. CE RECRUE DH mare mm rer? re rn ee are CCE TE CORRE ROMEO nm ent tnt anne 28 Debut gd ae CETTE PEL ot et aed oh Gr us Tae e nt SPAR DEE SI SPARTEN onu Gosia ae weer Tre CHERE CRC RL 2 are PR wa Donner nore) OCT eee COCO 4 te D tr m OP nt - . 1 ans - , - > à ve ‘ ‚ + = : San . x 5 5 ’ -o-8 4 4 . Det ‘ + ' 7 ‘ - » > | > = - a tes 1 . . ri * " ; . + sn 4 ; ’ i > > “ " . . - ‘ .. ‘ prés ® - anteaters vor + : R . „ . ss. ve we ’ - + ¢ +» eye Meee mn . : z - « > + - - ‚ ’ ‘ 4 NEA RR Vra amas AARAAS MAR NZ OR a À nn nn == PR nn RE AA AT A nz ARRAARRRAZ FF, AA > ABRMARAARRARAAAAN RCE IE ann aaa ar appre aan nanan PAA ANNAN ZAR CAN RAACEER n AAC FRE PS PS PS TN — 2 AA POS ES A —_ Ss ZISMASSRABFAARAR na ER x ENS AA AARRRAR2RRARSACT aan IA mann Tnnman Mar AAA AAA PPS ù . 2 PAR AAA mA RAR ASSET NASA ae WY YYYYY’Y annne Rear? AARARARRAARA FAR, ann oN ARAN m ep ARR ass RE Dana AAA À TERRES RAA AAA AR aaP AR Ian PAARnn, 2 AAA RAR an ARAR PNA A a Pa a a NN NN SZ. an ne AA Fa En ET PA a PA Pa PA PS Fes SESARRAR == ARRET: BRARARARAARAAE AA anne ee es AAA 22 AAA AR ARR ann CARRE EIER PA Be PAPA À À SARPARBAERARA- AAnAmaman ARR AAA RAR RAR RRARE 22 RARE EEE RESTE REC ae ee eS „ren FA ZAARARFARAARGSSZ arr BAARR EP PRRRRARARAS \ AT RP PA PP PA PA Pa a PA A PA aß = PARC). PARRERE NAN Ann AAARRERRARR AAR RAARRRAAL aaa. Nm RARE AAARALAL PARA AAAAAAAAARAAAARARRH A A: Anar AREAS RAM AA SA AAA REA EG NAAR AAA PARRA AAAs PARAS AY AAIPRIAARRBARRBAINTENT RAA TS | ERRANG RAR AR AR MR AAARAAR FRERES APA RAA AAA An a A ARARRAARASER ass Ra Ve BERLIN nanas Ana Mann = SNA AT RAS An Nana anna =a YONA ee AAN RRAARA Aan Fr ap! | RAR Ce AAAAAAAAAAA PAPA Ay Ae AFFFFRERA = CELA E EEE EEE EEE AAA Ne ere Fi FA Fa aaa \ À \ en RARE ey ER \ pes Ÿ D AAA RRRA AAA ER A ARR RA AoARRARAP AAA AAFr aan PRR AR AAA A ATTY & AR | \ PR PP PP PS TS AA pa’ AAA \ à AR AA A PA PA A PS RN aaa Qi PP aN AAARAARRAMAAAA en aAmanA’ ARTE RARE aa annn annee RaRARARRARE sRCeCrReRR ery = a RAA AA A à A à à à == == RER RS ere re PRÉ Aaanaa © NA a pr pp Me ae BRAD SM AMARAAABRAAAARASASSSSITSSEISS aap ae a QT AABARBAHAGNE Pr AR ar 4 Mr A’ mm RAR AA RAM RARES A Banna! = RES NNN INN pr pa PS Pa PA nn 3 ES ER PA PASS NN EN EN ENS aAAR (ru | mms u À tin À, mm £ i ae a a a = = ee oom = = PY alalel anne CEPT ARARAA 7 — A | { f i, / eae aa eee RCE RAR ARR CEE RER A a Pa a PP on) BRERRARRRRETTN \ PAP Ai pase Naar PARAS S555 RAP | | \ | PSE el = ma a | Ar ARARBARE;, f ay ‘ AAA RER aaa a RAESSARRSSF = ARR; 11 A an pri PA PS PSP pr pr MAMA | SAR AA “ann SARA RARARRARR SA AA Beeeapecocsce Pl ps rs ARR ja Pa PA PA pa pa AAA A a i = An pe Bapap PET as RR - EE RAA ARARAAAAA RON PAP RRE AAAT ARR Ce iin ann AAA = = = — | nl en) an) an) A - ARABES R NN N ‘i NES RA an FF AARRERARA A N FAR ARE RE AAA ARAAAAT, A, N A An TRIO A IA AAA SR AR Sonn MAI AAA ARanaaas anannanananease | Bae SAAR AARAAARAAAY ARR RAA RAR RAR AAPAA 7 aay a a N Te = FA Be NAA AAAAA AAA AARAAEAAAAA = AA Ar SRAGRARGR Ar | SER SAAR ABA menage ei ee AARS Senßti = ie SARA ARBRE C a AIAA / a Anne SEA ARR ann n vs Shane RAR A N Mm en nares ee are IRA SEN pt aN | As: sa RAP a ARABE a A TEN ET. DAPRARRRREER AAAAE FRÈRES AAA AAA AAA AA RAAAAY YT en enennnnnn nana nn ang TS ee aa | Ras rwey oy Ti) DR u PA PA pa | AAAAAAAAAAAAAAAAAAA San AEA SARRFF I! ARRARAR aaa AAFFFER ARRAARFASS SITE Annan aren \ N als EN { | ) ‘ Er f \ am aa aA) BR ÉERER AAA RARE eS AAA AA AREA aa NA aga AAA N “AYP ARAB EE | ARR DLL RARE Ran ay YY 4 + | QU SPA Je a Pa PA PAP PA PN ie Rs AAAS ARR RRARARAR SA ER ne | un RARE FRERE a2 men Ian ARR RRRERES | | TE ER ARAA 22222222 ar ar amen anaaa= ea Aa RITTER SRR RRRRARAR aaron a nnn ARR RSA AR AAA AIRE 222aaaR | | Bm AR: SEIEN Sen - a7 \ SARA ER anna ~~ AAAARA ea RARER SANA Aa PN PAR pa pr pu AA ee AAA AA Am aa Lan lana RER SPAR RAR PARA RRR ER RAR RA 2 a 2 = AREAS amnanammnn rn RER RR RAS AAA ARRETE A AAA AA ANR AARA a RAR ===" RFA a ARE RARE PAPY BA af AN Um en 1 fm ES EN FN rn ARR RA AAA RAARA ERA ARPA Eee — NA A A A A A A 2 SNA SIASTISSAFFFFEZ NANA AAR Rp ARARRAPRPFP ARR AR | NA Am nn NRA © nee RAR Re RAT Emme Anna RAAR AA = | sA m) Desaaasspara- AAA AA RAAA AA AAA AE RAA AREA A AAA RES EE — fp) PA OR RaRRARAAA" SE Ca N aasan N on : , - N \ ÿ ÿ \ J = Em ge (ED FR 2 ARM Vem PR / Y l : } non af SR) RS SARA APS BARBARA paar SSa= AA’: AAA NN VANNES ANS RAR AUS = S = = Ja { f ARPAI à ARE le Penn a OF NARA RRA AAA, Anna m FR PR LT i Bo ie MEMOIRS OF THE INDIAN MUSEUM, VOL. V. FAUNA OF THE CHILKA LAKE No. 7. OCTOBER, 1920. PAGE . Hirudinea .. 509 Csonian Ing ASS y a Fe + ie Pe Setional Museo Calcutta : PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA. PRINTED AT THE BAPTIST MISSION PRESS. — 1920. Price Que Rupee. FAUNA OF THE CHILKA LAKE. HIRUDINEA. By W. A. HARDING, M.A., F.L.S. (With 2 text-figures.) a x 4 Glossosiphonia heteroclita, Linn. Piscicola olivacea, sp. nov. .. Placobdella emydae, sp. nov. . : HIRUDINEA. By W. A. HARDING. Dr. Annandale has placed in my hands a large collection of Indian Hirudinea and my hearty thanks are due to him for the opportunity of examining so large a body of interesting material. The results of my work upon this collection will appear in due course and I confine myself here, at Dr. Annandale’s request, to a report upon the leeches obtained during the investigations by Dr. Kemp and himself on the fauna of the Chilka Lake. These consist of examples of three new Rhynchobdellid species, whilst the general collection of Indian Hirudinea just refer- red to contains examples from this lake of another Rhynchobdellid species, a colour-variety of the widely distributed Glossosiphonia heterochta. It will be convenient to deal with this leech first. Glossosiphonia heteroclita, Linn. (1761). This well-known leech occurs in North America and throughout the greater part of Europe and itis now recorded from India for the first time. It is not peculiar to the Chilka Lake, and in the general collection are examples from many parts of India, showing it to be widely distributed there. The Indian examples agree with the normal type, with the exception of black markings present on the pale golden-yellow dorsal surface. European specimens are generally free from black pigment, but Apathy (1888, p. 790) describes a variety (siviata) having black transverse stripes, more or less broken, on every third ring. Castle (1900, p. 42, pl. viii, fig. 38) finds in the United States various gradations between the typical yellow form, and a form with transverse striae and an irregular longitudinal mid-dorsal band. The Indian examples have the black pigment disposed in a broken mid-dorsal line and further investigations may prove that they represent a new variety. The examples of this species, collected by Dr. Annandale at Lake Chilka in March, 1910, are accompanied by the following notes : — (a) Found “ among weeds in pond. Rambha, S. end of Chilka Lake.” (OE romeo eud of Chilka Take,” [This leech appears to be very common on freshwater molluscs of the genera Vivipara and Pachylabra all over India. In the Chilka Lake it is only found in flooded areas of fresh or practically fresh water.—N.A. | 512 Memoirs of the Indian Museum. NOW Piscicola olivacea, n. sp. (FIG. I.) FORM, SIZE, COLOUR. The eircular and exceedingly long and slender body of this little leech re- sembles in general form that of the Prscicola geometra of Europe. It is about half the size of the European species, the largest example measuring, when fairly extended, approximately 10°75 mm, in total length, the greatest width of the body being about 1°50 mm. For information as to colour, I am dependent upon Dr. Annandale, who has been good enough to send me notes upon the external features of several different individuals. From these notes it appears that the general colour of the body varies from bright to pale olive green, minutely speckled with black, or with a darker shade of green. A series of conspicuous white spots occur, one on either side of each somite, on the margins of the body, and these are connected across the dorsal surface by whitish and often indistinct bands. Another series of somewhat irregular elongated spots or blotches lie in the mid- dorsal line, one in each somite (on a level with the marginal spots) and these median spots, which may or may not be joined together at their extremities, give the appear- ance of a somewhat ill-defined whitish mid-dorsal streak. Anterior sucker circular, whitish, with three brownish bands on the dorsal surface; one band following the junction with the body, one near the anterior tip and a third and broader one between the two, in the posterior part of the sucker, which contains the eyes. The mouth-opening is situated in the centre of the interior cup. Posterior sucker somewhat heart-shaped, of the same green colour as the body, with seven pairs of whitish rays, corresponding to the seven somites xxvili-xxxiv of which it is composed. Several individuals examined showed traces of the original dorsal pattern, and from one of these the arrangement of spots and bands shown in Fig. I was drawn. It is to be understood that the arrangement indicated is schematic and subject to a good deal of variation in its details. RINGS, SOMITES, GENITAL ORGANS, EYES. The complete somite is formed of 14 annuli. The transverse middle line of the complete somite passes through a ventral ganglion, and also through the middle of the white mid-dorsal and marginal spots. The ganglion occupies two rings (7 and 8) of the complete somite. | The pulsating vesicles have collapsed in the specimens examined, but traces of them occasionally may be seen, and their presence is placed beyond doubt by Dr. Annandale, who has noted that they occur in the whitish spots on the margins of the body. The first pair lie in somite xiii, and there appear to be eleven pairs, the last pair lying in somite xxiii. 1920.] The male genital orifice lies in somite xi, and the female orifice in somite xii. There are two pairs of eyes on the anterior sucker, one pair on either side of the mid-dor- sal line. The component eyes of each pair are linear in form and inclined together at an acute angle in such a way as to resemble the equal sides of an isosceles triangle lying horizontally, with its apex pointing towards the margin of the sucker. The eyes in each pair do not actually touch at the point of inclination; they may be somewhat curved outwardly or, again, so closely approximated as to give, at first sight, the appearance of one linear eye. Hosts, HABITAT. This leech is a fish parasite and has been recorded from Hypolophus sephen, Tetrodon reticularis and Dorosoma indicum (Chatoessus chacunda). It has so far only been found in the Chilka Lake. The following examples from this lake were examined :— (a) “Station 52. 4 mi E.4N. of Patsa- hanipur. On AHypolophus sephen (Forsk.) found crawling on lower surface of body, in the gill-slits, near the anus and within the mouth on the palate. Sp. gravity of water 1°007—I'OII.” “ Station 22. E. side of Rambha Bay, base of Ganta Sila. Sp. gravity of water I’007—I'0II. Found on a man’s foot after wading.”’ “Station 49. 4-9 mi. E. 4S. of Barkul bungalow.” Id) “Station 85. Satpara. On Tetrodon velicularis (Bl.). Sp. gravity of water about I'026.”’ “Station 14. E. side of Rambha Bay. Sp. gravity of water I‘007—I"011. 7 Stations 554 28 mi NEI!E of Kalidai.”’ Fauna of the Chilka Lake : Hirudinea. | 319 eyes 1 I-V< Se 7 L->brown bands VISE VII< x VIII<= white bands 1 and spots IX<-: X<< XI<- & XII< Q XUI< -p.ves.] XIV< XV<_ XVI< XVII< XVIII< Ex x + > . 5 a bands: XX à and spots DR N 5 7} XXE XXI<. XXII<. XXII. b---p.ves.11 XXIV-<- NX VE SOK Willis = g XXVH Sc Sau an Fic.1.—Piscicola olivacea, n. sp. Diagram showing dorsal pattern, somites (numbered in Roman figures), eyes, etc An Anus Pp. ves, z land D. ves, IT, Pulsating vesicles (shown only on one side, for the sake of clearness) of first and eleventh pairs. 514 Memoirs of the Indian Museum. (Vor. V, (g) “Rambha Bay. Among weeds.” ) “Station 51. 5-7 mi. E. by N. of Patsahanipur.” ) “Station 64. Ghiakhala headland and neighbouring islands.” ) “Station 73. 7-4 mi. E.4S. of Patsahanipur..On Hypolophus sephen (Forsk.)” ) “Station 54. 2 mi. N.E. by N.4N. of Kalidai. (2) “Off Barkul. From Chatoessus chacunda.”’ [A leech apparently identical with Piscicola olivacea is very abundant among algae on the shore of Barkuda Island, Chilka Lake, and I have also obtained specimens from a small pool of almost fresh water on the same island. Inthis pool the only vertebrates were frogs (Rana cyanophlyctis) —N. A.] Additions to this preliminary description of P. olivacea will, it is hoped, appear at a later date. N.B.—Along with the examples of P. olivacea were several specimens of another and apparently new species of leech closely resembling the former in size and form, without eyes and having at least fourteen rings to the complete somite. I give here Dr. Annandale’s notes upon it, and it is doubtful if the material will yield much more information ; the muscular expansion of the margins of the body giving the fin-like appearance described below is not evident after death. “Leeches on lips of Hypolophus sephen (Forsk.) .. attached outside, close to junction of skin and teeth on both upper and lower jaws. No visible eye spots. Specimens seem to have a lateral fin on anterior two-thirds of body. Colour whitish Occasionally minute spots of faint pink on dorsal surface.” Collected at “Station 43. 8 mi.S.S.W. of Kalidai, Chilka Lake. Sp. gravity of water 1:°007— TROT Placobdella emydae, n. sp. (FIG. 2). EXTERNAL FEATURES—FORM, SUCKERS, ANNULI, SOMITES, EYES. Body flattened, in extension elliptic-lanceolate, with the head region slightly dilated. Dorsal surface with a roughened appearance due to the presence on each annulus of numerous small papillae. Ventral surface without papillae and smooth. The dorsal papillae vary in size and arrangement. A row of about 16-20 papillae are present on the first and third ring of the complete somite, whilst about twelve are more regularly disposed upon the middle ring. The middle ring of the somite (which lodges a ganglion of the ventral chain) has, amongst the others, three pairs of meta- meric papillae ; a paramedian, an intermediate and a paramarginal pair, the inter- mediate pair being the largest. N.B.—AIl papillae tend to disappear and may even be absent in improperly preserved specimens. Anterior sucker pierced on its anterior lip by the mouth-opening ; with a shallow 1920. | Fauna of the Chitka Lake : Hirudinea. 515 imperforate interior cup having a finely rib- bed surface somewhat resembling that of the tip of the human finger. Posterior sucker circular, centrally at- tached, narrower than the greatest width of the body, broader in proportion to the body- width in immature than in adult examples. Annuli 71, two being pre-ocular. The second and third annuli may show some division at their margins: the fifth is confluent with the free ventral edge of the anterior sucker: annuli 6 and 7 are separate above but some- times so slightly divided as to appear as one ring: annulus 6 disappears ventrally, leaving the seventh (distinguished by papil- lae) to form the first ventral annulus fol- lowing the anterior sucker. Complete somite formed of three annuli. Somites i, ii and iii uniannulate; iv, xxv, xxvi and xxvii biannulate. The twenty somites v-xxiv are complete with three: rings. Eyes, one pair, closely approximated, situated normally in the third annulus, but sometimes between the third and second rings. REPRODUCTIVE ORGANS, DIGESTIVE TRACT, NEPHRIDIOPORES. Male genital orifice situated between annuli 26 and 27, that is, between somites xi and xii; female genital orifice situated two rings behind the male, between annuli 28 and 29, being the second and third rings of somite xii. ; Epididymis large, long and thrown into complex coils and loops, the last loop ex- tending as far down as the xvth somite, and returning upwards through two somites before again turning down to join the vas 45 X1X¢ XXC 60 hr ' 2 =. = Fa >65 Las} = = = i=) EN 5 2 FC 2 = 2 2 EN, \ Nu Vp : Ver N \ roa = PAST = Les = : | EN À [| XXVII- NT on Fic. 2.—Placobdella emydae, n. sp. Diagram showing annulation, genital organs, digestive tract and other internal and external features. The testes are shown on one side only, and the nephridiopores on the other for the sake of clearness, and other organs have been treated in the same manner and for the same reason. Somites numbered in Roman, and rings in ordinary figures. prb: Proboscis, sal: g, Salivary glands. Cr. 1 and 7, First and seventh diverticula of Crop. St, Stomach. ini, Intes tine. an, anus, gang, one of the ventral ganglia, 7. 1 and n. 14, Nephridiopores of the first and fourteenth pair. Ef, Epididymis. vd, vas deferens. 7, and t, 6, Testes of the first and sixth pairs. ov, ovary. A 516 Memoirs of the Indian Museum. - [VoL. V, deferens. The large epididymes form a characteristic feature of the species and are not apparently symmetrical: one is usually shorter than the other. Testes 6 pairs, of kidney-like form. The ovaries are a pair of simple sacs prolonged anteriorly into horn-like processes. Crop with seven pairs of lateral diverticula of fairly simple form and all posterior to the male genital orifice: the first pair reflected anteriorly, the last and longest pair reflected posteriorly. Stomach with the usual four pairs of lateral diverticula. Ex- tensive salivary glands are present and the mouth, as already noted, occupies the nearly terminal position characteristic of the genus. Nephridiopores, 14 pairs, situated upon annuli 16, 19, 22, 34, 37, 40, 43, 46, 40, 52,55, 58, 61, and 64. A pair thus occurs upon the middle ring of somites viii-x and xiv-xxiv, none being present in somites xi-xili. Each pore perforates the middle part of the ring in which it is located and lies about midway between the ventral margin and middle line. HABITAT, HOSTS. This little species has so far been found upon Mud-turtles of the genus Emyda and is not peculiar to Lake Chilka. In addition to (a) four individuals from this lake taken from Emyda granosa, Gtinth., subsp. intermedia, Annandale, I have the ex- amples detailed below. (Information from notes enclosed with the material.) (b) “From Emyda granosa, Günth. Outskirts of Calcutta.” (c) “From carapace of Emyda granosa, Günth. Gatiagurh, Dist. Hughly, Bengal. (d) ‘From carapace of Emyda granosa, subsp. intermedia, Annandale. R. Mahanaddi, Sambalpur, Orissa.” (e) “ From Emyda granosa, subsp. intermedia, near Purulia, Chota Nagpur Div., Bihar.”’ (f) “From the turtle Emyda vittata, Nagpur, C.P.’ Dr. Annandale adds “ The mud-turtle that occurs at Nagpur is probably E. granosa intermedia, and not Ep vittata.” The exampies from Lake Chilka, according to Dr. Annandale’s notes, were collect- ed at “Station No. 52, 4-9 mi. E.4N. of Patsahanipur .. sp. gravity of water 1:007—1'011.” This leech, therefore, is able, like its hosts, to accommodate itself to water of a certain degree of brackishness. COLOUR AND SIZE. For information as to the colour of this species, I am dependent upon Dr. Annandale who has provided me with notes taken from five different living indivi- duals. From these notes, I am enabled to state that the ground colour varies from greyish green to pale olive brown or pale brown, the gorged crop appearing dull green through the semi-transparent body. Dorsal surface with a distinct white median line, closely speckled with white and a darker green, the white specks being prominent on the margins of the body. Ventral surface olivaceous. Posterior sucker with whitish radiating lines, the inter- > 1920. | Fauna of the Chilka Lake : Hirudinea. 517 spaces sometimes minutely speckled with grey-green. One of my examples in alcohol shewed that the speckled or tessellated appearance of the dorsal surface was due to the numerous small papillae covering it, which were tipped with white. Size of the largest specimen examined: total length 13°5 mm., greatest width 9 mm. LITERATURE. APATHY, S. (1888). Süsswasser-Hirudineen. Zool. Jahrb. III, p. 790. CASTLE, W. E. (1900). Some North American Fresh-water Rhynchobdellidae, and their parasites. Bull. Mus. Zool. Harvard, XXXVI, No. 2, pp. 42-43, pl. viii, fig. 38. MEMOIRS OF THE INDIAN MUSEUM, VOL. V. FAUNA OF THE CHILKA LAKE No. 8. MAY, 1921. PAGE Amphipoda = oye ae Ke 519 a os 450 On a larval Cestode from the umbrella of a Jelly-Fish Polychaeta of the Chilka Lake and also of Fresh and Brackish Waters i ie 8503 in other parts of India — AT TN nstirge N LS VEN Ss 9 3K | R GS fee ct ae f n FD (x PS à es 2 N ee Calcutta: PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA, PRINTED AT THE BAPTIST MISSION PRESS. 1921. Price Jew Rupees. fe Je (3 ; 4 : ee we | ' ‘pt j r , Le { j x u 3 $ : ¥ x + 2 7 ; i Wu ñ { | > ‘ ; 4 Re Ke { on ’ a “ > x 4 * J 4 Mi * incl Wg PME à ” ee oe “fh # et a lp y \ , k N N + J £ r J “ A bi Tg! f ! { ; 5 j { K j j | CAT ¥ 2 £ % i 4, fi { Kari Fy £ ¢ M may é 2 E ER Dr. 2 ÿ & AT HAN 1 > 1 Ÿ Pied Ne eh "able oe FAUNA OF THE CHILKA LAKE. AMPHIPODA. By CuAs. CH ton, M.A., D.Sc., M.B., C.M., LL.D., F.L.S., CAMES". F.N.Z. Inst., Hon. Member Roy. Soc. N.S.W., Professor of Biology, Canterbury College, University of New Zealand. \ CONTENTS. Page Introduction ae we Ä ie ay D NE oa List of Species with Distribution ce = as ie 22 Ampelisca pusilla Sars bs er >. A 2 oe yr 528 Amphilochus brunneus Della Valle u = Ne A at Idunella chilkensis sp. nov. .. 2 Bis aS ba? tee 525 Perioculodes longimanus (Bate and Westw.) I + I 77 Synchelidium haplocheles (Grube) IR: Er Me its an 528 Paracalliope fluviatilis (G. M. Thomson) .. 7 Ss or fe ZO) Niphargus chilkensis sp. nov. il, he Fe de ei SI Melita inaequistylis (Dana) .. ie ue ae RATE: > = 0585 Maera othonides, Walker Pe Le Le À + Loh 535 Quadrivisio bengalensis Stebbing se Be de > fe. 2537. Orchestra platensis Kroyer .. a te 93 a .2 2588 Talorchestia martensii (M. Weber) Be es ab: 2 one SAE Hyale brevipes Chevreux au er ae ihe Be LS 12545 Grandidierella megnae (Giles)... + IK Bie ESS AS Grandidierella gilesi sp. nov. .. ie Le se + 2 552 Photis longicaudata (Bate and Westw.) .. a Rs ue OMS SA) Corophium triaeonyx Stebbing EN i m e EN Bibliography à: à, = oe hee 10550 AMPHIPODA. By CHAS. CHILTON. INTRODUCTION. The number of species of Amphipoda collected from Chilka Lake is not great, comprising only seventeen (17) species, but the collection is nevertheless of very considerable importance and interest. Most of the species were gathered at many different localities and often in very great numbers. They are nearly all of small size and in several cases the task of sorting them out and of distinguishing allied species was somewhat laborious. Most of the specimens have been referred to species already known, but I have had to establish three new species. A considerable amount of additional information is also given with regard to species previously described. Several questions of interest arise with regard to the distribution of the species. Thus Ampelisca pusilla Sars, which was originally described from deep waters in Arctic Seas, is found in great abundance in the Lake. It also occurs in the river Ganges at Buxar about six hundred miles from the mouth and had previously been recorded from the east coast of Australia by Stebbing. The Indian specimens differ from the Arctic in having corneal lenses present, but in all other respects agree very closely with the Arctic forms. Quadrivisio bengalensis Stebbing, originally described from Port Canning, Lower Bengal, evidently occurs in great abundance in some parts of the lake. It is found also in the island of Zanzibar, East Africa, and Dr. Annandale has sent me specimens from four different localities in the Tale Sap, Siam. Grandidierella megnae (Giles) seems hardly distinguishable from G. mahafalensıs Coutiére from Madagascar. The original specimen from which Dr. Giles described the species was taken at Megna Shoals, Bay of Bengal. In this species there appear to be two forms of the male, differing from one another in the characters of the first gnathopod. An allied form which agrees. with most of the characters of Grandidi- evella megnae except in the second gnathopods, is described below as a new species, G. gilesi sp. nov. | Melita inaequistylis (Dana) is a very widely distributed species occurring in India, South Africa, New Zealand, etc. Perhaps the most striking example as regards the geographical distribution of the Chilka Lake Amphipoda is the species Paracalliope fluviatilis (G. M. Thomson). This species was described many years ago from New Zealand, where it is the common form occurring in nearly all freshwater streams, though it is also found in brackish waters and sometimes in water that is perfectly salt. It is evidently abundant in 522 Memoirs of the Indian Museum. [VoL. V, - Chilka Lake and in the collection there are also specimens from the lower reaches of * the river Adyar near Madras. The species is almost certainly identical with Pherusa australis Haswell described from Botany Bay, New South Wales, and this helps to connect the extreme points at which the species is found.' This case forms a parallel to that of Melita inaequistyhs, which is found in brackish waters in New Zealand and in India and Ceylon, but the latter species has been recorded from a greater number of marine localities. Discussion of the importance of the distribution of the species, as mentioned above, must be held over for a later occasion, but it may be mentioned in connection therewith that X1phocaris(—Paratya) curvirostris (Heller), the freshwater shrimp com- mon in New Zealand streams, also occurs in Assam and that Paracorophium excavatum (G. M. Thomson), an Amphipod which occurs in brackish and fresh water on the coast of New Zealand, has recently been found in similar localities in Brisbane river, Queensland.” 2 One of the most interesting Amphipods in the Chilka Lake fauna is Niphargus chilkensis sp. nov. This form was found at various localities apparently under the same conditions as the other species. It possesses eyes, though these seem to be somewhat imperfect. It differs in several points from the species of Niphargus found in Europe and Northern Africa. A closely related species has recently been sent to me from the underground waters in the Philippine Islands. Under each species I have given only those synonyms and references that appear to be of importance in connection with the present paper. The names of authors followed by a date refer to the bibliographical list on p. 556. I wish to express my grateful thanks to Dr. Annandale for the opportunity of examining and reporting upon this fine collection and to Miss E. M. Herriott, M.A., Assistant at the Biological Laboratory of Canterbury College, for preparing the figures and assisting me in other ways. LIST OF SPECIES, WITH DISTRIBUTION. I. Ampelisca pusilla Sars. Chilka Lake; Buxar, R. Ganges; Arctic Seas ; North Atlantic; East coast of Australia. 2. Amphilochus brunneus Della Valle. Chilka Lake; North Sea ; Mediterranean. 3. Idunella chilkensis sp. nov. Chilka Lake. 4. Perioculodes longimanus (Bate & Westw.). Chilka Lake; Ceylon; Arctic Ocean ; North Sea ; North Atlantic; Mediterranean. 5. Synchelidium haplocheles (Grube). Chilka Lake; Ceylon; North Sea ; North Atlantic; Mediterranean. 2x 6. Paracalhope fluviatilis (G. M. Thomson). Chilka Lake; Philippine Islands ; New Zealand ; Australia; (in fresh and brackish water). 7. Niphargus chilkensis sp. nov. Chilka Lake. ! Since this was written the species has been found in the Philippine Islands. * See alsoS. Kemp’s remarks on the distribution of certain Onychophora and other terrestrial animals (Rec. Ind. Mus., Vol. VII, p- 491). 1921.] Fauna of the Chilka Lake : Amphipoda. 523 8. Melita inaequistylis (Dana) Chilka Lake; Ceylon; South Africa; New Zealand. 9. Maera othonides Walker. Chilka Lake; Ceylon. 10. Quadrivisio bengalensis Stebbing. Chilka Lake; Port Canning, Gangetic Delta ; Zanzibar Island ; Talé Sap, Siam. II. Orchestia platensis Kroyer. Chilka Lake; Philippine Islands ; widely distri- buted on warmer shores of America; Mediterranean ; Hawaiian Islands ; Tonga, Low Archipelago, etc. 12. Talorchestia martensii (M. Weber). Chilka Lake; Flores in Malay Archipelago. 13. Hyale brevipes Chevreux. Chilka Lake; Ceylon; Laccadive Archipelago : Seychelles. | 14. Grandidierella megnae (Giles). Chilka Lake ; Bay of Bengal; Madagascar. 15. Grandidierella gilesi sp. nov. Chilka Lake. 16. Photis longicaudata (Bate & Westw.). Chilka Lake; Philippine Islands; North Atlantic; North Sea; South Africa. 17. Corophium triaeonyx Stebbing. Chilka Lake; Ceylon. Ampelisca pusilla Sars. Ampelisca pusilla Sars, 1891, p. 181, pl. 63, fig. 2. Ampelisca pusilla Stebbing, 1906, p. 105; IQIO, p. 576. Ampelisca chevreuxi Walker, 1904, p 254, pl. 3, fig. 15. Localities : — 1 mile S. of Kalidai. Several. 3 to 2 miles S.E. by E. 4 E. of Patsahanipur. Several. 2 to 8 miles N.E. & E. of Kalidai. Many. 2 to ı miles S.E. by S. of Patsahanipur. Several. 4 miles N.E. by & E. of Kalidai. Many. 2 to 1 miles S.E. by S. of Patsahanipur. Several. These specimens are all small, not more than about 5 mm. in length. I feel pretty confident in referring them to Ampelisca pusilla. Both male and female agree closely with the figures given by Sars in the proportions of the antennz and the shape of the appendages and particularly in the fact that the fourth segment of the pleon is scarcely carinate in the female but is distinctly carinate in the male, the projection agreeing closely with Sars’ figures. The specimens differ, however, from Sars’ description in having corneal lenses present in the normal manner. Stebbing has identified this species from Australia and remarks that in his specimen also the corneal lenses are present. Sars says that the species occurs off the coasts of Norway in considerable depths. The absence of corneal lenses in his ' specimens is doubtless due to a degeneration of the eyes caused by the depth at which it lives. I have little doubt that A. chevreuxi Walker should also be referred to this species. ‘The Chilka Lake specimens agree closely with Walker’s description except in having the first antenna about as long as the second, while he describes it as 524 Memoirs of the Indian Museum. [Mou V, reaching to one-third the length of the last joint of the peduncle of the lower antenna. The two antennæ are subequal in the female specimens I have examined, but the upper one is shorter than the lower in the male. In some other species the upper one varies much in length at different stages of development. As regards the telson the vari- ous descriptions agree as to its being divided almost to the base and having the divisions pointed. Walker states that there are three spines before the point on the outer margin. Sars gives two and Stebbing says that there is only one apical spinule in the Australian specimen; there is only one in the Chilka Lake specimens that I have examined. Sars says the telson is without dorsal denticles ; in one female examined I found one minute dorsal denticle on one lobe but none on the other. These and other points in the structure of various species of Ampelisca have been dealt with by myself in another paper.‘ Stebbing says Ampelisca pusilla closely approaches A. rubella A. Costa which has been described from the Mediterranean, and Sars says it is nearly allied to A. amblyops which again somewhat resembles A. anomala but differs in the absolute want of any corneal lenses. A. anomala has been recorded from South Africa by Stebbing (I9I0A, p. 450). Dr. Annandale has sent me specimens from the River Ganges, Buxar, 600 miles up the river, which seem to be the same as the Chilka Lake specimens; in some the first antenna is shorter in comparison with the second antenna; the eyes are dis- tinctly red, in some the whole eye is red, in others patchy. [Taken commonly in the main area on or just above a muddy bottom in 4 to 8 feet of water some distance from shore. N.A.] Amphilochus brunneus Della Valle. Amphilochus brunneus Della Valle, 1893, p. 596, pl. 4, figs. 5; pl. 20, figs. 1-15. Amphilochus brunneus Stebbing, 1906, p. 151. Amphilochus brunneus Chevreux, III, p. 192. Amphilochus neapolitanus Walker (part), 1001, p. 301. Amphilochus neapolitanus Walker, 1904, p. 255. Amphilochus melanops Walker, 1895, p. 298, pl. 18, fig. 12 and pl. 19, figs. 13-15. Localities :-— 2-8 miles N.E. + E. of Kalidai. Several. 1 mile E. by N. of Patsahanipur. One. 2-6 miles E. by S. 4 E. of Patsahanipur. Several. Near Samal Island. Several, from Medusa. Barkul. Several, from large Medusa. I have no hesitation in referring these specimens to the species named above. They agree closely with the description and figures given by Della Valle and Walker. Walker has united both A. melanops and A. brunneus with A. neapolitanus Della Valle, but in that species as described by Della Valle the process of the fifth joint of the © ! The identity of the two Amphipods, Ampelisca eschrichtii Kroyer and A. macrocephala Liljeborg. Jour. Zool. Research, Vol. IT, p. 75. 1021. | Fauna of the Chilka Lake : Amphipoda. 525 second gnathopod overlaps the palm and Chevreux, who was able to examine a large number of specimens, found this to be the case in all his specimens of A. neapolitanus, even those only 1 mm. long, while in A. brunneus it varied between the half and two- thirds and did not overlap the palm. Hence I refer theChilka Lake specimens to A. brunneus, for in none of them does the process overlap the palm. Walker’s speci- men from Ceylon doubtfully referred to A. neapolitanus is probably the same. [This species was habitually taken on the subumbrella and among the tentacles of the only large medusa found in the lake, Acromitus vabanchatu, Annandale. The amphipods were almost invariably present in this situation in adult medusae. N.A.] Idunella chilkensis sp. nov. (Text-fig. I). Localities : — 1 mile E. by N. of Patsahanipur. Five males, one female. 2-6 miles E. by S$. 4 S. of Patsahanipur. One male, one female. Specific Diagnosis. Male. Pleon segments 2, 4 and 5 produced into minute dorsal teeth. Rostrum short or absent. First side plate large, broadly rounded in front, side plates I to 3 with minute denticle at lower hind corner. Postero-lateral angle of third pleon segment acute, a small sinus between the point and the convex hind margin. Antenna I (fig. ra) about three-fourths as long as the body, stout; first joint of peduncle about as long as the second but much stouter, third joint very short, not longer than first joint of flagellum; flagellum stout, consisting of about 35 joints, accessory flagellum small, of two slender joints. Antenna 2 (fig. ra) shorter than the first antenna, ultimate and penultimate joints of peduncle subequal, flagellum stout, about as long as peduncle. Mouth parts closely resembling those of I. aequicorms. First gnathopod (fig. 1d) large, basis slender, with two or three long setules near base of hind margin and a large group about the middle of anterior margin: ischium and merus short, subequal, two long setules at distal angle of merus; carpus very short, triangular, propod very large, longer than the rest of the limb, broadly oval, anterior margin regularly convex and without setae, palm very oblique, irregularly defined by 3 or 4 stout setules, having near the finger a prominent lobe ending in two or more blunt teeth, a depression near the lobe followed by a convex serrated and setulate portion leading to the defining setules; hind margin much shorter than palm, convex, fringed with minute setules; finger long and strongly curved, having a concavity near the base followed by a slight prominence, inner concave margin slightly irregular. Second gnathopod much smaller, carpus produced near the antero-distal angle ; propod oval, longer and broader than carpus, anterior margin convex, bearing tufts of slender setules; palm not very oblique, convex, with minute setules; hind margin longer than palm, with 4 or 5 tufts of setules; finger strongly curved, fitting closely on to the palm. 526 Memoirs of the Indian Museum. [Vor "ar Peracopods x and 2 as in I. aegmcornis, slender, with long slender dactyls. Peraeopods 3 to 5 increasing in length posteriorly, basal joint not greatly expanded, its hind margin straight and sharply serrate. Third uropods (fig. ıg) with the inner branch much larger and ne than the L ( Ar eo CSN oe 4 1 (os h Lf ER SL GE Vos = FY Fic. 1.—Idunella chilkensis sp nov. Antennee of male. e. First gnathopod of female. Upper antenna of female. J. Second gnathopod of female. a b c. Lower antenna of female. g. Third uropod of female. d. First gnathopod of male. h. Telson of female. outer, outer with tufts of setules on both margins, inner one with fewer setules and these confined to the outer margin and apex. Telson cleft to the base, each lobe narrow with outer angle strongly produced into a sharp tooth, two stout setules arising from the concave extremity. Female, similar to the male except in the antennæ and gnathopods 1021. ] Fauna of the Chilka Lake : Amphipoda. 527 Upper antenna (fig. 1b) shorter, subequal with the lower (fig. 1c), flagellum not so stout as in the male. First gnathopod (fig. 1e) much larger than the second, but not so large as in the male; propod oval, palm regularly convex, fringed with stout setules, hind margin not so abundantly fringed with setules as in the male; finger smaller, inner margin regularly concave and bearing a few minute setules. Second gnathopod (fig. 1f) as in the male, but with propod rather smaller. Length of body, 4 mm. Colour. Pale yellow, with a few darker markings. Remarks. This species agrees well with the characters of the genus and evidently comes pretty close to I. aequicorms Sars, from Arctic Seas. It differs in the male chiefly in the greater length of the antennæ, especially of the upper, and in the shape of the first gnathopod. In both sexes it differs also in the smaller accessory flagellum, the shape of the carpus of the second gnathopod and in the unequal branches of the third uropod. The sexual differences appear to be confined to the antennz and the gnathopods. In the male specimen examined the third uropods also differed from those of the female shown in fig. 1g in having both margins of the inner branch free from setae _ except for the small tuft near the base of the inner margin and the two or three towards the apex on each margin. [Only taken in the main area of the lake on a muddy bottom some distance from shore in 44 to 54 feet of water. N.A.] Perioculodes longimanus (Bate and Westw.). (Text-fig. 2.) Perioculodes longimanus Stebbing, 1906, p. 237. Perioculodes longimanus Sars, 1892, p. 313, pl. 110, fig. 2; pl. 111, fig. 1. Monoculodes megapleon Giles, 1888, p. 235, pl. 7, fig. 12. ? Oedicerus puliciformis Giles, 1888, p. 248, pl. 7, fig. 5 and 6. Localities :— 2 to 4 miles N.E. #E. of Gunta Sila. Many. 1 mile N.N.E. of Breakfast Island. Several. 1 to 9 miles N.E. LE. of Breakfast Island. Several. 2 to 8 miles N.E. LE. of Kalidai. Several. 1 mile E. by N. of Patsahanipur. Several. 2 to 6 miles E. by $. 4S. of Patsahanipur. Several. 4 miles N.E. $E. of Kalidai. Several. I to 2 miles S.E. by S. of Patsahanipur. Several. This species was obtained in great abundance at some of the stations, the male specimens, however, being very rare. I refer them to Perioculodes longimanus (Bate and Westw.) with some hesitation. They agree on the whole well with Sars’ figures except as regards the antennæ (fig. 2), which in both sexes have the third joint of the peduncle of the upper antenna considerably shorter than the second and the 528 Memoirs of the Indian Museum. - [Vor. V, flagellum rather stout and usually with 8 or 9 joints; the second antenna has the last two joints of the peduncle somewhat thickened and provided with long setae ; the flagellum in the female is shorter than the peduncle and consists of four joints; in the male it is very slender and very long, the whole antenna being about three- fourths the length of the body. The branches of the uropods are slender, sub- equal, and with the extremities very acute; a few setules are present on the pedun- cles and sometimes also on the branches, particularly on those of the third ; Sars says those of the third are quite unarmed. Walker says that his P. serra from Ceylon much resembles P. longimanus but he describes the rostrum as being much longer than it is in the Chilka Lake specimens. The differences in the proportions of the segments of the peraeon are perhaps not important but he describes the outer ramus of the first uropod as half as long as the inner and the upper margins of the rami in adults strongly serrate. In my specimens the branches are subequal and though there may be a slight appearance of serration Dolls ee iat formed by the shallow notches from which the antennee of female. setules arise these are much less numerous and less conspicuous than in his figure. Monoculodes megapleon Giles was described from a single specimen obtained off Chittagong. From Dr. Giles’ description it seems to be the same species as the Chilka specimens and in some points to agree better with P. longimanus. Stebbing (1906, p. 238) says of it “perhaps identical with P. longimanus.”’ Oedicerus puliciformis Giles (1888, p. 248) is merely mentioned by Stebbing (1906, p. 742); if we disregard the difference in the description of the eyes, which may be due to the condition of the specimens, there does not seem to be anything in the figure to prevent Dr. Giles’ specimen being a male of the species now under consi- deration, 2.e. P. longimanus, the enlarged figure he gives of the terminal portion of the gnathopod agrees precisely with that of the Chilka specimens. Unfortunately for this suggestion, however, his specimen was described as a female carrying ova, though it was only 2 mm. long. [A very abundant species in the main area on a muddy bottom some distance off shore. N.A.] Synchelidium haplocheles (Grube). Synchelidium haplocheles Stebbing, 1906, p. 242. Synchelidium haplocheles Chevreux, IQII, p. 206. Synchelidium brevicarpum Walker, 1904, p. 263. Locahty -—-2 to 8 miles N.E. LE. of Kalidai. A few specimens. I think these specimens must be referred to the species named above. ‘The male specimen dissected has the rostrum and antenna agreeing exactly with the description given by Stebbing and with Sars’ figures, except that the first joint of the 1921.] Fauna of the Chilka Lake : Amphipoda. 529 flagellum of the first antenna is not densely clothed with fine hairs; perhaps the specimen is not fully matured, though the second antenna has the long slender flagellum characteristic of the male. The gnathopods are in minute agreement, even as regards the armature of the palm of gnathopod 1. So are the peraeopods, uropods and telson. The other specimens also agree, but in some of them, which have not the characteristic second antenna of the male, this appendage is distinctly longer than antenna I. Stebbing says, “Antenna 2 in the female shorter than antenna 1.” Possibly my specimens are immature males. Walker has recorded this species from Ceylon saying that his single specimen agrees with British examples even to the dark brown blotches on the fifth and sixth segments of the peraeon. My specimens are whitish with irregular reticulate patches of black on various parts of the hody and appendages, the colour being well retained in the spirit specimens; the eye is black. [Taken among filamentous algae on a muddy bottom in 5 to 6 feet of water. N.A.] Paracalliope fluviatilis (G. M. Thomson). (Text-fig. 3.) Calliope fluviatilis G. M. Thomson, 1879, p. 240, pl. toc, fig. 4, 4a-c. Paracalliope fluviatilis Stebbing, 1906, p. 297. Paracalliope fluviatilis Chilton, 1g09A, p. 55. Pherusa australis Haswell, 1880, p. 103, pl. 7, fig. 1. Localities :— Off Sama] Island, 8-15 ft., 22-ix-13. Several. Off Barkul, 3-4 ft., 21-31-vii-13. Several. Off Barkul, at edge of lake, 2I-vii-13. Several. Barkul, 18-iv-05. Several. East side of Rambha Bay. Several. Adyar River, outskirts of Madras town, 3-4 ft. Several. However unlikely it may at first appear these specimens agree in size, structure and colour with the species, Paracalliope fluviatilis (G. M. Thomson), which is the common one inhabiting freshwater streams in New Zealand. When first sorting out the Chilka Lake specimens I was struck by the superficial resemblance of some of the Specimens in the pale orange colour and the long terminal peraeopods to the form I was familiar with in New Zealand and careful comparison of specimens has convinced me that they must be referred to the same species. I have previously pointed out (1909A, p. 55) that the New Zealand species, in addition to inhabiting the fresh- water streams, is also to be found in brackish water and at times in water that is quite salt. It is found all over New Zealand and I have recently had specimens sent to me from Cape Maria van Diemen in the very north. I have also specimens collected in quite salt water in Auckland Harbour as well as those previously recorded from Dunedin Harbour. Stebbing (1906, p. 297) referred to this species the form described 530 Memonrs of the Indian Museum. [Vor. V, under the name Pherusa australis by Haswell from Botany Bay, New South Wales. Unfortunately I have been unable to secure specimens from Australia and the types of Haswell’s species are no longer available, but after carefully going through Has- well’s description I am convinced that Mr. Stebbing is right; the description on the whole agrees well with the specimens and seems to be confirmed by the statement and by the figure showing that there are three spines on the inner border of the inner ramus of the third uropod; for these are certainly present as shown in Haswell’s Se Fic. 3.—Paracalliope fluviatilis. a.—Whole animal, side view. c.—Lower antenna. b.—Upper antenna. d.—First gnathopod of male. e.—Second gnathopod of male. figure in ordinary fully adult specimens, though the number may sometimes be greater or less than three. Probably the species will be found to occur in other fresh- water or brackish localities on the east coast of Australia or between Australia and India.! : : en OT ue ER er me the MS. Santa ne the remarks made above I received from Professor C. F. Baker of Los Banos, Philippine Islands, a few amphipods from Nasugbu, south coast of Luzon, most of which proved to be specimens of Paracalliope fluviatilis (G. M. Thomson) quite 1921.] Fauna of the Chilka Lake : Amphipoda. 531 Paracalliope fluviatilis was described by Thomson as long ago as 1879. Unfortun- ately his figures were so greatly reduced in reproduction that they do not show clearly the structure of the various appendages and no other figures have yet been published. I have, however, the tracings that were made of Mr. Thomson’s original figures for the purpose of reproduction and find that the main points in the structure of thegnathopods and of the telson and uropods are quite clearly shown. A descrip- tion was given by Stebbing in 1906 and on the whole agrees well with the speci- mens. In 1899 Stebbing had established the genus Paracalliope for the species. There is one point in his generic diagnosis that requires alteration, for calceoli are certainly present in the males on some of the basal joints of the flagellum in both upper and lower antenne. The species can generally be easily recognised by the greatly elongated fifth peraeopods and by the peculiar inverted position of the second gnathopod ; this appendage seems to have a very loose articulation between the ischium and the carpus, so that the distal portion of the limb often faces in the direction opposite to the normal one (see fig. 3a). Stebbing has pointed out that this recalls the curious torsion in the first gnathopod of Trischizostoma nicaeense. I give figures of the whole animal (fig. 3a) and of the antenne (figs. 30, c) and the gnathopods (figs. 3d, e) which will render a detailed description unnecessary. It is difficult to represent the gnathopods accurately, for as Stebbing points out there are two margins to the palm and the outline therefore varies according to the posi- tion in which the gnathopod is mounted. Stebbing places the genus under the family Calliopiidae, and this is probably its proper position, though the elongated fifth peraeopods are peculiar and the characters of the family are Somewhat indefinite. It was ‘doubtless from the elongated fifth peraeopods that Stebbing suggested that the species Oedicerus novi-zealandiae Dana was probably identical with Paracalliope fluviatilis. I have, however, pointed out elsewhere that Oedicerus novi-zealandiae is a distinct species and is in all probability identical with the species afterwards described by Stebbing as Carolobatea schneider: (1909, p. 620). [Not uncommon on a muddy bottom off shore in the main area. N.A.] Niphargus chilkensis sp. nov. (Text-fig. 4.) Localities :— Off Samal Island, 3-15 it., 22-ix-13. One. Off Barkul, 21-vii-13. Two. One mile S. of Kalidai. Several. 4 to 9 miles E.4S. of Barkul bungalow. Several. 3 to 2 miles S.E. by E.4E. of Patsahanipur. Four. similar to those from Chilka Lake and New Zealand. These are stated to have been collected in “shallow water,’ but there is nothing said as to whether the water was fresh, brackish or marine. Along with them was a single specimen which, though smal! and perhaps immature, appears to be Photis longicaudata (Bate and Westw.). 532 Memoirs of the Indian Museum. [Voz. V, 1 mile E. by N. of Patsahanipur. Five. 2 miles E. by S.iS. of Patsahanipur. Several. I to 9 miles N.E. by E. of Kalidai. Several. Specific Diagnosis :— Body long and narrow, side plates small, much shallower than the segments. Pleon segment 3 with the postero-lateral angle quadrate. Eyes small, irregular, apparently imperfect. Tpper antenna about three-fourths the length of body, first and second joints of peduncle subequal, elongate; third short, flagellum slender, longer than peduncle, accessory flagellum of two slender joints. Lower antenna a little longer than peduncle of upper, last two joints of peduncle subequal, flagellum shorter than last joint, consisting of one long joint followed by two or three indistinct ones. First gnathopod with the merus produced posteriorly into a rounded lobe, carpus much longer than the propod, with numerous tufts of setules on the posterior margin and on the surface near to it; propod widening distally, palm neatly transverse, evenly convex. Second gnathopod much larger than the first, carpus short, about one-third the length of the propod ; propod irregularly oval, palm oblique, sinuous, about equal in length to the hind margin; finger strongly curved, bulging on the inner margin near its base. Third, fourth and fifth peraeopods increasing in size posteriorly, basal joint in the third narrow, slightly wider in the fourth and greatly widened and enlarged in the fifth, its posterior margin irregularly serrate, most distinctly so in the fifth peraeopod; in the fifth peraeopod the merus is dilated posteriorly. Third uropods greatly elongated, peduncle longer than the telson, outer branch formed of two subequal joints, somewhat broadened, inner branch small, tapering, tipped with one or two setae. ‘Telson cleft to the base, each lobe narrowing posteriorly and bearing a stout setule at the extremity. Colour (in spirit), whitish. Length of body, about 10 mm. I have referred this species to the genus Niphargus with which it agrees in most characters. It differs, however, from Stebbing’s generic diagnosis (1906, p. 405) in the following points :— I. The eyes are moderately well developed. 2. The third joint of the mandibular palp is hardly longer than the second. 3. The inner plate of the first maxilla is large and bears numerous setae. 4. The outer plate of the maxilliped bears numerous setae all of the same character, but has no spine teeth. 5. The second gnathopod is larger than the first and differently shaped. The following are additional notes cn the structure of this species. Most of the figures are taken from the type specimen which was the first one examined. Some specimens, rather larger, show slightly more developed characters in the antennae, gnathopods and in the fifth peraeopods in which the basal joint may be larger in proportion to the rest of the limb than is shown in figure AP. The basal portions of the antennæ are shown in figure 4b from which the 1021. ] Fauna of the Chilka Lake: Amphipoda. 533 J proportions of the different joints can be readily seen. The tufts of setae on the second joint of the peduncle of antenne 1 towards its extremity seem to be a characteristic feature and may be more developed than is shown in the figure. The upper lip (fig. 4c) is regularly rounded and bears a few minute setae. In the mandible (fig. 4d) the molar tubercle is prominent and projecting, the cutting edge and spine row of the normal character, the palp has the first joint short, the next two joints subequal, bearing few setae except the distinct tuft at the end of the third joint. The first maxilla (fig. 4f) differs considerably from the generic description given by Stebbing in having the inner lobe broad and well developed and fringed with about 12 plumose setae. The second maxilla (fig. 4g) has the two lobes subequal, both with apical setae and the inner one with a few setae on its inner margin. The maxilliped (fig. 44) has the inner lobe large with three stout spinules and a number of plumose setae, the outer lobe reaches only about half-way along the carpus and has its inner margin provided with numerous short setae, none of which are developed into spine-teeth. In the first gnathopod (fig. 4k) the merus is produced posteriorly into a slight rounded lobe and appears to be covered with very minute setae, the rest of the appendage seems to be normal, having the characters already mentioned in the specific description. The second gnathopod differs de froın the a both in size and in structure and is rather different from the gnathopods of other species of the genus. The general character will be best learnt from the figure 4/; in larger specimens the propod may be somewhat larger and the swelling on the inner margin of the finger more pronounced. The peraeopods are fully shown in the figures (figs. 4m, n, 0, p) and do not call for further detailed description. The branchiae (fig. 4m) are all somewhat large in size, rectangular at the base and narrowing a little towards the extremity. | In the first uropod (fig. 4g) the peduncle is much longer than the rami which are subequal; in the second (fig. 47) the peduncle is only slightly longer than the rami; in both there are numerous spinules on the upper surface of the peduncle and the rami. The third uropods appear to vary in length in different specimens being sometimes as much as one-third of the totallength of the body. In figure 4s they are shown in side view as attached to the animal, when viewed from above they appear somewhat broader. The telson (fig. 42) is cleft to the base, each lobe narrowing posteriorly, the inner margin being nearly straight except towards the extremity, the outer strongly convex. A long spinule arises near the extremity of each lobe and on its inner side there are three minute setae; from the upper surface of each lobe towards the extremity arise two delicate sensory plumed setae. In the large inner lobe of the first maxilla and in the gnathopods this species differs considerably from the species of Niphargus hitherto described. I have, however, specimens sent to me by. Professor C. F. Baker from springs in the Philippine en which closely approach the Chilka Lake specimens in these Fic. 4.—Niphargus chilkensis sp. nov. e. Tower lip. 1. Second gnathopod. 7. First maxilla. m. First peraeopod. g. Second maxilla. n. Third peraeopod. h. Maxilliped. o. Fourth peraeopod. k. First gnathopod. p. Fifth peraeopod, s. Third uropod and telson NO NR Fauna of the Chilka Lake : Amphipoda. 535 characters. The gnathopods of Niphargus chilkensis present a somewhat striking resemblance to those of Phreatogammarus propinquus Chilton from New Zealand. That species, however, differs very considerably in the third uropods which, though elongated, have two branches each consisting of a single joint. ‘The third uropods in Niphargus and allied species are subject to special development in the different species and it is possible that the resemblance in the gnathopods is of more import- ance from the point of view of relationship than the differences of the uropods. [One of the commonest bottom species in the main area. NA. Melita inaequistylis (Dana.) Melita inaequistylis Stebbing, 1906, p. 429. Melita inaequistylis Chilton, 1909, p. 630 and 1911, p. 564. Melita inaequistylhis Barnard, 1016, p. 191. Melita zeylanica Stebbing, 1904, p. 22, pl. 5. Melita tenuicornis Walker, 1904, p. 273, pl. 5, fig. 33. Locality. Off Barkul, in fresh water. A few specimens, males and females. All small, largest about 4 mm. _ This species has already been recorded from Ceylon by Stebbing and Walker and it is interesting to find that it extends into fresh and brackish waters in India, just as it does in New Zealand. I have discussed the species to some extent else- where and compared it with M. palmata to which I referred specimens from Kermadec Islands (1911, p. 564); since then I have found that the same form occurs on the coast of New Zealand along with the typical M. imaequistylis. The latter Species is very wide spread and has been recorded from Cape Colony by Barnard, who gives a description of his specimens and says that it will ultimately have to be united with M. palmata (1916, p. 192). I have little doubt that the species des- cribed by Fritz Muller from Brazil (1869, pp. 27, 28) under the names M. messalina and M. insatiabilis will also prove to be identical with one or other of these forms. Maera othonides, Walker. (Text-fig. 5.) Maera othonides Walker, 1904, p. 271, pl. 5, fig. 29. Localities :— Off Samal Island, 8-15 ft., 22-ix-13. One. _ North side Chirriya Island. One. Chirriya Island. One. 8 miles W. by S. of Breakfast Island. Several. Barkuda Island. One. Chirriya Island. Three. Maludaikuda Island. One. Ennur backwater, near Madras town, 4-5 ft., Oct. 1913. One. These specimens undoubtedly belong to Walker’s species, though in one or two 536 Memoirs of the Indian Museum. VOLANT 93 / ; points they do not quite agree with his description which was drawn up from a - specimen only 8 mm. long and probably immature. In the third pleon segment there may be one or two teeth on the lower margin and usually there are three or four very distinct teeth on the posterior margin; the anterior portion of the lower margin bears a series of spinules. The accessory appendage of the first antenna may contain as many as five joints. Walker describes the hand of the second gnathopod in the female as being concave. In my specimens the palm is slightly convex (fig. 5c); in the male the palm is slightly concave towards the defining setules as shown in figure 55. Fic. 5.—Maera othonides Walker. a. First gnathopod of male. c. Second gnathopod of female. b. Second gnathopod of male. d. Third uropod. The third uropods (fig. 5d) especially in the older specimens bear a number of delicate woolly hairs and similar hairs are found on the dorsal surface of the posterior portion of the pleon. In the telson each lobe narrows greatly towards the acute extremity which bears one long setule and two or more smaller ones; in my specimens I cannot make out the second notch on the inside of each division which is described and figured by Walker; it is probably present only in immature specimens. | [Usually taken under stones just below water level, but also among algae of Shore Ne] 1921.] Fauna of the Chilka Lake : Amphipoda. 537 Quadrivisio bengalensis, Stebbing. (Text-fig. 6.) Quadrivisio bengalensis Stebbing, 1907, p. 159, pl. 7. Quadrivisio bengalensis Chevreux, 1013, p. I5, fig. I. Localities :— Off Samal Island, 8-15 ft., 22-ix-13. Many specimens, males and females, the largest about 12 mm. in length. Off Barkul, 3-4 ft. Two specimens. — Off Satpara, 4-5 ft., 17-ix-13. One specimen. Ghiakhala Headland and neighbouring island. Four specimens. Main Channel, W. of Satpara Island. Four specimens. Fic. 6.—Quadrivisio bengalensis. a. First gnathopod of female. b. Second gnathopod of female. c. Telson. I have compared these specimens both with Stebbing’s description and with a co-type of his species from Port Canning, and consider that they should be referred to the same species although in one or two points they show slight differences. The antennæ differ very much in the two sexes, the peduncle of the lower one in particular being greatly elongated in the adult males. The widely spaced pair of denticles on the dorsal surface of the pleon segments are quite distinctly to be made out in adult specimens from Chilka Lake; in Stebbing’s specimens they are said to be very small and difficult to observe. The main difference appears to be in the telson, Stebbing’s figure is probably from a slightly abnormal specimen for it shows the two halves distinctly unlike, the right being shorter and more rounded at the end than the left half and with a different armature of setae. In specimens I have examined both halves are alike (fig. 6c) and in general shape resemble the left half of Stebbing’s figure, narrowing towards the extremity, but they bear fewer setae, 538 Memoirs of the Indian Museum. [Vor. V3 namely three large ones at the apex and three minute ones on the inner margin of each lobe. Chevreux has recently (1913, p. 15) recorded the occurrence of Quadrivisio bengalensis in the island of Zanzibar; his specimens are found in a cave and have the eyes imperfect. He does not state whether there are any differences between his specimens and the description given by Stebbing. Dr. Annandale has sent me specimens from four localities in the Talé Sap, Siam, which appear to be quite the same as the Chilka Lake specimens. The peculiar character of the eye by which it is divided on each side into two separate pigmented portions is best marked in adult specimens. In one small specimen, 2 mm. in length, the eye on each side is single, being somewhat irregularly rounded and situated in the usual position slightly below the base of the upper antenna; in another specimen, slightly larger, the main portion of the eye corres- ponds with that in the first specimen, but it is continued upwards towards the dorsal surface of the head as a narrow pigmented projection. In older specimens this portion becomes enlarged, at first remaining connected with the older part of the original eye but later on becoming quite distinct. The appearance of the eyes in adults as seen from above is well shown by Stebbing (1907, pl. VII oc). [Common under stones just below water-level all round the lake. Also seen swimming in pairs at the edge. N.A.] Orchestia platensis Kroyer. (Text-fig. 7). Orchestia platensis Kroyer, 1845, Naturh. Tidskr. ser. 2, v.l, p. 304, pl. 2, f. 2. Orchestia platensis Stebbing, 1906, p. 540. Orchestia platensis Stebbing, 1900, p. 527, pl. 21a. Orchestia platensis Chevreux, 1908, p. 494, fig. 14. Orchestia agilis Kunkel, 1918, p. 118, fig. 31. Orchesha picheringii Dana, 1852, p. 882, pl. 59, fig. 9. Orchestia pickeringit Stebbing, 1900, p. 528. Orchestia pickeringii Stebbing, 1906, p. 538. Localities :— Barkul. Several specimens, males and females. Chirriya Island. Several. After much consideration I am referring these specimens to Orchestia platensis Kroyer, a species which according to Stebbing has a very wide distribution. Kroyer’s original specimens were from the Rio de la Plata, north-west of Monte Video, and Stebbing records the species from the Atlantic Coast of North America (where it has long been known as Orchestia agilis S. I. Smith), the Bermudas, the Mediterranean, Lake Tiberias and also from the Hawaiian Islands where it was obtained at two localities at heights of 3,000 and 2,000 ft. respectively. I had first identified the Chilka Lake specimens with O. pickeringu Dana which had been recorded from Hawaiian Islands, California and from New South Wales. « 1921.] Fauna of the Chilka Lake : Amphipoda. 539 J In the short antenne (figs. 7a, b) thickened in the males, and in the gnathopods (figs. 7c, d) and other appendages they seem to agree very closely with Dana’s species as described by Stebbing in 1906, and they certainly correspond with other specimens in my collection from Tonga which I -had also referred to this species. Among the Chilka Lake specimens there were only Y a few males and probably not more than one quite fully developed; the second gnathopod of the largest and presumably the oldest one is represented in figure 7d and shows two low convex spinulose processes as described by Dana, though these are both about the same breadth. Stebbing in his figure of a male of O. Pickeringii from the Hawaiian Islands shows the processes on the palm more separated and more pronounced, with the _ finger thickened about the middle, but as he points out, in younger specimens the inner margin of the finger is smoothly concave. Later on I had occasion to examine and mount some specimens sent to me from Cold Spring Harbor, CSA as OF agıhs es reducta, sp. nov. Perinereis marjorüi, sp. nov. Dendronereis aestuarina, sp. nov. .. Dendronereides, gen. nov... Dendronereides heteropoda, sp. nov. NEPHTHYDIDAE. Nephthys polybranchia, sp. nov. En oligobranchia, sp. nov. EUNICIDAE. Diopatra variabilis, sp. nov. Marphysa gravelyi, sp. nov. Lumbriconereis polydesma, sp. nov. 5 simplex, sp. nov. Page 565 565 568 515 578 582 584 589 593 595 598 602 603 607 610 6II 617 622 625 CONTENTS. GLYCERIDAE. Glycera alba cochinensis, var. nov. Glycinde oligodon, sp. nov. ARICHDAE. Scoloplos marsupialis, sp. nov. SPIONIDAE. Polydora hornelli, Willey .. Polydora (Carazzia) kempt, sp. nov. AMMOCHARIDAE. Myriochele picta, sp. nov. CAPITELLIDAE. Heteromastus similis, sp. nov. Barantolla, gen. nov. Barantolla sculpta, sp. nov. Mastobranchus indicus, sp. nov. MALDANIDAE. Euclymene annandalet, sp. nov. STERNASPIDAE. Sternaspis costata, Marenzeller SABELLIDAE. Potamilla leptochaeta, sp. nov. Laonome indica, sp. nov... oA Fabricia (Manayunkia) spongicola, Sp. nov. SERPULIDAE. Ficopomatus; gen. nov. Ficopomatus macrodon, sp. nov. .. Page 627 ‘ 629 632 634 636 638 640 642 643 645 648 649 651 652 653 655 655. POLYCHAETA. By ROWLAND SOUTHERN. INTRODUCTION. It was considered advisable, whilst describing the Polychaeta from the Chilka Lake, to examine also those species of the group in the Indian Museum which, at various localities in India, had been found in water which was either fresh or periodi- cally mixed with fresh water, or in other words was not distinctively marine. The collection contains very few species which spend their whole lives in fresh water. The majority live either in brackish water of low salinity, or belong to the “ Eury- haline” group. The latter term was first used by Moebius to designate those species which can live in water the salinity of which varies between wide limits. In Europe very few species of Polychaeta can tolerate marked changes in the salinity of the sea- water and fewer still can reproduce under such conditions, but in India, judging from . the list of species dealt with in the present paper which is obviously far from ex- haustive, they are relatively far more numerous. This adaptation may be correlated with the sharp division of the climate into wet and dry seasons, whereby the littoral region is periodically flooded with water of low salinity, especially in bays and estuaries. From the whole of India only two species of Polychaeta have been recorded from fresh or brackish water. These are Matla bengalensis Stephenson (1908, p. 39 and 1910, p. 82) and Spio bengalensis Willey (1908, p. 389) both from brackish pools at Port Canning, Lower Bengal. Matla bengalensis is based on juvenile speci- mens of a Capitellid, and neither it nor Spio bengalensis were represented in the present collection. Records of littoral marine Polychaeta from Indian shores are practically absent. A number of species have been recorded from Ceylon by Schmarda, Grube, Michaelsen, and Willey. It is therefore not surprising that almost all the species in the present paper are new to science, especially when one remem- bers the peculiar nature of the habitat in which they have been found. Many of the species of Polychaeta known from the Indian Ocean and South Pacific have been very imperfectly described and inadequately figured, judged by modern standards. Any specimens now referred to such species would be involved in a cloud of uncertainty, and it seems preferable, where there is any doubt, to ignore them until they have been redescribed from trustworthy material. List OF SPECIES. Ancıstrosyllis constricta, sp. nov. Chilka Lake. Lycastis indica, sp.nov. Gangetic Delta ; Cochin Backwater. 566 -Memoirs of the Indian Museum. [VoL. V, Tylonereis fauveli, sp. nov. Chilka Lake. Nereis (Nereis) chilkaénsis, sp. nov. Chilka Lake. Nereis (N.) glandicincta, sp. nov. Gangetic Delta. Nereis (N.) reducta, sp. nov. Chilka Lake. Perinereis marjorii, sp. nov. Chilka Lake. Dendronereis aestuarina, sp. nov. Gangetic Delta. Dendronereides heteropoda, gen. et sp. nov. Gangetic Delta. Nephthys polybranchia, sp. nov. Chilka Lake. Nephthys oligobranchia, sp. nov. Cochin Backwater ; Chilka Lake. Diopatra variabilis, sp. nov. Chilka Lake. Marphysa gravelyi, sp. nov. Chilka Lake. Lumbriconereis polydesma, sp. nov. Chilka Lake. _Lumbriconereis simplex, sp. nov. Chilka Lake. Glycera alba, Rathke, var. cochinensis, var. nov. Cochin Backwater. Glycinde oligodon, sp. nov. Chilka Lake. Scoloplos marsupialis, sp. nov. Chilka Lake. Polydora hornelli, Willey. Chilka Lake. Polydora (Carazzia) kempi, sp. nov. Gangetic Delta. Myriochele picta, sp. nov. Chilka Lake. Heteromastus similis, sp. nov. Chilka Lake. Barantolla sculpta, gen. et sp. nov. Gangetic Delta. Mastobranchus indicus, sp. nov. Gangetic Delta. Euclymene annandalei, sp. nov. Chilka Lake. Sternaspis costata, Marenzeller. Chilka Lake. Potamulla leptochaeta, sp. nov. Gangetic Delta. Laonome indica, sp. nov. Chilka Lake. Fabricia (Manayunkia) spongicola, sp. nov. Chilka Lake. Ficopomatus macrodon, gen. et sp. nov. Cochin Backwater. The species represented in the collection fall naturally into three geographical groups, viz. (I) those living in the Gangetic Delta and in the neighbourhood of Cal- cutta, (2) those collected in the Cochin Backwater, near Ernakulam, on the south- west coast of the Madras Presidency, and (3) those collected in the Chilka Lake. Le THE GANGETIC DELTA. Specimens were collected at the following localities :— Barantolla. In brackish pools and salt lakes. Chingrighatta. Ina canal of slightly brackish water. Durgapur. Ina canal of slightly brackish water. Beliaghatta Canal. Water slightly brackish. Garia. Pools of slightly brackish water. Dhappa. Salt lake and adjacent ditches of brackish water. All these places are within 10 mileMof Calcutta. The salinity of the water is very variable, but never high, probably never exceeding sp. gr. 1-015 at 15°C. 1921.] Fauna of the Chilka Lake : Polychaeta. 567 The Sunderbans. In a pool of brackish water inside the protection em- bankment of a clearing high up the Munda River. Vide p. 602. Port Canning. The variation in the salinity of the ponds at Port Canning is very great, ranging from 9°82 to 22°88 parts per thousand of saline residue. The following species were collected in the Gangetic Delta :—- Lycastis indica, sp. nov. Beliaghatta Canal; Garia. Nereis glandicincta, sp nov. Barantolla; Garia ; Dhappa. Dendronereis aestuarina, sp. nov. Sunderbans. Dendronereides heteropoda, gen. et sp. nov. Barantolla. Polydora (Carazzia) kempi, sp. nov. Barantolla sculpta, gen. et sp. nov. Mastobranchus indicus, sp. nov. Potamilla leptochaeta, sp. nov. Chingrighatta Canal. Barantolla. Barantolla. Port Canning. In addition, Matla bengalensis Stephenson, and Spio bengalensis Willey, have been recorded from Port Canning. 2. THE CocHIN BACKWATER. Collections were made in the Cochin Backwater near Ernakulam, on the south- west coast of the Madras Presidency. These backwaters are part of a system con- nected both with the sea and with large freshwater lakes. The salinity is probably very variable, according to the season of the year, but precise information is not available. The following species occur here :— Lycastis indica, sp. nov. Nephthys oligobranchia, sp. nov. 3. THE CHILKA LAKE. Glycera alba, Rathke, var. cochinensis, var. nov. Ficopomatus macrodon, gen. et sp. nov. The following species were obtained here :— Ancistrosyllis constricta, sp. nov. Tylonereis fawveli, sp. nov. Nereis chilkaensts, sp. nov. Nereis reducta, sp. nov. Perinereis marjori, sp. nov. Nephthys polybranchia, sp. nov. Nephthys oligobranchia, sp. nov. Diopatra variabilis, sp. nov. Marphysa gravelyi, sp. nov. Lumbriconereis polydesma, sp. noV. Lumbriconereis simplex, sp. nov. Glycinde oligodon, sp. nov. Scoloplos miarsupialis, sp. nov. Polydora hornelh, Willey. Myriochele picta, sp. nov. Heteromastus similis, sp. nov. Euclymene annandalet, sp. nov. Sternaspis costata, Marenzeller. Laonome indica, Sp. nov. Fabricia spongicola, sp. nov. No species is common to the three distriets. Lycastis indica was found in the Gangetic Delta and the Cochin Backwater, whilst Nephthys oligobranchia was found 568 M emoirs of the Indian Museum. [Vor. V, in the Chilka Lake and the Cochin Backwater. So few species were found in the latter locality that it may be left out of further consideration. Asregards the Chilka Lake and the Gangetic Delta, in the almost complete absence of any knowledge of the species comprising the Polychaete fauna of the Indian littoral, it is quite impos- sible to reach any reliable conclusions as to whether the species here described are peculiar to, or specially adapted to survive in the euryhaline conditions in which they occur. Nevertheless, certain opinions which I have formed may be stated for what they are worth. The collection of species from the Chilka Lake has a typically marine facies. It probably represents an impoverished remnant of the Polychaete fauna which inhabited the open bay before the present lake was almost completely cut off from the sea by the spit of sand which forms its eastern boundary (vide Annandale and Kemp, 1915, p. 5). How far the species living in the Chilka Lake have been differentiated from their relations still probably living in the adjacent open sea, in order to fit themselves for the altered environment, cannot at present be stated, but the changes have probably not been great. It seems likely that only those species survived which already had the capacity to withstand great variations of salinity. This point receives further consideration below (p. 571). The group of species found in the Gangetic Delta conveys the impression, in a rather vague manner, of having undergone greater modification than the Chilka Lake group. The eight species found here are all new to science, and two of them belong to genera previously undescribed. All the species possess branchiae, or analogous organs for performing the function of respiration. Six of the species belong to genera which have seldom or never been found before. Speculations of this nature, however, will have greater claims to validity when the whole fauna of the two areas is taken into consideration. In other groups of animals, the species common to the Chilka Lake and the Gangetic Delta are more numerous (Annandale and Kemp, 1915, p. 15). DISTRIBUTION OF POLYCHAETA IN THE CHILKA LAKE. For a full account of the conditions, topographical and hydrographical, prevail- ing in the Chilka Lake, reference must be made to the account given by Annandale and Kemp (1915, p. 1). For our purposes, a very brief description will suffice. The Chilka Lake is on the east coast of India, and is connected by a very narrow opening with the Bay of Bengal. It lies mostly in Orissa, but the southern extremity extends into the Madras Presidency. It is about 40 miles long, the greatest breadth is 124 miles, and it occupies an area of about 350 square miles. It is divisible into two well- defined areas, where the biological and physical conditions are very different. There is the main area, a shallow lagoon 4-8 feet deep in the dry season, 10-15 feet in the wet season. The northern part of the lake is very shallow, and very few Polychaeta were collected there. The bottom of the lake is predominantly muddy, though on the shores of some of the islands, and at the south-western extremity there is an admix- ture of sand. Near the middle of the eastern shore there is a shallow bar leading into a narrow channel about 12 miles long and a mile wide, which connects the lagoon with the Bay of Bengal, the outer opening being only several hundred yards wide. The Æ 1021. |- Fauna of the Chilka Lake : Polychaeta. 569 channel is separated from the sea by a narrow spit of sand. The inner part of the channel has a bottom of mud and sand, whilst the outer part is clean sand, and has an extensive oyster bed at Manikpatna. The dominating feature, at any rate in the channel, is the great seasonal change in the salinity of the water. All the main rivers enter the lake at the northern end. In the rainy season an enormous amount of fresh water enters this part of the lake, and gradually expels the salt water, so that from the middle of August to the middle of October, the northern half of the lake and the whole of the channel are occupied by water which is quite fresh. In the dry season the volume of water in the lake is reduced by evaporation, and salt- water from the Bay of Bengal enters the channel, and occupies it entirely as far as the inner entrance to the main area. Here the transition to water only slightly brackish occurs in a small space, and the main area is filled with water of low salinity. The southern half of the lake receives no large rivers, and throughout the year contains water only slightly brackish, varying between the limits sp. gr. I-oor-I'015. The variation at any particular spot in this part of the lake is not more than ‘oro, and is usually less, or one-third to one-fifth of that in the channel, where the sp. gr. ranges from 1:000-1'0270, The level of the water in the lake is 5 or 6 feet higher in the freshwater season (August to December) than in the salt-water season (December to August). The following table shows the distribution of the various species in the lake, the date of capture (salt or freshwater season), the specific gravity of the water where they were found, and the extreme observed range of the specific gravity they would have to endure if they lived there throughout the year. The true range is probably greater, as the limits represent observations made only in February and March (salt-water season) and in September (freshwater season) in 1914 (Annandale and Kemp, 1915, p. 6). The majority of captures at the south-west end of the lake were made during the salt-water season, but this is largely due to the lower level of the water at that period rendering it easier to reach the habitat of the various mud- and sand-dwelling species. In the outer channel the bulk of the records were made in the fresh-water period, because it was only at that time that the launch used for dredging and trawling, could cross the shallow bar which separates the main area from the outer channel, at Mugger-Mukh. These circumstances limit to some extent the conclusions which might be drawn from the distribution of species as known at present. 570 Memoirs of the Indian Museum. [Vor. V, TABLE I. | SPECIFIC GRAVITY OF WATER] | | Species. à Distribution. =. At time of Annual range at capture. place of capture. | Ancistrosyllis constricta .. I°000—I"0I0 1:000—1"027 | South and south-west | S. F.? shores ; outer channel. | Tylonereis fauveli ve 1'000—I'0265 1'000— 1'027 | Outer channel. Sis Re Nereis chilkaénsis aK I°‘002—TI'OIT I’00I—I'0I5 South of lake. SR, Nereis reducta ys I°000 I'000—1'027 | Outer channel. 18, Perinereis marjorii ee I°000 1'000—I'027 Outer channel. | F. Nephthys polbranchia CNT OOE- TORE I'00I—I'0I5 | South of lake. SE: Nephthys oligobranchia .- | TOOTS-T'OITS I'00I—I'0I5 South of lake. Ss IR, Diopatra variabilis TE 1‘000—I"OIO I'000—I'0I5 | South and north-west of | lake. Ste . Marphysa gravelyi iR: 1'008—I'0II5 I'001—1'015 | South of lake. Sı Lumbriconereis polydesma .. | I’OII 1'006—I'0I5 South of lake. Se Lumbriconereis simplex... I’OII 1'006—1'015 | South of lake. S: Glycinde oligodon : ha I'000—I'OII I 000—I'026 South of lake : once atin- | nerendofouter channel.| S.F. Scoloplos marsupialis Ye 10264 1:000—I'027 | Outer channel. S. Polydora hornelli ae T:000 1'000—1'027 | Outer channel. | 18% Myriochele picta .. | ‚2'005 1 010 I'00I—I'0I5 | South of lake. |S. F. Heteromastus similis ce 1’000—I‘0261 1'000—1'0265 Inner end of channel, | So JE, | | near bar. | Euclymene annandalei a | 1'002—-I'OII I'00I—I'0I5 | South of lake. SR, Sternaspis costata | I‘000 ; I'000—1'027_ | Outer channel. F. Fabricia spongicola | I'O06—I'OIT I'OOI—I'O15 South of lake. SE 8. Laonome indica | I’009 I'004—1'009 | South of lake: | Sı | L At 15°C. 2 S — Salt-water season (February —July); F = Freshwater season (September—November). When the distribution of the species in Table I is studied, two main groups are easily distinguished. In the first are those species confined to the south end of the lake, which have to withstand variations in the salinity ranging from 1I'O001-I1'015, if spread over the whole area. If very localised in range, the variation may be much less The species in this group are :— GROUP I. Nereis chilkaénsis. Lumbriconereis simplex. Nephthys polybranchia. Myriochele picta. Nephthys oligobranchia. Euclymene annandalei. Marphysa gravelyi. Fabricia spongicola. Lumbriconereis polydesma. Laonome indica. In the second group are those species which have only been found in the Outer Channel. These are :— GROUP II. Tylonereis fauveli. Scoloplos marsupialis. Nevers reducta. Polydora hornelli. Perinereis marjorüi. Sternaspis costata. 1921.] Fauna of the Chilka Lake : Polychaeta. 571 There remain four species not included in the above groups. Glycinde oligodon was found at eleven stations in the southern end of the lake, and at one station in the inner part of the Channel, south of Mahosa, so that it may be considered as essentially belonging to Group I. Heteromastus similis was taken at three stations between Nalbano and Barhampur, that is to say, near the inner entrance to the lake. It is thus intermediate in distribution between the two groups, and probably likes the bottom of muddy sand which occurs there. Diopatra variabilis was found at eight station in the south end of the lake, twice near the middle, and once at the north- west corner, off Kalapara Ghat, so that it probably belongs to Group I. Ancistrosyllis constricta was taken four times at the south end of the lake, and twice in the Outer Channel during the freshwater season, though it probably lives there the whole year round. Of all the Polychaetes in the lake, this is the species which can endure the greatest range of conditions. A more complete knowledge of the distribution of the species in the lake might cause some changes in this grouping, though probably not to a great extent. The species in Group II have to endure much greater variations in the salinity of the water than those in Group I. It would, however, be rash to claim that the salinity is the determining factor. The species found in the Outer Channel are sand-loving forms, whilst those in the lake are chiefly mud-loving, though some of them live in sponges, etc. The faunistic boundaries are probably two . in number, one lying close to the inner entrance of the lake near Barhampur, and limiting the inward range of the sand-loving euryhaline species of the Outer Channel ; the other lying in the area roughly bounded by Patsahanipur—Nalbano— Parikudh — Kalidai Id.—Barkul Point, and cutting off, in the south end of the lake, the mud- loving species living in water of low, but not greatly varying salinity. There is not yet sufficient evidence to decide whether the determining factor is the salinity of the water or the nature of the bottom. Annandale (1915, p. 7) has stated, with reference to the Chilka Lake fauna, that “the great majority of what we may call the permanent residents are, for the reasons that I have already given you, very ordinary in appearance. There is an absence both of brilliant colours and of colouration specially in a high degree adapted for concealment. There is a lack of bizarre form, and the majority of the animals are not modified structurally to a visible extent.” Our complete ignorance of the littoral Polychaeta of the Bay of Bengal prevents any analogous statements being made here. To anyone accustomed to examining the Polychaeta of the European coasts, a collec- tion which, out of 17 genera, contains representatives of Ancistrosyllis, Tylonereis, Glycinde, Myriochele, Heteromastus, Sternaspis, Laonome, and’ Manayunkia, cannot be considered as ordinary in appearance. This impression is also greatly strengthened if the species from the Gangetic Delta are added, though it might be dissipated by a knowledge of the species in the adjacent sea. It may be stated generally that the species in Group I are adapted for life in waters of low, but not very variable salinity, and the majority of them prefer a muddy bot- tom. The species in Group II can live in water ranging in salinity from quite fresh to normal sea-water of the Bay of Bengal, and they seem to prefer a sandy bottom. 572 Memoirs of the Indian Museum. Nor; Sexual MATURITY. In an environment like that of the Chilka Lake it may well be that some species are able to live throughout the later larval and adult periods of their life, without being able to reproduce there (Annandale and Kemp, 1916, p. 337). Free-swimming adults and larvae, or eggs, may be periodically carried into the lake by favourable winds, currents and tides, to places which are suitable for their further growth. This is much more probable in the case of those species which live in the Outer Channel than in those which occupy the main area of the lake. Such a sterile migra- tion is much more probable in the case of mobile animals like the fishes and Crustacea than in such a relatively sedentary group as the Polychaeta. Of the 20 species found in the lake, 8 contained genital products more or less mature, the species Nereis chilhaënsis being represented by the Heteronereis in all stages of maturity. A number of quite small and immature individuals of several species were also collec- ted. Probably the majority of the Polychaeta, if not all, are permanent residents. The subject is worthy of further investigation. | HABITAT OF THE CHILKA LAKE POLYCHAETA. Five types of habitat may be provisionally distinguished. Information on this subject is very defective, and the scheme may be greatly modified by further obser- vations. Species found mainly or wholly on a muddy bottom. Ancistrosyllis constricta. © Lumbriconereis simplex. Nephihys polybrancma. Glycinde oligodon. Nephthys oligobranchia. Moyriochele picta. Diopatra variabılıs. Euclymene annandale:. Marphysa gravelyı. Laonome indica. Species found mainly or wholly on a sandy bottom. Tylonereis fauveli. Lumbriconereis polydesma (?). Nereis chilkaénsts (?). Scoloplos marsupiahs. — Nereis reducta. Sternaspis costata. In the case of the two species queried, the distribution in the lake would lead one to expect a muddy habitat, though the labels attached indicate sand. Species found on the oyster bed at Manikpatna in the Outer Channel. Ancıstrosyllis constricta (on one occasion). Perinereis marjorii. Polydora hornelli. The last named only is a true shell-borer ; the two former species probably only take shelter amongst the oysters. Species living in sponges (Spongilla and Laxosuberites). Nereis chilkaénsis. Fabricia spongicola. Species living amongst weeds. | Nereis chilkaénsis. Fabricia spongicola. 1921.] Fauna of the Chilka Lake : Polychaeta. 573 DIAGRAMMATIC REPRESENTATION OF THE PARAPODIA. An effort has been made, by means of diagrammatic text-figures, to give a more precise idea of the structure of the parapodium, and the distribution of the various types of setae, than is possible in the usual method of illustration. In the latter, the foot—the horizontal axis of which is at right angle to the main axis of the trunk —is shown either from the anterior or posterior point of view. The text-figures now endeavour to represent the foot diagrammatically as seen from the side, that is to say, when looking at the distal end of the various lobes. The observer is supposed to be looking at the parapodium on the right side of the specimen, so that the ante- rior face of the parapodium is on the right, facing the head, and the posterior face on the left, facing the tail. Further details are given on p. 576 and text-fig. ı. MATERIAL NOT FIT FOR IDENTIFICATION. For various reasons a small number of specimens in the collection were not iden- tified. They were either too immature, or too imperfect, or (in one case) the preserv- ing fluid had evaporated and the specimens had shrivelled up. They furnished sufficient evidence, however, to prove the presence in fresh or brackish water, of a number of additional species not described in the present report. Of special interest was a tube of dried specimens from a canal of slightly brackish water at Durgapur, near Calcutta, containing species of the genera Perinereis, Lepidonotus and Phyllodoce, the two latter belonging to families not otherwise represented, whilst the Perinereis sp. was distinct from P. marjori of the Chilka Lake. Family HESIONIDAE. Ancistrosyllis constricta, sp. nov. (Plate XIX, figs. IA-IG.) This species, of which many individuals were taken, was found only in the Chilka Lake. The general appearance of the body is very characteristic. The greatest width is at the anterior end, the peristomial segment being the widest. The body narrows tapidly as far as the 4th setigerous segment, this anterior region enclosing the large thick-walled pharynx, which causes it to be circular in section. The body then becomes flat, and on account of the greater size and length of the parapodia, it seems to increase considerably in width, but in reality the trunk, after a slight increase, remains the same width for some distance, and then gradually tapers towards the tail. The body has thus a distinct neck, or waist, and hence the specific name. Towards the posterior end the ventral surface becomes rather more convex, and the feet appear to be on the dorso-lateral surface. The peristomium and three anterior setigerous segments (fig. IA) are longer than the succeeding ones, and are only separated from each other externally by very indistinct grooves. In the largest individuals the posterior region is absent. A fragment, 23 mm. 374 Memoirs of the Indian Museum. [Vor. V, long, had 120 setigerous segments. Another fragment, full of sperm, was 16 mm. long, with 63 setigerous segments, and was rather stouter than the other immature speci- mens. A complete individual 19 mm. long had 105 setigerous segments. The head (fig. IA) bears in front two flattened palps, fused in the median line where they are very thin and transparent. The front margin is concave. Each palp has a smail terminal joint. The postero-lateral angles, where the head is widest, are rounded and indented. The posterior border is concealed beneath the very thin pro- jecting margin of the peristomium. . There are three tentacles, the median, which is nearly twice as long as the laterals, being slightly in front of them. The lateral ten- tacles project a little beyond the palps. As in A. robusta, Ehlers, the entrance to the pharynx is guarded by large pear- shaped papillae, 14 in number. In À. robusta there are 16 of them. When the pro- boscis is fully extruded, the mouth is a vertical slit, but when retracted the mouth is large and gaping, bounded behind by the much folded lower lip, in front and at the sides by the folded and flattened palps. No jaws are present. In the dorsal median line the projecting collar of the peristomium is fused with the head, and only sepa- rated from it by a faint groove. At the sides of the head, and in front of the tenta- cular cirri, the peristomium has two forwardly projecting rounded lobes. There are two pairs of tentacular cirri placed near the front of the peristomium. The dorsal pair are about one quarter longer than the ventral pair, and all have filiform tips and swollen bases. Neither setigerous lobes, setae, nor spines are present. The succeeding segment bears the first pair of parapodia. They are distinguished from the remaining feet by the very long and tapering dorsal cirri (fig. IB), which doubtless function as tentacular cirri, and are as long as the median tentacle. The ventral cirrus is clavate, less than one-third as long as the dorsal cirrus. The seti- gerous lobe is pointed at the tip, where the spine terminates. Two rows of simple capillary setae lie obliquely above and below the spine. The anterior setae in each row are short, strongly curved, and coarsely serrated (fig. rx). The posterior setae are three times as long, gently curved and very slender, the upper edge being smooth and very thin. There is a gradual transition from one extreme to the other, and the two groups, above and below the spine, are similar. The dorsal division of the foot is represented by a single short and very slender spine, which pierces the base of the dorsal cirrus. | The second foot (fig. IC) is very minute. The clavate dorsal cirrus’ projects be- yond the conical setigerous lobe, and there is no ventral cirrus, a rather remarkable feature. The dorsal spine is very small. In each group of setae there are six or seven with coarse teeth, two with short teeth, and three or four smooth setae. The next few feet increase till the normal size is attained. The setae increase in numbers, and there are very few intermediate between the coarsely serrate and the smooth forms, except as regards length. One of these intermediate forms is shown in fig. IG. Some of the long posterior setae show minute serrations, and this is probably the original condition, the smooth ones having lost their teeth by abrasion, as the edge of the seta on which they occur is very thin. In the 4th foot the dorsal spine is } 1921.] Fauna of the Chilka Lake : Polychaeta. 575 small and straight, just reaching the base of the dorsal cirrus. In the 5th foot it is much longer, passing into the cirrus, but not piercing the epidermis. In the roth foot all the setae show at least traces of serrations, and the anterior setae have longer teeth than those in the example figured from the Ist foot. In the 20th foot the dor- sal spine is curved at the tip, but still completely embedded in the foot. ‘There are two minute papillae on the base of the dorsal cirrus. The setae in the upper group are longer than those in the lower group, and the number of coarsely serrate setae is relatively less. Between the 30th and 4oth feet a stout sickle-shaped seta appears in the dorsal division (fig. ID). It is in addition to the dorsal spine, which still persists, thin and curved, and does not pierce the epidermis. Ehlers (1908, p. 60) says that in A. robusta, the stout sickle-shaped seta replaces the slender spine. The sickle-shaped seta tapers gradually towards the tip, which pierces the skin, and there is no nodulus such as MacIntosh shows in 4. groenlandica (1877, pl. 65, fig. 3). Behind it are a few small conical papillae on the base of the dorsal cirrus. The succeeding feet do not show much change. In the posterior 27 feet of the specimen examined the spine and hook of the dorsal division are joined by a slender capillary seta (fig. IE) which pierces the skin. The spine tapers very rapidly near the curved tip. The papillae on the base of the dorsal cirrus are larger and more numerous than in the anterior and middle feet, and the setigerous lobe is shorter and more rounded. ‘The anus is terminal, and the anal segment bears two slender ventro- lateral cirri, equal in length to the last six segments. A specimen with the body-cavity full of ripe sperm was found in September, in the Outer Channel, during the freshwater season. This species greatly resembles in many respects the A. robusta, Ehlers (1908, p.59), found at Great Fish Bay, S.W. Africa, but differs in the following characters :— (x) the shape of the body; (2) the palps are fused almost throughout their length, and are flat, not cylindrical ; (3) the 1st foot has a well-developed setigerous lobe, and the dorsal cirrus is 3% times as long as the ventral cirrus. In A. robusta the seti- gerous lobe is small, and the dorsal cirrus twice as long as the ventral; (4) the papil- lae on the base of the dorsal cirrus, and the dorsal capillary seta of the posterior feet were not observed by Ehlers in A. robusta; (5) the proboscis has 14 papillae, that of A. robusta 16 ; (6) the absence of the ventral cirrus of the second pair of feet is not mentioned by Ehlers nor MacIntosh. The 4. groenlandica of MacIntosh (1877, p.502), dredged in 410 fathoms in Davis Strait, is very different from A. constricta or A. robusta, but in several characters of the feet it shows the greater affinities with the former species. The setigerous lobe and dorsal and ventral cirri bear many small cylindrical papillae. More important, however, is the presence in the dorsal division, together with the stout sickle-shaped seta, of a slender bristle, generally enclosed in the tissues. According to Ehlers the sickle-shaped seta is a modified. aciculum, which, in A. robusta, has replaced the simple curved spine of the anterior segments, and he does not regard the presence of both forms in the same foot in A. groenlandica as affecting this theory. It is clear, 576 Memoirs of the Indian Museum. [Vor. V’ however, in À. constricta, that the slender curved aciculum is present in all the feet, and that the sickle-shaped seta is an addition in the middle and posterior segments, in the latter of which it is accompanied by a slender capillary seta. It is extremely probable, therefore, that the sickle-shaped seta is a modified dorsal seta, and not an aciculum. The À. albim of Langerhans (1881, p.107) is quite distinct from any of these species, if, indeed, it is a true Ancistrosyllis at all. Habitat.—This species was found on six occasions in the Chilka Lake. Twice it was taken in the outer channel, three times near the southern extremity of the lake and once off Balugaon. On one occasion it was found in crevices of oyster shells at Manikpatna, but usually it occurred on a muddy bottom in a few feet of water, and once it was taken on a sandy bottom south of Mahosa. | The salinity of the water varied from 1‘000-1‘015, but the occurrence in the outer channel at Manikpatna in oyster shells suggests that it can survive the salt- water season when the gravity is as high as 1026. Family NEREIDAE. The wide variation usually to be observed in the structure of the parapodia from various parts of the same specimen of Neveis necessitates a very close examination: The difficulty of adequately representing the structure of the foot by the usual method of illustrating the anterior and posterior aspects led to an attempt to depict it dia- grammatically from the terminal aspect. In text-fig. 1, a typical foot from the right side of a specimen of Nereis pelagica, L. is shown from the anterior (A) and the posterior (B) points of view and C is a diagram representing the same foot as it would appear if examined from the terminal aspect. The arrangement of the various lobes and different types of setae is obscure in A and B, but is clearly shown in C. The position of the setae and their differing structure are indicated by the use of various symbols. | The posterior side of the foot is on the left of the vertical axis and the anterior on the right. Thus the diagram represents the foot as it would appear if looked at in situ on the right side of the worm, with the head to the right and the tail to the left. : Complete uniformity has not yet been attained in the nomenclature of the various lobes which constitute the foot of a Nereid and the following explanation of the terms here used may conduce to this end. There is considerable variation in detail in the parapodia of the different species of the Nereidae, but the fundamental plan remains fairly constant. The simplest and most typical form is seen in the sub- genus Nereis, as represented by Nereis pelagica, L,. (text-figs. 1 A-C). Here the foot consists of dorsal cirrus a, a ventral cirrus b, and three lobes designated in this paper the dorsal ligule c, the median ligule d, and the ventral ligule e. Between the dorsal and median ligules emerge the setae here known as “ the setae of the dorsal division.” Where they emerge from the foot, these setae are surrounded by a mem- moan. | à Fauna of the Chilka Lake : Polychaeta. 377 brane, the “ dorsal fillet ”’ f. Beneath the setae, and embedded in the upper part of the median ligule, is the dorsal spine k. The dorsal fillet is attached by its edges to the median ligule. Between the median and ventral ligules lie the “setae of the ventral division,” enclosed within the ventral fillet g. The latter projects outwards in two lobes, a larger anterior and a smaller posterior. The “ventral spine’’ k terminates in the anterior lobe of the fillet. Three types of setae are most commonly found in the Nereidae, and their position in the foot is fairly constant, the arrangement in N. pelagica being typical. Homo- gomph setae with elongate finely-serrated terminal pieces (plate XXT, fig. 6K) occur in a C =F; = de to IT IN —_ ae | (a A b TEXT-FIG. 1.—Foot of Nereis pelagica, L. A. toth right foot, front view. B. Loth right foot, posterior view. C. toth right foot, diagram. The lettering is identical in the three figures. a = dorsal cirrus; b = ventral cirrus; c = dorsal ligule; d = median ligule; e = ventral ligule ; = dorsal fillet ; g. = ventral fillet ; À = dorsal spine ; À = ventral spine. : = Homogomph setae with elongate tips; o = Heterogomph setae with elongate tips; x = Heterogomph setae with falcate tips. the dorsal division, and in the upper posterior part of the ventral division. Hetero- gomph setae with elongate terminal pieces (plate XIX, fig. 2H) occur in the lower posterior part of the ventral division. Heterogomph setae with short falcate termi- nal pieces (plate XXI, fig. 7K) occur in two groups on the anterior side of the ventral division, one group above, and the other below, the spine. Variations from this typical structure take the form either of reduction of the number of lobes on the one hand, or on the other of increased complexity due to branching. The condition of the various genera may be passed in review. In Lycastis the footis very simple, consisting only of dorsal and ventral cirri and the setigerous divisions. This simplicity is probably not primitive, but derived, as 578 Memoirs of the Indian Museum. Domenie the dorsal setigerous division is to be found in various stages of reduction. In Micronereis the foot has not yet been adequately described, but appears to consist of dorsal and ventral cirri, ligules, and setigerous divisions, the median ligule being absent. In Dendronereis the foot appears to be normal, except that in some anterior segments the dorsal and ventral fillets project in a large and variable number of lobes. In Dendronereides the ventral ligule is absent. In some anterior segments the dorsal ligule is richly branched. In the posterior parapodia only dorsal and ventral cirri and median ligule are present. In T'ylonereis the median ligule is apparently absent, but it may be represented in the anterior segments by the _ retractile lobe, and in the posterior segments by the lower lobe of the dorsal setige- rous division. Ceratocephala, Malmgren, with which I am not acquainted, seems to be very abnormal. Judging from Malmgren’s figures, it has a dorsal cirrus, a bifid ventral cirrus, a ventral ligule, dorsal and ventral setigerous divisions, and a lobe near the dorsal setae, which may be either the dorsal or median ligule. Tylorrhynchus, Grube, has dorsal and median ligules but no ventral ligule. The remaining genera of the family, comprising Leptonereis, Leonnates, Nereis, Perinereis, Platynereis, and Cheilonereis, have parapodia of typical structure. Lycastis indica, sp. nov. | (Plate XIX, figs. 2A-J, and text-figs. 2 a-d.) This species was taken in three different localities. It was not found in the Chilka Lake. The five specimens from the Beliaghatta Canal are small, and the largest is incomplete. It is 26 mm. long, and consists of the head and 120 setigerous segments. Probably the perfect individual had at least 200 segments. The smallest specimen, 8 mm. long, has 40 setigerous segments. The specimen from the Cochin Backwater differs slightly from the others, and is described below. The specimen selected for the examination of the feet, from the Beliaghatta Canal, was complete, 16 mm. long, having 115 setigerous segments. The dorsum is pale reddish brown in the anterior region, and the pigment increases in density and redness towards the tail. The ventral surface is colourless and deeply grooved. The head is thickly speckled with dark reddish brown pigment, except in the median groove and the posterior median part, where it is colourless. The bases of the tentacular cirri are pigmented, the remainder being colourless. The head (fig. 2A) is much broader than long, with two short tentacles in front. In the anterior median region there is a deep groove which widens to form a pit on the middle of the head. The eyes are on the posterior margin, almost in a straight line, and are pro- vided with lenses. The palps are short, thick and stumpy, and the tentacular cirri are also rather short. The jaws are short, with 9 teeth. The everted proboscis has neither papillae nor paragnaths. The buccal segment is rather narrower than the succeeding ones. Rose Fauna of the Chilka Lake : Polychaeta. 570 The feet increase in size up to the 7th pair. The two anterior feet differ from those of most species of the Nereidae in having a dorsal spine present. The Ist foot (fig. 2B and text-fig. 2a) has a short stumpy dorsal cirrus, jointed at the base, and is shorter than the finger-shaped ventral cirrus. The setigerous lobe is broad and truncated, with a papilliform retractile tip. The dorsal division is only represented by a black spine. The ventral setae are in two groups, one above and one below the spine. The upper group, emerging from the front of the foot, consists of two setae, one a short falcate heterogomph (fig. 2F), the other a hemigomph with a long finely serrated terminal piece (fig. 26). The lower group consists of heterogomphs, the upper posterior one having a long finely serrated terminal piece, the others short falcate tips. In the 2nd foot the dorsal cirrus is larger than the ventral. The upper group of Tex'r-FIG. 2. —Parapodial diagrams of Lycastis indica, sp. nov. a. Ist foot of specimen from the Beliaghatta Canal. DOTE ÿs ms 5 5 5 Cs TOWN à ir » » Cochin Backwater. de) 70h". re de , Beliaghatta Canal. - = Hemigomph setae with long finely serrated tips; x = Heterogomph setae with falcate tips; — = Hetero- gomph setae with finely serrated tips; o = Heterogomph setae with coarsely serrated tips. setae consists of two hemigomphs with long finely serrated tips, and two falcate heterogomphs. The 3rd foot is similar, the dorsal cirrus being a little larger. In the roth foot (fig. 2c and text-fig. 2b) the upper group consists of three anterior falcate heterogomphs and three posterior hemigomphs with finely serrated tips. In the lower group the upper posterior seta is a heterogomph with a long tip which has very coarse serrations on its lower edge (fig. 2H). This has replaced the finely serrated heterogomph of the anterior segments. Beneath it are 4-7 falcate heterogomphs. The dorsal cirrus is jointed, with a cylindrical base, and the appa- rent increase in size of the dorsal cirrus in this and the succeeding segments is due to the enlargement of the base, which carries the dorsal cirrus at its tip. In the succeeding feet there is little alteration, except as regards the gradual enlargement of the base of the dorsal cirrus (figs. 2D, 2E and text-fig. 2d). This enlargement is greatly accentuated about the 6oth foot. In the posterior segments 580 Memoirs of the Indian Museum. Lies WY 5) it attains a relatively enormous size, and its length exceeds the width of the body. It is traversed by three main blood-vessels, which give off numerous capillaries lying beneath the cuticle. The dorsal spine emerges in a small papilla. In the posterior feet the spines are only black near the tip. The falcate heterogomphs in the upper group have rather longer and more spinous tips than those in the lower group. There may be as many as three heterogomphs with coarsely serrated tips in the posterior part of the lower group. The setae are fewer in the posterior segments. The anal segment is button-shaped, with very bright and conspicuous streaks of reddish brown pigment. The anal cirri are as long as the posterior dorsal tenta- cular cirri. The single specimen from the Cochin Backwater is 30 mm. long, consisting of 150 setigerous segments. The head, bases of the tentacular cirri, and anterior dorsum ‚are bright rusty red in colour. The middle region is reddish brown, and the posterior region is bright red, especially in the lateral intersegmental areas. The anal seg- ment is bright reddish brown. The tentacular cirri, feet, and ventral surface are colourless, except near the tail, where the ventral surface is pale reddish brown. The colour is due to granular pigment in the skin. The most striking difference in the feet of this specimen is the presence, with the dorsal spine of the Ioth to about the 6oth foot, of a single slender hemigomph with a long finely serrated tip (fig. 27). The arrangement of the setae in the roth foot is shown in text-fig. 2c. The dorsal seta is guarded by a small fillet. The specimen from Garia, Lower Bengal, is in bad condition, but resembles this form in having a single hemigomph in the dorsal division of the foot. The head is similar in all the specimens, except that the tentacles and tentacular cirri are shorter in the Cochin specimen. All were immature. This species is closely related to the Lycastis hawatiensis, first described by Johnson (1903, p. 210) whose specimens were found in a spring near Honolulu, Hawaii. It was afterwards described by Horst (1909, p. I), from specimens found in a fresh- water pond in the Botanical Garden at Buitenzorg, Java. The Lycastis indica differs from that species, as described by Johnson and Horst, in the following characters :— (I) the groove in the head ends in a pit, and does not run to the posterior border ; (2) the eyes are almost in a line and are provided.with lenses. According to John- son, in the type the external pair are slightly in front of the internal pair, and have no lenses ; (3) in the presence of much reddish brown pigment. Johnson says that the only pigment present in his specimens was in the yellow tips of the posterior dorsal tentacular cirri. Horst says: “The body of the preserved worms is quite colourless, except the underside of the head and of two or three anterior segments, having a yellowish hue....... During life the worms appear to be flesh-coloured, due to the blood of the peripheral vascular system shining through the skin’’ (4) The atrangement and structure of the setae differ considerably. Neither Johnson nor Horst mentions the occurrence of the heterogomph setae with the long coarsely serrated tips. According to Johnson, the setae are in two groups, the group above the spine consisting of two or three moderately heterogomph setae (similar to the 1921.] Fauna of the Chilka Lake : Polychaeta. 581 hemigomphs shown in fig. 2G), the lower group consisting of a number of falcate heterogomphs, the lower ones having shorter tips than the upper ones. Horst’s description is substantially similar. Thus the present species differs from L. hawait- ensis in having falcate heterogomphs above the spine, and heterogomphs with long coarsely serrated tips below the spine. Horst found in the dorsal lobe of the 41st foot of one specimen “‘a faintly developed setose bristle.”’ In the present state of our knowledge of the genus Lycastis, this form has as much claim to specific rank as any of the previously described species. Sufficient attention has not been given to the precise arrangement and structure of the setae. On the other hand, too much significance has been attached to slight differences in the arrangement of the eyes. The following six species of Lycastis, characterised by the great enlargement of the posterior dorsal cirri, have been described :— L. brevicornis, Audouin and Milne-Edwards. Noirmoutiers (France). Marine. L. abiuma, Grube. Desterro, Brazil. Marine. L. senegalensis, de Saint-Joseph. Marsassoun (100 kilometres from the sea) Senegal, West Africa. Brackish water. . ouanaryensis, Gravier. French Guiana. Marine and freshwater. . Geayi, Gravier. French Guiana. Freshwater. . hawaiiensis, Johnson. Hawaii and Java. Freshwater. . Geayi is sufficiently characterised by the complete absence of falcate setae. In L. brevicornis, L. abiuma, L. senegalensis, and L. hawaitensis, the heterogomph setae with coarsely serrated tips have not been described. There remains L. owanar- yensis, which was found both in the sea and in freshwater in French Guiana, on the north-east coast of South America. The similarities between this species and L. indica are very striking. The setae are very similar in arrangement, and the ventral division of the middle and posterior feet contains heterogomph setae with slender coarsely serrated tips, exactly as in L. indica. The anterior part of the bedy is rose coloured, the posterior green. The very slight distinctions between the two forms are as follows :—In Gravier’s species (1) the longitudinal median groove on the head runs back to the buccal segment ; (2) the eyes are devoid of lenses; (3) the tentacular cirri are shorter than in L. indica ; (4) the jaws have six teeth, as against nine in L. indica; (5) the dorsal cirri of the posterior segments are longer than in L. indica ; (6) the dorsal division of the foot contains 2 or 3 setae, and the falcate setae have tips of rather different shape. L. indica is intermediate between L. hawaiiensis and L. owanaryensis as regards the setae. It agrees with the former in the great reduction of the dorsal division of the foot, and with the latter in the presence of the coarsely serrated heterogomph setae. Habitat. This species was obtained from the three following localities :— In rotting cocoa-nut shell floating in the Beliaghatta Canal, near Calcutta, 26. viii. ‘oo. Water slightly brackish. In the same tube was a specimen of the freshwater Oligochaete, Branchiura sowerbyi, Beddard. FREE 582 Memoirs of the Indian Museum. [Vor. V, Garia, near Calcutta, burrowing in mud. The Cochin Backwater, near Ernakulam, in the Madras Presidency, September 1014. In the two former localities the salinity of the water is very variable, but never high, probably never exceeding 1r'o15 at 15°C. The Cochin Backwater has also water of very variable salinity, but precise information regarding it is not available. It is part of a system connected both with the sea and with large freshwater lakes. Tylonereis fauveli, sp. nov. (Plate XIX, figs. 3A-], and text-figs. 3a-c.) Two complete specimens and fragments of four individuals of this species are available for examination. The preserved specimens have retained very little pigment. The dorsum is pale yellowish brown, and the head is faintly marked as shown in fig. 3A. The anal segment (fig. 3B) is, however, of a conspicuously bright red colour. The body is slender, and rather more delicate and transparent than is usual in the Nereidae. Of the two complete specimens, both of which are immature, one is 95 mm. long, composed of 200 segments, the other is 55 mm. long, with 150 fully formed segments and a number of developing segments at the posterior end. Several of the fragments apparently belong to larger specimens than either of these. The head (fig. 3A) is rounded at the angles, divided into two lobes in front by a shallow indentation. The width of the posterior part slightly exceeds the length. The eyes are black, and the posterior pair are larger and closer together than the anterior pair. Allare provided with lenses. The tentacles and palps are normal. The tentacular cirri are rather short, the anterior dorsal being two-thirds of the length of the posterior dorsal cirri, and a little longer than the posterior ventral cirri. The peristomial segment is only a little longer than the succeeding segments, and like them is grooved in the lateral regions. The proboscis was not everted in any of the specimens, and its condition could only be observed by dissection. No horny paragnaths are present, but two large papillae were seen on the dorsal surface of the basal ring, that is to say, one in each Group VI. An interrupted row of smaller papillae was seen on the ventral side of the maxillary ring, in Group IV. None could be seen in Groups V, VII or VIII, but they may be present, as it is difficult to observe the soft papillae in dissected speci- mens. The jaws are small and slender, pale except at the tip, with 12 teeth. The feet increase in size up to the 6th pair. In the rst and 2nd feet the dorsal setigerous division is absent. The 1st foot (fig. 3c and text-fig. 3a) is relatively small. The dorsal cirrus is a little longer than the dorsal ligule. The ventral setigerous lobe is somewhat fusiform, and its base is enveloped on its anterior and ventral sides by a prominent fillet, which in side view resembles an additional lobe, and gives a bifid appearance to the setigerous lobe. In T. bogoyawlensky Fauvel represents the setigerous lobe as terminating in three lobes (1911, Pl. XIX, fig. 2). The setae are in three groups, 1921.] Fauna of the Chilka Lake : Polychaeta. 583 two anterior, above and below the spine, and a posterior ventral group. They are all homogomphs, with slender finely serrated terminal pieces, those in the lower anterior group having shorter tips than the rest. The 2nd foot resembles the Ist, except that it is a little larger, the lobes are larger, and the setae more numerous. In the 7th foot (fig. 3D, and text-fig. 3b) the dorsal cirrus is relatively smaller. The dorsal ligule is larger and flatter, pointed at the tip, and filled with glands. The dorsal setigerous lobe is long and cylindrical, rounded at the tip, and its base is surrounded by a sheath or fillet, within which it can be retracted by powerful muscles. It is penetrated almost to the end by . the dorsal spine, the tip of which is surrounded by a sheath of columnar cells. There are three to five setae in the dorsal division. The ventral setigerous division remains as in the ıst foot, except that the setae are more numerous, and apparently only in two groups, and the ventral ligule is relatively thinner. The 15th foot differs from the 7th only in having the dorsal and ventral cirri somewhat smaller, the dorsal ligule rather larger and flatter, and the lobes of the ventral setigerous division thinner and more elongate. The dorsal division has four setae, the ventral division ten in the upper group, sixteen in the lower group. At the 22nd foot a slight indenta- tion appears in the tip of the elon- ee oor ey bre TEXT-FIG. 3.—Parapodial diagrams of Tylonereis gradually deepens till in the 26th or fauveli, sp. nov. 27th foot the bifid condition shown a. Ist foot. 6. 7th foot. c. 30th foot. in fig. 3E is attained. In smaller Dena a tips. specimens the change may be com- plete in the 24th foot. A similar change occurs in T. bogoyawlenskyi, but accord- ing to Fauvel only in the posterior segments, the exact position not being stated. In the 30th foot (fig. 3e, and text-fig. 3c) the dorsal and ventral cirri and the ventral ligule are greatly reduced in size. The dorsal ligule is large and long, pointed at the tip, and full of glands. The dorsal setigerous lobe is deeply bilobed at the tip, the spine lying between the two lobes. Its tip, enclosed in the gland, is shown in fig. 3F. The dorsal setae, eight in number, lie above the two lobes, whereas Fauvel says that in T. bogoyawlenskyi they emerge between the two lobes. The dorsal setae now differ markedly from those in the ventral division. The 584 Memoirs of the Indian Museum. [MOT terminal pieces are short, and they taper rapidly to a fine point (fig. 3y). The terminal pieces of the ventral setae are 2-4 times as long, and taper very gradually to the tip. The space between the ventral ligule and the ventral cirrus has greatly increased. The 7oth foot (fig. 3G) is smaller, and the setigerous lobes and fillets are reduced and flattened. The setae in the dorsal division have stouter shafts than those in: the ventral division. In the posterior feet (fig. 3H) the two setigerous divisions are more widely separated. In the dorsal setigerous lobe the upper lobe is reduced to a small papilla, but the lower lobe has relatively increased. The ventral fillet is still very obvious, giving a bifid appearance to the ventral division. The dorsal setae have longer tips, which taper more gradually than those in the middle region of the body. The anal segment (fig. 3B) is bright red in colour. It is button-shaped, and the anus is round and terminal. The ventral anal cirri are short and slender. The dorsal setae are all pure homogomphs, but some of the ventral setae show a tendency towards the hemigomph condition, the enlarged cup-shaped tip of the shaft having a process on one side, of variable size. This species is very closely related to the 7. bogoyawlenskyi described by Fauvel (1911, p. 376) from a single specimen collected im the Persian Gulf, for which he created the genus T'yonereis. This genus is characterised by the presence of soft papillae only on the proboscis, by the structure of the dorsal setigerous division of the feet, and by having setae of one type only. T. bogoyawlenskyi differs from the present species chiefly by the trilobed condition of the ventral setigerous lobe in the anterior feet. Other minor characters distinguishing the present species are the shape of the head and tentacular cirri, the red anal segment, the shape of the setae, and the proportions of the various parts of the feet. The change in the condition of the dorsal setigerous lobe apparently occurs in a more anterior position in T. fauveli. According to Fauvel there is no difference between the dorsal and ventral setae. A noteworthy character of the two species of Tylonereis is the apparent absence of the median ligule of the foot. Possibly, how- ever, it may be represented in the anterior segments by the retractile lobe, and in the posterior segments by the lower lobe of the dorsal setigerous division. Habitat.—This species was taken on two occasions, in the Chilka Lake, both times in the outer channel. It was taken in September, when the water was quite fresh, and in March, when the specific gravity was 1:0265, showing that it can withstand the full rigours of the violent seasonal alternation of conditions in the outer channel. Nereis (Nereis) chilkaensis, sp. nov. (Plate XXII, figs. 8A-R, and text-figs. 4a-c.) This species was found in great abundance in the southern half of the Chilka Lake. It is evidently quite at home there, for both immature and heteronereid stages were lound. 1921.] Fauna of the Chilka Lake : Polychaeta. 585 The largest specimen is 92 mm. long, and has 8: setigerous segments. Others are 65, 56 and 44 mm. in length, with 84, 66 and 81 segments. The dorsum is deeply coloured with purplish brown pigment, dark in front, and growing paler behind. The ventral surface is colourless. The head (figs. 8a, 8B) is deeply coloured, with a median pale streak which broadens out between the eyes. The palps are more faintly coloured. The basal part of the proboscis is tinged with the purple pigment. The bases only of the tentacular cirri are deeply coloured. The first few segments have the dorsum pigmented all over, except for a few pale oblique lateral grooves. Further behind, the pigment forms a definite pattern on each seg- ment (fig. 8p). The central part becomes paler, leaving a dark band on each side, separated from the central part by a narrow pale band. The intersegmental areas are pale. In the anterior segments the base of each foot is pigmented, but in the middle and posterior regions the feet are colourless. The pigment is not uniformly diffused, but is more or less granular, especially in the dark lateral bands. The head (fig. 8a) is considerably narrower in front than it is behind, and the length is about equal to the greatest width. It projects a little in front between the tentacles, which are 4rd to 4th as long as the head. In small specimens the palps are separated from the head by grooves, but in large specimens these grooves become indistinct. The palps are large and stout. The eyes, in immature individuals, are small and dark, equal in size, the anterior pair being a little further apart than the posterior pair. The tentacular cirri have dark ringed ceratophores. The pos- terior dorsal pair are much longer than the others, reaching back usually to the mid- dle of the 6th setigerous segment, in some cases to the 8th and even the 12th seg- ment. The anterior dorsal tentacular cirri are one-half to two-thirds as long. The ventral cirri are approximately equal in length, being half the length of the rst dorsal cirri. These sizes and proportions are subject to great variations. The paragnaths have the normal form and arrangement for the subgenus. Those oti the basal ring are longer than those on the maxillary ring. They are arranged as follows :— Group I, 6-10, usually 7 or 8. Group V, absent. 820; , VI, 3-5 usually 4, larger than the „ HI, 26-34, usually 28. | others, in a curved row. » IV, 35-41, varying greatly. VII and VIII, forming a continuous band in two rows. 3) The paragnaths of the anterior row in Groups VII and VIII are large; those forming the posterior row are more numerous and of different sizes. The jaws are of the usual shape, with eight teeth. The peristomium (fig. 88) is nearly twice as broad as the succeeding segments. In fig. 8A it is crushed by the extruded proboscis. - The feet were examined in a specimen 56 mm. long, having 66 setigerous seg- ments. The feet increase in size up to the 6th or 7th. 586 Memoirs of the Indian Museum. [Vor. V, The ist foot (fig. 8E, and text-fig. 4a) contains only the ventral setigerous divi- sion. The dorsal and ventral cirri are long and slender. The median (?) and ven- tral ligules are stout and conical. The ventral setigerous region consists of a fillet forming a dome over the setae. The fillet is produced outwards into two conical lobes, one in front of, and one behind the setae, and its lower edges are attached to the ventral ligule. The spine pierces the anterior lobe of the fillet. In some speci- mens there are two ventral spines. The setae are in three groups, a dorsal group consisting of a single row with falcate heterogomphs in front, long-tipped (spinose) homogomphs behind, and two ventral groups, an anterior group of falcate heterogomphs, and a posterior group of spinose heterogomphs. The 2nd foot is similar. The dorsal division appears in the 3rd foot. The roth foot (fig. SF, and text- fig. 4b) exhibits the normal condi- tion. ‘The dorsal cirrus is long, great- ly surpassing the dorsal ligule, but the ventral cirrus is small. The dor- sal, median, and ventral ligules are large, thick, and bluntly rounded, the dorsal being the largest. In the dorsal division the fillet has a pro- minent pointed anterior lobe, and a smaller rounded posterior one, and its lower edges are attached to the median lobe, before and behind the : © a RER spine. The setae are all homogomphs TEXT-FIG. Te cn of Nereis chilkaén- (fig. 8x) ec long slender tel a. Ist foot. b. roth foot. c. 50th foot. pieces. The ventral division is asin * = Homogomph setae with long spinous tips; o = Hetero- the first foot. gomph setae with long spinous tips; x = Heterogomph setae ; with long falcate tips : e = Heterogomph setae with thick shafts In the succeeding feet the lobes and tips. become gradually more elongate and pointed. The dorsal setigerous lobe decreases in size, and fuses more oF less com- pletely with the median ligule. The dorsal part of the foot, including the median ligule, increases in size relatively to the ventral part and projects beyond ie: In the soth foot (fig. 86 and text-fig. 4c) the dorsal cirrus 1s long and slender, the. ventral cirrus small and thin. The dorsal and median ligules are large and pointed. The dorsal setigerous lobe is represented by a small swelling, pierced by the spine, on the upper side of the median ligule, and by a small and indistinct fillet guarding the dorsal setae. The ventral ligule 1s much smaller. The ventral fillet, with its two lobes, is as before, and forms an arch above and below the setae. 1921.| Fauna of the Chilka Lake : Polychaeta. 587 In the 63rd foot the dorsal cirrus is 24 times as long as the dorsal ligule. The dorsal division is relatively still larger and more prominent than the ventral division. The upper group of the ventral division consists only of a single thick falcate heterogomph, the spinose homogomphs having disappeared. There are four types of setae present in this species, in addition to the modified setae of the Heteronereid phase. In the dorsal division only homogomphs with long tapering spinose tips occur (fig. 8N). In the ventral division, the anterior setae, above and below the spine, are falcate heterogomphs (fig. 8p). They have moderately long terminal pieces, smooth at the tip, spinose below. The posterior setae consist of an upper group of spinose homogomphs, similar to those in the dorsal division, and a lower group of spinose heterogomphs. Between the 20th and 30th feet two or three setae appear in the upper anterior group of falcate heterogomphs of the ventral division, which have shafts rather thicker than those of the other. setae. In the subsequent feet these setae increase in thickness, but diminish in number, and in the last few feet, as already stated, the upper ventral group is represented only by a single falcate seta with a shaft three times as thick as that of the normal type of seta. The terminal piece is relatively shorter, with only a few strong spines on the edge (fig. 80). The shafts of the falcate setae of the lower group also increase in thickness in the posterior feet, but not to the same extent. There is considerable variation in this species. In some individuals the tentacu- lar cirri and dorsal cirri are much longer than usual (fig. 83). The terminal pieces of the very thick ventral falcate setae have sometimes only three or four spines, and rounded tips, as in fig. 80, but in other specimens they may have ten or twelve spines, and pointed tips. This character doubtless depends on the amount of wear they have undergone. The eyes vary considerably in size, as do the paragnaths, and the intensity of the epidermal pigment. Some or all of these characters may be associated with approaching change to the Heteronereid condition. Three specimens of this species were taken which had, in varying degrees, assum- ed the Heteronereid condition. The more advanced of the two female specimens was selected for examination. It is 40 mm. long, with 63 setigerous segments. The colour of the dorsum is much deeper than that of the immature individuals, and the eyes are much larger, but not connate. The paragnaths on the basal segment of the proboscis are larger than usual. The anterior 18 pairs of feet show no change. In the 19th foot the ventral cirrus shows indications of a small ventral lobe near the base. The 20th foot has a few of the characteristic swimming setae in both divisions of the foot, and the basal lobe of the ventral cirrus is well developed. The setigerous lobes and their fillets are enlarged. In the 25th foot (fig. 87) the dorsal cirrus remains unchanged. The ventral cir- rus has a large ventral and a small dorsal wing near the base. The three ligules are elongate and pointed. The two lobes of the ventral! setigerous division are enlarged and flattened, and the posterior one has two flat wings. The setae are mostly of the natatory type, but a few of the old setae remain. In the 30th foot the flattening of the various lobes is more obvious. In the 4oth foot (fig. 8k) the dorsal cirrus has a 588 Memoirs of the Indian Museum. "Vor AVE small flat wing above its base, and the wings of the ventral cirrus are larger. The posterior lobe of the dorsal fillet is greatly enlarged, displacing by its growth the median ligule. The ventral setigerous lobes are also greatly enlarged and flattened, and the ventral ligule is directed downwards. The 58th foot is much the same, but it is smaller, and has a larger proportion of the normal setae. The eggs escape from the body cavity through the tips of papillae on the ventral surface near the base of the feet. The swimming setae (fig. 8R) are modified homo- gomphs. They are very transparent. The end of the shaft is cup-shaped, with an anterior spinous projection. The blade, which is minutely serrate, increases in width to near the tip, and then narrows suddenly to a sharp point. The two female Heteronereids were found under stones on the shore. Another Heteronereid was taken on the surface of the lake, where it was gyrating in a spiral course. It is a fully developed Heteronereid, but the sexual products have been completely extruded. It is probably a male, as it differs in several important points from the specimen just described. It is small in size, being only 16 mm. long, but it consists of 72 setigerous segments. The eyes are very large, and the head is deeply pigmented, the white patch, which occurs in all other specimens, being absent. The anterior end of the body is dorsally very deeply coloured with metallic brown pigment, and is without the characteristic pattern. The tentacular cirri are unusually long. The anterior 19 pairs of feet are unmodified, except that the roth has a single swimming seta in the dorsal division, and that the dorsal setigerous division is in all the feet fused with the median ligule (fig. 81). In the 2oth foot the lobes are not altered, but the dorsal division contains only swimming setae, and a few of these are also in the ventral division. The only unchanged setae are the two thick falcate heterogomphs of the upper ventral group. In the 25th foot only one thick falcate seta remains. The setigerous lobes are enlarged and flattened, and the ventral cirrus has basal wings. In the 30th foot (fig. 8m) the median ligule, to which the dorsal setigerous lobe is fused, is enlarged and flattened. The ventral setigerous lobes are also flattened, the anterior one being large and foliate. The 18 posterior feet are not modified, and have only normal setae. The dorsal cirrus is very long, and the ven- tral cirrus has increased and projects beyond the ventral ligule. The spines are very thick and dark. | À specimen taken on the shore seems to represent a stage in the development of the last described form. It is a slender worm, 22 mm. long, with 62 setigerous seg- ments. The body cavity is full of sperm morulae in an early stage of development. The head of this specimen is shown in fig. 8B. It has very large eyes, and the tentacular cirri are unusually long. In both these points it resembles the male Hetero- nereis. The colour, however, is normal. The most exceptional character of this specimen is that in a few of the posterior feet the dorsal ligule decreases very much in size (fig. 8H), a character noted in no other individual of this species. The jaws of this specimen have 10 teeth. It is remarkable that all the partially or completely metamorphosed Hetero- 1921.] Fauna of the Chilka Lake : Polychaeta. 589 nereids are smaller than the average immature individual in the collection, though they have the normal number of segments. The two female Heteronereids were taken in January and February, the small immature male in February, and the spent male in November, so that the period of sexual maturity is evidently prolonged. The specific gravity of the water at these four stations varied only from I’006-1'009. Habitat.— This species was found at 15 stations, all south of a line drawn from Nalbano to Patsahanipur. Twice it was found in November, and on thirteen occa- sions in January to March. In this part of the lake, however, the season makes little difference in the specific gravity of the water, which ranged at the various sta- tions from 1002-1011. The habitat of this species varies widely. It was often found in sponges, both Spongilla and Laxosuberites. It also lives under stones on the shore, amongst algae on the rocks, and in sand above and below high-water mark. Nereis (Nereis) glandicincta, sp. nov. (Plate XXIII, figs. ga-L, and text-figs. 5a-e.) This species was taken at four localities near Calcutta, in brackish lakes or pools. Twenty-six specimens are available, most of them approaching a state of maturity. The preserved specimens are pale buff brown. Running transversely across the dorsal and ventral surface of each segment is a very conspicuous row of dark glands. The glands are continued into the feet, and are especially prominent in the dorsal ligule. They occur to a lesser degree in the median and ventral ligules and the base of the ventral cirrus, but are not found in the dorsal and ventral cirri nor in the setigerous lobes. There is usually a band of glands on the median dorsal area of the peristomium. There is considerable variation in the number of glands in each row. Sometimes they are absent from the median dorsal and ventral areas, and in other specimens they form complete and very thick bands round each segment. The specific name is derived from this character. The dorsal vessel shows very dis- tinctly. The ventral surface is slightly grooved, and the nerve-cord shows as a pale line. The type-specimen is 88 mm. long, and comprises 123 setigerous segments. It is a female, full of eggs, but shows no sign of assuming the Heteronereis condition. In all the specimens which have the pharynx extruded, the peristomium is large, oval, and greatly inflated. Even when the pharynx is retracted, the peristomium is large, bulging out at the sides and in front of the head. The head (fig. ga) is narrow in front, wide behind, where the width exceeds the length. In front are the two short tentacles. The posterior angles are rounded. The eyes vary considerably in size in the different specimens, probably owing to the approaching condition of sexual maturity. They are provided with lenses. The pigmentation of the head also varies. In some specimens there is a conspicuous band on each side of the anterior part, reaching from the base of the tentacles to the front eyes. In others there are only two oval patches of pigment between the posterior pair of eyes. Others have two short lateral and a narrow median band as 590 Memoirs of the Indian Museum. ONE in fig. 9A. The palps are very large and flat. The posterior dorsal tentacular cirri are twice as long as the anterior dorsal pair, and reach back to the 3rd setigerous segment. The armature of the pharynx (figs. 9A, 9B) is of considerable interest. The maxillary ring has its full complement of teeth, but the basal ring is greatly reduced. The arrangement is as follows :— Group I, 10 scattered unequal teeth. Group V, absent. 10-13 large curved teeth. VI, one small tooth. III, transversely elongated band >, WIE ar single row Wore 7m me of 50 teeth, in 4 rows. | ON ARTE teeth. 10-12 large teeth. | +) In group III the two middle rows of teeth are larger than those in the outer rows. All the teeth on the maxillary ring are of the normal size and conical shape. On the basal ring, however, they are rudimentary. Dorsally there are two small pale paragnaths seated on rounded papillae (figs. 94, 9€). Ventrally there is a single row of 7 similar minute paragnaths. In two other specimens there were only 3 and 4 paragnaths respectively in the ventral row. These rudimentary paragnaths easily fall off, and consequently some or all of them are frequently missing. The condition of things characteristic of the sub-genus Ceratonereis is thus attained. The two dorsal papillae can always be distinguished, even when the paragnaths have fallen off. The jaws (fig. 9D) are slender, and have 15-16 teeth, an unusually large num- ber. Ina specimen from the Salt Lake at Dhappa, however, there are only Io teeth on each jaw. The feet gradually increase in size up to the 8th. The Ist foot (fig. ge, and text- fig. 5a) has pointed dorsal and ventral cirri, equal in size and shorter than the adjacent ligules. The latter are stout pointed lobes. The ventral division consists of three slender lobes, two in front more or less completely surrounded by the fillet and setae, the third behind the setae. The edges of the fillet are attached to the lower anterior lobe. The setae are of three types. Above the spine there is a group consist- ing in front of hemigomphs with rather short coarsely serrate tips, behind of homo- gomphs with long slender finely serrate tips (fig. 9J). Below the spine the anterior and ventral group consists of hemigomphs as in the anterior dorsal group. Behind the spine is a group of hemigomphs with long finely serrate tips. The anterior hemi- gomphs with coarsely serrate tips resemble those shown in fig. 9K, except that the tips are shorter. | The 2nd foot resembles the Ist, except that it is larger and more glandular. In the 3rd foot the dorsal division appears. It is deeply bifid, with a few slender homo- gomphs. The roth foot (fig. gF, and text-fig. 56) has small clavate dorsal and ventral cirri, much smaller than the adjacent ligules. The dorsal ligule is pointed and triangular in shape, richly supplied with glands. The dorsal setigerous division consists of a slender lobe lying in front of the dorsal setae The latter are homogomphs (fig. 9J) 1921. | Fauna of the Chilka Lake : Polychaeta. 591 with long finely serrate tips. They are guarded behind by a fillet which is attached to the dorsal and median ligules. The tip of the spine, which is curved and enclosed in a sheath of gland cells, lies between the median ligule and the dorsal setigerous lobe. The ventral division resembles that of the 1st foot, except that the setae are much more numerous, form- ing a continuous row. At the 18th foot a number of falcate hemigomphs appear in the TEXT-FIG. 5.—Parapodial diagrams of Nereis glandicincta, sp. nov. a. Ist foot. b. Ioth foot. c. 30th foot. d. 5oth foot. e. Iooth foot. : = Homogomph setae with long finely serrated tips ; — = Hemigomph setae with short coarsely serrated tips; x = Hemigomph setae with long slender finely serrated tips ; o= Hemigomph setae with falcate tips. ventral division. ‘The 20th foot is very like the roth in shape. The median ligule is relatively larger. In the 30th foot (text-fig. 5c) the dorsal and ventral cirri, and the dorsal and ventral setigerous lobes have decreased relatively in size, whilst the dorsal, median and ventral ligules have increased. In the ventral setigerous lobe, as compared with the 1oth foot, the setae are fewer and some of the hemigomphs with short coarsely serrate 592 Memoirs of the Indian Museum. Vor, tips have been replaced by falcate hemigomphs (fig. 9L) especially in the lowest group. These setae have very elongate spinous tips, and the point is boldly curved. In the 50th foot (fig. 9¢, and text-fig. 5d) there is no marked change in the dorsal division. Inthe ventral division the upper anterior group of setae are all falcate hemigomphs. A few hemigomphs with short coarsely serrate pointed tips appear in the posterior row behind the spine. In the 60th foot the two anterior lobes of the ventral division are smaller, and the setae are fewer in number. The falcate hemigomphs have disappeared and are not present in the succeeding feet. In the Soth foot the upper anterior lobe of the ventral division is reduced to a small papilla, scarcely visible. In the goth foot it has quite disappeared. In the 1ooth foot (fig. 9H, and text-fig. 5e) all the foot lobes are slender and sharply pointed. In the last few segments the dorsal cirrus increases in length tillit surpasses the dorsal lobe. The dorsal setigerous lobe becomes small and rudimentary. There appear to be no true heterogomph setae in this species, their place being taken by setae having the shafts ending in the intermediate condition which has been termed“ hemi- gomph.” There is no sharp distinction to be drawn between these various types, the faicate seta shown in fig. 9L being nearer the true heterogomph condition than the coarsely spinose seta shown in fig. OK. Ra The anal segment is conical, the anus forming a terminal slit. The anal cirri are short and slender, a little shorter than the anterior ventral tentacular cirri. There is a ting of reddish brown pigment (or glands) round the middle of the anal segment. In a number of the specimens, the body cavity was full of eggs, but no indica- tion of change to the Heteronereis condition was observed. This species is charac- terised by the armature of the proboscis, the structure of the feet, and the shape and arrangement of the setae, especially by the absence of falcate setae from the anterior and posterior segments. The girdle of glands on each segment is also very charac- teristic. Nereis kerguelensis, McIntosh, has some points of resemblance to this species but isotherwise not closely related toit. Of greater interest is a comparison with Nereis veducta (p. 593). In both species the paragnaths of the basal ring are greatly reduced in size and number, so as to be almost rudimentary. The dorsal setigerous division in N. glandicincta, though small, is much more developed than that of N. veducta. In both species the dorsal and ventral cirri are short, but N. glandicincta has a prominent dorsal ligule, whilst that of N. veducta is small in the anterior and poste- rior segments. In N. veducta the spinous heterogomph setae are absent in the anterior segments, and only in small numbers elsewhere, whilst in N. glandicincta it is the falcate setae which are absent from the anterior and posterior segments. These resemblances indicate no close affinity, but are probably due to convergence. They are mostly of a negative character, and may be due to modification fitting for life in brackish or freshwater. It is significant that Lycastis, the genus found most fre- quently in fresh and brackish water, shows great simplicity in the structure of the foot, little variety in the shape of the setae, and has no paragnaths on the proboscis. 1921. | Fauna of the Chilka Lake : Polychaeta. 593 Habitat.—Twenty-six specimens were found in the four following localities :— Salt lake, Barantolla near Calcutta. Netted in 5-6 feet. Brackish pools near the Salt lake, Barantolla, November 1913. From mud in a ditch containing brackish water, at the edge of a salt lake at Dhappa, near Calcutta, January Igı1. Burrowing in mud at Garia, Lower Bengal. These localities are all within 10 miles of Calcutta. The specific gravity of the water in which the specimens were found is very variable but never high, probably never exceeding I°015. Nereis (Nereis) reducta, sp. nov. (Plate XXI, figs. 7A-7K, and text-figs. 6a—d.) Only a single specimen of this species was obtained, about one mile inside the mouth of the Chilka Lake. The body is 50 mm. long, and is composed of 96 setiger- ous segments. It is relatively narrow, and the ventral surface is grooved. The segments are three times as wide as long in the anterior region, but the length increases and the width decreases till they are less than twice as wide as long. The body does not taper very much towards the posterior end. The head and dorsal surface of the anterior segments are pale brown in colour; the rest of the body is colourless. . The head (fig. 74) is narrow in front, broad behind, and the length slightly exceeds the width. The eyes are small and distinct, with lenses. The palps are large and pointed. The tentacular cirri are rather short, the posterior dorsal pair being a little longer than the anterior dorsal. The proboscis of the unique specimen is fortunately fully extruded (figs. 7A, 7B). The paragnaths are distributed as follows :— i M Group I, A single large paragnath | Group ‘ le peace is II, 6 paragnaths of varying size. fe N 7 Be Ob Bis: Es En 3 PACE CNT, ARE paragnaths in er IV, 8-10 ,, i. N. Bl, WEEE, longitudinal rows. The paragnaths of the distal or maxillary segment are of the normal dark conical type (fig. 7c). Those of the basal segment are small, flattened, and circular, of a pale amber colour (fig. 7p). The dorsal group consists of two paragnaths close together in the median line, and may both be in Group V, or one each in Group VI. They are very minute, and easily overlooked. The ventral Groups VII and VIII occupy a large area, and are composed of numerous longitudinal rows, each contain- ing 4-7 minute paragnaths. The jaws are provided with 7 or 8 teeth. The feet gradually increase in size up to the 6th. The rst foot (fig. 7E, and text-fig. 6a) as usual, is represented by the ventral division, the dorsal cirrus and a lobe between them. This latter lobe, as a considera- tion of figs. 7m, 7F, and 7G clearly shows, represents the median ligule. The dorsai and ventral cirri are rather short and finger-shaped, the median and lower ligules 594 Memoirs of the Indian Museum. Vor. V, are still a little shorter and thicker. The ventral setigerous division consists of a stout central papilla, pierced at the tip by a spine, and surrounded by a stout fillet, which has an entire margin behind, but projects in front, forming a small lobe. The upper group of setae consist of 3 spinous homogomphs behind, and 2 falcate hetero- gomphs in front, the lower group of 7 falcate heterogomphs. The 2nd foot closely resembles the Ist, except that the central papilla of the setigerous lobe is rather larger. In the 3rd foot (fig. 7F, and text-fig. 6b) the dorsal division appears. It consists of a small lobe, the rudiment of the dorsal ligule, and a small fillet in front of the spine, and a single spinous homogomph seta. In the ventral division the chief change is that the posterior margin of the fillet is produced into a papilla similar to but smaller than the one on the ante- rior margin. The setae are un- changed. In the roth foot (fig. 76, and text-fig. 6c) the normal condition is shown. The dorsal and ven- tral cirri are relatively much smaller, whilst the upper and median ligules are considerably enlarged, especially the former. The dorsal setigerous division con- tains two spinous homogomphs, the ventral division showing little change in the setae. The dorsal ligule gradually increases in size from the 3rd to about the 23rd foot, and attains a size equal to that of the median ligule. TEXT-FIG. 6.—Parapodial diagrams of Nereis reducta, sp. zus sea un ey lately Uns nov. same. The papillae on the fillet a. ist foot. b. 3rd foot. c. 10th foot. d. 6oth foot. of the ventral division are not so : = Homogomph setae with long tips; x = Heterogomph setae with i 1 i falcate tip; o = Heterogomph setae with long tips. prominent, and the mei 18 more vertical. The dorsal ligule has now begun to decrease again slightly. In the upper posterior part of the lower ventral group of setae there is a single spinous heterogomph. In the 4oth foot the dorsal ligule is stili smaller, and in the 50th foot is just shorter but thicker than the dorsal cirrus. In the 6oth foot (fig. 7H, and text-fig. 6d) the dorsal cirrus, dorsal and ventral ligules are about equal in size, the ventral cirrus being smaller. The largest lobe is the median ligule. The dorsal division contains three spinous homogomphs. The ventral setigerous division is flattened from side to side, and the margins of the fillet are rounded, the papillae noted in the anterior segments having disappeared. The 1921.] Fauna of the Chilka Lake : Polychaeta. 595 setae have the same arrangement, only a single spinous heterogomph being present. The remaining feet show no change, except that they decrease in size and the setae become fewer in number. The anal segment is button-shaped, with two very short anal cirri, only as long as the segment is wide. The anal segment and cirri are possibly regenerated, as the posterior feet are quite large. The dorsal homogomphs are slenderer than any of the ventral setae (figs. 7) and 7K). The falcate setae have tips of medium size with spinous edges. The most interesting characters of this species are those which exhibit a tendency towards degradation from the normal type. These are (1) the insignificance of the paragnaths on the basal segment of the proboscis. In this respect the species is approaching the condition found in the sub-genus Ceratonereis, Kinberg, in which the basal segment is devoid of paragnaths; (2) the insignificance of the dorsal setigerous division, indicated by the small number of dorsal setae, and the reduction in size of the dorsal ligule in the anterior and posterior segments ; (3) the rarity of the spinous heterogomph setae, which only occur singly in the middle and posterior segments, and are absent from the anterior segments. According to the classification of de Saint-Joseph (1898, p. 285), based on the structure and arrangement of the paragnaths, this species belongs to the sub-genus Nereis of the genus Nereis. A point of considerable interest, clearly established in this species at any rate, is that the lobe beneath the dorsal cirrus in the Ist and 2nd feet represents the median ligule. A consideration of figs. 7E, 7F, and 7G leaves no room for doubt. The dorsal ligule appears in the 3rd foot as a small lobe. Habitat.—Only a single specimen of this species was obtained, on the shore about one mile from the mouth of the Chilka Lake. It was taken in September, during the freshwater season, and the water was quite fresh. Perinereis marjorii, sp. nov. (Plate XXIII, figs. roa-c, and text-figs. 7 and 8a-c.) The largest of the 10 specimens of this species, from the Chilka Lake, is 61 mm. long, and has 80 setigerous segments. Another is 53 mm. long, with 76 setigerous segments. The body is long and slender, 2°5- 2°75 mm. wide, including the parapodia at the widest part. The segments are very distinct, 2-3 times as wide as long, and the feet are comparatively short. The colour is pale pur- plish brown, and is strongest on the head and anterior dorsal region (fig. 10A). On the head Text-ric. 7.—Perinereis marjori, sp. nov. there are three longitudinal bands of pigment, Ds! view of ee ua two of them marginal, and a shorter median one. Behind the latter, between the eyes, there isa V-shaped band. In the anterior region of the body there are three short transverse bars on the dorsum of each seg- ment, the median being a little behind the other two, leaving a colourless patch in 596 Memoirs of the Indian Museum. (Vor we front. From each of these bars a band passes to the posterior margin of the seg- ment. In the middle and posterior regions the two lateral longitudinal bands dis- ‘appear, but the median band remains very distinct (text-fig. 7). The ventral surface is colourless. The anal segment is short, cylindrical, and colourless, and the anus is terminal. There are two short tapering anal cirri equal in length to the last three segments. The head (fig. 104) is narrow in front, with an abruptly widening posterior part which bears the two pairs of dark eyes, each provided with a lens. The width of the posterior part exceeds the length of ® the head. The tentacles and palps Cy are normal. The tentacular cirri : are rather short, and of the usual relative proportions. All the parag- naths are large, well-formed, and dark in colour (figs. IOA, IOB). fen Group I consists of 5-12 teeth; Group II of curved bands each con- “> taining 18-22 teeth; Group III of 19 teeth ; Group IV of 16-22 teeth; Group V of 3 large teeth curved backwards and arranged in a tri- angle; Group VI on each side of a single large semi-circular smooth tooth with the thin edge in front ; Groups VII and VIII of a double ‚row of fairly large teeth with oval bases, and approximately equal in size. B The jaws are of the usual shape, TEXT-FIG. 8—-Parapodial diagrams of Perinereis marjorii, and have 10-12 small teeth. Sp. nov. The feet increase in size up to the a. Ist foot. 0. 10th foot. c. 50th foot. th - = Homogomph setae with spinous tips; o = Heterogomph setae | with spinous tips; x = Heterogomph setae with falcate tips. The rst foot (fig. LOC and text- fig. 8a) has slender dorsal and ven- tral cirri. The upper lobe (median ligule ?) is clavate, shorter than the dorsal cir- rus. Only the ventral setigerous division is present. It consists of two slender lobes, and the black spine lies in the upper part of the lower and longer one. Almost surrounding these lobes and the setae is a large and prominent fillet, its edges attached in front to the lower of the two lobes. The setae above the spine consist of 2 anterior falcate heterogomphs and 3 posterior long-tipped homogomphs. Below the spine is a single posterior long-tipped heterogomph and 10 inferior falcate heterogomphs. The 2nd foot resembles the Ist, but in the 3rd foot the dorsal division appears. 1921.] Fauna of the Chilka Lake : Polychaeta. 597 In the roth foot (fig. 10D, and text-fig. 8b) the dorsal cirrus is equal in length to the dorsal ligule, the latter being stout and conical. The dorsal setigerous lobe consists of a small papilla lying above the dorsal spine. From the front face of this papilla the fillet runs above and behind it, and is attached below to the median ligule. The setae are all long-tipped homogomphs. The median ligule is stout and rounded. The ventral division is very much asin the Ist foot. The ventral ligule is finger-shaped, and the ventral cirrus is small and slender. The long-tipped homo- gomphs and heterogomphs are of the usualtype. The falcate heterogomphs (fig. 10G) have short terminal pieces, with smooth tips, and the spinous portion is short. The 2oth foot is very similar, except that the two lobes of the ventral division are much smaller. In the 30th foot the base of the dorsal ligule has begun to elongate, carrying out the dorsal cirrus with it. The median ligule is also a little longer. In the ventral division the lower of the two lobes is broad and thin, and forms an almost indistinguishable part of the setigerous fillet. The upper lobe is very much reduced. The two upper anterior setae in the ventral group of fal- cate heterogomphs are much stouter than the other setae. In the succeeding feet these tendencies are accentuated. The dorsal division of the foot is greatly enlarged (figs. 10E, IoF). The papilla in the dorsal setigerous division disappears, but the fillet remains (text-fig. 8c), both edges being fused to the median ligule. In the ventral division the small upper lobe disappears, and the lower one completely fuses with the fillet. In the 7oth foot (fig. ror) the dorsal division is very large and conical, with many glands. The dorsal and ventral setigerous lobes are represented only by the fillets. In the anterior part of the ventral division two of the falcate hetero- gomphs just above the spine, and one below it are much thicker than the rest. All the feet are very vascular. This species has some points of re to N. variegata, Grube (Paranereis elegans, Kinberg, 1910, p. 53), Ehlers (10017, p. 112), especially in the colour pattern, but differs in the arrangement of the paragnaths, the length of the dorsal cirri and other foot lobes, etc. In the shape of the feet, especially in the position of the dorsal cirri, it is more closely related to Pseudonereis novae-hollandiae, Kinberg (1910, p. 52), from Port Jackson, Australia, but differs from the latter in having much shorter dor- sal cirri and in the paragnaths of Groups I and V. From N. camiguina, Grube (1878, p. 87), it differs in the shape of the head, length of the tentacular cirri, colour pattern, ete. Habitat.—Ten specimens of this species were found, living in burrows or crevices in oyster shells, at Manikpatna in the outer channel of the Chilka Lake. They were taken in September, during the freshwater season, when the water in the outer chan- nel was quite fresh. From the nature of the habitat, it is probable that the species lives there throughout the year. 598 M emoirs of the Indian Museum. (Vor. V, Dendronereis aestuarina, sp. nov. (Plate XX, figs. 4A-N, and text-figs. 9a-h.) Nineteen specimens of this species were obtained in brackish water in the Gangetic delta, in very good condition. The type specimen is 160 mm. long, and consists of 160 setigerous segments. It is a female, full of immature ova. In life the body is reddish brown, the colour being probably due to the blood, and the gills and dorsal vessel are bright red. The body in front of the gills is much darker, and appears to be coloured with brown pigment.' The preserved specimens are colourless. On the ventral surface of segments 3—13, in the median line, there is in all specimens a definite pattern of grooves (fig. 4c). The head (fig. 4a) is deeply indented in front, where it bears two small tentacles. The posterior part is short and wide, with rounded corners. It bears two pairs of large eyes, the pair on each side being close together. In front the head is prolonged into the two large conical palps. The peristomium bears four pairs of tentacular cirri, with large jointed ceratophores. The posterior dorsal pair are much the longest, and when laid backwards they reach to the end of the oth setigerous segment. They are nearly three times as long as the posterior ventral pair, and 24 times as long as the anterior dorsal pair, whilst the latter are nearly twice as long as the anterior ven- tral pair. The proboscis (figs. 4A, 4B) is not armed with horny paragnaths, but on the anterior border of the basal segment it carries a number of papillae. Of these the two largest are near the median dorsal line. On the ventro-lateral region there are two on each side, and there are three smaller ones in the median ventral area. The papillae are pear-shaped or fusiform. The whole of the basal segment is covered with low rounded papillae, but these are probably produced by the contraction of the proboscis in the preservative, and have no relation to the papillae on the anterior margin. The maxillary segment is devoid of papillae. The jaws have 14-17 teeth, 2 or 3 at each end of the row being indistinct. 2 The branchiae commence on the I5th foot (fig. 4F). Each branchia consists essentially of the greatly enlarged, flattened, aud branched base of the dorsal cirrus. The dorsal cirrus itself is, as usual, a simple lobe, attached to the tip of the enlarged base. Two large blood-vessels traverse the main stem, giving off or receiving numerous branches which divide several times, and penetrate the filaments. Apart from the vessels in the filaments, there are numerous capillaries in the walls of the main stem of the branchia. In the Ist branchia (fig. 4F) there are 8 clavate lobes on the outer edge of the main stem. They are slightly contracted at the base, and diminish in size towards the tip of the stem. ‘The 2nd pair of branchiae resemble the Ist pair. The 3rd pair, on the I7th feet (fig. 46), have each Io filaments, and indica- tions of another row on the outer side of the stem. ‘The 4th pair, on the ISth feet, show the complete development of the branchiae (fig. 4H). The main stem is almost ‘ 1 This description is from a coloured figure, drawn from a living specimen, by A. Chowdhary. 1921.] Fauna of the Chilka Lake : Polychaeta. TEXT-FIG. 9.—Parapodial diagrams of Dendronereis aestuarina, sp. nov. a. Ist foot (lobes a and b sometimes absent) ; b. 2nd foot; c. 3rd foot; d. 4th foot; e. roth foot; f. 15th foot; g. 22nd foot; h. 7oth foot. : = Homogomph setae with long smooth tips; x = Homogomph setae with coarsely serrated tips; o =Homogomph setae with thick shafts and short tips. 600 Memoirs of the Indian Museum. Vor. V, completely occupied by the two blood-vessels. There are 13 pairs of branches, each bearing a double row of slender filaments which are nearly as long as the branches. The slender dorsal cirrus is carried at the tip of the stem. The 5th, 6th and 7th pairs are similar, with 14 pairs of branches. The 8th and last pair, on the 22nd feet, are much smaller, with 9 or 10 pairs of branches. The base of the dorsal cirrus of the 23rd foot is short, simple, and slender. In the anterior region of the bedy the feet differ so much that it is necessary to examine them all carefully. The first foot (fig. 4p, and text-fig. ga) consists only of the dorsal cirrus with a lobe beneath it, which is either the dorsal or the median ligule, and the ventral setigerous division. ‘The dorsal cirrus, seated on a jointed base, is long and tapering. The ventral division is composed of 6 or 8 lobes and the ventral cirrus. The setae are in two groups, each guarded by a fillet, the single spine lying between the groups. They are all homogomph setae (fig. 41), with long minutely serrate or smooth terminal pieces. A single seta in the posterior row of the upper group differs from the rest in having a shorter terminal piece with long stout teeth (fig. 4m). The arrangement of the lobes and setae is shown in text-fig. ga. Sometimes the lobe b is absent, sometimes a and 0. The foot is very vascular, blood-vessels passing into each lobe. The 2nd foot (text-fig. 95) resembles the Ist, but it has 11 lobes in addition to the ventral cirrus in the ventral division, and two setae with coarsely serrate terminal pieces. The lowest ventral setae, as in all other segments, have shorter terminal pieces than those in the upper part of the foot. In the 3rd foot (text-fig. gc) the dorsal setigerous division appears. It consists in addition to the dorsal cirrus and dorsal ligule, of two other lobes, the posterior one being the median ligule, between which lies the spine. Above and behind the anterior lobe is a group of a few slender homogomphs like those in the ventral division, guarded externally by a fillet. The ventral division consists of 15 lobes, in addition -to the ventral cirrus. The 4th foot is very similar (text-fig. gd), especially the dorsal division. In the ventral division the setae now form a continuous series, guarded externally by a fillet. There are 16 lobes in the ventral division, of which 9 are behind, and 7 in front of the setae. The coarsely serrate setae are more numerous. The base of the dorsal cirrus shows signs of flattening. The 5th foot is similar to the 4th, wi more numerous setae. The roth foot (fig. 4E, and text-fig. ge) shows still greater complexity. The base of the dorsal cirrus is large and flattened. In the dorsal division the setae are more numerous, and an additional small lobe lies in front of the spine. The ventral division has 18 or 19 lobes, of which 12 form a fringe behind the setae, whilst the anterior 6 or 7 lobes are almost surrounded by the setigerous fillet. The upper posterior lobe of the ventral division is flattened at the base and pointed distally. The rith-r4th feet are similar. At the 15th foot (fig. 4F, and text-fig. of), which carries the first branchia, there is a marked change. The dorsal division is reduced to the two posterior lobes, the upper one corresponding to the dorsal ligule of the anterior feet, the lower one to a TO2T.] Fauna of the Chilka Lake : Polychaeta. 601 the median ligule. The setae lie above and in front of the median ligule, to which the edges of the setigerous fillet are attached. The ventral division consists of two anterior lobes, almost surrounded by the fillet, three posterior lobes, the ventral ligule beneath the setae, and the ventral cirrus. The coarsely serrate setae are few in number. In the 16th foot one of the posterior ventral lobes has disappeared. The 17th and 18th feet (fig. 4H) are similar. The two posterior lobes of the ventral division are flattened at the base, and the spine terminates inasmallpapilla. Inthe roth and 20th feet (fig. 43) the upper anterior lobe of the ventral division has a small papilla beneath it, which disappears in the 21st foot. The posterior lobes are still more foliate. In the 21st foot the lower posterior lobe is almost entirely fused with the upper one. In the 22nd foot (text-fig. 9g) the two lobes of the dorsal division show signs of flattening. In the ventral division, the two posterior lobes have fused to form a single foliate lobe. 32 In the 23rd-4oth feet (fig. 4K) the two dorsal lobes gradually grow thin and foliate, and the lower one assumes an anterior position. In the ventral division the upper anterior lobe is gradually reduced to a small papilla, and finally disappears, the ventral division then consisting of two foliate lobes, one on each side of the setae, a conical lobe between them, the ventral ligule and the ventral cirrus. Between the 5oth and 6oth feet a new type of homogomph seta appears (fig. 4N). It is stouter than the others, and the terminal piece is short and smooth, tapering rapidly to a very slender tip. It occurs in the lower part of the dorsal division, and in the upper anterior part of the ventral division. There is no sharp distinction between these setae and the normal homogomphs, intermediate forms occurring both in shape and position. In the 7oth foot (text-fig. 9%) these thick setae occur also on the lower anterior end of the ventral division. None of the coarsely serrated homo- gomphs were observed after the 35th foot. In the posterior feet the setae become fewer in number, the lobes of the feet greatly reduced, and the whole foot decreases in size. The tips of the shafts of the setae are more swollen than those of the ante- rior segments. The anal segment bears two slender anal cirri, equal in length to the last six segments. Members of this genus have rarely been found, and only two species have been described. ‘The first of these Dendronereis arborifera, Peters, from the coast of Mozam- bique differs conspicuously in the structure of the branchiae, the main stem of which carries only simple branches. The second species, D. pinnaticirris, Grube (1878, p. 92), from the Philippines, agrees with the present species in the structure of the branchiae. It differs, however, according to Grube, in having no papillae on the proboscis, and in having only 6 teeth on the jaws, as compared with 14-17 in the present species. _ In D. pinnaticirris, the branchiae begin on the 12th or 13th segments, and there are 10-12 pairs of branchiae. In D. aestuarina they begin on the 15th foot, and there are only 8 pairs. In D. pinnaticirris all the branchiae appear to have two rows of com- 602 Memoirs of the Indian Museum. [Von WV, pound branches, whilst in D. aestuarina the three anterior pairs have only a single row of simple branches. The fully formed branchia of D. aestuarina has more branches than that of D. pinnaticirris. The anterior feet of the present species have many more lobes than those of D. pinnaticirris, but the brief description and figures given by Grube do not permit a close comparison as regards the structure of the feet. A re-examination of D. pinnaticirris would doubtless show other important differences. Only a single specimen of D. arborifera has ever been examined, and according to Ehlers (1864-68, p. 581) the proboscis is retracted, and has no paragnaths. It is possible, however, that papillae may have been overlooked on the retracted proboscis, as they are often difficult to observe in this condition. In D. pinnaticirris, according to Grube, there are neither paragnaths nor papillae on the proboscis. ‘Their presence then, in D. aestuarina, was rather a surprise, and I still think it possible that they have been overlooked in the other two species of the genus. Habitat.--19 specimens of this species were collected onthe 27th of November, IQII, by Mr. S. G. Platts in the Sunderbans, a district in the Gangetic Delta, and sent by him to the Indian Museum. In a letter accompanying the specimens he gave the follow- ing information :—‘‘ The Polychaetous worms I sent you this morning were found by me in a small pool of brackish water inside the protective embankment of a clear- ance fairly high up the Munda river. There were hundreds swimming round and round three or four vortices, and it looked as if they were coming up from the ground at these points. A few minnows were hovering about. These used to occasionally pull down a worm, but the fish were either not strong enough to swallow the worms, or the worms were not relished, since they were invariably let go.” The water in which the worms were despatched was analysed by Mr. David Hooper, who reported as follows :— “The sample of brackish water you left with me yesterday contains 833 parts of solid matter and 376°3 parts of chlorine per 100,000 parts. This approximately represents a mixture of three parts of fresh water with one of sea water.” The two species of Dendronereis previously described were found in sea-water, so that the occurrence of the present species in water almost fresh is of considerable interest. The three known species live on or near the shores of the Pacific and Indian Oceans, and the genus is not known elsewhere. Genus Dendronereides, gen. nov. The following will serve as a preliminary diagnosis of the genus, until additional species are known :—Proboscis armed only with soft paragnaths. Dorsal setigerous lobe absent in first and second feet. In some of the anterior feet, branchiae are present, vn the form of numerous filaments situated below the dorsal cirrus. They are not provided with blood-vessels. Setae of two kinds, falcate homogomphs, and spinose homogomphs. In all feet except a few anterior ones there is a peculiar gland opening to the exterior beneath the dorsal cirrus. The ventral ligule is absent. In the post-branchial region the foot is greatly simplified. The presence of parapodial branchiae in this genus at once suggests relationship 7927] Fauna of the Chilka Lake : Polychaeta. 603 to Dendronereis. The obvious distinction that the branchiae of Dendronereis are formed by modification of th, dorsal cirrus, whilst those of Dendronereides lie beneath and do not involve the dorsal cirrus, is not so decisive as it appears at first. I have shown above that the branchiae of Dendronereis are attached to the base of the dorsal eirrus, the latter structure remaining unchanged, at the tip of the swollen base. If figs. 6E, 6H, are examined, it will be seen that the branchiae of Dendronereides also are attached to the base of the dorsal cirrus, though in this case the base is not elongate, and the area of attachment is condensed. A more important distinction between the two genera lies in the fact that the branchiae of Dendronereis are sup- plied with an elaborate system of blood-vessels, whilst those of Dendronereides are apparently not. In both genera the gills are concentrated on the anterior part of the body. These resemblances, however, may be due to convergence. It is remarkable, however, that in both genera there is great simplification of the foot behind the bran- chial region, and an unusually large number of foot-lobes in the anterior segments. In both genera the proboscis is provided with soft papillae, and devoid of horny paragnaths. In both genera also, heterogomph setae are absent, and a number of the spinose setae have very long slender teeth. This genus also shows rather more distant affinities with Tylonereis, Cerato- cephala, and Tylorrhynchus. Dendronereides heteropoda, sp. nov. (Plate XXI, figs. 6A—N, and text-figs. roa-}.) Nine specimens of this species are available. They are all immature, with new segments in process of formation in the posterior region. The type specimen is 66 mm. long, and consists of 140 segments. The body is long and slender, and attains its greatest width at the 8th setigerous segment. Dorsally the anterior margin of each segment is distinctly concave. The anterior segments are two or three times as wide as long. Further back, the length of the segments increases till it exceeds the width, but in the posterior region the segments become short again. The anal segment is large and button-shaped, and the anus is terminal. The two short anal cirri are ventro-lateral in position, and taper to filiform tips. The head, palps, tentacular cirri, and anterior dorsal region of the body are rusty red in colour, the pigment being most marked on the head. In some of the specimens the colour is quite gone, except for the three dark patches on the head. The head (fig. 6A) is relatively very small, and the width greatly exceeds the length. In front it is deeply bilobed, with the small tentacles at the tips of the lobes. The eyes are rather small in the specimen figured, and are larger in other individuals. They are placed near the postero-lateral angles of the head, and the anterior pair are more widely separated than the posterior pair. The palps are short, stout, and contracted. At the back of the head, in the median dorsal line, . is a narrow transverse band of deep reddish brown pigment. At the base of each tentacle there is a more diffuse patch of the same colour, The bases of the palps, 604 Memoirs of the Indian Museum. [Vor. V, and the ventral surface of the head and body are colourless. ‘The posterior dorsa tentacular cirri are stout at the base, and are twice as long as the anterior dorsal pair, three or four times as long as the posterior ventral pair. The proboscis is only partially protruded in one specimen, showing the basal or oral ring (fig. 6B). In the dorsal region there is a group of 7 papillae; 3 median of which the largest is in front, the smaliest behind; and lateral groups of 2 papillae on each side. These papillae may represent groups V and VI, or group V only. The three anterior papillae are seated on large hemispherical bases. Ventrally and laterally, in groups VII and VIII, there is a ring of papillae in two rows, 9 in each row, those in the anterior row being the largest and the lateral ones larger than the ventral. The dis- tal, or maxillary ring had to be dissected, and as the proboscis is small, and the papillae soft, inconspicuous and crushed, their numbers and arrangement can only be stated approximately. Group I consists of about 12 papillae of different sizes, II ofa small number of larger and more conical papillae. Ventrally are indistinct groups of very large conical papillae, which may represent two groups on IV, or pos- sibly three groups on III and IV. The jaws are of a pale amber colour, with numerous small teeth, 18-20 in num- ber in one specimen, only 10 in another. In the 1st and znd pairs of feet, the dorsal setigerous division is represented only by the dorsal cirrus. Above the ventral setigerous region (fig. 6c, and text-fig. 10a) is a clavate lobe representing either the dorsal or the median ligule. The dorsal and ventral cirri are short, stout, and conical. In the ventral setigerous division there are three lobes, two slender ones in front, and a posterior foliate lobe. The setigerous fillet curves right round the two anterior lobes, projecting outwards to form the posterior lobe behind, and a small papilla in front, pierced by the black spine. ‘Two kinds of setae are present, both having homogomph shafts. At the front of the foot are two small groups of setae, above and below the spine, with short smooth terminal pieces (fig. 6m), rounded at the tip. The remaining setae have relatively long tapering ter- minal pieces with very long slender teeth on the lower half (fig. 61). In the 2nd foot the upper ligule is nearly as large as the dorsal cirrus, and the line of setae is continuous behind the anterior lobes. | In the 3rd foot the dorsal setigerous lobe appears. Between the dorsal cirrus and the ventral setigerous lobe are the dorsal and median ligules, and a setigerous fillet lying in front of and just above the median ligule. There are three homogomph setae, having slender terminal pieces with long serrations, as in the ventral division. The spine lies beneath the setae. In the 4th foot (fig. 6p, and text-fig. 10b) the dorsal division has now two lobes between the dorsal cirrus and the median ligule, and the setae are more numerous. In the 5th and 6th feet there are three of these dorsal lobes, four in the 7th foot, five in the 8th foot. In the oth foot (fig. 6E, and text-fig. roc) there are six dorsal lobes. The me- dian ligule is now larger than the dorsal cirrus. The ventral division remains practi- Fauna of the Chilka Lake : Polychaeta. Le] OO 02 2200 388 © 000 802 O 0000 09% = 000 0Q Q 4 000 000 TEXT-FIG. 10.—Parapodial diagrams of Dendronereides heteropoda, gen. et sp. nov. a. Istifoot ; b. 4th foot ; c. oth foot; d. 21st foot; e. 27th foot; /. 30th foot. - = Homogomph setae with long tips; x = Homogomph setae with short tips. oo D 605 606 | Memoirs of the Indian Museum. Mer, cally unchanged. The setae of the dorsal division have longer and more slender ter- minal pieces than those of the ventral division, and the serrations are not so long (figs. 6K, 61). The falcate homogomphs are confined to the ventral division. In some of the long-tipped setae the shaft is quite homogomph, but in others it is produced into a point at one side (fig. 6N), the so-called “hemigomph’ condition. The 11th foot has 8 lobes, the 12th has 9 lobes, in the dorsal division. In the ventral division the upper anterior lobe has increased, and the posterior lobe decreased in size relatively. In the 13th foot the dorsal division has 11 lobes, in the r4th foot 21 lobes, in the 15th foot 28 lobes, and in the 16th foot 32 lobes. The number of these lobes or filaments reaches its maximum in the 21st foot, where there are about 65-70 (fig. 6F, and text-fig. 10d). They appear to be arranged in three groups, each group having a common stout stem, lying side by side. The median ligule is large, and the setigerous fillet of the dorsal division rises from its upper side, curves for- wards and upwards, and is attached beneath the filaments. In the ventral division the posterior lobe is greatly reduced. In the 23rd foot the lower anterior lobe of the ventral division has disappeared, and the upper one is small. The posterior lobe still has a pointed tip. The setae are much fewer in number, and there are about 55 filaments in the dorsal division. In the 25th foot the dorsal filaments are much fewer in number, and appear to spring from four main stems. The ventral division is now without pointed lobes, and the setigerous fillet forms a broad rounded flap behind the setae, and a smaller one in front, the setae curving round the minute remnant of the upper anterior lobe, which is first visible in side view. The spines are only black near the tip, having grown gradually paler from the front backwards. In the 26th foot (fig. 6H) the filaments rise from four stems, the lower one being unbranched. In the 27th foot (text-fig. 10e) there are only three filaments rising from a single stem. The ven- tral division consists of two fillets attached to the median ligule. The remnant of the anterior upper lobe, round which the setae form a circle, is very small, and is not visible in side view. In the 28th and 2oth feet there is only a simple short filament beneath the dor- sal cirrus, and in the 30th foot this also has disappeared (fig. 67, and text-fig. 10/). The lobe inside the ventral setae has also vanished. In the middle and posterior parts of the body the foot does not change much in shape, but gradually grows smaller, the setae become fewer in number, and the median lobe and setigerous fillets smaller and more indistinct. The falcate homogomph setae, with smooth terminal pieces (fig. 6M), are regu- larly present in the ventral divisions of the anterior 25 feet or so. They then become fewer in number, and are frequently absent in many adjacent feet. They were noted, however, in the 7oth and 8oth feet. The ventral and spinose setae have, as a rule, shorter terminal pieces with longer serrations, than those of the dorsal division (figs. 6K, 61), but both kinds frequently occur in the ventral division. It can hardly be doubted that the dorsal filaments of the anterior feet have a 1921. | Fauna of the Chilka Lake : Polychaeta. 607 respiratory function. They are not, however, supplied with blood-vessels as in Dendronereis. The filaments have an ill-defined central lumen (fig. 66). Morpho- logically they appear to represent the proliferation of the dorsal ligule. A remarkable glandular organ occurs in most of the feet. They first appear in the I4th setigerous segment, and continue to the end of the body. This distribution shows that they have no functional connection with the branchial filaments. At first they are small, but they rapidly increase in size. They lie in the upper part of the foot, and open to the exterior just beneath the dorsal cirri. Their usual shape and position is shown in figs. 6F, 6H, and 67. Each consists of a number of elongate cells with granular contents, the whole being surrounded by a thick muscular coat. There is no common duct to the gland, and all the elongate cells apparently open to the exterior through the thickened lip ofthe gland. In the middle and posterior seg- ments the gland decreases considerably in size, and consists only of a few elongate cells with a very thin muscular coat. In the segments where the glands occur, the nephridia are very large, with many black granules in the duct cells. The nephridia occupy most of the cavity of each foot, and are closely surrounded by branched blood-vessels. In many respects this species is one of the most aberrant of the Nereids. The parapodia are very heterodox in structure, and it is difficult to homologise the various lobes. One of the most remarkable features is the presence of only a single lobe beneath the ventral setae. This may be either the ventral ligule or the ven- tral cirrus. As the latter is a much more constant structure than the former, it is highly probable that this lobe is the ventral cirrus, and that the ventral ligule is absent. As already mentioned, the dorsal lobe in the ıst and 2nd feet may be either the dorsal or median ligule. In all the other feet the median ligule is present. The dorsal ligule is absent at least after the 2gth foot. Habitat.—In brackish pools, salt lakes, Barantolla, near Calcutta. The salinity is very variable, but never high, probably never exceeding sp. gr. 1.015 at 25°C. The specimens were collected in November. Family NEPHTHYDIDAE. Nephthys polybranchia, sp. nov. (Plate XXIV, figs.1rA-G, and text-figs. I1a—0.) Twelve specimens of this species were taken in Chilka Lake. They were all approaching maturity, full either of ova or sperm. One specimen was 22 mm. long, and had 52 setigerous segments. Another was 19 mm. long, with 50 setigerous seg- ments. No trace of colour remains in the specimens. The body is widest near the anterior end and tapers gradually towards the tail. The head (fig. ııA) is rather elongate, with parallel sides. The dorsal tentacles are atthe anterior angles, and the ventral pair are a little further behind on the lower surface of the head. At the posterior dorsal angles of the head are two small rounded papillae. The brain is clearly visible behind the head. It is bilobed, and on its 608 Memorrs of the Indian Museum. [VoL. V, upper surface lie two small but distinct black eyes. No individual had the proboscis extruded, and so far as could be ascertained by dissection there are 20-22 rows Of papillae, 4 in each row. There is no spe- cially large one in the mid-dorsal line, but in the two median rows the anterior papilla is a little in front of the other rows, although only of the same size. This point needs to be verified with better material than was available. The ıst foot (figs. ITA, IIB, and text- fig. Ila) projects obliquely or directly to the front, and does not attain the ante- rior border of the head. It has well- marked dorsal and ventral divisions, with spines, but the lamellae are minute and the branchiae are absent. On the posteri- or face of each division there is a small cirrus. The dorsal division is conical. TEXT-FIG. 11.—Parapodial diagrams of Nephihys Above the spine there is a group of long polyhranchia, sp. nov. : Fe one D mother. slender capillary setae, with smooth or "= slender capillary setae; x = camerated setae. minutely serrated flattened blades. Below the spine is a group of camerated setae. There is a minute posterior lamella. The ventral division is truncated at the tip. The spine pierces a conical papilla, which is completely surrounded by the setiger- ous fillet or lamella. Emerging between the papilla and the fillet is a ring of long smooth capillary setae. No camerated setae are present in the ventral divi- sion. ; The peculiar nature of the Ist foot may perhaps be explained by regarding it as composed of two imperfect feet. If the text-fig. 11a is turned round till the cirri are ventral in position, the resemblance of each division to a single division of a typical foot (text-fig. 11d) will be at once apparent. The ventral division, which does in fact lie somewhat in front of the dorsal division, would represent the rst foot, armed only with capillary setae, whilst the dorsal division would be the 2nd foot, with an anterior row of camerated setae and a posterior row of capillary setae. The two cirri would correspond to the ventral cirri. The papillae at the posterior dorsal angles of the head may be the reduced dorsal divisions of the 2nd pair of feet. On this hypothesis the two anterior feet would be represented only by the ventral division of the typical foot, as in Nereıs. In the 2nd foot (fig. IIC) the dorsal and ventral divisions are widely separated from each other, and the various lamellae of the normal foot are present, though still very small. There is a small ventral cirrus, and camerated setae are present in both divisions. At the ventral side of the dorsal division there is a small lobe (fig. IIC, a), which appears to be the rudiment of the branchia. 1921.] | Fauna of the Chilka Lake : Polychaeta. 609 In the 3rd foot the lamellae are more prominent, the branchia is rather larger, and there is a slight indication of the dorsal cirrus. The anterior row of setae in the ventral division are mostly barred, but a few smooth setae occur at the upper and lower ends of the row. The setae are now in two rows, before and behind the spines, not above and below them as in the Ist and 2nd pairs of feet. The 4th foot is very similar, but a distinct change occurs in the 5th foot, which is almost normal. The branchia is large and thick, with a small cirrus on its upper basal portion. The normal condition of the anterior feet is shown in fig. IID (also text-fig. 116) of the Ioth foot. The spines, which are ringed at the tip, terminate in each division at the tip of a conical papilla, which is surrounded by a ring of setae. The setae in front are of the usual camerated type, rather short and stout. The posterior setae are very long and slender capillary setae, with slightly flattened blades very finely serrated along one edge (fig. 11F). They are more numerous than the camerated kind, and tend to invade the front row at the upper and lower ends of the bundles. Surround- ing the setae is a fillet which, in both divisions, projects a little beyond the spine on the posterior side of the foot, but not on the anterior side. In the dorsal division the lower ends of the fillet are attached to the branchia, in the veritral division to the ventral cirrus. The dorsal cirrus is small but distinct. The vascular system of the foot consists of a vessel which traverses the branchia as a single loop (fig. 11D). On emerging, it divides, and the smaller branch runs down to the ventral divi- sion. It then curves back and terminates in a flask-shaped ampulla. In fully grown specimens the barred setae are found only in the anterior 16 feet, ' whilst smaller specimens may have them in 12-15 feet. They are then replaced by smooth capillary setae (fig. 11G), which are smaller than those in the posterior row. The blade widens rather suddenly, and then tapers to a fine tip. The 20th and 30th feet (fig. IIE) are very similar, except that the anterior dorsal lobe of the fillet is rather larger, the posterior lobes are smaller, and the branchia is relatively very large. In the posterior feet the fillets are small, and are considerably surpassed by the spinal lobes. At the 45th foot the branchia suddenly becomes small, and is absent on the succeeding 7 posterior segments. The anus is terminal, and there is a single short tapering anal cirrus. This species is characterised by the structure of the proboscis, the shape of the head, and by the condition of the feet. As regards the latter, the most important characters are the size and shape of the lamellae; the restriction of the camerated setae to the anterior feet, and their replacement in the middle and posterior feet by peculiar capillary setae; the distribution of the branchiae, and the presence in them of only a single vascular loop. Habitat.—The 12 specimens, all mature or approaching maturity, were taken at 4 stations, all in the south-west end of the lake between Rambha and Nalbano, the shore being mud or sandy mud. Two of the stations were worked in February, the salt-water season, and two in September and November, the freshwater season, but in this part of the lake the specific gravity only ranges from I’00I-I’'OII5. 610 Memotrs of the Indian Museum. More Nephthys oligobranchia, sp. nov. (Plate XXIV, figs. 12A-I2C.) This species is unique in the collection, insomuch as it is represented by specimens from the Chilka Lake and also from another locality, viz. the Cochin Backwater, near Ernakulam, in the south-west of the Madras Presidency. The superficial resemblance of this species to Nephthys polybranchia is so strong that at first it was regarded as a mere variety of that species, but a closer examina- tion revealed such striking differences as to indicate that the two forms are not closely related. The description will be confined to these points of difference. A mature male specimen was 17 mm. long, consisting of 51 setigerous segments. - An- other had 47 setigerous segments. The head (fig. 12A) is rectangular, and the length slightly exceeds the width. At the posterior dorsal angles are two small rounded papillae. The brain is rounded behind, with a deep posterior indentation, and carries two small black eyes. The front margin is almost straight. The head in fig. 12A is distorted and widened, owing to the protrusion of the pharynx. The latter organ has in the distal region, I4 rows of papillae, 3-5 in a row. In the anterior dorsal line there is a conspicuously long median papilla, but no corresponding ventral papilla. The mouth is surrounded by 16 bifid papillae. In small specimens the Ist pair of feet are directed forwards, alongside the head, which they slightly surpass, but in adults they point obliquely outwards, and do not reach the front of the head. This species differs only slightly from N. polybranchia as regards the shape and arrangement of the setae. Camerated setae are as a rule confined to the anterior 13-15 pairs of feet, but in one specimen they persisted as far as the 18thfeet. The setae which replace them subsequently in the anterior rows, differ slightly from those of N. polybranchia, the blade being narrower and more uniform in width. There is a marked difference between the two forms in the structure of the feet. The first fully formed branchia is on the 6th foot, that on the 5th foot being very smali. They cease abruptly on the 2oth-23rd foot, and are quite absent from the feet of the middle and posterior regions of the body. Thus in a specimen having 47 seti- gerous segments, branchiae are present on 17 pairs of feet only, the 6th to the 22nd, and one specimen with 5I setigerous segments has 30 posterior abranchiate pairs of feet. A typical branchiate foot, the roth, is shown in front view in fig. 12B. The posterior lamellae are very small and are surpassed by the spinal lobes, so that in front view they are not visible. The anterior lamellae resemble those of N. poly- branchia, and the ventral cirrus is equally small. The dorsal cirrus is very insigni- ficant, being represented only by a small lobe on the upper side of the base of the branchia. The vascular system of the foot differs from that of the preceding species. The vessel which penetrates the branchia, after emerging, sometimes for a considerable ni Oyases| | Fauna of the Chilka Lake : Polychaeta. 611 distance, returns and traverses the branchia again, so that there appear to be four vessels in the branchia. The branch to the ventral division of the foot terminates as in the preceding species in a rounded ampulla. Ina few of the anterior feet, branchiae containing only a single loop were occasionally noted, but a double loop is the nor- mal condition. The presence of two loops in the branchiae of this species may be correlated with the relatively small number of branchiae. One of the abranchiate feet is shown in fig. r2c. In the posterior region of the body the feet are large and deeply bilobed, the lobes being large, conical, and widely divergent, giving a characteristic appearance to these specimens. The anus is terminal, and there is a single anal cirrus. Mature specimens, con- taining ova and sperm, were taken in January, February, March and September, in both the fresh and salt-water seasons. This species differs from N. polybranchia (I) in having 14 rows of papillae on the proboscis, as compared with 22; (2) in having a long anterior unpaired papilla on the median dorsal line of the proboscis ; (3) in the distribution of the branchiae, which occur fully developed on the 6th foot, and disappear on the 20th to the 23rd foot, whereas in N. polybranchia, the branchiae are large on the 5th foot, and persist almost to the end of the body; (4) inthat the branchiae contain a double vascular loop, whereas there is only a single loop in N. polybrancha ; (5) in that the posterior lamellae of the feet are considerably surpassed by the spinal lobes. Habitat.—18 specimens were collected in the Cochin Backwater, near Ernakulam, in the south-west of the Madras Presidency. The salinity of the water is probably very variable in this locality, but precise information is not available. In the Chilka Lake this species was taken at ten stations, all in the south-west end of the lake, between Rambha and Nalbano. The species was apparently taken usually on a muddy bottom, both on the shore, and dredged down to 15 feet. Eight stations were worked during the salt- water season, and two during the freshwater season, but the gravity of the water in this part of the lake only varies between the limits of 1°0015 and 1'O115. Family EUNICIDAE. Diopatra variabilis, sp. nov. (Plate XX V, figs. I4A-I4R, and text-figs. I2a-e.) Of this species there are fragments, more or less complete, of seven large speci- mens, all from the same station in Rambha Bay, and twenty-two small specimens from various parts of the lake. The nature of the habitat, and the relation between the large and small specimens, will be discussed below, when the large individuals have been described. The largest (type) specimen, inthree fragments, is 312 mm. long, composed of 451 segments. There is a short gap, probably consisting of about 20-30 segments, between the first and second fragments, as is shown by the condition of the branchiae, and a few posterior segments are missing, so that the animal originally had probably about 612 Memonrs of the Indian Museum. [VOL 480 segments. The greatest width of the body, not including the feet, is 4-5 mm. The whole body is suffused with iridescent purplish green, especially marked in the anterior region. The basal parts of the various appendages of the head are deeply coloured. In several specimens the anterior part of the body (10-14 seg- ments) is paler in colour than it is further behind, especially the branchiae. This may be due to regeneration of the anterior end. The body is round in front, some- what flattened behind. The head (figs. 144, 14B), as usual in this genus, is partly concealed dorsally by the projecting peristomium, ventrally by the palps. The median antenna is II mm. long, very slightly exceeding the inner paired antennae, and nearly twice as long as the outer pair. The median antenna if bent back reaches to the middle of the roth setigerous segment. The swollen bases of the antennae are composed of 12-13 rings. The cephalic lobe is small, and the posterior part is hidden by the peristo- mium. It bears two dark pigment spots which may represent eyes. The outer antennae are attached beneath the median pair. The frontal palps are fusiform, shorter than the ringed bases of the antennae. The palps are large, and are grooved on the dorsal surface. The lower surface shows two portions differing in appearance, the inner (anterior) half being the more tumid, and marked off by a deep groove from the outer half. The peristomium is very little longer than the succeeding segments, and slightly narrowed. The tentacular cirri are about as long as the ringed bases of the antennae. At first sight they appear to be attached to the anterior margin of the peristomium, but a closer examination shows that they really spring from the dorsal surface, a short distance behind the front margin. Ventrally the peristomium forms a wide loose lower lip. In six out of the seven large individuals the first branchia is on the 4th foot. In the other one, which is the smallest of the seven, the first branchia is on the 5th foot. The branchiae (figs. I4A, I4F) increase in size as far as the 4th—6th pair. The largest have about seven whorls of moderately long filaments, attached to a stout stem, the base of which is more or less distinctly ringed. The anterior 77 pairs of branchiae are branched and are approximately as long as the dorsal cirri, the two organs gradually decreasing together. The 78th pair consists of simple filaments. There are 20-40 segments with these simple filiform branchiae, the last few being like minute papillae. The mandibles (fig. 14D) are fused at the anterior end. The front margin has a rounded notch at each side and a more acute notch on each lateral margin. The posterior limbs are black and slightly expanded distally. The maxillae are very broad (fig. 14C), the supports being rather short and rounded, the forcipate processes stout and boldly curved. The great dental plates have 6-8 teeth on the right side, 7-9 on the left. The azygos plate has 6-9 teeth. The crescentic plates have 7-10 teeth. On the outer side of the latter are two rectangular plates, without teeth. The feet are all provided with long slender dorsal cirri, each of which has a bundle 1921.] Fauna of the Chilka Lake : Polychaeta. 613 of slender spines running into its base, but not piercing the skin. In the four ante- rior pairs of feet the setigerous lobe is relatively much larger than in the others. The Ist foot (fig. I4E, and text-fig. 12a) has a large fusiform lobe behind the setae. In front of it lie two low rounded papillae, the dorsal one being the larger. A single capillary seta emerges behind the dorsal papilla. Ventrally there are 4 bifid hooks, 2 thin ones in front of the ventral papilla, and two thicker ones between the two papillae. In addition there are many stout and slender setae in the setigerous lobe and in the base of the dorsal cirrus which do not pierce the skin. In the bifid TEXT-FIG. 12.—Parapodial diagrams of Diopatra varıabılis, sp. nov. a. Ist foot; b. roth foot; c. 20th foot, setigerous lobe only shown; d. 7oth foot; e. 150th foot. — = dorsal capillary setae, without wings, in segments 1-4; x = bifid setae in segments 1-4; -’ =spines or setae of varying thickness, which are embedded in the tissues and seldom pierce the skin; o = brush setae; | = capillary setae, winged and smooth in anterior segments, serrate in posterior segments; y= ventral capillary setae, with flat smooth wingless blades; A = stout ventral hooks of middle and posterior segments. setae (fig. 14m) the tip is enclosed between two wings, and the shaft is obliquely striated. A short distance behind the tip there is in some cases a pseudo-articulation. The dorsal capillary seta is strong, broad, curved, and finely pointed, with smooth edges. The blade is flattened, but not winged. The 2nd and 3rd feet are similar in shape, with two dorsal capillary setae, three stout bifid setae between the anterior papillae, and two thin ones in front of the lower papilla. The posterior lobe is rather more elongate than in the rst foot. On the 4th foot is the first branchia, rather shorter than the dorsal cirrus. The rest of 614 Memorrs of the Indian Museum. [Vous Vi the foot does not differ from those in front. In the 5th foot there is a marked change. The ventral cirrus is much shorter and stouter, and the bifid setae are replaced by capillaries. The upper capillary setae lie above and behind the posterior lobe, and now show a narrow wing (fig. 14H). They are accompanied by a single brush seta (fig. 140), which has an expanded oblique tip, with about 13 coarse teeth and two external slender teeth, the latter differing very littlein length from the others: The ventral capillary setae (fig. I4J) are shorter than the dorsal, and have no wings. They taper suddenly to a very slender tip, and are obliquely striated with smooth edges. : In the 6th foot the ventral cirrus is reduced to a short stout stump and ventral pad. There are two brush setae in the dorsal division. ‘Three spines lie between the two anterior papillae. They taper rapidly at the distal end to filiform tips, which pierce the skin. In the 7th foot the ventral cirrus is still more rounded, and in the follow- ing segments it consists of a flat glandular pad, large in the anterior segments, but gradually becoming smaller towards the tail. In the roth foot (fig. 14F, and text-fig. 125) the setigerous lobe forms a rela- tively small part of the foot. The setigerous fillet runs from beneath the anterior ventral papilla round the front of the foot and is connected to the back of the poste- rior lobe. There are four stout spines in the ventral division, and a bundle of fine spines enters the base of the dorsal cirrus. There are three brush setae. The dorsal capillary setae are long, stout, tapering gradually, with narrow wings, only one of which can usually be seen. The ventral setae are shorter with much. flatter blades, _ Without wings, and taper much more rapidly to very slender tips. In the 15th foot there are 8 brush setae, in the 2oth foot 10. In the 15th foot the lower anterior papilla has disappeared (text-fig. 12c), and the fillet is attached to the ventral side of the upper.anterior papilla. In the 28th foot a stout hook appears in the ventral part of the foot, below the setae.. The tip (fig. 14N) is bifid, and the lower tooth, which is slightly bent, is stouter and longer than the upper one. The tip is guarded by two delicate flexible wings. In the 4oth foot there are 18 brush setae and 4 or 5 capillary setae in the dorsal division. There are two ventral hooks, the lower one emerging far down on the ventral side of the seti- gerous lobe. In the 7oth foot (text-fig. 12d) the branchia and ventral pad are greatly reduced in size. The posterior lobe is still conspicuous, but the anterior papilla is not visible in side-view. Of the ventral setae, only two or three project beyond the skin. They have short broad blades, with slender tips. Some of them seem to have been converted into stout spines with filiform tips. Outside the setigerous fillet there is another fold of skin, which completely surrounds the setigerous lobe. One of the stout ventral hooks emerges with the ventral setae within the fillet, the other is more ventral in position and emerges outside the fillet. There are more brush setae than capillary setae in the dorsal group. In the rooth foot the posterior lobe is greatly reduced, and the anterior papilla has disappeared. The setae form a single series, arched round the posterior lobe. 1921. | Fauna of the Chilka Lake : Polychaeta. 615 The 150th foot (fig. 146, and text-fig. 12e) is mammillate, with a tapering dorsal cirrus. The posterior lobe is small, and is almost completely surrounded by the inner fillet. There are 12 brush setae and 4 capillaries in the dorsal group. In the ventral group there are 3 stout capillaries with acutely tapering tips, and 2 very thick bifid hooks. In the latter (fig. r4P) the lower tooth is now much longer than the upper one. Towards the posterior end of the body the foot becomes smaller and more elongate, the setae fewer in number, and only one ventral hook occurs. The spines which pierce the dorsal cirrus are fewer innumber and much finer and shorter. In none of the large specimens is the anal end complete. The dorsal capillary setae show certain interesting changes as one passes from the anterior end backwards. In the 4 anterior feet they are devoid of wings. A certain number of feet follow in which the setae have narrow wings and smooth edges (fig. 14H). In the middle and posterior feet they are as shown in figs. 14K, 141. The wings are not quite opposite each other, so that in side view only one is seen. The proximal part of the wing expands suddenly, and the margin is finely but distinctly toothed. In setae from the posterior feet this serrate part is more elongate than in those from the mid-body. The ventral capillary setae also show changes. The blade becomes wider and shorter and the abruptly filiform tip more obvious in the posterior feet. The tapering part of the seta is minutely spinous. The acicula are not wider than the ventral capillaries, and approach them in shape, the filiform tip being very notable. The brush setae are not flat, but have the edges curved towards each other. The smaller specimens differ in several respects from the above description. The most obvious difference is in the branchiae. Fig. 14R shows the roth foot of one of these specimens. The branchia is many times longer than the dorsal cirrus, which is only about twice as long as the posteriorlobe. The filaments are longer, and not so crowded on the stem. In a few specimens the Ist branchia is on the 4th foot, but in the majority it is on the 5th. The anterior branchiae are relatively much longer than in the large specimens. The last branchia occurs on the 37th-3gth foot, and there are very few with only a single filament. In some specimens the filaments are long, and loosely arranged, but in others they are bushy, as in the large specimens. In the smallest of the large specimens the Ist branchia is on the 5th foot, and the filaments are long and loosely arranged, but the stem is only as long as the dorsal cirrus. In the small specimens the serrate part of the dorsal capillary setae is more elongate and not so wide asin the large specimens. The peristomium - does not project over the posterior margin of the head. Only one of the small specimens is complete. It measures 64 mm. in length, and has 150 segments. The rest are incomplete, some larger than this, and some smaller. The anal segment is button-shaped, and bears 4 ventral cirri, the upper pair being 4 times as long as the lower pair, and equal in length to the last 6 seg- ments. In one of these small specimens, one of the ventral hooks of the 3rd foot has 2 teeth beneath the main fang, but this is exceptional. In the structure of the head, jaws and tubes, the two forms agreeclosely. If these specimens had come from 616 Memoirs of the Indian Museum. NONE different localities they would probably have been regarded as distinct species. It seems very probable, however, that the distinctions observed are due to differences in size and age. The most important difference is in the relative lengths of the branchia and dorsal cirrus, and it is difficult to see how this can be a function of age and size. The large specimens were found only at one locality, in March, though empty tubes, apparently belonging to this form, were obtained in September. The small specimens were found both in March and September. The question of whether one or two species are involved can only be decided by a more ample supply of material, collect- ed at different seasons of the year. If all the specimens are regarded as belonging to the same species, then the follow- ing variations were observed :— I. In the anterior segments the dorsal cirrus may be longer than, equal to, or only one quarter as long as, the stem of the branchia. 2. The branchiae may commence on the 4th or 5th foot. 3. The branchial filaments may be long and open, or short and bushy. 4. The stem of the branchia may be thick or thin, and the base may be clearly or obscurely ringed. 5. There may be few or many branchiae pairs of (33-120), according to the size of the individual. 6. There may be few or many feet with simple branchiae. 7. Ihe peristomium may or may not project over the posterior part of the head. 8. There may be considerable variation in the number of teeth in the dental plates. These variations show that many of the criteria which have hitherto been used to discriminate the various species of this genus, are of little or no value. Crossland (1903, p. 132) has already commented on this point. In these circumstances it is probably useless to compare the present species with the majority of the species previously described, which will need careful examination, and will probably be con- siderably reduced innumber. The most characteristic features of this species are the long dorsal cirri (in the large specimens), and the shape of the dorsal serrate capillary setae in the middle and posterior segments. In the possession of long dorsal cirri, the species agrees with D. amoena, described by Kinberg (1910, p. 38), from the Atlantic, near the mouth of the Rio de la Plata. The text is too brief to be of any value, but judging from the figures recently published, D. amoena differs from the present species in having smaller, and probably fewer branchiae, with fewer fila- ments, more elongate feet and larger palps. The upper jaws also differ in shape, especially the supports and forcipate processes. The smaller specimens, however, closely resemble certain species already des- cribed, such as D. chliensis, Quat. (Ehlers 1901, p. 123), and D. leuckarti, Kinberg (1910, p. 38). The structure of the dorsal capillary setae, which apparently differen- tiates the present species, may have been overlooked by previous workers. Habitat.—Dr. Annandale writes of the large specimens of this species— “I think it is the largest worm in the collection. Specimens were taken in February or March 1921.] Fauna of the Chılka Lake : Polychaeta. 617 with their tubes, near Ganta Sila, Rambha Bay. This worm lives at the edge of the lake in muddy sand or sandy mud, at a depth of about two feet below the surface of the bottom. When the water is high, its burrows are covered by 5 or 6 feet of water, but the fore part of the tube is exposed as the level sinks in the dry weather. ‘The upper part of the tube is shaped like the ventilation funnel on a steamer, and in places where suitable molluses occur, the worm fixes a single white shell on the top of the funnel, rendering the structure very conspicuous. The specimens of the animal are very difficult to obtain, and so far as I remember, we only succeeded in digging them out on one occasion, on which the tops of the tubes were exposed.” | The tubes are large and thick-walled, formed of mud and sand on a membran- ous basis. These specimens were taken in water of specific gravity about rort. Empty tubes of similar form, though smaller, were taken in 10-12 feet of water, off Nalbano. The small specimens were taken at nine stations, usually in the dredge, in as much as I5 feet of water. Seven of these stations were south of a line joining Patsahanipur with Nalbano, one station was a little north of this line, and the remain- ing station was in the north-west corner of the lake, off Kalupara Ghat. The water was always of low salinity, the specific gravity ranging from I’000-I'010. Four stations were worked during the salt-water season, and five during the fresh- water season. j Marphysa gravelyi, sp. nov. (Plates XXIV, figs. I3A-6, and X XV, 13H-1, and text-figs. 13a-d.) Twenty-seven specimens of this species, all from the Chilka Lake, are available for study. They are of various sizes, but fall roughly into two groups, with probably an age difference of a year between them. A specimen from each of these two groups was examined, and there is no doubt but that they belong to the same species, the dis- tinctions noted being all explicable by the difference in size and age. The larger (type) specimen is 172 mm. long, composed of 360 segments. The body is cylindrical so far back as the 5th segment. It then becomes flattened dorso- ventrally. The skin is iridescent, and the dorsum shows faint white spots in front, which are probably much more distinct in living specimens. The head (fig. 134) is rounded at the sides, deeply indented in front. The inden- tation is continued as a groove, running dorsally to the base of the median tentacle, ventrally to the mouth. The median tentacle is slightly longer than, and the ex- ternal tentacle slightly shorter than, the other two. Two small black eyes lie between the bases of the lateral tentacles, hidden by the projecting collar of the peristomium (fig. 134 is from a small specimen). The head is followed by two achaetous segments, of which the first—the peristomium—is about three times as long as the next and succeeding segments. The anal segment (fig. 13B) is button-shaped, with a crenate margin. On the ventral border are two long cylindrical anal cirri, and beneath these another pair of very small stumpy cirri. 618 Memoirs of the Indian Museum. [Vor. W, The branchiae are distributed in the type specimen as follows : — No. of No. of Position of No. of No. of Position of filaments. segments. segments. filaments. segments. segments. 0 aa 130 7 38 268-305 I 6 33-38 6 31677 306-322 2 4 39-42 5 2 323-324 3 22 43-64 4 12 325-336 4 19 65-83 3 Io 337-340 5 44 84-127 2 5 347-351 6 I2 128-139 I 5 352-356 7 27 140-166 0 4 357-360 8 IOI 167-267 ' There is, of course, considerable variation in the distribution of the branchiae, and the numbers of filaments frequently differ on the two feet of the same segment. In three other specimens the first branchia appears on the 36th, 42nd, and 52nd feet res- pectively. The maximum number of filaments is 8 or rarely 9. They are much longer than the dorsal cirrus, and are crowded together on a short stem (fig. 136). The mandibles (fig. 13D) are blackish at the posterior ends, paler in front. The ringed anterior ends are small, and the two halves are not very firmly attached to each other. The maxillae (fig. 13C) are stout and boldly curved, the posterior pro- cesses being spatulate. The maxillary plates have 6 teeth on the right, 5 teeth on the left side. The azygos plate on the left has 7 ot 8 teeth, whilst the anterior crescen- tic plates have 9 teeth on the right, 6 on the left. Dark horny patches of indefinite shape occur in front of and at the side of the anterior toothed plates. The right anterior crescentic plate bears less resemblance to its corresponding left plate than it does to the azygos plate. The feet increase gradually in size up to the Ioth. The Ist foot (fig. 13E and text-fig. I3a) has stumpy rounded dorsal and ventral cirri, and a short rounded median lobe behind the setae. In front there is a thin fillet guarding the setae. The foot is very vascular. There are three black spines, lying horizontally. The setae are in two groups, above and below the spines. The dorsal group consists of about 9 capillary setae of various lengths, with flattened, very finely serrate blades (fig. 13H). Ventrally there are 7 or 8 compound setae (fig. 137). The tip of the shaft is bevelled, and bas a row of spines along the upper edge. The blade is very finely serrated, and tapers to a long filiform tip. The 2nd foot resembles the Ist, but the setae are more numerous. The 5th foot (text-fig. 136) has 4 black spines. The setigerous lobe is relatively larger and the setae much more numerous. The setae are enclosed within two fillets, which are attached by their edges to the middle lobe, and lie outside the spines. The roth foot has 5 spines, and the ventral setae are still more numerous. In the 25th foot there are 4 spines, and 2 brush setae (fig. 13K) appear, lying in front ! A few of these have g filaments. 1921.] Fauna of the Chilka Lake : Polychaeta. 619 of the dorsal setae. They have expanded oblique tips, with numerous fine teeth, of which the two outermost are much longer and stronger than the others. At the 35th foot a powerful hook (fig. 131) appears on the lower edge of the ventral setae. It is bifid at the tip, but in the anterior segments the points are either rounded or worn away. The tip is guarded by two delicate flexible wings, and the shaft is finely dotted. The 4oth foot (fig. 13F, and text-fig. 13c) has 3 or 4 spines, the upper one being colourless. The dorsal and ventral cirri are relatively smaller than in the ıst foot. In segments 50-120 there are two ventral hooks. d TEXT-FIG. 13.—Parapodial diagrams of Marphysa gravelyi, sp. nov. a. and foot; 6. 5th foot; c. 30th foot of small sp., 4oth foot of large sp.; d. 160th foot of small sp., 240th foot of large sp. ‘= capillary setae; @ = spines; x = compound setae; >= simple bifid hook. After the Iooth foot, capillary setae appear in the ventral bunch, replacing the — compound setae. A single compound seta remains in many feet, at the upper ante- rior border of the ventral bunch (text-fig. 13d), but the posterior 100 feet have no compound setae. At the 240th foot (fig. 13G, and text-fig. 13d) the branchia has its maximum of g filaments. The dorsal cirrus is short and conical, the ventral cirrus stumpy and round. The middle papilla is very small, and the internal fillet runs out to the tip of the black spine or spines. The external fillet now runs completely round the setigerous lobe, The dorsal division has 4 brush setae in front, arranged in 620 Memoirs of the Indian Museum. |VoL. V, a gradual sequence, the uppermost being the shortest and smallest. The greatest number of brush setae noted in any foot was 4. The dorsal capillary setae lie close to the inner fillet, emerging inside it in the anterior region (text-fig. 13d). In the ventral division there is a single compound setae in the front of the foot, and a bifid winged hook on the lower margin. All the other setae are simple capillaries. The ventral setae are shorter than the dorsal, and the blades are more distinctly flattened. In the posterior segments the setigerous lobe is smaller in proportion to the dorsal and ventral cirri. The setae are few and simple. There is one spine, one ventral hook, and 1-3 brush setae with somewhat coarser teeth than those in the anterior segments. | The smaller specimens differ from the above description in many respects, but these differences are due to age and size. The specimen examined was 66 mm. long, composed of 220 segments, with traces of others forming in front of the anal segment. The eyes are larger and more distinct, and are not hidden by the collar of the peri- stomium (fig. 134). The first branchia appears on the 22nd foot, and the maximum number of filaments is 5, on segments 145-165. The branchiae are absent from the -posterior 20 segments. The maximum number of spines present in any foot is 4. The first brush seta appears in the 20th foot, and the first ventral hook on the 26th foot. The capillary setae appear in the ventral division of the 8oth foot, and there are no compound setae behind the I6oth foot. Genital products were not observed in any of the specimens. | | This species belongs to the M. sanguinea group. The most remarkable character is the appearance of the capillary setae in the ventral division of the feet in the middle and posterior parts of the body. It is improbable, considering the size of the larger specimens, that this is a larval character, especially as these setae appear at relatively the same position in both large and small specimens, viz. near the begin- ning of the middle third of the body. The same proportion is also seen in the posi- tion of the last compound seta, viz. segment 150 in the small specimen, and segment 240 in the large one. In this character there is a tendency towards M. mossambica, Grube, which has no compound setae. This species has also some resemblance to M. californica, Moore (1909, p. 251), but differs in the shape of the head, relative length of the tentacles, number and dis- tribution of the branchiae, etc. M. californica has two kinds of brush setae in the middle segments, one kind with many, the other with few teeth. Moore’s speci- mens were incomplete behind, so that no comparison can be made as to the arrange- ment of the setae, a feature so characteristic for M. gravely. Dr. Annandale writes in litt. of this species :—‘‘ Another interesting Eunicid was obtained only in very small numbers, though actually abundant everywhere round the lake where the shores are muddy. This species produced large balloon-like masses of gelatinous spawn anchored to the mud by a tubulous structure and known to the Ooriyas as nowdar.’’ It is possible that this may be the same species as was found by Borradaile (1901, p. 714). In a lagoon at Jaffna, on the north coast of Ceylon, he found a species of Marphysa which liberated its eggs in a pear-shaped mass of 1921.] Fauna of the Chilka Lake : Polychaeta. 621 _ jelly, attached to the bottom by a short stalk. He has described the early develop- ment of the worm most carefully, but unfortunately all the adult specimens were lost, and the species was never definitely identified, but was said by Willey, who saw a damaged specimen, to be identical with, or closely allied to, M. teretiuscula, Schmarda, a species differing distinctly from the present form. Dr. Annandale further writes that similar spawn is a conspicuous object in shallow creeks and lagoons of brackish water both on the east coast of India and on TEXT-FIG. 13'.—Marphysa gravelyi, sp. nov. A. Egg-mass, x3. The outer layer has been removed at one spot, showing the clear jelly contain- ing the eggs. B. Larva, highly magnified. that of the Malay Peninsula. It was common at the outer edge of and in small ditches connected with the outer section of the Tale Sap on the Gulf of Siam' in January 1916, where the mass seemed as a whole more cylindrical than those in the Chilka Lake. In the Chilka Lake the masses are pear- or egg-shaped, often as much as six inches long, and are anchored to the mud, in which the worm burrows to a considerable depth, by a tube at one end. The mass consists of jelly, which is quite hyaline and colourless, and contains the eggs or larvae to the number of several hundreds. The eggs are minute and scattered. The cuter part of the gelatinous mass is stiffer than the central part, to which the eggs are confined, and the whole is 1 Concerning this lake see Mem. Asiatic Soc. Bengal, VI, pp. 3-6, text-fig. 3 (map). 622 Memoirvs of the Indian Museum. VOLANT covered by a thin, soft, sticky layer, transparent in itself but soon powdered with fine silt, which adheres to it firmly. This soft external layer and also the outer, stiffer layer of the central mass are continued at the base to form a narrow tube, which is usually filled with the excreta of the worm. The tube goes straight down into the mud for a foot or more and is inhabited by the parent worm. In the photograph of an egg-mass of M. gravelyi here reproduced the basal tube is well shown; also the. thin externallayer, which is torn at one place, revealing the egg-mass with the minute contained eggs or larvae. Dr. Gravely adds that the unhatched larvae (text-fig. ı 3'b) can move freely in the jelly in which they are embedded. ‘They closely resemble Borradaile’s larvae, but minute examination revealed somewhat great- er differentiation among the cilia than is indicated in his figures. Two groups of setae crushed out of a larva with two seti- gerous segments are shown in text-fig. 13°. It will be observed that of the simple setae in the anterior group one is pointed at the end and the other somewhat blunter, as described by Borradaile. ‘There is, moreover, a single jointed seta associated with them. Neither this jointed seta nor the one in the second group is, however, quite like either of the TEXT-FIG. 13”.—Parapodium of larva of Mar. . . EN physa gravelyi, sp. nov., highly magnified, Jointed setae figured from -Borradaile's older larva. The figures of the Chilka larva have been prepared under Dr. Gravely’s supervision. A minute nematode worm Monhystera uria, Stewart oe Ind. wee xp. 247) is found in the jelly with the larvae. Habitat.—27 specimens of this species were taken at 7 stations in Chilka Lake. All the stations were in the southern end of the lake, from Nalbano to Rambha, and the bottom was muddy. The species was apparently only taken during the salt- water season, from January to March. In this part of the lake, however, the specific gravity is only from I'008 to 1‘0115 in the salt-water season. It is certain that the species lives in this area throughout the year, when the gravity is as low as I’00I, and is only obtained in January to March because, at that time, the level of the water has sunk so low that the mud flats and banks, in which the worm lives, can be examined. Lumbriconereis polydesma, sp. nov. (Plate XXVI, figs. I5A-L, and text-figs. I4a-c.) Only one specimen, fortunately quite complete, and a fragment of a second of this species were taken in the south-western end of Chilka Lake. It is 185 mm. in length, and is composed of 386 segments. It is not fully grown, as genital products are absent, and new segments are in process of formation in front of the anal segment. TO2L.] Fauna of the Chilka Lake : Polychaeta. 623 The body is very slender, the greatest width being 1°0-1:2 mm., not including the feet. It tapers very gradually towards the tail, more rapidly, but to a lesser degree towards the head. The skin ofthe preserved specimen is opaque white in colour, with a faint green iridescence. The head (figs. I5A, 15B) is rounded in front and at the sides. Dorsally the posterior border is covered by the projecting front border of the peristomium. Noeyes were observed, though there is a small patch of pigment near the posterior border of the head, a little on the right of the median line. The mouth is bounded in front and at the sides by two large lips formed by the infold- a ‘ing of the lateral posterior mar- TEXT-FIG. 14.—Parapodial diagrams of Lumbriconereis à polydesma, sp. nov. gins of the head. Between the a. Ist foot; b. roth foot; c. 80th foot. head and Ist setigerous segment - = Simple capillary setae; x = crochets. ate two achaetous rings, the first being twice as broad as the second. Ventrally the 1st ring is incomplete, being interrupted by a projecting lobe of the 2nd ring. Both rings are furrowed on the ventral side. The feet increase gradually, and only very slightly, as far as the 10th. The rst foot (text-fig. 14a) has a prominent rounded ventral lobe. The setae lie above and in front of this lobe, and are guarded externally by a conspicuous curved fillet. There are two colourless spines. The setae are stout winged capillaries, with bilimbate obliquely striated blades. There are 3 of them above the spines and 4 below. In the roth foot (fig. 15F and text-fig. 14b) the fillet forms a hood over the ventral lobe. It is attached to the middle of it in front, then curves over it, and is attached to it again on the lower posterior border. There are 3 spines in a horizontal row. They are colourless, except the tips of the two anterior ones, which are dark brown. There are 4 capillary setae above the spines and 5 below. The upper setae are rather longer than the lower, and they all curve towards the spines. At the 29th foot the first crochet appears in the ventral division. The 30th foot has 2 colourless spines, 3 capillary setae above the spines, 2 capillary setae and 1 crochet below the spines. ‘The fillet lies more in front of the ventral lobe than in the roth foot, and its anterior line of attachment is near the ventral border of the lobe. The 40th foot is similar in shape. It has 2 spines, 2 capillary setae and I crochet above the spines, I capillary seta and 2 crochets below. In the 80th foot (fig. 15¢, and text-fig. I4c) the ventral lobe is smaller and slenderer, and curves upwards. ‘The fillet lies almost altogether on the anterior side of the lobe, and is hardly visible from the posterior aspect. There are 2 spines, I capillary seta and 1 crochet above the spines, 3 crochets below. The shape of the feet is very uniform in the middle and posterior parts of the body. At the 300th foot (fig. 15H) the ventral lobe is rather longer and thinner. : 624 Memoirs of the Indian Museum. [Vor. V, The fillet is attached near the base of the lobe, and projects outwards, in front of it. There is a single colourless spine, one capillary seta and one crochet above the spine, and one crochet below. This capillary seta persists in all the feet. The anus is terminal. On its ventral margin are the remains of 4 short stumpy cirri, the upper pair being the longer. The capillary setae (figs. 15J, I5L) have broad wings which tend to curve inwards, so that frequently only one is visible. They are obliquely striated, and taper to a slender tip. In the posterior feet the wings are broader than those in the anterior feet. The crochets (fig. 15K) have each two broad wings over the tip, with delicate oblique striations. The boldly curved terminal tooth has a crown of 6-10 slender spines. The mandibles (fig. 15E) are fused throughout their whole length. The biting plate is composed of 5 or 6 half-rings. The shafts are striated longitudinally and also obliquely. The whole structure is very delicate and the edges are so thin that it is impossible to indicate the precise shape. The forcipate processes of the maxillae are normal in size and shape, and are equal in length to the posterior processes or “supports”, which have a distinct waist. Lving alongside the forcipate processes are two narrow dotted strips of chitin. The main dental plates are each composed of a dark-coloured ventral part, rather narrow and bearing teeth, and a larger and paler reflexed plate lying above it. There are 5 teeth on the right side, 4 on the left. The 3rd pair of jaws have bifid tips, and the 4th pair strong simple tips. Fig. 15D shows them after they have been flattened. Lying alongside the 3rd and 4th pairs are two rectangular dotted plates. This species is characterised by the very slender elongate body, the structure of the jaws, the shape of the head and setae, and the presence of capillary setae in all dns IE. In the shape of the jaws it resembles the L. Zetraura (Schmarda) as described by Ehlers (1901, p. 137), but the two species differ widely in the distribution of the capil- lary setae and crochets. Inu the form of the setae, and in having two teeth on the 3rd pair of jaws, it resembles L. indica, Kinberg (1910, p. 48), from the Bangka Straits, but differs in having only a single tooth on the 4th pair of jaws, and in the presence of capillary setae in the posterior segments, etc. This species rather closely resembles the L. heteropoda of Marenzeller (1879, p. 30), from Japan. The latter differs in having a longer and more pointed head, shorter and broader body, with fewer seg- ments, feet of somewhat different shape, and in the relative length of the two lobes. In the middle of the body the foot of L. heteropoda carries a rudimentary dorsal lobe with a few simple setae embedded in the tissues. The capillary setae are distributed as in the present species, and occur in the posterior feet. ‘The jaws are similar as a whole, but differin details, especially in the shape of the maxillae and mandibles. The tips of the crochets also differ. | Habitat. —A single entire individual and a fragment of a second were obtained by digging in sand just above high-water mark. on the shore of Chiriya Island, in the south-western extremity of the Chilka Lake. They were collected during the salt- water season, in February, and the salinity of the adjacent water was I’OII. 1921.] Fauna of the Chilka Lake : Polychaeta. œ D Ou Lumbriconereis simplex, sp. nov. (Plate X XVI, figs. I6A-I6M, and text-figs. 15a—b.) Six specimens of this species, from the Chilka Lake, all from the south-western end, are available for description, all of them in a fragmentary condition, and the posterior end is missing in all cases. The largest fragment is 32 mm. long, and con- sists of 125 segments. It is probably half the total length, or rather more. The preserved animals are colourless, except for a delicate greenish iridescence. There is a dark amber-coloured spot near the base of each foot, on the ventral side, composed of granular pigment, like dried blood. Similar granules are scattered over each foot, especially near the tip of the setigerous lobe, and just above the foot. The body expands gradually up to the 15th setigerous segment, and then tapers still more gradually towards the posterior end. The feet are rather prominent. The width of the body at the 15th foot is 17 mm. or including the feet and cirri 27 mm. The Ist achaetous ring is 24 times as long as the 2nd ring. The head (figs. 164, 168) is triangular, with rounded angles, the length slightly exceeding the width. No trace of eyes could be found, though there is some subcutane- ous pigment on the head. The mouth is bounded in front by two tumid lips belong- ing to the head, laterally by the first ring, > and in the mid-ventral region by the second ring, which interrupts the first ring and pro- jects forwards to the mouth. a 8: The ıst foot (fig. 16G, and text-fig. I5a) Texr-Fic. 15.—Parapodial diagrams of Lum- consists of a setigerous lobe composed of D a..ist foot; 5b. 1oth foot. two lamellae or fillets, and a rounded lobe De ne eue lying behind and beneath the fillets. : Be- tween the fillets lie four colourless spines. The setae are in two groups, a dorsal bundle of two long bilimbate setae above the spines, and four similar but shorter setae beneath the spines. The upper ventral seta emerges inside the inner fillet, all the others between the two fillets. The succeeding feet increase in size and in the number of setae. In the roth foot (fig. 16H, and text-fig. 155) the posterior lobe is longer, bluntly pointed, and flattened, and the two fillets lie in front of it. There are four bilimbate setae in each bundle. The succeeding feet up to the 30th show little change. The base of the posterior lobe then becomes gradually fused to the fillets, and the terminal portion projects backwards as a finger-shaped lobe (fig. 16K). The number of spines in the 30th—5oth feet is 3, in the rooth foot there are 2. From the 30th foot there are only 3 setae in each foot, one dorsal and 2 ventral. In the anterior segments the setae (fig. 161) are long and tapering, with smooth flattened obliquely striated blades. They are bilimbate, but the two wings are not directly opposite each other, so that usually only one is seen. The dorsal setae are longer than the ventral. In the middle and posterior feet the setae (fig. 16m) have wider and 626 Memoirs of the Indian Museum. Do. flatter blades, and the wings are very narrow or quite absent. In a fragment, evi- dently from near the posterior end of a specimen, the feet are more translucent and more elongate. They have I or 2 spines, and I dorsal and 2 ventral setae. The blades of the setae are short, broad, and flat, with very narrow wings, and they taper rapidly to a long fine tip. The setigerous lobe is more conical at the tip, and the posterior lobe more elongate. The feet of this species are remarkable for the richness of their blood supply, and they evidently function as branchiae. In many of the feet there is in the dorsal region of the base a large heart-shaped structure full of blood (fig. 167), into which a number of vessels open. It is not present in all the feet, and is apparently formed by the swollen junction of the blood-vessels. When present, it forms a conspicuous spot on the dorsal side of the foot. From this ‘heart’ a large vessel runs into the foot, giving off numerous branches which lie under the cuticle on the front side of the foot. A large branch runs round the margin of the setigerous lobe, and another passes into the posterior lobe. All these vessels branch repeatedly, and the capillaries finally unite to form large vessels lying on the posterior side of the foot. In the fragment from the posterior end of the body, mentioned above, the blood-vessels are greatly reduced in size and number. The mandibles (fig. 16F) are translucent, broad, and fused throughout almost the whole length. The anterior or biting end is composed of numerous semi-rings. There is very little pigment, which is confined to the lines of growth. The maxillae (fig. 16c) are stout and boldly curved. ‘Their posterior portions (the ‘supports,’ or ‘carriers’) are laterally deeply indented. This part varies greatly in length, as is seen by comparing figs. I6C and 16D, but is always shorter than the forcipate processes The posterior oblique margin is thin and frayed. Lying on the outer margin of the forcipate processes is a long thin horny plate. The great dental plates have each 4 large teeth. At the anterior end each has a flat cap (fig. 168). The dental plate itself is rather narrow, but it has a broad dorsal flange lying behind the forcipate process. The 3rd pair of jaws are bidentate. The 4th pair have each a stout tooth, which may be slightly bifid at the tip, possible through abrasion. Lying ventrally to the 3rd and 4th pairs of jaws is a thin rectangular coarsely dotted plate on each side. So far as one can judge from the brief description and inadequate figures, this species resembles L. atlantica, Kinberg (1910, p. 47), from the mouth of the Rio de la Plata, especially in having only slender capillary setae. It appears to differ in the shape of the head, and in having two teeth on the 3rd pair of jaws. - Habitat.—Fragments of six specimens were taken at two stations, both of them on the shore of the south-western extremity of the lake, at Rambha, in mud. Of one station, no further information is available. The other station was in March, and the specific gravity of the water was I’OII. TO2T.| Fauma of the Chilka Lake : Polychaeta. 627 Family GLYCERIDAE. Glycera alba, Rathke, var. cochinensis, var. nov. (Plate XX VII, figs. 17A-J, and text-figs. 16a-e.) Two specimens of this species were collected in the Cochin Backwater, near Ernakulam. One of them is in good condition for description, the other is softened and lacks the posterior region. This species closely resembles the European Glycera alba in appearance. The preserved animals are buff coloured. The type specimen is 39 mm. long, and consists of 120 setigerous segments. The body attains its greatest width between the 30th and 40th segments. From this region it tapers abruptly towards the head, and only very gradually towards the posterior end. On approaching the tail, however, it decreases rapidly in size. The segments are biannulate, the two rings being almost equal in length, or the ring bearing the feet is very slightly the longer of the two. The mouth (fig. 17A) is bounded laterally by two large tumid lips, formed of two rings. In front of it are Io rings which gradually taper towards the tip. The terminal ring bears four small tentacles. The large segment which forms the posterior mar- gin of the mouth is much folded ventrally, and bears a fairly large pair of parapodia. In front of this segment are two others which are very narrow and incomplete ven- trally. Each of them bears a pair of rudimentary parapodia, the anterior pair espe- cially, being very small. The proboscis in both specimens is completely retracted, and the jaws lie opposite the 21st pair of parapodia. The posterior part of the retracted proboscis is shown in fig.17B. The expanded part—the gizzard—containing the jaws, is almost square in section, the rounded angles being formed by four large glands, whilst there is a smaller gland on each face. Attached to the front margin of the gizzard are four flat, almost rectangular lamellae, two directed forwards and two folded backwards. They are composed of large rounded cells with large nuclei. The jaws (fig. 17E) are stout and curved. The fixing process is slender, and is attached near the middle of the jaw. It bearsa thin triangular wing on the basal part. In the specimen figured, the posterior end of the jaw is bifid, but this is probably due to fracture, as it is entire in other Jaws which were examined. The internal surface of the retracted proboscis is covered with small papillae of various shapes. The most numerous variety has a cylindrical stem (fig. 17C), with an oblique mammillate tip, bounded by flat wings. The diameter of the stem is equal to the width of the wings, and the papilla is 2} times as high as the diameter. Another variety is pear-shaped (fig. 17D), whilst others are similar in shape, but slenderer. All are traversed by ducts opening at the tip. The 1st foot is very minute, and consists only of a setigerous lobe with a group of compound setae, and a flat heart-shaped posterior lobe. The 2nd foot is a little larger. The setigerous lobe has two bundles of compound setae, an anterior finger-shaped lobe, and a posterior cordate lobe. In the 3rd foot, which is almost normal in size, there is a rounded dorsal cirrus, 628 Memoirs of the Indian Museum. Mer and a group of dorsal capillary setae. There are two long slender anterior lobes. The posterior cordate lobe has now moved towards the ventral border, and there isno posterior lobe. There are two spines. In the 4th foot (fig. 17F, and text-fig. 16a), the posterior cordate lobe of the Ist and 2nd feet has now definitely assumed its position as the ventral cirrus. The fillet, which lies behind the setae, is attached to the two anterior lobes in the region of the spines, and is thus broken into three sections. The dorsal capillary setae are in two bundles, above and below the dorsal spine, whilst the compound setae have a similar relation to the ventral spine. The 6th and 8th feet (text-fig. 160) are similar, except that the fillet runs further out along the upper anterior lobe. It then passes direct to the ventral cirrus, and is not fused to the lower anterior lobe. TEXT-FIG. 16.—Parapodial diagrams of Glycera alba var. cochinensis, var. nov. a. 4th foot; b. 6th or 8th foot; c. 10th foot; d. 4oth foot; e. r1oth foot. * = capillary setae; x = compound setae. In the roth foot (text-fig. 16c) the dorsal projection of the fillet now forms a posterior lobe, and is quite free from the upper anterior lobe. ‘In the 12th foot the posterior lobe is half as long as the upper anterior lobe. The fillet projects slightly outwards behind tke lower anterior lobe. The first branchia appears on the 17th foot, as a short stumpy lobe above the posterior dorsal lobe. It increases in size up to the 30th foot. The 40th foot (fig. 17H, and text-fig. 16d) represents the normal condition. The branchia is very large, 3 or 4 times as long as any of the foot lobes. The upper posterior lobe is always much shorter than the anterior lobes, whilst the lower poste- rior lobe is only just indicated by the outward curve of the fillet. Towards the posterior end the various lobes (excepting the dorsal cirrus, which remains short and rounded) become thinner and more pointed. The rroth foot (text- 1921.] Fauna of the Chilka Lake : Polychaeta. 629 fig. 16e) consists only of the dorsal cirrus, the two anterior lobes, and the ventral citrus. The branchia and posterior lobes have disappeared, and the fillet is perpendi- cular. The posterior 12 feet are without branchiae. The anal segment bears a pair of long tapering cirri, stout at the base. The setae are of the usual type. The dorsal capillary setae have slightly flattened blades, with slight traces of a wing. The compound setae have the ends of the shafts unequally bifid (fig. 17J), and slender terminal pieces. The edges of the latter are minutely crenulate, rather than serrate. I have compared these two individuals carefully with specimens of Glycera alba, Rathke, from Ireland, and have been unable to find any differences worthy of specific rank. In the Cochin specimens the branchiae are a little longer, and the lobes of the feet rather more acute, the posterior lobes being shorter relatively. Itis also possible that examination of more material would show differences in the struc- ture of the anterior feet, especially in the origin of the posterior lobes. The structure of the proboscis is very similar in both forms. Glycera africana, Arwidsson (1808, p. 21) and Fauvel (1902, p. 75), seems to be a similar variety of G. alba, hardly distinguishable from the present form. It has been recorded from Senegal, Cape of Good Hope, and the Red Sea. The branchiae of the European forms of G. alba vary considerably in size, G. convoluta, Keferstein, having apparently been described from specimens with very large branchiae. G. alba, var. macrobranchia, Moore (IQII, p. 301), from San Diego Bay, California, is very closely related to the present form. A number of other species will probably eventually be referred to G. alba, which appears to have an almost world-wide distribution. It has not, however, previously been found in fresh or brackish water, though its wide dis- tribution indicates great capacity for adaptation. Habitat.—-Two specimens were collected in September, 1914, by F. H. Gravely, . in the Cochin Backwater, near Ernakulam, south-west coast of the Madras Presidency. The water is of variable salinity, but precise information on this point is lacking. Glycinde oligodon, sp. nov. (Plate X XVII, figs. 18A-0, and text-figs. 17a-c.) This species is widely distributed in the Chilka Lake. The type specimen is an immature female, 20 mm. long, having 97 setigerous segments. The colour of the body is a dark greenish yellow. . In the intersegmental areas of the lateral region the pig- ment is much darker, except between the parapodia. In the median ventral line there is a row of dark spots—the neural eye-spots—not very distinct in the anterior segments. Each spot is formed, sometimes by a small bar on each side of the in- tersegmental groove, but more often by two short parallel longitudinal bars crossing the groove. The body is narrow in front, and gradually expands till the beginning of the posterior third. It then tapers a little, more rapidly in the last few segments, but still the tail terminates rather bluntly, especially in small specimens (fig. 188). The anal segment bears two long cirri, swollen at the base, with long filiform tips. The 630 Memoirs of the Indian Museum. IVOEAVE anus is terminal, and the cirri are ventral, equal in length to the last 9 or 10 seg- ments. | The segments increase in length up to the 22nd, where the length of each is two- thirds of the width, then slightly decrease up to the 28th. Then they suddenly de- crease greatly till the length is only ¢th—{th of the width, the latter having at the same time increased. The anterior part of the body is rounded, the middle and posterior regions are flat. Inthe anterior segments the nerve cords are widely sepa- rated (fig. 18a), and are indicated externally by faint lines of pigment, which are thickened to form a dot in the middle of each segment. _ The basal ring of the head (fig. 18a) is broad, and has a small deeply seated black eye at each side. No other eyes were observed on the head. The distal ring of the head has four small tentacles, each having a small terminal joint furnished with palpocils. The ventral tentacles are a little in front of the dorsal. The head is com- posed of the basal ocular segment and eight tapering rings. The dorsal region of the head is defined by lateral grooves. The everted proboscis is armed with two dorsal and two ventral bands of tooth- like papillae or paragnaths (fig. 18C). Each ventral band consists of two rows of papillae. The inner or median row is composed of soft mammillate papillae (fig. 185), each with a pore at the tip. The outer row iscomposed of harder thick-walled parag- naths (fig. 18F), each with one large and one small tooth. These outer paragnaths are largest at the proximal end of the everted proboscis, and become smaller and more shapeless towards the distal end. Each dorsal band consists of four irregular tows of paragnaths. Between the bands, in the median dorsal line (fig. 18c), is a row of small rounded horny papillae with transverse apical pores. The paragnaths composing the inner row of each dorsal band are distinguished clearly from the . others by their larger size and peculiar shape (figs. 18C and 18p). Each is attached near the middle, and the outer end is expanded to form a knob. ‘The inner end is poiuted, and a short distance from the tip on the upper side is a round pore. Lying externally to this row, are three very irregular rows of paragnaths, those in the outer row being usually the smallest. They have simple bases, and the tips may be simple or bifid, with terminal pores. The paragnaths, in addition to being in longitudinal bands and rows, are also arranged in transverse rows, each row being completed in the lateral region of the proboscis by aridge. On this ridge, on each side, are two flat papillae with pointed angles (figs. 18c, 186). These ridges and papillae are absent on the basal part of the proboscis, largest in the middle and distal regions. The jaws (fig. 18H) are disposed in a single row, and are very few in number. There are two large ventral jaws (a) each with two or three large blunt roots and six teeth. There are no small jaws between the large pair. There is a dorsal row of four small denticles, each consisting of several rounded lobes (b), sometimes with two sharp teeth (c), or with four teeth and a larger bifid tooth (d) occasionally there is an additional denticle, still smaller and simpler. The rst foot is very small, the 2nd a little larger, and the 3rd almost normal in size. 1921.| Fauna of the Chilka Lake : Polychaeta. 631 The ist foot (fig. 18], and text-fig. 17a) consists of three lobes, one in front of the setae and two behind. The small spine, with bent tip, lies behind the front lobe. The setae are all compound, and lie in two groups, 4 beneath the upper lobe, and 3-4 between the anterior and lower lobes. The 2nd foot is similar, and there is an additional seta in each group. The 3rd foot has 9 dorsal and 4 ventral setae. In the 6th foot the anterior lobe is much flatter and wider, and a small lobe appears behind it. The two posterior lobes have moved further apart, and now obviously represent the dorsal and ventral cirri. At the roth foot (fig. 18K, and text-fig. 17) all the lobes are flatter. ‘The dorsal cirrus has taken up a more elevated position. It has a rounded ventral projection near the base, and is also indented near the tip, the latter characteristic being very = xx * x x ue N f\ * % a + ¥ TEXT-FIG. 17.—Parapodial diagrams of Glycinde oligodon, sp. nov. a. Ist foot; b. roth foot; c. 30th foot. ‘= simple dorsal setae; x = compound ventral setae. % x x % x Ey See constant in all the subsequentfeet. Theanterior lobe is very much enlarged, flat and tapering to a slender tip. There are 8 setae above and 7 below thespine. The 15th foot is very similar, the dorsal spine and setae being still absent. In the 20th foot the dorsal cirrus is still more obviously enlarged near the base, which projects out- wards as two small papillae, between which terminate the dorsal spine and one or two slender spines. At the 30th foot (fig. 181, and text-fig. 17c) the dorsal division consists of the short stout swollen base of the dorsal cirrus, the latter organ, with its indented tip, being carried on the upper surface. Beneath the dorsal cirrus lies the dark reddish brown spine; accompanied by two or three dark brown setae (fig. 18N), which do not ‘ pierce the skin. In front of the spine is a small rounded papilla. The ventral divi- 632 Memoirvs of the Indian Museum. ; Dir, sion shows little change, except that the posterior lobe is rather larger and wider. In some specimens, however, this lobe is narrower and longer than that shown in fig. 181. The succeeding feet undergo little change, except that the various lobes gradually become shorter, and the dorsal division relatively increases in size, till at the goth foot (fig. 18m) it slightly exceeds the ventral division. | _ The ventral compound setae are of the usual type (fig. 18P, 180). The end of the shaft is swollen, and the terminal piece is long and slender, with very minute serrations. The dorsal setae (fig. r8N) are of the shape characteristic for the genus Glycinde, having a curved tip with a long slender curved spine on the crest. In none of the feet do they pierce the skin. This species is characterised (1) by the small number of rings composing the head; (2) by the structure of the proboscis, and especially by the small number of dorsal jaws; (3) by the structure of the feet. G. oligodon seems to be most closely related to the G. armigera described by Moore (1911, p. 307), from California, in rather deep water. It differs in the shape of the paragnaths, the smaller number of dorsal jaws, the smaller number of rings composing the head, and in the shape of the foot-lobes. Habitat.—Numerous individuals of this species were taken at 12 stations in the Chilka Lake. With one exception these stations were in the south-western half of the lake. The remaining station was in the outer channel, and was worked during September, when the water was quite fresh. At the other stations, 9 of which were worked in the salt-water season, and 2 in the freshwater season, the specific gravity of the water varied from L’00I-I’0II. ‘The species was usually taken on a muddy bottom, at some distance from the shore. Family ARICIIDAE. Scoloplos marsupialis, sp. nov. (Plate XXVII, figs. 194-G, and text-figs. 184, b.) Only a single specimen of this species was obtained, in the Chilka Lake, but fortu- nately it is complete. It is a male, and the body cavity is full of nearly ripe sperma- tozoa. The body is wide and flattened in front, attaining its greatest width at the 13th setigerous segment. It is 50 mm. long, and is composed of 210 segments. The ventral surface is flattened in front for a short distance, but soon becomes markedly convex, and is traversed by a median groove which commences in the 15th segment, and runs to the posterior end. ‘The dorsal surface is convex up to the 5th setigerous segment, and then becomes flat, but the crowding of the parapodia on the dorso- lateral borders makes it appear concave. In the anterior region there is a transverse ridge running round the middle of each segment, dorsally and ventrally. In the middle and posterior regions, each segment is biannulate. The prostomium (fig. 194) is composed of two rings. The anterior ring is a slender cone, separated by a groove from the posterior ring, which is much wider. The mouth is at the anterior end of the buccal segment, and has two lateral semicircular lips, with a longitudinal sn 3 al Sl 1921.] Fauna of the Chilka Lake : Polychaeta. 633 slit—the mouth— between them. These lips may possibly represent the proboscis as it begins to be extruded. No eyes were observed. The intestine is full of sand grains. The rst foot (text-fig. 18a) has dorsally a nu- merous group of capillary setae, with serrated blades, diminishing in size from above down- wards. They are guarded in front by a low in- significant fillet, behind by a rather more conspi- cuous one. The ventral bundle consists of three or four rows of stout hooks (fig. 19F). Above and below the hooks, and posterior to them are two groups of capillary setae like those in the dorsal bundle. The whole ventral bundle is sur- rounded by a low and incoaspicuous fillet. The and foot is similar. In the 3rd foot the dorsal bundle is relatively a little smaller, and has a small flat cirrus behind it. No cirrus was noted in the ventral division. In the 4th foot (fig. 19B) there is a conical cir- rus behind the middle of the ventral division, as well as the flat finger-shaped dorsal cirrus. a b The short ventral hooks are only present in Texr-ric. 18.—Parapodial diagrams of numbers in the anterior 8 setigerous segments. Scoloplos marsuptalis, sp. nov. In the 9th segment only 2 were observed, and in He De sath fear Ê ‘= capillary setae; x = short hooks. the roth, none. This type of seta has a number of coarse serrations beneath the rounded tip, which is guarded by two very delicate wings. The shaft shows a double curvature. In the anterior segments the ventral capillary setae are shorter and stouter than the dorsal setae, and the blade is more abruptly widened. No further change occurs in the feet till the 13th setigerous segment, where the branchiae appear as minute papillae, and the dorsal cirrus commences to elongate. At the 17th foot (fig. 19¢) the dorsal cirrus and branchia are very similar in size and shape, but the ventral cirrus is still small. In the 18th and roth feet (fig. 19D) marked changes commence. The dorsal cirrus and dorsal setae are much longer. The ventral cirrus is longer also, and behind and beneath it is a large thin membrane shaped like a pocket. The latter commences on the 18th segment, and increases a little in size in the succeeding segments. All the setae are now longer, slenderer, and with finer serrations than those in the anterior region, the dorsal setae being longer than the ventral. Each bundle is supported by a number of pale spines. The ven- tral cirrus, however, still remains entire, and the area from which the ventral setae emerge is still elongate, as in the preceding feet. In the 20th foot the metamorphosis is complete. The ventral cirrus is bifid at the tip, and the setae emerge in a small area in front of and between the two lobes. The dorsal cirrus is constricted near the base, The ventral setae have also changed 634 Memoirs of the Indian Museum. [VoL. V, again slightly, the blade, instead of tapering very gradually, narrowing abruptly near the tip (fig. 196). The branchiae have gradually increased, and the 30th foot may be taken as typical of the median and posterior regions of the body (fig. IgE, and text-fig. 18d). | The branchiae are large and tapering and covered with cilia. They lie close to- gether near the median dorsal line. The dorsal cirrus is fusiform, and a thin fillet at its base lies behind the dorsal setae. The ventral setigerous lobe is also constricted at the base and bilobate at the tip. The ventral setae emerge in a bunch just in front of the bifid tip, and are guarded by a low anterior fillet. Beneath the ventral divi- sion is the delicate pocket-shaped membrane, from which the specific name is derived. It is attached by its anterior edge in a line with the setigerous lobes, and its shorter posterior margin is thus behind the foot. The setae are almost equally long in both divisions. Very little change takes place in the succeeding feet. In the 7oth foot the dorsal cirrus is half as long again as the branchia, which is very short and stout, but in the 150th foot the two are equal in length. As only a single specimen was available it is impossible to say whether these variations have any significance. The posterior 10 pairs of feet decrease rapidly in size. The anal segment is large and cylindrical and the anus is terminal. There are two very slender dorso-lateral anal cirri. This species differs from S. armiger (Müller) in many respects. In the latter species the pockets below the feet are absent. The position of the first branchia differs, as does the shape and arrangement of the anterior feet, the shape of the sétae ete: | S. kerguelensis, McIntosh, is easily distinguished from the present species by the shape of the setae, the position of the first branchiae, etc. Habitat—A single specimen was found in the outer channel at Manikpatna Island, during the salt-water season, when the specific gravity of the water was about 1'0264. It probably lives there throughout the year, though this remains doubtful, in the absence of material collected during the freshwater season. Family SPIONIDAE. Polydora hornelli, Willey. (Plate XXVIII, figs. 2IA-D.) 1905. Polydora hornelli, Willey, p. 286. A single specimen, the posterior part of which is missing, was found in an oyster shell at Manikpatna, Chilka Lake. It exhibits no obvious characters by which it can be distinguished from the P. hornelli described by Willey (tom. cit.), who obtained his specimens from pearl oysters in the Gulf of Manaar, Ceylon. Unfortunately Willey’s description is rather brief, and is illustrated only by a single figure showing one of the modified setae of the 5th segment. He gives no account of the structure of the posterior region, so important in diagnosing the species of this genus, and the absence of the tail in the Chilka specimen prevents me from completing his account. 1921.] Fauna of the Chilka Lake : Polychaeta. 635 The single incomplete individual is 31 mm. long, consisting of 92 segments. After preservation in alcohol, it is quite colourless. The head (fig. 21A) corresponds fairly well with the description given by Willey. The prostomium is slightly indented in front. Between the bases of the tentacles there is an oval area of the prostomium, separated by grooves, which seems to corre- spond to the ‘ ocular area’ of Willey. The prostomial ridge, or caruncle, runs back to the end of the 3rd setigerous segment. Noeyes were observed. The tentacles are stout, long and firmly attached. The Ist foot has a bundle of ventral per setae, with a cirrus (the ventral lamella) behind them. There are no dorsal setae, but just outside the base of the tentacles on each side is a small conical lobe, representing the dorsal lamella. The dorsal bundles of the succeeding feet have two rows of setae, the anterior row consisting of short setae, the posterior row of long setae. The 5th setigerous segment (fig. 21B) bears the modified setae. In the middle of the lateral region is a curved row of 8 or 9 stout hooks, alternating with as many pointed setae. At the anterior end of the row is a bunch of capillary setae, the dorsal superior setae. According to Willey (tom. cit.) these setae are absent in his speci- mens. There is no trace of dorsal or ventral lamellae. Beneath the middle of the modified setae is a tuft of capillary setae, representing the ventral bundle. The tips of the hooks differ according to their position in the row. The oldest hooks, at the anterior end of the row, are worn at the tip (fig. 2IC, a, a‘), which is spoon-shaped, with a distinct excavation, and a neck below. The unworn posterior hooks (fig. 21C, b) have a closely applied tooth, which occupies the concavity of the tip. The latter type is figured by Willey (1905, Pl. V, fig. 117). The capillary setae which alternate with the hooks, and which represent the anterior row of dorsal inferior setae, have rather short flattened blades and slender tips, and the dorsal superior setae resemble them. 5 The branchiae and ventrai crochets appear on the 7th setigerous segment. The branchiae continue to the end of the imperfect worm (the 92nd setigerous segment at least).. The crochets (fig. 21D) have a distinct constriction beneath the neck. The main tooth is long and pointed, making an angle somewhat less than 90° with the shaft. There is a simple tooth on the crest, making an angle of 45° with the main tooth. This specimen is very close to the P. hornelli of Willey. The chief differences are (1) the caruncle extends over the 3 anterior setigerous segments, whereas Willey gives 2 segments for his specimens ; (2) the dorsal superior group of setae of the 5th setigerous segment is present in the Chilka specimen, absent in the Ceylon specimens. It is possible that a closer examination of the Ceylon specimens would show other differ- ences, which would necessitate the creation of a new species for the Chilka form, but meanwhile it may be regarded as a variety of Willey’s species. There is no doubt but that this species is distinct from P. ciliata (Johnston). Habitat.—In crevices of oyster shells, Manikpatna, Chilka Lake. The specimen was collected in September, when the water was quite fresh. From the nature of its habitat it is highiy probable that it survives through the salt-water season, 636 Memuirs of the Indian Museum. Ver V,, Polydora (Carazzia) kempi, sp. nov. (Plate XXVIII, figs. 204-J.) Five specimens of this species were collected from a canal near Calcutta. Un- fortunately the specimens are all incomplete, and nothing is known of the posterior end. The longest fragment is 6°5 mm. long, composed of 27 setigerous segments. The body is of the usual shape, the anterior region being flattened. dorsally, some- what rounded ventrally. Near the 17th segment the body becomes round in section. The anterior segments are short, but gradually elongate, till at the 2oth they are nearly as long as broad. The anterior segments have each a single transverse line of black spots on the dorsum, but the line is sometimes reduced to two median spots. The head (fig. 20a) is rather small and broad. It bears in front two short but distinct anterior lobes. There are four black eyes, the two posterior being small and round, whilst the anterior pair, which are more widely separated than the others, are crescentic. The posterior margin of the head is rounded, and there is no caruncular prolongation, but instead there is a large erect occipital tentacle, lying between the bases of the tentacular cirri. The latter organs are relatively large. The Ist foot (fig. 20C) consists of a stumpy round dorsal papilla, without dorsal setae. The ventral setigerous lobe is vertically elongate, with a row of slender capil- lary setae, which have only slightly flattened blades. The 2nd foot has a fairly large rounded dorsal lamella lying behind the dorsal setae. The dorsal setigerous lobe is conical, and the setae are long and slender, with only slight indications of a wing. The ventral setae are shorter and wider than the dorsal, with narrow wings. In the 3rd foot the lower dorsal setae are wider than those in the 2nd foot, and have wings. The ventral setae have increased in size, and are very distinctly flattened, with slender tips. These changes are much more marked in the 4th foot (fig. 20D), especially in the dorsal division. The upper dorsal setae remain long and slender, with very narrowly winged tips, but there is a gradual transition to the lower ones (fig. 206), which are short and boldly curved, with widely winged blades. The ventral setae, in two rows, are all lance-shaped (fig. 20H), those at the lower border being very slender. The lower edge of the dorsal bundle curves backwards a little, as in P. antennata, Clap. (Mesnil, 1896, p. 228), and this tendency is much more obvious in the next segment. The 5th foot (fig. 20E) seems to be less modified than that of any other species of Polydora, and is very similar in structure to the 4th foot, as may be seen from a comparison of figs. 20D and 20E. Mesnil (1896, p. 235) on the contrary, ex- presses the view that in P. polybranchia, Haswell, and P. (Carazzia) antennata, Clp., the 5th segment has undergone greater modification than in any other species of Polydora. ‘This opinion is apparently based on the presence, in these two species, of two kinds of modified setae in the 5th segment. Ifthe 4th and 5th feet of P. kempi (a species very closely related to P. antennata) are compared, it will be apparent that 1921.] Fauna of the Chilka Lake : Polychaeta. 637 the only difference of note is in the shape of the dorsal inferior setae. In both feet the dorsal and ventral lamellae are similar, the dorsal and ventral bundles are verti- cally elongate, the dorsal superior setae are very long, with narrow wings, and the ventral setae are lance-shaped. In the 5th foot the dorsal inferior setae are sharply differentiated from the other setae. They consist of two rows, II-I5 in a row, of modi- fied setae (figs. 20E and zor). The anterior setae are bilimbate capillaries, short, with very broad wings and rapidly tapering filiform tips. The posterior row consists of rather stout hooks with curved tips. There is not much difference between the setae of the anterior row and the dorsal inferior setae of the 4th foot. Moreover, though the row of dorsal setae is distinctly curved at the lower end (fig. 208), a similar tendency is observed in the 4th foot. If the 5th foot of P. kempi is compared with that of a typical Polydora (as in fig. 21B), it is difficult to see how Mesnil’s assertion can be maintained. In the typical 5th foot, the dorsal and ventral lamel- lae are absent, and the general disposition of the dorsal and ventral bundles is more modified. The anterior row of dorsal inferior setae are usually modified as much as, or more than, those of P. kempi. The curvature of the dorsal row of setae of the 5th foot in P. kempr is much less than in the case of P. anten- nala, as figured by Mesnil (1896, Pl. XIV, fig. 22), or Carazzi (1893, Taf. 2, fig. 12) The 6th foot is very like the 4th, but the dorsal lamella is not quite so large. The setae are as in the 4th foot, except that the posterior row of the ventral divi- sion is composed of very slender thread-like setae. There are two pairs of glandular pouches in the 6th and 7th segments, as in P. antennata. In the 7th setigerous segment the branchiae appear. They are quite free from the dorsal lamellae, and at first are large. They are only 10-12 pairs of them, and the last few pairs are very small. In the 8th foot the ventral crochets appear (fig. 20J). They are 18-20 in each foot, and they are not accompanied by any capillary setae. On the crest there is a slender tooth so closely applied to the main fang that there is no angle between them. The shaft swells gradually above, and abruptly below, the narrow waist. The longest crochets are in the middle of the row. These crochets greatly resemble those of P. antennata. The dorsal setae are longitudinally striated, slender, and very narrowly winged, the lower ones being short but narrow. The wide form of the 4th foot has disappeared. — -In the 18th foot the dorsal setae, 8 or 9 in number, are very long and slender, wingless, or with indistinct wings. They emerge in a bunch, and no longer form a transverse row. The branchia is reduced to a small papilla, and the lamellae and fillets are small. In this region of the body the feet are insignificant, and project only slightly. In the absence of the posterior region it is impossible to be certain, but the insignificance of the feet in the middle of the body would seem to differen- tiate this species from P: antennata, where, according to Mesnil, the feet of the poste- rior region are very prominent, and the dorsal lamellae and fillets are well developed. 638 Memoirs of the Indian Museum. Vor V; Moreover, in the latter species the branchiae extend from the 7th foot to the 40th or 5oth. There is no pharynx in this species. None of the specimens was sexually mature, and probably they are not fully grown, so that changes may occur as growth proceeds. There is no doubt that this species has its closest affinities with P. antennata. It resembles it, and differs from other species of Polydora, in the structure of the 5th setigerous segment, which is only slightly modified, and in the appearance of the crochets in the 8th setigerous segment and in their shape. In other ways the two species are alike. I agree with Mesnil that these differences justify the separation of P. antennata and the creation for it of the genus Carazzia, in which the present species must also be placed. Habitat. —Five specimens were collected in October IgI4, in a canal at Chingri- ghatta, near Calcutta. The specific gravity of the water was 1'004. Family AMMOCHARIDAE. Myriochele picta, sp. nov. (Plate XXXI, figs. 30A-F.) Very numerous specimens of this species were included in the collection, all from the Chilka Lake. Unfortunately, the tubes are so narrow and tough, and the animals so fragile and firmly attached to the tubes, that it is almost impossible to extract an entire specimen. Frequently, also, the posterior extremity seems to be missing, or to have been regenerated to a certain extent. The tubes are very uniform in size, 7-11 mm. long, and only 2 mm. in width. They are composed of a tough inner membrane, covered on the outside with small quartz grains (fig. 30B). The posterior end of the tube is somewhat narrowed, and is covered with rather smaller quartz grains than the anterior end. The worms themselves are 3-4 mm. long. No specimens containing sperm or ova were observed, so that possibly the worms attain greater dimensions, but their uniform size, whether found in February or September, tends to show that they are fully grown. On the back of the head, opposite the anterior end of the mouth, is a conspi- -cuous patch of reticulate purplish brown pigment (fig. 30€). This patch is fairly constant in occurrence, though of variable size and density. At the posterior end of the buccal segment is a transverse dorsal band of similar pigment. A fainter band may occur near the anterior dorsal margin of the first thoracic segment, and smaller patches are sometimes found on other parts of the head, though these are frequently absent, at least in the preserved animals. A specimen, apparently complete (fig. 30A), was extracted from its tube. It is 3°5 mm. long, and consists of the head and 16 setigerous segments, of which 3 are thoracic and 13 abdominal. The head and buccal segments (fig. 30C) are quite fused, and frequently marked off from the thorax by a constriction. The head is some- 021] Fauna of the Chilka Lake : Polychaeta. 639 times swollen and bent, as in fig. 304. The mouth forms a conspicuous opening on the ventral surface (figs. 30C, 30D). It is surrounded by a thick folded lip, which is richly ciliated. The thoracic segments, 3 in number, are provided on each side with a group of capillary setae, 2-4 in number, and lateral in position. No hooks are present on the thorax. The first abdominal segment is as long as, or a little longer than, the three thoracic segments. The second abdominal segment is still longer, and is the longest segment of the body. The succeeding eight segments diminish only very slightly in length, but the three posterior segments are much shorter, especially the two last. The anal segment is conical. All the abdominal segments are provided with dorsal capillary setae and ventral hooks. The capillary setae are 1-5 in number. The proximal part of the seta is smooth, the distal part spinous, but the spines are so small that even under a magnification of 1300 diameters they are indistinct. The thoracic setae are slightly thicker than those of the abdomen. In the ventral tori the hooks are arranged in irregular transverse rows. The anterior segments contain 25-35 in each torus, the middle segments sometimes 55-65, the posterior segments as few as 5-10. The hook (fig. 30E) has a straight tapering shaft, a distinctly constricted neck, and above this a boldly curved bifid tip. The teeth at the tip vary considerably in thickness, the two represented in fig. 27E being the extremes. One hook is placed above the other, so that when seen from above (fig. 30F) only a single tooth is visible. In the anterior abdominal segments the setae are near the front part of the segment, but in the posterior segments they are at the back. . The head and buccal segment are almost entirely occupied by the buccal mass (figs. 30C, 30D). Passing backwards from the mouth, and lying beneath the buccal mass, is a clavate diverticulum, probably homologous with the “ Lippen-organ’’ described by von Drasche in Owenia fusiformis, D. Chiaje, and later by Watson, but not previously noted in Myriochele. The narrow oesophagus lies in the two anterior thoracic segments. In the third thoracic segment it passes into a somewhat swollen ‘pharynx. This is succeeded by the swollen intestine, which, in the first abdominal segment, is composed of a mosaic of flattened granular cells, and differs markedly in appearance from the succeeding and narrower part, though the transition in the second abdominal segment is gradual. The intestine is full of organic debris, and contains numerous diatoms, but no sand is present. A septum divides the buccal from the first thoracic segment, and another lies between the second and third thoracic segments. No septa could be distinguished in the anterior abdominal seg- ments, but they are present in the posterior segments, just behind the bundle of setae. In Owenia fusiformis all the segments, except the three thoracic, are separated by septa. Five pairs of ‘thread glands’ are present (fig. 30D), a pair in each of the three thoracic segments and in the two anterior abdominal segments. The thoracic glands are small, those in the third segment being the largest, whilst those in the abdominal 640 Memoirs of the Indian Museum. [Vor.. V, segments are very long. The thoracic glands open to the exterior by minutes pores near the setae, the abdominal glands between the dorsal and ventral setae. In Owenia there are 6 or 7 pairs of glands in the anterior segments, those of the third thoracic segment being sometimes small or absent. In Myriochele heeri, according to - Hansen (1882, pp. 41, 42), there is a pair of glands in each of the long segments, and he shows 14 such segments (tom. cit. T. vi, fig. 6). Fauvel (1014, p. 264) says that in M. heeri there is a pair of glands in each segment. McIntosh (1885, p. 410) refers a form dredged in 2975 fathoms, east of the Antilles, to the same species. The great depth at which it was taken, the large size of the specimens, and the fact that both extremities were missing when McIntosh examined the material, make the identifica- tion more than doubtful. McIntosh quotes the statement of Dr. v. Willimoes- Suhm, who examined the specimens when they were captured, that “ there is a pair of glands in each of the segments, from the second to the seventh.” This agrees more closely with the present form, in which they occur in segments 2-6. The Psammocollus australis of Grube (1867, p. 30) from St. Paul, appears to differ _ from the present species, so far as one can tell from the rather imperfect description and figures, in a few points, such as the size, position of the thoracic setae, shape of the hooks, which have only a single hook at the tip, and the larger number of capil- lary setae per bundle. Glands were observed to accompany the first and three follow- ing girdles of hooks, that is to say, in the four anterior abdominal segments. Habitat.—This species was found on 6 occasions, in the southern part of the lake, between Kalidai I. and Rambha. Five occasions were during the salt-water season and one during the freshwater period, but the specific gravity of the water only varied between I’005-I’01. The specimens were taken on a muddy bottom, in a fine-meshed net attached to the trawl. Family CAPITELLIDAE. Heteromastus similis, sp. nov. (Plate X XTX; figs. 23A-23H.) In addition to two very small specimens, one complete individual and a number of fragments are available for study. The description is drawn up from the complete individual (the type specimen). It is 55 mm. long, composed of 212 segments. Some of the other fragments apparently belong to rather larger individuals, and as no trace of sexual products was found, the mature individual is doubtless larger still. The body is long and slender, swollen at the anterior end, and tapering very gradually to the tail. In the preserved specimens the thoracic segments are pale buff colour, the abdominal segments greyish brown with a tinge of purple. The head (fig. 23A) is small and pear-shaped, consisting of a wide basal part and a much narrower tip. It usually projects freely, but may be partly contracted beneath the anterior border of the peristomium. No eyes were observed. The basal part of the proboscis is thin-walled, and is covered regularly with mammillate . papillae. The walls are full of clear oval refringent bodies. The anterior five seg- IQ21.] Fauna of the Chilka Lake : Polychaeta. 641 ments are usually swollen, and the inter-segmental grooves are faint. Behind the 5th (the 4th setigerous) and the succeeding segments the grooves are deeper, and there are faint transverse grooves in line with the setae. Beginning on the Ist seti- gerous segment there is on each side of the body a shallow groove, and these, com- bined with the broad and shallow ventral groove, produce two ventro-lateral ridges, carrying the ventral setae, somewhat in the manner of Ammotrypane. The ridges fade away in the middle of the body. The thorax consists of 12 segments, the Ist being achaetous. The 5 anterior setigerous segments have rather short capillary setae in both dorsal and ventral bundles. ‘The dorsal setae (fig. 23F) have broad tapering wings. The ventral setae are very similar, except that the wings are rather narrower, more elongate, and proximally do not taper so gradually. There are 7-15 setae in the dorsal bundles, 4-14 in the ventral bundles. In the 1st and 2nd feet the dorsal setae are in two rows, in the other bundles only in a single row. In the 7th-ı2th segments all the bundles are composed of long crochets (fig. 236). The distal quarter is widened, and enclosed in a sheath. The shaft narrows beneath the tip, which has a large tooth, with two small and indistinct spines on the crest. The shaft is long, doubly curved, and has no nodular swelling. . In the Ist setigerous segment the dorsal bundles of setae are separated by a space rather greater than that which separates the dorsal and ventral setae on each side. In the subsequent thoracic segments this interval is only slightly reduced, but in the 12th setigerous segment—the Ist abdominal—the dorsal bundles lie close together near the median line. The setae in the abdominal segments are surrounded by tumid lips, which are absent in the thoracic segments. ; In the thoracic segments the setae lie in the middle of the segment, but in the abdominal region the setae are near the posterior border. Furthermore, in the anterior region the segments are as long as broad, but they gradually shorten, till in the mid-region they are three or four times as wide as long. The length again increases in the posterior region till it equals half the width. In the middle of the body the constrictions between the segments are very deep, especially in the dorsal region. The abdominal setae are all very short crochets (fig. 23H), and have a nodu- lar swelling in the middle of the shaft. There are 10-15 of them in each bundle. The structures which, in this genus, are supposed to have a branchial function, consist of a number of lobes projecting from the posterior part of each segment. There are two of them on the dorsal surface (fig. 2 3c), carrying the dorsal setae, two in the latero-ventral region (figs. 23D, 23E) carrying the ventral setae, and one in the mid-ventral region, overlying the nerve-cord. The dorsal lobes, which make their appearance near the Iooth segment, are by far the most conspicuous. The ventral lobes appear near the 140th segment, and are smaller and more rounded than the dorsal pair. In the last few feet the branchial lobes are small and inconspicuous. In other specimens the branchiae are on fewer segments, but this may be due to re- generation of the posterior region. The anal segment bears ventrally a slender clavate tail, 33 mm. long (fig. 238). 642 Memoirs of the Indian Museum. [VOL-M, The type of this genus is Meteromastus filiformis (Clap.). It is known from the Mediterranean, the west coast of France, and Belgium (Ostend), and from several localities on the east coast of North America (Eisig, 1887, p. 839). The two forms are apparently so closely related that it is difficult to distinguish them. The Chilka Lake species is the smaller (55 mm. as against 100 mm.), but has more segments (212 against 140). The head, arrangement and structure of the setae, the branchiae and tail are almost identical. Eisig’s figures of the branchiae (1887, Taf. 27, fig. 18) shows a lobe on each side, between the dorsal and ventral lobes, which is absent in the Chilka Lake specimens (fig. 23E). In A. filiformis the anterior abdominal seg- ments are much longer than the thoracic segments, whereas in H. similis the lengths do not differ. Moreover, though considerable work has been done in the intervening area of the Red Sea and Persian Gulf, Heteromastus has not been found there. I have not been able to examine specimens of H. filiformis, but it is probable that a comparison of the two species would reveal other differences, whilst possibly reduc- ing the importance of those mentioned above. Habitat.—This species was taken at three localities in the Chilka Lake, between Nalbano and Berhampur, that is, near the inner end of the outer channel. Two of these stations were worked in March and one in September. The specific gravity of the water ranged from 1:000-1'0261. The species was taken once on the shore, twice in a few feet of water. Genus Barantolla, gen. nov. Capitellidae having 12 thoracic segments, of which the 1st is achaetous. Segments 2-7 have only capillary setae, segments 8-12 only elongate crochets. The abdominal segments have short crochets only. The anterior thoracic segments have reticulate markings on the skin, and the sculpture of the thoracic segments is rather elaborate. Branchiae in the form of short finger-shaped lobes behind the dorsal setae of the middle and posterior segments. These segments ave provided each with a membranous collar, produced into jour shallow parapodial lobes. This genus shows marked affinities with Heteromastus and Mastobranchus, and it is evident that the three genera are closely related. They all possess 12 thoracic segments, of which the Ist is achaetous. Barantolla agrees with Heteromastus in hav- ing capillary setae on the anterior, long hooks on the posterior thoracic segments, and only short hooks on the abdominal segments, but differs in having capillary setae on the segments 2-7, whereas Heteromastus has them only on segments 2-6. Mastobran- chus is peculiar in having capillary setae only, on the thoracic segments, and in the dorsal bundles of a number of the anterior abdominal segments. On the other hand, Barantolla agrees with Mastobranchus in having finger-shaped branchiae, situated behind the dorsal setae of the middle and posterior segments. The parapodial lobes (branchiae) of Heteromastus are very like those of Barantolla. Heteromastus and Mastobranchus are peculiar in having finger-shaped appendages on the anal segment, and it is unfortunate that no information is available on this point for Barantolla. The general appearance of the abdominal segments is very similar in all three 1921.] Fauna of the Chilka Lake : Polychaeta. 643 genera. The affinity between Barantolla and Mastobranchus is also indicated by the crenulated skin of the anterior thoracic segments, and by the discovery in the species of Mastobranchus described below of two elongate crochets in the ventral bundle of the 12th segment (vide p. 646). A comparison of these three genera, based on a much more ample supply of material, would probably lead to considerable modifications. The form described by Stephenson (1908, p. 39) from brackish ponds at Port Canning, Lower Bengal, as Matla bengalensis may be related to one of the above genera. The specimens examined were very young, I'’5-4°5 mm. in length, and there is little in the description to indicate what the adult would be like. Barantolla sculpta, sp. nov. (Plate XXIX, figs. 24A-K.) This species is represented by a number of fragments, including the anterior ends of four individuals. Unfortunately the posterior end is not present in any specimen. The largest fragment is 117 mm. long, composed of 135 segments. Judging from the appearance of the last few segments, the missing portion of the tail is quite short. The body is widest near the 4th or 5th setigerous segment (1°7 mm. wide), and gradually tapers till near the 30th segment. It then remains fairly uniform in diameter for a long distance, expanding again a little near the posterior end. The length of the segments varies considerably, according to their-position. In the type- specimen the approximate ratio of the width to the length is in segments 1-5, 3 tor; segments 6-14, 4 to I, segments 15-20, 3 to I; segments 21-35, 2 to I; segments 36-43, 14 to I; segments 44-55, I to I; segments 56-70, a little less than 1 to 1; segments 71-80, 1 tor. The segments then gradually become shorter towards the tail, and the last segments are 6-8 times as wide as long. The head in all the specimens is contracted, and withdrawn under the peristo- mium, and hence it is difficult to describe accurately. It is conical, composed of two rings, a wide basal, and a narrower terminal. The latter, in its contracted condi- tion, is rounded. No eyes could be seen, but behind the basal ring of the head is an irregular transverse row of deeply embedded black pigment spots, which, however, may have no optical function. The proboscis is partially extruded (fig. 24A), and this part is covered with very minute papillae. The thorax consists of 12 segments, of which the first is achaetous. In the first four segments, and to a lesser degree, in the anterior part of the 5th, the skin is dis- tinctly tessellated (figs. 24A-C), but on the succeeding segments it is smooth, except for ridges caused by unequal contraction of the skin. Segments 2-7 have capillary setae only, segments 8-12 long crochets only, and the abdominal segments short crochets only. In the thoracic segments the setae are neat the middle of the segment, but in the abdominal segments they are near the posterior end. Each of the abdominal segments is in two parts, an anterior narrow 644 Memoirs of the Indian Museum. NOS smooth part, and an expanded posterior part, the latter bearing the swollen tori. The dorsal tori (fig. 24C) are widely separated in segments 2-4, and then gradually approach each other, till from the I2th segment backwards the tumid lips of the tori form a dorsal ridge. | The sculpturing of the integument of this species is very characteristic. On the ventral surface (fig. 24A) there is, in the median line of segments 2-5, a row of hexa- gonal areas, 2 or 3 in each segment. These are replaced in segments 6-12 by raised areas not clearly marked off except by a median transverse groove. In segments 5-I2, on each side of the raised median areas, are narrow hexagonal areas, the outer ends of which reach to the ventral setae. In side view (fig. 24B) the lateral organs are clearly shown in segments 5-12, less clearly in segments 2-4, and are very in- distinct on the abdominal segments. Dorsally (fig. 24C) there are median hexagonal areas on segments 6-II. A shallow lateral groove runs backwards on each side from the 13th segment, but occasionally it begins near the anterior end. It is probably due to the contraction of the specimens, and the presence of very powerful longitudi- nal latero- ventral bands of muscle. The branchiae commence between the 55th and 60th segments. They lie under the dorsal parapodial lobes (figs. 24D-E). The anterior branchiae consist of 3 or 4 short rounded lobes, hidden by the parapodial lobes. Towards the tail the bran- chiae increase in number and length. The largest, near the posterior end, have 9-II finger-shaped lobes, attached to the underside of the dorsal parapodial lobes (fig. 24E). The latter structures are part of a thin membrane which surrounds the segment like a collar, the anterior margin being fixed, the posterior one free and pro- duced into four rounded lobes, the two larger carrying the dorsal setae, the two lateral, which are not so pronounced, carrying the ventral setae. The collar is attached to the body-wall near the middle or the beginning of the posterior third of the segment. The appearance of this part of the body is very like that of Heteromas- tus. In segments 2-7 the dorsal and ventral tori contain only capillary setae. They have narrow wings (fig. 24F), and taper to a slender tip. he bundles of the 2nd segment (Ist setigerous) have about 16 setae in each, the 3rd-7th segments have 20-40 setae in each bundle. In segments 8-12 the bundles contain only long crochets (fig. 24G), which closely resemble the similar setae of Heteromastus. The tip (fig. 24H), which is enclosed in a long narrow sheath, is not expanded as in Hetero- mastus. It terminates in a strong tooth, with 5 or 6 slender spines on the crest. There is a very slight nodulus where the seta pierces the skin. The nodulus lies just within the proximal half of the seta. Each bundle contains 20-40 of these setae. The transition between the thoracic crochets and the typical abdominal crochets is not so abrupt as in other species of the family. The typical abdominal crochets are much smaller than those of the thorax (fig. 24K), they have a more distinct swelling below the neck, the sheath over the tip is relatively shorter, and the nodu- lus is more distinct, and is in the distal half of the seta. On the anterior abdominal segments, the setae (fig. 24J) are intermediate in size, rather more like the thoracic OT] Fauna of the Chilka Lake : Polychaeta. 645 than the abdominal crochets in shape, with only a slight expansion below the neck, but they have short wings over the tips, and the nodulus is in the distal half, though only slightly developed. The length of these crochets is as follows :— On the 8th segment, ‘33 mm. 93.99 da ” ) 25— 20 mm. ” )) 14th ” D #17 mm. 30899 both ” ) ‘05—"06 mm. Passing towards the tail the abdominal setae become fewer and smaller. In the 13th foot there are 25 ina bundle, on the 60th foot Io, and on some of the posterior feet 2 or 3, or none at all. The spines on the crest of the crochets form a transverse curved row, the concavity facing the terminal tooth. Habitat. —This species was collected in some brackish pools near the salt lake at Barantolla, near Calcutta, in November. The salinity of the water is very variable, but never high, probably never exceeding I’0I5. Mastobranchus indicus, sp. nov. (Plate XXX, figs. 254—F.) This description is founded on a single specimen from brackish pools, near Barantolla. Unfortunately only the anterior end is available, in a very contracted condition. It is 46 mm. long, and consists of 90 segments. The widest part is near the anterior end, where it is 3 mm. wide. The fragment evidently belongs to an individual considerably larger than the specimens of Barantolla sculpta. The head (fig. 254) consists of a small rounded lobe, withdrawn under the buccal segment. No trace of eyes or nuchal organs could be found. The skin of segments I-6 and the anterior part of 7 is tessellated, the grooves being deeper in the anterior segments. The thorax consists of 12 segments, of which the Ist is achaetous. The sculpturing of the thorax recalls that of Barantolla sculpta (compare fig. 254 and fig. 24a), but is not so distinct or general. On the ventral surface hexagonal areas occupy the median line in segments 5-8 and the anterior part of 9. The ventro-lateral narrow hexagonal areas are clearly differentiated on segments 6-9, less clearly on segments 3-5. In segments 10-13 these areas gradually coalesce with the tumid tips of the ventral tori. The lateral organs are not very distinct, but narrow hexagonal areas, clearly marked off by grooves, lie between the dorsal and ventral setae in segments 5-8. They probably exist in other adjacent segments, but are not distinctly differentiated in this specimen. On the dorsum narrow trans- verse hexagonal areas occur between the dorsal bundles of setae in segments 5-11. As is the case with all the other areas, lateral and ventral, it is the posterior groove which disappears first, the anterior grooves remaining on many posterior segments, forming a complete groove round each segment. Four pairs of genital pores were observed, behind segments 8-11. In the lateral region there is a deep groove on each side, commencing in the 6th segment, and these, combined with a shallow 646 Memoirs of the Indian Museum. [ Vor. V, ventral groove, help to form two ventro-lateral ridges, containing the stout longi- tudinal neural muscle bands. The tori in segments 2-4 are very short, containing about 20-30 setae. Segment 5 has rather longer tori, and on the 6th and subsequent segments the tori are 3-4 times as long, and contain 50-60 setae. There is no very obvious external difference between the posterior thoracic and the anterior abdominal segments. The latter are divided into two rings, asmooth anterior and a slightly enlarged posterior ring, the latter bearing the tori with their tumid lips. The setae of segments 2-4 are rather short, with pointed tips, bilimbate, with narrow wings. The ventral setae closely resemble the dorsal setae (fig. 25B), and they are very similar to the capillary setae of Barantolla sculpta. The thoracic setae are all capillaries with the exception of two in the right ventral bundle of the I2th segment (11th foot). This bundle contains 45 setae, of which 43 are normal capil- laries. The other two are elongate hooks, very similar to those in the posterior thoracic segments of Barantolla sculpta. The nodulus is only slightly indicated. It is well in the distal half of the shaft, thus differing somewhat from the similar setae of Barantolla sculpta. The shaft is widest near the lower end of the wing, and narrows gradually towards the neck, beneath the hooked tip. In the absence of any marked widening of the shaft below the neck, these setae resemble those of Barantolla sculpta and differ from those of Heteromastus similis (compare figs. 236, 24G and 250). The tooth at the tip is rather slender, and makes an obtuse angle with the shaft. Behind the tooth the crest bears 5 or 6 slender spines. These setae are 52 mm. long, a little shorter than the capillaries of the same bundle, which are ‘6 mm. long. Except for the hooked tip, they greatly resemble the capillary setae in general appearance. No other crochets were found in the thoracic tori. The setae of the corresponding ventral bundle on the left side of the body were all broken, except a few, which had capillary tips. In the 11th segment all the setae were capillaries. Owing to the lack of material it is impossible to elucidate the significance of the pre- sence of these two elongate crochets. The resemblance they exhibit to the similar setae in the posterior thoracic segments of Heteromastus and Barantolla indicates the close affinity existing between these three genera, and their presence in this specimen may represent a reversion to an ancestral type which possessed them normally in the posterior thoracic segments. On the other hand it is possible that these setae occur normally in the young and immature stages of Mastobranchus indicus, and are replaced by capillaries as the individuals approach maturity. The dorsal bundles of the 13th and 14th segments each contain about 40 capil- lary setae, similar to those in the thoracic feet, and there are no crochets accompany- ing them. ‘The ventral bundles contain only crochets, about 40 in each bundle. These differ considerably from those in the last thoracic ventral bundle (figs. 25D, 25E). The nodulus is distinct, in the distal part of the shaft. The shaft shows a distinct swelling below the neck. These setae are only ‘28 mm. long. The wing, enclosing the tip, is relatively short. The terminal tooth makes rather a wide angle with the shaft, and on the crest behind it are two transverse rows of spines. In the 1921.] Fauna of the Chilka Lake : Polychaeta. 647 dorsal bundle of the 15th segment there are only crochets, ‘33 mm. long, in shape like those of the ventral bundles. In the 20th segment the dorsal crochets are ‘174 mm. long, the ventral crochets ‘108 mm. long. The nodulus is very distinct, but the swelling below the neck is not so evident. In the 50th segment the dorsal crochets are ‘12 mm. long. In the ventral bundle there are 63 crochets, ‘I mm. -long (fig. 25F). The nodulus is very distinct, in the middle of the shaft. From Garia, near Calcutta, in December 1910, were obtained two fragments of a Capitellid, which may belong to the present species or to Barantolla sculpta. They are both from the middle of the body, but one is evidently from a more posterior region than the other. Both fragments have very powerfully developed ventro-lateral muscles, which are more prominent than those of Barantolla sculpta. One fragment has branchiae on each segment, behind the dorsal setae. They are very similar to those of Barantolla, but rather longer, consisting of 5-7 finger-shaped lobes. The dorsal tori are more elongate and the setae more numerous than those of Barantolla. In the absence of any knowledge of the structure of the middle and posterior regions of Mastobranchus indicus, it is not possible to assign these fragments to any species at present. The present species is assigned to the genus Mastobranchus by virtue of the structure of the thorax and anterior segments of the abdomen. A knowledge of the structure of the anal appendages, and branchiae (if any are present) is necessary to modify or confirm this conclusion. In view of the great resemblance between this species and Barantolla sculpta, and the presence of both forms in the brackish ponds near Barantolla, the question arises whether the latter does not represent a stage in the growth of the former. All the specimens of Barantolla available are immature. In the sculpture of the skin, shape of the anterior end, and the form of the setae there is close resemblance. On the other hand, all the specimens of Barantolla sculpta agree in having capillary setae on segments 2-7, long crochets on segments 8-12, and only short crochets on the anterior abdominal segments. Mastobranchus indicus has capillary setae only, on the thoracic segments 2-12, with the exception of the two long crochets noted above, and capillary setae in the dorsal bundles of the two anterior abdominal segments. If B. sculpta is only a stage in the development of M. indicus, then the completion of the metamorphosis has been postponed to an unusually late period, as the specimens are at least 120 mm. long. On the whole I am inclined to regard the two forms as distinct, but the question can only be definitely settled by the examination of more material, at different stages of growth. Habitat.—A single imperfect specimen from brackish pools, salt lakes, Barantolla, near Calcutta. The specific gravity of the water is very variable, but never high, probably never exceeding 1'‘015. 648 Memoirs of the Indian Museum. Ver N; Family MALDANIDAE. Euclymene annandalei, sp. nov. (Plates XX VIII, figs. 22A-G, and XXIX, figs. 22H-K.) This species is represented by numerous specimens, collected at 11 stations, all in the south-western end of the Chilka Lake. Apparently they live in tubes, but these are so brittle that only a few fragments remain, composed of sand grains. The body varies from 40-80 mm. in length, and each complete individual is, without exception, composed of the head, buccal segment (achaetous), 21 setigerous segments, 2 posterior achaetous segments, a funnel-shaped ring, and the caudal ring. The preserved specimens are almost colourless, except for the ocelli. Epidermal glands were diffusely scattered on the head and anterior segment (fig. 224). On the Ist, 2nd and 3rd setigerous segments there is a narrow ring of glands in the anterior part, in front of the setae, the 3rd also having scattered glands over the whole skin. The 4th setigerous segment has bands of glands in front of and behind the setae. In the 5th-8th setigerous segments the epidermis in front of the setae is thickly covered with glands, as well as the parapodial pads. In the subsequent segments there is a strong longitudinal mid-dorsal band of glands and the parapodial pads are thinly covered. There are also narrow diffuse longitudinal bands running in:the line of the dorsal setae. Beginning in the 7th setigerous segment there is a conspicuous double band of glands on the mid-ventral surface, lying over the ventral nerve-cord, and running back to the caudal ring. The head (figs. 224-c) has a large and concave dorsal cephalic plate. The frontal process is fairly large and rounded. The sides are thin and broad, divided by indentations into two lateral smooth areas with entire borders, and a posterior (dorsal) crenate portion of eight rounded lobes. The nuchal grooves are rather long, almost parallel, and are separated by a narrow high keel. On the tip of the ventral side of the prostomium (fig. 22D), in front of the mouth, are 4 elongate patches of very numerous reddish-brown ocelli. The two median patches lie partly under the frontal process, and in some cases are fused in the median line. The two other patches are more lateral in position. The proboscis is partly extruded in several specimens. Its proximal portion is covered with low conical papillae, which are largest near the mouth, and the distal region is smooth. The buccal segment and prostomium combined are rather longer than the Ist setigerous segment. Setigerous segments I-6 are approximately equal in length, the proportions varying greatly in different individuals, according to the degree of con- traction. ‘The 7th setigerous segment is rather longer, and the 8th is the shortest of all. ‘The following segments are considerably longer, diminishing a little towards the tail, the last setigerous segment—the 21st—being rather short. Then follow two achaetous segments, the second of which is very short. The latter is prolonged into a bell-shaped ring, which has no parapodial pads. The caudal segment is funnel- shaped (figs. 22E-F), with a deep posterior depression, at the bottom of which lies the anal cone. The height of the latter varies according to the degree of contraction 1921.] Fauna of the Chilka Lake : Polychaeta. 649 of the intestine. It is folded, especially round the anus. The caudal ring is fringed with short bluntly rounded cirri. The median ventral cirrus is stouter than the rest, and 2-4 times as long. In I5 specimens examined, the number of cirri, including the large ventral one, is as follows :—24, 20 (2), 19 (2), 18, 17 (3), 16 (3), 15, 14 (2). Sometimes several of the cirri are fused together, forming a flat plate. In setigerous segments I-3 the parapodia are in the anterior half of the segments, in 4-8 in the middle, and in the subsequent segments near the posterior end. The three anterior setigerous segments have dorsally a double row of capillary setae. They are longitudinally striated, with two narrow wings and slender tips. Ventrally there is a single uncinus (fig. 22H) with a simple boldly curved tip. In the two anterior bundles the setae are rather more numerous than in the 3rd. In the 4th setigerous segment the dorsal setae are similar, but fewer in number. Ventrally there is a row of uncini, each with 5 teeth above the main fang (fig. 22K). In the 8th foot several of the dorsal capillaries have plumose tips (fig. 227), the spinelets being rather long. These setae are much shorter than the smooth capillaries. There is a ventral row of 14 uncini. In the roth segment there are 17 uncini, in the 17th segment I4 uncini. In the anterior segments the row of dorsal capillary setae gradually shortens as the setae become fewer in number, and at the same time the papilla from which they emerge increases in length. It is low and insignificant on the Ist setigerous segment, but on the 5th or 6th it forms a distinct swelling. In segments 1-0 the capillary setae form a transverse row. At the roth segment the row tends to be oblique, and this tendency increases till at the 15th segment the setae form an arch above the uncini (fig. 226). The latter are embedded in a long narrow slit surrounded by prominent tumid lips. The uncini (fig. 22K) have 5 or 6 diminishing teeth above the main fang. Beneath the latter there is on each side a bundle of bristles, but their point of attachment is not indicated by a prominent boss, as in many other species. The uncini of segments 4-21 are all similar, and there is no obvious difference between the uncini of the same foot. Habitat.—This species was taken in large numbers at 11 stations, all in the south- west end of the lake, south of a line drawn from Patsahanipur to Nalbano. Usually they were obtained by digging in mud or sand at the water’s edge, but on three occa- sions they were taken some distance off the shore, on a muddy bottom. As might be expected from the habitat of this species, it was usually taken in the salt-water season, when the water level was low, but on one occasion it was taken in the dredge in the freshwater season. The specific gravity of the water ranged from I’002- TOLT. Family STERNASPIDAE. Sternaspis costata, Marenzeller. (Plate XX, figs. 5A, 5B.) 1879. Sternaspis costata, Marenzeller, p. 34, taf. vi, fig. 4. 1890. Sternaspis costata, Sluiter, p. 108. 650 Memoirs of the Indian Museum. [Vor Ve It is with some hesitation that the numerous specimens of Sternaspis from the Chilka Lake are referred to the species from Japan, described by Marenzeller, and rediscovered at Batavia by Sluiter. According to Marenzeller, it agrees closely with the widely distributed S. scutata (Ranzani) of European seas, the most marked difference being in the shape of the posterior ventral shield. In the latter respect the Chilka Lake specimens show a fairly close agreement with the description and figures given by Marenzeller. Not one of the 316 specimens which I examined showed any trace of an elongate proboscis such as Sluiter found in S. spinosa, and it is obvious that the latter species should be removed to a new genus. The prostomial lobe is small and rounded, exactly as in S. scutata. The largest specimens were 8 mm. long, and the greatest width was 3-35 mm. The largest specimen examined by Marenzeller was 12 mm. long and 5 mm. wide. Sluiter examined three specimens, 18 mm., 22 mm., and 35 mm. long respectively. In the smallest of these, the ventral shield closely resembled that of S. costata in shape and colour, but the largest individual varied considerably in the direction of S. scutata, the other specimen being intermediate in size and structure. European specimens of S. scutata which I have been able to examine were larger than those from the Chilka Lake, and differed in the shape of the ventral shield. Some of the specimens, taken in the outer channel in September, had the body cavity full of ripe eggs, so that apparently the Chilka Lake species does not attain a much greater size than 8 mm. The body as a whole is brownish yellow in colour. The yellow pigment is specially marked on the gills, the intersegmental areas, and in the area surrounding the mouth and prostomial lobe. The wide posterior part of the body is more deeply coloured than the narrow anterior part, and the body-wall of the former is also thicker and less transparent. The prostomial lobe is quite transparent. The ventral shield (fig. 5A) is brightly coloured, the colour ranging from rusty red to scarlet. There is a narrow colourless margin on the lateral and posterior borders. It easily separates into two halves, and each half is slightly longer than broad. It has a series of ridges radiating from the centre, each ridge forming a groove over one of the bundles of setae, which emerge from the body-wall beneath the shield. A series of concentric markings is also obvious. ‘The anterior and posterior borders are apparently deeply indented, the indentations, which are not so rounded as in Marenzeller’s figure, being occupied by paired triangular plates. The anterior pair, which Marenzeller calls accessory plates, and which are absent in S. scutata, really form part of the large plates, being only differentiated by their delicate transparency and smoothness, and by the conspicuous ridge which separates them from the large plates. They only occupy part of the anterior concavity. The posterior triangular plates make the posterior border almost straight. . They show the same ridges and concentric markings as the large plates, and five pairs of bundles of setae emerge from beneath them. From beneath the large plates 9-11 pairs of bundles of setae emerge. There are thus 14-16 pairs altogether, those at the outer posterior angles being the largest and longest. All 1921.] Fauna of the Chilka Lake : Polychaeta. 651 these setae are longitudinally striated, and some are smooth, whilst others have the terminal part of the shaft covered with fine short hairs. The three anterior segments of the body have two bundles of setae each. There are 14-26 setae in each bundle, the youngest being on the ventral margin of the bundle. The setae have a double curvature, and are longitudinally striated the striations being in the superficial layer. The core of the seta is abruptly truncated some distance from the tip (fig. 5B). This appearance is not caused by injury, as it was observed in cases where the seta tapered to a fine point. Distribution.—Miya Bay, Japan; Bai von Batavia, in mud, 10-12 fms. Habitat.—316 specimens were taken at four stations in the outer channel near Mahosa, in September, when the water was quite fresh. The bottom here is sand and sandy mud. Mr. Kemp says im hit. : “As regards the Sternaspis, we only found it in the outer channel at the season when the water was quite fresh. But our observations in March were a little incomplete, for the water was too shallow to permit the passage of the launch across the bar separating the main area from the outer channel, and we had only a dinghy to work from. The species is doubtless present all the year round in salinities varying from 1:000 to 1°027.” Family SABELLIDAE, Potamilla leptochaeta, sp. nov. - (Plate XXXI, figs. 28A-N.) A number of small specimens of this species were found in masses composed of Entoproct Polyzoa (Loxosomatoides colonialis, Annandale), accompanied by Bowerban- kia caudata, Hicks, and by a number of Nematode worms (Oncholaimus indicus, von Linstow, Rec. Indian Museum, Vol. I, 1907, p. 45). The Potamilla lives in rather brittle membranous tubes, coated with fine mud, very like those of P. torelli, but not quite so hard and hyaline. The stolons of the Polyzoan are attached to the worm tubes, the whole forming a tangled mass. Most of the specimens are incomplete, the posterior extremity usually being absent. A moderate sized individual, not the largest, measured 4°5 mm. in length, of which the branchiae were 1°3 mm., the thorax I’2 mm., and the abdomen 2 mm. The branchiae are marked by two broad bands of reddish brown pigment. The total number of setigerous segments is about 33 or 34, of which 6 compose the thorax, and 27 or 28 the abdomen. The thorax is of uniform width, but the abdo- men tapers gradually towards the anal segment. The latter is rather flattened, and heart-shaped, with rounded posterior border. No eyes were observed, either on the anterior or anal segments or on the branchiae. In all specimens examined, there were 9 branchiae, 4 on the right side, 5 on the left. Each branchia has 40-50 filaments. The latter are nearly all of the same length, decreasing slowly in size towards the tip. In most of the specimens the tip is short, bare, and blunt (fig. 28c), but one specimen was found in which all the branchiae had rather long bare tips (fig. 280). A more careful examination then showed that branchiae with long tips were by no means uncoiamon, even in speci- 632 Memoirs of the Indian Museum. Nera we mens where the majority had short tips. Moreover the left ventral branchia, which was always much shorter than the others, always had an elongate tip. It seems probable, therefore, that the long tip is the normal condition, and that the short tips are due to accidents. Dorsally the front margin of the collar slopes backwards, and the two ends are attached near the median line (fig. 28a). Ventrally the collar inclines forward (fig. 288), and in the ventral region it is deeply cleft and bilobed. | The thorax is composed of 6 setigerous segments. The faecal groove is narrow - shallow, and very indistinct, and can hardly be seen on the abdomen. Ventral gland shields are very inconspicuous. On the mid-dorsal area there is a slightly elevated ridge (fig. 284), especially distinct at the anterior end, where it narrows, and lies between the ends of the collar. The first thoracic segment bears only dorsal bundles of setae. Each bundle contains 7 setae, the 3 upper being long and bilimbate, as in the succeeding thoracic segments (fig. 28F). The 4 lower setae are shorter (fig. 28E), with wider and shorter bilimbate blades and filiform tips. They are intermediate in shape between the elongate capillaries and the spatulate setae. In the remaining thoracic segments, the dorsal bundles contain two types of setae, capillary and spatulate. The capillary setae (fig. 28F) have elongate narrow wings and very long filiform tips. There are 3-6 in each bundle. Beneath and behind them are the spatulate setae, 5-10 in number. They have pear-shaped blades (fig. 286) with finely pointed tips. The ventral bundles also contain two types of setae. The anterior row (figs. 28H, 287) consists of 7-10 cuspidate setae (soies en pioche). The posterior row contains 7-0 avicular uncini (fig. 28K). The anterior part of the base is large and swollen, and the posterior process is rather short. In the anterior abdominal segments the ventral bundles contain 5-7 capillary setae, with short blades and very long filiform tips. They vary in length (figs. 281, 28m), but are all bilimbate, and the blades are shorter and wider than those of the thoracic setae. The dorsal bundles contain 9 or 10 avicular uncini (fig. 28N), differing very slightly from those of the thorax. The posterior process is shorter, and the spinose crown is higher and narrower. The uncini vary much in size in each bundle, the largest being ventral, the smaHest dorsal. In the posterior part of the abdomen the ‘setae are similar, but fewer and smaller. The ventral bundles contain 1-3 capillary setae, the dorsal bundles 1-3 uncini. The most remarkable feature about the setae of this species is the ee length and extreme slenderness of the tips of the capillary setae, from which character the specific name is derived. Habitat.—From a brackish pool, Port Canning, Lower Bengal, December 1908. The water here is of very variable salinity (Annandale, 1907, pp. 35 and 197). Laonome indica, sp. nov. (Plate XXX, figs. 26A-H.) Only a single individual of this species is available. It is 28 mm. long, the trunk being 26 mm., and the unusually short gills only 2 mm. The body is slender, TOP Fauna of the Chilka Lake : Polychaeta. 653 composed of the 110 segments, and the anal segment is large and conical. The posterior 15 segments are much shorter than the preceding ones, and may have been regenerated. There are seven pairs of branched gills, unconnected by a membrane (fig. 264). Each gill is penetrated by a single blood-vessel (fig. 26C), which sends off a single branch to each barbule. The barbules arerather short, and the distal quarter of each gillis devoid of them. In addition there are two smooth unbranched ventral tentacles, or palps. The upright collar is equal in length to the Ist setigerous segment. On the dorsal side (fig. 264) the two halves of the collar are simply folded inwards, and _ attached to the body wall. On the ventral side (fig. 26B) the collar projects forwards in two peculiar lobes, each pointed at the tip. Between these two lobes and the: palps is a median lobe rounded at the tip. No eyes or otocysts were observed. The thorax is composed of six setigerous segments. Along the median dorsal line there is a narrow groove, which passes to the ventral side in the 6th setigerous segment, and runs to the posterior end. Ventral thoracic gland plates are incon- spicuous or absent. The Ist thoracic segment bears only dorsal bundles of setae, which lie nearer the median line than those of the succeeding segments (fig. 26a.) The setae are all capillaries, mostly long and slender, with narrow bilimbate blades. A few are shorter, with relatively broader blades. In the znd segment the dorsal setae are of two kinds. The upper setae are long slender capillaries (fig. 26D), 6 or 8 in number, with narrow wings and long tapering tips. Below these are 9-11 setae with spatulate tips (fig. 26E) terminating in a long fine point. The ventral setae consist of a row of 20 uncini (fig. 266). Each of the latter has a stout rounded base. In front view (fig. 266, a) there are 4 or 5 rows of teeth on the crest above the main fang. The side view (fig. 266, 6) shows that the main fang is very acute, with 4 teeth above it. The remaining thoracic segments, 3-6, have similar setae. Of the abdominal setae those on the 8th setigerous segment may be taken as typical. The dorsal division consists of 17 uncini (fig. 26H) differing only very slightly from those of the thorax. The rounded base is rather more oblique, and the gap between it and the main fang is larger. The ventral group contains 4 or 5 capillary setae (fig. 26F). The expansion formed by the wings is shorter and broader than in the thoracic setae, and the tip is long and slender. In the posterior segments the setae are similar in shape, but the capillaries are rather longer. Nothing is known about the tube of this species. Habitat.—A single specimen was taken in the south-western extremity of Chilka Lake, 1-8 miles N.W. by N. of Sanad Point. The specific gravity of the water was 1'009 during the salt-water season. Fabricia (Manayunkia) spongicola, sp. nov. (Plate XXXI, figs. 29A-E.) A number of specimens of this species are available, in a rather contracted condition. They inhabit tubes, consisting of a delicate membrane covered with 654 Memorrs of the Indian Museum. [VoL. V, flocculent mud. The length of the animal varies from I’5-3 mm. In a specimen 1:65 mm. long, the branchiae constituted ‘44 mm., the thorax I mm., and the abdomen ‘25 mm. In a second specimen 2'3 mm. long, the branchiae were only "33 mm. in length. The branchiae (fig. 294) consist of about 18-20 slender unbranched filaments on each side (i.e. about 36-40 in all). Occasionally they present the appearance of being arranged in groups attached to a short common stem, but this is probably due to the contraction of the basalmembrane. In appearance and structure they closely resemble those of Manayunkia speciosa, as described and figured by Leidy (1883, p. 206, pl. ix, fig. 1). Within the circle of branchiae are two short clavate processes, the “palps’” or “prostomial tentacles.” The body (fig. 294) is cylindrical, and consists of 13 segments, the first and last being without setae. The head is conical in front, and bears two black eyes. There is a prominent collar, with an entire convex border ventrally. Dorsally the ends of the collar are rounded and folded inwards. The thorax consists of 8 setigerous segments, the anterior three being short, but gradually increasing up to the 8th, which is the largest in the body. No otocysts are present. There are 3 setigerous abdominal segments, and an achaetous anal segment, the latter being usually spatulate, sometimes pear-shaped, and bearing two black eyes. The Ist thoracic segment has a bundle of 4 slender capillary setae on each side, but no crochets. On the succeeding 7 segments there are on each side 2-4 dorsal capillary setae, and 2-5 ventral crochets in a single row. The capillary setae (fig. 29B) have short flattened blades and long slender tips. The crochets (fig. 29C) are rather stout, with three teeth above the main hook. In the abdominal segments the ventral bundles contain one or two very slender capillary setae (fig. 29D), which show only very slight flattening of the blade and no wings. The dorsal bundles contain 9-II crochets of peculiar form (fig. 29E). They are rather small, with elongate shafts, and with numerous very fine long teeth, in several rows, at one end. Beneath the teeth there is a slight swelling of the shaft or base of the crochet. These setae closely resemble those cf Haplobranchus aestuarinus, described and figured by Bourne (1883, p. 171, pl. ix, fig. 14), and in a lesser degree, those of Manayunkia and Fabricia. They are obviously intermediate in shape between the elongate crochets of the thorax and the avicular uncini such as are found in Potamilla and other genera (fig. 28k). _ The species agrees closely with the three genera Fabricia, Manayunkia, and Haplobranchus. The characters held in common are: (I) the possession of II seti- gerous segments, 8 thoracic and 3 abdominal, (2) the shape of the dorsal and ventral setae in the thoracic segments; (3) the peculiar elongate crochets in the abdominal segments; (4) the presence of eyes on the first segment; (5) and the absence of otocysts. The present species agrees with Manayunkia in having a well-developed collar and unbranched branchiae, and differs in the presence of caudal eyes. It resem- bles Fabricia in having caudal eyes, and differs in having unbranched branchiae and a well-developed collar. It agrees with Haplobranchus in having unbranched 1921.] hauna of the Chilka Lake : Polychaeta. 655 branchiae, though in much larger number, in the presence of a well-developed collar, and in the shape of the abdominal crochets, and differs in the presence of caudal eyes and in the shape of the abdominal capillary setae. Leidy (tom. cit.) figures but does not describe the collar of Manayunkia speciosa, and his figures are not decisive as to the presence or absence of the dorsal indentation. Bush, however (1905, p. 188), in a generic diagnosis of Manayunkia, says that the collar is entire, without in- cisions or clefts. If this is the case it resembles Haplobranchus and differs from F. spongicola. ‘The latter reduces the gap between Manayunkia and Haplobranchus on the one hand, and Fabrica on the other, and it has as much or as little right as these forms to generic rank. All the characters cited above, which differentiate the various forms, are probably only of specific rank, with the possible exception of the simple or branched branchiae. In the latter respect the Chilka species agrees with Manayunkia. The latter title has priority over Haplobranchus (1883), having been described by Leidy in 1858, and it may itself be regarded as a sub-division of the genus Fabricia. Habitat.—Taken on two occasions, in the extreme south-west end of the Chilka Lake. On one occasion it occupied tubes embedded in the sponge Laxosuberites lacustris, Annandale ; on the other, it was living amongst algae on the lower surfaces of rocks on the shore. It was taken both in the fresh and salt-water seasons, but the specific gravity of the water only varied from I’006-I'01I. Family SERPULIDAE. Genus Ficopomatus, gen. nov. This genus may, for the present, be defined by the following combination of characters :-— | à Modified setae present on the first thoracic segment, having blades provided with very stout teeth. Beneath the blades 1s a transverse row of more than two teeth. Uncini with relatively few teeth, the lowest of which is in the form of an elongate bifid spine. Ventral abdominal setae geniculate. Operculum fig-shaped, without any outgrowths. The modified setae of the first thoracic segment are very peculiar. They differ from those of Serpula, Hydroides and Crucigera, in having teeth on the blade, and in - having more than two stout teeth below the blade. In these genera the stout teeth are paired, whilst in Ficopomatus the largest tooth is median and unpaired. These setae also differ markedly from those of Spirorbis, Chitinopoma, etc., which have a crenulated wing below a finely serrated blade. The uncini resemble those of Poma- toceros, Spirobranchus and Pomatostegus in having few teeth, of which the lowest is bifid, but Ficopomatus differs markedly from these genera in the shape of the oper- culum and the modified setae of the first thoracic segment. Ficopomatus macrodon, sp. nov. (Plate XXX, figs. 27A-M.) Two small pieces of wood from the Cochin Backwater were covered with the interlacing tubes of this species. ‘The tubes form masses as in Hydroides norvegica, 656 M emoirs of the Indian Museum. De Gunn., or Pomatoceros triqueter, 1,. When free or erect the tube is circular in section, with a single dorsal ridge, which is usually worn away in the older part of the tube (fig. 27M, a and b). The ridge terminates over the orifice in a small sharp tooth. Where the tube is attached, it tends to flatten on the ventral surface, and spreads out two latero-ventral ridges (fig. 27M, c). This flattening of the attached tube is due to the thickening of the walls in the latero-ventral deg the bore remaining almost circular in section. A rather elongate specimen is 11 mm. in length, of which the abdomen is 8 mm., the thorax 2 mm., and the branchiae ‘5-1 mm. The width of the body is ‘5-75 mm. The thorax is composed of 7 setigerous segments, the abdomen of 43-57 segments and in addition there are 6 or 7 narrow rings forming the tapering posterior end. The specimens are all stained a deep reddish brown by pigment extracted by the preservative spirit from the wood to which they are attached, and the original colour pattern is obscured. Traces of deep blue pigment bands remain on the branchiae and thorax, and other pigment is present, so that probably the ne animals were coloured rather after the manner of Pomatoceros trigueter, L. The branchiae are 13-17 in number, 7-9 on the right side, 6-8 on the left. In addition, the left dorsal branchia forms the operculum. Each branchia has 3-5 diffuse bands of dark pigment, which spread on to the adjacent filaments. There are 18-20 pairs of rather long filaments on each branchia. The naked tip varies greatly in length, and may be longer or shorter than the filaments. The longest filaments are attached just above the middle of the stem, and those at each end are shortest. The stem of the branchia bears no other appendages. The operculum usually exceeds the branchiae in length. The stem is rather flattened (figs. 27A, 27C), and passes more or less abruptly into the swollen head. The latter varies in shape according to the manner in which it is compressed within the tube, the two extreme forms being shown in figs. 27A and 27c. It is rather fig- shaped, and tends to spread out over the branchiae, so as to protect them when withdrawn into the tube. The distal end is almost flat in some specimens, as in fig. 27C, in others it is markedly convex. There are no outgrowths either on the stem or the head of the operculum. Usually there are patches of pigment, a narrow band just beneath the swollen head being rather constant. The thorax (fig. 27A) is composed of 7 setigerous segments. The free ey of the thoracic membrane is entire, except for adventitious lobes. The Ist setigerous segment bears dorsal bundles only. They are much nearer the mid-dorsal line than those of the succeeding segments, and are lodged in small depressions, the setae being directed forwards. There are two kinds of setae in these bundles. The stouter kind, usually 3 in a bundle (figs. 27D, 27E), have very slender tips, with a series of very coarse teeth, diminishing in size towards the smooth tip. For some distance beneath these teeth the shaft is smooth. Then comes a transverse row of teeth, varying in number, but there is always present a large median tooth, with a smaller one on each side. In addition there are usually 2-4 smaller teeth on each side, but sometimes these are indistinct. These setae are r027T.| Fauna of the Chilka Lake : Polychaeta. 657 obviously homologous to the peculiar setae found in the Ist thoracic segment of many other genera of Serpulidae. They are accompanied by a number, usually four, of slender setae (fig. 27F), with finely tapering tips and minuteiy hispid edges. The dorsal setae of the 2nd—7th bundles are all capillary, in two rows, the anterior row composed of 8 slender setae resembling those in the Ist bundle (fig. 27F), the posterior row of 7 shorter capillaries, not winged, but with slightly flattened blades (fig, 276). Sometimes these setae appear quite smooth, but in other cases the blades are minutely hispid, the edge easily fraying, so as to appear serrated. The ventral bundles are composed of uncini (fig. 27H, a and 6), having 7, rarely 8 teeth. The lowest tooth is widely bifid. The abdomen is traversed by a ventral groove, shallow in front, but large and deep behind. The posterior end, in most specimens, narrows abruptly to a conical tail, terminating in two rounded lobes, between which the anus lies (fig. 27B). The abdominal parapodia project prominently as flat lobes (fig. 27J) bearing a row of uncini on the upper external margin, whilst the ventral capillaries project beneath the lower edge. The ventral capillaries are usually two or three in a bundle, one or two in the posterior segments. They project from the parapodium so far that the exposed part of each seta is usually twice as long as the part embedded in the tissues. The shaft expands suddenly towards the tip (fig. 27K), and then bends abruptly, the bent portion being serrated. In the anterior abdominal segments there are about 26 uncini in each parapo- dium. They resemble those in the thoracic segments, but have 11-12 rather slenderer teeth (fig. 271). They are also thinner and larger than the thoracic uncini, and when viewed on edge the forks of the bifid spine are smaller, In the mid-abdominal region there are 45 uncini in each foot, each with 12-14 teeth. In the posterior segments there are 6-10 uncini in each foot, and they are smaller, with fewer and finer teeth. Habitat.—Two pieces of wood, covered with the interlacing tubes of this species were found in the Cochin Backwater, near Ernakulam, on the south-west shore of the Madras Presidency, in September 1914. The masses of tubes resemble in general appearance those of Pomatoceros triqueter, L., or Hydroides norvegica, Gunn. They were accompanied by several specimens of Balanus amphitrite. The salinity of the water is probably very variable, but no precise information is available. LIST OF REFERENCES. Annandale, N., 1907.—‘“‘ The Fauna of Brackish Ponds at Port Canning, Lower Bengal.” Rec. Indian Museum, Vol. I, pp. 35 and 197. 5 „ 1915.—‘‘ A Naturalist’s view of the Chilka Lake.’ Reprinted from The Calcutta Review, 1915. Annandale, N., and Kemp, S., I915.—‘“ Fauna of the Chilka Lake. Introduction.” Mem. Indian Museum, Vol. V, p. I. 5, , and Kemp, S., 1916.—“Fauna of the Chilka Lake. Mollusca Gas- tropoda and Lamellibranchiata.”’ Mem. Indian Museum, Vol. V, Not 627 658 Memoirs of the Indian Museum. [ Vor. V, Arwidsson, I., 1898.— “Studien über die Familien Glyceridae und Goniadidae.” Bergens Museums Aarbog, 1898, No. xi. Borradaile, L. A., 1901.—"On the spawn and young of a Polychaete Worm of the Genus Marphysa.’ Proc. Zool. Soc. London, 1907, Vol. II, p. 714. Bourne, A. G., 1883.—“ On Haplobranchus, a new Genus of Capitobranchiate Annelids.”’ Quart. Journ. Microscop. Science, Vol. XXIII, p. 169. Bush, K. J., 1905.—‘“ Tubicolous Annelids of the Tribes Sabellides and Serpulides from the Pacific Ocean.” Harriman Alaska Expedition, Vol. XII. Carrazzi, D., 1893.—‘“‘ Revisione del Genere Polydora Bosc, E cenni su due specie che vivono sulle ostriche.”’ Mit. Zool. Stat. zu Neapel, Bd. XI, S. 4. Crossland, C., 1903.—‘‘On the Marine Fauna of Zanzibar and British Kast Africa. Polychaeta, Part Il.” Proc. Zool. Soc. London, Vol. II, ps 120: Ehlers, E., 1901.—“ Die Polychaeten des magellanischen und chilenischen Strandes.” Festschrift Ges. Göttingen. Berlin, 1901. Eisig, H., 1887.—“ Die Capitelliden des Golfes von Neapel’’ Fauna und Flora des Golfes von Neapel, Vol. XVI. Fauvel, P., r901.—“‘ Annélides Polychétes de la Casamance.” Bull. Soc. Linn. de Normandie (5), Tome V, p. 59. » » Igıı.— “ Annelides Polychètes du Golfe Persique.” Arch. de Zool. Expér. 5€ Ser., Tome VI, p. 353. » » 1914.—‘"Annélides Polychetes non pélagiques provenant des campagnes de l’Hirondelle et de la Princesse-Alice (1885-1910). Res. Cam- pagnes Scientifiques. Monaco. Gravier, C., 1901.— Sur trois nouveaux Polychètes d’eau douce de la Guyane Fran- gaise.” Bull. Soc. Hist. Nat. Autun, Tome XIV, p. I. Grube, E., 1867.—‘‘ Beschreibungen neuer von der ‘‘ Novara’’-Expedition mitge- brachten Anneliden.” Verhandl. zool. bot. Gesell. Wien, Bd. XVI, 1866. 5 _» 1873.—“ Die Familie der Lycoriden.” Jahresb. der Schles. Ges., 1873, p- 56. by»: 1878.—“ Annulata Semperiana.’’ Mém. Acad. Imp. de St.-Petersbourg, 7° Sér., Tome XXV, No. 8. Hansen, G. A., 1882.—“The Norwegian North-Atlantic Expedition 1876-1878.” Zoology. Annelida. Horst, R., 1909.—‘‘ On fresh-water Nereids from the Botanical Garden at Buitenzorg, belonging to Lycastis hawaiiensis, Johnson.” Bull. du Départ. de l'Agriculture aux Indes Néerlandaises, No. XXV. Johnson, H. P., 1903.—“ Fresh-water Nereids from the Pacific Coast and Hawaii, with remarks on fresh-water Polychaeta in general,’ Mark Anniver- sary Volume, Article X, p. 205. Kinberg, J. G. H., 1910.—“ Kongl. Sven. Fregatten ‘Eugenies’ Resa omk. Jorden, 1851-1853.” Zoologi III. Annulater. 1921.] Fauna of the Chilka Lake : Polychaeta. 659 Langerhans, P., 1881.—“ Ueber einige Canarische Anneliden.”’ Nova Acta d. Leop.- Carol Acad bd 42) No. 3, D 107. Leidy, J., 1884.—‘‘ Manayunkia speciosa.’ Proc. Acad. Nat. Hist. Philadelphia, 1883 (1884), p. 204. McIntosh, W. C., 1878.—“On the Annelida obtained during the cruise of H.M.S. ‘Valorous’ to Davis Strait in 213. Trans. Linn. Soc. London (2), Zool., Vol. I, p. 499. Poon Report on the Annelida Polychaeta collected by H.M.S. ‘ Challenger” during the years 1873-1876.” Challenger Reports, Zoo- logy, Vol. xIT. Marenzeller, E., 1879.—“‘Stidjapanische Anneliden, I.” Denkschr. Akad. d. wissensch. Math.-Naturw. Classe, Bd. XLI. Mesnil, F., 1896.—“ Études de morphologie externe chez les Annélides. Les Spioni- diens.” Bull. Scient. de France et de Belgique, Tome XXIX. Moore, J. P., 1909.—‘“‘ Polychaetous Annelids from Monterey Bay and San Diego, California.” Proc. Acad. Nat. Sci. Philadelphia, April, 1909, p. 235. 1911. —"* The Polychaetous Annelids dredged by the U.S.S. “ Albatross” off the coast of Southern California in 1904, III. Euphrosynidae to Goniadidae.”’ Proc. Acad. Nat. Sci. Philadelphia, April, IQII, p. 234. Saint- Joseph, Baron de, 1898.— “Les Annelides Polychetes des côtes de France.” Ann. Sci. Nat. Zool. (8), Tome V. Sluiter, C. Ph., r890.— Die Evertebraten aus der Sammlung des Königlichen Natur- wissenschaftlichen Vereins in Niederländisch Indien in Batavia.” Natuurkundig Tijdschrift voor Nederlandsch Indie, Band L, p. 102. Stephenson, J., 1908.—“‘ Preliminary description of an Oligochaete worm of uncertain position.” Rec. Ind. Mus., Vol. II, p. 39. », 1910.—" Matla bengalensis: A correction.” I/brd., Vol. V, p. 82. Willey, A., 1908.—‘‘ Description of a new species of Polychaete Worm of the genus Spon ine. Ind. Mus., Vol. II, p. 389. 3) 3) 32 3) EXPLANATION OF PLATE XIX. Ancistrosyllis constricta, Sp. nov. Fic. Ia.— Anterior end, dorsal view. X 40. IB.— Ist right foot. X 100. Ic.—2nd right foot. Setae not shown. X 100. Ip.—4oth right foot. X 100. IE.—Part of dorsal lobe of 80th right foot, posterior view. X 330. IF.—Anterior short seta from Ist foot. X 560. 1G.—Intermediate type of seta from 4th foot. x 560. Lycastis indica, sp. nov. FIG. 2A.—Anterior end, dorsal view. x 40. 2B.—Ist right foot, anterior view. x 140. 2C.—10th right foot, anterior view. x 56. 2D.—70th foot. X 100. 2E.—Iooth right foot, posterior view. Setae not shown. X 100. 2F.—Falcate heterogomph seta from upper part of 50th foot. X 560. 2G.—Hemigomph seta with long finely serrated tip, from upper part of Ioth foot. xX 560. 2H.—Heterogomph seta with long coarsely serrated tip, from middle of ıoth foot. x 560. 2J.—Tip of shaft of hemigomph seta from dorsal division of roth foot. Specimen from Cochin Backwater. x 560. . Tylonereis fauveli, sp. nov. Fic. 34.—Anterior end, dorsal view. X 30. 3B.—Posterior end, dorsal view. x 27. 3c.—Ist right foot, anterior view. x 56. 3D.—7th right foot, anterior view. X 56. » 3E.—30th right foot, anterior view. X 56. » 3F.—Tip of dorsal aciculum and gland, in 30th foot. X 330. „ 36.—7oth right foot, anterior view. x 56. » 3H.—150th foot. x 56. | » 3J.—Seta from dorsal division of 30th foot, front view. x 560. | Mem. Ind. Mus.,Vol.V. 1921. R.Southern del. DOLYSCEAETA OF Taz CHINE A ARE Eee Plate XIX. S.C Mondul lith. ‘: aa \ Fic. JEG, EXPLANATION OF PLATE XX. Dendronereis aestuarina, sp. nov. 4A.—Anterior end, dorsal view. X 17. 4B.—Everted proboscis, ventral view. X 17. 4C.—10th segment, ventral surface. X 20. 4D.—Ist right foot, front view. X 40. 4§.—10th right foot, front view. X 40. 4F.—15th right foot, front view. x 40. 4G.—Dorsal cirrus of 17th foot. X 30. 4H.—18th right foot, front view. X 25. 4J.—Ventral division of the 2oth right foot, front view. Setae not shown. X 40. R AK.— 30th right foot, front view. Setae not shown. x 25. 41. —Homogomph seta with long finely serrated terminal piece, as found in all the feet. x 330. 4M.—Homogomph seta with long coarsely serrated terminal piece, from the 3rd foot. X 560. 4N.—Homogomph seta with moderately short smooth tip, from upper divi- sion of the 6oth foot, X 330. Sternaspis costata, Marenzeller. 5A.—Ventral shield. Only the basal portion of the lateral bundles of setae shown. X 40. 5B.— Tip of dorsal seta from Ist bundle. X 330. Mem. Ind. Mus. Vol.V, 1921. Plate XX. = CIR, VE = ; f D, Ba gchi Yhulaı, POLYCHAETAOF THE CHILKA LAKE, E*. 7 m PET 0 NS M EXPLANATION OF PLATE XXI. Dendronereides heteropoda, gen. et sp. nov. Fic. 6A.—Anterior end, dorsal view. x 40. 6B.—Front view of partially extruded proboscis, showing the basal ring of papillae. 6c.—Ist right foot, front view. X I00. 6p.—4th-right foot, front view. X 100. 6E.—gth right foot, front view. X I00. 6F.—2Ist right foot, front view. X IO00O. 66.— Single bunch of gill branches from the 25th foot.x 100. 6H.—-Dorsal division of 26th right foot, front view. X 100. 67.— 30th right foot, front view. Setae not Shown. X 100. 6K.—Dorsal seta from the 14th foot. x 560. 61.— Ventral seta with long teeth, from 14th foot. X 560. 6M.—-Homogomph falcate seta from the ventral division of the 14th foot. x 560. 6n.— Tip of shaft of seta from the ventral division of the 22nd foot, show- ing the ‘hemigomph’ condition. X 560. Nereis reducta, sp. nov. .74.—Anterior end, dorsal view. x 26. 78.--Anterior end, ventral view. x 26. 7¢.—Paragnath from Group I. xX 330. 7D.—Paragnath from Group VIII. a=side view; b=surface view. X 330. 7E.-—-Ist right foot, front view. X.- 100. 7¥.—31d right foot, front view. X I00. 7G.—10th right foot, front view. x I00. 74.—60oth right foot, front view. X I00. 7).—Homogomph seta with long tapering tip, from dorsal division of 4oth foot. X 560. 7K.—Heterogomph falcate seta from lower part of ventral division of 4oth foot. x 500. Plate XXI. Mem. Ind. Mus., Vol.V.192! Dorsal. N = © S SK RS SS SS ASS IN 7C. S.C.Mondul lith. AN POLYGHAE TA OF THE GHILKA LAKE Eee R. Southern del. EXPLANATION OF PLATE, XXI. Nereis chilkaënsis, sp. nov. Fic. 8a.—Anterior end, dorsal view. x 12. +) 8B.—Anterior end, dorsal view, of a small male specimen, apparently com- mencing its metamorphosis, with partially extruded proboscis. The eyes are larger and the tentacular cirri relatively longer than in fig. U ES 20: 8C.—Anterior end, ventral view, of a small specimen. X 12. 8p.—Dorsal view of 26th and 27th segments. X 12. 8E.— Ist left foot, front view. x 56. 8F.—10th right foot, posterior view. x 56. 86.— 50th left foot, front view. x 56. 8H.—6oth right foot, posterior view, of small male specimen oh in fig. dB OMS 0! 8j.—25th left foot, posterior view, of 2 Heteronereis. X 40. 8K.—4oth left foot, posterior view, of same specimen. X 40. 81.— Igth right foot, posterior view, of ¢ Heteronereis. x 56. 8M.— 30th right foot, posterior view, of same specimen. x 56. 8N.—Homogomph seta with slender terminal piece, from the dorsal division Of the 20th foot. X 560: 8p.—Heterogomph seta with falcate terminal piece, from the ventral division of the 20th foot. x 560. 80.— Terminal piece of one of the short thick heterogomphs from the upper part of the ventral division of the 65th foot. X 560. SR.— Tip of the shaft of one of the swimming setae of the Heteronereis, from the 30th foot. X 560. Mem. Ind. Mus, Vol V,1921. Plate XXII. ay 73 = 4 pe 4 R.Southern del. A Chowdhary lith. EOEVGnARAVOR hE GHliLKA LAKE TETE: > > u Nj EXPLANATION OR PEATEZISITITE Nereis glandicincta, sp. nov. FIG. 94.—Anterior end, dorsal view. x 17. 9B.— Everted proboscis, ventral view. X 17. gc.—Papilla, with paragnath, from Group VI. x 100. 9D.—Jaw. xX 26. QE.—-Ist right foot, front view. x 56. 9F.—10th right foot, front view. x 56. 9G.—50th right foot, front view. x 56. gH.—100th right foot, front view. x 56. 9J.—Homogomph seta with long finely serrated terminal piece, from the dorsal division of the Ioth foot. X 560. 9K.—Hemigomph seta with shorter, more coarsely serrated terminal piece from the ventral division of the roth foot. x 560. 9L.—Hemigomph seta with falcate terminal piece, from the lower part of the ventral division of the 20th foot. X 560. Nereis (Perinereis) marjorii, sp. nov. . 10A.——Anterior end, dorsal view. X 12. 10B.—Anterior end, ventral view. x 12. 10C.—1st left foot, posterior view. x 56. 10D.—10th left foot, posterior view. x 56. 10E.—50th left foot, front view. x 56. 10F.—70th right foot, front view. X 56. 10G:—Heterogomph seta with falcate terminal piece, from the 1oth foot x 560. Plate Xx) Mem. Ind. Mus, Vol V1921. CLECP LE LT "HT: Vette Me chat DDC MEN re | \ | N | N \ ! Ÿ N | Pt, =——_ “tz = = 2a 7 >> nr es ee er oe ———— Le] © © rn A.Chowdharylith. R.Southern del. POLYCHAETA OF THE CHILKA LAKE,ET. A ) N v 5 N > 7 1 ‘ à rte { RARES > + x a ‘ À » Er r on = 1 Fe LA LS une = ke to dent +. ‘ ; i) ae EXPLANATION OF PLATE XXIV. Nephthys polybranchia, sp. nov. FIG. 11A.—Anterior end, dorsal view. x 56. „ 11B.—1st right foot, posterior view. X I00. 11C.—2nd right foot, posterior view, setae not fully shown. a—rudiment of branchia. X 100. 11D.—10th right foot, front view. X 70. IIE.—30th right foot, front view.- Setae not fully shown. x 56. IIF.—Seta from posterior row of 20th foot. X 330. IIG.—Seta from anterior row of 20th foot. X 560. Nephthys oligobranchia, sp. nov. FIG. 12A.—Anterior end, dorsal view. The head is distorted by the extruded proboscis. X 38. 12B.—10th right foot, front view. X 70. 12C.—21st right foot, front view. Setae not fully shown. x 70. Marphysa gravelyi, sp. nov. FIG. 13A.—Anterior end, dorsal view, of a small individual. x 18. 13B.—Posterior end, dorsal view, of a small individual. x 25. 13C.—Maxillae. x 25. 13D.—Mandibles. x 25. I3E.—Ist left foot, front view. X 100. 13F.—30th right foot, front view, of a small individual, equivalent to the 4oth foot of a large individual. X 56. 13G.—240th right foot, front view, of a large individual. x 38. Plate XXIV Mem. Ind. Mus „Vo1.V, 1921. A.Chowdhary lith. R. Southern del. . POMVCHAETA OR THE CHIEKA LAKE, E = . EXPLANATION OR PAIE Marphysa gravelyi, sp. nov. Fic. 13H. —Dorsal capillary seta from the 20th foot. x 560. , 13J.—Ventral compound seta from the 20th foot. X 560. , 13K.—Brush seta from the I6oth foot. x 560. , 13L.—Ventral hook from 50th foot. a=side view; b—front view. X 560. Diopatra variabilis, sp. nov. FIG. 14A.—Anterior end, dorsal view. x 8. » 14B.—Anterior end, ventral view. x 6. » 14¢.—Maxillae.. x 18. „ 14D.—Mandibles. x 18. » 14E.—Ist right foot, front view. xX 25. , 14F.—1Ioth right foot, front view. x 18. , 14G.—150th left foot, front view. X 40. », 14H.—Dorsal winged capillary seta from 6th foot. x 166. » 14J.—Ventral seta, without wing, from 5th foot. x 166. „ I4K.—Capillary seta from the dorsal group of the 56th foot, side view. X 166. ,, I4L.—Front view of the same seta. X 166. ,, 14M.—Ventral seta from Ist foot. X 560. , 14N.—Ventral hook from 28th foot. x 166. „ 14P.—Ventral hook from Iooth foot. Xx 166. ,, 140.—Brush seta from the 5th foot. X 560. ,, I4R.—10th right foot, front view, of a small individual. x 18. Plave DOW. Mem. Ind. Mus., Vol. V. 1921. —— _— — Ag. eee Q BI ae Nu en Se Se / : SE à um Se S A.Chowdharylith. ‚Southern del. IR MONO SERA Os THE CEIEKATLAKE EFS. = See. 7 Ben. EXPLANATIONT OEZELFUT SOSE Lumbriconereis polydesma, sp. nov. Fic. 154.—Anterior end, dorsal view. x 26. 15B.—Anterior end, ventral view. X 26. Ce Maxillae x 56; 15D.—3rd and 4th pairs of jaws, flattened. x 56. I5E.—Mandibles. x 56. I5F.—10th right foot, front view. x Ioo. 15G.—8oth right foot, front view. X 100. 15H.—300th right foot, front view. X I00. 15J.—One of the shorter capillary setae from the rothfoot. X 330. 15K.—Tip of crochet from ventral part of the 69th foot. x 569. 15L.—Capillary seta from the upper part of the 350th foot. X 330. Lumbriconereis simplex, sp. nov. . 164.—Anterior end, dorsal view. x 26. 16B.—Anterior end, ventral view. x 26. 16c.—Maxillae. x 56. 16p.—Left half of Ist pair of maxillae, with more elongate base than that shown 1m, 10e 27750: I6E.— Left half of 2nd pair of maxillae. x 56. 16F.—Mandibles. X 40. 16G.——-1st right foot, front view. X 86. 16H.—ıoth right foot, front view. x 86. 167.—100th right foot, front view, showing blood-vessels. Setae only partly indicated 7x286: 16K.—Foot from mid-body, seen from above. 161.—Short capillary seta from lower part of the roth foot. x 330. 16M.—Seta from a posterior foot. X 330. Plate XXVI. Mens mer Mus. VolvV 1921. en — RE ee) 22228227" \ Lil REP a A.Chowdharylith. R.Southern del. POLYCHAE TA OR Tee TEA GERS FIG. EXPLANATION OF PLATE XXVI. Glycera alba, Rathke, var. cochinensis, var. nov. 17A.—Anterior end, ventral view. x 40. 17B.—External aspect of inverted pharynx. X 18. 17¢.—Winged papilla of the proboscis. X 560. 17D.—Conical papilla of the proboscis. X 560. I7E.—Jaw. X 70. 17F.—4th right foot, posterior view. X 150. 17G.—10th right foot, posterior view. X 100. 17H.—40th right foot, posterior view. X 100. 17J.—Tip of shaft of compound seta from 4oth foot. X 560. Glycinde oligodon, sp. nov. . I8A.—Anterior end, ventral view. x I00. 18B.—Posterior end, dorsal view. x 70. 18c.—Transverse section of proboscis, mid-region. X 150. 18D.—Paragnath from inner dorsal row of proboscis. X 230. ISE.—Paragnath from inner ventral row, near base of proboscis. X 560. I18F.—Paragnath from outer ventral row, near base of proboscis. X 560. 18G.—Paragnath from transverse lateral ridge of proboscis. X 560. 18H.—Jaws. X 330. 187.—1st right foot, anterior view. X 330. 18K.—10th right foot, posterior view ; setae omitted. X 330. 18L,.—30th right foot, posterior view. X 150. 18M.—00th right foot, posterior view; setae omitted in part. X1150. 18N.—Dorsal seta from goth foot. X I120. 18P.—Ventral seta from 20th foot. X 560. 180.— Tip of shaft of same seta. X 1120. Scoloplos marsupialis, sp. nov. . IgA.— Anterior end, ventral view. x 56. 10B.—4th right foot, posterior view. X 100. _ 19C.—I7th right foot, posterior view. x 56. I9D.—Igth right foot, front view. x 56. 10E.—30th right foot, posterior view. X 70. 19F.—Short thick seta from 6th foot. X II20. I9G.— Tip of ventral seta from I5oth foot. X 1330. ee Plate XXVI. Mem. Ind. Mus. Vol. V, 1921 SS I Re: EXPLANATION OF PLATE XXVIII. Polydora (Carazzia) kempi, sp. nov. 20A.—Anterior end, dorsal view. x 56. 20B.—Anterior end, lateral view. x 56. 20c.—Ist right foot, front view. X 230. 20D.—4th right foot, front view. X 150. 20E.—5th right foot, front view. X 230. 20F.—Lower dorsal setae from 5th foot. X 560. 20G.—Seta from the lower part of the dorsal bundle of the 4th foot. X 800. 20H.—Seta from the middle of the ventral bundle of the 4th foot. X 800. 207.—Ventral crochet from the 8th foot. x 800. Polydora hornelli, Willey. . 2IA.—Anterior end, dorsal view. x 56. 2IB.— 5th segment, lateral view, showing arrangement of setae. 21C.—Modified setae from the 5th segment. X 360. 21D.— Ventral crochet from the 38th foot. x 560. Euclymene annandalei, sp. nov. . 22A.—Anterior end, ventral view. x 8. 22B.—Anterior end, dorsal view. x 18. 22C.—-Anterior end, lateral view. x 18. 22D.—Anterior border of the head, ventral view, showing the arrangement of the eyes. x 18. 22E.—Posterior end, ventral view. X 12. 22¥.—Anal funnel, posterior view. X 25. 22G.—Posterior part of the 16th setigerous segment, right side, showing arrangement of setae. X 40. IL. Tlaste 39 N \ \ K N Mem. Ind. Mus ‚Vol.V, 192°. ARE. ARF. S.C: Mendul lithy R.Southern. del. RPOEYGHAETA OF THE GHILKA LAKE, E78: ry ar GS! = r 1% m EXPLANATION OF PLATE XXIX. Euclymene annandalei, sp. nov. FIG. 22H.—Ventral hook from the Ist setigerous segment. X 230. 22J.—-Tip of a plume seta from the roth setigerous segment. X 560. 22K.—Ventral crochet from the 8th setigerous segment. X 560. Heteromastus similis, sp. nov. Fic. 23A.—Anterior end, side view. 0 „ 23B.— Posterior end, side view. x 56. 23C.—180th segment, dorsal view. xX 56. 23D.—190th segment, ventral view. x 56. 23E.—Igoth segment, side view. X 56. 23F.—Dorsal seta from 4th setigerous segment. X 560. 23G.—Dorsal crochet from 7th setigerous segment. X 660. 23H.—Ventral crochet from 86th segment. x 660. Barantolla sculpta, gen. et sp. nov. Fic. 24A.—Anterior end, ventral view. X 12. 24B.—Anterior end, lateral view. X 12. 24c.—Anterior end, dorsal view. XiI2. 24D.—Dorsal view of segments 115—118. X 40. 24E.—Lateral view of branchiae and dorsal lobes in the posterior segments. x TOO 24F.—Dorsal capillary seta from the 3rd foot. X 560. 24G.—Dorsal crochet from the 7th foot. X 360. 24H.— Tip of the same seta. x 870. 24J.—Dorsal crochet from 14th foot. X 360. 24K.—Dorsal crochet from 60th foot. X 360. Mem. Ind. Mus., Vol.V, 1921. AA SAAL EL 0 RASS | | ; Vi | ws Vv, À \ NY \ N 7 GLE ua ER A777, GL R.Southern del. D Bagchi lith. POLYCHAETA OF THE CHILKA LAKE, ET. PARLER. 2 NE A 3 DR A Mr Fic. EXPLANATION OF PLATE XXX. Mastobranchus indicus, sp. nov. 25A.—Anterior end, ventral view. x 8. 25B.—Dorsal capillary seta from the 12th segment (11th foot). X 485. 25C.—Tip of long crochet from the ventral division of the 12th segment (umth OOL) RS x 22120: 25D.—Ventral crochet from the 13th segment (12th foot). x 360. 25E.—Tip of the dorsal crochet from the 15th segment (14th foot). X I120. 25F.—Ventral crochet from the 50th segment. X 360. Laonome indica, sp. nov. . 26a.—Anterior end, dorsal view. x 25. 26B.—Collar segment, ventral view. x 56. 26€.—Tip of a branchia. X 56. 26D.—Dorsal capillary seta from the 2nd setigerous segment. X 660. 26E.—Spatulate dorsal setae from the 2nd setigerous segment. X 560. 26F.—Capillary seta from the 8th setigerous segment. X 660. 26G.—Uncinus from the 2nd setigerous segment. a=front view; b—side WACW 1 %X2800: 26H.— Uncinus from the 8th setigerous segment. X 800. Ficopomatus macrodon, gen. et sp. nov. . 27A.—Anterior end, dorsal view. xX 26. 27B.—Posterior end, ventral view. x 56. 272 — Opereulum from taedert side >70: 27D.—Modified seta from the Ist setigerous segment, side view. X 560. 27E.—The same seta, front view. X 560. 27¥.—Simple capillary seta from the Ist setigerous segment. X 560. 27G.—Capillary seta as found in the dorsal bundles of segments 2—6. x 560. 27H.—Uncinus from a thoracic segment. a=face view; b—side view. X 840. 27J.—Anterior abdominal parapodial lobe. 27&.—Ventral seta from an anterior abdominal segment. X 560. 271,.—Uncinus from an anterior abdominal segment. x 840. 27M.—Diagrammatic sections of the tube. a—old portion of a free tube; b=recent portion of a free tube; c—front view of the aperture of an attached tube. | Plate XXX. Mem. Ind. Mus.,Vol.V, 1921. a 4 Ss % / > Te FY ACE Ce 7 i 7 SN A A VS Ups || LALA ZH / Y fff, f / ae ll RTS ES TS Sr D.Bagchi lith. Ü TD =) fa (D) G 5 D le) 2 te POLYCHAE TA OF THE CHILKA LAKE ET. FIG. EXPLANATIONMNOP PI ARE RSI Potamilla leptochaeta, sp. nov. 28A.—Anterior end, dorsal view. X 70. 28B.-—Anterior end, lateral view. X 50. 28C.—Branchia, almost entire. X 50. 28p.—Tip of another branchia. x 50. 28E.—One of the shorter setae of the Ist thoracic segment. X 840. 28F.—A dorsal capillary seta from the 6th thoracic segment. X 840. 28G.—Spatulate setae from the 6th thoracic segment. X 840. 23H.—Cuspidate setae from the upper part of the anterior row of ventral setae of the 4th thoracic segment. X 840. 28y.—Cuspidate seta from the lower part of the anterior row of ventral setae of the 6th thoracic segment. X 840. 28K.—Avicular uncinus from the 6th thoracic segment. X 840. 281.—Side view of one of the longer capillary setae from one of the anterior abdominal segments. x 840. 28M.—Face view of one of the shorter capillary setae from one of the ante- rior abdominal segments. X 840. 28N.—Avicular uncinus from one of the anterior abdominal segments. X 840. < Fabricia (Manayunkia) spongicola, sp. nov. . 29A.—Entire animal, dorsal view. x 70. 29B.—Capillary setae from a thoracic segment. X 1050. 29€. —Ventral crochet from a thoracic segment. X 1700. 29D.—Capillary seta from an abdominal segment. X IO050. 298.—Crochets from an abdominal segment. x 870. Myriochele picta, sp. nov. . 30A.—Complete specimen (?). X 50. 30B.—Anterior end, in tube. X 50. 30C.—Lateral view of the anterior end. X 70. 30D.—Ventral view of the anterior end, showing internal structure. X 90. 30Ë.—Crochets. X 1730. 30F.—Crochets viewed from above. X 2000. j : ‘ : PATENT Mem. Ind. Mus .,Vo1l.V, 1921. NT D.Bagchi lith. POLYCHAETA OF THE CHILKA LAKE ,E%: R.Southern del. ce N re 1 fi BL as 4 Lan : J [0 n or nal Mm pec Sot PR rise pe Er yee FAUNA OF THE CHILKA LAKE ON SOME LERCHES FROM THE CHILKA LAKE. “. By 'ToKIO KABURAKI, Research Student, Imperial University, Tokio. (From the Zoological Laboratory, the Museums, Cambridge.) (With five text-figures.) CONTENTS. Piscicola olivacea, Harding RES dise Piscicola caeca, n. sp. - Se ne Pterobdella amara, n.g., n. sp. .- oe Glossosiphonia ceylanica, Harding Res Limnatis (Poecilobdella) granulosa (Savigny) ON SOME LEECHES FROM THE CHILKA LAKE. By Tokio KABURAKI. In his paper dealing with the leeches from the Chilka Lake, W. A. Harding! puts on record three species, viz. Piscicola olivacea, Glossosiphonia heterochta and Placob- della emydae. Recently. I have had the opportunity of examining a large collection of Indian leeches, in which some specimens from the same locality are included. On examination some of these proved to be of great interest on account of the hitherto undescribed species representing it. The following is a list of the species dealt with in this paper :— Piscicola olivacea, Harding. Piscicola caeca, n. Sp. Pierobdella amara, n. g., n. sp. Glossosiphoma ceylanica, Harding. Limnatis (Poecilobdella) granulosa (Savigny). Of these species Glossosiphonia ceylamca appears to be identical with the form referred by Harding to the well-known leech G. heteroclita (Linn.), but it may be distinguished from this by some different characters as will be mentioned later. No examples of Harding’s Placobdella emydae have been brought under my own observa- tion. Lu Before proceeding further, I should like to express my hearty thanks to Dr. Sir -A.E. Shipley for his suggestions in respect to the present investigation and to Dr. N. Annandale for the privilege of examining interesting material. My thanks are also due to Mr. W. A. Harding for his kind help in many respects, and to Professor J.S: Gardiner for allowing me the use of a table at the Zoological Laboratory. Piscicola olivacea, Harding. (Fig. I.) Numerous specimens of the species apparently identical with Prscicola olivacea are found included in the collection. This species generally occurs adhering to the body, or to the palate within the mouth, of fishes such as Hypolophus sephen, Tetradon reticularis and Dorosoma indica and appears to be fairly common in the Chilka Lake, it having been secured, besides the localities recorded by Harding, from several places, viz. in the neighbourhood of Mahosa, in the channel between the Satpur and Barhampur Islands, at the mouth of Barkul Bay, off Parikudh, and off the island in the bay beyond Grakala opposite to Kitrapal Village. ' Harding, W.A., 1920. Hirudinea. Fauna of the Chilka Lake. Mem. Ind. Mus., Vol. V, p. 509. 664 Memoirs of the Indian Museum. [Worn i; As Harding states, there is little more to mention in respect of the external features of this species. The body, which commonly measures 8 mm. long, inclusive of the Fic. 1.—Diagrammatic represen- tation of the organization of Pis- cicola olivacea, Harding, as seen from the dorsal side. an = anus, cg = cephalic ganglionic mass, cr=crop, in =intestine, m= mouth, oe=oesophagus, ov =ovary, ph=pharynx, sg=salivary gland, st=stomach, ¢=testis, v4= vas de- ferens. suckers, by 1 mm. across at the middle, is almost cylindrical, sometimes flattened and of nearly simi- lar breadth for its greater length, though it consider- ably tapers forwards. On some occasions it shows a slight constriction a short distance in front of the genital opening, so that there can be distinguished a neck and a trunk. The anterior sucker is of a circu- lar shape and rather less than half the posterior sucker, which is circular or oval in shape, the diame- ter being about | mm. So far as I could count with certainty, the com- plete somite is, as described by Harding, made up of fourteen rings, and the eleven somites XIII-X XIII, according to Dr. Annandale’s note, are each provided with a pair of pulsating vesicles, which in the pre- served specimens have collapsed. Dorsally situated on the anterior sucker are found two pairs of eyes which on each side lie so close to- gether as to give the appearance of a single eye. The first pair are directed obliquely forwards, while the second pair are directed obliquely backwards. The ground colour is subject to variation, gener- ally being bright or pale olive green. Both dorsally and ventrally the body appears more or less dark brown on account of irregular pigments present all over in reticular distribution, leaving only a small space free on the mid-lateral sides of each somite. As has been mentioned by Harding, the anterior sucker shows three transverse pigment bands on the dorsal surface, one band following the junction with the body, one near the anterior tip and a third and broader one between the two, in the posterior part of the sucker, which contains the eyes. Medially these bands are traversed by a longitudinal band. The posterior sucker is marked with seven pairs of radiat- ing pigment bands, corresponding to the seven som- ites of which it is composed. The epidermis, as is well known in Piscicola, presents the most primitive arrangement of cells which are not closely packed. Below the epidermis, in the parenchyma, are found enormous quantities of pigments which are widely distributed as mentioned above. 1921.] Fauna of the Chilka Lake : Leeches. 665 The glandular cells in the parenchyma are very weakly developed, and in no case have I been able to demonstrate such large and round cells as are observable in the following new species closely resembling this leech. The mouth lying nearly in the centre of the anterior sucker leads directly into the pharyngeal sheath, which reaches behind somite VIII. In it lies the pharynx which is of a short cylindrical shape and at the base receives the ducts of the salivary glands widely distributed in the anterior parts of the body. The pharynx gives rise posteriorly to a long passage, the oesophagus, which connects with the crop in somite XIII. The crop is a distensible part of the digestive organ, extending backwards nearly to the end of the trunk and ending blindly, and gives off a pair of slightly sub- divided lateral pouches in each somite. At the end of somite XVIII the crop passes dorsally into the stomach which bears four pairs of wide lateral pouches, lying within somites XIX-XXI. The intestine is a narrow passage leading from the stomach backwards to the anus which is situated dorsally just between somites XXVI and XXVII. The vascular and coelomic systems are well developed, and appear to be constructed on the same plan as in other species of Piscicola. The dorsal vessel is single and lies in the dorsal sinus, forming some dilations in its course. In no section have I been able to demonstrate the connection of the dorsal vessel with the ventral. The dorsal and ventral sinuses give off metamerically arranged transverse branches which communicate on each side with lateral sinuses and vesicles. The cephalic ganglionic mass lies for the most part in somite VI, consisting, as is usual, of the fused ganglia of the first six somites. The acetabular mass is composed of seven ganglia of the last seven somites forming the posterior sucker. Between these masses there can be found twenty-one ganglia in the ventral chain, each lying in the middle of the corresponding somite. | The male genital aperture lies in somite XI, directly leading into the tubular vestibulum, and the female aperture in somite XII. There exist six pairs of testes which are ovoid in shape and are situated inter- metamerically on either side of the crop in somites —. — ; to a short vas efferens communicating with the vas deferens, which pursues a for- wardly directed tortuous course and at the same time dilates into a thick-walled glandular canal. In the region of the male genital aperture the canal nears the median line and finally unites with its fellow of the opposite side to form a wide pas- sage, the “ prostate,” which opens into the vestibulum. The prostate is lined with an epithelium made up of high columnar cells closely packed, and is supplied with the ducts of glandular cells forming a large accumulation. The ovaries are represented by a pair of simple sacs lying in somite XIII. Not far behind the female genital aperture they unite into a short common duct which makes its way to the exterior. Ventrally occupying somite XIII is found a swelling of the body-wall in some specimens examined ; this may be regarded as representing a part of the clitellum. Each testis gives rise 666 Memoirs of the Indian Museum. VOL, Piscicola caeca, n. sp. Vs za). In his paper Harding puts on record a note upon a leech closely resembling P. olivacea in form and size but wholly destitute of eyes. To my mind, that note appears, however, to be partly applicable to the fol- lowing new genus and species. The species des- cribed below is based upon the material which has been placed by him in my hands. The three speci- mens of this new species were found attached out- side close to the junction of the skin and teeth on both upper and lower jaws of Hypolobhus sephen which was collected about eight miles S.S.-W. of Kali- dai in March, 1914. In the general collection is inclu- ded an example of this species, which was procured near the landing stage at Rambha. The body is much flattened, slender and in the middle of a nearly uniform breadth, though it is at- tenuated more anteriorly than posteriorly. The an- terior sucker is cup-shaped, alniost circular and about half as broad as the posterior sucker, which is also circular or heart-shaped, measuring about I mm. in diameter. In the specimens examined I have been ‘unable to demonstrate any trace of lateral pulsating vesicles. ‘The worm measures 13 mm. in length, in- clusive of the suckers, and about I mm. in breadth. In the preserved specimens the rings are mer- ged into irregular groups, and to obtain the correct number of them is extremely difficult. But a closer examination proves it to be grouped into twenty-one somites, each made up of fourteen rings. This species is wholly devoid of eyes. No trace of eye-like organs could be demonstrated even in sec- tions. é The ground colour is greyish white in spirit, wihout being marked with any trace of pattern. The Fic. 2.--Diagram showing the or- ete : ye . SH ER owin ganization of Piscicola caeca, n. sp, TOP 18 visible with more or less distinctness, g dorsal view. Index letters as in to the ingestion of blood. nern On the body-wall I have nothing peculiar to mention, excepting enormous quantities of unicellular glands which occur all over the body, just below the dermal musculature. These cells are round and far larger than those found in P. olivacea. The organization agrees in the main with that of the preceding species. The 1021. Fauna of the Chilka Lake : Leeches. 667 mouth is situated at about the middle of the anterior sucker, leading dorsally into the pharyngeal sheath which continues backwards through the cephalic ganglionic mass, extending into somite VIII. Within the sheath lies the pharynx which is of a short cylindrical shape and is supplied at the base with numerous ducts of the sali- vary glands. These glands ate always unicellular and represent large and round cells, which are extensively distributed in the anterior parts of the body. Posterior- ly the pharynx is continuous with the crop filled with blood. The crop represents a distensible part of the digestive tract, extending over ten somites XIII-XXII, without being divided into two branches, and giving off ten pairs of sub-divided lateral pouches which come off metamerically in these somites. Near the end of somite XVIII the crop gives rise dorsally to the stomach which bears four pairs of lateral pouches, a pair in each of somites XIX-XXII. The intestine extends through five somites, being distinguished by a slightly winding proximal part and a simple wide distal part, opening dorsally at the anus just between somites XXVI and XXVII. The coelome, though resembling that of the preceding species, is considerably reduced, and no trace of lateral pulsating vesicles could be detected even in sections. Of the nephridia I have been unable to obtain any more insight than a few ducts in sections available. | The nervous system is very closely similar in its arrangement to that observed in P. ohvacea, there being twenty-one ganglia in the ventral chain, not counting the cephalic and acetabular ganglionic masses. Each ganglion occupies a position in the middle of the corresponding somite. The cephalic ganglionic mass is, as usual, com- posed of six ganglia, chiefly situated in somite VI; the acetabular mass is made up of seven ganglia. ; The male genital aperture lies in somite XI, as it seems to me, between rings 79 and 80, leading into the tubular vestibulum. The female aperture is situated fourteen rings behind the male, in somite XII, being much smaller than the latter. The six pairs of testes exist anterior to and partly below the first six pairs of XI VTT KOT Moxey nny The testes on each side are connected by short vasa efferentia with the vas deferens which runs forwards and, after pursuing a somewhat tortuous course and dilating into a thick-walled glandular canal in somites XI and XII, opens in common with its fellow into the vestibulum as in the preceding species. Around the prostate is a large accu- mulation of glandular cells which make their way into its lumen. The ovaries represent a pair of simple sacs lying in somite XIII, usually lateral to the ventral nerve chain. Before opening to the exterior they unite into a short common duct. Br In the specimens examined there can be also observed, as in P. olivacea, a swell- ing of the body-wall, which occupies ventrally about one somite (XIII). . In spite of the entire absence of eyes and the reduction of the coelomic cavity this leech may be placed in the genus Piscicola Malm., on account of its great resemblance in external and internal features, as is evident from above. It is also the crop-pouches, appearing to lie intermetamerically in somites 668 Memoirs of the Indian Museum. [Vor. V, nearly allied to the genus Platybdella Malm., but it is distinguishable from this by the difference in the number of rings forming the complete somite as weli as in some structural respects. It is, I think, better to regard this leech as a member of Piscicola occupying a position on the border between this genus and Platybdella. Pterobdella amara, n.g., n. sp.' (Figs. 3, 4.) Some remarkably interesting examples appearing to represent a new genus and species were obtained near Kalidai and four miles E. 4 N. of Patsahanipur, from the mouth of Hypolophus sephen as well as near Manikpatna, from the mouth of Trygon uarnak. Usually they are to be found firmly adhering to the gums of their hosts.. This leech is of some resemblance in its features to Blanchard’s Piscicola elegans? described by that author from Kiu-Kiang and Vang-tse-Kiang in China. The body is depressed anteriorly and almost cylindrical posteriorly. As is seen from Fig. 3, it presents a peculiar shape, being sharp- ly divided into three distinct regions, each of which is on some occasions about one-third the entire length of the body, though the anterior region is usually shorter than the other. The two anterior regions are each provided with a pair of conspicuous lateral fin-like bodies, extending almost through- out the length of each region, while the posterior region is bare, without being marked with any trace of appendages. The anterior sucker is somewhat excentrically attached, near- ly campanulate and much smaller than the posterior sucker, which is centrally attached and represents a thick circular disc, as is the case with Pontobdella, Ichthyobdella and Ancyrobdella, with the diameter about as broad as the posterior part of the trunk. The specimens from Hypolophus are large, mea- suring about 12 mm. long, inclusive of the suckers, by about 3 mm. across at the middle, while the examples from 77ygon are much smaller, being about 10 mm. in length and 2. mm. N | _ in breadth. ae ee A correct count of rings forming the trunk is almost an sal aspect. impossibility. So far as I have examined the posterior region of the body where the rings are more or less distinct, the complete somite appears to be formed of some fourteen rings, which are merged into irregular groups. The ground colour is white, occasionally marked, according to Dr. Annandale’s ' Here I beg to express my indebtedness to Dr. H. A. Baylis of the British Museum for his sugges- tion of the generic name “ Pterobdella.” * Blanchard, R., 1896. Description de quelques Hirudinées asiatiques. Mèm. Soc. Zool. France, TX. 1921.] Fauna of the Chilka Lake : Leeches. 669 note applicable to this leech, with numerous minute spots of faint pink on the dorsal surface. The eyes are entirely absent. No trace of visual organs could be detected even in sections. The epidermis consists of a layer of cells covered with a cuticle. Among the epidermic cells are, as is usual, glandular cells for the sec- retion of mucus. Immediately beneath the epi- dermis lies the dermal musculature which is strongly developed, consisting of two sets of fibres, outer circular and inner longitudinal. The circular muscles form a continuous sheet of great thickness, while the longitudinal muscles occur in distinct and thick bundles. Widely scattered in the parenchyma are numerous glands, which represent a large and round cell as is seen in P. caeca. ; The mouth is placed in the centre of the anterior sucker, leading into the usual pharyn- geal sheath, which extends backwards into about somite IX. Within the sheath lies the pharynx, which commences as a short slender and cylin- drical tube just behind the brain. Opening into the posterior end are a pair of groups of the salivary glands, each group consisting of a bunch of numerous large gland-cells. The long oesop- hagus leads into the crop, which is produced laterally into five pairs of irregularly outbulged pouches in somites XV-XIX, the last of which connects with the stomach, without being pro- longed posteriorly into a blind sac. The stomach is provided with four pairs of simple pouches, coming off somewhat metamerically in somites XX-XXIII The intestine follows, and passes to the dorsally situated anus between somites , 1 Fic. 4.—Diagrammatic representation XXVI and XXVII. ; of the organization of Pterobdella’amara, The vascular system agrees in the main with ei ce ae dorsal side. Index let- that of Pontobdella and some others, the dorsal vessel being situated outside the dorsal sinus, but coming inside it occasionally. It becomes dilated in the anterior and middle regions of the body into a spaci- ous sac to surround the pharynx and oesophagus with its coeca. In the region of the pharynx the dorsal coeca send out on each side lateral branches, communi- cating with the ventral vessel, where the latter forms a complete loop. In some sec- 670 Memoirs of the Indian Museum. [Vor. V, tions of the middle parts of the body there can be found the connection of the dorsal vessel with the lateral vessels which in some parts give off branches running into the lateral fin-like bodies. The ventral vessel is simple for the greater part of its length usually inside the ventral sinus. Its posterior communication with the dorsal vessel could not be traced out, owing to my careless manipulation while sacrificing the body to the microtome. So far as my observation goes, the vascular system appears to be in communication, as in other Ichthyobdellids, with the coelomic cavity. The coelome represents a system of very complicated sinuses, especially in the anterior two-thirds of the body. In the anterior region there is a large sinus which surrounds the pharynx and its sheath, the brain and nerve cords, and communicates on each side with lateral sinuses. This large sinus is divided in the middle region into two sinuses, dorsal and ventral, which connect with the lateral sinuses, after uniting into a common canal. The lateral sinuses in these two regions are widely spread in the lateral fin-like bodies, forming a system of anastomoses. They com- municate with each other by metamerically arranged transverse canals just beneath the epidermis and thus form a complete circle. In the posterior region the arrange- ment of sinuses closely resembles that of Pontobdella and does not present any special features. No trace of pulsating vesicles could be detected even in sections. Judging from its arrangement and extension, the lateral sinus seems to play an important part in the respiratory function. The nephridia, though seeming to be similar in their arrangement to those of Pontobdella, were not clearly made out in sections available. The nervous system differs from that of the preceding two species in so far as the anterior ganglionic mass contains a ganglion more than ordinal. The cephalic ganglionic mass lies in somite VII, consisting, as in the case of Ancyrobdella, of seven ganglia, this is apparently due to the addition of the seventh ganglion. The acetabular mass is, as usual, composed of seven ganglia. Between these masses there are twenty ganglia which are metamerically arranged and joined by paired connectives. The ordinal position of the ganglion is in the middle of each somite. In the posterior end of the body, however, there can be noticed a slight centripetal displacement of the ganglia. The common genital aperture lies in somite XI, directly leading into the wide, upwardly directed vestibulum of an irregular contour, which receives the male duct from the front as well as the oviduct from behind. There are five pairs of testes, lying immediately in front of the five pouches of the crop; they are placed intermetamerically, as in the preceding two species, in sn, SON, Nea : : somites = >-—qx : The testes on each side communicate by short vasa efferentia with the vas deferens pursuing a tortuous course. In its course this duct becomes gradually dilated into a thick-walled glandular canal, which in front of the genital opening enters a large muscular, thick-walled and gland-covered “prostate,” which soon unites with its fellow into a common duct, passing upwards and then backwards dorsally to open into the vestibulum. 1921.] Fauna of the Chilka Lake : Leeches. 671 The female organs consist of the usual pair of small ovarian sacs which com- municate by a short narrow duct with the vestibulum. This remarkable leech is nearly allied, as is evident from the above, to Piscicola elegans from China, but a closer examination has revealed the fact that it cannot be ranged under the genus Piscicola, as dealt by Blanchard with elegans. To my mind, a minute investigation of the latter species would necessarily result in its generic dis- tinction from Piscicola. This leech is also closely allied to the genus Trachelobdella Diesing, but stands distinctly at variance from this in the general shape of the body» not to speak of other points of differences. The following are the chief characters which distinguish this new genus Pierob- della tounded on a single species. Brackish water leeches, ectoparasitic on fish. Body smooth, formed of three distinct regions, of which the anterior two are flattened and each provided with a pair of lateral fin-like bodies ; but the posterior region is cylindrical, without pulsat- ing vesicles. Anterior sucker nearly campanulate, excentrically attached ; posteriot sucker circular, disc-like, centrally attached. Without eyes. Complete somite, though still uncertain, may be said to consist of some fourteen rings which are merged into irregular groups. Crop produced into five pairs of pouches, without posterior blind sac. Male and female genital organs opening in common. Testes five paired. - Glossosiphonia ceylanica, Harding. (Fig. 5.) The material was collected by Dr. Annandale and also by Dr. F. H. Gravely from the pond in the island of Barkuda. This leech was found occasionally at- tached to the body of Rana cyanophlyctis. At a glance some examples appeared to be identical with G. heterochta, but a closer examination has revealed the fact that this is not so. After some hesitation I have referred it to Harding’s G. ceyla- nica,' which has not been adequately described. This species appears to be fairly common in India, some examples being found in the general collection. The body is nearly ovate-elliptical in contraction, and in some preserved speci- mens the head is seen separated from the trunk by a slight neck-like narrowing. No trace of papillae have I been able to demonstrate in the specimens examined. The anterior sucker lies on the ventral side of the head, within the limits of rings 1-6. The mouth occupies a position slightly anterior to the centre of the sucker. The posterior sucker is somewhat ventral in position and oval or circular in shape, the diameter measuring about I mm. The specimens measure about 8 mm. long by 2,mm. across at the middle of the body. On the dorsal side seventy-one rings are counted in front of the posterior sucker. They are grouped into twenty-seven somites, of which somites I, XXVI and XXVII ! Harding, W. A., 1909. Note on two new Leeches from Ceylon. Proc. Camb. Phil. Soc., Wol XV „Pt, IIE 672 Memoirs of the Indian Museum. [VoL. V, are uniannulate, somites II, III, XXIV and XXV biannulate; the twenty somites IV-XXIII are complete with three rings. The six eyes are disposed in two close sub- a el rows, as is the case with G. complanata. The first and small- PONT est pair lie in ring 3. The second and ics ja seas El larger pair occur in ring 4 but may be = ns shifted somewhat further back and a ee appear to extend into ring 5. The urn > third pair are usually one ring behind ey the second, that is in ring 7. mas The colour in spirit is pale buff = Tea or pale grey, being marked with some RR three longitudinal rows of dark brown NT pigment patches, one in each half of MILL the body and one median in position. xv ; The patches are arranged metameri- TR cally, marking the middle ring of each ARS somite. Harding speaks of there V1 being traces of four longitudinal brown Ey stripes on the dorsal surface, but no XVII trace of such stripes was shown in the -- specimens examined. XVIII In structural respects the bodv- 24: wall presents no noteworthy feature, XIX being constructed in a similar manner = ri to that found in other Glossosipho- sare nids. XXI The mouth lying at the level XXIT nearly between the anterior two pairs HET : of eyes leads into the pharyngeal APE Lan A ' sheath with the pharynx, which ex- aoe = tends over about four somites VIII- ARE XI. The salivary glands are closely aggregated into a compact group in FIG. 5.—Diagram showing the organization of Glosso- each half of the body, the group lying siphonia ceylanica, Harding, as seen from the dorsal i BE > side. Index letters as in fig. 1. symmetrically, within somites X and XI. The oesophagus represents a short passage connecting the base of the pharynx with the crop. The crop, differ- ing from the general form of this genus, possesses seven pairs of lateral diverticu- lae, a pair arising in the middle of each of somites XIII-XIX. All of the first six pairs are usually bilobed distally, and each of the seventh pair is reflected pos- teriorly and extends into somite XXII, giving off four secondary lateral diverticu- lae, which come off metamerically in somites XIX-XXII. The stomach is pro- TOZ T4] Fauna of the Chilka Lake : Leeches. 673 vided with its four pairs of lateral diverticulae, lying within three somites XX- XXII. Posteriorly the stomach is continuous with the intestine which passes to the anus between somites XXVI and XXVII. The vascular and coelomic systems, though not being traced out clearly, appear to be constructed on the same plan as in other Glossosiphonids. The nephridia form several coils in their central portion and open out ventrally on the middle ring of the somite. The total number could not be definitely made out. The cephalic ganglionic mass occupies two somites VI and VII, and consists, as usual, of the fused ganglia of the first six somites. Behind this there exist twenty- two distinct ganglia in the ventral chain, the last one representing the acetabular ganglionic mass. Hach ganglion usually lies in the middle of the somite, though it is slightly displaced towards either end of the body. The male genital aperture is placed between somites XI and XII. The female aperture is two rings behind the male, between the first and second rings of somite RUN: The six pairs of testes lie between the lateral diverticulae of the crop, inter- SAL EX YIIT NDS wards, and, after, forming several coils in somites XI-XIII, unit in the median plane into a thick-walled common canal, the ‘ prostate,” opening to the exterior. The ovaries are a pair of simple sacs extending over four somites XIII-XVI, and open in common as uswal. This leech may be placed in the genus Glossosiphonia, Johnson, with which it agrees in almost all of the diagnostic characters, excepting only the difference in the number of lateral diverticulae of the crop. In my opinion the difference may be regarded as being of not more than specific value. metamerically in somites The vasa deferentia on each side proceed for- Limnatis (Poecilobdella) granulosa (Savigny).' This species, as is well known, is used for blood-letting in India and is very wide in its distribution, a considerable number of examples having been recorded by several authorities from Celebes, Sumatra, Borneo, Java, Cochin, Siam, China, India, Ceylon and elsewhere. In addition to a single individual from this lake, I have examined numerous examples from various parts of India, Ceylon and Burma, which are included in the general collection. The body in the preserved condition is flattened and has a rough or granular appearance on both sides, owing to the presence of small closely-set papillae dis- posed transversely on the ring. In some cases the papillae are so small that a com- paratively smooth surface is seen. ‘This leech is of extremely varying size, the largest ! Blanchard, R., 1893. Revision des Hirudinées du Musée de Turin. Boll. Mus. Zool. Univ. d Torino, Vol. VIII, No. 146. 1897a, Hirudinées du Musée de Leyde. Notes from the Leyden Muse- um, Vol. XIX. 1897b, Hirudinées des Indes Néerlandaises. Zoologische Ergebnisse einer Reise in Niederländisch Ost-Indian herausgegebenen von Dr. Max Weber. 674 Memoirs of the Indian Museum. | [Vol ave examples measuring about 160 mm. in length in front of the faim is sucker and 20 mm. in breadth at the middle of the body. The anterior sucker presents a cup-shaped deepening on the ventral side of the head and is very rough, due to the presence of numerous minute papillae. The upper lip is divided on its inferior surface into two lobes by a longitudinal groove, at the base of which are found three small jaws arranged in the usual manner. The jaw is provided with numerous papillae and armed with a single row of numerous minute teeth. | The posterior sucker is of a circular shape, not wider than the greatest breadth of the body and also marked with some seven rows of transversely arranged papil- lae. The rings are very conspicuous, on the dorsal side counting 102 in front of the posterior sucker, of which rings 6 and 7 are fused ventrally to form the posterior boundary of the anterior sucker. The same is true of rings 8 ando. The last ring is pierced by the anus, thus showing signs of subdivision. According to Oka’s procedure! in determining the boundaries of somites, all the rings are grouped, as is usual, into twenty-seven somites, of which somites I, II, III and XX VII are uniannulate; somites IV, V, and XXVI biannulate; somites VI, VII and XXV triannulate; somites VIII and XXIV quadriannulate; the fifteen somites IX-XXIII are complete, consisting of five rings. There are five pairs of eyes, lying, as in all species of Hirudo, respectively in rings 2, 3, 4, 6 and 9. The segmental papillae are arranged, as has been stated by Blanchard, in eight dorsal and six ventral rows. | The male genital aperture is situated between rings 31 and 32, that is between the fourth and fifth rings of somite XI. The female aperture is five rings behind the male, that is between the fourth and fifth rings of somite XII. The nephridial pores in some specimens could be somewhat easily demonstrated from the exterior. There are in all seventeen pairs, the pore lying in the furrow separating the second and third rings of somites VIIIXXIV. The clitellum embraces the four somites X—XIII. This species, as has been described by Blanchard (loc. cıt., 1897b) in detail, exhibits great variation in colour and markings, so that their differences are purely individual, and not such as to authorise even the distinction of “varieties.” In the majority of cases the ground colour of the dorsal surface in spirit is a brownish olive with a slight touch of bluish grey. There are seven longitudinal rows of black patches, one median and three lateral on each side. The median stripe is sometimes of a dark brown colour and is in some cases continuous, the side of the median stripe extending almost throughout the whole length of the body. On each is seen a row of patches which mark in most instances the first and last rings of each complete ! Oka, A., 1917. Hirudinea. Zoological Result of a Tour in the Far East. Mem. Asiat. Soc. Bengal, Vol. VI. + THO VAC | | Fauna of the Chilka Lake: Leeches. 675 somite. The black patches forming the outer lateral stripes usually fall on the second and fourth rings, while those composing the inner lateral stripes are some- times entirely absent. The ventral surface is generally a dull brown or a bluish grey, sometimes marked with marginal bands of a deep colour. r = a « “ * ‘ Er r : : Le ve Spanes small en =" 1 x : ei e £ “2 ~ N r Cod io a 2 : , ig le | Wi. = MEMOIRS OF THE INDIAN MUSEUM, VoL. V. FAUNA OF THE CHILKA LAKE x am #24 0 No. 10. : | ee | = AUGUST, 1922. | i f RES PAGE The Hydrography and Invertebrate Fauna of Rambha Bay ene OF > ‘eo We) ‘ 7 Calcutta: PUBLISHED BY THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA. | _ PRINTED AT THE BAPTIST MISSION PRESS. saa | 1922. y 1 er " cw ! we wed ù à À 26 4 2 2 N KL yy Price Five Rupees. i. dy + : N | À | vai ER ng Karte * LUE e J vat ah rs j kts Mi. ‘ N ; Pu 2 x ted eee 5 fir a Ling | > 2 4 POS + LE AS N à BL 4 dss nie! Geek) Tr hire * ER ! FAUNA OF THE CHILKA LAKE. THE HYDROGRAPHY AND INVERTEBRATE FAUNA OF RAMBHA BAY IN AN ABNORMAL YEAR. = _ By R. B. Seymour SEWELL, B.A., F.A.S.B., Major, I.M.S., Surgeon Naturalist 12 to the Marine Survey of India AND : N. ANNANDALE, D.Sc., F.A.S.B., Director, Zoological Survey of India. (Plates XXXII-XLIII.) Ur PEL. VERT EL Tis = Se eee” 6 NN, 4 % CONTENTS. Page Part I. Hydrography. By R. B. SEYMOUR SEWELL 58 Ns 10070 Observations in August, 1919 a a ni on en Kan Observations in April, 1920 ai Be +h te a4 (SO Part II. Fauna. By N. ANNANDALE Porifera 35 a ie oh oh Re 0000 Coelenterata .. fe oad X a ne 9. oo Ctenophora .. is a as ao 30 222692 Polyzoa ee de ia AR, de ce .. 692 Annelida > ae oh es oe 2% 00693 Crustacea N > x fe Ba ays 1093 Insecta te ie ts ne an de .. 607 Mollusca a ay 26 Si Er es Se 1007, OBSERVATIONS IN RAMBHA BAY. The main object of this paper is twofold; firstly to describe the differences in density found in the water of the inner part of the Chilka Lake at different levels and secondly to put on record changes in the composition of the invertebrate fauna noted in the same locality, in a year in which the density of the water as a whole was considerably below normal. It was noted in August, 1919, that the water of Rambha Bay, which forms the south-western extremity of the Chilka Lake, was apparently much less salt than it had been at the same season or a little later in 1914, when Annandale and Kemp concluded their investigation of the fauna of the lake. For various reasons it was impossible to carry out another survey of the lake as a whole, but arrangements were made for observations on the bay at different seasons with Barkuda Island as our headquarters. We were unable to visit the mouth of the lake, but trusted to obtaining informa- tion from official sources as to its condition at the time. In this we have been to some extent disappointed. We have to thank the Director of Fisheries, Bengal, Bihar and Orissa, and also the Executive Engineer, Cuttack, for supplying us with such facts as were known to the departments concerned. They have spent much time and labour in doing so, but unfortunately no inspection seems to have been made at the critical period at which we believe that channel became closed, :.e. towards the end of the dry season. The mouth is known to have been open in April, 1919 and in the same month of 1920, and changes in its position are known to have occurred in the former year. We have little doubt that it was closed for a considerable period in the summer or autumn of 1919, in which low density was associated with exceptionally high water in a season of by no means excessive rainfall. We were informed by the serang of the Rajah of Kallikota’s steam launch, who is well acquainted with all parts of the lake, that the channel was entirely blocked in November, 1919, but this is contradicted by other statements. In the rainy season of 1920 conditions apparently became normal again. BART SEND ROG RAPE Ys. By R. B. SEYMOUR SEWELL. The present investigations were undertaken by me in order to supplement the work done by Dr. N. Annandale and Dr. S. W. Kemp' during their survey of the Chilka Lake in 1914 ‘The conclusions at which these investigators arrived were 1 Annandale and Kemp: “Fauna of the Chilka Lake, Introduction, Hydrography of the Lake,” Mem. Ind. Mus. pp. 5-12, Vol. V (Calcutta, 1915). 680 Memoirs of the Indian Museum. (Vor based entirely upon observations made on the surface waters, and it was felt that this was not altogether satisfactory. I therefore undertook to carry out a series of observations on the water at different levels, but as in the time at my disposal it would have been quite impossible to conduct such a hydrographic survey of the whole lake, I decided to confine my investigations to a small area, and Rambha Bay at the extreme south end was selected as being the most convenient locality. The water-samples were taken by means of an ‘Ekman’ reversing water-bottle and temperatures were simultaneously recorded by a reversing thermometer and were subsequently reduced in accordance with the formula given by W. Walfrid Ekman.' The density of the water-samples was estimated— usually on the day follow- ing their collection—by a ‘ Buchanan’ hydrometer and the findings were subsequent- ly reduced to o°C, Standard Temperature, and temperature im situ by means of Knudsen’s Tables. I have throughout adopted 25:0°C as Standard Temperature, so that my results are not directly comparable with those given by Annandale and Kemp who reduced their results to 15°C.’ The various stations--numbered consecutively—were made as nearly as possible in lines, running roughly at right angles to the main length of the bay and more or less parallel with each other, so that the results obtained could be plotted out in sections. The positions of the stations were in every case taken by means of a pris- matic compass, and, in order to fix them as accurately as possible, three observations were taken at each station of the bearings of prominent land-marks; their positions are shown in Chart 1. The results obtained by Dr. N. Annandale and Dr. S. W. Kemp in 1914 showed very clearly that there was during the course of that year a very considerable range of variation in the degree of salinity of the surface water, correlated with and in the main due to two causes, namely, the influx of fresh water from the Mahanaddi River system during the S.W. monsoon season and the influx of sea-water during the winter months. To quote their own words (loc. cit., pp. 11-12), the conclusions at which they arrived were that “the annual sequence of events, as it concerns the lake as a whole, may be stated briefly as follows : — “The floods that enter the lake at the close of the monsoon from the Mahanaddi delta expel all salt water from the northern portion, driving it through the outer channel to the sea, and are of sufficient volume to raise the level of the lake some 5 or 6 feet above the mean of the dry season. There being no outlet at the southern end, the comparatively saline water which had accumulated there is banked up by the flood, becoming, however, diluted to a considerable extent both by admixture with water from the north and by surface drainage from the land in the vicinity. Towards the end of the year the floods subside. The first effect of the alteration in ! W. Walfrid Ekman: ‘On the use of insulated water-bottles and reversing thermometers,’ Pub. de Circonstance, No. 23. (Conseil permanent Internat. pour l’explor. de la mer. Copenhagen, April, 1905.) | * Table IV (p. 710) shows the corresponding values of the different densities at 25°C and 15°C respectively. 1022; Fauna of the Chilka Lake : Observations in Rambha Bay. 681 level is that the water of low salinity, hitherto confined at the southern end, spreads further north. In course of time the level sinks to a minimum and subsequently, under suitable conditions of wind and tide, volumes of salt water enter from the sea and entirely fill the outer channel. This in 1914, had already taken place before the month of February. Under normal conditions the waters of the main area probably rise in salinity owing to successive inflows from the Bay of Bengal, until a maximum is reached in July. By August the monsoon floods have commenced, the water level rises rapidly and a repetition of the annual cycle begins.”’ In order, therefore, to obtain so far as was possible an insight into the annual changes that take place in the salinity of the water in Rambha Bay, two series of observations were carried out, in August, 1019, and in April, 1920, respectively. These months were selected as being the most likely periods in which to get the desired information, since in August the influx of fresh water would be causing a rise in level and a fall in salinity, while in April, there should be a large influx of salt water entering the lake from the sea. In the following pages I propose first to consider these two periods separately and afterwards to compare and correlate the results obtained. I. OBSERVATIONS IN AUGUST, 1910. During the early part of my visit to the Chilka Lake, from August 11th to 16th, 1919, the local conditions were more or less constant and, on the whole, the wea- ther was fine with light winds from the south and south-west; there were occasional light showers of rain and a somewhat heavier shower, accompanied by a north-easterly wind, occurred on August 16th, and lasted for about an hour. On August 17th, however, there was a heavy storm with strong wind from the north- east which caused a considerable rise in the height of the lake. The reverse effect, of a strong wind from the south-west lowering the level of the water in Rambha Bay, has been recorded by Annandale and Kemp (loc. cıt.,p. II). Unfortunately I had no means available of measuring the actual amount of the rise of leyel that occurred, but it must have been several inches, and simultaneously the waves raised by the wind must have caused a considerable disturbance and admixture of the water in the upper levels. Obviously then observations taken after this date (August 17th) are not comparable with those recorded earlier, and I have therefore considered them separately. | | Wherever it was found possible, three observations —on the surface, at 5 feet and 10 feet depth—were made at each station, but where the shallowness of water did not permit of observations at this latter depth, I took two observations, at the surface and at 6 feet depth respectively, and the results obtained are given ın extenso in the Appendix, Table 1. As my investigations were of necessity made on different days and therefore under slightly different conditions of temperature, etc., it might be argued that the results obtained are, in view of the shallowness of the water in the Bay and the varying effects of radiation and conduction of heat through the different layeıs of water, not strictly comparable with each other. I have, therefore 682 Memoirs of the Indian Museum. NOR AN in addition to giving the densities at the temperature in situ in the various sections also given corresponding sections after all densities had been reduced to standard temperature, 25:0°C. In an investigation of the physical conditions of so small an area of water it was expected that it would prove necessary somewhat to reduce the usual intervals between contour lines, and this proved to be correct especially as regards temperatures and I have in places drawn them as close as 0°25°C; but as regards the densities obtained, these were found to differ so widely and in such short distances that as a rule I have given them in 0'5 intervals. Annandale and Kemp (loc. cıt., p. 5) give the depth of water in Rambha Bay as having a maximum of 8 feet during the dry season, with a rise of 5 to 6 feet during the floods. In September, 1914, the density of water in this area was found to be 6:5 [they give it as 10065, distilled water being 1'000 (loc. cit. p. 8)] and again in Novem- ber of the same year they found the density to be 6°0, though on this occasion they state that “ water of appreciable salinity was, however, not so closely restricted to the southern area (of the lake) for a sample obtained off Kalidai' gave a reading of 1'0035 and others off Barkul! of 1'003. The flood waters had somewhat abated, with the result that the level had decreased and the saline water confined during Septem- ber at the extreme south had spread further north.” At the time of my visit to the lake the depth of water in the middle of the Bay was II feet and during September it rose still higher. On August IIth, 1910, a sample of water taken at the surface at Barkuda Island showed a density of 9°29 in situ at a temperature of 29° 9°C or of 10°78 at standard temperature. Throughout the whole period between August 11th and December I6th, 1919, there seems to have been a steady fall in the density of the surface layers (vide Appendix, Table II), and a sample taken near Breakfast Island on this latter date and sent to me for examination gave a density at standard temperature of only 2109. Allowing for such annual variations as are bound to occur owing to variations in the annual rainfall, my results up to this point seem to agree fairly well with those obtained by Annandale and Kemp in 1914. It was, however, found that the surface conditions of the water in the bay in August were by no means constant. In plate XXXVIII I have given the densities of the various water-samples taken at the surface at the temperatures that existed in situ, and it shows very clearly the way in which the density decreases as we pass from the mouth of the bay between Barkuda and Chiriya Islands in a south-westerly direction towards the village of Rambha, situated a the head of the bay to which it gives its name. This distribution of the water in the bay in 1910, is the exact opposite of that found by Annandale and Kemp in November, 1914, who record (loc. cit. p. 10) that “‘throughout the southern part of the lake the water in the middle was of lower specific gravity than that nearer the shores.’ During the period of my observations between August 11th and 18th, 1919, surface- ' As neither Kalidai or Barkul are shown on my charts I may mention that the former is an island some 8 miles away to the north-east of the entrance to Rambha Bay, while the latter lies on the main- land about 2 miles to the north of Kalidai. 1922. | Fauna of the Chilka Lake : Observations in Rambha Bay. 683 water having a density of 9:0 or over was found only at the mouth of the bay and in a small area in the southern prolongation of the head of the bay to the south-east of Rambha village; throughout the whole length of the bay the density was found to decrease steadily as we approach Rambha village, and the contour lines of equal density drawn on the plate show very clearly that this fall is due to an influx of fresh water. The map shows that at the head of the bay in the vicinity of the village two streams enter the lake and at this period of the year, and even on into Decem- ber a quantity of warm fresh water was being brought down, which flowed out into the lake over the top of the more dense water below. As this current of water passes out into the lake its own density is raised by gradual admixture with the denser water of the lake, while at the same time the salinity and consequent density of the lake-water is diminished, and the effects can be traced even as far as the mouth of the bay. The effect of this surface current of fresh and comparatively warm water is seen extremely well in sections 1 and 2, plate XXXII, at stations 14 and 20 respectively. The current runs from near Rambha village across the mouth of the southern conti- nuation of the bay towards Ganta Sila, and in August its course was plainly marked by floating leaves andscum. At this point the current alters its course and sweeping round the northern promontory of Ganta Sila appears to divide. A study of section 4, plate XXXII, shows that we have two areas of lower density situated at the two sides of the entrance to the bay, at stations 8 and 11 respectively, and, furthermore, ‘in both these areas the temperatures recorded are higher than that of the surrounding water. That the lowered density is not merely the result of this raised temperature is, however, clearly' shown in section 4, plate XX XIII, in which I have plotted out the densities after reduction to standard temperature. One area of lowered density occur on the surface at station 11, and it has a temperature of 0'2°C higher than the surrounding water, while at a depth of 6 feet the temperature is as much as 061° higher than the water at the corresponding level at station 10. This area is a direct continuation of the surface current noted above and on a calm day the course of the current could be easily traced from Ganta Sila to the south-west corner of Barkuda Island by the presence of the leaves and scum referred to above; it is obvious that any breeze from the S.W. or S.S.W. will very materially assist, even if it is not the actual determining factor, in the causation of this current. A strong breeze will cause a flow of surface water of low density out of the bay and this must natu- rally be compensated for by an influx of water from outside the bay at the deeper levels. At station 8, the second area of lowered density is found, not on the surface, but at a depth of from 5 to 6 feet, and in order to account for its presence at this depth we must first study the conditions existing in the deeper waters of the lake. One of the earliest points that attracted my attention during my survey was the extremely rapid rise in the density of the water of the bay as we pass from the surface towards the bottom. This alteration of density is most marked, as one would expect, at stations 13, 14 and 15 at the head of the bay where owing to the influx of fresh-water the surface density is markedly lowered, and in this region a re- ference to section I, plate XXXII, shows that the density in situ increases from 4°46 684 Memoirs of the Indian Museum. [Vor on the surface at station 14 to over 7:0 at 5 feet and to nearly 10:0 at 9 feet depth ; but even at the mouth of the bay we still find a very considerable rise in density as we pass from the surface to the Io-feet depth-level. If now we examine the density contour lines at the three levels—surface, 5 feet and Io feet depth— shown in plates XXXVIII, XXXIX and XL respectively, we see that there is at all levels, but especially at the Io-feet depth-level, a distinct curvature indicating a slow but steady flow of water into the bay. This is particularly noticeable at the 5-10 feet depth-levels, where the inflowing current can be traced throughout the whole length of the bay except in its southerly prolongation near Rambha village. This inflowing deep current, after passing through the comparatively narrow channel to the south-east of Samal Island, enters Rambha Bay and flows along the bottom till it strikes the shore in the neighbourhood of Chiriya Island and Ganta Sila, and is then deflected towards the surface. This deflection is well seen in section 4, plate XXXII. On reaching the surface the denser water spreads out and flows over the top of the southern branch of the surface current of low-density water that we saw was coming from the head of the bay, where the streams enter the lake, and in this way we get the formation of the area of low density at a depth of 5 feet, that we found to be present at stations. The only feature in the conditions existing during the first part of my observa- tions prior to August 17th, 1919, that remains to be considered is the area of higher density that was found in the southern extremity of the bay near Rambha village at stations 16,17 and 18. This patch of water appears to lie outside the influence of the circulation of water that was going on in the rest of the bay, and station 18 showed particularly clearly an area of raised density on the surface. I attribute the comparatively high densities found in this area to the effects of evaporation; as Annandale and Kemp have pointed out (loc. cit., p. II) “in a lagoon of the size and shallowness of the Chilka Lake evaporation must, especially in a tropical climate, be more than considerable and doubtless plays a great part in the phenomena we have been discussing.” | | I now pass on to the conditions observed after the occurrence of the storm on August 17th. During the time when my observations were made on August 18th, a wind was blowing from the south and south-west and this gradually increased in strength so that I was eventually compelled to discontinue my work. The results obtained are shown in section 3, plate XXXIII, and the first fact that strikes one is that in the centre of the bay extending from station 23 to station 25 and having a depth in the centre of about 6 feet is an area of water having a lowered density—7'0 and under— and raised temperature—29'5°C and over. It seems to me that owing to the wind the surface current coming from the southern end of the lake, which we have already studied, is modified, so that instead of dividing into two areas, the whole of the low- density water is now carried in a north-easterly direction across the bay. We have here a condition of affairs very similar to that recorded by Annandale and Kemp in 1914 (loc. cit., p. 11) except that in this case the effect on the density of the water is the 1922. | Fauna of the Chilka Lake : Observations in Rambha Bay. 685 reverse of that seen by them, because, as I have already pointed out, the water at the head of the bay in 1919 had a lower density than that in the centre. At the same time I found that the density of the lower layers of water near the bottom showed a considerable diminution below what had been found in correspond- ing situations previously, the bottom water at a depth of ro feet being now only a little over 9:0 in density 2n situ or a little over I0'0 at standard temperature of 25°C, whereas previously in sections 2 and 4 I had found that at this depth it was 10-0 or over in situ, or II’O and over at standard temperature. Furthermore, the temper- ature of the water at this depth showed an increase of more than I’o°C above that 15. 16. 17. : 18. Salinity. 29-0 € 29:5° 30:0 Temperature. emcee, --" wee, e .. © Depth in feet. 3S 11 Ve, Fic. 1.—The vertical distribution of temperature (¢) and salinity (s) at station 23, August 18, 1910. previously noted. The cause of these changes is by no means obvious, and it is probable that more than one factor has contributed towards their production; the purely mechanical effect of the waves raised by the storm must have caused a con- siderable amount of admixture between the various layers and would thus tend to cause a fall in density and rise in temperature, but if this had been the sole cause the temperature at 5 feet depth would have been higher than that at 10 feet depth which is not the case. It seems, therefore, that it must be due to an influx along the bottom of water having a higher temperature than that of the mid-water. It is possible that this flow may be merely an increase in the inflowing deep current that we have already 686 Memoıirs of the Indian Museum. [More seen to be normally taking place at the 10 foot level, brought about in order to com- pensate for the increased outflowing surface-current caused by the wind, since there seemed to be no appreciable fall in the surface level of the water of the bay; but I am inclined to think that we have here an additional inflow from a source not indi- cated in my previous observations, namely from the extensive and somewhat enclosed area lying between the mainland and Cherriakuda, Barkuda, and Samal Islands. The surface current due to the wind must have blown a quantity of water inte this area through the passages between the mainland and Cherriakuda Island and between this latter and Barkuda Island, and this must produce a corresponding outflow.' A study of the temperatures and salinities found at stations 2 and 3 in section 2, August, 1919, and again at station 23, section 3, August 1010, further seems to indi- cate that we have present in the lake at this season of the year a condition of affairs that is very similar to that which is known to exist in the Oyster ‘ Polls’ in Norway. The depth of the water in the lake is considerably less than that present in a typical “Poll’, such as the Kverne-Poll, but a comparison of the vertical distribution of temperature and salinity atthe above stations in the Chilka Lake with the distri- bution present in the Kverne-Pollin June as given by Murray and Hjort (1912, p. 226)’ reveals many features in common: and this is particularly well shown in the chart of station 23 which I give above. In both cases the water on tlıe surface is of appreci- ably higher temperature and slightly higher salinity than that present at a depth of about 4-5 feet and below this level both temperature and salinity steadily increase. The differences present in the Chilka Lake are not so marked as those in the Kverne- Poll but the conditions present are essentially similar. 2. OBSERVATIONS IN APRIL, 1920. My second visit to the Chilka Lake was from the 27th-2gth April, 1920. Throughout this period the weather was fine. Each morning there was little or no wind, but, as the day wore on, light airs sprang up from the S.S.W. and by midday a strong breeze was blowing from this direction. As the weather conditions on all three days were exactly similar, the results obtained are all comparable with each other. A very considerable change had occurred in the level of the lake since my visit in August, 1910, and the depth of water in the bay instead of being about LI feet was now only 5 feet—that is to say was 3 feet lower than the maximum depth given by Annandale and Kemp for the dry season (vide loc. cit., p. 5). In consequence it was ! From observations made in August, September, October, and December 1919, and in April June and July, 1920, Dr. Annandale tells me that he is convinced that the normal direction of the wind at Barkuda island is approximately S.W. throughout the year. This wind blows the water of the bay past the island during the day so that the water-level of the lake sinks appreciably. The breeze tends to fall in the evening, however, and is usually less strong at night. The water then returns to the bay, with the result that the level is usually higher in the morning than it is in the evening. » Murray and Hjort, “ The Depths of the Ocean ” (London, 1912). 1022] Fauna of the Chilka Lake : Observations in Rambha Bay. — 687 only possible to obtain two samples of water at the great majority of the stations namely at the surface and at a depth of four feet. As before, observations were taken in a series of traverses across the bay, and the positions of the various stations—numbered 26 to 40—are shown in plate XLI. On this occasion, the prismatic compass used for taking bearings was not so satisfac- tory as the one employed in August, 1919, and in several cases the bearings do not meet very satisfactorily, so that the positions of the stations shown on the plate are to be regarded as approximate only. The results obtained are given in extenso in the Appendix, Table III. The first point to be considered is the general density of the water in the bay, and here at the outset we find a most surprising difference from the results obtained in April, 1914. In that year Annandale and Kemp (oc. cıt., p. 8) found that the den- sity in Rambha bay was 10'0 except at the southern prolongation near Rambha village, where it was as high as 11‘0. A reference to plate XLII shows that in 1920, the surface water of the bay was everywhere less than 50, except just at the mouth and in the southern extension at the head of the bay where it was only little over 5:0. As was found to be the case in August, 1910, the surface density shows an appre- ciable range of variation in different areas. In plate XLII, I have plotted out the various densities, at the temperature im situ, of the surface samples taken at each of the stations. To the north-east of Chiriya Island is an area of higher density that appears to be slowly making its way into the bay. The source of this area is by no: means obvious for two observations taken to the north and east of Samal Island (vide samples 94 and 95, Appendix 8, Table III), show that in that region the density of water is on the whole less than that in Rambha Bay. Another area of higher density occurs in the southern extension of the head of the bay, between Rambha village and Ganta Sila, and this I attribute, as before, to evaporation. In the centre of the bay at stations 32 and 33 we find a peculiar circumscribed area of lowered density, 4°5 and under, and a similar lowering of density of the surface water was found at station 35 in the channel between Cherriakuda Island and Barkuda Is- land, though separated from the area in the centre of the bay by a band of water having a density of 4°58. A reference to the corresponding stations in section 2, plate XXXVI, in which I have shown the densities at standard temperature, shows that at station 32 the lowering of the density on the surface is due entirely to a rise in temperature, for the density at standard temperature in this area is uniform with those on each side. As to the cause of this very local rise in temperature I am un- able to offer any explanation. Turning now to the results obtained from samples taken at a depth of 4 feet, I have plotted the results, as before, in plate XLIII and here again we find the same areas of increased density both at the entrance and at the extreme head of the bay, while we still get evidence of the area of low salinity in the centre of the bay at sta- tion 32. But, unlike the surface area, the lowered density at this depth is not the result of a raised temperature. It is still clearly seen in section 2, plate XXXVI, after the density has been reduced to standard temperature. No sign of any similar lowering of 688 Memoirs of the Indian Museum. [Worn density is seen in either sections I or 3 so the condition is obviously local, and it seems to me that the most probable explanation is that there is in the neighbourhood of this station an under-water spring. If this be so, the spring appears to be an intermittent one, for no trace of it is visible in the results obtained in August, IQIQ. We now find that in the channel between Cherriakuda and Barkuda Islands there is an area of higher salinity and raised temperature. The sample was taken at 12-20 p.m. on the 17th April, 1920, when a strong breeze was blowing across the bay from the S.S.W. and it seems to me that we have here corroborative evidence of the view put forward by me above that a strong breeze from this direction in addition to causing a surface outflow from the bay also sets up a counter-current flowing along the bottom into the bay from the shallow area enclosed by Cherriakuda, Barkuda, and Samal Islands. A sample of water sent to me by Dr. Annandale in June, 1920, showed that yet another great change had taken place in the density of the water. This example was taken at I0-30 a.m. at the end of the jetty on Barkuda Island and at the time a strong breeze was blowing from the $.S.W. Unfortunately, no record of the temper- ature was taken simultaneously, but the density of the sample at standard tempera- ture was 9:66 so that it is obvious that there had been a great increase in the sali- nity of the water in the lake since April. A comparison of my observations in 1919 and 1920 with those made by Annan- dale and Kemp in 1914, shows clearly that the conditions existing in Rambha Bay during these two periods were vastly different. At the commencement of my work in August, 1919, my results agree very fairly well with those obtained five years pre- viously ; but between August, 1910, and June, 1920, changes occurred in the Lake that were quite unlike anything observed in 1914. I have already given Annandale and Kemp’s account of what, judging from their experience, is the normal sequence of events in the lake during successive seasons of the year, and a comparison with my results show that the abnormal features present during the end of 1919 and early part of 1920 are: (I) the steady and progressive diminution in the density of the lake-water long after the close of the monsoon season, so that in December the density was only a little over 2°0; (2) the steady fall in the level of the lake up to and possibly even beyond April, 1920; and (3) the subsequent rapid rise in salinity in May-June, 1920. These changes in the density and salinity of the water in the lake produced a very noticeable effect on the fauna, which Dr. Annandale has dealt with in the second part of this paper, and we are here merely concerned with the changes themselves. As I have already mentioned the height of the lake was steadily rising through- out the whole period of my survey in August, 1919, and for some weeks afterwards. As Annandale and Kemp have pointed out, by far the most potent factor in producing changes of level and reduction of density is the monsoon flood-water brought down by the Mahanaddi river-system and poured into the north end of the lake, and as a result of this influx the water in the greater part of the north-east end of the lake becomes fresh or almost fresh. According to their view (loc. cit., p. 8) “the great volume 1922. | Fauna of the Chilka Lake : Observations in Rambha Bay. 689 of silt-laden water brought down into the northern end by the branches of the Maha- naddi System had expelled all that of higher salinity—a phenomenon already noted with reference to the outer channel. It is evident that, in these parts of the lake, the changes are not due to admixture so much as to the expulsion of one volume by another.” They appear to have overlooked the possibility that the fresh-water com- ing into the lake might flow over the top of the denser water, and thus produce the diminution in the density that they found during the monsoon months, and since all their observations were confined entirely to samples taken from the surface, it is impossible to say whether the same rapid rise in density as we pass from the surface to the bottom, such as we have seen to be present in Rambha Bay in August, 1919, and to a less extent though still recognisable in April, 1920, existed at all in 1914 though I see no reason to suppose that it was not present at any rate to some degree, or that it is a phenomenon confined to the area at the south-west end of the lake. On the occasion of both my visits we obtained evidence that this bottom layer of denser water is steadily flowing into Rambha Bay from the main area of the lake. This inflow was vastly greater in August, when the lake was rising, than it was in April, and it seems probable that there are two factors concerned in its production. The first of these factors is in my opinion the steady and continuous breeze that blows throughout the greater part of the year across the lake from the south-west or south-south-west. We have already seen that the effect of this breeze is to produce a strong surface current out of the bay, and it is to this that I attribute the deep influx of water of higher density into the bay along the bottom that was occurr- ing in April, 1920. In August, 1919, however, the effect was too great to be accounted for by this agency, and was, moreover, accompanied by a rise in the surface level, and it seems to me that we have here evidence of the ‘banking up’ of the more saline water, as far as the deeper layers are concerned, such as Annandale and Kemp (loc. cit., p. II) postulated in 1914. The steady influx of fresh water from the Mahanaddi and local streams causes a correspondingly steady fall in the density of the lake water so that the minimum appears in a normal year to be reached about February when the influx of salt water from the sea takes place, and a corresponding rise in density follows, reaching its maximum about July, during which month Annandale and Kemp found the density in Rambha Bay to be as high as 145 to 15°0. During the period of my observations in 1919-1920, we have two outstanding features of great interest. As I have already mentioned between the months of August-December, 1910, the density of the water in the bay had fallen very considerably, till at the end of the year the density at standard temperature of a sample taken off Breakfast Island was only 2:19. In April, 1920, the density of the water off Barkuda Island at standard temperature had risen to more than twice this and varies from 5'5 to 6‘1. There had, therefore, been no appreciable influx of fresh water into the bay and presumably into the whole lake, and the explanation of the raised density combined with a marked lowering of the surface level that existed in April seems to be found in the absence of the normal inflow from the sea, which usually occurs in February and March, combined with 690 Memoirs of the Indian Museum. (Vor. V, extensive evaporation from the surface. As Annandale aud Kemp point out (loc. cit.,p. II). “Ina lagoon of the size and shallowness of the Chilka Lake evaporation must, especially in a tropical climate, be more than considerable and doubtless plays a great part in the phenomena we have been discussing. We have no means of estimating the exact influence of this factor, but it is not unreasonable to suppose that beyond compensating for the comparatively small amount of fresh water that comes from the Mahanaddi system in the dry season, it also plays an important part in inducing an inflow from the sea.” Any inflow from the sea must, however. de- pend on the existence of a free channel between the two areas, and although I have been unable to obtain any direct evidence that such was actually the case, the serang of the launch informed me that this channel was closed in November, 1910. PART II ENUNN By N. Annandale. The chief object of these notes is to put on record changes in the invertebrate fauna of the south-western extremity of the Chilka Lake that took place between the years 1914 and 1919-20. I have also added one or two particulars that sup- plement statements made in former parts of this volume. My observations were made in August, September, October and December, 1919, and in April and June, 1920. A few notes were also obtained in July and August of the latter year. PORIFERA. Three species of sponge were found in Rambha Bay in 1914, namely Cliona vastifica Hancock, Suberites sericeus Theile, and Laxosuberites lacustris (Annandale). All of these belong to marine genera. In the rainy season and in December, 1919, Laxosuberites lacustris was still common on the lower surface of stones in the landing stage at Barkuda, but no specimens of the other two sponges were found, and by April, 1920, the Laxosuberites had also disappeared. One of the most noteworthy changes in the fauna of the bay was the appearance in it of the freshwater sponge Spongilla alba Carter in great abundance. In 1915 I wrote of this sponge, “In the Chilka Lake its distribution is somewhat remarkable. It occurs on all the rocks of the northern region, often growing luxuriantly and covering considerable areas, and is found among loose algae in the outer channel. In sheltered inlets among the rocks its gemmules often form a scum on the surface. South of Kalidai I. it is not present in the lake, although many rocks apparently suitable for its growth are si- tuated round Rambha Bay. ... We found it growing actively and producing larvae in water of sp. gr. of 1°0065, but it cannot exist in water that never becomes fresh or practically fresh.” At the end of July, 1919, and on several occasions in the next month I found dead sticks on both the north and the south shore of Barkuda covered with dead sponge of this species containing numerous gemmules. In September and October small sponges were common on the lower surface of stones on both sides of the bay. and my nets were on several occasions completely blocked up by masses of weed 1922.] Fauna of the Chilka Lake : Observations in Rambha Bay. 607 coated with Spongilla alba. In the dry season of 1919-20 it died down, as it also did in a small pool of nearly fresh water on Barkuda, but dead sponges containing gemmules were common on the under side of stones at the edge of the lake. At the end of June, 1920, when the water had again become much salter, gemmules, which showed no tendency to germinate, were common in the same position and also on the surface of the bay. A single young sponge was, however, found on Potamogeton on June 26th. COELENTERATA. The following species of coelenterates were found in Rambha Bay in 1913 and 1914: Gyrostoma glaucum, Phytocoetes' chilkaeus, Pelocoetes exul, Halianthus limnicola, Edwardsia tinctrix, Virgularia sp., Acromitus vabanchatu, Dicyclocoryne filamentata and Bimeria fluminalis. Of these species, all of which but the Virgularia were originally described by myself, the first five are, in a wide sense, Actiniaria, the sixth is an Alcyonarian, the seventh a Rhizostomatous medusa, while the eighth and ninth are gymnoblastic Hydrozoa. Most of the species belong to marine genera and all to marine families, but Phytocoetes, Pelocoetes and Dicyclocoryne have been taken only in brackish water. In 1919-20, the only species included in this list that were observed were Halianthus limmcola, Acromitus rabanchatu, Dicyclocoryne filamentata and Bimeria fluminalis. These I will discuss in order, but first I must put on record additions to the coelenterate fauna of the bay. On August 14th, 1919, Major Sewell took, in a sample of water from the surface, a single medusa of Campanulina ceylonensis (Browne) of the typical form. This medusa was obtained in the outer channel of the lake in IgI4, but not in the main area. In the neighbourhood of Calcutta the hydroid is abundant in canals of brackish water at the beginning of and just before the rainy season and produces medusae in great abundance in water having a low specific gravity. Both hydroid and medusae disappear when the specific gravity sinks below 1'006 C. (corrected to standard temperature of 15°C) owing to increase in the rainfall. In the Gangetic delta this usually takes place about the end of July: On June 26th, 1920, numerous specimens of the medusa Phialidium cruciferum, a form hitherto known only from the outer channel of the Chilka Lake, were taken ina tow-net round Barkuda. In one of them one of the radial canals branched a short distance from the proximal extremity. The branch, which reached the edge of the umbrella, bore an imperfectly formed subsidiary manubrium. On the inner surface of the umbrella the Protozoon Trichodına was abundant. The medusae were feeding on pelagic fish-eggs, which were observed in the stomach of sevéral. 1 Mr. T. A. Stephenson has just published an important paper on the classification of the Actiniaria (Quart. Journ. Mic. Sci. LXIV, No. 256) in which he discusses the brackish-water species assigned by me to the Sagartiidae, subfamily Metridiinae. He considers these species as constituting a new family (Diadumenidae) with Sagartia schilleriana Stoliczka, as type, under the name Diadumene (gen. nov.) schilleriana. He further separates my Phytocoetes chilkaeus from P. gangeticus generically under the name Mena (gen. nov.) chilkaea. As to the generic distinction of Stoliczka’s species he is undoubt- edly right, but I am not at present prepared to follow him in the other changes suggested. 602 Memoirs of the Indian Museum. [Vor. V, Halianthus limnicola.—No sea-anemones were dredged in the rainy season of 1919, but several specimens of this species were obtained in April, 1920. They were much less active and not so contractile as those observed in the cold weather of IQI4. Acromitus rabanchatu.—This medusa was common in the bay in December, 1919, but none were observed in April or the early part of June, 1920. In the latter half of June young and half-grown individuals were common, including specimens in the Semostoman stage. Dicyclocoryne filamentata.—A young medusa was taken on the surface of the bay in August, 1919, by Major Sewell and Dr. Baini Prashad. It is about twice the size of those previously described, which were killed when just set free in an aquari- um, and agrees with them in all but two important characters. These two charac- ters are (1) that an eye-spot is well developed in the sensory mass at the base of each tentacle, and (2) that each tentacle bears, about half-way along its length, a pair of lateral branches precisely like itself. It is, therefore, probable that the adult medusa of this species has branched ambulatory tentacles and resembles Cladonema. Less important differences from the original specimens observed are that the manu- brium is narrowly bell-shaped and situated on a short peduncle, the mouth being now open in the form of a small pore, and that the umbrella expands a little towards the base. The subquadrate cross-section and minutely tuberculate surface are preserved. There is still no trace of the gonads. The medusa of this species had not been hitherto observed in the Chilka Lake, or indeed, anywhere in natural conditions, the youngest stage having been described from captive specimens. Bimeria fluminalis.—The dwarfed form of this species was common on the lower side of stones near the edge of the lake with the sponge Laxosuberites lacustris as late as December, 1910, but I was not able to find any specimens in the early part of June, 1920. By the end of that month, however, very small colonies were abun- dant on the lower side of stones on the landing-stage on Barkuda. The species is able to live for short periods in fresh water, but apparently not permanently. CTENOPHORA. The only member of this group that occurs in the bay is Pleurobrachia globosa bengalensis, which, however, in 1914 was found only when the water was fairly salt. It was not found in December or April, 1919, but was common at the end of June with Phialidium cruciferum and the young of Acromitus vabanchatu. POLYZOA. The only Polyzoon I saw in the lake in 1919-20 was Membranipora hippopus Levinsen. Loxosomatoides laevis, which was common on stones at Barkuda in 1914, was not found. Small colonies of the Membranipora were common on Potamogeton pectinatus as late as December, 1919, but I could find none in April, 1920. At the end of June in the latter year it was fairly common on Potamogeton. 1922. | Fauna of the Chilka Lake : Observations in Rambha Bay. 693 ANNELIDA. Hirudinea.—Two species of Hirudinea were found in Rambha Bay in 1914, viz. Glossosiphonia ceylonensis (Harding), and a new species of Piscicola. They have been described by Mr. W. A. Harding, whose report on the leeches, as well as further remarks by Dr. T. Kaburaki, have been already published in this volume. None were collected in 1919-20, but no special search was made for them. I saw the Piscicola in June, 1920, among algae at the end of the landing-stage at Barkuda and obtained specimens in August from a pool on the island. The Glossosiphonia was collected in 1914 in flooded country at the head of the bay and is a true freshwater species closely allied to G. heterochta (Linn.) of Europe and N. America. Oligochaeta.—The following species of oligochaete worms were taken on the shores of Rambha Bay below water-level in 1914:—Enchytraeus barkudensis Stephen- son and Pontodrilus bermudensis Beddard, f. ephippiger (Rosa). To these was added in July, 1916, Monophylephorus parvus Ditlevsen, which was found in masses of rot- ting weeds just afloat. This species has not been found since 1916, but conditions for its propagation, apart from changes in salinity, were perhaps unsuitable in 1919- 20, owing to storms which washed away the dead weeds. The other species are apparently still common, as was to be expected. Polychaeta.--Mr. R. Southern has recently published in this volume a full des- cription of the species collected in 1914. The littoral Nereidae mentioned in his paper as occurring in the main area do not seem to have been affected by the physical changes that have taken place in the lake since that date, and no change has been observed in the common species dredged off shore. CRUSTACEA. Amphipoda.—The species collected in 1914 have been worked out by Prof. Char- les Chilton, whose account has been published recently in this volume. Two am- phibious species are common as “sandhoppers’’ on the shores of Rambha Bay, namely Orchestia platensis Kroyer, and Talorchestia martensu (Weber), the former being much the more abundant of the two. Nine true aquatic species occur. Of these at least three are easily recognized and are well known to me. They are, the new species of Niphargus, Quadrivisio bengalensis Stebbing, and Grandidierella _ megnae (Giles). In saying that I am acquainted personally with the last I should state that I have not attempted to distinguish between it and a closely allied new species described in Prof. Chilton’s paper. The two forms, however, occur com- monly together. Niphargus, Quadrivisio and Grandidierella are all very abundant among algae and stones on the shore of Barkuda Island and I noticed no diminution either in their numbers or in those of the sandhoppers in 1019 or 1920. Isopoda.—The aquatic isopods collected in 1914 have not yet been identified, but I may note some further facts about the terrestrial but strictly littoral species, Ligia exotica Roux, which Prof. Chilton has already described in great detail in our report on the fauna of the lake. In the rainy season of 1914 we observed this species on the shore of Barkuda I. in great droves. They were also present in 694 Memoirs of the Indian Museum. [Wore August and for the greater part of September, 1919. About the end of that month they began to disappear and by the end of October not a single specimen could be found. In December also the species appeared to be entirely absent, but in April a few young individuals were seen on rocks and stones at the edge of the lake’. At the beginning of June the species was again becoming fairly common and fairly large, but no full-grown individuals occurred, while at the end of that month full-grown individuals were abundant. In August the individuals were as large and the droves as populous as in the same month in 1919. These phenomena are evidently seasonal and it isimprobable that they are affected by changes in the salinity of the water, for L. exotica, though it takes to water readily and is a good swimmer, habitually lives on dry land. Our former notes on its habits were made in the rainy season of 1914. Cumacea.—Only two species of Cumacea (Iphinoe sanguinea and Paradiastylis culicoides Kemp) were taken in the lake in 1914, and of these only the latter was found in Rambha Bay. In April, 1920, Major Sewell and I dredged specimens, pro- bably belonging to the /phinoe, just outside the bay. Stomatopoda.—Dr. Kemp found only one species of Stomatopod in Rambha Bay, namely Squilla scorpio Latreille. The very great majority of the specimens belonged to this var. immaculata. In 1914, many adult individuals of this variety were observed under stones on the shore of Barkuda I. and in similar situations, but none were found in 1919 or 1920. In the rainy season of the former year, how- ever, and as late as December, young specimens, an inch or more in length, were com- monly dredged from the bottom of the bay. None were obtained in April, 1920, but a few larvae were observed on the surface with Lucifer hanseni. Mysidacea.—Dr. Tattersall has recorded three species of this group as occurring in Rambha Bay. They are Rhopalopthalmus egregius Hansen, Macropsis orientalis and Potamomysis assimilis Tattersall. These species, which are easily recognized, were abundant in 1914, and equally so in 1919 and 1920. The first has been taken in the open sea and belongs to a marine genus, but the two latter are brackish-water forms occasionally found in fresh water at a considerable distance from the sea, but not as a rule in isolated bodies of water. Decapoda.—-No less than 23 sepcies of Crustacea Decapoda were taken in Ram- bha Bay in 1914, and, thanks to Dr. Kemp’s enthusiastic study, we know more about the distribution in the lake of this group than about that of any of the other larger groups of animals represented in its fauna. Of the 23 species 10 belong to the Rep- tantia, 6 are crabs in the ordinary sense of the word, 3 hermit crabs and one a burrow- ing shrimp-like form belonging to the tribe Thalassinidea. Of the six crabs, two (Ebalia malefactrix and Philyra alcockt Kemp) are small and rather scarce species dredged occasionally in 1914 from the bottom of the lake. My investigations in 1019 and 1920 were not sufficiently comprehensive to demonstrate their absence in these years beyond doubt, but Iam convinced that they were at any rate scarcer than before. A single specimen of the Philyra was dredged off Nal- bano, which lies some little distance to the north-east of Rambha Bay near the opening into the outer channel of the lake, in December, 1920. Two other species ! I have observed this also in February, April, May and June 1922. 1922. | Fauna of the Chilka Lake : Observations in Rambha Bay. 695 (Varuna hiterata (Fabr.) and Pachygrapsus propinquus de Man) are littoral, in damp weather practically amphibious, species. In 1914, they were common on the shore of Barkuda I., which I made my headquarters in 1919 and 1920 on several occa- sions for several weeks at a time. Ifound neither at any season in these years. The two remaining crabs, Scylla serrata (Forsk.) and Neptunus pelagicus (Linn.), are both free-swimming species, but Scylla serrata at any rate is often found in holes at the bottom. In the Chilka Lake both are caught for food. A fisherman I employed at Barkuda occasionally brought me the Scy//a both in the rainy season of 1919 and in the hot weather of 1920. I saw fresh claws of N. pelagicus dropped by a fishing- eagle on Barkuda in December, 1919. The three hermit-crabs (C/ibanarius padavensis de Man, C. longitarsis (de Haan) and C. olivaceus Henderson) are all conspicuous species, partly amphibious in habits, and were found in 1914 on the landing-stage at Barkuda and among the rocks at the base of Ganta Sila across the bay. In the main area of the Chilka Lake the only Gastropod shell large enough for the body of adults is that of a species of Cuma closely allied to C. carinifera (Lamarck). Subfossil shells of this mollusc, found living in Rambha Bay in 1914, are still abundant all round Barkuda, but the hermit- crabs entirely disappeared as the lake became fresh. Upogebia heterocheir Kemp, the only other representative of the Reptantia found in Rambha Bay, burrows in the soft mud in the open parts of the lake. It was if anything more abundant in April, 1920, than it was at any season in 1914. The thirteen species of Decapoda Natantia collected in Rambha Bay in 1914 be- long to five families, which it will be convenient to deal with separately to some extent. These families are the Palaemonidae, the Alphaeidae, the Atyidae, the Penaeidae and the Sergestidae. I shall discuss these not in taxonomic order but rather in reference to their habits. The Palaemonidae and Atyidae may be considered together first, as they are in the main freshwater Crustacea whose presence in the lake provides perhaps the chief lacustrine or fluviatile element in itsfauna. The Palaemonidae were represented in the bay in 1914 by three species of the freshwater genus Palaemon and by one of Uroca- ris, which is mainly marine. ‘The names of the species are Palaemon lamarrei Milne- Edwards, P. rudis Heller, P. scabriculus Heller and Urocaris indica! Kemp. The three former were seasonal visitors to the lake, which they probably entered from bodies of fresh water for the purpose of breeding. P. rudis did so regularly and in large numbers, while the other two species were rarely met with. P. rudis, indeed, was one of the common prawns of the lake in a commercial sense in the rainy season. I have no reason to think that any change took place in this respect in the monsoon of 1919, in which the prawn-fishery was said to be particularly good. The specimens of prawns I saw in Rambha Bay at this season seemed to me to be P. rudis. The habits of Urocaris indica, which has also been taken in the sea, are different. Itisa 1 Dr. Kemp has recently revised the subfamily to which this species belongs and considers that it should be known as Periclimenes (Periclimenes) indicus. See Rec. Ind. Mus. XXIV, p. 144 (1922). 696 Memoirs of the Indian Museum. More permanent inhabitant of the thickets of Potamogeton pectinatus that occur all over the Chilka Lake. What I take to be this prawn was common in these thickets in Ram- bha Bay in the rainy season of 1919, and I have no reason to think that it is less so now. The two species of Atyidae found in Rambha Bay in 1914 were Caridina nilotica (Roux) and C. propinqua de Man. Their genus is fluviatile and lacustrine, but both species have been found at more than one locality in brackish as well as fresh water. C. nilotica is represented in India by a race that was called bengalensis by de Man! in 1908, but Dr. Kemp’ has recently shown that this race cannot be separated from the same author’s var. gracilipes from Celebes and China. Both C. milotica gracilipes and C. propinqua are still abundant in the Potamogeton thickets of Rambha Bay. The family Alpheidae was represented in Rambha Bay in 1914 by two species ‚of Alpheus, namely A. crassimanus Heller and A. paludicola Kemp, the former a marine species, the latter only known from brackish water. In 1910 and in April, 1920, A. paludicola was dredged in considerable numbers in the bay with Upogebia heterocheir, which it resembles in habits. A. crassimanus was not obtained, but in 1914 this species was less abundant and less liable to be caught in our nets than the other. | As regards the Penaeidae my observations are very incomplete. This is the more unfortunate as we obtained evidence in 1914 that these prawns did not breed in the lake. Four species were obtained in the bay in 1914, viz. Penaeus carinatus Dana (de Man), P. indicus Milne-Edwards, Peneopsis monoceros (Fabr.) and P. dob- sont Miers. I am not sufficiently well acquainted with these species to distinguish them in the field and can only say that Penaeidae were found in the bay in the cold weather of 1919-20. The last Decapod Crustacean to be mentioned is the little Sergestid Lucifer han- sent Nobili, the only completely pelagic species of its class found in the Chilka Lake. It was enormously abundant on the surface of Rambha Bay in 1914 on certain occasions but could not always be obtained. Its appearance and disappearance were not correlated with changes in season or salinity. This was observed also in the rainy season of 1919, in December of the same year and in April and June, 1920. There was, however, certainly no diminution in the abundance of the species on the whole. The genus Lucifer is marine, but L. hanseni is known to be very tolerant of changes in salinity. It will be evident from what has been said that there was a considerable reduc- tion in the Crustacean fauna of Rambha Bay in the dry season of 1919-20 as com- pared with that of 1914-15, doubtless in correlation with the reduction of the salinity of water demonstrated by Major Sewell. Some species, moreover, proved more tol- erant than others and while, as might be expected, those of freshwater origin suf- fered less than those belonging to marine families and genera, it is somewhat remark- ! de Man, Rec. Ind. Mus. II, p. 265, pl. xx, figs. 6, 6a, 6b (1908). * Kemp, Mem. As. Soc. Bengal V1, p. 275 (1918). 1922.] Fauna of the Chilka Lake : Observations in Rambha Bay. 697 able that the amphibious and littoral forms seem to have been affected more than any others. It is also interesting to note that none of the pelagic species (except those only found previously when the water was at its saltest) have apparently diminished in numbers, though one of the most abundant (Lucifer hansent) belongs to a marine genus. I have been able to obtain no evidence of any fresh invasion of lacustrine or fluviatile species to take the place of those that have died out. INSECTA. Ephemeroptera.—A swarm of large may-flies was observed on the surface of the lake off Barkuda on Sept. 23, 1920. They were, however, fully mature and may have emerged from some body of fresh water in the neighbourhood and have been blown out on to the lake. They were evidently in a moribund condition. No members of this order were observed in the bay in IQI4. Odonata.— The dragon-fly Pseudagrion microcephalum (Ramb.) was as common in I9Ig--20 in Rambha Bay as it was in all parts of the lake in 1914. It was observed to breed in the lake in preference to the pond on Barkuda. The imago is common at all seasons, but particularly so at the beginning of the rains. The nymph emerges in the evening, as a rule on masses of dead weed on the shore, and there undergoes its final metamorphosis. Numbers of teneral adults, only able to flutter, can often be seen in the early morning at the edge of the water. Anax guttatus (Burm.) also breeds in the lake and probably Brachythemis contaminata Brauer, which skims along the shore in the evening, catching the sandhoppers (Orchestia platensis) as they leap into the air. Rhynchota.—Eurates formidabilis Distant, was just as common on the surface of the bay in 1920 as it was in 1914. A single winged specimen of Gerris spinolae was observed from the landing-stage at Barkuda on the first day of the rains, June 11th. It was first noticed on a stone at the margin and may have just alighted from flight. Diptera.—Anopheles rossi Giles, as in 1914, was the only mosquito found breed- ing in the lake in 1919-20. Imagines were enormously abundant on Barkuda, ap- parently coming from the lake, in the latter part of the rains of 1019 and the hot weather of 1920. Fortunately they are very easily affected by movements of the air and do not become really active till fairly late in the evening. In the day-time they hide in hollow trees, in crannies among the rocks and in corners of badly-ven- tilated rooms. MOLLUSCA. \ Twenty species of living Mollusca were taken in Rambha Bay in 1914, 12 species belonging to the Gastropoda and 8 to the Pelecypoda. Among the gastropods no less than 8 families are represented, and among the bivalves 6 families. Of the 20 families only one (the Hydrobiidae) is not essentially marine, and Stenothyra, the genus present, is essentially estuarine. The Mollusca, therefore, afford particularly good material for estimating the results of changes in salinity of the water of the Chilka Lake on the fauna, the more so as many of the species were extremely abun- dant in 1914 and are easily recognized. 698 Memoırs of the Indian Museum. Vor OY, Unfortunately several changes in nomenclature, affecting both genera and species, have been rendered necessary by further investigations undertaken since the publica- tion by Dr. Kemp and myself of our report on the Mollusca in this volume (1916). These changes I will explain in footnotes. Gastropoda.—Two of the species found in Rambha Bay in 1914, Cuthona henrici Eliot (fam. Aeolidae) and Litiopa kempi Preston (fam. Litiopidae) were of such rare and sporadic occurrence that they need not be considered further. Neither was found in 1919-20. At the end of this paper Dr. Baini Prashad and I describe a Nudibranch mollusc of the family Hermaeidae discovered in the bay in June, 1920. The Opisthobranchia, apart from the Cuthona, were represented in 1914-15 by a single species (Tornatina estriata Preston, fam. Tornatinidae). This in 1914 was one of the commonest molluscs all over the lake. In August, 1910, I dredged numerous dead shells of this species, especially between Cherriakuda and the main- land. Among them I found two exceptionally small living individuals. In Septem- ber and December of the same year I still found dead shells, but no living speci- mens, and in April, 1920, even empty shells seemed to have disappeared. The Nassidae are represented in the bay by two very abundant species, Nassa denegabilis and N. orissaensis Preston. These were among the most abundant mol- luscs in 1914, and in 1920 showed no marked diminution in numbers. By far the largest living gastropod in the main area of the lake in 1914 was a species (Cuma disjuncta) closely allied to C. carinifera (Lamarck)' of the family Muri- cidae. It was not uncommon on the landing-stage at Barkuda and among the rocks at the base of Ganta Sila, where the egg-capsules were found in February. Subfossil shells are still abundant on the shore at Barkuda but living individuals have com- pletely disappeared. The Cerithiidae, though many species live in brackish water, were represented in 1014 in the main area of the Chilka Lake by a single species, Potamides cingulata (Gmelin). This species was found living at only one spot in Rambha Bay, viz. in a ditch at Rambha. I have not been able to find it at this spot recently, but con- siderable changes have been made artificially in the ditch. A small elongate shell from Rambha Bay, the true systematic position of which s still somewhat doubtful, was placed in the genus Vanesia and the family Turri- tellidae by Preston. Living specimens were found, occasionally in some abundance, in 1914. In 1919-20 only dead shells were obtained. Preston recorded four species of the genus Séenothyra (family Hydrobiidae) as occurring in Rambha Bay. A re-examination ot his material has convinced Dr. Baini Prashad and myself that great confusion prevails in his identifications. We believe that the following species actually occurred in the Bay in 1914. Steno- 1 Identified by Preston as Thais carinifera, but really distinct. I have described it in the Memovrs of the Asiatic Society of Bengal VII, pp. 266-8, fig. 2 (1922). 2 Named Potamides (Typanotonos) fluviatilis Pott. and Mich. by Preston. The name given above seems to be the correct one, as Dr. Baini Prashad has demonstrated to me. 3 Annandale and Prashad, Rec. Ind. Mus. XXII, pp. 121-136, pl. Xvi (1921). 1922. | Fauna of the Chilka Lake : Observations in Rambha Bay. 699 thyra blanfordiana Nevill, S. minima (Sowerby) and S. (Astenothyra) miliacea (Nevill). No change in the numbers of these abundant forms, comparative or actual, was observed in April, 1920. Pelecypoda.—All the bivalve molluscs taken in Rambha Bay in 1914 were abun- dant forms. The Mytilidae were represented by two closely allied species of Modiola, M. undu- lata (Dunker), which was found attached to algae and other floating objects, and M. striatula Benson, which was taken on stones and other solid fixed objects. Both were extremely abundant, as they still are. Practically every stone at the edge of the bay, if not buried in mud, has several or many individuals of M. striatula fixed to it tightly, while in stormy weather innumerable shells of M. undulata are washed ashore. In suitable situations masses of living individuals are found attached to weeds. There has certainly been no diminution in the numbers of these species. M. striatula, an extremely plastic species, is known to be very tolerant of changes in salinity and allied forms are found in freshwater lakes in the centre of China. The Veneridae include certain species of the genus Meretrix that are eminently characteristic of estuarine and similar waters and M. casta Chemnitz is abundant in a subfossil condition with Ayca granosa (Linn.) at several places on the shores of Rambha Bay. The only species of the family found living in the bay in 1914 was, however, one of the genus Clementia, which is also characteristic of brackish water but differs from Meretrix in having an extremely delicate shell. This species (C. annandalei Preston) was one of the commonest molluscs all over the main area of the Chilka Lake. In September, 1919, I found many dead shells and a few living in- dividuals, the latter all of small size. In December of the same year only dead shells were found, while in the following April very few even of these were left. The Solenidae, a marine family with one freshwater genus, Novaculina Benson. were also represented by a single species, which was identified by Mr. Preston as possibly a form of Solen fonesi Dunker. We, therefore, referred to it as ? Solen fon- est, being doubtful as to the identification. Dr. Ekendranath Ghosh ' has recently- shown that it belongs to a new genus and species, which he has called Neosolen aquae- dulcioris. This was another very common mollusc in the bay in 1914, and nearly every haul of our nets brought up, if not complete specimens, at any rate fragments of the very characteristic double siphon, which is segmented and very readily cast off. In the rainy season of 1919 no specimens were obtained ; in December a couple of broken siphons were observed, while in April, 1920, no specimens of any kind were seen. Theora opalina (Hinds), belonging to the family Scrobiculariidae, was the most abundant of all the bottom-haunting molluscs of the bay in 1914. It survived in greatly diminished numbers at any rate till April, 1920, when a few small individuals were taken with many dead shells. The Cuspidariidae also were represented in 1914 by one abundant species, Cuspi- ! Ghosh, Rec. Ind. Mus. XIX, p. 57 (1929). 700 Memoirs of the Indian Museum. [VOLE Y, daria annandalei Preston. It was usually taken with Modiola undulata on weeds or algae. Until December, 1910, it was still common, but in April only a few individuals were found alive. The two remaining species, or nominal species, of the Pelecypoda taken in 1914 belong to the family Anatinidae and are doubtfully distinct. Mr. Preston has named them Anatina barkudensis and A. barkulensis. These molluscs can as a rule be secured alive only by digging in sand when the lake is low, 7.e. in the cold season and the hot weather. None were obtained in 1919-20, but no special search was made for them. Fairly fresh shells were observed on Cherriakuda in April, 1920. Considering the molluscs as a whole, we observed in 1914 that, in them as in other groups of animals, changes in the physiography and hydrography of the lake had a selective influence. This was also so in 1919-1920. Some species have disappeared, while others have survived without decrease of reproductive vigour. ‘The species moreover, that have survived are not always those that might have been expected to do so à priori. Those of Modiola are, we know, tolerant, but Nassa is mainly a marine genus. The changes observed in the invertebrate fauna of Rambha Bay in 1919-20 were doubtless correlated with the occurrence of abnormally low salinity of the water, and were almost certainly of a temporary nature. It is improbable that all individuals of the species that were not observed in these years had perished, and, indeed, observations made in June, July and August, 1920, proved that several of the more conspicuous forms were already recovering their old status.' It will be convenient to add here the description of a new species of Nudibranch- molluse discovered in Rambha Bay by Dr. Baini Prashad and myself. Stiliger pica Annandale & Prashad, sp. nov. The molluse we call Stiliger pica is a very small aeolidiform Hermaeid with simple ungrooved rhinophores and without other tentacular processes on the head or foot. The total length of the living animal in an expanded condition was 8°7 mm. and the breath of the foot 15mm. The foot was long and narrow with parallel sides and sharply pointed but not produced into a filamentous process behind. In front it was somewhat expanded and its antero-lateral angles formed small, broad, blunt lobes separated by a deep emargination in front. The snout was truncate as a whole and depressed over the front of the foot ; its anterior margin was separated into two lobes like those of the foot but smaller. The dorsal surface of the anterior part of the head was deeply concave. The rhinophores were long, slender, pointed and filiform; the eyes small but distinct and situated immediately behind the bases of the rhinophores. The cerata were relatively long and slender, almost filiform and sharply pointed at the tip. The longest were of about the same length as the rhinophores. ‘They were arranged in two bands, one running down each side of the notum froma little behind ! In October, 1920, Pachygrapsus propinguus and a species of hermit-crab reappeared at Barkuda. 1922. | | Fauna of the Chilka Lake : Observations ın Rambha Bay. 701 the eye to the posterior extremity. Each band of cerata consisted approximately of | three or four rows, but the arrangement was irregular. As a rule the longest cerata were those of the innermost row. Some very small cerata, broader in proportion than others, were situated in the anterior third of each band. The central region of RES TETE a FÉDÉRAL NO Kar eet re CS = RÉ RARES NE Bee Fic. 1.—Photographs of the ventral and dorsal views of the type-specimen of Stiliger pica Annandale & Prashad (magnified). the notum was bare and the anus opened on this region in the anterior third of the body. ‘The sides were vertical and a little concave. The male pore lay below and a short distance behind the right tentacle, the female pore immediately behind ib The penis was observed extruded in the living specimen. Itwas a slender organ about as long as one of the rhinophores and termina- ting in a slightly asymmetrical cup-shaped structure. The stylet was not observed in the living animal. In life the back and sides were dull olive-green tinged with black between the rhinophores and obscurely ver- miculated on the antero-lateral regions. There was a small pale spot round the anus. The sole of the foot was white, clouded with olive-green. The tip of the snout, rhinophores and a broad stripe extending back- wards from their bases on either side to a point a little behind the eye, which was included, were translucent white with minute specks. The cerata were sooty black, obscurely speckled with grey, and had conspicuous white tips. The animal was killed with boiling formalin (5 % sol.) and has retained the form and colouration described, ees ot ee except that the foot and head are slightly contracted diverticulum. and the latter somewhat depressed. With a single minute specimen in our hands we have not been able to make out details of the internal anatomy. The genital system in general agrees with Bergh’s 702 Memoirs of the Indian Museum. [Vor V>7r922] description and figures of Stüliger mariae (M. & M.), but the stylet of the penis is extremely short, slender and minute. The cup-like structure of the tip is with- drawn in a strongly muscular sheath. The alimentary system is also of the same type as in S. mariae and the radula is closely similar. The hepatic diverticula in the cerata (fig. 2) are not much coiled and bear only a few short branches. S. pica differs from most species of Stiliger in its slender cerata, but otherwise seems to be a typical member of the genus. The only other species described from brackish water is S. tentaculatus Eliot. This peculiar form differs greatly from other species in having well-developed oral tentacles and also tentacle-like processes at the antero-dorsal angles of the foot. The only specimen of S. pica we have seen was found in the latter part of June, 1920, in a small aquarium containing stones from the end of the landing-stage at Barkuda in the Chilka Lake. These stones were covered with both green and dark purple algae and it was not until they had been under observation for some days that the mollusc was observed, owing largely to the chance that it had strayed into a mass of green filamentous algae in which its dark colouration rendered it very conspicuous. The number of Nudibranch molluscs known from brackish water in the Oriental Region is still small, but is increasing with our knowledge of the estuarine faunas and of those of maritime lakes. The following list may be of interest :— Tribe Cladohepatica. Family Aeolididae. Cuthona annandaleı Eliot; Gangetic Delta. Rec. Ind. Mus. V, pp. 248, 249, pl. xix (1910). Cuthona henrici Eliot ; Chilka Lake, east coast of India. Mem. Ind. Mus. V, p. 377, fig. I (1916). Family Hermaeidae. Stiliger tentaculata Eliot ; Talé Sap, Gulf of Siam. Mem. As. Soc. Bengal VI, pp. 179-182, figs. I, 2 (1917). Stiliger pica, sp. nov., Chilka Lake. Family Elysiidae. | Elysia chilkensis Eliot ; Chilka Lake. Mem. Ind, Mus. V, p. 378 (1916). 1 Semper’s Reisen in Phillipinen, Wiss. Res. Th. II, Cd. II, pp. 139-144, pl. xxvi, figs. 1-17 (1870- 1875). See also Eliot, Brit. Nud. Moll, pt. VIII (suppl. pp. 136-137). According to Eliot, the species should be known as S. bellulus M. & M. [VoL. V, Memoirs of the Indian Museum. 704 0S.6 b¥.or | @z.pr | ,S0.0€ 94.8 | 606.6 |SoZ.£r | „o.22 08.6 66.01 | Zg.bı | ,Sg.0€ 96.6 || Hour | HOLS 06.6 os | Loc | OT 99.6 gg.or | EZ.VI 9.0€ 00.01 | 96.01 | 99-¥1 | SZ.0€ 6p.6 69.01 | SG.br | 6.62 G9.6 SOx | OS Ae | Ode LS.8 Oo | AAV | oué BOO Con oe AGE L9.6 90.11 | OC6.PI | 6.0€ 98-8 98.01 | OZ.VI | „o.IE HOw | men! Eon: | Se zg.0 €g.or | o2.ÿ1 | .o.ıE 0£.g zb.or | Sı.vı | ,SS.6z TO.On | Coco | CO Gr | Oars 62.6 OW | Loi || gee Z0.Q +o.or | Sg.£r | S6.Zz 12.6 Veda | Ko | whee 92.8 Sore | ze | ar nis UL | “Jost FE | “Do 1” Fe | | Sump SEVEN ordures. “ordures Jo AJısuaq jo eınye -rodmoy, 906.8 z£.0 61.6 12.6 91.6 ÿ1.0 02.6 Gz.6 846.8 zv6.9 cz 6 g1.6 L1.6 zr.6 C£v.6 ‘JoJomorpAH s.ueueygong Aq punoy se apdures jo Ayısusq € z en ae 4 I „SgI.gZ | 1997 OI J 7 „261 puejs] eArıyy 0824-82 | 3297 S |\ 3007 11€ |,02 .12 ‘SI “yreq ‘Ayyol „0.62 90eFING | ‚ob „IZE ‘ISI 15874801 ene Se | 399} a ; | ‚OP LT PUEISL BAD 084.8 | 309] ‘17 Zor) er | pues] epnyıeg ‘Ayo ( „0.62 panes | ‘Ob „282 “IST Isergeosg SL2.g2 | 3997 9 °° ‚0 008 Puejs] eArıyy 499] OT)| ‚02.22 [sI ‘4184 ‘Ag f „0.62 »ejıng ‚ob „poE ‘IST 4sergessg { "= (0% OI BIIG ejuen) SEES aan coulizos Bean 00.2€ | : s.ueuegaong Sınyesadus] "noye} sem apdımes HOIgA wolf de ‘HOT}ISOg Fauna of the Chilka Lake : 1922. | ‘6161 ‘429999 pun Ismony U99M2Q Avg vygmoy ur aovfans ayy ouf woyvg 1990 M [0 Aqisuaq ayy SuımoyS— "II AIAV\, = “SUL 062 uoneISeyquiey € . “bogsoi;| po.S Vz.9 LL.6 L.ve 6£.€ ol -62 | Part "80€ purysy ss ss zg © Ajipeseqs BPNAILIOUT JO pua4sam 2 ve UOON ZI = 229914 | 9S.v zl.S 02.6 g.IE Gg.€ 0.82 |ooeyins | ‘och purysy x se Ig = epnyieg JO PUS JS ‘I à 0812 BIg eyuen) ‘€ £oz 66 ce 0g eS.r 89.S 91.6 G.1€ zL.€ 9.987 | Par h nor 2 ? Mat eet ° 13249 eyquwey ‘z €€ | ue or-ı1 6+. 65.5 | Eure €9.€ 08.87 |°08FINS “QZ PUIST « ce 64 PNAEILIOUTD JO pus ‘M ‘I "+ €6 uoryeis eqquex ‘€ À à = ann ae Re £ | "622 eig mueg 'z 26 es | | x She ï EE "* (Ogi puejsj BAÂHIU)D “I s | "LE PUEISI > er ur6<| epnyıeg Jo pus yseq €. PA 3 ce ce S gs. 29.5 CO || TE Go.€ 69-92 qth || Sb eng eyueg :z Toes NUE NOETT gL oS ** “GT purysy ey “1 = “9S PUEIST 2 epnyieg Jo pue ‘Wy € È 84.Y o£. 61.6 of TE L9.€ oe ge |ooeyans |\ 377 - EEE PUEISI epny oe [wre Sp-or se 5 GL EN -BIIIOU) JO pus JS2M “Z = "+ (942 purysy eAıyg ‘I & °° „zIE pues] epny | Ss oL.b 85.5 60.6 Aue 19.€ of: | Joos ÿ |\ ‘ur 6 }| -ermıoyy jo pue sem ‘€ Be ume oa i ie vl Oe ae en WIC. v1.9 69.6 LAS 6€.€ 02-82 |eJıns\f ‘37 S 0982 voryegseyquey ‘2 ss s € S Ae „222 PUEISI erg) I os FIR : : .££/z woryeys eyqmery ‘€ ce « > . ee ern | 4 4 "= Obz ‘ISI wseysqvarg 2 Ve jue co | 2 on TS | 90. 86. 0 I. ae RO “IES |. Cre puerst EAE x f | = 69.¥ 59.5 80.6 ACTE rg.E 50-92 | yor + UT cf à CEz nn . \ (Pas ee : ue of 09. 25.5 66.8 GE 99.€ 02-82 | 90RJING | “33 S | be {302 purysy Arty ‘1 f 69 OO © . abe é 29.» bS.S 10.6 0.1€ vL.E Gz.gz | Joop V (W Re HL One ERFEIT FE : \ RE a : v89 S 5 y S Gzz BIS eyuen “Zz gz ['ure Sz-6 9S.p 09.5 16.8 COTE oL.E, „Sz.ge |BeJıng f ie 1 if Lz nıdy | 89 | 2 ns u | Io-5C YE} 9,0 Ye wae Jojomo1pAH en Naeh SU 3 ae s ueuegsng | >fdures x es ‘19JEM ‘H01J89S r "SAAyNTA 5 Ba Aq punoy sup RR jo “pue’y jo ssursesg jo “moy eme ee opdınes jo Aısuaq = a | 22 us ye Joye OI }d°€ zaqgunNn “eq D soda | 9 Ame | 93? | endo 2 ‘0201 “udp ‘zh-gz SUOWPIS 10 uayv} SUOYVALISQO d1yg¢vAsoApK g— III FIEVL 708 Observations in Rambha Bay. 709 Fauna of the Chilka Lake : 1922. | WSS mol] 229914 SU011S "Zur -U2US21} 229914 “ASS TOI} 279914 1431S “Wye PES Gh “MW S'S mo1J 27991q 3uong | Cr.p 1g. 80.6 6€.6 3d9x jou 6.6 88.6 16.6 £0.6 Coy | eng 399} Ÿ | 228JINS 3997 Ÿ | 29EJInS 3007 Ÿ | 298TINS 3997 V | 208fInS 1997 p 2087INS U — — m ———— —— _— —>=-9-0 eontour---- 277 "# 5 er Liissilri1111 Depthin feet. Section 1 August, 1919. & = a = > 5 ¥ 2 & 8 = gs & ® © 3 5 7-11 PA TEE I BET ER PE RE N ng oo“ Te a ad D a ic 5 ggg 89 contour nt ST aren ee coe pen Gay ee 7 „es ee TES 9-0 contouf”” We son 010 20:0 on Etre Section 2 August, 1919. Section 4 August,1919 1 2 SS EN Scale imam es: R. B. S. 8. del. Sections of Rambha Bay, August 1919, showing densities zn situ. ~ Las CAES EE fount yay, MEM. IND. MUS., VOL, V, 1922. » ® oe & ic + [or A Surface | 10 PLATE XXXIII. Section 1. August, 1919. 5 10 Section 2. August, 1919. = ee S = & aes 3 SE ® n n n n n un, Ban Sse SS Sg 5 1:03 / 991 à | 10:85 10:58 da | oe f RC ab RSS ee if TS TE En PO OT Mr PY CET Se \ © 10 co Fo beri TE das EE convo PRE ns Section 4. August, 1919. 1 a ce in miles R. B. $, S. del. Sections of Rambha Bay, August 1919, showing densities reduced to standard temperature [25. O°C] MEM. IND. MUS. VOL, V, 1922. AMIR XO XONGTIN Surface. 8 en ; Litiitiiiis Depth in feet. Section 3. August, 1919. Surface. Section 3. August, 1919. 1 2 RE ee ey eee eet Ee er re Soalesimamntles: R. B.S. S, del. Sections of Rambha Bay, August, 1919, showing (above) density 7n situ, (below) density reduced to standard temperature [25. O°C.] MEM. IND. MUS., VOL, V. 1922. PLATE XXXV. th in feet. ’ te Dep Samal Is] Surface. SSS5 = Section 1. April, 1920. Section 2. April, 1920. Re) N ) rp) ise) Ca) 4 ine) & cs) cd is) Ed + + + un un un un mm mm mm nn Emm dub nn dag nn 5-17 0 con 4-69 4:19 4-52 tour. 5-43 es 4-86 4-19 5 == Section 3. April, 1920. 1 2 Sealei Me | | eier : J -In miles. R. B. 8. S. del. Section of Rambha Bay, April 1920, showing densities in situ. MEM. IND. MUS. VOL, V. 1922. PLATE XXXVI. Depth in feet. Surface. ts OR ao es + =) > + u un Op) dp) un he 7 5-56 L Br Contour, se 02 a Section 2. April, 1920. Re) > ce] œ 2 cs 2 5 @ @ © 8 + 2 + ee Fan & 5:98 S7 501. 5:81 ‘ ~ 6.0 con® nen Contour-----—-- 6-35 6-06 = 6) i i 1 2 NUR: Section 3. April, 1920. | US na les R. B. 8. S. del. Section of Rambha Bay, April 1920, showing densities reduced to standard temperature [25. O°C.] MEM. IND. MUS. VOL. V, 1922. PLATE XXXVII. == ‘ G wily (ilu Hy / z GTS 7. Ir Mr D, VE OU tll CAI lu nt ‘ Wily Soe fa CASTE AO ANR = ANU NZZ PV no NV 4 WINN UIVAN YW WE S35 SHUNNING yy SSS = men NNN Nue CO AE TS RSS 7j 11) N i VS TPT VANS WW Mn CT DAT DB ES 1 yy, à URI AA 2 ré KE N 1197) His E77 ANNE) ANUS WE LS Betty UTS oS ae ER N TE ns oi * II SINN "By =» 1% un == + awe IS Z arkuda Island; 4) \ At, anes AA, 4 > ZN y Ba A 7 WA 73 pe of toe ; it ML sd) ast Island. ; 10% RDS s.._ Sielzz. et A. 10%. 7 N, © 5 DS Sta. 7% à \ \ “ u 10 Nuchiriya Island R z \ 4 Sta:6 CS \ S Mr MinchinsHouse cP! \ \ “ FA S 2.20, D \ à D 512.16 = # LPC 2 \ . SE a 7) “Ay =F o : er as Le \ S ta.l7 y, === 5 \ à [3 9% + 'S © . S Sta 19 \ N L274! N > DIR à 4 a à Sta.18. ‘ 4 À a é \ = , = \ Kaduakani à 1 D url fad, A ye oe: Fu rlongs. Le . ot adharchna PP h2 ER RN se \ N = . RE 2 SS S u. \Malk AD AN SS 9 SS sur MN DS Rene NS R, B.S. S. del. - Chart showing position of stations 1—25. August 1919. en £ + > LE ' F Ba \ N 5 N ; = Tr … : a = De à: & 3 4 > QE . “ { € r > 2 3 1 { 3 7 . a + > € Le z : R € z * i) ws < . < . = ( . = 6 À ; Le § ee + a ie té + M is Pete = £ fy 2 STE 2 j É x F en MEM. IND. MUS. VOL. V, 1922 PLATE XXXVIII. Lire. , lg Ar Hy COURT SS A A \ FA 9118. 24 Railway rest House i BE > | 672. \ Nase va a u \ ; Se D ev \ ‘ 8-02 % CA ! 3 co so ee ===> I Breakfast Ist and, on 2 we hen {Oy ce. 80. L 2 "go © oS © a D Mr Minchins House 9.30. 4 3 2 1 Q 1 Sane sc oo Seoeoe sel \ aduakant \ ae me ‘ P.P. & N au " Malkude, a R. B.S. S. del. Chart showing density at surface, August 1919 MEM. IND. MUS. VOL. V, 1922. PLATE XXXIX- \ ile at nti ua ZU AN À rin GEN I-55 ei Soutour-—= = 7/1 FEN 2 24 1 % X A es FL \ 7 \ 9-67 a See eg £ 04 Me; reakfast Island Wis £ ce kK ; @: 8-78. 04 G rl AA N ss F per 9:47. - 950, BR \Rambha” RE 5 N Nt À eu N À a Er Mr: irchins House 9-88. „rom Ganjam furlongs. IR. B.S. S. del. Chart showing density at 5 feet depth, August 1919. MEM. IND. MUS. VOL. V, 1922. PLATE XL. és ; x | xD ft 19 Ri 10:09 ie ae 2 à DD © ze (COD) „ee Tossa WA a fe 2 o : > a 24 9-23 ; 10-01 4 WA cot | a „Breakfast 1stana Ir I Rotes / ; : wo br i 10-22 4 ares u. opo\F \ \ Se D 5 SeNChiriya Island D re N CY = À rire > “a r ple à \ Vera: 2 \ \ \ rom Ganja 2% RAT 7 NS BEN UL ER BORN PART nn R. B.S S. del. Chart showing density at 10 feet depth, August 1919. # fe PLATE XL]. MEM. IND. MUS., VOL. V, 1922 Gon U hi. ren : St2.42, mas 7 7 NW 71] 1 N oe Railway rest u Ram ba fy. Stn | ee re oe hi À oo © | wo Eu 2 See = 9° | _ “0 Ve ; 5 a ue „Breakfast leland va Saag ä BEN ee Sta.31 ne i | | N 2 i 0 Sr = Ram Dine & N i Nez Island (8) & à Mr. Minchins House Dr 8 N [3 furlongs. 8 206.524: 3) 2 "1.0 k. B.S. 8. del. Chart showing position of stations 26—40, April 1920 MEM. IND. MUS. VOL. V, 1922. Vj bs} 7 7 = sem Magy Ny >> ZE À T: TE = SR CIN OD u NN À oy, DAS iss à ERS AU = ee Er 4.58. 4158. a Er = 4-82. es ne . ETA AIR a 8 # \ i 7 er Stn. Z re Railway rest House ws ~ 2 & er ne A ft 119 asi Fi 4:28: i is Va 00 4 S ir ; we 2 F4 io. ee à AE x go Bar nae or n = yan Z 2 Win KS ne £ reat fast Island ~~ 5 Se À so 5-17. ©) 4:58. : Gi < DO LS a va 2S [0 : 7 er UAlERo Rambha \ Den Chirıya I; Island 7 = 8 r Minchins House \ D \ \ He ; Ss : re \ SS À N : 7 \ N o Furlongs. N aduakanı à oc 5 4 @ 2-40 1 adharchua a Bo ER Seel S. S. del. Chart showing density 4 7 Ny i 7 My, ig i on the surface in situ. CAME) LT 443- MEM. IND. MUS. VOL. V, 1922. PLANE. DCT, CMD, de Adee Te, GEL: à x BET, Wh la” DE SS pony b "y, 4 Sy} 4 cmd VW S = 7 ~ SIEH y, NZ Dili GV ANNI de, {0 yeas we Are nn ee PAT SI. any] De NS “Up 5 7, < Ly IN ’ J . , = MSY WAS wie DONS DOS S QE DD \ We 708 Û NW Sy 3 0005 OT TER REN) EN RUN TERN ARTS IN ATER HER TH THAN — << 2 Tnt um 0,5) fd EWS Liga CQ EASY 1 [LUE yb aie % ñ} Yay x la 1 4-62. Ya 4:79. ., x \ L Ya f 2 510. 4 “ 4.23. } N L Samen. Î Va = gr ne = 9) go con our... Z A oo EN x LA ‘ wf reakfast Island D. 5.48. D Mr Minchins House furlongs. FING 154) 3) RE 1 eee eT R. B.S. S. del. Chart showing density at 4 feet depth in situ. MEMOIRS OF THE INDIAN MUSEUM, VOL. V. / : 5 v FAUNA OF THE CHILKA LAKE © a _No. 11. APRIL, 1923. Fish, Pt. IV Fish, Pt. V aa Insley ait | Hatton afk ous 4: Vpn muet : | | ‘i +) yak Calcutta : BF | KA _ PUBLISHED By\ THE DIRECTOR, ZOOLOGICAL SURVEY OF INDIA Bad ok | | SUPERINTENDENT GOVERNMENT PRINTING, INDIA SS tee ES MORE AR TE. Ir Sacs Ihres Rupee, Sight anman. Cal toi 4) Gé NOIRS Ze pa ; ) FAUNA OF THE CHILKA LAKE. | FISH | PART IV. By B. L. Cuaupuurt, D.Sc. (Edin.), F.R.S.E. 7 Le “ | if ; - “ E L | B “ | Lt a j a of | [| © 7 5 et av CONTENTS. Coius quadrifasciatus (Sevastianot) Lates calcarifer (Bloch) Chanda ambassis (Lacépède) Priopis gymnocephalus (Lacépède) Lutjanus gohnii (Bloch) Therapon jarbua (Forskäl) » puta, Cuvier Sillago sihama (Forskal) Sciaena coibor (Hamilton Buchanan) Umbrina indica (Kuhl & Hasselt) Gerres öyena (Forskäl). ” setiter (Hamilton Buchanan) ” punctatus, Cuvier & Valenciennes Leiognathus equulus (Forskal) + blochii, Cuvier & Valenciennes Gazza minuta (Bloch) Monodactylus argenteus (Linnaeus) FISH (PART IV). By B. L. CHaupaurr. This paper contains a systematic treatment of the division Perciformes of the sub-order -Acanthopterygu. The total number of specimens examined and recorded is 281. They belong to seventeen known species, to thirteen genera and seven families. Sub-Order ACANTHOPTERYGII. Division PERCIFORMES. Family LOBOLIDAE. Genus COIUS ! Hamilton Buchanan. Coius quadrifasciatus (Sevastianof). 1809. Chaetodon quadrifasciatus, Sevastianof, Mém. Acad. Imp. Sci. St. Petersbourg 1, p. 448, tab. xviii, fig. 2. 1822. Coius polota, Hamilton Buchanan, Fish. Ganges, pp. 95 and 370, pl. xxxviii, fig. 2. 1842. Anoplus polota, Temminck and Schlegel, Faun. Japon., p. 17. 1844. Anoplus polota, Richardson, Zool. Voy. ‘Sulpher,’ p. 83. 1849. Datina polota, Cantor, Journ. Asiat. Soc. Bengal, p. 928. 1851. Lobotes hexazona, Bleeker, Nat. Tijdschr. Ned. Ind. I, p. 9. 1853. Datniordes polota, id., vbid., V, p. 441. 1859. Datnioides polota, Günther, Cat. Fish. Brit. Mus. I, p. 339. 1876. Datnioides quadrifasciatus, Bleeker, Arch. Neerl. Sc. Nat. XI, i, p. 272. 1877. Datnioides quadrifasciatus, id., Atl. Ichthyol. Ind. Orient. Néerl. VIII, p. 32, pl. ccev. hs. 1878. Datnioides poluta, Day, Fish. Ind. p. 96, pl. xxiv, fig. 6. 1889. Datnioides quadrifasciatus, Day, Faun. Brit. Ind. Fish. I, p. 535, fig. 162. 1890. Datnioides polota, Vinciguerra, Ann. Civ. Stor. Nat. Genova (2) IX, p. 162. 1905. Covus quadrifasciatus, Fowler, Proc. Acad. Nat. Sci. Philadelphia LVIL, p. 504. 1907. Datnioides polota, Lloyd, Rec. Ind. Mus. I, p. 227. The original specimen of Sevastianof (not Sebastian as given by Day in the Fauna volume) ‘must have been a young one as his figure shows all the three radiating brown bands from the orbit which are conspicuous in the young specimens only. The figure of Sevastianof is apparently life size, measuring 55 mm. in length. In colouration and marking it resembles most of the young specimens in the collection. 1 Goius is one of Hamilton Buchanan’s composite genera (Fish. Ganges, p. 85). As Bleeker’s Datnioides is the last "name proposed, it gives precedence to Cotus, of which Coius polota of Hamilton Buchanan is the type. else || B2 714 Memoirs of the Indian Museum. [ Von. VW, Hamilton Buchanan’s figure does not show the round marking on the post-opercle.. Day’s figures—both for D. polota in Fish. Ind. and for D. quadrifasciatus in the Fauna—show these markings, though no mention is made of them in the text. This round marking on the post-opercle is very conspicuous in all the young ones but is not traceable in the larger specimens in the collection. The ventral fin, the base of which is almost directly below the root of the pectoral fin, has one spine and five branching rays; the spine is outermost and of the rays the two next the spine have filiform endings, the inner one having a much more elongated ending than the one next to the spine. The fish is a per- manent inhabitant in the main area of the lake, where it breeds at the end of the rainy season. There are altogether eleven specimens in the collection of which eight afe young. The following list gives the different parts of the lake from which the specimens were collected, together with their number and size :— mm. 4 specimens ... Mouth of Barkul Bay ... 18th September, 1914 cee, aie 3 en ... Off Mottapur ... 14th March, 1918 ... 50—60. 1 specimen ... Off Nalbano ... ... 18th September, 1914 Se 2 specimens ... Rambha cee .. 21—51st July, 1913 225 75: 1 specimen wi ‘ sie ... Ist January, 1915 wee LAD. Distribution.—The estuaries of the Ganges and the rivers of Burma, Siam, the Malay- Peninsula and the Malay Archipelago. Family SERRANIDAE. Sub-family CENTROPOMINAE. Genus LATES Cuvier and Valenciennes. Lates calcarifer (Bloch). 1790. Holocentrus calcarıfer, Bloch, Ausl. Fisch. IV, p. 100, pl. cexliv. 1801. Perca calcar, Bloch and Schneider, Syst. Ichthyol. I, p. 89. 1802. Holocentrus heptadactylus, Lacépéde, Hist. Nat. Poiss. IV, p. 344. 1803. Perca sp. (pandoomenoo), Russell, Fish. Vizagapatam II, p. 23, pl. exxxu. 1822. Coius vacti, Hamilton Buchanan, Fish. Ganges, pp. 86 and 369, pl. xvi, fig. 28. 1828. Lates nobilis, Cuvier and Valenciennes, Hist. Nat. Poiss. II, p. 96, pl. xiii. 1828. Lates calcarifer, id, ibid., IL, p. 100. 1845. Lates nobilis, Bleeker, Nat. Geneesk. Arch. Ned. Ind. IL, p. 524. 1846. Lates nobitis, Richards, Rep. Brit. Assoc. Adv. Sc. 1845, p. 222. 1846. Lates calcarifer, id., ibid. 1849. Lates nobilis, Bleeker, Verh. Batav. Gen. XXII, p. 27. 1849. Lates heptadactylus, Bleeker, Journ. Asiat. Soc. Bengal, p. 983. 1853. Lates nobilis, Jerdon, Madras Jour. Lit. Sci. XVII, p. 128. 1859. Lates calcarifer, Günther, Cat. Fish. Brit. Mus. I, p. 68. 1865. Lates calcarıfer, Day, Fish. Malabar, p. 2. 1870. Lates calcarifer, Günther, Proc. Zool. Soc. London, p. 824. 1923.] Fauna of the Chilka Lake : Fish. 715 1876. Lates calcarifer, Day, Fish. Ind., p. 7, pl. i, fig. 1. 1876. Plectropoma calcarifer, Bleeker, Atl. Ichthyol. Ind. Orient. Neerl. VIT, p. 109, pl. ecexxu, fig. 3. 1877. Pseudolates cavifrons, Alleyne and Macleay, Proc. Linn. Soc. N. 8. W. I, p. 262, pl. i. 1878. Laies darwiniensis, Macleay, vbrd., IL, p. 345. 1889. Lates calcarıfer, Day, Faun. Brit. Ind. Fish. i, p. 440, fig. 139. 1890. Lates calcarifer, Vinciguerra, Ann. Mus. Cw. Stor. Nat. Genova, (2) IX, p. 162. 1895. Lates calcarifer, Boulenger, Cat. Perc. Fish. Brit. Mus. 1, p. 363. 1906. Plectropomus calcarifer, Jordan and Seale, Bull. U. S. Bur. Fish. XXV, p. 255. 1907. Lates calcarifer, Lloyd, Rec. Ind. Mus. I, p. 225. 1907. Plectropoma calcariferum, Evermann and Seale, Bull. U. S. Bur. Fish. XXVI, p. 78. 1910. Lates calcarifer, Jenkins, Rec. Ind. Mus. V, p. 13i. 1911. Lates calcarifer, Willey, Spol. Zeylanica VII, p. 100. 1912. Lates calcarifer, Jenkins, Rec. Ind. Mus. VII, p. 54. 1913. Lates calcarifer, Weber, Siboga-Exped. LVIT, Fische, p. 215. 1916. Lates calcurifer, Sundara Raj, Rec. Ind. Mus. XII, p. 278. There are two specimens in the collection, both from Satpara ; one measuring 262 mm. in length was secured in March 1914, and the other, measuring 137 mm. in length, was cap- tured on the 10th October. The fish is thus reported only from the outer channel.! Distribution. —ÜCoasts and mouths of rivers of South Eastern Asia from India to Southern China, Malay Archipelago, the Philippine Islands, Australia and New Guinea. Sub-family OHANDINAE. Genus CHANDA? Hamilton Buchanan. Chanda ambassis (Lacépède). 1775. ? Sciaena safgha, Forskal, Descrip. Animal, p. 53. 1801. ? Perca safgha, Bloch and Schneider, Syst. Ichthyol. 1 It is, however, common in Rambha Bay in the main area. JN. A. 2 The generic name Chanda of Hamilton Buchanan [ Fish. Ganges, 1822, pp. 103 and 370 ] has priority over Ambassis of Cuvier and Valenciennes [ Hist. Nat. Poiss., II (1828), p. 1751. This was pointed out by M’Clelland and Cantor as well as by Waite, although Fowler (loc. cit.), the first reviser of Chanda, had regarded its type identical with the type of Bleeker’s genus Pseudoambassis. Chanda of Hamilton Buchanan, which is the same as Bogoda of Bleeker, is characterized by the uninterrupted lateral line, small or minute scales and strong eurved canines and ıs distinguished from the related genera by the serrated pre-orbital, small teeth, comparatively larger scales, complete lateral line and the presence of about ten rays in the dorsal fin. A procumbent dorsal spine is always present but in some cases itis small and concealed in the fiesh [ The Fishes of Samoa by Drs. Jordan and A. Seal, Bull Bur. Fish. (U.S.) xxv, p. 175]. Fowler, and long before him Cuvier and Valenciennes, observed that the two first species under Chanda as described by Hamilton Buchanan belonged to a different genus altogether and for this reason Cuvier and Valenciennes suppressed the name Chanda, but they often showed themselves zeaious in cancelling valid names without any justification. It should be remembered that Hamilton Buchanan clearly expressed his doubts as to the propriety of placing these two species in his genus Chanda. The fact that he placed these two admittedly doubtful species under the generic name cannot therefore vitiate it. As to the first doubtful species, Hamilton Buchanan himself proposed to place it in another genus : “ This species is ill defined, and might, perhaps, be placed asa Coius.” (Fish. Ganges, p. 105). He further pointed out, ‘ As in the genera already described there are, as it were, certain intermediate species, so in this the two first, which I have described, together with the Zeus insidiator, have but little of the transparency, which forms part of the generic character.” He further stated that his excuse for including these two 716 Memoirs of the Indian Museum. [ Vor. V, 1802. Centropomus ambassis, Lacépéde, Hist. Nat. Powss. IV, p. 273. 1828. Ambassis commersonii, Cuvier and Valenciennes, Hist. Nat. Poiss. II, p. 176, pl. xxv. 1837. Ambassis commersomi, Rüppell, Neu. wirbel. Fisch., p. 89. 1849. Ambassis commersonii, Bleeker, Verh. Batav. Gen. XXII, p. 30. 1849. Ambassis macracanthus, id., ibid., p. 30. 1859. Ambasssis commersonii, Günther, Cat. Fish. Brit. Mus. I, p. 223. 1859. Ambassis macracanthus, id., 1bid., p. 227. 1865. Ambassis commersonii, Day, Fish. Malabar, p. 15. 1866. Ambassis productus, Guichenot, Mém. Soc. Sci. Cherbourg XII, p. 130. 1868. Ambassis commersonit, Peters, Reis. Mossamb. IV, p. 10. 1870. Ambassis macracanthus, Day, Proc. Zool. Soc. London, p. 681. 1875. Ambassis commersonüt, id., Fish. Ind., p. 52, pl. xv, fig. 3. 1877. Ambassis commersonti, Bleeker, Ail. Ichthyol. Ind. Orient. Néerl. VIII, pp. 133 and 136. 1889. Ambassis commersont, Day, Faun. Brit. Ind. Fish. I, 488. 1905. Ambassis ambassis, Fowler, Proc. Acad. Nat. Sci. Philadelphia LVIT, p. 500. 1915. Ambassis commersoni, Boulenger, Brit. Mus. Cat. Freshw. Fish. Africa III, p. 112, fig. 85. 1916. Ambassis ambassis, Sundara Raj, Rec. Ind. Mus., XII, p. 279. 1916. Ambassis commersoniv, Boulenger, Brit. Mus. Cat. Freshw. Fish. Africa, IV, p. 326. There are only two specimens in the collection, both secured from a fisherman at Kalupara Ghat on 7th April 1914; these are 54-6 and 64-5 mm. in length. Distribution—Hast coast of Africa, shores of India and the Malay Archipelago, North coast of Australia. The species ascends rivers and estuaries. Genus PRIOPIS! Kuhl and van Hasselt. Priopis gymnocephalus (Lacépéde). 1802. Lutjanus gymnocephalus, Lacépéde, Hist. Nat. Poiss. III, p. 479, pl. xxiii, fig. 3 and IV, p- 216. 1828. Ambassis dussumiert, Cuvier and Valenciennes, Hist. Nat. Poiss. II, p. 181. 1831. Lutjanus gymnocephalus, Lacépéde and Desmarest, Hist. Nat. Poiss. V, p. 108. 1834. Ambassis dussumieri, Quoy and Gaimard; Voy. “‘Astrolabe” Poiss. III, p. 645, pl. i, fig. 3. 1845. Ambassis dussumiert, Bleeker, Nat. Geneesk. Arch. Ned. Ind., II, p. 520. 1849. Chanda dussumieri, Cantor, Journ. Asiat. Soc. Bengal, p. 988. 1849. Chanda gymnocephala, id., ibid., p. 989. 1859. Ambassis dussumieri, Günther, Cat. Fish. Brit. Mus. I, p. 225. 1865. Ambassis dussumieri, Day, Fish. Malabar, p. 16. doubtful species (viz., C. setifer and C. ruconius.) in his genus Chanda was that the name was a common local appellation for all the species he included. A very curious mistake may be noted here. In Agassiz’s Nomenclator Zoologicus Pisces, p. 15) Hamilton Buchanan’s generic name Chanda is stated to have been derived from “the Greek word xauècc =hians!” Itisin reality a vernacular name. The word is not derived from the Hindi word Chandi (=silver), as wrongly supposed by Cantor and Day, but from the Bengali word Chand=moon, from the moon-like rounded shape and the moon- like semitranslucent lustre exhibited by these fishes in their sudden sallies to the surface of the water and quick retreat. 1 The genus Priopis, Kuhl and van Hasselt (Cuvier and Valenciennes, Hist. Nat. Poiss. VI, p. 503) is defined as Chanda with the lateral line interrupted (Jordan and Seale, Proc. U. S. Nat. Mus. XXVIII, p. 780). 1923.] 1865. 1869. 1870. 1874. 1877. 1878. 1879. 1889. 1905. 1905. 1913. Fauna of the Chilka Lake: Fish. 717 Ambassis dussumieri, Kner, Reis. ‘Novara’ Fisch., p. 41. Ambassis vachelli, Peters, Monatsb. Königl. Preuss. Akad. Wiss., Berlin (1868), p. 255. Ambassis dussumieri, Day, Proc. Zool. Soc. London, p. 681. Ambassis gymnocephalus, Bleeker, Nat. Verh. Holl. Maatsch. Wetensch. II, p. 15. Ambassis gymnocephalus, Bleeker, Atl. Ichthyol. Ind. Orient. Néerl. VIII, pp. 133 and 138, pl. ccelü, fig. 3. Ambassis gymnocephalus, Day, Fish. Ind. p. 54, pl. ceclu, fig. 3. Ambassis gymnocephalus, Bleeker, Verh. Akad. Amsterdam, XVIII, p. 13. Ambassis gymnocephalus, Day, Faun. Brit. Ind. Fish. I, p. 489. Priopis gymnocephalus, Jordan and Seale, Proc. U. S. Nat. Mus XXVIII, p. 780. Ambassis gymnocephalus, Fowler, Proc. Acad. Nat. Sci. Philadelphia LVII, p. 501. Ambassis gymnocephalus, Weber, Fisch. ‘ Siboga’-Exped., p. 217. Lacepede’s figure is defective as it does not show that the lateral line is not continuous. In many specimens there is no external appearance of the horizontal spine in front of the first dorsal fin. In some specimens the tips of the pelvic fins reach the vent, covering the anal opening. The caudal fin is not tipped with black in some, and in some the skin between the first dorsal fin and the body is black. There are altogether forty-seven specimens in the collection ; the following list gives the distribution of the species in the lake :— mm 9 specimens Off Barkul ... ... 18-21st September, 1914 39—48 16 3 Off mouth of Barkul Bay 18th September, 1914 35°5—48 1 specimen Off Barkul ... ... 13th November, 1914 35 1 3 Chirriya Island (shore collecting) ... 13th February, 1914 16-5 3 specimens Chirriya Island (Toward Samal Point) 17th February, 1914 42-5—47-5 1 specimen Between Chirriya Island and Barkuda Island ... 17th November, 1914 47 3 specimens Rambha Bay February, 1914 46—51 1 specimen a3 x … March, 1914 44 1 specimen Satpara (shore collect- ing) Se ... 13th March, 1914 17-5 2 specimens West of Satpara (tow netting) 20th March, 1914 10—12 1 specimen Satpara March, 1914 „45 5 specimens ER 12-13th September, 1914 ... 38—50 3 % 4h October, 1914 40—43 1 specimen er 40-5 The species occurs in the main area of the lake as well as in the outer channel, where it breeds. It is a permanent inhabitant in the lake. Distribution. — Coasts of Orissa and Malabar, entering rivers and estuaries in India, Seychelles, Penang, Javanese and Chinese seas, Celebes and Isle de France. 718 Memoirs of the Indian Museum. [Voz. V, Sub-family LUTJANINAE, Genus LUTJANUS Bloch. Lutjanus johnii (Bloch). 1795. Anthias john, Bloch, Ichthyol. IX, p. 97, pl. ecexviii. 1801. Anthias johnii, Bloch and Schneider, Syst. Ichthyol., p. 303. 1802. Lutjanus johnii, Lacépède, Hist. Poiss. IV, p. 235. 1803. Sparus sp. (doondiawah), Russell, Fish. Vizagapatam, I, p. 76, pl. xevii. 1803. Sparus sp. (mungimupudee), id., ibid., IE, p. 8, pl. ex. 1803. Sparus tranquebaricus, Shaw, Gen. Zool., Pisc. IV, p. 471. 1822. Coius catus, Hamilton Buchanan, Fish. Ganges, pp. 90 & 369, pl. 38, fig. 30. 1824. Mesoprion unimaculatus, Quoy and Gaimard, Zool. Freycin., pp. 304-441. 1828. Mesoprion johnii, Cuvier and Valenciennes, Hist. Nat. Poxss. IT, p. 443. 1831. Mesoprion flavipinnis, id., 1bid., VII, p. 475. 1831. Serranus pavoninus, id., ibid., VII, p. 443. 1831. Lutjanus johnii, Desmarest, Oeuvres Lacép., IX, p. 127. 1834. Mesoprion unimaculatus, Quoy and Gaimard, Voy. de l’Astrolabe., Poiss. III, p. 665, POW), tie. >. 1836. Mesoprion unimaculatus, Cuvier, Reg. Anim., Poiss. p. 35. 1846. Mesoprion unimaculatus, Richardson, Rep. Brit. Assoc. Adv. Sc. (1845), p. 229. 1849. Mesoprion johnii, Cantor, Journ. Asiat. Soc. Bengal, p. 995. 1849. Mesoprion unimaculatus, Bleeker, Verh. Batav. Gen. XXII, pp. 4, 20 and 42. 1853. Mesoprion unimaculatus, Jerdon, Madras Journ. Lit. Sa. XVII, p. 130. 1859. Serranus pavoninus, Günther, Cat. Fish. Brit. Mus. I, p. 126. 1859. Mesoprion johnii, id., ibid., p. 200. 1865. Mesoprion johnii, Day, Fish. Malabar, p. 11. 1865. Mesoprion johnii, Kner, Reis. ‘Novara Fisch., p. 35. 1876. Lutianis johnu, Day, Fish. Ind., p. 42, pl. xii, fig. 1. 1877. Lutjanus johni, Bleeker, Atl. Ichthyol. Ind. Orient. Neérl. VIII, p. 49, pl. ccexxxviii, ol ; 1889. Lutjanus johnu, Day, Faun. Brit. Ind. Fish. I, p. 476. 1904. Lutianus johnii, Fowler, Journ. Acad. Nat. Sci. Philadelphia (2) XII, p. 325. 1906. Lutianus johni, Jordan and Seale, Bull. U. 8. Bur. Fish. XXV, p. 264. 1907. Lutianus johnii, Lloyd, Rec. Ind. Mus., I, p. 226. 1907. Lutianus johnii, Evermann and Seale, Bull. U. S. Bur. Fish. XXVI, p. 79. 1908. Lutianus john, Gilchrist and Thomson, Ann. South African Mus. VI, p. 213. 1913. Lutianus Johmm, Weber, ‘Siboga’-Exped. LVII, Fisch, p. 247. There is only one specimen (young) measuring 50 mm. in lensth, captured on 17th November 1914, when proceeding across the mouth of Rambha Bay between Chirriya Island and Barkuda Island. There are three broad but faint transverse bands ; the black ocellus commences at the thirty-fourth scale from the snout and on the twentieth scale of the lateral line ; the ocellus measures 10 mm. x 8 mm. The species is only an occasional visitor in the main area of the lake. 1923.] Fauna of the Chilka Lake: Fish. 719 Distribution.—Coasts of Africa, Red Sea, seas of India, ascending some distance up tidal rivers, Malay Archipelago, coasts of China and Australia. Genus THERAPON : Cuvier. Therapon jarbua (Forskäl). 1775. Sciaena jarbua, Forskäl, Deser. Anım., p. 50. 1788. Sciaena jarbua, Linnaeus, Syst. Natur., Gmelin, Ed. XIII, p. 1303. 1790. Holocentrus servus, Bloch, Ausl. Fisch. IV, p. 80, pl. cexxxvii, fig. 1. 1797. Holocentrus servus, id., Ichthyol., taf. ccexxxvii, fig. 1. 1801. Grammistes servus, Bloch and Schneider, Syst. Ichthyol., I, p. 185. 1802. Holocentrus jarbua, Lacépède, Hist. Nat. Poiss. IV, pp. 348 and 355. 1817. Terapon servus, Cuvier, Reg. Anim., Ed. I, II, p. 295. 1824. Therapon timoriensis, Quoy and Gaimard, Voy. “Uranie” et“ Physicienne, ’’ p. 341. 1829. Therapon servus, Cuvier and Valenciennes, Hist. Nat. Poiss. III, p. 125. 1831., Therapon servus, id., 1bid., VII, p. 479. 1836. Therapon servus, Cuvier, Rèa. Anim., Poiss. p. 43. 1846. Therapon servus, Richardson, Rept. Brit. Assoc. Adv. Sc. (1845), p. 238. 1848. Therapon servus Bleeker, Journ. Ind. Arch. Il, No. IX, p. 632. 1859. Therapon servus, Günther, Cat. Fish. Brit. Mus. I, p. 278. 1865. Therapon servus, Kner, Reis. “Novara, Fisch. p. 45. 1865. Therapon servus, Day, Fish. Malabar, p. 17. 1867. Therapon servus, Jouan, Mem. Soc. Imp. Sci. Nat. Cherbourg XIII, p. 251. 1868. Therapon servus, Peters, Reis. Mossambique, IV, p. 10. 1870. Therapon jarbua, Klunzinger, Verh. Zool.-bot. Ges. Wien XX, p. 729. 1873. Therapon servus, Günther, Fisch. Sudsee I, p. 26. 1873. Therapon (Batnia) jarbua, Bleeker, Ned. Tijdschr. Dierk. IV, p. 377. 1875. Therapon servus, Bleeker, Ail. Ichthyol. Ind. Orient. Neérl. VII, p. 112, pl. xxxiv, 1922, 1876. Therapon jarbua, Day, Fish. Ind. p. 69, pl. xviii, fig. 4. 1876. Therapon (Datnia) jarbua, Bleeker, Arch. Neerl. Sc. Nat. XI, I, p. 267. 1878. Therapon (Drama) jarbua, Bleeker, Arch. Néerl. Sc. Nat. XIII, p. 42. 1884. Therapon jarbua, Klunzinger, Fisch. Roth. Meer, p. 729. 1889. Therapon jarbua, Day, Faun. Brit. Ind. Fish. I, p. 505, fig. 153. 1903. Therapon jarbua, Jordan and Evermann, Proc. U. 8. Nat. Mus. XXV, p. 348. 1904. Therapon jarbua, Fowler, Journ. Acad. Nat. Sci. Philadelphia (2) XII, p. 527. 1905. Terapon servus, Jordan, Guide Study Fish. IL, p. 342. 1906. Terapon jarbua, Jordan and Seale, Bull. U. S. Bur. Fish. XXV, p. 266. 1907. Terapon jarbua, Smith and Pope, Proc. U. S. Nat. Mus. XXX], p. 476. 1907. Terapon jarbua, Jordan and Seale, Bull. U. S. Bur. Fish. XXVI, p. 23. 1907. Therapon jarbua, Lloyd, Rec. Ind. Mus. I, p. 226. 1907. Terapon jarbua, Evermann and Seale, Bull. U. S. Bux. Fish. XXVI, p. 83. 1908. Therapon jarbua, Gilchrist and Thomsor, Ann. South Afric. Mus. VI, p. 150. 1910. Therapon servus, Franz, Abhandl. Akad. Wiss. IV, p. 46. 1 Terapon, Cuvier, Reg. Anim. Ed. I (1817), p. 295, was a misprint for Therapon and was subsequently corrected in a later edition. Djabub, Forskäl, Desc. Anim. (1775), p. 44, though an earlier generic name, is held not eligible. Cc 720 Memoirs of the Indian Museum. [Voz. V, 1911. Terapon jarbua, Jordan and Richardson, Mem. Carnegie Mus. IV, p. 187. 1912. Therapon jarbua, Bean and Weed, Proc. U. S. Nat. Mus. XIII, p. 605. 1913. Therapon jarbua, Weber, ‘Siboga’-Exped. LVII, Fisch., p. 254. 1913. Therapon jarbua, Sewell, Journ. Proc. Asiat. Soc. Bengal, (n. s.) IX, pp. 334 and 344. 1913. Therapon servus, Jordan, Tanaka and Snyder, Journ. Coll. Sci. Imp. Univ. Tokyo, XXXII, p. 168. 1915. Therapon jarbua, Boulenger, Cat. Freshw. Fish. Afric. Brit. Mus. LIT, p. 118, fig. 114. 1917. Therapon jarbua, Hornell, Madras Fish. Bull. XI, p. 91. Hamilton Buchanan has left an excellent figure of this fish in the plate No. 67 of the volume of his manuscript drawings ; ! the name “Holocentrus (?) katkaya” is on the back of the plate in his own handwriting. This drawing is evidently the original of the badly copied figure in Hardwicke’s Illustrations” The figure was named Pterapon trivittatus | and was published without any acknowledgment of the source, the name also evidently was borrowed without acknowledgment from Hamilton’s Fishes of the Ganges (p. 92) on a mistaken identity of the published species with the unpublished manuscript figures. There are altogether five specimens in the collection, four of which are from Satpara, but no special locality is known for the fifth which measures 88 mm. and was collected at the end of July 1913. Ofthe Satpara specimens the biggest measures 115 mm. in length and was collected on 12th September 1914 and the remaining three on March, 1914, measuring 83 mm., 85 mm. and 95 mm. The biggest specimen has eleven spines in the first dorsal, the one measuring 95 mm. in length has ten prominent spines and a rudiment- ary one anteriorly. Of the rest one has a trace of a spine but the other two specimens have only ten prominent spines in the first dorsal fin. These facts satisfactorily explain the differences in the observations of Günther and Klunzinger on the number of spines. Distribution.—Red Sea, east coast of Africa, seas and estuaries of India, the Malay Archipelago, north coast of Australia, Formosa, Japan, Samoa, Fiji, New Britain, New Guinea and the Solomon Islands. Therapon puta, Cuvier. 1803. Perca sp. (keelputa), Russell, Fish. Vizag. IT, p. 19, pl. exvi. 1817. Terapon puta, Cuvier, Reg. Anim. Ed. I, II, p. 295. 1822. Coius trivittatus, Hamilton Buchanan, Fish. Ganges, pp. 92 and 370. 1829. Therapon puta, Cuvier and Valenciennes, Nat. Hist. Ponss. III, p. 131. 1829. Therapon ghebul, ıd., zbid. III, p. 133. 1836. Therapon puta, Cuvier, Reg. Anim., Poiss., p. 43, pl. xu, fig. 2. 1849. Therapon trivittatus, M’Clelland, Journ. Asiat. Soc. Bengal, p. 1001. 1853. Therapon puta, Jerdon, Madras Journ. Lit. Sci. XVII, p. 130. 1859. Therapon trivittatus, Giinther, Cat. Fish. Brit. Mus. I, p. 281. 1859. Therapon ghebul, id., ibid., I, p. 281. 1865. Therapon trwitiatus, Kner, Reis. ‘Novara,’ Fisch., p. 45. 1865. Therapon trwittatus, Day, Fish. Malabar, p. 17. 1873. Therapon (Datina) trivittatus, Bleeker, Ned. Tijdschr. Dierk. IV, p- 375. 1 Chaudhuri, Mem. Ind. Mus. V, p. 444 and foot-note. 2 Gray, Illustrations of Indian Zoology from the collection of Major-General Hardwicke, IL, pl. Ixxxvüi, fig. 1. are young. zontal bands. 1923.] Fauna of the Chilka Lake: Fish. 1875. Therapon puta, id., Fish. Ind., p. 68, pl. xvii, fig. 3. 1884. Therapon trivittatus, De Vis, Proc. Linn. Soc. N. 8. W. VIII, p. 457. 1889. Therapon puta, Day, Faun. Brit. Ind. Fish. 1, p. 505. 1906. Terapon puta, Jordan and Seale, Bull. U. S. Bur. Fish. XXV, p. 266. 1907. Terapon puta, Jordan and Seale, ibid. XXVI, p. 24. 1907. Terapon puta, Evermann and Seale, sbid. XXVI, p. 83. 1912. Therapon puta, Bean and Weed, Proc. U. 8. Nat. Mus. XLII, p. 605. i913. Therapon puta, Weber, ‘Siboga’-Exped. LVIL, Fisch., p. 91. 721 1913. Therapon puta, Sewell, Journ. Proc. Asiat. Soc. Bengal (n. s. ) IX, p. 352. 1917. Therapon puta, Hornell, Madras Fish. Bull. XI, p. 91. There are altogether thirty-six specimens in the collection, more than twenty of which 2 specimens 1 specimen 8 specimens 1 specimen 1 specimen 1 specimen 7 specimens 7 specimens 3 specimens 1 specimen 2 specimens 1 specimen 1 specimen Off Bamikuda Cherriakuda Samal Point Mahosa (Barhampur Island) ... IR Rambha Bay Between Sama! Island and mainland towards Satpara Satpara Bay South side of Satpara Island he: West of Satpara Seruanaddi From Seruanaddi going towards Barnikuda 6th September, 1914, 17th February, 1914 18th March, 1914 February, 1914 September, 1913 March, 1914 13th March, 1914 17th March, 1914 13th March, 1914 20th March, 1914 8th September, 1914 4th September, 1914 The list given below will show the time and place of their occurrence in the lake. mm. 54—62 32 ae 68—87 15—25 17, 18 & 21 14 15 & 17 60 67 The young ones were mostly obtained in shore-collecting in the neighbourhood of Sat- para Island. Numerous round light spots are found in these specimens between the hori- The caudal fin is immaculate in most of the young specimens. In some of the young specimens there is a black spot at the root of the caudal fin. This fish appears to be a permanent inhabitant in the main area as well as in the outer channel, breeding in the latter area in winter. Distribution.—Red sea, seas of India, Malay Archipelago, the Philippine Islands, 1801. Platycephalus sihamus, Bloch and Schneider, Syst. Ichthyol., p. 60. Family SILLAGINID Æ. Genus SILLAGO, Cuvier. Sillago sihama (Forskäl). 1775. Atherina sihama, Forskäl, Descrip. Anim., pp. xiii and 70. coast of Australia, sea of Timur and South Pacific Ocean (the island of Samoa). 722 1801. 1803. TENTE 1827. 1829. 1829. 1835. 1836. 1845. 1849. 1853. 1860. 1861. 1861. 1865. 1866. 1867. 1868. 1870. 1874. 1876. 1880. 1885. 1885. 1889. 1902. 1904. 1905. 1905. 1907. 1907. 1908. 1910. SV, 191% One 1912. 1913. 1913. 1913. 1913. 1913. 1914. LOT Memoirs of the Indian Museum. [Vor. V, Sciaena malabarica, id., ibid., p. 18, pl. xix. Sparus sp. (soring), Russell, Fish. Vizag. II, p. 9, pl. ex. Sillago acuta, Cuvier, Reg. Anim. (Ed. I) IE, p. 258. Sillago sihama, Rüppell, Atl. Reis. Nord. Afrik. Fisch. Meer. p. 9, pl. ii, fig. 1. Sillago acuta, Cuvier and Valenciennes, Hist. Nat. Poiss. III, p. 400. Sillago erythroea, id., ibid., p. 409. Sillago sihama, Rüppell, New. Wirbel. Faun. Abyssin. Fisch., p. 100. Sillago sthama, Cuvier, Reg. Anım., Poiss., p. 45. Sillago acuta, Bleeker, Nat. Geneesk. Arch. Ned. Ind. II, pp. 524 and 527. Sillago malabarica, Cantor, Journ. Asiat. Soc. Bengal, p. 1002. Sillago acuta, Jerdon, Madras Journ. Int. Sa. XVII, p. 131. Sillago sihama, Günther, Cat. Fish. Brit. Mus. II, p. 243. Sillago sihama, Gill, Proc. Ac. Nat. Sci. Philadelphia, p. 504. Sillago malabarica, id., ibid. Sillago sihama, Day, Fish. Malabar, p. 47. Sillago sihama, Playfair, Fish. Zanzibar, p. 69. Sillago sihama, Jouan, Mém. Soc. Imper. Sci. Nat. Cherbourg XIII, p. 252. Sillago sihama, Kner, Reis. Oster. Novar. Fisch., p- 128. Sillago sihama, Klunzinger, Verhandl. Zool.-Bot. Ges. Wien. XX, p. 818. Sillago sihama, Bleeker, Verh. Akad. Amsterdam XIV, p. 67. Sillago sihama, Day, Fish. Ind., p. 265, pl. lvii, fig. 3. Sillago sihama, Günther, Rep. Voy. H. M. S. “ Challenger,” Zool. I, p. 56. Sillago sthama, Macleay, Proc. Linn. Soc. N. S. W. IX, p. 28. Sillago sthama, Steindachner and Doderlein, Denk. Akad. Wiss. Wien. XLIX, p. 19. Sillago sihama, Day, Faun. Brit. Ind. Fish. II, p. 224. Sillago sihama&, Jordan and Snyder, Proc. U. S. Nat. Mus. XXIV, p. 486. Sillago sihama, Fowler, Journ. Acad. Nat. Sci. Philadelphia XII, p. 549. Sillago sihama, Jordan, Stud. Fish. II, p. 358. Sillago sihama, Jordan and Seale, Bull. U. S. Bur. Fish. XXV, p. 277. Sillago sihama, id., ibid, XXVI, p. 25. Sillago sihama, Smith and Pope, Proc. U. S. Nat. Mus. XXII, p. 478. Sillago sihama, Gilchrist and Thomson, Ann. S. Afric. Mus. VI, p. 192. Sillago sihama, Jenkins, Rec. Ind. Mus. V, pp. 132 and 136. Sillago sihama, Franz, Abhandl. Bayer. Akad. Wiss. IV, suppl. I, p. 83. Sillago sihama, Jordan and Richardson, Mem. Carnegie Mus. IV, p. 192. Sillago sihama, Bean and Weed, Proc. U. S. Nat. Mus. XLII, p. 607. Sillago sthama, Jenkins, Rec. Ind. Mus. VII, p. 60. Sillago sihama, Tanaka, Fig. Descrip. Fish. Japan XIV, p. 241, pl. Ixviii. Sillago sihama, Jordan, Tanaka and Snyder, Journ. Coll. Sci. Imp. Univ. Tokyo XX XIII, De UST Sillago sihama, Jordan and Metz, Mem. Carnegie Mus. VI, p. 41. Sillago sihama, Sewell, Journ. Asiat. Soc. Bengal (n. =.) IX, pp. 338 and 344. Sillago sihama, Weber, ‘Siboga’-Exped.. Fisch., p. 267. Sillago sihama, Jordan and Thomson, Mem. Carnegie Mus. VI, p. 259. Sillago sihama, Hornell, Madras Fish. Bull. XI, p. 91. 1923.] | Fauna of the Chilka Lake : Fish. 723 Dr. Gill separates S. malabarica as a distinct species having the soft dorsal spotted.! Besides colouration specimens of this species show great variation in the depth of the body, attenuation of the head and snout and height of the spinous dorsal. That M. Les- chenault saw “single individuals upwards of three feet in length” was first given currency by Cuvier.? Most subsequent writers, including Day, quoted this statement without any corroboration or acknowledgment. The species is, comparatively speaking, a small sized one.” As the species is of wide distribution and as no one else has observed it to reach anywhere near the size recorded by Leschenault, itis probable that his observation is erroneous. It is not unlikely that Leschenault mistook some species of Sphyraena for Sillago sihama as both the genera have two dorsal fins, a long pointed snout as well as a ‚similar nature and arrangement of scales. Cuvier and Valenciennes partly confounded S. panyus (Hamilton Buchanan) with S. sihama (Forskäl), for they remark that the vernacular name for the species in Calcutta is Panjı mas.* There are altogether seventeen specimens in the collection, seven of which are quite young and were obtained only a mile south-west of the mouth of the lake on the outer bar. The rest are all adult and were found distributed over the main area and collected throughout the year. The caudal fin is in most cases square-cut and in some emarginate but never deeply indented as represented in Russell’s figure, which in all probability is defective. Most of the specimens have a broad longitudinal silvery band about the middle -of the body not conspicuous in the young specimens. In some there are black blotches on the opercle. There is a horse-shoe shaped black marking on the occiput with white border in front. The following statement shows the number and size of the specimens in the collection together with the localities in the lake from which they were obtained :— : mm. 4 specimens 2. Balusapıy 2. … 21-31st July, 1913 .... 112 —144 2 specimens Ce nieud Le ... 28th November, 1914 nn SR 1 specimen ... Rambha Bay ... February, 1914 sty 13D 3 specimens ... (purchased) ... ... 19th November, 1914 ... 107—152 7 specimens ... Outer Bar, one mile south- west of the mouth of the lake. BS. ... 19th March, 1914 a 25—44 The species is a permanent inhabitant of the main area of the ‘ake going out to the sea or near the mouth of the lake to breed. In all probability the breeding time is about the month of February. Distribution.°—Coasts of Abyssinia, Zanzibar, North and East Africa, Red Sea, seas of India, Bay of Bengal and estuaries of the Ganges, Malay Archipelago, seas of the 1 Proc. U. S. Nat. Mus. XXIV, p. 487. 2 Hist. Nat. Poiss. III, p. 407. 3 “Brreicht fast 30 em. lange” Weber, ‘Siboga’-Exped. Fisch., p. 267. 4 Hist. Nat. Poiss. III, p. 401. 5 Günther in his catalogue enters “one skin (bad state)” of this fish as belonging to Nepal, presented by B. H. Hodgson ; this is undoubtedly a mistake. The locality of the donor who was for a long time a resident in Nepal must have been mis- taken for that of the fish. Günther has similarly referred a few more marine fish to Nepal which led T. C. Jerdon to contribute his paper “ On the extension of certain marine fishes to the freshwater Rivers of India.” Ann. Mag. Nat. Hist. (3) XVIL, p. 153. 724 Memoirs of the Indian Museum. [Voz. V, Philippines, China, Formosa, Japan and Korea, coast of Queensland and also that of Samoa. Family SCIAENIDAE. Genus SCIAENA Linnaeus. The genus Johnius (including Bola) was restricted to J. carutta by Gill.“ Bleeker proposed to separate those species which had enlarged teeth in the lower jaw from Sciaena and wanted to group them under a new genus Pseudosciaena,? for which he made S. aquila (Lacépéde) the type. If Bleeker’s arrangement be adopted the generic name of the group should for reasons of priority become Argyrosomus? of De La Pylaie, who founded the latter genus in 1832 on the same species. Sciaena is undoubtedly a large genus comprising a great variety of forms which, though differing widely among themselves, form an almost continuous series from one extremity to the other. The inter-relations of these forms have been fully discussed by Jordan and Eigenmann and no useful purpose would be served by upholding the number of these artificial genera. The genus Scaena is now therefore deiinitely restricted to Cheilodipterus aquila of Lacépède.5 This species thus becomes the. type of Sciaena which replaces the genera Argyrosomus of De La Pylaie and Pseudosciaena of Bleeker.® Sciaena coibor (Hamilton Buchanan). 1822. Bola coibor, Hamilton Buchanan, Fish. Ganges, pp. 78 and 368. 1830. Corvina albida, Cuvier and Valenciennes, Hist. Nat. Poiss. V, p. 93. 1830. Corvina anei, id, rbid., p. 131. 1834. Corvina albida, Belanger, Voy. Indes-orientales, p. 355. 1860. Johnius anei, Blyth (not Bloch), Proc. Asiat. Soc. Bengal, p. 141. 1863. Pseudosciaena albida, Bleeker, Ned. Tijdschr. Dierk. I, p. 145. 1865. Corvina albida, Day, Fish. Malabar, p. 54. 1865. Corvina neilli, id., ibid., p. 55. 1876. Sciaena albida, id., Fish. Ind., p. 188, pl. xliv, figs. 4 and 6. 1889. Sciaena albida, id., Faun. Brit. Ind. Fish. II, p. 117. 1910. Sciaena albida, Jenkins, Rec. Ind. Mus. V, p. 136. There is only one specimen in the collection, 462 mm. in length without the caudal fin. It was caught off Barkul Point at the end of November, 1914. Another specimen was re- ported from Gopkuda in August, 1907. In the Barkul specimen the muciferous pore below the symphysis of the lower jaw (the centrally situated one behind the bluntish knob) is semi- lunar in shape with a short hanging fold in front, the two lateral pores are deep and elongated and the outer pores are almost slit-like. The barbel between the right corner of the semilunar pore and the right lateral elongated pore is very slender and thin and is only 5 mm. in length 1 Proc. Acad. Nat. Sci. Philadelphia, 1862 (published 1863), pp. 16-18. 2 Ned. Tijdschr. Dierk. I (1863), p. 145., and Arch. Néerl. Sc. Nat. XI (1876), p. 329. > Compt. Rend. Congr. Sci. France for 1834 (published 1835), p. 534. 4 Bull. U. S. Fish Comm. for 1886 (published 1889), p. 395. 5 Lacépède, Hist. Nat. Poiss., Nou. Ed., IV, p. 373. 6 Jordan, The Genera of Fishes, 1917, p. 94. 1923.] Fauna of the Chilka Lake : Fish. 725 and is contained three times in the short vertical diameter of the eye. There is a minute (but thick) barbel-like growth near the left lateral pore. The eye is oval, the short vertical diameter is contamed in the horizontal diameter one and one-third times. The longer dia- meter of the eye is contained twice in the interorbital distance. The fish is an occasional visitor to the lake, appearing in the main area during the flood and also soon after the freshets are over. It grows to four feet and more in length.! Distribution. —Seas of India, coast of Malabar, larger estuaries of the Ganges, and the estuary of the Sittang river ; seas of China and the Philippine Islands. Genus UMBRINA Cuvier. Umbrina indica (Kuhl and Hasselt). 1803. Labrus sp. (qualar katchelee), Russell, Fish. Vizag. II, p. 13, pl. xviii. 1824. Sciaena indica, Kuhl and Hasselt, Bull. Sci. Nat. (Ferussac), II, pp. 374 and 377. 1830. Umbrina russelüi, Cuvier and Valenciennes, Hist. Nat. Poiss. V, p. 178. 1830. Sciaena indica, id., rbid., p. 179. 1830. Umbrina kuhlii, id., ibid. 1836. Umbrina russel, Cuvier, Reg. Anim., Poiss. p. 82. 1846. Umbrina russel, Richardson, Rep. Brit. Assoc. Adv. Sc. (1845), p. 226. 1849. Umbrina russell, Cantor, Journ. Asiat. Soc. Bengal, p. 1053. 1850. Umbrina kuhliv, Bleeker, Verh. Batavia Gen. XXIII (5), pp. 4, 5, 11 and 19. 1852. Umbrina russelli, Bleeker, Nat. Tijdschr. Ned. Ind. III, p. 56. 1853. Umbrina russelli, Jerdon, Madras Journ. Lit. Sci. XVII, p. 132. 1860. Umbrina russellu, Giinther, Cat. Fish. Brit. Mus. IT, p. 278. 1868. Umbrina russell, Kner, Reis. ‘ Novara, Fisch. p. 131. 1873. Umbrina russell, Day, Rep. Sea Fish and Fisher, p. cel. 1874. Sciaena russelli, Bleeker, Verh. Akad. Amsterdam XIV, p. 58. 1876. Umbrina russelli, Day, Fish. Ind., p. 183, pl. xiii, fig. 4. 1889. Umbrina russell, id., Faun. Brit. Ind. Fish. IL, p. 110. 1907. Umbrina russelli, Jordan and Seale, Bull. U. S. Bur. Fish. XXVI, p. 25. 1907. Umbrina russell, Evermann and Seale, ibid., p. 87. 1910. Umbrina russellii, Jenkins, Rec. Ind. Mus. V, p. 136. Russell described and figured this fish from Vizagapatam under its local name as a “ Labrus with a pentagonal tail.” Twenty-one years later Kuhl and van Hasselt rede- scribed it from a specimen obtained at Java and supplied a name under approved methods and called it Scoaena indica. Cuvier and Valenciennes, though acknowledging the name given by them, rechristened the species after the senior author and called it U. kuhlii and, thinking :t a different species, invented the name U. russellii for Russell’s species from Vizagapatam. Later writers, finding out the identity of U. kuhlii with Russell’s species, dropped the name U. kuhlu, but did not restore the earlier name. Moreover they copied the inadmissible name with its incorrect spelling. The law of priority demands that the name given by Kuhl and van Hasselt should be restored. 1 “ This is a very beautiful fish, found in the larger estuaries of the Ganges. I saw only one specimen, which was four feet in length; but it is said to grow considerably larger.” Fish. Ganges, p. 79. 726 Memoirs of the Indian Museum. [Vor. V,. There are altogether eight specimens in the collection. In all the specimens the second ray of the ventral fin ends in a prolonged filamentous extension, similar to that shown in Russell’s figure though not mentioned by later authors. The upper ridges over the eyes are black in all the specimens ; the upper two-thirds of the anterior dorsal fin is black or ashy brown in many and in some the membrane joining the spines is covered with minute dark brown spots, the other fins having yellow spots ; a faint black blotch is noticed on the opercle of almost all the specimens. The following list shows the different localities in the lake from which the specimens. were obtained and their number and size :— mm. 1 specimen … Barkul ag: ... 13th November, 1912 101199 I Barnes ... Ist March, 1914 AS 2 specimens ... Chiriya Island ... 18th February, 1914 TS 1 specimen … Between Maludaikuda and Kalidai ... 21st September, 1914 Oo il 5 ... Samal Island ... 22nd September, 1913 ae Oo 2 specimens ... Channel between Satpara and Barnikuda ... 4th September, 1914 sae) 195) C5. BS) The fish appears to be common in the main area and in the outer channel from September to March, the young ones bemg found in the outer channel after the floods are over. In all probability the fish is a permanent inhabitant of the lake, breeding in the outer channel. Some of the specimens grunted loudly on being removed from the water. Distribution.—Seas of India, Bay of Bengal, coasts of Ceylon, Penang, Malay Archi-. pelago, seas of China (Canton) and coasts of the Philippine Islands. Family GERRIDAE. Genus GERRES,! Quoy and Gaimard. Gerres oyena (Forskäl). 1775. Labrus öyena, Forskäl, Deser. Anim., p. 35. 1788. Labrus öyena, Linnaeus and Gmelin, Syst. Natur., (Ed. 13th), I, p. 1287. 1802. Labrus öyena, Lacépède, Hist. Nat. Poiss. III, p. 463. 1802. Labrus longirostris, id., 1bid., p. 468, pl. xix, fig. 1. 1802. Sparus britannus, id., ibid., IV, pp. 132 and 134. 1827. Smaris öyena, Rüppell, Atl. Reis. Nord. Afrika, p. 11, pl. iii, fig. 2. 1829. Gerres öyena, Cuvier, Reg. Anim., Poiss. (2nd Edit.), p. 104. 1 In 1824 the authors (viz., Quoy and Gaimard) published the ‘ Voyage autour du Monde” shortly before the- second edition of the Règne Animal. In this publication (p. 293) they adopted the genus Gerres from Cuvier’s manuscript. In 1829 in the second edition of Reg. Anim. Cuvier established the genus based on seven species, including - G. öyena. In 1850, thinking Gerres pre-occupied by Gerris Fabricius—a genus of Hemiptera (1794), Cantor proposed Catochaenum in its place. Gerres, being spelled differently from Gerris, is not pre-occupied (See Proc. California Acad. Sci. (2) V, p. 470.) it should be noted here that the new name Xystaema created by Jordan for some species of Gerres - has been withdrawn by the author (The Genera of Fishes, p. 118). Jordan further points out that Gerres and Gerris are words from different roots. Podager proposed as a substitute for Gerres [Natur. Thier. Schul., p. ix] is pre-occupied _ in birds and thus could not replace Gerres. [Proc. Acad. Nat. Sci. Philadelphia, LXX p. 338 (1918)]. 1923.] Fauna of the Chilka Lake : Fish. 727 1830. Gerres öyena, Cuvier and Valenciennes, Hist. Nat. Poiss. VI, p. 355. 1836. Gerres öyena, Cuvier, Reg. Anim., Poiss. p. 104. 1845. Gerres öyena, Bleeker, Nat. Geneesk. Arch. Ned. Ind. II, p. 521. 1850. Gerres öyena, id., Verh. Batav. Gen. XXII, p. 12. 1859. Gerres öyena, Günther, Cat. Brit. Mus. Fish. IV, p. 261. 1863. Diapterus öyena, Bleeker, Ned. Tijdschr. Dierk. 1, p. 232. 1866. Gerres öyena, Günther, Fish. Zanzibar, p. 111. 1875. Gerres öyena, Day, Fish. India, p. 99, pl. xxv, fig. 4. 1877. Diapterus öyena, Bleeker, Atl. Ichthyol. VIII, p. 129. 1884. Gerres öyena, Klunzinger, Synops. Fisch. Roth.-Meer, p. 49. 1889. Gerres öyena, Day, Faun. Brit. Ind. Fish. I, p. 538. 1890. Gerres öyena, Thurston, Notes Pearl Fish. and Marine Faun. Manaar, p. 91. 1907. Xystaema öyena, Smith and Pope, Proc. U. 8. Nat. Mus. XXXI, p. 478. 1908. Xystaema öyena, Seale and Bean, ıbid., XXXIH, p. 244. 1913. Gerres öeyena, Weber, ‘Siboga’-Exped. Fisch., p. 273. 1913. Gerres öeyena, Sewell, Journ. Asiat. Soc. Bengal (n. s.) IX, p. 344. Most of the later authors from Cuvier down to Day and Smith have included under the synonymy of this fish, Gerres equula of Temminck and Schlegel,! which is a distinct Japanese species.” G. equula is, however, identical with @. erythroarum (Bloch).3 It was first described from Japanese specimens. Both these names therefore have been excluded from the list of synonymy though they are found included in many of the previous lists. There are altogether three specimens, more or less damaged, measuring in length 76 mm., 77 mm. and 85 mm. without the caudal fin. They were all obtained at Satpara on the 10th October, 1914. In all probability the fish is not a permanent inhabitant of the lake, but is a casual visitor to the outer channel after floods. Distribution. —East coast of Africa, Red Sea, seas of India, Malay Archipelago. the Philippines; the Fiji Islands and Japan. Gerres setifer (Hamilton Buchanan). 1822. Chanda (2) setifer, Hamilton Buchanan, Fish. Ganges, pp. 105 and 370. 1830. Gerres Cuvier and Valenciennes, Hist. Nat. Poiss. VI, p. 477. 1849. Catochaenum lucidum, Cantor, Journ. Asiat. Soc. Bengal, p. 1038. 1853. Gerres lucidus, Bleeker, Verh. Batav. Genoot., XXV, p. 40. 1862. Gerres altipinis Günther, Cat. Fish. Brit. Mus. IV, p. 258. 1867. Gerres lucidus, Jouan, Mém. Soc. Imp. Sci. Nat. Cherbourg XIII, p. 263. 1875. Gerres setifer, Day, Fish. Ind. p. 97, pl. xxv, fig. 1. 1875. Gerres lucidus, id., ibid., p. 99, pl. xxv, fig. 5. 1889. Gerres setifer, id., Faun. Brit. Ind. Fish. I, p. 536. 1889. Gerres setifer, id., 1bid., p. 539. 1910. Gerres lucidus, Jenkins, Rec. Ind. Mus. V, pp. 131 and 135. 1913. Gerres lucidus, Sewell, Journ. Asiat. Soc. Bengal (n. s.) IX, p. 344. 1 Temminck and Schlegel, Faun. Japon. Poiss., p. 76, pl. xl, fig. 1. 2 Jordan, Tanaka and Snyder, Journ. Coll. Sci. Imp. Univ. Tokyo XX XIII, p. 177, fig. 129. 3 Bloch, Ichthyologie VIII, p. 23, pl. celxi. 728 Memoirs of the Indian Museum. [ Vor. V, Hamilton Buchanan found this fish in the estuaries of the Ganges and had a drawing made of it which is still preserved in the volume of manuscript drawings (Plate lxvi) in the library of the Asiatic Society of Bengal! The name Katchanda is written on the page both in Bengali and Roman characters. This is the local name for the fish. In the absence of the type specimen this manuscript plate becomes the protograph. Hamilton Buchanan doubted the propriety of including it under the genus Chanda and suggested its removal to the genus Coius. In the drawing also, the number of spines in the dorsal'fin is ten and that of rays only nine. Gerres lucidus of Cuvier and Valenciennes is described from specimens received from Pondicherry having nine spines and ten divided rays. Giinther considers G. lucidus as a doubtful species and does not recognise G. setifer at all. He, however, described this fish as a new species under the name of @. altipinnis, from a speci- men from the Ganges, which perhaps was Hamilton Buchanan’s type as it was out of a col- lection presented by G. R. Waterhouse which is suspected to contain some of Hamilton Buchanan’s types. Day has admitted both the names @. setifer and G. lucidus though he was strongly of opinion that they referred to the same species, Jordan? on the other hand proposed a new genus, which he styled Gerreomorpha, for specimens with ten instead of nine dorsal spines (viz., G. Japonica and G. setifer).. Though in other respects quite similar, some of the specimens in the present collection have ten and others nine spines. This is, therefore, a variable character in the species. There are altogether one hundred and twenty-two specimens in the collection, all obtained during the months of February and March. The species is found during this restricted period throughout the main area as well as in the outer channel of the lake. The following statement shows the different localities whence the specimens were obtained and their number and size :— mm. 1 specimen ... Off Barkuda Island ... 17th February, 1914 1000 2 specimens ... Off Barkul ... ... Ist March, 1914 .… 39 and 44 Ihe zs ... Barkul Bay ... .. Ist March, 1914 … O2—54 10 3 ... Chilka lake ... BER ... 49—98 60 = ... Chirriya Island towards Samal Island … 17-18th February, 1914 ... 28-56 7 À … Off Kalidai ... ... Ist March, 1914 oA 1 specimen ... Between Kalidai and Samal Island .... 20th February, 1914 1130 6 specimens ... Off Patsahanipur ... 3-9th March, 1914 u... 18—55 8 N .. From Sankuda towards Samal Island ... 17th February, 1914 … 26—43 fl a ... Rambha Bay … February, 1914 … 45—74 1 specimen = hs = ... March, 1914 Bee ow 2 specimens ... Satpara oes ... 7th March, 1914 ... 33—60 The species appears to be a dry-weather visitor to the lake and does not breed in it. It is said to be the most common Indian species, visiting the coasts in enormous numbers and going up the estuaries. 1 Chaudhuri, Mem. Ind. Mus. V, p. 444. 2 Jordan, The Gerrid fishes of Japan. Proc. U. S. Nat. Mus. XX XIII, p. 247 (1908). EL ” 1923.] Fauna of the Chilka Lake: Fish. 729 Distribution.—Seas and coasts of India including estuaries, the Malay Archipelago and China. Gerres punctatus, Cuvier and Valenciennes. 1803. Zeus sp. (wodawahah), Russell, Fish. Vizag. I, p. 52, pl. xvii. 1803. Zeus sp. (woodan), id., ıbid., p. 53, pl. Ixviii. 1830. Gerres punctatus, Cuvier and Valenciennes, Hist. Nat. Poiss. VI, p. 480. 1830. Gerres filamentosus, id., ibid., p. 482. 1849. Catochaenum filamentosum, Cantor, Journ. Asiat. Soc. Bengal, p. 1038. 1850. Gerres filamentosus, Bleeker, Verh. Batav. Gen. XXIII, p. 10. 1853. Gerres filamentosus, id., ibid., XXV, p. 40. 1853. Gerres punctatus, id., ibid. 1859. Gerres filamentosus, Günther, Cat. Fish. Brit. Mus. I, p. 345. 1859. Gerres punctatus, id., rbid., p. 346. 1862. Gerres punctatus, id., ıbid., IV, p. 260. 1862. Gerres filamentosus, id., 1bid., p. 261. 1863. Diapterus filamentosus, Bleeker, Ned. Tijd . Dierk. 1, p. 231. 1867. Gerres filamentosus, Jouan, Mém. Soc. Imp. Sci. Nat. Cherbourg XIII (Ser. 20), p. 263. 1873. Diapterus punctatus, Bleeker, Ned. Tijds., Dierk. IV, p. 140. 1875. Gerres filamentosus, Day, Fish. Ind., p. 98, pl. xxv, fig. 3. 1877. Diapterus filamentosus, Bleeker, At. Ichthyol. VIII, p. 124, pl. ecclsii, fig. 31. 1889. Gerres filamentosus, Day, Faun. Brit. Ind. I, p. 98, fig. 163. 1903. Xystaema filamentosus, Jordan and Evermann, Proc. U. 8. Nat. Mus. XXV, p. 352. 1904. Gerres filamentosus, Fowler, Journ. Acad. Nat. Sci., Philadelphia (2) XII, p. 530. 1905. Xystaema filamentosum, Jordan, Stud. Fish. II, p. 348. 1905. Xystaema punctatum, Jordan and Seale, Proc. U. S. Nat. Mus. XX VIII, p. 782. 1906. Xystaema punctatum, id., Bull. U. S. Bur. Fish. XXV, p. 272. 1907. Xystaema punctatum, Jordan and Seale, Bull. U. S. Bur. Fish. XXVI, p. 24. 1908. Xystaema punctatum, Seale and Bean, Proc. U. S. Nat. Mus. XXXIIL p. 244. 1911. Xystaema punctatum, Jordan and Richardson, Mem. Carnegie Mus. IV, p. 190. 1912. Gerres filamentosus, Snyder, Proc. U. S. Nat. Mus. XLIT, p. 501. 1913. Gerres filamentosus, Weber, ‘Siboga’-Exped. Fisch., p. 271. 1913. Gerres filamentosus, Gilchrist and Thompson, Ann. S. Afric. Mus. XI, p. 33. 1917. Gerres filamentosus, Hornell, Madras Fish. Bull. XI, p. 9. G. punctatus is evidently the same as G. filamentosus (the depth, 34 in the total length with caudal, in G. punctatus is true of the very young stage only till 24 inches long—in the adult it is 3 or a little less). The name punctatus appears in the same work as filamentosus, but, being on an earlier page, has priority.! There is only one specimen in the collection, 76 mm. in length, caught near Satpara in October, 1914. The dorsum is brown and the fins are dull yellow (in spirit) and the sneut is not black. The specimen possesses an adipose eye-lid. The species is only a casual visitor to the lake, and does not proceed further inwards than the outer channel. 1 Jordan and Seale, Bull. U. S. Bur. Fisher. XXV, p. 272. 730 Memoirs of the Indian Museum. | [Vou V, Distribution. —-Red Sea, seas of India, Malay Archipelago, Indo-Australian Archi- pelago, China, Philippines and Formosa. Genus LEIOGNATHUS, Lacépède. Leiognathus equulus (Forskäl). 1758. Scomber flavescens latitudine ad longitudinem dimidea denticulis piliformibus, Artedi, Descrip. Exac. Princ. Curios. Natur. Cab. Seba, III, p. 75, pl. xxvii, fig. 4. 1775. Scomber equula, Forskäl, Descrip. Anim. Pisc. p. 58. 1785. Scomber edentulus, Bloch, Allgem. Natur. Fisch., pl. cecexxviii. 1788. Centrogaster equula, Linnaeus and Gmelin, Syst. Natur. (Ed. xiii), I, p. 1337. 1801. Scomber edentulus, Bloch and Schneider, Syst. Ichthyol. p. 36. 1802. Caeses equulus, Lacépéde, Hist. Natur. Ponss. III, pp. 85, 90. 1803. Leiognathus argenteus, id., 1bid., IV, pp. 448 and 449. 1803. Zeus sp. (tottah karah), Russell, Fish. Vizagapatam, I, p. 49. pl. xii. 1804. Scomber equula, Shaw, Gener. Zool. IV, p. 596. 1835. Equula ensifera, Cuvier and Valenciennes, Hist. Nat. Poiss. X, p. 66. 1835. Equula caballa, id., ıbid., p. 73. 1836. Equula ensifera, Cuvier, Reg. Anim., Poiss., p. 139. 1836. Equula totta, id., rbid. 1838. Equula caballa, Rüppell, Neu. Warbel. ‘Faun. Ab: yssinien Gehör. Fisch. (Roth.-Meer), pp. 51 and 52. 1848. Equula serrulifera, Richardson, Zool. Voy. ‘Erebus’ and ‘Terror’ II, Ichthyol. p. 137, pl. lix, figs. 12 to 14. 1849. Equula caballa, Cantor, Journ. Asiat. Soc. Bengal, p. 1128. 1851. Equula caballa, Jerdon, Madras Journ. Lit. Sci., p. 138. 1852. Equula ensifera, Bleeker, Verh. Bat. Gen. XXIV, p. 8. 1860. Equula edentula, Günther, Cat. Fish. Brit. Mus. II, p. 498. 1860. Equula caballa, id., ibid., p. 499. 1863. Leiognathus edentulus, Bleeker, Ned. Tijdsch. Dierk. I, p. 235. 1865. Leiognathus edentulus, vd., ibid., II, p. 148. 1865. Equula ensifera, Kner, Reis. ‘ Novara,’ Fisch., p. 166. 1865. Equula edentula, Day, Fish. Malabar, p. 103. 1866. Equula edentula, Playfair, Fish. Zanzbar, p. 65. 1869. Equula ruconius, Day (not Hamilton Buchanan), Proc. Zool. Soc. London, p. 302. 1871. Equula caballa, Klunzinger, Verh. Zool.-Bot. Ges. Wien, p. 467. 1871. Equula edentula, id., ibid. 1873. Leiognathus edentulus, Bleeker, Versl. Akad. Amsterdam (2) VII, p. 37. 1875. Leiognathus edentulus, id., Poiss. Madagascar et Réunion, p. 98. 1876. Equula edentula, Day, Fish. Ind., p. 238, pl. lu, fig. 1. 1879. Levognathus edentulus, Verh. Akad. Amsterdam XVIII, p. 18. 1885. Equula edentula, Vinciguerra, Ann. Mus. Civ. Stor. Nat. Genova (2) II, XXII, p. 88. 1889. Æquula edentula, Day, Faun. Brit. India, Fish. Il, p. 186, fig. 65. 1890. Equula edentula, Vinciguerra, Ann. Mus. Civ. Stor. Nat. Genova (2) IX, p. 171. 1903. Letognathus edentulum, Jordan and Evermann, Proc. U. S. Nat. Mus. XXV, p. 338. 1905. Leiognathus edentulus, Fowler, Proc. Acad. Nat. Sci. Philadelphia LVII, p. 510. 1923.] Fauna of the Chilka Lake: Fish. 731 1906. Leiognathus edentulus, Jordan and Seale, Bull. U. 8. Bur. Fish. XXV, p. 273. 1906. Leiognathus equula, Jordan and Seale, Bull. U. S. Bur. Fish. XXV, p. 273. 1907. Leiognathus edentula, Evermann and Seale, Bull. 1. S. Bur. Fish. XXVI, p. 69. 1907. Equula edentula Lloyd, Rec. Ind. Mus., I, p. 228. 1908. Leiognathus edentulus, Seale and Bean, Proc. U. S. Nat. Mus. XXXIII, p. 242. 1908. Equula edentula, Gilchrist and Thompson, Ann. 8. Afric. Mus. VI, p. 188. 1911. Leiognathus edentulum, Jordan and Richardson, Mem. Carnegie Mus. IV, p. 180. 1912. Leiognathus argentium, Snyder, Proc. U. S. Nat. Mus., XLII, p. 412. 1912. Leiognathus edentulus, Bean and Weed, Proc. U. S. Nat. Mus. XLII, p. 604. Artedi’s specific name, although the earliest, being polynomial in form is inadmissible. Cuvier created the genus Æquulat, taking Centrogaster equula of Linnaeus and Gmelin? (which is Scomber equula of Forskäl) as its type and named his newly created genus, as was his wont, after the specific name of the type, at the same time supplying a new name for the already named species by dropping the old and earliest specific name. Therefore the name of the species should have been Equula equula, even if there were any justification for the newly created Cuvierian generic name. Cuvier’s objection to “Lesognathus ” of Lacé- péde was its etymological meaning, 2. e., “toothless.” Lacépéde in separating the new genus Leiognathus from the old genus ““Scomber’’ meant to take out all those species which did not possess any conspicuous teeth.? Cuvier and Valenciennes contended that as the group thus taken out actually possessed teeth, though minute, the name Leiognathus was not only inappropriate but also ineligible and therefore must go.* Thus Lacépéde’s generic name was discarded and Cuvier, after raising the specific name of the first author (¢.e., Forskäl) to that of a genus, substituted the specific name caballa tor equula of Forskal and ensifera for edentulus of Bloch, considering these two to be two distinct species and paying no regard to the law of priority. Giinther, though he remarked that he had no hesitation 5 in considering the two species as identical, recorded them under different names as distinct species. He, however, restored Bloch’s name edentulus in place cf ensıfera, but left the Cuvierian name caballa for equula of Forskal. The argument against the earlier name Leiognathus is no longer considered valid, hence the generic name Æquula is ineligible. It is regrettable that the familiar name of a well-known species must be altered.® There are altogether thirteen specimens in the collection. The fish is found all over the lake, including the outer channel, throughout the year. It is a permanent inhabitant, probably breeding in the lake during the flood-season. 1 Cuvier, Reg. Anim. (Ed. I), II, p. 323 (1817). ? Linnaeus and Gmelin, Sys. Nat. III, p. 1337 (1788). ® Lacépède, Hist. Nat. Poiss. IV, 449. 4 Cuvier and Valenciennes, Hist. Nat. Poiss. X, pp. 60. 61, and 67. > Günther, Cat. Fish. Brit. Mus. II, p. 499. 5 Houttuyn in 1782 reported “‘Centrogaster argentatus” from Nagasaki. (Verh. Hollandsche Maatsch. Weelen. Haarlem XX, pp. 311—346). As Houttuyn’s descriptions represent the earliest record of Japanese fishes his names must have precedence over all others when his descriptions can be identified. Jordan and Snyder in their “List of Japanese Fishes” point out that it is identical with Æquula nuchale of Temminck and Schlegel (Faun. Japonica, Poiss., p. 126, pl. Ixvii, fig. i), which is one of the commonest of Japanese fishes; but the name should be Leiognathus argentatum (Proc. U. $. Nat. Mus. XXIII (1901), p. 747) and the name should be restricted to Japanese species. Forskal’s name is applicable to the species from the Rea Sea and the Seas of India. 732 Memoirs of the Indian Museum. [Vor. V. The following statement shows the different localities in the lake whence the specimens. were obtained, and their number and size :— mm. 2specimens ... Off Barkul ... … 9-13th November, 1912 ... 25 and 41 2 55 … Barkul Bay ... 1st March, 1914 on ot and'38 1 specimen … East of Barkul bung- low ae ... 3rd March, 1914 OL eo 2 specimens ... Chirriya Island ... 18th February, 1914 … 30 and 32 2 5 ... Rambha Bay ... February, 1914 .. 60 and 54 2 Ped ae A + … March, 1914 doo EM incl BY 2 = ... Satpara sae ... 10th October, 1914 MAMAN Distribution. —Red Sea, seas of India, Malay Archipelago, Australian coasts, New Guinea, Formosa and Japan. Leiognathus blochii (Cuvier and Valenciennes). 1835. Equula blochii, Cuvier and Valenciennes, Hist. Nat. Poiss. X, p. 84. 1835. Zeus notatus, id., 1bid., (from ms. of Bloch). 1853. Equula blochii Bleeker, Verh. Bat. Gen. XXV, p. 46. 1865. Hquula blochu, Day, Fish. Malabar, p. 105. 1876. Equula blochi, id., Fish. India, p. 241, pl. lu, fig. 3. 1389. Equula blochu, id., Faun. Brit. India Fish. TI, p. 189. Bloch named this fish Zeus notatus from specimens sent to him from Tranquebar. This name, however, remained in manuscript until it was noticed by Cuvier and Valenciennes ! who identified Bloch’s species with specimens from Malabar. But Cuvier and Valenciennes renamed it, as was usual with them, out of respect for the author who first named the species. Cuvier and Valenciennes were the first to publish Bloch’s name along with the new name they substituted for it. Bloch only named the species but did not describe it ; more- over, Bloch’s name in its first publication is printed after the name given by Cuvier and Valenciennes. Günther recorded it as a doubtful species,? but it is generally regarded to be a valid one and is believed to be restricted to Indian waters. The name given to it by Cuvier and Valenciennes must stand though it is regrettable that Bloch’s original name was not adopted. There are altogether seven specimens in the collection. This fish appears to be a permanent resident in the lake and is found throughout the main area as well as in the outer channel during the dry months. The following statement shows the different localities where the specimens were obtained and their number and size :— mm. 3 specimens … Barkul Bay ... ... Ist March, 1914 .. 985, 40 & 41 1 specimen ... Between Kalidai and | Samal Island ... 20th February, 1914 u 32 1 = ... South of Kalidai ... 2st February, 1914 a 26 1 2 … Kaluparaghat ben ca pe 49 1 u ... Satpara 52 ... March, 1914 a 57 Distribution.—Seas of India. 1 Cuvier and Valenciennes, Hist. Nat. Poiss. X, p. 84. 2 Günther, Cat. Fish. Brit. Mus. II, p. 498. 1923.] 1788. 1193. 1801. 1803. 1835. 1835. 1835. 1835. 1844. 1845. 1849. 1851. 1851. 1851. 1853. 1860. 1860. 1860. 1863. 1865. 1866. 1871. 1871. 1876. 1876. 1881. 1881. 1888. 1889. 1889. 1889. 1903. 1905. 1905. 1905. 1905. 1905. 1907. LORIE 1912. 1913. OKT. STE Fauna of the Chilka Lake : Fish. 733 Genus GAZZA, Rüppell. Gazza minuta (Bloch). Zeus argentarius, Forster, Descrip. Animal. Mar. Australis (ms.) Scomber minutus, Bloch, Natur. Ausl. Fische, pl. ecccxxix, fig. 2. Zeus argentarius, Bloch and Schneider, Syst. Ichthyol. I, p. 95. Zeus sp. (komah karah), Russell, Fish Vizagapatam I, p. 60. pl. Ixüii. Equula coma, Cuvier and Valenciennes, Hist. Nat. Powss. X, p. 76. Equula minuta, id., ibid., p. 88. Equula dentex, rd., ibid., p. 91. Gazza equulaeformis, Rüppell, Neu.-Wirbelth. Fische, p. 4, pl. 1, fig. 3. Zeus argentarius, Forster and Lichtenstein, Descr. Anim., p. 288. Gazza equulaeformis, Bleeker, Nat. Geneesk. Arch. Ned. India II, p. 518. Gazza equulaeformis, Cantor, Journ. Asiat. Soc. Bengal, p. 1135. Gazza minuta, Bleeker, Nat. Tijdsch. Ned. India II, p. 213. Equula coma, Jerdon, Madras Journ. Lit. Sa., p. 138. Equula minuta, id., ibid. Gazza tapeinosoma, Bleeker, Nat. Tijdschr. Ned. India IV, p. 260. Gazza minuta, Günther, Cat. Fish. Brit. Mus. IT, p. 506. Gazza equulaeformis, 1d., ibid. Gazza argentaria, id., 1bid. Gazza argentarius, Bleeker, Ned. Tijdschr. Dierk. I, p. 242. Equula dentex, Kner, Reis. ‘Novara, Fisch., p. 170. Gazza equulaeformis, Playfair, Fish. Zanzibar, p. 65. Gazza argentaria, Klunzinger, Verh. Zool.-Bot. Ges. Wien XXI, p. 467. Gazza equulaeformis, rd., ibid., p. 468. Gazza minuta, Day, Fish. India, p. 244, pl. ln, fig. 1. Gazza aequulaeformis, id., ibid. Gazza equulaeformis, Günther, Journ. Mus. Godeff. IV, p. 144. Gazza argentaria, id., ibid., p. 144. pl. xci, fig. B. Gazza argentaria, Day, Fish. Ind. Suppl., p. 790. Gazza minuta, id., Faun. Brit. India, Fish., p. 194, fig. 66. Gazza equulaeformis, 1d., ibid. Gazza argentaria, id., vbid., p. 195. Gazza equulaeformis, Jordan and Evermann, Proc. U. S. Nat. Mus. XXV, p. 338. Gazza minuta, Jordan and Seale, Proc. U. S. Nat. Mus. XXVIII, p. 777. Gazza minuta, id., Bull. U. S. Bur. Fisher. XXV, p. 273. Gazza argentaria, id., ibid. Gazza equulaeformis, id., ibid. Gazza minuta, Jordan, Guid. Stud. Fish. Il, 287. Gazza minuta, Smith and Seale, Proc. Biol. Soc. Washington, XIX, p. 77. Gazza equulaeformis, Jordan and Richardson, Mem. Carnegie Mus. IV, p. 181. Gazza minuta, Bean and Weed, Proc. U. 8. Nat. Mus. XLII, p. 604. Gazza argentaria, Weber, ‘Siboga’-Exped. Fisch., p. 270. Gazza minuta, Jordan and Starks, Ann. Carnegie Mus. XI, p. 444. Gazza equulaeformis, Hornell, Madras Fish. Bull. XI, p. 92. 734 Memoirs of the Indian Museum. FVor. V, Weber sinks G. minuta, G. equulaeformis and G. tapeinosoma, in the synonymy of @. argentaria, J. R. Forster (1729-1798), described by him in his Descriptiones Animalium. But this work remained in manuscript till 1844 m which year it was published by Lichtenstein. In 1801, Schneider published Bloch’s Ichthyology, in which Forster’s name, Zeus argentarius, was first published with his description. Thus the name “Scomber minuta” was the earliest, being published in 1795, and has therefore priority. There is only one specimen in the collection 69 mm. in length. It was secured near Nalbano on 25th November, 1914. Probably the fish is only a casual visitor to the lake. Distribution.—Zanzibar, Red Sea, Hast Indian seas, Malay Archipelago, Indo-Aus- tralian Archipelago, Polynesia (Samoa), New Hebrides and the Philippines. Family SCORPIDIDAE. Genus MONODACTYLUS Lacépéde. Monodactylus argenteus (Linnaeus). 1754. Chaetodon argenteus, Linnaeus, Chinens. Lagerstorm. Amoen. Acad. IV, p. 249, No. 26. 1758. Chaetodon argenteus, id., Syst. Natur. Hd. X, p. 272. 1775. Scomber rhombeus, Forskäl, Deserip. Animal. Pisc., p. 58. 1788. Chaetodon argenteus, Linnaeus, Syst. Natur. (Gmelin), I, p. 1242. 1788. Centrogaster rhombeus, id., ibid., p. 1338. 1800. Monodactylus falciformis, Lacépéde, Hist. Nat. Poiss. II, pl. V, fig. 4. 1801. Chaetodon argenteus, Bloch and Schneider, Syst. Ichthyol., p. 230. 1802. Monodactylus falciformis, Lacépéde, Hist. Natur. Poiss. III, pp. 131 and 132. 1802. Centropodus rhombeus, id., ibid., pp. 303 and 304. 1802. Acanthopodus argenteus, id., ibid., pp. 558 and 559. 1803. Zeus sp. (kankı sandwa), Russell, Fish. Vizagapatam I, p. 47, pl. lix. 1803. Scomber rhombeus, Shaw, Gen. Zool. IV, p. 595. : 1830. Centropodus rhombeus, Desmarest, Oeuv. Lacép. VIII, p. 132. 1831. Psettus rhombeus, Cuvier and Valenciennes, Hist. Natur. Poiss. VII, p. 245. 1831. Psettus commersonw, id., ibid., p. 250. 1834. Psettus rhombeus, Cuvier, Reg. Anim., Ponss., p. 111, pl. xlüi, fig. 2. 1834. Monodactylus rhombeus, Griffith, Cuv. Anim. K ngdom X, pl. lv, fig. 2. 1839. Monodactylus rhombeus, Swainson, Nat. Hist. Fish. Amphib. Rep. II, p. 212. 1846. Psettus argenteus, Richardson, Rep. Brit. Assoc: Adv. Sci. (1845), p. 246. 1848. Psettus argenteus, id., Zool. Voy. ‘Erebus’ and ‘Terror, Fish, p. 57, pl. xxxv, figs. 1-3. 1849. Monodactylus rhombeus, Cantor, Journ. Asiat. Soc. Bengal, p. 1154. 1853. Psettus rhombeus, Bleeker, Verh. Batav. Genoot. XXV, p. 40. 1855. Psettus argenteus, rd., Verh. Akad. Amsterdam II, p. 10. 1855. Psettus argenteus, Peters, Arch. Naturgesch. p. 247. 1860. Psettus argenteus, Günther, Cat. Fish. Brit. Mus. II, p. 487. 1860. Psettus falciformis, id., ibid., p. 488. 1863. Monodactylus argenteus, Bleeker, Ned. Tijdschr. Dierk. I, p. 242. 1865. Psettus argenteus, Kner, Reis. ‘Novara, Fisch., p. 164. 1865. Psettus argenteus, Day, Fish. Malabar, p. 99. 1865. Psettus faleiformis, id., ibid., p. 100. 1923.] Fauna of the Chilka Lake: Fish. 735 1866. Psettus argenteus, Playfair, Fish. Zanzibar, p. 64. 1871. Psettus argenteus, Klunzinger, Verh. Zool.-Bot. Ges. Wien XX, p. 794. 1875. Monodactylus argenteus, Bleeker, Poiss. Madagascar, p. 65. 1876. Psettus falciformis, Day, Fish India, p. 234, pl. LA, fig. 6. 1876. Psettus argenteus, id., ibid., p. 235, pl. li B, fig. 5. 1876. Psettus argenteus, Günther, Journ. Mus. Godefroy, Fisch., p. 140. 1879. Psettus argenteus, Bleeker, Verh. Akad. Amsterdam, p. 18. 1880. Psettus argenteus, Günther, Introd. Study Fish., p. 448, fig. 199. 1889. Psettus falciformis, Day, Faun. Brit. India, Fish. IT, p. 180. 1889. Pseitus argenteus, id., ibid., fig. 62. 1905. Monodactylus argenteus, Jordan, Guide Study Fish. II, p. 398. 1906. Monodactylus argenteus, Stead, Fish. Australia, p. 133, fig. 49. 1906. Monodactylus argenteus, Jordan and Seale, Bull. U. S. Bur. Fish. XXV, pp. 178 and 237, fig. 30. | 1907. Psettus argenteus, Lloyd, Rec. Ind. Mus. I, p. 227. 1907. Monodactylus argenteus, Evermann and Seale, Bull. U. S. Bur. Fish. XXVI, p. 32. 1907. Monodactylus argenteus, id., ibid., p. 71. 1909. Psettus argenteus, Goodrich, Treat. Zool. Cyclost. Fish., p. 432, fig. 440. 1917. Psettus argenteus, Hornell, Madras Fish. Bull. XI, p. 91. Jordan and Fowler proposed a new family Platicidae to include the genera Monodactylus, Platax and Psettias (a new genus created by Jordan). This is a small group of fishes of the Asiatic seas related to the Chaetodontidae, but showing differences in the skeleton.1 In Monodactylus and Psettias the ventral fins are rudimentary and the body is still deeper in both than in Platax. In Monodactylus it is less deep than in Psettias and is not deeper than long, whereas in Psettias it is deeper than long.” Commerson proposed to unite three genera of Lacépède, viz., Monodactylus,® Centropodus* and Acanthopodas,? erroneously thinking that these had “no teeth in the palate,’ under the name of Psettus. The mistake is repeated by Günther® and Day. Commerson’s description of the genus was in manuscript until it was published by Cuvier and Valenciennes in 1831.7 Moreover, Monodactylus has priority, having been created by Lacépéde as early as 1802. There are three specimens in the collection, all caught off the coast of Parikud. The dates of their capture are, however, not recorded. Three specimens measure respectively 85 mm., 89 mm. and 97 mm. in length and all are evidently young as the full grown adults of this bat fish are said to reach a length from 180 mm. to 250 mm. and over. All the three specimens show the orbicular and the opercular dark-brown or black bands charac- teristic of the young. The disappearance of these in more mature forms and the alteration of proportion of parts consequent on growth have led to the creation of a very large number of species and even a few genera out of this one single fish, as the long list of its synonymy 1 Jordan and Fowler, Proc. U. S. Nat. Mus. XXV. p. 525. 2 Jordan and Seale, Bull. U. S. Bur. Fish. XXV, p. 236. 3 Lacépède, Hist. Nat. Poiss. III, p. 131. j 4 Id., op. cit., III, p. 303. 5 Id., op. cit., IV, 558. 5 Günther, Introd. Study Fish., p. 447 and Cat. Fish. Brit. Mus., II, p. 486. 7 Cuvier and Valenciennes, Hist. Nat. Poiss. VII, p. 240. Pr 736 ‘Memoirs of the Indian Museum. [Vor. V, 1923.] proves. The colour of the curved portions of the free anterior ends of the dorsal and the anal fins is dark brown in all the three specimens and that of the pectoral and the caudal fins is dull yellow. The fish is an occasional visitor to the main area of the iake and does not appear to breed in it. . Distribution. —The geographical range of this fish is very extensive; from the Red Sea through the east coast of Africa, Zanzibar and Aden to Indian seas, the Malay Peninsula, the Malay Archipelago, Polynesia (Samoa), seas of Australia and China and the Philippines. It is reported to be most common in Malabar and Coromandel during monsoon months and rather abundant in the harbour of Apia, Port Jackson and Singapore. FAUNA OF THE CHILKA LAKE. FISH PART V. By Sunper Lat Hora, D.Sc. (With 14 text-figures.) a CONTENTS. Page. Introduction... 3 ie ee ihe an wee a Re) Gobius ostreicola eine ch Eos me “i oe see HAO ,, albopunctatus Cuvier and Valenciennes see oan ae oa Br) Glossogobius giuris (Hamilton Buchanan) ... 300 bo a soap TAN i biocellatus (Cuvier and Valenciennes) ane a eae wy el 59 mas, Sp. NOV. os ey 3e de Sr ey ee Cienogobius acutipinnis (Cuvier and Valenciennes) Be er ag cc, WSS = chilkensis (Jenkins) ere ue LA ER 144 8 alcocki (Annandale) ... Bes 33, EN is … 144 a globiceps, sp. nov... wk sis nes Acs … (44 = cylindriceps, sp. nov. eae 503 one ae … 7145 Re dentifer, sp. nov. nee bs ae be a vat A Br minima, SP. NOV. A ; ee ve at sae ue ao Oxyurichthys tentaculatus (Cuvier and No don er sae sole! OT) Apocryptes rictuosus (Cuvier and Valenciennes) Me ash me TOI lanceolatus (Bloch and Schneider) bse BAC Br a Micrapocryptes fragilis, gen. et. sp. nov. M 00 ie wi oa eon Eleotris cavifrons Blyth a =a so ae he oh wie) » fusca (Bloch and Schneider) er a Bele a 29705 ESP: u Re Er ar ae a Ro US Butis butis (Hamilton Buchanan) 2% oa I ve ee Periophthalmus koelreuteri (Pallas) ce ai cai an a 90 Taenioides chilkensis, sp. nov. … Bae sae acs 5X fae CONS Pseudorhombus arsius (Hamilton Buchanan) ... cae ste a eee OS Synaptura orientalis (Bloch and Schneider) ... a ae ee BRE 200) Cynoglossus brevis Günther Inn a de Le Fa = O0 Caranz carangus (Bloch) a ne 0 Bee BR nie COM Equula edentula (Bloch) sae fs is de sae ae ail Petroscirtes bhattacharyae Chaudhuri iy aie Be is en Le ee JM 86 No. of rays in anal Fe oa jr Le fe 55 No. of rays in caudal Ve ee Mos sss oe de 15 No. of rays in ventral BA De La oa see 5 Pseudorhombus arsius extends from the east coast of Africa, through the seas and estua- ries of India, to Australia and China. Family SOLEIDAE. Genus SYNAPTURA Cantor. Synaptura orientalis (Bl. & Schn.). 1878. Synaptura orientalis, Day, Fish. India IT, p. 429, pl. xeiüi, fig. 4 ; pl. xciv, fig. 2. In the collection from the lake there is only one specimen about 17-6 cm. in length includ- - ing the caudal fin. It was collected near Parikudh. There are 67 rays in the dorsal fin and 51 in the anal. The number of scales along the lateral line is 91. The depth of the body is contained about two and a half times and the length of the head five and a half times in the total length includingthe length of the caudal fin. The scales in the anterior part of the head on the blind side and also those along the gill-openings on both sides have their ctenoid processes greatly produced and the body surface in that region appears as if covered with soft cutaneous filaments. The scales of this type have already been figured by Gilchrist for his species, Synaptura ciliata. In S. orientalis the processes are much longer than those figured for S. ciliata. In addition to these there are tufts of black hair-like filaments coming out from between the scales on the coloured side. S. orientalis is found in the seas of India and China. The following are the measurements of the Chilka Lake specimen in millimetres. Total length, including length of caudal aap he ge G20 Length of caudal os aM ke We 23:5 Length of head ... Ace >20 nae sé +. 31-2 LE 1 Gilchrist, Mar. Invest. South Africa III, p. 14, pl. xxxiv (1905). H 760 Memoirs of the Indian Museum. [Vor.V, Diameter of lower eye, Le a a idl 2. 4-5 Length of pectoral fin of right side > ak He bile 12-0 Length of pectoral fin of left side Een Su Fe ie 9-0 Longest ray of dorsal ds aie ae ae oie 14-0 Longest ray of anal a ee ne iy bi 13-5 Greatest depth of body ae 900 UM ne hi 71-0 Length of pelvic fin Ae I De ER ne 8-5 Family CYNOGLOSSIDAE. Genus CYNOGLOSSUS Hamilton Buchanan. Cynoglossus brevis Günther. 1862. Cynoglossus brevis, Günther, Cat. Brit. Mus. Fish. IV, p. 500. 1878. Cynoglossus brevis, Day, Fish. India II, p. 437, pl. xevii, fig. 2. There are sixty specimens of this species in the collection. It has so far been known from the river Hughli at Calcutta and is mainly a brackish water form. There is, however» one specimen in the old collection of the Indian Museum obtained by the Marine Que from along the Orissa Coast (off Chilka Lake). The colour varies greatly with age. In the young there are several short black bands on the body, which with the growth of the fish become thinner and more numerous and ulti- mately produce a reticulum in the adult. There is one very characteristic feature of the species, that several rays of the vertical fins at intervals are dark in colour. The longest specimen is about 14-5 cm. in length including the caudal fin. The list given below shows the place and time of occurrence of the species in the lake :— 1 specimen ... . Gopkuda Bays 22 … 14 February 1914. 1 A Rn 0) (Otte Banlcucla lan ... 17 February 1914. 1 5 a … South of Kalidai ... 21 February 1914. 1 3 gee MORT ie Baliye. a ... 23 February 1914. 1 Ps Ake … mile off Kalidai fe 1 March 1914. 3 specimens … 1 mile South of Kalidai Der 1 March 1914. 3 ps ade | | Ott Barkulabayen bea 1 March 1914. 4 4 a ... Palari to Gantasila Ap 3 March 1914. 4 young «. … West of Satpara I. ... 20 March 1914. Be) ws a … Shore of Mahosa .. ' 20 March 1914. 13 specimens ... Channel between Satpanı ai Barhampur I. 2 2 September 1914. 5 ce wee .. Channel off un Le 2 September 1914. 14 53 wae .... Channel off Barhampur I. ... 2 September 1914. 24 ” er . Channel between Satpara and Barnikuda aa bes 4 September 1914. 1 young. a. … Island near Manikpatna ... 7 September 1914. 1 specimen ... … Serua Nadi 3% 8 September 1914. il N ees … About 8 miles South-East ‘of Kalupara see … 16 September 1914. 1 is AR ... Mouth of Rambha Bay … 17 November 1914. 1923.] Fauna of the Chilka Lake : Fish. 761 Family CARANGIDAE. Genus CARANX Lacépède. Caranx carangus (Bloch). 1876. Caranx carangus, Day, Fish. India I, p. 215, pl. L fig. 4. The species is fairly common in the lake and is found both in the main area and the outer channel. It appears to be a permanent inhabitant of the lake. The largest specimen is 12-6 cm. in length and was obtained in Rambha Bay in ‘February, 1914. | Distribution.—The seas of India, the Malay Archipelago and the Atlantic coasts of tropical America. Genus EQUULA Cuv. and Val. Equula edentula (Bloch). 1876. Equula edentula, Day, Fish. India I, p. 238, pl. lu, fig. 1. The species is represented by 14 specimens, of which 12 were collected in the main area, while the remaining two were obtained near Satpara. The species appears to be a permanent inhabitant of the lake. Distribution —The Red Sea, the seas of India, the Malay Archipelago and beyond. Family BLENNIIDAE. Genus PETROSCIRTES Riippell. Petroscirtes bhattacharyae Chaudhuri. 1916. Petroscirtes bhattacharyae, Chaudhuri, Rec. Ind. Mus. XII, p. 107. 1916. Petroscirtes bhattacharyae, Bhattacharya, Mem. Ind. Mus. V, p. 385, pl. xvii, figs. 8-11 (young stages). The species occurs in the lake throughout the year and breedsin it. The specimens were collected both in the main area and the outer channel. TExT-FIG. 35.—Petroscirtes bhattacharyae Chaudhuri: X 3. The fish has so far been found only in the Chilka Lake. 2 762 Memoirs of the Indian Museum. [Voz. V, Family PLATYCEPHALIDAE. : Genus PLATYCEPHALUS Bl. & Schn. Platycephalus insidiator (Forskäl). 1876. Platycephalus insidiator, Day, Fish. India, I, p. 276. A few specimens of this species were taken in the outer channel, to which it is evidently an occasional visitor. - Disribution—The Red Sea, east coast of Africa, the seas of India, the Malay Archi- pelago and beyond. Sub-order OPISTHOMI. Family MASTACEMBELIDAE. Genus MASTACEMBELUS Cuv. and Val. Mastacembelus armatus (Lacépède). 1878. Mastacembelus armatus, Day, Fish. India, II, p. 340, pl. Ixxiii, fig. 2. The species is represented by two specimens captured near Patsahanipur. The largest is about a foot and half in length. The fish is probably an occasional immigrant from fresh waters in the rainy season. Distribution.—The fish occurs in the fresh and brackish waters of India, Ceylon and Burma. Its range extends as far as China. SUMMARY OF THE REPORT ON THE FISH OF THE CHILKA LAKE. The fish-fauna of the Chilka Lake comprises in all 118 species, of which 13 have been found to be new. Most of the new species, as many as seven, belong to the family Gobiidae, while the remaining six are distributed among the families Clupeidae, Siluridae, Ophich- thyidae and Sphyraenidae. A new genus, Micrapocryptes, has been erected in the family Gobiidæ to accommodate the small transparent Gobies of the lake and the Gangetic Delta. In the genus Hleotris a specimen is recorded without any specific name. It probably represents a species hithereto undescribed but | have refrained from naming it on account of the immaturity of the single specimen. Of the 118 species, 49 were taken in the main area only, 24 in the outer channel only and 39 from all over the lake. The distribution of the remaining six species in the lake is not known because they appear to have been purchased from local fishermen from time to time. The most noteworthy feature in the physical environment of the fauna of the lake is the great periodic change in salinity. This has been fully discussed in the introduction to this volume by Annandale and Kemp (pp. 6—10) and also by Sewell in his paper on Rambha Bay (pp. 680-690). Most of the fishes mhabiting the lake are known from the estuarine waters of India and the Malay Archipelago, but certain forms such as Pseudo- rhombus arsius, Gerres punctatus, G. öyena and Gobius albopunctatus, which have hitherto 1923.] Fauna of the Chilka Lake: Fish. 763 been known from sea water, were only captured in the outer channel when the water was quite fresh. Other sea fishes such as Bleotris fusca, Therapon puta, Priopis gym- nocephalus, Tetrodon reticularis, Mugil cunnesius, Dorosoma nasus and Trygon uarnak, have been found to live in salinities varying from fresh water to water as saline as that of the Bay of Bengal. Several species immigrate into the lake at a certain period. Purely freshwater fishes such as Mastacembelus armatus and Ophiocephalus punctatus visit the main area during the rains, while forms like Periophthalmus koelreuteri and Rhabdura macrura are found in the outer channel during a period of drought when the water is as salt as that of the sea outside. I hope to have an opportunity shortly of comparing the fish-fauna of the Chilka Lake with that of the Tale Sap, a somewhat similar lagoon connected with the Gulf of Siam, and to publish a paper in vol. VI of the Memoirs of the Asiatic Society of Bengal. In the following table the species are arranged in the order in which they have been treated in the report, with a statement as to the specific gravity of the water in which they were taken, their distribution in the lake and their geographical range. DISTRIBUTION IN Specific gravity LAKE. Species. en es Further distribution. taken. Main Outer area. channel. PLAGIOSTOMI. SELACHII CARCHARINIDAE. Physodon muller Müller and Henle . 1-01075 x Bengal and China. Carcharinus gangeticus (Müller and | 1-01075—1-02375 X Seas and estuaries of India, Henle). Japan and Fiji. Pe melanopterus (Quoy and 1-00325 X Seas of India and Malay Gaimard). Archipelago. BATOIDEI PRISTIDAE. Pristis pectinatus Latham 1-0010 X Tropical and temperate seas. TRYGONIDAE. Trygon uarnak (Forskäl) 1:000—1:0260 x Red Sea, Indian Ocean, Gulf of Siam and East Indies. ,, pareh Bleeker 1-000—1-0020 River Hughli, Bay of Bengal and Malay Archipelago. > wmbricata (Schneider) 1-00750—1-02375 X East Indies. Hypolophus sephen (Forskäl) 1:00750—1-02375 X X Red Sea, Indian Ocean and East Indies. MYLIOBATIDAE. Aetobatis flagellum (Schneider) . . ? X Tropical and semi-tropical waters of the world. 5 guttata (Bloch and Schneider) ? X Tropical parts of Indian Ocean. Aetomylaeus nichofii (Schneider) 1:0110 X Seas of India, East Indies and Japan. TELEOSTEI. MALACOPTERYGII. ELOPSIDAE. Elops indicus Swainson 1-00750—1-01050 Arabian Sea and Bay of Bengal, entering estuaries. Megalops cyprinoides (Broussonet) 1-00725 Indian and Pacific Oceans and their estraries. 764 Memoirs of the Indian Museum. [Voz. V, Species. TELEOSTE!—conid. MALACOPTERYGII—contd. CHANIDAE. Chanos chanos (Forskäl) . CLUPEIDAE. DOROSOMATIN AB. Dorosoma nasus (Bloch) . > indicus (Russel) ENGRAULINAE. Engraulis annandalei, sp. nov. 3; kempi, sp. nov. “A rambhae, sp. nov. - . 55 purava (Hamilton Bucha- nan). mystax (Bloch and Schnei- der). Stolephorus indicus (v. Hasselt) 35 commersonü Lacépède ss tri Bleeker. ° . CLUPEINAE. Clupeoides lile (Cuvier and Valen- ciennes). ie ilisha (Hamilton Bucha- nan.) OSTARIOPHYSI SILURIDAE. CLARIINAE. Platosus canius Hamilton Buchanan SILURINAE. Wallago attu (Bloch and Schneider) . Callichrous bimaculatus (Bloch) Pangasius pangasius (Cuvier and Valenciennes). Osteogeneiosus militaris (Linnaeus) . BAGAPINAE. Arius satparanus, Sp. NOV. > arius (Hamilton Buchanan) Specific gravity of water in which specimens were | taken. 1-000—1-0075 1-000—1-02650 1-000—1-0030 1-000 1-00750—1-00775 1-00750—1-00800 1-00750—1-02650 | | | 1-00750 1:000—1:00750 1:00750 1-00750 1-00750 | | | | | | | 1-000—1-02650 | 1-000—1-02650 1-0020 1-0020 1-000 1-000—1-0020 1-000 1-000 DISTRIBUTION IN LAKE. Main area. x = XS ” x Cv) x x Outer channel. Ka} Further distribution. Red Sea, East Coast of Africa, Indian and Pacific Oceans and their estuaries. South Arabia and Socotra, seas of India to Malay Archipelago, Philippines, Formosa and China. Seas and estuaries of India, Siam, Malay Archipelago and Philippines. Seas and estuaries of India, Malay Peninsula and Archi- pelago. Seas and estuaries of India, Malaysia and China. Seas of India and Malay Archipelago, Philippines, . Formosa, Japan, Samoa and Tahiti. R Madagascar, seas of India, Malay Archipelago and Philippines. Seas and estuaries of India, . Malay Archipelago and Philippines. West coast of India, Ceylon. Burma, Siam and Malay Archipelago. Persian Gulf, coasts and estua- ries of India and Siam. Seas, estuaries and rivers of India and the Malay Archi- pelago. Rivers and estuaries of India and the Malay Archipelago. Ditto. Ditto. Ditto. Estuaries of Orissa, Bengal and Burma. 1923.] Fauna of the Chilka Lake : Fish. 765 I I sn mn er en DISTRIBUTION IN Species. Specifie gravity of water in which specimens were taken. ° TELEOSTEI—conid. "OSTARIOPHYSI—contd. BAGARINAE—contd. Arvus caelatus (Cuvier ciennes). and Valen- ,, falcarius Richardson Macrones cavasius (Hamilton Bucha- nan). % gulio nan). vittatus (Bloch) (Hamilton Bucha- LE] CYPRINIDAE. CYPRININAE. | Cirrhina latia (Hamilton Buchanan) Barbus sophore (Hamilton Buchanan) | | ticto (Hamilton Buchanan) 22 vittatus Day . fi : APODES ANGUILLIDAE. Muraenesox cinereus (Forskäl) . MURAENIDAE. | Rhabdura macrura (Bleeker) . R OPHICHTHYIDAE. | Ophichthis chilkensis, sp. nov. . | hijala (Hamilton Bucha- nan). boro (Hamilton Buchanan) 2 LE] HAPLOMI CYPRINODONTIDAE. APLOCHEILINAE. Panchax panchax (Hamilton Bucha- nan). Aplocheilus melastigma McClelland CATOSTEOMI SYNGNATHIDAE. 3 Tchihyocampus carce (Hamilton Bu- chanan). Hippocampus brachyrhynchus, Dunker 1-0020—1-0070 1-0060 1-0020 1-0020—1-0260 1-000—1-0020 1-0080 1-000—1-0110 1-000—1-0110 1-0060—1-0110 1-000 1-000—1-0020 1-0020—1-0110 1-000—1-0080 1-0080 1-000—1-0110 LAKE. Main Outer area. channel. | X X x x x X X x de X ? 2 x x = X 4 x x x x X X X x Further distribution. Seas, brackish waters and rivers of India, Siam and the Malay Archipelago. Seas of East Africa, India and China. Fresh waters of India and Burma. Seas and estuaries of India and the Malay Archipelago. Fresh water of Tranquebar, India and Siam. Fresh waters of India. Fresh waters of India, also found within tidal influence. Fresh waters of India and Ceylon, also occurring in brackish waters. Cutch, Madras and Ceylon. Coasts of Africa and Arabia ; seas and estuaries of India, Malay Archipelago, Austra- lia, Japan and China. Natal, seas of India, Malay Archipelago, Australia and Formosa. Estuaries of Bengal and the sea of Penang. Seas and estuaries of India and the Malay Archipelago. Fresh waters and estuaries of India, Andamans, Siam, Malay Peninsula and Archi- pelago. India, Formosa, Korea, Japan and Kiangst in China. Seas, estuaries and fresh waters of India and the Malay Archipelago. Mekran Coast (Arabian Sea). 2 UU‚‚‚‚‚[3[1 SSS SS SS SS 766 Memoirs of the Indian Museum. [Vor. V, DISTRIBUTION IN Specific gravity LAKE. Species. eu oe Ha wines Further distribution. specimens were - taken. Main Outer area. | channel. TELEOSTEI—contd. | PERCESOCES | SCOMBRESOCIDAE. | Belone strongylura v. Hasselt . 1-0030—1-00750 | X Coasts and estuaries from | Bengal to China. Hemirhamphus limbatus Cuvier and 1-000—1-0110 | X Indian Ocean and the Sea of Valenciennes. | Penang. MUGILIDAE. Mugit cephalus Linnaeus : 1-000—1-0060 X Mediterranean Sea, West Coast of Africa, Red Sea, Indian Ocean, seas of China and Japan including estuaries, Pacific and Atlantic coasts of America. » gymnocephalus Swainson 2 2 ? Seas of India and the Malay | Archipelago. „ cunnesius Cuvier and Valen- 1:000—1-:0260 x X Abyssinia, Red Sea, seas of ciennes. India to the Malay Archi- pelago and beyond. , subviridis Cuvier and Valen- | 1-0020—1-0260 X Seas of India entering fresh ciennes. | waters: > caeruleo-maculatus Lacépède . 1:000 | |! Coasts of Mauritius, through | seas of India to the Malay Archipelago. » jerdont Day . 1:0110 X Seas of India. » Speigleri Bleeker 1-000—1-0080 x Seas of India, Malay Archi- pelago and Shanghai. Liza borneensis (Bleeker) ? Seas of India, Malay Archi- pelago and East Indies. > corsula (Hamilton Buchanan) . 1-000 X Fresh waters and estuaries of India. » troschelw (Bleeker) : | 2 X Seas of India and Indo-Aus- tralian Archipelago. POLYNEMIDAE. | Eleutheronema tetradactylum (Shaw) | 1-:000—1-0060 X | Seas of India, Indo- Australian Archipelago, North Austra- : lia and China. SPHYRAENIDAE. Sphyraena raghava, Sp. nov. % X OPHIOCEPHALIDAE. | Ophiocephalus punctatus Bloch 1-000 x | Fresh waters of India, Ceylon and Yunnan. PLECTOGNATHI SCLERODERMI. TRIACANTHIDAE. Triacanthus brevirostris Temm. and 1-000—1:0070 x x Seas of India, Malay Archi- Schl. pelago, Australia, China and Japan. GYMNODONTES. TETRODONTIDAE. Tetrodon fluviatilis (Hamilton 1:000—1-0110 X Seas and estuaries of India Buchanan). and Malay Archipelago. En oblongus (Bloch) ? Seas of India, Malay Archi- : pelago, China and Japan. ae patoca Hamilton Buchanan 1-0260 Seas of India to China. » rehñcularis (Bloch and 1-000—1-0260 see Seas of India, Malay Archi- Schneider). pelago and New Guinea. EE GE 1923.] Fauna of the Chilka Lake: Fish. 767 DISTRIBUTION IN Further distribution. Specific gravity LAKE. Ces + of water in which : = i specimens were | taken. Main Outer area. | channel. TELEOSTEi—contd. ACANTHOPTERYGII PERCIFORMES. LOBOTIDAE. Coius quadrifasciatus (Sevastianof) 1-000—1-00750 X | SERRANIDAE. CENTROPOMINAE. Lates calcarifer (Bloch) About X | x 1-0000—1-0260 CHANDINAE. Chanda ambassis (Lacépède) About 1-0150 - X Priopis gymnocephalus (Lacépède) 1:000—1-0260 X X LUTJANINAE. _ Lutjanus johnit (Bloch) . 1-0060 X Therapon jarbua (Forskël) 1:000—1-0260 x | | a puta Cuvier 1-000—1-0260 x x SILLAGINIDAE. Sillago sihama (Forskäl) > ® 1-000—1-260 X X Sciaena coitor (Hamilton Buchanan) 1:0030 Limbrina indica (Kuhl and Hasselt) 1:000—1-0110 | x GERRIDAE. Gerres öyena (Forskäl) 1:000 X » setifer (Hamilton Buchanan) . 1-0070—1-0260 x x „ punctatus Cuvier and Valen- 1:000 x ciennes. Leiognathus equulus (Forskäl) 6 1:000—1-0110 X X FA blochii (Cuvier and Valen- 1:00750—1:0260 X | x ciennes). | i Estuaries of the Ganges, rivers of Burma, Siam, the Malay Archipelago and Peninsula. Coasts and mouths of rivers of South Eastern Asia from India to Southern China, the Philippines and Austra- lia. East coast of Africa, shores of India, Malay Archipelago and North coast of Aus- tralia. Coasts of Orissa and Malabar, Seychelles and Sea of Penang. Coasts of Africa, Red Sea, seas of India, Malay Archi- pelago, coasts of China and Australia. East coast of Africa, Red Sea, seas and estuaries of India, Malay Archipelago, North coast of Australia and Japan. Red Sea, seas of India, Malay Archipelago, coasts of Aus- tralia, the Philippines and South Pacific Ocean. North and East Africa, Red Sea, seas of India, Malay Archipelago, coasts of Queensland, the Philip- pines, Japan and China. Seas of India, China and the Philippines. Seas of India, Malay Ar:hi- pelago, the Philippines, Sea of Penang and China. East coast of Africa, Red Sea, seas of India, Malay Archi- pelago, Fiji, Japan and the Philippines. Seas and coasts of India, Malay Archipelago and China. Red Sea, seas of India, Malay Archipelago, China and the Philippines. Ditto Seas of India. I 768 Memovrs of the Indian Museum. [Vor. V, Species. TESEOSTEI—conid. ACANTHOPTERYGII—contd. GERRIDAE—contd. Gazza minuta (Bloch) SCORPIDIDAE. Monodactylus argenteus (Linneaus) GOBIIFORMES GOBIIDAE. GOBIINAE. Gobius ostreicola Chaudhuri 5 albopunctatus Cuvier and Valenciennes. Glossogobius giuris (Hamilton Bucha- nan). 5 biocellatus (Cuvier and Valenciennes). 55 mas, SP. NOV. 6 Ctenogobius acutipinnis (Cuvier and Valenciennes). 35 chilkensis (Jenkins) ba alcockı (Annandale) 55 globiceps, sp. nov. . 5 cylindriceps, sp. nov. a dentifer, sp. nov. 5 minima, SP. nov. . 5 Oxyurichthys tentacularis (Cuvier and Valenciennes). Apocryptes rictuosus (Cuvier and Valenciennes). 5 lanceolatus (Bloch and Schneider). Micrapocryptes fragilis, gen. et sp. nov. ELEOTRINAE. Bleotris cavifrons Blyth . : : » fusca (Bloch and Schneider) . USD) Ce ; : 5 Butis butis (Hamilton Buchanan) PERIOPHTHALMINAE. Periophthalmus koelreuteri (Pallas) GOBIOIDINAE. Taenioides chilkensis, sp. nov. ZEORHOMBI BOTHIDAE. Pseudorhombus arsius (Hamilton Buchanan). Specific gravity of water in which specimens were taken. DISTRIBUTION IN LAKE. Main area. Outer channel. 1:0020 1-000 1-000 1-000—1-0260 1-0010—1-0060 1-000—1-0080 | 1-000—1-0080 | 1:0075—1-028250 | 1-000—1-028250 | 1-000—1-028250 1-000—1-0110 1-000—1-0260 1-000—1-0260 | | 1-0020—1-0260 | ? 1-0020—1-0080 À 1-000—1-0260 1:0280 ? 10280 1-000—1-0280 1-000 MK KOK KR KK KOK xx X XXE KK OKO: xx Further distribution. Zanzibar, Red Sea, East Indian Seas, Malay Archipelago, New Hebrides and the Philippines. East coast of Africa, Red Sea, seas of India, Malay Archi- pelago and of Australia, Philippines and China. Seas of India, Australia and Fiji. East coast of Africa, seas and fresh waters of India, the Malay Archipelago, Aus- tralia and beyond. Seas and coasts of India, the Malay Archipelago and China. Seas of India. Brackish water near Calcutta. Seas of India and the Malay Archipelago. Seas and estuaries of India. Seas of India and the Malay Archipelago. Andamans. East coast of Africa, coasts of India and the Malay Archi- pelago. Seas and estuaries of India and the Malay Archipelago. Seas and estuaries of India. East coast of Africa, seas of India to Australia and China. : Tr TS 1923.] Fauna of the Chilka Lake: Fish. 769 DISTRIBUTION IN Further distribution. Specific gravity LAKE. Shares of water in which P ; specimens were taken. Main Outer area. channel. TELEOSTEI—concld. ACANTHOPTERYGII—concld. | SOLEIDAE. | Synaptura orientalis (Bloch and ? x | Schneider). CYNOGLOSSIDAE. | Cynoglossus brevis Günther 1-000—1-0280 X | X SCOMBRIFORMES. | CARANGIDAE. | Caranx carangus (Bloch) . 1:000—1-0110 X | X Equula edentula (Bloch) . 1-007750—1:0110 x X | JUGULARES. BLENNIIDAE. Petroscirtes bhattacharyae Chaudhuri 1-000--1-0110 x ER SCLEROPAREI, PLATYCEPHALIDAE. Platycephalus insidiator (Forskal) 1-02825 X OPISTHOMI | MASTACEMBELIDAE, | Mastacembelus armatus (Lacépède) 1:000 X Seas of India and China, R. Hughli at Calcutta and along Orissa Coast. Seas of India and the Malay Archipelago to Atlantic Coast of Tropical America, Red Sea, seas of India to the Malay Archipelago and beyond. East Coast of Africa, Red Sea, seas of India to the Malay Archipelago and beyond. Fresh and brackish waters of India, Ceylon and China. : 9 tial MORE | 1 A el y RE NE PCR AU À | | FUN ui the Psi | a ; = tie i f Lu EN 1 j x or | Ane Durs" ; . Hi KT À ts j ny N h rss AUDE ARE Le Rire NW Se . a - \ L L at SE Woes À 2 abe MEA Te To à f | ESS UNE WO oO : Ka \ { ur te r AUOT babe DA Damart r . 7 d } j ® i By à Bs. ; x M LOS A tat lue ) Satan Ep hy B ] NA beads IK ARTE » jo } N rte Wu piace at a ‘sere ss Gh A tn tae ee amend oi NOTE NEI ree Ue: N Pi K 2 a ‘ = a i CALE L ÿ Br Et { * . x a : À * d 2 + ll SE ; À A \ sr | > L'ile ‘ rel . 4 Î ub 3 à $ r . NE a " ax ; N ‘ a shine wi er" ® ur t + hae Je P = An AAPAPAOAARE —NIIIN saa grrr ren RAAR A ER een aan N AASAMANAMA SS N SARRRRARRAN a a setannsasasse NE aman nannee npn PPS an Bann 5 met AR AN ES aa RT 222 22 222222 AACS Manz ARAA RAR AARARARAARRARASSSSS DAARRARAR RAR PAP ON PPS PPS pa RAA "= SEEN bag. AAA AR ARR RAT MASSSBE RR ARAARRAA RSS 2 N EN N gun PR 2e BE nennen M 2} Ras: AAA anna nnnnn nenn RR N 5 A À À A AA AA PA Pi ENST = SORT FF == SAMT. perenne 7 ZA A A A A A A AP à FA f A nannananannaan ER à AA SAS IN? 2222232232 AR 2e === BARAT i NAAR ann RAR Mii oa azaalalalelalelel AARAR AR aa aaa AA aan ER R PA AR pee AAA AR RAR Am PP AAR pA ARRARRARART ee er RAR BETZ m, RR hee 222203 22D3) YAAARF panaaes SAAR AAR AD AFFrT, ARE Re meoerenere ty an PRAAR 3 = an ao oe FRI à EEN F N N SAARRFRARRREFFZ AR S\N NN ARR AAAA YAY Dee 2 z SARA AR A An APP RARE RER Re ANS FACE fr’ FT Lee 2 PEN anna | In aaa mm mn in on A CE Co RR PA RAAAAAAARRR RARES Alain AT if i V D a! on f MAnaaaamaaana III BER Bes ge ART RARE BRRFEBREN en Van Rallye ar TARA nn anne ER anaes AR M SSS SS R || | || TT AAA ARR RAR RS aaa FAR éeracaRas" AAA Sopananannananan N AA AP le en nn A AS gaanannc - NN = NAS S ~~ AAAI A | Pa ER CRE == na nana ARR an a AAA a I || lm len, es A ne - Yan Nnnamaaacz RARAARAGÉ ERA y IN AA AAA PAP Yen I AAA SAA AAA SaAARAaT Anz aan wer | PA PR PA Pa an Pa PP war RRR RRR CA STASASS NAS SAR RFAATY CN PARARARRARRR AARAAA ))) 33 3 =v \ARAR HY pees 5345 RE mere nn nn REREE EEE Ju PA PA PA PA jan PA u jan PA PA jan PA PA PA PR A 3 === RSA PARARRERREE AA SRRRARARRERRARRRRRRRRFFFERTTER am. Re DRBRA Anna nn - 722222222222 RER NRA ARR RE AAR An | AAAAAAR aaa, 7-77 à FFF AREA R AAA SAS VAI am ASIAN, a, yo mn | rer ee m AP RARE RAR EREREE EYE an Se N AA N A BAR, a AAR EN 0 SRARERERAR VA, ARR RAR? ARR = 722 — esse AAA FN PA A A fa A PA , ARE E EIRE RER ARRETE Asa VON A AAA, sal J AAA RAA AA "AP AA mn i ASS AMAR lsialela la als Nam \ FR FR ur nr PAPE LU a Æ AAAA DRE am = SITES SA AAA AAA À À AA Pa A PA AAR SEE SQ = RAA AAR BARRA ARR RRs See = POR aR SG error yyy) an FA PR PR 26 : | £ & FE 0 A EE Qe ee a Ss Si Ee DER En que a PP AA AL 01261 7718