nie ep hadtetathetota! gihqintalalabet eta neatnl mipeieliwtio avohatetabotopode tate Pp v shee + oe D vi . v" aight Pairen a! hahaa homameionhpRatatlint sts arletre- atetiaiunbeyttasigtoihs>ehb -ihen git thr \miuaLaage ap ibapatiphade nate heietiobeim se) «uel . Dosahol piteipephatittehetulshihentboteysinttatatintoteisiarasetatatetesegetet haat bets te se eee eee ipicinlominentsncheoat ehetel ore eteteetenaieteh sh thulbele oteia inane tehedndettthinishelRedehetelletabaleheieiwicietanr «= )ehabe ateratsiehsvotsbatenshotatalintiots ish vlehe: ateinhnheiahl Nat senalaargra tydafbatigtnie enaiieretiahet aie oe) Li. oboaked a ~ a wires rs aohegeesuat wv bss heh or A where ie on oF Uy ne ENN om ipl Bie ® 9 heme ie 96 ol earths Nat ei yh RECORDS of the INDIAN MUSEUM (A JOURNAL OF INDIAN ZOOLOGY) Vol. XVIII, 1919-1921. EDITED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA. Calcutta : PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA. PRINTED AT THE BAPTIST MISSION PRESS. 1921. REPORT ON THE AQUATIC FAUNA OF SEISTAN WITH SUBSIDIARY STUDIES. baat viet a mate a ie y tN pti, ae ee * qa» ys 7 : oe * a * Se ee, ad ¥ a CON DIENT. Sk Part I. Published 11th September, 1919. Geographical Introduction. (N. Annandale.) The Molluscs of the Inland Waters of Baluchistan and of Seistan, with a note on the Liver-fluke of Sheep in Seistan. (N. Annan- dale, B. Prashad and S. W. Kemp.) : as Notes on Fish of the genus Discognathus from India and Persia. (N. Annandale.) ae Be Notes on Odonata collected in Seistan and Baluchistan in winter. (He Cs Fyvaser.) .: Part Il. Published 20th December, 1919. Sponges, Hydrozoa and Polyzoa of Seistan. (N. Annandale) Carabidae from Seistan. (H. E. Andrewes.) Report on the Fresh-water Gastropod Molluscs of Lower Mesopo- tamia. Part I.—The genus Limnaeca. (N. Annandale and B. Prashad.) Ae ta On a new genus and species of Coccidae from North-Western India and Eastern Persia. (E. E. Green.) ath ; Notes on two collections of Birds from Seistan. (E. C. Stuart Baker.) a ae =h: ae = Part IIL. Published 29th April, 1929. Note on the occurrence of the Leech Limnatis nilotica in Seistan and the Afghan-Baluch Desert. (N. Annandale and Amin-ud- Din.) Be aes : ae oa Notes on some Asiatic species of Palingenta {Order Ephemerop- tera]. (F. H. Gravely.) - : a ae List of Entomostraca collected in Seistan and the Baluch Desert. (R. Gurney.) .. : ; Report, on the Fresh-water Gastropod Molluscs of Lower Meso- potamia. Part II.—The Family Planorbidae. (N. Annandale.) Part IV. Published 15th September, 1920. The Fish of Seistan. (N. Annandale and Sunder Lal Hora.) The Rhynchota-Heteroptera (Notonectidae and Corixidae) of Seistan. (W. L. Distant.) .. oe ie The Gross Anatomy of Corbicula fluminalis. (B. Prashad.) 103 117 na Contents. Part V. Published 29th April, 1921. Note on the Leech Limnatis nilotica. (Tokio Kaburaki.) sty AB Tes Report on the Fresh-water Gastropod Molluscs of Lower Mesopo- tamia. Part III.—The Families Neritidae, eae and Melaniidae. (B. Prashad.) ay 713 Notes on Larval Trematodes from Seistan. (S. Kemp.) ot « (me22G The Aquatic Fauna of Seistan: A Summary. (N. Annandale.).. 235 LIST OF PLATES. ee Follow page Plate I (Map of Seistan and the neighbouring country) Plate II (Cliff on the Western shore of the Hamu} 16 Helmand, near Lab-i-Baring Plates III—VIII (Mollusca) .. cts a oe od Plates IX—XI (Fish) ae = Es ee eta7ie: Plate XII (Polyzoa) a a so 25-208 Plates XIII—XIV (Mollusca) so east ee rs. VA LEO Plates XV— XVI (Fish) Plate XVII (Shaped Rafts from India and Seistan) 5 tO ies 7 Plates XVITI—XX (Palingenia) et os So ae Plates XXI—XXII (Coccidae) i a Jaf a LZO 4 ae? ie ay”) Lu ANA, ~ f ‘i. é Fs Pia ine W | a hie f a a) : Th ; ERRATA. SP In the twenty-first line from top of page 45 for ‘‘are neither swollen ”’ read ‘‘ are neither greatly swollen.” In the twenty-fourth line from bottom of page 59 for ‘‘L. jen- kinsontanus’’ read ‘‘ L. jenkinsianus.” In the twenty-third line from bottom of page 81 for ‘‘ rice-beds ”’ read ‘‘ reed-beds.”’ . INDEX. ——<>— [N.B.—An asterisk (*) preceding a line denotes a new ae or subspecies; a dagger ({) indicates anew species; a double dagger (tf nyms are printed in italics. | A Page Achtea bimaculata .. : 243 tAdiposia 152,155, 157, 182, 185, 186, ; 247 fongicauda 188 macmahoni 151, eres NS 4s Sk Oues 190, I91, 239, 240, 249 riadinaea 151, 154, 157, 189, 239, 249 Aedon familiaris Bit 123 Aegialitis dubia 129 dubia jerdoni 129 Aeschninae 79 Afrindella gt Agrioninae : 80 Agrobates galactodes familiaris 123 Alaemon alaudipes pallida 125 desertorum 125 Alcyonella 94. tAlocinma 222 tw223 Alveolina 12 sphaeroidea 12 Amblyceps 5¢ . 184 Amuicola Dei, Daly Pliy PG ZAM BAB} orcula ee 23 parvula 17 7 LON 27; (Alocinma) alticola 24 ejecta 56 223 o.cula 56 ad 27 {sistanica T7ipelOn 215) 244 20249 Ammomanes deserti iranica 125 phoenicura zarudnyi 126 phoenicuroides .. ie 125 Anagenesia 138, 140 Anisops fieberi ZOOS 237 25L A piostoma zarudnyt ae 170 | Ardea cinerea ots 132 Ardetta minuta a2 | Argya caudata 122 caudata huttoni 122 Arthropoda 238 Athene bactriana 126 noctua bactrianea 126 Australella 93 B Balitora 164 maculata 164. Bembidium 99 latiplaga 100 niloticum 100 varium Sc 50 100 Bithinella oe s22L 222 a new genus; syno- Page Bithinella miliacea 222 palmyrae . 222 Bithynia 22 222 22),.223 badiella 222, 223 orcula 5 24 orcula parvula .. 2 rubens 223 tentaculata 246 Bosmina longirostris 145, 252 Botaurus stellaris Tse Brachinus IO! Bubo bengalensis 126 bubo bengalensis 126 bubo turcomanns 126 Buccinum scabrum.. Ba BY, Buto viridis -244, 249 Bullininae aie ne 148 Bullinus 147, 148, 246 contortus .- 149, 215 contortus approximaus £49 dybowskii 149 Buteo ferox 126 Bythinella : 221 Bythinia ejecta 223227 Cc Caprimulgus mahrattensis 126 Carabidae ae 99 Caridea = 244 Cercaria 229, 231, 232 Ceriodaphnia pulchella 145, 252 reticulata 145, 146, 252 Characeae ais OMON 70.1242 Chironomus ats 242 pictulus 250 Chlaenius semicyaneus 100 spoliatus 100 Chaetogaster ae 23 bengalensis Dd, Ahi, Bis punjabensis 238, 253 Chelonia 244 Chettusia leucura 129 Chondrostoma 164 Circus aeruginosus .. 127 macrurus 127 Cirrhina 69, 162—167 afghana oe 162 latia 67 Cladocera 237 Clangula glaucion 134 Cobitidae I5I, 152, 154—157, 178, 182, 188, 247 Coccidae 116 Page Columba eversmanui 127 intermedia t2 livia intermedia 127 Copepoda 237 Coracias garrula semenowi 1205) 226 Corbicula By ily Gy/g GOO Uy cor 58, 236 crassula : 58 fluminalis 18— se ie SiS 189 » 209, QUT, (235, 23/— 242 3-250 fluminalis coy .. 30 58 iagillierti 209, 211 Corbitis stoliczkae Sc 178 Corixa atfinis 265) 23713) 2405251 hieroglyphica 205, 240, 251 ragenhoteri as 206 yrandaua Se 206, 240, 251 yseistanensis 209, 251 substriata 205, 251 Corixidae ; ZO5 2805 237; Corvus cornix Bh 0 122 cornix sharpil 121,122 frugilegus 122 frugilegus tschusii 1222 Coturnix communis 128 coturnix ao 8a Crossochilus in 67, 69, 162, 163 Crustacea 145, 236, 240 Culex fatigans. . : 251 Cursorius gallicus 50 129 Cyclops leuckarti DAS N23 7252 strenuus W455 2379 252 viridis 145, 252 Cygnus cyguus an ae 132 MUSICUS 3 132 Cyperus alterifolius — 3 196 Cypridopsis dentatomarginatis 146 Cyprinidae 5 Uy UGLY UEOs uy Cyprininae 151, a 15s, WHO, bs WOS Cyprinion kirmanensis : 159 Cyprinodontidae 152 Cyprinoidea : 156, 186 Cyprinotus incongruens 146, 236 Cyprinus : 104. godiyari af gotyla Ac 75 lamta Win Gis Mon (Garra) lamta .. oe 164 Cyvena coy crassula 58 Cyrenidae 18, 57 D Dafila acuta 133 acuta acuta ae m38 Daplinia longispina rosea 145, 252 magna 145, 146, 237, 252 Diaphanosoma brachyurum 146 Diaptomus salinus .. 146 Diplocheila transcaspica 100 Diptera .. 239 | Discognathus 65-67, 69, 78, 156, 158, 161, 162, 164—166, 245 jadiscus 66, 68,69, 71,151, 152, 154, 156, 162, 163, 165—167, 238, 240—42, 249 | Page Discognathus bicornutus 65, 66 ceylonensis a 77 y elegans 66, 76, 77 sory 66, 72, 75 gravelyi 66, 67 imberbis ae 66 jerdoni : 66, 73—77 jerdoni kangrae Vi Gel hangvae 74 lamta 65, 66s Fil AZ 74, 7%, 158 macrochir ahs 67, 78 nasutus 67, 78 persicus 72 jphryne 6—70, 151 pene 156, 164, 166, 238, 239, 249 platycephalus 65, 66, 77 quadrimaculatus 67, 69 rufus Sas 158 stenorhynchus 66, 74, 76 variabilis 67, 69—77, 154, I6!, 164, Met 166, 167 vinciguerrae ae 67 _ wauae 65, 66, 69 Distichus planus 99 DOstiaws 215 Dyschirius 99 Dytiscidae 237 E Entomostraca 145, 237, 238, 242 Ephemera : a 138 vulgata PLAT pia Ephemeroptera : Lay Ephydatia fluviatilis "83, 84, 87, ee 96, a 242, 243, 247+ 253, fluviatilis himalayensis oes 87 fluviatilis intha Se 83 fluviatilis syriaca 83, 87, 90 meyeni ae 247 mulleri go multiformis go Ephydridae 239 Erium Ii7 Eucypris clavata Wl, Bas DRZ pubera i 146 F Falco aesalon insignis T2027, Fasciola 22 AT OZ OB aegyptiaca 63 augusta a 63 gigantea De OVNs (O55 53 hepatica 51, 62, 229, 231 Francolinus francolinus henricii 128 vulgaris : 128 Fredericella sultana LOO MeZAe sultana joidanice 83, or, 242, BARE 247, 253 | Fulica atra atra en IS) G Galatea 215 Gabrida cristata chendoola 121 cristata magna . 125 Page Gallinula chloropus parvifrons 121, 128 Gammarus ae - 243 Garra 158, 161, 162, 164, 165 245 crenulata . 73 gotyla : 75 jevdont an 73 lamta 3 73 nasutus 162, 163. 165 obtusa 73 persica Go 72 rufa :: 73 (Garra) cevlonensis 77 Gastropoda ae Pd yA > Gerris 239, 243 Glaucion 134 Glaucionetta clangula 134 Glycia ornata a IOt Glyptosternum 185, 186 Goniorhynchus 164 gotyla 73 stenorhynchus 74 Grus grus 130 Gyraulus 20, 52, 56, 147, 246, albus 535 57 convexiusculus 18, 19, 2I—23, 40, 52—57, 147, He ro 231, 232, DE] o ba 242, 246, 250 devians auphintica,” 55 euphraticus 18—22, 40, 52—55, TAG 148, 215, 235, 2375 238, 240, 242, 246, 250 intermixtus TAGE RUA GR 2 sic saigonensts 52h 55 velifer A 56 H Halmopota viridescens 239, 244, 25: Hemiarax a 79 ephippiger Ne 79 Herpetocypris reptans 145, 252 Himantopus candidus 23 12 himantopus.. 129 Hirudinea 230 Hyalinella 3 so Oy OA Hydra vulgaris 83,90, wh 253 Hydrobia 221 Hydrobiidae 17 23,2 2058 221 Hydrocena Bas 27 Dydrometridae 239 Hydrophilidae 237 Hydroprogne Se 131 F ydrozoa : 83, 90 ! Ilyocypris bradyi 145; 146, 236, 237, 252 Insecta 236, 240 Ischnura 80, 81 elegans ’ 80, 242, 250 senegalensis 80 Isidora brocchii 149 Ixobrychus minutus 132 L Labeo rohita 169 Lamellibranchia 18, 57 x1 Page Lamellidens Sin ay ZA SO jenkinsonianus Ss 59 marginalis NOp ZO 2 bs 57159239 marginalis candaharica . 60 marginalis corrianus PF 59 *marginalis rhadinaeus TOROS Os 60; 625,160, 189, 239), 240, 242), 247, 250 Lanius cristatus isabellinus 123 cristatus phoenicuroides 23 phoewncuroides 123 Larus argentatus cachinnans 131 gelastes 131 ichthyaetus 731 ridibundus . 131 Leptestheria tenuis 145, 238, 252 Libellulinae 5 A 79 Limtaea 19, 20, 22, 39, 41, 42, 45, 49, 103, 104, 246 auricularia 305,40, 43501045 05s auricularia persica ae 4! bactriana 18, 19, 22, 39, 40—az, 44, ASP Act OPelOd a OO, 1075) Tle oie. 237, 238, 246, 250 canalifera 108 chlamys - 47, 246 fcor LOAS LOOM Tee 1st euphratica 113 fgedrosiana 18, 19, 21, 22, 39, 41, 42, 48—50, 63, 104, ee nie Wie SOUT 239, 241, 242, 246, 250 *oedrosiana fectlabens DOs LOR) 225 23, 40,41, 42, 49, 106, 241 250 hordeum 18—21, 39, 41, 51, 106, 114, 246, 250 intermedia ays 45 Tiranica 18, 19, 39, 41, 42, 43, 49, 106 lagotis 45, 47, 113 lagotis persica .. a2 43 lagotis subdisjuncta 47 ovata inflata Lil peregva 48, 49 peregra canalifera 104, 106, 108, 113, 215 peregra ovata .. St 45 peregriforvmis 108, III persica . . 18—26, 39, 41, 106 stagnalis Se 20, 246 subpersica 41, 107 tenera : DUS ys! tenera angustior 114 tenera euphratica 104, 106, 113, 114, 215 truncatula 18, 19, 39, 41, 50,51, 104, 106, 115 truncatula longula BI 51 Limnaeae ee 21139 Limnaeidae 18, 21, 39, 47 Limnaeus lagotis persica ae 41 Limnatis nilotica 135, 136, 213, 214, 236, ; : 247, 253 Limnatis (Poecilobdella) granulosa 136, 214 Limnicythere inopinata : 146 Limosa belgica oe 129 limosa 126 Page Lithoglyphus ae ate 221 Lophopodella ot bn Clos Qa capensis 56 gO capensis michaelseni ne 96 carteri 83, 91, 96, 238, 239, 247, 253 carteri davenporti 96, 247 carteri himalayana eas 97 stuhlmanni ae, aa 96 thomasi an ae 96 M Machetes pugnax .. Sc 130 Macrocorisa geffroyi 205, 237, 251 Marmaronetta augustirostris 133 Melania 28335 222 adspersa $0 5.0. Biba. BS elegans ai Deiehel 27 pyvamis 28, 32 pyvamis adspersa si 33 scabra elegans .. es 37 tigrvina Pi Bi 30 tuberculata oe 3 224 tuberculata luteomarginata 35 (Plotia) scabra elegans oe 33 (Strviatella) tuberculata ie AS (Striatella) tuberculata flavida 34. Melaniidae I7, 21, 28—30, 215, 224 Melanoides PBN hoe RYO Pepa elegans Hig es 2 fascrolata 31 pyramis 17, 18, 20, 32, 34—37 235 Ppyramis flavida 17, 195120, 225,20), 30; 34—36, 225, 220, 235, 244 *pyramis leopardina 30, 33, 36 pyramis luteomarginata 17, 19, 29, 30, 34, 35 *pyramis puteicola 33, 34 scabra 19, 29, 34, 37 scabra elegans 18, 29, 32, 37 tigrina 18, 29, 30, 36 tuberculata 29, 31,32 35—37, 149, 224 Melanopsis 20, 37, 225, 240 costata 225, 226 deserticola Oe 29, 30,37 nodosa oo 225 praerosa nana .. i 39 subtingitana 226 subtingitana jaevis 226 turcica 225 turcica bellio 226 Merops persicus persicus Bc 126 « Metabletus fuscomaculatus .. IOI Microlestes corticalis ae IO Micronecta biskrensis 206, 236, 251 jdesertana 206, 236, 251 Microvelia 515 243 Mollusca ie 205 glauca 205, 237, 251 glauca marmorea 50 205 Notonectidae 205, 237 Notonectinae 3 ae 205 Numumulites atacicus ere II Nyroca ferina 134 O Odonata .. : ate 79 Oenanthe alboniger Se 12 deserti atrogularis 124 isabellina Ae 124 picata se as 124 Oligochaeta a oh 238 Orthetrum at ae 79 taeniolatum ats Ae 79 Ostracoda 237 P Palingenia r/o ugTo)n SUehe) ie 141, 142, 240, 250 longicauda 138, Xili Page Page Falingenta minor are 143 Pogonus micans 1 “* 99 vobusta oe ors 140 | Polyzoa 14, 83, 91, 239, 240, 242 (Anagenesia) greent : 141 | Porifera .. vs -: 84 (Anagenesia) lata a 140 | Porzana parva -- + 128 (Anagenesia) minor a8 143 potzana - 128 +(Anagenesia) picta Sc 140 | Potamocypris villosa 145, 146, 237, 252 (Anagenesia) robusta 3 140 | Potamogeton 6,7, 10, 47, 50, 238, 240, Paludestrinidae ae Sc 23 241 Paludina ie 27221 222 lucens : -- 9, 241 badiella Ss we 27 pectinatus °9, 49, 50, 145, 242 parvula ies ate 27 perfoliatus.. en O42 Paludinella fs ve 221 | Potamon gedrosianum 240, 251 Paludomus a0 222° (Potamon) potamios gedrosianum Passer moabiticus yatti Git, Pals v4 ae ; Bae 189 montanus ; i 124 Prinia gracilis lepida 123 montanus dilutus ae Se 124 lepida -- - 123 Pectinatella davenporti Ss 97 | Pseudamnicola a 235 24, 223 Pectinibranchia ae a 17 macrostoma ... - 277, Pelecanus onocrotalus eS 131 | Pseudococcinae Sc wi 3) LIZ, onocrotalus onocrotalus .. 131 | Psilorhynckus .° 162, 164 Phalacrocovax carbo AF is 132 tentaculatus .. .- 163 carbo subcormoranus ee 132 | Psychoda bengalensis mC 220 Phoenicopterus minor ae 132 Pterocles avenarius .. ae 128 Phoeincurus erythronotus 124 arenarius caudacuta oD 128 Phragmites Dh 8, 26, 49 Pterostichus a ve 101 communis rn 3. 194, 241 Pulmonata 5¢ 5c 18 kharka ame ar 8 Physa tiberidensis .. oe 149 Q (Isidora) Brochii oe 149 (Isodora) lirata ss 149 | Querquedula crecca.. =o 133 Physopsis 147 Pisidium paludosum 18, 195 57 Planorbidae So Me Bis a6 BG. yy R eases of os a Rana cyanophlyctis a 245 Cais bs Eos 6 Sheree: seistanica 245, 249 i RVESSUS iri ; esculenta : SC 244. ET oes ii i : : ay esculenta ridibunda 244, 249 euphraticus 2: 53 52 Se 23912377240 nities ise 3 = Rhynchota Heteroptera a 205 saigonensis sic 56 EGG vortex 54 S Planorbis (Gyraulus) ‘devians euphratica 54,148 | Salix acmophylla .. a6 10 (Gyraulus) intermixtus of 148 | Salmonidae Be 154, 168, 186 Platycava a: -» 71,164 | Saxicola albinigra .. 50 124 nasutus 50 164 capistrata aie 56 124 Plotia.. & 28, 30, 34 deserti ie sc 124 Plumatella ys A gi tsabellina *% 124 bigemmis fe 94, 242, 247 | Scaphiodon a 156, 158, 245 indica as as 94 | macmahoni I51r, 152, 154, 156, 158, jheringi oe os 94 160, 239, 249 lendenfeldi she ae 94 muscatensis ae 56 159 longigemmis .. oC 94 | Schistosoma japonicum 229, 231 persica rs -- 92, 247 | Schistosomatidae .. -, 22 philippinensis .. Es 2 | Schizocypris ae 154, 156, 175 punctata a0 Os OL: bruce (5p Esse l5Osn 5S) 70s 2408 tanganyikae .. 50 g2 241, 245, 249 testudinicola .. -- 83,92 | Schizopygopsis 154, 150, 176, 246, {(Afrindella) persica 83, 92, 240, 253 stoliczkae 151,154, 156, 173, 239, (Afrindella) tanganyikae .. QI | 245, 249 {(Hyalinella) ee ea 94, 253 | Schizothoracinae I51I, 152, 154—I156, Plumatellidae QI | 168, 178, 246 Podiceps albipennis. as 134 | Schizothorax 154, 156, 170, 176,198 cristatus 5 ae 134 | affinis = 170 fluviatilis albipennis 50 134 altior ae as 173 Poecilobdella Be a lSO, 2041 brevis we 170 Page Schizothorax intermedius 170 macrolepis 170 minutus 170 ritchianus Se : 170 zarudnyi I5I—154, 156, 165, 169, 170, 173, 176, 177, 192, 198, 232, 239—242, 245, 249 Scirpus Sa 36 10 littoralis 8, 194, 241 Segmentina 20, 52, 50 calathus 18, 19, ai 40, 55, 36, 215, 2 38, 240, 250 Siagona depressa 100 europaea 100 Siluroidea 186 Simocephalus exspinosus sic 146 vetulus Wil 5 22375 GAGA) Sisoridae bc 186 Spatula clypeata 133 Spirogyra 79 Spougilla alba 83, 84, 86, 96, 24, 247 *alba rhadinea .. 85, 253 carteri 83, 96, 247 fragilis eo 90 lacustris 3 84, 86 *“lacustris ineptorum 86 lacustris montana — 86 (Eunapius) carteri ST 2Ae253 Stenothyra ec oe 222 Steyvna anglica 131 fluviatilis 131 hirundo 131 minuta 131 minuta gouldi 131 nilotica 131 Streptopelia turtur decaorta 127. Striatella : 28, 34 | Sturnus menzbieri 123 vulgaris nobilior 123 vulgaris poltaratskii 123 Succinea bensoni 114 Sylvia jerdont 123 mystacea 3 121, 123 Symplecta elongata .. 2385 251 punctipennis 243, 251 T Tadorna tadorna .. aie 133 Talorchestia sie ae 243 Tamarix articulata .. 118 stricta ore Se 118 a Page Tellina .. an 4 209 fluminalis ; 2 5051209 Theobaldia longiareolata 236, 251 Theodoxia jordant 50 218 Theodoxts An 216, 217 jovdant 218,-21¢ michont ee 218—220 Theodoxus Sc ws 2157. fluviatilis ae ats 217 Tichodromia muraria 123 Tinnunculus tinnunculus saturatus 127 Totanus glareola_.. 130 totanus eurhinus 130 Tricula D2 222 montana 221 palmyrae : 222 Tringa alpina alpina 130 Turtur visorius dbo 510 127 Typha “es 8, 242 angustifolia 241 angustata 194 latifolia 196 U Unio as be Bo Ses SO Unionidae 18, 59, 160, 210 V Vallisneria 9 spiralis 9 Valvata 27 Vivipara 19, 27, 246 dissimilis : 27 (dissimilis) hilmendensis oy hlimandica 239, 246, 247, 250 hlimandensis 7s 19—21, 27, 28, 240 sindica 50 246 Viviparidae se Be) Gy Zi Xx | Xiphidiocercaria 22 Z Zannichellia Fc aie 145 palustris 5c 6, 47, 237, 238 Zaponia parva parva ne 128 pusilla a -- 128 ee a ed REPORT ON THE AQUATIC FAUNA OF SEISTAN. PREFATORY NOTE. The series of papers of which the first instalment is now pub- lished will be based mainly on the collections made by Mr. S. W. Kemp and myself in Seistan and Baluchistan in November, De- cember and January, 1918-19. The main object of our tour was to enquire into the etiology of the disease Bilharziasis or Schistoso- miasis, and particularly to discover whether the parasite occurred in Seistan, or whether any known intermediate molluscan host was found to be in that country. For this and other work on the zoological aspect of the disease a generous grant of money was made by the Council of the Indian Medical Research Fund. So far as it was concerned our results were entirely negative; but we took the opportunity to make as large and complete a collection of the aquatic fauna as the time at our disposal permitted. As this fauna is a limited one (7.e. includes few species but large numbers of individuals) we were able to obtain a more representa- tive set of specimens than would have otherwise been the case. Our material, moreover, is supplemented by the specimens col- lected by the Seistan Boundary Commission of 1902-03 under the command of Sir Henry McMahon. To Mr. Kemp my first acknowledgments are due. Indeed, a great part of the zoological success of our little expedition must be attributed to him. I only regret that it is impossible for his name to be more intimately associated with my own in the prepa- ration of this report. The reason is the stress of other work and the necessity for a visit to Europe on his part. I have also to thank the political officers in Baluchistan and Seistan, in par- ticular, Mr. B. J. Gould, I.C.S., His Britannic Majesty’s Consul in Seistan, Major D. Heron, I.M.S., C.I.E., Medical Officer and Vice- Consul, and Major. W. G. Hutchinson, Political Agent, Chagai. The officers serving on the Eastern Persian Cordon also gave us much assistance in the way of transport and supplies. Babu J. N. Bagchi, Head Clerk, Zoological Survey of India, accom- panied us and did good work. I have to thank Mr. EK. Vreden- burg of the Geological Survey of India for valuable suggestions embodied in the Introduction, while Dr. Baini Prashad, Offg. Director of Fisheries, Bengal, Bihar and Orissa, has helped me in various ways. 2 Records of the Indian Museum. [Vor. XVIII, 1919.] In examining the collection from Seistan it has been neces- sary also to examine much material of other provenance, and I have found it convenient to include in this volume certain studies of wider geographical scope. N. ANNANDALE. Calcutta, 17-v-19. GEOGRAPHICAL INTRODUCTION. By N. ANNANDALE, D.Sc., F.A.S.B., Director, Zoological Survey of India. (With Plates I-I1.) Seistan and the Helmand River. The Persian district of Seistan, at periods in its history an independent state and at others a part of Afghanistan, lies roughly between long. 61° and 62° E., lat. 30°50’’ and 31°50” N.; but its precise boundaries are not delimited to the west and south. It consists of the delta of the R. Helmand and the Hamun or basin into which that river flows. These lie, surrounded on all sides by stony desert, in a depression less than 2000 feet above sea level. The Helmand rises in the Hindu Khush in about lat. 68°40” and long. 34°30” and flows for three hundred miles through the moun- tains of Afghanistan, receiving many tributaries on its way. It then debouches on the desert plateau of Registan and, some distance after doing so, is joined by its largest tributary, the Arghandab. The course of the united waters, which flow in a deep bed through the desert, is S.S.W. for some seventy miles. They are then deflected by a small range of hills through which they have been unable to cut their way, and continue westward, with a distinct southward bend, for about another hundred and fifty miles. Then, on reaching the southern limits of the old delta, the river breaks up into innumerable channels, partly natural, partly artificial, which turn northwards. In these channels, the chief of which is ultimately known as the Rud-i-Pariun, much of the water is dissipated, but what remains finally drains westward into the Hamun-i-Helmand or Hamun-iSeistan, a large basin (or rather series of basins) which occupies a- considerable but extremely variable area. After the junction of the Arghan- dab and the Helmand very little additional water, not nearly enough to compensate for evaporation, enters the system ; for the desert is practically rainless, even in Seistan the rainfall is only a 4 Records of the Indian Museum. [Vor,. XVIII, little over 2} inches a year,’ and there are no permanent affluent streams. Seistan is a well-watered country, but its water-supply, like that of Egypt, depends not on local rainfall but on the rainfall and the snows in a mountainous region many miles away. The fact that, unlike most of the lake-systems of Central Asia and Persia, that of the Hamun-iSeistan has not dried up is to be explained only by the peculiar course of the Helmand, the greater part of which traverses comparatively damp mountainous country. The Hamun, however, is not the final repository of the soluble matter which the river inevitably brings into it. If it had been so, with its limited area, all its water would have been salt long ago. On maps of Persia and the adjacent countries a river, named the Shelagh (or Shila) river, is marked running southwards and eastwards from the Hamun into another basin, the Gaud-i-Zirreh, which occupies a very large area in the great desert of south-western Afghanistan. This is the Dead Sea of the | The following tables of rainfall (in inches) are derived from official sources :— ig | S ee 4 : ¢ [aa ; =e ee Season. 3 : E Sell Sige 3 iy | 3 © 3 = iS D mn Es op =) ates = is D = O) = a OQ O a5 DG op) & Siac) October to March sot E750 sOROO et eGi ae OediG al 2206 | SOL 79S ames April.to September... | 2°40 | 1°72) O°81 |.3°40 |\"0745 |) 374107! 1 -G25 i aiees TOTAL we | O'45- | 9°78! | 5°32 11°85 | 2°51 | 8:50" 1) Osha aaeam RAINFALL AT NASRATABAD, SEISTAN. Month 1902 190 190 1905 Total Avera sia ; phils 2e3: Oe DoE ; month. January aaa a: 1°46 0°60 0°44 2°50 0°83 February seul Notane= | corded. 0°42 O'15 o'28 | 0°85 0°28 March As be 0°07 o'19 ORV wa56 o'51 April Aa 0°03 I‘19Q ia aA Te22 O’4i May a6 a 0°05 sia ae 0°05 002 June a ate | 0:07 Stic =o) |, O07 ee July ae August September a a um Sn sek Age vee October By o'2 me rae a |) 0727, 0°09 November es O'17 0°07 0°06 ve 0°30 o'Lo December $i 0'60)7 ||; org oO'ol as O'74 O25 ToTAL re 1°07 || s4326 L‘OI 1°09 7°43 2°95! IgIQ.] Seistan and the Helmand. 5 Helmand system.' Except in times of exceptional flood the Shelagh is a deep stream-bed, dry except for scattered pools of water so saline that tamarisk twigs at the edge are coated half an inch thick with salt ; but the water in flood-times, at the very site of these pools, becomes readily drinkable and the river is a raging torrent. Similarly the Gaud-i-Zirreh is at most seasons a wide plain covered with a thick deposit of salt and containing pools and swamps of saline water ; but it may become a real lake for the time being, with water of comparatively but not actually low salinity. The climate of Seistan, though regular from year to year, is one of considerable extremes and with only two seasons, summer and winter. In December and January the temperature usually falls below freezing point at night, while in summer it frequently rises to T15°F. in the shade by day, with a maximum of about 117°. It is, however, very exceptional for the larger bodies of water to freeze completely, while the summer heat is tempered by an almost unceasing wind. Wind,” indeed, seems to be the most con- stant feature of the climate, and its direction seldom changes. The direction is about N.N.W. For weeks on end in summer time it does not stop, and even in winter windy days are commoner than calm ones, and the wind is as arule practically N.N.W. Such rain as falls, falls mainly between the end of December and that of April, in which thunderstorms accompanied by hail and causing sudden floods sometimes occur; but the real flood-season takes place when the snows begin to melt in Afghanistan between March and May. Different Types of Aquatic Environment in Seistan. A summary description of the Helmand-system and the cli- mate of Seistan has been necessary to explain the very existence of the Hamun-i-Helmand as a lake, and of an aquatic fauna in the country. A detailed account® of the system would be impossible | In high floods a certain amount of water runs from the Helmand direct into the southern part of the Hamun system through a channel probably of artificial origin, and in exceptionally high floods some may flow direct into the Gaud-i- Zirreh, but this is not the normal course. 2 ** At the end of May, or middle of June, the celebrated bad-i-sad-o-bist rug (120 days’ wind) sets in and blows with but little cessation till the middle or end of September. It blows unceasingly for four or five days at a time, usually attaining its maximum daily velocity between midnight and 5 a.M and again between 8 A.M. and5p.M. It moderatesa littlein the early morning andevening. After four or five days it drops a little for a day or two, only to recommence with renewed violence. It blows with appalling violence, reaching the maximum velocity. as recorded by the Mission anemometers, of 72 miles per hour. It blows always from one direc- tion, viz. a little west of north, i.e. between 3163° and 3332°.’’ (From an official document). 8 The most complete account of the topography of Seistan yet published is to be found in Tate’s Sefstan. A Memoir on the History, Topography, Ruins and People of the Country, Parts I-IV (Calcutta, 1910 and 1912). See also McMahon “ The Southern Borderland of Afghanistan,’’ Geog. Fourn. 1X, pp. 393- 415 (1897), zd., ‘‘Recent Survey and Exploration in Seistan,’’ Vol. XXVIII, PP: 3337340 (1906), and Rawlinson ‘‘ Notes on Seistan,”’ Fourn. Roy. Geog. Soc. II, pp. 272-294 (1873). 6 Records of the Indian Museum. [ Mor. IF Nos. 10 and 11, p. 437 (1915). 2 See Annandale, Rec. Ind. Mus. XIV, p. 172 (1918). ® See the works of Blanford and Huntington already cited, and also de Morgan’s note in Revue de L’Ecole D' Anthropologie for 1907 (Paris), pp. 214—- 215. 1919. ] Seistan and the Helmand. 15 The Water of Seistan. We collected a considerable number of samples of water in the Hamun-i-Sabari and other bodies of water in Seistan, but unfor- tunately several of the bottles were broken on our journey. By the kindness of Dr. H. H. Hayden, F.R.S., Director of the Geo- logical Survey of India, the samples that remain have been examined in the laboratory of that department. to ensure an accurate and detailed analysis, sufficient clearness the general character of the salts present. They are not sufficient but indicate with Our sample of water from the Shelagh river was lost, but one of the salts crystallized on its shore has been analysed. The following are the results :-— Analysis of Water Samples trom Setstan. From south shore of Hamun about From a small sa- | line stream run- | From edge of Ha- ning down to but | mun about one i a aoe oe not quite reach- | mile north of Lab- Bet ae Eich ing the Hamun. | i-Baring. On b et ae on About a mile and| stony shore be- h = ane mueey | halfmorthiot bab= les low clits SORE i- Baring. Quantity received 560 c.c. 500 c.c. 470 C.c. Given in grammes per 1000 €.C¢. AlgO3 and Fe2Os nil Trace nil (ee ae 0°0496 0°8629 _ 0°0345 MgO | 0°2220 2°4476 O'1542 Alkalies weighed as chlorides 1°4640 20°2650 0*9100 Na,O o'721 10°683 0°313 K,0 . 0'0664 0°009 0'201 Cie: 0°5293 10°8985 0°3399 SO, 0°4759 5°2136 0°2948 Suspended matter 0°0426 grammes | O'1200 grammes | 0°0872 grammes. Analysis of Salt from the edge of Shelagh River, near Girdhv, Seistan. Amount insoluble in boiling water 29°78 Al,O, and Fe,O, o°Io cam 4°64 Mg 0°27 Na 13°10 K 2°57 Ch x: 15°62 SO, 22°39 co, Tks Water of Composition 10°75 Total 16 Records of the Indian Museum. [Vou XVIII, 1919.] From this report it is evident that considerable amounts of salt are present even in waters that are potable, as that of the Hamun near Lab-i-Baring; that the salt is not merely sodium chloride but of mixed composition, and that its composition varies greatly even inthe same part of the Hamun in different circum- stances. I have already noted the changes in salinity produced in the water by a cessation of the normal wind (p. 13, antea). In both samples from the Hamun salts of sodium are the most prevalent, but salts of magnesium, which are usually more dele- terious to animal life, are present in considerable quantity, and in one sample those of potassium are also fairly well represented. The sample of water from the small stream was taken about a hundred yards up from the beach of the lake, in a little gully in the clay cliffs. The stream was a very small one and rose in clay among small hills at no great distance from the lake. We may take it as representing a solution of the soluble salts in the clay of the cliffs. The salt from the margin of the Shelagh river, on the one hand, represents the offscouring! of the whole Hamun system. Here magnesium is poorly represented, while both sodium and potassium are present in fairly large amounts. We have as yet no data, therefore, to estimate the differen- tial effect of water of different chemical composition on the aquatic fauna of different parts of Seistan, and, indeed, to arrive at any results of the kind would necessitate a very long and arduous investigation carried out through the seasons and under all pos- sible conditions of flood and the reverse. All that can be said is that the aquatic fauna throughout the country lives in abnormal and variable types of environment so far as the composition of the water is concerned. ! It cannot, however, be composed of the same proportions of the different salts that occur in the water in solution, for some salts crystallize out before others. a ats 7 sriponiaty , “while ; melee a : ~ Gx Ray 28 Feit 2708, - i ee ta MY) Ty!) ieee: 1 EXPLANATION OF PLATE II. CLIFF ON THE WESTERN SHORE OF THE HAMUN-I-HELMAND NEAR LAB-I-BARING. DECEMBER, I918. The photograph was taken on a calm day two days after the wind had fallen. SoH BY a ae Rae od Uniform lacustrine deposit of stiff clay. Superficial layer of more friable earth and water-worn pebbles. Blocks of clay fallen from the cliff. Beach of water-worn pebbles derived from superficial layer of cliff. Normal flood-level marked by fragments of reeds and other vegetable debris. Winter water-level in windy weather. Water-level after two days’ calm. Footpath. Reed-beds. PRrAP ET), UU aTAIA PLATE II. REC. IND. MUS., VOL. XVIII, 1919. | JO SOA 84 JO SAO 944 Ye pau d oY porrisuo-0joud ‘si6L “99 “ONIYVE-I-GV1 YVAN NNAWVH - ojOYd ‘duweay AAS 25 |. a Dead or ag THE MOLLUSCA OF THE INLAND WATERS OF BALUCHISTAN AND OF SEISTAN. (With Plates ITI—VIII). By N. ANNANDALE, D.Sc., &.A.S.B., Director, Zoological Survey of India, and BAInt PRASHAD, D.Sc., Offg. Directory of Fisheries, Bengal, Bihar and Orissa, with a note on the Liver-Fluke of Sheep in Seistan, by S. W. Kemp, B.A., Superintendent, Zoo logical Survey of India. In studying the aquatic Mollusca of Seistan we have found it necessary to study also those of Baluchistan, from certain parts of which abundant material was available. These parts are the hill-country of the Quetta and Pishin districts in Northern Balu- chistan, in which one of us collected the material himself, the great Baluch-Afghan desert, in which a collection was made by Dr. N. Annandale and Mr. S. W. Kemp, and Persian Baluchistan, in which, many years ago, the late Dr. W. T. Blanford! obtained many specimens now in the Indian Museum. We have seen no shells from the British district of Mekran, which, however, marches with Persian Baluchistan on the west, from Las Bela and the neighbouring states, or from the Indus plain south of Sibi, in which the fauna is probably quite Indian. The following is a list of the species and varieties now known, arranged according to the classification set forth by Pelseneer in Lankester’s Treatise on Zoology, Vol. V (1906). List of Freshwater Molluscs of Baluchistan and Seistan. é Gastropoda. PECTINIBRANCHIA. Fam. Hydrobiidae. Amnicola (Alocinma) sistanica, sp. Seistan. nov. ? Amnicola parvula (Hutton). N. Baluchistan. Fam. Viviparidae. Vivipara hilmandensts, Kobelt. Seistan ; Afghan desert. Fam. Melaniidae. Melanoides pyramts (Hutton). N. Baluchistan. M. pyramts var. flavida (Nevill). |W. Baluchistan. M. pyramis var. luteomarginata Persian Baluchistan; S. (Nevill). Persia. ! All the zoological collections made by Dr. Blanford and labelled ‘“ Baluchis- are from this district, in which there happens to be a place called Pishin. tan”’ 18 Records of the Indian Museum. [Vor. XVIII, M. tugrina (Hutton). N. Baluchistan. M. scabra var. elegans (Hutton). N. Baluchistan. Melanopsis deserticola, sp. nov. Persian Baluchistan. PULMONATA. Fam. Limnaeidae. Limnaea persica, Issel. Baluchistan desert; S. Per- sla. Limnaea tvanica, sp. nov. Persian Baluchistan. Limnaea bactriana, Hutton. N. Baluchistan ; Seistan. Limnaea gedrosiana, sp. nov. Do. do. L. gedrosiana var. rectilabrum, nov. Do. do. Limnaea truncatula, Gray. ~ _N,. Baluchistan. Limnaea hordeum, Mousson. Seistan. Fam. Planorbidae. Gyraulus convextusculus (Hutton). N. Baluchistan; Seistan. Gyvaulus euphraticus, Mousson. Do. do. Segmentina calathus (Benson). Seistan. Lamellibranchia. Fam. Unionidae. Lamellidens marginaits, subsp. Seistan ; Afghan desert. vyhadinaeus, subsp. nov. Fam. Cyrenidae, Corbicula fluminalis (Miller). Baluchistan ; Seistan. 2 Pisidium paludosum, Hutton. N. Baluchistan. GEOGRAPHICAL DISTRIBUTION. The molluscs to be considered in this paper come mainly from three districts, (1) the hill-country of Baluchistan watered (so far as it is watered at all) by the Lora or Pishin river, which has more or less saline water and makes its way down to the desert, where it disappears in the Zanginawar lakes ; (2) Seistan, the delta and basin of the Helmand river, which rises in the Hindu Kush a considerable distance to the north of Baluchistan and passes a great body of fresh water, and (3) Persian Baluchistan, a mountainous district lying some considerable distance south ot Seistan and reaching to the Arabian Sea. These three districts are separated by the great Afghan-Baluch desert, through which the Helmand flows. As even the desert is not devoid of aquatic molluses, we have actually four districts to consider. MOLLUSCA OF THE HILI-COUNTRY OF BALUCHISTAN. The molluscs of this district were described seventy years ago by Hutton, and we have no species to add to his list, though we do not accept all his identifications. The forms known from this district are: Melanoides pyramis. Melanoides scabra var. elegans. Melanoides tigrina. ? Amnicola parvula. 1919. ] N. ANNANDALE & B. PrasHaD: Mollusca. 19 Limnaea bactriana. Limnaea truncatula. Limnaea gedrostana. Gyraulus convextusculus. Limnaea gedrostana vat. rectilab- Gyraulus euphraticus. yum. Corbicula fluminalis. 2? Pisidium paludosum. MoLLuUscA OF THE AFGHAN-BALUCH DESERT. ‘The following forms have been found in this desert :-— Melanoides pyramis var. flavida. Gyraulus euphraticus. Vivipara hilmandensis. Corbicula fluminalts. Limnaea persica. Lamellidens marginalis subsp. rhadinaeus. MoLLusca OF SEISTAN. In the alluvial plain of Seistan the following forms occur :— Amntcola (Alocinma) ststanica. Gyraulus convexiusculus. Vivipara hilmandensis. Gyraulus euphraticus. Limnaea bactriana. Segmentina calathus. Limnaea gedrosiana. Corbicula fuminalis. Limnaea gedrosiana vat. rectilab- Lamellidens marginalis subsp. yum. - vhadinaeus. Limnaea hordeum. Mortiusca OF PERSIAN BALUCHISTAN. Most of the molluscs known from this district were obtained by the late Dr. W. T. Blanford, but Mr. W. J. Good has recently added an interesting species. The following is a list of the known forms :— Melanordes pyramis var. flavida. Melanoides scabra. Melanoides pyramis var. luteo- - Melanopsis deserticola. marginata. Limnaea tranica. Of the lists, those of the species of the hill-country of Balu- chistan and of Seistan are probably the most complete. Moreover, these two are very similar and the discrepancies between them are probably more apparent than real. In the Seistan list five (out of eleven) specific names are, indeed, present that are absent from the other, viz. Amnicola sistanica, Vivipara lilmandensis, Limnaea hordeum, Segmentina calathus and Lamellidens marginalis. The Vivipara, however, seems to be essentially a fluviatile species, probably unable to live in any but pure fresh water, and rivers with pure fresh water (or water of auy kind) are proverbially scarce in Baluchistan; the Limnaea, which is also known from Mesopotamia, is exceedingly rare; Segmentina calathus, which is not uncommon at some places in the plains of North-western India, is local in its distribution, and the bivalve, though not yet found in Baluchistan, is common (in a distinct racial form) in the neighbouring country of Afghanistan. Of the species probably common to the two countries this bivalve is the only one in which even subspecific distinction is possible, but several others (e.g. Limnaea gedrosiana) exhibit slight local dfferences. Ammicola sistanica is the only species probably peculiar to Seistan, and 20 Records of the Indian Museum. Vor. XVILE Limnaea hordeum (otherwise known only from lower Mesopo- tamia) the only true western form. The Helmand, which, so far as the aquatic fauna is concerned, is the connecting link and the only highway for aquatic animals between the mountains of Baluchistan and eastern Afghanistan and the Seistan basin (or, indeed, between Seistan and all other countries), has naturally a molluscan fauna identical with that of the basin into which it flows. In the small springs of the desert, the water of most of which is more or less saline, only a few very tolerant species (e.g. Gyraulus euphraticus and Corbicula fluminalis, both of great geographical range) can live, and we found but one form (Melanoides pyramis var. flavida) probably peculiar to situa- tions of the kind. It occurs in Persian Baluchistan as well as the Afghan-Baluch desert. Limnaea persica has only been found as yet in the southern part of the Persian plateau and in the basin of the Lora river in the eastern part of the desert. The molluscan fauna of Persian Baluchistan is still imperfectly known, but the inclusion of a species of Melanopsis indicates the presence of a true western Asiatic element absent from other parts of the area under consideration. The general absence of this western Asiatic element is per- haps the most striking feature of the fauna considered as a whole. Another point to be noted is the absence of certain large and con- spicuous Palaearctic forms (e.g. Limnaea stagnalis) common in Kashmir at altitudes no higher than those of the Quetta district. This, however, does not imply that the fauna is essentially Ori- ental in the strict zoogeographical sense, for conditions of life are clearly inimical to large forms. The fauna is a starved one in which only species of great adaptability can survive. Such mol- luscs as Vivipara hilmandensis and Lamellidens marginalis, though evidently of Oriental origin, have spread into Palaearctic districts on the limits of their range, while the species of Limnaea, though here described as distinct, have a clear Palaearctic facies and most of them are probably descended from Palaearctic rather than true Indian forms, from which they differ considerably in all but one instance. The species of Gyraulus and Segmentina, though they occur in Oriental districts, are closely related to and probably de- rived from European forms. The molluscan fauna of Seistan and Baluchistan, therefore, has little true geographical significance. BIONOMICS. The bionomics of the molluscan fauna of the inland waters of Baluchistan and Seistan are perhaps more interesting than its geographical distribution, but here again characters are negative. The fauna is one that lives habitually in water of abnormal chemical composition, for even potable water in these countries contains far more than its usual allowance of mineral salts (see p. 15, antea). ‘The molluscs have not, however, responded to the 1919. ] N. ANNANDALE & B. PrasHaD: Mollusca. 25 chemical stimulus with the facility sometimes associated with this factor in their environment, and are not particularly plastic or variable or at all exuberant in shell-sculpture. This is probably because other conditions are unfavourable—extremes of tempera- ture, drought and lack of food—and the struggle for existence is too keen. The composition of the water has, indeed, had one effect, physical rather than biological, in preserving such sculp- ture as the shells possess intact from erosion. Probably it has acted indirectly, by discouraging the growth of corroding algae. But, even so, the Limnaeae do not develop the strong longitudi- nal ribs formed on the shells of those that live in saline waters in North America.! The shells of all the species of Gastropods are small and in most cases very thin. They are of perfect form, neither distorted nor abnormal in other respects. Their colours, both of shell and soft parts, are pale. They constitute, in fact, alimited but very normal paludine fauna such as might be found in any temperate region. No peculiar lacustrine species or even well-marked phases have as yet been evolved in the Hamun-i-Helmand. In Seistan the recent geological history of the country has been of such a nature that subfossil shells are extremely abun- dant nearly everywhere except in the central parts of old lake- basins, while owing to the annual floods enormous numbers of quite recent shells are scattered over the country. In the deposits, both recent and historically ancient, examined both in the northern and the southern districts only two species. were found (Limnaea hordeum, evidently a scarce form, and Segmentina cala- thus, a widely distributed but somewhat sporadic one) which were not found living in the country. The absence of Melantidae from these deposits was a noteworthy feature. The commonest species in them at most places were Amnicola sistanica, Limnaea gedro- siana, Gyraulus convexiusculus, G. euphraticus and Corbicula fuminalis. At some spots, evidently those reached with fair regularity by recent floods, Lamellidens marginalis was also present in large numbers, and at one place Vivipara hilmandensis was common. An interesting question in the bionomics of freshwater mol- lusecs in a country like those under consideration is that of hiber- nation and aestivation and their effect on sexual activity. We give evidence below that certain species (mainly those of the genera Melanoides, Corbicula and Lamellidens) burrow into mud or sand either at the approach of winter or on the sinking of the annual floods. Perhaps this is also true of Amnicola ststanica. The Limnaeidae and Planorbidae, however, remain active through- out the winter. In Seistan and northern Baluchistan, and also at certain places in the North West Frontier Province, the eggs of ae Baker, Chicago Academy of Sciences, special publication 3, p. 30 IQI1). 22 Records of the Indian Museum. [VoL. XVIIE, Limnaea were observed in great abundance in November, Decem- ber and January, and in females killed at this time of the year the female part of the reproductive organs was found to be in a state of activity. No individuals were, however, observed paired, and the male part of the hermaphrodite gland seems to be aborted. It is probable, therefore, that Limnaea is protandric in the pecu- liar conditions which exist in Seistan and Baluchistan, that pairing takes place in summer, and that the spermatozoa are stored up for considerable periods. PARASITES AND INCOLAE. The main object of the tour on the collections of which this paper is mainly based was to discover what could be discovered about the distribution of the aquatic molluscs and their trematode parasites in reference to the etiology of the disease Bilharziasis or Schistosomiasis. Living molluscs were examined in the field by Mr. S. W. Kemp, who has been kind enough to supply us with the following information. His examinations were made in Nov- ember and December. The only molluscs in which trematodes were found in water brackish to the taste was Melanoides pyramis var. flavida. Of twenty-five individuals of this form from a small water-course at Saindak in the western part of the Baluchistan desert one was infected by the voung rediae (indeterminate) probably of a Xiph- idocercarta. ‘The water was potable but tasted salt and bitter. Sixty specimens of Limnaea gedrosiana from the reed-beds of the Hamun near Lab-i-Baring were examined and none found infected. Of another sixty of the var. rectilabrum of the same species from a small pool in the desert near Nasratabad only one was parasitized, its parasite being a small cercaria of the family Schistosomatidae. One hundred specimens of L. bactriana (fifty of the long-spired and fifty of the short-spired form) were examined at Nasratabad, from an irrigation channel. Ten contained tre- matode cercariae; in eight of these the parasite was a large monostome, while in one it was a Schistosomatid (apparently the same species found in L. gedvosiana) and in one the infection was a mixed infection of these two parasites. Seventy-four specimens of Gyvaulus convextuscuius from the reed-beds of the Hamun were examined and seven were found infected by Trematodes, two by indeterminate elongate sporocysts, three by Schistosomatid cercariae without eyespots and two by similar fork-tailed cercariae with eyespots. Only two specimens of Gyraulus euphraticus were examined, from the same locality; one was uninfected, while the other contained Schistosomatid cer- cariae without eyespots, apparently of the same species as was found in G. convexiusculus. The Schistosomatid cerearia found in aquatic Pulmonate molluscs in Seistan does not appear to belong to any of the species known to parasitize man. A liver-fluke of the genus Fasciola (s.s.) is common in that country, but its cercariae were 1gIQ. | N. ANNANDALE & B. PRasHAD: Mollusca. 23 not found, probably because their incidence is seasonal. Mr. Kemp has given us a note on this species (f. gigantea) which is ap- pended to our paper. He hopes to describe all other parasites later. Small red nematodes were common in Gyvaulus convexiusculus in the Hamun in December. The Oligochaete worm Chaetogaster was found in abundance at the edge of the mantle and in the branchial chamber of Lim- naea gedrosiana var. rectilabrum in a pool in the desert near Nasrat- abad in the same month. Col. Stephenson has identified the worm as Ch. bengalensis, Annandale, a species common in Northern India. SYSTEMATIC ACCOUNT OF THE FAUNA. Class GASTROPODA, Fam. HYDROBIIDAE. 1915. Paludestrinidae, Preston, Faun. Brit. Ind., Freshw Moll., p. 67, 1919. Paludestrinidae, Godwin-Austen, Rec. Ind. Mus. XVI, p, 209. Genus Amnicola, Gould and Haldeman. 1865. Amnicola, Stimpson, Smiths. Misc. Coll., 201, p. 12. The excellent account of this genus given in the work cited enables us to relegate to their proper genus certain Indian, Burmese and Persian species that have usually been placed in Bithynia. These species, however, differ in some respects from the American forms—sufficiently in our opinion to be regarded as constituting a new subgenus, for which we propose the name :— Subgenus Alocinma, nov. The shell agrees precisely with that of the American A mnicola and the Palaearctic Pseudamnicola, being globose or subglobose or slightly elongate, imperforate or subumbilicate, small, thick and smooth, with swollen whorls and having its mouth oval or ovate with a continuous but not greatly thickened peristome. The animal has a relatively short foot, which projects very little if at all beyond the shell. It is rounded or pointed behind and angulate in front. The snout is long and narrow. ‘The tentacles are hardly longer than the shell, thin and filiform, and bear the eyes, which are small, at their base externally. The edge of the mantle is simple. The penis is large, flattened, lunate in outline and provided with a long and stout lateral process, which projects on the left side from its concave margin almost at right angles. It is situated almost in the middle of the ‘‘ neck.’’ The opercu- lum 1s of large size, and incapable of withdrawal into the shell, thick and calcareous but usually hyaline or subhyaline, distinctly spiral and with a siightly eccentric nucleus, but ornamented round the margin with concentric lines. ; 24 Records of the Indian Museum. [VOL. XVIII, The radula is very like that of Amnicola, s.s., its central tooth being produced at either side and bearing a central process on its disk which projects downwards below its lower margin, as well as two or three lateral basal denticulations on each side. Type-species. Ammicola ststanica, sp. nov. This new subgenus is very closely related to Amnicola, s.s., but differs in its long snout, calcareous oval or ovoid operculum and lunate penis. From Pseudamnicola it differs in its much larger, spiral operculum and in having more than one basal denti- culation on each side of the central tooth of the radula. It is, indeed, a link between the two groups of species, each of which we regard as having subgeneric rank. (B= Woe Fic. 1.—Radular teeth of Amnicola { X 500). A. Teeth of Amnicola (Alocinma) sistanica, sp. nov., from the reed-beds of the Hamun-i-Helmand. B. Teeth of Amnicola (Alocinma) alticola (Annandale), from the Inlé Lake, Southern Shan States. Among the species that must be placed in the new subgenus are the common “‘ Bithynia’’ orcula of Bengal and also Ammnicola alticola (fig. IB) from the Southern Shan States. In the former as well as the latter the operculum is distinctly spiral and both have the other subgeneric characters. Amnicola (Alocinma) sistanica, sp. nov. (Pl. iti, figs. 1-5.) This species is very closely related to Amnicota orcula, but the shell is more hyaline and more globose and has the suture more oblique and more impressed. It may be described as follows :— 1919. ] N. ANNANDALE & B. PrasHap: Mollusca. 25 The shell is small, ovato-conical, short, broad and obese, with five whorls, fairly thick but naturally quite colourless and hyaline, rapidly, however, becoming milky after the death of the animal, polished and ornamented with delicate longitudinal striae on the surface. ‘The suture is impressed and very oblique, so that the spire is much longer in dorsal than in ventral view. The whorls increase rapidly in size and are highly convex; they are flattened outside the suture but rounded externally. The apex is minute but blunt and flattened. The basal whorl in ventral view is at least nearly three times as long as the spire, but these proportions are variable The mouth is large and somewhat expanded, ob- lique, never much longer than broad, ovate, rounded or bluntly pointed above and broadly rounded below. The peristome is con- tinuous and there is a band-like callus on the inner lip. The outer lip is hardly thickened, but distinctly though slightly flattened and very narrowly retroverted; its inner corner is sub- angulate. The umbilicus is rimate and the columellar callus is expanded over it. The columella is straight. Measurements of Shells (in millemetres). A BB” Coppa: F Length af 105 8°27 Oaeemns 38 Breadth 42>, 5:2 Ort Ae ee 2°7 enagrivolumouth ~... 275: 3° 13:7) 32es2 ame Breadthormouth .. “2, =2°6 2:6 273592 I'5 The operculum, which cannot be retracted into the shell, is when fresh of glassy transparency. It has three whorls. The nucleus is situated some distance from the antero-internal border, The shape is ovate, broadly rounded anteriorly and bluntly pointed posteriorly. The radula is of the type normal in Ammnicola, s.s. The cen- tral tooth is small and transverse, produced at each side into a bluntly pointed, narrow process, which points downwards and a little outwards. The disk bears a large quadrate raised area and a series of basal denticulations at each side, remote from both the lateral and the basal margin. In each series there are two distinct denticulations and an obscure, imperfectly developed ex- ternal one. The upper margin of the tooth is slightly concave. The central cusp is enlarged and there are several smaller denticula- tions on each side of it; all are triangular, at least as long as broad at the base, and pointed at the tip. The lateral tooth is broad above but slender towards the base. It bears numerous small sharp cusps, the central one of which is slightly enlarged on the inner lateral. This tooth bears a broad blunt process on its disk and has its inner margin broadly and irregularly emarginate below. The outer marginal is very long and narrow, but broadens some- what at the base. Its cusps are very minute and sharp. The intromittent organ of the male is of relatively large size and distinctly flattened. Its external (right) margin is semi-circu- 26 Records of the Indian Museum. PWor. VILE lar and its internal margin concave in the same degree. The tip is pointed. The internal margin bears near its middle a long straight process terminating in a crateriform sucker-like structure, from the centre of which protrudes an elongate muscular papilla. The main body is smooth, the process obscurely annulate. Before entering the penis the vas deferens is highly convoluted and would be of immense length if unravelled. In the penis it pursues a sinuous course near the concave margin and remains distinct nearly to the aperture at the tip of the organ. The outer part of the penis appears to be glandular internally, but is provided with a well- developed layer of transverse muscle fibres externally. The living animal was thus described in the field :—‘‘ Animal white with black clouding and minute goiden yellow specks on the mantle; a suffusion of black pigment on the snout and tentacles. Tentacles slender, when fully expanded no longer than shell. Eyes small, black, prominent, situated near the base of the tentacles Fic. 2.—Male intromittent organ of Amnicola (Alocinma) sistanica, sp. nov. ( X 20), seen from below. a.p.=accessory appendage: g.=glandular region: v.d.=vas deferens. externally. Snout rather long and narrow, with parallel sides, slightly notched in front. Foot relatively small, rounded in front and bluntly angulate at the anterior corners, bluntly pointed behind.” Type-spectmens. M 11538/2, Zoological Survey of India (Ind. Mus.). Distribution. All over the dry Nazzar or reed-country of nor- thern Seistan shells of this species are common in the soil, as they are also in the debris of floods. We found fresh shells in the larger pools in the reed-beds of the Hamun near Lab-i-Baring and a few living individuals among the algae on the roots of Phragmites in narrow channels in the same beds.. . These living individuals, which were very scarce, were all small and were only found in. protected situations. Very large numbers of the mollusc evidently perish annually with the sinking of the floods and the majority of those that survive probably bur- row into the mud and hibernate in winter. 1919. | N. ANNANDALE & B. PrAsHap: Mollusca. 29, The shell of this species bears a close resemblance, perhaps superficial, to that of the Syrian ‘‘Paludina ’’ badiella as figured by Ktister,! but the mouth is broader, the umbilicus narrower, the whorls less flattened above and pigment entirely absent. There has been much confusion about this Syrian species and we have no means of estimating the true relationship, if any exists, be- tween it and our Seistan mollusc. There is also a resemblance to Pseudamnicola macrostoma from Greece, specimens of which we have examined ; but the operculum of that species is much thinner, of different structure and of small size as compared with the mouth of the shell, which is much smaller than that of the Persian species. ? Amnicola parvula (Hutton). 1850. eG oon parvula, Hutton, Fourn. As. Soc. Bengal (2) XVIII, 5 LOSS) 1876. ee, parvula, Hanley and Theobald, Conch. Ind., p. XVII, pl. cli, figs. 8, 9. We have not seen this species and have no means of ascertain- ing its true generic and subgeneric position. Hutton says that the operculum is horny, and it would appear, therefore, to belong pos- sibly to Amnicola, s.s. The other known species of this subgenus are, however, American. Nevill’s* Bithynia orcula var. parvula is quite distinct and probably a true variety of Amnicola (Alocinma) orcula (| Frauenfeld). Hutton’s species was found in a marshy patch of ground in the Kojak Pass at Chaman (Chammun), now on the Afghan fron- tier of Northern Baluchistan. | ) Family VIVIPARIDAE. Genus Vivipara, Montfort. Vivipara hilmendensis, Kobelt. 190g. Vivipara (dissimilis var.?) hilmendensis, Kobelt, Paludina in Martini and Chemnitz’s Conch-Cab. (ed. Ktister and Kobelt), p- 289, pl. lix, figs. g-12. The complete synonymy of the forms included by Kobelt under the name Vivipara dissimilis has not yet been worked out, and we leave the Helmand form provisionally as a distinct species. We have very little to add to Kobelt’s description except that the natural colour of the shell is dark olivaceous with curious round whitish spots, and that the opercula of the specimens he examined were unnaturally thin owing perhaps to sand erosion. The species was described from the Afghan desert and is 1 Kuster, “ Paludina, Hydrocena and Valvata in Martini and Chemnitz's Conch. Cab. (ed. Schubert and Wagner), p. 62, pl. xi, figs. 25-28 (1852). 2 Nevill, Hand List Moll. Ind. Mus. U1, p. 37 (1885). 28 Records of the Indian Museum. _[Vor. XVIII, evidently rare in most parts of Seistan. Single fairly fresh but empty shells were collected at the edge of pools near Nasratabad and Jellalabad and at that of the Hamun near Lab-i-Baring, while a considerable number of bleached specimens were also observed in the soil of occasionally flooded country near Chilling towards the south of Seistan. It is not improbable, therefore, that V. hilmendensis is common on the banks of the lower Helmand and is essentially a fluviatile species. Tvpe-series. M 5087/1, Zoological Survey of India (Ind. Mus.). Family MELANIIDAE. Melanoides, Olivier. 1807. Melanoides, Olivier, Voy. l’Emp. Ottoman II, p. 40. 1854. Plotia, N. and A. Adams, Gen. Rec. Moll., p. 295. 1874. Plotia+Striatella (? in part), Brot, Melanztaceen in Martini and Chemnitz, Conch. Cab. (ed. Kuster), p. 7. 1897. Melanoides+ Plotia, v. Martens, ‘‘ Suss. und Brackw. Moll.’’, pp. 50, 62 in Weber’s Zool. Ergebn. Niederl.-Ost-Indien 1V. 1898. Neomelanien, P. and F. Sarasin, Sussw. Moll. Celebes, p. 38. 1915. Striatellat+Plotia, Preston, Faun. Brit. Ind., Freshw. Moll., pp. 15, 35: In discussing the species of Melania (s.l.) that occur in Balu- chistan and Seistan we have had to overcome two difficulties, firstly to settle the somewhat complicated specific synonomy, and secondly, to decide what characters should be regarded as of generic importance. So far as the second of these questions is concerned we have followed in the main the classification adopted by the Sarasins in the work cited above. We have, however, regarded the groups that they include under the name Melama as of generic value, believing that by so doing we are following sound lines in estimating such structures as the operculum and radula as of equal value in this family to the sculpture of the shell and the precise form of its mouth. The groups or subgenera Plotia, and Striatclla (== Melanoides) as defined by Brot in his monograph and accepted by Preston in the Fauna of India, fade imperceptibly one into the other, and Brot’s diagnosis of Ploiia is, as we have pointed out elsewhere,'! by no means applicable to all shells even of the type-species. The question of specific identity and nomenclature in the Indian and Persian species of the genus is complicated by imper- fect descriptions, particularly on the part of Troschel and Philippi. In considering the question due but not excessive attention must be given to the locality of specimens and it must be remem- bered that the names Melania pyramis and M. elegans are due to Hutton, and not to Benson, who merely distinguished certain forms by letters, and that, further, Hutton was dealing very largely 1 Rec. Ind. Mus. XIV, p. 147 (1919). IgIQ.] N. ANNANDALE & B. Prasuap: Mollusca. 29 when he first used these names with a collection from what is now British Baluchistan and the Afghan frontier. We have been great- ly helped by an examination of specimens named by Hutton himself. The conclusions to which we have come are these, (1) that three species, one of which has several varieties, have been as yet found in Baluchistan and the extreme south of Persia; (2) that Fic. 3.—Opercula of Melaniidae. Melanoides tuberculata (Miiller), from large artificial reservoir in Hyderabad State. M. pyramis var. flavida (Nevill), from spring in the Afghan-Baluch desert. M. pyramis var. luteomarginata (Nevill), from Persian Baluchistan. M. tigrina (Hutton) from Quetta. Melanopsis deserticola, sp. nov. from a spring in Persian Baluchistan. BOO Ww > the correct names of these species are Melanoides scabra var. elegans (Hutton), M. tigrina (Hutton) and M. pyramis (Hutton) ; (3) that Hutton has included forms of two distinct species under the name M. elegans, and (4) that while one of these forms must be regarded as a variety of M. scabra, the other is a variety of M. pyramis. In our opinion the last is a transitional form be- 30 Records of the Indian Museum. MVOL, X, Curve of outer arc of mouth flattened toa straight line ee ... LL. gedrosiana var. vectilabrum. >. Shell moderately or very thick, of very small size, elongate, with the length of the mouth much less than half the total length. 1. Apex pointed ; whorls of spire moderate .., L. tyuncatula. 2. Apex blunt; whorls of spire very large... L. hordeum. Limnaea persica, Issel. (PI. v, figs. 3-6.) Limnaea auricularia vat. persica, Bourguignat, Issel, Moll. Miss. Ital. Persia (Torino), p. 47. 1883. Limnaea persica, Locard, Arch. Mus. Hist. Nat. Lyon I\I, p. 285. 1865. on Issel gave a very brief description of this form, which he did not figure, merely comparing it with L. auricularia and stating the length and breadth of the shell. Locard had probably seen specimens, as he compared his L. subpersica with it, but gave no further details. It seems to us impossible that the form which Nevill! called Limnaeus lagotis var. persica, Issel (= L. trvanica, nobis) can be identical, as that form has no particular resemblance to L. auricularia. We have before us a series of shells? collected by the late Dr. W. T. Blanford at Saidabad, S.W. of Kerman (the type locality). The measurements of one of these agree almost precisely with those given by Issel. We have also a much larger series from the Balu- chistan desert clearly belonging to the same species but differing slightly. Fully adult shells from the former locality are somewhat broader than Issel’s type, the specimen that agrees with it being not quite full-grown, and we do not know from what kind of l Nevill, Hand List Moll. Ind. Mus. \, p. 237 (1879). 2 Identified by Nevill as L. auricularia, var. (op. cit., p. 238). 42 Records of the Indian Museum. [Vor XV, environment either Philippi’s or Blanford’s specimens came. It is possible, therefore, either that Issel’s type was immature, or that the specimens from the desert are more near the typical form than those which chanced to be collected in the same district as it. We will describe and figure shells from the Kerman district and point out the characters wherein those from the desert (examples of which we also figure) differ from them. NMOBOOE iN EAN CMA OSA Fic. 6.—Radular teeth of Limnaea (x ca. 185). A. Teeth of Limnaea gedrosiana, sp. nov from pond in garden in Quetta. B. Teeth of same species from the reed-beds of the Hamun-i-Helmand. C. Teeth of L. gedrosiana var. rectilabrum, var. nov from siones at edge of Kushdil Khan reservoir, Northern Baluchistan. D. Teeth of same variety from small pool in desert near Nasratabad, Seistan. E. Teeth of Z. bactriana, Hutton, from irrigation channel, Nasratabad, Seistan. EK. ‘Teeth of Z. tyanica from Persian Baluchistan, Form from the Kerman district. The shell is small and thin, of a pale horny colour and (perhaps through age) opaque. The spire is short but prominent, acutely pointed and-slightly oblique as a whole. It is considerably longer on the dorsal than on the ventral aspect of the shell, occupying about 4 of the total length on the former and on the latter nearly 4. The body-whorl is relatively large but not greatly inflated; its outlines are sinuous and it has considerable obliquity. ‘The mouth of the shell is large 1919. | N. ANNANDALE & B. PrasHapD: Mollusca. 43 and oval; its main axis lies nearly parallel’ to but considerably outside that of the shell; it is hardly pointed posteriorly. The outer lip is sharp, broadly and fairly regularly arched; it extends far beyond the body-whorl posteriorly. The antero-internal angle is broadly rounded aud a little expanded, the anterior border slightly flattened in its immediate vicinity. The columella is straight, its callus well-developed and broad but not coarse; it is slightly or not at all folded and is extended over the narrowly timate umbilicus. The sculpture consists of fine, somewhat ir- regular longitudinal ridges with microscopic striae running parallel to them ; the former are often developed only on the body-whorl. Form from the Baluchistan desert. Our specimens from the Baluchistan desert were collected in December, 1918 by Mr. S. W. Kemp at the edge of the Zanginawar lakes, 20 miles east of Nushki. These lakes are a series of small basins which retain the overflow from the Pishin river and contain water that is nearly fresh in winter and supports a luxuriant submerged vegetation. Mr. Kemp found only empty shells. These were abundant on the shore of the lakes one to two feet above the water-level. The shells from Zanginawar only differ from those from the Kerman district in being rather smoother and distinctly narrower and in having the columellar callus narrower and slightly more folded. As may be seen from the figures on plate V, they vary somewhat in outline, and young shells are narrower and have the mouth less expanded than old ones. Unfortunately we have no information about the radula or soft parts of either race of this species. Measurements of Shells (in millimetres). Specimens A-—C are from Saidabad, S.W. of Kerman, S. Persia: specimens D—F from Zanginawar in the Baluchistan desert, 20 miles west of Nushki. Ac Maree tak © eet Der bie EG Length ff ay L323 EI Oe LOO ESO aett Ea29 Breadth am 5, LOTOr Oren a7 a. a0 Length of aperture ia ELE 728 98-Oe 407596 Gi2 = Orn Breadth of aperture £3 70" 75) OF 95,0) Or Aa Op eenoes eaeth of spire (dorsal) '".. 9.295, 2:7) 202i ey Breadth or pase Of Spite 2. “2°50 °2°5.° 2:5" 2:01 2-4 2-6 It seems better, in the lack of anatomical information, not to dogmatize as to the precise relations of this form. There can, how- ever, be little doubt that it belongs to the same group as L. auricu- larva. Limnaea iranica, sp. nov. (Bis vit ice si) 1878. Limnaea lagotis var. persica, Nevill (nec Issel), Hand List Moll. Ind. Mus.. 1, p. 237. Shell. ‘he shell is of small or moderate size, fairly thin, dull on the external surface, of a pale, dull opaque buff colour, orna- 44 Records of the Indian Museum. [VoL. XVIII, mented with very fine longitudinal tidges and striae but without transverse striae. The shape is regularly, narrowly ovate, with the apex sharply pointed. The suture is not impressed and slightly oblique. The whorls, of which there are 44, increase gradually in size. The spire is prominent, but short, slightly oblique as a whole. It occupies a little less than } of the total length in dorsal view and is only a little longer in ventral view. The body-whorl is ovate and not at all swollen, almost bilaterally symmetrical. In dorsal view the internal profile forms for the greater part of its length from the base of the spire a regular arc a little less than a semi-circle but is distinctly constricted in front of the internal anterior angle, which is broadly rounded and slightly produced. The mouth of the shell is long and narrowly pear-shaped, pointed and slightly introverted posteriorly. It runs backwards for about the length of the body-whorl. The outer lip is sharp, not at all thickened internally, broadly and regularly arched but not ex- panded. The columella is straight and slightly folded. Its callus, which is continuous with the outer lip posteriorly, is moderately developed. The anterior margin of the mouth is rounded and projects slightly. The columellar callus completely covers the very narrowly rimate or altogether closed umbilicus. Measurements of shells (in millimetres). Specimens A—C are from Persian Baluchistan (Blanford), specimens D—F from Magas in §S. Persia (also Blanford). jae oven Ore Bend 8 a SIR) wale 2 Length th a TB8iS) 1465. 15°39 14-4 1374 or Maximum breadth ae, ably = Oo Te O11. Ore mga aoe Length of aperture Pa (7: ey As G7 le Gato ete GIGI eye G95) Breadth of aperture PV ORS! POW GER, O72) Oa ager Vength ot spire(dorsal)' 23 0) 2:0 6 3°3u feo tao Breadth ofsbasetol spire). -4°2. 395) 3°7e3-5 ny aan eadG Shells from the two localities only differ in size. Our speci- mens are those examined by Nevill. Radula. We have extracted the radula from one of Blanford’s specimens in which remains of the animal persisted. It is of the same type as that of L. bactriana but differs in several particulars. ‘The base of the central tooth is distinctly bilobed. The three cusps of the inner lateral teeth are long and sharp, but those of the intermediate bicuspid teeth are short and blunt, the inner cusp being broadly truncate. The marginal teeth are somewhat remarkable, having the outer margin coarsely denticulate, one of the denticulations being often of a lobular nature. The dental formula is approximately 12.7.1.7.12. Type-specimen. No. M11545/2, Zoological Survey of India (Ind Mus.) [Persian Baluchistan]. Localities. ‘The only specimens we have seen are those col- lected by Dr. Blanford in Persian Baluchistan and at Magas in Southern Persia. There is a good series from each locality. 1919. | N. ANNANDALE & B. PrasHap: Mollusca. 45 The position of this species is a little doubtful. The shell is in some respects intermediate between that of L. peregra (ovata) and that of L. lagotis. It resembles somewhat that of L. intermedia as figured by Kobelt in the new edition ot Rossmassler’s ‘* Icono- graphie” (pl. 488, No. 2602), but the whorls increase more gradu- ally in size and the mouth is broader. Limnaea bactriana, Hutton. (Plo vores: 1.2 capl. wit, te.)6.) 1850. Lee bactriana, Hutton, Fourn. As. Soc. Bengal XVIII (2), p- 656. The shell is moderately small, thin, fragile, of a pale, dull brownish colour, polished when clean but coated with calcareous algae in most of the specimens examined. The surface is often irregularly decussated and always bears, at any rate on the body- whorl, prominent but narrow longitudinal ridges and correspond- ing striae set close together. No transverse striae can be detected with a lens. The sculpture is often concealed by the calcareous algal coat. The apex is acuminate but not very acute, the spire prominent, but occupying less than 4 of the total length in dorsal view. The suture is impressed and moderately oblique. There are 4 or 44 whorls, which are neither swollen nor shouldered ; those of the spire increase gradually in size and the penultimate whorl is relatively large. The body-whorl is large and of ovoid form; its inner outline is markedly sinuate and somewhat emarginate towards the anterior extremity, but the antero-internal angle is broadly rounded, the outer outline is evenly and not very strong- ly curved. The mouth is large but not expanded, extending backwards for more than # the length of the body-whorl and being less than twice as long as broad; it is of symmetrical ovoid form, pointed posteriorly and with its main axis parallel to that of the shell. The outer lip is sharp and neither introverted nor expanded ; it has a regular and considerable outward curvature and extends forwards considerably beyond the limits of the body-whorl. The peristome is continuous, the callus broad but thin, extending over the narrow but profoundly perforate umbilicus ; the columella is distinctly folded. As in many species of Limnaea the shellis dimorphic. We shall call the two forms (a) and (b) and describe the commoner (@) first. (a) In this form the shell has a comparatively long spire, occupying nearly } of its length in dorsal view. The first 34 whorls increase in size gradually, but the basal whorl of the spire is enlarged, the spiralis by no means uniform and the body-whorl comparatively narrow. (b) The shell differs from that of (@) mainly on account of the fact that there is a distinct change in the direc- tion of the spiral between the ultimate and pen- ultimate whorls. ‘The base of the spire is therefore concealed in the body-whorl so that its visible part 46 Records of the Indian Museum. hVGL. x Vinbk becomes comparativey short (slightly more than 4) of the total length, while the body-whorl being more transverse appears broader and has a more expanded mouth, the posterior extremity of which is situated at.a higher level on the shell. Measurements of Shells (in millimetres). Type (a).. Type (0). Young. Length As TGESEO 16°0 3 Breadth : eet, II‘3 4°6 Length of aperture IOS) TSHOL 4'8 Breadth ot aperture _ Sipe OS 6°3 Ee, Length of spire (dorsal Ciem) 50 3°4 25 Breadth of base of spire (dorsal view) a BRL aoa 22 2'0 The soft parts of the tiving anima! have no noteworthy peculi- arity. The foot and head are pale greenish yellow with minute whitish specks. The lower part of the mantle is black with large rounded yellowish spots; higher up the yellow predominates and the dark pigmentation is reduced to a delicate network. The alimentary canal. ‘The jaw is not strongly developed ; the central piece is narrow and lunate, only its outer or marginal half is fully cornified and of a brown colour. The side pieces are feeble. The buccal mass is large and powerful, deeply rounded in lateral view. The salivary glands are smaller than in some species; they enter the alimentary canal just behind the buccal mass. The radula is broad, its dental formula being approximately 8.7.1.7.18. The central tooth, which is comparatively large, has a distinctly tridentate cusp, which is very asymmetrical and is provided at the tip of the central denticulation with a minute thickened spine or tooth. The base of the tooth is elongate and only a little emarginate proximally. The lateral teeth are broad and tridentate; those on either side of the central tooth have the innermost denticulation subequal to the outermost and the central denticulation long and sharp. The outer lateral or intermediate teeth have two rather blunt cusps; the true marginals have from 3 to at least 7 very short and blunt denticulations, the outermost of which is the broadest. These denticulations are arranged in an almost straight transverse line. The oesophagus is slender and elongate, marked on the sur- face with lines of dark pigment; it forms a well-defined double loop at about half its length between the buccal mass and the. chyle stomach. ‘This structure is short and by no means clearly differentiated externally. The lateral muscular masses of the gizzard are large but quite distinct in the middle line and slightly unequal in size; they extend very little over the true stomach, which is elongate and merges very gradually into the intestine Ig19Q. | N. ANNANDALE & B. PrasHaD: Mollusca. 47 near the point at which the liver-duct enters the alimentary canal. The proximal part of the stomach is somewhat sacculated. The liver is large and the intestine rather stouter than in some species. The genitalia are very like those of L. chlamys,' but the upper part of the male duct above the prostate, the lower part of the same duct and also the svermathecal duct are all considerably longer. Other apparent differences (the larger size of the female accessory glands and the more lobate form of the hermaphrodite gland) are probably due to the state of sexual activity in the specimens examined. The ovarian part of the hermaphrodite gland is particularly well developed. The prostate is spindle- shaped, but very small. Habitat. This species was described from Quetta and was found in considerable abundance in an irrigation-channel leading to the garden of the British Consulate at Nasratabad (Seistan town) and also in pools on the parade-ground at the same place. In the channel the water was usually still, but it was allowed to flow freely every few days. It was always more or less turbid. The bottom was composed of stiff clay and supported a rather scanty growth of Characeae and of a narrow-leafed Potamogeton. In the pools, which had a similar bottom but contained a some- what more luxuriant vegetation consisting chiefly of Zannichellia palustris, the water was extremely foul, being frequented by camels, donkeys and mules. ‘The basins had been excavated in obtaining clay for bricks and the water had probably entered by percolation. Shells were also found subfossil in the banks of old water-channels near Nasratabad. Habits. It is noteworthy that these molluscs, though living in water the surface of which was frequently frozen at the season at which they were observed, were in a state of sexual activity so far as the female organs were concerned. Egg-masses were abun- dant on the water-weeds. The adults seemed to feed chiefly on minute algae growing on the mud. No difference was observed between those individuals from the dirty pools and those from the irrigation channel. Affinities. Until the anatomy of the Asiatic Limnaeidae is better known some doubt must remain as to the affinities of this species. The shell resembles those of the group L. Jagotis, but is more distinctly perforate. The structure of the spire somewhat resembles that of L. lagotis var. subdisjuncta,? Nevill, but the penultimate whorl is relatively large and the structure of the mouth is quite different. Hutton in his original description com- pared the shell to that of the young L. chlamys, Benson, and curi- ously enough, before we recognized the identity of our specimens, we did the same so far as the genitalia were concerned. ‘There can be little doubt, therefore, that a relationship with the Indian ! Annandale and Prashad, Rec. Ind. Mus., XVI, p. 143, fig. 4 (1919). 2 Sci. Res. Yarkand Miss. Mollusca, p. 9 (1886). For figures see Weber, Wiss. Ergeb. Reise. Thian-Schan, Mollusken, pl. i, figs. f- (Ab. Bayer. Ak. Wiss. Math.-phys. Klasse, XX VI, 1913). 48 Records of the Indtan Museum. [Voy. XVIII, species exists, but what precisely that relationship is still remains to be discovered. We have compared our specimens from Seistan with one named by Hutton from the old collection of the Asiatic Society of Bengal This shell is stated to be from Kandahar, but Hutton records the species only from Quetta, which was in Afghanistan when he wrote. It is very possible that the locality is incorrect. The specimen is not mature and agrees very closely with the young shell figured by us on pl. vii. Limnaea gedrosiana, sp. nov. (Pl. vii, figs. 2-4.) 1850. Limnaea peregra, Hutton, Fourn As.Soc Bengal (2) XVIII. p. 655. This species, so far at any rate as the shell is concerned, closely resembles L. tvanica, but differs in that the shell is smaller, thinner, paler in colour, smoother, less regular in outline, with a blunter apex, more oblique spiral and slightly more impressed suture. The colour is a faint greenish yellow, the shell is extremely fragile and when fresh quite transparent. There are 34 or 4 whorls. The spire is twice as long in dorsal as in ventral view, occupying at least 1 of the length of the shell in the former. The curvature of the inner profile of the body-whorl is less convex and not so regular as in L. ivanica, and this whorl as a whole is much less symmetri- cal. The mouth of the shell is ovate, pointed but not retroverted posteriorly. It is almost bilaterally symmetrical and narrower than in L. ivanica; its main axis is parallel to that of the shell. Though relatively as long as in the preceding species, it only ex- tends backwards for about Z the length of the body-whorl. The curvature of the outer lip is afiehily flattened in adult shells. The anterior margin projects considerably beyond the body-whorl. Measurements of Shells (in millimetres). Quetta Lab-i-Baring. Length a2 Slee tO tl hen apr O eC Maximum breadth Me 165356. AO AC ARG Length of aperture oe 202) OF 576 @ CO eat Breadth of apetture ">. 5:4 4°r 3:673°38 3°6 3-3 Length of spire (dorsal)... 2°3 * iCal ge ai ee a | Breadth of base of spites. 3: 29) f OZ 2 oa27 The living animal resembles that of L. bactriana, but the foot is perhaps rather smaller, the tentacles longer and more pointed and the colour pale. The alimentary canal also resembles that of L. bactriana, but the muscular gizzard is more uniformly developed and more com- pact. It can be seen in the living animal through the shell as a globular shining mass. 1919. | N. ANNANDALE & B. PrasHaD: Mollusca. 49 The jaw and radula are of the same type as those of the species already discussed, and closely resemble those of L. bactriana in particular. They exhibit, however, great placticity and individual variability (see figures, p. 42). The marginals never have the pecu- liar form of those of L. ivanica. Genitalia. The genitalia are of the same type as those of L. bactriana, but have one important difference, viz. that whereas in that species the male and female parts of the system are approxi- mately equal in length, in L. gedrosiana the vas deferens is greatly elongated, while the female ducts are short. This difference is not correlated with any difference in the position of the external sexual apertures, which in both species are situated almost on a level, but in L. gedrosiana the male duct is strongly convoluted. Type-specimens. M. 11533/2, Zoological Survey of India (Ind. Mus.). Localities. ‘This species is common, as Hutton noted, in the Pishin district (Chaman) and at Kandahar. It also occurs in abun- dance at Quetta and in the Hamun-i-Helmand in Seistan. Habits. L. gedrostana can apparently live only amidst dense submerged vegetation. Hutton found it in brick tanks at Kanda- har and ina small marsh at Chaman. Our specimens from Quetta were taken among weeds in a pool supplied by water from an underground source in the Residency gardens. The submerged vegetation in this pool was dense and the water, in November and January, perceptibly warmer than the air. In the Hamun this Limnaea occurs mainly amidst algae growing on the roots of Phragmites and also on Potamogeton pectinatus in small pools in the reed-beds. Affinities. L. gedrosiana is apparently related to L. peregra, but differs in the blunter spire, more oblique spiral and longer mouth of its shell. We do not feel justified in uniting the two forms, and the shell differs considerably from that of any of the ‘‘ varieties”’ from Central Asia ascribed to L. peregra by former authors. Variation and plasticity. There is not much individual varia- tion in shells from the same environment, except that correlated with age. Shells from Quetta, however, are larger and a little broader and have distinctly larger mouths than those from the Hamun. Moreover, their spire is distinctly shorter. var. rectilabrum, nov. (Pl. vi, figs. 1-6.) This variety or phase differs from the forma /ypica so far as the shell is concerned mainly in having the outer lip distinctly flattened so that it slopes outwards in a straight line. Its margin is some- times slightly turned inwards towards the aperture. The precise form of the mouth is, however, subject to considerable individual variation (see pl. vi). The radula does not differ more from that of the forma typica than that of the latter varies. The genitalia 50 Records of the Indian Museum. [VoL. XVIII, are practically identical, but the vas deferens differs slightly in proportions. This variety exhibits both individual variation and plasticity in a higher degree than L. gedrosiana, s.s. The shape of the mouth of the shell and of the whole body-whorl differs considerably in individuals from the same environment, while individuals from one environment differ in having a narrower shell than those from another. The two localities from which we have examined fresh specimens are the Kushdil Khan reservoir, situated at an altitude of 5000 {t. in the hill-country of Baluchistan north of Quetta, and a small pool in the desert some miles south of Nasratabad in Seistan. The Kushdil Khan reservoir is a large body of clear shallow water artificiallv confined and liable to dry up in summer. In winter it contains at certain places a fairly dense submerged vegetation consisting of Potamogcton pectinatus, Naias major, ete. The pool near Nasratabad was quite a small one. At the time of our visit it was completely isolated in the desert, but in flood-time is evidently connected with alarge backwater of one of the effluent channels of the Helmand. Its vegetation consisted in December of dead reeds and a scanty growth of broad-leaved Potamogeton. Specimens from Kushdil Khan were much larger and asa rule considerably broader than those from Seistan (cf. figs. I-3, 4-6 on pl. vi). All the shells from each locality belonged, though differing considerably, quite definitely to the variety. At Kushdil Khan dead shells, some with remains of the animal, were collected in flood-drift on the edge of the reservoir, while a few individuals were found adhering to the lower surface of stones near the margins. ‘The molluscs in the desert near Nas- ratabad were attached in large numbers to dead reed-stems and to the droppings of goats, flocks of which watered at the pool. Measurements of Shells (in millimetres). Specimens from A—C are from Kushdil Khan reservoir (Baluchistan), and specimens D—F from a small pool, some miles south of Nasratabad, Seistan. A B IG D E F Length a PS TESST EAR EATS 2 SEP eae ae Maximum breadth a GIO: Oar 828.7 37 OS PEs Length of aperture , MEE. “TO TO°7-- *UOr2'= 6:9" S72 Breadth of aperture SE O45) 70 530 95 5 eee Length of spire (dorsal) 3 B75 394) 28" 250" Saas Breadth of base of spire 3°) S34. Bt5) -32t Soran eae Type-series. M 11534/2, Zoological Survey of India (Ind. Mus.). Limnaea truncatula, Gray. 1850. Limnaea truncatula, Hutton, fourn. As. Soc. Bengal (2) XVIII, p. 656. We have seen no specimens from Baluchistan or Seistan, but have examined a large series from different parts of the Western 1g1Q.] N. ANNANDALE & B. PRASHAD: Mollusca. 51 Himalayas. ‘These specimens provide evidence of much local plas- ticity, some agreeing with European shells, others having the form of the var. longula figured by von Martens in the report (in Rus- sian) on the molluscs in Fedtschenko’s Rezse in Turkestan (vol. 1, pl. ii, fig. 26, 1874). Specimens from some Himalayan localities are much larger than those from others. Hutton states that L. truncatula is common in the marshlands bordering the Helmand at Girishk and also in similar situations at the Kogrick Pass and at Quetta. ‘The fact that it was not found at Quetta in winter is, therefore, interesting. The species seems more susceptible to drought and unfavourable conditions than most of the genus! and probably conceals itself in cold weather. Had it occurred at all commonly in Seistan, however, dead shells would probably have been recovered from the recent deposits examined at the edge of the Hamun and elsewhere. The presence of a liver- fluke of the genus Fasciola in the country does not necessarily imply that this mollusc is the intermediate host of the liver-fluke, for L. truncatula does not occur in North America and yet Fasciola hepatica is prevalent in some districts”; moreover, as Mr. Kemp points out in a note appended to this paper, the Seistan liver-fluke is not identical with the European one. Limnaea hordeum, Mousson. (Pie vite fies 55) 1874. Limnaea hordeum, Mousson, Fourn. Conchyl. XXII, p. 42. The shell is extremely small and rather thick, narrowly elon- gate but blunt at the apex. Our single specimen is bleached white and has a somewhat porcellaneous appearance. There are four whorls but the apical one is very small and projects little. The suture is impressed and very oblique, so that the spire is much shorter in the ventral than in the dorsal view. The third whorl is more than 3 times as long as the second and the body-whorl consider- ably longer than the spire. ‘The mouth is small and rather narrow, almost straight and practically oval, being little contracted and not at all pointed posteriorly. The lip is somewhat expanded and has a thickened appearance due to a blunt ridge running round it a short distance inside the margin. The columella is slightly folded, its callus narrowly expanded over the rimate umbilicus. The callus is joined posteriorly to the outer lip. The surface of the shell is marked with faint longitudinal striae, which are regular and set close together. We have compared our single specimen with two of Mousson’s species from the edge of the river Euphrates and can find no difference. The species was originally described from that river. Our Mesopotamian specimens are from Nasariyeh. We obtained a single dead shell in a drift at the edge of a small pool in the ! Walton, Parasitology X, p. 243 (1917) 2 Ward, Fresh Water Biol. N. America, p. 389 (1918). 52 Records of the Indian Museum. [VoL. XVIII, desert some two miles south of Nasratabad in Seistan. The pool in flood-time is connected with a branch of the Helmand river. The following are the measurements (in millimetres) of our Persian specimen :— Length ue ss Set § Maximum breadth Hs vie 27 Length of aperture ; ne Bee Breadth of aperture ah ao. ea Family PLANORBIDAE. The three species belonging to this family and known from Baluchistan and Seistan are all small and all occur commonly throughout Northern India and the adjacent countries. It is with some reluctance that we feel obliged to recognize the two groups represented by the three species as distinct genera, but they differ so much not only in shell but also in anatomy that no other course seems possible to us. We assign, therefore, two of the species (Planorbis euphraticus, Mousson and P. convexiusculus, Hutton) to the genus Gyraulus, Agassiz, and one (Planorbis calathus, Benson) to the still more distinct genus Segmentina, Fleming. Genus Gyraulus, Agassiz. 1837. Gyraulus, Agassiz, Nouv. Mem. Soc. Helv. 1 (fide Preston, Faun. Brit. Ind. Freshw. Moll., p. 118, 1915). In this genus the shell is small, thin, flat, pale, translucent or transparent, without strong transverse ribs, with or without spiral epidermal cilia, with or without peripheral keel, with few whorls, with'a simple lip, without teeth or partitions on the internal surface, with a dextral spiral. The radula has the central ~ tooth bicuspid and the laterals bi- or tricuspid, the marginals with several sharp cusps. The edge of the mantle is not thickened. The vas deferens is continued distally into a narrow penis, which projects straight into an elongate bulbous chamber or penis-sheath and is armed at its termination with a well-developed horny stylet. Type-spectes. Planorbis albus, Miller (Palaearctic). There has been much confusion about the two species of this genus that occur in Baluchistan and Seistan, chiefly because conchologists have rarely seen specimens from the original locali- ties. The correct names for these species are in our opinion G. convexiusculus (Hutton), of which G. saigonensis, Crosse and Fischer, is a synonym, and G. euphraticus, Mousson, to which Hutton and later Benson applied the preoccupied name Planorbis compressus. Gyraulus convexiusculus (Hutton). 1850. Planorbis convexiusculus, Hutton, Fourn. As. Soc. Bengal (2), XVIII, p. 657. 1864. Planorbis saigonensis, Crosse and Fischer, Fowrn. de Conchyl., XII, p. 362,pl. xii, fig. 7. 1919. ] N. ANNANDALE & B. PrasHapD: Mollusca. 53 1876. Planorbis convexiusculus, Hanley and Theobald, Conch. Ind., p- 48, pl. xcix, figs. 8-10. 1886. Planorbis convexiusculus, Clessin, Die Fam. Limnaeiden in Martini and Chemnitz’s Conch. Cab. (ed. Kiister), p. 127, pl. xvii, fig. 9. 1897. Planorbis compressus, v. Martens, Suss-u. Brackw. Moll. in Weber’s, Zool. Ergebn. Niederl. Ost.-Ind. IV, p. 13, pl. 1, figs. 17-22, pl. xii, figs. 7, 10. 1909. Planorbis saigonensis, Germain, Rec. Ind. Mus., Lil pe rr: 1918. Planorbis saigonensis (?), Annandale, Rec. Jnd. Mus., XIV, p. 112, pl. xi, fig. 12. 1919. Planorbis, convexiusculus, id., Rec. Ind. Mus., XV, p. 166. ee NS Fic. 7.—Shells of Gyraulus from pool in Residency garden, Quetta, Baluchistan. A. Gyraulus euphraticus, Mousson. B. G. convexiusculus, Hutton. This shell is apparently very like that of G. albus, the type- species of the genus, but the whorls are deeper. It rarely exceeds 54 Records of the Indian Museum. [Vo,. XVIII, 5mm. in maximum diameter. The figure of the mouth in Concho- logia Indica is not very good, the upper margin being represented as too much elevated. Some variation, however, exists in this respect. It is undoubtedly to G. convexiusculus and not to Hutton’s P. compressus that Crosse and Fischer’s P. saigonensis belongs. Specimens from Tibet identified by Germain (who had probably had access to the collections described by the latter authors) as P. saigonensis agree closely with shells from Quetta, one of the type-localities of Hutton’s species. The radula has approximately the dental formula I1.9.1.9.11. The two cusps of the central tooth are well developed and sharply pointed. The inner laterals have two stout, sharply pointed cusps, the outer laterals or transitional teeth three. The marginals have from four to six similar but more slender cusps. Von Martens’ figure of the teeth is on too small a scale to show their structure clearly. The genitalia (fig. 5 E, p. 40) belong to Simroth’s ' Typus III and closely resemble his figure of those of Planorbis vortex in general structure. All the ducts are, however, much shorter, the penis- sheath is larger and more elongate and the spermatheca smaller and also more elongate. Planorbis convexiusculus is common with the succeeding species among weeds in water-channels and in pools in the reed- beds in Seistan, also in ponds near Quetta, where it occasionally occurs in large numbers on the muddy bottom of open water- channels. It is almost invariably found with P.saigonensis. Its geographical range extends from Lower Mesopotamia through Eastern Persia, Afghanistan and, Northern India to Upper Burma, French Indo-China, China, Japan, and the Malay Archipelago. Like many aquatic Pulmonates this species rises to the surface of the water in the evening and crawls shell-downwards on the surface film. It is, however, apparently unable to swim actively in this position as G. euphraticus does (post., p. 56). Gyraulus euphraticus, Mousson. 1834. Planorbis compressus, Hutton (nec Michaud), Fourn. As. Soc. Bengal (2) Ill, p. 93. 1850. Planorbis compressus. id., ibid., XVIII, p. 117. 1874. Planorbis (Gyraulus) devians var. euphratica, Mousson, Fourn. de Conchyl. (3) XIV, p. 44. 1918. Planorbis saigonensis, Annandale, Rec. Ind. Mus, XV, p. 166. 1918. Planorbis satgonensis, id., Mem. As. Soc., Bengal V1, p. 304. The shell of this species is so like that of G. convexiusculus, and the two are so frequently found together, that we would have felt inclined to regard them merely as dimorphic forms had it ! Simroth, ‘ Mollusca (Weichtiere) ’’ III, p. 502, fig. 165, pl. xxvi, figs. 4, 6 in Bronn’s Tier-Reich (1912). 1919.] N. ANNANDALE & B. PrasHaD: Mollusca. 55 not been for certain differences in the radula and for the fact that the habits are to some extent distinct. The shell, as one of us has pointed out (op. cit., 1918), differs from that of G. convexiusculus not only, as Hutton noted, in being more compressed and more strongly carinate, and having the lip and whorls of a slightly different shape, but also in being larger, more opaque and more coarsely and irregularly sculptured. The last whorl moreover as a rule deviates from the spiral of the upper whorls. These charac- ters are to some extent variable, but the radula differs in having all the teeth narrower, all the laterals tricuspid and the marginals with smaller cusps. The genitalia have all the ducts longer than those of L. convexiusculus and the spermatheca much larger. Otherwise they are very similar. (9 19 09,99 9 [ho pPrepPperur 7 VA Fic. §.—Radular teeth of Planorbidae. A. Teeth of Gyraulus euphraticus, Mousson, from Quetta (<5 700): B. Teeth of G. convexiusculus, Hutton, from the same locality (x goo). C. Teeth of Segmentina calathus (Benson) from swamp near Gurdaspur, Punjab (very highly magnified). Germain! regards Hutton’s Planorbis compressus as synony- mous with P. saigonensis, Crosse and Fischer, but specimens from Northern India have the sculpture coarser and more irregular, the last whorl more oblique, the mouth larger and more oblique and the inner whorls more concave on the lower surface than is shown of G. saigonensis in Crosse and Fischer’s? original figure, with which specimens agree. Specimens of the carinate form from Quetta, however, agree closely with shells of Mousson’s G. devians var. euphratica recently collected by Captain C. L. Boulenger in Mesopotamia. The species would, therefore, appear to be essen- tially a Palaearctic one, but there has been much confusion as to eee ee ee ee ! Rec. Ind. Mus., Ul, p. 117. 2 Crosse and Fischer, Fourn. de Conchyliologie, XI, p. 362, pl. xitl, fig. 7. 56 Records of the Indian Museum. EVOL. XVIIE, its real distribution. It certainly extends from Mesopotamia to the Kiangsu Province of China. The habits differ from those of G. convexiusculus in that the animal swims actively on the surface in the evening. This we recently observed at Chakradharpur in Chota Nagpur, and we have been able to confirm the observation in Calcutta. While floating shell-downwards like other species of the family with its foot applied to the surface film, it moves forward rapidly ina jerky manner by repeated strokes of its shell in the water. The sole of the foot adheres to the surface-film and the shell is raised almost to a horizontal position with its major axis parallel to but well below the film. It is then rapidly depressed, so that momentarily the axis forms almost a right angle with the surface. After this downward stroke it is rapidly raised again to a hori- zontal position, and the animal is propelled forwards a little ob- liquely. The manoeuvre is frequently repeated, each time with a jerk, leverage being provided by the friction between the sole of the foot and the surface-film. Doubtiess the flattened, carinate form of the shell is of advantage in its use as an oar, and pro- bably all species with this character behave in the same way. The bacterial “ velum’’, noted in the Burmese species G. velifer,' is commonly present in G. euphraticus also. Genus Segmentina, Fleming. 1828. Segmentina, Fleming, Hist. Brit. Anim., p. 270. This genus consists like Gyraulus of small or minute, thin- shelled species, but the whorls are as a rule of a different form, convex above and flattened below and the shell is characterized by the production at intervals on the inner surface of the main whorl of curious opaque, white, transverse teeth or ridges of an enamel-like substance Correlated doubtless with the presence of these is the fact that the edge of the mantle is thickened. The radula differs from that of Gyraulus in having the teeth narrower and with smaller cusps and all the lateral multicuspid. The geni- talia are also of a different type in that the penis, though pro- duced into the penis-sheath, is directed into it from one side and is not provided with a horny stylet. Type-species. Planorbis nitidus, Miiller (Palaearctic). Segmentina calathus (Benson). 1850. Planorbis calathus, Benson, Ann. Mag. Nat. Hist. (2) V, p. 348. 1876. Planorbis calathus, Hanley and Theobald, op. cit., pl. xxix, figs. 4-6. 1918. Planorbis calathus, Annandale, Rec. Ind. Mus. XIV, p. 113. The shell is very much like that of the type-species of the genus (Planorbis nitidus, Miller), to which it apparently bears ! Annandale, Rec. /nd. Mus., X1V, p. 112, pl. xi, figs. 7-11. 1919. | N. ANNANDALE & B. PrAsHaD: Mollusca. 57 much the same relation as Gyraulus convexiusculus does to G. albus. The radula has approximately the formula I3.10.1.10.13. The teeth are rather small. The terminal part of the central tooth is distinctly bilobed. The inner laterals have six small, sharp cusps, the other teeth of the same series five. The inner marginals have six cusps very similar to those of the inner later- als, while the outer marginals have seven cusps, the central cusp being considerably larger than the others. Our material is not well preserved for anatomical investiga- tions, but in a specimen from the Punjab the terminal part of the male duct belongs to Simroth’s Typus II. ‘The blind sac-like appendages shown in his diagram (op. cit., p. 502, fig. 165) are, however, very poorly developed. The species is common in swamps in Northern India, but somewhat local in its distribution. It is recorded from several places in the Himalayas, Bengal, Assam, Burma, Ceylon and Siam. Oneof us recently found it to be common in a swamp near Peshawar on the North-West Frontier of India and also obtained a specimen in a similar situation at Gurdaspur in the Punjab. Two dead and whitened shells were found at the edge of a pool in the desert near Nasratabad, Seistan. The shells are very small, but otherwise fairly typical. Crass LAMELLIBRANCHIA (==PELECYpopa). This class is represented in our collections from Seistan and Baluchistan by large series of specimens of two species, Corbicula fluminalis (Miller) and Lamellidens marginalts (Iramarck). Hutton (Journ. As. Soc. Bengal, XVIII, p. 659: 1850) has described under the name Pisidium paludosum a third species from Chaman, now on the Afghan frontier of Northern Afghanistan, but we have seen no shells of this form. It may be a young Corbicula. Corbicula fluminalis, which also occurs in ponds and streams at Quetta, and the Unzo are both so abundant in Seistan, both recent and subfossil, that their shells are a feature of the country. They are found lying on the surface, wherever the land is occa- sionally flooded, in thousands and the banks of ancient streams and lakes are full of the shells of Corbicula. This is the case at many places now completely desert. Though the shells are fre- quently bleached and sometimes wind-worn and sand-eroded, they are usually in a remarkably perfect condition. Family CYRENIDAE. Genus Corbicula, Megerb. This genus provides many difficult problems in taxonomy, increased by the fact that there is no recent, well-illustrated monograph. Undoubtedly many of the socalled species now generally accepted will have to retire into the synonymy of others. 58 Records of the Indian Museum. [VoL. XVIII, Corbicula fluminalis (Miiller). (Pl. viti, figs. 1-6.) 4. Tellina fluminalis, Miller, Verm. terr. et fluv. Hist. \I, p. 205. 8. Cyrenacor, Lamarck, Anim. sans. Vert. V, p. 552. 4. Cyrena crassula, Mousson, Bellardi’s Cat., p. 54 4. Corbicula cor, Prime, Ann. Lyc. Nat. Hist. N. York, VIII, p 7, fig. 8. 1866. Corbicula ecrassula, id., ibid., p- 216, figs. 44, 45. 1883. Corbicula fluminalis+C. crassula, Locard, Arch. Mus. Hist. Nat. Lyon III, pp. 222, 256, 258, pl. xxi, figs. 17, 18, 25, 26. 1913. Corbicula fluminalis, with var. cor, Germain, Bull. Mus. Hist. Nat. (Paris), p. 472. The species is a very variable and plastic one and has a very wide geographical range in Africa and Asia. It was originally described from the Euphrates. We have examined a large series of fresh and subfossil shells from Seistan, the Afghan desert and Northern Baluchistan. Those from the Afghan desert were col- lected by the Seistan Boundary Commission of 1902-1903. Several small shells were also obtained in a spring of distinctly brackish water at Saindak in the west of the Baluchistan desert. In most of the series comprised in this collection, including those of fresh shells from the Hamun-i-Helmand, the specimens can be separated easily enough into two groups, one agreeing well enough with the majority of sheils from the Euphrates, the other with the breadth proportionately narrower and the umbonal region more prominent. The former form is undoubtedly the true Tellina fluminalis of Miiller, while the latter agrees closely with Prime’s figures of Corbicula cor (amarck). In one large series from the desert on the banks of the Helmand in Afghan territory some of the shells are still narrower and come very near the same author’s figures of Corbicula crassula, Mousson; while in several series shells inter- mediate between C. fluminalis and C. cor, C. cor and C. crassula are readily selected. C. cor and C. crassula may, therefore, be recognized at most as varieties, if this be convenient, but not as distinct species. All the shells we have examined from Mesopotamia, Seistan or Persia are small, and the species in Asia Minor (in which also, however, it is (fide Locard) plastic in size) apparently attains larger dimensions in favourable circumstances. The largest fresh speci- mens we obtained in Seistan is a single valve 27°5 mm. broad by 24°5 mm. high. It exhibits an interesting abnormality in the hinge, in which only one, the central, cardinal tooth is developed. The colour of the periostracum varies from bright green to black ; that of the inner surface is violet. Mesopotamian shells are often decorated with broad, whitish, transverse bars externally. The species commonly hides itself in mud or sand in winter. We found very few living examples in Seistan, in spite of the abundance of fresh shells everywhere. Several living individuals were, however, dug from mud at the bottom of small pools in the Randa stream near Jellalabad, about 12 miles north of Nasratabad, in November. 1919.] N. ANNANDALE & B. PrasHAD: Mollusca. 59 Family UNIONIDAE. Genus Lamellidens, Simpson. 1900. Lamellidens, Simpson, Proc. Nat. Mus.(Washington), XXII, p. 854. 1911. Lamellidens, Ortmann, Nautilus, XXIV, p, 106 1918. Lamellidens, Prashad, Rec. /nd. Mus. XV, p- 145. 1919. Lamellidens, id., ibid., XVI, p. 293, fig. 4. The occurrence of this genus in Seistan proves the existence of a distinct Indian element in the fauna. Lamellidens marginalis, Lamarck. Subsp. rhadinaeus, nov. (Pl.ifi, figs; 9,10; plowilietes..7 Pr.) Shells from Seistan only differ from those of the forma typica from Bengal in a few particulars, but the differences are constant in a large series. In shape the shell is somewhat variable, but is close to that of the var. corrianus, Lea, being more transverse and having the upper margin straighter than that of the forma typica. In this respect it is intermediate between the two varieties. It is slightly more tumid than either and in old shells one or other of the valves is as a rule bent outwards slightly at the point at which the foot emerges, causing the shell to gape at this point. The sculpture of the young shells is finer and sharper than in the com- mon Indian varieties. The teeth of the hinge are more prominent than in either form, the pseudocardinals stouter and the laterals more bent. In their stoutness the teeth approach those of the spe- cies or variety L. jenkinsonianus (Benson) of Bengal and Assam. The older part of the shell is pale cream-colour or silver-grey externally. This usually deepens, as the shell grows, to chestnut, and sometimes even to black. Some shells, however, are pale greenish; there is usually at least one pale yellowish zone on the darker region, and the margin is sometimes pale. The nacre is bluish white and has a china-like lustre with very little iridescence. Two phases can be distinguished, the difference being visible even in very young shells. Phase A.—The shell is of considerable size and thickness, with the growth-lines strongly developed, the epidermis of the lower part dark and the hinge-teeth unusually stout and prominent, approaching those of some species of Unio in development. Phase B.—The shell is smaller, thinner, less inflated and paler in colour and the hinge-teeth are less stout and not so prominent. The young shell seems to be slightly more elongate in phase B than in phase A and is distinctly smoother in the former. It does not differ in essential characters from that of the typical form of the species, but has the sculpture less well developed ; it is rather shorter in proportion than that of the var. corrianus. Its sculp- ture is very well preserved even in old shells, the glochidial shell often remaining as a minute tubercle. When about 25 mm. long the shell has a fairly prominent dorsal “‘ wing” and is thin, almost translucent and of a greyish colour. 60 Records of the Indian Museum, [VoL. XViar Measurements of Shells (in millimetres). Specimens A—C belong to phase A: specimens D—F to phase B. A B C D E F Length 07:84. 968 2790321271 1407 30. 15 ato Height 55.5) 40°9-47 6.304 3078 20g Thickness BTA Boies 3BNe t25 25°3.. 160% Lamellidens marginalis subsp. rhadinaeus is common all over SP eS ad SURG SSE RET SEN. "oe ve, < <5 Pp fOr . LO OE SS yore f Ste, et FG. 9.—Shells Lamellidens marginalis rhadinaeus, subsp. nov. (natural size). A. Dorsal view of type-specimen from a backwater of an effluent of the Helmand near Nasratabad, Seistan. B. Hinges and scars of the same specimen. Seistan and broken shells from the Afghan desert evidently belong to the same form. It appears to be quite distinct from the var. candaharica, with specimens of which we have compared our series. The range of the species, which is a very plastic one, is more 1919. ] N. ANNANDALE & B. PRASsHAD: Mollusca. 61 extensive than Simpson (oP. cit., p.855) stated, extending from the mountains of Afghanistan to Ceylon and from Seistan to Burma, possibly even to Java and South China. Dead shells were found in very large numbers at the following places in Seistan :—at the edge of the Hamun near Lab-i-Baring, in pools in the desert near Nasratabad, in the bed of Randa stream near Jellalabad about twelve miles to the north and in a large backwater of an effluent of the Helmand some miles to the south-east of the capital. They were also observed on the surface and buried in the stiff clay of open plains near Jellalabad which are periodically flooded. Specimens from still water, including the Hamun, belong to phase B. those from water directly connected with larger streams to phase A. All these shells were in a remarkably good state of preserva- tion, and their surface was not at all eroded. In many instances the valves adhered tightly together and the molluscs had every appearance of being alive. Indeed, many of them were brought us as containing the animal by herdsmen at Lab-i-Baring, and the people were evidently surprised when we opened them and found only mud inside. It is probable that the animal burrows deep into the mud at the approach of winter and in this connec- tion the gape in the lower margin of the shell of many old indivi- duals is of considerable interest, indicating that the foot is particu- larly large and powerful. The shells give an indication of the age which the form habi- tually reaches, but this indication is open to two interpretations. On all the larger shells examined three very distinct regions can be distinguished. Round the umbo there is a region about 25 mm. wide by 13 mm. high in alarge shell. The sculpture of this region, though clear-cut, is almost microscopic. It includes the nodulose and sloping ridges characteristic of the species, and also numerous (about 16) concentric longitudinal striae, each of which is com- pound. This region has a smooth appearance as a whole and is always of a pale colour. The next region is a broad band about 60 mm. broad by 24 mm. high in a large shell. Its general appear- ance is similar to the first region, but the epidermis becomes darker towards the lower margin, and it bears about the same number of compound striae. The third or outer region is about 30 mm. deep and occupies the full width of the shell. It has a much rougher appearance than the other two and bears four to six bands or groups of compound striae separated by smooth grooves. We are able to state definitely that the first region represents the growth of more than a year From the situations in which fresh shells are found in winter, where we may confidently believe they occur living in the flood-season, it is clear that the animal commonly undergoes a fairly prolonged period of hibernation, and in all probability it breeds when the floods are at their height in April or May. We found several shells in November that correspond precisely with the first region in the adult shells, and one of them still contained remains of the soft parts. Allowing for the period of free larval 62 Records oj the Indian Museum. [ VoL. s 2 EXPLANATION OF PLATE VI. Limnaea gedrosiana var. rectilabrum, var. nov. Fics. 1-3.—Shells from the Kushdil Khan reservoir, Pishin District, Northern Baluchistan. 4, 6.—Shells from a pool in the desert near Nasratabad, Seistan. Fic. 5.—Young shell from the edge of a backwater of a large water-channel in the same district. +) Plate VI. REC. IND. MUS., VOL. XVIII, 1919. ‘NVLSIHONTVE GNV NVLSISS JO AVGIAYNIIT ‘}9p Aseypmoy OV i a Aint Fa *%. ay » ee Lt, Fx ee bone a ite i a 7 a oie», = © | “5 ee Pe KC Toei ie iy ee re 7 rey IR oe Sf a i J WA By AV: At BAL PENT, want ie fate ih Seda canntntl _ “et “) ARG ° liek anes ot: — er reat Lo . | us e.,. Vol ad sesieicberth) Sant =. Rabi” oy ni bron -£ Le te = a h)) Hotta ar a ie iF “2 7 os if : pele rare hi % i iJ ASU Wik : : Bangi vii PAGS SG EU eters OF OT eT ee ab gs a: ae — 0 ; a) - 1h. Vea : a Ao TOL Re rst 7 peta opel #7 As ie “isk : ; Pe ecl Fa awit: ifn] e So =— Tied) a} Neg © iy v4 i? hid 4 ities enti id SUBUNIT By) Yi! ites nnay= 0 ord . -* pre ., aes EXPLANATION OF PLATE VII. LIMNAEAE OF BALUCHISTAN AND SEISTAN. Limnaea iranica, sp. nov. Fic. I.—Type-specimen from Persian Baluchistan. Limnaea gedrosiana, sp. nov. Fic. 2.—-Shell (type-series) from a pond in the Residency garden, Quetta, Baluchistan. FIGs. 3, 4.—Shells from the reed-beds of the Hamun-i-Helmand near Lab-i-Baring. Limnaea hordeum, Mousson. Fic. 5.—Shell (? subfossil) from the desert near Nasratabad, Seistan. Limnaea bactriana, Hutton. Fic. 6.—Young shell from the desert near Nasratabad. Plate VII. REG. IND. MUS., VOL. XVIII, 1919. ‘NVISIHONTVE GNV NVLSIAS 40 AVGISWNWIT "9p AIBYPMOY ‘ ) Vv EXPLANATION ©OF PLATE Vil: LAMELLIBRANCH MOLLUSCS OF SEISTAN AND THE AFGHAN DESERT. Corbicula fluminalis (Miller). Fics. I1-6.—A series of shells from the edge of the Helmand in the Afghan desert, illustrating variation in form and size. Lamellidens marginalis, subsp. rhadinaeus, subsp. nov. Fic. 7.—Type-specimen (phase A) from the desert near Nas- ratabad, Seistan. Fics. 8-11.—A series of shells of different ages from the Hamun-i-Helmand near Lab-i-Baring. All the shells are shown of the natural size. Rec. IND. Mus., Vou. XVIII, 1919. PLATE VIII. Ss. C. Mondul photo. MOLLUSCA OF THE HELMAND. NOTES ON FISH OF THE CENUS DISCOG- NATHUS FROM INDIA AND PH RSEA, By N. ANNANDALE, D.Sc., F.A.S.B., Director, Zoological Survey of India. (With Plates IX—-XI.) Although it is only a few months since I published notes on the Indian species of this genus,! a large amount of additional material is already available and I have been able to examine living specimens in districts so far apart as Seistan in Eastern Persia and the Nilgiri Hills in Southern India. The latter district is particularly important as it is the type locality for several of the forms described by Jerdon and Day. It is not yet possible to clear up all the difficulties concerning the Indian species and much more material is still necessary before the Assamese and Burmese forms can be adequately discussed, but I hope that these notes may lead to further investigation of the genus, which is a particu- larly interesting one from a biological point of view. In my former notes I neglected to mention the species des- cribed by Tate Regan from the North Western Frontier of India under the name Discognathus wanae,* aud I gave no reference to the Persian form described by Berg as Garra persica.’ The latter is probably a local race of D. lamta, but the former seems to be a very distinct species. Two species from South India have been brought to my notice too late to be discussed in this paper. They will be des- cribed shortly by Mr. C. R. Narayan Rao under the names D. platycephalus and D. bicornutus. ‘These species must be attributed to Mr. Narayan Rao, but I have noted some of their more salient characters in the key to the species printed here. These charac- ters he has demonstrated to me. The genus as a whole falls into two sections, distinguished by the degree of differentiation of the adhesive disk behind the mouth on the ventral surface. Both these sections are found in Africa as well as in Asia and both extend into the Palaearctic part of the latter continent, but the most highly differentiated forms occur mainly in India and Malaysia, in which countries those with the simpler type of disk are practically absent. Among those in which the disk is best developed, a secondary differentiation occurs in 1 Annandale, Rec. Ind. Mus. XVI, pp. 113, 129 (1919). * Tate Regan, Ann. Mag. Nat. Hist. (8) XIII, p. 263, fig. A (1914). 3 Berg, Ann. Mus. Zool. St. Petersbourg XVIII, p. xi (1913). 66 Records of the Indian Museum. [Vor; X Vitae certain species in the naked chest and the specialization of the muscles of the thorax. In the following key I have made use of the first of these characters in my primary division and of the second in the first secondary division. It is possible that the two primary groups will have to be regarded ultimately as distinct genera. Key to the Indian and Persian species of Discognathus. I. Mental disk small, less than half as broad as head, without a specialized posterior border; margins of opercula meeting in the middle line well behind the disk. A. Disk very imperfectly differentiated, without free pee border; barbels 4; chest covered with scales D. adiscus. B. Disk with posterior border free but lateral borders not so: barbels 4; chest covered with scales D. wanae. C. Disk with both lateral and posterior margins ‘free ; 2 bar- bels ; the whole of the ventral surface naked D. phryne. Il. Disk large, more than half as bread as head, with a free lat- eral and posterior granular border, entirely or almost separat- ing the opercular margins. A. Chest covered with scales, not much flattened ; its muscles not highly differentiated. tr. Nobarbels. Zid. 44 .. ese . D. imberbis. 2. 4barbels. J/.1. 36-41. a. Pupil of eye in middle or anterior half of head: eye usually visible from below; snout tuberculate in male but not forming a very distinct process. i. Opercular borders approaching one another at an obtuse angle on the ventral surface, widely separated by mental disk D. lamta. . Opercular borders approaching one another closely at an acute angle behind the mental disk ; disk sub- circular ws, Di potyla: b. Pupil of eye distinctly in posterior half of head ; eye (except in D. gravelyi) invisible from below. . Snout moderately rounded, more or less produced and tuberculate, at any rate in adult male, immediately in front of or outside nostrils. a. Snout forming a single free conical process in front of nostrils in all individuals ; opercular borders ap- proaching one another at an acute angle on the ven- tral surface ; length of head nearly 5 times in total length * .. D.stenorhyn- g. Snout produced into a pair of conical processes, chus. one outside each pair of nostrils a3 . D. bicornu- y. Snout produced in male only; its process single, tus. usually shorter than in a; opercular margins nearly transverse and widely separated on ventral surface ; length of head at least 54 times in total Jength ... D. jerdoni. . Snout bluntly pointed, never produced in iront of or outside nostril ; slightly retroverted at tip. a. Dorsal profile from tip of snout to behind dorsal fin forming an even rather high curve; depth of body 5 times in total length . a .. D. gravelyt. 8. Dorsal profile nearly straight ; that of head sinu- ous ; depth of body (in adult) nearly 7 times in total length ae. D. elegans. . Top of head obliquely flattened ; profile of head and anterior part of body a slanting straight line ... D. platyce- phalus. B. Chest naked, flattened, with specialized muscles. 1919. ] N. ANNANDALE: Fish from India and Persia. 67 1. Pectoral fins hardly longer than head, not nearly reaching ventrals rs Sie = ... D. nasutus. 2. Pectorals distinctly longer than head, nearly reaching ven- trals ee oe A .... D. macrochir. SECTION I.—Gyroup of Discognathus variabilis. 1863. Discognathus (s.s.), Bleeker, Atl. Ichth. II], p. 24. In this group the adhesive apparatus on the ventral surface of the head is comparatively little differentiated, the snout projects little beyond it, and the general facies of the fish is less peculiar, the ventral surface being less flattened and the caudal peduncle more dis- tinct. As Boulenger has pointed out in discussing an African species (D. quadrimaculatus ') belonging to the group, it has a close resemblance to the eastern Asiatic genus Crossochilus, a genus which, according to most recent authors, is not found west or north of the Malay Peninsula; but there is a distinct difference in the structure of the mouth and in particular of the lower lip.» In this section of Discognathus, as also in the other species, the jaws are much less sharp than in Crossochtlus and are never horny or bony, but always cartilaginous and covered with a thin epithelium. The upper lip is joined to the lower lip directly by a frenulum, but the lower lip itself is vestigial, disappearing entirely in most species in the middle of the jaw, though in some, as in D, phryne, it is represented by a delicate fold of integument that may be ex- tended over the whole jaw. The transverse band of tissue which stretches in many species across the anterior margin of the mental disk just behind the lower jaw appears to be quite distinct, as is shown by the condition in D. phryne, from the lower lip. It is this band of tissue, however, which is labelled /.J. in my figure on p. 114, Rec. Ind. Mus., Vol. XVI. The snout does not appear to be produced or tuberculate in any of these species. It is possible that Discognathus is derived from Crossochilus and that the species of the first section are closely related gene- tically to that genus. The geographical difficulty is not so great as might appear at first sight, for the species Cirrhina latia is closely related to Cyrossochilus, in which indeed it is placed by Giinther* and Vinciguerra,* and the range of Cirrhina latia ex- tends from Upper Burma to Baluchistan. The eponymous species of this group is found in Palestine and Mesopotamia. The three discussed here occur in the extreme east of Persia or on the North West Frontier of India, while D. quadri- maculatus is recorded from various lakes and rivers in the upper watershed of the Nile and D. vinciguerrae (which, so far as I can judge from Boulenger’s figure,> also belongs to the group) from 5 Boulenger, Fish. Nile, p. 185, pl. xxxi, fig. 4 (1907), and Cat. Fresh-w. Fish. Africa \, p. 347, fig. 261 (1909). 68 Records of the Indian Museum. fvior. “vay Discognathus adiscus, sp. nov. (PIX Mio NOEs pla iene) L. 1.36—385 9 D532) Sew ese This is a small species of slender habit. The dorsal profile from the tip of the snout to behind the dorsal fin is strongly and evenly arched; that of the caudal peduncle slightly concave. The ventral profile as far as the base of the caudal peduncle is convex. The head is of moderate size, its length contained about 5 to 54 times in the actual total length. It is somewhat flattened above. The snout is rounded and smooth. It is shorter than the part of the head behind the eye. The nostril is situated nearer the tip of the snout than the eye. The eye is large and rather prominent ; its length contained about 34 times in the length of the head, and a little less than twice in the interorbital breadth; it is not visible from below. ‘The pupil is situated near the middle of the head, and the upper margin of the eye near the dorsal profile. The mouth is large and almost straight. It is situated only a short distance behind the tip of the snout. The upper lip is well developed, covering the upper jaw, fringed at the margin and minutely granular. There are 4 short barbels. The lower lip properly so called is only developed at the sides and there is no transverse band of specialized structure behind the lower jaw, which is fully exposed. The mental disk is very imperfectly developed. It consists of a granular pad truncate or subtruncate in front, free at the sides, and more or less emarginate, but not free, posteriorly. The margins of the opercula meet at an acute angle on the mid-ventral line some distance behind it and the branchial openings extend well on to the ventral surface. The dorsal fin is nearly as high as the body. Its last undivided ray is cartilaginous and articulated. The pectorals, which do not nearly reach the ventrals, are rather narrow, pointed and distinctly shorter than the head. The caudal is long, distinctly lobed and with the lobes pointed. The scales are well developed but some- what deciduous. They cover the whole of the body. There are 54 scales between the lateral line and the dorsal, and 4 between the former and the ventral. : The pharyngeal bones and their teeth closely resemble those of D. phryne (v. postea), but the teeth are more slender. The dorsal margin is pale bluish grey ; there is a more or less distinct bluish mid-lateral streak, running from behind the head to the base of the caudal fin. The lower part of the head and sides and the ventral surface are yellowish-white. Type-specimen, No. 2785 F, Z.S.1. (Ind. Mus.). Distribution.—This fish is extremely abundant in small water- courses and pools in the plains of Seistan. We obtained speci- mens from the following localities :—Nasratabad, irrigation chan- nel in Consulate garden; pool in the desert 5 miles south of Nasratabad; pools in stream-bed 12 miles north of Nasratabad ; channels in the reed-beds of the Hamun-i-Helmand near Lab-i- 1910. | N. ANNANDALE: Fish from India and Persia. 69 Baring, and channel leading out of the Hamun 12 miles east of Iab-i-Baring ; small watercourse, Lutak, southern Seistan. Habtts.—D. adiscus is gregarious and always lives in large shoals. In the day-time it stays at the bottom, feeding apparent- ly on algae, but in the evening I have seen shoals swimming on the surface. All the individuals we found in the Hamun, which were not numerous, were dead or dying, and we found enormous numbers in a moribund condition, in which they floated on the surface, in pools of very foul water in a stream-bed north of Nasratabad. We did not find the fish in any but still or slow- running water. I have placed this species in the genus Dzscognathus with some doubt, but I do not know where else to placeit. The mouth differs distinctly from that of both Cirrhina and Crossochilus, though the structure of the gill-openings resembles that found in these genera. Moreover, the mental disk, though poorly developed and differing in shape from that of other species, is present, and the general facies is not unlike that of D. quadrimaculatus. On the whole I think that the species must be accepted as an ex- tremely primitive representative of Discognathus. If this be so, its provenance, together with that of the other primitive species D. variabilis, D. phryne and I). wanae, would suggest that the genus perhaps originated in South Western Asia. It is noteworthy that it does not occur in Central Asia, and possibly the African species that are apparently allied, may be degenerate rather than primitive. Without examininig specimens I cannot express an opinion on this point. Discognathus wanae, Regan.! “Depth of body 4 in the length, length of head 44 to 42. Snout rounded, nearly as long as postorbital part of head; diameter of eye 5 in length of head; interorbital region flat, its width nearly 4 length of head. Width of mouth + length of head; two barbels ‘on each side, shorter than diameter of eye. Upper lip with minute papillae near the margin; lower very narrow; behind it a circular disc divided into a papillose anterior and a smooth pos- terior portion, and with only the posterior edge free. Dorsal III 7 ; origin equidistant from tip of snout and base of caudal; first or second branched ray longest, nearly as long as head. Anal II 5. Pectoral extending 2 of distance from its base to pelvics, which nearly or quite reach vent. Caudal deeply emarginate. Greyish, mottled with darker. “‘ Five specimens, the largest 80 mm. in total length.’’ (Tate Regan). Mr. Regan has kindly informed me that there are about 36 scales in the lateral line and that both back and belly are covered with scales smaller than those on the sides. I have not seen this species, the description of which I quote in full. It was described from Waziristan in the hills of the ! Tate Regan, Ann. Mag. Nat. Hist. (8) XIII, p. 263, fig. A (1914). 70 Records of the Indian Museum. (Vou. XVILE North West Frontier of India and comes from the same geographi- cal district as the species (D. phryne) next to be described. Discognathus phryne, sp. nov. (PI. satis 3.9 pli ie 2.) Di. 3/7." Ae2i5e- T3636. Ie stays 20: This species is apparently allied to D. variablis, Heckel, from which it differs in proportions and in its naked chest and back. Like D. variabilis it has only two barbels. The size is small and the habit rather stout. The length of the head, the greatest depth of the body and the length of the caudal fin are approximately equal and are contained from 4} to 54 times in the complete total length. The dorsal profile is sinuous but nowhere strongly arched, rising in an almost even low curve from the tip of the snout to the anterior margin of the dorsal fin. The abdomen is convex. The snout is blunt and rounded and projects slightly beyond the mouth. Secondary sexual characters were not observed on the heads of specimens captured in winter. The nostrils are large and situated nearer the eye than the tip of the snout. The eye is small, its length being contained 3 to 5 times in the length of the head; it is lateral in position, its upper margin approaching the upper profile, and is situated near the middle of the length of the head. The upper lip is comparatively narrow and indistinctly fringed. The mouth is large and broadly arched. The lower lip is represented by a narrow, linear flap of tissue; posterior to this there isa transverse, minutely tuberculate band, obliquely truncate at either end and much narrowet than the anterior flap; it is about as long as the upper lip. Posterior to this again lies the true mental disk, which is smooth and by no means highly developed. It is somewhat lozenge-shaped in the adult fish and considerably broader than long ; its posterior and lat- eral margins are free. In shape and proportions it is somewhat vari- able. There is a small, blunt barbel at each angle of the mouth; its size is variable and it is sometimes reduced to a mere tubercle. There is no trace of anterior barbels. The dorsal fin starts con- siderably nearer the base of the caudal than the tip of the snout and slightly in front of the ventrals. Its undivided rays are soft and slender and the last, which is almost as long as the head, is articulated in its distal third. The pectorals are a little shorter than the head and do not nearly reach the ventrals when ad- pressed; they are set obliquely on the side of the body. The caudal peduncle is not clearly marked off. The caudal fin is large, distinctly cleft and with the two halves subequal or equal and bluntly pointed. The scales are rather small. ‘There is a relatively broad mid- dorsal streak which is entirely bare and so also are the chest and abdomen. The muscles of the chest, however, are not highly specialized. The lateral scales are deciduous. In the young those 1919. | N. ANNANDALE: Fish from India aud Persia. 71 beneath the lateral line are poorly developed. The lateral line is conspicuous. The pharyngeal bones are broad and very convex. They each bear twelve teeth, but the second tooth of the outer row is very short, though broad, and almost hidden by the others. The formula appears to be 6.3.3/3.3.6, but the teeth are very closely congregated and the rows difficult to distinguish. The teeth are fairly long and slender but shorter than those of D. adiscus, sharply pointed (except the second of the outer row) and slightly retroverted at the tip, which is obliquely truncate. The colouration varies with the environment, but the scales, the upper part of the cheeks, the operculum and the dorsal surface of the head and body are always minutely speckled with black, and the specks are always more numerous on the back and on the top of the head than elsewhere. In individuals from very clear water they are so numerous as to give these regions a blackish colour. Larger black spots are sometimes present on the upper part of the sides, and a narrow blackish vertical bar can usually be distinguished on the distal end of the caudal peduncle. The ven- tral surface and the lower part of the head are white. The iris is speckled like the scales. The fins are colourless. In the young there is a bluish mid-lateral streak running along the body. Type-specimen, No 2787 F., Z.S.I. (Ind. Mus ) (from Seistan). Distribution.—This species is very abundant in the hill country of Baluchistan at altitudes between 5,000 and 6,000 feet. A single specimen was taken by Mr. S. W. Kemp and myself, with many of D. adiscus, in an irrigation channel at Nasratabad, Seistan. Habits.—D. adiscus is gregarious and lives as a rule among algae on the bottom of slow-running water-channels and pools. In the outflow of the Kushdil Khan reservoir in the Pishin district north of Quetta large numbers were observed opposite the places where water flowed in from underground sources. The weather was very cold at the time and this water was warmer than that which came from the reservoir. ‘The fish were feeding on a green filamentous alga. The species seems, as already stated, to be closely related to D. variabilis, Heckel, and is doubtless the one referred to by Zug- mayer! as intermediate between that form and D.lamta. It is almost certainly identical with the D. variabilis mentioned in the editorial note prefixed to Tate Regan’s account of fish from Seistan in Journ. As. Soc. Bengal, (n.s.) I, p. 8 (1906). SEcTION II.—Grvoup of Discognathus lamta. 1838. Platycara, McClelland, Fourn. As. Soc. Bengal, VII, (2), p- 944- This group is certainly more highly developed than that of D. variabilis, The mental disk is always relatively large and is a 1 Zugmayer, Abh. Bayerisch. Ak. Wiss. Math.-phys. Klasse XXVI (6), p- 24(1913). These specimens were from the Pishin River in northern Baluchistan. There are two Pishins in Baluchistan, the one north of Quetta and one, referred 72 Records of the Indian Museum. (Mor. 3< VDEr highly specialized structure consisting essentially of three parts—an anterior transverse band of soft tissue covered with minute tuber- cles, a central almost cartilaginous disk with asmooth surface, and a posterior and lateral free border of soft tuberculated integument. It is therefore a much more efficient organ of adhesion. In all the Indian forms with which I am acquainted the disk completely separates the antero-ventral margins of the opercula, but Gray and Hardwicke in their ‘‘ I/Justrations of Indian Zoology”’ figure these borders in D. gotyla as meeting behind the disk (Vol. I, pl. lxxxviii, fig. 3) and this also appears to be the case in certain African forms. In specimens I assign to D. gotyla the borders nearly meet. In the Indian forms the snout, either in the adult male or in both sexes, is tuberculate and often produced between or outside the nostrils. In the Indian species the number of rays in the dorsal fin and of scales in the lateral line as a rule affords little or no assistance in specific diagnosis. The fish of this group are mostly tropical, but a local race of D. lamta is found as far north as Palestine, while either D. jerdoni kangrae or a closely allied form inhabits mountain streams in the Aden hinterland. In Africa species are found in the Nile valley, in the great African lakes and in the eastern waters of Abyssinia. In Asia the range of the group extends from Palestine to Yunnan, Southern India and Borneo. It seems to have its headquarters in the hill country of Southern India and Assam, but the Assamese species or races have not been investigated since the time of McClelland. In streams at the base of the Nilgiris I found four distinct species. Discognathus lamta, Day'. 1919. Discognathus lamta, Annandale, Rec. Ind. Mus. XVI p. 114, text-fig. 1 and p. 131, pl. i1, figs. 1, 1a. Dr. Chaudhuri has recently taken specimens of this species near Seringapatam in Mysore. ‘They differ slightly from North Indian specimens, but I have not sufficient material to establish their racial identity. Discognathus persicus (Berg.). ‘““Garra persica, Berg, sp. n.? ‘* Discognathus lamta (non Ham. Buch.) Nikolsky, Ann. Mus. Zool. St. Petersbourg, IV, 1899, p. 411 (No. 11706,11707). Do 7,, pase igies sear 3314 11707. River Bampur in Eastern Persia. N. Zarundy 1808, 15-27. VII (6). to by W. T. Blanford in his ‘Zoology and Geology of Eastern Persia,” ir Per- sian Baluchistan some little distance inland from the Arabian Sea. “ Buchanan's Cyprinus (Garra) lamta was probably, from its habitat, identi- cal with McClelland’s Platycara nasuta (1838) rather than with the D. lamta of authors. It is, however, impossible to establish this with absolute certainty. 2 Berg, Ann, Mus. Zool. St. Pétersbourg XVIII, p. Ixi (1913). é 1919. | N. ANNANDALE: Fish from India and Persia. 73 11706. Kiabad in Zirkuh (Eastern Khorassan). N. Zarundy 1898, 3.V (1). “ Near Garva lamta (Ham. Buch.), from which differs in having 7 branched rays in dorsal (in /amta Ham. Buch. 8, as much in crenulata Heck., rufa Heck., obtusa Heck.). Lower lobe of caudal shorter than head. Total length 75 mm. ‘““ Barbels 4, very short, uppers $ diameter of eye, lowers 2. Pupil in the second half of the head. Ventrals below anterior 4 of dorsal. Snout projecting strongly beyond mouth. Upper lip well developed, not fringed. Width of the mouth less than 4 length of head, rather equals the interorbital width. Caudal peduncle 14-1 times as long as deep. Eye supero-lateral, not visible from below. Depth of body 5'0-4'6 in its length (without caudal), head 4°5-4°3. Diameter of eye 4°I-4'0 in the length of head, 1°8 in the interorbital width. Pectorals 5:o-4'8 in the length of body (without caudal). Belly covered with scales. 4-5 round black spots on the dorsal near its base. A dark vertical bar on the caudal peduncle near the base of the caudal. Snout of breeding males with numerous conical horny tubercles. ‘“ Kastern Persia.’’ (Berg). Discognathus jerdoni (Day). GPx, es. 2 pl. x1, fon a). 1849. Gonorhynchus Gotyla, Jerdon (nec Gray), Madras Fourn. Lit. Sez, XV, p. 309. 1867. Garra Ferdoni, Day, Proc. Zool. Soc. London, p. 288. 1878. Discognathus Ferdoni, Day, Fish. Ind. 11, p. 528, pl. cxxu, fig. 6. 1889. Discognathus zerdoni, Day, Faun. Brit. Ind. Fish. 1, p. 247. This species is remarkable for its stout facies and very short, broad head, which differs considerably so far as the shape of the snout is concerned in the two sexes. The dorsal profile behind the snout is very nearly straight in the male and only slightly arched in the region of the dorsal fin in the female.' The snout is rounded and much longer, especially in the male, than the part of the head behind the eye. The nostril is very much nearer the eye than the tip of the snout. ‘The eye is relatively large, especially in the male, in which its length is contained in the length of the head about 5 times and in the interorbital breadth twice. In the female the length of the eye is contained at least 54 times in the length of the head and from 23-3 times in the interorbital breadth. The branchial openings extend on to the ventral surface, but on the sides do not reach much more than half way up the head In the female the snout is smooth and very slightly concave in lateral profile. In the male it is traversed by two semicircular grooves The first of these, which is deep and undercut though narrow, lies a short distance in front of the nostril, while the second is about equidistant from the first and from the tip of the snout. The short projection 1 Day’s figure (Fish. Ind., pl. cxxii, fig. 6) is taken from a badly preserved specimen. 74 Records of the Indian Museum. (VoL. XVIII, caused by the first groove bears several spiny tubercles and there is a short row of smaller horny tubercles on each margin of the second groove. The upper lip is broad, concealing the upper jaw, granular and minutely fringed. There is a narrow semicircular transverse granular band in front of the disk, which is transverse and more strongly arched anteriorly than posteriorly. Behind the disk there is a broader semicircular free border. There are four short tentacles. ‘The opercular margins are almost transverse on the ventral surface. The chest is flattened but scaly and without specialized muscles. The dorsal fin is not so high as the body. Its last undivided ray is moderately stout and it has nine or ten rays in all. The pectorals are broad and expanded and have the outer ray flattened. ‘They are shorter than the head and their base is oblique. The scales are large. There are 3 or 3% above the lateral line and the same number between it and the ventral. The colour varies with the environment. Specimens from the Bhavani River at the base of the Nilgiris are very dark olivaceous on the sides and back and white on the ventral surface. All the fins are greyish but the pectoral fins have white borders. The rays of the caudal are white but the middle third of the mem- brane is blackish. In a specimen from a small muddy stream run-~ ning into the Bhavani the colours are much paler, but there is no dark mid-lateral streak and no spot behind the operculum. The largest specimen I have seen, an adult male from near Mettapolaiyam, is 184 mm. long. The species is common in the Bhavani River near the base of the Nilgiris both before and after the stream leaves its gorge. It lives in places where the stream-bed is rocky and the current strong. Jerdon found it in the Manantoddi as well as the Bhavani and Day records it also from the Wynaad. I have seen a small and probably immature specimen which seems to belong to the species from the Nasik district of the Bombay Presidency. Subsp. kangrae, Prashad. 1878. Discognathus lamta, Day, Fish India II, p. 528 (in part), pl. EXKIN, gS. 1, Ta: 1919. Discognathus kangrae, Prashad, Rec. Ind. Mus. XV1, p. 163, figs. I, Id. This form seems to be no more than a local race of D. jerdont, distinguished by its longer head and smaller eye. Capt. Donald, Warden of Fisheries in the Punjab, has recent- ly presented to the Indian Museum through Dr. Baini Prashad a series of specimens from hill-streams in the Kangra valley. They establish the fact that the fully developed adult male is identical with the form figured by Day in the plate cited. Discognathus stenorhynchus (Jerdon). (Rl: tx ahi ort p Isao): 1848. Gonorhynchus stenorhynchus, Jerdon, Madras Fourn. Lit. Sct. XVI, p. 310. 1919. ] N. ANNANDALE: Fish from India and Persia. 75 This species is closely related to D. jerdoni, but can be dis ~ tinguished at once by the structure of the snout and by its relatively longer head. ‘The snout, im both sexes and at all ages, is greatly pro- duced in front of the nostrils, forming a regular conical forwardly- directed process, which however does not extend as far forward as the actual tip. As in D. jerdoni, this process is formed by a semi- circular groove which passes below it. It bears at its free extrem- ity two rows of spiny tubercles. There is a second groove some distance in front of the first which transforms the actual tip of the snout into a second process, which is directed forwards and slightly upwards and bears a number of small spiny tubercles on its posterior surface. The dorsal fin is rather less high than in D. jerdoni and the two larger unbranched rays are thicker and stouter. A third (anterior) unbranched ray may be present or absent. The pec- toral fins are relatively short and narrow, rounded at the tip and oblique. The chest is not so flat as in D. jerdoni. ‘The scales are rather smaller, though of the same number in the lateral line. There are 33 between the lateral line and the dorsal fin and 24 or 3 between the former and the ventral. The colour of fresh specimens is as follows :—the sides and back pale yellowish above changing to pink below. All the fin- rays are somewhat infuscated and there is a row of dark spots along the base of the dorsal fin. An obscure dark mid-lateral line extends from behind the head on to the caudal fin. ‘The iris is golden yellow. The colours have faded considerably in specimens in spirit. I have examined thirteen specimens, the largest of which is go mm. long. The snout is produced in individuals less than two inches long. Distribution.—vThis fish is only known from the base of the Nilgiri Hills. My specimens were taken in a small muddy stream (the Nierolay) which runs into the gorge of the Bhavani River some I2 miles above Mattapolaiyam in August. Discognathus gotyla (Gray and Hardwicke). Wires eno ri Olax, 1 410), 1832. Cyprinus gotyla, Gray & Hardwicke, ///. /nd. Zool. I, pl. Ixxxviii, figs. 3, 3a. 1867. Garra gotyla, Day, Proc. Zool. Soc. London, p. 288. This little species is also closely related to D. jerdoni, but the eyes are in the middle of the head, the head is large and the structure of the snout is different. The three specimens I have examined are perhaps not fully adult, but the secondary sexual characters are fairly well developed. The length of the head is contained in the total length a little more than five times. The eye is very large, its length being contained a little more than four times in the length of the head. The dorsal profile of the head is convex in the female. In the male there is a short process between the nostrils, bearing several relatively large spiny tubercles. ‘The 76 Records of the Indian Museum. EVOL, <“vadiir anterior semicircular groove on the snout is not strongly developed. The mental disk is subcircular and of very large size. It is com- pletely surrounded by a granular border. The opercular and pre- opercular margins are adherent on the ventral surface. The former approach one another at an acute angle and almost meet behind the disk. The dorsal fin is higher than the depth of the body. The pectorals are large and pointed and have the outer ray somewhat expanded. ‘They are nearly as long as the head. The colour is dark olivaceous with traces of several paler longitudinal streaks on the caudal peduncle. There is a dark spot behind the operculum and a dark median streak on the caudal fin. The ventral surface is pale. Day states that the species grows as long as 5? inches. My specimens are about 50 mm. long. Distribution.—Day states that the species is abundant at the base of the hills in the Bhavani River but rarer in the Sigur. I took two males and a female in the Nierolay stream at the base of the Nilgiris in August, with a number of specimens of D. stenor- hynchus and one of D. jerdont. Discognathus elegans, sp. nov. (Pl ixene. 42 pl. xk figs15): This species is distinguished by its elongate form and by the structure of the snout, which bears numerous patches of horny tubercles but is not produced between or outside the nostrils. The dorsal and ventral profiles of the body and the tail are nearly parallel and the depth of the body is contained more than six times in the total length. The head is short and rather narrrow, its length being contained nearly six times in the total length. The snout is more than twice as long as the part of the head behind the eye. Its dorsal profile is concave, the posterior transverse groove found in certain other species between the nostrils being represented by a broad depression. There is a narrow anterior transverse groove, which extends backwards on the sides of the head nearly as far as the anterior margin of the eye. ‘There are about seven groups of horny tubercles on the snout, but some of them may coalesce or be subdivided. The nostril is very much nearer the eye than the tip of the snout. The eye is of moderate size, invisible from below. The snout is rounded in ventral view. The upper lip, which is fringed and granular, is relatively small, exposing both jaws. There are four very small barbels, those at the angle of the mouth being almost vestigial. The mental disk is transverse and lens- shaped. ‘There is a narrow band of granular tissue in front of it and it is surrounded on three sides by a broad granular free border, which is slightly emarginate near the angle of the jaw on each side. The opercular borders are practically transverse on the ventral surface and are widely separated. The chest, and indeed the whole ventral surface, is flat but scaly and without specialized muscles. ‘The dorsal fin is higher than the depth of the body; its 1919. | N. ANNANDALE: Fish from India and Persia. Gg undivided rays are weak. The pectorals are as long or very near- ly as long as the head, pointed and not much expanded. The ventrals, the anal and the two lobes of the caudal are also elon- gate and pointed. The scales are rather small. There are 4} between the lateral line and the dorsal and 34 between the former and the ventral. ‘The number in the lateral line is the same as in other Indian species of the group. The colour is dark olivaceous without definite markings. The ventral surface is yellowish-white The fins are infuscated, but the paired ones have a pale border. This is the largest species of the genus with which I am acquainted. The type-specimen is 216 mm. long and the local fishermen state that individuals one cubit long are sometimes captured. Type-specimen, No. 2725 F, Z.S.I. (Ind. Mus.). Distribution.—1 have seen this species only from the gorge of the Bhavani River at the base of the Nilgiris, where it was taken with D. jervdont in August. I have examined six specimens. D. elegans is related to D. gravely: from Burma, but the form is more elongate, the snout is tuberculate and there are considerable differences in the structure of the mental disk. From D. platy- cephalus, Rao it is distinguished by its more elongate form and more convex head. D. ceylonensis (Bleeker)! seems to be an allied species. ADDENDUM. The True Cyprinus lamta of Buchanan. Dr. B. L. Chaudhuri has kindly drawn my attention to a quotation from Buchanan’s manuscript notes which casts some light, in conjunction with the same author’s original figure of ‘‘Cyprinus godiyari,’’ on the identity of his Cyprinus lamta. This quotation will be found on page 81 of Day’s volume on the fisheries and botany of Bengal in Hunter’s Statistical Account of Bengal (1876). He says, quoting Buchanan, ‘‘ the Godiydéri is another small Cyprinus found in the same places,” 7z.e. in small streams among rocks in the Bhagalpur district ; while in a foot- note to the name Godtyarzt he adds, apparently on his own authority, “Cyprinus lamta, Ham. Buch. Fish. Ganges, p. 343, and MS. drawings No. 105, as Cyprinus godiyart.”’ Buchanan’s MS. drawing No. 105 is still in the possession of the Asiatic Society of Bengal. It comprises three figures, one a finished coloured drawing* of the whole fish, the others outlines of the dorsal view and of the ventral surface of the head. These figures represent a species unknown to me but apparently allied to 1 Garra (Garra) ceylonensis, Bleeker, Versl. en Meded. Afd. Natuurk. XV, p- 239 (1863). 2 McClelland gives a rather poor reproduction (sufficiently accurate in essen- tials) of this drawing in Asiatic Researches XIX, pl. xliii, fig. 2 (18309). 78 Records of the Indian Museum. [VoL. XVIII, t919.] D. nasutus. Possibly it is D. macrochiy (McClelland), but the figure of the head is unfinished and shows very little detail. The figure is labelled ‘‘ Cyprinus godyari’’ in Buchanan’s handwrit- ing. ‘The species figured is not the D. lamta of Day’s Fishes of India and of subsequent authors. If we are to accept Day’s identifica- tion on this occasion, the D. lamta of his later works will have to receive some other name; but the only point in favour of this is the fact that Buchanan himself was of the opinion that the Lamta of the Gorakhpur district was identical with the Godiyari of the Bhagalpur district (of. cit., p.103). Considering the universal confusion of species that has followed, it is by no means improbable that Buchanan himself did not distinguish them clearly and that his Cyprinus lamta was, as I have suggested else- where, a composite group rather than a single species. It must be remembered that Day, who had himself collected different forms of Discognathus in the Bhavani River (where at least four quite distinct species occur), failed in the end to recognize their diversity. The only way in which the point can be settled is by a thorough ichthyological survey of the small streams of the Bhagal- pur and Gorakhpur districts. July atst, 1919. vege x rey ipt ales i : ; Trey i ose) ae iG Lalit Ws subieiqgiee onl aetna 6 eee Bie: HIG, Pie: EXPLANATION OF PLATE ie Species of Discognathus from South India. Discognathus jerdoni (Day). 1.—Adult male from the Bhavani River near Mattapolaiyam, Coimbatore District. Actual length 184 mm. 2.—Adult female from the same locality. Slightly smaller than male. Discognathus stenorhynchus (Jerdon). 3.—Adult male from small stream running into the Bhavani gorge at the base of the Nilgiris. Actual length 90 mm. Discognathus elegans, sp. nov. 4.—Type-specimen (adult male) from the Bhavani gorge Actual length 216 mm. Plate IX, REC. IND. MUS., VOL. XVIl,1919. A. Chowdhary del. DISCOGNATHUS FROM SOUTH INDIA. HOT hy Sk eaalg estiid Sak Fe) SOF EXPLANATION OF PATE. Xe Species of Discognathus trom South India and Eastern Persia. Discognathus gotyla (Gray and Hardwicke). Fic. 1.—Young male from small stream running into the Bhavani gorge at the baseofthe Nilgiris. Actual length 50mm. Discognathus adiscus, sp. nov. Fic. 2.—One of the type-specimens from Seistan (X 14). Discognathus phryne, sp. nov. Fic. 3.—Type-specimen from Seistan (xX I4)- Plare X. REC. IND. MUS., VOL. XVIJI,1919. “i ? ~ rs 0 enon A. Chowdhary del. DISCOGNATHUS FROM INDIA & PERSIA. Le at: 7 EXPLANATION OF PLATE, XI- Mental disk and pectoral fins of Indian and Persian species of Discognathus. Fic. 1.—D., adiscus, sp. nov. 2.--D. phryne, sp. nov. 3.—D. jerdoni (Day). Adult male. 4.—D. stenorhynchus (Jerdon). Adult male. 5.—D. elegans, sp. nov. Adult male. 6.—D. gotyla (Gray & Hardwicke). Young male. +) REC. IND. MUS., VOL. XVIII,1919. Plate XI. A Chowdhary del, DISCOGNATHUS FROM SOUTH INDIA & PERSIA. NOTES ON’ODONATA COLE Ca Bay any SBRISTAN AND BALUCHISTAN IN WINTER. By Major F. C. Fraser, 1.M.S. LIBELLULINAE. Genus Orthetrum. O. taeniolatum, Kirby. Two males taken at Saindak near the Persian frontier, W. Baluchistan, 17-11-18. ae [Dragon-flies I believe to have be. longed to this species were also seen near the Hamun-i-Helmand in December. JN. A.]. One specimen is normal, the other departs from the type by having only one row of cells between Rs and Rs in all wings. Both specimens are fully adult and pruinosed. In the collection are two specimens of Libelluline larvae, taken in the Residency garden, Quetta, 10-11-18. These belong to the genus Orthetrum and are probably taentolatum. They closely resemble larvae of the latter taken in the Deccan, where they are common in shallow streams, concealed beneath curtains of Spirogyra and other water-weeds. The eyes are prominent and strongly angulated outwards; the body is depressed and squat and the dorsal carina of each segment, except the roth, ends with a sharp, robust spine. The whole body is hairy, this coat serving to pick up debris and flocculent detritus during life, which serves both for concealment from its foes and as an ambush for its prey. [The larvae were taken among dense weeds in a pond fed by an underground source of water distinctly warmer than the air at the time. WN. A.]. 4 AESCHNINAE. Genus Hemianax. Three Aeschnid larvae from Hanna Stream, ca. 6,000 ft., near : 8 Quetta, Baluchistan, 10-11-18. 4;. None are fully grown so that it is impossible to tell to what the insects belong by a study of the tracheation, but I think that there is no doubt that these are the larvae of Hemianax ephippiger, Selys, a common insect around Quetta. [These larvae are from a small stream with a luxuriant growth of Characeae, amongst which they were concealed. JN. A.] 80 Records of the Indian Museum. [Voy. XVIII, AGRIONINAE. Genus Ischnura. I, elegans, Van der Lind, —= rtd ‘ —_ Three females, all differing in their colour scheme; this due partly to a teneral condition and partly due to the polychroism customary in this species. I have compared these with a series from Mesopotamia where the insect is very common and find that teneral forms are usually of a bright orange colour, especially as to the thorax. The orange pigment is soon absorbed and replaced by a greenish-yellow. Progressively with the absorb- tion of the yellow pigment, blue is laid down, so that a series of forms is met with, passing from orange and yellow, through green to blue. Pari-passu with this, black pigment is deposited until it largely obscures the ground colour. Thus the eye-spots are often absent in the very early stage, being replaced by a broad orange fascia which soon changes through yellow and green, to blue, the change beginning from the frout and extending backwards. At the same time, the black fascia which crosses the vertex, extends backwards and gradually laps round the area which is eventually to form the eye-spot. Evidence of this may actually be seen in the specimens quoted. The humeral fascia, usually found in this species, is unenclosed in all three specimens, but two small, black spots on the sides indicate the genesis of a posthumeral stripe. In one specimen, the second abdominal segment bears a some- what quadrate, black spot on the dorsum which is absent in the other two. ‘This specimen is a bright orange colour and has the eye-spots fully developed. ‘There is no doubt that some speci- mens retain the original orange colouring throughout imaginal life. The other two specimens are orange and blue respectively but have no eye-spots nor the quadrate spot on the second ab- dominalsegment. All other mark- ings are the same as the first specimen. In my Mesopotamian specimens, a regular series gradu- ating from the one to the other may be seen so that there is no doubt that the insects are identi- cal. It is possible to divide up a number of Agrionine larvae in Fic. 1.—Mask of larva of Ischnura this small collection into two elegans * species. (The age of the larvae ; varies somewhat widely). One of these forms closely resembles the larvae of Ischnura senegalensis but it is probably the larvae of J. elegans. ‘The mask (fig. 1) is long 1919. ] F.C. Fraser: Odonata from Seistan, etc. 8I and flat; the anterior border projects well forward between the palps and is furnished along its free border with minute teeth. The opposed borders of the palps are similarly armed and each is furnished with a strong movable hook and five long, stout setae, directed inward. In addition, each has a robust spine at its end and a smaller one between this and the movable hook. The caudal lamellae are more or less lanceolate and acuminated at the distal end. ‘The tracheal ramification is pigmented and the lateral borders of the lamellae spined, the spines being robust on a little less than the basal half, and rather minute for rather more than the distal half. At the junction of these two different kinds of spines can be seen, in some specimens at least, a very faint suggestion of a transverse line, the only sign of the nodate charac- ter of the lamellae. The ends of the tibiae are richly tufted with spinous hairs and spines. These larvae were taken along with the two Libelluline larvae mentioned above, in the Residency garden, Quetta. One imago was taken at Lab-i-Baring, Seistan, ‘‘ lying on the surface of the water in channel in reed-beds in the Hamun,” 10-12-18, and the other two at Kirtaka, W. Baluchistan, near Afghanistan frontier, ‘““among coarse grass growing on sand round a small spring.” The females of Jschnura commonly lie up among coarse herbage, in the neighbourhood of water, whilst the males rarely stray far from the neighbourhood of water. INDETERMINATE LARVAE. The other larvae (from the rice-beds of the Hamun-i-Hel- mand near T,ab-1-Baring) com- bine some of the features of an Agrionine larva with those of a Lestine. The middle-lobe (fig. 2) is typically Agrionine ; there is no suggestion of a middle notch as seen in the other family but the palps are highly differen- tiated and the movable hook is of great length. There is a stout hook on both sides of the movable one and the space between the outer hook and this is deeply serrated. The border of the inner hook is lamellated, its edge being fur- nished with a row of closely-set, teeth-like processes, similar to F1G. 2.—Mask of indeterminate larva rows on the border of the middle from the Hamun-i-Helmand. lobe and outer border of the palp. The palps appear to be more or less cupped by the lamellated extension. 82 Records of the Indian Museum. [Voy. XVIII, r919.] The anal appendages are also of unusual character and ob- scurely nodate. They are lanceolate in shape and doubled on themselves like a half-opened leaf. The outer border and midrib are spined for about two-thirds of the proximal en. und the distal portion is deeply pigmented in its outer half, in fact almost black in some specimens. The specimens are young so that nothing is learnt from the tracheation. The larvae of dragonflies usually hibernate during the winter, and it is surprising to find that so many were taken in an active condition during two of the coldest months of the year. [The temperature was as a rule well below freezing-point at night at the time they were captured. The water of the streams and pools at Quetta in which the larvae were found was, however, dis- tinctly warmer than the air, while the channels and pools in the Haniun-it-Helmand were protected from wind and frost by the reeds. IN: A]. a SPONGES, HYDROZOA AND POLYZOA OF SEISTAN. By N. ANNANDALE, D.Sc., F.A.S.B., Director, Zoological Survey of India. (Plate XXII). Specimens of eight species of the groups discussed in this paper were collected in Seistan—three sponges, one Hydrozoon, and four Polyzoa. While the sponges and the Hydrozoon are widely-distributed species, all of which also occur within the limits of the Indian Empire, two of the four Polyzoa are new to science, though related to Indian species; one of the other two is characteristically Indian and Eastern Asiatic, the other cosmo- politan as a species. The three sponges are Spongilla alba, Spongilla cartert and Ephydatia fluviatilis. The range of the first extends from Egypt to Bengal, of the second from Hungary to Mauritius and the Malay Archipelago, while the third is cosmopolitan in non-tropical coun- tries. The specimens of S. alba are sufficiently distinct to be made the types of a new variety; of S. cartert only gemmules, which do not differ from those of Indian sponges, were obtained ; but the Ephydatia, while differing in certain respects from Euro- pean forms, is not definitely enough different for nominal distinc- tion. It is, however, quite distinct from the two Indian varieties or races, himalayensis from the Western Himalayas (which should perhaps be united with syriaca, Topsent) and intha from the Shan States of Burma. The Hydrozoon is Hydra vulgaris, a cosmopolitan species not uncommon in India, The four Polyzoa are somewhat remarkable forms. They are Fredericella sultana var. jordanica, Plumatella (Afrindella) persica, sp. nov., Plumatella (Hyalinella) bigemmis, sp. nov. and Lopho- podella cartert. The Fredericella is a race of a cosmopolitan species the known range of which includes the Volga and Jordan systems; one Plumatella is closely related to a Gangetic species (P. testudinicola) associated, unlike its Persian relative, with fresh- water tortoises; the other differs from the cosmopolitan P. punc- tata in one important structural and physiological character, while the Lophopodella is a characteristic Indian species with a local race in China and Japan and related to Tropical African forms. The representatives in Seistan of the three groups have, therefore, mixed geographical relationships, partly Indian, partly 84 Records of the Indian Museum. [Voy. XVIII, southern Palaearctic, partly cosmopolitan. ‘The Indian element is strongest, or at any rate most conspicuous, in the Polyzoa. It is remarkable to find these groups (or at any rate the spon- ges and Polyzoa) so well represented in a country that seems in almost every respect unsuitable for them. In the Inlé lake- system in the Shan States,! a district apparently in all respects favourable to such organisms, only three sponges and two Polyzoa, both of which belonged to the same genus, were found ; whereas in the Hamun system, in which the water is of extremely variable composition and amount, in which extremes of climate occur in regular succession, the same number of sponges and twice as many Polyzoa (which belonged, moreover, to three genera) were ob- tained. It might seem at first sight that it was necessary for gem- mules and statoblasts to undergo desiccation, of which there is the greatest possible chance in Seistan, just as it is necessary for the eggs of many “ Phyllopod’’ Crustacea; but against this theory must be placed the richness of the fauna of these groups in the comparatively equable conditions of Lower Bengal. We are still far from understanding the factors that encourage growth and reproduction in the lower aquatic invertebrates, and the only possible way to gain light is to keep careful records of the modes of occurrence of the living animals and of the provenance of specimens. The Hamun is the seventh large Asiatic lake that I have had an opportunity of examining in the last seven years (as well as innu- merable smaller bodies of water), and in each place I have paid particular attention to the sponges and Polyzoa; but I must con- fess myself still as far as ever from understanding many of the fundamental factors in the biology of these groups. The lakes have been of diverse kinds and situated in diverse countries—Lake Biwa in Japan, the Tai Hu in China, the Talé Sap in Siam, the Inlé Lake in Burma, the Chilka Lake in India, the Hamun in Persia and the Take of Tiberias in Palestine. But they are not sufficient. PORIFERA, Of three sponges collected in Seistan in winter, one (Ephy- datia fluviatilis) was found in an active state; of one ofthe others only dried specimens were found, and of the third only gemmules. The Ephydatia is interesting because it occurred in the Hamun-i- Helmand in two phases each correlated with a different type of environment. I take this opportunity to describe’a new variety of Spongilla lacustris from Mesopotamia. Spongilla alba, Carter. 1915. Spongilla alba, Annandale, Mem. Ind. Mus. V, pp. 25-32, figs. 1, 2 pl. play, figs. a2 pl. wv, hie: a. ' Annandale, Rec, Jnd. Mus, X1V, p. 75 (1918). 1919. ] N. ANNANDALE: Sponges, Hydrozoa and Polyzoa. 85 var. rhadinaea, nov. The chief diagnostic character of this variety lies in the shape of the skeleton-spicules, a large proportion of which are bluntly pointed. This I have not seen in any Indian specimen. ‘The flesh-spicules, which are scattered singly among the interstices of the skeleton, are very thin and vary in length; they taper to the extremities and have their spines, which are extremely minute, congregated in the central region. The gemmule-spicules have all their spines straight. The sponge is compact but friable, containing little chitinoid substance. The external membrane Léa, NOV., X 250. Fic. 1.—Spicules of Spongilla alba var. viiad LH has disappeared from my specimens, which coated the stems of reeds in a layer not more than 1 cm. thick. They contain abun- dant gemmules of a perfectly normal character, but rather small and bleached white. ; Measurements of spicules, etc. Length of skeleton- spicules . 0°402-0°414 mm. Greatest diameter of skeleton spicules 0°0123-0°025 mm. Length of flesh-spicules .. .. 0°'0826-0°135 mm. Length of gemmule-spicules .. 0°082-0°094 mm. Diameter of ee Bs 0°425-0°5I mm, Type-specimen.—P. %*, Z.S.1. (ind. M us.). 86 Records of the Indian Museum. [VoL. XVIII, Locality, etc.—This sponge was found in abundance in the dry Naizar or reed-country round the Hamun-i-Helmand in Decem- ber, 1918. Gemmules were also observed in drift near Nasratabad with those of S. carteri. ‘The sponge grows on the stems and roots of reeds in country desiccated for a considerable part of each year. No living examples were observed in winter. [Spongilla lacustris var. ineptorum, nov.] Fragments of sponge from the edge of a creek running into the Tigris at Baghdad must be assigned, on account of their yellowish colour in a dry condition, to S. lacustris rather than S. alba, but they represent a very distinct new variety, for which I propose the name ineptorum. When fresh they were evidently green. ‘Their skeletal support is fragile and all the elements in the skeleton feebly developed. The skeleton-spicules are very thin, resembling those of the var. montana, Potts,! a variety which ee = EE ER) Se Ee ie Fic. 2.—Spicules of Spongilla lacustris var. ineptorum, nov. X 250. lives at high altitudes. At the nodes of the skeleton, however, there are dense masses of microscleres, most of which are covered somewhat sparsely with rather stout spines. The spines at the extremities are retroverted. These microscleres are indistinguish- able from those of the gemmules. Others also occur, however, more sparingly in which the spines are all quite straight and the ends more pointed. The two types of flesh-spicules are found to- gether. The gemmules are normal, with the pneumatic wall well developed and the spicules abundant and arranged in the usual manner. Measurements of spicules, etc. Length of skeleton-spicules .. 0°348-0°373 mm. Greatest diameter of skeleton-spicules 0°0082-0'0123 mm. Length of flesh and gemmule-spicules 0°082-0°095 mm. ‘Diameter of gemmules .. 0°394-0°476 mm. 1 Potts, Proc. Nat. Sci, Philadelphia, 1887, p. 192, pl. 6. 1919. | N. ANNANDALE: Sfonges, Hydrozoa and Polyzoa. 87 lype-specimen. P. % Z.S.1. (Ind. Mus.). These specimens were collected in December, 1918 by Bomba- dier R. Hodgart of the Anglo-Indian Battery (Zoological Collector in the Z.S.I.) and presented by him to the Zoological Survey of India. Spongilla (Eunapius) carteri, Carter. 1911. Spongillacartert, Annandale, Faun. Brit. Ind., Freshw. Sponges, etc., p: 87, fig. 14. Gemmules, which do not differ from those of Indian specimens, were found among drift at the edge of a pool in the desert near Nasratabad, Seistan, in December, 1918. The pool in flood-time is connected with an effluent of the Helmand. This sponge is by far the commonest species in the plains of India. It has also been found in Hungary, Mauritius and several of the Malay islands. Its occurrence so far west in Asia as Eastern Persia is interesting in view of the fact that it has been found in Eastern Europe. Specimens from Lake Balaton in Hun- gary differ somewhat in structure from any Indian form, but their gemmules are closely similar. Ephydatia fluviatilis, auct. 1911. Ephydatia fluviatilis, Weltner, Trans. Soc. Nat. St. Pétersbourg XLII, p.'59, pl. i. 1916. Ephydatia fluviatilis, Annandale, Fourn. As. Soc. Bengal (n.s.) XI, Pp. 445. Sponges of this species we1e found in the Hamun-i-Helmand in two different types of environment, on the lower surface of blocks of hard clay at the edge of the lake and on the stems of bulrushes in the reed-beds. Specimens from these two habitats differ con- siderably, but neither affords any very definite diagnostic character whereby it might be distinguished nominally from the forma typica of the species. Both phases differ from the Himalayan var. himalay- ensis (which is so near the Syrian var. syrvaca that it is hardly worth while to distinguish them) in the almost complete absence of spines or tubercles, however minute, on the skeleton-spicules. Sponges on the stems of bulrushes form a layer 2 to 3, rarely 5 mm. thick. The outline of each mass is oval, following the iong axis of the reed, which it rarely, if ever, completely encircles. Few are more than about 70 mm. long. Their colour is dirty white. The external surface is smooth and rounded with but moderately conspicuous exhalent orifices and radiating superficial channels. The consistency is very soft and friable. The skeleton contains little binding substance and is not particularly regular in structure. There are no bubble-cells. The skeleton-spicules are short and slender, sharply but abruptly pointed, often a little irregular in outline and sometimes bearing a few widely scattered extremely minute tubercles, as a rule gently curved but sometimes bent in the middle or elsewhere almost abruptly. In some parts of the sponge there are groups of very small and slender spicules. Measurements of these are not included in the table given below. The gemmule- 88 Records of the Indian Museum. [Vor. XVIII, spicules are well-developed and normal in appearance. A few are scattered in the parenchyma. ‘The shafts are stout and as a rule considerably longer than a single rotule, with the spines upon them by no means strongly developed. ‘The rotules are deeply but irregularly divided, their denticulations having the form of flattened spines more or less welded together at the base but without any trace of webbing. The gemmules are small and somewhat depressed. Their pneumatic layer is thin and they bear a single layer of spicules. These specimens are from station 21 of our expedition. I give with their measurements those of specimens from station 20, which I will describe, for comparison. 4 \ [ i | | : B. Fic. 3.—Spicules of Ephydatia fluviatilis from the Hamun-i-Helmand, x 250. A=spicules of specimens from the lower surface of a block of clay at the edge of the lake. B= spicules of a specimen from the stem of a bulrush in a reed-bed. Measurements of spicules, etc. Stas 20: Stasi2r. Length of normal skeleton- spicules »- 0°246-0°328 mm. 0°2624-3075 mm. Maximum breadth of normal skeleton-spicules .. 0°006-0°025 __,, 0°0123-0°01435 ,, Length of normal gemmule-spicule .. 0°020-0'0369__,, 0°0287-0'0369_,, Diameter of rotule 0'016-0°0240 _,, 0°0246-0°0328 _,, Diameter of gemmule 0°340-0°375__,, 0°345-0°359 ” 1919. | N. ANNANDALE: Sponges, Hydvozoa and Polyzoa. 89 The specimens from sta. 20 are from the lower surface of blocks of hard clay which had fallen into the lake from cliffs of that sub- stance. Their skeleton-spicules exhibit less variation and are as a tule shorter and relatively stouter than those from the stems of bulrushes. Their normal gemmule-spicules are also usually stouter and shorter with relatively larger rotules, but very long spicules of the same type occur occasionally. The whole sponge is so full of particles of clay that it is almost impossible to study the structure of the skeleton in detail, but it is certainly (doubtless for this Fie. 4.—Gemmules and gemmule-spicules of Ephydatia fluviatilis from the Hamun-i-Helmand. A = outer rotules of a specimen from a block of clay at the edge of the lake: highly magnified. B—=agemmule of a specimen from the stem of a bulrush as seen from above: X 20. reason) very lax and amorphous and the groups of small spicules characteristic of the other phase seem to be absent, though small amphioxi occur scattered in the parenchyma. ‘The colour is that of the clay. The gemmules are normal except that in some single birotulate spicules are, as it were, plastered on outside the normal single row in a vertical or slanting position. ‘They are held in position by an extension of the outer horny coat, which covers them completely. It is probable that there was a difference in the chemical com- position of the water from which these two sets of sponges came go Records of the Indian Museum. [Vor Lie (see p. 97 antea), and they were growing in very different types of environment, though in the same lake at a distance of less than five miles apart. The sponges from sta. 2I were living in most unfavourable conditions on the lower surface of blocks of clay part- ly embedded in soft mud, which permeated their whole substance, and in a situation liable to desiccation with a change of wind, and also to the effects of nightly frost. Those from sta. 20 on the other hand were living in much deeper water, protected from frost and wind and not in any danger of being choked by mud (v. p. 91). The sponges from the blocks of clay were in a somewhat similar situa- tion to those of the var. syriaca I collected in the Lake of Tiberias,! except that the latter were attached to solid stone. In the other instance the method of growth is similar to that of Spongilla fragilis in Japan.* The specimens from the edge of the Persian lake are so en- veloped in and permeated by mud that their whole structure is dis- torted, whereas those of the Lake of Tiberias were normal in struc- ture but small. Moreover, there is no trace of green corpuscles in the Persian specimens, though minute extracellular algae of various kinds are found in their parenchyma. The specimens from both types of environment in the Hamun- i-Helmand were in an active vegetative condition in December, ~but both contained numerous gemmules. I can find no trace of embryos. No specimens of EF. fluviatilis from the Hamun bear any particu- lar resemblance to those of the same species described from lakes in Central Asia by Weltner (0p. cit.), except that the skeleton-spicules of those growing on reeds have a somewhat similar outline to those from Issyk Kul figured by him in figs. 8-14 on p. 65 of the work cited. In the occurrence in the sponges from the margin of the lake of occasional abnormally large birotulate spicules they resemble the Australian E. multiformis,? but that species (? or variety of EF. miillert) possesses bubble-cells in its parenchyma. I have examined a cotype or schizotype sent me before the war by Dr. Weltner and have found in it a spicule of this type, but neither in Persian nor in Australian specimens have I discovered such spicules 7m situ on the gemmule. I see no reason to regard them as adventitious but believe that they are produced free in the paren- chyma, perhaps as a result of abnormal environment. HYDROZOA. Hydra vulgaris, Pallas. 1911. Hydra vulgaris, Annandale, Faun. Brit. Ind., Freshw. Sponges, etc., p. 148, fig. 29, p. 131, fig. 27A. ! Annandale, Fourn, As. Soc. Bengal (n.s.) 1X, p. 59 (1913) and XI. p. 455 (1916). 2 Annandale and Kawamura, YFourn. Coll. Sci. Univ. Tokyo XXXIX, p. 13 (1916). 8 Weltner in Michaelsen and Hartmeyer’s Faun. Siidw. Australiens III, p- 138 (1910). 1919. | N. ANNANDALE: Sponges, Hydrozoa and Polyzoa. gt A single small specimen of a pale brownish colour and without buds or reproductive organs was taken amongst green filamentous alga in an irrigation-channel at Nasratabad, Seistan, in December. POLYZOA. The four species of Polyzoa collected in Seistan all belong to the Phylactolaemata and all but one are sessile, branching species. Otherwise they have little in common. ‘The two Plumatellinae are remarkable for the differentiation exhibited between the zooecia that produce free and those that produce fixed statoblasts. All the species were found in full activity in December, except the Afrindella, which was taken in foul water and was in a degenerate condition, densely packed with statoblasts of both kinds. Lopho- podella cartert, which elsewhere has been found associated with algae,! was only observed in an active state in Seistan attached to tubes inhabited by a small Oligochaete worm, but perhaps stolen by the worm from a Dipterous larva. Fredericella sultana subsp. jordanica, Annand. 1913. Fredericella sultana jordanica, Annandale, Fourn. As. Soc. Bengal (is: pp. 223, pl. vil, figs. 51a, £0, te 1915. Fredericella sultana subsp. jordanica, td., Trav. Sta. Biol. Volga (Saratow) V, p. 74. Specimens that may be assigned to this race were abundant in December both on the stems of bulrushes in the reed-beds of the Hamun-i-Helmand near Lab-i-Baring and on the lower surface of blocks of clay at the edge of the lake near the same place. It was also found on empty Unionid shells in the open lake: Though many of the colonies were degenerate they contained few stato- blasts and the peculiar thickening of the ectocyst noted in associa- tion with the formation of gemmules in the Lake of Tiberias was not observed. ‘The zooecia were narrow and still more strongly keeled and emarginate than in specimens from Palestine or the Volga. F, sultana is apparently cosmopolitan as a species. ‘The Pales- tinian race has hitherto been found only in and near the Lake of Tiberias and in the lower Volga system in eastern European Russia. In the plains of India it is replaced by the race ¢ndica, while the typical form occurs in the lakes of Kumaon in the Western Himalayas. Fam. PLUMATELIIDAE. Genus Plumatella, Lamarck. Subgenus Afrindella, Annandale. 1912. Afrindella, Annandale, Rec. /nd. Mus. VII, p. 140. This subgenus has hitherto been found only in tropical Africa, India, Siam? and the Philippines. Its occurrence in Seistan is, L West and Annandale, Fowrn. As. Soc. Bengal (n.s.) VII, p. 83 (1911). ® Plumatella (Afrindella) tanganyikae occurs in the inner lake of the Talé Sap in the Siamese province of Singgora or Sunkla. 92 Records of the Indian Museum. [Vo XVIII, therefore, evidence for the existence of a tropical element in the aquatic invertebrate fauna of that country. Key to the species of the subgenus Afrindella. A. Zoaria forming a single layer. 1. Ectocyst smooth ; zooecia regular in growth, with a strong continuous keel ; statoblasts (free) elongate .. Plumatella philippinensis. 2. Ectocyst obscurely annulate, densely covered with minute sand-grains ; zooecia with- out keel; statoblasts (fixed) broad, vari- able in shape... “a ... P. testudinicola. 3. Ectocyst rough, irregularly annulate on the distal region; proximal region of zooe- cium strongly keeled; statoblasts (free and fixed) moderately elongate ... P. tanganytkae. B. Zoarium in two layers, in the lower of which fixed, in the upper free statoblasts are produced. Ectocyst more or less irregular; zooecia without keel ; statoblasts (free and fixed) elongate ae Ay nee 70. PEVSUGH. Piumatella (Afrindella) persica, sp. nov. This species closely resembles the Gangetic Plumatella testu- dinicola' in structure, but is differentiated (apart from the method of growth) by the possession of free as well as fixed statoblasts and by the more elongate form of the latter. The specimens examined consist of rather dense growths on the woody roots and stems of water-plants which were in a condition of rest in winter. Each growth is separated quite definitely into two layers. Most of the zooecia of the outer layer are degenerate but some still contain polypides, while all except the youngest are packed with free gemmules. Those of the lower layer are filled with fixed statoblasts arranged in single longitudinal rows. The zoarium is everywhere too congested to reveal its precise method of growth, but even round the margins of the colony, where the youngest zooecia occur, the two layers are distinct and the lower zooecia contain fixed statoblasts. In this part of the zoarium the zooecia are arranged roughly in parallel lines and it is clear that the system of budding was that of a terminal and a latero- terminal bud being given off almost simultaneously by each termi- nal zooecium, and that owing to the congested state of the colony the latero-terminal buds have been closly adpressed to the terminal ones. ‘The zooecia lie practically flat, all orientated in one direc- tion and each with its orifice opening almost horizontally. The base of the buds being somewhat flattened and issuing from the lower part of the parent zooecium permits the latter to open in this way. When the polypides expand they doubtless bend up- wards, which the softness and laxness of the distal part of the zooecia would readily permit them to do. The zooecia are nearly cylindrical but flattened on the attached surface. They are about 1 Annandale, Rec. Jnd. Mus. VII, p. 148, pl. xin (1912). 1919. | N. ANNANDALE: Sponges, Hydrozoa and Polyzoa. 93 0°77 mm. in diameter and not longer than 255 mm. In the denser parts of the colony they are often bent or twisted in their long axis. The proximal region of each zooecium is brownish, smooth or irregularly annulated and translucent, the softer distal region colourless, transversely wrinkled when the polypide is retrac- ted and transparent in fresh specimens. In degenerate colonies this region disappears with the polypides, but in living areas it is of relatively large extent. The polypides are nearly colourless throughout. I have not been able to detect any distinctive feature in their anatomy. The statoblasts are of the elongate type. Those of the free kind are from 14 times to over twice as long as broad. ‘They have the sides nearly parallel and the ends broadly rounded. The ring of air cells is narrow and not much broader at the ends than at the sides. It encroaches little on either surface. ‘These statoblasts are somewhat curved in their long axis. The fixed statoblasts resemble the free ones in shape, but are larger and usually broader and more variable in outline. They are very flat but slightly con- vex on the dorsal surface, black, smooth and polished. Each is surrounded by a delicate peripheral crenulate carina separated from the body of the statoblast by a deep but narrow groove. Both kinds of statoblast are large compared with the calibre of the zooecium. Measurements of statoblasts (tn millimetres). Free. Fixed. Length oP, .. 0°289-0°374 0°34-0°544 Breadth xe te Orl7-0: 204 0°255-0°272 Type-specimen. P. 2 Z.S.1. (Ind. Mus.). Locality, etc.—Our specimens were found at the bottom of a pool of very foul water in the nearly dry bed of the Randa stream near Jellalabad, Seistan. They coated the peculiar nodular roots and the stems of some plant which grew in the mud but had died down completely in winter. Subgenus Hyalinella, Jullien. 1885. Hyalinella, Jullien, Bull. Soc. zool. France V, p. 133. 1910. Hyalinella, Loppens, Ann. Brol. Lacustre, 1V, p. 147. 1git. Australella, Annandale, Faun. Brit. Ind., Freshw. Sponges, etc., 4 PA 6 19tt. Plumatella (in part), td., ibid., p.212. 1914. Australella, Kraepelin in Michaelsen’s, Lauwd-u. Siisswass. Fauna Deutsch-Sudwestafrikas, X1, p. 61. 1916. Australella, Annandale, Rec. Ind. Mus. XI, p. 163. The one diagnostic feature of this subgenus is that the true ectocyst is transformed into a gelatinous layer, which may be so thick as to produce a synoecium analogous to that of the Lopho- podinae. The growth, however, is always dendritic, and the tent- acles never of great length. Until now no fixed statoblasts have 94 Records of the Indian Museum. [Vor. XVIII, been observed, but they are highly developed and specialized in the species to be described here. Some species of the genus bear a close external resemblance to those of the group of Pluma- tellae that has been named Alcyonella, but in Alcyonella a horny ectocyst is present as well as the gummy substance by means of which the zooecia are agglutinated together. In preserved speci- mens of Hyalinella the ectocyst is apt to shrink and lose its gela- tinous character, butin the natural condition its structure and appearance are most characteristic. It is usually much harder than the zooecium of the Lophopodinae. Six species may now be assigned to the subgenus:—Plumatella punctata, Hancock, the type-species; P. bigemmis, sp. nov.; P. indica, and P. longigemmis (Annandale); P. jheringi (Meissner), and P. lendenfeldi (Ridley). The first of these is widely distributed in Europe and North America and occurs also in tropical Africa and in India; the second is here described from E. Persia; P. indica and P. longigemmas are Indian ; P, jheringi comes from Brazil, and P. lendenfeldz from Australia. These species may be distinguished by the following key :— I. Ectocyst not greatly swollen, fairly soft, not concealing the identity of the zooecia. A. Statoblasts of one type only, all free. 1. Statoblasts not much longer than broad, very variable — 3 ... P. punctata. 2. Statoblasts nearly 1} times as long as broad, not particularly variable... .. P. longigemmis. B. Fixed statoblasts present as well as free .. BP. bigemmis. II. Ectocyst stiff, not greatly swollen but compacting the zooecia together into a solid mass. Statoblasts oval, rounded at the ends we De dndica. III. Ectocyst very copious, soft; the distinction between zooe- cia entirely obliterated. A. Statoblasts oval, subtruncate at the ends ... H. lendenfeldt. B. Statoblasts subcircular or polygonal ... HA. pheringt. As is shown in this key the distinction between my Australella and Jullien’s Hyalinella is merely a matter of degree. Some speci- mens of the species now to be described might be assigned with equal propriety to either. Nor does the one differential character, considered in this light, seem sufficient for generic separation from Plumatella. Plumatella (Hyalinella) bigemmis, sp. nov. The zoaria grow prone on the stems of plants and have much the appearance of those of H. punctata, except that the ectocyst is still more transparent and swollen and quite smooth on the external surface. Young zoaria have an almost linear growth, slightly zig-zag owing to the subterminal buds being produced on opposite sides of alternate zooecia. Though these buds are lateral in origin they are directed almost straight ahead, so that the deviation from a straight line is not great. As the colony develops, the zooecia are pressed together into a compact layer. This is brought about by the production of lateral branches which form an acute angle with the main axis of the colony. A radiating IQI9Q. | N. ANNANDALE: Sponges, Hydrozoa and Polyzoa. 95 zoarium may also be produced and in the mature colony the orientation of the zooecia is often radial, mainly in four directions. In any case a large number of the zooecia always point in the same direction. The colony as a whole is quite flat, the gelatin- ous ectocyst filling in the interstices between the zooecia. The individual zooecia maintain their identity distinct, but their ectocvst is so thick that their openings have in more conges- ted parts of the colony a honeycomb-like appearance. The ectocyst is, as already stated, usually quite hyaline and colourless, but it is sometimes darkened towards the distal extremity of the zooecia It is hard and almost cartilaginous for the greater part of its length, but the harder region ends abruptly near the aperture, which is surrounded by a thin, soft, mainly retractile membrane. The margin of the former region is well-defined, of an oval form and somewhat oblique in its long axis, which is mainly vertical in direction. Sometimes the external surface is covered with minute algae. The zooecia: are long, but somewhat variable in length, sometimes bent or curved in their long axis. They are distinctly flattened. Their transverse diameter (internal) is about 0°5 mm. and even when quite young they are of almost equal calibre throughout their length. Their long axis is parallel to the surface to which they are attached. This description applies to the normal zooecia which consti- tute the greater part of the colony, but in old zoaria zooecia of another type are produced at or near the terminal points of the branches. These are variable in shape and sometimes shorter, occasionally longer, than the normal zooecia. Zooecia of this type (which are only produced when the vegetative period of growth nears its end) never contain a fully developed polypide but only one or more statoblasts partly embedded in a strand of undifferen- tiated tissue, which broadens out towards the distal extremity of the zooecium. The polypide is much like that of H. punctata and offers no particular diagnostic characters. The tentacles are moderately short and not very numerous and the whole body is almost colourless. The free statoblasts are very like those of H. punctata, but not so variable in shape, a little more rhomboidal, and with a broader ring of air-cells at the extremities. The fixed statoblasts are large, broadly oval, of a dark brown colour and densely punctured on the surface. Each is, however, surrounded by an amorphous mass of dark horny material that obscures its true shape and ornamentation. ‘The statoblasts are not very numerous. Measurements of statoblasts (in millimetres). Free. Fixed. Length .. ae). 0°357-0° 374 0°459-0°561 Breadth 2. 0°255-0°272 0°425-0°459 Type-specimen. P. %, Z.S.1. (Ind. Mus.). 96 Records of the Indian Museum. [VoL. XVIII, Locality.—Our specimens were found on the stems of bul- rushes in the reed-beds of the Hamun-i-Helmand near Lab-i-Baring in December, 1918, with Fredericella sultana and Ephydatia flu- viatilis. Genus Lophopodella, Rousselet. 1904. Lophopodella, Rousselet, Fourn. Quek. Micr. Club (2) IX, p. 45. 1911. Lophopodella, Annandale, Faun. Brit. Ind., Freshw. Sponges, etc., 1914. Tubbs poeta. Kraepelin in Michaelsen’s Land-u. Susswassevfauna Deutsch-Sudwestafrikas 1, p. 64. Kraepelin gives in the work cited a useful key to the species and figures the statoblasts. He points out that the African species hitherto confused with L. carteri is distinct, and describes it under the name L. stuhlmanni. He also describes a new variety of L. capensis (Sollas) under the name var. michaelsent. The forms that must now be referred to the genus are L. cartert (Hyatt), L carteri subsp. davenporti (Oka), L. thomasi, Rousselet, L. cap- ensis (Sollas), L. capensis var. michaelsent, Kraepelin and L. stuhl- manni, Kraepelin. ‘The range of the genus extends from Eastern Persia to Japan, Brazil and South Africa, but is mainly tropical. The following key, though not actually based on Kraepelin’s, owes much to it. Key to the species of Lophopodella. 1. Each extremity of the statoblast produced into a long slender process bearing books along each margin ... L. capensis. 2. Extremities of statoblast truncate or subtruncate, with a single row of hooked processes. A. Extremities of statoblast broadly truncate, little narrower than the greatest transverse diameter .., L. stuhlmannt. B. Extremities of the statoblast broadly rounded, much narrower than the greatest transverse diameter ,,, L. cartert. C. Extremities of the statoblast very narrow, concave L. thomast. All of these species except L. carteri are African. Lophopodella carteri (Hyatt). 1911. Lophopodella cartert, Annandale, Faun. Brit. Ind. Freshw. Sponges, etc., p. 233, fig. 46, pl. iii, figs. 4, 4a. 1912. Lophopodella cartert, id., Rec. Ind. Mus. VII, p. 143. Specimens from an irrigation channel in the Consulate garden at Nasratabad, Seistan agree well with Indian specimens. Stato- blasts were also taken, with gemmules of Spongilla carteri and S. alba, amongst drift at the edge of a pool in the desert in the same district. The statoblasts did not differ in any respect from those of Indian colonies. Statoblasts were found in active colonies in all stages of development in December. The animal was living among green fila- mentous algae. The most interesting feature of these colonies was, however, that each was attached to a fine mucilaginous tube and that each tube was inhabited by a small Oligochaete worm 1919. | N. ANNANDALE: Sponges, Hydrozoa and Polyzoa. 97 identified by Tt.-Col. J. Stephenson, I.M.S., as Nats communis var. punjabensis. ‘This worm does not ordinarily construct tubes, and was found free in the reed—beds of the Hamun, but Col. Stephenson has seen it in tubes made by insect larvae (probably those of Diptera) in the Punjab. It is possible that in Seistan also it occupied the dwellings of larvae which it had dispossessed or succeeded after their departure, but as to the association between it and Lophopodella there can be no doubt, for it was noted repeat- edly on more than one occasion. L. carteri has been observed in association with certain algae,' and it is not uncommon for Chirono- mid larvae to construct their tubes at the base of its colonies ; but I have not hitherto found it associated with Oligochaete worms. The known geographical range of L. carteri now extends from Eastern Persia to Japan. The Japanese and Chinese race (daven- porti, Oka *) is distinguished from the forma typica by the greater development of the terminal processes of the gemmule, but nothing is known of the species in the countries intermediate between India and China.’ c : a ' gies A Fear a yt ay f i < ae es . a Pra J + 2 2 és a a = i Bei pe ‘ » # : 7 ia: a m4 - pe a : . i n y ’ = iy F Ls I ‘ < i - r, PS 5 J i ‘ ja ; call ‘ 4 4 3 » ¥ mn a - fr an i x iP — ye Hie: Fic. EXPLANATION OF EPRALE X1r POLYZOA OF SEISTAN. ° Plumatella (Hyalinella) bigemmis, sp. nov. Len! .—A part of the peripheral region of a mature but growing colony on the stem of a bulrush, showing the resting statoblasts, X 5%. .—A part of the peripheral region of a more compact and vigorous colony, X 5%. The white substance in the interstices between the zooecia is composed of microscopic algae. N .—Three zooecia containing statoblasts, X 40. On the right a zooecium in which the polypide has completely degen- erated and a single free statoblast has been produced, is shown growing over aterminal zooecium in which a single fixed statoblast has been formed. The degenerated polypide of the latter is also shown. Its roof has largely disappeared. On the left a zooecium with a single statoblast is figured. Oo Plumatella (Afrindella) persica, sp. nov. . 4.—Fixed statoblasts, X 40. 5.—A part ofa still vigorous colony growing ona rootlet, X 5%. 6.—A part of a colony in which most of the zooecia of the upper layer have degenerated and disappeared, leaving those of the lower layer exposed. The roof of the latter has also degenerated, exposing the fixed stato- blasts, X 5%. 7.—Free statoblasts, X 40. REC. IND. MUS., VOL. XVIII, 1919. Plate XIl. 7x40 A. C. Chowdhary del. . POLY ZOA OF SEISTAN. CA RAB LT DAE 3F ROM “SB VS i AN. By H. FE. ANDREWES. [The Carabid beetles on which Mr. Andrewes has been kind enough to report were collected, mostly by Mr. S. W. Kemp, in the depth of winter. With the exception of Metabletus fuscomaculatus and Microlestes corticalts, which come from the desert, they were found at the edge of water-channels, streams or pools, or at that of Hamun-i-Helmand or lake of Seistan. On the shore of this lake there is in winter a drift-line, marking flood-level and consisting mainly of fragments of reeds. Under these fragments a fairly large fauna (consisting mainly of beetles and Heteroptera, but also including earwigs, a toad, etc.) finds shelter, and here we took no less than seven of the sixteen Carabidae found in Seistan.—N. Annandale. | The species of Carabidae taken in Seistan number sixteen only. These show little relationship to the Indian fauna, though there are two species of which I have records from Karachi and the United Provinces respectively. A third species has not yet been recorded as coming from India, but I have examples of it in my collection from the Himalayas as mentioned below. Seistan—— judging by the Carabidae—seems to be near the boundary (in so far as it exists) separating the fauna of Central Asia from that of the long sandy tract which stretches from Morocco to Sind. Ihave been able to determine most of the species, but have had to leave one or two unidentified. Distichus planus, Bon. I Ex. No. 8782. lLab-i-Baring, a 7-xii-18, under drift at edge of Hamun. Common in S.E. Europe, S.W. Asia, Egypt, etc. Dyschirius sp. 3 Ex. No. 8795. Edgeofstream of saline water near Lab-i- Baring, Seistan, I1-xii-18. I have not been able at present to identify this species. Pogonus micans, Chaud. It Ex. No. 8791-3. Wab-i-Baring, Seistan, 7-x1i-18, under drift at edge of Hamun. N. Persia: Astrabad. Kopet-Dagh: Germab. Bembidium sp. I Ex. No.8779. Nasratabad, Seistan, ca. 2,000 ft., 26-xi-18. I cannot identify this species, but a large number have been described from Central Asia, and this may be among them. 100 Records of the Indian Museum. [VoL. XVIII, Bembidium niloticum, Dej. 1 Ex. No. 8799. Water-channels near Nasratabad, Seistan, x1-18. Described by Dejean from Egypt, whence it extends east- wards as far as Japan, and south to the Malay Peninsula; I have seen Indian specimens from the Himalayas, but not further south. Bembidium varium, Oliv. Ir Ex. Nos. 8798-9, 8802. Near Jellalabad, Seistan, 2-xii-18. Channel 8 miles E. of Lab-i-Baring, Seistan, 16-xii-18. Extends apparently over the whole of Europe, and the greater part of temperate Asia. Bembidium latiplaga, Chaud. 5 Ex. Nos. 8794, 8797, and 8800. Channel 8 miles E. of Lab- i-Baring, Seistan, I6-xil-18. Water channels near Nasratabad, Seistan, x18. Edge of stream of saline water near Lab-i-Baring, Seistan, L1-xti-18. Mediterranean basin and S. Russia. I am not aware that the species has been recorded yet from a locality so far east as Seistan. Siagona europaea, Dej. 1 Ex. No. 8789. lab-i-Baring, Seistan, 7-xii-18, under drift at edge of Hamun. This is the furthest point east from which I have seen an example of this species, which is commonly found in the Mediter- ranean basin. I have quite recently (Ann. and Mag. Nat. Hist., (9) iii, I919, p. 470) given my reasons for considering it distinct from the Indian S. depressa, F. Chlaenius semicyaneus, Solsky ? 1 Ex. No. 8777. Hurmuk, Perso-Baluch-Afghan frontier, Ca. 3,000 ft., 22-xi-18. A single @ example agreeing fairly well with Solsky’s description, but I have no specimen available for comparison. Chlaenius spoliatus, Rossi. 3 Ex. Nos. 8780, 8783-4. lWab-i-Baring, Seistan, 7-xii-18, under drift at edge of Hamun. This species has a very wide range, from S. Europe and N. Africa to Japan; it does not extend to Southern Asia. Diplocheila transcaspica, Sem. 2 Ex Nos. 8781 and 8785. lLab-i-Baring, Seistan, 7-x1i-18, under drift at edge of Hamun. 1919. | H. E. ANDREWES: Carabidae. IOI Hitherto known only from the transcaspian provinces of Russia. Pterostichus sp. 3 Ex. 922. Nos. 8786-8. Lab-i-Baring, Seistan, 7-xii-18, under drift at edge of Hamun. Possibly one of the numerous species described from Central Asia. Brachinus sp. I Ex. No. 8790. Lab-i-Baring, Seistan, 7-xii-18, under drift at edge of Hamun. Unknown to me. Chaudoir’s Monograph of this genus un- fortunately does not embrace the palaearctic species. Glycia ornata, Klug. I Ex. No.8778. Nasratabad, Seistan, ca. 2,000 ft., 26-xi-18. Ranges from the Mediterranean basin through the desert regions of Egypt and Arabia as far as Sind. Metabletus fuscomaculatus, Motch. I Ex. No. 8796. Wab-i-Baring, Seistan, 15-xii-18, at base of desert plants. South Russia, Mongolia, Caucasus, Syria; I have specimens in my collection from Chamba, Himalayas, but the species has not previously been recorded from the Indian region. Microlestes corticalis, Dufour. 2 Ex. oo No. 1401. Lab-i-Baring, Seistan, 15-xii-18, at base of desert plants. Extends from the Mediterranean basin to Central Asia. - eit Lite ae : oF ate fy oe gle ecg teow om ees eo NS, r SY BAL Se ilps. hrs 4. 3 oe > iP barat ise So - % mrt tsa : - sar - re Th Oiwtso ae clei” ion fair HVE) hth Pes df) ef ¢ SiS oye es ‘ ay 4 ' ; 4 : i_<™ 7 ; Z a ve a) ae =4 me ‘« Bets’ Pal hes Bay 7 Oa ae / 2 J it, 42 bed: Sis ae: = 68 es ot au st ‘< Be # ao Lait cs rr wt fe ; ee ant 44 Wipil it she ate "isthe : =e Or ox: ” v" . % Wie a iat $e ee i eee penta iO) hee : 7 | mies Otive2 a ~4 sige en Le mitch a 9 Stith ae eotiag? Galante. ' ‘ a rs 19a s # j av Iey jaa Ms Enee oe tt oe eae < - ‘ a0 a 7 a S iy ae vi’, aA : ~ Se Hh Vath Gh Oe Rca Oe L 7 ae) i) A.4 a 7 t + - aa - > = REPORT .ON, DEE PER See We Acie GASTROPOD MOLLUSCS OF LOWER MES OP OT AMTA: PART I. ‘THE GENUS LIMNAEA. By N. ANNANDALE, D.Sc., F.A.S.B., Director, Zoological Survey of India, and BAtnt PrasHap, D.Sc., Offg. Director of Fisheries, Bengal, Bihar and Orissa. (With Plates XIII, XIV.) In the report of which the first part is now published we pro- pose to discuss the freshwater Gastropod molluscs of the delta of the Tigris and Euphrates and of the lower reaches of the two rivers. Stress of other work and official duties have already greatly delayed its preparation, and as the medical authorities are anxious for any information about the freshwater molluscs as soon as possible, we have decided to issue it in parts dealing with separate genera or larger groups. It will be convenient to include this report in the same volume as that on the molluscs of Seistan, as the two faunas are reiated. The material from Mesopotamia that we now have in our hands consists of three collections, ail presented by the generosity of their collectors to the Zoological Survey of India. Two of these have already been discussed! They were made by Lt. Col. W. H. Lane and Bombadier R. Hodgart and consist of empty shells, most of which were probably subfossil. The third collection, made by Capt. C. L. Boulenger, adds greatly to our knowledge as it includes specimens preserved in spirit. It has been possible with its aid to correct and expand the results based on shells previously examined. We have, so far as possible, consulted the literature on the Persian, Central and Western Asiatic molluscs, as well as that on those of India and Europe, but it is possible that some Russian works, or memoirs published in Europe during the war, may have escaped our notice. The only paper dealing specifically with Lower Mesopotamia that we have found is Mousson’s, ‘‘ Coquilles ters. et fluv. rec. Dr. A. Schlefli en Orient” in the Journ. de Conchyl. XXII (1874). The descriptions in this paper are fairly full and it has not been difficult with specimens before us to dis- criminate the species. The paper, however, is not illustrated, and without figures or specimens verbal descriptions of freshwater molluscs have little value. We are strongly of the opinion that t Annandale, Rec. Ind. Mus. XV, pp. 159-170, pl. xx (1918). 104 Records of the Indian Museum. [VoL. XVIII, among the Pulmonata at any rate no description unaccompanied by a recognizable figure should be considered valid. Genus Limnaea, Lamarck. The species that occur in Lower Mesopotamia are of con- siderable interest. They fall into two of the main groups of the genus, namely those that may be called, in quite a general sense, those of L. auricularia and L. truncatula. Most of them we assign to the former group. These species are all variable and at first sight it might seem that most of those of southern Persia and the lower valley of the Tigris and Euphrates were conspecific, merely separable into numerous races and varieties. A careful study ofa large amount of material has, however, convinced us that this is not the case, and that a number of species actually exist which can be separated by anatomical as well as merely conchological characters. The examination of young shells is most important for this reason, and we find that adults which resemble one another rather closely can often be traced back to young forms that differ considerably. In conditions such as those found in Lower Mesopotamia, and also in most parts of Persia, convergence seems to have taken place between different species of the genus on a fairly extensive scale, and forms which are perhaps not closely related genetically have come to resemble one another owing to the plastic influence of environment. In Seistan we know of two species of the L. auricularia group (L. gedrosiana and L. baclrianu) which at first sight appear almost identical, but on a detailed investigation have proved quite dis- tinct. We are of the opinion that these species each had a differ- ent line of ancestry, but have thought it best in the present state of our knowledge not to discuss these lines of ancestry in detail. Similarly in collections from Tower Mesopotamia we have found five forms that seemed to us on a superficial examination to grade the one into the other. ‘Two of these (L. gedrosiana and L. bactriana) are the same as the two Seistan species. The others are L. peregra canalifera, L. tenera euphratica and L. cor. It is important, therefore, to consider what common features are to be found in the forms of each country, and to what extent these features can be correlated with peculiarities of environment. Both Lower Mesopotamia and Seistan are countries that have a prolonged dry season, when the temperature is high, and are nevertheless liable to floods. In Seistan the water always contains a high percentage of mineral salts, while the delta of the Tigris and Euphrates is an estuarine tract in which even water that is practi- cally fresh is probably liable to be contaminated with water of com- paratively high salinity. Moreover, this tract has been gradually extending southwards and receding from the north for a consider- able period, and many of our specimens are from old deposits that must have been laid down in conditions far more estuarine than 1919.] N. ANNANDALE & B. PrasHap: Gastropod Molluscs. 105 those that now prevail in the districts where the specimens were collected. Both the species of Limnaea from Seistan and those from Mesopotamia have in most instances small and very thin and fragile shells, which are sculptured with more or less prominent longitudinal tibs, but the ribs on those from Mesopotamia are much more pro- nounced than on Seistan shells. This is so even in identical species. The shape of the shell in most forms exhibits slightly different modifications in the two countries. The two Seistan species are both rather narrow but have relatively large and patent mouths and short, pointed spires. The three distinct Meso- potamian forms exhibit a tendency, on the other hand, to grow broad in proportion to their height, while their spires are still shorter. ‘Their mouths, in the broader forms, are relatively narrow but on the whole resemble those of L. gedrosiana and L. bactriana. An entirely new character appears in Mesopotamia in the canaliza- tion of the suture above the body-whorl found in all the endemic species and well-defined races. How far can these differences and resemblances be correlated with similar peculiarities of the shell from different or identical types of environment in other countries ? Baker,’ in his fine mono- graph of the North American Limnaeidae, states that shells from saline districts in that country are prominently ribbed, but whether this ribbing is associated with the presence of any particular salt we are not informed. We know ina general way what salts are present in the waters of Seistan,? but no information is available about the composition of those of the Euphratic delta. Thinness, fragility and paleness of shell in Lemmaca are usually associated with unfavourable conditions of life, but more precise information is needed on the point. The reduction of the spire and enlargement of the body-whorl and mouth provide additional space for the branchial chamber and allow a larger portion of the animal to be extended from the shell. This character, moreover, seems to be definitely correlated with difficulty in obtaining the necessary oxygen, and we find different forms of the same species 3 (living in different types of environment in the same localities) in which it is more developed in those that live in still water than in those that inhabit streams. Both in Seistan and in Mesopotamia the Lamnaeae live mainly in flooded country, perishing in large numbers when the floods subside. The few individuals that survive to perpetuate the species from year to year retire to shallow pools, backwaters or lakes. In the course of such an existence conditions must be encountered in which life is difficult and water lacking in free oxygen. The Limnaeae of Mesopotamia are, except L. peregra canalifera, clearly depauperated forms, and the considerably larger size attained { Baker, Spec. Pub. Chicago Ac. Sci. U11 (1911). i Annandale, Rec. Ind. Mus. XVIII, pp. 10, 15 (1919). 8 Annandale, Rec. Ind. Mus. XIV, p. 149, fig. 4 (1918). 106 Records of the Indian Museum. [Vo,. XVIII, by this race is in itself evidence of a different origin from that of the others associated with it—evidence which is greatly strength- ened by an examination of the young shell. In this discussion we have left L. hordeum out of account. It is so different from the other species of the fauna that there can be no doubt of its different origin ; there can be little doubt of its re- lationship to L. truncatula. It has, however, a larger mouth to the shell than that species and more tumid whorls, especially of the spire. It has, therefore, followed a line of evolution in some res- pects parallel to, in others divergent from, that which has resulted in such forms as L. cor. Key to shells of Limnaea from S. Persia and Lower Mesopotamia. A. Height of mouth less than 3 that of shell.1 1. Whorls of spire tumid; suture oblique _ .., ... L. hordeum. 2. Whorls of spire not tumid; suture much less oblique... L. truncatula. B. Length of mouth more than % that of shell. 1. Upper surface of body-whorl not flattened; suture above it not canalized. (Height less than 20 mm.) a. Mouth of shell projecting strongly in ventral view ; arc of lip practically a semicircle... ... WL. persica: b. Mouth less expanded, projecting less; arc of lip less than a semicircle. i. Whorls of spire somewhat tumid 2; suture impressed ; main axis of mouth forming an acute angle with that of shell. ... ... L. bactriana. ii. Whorls of spire not at all tumid; suture little or not at all impressed; main axis of mouth parallel to that of shell. a Arc of lip quite regular; apex sharply pointed 2a Sas ... L. tranica: 8 Arc of lip irregular; apex bluntly pointed. + Arc of lip never more than slightly flattened... Ae ... L. gedrosiana. * Arc of lip so flattened as to form a slanting straight line ... ... L. gedrosiana var. vectilabrum. 2. Upper surface of body-whorl distinctly flattened and deeply depressed round suture. a. Shell over 15 mm. high, not very fragile, with the spire usually about + as long as the body-whorl [fera. and the latter much narrower than high .. L. peregra canalt- b. Shell not higher than 15 mm., fragile, with the spire always less than 1 the height of the body-whorl, which is as broad or nearly as broad as high. i Spire very small; upper surface of body-whorl transverse, almost at right angles to the main axis; deeply depressed round suture Br Fay A ii Spire small; upper margin of body-whorl cutting main axis obliquely, less depressed round suture. L. tenera euph- vatica. The key applies only to adult shells; for young shells our figures in the plates issued with this paper and our former one 1 Cf. p. 41, Rec. Ind. Mus. XVIII. The specimens now before us show that the mouth may be, and probably always is in the adult, more than 3 as long as the shell. 2 On p. 45, line 19 of this volume, the word ‘‘ greatly” has slipped out be- tween ‘‘not’’ and “ swollen.’’ Ig19.] N. ANNANDALE & B. PRAsHAD: Gastropod Molluscs. 107 on the molluscs of Seistan may be consulted (see plates vi, vii, Xiii, xiv of this volume). Limnaea gedrosiana, Annand. and Prashad. 1918. Limnaea subpersica, Annandale, Rec. Ind. Mus. XV, p. 146, Plex Ho he 1919. Limmaea gedrostana, Annandale and Prashad, Rec. nd. Mus. XVIII, p. 48, pl. vil, figs. 2-4. There is in Captain Boulenger’s collection a fairly good series of specimens in spirit which we cannot separate from our recently described species. The shells, however, though not thicker or less fragile, possess much stronger longitudinal ridges on the body- whorl than specimens from Seistan or Baluchistan. ‘There is no spiral sculpture. The mouth of the shell is also as a rule a little narrower, but this difference is hardly beyond the limits of normal variation and is not so great as that observed between shells from Baluchistan and those from Seistan. The largest shell is 10 mm. high and its maximum diameter is 7 mm. The specimen recently figured by one of us provisionally as L. subpersica, Locard, is a very young shell of this species. The radula is so variable in L. gedrosiana that it cannot be regarded in this species as possessing sound diagnostic characters. In a specimen from Mesopctamia it is very like that of some individuals from Baluchistan. The genitalia resemble those of the Seistan form figured and described by us in the original description of the species. Some differences exist, but these are due to the fact that the Seistan specimen we figured was abnormal in certain respects, as is borne out by dissection of another specimen from the same country. This specimen was found to have the genitalia quite similar to those of specimens in the present collection. The abnormality in the indivi- dual figured consisted in the large development of the accessory gland and in the poorer development of the hermaphrodite gland, its duct and the uterine duct; all these latter structures are much better developed in normal specimens, while the accessory gland is usually a small structure. The proximal part of the vas deferens also is rather thicker in normal specimens. It is clear, therefore, that individual differences must be looked for in the genitalia as well as in the shell and radula of species belonging to this group of Limnaea. The precise locality of Capt. Boulenger’s specimens is given by him as ‘‘ higher reaches of Khandag Creek, Basra, Mesopota- mia.’’ The species is not uncommon in swamp-deposits in the delta of the Euphrates. Limnaea bactriana, Hutton. (Ph SIV, fig. 33) 1919. Limnaea bactriana, Annandale and Prashad, Rec. Ind. Mus. XVIII, p. 45, pl: v,-figs. 1, 2% pl. vil, fig. 6. 108 Records of the Indian Museum. [VoL. XVIII, Three specimens in spirit collected by Capt. Boulenger in ponds connected with the Khandag Creek in a palm-grove near Basra seem to belong to this species. The shells are, however, thicker and are sculptured with curious flattened opaque ribs on the body-whorl. The form is also narrower, perhaps because the specimens had not reached their full growth, and the basal whorl of the spire is smaller and not so distinctly separated from the body-whorl. Otherwise the spire has the characteristic features of Hutton’s species. A shell is I0o°5 mm. long, and its maximum diameter is 6 mm. The three shells are all more or less broken. With more abundant material racial differences might perhaps be found between shells from the eastern parts of the range of the species and those from Mesopotamia. The radula of a specimen has the approximate formula 18.8.1. 8.18. The asymmetry of the cusps of the central tooth is very distinctly marked, as also is its tridentate character. ‘The inner cusp. of the lateral tooth though situated at a higher level than the outer is not much larger; all the cusps of the laterals, however, are pointed, differing in this respect from those of the Seistan specimens. The marginals have four to seven blunt cusps, all situated in the same straight line, and one or two small pointed cusps situated on the outer margins of the teeth near the base. The genitalia. Owing to paucity of material and to the speci- mens being very much contracted we are not able to add much to our previous account. The genitalia of a Mesopotamian speci- men resemble those of specimens from Seistan except that the uterine duct is much thicker at its commencement, the prostate is better developed and lies a little higher up on the male duct, which also is much thicker in its proximal part. ‘These differ- ences may be due at any rate in part to the state of sexual activity in which the molluscs were killed. Limnaea peregra race canalifera, Mousson. (BE XE, fies. 1,72.) 1874. Limnaea canalifera, Mousson, Fourn. de Conchyl. XXI, p. 41. 1918. Limnaea peregriformis, Annandale, Rec. Ind. Mus. XV, p. 165, Diese ares eal, This is much the largest form of Limnaea known to us from Mesopotamia, and the only one in which the shell grows more than 20 mm. long. The shell is also stouter and more coarsely sculptured than that of other species from the lower Euphrates. It has as a rule—though the fact is not mentioned in the original description—one more whorl, 7.e. five whorls instead of four. In dorsal view the shell is very asymmetrical bilaterally. The spire is acuminate, conical, vertical, exserted but short, but not so short as that of other shells of the same group from Meso- potamia, being at least (in adult shells) 1 as long as the body- whorl. Its whorls increase rapidly and evenly iu size and the spiral between them is oblique, linear and somewhat impressed. 1919.]| N. ANNANDALE & B. PrasHap: Gastropod Molluscs. 109 Each whorl is nearer the inner than the outer margin of the one that succeeds it. They are slightly swollen and slightly flattened above. The apex is small and rounded. The body-whorl is long but rather narrow and not very convex. Its upper surface is flattened and oblique but not angulate. The inner margin is Z-shaped, the upper half of the outline consisting of a somewhat flat curve. It then slopes in fairly abruptly and finally forms a broad projecting lobe, which corresponds with the inner anterior angle of the mouth. This lobe does not project so far sideways as the outline of the upper part of the whorl. ‘The outer margin of the whorl forms an arc of wide diameter and less than a semi- circle. The surface is not highly polished. It is ornamented with rather coarse and irregular longitudinal ridges, some of which may be called low costae, and striae. Minute spiral striae are also ‘present. ‘The first three whorls are, however, almost smooth. In ventral view the shell is ovoid. The body-whorl is swollen above and transverse at its upper margin. It disappears behind the mouth some little distance above the anterior extremity of the latter. The shell is very narrowly rimate. The mouth is long and somewhat expanded, ovate, nearly vertical, sometimes narrowly rounded or truncate both above and below, sometimes pointed above and rounded below. ‘lhe lip is thin and sharp and its curvature is often uneven but never highly convex. The callus is well developed above, joining the columellar border to the lip. The columella is long and twisted but not curved. The margin of the mouth is long, straight and vertical below the umbilicus. The columellar border is expanded and flattened over the umbilicus. The inner anterolateral border of the mouth is very slightly ex- panded and produced. The sculpture of the surface is not so well developed in this view as in the dorsal. As seen from above, the spire increases gradually but often irregularly and the suture becomes gradually more impressed until it is practically canaliculate as it approaches the outer lip. The radula has the dental formula 13.8.1.8.13. The central tooth though minute is distinctly bicuspid, having a small, sharp subsidiary cusp at the base of the main one, which is narrow and sharp. The laterals are tricuspid and their cusps are long, slender and sharply pointed. The lateral cusps are equal and a little smaller than the central one, but the inner cusp arises at a considerably lower level than the outer one. ‘The marginals are very uniform in structure. Each has three (or occasionally four) subequal cusps arranged in a slanting line. The only difference between those of the inner and the outer rows is that in the later the cusps are blunter and a little less regular in shape. The jaw is broad, dark and stout. The free margin of the upper part is broadly truncate and the internal surface is concave in the middle region. The animal, as seen in a highly contracted state and preserved in spirit, offers no particular external diagnostic features. The mantle is pale with large dark spots. IIO Records of the Indian Museum. [Vo.. XVIII, The genitalia. On comparing the genitalia of this form with those of L, gedrosiana' we find that the hermaphrodite gland is much larger and more lobose. The hermaphrodite duct is much longer, more convoluted and more swollen in its distal portion. The situation of the prostate is similar but the gland itself is larger and the upper part of the male duct above the prostate much more swollen; the penis-sac is also more elongate. The uterus is swollen in the middle but pointed at the two ends. The sper- matheca has a shorter duct than in L. gedrosiana, but nearly equal in length to the spermatheca itself. HG. Phy AL.G. Ac. Sits “Sp.D. \ << 2 PS. 3 Fig. 1.—Genitalia of Limnaea peregra canalifera, Mousson. Measurements (in millimetres) and Proportions of Shells. Height is xh 21 21 17 17 Maximum diameter BS: 15 13 3 13 1 Height of mouth .,, 4: 16 17'5 15 14 I2 Maximum diameter of mouth 8 8 i ff 5 Maximum diameter to height of shell a6 Teale [ie 06 Oh sess Maximum diameter to height of mouth a i ne) 1 OPT Pe Pale | kB PAL Height of mouth to that of shell see hoe in WBN Teel a2 eo mliaett hele Maximum diameter of mouth to that of shell ... i! PesLO7 1 oe HeplsO2en aly: Wag lon pega eae 1 Rec. Ind. Mus. XVIII, p. 40, fig. 5B (1919). 1919.] N. ANNANDALE & B. PrasHapD: Gastropod Molluscs. 111 The height of the shell is from 12 to 13 the maximum dia- meter; the height of the mouth at least twice its maximum dia- meter. The height of the shell is from 12 to 12 that of the mouth and the maximum diameter from 13 to 21 that of the mouth. Capt. Boulenger obtained a number of living specimens in the Khandag Creek at Basra. The specimen recently figured by one of us and assigned pro- visionally to L. peregriformis, Iocard, is a young shell of this species. It differs considerably from older ones. Adult shells only differ from a series from the Rhineland in the Indian Museum, labelled L. ovata var. inflata, in having the upper surface of the body-whorl flatter and more depressed round the suture. Limnaea cor, sp. nov. (PL LE figs, as) The shell is moderately small (less than 15 mm. high), thin and fragile, diaphanous, tinged with rose-pink when fresh, con- spicuously striate longitudinally, and remarkable for its short, erect acuminate spire, the base of which is deeply depressed, and its large, irregularly heart-shaped, transverse body-whorl, the maxi- mum diameter of which is practically the same as the height. In dorsal view the following particulars are apparent: The spire consists of three whorls and is not more than ¢ as high at the body-whorl. The apical whorl is minute and rounded, the second at least three times as deep and broad as the first, which is set upon it nearer the inner than the outer margin of the shell. It is somewhat convexly flattened above and its outer margin slopes gradually outwards and downwards, while its inner margin is vertical. ‘The third whorl, which is at least five times as large as the second, possesses the same characters in a more developed manner. ‘The suture of the spire is oblique, linear and depressed, but not very deeply. The body-whorl is bilaterally very asym- metrical and irregularly heart-shaped. Its upper surface is broad- ly flattened and decidedly angulate at its outer margin. The suture just above it is deep and broad. ‘The inner margin is con- spicuously Z-shaped. In its upper half it is very convex, repre- senting an arc of nearly a semicircle. It then slopes gradually inwards for some distance, and finally projects outwards to form a semicircular lobe corresponding to the outer anterior extremity of the lip. Just above this lobe there is a broad, slanting longi- tudinal depression on the surface of the shell. The outer margin of the whorl possesses a fairly even convexity in an arc greater than a semicircle, but its middle region is often more or less flat- tened and straight. The upper and outer part of the whorl is tumid, but the surface slopes somewhat abruptly towards both the lip and the anterior extremity of the shell. The sculpture on this surface is conspicuous even to a good naked eye, but has a very delicate appearance. It consists, as seen under a powerful lens, LIZ Records of the Indian Museum. [Vor. eee Sra 3 a | Wy dite -: dee, ead ela ee 26: siopenane iat eee Hey aoe = Li eal ie heath ee aie , ele oie eee See: : - OE : tC @e Q > hag one a ee = o cr a OS ive of _ 7 @is if i = SS . j rf wy aay = is ‘ 7 .- : aS n : ya \-4 ’ i: wes ite Cay ee - = - { aa jeg 3550 , ~ - oe PD. dawnt , Fews ' ; 7a ho @ sb. Geereasaay yee E Sader - Sa ed ty = US oy ea sate “me Se- hee rey Sight nates cies Sem ; ew ye 4 ae, wesrT thier afte Geetditan aie aah ahi aes Be ae EXPLANATION OF PLATE XIII. Limnaea cor, sp. nov. Fic. 1.—Type-specimen from marsh near Samara, Mesopotamia. Actual height 12 mm. 2.—Young shell from the same locality. Actual height 6 mm. Limnaea tenera euphratica, Mousson. Fic. 3,34.—Shell of the phase angustior, Mousson, from a marsh near Samara. Actual height 10°5 mm. 4.—Shell of the forma typica of the race from the same locality. Actual height 10°5 mm. 5.—Young shell from the same locality. Actual height 6°5 mm. REC. IND. MUS., VOL. XVIII, 1919. Plate XIll. A. Chowdhary del. LIMNAEIDAE OF MESOPOTAMIA ith 2 il whey fists: =} ; t Piste ¥ fteite Pi oe a agi, ; Asie ot eemehiiai ty: b-s\7 1 5) Seger 8 Pe nen ji olf Qo e ile EXPLANATION OF PLATE XIV. Limnaea peregra canalifera, Mousson. Fic. 1.—Shell from Khandag Creek near Basra. Actual height 22 mm. 2.—Young shell from the same locality. Actual height 7°5 mim. Limnaea bactriana, Hutton. Fic. 3.—Shell from Khandag Creek near Basra. Actual height 8 mm. Limnaea hordeum, Mousson. Fic. 4.—Adult shell (? subfossil) from the Gaud-i-Zirreh in the Afghan Baluch desert. Actual height 7 mm. 5.—Young shell (? subfossil) from the banks of the Euphrates at Nasariyeh, Mesopotamia. Actual height 4°5 mm. Plate XIV. REC. IND. MUS., VOL. XVIII, 1919. ‘VINVLOdOSAW AO AVGCIFVNWIT . [EP SeU PASH OY NAY ee a hay ae pit 7 ve y ris ~ : is - aa : - - - : 7 = ys el ev SET Ls |p3, oper ct Vira neu ON A NEW. GENUS AND) SPECIES. .OF COCCIDAK FROM NORTH-WESTERN. INDIA AND. EA STE RNG) PaeeR Sian. By E. ERNEST GREEN, F.E.S., F.Z.S. (With Plates XXI—XXII.) Genus Naiacoccus, nov. Characters as in Eviwm (of the subfamily Pseudococcinae) : but with an enormously elongated ovisac, within the anterior extremity of which the adult insect lies concealed. Type serpen- dimus, Green. Naiacoccus serpentinus, sp. nov. Adult female occupying the extremity ofa long, white, tubular ovisac which may form either a simple loop or be twisted into an irregular coil (see fig. 1). When uncoiled and extended the ovisac may attain a length of approximately one and three-quarter inches, the average length being well over one inch. Adult female, removed from the ovisac, dull slaty grey or pur- plish brown: broadly ovoid when viewed from above (fig. 20) : irregularly tumescent when viewed from the side (fig. 2a): the dorsai area of the abdomen contracted and transversely wrinkled, the posterior segments assuming a dorsal position. Antennae small, 7- to 9-jointed (see fig. 3), the proportionate lengths of the several joints varying considerably. The normal number is ap- parently 8 (fig. 3b, c, d), the larger number (fig. 3a) being excep- tional and produced by a fracture of the normal 4th joint, while the smaller number has presumably resulted from the fusion of two joints (fig. 3f). In some instances the division between the 7th and 8th is incomplete (see fig. 3e). Apex of terminal joint truncate or obtuse, with 5 or 6 stout hairs: each of the remain- ing joints usually witli r or 2 small hairs on one side. Limbs small but comparatively stout (fig. 4): the tibia and tarsus together markedly shorter than the femur and _ trochanter; coxa unusually large, especially in some examples (see fig. 40) ; digitules simple. Anal ring (fig. 5) with 6 stout setae: some smaller setae immediately above and below the anal aperture. Anal lobes inconspicuous, represented by two small rounded prominences on the dorsum—one on each side of, and slightly below the anal aperture (see fig. 2b). Spines of the form shown at fig. 5b occur in transverse series across the dorsum of the abdomi- nal segments—sparsely on the basal but more numerously on the 118 Records of the Indian Museum. [Vou. XVIII, posterior segments. Similar spines are clustered on the small anal lobes and on a corresponding tract on the preceding segment. Intermingled with the spines are many minute, obscurely trilocular pores, which occur (rather more abundantly) on the venter also. There are some larger circular pores near the posterior extremity of the venter. Length (under compression) 2°5 to 4mm. Breadth 2 to 3 mm. The early adult insect (before the production of the ovisac) is enclosed in a more compact, felted covering, of a grayish ochreous colour, which may be observed—even in older examples—as a pointed cap at the anterior extremity of the ovisac. The freshly deposited eggs are of a pale yellow colour, but become reddish before hatching. The number of ova produced by a single female probably amounts to several thousands. On a large arboreal Tamarisk (Tamartx articulata). Y.ahore. Collected by Dr. N. Annandalein May. ‘‘ T. articulata is one of the chief shade-trees in Lahore. Mr. Sundar Lal Hora, M.Sc., Research Assistant, Z.S.1., found the Coccid upon it in abundance in Octo- ber, but I have failed to do so on several occasions in January. INAS The extraordinary masses of tangled ovisacs must be very conspicuous objects on the trees, but might easily be mistaken for collections of bird droppings, st.ch as may often be seen on branches beneath the roosting places of flocks of sparrows. The insects excrete a considerable amount of viscid fluid which soon assumes a brownish colour. Lt.-Col. Stephenson (of the Government College, Lahore), to whom I am indebted for some excellent photographs of the insect and for fresh living material, informs me that—in nature—the long ovisac is always attached at each end, forming a simple loop. The subsequent tangled condition may be the result of wind. Naiacoccus serpentinus var. minor, nov. Distinguishable from the type by its smaller size and by the greater number of dorsal spines and dermal pores (fig. 6f). The circular pores of the venter are particularly conspicuous and are densely crowded on the posterior segments (see fig. 6g). The average size of the insect ranges from 1°5 to 3 mm.; but little difference can be observed in the length of the ovisacs which have precisely the same appearance as those of typical serpentinus. The antennae (figs. 6a-d) are shorter, the number of joints varying from 5 to 7, being usually reduced by complete or partial fusion. In some examples only 5 complete joints can be distin- guished ; but the 6-jointed from (fig. 6b, c) is the most frequent. On Tamarix stricta. Collected by Dr. N. Annandale in two localities several hundred miles apart, viz. :— No. 8740, “‘In desert north of Nasratabad in Seistan, Eastern Persia; November. The roots and stems of T. stricta, which 1919. | HK. K. GREEN: New Coccidae. IIg rarely grows larger than a small bush, are the chief source of fire- wood in Seistan.”’ No. 8745, ‘From Kushdil Khan, in the Pishin District north of Quetta, Baluchistan ; December.” oni seats (es Kees vole eared ae i Oe! (ewe es 7 : ae . Beir rnriin nbtaks seh an eae al ~ wr ey 5 . a Pie ee: a ta 2 ~ F 2 Mec ae te vie? aie an Sonos me 2 meres "Sig “s ) wT 7s) | ria Sar = r = a ee . = oe ra eh Aaa Ae ~ EXPLANATION OF PLATE XXI. Fic. 1.—Natacoccus serpentinus. Collection of ovisacs, on branch of Tamarix articulata; Xabout 2. (From photograph by Lt-Col. Stephenson.) ,, 2.—Natacoccus serpentinus. (a) Diagrammatic view from side of adult female ; (0) dorsal view; X 15. 55 3-—Naiacoccus serpentinus. (a-f) Different forms of anten- nae of adult female; X 220. REG. IND. MUS., VOL. XVIII, 1919 Plate XXI. 3. COCCIDAE FROM N. W. INDIA & PERSIA. ) Ore Atte ’ : ‘< net “ Oyen ASTOR EN ib Ar le cgegy Si abip ln ee 7 eee en Peet ii} ; 1 hie ite e “> 2) ay Pees a® a i erect Pre 317 it ur — | ‘aoe net, 20h pine et _ ey Ca eer gs 3(ht 4 Py y : yeh 0 ’ ¥ | Ons ‘ oliver aha s- Be pert clr) au) aes : ot : EXPLANATION “OF PLATE Xr Fic. 4.—Natacoccus serpentinus. (a-c) Various forms of mid leg of adult female; X 220. 5.—Naiacoccus serpentinus. (a) Anal ring of adult female and surrounding parts; XxX 130. (0) Spines and pores from dorsum ; X 450. 9 ,, ©.—Nazacoccus serpentinus. var. minor. (a-d) Ditferent forms of antennae of adult female; X 220. (e) Mid leg of adult female; xX 220. (f) Dorsal spines and pores; xX 450. (g) Circular pores, from venter of abdomen; X 450. Plate XXIl. MUS., VOL. XVIII, 1919. REG. IND. INDIA & PERSIA. COCCIDAE FROM N. W. ny MOTES ON TWO; COLEBC ELON S, OF BlL RDS FROM Sis TAN: By i. (C. STUART BAKERY P.L.S%,) FZ:S3 BOs CP ALOU. In the years 1903-05 a small collection of birds numbering 106 specimens was made by Mr. J. W. N. Cumming and other mem- bers of the Seistan Arbitration Commission and has been described by him in Journ. Bombay Nat. Hist. Soc. XVI, pp. 686-699 (1905). Another, consisting of only 31 specimens, was obtained by Dr. N. Annandale and Mr. S.W. Kemp on their recent visit to Seistan. Both these collections belong to the Indian Museum. As it was throughout advisable that the specimens should be named after comparison with the magnificent material available in the British Museum, especially with that in the Hume collection, the Director of the Zoological Survey of India kindly sent me the birds to work out The results show that the collection of 137 specimens contain examples of 79 species all of which belong to the Indian avifauna with the exception of Sylvia mystacea and Passer moabtticus vatit. The geographical affinities are Indo-Palaearctic, the races of resident birds nearly all belonging tothe Palaearctic rather than to the Indian forms; for instance Corvus cornix sharpit, Corvus frugilegus tschusw, Coracias garrula semenowt, Falco aesalon insignts. On the other hand a few sub-species, apparently resident, are typically tropical Indian, such as Gallinula chleropus parvifrons. As many recent alterations in names have been made since Blanford’s time, owing to discoveries by ornithologists of earlier names having priority, etc., it has been thought advisable to add in brackets the number of the bird according to Oates and Blan- ford’s Avifauna. This will it is hoped facilitate recognition by those field workers who might otherwise be puzzled. The field-notes in brackets over the initials N. A. have been added by Dr. Annandale. [The birds of Seistan, as might be expected from the peculiar conformation of the country, are, with few noteworthy exceptions, either water-birds or desert-birds. As our work in Seistan was chiefly connected with water and occupied only a few weeks in the middle of winter, such observations as we made on bird-life were necessarily meagre and concerned only the birds of the Hamun-i- Helmand, for a brief account of which the Geographical Introduction to this volume may be consulted. Thirty-four of the seventy-nine races and species enumerated by Mr. Baker may be classed as water-birds. What struck us most in a general way about these birds was their enormous numbers and the apparent paucity of food for them, Nevertheless, both ducks and the wading-birds 122 Records of the Indian Museum. [Vor XVile were exceptionally fat. We watched large flocks of the Black- tailed Godwit and other species continually grubbing in the mud where we ourselves found no living thing. An examination of their stomach-contents proved that they were feeding on the seeds of water-plants, which were doubtless lying ready to sprout when the flood-season began. Water-birds, especially the Coot and the Gray Lag Goose (of which no specimens were preserved) play an important part in the economy of the Saiyad or Hunters, a peculiar almost outcaste tribe, who live on the shores of the Hamun in primitive reed-huts and navigate its reed-beds in skiff-like rafts (tution) of bulrush leaves. In winter the most abundant and conspicuous birds in Seistan peculiar neither to the desert nor to the lake is the Rook. Enor- mous flocks, so large that we frequently mistook them for clouds in the distance, fly about the country at this season and settle on the ground in open spaces, often among tamarisk bushes round small pools in the desert. What they get to eat is a mys- tery. We were told that these flocks arrived about October and were popularly believed to eat up the “‘Hamun fly,” a blood- sucking fly (2? Tabanus or Haematopota) that spreads surra among camels and horses. It is quite possible that the birds do devour these insects, but it is also probable that their arrival coincides with the close of the flies’ season of aerial life. N.A.] 1, Corvus cornix sharpii Oates. (6). 1905. Corvus cornix, Cumming, of. cit., p. 686. Oo. No. 24685; Rudbar, R. Helmand, Feb. 03. 9. ,, 25492; Lab-i-Baring, ca. 1600 ft., 15.12.18. 2. Corvus frugilegus tschusii Hartert. (5). 1905. Corvus frugilegus, Cumming, op. cit., p. 686. 9. No. 24707; Nad-i-Ali on the Helmand, 13.3.05. Or 178 2488 Er) Kulak 920. 2405¢ @. ,, 25469; Lab-i-Baring, ca. 1600 ft., 15.12.18. om th 2547016 * af Ej oL2.LOe Of these Rooks the first two specimens have nearly completed their facial moult but still have all the nareal bristles present ; No. 3 has the face quite bare whilst the fourth is a young bird with the face fully feathered. [See notes at end of introduction. N,A.] 3. Argya caudata huttoni (Blyth). (105). 1905. Argya caudata, Cumming, op. cit., p. 686. o. No. 24880; Kuhak, 26.4.05. This is a typical huttont in appearance but is a very small bird with a wing of only 83 mm. ‘The striae on the breast are almost obsolete. Ww 1919. | BE. C. Stuart BAKER: Birds. 12 4. Tichodroma muraria (Linn.). (348). 1905. Tichodroma muraria, Cumming, op. cit., p. 686. 3 »P 2. No. 24877; Peshawaran, 20.12.06. 5. Sylvia mystacea (Ménétr.). 1905. Sylvia serdont, Cumming, op. cit., p. 687. @. No. 24883 ; Kuhak, 21.3.05. 6. Agrobates galactodes familiaris (Ménétr.). (359) 1905. Aedon familiaris, Cumming, op. cit., p. 686. 9. No. 24875 ; Kuhak, 22.4.05. 7. Prinia gracilis lepida, Blyth. (462). 1905. Printa lepida, Cumming, op. cit., p. 087. oe eNorn24679 >, Kuliak 2.10.04: 8. Lanius cristatus isabellinus, Ehrenb. (479). 1905. Lantus phoenitcuroides, Cumming (in part), op. cit., p. 687. ?. No. 24709 ; Kaod on the Asinki Canal, 28.3.03. g. Lanius cristatus phoenicuroides (Hemp. and Ehrenb.). (480). 1905. Lantus phoenicuroides, Cumming (in part), op. cit., p. 087. @. No. 24878, Kuhak, 22.4.05. The former of these shrikes is presumably only a eer but the latter is resident and breeds, for General R. Betham took many of its eggs and nests when stationed at Quetta. 10. Sturnus vulgaris poltaratskii, Finsch. (532). 1905. Sturnus menzhiert, Cumming (in part), of. c7t. p. 687. 905 In | P O. No. 24686; Hamun, Seistan, Jan. 04. O. ,, 25489; Lab-i-Baring, ca. 1600 ft., 12.12.18. 11. Sturnus vulgaris nobilior, Hume. (533). 1905. Sturnus menzbiert, Cumming (in part), op. cit., p. 687. a. No. 25472; Lab-i-Baring, ca. 1600 ft., 11.12.18. Hartert has recently written a long article on the races of Sturnus vulgaris (Novitates Zoologicae XXV, p. 327) but gives no key and in some cases does not even say how they can be discriminated. The deep colour of No. 25472, its almost black white-edged under wing-coverts and deep red wings satisfy me that this specimen must be referred to Hume’s nobilior. The other two birds are cer- tainly both poltaratskit with which menzbiert is now generally placed as a synonym. 124 Records of the Indian Museum. (Vou, XVIII, [Both No. 25472 and No. 25489 were shot at the edge of the Hamun near Lab-i-Baring in December. Small flocks of starlings are not uncommon in the fields at this season, at which ploughing is just commencing in Seistan. N.A.] 12. Oenanthe deserti atrogularis (Blyth). (626). Saxicola deserti, Cumming, op. cii., p. 688. 1Q05- e. No. 24712; Khwaja Ahmed, 3.4.03. oe. 24868 ; Khwaja Ahmed, Jan. 05. O. 25481 ; Lab-i-Baring, 7.12.18. oe. 25485 ; Lab-i-Baring, 8.12.18. 13. Oenanthe alboniger (Hume). (617). 905. Saxicoia albinigra, Cumming, op. ctt., p. 687. @. No. 247:0; Kaod on the Asinki Canal, no date. 2. ,, 25476; Lab-i-Baring, 7.12.18. [Very common with the last in the Seistan desert in December. WN. 4A.] 14. Oenanthe picata (Blyth). (618). 1905. Saxicola capistrata, Cumming, op. cit., p. 687. @. No. 24711 ; Kaod on the Asinki Canal, 26.3.03. This specimen belongs to the white-headed form which has generally been separated as Oenanthe capistrata. 15. Oenanthe isabellina (Cretzschem). (625). 1905. Saxicola tsabellina, Cumming, op. cit., p. 688. °. No. 24866 ; Kuhak, 31.3.04. @. ,, 24867; Kuhak, 3.4.04. 16. Phoenicurus erythronotus (Eversm.). (642). oO. No. 25480; In garden at Nasratabad, 18.12.18. This bird although not sexed is obviously an adult female; the wing is rather small, only 72 mm. 17. Monticola solitaria pandoo (Sykes). (693). @. No. 24713; Khwaja Ahmed, 7.4.03. This specimen is marked ‘‘ male” but is undoubtedly an old female. 18. Passer montanus dilutus, Rechm. (779). 1905. Passer montanus, Cumming, op. cit., p. 688. @. No. 24869; Kuhak, 20.4.05. 1919.] E. C. Stuart BAKer: Birds. 12 On 19. Passer moabiticus yatii, Sharpe. 9. No. 24714; Nad-i-Ali on the Helmand, 13-3.03. This specimen is not in adult plumage and has very little yel- low on the lower plumage but is almost certainly of this race of the Palestine Sparrow. 20. Motacilla alba personata, Gould. (829). 1905. Motacilla personata, Cumming, op. cit., p. 688. O. No. 24870; Kuhak, 28.4.05. @. ,, 25477; Lab-i-Baring, 13.12.18. Gs (23°25 482> = E3128: Of the three specimens the first is in full summer plumage, the two others in winter plumage. [Very common in the desert, and at the edge of the Hamun in December, N.A.] 21. Alaemon alaudipes pallida (Blyth). (854). 1905. Alaemon desertorum, Cumming, op. cit., p. 688. @. No. 24871; Kuhak, 21.9.04. Though marked ‘‘ male” this is probably a female and even for that sex is an unusually small one, wing 126, bill from front 25 mm. ‘This is considerably below normal measurements which are (vide Hartert) #7, wing 138-141 mm., ? , 128-130 mm. Larger series of pallida, Blyth and desertorum (Stone) might possibly prove that they are divisible but at present most ornithologists lump them together. 22. Galerida cristata? magna, Hume. (874). 1905. Galerita cristata, Cumming, of. czt., p. 689. @. No. 24874; Kuhak, 20.4.05. O. ,, 25478; Lab-i-Baring, 13.12.18. These birds are very small both having a wing of 106 mm., whilst Hartert gives the minimum for this sub-species as Ilo mm; on the other hand he gives the maximum for chendoola, the common Indian form, as 105 mm. As a matter of fact throughout the intervening country between the Northern and Southern areas the birds are intermediate both in size and general appearance. 23. Ammomanes deserti-iranica (Hartert). (878). 1905. Ammomanes phoentcuroides, Cumming, op. cit., p. 63¢ @. No. 24873; Khwaja Ahmed, 24. 4. 05. ‘Quite a typical specimen. 126 Records of the Indian Museum. [Vo,. XVIII, 24. Ammomanes phoenicura zarudnyi, Hartert. O. No. 25484; Lab-i-Baring, 14.12.18. I have compared this specimen with others in the Tring Museum as I was rather doubtful of some of those in the British Museum where the forms of phoenicura have not yet been thoroughly worked out. 25. Coracias garrula semenowi, Loudon and Tschusi (1024). 1905. Corracias garrula, Cumming, op. cit., p. 689. O. No. 24684; Khwaja Ahmed, May 1903. a, 24861 ; s 24.4.05. +) 26. Merops persicus persicus, Pall. (1028). 1905. Merops persicus, Cumming, op. ctt., p. 68a. @. No. 24862; Kuhak, 22.4.05. 27. Caprimulgus mahrattensis, Sykes. (10809). 1905. Caprimulgus mahrattensis, Cumming, op. cit., p. 690. 9. No. 24688; Helmand, no date. ? " ” 2.4689 ; ” ” 28. Bubo bubo bengalensis (Frank.). (1168). 1905. Bubo bengalensis, Cumming, op. cit., p. 690. @. Juv. No. 24864 ; Shaharistan, 6.5.05. This appears to be bengalensis, which has already been obtained in Afghanistan. The wings measure respectively, right, 380 and left, 370 mm., whereas turcomanus (Eversm.) has a wing varying be- tween 425 and 500 mm. From the date on which killed, and also from its juvenile appearance, it is evidently a bird locally bred and it raises an interesting doubt as to the specific or sub-specific value of some of the differences hitherto held to be sub-specific only between some of the nearest allies of Bubo bubo. We cannot have two geo- graphical races breeding in the same area and if turcomanus and ben- galensts do breed in the same, then they are good species and not sub-species only. 29. Athene noctua bactriana, Blyth. (1180). 1905. Athene bactriana, Cumming, op. cit., p. 691. a. No. 24863; Kuhak, 21.10.04. 30. Buteo ferox (Gmel.). (12309). O. No. 24722; “‘ Caught in Seistan,’’ Feb. 1904. An adult but beautifully pale bird in very fine plumage, ap- parently a male. 1919. ] EH. C. Stuart BAKER: Birds. 127, 31.‘ Circus macrurus (Gmel.). (1233). O. No. 24716; Band-i-Seistan, Feb. 03. O. ,, 24865; Shaharistan; no-date. Although these specimens have not been sexed the first is cer- tainly a » and the second almost equally certainly an adult female. 32. Circus aeruginosus (Linn.). (1237). O. No. 24717; Takht-i-Shah, March 04. One ns 247 10)3 Nadalis March 03. On e247 20.44 no date. 2. ,, 25494; Lab-i-Baring, 10.12.18. [The commonest bird of prey over the reed-beds of the Hamun. We saw one capture a small water-vole or shrew among the'reeds, .NV.A'3| 33. Falco aesalon insignis (Clark). (1263). O. No. 24724; Takht-i-Shah, March o4. The wing measures I90 mm. 34. Tinnunculus tinnunculus saturatus, Blyth. (1265). O. No. 24726; ? no date. The wing of this specimen measures 236 mm. 35. Columba livia intermedia, Strick. (1292). 1905. Columba intermedia, Cumming, op. cit., p. 691. @. No. 24850; Kuhak, 22.4.05. O. ,, 25463; Lab-i-Baring, no sex; December. Both these specimens are very typical va intermedia with the lower back quite grey, not white. These geographical races of Columba livia are generally very inconstant, partly due, no doubt, to crosses with domesticated birds but also due to the very wide range of individual variation which occurs. Major C. R. S. Pit- man informs me that even well North in Mesopotamia he found birds breeding together some of which were the true white rumped livia, whilst others were as dark as the darkest Indian intermedia. 36. Columba eversmanni, Bonap. (1295). 1905. Columba eversmannt, Cumming, op. cit., p. 691. O°, No. 24851; 2? 28.4,05. This is a resident Pigeon in Afghanistan and Persia. 37. Streptopelia turtur decaorta, Friv. (1310). 1905. Turtur risorius, Cumming, of. cit., p. 691. a, No. 24858; Khwaja Ahmed, 7.5.05. 128 Records of the Indian Museum. [VoL. XVIII, 38. Pterocles arenarius caudacuta, Pall. (1316). 1905. Pterocles arenarius, Cumming, op. cit., p. 691. oO. No. 25488 ; in desert S. of Lutak, 21.12.18. This specimen is not sexed but is a female adult. 39. Francolinus francolinus henricii, Bonap. (1372). 1905. Francolinus vulgaris, Cumming, op. cit., p. 692. Q. Juv. No. 24847; Kuhak, Augt. 04. Hartert in 1917 reviewed the races of Francolinus francolinus in Novitates Zoologicae and there points out that the small pale bird from Sind to Afghanistan must bear Bonapartes’ name henricit. The present specimen is very typical, a small, decidedly pale bird. 4o. Coturnix coturnix (Linn.). (1355). 1905. Coturnix communis, Cumming, op. czt., p. 692. @. No. 24848; Kuhak, 22.9.04. At. Zaponia parva parva, Bechst. (1392). 1905. Porsana parva, Cumming, of. cit., p. 6093. 2. No. 24859; Kuhak, 6.9.04. The wings are very imperfect, one moulting, and from the other several quills missing. They measure approximately 81 mm. In spite of its comparatively small size I have no doubt that it can be correctly assigned to this race. The colour generally is decidedly light and the bill is longer than in any specimen of pusilla in the British Museum Collection. 42. Porzana porzana (Lin.). (1394). No. 24696; ? no date or sex. 43. Gallinula chloropus parvifrons (Blyth). (1402). No. 24727; Band-i-Seistan, no date or sex. This specimen is somewhat intermediate between typical chloropus of Kurope and parvifrons of India, but is on the whole nearer the latter and should come under that name. 44. Fulica atra atra, Linn. (1405). 1905. Fulica atra, Cumming, op. cit., p. 693¢ 2? .. No. 25468; Lab-i-Baring, 9.12.18. 7. 5, 25471; ” Opiaeiuel, These two specimens, both fully adult, were obtained at Lab- i-Baring, Seistan, E. Persia, at an elevation of some 1600 feet. I cannot separate these in any way from typical aira. I919.] E.C. Stuart BAKER: Birds. 129 [Coot live in enormous flocks among the reed-beds of the’ Hamun. Large numbers are caught for food in nets stretched across channels in the reed-beds into which the flocks are driven by men on tutin or rafts. The nets are stretched on sticks stuck into the mud, their lower edge being in the water and their upper edge about 2 feet above the surface. N.A.] 45. Cursorius gallicus (Gmelin). (1423). 1905. Cursorius gallicus, Cumming, op. cit., p. 694. No. 24681; Khwaja Ahmed, April 1903, not sexed. 2.4082 ; 3 », 1903, ,, * 24683 ; Nadali, Helmand, March 1903, z ”) +) 46. Chettusia leucura (Licht.). (1438). 1905. Chettusia leucura, Cumming, op. cit., p. 694. @. No. 24677; ? no date. @. ,, 24678; Shahgul on Rud-i-Pariun, 23.3.1903. 9. ,, 24679; Nadali, Helmand, March 03. @. ,, 24849; Nasratabad, 6.5.05. The White-tailed Plover is an exceedingly common bird in suitable parts of Persia and has been found breeding in great num- bers during the Mesopotamian campaign, the nests being placed, as a rule, in small patches of raised land in the marshes. The eggs number three or four, most often the latter. 47. Aegialitis dubia jerdoni, Legge. (1447). 1905. Aegialitis dubia, Cumming, of. cit., p. 694. @. No. 24854; Kuhak, 1.4.04. @. 4, 24855; » 9.5.05. These specimens, which are both in full breeding plumage, be- long, as we should expect in birds obviously breeding, to the Indian race of Little Ringed Plover. The wings measure about IIo mm., the yellow base to the bill is very distinct, especially in one bird, and the black on the forehead is very narrow. (See Hartert and Jackson, Ibis 1915, pp. 531-3.) 48. Himantopus himantopus (Linn.). (1451). 1905. Himantopus candidus, Cumming, of. cit., p. 6y4. @. No. 24836; Kuhak, 3.9.04. Q. ,, 24837; Khwaja Ahmed, 24,4.05. @. ,, 24838; Kuhak, 3.3.04. Of the two females only the second is fully adult. 49. Limosa limosa (Linn.). (1456). 1905. Limosa belgica, Cumming, op. cit., p. 695. o@. No. 24852; Nasratabad, 6.5.05. o. .,,. 25491; Lab-i-Baring, 12.12.18: Ors Pe 25 AO2 > - E2:12.15. 7 130 Records of the Indian Museum. [VoL. XVIII, The specimen killed in May is in an interesting stage of plum- age, the breast being partly red and the plumage above also in rather more than half summer plumage. The two birds killed in winter are, of course, in winter plumage. [Feeding in large flocks at the edge of the Hamun in December. NA] 50. Totanus glareola (Linn.). (1461). 1905. Totanus glareola, Cumming, op. cit., p. 695. O. No. 24730; ? no sex or date. @. ,, 24856; Khwaja Ahmed, 5.5.05. 9.-,, 24857. Nasratabad, 7.5.05: 51. Totanus totanus eurhinus, Oberholser. (1455). oO. No. 24731; Seistan, Jan. 04, no sex. O. 5B) 24732; ” 7-1.04. ” On bs, © 247335 sgt) 0 date orsex. The Eastern form breeding from the Himalayas to Eastern Siberia has been separated as a geographical race under the name of eurhinus by Oberholser (Proc. U. S. Nat. Mus. XXII, p. 208; 1900) on the grounds of its greater size. He gives the average size for his new sub-species as follows: wing 163 mm., culmen 46°6 mm., and tarsus 47°2 mm. The three specimens noted above have the wings 157-165 mm., bills 43 to 45 mm., and tarsi 52 to 55 mm. They must all therefore be accredited to the Eastern rather than to the Western form. Specimens of both birds and eggs obtained in Tibet fully bear out Oberholser’s diagnosis of eurhinus. 52. Machetes pugnax, Linn. (1468). o. No. 24734; Shaharistan, Seistan, 1.3.03, not sexed. The wing measures 167 mm., large for a female and small for a male, but probably the latter not yet adult. 53. Tringa alpina alpina (Linn.). (1478). o@. No. 25474; Lab-i-Baring, 8.12.18. ALR 525475 s 7 SE2ES. OP.) 5 25470); uf BBN 6278 oe O.| 430125403) i 132.06. The wings of these four specimens vary from 108 to II5 mm. so that they must belong to the smaller Western sub-species rather than to the larger Eastern one. 54. Grus grus (Linn.). (1407). oO. No. 24728; Khwaja Ali, Feb. 03. This specimen is that of a not quite adult bird with the crown of the head still fairly well covered with feathers as in the young bird. 1919.] E. C. Stuart BAKER: Birds. 131 55. Larus ichthyaetus, Pall. (1489). 1905. Larus ichthyaétus, Cumming, op. cit., p. 695. O. No. 24691; Seistan, March 1904. In full breeding plumage. 56. Larus ridibundus, Linn. (1490). o. No. 24736; Seistan, Jany. 1904. (). 25464; Lab-i-Baring, 11.12.18. +18) 57. Larus gelastes, Licht. (1493). 1905. Larus gelastes, Cumming, op. cit., p. 695. @. No. 24841; Kuhak, 26.4.05. Or 2 40588 ox 20:4.05. 58. Larus argentatus cachinnans, Pall. (1495). oO. No. 24738; Landi Barech, Feb. 03. 59. Hydroprogne caspia (Pall.). (1498). 1905. Hydroprogne caspia, Cumming, of. cit., p. 696. 9. No. 24842; Kuhak, 26.4.05. 60. Sterna nilotica, Gmelin. (1499). 1905. Sterna anglica,, Cumming, op. cit., p, 696. @. No. 24839; Kuhak, 22.4.05. 61. Sterna hirundo (Neum.). (1506). 1905. Sterna fluviatilis, Cumming of. cit., p. 696. 9. No. 24860; Kuhak, 28.4.05. The wing of this bird measures 265 mm. 62. Sterna minuta gouldi, Hume. (1510). 1905. Sterna minuta, Cumming, op. cit., p. 696. 9. No. 24843 ; Khwaja Ahmed, 5.5.05. @. ,, 24844; Nasaratabad, 7.5.05. PF . 8248455 v 6.5.05. The wings of the two males measure 163 and 175 mm. respec- tively and that of the female 170 mm. All these specimens appear to be the same as the Indian S. m. gould: rather than true S. m. minuta. ‘They are identical in shade of grey on the upper parts and in colour of outer primaries with specimens from N.W. India. 63. Pelecanus onocrotalus onocrotalus, Linn. (1521). 1905. Pelicanus onocrotalus, Cumming, op. cit., p. 696. @s No 24730); ? no sex or date: 132 Records of the Indian Museum. [VoL. XVIII, A very typical specimen of this species with a bill fully 17 inches, or 435 mm., in length from forehead to tip of culmen. 64. Phalacrocorax carbo subcormoranus (Brehm). (1526). 1905. Phalacrocorax carbo, Cumming, op. cit., p. 696. oO. No. 24740; Seistan, Feb. 04. This is the form which Hartert has recently shewn (Novitates Zoologicae, XXIII, p. 294; I916) must bear Brehm’s name of subcormoranus. [Cormorants are captured or shot in large numbers by the Saiyad. ‘The down from the breasts is sometimes mixed with the soft wool out of which the felt hats worn by Persians are made. N.A.] 65. Ardea cinera, Linn. (1555). 1905. Ardea cinerea, Cumming, op. cit., p. 666. o. No. 24741; Khwaja Ali, Feb. 03; no sex. ©. ,, 24882 ;°Farrah Rud) Dec.704" These are both adult birds. 66. Botaurus stellaris, Linn. (1574). 1905. Botaurus stellaris, Cumming, op. ctt., p. 696. ©. No. 24846; Fartah Rud, ;Dec. 04.; No sex. On 24744008 Decors No sex: These are both adult birds with wings of 335 and 342 mm. respectively, but otherwise call for no remark. 67. Ixobrychus minutus (Linn.). (1570). 1905. Ardetta minuta, Cumming, of. cit., p. 696. o. No. 24687; Khwaja Ali, Seistan, April 03. An adult bird and evidently a male though it has not been sexed. The generic name [xobrychus, Billberg of 1828 antedates that of Ardetta, Gray of 1842, which must therefore be discarded. 68. Phoenicopterus minor, Geoff. (1575). 1905. Phoenicopterus minor, Cumming, op. cit., p. 697. o. No. 24840; Kuhak, June I904. A young bird, but exceptionally large with a wing of 13-6 inches (345 mm). 69. Cygnus cygnus (Linn.). (1578a). 1905. Cygnus musicus, Cumming, op. cit., p. 697. o. No. 24884; Hamun-iSabous, Seistan, Dec. 04. This is a very large specimen with a wing of 602 mm. (23°75 inches) ; bill r1z mm. (4°4 inches). Although not sexed it is un- doubtedly a male. 1919. | H.C. Stuart BAKER: Birds. 133 70. Tadorna tadorna (Linn.). (1587). 1905. Tadorna cornuta, Cumming, of. cit., p. 697. @. No. 25466; Lab-i-Baring, 14.12.18. oe 5, 24818; Seistan, 3.10.04. @. ,, 24819; Kuhak, Sept, 04. @. 4, 24820: Seistan, 3:10:04. Juv. ,, 24821; Kuhak, Aug. 04. iy 9p 24822 ; Seistan, no date: The first bird isan adult in full plumage, the next three are ducklings of about a month old, or rather less, and the last two are still younger. 7I. Querquedula crecca (Linn.). (1597). 1905. Nettiuwm crecca, Cumming, op. cit., p. 697. @. No. 24830; Kuhak, 3.10.14. @. .,, 25473 ; Labia-Bariies 1 r9e06, Wa. 2AQO': a SoTL 228: Of these three birds the first is in eclipse plumage, the second in full breeding plumage, whilst the third appears to be a female and nota male. [One of the commonest ducks among the reed- beds of the Hamun. JN. 4.] 72. Dafila acuta acuta (Linn.). (1600), 1905. Dafila acuta, Cumming, op. cit., p. 698. @. No. 25487; Lab-i-Baring, 8.12.18. In full, but not very bright, breeding plumage. [Another very common species in the reed-beds. JN. A.] 73. Spatula clypeata (Linn.). (1602). 1905. Spatula clypeata, Cumming, of. cit., p. 608. @”. 24832; Hamun-i-Sabari, 29.12.04. This drake is still in eclipse plumage. 74. Marmaronetta angustirostris (Ménétr.). (1603). 1905. Marmaronetta angustirostris, Cumming, of. ctt., p. 698. ©: No: 24823; Kuhak, juv:, 6.7.04. O- 248240: Fs . 6.7.04. Ges) 2AG25 ; pempadult. 22.4,05: OF. §3/\248260. Poi. 0.7048 O. ,, 24827; ,, » 6.7.04. Apparently the four ducklings, though all young birds recently hatched when obtained, are in two stages of growth, the first two being some days older than the other two. This little duck breeds freely from as far South as the Mackran coast and Sind throughout South, Central and Eastern Persia wherever the country is suitable, 134 Records of the Indian Museum, [Vou XVIII, 1919.) The two youngest birds of those above enumerated have the wing quills only just beginning to sprout. 75. Netta rufina (Pall.). (1604). 1905. Netta rufina, Cumming, op. cit., p. 608. 9. No. 24833; Farrah Rud, Dec. 04. 76. Nyroca ferina (Linn.). (1605). 1905. Nyroca ferina, Cumming, op. cit., p. 698. 7, No. 25465; Lab-i-Baring, 10.12.18. chee pe Asy: Oper - LO:L2U6, 77, Glaucionetta clangula, Linn. (1610), 1905. Clangula glaucton, Cumming, op. cit., p. 699. o. No. 24829; Hamun, Seistan, Dec. 04. In full breeding plumage. Stejneger has shown that we cannot use the generic name Glaucion for the Golden-eye and has substituted Glauctonetta in its place (Proc. U.S. Nat. Mus. VIII, p. 409; 1885). 78. Podiceps cristatus (Ijinn.). (1615). 1905. Podiceps cristatus, Cumming, op. cit., p. 699. 9. No. 24692; ? no date, adult. 9." |, 24604; 2 no'date;quv. 2. ,, 246935 ° no date, adult: 79. Podiceps fluviatilis albipennis (Sharpe). (1617). 1905. Podiceps albipennis, Cumming, of. cit., p. 699. o. No. 24835; Seistan, juv., no date. @. ,, 25486; Lab-i-Baring, 11.12.18., adult. [Very common among the reed-beds of the Hamun. N.A4.] NOARE ON TEE OCCURRENCE Of LHe TCH LM NA gens NAOT AEC AT Ni, SEISTAN AND Pen G oy AN BALUCH DESERT. By N. ANNANDALE, D.Sc., F.A.S.B., Directoy, and AMIN-UD-DIN, M.Sc., Research Assistant, Z oological Survey of India. Makki in Western Baluchistan, close to the Afghan border. When asked why he did this, the man said that he was afraid of leeches. No leech was seen in this spring, but many were observed at Robat close to the point at which the Afghan, Baluch and Persian frontiers meet, and also at Hurmuk across the last. Specimens were not captured at N awarchah, a place some distance north of Hurmuk and well within the district of Seistan, on the tongue of a horse. The specimen js small, being only 2°5 cm. long and ‘5 em. broad as preserved in 90% alcohol, but it agrees in all essentials with small specimens of L. nilotica (Savigny) from Palestine. It belongs to the colour-form in which the dark markings are obscure or obsolete. The posterior sucker is of a characteristic size, the 1 Boll. Mus. Zool. Torino IX, No. 189, p. 43 (1894). * Parasitology I, Pp. 282 (1908). ° Fudges VIT, 6, Frazer in his Folk-lore in the Old Testament adopts a ritualistic explanation. 136 Records of the Indian Museum. [Vo1. XVIII, 1920.] diameter being about 5 mm. and that of the anterior sucker only 2mm. L. nilotica is easily distinguished from the common Indian species, Limnatis (Poecilobdella) granulosa (Savigny), by the ab- sence of the colour-pattern characteristic! of the sub-genus Poeci- lobdella and by its smaller size. 1 See Blanchard in Weber's Zool. Ergeb. Niederland. Ost.-Ind. 1V, p. 346, figs3. OD MOTES-.ON SOME) a5 Pe TiC: SPE CEES: OF PALINGENIA (ORDER EPHEMEROPTERA). By F. H. Gravey, D.Sc., Asst. Superintendent, Zoological Survey of India (now Superintendent, Government Museum, Madras). (With Plates XVIII—XX.) The Ephemerid genus Palingenia has attracted the attention of many naturalists on account of the way in which immense swarms of adults hatch and die annually during the evenings of a few consecutive days only.! The larvae are fossorial. They have six pairs of double gill- plumes arched upwards over the back and protected by five pairs of lamellae which are covered on the outer side by long hairs. According to Swammerdam (1758, p. 109), these lamellae are “ oars that serve the creature for swimming.” Male larvae may be distinguished from females by their larger eyes and, in the later stages, by their longer caudal appendages and developing forceps. The genus has been provisionally divided by Eaton (1883, p. 23) into three subgenera as follows:—‘‘ Palingenia (typical), Burmeister, containing European and Western Asiatic species; Anagenesia containing Indo-Malay and a Siberian species; anda nameless subgenus containing Brazilian species,’ concerning the adults of which scarcely anything appears to be known.” Eaton’s system of reference to the venation of the wings ® has been adopted throughout the following notes in order to facilitate comparison with his monograph. In the figures this system is supplemented by that used in Comstock’s book ‘‘ The Wings of Insects ’’ (New York, 1918). Subgenus Palingenia, Burmeister, s. stv. Adult with the fore-tarsus of the male about 2} times as long as the femur ; the praebrachial nervure (6) of the forewing forked beyond the middle; two conspicuous couples of longitudinal nervures (midway between 4 and 5, and 5-6) proceeding to the terminal margin of the fore-wing (pl. xx, fig. 21); the forceps with a long slender basal joint grooved on the inner side and (? always) at least five shorter terminal ones (pl. xx, figs. 22-3). | For references see Eaton, 1883, pp. 24-28; also Swammerdam, 1758, p- 104, concerning references by more ancient writers. 2 The larva is figured by Eaton, 1883, pl. xxv, figs. 20-24. ® Eaton, 1883, p. 4. 138 Records of the Indian Museum. [VoL. XVIII, Larva with the dorsal margin of the fore-tibia ( ? always) strongly toothed (pl. xviii, fig. 4). Palingenia (s. stv.) ? longicauda, Olivier.! Palingenta sp. (? robusta), Needham, 1909, p. 191, pl. xx, fig. 8.2 Adult (pl. xx, figs. 21-23).—Nine pinned specimens from Seis- tan, all in bad condition. One of these was erroneously referred by Needham to the subgenus Anagenesia, to which P. robusta, Eaton, belongs, with the remark :—“ The species will be recognized by the male forceps....which is different from the forceps of any species that has been figured hitherto.” His figure (1909, pl. xx, fig. 8) differs from Eaton’s figure (see 1883-8, pl. i. Ia o) of the male genitalia of P.longicauda in having five instead of six terminal joints to the forceps. But Cornelius (1848, pp. 28-29) states that P.longicauda has only three such joints and shows only three in his figure (1848, fig. 4K), though the terminal one could be almost better interpreted as two joints than as one but for the accompanying statement, and might even be com- posed of three. The probability, it seems to me, is therefore that normally the forceps of P. longicauda consists of at least five terminal joints and one long basalone. Inthe specimens now under consideration most of the forceps are broken. In one the terminal joints appear to be six in number, the last being very small and imperfectly separated, in another they appear to be seven, the terminal one again being very small. Unfortunately no European specimens are available for comparison. The colour of the Seistan specimens appears to be duller and more uniform than in European specimens, but this may be due to their poor state of preservation. The species probably occurs also in Mesopotamia, as Major Connor writes that he ‘‘saw millions of the large Mayflies on the Euphrates at about the beginning of April. They were being eaten up by the ordinary Caspian river tortoise as they lay in heaps in eddies and slack water. They swarm in the river even as far down as Basra.” Larva (pl. xviii, figs. 1-4).—Six males and eight females, none full-grown (length, excluding mandibles and caudal appendages, 21-25 mm.), from thick mud of pools in the bed of the Randa stream (otherwise dry) four miles northwest of Jellalabad, Seistan. Dis. G ti oe Dr. Annandale has supplied the following note on the finding of these larvae :— 1 For P. longicauda, see Eaton, 1883, p. 24, pl. i, fig. 1@ ; and Cornelius, 1848, pp. 22-20, figs. 4-4K. ? The nymphs referred by Needham (doc. cit.) to this genus belong, in all probability, to the genus Ephemera. (See Vayssiére, 1882, pp. 38-42, pl. i, figs. 3-7; Eaton, 1883-8, pp. 58-59, pl. xxx, figs, I-19; and Klapalek, 1909, pp-29-30). They are very different from those o Palingenta. 1920. | F. H. GrRAvELy: Asiatic Palingenta. 139 ‘“ The larvae of Palingenia were collected about the end of November at the edge of small pools of very foul water left by the retreat of the annual floods in the bed of the Randa stream near the ruined city of Jellalabad, some twelve miles north of Nasratabad, the capital of Seistan. This stream is filled with water only in flood-time. For some considerable distance south the country consists of a flat barren plain the surface of which in winter is formed of bare and extremely hard grey clay. It isin fact one of those shallow basins, so common in Seistan, which are flooded every normal year by the rise of the Helmand. ‘Towards the edge of this basin we noticed that the whole ground was pitted with little holes as though a rather narrow pen-holder had been re- peatedly thrust into it. We were unable to account for this phenomenon until we examined the edge of the pools, where each hole was occupied by a Palingenia larvae. The mud was here fairly soft but was caking rapidly and the larvae, the gills of which were pressed tightly against the sides of their abdomens, were, though still living, apparently being gradually asphyxiated. The foulness of the water was due partly to the presence of large numbers of dead fish and partly to the fact that the several tribes of the district watered their flocks of sheep and goats at the pools. Doubtless the Mayflies of the same genus collected in large numbers in Seistan by the officers of Sir Henry MacMahon’s Com- mission were captured in spring or summer.” The identity of Dr. Annandale’s larvae with these adults from Seistan is extremely probable, but has not been definitely proved. The larvae appear to be identical with those of P. /ongicauda des- cribed by Swammerdam (1758) and Cornelius (1848). The caudal appendages are not very well preserved ; they seem to be of almost equal length in both sexes, but this is probably due to their not being fully developed. ‘These larvae differ from all other Palin- genta larvae yet known in having the dorsal (outer) margin of both the mandibles and fore-tibiae very strongly toothed. I have not been able to distinguish the second spine figured by Cornelius on the blade of the maxilla; but this may be a variable character (see below, p. 142). The labial palps bear hairs and spines like those of P. robusta. The front legs closely resemble those of P. robusta. The group of spines on the inner side of the lower distal angle of the tibia is, however, composed entirely of simple spines somewhat longer and slenderer than the very stout simple spines of the outer half of the group in that species. The serrate spines on the tarsus, on the contrary , are somewhat coarser ; they are also more numerous. The middle legs differ from those of P. vobusta chiefly in the pre- sence of a large conical tooth on the dorsal surface of the distal end of the tibia. The hind legs differ chiefly in having the outer distal angle of the tibia less produced. In both middle and hind legs the spines on this angle resemble those found in the same position on the front legs; they are much thinner than in P. robusta, 140 Records of the Indian Museum. [VoL. XVITI, The gills and their protective lamellae have been somewhat crushed together, but appear to resemble those of P. robusta. Subgenus Anagenesia, Eaton. Adult with the fore-tarsus of the male shorter than the femur; the praebrachial nervure (6) of the fore-wing forked before the middle ; three conspicuous couples of longitudinal ner- vures (midway between 4 and 5, 5-6, and midway between 6 and 7) proceeding to the terminal margin of the fore-wing (pl. xx, fig. 24) ; the forceps with a long flattened basal joint and two shorter termi- nal ones, the latter sometimes with a small and imperfect third joint between them (pl. xx, figs. 17-20 and 25). Larva with the dorsal margin of the fore-tibia ( ? always) without conspicuous teeth (pl. xviii, fig. 8). Palingenia (Anagenesia) lata, Walker. Adult (pl. xx, fig. 17).—The Indian Museum collection con- tains pinned males from Sibsagar. See Eaton, 1892, p. 407. Palingenia (Anagenesia) picta, n. sp. Adult (pl. xx, figs. 24-25).—Two pinned males each about 22 mm. long, from Kapit, Sarawak, 24.vil.10. This species differs from P. lata in its slightly larger size and in having the terminal joint of the forceps distinctly smaller than the penultimate. The most striking difference, however, is in colour, P. /ata being of an almost uniformly dull brown colour, whereas in P. picta the general colouration is bright reddish or yellowish brown, while the head is more or less black between the eyes, except for a strongly marked median brown line on the vertex; the mesonotum is almost equally dark; the wings are whitish with yellowish veins, the anterior pair having infuscate margins, especially in front and at the tips; the dorsal plates of the abdomen are somewhat. dark, except for a narrow posterior border. ‘The relatively pale prono- tum stands out in marked contrast to the dark head and mesono- tum. Palingenia (Anagenesia) robusta, Eaton. Palingenia robusta, Eaton, 1892, pp. 407-408. Adult (pl. xx, figs. 19-20).—The imperfect type male from Cachar, two dry and six spirit males and three spirit females from the Dikko River, Nazira, Assam. ‘The Dikko specimens were sent by Mrs. Maxwell, with the following information. They emerge annually at about the end of October, and for three or four days float down the river in countless millions. The natives say that they also appear on other rivers, such as the Desoi, Desang and Dihing, and that they come out at and under the edge of the water in the shallows after the rivers have left the hills and where they run through silt only ; but Mrs. Maxwell says that so far as 1920. | F. H. GRAvELY: Asiatic Palingenta. I4Ir she knows they do not occur on rivers actually rising in flat dis- tricts. The natives believe that until they have appeared there is always a chance of further floods and that consequently it is no use building the temporary bamboo bridges which they put up every cold weather until these ‘‘ pani-pooka’’ (water insects) have gone. The caudal appendages of the males were 3 inches long and semi-transparent when fresh. The insects are so light and hollow that they cannot be kept under water; when just out they are white or creamy and look like foam when blown together by the wind. All the specimens collected as adults were males; the females were caught as nymphs and hatched in captivity. This species is of about the same size as the last, but lacks its rich warm colour. The general colour of the male is whitish, with the upper surface of the head, mesothorax and posterior end of abdo- men tinged with dull brown ‘The margins of the fore-wings are natrowly tinged with the same colour. The second joint of the foretarsus is more distinctly longer than the first and third than in either of the two preceding species. The forceps seems normally to consist of the one long basal and two terminal joints character- istic of the subgenus ; but the second of the latter joints, which is fully as long as or even a little longer than the first, is often divided quite definitely into two near the base, at least on the outer side. The legs and caudal appendages of the female are smaller and feebler than those of the male, especially the caudal appendages, and the dorsal surface of the body is much darker in colour, being of a dull brown tint. ‘The wings are whitish as in the male. Larva (pl. xviii, figs. 5-8, pl xix, figs 9-16).—One male and two female cast skins, found floating in the surface water of the Dikko River, Nazira, Assam when adults were emerging, Oct. 26-30, 1918; three males and one female insect from the same locality, Oct. 1919.! The total length (excluding the mandibles but including the caudal appendages) is 45-47 mm. in both sexes. The caudal appendages are 13 mm. long in the male and only g in the female, the body being therefore 4 mm. longer in the female than in the male. The teeth on the anterior margin of the head are somewhat more scattered than in P. longicauda. The mandibles (pl. xviii, fig. 7) are very hairy, long and slen- der and are upturned distally ; they have a number of small teeth, much smaller than those of P. /ongicauda, scattered along the basal 2 of the upper margin. ‘They are intermediate in form between those of Palingenia (s. str.) longicauda (see Cornelius, 1848, fig. 3B) and those of Ephemera vulgata (see Eaton, 1883-8, pl. xxx, figs. 7-8). They are very different from the mandibles of the Palingenta (Anagenesia) larva from Ceylon’ figured by Eaton ees 8, pl. XXV_ Eire aeeeaons: is Seitea from the cast skins, as the larvae were not received till it had been completed. 2 No adult from Ceylon was known to Eaton, Banks (1914, pp. 612-613), has since described Palingenta | Anagenesia) greeni trom there. 142 Records of the Indian Museum. PViOn Xe VEEL. figs. 8-9), which are much shorter and stouter, are distally some- what wedge-shaped and irregularly dentate instead of slender and pointed, and appear to be without the two small laminar teeth found below the molar tooth in the present species and in P. longi- cauda and E. vulgata. ‘here is little or no difference between the teeth on the right and left mandibles. The maxillae (pl. xix, fig. 9) and labium closely resemble those of Eaton’s Ceylonese larva but are less pointed, especially the maxillary palps. The blade of the maxilla bears two spines dis- tally as in Cornelius’s figure of P. longicauda, but the lower one is more transparent than the terminal one, and is sometimes very hard to distinguish and possibly absent. The labial palps (pl. xix, fig. 10) bear a number of transparent stout curved spines at the distal extremity and a tuft of spines on a tubercle at the base of the penultimate segment. The front legs (pl. xviii-xix, figs. 8 and 11) are clothed with hairs and spines arranged in very definite series. ‘The transverse line of hairs at the base of the femur and the two transverse lines at the base of the tibia are finely pectinate (pl. xix, fig. 14). The lateral filaments on these hairs are extremely minute, but probably form two series more nearly at right angles to each other thanin one plane. The ventral aspect of the outer distal angle of the tibia bears a group of strong spines of which the outermost half are somewhat less stout than the innermost, and are coarsely biserrate Owing, however, to the angle which the two rows of serrations bear to one another not more than one of them can be clearly seen from any one point of view (pl. xix, fig. 12). The tarsus bears a number of more slender biserrate spines below its outer margin. Their serra- tions are more nearly in one plane (pl. xix, fig. 13). The remaining hairs and spines are simple. On the last two pairs of legs the lines of pectinate hairs are absent, and there are no serrate spines, simple spines and hairs being more extensively distributed in place of them. The spines are strongest and most numerous on the third pair of legs. Except for a group of very stout curved spines of moderate length on the ventral aspect of the outer distal angle of the tibia, the spines are confined to the dorsal surface (pl. xix, figs. 15-16). The first abdominal segment bears a pair of gills but no protec- tive lamella. The five following segments bear both gills and lamellae. The three remaining segments are without appendages. The first two of these bear spines and hairs laterally. The last has hairs distributed over almost the whole of its dorsal surface; these hairs are thick behind, but there are no spines at all com- parable in strength with those on the two preceding segments. Each gill consists of two plumes, one situated behind and to the inner side of the other. The former is of about equal size on all segments, and being directed backwards it conceals the latter, which is considerably smaller—more so in the anterior than in the posterior segments. All five protective lamellae are of about equal size. Each 1920. | F. H. GRAVELY: A static Palingenia. 143 consists of a finger-like process of the body-wall, bare on the inner side but fringed and entirely covered on the outer side with very long hairs. Palingenia (Anagenesia) minor, Eaton. Palingenia minor, Eaton, 1882, p. 408. Adult (pl. xx, fig. 18).—Two specimens labelled ‘‘ Karachi Museum ”’ and one from Nattor, whichis in the Rajshahi Division of Bengal, were described by Eaton from the Indian Museum col- lection. A specimen from Sara Ghat in Bengal, and one from Pakokku in the oil-fields of Upper Burma, have since been added. The species would seem, therefore, to be very widely distributed over the Indian Empire. LIST OF PAPERS QUOTED. 1758. Swammerdam, J. ‘‘ The Book of Nature, or the History of Insects,’’ translated from the Dutch and Latin original edition by Thomas Flloyd (Ephemera = Palingenta longt- cauda), part i, pp. 103-119, pl. xill-xv. 1848. Cornelius, C. ‘‘ Beitrége zur nahern Kenntniss der Palin- genia longicauda, Olivier,’’ 38 pp., 3 pl., Elberfeld, 1848. 1882. Vayssiere, A. ‘‘ Recherches sur l’Organisation des Larves des Ephémérines.”” Ann. Set. Nat., Zool. (6) XIII, 1882, pp. 1-137, pl. i-xi. 1883-8. Eaton, A.E. “A Revisional Monograph of Recent Ephe- meridae or Mayflies.” Trans. Linn. Soc. London (2, Zool.) III, 1883-8, 352 pp., 65 pl. 1892. Eaton, A. E. ‘‘ Notes on some Native Ephemeridae inthe Indian Museum, Calcutta.” Journ. Asiatic Soc. Bengal, LX (II), 1891 (1892), pp. 406-413. 1909. Klapalek, F. ‘‘ Ephemerida, Plecoptera [Lepidoptera’”’ by K. Grtinberg] in Susswasserfauna Deutschlands, Jena, 1909, 163 pp., 260 text-figs. 1909. Needham, J. G. ‘‘ Notes on the Neuroptera in the Collec- tion of the Indian Museum.” Rec. Ind. Mus. III, 1909, pp. 185-210, pl. xix-xx1. 1914. Banks, N. ‘‘ New Neuropteroid Insects, Native and Exo- tic.” Proc. Ac. Nat. Sci. Philadelphia, UXVI, 1914, pp. 608-633, pl. xxviii. EXPLANATION ‘OF, PLATE, Vili: Palingenia larva from Seistan. Fic. 1.—Head of male. 2.—Head of female. 3.—Right mandible. 4.—Leit front tibia and tarsus from above. 33 Palingenta robusta, larva. 5.—Caudal appendages of female. 6.—Cast skin of male. 7.—Right mandible. 8.—Left front tibia and tarsus from above. REC. IND, MUS. VOL. XVIII 1920. Plate XVIII. f \ {] Ny | \ } \ {] \ f \ {] \) () () \ \Y Ay, ,\N ) e) \ a' LJ WN q OQ) D. Bagchi del. PALINGENIA LARVAE _ ' 1 ‘ ¢ 1¢ e & ; aa) . ~ " ears 4) EXPLANATION OF PLATE XIX. Palingenia robusta, larva. Fic. 9.—Right maxilla from below. 10.—Right labial palp from below. 11.—Left fore leg from below. 12. », 13. ¢Spines and hairs from fore leg. eee 15.—Left middle tibia and tarsus from above. 16.—Left hind tibia and tarsus from above. REG. IND. MUS., VOL, XVIII 1920. Sas beter | (Ae wily Mi MA i Hi HY \\} U/) 3 18. AN ) YY \ \\ SS SN ~ S Z Vil y lA y G SS iH I, Yj eS oe ZZ, YS SS SS — \ \\ MRA “\ SS AY 14, D. Bagchi del. PALINGENIA LARVAE, Mai rb aad coe - apt oe EXPLANATION OF PLATE XX. Fic. 17.—Palingenia lata, terminal joints of forceps of male. ,, 18.—Pahingenia minor ,, 5 fe 5, 19.—Palingenia robusta ,, - a ,, 20.—Palingenta robusta ,, =H 3 21.—Palingenia from Seistan, right fore wing.! 22.—Palingenia 5 , forceps of male. 23.—Palingenia ae , terminal joints of forceps of male. 24.—Palingenia picta, male from above.' » 25.—Palingenia picta, terminal joints of forceps of male. 1 The numbers at the ends of the veins are the symbols used by Eaton in his monograph. The letters are those used by Comstock in his book on ‘‘ The Wings of Insects’’ (see above p. 137). REC, IND.,MUS., VOL. XVIII 1920. Plate XX. Tee aoe eases == es 5 ART ea weny AW, pee ANS BP ORI UL) g Uys RK Us 78. s = 23 i fo x f | j fj Y reg di [es ; ? } RO : ill = Se ~ Se : il ib Pee Ei, VE Zi - be ee AK f i \\ PALINGENIA. D. Bagchi del a el dat h pee ee a 0 Set Pe ree a i 4 WP falta Wi ULL ok Mill Melia eal mist OF EN TOMOST RA CA COM DE CLE Do EN SEES TAN AND “THe BAY CH Ser SB Ri By RoBERT GuRNEY, M.A., F.Z.S. [The few Entomostraca mentioned in Mr. Gurney’s list are the only fresh- water Crustacea we saw in Seistan except the crab Potamon (Potamon) Pota- mios gedvostanum, Alcock, and certain small Ostracods abundant in springs in the desert. Owing to some accident no specimens of the latter were collected. The crab was found in considerable numbers buried in the mud at the bottom of pools of foul water in the bed of the Randa stream near Jellalabad. No trace of it was, however, observed in the Hamun-i-Helmand. Mr. S. W. Kemp has compared a series of specimens from Seistan and from Quetta, where it was seen in an active condition in winter, and can find no local difference. Inthe hill-country of Baluchistan, which is not represented in the collection examined by Mr. Gurney, at least two species of Amphipods are common in, springs.— N. Annandale. | (1) NASRATABAD, SEISTAN. Pools on Parade ground. 27-xi-18. These pools were pits from which clay had been dug for bricks. There was a luxuriant growth of Zannichellia, but the water was very foul, being visited by large numbers of donkeys and camels. Daphnia magna Straus. Abundant. Simocephalus vetulus O.F.M. Abundant. Cyclops strenuus Fischer-Sars. me leuckartt Claus. Eucypris clavata Baird. Common. Ilyocypris bradyi Sars. Potamocypris villosa Jur.. Common. (2) NASRATABAD, SEISTAN. Irrigation channel. Ig-xii-18. Leptestheria tenuis Sars. A single male specimen. . I do not feel at liberty to dissect this specimen, but it agrees so closely in external appearance with L. tenuis that I have no hesitation in so naming it. (3) LAB-I-BARING, SEISTAN. Channelsin reed-beds in Hamun. Io-xti-18. The collection was made in small pools about 8 feet deep, blocked with Potamogeton pectinatus. Daphnia longispina var. rosea Sats. Rare. Certodaphnia puichella Sars. Common. Ephippial females seen. aa reticulata Jur. A few. Simocephalus vetulus O.F.M. Bosmina longirostris O.F.M. Two seen. Cyclops viridis Jur. One female only. Herpetocypris reptans Baird. One only. Free ephippia of Daphnia magna. 146 Records of the Indian Museum. [Vo.. XVIII, 1920.| (4) YAKMATCH, W. BALUCHISTAN. 13-x1-18. Yakmatch is a station on the Baluchistan-Persia Railway in the middle of the desert. The specimens are from a small artifi- cial pool lined with cement into which water is pumped from a spring. The only vegetation was a spongy grey alga. The bottle contained a number of shells of Ostracods, but no complete animals. The shells of Cyprinotus incongruens Ramd. and of Ilyocypris bradyi Sars were identified. (5) ZANGI NAWAR, 20 MILES FROM NUSHKI, BALUCHISTAN. 29-xi-18. From small lakes of practically fresh water into which the Peshin-Lora river drains in the Baluchistan desert The pools were full of submerged weeds. Diaphanosoma brachyurum Liévin. Common. Males and females with resting eggs present. Simocephalus exspinosus Koch. Abundant. Certodaphnia reticulata Jur. Common. Diaptomus salinus Daday. Common. Eurycypris pubera O.F.M. Young only. Cypridopsts dentatomarginatus Daday ? One specimen only. Limnicythere inopinata Baird ? One young specimen. Potamocypris villosa Jur. Shells only. Ephippia of Daphnia magna found. ee Se ee ES ROE ORT? ON - This statement is not in verbal agreement with that of Cockerell, Bz. Bur, Fish. (Washington), XXXII: 1912); but the question is one of degree. * See Annandale, Rec. Jnd. Mus. XVI, p. 132, pl. ii, fig. 2. 166 Records of the Indian Museum. (Vor. XVAaE dry stream-beds in December. They perish annually in enormous numbers at this season as the water grows salt or foul owing to evaporation or to the excreta of large flocks of sheep and goats belonging to the nomad tribes who camp near the stream-beds. The Discognathus is found, alone or with D. phryne, also in permanent. irrigation channels and is very abundant in those that supply the garden of the British Consulate at Nasratabad or Shahr-i-Seistan. A few moribund individuals were caught at the same season in the reed-beds of the Hamun-i-Helmand, but the reason why they were dying was not apparent, for the water was neither salt nor foul. Numerous healthy individuals were captured in a reedy canal leading out of the Hamun a few days later. Discognathus adiscus, Annandale. Measurements (in millimetres). Number of Fin-rays, Scales and Proportions. | | | a 1} Total length (including | | “| caudal) .. . + | §0°5| 58°9) 41-0) 70°4| 41°5| 56°3) §7°3, 54°38 63°7 600) 52°3| 46°0 2) Length of caudal 5. | ELERO)! 5CC HO}! “(ORO} 14°6) “8-2| 13-0] 13°1, 11°4 140, 13°0| 11°8, 10°0 3| Greatest depth of Pe 9'0, 10°7| 8:0) 13°6| 8-1] 11-0] 11-0) OVI git oy 10°2)10°0, 7°O 4| Length of head : 9°s|11°5| 7°5 12°9| 8-0] 10°6 10°8 103] 11°5| TO 6 10°4 9:0 5 Width of head .. | 78) 8:2] 5:6, g:o| 61 7:8! 7-8) 7:0| 7-8 8:6] 7-2] 6-2 6) Length of snout 3°3|!4°6| 370) 50, 3-2/2 3:9) 14255 325) taro! “AO aeslunsca 7| Diameter of eye 2°5| 2:9).)2:2|| -3°3)'- 2:4! 3:0| 328] 370; 43°3h= 350) anne 8) Interorbital width 4°3| 5°0| 4°0| 6°8, 4:0] 5.0| 5-0 4:9! 5:0 5°25 -Clesaie 9 Longest ray of dorsal. g-2'12°5| 972/14°6) 8°3) 11-0 I1°3) 10°6 130] 12°2) 10°8] 9°38 10, Longest ray of anal ..:| 7:0; 8°4) 5°1;10°3| 6:0! 7°5| 7°7| 69, 9°6| 8:8) 8-2] 6.5 1 Length of pectoral | 7°4|10:7| 6:9 12°8| 7°1| 9:2) 10:0} 9.3! 11-2| 10:7; 8-6) 8-0 12| No. of branched rays | | | | in dorsal 8 a) Beale re: d Tail “gains | Telstar 13, No. of branched rays | | | in anal .. eae S ab ah sole S oi gSeaia 5 lS elpoe seas 14| No. of scalesin L-L.;.»| 37 | 36 | 36 | 37 | 37°} 37 | 37 | 37-1 37 | 35 | 35 30 IS No. of scales in ane: | | | verselineaboveL.L. | 5 5 Gop g Gt Bie 5 5 5 5 5 16 No. of scales in trans- verse line below L.L. | 63 | 6} | 6$ | 6} | 63 | 63 | 63 | 62 | 63 | 64 | 6% | Of Vy, z- | 4°59| £153| 4°55 482! 5° 06 4°33| 4°37, 48| 4°55) 4O1| 4°43] 4°3 18 +. | 5°6 | 5S | 5°22 5°17) 5°12 S11) 52 | 5°76) 5°79| 588) 5°23) 0'57 19| a: 5°31| 5° | 5°49) 5°45| 5°18) 5°31) 5°3 | 5°32| 5°53| 5°00, 5°03) 5-11 he + + | 3°8 |3°9"| 3°41) 3°9 |3°3 | 3°53! 3°48) 3°43] 3°5 | 3°53) 2°73) 3°40 ss 1-Caudal | | | } : | semezaliey ell aes ’ 21 neat : . = | 3°59| 3°53'3°55| 3°62| 400! 3°33] 3°37|3°S | 3°55] 3°O1| 3°43] 3°0 pi FCaudal | | | | _pelliMe ; : Pere le ; a2 fan ; +. |44|43 40 |4°T (4°13 3°93| 40 |4°57| 4°51) 40 | 4°05) 5°4 3 | | | a 1-Caudal | | ‘ 23 a ee .. |4°3 | 40 | 4°26] 4°32, 4:16 4°08] 4:07, 4°21] 4°32, 4°43) 3.9 | 4:0 Discognathus phryne, Annandale. ? 1897. Discognathus variabilis, Nikolsky, Ann. Mus. Zool. Ac. Sci., St. Petersburg, II; p. 347. 1899. Discognathus variabilis, Nikolsky (? in part), 7bid., IV, p. 412. 1906. Discognathus variabilis, editorial note to Regan, Fourn. As. Soc. Bengal, il, p. 8. 1919. Discognathus phryne, Annandale, Rec. Ind. Mus., XVIII, p. 70, pl, mes; appl isi tis 2. 1920.| N. ANNANDALE & S. LL. Hora: The Fish of Setstan. 167 The arrangement and structure of the pharyngeal teeth is very similar to that in D. adiscus, but they are a little stouter. We find in two specimens of a large series that small vestigial scales occur on the sides of the abdomen. In these specimens 94 scales, including the vestigial ones, can be distinguished below the lateral line on each side. We have not found any trace of scales on the dorsal line. Fully formed scales are shorter in proportion than those of D. adiscus and differ in being ornamented with radit below as well as above the nucleus. The circular striae are more numerous and less regular and the scale has a much more reticulate appearance. ‘The following are the measurements of a large scale from just above the lateral line:—length 1:8 mm.;_ breadth 2mm.; distance of nucleus from base 0°3 mm. The specimens of which measurements are given in the table are from the Pishin district of northern Baluchistan, except No. 6, which is the type- specimen from Seistan. This species has been generally confused with D. variabilis, Heckel, from which it differs, according to the description given by Gunther,' in the size of the eye as well as in its naked ventral and dorsal surfaces. It is impossible, therefore, to discuss the geographical distribution in detail. D. variabilis has been recorded from several localities in Syria, Mesopotamia and eastern Persia. Records from the last district probably refer to D. phryne. D. phryne is, with the exception of Nemachilus montanus (McClell.) (not the N. montanus of Day), by far the most abundant fish in the small streams of the Quetta and Pishin districts of northern Baluchistan at altitudes between 5,000 and 6,000 feet. It is not found in very rapid water but lives in thickets of Chara- ceae and other algae growing on a muddy bottom. Its food con- sists mainly of soft filamentous algae. At the Kushdil Khan reservoir it was observed in winter to collect in large numbers in pools into which water of a comparatively high temperature was flowing from underground sources into the outflow. The colour is much darker in very clear than in muddy water. In Seistan the species occurs in irrigation channels and probably (fide Nikolsky) in the reed-beds of the Hamun. Several specimens were captured by the members of the Seistan Arbitration Commission in the delta of the Helmand. ! Gunther, Cat. Fish. Brit. Mus., Vil, p. 71 (1868). 168 Records of the Indian Museum. [VoL. XVIII Discognathus phryne, Annandale. Measurements (in millimetres). Number of Fin-rays, Scales and Proportions. 1, Total length (including couse) «+ | 53°0) 63°3| 46°4) 55°3] 74°7| 660} 70°0) 38-0) 39°0 2) Length of caudal «+ | 12°4/14°4| 10°7| 12°6] 150} 15°0| 12-0] g*0| 8:0 3, Greatest depth of body .. +e | 9°2)12°9} 8-8! 10°6) 12°8/ 13°2/ 15°0| 8°3] 8:5 4| Length of head Bie ea MUOzs a2 Z 6] 114] 13°8] 12°8] 13-0] 8:1] 8:0 5| Width of head aie -» | 779] 10°3] 6:8] 8:1] 10°9| 9°6) 11°70] 6:0] 6:0 6) Length of snout ne -- | 4°0) 5°4| 3°8) 4°8] 5:0) 5°3} 5°5| 2°8] 3°0 7| Diameter of eye se ser! 222) 38°C!) 12-0} 223199 3°0) §2:0| 3732-0, meen 8) Interorbital width xe ~» | 5°0] 6:8) 4°6) §°1| 6:0) 5°38} 6°8/ 4:0] 4°0 9| Longest ray of dorsal .. -- | O70} 12°2} 8-0} 8°6)11°9/12°5) 12°9] 775! 6rx to) Longest ray of anal a sol! FAR 10's| 7°5| 8°6) 9:7) 975] 10°0| §°2} 5°3 11) Length of pectoral aie -+ | 9°9/T1°5| 8:0] 9°7| 1175) 10°4) 11's] 8-0] 66 12} No. of branched rays in dorsal ae er Sas ae MANLY fen heel a ke Naveen, oh fe | 3 13} No. of branched rays in anal 50 [8 5 5 5 5 Baines Ralesl 14) No. of scales in L.L. . 24935 1533 | 986340] 40.135 3/030" shales 15|No. of scales in transverse line | | above L.L. 54 | 42 | 23 | 42 | 2 | 53) 68 | 58 | 33 16) No. of scales in transverse line be- | tween LL. and ventral -- | 42 | 5% | 53 | 53 | OF | 43 1 53] 53 | 52 17 $ . +s -- | 4°27) 4°4|4°33|4°37 4°98 4°4 | 5°81| 4°22| 487 18 4 5°76| 4°9| 5°27| 5°21 5°83) 5°0 | 4°66| 4°57) 4°58 9 d ar Se -+ | 5°05) 4°38! 4°83) 4°85 5°83] 5°15) 5°38] 4°09| 4°85 20 4 as ike -- 14°77| 44148 | 4°95, 4:0 | 4°41] 4°03] 4°05] 3°81 1-Caudal | ! i —— at ++ 13°27) 3°4| 3°33) 3°37] 5°98) 3°4 | 4°81) 3°22] 3°87 | 1-Caudal | Bal are ore ae ++ [4°41 3°8| 4:05} 4'02| 4°66} 3°80) 3°86] 3°49) 3°64 1-Caudal | 23 Se are 3°86) 3°7| 3°71| 3°74) 4°32| 4°0 | 4°5 | 3°58| 3°87 Subfam. SCHIZOTHORACINAE. This subfamily is distinguished from the Cyprininae by the Salmonoid facies of the species, their small or degenerate lateral scales and the presence of an anal sheath consisting of folds of skin covered with greatly enlarged scales arranged in two parallel longitudinal rows. We have already commented on the two most noteworthy features of the Schizothoracinae, their geographical isolation and their superficial resemblance to the Salmonidae. It may be well, however, to state more precisely the characters wherein this resemblance consists, and those whereby the subfamily is linked to the Cyprininae. The resemblance to the Salmonidae is entirely external. It consists in the graceful but powerful frame of the fish, their small scales and usually silvery, often spotted colouration. ‘The close relationship to the Cyprininae is manifested in the whole structure. One or two important features of agreement may be noted. The air-bladder in both subfamilies is normally very large and is divided into a larger posterior and a smaller anterior region by a transverse ee al 1920.) N. ANNANDALE & S. L. Hora: The Fish of Seistan. 169 constriction. The pneumatic duct is long and slender and opens into the posterior part of the bladder just behind the constriction. In Schizothorax zarudnyt the weberian ossicles closely resemble those of so different-looking a Cyprinid as Labeo rohita, to which one of us has recently devoted special study in reference to these bones. The alimentary canal also is closely similar in the two fish. There is a strong probability that the Central Asiatic sub- family is related to the Labeo section of the Cyprininae, from which it has been derived as a result of isolation in mountain rivers flowing rapidly at high altitudes. An important factor, TExtT-FIG. 5.—Scales of Schizothoracinae. a. Dorso-lateral scale of Schizothorax zarudnyt, X 17%. 6. Dorso-lateral scale of Schizocypris brucei (adult specimen), X 373. c. Anal scale of Schizopygopsis stoliczkae from Siestan, X 17}. noticed by Stewart! in Tibet, is probably the necessity for long and arduous migrations at different periods of life. The three species (each of a different genus) that live in the lowlands of Seistan are either identical with or very closely related to mountain forms, but, as we have already noted, their isolation in a depression has not produced any very noteworthy structural modification of a general kind, perhaps because it is still too recent. 1 Stewart, Ree. Ind. Mus., V1, p. 73 (1911). 170 Records of the Indian Museum. [VoL. XVIII, Genus Schizothorax, Heckel. 1888. Schizothorax, Herzenstein, Fische, p. 96, in Wiss. Res. Przewalski Central-As. Reis.,; Zool. (II (2). 1916. Schisothorax, Vinciguerra, Ann. Mus. Civ, Stor. Nat. Genova, @), VALI spe 123- The genus is well represented in the Helmand system, whence Vinciguerra (loc. cit.) has given the names of the following five species :—S. brevis, McClell., S. macrolepis (Keys.), S. minutus, Kessler, S. ritchianus (McClell.), and S. zarudnyit (Nikolsky). There is also in the Indian Museum a mutilated skin from the old collec- tion of the Asiatic Society of Bengal labelled ‘‘ Schizothorax labiatus, McClell. Helmund R., Afghanistan.’’ The specimen is too imperfect to substantiate the identification, but the species to which it has been assigned is too distinctive to have been readily mistaken. We have thus six species known from this river- system, but except S. zarudnyi all these species have been found only in the upper waters at comparatively high altitudes. S. zarudnyt, moreover, is so closely allied to S. intermedius, McClell., a species common in some parts of the mountains of Afghanistan, that there can be little doubt as to its having originated as an isolated race of that species. Schizothorax zarudnyi (Nikolsky). (Plate Sk V5 figs: 1-2), 1897. Apiostoma sarudnyt, Nikolsky, Ann. Mus. Zool. Ac. Sct., St. Petersburg, II, p. 346. 1899. Schizothorax zarudnyt, id., ibid., IV, p. 409. This species is, as we have already stated, very closely allied to S. intermedius, McClell.,' but the following differential charac- ters are constant in a large series of adult specimens :— 1. The paired fins are much smaller. The branchial isthmus is longer and narrower. The scales are slightly enlarged at the base of all the fins, especially the dorsal and the anal. Oo Among the races assigned to S. intermedius by Herzenstein S. zarudnyi comes nearest affinits, Kessler (op. cit., p. 113, pl. xiv, fig. 1), but the snout is more pointed and the paired fins smaller and there are no greatly enlarged scales behind the opercular border. These differences may seem to some ichthyologists of no more than racial value and we have already admitted that we believe S. zarudnyi to have originated from S. intermedius as a local race. The differences are, however, so constant that we consider it more convenient to regard the Seistan fish as now specifically distinct. The colouration varies with the environment. In muddy water the back and fins are pale olive-green, the sides faintly | Herzenstein (op. cit., p. 106) does not regard the form identified with McClelland's species by Day as the forma typica, but see Giinther, Cat. Fish. Brit. Mus. 1920.] N. ANNANDALE & S. L. Hora: The Fish of Setstan. 171 tinged with green and the belly pure white. In the yellow water of the reed-beds the back and sides are much darker, sometimes almost black. A few silvery scales are always present on the back and some adult males have the fins reddish, and dull red specks scattered on the dorsal surface. The following measurements of a large male and female were taken from freshly killed fish :— g oe Total length ve a 460 mm. 490 mm. Length of head :. oe OZ, TOO! Length of-eye .. ie biter ea Tle Su Length of caudal fin... TA) 4s TAS a a Depth of body .. ae) 86 86 ”) TExT-FIG. 6.—Pharyngeal teeth of Schizothoraz zarudnyt, a. Lateral view of lower pharyngeal bone (x 3). b. c. Internal view of the bones of two sides in another specimen showing lateral variation. There are great differences in appearance, proportions and lepidosis between young and adults of this species, the chief being that the young are more slender, more silvery, have very much larger dorsal and caudal fins and eyes, and more imperfectly developed scales. In specimens between 56 and 66 mm. long we can detect no scales at all, while in those from 91 to 95 long they are much smaller in proportion than in the adult and are devoid of circular striae. In specimens up to 123 mm. long the caudal fin occupies about 1 of the total length, while in the adult it occupies only from :to.1. In specimens up to 93 mm. long the dorsal fin is considerably deeper than the body; in one 123 mm. long it is almost as deep, but in the adult it is distinctly less deep. The Records of the Indian Museum. [Vor. XVIII, 172 greatest depth of the body is contained from 8: to 6: times in the total length in young fish less than 124 mm. long, while in adults it is contained only from 45 to 5: times. In the proportions of the total length without the caudal to the greatest depth the differences are smaller, the figures 6 to 61 for fish under 67 mm. long, 5 for individuals between 90 and 123 mm. long and from 4 to 48 in the adult. The proportion between total length and length of head is less different at different ages, and that between head and body (without the caudal) and head is still more uniform, practically no difference existing between young and adult. In length of eye in that of head there is a great difference. In speci- mens between 56 and 92 mm. long it is roughly from 2s to 3s times, in one 123 mm. long 4s times, in the adult 71 to 8 times. In the young the spiny dorsal ray is also proportionately more slender and bears relatively much longer denticulations than in the adult. In the young these denticulations have a spiny char- acter. Measurements (1m millimetres). Schizothorax zarudnyt, Nikolsky. Number of Fin-rays, Scales and Proportions. | Total length (including | | | caudal) »- 155°7 |65°9 |91°3 | 92°I| 122°5| 320°3) 421-0) 285°5| 252°0 2| Length of caudal 11°4 |13°r | 18-15] 81:4 24°1| 47:2) S9°1| 46°5| 4371 3| Greatest depth of body 6:7'| 8:9 |14°4 | 14:5} 19°5| 67°0 83:2 56° 5) 45°3 4| Length of head T1254| 1454) 13280) V1O+7 2562172509355) 64°9' 55°4 5| Width of head 6:2 | 7°7 | 10°12] 11°3| 14°2| 39°8| 5179] 35°5| 20° 6} Length of snout PAE BOG | 26) 6'1 7°8\ 24°5| 30°4| 22°7) 18°6 7| Diameter of eye ScOnle4iesnl mbar BT) 5s) Oss 12°C] OnllanaAg 8) Interorbital width Steal eScAS| inact en| ania 6°7| 8:9] 24°8| 29°2) 20°2] 17°6 9| Length of caudal peduncle. | | 10| Depth of caudal peduncle .. | 11} Longest ray of dorsal 12°25/14°6 |19°5 | 20°4| 24°6| 47°5|- 65°3! 46°0, 41°7 12} Longest ray of anal 7-5 |°8:45|12°6 | 12°0| -17°3] 41°z| °58°O) 37-7 33'S 13} Length of pectoral .. | 8:4 | 8:9 | 13°6 | 13:5] 18-1} 42-2) 62°3] 3670) 36°8 14] No. of. branched rays in| pares WG . dorsal oe fies) 8 8 8 8 8 8 8 8 15| No. of branched rays in| anal So fe 85 5 5 5 5 5 5 5 5 16| No. of scales in L.L. BARU SO all genet nim al a .. | 107 | 110 | 108 | 106 17| No. of scales in transv erse | line above L.L. aa 354 354 | 334 | 333 | 328 18) No. of scales in transverse line below L.L. ae SS ae a3 ..- | 34% | 354 | 354 | 349 Hg 3 49 | 5:0 | 5:0 | 5:0 | 5:08) 6:78| 771 | O't4) 00 20 1 83 | 7:4 | 6°34 6°35| 6:3 | 478 5°00) 5.0 | 5°5 21, t 485 455| 485 47 | 48 | #£4| 45) 44) #5 22| + g2 | 2°82] 3°68| 3°86] 4°75, 777 | 778 | SO] 7°43 | - | 23 a | 3°66] 3°88] 3°74, 3°9 | 376 3°87) 37 | 377 | = .- 3°99 | #0] 3:9; 4:0 | 4:08) 5°78 | Saeed 25) — 2 676 | 5:9) 5:0)| 5:0) -5- |. £07 4iga) 4 Ol) ae | | a ea a a ae Es Se ee 1920.] N. ANNANDALE & S. L. Hora: The Fish of Setstan. 173 Schizothorax zarudnyi is a gregarious fish abundant in an adult condition in the pools among the reed-beds of the Hamun- i-Helmand. ‘The roe appeared to be ripe in specimens examined in December. Its food, unlike that of most species of its genus, consists largely if not exclusively of other smaller fish. From the fact that only adults were taken in the Hamun in winter, it is probable that the young make their way up stream in the flood- season. ‘They are extremely abundant in pools left in the beds of effluents of the Helmand or in the desert near these effluents, when the floods subside. It seems probable that the specimens we have examined represent the growth of at least five years and that sexual maturity is not obtained in a shorter period than four years. If this be so, the young of a year old are about 56- 66 mm. long; those of two years from gt to 95 mm., and those of three years about 125 mm. Both large individuals from the Hamun and young ones froni small pools were infested by an immature Trematode, which was encysted in their skin, in the superficial muscles, in the membrane of the fins and on both the outer and the inner aspect of the operculum. ‘The cysts were of a blackish colour and resembled those shown in Herzenstein’s figure of S. alttor (op. cit., pl. xii, fig. 1). We hope that a description of this parasite will be pub- lished later. S. zarudnyi is the only fish commonly caught for food in Seistan. A description of the methods by which it is caught will be found in the appendix to this paper. Schizopygopsis stoliczkae, Steind. 1888. Schizopygopsis stoliczkae, Herzenstein, op. cit., p. 191, pl. xvi, eee IQII. Siiisony corse stolicskae, Stewart, Rec. Ind. Mus., VI, p. 73, pl. iii, figs. 1, 2 and 3. Specimens from Seistan, whence we have examined a faily large series, apparently represent a dwarfed race. The largest we have seen is only 195 mm. long, and even smaller individuals are sexually mature. The two types of head referred to by Stewart (op. cit.) as the stoliczkae and the sevewzovi type are both found, without intermediates, in our series, but the former occurs only in two specimens and is not correlated with differences in proportions. We can discover no structural peculiarity in this low-altitude race except that there is a regular double row of large scales extending forwards in continuity with the anal sheath as far as the base of the ventral fins. Traces of a similar forward extension of the sheath are, however, to be found in certain speci- mens from high altitudes in the large collection from various. localities preserved in the Indian Museum. We do not, therefore, consider it advisable to give the Seistan fish a racial name. The series was collected by the Seistan Arbitration Commis- sion in the delta of the Helmand. The species has a wide range in £74 Records of the Indian Museum. [Vo~u XVHE, the headwaters of streams and rivers on the north side of the Himalayas and Hindu Kush, but except in Seistan is only found at high altitudes. TeExT-FIG. 7.—Adult specimen of Seistan race of Schtzopygopsis stoliczkae (reduced in size). 1920.]| N. ANNANDALE & §. L. Hora: The Fish of Seistan. 175 Schizopygopsis stoliczkaec, Steind. (Seistan). Measurements (in mallimetres). Number of Fin-rays, Scales and Proportions. ; | 1) Total length (including | | | caudal) 6 6 chon lumeh 202 ier? 7O 3| 179°8] 122°6, 195-2! 141°2| 133°2| 107°4 2| Length of caudal -- | LO°L] 13°6} 14-4] 26°3] 2275) 32:8] 2609] 26-3] 20:0 3| Greatest depth of body Q'O, 12°4| 12°4| 24°6| 23°0 32°4) 24°5| 25-4] 17°3 4| Length of head 116] 15°) 15°3| 34°1| 26°8' 39°6| 28-3] 30:2| 22-4 5| Width of head .. 5°7| 83] 7°5| 20:0 14°09) 24°8| 17:6] 17-0] 11-1 6 Length of snout 3°93! 4°7| 49] 9°8} 8:2) 12:0] 9-2) 95/2 10-2 7 Diameter of eye 375] 4:2|- 4°5) 6:8). 5:2] 7:8) 6:2) Grol - 5-2 8 Interorbital width oe | eS Gi 5°OF S5C| OB eRe | or2-sIk Ste Toole are 9, Length of caudal peduncle. PD Me she 20°0; 16:2) 23°0) 9°8| 17°1! 10} Depth of caudal peduncle. 3:0; a; S26)= = 7eO} 10:0) Aca ie So Ir, Longest ray of dorsal 9:3} I1°5| 12°0| 25-4| 19:3) 34:2! 25-0l 23-0] 16:0 12) Longest ray of anal 676) 9°2| 8:2) 24°6)-18°2| 29°3| 24-0! 24'1, 15:4 13) Length of pectoral Sf EGS, PG WOR! Pe olh srs}Oh iy acti 22°7| 22°0} 1598 14, No. of branched rays in | | dorsal IN. Chaudhuri, B. L. Cockerell, TD. Az .. 3) Gray, J. E., and Hardwicke, Major- General. Gravely, F. H. Giinther, A. 3) Hamilton, F. B. Eteckeliy. )\. ANNANDALE & S. L. Hora: The Fish of Seistan. 201 “ Description of new species of Botia and Nemachilus.” Rec. Ind. Mus. III, p. 341 (1909). ‘Zoological Results of the Abor Expedi- tion. Fish.’ Rec. Ind. Mus. VIII, pp. 243-257 (1913). “Report on a small collection of fish from Putao (Hkamti Long) on the Nothern Frontier of Burma.’’ Rec. Ind. Mus. XVI, pp. 271-287 (1919). “ The scales of the African Cyprinid Fishes: with a discussion of related Asiatic and European species.” Proc. Biol. Soc. Washington, XXIII pp. 141-152 (1910). ‘“ Observations on Fish Scales.’’ Bull. Bur. Fisheries Washington, XXXII, pp. IIg- T7A) (LOI). “ On the Fishes of the Neilgherry Hills and Rivers around their Bases.’ Proc. Zool. Soc. London, pp. 281-347 (1867). Vol. on Fisheries and Botany of Bengal in Hunter’s Statistical Account of Bengal (1871). “‘Tchthyology.” Scientific results of Sec- ond Yarkand Mission, (Calcutta: 1878). Fishes of India, YI (London: 1878). ‘“On the Fishes of Afghanistan.” Proc. Zool. Soc. London, pp. 224-232 (1880). Fauna of British India. Fish, I (London: 1889). Til. Ind. Zoology I, pl. \xxxviii, figs. 3 and 3a (London: 1832). ‘“‘Notes on some Asiatic species of Palingenia.” Rec. Ind. Mus. XVIII, pPsES7 (£920): [1868). Cat. Fish. British Mus. VII (London: Fishes in Aitchison’s ‘‘ The Zoology of Af- ghan Delimitation Commission.’’ Tyans. Linn. Soc. London, V (2), pp. 106-109 (1889). An account of the Fishes of the Ganges. (Edinburgh : 1822), Fishe aus Caschmir Gesammelt und Her- ausgegehen von Carl Fretherrn v. Hiigel (1838). Russegger, Rezsen, 1,2, p. 1027 (Wien: 1843): (not seen). 202 Records of the Indian Museum. Vor. XVI Heckell J: ji; Herzenstein, S. Jenkins, J. T. jerdon; 4c +) Jordan, D. S., and Evermann, B. W. Jordan, D. S., and Fowler, H. W. Kessler, K. T. Llyod, R. E. McClelland, J. oB) Nikolsky, A. M. Prashad, Be Regan, C22: Fishche Kaschmir’s nebst einem Anhang von dret neuen Arten aus indien, gesammelt von Freiherrn Carl v. Hiigel (Date later than 1843). “‘Fische,” in Wess. Res. Przewalski Cen- tral As. Reis. Zool. Theil III (2), (1888). ‘“Notes on Fish from India and Persia, with description of new species.”’ Rec. Ind. Mus. V, pp. 123-140 (1910). “On the Fresh-water Fishes of Southern India.” Madras Journ. Lit. Sci. XV, p. 309 (1848). ‘“On the Fresh-water Fishes of Southern India.” Ilid. XVI, p. 310 (1849). The Genera of Fishes. (California: Stanford University, 1917). ‘“A review oft he Cobitidae, or loaches, of - the: Riverss of Japan. = PaUase National Mus. XXVI, pp. 765-774 (1903). “Pisces’’ in Fedtschenko’s “Reise in Tur- kestan.’’ (1874). “ Report on the Fish collected in Tibet by Cap. F. H. Stewart, I.M.S.’’ Rec. Ind. Mus. II, pp. 341-346 (1908-09). ‘“ Observations on six new _ species of Cyprinidae, with an outline of a new classification of the family.” Journ. As. Soc. Bengal VII (2), pp. 941-948 (1838). “Indian Cyprinidae. XIX, p. 246 (1839). “Voyage de Zaroudny eétc., Reptiles: Amphibies et Poissons,’ Ann. Mus. Zool. -Ac. Sct; St. Petersburgh 11, pp. 346-347 (1897). Ibid., IV, p. 412 (1899). ‘““On a new species of Discognathus from the Kangra Valley.”’ Rec. Ind. Mus. XVI, pp. 163-165 (1919). “Description of five new Cyprinid Fishes from Lhasa, Tibet, collected by Capt. H. J. Walton, I.M.S.” Ann. Mag. Nat. Hist. XV, pp. 185-88 (1905). “ Description of two new Cyprinid Fishes from Tibet.”” Ibid., XV, pp. 300-301 (1905). ‘“New Cyprinoid Fishes from the Helmund Basin.’ Journ. As. Soc. Bengal, I, pp. 8 and 9 (1906). %) Astatic Researches, 1920.] N. ANNANDALE & S. L. Hora: The Fish of Setstan. 203 Regan, C.T. Steindachner. F. Stewart, F. H. +) Vinciguerra, D. Weber, M., and De Beautort, 1. EF. Zugmayer, E. e “Two new Cyprinid Fishes from Waziristan, collected by Major G. EK. Bruce.” Ann. Mag. Nat. Hist. (8) XIII, pp. 261-263 (1914). “Tchthyologische Mittheilungen (IX).”’ Verh. «Zool: bot. Ges= Wem. ’ : : é + ha > if ’ re 3 ; = ; > ble ‘ la Pt ASA, hy ys 4 Re f A \ aC ES EXPLANATION OF PLATE XVI. Soft fin of Adiposia, etc. Fic. 1.—Vertical section of soft fin of Adiposia macmahont, stained with haematoxylin, x 60. 2.—Part of the lateral region of the same preparation, X 200. 3.—Vertical section of the dorsal fold (behind the dorsal fin) of a post-larval Nemachilus evezardi 1 cm. long, stained with haematoxylin, X 200. 4.—Part of the lateral region of the same preparation, X 300. 5.—Part of the lateral region of a vertical section of the soft fin of Glyptosternun sp. (fam. Sisoridae, suborder Siluroidea), stained with haematoxylin, x 80. 6.—Part of the same preparation, X 200. EXPLANATION OF LETTERING. a=vacuolated tissue. b=epithelial region. c=fibrous connective tissue. d=large blood-vessel. d'=small blood-vessel. e=dorsal muscles. f=gland- cell. g=pigment-masses. g/=dendritic pigment-cell. i=unstriped muscle cells. Rec. INpb. Mus, VoL. XVIII, i920. PLATE XVI. wrvey of Tr a : S IATY iE ER 7 ee BES well ie VDE GSAS sot Se a Py oy a 4 al’ : ay Hh KO Lae aki tee bic [ i q Wie, ( eae if LA ae) 7, pie fee ae (a EXPLANATION OF PLATE XVII. SHAPED RAFTS FROM INDIA AND SEISTAN. Fic. 1.—Fisherman’s tuéin among the reed-beds (Scirpetum) of the Hamun-i Helmand. 2.—Men and boys making a futin on the shore of the same lake. 39 be) 3.—Fisherman’s bindt on swamp near Roorkee, U.P., India. REC. IND: MUS., VOL. XVIII, 1920. PLATE XVII. 3 SHAPED RAFTS IN INDIA AND SEISTAN. Photo.-engraved & printed at the Offices of the Survey of India, Calcutta, 1920. Bere RACV NC A O TA. Ee Pei ORE RA (WO LONET CLI DAE. &— COR TCE pA) OF SH LS PAN. By W.U,. DIstTantv. Fam. NOTONECTIDAE. Subfam. NOTONECTINAE. Genus Notonecta. Notonecta glauca Linn. Linn., Syst. Nat. ed. X, p. 439 (1758); Dist., Faun. Brit. Ind. Rhynchota III, p. 41 (1906). Var. marmorea, Fabr. Fabr., Syst. Rhyng., p. 103 (1803). Daulatabad, Seistan (village pond). Anisops fieberi, Kirk. Kirk., Entomologist, 1901, p. 5; Dist., Faun. Brit. Ind. UII, p. 46 (1906). Nasratabad, Seistan (pools in parade ground); Lab-i-Baring, Seistan (channels in reed-beds in Hamun). Fam. CORIXIDAE. Macrocorisa geffroyi, Leach. Leach (Corixa), Trans. Linn. Soc. X\I, p. 17 (1818) ; Dist. (Macrocorisa), Faun. Brit. Ind. V, p. 339 (1910). Daulatabad, Seistan (village pond). Corixa hieroglyphica, Duf. Duf.,; Hem. p. 86, f.f. 85—87 (1838); Fieb., Eur. Hem., p.93 (1861) ; Dist., Faun. Brit. Ind. II, p. 40, fig. 29 (1906). Randa Stream, 4 mi. N.W. of Jellalabad, Seistan ; Nasratabad, Seistan. : Corixa affinis, Dist. Dist., Faun. Brit. Ind. V, p. 341, fig. 202 (1910). Channel 8 miles east of Lab-i-Baring, Seistan ; Randa Stream, 4 mi. N.W. of Jellalabad, Seistan ; Lab-i-Baring, Seistan (channels in reed-beds in Hamun); Nasratabad, Seistan (pools in parade ground). 206 Records of the Indian Museum. DWOrexcV le Corixa substriata, Uhlet. Uhlet (Covisa), Proc. U.S. Nat. Mus. X1X, p. 275 (1896); Dist., Faun. Brit. Ind. V, p. 340 (1910). Nasratabad, Seistan (irrigation channels). Corixa seistanensis, sp. nov. Body, elytra and legs ochraceous ; pronotum with transverse black lines, the clavus with moderately oblique black lines, remain- inig elytra with somewhat close black mottlings; eyes black ; posterior tarsi more or less suffused with black ; margins of the clavus and elytra ochraceous, impunctate. Long 64 millimetres. Water-channels near Nasratabad, Seistan. Allied to the Palearctic C. rogenhofert Fieb. Corixa randana, sp. nov. Head ochraceous between eyes, which are black; pronotum dull dark greenish ochraceous with closely arranged transverse black lines; elytra dull dark greenish ochraceous mottled with black, the clavus with transverse black lines, the lateral margins ochraceous, impunctate; body beneath and legs ochraceous ; posterior tarsi with black or blackish suffusions ; head between eyes a little longer than broad, eyes much broader than long, their apices deflected a little beyond the anterior angles of the prono- tum. Long 73 millimetres. Randa stream, N.W. of Jellalabad, Seistan ; Lab-i-Baring, Seistan (channels in reed-beds in Hamun). Micronecta desertana, sp. nov. Head dull paie ochraceous with a central longitudinal, reddish linear spot, eyes black; pronotum dull pale ochraceous with a small dark obscure spot near each lateral margin which is paler and brighter in hue; elytra dull dark ochraceous with some scattered obscure fuscous shadings, the margins paler and brighter ochra- ceous; body beneath and legs dull pale ochraceous ; head shorter than breadth at base, subequal in length to pronotum; eyes not projecting beyond the anterior lateral angles of the pronotum. Length 3 millimetres. Hurmuk, Persian side of Perso-Afghan-Baluch boundary, 3,000 ft. (desert-spring). Micronecta biskrensis, Horv. Horv., Rev. d’Ent. 1899, p. 104. Hurmuk, Seistan 3,000 ft. (desert-spring); Algeria (Brit. Mus.). 1920. | W. L. Distant: Rhynchota Heteroptera. 207 The British Museum possesses an example of this species from Algeria, and determined by Dr. Horvath. With this speci- men the examples from the Seistan desert entirely agree. SOS OS Oe OOO OOOO ont BU, 10y we THE GROSS ANATOMY OF CORBICULA FLUMINALIS (Miller). By B. PrasHapD, D.Sc., Assistant Superintendent, Zoological Survey of India. In a recent paper on the molluscs of Baluchistan and Seistan ! Dr. Annandale and I have discussed the synonomy of Corbicula fuminalis atsome length. A few additional remarks may, however, be made regarding the type-species of the genus Corbicula. This genus was established in 1811 by Megerle? with Miiller’s species Tellina fluminalis? as the type. Preston‘, admitting this, referred to “‘ Corbicula fluminalis, Megerle’’ as the type-species of the genus Corbicula ; this is evidently a mistake since the author of the species C. fuminalis was Miiller and not Megerle ; the latter author only separated some of Miiller’s species of Tellina under the new name Corbicula. The anatomy of the various species of Corbicula is very imperfectly known, the only work of any importance being a paper on the anatomy of a Chinese species (C. Jagiliterti) by Fischer.’ I have therefore thought it desirable to describe the soft- parts of C. fluminalis, specimens of which, collected by Captain C.L. Boulenger at various places on the Euphrates, Mesopotamia, the original locality of Miiller’s species, and from Seistan, collected by Dr. N. Annandale and Dr. S. W. Kemp, are now available. Animal. Corresponding to the shape of the shell the animal is trigonal, very much swollen in the umbonal region and greatly depressed below. Specimens preserved in spirit are of a creamy colour, the muscles and foot being of a much darker shade of yellow. The mantle is transluscent and thin up to the palleal junction, the further lower part is much thicker owing to well-developed radial palleal muscles, while the free border is still thicker. The margin of the mantle is entire and without any papillae along the edge. There is, however, a row of small conical papillae on the inner surface a little behind the edge. The papillae are much reduced or even absent in the middle region of the pedal orifice. In the siphonal region also there are papillae in the same situation, but these are much smaller in size. The mantle-flaps of the two sides are not free from one another but owing to the absence of a supra- Rec. Ind. Mus. XVIUII, p. 58, pl. viti, figs. 1-6 (1919). Mag. Ges. Naturf. Freude Berlin, V, p. 56 (1811). Miiller’s Verm. Terr. Fluv. II, p. 205 (1774). Faun. Brit. Ind., Freshw. Moll., p. 210 (1915). Fourn. Conchyliol. XI, pp. 1-10, pl. i, figs. 1-3 (1863). Are Ye 210 Records of the Indian Museum. [VoyL. XVIII, anal aperture and the formation of a siphonal orifice consequent on the development of the two siphons, the arrangement is different from that in the Unionidae. Anteriorly the two flaps are united with one another to a point a little below the anterior adductor muscle. From this point to an imaginary vertical line drawn in continuation of the anterior border of the posterior adductor muscle to the free edge of the mantle, the two flaps are separate, forming the large pedal orifice for the protrusion of the foot. At the posterior termination of the pedal orifice the two flaps are united by a well-developed muscular connection, thus separating off the pedal from the siphonal orifice. The siphonal orifice extends to a little below the upper margin of the posterior adductor muscle, and encloses the anal and the branchial siphons. The mantle in the siphonal region is notched a little below the middle, marking off the regions for the two siphons. Above the siphonal orifice the mantle flaps « are united with one another as on the antero-dorsal border. There are two adductor muscles. Of these the anterior one lies a little below the anterior margin of the pedal orifice, and is circular in outline and of about the same size as the posterior adductor. The latter lies just above the anal siphon. A small pedal retractor muscle, ovoidal in outline, is situated above the posterior adductor muscle; its fibres, which by their contraction retract the foot as the shell-valves close, are easily traceable to the foot. The well-developed radiating palleal muscle-fibres of the mantle have been referred to already; they originate near the palleal line and end slightly behind the free edge of the mantle. In consequence of the absence of a distinct siphonal sinus on the shell the siphonal contractors are feebly developed and appear as specialized radial palleal fibres, which are more numerous in this region, are specially thickened and have a distinct antero-posterior course. In all the preserved specimens the siphons are fully contracted. It is not possible, therefore, to decide as to their respective lengths. The two siphons are, however, quite separate from one another, the upper or anal siphon being the smaller of the two. It has a rounded orifice with one or two small papillae surround- ing it, and has the anus opening into it anteriorly just behind the posterior adductor muscle. The lower or branchial siphon is much larger, with an ovoidal orifice in the contracted condition, and bears three to four rows of elongate papillae on its external orifice. The attachments of the gills are quite normal. The outer lamellae of the outer pair are attached to the mantle, the inner lamellae of the outer pair to the outer lamellae of the inner, while the inner lamellae of the inner pair are attached along a little more than the anterior half to the abdominal mass, the rest becoming fused with the same part of the lamella of the gill of the opposite side. The outer pair of gills are much shorter in both length and width than the inner pair. 1920. | B. PRASHAD: Anatomy of Corbicula. 20 The palps are comparatively short, fleshy structures some- what triangular in outline and the two palps of each side are of about the same size. The abdominal mass is much larger com- paratively than that of C. lagillierti. ‘The foot is small and feebly developed, it is rounded posteriorly and has a slightly pointed tip on the anterior side. The rest of the digestive system is very similar to that of the genus Galatea.! The rectum and the heart, lying in the pericardium, are plainly visible through the mantle. There is nothing special to note about the nervous, circulatory and excretory systems. Regarding the genital system the only point of interest is the more swollen condition of the abdominal mass in the females. Affinities, etc. The animal closely resembles that of C. lagil- iertt described by Fischer (loc. cit.), only differing in the better development of the siphons, the abdominal mass and the palleal muscles. The siphonal muscle is poorly developed and the palps are much smaller. These differences may possibly be correlated, as Dr. Annandale has suggested to me, with the peculiar conditions in which Corbicula fluminalts is found. These are its living buried in soft mud and the long period of hibernation during the dry weather. Fischer’s remark as to the main distinguishing characters of the genus Corbicula, when he says, “ On distinguera aisément les cyrénes a levrs branchie non réunies en arriere avec celles du cété opposé,’’ is far from a correct description for, as has been des- cribed, the inner lamellae of the inner pair of gills of the two sides are united with each other in the posterior part and are not free. Fischer himself had recognized this, for in his ‘‘ Manuel de Conchy- liologie ’’ (p. rogr) he described the animal of the genus Corbicula as having “‘ branchie réunies en arriére.’’ ! Rang in Ann. Sct. Nat. XXII, pp. 152-164, pl. v (1832). shootin. eet Pale ee” Cat itan - ce ok Pie 2s aes pny are ere = . Pipad sit Dak ee ane 3 rae ye Sort ott OL A ae Wes ee £) To itedeiii ees, agers co Shee & a 4 - - seis %y ‘ BEGGS er te ESS ‘ By race ce a = P ae : P eek Py. Sin Stasi ae: mes bONGeh, ER > x 3 x or Beit ats ay 16 ss ae, bie pe J ie ; DEO > et to) paeay ts rec oles bbe ns 19558 t - . Pa a. earth tw Ae oc —<— Es? “ae ft 2 7 : {set 2 ibe woteope ag FSH save we RS BOBS * “e Teo = ' : ar ie . = > +4 . - i S - > Tas , we © ‘ af : as * ~ > » a y cit ad a » _ can © ROE Be OON 4 Ele Cane Se ie MING Are tS: NPIS O Ie Ae By Tokio KaBurakl, Research Student, Imperial University, Tokyo. (From the Zoological Laboratory, The Museums, Cambridge.) A considerable number of accidents caused by the leech Limnatis mlotica (Savigny), which is well known under the term ‘* Horse-leech,”’ exist in literature.! In the process of being swal- lowed, the leech attaches itself to the mouth, throat and nasal cavity of men and beasts, causing hemorrhage as well! as hinder- ing respiration. When it penetrates deeper the hemorrhage may sometimes be very serious and even fatal. IT owe to Dr. N. Annaidale of the Indian Museum the oppor- tunity of examining one specimen of leech, which seems to be identical with the species mentioned. The material was obtained, apparently at Quetta, Baluchistan, from the throat of an Austrian prisoner, who had been brought from Persia. In drinking from dirty pools in Persia, he sucked up six individuals, all of which had been at the back of his throat for eight days. This informa- tion comes from Capt. A. G. R. Hardwick, R.A.M.C., who has communicated it to Dr. Annandale. The specimen is of large size, measuring 85 mm. long, exclu- sive of the posterior sucker, by 16 mm. across, taken almost in front of the posterior fifth of the body, from which the trunk tapers more gradually to the anterior end than to the posterior. The trunk is subcylindrical, presenting on the ventral surface of its anterior end the sucker which is destitute of the three powerful jaws, and in this respect it is unlike the medical leech. The upper lip of the sucker is divided on its inferior surface into two lobes by a deep longitudinal groove. ‘The jaws are covered by papillae and provided with more than roo minute teeth. The posterior sucker, which is distinctly separated from the trunk by a constriction, is of a circular shape, the diameter being about 12 mm. The leech, being preserved in spirit, cannot be expected to have retained its original colour. The body is of a uniform brown- ish grey colour, without being traversed on the dorsal surface by any trace of such four black lines and a median yellow or green stripe as has been described by Blanchard.? Along each side, ! Blanchard, R., Courtes notices sur les Hirudinées, 1. Bull. de la Soc. Zool. de France, XVI, 1891, p. 218. Hirudinées de I’Italie continentale et insulaire. Boll. Mus. Zool. Univ. di Torino, 1X, 1894, p.42. Masterman, E. W. G., Hiru- dinea as Human Parasites in Palestine. Parasitology, |, 1908, p. 282. 2 See Blanchard, Joc. cit., 1894, p. 43. 214 Records of the Indian Museum. [Vou. XVIII, 1921.} separating the dorsal from the ventral surface, is a well-defined, dull orange lateral stripe. In colouration the present specimen can be easily distinguished from Limmnatis (Poectlobdella) granulosa (Savigny), which presents the colour-pattern peculiar to the sub- genus Poecilobdella.' ‘The latter species is common in British India. The trunk is made up of ror annuli, of which the 5th and 6th are fused ventrally to form the posterior margin of the anterior sucker. The same is true of the 7th and 8th annuli. In the 95th and g6th annuli I have been unable to demonstrate such a tendency of dividing into two as has been put on record by Blanchard.? There are five pairs of eyes, of which the first three pairs are arranged on the first three annuli, forming a semicircle; the 4th pair lie on the 5th annulus; the 5th pair on the 8th annulus. The male genital aperture lies in the posterior edge of the 30th annulus, appearing to exist between the 2nd and 3rd annuli of the X somite; the female aperture occurs five annuli behind the male, between the 2nd and 3rd annuli of the XI somite. The nephridial pores lie in the posterior edge of the last annu- lus of each somite, but their total number could not be enu- merated. The alimentary tract agrees in the main with the medical leech, the crop being provided with some ten pairs of coeca, which are filled with blood, revealing a deep red colour. The intestine opens out on the dorsal surface of the last annulus, just in front of the posterior sucker. Limnatis nilotica is of wide distribution, extending from the Azores, through part of Western Europe as well as Northern Africa, to part of Western Asia. It is of interest that its range extends, as stated by Dr. Annandale,® into the borders and even within the boundaries of the Indian Empire. !/% Blanchard, Révision des Hirudinées du Musée de Turin. Soll. Mus. Zool. Univ. di Torino, VIXI, 1893, pp. 27, 28. 8 Annandale, N., Note on the Occurrence of the Leech Limnatis nilotica in Seistan and the Afghan-Baluch Desert. Rec. /ndian Mus., XVIII, pt. 1, 1920, Pp: 135. ns ens OOS REE POR’ -O,Ne st, SERGE SEW. sAcsiesis ae GAS ROP © DM OL CrSiGsm Olr LOWER MESOPOTAMIA. ParT IIJ.—TuHe Faminies NERITIDAR, HyDROBIIDAE AND MELANIIDAE. By B. Prasuap, D.Sc., Assistant Superintendent, Zoological Survey of India. Since the publication of the first two parts of this series, the Indian Museum has received a valuable collection of Mesopota- mian molluscs from Dr. KF. W. Bowell, made by him at Basra during August I917—February 1919, and placed with great generosity at our disposal. This collection, together with the collections mentioned in the previous parts of this report has rendered the working out of the Mesopotamian molluscs much easier and more satisfactory. Specimens of the families reported cn in the first two parts of the report are also represented in this collection, but these do not offer any specially interesting points for further discussion. They belong to the following species :— Limnaea tenera ewphratica. Gyraulus convexiusculus.' Limnaea peregra canalifera. Gyraulus euphraticus. Bullinus contortus. Gyraulus intermixtus. The only point to be noted in reference to these is that the large series of L. peregra canalifera exhibits much greater indivi- dual variability than the specimens previously examined. Family NERITIDAE. Genus Neritina, Lamarck. Subgenus Dostia, Gray. 1919. Dostia, Annandale and Prashad, Rec. Jud. Mus., XVI, pp. 2ZAZ, 243 In the paper cited above Dr. Annandale and I considered Dostia, Gray, as distinct from Neriztina, Lamarck, but asa result of further examination of the rich collections in the Indian Museum I am inclined to consider it as worthy of subgeneric rank only. ! We would invite attention to the fact that there has been an unfortunate transposition of lettering in reference to Gyraulus euphraticus and Segmentina calathus in fig. 5 on page 40 of this volume. The figure D refers to the Seg- mentina and the figure F to the Gyraulus. [N. A. and B. P.}. 216 Records of the Indian Museum. [VOLES Deve. In the Mesopotamian collection before me it is represented by Mousson’s species N. schlaeflii which was originally described from the Persian Gulf. Neritina schlaeflii, Mousson. 1874. Neritina (Mitrula) schlaefliit, Mousson, Fourn. Conchyliol. XXII, pp. 49, 50. 1874. Neritina crepidulavia? var. Schlaeflii, von Martens, Vorder Asiat. Conchyl., pp. 44, 67, 124. 1879. Neritina crepidularia (in part), von Martens, in Chemnitz, Conch.-Cab., Nevitina, pp. 37-45, pl. vii, figs. 9-11. This species was described by Mousson from shells collected on the island of ‘‘Ghaes” in the Persian Gulf. He noted the resemblance between it and the Oriental species, N. crepidularia and N. depressa, but considered it distinct from either owing to the much smaller size of the shell, the much less convex and laterally compressed dorsal surface, in the spire being more recurved and flattened, the nucleus being more prominent and in having a much smaller though comparatively more elongate mouth-opening. According to von Martens, it is only a variety of N. crepidularia, but the two Mesopotamian specimens collected by Dr. Bowell at Basra leave no doubt in my mind that it is speci- fically distinct. The Mesopotamian specimens have a purplish background with a large number of irregular white spots resulting in a network of rather broad purple lines surrounding the white spots ; near the margins of the shell it assumes a blackish tint owing to the dark- ness of the ground colour and absence of the white spots. Through the kindness of Major Froilano de Mello of the Portuguese Medical Service in Goa, the Indian Museum has received a specimen of this species from near Goa on the west coast of Peninsular India. This specimen is of a uniform dark brown colour. This record greatly extends the range of N. schlaeflit. Subgenus Neritaea, Roth. 1879. Neritaea, von Martens, op. cit., p. 6. 1899. Neritina (Neritaea), Kobelt in Rossmassler’s /con. Land. -u. Stssw.-Moll. (n. f.) VIII, p. 1. 1913. Theodoxis, Preston, Fourn. As. Soc. Bengal, 1X, pp. 470, 471. 1915. Theodoxis, Preston, Faun. Brit. Ind. Freshw.-Moll., p. 5. In his monograph of the genus Neritina, von Martens divided it into six subgenera, and included the species dealt with here in the subgenus Neritaea, Roth. This subgenus he further subdivid- ed into eight groups according to the shape, sculpture, etc., of the shell. In the synopsis of these groups he included N. jordans and the other Mesopotamian species of the genus in the group Pictae, but further on in the descriptive part of his monograph included them in the account of the group Semicirculatae. This is evidently a mistake since the group Semicirculatae is confined to Central America and South Africa, while the Pictae group is found in the tropical and subtropical regions of Asia, Africa and 1921I.] B. PRASsHAD : Gastropod Molluscs. 227, America, Moreover the structural characters of N. jordani and its allies do not justify their inclusion in the group Semicirculatae. Preston, following Locard! and Dautzenberg,® assigns these species to the genus or subgenus Theodoxis or rather Theodoxus, Montfort. This is certainly wrong, since Montfort’s name Theo- doxus, as both von Martens and Kobelt have shown, should be reserved for the Palaearctic species of the fluviatilis type; the shells of the jordant type being included in the subgenus Nert- taea, Roth. Dr. Annandale recorded four species of the genus Nerttina from Tower Mesopotamia. I have, however, found specimens of another in the collections made by Dr. Bowell at Basra. This form (N. macrii var. michom) was only known hitherto from Syria and Palestine. Neritina mesopotamica (Mousson), Martens. 1874. Neritrna meridionalis, var. Mesopotamica, Mousson, op. cit., p. 36. 1874. Were anatolica, var. Mesopotamica, v. Martens, op. cit., PP- 33, 34) pl. v, fig. 42. Ye 1879. Neritina mesopotamica, v. Martens, of. cit., pp. 90, 91, pl. xin, figs, 20, 21. 1899. Neritina mesopotamica, Kobelt, op. cit., p. 4, pl. ccxi, fig. 1325. Mousson confused this species with the Sicilian N. meridion- alis and gave a very incomplete description of his new variety mesopotamica. ‘The same form was also described in the same year by von Martens under the same name as Mousson’s, but apparently in ignorance of his work. Von Martens considered it to be a form of N. anatolica, but published a full description and a good figure, and also compared the variety with the species N. michoni and N. bellardit. Later in his monograph of the genus he considered it to be a species distinct from N. anatolica and worthy of specific rank. I have compared the single Mesopota- mian specimen before me with specimens of N. macriz, var. michonz, from Palestine and Mesopotamia and with those of N. anatolica, var. bellardit, {rom Damascus, and am of opinion that the species, as von Martens decided, is distinct from either. The single Mesopotamian specimen was collected by Dr. Boulenger on the banks of the Khandag creek, Basra. The spire of this specimen is not so prominent as is shown in von Marten’s figures, but in other respects closely agrees with his figures and description. ‘The inner lip is straight and has minute denticula- tions on its inner border. Neritina macrii, var. michoni (Bourg.). 1879. Neritina Macrii, var. michoni, von Martens, of. cit., pp. 88-90, pl. iv, figs. 11-13 and pl. xiii, figs. 27-29. 1 Arch. Mus. Hist. Nat. Lyon, UI, p. 231 (1883). 2 Rev. Biol. Nord. France, V1, p. 340 (1894). 218 Records of the Indian Museum. [VOreVEiE 1883. Theodoxis michont, Locard, Arch. Mus. Hist. Nat. Lyon, U1, pp. 1894. Neen. (Theodoxia) michoni, Dautzenberg, Rev. Biol. Nord. France, V1, pp. 351, 352. 1913. Theodoxis michont, Preston, op. cit , p. 471. There has been a great deal of confusion in literature as to the exact status of this form. Von Martens, who gives complete references to previous literature, was the first to recognize its relationship with N. macrit, Recluz, but in later works Locard, Dautzenberg and Preston have treated the species as distinct, and as belonging to the subgenus or genus Theodoxis. ‘This view, as I have pointed out above, is not correct and the species should be assigned to the genus Neritina and the subgenus Neritaea. Kobelt ! in his account of N. macriz says, ‘‘Martens hat’ Neritina karasuna und michonw mit macrii vereinigt, und zwar mit Recht,’’ and therefore includes michoni only as a synonym of N. macrit. However, owing to the differences between the typical N. macrii and the form michoni I consider the latter as a distinct variety of the species. The shells of this variety are less ovate than the typical form, have the spire a little more pronounced, the suture more impressed, the outer lip of the aperture extending much further over the columellar region, a relatively broader and more flat columellar region and the mouth shorter but broader. In the Mesopotamian collections there are five specimens of this form, four collected by Dr. Bowell at Basra and the fifth from the Khandag creek, collected by Dr. Boulenger. It may also be pointed out here that some of the specimens, collected by Dr. Annandale from the exit of the R. Jordan, Palestine, were described by Preston as being of an intermediate character between N. michom and N. jordani. I have examined these specimens but can find no resemblances between them and UN. jordant except for their colouration. ‘This point is considered further under N. jordani. Neritina jordani, Sowerby. 1861. Nevitina jordani, and var. turris, Mousson, Vierteljahrsschr. Naturf, Ges. Zurich V1, pp. 151-152. 1879. Neritina jordani, von Martens, op. cit., pp. 84-86, pl. 11, figs. 14-10. 1883. Theodoxia jordani, Locard, op, cit., pp. 231, 232. 1894. Neritina (Theodoxia) jordani, with var. aberrans, Dautzenberg, op. cit., PP. 349, 350. 1899. Neritina jordant, var. turris, Kobelt, op. cit., pp. 2, 3, pl. ccxi, figs. 1319, 1320. 1913. Lheodoxis gordani, Preston, op. cit., p- 470. 1918. Neritina gordani, Annandale, Rec. Ind. Mus. XV, ps. 162. Dr. Annandale recorded the occurrence of this species in Mesopotamia from specimens sent to him from Nasariyeh. Since then we have received several shells from Basra collected by Dr. Bowell. This large series, together with the Palestine shells ! Rossmassler's Fon penne wu, Stssw. Moll. (n. f.) VIII, p. 5. 1g21.| B. PrAsHaD : Gastropod Molluscs, 219 already in the Indian Museum, makes it possible to discuss the individual variation in form and colour. Sowerby ! figures three shells of what he calls the three varie- ties of N. yordant, without indicating a typical form; his figures also are very poor. Reeve’s® figures show a rather ovoid shell with a relatively small and scarcely exserted spire and with the body-whorl nearly smooth or with only a slight constriction on it. Mousson did not figure the shells he examined, but described a new variety under the name turris, which he distinguished from the typical form by its much larger size, more elevated apex, which makes the shell almost subcylindrical in shape, and by the body- whorl having a more prominent constriction. The figure of the typical form, the only one illustrated in von Marten’s monograph, differs from that of Reeve’s in having a comparatively shorter but more prominent spire, the columellar border narrower, the mouth relatively smaller and the constriction on the body-whorl much more pronounced. Dautzenberg, who followed Locard as to nomen- clature, added a new variety (aberrans), which, according to him, is distinguished by the almost complete absence of the constriction on the body-whorl and by its variable but different colouration. Kobelt has given good figures of the typical form and of var. ¢urrts, Mousson. His figure of the typical form is quite similar to that of von Martens. Preston does not add any notes on the specimens examined by him, but says in his account of T. michoni that some of the specimens of this species link up the two species, T. jordant and T. michoni, and that these two are probably extreme forms of the same species only. Annandale, however, refers to some of the shells from Palestine and reported on by Preston as belonging to the var. turvis of Mousson. In view of the above remarks it is clear that we have to deal with three forms. (i) N. jordani s.s. which von Martens’ and Kobhelt’s figures may be taken to re- present, and which appears to be a tiue lake-form occurring in the Lake of Tiberias and probably in the Lake of Homs. (ii) N.jordanz, var. turris, described by Mousson and of which Kobelt’s figure is a good representation. ‘This form is stated to have been taken in the Lake of Tiberias, but the exact biological conditions under wnich it was found are not stated by either Mousson or Kobelt. In Dr. Annandale’s collections and those of the late Dr. Anderson from those areas the form is only represented in the collections from the exit of the River Jordan. The form is probably a true stream phase. (iii) N. jordant, var. aberrans, described by Dautzenberg, but of which no figures have been published. The type speci- mens of this form were collected in the Lake of Homs and all the specimens from Mesopotamia in my opinion belong ton tt. his form appears to be confined to closed or slow-running waters. The three phases discussed above may be distinguished by the following key :— | Conchological illustrations, Neritina, pp. 4, 6, fig. 49 (18.41). 2 Conch. Icon., Neritina, species 129 (1856). 220 Records of the Indian Museum. [Vou. XVIII, I. Shell with a distinct transverse constriction on the body-whorl. A. Shell not more than 8 mm. in maximum diame- ter; more or less ovoidal in shape .. WM. gordant typtca. B. Shell measuring up to 15 mm. in maximum diameter ; almost subcylindrical in shape and with the constriction of the body-whorl better marked than in the typical form —.,..- NV. gordant, var. turrts. II. Shell with the body-whorl almost smooth or with only a faint transverse constriction across it ... NN. jordant, vat. aberrans. The major part of the Palestine collection reported on by Preston and the shells from the Lake of Tiberias from the late Dr. Anderson’s collection belong to the typical form (fig. Ia). They have the shell of an ovato-conical form with a prominent spire and a distinct though not very deep constriction across the Ay. 2\k\ SS if Fic. 1.—Neritina jordani, Sowerby. a. Shell of the typical form from Palestine. 6. Shell of Mousson’s var. turvris from River Jordan. c. Shell of var. aberrans, Dautzenberg, from Mesopotamia. body-whorl. The colouration is. variable. Most of the speci- mens have zigzag vertical stripes of a red, chocolate or dark brown colour alternating with white stripes of the same shape, in a few cases the stripes coalesce here and there to produce a reticulate pattern ; still further, a few have a uniform dark-brown or black colour with only a few pale spots. The specimens which Preston regarded as being of an intermediate character between his 7. michont and T. jordani are no more than uniformly coloured speci- mens of this form of N. jorvdani and have no relationship with N. macrii, var. michont. The var. ¢urris of Mousson (fig. 18) is, as I have stated above, represented by specimens from the River Jordan only. These are much larger, of a subcylindrical shape, have a much less promi- nent spire and the constriction on the body-whorl is much more impressed. In colour the shells show great variation ; the stripes are of various colours as in the typical form but are narrower and 1921. | B. PrasHapD: Gastropod Molluscs. 221 much closer ; a reticulate pattern is also formed in some cases by the fusion of the stripes with one another. I assign all the Mesopotamian specimens to the var. aberrans, Dautzenberg (fig. 1c.), since the constriction on the body-whorl in most cases is quite absent or only faintly indicated. The speci- mens are, further, less elongate than the typical form and have a comparatively broader columellar area. The colouration is very vatiable. In some cases the shells are uniformly dark-brown or black, in others they have purple, red, chocolate or brownish wavy stripes alternating with much broader white stripes, while most of them show a distinct network of white spots alternating with coloured ones, the latter formed by the fusion of the stripes with one another. Most of the specimens are of a smaller size than those of the typical form. Family HYDROBIIDAE. Specimens of the genera Tvicula, Bithynia and Amnicola (Alocinma) are represented in the Mesopotamian collections before me. Dr. Annandale (loc. cit., p. 163) assigned an imperfect shell from Nasariyeh to the genus Lithoglyphus, Mihl., with some doubt. I have examined this specimen and agree with Dr. Anna.\dale in considering it as possibly belonging to the genus Lithoglyphus, but it is too imperfect for a precise diagnosis. Genus Tricula, Benson. 1843. Tricula, Benson, Cal. Fourn. Nats iist., p. 407. —: 1851. Bithinella, Moquin Tandon, Fourn. Conchyliol. Ul, p. 239. 1852. Paludina (in part), Kiister, Mart., Chemn. Conch.—Cab., Palu- Gina, ebosyPs le 2 : 1856. Bythinella, Moquin Tandon, Hist. Moll. Terr.—Fluv. France, Presbo: 1858. Tricula, H. and A. Adams, Gen. Rec. Moll. 1, p. 300, pl. xxxi, figs. 5, 5a, 5b. 1862. Tricula, Benson, Ann. Mag. Nat. Hist. X, pp. 415, 410. 1863. Paludinella (in part), Frauenfeld, Verh. Zool.-bot. Ges. Wren, XIII, p. 199. 1885. Tvicula and Hydrobia_ subgen. Bythinella (in part), Nevill, Hand-List. Moll. Ind. Mus. M1, pp. 62, and 49 respectively. 1887. Tricula, Fischer, Man. Conchyliol., p. 727: 1892. Sythinella, Kobelt, Rossmassler. /con. Eur. Moll. (n. s.), V, Pp- 36, 37: 1915. Tvricula, Preston, Faun. Brit. Ind. Freshw.-Moll. p. 608. I have carefully compared shells of the Himalayan species Tricula montana, on which Benson founded his genus, with those of certain European species assigned by most recent authors to Bithinella, Moquin Tandon, and can find no generic difference. Kobelt’s figures of the various species of Bithinella, moreover, strongly support the view that the two are identical. A short review of the confusion that has existed in literature regarding the exact status of the genus Tvicula may be given before consi- dering the question of its synonymy. Li) th No Records of the Indian Museum. [Vor XVIII, Benson in his original description described the genus as belonging to the Melaniidae, basing his argument partly on shell- characters, which according to him showed the same relationships to Melania s.s. as ‘‘ certain Egyptian and Syrian species of Palu- domus’’ bore to that genus; and partly on anatomical grounds, the animal being, according to him, Melania-like. In this sup- position he was followed by Gray and H. and A. Adams, but Brot in his revision .of the Melaniidae, differing from them, stated that the genus was probably referrable to the Paludinidae. Benson in his subsequent work slightly elaborated his original arguments, and considered Brot’s position untenable owing to the ‘‘ very funda- mental difference between the concentric operculum of Paludina and the subspiral one of Tvicula” ; he again laid stress on the resemblance of the animal of Tvicula to that of Melania. Stimpson, Stoliczka, Blanford and Nevill considered Tricula to be a Rissoid genus. Fischer in doubtfully placing it amongst the Hydrobtiidae compared it with Acicula, a genus of land-molluscs, and re- marked, ‘‘ I,a classification de ce genre est embarrassante.”” He had, however, come to a nearly correct conclusion. Preston followed him in assigning Tyvicula to the Hydrobiidae, or what he calls Paludestrinidae. I have pointed out above the resem- blance between the shell of T. montana and that of the various species assigned to the genus Bithinella. ‘The resemblance between its animal and that of the genus Melania, on which Benson laid so much stress, is only superficial and his own description does not show any differences between the animal of Trvzcula and that of any other Hydrobiid. Assuming, therefore, that the genus is a true Hydrobiid and that there is no difference between it and the genus Bithinella, Moquin Tandon, the name Tvicula, Benson will have priority over Moquin Tandon’s. It may be noted here that the Indian brackish-water species referred to as Bithinella miiiacea in a recent paper! by Dr. Annan- dale and myself is not assignable to the genus Tvzcula, but has relationships with the genus Stenothyra, Benson. I have here to express my indebtedness to Dr. N. Annandale for pointing out to me the exact status of the genus Tyvicula and for the help he has so generously given me in clearing up its synonymy. Tricula palmyrae (Dautzenberg). 1918. Bithinella palmyrae, Annandale, op. cit., p. 162. The only specimens of this species are the shells referred to in Dr. Annandale’s paper cited above. I have carefully compared them with Dautzenberg’s description and figure of the species and can find no differences. Genus Bithynia, Gray. In the Mesopotamian collection this genus is represented by two species: (i) B. badiella, a species common in Palestine and | Rec. Ind. Mus., XVI, p. 248 (1919). G07 104 Naa B. PrasHApD : Gastropod Molluscs. 223 Syria and recently recorded by Dr. Annandale from Mesopotamia and (ii) B. rubens, a species widely distributed in Italy, Sardinia, Greece, Algeria, Syria and probably in Upper Mesopotamia. Bithynia badiella, Parreyss. 1919. Bithynia badiella, Annandale, op. cit., p. 162. In addition to the specimens reported on by Dr. Annandale there is a shell in the collection made by Dr. C. L. Boulenger from the area between Nasiriyeh and Hama Lake, Lower Mesopotamia. These specimens agree closely with the specimens from Palestine and Syria in the Indian Museum collection. Bithynia rubens (Menke). 1892.. Bithynia rubens, Kobelt, op. cit., pp. 70,71, pl. cxxxvu, fig. 806. Specimens of this very variable species collected by Dr. Boulenger and Dr. Bowell at Basra agree with Kobelt’s figures and with the specimens in the Indian Museum collection from Damas- cus. Nevill! gave names to a number of well characterized forms of the species from various localities and the specimens he marked are in the collection of the Indian Museum. As he did not describe these forms and some of them have since been described by Preston under other names, Nevill’s names will have to be taken as nomina nuda only, but the whole question is too complicated to be dealt with here. Genus Amnicola, Gould and Haldemann. Subgenus Alocinma, Annandale and Prashad. 1919. Alocinma, Annandale and Prashad, Rec. /nd. Mus., XVIII, pp. 23, 2 1920. Alocinma, Annandale Rec. Ind. Mus., XIX, pp. 42, 43: Dr. Annandale and I recently proposed this subgenus for certain Indian and Persian species hitherto assigned to the genera Bithynia aud Amnicola, We regarded this subgenus as being of an intermediate character between 4 mmnicola s.s. and Pseudamnicola, both of which also must be considered as subgenera only. Re- cently Dr. Annandale has pointed out that ‘‘ Bythenia ejecta,”’ a species described by Mousson from Lower Mesopotamia, also belongs to this subgenus. Amnicola (Alocinma) ejecta (Mousson). 1874. Bythinia ejecta, Mousson, op. cit., p. 40. A few of the specimens in Dr. Boulenger’s collection from the banks of the Euphrates at Nasiriyeh and at Feluja agree with Mousson’s description and are, therefore, assigned to his species. | Hand-List Moll. Ind. Mus., \\, pp. 40, 41 (1885). 224 Records of the Indian Museum. [Vo1. XVIII, At the end of his description of the species he adds the following: ‘‘ D’aprés la forme de ]’ouverture et surtout du bord, que est obtus a sa terminaison, je considére cette espéce comme une Bythinie, bien que l’opercule manque et que sa petitesse rappelle plut6t les Amnicoles.’’ It is clear from this quotation that Mousson was not quite definite about the generic position of his form. ‘The specimens before me, however, leave no doubt that the species B. ejecta, Mousson, as was considered by Dr. Annandale, is not a member of the genus Bithynia, Gray, but belongs to our new subgenus Alocinma. Bo Fic. 2.—Ammnicola (Alocinma) ejecta (Mousson) from the banks of the Euphrates at Nasiriyeh. Family MELANIIDAE. Genus Melanoides, Olivier. Melanoides tuberculata (Miller). 1874. Melania tuberculata, Mousson, op. cit., pp. 47, 48. 1887. Melania tuberculata, Schepmann and Snellmann, Moll. iz Veth. Mid.-Sum. Reiz. Ind. Sumatra-Exped., pp. 16, 17, pl. it, fig. 11 (radula). 1919. Melanoides tuberculata, Annandale and Prashad, Rec. Ind. Mus., OV TT pp: 3,325 pl. vi, fig. 1. In the paper cited above Dr. Annandale and I have given the diagnostic characters of this widely distributed and very variable species. The radula of the mollusc had been figured by us pre- viously in another paper.! Schepmann had previously figured and described the radula of the Sumatran form in the obscure publica- tion cited above. The two differ from one another, but the differences are more apparent than real, being due mainly to different views of the teeth having been figured; other differences in the number of denticulations are only of the nature of a varia- tion exhibited by the species. 1 Annandale and Prashad, Rec. Jnd. Mus., XVI, p. 146, pl. v, fig. 5 (1919). 1921. | B. PrasHap: Gastropod Molluscs. 225 The species is represented in the Mesopotamian collection by a large number of dry shells from Basra and a few preserved in spirit from the banks of the Euphrates. Most of the specimens are rather small but a few measuring up to 35 mm. in length are also represented. Melanoides pyramis, var. flavida (Nevill). 1919. Melanoides pyramis var. flavida, Annandale and Prashad, of. cit., Pp. 34, 35> In the paper cited above we referred to a rather dark speci- men of this form collected by Dr. C. I,. Boulenger from the flooded area near Khandag creek, Basra, Lower Mesopotamia. ‘This speci- men agrees in all particulars with the specimens from the Persian frontier and Baluchistan, but is darker in colour. It measures 27°5 mm. in length by 99 mm. in breadth, the aperture measures 9°8 mm. by 5°7 mm. Genus Melanopsis, Ferussac. Melanopsis nodosa, Férussac. 1874, Melanopsts nodosa, Mousson, Fourn. Conchyliol. XXII, p. 48. 1874. Melanopsis nodosa, Brot, ‘* Die Melantaceen’’ in Chemnitz, Conch.-Cab. (ed. Kiister), p. 432, pl. xvi, figs. 17-24 (in part). 1918. Melanopsis nodosa, Annandale, Rec. Ind. Mus., XV, p. 163. The species was recorded by Mousson from the Euphrates and the Tigris below Mosul on the basis of the collections made by Dr. A. Schlaefli in this region, and Annandale has recently recorded its occurrence in ower Mesopotamia. In the collections made by Dr. C. L. Boulenger and Dr. B.W. Bowell the species is represented by a large number of shells from Nasariyeh and from Khandag creek, Basra. All the specimens are of the typical nodosa-type and none belong to Mousson’s vat. moderata, described in his paper cited above. Most of the speci- mens are quite fresh and distinctly show the three rows of large tubercles on the body-whorl—a characteristic of this species. The specimens vary in colour from chestnut-brown to black, except for the subfossil shells which are white and chalky. The largest specimen measures 234 mm. in length and 11°3 mm. in maximum breadth ; the aperture measures Io°8 mm. by 5°2 mm. Melanopsis costata (Olivier). 1874. Melanopsis turcica, Mousson, op. ctt., pp. 48, 49. 1879. Molanopsts costata, Brot, op. cit., pp. 420-429, pl. xlvi, figs. 4-7. 1913. Melanopsis costata, Preston, Fourn. As. Soc. Bengal, 1X, p. 467. Melanopsis costata is a widely distributed species throughout Syria, Palestine and Mesopotamia. In the Mesopotamian collec- tions under report, it is represented by shells from the banks of a dry creek connected with Diala River between Baguba and Sharo- ban, and from the Khandag creek, Basra. 229 Records of the Indian Museum. [Vox. XVIII, The species, though closely allied to M. nodosa, is distin- guished from it by the costae on the surface of the shells being more regular and continuous and in having two instead of three rows of tubercles on the body-whorl. In the large series before me there are shells corresponding to Brot’s figures of the typical form, var. Bb and M. turcica bellio, Parreyss. There are, however, intermediate forms connecting the various varieties and it is therefore not necessary to distinguish them as different forms. Melanopsis subtingitana (Nevill), Annandale. 1918. Melanopsis subtingitana, Annandale, of. c’t., pp. 163, pl. xx, fieSeet. 2, This species has only recently been described by Annandale from. two shells in the Indian Museum recorded as Melanopsis 3. FrG. 3.—Type-specimen of Melanopsis subtingitana, var. laevis from the creek connected with Diala River, Mesopotamia. costata, var. by the late Mr. G. Nevill, and from two others from Mesopotamia presented to the Indian Museum by Colonel W. H. Lane. Dr. Bowell’s collection from Basra also contains a good series of the species. These shells agree closely with the type-shells, except that most of them are a little Jarger and have the ribs more obsolete. Var. laevis, nov. Two specimens from Mesopotamia, one from the banks of a dry creek connected with Diala River between Baguba and Sharo- ban collected by Dr. Boulenger, and the other from Basra by Dr. Bowell, are so different from the typical M. subtingitana as to deserve varietal rank. Both the shells ‘are nearly smooth and have the whorls much more regular than in the typical form. In other respects the two shells resemble the forma typica. Type-specimen No. M 11807/2 in the Zoological Survey of India (Indian Museum, Calcutta). i) No NI 1921.] B. PRASHAD: Gastropod Molluscs. Measurements of shells (in millimetres). Specimen A is from Basra, and B (the type) from the creek connected with the Diala River. A. B. Length of shell ay 16°8 17°6 Breadth of shell oe ont eal 8 Length of aperture 8:2 9° Breadth of aperture as of 3°9 5 NOLES, ON LARVAL, TREMA TO DHS Ho ROM S.#.1S TAN. By STaNLEY Kemp, Sc.D., Superintendent, Zoological Survey of India. The following notes on larval trematodes obtained during our visit to Seistan are admittedly very incomplete. In the field it was not possible to follow out the detailed anatomy of the various forms from living specimens and for this, it is to be feared, no subsequent work on preserved material can compensate. My notes are in the main based on preserved cercariae and on sections of infected livers. As a fixative Schaudinn’s solution, used hot, was employed ; while for staining haematoxylin followed by eosin gave the best results. In some instances cercariae were obtained in such small numbers that it has not been found possible to give any account of their anatomical characters. Of these one is a Xiphidiocercaria obtained in Melanotdes pyramis var. flavida at Saindak in the extreme west of Baluchistan and the other a furcocercous form with extremely long tail-flukes obtained in Gyvaulus euphraticus in the Hamun-i-Helmand, Of each of these only a single infection was discovered. The other three forms of cercariae are partially described below. One of them (Cercaria A) is a leptocercous form with rediae resembling those of Fusciola hepatica'. The other two (Cercaria B and C) are furcocercous forms ; one of these (B) bears a close resemblance to the larva of Schistosoma japonicum, but is distinguished by a number of characters. For the names of the molluscan hosts I am indebted to Dr. Annandale and Dr. Baini Prashad (see page 17 of this volume). Cercaria A (text-figs. I a—c). In well-preserved specimens the length of the body is from 290 to 360 » (average 320 ») and the breadth 170 to 220 » (average 200 /). The tail is twice or more than twice as long as the body, being from 670 to 780 » in length (average 730 »). The body (figs. 1 a, 6) is oval in outline, distinctly exca- vate posteriorly at the insertion of the tail, and is relatively thick- walled. ‘The oral sucker is about 0°05 mm. in diameter. The acetabulum is approximately the same size and is situated slightly behind the middle of the body. ‘There is a small pharyngeal bulb situated on the undivided anterior portion of the gut close to 1 Erroneously recorded as a monostome on-p. 22 of this volume. 230 Records of the Indian Museum. [VoL. XVIII, the oral sucker. The coeca are long reaching to the distal quarter of the body length. Near the junction of the coeca there is a cluster of cells which have the appearance of glands, but no ducts leading to the oral sucker were detected in the preserved material. Close to the tail a small bladder can sometimes be observed with traces of a pair of ducts directed anteriorly and one running posteriorly down the middle of the tail. The most characteristic feature of the anatomy is, however, the partially developed gonad which lies immediately above the acetabulum. ‘The gonad is very conspicuous in stained preparations and in dorsal or ventral view appears T-shaped with a fine strand joining each cross-piece to the stem. In reality the structure is more complex than this view indicates. The principal portion consists of a rod, about twice as long as the diameter VExT-FIG. 1.—Cercariae from Seistan. a-c. Cercaria A. d-e. Cercaria B. Feo CercanianGs \ of the acetabulum, which is dilated anteriorly and in its posterior third At its hinder end a transverse portion, consisting of a slender shaft and swollen head, joins it on either side and each of these portions curves downwards, so that the head itself is on a much lower level than the central rod from which it arises (fig. tc). From the head of each transverse portion a fine strand runs forward ; the two unite and the single strand so formed meets the dilated anterior part of the main axis on its ventral side. The disposition of the parts, as seen obliquely from the side, is indicat- ed in text-fig. tc. Thestructure is remarkably constant in form, showing little variations in fully grown and well preserved individuals. __ The space between the gut and the body-wall is filled with large cells which lie with their long axes placed transversely. 1921. ] S. Kemp: Larval Trematodes. 231 The cercariae develop in rediae which, as in those of Fasciola hepatica, bear a pair of processes near the hinder end ; they reach a maximum length of about 2°5 mm. Cercaria A was found in specimens of Limnaea bactriana, Hutton, living in small irrigation channels close to the British Consulate at Nasratabad in Seistan. Of one hundred molluscs which were examined for parasites nine contained this Cercaria. Cercaria B (text-figs. 1 d, e). This is a furcocercous form which appears to resemble rather closely the larva of Schistosoma japonicum as described by Cort.' The body in well-preserved specimens is from 167 to 193» in length and the mean of a number of observations is 186. Its breadth varies from 41 to 52, the mean being 48. The undivided part of the tail is from 192 to 222, with mean of 208; the furca are from 62 to 67 , with mean of 64h. The oral sucker is from 31 to 33, in diameter and the acetabulum from 22 to 23 p. The alimentary canal appears to be altogether absent and the body is for the most part filled with large gland cells (figs. I d, e). The four anterior gland cells differ very decidedly in character from the remainder, for in preparations treated with eosin the posterior cells are always very heavily stained, while the four anterior remain colourless. No ducts leading forwards from the four anterior cells could be traced in preserved material, though on analogy with similar forms it is probable that they really exist. The ducts from the posterior cells stain readily with eosin and are very conspicuous. ‘The number of posterior cells appears to be six, making five pairs of cells in all; but two are usually concealed by others which overlie them. On either side of the ducts from the gland cells, in the anterior half of the body length there is a small semitransparent area which perhaps represents an unpigmented eye-spot. _ Poster- ior to the acetabulum the beginnings of the gonad are visible, consisting of a mass of cells which is usually crescentic in form when seen in dorso-ventral view. This cercaria, so far as can be seen from preserved material, seems akin to that of Schistosoma japonicum or, more nearly, to the very closely allied form from Bengal recently described by Major Sewell (loc. cit.). There are, however, marked differences. In the Seistan cercaria the tail-flukes appear proportionately shorter, the acetabulum larger and unpigmented eye-spots are perhaps present. No trace of the gut could be found, nor of teeth on the anterior sucker. Cercaria B was found in specimens of Gyraulus convexiusculus (Hutton) obtained in the Hamun-i-Heimand, Seistan. Forty-three ! Cort, Univ. Calif. Publ. Zool., XVIII, p. 485 (1919) ; see also Sewell, Rec. Ind. Mus., XVI, p. 425 (1919). 232 Records of the Indian Museum. [VoL. XVIII, G. convexiusculus were examined of which three contained examples of the parasite. Cercaria C (text-fig. 1 /). The cercaria is a furcocercous form, similar in most respects in its internal anatomy to Cercaria B. It is, however, a much larger form and possesses pigmented eye-spots. In well-preserved specimens the body is from 188 to 230 p in length and from 56 to 80, in breadth, the mean of a number of observations being 206 by 70. The undivided part of the tail is from 247 to 360,, with mean of 305, and the furca are from 106 to 136 », with mean of 120. The internal anatomy, so far as it can be made out, is very similar to that of Cercaria B; the same gland cells are to be seen and in stained specimens the four anterior cells differ in the same way from those placed further back. The gonad, however, is not crescentic in dorso-lateral view, the acetabulum is smaller and the eye-spots are deeply pigmented and black in colour. This cercaria was found with the preceding in specimens of Gyraulus convexiusculus (Hutton) obtained in the Hamun-i- Helmand, Seistan. ‘Iwo molluscs, out of forty-three which were examined, contained examples of the parasite. ADDENDUM. With Dr. Kemp’s permission I add here three figures of the parasite of Schizothorax zarudnyi referred to by Mr. S. L. Hora TexT-F1G, 2.—Parasite from Schizothorax zarudnyt. and myself on p. 173 of this volume. The figures have been placed at my disposal by Major R. B. Seymour Sewell, I.M.S. and represent three views of the animal as seen as a solid object (A) and mounted in glycerine (B & C) after extraction from the cysts in the muscles of the fish. Their magnification is not stated, but it is at least 12 as reproduced. ‘The structure of the organism is so enigmatic that none of us are able even to suggest its taxonomic position. - The preservative had apparently failed to penetrate the 192T.] S. Kemp: Larval Trematodes. 233 peculiar cylinder in the interior of the ‘animal and sections were a complete failure. Major Sewell notes in explanation of his figures that there is a more or less well-defined groove round what appears to be the anterior extremity and sucker-like disc (s) in the middle of the (?) ventral surface. The (?) posterior part of the organism is divided by faint grooves (g), as shown in fig. A. These are visible only on the (presumed) ventral surface. On the dorsal surface, near the posterior extremity, there is a small aperture (). The outer parts seem to form a kind of test, lined by a thin membrane, and inside this there is a cylindrical body (c) with an apparently chitinous investment. At the end nearest the (?) dorsal grooves the extremity of this body is contracted to form a collar (7). Though the specimens had been fixed in Schaudinn’s fluid and were apparently well preserved, no further structure could be made out.—N. Annandale. eee eae aes ts ree PEA OU A CCEA UN ARO (S Eats a AGN A SUMMARY. By N. ANNANDALE, D.Sc., F.A.S.B., Director, Zoological Survey of India. The aquatic fauna of Seistan is a scanty one, fairly rich in individuals, as is usual in a fauna living in abnormal conditions, but poor in species. The most salient fact about it is that it is essentially a mountain fauna acclimatized to live in a swampy depression. Before discussing its geographical and biological rela- tions as a whole I will first consider briefly the different species found in the different bodies of water that exist in the country, viz. desert springs, stagnant pools, rivers and other water courses, and the Hamun-i-Helmand or basin into which the Helmand finally drains. An account of these will be found in the Introduc- tion to this volume. 1 have given in foot-notes references to the species of animals not otherwise noticed in the volume. THE FAUNA OF SPRINGS IN THE SEISTAN DESERT. In our hurried journey across the desert of Seistan and the Afghan-Perso-Baluch frontier we had little opportunity to examine many of the springs we passed, but those we did examine were very uniform in their animal life, which did not appear to be much affected by the varying degrees of salinity of water accepted under stress of necessity as potable by man and beast. None of the springs contained fish, but all the larger ones had a small but healthy fauna of molluscs, insects, Entomostraca and leeches. MoxiLusca. Only three species of molluscs were found, Mela- noides pyramis (Benson), Gyraulus euphraticus, Mousson and Cor- bicula fluminalis (Miller). Of these the most commonly present was the Melanoides. Of this species all our specimens belonged to Nevill’s var. favida, a form perhaps peculiar to situations of the kind and originally described from Persian Baluchistan. The shell, though not of the largest size attained by the species, is not dwarfed or in any way distorted. A single specimen of this race is also known from Mesopotamia. The Gyraulus is perfectly normal, though found in water distinctly bitter to the taste. It is noteworthy that the species has an extraordinarily wide range (from Mesopotamia to China) and must, therefore, be extremely adaptable. Thesameis true of the Corbicula, which is found practi- cally all over Africa and southern and middle Asia, was originally 236 Records cf the Indian Museum. [Vou Sey ida described from Mesopotamia and occurs fossil in Great Britain. Shells from the springs are smaller and rather thinner than usual and perhaps a little broader in proportion to their height, but specimens from larger bodies of water exhibit great individual variation in these respects, merging gradually into the thick narrow form called C. cor by many conchologists. The molluscs of the springs, therefore, may be said to show very little response to their environment in structure. Insecta. Insect life, even in the middle of winter, is by no means deficient in the springs. The most abundant and the most commonly observed species was the mosquito,' Theobaldta longi- areolata, the larva of which was found in water so saline as to be almost undrinkable, as well as in the excellent spring at Hurmuk, claimed locally to contain the finest water in all Iran. Both larvae and pupae were seen in December and imagines were observed hatching out. The latter, which were common also at Nasratabad, were, however, very sluggish at this season and made no attempt to suck blood. Small Rhynchota of the family Corixidae were frequently noted in the springs and in that at Hurmuk two species of Micronecta (M.desertana and M. biskrensts) were abundant. The former has been described by Mr. Distant as a new species, while the latter was described by Horvath from the oasis of Biskra in the Algerian desert. Small Hydrophilid and Dytiscid beetles are not uncommon, and a large species of the latter family was captured in the spring at Hurmuk. CRUSTACEA. ‘The only Crustacea seen in the springs were Ostracods, which often swarmed on the bottom. In asample from an artificial tank containing water pumped from a spring in the Baluch desert Mr. Gurney found tests of Cyprinotus incongruens and flyocypris bradyi, both widely distributed species. HIRUDINEA, So far as I can judge, oniy one species of leech (Limnatts nilotica) occurs in the springs. It is common in all those at which transport animals are watered and is dangerous both to men and animals owing to its habit of entering their mouths while they are drinking and sticking to the tongue or pharyngeal wall. THE FAUNA OF STAGNANT - POOLS: Under this heading I propose to consider pools connected neither with water-courses nor with small desert springs. Such pools are found occasionally in depressions, probably always of artificial origin, in Seistan and apparently owe their water to percolation. ‘The largest pool of the kind we saw was at the village of Daulatabad towards the south of the country. It formed a fairly large village pond and lay at the base of a low ridge of stiff ! For the identification of this mosquito and of the other Nemocera mentioned in this paper | have to thank Mr. Edwards of the British Museum. 1921. |] N. ANNANDALE: Fauna of Seistan. 237 clay on which the village was built. Even in December it contained | a considerable quantity of water, which was very foul but ap- parently slightly saline. It was used by the villagers for all domestic purposes. There was no macroscopic vegetation, but much evidence of the presence of a luxuriant growth of microscopic algae was present. The macroscopic fauna of this pond consisted, so far as we were able to discover, of arthropods only. Cladocera, including large Daphniids aud Copepoda, were abundant, but circumstances did not allow of their collection. The most noteworthy features were the wealth of insect life and the large size of some of the species present. This was particularly noteworthy in the Rhyn- chota, the representatives of which are mostly very small in Seistan. As at other places the only families of this order that we could find were the Corixidae and the Notonectidae, but in the latter family the relatively large and very widely distributed Notonecta glauca, which we did not see elsewhere in Seistan, was common, while in the Corixidae Macrocorisa geffrovi was also present in large numbers. Dipterous larvae were abundant, the most conspicuous being an exceptionally large Chironomid, the imagines of which were observed hatching out from the pupae on the surface. Water-beetles, including large Dytiscidae and Hydrophilidae of moderate size, were numerous. The only other pools of the sort in which we collected were those in the parade-grcund at Nasratabad. They occupied pits from which clay had been extracted for brick-making. Their water was fouled by the camels and donkeys that frequented them, but not or hardly saline. It was six or seven feet deep in places but blocked up by a profuse growth of the water-weed Zannichellia palustris. Insects, Entomostraca and molluscs were rich in indivi- duals, but the number of species was small. No large species of Rhy nchota were seen, but several species of Notonectidae and Corixidae were abundant. Mr. Distant has identified the following forms :—-A nisops fieberi and Corixa affinis. Mr. Gurney found the following species of Entomostraca in the collections made :—CLADOCERA : Daphnia magna, Simocephalus vetulus. COPEPODA: Cyclops strenutus, C. eek OSTRACODA : Eucypris clavata, Ilvocypris bradyt, Potamocy pris villosa ; all com- mon and widely ‘distributed forms. The molluscs present were Limnaea bactriana, Gyraulus ecu- phraticus, G. convextusculus and Corbicula Auminalis, all common species in Seistan and found, with the exception of the Limnaea, in all bodies of water containing luxuriant submerged vegetation in the country. L. bactriana, which appears to be mainly a pond mollusc, was found only in those pools and in small water-courses at the same place. FAUNA OF IRRIGATION CHANNELS. The small irrigation channels that form a close net-work over the whole of the habitable part of Seistan have, at any rate in 238 Records of the Indian Museum. [Vor. XVInE December, a very poor fauna. This is not altogether surprising, for the water in most of them is at that season intermittent, being shut off for several days each week. Even, however, where the stoppage of the flow does not cause desiccation and where a spe- cies of broad-leafed Potamogeton flourishes very few macroscopic animals could be found and even insects and Entomostraca were extremely scarce, if not altogether absent. The only crustacean obtained from this habitat was a single specimen of the Concho- stracous Leptestherta tenuis. A small Tipulid fly (Symflecta elong- ata), the only species in our collection described from Persia proper, was common on the damp mud at the edge of similar channels and probably bred in them. In the garden of the British Consulate at Nasratabad we examined an artificial water-course which had, however, a much richer vegetation and fauna, probably owing to the fact that it was less liable to desiccation and received the benefit of manure applied to the garden through which it flowed. Its flow was, how- ever, intermittent like that of the irrigation channels outside, from one of which it received its supply. The vegetation consisted mainly of a narrow-leafed species of Potamogeton and of Zanni- chellia palustris, but a filamentous green alga forming cloud-like masses was also abundant. The water was fresh or practically so. The fauna was sufficiently rich to be treated group by group. Fisu.—Shoals of the small Cyprinid Discognathus adiscus oc- curred, remaining at the bottom in the day-time but rising to the surface in the evening. Among them was found a single specimen of D. phryne. The first species is known only from Seistan, while the second is common in the hill-country of Baluchistan. MoLLuscA.—The same molluscs were found as in the pools on the parade-ground hard by, namely Limnaea bactriana, Gyraulus euphraticus, G. convexiusculus and Corbicula fluminalis. No differ- ence in the shell of any of these species could be discovered. A few empty shells of Segmentina calathus, a widely distributed North Indian species, were also obtained from this channel. ARTHROPODA.—The insects and Entomostraca of the channel were the same, or practically the same, as those of the pools on the parade-ground. OLIGOCHAETA.—The little Oligochaete worm Nats communis var. pumjabensis was found in considerable abundance and in in- teresting circumstances. It inhabited small mucilaginous tubes, probably stolen from a Dipterous iarva, in masses of filamentous algae and to each of the tubes a colony of the polyzoon Lopho podetla cartert was attached. N. communis is a cosmopolitan species and the var. punjabensis is common in Northern India. Col. Stephenson! found among our specimens of this little worm a single individual probably belonging to another species common in North India, namely Chaetogaster punjabensis. | Stephenson, Mem. Ind. Mus. VII, p. 196 (1920). 1g2t.] N. ANNANDALE: Fauna of Seistan. 239 Poiyzoa.—Lophopodella carteri was the only member of this group observed. Its geographical range is now known to be exclu- sively Asiatic, but to extend from Eastern Persia to Japan, the Chinese and Japanese race being slightly differentiated. It is not by any means always associated with Oligochaete worms, but a Chironomid larva not infrequently establishes itself at the base of the colony. DELTAIC FAUNA OF THE HELMAND AND ITS EFFLUENTS. We were unable to visit the main branches of the Helmand in its inland delta, but collections of fish and molluscs were made in this region by Sir Henry MacMahon and other officers of the Seistan Arbitration Commission. ‘The fish they obtained were :— Scaphiodon macmahont, Discognathus phryne, Schizothorax zarudnyt, Schizopygopsis stoliczkae, Nemachilus stoliczkae, Adiposia rhadinaea and A. macmahoni. Of these only three (D. phryne, Sch, zarudnyt and A. macmahoni) were found in the smaller streams of Seistan. The molluscs collected by the Commission in the Seistan delta were Vivipara helmandica,' Lamellidens marginalis and Corbicula fluminalis. Of the first of these only single empty shells were found by us in other parts of Seistan, except for a number of com- pletely bleached specimens found in a flood-deposit. There is, therefore, some reason to regard the species as peculiar to the estuaries of the Helmand, as it has not been found except in Seistan and the immediate vicinity to the east. The two bivalves are common throughout Seistan, in which the Lamellidens has become differentiated into a distinct race (vhadinaea). Shells from the Heimand are thinner and smaller than those from other parts of the country. The Corbicula is remarkable for its extreme variability. We examined smaller streams of the deltaic system of the Helmand in the immediate vicinity of Nasratabad, near the ruined city of Jellalabad some 12 miles to the north and at Chilling consi- derably futher south. Where the water was actually flowing the fauna was very scanty, but the high clay banks were full of bleached shells of Corbicula fluminalis and often of Limnaea gedro- siana and the diferent Planorbidae found in the country. Insect-life is usually scarce in such streams, but a noteworthy feature of those of Seistan is that the Hydrometridae often seen on the surface of the stiller pools are replaced, at any rate in winter, by Diptera of the family Ephydridae, which were often observed in large numbers resting on the surface film. Mr. Brunetti’ has described a species (Halmopota viridescens) from Seistan that ‘*skates’’ much like Gerrits. Near Jellalabad we found the Randa stream practically dry, except for shallow pools left in the bed and completely isolated. ! See Annadale, Rec. /nd. Mus. XIX, p. 114 (1920). ? Brunetti, Rec. Ind. Mus. XVI, p. 300 (1919). 240 Records of the Indian Museum. [VoL. XVIII, he water in these pools was apparently fresh but very foul as wandering Baluchis watered their sheep, goats and donkeys at them. Most of the fish and other animals were dying in them in large numbers. The fauna was, or had been, fairly rich, but there was no macroscopic vegetation in an active state of growth. Remains of reeds and a few moribund plants of a broad-leafed Potamogeton were observed in several of the pools and the nodular roots of some plant were common in the water, evidently in a resting state. The fauna may be dealt with group by group. Frsu.—Four species of fish were found in the pools, viz. Dis- cognathus adiscus, Schizothovax zarudnyt, Schizocypris brucei and Adiposia macmahom. Of the Schizothorax only young and halt- grown individuals were obtained, although we had the pools netted by Seistani fishermen. Both this species and D. adiscus were extremely abundant. The Schizocypris, of which we saw only young specimens, were much less abundant. The Adzpfosia was present in large numbers. This fish differs from the others in being a burrowing form. It was the only species in the pools that was ina healthy condition in December, the majority of the others being dead or moribund. Its stomach-contents consisted of the remains of Cyprinid fish and of May-fly larvae. Moztusca. With the exception of afew dead shells of Gyraulus euphraticus and G. convextusculus and one of Vivipara hilmandensis found among the remains of reeds inone or two of the pools, the only specimens of Mollusca we obtained from them were shells of Lamellidens marginalis rhadinacus and Corbicula fluminalis. AN those of the former species were empty, but a few living individuals of the Corbicu/a were dug from the mud, in which they were buried four to six inches deep. InsEcTA. The insect-fauna of the pools was scanty so far as most groups were concerned, but the larva of a large May-fly, probably identical with the common European Palingenia longicau- da, was abundant in the mud and we obtained three species of Rhynchota from the pools, namely Corixa hieroglyphica, C. affinis andC. vandana, anew species described by Mr. Distant. C. hierogly- phica and C. affinis are both Indian. ‘CRUSTACEA. Numerous specimens of the crab Potamon gedro- slanum were dug from the mud, in which they were apparently hibernating with Corbicula, Palingena larvae and the fish Adiposia macmahont. The range of the crab extends from the Punjab Salt Range to Seistan. PotyzoA. An interesting member of this group [Plumatella (Afrindella) persica, sp.nov.] was found in one of the pools, coating the stems and nodular roots of the plant to which I have referred above. ‘The animal has not as yet been found elsewhere. Near Nasratabad we visited a number of shallow pools which in the flood season were evidently backwaters of streams or large irrigation channels. In most of them the only traces of vegetation were the roots and dead stems of reeds and dried masses of fila- mentous algae stranded on land plants growing near the margin. Tg2t. | N. ANNANDALE: Fauna of Setstan. 241 In drift of fragments of reeds, etc., that surrounded most of these pools, dead shells of Limnaea gedrosiana and the other common molluscs. of the country, including one of the few endemic species (Amnicola sistanica), were abundant, with the statoblasts of Lopho- podella and the gemmules of the sponges Sfongilla alba and S. (Eunapius) carter. In most of the pools we found no macroscopic life, but in one, in which a broad-leaved Potamogeton was common though not in a flourishing condition, Disocgnathus adiscus and young specimens of Schizothorax zarudnyi were abundant. Schizocypris brucei were also found, but in much smaller numbers. A peculiar form of Limnaea gedrosiana was also common in this pool. It is distin- guished from the forma typica of the species by its much greater individual variability and by the fact that the curve of the outer lip of the shell is flattened to a straight line. This mollusc, for which the varietal name vectilabrum has been proposed, has been found elsewhere only in the Kushdil Khan reservoir in the north of the hill-country of Baluchistan. The reservoir is a large, shallow artificial lake witha luxuriant submerged vegetation in winter, but liable to complete desiccation in summer. ‘The speci- mens of the mollusc from Seistan were mostly infected bv the common North Indian Oligocheate worm Chaetogaster bengalensts, which frequented their pulmonary chamber in large numbers. THE FAUNA OF THE HAMUN-I-HELMAND. The Hamun-i-Helmand, or rather that part of it which is permanently filled with fresh or nearly fresh water, may be divided into three zones of life, that of the open lake, that of the reed- beds and that of the bare margin. The zone of the open lake may be called more appropriately the Central Region. It is that part of the lake which is free from reeds and always, except in abnormal droughts, contains several feet of water. The reed-beds form in winter what is called in Persian the matzar or reed-country, but the name maz is applied in Seistan particularly to Phragmites, which is the most abundant of the three species of which the reed-beds are composed, namelv Phragmites communts, Scirpus littoralis and Typha angustifolia. In the flood-season a great area in the mazzary is under water and even when the water is low, as it is in December, the reed-beds extend out into the lake for considerable distances. In discussing the fauna of this zone we must, therefore, consider both the species living in pools among the reeds and also those of which remains are found in a dead or dormant condition in the soil of the nazzar, By the zone of the bare region I mean the shore of the lake at or just below low-water level at places where there are no reeds. THE FAUNA OF THE CENTRAL REGION. In December this region is very poor in life, both animal and vegetable. The bottom is a stiff, sticky clay which supports but a scanty growth of water- plants. A few beds of Potamogeton lucens, none of them at all 242 Records of the Indian Museum. [VoL. XVIII, luxuriant, were all that were observed. On the bottom, shells of Lamellidens marginalis race rhadinaeus and Corbicula fluminalis were abundant, but no living specimens were obtained. On one of these shells a living colony of the Polyzoon Fredericella sultana race jordanica was observed. Fish were very scarce, but we saw fishermen catching Schizothorax zarudnyi in nets in the open lake, near a reed-bed. THE FAUNA OF THE REED-BEDS. This fauna is much the richest, or rather the least impoverished, in the lake. It is con-. centrated in small, comparatively deep pools which are choked even in December with submerged vegetation. In the composition of this, Potamogeton pectinatus is the dominant plant, but FP. perifoliatus, Nats major and at least one species of Characeae also occur. Among the reeds very few fish are found, and of those we caught all belong to one species (Discognathus adiscus) and seemed to be in a moribund condition; but the more open channels in the reed-beds are the proper home of Schizothorax zarudnyt, the largest fish found in Seistan and apparently the only one caught for food. Limnaea gedrosiani, Gyraulus cuphraticus and G. convexiusculus were the commonest molluscs in the small pools, but a few small specimens of Ammnicola sistanica were also found. Shells of this species, Jn much greater abundance and of a larger size, were dug from the soil of the natzar. The Limnaea belonged to the typical form of the species but the shells were smaller and a little narrower than those found in ponds at Quetta. Insect-life was less abundant in this region than might have been expected. Larvae of Chironomid Diptera were fairly common, and so were those of two species of dragonflies. Major Fraser thinks that one of these is probably the larva of the common Palaearctic Agrionid Ischnura elegans, while he states that the other ‘‘ combines some of the features of an Agrionine with those of a Lestine.”’ He remarks that it is unusual to find dragonfly larvae active in winter, as these species were. Adu!t insects were less abundant, both in the pools and among the reeds, than larvae. Mr. Edwards has found in our collection several species of Chironmus, all allied to, if not identical with European species but unfortunately, owing to an accident, not in sufficiently good condition for specific determination. The Entomostraca found in this habitat were cosmopolitan species common in similar situations in other countries. None of the higher Crustacea were seen. A sponge and two species of Polyzoa were fairly common on the stems of Typha. The sponge was a phase of the cosmopolitan Ephydatia fluviatilis and one of the Polyzoa an equally cosmopolitan species, Fvedericella sultana. ‘The latter belonged to a race (jorda- nica) hitherto known only from the Jordan and the Volga system, while the other member of the same group [Plumatella (Hyalinella) bigemmis] has been described as new in this volume. FOZ: | N. ANNANDALE: Fauna of Seistan. 243 MARGINAL FAauNA. The marginal fauna is very scanty. It includes no molluscs, fish ot crustacea, and in winter we found no aquatic insects. The lower surfaces of blocks of clay, however, were covered with a fairly luxuriant growth of Ephydatia fluvia- tilts, in a different phase from that found in the reed-beds, and of Fredcricella sultana jordanica. ‘The latter were covered with Vorti- cellid Protozoa. Just above the water-level certain insects were not uncommon in the same position, notably the Tipulid Symplecta punctipennis, at least two species of Ephydrid flies, and a cricket (Achtea bimaculata) also found in a similar habitat on the shore of the Lake of Tiberias. COMPOSITION OF THE FAUNA AS A WHOLE. The composition of the aquatic fauna of Seistan cannot be described as abnormal, but there are certain deficiencies that call for discussion. Some of these are due to geographical cause, which will be discussed later, but others are not so easily explained and evidently depend on some factor in the environment nct yet elucidated. The most noteworthy are the apparent absence of Crustacea Amphipoda and of most families of aquatic Rhynchota. Freshwater Amphipods are scarce in the plains of India and asa ruie occur only in the large rivers, in which the species are im- migrants from the sea.!. At even moderately high altitudes in the Himalayas, however, species of Gammarus and Talorchestia occur and in the Quetta district of Baluchistan, between 5,000 and 6,000 heetaralt least two species are abundant in every spring — stream. We could find none in Seistan. The absence of all aquatic Rhynchota except Notonectidac and Corixidae struck us very much in Seistan, particularly in reference to that of the surface-haunting Hydrometridae. It is of course possible that we failed to find these species in winter be- cause they were hibernating, but this is improbable for two rea- sons, fitstly, because we sought for them carefully in spots in which they might have been expected to conceal themselves had they left the water temporarily, and secondly, because they are not uncommon on the water at the same season in adjacent districts. Dr. Kemp found a Miucrovelia abundant on the Zanginawar I,akes in the eastern part of the Baluch desert in December, and I noted a Gerrzs on small streams near Peshawar in large numbers in January. In neither instance was the temperature higher than it was in Seistan in November and December. The line of vege- table debris that marks the flood level on the bare shores of the Hamun-i-Helmand would seem to be an ideal retreat for hiber- nating Hydrometridae and we found amongst the fragments of reeds, etc., two species of Reduviid Rhynchota, several species of Carabid and Staphylinid and one of Curculionid beetles, at least two species of Diptera, a cricket and an earwig, a wood- louse and a ! Cf. Chilton, Rec. Ind. Mus. XIX, p. 79 (1920). 244 Records of the Indian Museum. (Von. XVIII, toad (Bufo viridis), all in a more or less torpid condition; but no Hydrometrid. Indeed, it seemed to us that this family was toa large extent replaced by Diptera, such as Halmopota viridescen:, Brunetti, which skated on the surface of the water in almost the same way as these Rhynchota do. The absence of molluscs of the family Melaniidae from the Hamun-i-Helmand and the waters connected with it is another point worthy of note. In the extreme south of Seistan we found one form (Melanoides pyramis var. flavida) in a desert spring, but neither living molluscs nor empty shells were found at any place in the irrigated part of the country. The absence of species of this genus, one of which is not uncommon in adjacent districts, may perhaps be due to lack of nutriment or the presence of mineral salts in the mud of which they invariably feed. In other respects the limitations of the fauna seem to be due rather to geographical factors than to any peculiarities of the environment. GEOGRAPHICAL RELATIONS OF THE FAUNA. In considering the geographical relations of the aquatic fauna of Seistan five facts must be borne in mind :—firstly, that the country lies well within the limits of the Palaearctic Region and is separated from India not only by several hundreds of miles of desert but also by the great mass of mountains that occupies the more important part of Afghanistan and Baluchistan and juts down southwards almost to the Mekran coast west of the Indus; secondly, that the only waterways that reach it, and probably ever have reached it, com@from the east and the north; thirdly, that even these waterways are of recent origin in their present course ; fourth- ly, that it is much depressed below the surrounding districts, and fifthly, that the aquatic fauna, as follows from the third and fourth facts, is composed of immigrants from high mountainous tracts. These facts account for many of its deficiencies, for example for the absence of aquatic Chelonia and Caridea, both of which are unknown from the higher regions of Central Asia. To the same ~ facts we may trace the paucity of genera in the fish and molluscs, contrasted with the relative wealth of sponges and Polyzoa. It will be interesting to apply these deductions to the different groups of animals that are represented in turn. Of the three Batrachia known to inhabit Seistan two are per- haps the most widely distributed of all the Palaearctic frogs and toads, namely Rana esculenta and Bufo viridis. One of these has evidently been stayed in its eastward range by the mass of moun- tains to which I have already referred. It does not seem to have pene- trated beyond the eastern limits of the Baluch desert, or, from the north into the valleys of the western Himalayas, in which the toad has madeitself athome. The fact that Rana esculenta is represented in Seistan and western Baluchistan by the race dibunda hardly affects the situation, as this race itself has an immense range in 1921I.| N. ANNANDALE: Fauna of Seistan. 245 Eastern Europe, Western and Central Asia. The third Batrachian, Rana cyanophlyctis, has a much more peculiar geographical distri- bution—from near Aden to Penang. It is perhaps the commonest and most universally distributed of the Indian frogs, at all alti- tudes up to nearly 7,000 feet, but east of the Bay of Bengal becomes extremely rare. Throughout the greater part of its range no racial characters have been discovered, but in Seistan it is said to be distinguished by the size of its eyes and tympanum. If this -be so—I have seen no specimens of the race setstanica of Nikolsky—the race provides evidence of the complete isolation of Seistan from other parts of the range of the species. The number of fish (9 species in 7 genera) known from Seistan is small considering that the country possesses that rarest of phenomena in Central Asia and Persia, a freshwater lake; but here again the same facts are illustrated. This becomes clearer if we examine the fish-fauna in detail. Of the nine species three belong to the Central Asiatic subfamily Schizothoracinae, which are in a sense anadromous fish though far separated from the sea, three to the Cyprininae, which may be regarded as the dominant group in the great suborder Cyprinoidea, perhaps the most successful and characteristic of all the non-migratory freshwater fish, and three to the Cobitidae, a family of wide range in the Palaearctic and Oriental Regions and modified primarily for life on or in a soft bottom in water of no great speed. Of the three Schizothoracinae one (Schizothorax zarudmyt) is indigenous to Seistan, but is little more than a local race of a species found in mountain streams at much higher altitudes to the north-east, another is identical with a species of similar habitat, namely Schizopygopsis stoliczkae, while the third has been known hitherto from Waziristan in the extreme east of the mass of moun- tains that forms the ultimate barrier between the Oriental and Palaearctic Regions in the Indian Empire. This is Schizocypris brucei. ‘The Schizothoracinae are the most characteristic of the fish of the highlands of Central Asia, and particularly of the northern watershed of the Himalayas and Hindu Kush. A few species, including some of the least modified forms, have made their way across the great divide and live in the streams of the southern watershed and even in those on the lower slopes of the Himalayas and in the plains immediately at their base. It is not to these forms that the Schizothoracinae of Seistan are related, but to true Central Asiatic species. The Cyprininae of Seistan belong to two genera, Discognathus and Scaphiodon. The latter seems to have its headquaters in Baluchistan and not to be essentially a mountain-dweller, while Discognathus, which is replaced in India by the closely allied but more specialized genus Garva, occurs in Syria, Mesopotamia, E. Persia, Baluchistan and Waziristan on the North-West Frontier of India. Neither genus is found in the highlands of Central Asia, and though both live commonly in hilly country, neither inhabits high mountainous regions. 240 Records of the Indian Museum. [VOrL. 2 valida The Cobitidae of Seistan, on the other hand, must be associat- ed with the Schizothoracinae in origin. Two of the three species belong to the peculiar genus Adzposia, otherwise only known from Turkestan, and are apparently endemic as species. ‘The third (Nemachilus stoliczkae) belongs to a group in its genus characteristic of the Central Asiatic highlands, and resembles its namesake of the genus Schizopygopsis in geographical range. The majority of the fish of Seistan are, therefore, without doubt of Central Asiatic origin and can only have reached Seistan from the northern watershed of the Hindu Kush, while a minority have probably arrived in the district from the lower parts of Baluchistan. From a geographical point of view, the molluscs are perhaps the most interesting group in our fauna except the fish. They differ considerably from the true Eurasian species that have pene- trated from Central Asia as far south as the valley of Kashmir, and almost as much from those characteristic of the Persian Plateau. This fact is illustrated equally well by the species and genera that are present and by those that are absent. The Seistan fauna in- cludes none of the widely-distributed Eurasian species found in Kashmir, such as Limnaea stagnalis and Bithynia tentaculata, nor does it include any representative of the essentially Eastern Palaearctic genus Melanopsis, common in Persia proper and Mesopotamia, or of Bullinus, one species of which is common in Mesopotamia. ‘The species of Limnaea that do occur bear a distinct resemblance to European forms, but at least one of them (L. bactriana) also resembles an Indian form, L. chlamys. The three species of this genus, one of which (L.hordeum) is very rare and is only known from empty and possibly sub- fossil shells have all been found also in Lower Mesopotamia, though not in Persia proper, but are not dominant in the former country. Two of them (L. bactriana and L. gedrosiana) also occur commonly in the hill-country of Baluchistan and Afghanis- tan, but not, so far as we know, at high altitudes. The three Planorbidae have a wide range both in the Oriental Region and in neighbouring districts. That of the two species of Gyraulus (G. convexiusculus and G. euphraticus, extends at any rate from Mesopotamia to Burma and all over the Indian Empire, while the third species of the family (Segmentina calathus) is found in Burma and Sumatra as well as in northern India. The occurrence of a species of Vzvipara in Seistan is an interesting feature. The genus is practically cosmopolitan, but for some unaccountable reason is absent from Syria, Palestine, Mesopotamia, the greater part of Persia and Baluchistan. Among living species the Seistan form (V. helmandica) is most closely related to one from Sind (V. sindica). It is, however, still more closely related to a fossil (tertiary) species from the Bugti Hills in south-eastern Baluchistan. Indeed, it can be separated specifically from the fossil form only with difh- culty. Both V. helmandica and the only Unionid known from Seistan afford clear evidence of the existence of an Indian element 1g2I.| N. ANNANDALE: Fauna of Seistan. 247 in the fauna. The bivalve, indeed, Lameliidens marginalis rhadt- naeus, is only a local race of one of the commonest Indian species. The scarcity of endemic species of molluscs of Seistan is noteworthy as providing additional evidence for the recent origin of the fauna. Only two species apparently belong to this category, namely Amnzicola sistanica and Vivipara helmandica. The aquatic molluscs, therefore, are of more mixed origin than the fish, probably having had better opportunities for immigration, and include a much more distinct Indian element. They pro- vide less evidence, moreover, of derivation from a high mountain fauna. The only Decapod crustacean found in Seistan is a race of a species widely distributed in south-western Asia and clearly of western rather than eastern origin. The race is not.known from higher altitudes than about 6,000 feet, but is cominon in the Quetta district of Baluchistan and extends its range southwards and eastwards from Seistan to the Punjab Salt Range. The absence of Caridea from the fauna of Seistan, and also of aquatic Isopoda, is noteworthy, but is easily explicable on geographical grounds. That of Amphipoda I have already discussed The Entomostraca have little geographical significance. The only leech discovered in Seistan (Liamnatis nilotica) is distinctly south-eastern Palaearctic in range. It is common in Egypt, and in many parts of the Mediterranean basin, but is not known from within the limits of the Indian Empire except in the extreme west of British Baluchistan. The aquatic Oligochaeta are essentially Northern Indian. ‘Two of the three species recorded are known only from India proper, while the third form is an Indian race of a cosmopolitan species. Four species of Polyzoa have been found in Seistan, Two of these, both species of Plumatella, are apparently endemic. One of these [P. (Hyalinella) higemmis| belongs to a cosmopolitan subgenus, the other [P. (Afrindella) persica] to one of tropical range and strictly Oriental so far as Asia is concerned. Of the other two represen- tatives of the group, one ( Fredericella sultana jordanica ) is a race of a cosmopolitan species, formerly known only from Palestine and the Volga system, while the other is identical with the Indian race of a species (Lophopodella cartert ’ known from India, China and Japan, but represented in the two last countries by a distinct race (davenportt). The only Coelenterate collected is a cosmopolitan species (Hydra vulgaris) common in the plains of India. Three species of sponge were found, viz. Spongilla alba, S. cartert and Ephydatia fluviatilis. The ‘last is a cosmopolitan species common in most parts of the Holarctic Zone but represented by distinct races in the Himalayas and Upper Burma and replaced in Peninsular India by an allied species (EH. meyent). S. cartert is the commonest of the Indian freshwater sponges and has also been taken in Hungary, Mauritius and the Malay Archipelago. S. alba is known from Egypt and from India, where it is usually 248 Records of the Indian Museum. [ Vor. 2aVibiee found in slightly brackish water. The Seistan form belongs to a distinct race or variety (vyhadinaea) not found elsewhere. Our knowledge of the aquatic insects of Seistan is quite fragmentary, being based on a collection made in the middle of winter and only partially worked out. We obtained specimens of a considerable number of water-beetles, but have not succeeded in persuading any coleopterist to name them and our collection of Diptera met with more than one misfortune. The aquatic Rhynchota, as I have already pointed out, belong exclusively to the families Corixidae and Notonectidae. The genera represented (Micronecta, Corixa, Microcorisa, Anisops and Notonecta) are cosmopolitan and most of the species are known to be Palaearctic. Howsmall our true knowledge about the range of the less conspicuous water-bugs really is, is, however, illustrated by the fact that one of the Seistan species is otherwise known only from an oasis in the Algerian desert. What Ihave said about the Rhynchota also applies to the Diptera. One species of Tipulid (Symplecta elongata) is recorded as Persian and one Ephydrid (Halmopota viridescens) has been described from Seistan as new; the other flies are well known European species. So probably is also the May-fly (Palingenia) abundant in its larva! state on the banks of the Randa stream. The aquatic fauna of Seistan is thus, as might be expected from its geographical habitat, mainly Palaearctic. Particularly in - the fish, it has affinities with that of the highlands of Central Asia, but the molluscs belong to the geographical association I have recently called the Afghan type—not true Eurasian but belonging to species with both Palaearctic and Oriental relation- ships. They have, indeed, been introduced, with part of the fish- fauna, into Seistan recently, from the lower mountainous districts of Afghanistan and Baluchistan. It is among the less highly organized invertebrates that the tropical Indian element is most clearly manifest, but although this element is apparently absent in the fish, it appears (to go beyond the groups discussed in this paper) among the birds, of which Mr. Stuart Baker writes:—‘“ The geographical affinities are Indo-Palaearctic, the races of resident birds nearly all belonging to the Palaearctic rather than to the Tndian forms sss On Sone other hand a few sub- “species, appat- ently resident, are typically tropical Indian.. : 249 N. ANNANDALE: Fauna of Seistan. Ig2I.] (6381) gor -d ‘AT yaurelp TNoo $/P oxjatmMeIp ruedur4} ‘a1onrm opnysu ‘oJ ‘Ysny-npury wror1y SnIpaMmAgqur *YIS 0} par[[e ATeso[D ‘WOISOY orjore “ee[eq Jnoysnoiyi jsourye saroeds ‘e[nsulueg Ave 0} viqriy ‘S Wlory spueyxe va1dhz vutaso.y are Fe ueISIIG Ou St ais me4SI9S es UeySTOG “4sve IsY}IN} SuUIpus}xXe jnq aDYzZINOJS “YIS JO EY 0} ALTIUNIG ‘ueysIag { eIpUy jo I9IUONT “M-N ‘UezSIIIZEMY ‘Ysny-npurpyy ey} pue sede] “PUI F] 949 JO Spoys-19}eM UIdqION 13 sha We4SIIG TWeqSTOS Uv4sI9g ‘ UeYsIyon[eg jo sureyunoyy | .- = 8 UeISIIS TOseyY Ijo1vaeyeg Jo jred 1azva145 ‘BISW [e3} UID -BISY "M “S :odommy “q pue “Ss uPrysIaS “CANGSAIIT “IS “WS ‘IP "Oo ‘snpy “uu ‘AAS[OHIN ,,0UeISIOG Ur yeyqeH ‘yWoeyrp ‘azuenboe OJ l4yso1 wenb ozouNsIp Ieurpnyzsuo] orjamMeEIp TNI0 ‘oAred ofno0 votd4y BUIOT VY .. | *AyJe19ues A1jUN0D pepooy pue unmey ey} Jo spaq-pecxy se puemy[ay oq} Jo eyeq ‘peq-meaijs Surdip ur sjood Appnw pue pueulyoy 24} Jo eyeq she ** puewmyey oq} Jo eyeq ‘HOSB9S-Ppooy] UT surear}s PUL SIZATI YIM pajyoouu0d sfoog a puemyoyH 2} Jo ej[9q “* ‘oJ9 ‘unmeH ‘puemyax jo eqjoq cs me puemyeH jo vi9q | ‘unure Fy jO Speq-pse1 pue sjeuueyo r3zeE A, “IDAII SUI | -43p ur sjood pure sjauueyo 1078 A\ ‘ozo ‘anMeP jO v8pa je siiqap puev Avpo jo sdumy sepun 10yu1M uy "+ AIZuNOD 9y} AO [Te A[qeqorg ** UNWIeF JO spaq-pesy ‘peyseig pue ‘pueuuy ‘vo1uvjsis wjonump ‘epodoyjses) *VOSO'TIONW “+ (aesayY) vavurpyys visodipy (‘YpneyD) -woyvmovu visodipy (‘pule}s) avyzo107s snpryovma ny "* ‘aesey ‘2a9n1q stsdha0z149S5 "‘PUle}g ‘avyzorjojs sisdostdoz1yIs °* (XIN) 24upnavz xvsoyjzozyIs ‘TesOY “2u0yvmanm uoporydvas ‘pueuuy ‘audsyd snyywusorsig ‘pueuuy ‘snosipp snyywusorsiqg "SHOSIg oe “Ine’] ‘sipiara ofng (seed) vpunqiprs vjuaynosa puny TAIN ‘poupjsvas suokjydouvho vung ‘VIHOVALVG ‘asuel [eorydess0ar) "18 }SI19g UT yeyiqey “OMe NN ‘NVISINS JO VNOAV OMVAOY AHL JO IsSI'] [Vor. XVIII, Records of the Indian Museum. 250 ‘jouueys woes ur ~~ ESS * D> (i LIBRARY|2 CA >t nt & Vase: me 4 S t RECORDS of the INDIAN MUSEUM Vol. XVIII. REPORT ON THE AQUATIC FAUNA OF SEISTAN WITH SUBSIDIARY STUDIES. Part V. April, 1921. PAGE Note on the Leech Limnatis nilotica .. ‘iP ee Pat pepe ees Report on the Freshwater Gastropod Molluscs of Lower Mesopotamia. Part Ill. The Families Neritidae, Hydrobiidae and Melaniidae . . sty hs Notes on Larval Trematodes from Seistan . , ip ae a vin 220) The Aquatic Fauna of Seistan: A Summary ae Mie aS Sag aa Calcutta: PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA. PRINTED AT THE BAPTIST MISSION PRESS, 1g2r. Px l ce Jwo Rupees . oA. Ta 7G ry Mb ad Var iy ¥ tse 4 PTET RIT TET AAT LH TOW a \\f WN ] Ni 1 \f WN 1] AN Wit Hi Ai MAI WL naa pebadennidvebetubetereboreueieeds aeoae conn nesopaner elon eberdsabalebebnensedehebater giao yoininpe afenesd “Pet shore sjabnmdehene ie weer wane” rete preent easing “ Fit eye em ~ te ee - fee ee peuieteteliliads Aadabehebabntedaned actehoneteme seer DALAL. A) HERP Ot ett NETH PE st endtreit eet, on