: eee OE Km ntne awe Se ale bE Wea hot hime) pany reas rc ta aL thy te er: a4) ‘yoyehs Soca yatta, » mane seth whe Syne tee Vata a PRE. era te Npplaer, PPE gy, HAL AeA Sy same tebe pres ZF bids be | ELE FD pa Dory PS Nee gs Pat pepe ASP Pr Zh) ae pub pep boba a ppb Tih pede at ve Fir Repo tye ld eur Maou pute TRL i AE Aye iy FL tips a3 fe BP AT ad Pa payee a ld saudi wer aee Spe bbe wee dee awe ge ape eeheoe pi! Sp eA TERED Tae ep PEE iene ewe Satay fie 42 feeds apnabia WLP ibe ageg Cate pre Tamey end Fite ff ahi o Hage ag Pay sfs PO A LO ee a eV pases E ’ eu red wien A A pe in OAM ed HA ite ad et ana Se Pen Westy argh p Bar greby ie tig she po: oe 7 FO Wer Benin gh 7 at ee PEW ANE Cd i AEE cag one oe etd pera ee Bosnich Cpa psaepyce: Parra ceney Tanpetny iy SES aaa ee ee AE MTN ES Jaca phi ag ge Maw ge bey 98d ere Ob yi aarrew Pree L Ce wrer a ea) or er edad ay ee Se) OC rg fated ip aes y ae Foeyen Pru. Ee fae e aie ae Ors Ue SG Wb bia w SAD pate ene APP aed NET Fp age Pit esue ig FT Poa 5 wien cust gage? , (abe ewe Tp AD SE wre ae MLTR EE gle 8 Bri ecp fi ot AER rises oe Pm Un Hangs UNOS ERY RAE EI HOLT Gegrasingey ot wie ue Fe OND WT Gy re ca aie rae ae aes rererrerers PY PE ae ab Sees , uge fe Ancee er ge Peete REPO Lp ey + A OTA TG a Caoeees pale ete har rab ee 6 AD eb eA SEPP RSG : CRY Be fae ? re eee Peas Tse jane pnatee i we Ss Wee SANE WG 9 Le gongs «1h gd Soe) MCA 8a ts SGN Ge gry cag va: 6 iiss} BAR aa si * ebabede “! 4 i LL It ht) P aC ae Me ere Oe ay Be aN OE iene eee ser a ‘e ar etutary pea ory” (be MO Nee ee a paratgigte ed a Pew hers nes “a ea te eee ee) ag ah VEN Gie Wy fp ! VG ian gins ee heey Pe Bs. ewe pies waren POE ET Fe Loe CE a AE de eyon we, Fh 2 5 ans es SRAM foie fu r 2 POE Hee aT Vw secre ve ey eas Fey pas: SONOS Er Fn Cae iey Vipin Bak poste wy, “ Wika ehings Y Figs Ves feats We ates te ian ee ara SEU EAT eps mh Gee See SRE, a ease * — w — me?) — = a = 2 EY 2 E 0 = ) = > Py > : S = Zz = - = a = = - = = = — a ” ” rm ae i Z 5 Z i We | :S /,SMITHSONIAN ip NOLALILSNI ys tldVeaIT_LIBRARIES) SMITHSONIAN INSTITUTIOI = < = < = AGES ee = z < ~ = = z Spe Ay PN > < ; : 3 S 5GM IRS | NS 2) WY Uys eX ® Q : : : : sy EN 3 : N SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS z aS m2 (e) A ae = = 7) 2 & iu 2 td 2 tu z ” A 22) as 4 (op) EA 4 AS =t =a = me g = eS . a c o Cc ro} = 3 =) ae 3 = = “oe Zz XN = Fe =y} ae — ma xX a =S_ SMITHSONIAN_INSTITUTION_ NOILONLILSNI_NVINOSHLINS S3!1YVYHYSIT LIBRARIES SMITHSONIAN = 5 & z ee yis = im z a = fo) : — ~ — = yy oO a wo » — — w 5 ¥Y,3 5 2 Ne 5 2 5 > = Wu > = 22h ING ES , OE > = > - YY fe Za = ce AA - 2 a pc) Z Bilal: Z oe 5 2 : | NI NVINOSHLINS S3IYVYGIT_ LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI_NVINOSHLINS Sa1uvuaIt z a = 2 Zi 27) Zz a 5 < = < = < = : < = z = Zz = z = z = Se e) x= . Oo ae AK FAY OO sie fe) = z = = BP Sw ae = = = = 3 eS eee : S -S SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHLINS S31YVUYGIT LIBRARIES SMITHSONIAN INSTITUTION = n = Ores a 7) = a \ a ss a = a a z = ‘| <x 22 <x = te INS: = = < oS fs = = 4 ANN =f aie = = = = (oe) ca S = ro) Sarit -_— a S a 2 i. =3 2 =r 2 NI S31uYvVu¥d!i1 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31yuVugIq - 5 = S = S Sy AP ee z w = 2 Ar evs = 2 We = \ 2 5 2 2 _%YYy ze > 2D AVR = > * = > = ¥y Z > cS > = ee = is - “GY 2 = a = wn 58 ES (27) = wo = o Ue 2 S|, SMITHSONIAN _ INSTITUTION, | ELON EEN gr a eVEEIILLIBRARIES SMITHSONIAN INSTITUTION = Z « Z <= =e a Ne Ks = = ek. = 4 fA; Sip AWN = . = 5 + 4 S = wy, fe) NY 2a 5 : 2 2 if Mf 2 8GH 2N®& 2 2 NN 2 = Zit “if E Zz Gy = Y 2 = 2 = > GQ = > : = “tye S 2 7) Me oF 7) Z ” ee a a NI NVINOSHLINS S3IYNYHYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI S$ S3I1y¥vudit Fa ex 2 no n RS n” om ze V3 sss = = = 5) FSG N a z e ff (= Cc ew = =A sl = = fe) ) a a 4 x Zz = Bat x =S_ SMITHSONIAN _INSTITUTION NOILNLILSNI S31uYvVuadiT LIBRARIES SMITHSONIAN _INSTITUTION SaIYVNGIT LIBRARIES INSTITUTION NOILNLILSNI INSTITUTION NOILALILSNI ern rian we Ss Ny saiuvuaia S3iyvudl INSTITUTION Ni NVINOSHLINS S3I1YVYa!IT_ LIBRARIES ~~ t SMITHSONIAN INSTITUTION NOILOALILSNI NVINOSHLINS S3IYVYdIT NVINOSHLINS S3Z1YVYEIILIBRARIE SSS SMITHSONIAN NVINOSHLIWS NVINOSHLIWS ie { S43!YVYdIT LIBRARIES SMITHSONIAN INSTITUTION SMITHSONIAN =S SMITHSONIAN INSTITUTION NOILOLILSNI_ NVINOSHLIWS ES SMITHSONIAN ES 7d NVINOSHLIWS \. N AV x Wy SMITHSONIAN ES SN ESS SNI SNI ES SN! a mo.) — 1o9) = VE ee CV — ; yf yy = WS ‘3 = = - ¥YZ Uf es) 1“ | od wm { URS 8 a) (te yy aad 0] = > \ FE a eG, = ‘ = D m 2 ai ILNLILSNI_ NVINOSHLIWS $3 He 2 o = ” SMITHSONIAN — ie e = = SALMVU GIT LIBRARIES Ss z z AS ay NLILSNI 7) = NS = = = th Js 5 5 Yi e Sg rs SE 5 Mf > SY > 1 uf fy 5 NY = D S YY = SCY S pz = YW 5 ‘\ wD Z SYR, = Se i, a 2) thin = N = = : @ s = yy = 2 “ip = ‘ 2 ob a “LIBRARIES. SMITHSONIAN ~ INST! YY = > ; 7) z INOSHLINS S31YVudIT_ wn z 7 a AN“ INSTITUTION NOILNLILSNI_ NVIN Z g “3 2ARIES SMITHSONI = o Zz uw g ce. = n aad = < 3, Ee we we 4, ae w = 2 = o eal he e - : S Se ae « ““Yyy m2 5 ae 2 INOSHLIWS = C4 = = ARI ES SMITHSONIAN” INSTITUTION = NOLLALILSNI NI z yqgia Li ” fa re) wo = LILSNI” NVINOSHLINS Salay : z ; . 8 a E 5 $3 (fe) o E ae) =) > fa o = 0 > i > = a o Pe) = > E& “2 ie m Zz > Loa a at es ae _ , 2 i = 2 a = RARIES. SMITHSONIAN INSTI m Zz D EE AMINDSHLUNS Sees calc g z ‘* 7) ZINSTITUTION NOILAUILSN ws g yy < z aN n Bos = YY, = = Wy, = x By = . + GY; yy eS = * Ye H ie ae mS = = Y 4 vo tH fp. es, Oo : = z SY \\, ba od by; 92 n Y ia) oS Z E aN o (@) NYS arT,) VYys 2 (e) Jo 7 = z = : w 2) Sy SS \N .e) L, GREG = =) = 2 7p) IY eG ASR WS 2,7 7 = > a SNI_NVINOSHLIWS SJ Zz = IW > ov ONIAN INSTITUTION NOILN LIL ie: ey ‘ = = Z ye = INS Saluvualq LIBRARIES SMITHS = “ é = ILILSNI_NVINOSHLIWS S us a = = = wn = ra Ss y at = fea a ae + < Thy hy = = 5 \ ese E & = aay S) 3 S SONIAN” INSTI 4 ra) ro) - Zz [7_LIBRARIES SMITH r ie =I NOLLNLILSNIT NVINOSHLINS *Salayug Be S = NIAN INSTITUTION NOIL ¥ z = El eye ARIES SMITHSONIAN _IN' z is 5 B ey Diy x Se = od - EY Uy eo) a). ae) 2D 5 = = Wi 2 = n 5 > | = 2 5 Gig m =i > [= es) E m z n l= 3 bo) = =_ w an S31u' ie = ee m . N= 3 eae eae = ONIA ” x = ‘ = ea luvugia LIBRARIES SMITHS z < ‘Ss = Ge \LILSNI n = NE a= = a re) wo he z SE aie NS = S a = 2 Y LS 1 wh G FD Ke S o ee} EE Sy eS Up? 8 Gif IRR : : : z G se S “iy = XY > a 2 = LIBRARIES SMITHSONIAN _INSTIT Vy = 5 ; = 2 INS S31uvegIT_ a = a * INSTITUTION NOILMIILSNI_NVINOSHLIWS - 2 iM “eS ARIES SMITHSONIAN _ INST a > Ks td 4 Oc 4 a Ea Boa oc = SNS Ne = = r ne ee a 4 WN = rea O ‘ Pip SI s S SQ Ss 5 yi ile Z = Zs 5 co i) e i) ~ NOILALILSNINVINOSHLINS ~ $3 eed a lyf fe) ay 2 IAN~ INSTITUTION , Ee S) a SS31NVHG11_ LIBRARIES” SMITHSON a Q 2 WN ets LILSNI z i ro) WZ w = ee XS = o Ss 2 > ye fi > = = Wy ” ) = > i Ui He oa o m ~S = 5 = z o* ne o Sinema SMITHSONIAN _INSTIT os op) m z yuugi : ” Seon m z w INOSHLINS S31Y Zo = SS As 27) ZINSTITUTION NOILALILSNI_ NV @ 2 BS = z \ n Ca = - = = ro) wo & RN = = i Gy 2 a 72 S E OS \ Zz 2 WN ¢z Vien 2 = = S a \ D =z Ad 3 Ws e > Do * NVINOSHLINS S31uy = Sa tule ES SMITHSONIAN _INST near = A ~ > Saruvyai LIBRARI ys ZB are ST HLISNI_NVINOSHLINS Ze Ce Vi eter eek | ee ITU wn fo SU > nt THE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 26 July 1, 1983 to April 2, 1984 TABLE of CONTENTS Number 1 (1 July 1983) Our thanks to Rudolf Stohler. R. I. SMITH New molluscan species (Gastropoda: Neogastropoda) from the tropical eastern Pacific. LEROY H. POORMAN The biology of the northeastern Pacific Turridae. III. The habitat and diet of Kurtziella plumbea (Hinds, 1843). R. L. SHIMEK 10 Flow through and around the abalone Haliotis kamtschat- kana. ASNIGEBVIOUMZOW! Gigah cee oo as ke neil ee es 18 Shell strength in Corbicula sp. (Bivalvia: Corbiculidae) from the Potomac River, Maryland. VicTOR S. KENNEDY AND JAY A. BLUNDON .... 22 Relationship between beak morphometrics and live wet weight of the giant Pacific octopus, Octopus dofleini martini (Wilker). SHAWN M. C. ROBINSON AND E. BRIAN HARTWICK . . v6. a0. aa Os ed faked op ad ce Sra ne ee 26 Studies on the reproductive biology of some prosobranchs from the coast of Pakistan bordering the northern Arabian Sea. II. Egg capsules and larvae of four species of Thais. SOHAIL BARKATI AND MUZAMMIL AHMED 30 Comparison of northern and southern populations of Ep- itonium tinctum (Carpenter, 1864) on the California coast. CaREY RESCH SMITH AND AMY BREYER 37 The reproductive cycle of the trochid gastropod Oxystele variegata (Anton, 1839). M. A. P. JoskA AND G. M. BRANCH.......... 47 The larval biology of Brachidontes modiolus (Linné, 1767) (Bivalvia: Mytilidae). ANGELA FIELDS AND EUNA MOoRE 52 A comparison of the species richness and trophic roles of gastropods and chitons on rocky shores of temperate and tropical west America. ENILNN (Gi, IMOTUUTY slog ao soe ooudvianoay ooo gesn 62 Range extensions of three opisthobranch mollusks to the San Diego—La Jolla (California) Ecological Re- serve. Hans BERTSCH AND TOM SMITH............. Number 2 (3 October 1983) Radular patterns, systematics, diversity, and ecology of deep-sea limpets. CAR OUEBS MENCKMANE ane ceisc lair s skeen 73 Additions to the opisthobranch mollusk fauna of Marin County, California, with notes on the biology of cer- tain species. WILLIAM B. JAECKLE 93 Specialized feeding in mitrid gastropods: evidence from a temperate species, Mitra idae Melvill. ALLAN FUKUYAMA AND JAMES NYBAKKEN...... 96 Homing in the West Indian chiton Acanthopleura granu- lata Gmelin, 1791. DAVID ENT OOK eee ecieib ye Actin ssh nd alee eens 101 Aspects of reproduction in some enoploteuthid squids from Hawaiian waters. LIGA IME, JE\OW CSI vers bao os om Oe gee ome oe 106 Shell repair frequencies of two intertidal gastropods from northern California: microhabitat differences. A ONATHANEBS GBLELERMM a ae ane aor lS Morphological divergence and predator-induced shell re- il pair in Alia carinata (Gastropoda: Prosobranchia). Jutta BERGMAN, JONATHAN B. GELLER, AND VICTOR CHOWMAAE fasts ee eee EAE Nes cements 116 Thermal effects on some mangrove mollusks. Honc Woo KHOoo AND ELEANOR CHIN ....... 119 Barnacle attachment on live Nautilus: implications for Nautilus growth rate. NeIL H. LANDMAN 124 Distribution and abundance of Caecum cornucopiae (Gas- tropoda: Prosobranchia) on Cladophora crystallina mats in a Bahamian salt water lake. BESTS YD) Aas ape erates Mak pre Soterek ee, fy seal kee geet 128 The Bermudan and Caribbean sacoglossan mollusk Elysia flava Verrill now recorded from the Greek Aegean Sea. PIE H OMPSON Per pea ceva oocytes ae 136 Do oystercatchers influence limpet shell shape? P. A. R. Hockey aNnD G. M. BRANCH ........ 139 A range extension of Nassarius miser (Dall, 1908). IROBERM Iss EIOWLEY arse as) terior Shee oer 142 Number 3 (3 January 1984) Activity, dispersion, and size of Lanistes nyassanus and L. solidus (Gastropoda, Ampullariidae) over the depth gradient at Cape Maclear, Lake Malawi, Africa. S. M. Loupba, K. R. McKaye, T. D. KOCHER, AND C. J SSTACKHOUSE snot ee eee 145 The Recent Crassatellinae of the eastern Pacific, with some notes on Crassinella. EUGENE COAN 153 Description of five new species of Hawaiian Eulimidae. ANDERS WAREN, BEATRICE L. BURCH, AND THOMAS A. BURCH 170 A new species of Ischnochiton (Mollusca: Polyplacophora) in the Gulf of California. ANNINONNO) JJo JONNY, 6 occ 0cso0ccv0ss00as oc 179 Records of Cuthona pustulata (Alder & Hancock, 1854) from the Canadian Pacific. TERRENCE M. GOSLINER AND SANDRA V. MILLEN... Ee Ae eee eee ey ernest PY chats, Sh ga 183 The ecology of Parvilucina tenuisculpta (Carpenter, 1864) (Bivalvia: Lucinidae) on the southern California bor- derland. GILBERT F. JONES AND BRUCE E. THOMPSON... 188 Hybridization of two populations of a marine opistho- branch with different developmental patterns. HiLitary H. WEstT, JUNE F. HARRIGAN, AND SIDNEY K.. PIERCE. ..< 250. 409505005500 ee 199 The opisthobranch mollusks of Humboldt County, Cali- fornia. WILLIAM BS JJAECKILES 35 a0 6 ooo eee 207 Doridacean nudibranchs from Sri Lanka, with descrip- tions of four new species. NATHALIE YONOW 214 Technique for narcotizing and fixing veliger larvae of Am- phibola crenata. PENNY STIRLING, COLIN LITTLE, MARGARET C. PIL- Number 4 (2 April 1984) Supplementary information on the morphology of Phes- tilla melanobranchia Bergh, 1874, from Seto, Ki, Middle Japan (Nudibranchia: Aeolidacea: Tergipe- didae). ISIKUMARO BAB AGS a5 o/c cae Oise ey ais ee era ee 241 Distributional records for terrestrial and freshwater Mol- lusca of the Cascade and Coast ranges, Oregon. BRANLEY ALLAN BRANSON AND ROGERS MACGOWAN BRANSONEY a Potton ye eae eee ene ee 248 The biology of the northeastern Pacific Turridae. IV. Shell morphology and sexual dimorphism in Aforia circi- nata (Dall, 1873). IROINIALID) Ik, SIBUIMIDK occ ounscc vende ceuccsve ce 258 Distribution and radular morphology of various nudi- branchs (Gastropoda: Opisthobranchia) from the Gulf of California, Mexico. Hans BERTSCH AND ALEX KERSTITCH ........ 264 The diets of Alaskan Neptunea. IRONALEDEIL SHIMEK@s ay Athy score ee eee yak Spatial distribution of three species of bivalves on an, in- tertidal flat: the interaction of life-history strategy with predation and disturbance. Mark L. BOTTON 282 Orientational and anatomical trends related to detorsion among prosobranch gastropods. 1V KINGTON, AND JOHN B. PILKINGTON....... 229 New species of northeast Pacific archaeogastropods. JAMESTEIRI MICIZRAN Ia Aen eee eee 233 Louis F. GAINEY, JR. AND CHARLES R. STASEK ..... uygiey sao ee a ee ae ee 288 Reproductive cycle of Anomia simplex (Pelecypoda: Ano- miidae) from Cape Cod, Massachusetts. DD IANED py BROUSSEAU Se eee 299 Influence of varying oxygen tension on the oxygen con- sumption of the freshwater mussel Lamellidens mar- ginalis (Lamarck) and its relation to body size. V. Mapan Mouan Das aNnp S. A. 'T. VENKATACHARI 305 Gaimardia bahamondei, spec. nov., from central Chile (Mollusca: Bivalvia: Cyamiidae: Gaimardiinae). CECILIA OsoRIO R. AND PaTRICK M. ARNAUD .. 311 Egg masses and larvae of three species of Cerithium from the Arabian Sea. SOHAIL BARKATI AND MUZAMMIL AHMED 316 Male reproductive system of Chorus giganteus (Lesson, 1829) (Muricidae: Prosobranchia): anatomical and histological description. MartTaA AMin V., IRENE LEPEZ G., OSCAR Marin S., ASNID) IMUASTA, IDISIUAIN| Na op oc op uve as eeaace 320 Anesthetic methods for the moon snail Polinices lewrsit. GEORGE B. BOURNE 327 An earlier name for Nassarius corpulentus (C. B. Adams, 1852). RICHARD E. PETIT AUTHOR INDEX TIMID DNAS eerie ocr os Pe oo Aa eRe 30, 316 ENINSTONT,, Wo HINA Li ie Viet clea: eh Sn Bec a 320 AARRINUNTUID),: 12, JNA latounir te ene ete etna ota euoica a penen (cess mene 311 IBVAIBVA, UEC V2 Gralla et Rete: 6 SISGUrt non ee dice ree anes tae ar sere wueme 241 TBYAIRIRATI Ls Sig aks See Gace Been cee uence Gkencnc trie atone 30, 316 {BUST GINAUAINYS »| feaulnana eate-cnel es come tre ook ve ais yee teeanee nae 116 IBERINSCHA@EAR te me cass Ga whe eile taunts ocr 69, 264 BEWUNDONNG EY ACur Minty Sr arhees rn seein dd Sts 22 TBXOVITUROINT: NA La ya! oa ed a ees ee se eo 282 BOW CHER MMO M VIR Kin natin eater ein Wan Ae, he ae 106 BOURNE GB ges uaricih te ccttsurset tats ertla arava as 327 IBRANGHRE GINS, ey aiaccsn hoists ah os i Gaele dt 47, 139 BRANSON Dy ACH wa res ert 1a Aatiny a Js coma canes 248 IRAN © Neg Rem Lenraitre efits Sie ic telecine ele wes alia a 248 [BURTOWZTAIR, ANH al cece al RU ere Sere Be oe 37 BROUSSHAUS DEA Parents, welathacsiee a cuee sie sis en 3 299 WRC HBB olen tes ene eA eee 5 ole ee ues dodrsee sh 170 BWR CAME Ata Sle eGo auieie ts betes aon ana sa.s 170 (CRETE re ey tr oy OUR E A Gone UM eee Sena 119 (CHBIONY, Ws. a2: Bead eves rele Cone ee en ear 116 GOANRIE ee ee BA eae eee HS ti were oc 153, (232) IDAS, Wo IMIGUNG ei coereed aronetero nao tose chers cieeenet een 305 ID VAST, 1B)s: -o:ni bs abekor Ba siceeelee Seo E karen ceca enya ete tr aetna ean 128 1D) IARING ACMI ane enn inie sepcrtuneiaeneene hese atime oa os 320 HE RRETRA Nees mle Wart auer cnet aici Sei tis marek ei ss 179 IESIEVIATS) SMe Nagata ewes erent ere A ans 2 lieve tele, thio es os 52 TEAUIKSUWAN TA Nets str hee H Sety ie coe hese seg laneadg a 96 (GlATINTBN Ys, Las. 1 Be | RS" Soe) cepscltts ata Sache ciy oes tect le pane mre 288 (GIBILILIBIN | [o-IB3s ace chetonee eiguses ee encioe ae ecercaetaar™ 113, 116 (GOSTEINE Realy VIR es oe hu eons Hb koa lara eee 183 TRUATRIRIIGYAINT, - [ic] Se oe toa as basics ot aie Serer eee Sea ra 199 VARGA WiC Keen eta sleet Pyne eeu nese eset a ge ees eres = a 26 TEUIGRIMUNING: (Canty eck Sig co basics sare ok to Sen ee a ee W® VEU OGT RING) A) Re 139 LO WBE Ag Ee ed losin Sanu en cv terge CAN Sc ee hs 142 ASE CPE SAV VES Suir seer eictia ea che eectien: 93, 207 JOINTS, Gis TESTES axe choo Gola a eRe meee ene ae ne 188 A OSIKACEIN TEE Ace 2 sapere neta ara eases Sah ls Cl vate a8 47 TRANOOTESTS 7-91 De, lg eN is a Neer a ee ee (144) ISBNNEDYAIVES ont tie e ene is Bn deen Aa aa 22 IGERSMI CH SA oan came ase eaa Roetmionce ahs 2 264 IRHT OO SELES VN oie an Me een EMRE SE ie, ou thc et ied 199 IS @ CEE Route SO) 0 ears ell eee i aie gear Say ey By als 145 TEIN NUAIN Nea alison pera aitor creer ir Puree teor one ba cies 124 P74 G5. he en soe RS eres We eR vn Ho see 320 ISINDBERG wD RAGE aan ecauarir ity ses croes (334) TE TATeIGTA By OM ee ar ei oie Meee tee ees Ome eyes 229 TEOWD ALS NIG Cee ee ais wep oe eemuage eee MEMES ta ci ncee etn 145 INIVARITING St ©) sete cle ea as Rah a ge are ag ane a 320 IVI CIGARS RGe gio eee Soi rare ree a Nee ys 145 IMICIERAINGS Je bleu aiy) amen mec ten eon pasar casa nea 233 INAITTEICE: NERS 5 Weare cc miteee ts, maa tied cece eS ECs tare ePaee. 6 183 IMITTETIGE RE ACR Gin tee ae eins ore sash vee enare ns Ces Poe nto wrens 62 INALGYG) S155) B Yes een Ry ear eta Ran dR tl nie, th ntoN nl aac eH eter 101 INTOORE Bes ele Cer se entrar Np eReni gts, Ry we Weta beaters 52 INVIBAKIKENS Ji tas rosenusyoibae e mascara cae cla ete e 96 OSORIOBR GS Cie ee aula ere ye ase ac eae Sen 311 ] PDF UURIU a] Aan pray RENT ntetcer sem ok Le INST ae ale ere rete ned RO 330 | ES OCHO) aise De Sarna is ae rae aed Baca a eer (143) JEN TIO) Os) hor ena aa LOO VD me eu PEE Moa 199 RIEKING TONGS BS: ciate cee ath ube eiaasieee ela a 229 JPAGLISTINIGARON IML, Choco coeeannadcedsaccnscovuce 229 IROORINGAN Se ele Ava ee eames te pert are veers evar 5 INOBINSO Nags tales Cease ane nea ene 26 ROTHER B: Bie Scars Mieniea aac hee ates (143), (332) ISHIIMIR. Ke RES SIGS ioe tripe se eget Sista oaet nareas 10, 258, 274 IS MOTIDET a GROG ais oe a Oe el -mn RC EL ppe Nagra ee 37 S MINISEIERIRGOD oe toh, rere oe eee mer erate repent ee Ace 2, SMITE abl suetonttc penny eae aa ie SET Ae ae ee ine Scare el 69 STACKHOUSEMC A Wis rarest ean res ego ites ee Aevent aes 145 SIRASEK ti Ge Reais a elenyay Ye aiden Seu nnu mn myteN circ cum ee 288 SRIRIEING SiR aaretye eee Ln sien Syn anaiie co Lrer ene 1a 229 AHONMPSON a Bina a ees, este Wien ae irae 188 ATISCOWM IASON Men Bs sve, ga, Seto a peemae lore bouts oeeea neuer amee 136 WIDINTYNIPNGTEIN, Ss A Ws coctonscocsauosucosue 305 VOTEDZ OWI To senate cuehoke ante wa gamete ta dececa 18 VVEANRIE: NA (as noes ennara ihe Be depot err Maerua At, 170 DV VSESSoie ole] a El oar erras secant beurre mn crane SUN esc dmc ae 199 VION OWAN Nik eee re entender tie pert ns Sinuta ke 214 Page numbers for book reviews are indicated by parentheses. GW, pie ISSN 0042-3211 SYSNTION 4O NOISIAIC THE, = wasvesi toticcs % (IV H WITHA VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 26 July 1, 1983 Number 1 CONTENTS Our thanks to Rudolf Stohler Pap Ie OVNI Mee weep ae nee OSI AMM ta ice ie aout fel Bt adele a als 2 New molluscan species (Gastropoda: Neogastropoda) from the tropical eastern Pacific RO Napisy e OORINVAING eyes tet cae ore Yap cf wean toy al hen es, sy aa ev nie aha IS Imei euege Ye olle 5 The biology of the northeastern Pacific Turridae. III. The habitat and diet of Kurtziella plumbea (Hinds, 1843) [Re Lia, STEMI TSC FG tha louse seat teades Aas remain at te ADs RRO Pete na ok Raat e non 10 Flow through and around the abalone Haliotis hamtschatkana SAINI EMV OIMGZOW paar les icie re yh he aoee aired ese ay ee cis SUTIN, A UR ee at Be 18 Shell strength in Corbicula sp. (Bivalvia: Corbiculidae) from the Potomac River, Maryland WiICTORE SIKENNEDYSAND EJAY. Ac IBISUNDON) 2 qe) fst cele) ah neater 22 Relationship between beak morphometrics and live wet weight of the giant Pacific octopus, Octopus dofleini martini (Wilker) SHAWN M. C. ROBINSON AND E. BRIAN HARTWICK ................... 26 Studies on the reproductive biology of some prosobranchs from the coast of Pakistan bordering the northern Arabian Sea. II. Egg capsules and larvae of four species of Thais SOEATION DS AIRIKCASInIe ANID) BV MUHA MIN OTIS NEUE) es) oe SARE ee eee 30 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of July, October, January and April for $18.75 for existing members (plus mailing charges) and $37.50 for libraries and nonmembers (plus mailing charges). Mailing charges for all domestic addresses are $3.25 and $6.00 for all Canada, Mexico, and foreign addresses. Further membership and subscription information appears on the inside cover. The Veliger is published by the California Malacozoological Society, Inc., % Department of Zoology, University of California, Berkeley, CA 94720. Second Class postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to C.M.S., Inc., P.O. Box 9977, Berkeley, CA 94709. THE VELIGER Scope of the journal The Veliger is open to original papers pertaining to any problem concerned with mol- lusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS”; in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Donald P. Abbott, Emeritus, Hopkins Marine Station of Stanford University Hans Bertsch, Universidad Autonoma de Baja California James T. Carlton, Williams College—Mystic Seaport J. Wyatt Durham, University of California, Berkeley Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley A. Myra Keen, Emerita, Stanford University Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Washington Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley T. E. Thompson, University of Bristol, England Alex Tompa, University of Michigan, Ann Arbor Membership and Subscription Membership in the California Malacozoological Society is open to persons (no institu- tional memberships) interested in any aspect of malacology. Annual dues, which include a subscription to The Veliger, are US $18.75 plus mailing charges. An initiation fee of US $2.00 is required of new members; a reinstatement fee of US $1.00 will be required if membership renewals do not reach the Society on or before April 15 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign members, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $37.50 plus mailing charges. Mailing charges, both for individual memberships and library subscriptions, are US $3.25 for all domestic addresses and US $6.00 for all foreign addresses, including Canada and Mexico. Memberships and subscriptions are by Volume only (July 1 to April 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $30.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: C.M.S., Inc., Post Office Box 9977, Berkeley, CA 94709. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. TO OUR READERS The previous issue, Volume 25, Number 4, marked the completion of 25 years of growth and success of The Veliger, success due in great measure to the efforts of one man, Rudolf Stohler. The climb is by no means over. With Volume 26, Number 1, we begin again, advancing by standing on the shoulders of our predecessors. Some changes in format and style will be apparent to readers of the current issue. Our journal is now being produced by Allen Press, a quality commercial printing house specializing in scholarly journals and books. And there is a new Editor. Prospective authors may wish to consult the revised “Instructions to Authors” printed on the inside back cover. There are some changes. Despite these changes, our purpose remains the same. The California Malacozoological Society, through its journal The Veliger, remains committed to disseminating new information in the field of malacology and conchology as widely as possible at the lowest cost possible. We likewise reaffirm our continuing commitment to meeting ever higher standards for the publication of scientific information. There is no reason why The Veliger should not continue to grow for another 25 years. Our Society is vigorous and healthy. However, ominously escalating costs of publication have already threatened many a once healthy society. The simple truth is that we will need increased levels of income to balance our increased production costs. Readers can help ensure the continued publication of The Veliger by joining the Society (or by renewing their membership) and by encouraging their local and University libraries to subscribe. Donations are, of course, another extremely important means of ensuring the success of our Society (contributions to CMS are tax deductible). A new beginning is a time to reaffirm our purpose, to acknowledge the many splendid contributions that have brought us this far, and to seek ways to improve. The new Editor asks you to support the Society and invites the comments of readers, members, and prospective authors. He looks forward to serving you. D. W. Phillips, Editor Photograph by Leroy Poorman RUDOLF STOHLER OUR THANKS TO RUDOLF STOHLER With the completion of Volume 25 of The Veliger, Dr. Stohler has retired as Editor—stepping aside for some well- earned rest while still able to help us with his counsel and to inspire us with his unabated interest and enthusiasm. It is a good time to look back over the history of The Veliger, and to reflect gratefully how much we owe to the efforts of one devoted individual. To many who have read and contributed to The Veliger, Rudolf Stohler may be simply an exacting Editor, a stickler for accuracy and good order in writing; to those of us who from time to time have transgressed, he is one who can communicate in pithy language; to those of us who have appealed for help, he has been ready with advice and assistance. But Rudolf has done much more than found and edit The Veliger, and we should fill in the picture of his pre-Veliger years, for information on which we are indebted to Professor Emeritus Richard Eakin, long-time Chairman of the Department of Zoology at the University of California, Berkeley. Rudolf Stohler earned his Ph.D. at the University of Basel in his native Switzerland, and in the period of 1926-1932 published a series of papers on the chromosomes of European toads and their ovaries and Bidder’s organs. He came to California in 1928 as a Rockefeller Fellow to work with Dr. K. F. Meyer at the Hooper Foundation on the University of California’s San Francisco campus. The first day, as he arrived at the laboratory early, as was and is his habit, he picked up a large beetle that aroused his zoological curiosity. A young lady coming to work offered to obtain a bottle for the specimen. The Fellow was impressed with her kindness and friendliness and decided, then and there, that he wanted her for his wife. A year later Genevieve and Rudolf were married. Dr. Stohler had expected to continue his study of sex determination and differentiation in amphibians, but Professor Meyer had other plans for him, namely, that he work as an assistant on paralytic shellfish poisoning. At first it appeared that diatoms, found in abundance by Stohler in the guts of mussels and clams, might be the source of the lethal poison. Without more evidence Professor Meyer immediately published a paper, as sole author, to that effect. Later, Dr. Stohler disproved that hypothesis. The source of the toxin was found to be Gonyaulax or other red tide flagellates that are eaten by the bivalves. After the.completion of the fellowship, the Stohlers emigrated to Switzerland, but as the climate was unsuitable for the young bride the couple returned to Berkeley in 1932. Dr. Stohler was appointed a Research Associate (an honorary title but without salary) by Professor C. A. Kofoid, then Chairman of the Department of Zoology, University of California, Berkeley. Again Dr. Stohler was not free to pursue his own research because he was assigned to projects of Kofoid’s. To earn a living Stohler organized classes in German. Among his students were Alden Miller, later a Director of the Museum of Vertebrate Zoology, and Richard Eakin, later a Chairman of Zoology. In 1941 he was appointed to the position of Specimen Preparator and Collector for the Department of Zoology. In these capacities he ranged the California coast collecting for classes and for research, and developing a wide circle of friends and scientific colleagues. Later, under the Chairmanship of Richard Eakin, Dr. Stohler was relieved of some of his service responsibilities by appointment as a Research Zoologist (the research equivalent of a professorship) and was given an assistant and time to devote to his malacological interests. Dr. Stohler has also had a career of teaching. In 1934 he gave the departmental courses in cytology and assisted Professor Kofoid in courses in protozoology and parasitology while Professor Harold Kirby was on sabbatical leave. For many years he gave courses in the University of California Extension Division in Oakland, San Francisco, and Berkeley. And he has generously advised graduate students and faculty on procurement of research material, on nomenclatural and taxonomic problems, and on scientific writing. In the early 1950’s Rudolf Stohler and several colleagues of like interests founded the Northern California Malaco- zoological Club, the rather imposing name reflecting that the group was broadly interested in molluscan biology and not merely in “‘shell-collecting.” On June 27, 1958, appeared a mimeographed Club newsletter, ““The Veliger,” suitably named for an infant mollusc. The new publication grew rapidly and by its fourth year contained 220 pages and had appeared in printed format— made possible because Stohler had, out of his own pocket, acquired and set up in his basement an old linotype machine. Over twenty years later, on this machine, now supplied with a dozen different type-faces, Stohler still regularly set type for The Veliger, which since 1967 has run to over 400 pages annually. After many years’ investment of Stohler’s skill and labor, not to mention over $7,000 of his own money, The Veliger had clearly outgrown the needs of the Club and was incorporated by a small support group, the California Malaco- zoological Society, as a non-profit corporation. Stohler, with only a part-time business assistant and the help of various Photograph by Jane Scherr; courtesy of “California Monthly.” individuals (notably Mrs. Jean Cate), carried on all the essential functions of editing, printing, and distributing the journal, whose circulation is now over 800. The corporation undertook to repay Stohler his $7,000 investment. But, in setting up the corporation, Stohler had instituted an endowment fund, the capital of which cannot be touched, but the income from which can be used to help support publication, for example, of papers with expensive illustrations that cost more than an author can afford. As fast as the corporation paid back installments on the debt, Stohler put the money into the Endowment Fund, to aid the publication (contributions to this fund have come from many friends and are, of course, still enthusiastically welcomed). The success of The Veliger, which has never received a penny of foundation support, is a remarkable example, in this day of heavily-funded, group-supported ventures, of what one person with skill, energy, and devotion can accomplish. The first 25 volumes of The Veliger form a fitting and living monument, and it is our hope and intention that Rudolf will see it flourish for many years to come. To characterize Rudolf Stohler as a person is not easy. Even more solid than The Veliger, Rudolf radiates energy and enthusiasm. Generally at the Zoology Department well before anyone else, Rudolf moves with a brisk, perhaps bouncy, style. Extremely helpful to those needing and seeking his help, and sympathetic to personal needs of students, Rudolf is yet one who does not suffer fools gladly. Above all, he detests sloppiness in anything, physical or mental. His choicest comments have been reserved for his favorite béte noir, the U.S. Postal Service, but others have not escaped. Yet all who have known Rudolf know that behind the sometimes pungent manner there lies a deep decency, integrity, kindness, and high personal standards. Rudolf has not lightly retired as Editor of The Veliger, and doubtless feels like many a parent watching his child venture out into the world. His principal concern is that The Veliger maintain high standards. To this aim the Executive Committee of the California Malacozoological Society and the new Editor are committed. A final word needs to be said. In all his work for over 50 years Rudolf has had the support of his devoted wife, Genevieve, who has shared heavily in the labor and frustrations of producing The Veliger. Although inconspicuous in the background, she has nevertheless earned our deepest gratitude. R. I. Smith The Veliger 26(1):5-9 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 New Molluscan Species (Gastropoda: Neogastropoda) from the Tropical Eastern Pacific LEROY H. POORMAN 15300 Magnolia Street, Space 55, Westminster, California 92683 Abstract. Three new species are described: Murexiella venustula Poorman, spec. nov., off the south- ern coast of Isla Santa Cruz, Galapagos Islands, Ecuador, and probably the Gulf of California; Daph- nella levicallis Poorman, spec. nov., off Estero San Carlos, Sonora, Mexico; Anachis (Parvanachis) mullineri Poorman, spec. noy., in Bahia de Santiago, Colima, Mexico. TWENTY-NINE YEARS of observing and collecting mollusks along the west coasts of the Americas has resulted in the recognition of a number of new species, eight of which have already been published. Three additional species are described herein. Murexiella Clench & Pérez Farfante, 1945 Type species: Murex hidalgoi Crosse, 1869, by original designation. The small, stoutly fusiform shell has four or more var- ices with foliated spines connected by a laminated web- bing. The siphonal canal is moderately broad and extend- ed. The operculum is muricoid, with a sub-apical nucleus. Murexiella venustula Poorman, spec. nov. (Figures 1, 2, 5) Description: The shell is small and solid, consisting of five whorls and a turbinate protoconch of three turns. Axial sculpture is of seven strong, broad varices per whorl with narrow interspaces. Each varix crosses the shoulder area to the preceding whorl as a thin lamella but is not joined to it. The first several varices of the teleoconch cross the lower half of the last turn of the protoconch and are attached to it. Spiral sculpture on the body whorl is of five broadly flattened, strong cords that are made up of five scabrous threads. There are two cords on each whorl of the spire. All interspaces are narrow, wider at the bottom. At the base of the aperture is one minor cord and there are two strong, flattened cords on the canal. Anterior cords are hollow at their terminations and scarcely reflected as short, stout spines at the crests of the varices. The degree of reflection increases posteriorly along the varix. The cord at the shoulder is the heaviest and terminates as a large, reflected spine that is twisted toward the apex. There is a major, reflected, open spine at the center of the lamella on the shoulder and a much smaller one near the suture. Areas on the shoulder between the lamellae are flat and show only traces of the two cords causing the spines. The aperture is oval with a nearly complete peristome except along the parietal wall. A shallow, anal sulcus is apparent. Strong crenulations are along the erect outer lip. The spines behind the lip are roundly recurved and are joined for about half their lengths by intricate lamination. The an- terior canal is broad and moderately long, nearly straight, narrowly open to the right, and with the end distally re- curved. The top and left side of the canal each show four strong, longitudinal threads. Shell color is pinkish beige, with brown on the third spiral cord at the crests of the varices. Type locality: Off the southern coast of Isla Santa Cruz, Galapagos Islands, Ecuador; 0°47’S Latitude, 90°21/W Longitude; four specimens dredged in 150-200 m. Holotype: San Diego Natural History Museum, SDNHM 810610. Dimensions of the holotype: Height 19.6 mm, maxi- mum diameter 11.4 mm. Paratypes: Iwo paratypes are in the Carl and Laura Shy Collection, Seal Beach, California; one paratype is in the Rose Burch Collection, Seal Beach, California. One additional specimen was brought in by a fishing boat to Guaymas, Sonora, Mexico, in 1968, probably ex- tending the range to the Gulf of California. The specific name is taken from the Latin adjective meaning “pretty” or “charming little.” 5 INO: 1 No} N © > e o oe oO > o il a L. H. Poorman, 1983 Page 7 Discussion: This new species is closest to Murexiella mil- dredae Poorman, 1980, in general appearance and in the low, broad, spiral cords overhanging the narrow inter- spaces (POORMAN, 1980b). It differs in having two, not three, spiral cords on each whorl of the spire, in having numerous and broadly rounded varices, in coloring, and in other sculptural details. Murexiella venustula, spec. nov., also has a superficial resemblance to Murexiella laurae Vokes, 1970. However, the latter species has a different spine structure and has four varices per whorl (only three on some specimens) on the adult shell (POORMAN, 1980a). Also, the general col- oring of the two species is completely different. Hollow spines formed when the leading edges com- pletely circle to touch are not unusual in Murexiella. All three of the above species are of this type. Comparison was made with a specimen of Murexiella mactanensis Emerson & D’Attilio, 1979, from Bohol Strait, Philippine Islands. Although the shells differ in general appearance, the basic spine structure is similar. Daphnella Hinds, 1844 Type species: Pleurotoma lymneiformis Kiener, 1839-1840. The shell is cylindro-ovate, slowly contracted to a short, open, truncated, anterior canal. The body whorl is usually more than one-half the shell height. The protoconch is reticulated and of three to four turns. The sinus is sutural, reversed L-shaped. Spiral sculpture is of fine threads over- riding numerous fine, axial ribs. There is no operculum. The shell is irregularly maculated with brown. Daphnella levicallis Poorman, spec. nov. (Figures 3, 6) Description: The shell is larger and more inflated than most species in the genus. The protoconch (eroded on the holotype) is of four turns, narrowly turbinate and diago- nally reticulated by fine threads, with small beads at the intersections. The first turn of the protoconch is minute and has small beads (at 250X) arranged in spiral rows and diagonal lines which, by the second turn, become di- agonal threads. The last half turn of the protoconch de- velops a slight, peripheral angulation with the reticula- tions above the angulation sagging into a band and becoming the trace of the anal sulcus on the teleoconch. The angulation becomes a cord that develops into two l 0.5 mm Figure 5 Protoconch of Murexiella venustula Poorman, spec. nov. X 80. threads. These develop into the spiral cords on the teleo- conch. The protoconch terminates abruptly in a sinuosity advancing anteriorly. The teleoconch is of six rounded whorls constricted at the suture. Spiral sculpture is of strong threads throughout, 2 on the first turn and 25 on the body whorl. The trace of the anal sulcus is wide and flat, unornamented except for obscure fine threads. Axial sculpture is of numerous low rounded ribs, 10 on the first turn and 25 on the body whorl, extending to the anterior canal. Ribs are crossed by spiral threads in prominent nodes, strongest at the shoulder. The aperture is oval, with a deep J-shaped anal sulcus at the suture. The pillar is nearly straight anteriorly and has light callus. The outer lip is flaring but not produced forward, with a pronounced sinuosity at the lower part. A short, truncated, open, an- terior canal is differentiated from the aperture by an an- gulation in the lip. The outer lip is reinforced by a low ridge of callus just inside, thickest at the anal sinus and at the anterior part of the aperture. Shell color is pale brown maculated with red-brown except for an unmac- ulated band below the periphery. The third spiral cord below the shoulder is white on early whorls. Type locality: Five km south of Tetas de Cabra, Estero San Carlos, Sonora, Mexico; 27°54’N Latitude, 111°05’W Longitude; 16 specimens dredged in 80-100 m on broken shell, small rocks, and silt bottom. Explanation of Figures 1 to 4 Figure 1. Holotype of Murexiella venustula Poorman, spec. nov. x 5.6. Figure 2. Holotype of Murexiella venustula Poorman, spec. noy. x 5.6. Figure 3. Holotype of Daphnella levicallis Poorman, spec. nov. xX 6.0. Figure 4. Holotype of Anachis (Parvanachis) mullineri Poor- man, spec. nov. X 22. Page 8 0.5 mm l Figure 6 Protoconch of Daphnella levicallis Poorman, spec. nov. X 90. Holotype: San Diego Natural History Museum, SDNHM 810611. Dimensions of the holotype: Height 17.3 mm (apex eroded), maximum diameter 7.2 mm. Paratypes: Nine paratypes are in the Forrest and Leroy Poorman Collection; two paratypes are in the Carl and Laura Shy Collection, Seal Beach, California; four para- types are in the Paul and Carol Skoglund Collection, Phoenix, Arizona; one paratype will be placed at the Academy of Natural Sciences of Philadelphia. One additional specimen was dredged by Paul and Car- ol Skoglund in 100 m off Isla Danzante, Gulf of Califor- nia. The specific name is a Latin noun, masculine gender, and refers to the subsutural trace of the anal sulcus as a “smooth mountain path” much like the path up the Tow- er of Babel. Discussion: All of the recognized west American Daph- nella and the new species described here are very similar, differing only in size and details of ornamentation. The mechanism for developing the trace of the anal sulcus is observable on all. This subsutural band is also observable on the teleoconchs but is obscured except on Daphnella retusa McLean & Poorman, 1971, and the new species. Thickening of the outer lip occurs both internally and externally on the adults of all the species. Both Daphnella retusa and Daphnella levicallis, spec. nov., occur in significant numbers off Estero San Carlos but differ in their bathymetric ranges and habitats. Daph- nella retusa is found in 30 m on gravel bottoms; whereas D. levicallis is found in 100 m on shell and silt bottoms. Daphnella levicallis is larger, heavier, and of a darker color; the axial ribs are not obsolete on the body whorl, as they are on D. retusa, but extend nearly to the anterior canal. The ribs are crossed by strong spiral cords in prom- inent nodes not occurring on D. retusa. The trace of the anal sulcus is broader and flatter on the new species, with The Vielicers Volk ZG Nom Figure 7 Protoconch of Anachis (Parvanachis) mullinert Poorman, spec. nov. X 30. the axial ribs projecting slightly at their terminations to give a beaded effect to the shoulder. Anachis (Parvanachis) Radwin, 1968 Type species: Buccinum obesum C. B. Adams, 1845. The shell is small and obese, with a moderately high spire and flat-sided whorls with incised sutures. Body whorl and spire are of equal length. The apertural lip is thickened and denticulated. Sculpture is of prominent ax- ial ribs crossed by spiral cords. Anachis (Parvanachis) mullineri Poorman, spec. nov. (Figures 4, 7) Description: The shell is small and stout, consisting of a turbinate protoconch of four smooth turns and a teleo- conch of three and one-fourth whorls, terminating in a large, rounded, lip varix that decreases in size anteriorly. Transition from protoconch to teleoconch begins with weak, slanted ribs advancing anteriorly. The transition takes about one-half whorl when the ribs become longi- tudinal, rounded, and with equal interspaces. There are about 20 ribs on each whorl. The ribs are abruptly con- stricted just above the indented suture to leave widened areas in the interspaces. The ribs are rounded, protrude slightly above the shoulder, and are obsolete on the base. A spiral groove just below the indented suture cuts the ribs to produce a row of squarish beads. Below this is a region with no spiral sculpture. The middle half of the body whorl shows nine strong spiral grooves in the inter- spaces of the ribs. On the base, the spiral grooves override the diminishing ribs producing strong, flat-topped spiral cords. The entire surface of the teleoconch is covered with minute, spiral striae. The narrow aperture is somewhat trapezoidal. The outer lip is sharp, erect, and slightly crenulated by the spiral cords overriding the lip varix. Within the outer lip are six denticles. Columellar callus L. H. Poorman, 1983 is produced into a lamella with a chink behind, both of which extend to the end of the canal (chipped on the holotype). Within the aperture, along the pillar, is a lon- gitudinal ridge of callus with six denticles. The anal sul- cus is semicircular in cross section and slightly constricted by parietal callus. The sulcus penetrates the outer lip and varix at an angle of 60° with the axis and curves to ter- minate at an angle of 90°. The short, anterior canal is at an angle of 30° left of the shell axis, scarcely differentiated from the aperture, and broadly open to the right. The shell is light horn color with a band of darker brown above the periphery and a second indefinite band on the base. Type locality: Bahia de Santiago, Colima, Mexico; 19°02’N Latitude, 104°28’W Longitude; five specimens dredged in 20 m on sand and gravel bottom. Holotype: San Diego Natural History Museum, SDNHM 81612. Dimensions of the holotype: Height 5.0 mm, maximum diameter 2.8 mm. Paratypes: Four paratypes are in the Forrest and Leroy Poorman Collection. Additional specimens in the Paul and Carol Skoglund Collection, Phoenix, Arizona, are from: Cuastecomate, Jalisco, dredged in 25-33 m; La Cruz de Juanacaxtle, Bahia de las Banderas, Nayarit, dredged in 20 m; Playa Novellero, Nayarit, diving in 8-12 m. The specific name is chosen in recognition of David K. Mulliner, San Diego, California, a good friend known to all for his generous contributions of time and talent to the field of malacology. Page 9 Discussion: The massive lip varix and the unusual anal sulcus curving across it, together with the erect lamella along the pillar, are distinctive and make further compar- ison with other existing taxa unnecessary. ACKNOWLEDGMENTS I wish to recognize with thanks the contributions of An- thony D’Attilio, San Diego Natural History Museum, who made the protoconch drawings, and David Mulliner for technical assistance in preparing the illustrations. LITERATURE CITED CLENCH, W. J. & I. PEREZ FARFANTE. 1945. The genus Mu- rex in the western Atlantic. Johnsonia 1(17):1-56. EMERSON, W. K. & A. D’ATTILIO. 1979. Six new living species of muricacean gastropods. Nautilus 93(1):1-10. HInps, R. B. 1844{-1845]. The zoology of the voyage of H.M.S. Sulphur ... Mollusca, pt. 2, pp. 25-48 (published October 1844) London. McLean, J. H. & L. H. PooRMAN. 1971. New species of tropical eastern Pacific Turridae. Veliger 14(1):89-113. PoorMAN, L. H. 1980a. Reinstatement of two species of Mu- rexiella (Gastropoda: Muricidae) from the tropical eastern Pacific. Veliger 22(3):273-276. PooRMAN, L. H. 1980b. Two new molluscan species (Gas- tropoda: Muricidae) from the tropical eastern Pacific. Ve- liger 22(4):361-363. RapDwIn, G. E. 1968. New taxa of western Atlantic Colum- bellidae. Proc. Biol. Soc. Wash. 81:143-150. VoKEs, E. H. 1970. The west American species of Murexziella (Gastropoda: Muricidae), including two new species. Veli- ger 12(3):325-329. The Veliger 26(1):10-17 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 The Biology of the Northeastern Pacific Turridae. III. ‘The Habitat and Diet of Kurtziella plumbea (Hinds, 1843) by R. L. SHIMEK' Friday Harbor Laboratories, Friday Harbor, Washington 98250 Abstract. The habitat, diet, and some reproductive and mortality factors were determined for Kurt- ziella plumbea, a shallow-water turrid gastropod from the Puget Sound region. In a shallow subtidal area with four depth-related habitats, it was found more often in upper habitats, although all of the habitats available were sandy and had similar sedimentary characteristics. In a deeper area, it was found primarily in areas of silty mud. Kurtziella plumbea is a predator of polychaetes. At the shallow site, it specialized on the oweniid Owenia fusiformis and spionids. In the deeper area, it was more generalized, but the cirratulid Tharyx multifilis was the most abundant prey item. Egg capsule deposition occurred in the spring. There were no nurse eggs, and the time until veligers left the capsule was about 18-20 days. Settlement and metamorphosis were not observed. Growth rates of 1 to 2 mm/yr for median-sized individuals, and 3 to 4 mm/yr for small individuals were indicated. Potential predators may include crabs and fish, but mortality factors were not conclusively deter- mined. INTROIDUCTION THE MOST WIDELY distributed turrid gastropod in the shallow northeastern Pacific is Kurtzzella plumbea (Figure 1), which is found from Mazatlan (MCLEAN in KEEN, 1971) to southcentral Alaska (Shimek, unpublished data). I examined two populations of Kurtziella plumbea (here- after as Kurtzzella) to determine habitat, dietary require- ments, and aspects of predatory and reproductive behav- ior. I attempted to determine some of the reasons why this particular snail is so widespread. Two mutually contra- dictory statements about this species were examined. (1) The species is generalized in both its use of habitats and prey, and thus it can be found in many habitats. In effect, it would be a successful “Jack-of-all-trades.” (2) The species is specialized in habitat and/or diet, but the array of suitable habitats is widespread, thus permitting the snail to be widespread. Several major questions were addressed. (1) What is the relationship of diet to the potential dietary resources present? (2) Are these animals dietary or habitat ' Current mailing address: P.O. Box 6683, Bellevue, Wash- ington 98008. specialists or generalists? (3) What is the effect of these predators upon their prey populations? (4) Are there any particular traits that limit their choices of habitats or prey? MATERIALS anp METHODS Study Sites All sites were subtidal, since Kurtziella is rarely inter- tidal in this region. Two major study sites were chosen: Windy Point in Dyes Inlet in lower Puget Sound and off the University of Washington Friday Harbor Laborato- ries dock on San Juan Island (Figure 2). Additional spec- imens were collected from many other localities, particu- larly in the San Juan Islands (SHIMEK, 1977); however, at these latter sites I collected only distributional data; quantitative data were seldom obtainable. Distributional information was gathered by dredge or bottom tow. All quantitative field work was done using SCUBA. The Windy Point area, WP (47°37’25’N, 122° 40’30”W), ranges from —1.5 m to —9.0 m below MLLW. It is a topographically homogeneous, sandy subtidal re- gion unbounded laterally and divisible into four visually distinctive habitats: upper bench, upper slope, lower slope, R. L. Shimek, 1983 Page 11 Figure 1 A. Kurtziella plumbea (Hinds, 1843). Scale in millimeters. B. Radular teeth of Kurtziella plumbea. Scale bar = 10 um. These are stabbing toxoglossan teeth used individually to pierce the prey and to introduce venom. and lower bench, based upon the degree of slope, depth, and algal cover (SHIMEK, 1982). Kurtziella is found there with two other turrids, Ophiodermella inermis (Hinds, 1843) and Oenopota levidensis (Carpenter, 1864). The Friday Harbor Laboratories site, FHL (48° 32/38”N, 123°00’50”W), is topographically diverse, con- taining five visually distinctive habitats: wood chips, rock, shell fragments, and shallow and deep areas of silty mud. The site is located from —10 m to —25 m below MLLW. No discontinuities limit the site except at the upper edge where the boundary is established by the lower edge of an eelgrass, Zostera marina L., 1753, bed. No turrids were found in either the eelgrass or wood-chip areas in prelim- inary observations, consequently no quantitative sampling was done in either habitat. The remaining habitats were sampled quantitatively. Page 12 mrAAaAbPMo Figure 2 The Puget Sound region of Washington state showing the study sites. Habitat Analyses At the major sites, the physical and biological properties of the habitats were examined in detail. Sediment particle- size distributions were quantitatively determined and the remainder of the sample was washed through a 0.5-mm sieve. The animals were removed and sorted by taxon. Gastropods and polychaetes were identified to species whenever possible. Other taxa were identified to class and counted, but not detailed further (see SHIMEK, 1982, for a more complete description of sampling methods). Sediment parameters were tabulated and statistically compared between and within the areas. Seasonal vari- ability of the sediment particle distributions was insignifi- cant, thus no seasonal comparisons were made (SHIMEK, 1982). Polychaete assemblage abundances for each habitat were determined by the quantitative, infaunal sampling, and The Veliger, Vol. 26, No. 1 statistical comparisons were made between and within habitats on a seasonal basis (SHIMEK, 1982). Turrid Distribution, Collection, and Processing Periodic transect studies from November, 1973, until December, 1975, were used to determine turrid distribu- tion, seasonal or other distributional changes, and to pro- vide a reference for the quantitative infaunal samples. ‘The significance of Kurtziella distributions compared to ran- dom habitat utilizations were calculated using log-likeli- hood ratios (G-tests) (SOKAL & ROHLF, 1969). A hand-held, semiquantitative dredge was used at FHL to determine the relative fraction of the turrid populations buried in the sediment. Simultaneous, parallel, surface- transect surveys were conducted to compare the number of snails buried and on the surface. Kurtziella collected from FHL and WP were individ- ually washed in sea water, isolated for up to a week, and feces were collected. The snail was then measured to the nearest 0.1 mm, and marked (SHIMEK, 1982). After mark- ing, the animal was returned to fresh sea water at ambient sea-water temperature and observed to assure no notice- able effects of measuring and marking. The animal was then transferred to a “holding” aquarium and maintained in an artificial habitat similar to the normal one. All ap- parently healthy animals were returned to their habitat, albeit seldom to the point of capture, within two weeks. Measuring and marking mortality was about five percent. Following marking and measuring, any particulate ma- terial remaining in the collecting jar was placed on a slide, dried, mounted in polyvinyl lactophenol (A. Kohn, per- sonal communication), and examined microscopically. Identification of all fecal material was attempted. Feces consisted of mucus, radular teeth of the same animal, diatom frustules, and polychaete remains. Preliminary gut analysis by dissection indicated polychaetes swallowed whole to be the only prey. Thus only polychaete remains consisting of setae, Jaws, and occasional cuticular strips were accepted as indicators of feeding. These remains were identified by comparison with descriptions, drawings, and setal preparations of known animals, identified with stan- dard references (SHIMEK, 1982). Dietary heterogeneity was measured using H’ (KOHN & NyYBAKKEN, 1975). Turrids collected from other localities were preserved and their habitats noted, if the collecting was done with SCUBA, but no dietary analysis was attempted. Size-frequency histograms were constructed for both populations. These were normalized to percent collected to facilitate comparison between populations because of varying sample sizes. For the purpose of determining growth rate, collections were considered quarterly: No- vember through January as Winter; February through April as Spring; May through July as Summer; and Au- gust through October as Autumn. Generally these samples were too small and/or variable for quantitative determi- nations of recruitment cohorts (BLISS, 1967); however, I R. L. Shimek, 1983 Page 13 attempted to use seasonal shifts in histogram peaks to estimate growth rates. Laboratory Experiments A substrate-choice chamber was constructed, and filled to a depth of 2 cm with sediments (SHIMEK, 1982). The choices were sediments with a particle-size distribution from 0.250 to 0.500 mm, and a distribution in excess of 2.00 mm. Both sides had all detectable biota removed. Animals were placed in the chamber and one week later they were collected and their positions noted. These data were analyzed using cumulative binomial probabilities. Egg capsules were collected from the jars in which the turrids were stored. Capsular dimensions were measured, the number of eggs per capsule was counted, and the egg diameters were measured (SHIMEK, 1982). The capsules were examined periodically. After hatching, the veligers were fed a mixture of Isochrysis sp. and Dunaliella sp. RESULTS Habitat Descriptions Windy Point: The four Windy Point habitats have been described in detail elsewhere (SHIMEK, 1982). Briefly, these habitats were sandy, with moderately well-sorted, uncon- solidated sediments. They were similar to each other and to the nearby sandy low intertidal areas. Algal cover in the shallower (depth < —5 m) areas varied seasonally, being very abundant in the late summer, and was mostly ulvoid algae. In the deeper areas the algal cover consisted of various red and brown algae and was less variable or dense. Friday Harbor Laboratories: The FHL habitats are also described in SHIMEK (1982). Of the four major turrid habitats, only the rock areas could not be sampled quan- titatively for infauna and sediment. Of the three un- consolidated-sediment areas, only the lower mud was physically different, having a distinctly smaller median- sediment-particle size. Biology of Kurtziella at Windy Point Kurtziella was associated with the turrids Ophiodermella inermis and Oenopota levidensis at WP. All three were distributed in patches, and because of this, the mobility of these animals, asd the lack of physical boundaries to the study area, no adequate estimates of population sizes could be made. Capture and transect observational frequencies did, however, give an estimate of relative population sizes. I collected or observed 254 Ophiodermella inermis, 134 Kurtziella, and 108 O6enopota levidensis. Kurtziella and Oenopota levidensis appeared to have roughly equivalent populations. During quantitative surveys, the density of Kurtziella, when found, varied from 0.01/m? to 0.16/m?. No seasonal trends in abundances or habitats utilized were seen, but Kurtziella was found more often in the shallower Table 1 Windy Point Kurtziella plumbea habitat utilization. Proportion of turrids Proportion Proportion observed — of total of turrids proportion Habitat area per area expected Upper bench 0.25 0.34 +0.09 Upper slope 0.25 0.26 +0.01 Lower slope 0.25 0.17 —0.08 Lower bench 0.25 0.24 —0.01 Number observed 134 Significance (G-test) G = 9.40 P < 0.05 areas at WP (Table 1). The sediment-particle distribu- tions were not significantly different among most of these areas (SHIMEK, 1982). Depth was probably not a factor as Kurtziella was found in the deeper areas at FHL (Ta- ble 2). There was a patchy distribution of the polychaete fauna at WP, particularly regarding the turrid prey species Ow- enia fusiformis delle Chiaje, 1844, which was dense only in the upper bench areas, and Polydora socialis (Schmarda, 1861), which was more widely distributed than O. fuszfor- mis (Table 3). There was no seasonal pattern of predation at either site, consequently the prey-polychaete-abun- dance data were pooled. The high sample variability re- flected seasonal abundance patterns for the worms; see Table 2 Friday Harbor Laboratories Kurtziella plumbea habitat utilization. Propor- Proportion Propor- tion of of turrids tion of _ turrids observed — total per proportion Habitat area area expected A. All habitats Upper mud 0.25 0.06 —0.18 Shell fragments 0.34 0.35 +0.01 Rock 0.05 0.06 +0.01 Lower mud 0.36 0.52 +0.16 Total number observed 48 Significance (G-test) P < 0.01 B. Lower habitats only Shell fragments 0.45 0.38 —0.07 Rock 0.07 0.07 0.00 Lower mud 0.48 0.56 +0.08 Number observed 45 Significance (G-test) n.s. Page 14 A. Area: Windy Point Habitats: Prey species The Veliger, Vol. 26, No. 1 Table 3 Density of prey species (mean number/m? + 1 SD). Upper bench Upper slope Lower slope Lower bench Owenta fusiformis 713 + 413 IG se 37/5) ) ar DD 0 Polydora socialis 843 + 907 661 + 591 468 + 440 268 + 440 Spiophanes berkeleyorum Ole F 0 5 + 16 0 B. Area: Friday Harbor Laboratories Habitats: Upper mud Shell fragments Lower mud Prey species Myrvochele oculata 6 ae iY) 0 0 Cirratulus cirratus Sil ae 56) 23) 22 Si 7X0) ae 37 Tharyx multifilis 136 + 125 YS 22 T7/ 1B) = (35) Spiophanes bombyx not sampled SHIMEK (1977, 1982) for more complete listings of the polychaetes. Kurtziella was found to be widely, but unevenly, dis- tributed in the Puget Sound region. I dredged and/or surveyed by SCUBA 40 different sites and Kurtziella was found in only six of them (SHIMEK, 1977). All six were characterized as sandy or sandy-mud habitats. Dietary Analysis Dietary information obtained by fecal examination is summarized in Table 4. Kurtziella at WP ate three species of identified polychaetes. Densities of the most common prey, Owenza fusiformis, fluctuated dramatically in all but the upper-bench area, and this worm was absent from the lower-bench area completely (Table 3). Predation on Kurtziella at WP In the WP-slope habitats predation by crabs upon the snails may have been an important factor. Two predatory Table 4 Results of fecal sample analysis of Kurtziella plumbea. Area: WP FHL Total Prey species Myrtochele oculata 1 1 Owenia fusiformis 13 13 Polydora soctalis 1 1 Spiophanes berkeleyorum 2 2 S. bombyx 1 1 Cirratulus cirratus 1 1 Tharyx multifilis 3 3 Unidentified polychaetes 1 4 5 Number of snails examined 139 110 249 Percent feeding 22 9.1 10.8 H’ (identified prey only) 0.60 1.24 1.36 crabs were present, Cancer gracilis Dana, 1852, and C. productus Randall, 1839, although specimens of the latter were uncommon, probably because of too few suitable refuges (SHIMEK, 1982). During the summer when the crabs were common, Kurtziella became rare (Figure 3). Consequently, no laboratory verification of the attractive- ness of Kurtziella as a prey item was attempted. Cancer productus will eat both Oenopota levidensis and the larger Ophiodermella inermis in the same area, and Cancer gra- culis may eat small snails of all types. As Kurtziella was smaller than the other turrids at WP, it was a likely prey item for both crabs. Biology of Kurtziella at Friday Harbor Laboratories The turrid assemblage at FHL is diverse, in addition to Kurtziella, nine other turrids in three other genera are found there (SHIMEK, 1982, 1983). Habitat utilizations were determined from transect surveys. As with the other turrids at FHL, Kurtziella was less common in the upper- mud area than would be expected if it was randomly distributed (SHIMEK, 1982, 1983). If the upper-mud area is considered to be a boundary area that is incompletely utilized, and the data from this habitat are excluded, Kurt- ziella was collected most frequently in the lower-mud hab- itat (Table 2). When found, Kurtziella had measured abundances from 0.01/m? to 0.05/m*. The polychaete fauna at FHL was diverse and sparse. The most abundant turrid prey, Tharyx multifilis (Moore, 1909), were present in most of the samples and habitats (Table 3). The infaunal assemblages from all of these habitats were similar in the summer, but in the winter the upper- and lower-mud areas had many differences (SHIMEK, 1982). The substrate-preference experiments gave conclusive results, even with the small number of individuals avail- able at any one time for each test. Of the animals tested, R. L. Shimek, 1983 Page 15 17 animals made choices; 13 chose the sediment with smaller particle-size distribution (0.250 mm-—0.500 mm), whereas only 4 chose the sediment with particles larger than 2.00 mm. The two-tailed binomial probability of a deviation this large or larger, given an equal probability of choice, is 0.049. The hand-held, semiquantitative dredge was used in- frequently as it resulted in substantial habitat damage. Data from these dredge samples indicated equivalent numbers of Kurtziella buried and on the surface at the times of the surveys. During the surface surveys, five Kurtziella were found in 583 m?’. Simultaneous dredging parallel to, but 2 m lateral to the surface-survey-transect lines collected one Kurtziella in 116.6 m?® dredged to a depth of 0.10 m. Dietary Analysis Relatively few animals were found eating (Table 4), and no seasonal dietary trends were evident. At WP, Kurt- ziella was most often feeding on Owenia fusiformis, al- though spionids were also found in the diet. At FHL, no such clear-cut pattern of specialization was evident, and the snails had a more catholic diet consisting of cirratulids, spionids and oweniids. In contrast to the prey distribution at WP, the major prey taxon at FHL, the Cirratulidae, was common in all habitats (Table 3). Kurtziella Reproduction Collection and confinement for fecal-sample examina- tion acted as stimuli for egg-capsule deposition in some turrids (SHIMEK, 1982, 1983). Eight individuals of Kurt- ziella deposited egg capsules in captivity in March, April, and May, 1975 (Table 5). Single females deposited from one to three capsules. Capsules were smaller than the egg capsules of either Oenopota or Ophiodermella, and con- tained fewer, smaller eggs (SHIMEK, 1982, 1983). The egg capsules were deposited on the inside of the collection jar at the junction of the lateral and bottom surfaces. The eggs in 10 capsules hatched after about 18-20 days. As with Oenopota and Ophiodermella, there were no nurse eggs, and the number of veligers leaving the capsules was the same as the original number of eggs deposited. Four capsules contained eggs that did not develop at all. Pre- sumably the stimulus for oviposition was so strong that unfertilized eggs were deposited in the capsules. These capsules appeared normal in all other respects. No egg capsules of Kurtziella were seen in the field. Life Histories Seasonal size-frequency distributions can be examined for indications of growth rate, recruitment, and sizes of individuals in the population. A comparison of the FHL and WP populations yields some interesting observations. The WP population was composed of distinctly smaller individuals with virtually no animals exceeding 12 mm S ce) Oo Oo S fe) Ae) (e) ine) Cancer observed Kurtzie//a collected SPR SUM AUT WIN Season Figure 3 Mean (+1 SD) number of Kurtziella plumbea (open bars) col- lected and mean number of Cancer gracilis (black bars) observed (per 25 m?) by season. All habitats and seasons were pooled. Data are for Spring, 1974, through Autumn, 1975; therefore, Kurtziella was collected in only one winter. total length (Figure 4). On the other hand, the FHL pop- ulation was composed of larger animals, with almost no animals below 10 mm total length (Figure 5). Neither population showed appreciable seasonal shifts in the size- frequency histograms, although in both populations the mean individual length increased from spring to winter, and dropped again the following spring. The pattern of changes in the mean sizes in both distributions was vir- tually identical although consistently displaced by the amount the two populations differed in mean length. Ex- amination of the seasonal shifts for both populations in- Table 5 Kurtziella plumbea reproductive information. Mean capsule Mean egg SISO NB) Number/ Diameter Length Width capsule (um) 230 + 34 186 +22 180+ 43 137+8 Number examined 13 6 9 Number of capsules hatching: 10 Mean number of days in the capsule: 19.3 + 1.16 days Maximum length of survival post hatching: 27 days Page 16 The Veliger, Vol. 26, No. 1 WINTER 1974-75 N=19 SPRING 1974 N= 43 40 Ow 2 26 oO v Oo oO S = 2 SPRING 1975 SUMMER 1975 : 404 N=4! X& (e} aS QO 5 10 15 5 10 IS Length (mm) Figure 4 Windy Point seasonal size-frequency distributions. In seasons not shown (Summer, Autumn, 1974) too few snails were ob- served to make the data meaningful. dicated a growth rate, for a 7-8 mm-long individual, of about 1.5 mm/yr. No marked animals were recaptured from the WP population. Five previously marked individ- uals were ‘recaptured from the FHL population. Three decreased in total length, and two increased. One of those two had grown 0.5 mm in 287 days, indicating a growth rate of less than 1 mm/yr. This individual was 11.9 mm long when initially captured, however; and the larger in- dividuals may grow more slowly than smaller ones, which was a pattern seen in the other turrids examined from these areas (SHIMEK, 1982, 1983). In both populations, distinctly smaller animals were recovered only in the spring. If this species has a faster growth rate in the smaller size classes, the 4-6 mm-long individuals recovered in the spring of one year may rep- resent larvae settling the previous summer. Growth rate determinations were difficult for these animals due to their small size, the wide variation in lengths, and the lack of more than one clearly defined peak in the size-frequency histograms. DISCUSSION As with sympatric turrids in the genera Oenopota and Ophiodermella, Kurtziella eats tube-dwelling polychaetes; 604 SPRING 1974 N= 45 me} & Oo @ © O S = oe N WINTER 1974-75 SUMMER 1975 = N=12 N= 21 3 e 40 ss 5 © 5 ye (© 05 Length (mm) Figure 5 Friday Harbor Laboratories seasonal size-frequency distribu- tions. In seasons not shown (Summer, 1974; Spring, 1975) too few snails were observed to make the data meaningful. but the range of prey taken is quite broad, from oweniids to cirratulids, although at WP, Owenia predominates. The diversity of acceptable prey may be less at WP, possibly limiting animals in this population to fewer potential prey species. This might also explain the tendency of the Kurt- ziella to be found in areas where Owenia was common. Because manipulative experiments were not attempted, the turrid’s effect on its prey is uncertain, but predation by Kurtziella alone probably did not have any substantial impact on the populations of its prey. The species ap- peared to have ample food sources; indeed, Tharyx mul- tifilis, its major food, was one of the most abundant poly- chaetes in the FHL areas, and Owenia fusiformis was very abundant in some habitats at WP. Kurtziella appears to be restricted to sandy or sandy- mud habitats and is rarely found on nearby rocky or shell- fragment habitats. Additionally, it seems to have substan- tial latitude in potential prey, although more data are needed to confirm this. The skewed distribution of Kurt- ziella at WP may indicate tracking of its most common prey species, Owenia fusiformis. Qwenia was rarely found Rewer Shimek 1983 Page 17 below the upper-slope areas, and Kurtziella is more com- mon in these upper areas than in the lower ones. Predatory effects on this species are difficult to deter- mine. The major predators in the WP area are inferred to be the crabs, Cancer gracilis and C. productus. Both crabs were common and have been shown to eat other, larger, turrid gastropods in the same area (SHIMEK, 1982, 1983). Nonetheless, no direct evidence of this predation was encountered. Kurtziella is a small snail with a rela- tively fragile shell, and attacks by these large crabs prob- ably result in the complete destruction of the shell. Unlike Ophiodermella and Oenopota species, Kurtziella does not show any significant tendency to bury; this may result in substantial mortality when the predatory crabs are com- mon. The relative rarity of the species at WP during and after the summer population peak of the crabs is likely the result of predation. The smaller mean length of in- dividuals in the WP population, compared to the FHL population may result from some size-selective predation by the crabs on larger snails. Immigration from deeper habitats where the crabs are less common is probably re- sponsible for the recovery of the population. Mortality effects at FHL are unknown. During the course of study, over 100 turrid shells were recovered with hermit crabs in them: These shells proved useful in de- termining some of the causes of mortality of some of the other turrids in the region (SHIMEK, 1982, 1983), but no Kurtziella shells were recovered. This may be indirect evi- dence of predation by crushing predators like crabs, which can sometimes be found in the FHL habitat. Some fishes may also eat the snails. Except for Luidia foliolata Grube, 1866, predatory asteroids are uncommon in the area. In both populations, the snails appear to have a defined maximum size of about 15 mm. Growth may cease or slow as the animals approach this size, thus making mea- surement of growth rates difficult. There is no indication of semilparity in this species. After spawning, the females appear healthy, and all were marked and returned to their habitat. In conclusion, Kurtziella requires sandy or sandy-mud habitats, and overall is a dietary generalist, although it may specialize in some populations. Thus, the hypothesis that Kurtziella plumbea is a “ Jack-of-all-trades” for both diet and habitat is rejected. It is likely that suitable sandy or sandy-mud habitats are widespread on the Pacific coast of North America; this turrid should be expected in many of them. That this species will be rarely found in rocky or shell-fragment areas is also predicted. ACKNOWLEDGMENTS Portions of this paper are taken from a dissertation sub- mitted in partial fulfillment of the requirements for a Ph.D. in the Department of Zoology at the University of Wash- ington. Members of my graduate committee, particularly Dr. Alan Kohn, Dr. Eugene Kozloff, Dr. Paul Illg, and Dr. Ken Chew, made many helpful suggestions. The field work for this study was done with SCUBA, and while I cannot thank all 83 of my diving partners individually, Paul Raymore, Carl Nyblade, Larry Moulton, Steve Bloom, Ken Sebens, Kathy DeRiemer, and Ed DeMartini were especially helpful. The Pacific Northwest Shell Club, Elsie Marshall and Hal Scheidt in particular, provided much information of value in locating study sites. I thank Dr. Robert Fernald and Dr. Dennis Willows, directors, and Dr. Richard Strathmann and Dr. Eugene Kozloff, acting directors, of the University of Washington Friday Harbor Laboratories for allowing the use of facil- ities. Earlier drafts of this manuscript were read in part or totally by A. Kohn, E. Kozloff, R. Paine, and R. Fred- rickson, and two anonymous reviewers. I thank them all for their helpful suggestions. This work was partially supported by N.S.F. Grant 75-03303 to Dr. Kohn, by an N.S.F. doctoral dissertation grant GA-41814, by grants from the Friday Harbor Lab- oratories, and by two scholarships from the Pacific North- west Shell Club. A final special debt is owed to R. Fredrickson, for with- out her help this work would not have been completed. LITERATURE CITED Buiss, C. I. 1967. Statistics in biology. Vol. 1. McGraw-Hill, New York. 558 pp. KEEN, A. M. 1971. Seashells of tropical west America. 2nd ed. Stanford Univ. Press, Stanford, Calif. 1064 pp. Koun A. J. & J. W. NyBAKKEN. 1975. Ecology of Conus on eastern Indian Ocean fringing reefs: diversity of species and resource utilization. Mar. Biol. 29:211-234. SHIMEK, R. L. 1977. Resource utilization and natural history of some northeastern Pacific Turridae. Doctoral thesis, Zo- ology, University of Washington, Seattle. SHIMEK, R. L. 1982. The biology of the northeastern Pacific Turridae. I. Ophiodermella. Malacologia 23:281-312. SHIMEK, R. L. 1983. The biology of the northeastern Pacific Turridae. II. Oenopota. J. Molluscan Stud. (in press). SOKAL, R. R. & F. J. ROHLF. 1969. Biometry. W. H. Free- man, San Francisco. 776 pp. Editor’s note: After the issue had gone to press, we were informed that the paper referred to as “Shimek, 1982” throughout the present article was published in 1983. The Veliger 26(1):18-21 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 Flow Through and Around the Abalone Haliotis kamtschatkana by JANICE VOLTZOW Department of Zoology, Duke University, Durham, North Carolina 27706 Abstract. Fluorescein dye streams released along the shell and at the shell openings of living and dead abalones revealed the lines of water flow around and through the shell. Regardless of ambient flow speed, water entered the shell of live abalones at a region to the left of the left cephalic tentacle and also through the one or two most anterior shell openings and exited through the two or three most posterior openings. An identical flow pattern occurred through the shell of dead, intact animals when placed in an ambient flow of 6 cm/s, regardless of whether the anterior end faced upstream or down- stream. Since the exhalent openings occur at the site of the shell at which the ambient streamlines are maximally compressed, they experience a reduction in pressure relative to the anterior, inhalent ones. Thus, the design of its shell may enable the abalone to take advantage of an induced flow to move water more efficiently through its mantle cavity. INTRODUCTION Hahotis kamtschatkana Jonas, 1845, the pinto abalone, must deal with changes in environmental flow while maintain- ing respiratory currents through its mantle cavity. These animals have a row of four to six openings through the dorsal shell surface that, according to earlier workers (GARSTANG, 1928; CROFTS, 1929; YONGE, 1947) evolved to prevent wastes released in the mantle cavity from mov- ing forward over the gills and head. In addition to creating a flow through its shell by beating the cilia of its gills, Haliotis could use environmental currents to help drive water through its mantle cavity. Such an induced flow requires that the inhalent openings encounter fluid at a pressure higher than that of the exhalent openings (VOGEL, 1981). MuRbDock & VOGEL (1978) found that the keyhole limpet, Diodora aspera, experiences such an externally driven flow and may use this flow to move water more efficiently through its mantle cavity. Haliotis kamtschat- kana lives in areas of surge and strong currents (Richard Emlet, personal communication), where it seems possible that its series of shell openings could be exposed to pres- sure differences sufficient for induced flow to occur. To visualize the paths that lines of water flow follow around and through abalones I combined information from dis- sections, dye streams, and flow velocity measurements tak- en in and around living and dead individuals. MATERIALS anp METHODS Five adult individuals of Haliotis kamtschatkana ranging in maximum aperture diameter from about 9-12 cm and collected from various sites in the Friday Harbor region of San Juan Island, Washington were used. The flow these animals normally encounter may range from 0-150 cm/s or more (Kenneth Collier, personal communication). To relate the external flow information to the anatomy ~ of the animal, I dissected both fresh and narcotized ani- mals. To narcotize, I added increasing volumes of a 7% MgCl,-6H,0 solution to animals in cooled (8°C) sea water until the epipodial and cephalic tentacles no longer re- tracted when touched. To visualize the flow of water qualitatively through and around Haliotis kamtschatkana, | used a stream of fluores- cein dye controlled by a micro dye injector (an adjustable syringe fitted with a drawn plastic tip and moved by a micromanipulator). Pencil marks at 1-cm intervals along the edge of the shell served as reference points for follow- ing dye paths, technically referred to as streaklines. Dye was released at each of these points along both sides of the animals as well as at various sites near the anterior and dorsal surfaces of the shells. Studies of animals in flowing and in still water took place in a 15-cm flow tank designed by VOGEL & LABARBERA (1978). Observations of induced flow were made on animals J. Voltzow, 1983 Page 19 Borel eRe tye rt hata s x ae nie Se 1 PP aa cee Figure 1 Diagram of flow through and around an abalone facing upstream in an ambient flow of about 6 cm/s. l.c.t.: left cephalic tentacle; r.c.t.: right cephalic tentacle; m.t.: mantle tentacle; A: site at left side of shell where water enters mantle cavity; 1, 2: anterior, inhalent shell openings; 3, 4, 5: posterior, exhalent shell openings; dot-dash lines (—----): inhalent streaklines; dashed lines (— — —): exhalent streaklines; dotted lines (----- ): flow along anterior end and side of shell. Letters and numbers at ends of exhalent streaklines correspond to the points at which streaklines may have entered the mantle cavity. that were first relaxed in MgCl, as above and then quickly frozen to —20°C by placing them in the bottom of a cryo- stat. The animals were returned to ambient sea water temperature before any data were taken. RESULTS Dye released at specific points around an animal facing upstream in an ambient flow of about 6 cm/s formed a repeatable set of streaklines (Figure 1). Water approach- ing the anterior edge of the shell was either deflected up- ward (perpendicular to the general direction of flow), straight upstream (180° to the general flow), or was passed in a series of vortices along either side of the shell. Flow contacting the shell at either side continued downstream close along the shell and then became caught in the tur- bulent backwash behind the animal. In ambient flows ranging from approximately 2-15 cm/s the streaklines did not appear to be velocity-sensitive, although there was more turbulence at higher speeds. Abalones accepted incoming water only at specific sites along the shell (Figure 1: A, 1, 2). At the lower edge of the shell, water entered in a region 1~3 cm to the left of the left cephalic tentacle. It appeared that the animals could control the entrance at this region by waving the left cephalic tentacle. Dye released in the region between the two cephalic tentacles sometimes traveled along the edge of the shell, over the left tentacle and eye, and into the mantle cavity. At other times under the same flow conditions it was deflected by the tentacles and passed downstream alongside the shell. The two most anterior openings of the shell also served as sites of water intake. Depending upon the position of the mantle tentacles, dye released upstream, above, or be- side these openings entered the mantle cavity. If the an- terior opening was incomplete, resembling a fold at the Page 20 The Veliger, Vol. 26, No. 1 edge of the shell, dye released just below its dorsal edge proceeded into the mantle cavity, while dye released at other points within the same region traveled up and out away from the shell, as mentioned above. Only the two or three most posterior openings on the top of the shell served as the animals’ exhalent passages from the mantle cavity. In all cases the penultimate open- ing (4 in Figure 1), which ordinarily has no mantle ten- tacle, was the chief channel for wastes and exhalent water flow. The most posterior opening was often partially sealed and always had a tentacle, both of which reduced its out- put volume and rate. The center-most opening, which usually had a mantle tentacle, tended to be weakly inhal- ent, although at times dye streams also exited from it. Dye streams usually exited from the posterior or from the right side of the exhalent openings and were laminar. Except at regions of exhalent and inhalent currents, dye released at the sides of animals standing in still water did not move. The inhalent and exhalent regions were at the same locations in animals in still water as they were in moving water. Dye released at the inhalent openings of dead, intact abalones facing upstream in a flow of about 10 cm/s en- tered the mantle cavity at the site of release and exited through one of the exhalent openings. Shells of the same animals oriented downstream had similar patterns of flow through them, although the rate of the induced flow was not as high. In either orientation, dye released near ex- halent openings did not enter the shell, but continued downstream. DISCUSSION Although earlier reports indicated that water enters the abalone mantle cavity only under the margin of the shell and mantle flap on either side of the head (STEPHENSON, 1924; CRrorTs, 1929; YONGE, 1947), it is well known now that in at least some species the first hole serves for water intake as well (e.g., ABBOTT & HADERLIE, 1980, p. 233). In Haliotis kamtschatkana the streaklines show that flow enters the mantle cavity in a restricted region to the left of the head and through the first, and sometimes second, shell opening. Only the posterior openings are exhalent, and they are the only channels by which water and ex- cretory products may exit the mantle cavity. The fact that the shell of a dead animal shows’ flow patterns through its openings similar to those of a live animal suggests that by its design, the animal may take advantage of an induced flow. Because of the shell shape, the exhalent openings lie in the region at which the streamlines over the shell are maximally compressed, leading to a reduction of pressure over the posterior open- ings relative to the anterior ones, which should cause the fluid to enter at the front and to exit at the rear (VOGEL, 1981). Location of openings relative to any wake formed and the geometry of the openings may also determine local pressure, but these aspects of flow were not investigated. The orientation of the openings may further enhance the induced flow. As openings form at the anterior edge of the shell, they face the oncoming flow when the animal is facing upstream, and thus encounter dynamic pressure, or the pressure created due to stopping fluid (VOGEL, 1981). As the shell grows and new openings are added, older openings rotate relative to a horizontal surface, be- coming more nearly parallel to flow, so that by the time they reach the central region of the shell and begin to function as exhalent openings they experience the reduced pressure described above. Thus as an animal grows, each opening performs first an inhalent and later an exhalent function. I noticed that on each animal there is one open- ing (or sometimes two) that may be both slightly inhalent near its anterior edge and exhalent near its posterior edge. Such transitional openings are located on the cusp of the shell and are oriented at an angle intermediate to openings that are strongly inhalent or exhalent. It may be that there is some angle at which an opening may no longer serve as an entrance to, and another at which it may begin to serve as an exit from, the mantle cavity. It is also possible that the cephalic and mantle tentacles may valve the open- ings and help control the direction and rate of flow. Since induced flow may occur whether animals are fac- ing upstream or downstream, they may need only be par- allel to the general direction of surge to take advantage of it. This would make induced flow useful in regions of surge where flow reversal is common. It would be valuable to know if abalones orient to flow in their natural habitat and to what degree orientation is necessary for utilizing induced flow. Traditionally, workers have believed that the openings in an abalone’s shell evolved primarily as a way to avoid mixing fresh inhalent water with waste-laden exhalent water. I have shown that in addition to eliminating this problem, the shell determines the path by which water will enter, as well as exit, the mantle cavity. Whether or not this design increases the efficiency of moving water through the mantle cavity will require a study of the en- ergetics of water movement. ACKNOWLEDGMENTS This work began as a project for a course in biomechanics taught by Dr. Steven Vogel and Dr. Michael LaBarbera at the University of Washington’s Friday Harbor Labo- ratories. I thank them for their generous supply of equip- ment and encouragement. I also wish to thank the Pacific Northwest Shell Club and Dr. Stephen Wainwright for their support, and Vogel, Wainwright, and Dr. H. Fred- erik Nijhout for commenting on the manuscript. LITERATURE CITED AppotTtT, D. P. & E. C. HADERLIE. 1980. Prosobranchia: ma- rine snails. Jn: R. H. Morris, D. P. Abbott & E. C. Had- erlie (eds.), Intertidal invertebrates of California. Stanford Univ. Press, Stanford, Calif. pp. 230-307. J. Voltzow, 1983 Crorts, D.R. 1929. Haliotis. Liverpool Mar. Biol. Com. Mem. 29:1-174. GarsTaNnG, W. 1928. The origin and evolution of larval forms. Brit. Assoc. Adv. Sci. Report of the 96th Meeting: pp. 77- 98. Murpock, G. R. & S. VOGEL. 1978. Hydrodynamic induction of water flow through a keyhole limpet (Gastropoda, Fis- surellidae). Comp. Biochem. Physiol. 61A:227-231. STEPHENSON, T. A. 1924. Notes on Haliotis tuberculata. J. Mar. Biol. Assoc. U.K. 13:480-495. Page 21 VoGEL, S. 1981. Life in moving fluids: the physical biology of flow. Willard Grant Press, Boston, Mass. 352 pp. VoGEL, S. & M. LABARBERA. 1978. Simple flow tanks for research and teaching. BioScience 28:638-643. YONGE, C.M. 1947. Pallial organs in aspidobranch gastropods and their evolution throughout the Mollusca. Phil. Trans. R. Soc. Lond. B 232:443-518. The Veliger 26(1):22-25 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 Shell Strength in Corbicula sp. (Bivalvia: Corbiculidae) from the Potomac River, Maryland VICTOR S. KENNEDY Horn Point Environmental Laboratories, University of Maryland, Cambridge, Maryland 21613 JAY A. BLUNDON Department of Zoology, University of Maryland, College Park, Maryland 20742 Abstract. Forces required to crack intact shells of the freshwater Asiatic clam, Corbicula sp., were determined. They are higher than those which crack the wedge clam, Rangia cuneata, a globose, strong- shelled, oligohaline bivalve. Asiatic clam shell shape and strength may explain published data on crayfish predation on this animal in Oklahoma and Tennessee, in which predation was successful only on clams less than 6 mm in size or those with holes in the shell. Strong shells may also help prevent damage during periods of high river flow or strong water movement. INTRODUCTION CLAMS OF THE ASIATIC genus Corbicula were apparently introduced to western North America early in this cen- tury, becoming widespread in the ensuing years (BRITTON & Morton, 1979; CounTs, 1981). Recently, laboratory experiments by CovicH ef al. (1981) revealed that two species of freshwater crayfish were able to feed only on specimens of Corbicula that were less than 6 mm in size or that had damaged (perforated) shells. The Asiatic clam is globose in shape and the shell is relatively thick, giving the impression of strength. MACKIE (1978) noted that the thickness of the shell of Corbicula fluminea was greater than the shells of 22 other species of sphaeriacean bivalves he investigated. In an earlier study of crustacean predation on estuarine bivalves, we examined shell strength of eight species of clams (BLUNDON & KENNEDY, 1982). Here we present measurements of shell strength of the Asiatic clam, com- pare them with the strength of the estuarine bivalves, and relate the results to the findings of CovicH et al. (1981) concerning crayfish predation on Corbicula. TAXONOMY oF EXPERIMENTAL ANIMALS There has been much confusion associated with the tax- onomy of Corbicula in North America, with BRITTON & Mor Ton (1979) having declared the species to be Corbic- ula fluminea. However, HILLIS & PATTON (1982) have presented evidence that two species of Corbicula (a “white form” and a “purple form’) are present in the Brazos River, Texas. The specimens we tested in this report re- sembled the “white form” in color of nacre. Our speci- mens were collected from the Potomac River at Whites Ferry, Maryland (approximately 39°09’N; 77°31’W) in shallow water close to the river bank, a habitat in which HILuis & PATTON (1982) found the “white form” to pre- dominate. However, the mean number of growth rings (“annuli”) for our sample was less than for the Texas sample and, when number of annuli was plotted against shell mass for each clam, all our values fell below the “envelope” surrounding the values that HILLIs & PATTON (1982) derived (their Figure 1) for the “white form” in Texas. V.S. Kennedy & J. A. Blundon, 1983 Page 23 400 300 200 o Cc . ro) Corbicula sp. eel logY=1.96 log X-0.30 c uJ 2 86 3 log Y= 2.31 Le Rangia cuneata log Y= 2.42 log X -1.73 —~> 10 | 5 10 20 30 40 50 LENGTH (mm) Figure 1 Predictive regression ( ) and geometric mean functional regressions (----- ) for shell strength of Asiatic clam (Corbicula sp.) and wedge clam (Rangia cuneata) from Maryland. Wedge clam data are from BLUNDON & KENNEDY (1982). We agree with HILLIs & PATTON (1982) that the as- signment of a species name to populations of Corbicula is unwarranted until conclusive taxonomic studies are per- formed; thus we have called our bivalves Corbicula sp. in this report. In referring to the reported work of others, we have kept the species names they used. Specimens from our study population have been deposited with the Smith- sonian Institution’s National Museum of Natural History (USNM 804414). METHODS To test for shell strength, we used an Instron testing ma- chine, an industrial instrument that measures compression applied to a surface (BLUNDON & KENNEDY, 1982). Clams were crushed with a steel bar, 11 mm in diameter, which moved vertically downward at a velocity of 4 mm/s. Clams were crushed in the umbo region, parallel to the dorso- ventral axis. A chart recorder was used to record force (in newtons) required to crack the clam shell. A 10-newton weight was used to calibrate the Instron before and during the experiment. Clams for crushing were collected from Whites Ferry and were crushed immediately upon return to the laboratory (within 2 h of collection). RESULTS Initially, log,, force (Y) was regressed on log,, length (X), length being the maximum anterior—posterior axis in mm, and a regression line (log Y = 1.96 log X — 0.30) was fitted (Figure 1). The coefficient of determination, R?, was equal to 0.73 (n = 70). Because the measurements of force and size are subject to error of measurement, a geometric mean estimate of the functional regression may be a more appropriate linear regression (RICKER, 1973). The resul- tant equation (Figure 1) is: log Y = 2.31 log X — 0.76. This curve is significantly different from zero (P < 0.001). As a comparison with these data, the geometric mean Page 24 regression for Rangia cuneata, an oligohaline bivalve res- ident in Chesapeake Bay, is presented in Figure 1. Rangia cuneata, like the Asiatic clam, is a globose bivalve with a thick shell and closely fitting valves. It was the strongest bivalve we tested in our survey of shell strength of eight estuarine bivalves (BLUNDON & KENNEDY, 1982). How- ever, the Asiatic clam had a stronger shell than did R. cuneata (Figure 1). The slopes of the geometric mean regression for the two species were not significantly dif- ferent (P > 0.05), according to the test statistic of CLARKE (1980). However, the elevations of the two curves were significantly different (P < 0.001), as determined by Ho- telling’s T? (MorRISON, 1967). DISCUSSION As noted earlier, MACKIE (1978) found that the shell of Corbicula fluminea was the thickest of the shells of 23 species of sphaeriacean clams he studied. Neither MACKIE (1978) nor CoUNTs & PREZANT (1982) present evidence that the shell of Corbicula fluminea is unusual in its struc- tural material or in the arrangement of that material. In addition to shell thickness, the globose shape of the Asiatic clam, like Rangia cuneata, is probably an important reason for the crushing resistance being so high. CovicH et al. (1981) noted that freshwater crayfish, Procambarus clarkiu, attacked the edge of the shell of Asiat- ic clams with their mandibles. Repeated chipping of the shell led to eventual penetration. Such chipping away at shell edges should allow a relatively weak predator to open a strong-shelled bivalve. This chipping method was successful only with clams less than 6 mm long. Asiatic clams greater than 6 mm were successfully preyed upon by the crayfish Cambarus bartoni if the clams had suffered damage, such as perforations in the shell, which allowed the crayfish to reach their first walking leg into the soft clam body. BROWN et al. (1979) found that a 33.7-g specimen of Procambarus clarku could exert an average force of 9.9 newtons in the region of the base of the chelipeds, with force decreasing to 3.4 newtons near the tip of the che- lipeds. The P. clarkit used by Covicu et al. (1981) ranged in size from 21.0-34.6 g, averaging 27.8 g. Using our geometric mean regression equation, we find that a 4-mm and 6-mm long Asiatic clam (respectively, the minimum and maximum size of undamaged prey that P. clarki opened, according to COVICH et al., 1981) have an average shell strength of about 4 newtons (4-mm clam) to 11 new- tons (6-mm clam). Thus, if our data are transferable to southern clams, the crayfish used by CovIcH et al. (1981) may not have been able to crush the Asiatic clams greater than 6 mm long, even if they had used their chelipeds in a crushing attempt (Covich et al. do not report any at- tempts by P. clarki to use their chelae to crush shells). The Veliger; VoljZoyiNowm Theoretically, the larger crayfish (e.g., 34.6 g) could crush the smallest (4 mm) clams available, assuming that the chelae could grip the globose shell appropriately. With regard to other sources of shell damage that might leave Asiatic clams susceptible to crayfish predation, our clams were collected from a substrate of gravel and peb- bles covered with silt, with cobble stones and boulders also present. Fast river flow during floods (e.g., in spring) might cause substrate movement, with tumbling of clams or rocks and with grinding and pressure on shells. Strong shell structure would seem a useful protective measure under such conditions. CovicH et al. (1981) found damaged (perforated) shells in a rocky region of variable water flow below a dam. We have not noted much broken or damaged shell in our monthly surveys of Asiatic clams in our col- lecting area; most dead shell has consisted of intact valves. We conclude that the thickness and globose shape of the shell of the Asiatic clam, which probably accounts for its considerable strength, should provide protection from predator crushing attack, especially for larger clams. This strength may also protect the clams in situations where they, or rocks, are being tumbled about in fast-flowing waters. ACKNOWLEDGMENTS Partial support for this research was provided by Power Plant Siting Program Contract P87-82-04 from the Maryland Department of Natural Resources. We ac- knowledge the assistance of Laurie Van Heukelem. Deb- orah Kennedy drew Figure 1. Contribution No. 1378HPEL of the Center for Environmental and Estua- rine Studies. LITERATURE CITED BLUNDON, J. A. & V. S. KENNEDY. 1982. Mechanical and behavioral aspects of blue crab (Callinectes sapidus Rathbun) predation on Chesapeake Bay bivalves. J. Exp. Mar. Biol. Ecol. 65:47-65. BRITTON, J. C. & B. Morton. 1979. Corbicula in North America: the evidence reviewed and evaluated. Jn: J. C. Britton (ed.), Proceedings, First International Corbicula Symposium. pp. 249-287. Brown, S. C., S. R. Cassuto & R. W. Loos. 1979. Biome- chanics of chelipeds in some decapod crustaceans. J. Zool. (Lond.) 188:143-159. CLARKE, M. R. B. 1980. The reduced major axis of a bivariate sample. Biometrika 67:441-446. Counts, C. L. 1981. Corbicula fluminea (Miiller) on the Del- marva Peninsula. Veliger 24:187-188. Counts, C. L. & R. S. PREZANT. 1982. Shell microstructure of Corbicula fluminea (Bivalvia:Corbiculidae). Nautilus 96: 25-30. Covicu, A. P., L. L. DyE & J. S. Mattice. 1981. Crayfish predation on Corbicula under laboratory conditions. Am. Mid]. Nat. 105:181-188. Hiuis, D. M. & J. C. Patron. 1982. Morphological and electrophoretic evidence for two species of Corbicula (Bival- V.S. Kennedy & J. A. Blundon, 1983 Page 25 via: Corbiculidae) in North America. Am. Midl. Nat. 108: Morrison, D. F. 1967. Multivariate statistical methods. 75-80. McGraw-Hill, New York. 338 pp. Mackig, G. L. 1978. Shell structure in freshwater Sphaerea- Ricker, W. E. 1973. Linear regressions in fishery research. cea (Bivalvia:Heterodonta). Can. J. Zool. 56:1-6. J. Fish. Res. Board Can. 30:409-434. The Veliger 26(1):26-29 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 Relationship Between Beak Morphometrics and Live Wet Weight of the Giant Pacific Octopus, Octopus dofleint martini (Wiilker) by SHAWN M. C. ROBINSON anp E. BRIAN HARTWICK Department of Biological Sciences, Simon Fraser University, Burnaby, British Columbia, Canada V5A 1S6 Abstract. Twelve separate beak measurements were taken on 73 different sets of beaks and a relationship was established with the live weight of Octopus doflein: martin: (Wiilker). The pigment upper-lateral-wall length best predicted the weight of the octopus after taking the natural logarithm of each measurement. Pigment lengths were found to be slightly more accurate than the corresponding total lengths. We could find no set of measurements which would permit discrimination between sexes based on beak morphometrics. INTRODUCTION THE HORNY MANDIBLES, called beaks, found within the buccal mass of cephalopods have received attention for a number of different reasons. There has been considerable interest in their function during feeding (ALTMAN & Nixon, 1970), the organization of their movements (BOYLE et al., 1979a, b) and their potential in taxonomic work (CLARKE, 1962a, b; AKIMUSHKIN, 1965; MANGOLD & FIORONI, 1966; CLARKE & MacLEop, 1980; IVERSON & PINKAS, 1971). Studies also indicate the possibility of us- ing beak measurements to estimate cephalopod body size (CLARKE, 1962b; Nixon, 1969, 1973) and in the case of the squid Illex illecebrosus to differentiate between sexes (MERCER ef al., 1980). The ability to identify cephalopods by their beaks and to estimate their size from measurements of these hard parts provides considerable opportunity to extract infor- mation from stomach analyses of marine predators. Such information includes indications of predator migratory patterns (CLARKE & STEVENS, 1974), predator diet (see, for example, PITCHER, 1981; RANDALL et al., 1981), ceph- alopod distribution (CLARKE, 1962c), and the relative im- portance of cephalopods in the ecology of the oceans (CLARKE, 1977). Clarke’s report of 18,000 beaks in the stomach of one sperm whale gives some impression of their importance to these marine mammals. Much of the work on beaks has centered on various species of squids, although NIxon (1969, 1973) has car- ried out studies on Octopus vulgaris beaks. No information is available on beaks of the giant Pacific octopus, Octopus dofleini, although a brief description is given by WINKLER & ASHLEY (1954) and PICKFORD (1964). Pickford de- scribes possible growth lines on beaks and records rostral length indices for this species. Octopus dofleint martin: is an abundant, fast-growing form inhabiting the coastal waters of British Columbia and extending as far south as California. Its current sta- tus, as a candidate for fisheries development (HARTWICK et al., 1978), means that information on its distribution, mortality and predator-prey size relationships is desirable. Since examination of beaks may provide such information, an attempt was made to measure a series of beak dimen- sions and to relate these to body size and sex. MATERIALS anp METHODS During May, 1981, to February, 1982, 73 animals (15 males, 58 females) were captured in Clayoquot Sound on the west coast of Vancouver Island, British Columbia. SCUBA was used to collect the octopuses, and the first animals encountered were captured and taken to a vessel at the surface. After draining the mantle cavities, live wet weights were taken using spring scales sensitive to 0.25 kg for animals over 3 kg and 0.025 kg for animals less than 3 kg. The entire buccal mass was then removed from the animal and stored in 10% formalin in seawater. Later, beaks were gently excised from the surrounding muscle S. M. C. Robinson & E. B. Hartwick, 1983 Page 27 UPPER BEAK 1-UHL 2-PUCL 3-TUCL 4-PULWL 5- URW LOWER BEAK 6-LHL 7-PLWL 8-TLWL 9-TSLWL 10-PLCL 11-TLCL 12-LRW Figure 1 Measurements taken on upper and lower beaks of Octopus do- fleini. tissues and measured, using Vernier calipers, to the near- est 0.05 mm. Twelve measurements per beak were taken using terminology modified from CLARKE (1962b). As the octopus grows, new chitin is added to the edge of the wings, crest, and lateral wall (NIxon, 1969). This leaves a translucent margin on the edges where it has not become dark and horny. The interface of this translucent area with the pigment is used for the measurement of the pig- ment length. The measurements taken were: pigment up- per-crest length (PUCL), pigment lower-crest length (PLCL), pigment upper-lateral-wall length (PULWL), pigment lower-wing length (PLWL), total upper-crest length (TUCL), total lower-crest length (TLCL), total lower-wing length (TLWL), total standard-lower-wing length (TSLWL), upper hood length (UHL), lower hood length (LHL), upper rostral width (URW), and lower rostral width (LRW) (Figure 1). Analysis of the data was carried out using least-squares regression and discriminant analysis, both statistical com- puter package programs (MIDAS). Correlation tests were taken from SOKAL & ROHLF (1969, p. 521). RESULTS anp DISCUSSION In this study, we attempted to predict the live wet weight of the octopus from the beak measurements. This was chosen over other measurements, such as the dorsal man- tle length, since body weight seems to be a more accurate measurement of size than length in live material (NIXON, 1968). The weight range of animals measured was from Table 1 Regression statistics of the natural logarithm of beak mea- surements (mm) versus the natural logarithm of live wet weight (kg) for Octopus dofleini (n = 73). (r = correlation coefhicient; SE = standard error of regression; m = slope; b = y-intercept.) Measurement Toa SE m 19) In PULWL 0.970 0.044 0.274 2.674 In PLWL 0.967 0.045 0.265 2.969 In PUCL 0.965 0.048 0.277 B22 In PLCL 0.959 0.050 0.268 DO In TLWL 0.946 0.053 0.245 3.074 In TSLWL 0.938 0.058 0.248 2.926 In TUCL 0.934 0.060 0.247 BATS In TLCL 0.927 0.068 0.264 2.832 In UHL 0.921 0.061 0.228 2.382 In LHL 0.909 0.074 0.253 1.954 In URW 0.885 0.082 0.244 1.216 In LRW 0.784 E13) 0.261 1.059 * All r’s are highly significant at the 99% confidence level. 0.95 to 22.75 kg with a mean weight of 7.64 kg. There was a non-linear increasing function of beak measure- ments on wet weight so the natural logarithms of both axes were taken in order to obtain a linear function. A transformation to the cube root of the live wet weight was also attempted, to test whether a better fit of the data to a line was obtained, but the fit was not as good as taking the natural logarithm of both axes. The relationships between the twelve measurements on the upper and lower beaks may be seen in Table 1. It is apparent from this table that all the beak measurements are good predictors of octopus weight with the correlation coefficients (r) ranging from 0.784 to 0.970. Although the best measurement is the pigment upper-lateral-wall length (PULWL), shown in Figure 2, there is really very little difference among the first three measurements. All of the beak measurements in Table 1 have been grouped from the highest correlation coefficient to the lowest and this resulting hierarchy allows one to choose the best mea- surement that is available. A hierarchy of possible mea- Table 2 Significance of the correlation coefficient values between corresponding pigment and total length beak measure- ments (Fisher’s z transformation). Comparison 1 vs. 2 n z(1) z(2) IP << In PUCL In TUCL if 2.0191 1.6902 0.052 In PLCL In TLCL 73 1.9345 1.6376 0.080 In PLWL In TLWL 73 2.0484 1.7943 0.133 Page 28 3.6 An A A 3.45 Ae 3.2- A In Pigment Upper Lateral Wall Length (mm) 2.87 OA A 205 = Th Rae sa = 0 1 2 3 4 In Octopus Weight (Kg) Figure 2 The relationship between pigment upper-lateral-wall length and live wet weight of Octopus dofleini (natural logarithms of both axes). surements would be advantageous if only one of the beaks was available and/or it was damaged in some way neces- sitating an alternate measurement. Our results compare favorably with those of NIXON (1973) with Octopus vulgaris. She found that after boiling the beaks in a 5% potassium hydroxide solution to remove the surrounding muscle tissue, there was a high degree of correlation (r = 0.97 with n = 77) between the logarithm of live wet weight and the logarithm of crest length of the upper beak. It is interesting to note that the correlation coefficient for the pigment upper-crest length (PUCL) in our study is essentially the same as Nixon’s with approx- imately the same number of samples taken, and that an upper-beak measurement is generally more accurate in predicting weight than its corresponding one on the lower beak. In our study, essentially two types of measurements were made: pigment lengths and total lengths, although four measurements—UHL, LHL, URW, and LRW— were not readily assignable to either of these. As can be seen from Table 1, there is a distinct grouping of mea- surements with the pigment lengths in every case having higher correlation coefficients than the total lengths. To The Veliger, Vol. 26, No. 1 determine whether a significant difference existed between these two types of measurements, Fisher’s z-transforma- tion test was carried out (Table 2). Three sets of corre- sponding measurements were used, and the data show that there is only a slightly significant difference between them (ranging from P < 0.05 to P < 0.13). Despite the low levels of significance, the use of pigment lengths over total lengths is warranted by the fact that often the delicate translucent edge will be destroyed if the beak is not care- fully removed from the muscle tissue of the buccal mass. Although MERCER et al. (1980) were able to differen- tiate between sexes of the squid Illex illecebrosus using beak morphometrics in a discriminant analysis, our at- tempts to find a set of beak measurements in Octopus dofleint which would separate sexes using this technique were unsuccessful. Although this may have been due, in part, to a small sample size of males, there was a large degree of overlap between the generated male and female discriminant values. ACKNOWLEDGMENTS This work was supported by a National Sciences and Engineering Research Council grant and a Canada Fish- eries and Oceans grant to E. B. Hartwick. We gratefully acknowledge the help of D. Trotter, L. Tulloch, M. Walsh, and G. Cox for their valuable assistance in the field and to R. Lockhart for his advice with the analysis. LITERATURE CITED AKIMUSHKIN, I. I. 1965. Cephalopods of the seas of the USSR. Translation of 1963 Russian edition by A. Mercado, Je- rusalem. 223 pp. ALTMAN, J. S. & M. Nixon. 1970. Use of beaks and radula by Octopus vulgaris in feeding. J. Zool. (Lond.) 161:25-38. BoyLeE, P. R., K. MANGOLD & D. FROESCH. 1979a. The man- dibular movements of Octopus vulgaris. J. Zool. (Lond.) 188: 53-67. Boy.e, P. R., K. MANGOLD & D. FROESCH. 1979b. The or- ganization of beak movements in Octopus. Malacologia 18: 423-430. CLARKE, M. R. 1962a. Significance of cephalopod beaks. Na- ture 193:560-561. CLARKE, M.R. 1962b. The identification of cephalopod “beaks” and the relationship between beak size and total body weight. Bull. Br. Mus. (Nat. Hist.) 8:419-480. CLARKE, M. R. 1962c. The identification of cephalopod beaks and their significance in systematic and ecological studies of squids and their predators. Rep. Challenger Soc. 3(14):43- 44. CiarkE, M. R. 1977. Beaks, nets and numbers. Symp. Zool. Soc. Lond. 38:89-126. CriarkE, M. R. & N. MacLeop. 1980. Cephalopod remains from sperm whales caught off western Canada. Mar. Biol. 59(4):241. CLARKE, M. R. & J. D. STEVENS. 1974. Cephalopods, blue sharks and migration. J. Mar. Biol. Assoc. U.K. 54:949- 957. Hartwick, E. B., P. A. BREEN & L. TULLOCH. 1978. A re- moval experiment with Octopus dofleini (Wiilker). J. Fish. Res. Board Can. 35(11):1492-1495. S. M. C. Robinson & E. B. Hartwick, 1983 Page 29 Iverson, I. L. K. & L. Pinkas. 1971. A pictorial guide to beaks of certain eastern Pacific cephalopods. Calif. Dep. Fish. Game, Fish Bull. 152:83-105. MANGOLD, K. & P. FIORONI. 1966. Morphologie et biométrie des mandibules de quelques céphalopodes Méditerranéens. Vie et Milieu 17:1139-1196. Mercer, M. C., R. K. Misra & G. V. Hurvey. 1980. Sex determination of the ommastrephid squid Illex ilecebrosus using beak morphometrics. Can. J. Fish. Aquat. Sci. 37: 283-286. Nixon, M. 1968. Growth of radula in Octopus vulgaris. J. Physiol. (Lond.) 196:28P-30P. Nixon, M. 1969. Growth of the beak and radula of Octopus vulgaris. J. Zool. (Lond.) 159:363-379. Nixon, M. 1973. Beak and radula growth in Octopus vulgaris. J. Zool. (Lond.) 170:451-462. PICKFORD, G. E. 1964. Octopus dofleini (Wiilker). Bull. Bing- ham Oceanogr. Coll. 19(1):1-70. PircHer, K. W. 1981. Prey of the Steller sea lion, Eumetopias jubatus, in the Gulf of Alaska. U.S. Nat. Mar. Fish. Serv. Fish. Bull. 79(3):467-472. RANDALL, R. M., B. M. RANDALL & E. W. KLINGELHOEFFER. 1981. Species diversity and size ranges of cephalopods in the diet of jackass penguins from Algoa Bay, South Africa. S. Afr. J. Zool. 16(3):163-166. SOKAL, R. R. & F. J. ROHLF. 1969. Biometry. W. H. Free- man, San Francisco. 776 pp. WINKLER, L. R. & L. H. ASHLEY. 1954. The anatomy of the common octopus of northern Washington. Walla Walla Coll. Publ. Biol. Sci. Number 10 (November 10, 1954). The Veliger 26(1):30-36 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 Studies on the Reproductive Biology of Some Prosobranchs from the Coast of Pakistan Bordering the Northern Arabian Sea. II]. Egg Capsules and Larvae of Four Species of Thais SOHAIL BARKATI anp MUZAMMIL AHMED Institute of Marine Biology, University of Karachi, Karachi-32, Pakistan Abstract. Egg masses of four species of Thais, namely, 7. rudolphi, T. carinifera, T. bufo, and T. tissoti, are reported from the coast of Pakistan for the first time. Based on the incidence of their egg capsules, the respective breeding seasons of the four species are: June to September, February to November, April to August, and March to August. The four species deposit benthic egg capsules from which free-swimming veligers are hatched. A comparative account of numbers and dimensions of capsules, number and sizes of eggs, and developmental periods of shelled larvae is presented. INTRODUCTION THE EGG MASSES and capsules of prosobranch mollusks, apart from the fact that they are of scientific interest in themselves, yield information about the breeding seasons of the parent species, may prove of assistance in the precise identification of closely related species, and provide larval stages and juveniles should there be a need for their com- mercial rearing. Although the literature in this field is vast, only the following are the major publications on the gastropods of the Arabian Sea: (THORSON (1940) from the Persian Gulf; NATARAJAN (1957) from the Gulf of Ma- nnar and Palk Bay; GOHAR & EISAWAY (1962, 1967a, b) from the Red Sea; and DEsal (1962), GANAPATI & SASTRY (1973), RAMAMOORTHY & NATARAJAN (1973), and Ka- SINATHAN et al. (1974) from different parts of the Indian coast. No information is available on the egg masses and larval development of marine gastropods from the coast of Pakistan bordering the northern Arabian Sea. It was in view of this paucity of information that a project was initiated in 1976 at the Institute of Marine Biology, Uni- versity of Karachi, to study the same. Observations on the juveniles of the gastropod Planaxis sulcatus from the coast of Karachi have already been published (BARKATI & AHMED, 1982). The present report on four species of Thazs is the second in the series. MATERIAL anp METHODS In order to collect egg masses and capsules of local gas- tropods, regular biweekly visits were made during the period June, 1976, to July, 1977, to Buleji, which is a rocky and partly sandy-cum-muddy beach about 18 km northwest of Karachi. Egg capsules were also collected rather irregularly from some other beaches in the Karachi area, namely, Manora Island, Keamari Seawall, Paradise Point, and Korangi Creek, and from Gawader and Jiwani on the Mekran coast, about 600 km northwest of Karachi. Some egg capsules had been collected prior to June, 1976, and others after July, 1977. Except for a couple of rainy months in summer the salinity at all these sites is high, ranging from 36-40%. In most of the cases the spawning female snails were present near their egg masses. It was difficult to ascertain the number of egg capsules spawned by individual snails since the capsules are laid together. During the study, therefore, only representative samples of egg capsules were collected although some of these con- sisted of the entire lots of capsules spawned by individual snails. Attempts to spawn specimens of the different species of Thais in the laboratory in limited quantities of water did not succeed. The egg capsules, and the larvae hatching from these, were maintained in filtered and aerated sea S. Barkati & M. Ahmed, 1983 Page 31 water in 1600-ml glass bowls. They were fixed in 5% formalin and then preserved in 70% ethanol for subse- quent study. Measurements of the size of capsules, eggs, and larvae were made with an ocular micrometer on a microscope. Illustrations were prepared with a camera lucida. OBSERVATIONS Breeding Season The four species of Thais investigated occur near the low and mid-tidal zones of the rocky beaches of the coast of Pakistan. The number of egg masses of four species collected from different localities is given in Table 1. It is apparent from this table that 7. rudolphi (Lamarck) spawns from June to September, 7. tissot: (Petit) from March to August, and 7. bufo (Lamarck) from April to August. An egg mass of 7. carinifera (Lamarck) was taken at station 45 (63°46/N latitude) along the Mekran coast on March 1, 1977, in a purse seine haul of the Norwegian Fisheries Research Vessel “Dr. Fridtjof Nansen” (egg capsules of this species in THORSON’s [1940] collection from the Persian Gulf were obtained from a depth of 11- 22 m). Since some of the capsules in this egg mass were of purple color they must have been spawned in February. Recently, an egg mass of 7. carinifera was also collected from Port Qasim, about 30 km southeast of Karachi. The spawning season of 7. carinifera on the coast of Pakistan, therefore, seems to extend from February to November. Characteristics of Egg Capsules, Eggs, and Larvae Egg capsules of 7. rudolphi (Figure 1A, B), like those of other species of Thais, are creamy white in color when deposited and become dark gray with the advancement of embryonic development. A change in color from yellow to gray was also observed by NATARAJAN (1957) in 7. bufo. The capsules are deposited in clusters of several irregular, short rows imparting a circular shape to the egg mass in general. There is always one layer of capsules the basal plates of which unite to form a common basal membrane that is firmly glued to the substratum. Each capsule has translucent, tough leathery walls. An aperture for the re- lease of larvae, 0.32 mm in diameter, is present slightly off center on the flattened, apical surface of each capsule and is covered with a transparent membrane which rup- tures at the time of larval hatching. Dimensions of cap- sules, eggs, and larvae of this species are given in Table 2. The larvae just after hatching (Figure 1C, D) possessed shell lengths of 211 to 251 um and widths of 171 to 182 um. They swam actively and gathered near the walls of the glass bowls. Their shells consisted of one and a half whorls. They measured 353 to 376 wm in height and 274 to 285 wm in width 36 h after hatching. The egg mass of 7. carinifera consists of closely spaced, stalkless, somewhat curved and thin tubular capsules that Table 1 The incidence of egg masses of four species of Thais on the coast of Pakistan. Exposure of shore abbreviated as: Exposed, E; Protected, P; Semi-exposed, SE; Subtidal, ST. Expo- Egg Locality sure Date Year masses T. rudolphi Manora Island E June 9 1970 Many Buleji E June 15 1976 1 Manora Island E June 16 1976 2 Buleji IE; July 2 1976 2 Buleji E July 3 1977 4 Buleji E August 13 1976 1 Paradise Point E August 26 1975 1 Buleji E September 8 1975 1 Buleji E September 9 1976 2 Paradise Point E September 9 1976 1 T. carinifera Mekran Coast ST March 1 1977 1 Jiwani SE April 29 1979 g Korangi Creek P April 30 1978 1 Native Jetty le May 1972 Many Buleji E June 14 1980 1 Keamari Seawall 1? August 1 1976 3 Port Qasim P November 1 1982 1 T. tissot Sandspit Bridge P March 16 1979 1 Jiwani SE April 29 1979 2 Korangi Creek P April 30 1978 1 Buleji SE June 4 1980 1 Manora Island E June 9 1970 Many Keamari Seawall rR August 1 1976 1 Korangi Creek 12 August 21 1977 Many T. bufo Jiwani SE April 29 1979 1 Keamari Seawall P May 1975 3 Keamari Seawall P May 18 1976 1 Manora Island E June 9 1970 Many Keamari Seawall P June 13 1975 1 Manora Island E June 13 1979 1 Keamari Seawall P August 1 1976 1 have very smooth translucent walls without any ridges (Figure 2A, B). They are of uniform width but taper at the apex to form a nipple-like structure. The apex func- tions as an operculum which is shed at the time of hatch- ing so as to provide an exit for the larvae. This aperture is 0.4 mm in diameter. Dimensions of capsules, eggs, and larvae of this species are given in Table 2. The larvae 12 h after hatching (Figure 2C, D) measured 388 X 295 um. Their shell walls were smooth and transparent and con- sisted of one and a half whorls. They attained a size of 401 X 301 wm (Figure 2E) after 36 h of development. Very few larvae survived in the laboratory for 60 h with- out food. The egg mass of 7. carinifera was described Page 32 The Veliger, Vol. 26, No. 1 Figure 1 Thais rudolphi: A. dorsolateral view of a portion of an egg mass; B. dorsolateral view of an egg capsule; C. ventral view of a larval shell just after hatching; D. dorsal view of the same; E. ventral view of 36-h larval shell. earlier by ANNANDALE & Kemp (1916) from Chilka Lake, India, and by THORSON (1940) from the Red Sea. Na- TARAJAN (1957) also described an egg mass that closely fits the above description but he was not able to assign it to any species. Egg capsules of 7. tissot: are stalkless (Figure 3A, B) and attach themselves to the substratum by their basal plates, which unite to form a basal membrane. The cap- sules are smaller than those of 7. carinifera. They are cylindrical, somewhat broader in the middle, taper at both ends, and curve on one side. Their walls are smooth and without ridges. A preformed aperture for the release of larvae does not exist but the apex of the capsule sheds off at the time of hatching. The part shed measures 0:72 X 0.32 mm across. Dimensions of capsules, eggs, and larvae of this species are given in Table 2. The larval shell (Fig- ure 3C) soon after hatching consists of one and a half whorls. Changes in the size of the larval shell of this species with time are shown in Table 3. The larvae thrived in the laboratory without food for about nine days but showed very little increase in shell dimensions after 84 h. The maximum shell dimensions recorded were: height 467 pm and width 456 um. The capsules of 7. bufo (Figure 4A, B, C) have smooth and very thick walls, are tubular in shape, and have long thin stalks which unite at the base and are glued to the substratum. The proximal end of the capsules is swollen and carries a rounded aperture for the exit of larvae. The aperture measures 0.49 to 0.58 mm in diameter and is covered by a thin membrane. The larvae (Figure 4D, E) hatch from the capsules as free-swimming veligers. Their dimensions appear in Table 2. The egg mass of 7. bufo was earlier described from south Indian waters by Gravely (1942, in NATARAJAN, 1957), Chari (1950, in NATARAJAN, 1957) and by Na- TARAJAN (1957) and the above account describes it closely. An egg mass was collected from the bottom of a crevice of a steep rocky ledge at Paradise Point that might be from a fifth species of Thais (or another muricid). This ball-like egg mass measured 2.4 X 2.2 cm across and con- sisted of two layers of low rounded and irregularly shaped capsules which were stalkless, with their basal plates united to form a common membrane. DISCUSSION Observations made during the present study show that three of the four species of 7hais examined, namely, 7. bufo, T. tassoti, and T. carinifera, spawn for about six to seven months in the spring and summer during the period February to August, whereas 7. rudolphi has the shortest spawning season of four months, restricted to the summer months of June to September. There is evidence that spawning in at least two of these species, namely, 7. bufo S. Barkati & M. Ahmed, 1983 Page 33 Figure 2 Thais carinifera: A. dorsolateral view of a portion of an egg mass; B. lateral view of a single egg capsule; C. ventral view of veliger shell 12 h after hatching; D. dorsal view of the same; E. ventral view of larval shell 36 h after hatching; F. front view of the same. Table 2 A comparative account of the characteristics of egg capsules, eggs, and larvae of four species of Thavs. as averages with the range of values given in parentheses. Data are presented Characteristics T. rudolphi T. carinifera T. tissoti T. bufo Approximate size of females (cm) 6.8 4.6 32 5.8 (4.0-8.0) (3.5-6.0) (2.5-4.0) (4.0-7.0) Number of egg capsules per mass 170 247 257 80 (97-257) (220-268) (255-295) (65-92) Capsule height (mm) 4.6 91 4.2 7.9 (3.8-5.0) (8.0-10.5) (4.0-4.4) (7.2-9.0) Capsule width (mm) 533 13) 1.2 2.0 (4.0-6.1) (1.2-1.5) (1.1-1.2) (1.8-2.5) Number of eggs per capsule 1094 140 37 228 (813-1422) (120-160) (25-40) (183-268) Egg diameter (um) 142 230 215 252 (137-148) (216-250) (200-250) (198-350) Hatching time (days) UF 19 19 20 (16-18) (18-20) (18-20) (18-22) 8220x255 Larval size at hatching (Lx W) (um) 224 x 173 340 X 276 231 x 198 Page 34 The Veliger, Vol. 26, No. 1 Figure 3 Thats tissott: A. dorsolateral view of a portion of an egg mass; B. lateral view of a single egg capsule; C. ventral view of veliger shell taken out of a capsule; D. ventral view of 24-h veliger shell; E. front view of 84-h veliger shell; F. front view of larval shell 13 days after hatching. and 7. carinifera, may commence slightly earlier on the coast of Mekran than on the coast of Karachi. Be as it may, spawning in the four species of 7hazs seems to occur with the rising temperatures of spring and summer when a lowering of salinity may also occur due to the southwest monsoon rains (see AHMED, 1980). The study shows that 7. rudolphi produces the highest Table 3 Changes in the size of the larval shell of Thais tissott. Egg size 1s 200-225 um. Size (um) Just after hatch- ing 12h 36h 60h 84h 9d Length 231 281 313 338 385 460 Width 198 229 232 269 302 442 numbers of eggs per capsule, and 7. carinifera, T. bufo, and 7. tissote follow it in that order. Survival of the brood in the natural environment seems to be in the same pro- portion, since 7. rudolphi is the most abundant species of Thais on the coast of Karachi followed by 7. carinifera. Thais rudolphi has not been recorded from Gawader and Jiwani on the Mekran coast where 7. carinifera does occur (AHMED et al., in press). Thais rudolphi, the species with the highest fecundity, also has been found in the present study to have eggs and larvae of the smallest size among the four species. Larvae of the four species spend about 16-22 days within egg capsules but those of 7. rudolphi hatch the earliest. Species of Thais are known to display two types of larval development. The first is the indirect development in which pelagic planktotrophic or lecithotrophic veliger larvae hatch from egg capsules. The second is direct or non-pelagic development in which crawling miniature snails emerge from egg capsules and where a planktonic stage is missing. The four species of 7hais examined in the present study also possess indirect development, which S. Barkati & M. Ahmed, 1983 Page 35 Figure 4 Thais bufo: A. lateral view of an egg mass; B. and C. dorsal and ventral views of egg capsules; D. ventral view of a veliger shell taken out of a capsule just before hatching; E. front view of the same; F. tip of an egg capsule. is shown by a majority of the species of Thais studied so far in different parts of the world: for instance, 7. cari- nifera (THORSON, 1940); 7. fasciata (LEBOUR, 1945); 7. coronata (KNUDSEN, 1950); 7. bufo, T. tissotz, and three unidentified species of Thais from India (NATARAJAN, 1957); and 7. haemastoma (D’AsaArRo, 1970). Species that show direct development are 7. hippocastaneum (THORSON, 1940), 7. lapillus (see THORSON, 1940), 7. lamellosa, T. emarginata (=T. lima), and T. canaliculata (see AHMED & SPARKS, 1970), and 7. emarginata (LEBOEUF, 1972; SPIGHT, 1976). There are, however, instances where members of the same species of 7hais may behave differently under different environmental conditions. For instance, 7. hae- mastoma is believed to have pelagic larval development in Louisiana but direct development in the West Indies (see THORSON, 1950; NATARAJAN, 1957). Lyons & SPIGHT (1973) have considered the members of the Muricacea as sufficiently variable that geographically separated species may develop according to the local conditions. We wish to point out here that some workers feel that the above reference to direct development in the West Indian species of Thais is based on misinformation and should not be attributed to either Thorson or Natarajan. It is generally believed that species of prosobranch mol- lusks occurring in cold waters of higher latitudes show direct development but those occurring in warm tropical and subtropical latitudes possess pelagic indirect devel- opment (THORSON, 1950). However, MILEIKOvSKyY (1971) pointed out, as an exception to Thorson’s generalization, that pelagic development is not altogether absent in colder waters of higher latitudes and that an occasional species may display such a developmental pattern. Also as an exception can be mentioned the case of the gastropod Planaxis sulcatus which shows direct development on the coast of Karachi, a subtropical locality (BARKATI & AHMED, 1982). LITERATURE CITED AHMED, M. 1980. The breeding and recruitment of marine animals of the coast of Pakistan bordering the Arabian Sea. Proc. First Pakistan Cong. Zool: pp. 55-96. AHMED, M., H. N. Rizvi & M. MoazzaM. The distribution and abundance of intertidal organisms on some beaches of Mekran coast in Pakistan (northern Arabian Sea). Pakistan J. Zool. (in press). AHMED, M. & A. K. Sparks. 1970. A note on the chromosome Page 36 number and interrelationships in the marine gastropod ge- nus Thais of the United States Pacific coast. Veliger 12:293- 294. ANNANDALE, N. & S. Kemp. 1916. Fauna of the Chilka Lake. Mollusca, Gastropoda and Lamellibranchiata with an ac- count of the anatomy of the common Solen etc. Mem. Ind. Mus. Calcutta: pp. 328-366. BARKATI, S. & M. AHMED. 1982. Studies on the reproductive biology of some prosobranchs from the coast of Karachi (Pakistan) bordering the northern Arabian Sea. I. Obser- vations on Planaxis sulcatus (Born, 1780). Veliger 24(4): 355-358. D’Asaro, C. N. 1970. Egg capsules of prosobranch mollusks from South Florida and the Bahamas with notes on spawn- ing in the laboratory. Bull. Mar. Sci. 20:414-440. Desal, B. N. 1962. A preliminary note on the eggs and larvae of some marine molluscs of Bombay. Curr. Sci. (Bangalore) 4:158-159. GanapaTl, P. N. & S. R. Sastry. 1973. The spawn of a cymatid gastropod, Cymatium pileare. Curr. Sci. (Bangalore) 42(1):25. Gonar, H. A. F. & A. M. Elsaway. 1962. The egg masses and development of Trochus erythraeus from the Red Sea. Publ. Mar. Biol. Sta. Al-Ghardaqa, Egypt, No. 12:191- 203. Gonar, H. A. F. & A. M. Etsaway. 1967a. The egg masses and development of four taenioglossan prosobranchs from the Red Sea. Publ. Mar. Biol. Sta. Al-Ghardaqa, Egypt, No. 14:109-147. Gouar, H. A. F. & A. M. ElsAway. 1967b. The egg masses and development of five rachiglossan prosobranchs. Publ. Mar. Biol. Sta. Al-Ghardaqa, Egypt, No. 14:215-268. The Veliger, Vol. 26, No. 1 KASINATHAN, R., K. GOVINDAN & R. NATARAJAN. 1974. Notes on spawning and hatching of three species of Gastropoda. Malacol. Rev. 7:133-135. KNUDSEN, J. 1950. Egg capsules and development of some marine prosobranchs from tropical west Africa. Atl. Rep. 1: 85-130. LEBOEUF, R. 1972. Thais emarginata: description of the veliger and egg capsule. Veliger 14(2):205-211. Lrespour, M. V. 1945. The eggs and larvae of some proso- branchs from Bermuda. Proc. Zool. Soc. Lond. 114:462- 489. MILEIKovsky, S. A. 1971. Types of larval development in marine bottom invertebrates, their distribution and ecolog- ical significance: a re-evaluation. Mar. Biol. 10(3):193-213. Lyons, A. & T. M. SpicHT. 1973. Diversity of feeding mech- anisms among embryos of Pacific Northwest Thais. Veliger 16:189-194. NATARAJAN, A. V. 1957. Studies on the egg masses and larval development of some prosobranchs from the Gulf of Man- nar and the Palk Bay. Proc. Ind. Acad. Sci., Sect. B. 46: 170-228. RAMAMOORTHY, K. & R. NATARAJAN. 1973. Spawning in Te- lescopium telescoprum. Venus 31(4):157-159. SpiGHT, T. 1976. Hatching size and the distribution of nurse eggs among prosobranch embryos. Biol. Bull. 150:491-499. THORSON, G. 1940. Studies on the egg masses and larval de- velopment of Gastropoda from the Iranian Gulf. Danish Scientific Investigations in Iran, Part IJ:159-238. Ejner Munkegaard, Copenhagen. TuHorson, G. 1950. Reproductive and larval ecology of marine bottom invertebrates. Biol. Rev. 25:1-45. The Veliger 26(1):37-46 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 Comparison of Northern and Southern Populations of Epitonium tinctum (Carpenter, 1864) Onethe California Coast CAREY RESCH SMITH Santa Barbara Museum of Natural History, 2559 Puesta Del Sol Road, Santa Barbara, California 93105 AMY BREYER Department of Biology, Sonoma State University, Rohnert Park, California 94928 Abstract. Epitonium tinctum (Carpenter, 1864) was collected for several years from Bodega Head to the north and from Carpinteria Beach to the south of Point Conception, California. Northern populations contained longer shells than southern populations. Northern snails became sexually mature and changed sex at a larger size than southern snails. There were no distinguishable differences in the weight and width of specimens between northern and southern snails of similar length. Variation in shell shape (squat versus elongate) between populations did not exceed variation within each population. Shell length and the number of body whorls overlapped substantially between the two sites. The size and abundance of snails collected monthly fluctuated more throughout the year in the northern popu- lation than in the southern population. INTRODUCTION ALONG THE WEST coast of North America the wentletrap Epitonium (=Nitidiscala) tinctum is found intertidally in association with the aggregate or clonal anemone Antho- pleura elegantissima (Brandt, 1835). Epitonium tinctum ranges from Magdalena Bay, Baja California to Forrester Island, Alaska (DUSHANE, 1979). As protandric her- maphrodites, the snails mature first as malesyjthen, at a larger size they change sex and function as females (BULNHEIM, 1968; BREYER, 1982). Females produce a number of sand-encrusted egg cases which are strung to- gether with a strong elastic mucous thread attached to the substrate or to the shell. When exposed at low tide, snails bury themselves in the sand between anemones and are often aggregated around a cluster of egg cases. When sub- merged during high-tide periods the snails become active and feed on the expanded tentacles of the anemones (HocHBERG, 1971; SMITH, 1977; BREYER, 1982). In order to feed, the snails evert an acrembolic proboscis and either slip the distal end over a tentacle tip or attach it at some point along the length of the tentacle. Jaws and radula are used to sever a portion of the tentacle which is then ingested upon retraction of the proboscis (RESCH, 1972). The shell of Epztonium tinctum can grow to over 10 mm in length. It has three nuclear whorls and up to 8 post- nuclear whorls with about 12 thin axial costae continuous from whorl to whorl. STRONG (1941) examined shells from north and south of Point Conception, California, and not- ed that northern specimens were longer and appeared to be heavier and broader than southern individuals. He sug- gested the subspecific name of swbcoronatum for the north- ern form. Although DUSHANE (1979) described similar differences, she did not recognize the subspecies suwbcoro- natum because radulae from northern and southern snails were indistinguishable. Individuals of populations living in food-rich environ- ments or in warm climates may be expected to grow faster than those living in more severe environments. This faster Page 38 128° 126° 124° 122° —- DN 42° * Bodega Head / / ( 40° San 1 Francisco \ St 38° Point ‘ Conception \ SC Carpinteria Beach— Los \ e Angeles 36° 34° 7 A Figure 1 Map of California showing two study sites located north (Bodega Head) and south (Carpinteria Beach) of Point Conception. growth may lead to an altered size at maturity and per- haps an important difference in generation time. In the present work we document differences within and be- tween populations of Epitonium tinctum to the north (Bo- dega Head) and to the south (Carpinteria Beach) of Point Conception, California (Figure 1). Providing the major basis for comparison are shell parameters, which include: length, width, weight, and number of whorls and costae. Distinct population modes were computed from measure- ments of shell length. The mean minimum size at. onset of sexual maturity and sex change were calculated from lengths of juveniles, males, and females. Additionally, fluctuations in size and abundance were noted monthly from July, 1977, to July, 1979, for populations at Bodega Head, and from July, 1978, to June, 1980, for popula- tions at Carpinteria Beach. MATERIALS ann METHODS Collection Sites The northern collecting site was in Horseshoe Cove located on the exposed outer coast of Bodega Head, So- The Veliger, Vol) Zo3Noml 25] BE Bodega Head (n= 1692) [) Carpinteria Beach (n=840) no (eo) — = = fo) a ——_ — Percent of Total Number of Snails Collected o 4 1 2 3 4 5 6 7 8 9 ie) 4 Size Class (mm shell length) Figure 2 Comparison of E. tinctum shell length and the percentage of total number of snails collected from 1977 to 1980. noma County, California. In a protected area on the northern side of the cove, a 3-m-wide surge channel lo- cated about 10 m from shore is bordered on one side by a cliff. The channel is parallel to the cliff for about 15 m until its mouth opens toward the west. Wave disturbance in this channel is reduced, not only because waves break to the south, but also because it is protected by an exten- sive rock shelf. At this site clones of Anthopleura elegan- tissima cover extensive areas on the sides and bottom of the channel. In the summer and fall, when wave action is reduced, the sand between the anemones is fine grained and at least 10 mm deep. During the winter, when wave action is heaviest, the amount of fine sand between the anemones is reduced, disappears altogether, or is replaced by coarser sand. Snails were collected from the anemone beds on both sides of the channel. The rocky reef area at Carpinteria Beach State Park, Santa Barbara County, California was the southern col- lecting site. Carpinteria Beach, like much of the southern California coastline south of Point Conception, is char- acterized by sandy beaches interrupted by intermittent rocky outcrops or headlands. The Carpinteria Beach site consists of a sandy beach extending out from the cliff about 100 m to where the first rock outcrop begins. In the mid-intertidal region of the rocky reef both clonal and solitary Anthopleura elegantissima are present. Epitonium tinctum is most often found in protected areas on these rocks in sand which accumulates within pockets between clonal anemones and beneath the algal cover of Ulva lobata (Kutzing, 1849) and Gigartina canaliculata (Harvey, 1841). C. R. Smith & A. Breyer, 1983 Page 39 Bodega Head Number of Individuals 2 3 4 5 Shell Length (mm) 7 8 9 10 11 12 [] Juvenile Ao As [9 Carpinteria Beach i 8 9 10 11 12 Figure 3 Comparison of £. tinctum shell length versus the number of juveniles, males, and females in each population. Snails in transition from male to female (possessing both spermatozeugmata and ova) are also indicated. Although this area receives gentle surf year around, a perceptible change occurs in the structure of the beach between summer and winter. In summer, a sand bar fills in the area between the shore and the first rock outcrop. Sand, several centimeters in depth, fills in the crevices between clonal anemones and covers many of the solitary anemones, which are primarily located lower in the in- tertidal. Algae flourish during this time and cover exten- sive surfaces of rocks. During winter storms much of the sand cover is removed from between the clonal anemones and the algal cover on the rocks is reduced. Methods Epitonium tinctum was collected approximately month- ly at low tide from each site; from July, 1977, to June, 1979, at Bodega Head and from July, 1978, to June, 1980, at Carpinteria Beach. Additional collections were made in 1980 and 1981 at both locations. Snails were collected by carefully searching among aggregations of Anthopleura elegantissima in an area approximately 3 m? for 30 minutes. Similar areas, exposed at mean lower low water, were used for repeated collections at each site. Snails were located by visually scanning the area or by lightly stroking the anemones with fingertips. Pressure on the anemone body caused the release of a small amount of fluid which washed over the anemone bed and removed some sand and debris. When the tip of a snail or a clump of eggs was spotted it was carefully removed by hand. Often a mucoid thread was attached from the snail or eggs being removed to other snails and eggs in an aggregation. Care was taken to remove all snails and eggs in a partic- ular aggregation before searching for additional snails. Lengths and widths of snails were measured using a variety of dissecting microscopes with calibrated ocular micrometers. The number of post-nuclear whorls was de- termined by placing the shell with the operculum facing upward and counting from the largest body whorl to the smooth nuclear whorls. Axial costae were counted from the first costa behind the outer lip, around the body whorl, 994 504 Percent Sexually Mature Individuals e Bodega Head x Carpinteria Beach 0.01 1 2 3 4 5 6 7 8 9 10 Size Class (mm shell length) Figure 4 Mean minimum size of E. tinctum at onset of sexual maturity. A. Males replaced juvenile snails at an average length of 2.6 mm in Bodega Head and 1.9 mm in Carpinteria Beach. B. Females replaced males at an average shell length of 7.3 mm in Bodega Head and 5.4 mm in Carpinteria Beach. No significance test was run between sets of lines; an appropriate test is not yet in print (Wenner, in preparation). to the point where the first costa met the costa directly above it. Weights (to 0.1 mg) were obtained on a Mettler balance from animals and shells originally preserved in 50% isopropyl alcohol, which were removed from the al- cohol and allowed to air dry 24 hours at room temperature before weighing. ; To determine sex, snails preserved in 70% ETOH or 50% isopropyl alcohol were first measured and then placed in a vial filled with Bouin’s solution for 24 hours to dis- solve their shells. Individuals were then examined with a dissecting microscope at approximately 50. Males were recognized by the presence of spermatozeugmata, which were clearly visible as very white, almost iridescent strands densely packed in the gonad especially to the right of the stomach. Females were distinguished by the absence of spermatozeugmata and the presence of oocytes and ova. Several snails, termed hermaphroditic, were characterized The Veliger, Vol. 26, No. 1 by having proximally located spermatozeugmata and dis- tally located oocytes and ova in their gonads. Probability Paper Analysis of Polymodal Frequency Distributions The technique of plotting cumulative normal distribu- tions on probability graph paper, first descrbed by Har- DING (1949), with methodology presented by CAssIE (1950), is a valuable graphical method with which to ana- lyze bimodal or polymodal size-frequency distributions in a population. In a polymodal size-frequency sample, in- dividual modes may be expanded and the mean and stan- dard deviation calculated from each mode (see CASSIE, 1954, for method). If modes are compared from two or more populations, subtle differences may be distinguished. Probability paper analysis has been utilized to compare sex ratios and size in crustacean populations (WENNER, 1972) and to define and compare the mean minimum size at onset of sexual maturity in populations of the sand crab Emerita analoga (Stimpson, 1857) from different locations (WENNER et al., 1974). RESULTS A compilation of size-frequency data for all snails col- lected and measured over a two-year period in Bodega Head and Carpinteria Beach is presented in a histogram (Figure 2). From these data a polymodal analysis of each population was calculated by the use of probability graph paper. That analysis indicated larger snails were more prevalent at the northern site than at the southern site. The modes that appear in each population were then ex- panded, also by the use of probability paper. Three dis- tinct modes were calculated for each population consisting of small, medium, and large snails. For Bodega Head these values were, respectively, as follows: x = 2.20 + 0.55 mm, n= 152; x = 5.40 + 1.20 mm, n= 1134; x= 8.35 + 1.15 mm, n = 406. For Carpinteria Beach values were, respectively: x = 2.65 + 0.77 mm, n=193; x= 4.75 + 0.90 mm, n = 487; x = 7.20 + 0.90 mm, n = 160. Since these snails are protandric hermaphrodites, chang- ing from juveniles to males and then from males to females as they grow, the modes that appear possibly represent the juveniles, males, and females in each population. More than 600 snails were examined for the presence or absence of spermatozeugmata during spring and fall (Figure 3). The size at which these protandric hermaph- rodites became sexually mature and changed sex was greater in the northern population than in the southern population. In the transition from male to female in each population, snails were occasionally found with both sper- matozeugmata and ova present. The data for transition of juveniles to males and males to females were plotted on probability paper (Figure 4) and the mean minimum size was calculated (e.g., WENNER et al., 1974). The mean size of females was larger at Bodega Head (7.3 mm) than at C. R. Smith & A. Breyer, 1983 Page 41 Dh NO [o) ———e = oa fi = —— Log Snail Weight (mg) e Bodega Head x Carpinteria Beach T aa fo T 1 2 3 4 5 Vier aa ree T 6 7 8 9 10 11 Snail Length (mm) Figure 5 Comparison of E. tinctum length versus weight. A complete overlap existed between sets of points, so only one curve was fitted (by inspection). Carpinteria Beach (5.4 mm). The mean size of juveniles was also larger at Bodega Head (2.6 mm) than at Car- pinteria Beach (1.9 mm). The weights of snails (from 1.7 to 91.3 mg) in each population were obtained for snails ranging in length from 1.9 to 11.0 mm (Figure 5). The results revealed little variation in weight of snails at the same length between populations. The width of snails relative to shell length from both populations varied somewhat (Figure 6). We observed that some snails within each population ap- peared to have either a more squat or more elongate shell shape than the others. Although the range of variation within each population was notable, regression lines did not differ significantly between northern and southern populations. Snails of similar lengths varied considerably in number of body whorls (Figure 7), but did not differ appreciably between populations. The number of axial costae on the largest body whorl of shells from both lo- cations ranged between 11 and 14, which agreed with DUSHANE’s (1979) findings. The size and abundance (Figure 8) of Epitonium tinc- tum collected monthly at Bodega Head fluctuated more than in similar collections from Carpinteria Beach. Small snails (less than 3 mm) were found occasionally through- out the year at Bodega Head; however, they were most common from late winter to early summer. In 1978 many small snails were found in March and April; in 1979 few small snails were found until June; in 1980 many small snails were found in January. Every year during the sum- mer and early fall, snails were common. In 1978 the mean shell length increased every month during this period. In the fall large females (longer than 8 mm) were most com- mon, and particularly large clusters of egg cases associated with large aggregations of snails were found (Figure 9A). In the late fall and early winter the abundance of snails declined. Snail abundance decreased from December, 1977, to February, 1978, and from February to May, 1979. Snails remained rare until an influx of small snails ap- peared. No clear annual pattern was shown at Carpinteria Page 42 ithe Vieligern) Vols Zo Noms 5.5 5.0 4.5 4.0 Shell Width (mm) KS) a 2.0 e Bodega Head x Carpinteria Beach a a enn ne 7 T T T a 1 2 3 4 5 6 7 8 9 10 11 Shell Length (mm) Figure 6 Comparison of E. tinctum length versus width. The slopes of the lines were not significantly different (P > 0.01). The regression for Bodega Head was y = 0.22 + 0.43x (n = 95) and Carpinteria Beach was y = 0.33 + 0.39x (n= 94). Beach for size or abundance of snails collected from July, 1978, to June, 1980. The mean snail length (4.5 to 6.0 mm) remained fairly constant through time. Newly settled snails (less than 3 mm in length) occurred commonly at all seasons. The largest individuals were observed sporad- ically throughout the year. Egg masses were recorded in the field in all collections in equal abundance. Snails with eggs most commonly occurred as individuals or in small groups (2-4 individuals) with few egg capsules per egg mass (less than 50 to about 500). Larger snail groups (10- 15 individuals) and large egg masses (1000-5000 eggs) occurred without respect to season (Figure 9B). During the winter of 1981, severe storms washed away large amounts of sand from Carpinteria beaches. It was difficult to cross on foot from the cliff to the first rock outcrop, because of the diminished sand coupled with large waves. In February, 1981, the population of snails was greatly reduced. An influx of small snails (mean length 3.5 mm) was noticed in the spring (April, 1981). A population fluctuation occurred at Carpinteria Beach that year sim- ilar to fluctuations observed at the Bodega Head site. DISCUSSION Our results support the literature on physiological varia- tion in intertidal molluscs summarized by NEWELL (1964). He concluded that northern species generally attain a larger final size than southern ones. WEYMOUTH et al. (1931) studied the razor clam Stliqua patula (Dixon, 1789), and concluded that growth in southern localities was ini- tially more rapid, but less sustained and hence led to smaller total lengths. There are several reasons why snails may grow larger at higher latitudes. Higher latitudes gen- erally imply a decrease in temperature coupled with an increase in environmental stresses. CHOW (1975) conclud- ed that the largest Littorina scutulata (Gould, 1849) had C. R. Smith & A. Breyer, 1983 Page 43 5 Number of Individuals 5 Number of Post-Nuclear Whorls > | Bodega Head O Carpinteria Beach 1 2 3 4 5 7 8 9 10 11 Shell Length (mm) Figure 7 Comparison of E. tinctum shell length and number of post-nuclear whorls. Along each line the numbers of individuals are arranged by collecting locality. higher tolerances to desiccation, wave shock, and osmotic stress. Thus, a larger snail may be equipped to survive the harsher environment of higher latitudes better than a smaller snail. Recently Breyer (personal observation) observed that southern California Epitonium tinctum, collected from Santa Barbara and raised in the laboratory at the Uni- versity of California, Santa Barbara, are capable of grow- ing as large as northern California snails. At present we do not know why snails in the field in southern California do not grow as large as those collected in northern Cali- fornia. Latitudinal variations in populations of Epitonium tinc- tum closely resemble those found by FRANK (1975) for the black turban, Tegula funebralis (A. Adams, 1855). In ad- dition to finding larger snails in northern latitudes, Frank found that individuals in northern 7egula populations ma- ture at a larger size and have irregular recruitment. ROBERTSON (1981) reported similar size differences be- tween populations of Epztonium albidum (Orbigny, 1842) in the British Virgin Islands and Barbados. The largest Virgin Island males were 8 mm long and females ranged to 16 mm. The largest Barbados Island males were 6 mm long while females reached 14 mm although most were 11 mm or less in length. The northern California population of Epztonium tinc- tum fluctuated in size and abundance more than the south- ern California population and had a definite periodicity. In winter, snail populations declined markedly in the north. Severe winters are known to cause increased mortality in intertidal molluscs unless animals can migrate or are pas- sively washed to deeper waters, or unless they can escape into crevices to hide (CRISP, 1964). Epitonium tinctum was difficult to locate among anemones during winter. It was assumed that most snails were washed off the rocks and died, but others may have escaped and lived out the winter in protected areas. Even if adults do survive the winter months, there is evidence that they may not increase in size appreciably until the spring. Thais (=Nucella) lapillus (Linné, 1758) grows little from October to March (Lar- GEN, 1967), and shell growth in Tegula funebralis ceases completely from November to February (FRANK, 1975). The return of spring signifies a period of larval settlement and the cycle continues. Snails grow during summer and fall until the onset of winter storms. Southern California, in contrast to the north, generally has milder winters, calmer waters, and warmer temperatures. In the southern area it is common to see all stages in the life cycle of a snail throughout the year. In the north, Epztonium tinctum larvae may survive in Page 44 Bodega Head = (o} a Snail Length (mm) The Veliger, Vol. 26, No. 1 JASONDJIFMAMJJASONDJFMAMJJASONDJFMAMJJASONDJFMAMJJA Carpinteria Beach = fo) Snail Length (mm) a B JASONDJFMAMJJASONDJSFMAMJSJASONDJFMAMJJASONDJFMAMJJA = a (o} = fo} {>} Number of Snails Collected a ° > Bodega Head Carpinteria Beach JASONDJFMAMJJSASONDJFMAMJJSASONDJSFMAMJJASONDJFMAMJJA 1978 1979 1980 1981 Figure 8 Size (including mean and standard deviation) and abundance of snails collected at Bodega Head and Carpinteria Beach from June, 1977, to July, 1981. A. Bodega Head, size. B. Carpinteria Beach, size. C. Bodega Head and Carpinteria Beach, abundance. the plankton from late September until settlement, which often occurs as late as June. In the south, young snails and eggs are found throughout the year. Preliminary ob- servations indicate the larvae of E. tinctum grow consid- erably before they undergo metamorphosis. Before hatch- ing, veligers of E. tinctum measure about 70 um in length. In the laboratory we have kept them alive for more than two months, at which time they had attained a size of more than 250 um but had not yet metamorphosed. In the field the smallest snail observed during more than two years of study was 1.1 mm; apparently, veligers grow considerably in the plankton before they settle. DEHNEL (1955) looked at growth rates of embryos and larvae of northern and southern gastropod populations. He con- cluded that the rates of growth of embryos and larvae in northern gastropod populations are often 2 to 9 times greater than in southern populations of the same species at a given temperature. C. R. Smith & A. Breyer, 1983 Page 45 Figure 9 Comparison of the largest groups of snails and their associated egg masses found in fall 1980 collections. A. Bodega Head (26 September 1980; SBMNH 33878). B. Carpinteria Beach (23 October 1980; SBMNH 33879). Once veligers metamorphose, growth in general is faster in warm than in cold seas, but this does not necessarily mean the largest species live in warm seas. The largest snails may be the slow-growing species found in cold water; and within a species, an individual is likely to grow more slowly, but become larger and older, at the northern limits of its distribution (FRANK, 1969). This theory is consistent with our observations. Larger Epitonium tinctum were found north of Bodega Bay in Fort Bragg, and the largest specimens were recorded from Forrester Island, Alaska, the northern limit of distribution for this species. ACKNOWLEDGMENTS The authors would like to thank Drs. Roger Seapy, Jo- seph Rosewater, Demorest Davenport, Robert Robertson, James Nybakken, and Diane Perry for critically review- ing the manuscript, and Laurie Marx for preparing the text figures. Dr. Adrian Wenner assisted at all stages with the statistical interpretation. Our special thanks go to Dr. F. G. Hochberg for his continued support and help throughout this study. LITERATURE CITED BREYER, A. 1982. Observations on the reproduction, feeding and ecology of the wentletrap Epitonium tinctum (Gastrop- oda: Mesogastropoda). Master’s thesis, Sonoma State Uni- versity. 50 pp. BULNHEIM, H. P. 1968. Atypische Spermatozoenbildung bei Epitonium tinctum. Ein Beitrag zum Problem des Sperma- tozoendimorphismus der Prosobranchia. Helgol. Wiss. Meeresunters. 18:232-252. Cassie, R. M. 1950. The analysis of polymodal frequency distributions by the probability paper method. N.Z. Sci. Rev. 8:89-91. CassiE, R. M. 1954. Some uses of probability paper in the analysis of size frequency distributions. Aust. J. Mar. and Freshwater Res. 5:513-522. Cuow, V. 1975. The importance of size in the intertidal dis- tribution of Littorina scutulata. Veliger 18(1):69-78. Crisp, D. J. 1964. The effects of the severe winter of 1962- 63 on marine life in Britain. J. Anim. Ecol. 33:165-210. DEHNEL, P. A. 1955. Rates of growth of gastropods as a func- tion of latitude. Physiol. Zool. 28:115-144. DuSHANE, H. 1979. The family Epitoniidae in the northeast- ern Pacific. Veliger 22(2):91-134. FRANK, P. W. 1969. Growth rates and longevity of some gas- tropod mollusks on the coral reef at Heron Island. Oecologia 2:232-250. FRANK, P. W. 1975. Latitudinal variation in the life history features of the black turban snail Tegula funebralis (Proso- branchia: Trochidae). Mar. Biol. 31:181-192. HarDING, J. P. 1949. The use of probability paper for the graphical analysis of polymodal frequency distributions. J. Mar. Biol. Assoc. U.K. 28:141-153. HocuBerG, F. G. 1971. Functional morphology and ultra- structure of the proboscis complex of Epitonium tinctum (Gastropoda: Ptenoglossa). ECHO (West. Soc. Malacol.) 4: 22-23. LarGEN, M. J. 1967. The influence of water temperature upon the life of the dog-whelk Thais lapillus (Gastropoda: Prosobranchia). J. Anim. Ecol. 36:207-214. NEWELL, G. E. 1964. Physiological variation in intertidal mol- luscs. In: K. M. Wilbur & C. M. Yonge (eds.), Physiology Page 46 of Mollusca. Vol. 1. Academic Press, New York. pp. 59- 87. REscH, C. 1972. Chemical recognition of prey in the gastropod Epitonium tinctum (Carpenter). Master’s thesis, Univ. Cal- if., Santa Barbara. 61 pp. ROBERTSON, R. 1981. Protandry with only one sex change in an Epitonium (Ptenoglossa). Nautilus 95(4):184-186. SMITH, C. REscH. 1977. Chemical recognition of prey by the gastropod Epitonium tinctum (Carpenter, 1864). Veliger 19(3):331-340. SOKAL, R. R. & F. J. ROHLF. 1969. Biometry. W. H. Free- man, San Francisco. 776 pp. The Veliger, Vol. 26, No. 1 STRONG, A. M. 1941. Notes on Epitonium (Nitidiscala) tinctum (Carpenter). Nautilus 55(2):46-47. WENNER, A. M. 1972. Sex ratio as a function of size in marine Crustacea. Am. Nat. 106:321-350. WENNER, A. M., C. FUsARO & A. OATEN. 1974. Size at onset of sexual maturity and growth rate in crustacean popula- tions. Can. J. Zool. 52:1095-1106. WEyYMoOuTH, F. W., H. C. MCMILLAN & W. H. RicH. 1931. Latitude and relative growth in the razor clam Siliqua pa- tula. J. Exp. Biol. 8:228-249. The Veliger 26(1):47-51 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 The Reproductive Cycle of the Trochid Gastropod Oxystele variegata (Anton, 1839) M. A. P. JOSKA ano G. M. BRANCH Zoology Department, University of Cape Town, Rondebosch. 7700, South Africa Abstract. Oxystele variegata (Anton, 1839) is a trochid gastropod found commonly in the southern African intertidal region. Monthly samples over a one-year period were examined for gonadal devel- opment by comparing dry somatic and gonadal tissue weights and histological sections. Mature oocytes and sperm could be found in monthly samples of the gonads of all animals over 10 mm in size throughout the year, but based on changes in gonadal weights two spawning peaks were detected—one in February and a second in September to October. It seems likely that this temperate species is able to spawn throughout the year but that spawning is intensified during these peak periods. INTRODUCTION Oxystele variegata (Anton, 1839) is a trochid gastropod which occurs commonly in the rocky intertidal zone in estuaries and along the southern African coast from south- ern Angola (KENSLEY & PENRITH, 1972) to Durban in South Africa (Day, 1974). It is one of four species in the genus Oxystele that are found in the balanoid zone, where it is very prominent. Because of its abundance O. variegata must be a major contributor to the energy budget of this zone. Based on measurements of the metabolic rate of this species, KING (1974) estimated that the annual population production was 46.4 kJ-m™~?-yr'. McQualb (1980, 1982a) has described how the population of O. variegata is zoned according to size with the largest animals in the upper balanoid zone and the smallest lower on the shore. No attempt has yet been made to ascertain the breeding cycle with any certainty, but McQuarp (1980) hypothesized from settlement patterns that spawning occurred through- out the year, with a peak in February and two smaller peaks in May/June and September/October, 1977. Monodonta lineata, a related trochid of the northern cool temperate regions, has been investigated by DEsaI (1966), who found that this species spawned from May to August with male and female gonads showing similar cycles. UNDERWOOD (1972) investigated the breeding cycle of the British trochids M. lineata, Gibbula umbilicalis, and G. cineraria. He found that M. lineata and G. umbilicalis spawned throughout July and August and less in Septem- ber. However, G. cineraria did not provide clear evidence of spawning since histological sections of the gonads ap- peared the same throughout the year. UNDERWOOD (1974) also found that the trochid Auws- trocochlea constricta, which occurs on the west and south- west coast of Australia, did not show any seasonal change in the appearance or histology of the gonad, and he con- cluded that this species spawned throughout the year with a peak in October/November. PAINE (1971), using energy content as a measure of reproductive state, concluded that the western North American trochid Tegula funebralis spawned once a year during the summer. In this paper a description is given of the breeding cycle of Oxystele variegata, based on changes in the ratio of gonadal/somatic dry weight and gonadal histology during the course of a full year. MATERIAL ann METHODS Monthly samples of Oxystele variegata were collected at random from the upper balanoid zone at Dalebrook on the Cape Peninsula (47°07’S, 18°27’E) during the period April, 1980, to March, 1981. The animals were preserved in a 5% formalin solution until they could be examined. Fifteen to 20 animals of each sex were dissected each month. Every shell was measured across the base at the widest point and the animal was then removed and the operculum cut off and measured. The gonad was dissected out of the somatic tissue and both gonadal and somatic tissue were dried in an oven at 55°C for 24 h. The dry Page 48 Table 1 Mean monthly shell sizes and opercular diameters of Ox- ystele variegata sampled for gonadal development. Opercular Shell size (mm) diameter (mm) Month 3 Q é g April, 1980 14.9 14.8 5.4 5.5) May 16.4 16.0 6.0 5.8 June 16.0 16.0 5.8 5.6 August 15.5 16.2 5.4 Dod) September 16.8 16.7 6.0 6.0 October 16.6 16.5 6.0 59) November 16.8 16.4 5.8 5.8 December 15.6 16.0 6.1 59) January, 1981 15.6 15.5 5.6 5.8 February 16.5 16.5 5.9 5.9 March 15.3 16.5 5.4 6.0 Annual mean 16.0 16.1 5.8 5.8 weights of the gonad and of the somatic tissue for each animal were measured. In addition to the above animals, two to five specimens of each sex were collected each month and used to obtain histological sections of the gonads. The gonads of these preserved specimens were dehydrated in alcohol and blocked in paraffin wax. Transverse sections were cut through the center of the gonads at 10 um. The sections were dehydrated in alcohol and stained with hematoxylin and eosin. The male and female gonadal weights were linearly regressed on somatic tissue weights for each month. The data obtained from these regressions were used to calcu- late monthly gonadal indices. These indices were defined as the ratio of gonadal to somatic weight, expressed as a percentage for a standard animal of 0.07 g somatic weight (with an opercular diameter averaging 5.8 mm). RESULTS All specimens dissected, ranging from 13.6 mm to 21.6 mm shell width, had mature gonads throughout the sam- pling period. Individuals of Oxystele variegata less than 10.0 mm shell width lacked gonads and progressively ‘larg- er gonads could be found in animals with a shell width of 10.0 mm and upwards. In both males and females of O. variegata the gonads lie adjacent to and above the digestive gland, extending into the shell spiral up to the apex. As with most archaeo- gastropods, trochids release the unfertilized gametes di- rectly to the surrounding environment where fertilization occurs (FRETTER & GRAHAM, 1962; DESAI, 1966; PAINE, 1971; UNDERWOOD, 1972; BRANCH, 1974). Oxystele var- vegata probably conforms to this pattern since we observed no egg laying or egg masses. The male gonad is a pinkish, densely constituted tissue. Histological sections showed that The Veliger, Vol. 26, No. 1 FEMALE o +A HA . oe \ ! 60 somatic weight 0.04 gonadal index 0 0.02 20 is \ J A Ss gonad weight 0.00 A M J N OOL xX anssi} 91eWoSs/peUuoy 0.08 Gonadal & Somatic tissue weights (g) 0.06 gonadal index 0.04 0.02 0.00 1980 1981 Figure 1 Gonadal and somatic dry tissue weights (g) of “standard”? Ox- ystele variegata with an opercular diameter of 5.8 mm and the gonadal index (expressed as a ratio of gonad : somatic weight X 100) plotted against time (month). (Bars indicate 95% confidence levels.) there were ripe spermatozoa present throughout the year. The ratio of spermatozoa to spermatids and spermatocytes differed from individual to individual in the same month, but spermatozoa never constituted less than 50% of the area of the testis (as determined from grid counts of the gonadal sections). The female gonad contains a mass of white or creamy oocytes, each within a clear jelly coat which is a feature of the trochids (FRETTER & GRAHAM, 1962; Dersal, 1966; DucH, 1969; UNDERWOOD, 1972, 1974; SIMPSON, 1977; WEBBER, 1977; HESLINGA, 1981). Mature oocytes were present throughout the year and had an average diameter of 180 wm, not including the jelly coat. Histological sections showed immature oocytes were also present throughout the year, but not common from January to February or from August to September. No difference existed in mean shell or opercular diameters between the sexes of the animals sampled (Table 1). The gonadal indices show strong correlation between the sexes, both peaking from September to October, 1980, M. A. P. Joska & G. M. Branch, 1983 Page 49 and in February, 1981, and both dropping in November, 1980, and March, 1981 (Figure 1). Since this gonadal index can be affected by changes in somatic weight, an independent assessment of changes in somatic and gonadal weights was also undertaken. Both gonadal and somatic weights were linearly regressed on opercular diameter for each month. These regressions are shown in Table 2. From these regressions the gonadal and somatic weights of an- imals with a standard opercular diameter of 5.8 mm could be calculated for each month. The somatic weights fluc- tuated about a mean of 0.07 g for each sex. As shown in Figure 1 the somatic weights of both sexes peaked above the mean in September and November and dropped below the mean in April. The gonadal weights of both sexes reflected the gonadal indices almost completely, thus con- firming that in spite of slight changes in the somatic weight the gonadal index is a reliable measure of reproductive activity. The number of non-significant correlations between go- nadal weight and opercular diameter seen in Table 2 in- dicates a wide “scatter” amongst the population in certain months, particularly the males. This scatter is due to the fact that some individuals are sexually mature at a time when others have just spawned. Again, however, there was a strong correlation between the sexes for both the gonadal and somatic tissue weights, and we are confident that the gonadal index and changes of gonadal weights do reflect reproductive cycles. DISCUSSION It appears that Oxystele variegata, as with some other tro- chids (UNDERWOOD, 1972, 1974; SIMPSON, 1977) exhibits continuous gametogenesis once mature. Synchronization of spawning still remains important since fertilization is external and, indeed, the male and female gonadal weights do peak and fall in unison (Figure 1). Oxystele variegata congregates in large numbers in the upper balanoid zone at Dalebrook thus also ensuring proximity of the sexes during spawning. DeEsal (1966), DucH (1969), and SIMPSON (1977) also found this clustering behavior in tro- chids. Despite this, the gonadal tissue weights (relative to opercular diameter) showed marked variation between in- dividuals, particularly around the periods of peak spawn- ing. However, in all samples dissected during other months there were always some specimens whose gonadal weights were well above or below the average and these were considered to be in pre- or postspawn states respectively. Such variation in the gonadal weights within the popu- lation probably indicates that there are some individuals that spawn out of phase with the bulk of the population. The presence of some spawned individuals in virtually every month of the year suggests that spawning occurs year round although peaking twice a year. These varia- tions in gonadal weight are reflected in the number of months in which there were non-significant correlations between gonadal weights and opercular diameter (Table 2), in contrast to the generally high correlations between somatic tissue weight and opercular diameter. The two main gonadal peaks indicate spring and late summer spawnings with the build up to the spring spawning (Sep- tember to October) being more prolonged. At present we have no evidence whether these peaks repeat each year. It is also apparent from the histological sections, which showed a constant supply of ripe ova and sperm in the gonads, and from the irregular but continuous settlement of recruits to the low shore (McQualp, 1982b), that spawning probably occurs throughout the year, peaking in the spring and late summer. The somatic tissue weights of standard-sized animals of both sexes remained steady at a mean of 0.07 g, except for April when they declined. This fall in somatic weight is possibly because during this period there is a recruit- ment of new individuals to this zone (McQuaIp, 1982b) and we speculate that increased feeding competition may strain existing food sources. By May to June the popu- lation has declined due to mortality and the somatic weight increases (Figure 1) together with new algal growth in autumn. In the upper balanoid zone, from where our sam- ples were taken, there are only two main macroalgae— Porphyra capensis and Gelidium pristoides. MCQUAID (1980) found that both these species have varied calorific values during the year. Porphyra capensis declined in cal- orific value during autumn whilst G. pristoides had a max- imum value in autumn. The drop in calorific value of some of the food available for Oxystele variegata could affect the somatic weights, but we found that O. variegata were not commonly seen on macroalgae, and they possibly mainly feed on microalgae and sporelings rasped from the substratum. This is also the opinion of McQualp (1982a). A drop in the number of sporelings has been found during April in an analysis of monthly colonization carried out by Joska (in preparation) and this, too, could explain the lower somatic weight during this month. In archaeogastropods the larval stage is brief (FRETTER & GRAHAM, 1962; DEsal, 1966; UNDERWOOD, 1972; WEBBER, 1977) and settlement on the shore should take place within a week of spawning. McQualip (1982a, b) found that 5 to 6-mm specimens of O. variegata were de- tectable in large numbers in February with a smaller peak in June/July. In addition, small numbers could be found throughout the year. Since specimens of O. variegata take about four months to reach a size of 5 to 6 mm, these juveniles probably settled about four months earlier, and are likely to have been the result of, respectively, the spring and late summer spawnings that we have recorded. Temperature is a factor often cited as being of great importance for spawning or reproduction in marine in- vertebrates. During the period of sampling the sea tem- peratures rose during the spring spawning but fell at the time of the late summer spawning (Figure 2). Apart from suggesting that a minimum temperature exists for spawn- ing, there is no obvious influence of temperature on the Page 50 The Veliger, Vol. 26, No. 1 Table 2 Regression coefficients for regressions of gonad weight (y) on operculum diameter (x): y = a) + a,x and coefficients of determination (r’) with the probability that r? is significant (P). Month ao a, ie P g April, 1980 —0.0155 0.0382 0.2729 0.05 May —0.0340 0.0794 0.4638 0.001 June —0.0152 0.0436 0.2915 0.05 August 0.0194 —0.0046 0.0026 Not significant September —0.0290 0.1103 0.2988 0.01 October —0.0034 0.0630 0.0232 Not significant November —0.0252 0.0672 0.3400 0.01 December —0.0093 0.0399 0.2495 0.05 January, 1981 0.0058 0.0251 0.0214 Not significant February —0.0676 0.1691 0.5365 0.001 March : —0.0281 0.0666 0.5426 0.001 ) April, 1980 —0.0222 0.0531 0.1509 Not significant May : 0.0374 - =0,0317/ 0.0228 Not significant June —0.0661 0.1363 0.5817 0.001 August —0.0017 0.0439 0.1618 Not significant September —0.0310 0.1093 0.1910 0.05 October —0.0195 0.0870 0.1710 0.05 November —0.0152 0.0535 0.4046 0.01 December 0.0020 0.0231 0.1047 Not significant January, 1981 —0.0239 0.0707 0.1925 0.05 February —0.0820 0.1823 0.6314 0.01 March —0.0156 0.0433 0.0940 Not significant Regression coefficients for regressions of somatic tissue weight (y) on operculum diameter (x): y = a, + a,x and coefficients of determination (r?) with the probability that r* is significant (P). Month Aly ay r? P g April, 1980 0.0346 0.0317 0.9934 0.001 May —0.1638 0.4054 0.8411 0.001 June —0.0601 0.2269 0.4540 0.01 August 0.0622 0.0144 0.0049 Not significant September —0.0863 0.2825 0.7607 0.001 October —0.1844 0.4410 0.7384 0.001 November —0.0926 0.2923 0.5173 0.001 December —0.0940 0.2788 0.4958 0.001 January, 1981 —0.1105 0.3049 0.7235 0.001 February —0.1199 0.3335 0.5736 0.001 March : —0.1688 0.4166 0.8663 0.001 ) April, 1980 —0.0229 0.1282 0.4085 0.01 May —0.0199 0.1610 0.2331 0.05 June —0.1480 0.3823 0.5455 0.001 August 0.0370 0.0659 0.1417 Not significant September —0.1046 0.3129 0.6867 0.001 October =O1B27 © 0.3571 0.8634 0.001 November —0.0941 0.2979 0.6955 0.001 December —0.0296 0.1709 0.5703 0.001 January, 1981 —0.0758 0.2295 0.6760 0.001 February —0.1319 0.3389 0.8117 0.001 March 0) UZS7/ 0.3222 0.5280 0,001 M.A. P. Joska & G. M. Branch, 1983 Page 51 24 om O — ¢ 5 20 — 0 hh o a — 16 o ~~ 0 o YM 12 = 1980 1981 Figure 2 Mean monthly sea temperatures during the sampling period. Bars indicate maximum and minimum temperatures. (Temper- ature records kindly supplied by the Department of Maritime Defence, Simonstown.) reproductive cycle in this species. Dalebrook lies in False Bay, which because of its shallow waters and current pat- terns affords year round moderate conditions and pre- dictable seasonal changes in temperature. Rather more variant conditions on the west coast could possibly pro- duce different spawning peaks. ACKNOWLEDGMENTS We would like to thank Dr. John Pearse for his pains- taking criticism of an earlier manuscript of this paper. The research was supported by a grant from the South African Committee for Oceanographic Research. LITERATURE CITED BRANCH, G. M. 1974. Ecology of Patella Linnaeus from the Cape Peninsula, South Africa. 2. Reproductive cycles. Trans. R. Soc. S. Afr. 41(2):111-160. Day, J. H. 1974. A guide to marine life on South African shores. A. A. Balkema, Cape Town. 300 pp. Desai, B. N. 1966. The biology of Monodonta lineata (Da Costa) Proc. Malac. Soc. Lond. 37:1-17. Ducu, T. M. 1969. Spawning and development in the trochid gastropod ELuchelus gemmatus (Gould, 1841) in the Hawai- ian Islands. Veliger 11(4):415-417. FRETTER, V. & A. GRAHAM. 1962. British prosobranch mol- luscs. Ray Society, London. 755 pp. HEsLINGA, G. A. 1981. Larval development, settlement and metamorphosis of the tropical gastropod Trochus niloticus. Malacologia 20(2):349-357. KENSLEY, B. & M.-L. PENRITH. 1972. Monodonta (Oxystele) fulgurata Philippi, a synonym of Oxystele variegata (Anton). J. Conchol. 27:349-352. KING, J. M. 1974. Population respiration of Oxystele variegata (Gastropoda). Zoology Honours Project, University of Cape Town. McQuaip, C. D. 1980. Spatial and temporal variations in rocky intertidal communities. Doctoral thesis, Zoology, Uni- versity of Cape Town. McQuaip, C.D. 1982a. The influence of desiccation and pre- dation on vertical size gradients in populations of the gas- tropod Oxystele variegata (Anton) on an exposed rocky shore. Oecologia 53:123-127. McQuaip, C. D. 1982b. Population dynamics and growth of the gastropod Oxystele variegata (Anton) on an exposed rocky shore. S. Afr. J. Zool. 18:56-61. PAINE, R. T. 1971. Energy flow in a natural population of the herbivorous gastropod Jegula funebralis. Limnol. Oceanogr. 16(1):86-98. SIMPSON, R. D. 1977. The reproduction of some littoral mol- luscs from Macquarie Island (Sub-Antarctic). Mar. Biol. 44:125-142. UNDERWOOD, A. J. 1972. Observations on the reproductive cycles of Monodonta lineata, Gibbula umbilicalis and G. cine- raria. Mar. Biol. 17:333-340. UNDERWOOD, A. J. 1974. The reproductive cycles and geo- graphical distribution of some common eastern Australian prosobranchs (Mollusca:Gastropoda). Aust. J. Mar. Fresh- water Res. 25:63-88. WEBBER, H. H. 1977. Gastropoda:Prosobranchia. In: A. C. Giese & J. S. Pearse (eds.), Reproduction of marine inver- tebrates. Academic Press, New York. pp 1-114. The Veliger 26(1):52-61 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 The Larval Biology of Brachidontes modiolus (Linné, 1767) (Bivalvia: Mytilidae) ANGELA FIELDS anp EUNA MOORE Department of Biology, University of the West Indies, Cave Hill, Barbados Abstract. Larvae of Brachidontes modiolus were reared in the laboratory from eggs through to settled juveniles. Egg sizes ranged between 67.3 and 77 um. Straight-hinge veligers appeared 15 to 17 h after fertilization of the eggs. The length of shelled larvae increased from 96 to 221 um: the straight-hinge stage from 96 to 176 um, the umbo stage from 168 to 221 um, and the pediveliger stage from 180 to 221 um. Settlement occurred at lengths of 180 um and upwards. The larval hinge consists of small teeth along the length of the hinge with larger teeth at both ends. Larvae of B. modiolus develop more rapidly and settle at an earlier age than larvae of B. recurvus and B. granulata. INTRODUCTION Brachidontes modiolus (Linné, 1767) (=B. citrinus) is a small marine mytilid, the adults of which measure be- tween 38 mm (ABBOTT, 1974) and 46 mm (MCLEAN, 1951). The geographic range for this species is from Flor- ida to the West Indies (ABBOTT, 1974). In Barbados, members of this sublittoral species may be found recessed within beds of the seagrass Thalassia testudinum Konig or attached to the rocky surface of reef flats at depths of O- 2 m. Brachidontes modiolus is highly gregarious and, whether living epifaunally or infaunally, occurs in dense aggregates of individuals. The specific name Brachidontes citrinus (Réding, 1798) has been widely used. ABBOTT (1974), however, consid- ered this to be a synonym of B. modiolus (Linné, 1767). Indeed the name ‘modiolus’ is singularly apt, for B. mo- diolus exhibits characteristics of the Modiolus group of mytilids, namely the possession of subterminal umbones and a shell shape which is more conical than triangular (STANLEY, 1970, 1972). The larval biology of Brachidontes modtolus has hitherto been unreported in the literature under any of its syn- onyms. A knowledge of the duration of the planktonic stage of an aquatic larva, the time during which it is exposed to ocean or coastal currents, could contribute to an understanding of the distribution of a species in a given region. This aspect has been discussed by COE (1953) and SCHELTEMA (1971). In addition, a description of the larval stages of B. modiolus would aid in the identification of these larvae when encountered in plankton samples, as well as establish the life history pattern and strategy of the species. The identification of bivalve larvae has posed problems in the past. LOOSANOFF et al. (1966) drew attention to the inadequacy of “indirect methods” of identification of bi- valve larvae, and pointed out that these methods have led to discrepancies in the description of larvae of the same species when reported by different authors. Indirect meth- ods include monitoring the development of an unidentified larva found in the plankton through to settling and the assumption of identifiable features of a particular species. These authors recommend “direct methods” of identifi- cation, 1.e., the rearing of larvae from fertilized egg stage through to metamorphosis under controlled laboratory conditions. The development of reliable methods for ob- taining viable gametes and for the rearing of juveniles has resulted in the successful culture of the larvae of several marine bivalves. LOOSANOFF & Davis (1963) reviewed methods for the cultivation of larvae and detailed the spe- cific requirements necessary for the successful culture of 19 species of bivalves. As characters to be used in the identification of marine bivalve larvae, LOOSANOFF e¢ al. (1966) listed the dimensions of the larval shell (the pro- dissoconch), its general shape, prominence of the umbones during growth to metamorphosis, and ratios of length of hinge to maximum length or width of shell. CHANLEY & ANDREWS (1971) and CHANLEY & CHANLEY (1980) pro- vided useful terminology for describing larval shell form and, in order to provide a better description of larval shell length, height and depth relationships. LOOSANOFF et al. (1966) provided a guide to the dimensions and shapes of A. Fields & E. Moore, 1983 Page 53 20 species of bivalves, and CHANLEY (1970) gave a review of the larval characteristics of the Mytilidae. Within the genus Brachidontes, descriptions of the larvae of B. recur- vus from the western North Atlantic Ocean have been published in CHANLEY (1970), and of B. granulata from central Chile in CAMPos & RAMORINO (1980). The lar- vae of B. senhausi were described in YOSHIDA (1937), but according to KURODA ef al. (1971; cited in CAMPos & RAMORINO, 1980), this species belongs to the genus Mus- culus. The purpose of this paper is to provide a description of the larval stages of Brachidontes modiolus, with informa- tion on the development and length of the larval life of the species. MATERIALS anp METHODS Gametes were obtained from ripe mussels collected during the peak of the reproductive season (June to September). This peak was determined by continuous sampling and histological techniques over a period of two years. In the laboratory the animals were cleaned of sediment and en- crustations and placed in freshly collected sea water. Ini- tially, several methods were employed in an attempt to obtain viable gametes: (1) temperature shock within the range 24—34°C; (2) pricking of the posterior adductor muscles of the adults; (3) introduction of sperm and/or eggs obtained from stripped gonads to water containing ripe mussels; (4) treatment of stripped gonads with a 0.1 N solution of ammonium hydroxide; (5) exposing the mussels to hydrogen peroxide in alkaline (pH 9.1) sea water as per Morsk et al. (1977). Only this last method was successful; hence, the procedure is detailed below. Alkaline sea water was prepared with predetermined quantities of Trizma*, hydrochloric acid and natural sea water (pH 7.2). Three or four mussels were placed in- dividually or collectively in beakers containing 50 ml of alkaline sea water per animal (pH 9.1); hydrogen per- oxide was then added with a micropipette to a final con- centration of 5mM. This was the treatment solution. The animals were kept in this solution for one hour, after which time the liquid was decanted and replaced with 100 ml of fresh treatment solution. One hour after the second treatment, the mussels were removed from the al- kaline sea water, rinsed, and placed in beakers containing 100 ml of natural sea water. These were then placed in a water bath at a temperature of 33-34°C. Males and females received the same treatment. The onset time of spawning varied widely, from 30 minutes to several hours after treatment in the hydrogen peroxide solution. Mus- * A brand name. Use here does not necessarily imply endorse- ment of the product. sels seen spawning were removed and placed individually in beakers containing natural sea water at 28°C. The eggs were rinsed thoroughly with sea water filtered through glass fiber filter paper. Sperm was added to the egg sus- pension, and the two allowed to remain undisturbed for 10 min to encourage fertilization. In some instances a female in the process of spawning was placed in a weak sperm suspension, and so the eggs were fertilized as they were extruded. The fertilized eggs were rinsed to remove excess sperm and placed in 150 ml of filtered sea water in Erlenmeyer flasks of 250 ml capacity. Possible contam- ination of the cultures was reduced by placing tissue paper in the necks of the flasks and covering the opening of the flask with aluminum foil. If flasks containing larvae were left open to the air, a dense population of ciliates soon developed, with consequent death of the larvae. The em- bryos were left undisturbed for 24 h, at which time the strongly swimming larvae were pipetted off and intro- duced into fresh sea water to a density of 25-30 larvae/ ml. The water containing the larvae thereafter was changed every third day. This was achieved by washing the larvae over a fine nylon mesh which was glued securely over one end of a large open-ended glass tube. Mortality of the larvae during washing was reduced if the mesh was held in a beaker containing filtered sea water, so that the mesh was positioned below the surface of the water. ‘The bottom layer of water in the flask was discarded as dead or weak larvae accumulated there. The flasks were thoroughly washed at each water change, and care was taken to en- sure that no detergents contaminated the sea water. The sea water used for each change was freshly collected and filtered. Best survival rates were obtained when the larvae were reared in unaerated sea water. The larvae remained unfed for 48 h and thereafter were fed on a mixture of Dunaliella tertiolecta and Nephroselmis tibron. When possible, /sochrysis galbana was added to the diet. Feeding densities in the case of the larger alga, z.e., Dunaliella, was approximately 8000-10,000 cells per ml of culture solution. Concentrations of the smaller alga were higher, 70,000-80,000 per ml of culture solution. Algal concentrations of these magnitudes were cleared by the larvae in 24 h. Sub-samples of approximately 10 larvae were mea- sured every two days for two weeks. Length was measured as the maximum distance in the antero-posterior direction, height as the maximum distance from the hinge to the ventral margin of the shell, and depth as the maximum left-right dimension (CHANLEY, 1970). RESULTS The eggs of Brachidontes modiolus ranged in diameter from 67.3-77 wm, and were a uniform dark brown in color. When spawned, the eggs had an irregular appearance but soon became rounded. No measurements were made of the spermatozoa. The sperm obtained from stripped go- Page 54 The Veliger, Vol. 26, No. 1 A. Fields & E. Moore, 1983 Page 55 nads appeared less active than those obtained through the mediation of hydrogen peroxide. As in BAYNE (1965), the stages chosen to evaluate de- velopment times to the straight-hinge stage were those easily recognized and not those with particular embryo- logical significance (Table 1). Stage 1—First division: the time at which 50% of the sam- ple had undergone first cleavage. Stage 2—The ciliated blastula: first appearance of cilia, evidenced by slowly rotating larvae (Figure 1a). Stage 3—The early trochophore: appearance of an apical flagellum (Figure 1b). Stage 4—Veliger: appearance of long cilia on the apical plate (Figure Ic). Stage 5—Transitional stage: first appearance of the shell as a transparent object on the dorsal surface of the larva (Figure 1d). Stage 6—Straight-hinge veliger: the possession of a com- plete shell, the prodissoconch. Very early straight-hinge veligers are darker than the later ones (Figure le). First division occurred approximately 36 min after fer- tilization and the ciliated blastula appeared 3 h later. The ciliated blastulae swam slowly at first and as time pro- gressed swam upwards towards the surface of the water column. Development of the trochophore occurred 5 h after fertilization, and veligers were first seen 4 h later. Transitional stages between the veliger and the straight- hinge veliger appeared 12 h from zero time, and straight- hinge veligers were seen in cultures 15-17 h after fertil- ization. The color of the larvae underwent a change during development, and stages 1 through 5 were dark brown while stage 6, the straight-hinge veliger, was a pale yellow-brown. The rate of development of the larvae varied with tem- perature. Larvae reared at 34°C developed into straight- hinge veligers 13 h after fertilization. In cultures kept at 24°C, it was not until 25 h had elapsed that straight-hinge veligers were observed. At the time that straight-hinge veligers were recorded from cultures reared at 28°, 32° and 34°C (7.e., between 13-17 h), larvae reared at 24°C were still at the early trochophore stage. Larval Dimensions and Shape The early straight-hinge larvae measure 96 um in length, with a minimum height of 67 wm. No measurement of depth was recorded at the earliest straight-hinge stage. The smallest depth recorded was 56 um at a larval length Table 1 Summary of the observed development times of the larvae of Brachidontes modiolus. Temperature = 28°C. Stage Description Time 0 Fertilization 0 1 First division 34-36 min 2 Ciliated blastula 3h 40 min 3 Early trochophore 5 h 15 min 4 Veliger 9h 15 min 5 Transitional stage 12h 6 Straight-hinge veliger 15-17 h of 152 wm. Straight-hinge larvae attained a size of up to 176 um in 6 days. Heights then ranged from 126-144 um, and depths from 84-104 wm. At lengths greater than 176 um the hinge line invariably showed signs of rounding (Figure 2). Lengths increased faster than height, which in turn increased faster than depth. Between lengths of 96 and 115 wm, heights were 19-29 um less than length, while at a length of 176 um, heights measured 32-50 um less than length. The first indication of umbo development appeared at 168 wm, with heights ranging from 120-136 um and depths from 72-88 um. All observed combinations of larval heights and depths for a given length are given in Figure 3. The length of the hinge at a shell length of 152 um was 120 um, increasing to 128 um at a shell length of 168 wm. The results are summarized in Table 2 and Figure 2. The straight-hinge larva is roughly ““D” shaped. The hinge line is long relative to the length of the shell. The shoulders slope steeply, with the posterior shoulder short- er and sloping more steeply than the anterior. The pos- terior end is higher and more pointed than the anterior. The ventral margin of the shell is rounded. The hinge line becomes slightly rounded with the development of the umbo, and is at first “round” or “indistinct.”” The umbo may remain low and not clearly defined through lengths of 168-288 wm, becoming “broadly rounded” in later stages. The minimum length at which the umbo was seen to project above the shell margin is 240 wm. The shoulders at this stage are almost straight, the anterior shoulder not sloping as steeply as the posterior. The ventral margin is now markedly elongate, but still rounded. The larval hinge consists of a series of small teeth, fanked by two or three larger teeth. Figure 1 Stages in the early larval development of Brachidontes modiolus. A. ciliated blastula; B. early trochophore; C. veliger; D. transitional stage; E. straight-hinge stage. Page 56 A B 120 x82 E 191 « 146 I 224 x 184 150 x 109 208 x 176 ithe Veligers Volk ZomNomi Cc D 159x118 173 x 137 218 x 182 J 282 « 264 Figure 2 The larvae of Brachidontes modiolus at different stages of development from the early straight-hinge stage (A) to settlement (G-I). Larvae are positioned with anterior end to the left, except in J. The larva in J is an early juvenile just beginning dissoconch growth. The length and height of the larvae are indicated under each photograph; measurements are in microns. The smallest pediveliger larva seen using its foot mea- sured 180 wm (Figure 4) in length, but more generally, ambulatory pediveligers appeared at a length of 184 um. The eye spot first appeared in larvae at length 184 um when cultured at temperatures of 32° and 34°C. Internal Anatomy of the Larvae The internal anatomy of the larva was at first indistinct at magnifications of up to 500X. The gut became apparent on the first day as a straight tube running in a posterior direction. It soon became coiled (Figure 5a). The digestive gland soon became easily visible, as with the onset of feed- ing in the larvae, the organ developed a green-brown color (Figure 5b). The adductor muscles showed clearly by the third day, and the velar retractor muscles were also con- spicuous at this time. The foot was fully developed by day 6, with the pedal retractor muscles clearly visible (Figure 5c, d). The velum increased in size with the development of the larva to the pediveliger stage, and occupied a large portion of the shell cavity. The gill filaments were not clearly visible until after metamorphosis (Figure 5e). The A. Fields & E. Moore, 1983 Rage y), Figure 3 Larval dimensions of Brachidontes modiolus. The height and depth co-ordinates run parallel to the length axis. The dots represent length-depth or length-height measurements. The lines enclosing the dots were fitted by eye and represent maximum and minimum height and depth measurements. The three-dimensional figure represents all possible length, height, and depth combinations for B. modiolus (after CHANLEY & VAN ENGEL, 1969). The clear area represents the straight-hinge stage, the lined area the umbo stages, and the stippled area, the transitional stage between straight-hinge and umbo forms. larvae of Brachidontes modvolus settled at sizes between 180 and 221 um in length and settlement occurred from day 11 onward, although swimming larvae were still vis- ible in the medium up to the 30th day. DISCUSSION Induction of Spawning LOOSANOFF & Davis (1963) list methods used in the induction of spawning in 19 bivalves. They pointed out that where some species responded to thermofluctuation as a stimulus to spawning, others needed the additional stimulus of a sperm or egg suspension. In some instances special methods had to be employed, such as pricking the Table 2 Summary of larval dimensions of Brachidontes modiolus. Length Height Depth Stage (um) (um) (um) Early straight-hinge stage 96 76 2 Straight-hinge stage 96-176 67-152 *_100 First indication of umbo 168 120-136 72-88 Umbo stage 168-221 120-184 72-124 * Not measured. Page 58 The Veliger, Vol. 26, No. 1 Figure 4 Pediveliger of Brachidontes modiolus. adductor muscle in Mytilus edulis. Other species, like Modiolus demissus, did not respond to any treatment; LOOSANOFF & Davis (1963) were unsuccessful in induc- ing this species to spawn. WILSON & HODGKIN (1967) failed to induce spawning of Brachidontes cf. variabilis in the laboratory. CHANLEY (1970) was finally successful in causing spawning in B. recurvus by placing the mussels in sea water with temperatures fluctuating between 20 and 32°C. Previous attempts to induce spawning in this animal, by adding stripped gametes to the water or stretching or injuring the adductor muscles, had proved unsuccessful. Ripe adults of B. granulata spawned when placed in filtered sea water at 16°C after being held in an incubator at 6°C for 12 h. The account presented above demonstrates the individuality of the response of different species to various spawning stimuli. Stripping of the gonad of Brachidontes modiolus failed to produce viable gametes for the same reason that the eggs obtained through the mediation of 0.5 M KCl are not fertilizable. RAVEN (1958) stated that maturation of the eggs of most mollusks may begin spontaneously, in- dependent of fertilization, e.g., after spawning in sea water, and continues until metaphase of the first maturation di- vision. Unfertilized eggs are blocked at this stage. Matu- ration is evidenced in part by the dissolution of the ger- minal vesicle. The ova of B. modiolus are not mature while still in the gonad. Eggs from this species, when obtained by stripping or by injection of KCl, possessed an intact germinal vesicle and were therefore immature and were incapable of being fertilized without further treatment (e.g., the addition of NH,OH). In contrast, the eggs extruded from mussels stimulated by the addition of hydrogen per- oxide had started the maturation process and were easily fertilizable. Mors et al. (1977) found that the hydrogen peroxide- induced spawning of the abalone Haliotis rufescens may have resulted from a “direct activation of the enzyme- catalyzed synthesis of prostaglandin endoperoxide.” Pros- taglandin endoperoxide (PGEP) is produced from arachi- donic acid through a series of reactions, the first step of which is catalyzed by the enzyme fatty acid cyclooxygen- ase (=PGEP synthetase). These investigators showed that abalone eggs and gonads from ripe animals of both sexes contained large quantities of cyclooxygenase, and further that hydrogen peroxide directly increased the rate of the reaction catalyzed by the PGEP-forming cyclooxygenase from reproductive cells of the abalone. PGEP synthetase has been implicated in the control of spawning not only in abalones, but in Mytilus californianus (MORSE et al., 1977) and sea urchins (MorsE et al., 1978). It is probable that a similar control of spawning exists in Brachidontes modiolus. IWATA (1952) found that electrical stimulation induced maturation of eggs of Mytzlus edulis. He suggested that the stimulus was mediated by the ovary, probably by the secretion of some substance that caused the ova to mature. According to IWATA (1952), spawning in Mytilus appeared to follow automatically as soon as the eggs begin the maturation division, and therefore spawning in this animal depended entirely on whether the maturation pro- cess had taken place. A number of questions arise from a comparison of the results of MorsE et al. (1977) and Iwa- TA (1952). Could the proposed substance of Iwata (1952) be related to PGEP synthetase, or to any of the enzymes or intermediate products of PGEP production? Converse- ly, does PGEP synthetase act to promote maturation of oocytes and hence spawning? The role of PGEP synthetase in the control of spawning in Brachidontes modiolus would be of interest for further study. Morse et al. (1977) stated that the fatty acid-cy- clooxygenase reaction may be potentially rate-limiting in the physiological sequence of reactions leading to spawn- ing and therefore possibly under hormonal and/or neural control. Thus a knowledge of the levels of this enzyme in B. modiolus, together with information on environmental factors, could help in determining the ultimate factors con- trolling the release of gametes in this species. Larval Survival and Development Larvae survived well in unaerated cultures, and change of water every third day proved adequate for removing toxic waste products before harmful levels were reached. Some morphological malformations were observed in some of the larvae, for example concavity of the hinge line or un-equal valve growth; but it could not be determined to what extent these were due to culture conditions. Abnor- mality of the hinge line as seen in Brachidontes modiolus was also reported for other bivalves (LOOSANOFF & DAVIS, 1963). In addition, these authors described another type A. Fields & E. Moore, 1983 Page 59 PAM AAM Figure 5 Diagrams of conspicuous features of the internal anatomy of Brachidontes modtolus. a. straight-hinge veliger; b. 3- day old larva; c and d. 6-day old larvae; e. spat. Scale bar for a, b, c, and d is 40 um long; for e, it is 60 wm long. AAM, anterior adductor muscle; DG, digestive gland; ES, eye spot; F, foot; Gi, gills; I, intestine; PAM, posterior adductor muscle; PRM, pedal retractor muscle; St, stomach; U, umbo; V, velum. of larval abnormality in which there was no clear-cut anatomical malformation, but rather the larvae were un- able to feed. Such larvae developed to the straight-hinge stage but grew no further and eventually died. It is prob- able that the larvae found in the bottom layers of cultures of B. modiolus suffered from this type of “feeding” abnor- mality. LOOSANOFF & Davis (1963) reported that this type of abnormality in Mercenaria mercenaria was related in some instances to the type of food fed to the larvae. This aspect was not investigated in B. modiolus. That larvae of various species of bivalves show variation in growth rate among individuals reared from the same spawn under similar conditions is well established (LOOSANOFF & Davis, 1963). In addition to individual variation in growth rate, larvae of Brachidontes modiolus settle at different times after 11 days. LOOSANOFF & Davis Page 60 Whe Veliger, Vol] 2Z6@Noml Table 3 Summary of the major features of the larvae of Brachi- dontes modiolus, B. recurvus*, and B. granulata**. Size (um) Stage or B. B. B. distinctive feature modiolus recurvus granulata 67-77 62-68 63-73 Unfertilized egg Straight-hinge veliger 96-176 90-165 177+ Appearance of umbo 168 135 150 Conspicuous eye spot 184 180 — Pediveliger (first appearance of functional foot) 180 165 200 * Data taken from CHANLEY (1970). ** Data taken from CAMPOS & RAMORINO (1980). + Length of prodissoconch. (1963) found that the number of days needed for larvae of the same cultures to reach settling stage was not sharply defined, even for a culture reared under standard well- controlled conditions. The observed extended settling pe- riod of B. modiolus may be explained by the probability that glass is not the most desirable substrate for settling, as well as it may be related to factors governing settling density. Larvae possessing a functional velum could be seen swimming in the water column long after others in the same flask had settled and attached themselves by byssal threads. BAYNE (1965) discussed the delay of meta- morphosis in Mytilus edulis and showed that the larvae become capable of attachment and metamorphosis some time before they would attach to glass, and that the ma- jority died without attachment. The delay of metamor- phosis of laboratory reared larvae has been previously reported for other bivalves: CULLINEY (1971) for Lithoph- aga bisulcata, CAMPOS & RAMORINO (1980) for Brachi- dontes granulata. BAYNE (1965) established that the larvae of Mytilus edulis settled preferentially on filamentous al- gae (primary settlement). The larvae of B. modiolus did not settle on algae provided in the culture flasks. However, in another experiment not described here, a fine filament of artificial fiber of unknown origin precipitated mass gre- garious settlement of larvae of B. modiolus. In the wild population no sign of attachment to algae by the larvae of B. modiolus was recorded in this study. ; CHANLEY (1970) described the common characteristics of larval Mytilidae. The larvae of Brachidontes modiolus exhibited many of the features listed. The hinge line was long in relation to larval length and increased in size with growth of the larvae. The hinge possessed a mytilid den- tition, having a series of small teeth along the length of the hinge, with larger teeth towards the end. The umbo was late in developing and remained indistinct. In fact, it was not until after settling that the umbo of B. modiolus became pronounced. However, the larvae of B. modiolus set at a smaller size (180-221 um) than most mytilid larvae (300 um, in CHANLEY, 1970) as do larvae of other species of this genus (Table 3). A comparison of the development of the larvae of Brachidontes modiolus, B. recurvus, and B. granulata (Ta- ble 3) reveals differences in size at first appearance of the umbo and in minimum sizes of the pediveligers. The shape of the straight-hinge veliger of B. modiolus more closely resembles that of B. recurvus than that of B. granulata. The shape of this larval stage of B. granulata is more rounded than that of the former two. If, however, a com- parison is made of later stages, the position is reversed, with the larvae of B. granulata being more similar in shape to that of B. modiolus than B. recurvus. The most notable difference is in the developmental times of the three species. Pediveligers first appeared after 6 days and settlement of the larvae of B. modiolus occurred from the 11th day on- wards, similar to the time of the first appearance of ped- iveligers of B. recurvus. CHANLEY (1970) does not report on the time of settlement of B. recurvus. The pediveligers of B. granulata do not appear until 55 days after fertiliza- tion, and CAMPOS & RAMORINO (1980) report no evidence of metamorphosis up to 33 days after this. This lack of metamorphosis in the larvae of B. granulata was attributed to the unavailability of a substrate suitable for settlement of the pediveligers. The variation in the developmental times may result from the different temperature regimes to which the larvae were exposed. The larvae of B. granu- lata were reared at 12-16°C, those of B. recurvus at 25°C, and the larvae of B. modiolus at 28°C. BAYNE (1976) pre- sented data from various authors on the rates of cleavage and early development of Mytilus edulis at different tem- peratures (8-22°C). Those results showed that develop- ment proceeded more rapidly at the higher temperatures. In general the developmental progress and morphology of the larvae of B. modiolus closely resemble those of the larvae of the other two members of the genus Brachidontes described in the literature. In the early stages the larvae of B. modiolus bear strongest resemblance to those of B. recurvus while at later stages the more obvious resem- blance is to the larvae of B. granulata. These findings tend to reinforce the family or group relationships of these three species. The most outstanding difference is in the devel- opmental times, with B. modiolus settling the earliest of the three. This difference has been attributed to different temper- ature regimes at which the larvae were reared. Brachi- dontes recurvus was reared at 25°C, B. granulata at tem- peratures between 12° and 16°C, and B. modiolus at 28°C. ACKNOWLEDGMENTS The National Marine Fisheries Service in Galveston, Texas, supplied the monocultures used for feeding the larvae in this study. The work was financially supported by grants from the OAS (Barbados Office) and Board for Postgraduate Studies of the University of the West Indies. A. Fields & E. Moore, 1983 LITERATURE CITED Appott, R. T. 1974. American seashells. The marine molluscs of the Atlantic and Pacific coasts of North America. Van Nostrand Reinhold, New York. 663 pp. Bayne, B. L. 1965. Growth and delay of metamorphosis of larvae of Mytilus edulis. Ophelia 2:1-47. BAYNE, B. L. (ed.). 1976. Marine mussels: their ecology and physiology. Cambridge Univ. Press, London. 506 pp. Campos, B. M. & L. RAMOoRINO. 1980. Larval and early ben- thic stages of Brachidontes granulata (Bivalvia: Mytilidae). Veliger 22(3):277-281. CHANLEY, P. 1970. Larval development of the hooked mussel, Brachidontes recurvus Rafinesque (Bivalvia: Mytilidae), in- cluding a literature review of larval characteristics of the Mytilidae. Proc. Nat. Shellfish. Assoc. 60:86-94. CHANLEY, P. & J. D. ANDREWS. 1971. Aids for the identifi- cation of bivalve larvae of Virginia. Malacologia 11(1):45- i1i9), CHANLEY, P. & M. CHANLEY. 1980. Reproductive biology of Arthritica crassiformis and Arthritica bifurca, two commensal bivalve molluscs (Leptonacea). N.Z. J. Mar. Freshwater Res. 14(1):31-43. CHANLEY, P. & W. A. VAN ENGEL. 1969. A three dimensional representation of measurement data. Veliger 12(1):78-83. Cor, W. R. 1953. Resurgent populations of marine inverte- brates and their dependence on ocean currents and tidal currents. Ecology 34:225-229. CULLINEY, J. L. 1971. Laboratory rearing of the larvae of the mahogany date mussel, Lithophaga bisulcata. Bull. Mar. Sci. 21(2):591-602. Iwata, K. S. 1952. Mechanism of egg maturation in Mytilus edulis. Biol. J. Okayama Univ. 1:1-11. Kuropa, T., T. Habe, K. OYAMA & C. LENT. 1971. The sea shells of Sagami Bay. Biol. Labor. Imp. Household, Ma- ruzen Co. Ltd., Tokyo. (Billing edition, English.) 489 pp. LoosanorF, V. L. & H. C. Davis. 1963. Rearing of bivalve Page 61 molluscs. /n: F. S. Russel (ed.), Advances in marine biology. Academic Press, New York. Vol. 1. pp. 1-136. LoosANoFF, V. L., H. C. Davis & P. CHANLEY. 1966. Di- mensions and shapes of larvae of some marine bivalve mol- luscs. Malacologia 4(2):351-435. McLean, R. 1951. The Pelecypoda or bivalve molluscs from Puerto Rico and the Virgin Islands. Scient. Surv. Puerto Rico, XVII, Part 1:183 pp. Morse, D. E., H. DuNcAN, N. HOOKER & A. Morse. 1977. Hydrogen peroxide induces spawning in molluscs, with the activation of prostaglandin endoperoxide synthetase. Science 196:298-300. Morse, D. E., M. KAYNE, M. TIDYMAN & S. ANDERSON. 1978. Capacity for biosynthesis of prostaglandin-related com- pounds: distribution and properties of the rate-limiting en- zyme in hydrocorals, gorgonians and other coelenterates of the Caribbean and Pacific. Biol. Bull. 154:440-452. RAVEN, C. P. 1958. Morphogenesis. The analysis of molluscan development. Pergamon Press, New York. 310 pp. SCHELTEMA, R. S. 1971. The dispersal of the larvae of shoal- water benthic invertebrate species over long distances by ocean currents. /n: D. J. Crisp (ed.), European marine bi- ology symposium. Cambridge University Press, New York. Vol. 4. pp. 7-28. STANLEY, S. M. 1970. Relation of shell form to life habits in the Bivalvia (Mollusca). Mem. Geol. Soc. Am. 125:496 pp. STANLEY, S. M. 1972. Functional morphology and evolution of byssally attached bivalve molluscs. J. Paleont. 46:165- 212. WILSON, B. R. & E. P. HODGKIN. 1967. A comparative ac- count of the reproductive cycles of five species of marine mussels (Bivalvia: Mytilidae) in the vicinity of Fremantle, Western Australia. Aust. J. Mar. Freshwater Res. 18:175- 203. YosHiDA, H. 1937. On the pelagic larvae and young of Brach- idontes senhausi (Reeve). Venus 7:121-128. The Veliger 26(1):62-68 (July 1, 1983) THE VELIGER © CMS, Ine., 1983 A Comparison of the Species Richness and ‘Trophic Roles of Gastropods and Chitons on Rocky Shores of ‘Temperate and ‘Tropical West America by ALAN C. MILLER Department of Biology, California State University, and Southern California Ocean Studies Consortium, Long Beach, California 90840 Abstract. I examined local species richness and trophic roles of gastropods and chitons in rocky intertidal zones in Oregon, U.S.A., and in Costa Rica, Central America. Local species richness was quite variable. Tropical sites did not always have more species nor was there a significant difference in local species richness between regions. However, the species pool for the tropical transects (75) was larger than for Oregon (44). The density of individuals per square meter was much greater in the temperate sites. Between-habitat diversity differences may account for most of the increase in over-all gastropod species richness in Costa Rica. A regression analysis suggests that, among the variables tested, only the number of gastropod refuges explains a significant amount of the variation in local species richness among all sites. An increase in carnivorous and carrion-eating snail species accounts for most of the expansion in species richness in Costa Rica. This increase is principally in species that eat different food resources than do temperate zone snails (e.g., sipunculans and small fishes) and species that feed on polychaetes, gastropods, and dead or injured animals. INTRODUCTION MANny TAXA OF plants and animals, including gastropods (THORSON, 1952; KOHN, 1971), have an increasing num- ber of species along a decreasing latitudinal gradient (PIANKA, 1966). This ecological phenomenon has been the focus of many studies, and several hypotheses based on physical and biological processes have been developed to explain the generation and maintenance of this pattern (PIANKA, 1966; PIELOU, 1975; HusTOoN, 1979). One interesting question about this diversity pattern is whether specific tropical habitats have a greater diversity than similar habitats in the temperate zone. If so, what ecological processes allow more species to co-occur in the tropical habitats? Are there simply additional, different microhabitats present within these “similar habitats” in the tropics compared to the temperate zone? Are impor- tant resources (such as food and habitat) more finely par- titioned? Have physical and/or biological disturbances (2.e., ' Contribution Number 14. predation) kept species populations reduced in size so that episodes of interspecific competition, if they occur at all, are brief? The general purpose of this study was to examine quantitatively rocky intertidal areas in Oregon and Costa Rica in order to provide answers to the following ques- tions: A. How does the gastropod and chiton species richness in rocky intertidal areas differ between temperate and tropical regions of West America? B. What physical factors might explain some of the variation in species richness between these areas? C. Do tropical carnivorous gastropods use food re- sources differently than temperate snails? SITE DESCRIPTIONS ann METHODS Table 1 describes the study sites chosen to sample areas with different wave exposures and topographies. The Or- egon sites were studied from April to August, during 1970, 1971, and 1972, the Costa Rican sites, in January and February, 1972. A. C. Miller, 1983 Page 63 Table 1 Description of the exposed-coast rocky intertidal study sites in Oregon and Costa Rica. Tidal Transect name Locality range (abbreviation) General description Cape Arago, Oregon, U.S.A. 2.1m North Cove (NC) cliff with high splash pool Middle Cove 1 (MC1) ledge area with no loose boulders Middle Cove 2 (MC2) boulder field South Cove (SC) bedrock with ledges and loose boulders Cape Blanco, Oregon, U.S.A. 2.2 m Cape Blanco 1 (CB1) bedrock with ledges and loose boulders Cape Blanco 2 (CB2) bedrock with ledges and loose boulders Playa Hermosa, Guanacaste, Costa Rica 2.7 m Playa Hermosa 1 (PH1) bedrock with ledges and loose boulders Playa Hermosa 2 (PH2) bedrock with ledges and loose boulders Playa Hermosa 3 (PH3) bedrock with ledges Samara, Guanacaste, Costa Rica Dell soa Samara 1 (S1) flat bedrock, ledges, and loose boulders Samara 2 (S2) flat bedrock, ledges, and much algal turf Samara 3 (S3) flat bedrock with ledges and tidepools Quépos, Puntarenas, Costa Rica 2.5 m Quépos 1 (Q1) large boulders Quépos 2 (Q2) At each site, I chose a point at the high-tide drift line and placed the base point for a transect a random number (between O and 50) of meters from this point along the drift line. I stretched a metric tape from the base point toward the water’s edge at low tide and placed quadrats (1 m X 1 m) across the transect line at arbitrarily set in- tervals according to the length of the transect and the time available to complete the work. For instance, South Cove was sampled every meter along the transect line, while at Samara 3, the interval was every 5 m. Within each quad- rat I counted all snails, chitons, and nudibranchs and not- ed their positions (under rocks, on top of rocks, etc.). I did not include vermetid snails and octopuses. Due to difficulties in differentiating the numerous small Fossarus spp. from the similarly shaped and marked Littorina as- pera in Costa Rica, I lumped these species. I had similar problems distinguishing Collisella paradigitalis from C. digitalis in Oregon so I lumped them together. When Fos- sarus spp., Littorina spp., and limpets occurred in high numbers, I counted them within six 10-cm? quadrats placed randomly within the 1-m? quadrat and extrapolated their densities to 1 m* from the average of these counts. I quantified the intertidal height above O-m low tide at 20-cm intervals on the transect line within each sampled quadrat using a simple surveying technique (placing a pole on the transect line and sighting a horizontal line from a marked position on this pole, a flag on a stationary pole a known distance above the O-m tide point, and the horizon). Based on C. M. Yonge’s method of zonation (RICKETTS et al., 1968), I used 174 cm above mean lower low water (O-m tide) as the separation height between high and mid intertidal zones for Oregon. Since tropical intertidal areas are not as well studied and the tidal datum is different (mean low water spring tides), I based the separation height between high and mid intertidal zones in Costa Rica on gastropod assemblages. I determined it bedrock with ledges and loose boulders to be the lowest height reading (120 cm) of the quadrats that contained only those gastropod species (Littorina as- pera, L. modesta, Nerita funiculata, N. scabricosta, Planaxis spp., Tegula pellisserpentis, Purpura pansa, and Acanthina brevidentata) whose distributions are limited to the high intertidal zone and the extreme upper portions of the mid intertidal zone (KEEN, 1971; SPIGHT, 1976; GARRITY & LEVINGS, 1981). I quantitatively or qualitatively estimated some physi- cal factors that might influence species richness. I quali- tatively estimated wave exposure (0.0 = lowest exposure to waves; 1.0 = greatest exposure) for each transect by observing surf conditions and by assessing the number and size of loose rocks. Using the average height measurements for each quadrat, I computed the proportion of quadrats that occurred below the high intertidal zone in each tran- sect (Quépos 1 was excluded due to sampling problems). As a measure of the availability of gastropod refuges, I computed for each transect the proportion of quadrats in which snails were found on the undersides of rocks or ledges. Since all of the values of these variables vary between O and 1, I angularly transformed them before computing a stepwise multiple regression with species richness. I in- cluded a latitude variable in this analysis, computed as a ratio of the transect’s latitude divided by 90°; I also an- gularly transformed this ratio. The diets of the carnivorous and carrion-eating gastro- pods were determined from a literature review (MILLER, 1974), from field observations, and by collecting individ- uals in the field, isolating them in containers of sea water for up to 24 h, and then microscopically examining fecal pellets for prey hard parts. I assigned the carnivorous snail species to various diet categories based on the percentage of occurrence of specific food items in their diets. If 75% or more of the food items in a species’ diet were in one Page 64 The Veliger, Vol. 26, No. 1 Table 2 A comparison of the species richness and density of gastropods and chitons in the Oregon and Costa Rican transects. Oregon NC MC1 MC2~ SC CB1 Number of quadrats 8 14 14 5 19 Species richness 13 18 28 36 19 XK + SEM 22.7 + 3.34 U-test Number/m? 61 83 171 188 143 X + SEM 131.0 + 20.2 U-test diet category, I assigned that species only to that diet cat- egory. If less than 75% of the feeding observations were in one category, I assigned the species to the two diet categories that made up at least 70% of the diet. If the two most common diet categories comprised less than 70% of the observations, I included the third highest diet cat- egory. MILLER (1974) lists the chiton and gastropod species discussed in this study as well as their abundances and the diets of the carnivorous and carrion-eating snails. These lists may be obtained from the author. Voucher specimens reside in the author’s collection. RESULTS Species Richness The number of species of gastropods and chitons is extremely variable in both Oregon and Costa Rica (Table 2). The number of species found in a local area does not differ significantly between regions. The density of gas- tropods and chitons is, however, significantly greater in the Oregon transects. The species pool for the Oregon samples (44) is much smaller than that for the Costa Rican samples (75); thus, there is an increase in the number of gastropod species with decreasing latitude when large geographic areas are considered. Physical Factors In an attempt to ascertain what physical factors might explain some of the variation in species richness among the transects, I compiled a list of potentially important variables (Table 3) that I was able to quantify or quali- tatively estimate. I chose wave exposure (WAVES) as a variable since the degree of exposure of an intertidal area to wave shock is important in determining what kinds of organisms occur there (RICKETTS et al., 1968). I picked the proportion of Costa Rica PH1 PH2 PH3 S1 S2 $3 Qi Q2 14 15 13 23 25 32 14 9 27 28 15 33 34 29 38 26 28.8 + 2.43 N.S. 341 47 105 49 28 34 46 18 83.5 + 37.9 U = 40, P < 0.05 samples from below the high intertidal zone in a transect (HT) as a variable because there is a trend of increasing species richness with a decrease in intertidal height (JOHNSON, 1970; MILLER, 1974; SpIGHT, 1977). The presence of undersides of rocks, boulders, or ledges in- creases the topographic complexity of an area and pro- vides microhabitats for the snails that are more sheltered from mortality factors like desiccation and wave shock than areas with only flat bedrock (KOHN & LEVITEN, 1976; MENGE & LUBCHENCO, 1981). For this reason, I calculated a refuge variable (REF) by computing the pro- portion of quadrats in each transect that had undersides of rocks or ledges occupied by gastropods. I included lat- itude (LAT) as a variable since it could include some biological and physical parameters that are different in the tropics compared to the temperate zone, but which I did not measure (PIANKA, 1966; Huston, 1979). Although not a physical factor, I also used the number of sampled quadrats (QUADS) as a variable since, at least for Oregon, species richness (S) may be related to it in my study (S = —8.66 + 25.8 log QUADS; t = 3.21, P < 0.025; R? = 0.72). For Costa Rica, none of the regression analyses using log-transformed or untransformed values of S and QUADS resulted in a significant regression or an R? value greater than 0.14. A comparison of each of the variables in Table 3 for a significant difference between Oregon and Costa Rica in- dicated that only “Latitude” was significantly different (U = 48, P < 0.001). I did a stepwise multiple regression using these vari- ables to find out which ones might explain significant amounts of the variation in species richness (S) among all of the transects (excluding Quépos 1). The largest total adjusted R’, 0.83, was obtained using the logarithm of species richness; the only significant independent variable was the gastropod refuge variable (F = 6.0, P < 0.05). The regression equation was: log S = 0.004 QUADS — 0.003 LAT + 0.003 REF + 0.001 HT — 0.002 WAVES + 1.319. A. C. Miller, 1983 Page 65 Table 3 Summary of the physical factor measurements and estimates, the latitudes, and the sample sizes of the Oregon and Costa Rican transects. Proportion of Proportion of quadrats below quadrats with Number of high intertidal gastropod 1-m? quadrats Transect Wave exposure zone refuges Latitude sampled Oregon North Cove 1.00 0.27 0.00 43,32° 8 Middle Cove 1 0.80 0.38 0.14 43.32° 14 Middle Cove 2 0.40 0.75 1.00 43.32° 14 South Cove 0.40 1.00 0.53 43.32° 51 Cape Blanco 1 0.80 0.70 0.26 42.50° 19 Cape Blanco 2 0.80 0.92 0.54 42.50° 13 Costa Rica Playa Hermosa 1 0.60 0.57 0.50 10.50° 14 Playa Hermosa 2 0.60 0.84 0.40 10.50° 15 Playa Hermosa 3 0.80 0.60 0.08 10.50° 13 Samara 1 0.20 0.89 0.35 9.87° 23 Samara 2 0.40 0.97 0.60 9 :8ile 25 Samara 3 0.20 1.00 0.19 9.87° 32 Quépos 1 0.80 — 0.79 9.45° 14 Quépos 2 0.80 0.94 0.67 9.45° 9 The Between- and Within-Habitat Components of Diversity Total diversity in a large geographic region consists of two components: within-habitat diversity and between- habitat diversity (MACARTHUR, 1965). If within-habitat diversity has contributed significantly to the larger species pool in Costa Rica, then similar habitats should support more diverse assemblages of gastropods and chitons there than in Oregon. Two lines of evidence in my study suggest that the within-habitat component is not as important as the between-habitat component for rocky intertidal habi- tats. First, transects in Oregon and Costa Rica that are sim- ilar in wave exposure, proportion of quadrats below the high intertidal zone, and proportion of quadrats with gas- tropod refuges (Table 3) have similar diversities (Table 2)—e.g., South Cove and Samara 2; Cape Blanco 2 and Quépos 2. Second, the number of species found in indi- vidual 1-m? samples is not significantly different between the two regions (U = 37.5, P > 0.05). These values ranged from 4.1 to 7.8 (X =5.8 + 0.59 SEM) in Oregon and from 2.8 to 5.3 (X = 4.4 + 0.29 SEM) in Costa Rica. Between-habitat diversity appears to be different be- tween regions. To assess this, I compared for each region the percentages of (1) omnipresent species, (2) species found in all transects except one, and (3) species unique to one transect (Table 4). These results suggest that tran- sects in Oregon have more species in common and have fewer unique species. This may mean that the transects are sampling similar habitats in Oregon and more differ- ent habitats in Costa Rica. Comparison of the percentages of the species in each transect that are also found in one or more other transects (SALE, 1980) also suggests an increase in the between- habitat component in Costa Rica. Oregon transects have significantly greater percentages than Costa Rica (U = 44.5, P < 0.01). The average percentage is 93.1% for Or- egon transects and 88.3% for Costa Rican transects. Trophic Structure In the trophic structure analysis, I lumped carnivorous and carrion-eating snails together since I could not tell whether snails known to feed on carrion would also eat living prey. Some snails that normally attacked living prey were observed eating dead animals. Table 4 A comparison of the percentage of species found in all transects, the percentage in all transects except one, and the percentage unique to one transect. Oregon Costa Rica % omnipresent 9.1% 2.7% % in all but one transect 34.1% 10.7% % unique 29.5% 36.0% The Veliger, Vol. 26, No. 1 Table 5 Comparison of the trophic characteristics of the assemblage of species in each transect. Page 66 NG MCi MC2 SC _ CBI No. herbivorous species 10 12 20 22 12 X + SEM SS OREe220) No. carnivorous and carrion-eating species 3 6 8 14 7 X + SEM Tell 22 1.5 Proportion of carnivorous and carrion-eating individuals 06 03 .03 .06 .06 X + SEM 0.05 + 0.01 CB2 PH1 PH2 PH3 Si S2 S3 Qi Q2 14 14 7 10 22 21 13 20 13 16.2 + 1.6 8 13 11 5 11 13 16 18 13) 12,5 ae il.4 05 07 aSi/ 05 alts 2) 54 Lai .28 0.26 + 0.06 The assignment of some of the grazing mollusks to the carnivorous or herbivorous trophic level is difficult since some species may be indiscriminate grazers, feeding on both plant and animal material. For instance, some species of Calliostoma graze on sessile animals (MILLER, 1972; PERRON, 1975); but I did not find animal remains in the feces of seven C. ligatum from the Oregon transects. Ex- cept for Placiphorella velata, which entraps and eats mobile crustaceans (MCLEAN, 1962; Miller, personal observa- tion), most chitons graze on plant material (references in Morais e¢ al., 1980). However, there is evidence that some species of Mopalia graze on sessile animals and algae (BARNAWELL, 1960). In this study I have considered all chitons except P. velata as herbivores, C. ligatum as an herbivore, and all species of Cypraea as carnivores. There is no significant difference in the number of her- bivorous species found in transects in Oregon compared to Costa Rica (Table 5; U=29, P > 0.05). However, Costa Rican transects have significantly more carnivorous and carrion-eating species than Oregon (U = 38, P < 0.05). The Costa Rican transects also have a larger pro- portion of carnivorous and carrion-eating individuals than Oregon transects (Table 5; U = 44.5, P < 0.01). The pro- portion of carnivorous and carrion-eating species in the species pool is 0.41 in Oregon and 0.52 in Costa Rica. The Diets of Carnivorous and Carrion-Eating Snails Since the increase in species richness in Costa Rica occurs primarily in carnivorous and carrion-eating species, I examined the diets of these snails to compare food-re- source use between temperate and tropical species (Table 6). The main differences are threefold increases in the number of snail species eating polychaetes and gastropods in Costa Rica compared to Oregon and the exploitation by tropical snails of sipunculans and fishes, two food types not used by Oregon gastropods. There is also a large increase in species eating dead material in the tropics, due mainly to the numerous species of Columbellidae. Most columbellids are attracted to dead and/or injured prey (Marcus & MARCUS, 1962; SPIGHT, 1976; HATFIELD, 1979; Miller, personal observation), but not enough work has been done to show that columbellids are only carrion-eaters. DISCUSSION I examined the gastropod and chiton species richness at several similar, exposed-coast rocky intertidal zone sites in Oregon and Costa Rica. Local species richness is quite variable (Table 2). The tropical transects were not always richer nor was there a significant difference in the number of species found in Oregon and Costa Rican transects. However, the density of gastropods and chitons was sig- nificantly greater in the temperate areas (Table 2). The total number of different gastropod and chiton species in all Costa Rican transects (75) is much larger than in all Oregon transects (44). These results agree with SPIGHT’s (1977) comparisons of prosobranch species richness at two Washington State rocky intertidal locations and one Costa Rican site. The Table 6 Number of carnivorous and carrion-eating gastropod and chiton species feeding in each diet category. * Indicates that species exist but were not found in my samples. Live, mobile Coelen- Sipun- Poly- crusta- Bar- Bi- Gastro- Bryo- Echino- Tuni- Para- Dead Sponges terates culans chaetes ceans nacles valves pods zoans derms cates Fishes sitic animals Oregon 3 2 0 4 1 5 3 1 1 0* 1 0 2 3 Costa Rica 1 1 2 13 0* 7 3 4 0 1 1 1 1 10 A. C. Miller, 1983 difference in the number of tropical prosobranch species in Spight’s study (79) and mine (61) may be due to his deliberate sampling of as many habitats as possible at this one site (SPIGHT, 1976, 1977). It is possible that the Cape Arago region in Oregon, where I found 25 prosobranch species, has more habitats than the more wave-protected Puget Sound area (Shaw Island) in Washington, where Spight found 16 prosobranch species. The between-habitat component of species diversity ap- pears to contribute more than the within-habitat compo- nent to the larger total species richness that I found in Costa Rica. This suggests that I was sampling a greater diversity of habitats in Costa Rica and agrees with SPIGHT’s (1977) conclusion that there are more niches (defined by substratum and intertidal height) in Costa Rica with no temperate zone counterparts. Although many different physical and biological factors may control local species richness, the only significant variable of those that I measured or estimated as explain- ing significant amounts of variation in species richness (Table 3) was the presence of gastropod refuges. The availability of the undersides of rocks and ledges as shel- ters from wave shock, UV radiation, insolation, desicca- tion, and large, visual predators has also been found by others to be important in controlling species diversity and population sizes (KOHN & LEVITEN, 1976; JOKIEL, 1980; MENGE & LUBCHENCO, 1981). The most obvious differences between the trophic struc- ture in Oregon and Costa Rica are the significant in- creases in the tropical region in the number of species of carnivorous and carrion-eating gastropods and in their proportion of the total individuals. The number of her- bivorous gastropod and chiton species per transect is also greater in Costa Rica than in Oregon, but not signifi- cantly. Three of the tropical carnivorous snail species use food resources present in the Oregon rocky intertidal but not eaten by Oregon snails (sipunculans and fishes) and a large number (13) prey on polychaetes. There are also substantial increases in the number of snail species eating living gastropods (4) and dead or injured animals (10) in Costa Rica. Although the more diverse carnivorous gastropod as- semblage in Costa Rica has large numbers of species shar- ing some food resources (at least on a large scale) and other species eating prey not consumed by Oregon species, I do not have enough dietary information to conclude that competition has been more important in structuring the gastropod community in Costa Rica than in Oregon com- pared to some of the other possible structuring forces, such as predation (MENGE & LUBCHENCO, 1981). ACKNOWLEDGMENTS The fieldwork formed part of my Ph.D. dissertation re- search. Travel and equipment funds were provided by a National Science Foundation Grant (GA 29340), a Soci- ety of Sigma Xi Grants-in-Aid of Research award, and Page 67 the Department of Biology, University of Oregon. I ap- preciate the advice and field assistance given to me by Drs. Peter Frank, Susanne Lawrenz-Miller, Fred Bernstein, John Cubit, Tom Niesen, and Tom Spight. I also thank Dr. James McLean for helping with the identification of many of the species and Alan Kohn and an anonymous reviewer for their comments on the manuscript. LITERATURE CITED BARNAWELL, E. B. 1960. The carnivorous habit among the Polyplacophora. Veliger 2:85-88. Garrity, S. D. & S. C. Levincs. 1981. A predator-prey in- teraction between two physically and biologically con- strained tropical rocky shore gastropods: direct, indirect and community effects. Ecol. Monogr. 51:267-286. HATFIELD, E. B. 1979. Food sources for Anachis avara (Col- umbellidae) and a discussion of feeding in the family. Nau- tilus 93:40-43. Huston, M. 1979. A general hypothesis of species diversity. Am. Nat. 113:81-101. JOHNSON, R. G. 1970. Variations in diversity within benthic marine communities. Am. Nat. 104:285-300. JoxieL, P. L. 1980. Solar ultraviolet radiation and coral reef epifauna. Science 207:1069-1071. KEEN, A. M. 1971. Sea shells of tropical West America: ma- rine mollusks from Baja California to Peru. Stanford Univ. Press, Stanford, Calif. 1066 pp. Koun, A. J. 1971. Diversity, utilization of resources, and adaptive radiation in shallow-water marine invertebrates of tropical oceanic islands. Limnol. Oceanogr. 16:332-348. Konun, A. J. & P. J. LEvITEN. 1976. Effect of habitat com- plexity on population density and species richness in tropical intertidal predatory gastropod assemblages. Oecologia 25: 199-210. MacArTuHur, R. H. 1965. Patterns in species diversity. Biol. Rev. 40:510-533. Marcus, Ev. & Er. Marcus. 1962. Studies on Columbelli- dae. Bol. Fac. Filos. Cienc. Letr. Univ. Sao Paulo, Brazil, No. 261. Zoologia No. 24:335-402. McLEan, J. H. 1962. Feeding behavior of the chiton Placi- phorella. Proc. Malacol. Soc. Lond. 35:23-26. MENGE, B. A. & J. LUBCHENCO. 1981. Community organi- zation in temperate and tropical rocky intertidal habitats: prey refuges in relation to consumer pressure gradients. Ecol. Monogr. 51:429-450. MILLER, A. C. 1972. Observations on the associations and feeding of six species of prosobranch gastropods on antho- zoans. In: G. J. Bakus (ed.), Marine studies on the north coast of Jamaica. Atoll Res. Bull. 152:4-5 (abstract). Mixer, A. C. 1974. A comparison of gastropod species di- versity and trophic structure in the rocky intertidal zone of the temperate and tropical West Americas. Ph.D. thesis Biology, Univ. Oregon, 143 pp. Morris, R. H., D. P. ABporr & E. C. HADERLIE. 1980. In- tertidal invertebrates of California. Stanford Univ. Press, Stanford, Calif. 690 pp. PERRON, F. E. 1975. Carnivorous Calliostoma (Prosobranchia: Trochidae) from the northeastern Pacific. Veliger 18:52-54. PIANKA, E. R. 1966. Latitudinal gradients in species diversity: a review of concepts. Am. Nat. 100:33-46. PrELou, E.C. 1975. Ecological diversity. Harper & Row, New York. 165 pp. RickeTTs, E. F., J. CALVIN & J. W. HEDGPETH. 1968. Be- Page 68 ihe Veligery Vola ZowNoml tween Pacific tides. Stanford Univ. Press, Stanford, Calif. 614 pp. SALE, P. F. 1980. The ecology of fishes on coral reefs. Ocean- ogr. Mar. Biol. Ann. Rev. 18:367-421. Spicut, T. M. 1976. Censuses of rocky shore prosobranchs from Washington and Costa Rica. Veliger 18:309-317. SPIGHT, T. M. 1977. Diversity of shallow-water gastropod communities on temperate and tropical beaches. Am. Nat. 111:1077-1097. TuHorsON, G. 1952. Zur jetzigen Lage der marinen Bodentier- Okologie. Verh. de Deutsch. Zool. Ges. 1951, Zool. An- zeiger Suppled. 16:276-327. The Veliger 26(1):69-70 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 Range Extensions of ‘Three Opisthobranch Mollusks to the San Diego—La Jolla (California) Ecological Reserve HANS BERTSCH! Instituto de Investigaciones Oceanologicas, Universidad Autonoma de Baja California, Apartado Postal 453, Ensenada, Baja California, Mexico TOM SMITH 917 Thomas, Apartment 19, San Diego, California 92109 SINCE JUNE, 1980, we have studied the subtidal fauna (by scuba diving) off La Jolla Shores Beach, La Jolla, Cali- fornia (32°51’-52’N; 117°15’-16’W) in the San Diego- La Jolla Ecological Reserve, within which collecting of any kind is prohibited. The bottom is a sandy habitat that gently slopes to about 12 m, and then precipitously, at 60° to 90° angles, drops into the La Jolla Branch of the La Jolla Submarine Canyon. The cliff faces and ledges of the canyon at 12.2 to 18.3 m consist of soft mudstone. Al- though this habitat supports a generally depauperate opisthobranch fauna, three of the opisthobranch species observed constitute southward range extensions. Aglaja ocelligera (Bergh, 1894) The reported range of Aglaja ocelligera (Cephalaspidea : Aglajidae) is from southern British Columbia, Canada, to Santa Cruz Island and Coal Oil Point, Santa Barbara County, California (SPHON & LANCE, 1968, p. 76; SPHON, 1972, p. 55; GOSLINER, 1980; BEHRENS, 1980, p. 32). The following records from La Jolla Canyon extend the range southward more than 250 km: (1) One specimen; 10 mm long, 9.1 m deep; 8 November 1981. "U.S. mailing address: 416 W. San Ysidro Blvd., K-229, San Ysidro, California 92073. (2) Three specimens; 15 mm, 18.3 m deep; 16 mm and eggs, 15.2 m deep, inside the stomach of Astropecten armatus; and 20 mm, 7.6 m deep; 25 December 1981. (3) One specimen; 20 mm, 13.1 m deep; 7 February 1982. The finding of one specimen inside the sea star Astro- pecten armatus Gray, 1840, is another record of predation upon an opisthobranch, a group considered to have few natural predators. MAUZEy et al. (1968) reported that in the Puget Sound region, another sea star, Crossaster pap- posus (Linnaeus, 1767), fed upon Aglaja diomedea. Flabellina price: (MacFarland, 1966) The known northern range limit of Flabellina pricei (Nudibranchia : Aeolidacea: Flabellinidae) (generic placement follows GOSLINER & GRIFFITHS, 1981) has been expanded from the type locality of Monterey, California (MacFarRLAND, 1966, p. 316), to Duxbury Reef, Marin County, California (GOSLINER & WILLIAMS, 1970, p. 177), and to Pearse Island, northern British Columbia, Canada (LAMBERT, 1976, p. 295). Its known southern range limit is Point Fermin, San Pedro, Los Angeles County, Cali- fornia (SPHON, 1972, p. 60). The following specimens extend its known range south- ward about 145 km to La Jolla Canyon: (1) Three specimens; 17 mm long, 10 m deep; 13 mm, 15.2 m deep; and 18 mm, 8.5 m deep; 14 May 1982. Page 70 Cerberilla mosslandica McDonald & Nybakken, 1975 The known distribution of Cerberilla mosslandica (Nu- dibranchia : Aeolidacea: Aeolidiidae) is restricted to the type locality of Monterey Bay, in central California, from 16 to 63 meters of water. The specimens collected by the Moss Landing Marine Laboratory benthic sampling team ranged from 5 to 9 mm in total length when alive. The following animal seen in the La Jolla Canyon represents a southern range extension of more than 550 km, its first reported occurrence elsewhere than the type locality, a new size record, and a slightly shallower bathy- metric distribution: (1) One specimen; 18 mm long, 12.2 m deep; 16 July 1982. ACKNOWLEDGMENTS The present study was completed under the auspices of the Benthic Ecosystems Project of Baja California, sup- ported by the Secretary of Public Education and the Au- tonomous University of Baja California, Mexico. LITERATURE CITED BEHRENS, D. W. 1980. Pacific coast nudibranchs: a guide to the opisthobranchs of the northeastern Pacific. Sea Chal- lengers, Los Osos, Calif. 112 pp. BERGH, L.S. R. 1894. Reports on the dredging operations off the west coast of Central America to the Galapagos, to the The Veliger, Vol. 26, No. 1 west coast of Mexico, and in the Gulf of California, in charge of Alexander Agassiz, carried on by the U.S. Fish Commission Steamer “Albatross,” during 1891, Lieut. Commander Z.L. Tanner, U.S.N., commanding. Bull. Mus. Comp. Zool. (Harvard) 25(10):125-233. GosLINER, T. M. 1980. Systematics and phylogeny of the Aglajidae (Opisthobranchia: Mollusca). Zool. J. Linn. Soc. 68:325-360. GOSLINER, T. M. & R. J. GRIFFITHS. 1981. Description and revision of some South African aeolidacean Nudibranchia (Mollusca, Gastropoda). Ann. S. Afr. Mus. 84(2):105-150. GosLINER, T. M. & G. C. WiLiiams. 1970. The opistho- branch mollusks of Marin County, California. Veliger 13(2): 175-180. LAMBERT, P. 1976. Records and range extensions of some northeastern Pacific opisthobranchs (Mollusca: Gastropo- da). Can. J. Zool. 54:293-300. MAcFARLAND, F. M. 1966. Studies of opisthobranchiate mol- lusks of the Pacific coast of North America. Calif. Acad. Sci. Mem. 6:xvi + 546 pp. Mauzey, K. P., C. BIRKELAND & P. K. DAYTON. 1968. Feed- ing behavior of asteroids and escape responses of their prey in the Puget Sound region. Ecology 49(4):603-619. McDOona.p, G. & J. NYBAKKEN. 1975. Cerberilla mosslandica, a new eolid nudibranch from Monterey Bay, California. Veliger 17(4):378-382. SPHON, G. G. 1972. An annotated checklist of the nudibranchs and their allies from the west coast of North America. Opisthobranch Newsletter 4(10/11):53-79. SPpHON, G. G. & J. R. LANcE. 1968. An annotated list of nudibranchs and their allies from Santa Barbara County, California. Proc. Calif. Acad. Sci. 36(3):73-84. The Veliger 26(1):71-72 (July 1, 1983) THE VELIGER © CMS, Inc., 1983 NOTES, INFORMATION & NEWS California Malacozoological Society, Inc. is a non-profit educational corporation (Articles of Incor- poration No. 463389 were filed January 6, 1964 in the office of the Secretary of State). The Society publishes a scientific quarterly, The Veliger. Donations to the So- ciety are used to pay a part of the production costs and thus to keep the subscription rate at a minimum. Donors may designate the Fund to which their contribution is to be credited: Operating Fund (available for current pro- duction); Savings Fund (available only for specified pur- poses, such as publication of especially long and signifi- cant papers); Endowment Fund (the income from which is available. The principal is irrevocably dedicated to sci- entific and educational purposes). Unassigned donations will be used according to greatest need. Contributions to the C.M.S., Inc. are deductible by do- nors as provided in section 170 of the Internal Revenue Code (for Federal income tax purposes). Bequests, lega- cies, gifts, and devises are deductible for Federal estate and gift tax purposes under section 2055, 2106, and 2522 of the Code. The Treasurer of the C.M.S., Inc. will issue suitable receipts which may be used by Donors to sub- stantiate their tax deductions. Subscription Rates and Membership Dues At its regular Annual Business Meeting on October 13, 1982, the Executive Board of the California Malacozoo- logical Society, Inc. decided to adjust the subscription rate for Members (‘dues’) to conform with the U.S. Postal Service regulations pertaining to second class mailing privileges to the required minimum of 50% of the regular (institutional) subscription rate, while holding the latter to the current US$37.50 per volume-year (July 1 to April 1). This means that the “dues,” which include a subscrip- tion to The Veliger, will increase a minimal 25¢ to $18.75. On the other hand, the Society is forced to increase the mailing charges for members and subscribers to $3.25 for all domestic addresses and to $6.00 for all foreign address- es (including Canada and Mexico). Membership is open to individuals only—no institu- tional or society memberships. When sending for mem- bership application forms, please include a self-addressed, stamped envelope (in the case of foreign members, the envelope and two International Postal Reply coupons). There is an initiation fee of $2.00. Membership renewals are due on or before April 1 each year. If renewal pay- ments are made after April 1 but before March 15 of the following year, there will be a re-instatement fee of $1.00. Members whose dues payments (including the re-instate- ment fee) have not been received by the later date will be dropped from the rolls of the Society. They may rejoin by paying a new initiation fee. The volume(s) published dur- ing the time a member was in arrears may be purchased, if still available, at the regular full volume price plus applicable handling charges. Back numbers of the current volume will be mailed to new subscribers, as well as those who renew late, on the first postal working day of the month following receipt of the remittance. The same policy applies to new members. The Veliger is not available on exchange from the Cali- fornia Malacozoological Society, Inc. We are willing to accept requests for expediting our journal via air mail; however, in that case we must ask for an additional payment of $8.00 in all cases where The Veliger goes to domestic addresses, and a deposit of $25.00 for all foreign addresses (including PUAS). Of course, we will carry forward as a credit toward the postage charges of the following year any amount over the actually re- quired postage charges. We call the attention of our foreign correspondents to the fact that bank drafts or checks on banks other than American banks are subject to a collection charge and that such remittances cannot be accepted as payment in full, unless sufficient overage is provided. Depending on the American banks on which drafts are made, such charges vary from a flat fee of $1.00 to a percentage of the value of the draft, going as high as 33%. Therefore, we recom- mend either International Postal Money Orders or bank drafts on the “Berkeley Branch of First Interstate Bank.” This institution has agreed to honor such drafts without charge. Similarly, we are unable to accept UNESCO cou- pons in payment, except at a charge of $4.25 (to reimburse us for the expenses involved in redeeming them) and at $0.95 per $1.00 face value of the coupons (the amount that we will receive for them). We regret that these charges must be passed on to our correspondents; however, our subscription rates and other charges are so low that we are absolutely unable to absorb additional expenses. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: C.M.S., Inc., Post Office Box 9977, Berkeley, CA 94709. Page 72 To Prospective Authors Manuscripts and correspondence pertaining to them should be sent to: Dr. D. W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616. Instructions for the preparation of manuscripts to be submitted to The Veliger are printed on the inside back cover of each issue. Following the instructions carefully will prevent needless delays in the processing of your manuscript. In addition a few comments regarding postal service and the costs thereof seem in order. Because of some dis- tressing experiences with the Postal Service in recent years, we now urge authors who wish to submit manuscripts to our journal to mail them as insured parcels, with insur- ance high enough to cover the complete replacement costs. If the replacement costs exceed $400.00, the manuscript should be sent by registered mail with additional insur- ance coverage (the maximum limit of insurance on parcel post is, at present, $400.00). We are unable to advise prospective authors in foreign countries and would urge them to make the necessary inquiries at their local post offices. We also wish to remind prospective authors that it is our policy not to acknowledge receipt of a manuscript unless a self-addressed stamped envelope is enclosed (two International Postal Reply Coupons are required from addresses outside the U.S.A.). Furthermore, we are un- able to return manuscripts (either for reworking or with the recommendation that they be submitted elsewhere) other than by ordinary surface mail. In the cases where the prospective recipient desires our communications sent as a registered article, we must expect advance payment of that fee ($3.25 regardless of destination). If a manu- script is withdrawn by the author, sufficient postage for return by certified mail within the U.S.A. and by regis- tered mail to other countries must be provided. We regret that we must insist on these conditions; however, the ex- orbitant increases in postal charges leave us no other choice. We do mail all page proofs of accepted manuscripts by air mail, and they should be returned by air mail also. If not, slow surface mail service (a minimum of 6 weeks” from European countries, 8 to 12 weeks from India and Africa) may make a delay in publication inevitable. The Society will absolutely not publish a paper unless proofs have been corrected and returned by the authors. Reprints, Page Charges, and Donations While it was hoped at the “birth” of The Veliger that a modest number of reprints could be supplied to authors free of charge, this has not yet become possible. Reprints are supplied to authors at cost, and requests for reprints should be addressed directly to the authors concerned. The ihe Vichiger, VoleZom Now Society does not maintain stocks of reprints and also can- not undertake to forward requests for reprints to the au- thor(s) concerned. Although many scientific journals assess page charges, the Executive Board of our Society, for the time being at least, wishes to avoid this possible financial handicap to the younger contributors. However, because of the high cost of halftone plates, a suitable contribution to reimburse the Society must be sought. Similarly, contributions will be sought from authors (or preferably their institutions) to help defray costs of line drawings and tables in excess of four simple ones. We emphasize, however, that requests for donations are totally unrelated to acceptance or rejec- tion of manuscripts for The Veliger. Acceptance is entirely on the basis of merit of the manuscript. Donations are to be made after publication, if at all, and are strictly on a voluntary basis. Nevertheless, donations by authors and readers are nec- essary for the continued good health of our Society, the stated aim of which is to disseminate new information in the field of malacology and conchology as widely as pos- sible at the lowest cost possible. Authors and readers are reminded that contributions to the C.M.S., Inc. are de- ductible by donors as provided in section 170 of the In- ternal Revenue Code (for Federal income tax purposes). A.S.Z. The 1983 Meeting of the American Society of Zoolo- gists and American Microscopical Society, Animal Be- havior Society, Biological Society of Washington, Crus- tacean Society, and International Society of Astacologists will be held in Philadelphia, Pennsylvania from Decem- ber 27-30, 1983. A call for papers went out in April and the deadline for abstracts is August 26, 1983. There are 15 symposia tentatively planned, several of which may be of interest to our readers. For detailed in- formation contact Mary Wiley, Business Manager, Amer- ican Society of Zoologists, Box 2739, California Lutheran College, Thousand Oaks, California 91360. Telephone: (805)492-3585. Another Donation from the San Diego Shell Club We recently received another contribution from the San Diego Shell Club. We express our gratitude for this con- tinued support, which is particularly encouraging in these inflationary times with their financial uncertainties. As in the past, contributions of this sort are added to our Endowment Fund, from which the income materially assists us in our endeavor to keep the subscription rate of the Veliger, as well as the membership dues in our Soci- ety, at the lowest possible level. Information for Contributors Manuscripts Manuscripts must be typed on white paper, 82” by 11”, and double-spaced throughout (including references, figure legends, footnotes, and tables). To facilitate the review pro- cess, manuscripts, including figures, should be submitted in triplicate. The first mention in the text of the scientific name of a species should be accompanied by the taxonomic authority, including the year. Underline scientific names and other words to be printed in italics. Metric and Celsius units are to be used. The sequence of manuscript components should be as follows in most cases: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, lit- erature cited, figure legends, figures, footnotes, and tables. The title page should be on a separate sheet and should include the title, author’s name, and address. The abstract should describe in the briefest possible way (normally less than 200 words) the scope, main results, and conclusions of the paper. Literature cited References in the text should be given by the name of the author(s) followed by the date of publication: for one author (SMITH, 1951), for two authors (SMITH & JONES, 1952), and for more than two (SMITH et al., 1953). i The “literature cited” section must include all (but not additional) references quoted in the text. References should be listed in alphabetical order and typed on sheets separate from the text. Each citation must be complete and in the following form: a) Periodicals CATE, J. M. 1962. On the identification of five Pacific Mitra. The Veliger 4(3):132- B34. b) Books YONGE, C. M. & T. E. THOMPSON. 1976. Living marine molluscs. Collins, London. 288 pp. c) Composite works FEDER, H. M. 1980. Asteroidea: the sea stars. Jn: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal invertebrates of California. Stanford Univ. Press, Stan- ford, Calif. pp. 117-135. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Figures and plates Figures must be carefully prepared and should, be submitted ready for publication. Each should have a short legend, listed on a sheet following the tables. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photographs for half-tone plates must be of good quality. They should be trimmed off squarely, arranged into plates, and mounted on suitable drawing board. Where necessary, a scale should be put on the actual figure. Preferably, photographs should be in the desired final size. It is the author’s responsibility that lettering is legible after final reduction (if any) and that lettering size is appropriate to the figure. Charges will be made for necessary alterations. Processing of manuscripts Receipt of a submitted manuscript, except in the case of foreign authors, is not ac- knowledged unless a stamped, self-addressed envelope is enclosed. Upon receipt each manuscript is critically evaluated by at least two referees. Based on these evaluations the editor decides on acceptance or rejection. Acceptable manuscripts are returned to the author for consideration of comments and criticisms, and a finalized manuscript is sent to press. The author will receive from the printer two sets of proofs, which should be corrected carefully for printing errors. At this stage, stylistic changes are no longer appropriate, and changes other than the correction of printing errors will be charged to the author at cost. One set of corrected proofs should be returned to the editor. An order form for the purchase of reprints will accompany proofs. If reprints are desired, they are to be ordered directly from the printer. Send manuscripts, proofs, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. CONTENTS— Continued Comparison of northern and southern populations of Epztonium tinctum (Car- penter, 1864) on the California coast CAREY ARES CH 1S MIIAGEty AND NYY Gn RIENYsE/ RO eae ee eee The reproductive cycle of the trochid gastropod Oxystele variegata (Anton, 1839) M. ALPS JOSKA ANDGs) Ma BRYANGEIN. ieeean kre rica eee eh hy ices aoe eet ear The larval biology of Brachidontes modiolus (Linné, 1767) (Bivalvia: Mytili- dae) ANGELACFIEL DS! AND SEUNAGNIOORE = ye ia erent ica ele ene A comparison of the species richness and trophic roles of gastropods and chitons on rocky shores of temperate and tropical west America ALAN C.JMILGER TS oo ere Sei oh eet ath ded oes Sed big cc Sa ge er Range extensions of three opisthobranch mollusks to the San Diego—La Jolla (California) Ecological Reserve EVANS) BERTSCH AND) (OM: SS MAHHO YE 2 Ge 4a5.4 a mie cle se ee ne oe Notes, Information & News 37 47 52 62 oe ISSN 0042-3211 _ 0, Oo / THE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 26 October 3, 1983 Number 2 CONTENTS Radular patterns, systematics, diversity, and ecology of deep-sea limpets. CAROLE S. HICKMAN Additions to the opisthobranch mollusk fauna of Marin County, California, with notes on the biology of certain species. VIN TELIL TAIN 1835." NNSKCH SO EMT Gel A PON tee eS en eo 93 Specialized feeding in mitrid gastropods: evidence from a temperate species, Mitra idae Melvill. AELANGHUKUVAMACAND) JAMES INYBAKKEN ©... 20. ..05.0.05-.0. 0.54500 96 Homing in the West Indian chiton Acanthopleura granulata Gmelin, 1791. JDUANWIOD) IVCKORS S08 ss ORR a a eta Oy) ees nm ne EAN AI 101 Aspects of reproduction in some enoploteuthid squids from Hawaiian waters. ILIA IML: IKOUWIGTETST Ls, ile dial i cata ered er ape LaPeer > te ed mea eae 106 Shell repair frequencies of two intertidal gastropods from northern California: microhabitat differences. RON AGEIAN GIS | GEE RG i meme ein, «Ge OM cM ote MMR ING AR ake, NS) Morphological divergence and predator-induced shell repair in Alia carinata (Gastropoda: Prosobranchia). JULIA BERGMAN, JONATHAN B. GELLER, AND VICTOR CHOW ........... 116 Thermal effects on some mangrove mollusks. ION CAV OOM GHOOPAND ESE ANOR (CHIN eels. ay laa ary sys ees eee ee 119 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of July, October, January and April for $18.75 for existing members (plus mailing charges) and $37.50 for libraries and nonmembers (plus mailing charges). Mailing charges for all domestic addresses are $3.25 and $6.00 for all Canada, Mexico, and foreign addresses. Further membership and subscription information appears on the inside cover. The Veliger is published by the California Malacozoological Society, Inc., % Department of Zoology, University of California, Berkeley, CA 94720. Second Class postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to C.M.S., Inc., P.O. Box 9977, Berkeley, CA 94709. THE VELIGER Scope of the journal The Veliger is open to original papers pertaining to any problem concerned with mol- lusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS’; in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Donald P. Abbott, Emeritus, Hopkins Marine Station of Stanford University Hans Bertsch, Universidad Autonoma de Baja California James T. Carlton, Williams College—Mystic Seaport J. Wyatt Durham, University of California, Berkeley Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley A. Myra Keen, Emerita, Stanford University Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Washington Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley T. E. Thompson, University of Bristol, England Alex Tompa, University of Michigan, Ann Arbor Membership and Subscription Membership in the California Malacozoological Society is open to persons (no institu- tional memberships) interested in any aspect of malacology. Annual dues, which include a subscription to The Veliger, are US $18.75 plus mailing charges. An initiation fee of US $2.00 is required of new members; a reinstatement fee of US $1.00 will be required if membership renewals do not reach the Society on or before April 15 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign members, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $37.50 plus mailing charges. Mailing charges, both for individual memberships and library subscriptions, are US $3.25 for all domestic addresses and US $6.00 for all foreign addresses, including Canada and Mexico. Memberships and subscriptions are by Volume only (July 1 to April 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $30.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: C.M.S., Inc., Post Office Box 9977, Berkeley, CA 94709. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. The Veliger 26(2):73-92 (October 3, 1983) THE VELIGER © © CMS, Inc., 1983 Radular Patterns, Systematics, Diversity, and Ecology of Deep-Sea Limpets by CAROLE S. HICKMAN Department of Paleontology, University of California, Berkeley, California 94720 Abstract. Sixteen distinct radular patterns from deep-sea limpets are illustrated and discussed. Two of the patterns are docoglossan, nine are rhipidoglossan, and five do not fit the standard classification of gastropod dentition. The diversity of gastropods that have converged on limpet shell form is re- markable. The animals occur in three major deep-sea settings: at hadal depths in eutrophic nearshore trenches; at eutrophic hydrothermal vents at divergent oceanic plate boundaries; and associated with isolated ‘‘food falls” in oligotrophic regions of the bathyal and abyssal zone. Many of the taxa are opportunistic specialists on unusual substrates such as wood, sea-grass rhizomes, cephalopod beaks, empty polychaete tubes, and empty skate or shark egg cases. Radular morphology is not, however, convergent in groups that have converged on the same unusual substrates. Different taxa have adopted alternative morphological solutions to using the same substrate. The high diversity of minute limpets in the deep sea today is equalled only by the diversity of minute limpets in the initial Cambrian radiations of the Mollusca. The protective function of limpet shell geometry that is invoked for familiar shallow-water intertidal limpets is not applicable to minute deep-sea limpets or to Cambrian limpets. Freedom to experiment with reproductive and respiratory modes is proposed as an important factor in the evolution of limpet geometry. Radular characters provide powerful taxonomic tools at the family level, and above, because they reflect fundamentally different groundplans. At the generic and specific levels, however, radular char- acters must be used with caution because the tendency for morphology to degenerate or fail to develop exceeds evolutionary “perfection” or “fine-tuning” of elaborate structure. Pseudococculinidae fam. nov. is proposed to provide consistent familial level distinction of groups separated by major discontinuities in radular patterns. INTRODUCTION THE TERM “limpet” is applied to a variety of gastropods with low, broadly conical shells. It denotes not only a shell form that has evolved numerous times within the class, but also the accompanying habit of clinging or clamping to rocky substrates (YONGE & THOMPSON, 1976, p. 51). Accordingly, functional interpretations of the limpet shell form have emphasized the theme of protection at the ex- pense of mobility (YONGE & THOMPSON, 1976, p. 265). Protection has been identified as resistance to predation as well as resistance to the physical stresses of the inter- tidal zone on rocky coasts: notably wave shock, tempera- ture, desiccation, and salinity stress (VERMEIJ, 1978; BRANCH, 1981). Not all limpets live on rocky substrates, however. The purpose of this paper is to call attention to the extraor- dinary taxonomic diversity of gastropods that have adopt- ed this shell form in deep water (below 200 meters). At least 15 families are involved. Many of the deep-sea lim- pets have minute shells (less than 5 mm long) and are poorly studied anatomically. Examination of their radulae has revealed striking differences within taxa heretofore considered to be closely related. Several of the radular patterns are so unique that they do not fit the standard classification and terminology used to describe gastropod radulae. Although radular anatomy will contribute substantially to evaluating the taxonomic relationships of deep-sea lim- pets, comprehensive systematic revision will require a great deal of further study of the soft anatomy. However, it is appropriate at this time to summarize and illustrate the range of radular morphology and correspondingly high diversity of feeding modes in order to stimulate further interest in the biology and systematics of these animals. Page 74 Many are restricted to unusual substrates in the deep sea, including water-logged wood, rhizomes of Thalassia, emp- ty polychaete tubes, empty skate egg cases, cephalopod beaks, and bone. There are three major deep settings in which most of the diversity and abundance of limpets is concentrated. None of these settings is typical of the deep sea in terms of the main body of ecological theory concerning this vast portion of the marine realm. The first is the eutrophic nearshore trench or basin that predictably receives large volumes of plant material of terrestrial origin. At low latitudes where trenches border salt marsh or mangrove areas of high productivity, such as the Puerto Rico and Cayman Trenches, the abundance of terrestrial material transported in by turbidity flows may be particularly high (WoLFF, 1976, 1979, 1980; GEORGE & HiGGIns, 1979). The second is the eutrophic setting created by hydrother- mal vents, also regions of high productivity, although based on a chemosynthetic rather than photosynthetic pathway (JANNASCH & WIRSEN, 1979, 1981; RAU & HEDGES, 1979; Kart et al., 1980; CAVANAUGH et al., 1981; Rau, 1981). The kinds of limpets that are associated with these first two kinds of food sources are taxonomically distinct. Both groups, however, share opportunistic characteristics that are not normally associated with deep-sea organisms. Many of the limpets in the first group also occur in a third and more transient setting: as colonists of relatively large “low-quality food falls” (sensu STOCKTON & DE- Laca, 1982) in the form of waterlogged wood that pe- riodically reaches the ocean floor in otherwise oligotrophic regions. The wood-ingesting limpet guild occurs consis- tently with other invertebrates that also are specialized starting points in the conversion of wood to animal bio- mass, most notably wood-boring bivalves of the genus Xy- lophaga (TURNER, 1973). A striking feature of many deep-sea limpets is the small size of the shell. It is potentially of significance to paleon- tological and malacological debates about the origins and early diversification of the Mollusca. The earliest fossil record of mollusks, not only gastropods but also Mono- placophora and a number of groups of questionable affin- ities, is dominated by small limpets. Many of these minute forms may have been associated with unusual substrata or may even have lived infaunally. The protective. inter- pretation of shallow-water modern limpet shell form would not apply at this scale. The discovery of a minute: (less than 2.5 mm maximum shell length), monoplacophoran limpet associated with small rocks within muddy sub- strates at the edge of the continental shelf (MCLEAN, 1979) suggests another mode of life—a means of being function- ally epifaunal in an infaunal environment. Thus, an anal- ysis of available ecological data on deep-sea limpets and their unconventional substrates provides the basis for some new hypotheses for functional interpretation of the minute limpet form in early mollusks, even though Recent deep- sea limpets are secondarily flattened. The Veliger, Voll 2633NogZ MATERIALS anp METHODS The following survey of deep-sea limpet radulae is based on extractions from both wet and dry material from nu- merous sources (see acknowledgments). Dry specimens were rehydrated in saturated trisodium phosphate prior to maceration. Radulae from both wet and dry specimens were removed from the buccal mass following treatment in a 10% solution of sodium hydroxide. All radulae were removed before the buccal mass was completely broken down, so that the chitin of the teeth and radular mem- brane did not come into direct contact with the caustic solution. Radulae were washed thoroughly in distilled water and dehydrated through an ethanol series from 70% to 100% prior to air drying and mounting for scanning electron microscopic (SEM) examination. Before drying, radulae were cleaned by brief immersion in an ultrasonic cleaner. The intensity of cleaning was based on a combi- nation of the robustness of the radula and its ability to withstand ultrasonication, and the amount of extraneous material on the teeth. Radulae were manipulated during air drying to expose maximum dentition, which involves laying back the mar- ginal tooth complexes in the rhipidoglossan types. Dry radulae were transferred to SEM stubs and mounted on a thin film of Elmer’s Glue® thinned with distilled water and allowed to dry until only slightly tacky. It is important not to drop the radula onto wet glue, and even when the glue appears dry, good bonding may be achieved between the radula and the SEM stub by breathing gently on the preparation. All handling of radulae was performed using flexible minuten nadlen, extremely fine insect pins, glued to 2-mm-diameter wooden splints. - All specimens were coated with 200-400 A of gold and viewed at accelerating voltages of 10 kV or less to mini- mize charging and specimen damage. Preparations illus- trating each major radula type were rotated into a stan- dard orientation relative to the electron beam, with the stage maintained flat. The orientation selected allows a diagonal presentation that maximizes areal coverage and resolution. All manipulations were made mechanically with the microscope stage, and not electronically or by scan rotation, to ensure consistent shading. Stereographic pairs were taken of each radula to provide further basis for interpretation. This is particularly important in the anal- ysis of unusual radulae in which the topographic differ- ences between teeth of different types may be considerable, but are not readily perceived, in single electron images. Scanning electron microscopy is a valuable tool in the study of deep-sea limpet radulae, because their minute size places a great deal of morphological detail beyond the resolvable range of light microscopy. Previous treatments of the taxonomy of deep-sea limpet taxa that include rad- ular data have been illustrated by line drawings that are ambiguous, imprecise, and frequently erroneous in inter- pretation. The illustrations that follow will help in the C. S. Hickman, 1983 reinterpretation of many drawings in the literature and will enable future workers who do not have access to SEM to interpret light microscope preparations. The selection of radular material for illustration is made on morphological rather than systematic or typological grounds. The objective is to present the best possible il- lustration of all the major radular patterns known from the deep sea. In several cases, distinctive patterns are known from a single species and represent the only ma- terial available. In most instances, however, the patterns are known from a number of species, and the material illustrated is chosen to show central tendencies and ex- tremes of variation. Because the radulae of type species do not necessarily illustrate central tendencies, they are included only where morphologically appropriate. RADULAR MORPHOLOGY anp PATTERNS Fissurellid Limpets (Figures 1, 2, 3) The keyhole limpets are, as a family, predominantly shallow-water, hard-substrate dwellers. Those forms that do occur in deep water are particularly interesting because their radulae differ from the radulae of shallow-water members of the family, although little is known of their feeding habits or substrate preferences. The radulae of three deep-sea fissurellids are illustrated in Figures 1-3. Puncturella: Puncturella Lowe, 1827 (Figure 1) has the most typically fissurellid radula, and some species in the genus, including the type, P. noachina (Linné, 1771), are not restricted to deep water, but occur intertidally in the high latitude portion of their range. The massive, strong- ly cusped outer lateral teeth and reduced prominence of the rachidian and inner lateral teeth are hallmarks of the fissurellid radula. The radula figured here lacks the pro- nounced asymmetry that is characteristic of many shal- low-water fissurellids (HICKMAN, 1981), but it is not known whether lack of asymmetry is universal in Punc- turella. Fissurisepta: Fissurisepta Seguenza, 1853 (Figure 2), on the other hand, displays both the prominent outer lateral tooth and the asymmetry. Note that each of the left inner lateral teeth is more anteriorly positioned than the right inner lateral in the same row. This skewing of rows is an accommodation that permits an alternating or zipper-like close-packing arrangement of the large outer lateral teeth when the radula is retracted or folded for storage when the animal is not feeding (HICKMAN, 1981). The combi- nation of serrate cusps on the inner lateral teeth and the absence of both shaft and cusp on the rachidian in this radula is, however, unique. Zeidora: The radula of the deep-water genus Zezdora A. Adams, 1860 (Figure 3) is even more unusual, as judged from examination of one unidentified species. There is no trace of a rachidian tooth, and the outer lateral tooth is of Page 75 an unusual form in that there is a thin, secondary shaft and cusp that appear to be fused to the main body of the tooth. This suggests that the massive fissurellid outer lat- eral may, evolutionarily, be a compound tooth resulting from the fusion of what were once distinct elements. The radula is also unusual in that the form of the asymmetry is reversed (2.e., rows are skewed so that the right lateral and marginal teeth fold into the tooth row anterior, rather than posterior to the left lateral teeth in the same row). Based on shell characters, Zezdora has been interpreted to be the most primitive genus in the Fissurellidae (FAR- FANTE, 1947), but the radula of the species investigated is not of a generalized form that would readily permit der- ivation of other forms of fissurellid dentition. Limpets Questionably Allied to the Fissurellidae (Figures 4, 5) Laevinesta: The monotypic genus Laevinesta Pilsbry & McGinty, 1952, based on Nesta atlantica Farfante, 1947, is so bizarre in its radular morphology that it can no longer be included comfortably in the Fissurellidae, al- though it may be of fissurellacean derivation (HICKMAN, 1983). The radula (Figure 4) bears little resemblance to that of any living gastropod in its combination of a low central field of teeth with prong-like cusps and its flanking fields of plates. The presence of two, long and slender marginal teeth in each half row does identify it as being of rhipidoglossan derivation, but it is so highly modified and distantly removed morphologically from all other gas- tropod radulae that it is impossible to ascertain its affin- ities. The fragile shell is wholly internal, but has a mar- ginal slit and selenizone that are of fissurellid appearance. The animal occurs on sponges and presumably feeds upon them, although the precise nature of the association is not documented. The generalized form of the radula of Laevinesta, in which fields of robust plates replace normal teeth, repre- sents a grade of radular morphology that has been at- tained independently in several other major groups of deep- sea limpets (e.g., Bathysciadiidae, Addisoniidae) that are associated with unconventional substrates. Galapagos Rift slit lmpet: Another small, emarginate, slit limpet that has a shell of fissurellid appearance occurs at deep-sea hydrothermal vents in the Galapagos Fracture Zone, where it has been recovered from washings of larger sessile mega-invertebrates (ANONYMOUS, 1980). The shell is unusual in that it is asymmetric, with the slit and se- lenizone forming an acute angle with the line bisecting the shell. It is the radula (Figure 5) that leads to serious questions about the affinities of the animal. It is rhipi- doglossan, but there is no asymmetry, no enlarged outer lateral tooth, and no reduction of the prominence of the central tooth field. In fact there is an abnormally high number of lateral teeth (9 per half row in contrast to the traditional 4 of the fissurellid radula). There are other The Veliger, Vol. 26, No. 2 Cy Ssilickman, 1983 Page 77 peculiarities of the radula that will be discussed in greater detail by Turner and Hickman (in preparation). Patellacean Limpets (Figures 6, 7, 8) Patellacean limpets, like fissurellids, occur primarily in rocky intertidal habitats, with several offshoots that are specialized for life in the deep sea: notably Pectinodonta Dall, 1882, in the Acmaeidae, and Propilidium Forbes & Hanley, 1849, Bathylepeta Moskalev, 1979, and other members of the family Lepetidae. Deep-sea patellaceans have radulae that are at once recognizable under the light microscope as docoglossan, a highly specialized form of gastropod dentition in which a few, heavily cusped teeth appear brown or black pigmented, due to the addition of iron salts that substantially increase tooth hardness. The radulae of deep-sea patellaceans do, however, contain some peculiarities that are related primarily to fusion of ele- ments. Pectinodonta: This genus comprises a small, but widely distributed, group of deep-sea species that are primarily members of the wood-ingesting guild. They are frequently recovered from wood (WOLFF, 1979; personal observa- tion) that they cohabit with both cocculinid and pseudo- cocculinid limpets. The radula of Pectinodonta (Figure 6) is unusual in the elongation of the third lateral tooth into a multicusped, curved, serrate unit that compresses the first and second lateral teeth, which form, along with the third lateral, a curved diagonal cutting edge with an ap- pearance similar to that of a wood rasp (SCHOOLEY et al., 1982; HICKMAN, 1982). The teeth are set at a steeply acute angle to the longitudinal axis of the radula. Al- though each pair of compound cutting edges appears to be identical with the pairs anterior and posterior to it, careful inspection of Figure 6 shows that pairs are not strict mirror images and that one tooth set in each pair lies either slightly anterior to or slightly posterior to the other. There is an alternation along the length of each half row of the radula in anterior vs posterior placement. WoLFF (1979) has reported scars on wood from which Pectinodonta has been recovered. These probably repre- sent homing depressions produced as a result of activity distinct from feeding. The plane of the aperture is not flat in some species but curved, presumably to conform to a curved, wood surface. CHRISTIAENS (1979) reviewed the species of Pectino- donta and concluded that, because of the close similarity of shell form with that of some cocculinid limpets, some species originally assigned to Cocculina are actually Pec- tinodonta. However, LINDBERG (1979) has noted that, even without animals and radulae, the genera are easily sepa- rated by major differences in shell structure. Lepetid limpets: Although the family Lepetidae is by no means restricted to deep water, the comprising species are predominantly subtidal in habit and restricted to cold water, primarily at high latitudes, where they occur on pebbles and cobbles at the sediment-water interface (YONGE, 1960). The ctenidium is lacking (lost?) in lepe- tids, but without development of secondary gills, which YONGE (1960) has interpreted as requiring continuous submergence while freeing the animal for life on muddy substrates, an impossibility for so many gastropods that face the problem of sediment clogging of the respiratory apparatus. Although no firm evidence supports YONGE’S scenario for the evolution of the situation in which spe- cialized respiratory structures are lacking (the absence of morphology is always difficult to interpret), a number of unrelated deep-sea limpet groups discussed below also lack ctenidia or secondary gills. MOSKALEV (1977) has de- scribed species of Lepetidae that are restricted to deep-sea trenches at depths in excess of 5000 m. Little is known of the diet of lepetids. YONGE (1960) has observed that Lepeta (Cryptobranchia) concentrica (Middendorff, 1847) feeds on organic detritus that settles out of the water column. The lepetid radula is complicat- ed, and drawings in the literature, although sometimes meticulous (MOSKALEV, 1977), are difficult to interpret. For comparative purposes it is particularly important to Explanation of Figures 1 to 8 Figure 1. Puncturella sp. Zoological Museum, University of Co- penhagen, Galathea Station 436. Bar = 100 um. Figure 2. Fissurisepta granulosa Jeffreys, 1882. Specimen pro- vided by A. Warén; Biological Station, University of Bergen Reference Number E39-71. Bar = 20 um. Figure 3. Zeidora sp. U.S. National Museum (USNM) 761482. Bar = 40 wm. Figure 4. Laevinesta atlantica (Farfante, 1947). USNM 792397. Bar = 100 um. Figure 5. Galapagos Rift slit limpet, gen. et sp. nov. Deep Sea Research Vessel (DSRV) Alvin Dive 733. Bar = 40 um. Figure 6. Pectinodonta n. sp. U.S. Fish and Wildlife Service Bureau of Commerical Fisheries, R/V Oregon II, Station 11230. Bar = 100 um. Figure 7. Lepeta (Cryptobranchia) concentrica Middendorff, 1847. University of California Museum of Paleontology (UCMP) 2896. Bar = 100 um. Dentition from posterior portion of radula. Note the central and lateral teeth are still fused into a single unit that is beginning to separate medially. Figure 8. L. (C.) concentrica. Dentition from the anterior end of the same radula illustrated in Figure 7. Note that the laterals are distinct and separate from the double central tooth. C = central; L = lateral; M = marginal. Bar = 100 um. The Veliges VorZouNow fl ied C. S. Hickman, 1983 view the radula at a standardized point in ontogeny due to variation in the amount of fusion of elements along the length of the membrane. This is illustrated by two views of the radula of L. (C.) concentrica: in Figure 7, the po- sitionally central and lateral teeth are fused into a single element, although they are beginning to separate medially; and the marginal teeth are separate and fully formed. These teeth are from the posterior portion of the radula. In Figure 8, from the anterior end of the same radula, the teeth are fully formed and worn from use; and the central tooth is double and flanked by worn lateral teeth as well as by worn marginals. Although I have not been able to examine the radulae of any of the deep-sea lepetid species, drawings in the literature (e.g.. MOSKALEV, 1977; DELL, 1956) suggest a basic plan of dentition comparable with that of the shal- low-water species discussed above and illustrated in Fig- ures 7 and 8. Bathysciadiid Limpets (Figures 9, 37a, b) The family Bathysciadiidae Dautzenberg & Fischer, 1901, and the genus Bathysciadium Dautzenberg & Fisch- er, 1901, were originally based on specimens of a minute, deep-water Atlantic limpet recovered from several stations in the vicinity of the Azores on the remains of cephalopod beaks. The type species, B. conicum Dautzenberg & Fischer, 1901, was originally described as Lepeta costulata Locard, 1898, and the type is thus a synonym of B. cos- _ tulatum. The radula of the type species, as drawn by PELSENEER (1901), who described the anatomy from material sup- plied by Dautzenberg and Fischer, is clearly unusual in form and difficult to interpret. Although PELSENEER con- cluded that the species was a patellacean limpet, the rad- ula is not docoglossan, nor can it be derived from a doco- glossan form of dentition. Minute limpets subsequently recovered from cephalo- pod beaks in deep water in both the eastern and western Pacific and depicted as having radulae of similar config- Page 79 uration have been assigned not only to Bathysciadium (DALL, 1908; THIELE, 1908) but also serve as type species for the genera Bonus Moskalev, 1973, and Bathypelta Moskalev, 1971. On the grounds that specimens have been reported with and without gills and with one and two kidneys, MosKALEV (1971) distinguished two families (Bathysciadiidae and Bathypeltidae) and two superfami- lies (Bathysciadoidea and Bathypeltoidea), both allied with the docoglossan limpets (MOSKALEV, 1971, 1973, 1978). The literature treating the anatomy of Bathysciadium and its allies is written in four languages (English, French, German, and Russian), and reconciliation of existing dif- ferences of opinion is hampered not only by language dif- ficulties but also by the fact that no one has yet been able to compare the anatomy of specimens from the Atlantic and Pacific. The extraordinary radula of Bathysciadium pacificum Dall, 1908 (Figure 9) does not fit any of the major radula types described to date and is so highly modified that it is not possible to determine the type of radula from which it is derived. There is a broad central portion of radular membrane that is without teeth. This region is flanked by two elevated, lateral columns of heavy plates that are in- terlocked or partially fused, both within rows and within columns, to form continuous, flexible, jointed pavements. A great deal of topographic relief is revealed in paired stereomicrographs of the pavements (Figures 37a, b). There are four plates in each half row: a small inner plate that forms a steep inner wall and partially underlies the adjacent plate; two larger, broadly rectangular plates that are most elevated at the joint between them; and a small, but heavy, outer plate that is squarish in outline, but with a strong elevated ridge parallel to the column axis that divides the plate into steeply-inclined inner and outer faces. On the basis of the inferred striking similarity of rad- ular configuration and ecological association with remains of cephalopod beaks, it seems probable that the forms described to date are very closely related and may not merit taxonomic distinction of the magnitude proposed by MosKALEV (1971, 1973). The radula is neither rhipido- Explanation of Figures 9 to 17 Figure 9. Bathysciadium pacificum Dall, 1908. Los Angeles County Museum of Natural History (LACM) 66-152. Bar = 100 um. Figure 10. Addisonia paradoxa Dall, 1882. USNM 181201. Bar = 40 um. Figure 11. Lepetella tubicola Verrill, 1881. USNM 153171. Bar = 40 um. Figure 12. L. tubicola. Detail of newly forming rachidian teeth from posterior end of same radula illustrated in Figure 11. Bar = 20 um. Figure 13. Addisonia paradoxa. Enlargement of the outer two heavily cusped radular plates from same radula illustrated in Figure 10. Structure at arrow may be a degenerate marginal tooth. Bar = 20 um. Figure 14. Cocculina sp. A. Zoological Museum, University of Copenhagen, Galathea Station 65. Bar = 40 um. Figure 15. Cocculina sp. B. U.S. Fish and Wildlife Service Bu- reau of Commercial Fisheries, R. N. Oregon II, Station 11220. Bar = 40 um. Figure 16. Cocculina sp. C. LACM, uncataloged. Bar = 100 um. Figure 17. Cocculina sp. D. Zoological Museum, University of Copenhagen, Galathea Station 276. Bar = 40 um. No. 2 b) The Veliger, Vol. 26 Page 80 C. S. Hickman, 1983 glossan nor docoglossan, and determination of the system- atic position of the Bathysciadiidae and resolution of dis- crepancies in the literature await careful examination and comparison of the soft anatomy of the constituent species. Although the substrate of these limpets is potentially difficult to locate and colonize, it is now known that the beaks of cephalopods may form extensive and dense, but highly localized, accumulations at abyssal and hadal depths. BELYAEV (1966) has estimated that they exceed 1000 per m? over extensive areas. Beaks that have been occupied by bathysciadiid limpets are extensively pitted and scarred. There is one species of cocculinid limpet that also lives on cephalopod beaks in the lower abyssal zone, described by MosKALEV (1976) as Teuthirostria cancellata. Although cocculinids normally live on wood, the radula of the cephalopod-beak species is typically cocculinid and is in no way convergent with the bathysciadiid radula. Addisoniid Limpets (Figures 10, 13, 38a, b) The genus Addisonia Dall, 1882, includes small, deep- sea limpets with thin, markedly asymmetrical shells and a series of anatomical and radular peculiarities that war- rant distinction as a separate family, as recognized by DALL (1882). The anatomy is, however, incompletely de- scribed (DALL, 1882; VERRILL, 1884; THIELE, 1908), and requires further detailed study to clarify the affinities of the family. The radula (Figure 10), like that of bathysciadiids, con- sists of a pavement of radular plates rather than distinctly cusped teeth, although there is no relationship between the plate patterns that would suggest any commonality of descent. It is neither rhipidoglossan nor docoglossan in form, unless one wishes to interpret the outermost element (see arrow in Figure 13) as a degenerate rhipidoglossan Page 81 marginal. The number of elements in each transverse row is 15, as interpreted by MOSKALEV (1971, p. 59), although the base number is not clear in the ambiguous series of drawings in the literature (e.g., VERRILL, 1884, pl. 29, fig. 14; THIELE, 1908, pl. 1, fig. 9; DALL, 1889, pl. 25, fig. le); and the shapes of individual elements cannot be de- termined accurately from light microscopy because of their minute size and overlap. Stereographic-paired, scanning electron micrographs (Figures 38a, b) clarify relationships in the addisoniid radula. The rachidian is a well-developed, subcylindrical element that fits together with the rachidian elements an- terior to and posterior to it to form a continuous, jointed, cylindrical column along the central longitudinal axis of the radula. The rachidian is flanked on either side by a pair of solid plates of rhomboidal outline, which are in turn flanked by a pair of small, narrow plates of sigmoid outline. The outermost elements in each half row are two complexly interlocked and heavily cusped plates that are elevated above the remainder of the radular ribbon and a narrow, elongate element that may represent a degenerate marginal tooth shaft. Nothing is known of the substrate preference of the type species (Addisonia paradoxa Dall, 1882) but specimens of an Addisonia from the eastern Pacific were extracted from inside of a skate egg case (J. H. McLean, written communication, 1983). Lepetellid Limpets (Figures 11, 12, 39a, b) Another group of minute, deep-sea limpets with a unique radular pattern and unique substrate association is the family Lepetellidae Dall, 1882, typified by Lepetella Verrill, 1880. The Atlantic type species, L. tubscola Ver- rill, 1880, occurs consistently on and in the tubes of the cosmopolitan polychaete genus Hyalinoecra Malmgren, Explanation of Figures 18 to 27 Figure 18. Detail of rachidian tooth from cocculinid radula in Figure 15. Note that the cusp is triangular and undivided. Bar = 10 um. Figure 19. Detail of rachidian and left lateral teeth from coc- culinid radula in Figure 14. Note that rachidian cusp is 5-pronged. Bar = 20 wm. Figure 20. Cocculina sp. E. Australian Museum C.108570. Note the great length and complex morphology of the massive outer lateral tooth and the flattening of marginal tooth shafts. Bar = 40 um. Figure 21. Enlargement of the surface of the cusp of the massive cocculinacean outer lateral tooth illustrated in Figure 20, with its unusual divaricate surface sculpture. Bar = 4 wm. Figure 22. Enlargement of cocculinacean marginal tooth shafts from the radula illustrated in Figure 15 and showing prominent flexure at mid height. Bar = 10 um. Figure 23. Fedikovella caymanensis Moskalev, 1976. Zoological Museum, University of Copenhagen, ex Akademik Kurchatov Station 1189. Bar = 40 um. Figure 24. Pseudococculina sp. A. Zoological Museum, Univer- sity of Copenhagen, ex Akademik Kurchatov Station 1267. Note the particularly well-developed rachidian. Bar = 40 um. Figure 25. Pseudococculina sp. B: Australian Museum 79-05-07. The rachidian is a robust plate lacking the shaft seen in Figure 24. Note strong development of cusps on the large outer lateral. Bar = 40 um. Figure 26. Caymanabyssia spina Moskalev, 1976. Although poor- ly developed, this is recognizable as pseudococculinid dentition. Bar = 40 um. Figure 27. Pseudococculina sp. C. Zoological Museum, Univer- sity of Copenhagen, Galathea Station 664. Bar = 20 um. Page 82 1867 (VERRILL, 1880; DALL, 1882, 1889; MOSKALEV, 1978). Minute limpets have also been removed from Hy- alinoecia tubes from deep water off New Zealand. They were originally described by SUTER (1908) as Cocculina clypidellaeformis, but FINLAY (1927) subsequently desig- nated this species as type species of the genus Tectiswmen. Although the affinities of many monotypic, or allegedly monotypic, austral gastropod genera have not been re- solved, radular preparations from the type species of TJec- tisumen Finlay, 1927, and Tecticrater Dell, 1956, have revealed dentition of the same pattern as Lepetella tubicola. Although VERRILL (1880) described the radula of Le- petella tubicola as taenioglossate, this radula provides another example of dentition that does not fit the standard radular classification employing “glossate” terminology. The positionally central tooth of L. tubscola (Figure 11) is broad, with a pair of lateral cusps that may result from fusion of once distinct lateral teeth. Fusion is suggested by the pattern of ontogenetic development (Figure 12), in which the earliest stage is the appearance of two cusps separated by a space on the radular membrane. There are three additional elements in the lepetellid radula: a small, free-standing inner lateral tooth with a robust, curved, and inward-pointed cusp, and a massive, tricusped outer lateral that is closely articulated with a large marginal plate. The marginal plates are of an unusual broad and flattened form, and they interlock with one another as well as with the bases of the massive outer laterals. Each plate has four interactions: one each with the plates anterior and posterior to it, and one each with the posterior and anterior basal faces of adjacent outer lateral teeth. Paired stereo-micrographs (Figures 39a, b) demonstrate that the plates are not elevated pavements analogous to those of Bathysciadium, Addisonia, or Laevinesta, and they are in- ferred to function solely in mechanical adjustments of the inner laterals rather than in food preparation and collec- tion. Cocculinid Limpets (Figures 14-23) Cocculinid limpets are small-shelled (usually less than 5 mm maximum length) wood-ingesting gastropods with minute radulae that have been difficult to evaluate with light microscopy. Although two major genera are. tradi- tionally assigned to the family (KNIGHT ef al., 1960; MOSKALEV, 1971, 1976), Cocculina Dall, 1882, and Pseu- dococculina Schepman, 1908, the radulae in these two gen- era are of radically different plans that require the rec- ognition of separate families (HICKMAN, in press). The cocculinid radula has many features that make it easy to recognize. Figures 14-17 illustrate a range of vari- ation in this plan. The rachidian tooth ranges from well- developed and cusped (Figure 14) to completely absent, leaving a gap on the radular membrane (Figure 17). Like many presence/absence characters, however, taxonomic significance should not be attached to this. Note that in Figure 15 these two character states exist along the length The Veliger, Vol. 26, No. 2 of the same radula. Although the rachidian forms initially in some species, it frequently is lost from the anterior half of the membrane. When present, the rachidian consists of two basal limbs that resemble the shafts of lateral teeth that fuse medially to form a short shaft and cusp (Figure 18). The cusp may be entire and bluntly triangular (Fig- ure 18) or subdivided into 3-5 secondary cusps (Figure 1D). There are always three inner lateral teeth of similar morphology (Figure 19), with long, slender shafts and small, coarsely divided cusps. The robustness of the inner lateral teeth is variable, but robust laterals are not posi- tively correlated with the presence of a well-developed rachidian, as illustrated in Figure 16. This radula is par- ticularly interesting because there appear to be four well- developed, lateral tooth shafts. The tops of the inner pair of shafts in each row appear to be broken off, which could be interpreted as failure to fuse and form a rachidian. The largest and best developed tooth in the cocculinid radula is the massive outer lateral (Figures 19, 20), with a large primary cusp and a smaller secondary cusp on either side. The surface of the primary cusp is not smooth, but bears a unique pattern of divaricate wrinkles (Figure 21). These massive, major food-preparing teeth fold to- gether in alternating zipper-like fashion when the radula is enrolled during retraction at the end of the feeding stroke and for storage within the radula sac (HICKMAN, 1981). Because of the size of the outer lateral tooth, the cocculinid radula provides a classic example of lack of coincidence of what HICKMAN (in press) calls “base rows” and “cusp rows.” That is to say that the cusp of the outer lateral is in a row of cusps that is different from the row of bases, making for a much more complex form of func- tional integration of tooth interactions than is possible in the standard rhipidoglossan radula where base and cusp rows coincide. The marginal teeth in the cocculinid radula are also unique. The innermost marginals have greatly expanded and flattened shafts that narrow distally to disproportion- ately tiny, but strongly hooked, cusps. The outer margin- als retain this same basic form, with strong lateral flat- tening of the shafts, but the shaft develops a peculiar twist at mid-length (Figure 22) that is of a design that would facilitate rotation in the upper shaft while the lower shaft remains stationary. There also appears to be a tendency for lower shafts to become fused below the twist. These teeth are extraordinarily long, and like the massive outer lateral, the cusp rows and base rows do not coincide. MOosKALEV (1976) described six new cocculinid genera based on abundant material collected from abyssal and hadal depths on eight Soviet expeditions. Four of the gen- era are based on species with pseudococculinid radulae, and these will be discussed in the following section. The other two, Fedikovella and Teuthirostria, have typical coc- culinid radulae. I have examined the radula of Fedikovella caymanensis from a specimen collected on the Akademik Kurchatov Expedition (Figure 23) and do not feel that it C. S. Hickman, 1983 is sufficiently different to warrant separate generic rec- ognition. MOSKALEV based the new name on the triden- ticulate condition of the rachidian, but, as discussed above, the condition of the rachidian in cocculinids is more a result of how faithfully the instructions are carried out in forming the tooth rather than any major variation in the nature of the instructions. Teuthirostria cancellata Mos- kalev, 1976, from the Peru-Chile Trench, is an interesting cocculinid because it occurs on squid beaks, along with Bathysciadium, rather than on wood. There is, however, no sign of any convergence on bathysciadiid radular mor- phology. DALL (1889) proposed Coccopygia for Cocculina spini- gera Jeffreys, 1883, on the basis of the absence of epipo- dial tentacles. Coccopygia is, however, a preoccupied name, and the radula of C. spinigera is not distinct from that of other species of Cocculina (based on SEM examination of radulae from USNM 177890). Maoricrater Dell, 1956, which has been placed in the Cocculinidae (KNIGHT et al., 1960), has a docoglossan radula and belongs in the Lepetidae. The animal and radula of Notocrater Finlay, 1927, are not known, and it too may be a lepetid. Detailed studies of the anatomy of the cocculinids are badly needed in order to establish the systematic position of these minute rhipidoglossan limpets and their relation- ships to other deep-sea families. The only account of the internal anatomy of a cocculinid is that of THIELE (1903), who concluded that their closest affinities were with the Neritacea. Shell morphology of the cocculinids likewise stands in need of review. Although the apex of the shell is poste- riorly directed in all species, it varies in placement from far posterior to central; and the shell may be tall and narrowly conical or extremely flattened. Shell surfaces vary from smooth to highly ornamented and pustulose. How- ever, shell surfaces are frequently badly corroded when they reach the surface; and specimens that have been kept in formalin or alcohol for long periods seem to be partic- ularly susceptible to disintegration. Shells are best pre- served dry, after being separated from animals that are to be preserved wet. Occasionally, fresh specimens preserve extraordinarily elaborate spinose periostracum (Lipps & HICKMAN, 1982, fig. 14-13). Pseudococculinid Limpets (Figures 24-28, 40a, b) Pseudococculinidae fam. nov.: Pseudococculina and a group of allied small deep-sea limpets share a plan of radular morphology that is so distinct from that of the Cocculinidae as to warrant separate familial recognition. The details of this plan are described below. The mor- phological differences require fundamentally different patterns of movement and mechanical interactions of teeth within and between rows that are discussed elsewhere (HICKMAN, in press). Pseudococculinid limpets also occur on wood, and have Page 83 shells and animals that are similar to cocculinid limpets. SCHEPMAN (1908) proposed the genus Pseudococculina on the basis of the unique radula of P. rugosplicata and P. granulata. Although SCHEPMAN (p. 16) noted the presence of a “plumate gill” in the type species, it is absent or poorly developed in pseudococculinids I have examined. As noted earlier, presence or absence of well-developed respiratory organs does not seem to be of profound taxo- nomic significance in deep-sea limpets, although the na- ture of the respiratory structures, when present, can be diagnostic. There are no published accounts of the inter- nal anatomy of Pseudococculina, and detailed histological comparisons with other deep-sea limpets will be required in order to establish relationships. The pseudococculinid radula is, like the cocculinid rad- ula, so small that light microscopy has not been adequate to diagnose it. It is superficially similar to the cocculinid radula in the parallel trends toward degeneration of teeth in the central field (particularly the rachidian) and the elaboration of the outer lateral tooth as the major food- preparing structure, accompanied by the development of radular asymmetry. Figures 24-26 illustrate a range of variation in pseu- dococculinid radular morphology. The rachidian varies from an asymmetric element with a prominent horseshoe- shaped base and thinner, irregular, tapered extension that cannot be differentiated into shaft and cusp (Figure 24) to a thin, squarish or oval plate (Figure 26). The inter- mediate condition (Figure 25) frequently involves the de- velopment of a plate with thickened edges and irregular or ragged edges or extensions. The degree of development of the rachidian may vary along the length of a single radula, as illustrated in Figure 27. In contrast to the cocculinid condition, there are four, rather than three, inner lateral teeth in the pseudococcu- linid radula. The innermost lateral (Figure 27) is an un- usual, large, triangular element with a long, inner basal limb and shorter, outer basal limb that fits into a curved pocket on the small second lateral. Figure 27 also illus- trates the strongly asymmetric placement of these large teeth on either side of the rachidian. The second, third, and fourth laterals are all strongly curved elements that have small, simple, distal cusps and are designed to inter- lock closely when the radula is tightly enrolled. The outer lateral (fifth lateral) is a massive tooth with a thickened quadrate shaft with a deep pocket on the inner surface that fits over the smaller inner lateral teeth when the radula is not in use. The massive cusp is subdivided into four blunt, terminal denticles. Unlike the massive outer lateral in Cocculina, it is a short tooth, and its base and cusp coincide with the bases and cusps of adjacent inner lateral teeth. The marginal teeth (Figure 28) have short, thick shafts in contrast to the extreme flattening of the elongate and twisted shafts of cocculinid marginals. The shaft thick- ening is most pronounced in the inner four or five mar- ginals, which also have long, curved, and sharply pointed The Veliger, Vol. 26, No. 2 Page 84 C. S. Hickman, 1983 cusps. Outer marginals do, however, develop flattened shafts, and the cusps are complexly serrate. A stereographic pair of the pseudococculinid radula (Figures 40a, b) demonstrates the extreme difference in topography between the massive outer lateral and mar- ginal complex and the poorly-developed central field of teeth. Moska.LeEv (1976) described four genera that I believe to be simple variations on the basic pseudococculinid plan: Kurilabyssia, Caymanabyssia, Bandabyssia, and Tentaocu- lus. MOSKALEV’s drawings of the radulae (fig. 1, p. 10) do indicate differences, but the differences seen in light microscopic observations and drawings have to do with degree of development rather than fundamental differ- ences in plan. The radula of Caymanabyssia spina (Figure 20) represents the extreme case of not only an underde- velopment of the plate-like rachidian, but also failure to | develop the inner lateral tooth except in outline and slight thickening. MOsKALEV (1976) notes that Tentaoculus perlucida has eyes, a feature that is lacking in other pseudococculinids as well as cocculinids, addisoniids, and bathysciadiids. It may be that there will be suites of anatomical differences in many of the deep-sea limpet families that will justify separate generic categories. Pseudococculina is sometimes extremely abundant on pieces of wood. MOSKALEV (1976) recovered 204 speci- mens of Caymanabyssia spina from a piece of wood 800 mm X 150 mm in size, and large numbers of an unde- scribed species of Pseudococculina have been recovered from wood from deep-sea samples off the east coast of Australia (personal observation). Bathyphytophilid Limpets Wood is not the only form of plant material that reaches the deep sea floor in large quantities. There has been a Page 85 great deal of interest in recent years in the export of or- ganic matter from high productivity sea-grass beds along the continental margins; and it has been established that this material is transported by turbidity flows and accu- mulates in deep-sea trenches. It is particularly abundant in the Puerto Rico and Cayman Trenches, where blades and rhizomes of Thalassia testudinum comprise 80-90% of the plant material (MENZIES et al., 1967; MENZIES & Rowe, 1969; WoLFF, 1976, 1979, 1980; GEORGE & HIG- GINS, 1979). It is, therefore, not extraordinary that there should be communities of deep-sea organisms specialized for using sea grasses as a food source. MOSKALEV (1978) based the family Bathyphytophilidae on a species of lim- pet that inhabits Thalassia rhizomes in the Cayman Trench. WOLFF (1979, 1980) who examined the Soviet R/V Akademik Kurchatov material that formed the basis for MOSKALEV’s description, observed that the limpets not only occupy burrows in the rhizomes created by boring bivalves, but that they may also occur in their own feeding cavities within the pithy center of the rhizome. The bathyphytophilid shell has a cocculinacean ap- pearance, minute and with a posterior-directed apex, but the radula is distinct from all other minute abyssal limpet groups. I have not been able to obtain material of the Thalassia-inhabiting limpets, but MOSKALEV’S illustration clearly indicates that it is a rhipidoglossan radula with extreme reduction in the number of marginal teeth to 5- 20. There are ambiguities in the drawings, but the most striking feature is the narrowness of the central field, where there is a single (poorly-developed?) rachidian, and the two massive (?) overlapping major lateral elements, which are at least superficially similar to the large overlapping elements of the addisoniid radula. The radula is strongly asymmetric. In addition to the type species, Bathyphytoph- ilus caribaeus, MOSKALEV (1978) described Aenigmabonus kurilokamtschaticus from the Kurile-Kamchatka Trench, but the substrate association of the second species is not known. Explanation of Figures 28 to 36 Figure 28. Detail of marginal tooth complex from radula illus- trated in Figure 27. Note differentiation of five inner marginals with unusually large shafts and cusps. Bar = 20 um. Figure 29. Choristella n. sp. LACM 77-374. Bar = 40 um. Figure 30. Neomphalus fretterae McLean, 1981. Deep-sea Re- search Vessel (DSRV) Alvin Dive 733. Bar = 100 um. Figure 31. Newly formed rachidian and left inner lateral teeth of N. fretterae from posterior end of radula. Bar = 40 um. Figure 32. Worn rachidian and left inner lateral teeth of N. fretterae from same radula illustrated in Figure 31. Bar = 40 wm. Figure 33. Radula of an undescribed deep-sea hydrothermal vent limpet, illustrating a rhipidoglossan plan shared by five new vent species. DSRV Alvin Dive 982. Bar = 40 um. Figure 34. Radula of an undescribed deep-sea hydrothermal vent limpet illustrating a new rhipidoglossan plan shared by four undescribed species. DSRV Alvin Dive 880. Bar = 100 um. Figure 35. Radula of an undescribed hydrothermal vent limpet illustrating a new radular pattern not falling under standard classification and shared by two new species. DSRV Alvin Dive 879. Bar = 20 pm. Figure 36. Radula of a second specimen of the limpet illustrated in Figure 35. Note that the dentition from a comparable portion of the radula is poorly developed, with cusps lacking on the rachidian and laterals. DSRV Alvin Dive 879. Bar = 20 um. Page 86 Cocculinellid Limpets and Choristellids (Figure 29) THIELE (1909) proposed the genus Cocculinella for a minute, deep-sea limpet with a unique non-rhipidoglos- san radular pattern in which the rachidian is unusually small and narrow, with five heavier teeth on either side. THIELE’S drawing of a half row of the radula (THIELE, 1909, pl. 1, fig. 8) and his description (pp. 21-22) indicate that the best-developed tooth is the bicuspid second lateral, and that the three outer laterals are of similar morphology but decreasing size. THIELE placed the genus with Lepe- tella in the Lepetellidae, with the interpretation that the large outer plate in the radula of Lepetella may be thought of as a fusion of the three outer laterals of Cocculinella. Unfortunately, no subsequent collections of animals of Cocculinella minutissima (Smith, 1904) are available for study. However, the radula of a minute, coiled deep-sea gastropod that is commonly associated with unusual sub- strates shows the same basic pattern (Figure 29). These coiled gastropods have been placed in the genus Chorvstella Bush, 1897 (Choristes auctt., non Carpenter, 1872). The radula has never been well understood, and there is a great deal of variation in the three drawings that have been published for the Atlantic species (VERRILL, 1882; BusH, 1897; BOUCHET & WarEN, 1979). The ambigui- ties, however, are reconciled by the scanning electron mi- crograph of an undescribed species (J. H. McLean, in preparation) from the Eastern Pacific (Figure 29). Ponder (unpublished) has scanned the radula of an undescribed species from off the Queensland coast, and it is compa- rable in numerous details (personal observation). Although substrate data are not available in most in- stances, where known, choristellids occur on or in the egg cases of sharks (KURODA e¢ al., 1971) or skates (VERRILL, 1884). The radular evidence for close affinities of a limpet and a coiled, operculate gastropod poses problems of classifi- cation that cannot be resolved within the scope of this paper and without reference to anatomy. Separate familial names are available: MOSKALEV (1971) proposed the Coc- culinellidae, placing it in the Lepetelloidea, and BOUCHET & WAREN (1979) have justified the change of the familial name Choristidae to Choristellidae. KURODA e¢ al. (1971) proposed the Choristiacea (=Choristelliacea) as a super- family, making this name available for hierarchically. re- lating these two families of disparate shell morphology. Hydrothermal Vent Limpets (Figures 30-36) The remaining four groups of deep-sea limpets are all associated with unconventional substrates and food sources at abyssal hydrothermal vents that have developed at di- vergent oceanic plate boundaries. The animals themselves occur on basalt surfaces, on polymetallic sulfide mounds and chimneys, or on the surfaces of sessile vent organisms such as the giant vestimentiferan worms or mussels. Che- mosynthetic bacteria, which are the only known primary The Veliger, Vol. 26, No. 2 producers in the deep sea, form the base of the unique food chain in this environment (JANNASCH & WIRSEN, 1979, 1981; Rau & HEDGES, 1979; Karu et al., 1980; CAVANAUGH et al., 1981; Rau, 1981). They are consumed directly by many of the vent organisms, which may either scrape bacterial films from hard substrata or filter bacteria that are suspended abundantly in the water column. Of the extraordinary diversity of limpets that have been re- covered to date from vent communities, only one species, Neomphalus fretterae, has been described (MCLEAN, 1981). Nine additional limpets remain to be described (McLean, in progress) as well as a secondarily flattened operculate species that is closely related to one of the lim- pet groups. McLean (written communication, 1982) ten- tatively recognizes three new families among the remain- ing 10 species. I have prepared and examined radulae of all the species and find three corresponding new radula types. Scanning electron micrographs are presented for each of the major radula patterns to facilitate comparison with patterns in deep-sea limpets from other environmen- tal settings. Detailed descriptions of these radulae will be published as an adjunct to the descriptions of the new taxa (Hickman, in preparation). Neomphalid limpets: One of the most abundant limpets at hydrothermal vents, particularly along the Galapagos Fracture Zone, is Neomphalus fretterae, a gastropod of such distinctive anatomy and shell morphology as to re- quire separate familial and superfamilial status (Mc- LEAN, 1981; FRETTER ef al., 1981). As illustrated in Fig- ure 30, it is rhipidoglossan in form, and the most remarkable feature is the robust development of the den- tition, with no indication of degeneration of teeth in the central field. The rachidian and five laterals are not only supplied with prominent pointed cusps, but the shafts are heavily reinforced, as are the complexly interlocking bas- es. The robust development is all the more remarkable in view of the enlarged mantle cavity and ctenidium and accompanying modifications for suspension feeding. Although the animals seem to be sedentary, conforming in shell growth to irregularities in the substrate at their home sites, they may derive a significant proportion of their nutrition from the bacterial films that develop on the basalt. FRETTER et al. (1981) note larger particles in the fecal string, suggesting a mixed feeding strategy. It is clear from comparison of micrographs of newly formed teeth and teeth from the working portion of the same radula (Figures 31, 32) that the teeth are used and sustain sub- stantial abrasion. The marginal teeth are delicate and finely serrate, frequently showing complex patterns of incom- plete separation of teeth during ontogeny and irregulari- ties that suggest lack of strong selection for normal devel- opment (HICKMAN, 1980, p. 292, fig. 6c). Hydrothermal vent Group-A limpets: Five hydrothermal vent gastropods share the rhipidoglossan radular pattern illus- trated in Figure 33. The shell morphology within this group is diverse and ranges from a minute, conispirally Crs rickman 1983 coiled, operculate species (Hickman, MS) to flattened forms that properly may be called limpets. The animals have numerous features indicating that their closest rela- tives are trochacean gastropods. I have tentatively placed the coiled species in the Trochacea, although anatomical evidence may lend support to establishment of a separate superfamily within the suborder Trochina. Vera Fretter is currently studying the anatomy of the limpet members of this group. In spite of the broad range of shell form, the radula is remarkably constant in morphology throughout the group. It is similar to the trochacean radular plan in the small base number of well-developed teeth in the central field and in the pattern of shaft reinforcement and within-row basal interlock. It is a relatively narrow radula, in part because the marginal tooth rows, although long, are set at a very low angle to the longitudinal axis of the ribbon. One of the most striking features of the radula is the wide spacing between rows in the central field, with tooth bases completely exposed. The wide spacing is correlated with an alternation in the position of the ends of lateral tooth rows and the beginnings of marginal tooth rows. A similar kind of accommodational alternation occurs in the neri- tacean radula (HICKMAN, in press). The substrate preferences and food sources of members of this group are not well understood. Three of the five species occur at high-temperature vents on the East Pacific Rise, where they have been removed from polymetallic sulfide build-ups surrounding the high-temperature ef- fluents. At these high-temperature vents, where water may be emitted at temperatures in excess of 350°C (in contrast to some of the ‘“‘warm” vents in the Galapagos that do not exceed 22°C), there are many fewer animals. Members of this limpet group also have been recovered from washings of the giant tube worm Riftia, and there may be significant microhabitat specialization within the complex. Hydrothermal vent Group-B limpets: Four limpet species share the basic radular pattern illustrated in Fig- ure 34. These species also share a combination of anatom- ical features that is unique to the Gastropoda (McLean, written communications, 1982-1983). Fretter is currently studying the anatomy of these limpets as well. The radula is rhipidoglossan, and as in the other rhipidoglossan vent limpet groups, the rachidian and lateral teeth are very well developed. The rachidian is a relatively small element with a sturdy basal support for the sharply pointed and finely denticulate cusp. The most unusual element in the radula is the inner lateral tooth with its massive, com- plexly divided, and finely denticulate compound cusp and broadened shaft. The cusp is oriented on a strong diagonal to the longitudinal axis of the radula, connecting the ra- chidian cusp with the four outer lateral cusps that are situated considerably anterior to the rachidian. This ar- rangement produces a strong, posteriorly directed dip in the cusp row. The shafts of the four outer laterals all have Page 87 a characteristic sharp bend at mid height that accommo- dates the cusps of the next row posterior. Shafts also have lateral expansions to facilitate within-row interlock. The marginal teeth are numerous, with flattened shafts and finely denticulate cusps. Hydrothermal vent Group-C limpets: Two of the lim- pet species collected to date from deep-sea hydrothermal vents have the non-rhipidoglossan radular plan illustrated in Figure 36. Like the radulae of bathysciadiids, addisoni- ids, and lepetellids, it does not fit into the standard clas- sification of radula types. The bases, shafts, and cusps of the five radular elements are unique in form. The single most unusual feature of the radula is the interaction be- tween the base of the second lateral and the cusp of the third lateral of the same row. Within-row interactions in other radulae are cusp-cusp or base-base, but in this group there is a prominent basal projection or fold on the second lateral that fits or locks beneath the cusp of the adjacent third lateral. The rachidian is a small tooth with a pointed triangular cusp. Its base is a short transverse bar that fits into deep pockets on the bases of the inner lateral teeth. The three inner laterals are of similar general form with thin, rel- atively broad and flexuous cusps, complex sinuous shafts, and interlocking basal folds. The outer lateral is a broad, flattened, spoon-shaped element with a semicircular rim of folded-over or enrolled chitin that is perhaps best not termed a cusp. The entire radula is very small and insubstantial, and in one specimen that I have prepared (Figure 36) failure of normal cusps to develop suggests that selection pressure to reproduce faithfully the standard plan is not great. DISCUSSION Radular Characters and Systematics Sixteen distinct deep-sea radular patterns are illustrat- ed in this paper. Although two of them are docoglossan and nine are rhipidoglossan, five do not fit the standard classification scheme. Even the docoglossan and rhipido- glossan radulae have patterns that are distinct in many details from those of related shallow-water groups, and in some cases deep-water families cannot be related to extant shallow-water families or to one another on radular char- acters. Some discussion of the role of radular characters in further systematic evaluation of deep-sea limpets is in order. Although a primary purpose of this paper is to provide data for dealing with the systematics and classification of deep-sea limpets, the radular data are biased toward an emphasis of differences rather than similarities. Radular characters will continue to be important in dealing with these taxa, but they will be most valuable in differentiat- ing genera and species within families. The major pat- terns are too distinct from one another and from patterns Page 88 The Veliger, Vol. 26, No. C. S. Hickman, 1983 in other gastropod groups to provide reliable data for es- tablishing higher relationships. This leaves us with something of a dilemma as to where to seek shared characters that will enable us to place many of these taxa into phylogenetic perspective. The shells are extraordinarily similar, but in shell form we are dealing with strong evolutionary convergence. Shell microstruc- ture may prove useful, although the emphasis should be on details of the mineral species and fabrics that are pres- ent rather than on absences that may be due to a secondary simplification of structure. Loss of shell layers, particu- larly the inner nacreous layer, appears to be strongly cor- related with small size (Hickman, MS), and therefore should not be heavily weighted in taxonomy. The most reliable clues should be revealed through detailed com- parative anatomical studies. Here too, however, it must be remembered that the limpet grade of evolution requires modifications of anatomical blueprints that were accom- modated to life in a coiled shell. Detailed comparative histological analyses are a formidable task, but will prob- ably be necessary because gross anatomical features are inadequate to establish relationships. Ecological and Evolutionary Implications What we now know about the radular morphology, habitats, and substrates of deep-sea limpets has some im- portant evolutionary and ecological implications. It has long been held that feeding structures in organisms are easily modified in relationship to substrates and food pref- erences and, thus, are not likely to provide good characters for differentiating taxa (see HOUBRICK, 1978, and refer- ences therein). Although this may be the case in some groups of gastropods, it is not a generalization that can be extended to the entire class (HICKMAN, 1980, 1982). This is nowhere better demonstrated than in the taxa of deep- sea limpets that have converged on feeding on unconven- tional substrates. For example, cocculinid limpets re- covered from wood have the same radular pattern as the cocculinid limpet recovered from squid beaks. Bathyscia- diid limpets from squid beaks have a radular pattern that shows no convergence with the cocculinid squid-beak species. Likewise, wood-ingesting docoglossans, wood-in- gesting cocculinids, and wood-ingesting pseudococculinids retain highly distinctive dentition patterns. In each of these three groups, there are formidable, major food-preparing Page 89 teeth that we may infer to be adequate designs for dealing with water-logged wood. However, it would be difficult to argue that any one of these solutions is an optimal one. Many evolutionary biologists and ecologists have based their research on the assumption that evolution involves a great deal of fine-tuning and optimization; but ideal so- lutions may be prohibitively costly to produce and main- tain, and many organisms may be built simply to “get by” (CowEN, 1981; HICKMAN, 1982). Limpets in the deep sea seem to have developed a number of alternative solutions to feeding on a number of difficult or unusual substrates. What is peculiar is the strong convergence in shell shape. It is not at all clear what the advantages of secondary flattening are to so many forms that colonize unconven- tional substrates in the deep sea. It is certain that the limpet shell form does not have the same protective func- tions in small deep-sea limpets that it does in shallow- water limpets. Connections Between Cambrian Limpets and Deep-Sea Limpets This brings us again to the question of size and the proposition in the introduction that there may be some connection between the predominance of small cap-shaped shells in the early evolution of the Mollusca and the pleth- ora of such forms in the deep sea today. If we accept the proposition that the protective mechanical functions of shell form that work in large shallow-water limpets do not apply to minute deep-sea limpets or to the Cambrian early mollusk radiations, what similarities do these forms share that might suggest alternative hypotheses? Although VERMEIJ (1971, 1973) has correctly empha- sized the increase in the morphological diversity of gas- tropod shells as a result of increase in the number of pa- rameters controlling shell geometry, it does not follow that more complex shell geometry permits greater diversity of anatomical organization. The simple geometric form of the limpet permits an extraordinary variation in anatom- ical organization. Lindberg and Wright (in MS) have shown that shell shape can place interesting volumetric constraints on the ways that anatomical plans can be varied with respect to reproductive modes and that some modes are more likely in limpets. Another important constraint, I believe, is that on respiration. The uncoiled limpet shape, especially when Explanation of Figures 37 to 40 Paired stereo micrographs illustrating topographic differences in four deep- sea limpet radular groundplans. Figure 37. Bathysciadium pacificum (same radula illustrated in Figure 9). Bar = 50 um. Figure 38. Addisonia n. sp. LACM, uncataloged. Bar = 50 wm. Figure 39. Lepetella tubicola (radula illustrated in Figure 11). Bar = 20 um. Figure 40. Pseudococculina sp. C (radula illustrated in Figure 27). Bar = 40 um. Page 90 coupled with small size, provides a great deal of versatility in modes of respiration. Discrete respiratory organs are not essential in small flat animals. And even when present, such organs do not require active control of water cur- rents. Although much has been made of the position of the “mantle cavity” in the interpretation and classification of early mollusks, the concept of the cavity needs critical reexamination. The concept derives in part from the standard drawings and discussions of the Hypothetical Ancestral Mollusk (YONGE & THOMPSON, 1976). A number of authors have advised of the dangers in using HAM to understand mol- luscan evolution (e.g., MORTON, 1967; STASEK, 1972; PojJETA, 1980), but many of the implications of the con- cept have been perpetuated by its harshest critics. A man- tle cavity need not be present at all. In flat, uncoiled mono- placophoran limpets the respiratory structures are situated in a pair of pallial grooves, not in a cavity. The notion that the development of a dorsal calcified covering created a mantle cavity is an oversimplification, and the conclu- sion that it necessitated the development of gills (POJETA, 1980) is unwarranted. The development of a distinct mantle cavity as opposed to grooves, and increasingly efficient gill structures and methods of passing water currents over them, are major themes in gastropod evolution and have undoubtedly per- mitted them to become large and active animals. It is significant to note, however, that, in animals with the limpet form, we can find everything from no discrete respiratory organs at all to animals with true molluscan ctenidia. And, in between, we can find animals with both ctenidia and secondary gills and animals with secondary gills but no ctenidia. We also see a range in the numbers of these structures and the way that they are housed. Modern limpets are much more adventuresome with res- piration than are coiled gastropods, and there is some good reason to believe that experimentation with respiration was important in Cambrian mollusks. This is suggested, for example, by the development of tubular structures (RUNNEGAR & POJETA, 1974; RUNNEGAR & JELL, 1976) in a number of tall cone-shaped, but minute, Cambrian shells. It must be remembered that modern deep-sea limpets are not living fossil holdovers from the Cambrian: There is too much evidence for secondary derivation from coiled ancestors. But, keeping this fact in mind, the living minute deep-sea limpets provide a wealth of anatomical infor- mation about the possibilities for organization and exper- imentation in the context of limpet geometry on a small scale. ACKNOWLEDGMENTS I thank the following persons and institutions for assis- tance in locating and borrowing specimens of deep-sea limpets upon which this work is based: Torben Wolff and Jorgen Knudsen (Zoological Museum, University of Co- The Veliger, Vol. 26, No. 2 penhagen), Ruth Turner (Harvard University), Anders Warén (University of Gothenberg), Eve Southward (Ma- rine Biological Association of the United Kingdom), James McLean (Los Angeles County Museum of Natural His- tory), Winston Ponder and Phillip Colman (The Austra- lian Museum), Joe Houbrick and Joe Rosewater (U.S. National Museum of Natural History), Paula Mikkelsen (Smithsonian Institution Fort Pierce Bureau), K. D. Spring (Continental Shelf Associates, Inc.), and the (U.S.) Bu- reau of Land Management. David Lindberg, James McLean and Winston Ponder provided helpful criticism and suggestions for improve- ment of the paper. I am particularly grateful to David Lindberg for his advice on interpretation of docoglossan radulae and for stimulating debates over the interpretation of limpet anatomy and morphology. Scanning electron mi- crographs were taken by the author in the Department of Anatomy, University of California, San Francisco; and the assistance of John Long and Jerry Morgan is grate- fully acknowledged. I thank George Shkurkin for English translations of the important Russian papers by Moska- lev. Copies of these translations are in the Earth Sciences Library, University of California. James Hickman as- sisted in preparation of the final draft and with intercon- tinental communications. The expeditions that have pro- vided limpet material from deep-sea hydrothermal vents have been supported by generous grants from the U.S. National Science Foundation, and my research on gastro- pod radulae by NSF grants DEB 77-14519 and DEB 80- 20992. Finally, I thank the Department of Malacology of The Australian Museum for the numerous courtesies ex- tended during the sabbatical leave that enabled completion of this project. LITERATURE CITED ANONYMOUS. 1980. New discoveries from the Galapagos Rift. Mus. Comp. Zool. Newsletter (Harvard University) 9(2): 1-2. BELYAEV, G. M. 1966. Hadal bottom fauna of the world ocean. Transl. by A. Mercado and published for the Smithsonian Institution and National Science Foundation by Israel Pro- gram for Scientific Translations (1972). 199 pp. BoucHET, P. & A. WAREN. 1979. The abyssal molluscan fau- na of the Norwegian Sea and its relation to other faunas. Sarsia 64:212-243. Brancu, G. M. 1981. The biology of limpets: physical factors, energy flow, and ecological interactions. Oceanogr. Mar. Biol. Ann. Rev. 19:235-380. Busu, K. J. 1897. Revision of the marine gastropods referred to Cyclostrema, Adeorbis, Vitrinella, and related genera; with descriptions of some new genera and species belonging to the Atlantic fauna of America. Trans. Conn. Acad. Sci. 10: 97-144. CAVANAUGH, C. M., S. L. GARDINER, M. L. JONEs, H. W. JannascH & J. B. WaTeRBURY. 1981. Prokaryotic cells in the hydrothermal vent tube worm Riftia pachyptila Jones: possible chemoautotrophic symbionts. Science 209:340-342. CoweEN, R. 1981. Crinoid arms and banana plantations: an economic harvesting analogy. Paleobiology 7:332-343. C. S. Hickman, 1983 CHRISTIAENS, J. 1979. Le genre Pectinodonta Dall, 1882 (Gas- tropoda, Acmaeidae). Informations de la Soc. Belge de Ma- lacologie, Serie 7, No. 1:17-22. DALL, W. H. 1882. On certain limpets and chitons from the deep waters off the eastern coast of the United States. Proc. U.S. Nat. Mus. 81(26):401-414. Dat, W. H. 1889. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico ... by the U. S. Coast Survey Steamer “Blake”... XXIX. Report on the Mollusca II. Gastropoda and Scaphopoda. Bull. Mus. Comp. Zool. (Harvard University) 18:1-492. DELL, R. K. 1956. The archibenthal Mollusca of New Zea- land. Dominion Mus. Bull. 18:1-235. FARFANTE, I. P. 1947. The genera Zeidora, Nesta, Emarginula, Rimula and Puncturella in the Western Atlantic. Johnsonia 2(24):93-148. FInuay, H. J. 1927. A further commentary on New Zealand molluscan systematics. Trans. N.Z. Inst. 57:320-485. FRETTER, V., A. GRAHAM & J. H. MCLEAN. 1981. The anat- omy of the Galapagos Rift limpet, Neomphalus fretterae. Malacologia 21:337-361. GeorGE, R. Y. & R. P. Hiccins. 1979. Eutrophic hadal ben- thic community in the Puerto Rico Trench. Ambio Special Rept. 6:51-58. HICKMAN, C. S. 1980. Gastropod radulae and the assessment of form in evolutionary paleontology. Paleobiology 6:276- 294. HICKMAN, C. S. 1981. Evolution and function of asymmetry in the archaeogastropod radula. Veliger 23:189-194. HICKMAN, C. S. 1982. The single helix: successful gastropods or successful paleontologists? Proc. Third North American Paleontol. Convention 1:237-242. Hickman, C. S. 1983. Ecological and phylogenetic implica- tions of the unusual radula of Laevinesta atlantica (Mollusca: Gastropoda). Veliger 25:323-325. HIckMAN, C. S. In press. Implications of radular tooth-row functional integration for archaeogastropod systematics. Malacologia. Housrick, R. S. 1978. The family Cerithiidae in the Indo- Pacific. Part I: The genera Rhinoclavis, Pseudovertagus and Clavocerithium. Monogr. Mar. Mollusca 1:1-130. JANNaSCH, H. & C. WirSEN. 1979. Chemosynthetic primary production of East Pacific sea floor spreading centers. BioScience 29:592-598. JANNASCH, H. & C. WiRSEN. 1981. Morphological survey of microbial mats near deep-sea hydrothermal vents. Applied and Environmental Microbiology 41(2):528-538. Kar, D., C. WiRSEN & H. JANNASCH. 1980. Deep-sea pri- mary production at the Galapagos hydrothermal vents. Sci- ence 207:1345-1347. KNIGHT, J. B. et al. 1960. Systematic descriptions. Jn: R. C. Moore (ed.), Treatise on invertebrate paleontology, Part I, Mollusca 1. Geol. Soc. Amer. and Univ. Kansas Press. pp. 169-331. Kuropa, T., T. HaBE & K. OyaAMa. 1971. The sea shells of Sagami Bay. Maruzen Co., Ltd., Tokyo. 489 pp. LINDBERG, D. R. 1979. Commentaire sur la revue de Chris- tiaens du genre Pectinodonta (Gastropoda, Acmaeidae). In- formations de la Soc. Belge de Malacologie, Serie 7, No. 4: 148. Lipps, J. H. & C. S. HICKMAN. 1982. Origin, age, and evo- lution of Antarctic and deep-sea faunas. Jn: W. G. Ernst & J. G. Morin (eds.), The environment of the deep sea. Rubey Series, Vol. II, Prentice Hall, Inc., New Jersey. pp. 324- 356. Page 91 McLean, J. H. 1979. A new monoplacophoran limpet from the continental shelf off southern California. Contrib. Sci. Nat. Hist. Mus. Los Angeles Co. 307:1-19. McLean, J. H. 1981. The Galapagos Rift limpet Neomphalus: relevance to understanding the evolution of a major Paleo- zoic-Mesozoic radiation. Malacologia 21(1—2):291-336. Menzies, R. J. & G. T. Rowe. 1969. The distribution and significance of detrital turtle grass, Thalassia testudinata, on the deep-sea floor off North Carolina. Int. Revue ges. Hy- drobiol. 54:217-222. Menzies, R. J., J. S. ZANEVELD & R. M. Pratr. 1967. Transported turtle grass as a source of organic enrichment of abyssal sediments off North Carolina. Deep Sea Res. 14: 111-112. Morton, J. E. 1967. Molluscs. Hutchinson University Li- brary, London. 264 pp. (5th ed.). MoskaLey, L. I. 1971. New data on the systematic position of the gastropod molluscs of the order of the Cocculinida Thiele, 1908. Abstr. of Repts., Fourth Conference on the Investigation of Mollusks. Acad. Sci. U.S.S.R., Nauka Pub- lishers, Leningrad. pp. 59-60. (in Russian) Moska.ey, L. I. 1973. Pacific Ocean Bathysciadiidae (Gas- tropoda) and forms similar to them. Zoological Journal 52(9): 1297-1303. (in Russian) MoskaLey, L. I. 1976. Concerning the generic diagnostics of the Cocculinidae (Gastropoda, Prosobranchia). Works of the P. P. Shirshov Institute of Oceanology, Acad. Sci. USSR 99:57-70. (in Russian) MoskaLey, L. I. 1977. Towards a revision of the molluscs (of the Family) of the Lepetidae (Gastropoda, Prosobranchia) from the world ocean. Works of the P. P. Shirshov Institute of Oceanology, Acad. Sci. USSR 108:52-78. (in Russian) MoskaLey, L. I. 1978. The Lepetellidae (Gastropoda, Pro- sobranchia) and forms similar to them. Works of the P. P. Shirshov Institute of Oceanology, Acad. Sci. USSR 113: 132-146. (in Russian) PELSENEER, P. 1901. Note sur Porganization du genre Bathy- sciadium. Bull. Zool. Soc. France 24:209-211. Pojeta, J. 1980. Molluscan phylogeny. Tulane Stud. in Geol. & Paleontol. 16(2):55-80. Rau, G. H. 1981. Hydrothermal vent clam and tube worm 8C/C: further evidence of non-photosynthetic food sources. Science 209:338-340. Rau, G. H. & J. I. HepGEs. 1979. Carbon-13 depletion in a hydrothermal vent mussel: suggestion of a chemosynthetic food source. Science 203:648-649. RUNNEGAR, B. & P. A. JELL. 1976. Australian Middle Cam- brian molluscs and their bearing on early molluscan evo- lution. Alcheringa 1:109-138. RUNNEGAR, B. & J. Poyera. 1974. Molluscan phylogeny: the paleontological viewpoint. Science 186:311-317. SCHOOLEY, C., C. S. HICKMAN & C. W. LANE. 1982. Com- puter graphic analysis of stereo micrographs as a taxonomic tool. Veliger 24(3):205-207. ScHEPMAN, M. M. 1908. The Prosobranchia of the Siboga Expedition, 1. Siboga Expeditie, monographie 49:1-107. STASEK, C. R. 1972. The molluscan framework. Jn: M. Flor- kin & B. T. Scheer (eds.), Chemical zoology, v. 7, Mollusca. Academic Press, New York. pp. 1-43. Stockton, W. L. & T. E. DELAcA. 1982. Food falls in the deep sea: occurrence, quality, and significance. Deep-Sea Res. 29(2A):157-169. SuTer, H. 1908. Additions to the marine molluscan fauna of New Zealand, with descriptions of new species. Proc. Ma- lac. Soc. 8(1):22-42. Page 92 THIELE, J. 1903. Die beschalten Gastropoden der deutschen Tiefsee-Expedition 1898-1899. B. Anatomisch-systema- tische Untersuchungen einiger Gastropoden. Berlin. pp. 147- 179. THIELE, J. 1908. Ueber die Anatomie und systematische Stel- lung von Bathysciadium, Lepetella, und Addisonia. Bull. Mus. Comp. Zool. (Harvard University) 52(5):79-89. THIELE, J. 1909. Cocculinoidea und die Gattungen Phenaco- lepas und Titiscania. In: Kuester, Systematisches Conch. Cabinet von Martini und Chemnitz. Bd. 2, Abt. 11a. Nurn- berg. 1-48. TurNER, R. D. 1973. Wood-boring bivalves, opportunistic species in the deep sea. Science 180:1377-1379. VERMEIJ, G. J. 1971. Gastropod evolution and morphological diversity in relation to shell geometry. J. Zool. (Lond.) 163(1):15-23. VERMEIJ, G. J. 1973. Biological versatility and earth history. Proc. Nat. Acad. Sci. USA 70(7):1936-1938. VERMEIJ, G. J. 1978. Biogeography and adaptation. Harvard Univ. Press, Cambridge, Mass. 332 pp. VERRILL, A. E. 1880. Notice on the remarkable marine fauna occupying the outer banks off the southern coast of New England. Am. J. Sci., Ser. 3, 20:115-120. The Veliger, Vol. 26, No. 2 VERRILL, A. E. 1882. Catalogue of the Mollusca added to the fauna of New England region. Trans. Conn. Acad. Sci. 5: 447-587. VERRILL, A. E. 1884. Second catalogue of Mollusca recently added to the fauna of the New England Coast and adjacent parts of the Atlantic, consisting mostly of deep-sea species, with notes on others previously recorded. Trans. Conn. Acad. Sci. 6:139-294. Wo LFF, T. 1976. Utilization of seagrass in the deep sea. Aquatic Botany 2:161-174. Wotrr, T. 1979. Macrofaunal utilization of plant remains in the deep sea. Sarsia 64:117-136. Wo rr, T. 1980. Animals associated with seagrass in the deep sea. In: R. C. Phillips & C. P. McRoy (eds.), Handbook of seagrass biology: an ecosystem perspective. Garland STPM Press, New York. pp. 199-224. YONGE, C. M. 1960. Mantle cavity, habits, and habitat in the blind limpet, Lepeta concentrica Middendorff. Proc. Calif. Acad. Sci. (4th Ser.) 31(4):103-110. YONGE, C. M. & T. E. THomMpson. 1976. Living marine mol- luscs. William Collins Sons & Co., Ltd., London. 288 pp. NOTE ADDED IN PROOF A paper has just appeared in which B. A. MARSHALL (1983. The family Cocculinellidae (Mollusca: Gastropo- da) in New Zealand. Nat. Mus. New Zealand, Records 2(12):139-143) provides new evidence on the biology and morphology of Cocculinella. It fails to support the rela- tionship with the Choristellidae proposed herein. The radula of a new shallow-water (13 m) New Zealand species is of a plan distinct from all the patterns illustrated in this paper, and the specimens were removed from fish bone. Marshall also reexamined syntypes of the type species and discovered that they were on fish bone, not on ‘water- logged wood’ as stated in the original description. The Veliger 26(2):93-95 (October 3, 1983) THE VELIGER © CMS, Inc., 1983 Additions to the Opisthobranch Mollusk Fauna of Marin County, California, with Notes on the Biology of Certain Species WILLIAM B. JAECKLE Department of Biological Sciences, University of Southern California, University Park, Los Angeles, California 90089 Abstract. Seven species of opisthobranch Mollusca from Marin County, California are discussed, and certain aspects of their biology presented. Three species are new additions to the faunistic list of this area; one sighting represents a geographical range extension. THE OPISTHOBRANCH mollusks of Marin County, Cali- fornia have been extensively documented in numerous re- ports (STEINBERG, 1960, 1963; Marcus, 1961, 1964; GOSLINER, 1968; GOSLINER & WILLIAMS, 1970, 1973; WILLIAMS, 1971; GODDARD, 1973; BEHRENS, 1980b; JAECKLE, 1981) in which a total of 74 species have been reported. Since 1977, this aspect of the littoral invertebrate fauna has been intermittently examined, and several species, previously unreported from this area, have been observed. These observations, as well as aspects of the biology of certain species found in Marin County are re- ported here. Species not previously reported from Marin County are denoted by an *; extension of the geographical range is denoted by a **. Systematic List of Discussed Species Subclass Opisthobranchia Order Nudibranchia Suborder Doridacea Family Cadlinidae Cadlina sparsa (Odhner, 1921)** Family Aldisidae Aldisa sanguinea (Cooper, 1863)* Suborder Dendronotacea Family Dotidae Doto kya Marcus, 1961 Suborder Arminacea Family Janolidae Janolus barbarensis (Cooper, 1863)* Janolus fuscus O’ Donoghue, 1924 Suborder Aeolidacea Family Flabellinidae Flabellina trilineata (O’ Donoghue, 1921) Family Tergipedidae Tenellia adspera (Nordmann, 1845)* Cadlina sparsa (Odhner, 1921) Two specimens of this eudoridoidean species were col- lected in a rocky, semi-protected littoral locality, the Fron- tier Arts Nature Reserve (latitude 37°52’22”N; longitude 122°36'56”W). These collections represent a geographical range extension, as well as an addition to the opistho- branch fauna of Marin County. The previous northern- most occurrence of Cadlina sparsa was Monterey Bay, Monterey County, California (BEHRENS, 1980a). The col- oration of these specimens differed from previously pub- lished accounts. Typically, the rhinophores are light yel- low to yellowish-brown in coloration (MCDONALD, 1977); however, the rhinophoral coloration of the collected spec- imens was light brown with numerous black spots on the rhinophoral lamellae. A voucher specimen has been de- posited in the invertebrate collection of the California Academy of Sciences bearing the voucher number CASIBP 030954. Aldisa sanguinea (Cooper, 1863) Although the geographical range of Aldisa sanguinea extends from Coos Bay, Oregon to San Diego, California and in the Gulf of California, Mexico (BERTSCH & Page 94 JOHNSON, 1982), this species has not been reported from Marin County. This species has been sighted in two rocky littoral areas, Muir Beach Overlook (latitude 37°52'21”"N; longitude 122°36'56”W) and the Frontier Arts Nature Reserve. Doto kya Marcus, 1961 This dendronotacean species is common in rocky litto- ral and estuarine habitats of Marin County. Despite the abundance of Doto kya in central California, the biology of this species is poorly known. In rocky littoral habitats of Marin County, specimens of D. kya have been observed preying upon the leptomedusan hydrozoan Obelia dichoto- ma (Linnaeus, 1758) and the anthomedusan species Sarsia sp. Doto kya, in estuarine areas of San Francisco Bay, feeds on O. dichotoma. The spawn masses of D. kya, de- posited on the hydrocauli of O. dichotoma colonies, have been found in both habitats. Janolus barbarensis (Cooper, 1863)* Janolus fuscus O’ Donoghue, 1924 GOSLINER (1981, 1982) synonymized the genus Antio- pella Hoyle, 1902, with Janolus Bergh, 1884, and rein- stated /. fuscus as a distinct species. Janolus barbarensis and /. fuscus have been viewed as a single species for 15 years (GOSLINER, 1982) and, as a result, the geographical ranges of both species have been combined by most au- thors. GOSLINER (1982) mentioned this range overlap but did not indicate their limits. In Marin County, /. barbar- ensis has been found on wharf pilings at Sausalito, San Francisco Bay (latitude 37°47'N; longitude 122°21’W) and J. fuscus has been collected at Muir Beach Overlook and the Frontier Arts Nature Reserve. Flabellina trilineata (O’ Donoghue, 1921) GOSLINER & GRIFFITHS (1981) synonymized the genus Coryphella Gray, 1850, with Flabellina Voigt, 1834, due to a lack of genus-specific morphological characters. In Marin County, Flabellina trilineata occurs in both estua- rine and marine habitats. At the Muir Beach Overlook study site, this species was observed feeding on the antho- medusan hydroid species Tubularia marina (Torrey, 1902) and Sarsia sp. The specimens of F. trilineata selectively preyed upon the tentacles of 7. marina, leaving the re- maining portion of the polyp intact. This feeding selectiv- ity differs from COOPER’s (1980) report that F. trilineata consumes the entire polyp of Tubularia crocea (Agassiz, 1862) in Elkhorn Slough, Monterey County, California. Tenellia adspera (Nordmann, 1845)* Large numbers of Tenellia adspera have been observed on wharf pilings in Sausalito, San Francisco Bay, feeding on Obelia dichotoma during May—August, 1982. Although T. adspera has been previously reported from San Fran- The Veliger, Vol. 26, No. 2 cisco Bay (STEINBERG, 1960, 1963; BEHRENS, 1980b), this is the first reported sighting from Marin County. The examined specimens were small (<5 mm in length) and exhibited a wide variation in the amount of black pigment spots present on the dorsal surface. Eastern Pacific spec- imens typically possess a light cream body coloration with “a few” black pigment spots on the dorsum (MCDONALD, 1977). BROWN (1980), however, reports that British spec- imens of 7. adspera exhibit a variable density in black “speckling.” In several specimens collected in Marin County, the black spots were so dense that the dorsum appeared black. Along with these darkly pigmented in- dividuals were collected specimens that exhibited the typ- ical coloration pattern. The radula of a darkly pigmented specimen agrees with the radula of 7. adspera as depicted in BRowN (1980). ACKNOWLEDGMENTS I would like to thank Dr. David W. Behrens and two anonymous reviewers for their critical review of this pa- per. LITERATURE CITED BEHRENS, D. W. 1980a. Pacific coast nudibranchs. A guide to the opisthobranchs of the northeastern Pacific. Sea Chal- lengers, Los Osos, Calif. 112 pp. BEHRENS, D. W. 1980b. A review of the literature on the opisthobranch fauna of San Francisco Bay. Opisthobranch Newsletter 12(4-12):34-37. BeRTSCH, H. & S. JOHNSON. 1982. Three new species of dorid nudibranchs (Gastropoda: Opisthobranchia) from the Hawaiian Islands. Veliger 24(3):208-218. Brown, G. H. 1980. The British species of the agdhdaczan Paral Tergipedidae (Gastropoda: Opisthobranchia) with a discussion of the genera. Zool. J. Linn. Soc. 69:225-255. Cooper, J. W. 1980. The influence of predation upon Tubu- laria crocea. In: P. Tardent & R. Tardent (eds.), Develop- ment and cellular biology of coelenterates. Elsevier North Holland Biomedical Press. pp. 281-286. GODDARD, J. 1973. Opisthobranchs of San Francisco Bay. Tabulata 6(4):8-10. GosLINER, T. M. 1968. A new record of Corambella steinbergae Lance, 1961. Veliger 11(2):147. GosLINER, T. M. 1981. The South African Janolidae (Mol- lusca: Nudibranchia) with descriptions of a new genus and two new species. Ann. S. Afr. Mus. 86(1):1-42. GOsLINER, T. M. 1982. The genus /anolus (Nudibranchia: Arminacea) from the Pacific coast of North America, with a reinstatement of Janolus fuscus O'Donoghue, 1924. Veliger 24(3):219-226. GosLinER, T. M. & R. J. GRIFFITHS. 1981. Description and revision of some South African aeolidacean Nudibranchia (Mollusca, Gastropoda). Ann. S. Afr. Mus. 84(2):105-150. GosLINER, T. M. & G. C. WILLIAMS. 1970. The opistho- branch mollusks of Marin County, California. Veliger 13(2): 175-180. GOSsLINER, T. M. & G. C. WILLLAMS. 1973. Additions to the opisthobranch mollusks of Marin County, California, with notes on their natural history. Veliger 15(4):352-354. W. B. Jaeckle, 1983 JAECKLE, W. B. 1981. New distributional records for two Cal- ifornia nudibranchs. Veliger 23(3):240. Marcus, E. 1961. Opisthobranch mollusks of California. Ve- liger 3(Suppl. 1):1-85. Marcus, E. 1964. A new species of Polycera (Nudibranchia) from California. Nautilus 77(4):128-131. McDonaLbD, G. R. 1977. A review of the nudibranchs of the California coast. Master’s Thesis, California State Univer- sity, Hayward. 1x + 337 pp. Page 95 STEINBERG, J. E. 1960. Rare and little known opisthobranch mollusks from the west coast of North America. Veliger 3(2):49. STEINBERG, J. E. 1963. Notes on the opisthobranchs of the west coast of North America III. Further nomenclatural changes in the order Nudibranchia. Veliger 6(2):63-67. WILLIAMS, G. C. 1971. New record of a color variation in Spurilla olwiae. Veliger 14(2):215-216. The Veliger 26(2):96-100 (October 3, 1983) THE VELIGER © CMS, Inc., 1983 Specialized Feeding in Mitrid Gastropods: Evidence from a ‘Temperate Species, Mitra idae Melvill ALLAN FUKUYAMA anpbD JAMES NYBAKKEN Moss Landing Marine Laboratories, Moss Landing, California 95039 Abstract. The neogastropod Mitra idae Melvill is the only member of the family Mitridae on the California coast. Although suspected of being a specialized predator, its food preferences have never been quantified. This study documents the feeding of M. idae on sipunculans. Laboratory experiments and gut analysis of field-collected specimens of M. idae have revealed that it is a specialized predator preferring the sipunculan Phascolosoma agassizu Keferstein. INTRODUCTION MEMBERS OF THE gastropod family Mitridae are most common in tropical waters, but some are present in tem- perate marine waters (CERNOHORSKY, 1970; ABBOTT, 1974; McLEAN, 1978; Kay, 1979). Although PONDER (1972) has given detailed descriptions of the digestive tract, little is known about the diet. The few species investigated have been shown to prey on sipunculans (KOHN, 1970, 1978; WEsT, 1979). Quantitative dietary information is available for only two members of the family, Mitra lit- terata Lamarck, of intertidal beaches in Hawaii, where it specializes on the sipunculan worms, Phascolosoma spp. and Aspidosiphon elegans Chamisso and Eyserhardt (KOHN, 1970), and Mitra flavocingulata of Easter Island, which KOHN (1978) has reported feeding on a species of Phas- colosoma. TAYLOR (1975) has found juveniles of Mitra assimilis in Hawaii to feed on small mollusks under lab- oratory conditions. There is some uncertainty as to the number of Mitra species occurring along the central California coast. WEST (1979) reports two, M. idae and M. catalinae, which, he reports, differ in radula structure and proboscis morphol- ogy, but which are otherwise alike externally except for size, M. catalinae being less than 30 mm in length. How- ever, MCLEAN (1978) considers M. catalinae synonymous with M. idae. ABBOTT (1974) questioned whether it was a dwarf form of M. idae, and Morris et al. (1980) men- tion only M. dae. If there are two species virtually indis- tinguishable externally except for size as WEST (1979) has suggested, it means that small animals (<30 mm) may be either juvenile M. idae or M. catalinae, and would be indistinguishable without dissection. We were not aware of the potential presence of two species until this study was completed and specimens had been discarded after gut dissection; thus, we were unable to undertake radula and proboscis dissections of small specimens to attempt to verify WEST’s (1979) contention. Therefore, those speci- mens that we have considered here as juvenile M. idae (<28 mm) could be a mixture of the two species if WEST’s (1979) contention that there are two species is substanti- ated by other malacologists. In either case, this does not invalidate our main argument of dietary specialization re- gardless of size. We prefer, pending presentation of more evidence, to follow MCLEAN (1978) and consider all Cal- ifornia mitrids as members of one species, M. idae. Mitra idae Melvill is restricted to kelp forests and the low intertidal zone (MCLEAN, 1978; Morris et al., 1980). Previous studies of this species have dealt with taxonomy and radular arrangement (CERNOHORSKY, 1966, 1970; Coan, 1966; CATE, 1967), functional morphology (WEST, 1979), and reproductive patterns and behavior (CATE, 1968; CHEss & ROSENTHAL, 1971; KELLOGG & LINDBERG, 1975). Although Morris e¢ al. (1980) report M. idae to be a specialized carnivore, no documentation is provided, nor is the prey identified. Moreover, CATE (1968) was unsuccessful in feeding a variety of potential prey items to M. idae in laboratory aquaria. She did not report whether sipunculans were offered to the M. idae. WEST (1979), however, has observed M. idae to feed upon the sipunculan Phascolosoma agassizu and the vermetid gas- tropod Petaloconchus montereyensis in the laboratory. His A. Fukuyama & J. Nybakken, 1983 accounts represent the most complete descriptions of feed- ing in the species. The present study further quantifies the diet of free- living Mitra idae and establishes preferences in laboratory experiments. This information should help to establish the role of predation by M. :dae in the kelp forest and the more general occurrence of sipunculan feeding within the family. Other species and genera remain to be studied to establish a general pattern of specialized feeding on si- punculan worms. METHODS The main study area was the kelp forest off Point Ca- brillo, Pacific Grove, California (36°37'N; 121°53’W). Additional collection of Mitra idae for gut content analyses was done off Del Monte Beach, Pacific Grove (36°37'N; 121°53’W), and Stillwater Cove (36°33’N; 121°56’W) (Figure 1). Descriptions of these kelp forests are given in Lowry & PEARSE (1973) for Point Cabrillo, in BURNETT (1972) and HOFFMAN (1981) for Del Monte Beach, and in FOSTER (1982) for Stillwater Cove. All sampling and collecting were done with SCUBA. Estimates of abundance of M. idae were made at Point Cabrillo in April and May, 1981. Water depth varied from 5 to 10 m. Random samples were taken along 25-m transects within a 0.25-m* quadrat. All M. idae found within these quadrats were counted and measured. Ani- mals used for gut-content analyses were collected haphaz- ardly during day and night, and were boiled shortly after collection to stop further digestion. After recording wet weight and measuring length from the apex of the shell to tip of siphonal canal, each shell was cracked open and the digestive tract dissected. Prey were identified and counted, and state of digestion noted. Laboratory feeding experiments were conducted to doc- ument prey preferences employing two species of sipun- culans and other potential prey, and to observe feeding behavior. One sipunculan species, Phascolosoma agassizu, was collected from local rocky intertidal areas and the holdfasts of drift kelp. The other sipunculan, 7hemuste pyroides (Chamberlain), was collected from subtidal shale. Prey wet weights were recorded before and after feeding experiments. One set of experiments evaluated the pref- erence of M. idae for the two species of sipunculans. In each experiment, one M. idae was placed in a dish with one P. agassizu and one T. pyroides. The experiment was repeated 2—4 times for each M. idae used. The dishes were covered with a fine screen to prevent escape and were held in laboratory aquaria with running sea water. Observa- tions continued until one of the prey was consumed or attacked. Sipunculans of similar size were used when pos- sible; however, individuals of 7. pyroides were generally larger than P. agassizu. In a second set of experiments, each species of prey was presented separately to a specimen of M. zdae. In addition to the two species of sipunculans, M. idae was also pre- Page 97 121° 55'W MONTEREY Pt. Pinos BAY PACIFIC OCEAN Pt Cabrillo 4 f Pacific. AN MN : Grove satay Monterey :: Bee Pt. Cypress 36°35 N | 2 3 oe S KILOMETERS Stillwater Cove Figure 1 Location of study areas in central California. sented with the echiuroid worm Urechis caupo Fisher and MacGinitie, and the holothuroids Eupentacta quinquese- mita (Selenka) and Cucumaria pseudocurata Deichmann. Urechis caupo is a close taxonomic relative of the sipun- culans, and the two holothuroids are similar in size and shape to sipunculans. RESULTS Individuals of Mitra idae were observed subtidally on rock walls, in crevices, in sand pockets, on old holdfasts of the kelp Macrocystis pyrifera, under the cover of red algae (Gigartina spp. and Rhodymenia spp.) and in tube beds of the polychaete Dzopatra ornata Moore. Although the abundance of M. idae was only 0.81/m? (SD = 0.54; n= 106) at Point Cabrillo, individuals were seen in aggre- gations of up to five individuals. KOHN (1970, 1978) ob- served 0.05-0.22 Mitra litterata/m* in Hawaii and 0.12 M. flavocingulata/m? at Easter Island. Size-frequency plots revealed two distinct size classes at Point Cabrillo, one near 15-20 mm and the other near 40 mm (Figure 2). CHESS & ROSENTHAL (1971) found mean shell lengths for copulating M. idae in southern California at 35 mm for males and 51 mm for females. Therefore, the smaller size group (<28 mm) probably was composed mostly of juveniles or a mixture of juveniles and M. catalinae. The gut contents of 41 M. :dae were examined (Table 1). Three of these individuals were collected at Stillwater Cove, six from Del Monte Beach, and the remaining an- Page 98 N=39 FREQUENCY O 8 I6 24 32 40 48 LENGTH (mm) Figure 2 Length-frequency histograms for Mitra idae collected at Point Cabrillo. Note the separation of a large (“adult”) and small (“juvenile”) size class. imals from Point Cabrillo. Food was found in 22 guts (54%). Sixteen guts contained whole sipunculans or frag- ments. Five had unidentifiable organic matter, and one gut had the remains of what appeared to be a nemertean worm. The sipunculan was identified as Phascolosoma agassizu in 13 of the 16 guts. Only five M. idae were collected at night. These were all collected at Point Ca- brillo. Two of these individuals had empty guts, two con- tained fragments of P. agassizi, and one had fragments of a nemertean. Most of the 16 juvenile M. idae had empty guts; however, one contained part of a P. agassizi, and The Veliger, Vol. 26, No. 2 Table 2 Choice of prey when Mitra idae was offered both sipun- culans, Phascolosoma agassizu and Themiste pyroides. Phas- colosoma agassizi was preferred in every experiment. Number of trials Prey consumed Mitra #1 5 Phascolosoma (5) Themiste (0) Mitra #2 6 Phascolosoma (4) Themiste (0) Mitra #3 3 Phascolosoma (1) * Themiste (Q) 14 10 0 * Phascolosoma attacked but later released (see text). five others contained unidentifiable organic matter. Among the 25 adults, two contained whole specimens of P. agas- sizit (<0.1 g), and 10 had parts of this species. The si- punculan prey in three adult guts could not be identified to species. The two-choice feeding experiments indicated that Mz- tra idae has a strong preference for one of the two sipun- culan species. Phascolosoma agassizii was attacked in 10 of 14 trials. Themiste pyroides was not attacked (Table 2). Feeding experiments containing one species of potential prey showed the same pattern. Phascolosoma agassizii was consumed in 30 of 45 trials (67%). Themuste pyroides was not consumed in any of five exposures to M. idae (Table 3). Fewer trials were made with 7. pyroides because of the longer duration of each experiment. In one experi- ment, a specimen of M. dae was starved for 17 days prior to exposure to a 7. pyroides. The T. pyroides was attacked and held by the proboscis of M. idae for over six hours before it was released. However, the M. idae did not con- sume the 7. pyroides, which suffered no apparent damage Table 1 Gut contents of Mitra idae collected at three areas in central California and in “adult” and “juvenile” sizes from all areas. Number of guts examined empty Area Stillwater Cove 3 0 Del Monte Beach 6 1 Point Cabrillo 3B) 18 41 19 Size of Mitra idae “Juveniles” <28 mm 16 10 “Adults” =>28 mm 25 ‘39. 41 19 * Fragment of nemertean worm. Number Amorphous Number with organic sipunculans matter Other* 0 3 0 5 0 0 1 2 i 16 5 1 1 5 0 15 0 i 16 5 1 A. Fukuyama & J. Nybakken, 1983 Page 99 Table 3 Feeding experiments offering Mitra idae either one or the other of the sipunculans, Phascolosoma agassizu or The- muste pyroides. Only P. agassizii was consumed. Number Total Number Pierced Attacked, of number — eaten and not Prey species trials consumed whole eaten eaten Phascolosoma 45 30 25 5 1 Themiste 5 0) 0) 0) 1 from the attack, and it remained alive in a holding tank. Mitra idae neither attacked nor consumed the three sipun- culan-like prey, Urechis caupo, Eupentacta quinquesemita and Cucumaria pseudocurata, in similar feeding experi- ments. Laboratory feeding observations indicated two methods of attack and consumption of the Phascolosoma agassizit. Prey either were ingested whole or the integument was punctured by the radula and the internal contents re- moved through the hole, presumably by a mechanism de- scribed by WEsT (1979). Considering all experiments, 81.5% of the P. agassizii were consumed whole (n = 40). Prey weighing less than 0.3 g were always ingested whole, but larger individuals (up to 0.8 g) were taken whole as well. Only seven individuals were eaten by puncturing the integument of the sipunculan. These prey were relatively large compared to those eaten whole, and ranged in size from 0.3 to 0.76 g (Figure 3). The weight loss of pierced prey averaged 64 + 17% (n =7) and all pierced animals died. DISCUSSION Mitra idae seems to be a specialized carnivore. Although we have tested only one other sipunculan species, gut con- tents and laboratory feeding experiments indicate that the sipunculan Phascolosoma agassizu is the dominant or pre- ferred prey. Phascolosoma agassizu is the most abundant sipunculan at all three study areas on the Monterey Pen- insula (HOFFMAN, 1981; personal observations). Despite the presence of another sipunculan, Themiste pyroides, in the field (HOFFMAN, 1981) and in feeding experiments, this species was not consumed by the gastropod (Tables 2, 3). An echiuran worm and two holothurians also were not eaten in laboratory feeding experiments. Moreover, the guts of M. idae collected in the field did not contain any of the abundant polychaete groups that co-occur with M. idae (e.g., onuphids, terebellids, sabellids, syllids). Specialized feeding on sipunculan worms was first es- tablished by KOHN (1970, 1978) for the tropical species Mitra litterata and M. flavocingulata. Now, evidence sug- gests a similar situation for the temperate species M. idae. Although three species is a small sample of the 500 or so OD Prey consumed whole n=32 W Prey with radula hole n=7 FREQUENCY O Ol 02 03 04 05 06 O7 08 PREY WEIGHT (g) Figure 3 The relationship between the size of Phascolosoma agassizi and the method of prey consumption. Smaller prey were generally eaten whole and only larger individuals were eaten by punctur- ing the integument of the prey and inserting the proboscis. species from this family, the geographical separation of M. idae from M. litterata and M. flavocingulata suggests that a general feeding pattern concentrating on sipuncu- lans may emerge, at least for this genus. Indeed, WEST (1979) has determined that the epi-proboscis of M. idae functions to locate, retrieve, and hold soft-bodied prey, in particular, sipunculans. Perhaps sipunculan worms will be the dominant prey for the entire family. Present know]- edge does not allow a critical evaluation of this hypothesis, but the distinct morphological similarities in the alimen- tary canal among members of the family Mitridae re- ported by PONDER (1972) would argue for this. Laboratory feeding observations indicated two modes of prey consumption for Mitra idae. A few sipunculans were eaten by puncturing the integument, inserting the pro- boscis, and removing the body contents, as WEST (1979) has described; but most were consumed whole, especially the smaller-sized prey. The proboscis of mitrids may be extended the length of the shell (WILSON & GILLETT, 1972; personal observations). The ability to extend the proboscis a long distance is of practical importance in feeding upon prey that lodge deep in crevices, holdfasts, or other substrata. The two modes of consuming prey also may be of significance in that the extraction of a strongly wedged sipunculan is not necessary. The frequency of puncture feeding is difficult to assess in the field, as empty guts can indicate either the absence of prey or faster diges- tion of the internal contents of a pierced prey compared to the outer integument of an animal consumed whole. Gut contents and laboratory experiments suggest that the feeding habits of Mutra idae may be more closely re- lated to behavioral preferences and not to prey availabil- ity. The ability to survive without food for at least 8-12 months (CATE, 1968; personal observations) supports the Page 100 idea that prey availability may not be of critical impor- tance. Size selective predation also may be feasible. Evi- dence from the few whole sipunculans found in gut con- tents of M. idae, as well as laboratory observations, seem to support this idea. Perhaps future studies will reveal important similarities in the foraging tactics of mitrid gas- tropods that will allow generalizations to be made about the family. Specialized feeding on sipunculan worms by tropical and a temperate Mitra suggests that similar evo- lutionary processes affected the development of these predator-prey interactions. ACKNOWLEDGMENTS We thank Michael Kellogg, Valerie Hironaka, Tim Herrlinger, John Oliver, and Mark Silberstein for re- viewing an early draft of the paper. We thank Hopkins Marine Station of Stanford University for use of their facilities. Special thanks go to Lynn McMasters for draw- ing the figures and Sheila Baldridge for her help in ob- taining library material. Finally, the discussions with Alan Kohn and John Oliver greatly aided in the preparation of the paper. We also appreciate the suggestions of several anonymous reviewers. LITERATURE CITED AsBBoTT, R. T. 1974. American seashells, 2nd ed. Van Nos- trand Reinhold Co. 663 pp. BuRNETT, N. A. 1972. The ecology of the benthic community of bivalve molluscs in the shale at the Monterey sewer out- fall. Master’s thesis, San Francisco State Univ. 56 pp. CaTE, J. M. 1967. The radulae of nine species of Mitridae. Veliger 10:192-195. CaTE, J. M. 1968. Mating behavior in Mitra idae Melvill, 1893. Veliger 10:247-252. CERNOHORSKY, W. O. 1966. A study of mitrid radulae and a tentative generic arrangement of the family Mitridae (Mol- lusca: Gastropoda). Veliger 9:101-126. CERNOHORSKY, W. O. 1970. Systematics of the family Mitri- dae and Volutomitridae. Bull. Auckl. Inst. Mus. 8:1-190. The Veliger, Vol. 26, No. 2 CHEss, R. J. & R. J. ROSENTHAL. 1971. On the reproductive biology of Mitra idae (Gastropoda: Mitridae). Veliger 14: 172-176. Coan, E. 1966. Nomenclatural units in the gastropod family Mitridae. Veliger 9:127-137. FosTeR, M. 1982. The regulation of macroalgal associations in kelp forests. In: W. Srivastava (ed.), Synthetic and deg- radative processes in macrophytes. DeGruyter and Co., Ber- lin. pp. 185-205. HoFFMAN, R. G. 1981. The ecology of the benthic and endo- lithic communities of a rocky reef in the kelp beds off Del Monte Beach, Monterey, California. Master’s thesis, U.S. Naval Postgraduate School. 117 pp. Kay, E. A. 1979. Hawaiian marine shells. Reef and shore fauna of Hawaii, Section 4. Mollusca. Bernice P. Bishop Museum Spec. Publ. 64(4). 653 pp. KeELLoGG, M. G. & D. R. LINDBERG. 1975. Notes on the spawning and larval development of Mitra idae Melvill (Gastropoda: Mitridae). Veliger 18:166-167. Koun, A. J. 1970. Food habits of the gastropod Mitra litterata Lamarck: relation to trophic structure of the intertidal ma- rine bench community in Hawaii. Pac. Sci. 24(4):483-486. Koun, A. J. 1978. Gastropods as predators and prey at Easter Island. Pac. Sci. 32(1):35-37. Lowry, L. W. & J.S. PEARSE. 1973. Abalones and sea urchins in an area inhabited by sea otters. Mar. Biol. 23:213-219. McLean, J. H. 1978. Marine shells of southern California. Science Series 24, Nat. Hist. Mus. of L.A. County. 104 pp. Morris, R. H., D. P. ABBoTT & E. C. HADERLIE. 1980. In- tertidal invertebrates of California. Stanford Univ. Press, Stanford, Calif. 690 pp. PoNDER, W. F. 1972. The morphology of some mitriform gastropods with special reference to their alimentary and reproductive systems (Neogastropoda). Malacologia 11(2): 295-342. TayLor, J. B. 1975. Planktonic prosobranch veligers of Ka- neohe Bay. Doctoral thesis, University of Hawaii, Hono- lulu. West, T. L. 1979. Structure, function, and functional mor- phology of Mitra idae Melvill 1893 and M. catalinae Dall 1920. Master’s thesis, University of the Pacific. 97 pp. WILSON, B. R. & K. GILLETT. 1972. Australian shells. Charles Tuttle Co. 168 pp. The Veliger 26(2):101-105 (October 3, 1983) THE VELIGER © CMS, Inc., 1983 Homing in the West Indian Chiton Acanthopleura granulata Gmelin, 1791 DAVID MOOK Harbor Branch Foundation, Inc., R.R. 1, Box 196, Ft. Pierce, Florida 33450 Abstract. The homing behavior of the chiton Acanthopleura granulata was studied on high-, moderate-, and low-energy rocky shores in the Bahamas. Chitons on low-energy shores tended to home more frequently than chitons on higher-energy shores. Chitons on higher-energy shores tended to make more frequent nighttime excursions from their home and to travel greater distances than chitons on low-energy shores, possibly because lower food availability on higher-energy shores may force the chitons to increase their grazing effort. INTRODUCTION HOMING IS A very common phenomenon in some groups of mollusks. It has been described in several species of gastropods (MCFARLANE, 1981; WELLS, 1980; Cook, 1979; CoLLINs, 1977; MacKay & UNDERWOOD, 1977; and others) and chitons (GLYNN, 1970; THORNE, 1968). Acanthopleura granulata Gmelin, 1791, a West Indian chi- ton, is very common on the intertidal limestone shores in the Bahamas. Individuals generally remain stationary on their rock substratum during the daylight hours and for- age for endolithic and surficial algal food at night (GLYNN, 1970). Unlike that of some other species of chitons, the foraging of A. granulata does not seem to be influenced by tidal height (NEWELL, 1979). Upon completion of its for- aging activity, A. granulata often returns (homes) to its original site of attachment. Because individuals do not always come to rest exactly on their old site of attachment and often orient themselves differently (GLYNN, 1970), the term “homing” as used in this study is not being used in the strictest sense (COLLINS, 1977). In this study, I de- scribe the differences in homing behavior of individual Acanthopleura granulata inhabiting high-, moderate-, and low-wave energy shorelines in the Bahamas. MATERIALS anp METHODS In order to compare homing activity of Acanthopleura granulata on high-, moderate-, and low-energy shorelines, observations were made on various rocky limestone (Kar- renfeld) (GARY et al., 1974) shorelines in the Bahama Island chain (Figure 1). High-energy zones (Stations 1, 4, 6, 7) were on shorelines exposed to the open ocean and were generally subjected to surf of one to several meters in height. These shorelines were usually wetted by spray and wave swash for a large part of the day. Moderate- energy shorelines (Stations 5, 8) were exposed to some waves and spray at high tides and during windy weather. Low-energy shorelines (Stations 2, 3) were exposed to little or no wave or spray activity and were wetted only at flood tides when water covered the substrate. Individual Acanthopleura granulata and their attach- ment sites (=homes) were marked with nail polish. The following day, the marked homes were checked to deter- mine whether the chitons were present or absent from where they were marked the previous day. If the chiton was absent from the home it occupied the previous day, the animal was located (if possible) and its new site of attachment (home) was marked. The distance between the original and new home was measured. If the animal could not be located, it was counted as absent. Observations were made from two (Indian Cay) to six (San Salvador) consecutive days. Because many chitons (especially in low- and moderate- energy zones) were generally at their original home when observed each day, observations also were made on nights when low tides occurred to determine whether those an- imals were foraging at night and returning to their homes or simply were not leaving their homes at all. The dis- tances between the foraging chitons and their homes were measured. Personal safety considerations prevented night- time observations on high-energy shores. A Student’s ¢ test was used to compare distances trav- eled between chitons (Acanthopleura granulata) in low- energy areas and chitons in moderate-energy areas. Be- Page 102 The Veliger, Vol. 26, No. 2 — is. 9 BLACK ROCK LN % 14 A 8 9 26°N ——4 30km ! o js 76°W € e, NEW PROVIDENCE SS. ‘> * & ‘ f - Ly ) oS Guile Ln FZ © 7 ae \/6 2 \ SAN SALVADOR \ \ <<) RUM Cay Figure 1 Localities where observations were made. 1: Indian Cay, near West End, Grand Bahama Island. 2, 3, 4: Black Rock, near Moors Island, Abaco Chain. 5, 6: North Point, San Salvador Island. 7, 8: Barkers Point, San Salvador Island. cause an F,,,, test indicated that distance data were not normal, all distance data were transformed using a log transformation. Caloric values of substrata in a moderate- (Station 8) and low- (Station 2) energy zone were compared. Samples of the limestone substrate were taken by scrubbing the substrate with a wire brush and rinsing the loosened lime- stone and its associated endolithic and surficial algae into a jar. The samples were kept frozen until they were re- turned to the laboratory where they were filtered, dried, and ground into a powder. Samples were mixed with ben- zoic acid (50% acid, 50% sample) and their caloric value measured in a Phillipson microbomb calorimeter (PHIL- LIPSON, 1964). Because the topography and density of the substratum were similar on all islands studied, caloric values are re- ported in calories per milligram of substratum rather than in calories per area of substratum. Irregularities in the substratum surface made it difficult to remove limestone evenly from a specific area. Twelve replicates were done for each sample. A ¢ test was used to compare the caloric value of samples taken from low- and moderate-energy zones. RESULTS From a total of 162 observations, about 43% of the Acan- thopleura granulata had moved from the point at which they were marked the previous day. Far fewer chitons were found at the point where they were marked in high- energy zones (15%) than in moderate- (56%) or low- (93%) energy zones (Figure 2). Although only a small number of chitons observed in low-energy zones were found absent from their home the following day, nighttime observations showed that 59% of the marked chitons in low-energy zones were away from their home at night. All of these chitons had returned to their home by morning. In mod- erate-energy zones, about 70% of the marked chitons were observed away from their home at night, but only 56% of these chitons had returned back to their home by morning (Figure 3). Since nighttime observations could not be made in high-energy zones, it is not known whether the chitons D. Mook, 1983 Page 103 100 a5) 5 O 50 co) ae 25 O Low Moderate High Energy Energy Energy N=44 N=86 N=32 Figure 2 Percent of chitons found at marked sites one day after marking. N = total number of animals observed. that were present at the point of marking the following day had homed or simply had not moved during the night. Acanthopleura granulata in low-energy zones which were not observed at their home the following day (only three) were observed from 5 to 35 cm away from their home. In moderate-energy zones, chitons which had not returned to their home by the following day were observed 5 to 500 cm from their home (Table 1). No distance measurements were done with high-energy-zone chitons because only a small percentage of the marked chitons in the high-energy zone were found the following day. The fact that marked chitons were not found within several meters of their home suggests that either these chitons could have traveled far- ther from their home than moderate-energy-zone chitons and relocated in areas that were inaccessible, or they were removed by predators or wave action. A ¢ test detected no significant difference between the distances of low-energy- zone chitons and moderate-energy-zone chitons from their home. The reason no difference was detected could be due to the small number of individuals that did not home in the low-energy zone; this would make a significant dif- ference difficult to detect. At night, chitons (Acanthopleura granulata) in moderate-energy zones ranged significantly 100 (GS) 100 pausinjar % 25 pauinjas %, @) Low Moderate Energy Energy N=13 N=23 Figure 3 Percent of chitons observed away from site of marking at night in moderate- and low-energy zones. Solid line inside bar indi- cates percentage of these animals which had returned to the marked area the following morning. farther (6-65 cm) from their home than chitons in low- energy zones (2~30 cm; P < 0.05) (Table 1). Caloric values of the substrate at the moderate-energy site were significantly lower (x = 0.1767 cal/mg) than the caloric value of the substrate at the lower energy site (x = 0.5056 cal/mg; P < 0.05). DISCUSSION The observations in this study suggest that Acanthopleura granulata living in lower-energy zones may have a greater tendency to stay attached to their home than those living in higher-energy zones. When chitons in lower-energy zones do leave their home to graze, they tend to travel shorter distances and to have a greater tendency to return to their home than do chitons in moderate- and high- energy zones. The reasons for development of homing in chitons are unclear. Several explanations, such as protection from dis- lodgment (LINDBERG & Dwyer, 1983; COLLINs, 1977), Page 104 The Veliger, Vol. 26, No. 2 Table 1 Distance (cm) that chitons were found away from the point of marking (home) the night and day following marking. NS means ¢ value not significant at P = .05. Day after Night after marking marking Low energy Range (cm) 5-35 2-30 X (n) 18.3 (3) 12.00 (11) Moderate energy Range (cm) 5-500 6-65 X (n) 61.12 (32) 36.22 (9) t value 1.37 (NS) DAZ predator avoidance (WELLS, 1980), maximum utilization of food resources (MacKay & UNDERWOOD, 1977), and protection from desiccation (VERDERBER et al., 1983; COLLINS, 1977; and others) have all been suggested for other homing mollusks. It is unlikely that protection from dislodgment is an important reason for homing in Baha- mian Acanthopleura granulata because chitons in high-en- ergy zones, where dislodgment is more likely due to high wave activity, have a lesser tendency to home. Whether homing is an important mechanism for predator avoidance was not determined. However, little evidence of predation and few predators (except humans at some sites) were observed at any of the study sites, suggesting that homing may not be important for predator avoidance in the Ba- hamas. It is also not clear whether Bahamian Acanthopleura granulata home for protection against desiccation. The fact that chitons home less in high-energy areas, which are frequently wetted by wave action, could indicate that homing may be an adaptation to prevent desiccation. However, because homing may be accomplished by olfac- tion and/or trail retracing (MCFARLANE, 1981; Cook, 1979; THORNE, 1968), the reduced homing in high-energy areas may be caused simply by trails and olfactory stimuli being washed away by wave action. If homing is impor- tant for the prevention of desiccation, the loss of a trail to home could be of little consequence for chitons in high- wave-energy areas where desiccation is not a major prob- lem. The lower caloric food value of the moderate-energy substrate zones may explain the more frequent and more distant nighttime excursions of chitons (Acanthopleura granulata) in moderate- (and possibly high-) energy zones. Studies with filter-feeding bivalves (WINTER, 1978) and grazing winkles (Littorina littorea Linné, 1758) (NEWELL et al., 1971) have shown that feeding effort (filtration rates or radular movement) tended to decrease as food became more available. For example, winkles tended to have higher radula movement rates when their food supply was lim- ited on a temporal basis (NEWELL et al., 1971). NEWELL (1979) suggested that when food availability was lower, feeding effort may have had to increase in order for the organism to continue to get enough food to sustain itself. In the case of the chitons examined in this study, the lower caloric value of the moderate-wave (and possibly high-) energy substrata may force animals in these zones to for- age at more frequent intervals and to travel longer dis- tances (increase grazing effort) in order to obtain the nec- essary calories to sustain themselves. Other factors in addition to wave exposure or food availability may affect homing behavior in chitons. Ele- ments such as inter-island variability, differences in sub- strate morphology, and differences in substrate productiv- ity have not been addressed in this study and would be interesting subjects for future studies with Acanthopleura granulata. ACKNOWLEDGMENTS I wish to thank Richard Houbrick, Paul Mikkelsen, and Robert Virnstein for their helpful comments on the manu- script. I am grateful to Pat Linley for proofreading the manuscript, and I am indebted to Charles Hoskin for his support during this study. I would like to thank Donald and Kathy Gerace of the College Center of the Finger Lakes field station on San Salvador for their hospitality and help. This is Harbor Branch Foundation Contribu- tion No. 339 and CCSL Contribution No. 64-B. LITERATURE CITED Co.uins, L. S. 1977. Substrate angle, movement and orien- tation of two sympatric species of limpets, Collisella digitalis and Collisella scabra. Veliger 20:43-48. Cook, A. 1979. Homing in the Gastropoda. Malacologia 18: 315-318. Gary, M., R. MCAFEE, JR. & C. Wor. 1974. Glossary of geology. American Geological Institute. Washington, D.C. xiv + 805 pp. + 52 pp. GLynn, P. W. 1970. On the ecology of the Caribbean chitons Acanthopleura granulata Gmelin and Chiton tuberculatus Linné: density, mortality, feeding, reproduction, and growth. Smithsonian Institution Press No. 60: 21 pp. LINDBERG, D & K. Dwyer. 1983. The topography, formation and role of the home depression of Collisella scabra (Gould) (Gastropoda: Acmaeidae). Veliger 25:229-234. MacKay, D. A. & A. J. UNDERWOOD. 1977. Experimental studies in the intertidal patellid limpet Cellana tramoserica (Sowerby). Oecologia 30:215-237. McFar ane, I. D. 1981. In the intertidal homing gastropod Onchidium verruculatum (Cuv.) the outward and homeward trails have a different information content. J. Exp. Mar. Biol. Ecol. 51:207-218. NEWELL, R. C. 1979. Biology of intertidal animals. Marine Ecological Surveys L & D, Faversham. ix + 781 pp. NEWELL, R. C., V. I. PYE & M. AHSUNULLAH. 1971. Factors affecting the feeding rate of the winkle Littorina littorea. Mar. Biol. 9:138-144. PHILLIPSON, J. 1964. A miniature bomb calorimeter for small biological samples. Oikos 15:130-139. THORNE, M. J. 1968. Studies on homing in the chiton Acan- D. Mook, 1983 thozostera gemmata. Aust. J. Mar. Freshwater Res. 19:151- 160. VERDERBER, G. W., S. B. Cook & C. B. Cook. 1983. The role of the home scar in reducing water loss during aerial exposure of the pulmonate limpet Siphonaria alternata (Say). Veliger 25:235-243. Page 105 WELLS, R. A. 1980. Activity pattern as a mechanism of pred- ator avoidance in two species of acmaeid limpet. J. Exp. Mar. Biol. Ecol. 48:151-168. WINTER, J. 1978. A review of the knowledge of suspension feeding in lamellibranchiate bivalves, with special reference to artificial aquaculture systems. Aquaculture 13:1-33. The Veliger 26(2):106-112 (October 3, 1983) THE VELIGER © CMS, Inc., 1983 Aspects of Reproduction in Some Enoploteuthid Squids from Hawaiian Waters! by LISA M. BOUCHER Mid-Pacific Research Laboratory, Enewetak, Marshall Islands, P.O. Box 1768, APO San Francisco 96555 Abstract. This study investigates aspects of reproduction of three enoploteuthid squid species, Pteryg- oteuthis microlampas Berry, Pyroteuthis addolux Young, and Abraliopsis sp., captured in Hawaiian waters during a 17-month, midwater-trawl sampling program. Pterygioteuthis microlampas differed from the other species in having smaller ova and higher relative fecundity. Juveniles of P. microlampas were most often caught in spring months, indicating a seasonal spawning tendency in this species. Seasonality of spawning in one or more species also was indicated by fall and winter peaks in enoploteuthid larval abundance. Although the majority of females in the larger size classes had mated prior to capture, only a minority of ova in any specimen were mature and the ratios of female gonad dry weight to total body dry weight were generally low, suggesting that full reproductive maturity was not observed in these specimens. In light of these results, net avoidance is considered, as is the possibility that partially spent females were captured. Also raised is the question of whether these species are able to spawn in more than one spawning season. INTRODUCTION APPROXIMATELY 50 species of pelagic squid are known to occur in Hawaiian waters (YOUNG, 1978). Most of these are small in size and low in abundance. Those most com- monly captured in mesopelagic trawl samples are of the family Enoploteuthidae. Previous studies of enoploteuthid species in Hawaii have primarily involved their system- atics (BERRY, 1914), vertical distribution (YOUNG et al., 1980; YOUNG, 1978; ROPER & YOUNG, 1975), photobiol- ogy (YOUNG et al., 1980; YOUNG, 1972, 1973), and occur- rence in the diet of larger carnivores (KING & IKEHARA, 1956; CLARKE, 1971). A few papers have treated aspects of reproduction of enoploteuthids occurring in temperate waters (e.g., SASAKI, 1914; OKUTANI & McGowan, 1969; PeEaARCY, 1965). However, the life histories and reproduc- tive biology of the Enoploteuthidae remain largely un- studied. The following study was undertaken to investigate fe- cundity, size at maturity, larval abundance, seasonality, and other aspects of reproduction of the three most com- monly sampled enoploteuthid squid species in Hawaiian waters. Many of the results must be considered prelimi- ' Hawaii Institute of Marine Biology Contribution No. 666. nary due to the sampling difficulties inherent in studies of this kind. Nevertheless, this study contributes original data that provide new insights into the life histories of these little-known squid species. MATERIALS anpD METHODS This study is based primarily on specimens collected at approximately monthly intervals 20 km west of Oahu, Hawaii, between August, 1977, and October, 1978. Ju- venile and adult enoploteuthids were collected in night- time oblique samples with a 3-m Isaccs-Kidd midwater trawl (IKMT) and a 1.25-m diameter bongo net (BB). Larvae (<5 mm dorsal mantle length) were taken pri- marily in standard bongo nets (SB) of 70-cm diameter. Each net was equipped with a flow meter and a time- depth recorder. The IKMT had a mouth area of 7.7 m? and a mesh of about 6 mm, except for a 1-m cod end of 333 um. The IKMT was towed at 2 m/s and was re- trieved at 40-50 m/min. The average maximum depth reached was 325 m (+30 m), and the average volume filtered per tow was 101 X 10? m®. The BB mesh was 2.5 mm. It was towed at about 1.75 m/s and retrieved at 40- 50 m/min. The average maximum depth was 304 m (+6 m), and average volume filtered per net was 9434