dactyli of last three legs with short blunt accessory claw and large spinules sac oe bas ... demani, sp. nov. graviert, sp. nov. mierst (de Man). Anchistus inermis (Miers). 1884. Harpilius inermis, Miers, Rep. Zool. Coll. H.M.S. ‘Alert,’ p. 291, pl. xxxii, fig. B. 1894. Pontonia pinnae, Ortmann, Denk. med.-naturw. Ges. Fena VIII, p. 16. pl. 1, fig. 3. 1906. Pontonia pinnae, Nobili. Ann. Sci. nat., Zool. (9g) 1V, p. 65. 1907. Pontonia pinnae, Nobili, Bull. Sct. France Belgique XL, p. 40, pl. iv, figs. 11-110. 1917. Anchistus inermis and Pontonia pinnae, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, pp. 388, 391. 1921. Anchistus inermis, Tattersall, Fourn. Linn. Soc., Zool. XXXIV, p- 391, pl. xxvii, fig. 4. Other references are given by Borradaile. ‘The principal characters of the species are as follows :— The rostrum is directed downwards, toothless and with the apex broadly rounded in lateral view. The lower limit of the orbit is defined by an acute projection from the frontal margin of the carapace; the antennal spine is either altogether absent or is represented merely by a minute pointed process. The basal seg- ment of the antennular peduncle is produced distally on its outer side in the form of a convex lobe, the outer margin terminating in a short spine. ‘he fused portion of the outer antennular flagellum comprises 5 segments. The antennal scale (text-fig. 81a) is broadly oval and little narrowed anteriorly; thestrongly convex outer border terminates in a rather small tooth which does not reach the distal end of the lamella. The antepenultimate segment of the third maxilliped (text-fig. 81b) is longer than the two distal segments taken together and is very broad; its least breadth is more than three times that of the penultimate segment. Thelower margins of the basis and ischium of the first peraeopods are heavily fringed with setae. The carpus 250 Records of the Indian Museum. [Vor, XXIV, is a little longer than the merus and nearly twice as long as the chela; the fingers are much shorter than the palm. ‘The structure of the chela, as Tattersall has pointed out, is very peculiar. The edge, both on its outer and inner side, is pro- duced to form a sort of flap which is bent downwards and is thickly fringed with long setae on its margin. The chela is thus deeply hollowed in a longitudinal direction when viewed from below and in a transverse section the lower surface would be semicircular (see Tattersall, Joc. cit., fig. 4). The second peraeopods are unequal, either the right or left limb may be enlarged. In the larger of the two the merus is from 2°0 to 2°4 times as long as broad; the carpus is very short, only one-sixth to one-eighth the length of the chela and the fingers are a little more than half thelength of the palm. The dactylus a. Tavwa\ TEextT-FiG. 81.—Anchistus tnerniis (Miers). a. Antennal scale. c. Dactylus of third peraeopod. 6. Third maxilliped. d. Telson. is strongly convex externally. On the inner margin it bears in its basal half a very large triangular tooth and a rounded knob close to the articulation ; when the claw is closed both the tooth and the knob are received into a large socket in the fixed finger. The inner margin of the fixed finger is obtusely produced in the middle and in the basal half, on a crest which borders the socket on its upper side, there are usually from 3 to 6 small denticles, the foremost placed at the summit of the obtuse prominence referred to above. In all well-developed specimens the distal half of each finger is internally concave. The fingers of the smaller limb are similar, but the tooth on the dactylus is usually less well developed. In the last three peraeopods the propodus is without spinules on its posterior edge. ‘The dactylus (text-fig. 81c) is strongly hooked, with the terminal claw bent at right angles to the proxi- | The character is not sexual as suggested by Tattersall. 1922. ] S. Kemp: Notes on Crustacea Decaboda. 251 mal portion. It is extremely slender, the basal breadth being only about half that of the distal end of the propodus. The apex of the telson (text-fig. 81d) is generally armed with six spines. The two forming the median pair are more slender than the intermediates; the outermost are very short and incon- spicuous and are occasionally missing. The dorsal spines are very small and are sometimes absent. When present the anterior pair is placed behind the middle of the telson; with the posterior pair midway between the first pair and the apex. An exceptionally large female is about 39 mm. in length; the majority of the specimens examined do not exceed 26 mm. Living specimens vary in colour from pale straw to bright orange vellow. In females the entire body and legs are covered with minute white dots and the eggs are pale straw, yellow, orange ot brown. Males are semitransparent and lack the white dots found in the female. Dr. W. T. Calman has been kind enough to compare certain specimens which I sent him with the holotype of Miers’ Harpilius inermis.. He writes that the type “‘ agrees exactly with your Indian specimens in the form of the chela of the first leg and in the dorsal spinules of the telson (these are very small, near the decurved lateral edge, and easy to overlook), as well asin all other characters that Ican see. I think there can be no doubt that your specimens belong to Miers’ species.’’ Tattersall is doubtless right in his suggestion that Ortmann’s Pontoma pinnae' is synonymous with this species. The only point of difference concerns the proportionate length of the palm and fingers of the second peraeopod as shown in the figure. On this no teliance can be placed, as Ortmann’s figures are usually inaccurate. I have examined specimens belonging to the Paris Museum which were obtained at the same locality in the Persian Gulf as those which Nobili recorded as Pontonia pinnae and find that they are typical A. tnermts. C 415/1. Port Blair, Andamans. S. Kemp, Feb., Thirty-four. March, 1921. C 4411. Andamans. A. R. S. Anderson. Nine. C 442/1. Paway I., Mergui Archi- ‘Investigator,’ Feb., Two. pelago. 1914. C 416 1. Cheval Paar, Ceylon, 6 T. Southwell, Jan., Four. fms. Feb., 1911. C 458.1. Pamban, G. of Manaar. S. Kemp, Feb., Five, 1913. I have also seen specimens belonging to the Paris Museum from the Pearl banks S.W. of Arzana I. in the Persian Gulf, obtained in Pinna (Bonnier and Pérez coll.) and from Vanikoro, in the Santa Cruz group, Polynesia. The.specimens from Port Blair were all obtained in the mantle- cavity of species of Pinna, a mollusc which occurs in abundance at low water at Brigade Creek and on the shore south of Viper I. ! The specific name used by Ortmann was 1s preoccupied by Lockington in 1879 (see Addendum, p. 287), 252 Records of the Indian Museum. [VoL. XXIV, Every large Pinna which was opened contained a pair of prawns belonging either to this species or to Conchodytes biunguiculatus. One pair of A. tnermis was found in Pinna nigrina Lam., one pair in P. vextllum Born. and the remainder in P. bicolor Gmelin.' The specimens from Pamban were also obtained in Pinna. The species was described by Miers from a specimen obtained in Pinna at Porte Molle in Queensland. It has been recorded from Shark Bay, W. Australia, in Pinna (Miers) ; from the Monte Bello Is.. N.W. Australia, in Pinna (Rathbun); from Penang, “ taken from the infra-branchial chamber of a large Gastropod’’” (Ivanchester) ; from Trincomali in Ceylon (Miiller) ; from the Ceylon Pearl banks (Pearson) ; from the Persian Gulf Pearl banks, in Pznna (Nobili)’; from Dar-es-Salaam, in Pinna (Ortmann)’; and, in the Red Sea, from Djibouti (Nobili)® and Suakin Harbour, in Penna (Tattersall). Anchistus mirabilis (Pesta). 1911. Marygrande mirab:lis, Pesta, Zool. Anz. XXXVIII, p. 571, : text-figs. 1-5. 1913. Marygrande mirabilis, Pesta, Denk. math-naturw. Kl. K Akad. Wiss. Wien LXXXIX, p. 675, text-figs. 31, 32. Pesta appears to have confused two forms when describing this species. The dactylus of the posterior legs is described and figured as simple, but he includes as a variety of the same species a form in which it is biunguiculate (v. Pesta, 1913, text-figs. 31d, e). Judging from the species of Anchistus that I have seen it does not seem possible that these two types of dactylus can be found in one and the same species. ~ The form with simple dactylus is closely related to Miers’ A. inermis, from which, so far as can be ascertained from Pesta’s account, it differs only in the less depressed rostrum with apex more pointed in lateral view. These characters are insufficient and re-examination of Pesta’s specimens is necessary before it is possible to reach any definite conclusions regarding the identity of the species. That it is not synonymous with Miers’ species may be inferred from the fact that it was found in the mantle-cavity of Tridacna gigas, whereas A. tnermts is apparently always associated with Pinna. Pesta’s specimens were obtained at Samoa. Anchistus gravieri, sp. nov. The rostrum (text-fig. 82) reaches to the end of the second segment of the antennular peduncle and is directed downwards. In lateral view it is rather deep, but obliquely truncate terminally | | am indebted to Dr. Baini Prashad for the identification of the species ot Pinna. 2 This is, I believe, the only record of a Macruran from a Gastropod and is doubtless an error. 3 Recorded as Pontonia pinnae. 1922.] S. Kemp: Noles on Crustacea Decapoda. 253 with the apex sharply pointed. On the upper edge close to the tip there are three sharp teeth, placed close together with setae in the interstices." On the lower border there is a small denticle placed Text-Fic. 82.—Anchistus graviert, sp. nov. Anterior part of carapace, rostrum, etc. near the distal end beneath the hindmost tooth on the upper edge. The lower limit of the orbit is-acutely produced and there is in addition a strong antennal spine. The cornea is a little narrower TEXT-FIG. 83.—Anchistus graviert, sp. nov. a. Antennule. 4. Antennal scale. c. Third maxilliped. Cc. than the stalk and the black ocular spot is distinct. The basal segment of the antennular peduncle is produced on the outer side of the articulation of the second segment much as in A. inermis, and the outer margin ends in a small tooth (text-fig. 83a). The 254 Records of the Indian Museum. [VoL. XXIV, fused portion of the two rami composing the outer antennular flagellum consists of 4 segments. The antennal scale is strongly narrowed distally (text-fig. 83b); the outer margin is ‘convex and terminates in a large spine which does not reach as far forwards as _ the sharply rounded distal end of the lamella. The antepenultimate segment of the third maxilliped (text- fig. 83c) is slender, as in A. mzersi, and does not contrast strongly in width with the two terminal segments ; in length it is slightly greater than these two segments combined. The first peraeopods (text-fig. 84a) reach beyond the anten- nal scale by the chela and half the length of the carpus. ‘There are a few setae on the lower borders of the basis and ischium. The merus and carpus are equal in length, each 1°5 times as long as the chela. The palm is normal in form, without the curious structure seen in A. inermis ; the fingers bear tufts of setae and TExt-F1G. 84.—Anchistus graviert, sp. nov. a. First peraeopod. c. Third peraeopod. 6. Second peraeopod. d. Dactylus of same. are somewhat spatulate, unarmed and longer than the palm. In the single specimen examined only the right leg of the sec- - ond pair is present (text-fig. 840). It reaches beyond the anten- nal scale by rather more than the length of the chela. The me- rus is 3 times as long as wide and about 1'5 times as long as the carpus. The carpus is conical, about two-thirds as broad as long and one quarter the length of the chela. The palm is 1°75 times the length of the fingers. In the dentition of the fingers the species resembles A. ineyvmis; the dactylar tooth is, however, smaller and the inner edge of the fixed finger is not angulate and bears six small denticles in the proximal half. The merus of the third leg (text-fig. 84c) is 4 times as lon g 1922. ] S. Kemp: Notes on Crustacea Decapoda. 255 as wide, that of the fifth 4°5 times. The propodus in all three is without spinules on its posterior edge. The dactylus (text-fig. 84d) is simple and short, broad at the base, and with a slender terminal claw which is bent at an angle of 45° to the main axis of the segment. The telson is similar to that of A. imermis. The single specimen is about 20 mm. in length. With this species I have associated the name of Prof. Ch. Gravier, to whom I am indebted for the opportunity of examining a most interesting collection of unnamed Pontoniinae belonging to the Paris Museum. A. graviert is distinguished from A. inermis (i) by the presence of teeth on the rostrum, (ii) by the strong - antennal spine, (iii) by the shape of the antennal scale, (iv) by the slender antepenultimate segment of the third maxilliped, (v) by the form of the chela of the first peraeopod and (vi) by the broad- based dactyli of the last three peraeopods. A. spinuliferus can never be recognised with certainty from Miers’ wholly inadequate description; it appears, however, to differ from the species des- cribed above in its unarmed rostrum. In A. mirabilis (Pesta) the rostrum is also unarmed and the proportions of the segments of third maxilliped are different. A. mtersi, A. demani and A. biung- urculatus are easily distinguished by the “form of the dactylus of the last three legs. The single specimen examined is from Vanikoro, one of the Santa Cruz Is. in Polynesia, and is the property of the Paris Mu- seum. The label does not indicate that the individua! was found in the mantle-cavity of a lamellibranch. Anchistus miersi (de Man). 1888. fearpsles Miersi, de Man, Fourn. Linn. Soc., Zool. XXII, pl. xvii, figs. 6-10. 1906. Anchistus mierst, Nobili, Ann. Sct. nat., Zool. (9) 1V, p. 63. 1917. Anchistus mierst, Borradaile, Trans. Linn. Soc. (2) Zool, XVH, p- 388, pl. lvi, fig. 25. -1921. Anchistus miersi, Tattersalt, Fourn. Linn. Soc., Zool. XXXIV, p- 391- Borradaile quotes other references. De Man has given an excellent description of this species and I have little to add to what he has said. The types of the species are in the collection of the Zoological Survey of India. In the specimens I have seen there are four or five teeth on the upper border of the ros- trum near the apex and one or two very small denticles on the lower side. In a young individual, about 1r mm. in length, an obsolete tooth can be seen at the distal end of the outer mar- gin of the basal antennular segment ; but in well-grown specimens no trace of this tooth remains. The antennal scale is anteriorly narrowed, much as in A. graviert. The second peraeopods vary somewhat in proportions and in the dentition of the fingers. In a specimen from Batavia the carpus, as in the types, is conspicuously longer than broad and the palm is only about 1°5 times the length of the fingers. There 256 Records of the Indian Museum. [Voy. XXIV, are two rather small teeth on the proximal part of the inner mar- gin of the dactylus' and 8 or g small denticles on the proximal half of the fixed finger. In a specimen from Pulo Con- dore the carpus is as broad as long and the palm is nearly twice the length of the fin- gers: there is only one large tooth at the base of the dac- tylus! and 5 denticles on the fixed finger. The distal part of the upper border of the dactylus of the last three legs is bent inwards the segment thus forming a sort of scoop Text-fig. 85.—Anchistus mievsi (de Man), (text-fig. 85). There is a Dactylus of third peraeopod. large tooth on the posterior margin. On the reflected part near the tip of the dactylus there are a number of extremely minute spinules, only visible under a high power of the microscope. The lateral margins of the telson are armed with two pairs of pes small spines arranged in the same way as in A. inermis. 8238/6. Elphinstone I., Mergui J. Anderson, March, Two, Tyres. Archipelago. 1887. C 418/1. Port Blair, Andamans. S. Kemp, Feb., 1921. One, young. The specimen from Port Blair, which is only II mm. in length, was found in a Tvidacna onthe shore at Aberdeen. It was almost transparent when alive, with large sparsely distributed red chroma- tophores. I have also seen specimens belonging to the Paris Museum from Batavia (Reynaud coll.) and from Pulo Condore (Germain coll.). The specimen from the latter locality was found in Tvdacna. <1. mterst has been recorded from Mangareva in the Gambier Is. in the pearl oyster (Nobili), from Funafuti in the Ellice Is. (Whitelegge), from the D’ Entrecasteaux Is., British New Guinea, in Tridacna squamosa (Borradaile), from Elphinstone I. in the Mergui Archipelago (de Man), the Maldives (Borradaile), the Seychelles (Borradaile), from the vicinity of ArzanaI.in the Persian Gulf, in Spondylus (Nobili), from the Red Sea, in Pinna (Nobili) and from Suakin Harbour, in Pinna (Tattersall).?_ The species has thus been recorded from four different genera of lamellibranchs. Anchistus demani, sp. nov. This species is closely related to A. mzersi, but is distinguished by the following characters :—- ! In addition to the rounded knob near the articulation. * Tattersall states that Henderson has recorded the species from the coasts of India, but I have not been able to find the reference. 1922.] A. mtersi (de Man). Rostrum apically pointed in lateral view, with 4 or 5 small teeth on upper border near tip and sometimes with ror 2 denticles on lower border. Antennal spine of carapace present. Dactylus of last three peraeopods scoop-shaped ; apex slender and sharply pointed; accessory spine sharp and conspicuous ; reflected portion of upper margin with fine microscopic spinules (text-fig. 85). Distance between the two pairs: of dorsa! spinules of telson (when present) about equal to distance between poste- rior pair and apex. S. Kempe: Notes on Crustacea Decapoda. 257 A demani, sp. nov. Rostrum squarely truncate at apex in lateral view, the upper end of the trun- cate margin armed with 2 or 3 small teeth (text-fig. 86). Antennal spine of carapace absent. Dactylus of last three peraeopods scoop-shaped; apex broader and less sharply pointed ; accessory spine very blunt and inconspicuous ; reflected por- tion of upper margin entirely covered with rather coarse spinules (text-fig. 88). Distance between the two pairs of dorsal ‘spinules of telson more than twice the distance between posterior pair and apex (text-fig. 87h). l.arger, ovigerous females 25 mm. or Smaller, ovigerous females 10 mm. in more in length. j | length. In other respects the species closely resembles A. mierst. The distal end of the basal segment of the antennular peduncle (text- Text-Fic. 86.—Anchistus demant, sp. nov. Anterior part of carapace, rostrum, etc. fig. 87a) is produced externally beyond the articulation of the second segment and the outer margin does not end in a spine. The fused portion of the outer flagellum is short and is composed of only 3 segments, as against 5 or 6 in A. mierst. The antennal scale (text-fig. 87b) is strongly narrowed distally and the ante- penultimate segment of the third maxilliped (text-fig. 87c) is little broader than the distal segments. In the first peraeopods (text-fig. 87d) the carpus is shorter than the merus and about one-sixth longer than the chela, the fingers are longer than the palm. The second peraeopods are unequal. In the larger of the two (text-fig. 87e) the merus is 2°5 times as long as broad; the carpus is little more than half the length of the merus and is as broad as long. The palm of the larger limb is 2°5 times the length of the merus and is twice or rather more than twice as long as the fingers. The dactylus (text- fig. 87/) bears a large triangular tooth in its proximal third and a knob close to the articulation; the fixed finger has a series of 4 258 Records of the Indian Museum. [VoL XXIV, Text-riG. 87.—Anchistus demani, sp. nov. a: Antennule. e. Second peraeopod. 6. Antennal scale. f. Fingers of second peraeopod. c. Third maxilliped. g. Third peraeopod. d. First peraeopod. h. Telson, or 5 small teeth in its proximal half. In the smaller limb the dentition is similar, but the palm is only about 1°5 times as long as the fingers. The merus of the third peraeopod (text-fig. 87g) is 5 times, that of the fifth peraeopod 5°5 times as long as broad. As in other species of the genus the propodites in all three pairs are without spinules on their posterior edges. TE: The three specimens exa- TeExt-F1G, 88.—Anchistus demani, mined ae from g to 10 mm. Sp. nov. in length: one is an ovigerous Dactylus of third peraeopod. female. 1922.] S. Kemp: Notes on Crustacea Decapoda. 259 - Two of the specimens, which were found together in a large Tridacna, were transparent when alive and dotted all over with pale green chromatophores. The female bore green eggs. The third specimen also found in Tridacna, was transparent with red chromatophores ; it differs structurally from the other two in the cornea of the eye, which is blacker and distinctly wider than the stalk. The affinity of this small species with A. miersz isclearly shown by the similarity in structure of the dactyli of the last three legs. Borradaile’s Anchistus biunguiculatus also possesses biunguiculate dactyli, but their detailed structure has not been described. In this species, however, the rostrum is toothless and the fixed finger of the second peraeopod is straight and is much shorter than the dactylus which is strongly hooked at the end. In A. miersi and A. demant the fingers are of equal length and both have inturned tips. C 419/1. Port Blair, Andamans. S. Kemp, March, - Two, Types. I1QI5. C 420/1. Port Blair, Andamans. S. Kemp, Feb., One. j 1921. - The specimens were all obtained from Tyidacna, found at low water on the shore at Aberdeen. Genus Pontonia Latreille. 1917. Pontonia, Borradaile, Tvans. Linn. Soc. (2) Zool. XXII, p. 380. This genus comprises species which live either in the mantle cavity of lamellibranch molluscs or in the branchial sac of ascidians. In structure Pontonia closely resembles Anchistus, with which it agrees in the very broad and hairy inner lacinia of the maxillula. The species of Pontonia are, however, rather more depressed in habit and the rostrum, though it may be dorsally carinate and with a small ventral keel near the tip, is always toothless and never exhibits the strong lateral compression found in the related genus. The two distal segments of the third maxilliped are frequently but not always broad and the plane of their greatest breadth, as in Conchodytes, is more or less at right angles to that of the preceding segment. This curious disposition is brought about either by a twisting of the antepenultimate segment or by a torsion at the articulation between the penultimate and ante- penultimate segments. The dactylus of the last three legs is simple and not strongly hooked, or biunguiculate, sometimes with a series of spines along the posterior margin. There is a tooth or spine at the distal end of the last abdominal somite on either side of the © base of the telson and the postero-lateral angles are acutely pro- duced. The dorsal spines of the telson are usually large. The antennal spine, as in Anchistus,is present or absent. The distal endite of the maxilla is divided into two lobes in the typical species, P. tyrrhena, but is slender and undivided in the Indian forms. 250 Records of the Indian Museum. (Vou. XXIV, Borradaile distinguishes the genus from Anchistus by the greater proportionate breadth of the two distal segments of the third maxilliped ; they are, however, slender in at least one species of Pontonia.' His statement in the generic diagnosis that the dac- tylus of the last three legs is simple is evidently a lapsus calam1, for it is biunguiculate in P. tyrrhena and in the majority of the species. Nobili? has pointed out that Forsk&al’s Cancer custos* was. obtained in a species of Pinna in the Red Sea and that the name cannot be applied, as it frequently has been, to the Mediterranean species more properly known as Pontoniatyrrhena. Ou the Indian coasts two Pontoniine prawns are found in Pinna, viz. Anchistus inermis (Miers) and Conchodytes biunguiculatus (Paulson). Both of these, if my identifications are correct, also occur in the Red Sea, but Forskal’s description is too indefinite to enable us to decide which was the original of his C. custos. Nobili, moreover, is of the opinion that the name custos is preoccupied by Forskal’s own use of the term on p. 89 of his work in reference to a Pinnotheres. To the genus Pontomia Borradaile assigns ten species, but of these Ortmann’s P. pinnae, as Tattersall has suggested, is synony- mous with Anchistus inermis, while Parisi has pointed out that P. nipponensts de Haan belongs to the genus Conchodytes. Two species from the W. Coast of America are to be added to the genus: P. pinnae Lockington* (mec Ortmann), which Borradaile appears to have overlooked, and P. margarita Smith* which he refers to the genus Conchodytes. Only four species have hitherto been recorded from the Indo- Pacific region, viz. P. brevivostrts Miers® from the Seychelles in “clamp shells,” P. ascidicola Borradaile ® from New Britain in a ascidian, P. minuta Baker’ from §. Australia, a species of un- known association, and P. quadvatophthalma, also of unknown association, recently described by Balss* from N.W. Australia. I have seen only two Indo-Pacific species of this genus, both of which appear to be undescribed. ‘They are related to P. ascidicola and were found lodged in the branchial sac of simple ascidians. Dr. Asajiro Oka, who found these specimens when examining the Indian Museum collection of Tunicata, remarks that judging from their size they ‘‘ must have entered the body of the host as larvae and grown up there to maturity.®’”’ The six Indo-Pacific species of Pontonia may be distinguished by the following characters :— P. okat, sp. nov. Nobili, Bull. scx. France Belgique X1., p. 49 (1907). Forskal, Descr. Anim., p. 94 (1775). See Addendum, p. 287. : Miers, Zool. Coll. H.M.S. ‘ Alert,’ p. 562, pl. li, fig. B (1884). Borradaile, in Willey’s Zool. Results, p. 409, pl. xxxvi, figs. 6a, 6 (1902). Baker, Zrans. R. Soc. S. Australia XXXII, p. 189, pl. xxiv, figs. 9-12 (1907). Balss, K. Svenska Vet.-Akad. Handl., 1.X1, no. to, p. 15, text-fig. 7 (1921). Oka, Mem. Ind. Mus. VI, p. 2 (1915). wonrrne eo oe 1922 ] S. Kemp: Notes on Crustacea Decapoda. 261 A. Dactylus of last three legs simple.! B. Rostrum reaching only to middle of eyes; last two segments of third maxilliped together fully as long as antepenultimate segment _... s a. B'. Rostrum twice as long as eyes; last two segments of third maxilliped together much shorter than ante- penultimate segment ss - Bae A’. Dactylus of last three legs biunguiculate and with a number of spines on its posterior margin. B. Rostrum well developed; eyes normal in form. C. Penultimate segment of third maxilliped 4 timés as long as wide and more than twice as long as ultimate ; carpus of first leg not longer than chela ; dactylus of last three legs with 11-13 spines behind biunguiculate apex *af & ss C'. Penultimate segment of third maxilliped 2°5 times as long as wide and 1'5 times as long as ultimate; carpus of first leg longer than chela ; dactylus of last three legs ? 1 an C’. Penultimate segment of third maxilliped less than twice as long as wide and shorter than ulti- mate; carpus of first leg not longer than chela ; dactylus of last three legs with 4-6 spines behind biunguiculate apex dat 508 Aor B’. Rostrum rudimentary; eyestalks flattened with inner margins contiguous, bearing cornea at outer distal angle _ Fe 5 brevirostris Miers. minuta Baker. okat, sp. nov. ascidicola Borr. anachoreta, sp. nov. qguadratophthalma Balss. Pontonia okai, sp. nov. The rostrum (text-fig. 89) is toothless, broad at the base and very slender at the apex which is acute in both dorsal and ventral views. Dorsally it is cari- nate throughout the greater: part of its length, but there is no suggestion of a ventral keel. It is curved down- wards and is a little shorter . than the eyes, reaching about to the middle of the basal antennular segment. : Text-F1G. 89.—Pontonia okai, sp. nov. The carapace is consider- Anterior part of carapace, rostrum and eye. ably depressed. It bears only.an antennal spine and the lower limit of the orbit is not angulate. The eves are short, with cornea slightly narrower than “ the stalk; the ocular spot appears to be absent. The antennular peduncle (text-fig. goa) is a little shorter than the antennal scale. The basal segment is broad at the base with a large lanceolate lateral process which reaches the middle of its length. The outer margin in front of the process is_ slightly concave and ends in a stout tooth which reaches beyond the middle of the second segment. ‘The second segment is broader than long, the third about the same length and as broad as long. The free portion of the stouter ramus of the outer flagellum is ex- tremely short and the fused portion comprises 3 or 4segments. The ! In P. minuta described as ‘‘simple—or perhaps a little bifid at the tip.”’ 262 Records of the Indian Museum. (VoL. XXIV, terminal segment of the antennal peduncle reaches nearly to the end of the anternular peduncle. The antennal scale (text-fig. 90d) is a little more than twice as long as troad; the outer margin is very slightly convex, terminating in a spine which reaches beyond the end of the lamella. The incisor-process of the mandible ends in 5 teeth and on the inner side near the apex there is a series of 5 or 6 spinules. The third maxilliped (text-fig. goc) reaches the end of the scale ; the exopod is slender and does not reach the distal end of the antepenultimate segment. The latter is about 2°75 times as long as wide; it contrasts strongly in breadth with the two termi- nal segments and is rather longer than the two combined. The penultimate segment is slender, rather less than 4 times as long as wide and more than twice as long as the ultimate segment. The inner edges of the three distal segments are thickly fringed with hooked hairs which retain debris. a. b c TrExt-F1G. 90.—Pontonia okai, sp. nov. a. Antennule. 6, Antennal scale. c. Third maxilliped. The first peraeopod reaches beyond the scale by almost the whole length of the chela. The carpus is about three-quarters the length of the merus and is slightly shorter than the chela; the fingers are about the same length as the palm. There are dense tufts of setae on the fixed finger. The second peraeopods are very large, unequal and dissimilar. The ischium in both legs bears a short tooth at the distal end of its lower border. In the larger limb (text-fig. 91a) the merus is 2°5 times as long as wide; the carpus is shorter than the merus and is very narrow at the base. ‘Ihe chela is swollen and very heavy and bears a few sparse hairs. The palm is about 2°5 times the length of the merus or fingers and is about twice as long as 1922.] S. Kempe: Notes on Crustacea Decapoda. 263 wide. The chela is carinate on its lower edge from the middle of the palm to the tip of the fixed finger. The finger-tips cross one another when the claw is closed; at the base the fixed finger is twice as broad as the dactylus. The dactylus (text-fig. 91d) has a very large triangular tooth in the proximal half of its inner margin. There are two large teeth on the fixed finger. ‘The foremost of these is very large and obtuse, the hindmost smaller, more acute and pointing forwards, the two being separated by a deep and nar- rownotch. When the claw is closed the dactylar tooth partially overlies the posterior tooth of the fixed finger. In the smaller sec- ond peraeopod the merus is a little broader, with more strongly convex borders; the palm is only 1°3 times the length of the merus or fingers. As in the larger limb the fixed finger is twice as broad as the dactylus, but the fingers are unarmed on the inner Text-FiG. 91.—Pontonta okai, sp. nov. a, \.arge second peraeopod. b. Fingers of same. c. Dactylus of third peraeopod. margin except for three very obscure teeth at the proximal end, one on the dactylus and two on the fixed finger. The last three peraeopods are long and slender. The third pair reaches beyond the scale by half the length of the propodus the fifth by the length of the dactylus. The merus is from 6 to 6'5 times as long as wide; the propodus bears a few spinules on its posterior border and is from 3°7 to 4°3 times as long as the dactylus. The dactylus (text-fig. gIc) is straight and slender, more than 4 times as long as wide; it is apically, biunguiculate and the large terminal claw appears to be articulated. Behind the two distal claws there is a series of 11 to 13 spines which are short and 264 Records of the Indian Museum. . [Vou, XXIV, broad where they begin, in the proximal third of the posterior bor- der, and become longer and more slender as they approach the apex. Excluding the terminal spines the telson (text-fig. g2a) is less than twice as long as its basal breadth; it bears two very large dorsal spines on either side. The median and intermediate apical spines are subequal and much longer than the outer. The outer uropod (text-fig. 920) is shorter than the inner, with the spine that wh TEx?-FIG. 92.—Pontonia okat, sp. nov. a. Telson. ~ b. Uropods. terminates the outer border placed close to the distal end. A single pair of specimens of this species has been exainined. The male is about 8°0 mm. is length and the female 8°5 mm. The species is related to Borradaile’s P. ascidicola, the des- cription of which is very meagre, but differs conspicuously in the proportions of the two ultimate segments of the third maxilliped. In P. ascidicola, also, the carpus of the first leg is longer than the chela and the fingers of the smaller second leg are said to be pro- vided with teeth just as in the larger limb of the pair. C 421/1. Off C. Negrais, Burma, ‘Investigator,’ Nov., Two, Tyres. 15°25' N., 93°45' E., 1909. 40-49 fms. The specimens were found by Dr. Asajiro Oka when working at the collection of Tunicata belonging to the Indian Museum. He discovered them in the branchial sac of the type-specimen of Ascidia willeyi, Oka. Pontonia anachoreta, sp. nov. This species is closely allied to P. okai and also lives in asci- dians. It differs from the description given above only in the following points :-— 1922. | S. Kemp: Notes on Crustacea Decapoda. 265 The apex of the rostrum is rather blunt in lateral view and is provided with one or two terminal setae (text-fig. 93). The antennal scale is rather broader, slightly less than twice as long as wide, and the terminal spine does not extend beyond the apex of the lamella (text-fig. 94a). The antepentultimate segment of the third maxilliped (text-fig. 940) is little more than twice as long as wide and the proportions of the two ultimate seg- ments are conspicuously different. The penulti- mate segment is about 1'°7 times as long as wide and is shorter than the ultimate segment. The fingers of the first peraeopod are consider- ably longer than the palm. The second peraeopods do not possess a tooth at the distal end of the lower border of the ischium. The chela of the larger limb (text-fig. 95a) is sharply carinate on the TEXT-F1G. 93.—Pontonia anachoreta, sp. nov. Anterior part of carapace, rostrum and eye. a. lower side throughout its TEXT-FIG. 94.—Pontonia anachoreta, sp. nov. length and is here thickly a. Antennal scale. fringed with very long se- b. Third maxilliped. tae. The dactylus (text- fig. 950) has a large tooth, as in P. okai, but the fixed finger is unarmed ii its distal half and bears at the base two bluntly rounded teeth separated by a broad notch. ‘The chela of the smaller limb is fringed with long setae on its lower border. The fingers have inconspicuous teeth at the base, much as in the related species; their inner margins are, however, concave they gape widely when the claw is closed and their length is almost or quite equal to that of the palm. The merus of the last three peraeopods is rather stouter, from 5 to 5°5 times as long as broad ‘The dactylus (text-fig. 95c) is broader, from 3 to 3°5 times as long as wide and bears only from 4 to 6 spines in addition to the two distal claws. The terminal claw, as in P. okat, is apparently articulated. The telson, excluding the terminal spines, is more than twice as long as its basal breadth, but is otherwise closely similar to that of the related species. 266 Records of the Indian Museum. [VoL, XXIV, A single pair of specimens has been examined; the female is about Io mm. in length and the male about 6 5 mm. In P. ascidicola, according to Borradaile’s figure the penulti- mate segment of the third maxilliped is about 2°5 times as long as > La LP : Y ; b. ie TEXT-FIG. 95.—Pontonia anclorete, sp. nov. a. \.arger second peraeopod. 6. Fingers of same. c. Dactylus of third peraeopod. wide and nearly 1°5 times the length of the ultimate segment. It is thus intermediate in form between P. anachoreta and P. oka. C 422/1. Off Madras Coast, 20 fms. ‘ Investigator.’ Two, Types. The specimens were found by Dr. Asajiro Oka in an ascidian which he has described under the name of Polycarpa annandalet. Genus Pontonides. Borradaile, 1917. Pontonides, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 387. This genus was established by Borradaile for Pontonia maldivensis,'.a species found at Fadiffolu Atoll in the Maldives, which is remarkable for the absence of exopods from all three pairs of maxillipeds. Periclimenes beaufortensis, Borradaile,' from Beaufort in North Carolina, appears from the description to be a related form, but with exopods absent from only the first two pairs of maxillipeds. In both species the rostrum is toothless, but in P. maldivensis the segments of the third maxilliped are broad, whereas in P. beauforten- sis the appendage is described as moderately slender. The dactylus of the last three legs is simple in both species. For the present at ! Borradaile, Ann. Mag. Nat. Hist. (9) V, p. 132 (1920). 1922.] S. Kempe; Notes on Crustacea Decapoda. 267 any rate P. beaufortensis is in my opinion more suitably accom- modated in Ponionides than in any other genus. P. maldivensis is not known to live in any particular associa- tion; P. beaufortensis was found on Gorgonians. Genus Balssia, nov. The remarkable species described by Balss under the name of Amphipalaemon egasti possesses three pairs of terminal spines on the telson and evidently does not belong to Amphipalaemon or to the family Anchistioididae in which, according to Borradaile,' that genus is included. The species is no doubt an aberrant member of the subfamily Pontoniinae and, in the rudimentary character of the exopods of the maxillipeds, resembles Pontonides. It differs from this genus, however, in many respects. Both carapace and abdomen are sculptured ; the rostral crest extends to the posterior end of the carapace and is armed with large teeth; on either side of the carapace there is a supra-orbital ridge armed with three teeth and further back there are two conspicuous tubercles placed one above the other ; mid-dorsally on the first abdominal somite there is a sharp forwardly directed tooth. There is a tubercle on the eye- stalk and a lateral spine on the fifth abdominal somite. In the sculptured carapace and abdomen Balssia bears some resemblance to Dasycaris, but it differs in the other points noted above as well as in the rudimentary exopods of the maxillipeds. Balssia gasti (Balss). 1921. Amphipalaemon gasti, Balss, Mitt. zool. Stat. Neapel XXII, p. 523, text-figs, 1-8. Balssia gasti is known from a single specimen only, obtained in the Gulf of Naples on Corallium rubrum. Genus Coutierea Nobili, 1got. Coutierea, Nobili, Boll. Mus. Torino XVI, no. 415, p. 4 This genus was established by Nobili for Coutiére’s Covallio- carts agassiz,? a species based on a single specimen dredged in 94 fathoms in the vicinity of Barbadoes. ‘The genus is readily dis- tinguished from all other Pontoniinae by the remarkable form of the supra-orbital spines, which are broad and connate with the ros- trun, concealing the eyes in dorsal view, by the huge antennal spines and by the presence of a pterygostomian spine. In the areolation of the carapace and abdomen Coutievea resembles Dasy- carts and Balssia. The two latter genera, however, do not possess the basal protuberance on the dactylus of the last three legs, which is well marked in Coutierea, and they differ also in many other respects. ' Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 405 (1917). 2 Coutiére, Bull. Mus. Paris VU, p. 115, text- figs. (1901). 268 Records of the Indian Museum. (VoL. XXIV, We know nothing of the oral appendages in this genus. Cou- tiére states that the apex of the telson is armed with only two short spines placed close together and it is thus possible that the genus does not belong to the subfamily Pontoniinae. ’ Genus Stegopontonia Nobili. 1907. i hee Nobili, Mem. Accad. Sct. Torino (2) I.VII, P- 3090. This genus was proposed for S. commensalis, Nobili, of which a single specimen, found on the Echinoid, Echinothrix turcarum, was obtained in Hao Lagoon, Paumotu Group, Polynesia. Stego- pontoma differs conspicuously from the related genera in the posses- sion of a double basal protuberance on the dactyli of the last three pairs of legs. ‘The rostrum is depressed, toothless, concave above, and wider near the middle than at the base; the only spine on the carapace is the antennal Nobili gives no description of any of the mouth-parts or of the telson. Genus Coralliocaris (Stimpson). 1852. O0edipus, Dana, U.S. Explor. Exped., Crust. |, p. 572. 1860. Coralliocaris, Stimpson, Proc. Acad. Sci. Philadelphia, p. 38. 1917. Coralliocaris (excluding subgen. Onycocaris), Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 381. Borradaile follows Nobili in recognising two subgenera of this genus, Coralliocaris and Onycocaris, but the proper position of the two species for which the latter name was proposed (see p. 278) appears to me to be very uncertain. I do not think there is any justification for including them in Stimpson’s genus. Thus restricted the genus Covalliocaris forms a compact group of species, all of which so far as is known live in association with madrepore corals, In general facies they agree very closely with Harpilius, which has adopted the same habitat, but they are at once distinguished by the presence of a very large basal protuber- ance on the dactylus of the last three pairs of legs. Coralliocaris is distinguished from Conchodytes by a number of well-marked characters. The rostrum is compressed, dorsally carinate and commonly bears teeth. The antennal spine of the carapace is always present, the hepatic present or absent. ‘The inner lobe of the maxillula is slender and the distal endite of the maxilla is narrow and furnished with setae only at the tip. The dactylus of the last three pairs of legs is provided with a single claw and the basal protuberance, in all the species I have examined, is swollen and hoof-shaped. Borradaile in his synoptic key separates the species of this genus mainly by the number of rostral teeth. In this character, however, there is much variation. Other and better characters will no doubt be found, but at present the descriptions of several species are very imperfect. Miss Rathbun’s C. adlantica' from the ' Rathbun, Bull. VU. S. Fish Comm. XX, p. 122, fig. 26 (1902). . 1922.] S. Kemp: Notes on Crustacea Decapoda. 269 West Indies does not belong to Coralliocaris ; the dactylus of the posterior legs is merely a little swollen and without the large basal process characteristic of the genus. The generic position of C. quadridentata, Rathbun, and C. truncata, Rathbun,’ both from the Hawaiian Is., also appears to me doubtful. The dactylus of the posterior legs is described as having ‘‘an accessory spinule ” in the former and ‘‘ a supplementary spine ” in the latter. Borradaile (loc. cit., 1917, p. 385) erroneously quotes Miss Rathbun’s C. quadridentata as “‘ C. tridentata” and, as the latter name has already been used by Miers, he substitutes ‘‘ C. vathbuni, n. nom.’’ In his key to the species, however, C. guadridentata is: used. The four species which I have myself examined may be separated by the following characters :-— 4d. Hepatic spine absent; fiist legs not remarkably slen- der, with fingers little if at all shorter than palm ; second legs similar in structure. B. Outer margin of dactylus of second leg semicircular ; fixed finger with large molar tooth fitting into cavity in dactylus; ultimate segment of third maxilliped more than 3 times as long as wide; R. 4-6: 1-2 ... gvaminea (Dana). B’. Outer margin of dactylus of second leg not convex ; fixed finger with 2 or 3 teeth which do not fit into cavities in dactylus; ultimate segment of third max- illiped not more than twice as long as wide. C. Rostrum usually with 4 or 5 dorsal teeth and 2 ventral ; merus of second leg with a series of small teeth at distal end of upper border, dactylus with outer margin abruptly angulate st . C'. Rostrum with 1 or 2 dorsal teeth and 1 ventral; merus of second leg unarmed at distal end of upper border, dactylus with outer margin slightly con- cave = be ats ... venusta, Sp. nov. A’. Hepatic spine present ; first legs remarkably slender with palm twice as long as fingers; second legs dissi- milar in structure; R. 3-6: 1-3 a ... lucina Nobili. superba (Dana). Coralliocaris graminea (Dana). 1852. Ocedipus gramineus, Dana, U.S. Explor. Exped., Crust. 1, p.573, pl. xxxvii, figs. 3a-e. 1889. O4cdipus gramineus, Pfeffer,? Fahrb.. Hamburg. wiss. Anstalt Il, p. 34. 1909. Coralliocaris graminea, Calinan, Proc. Zool. Soc. London, p- 706. 1915. Covolliocaris graminea, Balss, Denk. math-naturw. Kl. K. Akad, Wien XCI, p. 26. 1917. Coralliocaris graminea, Borradaile, Trans, Linn. Soc. (2) Zool. XVII, pp. 324, 383. Other references are given by Borradaile, who—no doubt correctly—includes as a synonym Ortmann’s C. inaequalis. ‘The more important specific characters are the following :— ; The rostrum bears from 4 to 6 teeth on its upper margin and ! Rathbun, Bull, U.S. Fish Comm. XXII, iii, p. 920, figs. 69, 70 (1906). 2 IT have not seen this paper. 270 Records of the Indian Museum. [Vor XXIV, I or 2 on its lower!; as a rule there are 5 above and 2 below. The hepatic spine of the carapace is absent. Thethird max- illiped (text-fig. 96) is short and stout and does not reach the distal end of the merus of the first peraeopod. The exopod reaches beyond the middle of the last segment. The penultimate segment is about 1°5 times as long as wide ; the ultimate segment is about 1°3 times the length of the penultimate and is nearly 3°5 times as long as wide. In the first peraeopods’ the merus is a little shorter than the carpus and much stouter, the greatest breadth of the former being about 1°75 times that of the latter. The chela Text-F1G. 96.—Corelliocaris graminea is half as long as the carpus (Dana). and the fingers are a little Third maxilliped. shorter than the palm. The s2cond peraeopods (text-fig. 97) are equal or unequal, but are similar in structure. In full-grown specimens the upper border of the merus is strongly convex in later al view and ends in one or two small spines. The lower border ends;on the outer side, in a large sharp tooth. The carpus bears a large tooth ventrally and the upper portion of the distal margin is cut into a series of 3 to 6 small teeth. The chela is swollen and its breadth near the proximal end is twice as great as at the base of the fingers. The palm is twice as long as wide and twice as long as.the fingers. The fixed finger (text-fig. 97b) is provided with a large blunt molar tooth which occupies the greater part of the proximal half of its inner edge and the margin in front of this tooth is strongly sinuous. The outer margin of the dactylus forms an almost perfect semicircle; at the base of its inner margin there is a large cavity to receive the molar tooth of the fixed finger. An exceptionally large specimen is 23 mm. in length, with chela 16 mm. in length. Specimens obtained at Port Blair, when alive, resembled Dana’s coloured figure. They were pale green throughout, minutely dotted with yellow and dark brown. In ovigerous females there were red streaks on the sides of the abdomen. ! Miers refers to a specimen with only 3 teeth above and none below. 2 This pair of legs in my specimens reaches beyond the scale only by the length of the chela. In Dana's figure they are much longer, but this is doubtless an error. > The chela is viewed obliquely in text-fig. 97a and the full breadth of the palm is not shown. 1922.] S. Kemp: Notes on Crustacea Decapoda. . By C. macrophthalma (Milne-Edwards), as redescribed by Nobili," appears to be closely related to this species, agreeing with it in the stout form of the third maxilliped, in the serration of the distal margin of the carpus of the second leg and in the remarkable form of the dactylus in the same limb. The rostrum, however, bears only 1 tooth above and is unarmed ventrally ; the chela of the first leg is less than half as long as the carpus and the spine at a. : TEXxT-F1G. 97.—Coralliocaris graminea (Dana). a. Second peraeopod. 6, Fingers of second paraeopod. the distal end of the upper border of the merus of the second leg is larger than that at the outer distal angle of the lower border. 235/7- Port Blair, Andamans. A. Alcock, Nov., One. 1888. C 423/1. Port Blair, Andamans. S. Kemp, Feb., Fourteen. 1915. C 424/1. Port Blair, Andamans. R. P. Mullins, June, Seven. 1918. C 425/1. Pamban, Gulf of Manaar. S. Kemp, Feb., Four. 1913. 7239/10. Seychelles. Bie Mi: ts. Alert 5 One. Brit. Mus. 1430. ‘South Sea.’ Purchased. One. I have also seen specimens belonging to the Paris Museum from New Caledonia and Pulo Condore (Harmand coll.). The species has a wide distribution in the Indo-Pacific region. It has been recorded from the Fiji Is. (Dana), Samoa (Ortmann), the Loyalty Is. (Borradaile), Kagoshima, Japan (Ortmann), Hong ! Nobili, Ann. Mus Univ. Napoli (n.s.) 1, 3, p. 3 (1901). 272 Records of the Indian Museum. [Vor XXIV, Kong (Stimpson), Ternate (de Man), Christmas I. (Calman), Pulo Edam in the Bay of Batavia (de Man), Coetivy (Borradaile), Sey- chelles (Miers), Zanzibar (Pfeffer), Dar-es-Salaam (Ortmann), Mozambique (Lenz), Red Sea (Balss). The species lives in associa- tion with madrepore corals. Coralliocaris superba (Dana). 1852. Oedipus superbus, Dana, U.S, Explor. Exped., Crust. I, p. 575, pl. xxxvil, figs. 2a-f. 1915. Coralliocaris superba, Balss, Denk. math.-naturw. Kl. K. Akad. Wiss Wien XCI, p. 26. 1917. Covalliocaris superba, Borradaile, Tvans. Linn. Soc. (2) Zool. II, p. 383. 1921. Coralliocavis superba, Tattersall, Fourvn. Linn Soc., Zool. XXXIV, p. 390. Paulson’s Oedipus dentirostris, as Nobili has pointed out, is a synonym of this species. Borradaile gives a full list of references. C. superba agrees with C. graminea in the absence of the hepa- tic spine of the carapace, in the stout form of the third maxilliped and in the possession of a series of small’teeth at the distal end of the carpus of the second peraeopod. The principal differences are the following :— (i) The ultimate segment of the third maxilliped (text-fig. 98) - 4 is expanded at the base; the inner margin is sinuous, convex proximally and concave distally. The segment is only twice as long as broad and is little if at all longer than the penultimate. (ii) The carpus of the first peraeopod is scarcely longer than the merus and its breadth at the distal end is little less than that of the merus. (iii) The merus of the second peraeopod (text-fig. 99a) is pro- vided with a series of 4 to 6? small teeth on the superior part of its distal margin and the tooth at the outer distal angle of the Text-Fic. 98.—Coralliocaris superba lower border 1S large and trian- (Dana). gular. There is a series of 7 to Third maxilliped. 10 sinall teeth ' on the upper part of the distal border of the carpus. The chela is less swollen; the palm is usually little less than 3 times as long as wide and is 2’5 to 2°8 times as long as the fingers. The form of the dactylus (text-fig. 99b) is entirely different. The outer margin is straight and abruptly angulate in the middle, while | These figures refer to well grown specimens ; in young individuals the teeth are less numerous. 1922.] S. Kemp: Notes on Crustacea Decapoda. 273 on the lower face of the segment in its distal half there is a sharp longitudinal keel. On its inner margin the dactylus bears a single sharp tooth just behind its middle point, which fits between two similar teeth on the fixed finger. In the specimens I have seen the rostrum bears 4 or 5 dorsal teeth and 2 ventral. The lateral process of the antennule is frequently much longer than in C. graminea and sometimes reaches the level of the articulation between the second and third segments. The largest specimen examined is 21 mm. in length. In living specimens the carapace and first four abdominal somites, except for a median intrusion from the fifth somite, were Text-FiG. 99.—Coralliocaris superba (Dana). a, Second peraeopod. 6. Fingers of second peraeopod. pure white. The antennal scale, antennules, rostrum, all the legs, the last two abdominal somites and the greater part of the tatl-fan were pale brown, dotted with very large dark reddish brown chroma- tophores, specially conspicuous on the antennal scales and large chelae. At the end of the tail-fan there was a narrow band of deep blue, bordered with white. This description agrees very closely with Dana’s coloured figure. 236-9/7. Port Blair, Andamans. - A. Alcock, Nov., Four. 1888. . C 427/1. Port Blair, Andamans. S. Kemp, Feb., Nine. 1915. C 428/1. Port Blair, Andamans. R. P. Mullins, June, Three. 1918. 274 Records of the Indian Museum. (VOL. 2Oxahy ; The species is recorded from Tongatabu (Dana), Tahiti (Stimp- son), the Bonin Is. (Balss), Christmas I. ? (Calman), the Noord- wachter Is. and Pulo Edam in the Bay of Batavia (de Man), the south coast of Arabia (Balss) and from numerous localities in the Red Sea (Nobili, Balss, Tattersall). The species is apparently always found in association with madrepore corals. Coralliocaris venusta, sp. nov. The rostrum (text-fig. 100) reaches to the middle or end of the second segment of the antennular peduncle and is directed slightly downwards. In dorsal view it is broad at the proximal end and is dorsally carinate throughout its length; in lateral view it is very slender. In the male there is a single dorsal tooth placed a little in front of the middle point; in the female there are two teeth, the foremost very small, both situated in the anterior third. On the lower border in each specimen there is one small tooth placed close to the apex. The lower angle of the orbit is acute. There is a strong antennal spine, but the hepatic is absent. The antennular peduncle reaches to three-quarters the length . TEXT-FIG. 100.—Corvalliocaris venusta, sp. nov. Anterior part of carapace, rostrum, etc., of male. of the antennal scale. The basal segment is broad. In the female the lateral process extends as far as the articulation between the second and third segments and the terminal spine of the outer margin reaches the middle of the third segment; in the male the spines are shorter. The outer margin of the antennal scale is almost straight, terminating in a spine which does not reach the end of the !amella. The third maxilliped (text-fig. to1@) is short and stout and does not reach the end of the merus of the first peraeopods. The exopod reaches the tip of the endopod. The antepenultimate segment is shorter than the two distal segments combined. The penultimate segment is scarcely longer than broad and is a little longer than the ultimate. The ultimate is much narrower than the penulti- 1922.] S. Kemp: Notes on Crustacea Decapoda. 275 mate and is rather less than twice as long as broad. The inner edges of the last three segments and the greater part of the lower face of the penultimate are thickly clothed with hair. The first peraeopods reach beyond the antennal scale by the chela and the greater part of the carpus. The merus and carpus are moderately stout, the latter slightly longer than the former and 1'7 times the length of the chela. The fingers are almost equal in length with the palm. The second peraeopods (text-figs. 1010, c) are a little unéqual, C. TextT-F1G. 101.—Corvalliocarts venusta, sp. nov. . Third maxilliped. c. Fingers of second peraeopod. . Second peraeopod. d. Third peraeopod. SR but are identical in structure. They extend beyond the scale by the greater part of the chela. The merus is unarmed at the distal end of its upper border, but bears a strong tooth externally at the end of the lower border. The carpus is very short and, as in C. graminea and C. superba, is provided with a stout ventral tooth. In the female the distal margin of the carpus on its upper and outer aspect is cut into anumber of very minute teeth; in the male, which is much smaller, these are not visible. The chela is swollen and is widest near the hase. The palm is from 2 to 2°3 times 276 Records of the Indian Museum. [Vor XXIV, the length of the fingers; in the female it is rather more than 2°5 times as long as wide, inthe male nearly 3 times. The fingers have acute inturnedtips. The dactylus (text-fig. Iorc) is longitudinally carinate in the distal two-thirds of its lower surface, much as in C. superba, but the outer margin is slightly concave and shows no trace of the abrupt angulation found in that species. At its base the dactylus is narrower than the fixed finger. On its inner margin it bears two rather short teeth, the anterior situated a little behind the middle of its length. When the claw is closed these teeth fit between three on the fixed finger; the foremost of the latter is placed a little in advance of the middle. The last three peraeopods (text-fig. 101d) are stout. The merus is. from 2°75 to rather more than 3 times as long as wide. The propodi are strongly curved and the dactyli are provided with a large hoof-shaped basal process and a very slender and strongly curved terminal spine. The telson is slender with the usual six apical spines. The anterior pair of dorsal spinules, as in the preceding species, is placed in the middle of the telson-length with the posterior pair rather nearer to it than to the apex. The female, which is ovigerous, is 10°5 mm. inlength, the male 6°5 mm, C. venusta is allied to C. superba, but differs in the smaller number of rostral teeth, in the form of the third maxilliped, in the absence of spines at "the distal end of the upper border cf the merus of the second leg and in the different form of the fingers in the same appendage. Nobili’s C. camevani' from Flamenco I. in the G. of Panama is perhaps also related, but differs in having no tooth at the distal end of the lower border of the merus of the second leg and only a single tooth on the inner margin of the fixed finger. C 4290/1. N.E. Tholayiram Paar, J. Hornell, Feb., Two, Types. Gulf of Manaar. IOI4. The specimens were found on a madrepore coral. Coralliocaris lucina Nobili. 1901. Covalliocaris lucina, Nobili, Ann. Mus, Univ. Napoli. (n.s.) I, no. 3, P- 5+ 1902. Coralliocaris lamellirostris, de Man, Abhandl. Senck. naturf. Ges. XXV, p. 842, pl. xxvi, figs. 55, 55a. 1906. Covalliocaris lucina, Nobili, Ann. Sci. nat., Zool. (9) IV, p. 57. 1915. Covalliocaris lucina, Balss, Denk. math.-naturw. Kl. BK: tee Wien XCI, p. 26. 1917. Coralliocaris superba var. japonica, and C. lucina, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 384, pl. lvi, fig. 23 1921. Covalliocarislucina, Tattersall, Fourn. Linn. Soc., Zool. XXXIV, P- 390. This species is readily distinguished from the three pEeeeani forms by a number of well-marked characters :— 1 Nobili, Boll, Mus. Torino, XVI, no. 415, p. 3 (1901). 1922.] S. Kemp: Notes on Crustacea Decapoda. 277 (i) The hepatic spine of the carapace is present. (ii) The third maxilliped (text-fig. I02) is very slender. The penultimate segment is fully 2°5 times as long as wide and is slightly shorter than the ultimate segment, the latter being about 5 times as long as wide. The exopod does not neatly reach the end of the endopod. ' (iii) The first peraeopods are extremely slender. Thecarpus varies from 1°8 to 2°5 times the length of the chela. The fingers are only half as long as the palm. (iv) The second peraeopods are unequal and dissimilar in structure. There is a tooth TEXT-FIG. 102.—Coralliocaris lucina externally at the distal end of Nobili. the lower border of the merus, Third maxilliped (arthrobranch omitted). but no terminal spine on the - upper border. The carpus does not possess the large ventral tooth found in the preceding species and the superior part of the distal margin is entire. In the larger chela the palm is slender and from 3°5 to 4 times the length of the fingers. The fingers are twisted, so that the chela opens almost vertically instead of horizontally. As in S. superba the dactylus is longitudinally carinate on its outer face and is abruptly angulate in the middle of its outer margin. On the inner margin of the dactylus there are 2 or 3 teeth placed near the middle and, when the claw is closed, the cutting edge of the dactylus fits between two slightly oblique crests on the fixed finger, that nearest the base bearing 2 or 3 small teeth. In the smaller chela the fingers are about two-thirds the length of the palm. The fingers have straight unarmed inner margins, but each is externally excavate, so that the whole chela, when viewed from the outer side is spoon-shaped. In the specimens I have examined there are from 3 to 6 teeth (usually 4 or 5) on the upper border of the rostrum and from I to 3 (nearly always 2 or 3) on the ventral border. De Man describes the apex of the telson as armed with 16 to 18 spines—a remarkable feature not known in any other Pontoniid. In most of my speci- mens only the usual 6 terminal spines are to be found, but I have seen an individual in which there are 9. The largest specimen examined is about 16 mm. in length. When alive the species is transparent, with colourless chelae and with the carapace and abdomen longitudinally streaked and speckled with bright red. It is possible, as de Man has suggested, that this species is the same as Stimpson’s C. lamellirostris. The description of the latter 278 Records of the Indian Museum. [Vo,. XXIV, is, however, very defective, so that it seems best to retain Nobili’s name. ‘The specimens which Borradaile referred to C. superba var. japonica doubtless belong to this species; his figures agree very closely with specimens I have examined. The only discrepancy is that Borradaile has apparently omitted to notice that his specimens are distinguished from C. superba by the presence of the hepatic spine. 8985/6. Rutland I, Andamans. ‘Investigator,’ Nov., - One. 1887. C 430/11. Port Blair, Andamans. S. Kemp, Feb., 1915. Fourteen. C 431/1. Port Blair, Andamans. J. Wood-Mason. Three. C 432/1. Cheval Paar, Ceylon. T. Southwell, Nov., One. IQIO. C 433/1. Red Sea. Mus. Zool. Napoli. Two, Co- TYPES. The species has been recorded from Ternate (de Man), from the S. Coast of Arabia (Balss) and from numerous localities in the Red Sea (Nobili, Balss, Tattersall). Borradaile (loc. cit., p. 324) has recorded the species under the name of C. japonica from the Mal- dives, the Chagos Archipelago and Saya de Malha. Like other species of the genus, C. lucina appears to be associated with madrepore corals. Genus Onycocaris Nobili. 1906. Covalliocaris subgen. Onycocaris, Nobili, Anz. Sct. nat., Zool. (9) IV, p. 60. Nobili has proposed Onycocaris as a new subgenus of Coraliio- carts for the reception of two species, C. aualitica and C. vhodope, both obtained at Djibouti in the Red Sea. In C. aualitica the dactylus of the last three pairs of legs bears a large accessory claw and is denticulate and slightly swollen at the base. In C. rhodope the accessory claw is very short, scarcely larger than the denticulations which exist on either side of it and the basal part is not swollen. I have already expressed the view that those two remarkable species cannot be included, even under a distinct subgeneric head- ing, in Stimpson’s Coralliocaris, and with the information we at present possess it appears to me to be impossible to arrive at any satisfactory conclusion regarding their true position. I have been obliged to omit Onycocaris from my synoptic key to the genera of the subfamily. Nobili, as usual, has failed to give any description of the mouth-parts and the two species seem to differ so widely from one another that it may be doubted whether there is any real generic affinity between them. In O. aualitica the spine at the distal end of the antennal scale is wanting and the outer margins of the uropods are said to be finely denticulate. ‘These characters do not occur in C. rhodope, nor so far as I am aware in any other species of the subfamily. 1922.] S. Kemp: Notes on Crustacea .Decapoda. 279 Genus Conchodytes Peters. 1917. Conchodytes, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 392. ; The species of this genus live, probably without exception, in the mantle cavity of lamellibranch molluscs. In the possession of a basal protuberance on the dactyli of the last three legs they resemble Coralliocaris, but they are easily distinguished by a num- ber of well-marked characters. The rostrum is depressed and toothless, without a dorsal carina. The lower angle of the orbit is produced, but neither antennal nor hepatic spines occur on the carapace. The inner lobe of the maxillula is very broad and the distal endite of the maxilla is broad and furnished with setae along the whole length of its inner margin. The dactylus of the last three pairs of legs is provided with two curved claws and the basal - protuberance is flat, not swollen and hoof-shaped as in Corallio- carts. , Borradaile recognises five species of this genus, but one of them, C. margarita (Smith),' belongs in my opinion to the genus Pontonia, in which it was originally described. Pontonia nippo- nensts, which Parisi has recently shown to be a true Conchodytes, must be added to the genus. If my identification is correct C. biunguiculatus is represented in the collection I have examined. This species was described by Paulson from an abnormal specimen in 1875 and has.not since been rediscovered. The host of C. nipponensis is unknown ; C. biunguiculatus lives in Pinna, while, C. ividacnae and C. meleagrinae are usually associ- ated with the genera of molluscs to which their specific names refer. All the species are closely related to one another and it is difficult to find valid characters for their separation. This is especi- ally true of C. ividacnae and C. meleagrinae which are perhaps not specifically distinct. The former is apparently restricted to Tridac- na; the latter is generally found in Meleagrina but according to Borradaile sometimes also occurs in Tridacna. The four Indo-Pacific species may be separated by the follow- ing characters :— A. Basal process of dactylus of last three legs with a small tooth on proximal side; posterior of the two pairs of spines on back of telson situated about mid- . way between first pair and apex. ‘ B. Antepenultimate segment of third maxilliped less than twice as long as broad ; fixed finger of second leg with foremost tooth very broad and low, occupy- ing greater part of distal half; lateral spines of telson tip situated at apex es AC a .. btunguiculatus (Paul- son). B'. Antepenultimate segment of third maxilliped rather more than twice as long as broad , feetlindoes of second leg with foremost tooth small and triangular ; lateral spines of telson tip shifted forwards on to dor- sal surface, not nearly reaching apex _... ... aipponensis (de Haan), ! See Addendum, p. 287. 280 Records of the Indian Museum. [VoL. XXIV, A’, Basal process of dactylus of last three legs without tooth ; posterior of the two pairs of spines on back of telson situated much nearer to apex than to first pair. B&. Rostrum reaching end of scale; outer distal angle of basal antennular segment rounded; carpus of first leg as long as or longer than merus . tvidacnae Peters. B’, Rostrum not reaching end of scale; outer distal angle of basal antennular segment acute; carpus of first leg conspicuously shorter than merus ... meleagrinae Peters. The two last-named species, as Borradaile has suggested, are perhaps not specifically distinct from one another. Conchodytes biunguiculatus (Paulson). 1875. Pontonia biunguiculata, Paulson, Crust. Red Sea, p. 111, figs. I, la-n. 21893. Pontonia tvidacnae, Henderson, Trans. Linn. Soc. (2) Zool. V, P- 438. 21905. Conchodytes meleagrinae, Pearson, Ceylon Pearl Oyster Rep. IV, p. 77 21906. Conchodytes meleagrinae, Nobili, Ann. Sci. nat., Zool. (9) IV, P. 77 (part). The specimen figured by Paulson possesses a large protuber- ance on the outer side of the dactylus of the right second peraeo- pod, but this, as Nobili has suggested, is probably an individual abnormality. If this be conceded there is little doubt that the specimens which I record here are correctly identified. TEXT-FIG. 103.— Conchod ytes biunguiculatus (Paulson). a. Third maxilliped. .c. Dactylus of third peraeopod. 6, Fingers of second peraeopod. d. Yelson. The characteristic features of the species are the following :— (i) The rostrum is sharply pointed in dorsal view and falls short of the apex of the antennal scale, usually not reaching the end of the antennular peduncle. (ii) The outer margin of the basal segment m the antennular peduncle terminates in an acute point. 1922.] S. Kemp: Notes on Crustacea Decapoda. 281 (iii) The antepenultimate segment of the third maxilliped (text-fig. 103a) is broad ; its greatest breadth is more than half its length and at the distal end it is conspicuously wider than the penultimate segment. The latter is rather less than twice as long as wide and is a little longer than the ultimate segment. (iv) The carpus of the first peraeopods is about equal in length with the merus. (v) There is one tooth on the dactylus of the second peraeo- pod (text-fig. 103d) and two on the fixed finger, all of which are rounded and, as a rule, finely serrate. The anterior tooth of the fixed finger has the form of a very broad and gently convex lobe. (vi) ‘The last three peraeopods are comparatively slender. In the third pair the merus is from 3°5 to 4 times and the propodus from 4°5 to 5 times as long as broad. ‘The terminal claw of the dactylus (text fig. 103c) is bent at an angle of about 45° to the main axis of the segment and the basal protuberance bears a short tooth on its proximal side. (vii) The dorsal spines of the telson (text-fig. 103d) are very large, fully one-sixth of the total length (terminal spines excluded). The distance between the posterior pair and the apex is equal, or almost equal, to the distance between the two pairs. The lateral apical teeth are comparatively large and are situated at or very near the distal end'; the intermediate pair is conspicuously stouter than the median. Large females sometimes reach a length of 35 mm.; males do not exceed 25 mm. Living specimens are semitransparent and colourless or pale yellowish when alive. Females are closely sprinkled with minute white dots, with the eggs and ovary very dark brown. 4910/10. Andamans. A. R. S. Anderson. Thirty-five. C 434/1. Port Blair, Andamans. S. Kemp, Feb., Twenty-five. March, 1921. The specimens I have myself found were all obtained in Pinna bicolor, Gmelin,? a mollusc which is common at low water in Bri- gade Creek and on the shore south of Viper I. The same lamelli- branch also harbours Anchistus inermis, a prawn which is almost identical with Conchodytes biunguiculatusin colouration. Practically every large Pinna which was opened contained a pair of either the Conchodytes or the Anchistus, but the two species were never dis- covered in the same mollusc. The species was described by Paulson from the Red Sea. I think it probable that the specimens from Pinna recorded by Nobili and Pearson from the Red Sea and from Cheval Paar in the G. of Manaar, under the name C. meleagrinae, belong to this species. The only other record of a Conchodytes from Pinna is that of ! They are placed further forwards in Paulson’s figure than in any specimen J have seen. 2 | am indebted to Dr. Baini Prashad for the identification of this species. 282 Records of the Indian Museum. [VoL. XXIV, Miers,' who refers to a dried and imperfect specimen obtained in this mollusc at Keppel I., Port Curtis, Queensland. Conchodytes nipponensis (de Haan). 1849. Pontonia nipponensis, de Haan, in Siebold’s Fauna Faponica, Crust., p. 180, pl. xlvi, fig. 8 (Hymenocera niponensis on plate) . 1914. Pontonia nipponensis, Balss, Abhandl, math.-phys. Kl. K. bayer. Akad. Wiss., Suppl. Bd. II, p. 53, fig. 33. ‘ Pontonia nipponensis Borradaile, Trans. Linn. Soc. (2) Zool. 1917. XVII, p. 391. 1919. Conchodytes nipponensis, Parisi, Atti Soc, ztal. Sct. nat. I.VII1, p. 75, text-figs. 5, 6. The principal characteristics of this species are the following :— (i) The rostrum is sharply pointed in dorsal view ; it falls short of the apex of the antennal scale, reaching to the base or middle of the second segment of the antennular peduncle. (ii) The outer margin of the. basal segment of the antennular peduncle terminates in an acute point. TEx?T-riG. 104.—Conchodytes nipponensis (de Haan). a. Third maxilliped. c. Dactylus of third peraeopod. b. Fingers of second peraeopod. d. Telson. (iii) The antepenultimate segment of the third maxilliped (text-fig. 104@) is comparatively narrow ; its greatest breadth is less than twice its length and at the distal end it is not much wider than the penultimate segment. The penultimate segment is about 1°6 times as long as wide and is equal in length with the ultimate. (iv) The carpus of the first peraeopods is equal to or slightly longer than the merus. (v) In the single specimen examined there is a large tooth, which is apically serrate, in the proximal half of the dactylus (text-fig. 1040) and, in front of it, another tooth,* much lower but more sharply pointed. ‘There are two teeth on the fixed finger, I Miers, Zool. Coll. H. M.S.‘ Alert,’ p. 291 (1884). 2 Parisi in his description states that there is only one dactylar tooth. 1922.] S. Kempe: Notes on Crustacea Decapoda. 283 ohe at the base which is small, rounded and serrate and another situated in the middle of the finger which is triangular and fits between the two dactylar teeth. The latter is not serrate and is very different from the low broad-based lobe found in C. biunguiculatus. (vi) The last three peraeopods are slightly stouter than in the preceding species. In the specimen examined the merus of the third pair is 3°I times and the propodus about 4 times as long:as wide. The dactylus (text-fig. 104c) is similar to that of the pre- ceding species and bears a short tooth at the proximal end of the basal protuberance. (vii) The dorsal spines of the telson (text-fig. ro4d) are very large, rather more than one-sixth of the total length (terminal spines excluded). The outermost pair of distal spines is shifted forwards on to the dorsal surface of the telson and, though they are very large, their tips do not reach the apex. There are thus in this species three pairs of dorsal spines and two at the tip. The intermediate pair of dorsal spines is situated just behind the middle of the telson and the distance between them and the apex is greater than that which separates them from the anterior pair. Of the two pairs of spines at the apex the outer are slightly stouter than the inner. The single specimen examined is a male 16 mm. in length. Parisi notes that one of his examples was 23 mm. in length. C. nipponensis is easily distinguished by the unusual position of the outermost terminal spines of the telson. The character. is evidently not an individual abnormality as it is shown in Parisi’s figure and referred to in his description. Except for the somewhat shorter rostrum the anterior parts of my specimen agree precisely with Balss’ figure. C 435/1. Misaki, Japan. N. Annandale, 1915 One. (Misaki Lab.). Although this species is here recorded for the fourth time, we are still without information as to the mollusc in which it lives. The species is known only from Japan. De Haan gives no definite locality for his specimens: those recorded by Balss and Parisi were from Sagami Bay. Conchodytes tridacnae Peters. 1917. Conchodytes tridacnae, Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 393. The specimens that I refer to this species agree in the following poisits :— (i) The rostrum in dorsal view is rather bluntly pointed and reaches to or a little beyond the end of the antennal scale. (ii) The outer margin of the basal segment of the antennular peduncle (text-fig. 105@) is distally rounded and not acutely pro- duced as in the other Indo-Pacific species of the genus. (ili) The antepenultimate segment of the third maxilliped is about 2°5 times as long as wide and at the distal end is not much e 284 Records of the Indian Museum. [Vo,. XXIV, wider than the next segment. The penultimate segment is about — twice as long as wide and is considerably longer than the ultimate segment. (iv) The carpus of the first peraeopods is equal to or longer than the merus. (v) There is a rounded tooth which is frequently serrate on the inner margin of the dactylus just behind its middle point. On the fixed finger there are two teeth, both of which are frequently low and inconspicuous. The proximal tooth is sometimes serrate ; the distal tooth is small, never broad at the base asin C. biunguicu- latus, and is occasionally acute. (vi) The last three peraeopods are stout. In the third pair the merus is from 2°5 ue 3 times and the propodus from 2°75 to 3 times as long as wide. The terminal claw of the dactylus is bent at right angles to the main axis of the seg- ment and its basal pro- tuberance is rounded, without a tooth on the proximal side. (vii) The dorsal spin- ules of the telson (text- fig. 1050) are small, only about one-ninth the total length (ter- minal spines excluded). In females the distance Ge, b between the posterior TrexT-riG. 105.—Coxchodytes tyidacnae Peters. pair and the apex is a Natennule: usually from one-third b. Telson. to one quarter,’ in males from one-third to two-fifths the distance between the anterior and posterior pairs. The outermost terminal spines are very small and are placed at the apex; the intermediate spines are not conspicuously stouter than the median. The largest Indian specimen is a female 27 mm. in length; a female from the Torres Straits is 34 mm.inlength. In an extremely young individual, about 7°5 mm. in length, the accessory spine on the dactylus of the last three peraeopods is not developed. Specimens obtained at Port Blair were semitransparent when alive. In females the carapace and abdomen were thinly sprinkled with small white chromatophores, with similar red chromato- phores on the rostrum and anterior parts of the carapace. The 1 The only exception is a large female from the Torres Straits in which the distance between the posterior teeth and the apex is slightly more than half that separating the two pairs. 1922. ] S. Kemp: Notes on Crustacea Decapoda, 285 eggs and ovary were orange or orange-red. In males the white chromatophores were usually absent and the red less numerous. C 436-7/1. Port Blair, Andamans. S. Kemp, March, Thirteen. 1915; Feb., 1921. C 438/1. Cherbaniani Reef. ‘Investigator,’ Oct., Four. Laccadives. 1891. 7421/10. Torres Straits. Brit. Mus. One. All the specimens were found in Tridacna. At Port Blair they were obtained on the shores of Aberdeen and North Bay in mol- luscs chiselled out of solid coral rock. The prawn was compara- tively scarce and was found in only a small proportion of shells that were opened. | C. tridacnae is apparently found only in Tridacna, but in view of Borradaile’s statement that C. meleagrinae sometimes oc- curs in this genus of molluscs it is difficult to determine the dis- tribution of the species with accuracy from the numerous pub- lished records. The species is in all probability widely distributed in the Indo-Pacific region. Conchodytes meleagrinae Peters. 1917. Conchodytes meleagrinae, Borradaile, Zvans. Linn. Soc. (2) Zool. XVII, p. 393. The question of the validity of this species and of the charac- ters by which it may be separated from the very closely allied C. tvidacnae has been discussed by Borradaile. I have myself seen only four specimens of Conchodytes from Meleagrina and two of these are in bad condition. They differ from C. tridacnae in two of the characters mentioned by Borradaile: the rostrum does not reach the end of the antennal scale and the carpus of the first peraeopod is conspicuously shorter than the merus. The third maxilliped is, however, similar in length to that of the related species and does not nearly reach the end of the scale. The specimens also differ from C. tvidacnae in the following points: (i) the outer margin of the basal segment of the antennu- lar peduncle terminates acutely ; (ii) the ultimate segment of the third maxilliped is a little longer, about equal in length with the penultimate; (iii) the last three peraeopods are rather more slen- der—the metus of the third pair is from 3°2 to 3'5 times and the propodus from 3°5 to 4°3 times as long as wide; (iv) the dorsal spinules of the telson are proportionately a little longer and the posterior pair is placed further forwards, the distance between the posterior pair and the apex being, in both sexes, slightly more than half the distance which separates the two pairs. These characters combined with those derived from the pro- portionate length of the rostrum and the carpus of the first leg are sufficient, if constant, to justify the retention of two distinct species. — The specimens I have examined are all small, the largest being only 21 mm. in length. 280) = Records of the Indian Museum. [VoL. XXIV, C 439/t. Port Blair, Andamans. — S. Kemp, Feb., Two. 1915. C 440/1. Andamans. A.R. S. Anderson. Two. 1417. Upolu, Samoa. Purchased. One. All the specimens from the Andamans were found in Mele- agrina and it is from this genus of molluscs that the species has generally been recorded. Borradaile notes, however, that it some- times occurs in Tvidacna. There is no note of the mollusc in which the Samoan specimen was found. ‘The species is probably one of wide distribution in the Indo-Pacific region. Genus Typton Costa. 1917. Typton, Borradaile, Tvans. Linn. Soc. (2) Zool. XVII, p. 304. Borradaile gives a full list of references to this genus and to the two species which belong to it. Typton is readily distin- guished from all other Pontoniinae except Paratypion by the rudi- mentary character of the antennal scale. In T. spongicola, which is found in sponges in the Mediterra- nean and western parts of the English Channel, the rostrum is “spine-like and toothless and there is a pair of very long supra- orbital spines. In T. bouvierz, which is known only from Djibouti in the Red Sea, the rostrum is short, with 2 or 3 teeth on its upper edge, and there are no supra-orbital spines. , In both species the dactylus of the three posterior legs is biunguiculate, but with- out a basal process. Genus Paratypton Balss. 1914. Paratypton, Balss, Zool. Anz. X1.1V, p. 83. 1915. Paratypton, Balss, Denk. math.-naturw. Kl. K, Akad. Wien XCI, p. 27. This genus agrees with Typton and differs from all other Pontoniinae in the rudimentary condition of-the antennal scale. It differs from Typton in a number of characters, of which the most important are (i) the complete absence of the rostrum, (ii) the absence of exopods from the second and third maxillipeds and (iii) the simple dactylus of the last three peraeopods. The distal endite of the maxilla is well developed in Typton, but quite rudi- mentary in Paratypton. The only known species of the genus, P. siebenvocki Balss, is recorded from the Red Sea, the south coast of Arabia and Samoa. It appears probable from its structure that it is para- sitic or symbiotic in its habits, but of this nothing is known. 1922.] S. Kemp: Notes on Crustacea Decapoda. 287 ADDENDUM. Prof. Ch. Gravier has recently sent me for examination a number of Macrura from the Gulf of California collected by M. L. Diguet. Among them I find two species of Pontonia which I identify as Pontonia margarita Smith! and Pontonia pinnae Lock- ington® (mec Ortmann). Of the former there are numerous speci- mens, obtained ‘‘ dans l’huitre perliére’’; of the latter a single pair obtained in Pinna rugosa. Miss Rathbun,’ when describing Pontonia californiensis remarks.—‘ This is the only Pontonia described from the west coast of North America, the P. margarita of Smith being a Con- chodytes.”’ ‘These statements call for correction, for P. margarita is in my opinion correctly placed in the genus Pontonia and P. pinnae was recorded by Lockington in 1879 from the Gulf of California. Schmitt,‘ in his valuable treatise on Californian Deca- poda mentions only P. californiensis and Borradaile,’ who also appears to have overlooked Lockington’s species, follows Miss Rathbun in referring P. margarita to the genus Conchodytes. In P. margarita the dactylus of the last three legs is broader than usual, with the two claws strongly curved; it thus bears a strong resemblance to Conchodyles but lacks the large basal process which is characteristic of that genus. P. margarita and P. pinnae are closely allied forms, but may be distinguished by the following characters :— P. margarita Smith. P. pinnae Lockington. Basal breadth of rostrum about half its | Basal breadth of rostrum about equal length. to its length. Eyes larger, almost reaching antennal | Eyes smaller, not nearly reaching an- spine when extended laterally. tennal spine when extended laterally. I.arge chela with palm scarcely more | Large chela with palm twice as long as than one and a half times as fate as broad. broad. Dactylus of last three legs very broad, | Dactylus of last three legs less broad with accessory claw strongly curved with accessory claw almost straight and directed slightly backwards. and directed obliquely forwards. Spines on dorsum of telson large; pos- | Spines on dorsum of telson small ; pos- terior pair almost equidistant between terior pair much nearer to apex than anterior pair and apex. to anterior pair. Size smaller, 20-28 mm. Size larger, 36-42 mm. Lives in Margaritophora fimbriata. Lives in Pinna rugosa. In the specimens of P. pinnae which I have examined the carapace is much’more strongly arched in lateral view than in P. margarita and in the ovigerous female the rostrum projects down- wards at an angle of 45°. P. californiensis Rathbun, which I have not seen, appears to be easily distinguished from both the above species by the very ! Smith, in Verrill, American Naturalist 111, p. 245 (1869). 2 Lockington, Bull. Essex Inst. X, p. 163 (1879). 3 Rathbun, Harriman Alaska Exped: X, p. 34 (1904). + Schmitt, Univ. Calif. Publ., Zool. XXI111, p. 38 (1921). 5 Borradaile, Trans. Linn. Soc. (2) Zool. XVII, p. 394. 288 Records of the Indian Museum. [Vol XXIV, 1922.] slender rostrum, the shorter carpus of the first leg, the form of the fingers of the second leg (which gape and are devoid of large teeth) and by.the position of the spines on the telson. In all three species the dactylus of the last three legs is biunguiculate. P. californiensis is known only from Santa Cruz I , California ; P. pinnae only from the Gulf of California! and P. margarita from the Gulf of California and the Gulf of Panama. ‘ The specimens | have seen are from Los Angeles Bay, one of the original localities. PNA PINS NINN es ; ° o e oa ae Ph, I ¢ i) ” 7 , ie : - - t i hes | 4 on . oe = = : i nae MO yy J a = } 5 ar , i t \ : bY y : \ ' cet ‘ an ’ i / ‘ , ¢ mn 4 tA, rt L ‘om ey vex al H t= = , * ~ o = ly , ay : , 3 & mo .vou ‘BE , eee ee: L) @ane evi! oo ; " : a _ thie t 7) ees ih ai 4\ outs i : , aes ; : ii \ ; ' Wi see le r ; = \ EXPLANATION OF PLATS III. Fic. 1.—Periclimenes (Periclimenes) impar, sp. nov., from a specimen about Io mm. in length. Fic. 2.—Palaemonella vestigialis, sp. nov., from a specimen about 18 mm, in length. «+ PLATE III. Rec. IND. Mus., VOL. XXIV, 1922. gv ) F ll A 3 ad > ‘ ZZ -_ i) (7 Z Ay "WP ui PY, ff VEZ >. M Woodward & A. Chowdhary del. Fig. 1. Periclimencs impar, sp. nov. Fig. 2. Palacmonclla vestigialis, sp. nov. ’ te WT ae tht Hhenakoaag \ ape i a uw META ATT iy be : ag) (aoe ayer \) keer \) 44 > Abi geral oF aca J , = oS . ie ge jl Ate »e . aly ne pd wa ——— ea eerie | {zunsumekve ye boc fs Ay nea ey t EXPLANATION OF PLATE IV. Fic. 3.—Periclimenes (Periclimenes) latipollex, sp. nov., from a specimen about 16 mm. in length. Fic. 4.—Periclimenes (Periclimenes) lanipes, sp. nov., from a specimen about 13 mm. in length. PLATE IV. Sj Rec. IND. Mus., VOL. XXIV, 1922. TIES G. M. Woodward del. Fig. 3. Periclimenes latipollex, sp. nov. Fig. 4. Periclimenes lanipes, sp. nov. Dies op KUT ASS BO etaranasiKs whee) aretha]: : P ot Ss gest im to riiedy* weini arya eatin) conven oO ant. ; Egil ot net a Mads aM 2 NE ba a és — AY , “ot | ‘ ; y { . 4 i 4 j 4 g a , : é , k a! ae ye - ; 4 a > | - sf a0 ae 7, oy a - Sa ke ewer \ _— ® ‘ — a _ - — f — a Wf EXPLANATION OF PLATE V. Fic. 5.—Periclimenes (Periclimenes) rex, sp. nov., from a specimen about 21 mm. in length. Fic. 6.—Periclimenes (Periclimenes) investigatoris, sp. nov., from a specimen about 15 mm. in length. REc. IND. Mus., VOL. XXIV, 1922. PLATE V. i \\ ( t \ Fig. 5. Periclimenes rex, sp. nov. Fig. 6. Periclimenes investigatoris, sp. nov. aa ae ‘ , SATE By Het i et, r | a. 7 i Mi aT i wa I ees n ba PY be eo rriree ; A Pio. 4 ey Ee ortega 4 i ; he i p 7 Rae eet a i ve + then “ay sari yr y, a > : } ne ie a, RY i? aig y eo 4 Bw dq » 5 beh A 1h hy be Ta = ri | Pe Ik CT 7 ve ; cae a ” ° — pn aa d ¥ aig E } f arf ~t a 1 Ay iS. | x | / : eft ' ‘ - ~ } ; “| a aes ‘ 4 ¥ *, 4 } ri Xs 7 r i , 4 a. ae ere De y bt} a UY, y / ; “ ny n = ‘ ~. » : i “ : 7 He J * ”~ A - oy ms y Ye aa ; Tina me i ’ a al a, 4 q “ ‘ : 5 iw i Vik J ' ‘ \ ‘ i d hy r | . 4 . 1é - ov ae - - a < | + 4 7 ree a he iy wT aA Vu A % oon Aw 4 ty ea we | : ss tay i 7, tha - * 7 G) ad i r se Wadraaivs : L i Prod ~ iS nee a ie o ey Ly fcisassicrel ; aes irrteress = =" he r 1 2 sitgewl oi unten da tus s fF ‘if a a } * Be j i \ ~*~ * Oa gate ‘ll i cs ~ ae, mir et we) of hak pe i] i} oe 2 EXPLANATION OF PLATE VII. Fic. 9.—Periclimenes (Ancylocaris) agag, sp. nov:, from a specimen about 16 mm. in length. Fic. 10.—Perichimenes (Ancylocaris) grandis (Stimpson), from a specimen about 20 mm. in length. a REc. IND. MuS., VOL. XXIV, 1922. PLATE VII. G. M. Woodward & A. Chowdhary del. Fig. 9. Periclimenes agag, sp. nov. Fig. 10. Periclimenes grandis (Stimpson). - 24 ak AY ON rte send or BP ne Fl i. | } ee ee te ee ee nee A= ee ee coe ae es ai nn’) _— — i = ai fa ul —- ; —— a. , _ — re “ i a % by oe )., ag se al ie aia’ Ge ee rh. < nha pee nm) 4 * = 4 + sfute ee OFF, WARTTYT KA nvig ee (erg ha? dornmnt i a ae iy eile ey — — i — — —_ —_ n —_ ; — ~ > — ! 7 \ iy ip i i} } » ii \ 4 i \ ~‘\ ’ i . . 4 a a \ ‘, | 7 be i) 2 , ¢ - i rh ie ’ a, ae i ' » LF : 7 ofl a= t nao e mi ol ban = a a =, “a ae he > > 7 EXPLANATION OF PLATE VIII. Fic. 11.—Periclimenes (Ancylocans) tenutpes Borradaile, trom a specimen about 22 mm. in length. Fic. 12.—Periclimenes (Ancylocaris) digitalis, sp. nov., from a specimen about 22 mm. in length. Rec. IND. Mus., VOL. XXIV, 1922. PLATE VIII. — \ oo Fig. 11. Periclimenes tenuipes Borradaile. A. Chowdhary del. Fig. 12. Periclimencs digitalis, sp. nov. ; , je “sy ni, ii j tae d Mets ‘ , ar lel ap ng ie ped one f TP mao FF en Poa, Oa oo Weg hall ieee ape ao Oy Sn Uw i. 02 7 t " j \ x : vr 1 ae | ; = ‘ / ‘% % F ¥ ¢ ¥ ; bs f : \ ¥ ae man I \ Ds , 7 ae ab ©, Vruimeasshnanviabacabitbvan ¥ =.) on 4 aK, t \ t et” Saliebaty Showers res ; : : ’ "2 *s: y a, Mr Reh ai teed et | W * ce i f b Y ani ’ M , the a Th it i * 7 } q ‘ f 1 ‘f _ . ; A an * pinot sae : - Para Upetas > » y . a i " a # Pa % A, ~ ; a’ am Ps ee ae Sh Oa Jar is, j ty ‘di | f d r 4} ‘4 i r ¢ 1 oe &. ; | 7 = a . i” L- i " i , D i* + : 9 % y oe oe « 7 , .s ¥° a : iN ; bh * ~, ne @ _ = , ae — _ ; * > os. a if - / = eps) ane =, a Fs 4 — , 7 ne a 7" en = are = a > es os a ‘ ¥ : i ae Py (= * nye a goo) | ya es U5 ry Lay EXPLANATION OF PLATE IX. Dasycaris symbiotes, gen. et sp. nov., from a specimen about I3 mm. in length. a REc. IND. MuS., VOL. XXIV, 1922. PLATE IX. biotes, gen. et sp. nov. SCarts SYM Day. A. Chowdhary del. ae ss oh genase . i re woul iy : meine | \ ~ RECORDS of the INDIAN MUSEUM (A JOURNAL OF INDIAN ZOOLOGY) Vol. XXIV, Part III. AUGUST, 1922. PAGE The Fauna of an Island in the Chilka Lake. Heteromera. I<. G. Blatr ae oe 289 Free-living Thysanura. C. Dover .. at aa Se oC a0 299 Dragonflies. F.C. Fraser and C. Dover .. 303 Notes on Fishes in the Indian Museum, 1V. On Fishes belonging to the genus Botta (Cobilidae). Sunder Lal Hora.. 5 313 New records and species of Membracidae from India. W. D. ubeioneee 323 Five new species of the phyychotag eine Coriza. ate C. A. Paiva and Cedric Dover .. oe Ac a0 ae a6 331 On some Indian Derbidae finmagters): F.Muir .. ae Zi a 335 New Indian Homoptera. F. Muir .. Me a ae ‘a AG 343 Materials for a generic Revision of the Freshwater Gastropod Molluscs of the Indian Empire. No.5. The Indian Planorbidae. N. Annandale ne 357 On a new Alycaeus from the Khasi Hills; H.H. Godwin-A usten es 365 A List of the Dragonflies recorded from the Indian Emerg With, special. reference to the collection of the Indian Museum. 9. the Subacily/, * Gomphinae. JF. I’. Laidlaw, With an Appendix. ne on a “ 367 pAXS 4 r QA, 3) ; J T AV wes ( 19 Vs Calcutta: Nhe onal ws” on oe PUBLISHED BY THE DIRECTOR, ZOOLOGICAL “SURVE OF INDIA, PRINTED AT THE BAPTIST MISSION PRESS. 1922. Pr ice Iwo K HUpees. wenicthions at at ~ i te deh asset bon Be aie bh Vek aasd es nn win “No ‘uae tonerat ‘esata wit fo, eutebestt viomay « Ri | lr A tliat aati oath 8 ert peri peter eG ie a ee naan henabe 34, REM Men lt: gst aaron: eae | - Nath even! yn Mahal plete oreapiteternal ridden ing min, Oty clivaisee meth tt tee gee hicentted abt wt maumaeettoe minee > | pleesns rene GIF tem Dehet% ik saat a . ) : Di hei q f . *) | bs aut” .°% atiniee Pare ih i Yet at | ’ 6 } 4 Sy SS XXX AMD Lo SSS ZS \ i Two types of scale in Botia. Scale from dorsal surface of Botia almorhae: X65. a. b. Scale from dorsal surface of Botia hymenophysa: X65. exception of Botta hymenophysa known from Burma, Siam and the Indo-Australian Archipelago, all the species from the Indian Empire possess eight barbels. On the other hand all the known species from China and Japan are characterized by six barbels only. In the intermediate regions, Burma, Siam and the Indo- Australian Archipelago, representatives of both the groups are met with. Jordan and Fowler! have regarded the two groups as dis- tinct genera and have adopted the name Hymenophysa McClelland,” ! Jordan and Fowler, Pvoc. U.S. Nat. Mus. XXVI, p. 772 (1903). 2 McClelland, Asiatic Researches p. 443 (1838). 1922. ] S. L. Hora: Fishes of the genus Botia. 315 for the species possessing six barbels. I have, however, retained the name Botta for both the groups mainly for two reasons, firstly, because in several Cyprinoid genera species are grouped irrespective of the number of barbels and secondly because Giinther’s two species, Botia pratfi and B. suferciliaris, possess “‘a pair of soft rounded buttons’’ on the chin; these may or may not be consi- dered as barbels and appear to afford a link between the two primary groups. I have examined the scales in the various species represented in our collection and find that those of B. hymenophysa differ greatly in structure from those of the remaining species. In B. hymenophysa they are almost circular with a big .central nucleus and a number of radii to all parts of the periphery, whereas in the other species the scale is ellipsoidal with an excen- tric nucleus and with a large number of long radii to the apex and a few short ones to the base. Both Giinther 'and Day * considered that the anterior division of the bladder in the genus Botia is partially enclosed in bony capsule, whilst the posterior division floats free in the abdominal cavity. This is true in all the species that I have examined with the exception of B. almorhae in which the anterior chamber is completely enclosed in bone and the posterior, though lying free in the ab- dominal cavity, is greatly reduced. In other species also the posterior chamber is somewhat reduced. Botta nebulosa, Blyth,’® is known from a single specimen from Darjiling, which is now preserved in the collection of the Zoological Survey of India Onexamination I am unable to refer it to the genus Botia. I believe that it belongs to Nemachilus and in ali probability is the male of N. botius. My reasons are as follows :— (i) I have not been able to find any trace of the suborbital spine in the unique specimen. Day* thought that the suborbital spine was damaged, but the groove that is present is not suffici- ently deep to justify the view that it ever contained a spine. The groove is of the nature of a shallow slit partly covered superiorly by a fold of skin. I° have already remarked in a previous paper that such grooves and folds of skin form the secondary sexual characters of the males of certain species of Nemachilus. (ii) The caudal fin of the specimen is now damaged, but Day, who examined it-in a better condition, remarks ‘‘ caudal. slightly rounded.’’ Some years ago Dr. B. L. Chaudiiuri had this speci- men figured and the manuscript drawing is now with me. It shows the caudal fin as slightly emarginate with both the lobes rounded. In the genus Botta the caudal fin is forked and the lobes sharply pointed. =a --- -— . 1 Giinther, Brit. Mus. Cat. Fish. V11, p. 366 (1868). 2 Day, Fourn. As. Soc. Bengal X1.1, part U1, p. 176 (1872). 3 Blyth, Fourn. As. Soc. Bengal XXIX, p. 165 (1860). $ Day, Proc. Zool. Soc. London, p. 549 (1869). 5 Hora, Rec. /nd. Mus. (in press). 316 Records of the Indian Museum. [VoL. XXIV, (iii) The air-pladder consists of two lateral chambers en- closed in a bony capsule. This type of bladder is characteristic of the genus Nemachilus and is not to be met with in any species of Botva. (iv) There are six barbels, four rostral and two maxillary, but the rostral barbels are not united at their base as is the case in Bota. : (v) The shape of the mouth, the structure of the lips, jaws and of the scales is quite different from any species of the genus Botva that I have examined. The following is an artificial key to all the known species of the genus Botza!:— Group I. BaARBELS six (Hymenophysa). {. Eye in middle of head [Commencement of dorsal equidistant from tip of snout and base of caudal) ... a .. B. multifascrata. II. Eye not in middle of head. A, Kye nearer end of operculum than that of snout or almost wholly in posterior half of head. 1. Length of head equals depth of body [Suborbital spine extending to below posterior margin of eye; a broad black bar at base of caudal]... B. modesta.* 2. Length of head greater than depth of body. a. Suborbital spine extending beyond eye in both directions Sc sea ihe ... B. superciliaris. 6. Suborbital spine extending to below middle of eye . B. hymenophysa.* B. Eye nearer end of snout than that of operculum. 1. Suborbital spine not extending to below hind margin of eye... 3 Ee ... B. curta.* 2. Suborbital spine distinctly extending to below hind margin of eye. . a. Interorbital space twice as wide as orbit ; ‘* ground- colour yellowish, the body ornamented with five black bands”’ ... ape a ... BB. variegata. b, Interorbital space three to four times as wide as orbit; ‘ground-colour brownish olive, without distinct marixings on the body ”’ ee .. B. pratti. Group II. BARBELS EIGHT (Botta s.s.). I. Eye in middle of head cs poe ... B. helodes. II. Eye not in middle of head. A. Length of snout considerably more than that of re- maining part of head. 1. Body marked with two broad bands... ... &. macracanthus.* 2. Body marked “ with irregular and partly confluent brown cross bands, which enclose larger and smaller round whitish spots ”’ adh ea ... SB. rostrata. B, \.ength of snout either equal to or less than that of remaining part of head. 1. Eye almost in posterior half of head. | a. Head and body marked with a number of narrow oblique vertical bands __... bs ... B. styata.* b. Head and body marked with a few broad vertical , bands or reticulations. 1 The species marked with an asterisk are present in the collection of the Zoological Survey of India. 1922. ] S. L. Hora: Fishes of the genus Botia. 357 i. Anterior origin of dorsal almost equidistant from ue of snout and base of caudal. o. Eye small, its diameter contained 4 to 4°5 times in length of snout B. birdi.* B. Eye moderately large, its diameter contained 3 times in length of snout B. dario.* ii. Anterior origin of dorsal not equidistant from tip of snout and base of caudal .. B. histvtonica.™ 2. Eye not situated wholly in posterior half of head. a. Head and body marked with reticulation, Air- bladder much reduced, anterior chamber wholly enclosed in bone B. almorhae.* b. Head and body marked with vertical bands. Air-bladder almost normal, anterior chamber partially enclosed in bone. i. Caudal marked with 2-3 bands, body marked with loops dorsally and with short vertical bands laterally ; B. lohachata.* ii. Caudal marked with two black spots, body mz irked with 6. 7 oblique vertical bands =f .. B. geto.™ Botia multifasciata Regan. 1905. Botia multifasciata, Regan, Rev. Suisse Zool. XIII, p. 380, pl. v, fig. 3. Hatitat :—China. Botia modesta Bleeker. 1864. Botia modesta, Bleeker, Nederl. Tydsch. Diek, p.\ 1868. Sotia modesta, Giinther, Brit. Mus. Cat. Fish. Vit, Pp. 368. 1870. Botia modesta, Bleeker, Versi. Meded. Ak. Un eteea: IV, p. 254 (figured). 1876. Botia modesta, Sauvage, Bull. Soc. Philom. XIII, p. go. 1876. Botia rubripinnis, Sauvage, Bull. Soc. Philom. XII, Pp. 99. 1881. Botia modesta, Sauvage, Nouv. Arch. Mus. Paris. (2) IV, p. 192. Habitat.—Siam. I have examined specimens from Lopburi sent to me by Dr. Malcolm Smith. Botia superciliaris Giinther. 1892. Botia superciliaris, Giinther, in Pratts’ ‘Snows of Tibet”, p. 250, pl. iv, fig. B. I have placed this species in the section comprising forms having six barbels. It possesses, however, according to Giinther, ‘a pair of soft rounded buttons’’ which are probably remnants of the additional pair. Habitat.—Kia-tiang-fu (foot of Amieshan), Province Sze Chuen, China. Botia hymenophysa (Bleeker). 1852. Cobitis hymenophysa, Bleeker, Nat. Tijdschr. Ned. Indié II, p. 602. 1858. i Aaa MacClellandi, Bleeker, Nat. 7ijdschr. Ned. Indié XVI, p. 358 1860. Aymenoplryse Was Clelland:, Bleeker, Jchth. Arch. Ind. Prody. . Cyprini, p. 63. 1860. genie a Res Blyth, Mg As. Soc. Bengal XXIX, p. 166. 318 Records of the Indian Museum. [VoOL. XXIV, 1863. Botia hymenoplysa, Bleeker, Atl. Jchth. III, p. 6, pl. cii, fig. 3 1868. Botia hymenophysa, Giinther, Brit. Mus, Cat. Fish. VII, p. 568. 1869. Botia berdmovei, Day, Proc. Zool. Soc. London, p. 549. 1872. Botta hymenophysa, Day, Fourn. As. Soc. Bengal XLT, part II, 178. 1878. Botia berdmorei, Day, Fish. India Xl, p. 607, pl. cliv, fig. 3. 1889. Botia berdmorei, Day, Faun. Brit. Ind. Fish. 1, p. 217." 1889. Botia berdmoret, Vinciguerra, Ann. Mus. Nat. Genova XXI1X, p. 345: ‘ 1903. Botia hymenophysa, Volz, Zool. Fahrb. Syst. XIX, p. 406. 1906. Botia hymenophysa, Popta, Notes Leyden Mus. XXVII, p. 207. 1916, Botta hymenophysa, Weber and Beaufort, Fish. [ndo-Austrai. Archipelago, I\I, p. 24, fig. 6. 1921. Botia beydmoret, Hora, Rec. Ind. Mus XXII, p. 195. This species is distributed over a very wide area. It occurs in the Indo-Australian Archipelago, Siam and Burma. Its range extends as far as the Manipur Valley (Assam), whence the waters flow into the Irrawaddi system. There has heen considerable confusion as to the occurrence of this species in Burmese waters. Day in 1872 (op. cit.) recorded it from ‘‘the northern portions of British and also Upper Burma,”’ but in his later works he referred fishes with the same Burmese names, ‘‘ Nga-tha-lay-doh,” and ‘‘ Shoay-Zagay” to Botia berd- moret which he considered to be *‘ closely allied to B. hymenophysa, Bleeker,’’ but differing “in its dorsal fin, and also in its colours, etc.” In his ‘‘Monograph of Indian Cyprinidae’”’ he gave the habitat of B. berdmoret as ‘‘ Darjiling and Bengal generally.’’ This is incorrect and it appears to me from the description of the species that the specimens referred to are not Bota at all. In the Manipur examples (0. cit., p: 195) I found great variation in the number of oblique bands and also in the general colouration of the body. On the character of the colouration, therefore, I am unable to recognise B. berdmorei as distinct from B. hymenophysa. Inmy conclusions I am supported by Vinciguerra (op. cit.). In the Siamese examples that I have examined, sent me from Lopburi by Dr. Malcolm Smith, the position of the anus is some- what different. It is situated half-way between the base of the anal fin and the posterior origin of the ventral fin. In another example the anus is much nearer to the base of the anal fin than to that of the ventral fin. There are, however, so many points of agreement between the Siamese and the Burmese forms that I do not think myself justified in separating them. Botia curta (Schlegel). 1850. Coens curta, Schlegel, Faun. Fapon. Pisces, Be 223, pl. ciil, 8-4. 1868. Botia curta, Ginther, Brit. Mus. Cat. Fish. VII, p. 368. 1903. Hymenophysa curta, Jordan and Fowler, Proc. U.S. Nat. Mus. VI, p. 772. Habitat.—Japan. I have examined a specimen from Yodo river, sent to the Indian Museum by the Otsu Lake Laboratory. 1922.] S. L. Hora: Fishes of the genus Botia. 319 Botia variegata Giinther. 1889. Aotia variegata, Giinther, Ann. Mag. Nat. Hist. (6) IV, p. 228. 1892. Botia vartegata, Giinther, in Pratt’s ‘Snows of Tibet,” p. 249. Habitat.—Ichang (China). Botia pratti Giinther. 1892. PRET. Ginther, in Pratt’s “Snows of Tibet,” p. 250, pl. iv, fig. A Habitat.—Kia-tiang-fu (foot of Omie-shan), province of Sze Chuan, China. Botia helodes Sauvage. 1876. Botia helodes, Sauvage, Bull. Soc. Philom. XIII, p. 99. 1881. Botiahelodes, Sauvage, Nouv. Archiv. Mus. Paris (2) 1V, p. 192. Habitat —Siam. Botia rostrata Gtinther. 1868. Botia rostrata, Giinther, Brit. Mus. Cat. Fish. VII, Pp: 367 (head figured). 1872. Botia rostrata, Day, Fourn. As. Soc. Bengal XI\.I, ii p. 178, Habitat.—Bengal and Assam. Botia striata Rao. 1920. Botia striata, Rao, Ann. Mag. Nat. Hist. (9) VI. p. 60, pl. ii, figs. 4. 4a, 46. Habitat.—River Thunga in Mysore State, South India. The range of the species extends as far as the Satara District in the Eombay Presidency, whence a single specimen, now in our collec- tion, was obtained by Dr. S. P. Agharkar. Botia birdi Chaudhuri. 1878. Botia geto, Day (nec Buchanan), Fish. Jndia II, p. 606, pl. cliv, 1889. Bune pan Day (mec Buchanan), Faun. Brit. Ind. Fish. p. 217, 1909. Bolts bird, Chaudhuri, Rec. Ind. Mus. III, p. 339. This species exhibits considerable variation in colour with the age of the fish. The dark bands on the body often break up to form an irregular reticulation on the dorsal surface and the sides. Recently a large series of specimens has been obtained from the Kashmir Valley. All forms of colour pattern from regular bands to reticulation are present in this series. The females contain a large number of minute eggs; in a ripe female the depth of body is considerably greater than the length of the head and the ventral profile is greatly arched. Habitat.—Sind in the Kashmir Valley and the Punjab. 320 1822. 1868. 1872. 1878. 1889. Records of the Indian Museum. [VoL. XXIV, Botia dario (Ham. Buch.). Cobitis dario, Hamilton Buchanan, Fish. Ganges, pp. 354, 394 pl. xxix, fig. 95. Botia dario, Ginther (in part), Brit. Mus. Cat. Fish. VII, p. 366. Botia dario, Day, Fourn. As. Soc. Bengal, X11, part II, p. 177. Botia dario, Day, Fish. India UL, p. 606, pl. cliv, fig, 1 Botia dario, Day, Faun Brit. Ind, Fish. I, p. 216 Habitat.—Upper Bengal and Assan. Hamilton Buchanan found this species in all the districts of Northern Bengal and Bihar that he visited. We have a number of specimens from Cachar. 1860. 186g. 1872. 1878. 1880. 1889. 1921. Botia histrionica Blyth. Botia histrionica, Blyth, Fourn. As. Soc. Bengal XXI1X, p. 160. Botia histrionica, Day, Proc. Zool. Soc. London, p. 550. Botia histrionica, Day, Fourn. As. Soc. Bengal XI.1, part If, p. 170. Botia histrionica, Day, Fish. India U1, p. 607, pl. cliv, fig. 4. Botia histrionica, Day, Faun. Brit. Ind. Fish. 1, p. 218. Botia histrionitca, Vinciguerra, Ann. Mus. Civ. Nat. Genova, p. 340. Botia histrionica, Hora, Rec. Ind. Mus. X X11, p. 195. Habitat.—The species was originally described from Pegu, but since then it has been recorded from several other places in Burma such as Bhamo and Mandalay and from the Manipur Valley in Assam. 1852. 1860. 1863. 1868. 1903. 1905. 1916. Botia macracanthus (Bleeker.) Cobitis macracanthus, Bleeker, Nat. Tijdschr. Ned. Indie Ul, D. 603. Py menenigee naar ANS Bleeker, /chth. Arch. Ind. Prodr. II, Cyprini, p. 62. Botia macracanthus, Bleeker, Atl. /chth. III, p. 5, pl. cil, fig. 2 Botia macracanthus, Giinther, Brit. Mas. Cat. Fish. VII, p. 368. Botia macracanthus, Volz, Zool. Fahrb., Syst. X1X, p. 405. Botia macracanthus, fowler, Proc. Nat. Sct. Philadelphia. (2) LVIT, p. 474. Botia macracanthus, Weber and Beaufort, Fish. /2do-Austral. Archipelago III, p. 23, fig. 7. I have examined a specimen of this species from Sumatra kindly sent me by Prof. Max Weber. Habitat—Sumatra and Borneo. 1831. 1838. 1868. 1872. 1878. 1889. Botia almorhae Gray. Botia almorhae, Gray, Zool. Misc. p. 8. Botia grandis, Gray, fll. Ind. Zool., pl. xciv, fig. 3. Botia almorhae, Gunther, Brit. Mus. Cat. Fish. VII, p. 367. Botia almorhae, Day, Fourn. As. Soc. Bengal XI.1, part. If, p. 178. Botia almorhae, Day, Fish. India II, p. 607, pl. cliv, fig. 5. Botia almorhae, Day, Faun. Brit. Ind. Fish. |, p. 217. This species is known from Almora (United Provinces). McClelland ' recorded a fish under the name of Botia (Schistura) ! McClelland, Calcutta Sibi nl Nat. Hist. Il, p. 586 (1842). 1922. ] S. L. Hora: Fishes of the genus Botia. 321 grandis from the Khasi Hills and later on Vinciguerra' found Botia almorhae in ‘‘Meetan” and ‘‘ Meekalan”’ (Burma). I think the later records require confirmation. Botia lohachata Chaudhuri. 1912. Botia lohachata, Chaudhuri, Rec. /nd. Mus. VII, p. 441, pl. xl, figs. 2, 2a, 2b. Habitat.—Gandak River in Saran, Bihar. Botia geto (Ham. Buch.). 1822. Cobitis geto, Hamilton Buchanan, Fish. Ganges, pp. 355; 394, pl- xi, figs, 96. This species is Buchanan’s* Gengto of Goalpara. I collected some specimens at Gorakhpur which correspond in every respect with the figure published by its author. Giinther*® considered it to be the young of Botia dario and Day* in his earlier works was of the same view. The specimens from Gorakhpur are not in good condition for detailed morphological investigation and I am therefore unable to confirm Giinther’s statement. The colouration is, however very distinct and seems to be characteristic of the species. ! Vinciguerra, Ann. Mus. Nat. Genova XXIX, p. 344 (1889). 2 Hunter's Statistical Account of Bengal XX, p. 41 (1877). 8 Ginther, Brit. Mus. Cat. Fish. VII, p. 366 (1868). # Day, Fourn. As. Soc. Bengal XLI, part II, p. 177 (1872). Le ee eee eOeoOEeeeeeeasS ee ia wr , he abies ebuow > "5 * af " > ] teseteys i Hain Bit ‘te pitas 3 . U ns 2a % (ene; | 5 ANTI Y ¥ rta Ls? ~tou ae £ rs , J — : i “3 i 7 * % o \ m f rare i rea vw i tte bad ba hy tt PF ; fay ¢'me, sé , . s facerd F Rah So \ 7 -” A ee o> # f (ny 4 a i 7 J m t < € X - an q 4 r if 7 f y = u 4 NEW RECORDS AND SPECIES OF MEMBRACIDAE FROM INDIA. By W. D. FuUNKHOUSER, University of Kentucky. (Plate X.) Through the courtesy of Professor C. F. Baker of Los Banos, P.I., I have had the privilege of examining a most interesting series of Membracidae belonging to the Zoological Survey of India. This collection contains five new species and furnishes a num- ber of very valuable records. The report on these insects follows :— Tricentrus pronus Distant. One female from Tura, Garo. Hills, Assam, 1400 ft., October 1917 (Mrs, Kemp). ~ Tricentrus projectus Distant. One female from Calcutta, Tollyganj, Nov. 11, 1916 (F. H. Gravely). Tricentrus resectus Distant. A pair, each specimen labelled ‘‘ Hills near Taiping, Perak, Dec. 26-30, 1915.’’ The male is very slightly smaller than the female. Tricentrus brevis Funkhouser. One male from Barkuda I., Chilka Lake, Ganjam Dist., Madras Pres., Aug. 3-19, 1919 (F. H. Gravely). Tricentrus albomaculatus Distant. ‘One male from Tura, Garo Hills, Assam, 1200-1500 ft., July 1917 (S. Kemp). Tricentrus allabens Distant. Two specimens, both females, one from Darjiling, 7000 ft., E. Himalayas, June 4, 1917 (E. Brunetti); the other from hills near Taiping, Perak, Dec. 26-30, 1915 (N. Annandale). Acanthucus minutispinus, sp. nov. (Pix; figi'n): Black with golden pubescence ; tegmina smoky ; legs ferrugin- ous; suprahumeral horns long, sharp, triquerate; median spine 324 Records oj the Indian Museum. [VoL. XXIV, very small, triangular, arising on dorsal line just back of suprahu- merals ; posterior process nearly straight, slightiy upturned at tip, reaching just beyond internal angles of tegmina. Technical description :— Head about as long as wide, black, densely pubescent with long golden hairs, roughly sculptured; base arched and sinuate; eyes large, prominent, gray mottled with brown; ocelli small, amber-coloured, not conspicuous, about equidistant from each other and from the eyes and situated above an imaginary line drawn through centres of eyes; clypeus longer than broad, black, densely pilose, extending for about half its length below in- ferior margins of genae; margins of genae nearly straight, slightly turned outward at edges. Pronotum black, finely punctate densely pubescent; metopi- dium vertical, as broad as high; humeral angles prominent, trian- gular, extending outward farther than the eyes; median carina strongly percurrent ; suprahumeral horns as long as the distance. between their bases, flattened dorso-ventrally, extending outward and upward and curving slightly backward, undersurface bearing central carina, tips sharp ; central spine very small, triangular, entirely black, pubescent, situated on median dorsal line just behind bases of suprahumerals ; scutellum only slightly exposed ; posterior process slender, thicker through the middle than at the base, tricarinate, tip sharp and slightly upraised, extending just beyond internal angles of tegmina. Tegmina long, narrow, smoky-hyaline, tinged with ferruginous ; base narrowly opaque and punctate; veins prominent, marked with brown; basal costal margin pilose; five apical and two dis- ‘coidal areas. Hind wings with four apical areas. Undersurface of body black and strongly pubescent. Legs uniformly ferruginous. Length from front of head to tips of tegmina 7°5 mm.; width between extremities of suprahumeral horns 3°5 mm. Type.-—-Female. In collection of Zoological Survey of India. Locality.—Sureil, Darjiling Dist., E. Himalayas, Oct. 11-31, 1917 (N. Annandale and F. H. Gravely). Gargara pulchripennis Stal. One female from Mujang, Sarawak, July 12, 1910 (C. Beebe). Gargara nigrofasciata Stal. One female from Talewadi, near Castle Rock, N. Kanara Dist., Bombay Pres., Oct. 3-10, 1916 (S. Kemp). Gargara nitidipennis Funkhouser. One male from Mujang, Sarawak, July 12, rg1o (C. Beebe). 1922. ] W. D. FUNKHOUSER: Indian Membracidae. 325 Gargara majuscula Distant. One female from Pashok, alt. 3500 ft., Darjiling Dist., FE. Himalayas, June, rg16 (L. C. Hartless). Gargara tumida Melichar. One female from Pashok, alt. 2500 ft., Darjiling Dist., E. Himalayas, May 26, 1914 (F. H. Gravely). Centrotypus asmodeus Distant. One female from Kapit, Sarawak, Aug. 9, rgto (C. Beebe). Centrotypus parvus, sp. nov. (Pl. X, fig. 2). Small, slender, black, pubescent; suprahumerals slender- sharp, projecting upward and outward, as long as the distance between their bases; scutellum entirely concealed; posterior pro- cess long, slender, decurved, extending beyond internal angles of tegmina and just about reaching end of abdomen; tegmina ferruginous-hyaline ; undersurface of body black; legs uniformly ferruginous. Technical description :— Head subquadrate, wider than long, black, impunctate, finely pubescent with short silvery hairs; base arcuate and sinuate; eyes large, prominent, dark brown; ocelli small, conspicuous, white, shining, equidistant from each other and from the eyes and situat- ed about on a line drawn through centres of eyes; inferior mar- gins of genae feebly sinuate; clypeus longer than wide, black, pilose, projecting for more than half its length below margins of genae, tip pointed. : Pronotum black, finely punctate, closely pubescent with short silvery hairs; gibbous above head; median carina percurrent ; dorsum nearly straight ; humeral angles small, triangular, sharp, inconspicuous; suprahumeral horns slender, sharp, extending outward and upward with tips bent slightly backward, as long as the distance between their bases; metopidium convex, broader than high, nearly vertical above the head, a smooth semicircular depression over each eye; scutellum entirely concealed; posterior process long , slender, tricarinate, nearly straight, tip slightly depressed, extending beyond internal angles of tegmina and just about reaching tip of abdomen. Tegmina long, narrow, smoky-hyaline, tinged with ferrugin- ous; base black, punctate, coriaceous and opaque; veins pro- minent, costal veins black, others brown; five apical and two dis- coidal cells. Undersurface of body black and densely pubescent. Legs entirely and uniformly ferruginous. 326 Records of the Indian Museum. [VoL. XXIV,. Length from front of head to tips of tegmina 5 mm.; width. between tips of suprahumeral horns 2°3 mm. Type.—Male. ‘Ihe type-specimen bears Professor Baker’s. duplicate number 16912. Locality.— Hills near Taiping, Perak, Dec. 26-30, 1915 (N. Annandale). Leptocentrus decipiens Kirby. One female from Calcutta, the label bearing the data ‘‘ Tolly- ganj, April 9, 1917 (Ff. H. Gravely).’’ Leptocentrus leucaspis Walker. One male from hills near Taiping, Perak, Dec. 26-30, 1915 (N. Annandale), and one female from Rawalpindi, Punjab, June— July, 1917 (R. Hodgart). Leptocentrus mephistopheles Buckton. One female from Garo Hills above ‘ura, Assam, alt, 3500-3900. ft., Sept., 1917 (Mrs. Kemp). I am very suspicious that Buckton’s species is merely a colour variety of L. leucaspis Walker. Leptocentrus longispinus Distant. One female from Mormugao, Portuguese India, Sept., 1916: (S. Kemp). k Leptocentrus obortus Distant. Four specimens, a male from Phagu, alt. gooo ft., Simla Hills, May 18-21, 1916 (N. Annandale and S. Kemp); two females from Tura, Garo Hills, Assam, Oct. 1917 (Mrs. Kemp); and a female from Barkuda I., Chilka Lake, Ganjam Dist., Madras Pres., Sept.. 20, 1919 (E. Brunetti). Ebhul maculipennis, sp. nov. (EM, DaGa sHiets 2s Near E. carinaius Funkh., but larger, with differently shaped metopidium and differently marked tegmina. Large, dark brown, not punctured, sparingly pubescent ; base of metopidium flaring forwards over the head; pronotum high and subarcuate ; posterior process long, sinuate, sharp, reaching just to internal angles of tegmina; tegmina opaque, richly marked with yellow and dark brown; trochanters, femora and bases of tibiae dark brown, rest of legs yellow. A beautiful, distinct and well-marked species. Technical description :— Head subtriangular, longer than wide, rough'y sculptured,. dark brown, not punctate, densely pubescent with short silvery 1922. ] W. D. FuNKHOUSER: Indian Membractdae. 327 hairs; base arcuate and nodose, partly hidden under overhang- ing margin of pronotum; eyes small, gray mottled with brown ; ecelli very small, opalescent, inconspicuous, twice as far from each other as from the eyes and situated well above a line drawn through centres of eyes; inferior margins of genae sinuate ; clypeus twice as long as wide, brown, densely pubescent, extending for two-thirds its. length below margins of genae, tip broadly rounded. Pronotum dark brown, roughly sculptured, not punctate, sparingly pubescent with short silvery hairs, elevated and gibbous over humeral angles, no indications of lateral carinae; metopidium wider than high, roughly sculptured, depressed at base, lower an- terior margin projecting forward over the head, upper margin keel- shaped; median carina strongly percurrent ; humeral angles large, triangular, blunt, projecting outward beyond the eyes as far as twice the width of the eyes ; scutellum well exposed, longer than wide, apex bifurcate; posterior process long, slender, sinuate, dark brown at both ends and yellow inthe middle, tip sharp and reach- ing just to internal angles of tegmina. Tegmina short, broad, opaque, basal half bright yellow, apical half dark brown, the two colours meeting along an irregular dia-. gonal line extending from the scutellum backwards and downwards; five apical and two discoidal cells ; veins not prominent. Undersurface of body dark brown, densely pubescent; tro- chanters, femora and bases of tibiae dark brown, apical two-thirds of tibiae and all of tarsi and claws bright yellow. i Length from front of head to tips of tegmina 6 mm.: width between extremities of humeral angles 2°7 mm. Type.—Female. In collection of Zoological Survey of India. Locality.—Pashok, alt. 2000 ft., Darjiling Dist., E. Himalayas, May 16-June 14, 1916 (F. H. Gravely). Otinotus oneratus Walker. One male from Coorg, S. India (F. Hannyngton). Antialcidas attenuatus, sp. nov. (Pl. X, fig. 4). Small, slender, brown, shining, punctate, pubescent ; supra- humerals large, triangular, sharp; posterior process extended up- ward in a plate before apex; tegmina shining smoky-hyaline marked with brown; undersurface of body dark brown; legs fer- ruginous. Technical description :— Head subquadrate, longer than wide, dark brown, densely pubescent with golden hairs, a white sericeous streak of longer white hairs down median line and another at right angles to it across genae; base arcuate and sinuate; eyes large, prominent, brown ; ocelli large, prominent, glassy, elevated, twice as far from each other as from the eyes and situated well above an imaginary 328 Records of the Indian Museum. [VoL. XXIV, line drawn through centres of eyes; inferior margins of genae rounded ; clypeus longer than wide, extending for two-thirds its length below inferior margins of genae, a broad white sericeous streak down median line tip rounded and pilose. Pronotum bright golden brown, finely punctate, densely pubes- cent with short golden hairs, a ‘white sericeous streak extending upwards from the head on each side the median line between the horns as far as the posterior process, another fainter streak on each side passing under the horns ; metopidium broader than high, almost vertical above the head, convex, a smooth depressed spot over each eye; median carina percurrent ; humeral angles large, triangular, sharp, extending outward as far beyond the eyes as the width of the eye ; suprahumeral horns large, heavy, triquerate, sharp, extend- ing outward and upward, about as long as their width at base but not as long as the distance between their bases, upper surface flat and white sericeous ; posterior process elevated at base in subtrian- gular crest extending upward about as high as suprahumeral horns, tip suddenly short, sharp, upturned, just reaching internal angles of tegmina. Tegmina smoky-hyaline, shining, mreed with brown at tips and just before internal angles ; base broadly opaque, coriaceous, punctate and pubescent ; five apical and two discoidal cells ; in- terior apical veins strongly bent upwards. Undersurface of body very dark brown, aims black, pubes- cent with silvery hairs ; legs uniformly ferruginous. Length from front of head to tips of tegmina 4°6 mm.; width between extremities of suprahumeral horns 2°2 mm. Type.--Female. In collection of Zoological Survey of India. Locality.—Sureil, 5000 ft., Darjiling Dist., E. Himalayas, Oct. 11-31, 1917 (N. Annandale and F. H. Gravely). Dograna suffulta Distant. One female from Castle Rock, N. Kanara Dist., Bombay Pres., Oct. 11-26, 1916 (S. Kemp). Emphusis malleus Walker. One male from Castle Rock, N. Kanara Dist., Bombay Pres., Oct. 11-26, 1916 (S. Kemp). Machaerotypus brunneus, sp. nov. (Pl. X, fig. 5). Large, entirely brown, coarsely punctate, densely pubescent ; pronotum much swollen and elevated above the head; scutellum entirely exposed ; humeral angles prominent ; no suprahumerals ; posterior process short, sinuate, sharp, arising high above the scutellum ; tegmina smoky- hyaline ; legs and undersurface of body ‘brown. 1922. ] W. D. FunKHOUSER: Indian Membracidac. 329 Technical description :— Head twice as wide as high, reddish-brown, nearly flat, coarse- ly punctate with black punctures, sparingly pubescent with long silvery hairs; base gradually arcuate; eyes large, prominent, brown; ocelli large, conspicuous, opaque white, equidistant from each other and from the eyes and situated about on a line drawn through centres of eyes; inferior margins of genae rounded; cly- peus longer than wide, depressed, projecting for more than half its length below inferior margins of genae, very densely pilose, tip rounded. Pronotum brown, coarsely punctate, densely pubescent, high- ly elevated above head, swollen subglobose; median carina nearly obsolete ; metopidium higher than wide, a smooth irregular depres- sion above each eye, convex in front, tectiform above as seen from the front ; humeral angles large, prominent triangular, extending outward farther than the eyes; no suprahumerals ; scutellum en- tirely exposed, wider than long, coarsely punctate, densely pubes- cent, apical margin weakly notched; posterior process short, sinuate, sharp, arising from highest point of pronotum well above the scutellum, not reaching internal angles of tegmina. Tegmina long, narrow, ferruginous-hyaline, sparingly pilose both on veins and between veins; apex clouded; base narrowly opaque, coriaceous, dark brown, punctate; veins heavy, promi- nent, brown ; five apical and two discoidal cells. Undersurface of body dark brown and pubescent ; legs light brown, the trochanters and femora marked with dark brown and ferruginous, the tibiae lighter brown and the tarsi and claws darker. Leugth from front of head to tips of tegmina Io mm. ; width between extremities of humeral angles 4 mm.; height of prono- tum above head 4 mm. Type.—Female. In collection of Zoological Survey of India. Locality.—Mungpoo, alt. ca. 3000 ft., Darjiling Dist., E. Hima- lavas, Oct. 11-31, 1917 (N. Annandale and F. H. Gravely). ee ee er er wee ee ESSE 1s a) ay orn | : a a ur hay ak BS vera" ie A sa ae x) or! 7 EN S vont Se, Phy pin stare Sed wei} ict pavereens bab ha’ +b, sah * eerie hehiyhoral hse spiltirn | ve , Soh nile 4 Sa § > J ~ ths _ nary aatberks shane 3) egrrws by . t EXPLANATION OF PLATE X. Fic. 1.—Lateral outline Acanthucus minutispinus, sp. nov. 2.—Lateral outline Centrotypus parvus, sp. nov. 3.—Lateral outline Ebhul maculipennis, sp. nov. 4.—IL,ateral outline Antialcidas attenuatus, sp. nov. 5.—Lateral outline Machaerotypus brunneus, sp. nov. ¢ “tap f : ve , : =. - - \ _ Sd - O° tt a. cmt \ ‘ 2 4 ‘ Pi 2 ee ad 7 a " i ah, rhvans I Tw) Aides iy “o a te ~ r~ i ~ . “ ) 7 ow! ~ : al” A ae ‘ * a ; f™ % iy ' . ; ‘ . u } { : 4 Li j —_* tS aes Maer - x, My - ) « ie Natiss Pomayed, SD aes eee ayy eh & ‘rah 9 Se * 7, y i 4 AS thst FIVE NEW SPECIES OF THE RHYNCHOTAN GENUS CORIXA. By the late C. A. Patva and CEpRic DOVER. When Mr. W. L. Distant’s third volume on the Rhynchota in the “ Fauna of British India” series was published in 1906, the widely distributed Corixa hieroglyphica was the only species of the genus then known from India. In 1910 Distant! described seven more species and Paiva’ recently added another two to the fauna of British India, thus making a total of ten species in all. ‘The present note adds five more species to the list, and many more species will no doubt eventually be described. This paper has been compiled from notes left by the late Mr. Paiva.’ My own share in its production has been mainly to check Mr. Paiva’s rough descriptions, give his species names, and compare them with the other known species of the genus. As at the close of this work I had obtained a good deal of knowledge of the genus I have ventured to incorporate the description of another species which I believe to be new. [C. Dover.] Corixa rambhaensis, sp. nov. Two specimens from an ornamental fountain in the palace of the Raja of Kallikota, Rambha, Ganjam District, Madras (N. Annandale, 3°xii*13). Head stramineous, longer than width at base between eyes, about half the greatest breadth of the pronotum; on each side of the centre of the vertex with a short line of shallow punctures, and a small tubercle on the middle of the hind margin; eyes large, posteriorly overlapping the anterior angles of the pronotum. Pronotum ochraceous, with six dark castaneous transverse fasciae ; about twice as broad as medial length ; obtusely angularly rounded, behind lateral angles posteriorly somewhat acutely pointed. Elytra very pale ochraceous, rather faintly mottled with castaneous ; costal margin pale white. The body beneath and the legs pale ochraceous. Length 6°15 mm. This species is closely allied to C. promontoria and C. affints, ' Faun. Brit. Ind. Rhyn. V, p 340, 1910. 2 Rec. Ind. Mus. XIV, p. 19, 1918. ® Dr. Annandale has published a short obituary notice of Mr. Paiva in his Report on the Zoological Survey India, for the years 1917-1920 (Calcutta: 1920), which has also been adopted by Mr. T. Bainbrigge-Fletcher, Imperial Entomolo- gist, in his Presidential address to the Fourth Entomological Meeting held at Pusa in February, 1921. (See Rep. Proc. Fourth Ent. Meeting Pusa, 1921, p. 4.) 332 Records of the Indian Museum. [VoL, XXIV, but differs from the former in the less produced head and the absence of any dark markings on the hind tibiae, and from the latter in the more regularly fasciate pronotum and less mottled elytra. Type in the collection of the Zoological Survey of India. Corixa annandalei, sp. nov. ‘Two specimens from Satpara, Orissa (NV. Annandale, 16°ix’13). Head ochraceous, about as long as width at. base between eyes, obscurely centrally carinate, the carina ending posteriorly in an obtuse tubercle at the middle of the hind margin of the head ; on each side of the carina a row of four or five shallow punctures ; face with a few scattered rather long silky white hairs ; a blackish spot at apex of clypeus. Pronotum ochraceous, with five distinct dark castaneous transverse fasciae, the second one short and not reaching the lateral margins ; breadth between humeral angles twice the medial length, posterior margin broadly rounded; anterior and posterior margins narrowly dark castaneous, the ochraceous interspaces somewhat broader than the fasciae. Elytra ochraceous, rather densely mottled with castaneous, the markings on the clavus linear, and transverse towards the base; costal margins dull white, very sparingly mottled with castaneous a little before apex, a linear castaneous marginal fascia at apex. Length 7°25 mm. Body beneath and legs ochraceous. Easily distinguished from all other Indian species of Corixa by its large size and the small number of fasciae on the pronotum. I have much pleasure in associating this species with the name of its collector, Dr. N. Annandale, to whom I am personally indebted for many favours. ‘Type in the collection of the Zoological Survey of India. Corixa dubia, sp. nov. A es lial from Mazbat, Darrang District, Assam (S. WW. Kemp, 4*i"I1). Head ochraceous, about as long as width at base between eyes, distinctly acutely tuberculate at middle of posterior margin, a short rather obscure carina on posterior area of vertex, on each side of which is a moderately long line consisting of shallow single punctures ; posterior margin narrowly spotted with castaneous ; eyes blackish-grey. Pronotum ochraceous with six transverse blackish fasciae, the second and third broken in the centre, anterior area with a short carina, posterior margin obliquely subacute. Elytra ochraceous, mottled with castaneous ; costal margin much paler, with three fasciate fuscous spots on outer margin, the apical one darkest and most conspicuous. Body beneath and legs ochraceous, posterior tarsi with two fuscous streaks on upper side. Length 6 mm. 1922.] C. A. Paiva & C. DovER: New Indian Cortxidae. 333 Closely allied to C. affinis, but differing in the nature of the markings on the pronotum. Type in the collection of the Zoolo- gical Survey of India. Corixa ribeiroi, sp. nov. A single example from Malwa Tal, 3000 ft., Kumaon, W. Himalayas. Head yellowish-white, shining, the basal margin reddish- brown, distinctly longer than width at base between eyes, wider than pronotum, a few scattered punctures on disk of vertex; eyes black, large, overlapping the anterior angles of the pronotum. Pronotum twice as broad as medially long, disk with about six castaneous, transverse lines, posterior margin broadly rounded. Elytra ochraceous, rather thickly mottled with castaneous, on the basal claval area the markings are more or less linear and transverse. Body beneath and legs ochraceous, posterior tarsi fuscous. Length 6 mm. I have named this species after Mr. Sydney Ribeiro, the Entomological Assistant of the Zoological Survey of India, in recognition of the assistance he has rendered me in various ways. Type in the collection of the Zoological Survey of India. Corixa paivana, Dover, sp. nov. Several specimens from Kalka, Umballa District, base of W. Himalayas (N. Annandale, 16'v’11); Dhurampur Kooa, Patiala State, base of Simla Hills (R. Hodgart, 21°vii17) ; Satpara, Orissa (NV. Annandale, 16°1x°13); from an ornamental fountain in the palace of the Raja of Kallikota, Rambha, Ganjam District, Madras. Anwargangi, Cawn- pore District, U.P. (F. Caunter, 1-13°x'11). Head pale yellowish, rather paler at base, short, about as long as width at base between eyes, a distinctly raised tubercle at middle of hind margin in front of which is a short blackish fasciate spot; a few shallow punctures on each side of the middle of the vertex, and another line of small punctures within the margin of each eye. Pronotum olivaceous-brown, unicolorous, minutely punctured ; anterior margin slightly sinuate in the middle, lateral margins truncate, its posterior angle acute; posterior margin rounded; a pale, somewhat obscure medial carina on middle of disk. Elytra olivaceous-brown, thickly but very finely punctured, having a large piceous spot on anterior area of clavus; the subcostal area dull ochraceous, the outer margins fuscous or black. The body beneath and legs rather pale ochraceous. Length 7-7°5 mm. A distinct species. I have named it after the late Mr. C. A. Paiva as a slight recognition of his services to entomology. Type in the collection of the Zoological Survey of India. Oe Se eee owe | RS a ier ON SOME INDIAN DERBIDAE (HOMOPTERA), By F. Muir, Hawaiian Sugar Planters’ Experiment Station, Honolulu, T.H. I have recently received for study a small collection of Indian fulgorids belonging to the Zoological Survey of India. The pre- sent paper deals with the Derbidae. This is an interesting family of mostly small and delicate insects found in forest lands. The eggs are at present unknown. The young live under bark and in rotten wood; what their food is is not at present known. ‘There are over ninety genera and nearly five hundred species described. The species generally have a limited geographical distribution, especially the island forms, but this cannot always be recognized unless the genitalia be examined, as the species ate often difficult to recognize by any other character. The types have been returned to the Indian Museum, cotypes have been retained by the describer. Measurements are from apex of head to anus and from base to apex of one tegmen. DERBINAE. CENCHREINI. Herpis turae, sp. nov. (Fig. 1.) Male. Length 3 mm.; tegmen 5 mm. Length of vertex equal to width at base, apex narrower than base, lateral margins thick and granulate, base angularly emar- ginate. Face fairly narrow. Sub- antennal process distinct but not very large. Pygofer produced angularly below anal segment. Anal seg- ment large; anus slightly distad of middle, in dorsal view slightly narrowed beyond anus to the truncate apex. Genital styles large, long, comparatively nar- row, margins subparallel, apex — Pgyxr-riG, 1.—Lateral view of male rounded, inner margin slightly genitalia of Herpts tuvae, sp. nov. sinuate, outer margin with a narrow border turned inward, two small processes near base, one pointed and curved, the other smaller with truncate apex. Head, thorax, legs and genitalia light brown, abdomen reddish. 336 Records of the Indian Museum. [VoL. XXIV, Tegimina stramineous, veins yellowish, all the apical cells slightly fuscous. Wings slightly fuscous with brown veins. Female. Length 4:.4mm.; tegmen5*4mm. The subantennal process is larger thaninthe male. In colouration similar to male. Pregenital plate longer than broad, hind margin evenly produced from the side to middle, the apex of the production rounded, the produced portion forming more than half the length of the plate. Described from one male and one female from above Tura, Garo Hills, Assam, 3500 to 3900 ft., July, 1917 (S. Kemp) ; and one female from Shillong, Khasi Hills, Assam, elevation 5500 to 6400 ft., August, 1915 (S. Kemp). I have not seen the type-species of Vekunia Dist., but the figures of it show the subcostal cells short and there is no suban- tennal process. The present species of Herpis differs from such species as Herpis vulgaris in having the vertex much longer and the lateral margins broad and granulate, as in Vekunta. Its division into two subgenera may be convenient. Vekunta flavipes, sp. nov. Female. Length 36 mm.: tegmen 5°4 mm. Subantennal process forming a very small flange below the antennae. Pregenital plate very short at sides, the middle half produced into a large process, the sides at first gradually curved then nearly straight with the apex rounded. Anal segment slightly longer than broad, anus near base, apex subtruncate. Genital styles fairly large, projecting beyond process of pregenital plate. Very dark brown or black; as antennae, face and clypeus light brown, legs yellow, hind margin of abdominal sternites yellow. Tegmina dark brown or nearly black, yellowish along costal margin, a small yellow spot at stigma. Described from one female from Tura, Garo Hills, Assam, 3500 to 3900 feet elevation (S. Kemp, August, 1917). OTIOCERINI. Lyricen vagans, sp. nov. (Fig. 2.) Male. Length 3 mm. ; tegmen 5 mm. The vertex and face slightly wider than in the type-species and the cubitus forks slightly lower down, otherwise quite typical. Light brown, darker over lateral portions of pronotum and mesonotum, face, genae, clypeus and front coxae. Tegmina fus- cous with darker markings in middle of clavus, fork of Cu, middle of M and fork of Sc and R; veins darker at base of median sectors and apical cross-veins. I,ateral margins of pygofer widely angularly produced below anal segment; medio-ventral margin angularly produced with a pair of curved spines just within the pygofer (these may pertain 1922.] F. Murr: Indian Derbidae. 337 to aedeagus). Anal segment short, broad, slightly narrowed at apex which is slightly rounded, anus near apex. Genital styles large, long, curved dorsad on apical third, apex rounded ; outer margin about middle produced into a wide angle, distad of that the margin is roundly excavate with a curved spine arising from the bottom of © the emargination, inner margin produced into a small, quadrate process about middle. Female. Length 4mm.; tegmen 73mm Text-FiG. 2.—Lateral view of Le 4 male pygofer of Lyvicen The hind margins of abdomen eae sp. Ag red. The tegmina lighter than the males, with the mottling more distinct. Pregenital plate broader than long, hind margin angularly produced to middle, the sides of the production slightly excavate, . apex rounded and slightly lipped, in lateral view flat. Described from two males and two females from above Tura, Garo Hills, Assam, elevation 3500 to 3900 ft., August, 1917 (S. Kemp). The genus has hitherto been known only from Fiji. Kamendaka (Eosaccharissa) albipennis, sp. nov. (Fig. 3.) Male. Length 2 mm.; tegmen 2°7 mm. In profile the vertex and face meeting at an angle of about 95, face curved, especially the apical half. Stramineous thickly encrusted with white waxy secretion ; a dark fuscous mark on genae in front of eyes continued over pronotum ; clypeus and front legs slightly fuscous. Wings thickly encrusted with white waxy secretion, slightly fuscous at apex of media, in the middle of tegmina and at apex of clavus. Wings white with waxy secretion, hyaline with yellow veins. Ventral margin of pygofer produced into a long, narrow an- gular process nearly half the length of genital styles. Anal segment long, narrow, apical third turned ventrad at right angle to base, anus at base of” the apical third, apex narrow with a small emargination making it minute- ly furcate. Genital styles long, nar- row, apex obliquely truncate and pieeryeae seer view of slightly emarginate, outer margin daka (Husabendpase) ol slightly produced just basad of middle hbipennis, sp. nov. with a small curved spine distad of the produced part; inner margin slightly sinuous. 338 Records of the Indian Museum. [Voy. XXIV, Described from one male specimen from Barkuda Island, Chilka Lake, Madras Pres., July, 1916 (Ff. H. Gravely). The species of the three subgenera that constitute this genus are mostly obscurely coloured and difficult to recognize unless the genitalia be examined. ‘The genitalia of none of the Indian species have been described. Niceta kanarae, sp. nov. Female. Length 5 mm.; tegmen 8 mm. In lateral view head produced in front of eye slightly more than the width of the eye, the width of the head greater than the depth. Antennae cylindrical not reaching to apex of head, a small curved knob at base. Pregenital plate wider than long, very short at sides, hind margin gradually and angularly produced to middle; the sides of _the production slightly sinuous and the apex slightly rounded. Stramineous ; fuscous over genae in front of eyes and over pronotum behind eyes, and over mesonotum, ‘Tegmina hyaline, slightly opaque with waxy secretions fuscous over apical half of clavus, over cubitus from the fork and over median apical cells, a little mark in radial cell at cross-vein and in apical cells, veins yellow. Wings hyaline, veins yellow, slightly opaque with waxy secretion. Described from one female specimen from Castle Rock, North Kanara District, Bombay Pres., October, 1916 (S. Kemp). Phra amplificata Distant. Phra, Wistant, Muir, Ext. Mo. Mag. 1918, p. 242. One female specimen from Castle Rock, North Kanara Dis- trict, Bombay Pres., October, 1916 (S. Kemp). I have already remarked upon the type oi this insect and the difference in the figures of the head. The present specimen agrees with the type-specimen and is apparently the same species. Mysidiides fuscinervis, sp. nov. (Fig. 4.) Male. Length 2 mm. ; tegmen 5 mm. Light brown ; mesonotum darker. ‘legmina hyaline, slightly opaque with waxy secretion, with slight fuscous marking over base and clavus, two or three marks in costal cell, over Cu 1, base of median sectors spreading out into the median cells, over apical cells; the veins darker where the membrane is fuscous. Wings slightly fuscous with brown veins. Lateral margins of pygofer produced into a broad angle below anal segment. Anal segment comparatively short, broad, lateral margins slightly flattened, produced into a small thin process be- fore apex; anus about middle, broadest distad of anus, apex 1922.] F. Murr: Indian Derbidae. 339 broad, slightly emarginate in middle making it broadly and slight- ly bilobed. Genital styles long, narrow, curved on apical third, a small projection on outer margin near base. Female. Lengh 2°9 mm.; tegmen 6 mm. In colour similar to male. Anal segment small, subtriangular, anus at base; pregenital plate wider than long, posterior margin widely angu- larly produced from sides to middle, |. the sides of the production slightly !*T:TIG. 4—_Lateval view ol excavate; in lateral view straight, Sees Neeser, hind margin not turned ventrad. Sp. nov. Described from a male and female (types), from Talewadi, near Castle Rock, North Kanara District, Bombay Pres., October, 1916 (S. Kemp); and two males and two females from Castle Rock, October, 1916 (S. Kemp). Mysidiides furcata, sp. nov. (Fig. 5-) Male. Length 1:8 mm.; tegmen 5 mm. Dark brown; lighter over middle of mesonotum, legs, lateral portions of pronotum and genital styles. Tegmina hyaline, slightly opaque with waxy secretion, veins yellow except Cu 1, fork of Cu and first (basal) median section, radial cross veins and base of second median sector and apical cross-veins which are fuscous, the fuscous spreading out into the membrane, fuscous at apex of subcostal cell and over apical cells. Lateral margins of pygofer rounded, not angularly produced. Anal segment large, in lateral view curved ventrad, anus about middle, in dorsal view gradually narrowing to near apex where it is produced into ‘a furcate apex. Genital styles large, apical half curved dor- sad, apex rounded, outer mar- gin with two processes near TEeExt-FIG. 5.—Lateral view of male base, the basal one subquad- genitalia of Mysidiides furcata, rate and ‘broader than long, sp. nov. a. Apex of anal the distal one small and thin, segment. inner margin with a subquad- rate projection about middle. Female. Length 2°77 mm.; tegmen 6 mm. In colouration similar to male. In lateral view the pregenital . plate concave, the posterior margin slightly and angularly pro- duced in middle and turned ventrad, a minute emargination in the middle, 340 Records of the Indian Museum. [VoL,. XXIV, Described from one male and two females from Castle Rock, North Kanara District, October, 1916 (S. Kemp). Mysidioides, sp. One female specimen from above Tura, Garo Hills, Assam, August, 1917 (S. Kemp), in colouration similar to M. fuycata, but having the pregenital plate flat and not turned ventrally at apex. In the absence of the male I refrain from naming it. DERBINI. Zeugma fuscinervis, sp. nov. Female. Length 5 mm.; tegmen If mm. Characteristic of the genus; face fairly narrow. Head and pronotum light brown, darker over vertex and down the middle of face and clypeus and in the middle of prono- tum ; mesonotum and tegulae dark brown or black, lateral carinae lighter ; abdomen dark brown; legs lighter brown. ‘Tegmina hya- line slightly yellowish, slightly fuscous over apical and hind areas, veins dark spreading into membrane, a small dark spot at base of cubitus, another in radial cell and one at fork of cubitus. Wings hyaline, slightly fuscous, veins brown. Pregenital plate large, in lateral view well rounded, middle third of posterior margin produced into a plate broadly conical in outline, a keel runs from apex to a little beyond the base of pro- duced portion. Anal segment short, ventral edge produced beyond apex with a small patch of short stout hairs at each angle of apex. Described from one female from above Tura, Garo Hills, Assam, 3500 to 3900 ft., July, 1917 (S. Kemp). RHOTANINI, Sumangala delicatula Distant. One male specimen from Castle Rock, North Kanara District, October, 1916 (S. Kemp). The figure of this species has no cross-vein between the media and cubitus (really the base of first or basal median sector) other- wise this specimen agrees with the figure and description. Levu iridipennis ? Melichar. Rhotana iridipennis, Melichar, Hom. Faun. Ceylon, p. 62. One male specimen from above Tura, Garo Hills, Assam, 3500 to 3900 ft., August, 1917 (S. Kemp). The specimen is not in very good condition, but it appears to agree with the original descrip- tion. £922. ] F. Murr : Indian Derbidae. 341 ZORAIDINAE. ZORAIDINI. Pamendanga pallata (Distant). (Fig. 6.) Phenice pallata, Distant, Ann. Mag. Nat. Hist. (8) VIII, p. 639 (1911) Faun. Brit. Ind. Rhynchota V1, appendix, p. 64 (1916). Two males and four females from above Tura, Garo Hills, Assam, 3500 to 3900 ft., August, 1917 (S. Kemp). This was originally described from a female from Kumaon, W. Himalyas. The male is coloured similarly to the female. The male pygofer is sunk within the preceding segment, ventral margin slightly and roundly produced. Anal segment large, anus before mid- dle, evenly curved ventrad, in dorsal view widest at base, gradually and slightly narrowed to the rounded apex. Genital styles small, short, slightly curved, apex rounded, at base on outer margin produced into a large, TExt-r1G. 6.—Pamendanga quadrate process, on inner margin pallata (Distant). into a stout pointed process. a. Se eye male In outline the pregenital segment in Re Matera Gee oftaiale | female roundly produced somewhat - pregenital plate. like a Phrygian cap. Zoraida brunnipennis, sp. nov. (Fig. 7.) Female. Length 5°3 mm.; tegmen 14 mm.; wing 6 mm. Antennae longer than face, cylindrical. Four cubital veins - reaching hind margin; radial cell not very narrow, slightly widened distad of second median sector. Pregenital plate large, in lateral view concave, hind margin roundly produced and with a small cleft in the middle. Anal segment large, reaching to apex of styles, broadest at base, slightly narrowing to the rounded apex, anus at Text-Fic. 7.—Dorsal view of female geni- base. : : 5 . talia of Zoraida Brown ; tegmina and wings hyaline, uni- brunnipennis, formly brown with brown veins. Sp. nov. Described from one female from Tura, Garo Hills, Assam, 1200 to 1500 ft., July, 1917 (S. Kemp). 342 Records of the Indian Museum. [VoL. XXIV, 1922.] Zoraida (Peggiopsis) kempi, sp. nov. (Fig, 8.) Male. Length 3 mm.; tegmen 8 mm.; wing 3°8 mm. Antennae slightly longer than face, flat, broad. Radial cell not very narrow, slightly widened beyond cross-vein. Stramineous inclined to salmon, more so over antennae and abdomen Tegminahyaline, slightly fuscous ; costal, subcostal and radial cells fuscous ; veins brown, apices of apical veins light. Medio-ventral process of pygofer longer than broad, apex trun- cate, slightly narrower than base. Anal segment large, anus before middle, in dorsal view narrrow at base, slightly widened beyond anus then narrowed to the acute apex, the apical third turned ventrad. Genital styles large. narrow at base, considerably widened on apical half, apex truncate; outer margin considerably produced be- yond middle. Aedeagus complex, ‘Trext-riG. 8.—Lateral view of male pygofer of Zoraida not dissected out. (Peggiopsis) kempi, Described from two males from Sp. Nov. Mormugao, Portuguese India, Sep- tember, 1916 (S. Kemp). NEW INDIAN HOMOPTERA. Ly F. Muir, Hawatian Sugar Planters’ Experiment Station, Honolulu, T. H. The material dealt with in this paper belongs to the Zoological Survey of India. Ina former paper the Derbidae were dealt with. This paper deals with the Cixiidae, Delphacidae, and Achilidae. Sixteen of the twenty-six species recorded are considered as new, which indicates the large amount of work still to be done in these families. The types have been returned to the Zoological Survey of India, and paratypes placed in the H.S.P.A. Experiment Station collec- tion, Honolulu. Measurements are from apex of vertex to anus and from base to apex of one tegmen. Family CIXIIDAT. Cixius gravelyi, sp. nov. Congeneric with C, nervosa ut the base of vertex more deeply and angularly emarginate. Male. Length 3°55 mm. ; tegmen 4°7 mm. Rlack ; lateral carinae of face and vertex and hind margin of pronotum and the legs light brown, basal portion of abdomen yellow, apical portion brown. Tegmina hyaline, base brown, a broad light brown or yellowish band across from middle of clavus to middle of costa; stigma brown; fuscous over apical cells, darker in apical radial cells; cross-veins infuscate; tubercles small, dark, bearing black macrotrichia. Wings hyaline with brown veins. Lateral margins of pygofer roundly produced beside anal seg- ment, medio-ventral process angular. Anal segment large, longer than wide, anus about one-fourth from apex, apex rounded and turned ventrad. Genital styles long, narrow, apex acute, the middle of the inner margin produced into an angle. Aedeagus large, complex. Described from one male from the Darjiling District, East Himalayas, India, elevation 4000 feet (F. H. Gravely, June, 1916). Genus Oliarus Stal. In describing the genus Mnemosyne Stal states that the meso- notal carinae are obsolete, but he placed M. philippina in the genus although it has five carinae. Fowler in describing the genus says, “‘ the three keels on the scutellum more or less obsolete.’’ Distant places Oliarus punctipennis and M. cingalensis in Mnemo- syne although they have five mesonotal carinae. 344 Records of the Indian Museum. [VoL. XXIV, Apart from the carinae of the mesonotum Mnemosyne and Oliavus appear to differ only in the width of the vertex, and as the width of the vertex varies in the different species of Oliarus it is difficult to draw the line between them. Until the type species of Mnemosyne is redescribed I shall consider that it has three meso- notal carinae and describe all those having five, and only differing in the width of vertex, as Oliarus. Oliarus kempi, sp. nov. Female. Length 4:3 mm.; tegmen 5°3 mm. Length of vertex 3°3 times the width at base, base slightly wider than apex, inner carinae leaving lateral carinae one-third from base, gradually converging to apex where they meet and touch the apical transverse carina. Face very narrow at base; median carina forked at base. Fronto-clypeal suture straight at sides, the middle half rounded. Median ocellus present a little distance before apex of face. Forking of Sc and R slightly distad of fork of Cu. Pygofer oblong, broader than long (1°8 to 1). Anal segment small, half the width of pygofer, ovate, anus at apex. Genital styles not quite so long as anal segment. Hind margin of pregeni-. tal plate slightly emarginate in middle, the corners forming a small angular projection. Dark brown; carinae of head and pronotum, abdominal pleura and hind margin of segments yellow. Tegmina hyaline very slightly infuscous, darker over apical portion; veins brown with small tubercles bearing small, black macrotrichia. Described from one female specimen from Talewadi, near Castle Rock, North Kanara District, Bombay Pres. (S. Kemp, October, 1916). : Oliarus kierpurensis, sp. nov. Male. Length 3°77 mm.; tegmen 5 mni. Vertex a little longer than width at base (1°3 to I), base 1°4 times the width at apex. Latero-median carinae leaving lateral carinae about one-third from apex, converging and meeting together a little before apical transverse carina to which they are joined by a short carina; base angularly emarginate. Fronto- clypeal suture forming a half circle. Median ocellus present at apex of face. Cw forking slightly basad of Sc and R. Lateral margins of pygofer rounded, medio-ventral margin produced into a small, sublanceolate process. Anal segment large, dorsal surface tectiform, ventral surface concave, anus at apex which is slightly narrowed and emarginate. Genital styles large, flattened, elongate, S-shape with the apex widened. Dark brown; carinae of head and pronotum and the middle of pronotum, legs and basal half of abdominal segments yellow or light brown; carinae of mesonotum slightly lighter than disc. Tegmina clear hyaline, veins light yellow, tubercles small, yellow, 1922.] F. Muir: New Indian Homoptera. 345 bearing white or yellowish macrotrichia; over the apical area the veins and tubercles are darker. Female. Length 3°7 mm.; tegmen 5'2 mm. In colour similar to male. Pygofer wider than long. Anal segment small, reaching about half-way across pygofer, in dorsal aspect quadrate, a little longer than wide, anus at apex. Hind margin of pregenital plate very slightly rounded and minutely emarginate in middle, the margin curved ‘slightly dorsad. Genital styles long, reaching nearly across pygofer. Described from one male and two females from Kierpur, Bihar, India (C. Paiva, October, 1915). Oliarus goae, sp. nov Female. Length 4 mm.; tegmen 5'4 mm. Length of vertex twice the width at base, base very slightly wider than apex, angularly emarginate; medio-lateral carinae leaving sides one-fourth from apex, straight, converging and meeting in middle slightly before apex. Face narrowed for some slight distance at base, fronto-clypeal suture obscure, median ocellus distinct. Cz forking some distance basad of fork of Sc and R. Pygofer large, wider than long. Anal segment subdiamond shape, slightly broader than long, apex small, truncate, on dorsal aspect a raised, longitudinal ridge down middle. Hind margin of pregenital plate very slightly curved. Genital styles reaching across pygofer. Dark brown or black; carinae of head and thorax lighter, more especially so on pronotum; legs lighter brown; hind margin of abdominal segments yellowish. Tegmina hyaline, very slightly opaque and whitish, veins light brown with darker tubercles bear- ing light brown macrotrichia; a dark mark on margin of clavus at apex of claval vein; fuscous over cross-veins and apical cross-veins and apical veins; stigma dark brown. Wings hyaline with brown veils. Described from one female from Mormugao, Goa, Portuguese India (S. Kemp, November, 1916). Oliarus turae, sp. nov. Female. Length 3°8 mm.; tegmen 5°7 mm. Length of vertex from apex to basal angles slightly greater than width at basal angles ; base deeply and angularly emarginate, I°4 times the width at apex ; medio-lateral carinae arising from the sides about one-third from apex, converging and meeting in middle at apex. Pygofer large, oval, width 1°5 times the length. Anal segment not reaching quite across pygofer, flat, length nearly twice the width, sides slightly curved, width about one-third the width of 346 Records of the Indian Museum. VoL. XXIV, pygofer, ovipositor incomplete, the styles reaching about two-thirds across pygofer ; pregenital segment small, hind margin straight or very slightly curved. Dark brown or black ; carinae of frons, vertex and pronotum and the margin of pronotum and margin of metanotum light brown, legs light brown. Tegmina with venation as in O. kurseon- gensts, Dist. ; clear hyaline with brown veins, stigma brown, tuber- cles brown bearing black macrotrichia. Wings hyaline with brown veins. Described from one female from Tura, Garo Hills, Assam, 3500 to 3900 feet elevation (S. Kemp, July, 1917). - Kuvera brunettii, sp. nov. Male. Length 3°4 mm.; tegmen 46 mm. I,ateral margins of pygofer slightly curved, medio-ventral process conicalin outline. Anal segment considerably longer than broad, anus about one-third from apex, sides straight to near apex then converging to pointed apex. Genital styles narrow at base, broadly round at apex, outer margin strongly concave, inner mar- gin nearly straight. This forms a sickle-shape organ with handle very thin and blade broad. Aedeagus large and complex. Dark brown or black; carinae of frons and vertex, the hind margin of pronotum, tegulae, legs, margins of pygofer and genital styles lighter brown. Tegmina hyaline with brown veins which are blacker over apical half; tubercles small with black macrotri- chia; stigma dark brown, light at base. Wings hyaline with brown veins. : é Female. Length 3:6 mm.; tegmen 5 mm. In colour similar to male. Pygofer small, slightly wider than long, concave, wax-bearing ; ovipositor complete, large, project- ing more than half its length beyond apex of pygofer. Anal segment small, about as wide as long, apex truncate, reaching about two-thirds along pygofer. Described from two males and three females from Darjiling, Eastern Himalayas, India, elevation 7000 feet (EZ. Brunetti, May, 1917). ‘This species is closely allied to the type species K. semi- hyalina, Dist. Mundopa vagans Dist. One female from Tura, Garo Hills, Assam, 3500 to 3700 feet elevation (S. Kemp, August, 1917). This agrees with the descrip- tion but the vertex is not so wide at apex as is indicated in the figure. . Mundopa pashokensis, sp. nov. Female. Length 2°7 mm.; tegmen 3°77 mm. Apex of vertex slightly narrower than base; width at apex three times the length in middle; base roundly emarginate. Pygofer small, much longer than wide; ovipositor complete, large, extending nearly half its length beyond apex of pygofer. 1922.) F. Muir: New Indian Homoptera. 347 Anal segment cylindrical, long, about four times as long as broad. Dark chocolate brown; lateral carinae of face except the apical third, clypeus except the basal sides, lateral portions of pronotum and the legs, lighter brown. Tegmen hyaline, the apical two-thirds, from slightly before stigma and apex of clavus, dark brown with four light areas, one on costa at stigma, one at apex of clavus, a larger central one stretching from radius to clavus, and one at apex of median veins; a dark band across base ; veins same colour as membrane, tubercles minute bearing fine macrotrichia the same colour as veins. Wings hyaline, fuscous, darker over apical half of costal area. Described from one female from Pashok, Darjiling District, India, 2v00 feet elevation (Ff. H. Gravely, May, 1916). There is a third species in the collection which appears to be undescribed, but as it has no abdomen and the sex is not known, I refrain from naming it. Brixia albomaculata Dist. Eight males and six females from Castle Rock, Kanara District, India (S. Kemp, October, 1916). Brixia plagosa Dist. One male and one female from Tura, Garo Hills, Assam, 3500 to 3900 feet elevation. These specimens agree with the description except that they have no median carina on face. If the description be correct then this identification may not be correct and Distant’s species may be a Letrioessa Kirk. Leirioessa pulchra, sp. nov. This genus differs from Brixia Stal, in the shortness of the autennae; the frons has a median carina and Sc, Rand M separate to the basal cell and do not form a stalk. Male. Length 3 mm.; tegmen 4°6 mm. Dark chocolate brown; antennae, carinae of face, lateral portion of pronotum and the carinae of mesonotum light brown or yellow, legs and genital styles light brown. Tegmina hyaline with light infuscations and darker markings. A dark mark at base over basal cell and half-way along suture and extending to first claval vein ; a large subquadrate mark from costa to cubitus commencing at apex of basal mark and ending at apex of clavus, on the costa it includes two light marks, the basal one triangular and the distal one smaller and round; the apical cells of R and M dark enclosing two lighter marks; three small marks in a row, one from hind margin to Cu 2a, the second from M 2+3 to K and the third at apex of stigma; veins light with minute brown tubercles bearing fine brown macrotrichia. Described from two males from Talewadi near Castle Rock 348 Records of the Indian Museum. [ Vor. XXIV, North Kanara District, India (S. Kemp, October, 1916). This species comes near to the type, L. tortricomorpha Kirk., from Australia. It is possible that Cotylecebs Uhler, and Leirioessa Kirk., are the same, as a specimen I identify as C. marmorata Uhler, from Japan is generically the same as the above species. It is also probable that both are the same as Andes Stal. Leirioessa mander (Walk.). Byixia meander (Walk.), Distant, Faun. Brit. Ind. Rhyn. Ill, p. 270 (1906). Two males from Castle Rock, North Kanara District, Bom- bay Pres. (S. Kemp, October, 1916). Leirioessa nubila (Walk). Brixia nubila (Walk.), Distant, Faun. Brit. Ind. Rhyn. Wl, p. 270 (1906). One male and two females from Castle Rock, North Kanara District, (S. Kemp, October, 1916). If my identifications of these two species be correct then they should not be placed in Brixa. \ Borysthenes fascialatus, sp. nov. (Fig. 1.) The genus Borysthenes is distinguished from most of the Cixiidae by the presence of ~a swollen subantennal process which is fringed with hairs. When at rest the hind margins of the tegmina overlap considerably, but not in the same manner as in the Achilidae and the claval veins do not reach the apex of the clavus. Male. Length 3 mm.; tegmen 4 mm. The lateral margins of pygofer produced broadly beside the anal styles, the production on right side being broader than that on the left, medio-ventral edge angularly produced. Anal segment large, anus near base, broad before anus then narrowed to bluntly Be dee iG pointed apex, the edges of PEXxT-FIG. es ACG LOSE the apical portion turned Lateral ae i aeTe genitalia. ventrad and produced into an angular process on each side, the one on the left the larger. Genital styles narrow, long, ‘ 1922.] F. Muir: New Indian Homoptera. 349 subequal in width throughout, bent at an angle before middle, apex bluntly pointed. Aedeagus complex, consisting of two parts, a large basal portion and an apical portion at an angle to the basal portion. A strong chitinous tube runs through the middle of the basal portion, its apex in connection with the apical portion of aedeagus and its base joined to the structure connecting with the base of genital styles. The outer portion of the basal part of aedeagus is membranous with two large sclerites and having two large spines arising from near the apex. The apical portion consists of a cup-shaped organ from the edge of which two long, slender spines arise, there is a third curved spine from the outer portion of the cup. Head, pronotum and legs light brown, mesonotum and abdo- men dark brown. Tegmina hyaline, fuscous with two light bands and a light mark, the first band from costa at apex of subcostal vein to hind margin, the second from before stigma to hind margin above clavus and the light mark over base of cubitus; veins same colour as membrane without tubercles or macrotrichia. Female. Length 3°6 mm.; tegmen 4°3 mm. In colour similar to male. Pygofer small, longer than broad, depressed down the middle, forming a wax-secreting area; ovipositor complete, moderate in size, reaching to apex of pygofer. Anal segment short, apical angles slightly produced. Described from five males and three females from Castle Rock, North Kanara District (S. Kemp, October, 1916). Kinnara spectra Distant. One female specimen which agrees with the original descrip- tion. Above Tura, Garo Hills, Assam, 3500 to 3900 feet elevation (S. Kemp, August, 1917). This genus, like Borysthenes, has a subantennal process. . Kinnara maculata Distant. One male from Talewadi near Castle Rock and two females from Castle Rock, North Kanara District (S. Kemp, October, 1916). These agree with Distant’s description but the dark mark at base of tegmina is slightly more extensive. Genus Commolenda, Distant. The genus is described as having two ocelli at the apex of the frons. This, I think, must be an error as no homopteron has a pair of ocelli in that position and if the normal lateral ocelli be present then it possesses four ocelli. Apart from this character I cannot separate it from Ptoleria Stal. In Australoma Kirk. the vertex is distinctly wider than long and there is no longitudinal median carina. 350 . Records of the Indian Musewm. [VOr. XEXaby Australoma brunnia, sp. nov. Female. Length 3°6 mm.; tegmen 5 mm. Width of vertex 2'7 times the length. Pygofer much longer than wide, depressed down the middle for the reception of ovipositor which is complete, curved and reaches a little beyond the apex of pygofer ; lateral plates short, broad at base, inner margin slightly concave apically convex basally; posterior margin of seventh sternite straight. Anal segment short, convex dorsally flat or slightly concave ventrally ; anus at apex. Head light brown, darker over apical half of lateral carinae of frons and in the basal half of middle of frons; pronotum light brown, darker on hind margin; mesonotum dark brown; legs and abdomen light brown. Tegmina light brown darker over posterior half including clavus, veins same colour as membrane with a double or treble series of minute dark tubercles bearing black macrotrichia. Described from one female from above Tura, Garo Hills, Assam, 3500 to 3900 feet elevation (S. Kemp, August, 1917). Kermesia parva, sp. nov. Female. Length 2 mm.; tegmina 3°6 mm. Stramineous ; tegmina and wings hyaline, milky white with waxy secretion, veins light yellow. ‘Tubercles along first claval vein and Sc + R. The M3+4 and Cw I are in contact for a short distance. Described from one fernale from Pashok, Darjiling District, 1000 feet elevation (Ff. H. Gravely, June, 1916). ‘The small size of this species distinguishes it from K. albida Mel. Family DEI,PHACIDAE. Nilaparvata sordescens (Motsch.). Delphax sordescens, Motsch. Bull. Soc. Nat. Mosc. XXXVI, p. 109 (1863). Liburnia sordescens (Motsch.), in Melichar’s Hom. Faun. Ceylon, p. 102 (1903); Distant, Faun, Brit. Ind. Rhyn. III, p. 486 (1906). Vilaparvata greeni, Distant, Faun. Brit. Ind. Rhyn. 111, p. 473 (1906); Muir, Can. Ent. Fan. p. 7 (1919). Kalpa aculeata, Distant, Faun. Brit. Ind. Rhyn. II, p. 474 (1906) ; Muir, Can. Ent. Fan., p. 8 (1919). Dicranotropis andervida, Wirkaldy, H. S. P. A. Ent. Bull. Wl, p. 133 (1907). Delphacodes anderida (\Niris.), Muir, Proc. Haw, Ent. Soc. Wl, 4, p. 335 (1917). One male specimen from Castle Rock, North Kanara District (S. Kemp, October, 1916). I have accepted Melichar’s identification of Motschoulsky’s species to be correct. If it be not correct then N. gveenz will be the name of the insect. It is only separated from Delphacodes by the presence of two or three small spines on the hind basitarsus. 1g22. | I’. Murr: New Indian Homoptera. 351 The genus includes D. bakevi Muir, and another species from Porto Rico not yet described. Kelisia fieberi Muir. One female specimen from Tura, Garo Hills, Assam (S. Kemp, July, r917). ‘This species is not typical of the genus, but had better remain until the genus is revised. This may be the same as Sogata pusana Distant. Euidella kashmirensis, sp. nov. (Fig. 2.) Male. Macropterous; length 2°8 mm; tegmen 3°7 mm. Head brown, lighter on vertex; pronotum and mesonotum light brown between carinae, dark brown on sides; abdo- men dark brown, yellowish at base; legs light brown. ' Tegmina hyaline, a dark brown mark over base of radial and median cells and a broad, semicircular mark from apex of costal cell, over subcostal cell the cross- veins and to apical margin between cubital apical veins and first median apical vein. In some specimens, includ- ing the type, the dark mark at base is more extensive b and extends along costal cell to the apical curved band ; veins dark brown Lexr-viG. 2.—Euidella kashmivensis, with minute tubercles bear- Sebie vee : a. lateral view of anal segment and ing very fine black macto- aedeagus. trichia. Wings hyaline, b. Kull view of right genital style. slightly fuscous with dark brown veins. Opening of pygofer round, a little wider than long, margins entire ; from the medio-ventral edge arises asmall process subconi- cal in outline and cleft down the middle. Phragma fairly long, dorsal margin straight, from the middle arises two small spines. Anal segment large with two spines on ventral margin, the one on the right large, strong, with several fine teeth on the apical half, the one on the left about one-fourth the size of the other. Aedeagus slightly flattened laterally, basal two-thirds straight, apical third bent at right angle and slightly wider, a small comb of teeth on dorsal aspect at bend and another on ventral aspect, both inclining to right side, a large flat flange-like spine at apical third on ventral 352 Records of the Indian Museum. [VoL. XXIV, aspect. Genital styles large, apex widely angular, inner margin roundly emarginate, outer margin angular. Female. Macropterous. lTength 3°8 mm.; tegmen 4°6 mm. In colour similar to male. Described from four males and four females from Kashmir, North-West Himalayas (H. T. Pease, 1915) and one male from Pashok, Darjiling District, East Himalayas (PF. H. Gravely, June, 1916). This species is very near to E. speciosa (Bohem) of Europe but the genitalia are different. I have only a brachypterous female of the European species for comparison. Sadia rostrata Melichar. One female specimen from Eden Gardens, Calcutta (F. 4H. Gravely, November, 1911). Family ACHILIDAE. Faventia pustulata (Walker). One female from above Tura, Garo Hills, Assam, 3500 to 3900 feet elevation (S. Kemp, August, 1917). Anal segment beyond anus pointed and curved ventrad. Faventia flava, sp. nov. Female. Length 5°4 mm.; tegmen 7°7 mm. Vertex broader than long, the base widely and angularly emar- ginate, apex widely angular, the carinae of face projecting in dor- sal view. Apart from the shape of vertex typical of genus. Yellow or light brown. Pronotum darker. brown with a narrow lighter hind margin, head darker between carinae. Teg- mina yellowish, veins same colour; fuscous betweén the oblique cross veins at apex of costal cell, a few small, scattered spots of brown over clavus and corium, a series of six minute dots in apical cells. Wings hyaline, veins light, fuscous over apical portion. Male. Length 5-4 mm.; tegmen 7 mm. In colour similar to female. Lateral margins of pygofer roundly produced ; medio-ventral angularly produced with the apex narrow and truncate. Genital styles large, apex rounded, produced into a narrow angular or spine- like projection on the outer margin near apex. Anal segment flattened horizontally, broadened slightly to apex which is round- ed. When at rest the genital styles come together in the middle line, the medio-ventral process of pygofer filling the space between their bases, the projection on outer margin of genital styles laying between the anal segment and the rounded projection of the lateral margins of pygofer, thus forming a ‘‘ closed”’ pygofer. Described from one male and one female from above Tura, xaro Hills, Assam, 3500 to 3900 feet elevation (S. Kemp, August, 1917). 1922. ] F. Murr: New Indian Homoptera. 353 Majella albomaculata, sp. nov. The genus Majella Kirkaldy differs from Gordia Melichar in having the vertex much longer than wide with its apex very much narrower than its base. Phenelia Kirkaldy differs from both the above by not having the ‘‘break’’ in the tegmina causing the apical portion beyond the apex of cortal cell and clavus to drop down over the end of the abdomen, there being a distor- tion of the veins along that line, especially of the cubitus. Female. Length 2°8 mm.; tegmen 3°7 mm. Vertex and face light brown, a dark mark on each side of the median carina of vertex, three dark marks across gena in front of eyes ; clypeus dark brown. Pronotum light brown with a series of five or six dark spots behind eyes; mesonotum dark brown with lighter carinae. Coxae dark brown, femora and tibiae light brown with two darker marks on hind tibiae; abdomen dark brown. Tegmina brown, darker over base, a dark mark at apex of costal cell with a small scarlet spot in the middle, veins darker brown with a number of small, white dots along them, some small light dots in costal cell. Wings light brown, veins darker. Anal segment oval, wider than long, anal style narrow, long. Male. Length 2.9 mm.; tegmen 3'4 mm. Colour similar to female. Pygofer short, broad, flattened horizontally; ventral margin produced in the middle into two right angle triangles which meet together on the middle line and appear as one angular projection. Anal segment broader than long, nearly semicircular with a small emargination at apex. Genital styles triangular with the apex forming the base, a small projection on the outer margin near apex. ‘he aedeagus complex, of the normal Achilid type. Described from one female and one male from above Tura, Garo Hills, Assam, elevation 3500 to 3900 feet (S. Kemp, August, IQt7). Rhotala gravelyi, sp. nov. (Fig: 3.) This species agrees with Walker’s generic description as far as it goes, only a comparison with the type will settle the question. In dorsal view vertex a little broader than long, apex broadly rounded, base roundly emarginate, a small de- pression in each side, ‘without carinae; the vertex stands up above the pronotum; frons longer than broad, apex TExt-F1G. 3.—Rhotala gravely, sp. nov. broader than hase, eee broadest in front of 354 Records of the Indian Museum. [| VoL. XXIV, antennae, smooth and shinv with small pits, no median carina and no true lateral carinae ; only the margins as seen in lateral view form carinae; clypeus with lateral and median carinae. Hind tibiae with six spines. Pronotum 5-carinate, the lateral outer ones in front of tegulae very distinct; mesonotum tricarinate with a small, round spot on each side near lateral carinae. Se and R parting about one-fourth from base, M not joining Sc+R at base, the cubitus with seven or eight apical veins. One of these may be M 3+ 4 touching the cubitus. Male. Length 7 mm.; tegmen 8°4 mm. Datk brown, face dark and shiny, light at apex, the front and middle tibiae banded. ‘Tegmina hyaline, brown along costa reach- ing back to fork of cubitus, slightly brownish over apical area, a series of seven dark spots along margin of clavus, veins darker. Wings fuscous with brown veins. Pygofer with ventral margin entire, not produced; genital styles large meeting together on middle line, apex rounded. Anal segment slightly flattened horizontally, slightly broader at base than apex; anus at apex which is rounded. Aedeaguscomplex but not dissected out. Described from one male from Pashok, Darjiling District, elevation 5500 feet (PF. H. Gravely, June, 1916). Magadha flavisigna (Walker) One female specimen and one with abdomen missing from above Tura, Garo Hills, Assam, 3500 to 3900 feet elevation (S Kemp, August, 1917). - Kempiana, gen. nov. Vertex slightly broader than long, apex slightly narrower than base, angularly produced, base angularly emarginate, no median carina, lateral carinae large. Frons considerably longer than broad, broadest at apex, tricarinate, the carinae continuing on to the clypeus. Pronotum short, 5-carinate, the carinae behind tegulae short and reaching a curved carina running from the medio-lateral carinae to the lateral margin. Mesonotum tricarinate ; the ante- rior portion between the carinae marked off by a different texture to the rest. The costa within the membrane forming a distinct costal membrane on basal half of tegmina. Sc and FR joined together for basal third, M with three apical veins. This genus is close to Magadha Distant, with the exception of the distinct costal area. Type, Kempiana maculata. Kempiana maculata, sp. nov. (Big: 4:) Female. Length 7°77 mm.; tegmen 8-4 mm. d 1922. F. Murr: New Indian Homoptera. 355 9 Dark brown; vertex lighter with a dark mark across the sides, frons with light marks down the sides which form a band across the middle; clypeus light at base and at apex; pro- notum lighter over middle; mesonotum lighter over lateral carinae; abdominal segments with light hind margins. TeEx-1iG. 4.—Kempiana maculata, sp. nov. Right tegmen. Tegmina hyaline speckled all over with brown, darkest over middle of subcostal and radial cell and middle of costal membrane and costal cell. Described from one female from above Tura, Garo Hills, Assam, elevation 3500 to 3900 feet (S. Kemp, August, 1917). —~_— Oeooearen ae aeeaeaSeoeeeSeeOSEOeSEOeesSe Alas, 5 in i [ae oy wosners ala pkey Rie ‘a. fee} 1 atti) fet, went wn fi Pe EOE: . f, oY ge out 4 ‘ fay Bice Eo wee: : sn ae - { a i Es ti, MATERIALS FOR A GENERIC REVISION OF THE FRESHWATER GASTROPOD MOLLUSCS OF THE INDIAN EMPIRE. No 5.—-THE INDIAN PI,ANORBIDAE. By N. ANNANDALE, D.Sc., F.A.S.B., Director, Zoological Survey of India. The Planorbidae are distinguished from their allies the Limnaeidae and Physidae by well-defined conchological, anatomi- cal and physiological characters. In their dextral bodies they come nearest the Physidae (which are not represented in’ the Indian fauna) and in the sinistral more or less ovate shells of one of their two subfamilies (the Bullininae) there is also a close resem- blance to the same family, but important differences are to be found in the radulae, the lateral teeth of which in the Planorbidae are simply cusped, while in the Physidae they bear a curious lateral process, A still more important difference is to be found in the colour of the blood, which is red in both subfamilies of the Plan- orbidae and colourless in the Physidae and also in the Limnaeidae and Ancylidae. The Bullininae, moreover, comprise comparatively few species and the much more numerous Planorbinae have disc- shaped shells quite unlike those of any of the other two families. The genitalia of the Planorbidae show great diversity in the structure of the male organ, but otherwise conform to the same type as those of the Limnaeidae. The digestive system is also similar, allowance being made for the more elongate type of body, correlated with the difference in shell-form, in the Planorbinae. The jaws in most genera consist, as in Limnaea, of a central more or less lunate or sublinear upper transverse piece and of two slender vertical sidepieces, but in some species they are broken up into many horny teeth as in the Ancylidae. The radulae bear smaller teeth with shorter and often more numerous cusps than in Limnaea. There is present on the left side of the body in the Planorbidae a vascular outgrowth (pseudobranch) of more complex structure in some genera than in others. Before discussing the subfamilies and genera I wish to say a few words about the colour of the blood. In all the Planorbid genera I have examined, including Bullinus and Camptoceras among the Bullininae, it is some shade of red or pink, but its colour is much more intense in some species than in others. In certain minute lacustrine forms, indeed, such as Gyrauius veltfer' it ! Annandale, Rec. Ind. Mus. XIV. p. 112, pl. xi, figs. 7-11 (1918). 358 Records of the Indian Museum. [VoL. XXIV, appears at first sight to be colourless, but even in G. velfer the tentacles of the living animal have a faint pink tinge under a high power of the microscope and if the mollusc be killed suddenly, as with hot corrosive solution, a distinct pink drop can be seen in the region of the heart through the transparent shell. Intensity of colour in the blood is, however, not correlated with size, for the tint is a deep scarlet both in Indoplanorbis exustus, the largest, aud in Intha capitis, the smallest Indian species known to me. It is, perhaps, correlated in some species and to some extent both with external pigmentation of the body and with habitat. G. velifer has as arule very little external pigment and even in pigmented indivi- duals from the Inlé Lake the blood is only a faint pink, though it is deep red in I. capitis from the same habitat ; but in individuals of the former species from canals and swamps, where pigmentation of individuals of G. velifer is more general and as a rule more intense, the blood is slightly pinker, but still much paler than it ever is in the closely allied G. convexiusculus and G. ewphraticus, species that are always pigmented. In both Bullinus and Camptoceras it is bright red. The Planorbidae may, as a matter of convenience, be separ- ated, as already indicated, into two subfamilies: the Bullininae {or Isidorinae) and the Planorbinae. In the former the shell is hardly to be distinguished from that of the Physidae, while in the latter it is disc-shaped or at teast discoidal. ‘These differences in the shell do not seem to be correlated with any important differences in the soft parts, which show considerable generic variation in both subfamilies. Subfamily PLANORBINAE. In dealing with the Planorbinae most European authors include the species in a single genus with many subgenera. These subgenera were founded, almost without exception, on shell- characters only, but subsequent investigations have shown that shell-characters are supported by others in the radulae and soft- parts, and it seems to me preferable to regard the ‘‘ subgenera”’ as true genera. In his invaluable Catalogue of the Planorbidae in the Indian Museum, of which only a part has yet appeared (Rec. Ind. Mus. XXI, 1921), Germain regards Segmentina as distinct from Planorbis, in which he includes as subgenera Gyrazlus, Diplodiscus and Hippeutis, here treated as genera; but in so do- ing he relies solely on conchological evidence. I have found it necessary not only to recognize the one large Indian discoidal Planorbid (Planorbis exustus Deshayes) as representing a distinct genus on anatomical grounds, but also to describe a new genus based both on shell and on anatomy, with a minute Burmese species as genotype. In the structure of the soft parts the Planorbinae show much greater diversity than the allied families. In the genitalia Simroth ! ' Simroth, in Bronn’s 7¥er-Reichs, Mollusca 111, p. 502, figs. (1912). i922.| N. ANNANDALE: Indian Planorbidae. 359 has recognized four distinct types of male organ. All but one of these are found in the Indian species, as well as a fifth. Slightly modifving Simroth’s definitions and adding one of the fifth type, they may be defined as follows :— TypPE I. Penis short, bulbous, asymmetrical, without a peni- al stylet, with an elongate, thick-walled praeputium. Sheath with two retractor muscles. (Indian genera, [m#tha, gen. nov.; ? Pla- norbts Geoffroy.) Typr II. Penis slender, elongate, asymmetrical at the tip, with a comparatively short, thick-walled praeputium, without a penial stylet. Sheath with a pair of ear-like processes above, with a single retractor muscle. (Indian genus, Segmentina Flemming.) Tyee III. Penis cylindrical, symmetrical, without a penial stylet, with a short, thin-walled but well differentiated praeputium and two retractor muscles. (No Indian genus.) TypE IV. Penis cylindrical, but asymmetrical at the tip, with a horny stylet and a praeputium of complex structure. Sheath with asingle retractor muscle. (Indian genera, Gyvaulus Agassiz ; ? Diplodiscus Westerlund.) TyprE V. As in type III, but without differentiated praepu- tium and with the penis very long and sometimes coiled in the sheath (Indian genera, Inudoplanorbis Annandale and Prashad ; Hippeutis Agassiz.) Type V is closely allied to type III but may for convenience be considered distinct. ‘The radulae of the different genera are not so distinct as the genitalia, but afford good characters in some instances. In Indo- planorbis the teeth are relatively large and the whole organ is broad. In Hippeutis the lateral teeth are arranged in pairs. The jaws differ more markedly. In Indoplanorbis and Gyraulus they consist of a comparatively stout but almost linear transverse upper piece and of a pair of slender vertical side- pieces, which in Indop/lanorbis, as my assistant Mr. Sri Navasa Rao has observed, are fragmented ; but in Segmentina (at any rate in soime species) and in /mtha the three pieces are completely broken up into numerous horny teeth as in the Ancylidae. Key to the Indian Genera of Planorbinae. 1. Shell compare ively thick and large (as a rule well over 1 cm. in diameter) ; whorls convex above and below without peripheral keel. No penial stylet. A, Young shell discoidal. Pseudobranch simple. Male organ of type | we . Planorbis. B. Young shell Physa-like. Pseudobranch when expanded ribbon-like with alternate depres- sions and projections, lobed in contraction. Male organ of type V .. Indoplanorbts. I]. Shell small and thin, usually less than 1 cm. in diameter, at least one surface more or less flattened A. Shell with internal ridges of enamel-like sub- stance, of a flattened and truncate conoidal shape. Male organ of type II... . Segmentina. 360 Records of the Indian Museum. {[Vor. XXIV, &. Shell without internal ridges. Male organ not of type IT. 1. Shell flattened and disc-like, often cari- nate, but never excessively so, its econ lunate. Male organ of type a. Whorls not more than four, in- creasing rapidly; the body- whorl much broader than the penultimate ee 7a 6. Whorls more than four, increas— ing gradually ; the body-whorl not much broader than the penultimate aes “s 2. Shell more or less of the form of a flattened conoid, with the aperture cordate. Male organ not of type IV. a. Spire partly exposed on upper surface of shell. Male organ of type V, but with the penis coiled inside the sheath ,., ... Hippeutis. b. Body-whorl completely occlud- ing spire. Male organ of type { La she ... tntha, nov. Gyraulus, Diplodiscus. Genus Planorbis Geoffroy (1776). 1921. Planorbis s.s., Germain, ‘ Catalogue of the Planorbidae in the Indian Museum,” Rec. Jud. Mus. X XI, pp. 619. There is great doubt as to the occurrence of the true Planorbis (taking Helix corneus I. as type-species) in the Indian Enipire. I include it here on the evidence of Clessin’s figure of Planorbis hindu,' but both the provenance and the generic posi- tion of this species are doubtful. It may be a Gyvaulus, and may not be Indian. Genus Indoplanorbis Annand. & Prashad (1920). 1920. J/ndoplanorbis, Annandale and Prashad, Fourn. Ind. Med. Res. VIII, p. 112. 1921. Jndoplanorbis, wd., Rec. Ind. Mus. XXII, p. 537- In our recent account of this genus we failed to observe the retractor muscles of the penis-sheath, and also to recognize the fundamental agreement in structure of the male organ with Simroth’s type III. The muscles are two in number, one situated at the upper end of the sheath, the other a short distance down its side, but relatively higher than in Simroth’s diagram. When not distorted by the presence of spermatophores the sheath is more sausage-shaped than our figure would indicate (op. cit., 1921, p. 579, fig. 14) and the penis can be mucli contracted, but without losing its straight cylindrical form. ‘The only species of the genus with which I am acquainted is Planorbis exustus Deshayes. Germain has discussed the variations and growth of the shell in a masterly manner (of. cit., 192, pp 34-41, figs. 2-16). ! Clessin on Planorbzs in Martini and Chemnitz's Conch. Cab. (ed. uster and Duncker). 1022.] N. ANNANDALE: Indian Planorbidae. 361 Genus Gyraulus Agassiz (1837). 1919. Gyranlus, Annandale & Prashad, Rec. Ind. Mus. XVINI, p. 52. 1921. Gyranlus, Germain, op. cit., p. 8. Most of the smaller Indian Planorbidae are comprised in this genus. ‘The species I have examined are G. ewphraticus Mousson, G. convexiusculus (Hutton), G. labiatus, G. cantori and G. rotula Benson, but Germain also assigns to the subgenus (as he conceives it) G. himalayanus (Hutton). I think he is wrong in assigning cantori to Segmentina, though he follows Benson and other early Indian conchologists in so doing.’ v The type-species of Gyraulus is Planorbis albus Miller, which is widely distributed in the Palaearctic Region. Genus Diplodiscus Westerlund (1897). 1921. Drplodiscus, Germain, op., cit., p. 7. I have not seen any Indian species of this genus, but Germain assigns to it Benson’s Planorbis hyptiocyclos from Ceylon. Accord - ing to Simroth® the type-species (Helix vortex L.) has the male organ of the same type as that of Gyvaulus, viz. P. albus Miller. Genus Hippeutis Agassiz (1837). 1921. /dippeutis, Germain, op. cit., p. 1921. Hippentis, Annandale and Prashad, op. cit., p. 584. Dr. Prashad nas been able to confirm our recent identification of Benson’s P. umbilicalis as belonging to this genus by an examina- tion of the anatomy of a European species. The male organ of the latter resembles that of Indoplanorbis except that the penis is coiled inside the upper part of the penis-sheath. This character is still more strongly marked in H. umbilicalis, in which, however, the praeputium is apparently longer. The radulae of the two species also agree in general characters and in particular in having the lateral teeth arranged in pairs as if twinned. The only Indian species I have seen is Planorbis umbtilicalis Benson. P. sindicus Benson also probably belongs to the genus, hut his P. cacnosus is a Segmentina. Intha, gen. nov. In this genus the body-whorl, though relatively smaller than it is in Hippeutis, completely embraces and occludes the rest of ' See Annandale and Prashad, Rec. Ind. Mus. XXII, p. 583 (1921). 2 Simroth, of. cit., p 503. * The specimens examined belong to a very large and well-developed phase from the mouth of the Var in the south of France. They seem to me, however, to be at least generically identical with typical shells of H. fontanus, the type- species, from England which Mr. Tomlin has been kind enough to give me, My French shells apparently belong to the form called euphaeus Bourg. by Germain in his Mollusques de la France, tom. 11 (1913). 362 Records of the Indian Museum. | Vor. XXIV, the shell in such a way that the spire is entirely coucealed, except in so far as it can be detected by transparency. ‘The shell is very minute and has few whorls, which increases in size rapidly. Those of the spire are cylindrical, but the body-whorl is flattened below and has the form of a flattened conoid slightly truncate above. The outer lip arises in the middle of the upper surface and forms a small lobe at its point of origin. The aperture is large and very oblique but with a cordate outline, There is a well-devel- oped simple callus on the inner lip, but internal ridges are com- pletely absent. The lower surface is narrowly umbilicate. The external surface is practically smooth. The animal is remarkable externally for the large upper and lower lobes into which the mantle is divided. ‘The pseudobranch is poorly developed. ‘The jaw is broken up into many horny teeth as in Segmentina and Ancylus. ‘The radula is very minute but appears to be quite normal. The male organ resembles that of Planorbis, except that the penial bulb is relatively very large. Tvpe species. Intha capitis, sp: nov. Intha capitis, sp. nov. ‘he type-species may be described here very briefly. I hope to discuss it at greater length shortly in a second paper on the Inlé fauna. — Shell minute (2°5*2X1I mm.), colourless, hyaline but rather thick, highly polished, with the upper surface somewhat convex, with about 3 whorls; a minute pinhole on the upper surface at the base of the outer lip, which bears a minute lobe at its point of origin. Aperture very large ; the callus of the inner lip broad and rather opaque, extending outwards on the shell beyond the lip, but not greatly: thickened. Lower surface quite flat, very narrowly umbilicate. External surface with fine curved vertical striae; no spiral sculpture. . Habitat. We-Ho and Inlé valleys (3000-3800 [t.), Southern Shan States, Burma (recent and subfossil). Type-specimen. No. M 11998/2, Zoological Survey of Indva. Segmentina Flemming (1828). 1gig. Segmentina, \nnandale anc Prashad, op. cit., p. 50. 1921. Segmentina, Germain, op. cit. p. ‘The Indian species examined are P. calathus aud P. caenosus Benson and an undescribed species from the Southern Shan States. Benson’s P. trochoideus also undoubtedly belongs to it. In P. caenosus the internal ridges are often poorly developed and con- cealed by the opacity of the shell, but at least traces of them can always be detected on close examination. The type-species is the Palaearctic Planorbis nttidus Muller, to which P. calathus is closely related. 19221. N. ANNANDALE: T[idian Planorbidae. 363 Subfamily BULLININAE (ISIDORINAE). The only living Indian genus is Camptoceras Benson, but in late cretaceous times the gigantic Bullinus (Platyphysa) prinsepii (Hislop) and the same author’s “ /’hysa’’ elongata, for which a new genus will ultimately have to be created in the same sub- family, were dominant forms in the Indian freshwater fauna. Genus Camptoceras Benson (1843). 1910. Camptoceras Annandale & Prashad, Fomrn. As. Soe. Bengal (n.s.) XVI, p. 457. 1919. Camptoceras, tid., ibid.. XVII, p. 27. Four Indian and one Japanese species are now known, and T have another, as yet undescribed, from the Southern Shan States, where it was found subfossil. The external structure of the animal is very like that of Gyraulus except that there is a large anal siphon constructed of a leaf-shaped epipodium which is coiled up spirally to form a funnel each time the animal expands. This has been observed both in the Japanese and in one of the Indian forms. The blood is deep red. Very little is known of the genitalia, but there is no penial stylet or flagellum. The jaws resemble those of Planorbis and the radula is of normal Planorbid type. The pseudobraiich is simple. The type-species is C. tevebva Benson from the United Pro- vinees. Other Indian species are C. austent and C. lineatum Blan- ford from Eastern Bengal, C. subspinosum Annand. & Prashad from Kashmir. C. lineatum has also been found in Manipur, Assam. The genus may be divided into two groups as follows :— CAMPTOCERATA LINEATA. Shell ovate but dissolute, with spiral lines of minute chaetae. Species—C. lineatum and C. subspinosuim. CAMPTOCERATA TEREBRAE. Shell definitely cornucopia-shaped, without chaetae. Species—C. terebra, C. austent, C. hiraset (Japan) and an undescribed subfossil Burmese species. ON A NEW ALYCAEUS FROM THE KHASI HILLS. By Lt.-Col. H. H. Gopwin-AusteEn, F.R.S. Ina tour made last year in the Khasi Hills Mr. Sunder Lal Hora collected a number of shells. These he has sent me together with others which he collected at Amingaon, across the tiver from Gauhati. These shells have been determined. ‘The most interesting specimens he sends me, belong to the genus Alycaeus and were “found under stones and damp leaves.’’ They turn out to be a new species, which I now describe and figure. Although I made the recess quarters of my Survey Party at Cherrapunji for two summers, I never collected near Maosmai cave and it has been left to Mr. Sunder Lal to discover the new species, which will no doubt be found on the same limestone both to the west and east of Cherrapunji in suitable places. I must regret the delay in publication, but I have been so much occupied with other matters that malacological work could not be touched. Remarks on the anatomy of the animal must find a place later on, Alycaeus maosmaiensis, sp. nov. Habitat. Khasi Hills, near Cherrapunji, at the mouth of the Maosmai cave. Shell turbinate, openly umbilicate, small; sculpture: fine costulation on the upper whorls, suddenly stronger and regular at the sutural tube, as far as its base. Colour dull ochra- ceous brown or very pale. Spire conoid, apex blunt. Suture impressed. ‘I'he su- tural tube rather short and large in diameter through- out. Whorls 4, the last slightly swollen midway be- tween the aperture and the tube, this portion smooth. Aperture oblique, circular, Alycaeus maosmaiensts, Sp. nov. a slight angulation above, rounded below. Peristome solid, double, well defined. Columellar margin rounded. Major diameter 4'0, alt. axis 2°0 mm. The species must be plentiful from the number sent to me in spirit. It finds its nearest counterpart in its thickened simple peristome in Alycaecus pachitaensis of the Dafla Hills, and may be regarded as a representative of this form on the Khasi Hills, south of the Brahmaputra. It is more tumid and globose and flatter behind the peristomie. a av foots’ gates in’ ber ue srheralen ean sath. niger) 7R9e ey jens a ath m A LIST OF THE DRAGONFLIES RECORDED FROM THE INDIAN EMPIRE WITH SPECIAL REFER- ENCE TO THE COLLECTION OF THE INDIAN MUSEUM. Patt V. THE SUBFAMILY GOMPHINAF. By F. F. Vamraw, M.A! M.RICLS., L.R.C.P. With an Appendix By F. C. Fraser, Major, [.M.S. CONTENTS. PAGE Introduction... ot ae oh am sacl 307) List of species oe mS or a we 369 Systematic notes, with description of new species “ae TA 372 Appendix... 5 bee oe er wee 4I5 INTRODUCTION. In 1907 Williamson published an account of the Gomphines of Burma and Lower Siam (see literature) in which he gave the first systematic review of the Gomphine fauna of the Oriental Region that has been attempted. Up to the present day no other account of this fauna as a whole has been published. There are very considerable difficulties to be faced in making such a review, difficulties that may be defined as the results of two sets of circumstances. Firstly, the subfamily as a whole shows a remarkably small range of important venational differences and, as it is on venation that systematic writers have so largely depended, this has not unnaturally resulted in a lack of clearness in the efinition of major series and even of genera in the subfamily. It has also resulted in the necessity of using for specific charac- ters those other than venational— characters in many cases con- fined to or drawn from a single sex; hence it has come about that the identification of specimens is at times a laborious matter, indeed sometimes impossible without the examination of type- specimens. Secondly, these insects are usually rare, in collections at any rate, the sexes are not often taken together and specimens are frequently teneral or damaged ; these factors depend probably rather on the habits of the insects than on any real scarcity. At any rate it is for this reason difficult to obtain adequate material in many cases. i} 368 Records of the Indian Museum. [VoL. XXIV, Williamson’s paper, one of the most important contributions to systematic and faunistic Odonatology that has yet appeared, has made it possible for me to deal with the material before me with greater confidence than would have been possible otherwise. The number of species available for this survey is so considerable that I consider myself exceptionally fortunate in having to deal with so rich a supply of material. The sequel will show only too plainly how much I have left undone, and how very much more remains for the collector to do. From want of leisure I have omitted any study of the genital structures of the second abdominal segment of the male, and for the present I have not attempted any account of the larvae in the collection. The Gomphinae are a very clearly defined group, not likely to be confused with any other of the subfamilies of the suborder Anisoptera, either in the larval or in the adult state. The larvae are essentially burrowers, living in mud, silt or sand at the bottom of streams, either sluggish or rapid-flowing. Fraser has given an account of several of the Indian species! and he makes the interesting statement that in the larvae of Macro- gomphus annulatus ‘‘ the syphon-like end of the abdomen projects from the mud and thus permits the easy inspiration of clear water for purposes of respiration.’’ Correlated with this burrowing habit no doubt are other characteristics of Gomphine larvae—small eyes, short limbs with two-jointed tarsi, short thick antennae and telatively long abdomen. The larvae of Ictinus, however, have an almost circular abdomen, with a flattened, disc-like ventral surface, which sug- gests that they live under boulders and rocks in rapidly running water. The adults though probably individually numerous, are not at all gregarious and probably disperse themselves widely over the country, resting high up in trees. Hence they are so usually captured soon after emergence. With regards to literature bearing on the subject ; the follow- ing is a list of the more important works consulted. It is not exhaustive. — E, de Selys Longchamps. 1854. Synopsis des Gomphines. Bull. Acad. Roy. Belg. (i) XXTI. 1857. Monographie des Gomphines. Mém. couron. Soc. Roy. Scr. Liége XI. 1859. Addition au Synopsis desGomphines. Bull. Acad. Roy. Belg. (ii) VII. 1869. Secondes Additions au Synopsis des Gomphines. Bull. Acad. Roy. Belg. (ii) XXVIII. J “l Rec. Ind. Mus. XVI, pp. 461-463, pls. xxxiil, Xxxiv (1919). . 1922. |] F. F. LAm Law: Indian Dragonfites. 369 1873. ‘Troisiémes Additions au Synopsis des Gomphines. Appendices. Bull. Acad. Roy. Belg. (ii) XXXV. 1878. Quatriémes Additions au Synopsis des Gomphines. Rull. Acad. Roy. Belg. (ii) XLVI. 1890. Odonates de Birmanie. Ann. Mus. Civ. Genova XXX. 1894. Causeries Odonatologiques, No. 7. Ann. Soc. Ent. Belg. 1894, pp. 163-181. Calvert, P. P. 1898. Odonata (Dragonflies) from the Indian Ocean and from Kashmir, collected by Dr. W. L.. Abbott. Proc. Acad. Nat. Sci.'Philadelphia 1808, pp. T4I-154. Kirby, W. L. 1890. A Synonymic Catalogue of Neuroptera Odonata or Dragonflies. 1894. Catalogue of the described Neuroptera Odonata of Ceylon with descriptions of new species. Kriiger, L,. 1898. Die Odonaten von Sumatra: II Thiel, Familie Aeschniden, iv, UnterfamilieGomphinae. Ste¢- tin entomol. Zeit. 1898, pp. 290-330. Martin, R. 1904. Liste des Neuroptéres de l’Indo-Chine. Mission Pavie (sep.), pp. 1-18. Ris, F. 1912. Neue Libellen von Formosa, Sudchina, Tonkin und den Philippinen. Supplementa Entomolo- gica, No. t. 1916. H. Sauter’s Formosa-Ausbeute: Odonata. Suf- plementa E-ntomologica, No. 5. Williamson, E. B. 1908. The Dragonflies (Odonata) of Burma and Lower Siam: II Subfamilies Cordulegastrinae, Chloro- gomphinae, and Gomphinae. Proc. U.S. Nat. Mus, XXXIII, pp. 267-317. 1920. A new Gomphine genus from British Guiana with a note on the classification of the subfamily. Occas. Papers Mus. Zool. Univ. Michigan, no. 50. LIst OF THE SPECIES. The fifty or so species of the subfamily I have to list are arranged according to Williamson’s suggested grouping, whichI have adopted as by far the most satisfactory classification so far pub- lished. I have even ventured to elaborate this classification to a 370 Records of the Indian Museum. (VoL. XXIV, limited extent as regards his series Gomphus, which remains still in an unsatisfactory state for the systematist. The characters I have given in my generic definitions are such as I hope will render the reference of species to their appropriate genera not a very difficult task. I confess I have myself not found it easy in many cases. As to terms employed, it is perhaps necessary to explain that as regards the colour pattern of the synthorax an imaginary typical species would have the dorsum (mesepisternites) black with a ‘ dorsal’ yellow stripe or band on either side of the median suture (or mid-dorsal carina) and external to this a juxta-humeral (or more shortly humeral) stripe just internal to the humeral suture. The sides of the synthorax of such a species would be yellow, with a fine black line marking the position of the two lateral sutures, The term meso-thoracic collar explains itself. De Selys uses also the term antehumeral in the same sense as I give here to the word dorsal, referring to the yellow bands placed near the mid-dorsal carina. For terms used in discussing venation I would refer the reader to Tillyard’s book ‘‘ The Biology of Dragonflies’’ or to Needham’s ‘‘Genealogic study of Dragonfly wing-venation’’! or to William- son’s paper already quoted on ‘ The Dragonflies of Burma and Lower Siam.’ For wing-photographs I am indebted, as on other occasions, to Mr. F. W. Campion. I fear that this paper is in danger of being too lengthy already. I will therefore not attempt to deal with the interesting questions of geographical distribution that suggest themselves. They may be postponed for future consideration, when our knowledge of the group is more complete. I have not been able to deal with certain questions of synony- my satisfactorily, especially as concerns some of the species of Icttnus. I hope they will be tackled and solved by field-workers in India. In the following list species marked with an asterisk have not been seen by me. Species of doubtful distinctness are put in brackets. I have throughout adopted the synonymy of Kirby’s ‘ Catalogue.’ Series HAGENIUS. Sieboldius japponcus Selys. Series DIASTATOMMA. Ictinus rapax Ramb. (praecox)* Selys. (mordax)* Selys. angulosus Selys. ,, (atrox)* Selys. Gomphidia T-nigrum Selys. 3) » ? | Proc. U.S. Nat. Mus. XXV1, pp. 703-764 (1903). 1922. | F. F. LaripLaw: Indian Dragonflies. Series EPIGOMPHUS. Macrogomphus annulatus Selys. ” vobustus Selys. ~ montanus Selys. Perissogomphus stevenst, sp. nov. Leptogomphus gestroi* Selys. is inclitus* Selys. (2) maculivertex* Selys. Heliogomphus nietnert (Selys). Microgomphus torquatus (Selys). Series GoMPHUS. Davidius (zallorensis)* Selys. 4 aberrans (Selys). es davidi assamensis Laidlaw. Anormogomphus heteropterus Selys. Gomphus xanthenatus* Williamson. e. personatus Selys. a (?) promelas* Selys. os milgivicus Sp. nov. 93 (?) ceylontcus* Selys. Platygomphus dolabratus Selys. 6 feae* Selys. ‘ Burmagomphus pyramidalis sp. nov. i, vermiculatus* (Martin). sivalikensis sp. nov. Cyclogomphus hypsilon Selys. is heterostylus Selys. - vestculosus* Selys. be (?) minusculus* Selys. Temnogomphus bivitiatus (Selys). Antsogomphus occipitalis Selys. as orites Sp. nov. Onychogomphus grammicus (Ramb). pe lineatus (Selys). od cevastis* (Selys). re bistrigatus* Selys. M-flavum Selys. 5 saunderst Selys. a8 aureus* sp. nov. a3 biforceps Selys. ag acimaces sp. nov. ¥ modestus* Selys. - jrontalis* Selys. a annularis* Selys. . maclachlani* Selys. (?) cercularis* Selys. Heterogomphus smithii* Selys. ceylonicus sp. nov. Ophiogomphus veductus Calvert. 371 372 Records of the Indian Museum. [Vot. XXIV, Types of new species unless euhonuce stated will be deposited with the Zoological Survey of India (Indian Museum, Calcutta). SySTEMATIC NOTES WITH DESCRIPTION OF NEw SPECIES. Series HAGENIUS Williamson. A small series characterized by the possession of a distinct trigonal supplement ; that is to say, the boundary nerve between the two rows of cells which foJlow immediately the triangles of fore and hinder-wings is straightened out to form a single continuous nerve and does not consist, as in other Gomphines, of a number of separate nerves meeting one another in angular fashion. (Or in other words, the cells immediately following the triangles are rectangular.) The distal margin of the triangle posterior to the attachment of this supplement, is distinctly concave. ‘Triangles crossed by a single nerve. Some reduction of the cross-nerves between M,_,and M, in hinder-wings. Legs very long, hinder femora as much as 17°5 mim. long in Steboldius. Head relatively small, thorax robust, wings rather pointed. Distribution : Holarctic and oriental. 6 Genus Sieboldius Selys. Genotype: Sieboldius japponicus Selys. Head very small relatively, abdomen black with yellony rings ; second segment of abdomen shorter than third. Sieboldius japponicus Selys. This fine species is chiefly Malayan in its distribution. It occurs in Lower Siam, and I have entered it on my list as there seems a possibility that it may turn up in Burma. In all probab- ility it does not occur in Japan. Its only near allies are another species of the genus, S. albardae Selys from Pekin and the two species of the closely related genus Hagenius, one, H. brevistylus Selys, from N. America, the other H. gigas Martin, from Tonkin. Kriiger’s species, S. grandis from Sumatra, does not appear to be separable from the Selysian species. Series DiaSTATOMMaA Williamson. (= Legion Lindenia Selys.) Venation dense. The series ‘‘is unique by the presence of a strongly developed sector (branch) of Rs and a usually less well developed sector of M,.’’ ‘Triangles of fore and hinder-wings dissimilar. Outer side of triangle of hinder-wing slightly concave. I,egs short. [922.] F. F. Lawiaw: Indian Dragonflies. 373 This series includes a small number of rather large insects ; it is represented in S. America by a single species of a peculiar genus ; otherwise it is peculiar to the old world, mainly to the Oriéntal and Ethiopean regions. - Genus Ictinus Selys. Genotype: Ictinus rapax (Ramb.). Species examined: I. rapax (Ramb.), I. angulosus Selys. Characters of the series. L,eaf-like expansion on either side of the tergite of the eighth segment of the abdomen. Upper anal appendages of male longer than the tenth segment, parallel to one another and straight or nearly so. Lower appendage much reduced, with a pair of small processes. A genus of large and handsome insects ranging from Africa to the warmer parts of Australia, but most abundantly represented in tropical Asia. Ictinus rapax (Ramb.). Ictinus rapax, Selys, Mon. Gomph. » praecox, Selys, Mon. Gomph. », mordax, Selys, Mon. Gomph. 26,12. Calcutta, Labelled in de Selys’ handwriting ‘ Jctinus vapax, Ramb. Calcutta." (154/6, 387/6, 5441/20). 43 ¢. Calcutta, April, Sept. and Oct. (6235/20, 6331/20, 6333/20). S22. Calcutta (7220/8, 6283/14, 9282/14, 6332/20, 6334/20). 12. Calcutta, 17-ix-19 (1488/8 2). 1Q. Calcutta, Maidan, 27-vii-14 (8289/20). 1d. Museum, Calcutta, 18-vili-16 (4246/H 1). 19. Calcutta, “sticking to railing in Museum Buildings,’ 28-ix-17 (7940/#. 1). 13,12. Murshidabad (1815/10, 1817/10). ia and exuviae. “Larva from hill-stream, Chakradharpur, Chota Nagpur, v-18. Adult emerged in Museum, 8-vi-18." F. H. Gravely (1428/H 2). 238¢. Chota Nagpur, tooo ft., 1-ix-i5. E. d’Abreu. 12. Barkuda I., Chilka Lake, 1-3-Vili-14. 16. Chalakudi, "Gechin State, 14-30-ix-14, F. H. Gravely (8236/20). 19. Trichur, Cochin State, o-800 ft., 4-x-14. F. H. Gravely (8226/20). 16 andexuviae. ‘ Larva from hill-stream, Chakradharpur, Chota Nag- a v-18. Adult emerged in Museum, Calcutta, 8-vi-18.’’ . H. Gravely (1428/H 2). gyi ee from Calcutta have the juxta-humeral stripe complete or interrupted. In the specimens from Cochin this stripe is obsolescent, as is the yellow mark on the metepisternite ; they are also slightly smaller than the more northern specimens. Possibly the S. Indian form may form a recognizable race. There seems a certain amount of individual variation, and in view of this I am strongly inclined to regard praecox Selys, and mordax Selys, as synonymous with the present species. Length of abdomen. | Length of hinder-wing. 3, Calcutta. 48+3'°25 mm. 40°5 mm. 2, Calcutta. 50 mm. 41 mm. 3, Cochin. 45+3 mm. 36 mm. 2, Cochin. 43 mm. 35 mm. 374 Records of the Indian Museum. [Vo1. XXIV. ‘This species seems to be one of the most widely spread, and pethaps one of the commonest of the Indian Gomphines. I have not seen specimens from Assam or Burma. A closely allied species or race, I. fallax Selys, is known from Shanghai. Ictinus angulosus Selys. 16. Dum-Dun, near Calcutta, 4-vi-11 (6371/20). Face entirely yellow save for fine black lines below the ante- clypeus and above the post-clypeus. A fine yellow line on the exterolateral surface of the middle pair of tibiae. The yellow markings, especially on the thorax and on segment 10, have a distinct red tone. Inter-alar spaces largely yellow, with a trans- verse black line between the wing-bases. Otherwise the specimen agrees with the description given of the type. Length of abdomen 54+ 2°25 mm., of hinder-wing, 42 mm. The species seems to me to be very near atvox Selys, of which only the female is known. Both are rare in collections, and I have not seen an example of aivox, nor the female of angulosus. I suspect the two are conspecific. Genus Gomphidia Selys. Genotype: G. T-nigrum Selys. Species examined: G. T-nigrum Selys. A genus of large and handsome insects, very similar in general to Ictunus, but readily distinguished by the absence of anv leaf-like lateral expansions on the eighth abdominal segment. ‘The members of the genus appear to be rarer insects of rather more restricted distribution than are the species of Ictimus. All are peculiar to the Oriental Region, and most of them are found in Malaya. G. T-nigrum is the only species recorded from within the limits of the Indian Empire. ‘The larvae breed in tanks and still waters, thus differing from those of Ictinus which breed in running water (Fraser). Gomphidia T-nigrum Selys. 283,22 @. Poona, May, 1917. H.C. Fraser. Recorded by de Selys as coming from the ‘ North of India.’ -The species, hitherto regarded as rare, is evidently fairly abun- dant near Poona, at any rate seasonally. Length of hinder-wing of #7 38 mm., of abdomen 50 mm.+4 mim. Series Epigomphus (Williamson). Characterized by free triangles, subtriangles and supra-tri- angles, and by the existence of more than two cross-nerves be- tween M,_, and M, in the hinder-wing. These cross-nerves are not widely spaced from one another as in the next series. Fork- ing of M,., and M; usually unsymmetrical. 1922.] F. KF. LArwiaw: Indian Dragonfies. 375 Five oriental genera can be distinguished. They may be very briefly separated as follows :— A. Large insects (h.wv.>36 mm.) ; ninth abdo- minal segment much longer than eighth .. Macrogomphus Selys. B. Size moderate or small (h.w. usually less than 36 mm) : ninth abdominal segment smaller than eighth, \ 7. Basal antenodal nervure of 2nd series present Leptogomphus Selys. /i. Basal antenodal nervure of 2nd series absent a. Sectors of arculus. approximated shortly after their origin. 1. Size moderate (h.w. 25 mm. or more) ; outer side of triangles straight ; upper anal appendages of male lyrate .. Heliogomphus, gen. nov. 2. Size small (h.w. about 20 mm.) ; outer side of triangle broken; upper anal ap- pendages of male chelate... ... Microgomphis Selys. 4. Sectors of arculus not approximated after their origin eh doe .. Perissogomphis, gen. nov. The unsymmetrical character of the forking of M,, and M, (which Williamson, doubtlessly the result of a lapsus calami, writes as M,_; and M,) is much more obvious in Leptogomphus, Helitogomphus and Microgomphus, than in Macrogomphus and in Pertssogomphus. As to the specialization of cross-nerves between M,-3; and M,, in all genera of the series Gomphus that I have examined, or of which I have seen photographs, the forking of M,_-, and M, is preceded in the hinder-wing by a single cross- nerve only. In Perrissogomphus, least typical of oriental Epi- gomphines, this forking usually occurs at the level of the third cross-nerve, occasionally distal to it. The genera of this series are confined so far as is known to the Oriental and Neotropical Regions. Genus Macrogomphus Selys. Genotype: Macrogomphus robustus Selys. Species examined: M. annulatus Selys; M. robustus Selys; M@. montanus Selys; M. decemlineatus Selys; M. quadratus Selys. \ Large insects (h.w. about 40 mm., abdomen still longer) easily recognized by the curious lengthening of the ninth segment of the abdomen in both sexes. Pterostigma unbraced. Forking of M,-, and M, not markedly asymmetrical, more so in fore than in hinder-wing. Cubital space usually with two cross-nerves. Upper anal appendages of males chelate (much as in Microgomphus). The lengthening of the ninth segment of the abdomen is diagnostic and at once distinguishes a Macrogomphus from any known oriental genus. The segment is roughly twice as long as the eighth. Macrogomphus robustus Selys. 2¢d¢. Assam. 19. “ Sibs. '’=Sibsagar (6331/1). Specimen identified by de Selys. 376. Records of the Indian Museum. [Vo.. XXIV, The males are undoubtedly conspecific with the female. The latter is very mature and in bad condition. The head is missing from one of the males, the occiput of the other differs from that of the type male described by de Selys in not possessing a conical tubercle, and to some extent in details of colouring. The different appearance of the occiput may be an individual peculiarity, the colouring pro- bably owes its distinctions to I'1G. 1.—Colour-pattern of synthorax of age and state of preservation. Macrogomphus rvobustus Selys. At any rate I propose to list the males under the Selysian species for the present. Mae. Head. Upper lip black save for a small yellow mark on either side at the base. Ante- and post-clypeus largely brown- ish-yellow margined with black, vertical parts of frons black, its horizontal part brownish-yellow on the synthorax, the juxta-hu- metal stripes are quite obsolete, and the whole of the metepis- ternite is black. The abdomen has segments 3-8 ringed with yellow, the yellow ring on the seventh segment covering the basal half of the segment, that on the other segments about the basal fifth of each. Segments 9 and 10 entirely black. Upper anal appendages yellowish-white darkening apically, the outer branch of each being the stouter, jointed atits apex; the inner branch is slender rather knobbed terminally and not quite so long. Length of abdomen of 751+2 mm., of hinder-wing 40°5 mm.; length of abdomen of $ 47 mm., of hinder-wing 40 mm. Macrogomphus annulatus Selys. 395,322. Poona. F. C. Fraser. Fairly similar to M. robustus. It differs slightly in size. In Colour its markings seem to be of a paler yellow than those of J. vobustus in which (in dried specimens) they are distinctly brown. M. annulatus has the upper lip largely lemon-yellow, its base and a median longitudinal line black. The vertex has a small yellow spot immediately behind the ocelli. The synthorax has a minute yellow spot, the last vestige of the humeral band, just in front of the alar sinuses to the outer side of the dorsal marks. This is present in all the specimens I have seen. The ninth segment of the abdomen has a small basal lateral spot of yellow. The metepisternite is entirely black. For an account of the larva of this species, and for an interesting note on its habits see Fraser. Length of abdomen of » 49+1°8 mm., of hinder-wing 38 mim.; length of abdomen of ? 51 mm., of hinder-wing 39 mm. 1922. ] F. F. Larwwraw: Indian Dragonflies. 377 Macrogomphus montanus Selys. Tic. Calcutta (5255/20). 12. No locality (5921/13). Both in poor condition. The male identified and labelled by de Selys. Readily distinguished from the preceding species by its gene- rally lighter colouration, which appears to mein some respects also to be more primitive. On the synthorax the juxta-humeral stripes are conspicuous, though incomplete below. The mete- pisternite is yellow, outlined by black lines along the lateral sutures. The basal yellow rings on segments 2-8 of the abdo- men occupy about the basal third of the segment. ‘The abdomen of the male specimen —_ DA eee t 7% ; : Tae 1G. 2.—Colour-pattern of synthorax of is missing, that of the female Macrogomphus montanus Selys. lacks segments 7-10. Length of hinder-wing in #7 38 mm., in 41 mm. In addition to these Indian species of Macrogomphus three other species apparently closely related to one another occur in Malaya. These species are M. parailelogramma Burm., M. albar- dae Selys, and M. decemlineatus Selys. Of these species M. parallelogramma has no trace of the juxta- humeral stripe on the synthorax. The metepisternite is black, but with a narrow vestige of yellow banding in its middle. It is recorded from Java and Sumatra. WM. albardae is very closely re- lated (possibly only a local race) and differs from 1. parallelogramma in having the metepisternite entirely black. M. decemlineatus has the dorsal stripes of the synthorax very narrow, juxta-humeral stripes present, complete but.very narrow, mesepisternite yellow, but with black bands on the fateral sutures as broad as the vellow. Metepimerite yellow with well-marked black posterior margin. Lastly, three Malayan species form a distinct section of the genus. This is characterized by the large size of its members, the well-marked curving and complexity of the sectors of the wings, and the specialized colouring of thesynthorax. The species forming this section are M. guadratus Selys, from Borneo and the Malay Peninsula, M. thoracicus McLach., from Sumatra and also from the Malay Peninsula, and M. abnormis Selys, probably from Borneo. In them the dorsum of the synthorax is black with a large squarish yellow spot occupying about. its anterior half. The way in which the species of the first section of the genus ‘ring the changes’ in the colour-pattern of the synthorax is very remarkable. It is parallelled in other genera of Gomphinae, but it seems to me that in other groups of Anisoptera at any rate the specific differences are of some other character. Here they 378 Records of the Indian Museum. DVOL., SOS. J suggest almost some sort of ‘‘ permutation and combination,” and I believe similar conditions may be traced in regard to other features, hinting as it were at the working of some Mendelian sorting out of characters. Genus Leptogomphus Selys (restr.). Genotype: L. semperz Selys. Basal antenodal nervure of 2nd series present. A maximum of two rows of cells between Cu, and the hinder-margin of the fore- wing, only one row of cells between M, and M,, at level of distal end of pterostigma. Proximal angle of triangle of fore-wing as far distant from the arculus as length of proximal side of subtriangle. Sectors of arculus not approximated after their origin. Forking of M,_. and M, unsymmetrical in both pairs of wings. One or two cross-nerves in submedian space (number apparently varying). Pterostigma unbraced. Upper anal appendages of male simple, rather Gomphus-like. Hamuli large, vesicle of penis small or moderate. Ninth segment of abdomen shorter than eighth. Size moderate. Unfortunately the Museum collection contains no example of this genus. I believe that L. gestroi Selys, from Burma, is certainly to be referred to it and also L. inchtus of the same author. The latter, of which a description of the female only is available, is said by de Selys to be very similar to L. sempert. In addition, Ris has described two species of the genus, L. sautert and L. ferforatus, the former from Formosa, the latter from S. China. The type-species, L. sempert Selys, has been recorded from the Philippine Islands, Borneo, and Tonkin. Lastly, from Malaya several species are known, of these L. lansbergi Selys, is recorded from Sumatra and Java, the closely allied L. assimilis Kriiger, is from Sumatra. L. kelantanensis (Laidlaw) is from the Malay State of. that name and L. wilsasnsore Laidlaw, is from Borneo. The position of one or two species referred to the genus is doubtful. L.? maculivertex Selys, from Burma, known from the female only, is possibly a Heliogomphus. L. parvus Kriger, from Sumatra would seem to belong rather to the Gomphus series; at any rate the number of cross-veins between M,_; and M, is reduced as in that series. Genus Heliogomphus, gen. nov. Genotype: L. ntetneri (Selys). Species examined: H. mietneri (Selys). No basal antenodal nerve of 2nd series. A maximum of two rows of cells between Cu, and the hinder margin of the fore-wing. Only one row of cells between M, and M,, at level of distal end of the pterostigma, Proximal angle of triangle of fore-wing as far dis- tant from the arculus as the length of the proximal side of the sub- 1922.] F. F. Larwiaw: Indian Dragonfites. 379 triangle. Sectors of arculus approximated about 1 mm. after their origin. Forking of M,_, and M, unsymmetrical in both pairs of wings. One or two cross-nerves in the submedian space (number apparently varying according to species and sex). ‘Triangle of hinder-wing occasionally crossed. Pterostigma unbraced. Upper pair of anal appendages of male carrying a remarkable apical process, so that the two processes are together somewhat lyre-shaped. Hamuli sinall, vesicle of penis large and prominent. Ninth segment of abdomen shorter than eighth. Size moderate, colouring rather sombre, with (usually) fine, longitudinal, dorsal line on segments 3-7 of the abdomen. Appearance of adult male, more particularly of the abdomen, very like that of Anisogomphus. The following species appear to me to be clearly referable to this genus: H. nietneri Hagen, from Ceylon and Assam (from the latter locality possibly as a geographical race); H. gracilis Kriiger, from Sumatra; H. retvoplexus (Ris), from Tonkin; and H. scorpio (Ris), from Yunnan. Thé generic separation of these species has been clearly fore- shadowed by Ris, and hinted at by Williamson and Kriiger. Though I cannot claim to have had the opportunity of making as detailed a study of the species of this genus and of Leptogomphus, s. vesiy., as any of these writers, it seems to me that a faunistic paper of the character of this on which I am now engaged affords a suitable opportunity for defining this new genus. Heliogomphus nietneri (Selys). 16,12. Tura, Garo Hills, Assam, 1500 ft. S. Kemp. (7977/H1). These specimens are clearly examples of the same form as that described in the ‘ Odonates de Birmanie’ by de Selys, which was regarded by him as being conspecific with Hagen’s species from Ceylon. Not having any examples from the latter locality with which to compare the Assa- mese material I follow de Selys in leaving the specimens under this name, though I believe it !'6. 3.—Anal appendages of Helio- not unlikely that they may prove eae ass, nietneri Selys, seen «| ew e. to belong to a distinct race or even to a distinct species. From Hagen’s description they differ chiefly in having the two lateral black bands on the synthorax along the lines of the lateral sutures, the posterior band not being ‘ presque terminale.’ The male has no lateral yellow markings on any of the segments beyond the fourth, whilst the pterostigma is brownish-black. In general appearance, and especially in the colouring and shape of the abdomen, the male of H. nictneri bears a curiously close resem- blance to that of Anisogomphus occipitalis Selys. 380 Records of the Indian Museum. [Vor SOx, The female of H. nietneri has not been described. In this sex _the colouring of the head is, as in the male, largely black, with a pair of yellow spots at the base of the upper lip, the bases of the mandibles marked with yellow, and a yellow line across the frons from eye to eye. The frons, as in Macrogomphus, is flattened so that there is not the usual distinction between a vertical and a hoti- zontal part. Behind the ocelli the vertex bears a pair of small conical tubercles. The occiput is depressed. The prothorax has the yellow markings more extensive than in the male. The synthorax has a small superior juxta-humeral spot not found in the mature male, in addition to the dorsal band, other- wise as in the male. Abdomen. First segment entirely yellow, the remainder black, marked with yellow as follows :—Dorsum of 2-8 with longitudinal band, broader on the second segment than on the others, and from 2-7 occupying the whole length of the segment, on 8 only the basal third. A broad lateral band, including the well-developed auricles, along the second segment, a narrower and partly inter- rupted band on the third. Basal triangular spots on 4-7. Anal appendages whitish-yellow. Tenth segment extremely reduced (?shrivelled) not one-quarter the length of the ninth, Vulvar scale small, about one-third the length of the ninth segment, shaped like a truncated triangle, only slightly bifid. Length of hinder-wing of 728 mm., of abdomen 29+1 mm. ; length of hinder-wing of 9 31 mm., of abdomen 32 mm. a Genus Microgomphus Selys. Genotype: M. chelifer Selys. Species examined: M. torquatus (Selys). No basal antenodal nervure of 2ndseries. Only one row of cells between Cu, and the hinder margin of the fore-wing. Only one row of cells between M, and M,, at level of distal end of ptero- stigma. Proximal angle of triangle of fore-wing as far distant from the arculus as the length of the proximal side of the sub- triangle. Sectors of arculus approximated after their origin. Forking of M,_, and M, unsymmetrical in fore and hinder-wings. Pterostigma unbraced or feebly braced. Upper anal appendages of male chelate. Vesicle of penis large, ninth segment of abdo- men shorter than eighth. Size small, colouring brilliant. In this definition there is little to separate Microgomphus from Heliogomphus, and I think the two genera are related. Microgomphus differs mainly in its smaller size, more reduced vena- tion, chelate upper anal appendages of the male, and in its richer colouring. | Microgomphus torquatus (Selys). Cyclogomphus torquatus, Selys, Mon. Gomph. 753,499. Poona. KF. C. Fraser. 1922.] F. F. LamwLaw: Indian Dragonflies. 381 MALE. Head.—Lower lip yellow. Upper lip yellow with black line at base produced asa triangular mark in the middle line. Genae and bases of mandibles yellow. Ante- and post-clypeus yellow, the latter with a pair of small black marks where it joins the frons, This has a black band running from one eye to the other just above the clypeus, but is otherwise yellow, its crest not inflated. Vertex and occiput black, the latter with a slightly concave mar- gin, without hairs. Prothorax black, its anterior margin lined with yellow, and with a small lateral spot on the posterior margin, and a median spot of the same colour. Synthorax largely yellow, the dorsum black marked with yel- low as follows: a mesothoracic collar, not interrupted, and in the middle line sending a short projection upwards along the mid-dor- sal carina; a pair of oblique dorsal markings, pointed above and below, on either side of the mid-dorsal carina, not reaching to the mesothoracic ‘collar’ nor to the ante-alar sinus; above and to the outside of these markings a pair of small triangular spots, with the apices directed anteriorly ; and lastly a minute yellow spot between the ante-alar sinuses. Laterally a black band less Fic. hae —Colour-pattern of Microgomphus torquatus (Selys) 3, semi-diagrammatic. than half a millimetre across, runs obliquely downwards from be- low the front pair of wings to the base of the posterior coxa. Legs black, the postero-lateral surfaces of the first, and the proximal half of the posterior surface of the third pair marked with vellow. The anterior surfaces of the femora are thickly studded with fine, irregularly arranged spines. Abdomen biack. Dorsum and sides of first segment with yellow marks. Second segment yellow on the sides. and with a yellow mark on the dorsum, not reaching the apex of the segment, pointed distally and widened at its middle rectangularly. The third segment has a yellow mark on either side at its base, extend- ing for rather more than a third of its length distally, and a dorsal triangle of the same colour, its base resting on the basal margin of the segment. In the middle of the distal half of the segment is a narrow oblong-oval yellow spot dorsally, and on either side of this a somewhat similar lateral mark. Segments 4-8 have each a basal ring of yellow narrowest on 4-5, lengthening on 7-8, more produced laterally than dorsally, especially on the two latter seg- ments. In addition 4 has a small, longitudinal oval spot, lying mid-dorsally, and occupying about the middle third of the seg- 382 Records of the Indian Museum. [Vor. XXIV, ment, The ninth has a rounded lateral spot, and the tenth is en- tirely black. The auricle is yellow and the genital structures of the second segment are black. The abdomen is slender, the seg- ments 7-9 rather wider and deeper than the rest. The upper anal appendages, which are pale yellow with dark- ened apices are a little longer than the 9th segment of the abdomen. Each is chelate; the outer branch is much the stouter and dimin- ishes gradually to its apex. The inner branch is slender, cylin- drical, and is widest at about its middle. The two branches separate at about the middle of the total length of the appendage and run at an angle of about 60° to one another. The outer branch is quite straight, the inner 0) = (om wcvo . . 5 Ki. obi ua ‘Selyen a, is turned a little upwards at its seen from above. apex. ‘The lower appendage is about two-thirds of the length of the upper pair. It is brownish black in colour, has nearly parallel sides, and at its apex divides into two short branches which run a little outwards and upwards. The vesicle of the penis is large and conspicuous, almost as striking a feature of the male as in Cyclogomphus. FEMALE. Colouring of the head, prothorax and synthorax as in the male. The abdominal colouring differs chiefly in that the two dorsal marks on the third segment are united, that on the fourth segment relatively longer, and in the fact that ‘lateral spots lying at about the middle of the length of the segment occur on segments 4 and 5, that on 5 being very small. The anal ap- pendages are minute, yellow in colour, and the abdomen is re- eularly cylindrical from the end of the third segment to the apex. Length of abdomen of @ 23°5 mm., of hinder-wing 21 mm. ; length of abdomen of 2 25°5 mm., of hinder-wing 22°75 mm. It is curious that in this the smallest of the old-world Epigom- phines the anal appendages of the males should bear a strong resem- blance to those of Macrogomphus, a genus which includes the largest species of the series. Microgomphus was founded by de Selys to include a species, M. chelifer from Malaya, which until Major Fraser’s discovery of the male of the present species remained the only known example of the genus. The finding of a second species in Peninsular India under climatic conditions which must be very different from those obtaining in the countries inhabited by M. chelifer is interesting and a little surprising. The larva of M. torquatus has been figured and described by Major Fraser under the name of Cyclogomphus minusculus Selys. The Malay species differs from M. torguatus in having the dorsal band of the synthorax confluent at its upper extremity with 1y22.] F. F. LAtpLAw: Indian Dragonflies. 383 the juxta-humeral band, which is apparently not reduced as in M. torquatus to a mere superior spot. The yellow rings on segments 4-7 of the abdomen are smaller, and the branches of each of the upper anal appendages of the male are nearly parallel to one another. The venation differs but slightly from that of M. cheléfer, the chief distinction being apparently that the present species has its venation a very little denser than in M. chelifer. 12 - 10| TO - 12-13, Nodal indicator pes eae l1G. 5.—Venation of Microgomphius torquatus (Sclys) 3. (Photo by F. W. Campion.) Cu., cubital space. Mc., area between M143 and My. Mf., Fork of My-2 and M3. Perissoguinphus,! gen. nov. Genotype: P. stevensi, sp. nov. An Epigomphine genus, without a basal antenodal nervure of 2nd series. A maximum of at least three rows of cells between Cu, and the hinder margin of the fore-wing. Only one row of cells between M, andM,, at level of distal end of pterostigma. Proximal angle of triangle of fore-wing not so far distant from the arculus as length of proximal side of sub-triangle. Forking of M,_, and M, unsymmetrical in both fore and hinder-wings. Two (sometimes three) cross-netves in the submedian space of all four wings. Triangle of hinder-wing of female frequently crossed by a nerve running parallel to long axis of wing. Pterostigma braced. Anal l neotaoos=redundant, excessive. 384 Records of the Indzan Museum. [VoL. XXIV, appendages of male rather like those of Gomphus,s. sty. Ninthseg- ment of abdomen shorter than eighth. Legs short, hindermost femora when adpressed barely reaching the end of the first ab- dominal segment; armed with a number of very small tubercles on the ventral surface, not arranged in definite rows. Vulvar scale of female small, deeply cleft. I have found it difficult to make up my mind as to the proper place of this genus. The character of the cross-nerves between M,_3; and M, shows a certain amount of variability even on the different wings of a single specimen. In some cases there is distinct evidence of a spacing out of the cross-nerves, sug- gesting an approach to the condition found in the Gomphus series ; in other wings there is no evidence of anything of the sort. So that it would, it seems to me, be reasonable to regard this form as a primitive member of the Gomphus series, or as an Epigomphine showing a tendency to specialization in the same direction as the Gomphus series. The presence of two or even sometimes three cross-nerves in the submedian space, the frequent existence of a cross-nerve in the triangle of the hinder-wing of the female are characters that incline me to think the position of the genus should be rather with the Epigomphines. The colouring and so far as I know the other characters of Perissogomphus, do not lend any assistance in settling the question. On the whole I be- lieve the genus is an annectant one and that it will ultimately prove to occupy a position not far from the base of both series. Another oriental genus, Merogomphus Martin, known from a single species from Tonkin, seems to me to be, like Perissogom- phus, intermediate between the series Epigomphus and the series Gomphus. It has fore-wings similar to those of the former series, hinder-wings with the differentiation of cross-veins characterizing the latter. The pterostigma is well braced. The anal appendages of the male closely resemble those of Heliogomphus. The posi- tion of the genus must be regarded as doubtful at present. Perissogomphus stevensi, sp. nov. 13,4 9 9. Gopaldhara, Darjiling District. 229 (fragmentary). Darjiling (cc/1060-cc/1061). Mare. Head. Lower lip black, upper lip, ante- and post- clypeus brownish-black, Bases of mandibles yellow. Frons pale yellow. Occiput and vertex black, the former with a straight margin, edged with long hairs. Prothorax black, with a fine yellow spot on its posterior margin, in the middle line. Synthorax yellowish-brown, with a large M-shaped black mark anteriorly. ‘The outer lines of the M run down along the humeral suture on either side from the ante-alar sinuses. The middle part is made up of a median black line running down from the sinuses on either side of the mid-dorsal carina, but not quite reaching the meso thoracic ridge. The carina itself is yellowish-brown. 1922.] EF. F. Lamriaw: Indian Dragonflies. 385 In addition, the second lateral suture is marked with a narrow black band. Legs black, the first pair of femora yellowish on their ventral surfaces. Abdomen black, marked with yel- low. . The first segment is yellow laterally, and has a dorsal band of the same colour, much contracted at its middle. The second is like- wise yellow on the sides, with yel- low oreillet, and it has a trilobed dorsal band of yellow extending the whole length of the segment. Seg- Fic. 6.—Colour-pattern of syntho- ments 3-7 have a narrow longi- rax of Pevissogomphus steven- tudinal band of yellow dorsally, ‘si, sp. nov., Q. commencing at the base of each seg- ment, but not reaching to the apex except in the case of the third segment. In each case this band is broadest at the base of the segment and narrows rapidly apically to a very fine line. It ' is broadest on 7. In addition 3 has a small basal lateral mark. The dorsum of segments 8 and g is entirely black, but these seg- ments and also 7 have each a complete lateral yellow band. Segment 10 is yellow with black margins. The abdomen is slen- der, the distal part of 7, and the whole of 8 and 9 rather in- flated. The anal appendages are pale yellow, and in general like those of Gomphus, s. sty. The upper pair are simple, almost regularly conical, somewhat acuminate, a little flattened below, and very slightly upturned at the apices. They are a little longer than segment to of the abdomen. The lower appendage is about two-thirds of the length of the upper pair. Its terminal half is deeply cleft so as to form two‘’branches, which are a little divaricate and upturned at their apices. FEMALE. The female differs from the male in - colouring chiefly as fol- Fic. 7.—Anal appendages of Perissogomphus lows:—On the head the stevensi, sp. nov., 3. upper lip is dark brown margined with black. The ante- and post-clypeus are dark-brown, the vertex and occiput are also very dark brown. Further, the hinder margin of the latter carries at either extremity a small spine. The prothorax has the posterior margin entirely yellow. On. the synthorax the only noticeable difference lies in the fact that the black line situate on either side of the mid-dorsal carina is truncate below, and not rounded as in the male. The abdominal markings are very much like those of the 386 Records of the Indian Museum. [VoL. XXIV, other sex. The lateral markings on the first three segments form a more definite band, reaching to the middle of the third segment. In general the colouring is not quite so bright as in the male, but the female specimens are all more mature and in worse preserva- tion. Length of abdomen of » 35+1°5 mm. (approx.), of hinder- wing 33 mm.; length of abdomen of @ 41 mm., of hinder-wing 37 mm. Antenodal cross-nerves on fore-wing 17, post-nodals 13; antenodal cross nerves on hinder-wing 14, post-nodals 14. Two of the six females examined have the triangles of the hinder-wing free. One female has three cross-nerves in the sub- rr da eeetnu & Ww SSaety OY ars iG. 8.—Venation of Perissogomphus stevenst, sp. nov. (Photo by Fk. W. Campion.) median space of the fore-wing the other specimens have two apiece. Series Gomenus (Williamson). : ** Associated with uncrossed triangles, supra-triangles, and sub-triangles is a reduction and specialization in the cross-veins between M,-; and M,. This last character is unique in the Anisoptera.”” The forking of M,_, and M; is approximately symmetrical. The difficulty of defining genera in a satisfactory way, and of indicating their proper systematic position in grouping them with their allies, obvious enough in all the larger series of Gom- phines, is especially so in the case of this large and dominant series which contains, as Williamson remarks, not less than thirty genera and two hundred and fifty species. 1922. | F. F. Lamwriaw: Indian Dragonflies. 387 Clear-cut venational characters are not to be found, though this is possibly to some extent due to the fact that the very large gents Gomphus is something of a ‘‘ dump’”’ for a certain number of species whose exact position is at present doubtful. On the other hand, anyone who examines a considerable number of specimens belonging to several of the genera of the series can scarcely fail to notice that whereas the specialization of cross- veins between M,_; and M,, used by Williamson as the most important character for the definition of the series, is tolerably constant in the hinder-wing, in the case of the fore-wing a certain tange of specific and even of individual variation may occur. Further, I believe that in certain genera, probably in those which are to be regarded as the most highly evolved of the series, this variability in the case of the fore-wing is by no means marked; it seems to me even to be absent in some cases, so that in such zenera the specialization of the fore- and hinder-wings is about equal. Unfortunately, owing to the occurrence of tolerably nu- merous cases of individual variations of the fore-wing, the degree of specialization exhibited by it cannot as yet be used as a generic character. Of characters available for grouping genera which appear to form natural assemblies in the series the most evident are those supplied by the anal appendages of the males. These characters, backed to some extent by colour peculiarities, and also here and there by features of the venation, give a grouping which I am per- suaded is tolerably natural, and of some practical value. I would again insist that in presenting such an arrangement here I am drawing largely on suggestions and hints made by de Selys, by Williamson and by Ris, and that I am dealing entirely with Indian or Oriental forms. I believe the tribes or sections of the series defined below to have approximately equal value. Davidius would seem venationally to be the least specialized, and perhaps the most easily defined. Onychogomphus and Heterogomphus ate the most advanced. In them itis rare or exceptional to find an individual in which the specialization of the cross-veins between M,-3; and M, of the fore-wing is not as fully developed as in the hinder-wing. I propose then to group the bulk of the oriental members of the series in the following five “ groups,” leaving out of account one or two genera whose position is to my mind doubtful, or which are not sufficiently known to me. From what I have seen of the British Museum collection I should be inclined to say that there are still several genera awaiting recognition, especially amongst the material from Tonkin,a country apparently very rich in Gomphinae. Groups: , IDAVIDIUS, CYCLOGOMPHUS, GOMPHUS, ONYCHOGOMPHUS, HETEROGOMPHUS. 388 Records of the Indian Museum. [Voy. XXIV, Group: DaviIpDIUvs. Number of cross-nerves between M,_; and M, not so constant, and not always showing the same amount of specialization in the fore-wings as in the hinder-wings. Sectors of arculus widely sep- arated at their origin, constantly though slightly divergent, and scarcely curved ‘Triangle of hinder-wing with costal margin about twice as long as the basal margin, its outer margin distinctly angled; often with a cross-nerve. Cu, and Cu, in hinder-wing distinctly divergent. Pterostigma braced. Basal antenodal nervure of 2nd series occasionally present. Hindermost femora when adpressed reaching almost to distal end of second abdominal segment (in D. davidi). Dorsum of segments 3-7 of abdomen entirely black, or marked with a fine longitudinal line of yellow. Upper anal appendages of male each with large ventral process. Lower appendage cleft at the apex, its branches not divaricated. Genus Davidius Selys. Genotype: D. zallorensis Selys (?).! Species studied: D. aberrvans Selys; D. davidi assam- ensis Laidlaw. ; Characters of the tribe. Distribution: Himalaya, Assam, Tonkin, China, Japan, Man- huria. Davidius aberrans Selys. 1 9. Binyar, Kumaon, 7700 ft., 24-v-1912. A. LD. Imms, Kor. Zool. Coll. This specimen has the triangles of both hinder-wings crossed by a single nerve, and in addition the triangle of the right forewing is traversed by a nerve lying parallel to the long axis of the wing. The triangles of the fore-wings in this specimen are relatively more elong- ated than appears to be the case in KG, 9.—Colour-pattern of syn- the species figured by Ris. thorax of Davidius aberrans Length of abdomen 34 mm., of Selys, 9. ; ; ‘” hinder-wing 28 mm. Davidius davidi assamensis Laidlaw. 16,299. Gopaldhara Darjiling District 9-v-14. H. Stevens. ‘Dorsum of synthorax entirely black save for the small meso- thoracic collar and a small extension from this running up the mid-dorsal carina for about the first half of its length. ' Should D. zallovensis and D. aberrvans prove to be conspecific the latter name must take precedence, and the genotype would then be D. aberrans (Scelys). 1922.] F. F. LarwLaw : Indian Dragonfites. 389. In the male the last five segments of the abdomen are entirely black. The seventh segment shows a peculiarity that I have not seen remarked on. A little beyond the middle of the segment the ventral border of the tergite is produced on either side to form a small tubercle-like process armed with stout hook-like spines directed backward. The ventral margins of the tergite of the eighth segment carry a series of spines, about the middle of the segment, rather larger than is usual in that position, and the sternite of that segment carries, close after its base, a small blunt projec- tion. The significance of these structures is unknown to me. The anal appendages of the male resemble rather closely those of D. cumiculus Ris, a Japanese species. Of the species referred to the genus, D. nanus Selys would appear to have a doubtful right to the position. As Ris has’ pointed out the anal appendages of the be r ghey Wipuachiuedtanade : : "1G. 10. ~ Apex of abdomen of Davidius davidi assa- male are quite unlike mensis \.aidlaw, @. - x, process of edge of tergite those of other spe- of the seventh segment. cies where known. This species is Japanese. D. fruhstorferi Martin, is in its venation clearly a member of the tribe Davidius. But the reduced anal area of the hinder-wing, and the characters of the male anal appendages suggest to me that it may require transference to a distinct (and new) genus, as. already hinted by Ris. For the rest D. lunatus (Bartenef), D. aberrans Selys, and D. zallorensis, have a yellow mark on the occiput; the first of these is Manchurian, the two others (probably conspecific) are from the Himalayas. D. davidt Selys, from S. China and Assam, D. ater Selvs, and D. cuniculus Ris, both from Japan, have a black occiput. Lastly D. bicornutus, also from Japan, is larger than the other species, is known only from a female specimen and is of doubtful generic position. Length of abdomen of o 31 mm., of hinder-wing 26°5 mm. ; length of abdomen of 9 28 mm., of hinder-wing 28°5 mm. [ Group CycLOGOMPHUS. It is impossible to give a really satisfactory definition of this tribe, apart from sexual characters. None the less I am per- suaded that it is a true phylogenetic entity. The tribe is characterized by the anal appendage of the male. The upper pair are small, nearly parallel to one another, and often brightly coloured. They are often provided with a strong ventral projection. The lower appendage has its branches. 390 Records of the Indian Museum. |Vor. XXIV, widely divaricated, often these are longer than the upper append- ages. In many cases the hinder femora are long, and have the spines of their apical half at least much longer than is usually the case with the other tribes of the series so far as I know. The occiput of the female is generally much reduced. A longi- tudinal dorsal stripe is often present on the last four segments of the abdomen. Lastly, the differentiation of the cross-nerves be- tween M,_,; and M, does not seem so firmly fixed a character as in the three remaining tribes of series. : In Anisogomphus in fact, some ‘individuals show scarcely more differentiation than does Leptogomphus in this respect, others in my short series show marked differentiation in the hinder-wing, less or none in the fore-wing. I have not enough material for a statistical study, but I believe that in Cyclogomphus about one individual in four shows lack of differentiation in the fore-wings. An examination of the males of Podogomphus praetorius and of Notogomphus sp. in the British Museuin has convinced me that these two genera at least, in addition to the Oriental forms here enumerated, belong to this tribe. Genus Cyclogomphus Selys. Genotype: C. hypsilon Selys. Species examined: C. hypsilon Selys; C. hetcrostylus Selys. A genus of rather small Gomphines, characterized by the pre- sence of a hasal antenodal nervure of 2nd series on all four wings (absent from one hinder-wing of a single female specimen of C. hypsilon only). The costal side of the triangle of the hinder-wing is much longer than the basal side (almost twice as long in C. hypsv- lon), only one cross-nerve in the submedian space of both wings. Costal nerve with a fine yellow line. The pterostigma is relatively long, more than one-quarter the length of the distance between the nodus and the distal end of the pterostigma. The hinder pair of femora when adpressed reach to the end of the proximal third of the second segment of the abdomen. They are armed with two, rows of spines on the ventral surface, and there are in addition a few scattered tubercles of minute size neat the base. The vulvar scale is small. Of the five species which have from time to time been re- ferred to the genus, C. ¢orquatus Selys is undoubtedly a Microgom- phus. The small species, C. minusculus Selys, known from a single specimen, a female taken at an elevation of between 4000 and 6000 ft. near Tenasserim, will in all probability prove to belong to a different genus. | C. vesiculosus Selys, described from an imperfect male, has been recorded from Poona by Major Fraser, but is unknown to me; it appears to differ, according to de Selys’ account, but little from C. hypsilon, chiefly in its smaller size. 1922.] F. F. Lawiaw: Indian Dragonflies. 391 Cyclogomphus hypsilon Selys. 120 6,929. Poona, Kartraj Lake, Aug., Sept., 1918. The pterostigma is uniformly brown, enclosed between dark brown nerves, whilst in the next species (C. heterostylus, Selys) it is dark in the centre, and distinctly paler for its outer third. Fur- thermore, the light markings on the dorsum of the terminal seg- ments of the present species are more extensive. Colouring varies much with age and state of preservation. os Re ed 1G. t1.—Abdomen of Cyclogomphus hypsilon Selys, 3, seen from above. Length of abdomen of # 28 mm., of hinder-wing 25 mm.; length of abdomen of 2 30 mm., of hinder-wings 26 mm. Cyclogomphus heterostylus Selys. 24 4,19. Poona, 1g-ix-19. 19. Darjiling, 4-vii-18. 12, St. Thomas Mount, Madras, 9-iii-18 (all from Major T°. C. Fraser). I cannot find any characters by which to distinguish the Dar- jiling specimen from the female from Poona. The males agree precisely with the figure given for this species in the monograph. Length of abdomen of ¢ 28°5 mm., of hinder-wing 26 mm. ; length of abdomen of 2 30 mm., of hinder-wing 26°5 mm. Genus Anisogomphus Selys. Genotype: A. occtpitalts (Selys). Species examined: A. occipitalis (Selys); A. orites, sp. nov. A genus of medium-sized Gomphines, distinguished from Cyclo- gomphus and Temnogomphus by the absence of a basal antenodal nervure of 2nd series. The costal nerve is black, and in general the colouring, especially in adult specimens, is rich but more sombre than in allied genera. Pterostigma relatively shorter than in Cyclogomphus. The inferior anal appendage of the male carries two stout rather widely divaricated branches, the superior appendages are coloured (in Indian species), each carries a stout black ventral process. The occiput of the female is much reduced. ‘The vulvar scale has a length of about two-thirds of the ninth segment, its apical quarter is cleft. Hindermost femora when adpressed reach to the middle of the second segment of the abdomen. Their armature consists of two rows of spines on the ventral surface, rather irregularly spaced, aud varying a little in length, those placed more distally being on the whole the longer. 392 Records of the Indian Museum. [Vou XXIV, Of the species referred to the genus in Kirby’s catalogue, the African A. praetorius Selys, has been since referred to an allied genus Podogomphus by its author. I have ventured to remove A. bivittatus Selys, to a separate genus. The two species from N.-E. Asia, A. maacki Selys, and A. M-flavum Selys, are unknown to me save from the descrip- tion. Both appear to belong to the genus. Lastly A. nietneri Selys is referred by me to the new genus Heliogomphus. Anisogomphus occipitalis (Selys). 233,229 (in poor condition). Gopaldhara, Darjiling district. H. Stevens. 13. Darjiling, 1-3000 ft., May, 1912. Lotd Carmichael’s collection (cc/1167). 3,19. Turzum Tea Estate, Darjiling, Mar. 1920. O. Lindgren. 13, Darjiling Dists. (1409/12). 13, 19. WDarjiling Dists. (3424/H1). Closely allied to the following species, from which it is readily distinguished by the following characters :—- (i) All the specimens of A. occrpitalis before me have two cross-nerves in the cubital space of the fore-wing. (ii) The upper end of the dorsal stripe of the synthorax is not confluent with the upper end of the antehumeral stripe. (iii) ‘The black bands on the lateral sutures of the synthorax are much broader than in A. orites. (iv) The anal appendages of male A. occipitalis also show marked differences. The ventral process of the upper pair is straight and dagger like, and this process, owing to the fact that the appendages are more closely ap- proximated than is the case in orites, is invisible from above. Further the branches of the lower appendage are straighter and rather more slender, with a more Fic. 12.—Colour pattern of syntho- pointed apex. ee of niece cml occipitalis . : elys a In young specimens of this ee pat species the antehumeral band of the synthorax is complete, in older specimens it is in part obliterated, and comes to consist of a dorsal spot, rather widely separated from a more anteriorly placed linear remnant. 1a Wir i ad Length of abdomen of #32 mm., hoe ee oven of hinder-wing 30 mm.; length _ of We abdomen of ? 33 mm., of hinder-wing 32 mm. 9g °o 1922.] F. F. LaiwiAw. Indian Dragonflies. 393 Anisogomphus orites,' sp. nov. 1¢,1Q. Shillong, Sept., 1919. T. B. Fletcher. (Type and allotype.) ‘* © Flying along hedge away from water.” Mare. Head black save for a small spot on either side of the upper lip, the bases of the mandibles, and a broad band across the top of the frons from eye to eye. The markings are lemon yellow. © Prothorax black, marked with yellow on its hinder lobe. Synthorax with black dorsum, yellow mesothoracic collar interrupted in the middle line. Dorsal bands meeting the collar anteriorly, running close up to the ante-alar sinuses, narrow, regular, of a pale greenish-vellow- colour, confluent at their upper extremities with a small rounded spot on either side, the vestige of the upper end of the ante-humeral band, which is otherwise obso- lete save for a fine line lying close against the humeral suture at fig. 13.—Colour-pattern of syntho- about the middle of its length. rax, Anisogomphus orites, sp. The black of the dorsum passes nov., o. on either side just beyond the humeral suture. The sides of the thorax are greenish-yellow in colour with a narrow black band on the first and second lateral sutures. The meso- and meta-notum are whitish-yellow. Legs black; the first pair of femora have a greenish-yellow kand on the extero-lateral surface. Abdomen black, of a particularly rich, almost velvety quality. Yhe first segment has a transverse mark of yellowish-white dorsally. Segments 2-7 have each of them a dorsal, longitudinal Ce Fic. 14.—Last six segments of abdomen of Anisogomphus ovites, sp. nov., @. band of the same colour, extending nearly.the whole length of the segment but not quite reaching its apex. That on the second segment is trilobed; on segments 4-7 the bands are narrow and linear, that on the seventh being the most conspicuous. ‘The sides of the first and second segments, including the oreillets, are greenish-yellow, also a small basal lateral spot on the third. Upper anal appendages small, white in colour and aculeate. Each carries on its ventral side a large, black process. This process appears in side view almost triangular, attached above by its apex, widening ventrally. Each is somewhat incurved so as to be visible l dpertas = mountaineer. 304 ltecords of the Indian Museum. [Vor. XXIV, between the upper part of the appendages when the abdomen is viewed directly from above. The lower distal angle of each of these processes ends in a small projecting point. Branches of lower appendage black, stout, curved a little upwards and out- wards, with rounded apices. FEMALE. Colouring in general very similar to that of the male. The third abdominal segment has a longitudinal lateral band of colour, and each of the segments 4-7 has a small lateral basal spot on either side. As in the male the upper anal appendages are white. Length of abdomen of @ 31°5 mm., hinder-wing 30 mm.; length of abdomen of 2 34 mm., of hinder-wing 31°5 mm. The venation of this species is in general very similar to that of its congener A. occipitalis. Neither of the two specimens be- fore me, however, have the additional cross-nerve in the cubital space of the fore-wing. ‘Though one cannot assert with any degree of certainty that the presence or absence of this additional vein can be used to distinguish the two species, what evidence exists points in this direction. 1 have accordingly omitted the presence of an additional nerve in this area from the generic characters. Genus Temnogomphus,! nov. Genotype: T. bivittatus (Selys). Species examined: T. bivittatus (Selys) ¢?. A genus containing a single species of moderate size. It is characterized by the possession of a basal antenodal nervure of the ‘and series, aud by the colour-line on the costal nerve which separate it from the closely allied Amusogomphus. From Cyclogomphus it differs in the relatively : : ae short pterostigma, and in having the atin a meet Or ERO triangle of the hinder-wing not elong- Q , outline. ated. It is also larger than in the species of Cyclogomphus. The occiput in the female is reduced to a marginal line. The hindermost femora when adpressed reach to the end of the first segment of the abdomen. The armature consists of relatively long spines, much as in,Anzsogomphus. Temnogomphus bivittatus (Selys). 2@. Mumaon, v-tgt1. A. 1D. Imms, For. Zool. Coll. The species is readily identified by the yellow face with two black bands running across it. The lower of these is the “ ante- clypeus ’’ the upper a black band across the lower part of the irons. The occiput in the female is reduced to a narrow yeilow margin between the eyes, fringed with long yellow hairs. The prothorax has its anterior and posterior margins yellow. On the syntho- ' reavw=cut oy separate. 1922. ] F. F. Larw.aw: Indian Dragonflies. 395 trax the mesothoracic collar is narrowly interrupted in the middie line, the dorsal stripes join it on either side, and the juxta-humeral stripes are complete and rather broad. ‘The sides are yellow with ery narrow black lines along the sutures. The abdomen has a longitudinal median band of whitish- vellow on all the segments except the tenth, complete save for the narrow black sutural rings, bounded on either side by a brownish- black band; this encloses on either side a lateral band of yellow, broad and complete on the first and second segments, divided into two parts on segments 3-7 by the black mark on the transverse carina of each of those segments. On 8-9 the yellow lateral band is complete, and on the tenth segment it passes entirely to the ventral side, whilst the dorsum of the segment is black. Anal appendages black. The femora are yellow, with brownish-black spines and a dorsal brown band on each, the tibiae are black. On the back of the occiput is a curious rounded tubercle of a bright yellow colour. Length of hinder-wing 33 mm., of abdomen 36 mm. The two males in the British Museum labelled as belonging to this species, one of them so labelled by de Selys himself, are evidently rather examples of A. occipitalis, as was recognized by de Selys in his latest account of the species (Causertes Odonatologi- gues No. 7, 1894). In this account he remarks further that A. bivitiatus has a basal antenodal nervure of the 2nd series. This. combined with certain other features emboldens me to remove hivittatus from the genus Anisogomphus, and to erect for it the new genus described above. Group GOMPHUS. Fore-wing and hinder-wiug showing about equal specialization in regard to the arrangement of cross-nerves between M,_, and M,. Sectors of arculus well separated as a rule, approximately parallel and well curved. Cu, and Cu, parallel or only slightly divergent on hinder-wing. M,and Cu, slightly divergent (or parallel in Burmagomphus) at the level of the nodus, in the fore-wing. Pterostigma braced, basal antenodal nervure of the 2nd series. usually absent. Legs variable, in Gomphus, s. str., rather long. The male often has a conspicuous yellow mark on the dorsum of the ninth abdominal segment. Upper anal appendages of male shaped much like the same structures in the Libellulinae. Lower appendage with its branches divergent, usually a little shorter than the upper pair, and roughly parallel with them. A. Hinder-wing of male with its anal border rounded. Small insects with reduced venation. Body sandy colour ae ee us 8B. Anal border of hinder-wing of male sharply angled. 1. Four rows of cells at least between Cug and hinder margin of hinder-wing _... . Gonphus (each. 2. Usually three rows of cells between Cug and hinder margin of hinder-wing. Anormogomphus Selys. 396 Records of the Indian Museum. [VoL. XXIV, a. Apical segments of male abdomen much dilated ie ag ... Platygomphus Selys. 6. Apical segments of male abdomen only slightly dilated. M4 and Cui parallel to beyond level of nodus of fore-wing ... Burmagomphus \Villiamson. Genus Anormogomphus Selys. Genotype: Anormogomphus heteropterus Selys. Species examined: A. heteropterus Selys. This genus may ultimately come to be placed in a distinct tribe, but there seems no reason to doubt that it is derived from ancestors belonging definitely to the Gomphine series. It is of course unique amongst the Gomphinae by reason of the rounded hinder wings of the males. Its geographical distribution is of greatinterest. With Vande- via (an ally of Ictinus) and with the Libelluline Selyszothemts, it is characteristic of the desert areas of §.-W. Asia, but has apparent- ly a more restricted distribution than either of the other genera. Anormogomphus heteropterus Selys. 1 6. f.akore. H. T. Pease. The small genus of which only two species are known, seems to be confined to the arid regions of W. Asia. The specimen before me agrees precisely with the account given of this species in the monograph. The other described species occurs in Mesopotamia and Turkestan, and is rather larger than 4. hetero- pterus. ‘This species is A. kivitshenkow Bart., of which I have seen a series in the British Museum, Genus Gomphus Leach. Genotype: G. vulgatissimus (Linn.). Species examined: G. pervsonatus Sely$; G. nilgivicus, sp. nov. My knowledge of this genus is very limited. A brief compari- son of the genotype with the species inhabiting India (and some of their more immediate Eastern allies) gives one the impression that characters for a subdivision of the genus may be forthcoming in the future. I confine myself to describing a species from the Nilgiri Hills that appears to be new, and to very short notes on the other Indian species. Gomphus personatus Selys. 1 6 (fraginentary). Assam. labelled by de Selys ‘‘ Gomphus promelas ” 3.” (5442/20). Differs from the type in haying only a superior spot, in- stead of a narrow vestige of the antehumeral stripe; in other respects what there is of the specimen agrees with the account given by de Selys;; and I have no doubt but that the specimen is 1922. | F. F. Larptaw: Indian Dragonflies. 397 correctly referred to the present species. The structure of the second pair of genital hamules is very similar to that found in G. nilgivicus, but the hamule is more curved forward at its apex and not so prominent as in that species. Gomphus nilgiricus, sp. nov. 1 *. Nilgiris, go00-4000 ft. IK. C. Fraser. Head. Lower lip yellow, upper lip brown, bases of mandi- bles and anteclypeus of the same colour, post-clypeus black. Vertical part of frons black, horizontal part lemon yellow. Ver- tex and occiput black, the latter with its hinder margin a little raised medially. Prothorax black, the middle lobe with a large lateral spot of yellow on either side, and a small mid-dorsal, paired, yellow spot ; the hinder margin likewise yellow. Synthorax black dorsally, with a pair of rather broad yellow dorsal stripes, meeting the yellow mesothoracic collar, which is widely interrupted in the middle line. Sides yellow, with black FG. 16. —Colour- -pattern of Gomphus nilgiricus, sp. nov., ¢, semi-diagrammatic, bands along the lateral sutures, the metepisternite between these bands is darkened so that at first sight the bands appear to coalesce. Legs black, rather long; hindermost femora 7°5 mm. Abdomen slender, cylindrical, the eighth segment a little wider and deeper than the rest. Black, marked with yellow as fol- lows :—Sides of first and second segments, including the oreillet, which are, however, heavily margined with black; lateral ventral mark on third segment ; dorsum of first segment, the yellow mark widening apically; a longitudinal band on the dorsum of the second segment, extending the whole length of the segment, widest basally ; a dorsal line on the third segment, wide at its base, nar- rowing rapidly, extending as a fine line almost to the apex of the segment. Segments 4-7 with fine paired spots lying basally on the dorsum of each. Distal half of 9 yellow. Anal appendages black ; upper pair about as long as the tenth segment, curved slightly ventralwards, ending in a fine upturned point. Branches of lower apperidage about equal to them in length and equally divaricated. Genital structures of second abdominal segment, black in \ 398 Records of the Indian Museum. [Vor XXIV, colour. Anterior pair of hamules, small, simple and partly con- cealed between the second pair. These are large and very con- spicuous, rather trigger-shaped, their apices lying ventrally to the triangular vesicle. Length of hind-wing 35 mm., of abdomen 39+1°5 mm. Venation that of a typical Gomphus, pterostigma braced ; BiB. I2-13/15-12 nodal indicator —— - 3} 75. . I2- 9| 10-13 G. xanthenatus Williamson, Burma. ‘Traces of the juxta- humeral band present, a broad black band on the two lateral sutures. Abdomen of 745 mm., hinder-wing 39 mm. G. personatus Selys, Assam. ‘Traces of juxta-humeral band present, narrow black bands on lateral sutures, the first incomplete above. Abdomen of #42 mm., hinder-wing 37 mm. G.? promelas Selys, Madras. Dorsal stripes of synthorax ‘“ presque confluentes ” with mesothoracic collar, no juxta-humer- al stripe. Pterostigma unbraced. Exact position doubtful. Fe- male only known. Abdomen of 2 42 mmm., hinder-wing 38 mm. G.? ceylonicus Selys, Ceylon. Narrow isolated dorsal stripes, juxta-humeral band represented by a superior spot. Pterostigma unbraced. Female only known. Abdomen of 2 41 mm., hinde-- wing 39 mm. ‘ Genus Platygomphus Selys. Genotype: P. dolabratus Selys. Species examined: P. dolabsaius Selys. Differs from oriental species of Gomphus in having the anal margin of the hinder-wing of the male only slightly excavated, so that the anal angle of the hinder-wing is not so bold as in that genus. In this respect it shows some approach to Anormogomphus. The number of rows of cells between Cu, and the margin of the hinder-wing is three only. The apex of the abdomen of the male is more dilated than in the oriental species of Gomphus, and the colouring is rather of the ‘ xerophilous’ type; i.e. more brown and vellow than is usual in the allied forms where yellow and black prevails. Platygomphus dolabratus Selys. Io Upper lip entirely yellow. Vertex with a diamond-shaped yellow spot between the posterior ocelli, occiput yellow glabrous. Dorsum of synthorax black, with dorsal stripe reduced to a large triangular spot on either side of the mid-dorsal carina. Juxta- humeral band remarkably broad, complete, mesothoracic collar not interrupted in middle line. Sides yellow, short dorsal black 1922.] F. F. Larpnaw: Indian Dragonflies. 399 mark at upper end of first lateral suture, very narrow complete black line on second suture. Abdomen with segments 1-6 black with longitudinal brown bands dorsally ; 7-10 mainly brown. Hindermost femora yellow. Length of abdomen 38+ 1°25 mm., of hinder-wing 30 mm. The species P. feae Selys, from Burma, is apparently closely allied. It differs in having no black on the clypeus, the yellow mark on the vertex is absent, and the short black band on the upper end of the first lateral suture of the synthorax is carried vertically downwards to join the black line of the second suture, forming with it a Y-shaped mark. Genus Burmagomphus Williamson. Genotype: B. vermiculatus Williamson? nec Martin. Species examined: B. pyramidalis, sp. nov.; B. sivals- kensis, sp. nov.; B. sp. Rather smal! species (h.w. about 27 mm.). ‘There is marked parallelism between M, and Cu, in the fore-wing to beyond the level of the nodus. As a rule only three rows of cells between Cu, and hinder margin of hind-wing. (Individually there are sometimes four rows). Generally only a single row of cells between M, and M,, at distal end of pterostigma. The genus probably contains a considerable number of small species which may shade off into Gomphus on the one hand and into Platygomphus on the other. Ris has pointed out that the parallelism of M, and Cu, isa feature common with Onychogomphus the Javanese species, B. jacobseni Ris, has the apex of the ab- domen very like that of a Platygomphus, though the venation is definitely that of a Burmagomphus. The genotype is the species described by Williamson under the name Burmagomphus vermiculatus (Martin). But his speci- mens are very probably not conspecific with Martin’s, in which case the genotype would be without a name. B. Pyramidalis, sp. nov., is closely allied to both Martin’s and Williamson’s species ; as in the latter the oblique yellow stripe of the dorsum of the synthorax is formed by the fusion of the upper part of the dorsal, with the lower part of the juxta-humeral stripe, a peculiarity which seems to occur in Martin’s species as well. The three may well be geographical races of a single species. On the other hand the Bornean form which I described under the name B. vermiculatus insularis must rank as a distinct species, it appears to be related to a form from the Nilgiris represented in the collection before me by a single very immature example. B. jacobsent Ris is quite distinct and as stated above in some respects approaches Platygomphus. Lastly the new species, B. sivalikensis, from Dehra Dun is again quite distinct, and possesses a complete "juxta- -humeral stripe. In size and general ‘proportions it resembles B. pyramidalis. 400 Records of the Indian Museum. [Voy. XXIV, Burmagomphus pyramidalis, sp. nov. 8 63,10 92%. Poona, Aug., Sept., 1919. F.C. Fraser. 1 od. Nilgiri Hills, June, 1917. F.C. Fraser. 1 9. Gopaldhara, Darjiling district, 1913. H. Stevens. Mate. Head. Lower lip yellow, black at its margins with lateral lobes greenish-yellow. Upper lip greenish-yellow, with narrow, black, anterior margin, and transverse black hasal line pro- longed forward in the middle line, but not meeting the black of the margin. Bases of mandibles yellow, genae black. Ante-clypeus black, post-clypeus black, with a median spot of greenish- yellow, and a pair of lateralspots of the same colour. Frons yellow, margined with black. Occiput and vertex black. The posterior margin of the occiput is slightly concave. Prothorax black, with a greenish-yellow spot on either side. Synthorax. ‘There is a broad, complete mesothoracic collar of yellow. Above this is a pyramid of black, at the apex of which lies a small median yellow spot against the ante-alar sinuses. The pyramid is outlined on either side by an oblique yellow band which runs from just below the median yellow spot, without ac- tually touching it, to the base of the second pair of legs. This Fic. 17.—Colour pattern of Burmagomphus pyramidalis, sp. nov., 3, : semi-diagrammatic. band is abruptly narrowed above from without inward, and ends dorsally in a fine point. To the outer side of its upper end and close below the lateral part of the ante-alar sinus is a small yellow triangle. ‘The black colouring extends from the outer margin ot the yellow band to just beyond the humeral suture. Laterally the synthorax is yellow, with a complete narrow band of black along the second lateral suture, and with a black mark along the lower part of the first lateral suture, extending upwards a very little beyond the level of the spiracle. A short black line descends vertically from the base of the first wing towards the black band of the second lateral suture, but does not meet it. Legs black, coxae of first and third pairs, and ventral surface of first pairs of femurs marked with yellowish-white. Abdomen black. First segment with a basal dorsal mark of yellow, and a lateral mark of the same colour. Second segment with a basal ring of yellow, a mid-dorsal triangle of the same colour, and the sides of the segment, including the oreillet, like- wise yellow: an apical ring of black. Segments 3-8 with a basal ring of yellow, extending for about one-sixth of the length of the segment on 3, about one-fifth on segment 7, very narrow on 8, 1922.] F. F. Larriaw: Indian Dragonflies. 401 where, however, it is produced at the sides for about one-half the length of the segment, and followed by a yellow apical mark. The ninth segment has a large apical dorsal brownish-yellow mark. Segment Io and the anal appendages are entirely black. The up- per pair are about as long as the tenthsegment, slightly divaricate, acuminate at their apices, the outer margin elbowed at its middle, with a minute ventral tooth at the level of the elbow. The lower appendage has two branches, slightly longer than the upper pair of appendages, more widely divaricated, and sharply upturned apically. The genital structures of the second abdominal segment are black in colour. The first hamules are small and inconspicuous, the apices pointed and hooked backwards. ‘The second pair are large, rather oval, and each carries a prominent forwardly directed hook near its apex. FEMALE. Colouring very similar to that of the male. Itdiffers chiefly in having the basal yellow ring of the third segment of the ab- domen broader, and in the presence of a lateral distal yellow mark on the same segment, whilst the lateral distal mark on the eighth segment is reduced and the dorsal yellow mark on the ninth is much smaller than in the male. Length of abdomen of #29+0°75 mm., of hinder-wing 23°5 mm,; length of abdomen of 2 33 mm., of hinder-wing 27 mm. Burmagomphus sivalikensis, sp. nov. 1 3. Dehra Dun, 4-ii-19. F.C. Fraser. Head. Lower lip pale yellow. Upper lip yellow with black margins, and a median black line dividing the yellow into two large lateral spots. Basesof mandibles yellow. Anteclypeus brownisb- black, post-clypeus the same, with a median and pair of lateral yellow spots. Frons yellow, with a black line extending from eye to eye across its vertical part. Vertex black, occiput yellow, margined with black, its posterior margin gently convex, carrying a fringe of long brownish-black hairs. Prothorax black, its anterior margin, a pair of lateral spots on the hinder margin and a minute median spot on the same margin yellow. Synthorax. Dorsum black marked with yellow as follows :—a broad mesothoracic collar, finely divided by a median black line; a pair of dorsal lines separated below from the mesothoracic collar and above from the ante-alar sinuses, and outside these on either side, a longer irregular juxta-humeral band, constricted a little below its apex, running down and continuing on to the mesinfraepisternite. This band is margined externally by a black line which runs on either side of the humeral suture. The sides of the synthorax are yellow ; a narrow black band, of about the same width as that lying along the humeral suture, runs along the position of each of the lateral sutures on either side. j Legs black, the inner surface of the first pair of femurs yellow- ish-white. 402 Records of the Indian Museum. [Vou. XXIV, Abdomen black, marked with yellow as follows :—Sides of first and second segments, including the oreillet; dorsum of first seg- ment, and a longitudinal band on dorsum of second segment, the latter mark narrowing apically; a basal ring, occupying nearly the first third of segment three, but contracted dorsally in the middle line, in addition a distal lateral yellow mark ; a similar basal ring on segments 4-7, but relatively smaller, and occupying only about one-quarter of each segment. Eighth segment entirely black, ninth with a dorsal triangle of orange-yellow, its base resting on the apical margin of the segment, its apex not quite touching the base of the segment. ‘lerminal segment black. Anal appendages black, very similar to those of B. pyramidalis. The upper pair about as long as segment 10, acuminate, very slightly upturned at the apex, elbowed onthe outer margin. Lower appendage a shade longer, its branches rather more divaricated than are the upper appendages, upturned at the apices. Genital structures of second segment of abdomen similar to those of B. pyramidalis, but the first pair of hamules are relatively larger and more prominent, the second pair not so oval but sloping obliquely backwards and ending in a forwardly directed hook. Length of abdomen of 733+1 mm., of hinder-wing 25°5 mm. Group ONYCHOGOMPHUS. Fore-wing and hinder-wing showing equal specialization in regard to the cross-nerves between M,_, and M,. Sectors of arculus as in preceding tribe. Cu, and Cu, nearly parallel as far as wing margin. Pterostigma braced. M, and Cu, parallel to level of nodus in fore-wing. A, separated from A, by two rows of cells from immediately below the subtriangle, whereas in the preceding tribe there is usually a single cell between them at their origin ; further A, lies nearer the wing base in this tribe than in Gomphus. Legs very short, hindermost femora scarcely reaching beyond the end of the synthorax when adpressed. Colour pattern not afford- ing any definite characters for separation of the tribe. Upper anal appendages of male longer than tenth segment of abdomen, often equal in length to the ninth and tenth segments together ; parallel or converging apically, arcuate. Genus Onychogomphus, Genotype: O. forcipatus (Linn.). Species examined: [0. forcipatus (Linn.)]; O. lineatius Selys; O. grammicus, Selys; O. saundersi Selys; O. aureus, sp.nov. ; O. M-flavum Selys; O. biforceds Selys: O. acinaces, sp. nov. Characters of the tribe. Distribution: Warmer parts of Europe, Africa, tropical Asia as far as the Celebes. . I group the species of this large and rather difficult genus in 1922. | F. F. Latwiaw: Indian Dragonjfites. 403 several sections, which arrangement will, I hope, facilitate to some extent their identification. Inthis grouping I do not follow exactly de Selys’ classification given in the monograph. Having to deal only with oriental forms I have adopted an order which bears more directly on them. As usually is the case with collections of Gomphines the males are more abundant and more readily recognizable than the females. Hence in this list, as elsewhere in the paper, I am forced to rely largely on male characters, a practice not by any means theoretically ideal but, at the worst, useful in practice. SECTION I, grammucus. Frons and front of head entirely yellow. Dorsal stripes of syn- thorax not meeting mesothoracic collar, antehumeral stripes com- plete. Apex of abdomen of male not dilated. Upper anal ap- pendages about twice as long as branches of lower appendage. A “xerophilous’ section. Onychogomphus grammicus Selys. 1d. Agra. S. Hankin. Front of head including the whole of the frons, entirely yellow. Vertex and occiput largely yellow. Synthorax with dorsal yellow stripes not meeting the mesothoracic collar, ante- humeral stripes complete. Femora and tibiae largely marked with yellow. Abdomen almost cylindrical; the eighth segment shows a slight increase in depth compared with the others. The first seg- ment is almost entirely yellow. The second has a trilobed longi- tudinal dorsal mark of yellow enclosed between lateral bands of brownish-black, Segments 3-6 are whitish-yellow as far as the transverse carina and beyond this black, but the black is marked with a prominent yellow spot dorsally, lying longitudinally. Segments 7-10 entirely sandy yellow. Anal appendages of the same colour, upper pair nearly twice as long as branches of lower appendage; very like those of O. lineatus, but flattened and truncate apically. Length of abdomen 32°5+3 mm., of hinder-wing 29 mm. Closely allied to O. flexuosus Schneider, Asia Minor. SECTION II, lineatus. , Front of head and frons entirely yellow save for a small transverse black line on the crest of the frons. Dorsal stripes of synthorax not meeting mesothoracic collar, antehumeral stripes complete. Apex of abdomen of male but little dilated from side to side, segments 8 and 9 carry leaf-like expansions. Upper anal appendages closely apposed, parallel and strongly decurved apically, lower appendage about one-quarter of the length of upper pair. To this group belong the species O. genei Selys, and O. pumilis {Ramb.). Further the two species of Mesogomphus that I have seen 404 Records of the Indian Museum. [Voy. XXIV, in the British Museum, M. coguatus Selys, and M. hageni Selys, both appear to be derived from it. Like the last this section con- tains ‘ xerophilous’ forms. Onychogomphus lineatus Selys. 283,299. Poona, May-Aug., 1918. F.C. Fraser. 284,299. Chota Nagpur, July, August, 1915. E. d’Abreu. 16 (with larval exuviae). Peradeniya, Ceylon, 1700 ft., 7-x-17. N. Annandale, Probably a very common species. The presence or absence of a row of small denticles on the posterior margin of the occiput seems to be an individual character ; two of the females are with- out the denticles, and one of the males (from Poona) has three or four denticles unsymmetrically arranged in this position, the other males lacking them altogether. It would be interesting to study this character in series from different localities, and also to determine how it is inherited. Young males have the last four segments of the abdomen entirely yellow, with increasing age a black band develops on the dorsum of these segments, and at the same time the colouring of other parts of the body deepens, making the fully adult insect differ considerably in appearance from younger specimens. O. lineaius seems to range over the Indian Peninsular but the limits of its distribution to the north and east are not known. Possibly allied to this species are O. reinwardti Selys, from Java and O. capitatus Martin, from Celebes. O. lineatus is the only Gomphus I know of that shows a develop- ment of the second abdominal segment resembling the ‘‘ genital lobes ”’ of the Libellulidae. SEcTION III, geometricus. Black markings on upper lip and head; frons largely black. Dorsal stripes of synthorax meeting mesothoracic collar to form a pair of inverted 7-shaped marks, Antehumeral stripes interrupted or represented by superior spot only. Segments 7 (distal half), § and 9 of abdomen dilated from side to side, the dilatation increas- ing regularly to the distal end of 8. These segments black above. Upper pair of anal appendages orange or yellow in colour, well separated at origin, regularly tapering, cylindrical, a little down- curved. Lower appendage with its branches approximately equal in length to the upper pair ; closely applied to one another for their whole length, moderately upturned. Specially characteristic of this group is the combination of inverted 7-shaped marks with a vestigial antehumeral band on the dorsum of the synthorax. Only one other Indian species of Onychogomphus, O. annularis Selys, shows this feature in addition to the species referred to the section. And as that species was described from imperfect specimens, and remains very imperfectly known, it is quite possible that it too may ultimately find its place here. 1922.] F. F. Larpraw: Indian Dragonflies. 405 O. capitatus Martin, from the Celebes shows the same combi- nation in the colour-pattern of the synthorax. But according to its describer the sides of segments 8 and 9 of the abdomen carry leaf- like expansions, which at once distinguish the male from any of the geometricus group. Onychogomphus saundersi Selys. 1 &. Burma. E. B. Williamson. 1 9. Toungo, Burma. E. B. William- son. 1 9 “Type.” Brit. Mus. The specimens from Burma have not been identified by Mr, Williamson as this species but I think there can be no doubt o{ their identity. I have not seen the specimen recorded by Williamson from Burma as belonging here. His figure of the anal appendages leave little room for doubt but that the specimen belongs to the geometricus section, but the colouring of the sides of the synthorax is not that of sawndersi, which has distinct bands along the lateral sutures, with the metepisternite yellow, not black. HenceI think his specimen must belong to a species distinct from O. saundersi Selys. Length of hinder-wing of 7 30 mm., of @ 32°5 mm. Onychogomphus aureus, sp. nov. 366. Tura, Garo Hills, Assam, 1200-1500 ft. S. Kemp. (7978/H1). Close allied to O. geometricus Selys. Head. Upper lip black with a pair of large yellow spots. Ante-clypeus and post-clypeus black, with a pair of lateral spots on the latter. Frons black with a broad yellow band across the horizontal part just behind its crest. Vertex and occiput black, the former with a pair of tubercle-like projections immediately behind the ocelli. Prothorax black with small paired median spot anteriorly and larger lateral spots; its hinder lobe yellow. Synthorax with dorsum black, yellow mesothoracic collar inter- rupted in the middle by the black of the mid-dorsal carina. Dor- sal stripes meeting it on either side. Antehumeral stripe inter- rupted, represented by a superior spot and a vestigial line separated from the spot, along the humeral suture. Laterally the synthorax is golden-yellow, with a black line along the second lateral suture. Legs black, but posterior femora largely brown deepening to black apically. : Abdomen with segments I and 2 yellow, with the yellow of the dorsum enclosed between longitudinal black bands. On the first segment this dorsal yellow widens apically. On the second it is trilobed diminishing from before backwards. Oreillets yellow margined finely with black. Segments 3-6 golden brown with apical black rings, which are progressively larger, on the third seg- ment occupying about the distal quarter of the segment, on the 4 406 Records of the Indian Museum. [Voy. XXIV, sixth the distal third. In addition these segments have each an obscure darker longitudinal mark occupying about the middle of the segment. The basal two-thirds of the seventh segment are yellow, the apical third is black. Segments 8, 9 black, 10 golden brown margined apically with black, and with black marking on either side of the middle line dorsally. Anal appendages yellow, very like those figured for O. geometri- cusin the monograph. Upper pair as long as.segments 9g and 10 of abdomen, curved downwards, cylindrical and tapering to a point. Branches of lower pair closely approximated, rather abruptly curved upwards at the commencement of their apical third, trun- cate at their apices, rather shorter than the upper pair ; and each carrying a dorsal tubercle at the end of the basal third of their length. . Length of abdomen of & 35+3 mm., of hinder-wing 30°5 mm. The species of this section of the genus may be distinguished as below. a. Narrow complete black band on second lateral suture only, position of first lateral suture unmarked. Segments 3-6 with apical two-thirds yellow or brown, distal third black .., ine ... ©. aureus, sp. nov. Garo Hills, Assam. B. Black bands marking the position of both lateral sutures of synthorax. 1. Lower anal appendage of male black, upper pair tipped with black... eh O. saundersi Selys. Burma, Sumatra (Malay Peninsula ?). 2. Anal appendages of male entirely orange ... O. geometricus Selys. 3 Java. C. Black bands marking the position of lateral su- tures confluent over the metepisternite ... O. saundersi Williamson (nec Selys?). Burma. There is a male specimen of a species of this group in the British Museum from Tonkin; I have not been able to identify it ; but on casual inspection it would appear to be O. geometricus. I have also had the opportunity of examining a specimen from Su- matra identified by N. H. Campion, and of discussing it with him. Section IV, biforceps. Colouring largely black, dorsal bands of synthorax confluent or not with mesothoracic collar, antehumeral stripe present or absent. Dilatation of apical end of abdomen begins abruptly at base of eighth segment, the apical half of the seventh being scarcely enlarged. The dilata- tion attains its maxi- mum at the middle of the eighth segment. Lower anal appendage 1G. 18.—Apex of abdomen of Onychogomphus of male longer than biforceps, Selys, 3, seen from above. upper pair ; its branches separated at their ori- gin by a circular space. ‘The branches project beyond the end otf 1922.] I’. F. LAmLaw: Indian Dragonflies. 407 the upper pair, or these latter may be sharply hooked downwards to lie along the dorsal surface of the lower appendage. The insects contained in this section would appear from their colouring to be forest-haunting and shade-loving forms. Onychogomphus biforceps Selys. 1 ¢. Pashok, Darjiling Distr., May, 1915. (3409/H1). This splendid species remarkable for the specialization of the anal appendages of the male, is known only from examples of that sex. Owing no doubt to a misreading of de Selys’ account, William- son in his key to the species of Onychogom- phus puts biforceps amongst the species in which the dorsal stripe joins the mesothoracic j4;. 19.—Anal appendages of Onychogomphus collar; this is not the biforceps, Selys, 3, side view. case, the dorsal stripe being isolated. The still larger Tonkinese species, O. camelus Martin, repre- sented in the British Museum collection, has the anal appendages in the male almost identical in shape with those of O. biforceps but entirely black. The dimensions of the Indian Museum specimen are as fol- lows, length of abdomen 35+4 mm., of hinder-wing 35 mm. Onychogomphus acinaces,' sp. nov. 1 @. Castle Rock, N. Kanara Dist. S. Kemp. (4392/H1). Very distinct from other described species. Head. Lower lip black, yellow at its base. Upper lip black with a pair of transverse, greenish-yellow spots. Ante-clypeus vellow, post-clypeus black. Frons, vertical part black, horizon- tal part yellow, the yellow divided into two distinct parts by a median triangle of black. Vertex and occiput SL black. Prothorax entirely black. Synthorax, dorsum black, mesothoracic collar of greenish- yellow, interrupted by black in the middle line. A dorsal band of greenish-yellow on either side, not joining the collar, narrowing to a point below, reaching the humeral suture above. WNo trace of ante- humeral band. Sides of synthorax largely yellow, but the met- episternite entirely black, the black extending to just behind the second lateral suture, so that laterally the synthorax has two yellow areas widely separated from one another by a broad belt of black. ! Acinaces = a scimitar. 408 Records of the Indian Museum. [Vou XXIV, Ventral surface black, meso and metanotum marked with lemon yellow. Legs entirely black. Abdomen longer than wings, swollen at its base and again enlarged from segments 7-9. Black in colour. Segments 1-2 with mid-dorsal yellow bands. 3-6 with basal rings of yellow occupy- ing about the basal sixth of each; these rings are contracted in Skee, sess MEAN AIO ¥ the middle line dor- aci1aces, Eee See ae sally. Basal half of seventh segment lemon yellow. First and second segments, including the oreillets, marked with yellow. Eighth segment with basal, lateral spots of the same colour. Anal appendages, upper pair as long as ninth and tenth abdo- minal segments together, yellow, darker at apex, internally, and ventrally, tapering, parallel, curving downwards in their distal half. Lower appendage one-quarter as long again, its outer four-fifths bifid. The branches are parallel, closely approximated, and curve upwards for their distal half. Seen in profile the lower appendage is rather scimitar-shaped, and projects well beyond the end of the upper pair. They are entirely black. The venation is that characteristic of the genus. The bases of both pairs of wings are distinctly tinged with dark brown, espe- cially in the sub-costal space, where the colouring extends as far as the first cross-nerve. Antenodals age ; post nodals eee Ir | 11 Ta, |e Length of abdomen 33 mm. +3°75 mm. (lower appendages) of hinder-wing 30 mm. Apparently related to O. btforceps Selys, but with the anal appendages less strongly curved. It differs also in the distinct colouring of the wing-bases, rather unusual in Gomphines, and generally in the great extent of black colouring on the thorax. The males of the species of this section may be distinguished as follows :— A. Dorsal stripe confluent with mesothoracic collar ; antehumeral stripe absent ; anal append- ages black, segment 8 of abdomen with a pair of dorsal prominences _... as ... O, camelus Martin. Tonkin. B. Worsal stripes of synthorax isolated. 1. Antehumeral stripecomplete ... .. O. biforceps Selys. Tonkin; India. 2. Antehumeral stripesabsent ,,, ... O. acinaces, sp. nov. W. India. The small species, O. modestus Selys, the hinder-wing of which has a length of 23 mm., would appear to have some right to be included in the present group on account of its generally dark colouration. The synthoracic colour-pattern is in fact very 1922.] F, F. Latwwiaw: Indian Dragonflies. 409 similar to that of O. acimaces. De Selys regarded it as*allied closley to O. saunderst. I have not seen a specimen but am in- clined to refer it here, rather than to the geometricus group. Onychogomphus, sp. 19. Garo Hills, Assam, 1200-1500 ft., June-July, 1917. S. Kemp, (7979/H1). This specimen is apparently quite distinct from any yet des- cribed, its exact position is doubtful, but it seems to me possibly allied to O. biforceps hence I describe it briefly here. Head: upper lip black, base of mandibles yellow, ante-clypeus yellow. Post-clypeus black, frons black with narrow yellow band across its crest, head otherwise black. Posterior margin of occiput slightly elevated in the middle line. Prothorax black above, yellow below. Synthorax with a black dorsum, marked with a yellow meso- thoracic collar, which is continuous across the mid-dorsal carina, and a pair of narrow, isolated dorsal bands. No trace of ante- humeral bands. Sides of synthorax yellow, with a broad black band on the first lateral suture, and a narrower band of grayish brown along the position of the second suture. Legs black, anterior femora yellow internally, posterior pair yellow extero-laterally. Abdomen black, sides of first and second segments yellow. Dorsum of first segment yellow, of second segment marked with a longitudinal trilobed band of the same colour. Segments 3-6 with ' basal ring of yellow, interrupted mid-dorsally, occupying about the first sixth of each segment. In addition segments 3-5 havea median longitudinal mark of yellow on the dorsum. The basal half of 7 is yellow; 8, 9 and fo are black, but 8 has a small basal lateral mark of yellow on either side. Anal appendages small, brownish-white. Apical half of seventh segment dilated from side to side, the succeeding segments progressively narrower. The sternite of the eighth segment is produced apically in a downward direction, so that it is visible in profile. The vulvar scale, pro- tected by this projection is small, bifid apically. A similar development of the sternite of the eighth segment of the abdomen occurs in certain species belonging to the Gomphus series (eg. Gomphus melampus Selys), but I do not know of any other Onychogomphus which shows a similar development. Vena- tionally and in colouration, however, I do not doubt but that this specimen is a true Onychogomphus. Trength of abdomen 37°5 mm., of hinder-wing 35 mm. The following species, known only from female examples, resembles this specimen to some extent in the colouring of the synthorax. O. frontalis Selys, Moolai, ‘‘ between Burma and Tenasserim.” Dorsal stripes of synthorax isolated, vestige of a juxta-humer- al band present. T-shaped mark on dorsum of frons. Occiput 410 Records of the Indian Museum. [VoL. XXIV, yellow. Length of abdomen 29 mm., of hinder-wing 25 mm. From de Selys account I should imagine that the vulvar apparatus of the female did not show the remarkable arrangement found in the female described above from the Garo Hills. Onychogomphus M-flavum Selys. 13. Darjiling Dist., 13000 ft., May, 1912. L.ord Carmichael’s collection. (cc/1319). y 19. Gopaldhara, Darjiling district, 1914. H. Stevens. The synonymy of this species is somewhat obscure. I have no doubt whatever but that the female specimen before me is properly referred to the Selysian species. The remarkable long vulvar scale, reaching to the apex of the abdomen is sufficient to make the determination certain, and the agreement in colouring is precise. The male is I believe unquestionably conspecific with the female, it differs only in having the costal-vein lined with yellow as far as the nodus. It is immature and badly crushed. I believe the colouring of the costal vein may well be lost in more adult specimens and need not be regarded as a specific character, A second, more serious difficulty lies in the fact that the male appears identical with that described by de Selys for O. bistrigatus (Selys) and figured by Hagen as such in the monograph. This is due to the fact, as I regard it, that the specimen figured in the monogtaph, and the male described in the second additions to synopsis are not conspecific with the type-specimens of O. bistrigatus, ‘but are really examples of O. M-flavwm, as is also the adult female (imperfect) noted in the monograph, in the Vienna Museum. Hence in my opinion O, bistyigatus remains known only from the type female from ‘India.’ The figure of the occiput of the female given in the monograph would appear to have been drawn from the Vienna specimen of M-flavum as it shows none of the ‘ dentellures ’ described for O. bistrvigatus. ; Mate. Dorsal stripes of synthorax meeting mesothoracic collar, mid-dorsal carina marked with yellow for its anterior halt. Juxta-humeral stripes complete. Sides of thorax yellow with narrow black bands on the lateral sutures. Abdomen largely black, with basal rings of yellow, constricted mid-dorsally, on segments 3-6; and a mid-dorsal yellow mark on each of the same segments. Basal two-thirds of the seventh segment yellow, distal third black. Segments 8, 9 yellow at the sides, dark dorsally, 10 yellow or light brown margined with black. Anal appendages yellow, upper pair darker apically, the distal third rather sharply angled downwards, and _lancet-shaped. Branches of lower appendage about two-thirds length of upper pair contiguous and parallel, curved upwards, with stout dorsal tooth at end of basal third, and smaller tooth at junction of middle and distal thirds (exactly as in pl. i, fig. 3, Mon. Gomph.). Seg- ments 8, 9 of abdomen, and distal half of 7 slightly dilated. 1922.] F. F. LawiLaw: Indian Dragonflies. 41 Costal nerve lined with yellow as far as the nodus (? colour line disappearing in more mature specimens). FEMALE. Colouring much as in the male; but the costal nerve is black (the specimen is much more mature than the male) and the dorsum of the tenth segment of the abdomen is black. The occiput is without dentellations, but has a fringe of long black hairs. The length of the vulvar scale, reaching to the end of the tenth segment, is very remarkable. Length of abdomen of @ ? +3.5 mm., of hinder-wing 32°5 min. ; length of abdomen of 2° 38 mm.,, of hinder-wing 34°5 mm. The two followiug species are known only from female speci- mens or from imperfect males, and I have not attempted to place them in any definite group. O. annularis Selys, North Burma. Dorsum of synthorax with dorsal stripes confluent with the mesothoracic collar, vestiges of the juxta-humeral stripe present. Described from two males, both imperfect. This species may belong to the geometricus group. Length of abdomen 37 mm., of hinder-wing 32 mm. ' QO. maclachlani Selys, North Burma. De Selys suggests that this may be the female of the last species. It is larger, lacks any trace of the juxta-humeral stripe, but otherwise, except for details of colouring of the head, is very similar. Length of abdo- inen 43 mm., of hinder-wing 38 mm. Another species, known only from the female, is O. cerastes Selys, from Nepal. This from its largely yellow colouring would seem to be rather a ‘ xerophilous’ species, like O. grammuicus. It is possibly allied to the female next described. Onychogomphus sp. 19. Kumaon, W. Himalayas, 22-vii-14. (8343/20) (lacking abdo- minal segments 4-10.) Differs from other species of the genus (except O. cerastes Selys, and O. grammicus Selys) that I have seen in having many of the cross-nerves of the wing whitish-yellow in colour. This is particularly the case with the rows of cross-nerves in the areas be- tween the subcostal nerve and M, proximal to the nodus, and with the cross-nerves of the cubital space. Head. Upper lip yellow, with fine black line at base. Ante- and post-clypeus yellow, the latter separated from the frolis by a transverse black line. Frons yellow, its extreme base black. Vertex black, with a small yellow spot between the posterior ocelli. Occiput yellow, its posterior margin slightly undulated. i Prothorax black, its anterior and posterior margins lined with yellow. Synthorax. Dorsum black, mesothoracic collar entire, joined by the lower ends of the dorsal stripes. Juxta-humeral stripes complete. Sides of synthorax yellow, with narrow black line on each of the lateral sutures. 412 Records of the Indian Mluseum. [VoL. XXIV, Legs yellow, with black spines and tarsi. Anterior surfaces of tibiae black. First pair ot femora with black antero-lateral band, which is much reduced on the succeeding pairs. Abdomen. First segment yellow, second with yellow sides and black dorsal marking enclosing a longitudinal median band of yellow, meeting a narrow apical ring of the same colour, Third segment yellowish-brown, its anterior margin finely marked with black, and with a lateral band of black not touching the anterior and of the segment. (The remainder missing.) Length of hinder-wing 31 mm. Lastly the N. Burmese species civcularis Selys, unknown to me except from de Selys’ account, is referred by its author to this genus with doubt. Group HETEROGOMPHUS. Large insects (h.w. 40 mm. or more). Cross-nerves between M,_, and M, of fore-wing as specialized as those of hinder-wing, sectors of arculus approximated at origin, curved, and nearly parallel. Cu, and Cu, in hinder-wing parallel to wing-margin. M, and Cu, divergent in the fore-wing. Brace of pterostigma occasion- ally wanting. Wings long and pointed resembling rather those of Ictinus. Windermost femora when adpressed barely reaching base of abdomen. Abdomen with basal rings on segments 3-7. Colour in some species tending towards a uniform brown. Segments 8 and g of abdomen slightly dilated from side to side. Upper anal appendages of male equal in length to the last two seg- ments of abdomen ; well separated, parallel and nearly straight. The two branches of the lower appendage also nearly straight, parallel, each carrying an internal tooth near its apex; each is neatly as long as an upper appendage. Genus Heterogomphus Selys. Genotype: H. smithii Selys. Species studied: H. ceylonicus, sp. nov. Characters of the tribe. Distribution: Himalayas, Ceylon, Indo-China, S. China, Great Sunda Islands. ; Heterogomphus ceylonicus, sp. nov. 1d. Ceylon. Col. Yerbury. Brit. Mus. Head largely black. Upper lip yellow, margined with black, and with a black projection in the middle line from the base. Ante-clypeus yellow, post-clypeus black, with a lateral yellow spot on either side. Vertical part of the frons black, horizontal part yellow, with narrow black base. Vertex black, occiput black with median yellow spot. Prothorax black with large lateral spot of yellow and a smaller -median paired spot of the same colour. 1922.} F. F. Lamiaw: Indian Dragonflies. 413 Synthorax black, mesothoracic collar yellow, interrupted by the black of the mid-dorsal carina in the middle line. Dorsal stripe isolated, broad, elongate oval ; antehumeral stripe represented by a small superior spot. Sides of synthorax black with a broad yellow band on the mesepimerite and a second on the metepimerite. These yellow bands are continuous on to the meso- and the meta- notum, Legs black, coxae and anterior surfaces of the femora marked with yellow. Abdomen black. ‘The first segment has a transverse yellow mark at its base dorsally. The second has a longitudinal dorsal band of yellow which is trilobed. Both these segments are marked with yellow laterally, including the oreillets. The third and fourth segments have a basal ring of yellow, and Fic. 21.—Anal appendages of Heterogomphus dorsally, from this a ceylonicus, sp. nov., & from type in the longitudinal basal ex- British Museum : side view. tending for about half the length of the segment. Segments 5 and 6 have a narrow basal ring only; the basal half of 7 is brownish-yellow, and segments 8 and 9 have dark brown lateral basal marks. The tenth segment is entirely black. ; Anal appendages gray-black, upper pair slightly down-curved. Lower appendage with branches about five-sixths the length of upper pair each branch carrying a small internal tooth at about the commencement of its distal quarter. Antenodals of fore-wing 17, 18, post-nodals 12-14. Length of hinder-wing 41 mm., of abdomen 42+5'5 mm. A handsome species rather resembling an Ictinus inits propor- tions, I fancy that the late Mr. Kirby must have overlooked the specimen for this reason when describing Col. Yerbury’s collection. H. ceylonicus is very distinct from other species of the genus in the markings of the synthorax, which are bolder and more sharply contrasted than is usual in the genus. It resembles in sizé the Malayan species of the genus. These are H. sumatranus Martin, from Sumatra and Borneo, and H. tcterops Martin, from Java and Borneo. The British Museum has a specimen of both these species. H. smithi Selys, is also represented in the Museum collection, it is very similar structurally to these species though considerably larger, and differs of course in details of colouring. It is found in the Himalayas. H. cochinchinensis Selys, is apparently an allied species from Cochin-China. H. sommert Selys, from China is still larger and differs from the other species of the genus in having the upper anal appendages of the male a little incurved to one another apically. H. untcoloy Martin, from Siam is allied to or identical 4T4 Records of the Indian Museum. [Vo. XXIV, 1922.) with H. swmatranus. Lastly H. naninus Foerster, from Tonkin, which IT have not seen, is almost certainly not a Heterogomphus at all. Genus not referred to any of the above defined ‘ groups.’ Genus Ophiogomphus Species examined: O. reductus Calvert. A genus belonging to the series Gomphus, holarctic in distri- bution, containing a number of robust species of moderately large size, mostly characteristic of mountainous country, whose larvae live in rapidly running rivers with sandy beds. Venationally the genus differs from other members of the series noted in this paper by the possession of a small but quite definite ‘anal loop.’ The arrangement of cross-veins between M43 and M, is constant and specialized ; the pterostigma is small and well braced, the triangles of fore and hinder-wings are sub- equal. The hindermost femora reach to the end of the first seg- ment of the abdomen, and are armed with short black spines arranged irregularly for the basal half of the femur, on the distal half in two rows. Upper anal appendages of the male asin the group Gomphus ; lower appendage not so long as the upper pate, cleft narrowly for about its distal half. Larvae with wing-sheaths divergent. This genus may very likely stand as a distinct tribe, but as its distribution is Holarctic and its inclusion in this paper depend- ant rather on political than on zoogeographical boundaries I leave its exact position to be defined in some more appropriate place. Ophiogomphus reductus Calvert. 17. (newly emerged, with exuviac), Kashmir, 1915. H. 1. Pease. (890/fH11). 19. IXashmir, 1915. H. I. Pease. (582/1-+1). 1@. Jhelum Valley, Kashmir, 5200 ft., July, 1916. H. T. Pease. (4819/1+1). Length of hinder-wing of ? 36 mm., of abdomen 40 mm, APPENDIX. By Major F. C. Fraser, I.M.S. (Plate XI.) Since I handed over a collection of Gomphines to Mr. Laidlaw in 1920 I have come into the possession of a further lot collected partly by myself and partly by friends who have kindly placed them at my disposal. Some of these have already been described, viz. Stylogomphus inglist, nearly related to Heliogomphus, Onychogomphus sp., and females of two distinct species of Heterogomphus, which will appear shortly in the Memoirs of the Department of Agriculture in India, Entomological Series. ‘The remainder, some I1I0 species are now before me and provide new material and add new light to what has already been given above by Mr. Laidlaw. Dr. Annandale has kindly given me this opportunity of adding to the most important paper which has appeared on the Indian Gom- phinae since Williamson’s paper was published in 1907. Some of the present genera will have to be further split up, I refer especially to the genus Onychogomphus, the larvae of two species of which, O. biforceds and O. lineatus, are contrasted below. The body of the former is broad and greatly depressed and its antennae broad, flat, triangular plates, the body of the latter is natrow and cylindrical, whilst its antennae conform to such as prevail amongst the majority of known Gomphine larvae (see text-fig., p. 426). With such wide variations in structure, it is impossible to believe that the two insects fall into the same genus. With a long and wide experience of these insects in their natural habitat, I am able to sav that few species occur commonly ; a few such as Cyclogomphus and Anormogomphus are locally common, only two are widely spread, the remainder are come upon at odd intervals and in unexpected places and then only as solitary individuals. With the exception of O. lineatus and Ictinus rapax they are single brooded, a few species emerge in swarms, generally after heavy rain and their appearance is of remarkably short duration, two to three weeks being the extent of their life on the wing. The types of new species described below will eventually be deposited in the national collection in the British Museum, at present they remain in my own collection. I hope to place para- types in the Indian Museum, at least as far as the Nilgiri species are concerned. 416 Records of the Indian Museum. [VoL. XXIV, Heliogomphus nietneri Selys. 1¢ Kalar, 1000 ft., March 1916, coll. F. C. Fraser. This specimen differs form the Ceylon form described by Hagen in the same particulars as the Assam species described by Laidlaw except that there are lateral markings as far as abdom- inal segment 6. The wings are saffronated at the base as far outwards as the node, this colour gradually diffusing outwardly. Stigma pale brown. Caught in dry jungle a mile or so away from the nearest water. Heliogomphus pruinans, sp. nov. A single pair taken together, Burliyar, Nilgiris, 1500 ft., 29'vii:21. Male. Hindwing 32 mm. Abdomen 42 mm. Head black, the labrum with two basal greenish spots whose opposing borders are deeply concave; bases of mandibles greenish white; a greenish white band across the frons which is rounded and flattened ; occiput depressed, black ; eyes bottle green. Prothorax black, the posterior lobe and an anterior band yellowish green, beneath pruinosed white. Thorax black on the dorsum marked by a complete, meso- thoracic collar and narrow dorsal bands lying close to and parallel to the dorsal crest, both greenish yellow ; laterally greenish yellow marked by two narrow black lines on the sutures, confluent above and below. Beneath and on lower part of sides pruinosed white. Legs black, the hind femora with a row of very closely set and very small spines. Wings enfumed ; stigma blackish brown, braced only in one wing ; no basal antenodal ‘of second series; only one cubital TI-15|15-12 I2-11|12-12 wing very elongate, the costal side twice as long as the basal; 3 rows of discoidal cells at level of node; all triangles entire; base of wings very oblique and closely resembling that of Anisogomphus ; other points as for genus. Abdomen black marked with pale greenish yellow as follows : a fine, middorsal line extending from segments 1 to 5, thickest at 2; acomplete basal annule almost encompassing segment 7 and occupying about the basal third; a minute triangular basal point on the dorsum of 8, the remainder black. Laterally a spot on 1, the oreillet and an apical spot on 2, and small basal spots on 3 to 6. Anal appendages as for genotype, the superior black at base, pale green to yellow at the apices, which are turned at first in, then up and finally out. Female very similar to the male but much larger. The occiput simple, depressed, exactly similar to that of the male. Sides of thorax vivid greenish yellow, the pruinescence on the lower part of sides and beneath more marked than in the male. Wings saffronated at the extreme base. ‘The Jateral spots on segments I to 3 forming a continuous unbroken line; on segments 4 and 5 the line is interrupted to form two elongate spots, whilst ; trigones of hind- nervure to all wings; nodal index 1922.] F. C. Fraser: Indian Dragonflies. 417 on 6 the line is represented merely by a basal and a subbasal spot. Legs black, hind femora with a row of long, robust, evenly spaced spines. (The sexual differences here are very striking.) Vulvar scale as for H. nietnert, but the anex scarcely notched. - The insects are jungle-loving creatures, haunting the rocky beds of wild, mountain streams. Microgomphus sp. A single, somewhat teneral female from Rangoon, coll. 1909. I am satisfied that this specimen is a true Microgomphus although it differs from the generic characters given above by Mr. Laidlaw in the two following particulars :— (i) The proximal angle of triangle in the forewing is not as far distant from the arc as the length of the proximal side of subtrigone. (ii) There are two rows of cells between Cu,;; and the hind margin of forewing. I have however examined a number of both sexes of M. torquatus and find that neither of these two features are constant. As regards the number of rows of cells posterior to Cu,;; I find that there are invariably two rows in the female and one in the male. The distance of the proximal angle of the triangle in the forewing is very variable in both sexes, being sometimes more and sometimes less than the length of the proximal side of the subtrigone, so that other generic characters being present one may rightly assume that the specimen is a true Microgomphus. Female. Abdomen 25 mm. Hindwing 22°5 mm. Head black marked with citron yellow as follows: bases of mandibles, small basal spots joined by a basal streak on labrum, small lateral spots and a very fine streak on upper epistome, a line on the front of frons interrupted in the middle; occiput black with paired minute spines on either side of its middle. Prothorax entirely black. Thorax black on the dorsum marked by a narrowly interrupt- ed, citron yellow, mesothoracic collar and an oblique dorsal stripe not joined to the collar; humeral stripe absent. Laterally bright citron yellow, the sutures finely outlined in black. Legs black, the anterior femora yellow outwardly, the hind femora with about 9 robust spines on either side of the limb. Wings hyaline, saffronated rather deeply at the base, venation as for genus subject to the exceptions already discussed ; stigma pale brown, feebly braced ; nodal index = = Abdomen black, marked with yellow as follows: segment I all yellow, 2 broadly on the sides, the dorsal carina finely, 3 with a basal dorsal spot and an elongate spot on the middle of dorsal carina, 4 similar but the spots much smaller, 5 to 7 with only the basal spots which are largest on 7 ; 8 to 10 all black. 418 Records of the Indian Museum. [Vor XXIV, Anal appendages yellow. Vulvar scale very small, cleft to its base. ; This species is probably the smallest Gomphine known and is _smaller than any individual M. torquatus which I have examined. Perissogomphus stevensi Laid. I have received several specimens of this species from Darjil- ing and Assam and from an examination of the venation, am able to add the following to the generic characters :— (i) Usually 1 but sometimes 2 rows of cells between M; and M jq at the level of the distal end of stigma. (ii) Triangles of both fore and hindwings of female and also hindwings of male frequently traversed by a nervure. Concerning the species, whilst the majority of those received have the ground colour yellowish brown, some specimens are a bright greenish yellow and I am satisfied that the former colour is the effect of decomposition and the latter the true colour during life. Davidius sp. . A single female from Gopaldhara, Assam, coll. H. Stevens. Abdomen 34 mm. Hindwing 32 mm. Head glossy black, the bases of mandibles, 2 spots at the base of the labrum and a stripe across the frons citron yellow. Occiput greatly depressed at its centre. Prothorax black with a small median spot on the posterior lobe, a geminate spot in front of it, the anterior border and a small spot on each side citron yellow. Thorax black on the dorsum marked with a dorsal stripe parallel to the dorsal carina and connected with an interrupted mesothoracic collar; a triangular spot above and to the outer side of the dorsal stripe; a fine, humeral line separated from the spot, the long axis of the latter being at right anglestoit. The sides citron yellow with narrow black stripes on the lateral sutures. Spots of yellow on tergum and at bases of wings. Legs long and slim, black, the anterior femora yellow out- wardly, the hind with a row of very long, robust, widely separated spines, 6 in number, mid femora with a row of closely set, short spines and a single long one at the distal end. Abdomen black, marked with yellow as follows: a triangular, apical spot on segment I, the base of the triangle at the apex of the segment, 2 with a bilobed spot on its middorsum, 3 to 7 and base of 8th with a fine, middorsal stripe, segments I to 3 broadly yellow at the sides. Anal appendages small, conical, yellow. Wings as for type, triangles of hindwings crossed, the left forewing has 2 basal antenodal nervures of the second series (a very rare occurrence), the other wings with one each. 1922.] F. C. FRASER: Indian Dragonflies. 419 Gomphus sp. A single female, Gudalur, Nilgiri Wynaad, 3500 ft., 8:vii'21, coll. F.C Fraser. Female. Abdomen 34 mm. Hindwing 30 mm. Head black, marked with citron yellow as follows: lateral lobes of labium, 2 large basal spots on labrum and the bases of mandibles. A minute spot in the centre of epistome and a larger one on each side against the eyes; the frons, which is rounded , with a broad stripe on its upper surface; occiput simple, black, fringed thickly with very long black hairs ; eyes bottle green. Prothorax black with a large triangular citron yellow spot on either side. Thorax black on dorsum with a complete, broadly joined, mesothoracic collar, prolonged slightly upward along the dorsal carina, an oblique, short, dorsal stripe well separated above and below from the alar sinus and the mesothoracic collar respectively. A humeral stripe represented by a small upper spot and a vesti- gial streak below. Laterally citron yellow, the sutures outlined in black. The thorax is coated with long, rather dense. black hairs. Legs all black, hind femora with a row of 11 to 12 robust, evenly spaced, gradually lengthening spines on either side of the limb. Wings slightly enfumed, saffronated at the extreme base as far out as the triangles which are all entire; stigma short, red- dish brown, braced ; no basal antenodal nervure of the 2nd series ; I to 2 cubital nervures in forewing, only 1 in the hind ; 2 rows of cells in the postanal area of forewing, 4 to 5 in the hind ; 2 rows 13-15[15-15 I3-I0|1II-12 (two of the postnodals of forewing are connected by a nervure and another is forked) ; 2 rows of cells as far as the level of node. Abdomen black, marked with brilliant citron yellow as follows : the dorsum and sides of segment 1 broadly, 2 with a trilobed, dorsal stripe and the sides very broadly. 3 with the middorsal carina finely and two large lateral spots, 3 to 7 with basal annules notched by the black on the dorsum except on 7 where the apical border of the spot is straight, on 8 the basal spot a mere fine line, g and 1o with apical annules, broad on 9, narrow on Io. Anal appendages black; segments 7 to 10 progressively shortening ; vulvar scale very tiny, barely evident. This handsome species will probably prove to be the type of a new genus but until the male is found will have to be confined to genus Gomphus sens. sty. of which it at present forms the smallest species. of cells between M; and Myjqin all wings ; nodal index Gomphus nilgiricus Laid. A single female, Coonoor, June 1917, coll. F. C. Fraser. Abdomen 43 mm. Hindwing 40 mm. This specimen was picked up dead, a small army of ants 420 Records of the Indian Museum. [Vor. XXIV, was bearing it away and had somewhat damaged it. I have seen a second female near Gudalur (August 1921) which was oviposit- ing in wet sand in the half-dried bed of a mountain stream. The insect was quite fearless and flew backwards and forwards several times passing under my outspanned legs. Occasionally it hovered a few inches above the sand and made stabbing motions with the end of its abdomen in the wet sand. The function of its long ovipositor was thus explained. I had no net on this occasion and though I attempted to take it with a sweep of my hand it eluded me. It differs in the few following respects from the above described male: the abdomen is of nearly even width throughout and perfectly cylindrical, the markings are a greenish yellow and on segments 3 to 5 are of the note-of-exclamation type, the dorsal stripe being swollen and rounded at the base of the segments, tapering pin-like to the apex ; segments 6, 8 and 10 are unmarked and g has a large, dorsal, yellow mark extending from apex to base in a narrowing point. Anal appendages small, conical, pointed, black. Vulvar scale of remarkable length, very narrow and pointed and somewhat analogous to the structure as seen in Cordulegaster. Gomphus o’doneli, sp. noy. A single male from Hasimara Tea Estate, Duars, Bengal, coll. H. V. O’Donel. Head very large ; labium pale yellow, the middle lobe bor- dered with black, rest of head black save for two small basal spots on labrum and a narrow stripe across the crest of frons. Occiput curled up at its border and fringed thickly with stiff black ‘hairs. Prothorax black, the posterior lobe, a small spot on either side of it, a geminate spot in front of it and the anterior border yellow. Thorax black, marked with yellow as follows: oblique. dorsal stripes meeting a slightly interrupted mesothoracic collar, humeral stripe represented only by a small spot above. Laterally a broad, posthumeral stripe and the anterior three-fourths of the metepi- meron, On the broad black between these two stripes there are three small yellow spots. Legs short and robust, entirely black. The hind femora with the surface covered with small spines and a single larger one at the distal end. Wings hyaline, stigma dark brown, braced, rather small; membrane very narrow, dark brown; 2 nervures between M;, and M;.;;; in the forewing, only 1 in the hind; only 1 row of cells between M; and Mj, at level of distal end of stigma; Cu; and g-16/14-9 fr-10|TTSro0\" rows of discoidal cells at level of node in the forewing ; all triangles entire ; 3 to 4 rows of post-anal cells in hindwing. Cu;; nearly parallel to wing border ; nodal index 1922.] F. C. FRASER: Indian Dragonflies. 421 Abdomen tumid at base, 3 to 7 very narrow and cylindrical, the latter broadening at apex, 8 and g very broad, especially the former (but not winged), 10 rather small, segments progressively smaller from 7 to Io. Abdomen black, marked with yellow as follows: a triangular dorsal spot and a large lateral spot on segment 1; a trilobed dorsal spot, the oreillet and a large apical lateral spot on 2; 3 with the dorsum narrowly at the base and a large lateral basal spot; 4 to 6 with dorsal basal spots meeting rather broadly over the carina; 7 with a broad basal annule prolonged apicalward along the dorsal carina; 8 with a small round spot on the dorsum at the base, the basal part narrowly expanded, also an L-shaped mark on the middle of the side, the ‘‘L’’ lying on its back ; 9 with the whole of the lateral border yellow; Io unmarked. — Anal appendages black, much the same as in the genotype, the superior, however, very broad, hollowed out below and corre- spondingly domed above. Hamuli projecting as two long foliate structures directed forwards ; lobe of penis of enormous size. The robust short stature and the general facies of this species are quite unlike any other Gomphus I know of from India and recall the size and shape of Ictinus. Burmagomphus pyramidalis, race. 1¢ Hasimara Tea Estate, Duars, Bengal, coll. H. V. O’Donel. Differs from type by the occiput being all black, by the yellow on the labrum being cut up into two spots by the median prolonga- tion of the basal black meeting the bordering black, and lastly by the inferior anal appendages being strongly recurved upwards at their apices almost like a fish-hook. The size is also much larger: abdomen 36 mm., hindwing 26°5 mm. Wings rather deeply enfumed. Burmagomphus duarensis, sp. nov. A single male from Hasimara Tea Estate, coll. H. V. O’Donel, Sept. 21. Male. Abdomen 34mm. Hindwing 26 mm. Head entirely black save for bases of mandibles and a broad, bright yellow line on the upper surface of frons; occiput straight, simple, fringed with a few black hairs. Prothorax black with a large citron yellow spot on either side. Thorax black on dorsum with dorsal oblique stripe united to a narrowly interrupted mesothoracic collar; humeral stripe repre- sented only by a small upper spot. Sides citron yellow, the sutures finely outlined in black. Wings hyaline, saffronated at extreme base; stigma light brown, braced, one or two rows of cells between M; and Mjg; a basal antenodal nervure of 2nd series in both the forewings; nodal II-15]16-11 index : —-_—-|—_—. II-I0]Io-II 422 Records of the Indian Museum. [VoL. XXIV, Legs entirely black except the anterior pair of femora which are pale whitish green outwardly ; hind femora with a row of long robust spines to the number of 6 on the distal half. Abdomen tumid at base, very narrow and cylindrical from 3 to 6, remaining segments dilated, especially apex of 7 and whole of 8. Black, marked with citron yellow as follows: Ist segment entirely yellow save for two black dorsal spots; 2 with a trilobed dorsal spot the oreillets and an apical spot; 3 to 4 with the dorsal carina finely yellow and a lateral stripe, broadest at the base and not extending quite as far as apex; 5 to 7 with the lateral basal spot meeting over the dorsum and on 7 occupying the basal half of.the segment; remaining segments black. Anal appendages as for type, the superior pale yellow, inferior black. Indogomphus, gen. nov. Wings with a basal antenodal nervure of 2nd series to all wings, all triangles entire, the triangles of hindwings elongate, sectors of arc approximating immediately after the arc, 3 trans- verse nervures between M;_;;; and Mj,, in forewing, only one in. the hind, Cu; and Cu;; a little divergent in the hindwing, 4 rows of cells posterior to Cuj; in the hind wing, forking of M;_j; and Mj;; symmetrical, discoidal field divergent, 3 to 4 rows of cells at level of node, 3 rows of cells between M; and Mj, at distal end of stigma, only 1 row of cells (occasionally 2) at base of forewing, base of wing rather oblique- as in Anisogomphus and angle not prominent. Anal appendages of male very similar to those of Heliogomphus. Vulvar scale broadly triangular, almost equilateral, the apex with a small rounded notch, Indogomphus longistigma, sp. nov. A single pair from the Nilgiri Wynaad, 3000 ft., 14‘viii'21, coll. F. C. Fraser. Male. Abdomen 44 mm. Hindwing 35 mm. Hindwing 37 mim. in the female. Head entirely black save for a broad greenish yellow line on the frons overlapping the front and a broad line on the occiput of the same colour. The frons finely black at the base and the occiput black at either end and fringed with remarkably long black hairs. Eyes bottle green. Prothorax black, the posterior lobe and a small oval spot adjoining it anteriorly and a band on its anterior border pale yellow. Thorax black, marked with bright yellow as follows: a com- plete mesothoracic collar which sends a prolongation up along the middorsal carina as far as the alar sinus; a narrow dorsal stripe close alongside the middorsal carina reaching the alar sinus above 1922.] F. C. Fraser: Indian Dragonflies. 423 but not meeting the mesothoracic collar; a vestigial antehumeral stripe represented by a subquadrate spot above and a mere trace of a line some distance below it, barely visible to the naked eye; a broad humeral stripe and a very narrow mid-lateral, both discon- nected from two larger spots below themselves ; finally, the posterior two-thirds of the metepimeron. Beneath black, marked by a fine, V-shaped, yellow spot. Legs slim, black and very long, the hind femora extending to the apical end of the 2nd segment and furnished with four pairs of very long equidistant black spines The anterior femora are greenish yellow on the flexor surface. Wings hyaline, long and narrow ; stigma pale brownish yellow, that of the hind much larger than that of forewing, 3°5 mm. to 5 perils Om membrane absent. I2-II|10-12 Abdomen a little tumid at the base, segments 3 to 6 slim and cylindrical, the apical part of 7 and segment 8 dilated and with rudimentary lateral leaves, 8 and 9 almost the same length, ro half the length of 9. Black, marked with yellow as follows: segment I with a large quadrate spot on the side and a broad stripe on the dorsum ; segment 2 with an L-shaped spot on the side, the under- side of the very robust oreillets and a small stripe on the upper surface of the samme structure and a triiobed dorsal band; 3 with a lateral basal triangular spot followed by a small oval stripe about the middle of the segment and its dorsum, widely at the basal third but less so afterwards and not extending to the apex of segment; 4 to 6 with the same markings but the lateral stripe absent and a wide gap between the basal, dorsal yellow and that following it (this dorsal yellow is peppered with minute, black spines); segment 7 with nearly its basal half yellow broadly and its middle third narrowly; 8 to 10 with merely the dorsal carina moderately, finely yellow. Anal superior appendages pale yellow, lvrate, broad at base, tapering to a fine point, a little upturned, at first divergent but then curling in so as to meet at the tips and enclose a circular open- ing. The outer side with a broad blunt spitie. Inferior append- age with widely divergent branches, projecting from below the superior so as to be seen from above, black. (It will be seen from this description that the appendages are very similar to those of Heliogomphus to which the genus is closely allied.) Female very similar to the male, differing as follows: bases of mandibles yellow (occiput similar to the male, simple but with only fine, spatse, short hairs) ; abdomen with gth segment about the same length as 8, which is a little dilated, and tapering rapidly to Io, which is very small and narrow (the tapering end of abdomen suggestive of that of Macrogomphus). Anal appendages small, conical, palest yellow, as is also a small, cone-like protuberance between them. Hind femora with 5 to 6 pairs of long black spines similar to mm. in the hind; nodal index 424 Records of the Indian Museum. [VoL. XXIV, but more numerous thanin the male. Between them are numerous smaller, evenly sized spines. Wings a little enfumed, stigma light aoe the difference in 12-17|17-12 II-I0|1I-I1’ size even more marked than in the male; nodal index otherwise similar to the male. Onychogomphus bistrigatus Selys. A single male from Kalar, Nilgiris, 1000 ft., v-1917, coll. K. C. Fraser. The specimen agrees so minutely with the Selysian descrip- tion of the type female that there can be no doubt that the insects are conspecific. Abdomen 38 mm. Hindwing 29 mm. Abdominal segments 8 and 9g are black with a small basal lateral yellow spot, whilst Io is all black. The anal appendages are yellow, the superior tipped with black, the inferior black externally, yellow internally. The supe- rior has a small spine inwardly at the junction of the last two- thirds, the inferior has only a basal tooth at its basal third, otherwise as for O. uncatus Selys. Onychogomphus biforceps Selys. A single female from Palghat, Malabar Dist., coll. T. N. Hearsey, 16:vi- 21. Female. Abdomen 42.mm. Hindwing 35 mm. The insect is somewhat stouter and larger than the female of O. biforceps nilgirvensts and the markings differ considerably ; the 7th and 8th abdominal segments are also more dilated. Differs in markings as follows: spots on labrum large, in fact it would be more correct to say that this structure is yellow, narrowly bordered with black, the narrow belt of black at base connected narrowly in the middle line with the anterior bordering black ; base of mandibles, lower part of epistome and a spot on either side of upper yellow; band on frons complete; occiput yellow at its centre and raised into a single point in the middle (humeral stripe absent, antehumeral oblique, not connected with the mesothoracic collar which is slightly interrupted). Laterally the broad median black stripe is traversed by a narrow irregular yellow stripe. Legs. Femora yellow mottled with black, the hind femora with a row of closely set, very short, very robust spines, g to ro in number. Abdominal markings similar but the basal spot on segment 8 very large. Anal appendage entirely yellow, as is also the apex of the intermediate conical process. Vulvar scale light reddish brown, very short and broad, deeply notched in the middle, reset in a hollow of the gth segment, 1922.] F. C. Fraser: Indian Dragonflies 425 Wings similar but only a single row of cells between M; and 10-16|17-10 g-II|I1i-II Mjq; nodal index and 3 rows of cells in discoidal field at level of node. Onychogomphus biforceps nilgiriensis, subsp. nov. 3 6 andi Q Gudalur, Nilgiris, 16-x'21, coll. F.C Fraser. (One male, the type, coll. by T. Bainbrigge Fletcher on the same date.) Male. Abdomen 35°4mm. Hindwing 30 mm. Differs from O. biforceps Selys as follows :— Yellow spot on occiput absent ; mesothoracic collar interrupted ; humeral stripes entirely absent ; a small yellow spot just below wings on the upper part of the broad lateral black stripe ; oval dorsal spots on abdominal segments 4 to 6 absent ; a minute apical spot in ad- dition to small lateral basal spot on segment 8 ; superior appendages yellow only on outer side, black inwardly ; legs entirely black. The female of O. biforceps Selys has not been described, but is probably somewhat like that of the present species which is as follows :— Abdomen 36 mm. Hindwing 33 mm. Colouring similar to the male with the following exceptions: spots on labrum very small; band on frons interrupted by the black in the floor of the suture so as to form two oval spots; occi- put with two robust spines situated close together at its middle; antehumeral band not connected with the mesothoracic collar ; yellow spots on sides of segment 2 and the oreillets confluent ; basal spots on dorsum of 3 to 7 interrupted by the black of dorsal crest ; basal spot on 8 very minute and the apical one absent. Anal appendages rather longer than segment 10, tapering, black with a bright yellow tip, a long triangular protuberance between them ; 7 to Io progressively shortening. Vulvar scale half the length of segment 9, bifid to its base so as to form two small triangular leaves. Wings enfumed and distinctly saffronated at the base ; nodal 12—14|15-I1 II-10|I10-13, . Jungle-loving insects, hiding up in shady mountain streams. Description of larva of O. biforceps nilgiriensis. Total length 23 mm. Length of hind femora6mm. Greatest breadth of body at abdomen 9 mm. Head moderately broad and quadrate, a postocular spine on either side; antennae remarkably specialized, basal segments small, cylindrical, 3rd se#ment broad, flattened and triangular, sloping downward and forward, 4th segment present only as a rudimentary tiny spine at inner lateral angle of 3rd segment. Prothorax small, a double tubercle on its dorsum. Wing sheaths broad, extending to middle of segment 6 or thereabouts. index: 426 Records of the Indian Museum. [VoL. XXIV, 1922.] Abdomen greatly depressed, segments 7 to Io with stout apical lateral spines, 4 to 8 with well marked dorsal ridge, raised up as robust spines on each segment. Mask very short, extending to base of first pair of legs only, middle lobe rounded and fringed with rather long stiff brissae. Mentum angulated, the whole lube nearly quadrate. Easily distinguished from the larvae of any other species of Gomphine by the shape of its antennae. Four specimens were £ \G a. Larva of Onychogomphus biforceps nilgiviensis. 6. Antennae of same. c. Mask of same. d. Larva of Onychogomphus lineatus. e. Antennae of same. found in the pool of a mountain stream, amongst debris consisting mainly of rotting leaves, twigs, etc., above Gudalur, 26:ii'22. Five adults insects were taken within Io yards of this pool and 3 others seen. No other species have been seen or taken over this stream and no other kinds of larvae found, so that there is no doubt as to the species to which they belong. If the breeding out of these larvae prove the correctness of the diagnosis, I propose to remove the group biforceps from Onychogomphus and erect a new genus for it with the name of Lamelligomphus. The head of the larva reminds one irresistibly of that of a cockchafer (Melolonthidae). ee - —— OES > jan \aegpeeenoll Gy t DESCRIPTION OF PLATE XI. Fic, 1.—Thoracic markings of Heliogomphus nietneri Selys. 2.—Markings of Heliogomphus pruinans. Male. 3.—Thoracic markings of Davidius sp. Female. 4.—Markings of Gomphus sp. Female. 5.—Abdominal markings of Gomphus nilgiricus Laid. Female. 6.—Markings of Gomphus o’doneli. Male. 7.—Markings of Burmagomphus duarensis. Male. 8.—Semi-lateral view of Indogomphus longistigma. Female. 9g.—Markings of Onychogomphus bistrigatus Selys. Male. 10.—Markings of Onychogomphus biforceps Selys. Female. 11.-—Markings of Onychogomphus biforceps nilgiviensis. Male. PLATE XI. Mus., Vou. XXIV, 1922. Rec. IND. “GE6L “BIINI]V) “vIpUy JO S9Aang ai Jo S990 2y2 Iv parutig ‘wuryduor ATrureyqns ay} Jo seruosviq uerpuy tad a) nae ’ ¢ oi 2 ' Aap larg ee fouh oF Comepl ni eh, i \ RECORDS of the INDIAN MUSEUM (A JOURNAL OF INDIAN ZOOLOGY) Vol. XXIV, Part IV. OCTOBER, 1922, PAGE Some Earthworms from Kashmir, Bombay, and other parts of India. J. Stephenson wen 427 Indian Mysidacea. W. M. Tattersall s 445 Parallel Evolution in the Fish and Tadpoles of Mountain “Toned N. Annandale and S. L. Hora 5 se «= 505 Some Oriental Ascalaphidae in the Indian Ausen, F. C. Fics 511 Hirudinea from the Inle Lake, S. Shan States. A. Oka 521 Descriptions of some Indo-Malayan species of Caprifermes (Termitidae). F. Silvestri oh 535 Zones Instiige. > a He: Ep aN tx JAM 71924 *} A vd wi Mo x3 yi ee Calcutta: PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA. PRINTED AT THE BAPTIST MISSION PRESS. 1922. Price Iwo Rupees. alii x ih ce Me i stoma anh La heya le singin ‘noon io ivan i ae oer Ee ee i insets au bs :} bebitqehae’ Yatra pis j 5) Kn Mee a a RON ANAL nel 8” oratnd) Vink mal enemy Bien | | .{ seberhaeys 1 seis inn te: aay vt si yatabiood® tome Ww eV * ack a Fi, yet ert " it ry ay re fee sae i J 4 Puan Reg) Clty tt Ren 1 Oe te ri Mer, | > DPy ; et 1 ee mie i 9h ; ay A hi : 7 a. ‘ ‘nn . \ ’ © + J » — : \ a i ‘ iS ‘ ih i PJ = " e i M , ’ 4 not na 4 ti ral < POOMMAL Shh +. CM vi . yy ey a \ 7] vi if oy i Ae. ty it Le ay? 3 i 4 SOME EARTHWORMS FROM KASHMIR, BOMBAY, AND OTHER PARTS OF INDIA. By J. Srepnenson, M.B., D.Sc., Lecturer in Zoology in the University of Edinburgh. CONTENTS. Page The Earthworms of Kashmir a ae a wee = 428 The Range of the Lumbricinae ats até w. =428 The Polyphyletic Origin of the genus Megascolex we ane bp 20 Some since species of interest ab an + 430 Systematic Part. Family Moniligastridae. Drawida nepalensis Mich. wilt Lo oma wpaedgo Drawida vosea sp. nov. at we sae vee 430 Family Megascolecidae. 7 Subfamily Megascolecinae. Megascolex konkanensis Fedarb se tis mol gl Megascolex mauritit (Kinb.) ... i Se AG? Megascolex hovrai sp. nov. Se cma io Asie ty 23 Pheretima elongata (E. Perr.) . 308 =a a 7433 Pheretima hawayana (Rosa) ... sch ade ee 9433 Phevretima heterochaeta (Mich.) ace jee Sean Paes Pheretima houlletti (E. Perr.) ... sia ae we 434 Pheretima posthuma (L. Vaill.) ot re A) AB4 Pheretima suctoria(Mich.) ..: oe re eee aaa: Perionyx excavatus E, Perr. ... sat pat das Perionyx modestus sp.nov. ... he te sua 435) Subfamily Octochaetinae. $ Octochaetus beatrix Bedd. Ay, re 430 Erythraeodrilus kempi var. bifoveatus (Ste ph. ) St 2437: Eutyphoeus orientalis (Bedd.) Se Piel] Eutyphoeus waltont Mich. oa Seb Ba 438 Eudichogaster mullani sp. nov. A: hee tie Gh) amily Lumbricidae. Subfamily Glossoscolecinae. Pontoscolex corethrurus (Fr. Mill ) ese at .-» 440 Subfamily Lumbricinae. 4 Helodrilus (Allolobophora) caliginosus subsp. trapezoides (Ant. ) a8 ... 440 Hela his (Allolobophova) prashadi sp. nov. Ad e440 Helodrilus (Dendvobaena) kempi sp. nov. ... aa ane eve Helodvilus (Bimastus) constrictus (Rosa) .. Bes 1. 442 Helodrilus (Bimastus) parvus (Eisen) FS Me aan 4A2 Octolasium lacteum (Orley) _... Hee Ate steA4g The present paper contains an account of some Oligochaeta receutly received for identification from the Indian Museum. In part these have been collected by the officers of the Zoological Survey, and in part sent tothe Museum by other naturalists ,—Prof. J. J. Asana of Ahmedabad and Prof. J. P. Mullan of Bombay. The collection of the last-named contained some Ue specimens. 428 Records of the Indian Museum. [VoL. XXIV, THE EARTHWORMS OF KASHMIR. Our knowledge of the worms of Kashmir has hitherto been meagre in the extreme. In the Report on the Natural History Results of the Pamir Boundary Commission, published in 1898 (1) Alcock states: “‘ Three species of earthworms were obtained, one in the Kishenganga Valley at 8r1oo ft., one in the Gilgit River Valley at over 7000 ft., and one in the Yasin Valley at 8000 ft. Specimens of all of these were sent to Mr. F. E. Beddard, F.RS., who writes as follows concerning them: ‘They are entirely European, i.e. Palaearctic species; they belong, in fact, to the usual British forms. This is of interest, as being an approxima- tion to discovering the limits of the Oriental region for worms.’ ’’ Michaelsen in 1909 (3) identified three species (Eisenia rosea (Sav.), Helodrilus (Allolobophora) caliginosus subsp. trapezoides (Ant. Dug.), and Helodrilus (Bimastus) parvus (Eisen) in collections received from the Indian Museum; and I had a Limnodrilus, species unrecognizable, sent to me from one of the high lakes (9). Thus the only identified species of Oligochaeta from this region are the three recorded by Michaelsen. ; This is perhaps surprising, seeing that Kashmir is a favourite summer resort, and is visited annually by large numbers of travellers from all parts of India, and indeed from other parts of the world also. There are possibly two reasons for the paucity of the collections. One is that Kashmir-is a holiday country, and zoologists who visit it are doubtless concerned rather in providing for themselves a change of interests than in pursuing their usual occupation ;—at least this -has been the case with myself. The other is that it has been recognized that Kashmir belongs to the Palaearctic region, and not to the Oriental, which is of greater interest to Indian naturalists. The present small collection contaits only three identifiable species, of which two have been previously recorded, while one is new. ‘These all belong to the Lumbricinae, a Palaearctic group. THE RANGE OF THE LUMBRICINAE. The Lumbricinae are a recently evolved and dominant group of earthworms, which possesses great powers of adaptation to new surroundings, and of which numerous species have been carried by man and have established themselves all over the world. The occurrence of these peregrine species of Lumbricinae gives no clue, therefore, to the zoogeographical affinities of the region where they are found ; and since Beddard’s and Michaelsen’s records are entirely of these ‘‘ world-wanderers,’’ it would have been permissible to regard Kashmir as possessing no proper earthworm fauna of its own, and therefore as not to be included in the territory of any particular family or group of Oligochaeta. Some little time ago, however, I described a new species of Lumbricine from Murree in the Himalayas (10), a few miles only from the southern border of Kashmir. In addition, one of the 1922.) J. STEPHENSON: Earthworms from Kashmir & India. 429 species here recorded from Kashmir appears to be new, and therefore possibly endemic. These justify the inclusion of this region, at least provisionally, in the territory of the Lumbricinae. Simla, the summer capital of India, and the surrounding area in the W. Himalayas, swarm, as might be expected, with peregrine Lumbricines, from which no zoogeographical conclusions can be drawn. But here too a new and possibly endemic Lumbricine is now found to occur (Helodrilus (Dendrobaena) kempi, v. infra) ; the W. Himalayas may thus probably be added to the proper territory of the Lumbricinae. The extreme outpost of the sub- family appears to be Calcutta, whence Helodrilus (Bimastus) ° indicus was described some years ago by Michaelsen (3). These four species are the only endemic Lumbricinae so far known in India. THE PoLYPHYLETIC ORIGIN OF THE GENUS MEGASCOLEX., That the genus Megascolex is polyphyletic, i.e. that different species of the genus kave arisen in different places, at different times, and from different ancestors, is already recognized. Mi- chaelsen (5) has pointed out the close relation of certain S. Indian species of Megascolex to certain S. Indian species of Notoscolex (the group of Megascolex travancorensis to that of Notoscolex ponmudianus), and argues that these species of Megascolex have in all probability arisen from the local representatives of Notoscolex. There is also a similar correspondence between species of Notoscolex and species of Megascolex in another restricted area, the N. Island of New Zealand; here, too, the inference is that the latter have arisen from the former. I have shown (11) that a worm which is by definition a Megascolex has descended from some species of Perionyx, in a region more than a thousand miles away from the proper Indian Megascolex territory ; and have in the same paper given some reason for thinking that certain species of Megascolex may be descended from still another genus Spenceriella. Perhaps the clearest case of an independent origin of a species of Megascolex, however, is that of Megascolex horat, described below. The worm has certain remarkable peculiarities; it is unique in the genus in having the male pores on segment xvii, one segment in front of the normal position for the genus; and both male and female organs are found, on dissection, to be shifted one segment forwards. There are also well marked and stalked calciferous glands in segments x, xi and xii; such stalked calci- ferous glands scarcely occur elsewhere amongst the Indian species of Megascolex. ; Now a small group of species of Notoscolex has exactly these same characters; and here too the species (Notoscolex oneilli, stewartt, and striatus) are peculiar in these respects in their genus. The species of Notoscolex wete found in the Abor country in the Assamese Himalayas ; Megascolex horat was taken at Cherrapunji, also in Assam (though not in the Himalayas), more than a thousand 430 Records of the Indian Museum. [VoL. XXIV, miles from the proper Indian Megascolex region. ‘There can be no doubt that while the majority of species of Megascolex have arisen from Notoscolex elsewhere, this species has had an independent origin from a local species of Notoscolex in Assam. SOME OTHER SPECIES OF INTEREST. A new Drawida and a new Perionyx are also to be recorded from Assam, and a new Eudichogaster from Bombay. It has been interesting to rediscover Eutyphoeus ortentalis (Bedd.) with its peculiar penial setae, and Octochaetus beatrix Bedd., in which I have found small penial setae, previously over- looked. Both the new species of Lumbricinae illustrate the fact, brought out by Michaelsen (4), that the subgenera of Helodrilus may run into each other, and have no sharp limits. Thus one of the species, which I assign to the subgenus Allolobophora agrees with Bimastus in having no spermathecae; while the other, which I place in Dendrobaena, agrees with Allolobophora in the characters of the seminal vesicles. Family MONILIGASTRIDAE. Genus Drawida Mich. Drawida nepalensis Mich. Dehra Dun, compound of the Forest Research Institute (serial No. 111 of 27'vili"1921). Dorsal pores are usually absent in the genus Drawida, but I have previously noted (10) that in this species vestiges are present, in the form of gaps in the muscular layers of the body-wall. These vestiges were very obvious in the present specimens, though there were no actual perforations. They occurred from furrow 4/5 onwards. . ae Drawida rosea, sp. nov. Cherrapunji, Assam; under stones and in muddy pools around Dak Bungalow. S.L. Hora. 28°x'1g21. A single specimen. ~ External Characters:—length 102 mm. Maximum diameter 3mm. Segments 149. Colour grey, but with a faint pinkish tinge dorsally. Prostomium prolobous. Dorsal pores absent, but vestiges are visible in the middle of the body. . Nephridiopores in the line of the lateral setae. The setae are closely paired; aa==4/5 bc; dd=4/7 of the circumference. ‘The setae begin on segment ii. The clitellum embraces segments x-xiii, but xiv and perhaps ix are slightly altered. 1 On the fusion of the genera Notoscolex and Megascolex by Michaelsen cf. Michaelsen (6) and Stephenson (11). 1922.] J. STEPHENSON: Earthworms from Kashmir & India. 431 The male pores are situated on small somewhat irregular transversely elongated papillae at the hinder border of segm. x; they are immediately outside—almost in—the line of setae BD. Immediately behind the papillae of the male pores, on the anterior part of segm. xi, are a pair of smaller and rounder papillae. The midventral region between the four papillae is somewhat sunken, and darker in colour. The female pores are minute, in furrow 11/12, in line with setae ab. The spermathecal pores are conspicuous, with swollen margins, in furrow 7/8, between the lines of setae ab and cd, but nearer the latter. The upper end of the pore reaches the line cd. Internal Anatomy :—-Septa 5/6-8/9 are much thickened. There are four gizzards, in segm. xiii-xvi, all well developed. The last heart is in segm. ix. The testis sacs are large, elongated in shape, and extend into segm. ix, though their greater partis in x; they reach as far back as septum 10/11, and are slightly constricted at septum 9/r0. The vas deferens lies on the posterior face of septum 9/10; it is narrow, and thrown into numerous coils. The prostates are elongated, cylindrical, and bent on them- selves; the ental end is rather thicker, and there is no separate duct ; the surface is soft and “glandular.’”’ The vas deferens enters near the ental end. Segm. xi is narrow antero-posteriorly, but there is no ovarian chamber ; segm. xi is fully opened up on opening the body in the usual way. The segment was full of genital products and the female organs were not separately distinguishable. The ovisacs are large, stoutly and irregularly ovoid in shape, and extend back to septum 13/14. The spermathecal ampulla is spherical; the duct forms coils on the posterior face of the septum. The atrium is large—not very much smaller than the ampulla; it is a pear-shaped sac, the lower and narrower portion marked by a number of slight annular constrictions, and prolonged into a narrow, bent, duct-like tube, joined at its termination by the spermathecal duct. Remarks:—The relations of this species seem to be to D. nepalensis and papillifer. Family MEGASCOLECIDAE. Subfamily MEGASCOLECINAE. Genus Megascolex Templeton. Megascolex konkanensis Fedarb. Bombay. Coll. Prof. J. P. Mullan. The actual male pores are not usually visible; in these speci- mens they appear to be on the transverse ridge which runs across the male area on each side, nearer the outer than the inner margin of the area, 432 Records of the Indian Museum. (VoL. XXIV, Megascolex mauritii (Kinb.), Bombay. Coll. Prof. J. P. Mullan. Ahmedabad. Coll. Prof. J. J. Asana. Megascolex horai, sp. nov. Cherrapunji, Assam; under stones and in muddy pools around Dak Bungalow. 28°x'1921, S. L. Hora. A single specimen. External Characters :—I,ength 110 mm. Diameter 2°5 mm. Segments 188. Colour yellowish grey, no difference between dor- sal and ventral surfaces. A long thin worm. Prostomium slightly epilobous (?). Dorsal pores begin in furrow 10/11. The setae are disposed in rings; they are of fair size, and form fairly regular longitudinal lines. In front of the male pores the ventral break is equal to 3-4 ab; in the middle and hinder parts of the body to 24-3 ab. The dorsal break is equal to 2-3 yz anteriorly, but is much smaller behind the genital region (13-14 yz). The following numbers were counted: v/26, ix/27, xii/32, xix/32, and in the middle of the body 28. The specimen was apparently not fully mature. There was no clitellum, and no genital papillae or markings of any kind. The male pores are on segm. xvii, on papillae, in line with setae b, about one-fifth of the circumference apart. The female apertures are not visible. The spermathecal pores are in furrows 6/7 and 7/8, in or just internal to the line of b, about one-fifth of the circumference apart. Internal Anatomy :—Septum 4/5 is thin, 5/6 is very thin, 6/7 is thin and attached to the ventral body-wall behind the normal site of insertion,—on the left side nearly at the level of furrow 7/8; the following septa as far as 13/14 are perhaps slightly strengthened, but not much; 7/8 is displaced backwards in the » same way as 6/7, 8/9 is attached slightly behind its normal line of insertion, but the rest are not displaced. All the septa as far back as 12/13 are strongly convex backwards. The gizzard is in segm. vi, firm and barrel-shaped. Calcareous glands are present in segms. x, xi and xii; they are of moderate size, are stalked, and the margins may be lobul- ated. The intestine ‘begins i in xv. The last heart appears to be in segm. xii. There are tufted nephridia in segm. v. In the body generally the micronephridia are arranged in a single transverse row in each segment. Testes and funnels are present in segms. ix and x. Seminal vesicles are found in xi only; they were small in size in the present specimen. The prostates are small—perhaps not fully developed; they are lobular and deeply bifid on the outer border. The duct is bent round sharply at its ectal end. Ovaries and perhaps funnels were seen in segm. Xii. 1922.] J. StepHENSON: Earthworms from Kashmir & India. 433 The spermathecae are small sacs sessile on the body-wall, without duct. There is asingle diverticulum, narrow and tubular, as long as or not quite so long as the ampulla; it arises from the inner side of the sac where it joins the body-wall. There are no penial setae. Genus Pheretima Kirb. em. Mich. Pheretima elongata (E. Perr.). Bombay. Coll. Prof. J. P. Mullan. Pheretima hawayana (Rosa). Dehra Dun; compound of the Forest Research Institute (serial no. 111 of 27'vili*1921). Pheretima heterochaeta (Mich.). Sariya Tal, about three miles from Naini Tal to the west, and at a lower level than Naini Tal; from underneath stone on the banks of the lake. A single specimen. Dr. B, Prashad. Cherrapunji, Assam ; under stones and in muddy pools around Dak Bungalow. 28°x'1921. S. L. Hora. Several specimens. Two of the latter batch of specimens showed certain abnor- malities, which led to my examining them more closely. In one specimen, the number of segments counted on the dorsal surface was one greater than those seen ventrally; this was due to the presence of a spiral groove in the anterior part of the body. There were other spiral abnormalities further back, behind the genital region. The clitellum: was incomplete on the tight side from above the lateral line of the body to the midven- tral line; it embraced segments xiv-xvi. ‘The male pore was on Xviii on the left side, and on xx on the right. There was a papilla midventrally situated on xvii. Internally, and reckoning by the segments seen on the dorsal surface (the above appearances are described as seen from the ventral surface), the spermathecae are situated between segms. 6/7, 7/8, 8/9 and 9/10; that in 7/8 on. the right side has a double ampulla with a single duct and diver- ticulum ; the last heart is in segm. xiv. Seminal vesicles are present in xii, xiii and xiv on the right side, and in xii and xiii on the left. Ovaries and their funnels are present in xiv on the left, and in xiv, xv, xvi and xvii on the right. The prostate of the right side is absent ; the male duct ends in xxi. In the other specimen there were two male pores on the right side, on segms. xviii and xix, each smaller than normal. Internally, the prostate was small on the left side, and there was none on the right. On the right side, both vasa deferentia were continued into the anterior prostatic duct ; there was a second prostatic duct in xix, but unconnected with the vasa deferentia. The right anterior prostatic duct was slightly stouter than the one behind it. 434 Records of the Indian Museum. (Vor. XXIV, Pheretima houlleti (E. Perrier). Bombay. Coll. Prof. J. P. Mullan. Dehra Dun; compound of Forest Research Institute (serial No. 111 of 27‘vill’1921). Cherrapunji, Assam; under stones and in muddy pools around Nak Bungalow. 28°x‘1921. S. |. Hora. Two specimens, Pheretima posthuma (I). Vaill.). Dehra Dun; compound of Forest Research Institute (serial No. 111 of 27'Viil' 1921). Pheretima suctoria Mich. Bombay. Coll, Prof. J. P. Mullan. As the species has been met with only once previously, and as the present specimen shows a few variations from the descrip- tion given by Michaelsen (3), I. add the following notes :— Length 205 mm. Diameter 6 mm. Colour a dark brown dorsally, paler ventrally. Prostomium small, epilobous 4/5; the grooves at the sides of the prostomium are hardly different from the numerous other longitudinal grooves round the mouth on the first segment ; the backwardly extending process (tongue) of the prostomium not cut off by a groove behind. Dorsal pores from 13/14; perhaps a rudimentary pore in 12/13,—a deepening of the intersegmental groove in the middorsal line. The setae are larger on the anterior segments (ij—vi). The characteristic male area may be described as follows :— On segment xviii, taking up the whole length of the segment, are a pair of raised circular dise-like areas with well defined margins ; the interval between these discs is less than the diameter of one of them, and shows seven setae intervening. There are also one or two setae on the inner and outer edges of the discs,—i.e. the setal ring is continued a little way into the discs at each side, The male pore is situated at the centre of the disc on a tiny papilla; and a faint ridge runs transversely across each disc in the line of the setae and of the male pore; behind the ridge, and also transverse in direction, is a slight depression. The discs are light in colour. Michaelsen places the male pores at the outer border of the discs, about one-third of the circumference apart. This did not seem to be the case here; and from internal examination also the male apertures cannot be so far apart as that,—scarcely, I think, as much as one-fourth of the circumference apart. The female pore appeared to be single, in a small depression. Septum 4/5 was slightly thickened ; 5/6, 6/7 and 7/8 were very stout; 10/11 was moderately thick, 11/12 and 12/13 decreas- ingly so. The testis sacs were large, and came up laterally round the alimentary canal on each side, leaving the dorsal surface of the gut uncovered. In segm. xi the sacs enclose the hearts and cover in the seminal vesicles also ; in x they enclose the hearts (though 1922.] J. StepHENSON: Earthworms from Kashmir & India. 435 at first sight this appeats not to be the case; but the large dark vessel running superficial to the sac is not the proper heart). The prostatic duct is looped or coiled ; it is thin at first, but becomes stout towards its ectal end. I found no ovisacs. The spermathecae differ from those of Michaelsen’s specimens (fig. 1). The ampulla is ovoid ; the duct, nearly as long as the ampulla, is narrow at first, then swells and becomes shining and firm. The diverticulum is as long as the ampulla and duct together; it arises from the ectal end of the duct, and is narrow and tubular; its inner (ental) portion consists of a number of short a closely adpressed loops and has a crenulat- % 5 ed appearance; the ectal portion becomes fig, 1.—Phevetima suc- smooth and shining towards its termina- toria ; spermatheca. tion. Genus Perionyx E. Perr. Perionyx excavatus E. Perr. Mashobra, Simla Hill States. 13:vit1g2t. S. .. Hora. ‘Two speci- mens and a few fragments. Below Kufri, Simla Hill States; near stream, under stones and in moss. 28'ix'1g21. Dr. S. W. Kemp Four specimens, two quite immature. Perionyx modestus, sp. nov. Cherrapunji, Assam; under stones and in muddy pools, around Dak Bungalow. 28ix'1921. S.L. Hora, A number of specimens. External Characters:—A long specimen measures 167 mm., but sexual specimens are found down to 85mm. Diameter, max. 4 mm. Segments 174. Colour deep purple dorsally ; lighter, of a violet tint, ventrally. The body is somewhat flattened dorso- ventrally. Prostomium epilobous $ ; tongue not closed behind. Dorsal pores begin from furrow 4/5. The setae are in rings, and are more closely set ventrally. The dorsal break is either absent or very small (zz—=ca. 1+ yz); the ventral break is absent or very small behind the genital region, and small (ca. 14 ab) in front of the genital region. The following numbers were counted: v/ ca. 38, ix/41, xti/40, xix/42, and 42 in the middle of the body. There was no clitellum to be seen in any of the specimens. The male pores are on segm. xviii, the anterior and posterior borders of which are bowed forwards and backwards respectively. The segment presents a transverse groove, usually shallow but occasionally deep. The pores are short longitudinally placed slits at the ends of the groove; they are not far from the midven- tral line, in line with about the fourth seta on each side. The female pores were not visible. 436 Records of the Indian Museum. [VoL. XXIV, The spermathecal apertures are in furrows 7/8 and 8/9, near together, about in line with the third seta on each side. There are no other genital marks, Internal Anatomy :—Septa 6/7 and 7/8 are slightly thickened. 8/9 and 9/10 moderately so. The gizzard is vestigial, in segment v. There are no calci- ferous glands; but the oesophagus is dark in colour, with trans- verse vascular striations, in segms. xii and xiii. The intestine begins to widen out behind the prostates. The last heart is in segm. xiii. The nephridia all end in the same line. Testes and funnels are free in segms, x and xi. Seminal vesicles are present in segms. xi and xii; they fill the length of their segments, and are apposed to their fellows in the middorsal line. Prostates are present in segm. xviii, but are very small; the duct is stout in relation to the size of the gland, is muscular and shining, and almost straight. Ovaries and funnels are present in segm. xiii. The spermathecae, in segms. viii and ix, are small elongated sacs, with no distinguishable duct, and a minute wartlike diverti- culum near their base (not present in all), There are no penial setae. Subfamily OCTOCHAETINAE. Genus Octochaetus Bedd. Octochaetus beatrix Bedd. Bombay. Coll. Prof. J. P. Mullan. The original description of this species was given by Beddard (2) in 1902. In 1914 I described as a new species a worm which I called Octochaetus dasi (8), but I now believe that this is identi- cal with Beddard’s species. The following notes fill in a few gaps in our knowledge. Prostomium epilobous 4; the tongue is very narrow, and not cut off behind. The setae in the present specimen were spaced as follows :— in the middle of the body, and behind the genital region ab=1/3 aa=2/5 be=2/3 cd; on seg. ix ab=} aa=4 bc=2/3 cd; dd=2/3 of the circumference. The male area is a somewhat quadrilateral depression with rounded angles, small, rather deep, midventrally on segms. xvii-— xix ; it indents the posterior border of the clitellum, and is com- prised within the lines of setae a. Small papillae are seen in the line of 6, on the borders of the depression, in segms. xvili and xix, perhaps the porophores of the male and posterior prosta- tic apertures respectively ; there is not distinct papilia or aperture on xvii. No seminal grooves were visible. 1922.] J. STEPHENSON: Earthworms from Kashmir & India. 437 The female pore or pores are indicated by a small midventral depression on the anterior part of segm. xiv. The spermathecal pores are two pairs, on minute papillae close to the middle line, the least trifle in front of the setal zones of segms. viii and ix. Septum 5/6 is somewhat strengthened ; the next is 8/9, slightly thickened ; 9/10, 10/11, and Ir/I2 are somewhat strengthened, 12/13 very slightly so, and the rest are thin. Previous statements regarding the absence of penial setae appear to be mistaken. There was very little to indicate their presence; but an endeavour to isolate and mount the sac was successful in revealing one. This was very small, 6 mm. long, and 13 thick at its middle ; the shaft has a slight double curve, and the tip ea is fairly sharply pointed; the ornamentation a consists of a few irregular indentations of the margin near the free end (fig. 2). No copula- pile hag cece nae tory setae from the spermathecal segments could oa xo. be obtained. Genus Erythraeodrilus Steph. Erythraeodrilus kempi var. bifoveatus (Steph.). Bombay. Coll. Prof. J. P. Mullan. I accept Michaelsen’s recent separation of this and allied species from the genus Hoplochaetella, and their union with Evythra- eodrilus (7). It now seems to me that the difference between the two species that I formerly described separately as kempi and bifoveata (10) is scarcely sufficiently marked to justify their being kept apart. Genus Eutyphoeus Mich. Eutyphoeus orientalis (Bedd.). Dehra Dun; compound of Forest Research Institute (serial No, 111 of 27'Vili'1921). This species, obtained from near Calcutta by Beddard and from Dehra Dun by Fedarh, has not been seen since 1898. It is interesting to find it now in a batch of worms from Dehra Dun, one of the previously recorded localities. The following details may be noted :— On segm. xvii are a pair of grooves or cracks, shaped like square brackets —[ ]— overhung on their outer side by a thick- ened ridge; the male pores, with penial setae projecting, are in the posterior corners of the brackets. The longitudinal part of the eee and the pores themselves, are a little outside the line of setae 0. 438 Records of the Indian Museum. [VoL. XXIV, The spermathecal apertures are transverse slits with their centre between b and c, but nearer c,—the outer end of the slit reaches the line of c. The spermathecae are somewhat ovoid sacs; the duct is very short and stout, from the under sur- face of the ampulla; the margin of the ampulla is crenated. The diverticula are two, attached to the beginning of the duct and rather on its posterior side; they appear to have one, two, or three small chambers. The peculiar penial setae are 25 mm. long. and 26, thick in the middle; the shaft is almost straight, and the tip is bluntly pointed. The oblique markings, closely set along the borders of the distal end of the seta, are interpreted by ee Beddard as ‘“‘chevron-shaped ridges”’ ; Fic. 3.—Eutyphoeus orien- but the appearance is almost as if there talis; penial setae: X300. were a cleavage along the oblique lines (fig. 3). Eutyphoeus waltoni Mich. Dehra Dun; compound of Forest Research Institute (serial No. 111 of 27'Vii'1921). Genus Eudichogaster Mich. Eudichogaster mullani, sp. nov. Bombay. Coll. Prof. J. P. Mullan. External Characters :—Uength 134 mm. Diameter 6 mm. Segments 200. Colour a light and even grey, no difference be- tween dorsal and ventral surfaces. Anterior end rather bulbous. Secondary annulation on the anterior segments; iv and v biannu- lar, vi triannular, vii and onwards to the clitellum with four, five or even more annuli. Prostomium small and prolabuEee a median dorsal groove divides segm. i throughout its length. Dorsal pores very small, the first in furrow 12/13; perhaps a small or rudimentary pore in 11/12. Setae are not visible in segments ii-iv, and only a few are seen in v and vi. In the middle of the body ab=2/7 aa=2/5 be=2/3 cd, and dd=ca. 4/7 of the circumference; behind the genital region ab = } aa=1/3 be = 4/7 cd, while dd = 2/3 of the circumference ; in the anterior segments the ratios are about the same as these last. The clitellum is not distinctly developed, but penne extends over $xiii—}xvii. 1922.] J. SterHENSON: Earthworms from Kashmir & India. 439 The midventral region of segments xvii—xix is depressed, the depression extending laterally from the line of setae 6 on one side to an equivalent extent on the other; in the bottom of the depression is an irregular slightly raised rough patch. The pro- static pores are perhaps on four small papillae at the angles of the depression, in or very slightly outside the line of setae 0, and very slightly anterior to the setal zone ‘of xvii and very slightly pos- terior to that of xix respectively. On the anterior border and behind the posterior border of the rough patch in the depressed area, in other words posteriorly on segments xvii and xix respectively, and in the midventral line, are two papillae of small size, each appearing to have a pore in its centre. The female pores were not seen. On segment viii is a roughened patch, slightly elevated and extended in a transverse direction on each side to a little beyond the line of seta d; it is narrow antero-posteriorly, and does not embrace the anterior two-fifths of the segment, nor, except be- tween the lines of setae b and c, the posterior fifth either. Both pairs of setae are thus included in the patch. The spermathecal apertures are not visible, but from internal examination they are two pairs, behind furrows 7/8 and 8/9, but slightly in front of the setal zones of segments viii and ix, between the lines of setae b and c, but nearer to that of b. Internal Anatomy:—Septum 4/5 is thin, 5/6 and all succeed- ing septa as far as 10/11 are moderately strong, 8/9 and g/10 being the thickest of the series; 11/12 is somewhat thickened, and the rest are thin. The gizzard in segment v is large, spherical, and very firm; that in vi is rather smaller. Calciferous glands are present in segments xi and xii; they are dark in colour, ovoid or kidney- shaped, well set off from the gut, not stalked but attached by one edge. The intestine begins in segm. xv. The last heart is in segm. xii. Behind the genital region the micronephridia are arranged in a transverse row in each segment, about nine on each side; there is no marked difference in size, the most internal being a little smaller than the rest and a little closer together. At the hinder end of the body the arrangement is much the same; there are about seven nephridia on each side, and the innermost of the series is scarcely larger than the rest—a little larger than the one next to it. Testes and funnels are free in segms. x and xi. Seminal vesicles are present in segms. ix, x and xii; in xii they are small, in ix smaller still, and in x there was one only, on the right side, and this was quite minute. The prostates, in segms. xvii and xix, are small; the gland- ular part is disposed in a few loose coils or loops ; the duct is thin, of the same diameter as the glandular part, but muscular and shining. 440 Records of the Indian Museum. [Vo,. XXIV, The spermathecae are in segments viii and ix; the ampulla is small and ovoid; the duct is short, and relatively wide. The diverticulum is a small wart-like swelling on the side of the duct. Copulatory setae are found on segm. viii, in the site of the ventral bundles. In length, measured across the bend, they are ‘7 mm. Or more, and _ their thickness at the mid- Fic. 4.—Eudichogaster mullani ; copulatory seta ; dle is 164. The distal a, whole seta: x50; 8, tip of seta: x 300. half is either curved through a quarter of a eiecies or bent and twisted more irregularly. The tip ends in a blunt. point ; there is no ornamentation (fig. 4). Family I.UMBRICIDAE. Subfamily GLOSSOSCOLECINAE. Genus Pontoscolex Schmarda. Pontoscolex corethrurus (Fr. Miull.). Ahmedabad. Coll. J. J. Asana. Several specimens. Subfamily LUMBRICINAE. Genus Helodrilus Hoffm. Helodrilus (Allolobophora) caliginosus subsp. trapezoides (Ant. Dug.). Sariya Tal, about three miles to the west of Nein Tal, and at a lower level than Naini Tal ; from underneath stone on the banks of the lake. Dr. B. Prashad. ‘Two specimens, one mature, one immature. Sukla Tal, almost a mile to the west of Naini Tal; 7000 ft.; from the margins of the lake. Dr. B. Prashad. Several specimens. Gandarbal, Kashmir; ponds in the course of a shallow irrigation streamlet. Ca.6000 ft. 14:vit1g2t. Dr. B. Prashad. Two specimens. Anchar Lake, Kashmir (an extensive marshy and weedy area in the course of the Sind River). 29'vi-1921. Dr. B. Prashad. A number of specimens. Helodrilus (Allolobophora) prashadi, sp. nov. Gandarbal, Kashmir; ponds in the course of a shallow irrigation streamlet. Ca. 6000 ft. Iy4'vitr921. Dr. B. Prashad. Several specimens, External Characters :—\Length 62mm. Diameter 3mm. Geg- ments 133. Colour grey, with a slightly pinkish tinge. _ Prostomium proepilobous. 1922.] J. STEPHENSON: Earthworms from Kashmir & India. 441 Dorsal pores begin in furrow 4/5 or 5/6. The setae are closely paired; aa is nearly twice bc; ab is greater than cd; dd is less than half the circumference. The clitellum extends from part of segment xxiii or segment xxiv to xxxii or xxxiii (— 9 to more than 10). The clitellum is saddle-shaped, and the clitellar region is swollen, and flattened ventrally. ‘‘ Ridges’’ or “walls’’ are present on segms. xxix- xxxi; and the ventral setae in these segments are implanted on minute papillae. The male pores are seated on vety prominent papillae, hemi- spherical in shape on xv and encroaching also on xiv and xvi; the centres of the papillae are just outside the line of setae b. Female pores were not visible. Spermathecal pores are absent. The ventral setae of segment xii, and sometimes those of segments xi and x, are situated on papillae. Internal Anatomy :—Septum 5/6 is thin, 6/7, 7/8 and 8/9 are much thickened, 9/ro is fairly thick, and succeeding septa as far as 13/14 gradually diminish in thickness ; the rest are thin. The gizzard comprises two segments, xvii and xviii. Testes and funnels are free in segms. x and xi. Seminal vesicles are present in ix, x, xi and xii; those in x are the smallest, though they are not much sinaller than those in ix; those in xi and xii have a nodular surface, indeed they are almost racemose, being composed of small spherical lobules. Spermathecae are absent. The lateral setae of segments xi and xii are seen on internal dissection to be contained in large sacs. On examination they are found to be ‘76 mm. long, and almost straight ; they are fairly sharply pointed, and the distal portion is grooved as in the clitellar setae of Lumbricus terresiris. Remarks :—This species disagrees with the great majority of the subgenus, and resembles Bimastus, in having no spermathecae. This peculiarity has been recorded by Michaelsen in the case of H. (A.) agatschtensis (4). Helodrilus (Dendrobaena) kempi, sp. nov. Kufri, Simla Hill States; 7800 ft. Oct. 1921. Dr. S. W. Kemp. Two specimens, one mature. External Characters :—lLength 91mm. Diameter6mm. Seg- ments 128. Nonpigmented, light grey in colour. Prostomium epilobous }$, the tongue not cut off behind. Dorsal pores from furrow 9/ro. The setae are small; anteriorly ab = 1/3 aa = tbc = cd; behind the male apertures the intervals between the individuals of a pair begin to widen, and behind the clitellum this separation becomes greater still, so that in the middle of the body the setae are no longer paired. Here ab = 4-3/5 aa = 1}-14 bc = 2 cd or nearly ; dd = 1/3-2/5 of circumference. 442 Records of the Indian Museum. [VoL. XXIV, The nephridiopores are just above the line of setae b. The clitellum is saddle-shaped, and extends over segms. xxix— xxxiv (= 6). The ridges of puberty are indistinct, and appear to be coextensive with the clitellum, or perhaps exclude the last segment and half of the first. The male pores are on segm. xv; they appear as transverse slits which extend from the line of setae 6 to that of c, with tumid anterior and posterior lips which cause the limits of seg- ment xv to bulge forwards and backwards. The spermathecal apertures are in furrows 9/10 and Io/I1, in the line of setae d, Internal Anatomy :—Septum 4/5 is thin, 5/6 - 15/16 are thick- ened, 6/7—8/9 most so, the rest only slightly. The gizzard takes up two segments; xvii and xviii. There are oesophageal pouches in segm. x; the calciferous glands in xi form large lateral widenings of the oesophagus which, however, are not set off from the tube; they are continued back, but are less prominent, in xii, and are not distinguishable behind this. The last heart is in xii, but this is much smaller than the one in xi, and is at a deeper level. Testes and funnels are free, in segms. x and xi. There are four pairs of seminal vesicles, in segms. ix—xii, all of quite moderate size ; those of segm. x are equal in size to those of ix. Spermathecae are present in segms, x and xi as small round sacs sessile on the body-wall. The ventral setae of segm. xv are slightly iG. 5.—Helodvilus(Den- modified. The points are apparently soft- drobaena) kempi; geni- ened; the characteristic feature is a faint tal seta from segm. XV? Souipturing of the distal portion of the X Ca. 200. 0 shaft by a numerous series of transverse markings, slightly jagged and convex towards the insertion of the seta (fig. 5). Helodrifus (Bimastus) constrictus (Rosa). Mashobra, Simla Hill States. 13°vitig21. S. L.. Hora. A single specimen, incomplete behind. Helodrilus (Bimastus) parvus (Hisen). Sulha Lal, almost a mile to the west of Naini Tal; ca. 7ooo it. From the margins of the lake. Dr. B. Prashad. Five specimens. Chenar Bagh Nullah (a very shallow slow-running stream with a sandy, and inuddy bottom), Srinagar, Kashmir. 6-8'vit1g21. Dr. Prashad. A single specimen, 1922.] J. SrePHENSON: Earthworms from Kashmir & India. 443 Genus Octolasium Orley em. Rosa. Octolasium lacteum (Orley). Mashobra, Simla Hill States. 13*vit1921. S. L. Hora. ‘Two speci- mens, immature, probably belonging to the above species. REFERENCES TO LITERATURE. (1) Alcock, A. W. Report on the Natural History Results of the Pamir Boundary Commission. Calcutta, 1898. (2) Beddard, F. E. On two new Earthworms of the family Megascolecidae. Ann. Mag. Nat. Hist., ser. 9, Vol. IX, Igoz2. (3) Michaelsen, W. ‘The Oligochaeta of India, Nepal, Ceylon, Burma and the Andaman Islands. Mem. Ind. Mus., Vol. I, 1909. (4) Michaelsen, W. Zur Kenntnis der Lumbriciden und ihrer Verbreitung. Ann. Mus. Zool. Acad. Imp. Sct. St.- Petersbourg, Vol. XV, 1910. (5) Michaelsen, W. Oligochaten von Travancore und Borneo. Mitt. Nat.-Hist. Mus. Hamburg, Vol. XXX (Beiheft 2), 1913. (6) Michaelsen, W. Oligochaten in: Results of Dr. E. Mjo- berg’s Swedish Scientific Expeditions to Australia 1910- 1913, pt. xiii. Stockholm, 1916. (7) Michaelsen, W. Oligochaten vom westlichen Vorderindien und ihre Beziehungen zur Oligochatenfauna von Mada- gascar und den Seychellen. Mitt. Zool. Mus. Ham- burg (Betheft z. Jahrbuch der Hamb. Wiss. Anst., Vol. XXXVIII, 1920). (8) Stephenson, J. On a Collection of Oligochaeta, mainly from Northern India. Rec. Ind. Mus., Vol. X, 1914. (9) Stephenson, J. On a Collection of Oligochaeta belonging to the Indian Museum. Rec. Ind. Mus., Vol. XII, 1916. (10) Stephenson, J. On a Collection of Oligochaeta from var- ious parts of India and Further India. Rec. Ind. Mus., Vol. XIII, 1917. (11) Stephenson, J. Contributions to the Morphology, Class- ification, and Zoogeography of Indian Oligochaeta. Proc. Zool. Soc. Lond., 1921. SF ONE SY YO INDIAN MYSIDACEA. By WALTER M. TATTERSALL, D.Sc., Keeper of the Manchester Museum. The following report deals with collections of Mysidacea sent to me from the Indian Museum for identification. The greater part of them were collected by Dr. S. W. Kemp in three localities, Kilakarai and Pamban, at the northern end of the Gulf of Manaar and at Port Blair in the Andaman Islands. At the latter place Dr. Kemp made collections at eight different stations and in order to save repetition I give here a list of these stations with full particulars, merely giving the station number under each species. Port BLair, ANDAMAN ISLANDS. List of stations at which Mysidacea were obtained. St. 3. I9-ii-15. North Bay, ca. 1 fm. Muddy sand. St. 5. 2I-ii-15. Brigade Creek, 2-5 fms. Bottom composed of mud with much decaying vegetable matter. St. 7. 22-ii-15. Off jetty, Ross I., 13-2 fms. Sand and a little weed. St. 8. 23-ii—1o-iii-15. Between Ross I. and Aberdeen, 2-10 fms. Bottom very varied, principally sand, weed, * small living corals, dead coral fragments and sponge. St. 11. 27-ii-15. Channel N. of Viper I., 1-2 fms. Mud. St. 13. 1-iii-15. Reef at N. end of Ross I., shore collecting. Dead coral. St. 19. 7-iii-15. Semiramis Bay and off Perseverance Pt. Fine mud. St. 21. I1-iii-15. Mid-channel, N. W. of Ross I., 10-12 fms. Muddy sand. St. 32. Feb., March, 1921. Ross Channel, 2-9 fathoms. Full particulars of all other localities are given under each species, Our knowledge of the Mysidacea of Indian waters has until within the last ten years been of the scantiest nature. Wood- Mason, Alcock and Anderson recorded the following deep-sea and bathy pelagic species from the collections of the ‘‘ Investigator ’’ :— Gnathophausia calcarata G. O. Gnathophausia zoéa Will.-Suhm. Sars. (=G. bengalensis Wood- (= G. sarst Wood-Mason.) Mason.) Eucopta australis Dana. Gnathophausia gracilis Will.- Eucopia sculpticauda Faxon. Suhm. (=—G. brevispinis Petalophthalmus armiger Will.- Wood-Mason and Alcock ) Suhm. 446 Records of the Indian Musewm. [VoL. XXIV, In 1906 I recorded Sivella paulsonm Kossmann from Ceylon and between 1908 and 1914 I have described six other species of Mysidae from the brackish waters of the coast of India :— Rhopalophthalmus egregius Han- Potamomysis assimilis Tatter- sen. sall. Gastrosaccus muticus Tattersall. Indomysis annandalei ‘Tatter- Gastrosaccus simulans ‘Tatter- sall. sall. Mesopodopsis orientalis (‘Tatter- sall). In the course of his report on the Siboga Mysidacea Hansen (1910) records the following species from the Bay of Bengal :— Siriella gracilis Dana. Gastrosaccus bengalensts Han- Siriella aequiremis Hansen. sen, Hemisiriella parva Hansen. Pseudanchialina pusilla G. O. Anchialina grossa Hansen. Sars. Pseudanchialina inermis IMllig. Zimmer (1915 (3)) described the Mysidae collected by Dr. Duncker during a voyage from Ceylon to New Guinea. No exact localities are given for the following species, which may or may not have been taken in Indian waters :— Anchialina frontalis Zimmer. Leptomysis aptops G. O. Sars? Anchialina penicillaia Zimmer, Dioptromysis perspicillata Zim- Gastrosaccus bengalensis Han- mer. sen, Uromysis armata Hansen. Gastrosaccus dunckert Zimmer. Lycomysis pusilla Zimmer. Of these forms, Anchialina frontalis is, in my opinion, a syno- nym of A. grossa Hansen, Leptomysis apiops of L. xenops, sp. nov., described below, and Lycomysis pusilla of L. sbinicauda Hansen. Finally Colosi (1920) has recorded Doxomysis zimmert Colosi, from Ceylon. The total number of species of Mysidacea known from Indian waters is therefore 27. In the present report I record 38 species of which twelve have been recorded previously from India, sixteen are described as new to science and ten are new to the Indian fauna. These last species are :—- Lophogaster intermedius Wan- Siriella dubia Hansen. sen. Anchialina typica (Kroyer). Sirtella brevicaudata Paulson. Gastrosaccus pacificus Hansen. Sirtella quadrispinosa Hansen. Evythrops minuta Hansen. Siriella vulgaris Hansen. Hypererythrops spinifera (Han- Siriella affinis Hansen, sen.) The new species described below are :— Sirtella hansent. Mysidopsis indica. Gastrosaccus kempi. Mysidopsis kempt. Evythrops nana Leptomysis xenops. 1922.} W. M. TatrmrRsALL: Indian Mysidacea. 447 Afromysis macropsis. Neomysis hodgarit. Prionomysis stenolepis (gen. nov.). Idiomysts inermis (gen. nov.) Doxomysis anomala. Heteromysis proxima. Doxomysis littoralis. Heteromysis zeylanica. Neomysts indica. Heteromysts gymnura. The total number of species of Indian Mysidacea is therefore brought up to 53 species. Much remains to be done with the deep-water fauna of Indian waters and many deep-sea species will doubtless be added to the list. Extended knowledge of the distribution of the shallow-water forms is desirable, and when it is remembered that the majority of the species reported here were collected during two short expedi- tions only, the results, if continuous observation and collection were possible, are distinctly promising. The failure of earlier expeditions to tropical waters to obtain shallow-water Mysidacea is not due, as one was almost beginning to suspect, to the fact that these forms are absent from tropical waters, but entirely to a lack of knowledge of how to collect them. They are much smaller than the species from temperate and Arctic regions and easily pass through dredges and trawls. They require to be collected by means of special hand-nets made of mosquito netting used vigorously among the weeds on the shores below low-water mark. The results recorded below area testimony to the successful use of such means by Dr. Kemp. In examining this collection the most striking fact which presented itself was its strong Mediterranean facies. Out of twenty genera, no fewer than twelve are represented in the fauna of the Mediterranean, and I have frequently had to refer to Sars’ work on the Mediterranean Mysidae for the nearest described form to many of the new species noted here. Several of them, indeed, are so closely allied to Mediterranean species that it was only ne- cessary to referto Sars’ work and to tabulate the differences found in the Indian species. These facts will be more clearly brought out by a study of the following list in which are given the Indian forms and their Mediterranean allies :— Indian species. Mediterranean species. Lophogaster intermedius. L. typicus. 8 es S. norvegica. Siriella ya Shatucae Sis lynne S. jaltensis. Anchialina typica. } A. agilis. * grossa. Gastrosaccus duncreri. - muticus. G. sanctus, ” kempi. a pacifi gy G. normant ” bengalensts. ? ; Erythrops minuta. Erythrops serrata. 9 «6 NANG, 5. elegans. 448 Records of the Indian Museum. [Vor XXIV, Indian species. Mediterranean species. Mysidopsis indica. M. gibbosa. Leptomysis xenops. L. aptops. Mesopodopsis orientalis. M. slabberi. Neomysis indica. N. longicorms. Potamomysis assimilis. P. pengot. Heteromysis harpax. | H microps. The superficial resemblance between the Mysidacean fauna of the two regions is thus seen to be most striking and the fact is further emphasised if actual numbers are considered, for by far the greatest numbers of specimens belong to those species which are related to Mediterranean forms. My thanks are due to Dr. Annandale for the opportunity of examining and reporting on this collection and to Dr. Kemp for his successful efforts to obtain Mysidae at my request. I am greatly indebted to my wife for the figures illustrating this report. _ Suborder LOPHOGASTRIDA. Family LOPHOGASTRIDAE G. O. Sars. Genus Lophogaster M. Sars. Lophogaster intermedius Hansen. Lophogaster intermedius, Hansen, 1910, p- 14, pl. 1, figs.1a—t1e. Locality.—‘ Investigator? St. 532: Mergui Archipelago, 62 fathoms, 16-iv-13. 68 specimens, 10-20 mm. Distribution.—Only known from specimens captured by the ‘Siboga ’ in the waters of the East Indian Archipelago. Suborder Mysipa. Family MYSIDAE Dana. Subfamily STRIELLINAE Norman. Genus Siriella Dana. Siriella hanseni, sp. nov. Text-figs. Ia-c, 2. Locality.—Pamban, Gulf of Manaar, from weeds, o—2 fathoms, February, 1913. Sixtyspecimens, 4-7 mm. (Types.) Description.—A Siriella belonging to Hansen’s group I and allied to S. guadrispinosa Hansen and S. nodosa Hansen. Carapace in both sexes without protuberances or tubercles, hardly at all produced into a frontal plate, anterior margin broadly and evenly rounded, leaving exposed a small spiniform pseudo- rostral process, 1922.] W. M. Tarrersat.: Indian Mysidacea. 449 Antennal scale subequal in both sexes, reaching the distal end of the antennular peduncle in the female, not quite extending thus far in the male, three and a half times as long as broad, terminal lobe not quite so long as broad, slightly overreaching the terminal spine of the outer margin. Tarsus of the thoracic legs without a secondary joint. Pseudobranchial rami of the second to the fourth pleopods of the male spirally twisted, the terminal setae of the fourth pair not modified. Telson short, not extending to the distal end of the proximal TExT-FIG. 1.—Stviella hansent, sp. nov. a, anterior end of female; 4, antennal scale, c, third thoracic limb, endopod. All x 65.. joint of the exopod of the uropods, scarcely one and a half times as long as broad at the base, apex broadly rounded, almost truncate, and armed in the middle line with three equal small spinules and a pair of plumose setae longer than the spines flanking them, three large spines on the lateral margins of the telson at the base, distal portion of the lateral margins armed with about 12 spines gradually increasing towards the apex, the three spines on each side of the apex not much longer than the preceding spines and more or less subequal in size. Inner uropod shorter than the outer, its inner lower margins with about ten somewhat distantly placed spines without any 450 Records of the Indian Museum. [VoL. XXIV, small spines between them, extending from the anterior edge of the statocyst almost to the apex. Outer uropod with the proximal joint more than twice as long as the distal, its outer margin with three or four spines at the distal end only ; terminal joint one and a quarter times as long as broad. Remarks.—This species is distinguished by (i) the ab- sence of a rostral projection (ii) the unjointed tarsus of the thoracic limbs and (3) the size and armature of the telson and uropods. It is most closely allied to S. quadrispinosa and S. no- dosa, but the combination c of the three characters TEXT-FIG, 2.—Siriella hanseni, sp. nov. named will serve to distin- guish it from both these species. It :lso shows many points of resemblance to S. brevicaudata described below, but dif- fers in the relative size and the armature of the telson. Both species agree in the absence. of a rostral projection and the un- jointed tarsus of the thoracic legs. Telson and uropods: x 65. Siriella brevicaudata Paulson. Text-figs. 3a-h, 4a-}. Siviella brevicaudata, Paulson, 1875 (1), p. 30, pl. t, figs. 15-16. if ap Paulson, 1875 (2), p. 123, pl. xx, figs. 1a—m. rn rr Czerniavsky, 1882, p. 109. he ie Czerniavsky, 1883, p. 32. Localities —Kilakarai and Pamban, Gulf of Manaar, from weeds, 0-2 fathoms, February 12th-25th, 1913. Abundant, adult males and females, 6 mm. long. Remarks.—The rediscovery of this species, not recorded since Paulson originally described it in 1875 from specimens taken in the Red Sea, is a matter of great interest. Paulson’s original descrip- tion is in Russian and I am obliged to rely on his figures, but these specimens agree so closely with Paulson’s figures that I am con- fident of the correctness of my determination. In his ‘Siboga’ report (1910) Hansen arranges the Asiatic species of Szviella into four groups, but he does not include S. brevi- caudata in his list. It belongs to his group I and is specially distinguished in that group by the size and armature of the telson, its chief character being reflected in its specific name. 1922.] W. M. TATrERSALL: Indian Mystdacea. 451 It seems advisable to give a brief description of the species. Body moderately robust. Carapace short, leaving the last three thoracic segments exposed in the mid-dorsal line, and barely TExt-FIG. 3.—Siriella bvevicaudata Paulson. a, anterior end of female; 6, antennular peduncle of female; c, antennular peduncle of male; d, antennal scale and peduncle ; ¢, first thoracic limb; f, second thoracic limb; g, third thoracic limb; h, eighth thoracic limb. All x 33. reaching the penultimate segment laterally. Frontal plate only slightly produced in both sexes as a broadly and evenly rounded semicircular plate with slightly upturned margin and a median 452 Records of the Indian Museum. [VoL. XXIV, depression in the mid-dorsal line. In the median line of the cephalothorax about midway between the cervical groove and the anterior end of the frontal plate there is a trace of a tubercle, more marked in the female than in the male but much less pro- minent that Hansen figures in S. nodosa. Cervical groove well marked, especially on the posterior margin. Eyes moderately large, pigment black. Antennular peduncle exhibiting sexual differences, the last joint in the male being longer and thicker than in the female and bearing the usual brush of setae. Antennal scale reaching the distal end of the antennular peduncle in the female, TExt?-FiG. 4.—Siriella brevicaudata Paulson. a, first pleopod of male; ¢, second pleopod of male; c,Zuropods; d, telson e, apex of the telson: 7, apex of the telson from below. a-d xX 33, e X 65, fi X 400. falling somewhat short of this in the male, three times as long as broad, terminal lobe about one quarter of the scale in length and as broad as long, outer margin terminating in a strong spine, one prominent spine on the outer distal corner of the joint from which the scale springs. First and second thoracic limbs (gnathopods) stout, with the dactylus remarkably long, robust and strongly curved. Third thoracic limbs much more robust that the re- mainder due to the expanded merus and carpus. Posterior thoracic limbs much more slender and linear, tarsus unjointed. Telson remarkably short, when in position not extending much 1922.] W. M. Tarrersaut: Indian Mystidacea. 453 beyond half way down the uropods, one and a half times as long as broad at its base, proximal portion of the lateral margins with two or three spines, apex broadly truncate or even slightly emarginate, its breadth equal to half the total length of the telson; distal portions of the lateral margins and the apex together bearing about five or six pairs of spines, the innermost pair of spines at the apex equal in length to three quarters of the breadth of the apex, the remaining spines grading smaller in size, the fifth and sixth spines quite small. The centre of the apex bearing the usual pair of plumose setae which are longer than the innermost spines, and, hidden in dorsal view but visible under the high power of the microscope (}) in ventral view, are three very small spinules, cor- responding to the three small spines usually present in that posi- tion, but greatly reduced in size and at first sight apparently absent. Uropods about twice as long as the telson, the endopod slightly shorter than the exopod, with a row of 10-12 spines along the entire margin, regularly graded in size with no smaller spines between; proximal joint of the exopod abott three times as long as the distal, with a group of five spines at the distal end, the rest of the margin naked, distal joint about as long as broad. Pseu- dobranchial rami of the second to the fourth pairs of the pleopods of the male spirally twisted. Third and fourth pairs of male pleopods with the endopod and exopod subequal in length and having a normal armature of plumose setae, none of which are modified. Length of adult males and females 6 mm. This species falls into Hansen’s group I, characterized by the spirally-twisted pseudobranchial rami of the second to the fourth pleopods of the male and the unmodified nature of the setae on the terminal parts of these pleopods. It is distinguished specially by the very short telson and its peculiar armature of spines and by the very reduced size of the three spinules at its apex between the innermost long pair of spines. It has been a source of great satisfaction to rediscover Paulson’s species and to find that it is a good species which its original discoverer described and illustrated adequately. The species was quite easy to recognise from Paulson’s figures, but in view of the rarity of his work I have thought it well to redescribe and figure it here and to indicate its true position in the light of recent work. As far as I can make out Hansen’s group I contains at present about 12 species. Of this group S. thompsoni and S. gracilis differ from the rest in having the endopod of the uropods longer that the exopod. The four species S. clausit, S. jaltensis, S. norvegica and S. brooki are distinguished from the remainder in having more than half of the outer margin of the proximal joint of the exopod armed with spines. The remaining species in this group are S. quadrispinosa, S. nodosa, S. vulgaris, S. affinis, S, watasei and S. longtpes (the last two doubt- fully placed here), and S. brevicaudata is most closely allied to the two first-named forms. Distribution.—Known previously only from the Red Sea. 454 Records of the Indian Museum. [VoL. XXIV, Siriella quadrispinosa Hansen. Siviella quadrispinosa, Hansen, 1910, p. 32, pl. ii, figs. 5a-/. Locality.—Pamban, Gulf of Manaar, from weeds, o-2 fathoms. February 24th, 1913. Four males and two females; largest male, 75 mm., largest female, 5°5 mm. Remarks.—These specimens are in substantial agreement with Hansen’s description and figures. ‘They differ mainly in having the spines on the outer margin of the proximal joint of the exopod of the uropod never more than six in number and confined to the distal third of the margin. In the smaller speci- mens the inner pair of spines at the apex of the telson is equal in size to the outer pair but in larger specimens these spines are as figured by Hansen. Distribution.—Hitherto only known from the ‘ Siboga’ speci- mens taken among the Islands of the East Indies, near Saleyer. Siriella vulgaris Hansen. Siviella vulgaris, Hansen, 1910, p. 34, pl. iil, figs. 2a-z. Locality,—Port Blair; Andaman Islands. St. 3. Seventeen specimens, 4-7 mm. St. 11. Seven specimens, 4-7 mm. St. 13. One male, 7 min. St. 19. One male and two females, 6°5-7°5 min. St. 21. Two males and one female, 6-7 min. St. 32. Two males, fourteen females. ‘Investigator’ St. 556:—12° 40’ N., 98° 26’ 30” E, one Specimen. Distribution.—Found at 21 stations in the East Indian Archi- pelago by the ‘Siboga.’ Hansen states that it is common near the shores throughout the area explored by the‘ Siboga.’ It is evidently a common species at Port Blair in the shallow waters down to 12 fathoms. Siriella affinis Hansen. ? Siviella affinis, Hansen, 1910, p. 35, figs. 3a-/. Locality.—Kilakarai, Gulf of Manaar, from weeds, 0-2 fathoms, February 1913. Four males and four females, 5-7 mm. Remarks :—I am doubtful about the identification of these specimens. They belong to Hansen’s group I and are closely allied to S. vulgaris and S. affints. From the former they are distinguished by the much smaller terminal lobe to the antennal scale and by having only 3-5 spines on the outer uropod. In these respects and indeed in most of their characters they agree with S. affinis but they differ as follows: (1) the male specimens agree with the females in the characters of the rostral plate and antennal scale, Hansen described marked difference between the sexes in these characters ; (2) the large spines on the lower inner 1922.] W. M. TarrerRsALL: Indian Mysidacea. 455 margin of the inner uropods have smaller spines in between them, so that the spines as a whole are arranged in series of 3-5. Hansen gives no particulars on this point in his text but his figure does not show the intermediate spines. In other respects the specimens agree with S. afints and provisionally I record them under that name. Distribution.—Known only from the waters of the East Indian Archipelago. Siriella dubia Hansen. Text-figs. 5a, 0. Siviella dubia, Hansen, 910, p. 44, pl. v, figs. 4a-e. Locality.—Port Blair, Andaman Islands, Station 19. Eight specimens: largest female, 8 mm., largest male, 7°5 mm. Remarks.—These specimens differ in one important feature from the description and figures of Hansen. There are three small spines, equal in size to one another, between the inner pair of large terminal spines at the apex of the telson, in addition to the usual pair of plumose setae. Hansen parti- cularly emphasizes the absence of these spines in his single specimen but the present speci- mens are so very closely in agreement with Hansen’s des- cription and figures that I cau only suppose that the spines - were overlooked or broken off in his type. ue HENS UBL OSS EIS EE a, fourth pleopod of the male, x SYS = the outer margin of 33; b, distal joints of the exopod the proximal joint of the outer further enlarged, X !00. uropod in the positions indica- ted in Hansen’s figure 4d, with the addition of an extra spine bet- ween the two proximal ones of his figure, as mentioned in his text. The serrations on the proximal margin of th: outer uropod are in reality the bases of plumose setae, which have become detached. Several of my specimens still retain some of these plumose setae, so evidentiy the serrations do not indicate the base of broken spines. In the smallest specimen, 4°5 mm. long, the outer and inner uropods are equal in size; there is a progressive difference in size between the two uropods, with an increase in the total length of the animal The last joint of the antennular peduncle in the male is longer and stouter than in the female and furnished with the usual brush of setae. In the adult male, therefore, the antennal scale is shorter than the antennular peduncle, whereas in the female it TExt-FG. 5.—Siriella dubta, Hansen. 456 Records of the Indian Museum, [Vo,. XXIV, extends to the level of the distal end of that appendage. In young specimens the terminal lobe of the scale is not so long as in fully grown animals. In the male the pseudobranchial rami of the second to the fourth pleopods are spirally twisted. In the fourth pair of pleo- pods, the endopod and exopod are about equal in size and the endopod is only slightly modified. The terminal joint bears two setae, a short plumose seta, and a very long stout seta slightly curved and not plumose. The presence of the three small spines at the apex of the telson brings this species more into line with the normal species of the genus, but the peculiar form of the outer uropod and the unusual shape and spinulation of the telson are unique. _ Distribution.—Hansen’s single specimen was taken off the coast of Obi Major in the East Indian Archipelago. No other records are known. Siriella paulsoni Kossmann ? Siviella paulsoni, Tattersall, 1906, p. 160, pl. i, figs. 3-7. ‘Localities—Pamban, Gulf of Manaar, exposed reef, from pools. One female, Io mm. Kilakarai, Gulf of Manaar, from weeds, 0-2 fathoms. One female, 12 mm. Remarks. —These specimens belong certainly to the same species as the single female I recorded from Ceylon under this name. In view of the recent advances in our knowledge of this genus I now think it doubtful whether the species is really the same as that described by Kossmann. Certainty on this point can only be obtained when male specimens are available for examination. In the meantime I record the present specimens under S. paulsoni to indicate that they are the same as the Ceylon specimen. Genus Hemisiriella Hansen. Hemisiriella parva Hansen. Hemistriella pavva, Hansen, 1910, p. 47, pl. vi, figs, 2a—c. AS ~ Colosi, 1918, p. 6. 1 " Zimmer, 1918, p. 16, text-figs. 5-7. Ap i Colosi, 1920, p. 236, pl. xviii, figs. 2a. Locality :—Port Blair, Andaman Islands. St. 3. One young specimen. St. 19. Three males and three females, 5-6°5 mm. St. 21. Seventeen specimens. Remarks.—These specimen agree rather with Zimmer’s descrip- tion than with Hansen’s. Particularly is this so with the form of the eyes which are longer and narrower than Hansen shows them. Distribution.--Waters of the East Indian Archipelago and Bay of Bengal (Hansen); Bay of Bengal (Colosi); Java (Zimmer). These specimens were alltakenin plankton. It is therefore interest- 1922.] W. M. TatTrERSALL: Indian Mysidacea. 457 ing to find it in quite shallow waters, obtained by dredging and in shore-collecting. Subfamily RHOPALOPHTHALMINAE Hansen. Genus Rhopalophthalmus Illig. Rhopalophthalmus egregius Hansen. Rhopalophthalmus egregius, Hansen, 1910, p. 49, pl. vi, figs. 3a-z, pl. vii, figs, 1a-d, A ye Nakazawa, 1910, p. 255, pl. vill, figs. 12, 22. ” ” Tattersall, 1915, p- 151. ae . Colosi, 1918, p. 6. * Colosi, 1920, p. 237, pl. xviii, figs. 3a, b. Necsiiiee —Port Blair, Andaman Islands. St. 3. Two specimens. St. rr. Many specimens. St. 19. Four specimens. Off Puri Beach, Orissa, India, 4-44 fathoms, muddy sand, one male, 15 mm. Vasco Bay, Mormugao Bay, Portuguese India, sixteen spe- cimens. Bay N.W. of Nazareth Point, Mormugao Bay, Portuguese India, four specimens. Chicolna Bay and stream at its southern end, Mormugao Bay, Portuguese India, abundant. ‘Investigator’ St. 604 :—11°17’20"N., 98°29’40’E., five spe- cimens. ‘Investigator’ St. 556:---12°40'N., 98°26'30”E., two speci- mens. Distribution.—East Indian Archipelago (Hansen); Japan (Nakazawa); Chilka Lake, India (Tattersall); Torres Straits and off New Zealand (Colosi). The last two records of Colosi are very interesting and indicate a very wide geographical range in the Indian and Pacific Oceans. Subfamily GASTROSACCINAE Norman. Genus Anchialina Norman. Anchialina typica (Kroyer). Anchialus typicus, Kréyer, 1861, p. 53, pl. ii, figs. 7a-l. Anchialina typica, Hansen, 1910, p. 52, pl. vii, figs. 2a—k. ” i Hansen, 1912, p. 196. En 15 Colosi, 1918, pi 7: a Colosi, 1920, p. 237. Lacaities, —Port Blair, Andaman Islands. St. 3. One male, one female, two young. St. 8. One male. St. 19. One female. St. 21. One male, one female. 458 Records of the Indian Museum. [VoL. XXIV, Kilakarai, Gulf of Manaar, from weeds, 0-2 fathoms, Feb- tuary, 1913. Two males and two females. ‘Investigator’ St. 556:—12° 4o’ N., 98° 26’ 30” E., six specimens. The largest specimen of either sex measured 5 mm. Distribution .— Hansen (1910 and 1912) has given a synopsis of the known distribution of this species, which is known from the tropical Atlantic, West Indies, Gulf of Siam, Hast Indian Archipe- lago and the Hawaiian Islands. The only record subsequent to Hansen’s paper is that of a single male from the Caribbean Sea by Colosi (t918 and 1920). Gough’s record from the English Channel (Publ. de Circonstance, No. 33, 1906, p. 105) is interesting but re- quires confirmation. Anchialina grossa Hansen. Text-fig. 6. gence grossa, Hansen, 1910, p. 54 pl. vii, figs. 3a-n, pl. viii, figs. la-d. Anchtalina grossa, Hansen, 1912, p. 196. 9 frontalis, Zimmer, 1915 (3), p. 159, text-figs. 1-6. Locality.—Port Blair, Andaman Islands. St. 3. Three specimens. St. 8. Twenty-six specimens. St. 19. Two specimens. St. 21. Three specimens. St. 32. Four males, five females. Altogether eighteen males and twenty-five females ; the larg- est male 9 mm., largest female 6°5 mm. Remarks.—The specimens agree closely with Hansen’s description and figures. A. frontalis, as described by Zimmer, differs from A. grossa (I) in the telson which is only two and a half times as long as broad whereas in A. grossa it is three times as long as broad; (2) in the antennal scale, which is twice as long as broad as against two and a third in 4. grossa; and (3) in the proportions of the an- tennal peduncle. These differences are very small and Zimmer was led to institute his species mainly on the structure of the third pleopod of the aastienits NO SeAR ana male. TI give herewith a figure of the grossa, Hansen. distal end of the outer branch of the Distal end of the exopod of titd pleopod of an immature male the third pleopod of a young of A. grossa measuring 8 mm. It dif- male, X 200, fers remarkably from the fully grown 1922.] W. M. TATTERSALL: Indian Mysidacea. 459 state and agrees very closely with Zimmer’s figure of the samme appendage in A. frontalis. This male specimen does not appear to be quite fully grown. In the proportions of the telson and scale it agrees with A. grossa. A study of my specimens of 4. grossa seems to indicate, however, that Zimmer’s male of A. fron- talis was not fully grown. Hansen (1910) has given a figure of the distal portion of the exopod of the third pleopod of an immature male of A. grossa which shows a stage earlier in deve- lopment to the one I figure here. I suggest, therefore, that A. frontalis will prove to be founded on not quite adult males of A. grossa Distribution.—Waters of the East Indian Archipelago, Bay of Bengal, Gulf of Siam (Hansen, tg10); Gilbert Islands (Hansen, IgI2). Zimmer’s specimens of A. frontalis were taken during a voy- age from Ceylon to the Dampier Straits, New Guinea and therefore in the same waters as A. grossa. Genus Gastrosaccus Norman. Gastrosaccus dunckeri Zimmer. Gastrosaccus dunckeri, Zimmer, 1915 (3), p. 165, text-figs. 13-18. Locality.—Off Puri Beach, Orissa, 4-44 fathoms, 57 specimens. Remarks,—-This species belongs to the same group of species of the genus as G. sanctus and is very closely allied to that species. It is distinguished mainly by the remarkably well-developed lobes on the carapace and the shape and armature of the telson. The lobes on the carapace are larger than in any other species of the genus ‘They extend forward to the centre of the dorsal surface of the carapace and are acutely pointed in shape. Distribution.—Zimmer’s specimens came from the Duncker collection made during a voyage from Ceylon to New Guinea. Gastrosaccus muticus Tattersall. Gastrosaccus muticus, Vattersall, 1915, p. 152. text-fig. 1. Locality.x—Off Puri, Orissa, 4-44 fathoms, muddy sand. Eight adult females, 7 mm. Several localities in the Matlah River, Gangetic Delta, abun- dant. Remarks.—tThere are no male specimens, but from the form and armature of the telson, which in all the specimens has fourteen spines on its lateral margin, and by the possession of a fringe of six to nine filaments on the posterior median dorsal margin of the carapace, I feel certain that these specimens belong to the same species as that which I have described from the Chilka Lake. No other records are known. A single mutilated female specimen in this collection from 460 ' Records of the Indian Museum. [VoL,. XXIV, Kilakarai, Gulf of Manaar, possibly belongs to this species or to G. simulans. The telson and uropods are broken so that certainty on this point is impossible. Gastrosaccus simulans Tattersall. Gastrosaccus simulans. Tattersall, 1915, p. 155, text-fig. Ic. Locality.— Vasco Bay, Mormugao Bay, Portuguese India, one female. Remarks,—In the absence of male specimens, this example seems to agree well with the species which I described from Puri Beach. No other records are known. Gastrosaccus kempi, sp. nov. Text-figs. 7a-d. Locality.—Off Puri, Orissa 4-44 fathoms, 24-iii-16, muddy sand. Two males, 9 mm., one female, 9 mm. (Types.) TEXT-FIGURE 7.—Gastrosaccus kempi, sp. nov. a, telson, X 65; 6, third pleopod of male, x 33; c¢, second pleopod of male, x 33, 2, one of the setae from the exopod of the second pleopod of the male, further enlarged. Description.— This new species belongs to the same group of 1922.] W. M. TaTrersart: Indian Mystdacea. 461 species as G. sanctus and is very closely allied to that species. It will be best to refer to Sars’ description and figures of G. sanctus (1877) and to point out the differences between the two forms. G. kempi agrees with G. sanctus in general form and in the details of the appendages of the head and thorax but differs in the following points :— (x) There are no lobes or lappets on the dorsal hinder margin of the carapace. . (2) The antennal scale is slightly more than three times as long as broad, the terminal spine on the outer margin not quite extending as far forward as the apex of the terminal lobe. A suture across the terminal lobe is present. (3) There are fifteen marginal spines on the outer uropod and fourteen spines on the inner lower margin of the inner uropod. (4) Telson somewhat less than two and a half times as long as broad, cleft about one eighth of the total length, eight spines on each lateral margin, the terminal spines about one-sixth of the total length of the telson, about 3-5 small spinules between the spines of the lateral margin from the third to the eighth (terminal) spine. This last character is unique in the genus. (5) First, fourth and fifth pleopods of the male exactly as in G. sanctus. Second pair with the endopod composed of seven * joints, exopod of eight joints, half as long again as the endopod, slightly curved, the proximal joints armed with peculiar plumose setae. Third pair with the endopod six-jointed, not as long as the first joint of the exopod. Latter very elongate, with the terminal spines extending to the base of the telson, four—jointed, the second and fourth joints each longer than the third, fourth joint termina- ted by a long feathered spine, almost as long as the joint, and a small simple spine. The form and spinulation of the telson will serve to distin- guish this species from any hitherto described. I know of no other species which has subsidiary spinules between the large spines arming the margin of the telson. Otherwise it agrees closely with G. sanctus except for the absence of lobes on the carapace and minor details in the number of spines on the telson and uropods. | Gastrosaccus pacificus Hansen. Text-figs. 8a, b. ’ Gastrosaccus pacificus, Hansen, 1912, p. 198, pl. 2, figs. 3a-g. Locality.—Port Blair, Andaman Islands. St. 3. Five males, four females. St. 7. Two males, one female. St. 32. One male. 462 Records of the Indian Museum. [Vor. XXIV, Remarks.—This species is very closely allied to G. sndicus and TExtT-FiG. 8.—Gastrosaccus pacificus, Hansen. a, third pleopod of the male, X 33; 4, joint of the same, X 100. the only real difference indicated by Hansen lies in the structure of the third pleopod of the male. I figure herewith one of these pleopods to indicate its form in these specimens and to complete Han- sen’s description, since the distal part of the exopod was missing in his specimens. The exopod is three-jointed, the second joint slightly longer than the first and one and a half times as long as the third. The basal joint bears on the outside a sharply pointed triangular process and on the inside two spines (Hansen figures three). The third joint narrows sudden- ly at about two-thirds of its length and terminates in three setae, one long with a few subsidiary hairs, a second some- what shorter and smooth and a third slender lash-like seta twisted as shown in the figure. All the specimens showed this peculiar feature. Distribution.—Gilbert Islands (Hansen). Gastrosaccus bengalensis Hansen. Gastrosaccus bengalensis, Hansen, 1910, p. 58. ” ” ” ” Zimmer, 1915 (3), p. 164. Zimmer, 1918, p. 15. Locality.—Port Blair, Andaman Islands. St. 8. ‘Two males. St. 32. One male. Distribuison.—Bay of Bengal (Hansen); voyage from Ceylon to New Guinea, and off Formosa (Zimmer). Subfamily MYSINAE. Tribe ERYTHROPINI. Genus Erythrops G. O. Sars. Erythrops minuta Hansen. Text-figs. 9a, b. Erythrops minuta, Hansen, 1910, p. 63. Locality.—Kilakarai, Gulf of Manaar, from weeds, 0-2 fathoms, February, 1913. Five males, 3 mm. 1922.] W. M. TATTERSALL: Indian Mysidacea. 463 Remarks.—-These speci- mens agree with Hansen’s description but since the author has not published any figures illustrating this spe- cies, I give herewith a figure of the antennal scale and of the telson and uropods to illustrate the salient charac- ters of the species. The dis- tal part of the lateral mar- ees of the telson is minute- Text-Fic. 9.—LZvythrops minuta, Hansen! ly serrulate. : a, antennal scale and peduncle; 3, telson Distributton.—Gulf of and uropods. Both x 65. Siam (Hansen). a, Erythrops nana, sp. nov. Text-figs. 10 a—c. Locality.—Port Blair, Andaman Islands, St. 3. Three females, one male, 3 mn. (Types.) Description.—Frontal plate produced into a short, pointed rostral process. Eyes small, depressed, scarcely broader than the eyestalks. Antennal scale scarcely as long as the antennular peduncle, four times as long as broad, outer margin smooth, ter- minating in a strong spine distally, beyond which the terminal lobe of the scale projects considerably ; terminal lobe longer than broad with a distal articulation. Telson broader than long, pos- terior margin nearly half as long as the greatest breadth, with three pairs of spines and a median pair of plumose setae; the inner pair of spines longer than the next pair while the outer pair are quite small; the lateral margins of the telson unarmed. Uropods more than twice as long as the telson, inner and outer branches equal in length, no spines on the Pr lower inner margin of the } inner uropod and its inner margin not serrulate. The specimens are very a mutilated but the remain- ing appendages do not call TEXT-FIG. 10.—Erythrops nana, n. sp. for special comment. They a, telson and uropods; 6, antennular ped- agree in the main with ee c, antennal peduncle and scale. All those of Erythrops elegans, the species to which E. nana is most nearly allied. It differs from this species in the shorter antennal scale, in the armature of the apex of the telson which has an extra pair of spines, and in the equal outer and inn>r 464 Records of the Indian Museum. [VoL. XXIV, uropods. It is distinguished from EF. minuta in having the outer margin of the antennal scale smooth and not serrate, as well as in baving an extra pair of spines on the telson. Genus Hypererythrops Holt and Tattersall. Hypererythrops spinifera (Hansen). Text-figs. Ila-1. Erythrops spinifera, Hansen, 1910, p. 62, pl. 9, figs. 3a-c. Locality .—Port Blair, Andaman Islands. St. 3. Two males, two females. St. 19. Six males, eight females. Remarks.—These specimens agree very closely with Hansen’s description and figures of Erythrops spinifera except in one point, the number of spines on.the lateral margins of the telson. Han- sen gives the number as 10-13 and his figure shows them to be arranged at practically regular intervals along the whole margin. In these specimens I find the spines to be fewer and to be more distantly and more irregularly arranged. The figure (text-fig. 11/) shows a typical telson among the Port Blair specimens. But the spines on the telson appear to be very variable in number and seldom the same number on both sides of a single telson. The smallest number of spines on each margin is five and the largest number nine. In only four specimens was the number of spines on each side of the telson the same and frequently the two sides differed by two spines. The apex of the telson in the Port Blair specimens bears two pairs of long stout spines, the inner pair of which is always longer than the outer pair but the proportion between the lengths of the outer and inner spines varies considerably, in some specimens approaching the condition as figured by Hansen in which the two pairs are nearly equal in size, in other specimens having the pro- portions shown in my figure in which the outer pair is considerably shorter than the inner pair, Between the latter are a pair of quite small spines and a pair of long plumose setae. Hansen found the . setae only in one of his specimens and the spines only in the other. In other respects these specimens agree so.closely with Han- sens’s: species that I feel that they cannot be considered as more than a variety, especially in view of the great variation which they show among themselves. I am content, therefore, to record them under Hansen’s name, to point out the differences I have found and to figure the more essential parts for comparison. Hansen had no mature males at his disposal. The Port Blair specimens include several males and an examination of them shows that the species must be referred to the genus Hypererythrops Holt and Tattersall (1905). This genus differsfrom Erythrops in having the telson much longer in shape with its lateral margins armed 1922.] W. M. TATTERSALL: Indian Mysidacea. 465 h TeExtT-Fic. 11.—Hypererythrops spinifera (Hansen). a, antennal peduncle and eye; 6, antennal scale and pean c, first thoracic limb ; d, endopod of second thoracic limb; e, endopod of third thoracic limb; f, first pleopod of the male; g, second pleopod of the male; h, telson; t, one of the processes arming the sterna of the thorax of the male. a@ X 30, b—h x 60. 466 Records of the Indian Museum. [VoL. XXIV, with spines instead of smooth, and in having the sterna of some of the thoracic and abdominal somites furnished with median pro- cesses. In the present species the sterna of the last six thoracic somites are furnished with long sharply pointed forwardly directed processes as shown in the accompanying figure (text-fig. 117), the lower margin of which is furnished with numerous spinous pro- cesses. The sterna of the first four abdominal somites are fur- nished with simple papilliform processes. _ The species, H. spinifera, therefore agrees absolutely with the characters of the genus Hypererythrops as distinguished from Erythrops and is closely allied to the type species, H. serri- ventey H. and T. It may, however, be distinguished froin the latter by its smaller size, the different form of the antennal scale, which is narrower and has the terminal lobe much less developed than in the type, and by the different shape and spinulation of the telson. One other feature of the genus Hypererythrops must be men- tioned. Inthe pleopods of the male there isa broad flat branchial plate, devoid of setae, at the base of the endopod, which appears to arise from the short setiferous lobe characteristic of the endopod of the pleopods of Mysidae. The form of this branchial plate in H. spinifera is shown in (text-figs. 11f, g.) I have re-examined some specimens of H. serriventer, and find that a similar branchial plate, larger in size, is present on the male pleopods. I do not know of a similar development among the Erythropini and it is interesting to note that it is present in the European and the Indian species of this genus. Its presence forms an additional character separating the genus from Erythrops. These branchial plates recall the pseudo-branchial processes on the pleopods of the species of Sivtella and, in point of fact, on the first pleopod of the male of H. spinifera the branchial lamella is bilobed as in so many of the species of Siviella. But in the remaining pleopods it is a simple broadly expanded plate without setae. Distribution.—Hansen’s specimens were found in the seas of the East Indian Archipelago. The occurrence of this genus, hitherto known only from European waters, in the waters of the Indian Ocean is a matter of great interest. Tribe LEPTOMYSINI. Genus Mysidopsis G. O. Sars. Mysidopsis indica, sp. nov. Text-figs. 12a-e. Locahty.—Port Blair, Andaman Islands, Station 5. Two males and two females, 4 mm. (Types.) Description.—Very closely allied to Mysidopsts gibbosa G. O. Sars. It will be sufficient to refer to Sars’ description of this species (1870-79) and to point out the following differences :— 1922.] W. M. TATTERSALL: Indian Mysidacea. 46 7 (1) There are three nodules in the median dorsal line of the catapace, two in the same positions as in M. gibbosa, the third one in front of the cervical groove. These nodules are present in both sexes though less marked in the male than in the female. (2) The frontal plate is more developed than in M. gibbosa, TEXT-FIG. 12.—Mysidopsis indica, sp. nov. a, antennular peduncle of male; 6, antennal scale; c, first pleopod of the male; d, fourth pleopod of the male; e, telson and uropods. All x 65. more broadly triangular, longer and completely covering the basal joints of the eye-stalks. (3) The antennal scale is two and a half times as long as broad, being thus broader than in M. gibbosa. It outreaches the antennular peduncle but not to the same extent as in M. gibbosa. : (4) Telson shorter than in M. gibbosa, not much more than half as long as the uropods, and hardly longer than wide at the base, 468 Records of the Indian Museum. [VoL. XXIV, The distal part narrows very suddenly and considerably and the apex is only as long as one quarter of the basal width. The apex is very shallowly notched and bears a pair of small spines on each side of the notch. There are no plumose setae. The lateral margins bear two spines at the widest part of tlie telson but otherwise are naked. (5) The uropods in the specimen figured have the endopod and exopod of approximately equal length, but there appears to be some variation in this respect since one of the female specimens has the endopod distinctly shorter than the exopod. There is a single spine on the lower inner margin of the endopod in the region of the otocyst. In M. gibbosa there are five such spines. (6) The pleopods of the male agree essentially with those of M., gibbosa, except that there is a small branchial plate at the base of the endopod. This lamella-like expansion is broad and flat on the second to the fifth pleopods, but is narrower and more finger-like on the first pleopod. The exopod of the fourth pleopod terminates in a single stout plumose spine. (2) M. indica is smaller than M. gibbosa, adult specimens of both sexes measuring only 4 mm., as against 6-7 mm. in the latter species. Mysidopsis kempi, sp. nov. Text figs. 13a-g. Locality.—Kilakarai, Gulf of Manaar, among weeds, 1-2 fa- thoms, February, 1913. Eleven females and four males, 5-6 mm. (Types.) ) . Description :—Carapace leaving the last two thoracic somites exposed dorsally, but laterally covering all but the last somite ; produced in front into a short triangular plate with a bluntly point- ed apex which does not project forward very much beyond the antero-lateral corners; no tubercles or nodules. Eyes large, pigment black, cornea as wide as the rest of the eye and occupying half the eye in dorsal view. Antennal scale narrowly oval in shape, four times as long as broad; setose all round, terminal joint distinct, extending for one- third of its length beyond the antennular peduncle. Mouth parts and thoracic appendages not differing greatly from those of M. didelphys; the inner lobe of the first maxilla has three terminal setae; in the posterior thoracic limbs the merus is about equal in length to the ischium but less expanded, the tar- sus is three-jointed, the second joint the smallest ; the basal joint of the exopodite has the outer corner rounded and the flagellum is composed of eight to ten joints. Telson (without terminal spines) as long as the last abdominal somite, one and a quarter times as long as broad at its base, apex quadrate with rounded angles, with four pairs of long stout spines, the inner pair nearly one-third the length of the telson, outermost _ pair of the four about half as long as the inside pair, lateral mar- 1922. } W. M. TarrersatL: Indian Mysidacea. 469 TEXT-FIG. 13.—A/ysidopsis kempt, sp. nov. a, antennular peduncle and eye; 4, antennal scale; c, first thoracic limb ; d, second thoracic limb ; e, endopod of third thoracic limb; f, fourth pleopod of the male; g, telson and uropods. All x 64. 470 Records of the Indian Museum. [ VoL. XXIV, gins with about ten spines distributed throughout their length, more distantly placed proximally, nearer together distally, the most dis- tal marginal spine less than half as long as the outer spines of the apex ; no plumose setae at the centre of the apex. Inner uropod half as long again as the telson, with a comb of ten spines on the lower surface in the region of the statocyst but not extending down the inner margin. Outer uropod nearly twice as long as the telson. Fourth pleopod of the male of the usual. type found in My- sidopsis, both rami six-jointed, the outer ramus longer than the inner and terminating in a single, long, stout plumose spine. Remarks.—This species is a very typical member of the genus Mysidopsis but is easily distinguished from all the other species by the shape and armature of the telson. Genus Leptomysis G. O. Sars. Leptomysis xenops, sp. nov. Text-figs. 14a-g. Leptomysis apiops?, Zimmer, 1915 (3), p. 167, fig. 19. Locality.—Port Blair, Andaman Islands. St. 5. Two. St. 11. Abundant. (Types.) Description.—Agreeing with Leptomysis apiops G. O. Sars, except in the form of the telson. ‘The latter is linguiform in shape, one and a half times as long as broad at the base; apex more or less truncate, half as long as the width of the telson at the base, armed with three pairs of stout spines, the innermost pair the longest, equal in length to two-fifths of the length of the telson ; between the inner pair of spines are two small spinules about one | quarter of the length of the spines; the spines immediately out- side the inner pair are about two-thirds of the length of the latter and the outer spines of the apex are slightly less than one half of the length of the inner pair; the lateral margins of the telson bear a single long spine at the point of the greatest width of the telson and from 14-17 spines on the rest of the margin, the prox- imal ones more distantly placed than the distal, the spines increas- ing in size in regular sequence towards the apex and not arranged in groups. For the rest of the characters reference may be made to Sars’ figures of L. apiops with which this species agrees in all its other characters. I have given figures of the principal appendages of L. xenops for comparison. Zimmer (1915 (2) ) has given a new figure of the eye of L. apiops to illustrate the elongation of certain facets which leads to the peculiar shape of the eye, trom which the species takes its name. The present species has eyes of exactly the same form. In fact it is very closely allied to L. apiops, but the latter has about 35 1922 ] 471 W. M. TaTrersALL: Indian Mysidacea. WIT Tt xppr Sy Text-Fic. 14.—Leptomysis xenops, sp. nov. a, antennal scale; 4, antennular peduncle; c, endopod of the first thoracic limb; d, endopod of the second thoracic limb: e, fourth pleopod of the male ; f, inner uropod; g, telson. a X 30, b—g xX 60. 472 Records of the Indian Musewm. [VoL. XXIV, spines on each of the lateral margins of the telson and moreover these spines tend to arrange themselves in groups of three to five smaller spines separated by larger spines. There are also differ- ences in the proportions of the spines at the apex of the telson. In L. apiops the inner pair of spines is twice as long as the next pair and the spinules between the inner spines are about half as long as the latter. Zimmer (1915 (3) ) has recorded L. apiofs with a query from the Indian Ocean. It seems probable that the single specimen at his disposal really belonged to the present species. Zimmer does not give the number of spines arming the lateral margins of the telson but the spines at the apex, judging from his figure, agree in their proportions rather with L. xenops than with L. apiops. The only difference I can see is that the small spinules between the large pair of spines at the apex are about half as long as the spines in Zimmer’s specimen and only one quarter as long as the spines in mine. The species is an abundant one in the neighbourhood of the Andamans, to judge by the large number of the specimens in this collection. Genus Afromysis Zimmer. Afromysis macropsis, sp. nov. Text-figs. 15a-g. Locality.—Off Puri, Orissa, 4-44 fathoms. One male, 9 mm. (Type.) Description.—Body sntooth, without spinules; carapace pro- duced into a short triangular rostral plate with an obtusely rounded apex; eye long and narrow, recalling the eye of the genus Meso- podopsis, more than twice as long as broad, cornea occupying less than the distal half of each eye and not wider than the rest of the eye, pigment black. Antennal scale shortey than the antennular peduncle, about seven times as long as broad, setose all round, distal articulation well marked, a prominent spine on the outer distal corner of the basal joint. Posterior thoracic legs rather short and slender, tarsal joint divided by a single transverse articulation. Telson one and a half times as long as broad at the base, cleft for one-third of its length, cleft wider proximally than distally and unarmed except for two long plumose setae, the lateral margins armed with three spines proximally at the widest part; these are followed by a short unarmed portion of the margin and distally there are about twenty spines; the proximal eight or nine of these spines are normal sharply-pointed short spines; the remainder are blunt spines increasing in size to the apical lobes, the two or three on the inner side of the apical lobes somewhat smaller but of the same type. Inner uropod one and a half times as long as the telson, its 1922. ] W. M. TaTrtTeRsALL: Indian Mysidacea. 473 inner margin armed with a dense row of about 35-40 spines, the proximal 30 of which are bluntly pointed, alternating larger and smaller sizes, nearly always one smaller one between two large ones, but distally there may be two or even three small ones Sy SS = Ce — TEXT-FIG. 15.—Afromysis macropsis, sp. nov. a, anterior end, X 33; 8, antennal scale, x 50; c, second maxilla, x 65; d, endopod of fourth thoracic limb, x 50; e, fourth pleopod of male; f, telson, x 50; g, uropods, X 50. between the large ones; the distal spines are normal and sharply- pointed. Outer uropod one and three quarter times as long as the telson. The second maxilla conforms to the type found in Afromysis hansont, with the outer distal corner of the second joint of the palp 474 Records of the Indian Museum, [VoL, XXIV, very much produced. The fourth pleopod of the male differs from that of the type-species in having the penultimate joint of the exopod of normal size and not unduly elongated. But the only specimen has a look of immaturity about it, since the lobe on the antennule lacks the dense tuft of hairs characteristic of adult males. It is possible therefore that adult males may be found to agree more closely with the type in the form of the fourth pleopods of the male. . Remarks.—This interesting species is a true Afromysis, differ- ing from the type-species in the different form of the eye, the more produced rostrum, the longer antennular peduncle and shorter antennal scale and in the different form and armature of the telson. Genus Prionomysis, nov. Antennal scale long and narrow, setose on both margins, terminal joint distinct. Terminal joint of the palp of the second maxilla longer than wide, without strong spines on its distal margin. First thoracic limb with a masticatory lobe on the second joint of the endopod only ; tarsal joint of the remaining thoracic limbs with two transverse articulations. Telson linguiform in shape, cleft at the apex, cleft furnished with a pair of plumose setae but without spines, lateral margins armed throughout their length by spines which increase in length posteriorly and are arranged in a regular saw-like formation on each of the apical lobes. Inner uropods with a dense row of spines on the inner margin, extending from the statocyst to the apex; outer uropods without a distal joint, and without spines. Pleopods in the male as in the genus Lepftomysis. Female with three pairs of incubatory lamellae. Type:—Prionomysis stenolepis, sp. nov. Remarks.—This genus is most nearly allied to the genus Lepiomysis. It agrees with that genus in the form of the second maxilla, antennal scale and pleopods of the male, but is distin- guished at once by the form of the telson. At first I was disposed to refer the species to the genus A/j- vomysis but the discovery of a second species of the latter, showing the same peculiar form of the palp of the second maxilla as in the type has led me to regard this character as of generic value. Prionomysis shows considerable resemblances to the genera Doxomysis and Bathymysis, but again the form of the second maxilla separates it. In Doxomvysis and Bathymysis the terminal joint of the palp of the second maxilla is broader than long, expanded distally and armed with stout spines. In both genera, too, the cleft of the telson is armed with spinules, whereas in Prionomysis the cleft is smooth. In Doxomysis the masticatory lobes on the endopod of the first thoracic limb are much more 1922.] W. M. TAYTERSALL: Indian Mystdacea. 475 developed than in Prionomysis, being present on the second, third and fourth joints and very much larger. In addition to the form of the second maxilla, Prionomysts also differs from Afvomysis in the less specialized form of the fourth pleopod of the male. Prionomysis stenolepis, sp. nov. Text-figs. 16a-7. Locality.—Port Blair, Andaman Isles, Station 3. Eight females, two males, 8-g mm. (Types.) Description.—Carapace produced in front in the form of a triangular plate with acutely pointed somewhat depressed apex which reaches forward almost to the middle of the first joint of the antennular peduncle; antero-lateral corners rounded; last two thoracic somites exposed dorsally. Eyes large, somewhat flattened, cornea wider than the remain- der of the eye, occupying more than one half of the eye in dorsal view, the anterior margin of the eye stalk longer than the pos- terior; eye at least as long as the first joint of the antennular peduncle. Antennular peduncle with the first joint longer than the remaining two combined; the last joint in the male with a well- developed hirsute lobe of normal form. Antennal scale exceedingly long and narrow and curiously twisted, about thirteen times as long as broad, twice as long as the antennular peduncle and four times as long as the antennal peduncle, setose all round, terminal joint distinct though small. Antennal peduncle much shorter than the antennular, second joint ‘onger than the third; mouth-parts agreeing on the whole with those of Ajfromysis hansont Zimmer, except that the terminal joint of the palp of the second maxilla is not expanded and produ- ced into a narrow process but is normal in shape. First thoracic limbs robust, masticatory lobe present only on the second joint and not nearly so well developed as in Afromysis or Doxomysis, nail robust ; second thoracic limbs having the nail long and stout ; tarsus of the remaining thoracic limbs three-jointed and terminated by a long stout nail; all the posterior thoracic limbs appear to be similar in size and form. Fourth pleopod of the male with both endopod and exopod six-jointed, but the exopod one quarter longer than the endopod ; each of the last three joints bears a stout plumose spine; the fourth pleopod of the male is very like that in the genus Lepffo- mysis and not nearly so specialized as in the genus A fromysis. Telson slightly shorter than the last somite of the abdomen and not reaching very much beyond the statocyst of the uropod ; lateral parts of the last abdominal somite produced rather acutely at each side of the base of the telson; telson much narrower than the last abdominal somite, not quite twice as long as broad at the base, narrowing gradually for almost three quarters of its length 476 Records of the Indian Museum. [Vor. XXIV, Text-Fric. 16.—Prionomysis stenolepis, sp. nov. a, antennular peduncle and eye, X 30; 6, rostrum, X 30; c, antennal scale, x 30; a, endopod of first thoracic limb, X 30; ¢, endopod of second thoracic limb, X 30; f, second maxilla, x 60; g, third thoracic limb, xX 30; h, fourth pleopod of the male, X 60; 72, uropods, X 30; 7, telson, X 60. 1922.] W. M. TATTERSALL: Indian Mysidacea. 477 and there widening and terminating in two broad lobes separated by a median wide cleft, about one-fifth of the length of the whole telson ; margins of the cleft unarmed except for two long plu- mose setae at the anterior end; margins armed throughout their entire length with spines; about seventeen small spines on the proximal part of the margin from the base of the telson to the narrowest part; from the narrowest part to the apex of each lobe there are about twenty-five closely packed spines, longer than those on the proximal portion of the margin and increasing in size to- wards the apex. Inner uropod about one and a half times as long as the telson with a very prominent spine on the dorsal surface of the stato- cyst, towards the outside; this spine is very prominent in lateral view; inner margin armed with a dense row of spines throughout its length from the statocyst to the apex, the spines arranged in series of larger and smaller ones, three to four in each series. ~ Outer uropod twice as long as the telson. Female with three pairs of incubatory lamellae. Genus Dioptromysis Zimmer. Dioptromysis perspicillata Zimmer. Dioptromysis perspicillata, Zimmer, 1915 (3), p. 168, text-figs. 20-22. Localities.—Port Blair, Andaman Islands. i St. 3. Five females, one male. St. rr. One male. Pamban, Gulf of Manaar, from weeds, o-2 fathoms, February, 1913. Nine females, one male. Kilakarai, Gulf of Manaar, from weeds, o-2 fathoms, February, 1913. Nine females, three males. Remarks —-The largest female measured 5 mm., and the largest male, 3°5 mm. The discovery of male specimens allows of the proper classification of this species. It belongs to the tribe Leptomysini. The pleopods of the male agree generally with those of the genus Lepiomysis. The exopod of the fourth pair is longer than the endopod, composed of six joints, the last one terminating in a single thick plumose spine, equal in length to the last four joints of the exopod, at the base of which is a small smooth spine. The endopod of the fourth pair and the exopods and endopods of the other pleopods are four jointed. Distribution.—The only known record is of a female taken on a voyage from Ceylon to New Guinea. Genus Doxomysis Hansen. This genus was established by Hansen (1912) for a species, D. pelagica, captured off the Galapagos Islands. Illig, however, in 1906, had described a species, ‘‘ Mysts’’ quadrispinosa, which is clearly referable to this genus, though as Illig had only a single female specimen at his disposal, he did not feel justified in 478 Records of the Indian Museum. [VoL. XXIV, establishing a new genus for his species. Hansen’s species was also represented by a single female and in consequence neither he nor Illiz was able to place the genus in its proper place in the classification. Colosi (1920) had more abundant material and was able to establish the fact that the genus belongs to the tribe Lep- tomysini. He described four new species, D. hanseni, D. zimmert. D. tattersallii and D. microps. These species do not seem to me to be founded on sufficient grounds. I regard D. zimmeri as a syno- nym of the earlier D. quadnspinosa (Illig) and I do not think D. tattersallii is separated from D. pelagica by any characters of specific value. This would leave four species in the genus. They all agree in having the spines arming the telson confined to the distal half of the lateral margins and thereby differ from the new species described below, in which the spines extend throughout the whole length of the lateral margins. All the hitherto described species are pelagic and were taken at the surface in the open sea, in contrast to the species in this collection which is littoral in habit. The genus is very closely allied to Bathymysis ‘Tattersall (1907 and 1911), and to Af/romysis Zimmer (1916). It differs from the former in the possession of well-developed eyes and from the latter in the form of the second maxilla. Doxomysis littoralis, sp. nov. Text-figs. 17a-e. Localities.—Port Blair, Andaman Islands. St. 3. Fifteen females, one male, 4 mm. St. 11. Nine females, one male, 5 mm. (Types.) St. 19. One male, 4mm. Description.—Body smooth, without spinules ; carapace pro- duced into a very short triangular rostral plate with the apex bluntly rounded, not covering the bases of the eyestalks. Eyes of moderate size, cornea more than half as large as the whole eye, slightly wider than the stalk, pigment black. Antennal scale outreaching the antennular peduncle by one- third of its length, seven times as long as broad, setose all round, terminal joint one-seventh of the total length of the scale, a promi- nent spine on the outer distal corner of the basal joint. Thoracic limbs with the endopods long and slender, increasing in length from the fourth to the eighth limbs, tarsus three-jointed, nail distinct. Telson one and a half times as long as broad at the base, cleft for one-third of its length, cleft wide, rounded at the apex, armed with a pair of plumose setae and fifteen small spinules on each margin, terminal lobes of more or less equal width through- out, with the apex rounded, almost truncate, and armed with four spines, the centre pair of which are subequal in size and slightly longer than each of the lateral ones, the lateral margins armed 1922.] W. M. ‘TATTERSALL: Indian Mysidacea. 479 with about thirteen spines extending throughout their entire length, the proximal spines more distantly placed than the distal ones, the latter gradually increasing in size to the apical lobes and grading off into the spines which arm them. Text-FiG. 17.—Doxomysis littovalis, sp. nov. a, anterior end, X 65; b, antennal scale, x 65; c, endopod of eighth thoracic limb, X 33; d, uropods, xX 65; e, telson, X 65. Inner uropods one and a quarter times as long as the telson, the inner margin armed with .a row of about thirty-two blunt spines extending from the statocyst to the apex, alternately larger and smaller in size, sometimes distally there may be two or even three smaller spines between a pair of larger ones. 480 _ Records of the Indian Museum, | [Vou XXIV, Outer uropod one and a half times as long as the telson. Length of adults of both sexes, 5 mm. There is a consider- able development of chromatephores on the antennular peduncle, antennal scale and its peduncle, mouth parts and first two pairs of thoracic limbs, brood lamellae, along the whole of the ventral sur- face of the abdomen, telson and uropods, and, as, at the time of death, these chromatophores were expanded, the preserved animals present a dusky appearance. Remarks.—This species differs from all the other described species in having the margins of the telson armed throughout their entire length with spines and in the very large number of spines on the inner uropod. Only D. hanseni, among the described species, agrees with the present one in having a smooth body devoid of spinules. Three specimens of a species of Doxomysis from Port Blair are held over to await further material. They differ from D. littoralis in having the whole of the carapace and abdomen covered with fine spinules which even extend to the eyestalks, in the shorter antennal scale which barely outreaches the antennular peduncle, in the slightly different arrangement of the spines on the telson (text-fig. 18) and in the longer uropods, the inner being one and a half and the outer twice as long as the telson. They agree with D. littorvalis and differ from all the other described species in having the spines on the lateral margins of the telson extending throughout their whole length. The rostral plate is trian- gular in shape, with the apex bluntly pointed and quite short, while the eyes resemble the smaller eyes of D. littoralis rather than the larger eyes of the pelagic species. The specimens are rather damaged and though they appear to represent a new spe- cies, it does not seem advisable to give them a name at present. LEXT-FIG. 18.—Doxomy- sts Sp. Telson, X 65. Doxomysis anomala, sp. nov. Text-figs. 19a-/. Localities.—Port Blair, Andaman Islands. St. 3. One hundred and twenty specimens, up to 5'5 mm, - (Types.) St. 32. One female and two males. Description.—Body smooth without spinules; carapace short, leaving the last three thoracic somites exposed dorsally, produced in front into a short triangular rostral plate with an obtuse apex, not covering the eyestalks. 1922. ] W. M. TarrerRsALL: Indian Mysidacea. 481 Eyes of moderate size, cornea occupying one half of the whole eye and wider than the peduncle, pigment brown. TExT-FIG. 19.—Doxomysis anomala, sp. nov. a, anterior end, X 65; 5, antennal scale, x 65; c, endopod of fourth tho- racic limb, X 65; d, fourth pleopod of male, x 65; e, telson, x 65; f, uro- pods, X 65. Antennal scale extending for about one-third of its length beyond the antennular peduncle, lanceolate in shape, six times a 482 Records of the Indian Museum. [VoL. XXIV, long as broad, setose all round, terminal joint about one-seventh of the whole scale, a prominent spine on the outer distal corner of the basal joint. Mouth-parts agreeing with those of the genus Doxomysis as described by Hansen except that the exopod of the second maxilla has more setae than Hansen figures though these setae are quite short and feeble ; setose lobes well developed on the second, third and fourth joints of the first thoracic legs. Endopods of the posterior thoracic limbs very slender, tarsus with three joints, the proximal articulation very oblique, the distal articulation slightly oblique, nail well developed. Telson three quarters of the length of the last abdominal somite, one and a half times as long as broad at the base, cleft for one-fifth of its length, cleft armed on each side by six small articulated spines, a pair of plumose setae at the base of the cleft longer than the cleft, lobes at the apex truncate, each armed with three spines, the outer spine stouter and twice as long as the inner pair, lateral margins armed with 7-8 spines, three larger and stout- er ones on the proximal portion at the widest part, 4-5 on the distal portion. Inner uropod one and a half times as long as the telson with row of about 20-25 closely set spines on the inner margin extend- ing from the statocyst almost to the apex. Outer uropod twice as long as the telson. Fourth pleopods of the male with the exopod longer than the endopod, the antepenultimate and penultimate joint each bearing a long stout seta feathered at the distal end, the terminal joint with a single short simple seta. Female with two pairs of incubatory lamellae. Length of adult specimens of both sexes, 5°5 mm. Remarks.—Hansen in his monograph of the ‘ Siboga’ Mysidae distinguished the tribe Erythropini from the tribe Leptomysini, among other characters, by the fact that the proximal articulation of the tarsus is oblique. The present species, which from the structure of the second maxilla, antennal scale and pleopods in the male is clearly a member of the Leptomysini, presents the anomalous character of two oblique articulations defining the joints of the tarsus of the thoracic limbs. The proximal articula- tion is very oblique, quite as oblique as in any of the Erythropini but the distal articulation is only slightly oblique. Tribe MysINI. Genus Mesopodopsis Czerniavsky. Mesopodopsis orientalis (Tattersall). Macropsis orientalis, Tattersall, 1908, 1914, 1915. Locality .—Balliaghatta Canal, near Calcutta, in brackish water. Abundant. 1922.] W. M. 'TarreRsALL: Indian Mystdacea. 483 Bay N.-W. of Nazareth Point, Mormugao Bay, Portuguese India, nine specimens. Chilcolna Bay and stream at its southern end, Mormugao Bay, Portuguese India, one specimen. Remarks .—It is unfortunately necessary to alter the name of the genus. The name Macropsis, proposed by Sars in 1877, had already been applied by Lewis in 1836 to one of the Hemiptera and the genus must therefore be known by the name proposed as a subgenus by Czerniavsky in 1882. Genus Neomysis Czerniavsky. Zimmer (1915(1)) has united with the genus Neomysis Czerniavsky, the genus Acanthomysis Czerniavsky (= Dasymysis Holt and Beaumont, Metamysis Nakazawa (not Sars), Ortentomy- sis Derzhavin) on the grounds that the distinctions between these genera have broken down in the light of the species described by Nakazawa and Derzhavin. In the structure of the pleopods of the male both genera are identical and the differences lie mainly in the antennal scale and in the tersus of the thoracic legs. In Neomysis the antennal scale is very long, with a sharply pointed apex, and the tarsus of the thoracic legs is many jointed. In Acanthomysis the antennal scale is short, the apex rounded or truncate, and the tarsus of the legs three-jointed. But Meta- nvysis mitsukurit Nakazawa, has the antennal scale of Acantho- mysis and the tarsus of the thoracic legs six-jointed, i.e. as found in Neomysis. Both the species of Ovientomysis described by Derzhavin have many joints, 4-8, in the tarsus of the thoracic legs, but the antennal scale is short with a rounded apex. On the whole Zimmer appears to be right in uniting these genera. It is difficult to seize upon any constant character separating them. The type of the genus Acanthomysis is A. longicornis (M. Edw.) from the Mediterranean and the new species J describe below could without difficulty be referred to this genus. It agrees very closely with 4. longicornis and is only distinguishable by characters which cannot be regarded as of more than specific value. But I have foilowed Zimmer in his arrangement and described the species under the genus Neomysis. Neomysis indica, sp. nov. Text-figs. 20a-7. Localities —Port Blair, Andaman Islands. Stations 3, 5, I1 and 1g. Ten females and 3 males, up to6'5 mm. Kilakarai, Gulf of Manaar, from weeds, o-2 fathoms, February, 1913. Six females and 10 males up to8 mm. (Types.) Chilcolna Bay, and stream at its southern end, Mormugao Bay, Portuguese India. One specimen. Description.—Very closely allied to Neomysis longicornis (M.-Edw.). Body, including the eyes, and last pair of brood 484 Records of the Indian Museum. [VoL. XXIV, TEXT-F1G, 20.—Neomysis indica, sp. nov. a, eye and antennular peduncle, x 33; 0, rostrum, X 33; ¢, antennal scale, x 65; d@. fifth thoracic limb, X 50; e, fourth pleopod of male, x 65; /, filth pleopod of male, x 65; g, uropods, x 65; h, telson of adult, x 65; 2, telson of young, X 65. £522;) W. M. TatrerRsauL: Indian Mysidacea. 485 lamellae in the female, hispid all over, the spinules thickest on the posterior segment of the abdomen and on the anterior part of the thorax. Fully grown males appear to be much smoother than young males and females. Carapace produced in front into a short triangular rostral plate with pointed apex. Eyes large, cornea wider than the rest of the eye, pigment black. Antennal scale barely outreaching the antennular peduncle, seven times as long as broad, terminal joint about one-tenth of the total length of the scale. Tarsus of the thoracic limbs three-jointed. Telson one and a half times as long as the last abdominal somite lanceolate in shape, entire, about twice as longas broad at the base, suddenly narrowing a short distance from the base and gradually narrowing from that point to a bluntly rounded apex, the proximal part of the lateral margins smooth except for three small spines on each side of the widest part of the telson, apex armed with from 6-8 strong spines of equal length with no smaller spines between them, distal part of the lateral margins armed with numerous spines of varying sizes, about seven to nine of these spines much larger than the rest and placed more or less at regular intervals, between them groups of smaller spines, 3-5 in a group, grading in size, the smaller ones anterior and the larger ones posterior. In small specimens the large spines arming the telson are relatively more prominent than in larger specimens and the telsons of both sizes look strikingly different. But I think the difference is entirely due to differences in size and the development of the subsidiary spines. Inner uropod slightly longer than the telson with a group of five graded spines on the lower inner margin, near the statocyst. Outer uropod one-seventh longer than the inner. In young specimens the uropods are more equal in size. Fourth pleopod of the male, with its terminal setae, not reaching as far as the telson, endopod well developed, of normal form, exopod two-jointed, the terminal joint about one-seventh of the Jength of the proximal joint and bearing two long stout plumose setae about three times as long as the joint itself. Fifth pleopod of the male much longer than the first, second and third, but of the same form. It is nearly as long as the sixth abdominal somite and the apex bears two specially strong and long plumose setae. Remarks .—This species is distinguished from N. longtcornts at once by the armature of the telson. In N. longtcornis the spines arming the telson are more equal in size and there are not any out- standing spines of much greater length than the rest. InN. stellevi Derzhavin (1913) the telson has special spines of outstanding length separating groups of spines, but the spines on the telson extend along the entire margin, whereas in N. indica the proximal 486 Records of the Indian Museum. [VoL. XXIV, portion of the margins of the telson is unarmed. Moreover N. stellert has the tarsus of the thoracic legs six-jointed. N. indica is the Indian Ocean representative of N. longicornis and provides additional evidence of the close similarity of the Mysidacean fauna of the Mediterranean and Indian Ocean, so clearly exemplified by this collection. Neomysis hodgarti, sp. nov. Text-figs. 21a-f. Locality.—Mouth of the Rajang River, Sarawak, Borneo, I-vii-I0. Four males and fifteen females, up to 7 mm. in length, collected by R. Hodgart. (Types.) escription.—Body smooth, without spinules on either thorax or abdomen. Carapace produced into a short broadly triangular rostral plate with bluntly pointed apex, antero-lateral corners rounded. . Eyes reaching to the end of the second joint of the antennular peduncle, normal in shape, pigment black. Antennal scale extending slightly beyond the distal end of the antennular peduncle in the female and level with the male process of the antennules in the male, narrowly oval in shape, setose all round, about five times as long as broad, terminal joint distinct, a strong spine on the outer distal corner of the joint from which the scale springs. Labrum with a very long sharp forwardly directed spine. Tarsus of the thoracic legs composed of five to six joints and terminated by a slender nail. Last segment of the abdomen slightly longer than the fifth. Telson one and a quarter times as long as the last abdominal somite, narrowly linguiform in shape, not quite twice as long as broad at the base, apex en- tire, without cleft, almost truncate, onv-fifth of the width of the telson at the base, bearing four equal strong spines in length about one-eighth of the total length of the telson, distal two- thirds of the lateral margins armed with about 26-28 short, closely set and regularly arranged spines increasing in length towards the apex. Inner uropod about one-sixth longer than the telson plus the terminal spines, with a single spine on the lower surface near the statocyst. Outer uropod one-sixth longer than the inner. First, second, third and fifth pleopods of the male simple uajointed plates as in the female. Fourth pleopod reach- ing almost to the base of the telson, endopod with well developed side lobe, exopod composed of three joints, first joint very long, three and a half times as long as the second, latter bearing a very long straight simple seta three times as iong as the joint, terminal joint very minute with a single short seta at the apex. Length of the largest specimens of both sexes, 7 mm. Remarks.—This species belongs to the N. longicornis group of the genus and is distinguished by the form of the telson with its apical armature of four equal stout spines, the smooth dermis, and the number of joints in the tarsus of the thoracic limbs. 1922. ]} W. M. Tarrersauy: Indian Mysidacea. 487 el Ee TEXT-F1G. 21.—Neomysis hodgarti, sp. nov. a, antennal scale and peduncle; 6, endopod of the second thoracic limb ; c, distal end of the endopod of the third thoracic limb; d, telson; e, uropods ; Ff, fourth pleopod of the male. All x 57. Genus Potamomysis Czerniavsky. Potamomysis assimilis Tattersall. Text-fig. 22. Potamomysis assimilis, Tattersall, 1908, 1914, 1915. i A Zimmer, 1915 (1), p. 215, fig. 19. Locality.—Ganges, near Buxar, Bihar, on surface. Several specimens (T. Southwell). ‘‘ These specimens, captured over 600 488 - Records of the Indian Museum. [VoL. XXIV, miles from the sea, were taken at a point much further inland than any previously recorded.’’ (N. A.). R.marks.—Zimmer in his revision of the Mysini was unable to place this species and genus because he was of opinion that the male from which I figured the fourth pleo- pod was not mature. I find that this opinion is correct. Among the present specimens are two or three fully grown males, somewhat lar- ger than I have examined before, and I find that the drawing I gave of the fourth pleopod of the male requires modification. The outer branch is five-jointed not three-jointed as I have described it, the third joint of my pre-- vious figure being subdivided into two. extra small joints at the distal end. Each of these last three joints bears a single plumose seta at its outer distal end, those on the terminal and ~ penultimate joints about equal in size, while that on the antepenultimate joint is much longer and stouter, and is plumose and not smooth as I had previously described it. Immature speci- mens have the fourth pleopod as I had previ- ously described it, that is with the two small terminal joints of the outer branch not marked off and the large seta on the third joint smooth and not plumose. The fourth pleopod in this species is remark- ably like that of the genus Stilomysis, but the Trxt-ric. 22— third pleopod of the latter is two branched Potamomysts whereas in P. assimilis it is rudimentary as in OTE Ss the female. Tattersall. Foutthieplesnod It is now possible to put this genus in its place of the adult male, 1m Zimmer’s key. It should be placed in Group x 65. III B in which it will form a separate section, characterised by the large number of joints in the outer branch of the fourth pleopod of the male. — Genus Idiomysis, nov. Body robust and gibbous, in the only specimen, flexed in the curious way shown in text-fig. 23a. Carapace produced in front into a large frontal plate with a broadly rounded apex. Eyes very large, stalks short, pigment golden brown. Antennular peduncle with the male lobe well developed and densely hirsute. Antennal scale very short and broad, the greater part of the outer margin smooth, without setae, no transverse suture distally. Second maxilla with the outer plate very reduced, with only four or five setae. First thoracic limbs with a masticatory lobe on the second joint only. ‘Tarsus of the third fourth, sixth, seventh and eighth thoracic limbs unjointed. Fifth thoracic limbs markedly 1922.] W. M. TATrEeRSALL: Indian Mystdacea. 489 longer than any of the others, tarsus two-jointed, nail prominent. Eighth thoracic limbs much reduced in size, shorter than the exopod and slender. Telsona very short broad triangular plate, not covering the statocyst, unarmed. Uropods short, robust, subequal, without spines, statocyst large. First, second, third and fifth pleo- pods of the male small unjointed plates, fourth pleopod of the male consisting of a basal joint from which the exopod and endopod are imperfectly separated, a small one-jointed endopod and a large exopod, apparently consisting of a single joint terminated by a single long stout seta, as long as the joint and imperfectly annulose at the tip, the whole appendage reaching to the tip of the uropods. Type :—Idomysts inermis, sp. nov. ; Remarks.—This curious genus is quite unlike any other Mysid known tome. Its robust gibbous body and the set of the telson and uropods remind one of Mystdopsts gibbosa, superficial.y, but it is widely removed from that species in structure. It belongs to the tribe Mysini and apparently to Zimmer’s group III B. It is abundantly characterized by the very short and broad antennal scale and by the short unarmed telson. Idiomysis inermis, sp. nov. Text-figs. 23a-/, 24a-g. Locality.—Kilakarai, Gulf of Manaar, among weeds, 0-2 fathoms, February, 1913. One adult male, 4mm. (Type) Description.—Body robust, gibbous, abdomen flexed between the third and fourth somites; carapace leaving the last thoracic somite exposed, produced in front into a prominent triangular fron- tal plate with a broadly rounded apex which reaches forward to the distal end of the first joint of the antennular peduncle. Eves very large, cornea much broader than the stalk, pigment golden brown. Antennal scale barely outreaching the short antennular pe- duncle, very short aud broad, one and a half times as long as broad, outer margin without setae and without a prominent distal spine, terminal lobe broader than long, without a suture, no spine on the outer distal corner of the joint from which the scale springs ; peduncle of the antenna shorter than the scale and composed of three short broad subequal joints. Mandible with a well-developed cutting edge and molar process, second joint of the palp broad. First maxilia with the inner plate reduced in size and tipped by four or five feeble setae. Second maxilla with the outer plate very much reduced in size and feebly armed with five small setae, second gis of the palp linear, not expanded at the apex. First thoracic limb (first gnathopod) with a masticatory lobe, tipped by two or three setae, on the second joint, no lobes on the third and fourth; nail not more Ce ee than the other setae arming the distal joint. 490 Records of the Indian Museum. [VoL. XXIV, Third, fourth, sixth and seventh thoracic limbs of similar structure, tarsus unjointed, equal in length to the merus and to the ischium, nail only distinguishable from the other setae by its swollen base. Fifth thoracic limb longer than any of the others, merus at Text-¥iG. 23.—/diomysis inermis, gen. et sp. nov. a, lateral view of adult male, x 21; 6, dorsal view of anterior end, X 22; c, antennal scale, x 65; d, mandible, x 100; e, first maxilla, x 100; 7, second maxilla, x 100. least twice as long as the preceding joint, tarsus about one-third longer than the merus but much narrower, two-jointed, the second joint one-third of the length of the first, nail with a swollen base. Eighth thoracic limb much reduced in size, endopod possess- ing the full number of joints, shorter than the exopod, joints narrow and feebly armed. Exopods of all the thoracic limbs with 1922.] W. M. 'TATTERSALI,: Indian Mvsidacea. 491 the outer distal corner of the hasal joint rounded, flagellum with about ten joints. TEXT-FIG. 24.—/diomysis ineymis, gen. et sp. nov. a, endopod of first thoracic limb, x 65; 4, second thoracic limb, x 65: c, endopod of third thoracic limb, x 65; d, endopod of fifth thoracic limb, x 65; e, endopod of eighth thoracic limb, x 65; f, fourth pleopod of male, x 65; g, telson, x 65. Telson very short, much shorter than the last abdominal somite and barely extending to and not covering the statocyst ; a 42 Records of the Indian Museum. [Vor. XXIV, broad triangular plate, rather broader than long, apex bluntly rounded and entire; telson quite unarmed. Inner and outer uropods short broad plates, equal in size, furnished with long setae on both margins, inner uropod without spines on its lower margin, statocyst large. First, second, third and fifth pleopods of the male simple small unjointed plates, fourth pair with the endopod of normal shape, consisting of a small unjointed plate with terminal setae and a well marked side lobe, imperfectly marked off from the basal joint, exopod consisting of a single long linear joint terminated by a single stout seta, annulose at its tip, the whole exopod extending to the tip of the uropods or slightly beyond. Length of the only speci- men, an adult male, 4 mm. Remarks.—This species shows no very great affinities with any described form. The form of the pleopods of the male clearly shows its place in the tribe Mysini, but the shape of the antennal scale and the very short unarmed telson are quite unlike any member of that tribe. I hope that female specimens will be forthcoming some day so that the sexual differences may be des- .cribed. It is the most interesting and distinctive species in this collection. Genus Lycomysis Hansen. Lycomysis spinicauda Hansen. , Text-figs. 25a-c. Lycomysis spinicauda, Hansen, 1910, p. 77, pl. xi, figs. 3a-f, pl. xii, figs. 2a-h. i Colosi, 1916, p. 194, text-figs. 1a-d. * A Colosi, 1918, p. 10. ; An Colosi, 1920, p. 251, pl. xx., figs. 10a-g. Mi pusilla, Zimmer, 1915, p. 175, figs. 30-37. Locality —Port Blair, Andaman Isles. » St. 19. One male, 4°5 mm. Remarks.—I have no doubt as to the identity of this speci- men with the species described by Zimmer. ‘The agreement is, in all points,complete. But I am somewhat puzzled as to the relation of L. pusilla to L. spinicauda and, after due consideration, I have reached the conclusion that the two species are identical, the sup- posed differences being due to a difference of interpretation of the structure of the male pleopods, the only character separating the two forms. Hansen (1910) describing L. sfinicauda irom an immature male says (p. 76), ‘‘ Pleopods in the male immature specimens small, biramous, with the exopod [endopod] increasing in length back- wards, being on the anterior pairs shorter, on the fourth pair some- what longer, than the exopod, on the fifth pair twice as long as the exopod, but very far from developed ’’ and later (p. 77) .‘‘ unfor- tunately the male pleopods are so imperfectly developed in my specimens, that they cannot afford any real help for deciding the systematic position of Lycomysis, yet it may be stated that they 1922.] W. M. TatrersaLi: Indian Mysidacea. 493 show that it cannot be referred to the Mysini, and that the exo- pod [endopod] of the fifth pair being twice as long as the endopod [exopod] is somewhat ano- malous.’’ ‘There is some- thing inconsistent in both these statements and Han- sen appears to have got his terms exopod and endopod mixed. I have given in square brackets the term which I think the author intended. At any tate, my interpretation is supported by Hansen himself earlier in the same paper (p. 13) where, when emphasising the supposed anomalous con- dition of the male pleopods, he states that Lycomysts differs from all hitherto known genera ‘‘in having the endopod of the poste- rior pairs of pleopods longer than the exopod and the en- dopod of the fifth pair longer than that of the fourth.”’ From these quotations it is clear, I think, that Han- sen regarded the pleopods of the male of Lycomysis as biramous, and that, in con- sequence, it could not be referred to the tribe Mysinz. Colosi (1916) gives a new diagnosis of L. spinicauda based on a male specimen from the China Sea. He walt Auer ec the gleopads reiie TEXT-FIG. 25. Bi Ailes spinicauda, male in the genus as follows a, third pleopod of male, x 100; 6, fourth “‘Primo, secondo, terzo e pleopod of male, x 100; c, fifth pleopod of quinto pajo con endopodite ale, X 100. ed esopodite rudimentali; quarto pajo con pedunculo piu lungo che largo, endopodite rudimen- tale ed esopodite lunghissimo composto di tre articoli, di cui il primo piu lungo degli altri, e terminato da due filamenti (spiniformi ?).”’ In the diagnosis of the species Colosi gives slightly fuller details especially of the fourth pair of pleopods, from an adult male; which have the exopod greatly elongated, three-jointed, the third joint bearing two terminal filaments. Colosi figures the fourth pair of 494 Records of the Indian Museum. [VoL. XXIV, pleopods but none of the others. His description, however, agrees with that of Hansen, in stating that the first, second, third and fifth pairs of pleopods in the male are bivamous with exopodite and endo- podite. In spite of this anomalous form of the pleopods, Colosi places Lycomysis spinicauda in the tribe Mysini of the subfamily Mysinae. But ifthe descriptions of Hansen and Colosi are correct this position for the species cannot be maintained, for in the Mysini at least the first and second and in most cases the fifth pair of pleopods of the male are simple unjointed plates as in the females of Mysidae generally, without any definite indication of a separate endopod and exopod. In the meantime Zimmer (1915 (3)) described a second species of Lycomysis, L. pusilla. It is evident that Zimmer was puzzled by Hansen’s description of the pleopods in his species for Zimmer’s new species is founded entirely on the characters of the pleopods of the male, the author stating that in all other characters his species was identical with L. spinicauda. Zimmer (1915 (3), p. 175) describes the pleopods of the male in L. pusilla as follows :—“'Die paare I, 2, 3 und 5 rudimentar, wahrend 4 einene eingliederigen Innen— und stark verlangerten Aussenast besitzt”” and later (p. 177) he states that the pleopods 1, 2, 3 and 5 of the male are as in the female. He gives a figure of the first pleopod of the female which shows this appendage as a simple unjointed plate, somewhat bilobed at the apex, each of the lobes bearing setae. His description of the pleopods of the female states that in the first three pairs the two lobes are more or less equal in size, but in the last two pairs the inner lobe is much longer than the outer. The fourth pair of pleo- pods of the male have an endopodite which corresponds with the female pleopod in shape and a very elongate exopod of three joints terminated by a single long plumose seta. Zimmet’s species is clearly referable to the Tribe Mysini and the present specimen agrees absolutely with his description and figures in the matter of the pleopods of the male. Colosi (1920) gives some further notes on L. spinicauda, Hansen, and compares it with L. pusilla, Zimmer. His description of the pleopods of the male of L. spinicauda is substantially as in his previous paper and he points out that the two species are distinguished not only by the pleopods but. by the characters of the mandibular palp, the terminal joint of which is longer and nar- rower in L. pusilla than in L. spinicauda and the teeth on the margin of the second joint less well marked in the former than in the latter. I regard these latter differences between the two species as of no moment and due mainly to the fact that Zimmer’s figure is taken from asomewhat more oblique point of view than Colosi’s. But the differences in the pleopods of the male are more puzzling. It is almost inconceivable that two species so essenti- ally alike in all other details that female specimens could not be distinguished one from another, should differ so profoundly in the structure of the male pleopods that adult males should require to 1922.] W. M. TATTERSALL: Indian Mysidacea. 495 be placed in separate subfamilies at least. I can only suppose that Hansen and Colosi are in error in describing the first, second, third and fifth pleopods in the males of L. spinicauda as biramous, with endopodite and exopodite defined. I give herewith a figure of the third, fourth and fifth pleopods of my specimen. ‘The outer lobe (seitenlobus) is remarkably well developed in all the pleopods and in the first three is as long as the inner lobe (haup- teile). Zimmer makes the same observation. In the fourth and fifth pair the inner lobe is much the longer, in the fifth pair longer than in the fourth. At first sight the appendages look biramous and it is only when dissected and examined under the high power of the microscope that they are found to be simple unjointed plates of the type usual in the females of Mysidae except that the outer (side lobe) is unusually well developed. If we suppose the words endopod and exopod in Hansen’s statements to be replaced by inner lobe and outer lobe, Hansen’s description of L. spinicauda applies equally well to L. pusilla. In fact, if my suggestion is correct the two species should be united under the name spinicauda and it is in this light that I have regarded them here. Disiribution:— L. pusilla was recorded by Zimmer from a collection made during a voyage from Ceylon to New Guinea. L. spinicauda is known from the waters of the East Indian Archipe- lago (Hansen) and the China Sea (Colosi). The distribution of the two forms is therefore not inconsistent with their suggested specific identity. Tribe HETEROMYSINI. Genus Heteromysis, S. I. Smith. Syn. Chiromysts G. O. Sars. Gnathomysis Bonnier and Pérez. Through the kindness of Professor C. Pérez, I have been per- mitted to see a series of unpublished drawings made by the late Dr. Jules Bonnier to illustrate the general form and the structure of the appendages of Guathomysis gerlachet (Bonnier and Pérez, C. R. Acad. Sci. Paris, T. 134, p. 117-I19, 1902), a preliminary description only of which, without figures, has so far appeared. I have not been able to examine the specimens from which the drawings were made but there is no doubt in my mind, after studying the drawings sent to me by Professor Pérez, that Gnathomysis gerlachet is identical with Chiromysis harpax Hilgen- dorf. The genus Guathomysis is therefore a synonym of Chiro- mysis G. O. Sars, which in turn must give way to the earlier Heteromysis S. I. Smith. Heteromysis harpax was described by Hilgendorf in 1879 in very summary fashion. Kossmann (1880) redescribed the species from examples collected in the Red Sea and figured the append- ages in some detail. It is difficult to be sure that Kossmann des- 496 Records of the Indian Museum. [VoL. XXIV, cribed and figured the same species as Hilgendorf, but the form of the third thoracic limb, more especially of its armature and the gap in the series of spines arming the inner margin of the carpus, which is indicated in Hilgendorf’s figures and clearly shown by Kossmann, renders it at least probable that both authors were dealing with the same species. The unpublished drawings of Bonnier agree in the greatest detail with Kossmann’s figures. The agreement is complete in the figures of the third thoracic limb, even to the peculiar form of the ischium and to the arrangement, number and forms of the spines arming the inner margin of the carpus. In only one respect do Bonnter’s and Kossmann’s figures differ, namely, in the armature of the telson. Kossmann gives no details in his description and his figuie indicates that the lateral margins of the telson are armed distally with five spines, that each apical lobe of the telson bears one long spine and that each side of the cleft bears five spinules. Bonnier’s figure shows that the lateral margins of the telson are armed distally with eleven spines, that each lobe of the telson bears two spines, the outer of which is stouter and twice as long as the inner, and that each margin of the cleft is armed with ten spinules. I conclude therefore, that Gnathomysis gerlachet Bonnier and Pérez must be considered undoubtedly as a synonym of Hetero- mysts harpax Kossmann, which is almost certainly the same as H. harpax Hilgendorf. Heteromysis proxima, sp. nov. Text-figs. 26a-e. Locality ate an pools on exposed reef at Pamban, Gulf of Manaar. One male and one female, 6-7 mm. (Types.) Description :—Kye small, longer than wide, without finger- like process, cornea occupying less than half of the eye in dorsal view and narrower than the rest of the eye. Antennal scale equal in length to the antennal peduncle and shorter than the anten- . nular peduncle, three times as long as broad, setose all round, terminal joint present. Third thoracic limb with the endopod massive and stoutly built, ischium with the inner distal corner not produced into an acute process, merus three times as long as broad with a prominent blunt process at the distal end of the inner margin, carpus stouter and longer than the merus, about two and a haif times as long as broad, inner margin armed with spines of three kinds (1) stout simple spines, three or four in number on.the distal part of the margin, (2) stout spines with a truncate apex, one or two in number at the distal end, and (3) slender spines bearing a single seta inserted some way from the tip, extending all along the margin in two double rows of eight or nine spines, between which the other spines are situated ; penultimate joint small with an acute process at the distal end of the inner margin; nail long and strongly curved, one-third of 1922.] W. M. TATTERSALL: Indian Mysidacea, 497 é. TEXT-FIG. 26.—Heteromysis proxima, sp. nov. a, eye, X 50; 3, antennal scale, x 65; c, endopod of third thoracic limb of male, X 65; d, endopod of fourth thoracic limb, x 65; e, telson, x 65. 498 Records of the Indian Museum. [VoL. XXIV, the length of the carpus. ‘Tarsus of the remaining thoracic limbs five-jointed, at least equal to the merus which is one and a half times as long as the ischium; nail small and setiform. ‘Telson slightly longer than the last abdominal somite, not quite one and a half times as long as broad at the base, cleft for one quarter of its length, the cleft armed on each side by ten coarse teeth, apex about one quarter of the breadth at the base, each lobe at the apex armed with an inner small spine and an outer larger and stouter spine which is twice as long as the small spine and about one-eighth of the length of the telson, lateral margins straight, distal half armed with 10-12 spines, proximal half of the margins smooth, without spines. Inner uropod slightly longer than the telson plus the termi- nal spine at the apex, a single spine on the lower inner margin near the statocyst. Outer uropod about one quarter as long again as the telson. Remarks:—Of the three Indian species of Heteromysis here recorded, this species approaches most closely to H. harpax (Hil- gendorf). I was inclined at first to regard my specimens as be- longing to Hilgendorf’s species, but after an examination of the unpublished drawings of Bonnier, illustrating the structure of Gnathomysts gerlachet which I regard as identical with H. harpax, I have decided that the Ceylon specimens represent a distinct species differing mainly in the form and armature of the third thoracic limbs. In H. harpax as figured by Kossmann and also by Bonnier among his unpublished drawings, the ischium of the third tho- racic limbs has the inner distal corner produced and acute and the distal margin minutely toothed or serrate. The carpus is armed on its inner margin with a group of four spines distally and two spines proximally, with a distinct gap, unarmed, between the two sets of spines. The distal spines are truncate at the apex and microscopically toothed. The proximal spines are bluntly pointed and bear two or three small blunt teeth. The inner distal angle of the propodus is bluntly produced. In H. proxima the ischium of the third thoracic limbs is not produced at its inner distal angle and the distal margin is not serrate. The carpus has the inner margin armed with two rows of eight or nine peculiar spines with a seta inserted near the tip, extending in a continuous line, without gap, along the greater part of the margin. Between these two rows of peculiar spines, on the distal part of the margin, are three or four stouter, blunter spines and at the extreme distal angle one or two stout spines with a truncate apex. The inner angle of the propodus is more acutely produced than in H. harpax. The close agreement between Kossmann’s and Bonnier’s drawings of the third thoracic limbs of the specimens they ex- amined is strong evidence of the identity of these specimens and also of the probability that the Indian specimens represent a distinct species. ‘The differences I have noted are not sexual, 1922.] W. M. TarrersaLL: Indian Mystdacea. 499 since both Kossmann’s and Bonnier’s figures were drawn from an adult male and my own figure is likewise drawn from an append- age belonging to that sex. In other respects H. proxima agrees closely with H. harpax and with H. microps from the Mediterranean. All three agree in having small eyes with processes, in having only one spine on the inner uropods, in having only the distal half of the margins of the telson armed with spines and in the number of joints in the tarsus of the thoracic limbs. ‘The form and armature of the third thoracic limbs is, however, quite distinct in all three. Heteromysis zeylanica, sp. nov. Text-figs. 27a-e. Localittv.—From pools on exposed reef at Pamban, Gulf of Manaar. One male, 5 mm. Kilakarai, Gulf of Manaar, from weeds, 1-2 fathoms. One male, 5 mm., two immature. (Types.) Description.—Eye small, longer than broad, a short pointed process on the upper distal border overhanging the cornea, latter occupying less than one half of the eye in dorsal view and nar- rower than the rest of the eyes. Antennal scale slightly shorter than the antennal peduncle and considerably shorter than the antennular peduncle, three and a half times as long as broad, setose all round, terminal joint present. Third thoracic limb with the endopod moderately stout, merus three times as long as broad without process at the distal end of the inner margin, carpus robust, shorter than the merus, twice as long as broad, inner margin armed with four or five spines each with an inserted seta, penultimate joint small without process, nail strongly curved about half as long as the carpus. Remaining thoracic limbs having the endopods moderately stout with the tatsus four-jointed, merus equal to the ischium and longer than the tarsus, nail setiform and curved. Telson about as long as the last somite of the abdomen, one and a quarter times as long as broad at the base, cleft for rather more than one quarter of its length, the proximal half of each margin of the cleft armed with seven teeth, distal half of each margin of the clefc smooth, apex rather less than one quarter of the breadth of the telson at its base, each lobe of the apex furnished with two spines, the inner about half as long as the outer which is about one seventh of the length of the telson, lateral margins lightly concave the proximal portion with five spines at the widest part, the central portion smooth, the distal portion with about eight or nine spines arranged at more or less regular intervals, the interval between the last marginal spine and the large apical spine not greater than that between the other distal marginal spines. Inner uropods about one quarter as long again as the telson 500 Records of the Indian Museum. [Vo.. XXIV, plus the terminal spines, inner lower margin armed with about eleven stout spines from the statocyst to just short of the apex, the spines increasing in size distally. . Outer uropod about half as long again as the telson. Remarks.—This species belongs to that group-cf species characterised by the presence of a distinct process on the eye. TExT-r1G. 27.—Heteromysis zeylanica, sp nov. a, eye, X 50; 6b, antennal scale, X 65; c, endopod of third thoracic limb, X 65; @, endopod of fourth thoracic limb, x 65; e, telson, X 65. H. odontops Walker is the type of this group. It is distinguished from H. harpax by this character, by the telson having spines on the proximal wide part of its margins, by the details of the armature of the cleft and by the row of stout spines on the inner margin of the inner uropod, as well as by the less robust form of the third thoracic limb. Heteromysis gymnuta, sp. nov. Text-figs. 28a-e. Locahty.—Kilakarai, Gulf of Manaar, among weeds, 1-2 fathoms, Five males and two females, 4-6 mm. (Types.) Descrniption.—Eye large, at least as wide as long in dorsal view, no finger-like process, cornea occupying more than half the eye in dorsal view and wider than the rest of the eye. Antennal scale Jonger than the antennal peduncle and equal in length to the antennular peduncle, three anda half times as long as broad, no terminal joint. 1922.] W. M. TATTERSALL: Indian Mysidacea. 501 Third thoracic limb less robustly built than in the other two species, merus nearly four times as long as broad without distal process, carpus shorter than the merus, three times as long as broad, the inner margin armed distally with three rather stout simple spines, penultimate joint short without inner process, nail strongly curved, about one third of the length of the carpus. Remaining thoracic limbs slender with the tarsus composed | eX, C. E. TExt-FiG. 28.—Heteromysis gymnura, sp. nov. a, eye, X 50; b, antennal scale, x 65; c, endopod of third thoracic limb, Xx 65; d, endopod of fourth thoracic limb, x 65; e, telson, x 65. of three joints terminated by a distinct nail, merus almost equal to the tarsus and nail together, ischittm- one and a half times as long as the merus. Telson slightly longer than the last abdominal somite, one and three quarter times as long as broad at the base, cleft for more than one third of its length, the margins of the cleft armed through- out with about 25 closely set teeth, apex one quarter of the base of 502 _ Records of the Indian Museum. [Vor XXIV, the telson in width, each lobe armed with two spines, the outer three times as long as the inner and equal to one sixth of the telson in length, lateral margins lightly concave, proximal portion smooth, distal portion armed with 12-15 spines increasing in length distally, the interval between the last marginal spine and the larger spine on the apical lobe greater than the interval between any other pair of marginal spines. Inner uropod equal in length to the telson plus the long ter- minal spine, without any spines on its lower inner margin. Outer uropod one-third as long again as the telson. Remarks.—This species is much less specialized than the other two, ‘The eyes are much larger, the antennal scale proportionately longer and the third thoracic limb much less robustly built. The posterior thoracic limbs are noteworthy for their slender build, for the great length of the ischium and for the distinct nail and few joints in the tarsus. The inner uropods are without spines, a character which marks this species as distinct from all other des- -cribed species of the genus. LIST OF REFERENCES. Alcock, A., and Anderson, A. R.S., 1894. An account of a recent collection of deep-sea Crustacea from the Bay of Bengal and Laccadive Sea.—Journ. Asiat. Soc. Bengal, vol. I-X1II, pt. II, no 3, pp. 141-185. Alcock, A., and Anderson, A.R.S., 1899. An account of the deep- sea Crustacea dredged during the surveying season of 1897-98. —Ann. Mag. Nat. Hist., ser. 7, vol. 111, pp. 1-27. Anderson, A. R. S., 1897. An account of the deep-sea Crustacea collected during the season 1894-3.—J ourn. Astat Soc. Bengal, vol. LXV, pt. 11, pp. 88-106. Colosi, G., 1916. Nuova diagnosi e posizione sistematica di Lycomysis spinicauda Hansen.—Mon. Zool. Ital., Anno xxvii, nr. 9, pp. 193-200. Colosi, G. 1918. Nota preliminare sui Misidacei raccolti dalla R. N. ‘‘ Liguria” nel 1903-1905.—Bull. Soc. Entom. Ital., Anno xlix, pp. I-II. Colosi, G., I9z0. Crostacei, Parte IV. Misidacei.—Racc. Planc- ton. R. N. “ Liguria,’’ vol. II, fasc. ix, pp. 227-260. Czerniavsky, V., 1882-3. Monographia Mysidarum imprimis Im- perii Rossici.—Avb. Nat Ges. Petersburg, V,12, 13 and 18. Derzhavin, A., 1913. Neue Mysiden von der Kiiste der Halbinsel Kamtschatka.—Zool. Anz., Bd. XLIII, Nr. 5, pp. 197-204. Hansen, H.J., 1910. The Schizopoda of the Siboga Expedition.— Siboga Reports, No. xxxvii. Hansen, H. J., 1912. The Schizopoda—Mem. Mus. Comp. Zool. H arvard, Vol. XXXV, No. 4.. Hilgendorf, F. , 1879. Die von Hrn. Peters in Mocambique ges- ammelten Crustaceen.—Monats. K. Preuss. Akad. Wiss. Berlin, Jahr. 1878, pp. 782-851, taf. 1-4. 1922.] W. M. TATTeRSALL: Indian Mysidacea. 503 Holt, E. W. L., and Tattersall, W. M., 1905. Schizopodous Crus- tacea from the NE. Atlantic Slope.—Rep. Sea and Inland Fish., Ireland, 1902-3, pt. II, app. IV. Illig, G., 1906. Bericht tiber die neuen Schizopoden-gattungen und Arten der Deutschen Tiefsee-Expedition, 1898-99.— Zool, Anz., Bd. XXX. no. 7, pp. 194-211. Kossmann, R., 1880. Zoologische Ergebnisse Reise in die Kusten des Rothen Meeres, Hft. II, Lief. I, Theil 2: Anomura, pp. 67-140, pls. iv—xv. Kroyer, H., 186r. Bidrag til Kundskab om Krebsdyrfamilien Mysidae.—WNat. Tidsskr., ser. 3, vol. I; pp. I-75, tab. 1-2. Nakazawa, K., 1910. Notes on Japanese Schizopoda.—Annot. Zool. Jap., vol. vii, pp. 247-261, pl. viii. Paulson, O., 1875 (1). Crustacea Mari Rubri, Pars I. Paulson, O., 1875 (2). Carcinological Notes.—Zap. Obshch. Estestv. Kieff, tom. 1V, pp. 27-32, tab. 1. Sars, G. O., 1877. Nye bidrag til Kundskaben om Middelhavets Invertebratfauna I. Middelhavets Mysider.—Arch. Math. Nat., Bd. II, pp. 10-119, tab. 1-36. Sars, G. O., 1870-79. Carcinologiske Bidrag til Norges Fauna, I, Monographi over de ved Norges Kyster forekommende Mysider. Tattersall, W. M., 1906. Report on the Leptostraca, Schizopoda and Stomatopnda collected by Professor Herdman at Ceylon in 1902.—Ceylon Pearl Oyster Fisheries, Suppl. Rep. no. XXXITI. Tattersall, W.M.,1907. Preliminary diagnoses of six new Mysidae. from the West Coast of Ireland.—Ann. Mag. Nat. Hist., ser. 7, vol. XIX, pp. 106-118. Tattersall, W.M., 1908. The Fauna of brackish ponds at Port Can- ning, Lower Bengal. XI. Two new Mysidae from brackish water in the Ganges delta.—Rec. Ind. Mus., vol. II, pt. III, no. 25, pp. 233-239, pls. xxi-xxii. Tattersall, W.M., 191z. On the Mysidacea and Euphausiacea col- ected in the Indian Ocean during 1905.—Tvans. Linn. Soc. London, ser. 2, Zool., vol. XV, pp. 119-136. pls. 6-7. Tattersall, W. M., 1911. Schizopodous Crustacea from the North- East Atlantic Slope. Second Supplement.—Fish. Ireland, Sct. Invest., 1910 (1911), pp. 1-77, 8 plates. Tattersall, W. M., 1914. Further records of Indian brackish water Mysidae with descriptions of a new genus and species.— Rec. Ind. Mus., vol. X, pp. 75-80, pls. xii-xiii. Tattersall, W.M., 1915. Fauna of Chilka Lake. The Mysidacea of the lake, with the description of a species from the coast of Orissa.—Mem. Ind. Mus., vol. V, no. 2, pp. 147-161. Tattersall, W. M., 1918. Euphausiacea and Mysidacea.—Austra lian Antarctic Expedition 1911-14, Sct. Rep., Ser. C, Zool. Bot., vol. Vi, PE..5. Thomson, G.M., 1900. On some New Zealand Schizopuda.—Journ. Linn. Soc.'London, Zool., vol. 27, pp. 482-486, pls 33 and 34. 504 Records of the Indian Museum. [VoL. XXIV, 1922.] Wood-Mason, J., and Alcock, A., 1891 (1). Note on the results of the last season’s dredging.—Ann. Mag. Nat. Hist., ser. 6, vol. VII, pp. 186-202. Wood-Mason,. 1891 (2). On the results of deep sea dredging during the season 18yo-91.—Ann. Mag. Nat. Hist., ser. 6, vol. VIII, pp. 268-286. Zimmer, C., 1915 (1). Die Systematik der Tribus Mysini H. J. Hansen.—Zool. Anz., Bd. XLVI, Nr. 7, pp. 202-216. Zimmer, C., 1915 (2). Zur Kenntnis der Schizopodenfauna Nea- pels.—M itt. Zool. Stat. Neapel, Bd. 22 Nr. io. Zimmer, C., 1915 (3). Schizopoden des Hamburger Naturhist- orischen (Zoologischen) Museums.—Mztt. Nat. Mus. Hamburg, Bd. XXXII, pp. 159-182. Zimmer C., 1916. Crustacea IV: Cumacea und Schizopoda.— Beitr. z. Kennt. d. Meerestauna Westafrikas, Herausg. von W. Michaelsen (Hamburg), Bd. II, pp. 55-66. Zimmer, C., 1918. Neue und wenig bekannte Mysidaceen des Berliner Zoologischen Museums.—M itt. Zool. Mus. Berlin, Bd. 9, Hit. I, pp. 13-26. PARALLEL EVOLUTION IN THE FISH AND TADPOLES OF MOUNTAIN TORRENTS. By N. ANNANDALE, D.Sc., F.A.S.B., Director, and SUNDER Lat, Hora, M.Sc., Assistant Superintendent, Zoological Survey of India. The structural modifications of the fish and of the Batra- chian larvae that inhabit the small mountain torrents of the Oriental Region afford a remarkable instance of parallel evolu- tion on a comprehensive scale. The phenomena they exhibit may indeed, be called communal convergence. One of us! has quite recently discussed these modifications in the fish, while the other ? has from time to time published observations on the external features of the tadpoles. We propose in the present paper to give a short general survey of the facts and to discuss a specific instance in anatomical detail. Speaking generally, modifications in the tadpoles of moun- tain torrents chiefly consist either in the formation of floats for floating away lightly on the surface of the flood, as in some spe- cies of Megalophrys, or in that of ‘‘ suckers ’’ for clinging to fixed objects. These structures in the tadpoles seem to have been evolved independently of any high degree of specialization in the adult frog or toad,’ just as the larva of an insect may be highly modified in correlation with a peculiar mode of life, while the adult remains of an unspecialized type. Identical structural resemblances in the tadpoles mean that a true genetic affinity exists, but similar structures are frequently evolved independ- ently for the same function but along different lines, for ex- ample the oral float on the mouth of the larvae of certain species of Megalophrys * and that on the mouth of Microhyla achatina,’ or the powerful oral sucker of the tadpole of Helophryne natal- ensis® and that of Bufo penangensis.’ In the floats of the two former tadpoles there is a general similarity in form and function, but in the Megalophrys the internal surface of the float bears rows of peculiar horny tooth-like processes and these are replaced ! Hora, Rec. Ind. Mus. XXIV, pp. 31-36 (1922). * Annandale, Rec. Ind. Mus. VIII, p. 29 (1912); ibid. XV, p. 17 (1918) ; Proc. As. Soc. Bengal (n.s.) XIII, p. clxxxvi (1917). * Boulenger, Rec. Ind. Mus. XV, p. 65 (1918). * Hora, Fourn. As. Soc. Bengal (1922). 5 Smith, Fourn. Nat. Hist. Soc. Siam Il, p. 37, figs. Ay-Ag (1916). 5 Hewitt, Ann. Natal Mus. Il, p. 477, pl. xxxix, figs. 5, 6, 7 (1913). ' Flower, Proc. Zool. Soc. London, p. 908, pl. |x, figs. 3, 3a (1899). 506 Records of the Indian Museum. [VoL XXIV, in the Microhyla by soft ridges. Moreover, it is much more evi- dent in the latter tadpole than in the former that the whole float is formed by a hypertrophy of the lower lip. Similarly in the Helophryne and the Bufo there is a structural difference, only to be observed on close examination. The sucker of the Bufo is formed (much in the same general way as the float of M. achatina but correlated with an entirely different function) by a hyper- trophy of the lower lip, while in the Helophryne both lips are equally developed. In this respect the Helophryne closely resembles an unidenti- fied larva discussed by one! of us from the Malabar Zone of Peninsular India. Indeed the structural analogy is so close that it does not seem too much to claim that there is also a homology, in other words that a genetic affinity exists.* We can claim genetic affinity only between those tadpoles in which similar structures are present with a similar function and produced by the same modifica- tions of structures or organs common to widely different forms, but when these conditions occur to such a degree as to produce morpho- logical identity it is not extravagant to do so. Suckers may be produced by the evolution of a new organ, as in the species of Rana described later, or by hypertrophy of the lips, as in the tadpoles of the Helophryne and the Bufo discussed above.. In the latter instance many different stages in the evolution of the perfect structure are known.’ But in species like Rana afghana the peculiar structure seems, so far as our present knowledge goes, to have arisen strictly de novo. That_ it has really done so is of course improbable, but the earlier stages have not been discovered and have perhaps been eliminated. There is no evidence for any homology between the adhesive apparatus and that found on the ventral surface of all very young Batrachian larvae,* It has been observed by one of us both in the Khasi Hills and the Nilgiris that the tadpoles abundant in the large pools of hill-streams, at spots at which the current is not rapid, belong to quite a different type from any we have mentioned, being very large and stout, with powerful tails, comparatively small mouths, no marked structural peculiarities of an obviously adaptive nature and either a very conspicuous or a very dense pigmentation. Good examples of this type are the tadpoles of Rana alticola® in Assam and of R. malabarica in South India. The former has a conspicu- ous black ocellus on the tail and possesses parotid glands which and the adult of this tadpole is a burrowing form and, therefore, liable to escape notice. The Ethiopian afiinities of the fauna of the Malabar Zone are well re- cognised by Zoogeographers. 8 Annandale, Rec. /nd. Mus. VIII, p. 19 (1912). 4 Boulenger, Rec. Ind. Mus. XX, pp. 100, 168 (1920). t Thiele, Zettsch. wiss. Zool. XLVI, p. 75, pl. x, fig. 6 (1888). 1922.)} N. ANNANDALE & §. L. Hora: Parailel Evolution. 507 dense and uniform black colour. Similar conditions are found as regards structure in the fishes of large pools of hill-streams, which are usually species with strong swimming-powers but not highly specialized, belonging to such genera as Barbus and Barilius. The general analogy between the structure and form of these less specialized members of the fauna of hill-streams becomes in maily instances particular as between fish and tadpoles when the more specialized members, living in more peculiar conditions, are. critically examined. We know of no exact parallel between tlie oral floats cf the larvae of Megalophrys parva and its allies and any similar structure in a mountain fish; but when wecome to the production of adhesive organs many specific parallels occur. Between the enlarged lips and ventral mouth of several fish of the genera Glyptothorax and Glyptosternum and of tadpoles like thase of Bufo penangensis there is a close analogy, and just as we find the oral suckers in different stages of evolution in different species of tadpoles, so also do we find them in different fish of the suborder Siluroidea. Inthe larvae, for example, of Rana assamensts aml its close allies the lips, though ventral and enlarged, are not greatly enlarged and the organ produced is not conspicuous. Almost every stage between this condition and that of Bufo penan- gcensts has been observed. Similarly in such species as Glyptoster- num andersont and G. feae the lips are comparatively small, while in other species of the same genus (e.g. G. labiatum and G. blytht) they are much more highly developed. In both groups the evolution can be correlated with life in waters of stronger and stronger current. It is, however, in the ventral suckers of certain tadpoles of the section Ranae Formosae! of the subgenus Hylovana on the one hand and similar structures in fish of the genera Garra (or Dis- cognathus) of the family Cyprinidae and Glyptothorax of the family Sisoridae on the other that the closest analogy is to be sought, especially, so far as the fish are concerned, in the former genus. We have thought it worth while not only to give a brief general account of the convergence that occurs between these fish and tadpoles but also to consider the minute structure of the adhesive disc of such Ranid tadpoles as Rana afghana (Giinther) [=R. latopalmata, Bligr.] and R. livida (Blyth) and to compare it in detail with that observed in the fishes of the genus Garra, in which the modifications are of a similar nature and occupy a similar position on the ventral surface just behind the mouth. The disc of Rana afghana®* is a well-marked structure; it is almost as broad as the body and a little more than half its length. The disc is provided with a free border except at the anterior end, where the border is replaced by the posterior lip. The comparatively thin central portion of the disc in preserved ‘ Boulenger, Rec. Ind. Mus. XX, pp. 1123s 130 (1920). 2 Annandale, Fec. Ind. Mus. VIL, pl. iv, fig. 3a. 508 Records of the Indian Museum. [Vor. XXIV, specimens is as a rule depressed in the form of a saucer. Through the skin of this region are visible three large promi- nences, which on dissection are found to be the extremeties of two muscles and a tendon. ‘They represent (i) a strong tendon (¢) attached internally to the middle of the disc: it proceeds up- wards for a short distance and then divides into two portions, which are both attached to the vertebral column: and (ii) two pairs of muscles composed of striated fibres which have similar attachments at both ends but are quite distinct from the tendon. By keeping the free border closely in touch with a fixed object and then taising the central portion of the disc by contracting the muscles, the animal can convert the whole structure into an efficient organ of adhesion by creating a partial vacuum between it and the fixed object. The function of the elastic tendon is to couuteract too strong contraction, which might tear the delicate surrounding tissues. 18-94 FIG. {.—Disc and its musculature in Rana afghana. a. Slightly oblique lateral view of disc with its muscles and tendon. hb. Anterior view of muscles and tendon after removal of disc. d.=disc ; ¢.=tendon; mus: 1., and mus, 2.=muscles of disc. (cf. Rec. Ind. Mus. XXIV, p. 47, fig. gb.) The structure of the disc is precisely similar in the tadpole of Rana livida to that in R. afghana. The mechanism of the disc is the same as that already des- cribed for the analogous structure in Garra!; except that in the fish the central portion of the disc is raised by the elevation of the urohyal, without direct muscular action in the disc itself, which is decidedly callous as a whole. The free borders are quite smooth in the tadpoles, but in a section of the tissue under a high power (fig. 2) it is observed that the outer cells are produced into minute processes which are greatly flattened near the base and are somewhat pointed towards the end. ‘These are covered by a chitinized cuticle. Each of the spine-like outgrowths (s) is provided with a nucleus at the base. The rest. of the epidermis (ep. d.) consists of a large number of nuclei irregularly scattered in a homogenous mass of cytoplasm. Below the epidermis is a loose connective tissue in which nuclei are present at irregular intervals, This tissue (¢. ¢.) is formed of a series of minute fibres, which in the outer zo! run paralled to the epidermis, while internally Bey ! Hora, Rec. Ind. Mus. XXIV, p. 47 (1922.) 1922.] N. ANNANDALE & S.J1,. Hora: Parallel Evolution, 509 run at right angles to it. In all essentials the structure described above is similar to that of the free borders of the disc of Garra. The spine-like outgrowths help to make the surface rough in such a way that better grip must be obtained. The minute structure of the disc is thus much less compli- cated than that of the adult Garra or Glyptothorax.' ‘The spine- like outgrowths on the organs of adhesion are strictly analogous in the fish and the tadpoles, but they occupy a different position and the structure of the underlying parts is completely different. One of the most interesting features of these instances of parallel evolution lies in the fact that whereas in the larvae of the Ranae Formosae we only know, so to speak, the finished product of evolution in the highly perfected organ of adhesion, in the ’ / et \ ij We Air Kg i | ee 1 a FIG. 2,—Transverse section through free border of disc of Rana afghana. (Highly magnified and slightly diagrammatic), s.=spine ; ep. d.=epidermis ; c. t.=connective tissue. (cf. Rec. Ind. Mus. ea Pp. 50, fig. 13.) genus Garra we have before us almost every possible stage alike in postembryonic development, in individual variability and in specific differentiation. One* of us has so recently given the facts that it is unnecessary even to recapitulate them here. The evolution of the mental disc of Garra is in this respect parallel to that of oral suckers in various tadpoles of the Himalayan streams. We have thus evidence that these particular structures have come into existence, not through mutation and not by any Mendelian segregation of characters, but through a gradual accu- mulation of small changes. The close correlation, especially in Garra, between these changes and differences in the flow of water in which species and even individuals live is at any rate suggestive. Whether we are witnessing the survival of the fittest in the Darwinian sense or must accept a frankly Lamarckian explanation only Sher ee can prove. ! Cf. Hora, op. ots figs. 12, 13, p. 50; also figs. 15, 17, 18, 19 on pages 53, 55) 57+ 2 /d., ibid., edit pp. 639-643, text-fig. 1 (1921). qttaah tran i SRG hy ue in ) a y vaenter Sut chet da) es ae if ¥ y Mit! SOME ORIENTAL ASCALAPHIDAE IN ‘THE INDIAN MUSEUM, By F. C, Fraser, Major, I1.M.S. With one exception, the whole of the specimens dealt with in this paper are from within the limits of the Indian Empire. Four new species are dealt with, of which one is from Siam and the other three from purely Indian localities. The types of these will be preserved in the Indian Museum. Individual species are difficult to determine from the wide variations met with in colouring both of body and wings. ‘They pass through a teneral stage analogous to that met with in dragonflies and markings may be obscure, well-defined or entirely obliterated according to the age of individual specimens. : Very little is known of the life-histories of this family of insects so that the descriptions and illustrations of three new latvae will be of interest even though it is impossible to say with certainty to which species they belong. Dr. Tillyard writing to me three or four years ago described a method of obtaining the larvae which is best given in his own word;: “There is a very simple trick not known to many, for the finding of these sorts of larvae and that is to go out into the dry bush (as we call it here), and study the large isolated trees, if you have them. I select a tree that is old and worn, with a good lean on it, and with loose rubbly soil around it (termite earth is very good). I then go down on my hands and knees, scoop the soil up in both hands, and let it run slowly through again, forming a mound slowly. Any larvae of Myrmeleontidae or Ascalaphidae hiding in the soil fal! out, and can be seen at once, as they give a kick and begin to burrow again very quickly. I also examine bits of bark, etc., for the Ascalaphidae, which are usually more sluggish and like to hide under bark, debris, etc.’? Dr. Tillyard states that he has secured larvae of nearly every known local genus in this manner around Sydney. I have adopted his methods from time to time but have not met with any success Situations such as he describes are very common in parts of the Deccan and Punjab, but I have not found them to yield fruit although I have copied his instructions to the letter. Dry seasons, he further states, promote the increase of these Neuropterous groups, whilst wet weather nea:ly wipes them out. This does not appear to be the case in India as I have found Ascalaphids more common ia the wet than the dry seasons, al- though their occurrence is scattered pretty well throughout the whole uf the year. The termination of the monsoon is probably the best time to take most species, so that the latter part of the rainy season is at least spent in the senior larval state. On A iS) Records of the Indian Museum. [VoL. XXIV, LARVAE. 1. A single specimen from Talewadi, near Castle Rock, N. Kanara District, Bombay Pres. (coll. S. W. Kemp), almost U “ Text-ric. 1.—Ascalaphid larva from Talewadi, Castle Rock, N. IXanara Dist. certainly the larva of Glyptobasis denti- fera which species was common in the same neighbourhood (Fig. 1.) Head quadrate, deeply fissured in front, coated with very short fine bristles and pigmented darkly save for a pale fascia which begins at the mid-line on a level with the eyes and runs out and backwards. Eyes deeply pigmented and furnished with a small chitinous horn- like process antero-laterallv. Maxillae long and curved inward at a right angle near the tips. They are furnished with short spines of which three are much longer than the rest; of these latter, the two anterior are close together and the middle one longer than the two others, Prothorax very short and rather hid- den, with no definite tubercles on the outer side. The remainder of the body-segment furnished laterally with twelve stout spines all of which are beset, with short, stiff setae. This arma- ture forms an impassable rampart around the insect which serves to protect it from the attacks of ants, j and is closely analagous to that found in various larvac of Euthalia. The presence of these spi.:es serves readily to distinguish the larvae from those of Myrmeleonidae which otherwise are closely similar in form and some- times size. Legs short and slim and entirely hidden beneath the body. With the exception of the last segment, the final seven are deeply pigmented as far inward as the sub- dorsum, including the stout spines. 2. Asingle specimen from Janakh- mukh, Abor country, 600 ft., 29*xii: Igit (coll. S. W. Kemp). Species ? (Fig. 2.) Generally similar to the last in shape but differing as follows:—The maxillae are rather shorter and more robust and deeply pigmented. ‘The Trexv-vic. 2,— Ascalaphid larva from Janakhmukh, Abor country, 1922.] F. C. FRASER: Oriental Ascalaphidae. 513 three spines on their inner side are equal in size and separated by approximately equal intervals. The head is not coated densely with short bristles; it is spotted above with dark brown and bears a fine, lateral streak of the same colour. The frontal notch is very shallow and wide. Eyes without the small, horn-like process. The prothorax is longer and more evident and has on each side a short, stout, blunt tubercle. The lateral spines on the remainder of the body-segment are longer and more slim. The mid-dorsum is marked with a dark pigmented line and each segment bears a couple of fine transverse lines of which the anterior is very sinuous and the posterior more or less straight. The whole body is paler and of a pale brown colour. 3. A single specimen from Rotung, Abor country, alt. 1300 ft., 7 iii rg12 (coll. S. W. Kemp). Found under stones. (Fig. 3.) Somewhat similar to the last but the head differing markedly in shape, much more rounded, the sides concave imme- diately posterior to the eyes. The fron- tal border somewhat crenate and the median notch narrow and shallow. Eyes without the horn-like process. Upper surface of head moderately deeply pig- mented save for a pale band which passes between the eyes and is convex posteriorly. Maxillae long, not so sharp- ly curved at the ends, one with its tip bifid, the three inner spines of equal length, the posterior one wider-spaced than the two anterior. Between them and posterior to the hinder, a short and a long series of smaller spines. Prothorax narrow and well defined, with similar lateral tubercles to the last. Lateral spines of the remainder ofthe | Text-ric. 3.—Ascalaphid body similar to those of the first des- [@tva from Rotung, Abor ctibed. Similar, but thicker and darker reray pigmented transverse lines traversing the dorsum. No imagines were collected on the Abor Expedition so that one cannot give any opinion as to which species the two latter larvae belong to. All the larvae are juveniles, so no measurements have been given. ADULTS. Tribe SUHPALACSINI Ris. Genus Suhpalomitus Ris. Suphalomitus serratus sp. nov. 1 male, Meetaw Forest, W. Rahang, Siam, alt. ca. rooo ft. 4'iv'1913 (C. S. Barton). 514 Records of the Indian Museum. [Vo.. XXIV, Antennae yellowish, club broad, black and spatulate. Head thickly coated with light brown hair, face rather paler, also coated with brownish hair. Eves reddish brown, ahout evenly divided. Thorax moderately broad and short, striped longitudinally and alternately with dark brown and pale ochreous, the mid-dorsum dark brown followed outwardly by a narrow ochreous band and then by a broader, dark brown, humeral band on a level with the wing insertions. laterally pale ochreous, chest pale brown, Legs short and moderately slim, femora yellowish red, tibiae reddish brown, tarsi black, claws reddish. Posterior tibial spurs about as long as the basal joint of tarsus. Abdomen long and slim, uniform:dark reddish brown, the sides of the three basal segments coated with very short, very stiff hairs which stand out perpendicular to the surface. Anal appendages very short, directed down and outwards, with a few long bristles projecting from the ends; genital flaps very short, conjoined, spatulate. Wings hyaline, faintly tinted with brown, reticulation moder- ately close, reddish brown, subcostal field brown; this colour prolonged along all the nervures radiating from the subcosta and radius so as to form a narrow, serrated fascia extending from the base of the wing to the stigma. Pterostigma brown, traversed by 4 to 5 nervures, broad and rather long, its outer border rather oblique. The brown of the stigma is alse prolonged along the costal margin as far as the apex of the wing. Apical field moderately broad, with 3 rows of cells. Axillary angle obtuse and very blunt. Antennae 25 mm.; forewing 39 mm.; hindwing 32 mm.; abdomen 31 mm. Suphalomitus verbosus Wa'k. Ascalaphus ver'osus, Walk., Cat. B.M. Ne-7., no. 36, p. 426 (1853). Ascalaphus prof inus, Walk., l.c., no. 39, p: 428 1853). Helicomius verbosus, MacLach., Fourn. Linn. Soc., dvol., XI, p. 262, no, 4 (1871. Helicomitus profunus, Macl.ach., l.c., no. 5, ». 262 (1871). Suphalomitus verbosus, Ris, Cut. Coll. Selys, Asculuphidae, p. 183 (1908). Type in Brit. Mus, from North India. 2 females, Coorg, 2000 ft., 1913, coll. Hannyngion, and Annarchardi, S. India, 1gog. 2 females, Pashok, Darjiling Dist., alt. 2500 ft., vi'1g16, (coll. L. C. Hartless). A widely distributed species reported from Northern India, Mysore, Rangoon and Ceylon. The present examples do not differ markedly from type, the markings varying according to the age of individuals, Genus Helicomitus MaclI,ach. Helicomitus dicax Walk. Ascalaphus dicax, Walk., l.c., no. 31, ~P. 423 (1853). Ascaluphus sinister, Walk., l.c., no. 32, p. 424 (1353). 1922. | F. C. FRASER: Oriental Ascalaphidae. 515 Ascalaphus immotus, Walk., l.c., no. 33, Pp. 425 (1853). Ascalaphus procax, Walk., l.c., no. 34, Pp. 425 (1853). Ascalaphus odiosus, Walk., l.c., no. 35, p. 426 (1853). Ascalaphus insimulans, Walk., l.c., no. 36, p. 429 (1853). Helicomtins tnsimulans, Walk, and MacLach., Fourn. Linn. Soc., Zool. XI, p. 262 (1871); Weele, Notes Leyden Mus. XXVI, p. 200 (1905); XXVIII, p. 153 (1907) Ascalaphus ceruvinus, Hagen, MacLach., /.c., p. 267 (1871). J Suphalasca cervinus, Hagen, Gerstaecker, Mitt. natur. Ver. Neu.- Vorpomm, und Rugen, 16, p. 88 (1885). Suphalasca placida, Gent l.c., p. 105 ('893); Weele, Motes Leyden Mus. XXVI1, p. 228 (1906) ; /.c., RX VIII, p. 156 (1907). Helicomitus dicax, Ris, Cat. Coll. Selys, Ascalaphidae. p. 172 (1908). Type in Brit. Mus, ; I@,Amarah, Mesopotamia (taken in a tent), x'1916 (coll. Ff. P. Connor); 3a 07, Eden Gardens (at light) and Museum Compound, Calcutta; 29 9, Chowringhee, Calcutta, 27°iv'1y14, 27‘v'Ig2t. I7'vi'tgir, and 28 v1gtr (coll. N. Annandale and F. H. Grauly); 19, Khargpur, Bengal, 17-30°vi'rgtr (coll. R. Hodgart). The specimens exhibit iu a small way the extreme variability of this insect. especially in the markings. The species is widely distributed. I have taken it on several occasions in Mesopotamia above Kerna, that is above flooding areas, and it becomes in- creasingly common towards Amarah and Kut. Extends from Asia Minor to the Philippines and has been reported trom Arabia, India, Ceylon, Java, China and Celebes. Genus Suhpalacsa Lefebr. Suhpalacsa obscura, sp. nov. I@ (rather teneral), Khemsa, 2650 ft., 5‘v' 1913. Head: Antennae pale yellowish brown, the club long and pyriform, finely ringed with blackish brown; about half the length of the forewing. Jaws and face pale yellow; vertex and occiput brown, coated thickly with long brown hair ; eyes dark purplish, iridescent brown. © Thorax dark brown, unicolorous, coated thickly with long dark brown hair, the sides paler. Legs moderately short aud robust, palest brown with a sub-basal spot of black on the tibiae, coated with long whitish hairs. ' Abdomen long and cylindrical, tapering at the end segments, shorter than the hindwing, purplish brown, with obscure apical black annules on each segment. Anal appendages tumid, elliptical, moderately long; genital flaps foliate, notched in the middle. Wings hyaline, reticulation moderately close, white spotted with black, the subcosta especially, which bzars- at its junction with each transverse nervure, a longitudinal black spot, so that it avpears alternately black and white; stigna short, very pale brown, traversed by 4 nervures; apical field broad, 3 rows of cells 516 Records of the Indian Museum. [ Vor. XXIV, in forewing, 2 to 3 in the hind; 7 rows of cells at outlet of the discoidal field. Forewing 37 mm.; hindwing 3r mm, ; abdomen 23 mm. ; antennae Ig mm. I place this species with some doubt in genus Swhpalacsa, which so far has not been shown to contribute any species within Indian limits. It is at least very closely allied to the genus. It bears a superficial resemblance to Idricerus decrepitus ftom which, however, it is of course easily separated by its bipartite eyes. Tribe HyYBRISINI Ris. Genus Acheron Lefebr. Acheron trux Walk. Ascalaphus trux, Walk., Cat. B.M., p. 432, no. 45 (1853). Ascalaphus loqguax, Walk., l.c., no. 48, p. 434, (1853). Ascalaphus antiquus, Walk., l.c., no. 49, P. 434 (1853). Ascalaphus longus, Walk., l.c., no. 50, p. 435 (1853). Helicomitus ctenocerus, Gerstaeck., Mitt. natur. Ver. Neu-Vorpom. und Rugen XXV, p. 101 (1893) ; Weele, Notes Leyden Mus. XXVI, Pp. 200, 228 (1900). Acheron trux, Ris, Cat. Coll. Selys, Ascalaphidae, p. 228 (1908). I@ and 2? 9, Tura, Garo Hills, Assam, 1200-1500 ft., vi-viii, 1917 (coll. S. Kemp); 1@, Pashok, alt. 3500 ft., Darjiling Dist.. E, Himalayas, 26°v'191¥4 (coll. I’. H. Gravely). Type in Brit. Mus. from Bengal, but without precise locality. This species is very constant and has a fairly wide distribution. Localities from which this insect has been recorded are Burma, Sylhet, Darjiling, Bhutan, Bengal, Assam, Sikkim, Malacca, China and Formosa. Genus Glyptobasis MacLach. Glyptobasis dentifera Westwood. Ascalaphus dentifer, West., Cab. Ori. Ent., pl. xxxiv (1848). Ogcogaster dentifer, West., l.c. Ascalaphus dentifer, Walk., Cat. B.M., p. 421, no. 26 (1853). Glyptobasis dentifera, Macl.ach., J.c., p. 268 (1871); Ris. Cat. Coll. Selys, Ascalaphidae, p. 241, fig. 197, 198, 199 (1908). 2e¢o and 1¢, Jalewadi, near Castle Rock, N. Kanara Dist., Bombay Pres., 3-10°x'1916 (coll. S. Kimp); 32 @, Mormugao, Portuguese Ind., ix*1916 (coll. S. Kemp); 12, Balugaon, Puri Dist., Orissa, 21-31'viii'1913 (coll. N. Annandale) ; 1 pair, Param- bikalam, Cochin State, 1700-3200 ft., 16-24'ix:1914 (coll. F. H. Gravely); 2¢7 ¢ and 2? 2, Trichur, Cochin State, 0-300 ft., I-4°x'I914; o, Kavalai, Cochin State, 300-3000 ft., 24—-27°ix"IgI (coll. F. H. Gravely). . This species is the type of the genus. Type in Brit. Mus. (East India). MaclIachlan’s specimen is from Bombay and speci- mevs have also been obtained from Goregaon, which is near Bombay. Bangalore is another locality mentioned for the species. 1922.] F.C. FRASER: Oriental Ascalaphidae. 517 The Deccan and Western Ghats appear to be the districts to which this species is mainly restricted. Glyptobasis nugax Walk. Ascalaphus nugax, Walk., Cat. B.M., p. 433, no. 47 (1853); Hagen, Verh. zoul.-bot. Ges. Wien. VIII, p. 481, no. 66 (1858). Ascalaphus incusans, Walk., l.c., p. 442, no. 63 (1853); Hagen, J.c., p. 481, No. 67 (1858). Glyptobasis incusans, Walk., l.c., p. 442, no. 2, p. 268 (1871). Glyptobasis nugax, Ris, Cat. Coll. Selys, Ascalaphidae, p. 243, figs. 200 and 201 (1908). Type, a 2 from Ceylon, Mus. Griefwald. One pair from Castle Rock, ix'1916; 22 2, Talewadi, near Castle Rock, Kanara Dist., Bombay Pres. (coll. S. Kemp) ; 1@ and 2° 2, Barkul, o—1000 ft., Orissa, I-3'viii'1914, ‘‘ flying in thick jungle, in rain” (coll. F. H. Gravely). This species appears to be restricted to Southern India and Ceylon. With regard to the two females from Orissa I am not altogether sure of the determination and think that they may be vatieties of the species. Females are, however, difficult to deter- mine satisfactorily. Glyptobasis brunnea, sp. nov. 12, foot-hills, Pegu Yomas, Thayetmyo Dist., Burma, Oct. 1git (coll. C. J. Rogers). Antennae dark reddish brown, extending nearly to the level of the stigma, about 4/5ths the wing-length ; club Jong and pyri- form, very dark brown. Labium bright yellow, rest of head and face dark blackish brown, coated thickly with coarse dark brown hair except the occiput which is pale yellow and nearly bald. Eyes brown with a metallic reflex, the upper hemisphere decidedly the larger. Thorax sparsely hairy, dark brown with a broad mid-dorsal stripe of yellow which is encroached upon by an angular process of the ground colour. Jaterally a moderately broad, bright yellow, oblique stripe. Legs dark brown, almost black, robust but short, the anterior tibiae with a cushion of short, thickly-set, yellow hairs on the flexor surfaces. Tibial spine of posterior tibia as long as the basal joint of the tarsus. Abdomen blackish brown. Each of the four basal segments bearing a bright ochreous spot shaped like a leaf with a crenate border, the stalk connected to an apical ring of the same colour. On the other segments this spot is more quadrate or like a leaf without its stalk. Wings long and broad, slightly enfumed throughout, but the apices for about the outer fourth of the wings dark reddish brown. There is also a streak of the same colour immediately posterior to the radius in the forewing. Stigma dark brown, with five ner- vures, very obliquely pointed outwardly ; reticulation close, black ; the appendix very acute and rather long; apical field moderately 518 Records of the Indian Museum. [VoL. XXIV, broad, with 3 rows of cells ; 10 cells at the outlet of the discoidal field. Anal appendages short, tumid, elliptical; genital valves, two rounded, convex, vertical flaps furnished with stiff black hairs. Abdomen 23 mm. ; forewing 33 mm.; hindwing 36 mm. : antennae 38 mm, Genus Siphlocerus MacI,ach. Siphlocerus Minius, Walk. Ascalaphus minius, Walk., l.c., p. 429, no. 4o (1853). Ascalaphus luctifer, Walk., l.c., p. 432, no. 46 (1553). Stphlocerus minizs, Walk., and MacLach., Fourn. Linn. Soc., Zool XI, p. 261, no. 1 (1871); Ris, /.c., p. 246, figs. 202, 203, 204, (1908). 1@, Dharampur, Patiala State, 12 vii'tgtt (Mus. coll.); 1@ and 2° @, Shahzadpur, Allahabad Dist., 29‘viii-rgto0 (Mus. coll.). Not differing from type, which is in Dr. Ris’ collection. Only so far reported from North India and North Bengal. Genus Ogcogaster Westwood. Ogcogaster tessalata West. Ascalaphus tessalatus, West., Cat. Orient. Ent. 34, fig. 1 (1848) ; Walk., lc., p. 420, nv. 24 1853). Ozcogaster tesselatus, West., MacLach., Fourn. Linn. Soc., Zool. XI, p- 265, no. 1 (1871). Ogcogaster tessaluta, Ris, l.c.. p. 253 (1908). 12 ,Kumnon, 6075 fi., W. Himalayas, vii°1914 (coll. Tvéler.) Type in Brit. Mus., a fémal:. No localities h.ve hitherto been recorded for this insect except the broad term ‘‘ India.’’ The single specimen is a very large one, its measurements being: abdomen 15 mm.; forewing 38 mm.; hindwing 32°5 mm.; antennae 25 mim. The discoidal spots of brown are missing but the brown mark- ing in conjunction to the stigma is well defined and forms with the stigma a ye1y consp.cuous arrowhead-shaped marking. Ogcogaster kempi, sp. nov. Two pairs from Talewadi, near Cis le Rock, N. Kanara Dist., Bombay Pres , 10*x"1916 (col! S. Kemp). Male. Antennae b-ight yellow, a little longer than half the length of the wing; club broad, dark brewn. Head bright yellow except the vertex which is black and coated thickly with long datk brown ha‘r; a tuft of bright yellow hair in fiont of and between the antennae; eyes dark reddish brown with an iridescent sheen. Thorax bright yellow marked with blak as follows :—a narrow, dorsal, transverse, black mark on middle ot prothorax, on m.ddor- sum of thorax f-om before back, a b-oal, ‘‘T’’ shaped mark, fol!owed by a marking shaped like an inverted anchor. ‘The arms 1922.] F.C. FRASER: Oriental Ascalaphidae 519 of the “T’’ prolonged out and back as far as the root of the forewing and on each side of its stem with a parallel stripe which runs back and joins the ams of the anchor-mark. Poste- rior to the latter, a transverse black line and then another, smaller “T” shaped mark. Laterally pale yellow marked with two fine, black and rather sinuous lines which pass down from the root of each wing to run between the first and second, and second and third pairs of legs respectively. A third line borders the metepimeron below and runs forward behind the Jast pair of legs. Legs golden yellow, the femora with a ferruginous tinge on the extensor surface; tarsi and extreme distal ends of tibiae black; a longitudinal broad stripe of brown on the distal two-thirds of aera ses A pre KOO GZ 3 Bi RAR ges Hb TEXT-FIG. 4.—Ogcogaster kempi, sp. nov. Q: X 2. tke extensor surface of femora. ‘Tibial spines as long as the first joint of tarsus. Abdomen yellow marked with b'ack. On the dorsum, each segment is outlined in black enclosing a bright yellow spot, and from the lateral black line, on each segment runs an apical and a medial fine, black line, the former passing right under the ventrum to join up with its fellow from the other side and the latter or medial stopping short at the level of the spiracles. Wings similar to those of segmentator, rather rounded at the apex, membrane hyaline, reticulation black except the main nervures which are bright yellow, especially the subcosta and radius. The transverse nervures along the costa of both wings, at the base and over a small area of the wing posterior to the 520 Records of the Indian Museum. [Vou XXIV, 1922.] stigma in the hindwing and a similar area of the same wing along the posterior border opposite the stigma suffused with dark brown ; stigma bright citron yellow, traversed by 5 yellow nervures which bear minute, black spines. Appendages very short, genital flaps projecting horizontally out and forwards. Female very similar to the male, differing as follows :— The eyes are puce coloured; legs black except the knee-joints and extensor surface of tibiae which are yellow, this colour being more extensive on the anterior tibiae than the middle paix and on the middle than the posterior pair. Genital flaps very broad, bluntly triangular, not differing markedly from those of segmentator. Abdomen o@ 16 mm.; forewing o 40 mm.; hindwing o@ 35 mm, ; antennae ¢ 27mm. Abdomen ? 16 mm.; forewing ? 37 mm.; hindwing ° 32 mm.; antennae 9 25 mm. This species is readily distinguished from ¢essalata and kirby: by the bright yellow stigma (black in the latter two species), and from segmentator by the absence of the broad, black, midventral line, by the antennae being yellow, the face quite unmarked and by the very short anal appendages of the male. HIRUDINEA FROM THE INLE LAKE, S. SHAN STATES. By Asajiro OKA, Tokyo. (Text-figs. I-7.) The collection of leeches from the Inlé Lake, S. Shan States, kindly placed in my hands for study by Dr. N. Annandale, is a small one, comprising only three genera and five species. It is nevertheless of an exceptional interest on account of certain structural peculiarities exhibited by some of the new forms, which seem to throw considerable light upon the question concerning the external morphology of the Hirudinea in general. One of the new forms, Glossiphonia inleana, sp. nov., is unique among the Glossiphonidae in having the three annuli constituting a somite easily recognizable at a glance. As is well known, the external annuli of the Hirudinea have, as a rule, an exactly similar appearance, making the determination of somite limits a matter of great difficulty. Until so late as I900, when Castle (5) and Moore (11) almost simultaneously pointed out its inadequacy, an entirely erroneous method of plotting out the somite limits by assuming the sensillae-bearing annulus to be the first annulus of the somite, was in use among students of leeches. Now, in our species, there is no danger of being mistaken in the determination, as the furrows separating the somites are decidedly deeper and more conspicuous than those separating the annuli of the same somite. This is particularly apparent at the margins of the body, where the rings form groups of three each, two fused together and one separate, projecting toward the side in the form of broad and narrow teeth placed alternately. Ancther new species, Glossiphonia annandalet, is also very interesting because of its having four of the six. eyes arranged transversely upon one and the same annulus. This character, though not uncommon among the Herpobdellidae, has, so far as I know, never been observed in any of the remaining families. The five species dealt with in the present paper are here systematically arranged :— Family GLOSSIFHONIDAE. Glossiphonia heteroclita (Linné). Gl. nleana, sp. nov. Gl. annandalei sp. nov. Piacobdella parasitica, juv. (?) 522 : Records of the Indian Museum. { Vor. XXIV, Family HERPOBDELLIDAE. Trocheta quadrioculata, sp. nov. Glossiphonia heteroclita (Linné). Syn.: Hivudo heteroclita, \.inné, 1761. HA. hyalina, O. F Miller, 1774. Hi. trioculata, Carena, 1823. Clepsine hyalina, Moquin- Tandon, 1826. Glossiphonia heteroclita, Moquin- ‘Tandon ,1846, Localities :—Loitan Tank, Yawnghwe Valley. One specimen. Marginal zone, Inlé Lake. Three specimens, from Pachylabra maura (Reeve, Mollusca Gastropoda, family Ampullariidae). These are all small snec mens, measuring only 6 mm. in length and 3-4 mm. in width, Preserved in alcohol, the specimen from Loitan Tank is almost whice; those from the other station are somewhat darker. In both cases there is no indication that the animal was striped or spotted during life. The identification of these specimens is chiefly based upon the disposition of the six eyes, so characteristic of the species. They are arranged, namely, in thre groups of two eyes each, one anterior median and two posterior lateral, in such a way that the animal, when examined superficially, appears to pvssess only three eyes. So far as I could ascertain, the constitution of the abbreviated somites at the anterior and posterior extremities of the body agrees fairly well with the minute description of American speci- mens of Glossiphonia heteroclita given by Castle (6). From Glosstphonia ceylanica Harding (9), some examples of which, according to Kaburaki (1c), appear to be identical with this species, the specimens examined by me could be easily distin- guished by comparing the position of the eyes. In Gl. ceylanica the second and third pairs of eyes are not so close together as in Gl. heteroclita, but are separated by at least one ring. Glossiphonia inleana, sp. nov. Locality :—Fort Stedman, Inlé Lake. Numerous (about 60) specimens, from a tortoise (Cycle nys dhor shanensis Annandale). Shipe and dimensions. As shown in fig. t the form is rather slender for a Glossiphon'a, being from 3 to 4 times as long as wide in moderately contracted specimens. ‘The broadest part lies posterior to the middle, from where the body tapers very gradually toward both ends. Both the dorsal and ventral surfaces are convex, so that the body is somewhat lens-shaped in cross sections. ‘The lateral margins are sharp and serrate. The posterior sucker is of moderate s.ze, circular in outline. Except the shape of the head, which does not projcct toward the side, this species closely resembles Hemiclepsis marginata, with which it also shares the habit of living as an ectoparasite on tortoises. The largest individuals measure about 9 mm, in length and 1922.] A. OKA: Hirudinea from the Inle Lake. 523 nearly 3 mm. in width, the posterior sucker is about 1°5 mim. across. External features. ‘There is a single pair of eyes on the second annulus. They are pretty large and distinct, and are placed close together, almost touching with the base at the median plane of the animal. The opening of the pigment-cup is directed obliquely forward and toward the side. The oral sucket occupies the ventral surface of the first six annuli, the ventral portion of annulus 6, which is very narrow on that side, forming the posterior boundary of the sucker. The mouth-opening is situated in the anterior half of the sucker just below the eyes, corresponding in position with annulus 2. The genital apertures are separated by two rings; the male opening lies between annuli 26 and 27, the female opening between 28 and 29, counting each of those annuli at the anterior end, =: which are double in larger specimens, as two. Both pores are small and inconspicuous. In most of the specimens the clitellum was not distinguishable from the rest of the body. The anus is a small transverse slit placed behind the last annulus. There is, however, a small ring-like portion of the body on the dorsal surface ef the sucker just behind the anus. The hinder sucker is directed ventrally and is attached to the body by the ventral surface of the last three or four annuli. The specimens preserved in alcohol are of a uniform pale grey colour, but there are indications that the animal was orna- mented during life with roundish spots regularly arranged on the dorsal surface, much in the same way as those of Hemiclepsts marginala (2). In many individuals the crop with its paired diverticula is recognizable externally on account of the dark brown mass (coagulated blood of tke host) it contains. Internal anatomy. The small mouth-opening leads into the TEXxT-FIG. 1.—Glossiphonia inleana, sp. nov. a. Outline of entire animal: x 3. 6. Somites i-ix, dorsal view: X 30. c. Somites i-ix, ventral view: x30. 524 Records of the Indian Museum. [VoL. XXIV, usual pharyngeal sac, in which is found the long muscular pro- boscis. ‘This organ is rather slender throughout, being only about twice as thick at the base as at the tip. A pair of groups of unicellular salivary glands open into the posterior end of the pro- boscis. ‘The oesophagus is quite short and leads into the crop, produced laterally into seven pairs of diverticula, of which the last is much larger than the rest and elongated posteriorly. The diverticula show a slight bifurcation laterally, except those of the b. TexT-¥1G. 2.—Glossiphonia tnleana, sp. nov. a. Somites xvi and xvii: X30. 6. Somites xxii-xxvil, with sucker: xX 30. last pair which have three metamerically arranged dilatations directed toward the side. The stomach is provided, as in all other species, with four pairs of simple finger-like caeca. There are six pairs of testes placed alternately with the. crop diverticula, the first pair occupying the space between the first and second pairs of the latter and partly covered by the second pair dorsally. ‘The vas deferens forms a mass of convoluted tubes on either side of the male opening. ‘The ovaries occupy the usual 1922. | A. OKA: Hirudinea from the Inle Lake. 525 position and present nothing particular compared with those of other well-known species. Of the nephridia I counted sixteen pairs, which are situated in the same position as those of Hemiclepsts marginata (12). The openings could not be detected in surface views, so that their posi- tion had to be determined by a study of sections. They lie, as in many other species of the genus, in the furrow separating the first and second annuli of the respective somites. The nervous system is composed of 34 ganglia connected by longitudinal nerve trunks The first six are fused together to form the circum-oesophageal ring. The last seven are likewise coalesced in a single mass at the base of the hinder sucker. The remaining ganglia are separate, arranged regularly one in each somite, except near the anterior and posterior extremities where they are more crowded. Annulation. As stated before, the somite limits are recogniz- able at a glance in this species, as the furrows separating the successive somites are more conspicuous than the interannular furrows. Examined under a low power of the microscope a little out of focus, the former alone are visible, while the latter disappear completely, in such a way that the worm has now the appearance of being uniannulate throughout. I have never seen a leech, except perhaps Myxobdella annandalei Oka (14) from Hongkong, which presented a somewhat similar appearance. Moreover, the interannuiar furrows are not all of the same depth, that separat- ing the first and second annuli of each somite being always less deep than that separating the second and third annuli, so that in many somites the first two annuli appear as a single broad ring, especially at the margins. In most cases the three annuli com- posing a somite have different widths; as shown in the accom- panying figure (fig. 2a) the second ring is the widest, then comes the third which is a little narrower, while the first ring is always the narrowest. There is apparently a certain amount of variation in the con- stitution of somites at either end of the body according to the age of the individual. This will be clear from the following table which shows the annulation of two individuals, 9 mm. and 5 mm. in length respectively. Somites. Number of rings in each somite. A. B. i, ii I I iii, iv Z I v, vi 2 Z Vil, Vili 3 2 ix-xxli 3 3 xxiii 3 Z XXiv 2 2 XXV 2 I XXvi, XXvii . a ae I 526 Records of the Indian Museum. [VoL. XXIV, Thus, the larger individual (A) has 67 rings, while the smaller one (B) has only 61. As the somites at the extremities are more abbreviated in younger individuals than in older ones, it is evident that the elaboration of the somites has progressed centrifugally | from the middle region toward both extremities. This is exactly the opposite of what we should expect, if the triannulate somite represented the primitive condition, from which both biannulate and uniannulate somites were derived by subsequent abbreviation. An examination of the individual somites seems to confirm this view. Somite i has only onering. It bears a faint transverse groove which, however, does not reach to the lateral margins. Somite ii is clearly uniannulate, the pigment cup of the eye occupies almost the entire breadth of this ring. Somites iii and iv are biannulate in large individuals, but uniannulate in smaller ones. Inthe former ~ case somite iii is composed of two rings of practically equal breadth, somite iv, on the contrary, of an anterior broader and a posterior narrower, the ratio of the breadths being about 2to 1. Somites v and vi are always biannulate, being composed of a broad and a narrow ring. Somites vii and viii are triannulate or biannulate according to the size of the individuals, the furrow separating the first and second rings being confined to the median area in smaller individuals. Somites ix to xxii are triannulate. Somite xxiii is triannulate in large individuals but biannulate in small ones; in some cases it is difficult to decide whether the somite should be considered as triannulate or as biannulate. Somite xxiv is biannulate ; somite xxv either bi- or uniannulate; somites xxvi and xxvii uniannulate. The biannulate somites at the posterior end of the body are invariably composed of an anterior broader and a posterior much narrower ring. At the margins all the triannulate somites appear as biannulate, as the rings I and 2 form a single broad tooth separated by a notch from the narrow tooth formed by the ring 3. The chief peculiarity in the external morphology of this leech is, as already stated, that the somite boundaries are recog- nizable at a glance and the three annuli forming a somite are not of the same size. By tracing the somites from the extremities toward the median region, we can observe the various stages through which the primitive uniannulate somite of the ancestral leech gradually became the typical triannulate somite of the Glossi- phonidae. First the somite became broader, then a narrow ring was separated off from the posterior margin. The biannulate somite thus formed next became triannulate by the separation of a still narrower ring from the anterior margin. Afterwards the three rings became equivalent in size, making it extremely difficult to find out where the somite boundaries really are. This species presents a certain resemblance to the Hse matic figure of Placobdella emydae Harding (9), with which it. agrees in the number of crop diverticula, the testes, and the posi- tion of the genital openings. Both forms were also found attached 1922.] A. OKA: Hirudinea from the Inle Lake. 527 to chelonians. In reality, however, the difference is very great, as Pl. emydae is a broad and flat species, measuring 13°5 mm. in length by 9 mm. across. Glossiphonia annandalei, sp. nov. Localities :—Central region of Inlé Lake. Eight specimens. Central region of Inlé Lake. Two specimens, on Tata intha Annandale (Mollusca Gastropoda, family Viviparidae). Shape and dimensions. Body rather long and slender, not unlike Gl. stagnalis, anterior portion having the appearance of a neck. The greatest width, which is about one-third of the length, lies at about one-third of the length from behind. ‘The head is slightly broader than the neck. The body is not much flattened, being convex on both sides, though more so on the dorsal thau on 8 TeExt-F1G. 3.—Glossiphonia annandalet, sp. nov. a. Outline of entire animal: x3. b. Somites i-ix, dorsal view: X 30. c. Somites i-ix, ventral view: X 30. the ventral surface. The margins are sharp and serrate, but not thinned out. The hinder sucker is small and directed ventrally and backward. The largest example, somewhat contracted, measures 6 mm. in length and 2°8 mm. in width. External features. The surface of the body is on the whole smooth, all the papillae being low and insignificant. Neither transverse nor longitudinal rows of particularly large papillae could be observed. The oral sucker occupies the ventral surface of the first five rings, ring 5, which is very narrow on the ventral side, forming its posterior border. The small mouth-opening lies a little in front of the middle of the sucker. There are three pairs of eyes, whose position is quite unique among the Glossiphonidae. Two pairs are situated in the pos- 528 Records of the Indian Museum. [VoL. XXIV, terior half of annulus 4, a pair of small eyes on either side of the median line and a pair of much larger eyes placed about midway between the median eye and the lateral margin. The opening of the pigment-cup of the median eyes is directed forward, that of the lateral eyes forward and slightly toward the side. ‘The re- maining pair is found on annulus 5 just behind the lateral eyes of the preceding annulus. The pigment-cup of this pair is directed backward and a little laterally. The two lateral eyes of either side are placed so closely together, that their pigment cups touch each other with their bottom, presenting the form of X in surface views. Very possibly the eyes which appear to be placed in annulus 5 belong in reality to annulus 6, the pigment cup having been displaced forward. So far as I am aware, there is no Glossi- phonid hitherto known which presents a similar arrangement of eyes. The genital openings are separated by two rings. The male pore lies between annuli 24 and 25, the female pore between 26 and 27. Both openings are small and inconspicuous. The clitellum could not be distinguished in any of the specimens. The nephridial pores were invisible in surface views. By a careful study of longitudinal sections I was enabled to locate them at the usual position, namely, in the furrow separating the first and second annuli of the res- pective somites. In all sixteen pairs of nephri- dia were observed. The anus is situated behind the last annulus on the dorsal surface of TEXxT-FIG. 4.—Glossiphonia annandalei,sp.nov. the sucker. Somites xxii-xxvii with sucker ; X3o. The colour of the specimens preserved in alcohol is a uniform pale gray. There is no indication of the animal having been spotted or mottled during life. Internal anatomy. The proboscis is very long and slender. A short oesophagus connects the base of the proboscis with the crop, which extends over six somites and is provided with as many pairs of diverticula. The latter are all simple except those - of the last pair which extend backward and are provided with three short branches directed toward the side. The stomach bears four pairs of simple tubular diverticula, of which the first two pairs are directed forward, the last pair backward, while the third pair lies about at right angles to the long axis of the body. There are six pairs of testes placed alternately with the crop diverticula, the first pair being in front of the most anterior of the latter. The ovaries are, as usual, a pair of simple sacs ex- tending back along the sides of the ventral nerve chain. 1922. ] A. Oxa: Hirudinea from the Inle Lake. 529 The nervous system agrees most closely with that of Glosst- phonia stagnalis. Annulation. Somites i, ii, and ili are uniannulate. Somite iv is biannulate, consisting of an anterior broad and a posterior narrow ring; it is in the hinder half of the broad ring, i.e. ring 4, that the four anterior eyes are transversely arranged, while the remaining two eyes are imbedded in the interior of the next ring. Somites v and vi are biannulate, with the rings of practically equal breadth. Somites vii-xxiv are triannulate; here the rings are of the same breadth throughout, so that there is no distinction between the inter-somital and interannular furrows. Somites xxv and xxvi are biannulate with the anterior ring about twice as broad as the posterior. Somite xxvii has but one ring, behind which is placed the anus. At the posterior extremity, where the somites are abbreviated, the somite boundaries can be deter- mined without difficulty, as it is always the first and second annuli that are fused. . This species can be easily distinguished from all other species of the genus by the peculiar arrangement of the eyes mentioned above. Placobdella parasitica, juv. (?). Syn.: Hirudo parasitica, Say, 1824. Clepsine parasitica, Diesing, 1850. Cl. plana, Whitman, 1891. Locality :—Canal on W. side of Inlé Lake. One specimen, on Taia shanensis (Kobelt). It is with much doubt that I assign this specimen to the above species. Judging from the size as well as from the condi- tion of the genital pores, it is certainly immature, and it is diffi- cult to ascertain whether the slight but obvious discrepancies existing between this specimen and typical P. parasitica are due to difference in age or to specific distinctness. The specimen is a good deal contracted. The form is oval, much arched dorsally, concave ventrally. The head is curved downward, so that the eyes cannot be seen when the animal is viewed from above. The lateral margins are similarly inflexed. The total length measured along the curved dorsum is 5'5 mm., the transverse diameter 3 mm., the widest part being a little behind the middle of the body. The posterior sucker is circular and measures about I mm. across; its margins are inflexed. There is a single pair of eyes in the anterior half of the third annulus. They are not so close together as in the normal speci- mens of Placobdella parasitica. As I could not study the unique specimen in sections, it was impossible to determine whether the eyes were really simple, as they appeared to be in surface view, and not composed of three eyes crowded together as is the case in that species. The genital pores are separated by two rings. The male opening is situated in the furrow between rings 27 and 28, the 530 Records of the Indian Museum. [VoL. XXIV, female opening between 29 and 30. They are both very small and hidden in the furrow, so that the specimen had to be strongly bent dorsally to make them discernible. The nephridial pores could not be observed. The annulation is practically the same as that given by Castle (6) for Glossiphonia parasitica. Somites i and ii are uni- annulate, somites iii and iv biannulate, somites y—xxiyv trian- nulate, somites xxv and xxvi biannulate, and somite xxvii uniannulate, giving the total of 71 rings. Castle regards somites xxv and xxvi as uniannulate each, which reduces the number of rings to 69, but as these somites were divided, in his speci- mens too, into a broad anterior and a nairow posterior portion at the margins, the difference is more apparent than real. The oral sucker occupies the ventral surface of somites i~iv. The anus lies just behind the last (71st) ring. The surface is on the whole rather smooth; in this respect the specimen comes nearer to var. plana than to var. rugosa. The dorsal surface is covered with numerous papillae, but they are all exceedingly low, and there seems to be no regularity as to their arrangement. The colour is a uniform pale gray. One striking peculiarity in this specimen is that, on the ventral surface, the furrow separating the first and second annuli of each somite is markedly less deep than the others, in consequence of which the body appears, when viewed from this side, to be composed of double and single annuli arranged alter- nately. This is one of the rare instances among the Hirudinae where the somite limits are externally recognizable at a glance. On the dorsal surface, however, all the furrows appear quite alike, rendering it impossible to distinguish the inter-somital from interannular furrows. A similar condition was also noticed by Castle in some of his specimens of Placobdella parasitica, in which the anterior two-thirds of a somite appeared at places like a single broad annulus, but this character seems to have been present in his case on the dorsal as well as on the ventral surface of the body and not confined to the latter as in the case of our specimen, Trocheta quadrioculata, sp. nov. Localities :—Central region of Inlé Lake. ‘‘ Colour blood-red.” One specimen. Central region of Inlé Lake on muddy bottom, 9-12 ft. One specimen, Shape and dimensions. Both specimens are small and seem to be immature. The body is long and slender, almost cylin- drical, being only slightly wider in the middle than near the extremities. The head is rounded in front, forming the anterior lip: of the spacious mouth, ‘The hinder sucker is almost circular, a little broader than long, and is directed ventrally and back- ward. The specimen from g—12 ft. measures 24 mm, in length and 1922. ] A. OKA: Hirudinea from the Inle Lake. 531 I°5 mm. in width; that from the first-named locality is 19 mm. in length and 1-3 mm. in width. The hinder sucker is about as wide as the body in both cases. External features. The mouth is very wide and occupies the ventral surface of the first six rings. It is a spoon-shaped hollow directly continuous with the oesophagus. No jaws nor so-called pseudognaths are visible externally. ; The eyes, in two pairs, are situated on the fourth and seventh rings, rather wide apart. In the individual from the first- named locality the posterior pair presents an anomaly in the fact that the right eye is placed one ring in front of the left eye, i.e. TExt-r1G. 5.—TZvocheta guadrioculata, sp. nov. a. Outline of entire animal: x 3. 6. Somites i-ix, dorsal view: X25. c. Somites i-ix, ventral view: X25. on ring 6. ‘There is no marked difference in size between the anterior and posterior pairs. The genital apertures are very small, almost invisible, except the male opening of the smaller specimen which is rather pro- minent. The male pore is situated close to the posterior bound- ary of somite xi, the female pore just in front of the furrow separating ring 6 and ring 7 of somite xii; they are separated, thus, by a space equivalent to four rings of a five-ringed somite. The clitellum is not developed in either specimen. The colour is pale grayish, a little yellowish anteriorly. There is no indication that the animal possessed any pattern during life. The surface is perfectly smooth all over. As I have not studied the anatomy of the specimens, no 532 Records of the Indian Museum. [Voy. XXIV, comparison can be made with allied forms in regard to the structure of internal organs. However, a minute study of the annulation of various similar-looking leeches from other sources has led me to the conclusion that the specimens under consi- deration can belong to no other genus than Tvocheta Dutrochet, Annulation. ‘The study of the annulation of this leech pre- sented a very great difficulty. Observed in alcohol, the furrows were almost invisible, the integument appearing entirely smooth throughout the whole extent of the body. It was necessary, therefore, to examine the specimens placed on a piece of blot- ting paper and illuminated from the side, when the furrows he- came visible as faint lines on the half-dried surface. TrexT-ric. 6.—Somites xvi and xvii of a. Trocheta quadrioculata, sp. nov. b. Herpobdella atomaria. The annuli are very numerous and of different widths. Ex- cept at the extremities they fall into groups repeated metameri- cally, each consisting of a definite number of broad and narrow tings. In the middle region of the body, we find two broad and seven narrow rings forming such a group, but where the somite limits exactly lie, it is impossible to tell for want of proper land- matks. A thorough investigation of those Herpobdellid genera, whose somites exhibit annuli of unequal widths, such as Tvocheta, Mimobdella, and Odontobdella (as yet unpublished) enables me to state with certainty, that the somite limit falls between the fifth and sixth narrow rings. In other words, a somite in this species typically consists of nine annuli arranged in the following order: two narrow, two broad, and five narrow. Compared with the 1922.] A. OKA: Hirudinea from the Inle Lake. 533 five-ririged somite of Hirvudo or Herpobdella (Nephelis) the first two rings of Tvrocheta correspond with the first ring of these leeches, the two broad rings with the second and third, the two narrow rings that come next with the fourth, and the last three natrow tings with the fifth. As shown in the accompanying figure, the narrow rings are not all of the same width, the seventh and eighth of each somite being always somewhat narrower than their neighbours either in front or behind. These rings have, in all pro- bability, been derived by subsequent division from the anterior half of the last ring of a five-ringed somite and are, conse- quently, each equivalent to one-fourth of the ori- ginal ring, while all the other narrow rings are each equivalent to one half of the original ring. In larger specimens of Trocheta from Japan these relations can be demons- trated so perfectly as to leave no doubt about the Texr-ric. 7.—Tvocheta quadrioculata, sp. nov. matter. The subject will Somites xxiv-xxvii, with sucker: x 25. be discussed more fully in my paper on the Herpobdellidae of Japan to be published shortly. I refrain from giving the exact number of rings in these specimens, as it was impossible to count them in some places. On the whole, the annulation appears to be very similar to that of Japanese species of the same genus. Remarks upon the genus Scaptobdella Blanchard. In my Synopsis of Japanese Leeches (13) I recognized thie ~ genus Scaptobdella Blanchard and described a new species under the name of Scaptobdella blanchardi. Subsequent studies, however, led me to cast doubts upon the validity of this genus, which I now regard as Synonymous with the European genus Tyochela. The reasons for this change will be given in my future paper referred to above. M. P. Gedroyé, in his paper on European leeches (7) expressed his doubts as to the systematic value of the genus Scap- tobdella, and in his Synopsis of Polish Leeches (8) published some years afterwards, he abolished that genus and placed Scaptobdella horsti, the type of the genus, in the genus Trochefa naming it Trocheta horstt, In this regard I am of exactly the same opinion as the Polish author. 534 Records of the Indian Museum. [Vol XXIV, 1922.] It may also be mentioned here that I possess a specimen of a small leech from Formosa, which agrees well with the speci- mens here described, in size, shape, annulation, and above all in the number and arrangement of the eyes. I regard it, therefore, as belonging to the same species. It is interesting to note that a new species of leech with characters different from those of any known form, has been discovered almost simultaneously from two localities so widely separated from each other as Burma and Formosa. LITERATURE. (1) Blanchard, R. Sur la présence de la Trocheta subviridis en Ligurie et description’ de cette Hirudinée. Actes Soc. Ligust. Sc. nat. 3me. Année, 1892. (2) Blanchard, R. Hirudinées de I’Italie continentale et insulaire. Boll. Mus. Zool. Anat. comp. Univ. Torino. Vol. IX. 1894. : (3) Blanchard, R Hirudinées des Indes néerlandaises. Zool. Ergebn. einer Reise in Niederl. Ost-Ind. Bd. IV. 1897. (4) Blanchard, R. WHirudinées du Musée de Leyde. Notes from the Leyden Mus. Vol. XIX. 1897. (5) Castle, W. E. The Metamerism of the Hirudinea. Proc. Amer. Acad. Arts and Sct. Vol. XXXV. 1900. (6) Castle, W. E. Some North American Fresh-water Rhyn- chobdellidae, and their Parasites. Bull. Mus. Comp. Zool. Harv. Univ. . Vol. XXXVI. 1900. (7) Gedroyé, M. Zur Kenntnis der Europaischen Hirudine- enarten. Bull. Acad. des Sci. Cracovie. 1913. (8) Gedroyé, M. Pijawki (Hirudinea) Polski. Tablica Sy- noptyczna. Rozpr. 1 Wiadom. Muz. Im. Diedusz. Tom III. 1918. (9) Harding, W. A. Hirudinea. In: Fauna of the Chilka Lake. Mem. Ind. Mus. Vol. V. 1920. (10) Kaburaki, T. On Some Leeches from the Chilka Lake. In: Fauna of the Chilka Lake. Mem. Ind. Mus. Vol. Vi. 1925, (11) Moore, J. P. A Description of Microbdella biannulata with special Regard to the Constitution of the Leech Somite. Proc. Acad. Nat. Sci. Philad. 1900. (12) Oka, A. Beitrage zur Anatomie der Clepsine. Zeitschr. f. wiss. Zool. Bd. VIII. 1894. (13) Oka, A. Synopsis der japanischen Hirudineen, mit Diag- nosen der neuen Species. Annot. Zool. Japon. Vol. Vit £Or0. (14) Oka, A. Hirudinea. In Zoological Results of a Tour in the Far East by N. Annandale. Mem. Asiat. Soc. Bengal. Vol. VI. 1917. Ee rr DESCRIPTIONS OF SOME INDO-MALAYAN SPECIES OF CAPRITERMES (TERMITIDAE). By F. SI,vEstrli. Among a collection of Termites kindly submitted to me for determination by Dr. N. Annandale, Director of the Zoological Survey of India, I have found four species of Capritermes which I am describing in this note and I have added, with Dr. Annandale’s kind permission, the description of two more species of the same genus which are in my collection from Sumatra. Capritermes gravelyi, sp. nov. (Fig. I.) Miles. Corpus ochroleucum capite ochraceo-ferrugineo, man- dibulis nigris. Caput aliquantum minus quam duplo longius quam latius, lateribus antice paullum convergentibus, fontanella parva glandula sat magna, ab occipite ad parvum tractum supra fontanellam gradatim parum altius et supra fontanellam ipsam convexiusculum, superficie setis nonnullis instructa, labro longo, a media basi ad medium marginem anticum menso, subaeque longo ataue lato, lateribus mediis parum convexis, margine antico profundiore sinuato, atigulis anticis elongatis subtilibus corniformibus, margine externo ad cornuum basi parum producto, breviter et irregulariter inciso, mandibula laeva capitis longitudinem aequante et quam mandibula dextera aliquantum longiore, forma vide fig. I, 1-2, antennis I4-articulatis, articulo tertio quam secundus parum breviore et quartum longitudine aequante. Pronotum quam caput magis quam 1/3 minus latum, lobi antici margine supero paullum sinuato, setis nonnullis sat longis et aliis brevibus et brevioribus instructo, meso- et metanoti margine postico medio subrecto vel vix sinuato, angulis rotundatis. Pedes primi paris coxis antice ad marginem externum setis 6-7 brevibus robustis instructis, tibiae margine interno serie setis robustis 9-10, calcare externo quam interna parum minore. Abdomen setis sat longis, setis brevibus et brevioribus numerosis instructum. Cerci breviores, parte distali subcylin- dracea. Long. corp. mm. 5, long. capitis 2°10, ejusdem lat. 1°30, long. mandibulae laevae 2°10, antennarum 2°50, tibiae III, 1°30. Operarius. Corpus stramineum capite ochroleuco. Caput parum latius quam longius, supra setis parum numer- osis brevibus instructum, fontanella circulari magna albicante, 536 Records of the Indian Museum. [Vor. XXIV, clypeo bene inflato, ejusdem dimidia parte aeque longa atque lata, antennis 14-articulatis, articulo tertio quam secundus parum breviore et quartum longitudine subaequante. Pronotum lobi antici margine supero rotundato, meso- et Fic. |.—Capritermes gvavelyi: 1. militis caput pronum; 2. idem lateral- iter inspectum; 3. labrum; 4. epicranii pars antica lateralis laeva prona: 5. eaclem lateraliter inspecta ; 6, tibia primi paris; 7. tibia secundi paris ; 8. operarii pes primi paris; 9. operarii tibia secundi paris. metanoti margine postico et pedibus eisdem militis similibus.sed pedum primi paris coxis tantum setis externis 3-4. Abdomen setis numerosis quam eaedem militis parum bre- vioribus. Long. corp. mm. 4°2, long. capitis I'ro, ejusdem lat. 1°18, long antennarum 1°60, tibiae III, 1°04. 1922. ] IF. StnvEstri: Indo-Malayan Termittidae. 537 Habitat. Exempla typica ad Kas, Satara Distr., 3700 ft., clar. F. H. Gravely legit, alia (paratypi!) idem Gravely ad Helvak et Vela, Koyna Valley, Satara Distr. ca. 2000-2100 ft. et E. side of Koyna Valley legit. Observatio. Species haec ad C. longivostris Wasm. valde affinis est, sed militis capite ab occipite ad frontis partem anticam gradatim parum altiore (in C. longivostris linea supera fere sub- recta), labro parum magis elongato, operarii tibiis primi paris setis internis robustis magis numerosis armata, bene distincta est, Capritermes longirostris Wasm. var. cornutella, nov. (Fig. IT.) Miles. Corpus cremeum capite ochraceo-ferrugineo, mandi- bulis nigris. Caput ca. 1/3 longius quam latius, lateribus antice paullum convergentibus, supra ab occipite ad parvum tractum ante fonta- nellam gradatim parum altius et supra fontanellam ipsam discen- dens superficie setis parum numerosis instructa, fontanella parva, epicranii margine antico sublaterali ad mandibularum basim in processum brevem conicum producto, labro longo a media basi ad medium marginem anticum menso subaeque longum atque latum, lateribus antice parum latioribus angulis anticis elongatis corniformibus, margine externo ad cornuum basim breviter et irregulariter inciso, mandibula laeva capitis longitudinem aequante et quam dextera parum longiore, forma vide fig. II, 1-2, antennis 14-articulatis, articulo tertio secundum longitudine aequante et quam quartus parum longiore. Pronotum quam caput fere dimidio minus latum, lobi antici margine antico sat profunde et anguste sinuato, setis nonnullis sat longis et aliis brevioribus instructum, meso- et metanotum margine postico medio subrecto, angulis rotundatis. Pedes primi paris coxis setis robustis 6-7 externis, tibia setis brevibus robustis 9-10 internis, calcare externo quam interna fere 2/3 breviore, setis ceteris vide fig. II, 5-6, secundi paris tibiae spinis duabus externis sat longis et sat robustis. Abdomen setis sat longis parum numerosis et setis brevibus et brevioribus magis numerosis instructuin. Cercibreviores, parte distali subcylindracea. Long. corp. mm. 4'6, long. capitis 1°84, ejusdem lat. 1°20, long. mand. Jaevae 1°84, antennarum 2, tibiae III, 1°ro. Operarius. Corpus cremeum capite ochroleuco, Caput parum latius quam longius supra setis brevibus et bre- vioribus parum numerosis instructum, clypeo bene inflato, ejusdem dimidia parte subaeque longa atque lata, antennis 14-articulatis, articulo tertio quam secundus ca. 1/3 breviore et quam quartus parum longiore. Pronotum lobi antici margine medio vix inciso. 538 Records of the-Indian Museum. [Vo1.. XXIV, Pedes primi paris tibia calcare externo perparvo, cetero eisdem militis subsimiles ut fig. II, 7-8, demonstrant. Long. corp. mm 4, long. capitis 0'90, ejusdem lat. 1‘Io, long. antennarum 1°30, tibiae III, o'go. militis caput pronum ; Fic. 11.—Capritermes longirostris var. cornutella: t. 2. idem lateraliter leuk 5 epicranii pars antica lateralis dextera parum oblique inspecta; 4. labrum; 5. tibia primi paris; 6. tibia secundi paris; 7. operarii pes primi paris ; 8. apes tibia secundi paris. Habitat. Milites duos et operarios pauces ad Akalpa, Randal Valley, Ratnagiri distr. collectos vidi. Observatio. Subspecies haec statura minore a forma typica C. longivosivis Wasm. distinctissima est. Capritermes santschii, sp. nov. (Fig. III-IV.) Femina alata. Corpus castaneum, sternorum parte mediana, sterniti sexti parte postica excepta, ochroleuca, alis fumosis. Caput parum longius quam inter oculos latius, supra setis 1922. ] F. S1nveEstri: Indo-Malayan Termitidae. 539 sat numerosis brevibus et setis brevissimis magis numerosis in- structum, fenestra sat magna, ovali, fusca, clypeo parum inflato, ejusdem dimidia parte parum latiore quam longiore, oculis sat magnis et sat prominentibus, ocellis ab oculis quam ocelli diametros transversalis parum minus distantibus, antennis 14-articulatis, articulo tertio quam secundus 1/4 et quam quartus fere duplo longiore. Fic. I1]].—Capritermes santschit, alatus: 1. feminae caput pronum; 2. idem lateraliter inspectum ; 3. thorax pronum ; 4-5. ala interior et ala posterior ; 6. ala anterior anomala; 7. alae particula:multo ampliata; 8. pes primi paris a tibia; 9, tibia secundi paris; 10. cercus. Pronotum quam caput cum oculis ca. 1/8 minus latum, antice brevi tractu sursum vergens, lateribus postice convergentibus angulis rotundatis, margine postico medio vix sinuato; meso- et metanotum lateribus convergentibus, margine postico angulatim profunde inciso, angulis posticis acutis. Alae superficie tuberculis minimis, 6-7 radiatis obsessis et setis sparsis brevibus instructa, venis vide fig. ILI, 4-5. Pedes primi paris tibia interne setis sat numerosis instructa, 540 Records of the Indian Museum. [Vo1L. XXIV, calcare externo quam interna multo breviore, secundi paris tibia spinis apicalibus duabus externis sat robustis. Abdominis tergita setis brevioribus sat numerosis et setis bre- vissimis numerosis sternita setis brevibus, brevioribus et bre- vissimis instructa, pleuris setis brevioribus numerosis vestitis. Long. corp. cum alis mm II, sine alis 6, long. capitis o’gr, ejusdem lat. inter oculos 0°82, diametros longitudinalis oculi 0°32, Fic. 1V.—Capritermes santschit: 1. militis caput pronum; 2. idem lateral- iter inspectum; 3. labrum; 4. tibia primi paris; 5. tibia secundi paris; 6. operarii pes primi paris a tibia; 7. tibia secundi paris. long. antennarum 1°70, alae anticae 9°5, ejusdem lat. 2°5, long. tibiae III, 1°05. Miles. Corpus cremeum capite ochroleuco, mandibulis nigris. Caput duplo longius quam latius lateribus parallelis, dorso usque fere ad fontanellam subplanum antice discendens, setis nonnullis instructum, fontanella perparva glandula cephalica parva, labro aeque longo (processibus anticis exceptis) atque lato, 1922.] F. Srtvestri: Indo-Malayan Termttidae. 545 latere laevo convexiusculo, dextero parum obliquo, angulis anticis processuum instar elongatis et margine ad processuum basim aliquantum producto ut fig. IV, 3, demonstrat, margine antico sinuato, mandibula dextera quam laeva aliquantum longiore et ‘quam caput ca. 1/4 breviore, antennis £4-articulatis, articulo tertio quam secundus paullum et quam quartus ca. 1/3 longiore. Pronotum quam caput ca. 1/3 minus latum, lobi antici margine parum sinuato, setis nonnullis sat longis et aliis brevissimis in- structum; mesonotum postice paullum sinuatum, metanotum postice subrectum. Pedes primi paris coxis setis 4-5 brevioribus, partum robustis et aliis brevissimis, tibia interne setis modice numerosis parum robustis, calcare externo quam interna multo breviore, secundi: paris tibia spinis duabus distalibus externis elongatis. Abdominis tergita et sternita setis brevibus, brevioribus et brevissimis simul sumptis numerosis instructa. Cerci articulo secundo elongato. ; Long. corp. mm. 5, long. capitis 2°08, ejusdem lat. 1-04, long. mandibulae laevae 1°50, antennarum 2, tibiae III, ogo. Operarius. Corpus cremeum, capite ochroleuco, abdomine cibi contenti causa cinereo. Caput paullum latius quam longius supra setis parum numer- osis brevibus et aliis brevioribus et brevissimis instructum, fonta- nella circulari straminea, clypeo bene inflato, ejusdem dimidia parte subaeque longa atque lata, antennis 14-articulatis, articulo tertio quam secundus fere 1/3 breviore et quam quartus patum longiore. Pronotum lobi antici margine vix sinuato. Pedes primi paris. tibia calcare externo perparvo. Abdominis tergita et sternita setis numerosis brevioribus et paucis brevibus vel sat longis instructa. Long. corp. mm. 3°4 long. capitis 0°88, ejusdem lat, 0°96, long. antennarum 1°30, tibiae III, 0°78. ; Habitat. Sumatra: Padang-Pandjang. Exempla nonnulla cl. F. Santschi mihi dedit. rts Observatio. Species haec ad C. nemorosus Hav. proxima est, sed adulti alis longioribus, militis capite parum angustiore, labro processibus parum brevioribus, mandibulis parum brevioribus sat bene distineta est. Capritermes distortus, sp. nov. (Fig. V.) Femina alata. Corpus rufo-castaneum, thoracis parte ven- trale et pedibus fulvescentibus, urosternitis ochroleucis lateraliter rufis, alis fumosis. Caput paullum longius quam inter oculos latius, supra setis sat numerosis sat longis et setis magis numerosis brevioribus instructum, fenestra patva longitudinali quam capitis cetera superficies parum pallidiore, clypeo sat inflato, ejusdem dimidia 542 . Records of the Indian Musewn, [Vou. XXIV, parte paullum longiore quam Jlatiore, oculis parvis bene convexis et prominulis, ocellis ab oculis quam ocelli diametros transver- salis paullum minus distantibus, antennis 15-articulatis, articulo tertio quam secundus parum magis quam dimidium breviore et quam quartus parum breviore. Pronotum quam caput cum oculis ca. 1/6 minus latum, antice brevi tractu sursum vergens, lateribus postice convergentibus, angulis rotundatis, margine postico medio vix sinuato; meso- et metanotum lateribus convergentibus, margine postico angulatim profunde inciso, angulis ipsis plus minusve rotundatis. Alae superficie tuberculis minimis 6-7 radiatis obsessis et setis nonnullis brevibus fere omnibus per venas et per marginem in- structa, venis vide fig. V, 4-5. Fic, V.—Capritermes distortus: 1. feminae caput pronum ; 2. idem lateral- iter inspectum; 3. thorax pronum ; 5, ala anterior et ala posterior; 6. alae particula multo ampliata; 7. pes primi paris a tibia ; 8. cercus; 9. militis caput pronum ; .10, idem ene inspectum ; II. labrum ; 12. tibia primi Paris ; 13. tibia secundi paris; 14. operarii pes primi paris a tibia ; 15. Operarii tibia secundi paris. Pedes primi paris tibia interne setis sat numerosis et sat robustis, calcare externo quam interna paullum breviore, secundi paris tibia spinis externis duabus distalibus sat robustis. Abdominis tergita et sternita setis brevibus sat numerosis et setis brevioribus pernumerosis instructa, pleuris setis brevioribus numerosis. Cerci breviores. Long. corp. cum alis mm. 10, sine alis 6, long. capitis 0°86, ejuscem lat. inter ocules 0°84, diametros longit. oculi 0°28, long. antennarum 1°60, alae anticae 8°5, ejusdem lat. 2°1, long. tibiae TIT, 0°97. Miles. Corpus cremeum capite ochraceo, mandibulis nigris. 1922:] F. Srrvestry: Indo-Malavan Termitidae. 543 Caput magis quam 1/3 longius quam latius lateribus paral- lelis, fontanella parva, ab hac aliquantum discendens, supra setis numerosis brevibus instructum, labro brevi aliquantum latius quam longius, lateribus aliquantum convexis, margine antico sinuato angulis ipsis acutis, setis haud distinctis (in exemplo typico), mandibula dextera quam laeva aliquantum breviore, subacuta, mandibula laeva quam caput 1/3 breviore, forma vide fig. V, g-Io, antennis ? (in exemplo typico articulis a decimo abruptis), articulo tertio quam secundus vix breviore et quam quartus aliquantum longiore. Pronotum quam caput ca. 1/3 minus latum, lobi antici margine supero vix sinuato, supra setis nonnullis sat longis instructum, mesonoti margine postico paullum sinuato, metanoti taptiedato: Pedes setis et calcaribus vide fig. V, 12-13. Abdominis tergita et sternita setis sat longis parum numerosis et setis brevioribus et brevissimis instructa. Cerci breviores gra- datim parum angustiores. Long. corp. mm. 5, long. capitis 1°45, ejusdem lat. 0°98, long. mandibulae laevae 1°15, antennarurn ? ; tibiae III, o'90. © . Operarius. Corpus cremeum capite ochraceo, abdomine cibi contenti causa cinereo. Caput aeque longum atque latum, supra setis sat inumerosis brevibus et nonnullis brevioribus instructum, fenestra parum distincta, clypeo bene inflato, ejusdem dimidia parte parum latiore quam longiore, antennis 14-articulatis, articulo tertio quam secundus aliquantum breviore et quam quartus aliquantum longiore. Pronotum lobi antici margine vix inciso, cetero thorace et abdomine eisdem militis similibus. Pedes setis et calcaribus vide fig. V, 14-15, notandum est primi paris tibiae calcar externum quam interna, fere dimidio brevius. Long. corp. mm. 3'5, lat. capitis 0°78, long. antennarum 1°25, tibiae III, 0°70. Habitat. Exempla nonnulla alata, paucos operarios et mili- tem unum cl. F. H. Gravely ad Kavalai (1300-3000 ft.), Cochin State, 24-27 ix'1g14 legit. Observatio. Species haec ad C. ceylonicus Holmgren affinis, sed statura minore, militis capitis labri forma bene distincta est. Capritermes tetraphilus, sp. nov. (Fig. VI.) Miles. Corpus ochroleucum capite ochraceo ferrugineo, antice ferrugineo testaceo, mandibulis nigris. Caput magis quam 1/3 longius quam latius, lateribus sub- parallelis, supra subplanum, antice gradatim parum descendents, setis nonnullis instructum, fontanella perparva, glandula cephalica parva, labro btevi aliquantum latiore quam Jongiore, lateribus irregulariter convexis, angulis anticis angustioribus aliquantum 544 . Records of the Indian Museum. [Vou, XXIV, productis, margine antico subrecto setis nunnullis longis instructo, mandibula laeva quam dextera aliquantum longiore et quam caput fere 1/3 breviore, forma vide fig. VI, 1-2, antennis 14—arti- culatis, articulo tertio quam secundus paullum breviore et quam quartus parum longiore. Pronotum quam caput aliquantum minus latum, lobo antico parvo margine medio vix sinuato, setis quatuor sat longis et setis nonnullis brevissimis instructum, meso- et metanotum margine postico vix sinuato setis 2+2 sat longis et setis brevissimis instructo. . Pedes primi paris tibia interne setis 9-10 sat robustis, calcare externo quam interna parum breviore, secundi paris tibia spinis externis duabus robustis (in pede nonnullo spina tantum una sistente). Fic. V1.—Capritermes tetraphilus: 1. militis caput pronum; 2. idem later- aliter inspectum ; 3. labrum; 4. tibia Sint paris; 5. tibia. secundi paris; 6. cercus ; 7. Operarii pes primi paris; 8, tibia secundi paris. Abdomen tergitis setis 2+2 sat longis, eae brevioribus 24+2 et setis brevissimis instructis, sternitis anticis setis sat longis paucis, posticis setis sat longis nonnullis, nec non setis brevioribus nonnullis et setis brevissimis numerosis instructis. Cerci breviores apice acuto, setis et sensillis vide fig. VI, 6. Long. corp. mm. 6, long. capitis 2°60, ejusdem lat. 1°50, long. mand. laevae 1°82, antennarum 1°90, tibiae III, 1-30. Operarius. Corpus stramineum capite cremeo, abdomine cibi contenti causa cinereo. Caput paullum latius quam longius, supra setis nonnullis. brevibus et brevioribus instiuctum, clypeo bene inflato ejusdem dimidia parte parum latiore quam longiore, fontanella sat magna straminea, antennis 14-articulatis, articulo tertio quam secundus ca. 1/3 breviore et quam quartus ca. 1/3 longiore. 1922.] F. SrnvEstri: Indo-Malavan Termitidae. 545 Pronotum lobi antici margine supero rotundato, setis nonnul- lis sat longis, brevibus et brevioribus instructum, meso- et metano- tum margine postico vix sinuato. Pedes primi paris coxis Setis robustis 2-3, tibia setis brevibus robustis internis 5-6, calcare externo quam interna aliquantum breviore, secundi paris tibia spinis externis parum robustis inter sese remotis (in pede nonnullo spina tina). Abdomen tergitis et sternitis setis paucis sat longis, setis brevibus parum numerosis et setis brevissimis magis numerosis instructum. Cerci parte distali subconica. Long. corp. mm. 4, long: capitis 1°02, ejusdem lat, ‘08. , long. antennarum 1°52, tibiae III, tr. Habitat. Rangamati, Chittagong Hill Tracts, Bengal (R. Hodgatt). = Observatio. Species haec ad .C. ceylonicus.Holmg. proxima est, sed militis statura majore, capitis parte postica altiore praesertim distincta est. i Capritermes modiglianii, sp. nov. Miles. Corpus ochroleucum capite ochraceo antice latericio conspurcato, mandibulis nigris. Caput minus quam 1/3 longius quam latius lateribus sub- parallelis supra latissime convexum, setis nonnullis instructum, fontanella perparva, glandula cephalica parva, labro brevi subae- que longum atque latum, latere laevo ad basim parum convexo, dextero parum concavo, margine antico subrecto, angulis anticis acutis patum productis, superficie setis nonnullis brevibus in- structa, atitennis I4-articulatis, articulo tertio quam secundus fere 1/3 longiore et quam quartus fere duplo longiore. ~ Prohotum lobi antici margine antico medio paullum sinuato setis nonnullis longis et aliis brevioribus instructum, mesonotum postice paullum convexum, metanotum vix sinuatum. Pedes primi paris coxis seta nonnulla sat robusta elongata, tibiae _margitie interno setis elongatis, calcarilus subsimilibus, secundi paris tibiae spinis duabus externis distalibus elongatis. Abdomen tergitis et sternitis setis nonnullis sat longis et setis brevioribus et brevibts sat numerosis instructis. Cerci breviores apice subconico. Long. corp. mnt. 6, long. capitis 2°10, ejusdem lat. rar Sans: mandibulae laevae I° 62, antennarum 1 80, tibiae III, 1°18. Operarius. Corpus ochroleucum abdomine cibi contenti causa cinereo. Caput paullum latius quam longius supra setis sat numerosis brevibus et brevioribus instructum, fontanella straminea sat magna, clypeo bene inflato ejusdem dimidia parte parum latiore quam longiore, antennis 14-articulatis, articulo tertio secundum longitudine aequante et quam quartus parum longiore. Pronotum lobi-antici margine rotundato cetero thorace et abdomine eisdem militis similibus. 546 Records of the Indian Museum. [Vou. XXIV, 1922.] Pedes primi paris tibiae calcare externo quam interna multo breviore. Long. corp. mm. 3°6, long. capitis 0°95, ejusdem lat. 104), long. antennarum 1°70, tibiae IIT, 1°30. Fic. VII.—Capritermes modiglianii: 1. militis caput pronum; 2. idem lateraliter inspectum ; 3. labrum, a tibia primi paris; 5. tibia secundi paris ; 6. operarii pes primi paris a tibia; 7. tibia secundi paris. Habitat. Militem et operarium tantum vidi ad Doloc-Tolong (Sumatra) a cl. KE. Modigliani, cui species dicata est, collecta. Observatio. Species haec statura minore militis labri forma et setis a C. ceylonicus Holmgr. bene distincta est et a C. bucten- zorgt Holmg. salfem statura minore etiam distincta est. RECORDS of the INDIAN MUSEUM (A JOURNAL OF INDIAN ZOOLOGY) Vol. XXIV. Appendix A. POPPA PDP LIAL PALA P DL DDE ee a . 3 ah se tae Ata r : me em. wontall Ins; ~ ort Gove Liiry, My if if Ken = - *., / > oma f “ rAQD | 4 ¢ f tA. PAM i Hi. yess ‘ a pea | ee | ie ae i 4. a ‘sy UB as yf >», -¢ scr SA ey) ee Dress u Monel ec Wks Ealcutta: PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA 1923 Sxuice Onc Rupee. LIST OF LITERATURE REFERRING TO INDIAN ZOOLOGY (EXCLUDING INSECTA) RECEIVED IN CALCUTTA DURING THE YEAR 1922. Compiled by SunpER Lat Hora, D.Sc., Assistant Superintendent, Zoological Survey of India. ANNANDALE, N.—Ethics of Zoology. Calcutta Review, pp. 1-16 (1922) ANNANDALE, N.—Museums and Taxonomic Zoology. Mus. Journ. XXI, pp. 141-146 (1922). ANNANDALE, N.—Introduction to the study of an Island in the Chilka Lake. Mem. As. Soc. Bengal VII, pp. 257-287 (1922). Ayyanaar, 8. R.—Notes on the Fauna and Fishing Industries of the Lac- cadive Islands. Madras Fish. Bull. XV, pp. 45-69 (1922). Bapavuit, G.—Recherches Zoologique (Paris: 1922). Baitey, J.—Museum Publications (A review of Vol. VIII, part 13, Vol. XXI, part 1, Vol. XXIV, parts 1,2, 3 of Rec. Ind. Mus.) Mus. Journ. London XXII, pp. 128, 129 (1922). Cauttery, M.—-Le Parasitisme et la Symbiose. Lncyclopédie Scienti- Jique (Paris: 1922). Hornet, J.—The Madras Aquarium. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 621-629, one plate (1922). Hornet, J.—The Madras Marine Aquarium (Madras: 1922). Pearson, J.—Marine Biology. Ceylon Administration Report for 1920. SEWELL, R. B. 8. and Annanvatr, N.—The Hydrography and Inverte- brate Fauna of Rambha Bay in an abnormal year. Mem. Ind. Mus. V, pp. 677-710, pls. xxxii-xliii (1922). Suersory, C. D.—Index Animalium. Part 1. 1801-1850 (London : 1922). Wooptanp, W. N. F.—On the Renal Portal System (Renal Venous Meshwork) and Kidney Excretion in Vertebrata. Journ. As. Soc. Bengal (n. s.) XVIII, pp. 85-193 (1922). Protozoa. Buatia, B. L.—Notes on Fresh-water Ciliate Protozoa of India.-II. Journ. Roy. Microse. Soc. pp. 27-37 (1922). CHATTERJEE, G. C.—A new flagellate parasitic in the human intestinal canal. Ind. Journ. Med. Res. X, pp. 523-525, pl. xxiv (1922). Cusuman, J. A—Foraminifera of the Philippines and Adjacent seas. Bull. U. 8S. Nat. Mus. No. 100, vol. 4 (1921). Faust, E. O.—A study of Trichomonas of Guinea-Pig from Peking. Archiv. Protistenkunde Jena XLIV, pp. 115-118, pt. v (1921). il Records of the Indian Museum. [ Vou. XXIV, Kamm, M. W.—A list of the New Gregarines described from 1911 to 1920. Trans. Amer. Microsc. Soc. Menasha (Wisconsin) XLI, pp. 122-152 (1922). Koroip, 0. A. and Swrzy, O.—Mitosis and Fission in the active and en- cysted phases of Giardia enterica (Grassi) of man, with a discussion of the method of origin of Bilateral symmetry in the Polymastigote Flagellates. University California Publications, Zoology XX, pp. 199-234, pls. 23-26 (1922). Porifera. Denpy, A.—Report on the Sigmatotetraxonida collected by H. M. &. “Sealark”’ in the Indian Ocean. Trans. Linn. Soc. London (2) XVIII, pp. 1-164, pl. i-xviii (1922). Coelentrata. Horst, C. J. van der—Madreporaria: Agariciidae. Trans. Linn. Soc. London (2) XVIII, pp. 417-429, pls. xxxi, xxxii. (1922). Horst, C. J. van der—The Madreporaria of the Siboga Expedition. II. Siboga Exped. XVIb, pp. 53-98, pls. i-vi (1921). Horst, C. J. van der—The Madreporaria of the Siboga Expedition. Part III, Eupsammidae. Siboga-Exped. XVIc, pp. 47-75, pls. vii, vili (1922). Jarvis, F..—.—The Hydroids from the Chagos, Seychelles and other islands and from the coasts of British Kast Africa and Zanzibar. Trans. Linn. Soc. London (2) XVIII, pp. 331-360, pls. xxiv-xxvi (1922). LitrscnwaGer, H.—Alcyonarien von den Philippinen. I. Die Gattung Alcyonium Linnaeus. Philippine Journ. Sci. Manila XX, pp. 519- 540, pl. i (1922). Pax, F.—Die Antipatharien der Deutschen Tiefsee Expedition: Wissen. Evrgebn. Deutsch. Tiefsee-Expedition, pp. 1-16, pl. xxiii (1922). Rosxin, Gr. von.—Uber den feineren Bau der Epithel-Muskelzellen von Hydra grisea und fusca. Anat. Anz. LVI, pp. 158-168 (1922). SueRRirrs, W. R.—Evolution within the Genus Dendronephthya (Spon- godes) (Alcyonaria), with descriptions of a number of species. Proc. Zool. Soc. London, pp. 33-77, pls. i-iii (1922). SrePHEeNsoN, T. A.—On the classification of Actiniaria. PartII. Quart. Journ. Microsc. Sci. LXV, pp. 493-576 (1921). Stiasny, G.—Studien iiber Rhizostomeen mit besonderer beriicksich- tigung der Fauna des Malaiischen Archipels nebsteiner revision des system. Cap. Zool. Hague I, pp. 1-179, pls. i-v (1921). Polyzoa. ANNANDALE, N.—Polyzoa in the Colombo Waterworks. Spolia Zey- lanica XII, pp. 207-209 (1922). 1923.] Last of Literature : Appendix. iil Crustacea. Batss, H.—Diagnosen neuer Japanischer Decapoden. Zool. Anz. LIV, pp. 1-6 (1922). Batss, H.—Crustaceca VI : Decapoda Anomura (Paguridea) und Bra- chyura (Dromiacea bis Brachygnatha). In Michaelsen’s Beitr. Kennt. Mereesf. Westafrikas III, pp. 39-67 (1921). Bourne, G. C.—The Raninidae : a study in Carcinology. Journ. Linn. Soc. London XXXV, pp. 25-79, pls. iv-vii (1922). Cuitton, C.—Report on the Amphipoda obtained by the F. I. 8. “Endeavour” in Australian Seas. Fisheries Dep. Australia V, pt. 2, pp. 33-92 (1921). Cnorra, B.—Preliminary note on Isopoda of the family Bopyridae parasitic on Indian Decapoda Macrura. Journ. As. Soc. Bengal (n. s.) XVIII, pp. 69-71 (1922). ' Cotost, G.—I Potamonidi del R. Museo Zoologico di Torino. Boll. Mus. Zool. Anat. Comp. Torino XXXV, pp. 1-39 (1920). Jackson, H. G.—A Revision of the Isopod Genus Jngia (Fabricius). Proc. Zool. Soc. London, part iii, pp. 683-704, pls. i-ii (1922). Kemp, S.—Notes on Crustacea Decapoda in the Indian Museum, XV. Pontoniinae. Rec. Ind. Mus. XXIV, pp. 113-288, pls. iii-ix (1922). Man, J. G. de.—The Decapoda of the Siboga Expedition. Part V. On a collection of Macrurous Decapod Crustacea of the Siboga Expedition, chiefly Penaeidae and Alpheidae. Stboga-Exped. pp. 1-51, pls. i-iv (1922). Parist, B.—Elenco degli Stomatopodi del Museo Civico di Milano. Atti Soc. Ital. Mus. Civ. Stor. Nat. Milano LXI, pp. 91-114 (1922). TATTERSALL, W. M.—Indian Mysidacea. ec? Ind. Mus. XXIV, pp. 445-504 (1922). TATTERSALL, W. M.—The Percy Sladen Trust Expedition to the Albrolhos Islands (Indian Ocean), under the Leadership of Prof. J. Dakin, F.L.S., F.Z.S.—Amphipoda and Isopoda. Journ. Linn. Soc. London XXXV, pp. 1-20, pl. i-iii (1922). TATTERSALL, W. M.—Zoological Results of a Tour in the Far East. Amphipoda with notes on an Additional Species of Isopoda. Mem. As. Soc. Bengal VI, pp. 435-459, pls. xvili-xxi (1922). WeE.LTNER, W.—Cirripedia der Deutschen Tiefsee-Expedition. Wissen. Ergebn. Deutsch. Tiefsee-Expedition, pp. 61-112, pls. ii-iv (1922). Echinodermata. CrarK, A. H.—Sea-lilies and Feather-stars. Smithson. Misc. Coll. LXXII, No. 7, pp. 1-43, pls. 1-16 (1921). CrarK, A. H.—A Monograph of the Existing Crinoids. I. The Co- matulids. Part 2. Bull. U. S. Nat. Mus. LXXXII, pp. 1-795, pls? i-lvii (1921). CiarK, H. L.—The Holothurians of the genus Stichopus. Bull. Mus.- Comp. Zool. Harvard LXV, pp. 39-74, pl. i, 1i (1922). iv Records of the Indian Museum, [ Vou. XXIV, D6épreR.e1n, L.—Die Asteriden der Siboga-Expedition. 1. Porcellanas- teridae, Astropectinidae,‘Benthopectinidae. Siboga-Exped. XLVI, pp. 1-47, pls. i-xii (1921). Fisoer, W. K.—A new Sea-star from Hongkong. Ann. Mag. Nat. Hist. (9) X, pp. 415-418, pl. x (1922). Korner, R.—Echinoderma of the Indian Museum. Part IX. Cly- peastridés. et Cassidulidés (Calcutta: 1922). “ Vermidea.” Anon.—Bibliography of Hookworm Diseases. Publication No. 11. The Rockfeller Foundation International Health Board (New York ; 1922). Bani, Karm Narayan.—On the Development of the “ Enteronephric ” type of Nephridial system found in Indian Harthworms of the genus Pheretima. Quart. Journ. Microsc. Scr. (n. s.) LXVI, pp. 49- 103, pls. v-vii (1922). Bayuis, H. A. and Dausney, R.—Report on the Parasitic Nematodes in the collection of the Zoological Survey of India. Mem. Ind. Mus. VII, pp. 263-347 (1922). Brinkmann, A.—Die Pelagischen Nemertinen (Monographisch Dar- gestellt). Bergens Mus. Skrift. (n. s.) TI, pp. 1-194, pls. i-xvi (1917). aust, E. C.—Notes on the Excretory System in Aspidogaster conchi- cola. Trans. Amer. Mircosc. Soc. Menasha (Wisconsin) XLI, pp. 113-117, pls. xiii, xiv (1922). Fauve, P.—Annélides Polychétes de 1’ Archipel Houtman Albrolhos (Australie Occidentale) recueillies par M. le. Prof. W. J. Dakin, F.LS. Journ. Linn. Soc. London XXXIV, pp. 487-500 (1921). Fiscuer, W.—Gephyreen der Deutschen Tiefsee-Expedition. Wissen. Ergebn. Deutsch. Tiefsee-Expedition, pp. 1-26, pls. i-ii (1922). Giroup, A.—Notes sur le Tube Digestif d’ Ascaris Holoptera (Rudolphi). Archw. Zool. Experiment. Paris LXI, pp. 17-20 (1922). Irnwin-Smitu, V.—Notes on Nematodes of the genus Physaloptera with special reference to those parasitic in Reptiles. Part ii. A review of the Physaloptera of Lizards. Proc. Linn. Soc. N. S. Wales XLVII, pp. 53-62 (1922). MicHaELsen, W.—Oligochaten von westlichen Vorderindien und ihre Beziehungen zur oligochéten fauna von Madagaskar und den Sey- chellen. Mitt. Zool. Staatsinst. u. Zool. Mus. Hamburg XXXVITI, pp. 27-68 (1920). M‘Intosu—Notes from the Gatty Marine Laboratory, St. Andrews.— No. XLIV. 1. On new and rare Polychaeta, Gephyrea, etc., from various Regions. Ann. Mag. Nat. Hist. (9) TX, pp. 1-12 (1922). Oxa, A.—Hirudinea from the Inlé Lake, 8. Shan States. Rec. Ind. Mus. XXIV, pp. 622-534 (1922). * Pocur, F.—Zur Kenntnis der Amphilinidea, Zool, Auz. LIV, pp. 276- 287 (1922), 1923.] List of Literature: Appendia. v Rao, C. R. N.—Some new Species of Earthworms belonging to the Genus Glyphidrilus. Ann. Mag. Nat. Hist. (9) IX, pp. 51-68 (1922). SEwELL, R. B. S.—Cercarae Indicae. Ind. Journ. Med. Research X (suppl.), pp. 1-366 (1922). SOUTHWELL, T.—Cestodes in the collection of the Indian Museum. Ann. Trop. Med. Parasit. XVI, pp. 127-152 (1922). ‘STEPHENSON, J.—On some Earthworms from India and Palestine belong- ing to the British Museum. Ann. Mag. Nat. Hist. (9) IX, pp. 129- 136 (1922). STEPHENSON, J.—A note on some supposed new species of Harthworms of the Genus Glyphidrilus. Ann. Mag. Nat. Hist. (9) 1X, pp. 387-389 (1922). | ‘STEPHENSON, J.—On the Septal and Pharyngeal Glands of the Mirodrili (Oligochaeta). . Trans. Roy. Soc. Edinburgh LIII, pp. 231-264, pl. iii (1922). SrepHenson, J.—Some Earthworms from Kashmir, Bombay, and other parts of India. Rec. Ind. Mus. XXIV, pp. 427-448 (1922). STEPHENSON, J.—Contributions to the Morphology, Classification, and Zoogeography of Indian Oligochaeta. Proc. Zool. Soc. London, pp. 109-148, pl. 1 (1922). Stizves, C. W. and Hassatt, A.—Index-Catalogue of Medical and Veterinary Zoology. Roundworms. Bull. Hygienic Laboratory Washington, No. 114, pp. 1-886 (1920). Mollusca. ANNANDALE, N.—Introduction to the Study of an Island in the Chilka Lake. Mem. As. Soc. Bengal VII, pp. 257-287 (1922). ANNANDALE, N. and Prasuap, B.—Description of a new species of Nudibranch Mollusc. Mem. Ind. Mus. V, pp. 700-702 (1922). Bavay, A.~-Coquilles des cables marins del’ Indo-Pacifique. Journ. Conchyliol. Paris, LXVI, pp. 155-161, pl. vi (1922). ExRMANN, P.»~-Land-und Siiswasser-schnecken aus den siidlichen Schan-Staaten, Hinterindien. Sitzungsber. Naturf. Ges. Leipzig, Jahrgg. 45-48, pp. 1-28, pl. 1 (1922). Gopwin-Austen, H. H.—Land and Freshwater Mollusca of India III, pp. 1-65, pls. CLIX-CLXV (London : 1920). ‘Gopwin-Avsten, H. H.—On a new Alycaeus from the Khasi Hills. Rec. Ind. Mus. XXIV, pp. 365 (1922). Haas, F.—Bemerkungen iiber Asiatische Najaden—Abh. u. Ber. d. Mus. Natur. u. Heimatkunde Magdeburg, III, pp. 287-316, pls: ix-xi (1922). : Haas, F.—Eine nene indische Najade, Trapezoideus prashadi. Sen- ckenbergiana, IV, pp. 101, 102 (1922). “Hepiey, C.—A Revision of the Australian Turridae. Rec. Austral, Mus. XIII, pp. 213-359, pls. xlii-lvi (1922), vi Records of the Indian Museum. [Vot. XXIV. HorneELL, J.—Pearl Formation in the Indian Pearl Oyster. Journ. As. Soc. Bengal (n. s.) XVIII, pp. 213-219 (1922). MarsuaLi, H.—List of Mollusca of Rangoon. Journ. Bombay Nat.. Hist, Soe. XXVIII, pp. 773-776 (1922). PrasnaD, B.—A Revision of the Burmese Unionidae. Rec. Ind. Mus. XXIV, pp. 91-111, pl. ii (1922). PRasHaD, B.—Observations on the Invertebrate Fauna of the Kumaon Lakes. III. The Freshwater Mollusca. Rec. Ind. Mus. XXIV, pp. 11-17 (1922). Sturoru, H.—Uber einige Nacktschnecken vom Malayischen Archipel von Lombok an ostwarts bis zu den Gesellschafts-Inseln. Abh.. Senckenb. Natur}. Ges. XX XV, pp. 261-306, pls. xviii-xx (1920). SterK1, V.—Some Notes on the Hinge of the Sphaeridae. Nautilus XXXV, pp. 104-117 (1922). Arachnida and Myriapoda. Borewiii, A.—Di alcuni Dermatteri raccolti in Malesia dal Prof. C. F.. Baker. Boll. Mus. Zool. Anat. Comp. Torino XXXV, pp. 1-9 (1921). Brotemann, H. W.—Myriapods collected in Mesopotamia and N. W.. Persia by W. Edgar Evans, B.Sc., Late Capt. R.A.M.C. Proc. Roy. Soc. Edinburgh XLII, pp. 54-74 (1922). Ewine, H. E.—The Phylogeny of the Gall Mites and a New Classifica- tion of the Suborder Prostigmata of the Order Acarina. Ann. Ent. Soc. America XV, pp. 213-222 (1922). Hrymons, R. von.—Beitrag zur Systematik und Morphologie der Zungenwiirmer (Pentastomida). Zool. Anz. LV, pp. 154-167 (1922). Hinesron, R. W. G.—The Snare of the Giant Wood Spider (Nephila. maculata). Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 642-649: (1922). | Hirst, §.—Mites injurious to domestic animals (with an appendix on the Acarine Disease of Hive Bees). British Museum (Natural History) Econonuc Series No. 13, pp. 1-107 (London : 1922). ManavaLaRAMANuJAM, S. G.—Occurrence of the Galeod Spider (Rha-- godes nigrocintus) in the South Arcot District, Madras Bevsiderty Journ. Bombay Nat. Hist. Soc. XXVIII, p. 814 (1922). Reimoser, E.—Katalog der Echten Spinnen (Araneae) des Palaark- tischen Gebietes. Abh. zool.- bot. Ges. Wien X, pp. 1-280 (1919). : Remy, P.—Sur le Rejet de Sang par les Argasidae. Archiv. Zool.. Experiment. Paris LXI, pp. 1-16 (1922). Savory, T. H.—The Spider Liphistius: a Study in the Biology of a: Primitive Animal. Ann. Mag. Nat. Hist. (9) X, pp. 444-449 (1922). Watersron, J.—The Louse as a menace to Man. British Museum. (Natural History) Economic Series No. 2, pp. 1-20 (London : 1921). 1923.] List of Literature : Appendia. shill Pisces. ANNANDALE, N. and Hora, S. L. —Parallel Evolution in the Fish and Tadpoles of Mountain Torrents. Rec, Ind. Mus. XXIV, pp. 505- 509 (1922). Arcurty, G.—A New Species of Shark. Trans. New Said Inst. LIII, pp. 195, 196 (1921). EIssELE, L.—Histologische Studien an der piesetins einiger Siisswasserfische. Biol. Zentralblatt XLII, pp. 125-138 (1922). Guosu, A. C. and Guosn, 8. N.—Notes on carp-breeding and culture . in confined waters of Bengal and Bihar. Department of Fisheries, Bengal. Bull. XVIII, pp. 1-9 (1922). Hora, 8. L.—Structural Modifications in the Fish of Mountain Hee Rec. Ind. Mus. XXIV, pp. 31-61 (1922). Hora, 8. L.—The Modification of the Swim-bladder in Hill-stream Fishes. Journ. As. Soc. Bengal (n.s.) XVIII, pp. 5-7 (1922). Hora, 8. L.—The Homology of the Weberian Ossicles. Journ. As. Soc. Bengal (n.s.) XVIII, pp. 1-4 (1922). Hora, 8. L.—Notes on Fishes in the Indian Museum. III. On Fishes belonging to the family Cobitidae from high altitudes in Central Asia. Rec. Ind. Mus. XXIV, pp. 63-83 (1922). Hora, 8. L.—Notes on Fishes in the Indian Museum, IV. On Fishes belonging to the genus Botia (Cobitidae). Rec. Ind. Mus. XXIV, pp. 313-321 (1922). Horn, A.—Der Schwimmblasenapparat bei Cobitis. Biol. Zentral- blatt XLII, pp. 118-125 (1922). 7 Manrrep!, P.—Descrizione di una carpa mostruosa. Natura XIII, pp. 20-28 (1922). Maxwett, C. N.—Malayan Fishes. Journ. Straits Branch Roy. Asiat. Soc. No. 84, pp. 179-280, pls. i-lxxii (1921). Mius, J. P—The Lhota Nagas. Fishing, pp. 70-74, pl. (London: 1922). Nicuots, J. T.—Carangoides Jordani from the Hawaiian Islands with notes on related species. Amer. Mus. Novitates No. 50, pp. 1-3 (1922). Nussavum - Hinarowicz, J.—Etudes d’anatomie comparée sur les Pois- sons provenant des campagnes scientifiques de 8. A. S. le Prince de Monaco. Res. Camp. Sci. Monaco LVIII, pp. 1-115, pls. i-xii (1920). Osuima, M.—A Review of the Fishes of the Family Mugilidae found in the Waters of Formosa. Ann. Carnegie Mus. XIII, pp. 240- 259, pls. xi-xiii (1922). ° Osuima, M.—A Review of the Fishes of the Family Centriscidae found in the Waters of Formosa. Ann. Carnegie Mus. XIII, pp. 260- _ 264, pl. xiii, figs. 2-3 (1922). Reaan, C. T.—Fishes of the Clupeid Genera Clupeoides and Potamalosa, and allied Genera. Ann. Mag. Nat. Hist. (9) X, pp. 587-590 (1922). vii | Records of the Indian Museum. [ Vou. XXIV, Waser, Max and Beavrort, L. F. Dr.—The Fishes of the Indo-Austra- lan Archipelago. IV, pp. 1-410, text-figs. (Leiden: 1922). . Waitenouse, R. H.—The Grey Mullets of Tuticorin. Madras Fish: Bull. XV, pp. 71-98 (1922). Batrachia and Reptilia. ANNANDALE, N. and Hora, 8. L.—Parallel Evolution in the Fish and Tadpoles of Mountain Torrents. Rec. Ind. Mus. XXIV, pp. 505- 509 (1922). Bargsour, T.—Sphaerodactylus. Mem. Mus. Comp. Zool. Harvard XLVII, pp. 217-278, pls. 1-26 (1921). , Buattacnarya, D. R.—Notes on the Venous System of Varanus bengalensis. Journ. As. Soc. Bengal (n.s.) XVII, pp. 257-261, pl. v (1922). Boutencer, G. A—Monograph of the Lacertidae II, pp. 1-449 (London : 1921). DevaNnEsEN, D. W.—Notes on the Anatomy of Cacopus systoma, an Indian Toad of the Family Engystomatidae. Proc. Zool. Soc. Lon- don, pp. 527-556, part ili (1922). Frere, A. G.—An aggressive Phoorsa (Echis carinata). Journ, Bombay Nat. Hist. Soc. XXVIII, pp. 291, 292 (1921). Hora, 8S. L.—Some observations on the Oral Apparatus of the Tadpoles of Megalophrys parva Boulenger. Journ. As. Soc. Bengal (u.s.) XVIII, pp. 9-15 (1922), JEUDE, TH. W. vAN LiptH DE.—Snakes from Sumatra. Zool. Meded. Leiden VI, pp. 239-253 (1922). Moutton, J. C.—The reported occurrence of Russell’s Viper in Sumatra and. the Malay Peninsula. Journ. Straits Branch Roy. As. Soc. No. 85, pp. 206, 207 (1922). Nosie, G. K.—Snakes That Inflate. Natural History XXI, pp. 166- 171 (1921). sient Nosiz, G. K.—The Phylogeny of the Salientia 1—The Osteology and the Thigh Musculature ; Their Bearing on Classification and Phylo- geny. Bull. American Mus. Nat. Hist. XLVI, pp. 1-87, pls. i-xxiii (1922). Procter, J. B.—-On a new Toad, Cophophryne alticola, collected on the Mt. Everest Reconnaissance Expedition, 1921. Ann. Mag. Nat. Hist. (9) IX, pp. 583-587 (1922). Procter, J, B.—Further Lizards and Snakes from Persia and Meso- potamia. Journ. Bombay Nat. Hist. Soc. XXVIII, No. 1, pp. 251-253 (1921). Rao, C. R. N.—Notes on Batrachia. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 439-447 (1922). Roos, N, p—.—Fauna Simalurensis. Reptilia. Zool. Meded. Leiden VI, pp. 217-238 (1922). 1923. ] List of Literature : Appendia. ix Smita, M. A.—Notes on Reptiles and Batrachians from Siam and Indo- China. No. 1. Journ. Nat. Hist. Soc. Siam IV, pp. 203-214, pl. viii (1922). Smita, M. A.—The Frogs Allied to Rana Doriae. Journ. Nat. Hist. Soc. Siam IV, pp. 215-225, pl. IX (1922). Smita, M. A.—The Frogs Allied to Rana Doriae. Addendum. Journ. Nat. Hist. Soc. Siam IV, pp. 227-229 (1922). Smita, M. A.—On a collection of Reptiles and Batrachians from the mountains of Pahang, Malay Peninsula. Journ. Fed. Malay States . Mus. X, pp. 263-282 (1922). Watt, F.—A Review of the Indian Species of Amblycephalus. Rec. Ind. Mus. XXIV, pp. 19-27 (1922). Watt, F.—A new Snake from the Northern Frontier of Assam. Rec.. Ind. Mus. XXIV, pp. 29, 30 (1922). Watt, F.—A New Snake of the family Uropeltidae. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 41, 42, one pl. (1921). Watt, F.—Notes on Some Notable additions to the Bombay Natural. History Society’s Snake collection. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 43, 44, one pl. (1921). Watt, F.—Notes on some Lizards, Frogs and Human Beings in the Nilgiri Hills. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 493-499: (1922). | Aves. Avoxrut, H—Uber den Brustkorb und die Wirbelsiule der Vogel. Zeitschr. Ges. Anat. Miinchen u. Berlin LXV, pp. 1-149 (1922). Baker, E. C. 8.—-The Game-Birds of India, Burma and Ceylon II, pp. 1-328, pls. i-xix (London: 1921). ' Baker, HE. C. 8.—A note on some Oriental Zosteropidae, and descrip- tions of new Subspecies. Jbis (II) IV, pp. 142-147 (1922). Baker, E. C. 8.—The Game Birds of India, Burma and Ceylon. Part XXX (Genus Arboricola). Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 1-22, one pl. (1921). ° BaxeEr, E. C. S.—Birds of the Indian Empire. Part IV. (Hand-list of the “Birds of India”). Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 85-106 (1921). Baker, E. C. S.—-The Game Birds of India, Burma and Ceylon. Part. XXXI. Journ. Bombay Nat. Hist. Soc. XXVIII. pp. 305-313, plate (1922). Baker, E. C. 8.—Birds of the Indian Empire, Part V. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 313-334 (1922). Baker, E. C. 8.—The Game Birds of India, Burma and Ceylon. Part, XXXII (Genus Perdiz). Journ. Bombay Nat. Hist. Soc. XXVIII pp. 571-575, one plate (1922). i Barer, E. C. 8.—Birds of the Indian Empire. Part VI. Journ. Bom- bay Nat. Hist. Soc. XXVIII, pp. 576-594 (1922). x Records of the I ndian Museum. [ Vor. XXTYV. Baker, H. R.—Occurrence of the Malay Bittern (Gorsachius melanolo- phus) at Ootacamund, 8. India. Journ. Bombay Nat. Hist. Soc. XXVIII, p. 547 (1922). Basit-Epwarpes, 8.—Behaviour of the White-cheeked Bulbul (Mol- pastes bucogenys) when its young is in danger ; and the (?) parental instinct of love for the offspring displayed by the dark grey Bush- Chat (Oveicola ferrea). Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 280, 281 (1921). ‘Botster, R. C.—The Occurrence, Habits and Breeding of the spotted Sandgrouse (Pteroclurus senegallus) in the Bahawalpur State, Pun- jab. Journ. Bombay Nat. Hist. Soc, XXVIII, pp. 807-809 (1922). Fieip, F.—Rough list and notes on the Birds found breeding in the Gonda District, Oudh. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 753-772 (1922). Finn, F.—The Water Fowl of India and Asia. Review in Journ. Bombay Nat. Hist. Soc. XXVIII, p. 254 (1921). Fietcuer, T, B. and Inaiis, C. Mi—Some common Indian Birds, Nos. 13-18. Agric. Journ. India XVII, pp. 3-6, 113-118, 221-224, 343- 346, 441-444, 543-545, pls. (1922). Frere, A. G.—Roosting Habits of the Common Babbler (Argya caudata). Journ. Bombay Nat. Hist, Soc. XXVIII, p. 280 (1921). Hieeins, J. C.—Manipur names of certain Birds. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 288-290 (1921). Kintocu, A. P.—The Avifauna of the Nelliampathy Hills. Journ. ’ Bombay Nat. Hist. Soc. XXVIII, pp. 279, 280 (1921). Kinnear, N. B.—On the Birds collected by Mr. A. F. R. Wollaston: during the first Mt. Everest Expedition. Zbis (11) IV, pp. 495- 525, pl. vii (1922). Law, 8S. C.—An Albinoid Otocompsa emeria. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 281, 282 (1921). Macrara, H. A. F.—Kashmir Bird Notes. Journ. Bombay Nat. Hist.’ Soc. XXVIII, pp. 276, 277 (1921). Osmaston, A. E.—Note on the Nidification and habits of some Birds in British Garhwal. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 140-160, pls. (1921). Osmaston, B. B.—The White-spotted Fantail Flycatcher (Rhipidura pectcvalis). Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 282, 283 (1921). Osmaston, B. B.—The Crested Swift (Macropteryx coronata). Journ. Bombay Nat. Hist, Soc. XXVIII, pp. 283, 284 (1921). Osmaston, B. B.—Occurrence of the Flamingo (P. roseuws) in the Central Provinces. Journ. Bombay Nat. Hist. Soc. XXVIII, p. 549 (1922). ‘Osmaston, B. B.—Birds of Pachmarhi. Journ. Bombay Nat. Hist. Soc., XXVIII, pp. 453-459, pls. i, ii (1922). Osmaston, B. B.—The Occurrence of the Blue-bearded Bee-eater (Nyc- tiornis athertont) in the C. P. Journ. Bombay Nat. Hist. Soc. XXVIII, p. 805 (1922). 1923.] List of Literature: Appendia. xi Rozrnson, H. CG. and Kross, C. B—The Birds of South-West and Peninsular Siam. Journ. Nat. Hist. Soc. Siam V, pp. 1-87 (1921). Rosinson, H. 0. and Kioss, 0. B.— Birds from the One Fathom Bank Lighthouse, Straits of Malacca. Journ. Fed. Malay States Mus. X, pp. 253-255 (1922). RoBInson, H.C. and Kuoss, C. B.—A list of Birds collected on Pala Rumpia, Sembilan Islands. Journ. Fed. Mala, y States Mus. X, pp. 255-259 (1922). Roszrnson, H. OC. and Kioss, C. B.—List of Birds collected on Pulau Jarak, Straits of Malacca. Journ. Fed. Malay States Mus. X, pp. 259-260 (1922). Rosrinson, H. C. and Kross, C. B.—Three New Oriental Birds. Journ. Fed. Malay States Mus. X, pp. 261, 262 (1922). Scuaupure, R. 8. van.—On a collection of Birds from Acheen ey (Sum tra). Ibis (11), IV, pp. 662-675 (1922). Stocktey, C. H.—Notes on Duck in the Rawalpindi District. Journ, Bombay Nat. Hist. Soc. XXVIII, p. 548 (1922). STRESEMANN, E. von—Hinige Bemerkungen Zur “ Synopsis of the Acci- pitres”? von H. Kirke Swann. Die indoaustralischen Tagraubvogel. Journ. Ornithologie LXX, pp. 487, 488 (1922). Swann, H. K.—A Synopsis of the Accipitres (Diurnal Birds of Prey) Parts II, III, IV (London : 1922). Ticrnurst, C. B.—Notes on some Indian Wheateaters. Jbis (11) IV, pp. 151-159 (1922). Trornurst, C. B.—The Birds of Sind (Parts i and ii). Jbzs (11) IV, pp. 526-571, 605-662, pl. viii (1922). Ticruurst, C. B.—The Birds of Mesopotamia. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 210-250, pls. i, 11 (1921). TickuursT, C. B.—The Birds of Mesopotamia. Part II. Journ. Bom- bay Nat. Hist. Soc. XXVIII, pp. 381-427, pls. 3, 4 (1922). TickHuRsT, 0. B. and Buxton, P. A. and Currsman, R. E.—The Birds of Mesopotamia. Part III. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 650-674, two plates (1922). Warr, W. E.—The Passerine Birds of Ceylon. Spolia Zeylanica XII, pp. 22-194, pls. i, ii (1922). Waite, H. W.—The Marbled Duck (Marmaronetta angustirostris) in the Punjab. Journ. Bombay Nat. Hist. Soc. XXVIII, p. 807 (1922). WuistLeR, H.—The Birds of Jhang District, 8. W. Punjab. Part I.— Passerine Birds. Ibis (11) IV, pp. 259-309 (1922). WauistitER, H.—The Birds of Jhang District, 8. W. Punjab. Part II. —Non-Passerine Birds. Ibis (11) IV, pp. 401-436 (1922). WnistiLrerR, H.—An Addition to the list of Indian Birds. Journ, Bombay Nat. Hist, Soc. XXVIII, pp. 544, 545 (1922), xii Records of the Indian Museum. — [ Vou. XXIV, Mammalia. CHATTERJEE, N. C.—The Indian Pangolin (Manis pentadactyla). Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 273, 274 (1921). Das, B. K.—On Truncated Umbilical Arteries in Some Indian Mam- mals. Journ. Anat. LVI, pp. 325-332 (1922). Fiscuer, C. EH. C.—The Habits of the Grey Mongoose. Journ. Bom- bay Nat. Hist. Soc. XXVIII, p. 274 (1921).. Hutton, J. H.—The occurrence of the Stripebacked Weasel (Mustela strigidorsa) in the Naga Hills. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 795, 796 (1922). LONNBERG, E.—Some Remarks about Hastern Hedgehogs. Ann. Mag. Nat. Hist. (9), TX, pp. 620-629 (1922). Mitner, C. E.—Wild Dogs in Burma. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 266, 267 (1921). Mitner, O, E.—Distribution of Serow in Burma. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 267, 268 (1921). Puituies, W. W. A.—Notes on the Habits of Some Ceylon Bats. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 448-452 (1922). Pirman, C. R. 8.—Notes on Mesopotamian Mammals. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 474-480 (1922). Pocock, R. I.—The External Characters of Scarturus and other Jerb- oas, compared with those of Zapus and Pedetes. Proc. Zool. Soc. London, part ii, pp. 659-682 (1922). Powar, G. B.--A Good female Chinkara Head (G. Bennetti). Journ. Bombay Nat. Hist. Soc. XXVIII, p. 266 (1921). Rosinson, H. C. and Ktoss, C. B.—New Mammals from French Indo- China and Siam. Ann. Mag. Nat. Hist. (9) IX, pp. 87-99 (1922). Rosinson, H. C. and Moutron, J. C_—The Bearded pig (Sus barbatus) in the Malay Peninsula. Journ. Straits Branch Roy. As. Soc., No. 85, pp. 202-205 (1922). SonntaG, C. F.---The Comparative Anatomy of the Tongues of the Mammalia. VII. Cetacea, Sirenia, and Ungulata. Proc. Zool. Soc. London, part iii, pp. 639-658 (1922). Srock.ey, 0. H.—Some notes on the Horns of the Thamin (Cervus eldi). Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 268-271, plate (1921). Suter, M.—Occurrence of the Tree Shrew (Anathana wroughtont) at Khandalla, Poona District. Journ. Bombay Nat. Hist. Soc. XXVIII, p. 537 (1922). ? Tuomas, O.—On some interesting Hedgehogs from the Persian Gulf. Ann. Mag. Nat. Hist. (9) 1X, pp. 142-144 (1922). Tomas, O.—On some new Forms of Ochotona. Ann. Mag. Nat. Hist. (9) IX, pp. 187-193 (1922). . Tuomas, O.—Some Notes on Ferret-Badgers. Ann. Mag. Nat. Hist. (9) IX, pp. 193-196 (1922). Taomas, O.—The Generic Classification of the Taphozous Group. Ann, Mag. Nat. Hist. (9) IX, pp. 266, 267 (1922). 1923.] List of Literature : Appendia. xili Tuomas, O.—On Mammals from the Yunnan Highlands collected by Mr. George Forrest and presented to the British Museum by Col. Stephenson R. Clarke, D.8.0. Ann. Mag. Nat. Hist. (9) X, pp. 391-406 (1922). Tuomas, O.—Scientific results from the Mammal Survey, Nos. XXX and XXXI. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 26, 27 (1921). Tuomas, O.—A New Arabian Hare. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 28, 29 (1921). Tuomas, O. and Hinton, M. A. C.—The Mammals of the 1921 Mount Everest Expedition. Ann. Mag. Nat. Hist. (9) IX, pp. 178-186 (1922). Warp, A. E.—Game Animals of Kashmir and Adjacent Hill Provinces. Part III. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 595-609, one plate (1922). Winer, H.—A review on the interrelationships of the Cetacea. Smith- son. Mise. Coll., LXXII No. 8, pp. 1-97 (1921). (Translated by G. 8. Miller, Junior.) Wroveuton, R. C.—Scientific Results from the Mammal Survey. No. XXIX. Journ. Bombay Nat. Hist. Soc. XXVIII, pp. 23-26 (1921). CALCUTTA SUPERINTENDENT GOVERNMENT PRINTING, INDIA 8, HASTINGS STREET This preservation photocopy was made and hand bound at BookLab, Inc. in compliance with copyright law. The paper, Weyerhaeuser Cougar Opaque Natural, meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Austin 1994 Ay 4 ee] e Mh ) lh i i i iii INSTITUTIO! 3 9088 01261 | —_— — ——— S=