me OF a 2 Lav r ELS wid Pee FA Eritsie hole hors hae SPCR RES £rKe heya Bory pears ae Seleaas (iyi eat ba Sra acy Pe Or RSS OEH : Hig te et gl ed, ry he Pett Rats, oN Bie yeoenhgieae Meee eatin anita elo les a4 7 = * 4 * 14 oh er reek ye TAR Risat eT Pernt Sain he 2 at Vella bates basa om, wah ans feat icin ives eit Bahihin aly Aa ae Tene UN TEED CLERER LA Taha Nageonga APE 8 dog phen Aika Tavs eee : ‘ Wadena eg E ARAB LTS BARS foe ORE LD DORE Gale fete : Mi esautuessiieaye ; ‘ Fit teats mi i ; ; ; SOUT Ae . te ; tit OR seb aety ERS wilted vane any Utne, bay Sptaelee ae rtd gre aba atgeiye i nani btsiye nt Sethe ts wt Bets Biri ty. oes Res aad a mon a Ptyath Frater, pager thatace Pigswie ria fig oat ne as ne, Syeeae 1 at Hee Mes eat i bes caee ur speetee Fa ANB) PRON Mal ALTE TREES EME Pde pat Sinbsd See t ‘ fhe an Aidit ih Br oe AST # es THE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 42 January 4, 1999 to October 71, 1999 TABLE OF CONTENTS Number 1 (January 1, 1999) Growth, seasonality, and dispersion of a population of Aplysia vaccaria Winkler, 1955 LisA ANGELONI, JACK BRADBURY, AND ALEXIS CHAINE . . 1 Metazoan parasites and pearls in coexisting mussel species: My- tilus califorianus, Mytilus galloprovincialis, and Septifer bi- furcatus, from an exposed rocky shore in Baja California, northwestern Mexico J. CACERES-MARTIiNEZ AND R. VASQUEZ-YEOMANS ..... 10 Pea crab, Pinnotheres ostreum Say, 1817, in the eastern oyster, Crassostrea virginica (Gmelin, 1791): prevalence and ap- parent adverse effects on oyster gonad development FRANCIS X. O’BERIN AND RANDAL L. WALKER ....... 17 The genus Littoraria Griffith & Pidgeon, 1834 (Gastropoda: Lit- torinidae) in the tropical Eastern Pacific Davip G. REID Genetic and environmental control of growth and reproduction of Phacosoma japonicum (Bivalvia: Veneridae) SHIN ICHIC SATO Hemera aconcu ie tain ucne wench omen ue 54 A worldwide review of the food of nudibranch molluks. Part II. The suborder Dendronotacea GARY MCDONALD AND JAMES NYBAKKEN .........-.-.- 62 The Giant Amazonian Snail (Pulmonata: Acavidae) beats them all FRANK P. WESSELINGH AND EDMUND GITTENBERGER .... 67 Gastropods and intertidal soft-sediments, the case of Chilina ovalis Sowerby (Pulmonata: Basommatophora) in south- central Chile PEDRO QUIJON AND EDUARDO JARAMILLO ........... V2 Designation of lectotype for Haliotis crebrisculpta Sowerby, 1914, with a discussion of H. clathrata Reeve, 1846 (non Lichtenstein, 1794) KATHERINE A. STEWART AND DANIEL L. GEIGER ..... . 85 Lioconcha (Sulcilioconcha) caledonensis sp. nov., a species of Veneridae (Bivalvia) from New Caledonia MAry ELLEN HARTE AND KEVIN L. LAMPRELL........ 97 Number 2 (April 1, 1999) Utilization of artificial diets and effect of protein/energy rela- tionship on growth performance of the apple snail Pomacea bridgesi (Prosobranchia: Ampullariidae) ROBERTO MENDOZA, CARLOS AGUILERA, JESUS MONTEMAYOR, AND GABINO RODRIGUEZ ........ 101 A new species of Depressigrya? (Gastropoda: Peltospiridae) from cold-seep carbonates in Eocene and Oligocene rocks of western Washington JAMEs L. GOEDERT AND STEVEN R. BENHAM ........ 112 Calyptogena diagonalis, a new Vesicomyid bivalve from sub- duction zone cold seeps in the eastern North Pacific JAMES P. BARRY AND RANDALL E. KOCHEVAR ....... 117 Histological description of the gonad, reproductive cycle, and fertilization of Pisidium amnicum (Miiller, 1774) (Bivalvia: Sphaeriidae) ReeARAUIOFPANDIMacAa RAMOS ite teins re ene ee ahaa 124 The Eastern Pacific Sportellidae (Bivalvia) EUGENERV GOAN) nce aval Ieee ecu en ee 132 Laboratory observations of the feeding behavior of the cirrate octopod, Grimpoteuthis sp.: one use of cirri JAMES. G=ELUNT) Bisritacs.o chess Oe er nae ae 152 Ontogenetic changes in boring behavior by the rock-boring bi- valve, Barnea manilensis (Pholadidae) Y ASUHIRO ITO li A new species of gastropod of the genus Trophon Montfort, 1810 (Mollusca: Gastropoda: Muricidae) from subantarctic waters GuIbO PASTORINO 169 Observations on epithelial mucocytes in the sole of Patella spe- cies and Littorina littorea (Linnaeus, 1758) Mark S. DAVIES 175 Observations on the winter spawning and larval development of the ribbed limpet Lottia digitalis (Rathke, 1833) in the San Juan Islands, Washington, USA ALAN R. HOLYOAK, DONALD J. BROOKS, AND SHAWNA R. COBLENTZ 181 The description of a new species of Favartia (Murexiella) from the South Pacific Ocean BARBARA W. MYERS AND CAROLE M. HERTZ........ 182 High performance thin layer chromatography determination of carbohydrates in the hemolymph and digestive gland of Lymnaea elodes (Gastropoda: Lymnaeidae) DANIEL J. CLINE, BERNARD FRIED, AND JOSEPH SHERMA — 185 On the egg capsules of Epitonium georgettinum (Kiener, 1839) (Gastropoda: Epitoniidae) from Patagonian shallow waters GuIbO PASTORINO AND PABLO PENCHASZADEH 188 Description of a new species of the genus Phidiana Gray, 1850 (Nudibranchia: Facelinidae) from Pacific Ocean waters of Panama FRANCISCO J. GARCIA AND JESUS S. TRONCOSO Number 3 (July 1, 1999) A new species of Doriopsilla (Nudibranchia: Dendrodorididae) from the Pacific Coast of North America, including a com- parison with Doriopsilla albopunctata (Cooper, 1863) TERRENCE M. GOSLINER, MARIA C. SCHAEFER, AND SANDRA V. MILLEN 201 Observations on the embryonic development of Octopus mimus (Mollusca: Cephalopoda) from northern Chile IS WW/AIRINIIS. S.A cpio ane ace a pedi Gla obi csien pieiaintono oie oad 211 Bathymodiolus (Bivalvia: Mytilidae) from hydrothermal vents on the Azores Triple Junction and the Logatchev hydrothermal field Mid-Atlantic Ridge RUDO VON COSEL, THIERRY COMTET, AND ELENA M. KRYLOVA 218 Shell form and color variability in Alia carinata (Neogastropoda: Columbellidae) JEFF W. TUPIN 249 Remains of the prey—recognizing the midden piles of Octopus dofleini (Wiilker) R. DODGE AND D. SCHEEL 260 Morphometric species recognition in Brachidontes darwinianus and Brachidontes solisianus (Bivalvia: Mytilidae) MARCEL OKAMOTO TANAKA AND CLAUDIA ALVES DE INTAGATERIAE Saricerart es acai ulsme onic ietmmar sen? fe,ye, se) ature 267 Early development of Fissurella picta (Gmelin, 1791) M. L. GONZALEZ, M. C. PEREZ, D. A. LOPEZ, J. M. Wintas, AND (Co YN IBINO) 5.0% 6 lb bo woo bobo a bee 275 Occurrence of the Asian clam, Corbicula fluminea (Miller, 1774) Bivalvia: Sphaeriacea: Corbiculidae in Colorado JAMES R. CORDEIRO AND SARAH MACWILLIAMS 278 Cytotaxonomic verification of a non-indigenous marine mussel in the Gulf of Mexico BRENDEN S. HOLLAND, DANIEL S. GALLAGHER, DAVID W. FLICKS VANDI S CODD KE DAVIS! fis.) ces sci Se eh 280 Rediscovery of the introduced non-indigenous bivalve Laternula marilina (Reeve, 1860) (Laternulidae) in the northeastern Pacific Topp W. MILLER, EUGENE V. COAN, AND JOHN W. (SHAPIIAIN IE ts pehetrertans fist ste emoanetre eden sells Js) tetisvss 18) cates 282 Chemoattraction of Lymnaea elodes (Gastropoda: Lymnaeidae) to leaf lettuce and Tetramin JASON T. FEDOK, BERNARD FRIED, AND ADITYA REDDY 284 Lindeman Lake, British Columbia, type locality of Zonitoides randolphi Pilsbry ROIs Ue (GION AES Ga ebeneedaunaecasmadu ons 286 A new species of Gastrocopta (Gastropoda: Pulmonata: Pupilli- dae) from the Deep River Formation, late Oligocene or early Miocene, Montana BARRY ROTH Number 4 (October 1, 1999) Another look at the muricine genus A/ftiliosa EMM VOKE Sica orrciaceine vesteaveis ot dre: =) Shee ye) ea eeamerecs 289 A systematic review of the hydrobiid snails (Gastropoda: Ris- sooidea) of the Great Basin, western United States. Part II. Genera Colligyrus, Eremopyrgus, Fluminicola, Pristinicola, and Tryonia ROBERT HERSHLER il Land caenogastropods of Mounts Mahermana, [lapiry, and Va- siha, southeastern Madagascar, with conservation statuses of 17 species of Boucardicus KENNETH C. EMBERTON AND TIMOTHY A. PEARCE 338 AUTHOR INDEX PANU IDS UNG NOSUNT (Cag bin ood gun Go oo dau nb oan oe 267 AA QUILERA; Ge. ce boats sa heer ee aad aotearoa ae ee 101 AANGELONI) os = G fsc, cts ee acre he eon to Toes deere ter ete real 1 PNY ONO BE ter RPGR Eo) een Be Ree ORO ELE Ge CAE Uaey Earn Geer 124 BARRYs. Ji Pee erase Roce hee cto Sano eases 117 BENHAM: SiR dices, casos altiche teu ee neni: Suniel eeewras 112 BRADBURY Jy ns cecus uel cheleeapcn th oeee Seo etne reaeaieos raneee ee iaiacins 1 BROOKS: Di Ux sccecetasie oe See hist, Apsara) Mm 181 C@ACERES= MARTINEZ obo tutose; se issirtaten Stet orieureney ceseed aval emcee wears 10 GHATINE'S A soe 5 sursins ce uses RO EE SOLS Ne he 1 GHAPMANS 2) Wise civccsine ice Sea OR ES 282 GESINES DJ. acta Se Sees oN ae eee BSE BAe a ee 185 GOAN SERVE | Bice eyelet eaten ge verte Une eae ott ae 132,282 GOBLEENTI Zs) S SERS ga curso Rae ot os nade euespeaey ees 181 GOMTET iy 58 Races ae seca HS ears wae ee coaches eee RT 218 CORDEIROS. J Resa sesso tik house re hence ates eR ee oe Mea ie 278 BYASVIES NDS ecesece asses tose one eerie aa 175 DVASVIS 55S seo Sct pee steed oR ois rad ee Oe ce Ee ras 280 DODGE RE ocx cese cena aca ater cee ope ee Rn oe ore ee 260 EMBER TON Gos (Cesar re cries eater ea eee 338 BEDOK As De > Sorceress eee ee eee aim ay Lae Re oes ecto Sa 284 HORS VADs Rs Gy ota pire aS Ce cic ere as mee ere 286 RIED SSB eae odore. ae case ee OR RT eS 185,284 GATIAGHER MD! OSs, e) en ee eee Cee ae 280 GARCIA, Gs a sgecnre fo eoce N Oe ihay Ae se Cae meet aes 190 GEIGER SD ATs etek etesosy tak rai Merten ees cece aah Saree intranets 85 GITTENBERGERS: Eis Pies Suess Ree) cuca te iene eed eta one 67 GOEDER TAN Ee geo a ary one Re nee ee 112 GONZALEZ 8 VIN es eae cere eee eee aE e Cr eam 275 GOSTHNERS ls VIS Encuentra 201 UATE sl Wise Us, ie louie hee ses sie ene eu eek Ee Ca ee een ea aa 97 ETERSHIEER SS Renny ari citer tee tee nee oe eI eR a ee 306 FIER TZ AC MIS oy Ris atte crn ein eae Rec ae Pea ne ne es 182 FIICKS 1D Weed fcc cise ben Se eee SE eee Rea 280 FLODIEAND SB ea tac ude eer ine eee ere eRe aT inter errs es 280 FIOLYOAK AG IRS Seuss austetey eden ie ear oe ogy 181 ELUNE se) Cee eee sees ca antes ett iae ee UC ae an er REN 152 1X0 RaS Gites oe ene ee i oat NR a fo teen Ane Nth Aen he aa 157 JARAMILLO AES) eon ear eto Porat SP entices, coe aA RE gee eee 72 IKOGHEVAR: IRE EG hie eo sett cde aes oracle aca MARE Un een 117 KRYROVAS (EI Miee ai. cen Setanta: cee ee 218 TSAMPREDE 2 SIL. Gate. oases eI Boe os LE Re 97 JEINDBERG; DD aIRe © area ass Sens one ls) Coa ee (194) LOPEZ Di Aci aul wait: sect ee ee Legace ee 275 MAG WiIEEIAMSS: Sa ieesesiee ence aie) ein ieee 278 MeEDONALD (Gs wa ah ee eee acne 62 MENDOZA:. Rio) econ Sco ae ee ee ee 101 MUILPENS Ss. Vie scot ene elite ae cee 201 Miabiers a Wee cy fa a Sea Ss ee 282 MONTEMAYORS Ro cyecrcoe cess tee cbt as a ee 101 MYERS: Bio We. io 2sccnecaisie iach sieas. cnn eee 182 NYBAKKENS Jinn Sader cg. eens aicss es bute Le OR eee en 62 O'BEIRN GER OXE Es acsgecsl ciety ase tlee Glue, Seen cue ORE eee 17 PASTORINOs¢Gi ei) occa iste eae eee 169,188 PEARCE ISAS © 0. aigis ky epee sues fest) SiR 338 PENGHASZADEHS Ps ass.aos a ost el Dae 188 PEREZ (MGs 5 eo wih end aa Aes ea ee ee DUS) PINO! 5 Ag tesa shase se coe ea keels Ge ee 275 QUUONS, Boda Gea setae cacke aie yale a EO eee 72 RAMOS) MivAs oo. 2Sscs aelencs 6 see 124 REDDYjAs., fete Shae S53 eb oes eee 284 RED; Dy iG). 0 irs 5 as 2 eee ee 21 RODRIQUEZ; (Gio Ses Re eee 101 ROTH Bis) Socras © ee eee 286,(198),(199) SATOS “Ss ei ccsl eos oie Sse os hela eae 54 SCHAEFER; MEG. eee bh 8 ae es cee) eee 201 SCHEEL;. [De Jess 2 es oe Ge eee 260 SHERMA\. Ji) iosuete aoaie gaits fia oy eee eee eke 185 STEWART; KevAs 53. 2608 2 ss is See eee 85 TANAKA Mi ©5555 bec, RA aay Gentes oe ae en 267 MTRONCOSO: JE Soe so oe Ree See 190 TURING Wi 2 SE SO 249 WRIBESS MS Roe vane ae nae crise) een oe) ene 275 WASQUEZ=YEOMANS, Rois be cle ee cece ee 10 VOKES BAe ee cer Seas) 289 VON COSEL, Rete Sas ora bn pe nhs RU eee 218 WiABKERS Rye eo 05 Scans Gidea, aude 6 one eee 17 WIARNKES Ker fiat aid ine aoe Onn sie ae 211 WESSEDINGH® | Fa IPL 7) ines sacle) eee eee 67 Page numbers for book reviews are indicated by parentheses. iV On ae ae ISSN 0042-3211. Vay a Motil : VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 42 January 4, 1999 Number 1 CONTENTS Growth, seasonality, and dispersion of a population of Aplysia vaccaria Winkler, 1955 IISAWANGELONI, JACK BRADBURY, AND ALEXIS (CHAINE .....-.0000+s.s00s+s05 1 Metazoan parasites and pearls in coexisting mussel species: Mytilus californianus, Mytilus galloprovincialis, and Septifer bifurcatus, from an exposed rocky shore in Baja California, northwestern Mexico en CAGCERES-IMIARTINEZ AND: Re WASQUEZ-YEOMANS 2.0... 5. 08ace6-eseceeneecs 10 Pea crab, Pinnotheres ostreum Say, 1817, in the eastern oyster, Crassostrea virginica (Gmelin, 1791): prevalence and apparent adverse effects on oyster gonad development BRAN CIS Xa: ©; BEIRN AND RANDAL) I WALKER Seo. ltl ls et ict cee cone oot 17, The genus Littoraria Griffith & Pidgeon, 1834 (Gastropoda: Littorinidae) in the tropical Eastern Pacific DAN AI CABINET Meeps ara were ter ian resiseeie meer ce sre ics Sucre otcer eevee aie opine alae eel an tia waiioneiets 21 Genetic and environmental control of growth and reproduction of Phacosoma ja- ponicum (Bivalvia: Veneridae) SHUNT WORM, SVGUO)" <5 serch eo Gin aeteece Seti mC na Pore ore a ra ar 54 A worldwide review of the food of nudibranch mollusks. Part II. The suborder Dendronotacea GarY MCDONALD AND JAMES NYBAKKEN CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly in January, April, July, and October by the California Malacozoological Society, Inc., % Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. Periodicals postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to The Veliger, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. THE VELIGER Scope of the journal The Veliger is an international, peer-reviewed scientific quarterly published by the Cali- fornia Malacozoological Society, a non-profit educational organization. The Veliger is open to original papers pertaining to any problem connected with mollusks. Manuscripts are considered on the understanding that their contents have not appeared, or will not appear, elsewhere in substantially the same or abbreviated form. Holotypes of new species must be deposited in a recognized public museum, with catalogue numbers provided. Even for non- taxonomic papers, placement of voucher specimens in a museum is strongly encouraged and may be required. Very short papers, generally not over 750 words, will be published in a “Notes, Infor- mation & News’ column; in this column will also appear notices of meetings and other items of interest to our members and subscribers. Editor-in-Chief Barry Roth, 745 Cole Street, San Francisco, CA 94117, USA e-mail: veliger@ucmp1.berkeley.edu Production Editor Leslie Roth, San Francisco Board of Directors Michael G. Kellogg, City and County of San Francisco (President) Hans Bertsch, National University, San Diego Henry W. Chaney, Santa Barbara Museum of Natural History Eugene V. Coan, California Academy of Sciences, San Francisco Terrence M. Gosliner, California Academy of Sciences, San Francisco Carole S. Hickman, University of California, Berkeley FE. G. Hochberg, Santa Barbara Museum of Natural History Matthew J. James, Sonoma State University David R. Lindberg, University of California, Berkeley James Nybakken, Moss Landing Marine Laboratories David W. Phillips, Davis Peter U. Rodda, California Academy of Sciences, San Francisco Barry Roth, San Francisco Geerat J. Vermeij, University of California, Davis Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (not institutions) interested in any aspect of malacology. New members join the society by sub- scribing to The Veliger. Rates for Volume 42 are US $40.00 for affiliate members in North America (USA, Canada, and Mexico) and US $72.00 for libraries and other institutions. Rates to members outside of North America are US $50.00 and US $82.00 for libraries and other institutions. All rates include postage, by air to addresses outside of North America. Memberships and subscriptions are by Volume only and follow the calendar year, starting January 1. Payment should be made in advance, in US Dollars, using checks drawn from US banks or by international postal order. No credit cards are accepted. Payment should be made to The Veliger or “CMS, Inc.” and not the Santa Barbara Museum of Natural History. Single copies of an issue are US $25.00, postage included. A limited number of back issues are available. Send all business correspondence, including subscription orders, membership applications, pay- ments, and changes of address, to: The Veliger, Dr. Henry Chaney, Secretary, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105, USA. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. Barry Roth, Editor, 745 Cole Street, San Francisco, CA 94117, USA. © This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). The Veliger 42(1):1—9 (January 4, 1999) THE VELIGER © CMS, Inc., 1999 Growth, Seasonality, and Dispersion of a Population of Aplysia vaccaria Winkler, 1955 LISA ANGELONTI, JACK BRADBURY anp ALEXIS CHAINE Department of Biology, 0116, University of California, San Diego, 9500 Gilman Drive, La Jolla, California 92093-0116, USA Abstract. The growth and spatial dispersion of an intertidal population of the California black sea hare, Aplysia vaccaria Winkler, 1955, was studied from October 1995 to October 1996 in North Cardiff Beach, California. Population size peaked in November and then declined to zero the following year, while mean weight peaked in June. Breeding was observed throughout the year. The sea hares were spatially clustered and the aggregation pattern was invariant over time. Individual movements and growth were recorded by tagging 19 animals with internal microchips. Tagged animals grew at a rate of 4.9 g/day and moved an average minimum distance of 2.3 m/day. INTRODUCTION Intensive research on the neurobiology of sea hares (Kan- del, 1979) has been complemented by field studies of their behavior and ecology (e.g., Carefoot, 1967; Usuki, 1970; Kupferman & Carew, 1974; Audesirk, 1979; Nish- iwaki et al., 1975; Susswein et al., 1983; Susswein et al., 1984; Carefoot, 1989; Pennings, 199la, b; Strenth & Blankenship, 1991; Yusa, 1996). One of the least studied of the sea hare species is Aplysia vaccaria Winkler, 1955 (Carefoot, 1987). This may be partly due to its perception as “‘a secretive animal ... much more difficult to obtain in numbers” than other sea hares (Winkler, 1957). A. vac- caria ranges from California to Baja California (Lance, 1967). The species is reported to be primarily nocturnal (Eales, 1960; Pennings, 1991b) and often immobile (Pen- nings, 1991b), inhabits rocky coasts and kelp beds (Kan- del, 1979), spawns in February and March under rocks in shallow water (Winkler, 1955), and feeds upon Egregia spp. (Winkler, 1955; Winkler & Dawson, 1963). There are no published data on the growth rate of A. vaccaria, its seasonal abundance or its movements in the field (Care- foot, 1987). In this study, we report on a dense intertidal population of A. vaccaria which we were able to monitor regularly for 1 year. Data collected included population size, indi- vidual body masses, and fraction of animals mating. The animals in this population appeared to be clustered into small aggregations. Aplysia aggregations have often been described (Kupferman & Carew, 1974; Achituv & Sus- swein, 1985; Pennings, 1991b), but rarely quantitatively, and their function is still unknown. Sea hares may aggre- gate primarily for mating purposes or other social func- tions (Susswein et al., 1984; Carefoot, 1987; Pennings, 1991b). On the other hand, aggregations may be caused by differential larval settlement on preferred habitats, at- traction of adults to patches of food, or attraction of adults to sites with preferred levels of exposure and tidal action (Pennings, 1991la). Because we had the opportunity to map every animal within a fixed study area on each cen- sus, we were able to monitor a number of measures of spatial dispersion to see how these changed with season, mean body size, mating frequency, and density of ani- mals. Microchip tagging, a method new to sea hare bi- ology, successfully provided data on growth and move- ment for a small number of individuals. The result is the first study of dispersion, growth, and survival on this spe- cies. MATERIALS anp METHODS Site and Study Period The study population was monitored from October 1995 to October 1996. The site is an intertidal rocky reef at North Cardiff Beach, San Diego County, California (33°1'N, 117°17'W). At the beginning of our study, there was little sand, much exposed bedrock, and extensive cobbling of the upper strand. A year later much of the beach was covered with sand, and most of the tidepools formerly occupied by A. vaccaria were covered. We se- lected a 15 X 19.85 m rectangular census site with deeply eroded channels and pools which harbored high densities of A. vaccaria. The tidal range of this site spans from 25 cm above mean low tide level to 90 cm below mean low tide level. The included channels remained filled with wa- ter during the lowest tides (—58 cm), and there were many rock ledges under which the sea hares aggregated. Algae in the study site included Ulva californica, Plo- camium cartilagineum, Laurencia sinicola, Ceramium sp., Pterocladia capillacea, Gelidium purpurascens, Ac- rosorium venulosum, Jania crassa, Herposiphonia sp., Centroceras clavulatum, Hypnea valentiae, Zonaria far- Page 2 lowii, Dictyopteris undulata, Sphacelaria sp., Colpomen- ia sinuosa, Egregia menziesii, and Macrocystis pyrifera drift. The animals in this area were not isolated from hu- man disturbance, although we usually arrived before the low tide and secured cooperation from onlookers in min- imizing disturbance to the study site. Sampling and Mapping Methods On average, we sampled the site every 2—3 weeks dur- ing low tides, which occurred sometime between 4 a.m. and 6 p.m. Each A. vaccaria was mapped by recording the distance and compass angle measurement from one corner of the study site to the animal; these were later converted to cartesian coordinates relative to the sides of the study rectangle. Each animal was weighed after re- moving any debris and as much water as possible from its body. Errors in wet mass measures were estimated by returning five individuals to the water, letting them move about for 5 minutes, and then reweighing them three times. The repeatability (a measure of correlation be- tween repeated measures) of mass measurements was very high (r = .998, SD = 7 g; Falconer, 1989). Mating status of closely opposed animals was determined by in- serting a finger under the parapodia to determine whether or not an everted penis joined individuals. Density Measures We used two different measures of density. “‘Absolute density”’ is the number of sea hares in the study area divided by the total area in the plot. We also computed an “effective density’’ by dividing the total number of sea hares counted on a census by the minimum convex polygon required to surround them all. This second den- sity measure thus reflects both the number of animals present and their dispersion. Dispersion Analysis The study area was partitioned into 81 contiguous quadrats. Twenty-three of these were considered uninhab- itable because of lack of sufficient tidepool area and none ever hosted an animal. All but one of the remaining quad- rats did harbor A. vaccaria at one time or another during the study. In order to determine whether the animals were significantly aggregated on each census, counts in the habitable 58 quadrats were compared to random (Poisson) expectations with a chi-square goodness-of-fit test. Where dispersions were significantly non-random, pat- tern was characterized using several measures of intensity and grain (Pielou, 1969). Intensity measures the differ- ence in sea hare density between cluster peaks and spaces between clusters; grain measures the typical distance be- tween cluster centers and the typical area occupied by a cluster. Lloyd’s index of patchiness was an intensity mea- sure computed from quadrat counts. This value indicates The Veliger, Vol. 42, No. 1 the average number of animals found in the same quadrat with a focal animal after correcting for differences in overall densities. A second measure of intensity was com- puted by assigning all animals within 1 m of a neighbor to a ‘cluster’? and averaging the resulting number of an- imals per cluster. Grain was measured in several ways. The first method was to impose a 10 X 10 cell grid on the study site and construct correlograms to characterize levels of autocor- relation between numbers of animals/cell at varying cell separations. These plots all showed initial positive auto- correlation (as measured by Moran’s I) which dropped to zero and then oscillated around the zero line with increas- ing cell separations. The farthest separation with a sig- nificant positive I (after a Bonferroni correction) and that at which I first crossed the zero line were both noted. The two values are rough estimates of average minimal and maximal cluster size (Upton & Fingleton, 1985). A sec- ond measure of grain relied on the number of clusters generated by the | m proximity rule: the larger the num- ber of clusters per unit area, the finer the grain. The clus- tering algorithm also drew minimum convex polygons around each cluster, identified the geometrical centers of the polygons, and computed the enclosed areas. The dis- persion of the cluster centers was examined using nearest neighbor methods. The areas of the polygons were used as additional measures of grain: larger mean cluster areas implies coarser pattern grain. Cluster areas could be larg- er because of more animals per cluster, larger distances between nearest neighbors, or both. To tease apart these effects, we measured average nearest neighbor distances for each census. Intensity and grain are both likely to vary with popu- lation density. We plotted a measure of intensity (the log- arithm of cluster size) against a measure of grain (the logarithm of the number of clusters) for successive cen- suses. Points have to move as overall densities change: which variable shifts least over time can be used as an indicator of the possible mechanisms governing disper- sion. Because densities steadily decreased from the fourth census on, we confirmed impressions from the grain vs. intensity plot by regressing the logarithms of animal den- sity, cluster size, and cluster number on time, and then comparing the slopes of the three regressions using AN- COVA. Finally, we examined the regularity with which differ- ent areas in the study site were used by ranking habitable quadrats according to the fraction of the total animals they hosted on each census, and comparing the consis- tency of quadrat ranks over time using Kendall’s index of concordance. Statistics were undertaken on Macintosh computers us- ing the commercially available Statview and JMP pack- ages. The analyses of intensity and grain were largely undertaken using our own dispersion program called An- L. Angeloni et al., 1999 telope (available on the Internet at http://www-biolo- gy.ucsd.edu/research/vehrenbury/programs.html). Tagging Methods Nineteen sea hares were tagged (12 in March and sev- en in May) using number-coded Trovan passive transpon- der tags, which were later detected and read with a Tro- van LID-500 Hand Held Reader held close to the body of the sea hare. Both the tags and reader were obtained from InfoPet Identification Systems, Inc. Transponder tags, weighing only 0.01% of the weight of a typical in- dividual, were injected under the mantle just inside of the left parapodium. This tagging method was selected for several reasons: we found it to be less likely to attract the attention of curious onlookers than external tags, reducing the human disturbance to the study; tagged animals ap- peared to be healthy and unaffected by the procedure, and continued to increase in mass, as did the rest of the pop- ulation; the reader was easy to use and detected micro- chips quickly, even when wrapped in a plastic bag for protection against moisture. The 14 tagged animals that were recaptured on sub- sequent censuses were mapped and weighed. Because our data are limited to those animals that stayed within the study site, and assume a straight line of travel between the two points on subsequent days, our estimates of in- dividual movements are highly conservative. Because we could sample the study area exhaustively, tagged animals not found on one census, but found later, must have em- igrated outside the site and then returned. For the census when they were not detected, we recorded the minimum distance between last capture site and the edge of the study area. This is again a conservative estimate of move- ment over that period given that we routinely searched the immediate area around the study site for tagged ani- mals; hence any sea hares moving out and back into the site must have gone even farther than the value recorded. Mean movement/day was calculated using data from cen- suses on 5 consecutive days in March (Days 156-160 of the study). Estimates of short-term movements were obtained by observing 16 animals during four low-tide periods in April (Days 183, 185-187). Each animal was followed for 2 hours between 10 a.m. and 3 p.m., and its location was mapped on a diagram of the study site every 15 min. RESULTS Density, Growth and Mating Patterns The number of sea hares in the study site peaked in November (Day 42) at 310 animals and an absolute den- sity of 1.04 individuals/m’; the corresponding effective density was 1.63 individuals/m*. The population then steadily decreased to zero by October of the following year (Figure 1). Page 3 a) Absolute density OND DF VS to on S On 1 300 9 = 8 250 i = v $ 200 : ee 150 & is] 5 100 3 ° 4 Be 3 2 50 Z, 0 0 0 50 100 150 200 250 300 350 400 Sample day b) Effective density 1.8 1.6 Number/m?2-MCP = hb A ph ® © 0 50 100 150 200 250 300 350 400 Sample day Figure 1 Density of A. vaccaria in the study area vs. sample day. a. Ab- solute density calculated by dividing the number of individuals by the total area of the study site. b. Effective density calculated by dividing the number of individuals by the minimum convex polygon around them (calculated in Antelope). Months are in- dicated at the top of the figure. Mean body mass increased in a roughly linear fashion from 372 g in October to the peak of 1105 g in June (Day 239); this corresponds to an average increase of 3.1 g/ day (Figure 2). June was also the only time that animals with weights below 180 g were observed; however, these were few in number. The smallest individual found was 30 g. The decrease in mean mass after the June peak was not due to further recruitment of small individuals, but instead to a rapid drop in maximum body size. Because minimal body sizes concurrently increased, this was a pe- riod in which the range of body sizes in the site was dramatically reduced. Body size histories for 13 tagged animals are summa- rized in Figure 3. All but two samples were taken before the mass peak in June. Although the general trend for the tagged animals is an increase in mass, we observed both rises and falls over the short term. Whether these reflect egg-laying bouts, food shortages, or other constraints on The Veliger, Vol. 42, No. 1 0 50 100 150 200 250 300 350 400 Sample day Figure 2 Mean mass of the population vs. sample day (solid line). Bars represent 1.96 standard errors of the mean. Dashed lines indicate maximum and minimum masses for each census. Months are indicated at the top of the figure. feeding is unknown. Note that some tagged individuals show synchronous increases and decreases in mass, whereas others show quite asynchronous patterns. The average increase in mass for tagged animals was 4.9 g/ day over the time period they were followed; this can be compared to a 3.2 g/day average increase in the unmarked population over the same time period. There was a great deal of variance in growth rate (SD = 8.7 g, SE = 2.3 g) with some tagged animals even losing weight over the period they were monitored. Growth rate of tagged ani- mals was unrelated to their initial weights (r* = 0.02, df = 13, P > 0.5). The lower growth rate for the population as a whole during this period when compared to the tagged animals is at least partly due to the appearance of small individuals in the population in June. In every census, some fraction of the population, be- tween 3% and 43%, was found mating. Both time of day (morning vs. afternoon; Figure 4a) and tide height (Figure 4b) were found to have a significant effect on the fraction of total individuals mating (analysis of covariance on transformed data, r? = 0.51, df = 23, P = 0.002). The results indicate that more individuals were mating in the morning censuses (P = 0.007) and when the low tide was relatively higher (P = 0.008). There was also a significant interaction effect between tide height and time of day on the fraction mating (P = 0.004). However, it is impossible to separate the effects of time of day and season, as most of the morning censuses occurred in the spring and sum- mer and most of the afternoon censuses occurred in the fall and winter. Although the number of egg masses was not quantified, newly laid eggs were observed in the study site throughout the entire year’s sampling. Tatten Mass (g) rae #1 #10 #5 #9 Talat 150 170 190 210 230 250 Sample day Figure 3 Body size histories for 13 tagged animals which were recaptured and reweighed. Months are indicated at the top of the figure. Dispersion Even after uninhabitable quadrats were removed from the analyses, animals were found to be significantly clus- tered in space on every census (all x? > 23, minimal df = 3, and all P < 0.0001). Correlograms showed strong positive autocorrelation of animal densities over an av- erage range of 3.4 m, and a drop to zero correlation for quadrats separated by an average 5.3 m (see example in Figure 5). Mean cluster size within a census (using a 1 m linkage rule) ranged from 4.1—12.1 animals/cluster when all individuals were considered, and from 6.2—23.0 when only clusters with more than two individuals were tallied. Mean numbers of clusters in the site ranged from 14—34 including singletons and pairs, and from 8—20 when only clusters larger than two were considered. Clus- L. Angeloni et al., 1999 a) ot) a rr = as 0 50 100 150 200 250 300 350 400 Sample day b) 45 40 35 ap 30 3S 25 is] & 20 J eae5 10 5 0 -G0NN-50) 6-40)-30) -20) 10° 0 Tide height (cm) Figure 4 a. Percent of total individuals which were mating in each census. Squares represent afternoon censuses, circles represent morning censuses. Months are indicated at the top of the figure. b. Percent of total individuals which were mating vs. tide height in each census. ter size and cluster number were uncorrelated (r = —0.159, t = 0.534, P > 0.5). Population density equals the product of mean cluster size and cluster density. Thus, variation in population density or population size (given the fixed area of our study site) can be completely explained by the indepen- dent variations in cluster size and number; whichever of these has the larger variation will dominate variations in density. For our samples, the coefficient of variation in cluster size was 41.4%, whereas it was only 26.6% for cluster number. This suggests that most of the variation in density was due to changes in cluster size. This is confirmed in Figure 6a, which summarizes how mean cluster size and number each varied as population size decreased over time. The relative stability of cluster num- ber when compared to cluster size is demonstrated statis- tically in Figure 6b. Here, the logarithms of density, clus- ter size, and cluster number are regressed against sample date over the period of population decline. The regres- Page 5 a) 15 Gg ke] a > Ua a X axis (m) b) rm ¢€ © {e) = 3 6 9 12 15 18 21 Distance (m) Figure 5 a. Dispersion and size of clusters using 30 cm linkage rule (dark stipple) and 1 m linkage rule (light stipple) for census on 19 December 1995 (Day 69). A total of 270 animals were recorded on this census. b. Correlogram based on a 10 X 10 grid for above sample. Dark circle at distance of 3.75 m corresponds to Moran’s I of 0.241 (P = 0.00275). This is just slightly greater than the P = 0.0025 required by a Bonferroni correction given an overall significance level of 0.05 and 20 tests. sions show a rate of drop in number of clusters which is significantly slower than that for cluster size or overall density, but statistically similar rates for drops in cluster size and population density. This again suggests that den- sity decreases were accomplished as reductions in num- bers of animals/cluster, not in the number of clusters. This linkage between density and cluster size is also indicated by a plot of Lloyd’s index of patchiness vs. sample day (Figure 6c). There is no significant trend here indicating that once variations in density have been taken into ac- count (a fundamental focus of this index), the intensity of the spatial pattern is invariant over time. The area of clusters as measured by the | m clustering rule is negatively correlated with sample day (1? = 0.735, P = 0.0002). Since it is also positively correlated with mean cluster size (In(cluster area) = 2.1 In(cluster size) — 5.3; r° = 0.871, P = 0.0001), the decrease in cluster area could simply reflect the demonstrated drop in mean cluster sizes over the season. However, the 2.1 coefficient a) 24.5 20.1 N 16.4 a7) 3’ 13.5 S bb mai 6 vu = 9.0}. 7.4 6 é as 6 74 9.0 11.0 13.55 16.4 20.1 24.5 Number of groups b) Oo 3 i] > Vv = s £ = 0 50 100 150 200 250 Sample day co) ONE D) dF SIMin Al Milde JineAvsShaOusN = Pe Wn Vv = cae 5 2 6 Sy le ° 5 sas} 5 & a2) NS 0 & = 5 0 50 100 150 200 250 300 350 400 Sample day Figure 6 a. Mean cluster size vs. number of clusters for successive cen- suses after population peak. Clustering uses 1 m rule and only shows groups greater than two individuals. Note logarithmic axes which cause isopleths of equal density to plot as straight lines with a slope of negative one (dotted lines with selected densities indicated). As density decreases, points must move closer to low- The Veliger, Vol. 42, No. 1 in the log-log regression implies that cluster area depends upon the square of the number of animals in a cluster. Were each animal to require the same amount of space around it, and animals settled in clusters with efficient packing, cluster area should depend only on the first pow- er of cluster size. A likely explanation is that the area added to a cluster per animal is not a constant. In fact, a regression of area/cluster member vs. sample date shows a highly significant decrease over the season (r* = 0.684, P = 0.0005). This could arise either because individuals crowd more closely together later in the season, or be- cause they do not pack into clusters efficiently. Mean nearest neighbor distances range from 14-25 cm, but show no seasonal effects (r7 = 0.035, P = 0.542). Thus, the answer is not variation in individual spacing. Because the animals tend to aggregate around the margins of large boulders, their within-cluster dispersion is often curvilin- ear. This could easily increase the area of enclosing poly- gons at rates faster than were animals to pack in a con- tiguous fashion. If the 58 habitable quadrats are ranked according to the fraction of animals they harbor on each census, there is a high degree of repeatability in quadrat rank over the season (Kendall’s index of concordance, x? = 249, df = 56, P < 0.0001). In fact, the same 17.5% of the quadrats host an average 48% (95% CL = 40-56%) of the animals on any census, and of these, the top 9% harbor an average 24% (CL = 18-30%) of the population. Individual Movements Minimum values for the cumulative distances traveled since first capture date are shown in Figure 7a. A con- servative estimate of the minimum mean distance trav- eled per day by 11 tagged animals on 5 subsequent days is 2.3 m/day (n = 42, SE = 0.3). Eleven of the 42 dis- e er left corner of graph. Diagonal from upper right to lower left indicates trajectory points would follow were decreases in den- sity accommodated by equivalent decreases in cluster size and cluster number. The fact that most points are below this diagonal indicates that drops in density are largely borne by drops in clus- ter size; cluster number remained relatively stable over the study period. b. Rates of seasonal decrease in overall density of animals on the study plot (open circles and dashed line), numbers of clusters (filled circles and dark solid line), and mean cluster sizes (squares and thin solid line). All measures normalized by divid- ing by maximum value for season and transformed using loga- rithms. Results of ANCOVA indicate a significant overall effect of sample day (F,5, = 44.1, P = 0.0001) and measure (F; 55 = 7.8, P = 0.0001). Post hoc tests using both Fisher’s LSD and Scheffe tests indicate significant differences in slopes of density vs. number of clusters (P = 0.014 and P = 0.046 respectively), and cluster size vs. number of clusters (P = 0.0006 and 0.0024), but not between density and cluster size (P = 0.222 and 0.478). c. Lloyd’s index of patchiness over time. Bars represent 1.96 standard errors of the mean. L. Angeloni et al., 1999 2 54 o7 Cumulative distance (m) hE = eS! ie) 2% NI b oO 14 1 a th YY ss 3) Gs} 8 Time since initial tagging (days) b — 20.0 12.2 7.4 4.5 Greatest distance (m) 2.7 1 4 10 21 39 65 Sample interval (days) Figure 7 a. Conservative estimates for cumulative distance traveled by each of the tagged animals over the time that they were recap- tured. Regression equation is In(cumulative distance) = 0.28 + .70 In(time) (1? = 0.57, df = 69, P < 0.0001). b. Distance be- tween two most distal capture locations vs. time interval between capture events for each of the tagged animals. Regression equa- tion is In(max distance) = 0.53 + 0.58 time®? (r? = 0.43, df = 14, P = 0.0078). tance measurements are based on estimates of the mini- mum distance traveled by animals leaving or returning to the study site, and two values are missing because the animals left the study site, but did not return on the con- secutive days. Figure 7b shows the greatest distance be- tween two recapture sites for each tagged animal as a function of the time interval between the corresponding recaptures. Maximum distances between recapture sites range from 3—12 m and increase significantly with the interval between recaptures. Direct observations of indi- vidual movements during low-tide periods indicate that A. vaccaria move an average of 0.92 m/hour (SD = 1.4, SE = 0.24), and that movement is restricted when the tide is especially low (ANOVA, mean for —3 cm and —6 cm tides = 1.5 m/hr, mean for —12 cm tides = 0.5 m/hr, p = 0.04). During these observation periods, sea hares were seen grazing on Ulva, smaller red and brown algae Page 7 on the sides of rock ledges, and drifting pieces of Ma- crocystis and Egregia trapped in deep tidal pools. DISCUSSION Most sea hares are thought to have maximum life cycles of 1 year (Miller, 1960; Carefoot, 1967; Audesirk, 1979; Carefoot, 1987; Strenth & Blankenship, 1991). While we were unable to follow tagged individuals for their full lifetime, the temporally changing weight distribution for the population does not contradict the possibility of a year-long life cycle for A. vaccaria. If we assume that the mean weight of the population was increasing at the same rate before this study as during the increasing por- tion of this study, the estimated recruitment time of this population would have been June—July, 1995. This, in combination with the presence of small individuals in June 1996, indicates a late spring or summer recruitment time for A. vaccaria. There is evidence of some overlap- ping of generations, as small animals were present with the largest individuals in June. The scarcity of small an- imals and absence of juveniles smaller than 30 g may reflect a low recruitment rate for the year of this study, or may indicate that juveniles recruit to other locations or habitat types. The increase in amount of sand within the study site was not quantified, but may have contrib- uted to the decline of the population. Very few dead an- imals were found during the last censuses, and these were quickly washed offshore. It is important to note that be- cause this study population was not a closed one, the measured changes in density and mass cannot be entirely attributed to the seasonal patterns of settlement, growth, and death, but could also be caused by migration into and out of the site. A. vaccaria is described as the largest gastropod in the world with record sizes of 14 kg and 99 cm (Behrens, 1991). To attain such large body sizes in 1 or 2 years would require a rapid growth rate. While both population averages and tagged animals showed rapid growth, indi- viduals in this study did not approach these record sizes. There are several possible explanations for the large dif- ference in body sizes between record animals and those in this study: there may be greater variance in lifespan than what is seen in most sea hares, such that record animals live longer than | or 2 years; the study site may be a marginal or lower quality habitat for A. vaccaria; while there was sufficient food in the site to attract and support growth for record densities of these animals, it may not have been the amount or quality required for record growth rates in such a dense population. Our ob- servations show that the A. vaccaria diet is considerably broader than suggested by Winkler & Dawson (1963). The data on individual movements support the claim that the activity of intertidal sea hares is often restricted during extremely low tides due to exposure to the air (Kupferman & Carew, 1974; Carefoot, 1987). Our finding Page 8 that members of this population are less likely to mate during the lowest low tides may be explained by the fact that many animals were partially exposed. The resulting reduction in movement is likely to reduce encounter rates with potential mates, and dessication may make it phys- ically difficult for sea hares to mate. The significantly higher levels of mating in the morning censuses of April, May, and June might be explained by time of day, season, or both. If A. vaccaria are indeed nocturnal (Eales, 1960; Pennings, 1991b), they may initiate mating during the night and then continue on into the morning. It is also possible that they mate more in the spring, allowing for high levels of recruitment in the summer. In any case, it is clear that A. vaccaria are not limited to reproducing in February and March (Winkler, 1955), but spawn year- round. The dispersion data show that, like other sea hares, A. vaccaria is characterized by dense aggregations, and that as densities vary seasonally, the number and spacing of clusters is strongly conserved. This could arise because the animals are willing to travel a limited distance to join a cluster, and thus the spacing of clusters depends only on the area of the site and this typical range, or it could arise because there are favored locations in which clusters might form. The consistency in location of clusters sup- ports the latter possibility. Pennings (1991b) noted that aggregations of A. californica often appeared in the same locations as previous aggregations, indicating a prefer- ence for certain sites, either because those sites are more environmentally suitable or because they were previously occupied, leaving olfactory cues as a basis for subsequent aggregations. Our data indicate that this site-fidelity is also true for A. vaccaria. Microchip tagging of individuals showed that they moved across an average of 6 m of the study site during the 2 weeks between censuses (as indicated by the great- est distances between recapture locations), and some moved completely across or even out of the 15 x 19.85 m study site. The minimum average daily movements of 2 m were themselves as great as the typical distances between clusters (about 1-3 m). Average hourly move- ments of 0.92 m also allowed for a great deal of move- ment between clusters, even during low tides. Tagged in- dividuals were found in different clusters in subsequent recaptures, and the sizes of clusters varied above and be- yond global density changes. All of these results suggest that the dispersion patterns are not simply the conse- quence of initially patchy recruitment of larvae, but rather that these very mobile animals are actively aggregating. Given normal movements, each animal thus has a choice of many groups that it could join. An earlier study (Winkler, 1955) suggested that repro- duction in A. vaccaria was limited to a few winter months. The fact that the animals aggregate year-round could then have been construed as evidence against clustering as a mat- ing strategy. Our data show clearly that clustering and The Veliger, Vol. 42, No. 1 breeding are both maintained year-round, normal ranging allows access to multiple clusters, and most clusters contain mating animals. It thus remains possible that clustering in A. vaccaria is related to mating strategies as has been sug- gested for other members of the genus (Audesirk, 1979; Carefoot, 1987; Pennings, 1991b). ACKNOWLEDGMENTS We thank the many people who helped in the field: Ian Billick, Dahlia Chazan, Marc Dantzker, Karen Fear, Paul Griffin, Darren Irwin, Laura Molles, Chris Nagy, Helen Neville, Jonathan Richmond, Amy Ritter, and Sukamol Srikwan. Susan Williams identified algae, and she, Jim Lance, Trevor Price, and Scott Rumsey provided tactical support and advice at critical junctures. LITERATURE CITED AcuitTuv, Y. & A. J. SUSSWEIN. 1985. Habitat selection by two Mediterranean species of Aplysia: A. fasciata Poiret and A. depilans Gmelin (Mollusca: Opisthobranchia). Journal of Experimental Marine Biology and Ecology 85:113—122. AUDESIRK, T. E. 1979. A field study of growth and reproduction in Aplysia californica. The Biological Bulletin 157:407—421. BEHRENS, D. W. 1991. Pacific Coast Nudibranchs. Sea Chal- lengers: Monterey. 107 pp. CAREFOOT, T. H. 1967. Studies on a sublittoral population of Aplysia punctata. Journal of the Marine Biological Associ- ation of the United Kingdom 47:335-—350. CAREFOOT, T. H. 1987. Aplysia: its biology and ecology. Annual Review of Oceanography and Marine Biology 25:167—284. CaREFOOT, T. H. 1989. A comparison of time/energy budgeting in two species of tropical sea hares Aplysia. Journal of Ex- perimental Marine Biology and Ecology 131:267—282. EALEs, N. B. 1960. Revision of the world species of Aplysia (Gastropoda, Opisthobranchia). Bulletin of the British Mu- seum (Natural History), Zoology 5:269—404. FALCONER, D. S. 1989. Introduction to Quantitative Genetics. Longman Scientific & Technical: Essex. 438 pp. KANDEL, E. R. 1979. Behavioral Biology of Aplysia: A Contri- bution to the Comparative Study of Opisthobranch Molluscs. W. H. Freeman and Company: San Francisco. 463 pp. KUPFERMAN, I. & T. J. CAREW. 1974. Behavior patterns of Aply- sia californica in its natural environment. Behavioral Biol- ogy 12:317-337. LANcE, J. R. 1967. Northern and southern range extensions of Aplysia vaccaria. The Veliger 9:412. MILLER, M. C. 1960. A note on the life history of Aplysia punc- tata Cuvier in Manx waters. Proceedings of the Malacolog- ical Society of London 34:165—167. NISHIWAKI, S., H. UEDA & T. MAKIOoKA. 1975. Tagging studies on the growth of the sea hare Aplysia kurodai on an inter- tidal rocky shore. Marine Biology 32:389-395. PENNINGS, S. C. 1991a. Spatial and temporal variation in re- cruitment of Aplysia californica Cooper: patterns, mecha- nisms and consequences. Journal of Experimental Marine Biology and Ecology 146:253-274. PENNINGS, S. C. 1991b. Reproductive behavior of Aplysia cali- fornica Cooper: diel patterns, sexual roles and mating ag- gregations. Journal of Experimental Marine Biology and Ecology 149:249—266. L. Angeloni et al., 1999 PreLou, E. C. 1969. An Introduction to Mathematical Ecology. Wiley Interscience: New York. 286 pp. STRENTH, N. E. & J. E. BLANKENSHIP. 1991. Reproductive pat- terns and seasonal occurrence of the sea hare Aplysia bras- iliana Rang (Gastropoda, Opisthobranchia) at South Padre Island, Texas. American Malacological Bulletin 9:85—88. SusswEIN, A. J., S. Gev, E. FELDMAN & S. MARKOVICH. 1983. Activity patterns and time budgeting of Aplysia fasciata un- der field and laboratory conditions. Behavioral and Neural Biology 39:203-—220. SusswEIN, A. J., S. Gev, Y. AcHITUv & S. MARKOVICH. 1984. Behavioral patterns of Aplysia fasciata along the Mediter- ranean coast of Isreal. Behavioral and Neural Biology 41:7— 22 Upton, G. & B. FINGLETON. 1985. Spatial Data Analysis by Example. John Wiley & Sons: New York. 410 pp. Page 9 Usuk1, I. 1970. Studies on the life history of Aplysiae and their allies in the Sado district of the Japan Sea. Science Reports of the Niigata University, Series D (Biology) 7:91—105. WINKLER, L. R. 1955. A new species of Aplysia on the southern California coast. Bulletin of the Southern California Acad- emy of Sciences 54:5—7. WINKLER, L. R. 1957. The Biology of California Sea Hares of the Genus Aplysia. University of Southern California. 201 Pp. WINKLER, L. R. & E. Y. DAWSON. 1963. Observations and ex- periments on the food habits of California sea hares of the genus Aplysia. Pacific Science 17:102—105. Yusa, Y. 1996. The effects of body size in mating features in a field population of the hermaphroditic sea hare Aplysia ku- rodai Baba, 1937 (Gastropoda, Opisthobranchia). Journal of Molluscan Studies 62:381—386. THE VELIGER © CMS, Inc., 1999 The Veliger 42(1):10—-16 (January 4, 1999) Metazoan Parasites and Pearls in Coexisting Mussel Species: Mytilus californianus, Mytilus galloprovincialis, and Septifer bifurcatus, from an Exposed Rocky Shore in Baja California, Northwestern Mexico J. CACERES-MARTINEZ* AND R. VASQUEZ- YEOMANS Centro de Investigaci6n Cientifica y de Educacioén Superior de Ensenada, Departamento de Acuicultura, Km. 107 Carret. Tijuana-Ensenada Apdo. Postal 2732, 2800, Ensenada, Baja California, México Abstract. Metazoan parasites, Modiolicola gracilis Wilson (Copepoda), Pseudomyicola spinosus Rafaelle & Monti- celli (Copepoda), Urastoma cyprinae Graff (Platyhelminthes), unidentified encysted trematode cercaria, and pearls were found in coexisting mussel species: Mytilus californianus Conrad, Mytilus galloprovincialis Lamarck, and Septifer bi- furcatus (Conrad) from the upper intertidal zone of an exposed rocky shore in Baja California, northwestern Mexico. Incidence of parasites and pearls was greater in M. californianus, the largest species of mussels examined, than in the other mytilids. The lowest incidence of parasites and pearls was in S. bifurcatus, the smallest species of mussels ex- amined. The highest parasite prevalence coincided with the autumn and winter months when M. californianus and M. galloprovincialis had a lower condition index and were reproductively active. All parasites produced histological alter- ations in their hosts; a hemocytic reaction and compression of tissues were commonly observed. In spite of M. califor- nianus being the most parasitized species, it is the dominant component in the upper intertidal zone. These results suggest that factors specific to the infesting metazoan parasites in M. galloprovincialis and S. bifurcatus reduce their competi- tiveness capacity against M. californianus. INTRODUCTION The exposed rocky intertidal component of Pacific North- west shorelines is dominated by structurally complex beds of the intertidal mussel Mytilus californianus Con- rad, 1837 (Suchanek, 1992). In mussel beds in California, M. californianus may coexist with Mytilus galloprovin- cialis Lamark, 1819 (after the works of Harger [1972a, b] in Santa Barbara, California, the dominant Mytilus ed- ulis-like species in southern California were indentified as M. galloprovincialis [see McDonald & Koehn, 1988; Koehn, 1991]), and Septifer bifurcatus (Conrad, 1837) (Harger, 1972a, b; Haderlie & Abbott, 1980). On the ex- posed rocky shores of Baja California, northwestern Mex- ico, the three species also coexist. Mytilus californianus is dominant in the middle intertidal zone where it forms a relatively uniform carpet. However, in the upper inter- tidal zone, M. californianus coexists with aggregations of S. bifurcatus and M. galloprovincialis. The latter may be found in clumps or may be scattered singly in tidepools. There are a variety of studies on distribution, competition, and ecology in mixed populations of M. galloprovincialis and M. californianus on the Pacific coast of California * Corresponding author: Centro de Investigacién Cientifica y de Educaci6n Superior de Ensenada, Departamento de Acuicul- tura, Km. 107 Carret. Tijuana-Ensenada Apdo. Postal 2732, 2800, Ensenada, Baja California, México. Telephone: (617) 45050 ext. 24303. fax: (617) 64225. e-mail: jcaceres@cicese.mx (Harger 1968, 1970, 1972a, 1972b; Petraits, 1978; Su- chanek, 1978; Witman & Suchanek, 1984). However, studies on the relationships among the coexisting mytil- ids, Mytilus spp. and Septifer spp. (Haas, 1942; Hoshiai, 1964) and on parasites of coexisting Mytilus spp. (Cous- tau et al., 1990) are scarce. No studies exist on the interaction of coexisting mytil- ids, M. californianus, M. galloprovincialis, and S. bifur- catus and their parasitic load. However, that interaction may be important from an ecological and epizootiological point of view. Suchanek (1992) showed the extreme bio- diversity associated with Mytilus californianus beds in southern California. Over 300 species of plants and ani- mals live within the layers and spaces of these mussel beds, one of the most diverse temperate communities de- scribed. Several of these species have been considered as ecto-symbionts (Laihonen & Furman, 1986). There is no information about the diversity of metazoan parasites af- fecting coexisting mytilids at the same time period in spite of the fact that Mytilus species may be infested by more than 50 kinds of organisms (Lauckner, 1983). Pearl formation in mussels has been associated with the pres- ence of trematode cysts and has been considered as an abnormality (Lutz, 1980; Lauckner, 1983). Consequently, their presence may be considered as a part of the parasitic load in mussels. The aims of the present work were to determine the diversity of metazoan parasites and pearls associated with the soft body of M. californianus, M. gal- J. Caceres-Martinez & R. Vasquez-Yeomans, 1999 Page 11 loprovincialis, and Septifer bifurcatus that coexist in the upper intertidal zone of an exposed rocky shore in Baja California northwestern Mexico, to corroborate histologic damage, and to discuss relationships among metazoan parasites and pearls and their hosts. MATERIALS AnD METHODS From August 1995 to July 1996, 30 adults of the blue mussel Mytilus galloprovincialis (mean total shell length 44.20 mm, SD = 7.25), the Californian mussel Mytilus californianus (mean total shell length 63.61 mm, SD = 4.23), and the branch-ribbed mussel Septifer bifurcatus (mean total shell length 34.65 mm, SD = 3.11 mm) were collected each month from the upper intertidal exposed rocky shore of La Mina del Fraile (31°19’N, 116°26’W) Baja California, Mexico, where the three mussel species coexist. After removal of any fouling organisms, each mussel was measured (total shell length) and weighed (total weight), and then placed in a Petri dish and opened. In- tervalvar water and mussel flesh were examined for the presence of parasites under a dissecting microscope. Par- asites and cysts were picked up with a dissecting tweezers from the gills and mantle, and all pearls were removed from the space between the mantle and the inner shell. Turbellarians were preserved in Steinmann’s fluid (1 part concentrated nitric acid, | part saturated solution of mer- curic chloride in 5% sodium chloride, 1 part distilled wa- ter) (Sluys, 1989). Copepods were preserved in 70% eth- anol. Pearls were dried and stored in glass vials. Cysts were excised with a bistoury under a dissecting micro- scope to extract metacercaria and were preserved in 70% ethanol. The following works were used for metazoan identification: Graff 1913; Rafaelle & Monticelli, 1885; Ho, 1980; Lauckner, 1983; Do et al., 1984; Do & Kaji- hara, 1986. Total weight (TW), wet meat weight (MW), and shell weight (SW) of mussels were recorded to obtain a con- dition index where CI = [MW/(TW — SW)] xX 100 (Aguirre, 1979). Parasite prevalence was estimated as the number of infested mussels/number of mussels examined x 100. Sixty mussels were used for histopathological evalua- tion; in this case, parasites were not picked out. The soft body of these mussels was removed from the shell and fixed whole in Davison’s fixative (Shaw & Battle, 1957) for at least 24 hr. An anterior transverse section including digestive gland, mantle, and gills was taken. Tissue sam- ples were embedded in paraffin wax and were sectioned at intervals of 5 um; histological sections were stained with hematoxylin and eosin (Shaw & Battle, 1957). Tis- sue analysis and measurements were made with a micro- meter eyepiece placed in an optical microscope (Olympus BH-2). A Kruskal-Wallis test was used to compare the preva- lence of parasites among mussel species, and one way ANOVA followed by a mean comparison test SNK was used to compare the CI and prevalence of parasites in different mussel species. RESULTS The same species of metazoan parasites were found in Mytilus californianus, Mytilus galloprovincialis, and Sep- tifer bifurcatus. They were identified as: Pseudomyicola spinosus (Raffaele & Monticelli, 1885) (Copepoda, My- colidae) (specimens deposited in the United States Na- tional Museum, Smithsonian Institution, USNM 274221; Modiolicola gracilis (Wilson, 1935) (Copepoda, Clausi- diidae), (USNM 274222; Urastoma cyprinae (Graff, 1913) Platyhelminthes, Urastomidae) (specimens depos- ited in the Zoological Museum Amsterdam, V.PI. 899 and Vpl. 900) (detailed morphological features of these spe- cies have been shown by Caceres-Martinez et al., 1996a, b; Caceres-Martinez & Vasquez-Yeomans, in press); and unidentified trematode encysted metacercaria, probably genus Himasthla (see Lauckner, 1983). This is the first record of these parasitic copepods and turbellarians in S. bifurcatus. Both copepod species were found crawling on gills and mantle; the turbellarians were observed among gill filaments and crawling on the gills where some de- coloration was found. Encysted metacercaria were ob- served in the base of the gills and labial palps. Pearls of different sizes were found between the mantle and inter- nal surface of the shell in the three mussel species stud- ied. Histopathological analysis revealed that both copepod species may be found inside the digestive tract. Identifi- cation was possible due to the characteristic body shape of both copepod species. The prosome of P. spinosus is more slender than the prosome of M. gracilis (see figures in Caceres-Martinez & Vasquez- Yeomans (1997) for the former, and Do & Kajihara (1986) for the latter). Al- though Mytilicola orientalis has been found in mussels from the Pacific coast of North America (Lauckner, 1983), its modified body is easily distinguishable from the copepods here reported and we did not find evidence of this species in the present study. There was no damage observed in the gills, only compression of gill filaments, but in the gut and stomach, appendages of M. gracilis and P. spinosus caused erosion of the epithelium and a hemocytic reaction (Figure 1). Some copepods were found within the connective tissue of the digestive gland where a strong hemocytic reaction (encapsulation) was observed (Figure 2). Urastoma cyprinae produced a com- pression, erosion, and rupture of gill filaments and a he- mocytic reaction (Figures 3, 4). Encysted metacercaria may or may not produce a hemocytic reaction from the host. They were observed in gills, labial palps, and con- nective tissue of the digestive gland and mantle (Figures 5, 6). Page 12 The Veliger, Vol. 42, No. 1 Explanation of Figures | to 6 Figure 1. Modiolicola gracilis in the gut of Mytilus californianus. Asterisk shows the copepod body, ap = copepod appendages. Thin arrows show the eroded epithelium of the gut and the wide arrow shows the completely destroyed epithelium and hemocytic reaction of the host in this area. Copepod may obstruct the gut. Scale bar = 50 pm. Figure 2, Granulocytoma in Mytilus californianus. Hemocytes are engulfing the copepod found within in the con- nective tissue of the digestive gland. Large arrow shows the granulocytoma and the small arrows show appendages of the encapsulated copepod. Scale bar = 50 wm. Figure 3. Urastoma cyprinae in the gills of Mytilus galloprov- incialis. Presence of the worm results in accumulation of hemocytes in gill filaments around the worm (large arrows), gill filament compression (thin arrow at the top), and small clusters of hemocytes around the turbellarian; these hemocytes appear to have ruptured from the gill filaments (thin arrow). Scale bar = 50 wm. Figure 4. Urastoma cyprinae in the gills of Mytilus californianus. As with Mytilus galloprovincialis, the presence of the turbellarian is characterized by swelling of the gill filaments. See hemocyte accumulation in gill filaments around the worm (big J. Caceres-Martinez & R. Vasquez-Yeomans, 1999 Page 13 Table 1 Minimum and maximum number of parasites (range) per mussel species observed from August 1995 to July 1996. Septifer bifurcatus Modioli- Modioli- Urastoma cola _Trematode Urastoma — cola Month cyprinae_ gracilis cysts Pearls cyprinae — gracilis A 0-0 0-0 0-0 0-0 0-2 0-0 S 0-1 O-1 0-0 0-0 0-2 0-2 O 0-1 0-1 0-0 0-0 O-1 0-4 N 0-0 0-2 0-0 0-0 0-3 0-1 D 0-1 0-2 0-0 0-0 0-8 0-5 J 0-0 0-1 0-0 0-0 0-20 0-1 F 0-0 0-2 0-0 0-0 0-18 0-5 M 0-0 0-1 0-0 0-0 0-4 0-2 A 0-0 0-0 0-1 0-1 0-42 0-0 M 0-0 0-0 0-0 0-0 0-3 0-3 J 0-0 0-0 0-0 0-0 0-1 0-0 J 0-0 0-0 0-0 0-0 0-0 0-0 Mytilus galloprovincialis Mytilus californianus Modioli- Trematode Urastoma cola Trematode cysts Pearls cyprinae — gracilis cysts Pearls 0-0 0-0 O-11 0-0 0-0 0-3 0-0 0-1 0-9 O-1 0-0 0-4 0-1 0-3 O-7 0-15 0-1 0-4 0-0 0-0 0-6 0-8 0-0 0-4 0-0 0-5 0-38 0-10 0-0 0-1 0-0 0-0 0-34 0-7 O-1 0-7 0-1 0-0 0-23 0-9 0-1 0-3 0-0 0-0 0-38 0-5 0-0 0-0 0-0 0-0 0-95 0-2 0-0 0-1 0-7 0-2 0-2 0-0 0-0 0-1 0-0 0-3 0-5 0-0 0-0 0-0 0-0 0-0 0-2 0-0 0-2 0-0 Table 1 shows the range of parasites and pearls in mus- sels, and Figure 7 shows their prevalence. Urastoma cy- prinae was the most abundant parasite in M. californianus and M. galloprovincialis, its number ranged from 0 to 95 in the former and from 0 to 42 in the latter. Their maxi- mum prevalence occurred during autumn-winter, and prevalence was similar in M. californianus and M. gal- loprovincialis but different in S. bifurcatus (Kruskal-Wal- lis Test, H = 2.25, P < 0.001, followed by SNK method, q = 2.5, ns). In S. bifurcatus the number of U. cyprinae ranged from O to 1 and prevalence was minimum and occurred in September, October, and December 1995. Modiolicola gracilis was also abundant, its number ranged from 0 to 15 in M. californianus and from 0 to 5 in M. galloprovincialis; its prevalence was similar in both hosts (Kruskal-Wallis Test, H = 4.77, P = 0.09). This copepod was found ranging from 0 to 2 individuals in S. bifurcatus, and it was only observed during autumn-win- ter. Two Pseudomyicola spinosus were found in M. cali- fornianus; only one was observed in M. galloprovincialis and one in S. bifurcatus during autumn. Encysted meta- cercaria were found scattered during the time of the study and in low numbers; their numbers ranged from 0 to 2 in M. californianus, from 0 to 7 in M. galloprovincialis, and Oto 1 in S. bifurcatus. Of the three mytilid species, pearls were found most frequently in M. californianus and ranged from 0 to 7; in M. galloprovincialis, they ranged from 0 to 5, and from 0 to | in S. bifurcatus. Larger pearls were found in M. californianus where the largest was 2.76 mm in diameter. In Mytilus galloprovincialis and S. bifurcatus, the largest pearls were 0.92 mm and 0.48 mm, respectively. In general, Mytilus californianus was the most parasit- ized species when comparing all parasitic loads and pearls in the three mussel species studied (Kruskal-Wallis Test, H = 1.98, P < 0.0001). The condition index of mussels studied is shown in Figure 7. One way ANOVA (F = 4.1, P = 0.026) and Student-Newman-Keuls Method showed that CI was dif- ferent between Mytilus californianus and Mytilus gallo- provincialis, and also between M. californianus and Sep- tifer bifurcatus. However, CI was similar between S. bi- furcatus and M. galloprovincialis (q = 1.34, ns). Lowest condition indices in M. californianus and M. galloprov- incialis were recorded from autumn to winter when the highest parasite prevalence also occurred and mussels were reproductively active (unpublished data). This pat- tern was not observed in S. bifurcatus. DISCUSSION All species from a community are in close relationship, and among them, parasites play an important role in the health status of community members. Living parasite arrow at the right of the worm), the rupture of the gill filaments (arrow at the top), and the clusters of hemocytes around the worm (thin arrow). Scale bar = 50 wm. Figure 5. Encysted metacercaria in the labial palps of Myzilus californianus. The cyst wall is wide and dark. There is no histopathological evidence of host reaction against the cyst. Scale bar = 50 wm. Figure 6. Trematode cyst in degradation by hemocytes (small arrows) in the Mytilus galloprovincialis connective tissue of the digestive gland. Note that the wall of the cyst is broken and clearly distinguishable from the digestive diverticula at the left of the picture. Scale bar = 50 ym. Page 14 & Urastoma cyprinae A Septifer bifurcatus [) Modiolicola gracilis 100 = 90 Encysted metacercaria 80 EJ Pearls 70 60 50 Nee a a 40 AGES OF NGeDa aks Mir Are Maes B Mytilus galloprovincialis Parasites prevalences (%) Month Figure 7 Prevalence of metazoan parasites and pearls in (A) Septifer bi- furcatus, (B) Mytilus galloprovincialis, and (C) Mytilus califor- nianus from La Mina del Fraile, an exposed rocky shore in Baja California, Mexico. Lines show the condition index (CI) of my- tilids studied. strategies may be specific for the host species or have multiple species as host (lack of host specificity). This is the case in the present study; all metazoan parasites ob- served infected all three coexisting mytilids. This sug- gests that observed parasites have at least three possible hosts. However, parasite prevalence and range for host The Veliger, Vol. 42, No. 1 species were different. In the first instance, the size of the host seems to be an important factor; the largest mus- sel, Mytilus californianus, was the most infested mytilid, while Septifer bifurcatus, the smallest species, was the least infested mussel. In Todos Santos Bay, Ensenada, northwestern Mexico, Caceres-Martinez et al. (1996) found a significant positive correlation between the size of Mytilus galloprovincialis and the presence of the co- pepod Pseudomyicola spinosus; mussels from 45 to 75 mm shell length had the highest number of copepods. In the Black Sea, Murina & Solochenko (1991) found a re- lationship between the number of the parasitic turbellar- ian Urastoma cyprinae and the size of M. galloprovin- cialis. Mussels from 50 to 70 mm were the most parasit- ized, while mussels under 30 mm had no parasites. This kind of relationship has also been found for the parasitic copepod Modiolicola insignis Aurvillius, 1882, in the Mediterranean Sea by Costanzo & Calafiore (1987). They pointed out that smaller mussels (under 33 mm shell length) were more likely to escape infestation. This ob- served relationship may be favorable to S. bifurcatus, in which the largest size was under 40 mm. Specific studies are needed to determine the reasons for this differential infestation. The condition index of Mytilus californianus and My- tilus galloprovincialis was related with their parasitic load and reproductive season during autumn-winter (unpub- lished data). There was lower CI when highest parasitic loads were recorded and mussels were reproductively ac- tive. Histopathological evidence showed that all parasites studied may produce tissue damage. Similar damage in tissues caused by Urastoma cyprinae and Pseudomyicola spinosus has been described in mussels and oysters (Din- amani & Gordon, 1974; Robledo et al., 1994). The in- crease in the number of parasites suggests a probable in- crement in the extension of their associated damages, which in turn could affect the host condition independent of its reproductive stage. High number of parasites related to a low condition index has been shown in a variety of studies (Cole & Savage, 1951; Dare, 1981; Theisen, 1987; Coustau et al., 1990; Murina & Solochenko, 1991; Caceres-Martinez et al., 1996). The condition index of S. bifurcatus, the less parasitized species, was not related to parasitic load, supporting the fact that a low number of parasites may not reflect changes in the CI. However, comparative studies on infected and uninfected mussels at the same reproductive stage must be carried out to determine effect of parasite load on condition of mussels. In accordance with Stunkard & Uzmann (1958) and Lauckner (1983), the trematode metacercaria cyst recov- ered from marine bivalves has frequently been misiden- tified and assigned to various species, mainly of the genus Himathla. Encysted metacercaria invade a wide range of bivalve species like Mytilus edulis Linnaeus, Cardium ed- ule Linnaeus, Mya arenaria Linnaeus, and Macomabal- tica Linnaeus (Lutz, 1980; Lauckner, 1983). Tissue dam- J. Caceres-Martinez & R. Vasquez-Yeomans, 1999 age observed was similar to that described by Bower (1992). Recorded prevalences are low compared to other studies on encysted metacercaria and mussels (Lutz, 1980). Presence of encysted metacercaria may be con- nected to the presence of pearls recorded in this study because pearl formation may be induced by encysted trematode infestation (Lutz, 1980; Lauckner, 1983; Bow- er, 1992). From a pathological point of view, encysted metacercaria resulted in damage to the host, and from a commercial point of view, the presence of pearls in edible mussels (Mytilus californianus and Mytilus galloprovin- cialis) may affect consumer acceptance (Lutz, 1980). If the parasitic load were to affect competitivenes of the host, it is surprising that the most parasitized mytilid species, M. californianus, is dominant. This suggests that other factors affect the competitive capacity of M. gal- loprovincialis and S. bifurcatus. ACKNOWLEDGMENTS The authors thank Dr. M. A. del Rio Portilla for English review and comments on the manuscript; K. Castaneda, P. Macias, and S. Curiel for their help in sample process- ing; and N. Flores and F Valenzuela for their help during field sampling. We thank O. Pedrin, O. Tapia, and J. Cas- tro from Centro Regional de Investigaci6n Pesquera de Ensenada, and O. Sosa and N. Lopez from the Ecology Department of CICESE for their help in histological pro- cessing and providing us with equipment and installa- tions. Thanks to L. Morales from the CICESE library for providing us with the bibliography. This work was sup- ported by CICESE project #623106. LITERATURE CITED AGuIRRE, M. P. 1979. Biologia del mejillon (Mytilus edulis) de cultivo de la Ria de Vigo. Boletin del Instituto Espanol de Oceanografia 5(3):107—160. Bower, S. M. 1992. Diseases and parasites of mussels. Pp. 543— 559 in E. Gosling (ed.), The Mussel Mytilus: Ecology, Phys- iology, Genetics and Culture. Elsevier: London. CACERES-MARTINEZ, J., R. VWASQUEZ-YEOMANS & R. SLUuYS. 1996a. Urastoma cyprinae in natural and cultured mussel (Mytilus galloprovincialis Lmk.) populations in México. Bulletin of European Association of Fish Pathologists 16(6): 1-3. CACERES-MARTINEZ, J., R. VASQUEZ-YEOMANS & M. E. SUAREZ. 1996b. Two parasitic copepods Pseudomyicola spinosus and Modiolicola gracilis associated with edible mussels Mytilus galloprovincialis and Mytilus californianus from Baja Cal- ifornia, NW México. Journal of Shellfish Research 15(3): 667-672. CACERES-MARTINEZ, J. & R. VASQUEZ-YEOMANS. In press. Pres- ence and histopathological effects of the copepod Pseudo- myicola spinosus in Mytilus galloprovincialis and Mytilus californianus. Journal of Invertebrate Pathology. ConrabD, T. A. 1837. Description of new marine shells from upper California, collected by Thomas Nuttall, Esq. Journal of the Academy of Natural Sciences Philadelphia 7:227— 268. Page 15 CosTANnzo, G. & N. CALAFIORE. 1987. Seasonal fluctuation of Modiolicola insignis Aurvillius, 1882 (Copepoda: Poecilos- tomatoida: Sabelliphilidae), associated with Mytilus gallo- provincialis in Lake Faro (Messina). Journal of Crustacean Biology 7(1):77—86. Cote, H. A. & R. E. SAvace. 1951. The effect of the parasitic copepod, Mytilicola intestinalis (Steuer) upon the condition of mussels. Parasitology 41:156-161. CoustTau, C., C. ComBes, C. MAILLARD, KF RENAUD & B. DELAY. 1990. Prosorhynchus squamatus (trematode) parasitosis in the Mytilus edulis-Mytilus galloprovincialis complex: spec- ificity and host-parasite relationships. Pp. 291—298 in O. Per- kins & C. Cheng (eds.). Pathology in Marine Science. Ac- ademic Press: New York. Dare, P. J. 1981. The susceptibility of seed oysters of Ostrea edulis L. and Crassostrea gigas Thunberg to natural infes- tation by the copepod Mytilicola intestinalis Steuer. Aqua- culture 26:201—211. DINAMANI, P. & D. B. GORDON. 1974. On the habits and nature of association of the copepod Pseudomyicola spinosus with the rock oyster Crassostrea glomerata in New Zeland. Jour- nal of Invertebrate Pathology 24:305—310. Do, T. T., T. KASIHARA & J.-S. Ho. 1984. The life history of Pseudomyicola spinosus (Rafaele and Monticelli, 1985) from the blue mussel, Mytilus edulis galloprovincialis La- marck in Tokyo Bay, Japan, with notes on the production of atypical male. Bulletin Ocean Research Institute University of Tokyo 17:1—65. Do, T. T. & T. KAsIHARA. 1986. Studies on parasitic copepod fauna and biology of Pseudomyicola spinosus, associated with blue mussel, Mytilus edulis galloprovincialis. Bulletin Ocean Research Institute University of Tokyo. 23:1—63. GrarF, L. von. 1913. Platyhelminthes. Turbellaria. II. Rabdo- coelida. In F E. Schulze (ed.), Das Tierreich. Eine Zusam- menstellung und Kennzeichnung der rezenten Tierformen. 35. Lieferung. Friedlander: Berlin. Haas, FE 1942. The habits of life of some west coast bivalves. The Nautilus 56(1):30-33. HADERLIE, E. C. & D. P. ABBotr. 1980. Bivalvia: the clams and allies. Pp. 355—411 in R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stan- ford University Press: Stanford, California. HarGeER, J. R. E. 1968. The role of behavioral traits in influ- encing the distribution of two species of sea mussel: Mytilus edulis and Mytilus californianus. The Veliger 11(1):45—49. Harcer, J. R. E. 1970. Comparisons among growth character- istics of two species of sea mussel, Mytilus edulis and My- tilus californianus. The Veliger 13(1):44—56. HarGER, J. R. E. 1972a. Competitive co-existence: maintenance of interacting associations of the sea mussels Mytilus edulis and Mytilus californianus. The Veliger 14:387—410. HarGER, J. R. E. 1972b. Variation and relative “‘niche”’ size in the sea mussel Mytilus edulis in association with Mytilus californianus. The Veliger 14(3):275—283. Ho, J.-S. 1980. Origin and dispersal of Mytilus edulis in Japan deduced from its present status of copepod parasitism. Pub- lication Seto Marine Biology Laboratory, 25(5/6):293—313. Hosuial, T. 1964. Synecological study on intertidal communities V. The interrelation between Septifer vigratus and Mytilus edulis. Bulletin of the Marine Biological Station of Asa- mushi 12(1):37—41. Koeun, R. K. 1991. The genetics and taxonomy of species in the genus Mytilus. Aquaculture 94:125—146. LAIHONEN, P. & E. R. FURMAN. 1986. The site of settlement Page 16 indicates commensalism between blue mussel and its epi- biont. Oecologia 71:38—40. LAMARCK, J. B. PR. A. DE 1819. Historie naturelle des animaux sans vertébres. Vol. 6. A.S.B. Verdiere Librarie: Paris. LAUCKNER, G. 1983. Diseases of Mollusca: Bivalvia. Pp. 477— 879 in O. Kinne (ed.), Diseases of Marine Animals, Vol. II. Introduction, Bivalvia to Scaphopoda. Biologische Anstalt Helgoland: Hamburg. Lutz, R. A. 1980. Pearl incidence: mussel culture and harvest implications. Pp. 193-222 in R. A. Lutz (ed.), Mussel Cul- ture and Harvest: A North American Perspective. Elsevier: Amsterdam. McDona_p. J. H. & R. K. KOEHN. 1988. The mussels Mytilus galloprovincialis and Mytilus trossulus on the Pacific coast of North America. Marine Biology 99:111—118. Mvurina, G.-V. & A. I. SOLOCHENKO. 1991. Commensals of My- tilus galloprovincialis in the Black Sea: Urastoma cyprinae (Turbellaria) and Polydora ciliata (Polychaeta). Hydrobiol- ogia 227:385-387. PetraltTs, P. S. 1978. Distributional patterns in juvenile Mytilus edulis and Mytilus californianus. The Veliger 21:288—292. RAFAELLE, FE & E S. MONTICELLI. 1885. Descrizione di un nuovo Lichomolgus parasita del Mytilus galloprovincialis Lmk. Atti della R. Accademia Nazionale dei Lincei, (4), Memorie della Classe di Scienza Fisiche, Matematiche e Naturali 1: 302-307. The Veliger, Vol. 42, No. 1 Roscepbo, J. A. F, J. CACERES-MARTINEZ & A. FIGUERAS. 1994. The parasitic turbellarian Urastoma cyprinae (Platyhelmin- thes: Urastomidae) from blue mussel Myzilus galloprovin- cialis in Spain: occurrence and pathology. Diseases of Aquatic Organisms 18:203-210. SHAW, B. L. & H. I. BATTLE. 1957. The gross microscopic anat- omy of the digestive tract of the oyster Crassostrea virginica (Gmelin). Canadian Journal of Zoology 35:325—346. SLuys, R. 1989. A Monograph of the Marine Triclads. A. A. Balkema: Rotterdam. 459 pp. STUNKARD, H. W. & J. R. UZMANN. 1958. Studies on digenetic trematodes of the genera Gymnophallus and Parvatrema. Bi- ological Bulletin Marine Biology Laboratory. Woods Hole 116:184—-193. SUCHANEK, T. H. 1978. The ecology of Mytilus edulis L. in ex- posed rocky intertidal communities. Journal of Experimental Marine Biology and Ecology 31:105—120. SUCHANEK, T. H. 1992. Extreme biodiversity in the marine en- vironment: mussel bed communities of Mytilus californi- anus. Northwest Environmental Journal 8(1):150—152. THEISEN, B. E 1987. Mytilicola intestinalis Steuer and the con- dition of its host Mytilus edulis L. Ophelia 27(2):77-86. WILSON, C. B. 1935. Parasitic copepods from the Pacific coast. American Midland Naturalist 16:776—797. WITMAN, J. D. & T. H. SUCHANEK. 1984. Mussels in flow: drag and dislodgement by epizoans. Marine Ecology Progress Se- ries 16:259—268. The Veliger 42(1):17—20 (January 4, 1999) THE VELIGER © CMS, Inc., 1999 Pea Crab, Pinnotheres ostreum Say, 1817, in the Eastern Oyster, Crassostrea virginica (Gmelin, 1791): Prevalence and Apparent Adverse Effects on Oyster Gonad Development FRANCIS X. O’BEIRN* VIMS-Eastern Shore Laboratory, PO. Box 350, Wachapreague, Virginia 23480, USA RANDAL L. WALKER University of Georgia, Marine Extension Service, Shellfish Aquaculture Laboratory, 20 Ocean Science Circle, Savannah, Georgia 31411-1011, USA Abstract. Incidence of pea crab, Pinnotheres ostreum Say 1817, infestation in the eastern oyster, Crassostrea vir- ginica (Gmelin, 1791), was recorded and related to oyster gametogenic activity over 18 months. Sampling occurred at two tidal heights (high intertidal HI and low intertidal LI) at two sites (House Creek, HC and Skidaway River, SR) in Wassaw Sound, Georgia. Overall, incidence rates were 3% HC LI, 1% HC HI, 8% SR LI, and 4% SR HI. At both tidal heights at HC, no differences were observed in gonad area between those oysters with and without pea crabs. At SR (where overall incidences were higher), oysters without pea crabs had significantly higher gonad area values than those oysters with pea crabs present. These results suggest that at higher incidences of pea crab infestation, oyster reproductive capabilities could be impacted, and support the claim that the pea crab/oyster relationship is a parasitic one. INTRODUCTION The brachyuran pea crab, Pinnotheres ostreum Say, 1817, has been observed in a number of bivalve species, e.g., Mytilus edulis Linnaeus, 1758, Geukensia demissa (Dill- wyn, 1817), Anomia simplex d’ Orbigny, 1842, and Pecten sp. (Williams, 1984). However, it is primarily a parasite (formerly considered a commensal) of the eastern oyster, Crassostrea virginica (Gmelin, 1791). This pea crab is found predominantly in the western Atlantic from Mas- sachusetts, United States, to Santa Catarina in Brazil (Williams, 1984). The prevalence of the pea crab in oys- ters along the eastern seaboard of the United States has generally been high, with prevalences of up to 100% in some subtidal oyster populations in the Chesapeake Bay (Galtsoff, 1964). However, records of pea crab occurrence in the southeastern United States and especially coastal Georgia are scant. Linton (1968) stated that the occur- rence of pea crabs in subtidal oysters in coastal Georgia was 100%. However, the vast majority of Georgia oysters occur intertidally (Harris, 1980). Parks (1968) reported that there were substantially higher proportions of pea crabs in oysters found subtidally than in those found in- tertidally. In the present study, oysters were sampled over a period of 1% years, and the gonads were examined his- * Corresponding author: Telephone: (757)787-5837, fax: (757)787-5831, e-mail: francis@vims.edu tologically. Pea crab presence and absence was recorded in the oysters and these data were then related to the gonad condition of the oysters throughout the sampling period. SITE DESCRIPTION anpD METHODS The two sites chosen for this investigation are shown in Figure 1. House Creek (HC), a shallow sheltered creek, is located on the northern end of Wassaw Sound, Georgia. This site is characterized by relatively high salinities (> 25 %oc) and is sheltered from wave action. The Skidaway River (SR) site, under the Skidaway Institute of Ocean- ography dock on the north end of Skidaway Island, has more variable salinities and is exposed to higher wave action from passing boats than the House Creek site. Two tidal heights were chosen for this study. The low- intertidal (LI) area was that area in and around the mean low water mark. The high-intertidal (HI) area was des- ignated as the area above the region designated by the tidal level at approximately 3 hours after mean low water. In coastal Georgia, the majority of oysters occur between these two intertidal boundaries. Sampling commenced in June 1993 and continued on a biweekly basis until the end of September 1993, when monthly sampling took place. Monthly sampling contin- ued until January 1994. Biweekly sampling recommenced in April 1994 through September 1994. ~ GEORGIAN STUDY 3 AREA Figure 1 Wassaw Sound, Georgia with the two sampling sites indicated: (1) House Creek and (2) Skidaway River near the Skidaway In- stitute of Oceanography (SkIO). At each sampling period, 20 (n = 20) adult oysters were taken from each tidal height at each site. Upon shucking, the tissue was examined and the presence or absence of pea crabs were recorded. A transverse tissue section (5 mm) was dissected from each shucked oyster and was processed for histological examination and qual- itative and quantitative analysis of the gonad tissue ac- cording to the methods outlined in O’Beirn et al. (1996). The quantitative parameter used in this study was gonad area which accounted for that proportion, in a standard viewing area of a histological section of the oyster’s tis- sue, occupied by gonad. Statistical Analysis Single factor repeated measures analysis was carried out on the data whereby all of the independent variables (oyster height, gonad area) were grouped into two cate- gories—pea crabs present or pea crabs absent. Two de- pendent variables were examined in the analysis-of-vari- ance (ANOVA): pea crab presence/absence and sampling periods. No interaction term was determined. The varia- tions from the grand mean due to pea crabs and sampling periods will have been accounted for with remaining de- viations being the source of error. All proportional data was arcsine square-root transformed prior to analysis. An arbitrary value of (a = 0.05) was chosen as the signifi- cance level for each ANOVA. RESULTS The highest recorded proportion of pea crabs in oysters was at the Skidaway River low intertidal site, where 8% The Veliger, Vol. 42, No. 1 Table 1 Percent of oysters, Crassostrea virginica, according to presence or absence of pea crabs, Pinnotheres ostreum. Also given (in parentheses) is the absolute number of oysters in each category. PEA CRAB Present Absent HOUSE CREEK HIGH INTERTIDAL 1% (4) 99% (394) LOW INTERTIDAL 3% (13) 97% (380) SKIDAWAY RIVER HIGH INTERTIDAL 4% (16) 96% (380) LOW INTERTIDAL 8% (33) 92% (364) of oysters sampled throughout the study contained pea crabs (Table 1). The lowest proportion of pea crabs was at the House Creek high intertidal site where 1% of the oysters contained pea crabs (Table 1). Within the sam- pling periods, the highest incidence of pea crabs in oys- ters was found in the Skidaway Low Intertidal oysters in April, 1995 where 21% (4 of 19) of the oysters contained pea crabs. No oysters were found containing more than one pea crab. There were no significant differences in gonad area be- tween those oysters with pea crabs and those without, at both tidal heights at House Creek (HI P = 0.4152 and LI P = 0.8366; Table 2). The high intertidal oysters at Skidaway River had sig- nificantly higher (P = 0.0085) gonad area in oysters with- out pea crabs, than those with pea crabs (Table 2). The low intertidal oysters also had significantly higher gonad area values (P = 0.0117) in oysters where pea crabs were absent than those with pea crabs present (Table 2). Table 2 Percent gonad area of oysters, Crassostrea virginica, ac- cording to presence or absence of pea crabs, Pinnotheres ostreum. Also given are the p-values of repeated mea- sures analysis using ANOVA. PEA CRAB Present Absent p-value HOUSE CREEK HIGH INTERTIDAL 38.4% 55.4% 0.4152 LOW INTERTIDAL 54.5% 56.8% 0.8366 SKIDAWAY RIVER HIGH INTERTIDAL 38.4% 56.9% 0.0085 LOW INTERTIDAL 42.0% 52.5% 0.0117 F. X. O’Beirn & R. L. Walker, 1999 Page 19 DISCUSSION The number of pea crabs found in oysters in our study is substantially lower that those reported previously for oys- ters in coastal Georgia. Not surprisingly, in our study, oysters located near the low-tide mark had higher num- bers of pea crabs than those located higher in the inter- tidal zone. A similar phenomenon was reported by Beach (1969) in North Carolina. However, the maximum pro- portions at any one intertidal height and site of 8% was substantially lower than that of 100% in subtidal oysters as reported previously by Linton (1968). Parks (1968) did record higher instances of pea crabs in subtidal oysters than intertidal oysters. However, the values in Park’s (1968) study were in terms of number of pea crabs ob- tained from a specific number of oysters necessary to give one pint of oyster meat. The number of oysters differed considerably between the sites (tidal heights). Therefore, comparison of Park’s (1968) data to those obtained in this study can only be cursory. The disparity between the re- sults of Linton (1968) and this study can be accounted for by the differences in sampling location (subtidal ver- sus intertidal, respectively). However, given that the ma- jority of oysters in coastal Georgia are located intertidally (Harris, 1980), the proportions reported herein are per- haps more reflective of pea crab incidence in oysters in the region. In Delaware Bay, Flower & McDermott (1952) noted that the proportion of oysters containing pea crabs was higher as they sampled from the upper reaches of the bay toward the ocean, which was concomitant with an in- crease in salinity. Such a pattern was not observed in this study. In fact, it appears that the higher incidences of pea crabs were found at the Skidaway River site, which tra- ditionally has lower salinities (O’ Beirn et al., 1995, 1996; Spruck et al., 1995). The reason for this apparent reversal in prevalence is unclear, but it might be related to the exact location of the House Creek sampling site. All oys- ters were removed from a small sheltered tidal creek, which is subject to high temperature fluctuations on a daily basis. O’Beirn et al. (1995) reported an 8°C water temperature change at this site in the space of 8 hours in 1991. Also, because of the shallow nature of the creek, it is subject to higher salinity fluctuations caused by fresh- water runoff from the marsh, originating from storms which are frequent in the summer months in coastal Geor- gia. Pea crab development is inhibited by salinities less than 15%c (Beach, 1969). Assuming the salinities will drop below 15%c, such factors might inhibit free-swim- ming invasive stages from surviving and hence infesting oysters, at this particular site. A more comprehensive in- vestigation of pea crab incidences along a salinity gra- dient in the Wassaw Sound, Georgia area would need to be carried out to confirm that our findings were not anom- alous. It must be noted that Kruczynski (1974) found no relationship between presence or absence of pea crabs in Mytilus edulis and salinity. The presence of pea crabs within the mantle cavity of bivalves has been determined to have an adverse effect on the host mollusk. Physical damage to the gills, palps, and gonads of the bivalves has been recorded by a variety of authors (Stauber, 1945; McDermott, 1962; Dix, 1973; Jones, 1977). The presence of pea crabs, Pinnotheres ma- culatus Say, 1818, was deemed responsible for adversely impacting filtration and oxygen consumption rates in My- tilus edulis (Bierbaum & Shumway, 1988), as well as having an apparent negative impact on growth rates in nutrient-poor environments (Bierbaum & Ferson, 1986). Tablado & Lopez-Gappa (1995) demonstrated that Myti- lus edulis individuals harboring mature female pea crabs, Tumidotheres (Pinnotheres) maculatus (Say), were sig- nificantly smaller and had lower dry weights than those mussels without pea crabs. Bay scallops, Argopecten ir- radians concentricus (Say, 1822), containing adult female pea crabs tended to weigh less and were smaller than those scallops without pea crabs in Bogue Sound, North Carolina (Kruczynski, 1972). Havert (1958) determined that oysters, Crassostrea virginica, containing pea crabs, Pinnotheres ostreum, had significantly lower dry meat weight and condition indices than oysters without pea crabs. Kruczynski (1972) noted that in the presence of large female pea crabs, the host bivalves tended to have reduced gametogenic output, which was attributed to physical pressure on the gonads. At both sites in our study, oysters with pea crabs pre- sent had lower gonad area values overall than oysters without the pea crabs (Table 2). At the House Creek site, no significant difference in gonad area was determined between those oysters with or without pea crabs. We at- tribute this to insufficient numbers of infested oysters ob- tained from this site. The differences at the Skidaway River site were statistically significant, at both tidal heights. In a parallel study (O’Beirn, unpublished stud- ies), oysters in the high intertidal zone tended to have higher quantitative gametogenic parameters than oysters lower down, suggesting that the high intertidal zone was less stressful to the oyster than previously hypothesized (O’Beirn, unpublished studies). In this study, this appar- ent negative impact of pea crabs on oyster gonad quantity was not confined to these supposedly more stressful en- vironments as was the case with Mytilus edulis infested with Pinnotheres maculatus (Bierbaum & Ferson, 1986). The observation in this study that the presence of pea crabs corresponded with lower gonad area measurements in oysters would question the classification of pea crabs as a commensal of oysters. Haines et al. (1994) proposed that the relationship between female pea crabs and their molluscan hosts be classified as true parasitism, as the female is rarely found free-living outside of the host. The results of the findings herein go further to suggest that the pea crabs negatively impact the fitness of oysters. The Page 20 significance of these results in terms of pea crab influence on oyster reproduction, must be tempered by the fact that the infestation rates observed were low. Consequently, the impact on the oyster populations in Georgia would appear to be minimal. However, given the high rates of pea crab infestation in oysters reported elsewhere, the ap- parent negative impact may be extensive and could have more far-reaching implications. ACKNOWLEDGMENTS The authors wish to thank Michelle L. Jansen who helped with the histological samples. This work was supported by the University of Georgia Marine Extension Service and the Georgia Sea Grant College Program under grant number NA 84AA-D-00072. LITERATURE CITED Beacu, N. W. 1969. The oyster crab, Pinnotheres ostreum Say, in the vicinity of Beaufort North Carolina. Crustaceana 17: 87-199. BIERBAUM, R. M. & S. FERSON. 1986. Do symbiotic pea crabs decrease growth rates in mussels? Biological Bulletin 170: 51-61. BIERBAUM, R. M. & S. E. SHUMWAY. 1988. Filtration and oxygen consumption in mussels, Mytilus edulis, with and without pea crabs, Pinnotheres maculatus. Estuaries 11:264—271. Dix, T. G. 1973. Mantle changes in the pearl oyster Pinctada maxima induced by the pea crab Pinnotheres villosulus. The Veliger 15:330—331. FLower, FE B. & J. J. McDeERMoTT. 1952. Observations on the occurrence of the oyster crab, Pinnotheres ostreum, as re- lated to oyster damage in Delaware Bay. Proceedings of the National Shellfisheries Association 43:44—S0. Ga tTsorF, P. S. 1964. The American oyster, Crassostrea virgin- ica, Gmelin. United States Fish and Wildlife Service, Fish- ery Bulletin 64:1—480. Haines, C. M. C., M. EDMUNDS & A. R. Pewsey. 1994. The pea crab, Pinnotheres pisum (Linnaeus, 1767), and its associa- tion with the common mussel, Mytilus edulis (Linnaeus, 1758), in the Solent (UK). Journal of Shellfish Research 13: 5-10. Harris, D. C. 1980. Survey of the Intertidal and Subtidal Oyster Resources of the Georgia Coast. Georgia Department of Nat- ural Resources, Coastal Resources Division (Project No. 2- 234-R), Brunswick, Georgia. 44 pp. The Veliger, Vol. 42, No. 1 Haven, D. S. 1958. Effects of pea crabs, Pinnotheres ostreum, on oysters, Crassostrea virginica. Proceedings of the Na- tional Shellfisheries Association 49:77—86. Jones, J. B. 1977. Natural history of the pea crab in Wellington Harbour, New Zealand. New Zealand Journal of Marine and Freshwater Research 11:667—676. KRUCZYNSKI, W. L. 1972. The effect of the pea crab, Pinnotheres maculatus Say, on growth of the bay scallop, Argopecten irradians concentricus (Say). Chesapeake Science 13:218— 220. KRucCZYNSKI, W. L. 1974. Relationship between depth and oc- currence of pea crabs, Pinnotheres maculatus, in blue mus- sels, Mytilus edulis, in the vicinity of Woods Hole Massa- chusetts. Chesapeake Science 15:167—169. Linton, T. L. 1968. Feasibility of raft culturing oysters in Geor- gia. Pp. 69-73 in T. L. Linton (ed.), Proceedings of the Oys- ter Culture Workshop, July 11—13, 1967. University of Geor- gia and Georgia Game and Fisheries Commission, Marine Fisheries Division, Contribution Series Number 6. McDermott, J. J. 1962. The incidence and host-parasite rela- tions of pinnotherid crabs (Decapoda, Pinnotheridae). Pro- ceedings of the First National Coastal Shallow Water Re- search Conference. 1961:162—164. O’BEIRN, F X., P. B. HEFFERNAN & R. L. WALKER. 1995. Pre- liminary recruitment studies of the eastern oyster, Crassos- trea virginica, and their potential applications in coastal Georgia. Aquaculture 136:231—342. O’BEIRN, F. X., P. B. HEFFERNAN, R. L. WALKER & M. L. JANSEN. 1996. ‘Young of the year’ oyster, Crassostrea virginica, re- production in coastal Georgia. Estuaries 19:651—658. PaRKS, P. 1968. A comparison of chemical composition of oys- ters (Crassostrea virginica) from different ecological habi- tats. Pp. 147-173 in T. L. Linton (ed.), Proceedings of the Oyster Culture Workshop, July 11-13, 1967. University of Georgia and Georgia Game and Fisheries Commission, Ma- rine Fisheries Division, Contribution Series Number 6. SpRUCK, C. R., R. L. WALKER, M. L. SWEENEY & D. H. HURLEY. 1995. Gametogenic cycle in the non-native Atlantic surf- clam, Spisula solidissima (Dillwyn, 1817) cultured in the coastal water of Georgia. Gulf Research Reports 9:131—137. STAUBER, L. A. 1945. Pinnotheres ostreum parasitic on the American oyster, Ostrea (Gryphea) virginica. Biological Bulletin 88:269—291. TABLADO, A. & J. Lopez-Gappa. 1995. Host-parasite relation- ships between the mussel, Mytilus edulis L., and the pea crab, Tumidotheres maculatus (Say), in the southwestern At- lantic. Journal of Shellfish Research. 14:417—423. WILLIAMS, A. B. 1984. Shrimps, Lobsters and Crabs of the At- lantic Coast of the Eastern United States, Maine to Florida. Smithsonian Institution Press: Washington D.C. 550 pp. THE VELIGER © CMS, Inc., 1999 The Veliger 42(1):21—53 (January 4, 1999) The Genus Littoraria Griffith & Pidgeon, 1834 (Gastropoda: Littorinidae) in the Tropical Eastern Pacific DAVID G. REID Department of Zoology, The Natural History Museum, London SW7 5BD, United Kingdom Abstract. Six species of Littoraria Griffith & Pidgeon, 1834, are recognized in the Panamic Province: L. pintado pullata (Carpenter, 1864), L. varia (Sowerby, 1832), L. zebra (Donovan, 1825), L. variegata (Souleyet, in Eydoux & Souleyet, 1852) (=L. fasciata of authors, not Gray, 1839), L. rosewateri new species, and L. aberrans (Philippi, 1846). The shell, reproductive anatomy, and radula of each are described and illustrated. Three possible interspecific hybrids are recorded. Littoraria pintado pullata occurs on oceanic rocky shores, but the others are found in mangrove habitats. Distribution maps are given. This diversity of species is much lower than in the Indo-West Pacific, and only L. pintado is common to the two provinces. Morphological comparison does not suggest any obvious sister-species pairs on either side of the Isthmus of Panama, supporting the idea that speciation and/or extinction since the formation of the Isthmus has obscured such relationships. Phylogenetic relationships with other members of the genus are discussed. Comparisons of shell morphology confirm trends previously demonstrated in Indo-West Pacific species: those species zoned at higher levels on mangrove trees have thinner shells and are more variable (or polymorphic) in shell color. Extreme intraspecific variation in radular morphology is described in three of these species. Littoraria aberrans is one of only four ovoviviparous species with intracapsular metamorphosis in the Littorinidae. INTRODUCTION The genus Littoraria Griffith & Pidgeon, 1834, consists of a group of 36 littorinid species. In cladistic analyses of morphological characters, the genus has been clearly recognized by two unreversed synapomorphies (closed prostate gland and lack of mamilliform penial glands) which, while not individually unique within the family, combine to define it as a monophyletic group (Reid, 1986, 1989). Its members are mainly tropical in distribution and, although some of the basal species occur in the an- cestral habitat of the upper eulittoral on rocky shores, the majority show a close and often obligate association with mangroves, wood, and salt-marsh vegetation. In early tax- onomic works, species in this group were considered dif- ficult to delimit and characterize, since shells often show interspecific similarities and intraspecific variability, and color polymorphism is common. However, anatomical features, particularly of the reproductive tract, are now known to provide consistent and reliable characters for the identification of Littoraria species (Reid, 1986). Littorinid gastropods are intensively studied because of their abundance, accessibility on the shore, and world- wide occurrence. Within this well-known family, Litto- raria species show several peculiarities which make them of particular interest. Their association with mangrove and other vegetation is shared by only one other littorinid genus (Mainwaringia Nevill, 1885), and aspects of their field ecology, diet, and zonation patterns on the trees have been described (e.g., Reid, 1985; Kohlmeyer & Bebout, 1986; Newell & Barlocher, 1993; Blanco et al., 1995). Living on trees, often above the regular reach of the tide, they show behavioral and reproductive specializations, in- cluding vertical migration (Reid, 1984), lunar spawning rhythms (Berry & Chew, 1973; Gallagher & Reid, 1974), and ovoviviparity (Reid, 1986, 1989). Where sympatric Littoraria species occupy different vertical zones on the trees, they provide a clear example of the correlation be- tween the architectural defense of shells and the intensity of crushing by aquatic predators such as crabs and fish (Reid, 1984, 1986, 1992; Cook et al., 1985; Borjesson & Szelistowski, 1989). The species that inhabit the highest levels, among the foliage, often show discrete polymor- phism (sensu Ford, 1945) of shell color; these provide a model system for the study of maintenance and adaptive significance of color polymorphism (Cook, 1983, 1986, 1990, 1992; Hughes & Mather, 1986; Reid, 1986, 1987; Cook & Garbett, 1992). The systematics of Littoraria are now relatively well understood, particularly in the Indo-West Pacific prov- ince, where the 20 mangrove-associated species have been the subject of a taxonomic monograph (Reid, 1986). The remaining species are mostly familiar and easily identified (species lists in Reid, 1986, 1989). The prin- cipal clades indicated in a phylogenetic analysis of mor- phological characters have been recognized as subgenera, although the species-level phylogeny is not well resolved (Reid, 1986, 1989). Nevertheless, those species in the tropical Eastern Pacific have been neglected. Since the earliest faunistic studies of the mollusks in this region (Adams, 1852; Carpenter, 1857b; Mgrch, 1860), the three larger mangrove-associated species have been familiar (generally under the names L. varia, L. fasciata, and L. zebra), although only from their distinctive shells. These have been illustrated in the few modern identification guides for mangrove mollusks from the Panamic province (Zilch, 1954; Keen, 1958, 1971; Petia, 1971b; Alamo & Valdivieso, 1987), but shell characters are variable, and some confusion has persisted. At least in Colombia, they are gathered for food, and are of potential commercial importance (Cantera & Contreras, 1978). A fourth man- grove-associated species, the enigmatic L. aberrans, was for over a century known only from the shell of the ho- lotype (Philippi, 1846a), until briefly redescribed by Rosewater (1980b). An additional species, hitherto vari- ously classified as L. pullata, L. pintado, or L. pintado schmitti, occurs on the rocky shores of remote oceanic islands and peninsulas; only the shell has been illustrated (Bartsch & Rehder, 1939; Keen, 1958, 1971; Palmer, 1963; Rosewater, 1970), and its relationship to the Indo- West Pacific L. pintado has been considered (Reid, 1986). The anatomy of all these Littoraria species of the Eastern Pacific has been examined during the course of recent studies of the phylogeny and classification of the genus (Reid, 1986, 1989), and two electron micrographs of their radulae have been made (Rosewater, 1980a, b). However, no comprehensive descriptions have yet been published. Furthermore, the geographical distributions of these spe- cies are not known in any detail. The present study therefore aims to provide full de- scriptions of the Littoraria species of the tropical Eastern Pacific (Panamic) province. Radular characters are shown to be extraordinarily variable within species. The repro- ductive anatomy of L. aberrans is uniquely modified in the genus, and this is one of only four members of the Littorinidae that are ovoviviparous with intracapsular metamorphosis (Reid & Geller, 1997). One new species, hitherto confused with L. aberrans, is described. Nomen- clatural revision necessitates a change in the name of L. fasciata. The limited ecological information is reviewed, and supplemented by field observations. Distribution maps are plotted for each species, and their biogeography and relationships discussed in the context of the geolog- ical history of Central America. MATERIALS AnpD METHODS This account is based on examination of all material in the collections of the Natural History Museum, London (BMNH), the National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM), the Academy of Natural Sciences of Philadelphia (ANSP), and the Museum of Comparative Zoology, Harvard Uni- versity (MCZ). Personal collections of all species were made in Costa Rica (1985) and Mexico (1994), and are deposited in BMNH. Additional material was borrowed The Veliger, Vol. 42, No. 1 from the Los Angeles County Museum of Natural History (LACM) and the California Academy of Sciences (CAS). All available type material was examined. Shell dimensions were measured with vernier calipers to 0.1 mm. Shell height (H) is the maximum dimension parallel to the axis of coiling, shell breadth (B) the max- imum dimension perpendicular to H, and the length of the aperture (LA) the greatest length from the junction of the outer lip with the penultimate whorl to the anterior lip. For the purpose of diagnosis, shell shape was quan- tified simply as the ratio H/B and H/LA (relative spire height, SH), and the range of these ratios quoted. Proto- conch whorls were counted as recommended by Reid (1996). To describe the coiling of the operculum, the opercular ratio was defined as the ratio of two parallel measurements, the diameter of the spiral part divided by the maximum length (Reid, 1996). The relative radular length was the total radular length divided by shell height. Living animals were relaxed in 7.5% (volume of hy- drated crystals to volume of fresh water) magnesium chlo- ride solution. Sperm samples were removed from the sem- inal vesicles of relaxed, living animals, fixed in 0.5% sea- water formalin, examined immediately by light microsco- py, and drawn by camera lucida. Animals were fixed in 10% seawater formalin buffered with borax, and stored in 80% ethanol before dissection. For general accounts of the male and female anatomy of Littoraria, see Reid (1986). Radulae were cleaned by soaking in a hypochlorite bleach- ing solution at room temperature for about 5 min, rinsed in distilled water, mounted on a film of polyvinyl acetate glue on glass, allowed to dry in air, and coated with gold and palladium before examination in a scanning electron microscope. Unworn portions of radulae were viewed in three orientations: in standard flat view from vertically above the radula (to show shapes of teeth), at an angle of 45° from the front end of the radula (to show shapes of tooth cusps), and at an angle of 45° from the side of the radula (to show relief). The shape of the rachidian tooth was quantified as the ratio of the total length (in flat view) to the maximum basal width. The “hood” of the rachidian is a sharp flange (presumably an additional cutting edge) anterior to the main cusps of the tooth. The supraspecific classification employed follows that of Reid (1989). SYSTEMATIC DESCRIPTIONS Family LITTORINIDAE Anon., 1834 Genus Littoraria Griffith & Pidgeon, 1834 Type species: Littorina pulchra ‘Gray’? Sowerby, 1832 [= Turbo zebra Donovan, 1825] Diagnosis: Littorinidae without nodulose shell sculpture, with paucispiral operculum, egg groove of pallial oviduct coiled in a single spiral, salivary glands constricted by nerve ring; defining (but not unique) synapomorphies are DAGy Reid= 1999 Page 23 closed prostate gland and absence of mamilliform penial glands (after Reid, 1989). Subgenus Protolittoraria Reid, 1989 Type species: Turbo pintado Wood, 1828 Diagnosis: Penis not bifurcate; scattered simple penial glands not forming discrete glandular disc; copulatory bursa opening at posterior end of straight section of pallial oviduct; spawn of cupola capsules sculptured by one concentric ring; hood of rachidian tooth slight or absent; six to eight elongate cusps on outer marginal tooth (diagnosis modified from Reid, 1989). Littoraria (Protolittoraria) pintado pullata (Carpenter, 1864) (Figures 1, 2A—C, 3A, 4A, B, 5A—E, 6A) Litorina sp. Carpenter, 1857b: 350 (see Carpenter, 1864a). Litorina pullata Carpenter, 1864a: 477 (Cape St Lucas [Cape San Lucas, Baja California, Mexico]; lectotype (here designated, 11.3 mm, Figure 1B) USNM 12661, seen; 2 paralectotypes USNM 635481, seen; 7 paralectotypes BMNH 1865.12.6.69, seen; 3 paralectotypes BMNH 1968357, seen; 3 paralectotypes ANSP 18627, seen). Carpenter, 1864b: 546, 618. Weinkauff, 1882: 106. Littorina (Melaraphe) pullata—Keep & Baily, 1935: 199. Littorina (Melarhaphe) scutulata pullata—Burch, 1945: 12. Palmer, 1958: 159. Littorina pullata—Keen, 1958: 282; fig. 177. Palmer, 1963: 335-336; pl. 61, fig. 6. Keen, 1971: 366; fig. 186. Ab- bott, 1974: 69. Littorina (Littoraria) pullata—Rosewater, 1970: 423, 447. Littoraria pintado pullata—Reid, 1996: 11. Littorina (Melaraphe) scutulata—Tryon, 1887: 250; pl. 45, fig. 3 (in part, includes Littorina scutulata and Littorina plena; not Gould, 1849). Littorina scutulata—Abbott, 1974: 67—68 (in part, includes Littorina scutulata and Littorina plena; not Gould, 1849). Littorina schmitti Bartsch & Rehder, 1939: 9-10; pl. 2, fig. 4 (shore south of landing, Clipperton Island; holotype USNM 472547, Figure 1F seen). Keen, 1971: 366; fig. 187. Littorina (Littoraria) pintado schmitti—Rosewater, 1970: 423, 449-450; pl. 346, figs 13-16. Littoraria (Littoraria) pintado—Reid, 1986: 64, 73 (not Tur- bo pintado Wood, 1828, which is the nominate subspe- cies). Littoraria (Protolittoraria) pintado—Reid, 1989: 96 (not Wood, 1828). Taxonomic history: Despite an adequate initial descrip- tion (Carpenter, 1864a), this subspecies has long re- mained misunderstood and poorly known. Following Tryon (1887), it has often been considered a color form or subspecies of Littorina scutulata (Burch, 1945; Palmer, 1958; Abbott, 1974; see Reid, 1996). Bartsch & Rehder (1939) gave the name Littorina schmitti to examples from Clipperton Island, and noted a relationship to “‘Littorina”’ pintado from the Indo-West Pacific. However, the con- specificity of Mexican specimens with those from Clip- perton Island, and with Littoraria pintado, was only pointed out much later (Reid, 1986). Diagnosis: Shell smooth with fine incised spiral lines, brown to black, often with pale flecks and spiral lines, aperture brown, columellar pillar white. Penis long, sim- ple, no glandular structures visible externally. Shell (Figure 1): Mature shell height 5—16.9 mm. Shape high-turbinate to elongate (H/B = 1.47—1.71, SH = 1.47-— 2.02); spire whorls only slightly rounded, sutures slightly impressed; indistinct angulation at periphery of last whorl; of moderate thickness. Mature lip not flared; col- umellar pillar long, straight and somewhat flattened, only slightly hollowed at base. Sculpture smooth except for fine incised spiral lines over whole surface, 8—13 above periphery of last whorl, but often indistinct or obsolete; entire surface with fine spiral microstriae if well pre- served; no discernible periostracum. Protoconch 0.30 mm diameter, about 3.5 whorls, terminated by sinusigera rib, sculpture not preserved. Color: densely pigmented, large- ly obscuring pale ground color; effect is chocolate brown to black with variable patterning of pale grey to white: finely flecked, marbled or tessellated, alternatively with narrow spiral lines (2-17 on last whorl), or combination of flecks and lines; pale patterning usually stronger on base and from shoulder to suture; shell rarely almost en- tirely black. Columellar pillar white, edged with choco- late brown; interior blackish brown with pale lines show- ing through. Animal: Head, tentacles, and sides of foot dark grey to black, sometimes a pale stripe behind eye and pale spot at inside of tentacle base. Opercular ratio 0.31—0.36. Pe- nis (Figure 2A—C) long, vermiform, tapering only near tip; fine annular wrinkles extend almost to tip, so that filament is not differentiated from base; base not bifur- cate, no glandular disc visible externally, but base prob- ably contains simple subepithelial glandular cells (as con- firmed by histological examination of nominate subspe- cies, Reid, 1989); sperm groove open (also anterior vas deferens from prostate), extending to tip of filament; un- pigmented except for small grey or blackish area at very base. Euspermatozoa 107-114 wm; paraspermatozoa (Figure 3A) spherical to oval, maximum diameter 15—22 jm, packed with large spherical granules (to 6 pm di- ameter), single rod-pieces small and often irregular (6— 14 wm long). Pallial oviduct (Figure 4A, B) with spiral section of 3.5 whorls, of which capsule gland (with prox- imal opaque and distal translucent portion) about two- thirds of a whorl; bursa small, at posterior end of straight section of pallial oviduct. Spawn and development not observed; presence of capsule gland suggests pelagic egg capsule (pelagic cupola capsule with single annular ridge and single ovum described in nominate subspecies; Page 24 The Veliger, Vol. 42, No. 1 Figure 1 Shells of Littoraria pintado pullata. A. Bahia Santa Maria, Baja California, Mexico (BMNH 1996198; small, black shell form from algal pools at top of eulittoral zone). B. Lectotype of Littorina pullata Carpenter, 1864 (USNM 12661); Cabo San Lucas, Baja California, Mexico. C. Socorro Island, Mexico (CAS 96233). D. Cabo San Lucas, Baja California, Mexico (BMNH 1996199). E. Punta Arena, Cerralvo Island, Gulf of California, Mexico (CAS 107727). E Holotype of Littorina schmitti Bartsch & Rehder, 1939 (USNM 472547); Clipperton Island. G. Cabo San Lucas, Baja California, Mexico (BMNH 1996199). H. Maria Madre Island, Tres Marias Islands, Mexico (CAS 32564). Scale bar = 10 mm. Figure 2 Penes and heads of Littoraria pintado pullata (A—C), L. variegata (D, E, L), L. varia (FG, J, K) and L. zebra (H, I). A-C. Penes of L. pintado pullata. A, B. Cabo San Lucas, Baja California, Mexico (BMNH 1996209; shell H of A = 9.7 mm, shell H of B = 10.4 mm). C. Bahia Santa Marfa, Baja California, Mexico (BMNH 1996210; shell H = 7.1 mm). D, E. Penes of L. variegata. D. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996211; shell H = 20.3 mm). E. Topolobampo, Sinaloa, Mexico (BMNH 1996212; shell H = 15.9 mm). E G. Penes of L. varia. F Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996213; shell H = 19.5 mm). G. Panama (BMNH 1867.5.22.27; shell H = 28.9 mm). H, I. Penes of L. zebra. H. Golfito, Costa Rica (BMNH 1996214; shell H = 26.8 mm). I. Puntarenas, Costa Rica (BMNH 1996215; shell H = 29.8 mm). J. Penis of L. varia; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996213; shell H = 19.1 mm). K. Head of L. varia; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996213). L. Head of L. variegata; Topolobampo, Sinaloa, Mexico (BMNH 1996212). Abbreviations: f, penial filament; pd, penial glandular disc. D. G. Reid, 1999 1mm } Lye = aN er 2mm K,L The Veliger, Vol WA? Nowl 20 um DG] Reidy 1999 Struhsaker, 1966); type of protoconch indicates plankto- trophic development. Radula (Figure 5A—E): Relative radular length 1.46— 4.09. Rachidian: length/width 0.92—1.37; cusps variable: central cusp pointed (Figure 5D), elongate leaf-shaped (Figure 5E) or very elongate with rounded tip (Figure 5B); smaller pointed cusp and outer denticle on either side; hood developed only as small ridge (Figure 5C—E), or sometimes absent (Figure 5A, B). Lateral: five to six cusps, largest central cusp elongate, but variable in shape (pointed, leaf-shaped or bluntly rounded); two small pointed cusps on inside and two to three on outside of main cusp. Inner marginal: four cusps, largest central cusp shaped like that on lateral; two smaller pointed cusps on inside and one on outside of main cusp. Outer mar- ginal: six to eight elongate pointed cusps, outermost larg- est. See Remarks. Material examined: Types as indicated; 26 lots; two pro- toconchs; nine penes; two sperm samples; six pallial ovi- ducts; eight radulae. (Of nominate subspecies: lectotype of Turbo pintado Wood, 1828, BMNH 1968368; 50 lots; one protoconch; seven penes; three sperm samples; four pallial oviducts; three radulae). Habitat: On rock (including granite, beachrock, and con- crete) in uppermost eulittoral and low littoral fringe, clus- tered in crevices and on sides of rocks; at one locality (Bahia Santa Maria, NE of Cabo San Lucas, Baja Cali- fornia) submerged or clustered at margins of small algal pools at top of shore; usually on strongly wave-exposed shores. Occurs only at sites with clear, oceanic water. Abundant only at tip of Baja California and on oceanic islands (e.g., 5375 per m? at Socorro Island (Mille-Pagaza et al., 1994). Range (Figure 6A): Southern Baja California from To- dos Santos (BMNH) to 35 km N of La Paz (24°21'N;: BMNH; but common only close to Cabo San Lucas); Clarion Island and Socorro Island in the Revillagigedo Islands (CAS, LACM); Tres Marias Islands (CAS); Clip- perton Island (USNM). There is also a single collection of four specimens from Cocos Island (5°33’N, LACM) much farther to the southeast; since this species is char- acteristically found on oceanic islands, this record is Page 27 probably reliable. Only occasional specimens have been found on the mainland of Mexico, e.g., 25 km SW Puerto Vallarta (BMNH), 10 km N of Melaque (18°48’'N; BMNH), and recorded from Mazatlan (Carpenter, 1857b, 1864a). Records from the state of California (e.g., Burch, 1945; Abbott, 1974) are believed to be misidentifications of Littorina scutulata and/or Littorina plena Gould, 1849, as in one lot in CAS. Remarks: The nominate subspecies, L. pintado pintado, has a very wide distribution in the Indo-West Pacific; this range is disjunct, with one area in the south and western Indian Ocean (southeast Africa, Madagascar, Mascarene Islands, Somalia) and another in the northern and western Pacific (Ryukyu, Bonin, Caroline, Mariana, Marshall, and Hawaiian Islands) (Rosewater, 1970; Reid, 1986). Al- though the two areas of occurrence are separated by about 8000 km, no consistent morphological differences in shells, anatomy, or radulae have been detected (personal observation). The Eastern Pacific records of L. pintado pullata are at least 4500 km from the closest known occurrence of the nominate subspecies in the Indo-West Pacific, in the Hawaiian Islands. The close relationship between Eastern Pacific and Indo-West Pacific forms was first noted by Bartsch & Rehder (1939) when they described shells from Clipperton Islands as a new species, schmitti, al- though they did not mention pullata. Curiously, Rose- water (1970) reduced schmitti to a subspecies of pintado, while remarking that pullata was an ‘“‘apparent analogue”’ of pintado in the Eastern Pacific. Reid (1986) found no differences in the reproductive anatomy of pullata and pintado, and synonymized all three names. This has been confirmed in the present study of additional material, which has included examination of sperm and radulae. However, there are consistent differences in shell color- ation. Since all known species of Littoraria differ from each other in penial shape (Reid, 1986), separation at spe- cific level does not seem warranted at present. The cat- egory of subspecies is appropriate for such a case of mi- nor differentiation combined with allopatric distribution, and carries the implication that although the differentia- tion probably reflects genetic isolation, there is no mor- phological evidence for reproductive isolation. In the Figure 3 Paraspermatozoa of Littoraria pintado pullata (A), L. varia (B, C), L. zebra (D), L. variegata (E, F), L. rosewateri Reid, sp. nov. (G, H) and L. aberrans (1). A. Cabo San Lucas, Baja California, Mexico (BMNH 1996209). B. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996213). C. Estero Aguadulce, Bahia Parita, Panama (USNM 733057; alcohol preserved). D. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996216). E. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996211). F Estero Aguadulce, Bahia Parita, Panama (USNM 733055; alcohol preserved, therefore granules indistinct). G, H. Golfito, Costa Rica (BMNH 1996217). I. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218). All samples from single individuals; unless otherwise noted, all samples fixed in 0.5% seawater formalin. Page 28 The Veliger, Vol. 42, No. 1 D. G. Reid, 1999 nominate subspecies, coloration is predominantly white to pale grey, with spiral rows of small black or brown spots or flecks (18-40 rows on final whorl); although the spotting may occasionally be quite dense, the spots do not fuse to form spiral bands. In contrast, in L. pintado pullata the predominant coloration is black to brown, and in addition to varying degrees of pale spotting or mar- bling, most specimens show white spiral lines. The geographical distribution of L. pintado is of partic- ular interest, since it is one of the few molluscan species, and the only littorinid, to span both Indo-West Pacific and Eastern Pacific provinces. These zoogeographic regions are separated by “Ekman’s Barrier,” a 5000 km expanse of deep ocean without island stepping stones, which appears to have acted as an effective barrier for most shallow-water benthic invertebrates (Vermeij, 1987; Richmond, 1990). However, the barrier is not complete; although almost no Panamic mollusks are known from the Indo-West Pacific, a small number of typically Indo-West Pacific species have been recorded in the Eastern Pacific. The most recent com- pilation of these listed 61 prosobranch gastropods, of which 56% are found only on the oceanic islands off the American mainland (Emerson, 1991). The rarity of most of these species, combined with their absence from the limited fossil record of western Central America, suggests that the majority are recent (post-Pliocene) arrivals derived by dispersal from the Central Pacific, and that many of the species may be unable to maintain viable populations with- out replenishment from the source areas to the west (Em- erson, 1991). Eastward dispersal is believed to take place mainly by larval transport in the North Equatorial Coun- tercurrent, in which drifter buoys have covered the distance from the Line Islands to the Eastern Pacific in as little as 100 days (Richmond, 1990). Most of the Indo-West Pacific immigrants, particularly the tonnoideans, are known to possess long-lived (teleplanic) larvae able to survive in the plankton for this length of time (Scheltema, 1988). During the periodic El Nifio events which considerably alter oceanographic patterns in the Central and Eastern Pacific, this passage may be accomplished in half the time, but the main source area is still considered to be in the Line Is- lands, lying in the eastward flow of the North Equatorial Page 29 Countercurrent (Richmond, 1990). The evolutionary con- sequences of trans-Pacific dispersal appear to have been limited. For ecological reasons, some arrivals in the East- ern Pacific may be unable to maintain viable populations there. In those that do, trans-Pacific dispersal is apparently frequent enough to prevent isolation and genetic diver- gence in the Eastern Pacific; among gastropods, only four Eastern Pacific species or subspecies (excluding L. pintado pullata) were interpreted as endemic derivatives of recent immigrants from the Indo-West Pacific by Vermeij (1990). In many respects, however, L. pintado does not con- form to the distributional and developmental character- istics shown by other gastropods with trans-Pacific dis- tributions. Quite clearly, L. pintado pullata maintains vi- able populations in the Eastern Pacific, at least in Baja California and on the oceanic islands where it is abun- dant. Indeed, throughout the range of the species as a whole, it seems to be found largely on oceanic islands, and this habitat specialization explains its almost com- plete absence from the mainland of Mexico and Central America. Whether the island populations are in genetic contact is unknown, but surface current patterns (Wyrtki, 1965) suggest that this is possible. During the period May to December, the strong North Equatorial Countercurrent could perhaps transport egg capsules and larvae from Clipperton Island to Cocos Island to the east, sweeping northwestward parallel with the Mexican coast toward the Revillagigedo Islands. During the winter this current dis- appears, while from February to June the California Cur- rent flows south and southeast, turning westward to join the North Equatorial Current, thus potentially connecting the populations of Baja California, the Revillagigedo Is- lands, and Clipperton Island. The spawning season in the Eastern Pacific is unknown, but in Hawaii L. pintado pin- tado produces egg capsules all year round (Struhsaker, 1966). The length of larval life has not been recorded; when reared in the laboratory in Hawaii, veligers sur- vived for 11 days from the time of spawning (Struhsaker, 1966). In a littorinid species with a similar protoconch, Nodilittorina hawaiiensis Rosewater & Kadolsky, 1981, the total time to larval settlement in the laboratory was 3—4 weeks (Struhsaker & Costlow, 1968, as Littorina pic- Figure 4 Pallial oviducts of Littoraria pintado pullata (A, B), L. varia (C), L. variegata (D) and L. zebra (E). A. Bahia Santa Maria, Baja California, Mexico (BMNH 1996210; shell H = 8.8 mm). B. Cabo San Lucas, Baja California, Mexico (BMNH 1996209; shell H = 12.6 mm). C. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996213; shell H = 25.2 mm). D. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996211; shell H = 22.0 mm). E. Golfito, Costa Rica (BMNH 1996214; shell H = 27.0 mm). Transverse sections taken at levels indicated, viewed from anterior (i.e., right side of figure). Abbreviations: b, copulatory bursa (dashed line; visible only by dissection); eg, egg groove (visible externally if darkly pigmented, as in E, then indicated by thick line); oag, opaque albumen gland (light stipple); ocg, opaque capsule gland (dark stipple); sg, sperm groove (leading ventrally to seminal receptacle); sr, seminal receptacle (darkest stipple); tag, translucent albumen gland (lightest stipple); tcg, translucent capsule gland (cross-hatched); in sections, spiral lumen is black. Page 30 The Veliger, Vol. 42, No. 1 D. G. Reid, 1999 Revillagigedo Is 10° ri Clipperton I. 10° A Page 31 B Cc D E F Figure 6 Distribution maps of six Eastern Pacific Littoraria species. A. L. pintado pullata. B. L. varia. C. L. zebra. D. L. variegata. E. L. rosewateri Reid, sp. nov. E L. aberrans. Key: closed circles, material examined; open circles, records from literature (1, Pena, 1970, 1971b; 2, Guerrini, 1990; 3, Pilsbry & Lowe, 1932). ta). This is sufficient to permit transport for approximate- ly 1000 km at some of the faster average current flows suggested by Wyrtki (1965), but is much shorter than the developmental times of long-lived teleplanic larvae (Scheltema, 1988). Larval dispersal between the more distant islands is therefore unlikely to be frequent in nor- mal seasons, but might take place under the exceptional conditions of El] Nifio events (Richmond, 1990). Litto- raria pintado pullata is notably absent from the Galapa- gos Islands (Finet, 1994) which, although only 750 km from Cocos Island, lie outside the path of the North Equa- torial Countercurrent (Finet, 1991). Another peculiarity of L. pintado in this context is that it is distributed in the northern Central Pacific (Rosewater, 1970; personal ob- servation of museum collections), the closest occurrence to the Eastern Pacific being in the Hawaiian Islands. In comparison, all but one of the other 60 trans-Pacific pros- obranchs listed by Emerson (1991) occur in the Line Is- lands and/or French Polynesia (although many do in ad- dition occur in the Hawaiian Islands). Since the Hawaiian Islands are so distant (about 4500 km) from the range of L. pintado pullata, and furthermore lie in the weak west- ward-flowing North Equatorial Current (McNally et al., 1983), it is improbable that there is any gene flow be- tween the populations in the Indo-West Pacific and the Eastern Pacific, even during the exceptional El Nino Figure 5 Radulae of Littoraria pintado pullata (A-E) and L. variegata (F—H). A, B. Cabo San Lucas, Baja California, Mexico (BMNH 1996209; two views of radula, flat and at 45°; shell H = 10.4 mm). C, D. Bahia Santa Maria, Baja California, Mexico (BMNH 1996210; two views of radula, flat and at 45°; shell H = 8.8 mm; small, black shell form from algal pools at top of eulittoral zone). E. Bahia Santa Maria, Baja California, Mexico (BMNH 1996210; at 45°; shell H = 8.7 mm; normal shell form from open rock surfaces). E Topolobampo, Sinaloa, Mexico (BMNH 1996212; at 45°; shell H = 15.9 mm). G, H. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996211; two views of radula, flat and at 45°; shell H = 20.3 mm). Abbreviation: h, “hood” of rachidian tooth. Scale bars = 50 wm. Page 32 events (Richmond, 1990). This likely genetic isolation may be reflected by the consistent differences in the shell pigmentation of the two, and these arguments support the assertion that L. pintado pullata should be recognized at least as a distinct subspecies. Information on the genetic interrelationships of L. pintado throughout its range would be most desirable. Unfortunately, littorinids of high-energy rocky shores are seldom preserved as fossils, and the marine Tertiary record of western Central America is poor, so that there is no fossil evidence for the history of L. pintado in the Eastern Pacific. Nevertheless, the evidence reviewed above does suggest that L. pintado is not in the same category as those other non-tonnoidean gastropods with trans-Pacific distributions, which have been interpreted as recent colonizers, sometimes tenuously established, which have in general not differentiated from their parent populations in the Central Pacific (Vermeij, 1987, 1990; Emerson, 1991). Instead, L. pintado pullata can be added to the four possible examples of Eastern Pacific endemic gastropods derived from Indo-West Pacific immigrants (listed by Vermeij, 1990). The origin of L. pintado pullata is still likely to have been relatively recent in geological terms, since the islands on which it occurs are all of Pleis- tocene age, and elsewhere in the Eastern Pacific only the Galapagos are older (Emerson, 1978). Although dispersal from the Hawaiian Islands to the Eastern Pacific appears to be unlikely under present conditions, L. pintado may have been more widely distributed in the Central Pacific in the past. For example, during the sea level fluctuations of the Pleistocene, suitable habitat (high oceanic islands lacking well-developed reefs) may have been more wide- spread. An alternative scenario, not requiring dispersal across the Pacific, is that the distribution of L. pintado is an ancient one, predating the Miocene division of the Tethys Sea and the Pliocene formation of the Isthmus of Panama. This is not credible, in view of the geological ages of the Eastern Pacific islands. Furthermore, the mor- phological identity of the two subspecies is unlikely to have been maintained if they have been separated since vicariance of an ancient Tethyan distribution. Once again, genetic evidence will be valuable in testing this assertion. Morphologically, this species is of interest as it shows a number of features suggesting that it is the basal branch within the clade Littoraria (phylogenetic analyses by Reid, 1986, 1989). These include the cupola-type egg capsule (biconvex elsewhere in the genus), lack of a dis- crete penial glandular disc, and the poorly developed or absent hood on the rachidian radular tooth (hitherto re- corded only as absent; Reid, 1986, 1989). If correct, this implies that L. pintado (or the clade of which it is the only surviving member) is at least as old as any other Littoraria species. Other members of this genus are re- corded from the Lower Eocene (Reid, 1989). This does not, however, affect the biogeographic scenarios dis- The Veliger, Vol. 42, No. 1 cussed above, which depend upon the age of the sepa- ration of the two subspecies, and not on that of the clade. Radular variation is striking in this species. Five spec- imens from Bahia Santa Maria, Baja California, suggest a possible correlation with shell form or habitat. Two ex- amples of a peculiar small (< 8.8 mm) almost black shell form, with eroded spire, collected from the unusual hab- itat of algal pools high on the shore (Figure 1A), showed radulae with relatively smaller and more pointed major cusps, and more well developed rachidian hood (Figure 5C, D). Three shells of typical form (7.2-14.7 mm), col- lected on open rock surfaces at the same locality (similar to Figure 1D), showed radulae with more elongate cusps and only slightly hooded rachidian (Figure 5E); these were similar to examples from Cabo San Lucas (Figure 5A, B), and to specimens of L. pintado pintado from Ha- waii and Mauritius. No other anatomical differences were detected among specimens from Bahia Santa Maria. Fur- ther investigation is required, but a possible explanation of these preliminary observations is an ecophenotypic ef- fect on radular tooth shape, as has recently been dem- onstrated in the littorinid Lacuna (Padilla, 1998). This species is common only near Cabo San Lucas and on the relatively inaccessible offshore islands, which ex- plains why it has remained poorly known, and has seldom been figured or described. Among other littorinids in the Panamic province, confusion is possible with Nodilittorina species such as N. aspera (Philippi, 1846) and N. penicil- lata (Carpenter, 1864); these shells are separated by their entirely brown columella and more pronounced oblique axial stripes of black and white; anatomically, Nodilittorina species ‘have a bifurcate penis with a glandular disc and single mamilliform gland, and a pallial oviduct with single loop in the albumen gland only (Reid, 1989). In the Cal- ifornian province, Littorina scutulata and Littorina plena are similar in shell outline, but generally show coarser tes- sellation and lack spiral lines except on the base; their penes have glandular protrusions and there are three con- secutive spiral loops in the pallial oviduct (Reid, 1996). The distinction from the nominate subspecies in the Indo- West Pacific has been described earlier. Subgenus Littoraria Griffith & Pidgeon, 1834 Diagnosis: Penis usually bifurcate, with differentiated pe- nial glandular disc; paraspermatozoa lacking pseudotrich (Healy & Jamieson, 1993; ‘‘flagellum” of Reid, 1986, 1989); spawn of biconvex discoidal capsules; rachidian tooth usually hooded (diagnosis modified from Reid, 1989). Note that phylogenetic analysis of Reid (1989) sug- gested this is a paraphyletic or polyphyletic assemblage. Littoraria (Littoraria) varia (G.B. Sowerby, 1832) (Figures 2F G, J, K, 3B, C, 4C, 6B, 7A-C, 8A-E) Littorina varia G.B. Sowerby, 1832: part 37; pl. 211, fig. 4 (Panama; lectotype (here designated) Sowerby, 1832: DG. Reid 11999 pl. 211, fig. 4). Adams, 1852: 400—401. Souleyet, in Eydoux & Souleyet, 1852: 561; atlas pl. 31, figs 43—45 (as ““Littorine costulée” in caption). Reeve, 1857: Lit- torina sp. 19; pl. 4, fig. 19a, b. Merch, 1860: 69. Dall, 1909: 231, 285 (in part, includes Littoraria variegata). Keen, 1958: 282; fig. 178. Keen, 1971: 366; fig. 188. Pena, 1971b: 47. Rosewater, 1980a: 5; figs 3, 4 (radula). Guerrini, 1990: 14. Litorina varia—Philippi, 1846b: 2: 99-100; Litorina pl. 1, figs 2, 3. Weinkauff, 1882: 53; pl. 6, figs 14, 15. Littorina (Melaraphe) varia—Tryon, 1887: 246; pl. 43, fig. 44 (in part, includes Littoraria variegata). Littorina (Littorinopsis) varia—von Martens, 1900: 580. Rosewater, 1970: 423. Alamo & Valdivieso, 1987: 25; fig. 38. Littorina (Algaroda) varia—Zilch, 1954: 81; pl. 3, fig. 8. Littoraria (Littoraria) varia—Reid, 1986: 73; figs 4j (penis), 18 (cladogram). Reid, 1989: 97. ?Littorina perdix King & Broderip, 1832: 345 (no locality; types lost). Littorina costulata ‘Souleyet’ Tryon, 1887: 246, 292 (nomen nudum). Littorina (Littorinopsis) fasciata—Abbott, 1974: 69; pl. 3, fig. 567 (in part; includes Littoraria variegata; not Gray, 1839 = Littoraria zebra). Taxonomic history: No type specimens are known to exist and a lectotype figure is here designated. Neverthe- less, there is no uncertainty about the identity of this tax- on, and the name varia has been employed by most au- thors, in various generic combinations, throughout its his- tory. Tryon (1887) and Dall (1909) had a broader concept of this taxon, including L. variegata. The identity of Lit- torina perdix King & Broderip, 1832, is uncertain; no original material has been located in BMNH. The original diagnosis was inadequate, but the dimensions given (equivalent to 20.6 X 13.5 mm), together with the raised spiral striae, and white aperture with brown-spotted mar- gin, support its synonymy with L. varia, and preclude all other South American littorinids. No locality was given; the title of the paper suggests that all specimens were collected on the South American voyages of the Adven- ture and Beagle between 1826 and 1830, neither of which visited the geographical range of L. varia (King, 1839). Nevertheless, in the same paper some species were also described from the Cuming and Sowerby collections, from localities such as Lima and Panama, which were not visited during these voyages. The identity of Littorina perdix with L. varia is therefore a possibility. Diagnosis: Shell thick-walled; sculpture of strong spiral ribs; color whitish with minute brown flecks, aperture and columella white. Penis with small filament, large glan- dular disc on branch of base, unpigmented. Shell (Figure 7A—C): Mature shell height 16-34.4 mm. Shape high-turbinate (H/B = 1.41-1.53, SH = 1.51- 1.65); spire whorls only slightly rounded, sutures slightly channelled; angulation at periphery of last whorl marked by largest rib; thick-walled. Mature lip not flared; colu- mella broad and hollowed. Sculpture of strong spiral ribs, Page 33 about 12—16 on last whorl, with one to two smaller cords in the broad spaces between each; axial growth lines dis- tinct, especially near end of last whorl, giving slightly cancellate appearance between ribs; surface shiny, with only faint spiral microstriae; no discernible periostracum. Protoconch 0.31 mm diameter, about three whorls, ter- minated by sinusigera rib, sculpture not preserved. Color whitish to cream, closely and minutely flecked with dark brown; pattern sometimes aligned near suture to form close, narrow, axial stripes; pattern occasionally faint or absent. Columella and interior of aperture white; apertural margin marked with dark brown spots where pattern shows through. Animal: Head, tentacles and sides of foot black; unpig- mented stripe behind eye (Figure 2K). Opercular ratio 0.33-—0.40. Hypobranchial gland exceptionally large, up to 3.0 mm wide in shell of 25.3 mm. Penis (Figure 2F G, J) with large, wrinkled base and small smooth filament (15-20% total length); base bifurcate, broader branch bearing large, slightly pointed, glandular disc; sperm groove open (also anterior vas deferens from prostate), extending to tip of filament; unpigmented. Euspermatozoa 136 wm; paraspermatozoa (Figure 3B, C) round to slight- ly oval, maximum diameter 21—26 ym, packed with mi- nute indistinct granules, one or two oval to elongate rod- pieces 4-17 wm long. Pallial oviduct (Figure 4C) with spiral section of 5.5—6.5 whorls, of which capsule gland (with proximal opaque and distal translucent portion) is two whorls; bursa long, opening near anterior end of straight section of pallial oviduct, extending back to spiral section. Spawn and development not observed; presence of capsule gland suggests pelagic egg capsule; type of protoconch indicates planktotrophic development. Radula (Figure 8A—-E): Relative radular length 0.87— 1.68. Rachidian: length/width 1.04—1.29; cusps extremely variable: one large rounded cusp with two pointed den- ticles on either side (Figure 8E), or five pointed cusps decreasing in size on either side of central cusp (Figure 8A), or central cusp may be short and blunt (Figure 8C); hood generally well developed, sometimes narrow. Lat- eral: four to six cusps, largest central cusp variable in shape and size (rounded, pointed or short and blunt); two to three small pointed cusps on inside and one to two on outside of main cusp. Inner marginal: three to four cusps, largest central cusp variable in shape and size (rounded, pointed or short and blunt); one to two small pointed cusps on inside and one on outside of main cusp. Outer marginal: two short broad cusps, either may be slightly larger, both bluntly rounded or pointed. Cusps of all teeth more elongate in smallest specimen examined (6.9 mm shell height), and three (not two) cusps on outer marginal. Material examined: 40 lots; one protoconch; nine penes; four sperm samples; three pallial oviducts; seven radulae. Habitat: At low levels on trunks and roots of mangroves, — fe Z a a > ial ob a O > 0 cS a DaG Reid (999 only rarely on leaves, up to 2.1 m above ground; common from seaward edge to landward fringe (personal obser- vation, Costa Rica); also on stones, logs, and grass among mangroves; muddy rocks on sheltered shores (Contreras & Cantera, 1978; Guerrini, 1990; personal observation). Remains at or below water level at high tide (Contreras & Cantera, 1978; Borjesson & Szelistowski, 1989). Range (Figure 6B): Specimens seen from El] Triunfo, El Salvador (13°34’'N; USNM; also Hernandez, 1979, from 13°14’N) to Paita, Peru (5°11’S; USNM). The southern limit in Peru requires confirmation; Pena (1970, 1971b) records the species from Puerto Pizarro (3°34’S). Never- theless, there is a relictual stand of mangroves at San Pedro (5°30’S), near Paita; although this species was not recorded in a survey of this site by Pefia & Vasquez (1985), occurrence there may be possible, perhaps only during El] Nifio events when warm equatorial water ex- tends this far south. Remarks: Of the three large, common L. (Littoraria) species found in the mangroves of western Central Amer- ica, L. varia occurs at the lowest levels on the trees (al- though there is considerable overlap among them), and is the only one that is regularly submerged by the rising tide (Pena, 1971a; Contreras & Cantera, 1978; Borjesson & Szelistowski, 1989; Blanco et al., 1995). In comparison with the higher-zoned L. variegata, the shell is thicker and the aperture more narrow. This makes it less suscep- tible to predators that forage during high tide, such as puffer fish, portunid and xanthid crabs, as shown by field tethering and laboratory predation trials (Borjesson & Szelistowski, 1989). The shells are nevertheless frequent- ly damaged during unsuccessful predation attempts, and most specimens bear the evidence in the form of one or more scars of repaired breakages. The intraspecific variation in the form of the radular tooth cusps is extreme. Radular variation has been de- scribed in other littorinid genera (e.g., Bembicium by Reid, 1988; Littorina by Reid, 1996), but L. varia is the most striking example. Although only seven radulae were examined, variation was evidently not correlated with sex, adult size, or locality. All specimens were from man- groves, so there was no obvious correlation with micro- habitat (cf. L. pintado pullata described earlier). There Page 35 may, however, be ontogenetic change in radular form; cusps of all teeth were relatively longest (although not as pointed as in one adult), and on the outer marginal more numerous, in the smallest (6.9 mm) specimen available; similar trends have been documented in Littorina (Reid, 1996; see also description of L. zebra). As in other studies of radular variation in littorinids, it is notable that tooth cusps vary together in the same way; in particular the major cusp on each of the five central teeth of each row are always similar in shape, suggesting a developmental constraint. There is a possibility of confusion among L. varia, L. zebra, and L. variegata, which are sympatric over much of their range (although L. variegata alone occurs in Mexico). Littoraria varia is easily recognized by the smaller apical angle of its more elongate shell, its pure white columella and interior of the aperture, sculpture of strong spiral ribs, the largest of which marks the angled periphery. Littoraria zebra is likewise thick-walled, but has a broader shell, distinctly angled at the shoulder, with bright coppery orange columella and inner apertural mar- gin, and striking broad brown stripes on the final whorl. Littoraria variegata is thinner in texture, has rounded whorls, columella edged with brown, and a variable shell pattern (usually of narrow oblique stripes, zigzags, or spi- ral lines). Penial shapes are diagnostic of each. The cop- ulatory bursa is similar in all three, but the spiral part of the pallial oviduct shows most numerous whorls in L. variegata and fewest in L. varia. Littoraria (Littoraria) zebra (Donovan, 1825) (Figures 2H, I, 3D, 4E, 6C, 7D—G, 8F—H) Turbo zebra Donovan, 1825: pl. 130; caption to pl. 131 (Panama; lectotype (here designated) Donovan, 1825: pl. 130). Littorina zebra—Morrison, 1946: 9. Keen, 1958: 282; fig. 179. Keen, 1971: 366; fig. 189. Guerrini, 1990: 14. Littorina (Littoraria) zebra—Rosewater, 1970: 423; pl. 326, figs 6, 7. Littoraria (Littoraria) zebra—Reid, 1986: 72; figs 41 (pe- nis), 18 (cladogram), 99g. Reid, 1989: 97; pl. 2, fig. 2a; figs 7e (penis), 10c (oviduct), 14e (radula). Littorina pulchra “Swainson” G. B. Sowerby, 1832: part 37; pl. 211, figs 2, 3 (no locality; types lost). Deshayes Figure 7 Shells of Littoraria varia (A-C), L. zebra (D-G), L. variegata (H—K) and possible hybrids (L—N). A. Guayaquil, Ecuador (BMNH 1996200, Cuming Colln; note repaired crab breakage on penultimate whorl). B, C. Panama (BMNH 1967698, Cuming Colln). D. Puntarenas, Costa Rica (BMNH 1996201). E. Panama (BMNH 1996202). E G. Panama (BMNH 1996203). H. El Salvador (BMNH 1996204; note eroded area where males attach in copulation position). I. Tumbez, Peru (BMNH 1996205). J. Lectotype of Littorina variegata Souleyet, in Eydoux & Souleyet, 1852 (MNHNP unregistered); La Puna, Guayaquil River, Ecuador. K. Tumbez, Peru (BMNH 1996206). L. Possible hybrid between L. varia and L. variegata; Puntarenas, Costa Rica (BMNH 1996154). M, N. Possible hybrids between L. zebra and L. varia; 2 miles west of Venado Beach, Veracruz, Panama (USNM 743087). Scale bar = 10 mm. < Wiese A ve Ta i o ay "Oo > 0) A= E DAG Reidy 1999 patch at inside of tentacle base and sometimes behind eye (Figure 11F); sides of foot mottled with black. Opercular ratio 0.48—0.50. Penis (Figure 11G—I) with wrinkled, bi- furcate base bearing two short papillae with blunt, puck- ered openings in place of glandular disc; two large, con- voluted, tubular glands visible by transparency, extending from papillate openings to very base of penis; filament large (40-60% total length), tapering; open sperm groove (also anterior vas deferens from prostate), extending to tip of filament; penis largely unpigmented, base faintly grey. Euspermatozoa 120 wm; paraspermatozoa (Figure 31) round, 9-13 wm diameter, surface minutely rough, contents indistinctly granular, no obvious rod-pieces. Pal- lial oviduct (Figure 11J) with spiral section of 2.5—3.5 whorls, capsule gland absent; bursa short, opening near anterior end of long straight section of pallial oviduct. Development ovoviviparous, with intracapsular metamor- phosis; in brooding females mantle cavity is solidly packed with embryos; one female (17.6 mm) contained 600 unencapsulated embryos, all at same stage of devel- opment, shell diameters 0.65—0.68 mm (Figure 10A—C). Gill leaflets similar in gross appearance to those in other members of genus; in a female of 15.0 mm, width of gills (from hypobranchial gland to osphradium) 4.0 mm, max- imum height of leaflets 0.5 mm. Radula (Figure 12F-H): Relative radular length 0.64— 0.74. Rachidian: length/width 1.11—1.47; central cusp largest, with mucronate point, one pointed cusp and a denticle on either side; hood well developed. Lateral: sev- en cusps, largest central cusp square; four short pointed or rounded cusps on inside and two small pointed cusps on outside of main cusp; anterior face of tooth is concave behind main cusp, so that inner and outer cusps are not aligned in same plane. Inner marginal: five to seven cusps, largest blunt; three to five smaller pointed cusps on inside and one on outside of main cusp. Outer mar- ginal: four to five cusps, outermost pointed, others bluntly rounded; neck and base of tooth unusually broad. Material examined: Type; 12 lots; two protoconchs; 10 embryonic shells; five penes; two sperm samples; three pallial oviducts; four radulae. Habitat: Leaves, branches and roots of Rhizophora at Page 45 landward edge of mangrove forests, up to 3.5 m above ground; apparently always scarce. Range (Figure 6F): The few available records are from Playa Tamarindo (10°19'N, BMNH), Puntarenas, and Quepos, all in Costa Rica, and Farfan River, Taboga Is- land and San José Island (8°15’N, USNM), all in Panama. Since the species is uncommon and occurs at high tidal levels, it may have been overlooked elsewhere. “‘Littorina scabra aberrans,” mentioned from the branches and fo- lage of mangroves in Colombia (Blanco et al., 1995), may refer to this species (or perhaps to L. rosewateri), but has been described as “‘very abundant’’ (Cantera et al., 1983). Remarks: This species is the only member of the genus, and one of only four species in the family (Reid & Geller, 1997), to show ovoviviparity with intracapsular meta- morphosis, so that crawling juveniles are released from the female. Twelve other species of Littoraria (the mem- bers of the subgenus Littorinopsis) are also ovovivipa- rous, but in these the larvae are released from the mantle cavity as planktotrophic early veligers with shells about 0.1 mm in diameter (Reid, 1986). Assuming that ovovi- viparity has arisen only once in Littoraria, the condition in L. aberrans has presumably been derived from that shown by Littorinopsis species (Reid, 1989). Elimination of the marine larval stage is presumably adaptive in this species which inhabits such high levels in the trees, often at the landward fringe of mangrove forests where contact with the tide is infrequent, making it effectively a terres- trial snail. This type of development might be expected to be advantageous for other Littoraria species which oc- cupy a similar high-level habitat elsewhere in the tropics; however, its absence may perhaps be explained by the consequent limitation of larval dispersal, which increases the likelihood of extinction (Reid & Geller, 1997). In im- mature or non-brooding females, ovoviviparous devel- opment is still recognizable because of the absence of capsule glands (and consequently small spiral section) in the pallial oviduct. Another anatomical peculiarity of this species is the unique structure of the penis. All other species of the genus possess a glandular pad or sucker, the penial glan- Figure 10 SEM details of shells of Littoraria aberrans (A—C, E-G) and L. rosewateri Reid, sp. nov. (D). A-C. Unencapsulated juveniles of L. aberrans from mantle cavity of brooding female; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218). D. Apex and protoconch of L. rosewateri Reid, sp. nov.; Golfito, Costa Rica (BMNH 1996217). E. Apex and protoconch of L. aberrans; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218). F Detail of sculpture and periostracum on last whorl of L. aberrans; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218). G. High-power detail of sculpture of larval shell of L. aberrans; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218). Scale bars: A-E = 200 pm; F = 400 pm; G = 50 pum. The Veliger, Vol. 42, No. I] e_— DAG Reids 1999 dular disc (although this is not clearly differentiated in the basal species L. pintado). The tubular glands of L. aberrans are superficially similar to the mamilliform pe- nial glands of other Littorininae, but their histology and staining reactions suggest that they are not homologous, and have probaby been formed by infolding of the penial glandular disc (Reid, 1989). The delicate shell of L. aberrans, with flared aperture, is similar to those of the members of the subgenus Lit- torinopsis that likewise inhabit high supratidal levels among the foliage of mangrove trees (Reid, 1986). A study of the behavior of this virtually terrestrial species would be interesting, for it is likely that it rarely comes into contact with the high tide (see Reid, 1984, for ac- count of the behavior of Australian Littoraria species). Consequently it avoids the powerful predators such as fish and crabs that forage at lower levels on the trees during high tide, and a thick protective shell is unneces- sary (Reid, 1992). The pigmentation of the shell is vari- able, and this is apparently a true polymorphism, since both unpigmented and variously patterned shells can be found together. However, none corresponding to the or- ange-pink morph of other polymorphic Littoraria species have yet been seen. Shell color polymorphism in Litto- raria is associated with a habitat among the foliage of mangrove trees (Reid, 1986), and the color polymorphism of both L. rosewateri and L. aberrans strengthens this correlation. It is believed to be maintained by visual se- lection against the varied background, and may be adap- tive in relation to predation (Cook, 1983, 1986, 1992; Hughes & Mather, 1986; Reid, 1987; Cook & Garbett, 1992). This species is one of the rarest littorinids in museum collections. This is not simply the result of its limited geographical distribution, for it appears always to be gen- uinely scarce in its mangrove habitat. This is in contrast to the high abundance attained by some other foliage- dwelling Littoraria species (Reid, 1985). The holotype was for long the only specimen known; this shell (Figure OF) is an aberrant example with exceptionally rounded Page 47 whorls, narrow spire, and fine spiral ribs. This partly ex- plains the early doubt about its generic allocation (Philip- pi, 1846a; Tryon, 1887; von Martens, 1900), for this shell does closely resemble some terrestrial prosobranchs of the genus Chondropoma Pfeiffer, 1847. Nevertheless, the protoconch, sculpture of the early whorls, columella, and dark pigmentation around the aperture are sufficient to identify it with others of the species, and intermediates with rounded last whorls (Figure 9H) connect it with the more typical form with keeled periphery and less marked sutures (Figure 9E, G, I). Among other Littoraria species in the Panamic prov- ince, confusion is only likely with L. rosewateri (see Re- marks on that species). In the absence of accurate locality information, confusion could easily arise with L. anguli- fera (Lamarck, 1822) from the Caribbean coast of Central America. That species reaches larger size (36 mm), is generally a broader, more solid shell with more rounded whorls, and the columella, while narrow, is excavated; most importantly the sculpture is finer (SO—90 ribs on last whorl, cf. 21-33 in L. aberrans), and the protoconch is of the planktotrophic type (0.35 mm diameter, about 3 whorls, sinusigera rib). Anatomically, the penis of L. an- gulifera has a bifurcate base bearing a glandular disc and a large filament (Reid, 1986: fig. 40), and in the female the mantle cavity contains numerous small eggs and em- bryos that are brooded only to the early veliger stage with shells about 0.1 mm in diameter. DISCUSSION The evolutionary history of the marine species of Central America has long been of particular interest, because of the opportunity for the study of processes of speciation and extinction that is provided by the Pliocene emergence of the Isthmus of Panama (review by Vermeij, 1993). Un- til recently, it was believed that the formation of the land bridge about 3 million years ago not only isolated the Eastern Pacific and Western Atlantic provinces, but also caused an episode of extinction that was most severe in Figure 11 Anatomy of Littoraria rosewateri Reid, sp. nov. (A-E, K, L) and L. aberrans (F—J). A. Head of L. rosewateri Reid, sp. nov.; Golfito, Costa Rica (BMNH 1996217). B—E. Penes of L. rosewateri Reid, sp. nov. B, D. Golfito, Costa Rica (BMNH 1996217; shell H of B = 5.1 mm; shell H of D = 4.7 mm). C. Penis of paratype of L. rosewateri Reid, sp. nov.; Topolobampo, Sinaloa, Mexico (BMNH 1996156). E. Rio Marina Lagoon, San José Island, Pearl Islands, Panama (USNM 588870). E Head of L. aberrans; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218). G—I. Penes of L. aberrans; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218; shell H of G = 12.0 mm; shell H of H = 9.1 mm; shell H of I = 9.7 mm; penis G is in a more contracted state then H and I; tubular glands visible by transparency are indicated by dotted outlines). J. Pallial oviduct of L. aberrans; Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218; shell H = 17.0 mm). K. Pallial oviduct of L. rosewateri Reid, sp. nov.; Golfito, Costa Rica (BMNH 1996217; shell H = 6.0 mm). L. Pallial oviduct of paratype of L. rosewateri Reid, sp. nov.; Topolobampo, Sinaloa, Mexico (BMNH 1996156; shell H = 9.3 mm). Shading conven- tions as in Figure 4. The Veliger, Vol. 42, No. 1 DaGe Reid: 1999 the Atlantic, resulting in an impoverished Caribbean fau- na (e.g., Vermeij & Petuch, 1986). It now appears that the extinctions began later, about 2.4 million years ago, perhaps as a result of changing patterns of upwelling and productivity (Jackson et al., 1993). Furthermore, extinc- tions were balanced by speciation and immigration, so that the overall diversity of Caribbean mollusks has not decreased since the Pliocene, and is not lower than that of the tropical Eastern Pacific (Allmon et al., 1993). The modern differences between the faunas of these two prov- inces are therefore the result not only of differential pat- terns of extinction, but also of origination. Unfortunately, neither the phylogenetic relationships nor fossil history of Littoraria are yet sufficiently well known to permit more than speculation on its evolutionary history in Cen- tral America. The genus Littoraria has a pantropical distribution, and the oldest fossils occur in the Lower Eocene of France (Reid, 1989). Of the 36 Recent species, 25 occur in the Indo-West Pacific province. This compares with the six species reported here from the Eastern Pacific, five from the Western Atlantic and two from the Eastern Atlantic (Reid, 1986). The Western Atlantic species are L. angu- lifera (Lamarck, 1822), L. flava (King & Broderip, 1832), L. irrorata (Say, 1822), L. nebulosa (Lamarck, 1822), and L. tessellata (Philippi, 1847). On the basis of shell resem- blance, Rosewater (1963) suggested the following pairs of “species analogues”’ on either side of the Isthmus of Panama: L. varia and L. irrorata, L. pintado and L. tes- sellata, L. fasciata and L. angulifera. However, shells are a poor guide to affinity among littorinids, and (with the possible exception of the first) none of these pairs is sup- ported by anatomical evidence. Later, using both shell and radular characters, Rosewater (1980b) classified L. scabra (L.) as a single pantropical species, with subspecies L. s. scabra in the Indo-West Pacific, L. s. angulifera in the Atlantic, and L. s. aberrans in the Eastern Pacific, but again anatomical evidence has contradicted the implied relationships (Reid, 1986). In an early attempt to use bio- chemical characters to define ‘‘geminate species pairs”’ on either side of Panama, Jones (1972) analyzed allozyme frequences and myoglobin banding patterns in 12 littorin- ids, but failed even to separate the generic groupings now recognized as Littoraria and Nodilittorina, much less to identify consistent species-pairs. Reid (1986) used a cla- Page 49 distic analysis of anatomical characters to define basal groups within Littoraria, and relied on subjective assess- ment of shell and penial form to suggest terminal group- ings; on this basis the only close relationships of species across the Isthmus were among L. varia, L. zebra, L. va- riegata, and L. irrorata. The present redescription of the Littoraria species of the Eastern Pacific has partly supported this earlier study. Littoraria pintado has no known sister-species among liv- ing members of Littoraria, and its subspecies, L. p. pul- lata is probably a relatively recent, Pleistocene, arrival in the Eastern Pacific from the west. The three species L. varia, L. zebra, and L. variegata are believed to form a clade, sharing likely synapomorphies of similar oviducts (tightly wound spiral; anterior bursa) and radulae (narrow posterior base of rachidian; only two to three cusps on outer marginal). This close relationship is also supported by the possible hybrids between them, discussed earlier. If they are indeed recently diverged from a common an- cestor, their diversity of penial form is noteworthy, sug- gesting that the size of the glandular disc and degree of bifurcation of the base are readily modified, and might be species-recognition characters. Elsewhere in the genus, the same combination of radular and oviduct characters is found only in the Western Atlantic species L. irrorata (also exhibiting a non-bifurcate penis similar to that of L. variegata), which may therefore belong to the same clade. These four American species were linked by Reid (1986) with the Indo-West Pacific L. vespacea Reid, 1986, but reexamination of its radula has shown that it does not share the same characters. The new species L. rosewateri shares the synapomorphy of the closed penial vas deferens with the two Caribbean species L. flava and L. tessellata (these are likely sister-species, sharing a uniquely elongated penial filament, although the base is bifurcate only in the latter). Radular characters are also similar among these three, as is the overall form of the pallial oviduct. However, the bursa opens in an anterior position in L. rosewateri, posteriorly in L. flava, and is variable in position in L. tessellata; this character may not be phylogenetically informative in these species with a relatively short straight section of the pallial oviduct (as also suggested in Littorina, Reid, 1996: 349). The penial glands and intracapsular metamorphosis of L. aberrans are unique in the genus; its ovoviviparity is a synapo- Figure 12 Radulae of Littoraria rosewateri Reid, sp. nov. (A-E) and L. aberrans (F—H). A-C. Radula of paratype of L. rosewateri Reid, sp. nov.; Topolobampo, Sinaloa, Mexico (BMNH 1996156; three views of radula, flat, at 45° and at 45° from side; shell H = 9.3 mm). D, E. Golfito, Costa Rica (BMNH 1996217; two views of radula, at 45° and flat; shell H = 6.0 mm; note aberrant inner marginal teeth on right side, with only two cusps). F—H. Punta Morales, Golfo de Nicoya, Costa Rica (BMNH 1996218; three views of radula, at 45°, flat and at 45° from side; shell H = 17.6 mm). Scale bars = 50 pm. Page 50 morphy shared with the subgenus Littorinopsis. No char- acters have yet been found which might indicate its re- lationships more precisely, and there is no reason to sup- pose that it is the sister-species of the Atlantic L. (Litto- rinopsis) angulifera. A reanalysis of the relationships of Littoraria species, using additional information on radular characters, is in progress. The trans-Panamanian relationships discussed above are mostly different from those predicted from shell char- acters (Rosewater, 1963, 1980b). The lack of phyloge- netically significant shell characters suggests that the fos- sil record may not be very helpful in reconstructing the evolutionary history of Littoraria. The pre-Pleistocene re- cord of this genus in Central America is meager: two specimens of “‘Littorina varia’ from the Pliocene of Cal- ifornia (Hanna, 1926), three specimens of ‘‘Littorina an- gulifera’’ from the Miocene of Panama and Costa Rica (Woodring, 1957), and numerous records of L. irrorata from the Upper Miocene and Pliocene of Florida, North and South Carolina (e.g., Smith, 1936; Bequaert, 1943). These will be examined in future work. During the emergence of the Panama land bridge, sep- aration of Pacific and Atlantic populations of all marine species did not take place simultaneously. Those able to tolerate inshore conditions appear to have remained in genetic contact until the final stages of the imposition of the barrier (Knowlton et al., 1993), which is estimated to have occurred about 3.2 to 2.5 Ma (Coates et al., 1992). It is likely that the Littoraria species of mangrove envi- ronments, tolerant of turbidity and reduced salinity, were among the last to be separated. The absence of any ob- vious pairs of sister-species on either side of the modern isthmus is therefore surprising, and suggests that subse- quent speciation, extinction, or migration may have ob- scured the expected pattern. It is also possible that there was already some differentiation of Pacific and Atlantic faunas before the isthmus appeared (Vermeij, 1993). For comparison, populations of Littorina squalida Broderip & Sowerby, 1829, in the Northern Pacific and Northern Atlantic were isolated about 4 to 2.4 Ma (as a result of climatic cooling following opening of the Bering Strait), and the resulting pair of sister-species, L. squalida and L. littorea (Linnaeus), are clearly recognizable on morpho- logical grounds (Reid, 1996); this pair of planktotrophic developers has not undergone further speciation during this time. In contrast, the trans-Panamanian relationships discussed above suggest that some speciation may have occurred over a similar time scale, in these likewise planktotrophic Littoraria species (e.g., L. varia, L. zebra, L. variegata in the Eastern Pacific; L. flava and L. tes- sellata in the Caribbean). However, this remains to be investigated by further phylogenetic and paleontological work. The geographical distributions of the Eastern Pacific Littoraria species clearly show limitation by habitat. Lit- toraria pintado pullata is restricted to oceanic high is- The Veliger, Vol. 42, No. 1 lands (although current patterns may have prevented its colonization of the Galapagos Islands); on the American mainland it occurs commonly only at the extremity of Baja California. The remaining five species are all found predominantly among mangrove vegetation. The southern limits of three of these (L. varia, L. variegata, L. rose- wateri) coincide with the southern limit of mangroves, in northern Peru, whereas that of L. zebra is a little farther north. Northern limits are less well established owing to a paucity of information about El Salvador and Guate- mala, but the long stretch of coastline without coastal mangroves between southern Mexico and the Gulf of Cal- ifornia appears to present a barrier to L. varia and L. zebra. Only L. variegata and L. rosewateri occur to the north of this barrier, thus showing markedly disjuct rang- es. Although egg capsules have not yet been described, all these five species have planktotrophic development (indicated by protoconch and capsule glands), with the corresponding potential for wide larval dispersal. Rafting might also be a common means of dispersal in the man- grove-associated species. Only the sixth species, L. aber- rans, has non-planktotrophic development, which might partly explain its restricted distribution. Bequaert (1943) reported that the Caribbean species L. angulifera had reached the Pacific coast of Panama through the Panama Canal. However, this has not been confirmed by the extensive museum collections from this area examined during the present study. The record may perhaps have arisen from confusion with L. aberrans. Among the mangrove-associated Littoraria species of the Indo-West Pacific, interspecific trends in shell archi- tecture and coloration, at successive levels on the trees, have been explained as adaptive responses to gradients in crushing and visual predation; species zoned at lower lev- els are thick-shelled and monomorphic, those found at higher levels are thinner-shelled, and those inhabiting the foliage are thinnest, and often color polymorphic (see In- troduction). As discussed in the Remarks on each species, these trends are also apparent in the Eastern Pacific, al- though L. zebra is somewhat anomalous. The ‘‘hooded”’ type of rachidian tooth is found in most Littoraria species, including all those known to occur on trees, driftwood and marsh plants, which led to the sug- gestion that it might be adaptive for grazing on such sub- strates (Rosewater, 1980a; Reid, 1986, 1989). However, the new discovery that a small ‘‘hood’’-like structure is also be present in some examples of the basal, rock- dwelling, species L. pintado suggests that it might per- haps be a synapomorphy of the genus, lost in a few spe- cies. Little is known about the diet of Littoraria species, but a mangrove-associated species (L. angulifera) and a salt-marsh species (L. irrorata) both include a significant component of fungal material in their diets (Kohlmeyer & Bebout, 1986; Newell & Barlocher, 1993; Barlocher & Newell, 1994). The intraspecific variation in form of the tooth cusps in L. varia (and to a lesser extent in L. va- D. G. Reid, 1999 riegata and L. pintado pullata) is the most remarkable example in the family. Such extreme variation has not previously been found in Littoraria species (Reid, 1986), but has been reported in other littorinid genera (Reid, 1988, 1996; Padilla, 1998). ACKNOWLEDGMENTS Field work in Costa Rica was undertaken during the ten- ure of a postdoctoral fellowship at the National Museum of Natural History, Smithsonian Institution. In Costa Rica, I thank the Consejo Nacional de Investigaciones Cientificas y Technoldgicas for permission to use the La- boratorio de Investigaciones Marinas at Punta Morales, and Prof. C. Villalobos for laboratory facilities at the Uni- versity of Costa Rica. I gratefully acknowledge the help of Dr W. A. Szelistowski in the field at Punta Morales, and thank the Organization for Tropical Studies for as- sistance with transport in Costa Rica. I am most grateful to S. A. Jeffcoat for help during fieldwork in Mexico. Loans of specimens were kindly arranged by R. Germon (USNM), L. T. Groves (LACM), V. Héros (MNHNP), and E. J. Kools (CAS), and Dr. A. M. Ellison generously pro- vided additional material from Costa Rica. For their as- sistance with electron microscopy I thank C. Jones and L. Jones, and for expert photography H. Taylor and N. Hayes (all BMNH). M. A. Rabi (Lima) very kindly as- sisted with the South American literature. LITERATURE CITED ABBOTT, R. T. 1974. American Seashells. 2nd ed. Van Nostrand Reinhold: New York. 663 pp. ApaMs, C. B. 1852. Catalogue of shells collected at Panama, with notes on synonymy, station, and habitat. Annals of the Lyceum of Natural History of New York 5:229—-549. ADAMS, H. & A. ADAMS. 1853-1854. The Genera of Recent Mol- lusca; Arranged According to their Organization. Volume 1. John van Voorst: London. xl + 484 pp. ALAMO, V. & V. VALDIVIESO. 1987. Lista sistematica de moluscos marinos del Péru. Boletin del Instituto del Mar, Volumen Extraordinario. 205 pp. ALLMOoN, W. D., G. ROSENBERG, R. W. PorTELL & K. S. SCHIN- DLER. 1993. Diversity of Atlantic coastal plain mollusks since the Pliocene. Science 260:1626—-1629. BARLOCHER, F & S. Y. NEWELL. 1994. Growth of the salt marsh periwinkle Littoraria irrorata on fungal and cordgrass diets. Marine Biology 118:109—-114. Bartscu, P. & H. A. REHDER. 1939. Mollusks collected on the presidential cruise of 1938. Smithsonian Miscellaneous Col- lections 98(10):1—18. BEQUAERT, J. C. 1943. The genus Littorina in the Western At- lantic. Johnsonia 1(7):1—27. Berry, A. J. & E. CHEw. 1973. Reproductive systems and cyclic release of eggs in Littorina melanostoma from Malayan mangrove swamps (Mollusca: Gastropoda). Journal of Zo- ology, London 171:333-344. BLANco, FE, A. BEJARANO, J. M. Diaz, E ZAPATA & J. R. CAN- TERA. 1995. Distribuci6n vertical del genero Littorina en manglares de la Bahia de Buenaventura. Pp. 291—299 in J. D. Restrepo & J. R. Cantera (eds.), Delta del Rio San Juan, Page 51 Bahias de Malaga y Buenaventura, Pacifico Colombiano. Volume 1. Cali. BorJesson, D. L. & W. A. SZELISTOWSKI. 1989. Shell selection, utilization and predation in the hermit crab Clibanarius pan- amensis Stimpson in a tropical mangrove estuary. Journal of Experimental Marine Biology and Ecology 133:213-228. BOULDING, E. G., J. BUCKLAND-Nicks & K. L. vAN ALSTYNE. 1993. Morphological and allozyme variation in Littorina sit- kana and related Littorina species from the northeastern Pa- cific. The Veliger 36:43-68. Burcu, J. Q. 1945. Family Littorinidae. Minutes of the Concho- logical Club of Southern California 55:9-13, 36, 37. CANTERA, J. R., PR. M. ARNAUD & B. A. THOMASSIN. 1983. Bio- geographic and ecological remarks on molluscan distribution in mangrove biotopes. 1. Gastropods. Journal of Molluscan Studies, Supplement 12A:10—26. CaNnrTERA, J. R. & R. CONTRERAS. 1978. Informe preliminar sobre el potencial malacologico aprovechable en el Pacifico col- ombiano. Pp. 440—474 in M. Vegas Velez & R. Rojas Bel- tram (eds.), Memorias del Primer Seminario sobre el Océano Pacifico Sudamericano. Cali, Sept. 1-5, 1976. Volume 2. Universidad del Valle: Cali, Colombia. CARPENTER, P. P. 1857a. Report on the present state of our knowl- edge with regard to the Mollusca of the West Coast of North America. Report of the British Association for the Advance- ment of Science 1856:159—368. CARPENTER, P. P. 1857b. Catalogue of the Collection of Mazatlan Shells in the British Museum: Collected by Frederick Rei- gen. British Museum: London. xvi + 552 pp. CARPENTER, P. P. 1864a. Diagnoses of new forms of mollusks collected at Cape St. Lucas by Mr. J. Xantus. Annals and Magazine of Natural History, Series 3 13:311-315. CARPENTER, P. P. 1864b. Supplementary report on the present state of our knowledge with regard to the Mollusca of the west coast of North America. Report of the British Associ- ation for the Advancement of Science 1863:517—684. Coates, A. G., J. B. C. Jackson, L. S. Coins, T. M. CRONIN, H. J. Dowsett, L. M. ByBett, P. JUNG & J. A. OBANDO. 1992. Closure of the Isthmus of Panama: the near-shore ma- rine record of Costa Rica and western Panama. Bulletin of the Geological Society of America 104:814—828. CONTRERAS, R. & J. R. CANTERA. 1978. Notas sobre la ecologia de los moluscos asociados al ecosistema manglar-estero en la Costa del Pacifico colombiano. Pp. 709-747 in M. Vegas Velez & R. Rojas Beltram (eds.), Memorias del Primer Sem- inario sobre el Océano Pacifico Sudamericano. Cali, Sept. 1-5, 1976. Volume 2. Universidad del Valle: Cali, Colom- bia. Cook, L. M. 1983. Polymorphism in a mangrove snail in Papua New Guinea. Biological Journal of the Linnean Society 20: 167-173. Cook, L. M. 1986. Site selection in a polymorphic mangrove snail. Biological Journal of the Linnean Society 29:101—-113. Cook, L. M. 1990. Systematic effects on morph frequency in the polymorphic mangrove snail Littoraria pallescens. Heredity 65:423—427. Cook, L. M. 1992. The neutral assumption and maintenance of colour morph frequency in mangrove snails. Heredity 69: 184-189. Cook, L. M., J. D. CURREY & V. H. Sarsam. 1985. Differences in morphology in relation to microhabitat in littorinid spe- cies from a mangrove in Papua New Guinea (Mollusca: Gas- tropoda). Journal of Zoology, London 206:297—310. Cook, L. M. & S. D. GARBETT. 1992. Selection in the polymor- Page 52 The Veliger, Vol. 42, No. 1 phic mangrove snail Littoraria pallescens. Pp. 247-253 in J. Grahame, P. J. Mill & D. G. Reid (eds.), Proceedings of the Third International Symposium on Littorinid Biology. Malacological Society of London: London. Cruz, R. A. 1989. Caracteres generales, edad y crecimiento de Littorina fasciata (Mollusca: Gastropoda). Brenesia 27:13— 22 DaL_, W. H. 1909. Report on a collection of shells from Peru, with a summary of the littoral marine Mollusca of the Pe- ruvian zoological province. Proceedings of the United States National Museum 37:147—294. DesHayeEs, G. P. & H. MILNE EDwarbs. 1843. Histoire Naturelle des Animaux sans Vertébres ... par J. B. P. A. de Lamarck. Edition 2. Volume 9. J.-B. Bailliere: Paris & London. Donovan, E. 1825. Turbo zebra. The Naturalist’s Repository 4: pl. 130, caption to pl. 131. EMERSON, W. K. 1978. Molluscs with Indo-Pacific faunal affini- ties in the Eastern Pacific Ocean. The Nautilus 92:91—96. EMERSON, W. K. 1991. First records for Cymatium mundum (Gould) in the Eastern Pacific Ocean, with comments on the zoogeography of the tropical trans-Pacific tonnacean and non-tonnacean prosobranch gastropods with Indo-Pacific faunal affinities in West American waters. The Nautilus 105: 62-80. FINET, Y. 1991. The marine mollusks of the Galapagos Islands. Pp. 253-280 in M. J. James (ed.), Galapagos Marine Inver- tebrates: Taxonomy, Biogeography, and Evolution in Dar- win’s Islands. Plenum Press: New York. FINeT, Y. 1994. The Marine Mollusks of the Galapagos Islands: A Documented Faunal List. Muséum d’Histoire Naturelle: Geneva. 180 pp. GALLAGHER, S. B. & G. K. REID. 1974. Reproductive behaviour and early development in Littorina scabra angulifera and Littorina irrorata (Gastropoda: Prosobranchia) in the Tampa Bay region of Florida. Malacological Review 7:105—125. Gray, J. E. 1839. Molluscous animals, and their shells. Pp. 101— 155 in F W. Beechey (ed.) The Zoology of Captain Beech- ey’s Voyage. H. G. Bohn: London. GRIFFITH, E. & E. PIDGEON. 1834. The Mollusca and Radiata. Pp. 1—601 in E. Griffith (ed.), The Animal Kingdom by the Bar- on Cuvier. Volume 12:1—601. Whittaker & Co.: London. GUERRINI, R. M. 1990. Occurrence of Littorina zebra (Donovan, 1825) (Gastropoda: Littorinidae) at Bahia de Cardquez, Ec- uador. Siratus 1(4):14. Hanna, G. D. 1926. Paleontology of Coyote Mountain, Imperial County, California. Proceedings of the California Academy of Sciences. Series 4 14:427—503. HEALY, J. M. & B. G. M. Jamieson. 1993. Euspermatozoa, par- aspermatozoa and spermatozeugmata of Littoraria (Palus- torina) articulata (Prosobranchia: Caenogastropoda) with species reference to the pseudotrich. Acta Zoologica 74: 321-330. HERNANDEZ, M. A. 1979. Range extensions of mollusk species found on the tropical coast of El Salvador. The Veliger 22: 204-205. HuGues, J. M. & P. B. MATHER. 1986. Evidence for predation as a factor in determining shell color frequencies in a mangrove snail Littorina sp. (Prosobranchia: Littorinidae). Evolution 40:68-77. JACKSON, J. B. C., P. JUNG, A. G. Coates & L. S. CoLuins. 1993. Diversity and extinction of tropical American mollusks and emergence of the Isthmus of Panama. Science 260:1624— 1626. Jones, M. L. 1972. Comparison of electrophoretic patterns of littorine snails of Panama: an attempt to define geminate species. XVIII Congrés International de Zoologie. Theme 3: 1-10. KEEN, A. M. 1958. Sea Shells of Tropical West America. Stan- ford University Press: Stanford. xi + 624. KEEN, A. M. 1971. Sea Shells of Tropical West America. 2nd ed. Stanford University Press: Palo Alto, California. xiv + 1064 pp. Keep, J. & J. L. Batty. 1935. West Coast Shells. Stanford Uni- versity Press: Stanford. xi + 350. Kina, P. P. 1839. Narrative of the Surveying Voyages of His Majesty’s Ships Adventure and Beagle, Between the Years 1826 and 1836. H. Colburn: London. 597 pp. Kinc, P. PR. & W. J. BRODERIP. 1832. Description of the Cirrhi- peda, Conchifera and Mollusca, in a collection formed by the officers of H.M.S. Adventure and Beagle employed be- tween the years 1826 and 1830 in surveying the southern coasts of South America, including the Straits of Magal- haens and the coast of Tierra del Fuego. Zoological Journal 5:332-349. KNOWLTON, N., L. A. WEIGT, L. ANIBAL SOLORZANO, D. K. MILLS & E. BERMINGHAM. 1993. Divergence in proteins, mitochon- drial DNA, and reproductive compatibility across the Isth- mus of Panama. Science 260:1629—1632. KOHLMEYER, J. & B. BEBouT. 1986. On the occurrence of marine fungi in the diet of Littorina angulifera and observations on the behaviour of the periwinkle. Marine Ecology 7:333-343. MaRTENS, E. von. 1890-1901. Land and freshwater Mollusca. Pp. xxvii + 706 in EF D. Godman & O. Salvin (eds.), Biol- ogia Centrali-Americana. R. H. Porter: London. MCNALLY, G. J., W. C. Patzert, A. D. KIRWAN & A. C. VAs- TANO. 1983. The near-surface circulation of the North Pacific using satellite tracked drifting buoys. Journal of Geophysical Research 88:7507—-7518. MILLE-PAGAZA, S. R., A. PEREZ-CHI & O. HOLGUIN-QUINONES. 1994. Malacologic benthic fauna of the Socorro Island lit- toral, Revillagigedo, Mexico. Ciencias Marinas 20:467—486. Morcu, O. A. L. 1860. Beitrége zur Molluskenfauna Central- Amerika’s. Malakozoologische Blatter 7:66—106. Morrison, J. P. E. 1946. The nonmarine mollusks of San Jose Islands, with notes on those of Pedro Gonzalez Island, Pearl Islands, Panama. Smithsonian Miscellaneous Collections 106(6):1—49. NEWELL, S. Y. & E BARLOCHER. 1993. Removal of fungal and total organic matter from decaying cordgrass leaves by shredder snails. Journal of Experimental Marine Biology and Ecology 171:39—49. PaDILLA. D. K. 1998. Inducible phenotype plasticity of the radula in Lacuna (Gastropoda: Littorinidae). The Veliger 41:201— 204. PALMER, K. VAN W. 1958. Type specimens of marine Mollusca described by P. P. Carpenter from the West Coast (San Diego to British Columbia). Geological Society of America Mem- oir 76:1-376. PALMER, K. VAN W. 1963. Type specimens of marine Mollusca described by P. P. Carpenter from the west coast of Mexico and Panama. Bulletins of American Paleontology 46:285— 408. PENA, G. M. 1970. Zonas de distribucion de los gasteropédos marinos del Péru. Anales Cientificos 8:154—170. PENA, G. M. 1971a. Biocenosis de los manglares Peruanos. An- ales Cientificos 9:38—45. PENA, G. M. 1971b. Descripci6n de los gasterépodos de los manglares del Péru. Anales Cientificos 9:46—55. DaGrReid. 11999 Pena, G. M. & P. G. VASQUEZ. 1985. Un relicto de manglar en San Pedro (Piura): estudio preliminar. Boletin de Lima 7: 27-32. Puiuippl, R. A. 1846a. Descriptions of a new species of Trochus and of eighteen new species of Littorina, in the collection of H. Cuming, Esq. Proceedings of the Zoological Society of London 1845:138-143. PuHiuipr!, R. A. 1846b, 1847. Abbildungen und Beschreibungen neuer oder wenig gekannter Conchylien. 2:99—105, Litorina pl. 1 (1846b); 221-227, Litorina pl. 5(1847); 3:9-18, Lito- rina pl. 6(1847). Theodor Fischer: Cassel. Pitssry, H. A. & H. N. Lowe. 1932. West Mexican and Central American mollusks collected by H. N. Lowe, 1929-31. Pro- ceedings of the Academy of Natural Sciences of Philadel- phia 84:33-144. REEVE, L. A. 1857-1858. Monograph of the genus Littorina. Pls 1-16 (1857), 17-18 (1858), in L. A. Reeve (ed.), Conchol- ogia Iconica. Lovell Reeve: London. REHDER, H. A. 1986. Joseph Rosewater (1928-1985) a tribute and bibliography. The Nautilus 100:9—13. Rep, D. G. 1984. The Systematics and Ecology of the Man- grove-dwelling Littoraria species (Gastropoda: Littorinidae) in the Indo-Pacific. PhD dissertation, James Cook University of North Queensland: Townsville. xxvi + 797 pp. Rep, D. G. 1985. Habitat and zonation patterns of Littoraria species (Gastropoda: Littorinidae) in Indo-Pacific mangrove forests. Biological Journal of the Linnean Society 26:39-68. Rep, D. G. 1986. The Littorinid Molluscs of Mangrove Forests in the Indo-Pacific Region: the Genus Littoraria. British Mu- seum (Natural History): London. xv + 228. Rep, D. G. 1987. Natural selection for apostasy and crypsis acting on the shell colour polymorphism of a mangrove snail, Littoraria filosa (Sowerby) (Gastropoda: Littorinidae). Biological Journal of the Linnean Society 30:1—24. REID, D. G. 1988. The genera Bembicium and Risellopsis (Gas- tropoda: Littorinidae) in Australia and New Zealand. Re- cords of the Australian Museum 40:91—150. REID, D. G. 1989. The comparative morphology, phylogeny and evolution of the gastropod family Littorinidae. Philosophical Transactions of the Royal Society of London, Series B 324: 1-110. REID, D. G. 1992. Predation by crabs on Littoraria species (Lit- torinidae) in a Queensland mangrove forest. Pp. 141-151 in J. Grahame, P. J. Mill & D. G. Reid (eds.), Proceedings of the Third International Symposium on Littorinid Biology. Malacological Society of London: London. REID, D. G. 1996. Systematics and Evolution of Littorina. Ray Society: London. x + 463 pp. REID, D. G. & J. B. GELLER. 1997. A new ovoviviparous species of Tectarius (Gastropoda: Littorinidae) from the tropical Pa- cific, with a molecular phylogeny of the genus. Journal of Molluscan Studies 63:207—233. RICHMOND, R. H. 1990. The effects of the El Nino/Southern Os- cillation on the dispersal of corals and other marine organ- isms. Pp. 127-140 in P. W. Glynn (ed.), Global Ecological Consequences of the 1982—83 El Nifo—Southern Oscillation. Elsevier: Amsterdam. ROSEWATER, J. 1963. Problems of species analogues in world Littorinidae. Report of the American Malacological Union and AMU Pacific Division 30:5-—6. ROSEWATER, J. 1970. The family Littorinidae in the Indo-Pacific. Part I. The subfamily Littorininae. Indo-Pacific Mollusca 2: 417-506. Page 53 ROSEWATER, J. 1980a. A close look at Littorina radulae. Bulletin of the American Malacological Union 1979:5-8. ROSEWATER, J. 1980b. Subspecies of the gastropod Littorina sca- bra. The Nautilus 94:158-162. SCHELTEMA, R. S. 1988. Initial evidence for the transport of te- leplanic larvae of benthic invertebrates across the East Pa- cific barrier. Biological Bulletin of the Marine Biological Laboratory, Woods Hole 174:145-152. SmiTH, M. 1936. New Tertiary shells from Florida. The Nautilus 49:135-139. SOULEYET, F L. A. 1852. Mollusques. Pp. 7-633 in J. FE T. Eydoux & F L. A. Souleyet (eds.), Voyage autour du Monde Exécuté pendant les Années 1836 et 1837 sur la Corvette La Bonite. Zoologie. Volume 2. Paris. SOWERBY, J. 1832. The Genera of Recent and Fossil Shells. Part 37. London. STUARDO, J. & M. VILLARROEL. 1976. Aspectos ecoldgicos y dis- tribucion de los moluscos en las lagunas costeras de Guer- rero, México. Anales del Centro de Ciencias del Mar y Lim- nologia 3:65—92. STRUHSAKER, J. W. 1966. Breeding, spawning, spawning peri- odicity and early development in the Hawaiian Littorina: L. pintado (Wood); L. picta Philippi and L. scabra (Linné). Proceedings of the Malacological Society of London 37: 137-166. STRUHSAKER, J. W. & J. D. CostLow. 1968. Larval development of Littorina picta (Prosobranchia, Mesogastropoda) reared in the laboratory. Proceedings of the Malacological Society of London 38:153—160. TRYON, G. W. 1887. Manual of Conchology. Volume 9. Phila- delphia. 488 pp. VERMEJI, G. J. 1987. The dispersal barrier in the tropical Pacific: implications for molluscan speciation and extinction. Evo- lution 41:1046—-1058. VERMEU, G. J. 1990. An ecological crisis in an evolutionary con- text: El Nino in the Eastern Pacific. Pp. SO5—517 in P. W. Glynn (ed.), Global Ecological Consequences of the 1982— 83 El Nino-Southern Oscillation. Elsevier: Amsterdam. VERMEU, G. J. 1993. The biological history of a seaway. Science 260: 1603-1604. VERMEU, G. J. & Petucu, E. J. 1986. Differential extinction in tropical American molluscs: endemism, architecture, and the Panama land bridge. Malacologia 27:29—41. Warwick, T., A. J. KNIGHT & R. D. Warp. 1990. Hybridisation in the Littorina saxatilis species complex (Prosobranchia: Mollusca). Hydrobiologia 193:109-116. WEINKAUFF, H. C. 1878-1882. Die Gattung Litorina. Angefaugen von Dr. Kiister, durchgesehen, erganzt und vollendet von H. C. Weinkauff. Parts 269 (1878, pp. 25—40), 315 (1882, pls 6-11, pp. 41-72), 318 (1882, pls 12-14, pp. 73-114) in H. C. Kiister, W. Kobelt & H. C. Weinkauff (eds.), Systema- tisches Conchylien-Cabinet von Martini und Chemnitz. Bauer & Raspe: Nurnberg. Woop, W. 1828. Supplement to the Index Testaceologicus; or a Catalogue of Shells, British and Foreign. London. WOOoDRING, W. P. 1957. Geology and paleontology of Canal Zone and adjoining parts of Panama. Geology and description of Tertiary mollusks (gastropods: Trochidae to Turritellidae). United States Geological Survey Professional Paper 306-A: 1-145. WyrTKI, K. 1965. Surface currents of the Eastern tropical Pacific Ocean. Bulletin of the Inter-American Tropical Tuna Com- mission 9:27 1—304. ZILCH, A. 1954. Moluscos de los manglares de El Salvador. Co- municaciones del Instituto Tropical de Investigaciones Cien- tificas de la Universidad de El Salvador 3:77—87. THE VELIGER © CMS, Inc., 1999 The Veliger 42(1):54—61 (January 4, 1999) Genetic and Environmental Control of Growth and Reproduction of Phacosoma japonicum (Bivalvia: Veneridae) SHIN’ ICHI SATO! Geological Institute, University of Tokyo; 7-3-1 Hongo, Tokyo 113, Japan Abstract. The venerid bivalve Phacosoma japonicum (Reeve, 1850) shows north-south gradients in the annual shell growth and sexual maturity patterns among the Japanese populations. However, one southern population from Ariake Bay, Kyushu, does not fit with this general trend, and life-history traits of this population are comparable to those of the northern population in Hakodate Bay, Hokkaido. In this study, to understand the genetic and environmental factors responsible for the life-history traits of this species, transplant experiments and analyses of monthly shell growth and reproductive cycles were conducted on the populations from Ariake Bay and its neighboring regions. A total of 128 living individuals were transplanted from the population in Tokyo Bay (central Japan) and those in Ariake and Kagoshima Bays (southwestern Japan) to Aburatsubo Cove, Sagami Bay (central Japan). Follow-up studies of the transplanted individuals for 3 years revealed that both annual and seasonal shell growth patterns of most individuals did not differ from those of animals in the habitats of origin. This fact suggests that the mode of shell growth of this species is not only controlled by environmental factors but also has some genetic background. Analysis of the seasonal patterns of growth and reproductive cycles revealed that shell growth and gonad development were active from winter to early spring in individuals from Ariake Bay, whereas growth and reproduction occurred from late spring to summer in indi- viduals from Tokyo and Kagoshima Bays. Seasonal changes in water temperature and salinity and also population density were similar among Ariake Bay and its neighboring regions. However, phytoplankton becomes most abundant in winter in Ariake Bay, in contrast to most other bays of central and southern Japan, including Tokyo and Kagoshima Bays, where phytoplankton flourishes in summer. These facts suggest that the growing seasons for both shell growth and reproduction of this species are strongly influenced by the annual pattern of food availability, and the geographical variations of annual shell growth and sexual maturity patterns can be explained by the difference in mean water tem- perature during the growing season among them. INTRODUCTION may not be genetically programmed, but instead may be a direct consequence of environmental controls. In this case, traits are not inherited by the descendants, so they never cause life-history evolution. Therefore, in order to understand the life-history evolution of organisms, it is Many bivalve species exhibit geographic variation in life- history patterns (e.g., Taylor, 1959; Ansell, 1968; Gilbert, 1973; Appeldoorn, 1983; Tanabe & Oba, 1988; Sato, 1994), and a number of studies have discussed controlling environmental factors such as temperature (Green & Hob- son, 1970; Noda et al., 1995), salinity (McLusky & Allan, 1976), food availability (Beukema et al., 1977; Thompson & Nichols, 1988; Irlandi & Peterson, 1991; Nakaoka, 1992), and density effects (Rae, 1979; Broom, 1982; Pe- terson, 1982). Recently, there have been theoretical stud- ies which explain how life-history traits have evolved by natural selection (Roff 1992; Stearns 1992). However, life-history traits are not usually adaptive to their envi- ronment (Futuyma 1986). For example, variation in life- history traits may be caused by genetic drift. In this sit- uation, life-history traits are not influenced by environ- mental factors at all. On the other hand, life-history traits ' Present address: Natural History Institute, National Science Museum, 3-23-1 Hyakunin-cho, Shinjuku-ku, Tokyo 169, Japan, fax: +81-3-3364-7104; e-mail: kurosato@kahaku.go.jp important to elucidate both environmental and genetic factors. The venerid bivalve Phacosoma japonicum (Reeve, 1850) is a common intertidal to subtidal species along the coasts of Japan, Korea, and China (Habe, 1977). Life- history traits of this species have been analyzed in detail (Tanabe, 1988; Tanabe & Oba, 1988; Sato, 1994, 1995), and a progressive change in life-history traits along a north-south gradient has been observed. Northern popu- lations generally exhibit more delayed sexual maturity and larger shell size at a given age than the southern ones. However, one southern population from Ariake Bay, Kyushu, does not fit this general trend. Age of sexual maturity and maximum shell size of the population in Ariake Bay are comparable to those of the northern pop- ulation in Hakodate Bay, Hokkaido (Figure 1). Because the genetic distance between the Ariake and Hakodate populations is much greater than those between popula- S. Sato, 1999 Legends Maximum shell height Age of sexual maturity i} i ! 6.49 cm 40°N------- 4 tea{olam {a} amjajamielem|a]em{alemialem|=imnini=mia 4 yrs poo i} Kagoshima Bay 8.02 cm >S yrs Ishikari Bay 7.21 cm 5 yrs Hakodate Bay Seto Inland Sea 4.95 cm Page 55 Tokyo Bay 5.80 cm 3 yrs AAburatsubo Cove 5.74 cm 3 yrs 3 yrs Figure 1 Sampling locations and life-history traits in each local population of Phacosoma japonicum. Numbers in box are maximum asymptotic shell height determined by the Bertalanffy equation (upper) and age of sexual maturity (lower), which were analyzed by Sato (1994). tions in Ariake Bay and its neighboring regions (Sato, 1996), the aberrant life-history traits in the Ariake pop- ulation cannot be explained solely by phylogenetic re- strictions such as genetic drift. In this study, to reveal the factors responsible for the variations in the life-history traits of this species, I at- tempted to estimate their genetic and environmental back- grounds by transplant experiments and analysis of month- ly shell growth and reproductive cycles for the popula- tions from Ariake Bay and its neighboring regions. MATERIALS AnD METHODS Transplant Experiments Transplant experiments were carried out on the inter- tidal sand flat of Aburatsubo Cove, Sagami Bay, central Honshu (35°9’20"N, 139°36'55”E). A total of 128 living individuals (2 to 4 years old) of Phacosoma japonicum were collected by digging with hands in April 1992 from the intertidal sand flat of (1) Nojima Coast, Tokyo Bay, central Honshu; (2) Nagahama Coast, Ariake Bay, central Kyushu; and (3) Shigetomi Coast, Kagoshima Bay, south- ern Kyushu (see Figure 1 for sampling localities). Each individual was tagged with a plate with an iden- tification number, and transplanted at several sites in the intertidal zone of Aburatsubo Cove. Shell height of each individual was measured annually between April 1992 and April 1995, and those of the 2 year old individuals measured bimonthly from April 1992 to October 1992 with a slide caliper (accuracy + 0.05 mm), and shell growth trajectory after transplantation was analyzed. Seasonal Patterns in Growth and Reproduction Seasonal changes in shell growth and reproductive cy- cle were examined for the native individuals from Tokyo, Ariake, and Kagoshima Bays. Thirty to forty individuals were sampled monthly from Ariake and Kagoshima Bays from January to September, 1995. Data on individuals from Tokyo Bay during January to October, 1992 are from Sato (1995). Gonadal tissue in each individual was excised and weighed using a dial scale (accuracy + 10 Page 56 mg). Then the dissected gonadal tissue was fixed for 48 hours in a solution of 10% formalin buffered with sea- water, followed by dehydration through a graded series of ethanols and benzols, and then embedded in paraffin (melting point: 58°C). Thin transverse sections of the go- nadal tissue were prepared at intervals of 8 im thickness and were stained with hematoxylin-eosin. The stained thin sections were subsequently observed using an Olym- pus model AHBS-515 optical microscope. Based on histological examination of thin-sectioned gonads, each individual was assigned to one of the spe- cific gonad developmental stages (early active [EA], late active [LA], ripe [R], partially spawned [PS], spent [S]) as previously defined by Sato (1995) (see Sato, 1995: figs. 1, 2). The frequency of each stage in monthly samples provided data on the temporal progression of the repro- ductive cycle. In addition, I calculated the mean gonad index percentage gonad mass to total mass of soft tissue for sexually mature individuals every month. The seasonal changes of shell growth were analyzed based on the annual increments produced by winter breaks (cf. Tanabe, 1988). Shell height from the umbo to the ventral margin of each winter break was measured using a slide caliper to an accuracy of + 0.05 mm. Sub- sequently, net growth (x), defined as the increase of shell height in the period from the time of the last winter break to the month of sampling, was standardized for each in- dividual by the expected annual growth (y), defined as the distance from the last winter break to the expected next winter break (cf. Goshima & Noda, 1992). The per- centage of the ratio x/y is defined as the ‘“‘growth index”’ (Sato, 1995). The extent of the expected annual growth (y) for each individual was estimated using the Ford-Walford equation (Ford, 1933; Walford, 1946). The equation is expressed as Hy; = aH, + b; where H, is the shell height at the Rth winter break (in mm), H,,, is the shell height at the R+/th winter break (in mm), and a and b are constants determined by a sim- ple regression between H, and Hp,, of different individ- uals of the same sample. Using this equation, shell height at the expected next winter break of each individual (H,.,) can be estimated by shell height at the last winter break (H,) and constants at each age class (a,b). Environmental Conditions of Each Locality Data of seasonal changes of water temperature, salin- ity, and contents of chlorophyll a near the sampling lo- calities were quoted from the unpublished data of the To- kyo Metropolitan Office (Tokyo Bay), of the Kagoshima Environmental Research and Service (Kagoshima Bay), and of the Kumamoto Prefectural Fisheries Research Sta- tion (Ariake Bay). Seasonal patterns of shell growth and The Veliger, Vol. 42, No. 1 reproductive cycle of Phacosoma japonicum were com- pared with the environmental data, and then the factors that directly influenced the life-history traits of this spe- cies were analyzed. RESULTS Transplant Experiments Annual shell growth patterns of the individuals of the four populations in the original localities and those of the transplanted individuals are shown in Figure 2. The mean shell growth pattern of the native individuals in Tokyo Bay is very similar to that in Aburatsubo Cove, but those in Kagoshima and Ariake Bays show earlier and later decline of shell growth than the individuals in Aburatsubo Cove, respectively (Sato, 1994). The annual shell growth patterns of most individuals transplanted from Tokyo Bay to Aburatsubo Cove were quite similar to those of the native individuals in Tokyo Bay and Aburatsubo Cove (Figure 2a). However, annual shell growth rate of the individuals transplanted from Ka- goshima and Ariake Bays to Aburatsubo Cove decreased at an earlier and later age than the native individuals in Aburatsubo Cove, respectively. Individuals transplanted from Kagoshima Bay attained the maximum shell size at 4 or 5 years old (Figure 2b). Most of the individuals transplanted from Ariake Bay continued to grow larger after 6 years old in contrast to the native individuals in Aburatsubo Cove (Figure 2c). The annual shell growth patterns of most transplanted individuals resembled those of the native individuals in their populations of origin. Mean bimonthly shell growth rate from April 1992 to October 1992 in the 2 year old transplanted individuals is shown in Figure 3. The shell growth of the individuals transplanted from Tokyo and Kagoshima Bays to Abu- ratsubo Cove occurred mainly from June to October, but shell growth in individuals transplanted from Ariake Bay to Aburatsubo Cove was observed during April and Au- gust (Figure 3). As mentioned, shell growth was active from winter to early spring in the native individuals from Ariake Bay, but growth occurred from late spring to sum- mer in the native individuals from Tokyo and Kagoshima Bays (see Figure Sb). These results indicate that most transplanted individuals attaining 2 years of age from the three geographically isolated habitats did not show any marked differences in both annual and seasonal shell growth patterns from those in their original habitats. Seasonal Patterns of Reproduction and Growth Seasonal variation in the frequency of each phase in the reproductive cycles of Phacosoma japonicum collect- ed from Tokyo, Kagoshima, and Ariake Bays is shown in Figure 4. The reproductive cycle of the sample from Kagoshima Bay was similar to that from Tokyo Bay. In contrast, gonad development in individuals from Ariake S. Sato, 1999 (a) Tokyo Bay ——> Aburatsubo Cove Ariake Bay : Tokyo Bay, Aburatsubo Cove 5.0 4 “Kagoshima Bay Shell height (cm) Shell height (cm) w oO io Page 57 : Tokyo Bay, Aburatsubo Cove Kagoshima Bay : (b) Kagoshima Bay ——> Aburatsubo Cove (c) Ariake Bay ——> Aburatsubo Cove Ariake Bay Ariake Bay Tokyo Bay, Aburatsubo Cove Kagoshima Bay pp 5.05 4.07 Shell height (cm) Figure 2 Results of the transplant experiments in Aburatsubo Cove. Each line represents annual shell growth trajectory of different individual transplanted from Tokyo Bay (a), Kagoshima Bay (b), and Ariake Bay (c). Bold lines indicate averaged annual shell growth patterns of the samples from the four original populations. Bay occurred earlier. Namely, ripe individuals appeared between March and April in Ariake Bay, whereas they occurred after May in Tokyo and Kagoshima Bays (Fig- ure 4). Some individuals also started to spawn in early June in Ariake Bay, but in Tokyo and Kagoshima Bays, spawning was delayed until July to August. The mean gonad index of the sample from Ariake Bay had already increased at a high rate (more than 40%) until April (Fig- ure 5a). In contrast, in the samples from Tokyo and Ka- goshima Bays, it increased at a lower rate (less than 40%) during January to April, and then increased rapidly in May and June. The growth index of each individual in the samples from Tokyo, Kagoshima, and Ariake Bays was calculated based on the constants of Ford-Walford equation at each age class. The seasonal patterns of growth index showed that at Ariake Bay the rate of increase in shell height was rapid from February to April, but declined from April to July (Figure 5b). In contrast, at Tokyo and Kagoshima Bays, shell growth rate was generally low from January to April and then high during April to September. These data suggest that both reproduction and shell growth of this species are active from winter to early spring at Ari- ake Bay, but those occurred from late spring to summer at Tokyo and Kagoshima Bays. Environmental Conditions of each Locality Seasonal variations in water temperature, salinity, and content of chlorophyll a near the sampling localities are shown in Figure 6. Seasonal changes in temperature and salinity are similar among them (Figure 6a, b). Also, spa- tial densities of individuals do not differ considerably among the habitats (Sato, personal observation). How- ever, the seasonal cycle in phytoplankton abundance varies markedly among the bays. In Tokyo and Kagoshi- ma Bays, the content of chlorophyll a increases in sum- mer (June—August), but remains at low levels in the other seasons (Figure 6c). In contrast, in Ariake Bay, the con- tent of chlorophyll a attains a maximum in winter (Jan- uary—March) and is at low levels in the other seasons. The seasonal change of chlorophyll a usually reflects the phytoplankton abundance in each habitat, and generally the phytoplankton bloom occurs in spring in embayments of northern Japan, and in summer in those of central and southern Japan (Iizumi et al., 1990; Yamashita, 1982). The winter bloom in Ariake Bay is, therefore, a peculiar phenomenon for central and southern regions in Japan. DISCUSSION Genetic Control of Shell Growth Patterns Both annual and seasonal shell growth patterns of the individuals transplanted from Tokyo, Ariake, and Kago- shima Bays to Aburatsubo Cove did not differ from those of the individuals in the original habitats (Figures 2, 3). These facts demonstrate that (at least for individuals of more than 2 years of age) transplanted individuals do not change their growth patterns in accordance with a new Page 58 = fo) Tokyo Bay Aburatsubo Cove (Nga) Growth rate (mm / 2 months) ie) o) 0.0 am 4.0 (b) Oe is Kagoshima Bay iS = = fat] 2.0 5 = Aburatsubo Cove Bg (Ni=a2) _ 4.0 o 2 BE Ariake Bay 420 - Aburatsubo Cove Ore (N = 8) CHOSE Apr. Jun. Aug. Oct. 1992 Figure 3 Mean bimonthly shell growth rate (April-June, June-August, August—October in 1992) among the 2 year old individuals trans- planted from Tokyo Bay (a), Kagoshima Bay (b), and Ariake Bay (c) to Aburatsubo Cove. Mean and the range of one standard deviation (vertical bar) are indicated. environment. The fact that the individuals from Ariake Bay could grow in Aburatsubo Cove as large as those in the original habitat, in spite of the stress expected for these transplanted animals, clearly indicates that the mode of shell growth in this species is not controlled directly by environmental factors. Study of the genetic structure of this species showed that the genetic distance between the Tokyo and Ariake populations is nearly zero (D < 0.0005), indicating a high gene flow between them (Sato, 1996). Nevertheless, the shell growth pattern and the age of sexual maturity differ markedly between them (Sato, 1994). These lines of ev- idence strongly suggest that shell growth and sexual ma- turity patterns of this species show phenotype plasticity against environmental fluctuations (Stearns, 1989). Avail- able data taken from the transplant experiments in the present study, however, could not account for the exis- tence of such phenotype plasticity in the shell growth patterns of this species. As a result, two possible alter- native explanations for the genetic control of the shell The Veliger, Vol. 42, No. 1 Tokyo Bay (1992) — [o) [o) ie) je) (ep) je) Frequency (%) i Kagoshima Bay (1995) {ee} oO D oO Frequency (%) aS oO Frequency (%) ASO JE Mia Maid Figure 4 Frequency of occurrence of each phase of the reproductive cycle in monthly collected samples of Phacosoma japonicum from Ka- goshima and Ariake Bays during January—September, 1995. Data from Tokyo Bay during January—October, 1992 from Sato (1995). (EA) early active phase, (LA) late active phase, (R) ripe phase, (PS) partially spawned phase, and (S) spent phase. growth patterns are suggested: (1) phenotype plasticity may function when the individuals are younger than 2 years old, and (2) rapid evolution of the life-history traits may occur within several generations, which is too short to cause the differentiation in allozyme loci among pop- ulations. In order to clarify this question further transplant S. Sato, 1999 (a) —o— Tokyo Bay (1992) ~~-}-~- Kagoshima Bay (1995) ----O---- Ariake Bay (1995) on (jo) £ oO MEAN GONAD INDEX (%) (a) $ a ro) MEAN GROWTH INDEX (%) NAS Ma od a Ay cS 0 Figure 5 Seasonal changes in development of gonads (a) and shell growth (b) in mature specimens (> 4 years old) of Phacosoma japoni- cum from Kagoshima and Ariake Bays during January—Septem- ber, 1995. Data from Tokyo Bay during January—October, 1992 from Sato (1995). Mean value and the range of one standard deviation (vertical bar) are indicated. studies focusing on the juveniles of this species need to be done. Relation of Life-History Traits to Environmental Factors The growing seasons for both shell growth and gonad development in Phacosoma japonicum are limited to the interval between winter and early spring in Ariake Bay, and between late spring and summer in Tokyo and Ka- goshima Bays (Figures 4, 5). Moreover, based on a scler- ochronological study of the marked and recovered spec- imens, Tanabe (1988) detected rapid shell growth be- tween April and September in the population of the Seto Inland Sea (southern Japan, see Figure 1). Tanabe & Oba (1988) also estimated the range of temperatures during the growing season from oxygen isotopic analysis of a specimen from Wakkanai Port, Hokkaido (northern Ja- Page 59 (a) Temperature ---O--- Tokyo Bay (1992) ssreen ©--- Kagoshima Bay (1995) —o— Ariake Bay (1995). 30 Water temperature (°C) Salinity Co) (c) Food availability 100 50 Chi a contents JA S ON OD J FM AM J Figure 6 Monthly mean water temperature (a), salinity (b), and seasonal variations of chlorophyll a (c) near the sampling localities. Data sources are unpublished data of the Tokyo Metropolitan Office (Tokyo Bay in 1992), of the Kagoshima Environmental Research and Service (Kagoshima Bay in 1995), and of the Kumamoto Prefectural Fisheries Research Station (Ariake Bay in 1995). pan, see Figure 1), and suggested that shell growth of the individual occurred between May and August. This dif- ference in growing season between the Ariake and other populations is more plausibly explained by the difference in seasonal change of phytoplankton abundance than by factors such as temperature, salinity, and individual den- sity. These data suggest that seasonal variations in shell growth and reproductive cycle of this species are primar- ily influenced by the seasonal change of food availability. In this species, northern populations generally display more delayed sexual maturity and attain larger shell size at a given age than do southern ones (Figure 1). This Page 60 north-south gradient of life-history traits suggests that the difference in water temperatures among sampling sites is one of the factors which affects life-history traits in this species. Sebens (1979, 1987) introduced a model for op- timum size at sexual maturation. According to him, the energy available for growth and reproduction is deter- mined by the difference between energy intake and met- abolic cost. The energy intake is primarily affected by food availability, and the metabolic cost in most poiki- lotherms is closely related to temperature (Sebens, 1979, 1980; Clarke, 1987). Because animals mostly utilize the energy during growing season for their growth and re- production, the optimum size for sexual maturation for each population is determined by the food availability and water temperature during growing season. In Phacosoma japonicum, both shell and gonad growth are generally limited to the interval between late spring and summer, except for the Ariake population. Because the phytoplankton bloom occurs in spring in northern Ja- pan and in summer in central and southern Japan (izumi et al., 1990; Yamashita, 1982), energy intake during the growing season is sufficient. However, at lower latitude, the water temperatures during the growing season in- crease, so that metabolic cost during the growing season increases, and the optimum size for sexual maturation be- comes smaller. This prediction from the optimum-size model is consistent with the north-south gradient of the life-history traits of P. japonicum. Moreover, in the Ariake population, rapid shell growth and gonad development occurred in a limited period be- tween February and April (Figures 4, 5). Since the phy- toplankton productivity increases in winter in Ariake Bay, food resources during the growing season of P. japoni- cum are also sufficient. However, the mean water tem- perature during the growing season for the Ariake pop- ulation (11—15°C: February—April) is much lower than those in neighboring regions (15—28°C: April-August) (Figure 6). Therefore, Sebens’ model predicted that the optimum size at sexual maturation for the Ariake popu- lation is much larger than those in the neighboring re- gions, because more energy can be allocated to growth and reproduction due to the low metabolic rate in the growing season. This prediction agrees with the fact that shell size at sexual maturity of the Ariake population is much larger than those of the neighboring populations. In conclusion, growing season in shell growth and re- production of this species is strongly influenced by food availability, and the geographical variations of annual shell growth and sexual maturity patterns can be ex- plained mainly by the mean water temperature during the growing seasons. Such intraspecific variations are genet- ically stable and are not influenced by short-term envi- ronmental changes. ACKNOWLEDGMENTS The author thanks Drs. K. Tanabe, T. Oji, and M. Nakao- ka (University of Tokyo) for their critical reviews of the The Veliger, Vol. 42, No. 1 manuscript and valuable comments, Drs. M. Sato (Ka- goshima University), K. Endo, and other colleagues at the Paleobiological Laboratory of the University of Tokyo for their fruitful advice, and the staffs of the Kumamoto Pre- fectural Fisheries Research Station and of the Kagoshima Environmental Research and Service for providing oceanographic data for this study. This study was partly supported by Grant-in-Aid for Scientific Research from the Japan Society for the Pro- motion of Science (Research Fellow of Doctoral Class, No. 4407) and Fujiwara Natural History Foundation. LITERATURE CITED ANSELL, A. D. 1968. The rate of growth of the hard clam Mer- cenaria mercenaria (L.) throughout the geographical range. Journal du Conseil 31:364—409. APPELDOORN, R. S. 1983. Variation in the growth rate of Mya arenaria and its relationship to the environment as analyzed through principal components analysis and the w parameter of the von Bertalanffy equation. Fishery Bulletin 81:75—83. BEUKEMA, J. J., G. C. CADEE & J. J. M. JENSEN. 1977. Variability of growth rate of Macoma balthica (L.) in the Wadden Sea in relation to availability of food. Pp. 69-77 in B. FE Keegan, P. O’Ceidigh & P. J. S. Boaden (eds.), Biology of Benthic Organisms. Proceedings 11th European Marine Biology Symposium Pergamon, Oxford. Broom, M. J. 1982. Analysis of the growth of Anadara granosa (Bivalvia: Arcidae) in natural, artificially seeded and exper- imental populations. Marine Ecology Progress Series 9:129— 144. CLARKE, A. 1987. Temperature, latitude and reproductive effort. Marine Ecology Progress Series 38:89-99. Forp, E. 1933. An account of the herring investigations con- ducted at Plymouth during the years from 1924-1933. Jour- nal of the Marine Biological Association of the United King- dom 19:305-384. FutuyMa, D. J. 1986. Evolutionary Biology. Sinauer Associa- tions, Incorporated: Sunderland, Massachusetts. 600 pp. GILBERT, M. A. 1973. Growth rate, longevity and maximum size of Macoma balthica (L.). Biological Bulletin, Marine Biol- ogy Laboratory, Woods Hole, Massachusetts 145:119—126. GOSHIMA, S. & T. NopA. 1992. Shell growth of the north Pacific cockle Clinocardium californiensis in Hakodate Bay, Hok- kaido. Benthos Research 42:39—48. [in Japanese with En- glish abstract] GREEN, R. H. & K. D. Hopson. 1970. Spatial and temporal struc- ture in a temperate intertidal community, with special em- phasis on Gemma gemma (Pelecypoda: Mollusca). Ecolog 51:999-1011. HABE, T. 1977. Systematics of Mollusca in Japan: Bivalvia and Scaphopoda [in Japanese]. Hokuryukan Book Company: To- kyo. 372 pp. lizumi, H., K. Furuya, I. TAKEUCHI, K. KursuwapDA, A. TADA, & K. Kawacuculi. 1990. Dynamics of nutrients and chlo- rophyll a at Otsuchi Bay: analyses of the data of monthly observation. Otsuchi Marine Research Center Report 16:63— 65. [in Japanese] IRLANDI, E. A. & C. H. PETERSON. 1991. Modification of animal habitat by large plants: mechanisms by which seagrasses in- fluence clam growth. Oecologia 87:307—318. McLusky, D. S. & D. G. ALLAN. 1976. Aspects of the biology S. Sato, 1999 of Macoma balthica (L.) from the estuarine Firth of Forth. The Journal of Molluscan Studies 42:31—45. NaAKAOKA, M. 1992. Spatial and seasonal variation in growth rate and secondary production of Yoldia notabilis in Otsuchi Bay, Japan, with reference to the influence of food supply from the water column. Marine Ecology Progress Series 88: 215-223. Nopa, T., S. NAKAO & S. GosHIMA. 1995. Life history of the temperate subtidal gastropod Umbonium costatum. Marine Biology 122:73-78. PETERSON, C. H. 1982. The importance of predation and intra- and interspecific competition in the population biology of two infaunal suspension-feeding bivalves, Protothaca sta- minea and Chione undatella. Ecological Monographs 52: 437-475. Rag, J. G. III. 1979. The population dynamics of two sympatric species of Macoma (Mollusca: Bivalvia). The Veliger 21: 384-399. REEVE, L. A. 1850. Conchologia Iconica. Vol. 6, Artemis sp. L. Reeve and Co. Ltd.: London. 61 pp. Rorr, D. A. 1992. The Evolution of Life Histories. Theory and Analysis. Chapman & Hall: New York & London. 535 pp. SaTo, S. 1994. Analysis of the relationship between growth and sexual maturation in Phacosoma japonicum (Bivalvia: Ve- neridae). Marine Biology 118:663—672. SaTo, S. 1995. Spawning periodicity and shell microgrowth pat- terns of the venerid bivalve Phacosoma japonicum (Reeve, 1850). The Veliger 38:61—72. SATO, S. 1996. Genetic variability and population structure of Phacosoma japonicum (Bivalvia: Veneridae). Venus The Japanese Journal of Malacology 55:51—63. Page 61 SEBENS, K. P. 1979. The energetics of asexual reproduction and colony formation in benthic marine invertebrates. American Zoologist 19:683—697. SEBENS, K. P. 1980. The control of asexual reproduction and col- ony formation in benthic marine invertebrates. Biological Bulletin 158:370—382. SEBENS, K. P. 1987. The ecology of indeterminate growth in an- imals. Annual Review of Ecology and Systematics 18:371— 407. STEARNS, S. C. 1989. The evolutionary significance of phenotyp- ic plasticity. BioScience 39:436—445. STEARNS, S. C. 1992. The Evolution of Life Histories. Oxford University Press: Oxford. 249 pp. TANABE, K. 1988. Age and growth rate determinations of an in- tertidal bivalve, Phacosoma japonicum, using internal shell increments. Lethaia 21:231—241. TANABE, K. & T. OBA 1988. Latitudinal variation in shell growth patterns of Phacosoma japonicum (Bivalvia; Veneridae) from the Japanese coast. Marine Ecology Progress Series 47: 75-82. TAYLOR, C. C. 1959. Temperature and growth. The Pacific razor clam. Journal du Conseil 25:93—101. THompson, J. K. & E H. NICHOLS. 1988. Food availability con- trols seasonal cycle of growth in Macoma balthica (L.) in San Francisco Bay, California. Journal of Experimental Ma- rine Biology and Ecology 116:43-61. WALFORD, L. A. 1946. New graphic method of describing the growth of animals. Biological Bulletin 90:141—147. YAMASHITA, T. 1982. Seasonal change of plankton in Ariake Bay—xXIV. Research report of the Fukuoka Prefectural Ar- iake fisheries experiment station in 1982: 125-132. [in Jap- anese] The Veliger 42(1):62—66 (January 4, 1999) THE VELIGER © CMS, Inc., 1999 A Worldwide Review of the Food of Nudibranch Mollusks. Part Il. The Suborder Dendronotacea GARY McDONALD Long Marine Laboratory, University of California at Santa Cruz, 100 Shaffer Road, Santa Cruz, California 95060, USA AND JAMES NYBAKKEN Moss Landing Marine Laboratories, PO. Box 450, Moss Landing, California 95039, USA Abstract. The prey items of 108 species representing all 10 families of the suborder Dendronotacea are presented in shortened form from the much larger electronic database accessible on the Web. INTRODUCTION This paper is the second paper in a series that will review the food of nudibranchs on a worldwide basis. As noted in the initial paper in this series (McDonald & Nybakken, 1997) these reviews are based on the published literature and are not the result of the investigations of the authors. Furthermore, each paper will be an abstracted version of a much larger database that contains all of the available information on the food of the nudibranchs extracted from an extensive search of the literature. This larger database has been deemed too large to publish in its entirety in hard copy and is available as the first electronic publication of the The Veliger. Electronic supplements and appendices of papers published in The Veliger are available via anony- mous FIP from ucmp1.Berkeley.Edu. These documents are available in three formats: PostScript (*.PS), Word- Perfect (*.WP), and ASCII (*.ASC). To retrieve a docu- ment, open an FIP connection to ucmp1.Berkeley.Edu (128.32.146.30). At the request for login enter “‘anony- mous.” At the request for a password enter your e-mail address (e.g., jsmith@veliger.amu.edu). At the prompt change directory to /pub/mollusca/veliger (Command = cd /pub/mollusca/veliger), set file transfer mode to binary (command = bin), and retrieve the desired file (command = get “filename.*’’). At the end of your FTP session close the connection (command = close) and quit. The electronic files associated with this paper are nudifood1.ps, nudi- foodl.wp, and nudifood1.asc. Introduction to the Dendronotacea The suborder Dendronotacea is the second smallest in the order Nudibranchia. The number of living species worldwide is not precisely known due in part to the pres- ence of several cryptic species of the speciose genus Doto that have recently been uncovered through both molecu- lar and morphological techniques (Morrow et al., 1992; Lemche, 1976; Goddard, 1996). This suggests that other single species in the genus may, in fact, be groups of cryptic species. We here report on the food of 108 species with representatives from all 10 families. According to Thompson (1988) and Thompson & Brown (1984), this suborder is characterized by having the rhinophores retractile into sheaths, the dorsum mar- gined with simple or branched processes and a mid-lateral anal opening. There has been no recent systematic review of the suborder since Odhner (1936). RESULTS Anp DISCUSSION As can be seen from Table 1, the dendronotaceans have a somewhat varied diet, and the suborder includes both specialists and generalists. All species, however, prey on one or more species of the phylum Cnidaria. The species of the Tritoniidae all prey upon octocorals of the orders Alcyonacea, Gorgonacea, Pennatulacea, and Stolonifera (note in the table that if alcyonarians is listed it means that the order is not known and when alcyonaceans is used it means the order). The families Lomanotidae, Embletoniidae, Hancocki- idae, Bornellidae, Scyllaeidae, and Dotonidae all con- sume thecate and athecate hydroids. The Dendronotidae also seem to be primarily predators of thecate and athe- cate hydroids, but there are two species which appear to be specialists. They are D. iris, which feeds on cerianthid anemones, and D. rufus, which preys on scyphozoan scy- phistomae. The family Tethyidae all appear to feed primarily on small crustaceans which they capture using a large oral hood to sweep the organisms out of the water column or off the substrate. G. McDonald & J. Nybakken, 1999 Page 63 Table | Summary of the food of the suborder dendronotacea. Family Genus Tritoniidae Marionia Species blainvillea Food on alcyonaceans (Alcyonium) and gorgonians (Eunicella, Lophogorgia, Paramuricea) cucullata alcyonarians occidentalis alcyonarians quadrilatera alcyonaceans (Alcyonium) Marioniopsis cyanobranchiata alcyonaceans (Xenia) platyctenea alcyonaceans (Parerythropo- dium) Paratritonia lutea gorgonians (Mopsella) Tochuina tetraquetra alcyonaceans (Gersemia), penna- tulaceans (Ptilosarcus) Tritonia alba on alcyonaceans (Alcyonium) aurantiaca bayeri Melitodes gorgonians (Briaeum, Pseudopterogoria) diomedea pennatulids (Ptilosarcus, Stylatu- la, Virgularia) festiva stoloniferans (Clavularia), al- cyonaceans (Gersemia), gor- gonians (Lophogorgia), and pennatulids (Ptilosarcus) griegi gorgonian (Paramuricea) hamnerorum gorgonian (Gorgonia) hawaiiensis stoloniferan (Anthelia) hombergi alcyonacean (Alcyonium) incerta alcyonarians (Alcyonium) lineata stoloniferan (Sarcodictyon) manicata stoloniferan (Cornularia) nilsodhneri gorgonians (Eunicella, Lopho- gorgia) pickensi gorgonians (Muricea, Psammo- gorgia) plebeia alcyonaceans (Alcyonium), gor- gonians (Eunicella, Lophogor- gia, Paramuricea) striata alcyonacean (Paralcyonium) vorax stoloniferan (Pachyclavularia) wellsi gorgonians (Gorgonia, Leptogor- gia) Tritoniella belli stoloniferans (Clavularia, Pachy- clavularia) Tritoniopsis alba frydis “soft coral’”’ on gorgonians Marianinidae Marianina rosea barlettai hydroids thecate hydroids (Kir- Lomanotidae Lomanotus chenpaueria, Ventromma) flavidus thecate hydroids (Antennularia, Nemertesia) genei on thecate hydroids (Antennular- ia, Nemertesia), athecate hy- droids (Tubularia) and bryo- zoans (Cellaria) marmoratus thecate hydroids (Antennularia, Nemertesia) vermiformis thecate hydroid (Lytocarpus) Embletoniidae Embletonia evelinae on thecate hydroid (Obelia) gracilis campanularid hydroids pulchra thecate hydroids (Antennularia, Nemertesia, Hydrallmania); athecate hydroids (Cordylo- phora, Tubularia) Page 64 Family Genus Hancockiidae Hancockia Dendronotidae Dendronotus Bornellidae Bornella Scyllaeidae Crosslandia Notobryon Tethyidae Scyllaea Melibe Tethys Dotonidae Doto Table 1 Continued. Species uncinata albopunctatus albus dalli diversicolor frondosus iris robustus rufus subramosus stellifer anguilla daedali wardi pelagica bucephala fimbriata leonina megaceras mirifica pilosa rosea fimbria acuta amyra arteoi aurita chica cindyneutes cinerea The Veliger, Vol. 42, No. 1 Food thecate hydroids (Campanularia, Clytia, Nemertesia, Obelia); athecate hydroids (Tubularia) on athecate hydroids (Tubularia) thecate hydroids (Abietinaria, Plumularia, Thuiaria); athe- cate hydroid (Tubularia) thecate hydroid (Abietinaria) thecate hydroids (Abietinaria, Hydrallmania, Sertularella) thecate hydroids (12 genera); athecate hydroids (7 genera); ascidiacean (Botryllus) cerianthid anthozoans (Pachycer- iathus); thecate hydroids (Obelia) beetles!; campanularid hydroids; oweniid polychaetes scyphozoan schyphistomae thecate hydroids (Aglaophenia, Obelia); athecate hydroid (Tubularia) thecate hydroid (Sertularia) thecate hydroid (Plumularia) hydroids campanularid hydroids hydroids detritus and microorganisms on algae crustaceans amphipods, bivalve spat, cope- pods, isopods, ostracods, small crustaceans, veliger larvae, zoea larvae, megalops larvae small crustaceans small crustaceans copepods, isopods, portunid crab, small crustacea amphipods amphipods, ophiuroids (Amphi- ura, Ophioglypha), copepods, crustaceans, brachyurans, iso- pods, ostracods, small echino- derms, small gastropods, sto- matopods, worms thecate hydroid (Obelia) thecate hydroids (Abietinaria, Aglaophenia, Bougainvillia, Obelia, Sertularia) thecate hydroid (Laomedea) on athecate hydroid (Tubularia) on athecate hydroid (Euden- drium) on thecate hydroid (Halecium) on thecate hydroids (Aglaophen- ia, Sertularia) G. McDonald & J. Nybakken, 1999 Genus Table 1 Continued. Page 65 Family Species columbiana coronata cuspidata doerga dunnei eireana floridicola formosa fragilis furva ganda hydrallmaniae hystrix Japonica koenneckeri kya lancei lemchei maculata millbayana oblicua onusta paulinae pinnatifida pita pontica rosea sarsiae Food on thecate hydroids (Aglaophen- ia, Obelia, Selaginopsis) and on athecate hydroids (Tubular- ia) 16 genera of thecate hydroids and 10 genera of athecate hy- droids and the bryozoan Al- cyonidium thecate hydroids (Nemertesia) thecate hydroids (Aglaophenia, Obelia) thecate hydroid (Kirchenpaueria) thecate hydroids (Dynamena, Sertularia) on thecate hydroids (Aglaophen- ia, Synthecium) athecate hydroid (Eudendrium) thecate hydroids (Antennularia, Halecium, Hydrallmania, Nemertesia) and athecate hy- droids (Tubularia) on thecate hydroids (Campanu- laria, Filellum, Halecium, Ser- tularia, Sertularella) on thecate hydroid (Obelia) thecate hydroid (Hydrallmania) thecate hydroid (Schizotricha) on thecate hydroid (Aglaophen- ia) thecate hydroid (Aglaophenia) thecate hydroids (Abietinaria, Aglaophenia, Obelia, Plumu- laria, Sertularella, Sarsia) and athecate hydroids (Euden- drium) thecate hydroids (Aglaophenia) thecate hydroids (A glaophenia, Synthecium) thecate hydroids (Halopteris, Plumularia, Schizotricha) thecate hydroids (Aglaophenia, Nemertesia, Plumularia) thecate hydroids (Amphisbetia, Sertularia) thecate hydroid (Dynamena) thecate hydroids (Aglaophenia, Obelia) and athecate hydroid (Eudendrium) thecate hydroids (Aglaophenia, Antennularia, Halecium, Nem- ertesia, Obelia, Sertularia) on thecate hydroids (Clytia, Obelia, Sertularella, Orthopy- xis) on thecate hydroid (Aglaophen- ia) on athecate hydroid (Euden- drium) athecate hydroid (Sarsia) Page 66 Table 1 Continued. Family Genus Phylliroidae Phyllirhoe Cephalopyge LITERATURE CITED GODDARD, J. H. R. 1996. Lecithotrophic development in Dota amyra (Nudibranchia: Dendronotacea), with a review of de- velopmental mode in the genus. The Veliger 39(1):43—54. LEMCHE, H. 1976. New British species of Doto Oken, 1815 (Mol- lusca: Opisthobranchia). Journal of the Marine Biological Association of the United Kingdom 56(3):691—706. McDonaLp, G. & J. NYBAKKEN. 1997. A worldwide review of the food of nudibranch mollusks. Part I. Introduction and the suborder Arminacea. The Veliger 40(2):157—-159. Morrow, C. C., J. P. THORPE & B. E. Picton. 1992. Genetic Species tuberculata unguis Uussi uva verdicioi wara bucephala mediterranea trematoides The Veliger, Vol. 42, No. 1 Food thecate hydroids (Abietinaria, Sertularella) thecate hydroid (Nemertesia) on thecate hydroid (Aglaophen- ia) on thecate hydroid (Pennaria) thecate hydroids (Aglaophenia, Laomedea) on thecate hydroid (Aglaophen- ia) thecate hydroid medusae (Aequo- rea); athecate hydroid medu- sae (Zanclea); larvaceans (Oi- kopleura) siphonophores (Halistemma) siphonophores (Nanomia, Halis- temma, Stephanomia) divergence and cryptic speciation in two morphs of the com- mon subtidal nudibranch Doto coronata (Opisthobranchia: Dendronotacea: Dotoidae) from the northern Irish Sea. Ma- rine Ecology Progress Series 84(1):53—61. ODHNER, N. H. 1936. Nudibranchia Dendronotacea a revision of the system. Memoires du Musee Royal d’Histoire Naturell de Belgique, series 2, fascicule 3:1057—1128. THompson, T. E. 1988. Molluscs: Benthic Opisthobranchs. Syn- opses of the British Fauna (new series) No. 8. 2nd ed. E. J. Brill: Leiden. 356 pp. THompson, T. E. & G. H. Brown. 1984. Biology of Opistho- branch Molluscs. Vol. If. The Ray Society: London. 229 pp. The Veliger 42(1):67—71 (January 4, 1999) THE VELIGER © CMS, Inc., 1999 The Giant Amazonian Snail (Pulmonata: Acavidae) Beats Them All FRANK P. WESSELINGH Biology Department, University of Turku, Turku, Finland and Nationaal Natuurhistorisch Museum, P.O. Box 9517, NL 2300 RA Leiden, The Netherlands AND EDMUND GITTENBERGER Institute for Ecological and Evolutionary Sciences, Leiden University & Nationaal Natuurhistorisch Museum, P.O. Box 9517, NL 2300 RA Leiden, The Netherlands Abstract. Pebasiconcha immanis gen. & sp. nov. is described from the Miocene Pebas Formation of Colombian Amazonia. Apart from a simple outer lip of the aperture and a prominent external “‘mytiloid’’ thickening shortly below the suture, behind the lip, the genus resembles Strophocheilus. It is the largest terrestrial gastropod now known to have ever existed, and one of only two species of terrestrial snails recorded from the Pebas Formation. Its shells are approx. 20% higher than those of the largest Achatinidae known. INTRODUCTION Species of the Achatinidae are generally considered the largest terrestrial snails that ever lived on earth. That gas- tropod family, including the Giant African Snails of pop- ular writing, is restricted to tropical Africa. Their fusiform shells may reach approx. 21 cm in height. South Ameri- can species of the Acavidae, Strophocheilinae, have long been second on the world list, with similar shells up to 16 cm high. During fieldwork in deposits of the Miocene Pebas Formation in Colombian Amazonia, in the autumn of 1991, some remarkably large shells of snails were found. These shells, locally common in the formation, were ini- tially classified with the genus Strophocheilus Spix, 1827 (Pulmonata: Acavidae: Strophocheilinae); later, it was ob- served that they differ in apertural characters, however. The working conditions did not permit the transport of complete shells (that were broken into pieces and disin- tegrated), but photographs were made of an entire spec- imen in situ. A second visit to the area in 1996 yielded only additional shell fragments. On this visit, however, in a shop in Iquitos, Peru, though not for sale, a single well- preserved specimen was discovered; this shell could also be photographed. Therefore, the current description of a new genus and species is based on photographs and shell fragments. The species is one of only two terrestrial pulmonate species recorded from the Miocene Pebas Formation, which is rich in fresh- and brackish-water mollusks. It is truly giant in size. The maximal height of the fusiform shell, directly observed, is 25.6 cm; on the basis of shell fragments, similar dimensions can be calculated. This dis- covery implies a 20% increase in the maximum shell size known for terrestrial snails, dethroning the Giant African Snails in favor of this ‘““Giant Amazonian Snail.” SYSTEMATIC PALEONTOLOGY Class Gastropoda Order Stylommatophora A. Schmidt, 1855 Family ACAVIDAE Pilsbry, 1895 Subfamily STROPHOCHEILINAE Thiele, 1926 Pebasiconcha immanis Wesselingh & Gittenberger gen. & sp. nov. (Figures 1, 2) Holotype (Figure 1a): a shell, complete when collected but secondarily broken into pieces (Instituto de Investi- gaciones en Geociencias, Mineria y Quimica [‘“‘Ingeom- inas’’], Bogota, Colombia, unnumbered). Paratypes: Nineteen shell fragments from the type lo- cality and type stratum (Figure 2a, b) (Nationaal Natuur- historisch Museum, Leiden, The Netherlands - RGM 394327/30); a complete shell from locality 2 (Figure 1b,c) (M. Callegari, private collection, Iquitos, Peru). Type locality and locality 2: The holotype was found September 1991 by the first author in Colombia, State of Amazonas, in the cliff at the northern side of the conflu- ence of the rivers Amazonas and Loreto-Yacu, 1 m above the water table. The locus typicus is in the Pebas For- mation, and is assigned to a late Middle to early Late Miocene age (Grimsdalea zone) (Hoorn, 1994). Locality Page 68 The Veliger, Vol. 42, No. 1 Figure 1 Pebasiconcha immanis Wesselingh & Gittenberger gen. & sp. nov. a. holotype [height 25.6 cm] (irreparably broken after the photograph had been taken in the field), with the “‘mytiloid’’ knob indicated by an arrow (“‘Ingeominas” Collection, Bogota, Colombia, unnumbered). b—c. paratype [height 21.9 cm] from the vicinity of Pebas, Loreto department, Peru (M. Callegari private collection, Iquitos, Peru, unnumbered). Photos: FE Wesselingh. 2 cannot be indicated more exactly than ‘‘vicinity of” Pebas, Loreto department, Peru; outcrops from that area have been assigned to a Middle Miocene age (Crasso- retitriletus zone) (Hoorn, 1994). Diagnosis: Shell very large, reaching over 25 cm in height; body whorl with a markedly constricted aperture; apertural lip neither thickened nor reflected; upper half of the body whorl, shortly behind the apertural lip, with a conspicuous knob, accompanied more posteriorly by a less prominent swelling. Description: The ovoid shell has up to nearly six mod- erately convex whorls, separated by a suture, which be- comes more strongly indented toward the aperture. The body whorl measures approx. four-fifths of and the ap- erture approx. one-third of the total shell height; it is slightly more flattened than the preceding whorls. The protoconch sculpture is unknown because the initial whorls are abraded or dissolved in all specimens. The teleoconch has irregular prosocline riblets that are most prominent at the apical side of the whorls. The penulti- mate whorl has about 170 such riblets. In between them, small malleated areas may occur, in particular on frag- ments with a weakly developed sculpture. On most of the body whorl the riblets are more or less obsolete, and the surface is somewhat malleated. There is a prominent, asymmetrical knob (Figures la, 2b) on the upper half of the body whorl, shortly behind the apertural border. This knob has a well-defined, narrow tip shortly below the suture and a broadly rounded basel part; in curvature it is reminiscent of a Mytilus valve. Toward the back, there is a far more obsolete roundish knob. The final quarter of the body whorl has a concave shoulder which is in- creasingly more prominent toward the aperture (Figure lc). The aperture is markedly constricted. Its outer lip is regularly curved and nearly circular, with the border nei- ther reflected nor thickened. The inner lip is straightened, with a relatively thin parieto-columellar callus. The um- bilicus is closed or slitlike; in some specimens a dam- aged, very narrow columellar canal is visible. Several fragments and the complete paratype bear ir- regularly arranged low ridges on the ultimate and the pen- ultimate whorls. On the lower half of the whorls these ridges are spirally arranged, whereas on the upper half F. P. Wesselingh & E. Gittenberger, 1999 Page 69 2cm Figure 2 Pebasiconcha immanis Wesselingh & Gittenberger, gen. & sp. nov., paratypes from the type locality & type stratum. a. apical fragment, RGM 394328. b. upper palatal wall fragment, with the “‘mytiloid knob,’ RGM 394329. G. A. Peeters del. Page 70 they converge obliquely toward the periphery. Some frag- ments have a brownish layer, possibly a remnant of the periostracum. Dimensions: The holotype suffered severely from trans- port. Because it had been photographed before, however, its size could be calculated as approx. 25.6 cm high and 14.5 cm broad; the aperture, measured outside, in frontal view, is 8.7 cm high and 8.2 cm broad. The locality 2 paratype measures 21.9 X 11.8 cm; its aperture is 7.0 cm high and equally broad. An apical fragment (RGM 394328: Figure 2a) is similar to the holotype in size. A fragment of the palatal apertural wall, with the “‘myti- loid’’ knob included (Figure 2b), is either of a specimen higher than 30 cm, or of a shell with a relatively large knob as compared to the holotype. Recently, Dr. A. J. de Winter (Nationaal Natuurhisto- risch Museum, Leiden) bought a shell of Archachatina marginata (Swainson, 1821), measuring 21.3 < 12.4 cm, from a villager at Nyangong, SW. Cameroun (De Winter, 1997). According to Dr. A. Mead (1961, and in litt., 24 April 1995), this specimen is larger than any shells of Giant African Snails recorded in his personal database; the shell is slightly aberrant in shape, however, which might be indicative of abnormal growth. A single shell of Achatina achatina (Linnaeus, 1758) is reported by Par- kinson et al. (1987:33), as “‘one exceptional specimen” of 27.3 cm, of this well-known species, normally mea- suring ‘‘up to 20 cm” only. Because of the magnitude of the difference in size with specimens hitherto reported, we prefer to consider this specimen an aberration, not to be considered in the regular species description. The “‘Gi- ant Amazonian Snail,’’ known from far fewer specimens, is considerably larger than the Giant African Snails. Classification and differentiation: The assignment of this species to the Acavidae, Strophocheilinae, is based on the size, shape, and sculpture of the shell, in particular the prosocline riblets at the apical side of the whorls, and its occurrence in Amazonia. Without anatomical data, this assignment has to remain poorly based. The closest rel- atives of this species might also be found among the Bu- limulidae or the Orthalicidae, mainly differing conchol- ogically by somewhat lower maxima in shell size (Zilch, 1960). The genus Strophocheilus Spix, 1827, is known from the neotropical region only, from the Paleocene on (Par- odiz, 1969; Zilch, 1960). Pebasiconcha immanis differs from all Strophocheilus species by its very large size and particularly by the simple outer lip of the aperture and the “mytiloid’”’ knob on the body whorl, shortly behind the lip. In Strophocheilus the straight outer lip is always clearly thickened, a generic autapomorphy, and the sur- face of the body whorl is not provided with any knobs. The simple outer lip in Pebasiconcha, found in many gastropod shells, may be considered a plesiomorphous The Veliger, Vol. 42, No. 1 character state (if not a reversal), whereas giant size and ‘““mytiloid’’ knob are seen as autapomorphies. Habitat: The Pebas Formation has been deposited in a fluvio-lacustrine to permanently lacustrine palaeo-envi- ronment (Hoorn, 1994). The deposits are rich in aquatic mollusks, i.e., cochliopine hydrobiids and pachydontine corbulids. Forty-four species are now known from these deposits (Nuttall, 1990). Despite predominantly lacustrine conditions in western Amazonia during the Middle Mio- cene, swamplike conditions and forested riverbanks must also have been present (Hoorn, 1994). Therefore, the number of only two terrestrial gastropod species, both pulmonates, P. immanis and Orthalicus linteus (Conrad, 1871) is remarkably low. The latter species probably lived on tree trunks and branches, the habitat occupied by the many congeneric Recent species. Pebasiconcha immanis might have lived on the humid bottom, where gravity is less problematic; it seems unlikely that such huge snails would have climbed trees. In outcrops of the Pebas Formation, concentrates of severely damaged specimens and fragments of P. imma- nis are common. Smaller, more clearly abraded fragments can be found dispersed throughout the deposits. The con- centrates are often part of lignitic lags. At the type lo- cality there are two such lags, overlain there by incursion layers containing abundant mangrove pollen, foraminif- era, and marine gastropods. The poor internal sorting of these lags and the preservation of delicate structures on some of the shell fragments (RGM 394327) indicate a rapid deposition without much reworking, suggesting ma- rine incursions in a fluviolacustrine environment. For oth- er lags with concentrates of P. immanis no relation with incursion events is seen. Etymology: Generic name after the Pebas Formation and concha, Latin for shell. Epithet immanis, Latin for enor- mous, huge, etc. ACKNOWLEDGMENTS We thank all those who supported our work in any way, in particular Prof. Dr. G. J. Boekschoten (Vrije Univer- siteit, Amsterdam, Netherlands); Mr. M. Callegari (Iqui- tos, Peru); Dr. C. Hoorn (University of Oman); Mr. A. Janssen (Nationaal Natuurhistorisch Museum, Leiden, Netherlands); Dr. A. Mead (University of Arizona, Tuc- son, U.S.A.); Dr. C. P. Nuttall (British Museum (Natural History), London, U.K.); Dr. M. Rasanen (University of Turku, Turku, Finland); Dr. L. Romero (Ingemmet, Lima, Peru); Dr. G. Sarmiento (formerly: Ingeominas, Bogota, Colombia); Dr. A. Villa (formerly: Inderena, Leticia, Co- lombia); and Mr. G. A. Peeters (Schiedam, Netherlands), who made the drawings. LITERATURE CITED DE WINTER, A. J. 1997. A giant specimen of Archachatina mar- ginata (Gastropoda Pulmonata: Achatinidae). Basteria 61: 41-42. F. P. Wesselingh & E. Gittenberger, 1999 Hoorn, C. 1994. An environmental reconstruction of the palaeo- Amazon River system (Middle-Late Miocene, NW Amazon- ia). Palaeogeography, Palaeoclimatology, Palaeoecology 112:187-238. MeEaD, A. R. 1961. The Giant African Snail; a Problem in Eco- nomic Malacology. The University of Chicago Press: Chi- cago & London. xvii + 257 pp. NUTTALL, C. P. 1990. A review of the Tertiary non-marine mol- luscan faunas of the Pebasian and other inland basins of Page 71 north-western South America. Bulletin of the British Mu- seum (Natural History), Geology 45:165—371. PARKINSON, B., J. HEMMEN & K. GROH 1987. Tropical Land- shells of the World. Verlag Christa Hemmen: Wiesbaden. 279 pp. Paropiz, J. J. 1969. The Tertiary non-marine Mollusca of South America. Annals of the Carnegie Museum 40:1-—242. ZILCH, A. 1960. Gastropoda. Teil 2. Euthyneura. Handbuch der Palaozoologie 6(2)3:401—600. The Veliger 42(1):72—84 (January 4, 1999) THE VELIGER © CMS, Inc., 1999 Gastropods and Intertidal Soft-Sediments: The Case of Chilina ovalis Sowerby (Pulmonata: Basommatophora) in South-Central Chile PEDRO QUIJON AND EDUARDO JARAMILLO Instituto de Zoologia, Universidad Austral de Chile, Valdivia, Chile Abstract. Field experiments (exclusion and inclusion of gastropods) were carried out in the intertidal of the Lingue River estuary (south-central Chile) during the summers of 1991 and 1992. Cages were used to analyze the effects of the snail Chilina ovalis Sowerby on the macroinfaunal community structure and the sedimentological properties of the top (1.5 cm) layer of the substrate. The experiments lasted 30 (1991) and 90 days (1992). We also studied the quality of sediment in snail trails versus sediment without trails, and the abundance and size structure of C. ovalis over a period of 15 months. In both field experiments, C. ovalis affected neither the macroinfaunal structure nor the sediment quality. Significant differences were detected for chlorophyll a content (phytobenthic biomass) when disturbed (trails sediment) versus undisturbed sediment were compared. The highest abundance of C. ovalis (up to 792 ind/m~*) occurred during summer months when the experiments were carried out. It is concluded that the disturbance of intertidal sediment by C. ovalis is quite local and of short duration, a situation which is discussed in connection with physical and biological factors involved in sediment stability and community organization of macroinfaunal assemblages. INTRODUCTION Epibenthic organisms such as decapod crustaceans and gastropod mollusks can play key roles in structuring the sediment and the macroinfauna of soft bottom substrates (Dayton, 1984; Wilson, 1990; Hall et al., 1993). Biolog- ical disturbance can alter the fabric and stability of the sediment (Brenchley, 1981; Posey, 1987) and affect the food availability by depletion of phytobenthic biomass (Connor et al., 1982). Epibenthos can also ingest mac- roinfaunal larvae or adults (MOller, 1986; Hines et al., 1990), bury and kill post-settlement infauna, or induce escape behavior (Ambrose, 1984; DeWitt & Levinton, 1985; Jensen & Jensen, 1985). Manipulative experiments carried out with prosobranch gastropods have shown effects of these organisms on var- ious components of the benthos (see Lopez & Levinton, 1987). Levinton & Stewart (1982) documented negative effects of Ilyanassa obsoleta Say and Hydrobia totteni Morrison on the population growth of the oligochaete Paranais litoralis Miller. The same gastropod species af- fected biomass and metabolism of benthic diatoms (Lev- inton & Bianchi, 1981; Connor et al., 1982), and the standing stock of bacteria adhering to the sediment (Bian- chi & Levinton, 1981). In salt marsh areas of eastern England, Frid & James (1988) found an increase in the abundance of oligochaetes and the polychaete Capitella capitata Fauvel, resulting from removal of the epibenthic gastropod Littorina littorea (L.). In muddy salt marsh ar- eas of Georgia (USA), Pace et al. (1979) found that the effects of /. obsoleta were related more to the ingestion of microorganisms than mechanical changes produced in the substrate. This snail species significantly affected the abundance of recently settled and juvenile meiofauna and macroinfauna in tidal flats of North Carolina, USA (Hunt et al., 1987). Similar results were reported by Wiltse (1980) who studied the effects of the snail Polinices du- plicatus (Say) on the tidal flats of Barnstable Harbor, USA. Knowledge of the ecology of epibenthic gastropods on other coasts is rather scarce. In estuarine areas of south central Chile (ca. 38—40° S), the endemic pulmonate gas- tropod Chilina ovalis Sowerby, 1842, resembles I/yanas- sa obsoleta, in spite of geographical and phylogenetic differences (Brown & Pullan, 1987; Castellanos & Mi- quel, 1991). Chilina ovalis gathers in muddy areas with high nutrient content, i.e., organic matter, and its move- ment and surface deposit-feeding habits produce notice- able changes in the substrate (trails). Deposit feeding in C. ovalis could be a secondary feeding mode, considering that other species of the genus have been described as periphyton consumers above hard sustrates (Miquel, 1986; Bosnia et al., 1990). In the Queule and Lingue River estuaries, C. ovalis coexists with polychaetes and amphipods (Richter, 1985; Bertran, 1989; Quijon & Jaramillo, 1993; Quijon et al., 1996). Seasonal studies carried out on the subtidal pop- ulations of such taxa show that their main periods of re- cruitment occur during the spring-summer months (Bra- vo, 1989; Quijén et al., 1996). Our unpublished data show similar trends for intertidal populations. Feeding trails of C. ovalis alter the surface layer of the sediment to an approximate depth of 0.5 cm where the highest abundance of adults and recruits of the macroinfauna are P. Quiyj6n & E. Jaramillo, 1999 ene, ad on, 2 0.05) among treatments (Tables 2, 3 and Figures 3-6). The results of analyses with sedimentological charac- teristics (Table 4) showed F values between 0.05 and 2.58 during the experiment carried out in 1991, and between 0.02 and 5.58 during 1992. The P values fluctuated be- tween 0.101 and 0.984 in 1991, and between 0.006 and 0.996 in 1992. Thus, only during the second experimental period were significant differences among the treatments detected (Figure 4): the sand content of the control treat- ment was significantly higher than those estimated for inclusion and exclusion sediments (day 8, P = 0.006); and the sand content of the control sediments was higher than that estimated for inclusi6n treatment (day 90, P = 0.037). When the abundance and species-richness of the mac- roinfauna were used for statistical comparisons (Table 3), the F values fluctuated between 0.03 and 2.29 (1991) and between 0.06 and 5.21 (1992). The P values fluctuated between 0.130 and 0.994 during 1991, and between 0.008 and 0.980 during 1992. Thus, as in the case of the sedi- mentological characteristics, only during the second ex- perimental period (1992) did the P values indicate sig- P. Quijon & E. Jaramillo, 1999 Page 75 (X+1s.e.) oD O O abundance-m2 1991 | NDJFMAMJJASO DJ 1992 | 1993 Figure 2 Temporal variability of the population abundance of C. ovalis in the sediments of the study area. No samples were collected in November 1992. nificant difference between the treatments (Figure 6): the abundance of C. capitata in control and wall sediments was significantly higher than that calculated in the exclu- sion sediments (P = 0.008, day 60); the abundance of the Littoridina species was significantly different when inclusion and exclusion sediments were compared (P = 0.023). The macroinfaunal analyses rendered a total of six spe- cies, the most abundant taxa being the ostracod Cyprideis beaconensis Leroy (2-122 ind/4 cm’), the polychaete Capitella capitata Fauvel (1—34 ind/4 cm?), and a species of Littoridina (O—4 ind/4 cm?) (Figures 5, 6). Other spe- cies were occasionally recorded: the polychaetes Perine- reis gualpensis Jeldes, Prionospio (Minuspio) patagonica Augener, and the amphipod Paracorophium hartmanno- rum Andrés. No species composition difference was de- tected when the treatments were compared. The total abundance of the macroinfauna (up to 137 ind/4 cm?) decreased toward the end of the summer of 1991, following the trends of C. beaconensis and C. capi- tata abundance (Figures 5, 6). Species richness (two to four species) did not show temporal variability. Sediment comparisons: In the comparison of sediments with snail trails versus undisturbed sediment, sand, ag- gregates, mud, water, organic matter, and chlorophyll a contents showed some differences in magnitude (Figure 7). However, only the chlorophyll a content was signifi- cantly lower (P = 0.001) in disturbed (87.2 ug/g) versus undisturbed sediment (101.1 ug/g). DISCUSSION Most of the results of the field experiments suggest that C. ovalis does not affect the structure of the intertidal Table 1 Mean sizes in mm of the cohorts of C. ovalis, detected by ‘“‘Modal Progress Analysis’” (MPA). Mean monthly growth rates are also presented. Cohort 1990 Cohort 1991 Cohort 1992 Date Size Growth Size Growth Size Growth November 1991 15.764 — 5.224 — — — December 16.665 0.901 5.448 0.224 — _— January 1992 16.884 0.219 6.902 1.545 — = February 17.020 0.136 7.852 0.950 — a March 18.355 1.335 10.103 2.251 — — April — — 11.908 1.805 — — May — — 12.973 1.065 — — September —- — 16.273 0.825 — = October — — 16.500 0.227 — — December — — 16.712 0.106 5.423 — January 1993 — — 17.593 0.881 7.545 DAQD Page 76 The Veliger, Vol. 42, No. 1 Table 2 Summary of variance analysis carried out with the sedimentological characteristics. Degrees of freedom were 3—12 for the period 1991 and 3-20 for the period 1992. The values of F and P (in parentheses) are given in the columns below each sedimentological characteristic. Asterisks indicate significant difference in sand content among sediments of control versus inclusion treatments (days 8 and 90) (cf. Figure 4). Sand Aggregates Mud Water Org. matter Chlorophyll a 199 1—start 0.59 (0.633) 1.19 (0.362) 1.19 (0.362) 0.74 (0.550) 1.30 (0.325) 2.58 (0.102) day 6 0.43 (0.737) 0.05 (0.984) 0.15 (0.928) 0.69 (0.579) 0.19 (0.900) 0.26 (0.853) day 11 0.24 (0.863) 0.62 (0.616) 0.18 (0.905) 1.57 (0.249) 0.09 (0.965) 2.59 (0.101) day 30 1.32 (0.313) 0.39 (0.762) 0.46 (0.714) 1.23 (0.346) 0.53 (0.672) 1.71 (0.217) 1992-start 0.40 (0.755) 0.59 (0.629) 0.82 (0.500) 0.06 (0.979) 0.62 (0.608) 0.36 (0.783) day 3 0.47 (0.708) 0.20 (0.894) 0.17 (0.919) 1.32 (0.297) 0.19 (0.901) 2.38 (0.100) day 8 5.58 (0.006)* 1.62 (0.217) 0.68 (0.576) 1.44 (0.260) 0.02 (0.996) 0.49 (0.700) day 30 0.49 (0.693) 0.21 (0.888) 0.11 (0.953) 0.71 (0.560) 0.61 (0.620) 1.60 (0.222) day 60 0.28 (0.840) 1.03 (0.400) 1.02 (0.406) 1.18 (0.341) 0.33 (0.802) 0.17 (0.913) day 90 3.55 (0.037)* 2.95 (0.062) 2.06 (0.144) 0.31 (0.821) 0.79 (0.513) 0.45 (0.720) sediment and macroinfauna in the Lingue River estuary. The experiments were set up intuitively expecting the dramatic changes that occur with the manipulation of some species (Paine, 1980). However, these changes did not occur or were trivial, leading to rejection of the hy- pothesis. In addition, the experiments did not detect the existence of confounding effects such as artifacts; i.e., no differences were found between the sediment of the con- trol and wall treatments on any occasion (cf. Hall et al., 1990). The results of the comparisons of sediment with and without feeding trails indicate that C. ovalis indeed af- fects the phytobenthic biomass on a time scale of minutes or hours. However, these effects do not persist longer than one tidal cycle. These results differ from those described by Pace et al. (1979), who detected effects of /lyanassa obsoleta on the phytobenthos only after the third day and lasting until at least 10 days after the start of their ex- periments. Laboratory studies have detected a negative effect on the phytobenthos by a high abundance of I. obsoleta, but a positive effect when abundance of the gastropod was lower (Connor et al., 1982). A similar re- lationship characterized the interaction between Hydrobia totteni and the phytobenthic standing-stock in tidal flats of Long Island, New York (Levinton & Bianchi, 1981). Our experiments were carried out during summer pe- riods characterized by the presence of greater abundance, higher growth rates of gastropods, and more trails on the sediment surface (personal observation). Later in the year because the snails are present in minor abundance , we can expect similar or smaller effects than those detected with these experiments (cf. Hunt et al., 1987; Cammen, 1989; Peterson & Black, 1993). Possible explanations for the absence of effects of C. Table 3 Summary of variance analysis carried out with faunal characteristics. Degrees of freedom were 3—12 for the period 1991 and 3-20 for the period 1992. The values of F and P (in parentheses) are given in the columns below each faunal characteristic. Asterisks indicate significant difference in faunal abundance. Among C. capitata in sediments of the wall and exclusion treatments versus that in control sediments (day 60); and between the abundance of Littoridina sp. in sediments of the exclusion versus that in inclusion sediments (day 90) (cf. Figure 6). C. beaconensis C. capitata Littoridina sp. Total abundance Spp. richness 199 1—start 0.79 (0.523) 0.35 (0.788) day 6 0.44 (0.734) 0.23 (0.870) day 20 0.70 (0.570) 0.03 (0.994) day 30 0.36 (0.787) 2.29 (0.130) 1992-start 0.39 (0.761) 0.93 (0.444) day 3 1.78 (0.184) 0.47 (0.704) day 8 0.37 (0.775) 0.08 (0.972) day 15 0.76 (0.531) 0.37 (0.777) day 30 0.79 (0.515) 0.60 (0.622) day 60 2.14 (0.127) 5.21 (0.008)* day 90 0.60 (0.621) 1.03 (0.401) 0.22 (0.879) 0.51 (0.685) 0.49 (0.699) 0.38 (0.773) 0.31 (0.816) 0.19 (0.896) 0.47 (0.706) 0.46 (0.717) 0.36 (0.781) 0.41 (0.746) 0.45 (0.724) 0.48 (0.705) 1.01 (0.412) 0.46 (0.712) 0.48 (0.703) 1.97 (0.151) 1.01 (0.408) 0.06 (0.980) 3.08 (0.051) 0.22 (0.885) 2.75 (0.070) 0.51 (0.679) 0.55 (0.654) 1.43 (0.264) 0.34 (0.795) 0.76 (0.533) 0.68 (0.578) 1.11 (0.368) 2.88 (0.061) 1.57 (0.228) 3.97 (0.023)* 0.99 (0.416) 0.79 (0.512) Page 77 P. Quijon & E. Jaramillo, 1999 30 days 11 6 start sand LLL exclusion \ \ N _ ry) oad S io) — > : E 2 S) ie) 2 oO fe rs) oO aggregates mu water NAAN inclusion ZL. LLL ISSSY Figure 3 LLL O Co) (8S;+X) % Ul JybIom Sedimentological characteristics resulting from field experiments carried out in the summer of 1991. Asterisks indicate the days when significant differences among the treatments (p < 0.05) were detected (see Table 2). The Veliger, Vol. 42, No. 1 Page 78 60 90 days 30 3 start sand LLL aggregates mud _ @ —_— 2) = YUL hh, Willh LLL (8S,+X) % ul JyBIOM Organic matter LLL chlorophyll a LLL \y \ N exclusion y) A, wall A inclusion | control Sedimentological characteristics resulting from field Figure 4 experiments carried out in the summer of 1992. Asterisks indicate the days when significant differences between the treatments (p < 0.05) were detected (see Table 2). P. Quijon & E. Jaramillo, 1999 start 6 aN} i} = oO t+ le ae Oo vo omen ¢2) (co fe Oy -- (Ss 5 |< Lei So nN i} = ro) Sea o ® Oo Ww S Oo + Cc ix AS fo} a a Ww no = ol hCUWt Cc ix “SS | control i wall Page 79 30 days C. beaconensis C. capitata Littoridina sp. total macroinfauna N N N j inclusion exclusion Figure 5 Abundance of C. beaconensis, C. capitata, the species of Littoridina genus, and total macroinfauna and species- richness resulting from field experiment, carried out in the summer of 1991. Asterisks indicate the days when significant differences between the treatments (p < 0.05) were detected (see Table 3). ovalis include factors that regulate the production of trails and others that inhibit their potential effects. The popu- lation size of C. ovalis and the consequent feeding pres- sure which it exerts could be below the level at which food becomes a limiting resource (see Peterson & Black, 1987). The high availability of nutrients in the area should allow an increase in population size and age reached by individuals (cf. Forbes & Lopez, 1986). How- ever, more than two cohorts were never directly observed simultaneously, while individuals reached an age of at least 18 months. By comparison, in the same genus ages of 2 years and two or more reproductive periods have been described for C. gibbosa in an Argentine freshwater reserve located at a latitude comparable to the Lingue River estuary (Bosnia et al., 1990). Other factors can diminish the possible effects of the Page 80 The Veliger, Vol. 42, No. 1 start 3 8 15 SOE SO) 90 days C. beaconensis . = (S) Se o © cw : Cr C. capitata D+ 5 I< ae) ~~ 0 A Littoridinag sp. N N N o = (S) +3 200 S2 total macroinfauna Oo + Lis a Q (=) ao ney N N CIx N N Z N A. N control wall 4 inclusion N exclusion Figure 6 Abundance of C. beaconensis, C. capitata, the species of Littoridina genus, and total macroinfauna and species- richness resulting from field experiment carried out in the summer of 1992. Asterisks indicate the days when significant differences between the treatments (p < 0.05) were detected (see Table 2). Table 4 Summary of variance analysis carried out with the sedimentological characteristics of sediments in areas with and without trails of C. ovalis. Degrees of freedom were 1—38. The values of F and P (in parentheses) are given in the columns below each sedimentological characteristic. Asterisk indicates significant difference in the contents of chlorophyll a. (cf. Figure 7). Sand Aggregates Mud Water Org. matter Chlorophyll # 3.83 (0.058) 1.42 (0.241) 1.80 (0.187) 3.23 (0.080) 0.03 (0.873) 12.40 (0.001)* P. Quijon & E. Jaramillo, 1999 10 5 2 0 (ep) + Ix 5© (e) ogo = SS ‘ ® = GO 40 20 aggregates mud | control Page 81 90 water (Ss organic matter * chlorophyll a 100 15 50 i trail Figure 7 Sedimentological characteristics in areas with and without trails (control sediments) of C. ovalis. Asterisk indicates the existence of significant differences (p < 0.05). trails and explain the lack of lasting and cumulative ef- fects of C. ovalis. The tides import, resuspend, redistrib- ute, and export sediments (Eisma & Li, 1993), fecal ag- gregates (Risk & Moffat, 1977; Taghon & Jumars, 1984), phytobenthos (de Jonge & van Beusecom, 1995), and or- ganisms of the macroinfauna (Butman et al., 1988a,b). The tides are also associated with bottom diatom prolif- eration and recolonization (Admiraal & Peletier, 1987) on recently altered patches of sediment (trails). The macroinfauna could also be responding to the local Page 82 disturbance exerted by C. ovalis. Capitellid species, for example, can rapidly recolonize recently altered sediment patches, supposedly in response to nutrient availability and the temporary absence of competing species (Tsut- sumi et al., 1990). In addition, other species can escape mechanical disturbance of sediment by burrowing to deeper layers in the sediment (refuge), avoiding the ef- fects of the disturbance (Roberts et al., 1989). Surprisingly, C. ovalis did not affect the texture or wa- ter content of the sediment, which suggests that this spe- cies alters neither the resuspension nor the stability of the substrate (cf. Rhoads & Boyer, 1982). This differs from the findings of Boyer (1980) who demonstrated through- out laboratory experiments that a larger snail (Polinices duplicatus) was able to destabilize the sediment of a tidal flat in Massachusetts, USA. The effects of P. duplicatus were first observed after 24 hours of experimentation and remained until at least 4 days after the snails were ex- cluded. Thus, it seems that C. ovalis does not form part of the biological component that affects water-sediment interaction (see Meadows & Tait, 1989; Paterson, 1989; Paterson & Daborn, 1991), at least in the time period studied. The role of C. ovalis contrasts with what is currently known about gastropods such as [. obsoleta, which in terms of size and habitat appears to be a functionally comparable species. J. obsoleta affects the structure of macroinfaunal (Hunt et al., 1987; Frid & James, 1988), bacterial (Bianchi & Levinton, 1981), and phytobenthic (Pace et al., 1979; Lopez & Levinton, 1987) communi- ties. These differences could result from factors such as dietary flexibility (Feller, 1984), behavioral aspects (Cranford, 1987), and the existence of intrapopulation in- teractions (Levinton, 1985; Forbes & Lopez, 1986). The absence of previous work on the natural history of C. ovalis remains the main obstacle to identifying the role of this species and the degree of its similarity to 1. ob- soleta and other species around the world. ACKNOWLEDGMENTS We thank Juan Carlos Castilla (P. U. Cat6élica de Chile, Santiago), Wolfgang Stotz (U. Catdlica del Norte, Co- quimbo), Carlos Moreno, Milton Gallardo, Mario Pino, Carlos Gallardo, Ramon Formas, Patricia Vallejos (U. Austral de Chile, Valdivia), and Fergus Kennedy (U. of Wales, Bangor) for suggestions on earlier versions of the manuscript. We also thank Victor Poblete, Jacqueline Munoz, Marcia Gonzalez, Luis Rojas, Antonio Low, Gid- eon Russouw, and Maritza Gutiérrez for field assistance. In addition, Ramon Formas provided laboratory facilities for this study. The financial support provided by CONI- CYT (FONDECYT Project 304-90) and Universidad Austral de Chile (DID Project S 92-36) is gratefully ac- knowledged. The Veliger, Vol. 42, No. 1 LITERATURE CITED ADMIRAAL, W. & H. PELETIER. 1987. The seasonal succession of diatom species on an intertidal mudflat: an experimental analysis. Oikos 42:30—40. AMBROSE, W. G. JR. 1984. Increased emigration of the amphipod Rhepoxynius abronius (Barnard) and the polychaete Nephtys caeca (Fabricius) in the presence of invertebrate predators. Journal of Experimental Marine Biology and Ecology 80: 67-75. ANDERSON, FE L. BLAck, L. M. MAYER & L. WATLING. 1981. A temporal and spatial study of mud flat texture. North Eastern Geology 3:184—-196. BERTRAN, C., 1989. Zonaci6n y dinamica temporal de la ma- croinfauna intermareal en el estuario del Rio Lingue (Val- divia, Chile). Revista Chilena de Historia Natural 62:19—32. BIANCHI, T. S. & J. S. LEVINTON. 1981. Nutrition and food limi- tation of deposit feeders. II: Differential effects of Hydrobia totteni and Ilyanassa obsoleta on the microbial community. Journal of Marine Research 39:547—556. Bosnia, A. S., FE J. Katsin & A. TaBLapo. 1990. Population dynamics and production of the freshwater snail C. gibbosa Sowerby 1842 (Chilinidae, Pulmonata) in a North Patagon- ian reservoir. Hydrobiologia 190:97—110. Boyer, L. 1980. Production and preservation of surface traces in the intertidal zone. Ph.D. Dissertation. University of Chica- go, Chicago. 248 pp. BRAvo, A. 1989. Variaci6n temporal de la macroinfauna sub- mareal del estuario del Rio Queule (IX Region, Chile). M.Sc. Thesis. Universidad Austral de Chile, Valdivia. 74 pp. BRENCHLEY, G. A. 1981. Disturbance and community structure: an experimental study of bioturbation in marine soft bottom environments. Journal of Marine Research 39:767—790. Brown, D. S. & N. B. PULLAN. 1987. Notes on the shell, radula and habitat of Chilina (Basommatophora) from the Falkland Islands. Journal of Molluscan Studies 53:105—108. ButMaNn, C. A., J. P. GRASSLE & E. J. BUSKEY. 1988a. Horizontal swimming and gravitational sinking of Capitella sp. I (An- nelida: Polychaeta) larvae: implications for settlement. Ophelia 29:43—57. BuTMANn, C. A., J. P. GRASSLE & C. M. WEBB. 1988b. Susbstrate choices made by marine settling in still water and in a flume flow. Nature 333:771—773. CAMMEN L. M. 1989. The relationship between ingestion rate of deposit feeders and sediment nutritional value. Pp. 156—167 in G. Lopez, G. Taghon & J. S. Levinton (eds.), Ecology of Marine Deposit Feeders. Springer Verlag: New York. CASTELLANOS, Z. A., DE & S. E. MIQUEL. 1991. Distribuci6n de los Pulmonata Basommatophora. Pp. 1-11 in R. A. Riguelet (dir), Fauna de agua dulce de la Republica Argentina; Mol- lusca, gastropoda. FECIC: Buenos Aires. Connor, M. S., J. M. TEAL & I. VALIELA. 1982. The effect of feeding by mud snails, //yanassa obsoleta, on the structure and metabolism of a laboratory benthic algal community. Journal of Experimental Marine Biology and Ecology 65: 29-45. CRANFORD, P. J. 1987. Behaviour and ecological importance of a mud snail (//yanassa obsoleta) population in a temperate macrotidal estuary. Canadian Journal of Zoology 66:459— 466. Day, R. W. & G. P. QUINN. 1989. Statistical comparisons of the treatments after analysis of variance. Ecological Mono- graphs 59:430—460. Dayton, P. K. 1984. Processes structuring some marine com- munities: are they general? Pp., 181-197 in D. R. Strong, P. Quijén & E. Jaramillo, 1999 Page 83 D. Simberloff, J. Abele & A. B. Thistle (eds.), Ecological Communities: Conceptual Issues and the Evidence. Prince- ton University Press: Princeton. De JoncE, V. N. & J. E. E. VAN BEUSEKOM. 1995. Wind- and tide-induced resuspension of sediment and microphytoben- thos from tidal flats in the Ems estuary. Limnology and Oceanography 40:766—778. DeWitt, T. H. & J. S. LEVINTON. 1985. Disturbance, emigration and refugia: how the mud snail, //yanassa obsoleta (Say), affects the habitat distribution of an epifaunal amphipod, Microdeutopus gryllotulpa (Costa). Journal of Experimental Marine Biology and Ecology 92:97-103. EIsMA D. & A. Li. 1993. Changes in suspended-matter floc size during the tidal cycle in the Dollard estuary. Netherlands Journal of Sea Research 31:107—117. FELLER, R. J. 1984. Dietary immunoassay of //yanassa obsoleta, the eastern mud snail. Biological Bulletin 166:96—102. ForBes, V. E. & G. R. Lopez. 1986. Changes in feeding and crawling rates of Hydrobia truncata (Prosobranchia: Hydro- biidae) in response to sedimentary chlorophyll-a and re- cently egested sediment. Marine Ecology Progress Series 33: 287-294. Frip, C. L. J. & R. JAMEs. 1988. Interactions between two species of saltmarsh gastropod, Hydrobia ulvae and Littorina litto- rea. Marine Ecology Progress Series 43:173-179. GayaNniLo, F.C. Jr., M. SorIANO & D. PAULy. 1989. A Draft Guide to the Compleat ELEFAN. International Center for Living Resources Management: Manila. 70 pp. HALL, S. J., D. RAFFAELLI & W. R. TURREL. 1990. Predator-caging experiments in marine systems: a reexamination of their val- ue. American Naturalist 136:657—672. HALL, S. J., M. R. ROBERTSON, D. J. BASFORD & R. FRYER. 1993. Pit digging by the crab Cancer pagurus: a test for long- term, large-scale effects on infaunal community. Journal of Animal Ecology 62:59—66. Hines, A. H., A. M. HADDON & A. WIECHERT. 1990. Guild struc- ture and foraging impact of blue crabs and epibenthic fish in a subestuary of Chesapeake Bay. Marine Ecology Pro- gress Series 67:105—126. Hunt, J. H., W. G. AMBROSE JR. & C. H. PETERSON. 1987. Effects of the gastropod, /lyanassa obsoleta (Say), and the bivalve, Mercenaria mercenaria (L.), on larval settlement and juve- nile recruitment of infauna. Journal of Experimental Marine Biology and Ecology 108:229—240. HUuRLBERT, S. H. 1984. Pseudoreplication and the design of eco- logical field experiments. Ecological Monographs 54:187— Ai ite JENSEN, K. T. & J. N. JENSEN. 1985. The importance of some epibenthic predators on the density of juvenile benthic mac- rofauna in the Danish Wadden Sea. Journal of Experimental Marine Biology and Ecology 89:157—174. LEVINTON, J. S. 1985. Complex interactions of a deposit feeder with its resources: roles of density, a competitor, and detrital addition in the growth and survival of the mudsnail Hydro- bia totteni Marine Ecology Progress Series 22:31—40. LevINTON, J. S. & T. S. BIANCHI. 1981. Nutrition and food limi- tation of deposit feeders. I. The role of microbes in the growth of mud snails (Hydrobiidae). Journal of Marine Re- search 39:531—544. LEvINTON, J. S. & S. STEWART. 1982. Marine succession: the effect of two deposit-feeding gastropod species on the pop- ulation growth of Paranais litoralis (Miiller, 1784) (Oligo- chaeta) Journal of Experimental Marine Biology and Ecol- ogy 59:231-241. Lopez, G. R. & J. S. LEVINTON. 1987. Ecology of deposit-feeding animals in marine sediments. Quarterly Review of Biology 62:235-260. MEapows, P. S. & J. Tait. 1989. Modification of sediment per- meability and shear strength by two burrowing invertebrates. Marine Biology 101:75—82. MIQUEL, S. E. 1984. Ovoposiciones de pulmonados neotropicales (Moll. Basom.: Chilinidae, Physidae y Ancylidae). Revista Museo de La Plata (Zoologia) 13:249—256. MIQUEL, S. E. 1986. El ciclo de vida y la evoluci6n gonadal de Chilina fluminea fluminea (Maton, 1809) (Gastropoda Ba- sommatophora, Chilinidae). Neotropica 33:23—34. MOLLER, P. 1986. Physical factors and biological interactions reg- ulating infauna in shallow boreal areas. Marine Ecology Pro- gress Series 30:33—47. Pace, M. L., S. SHIMMEL & W. M. Dar_ey. 1979. The effect of grazing by a gastropod Nassarius obsoletus on the benthic microbial community of a salt marsh mudflat. Estuarine Coastal & Marine Science 9:121—134. PAINE, R. T. 1980. Food webs: linkage, interaction strength, and community infrastructure. Journal of Animal Ecology 49: 667-685. PATERSON, D. M. 1989. Short term changes in the erodability of intertidal cohesive sediments related to migratory behaviour of epipelic diatoms. Limnology and Oceanography 34:223- 234. PaTERSON, D. M. & G. R. DABOoRN. 1991. Sediment stabilisation by biological action: significance for coastal engineering. Pp. 111-119 in D. M. Paterson & G. R. Daborn (eds.), Devel- opments in Coastal Engineering. University of Bristol: Bris- tol. PETERSON, C. H. & R. BLACK. 1987. Resource depletion by active suspension feeders on tidal flats: Influence of local density and tidal elevation. Limnology and Oceanography 32:143— 166. PETERSON, C. H. & R. BLACK. 1993. Experimental tests of the advantages and disadvantages of high density for two co- existing cockles in a Southern Ocean lagoon. Journal of An- imal Ecology 62:614—633. Posey, M. H. 1987. Influence of relative mobilities on the com- position of benthic communities. Marine Ecology Progress Series 39:99—104. Quon, P. & E. JARAMILLO. 1993. Temporal variability in the intertidal macroinfauna in the Queule river estuary, south- central Chile. Estuarine Coastal and Shelf Science 37:655— 667. QuuON, P., E. JARAMILLO & M. PINO. 1996. Macroinfaunal as- semblages associated to mussel and clam beds in an estuary of southern Chile. Estuaries 19:62—74. REISE, K. 1985. Tidal flat ecology. An Experimental Approach to Species Interactions. Ecological Studies, 54. Springer- Verlag: Berlin. 191 pp. Ruoaps, D. C. & L. EF Boyer. 1982. The effects of marine ben- thos on physical properties of sediments. A successional per- spective. Pp. 3-52 in P. L. McCall & M. J. S. Tevesz (eds.), Animal-Sediment Relationships: The Biogenic Alteration of Sediments. Plenum Press: New York. RICHTER, W. 1985. Distribution of the soft-bottom macroinfauna in an estuary of southern Chile. Marine Biology. 86:93—100. Risk, M. J. & J. S. MorraT. 1977. Sedimentological significance of fecal pellets of Macoma balthica in the Minas Basin Bay of Fundy. Journal of Sedimentary Petrology 47:1425—1436. Roserts, D., D. RitrscHor, D. J. GERHART, A. R. SCHMITD & L. G. HILL. 1989. Vertical migration of the clam Mercenaria Page 84 mercenaria (L.) (Mollusca: Bivalvia): Environmental cor- relates and ecological significance. Journal of Experimental Marine Biology and Ecology 126:271—280. SOKAL, R. R. & FE J. ROLHF. 1969. Biometria. H. Blume Edi- ciones: Madrid. 832 pp. STRICKLAND, J. D. H. & T. R. Parsons. 1972. A Practical Hand- book of Seawater Analysis. Fisheries Research Board of Canada. Bulletin 167: Ottawa. 310 pp. TAGHON, G. L. & P. A. JumMarRs. 1984. Variable ingestion rate and its role in optimal foraging behavior of marine deposit feed- ers. Ecology 65:549-558. The Veliger, Vol. 42, No. 1 TsuTsuMI, H., S. FUKUNAGA, N. Fusita & M. Sumipa. 1990. Re- lationship between growth of Capitella sp. and organic en- richment of the sediment. Marine Ecology Progress Series 63:157-162. WILSON, W. H. Jr. 1990. Competition and predation in marine soft sediment communities. Annual Review of Ecology and Systematics 21:221—241. WILTSE, W. I. 1980. Effects of Polinices duplicatus (Gastro- poda, Naticidae) on infaunal community structure at Barnstable Harbor, Massachusetts, U.S.A. Marine Biolo- gy 56:301-310. The Veliger 42(1):85—96 (January 4, 1999) THE VELIGER © CMS, Inc., 1999 Designation of Lectotype for Haliotis crebrisculpta Sowerby, 1914, with a Discussion of H. clathrata Reeve, 1846 (non Lichtenstein, 1794) KATHARINE A. STEWART 19 La Rancheria, Carmel Valley, California 93924, USA AND DANIEL L. GEIGER* Department of Biological Sciences, University of Southern California, Los Angeles, California 90089-0371, USA Abstract. The three syntypes of Haliotis crebrisculpta Sowerby, 1914, belong to two species. The figured specimen of Sowerby (BMNH 1919.12.31.19) is designated as lectotype. This species remains known from a single shell from New Caledonia. It is characterized by the eccentric apex (1/7.2 from the posterior margin of the shell) of the oblong shell and its strong spiral cords bearing numerous scales. The other two specimens in the syntype series of H. crebri- sculpta are referable to H. clathrata Reeve, 1846 (non Lichtenstein, 1794). Haliotis clathrata Reeve has been poorly identified in the literature. It is redescribed on shell, radular, and epipodial characters. Haliotis clathrata is not a sub- species of H. rubra Leach, 1814, but the fossil H. tuvuthaensis Ladd in Ladd & Hofmeister, 1945, is synonymized under it. The species ranges from Madagascar to American Samoa and from southern Japan to the Sydney area of southeastern Australia. INTRODUCTION In the family Haliotidae Rafinesque, 1815, approximately 200 species-level taxa have been described, of which 55 species with 11 subspecies are thought to be valid; an encompassing treatment of all those taxa was provided by Geiger (1998a). Several taxonomic issues have re- cently been addressed (Stewart, 1984; Herbert, 1990; Geiger, 1996, 1998b), but some uncertainties still await treatment. Here we deal with one source of confusion at the species level. The clarification of the taxonomy in this family is necessary for a forthcoming cladistic analysis of the family by DLG. The identity of Haliotis crebrisculpta Sowerby, 1914, has been controversial in the literature. One specimen of the syntype series is here designated as lectotype of H. crebrisculpta. The other two specimens are identified as Haliotis clathrata Reeve, 1846. As these three specimens are syntypes, the designation of a lectotype is necessary. In the following, H. clathrata refers to the taxon so- named by Reeve (1846) and not the overlooked, senior homonym of Lichtenstein (1794) discussed below and elsewhere (Geiger, 1998b; Geiger & Stewart, in press), unless specifically indicated. The variable H. clathrata has also been labeled H. crebrisculpta, causing much confusion. Those specimens with more pronounced spiral sculpture tended to be identified as H. crebrisculpta, whereas those with more or less radial ridges along the * Corresponding author: e-mail: dgeiger @scf.usc.edu growth lines have rather been called H. clathrata. Kuroda & Habe (1952) went as far as to synonymize H. crebri- sculpta with H. clathrata. The synonymy of H. clathrata with Haliotis rubra Leach, 1814, first suggested by Sow- erby (1882), is without merit as discussed below. Due to the long-standing confusion with respect to the identity of H. clathrata, we provide here a detailed re- description of the shell, report for the first time the mor- phology of the epipodium and the radula, and discuss the geographical distribution. MATERIALS anD METHODS The radulae of H. clathrata from dry specimens, which were rehydrated in 75% ethanol, were isolated in 4 M NaOH; the NaOH solution was changed daily until the body was completely dissolved. For wet-preserved spec- imens, the radula was dissected out, washed in water, cleaned with 4 M NaOH overnight, washed twice in wa- ter, dehydrated through two washes in 100% ethanol, mounted while drying in air on double-sided carbon ad- hesive (Ted Pella 16084—2) on a coin of 24 mm diameter, which in turn was mounted on a Cambridge stub with colloidal graphite, sputter-coated with gold, and viewed at an accelerating voltage of 10 kV and a probe current of 200 pA on a Cambridge 360 scanning electron micro- scope using the secondary electron detector. Epipodial characters from four specimens of H. clath- rata were assessed. A representative piece was cut from the preserved animal, washed in 100% ethanol, and trans- Page 86 The Veliger, Vol. 42, No. 1 ferred into 100% hexamethyldisilizane (HMDS: Polysci- ence). After 1 day the HMDS was changed and the next day the sample was air dried at room temperature. The specimen was mounted with colloidal graphite on a coin (see above). It was either sputter-coated, and viewed at an accelerating voltage of 2—5 kV and a probe current of 100—200 pA, or drawn with a camera lucida attached to a dissecting microscope. Statistical analysis was performed with Statistica™ for Macintosh 4.1 (StatSoft, 1994). Specimens from the following collections were exam- ined: AMNH, American Museum of Natural History, New York; ANSP, Academy of Natural Sciences, Phila- delphia; BMNH, The Natural History Museum, London; CASIZ, California Academy of Science, Invertebrate Zo- ology, San Francisco; DLG, D. L. Geiger collection, Los Angeles; DMNH, Delaware Museum of Natural History, Wilmington; KAS, K. A. Stewart collection, Carmel, Cal- ifornia; LACM, Los Angeles County Museum of Natural History; MNHN; Muséum Nationale d’ Histoire Naturelle, Paris; NM, Natal Museum, Pietermaritzburg, South Af- rica; NMW, National Museum of Wales, Cardiff, Wales; RP, Roger Pickery collection, Wilnjk, Belgium; SBMNH, Santa Barbara Museum of Natural History, California; UCMP, University of California Museum of Paleontolo- gy, Berkeley; USNM, National Museum of Natural His- tory, Smithsonian Institution, Washington, D. C. SYSTEMATICS Haliotis crebrisculpta Sowerby, 1914 (Figure 1A) H.. crebrisculpta Sowerby, 1914:—Sowerby, 1914: 478, pl. 14, fig. 2.—Kaicher, 1981: card no. 2878. Non H. crebrisculpta (Misidentifications of H. clathrata):— Talmadge, 1963: 137, pl. 14, fig. 1—Huinton, 1972: 1, fig. 4.—Hinton, 1978: 2, fig. 12.—Anon., 1975: 5.— Abbott & Dance, 1983: 22.—Dharma, 1988: pl. 1, fig. 5.—Wilson, 1993: 48, pl. 5 fig. 9 A, B—Pickery & Steppe, 1995: pl. 5, fig. 4. Identity of the three syntypes: Specimen BMNH 1919.12.31.19 (Figure 1A) agrees with the original illus- tration. The other two specimens NMW 1955.158.608 (Figure 1B) and USNM 341787 (Figure 1C) are identified as H. clathrata (see below for details). Designation of lectotype: Sowerby (1914) figured only one specimen in his description of H. crebrisculpta, in which he did not mention the number of specimens on which his description was based. He sold two specimens specifically as co-types (= syntypes) of this species (S. Greenhouse, personal communication; A. Kabat, personal communication), one of which is now housed in the NMW, the other in the USNM. The specimen in the BMNH is marked ‘‘type’”’ in Sowerby’s hand (K. Way, personal communication) and corresponds precisely with the figure of Sowerby (1914), particularly in the spiral sculpture with its numerous fine lamellae, and the place- ment of the apex. The description of the shell is some- what ambiguous as to which shell was being described, possibly owing to the fact that the taxon had been based on three non-conspecific syntypes. The most striking dis- crepancy is found in the number of open perforations. The original description mentioned four open perfora- tions; the specimen in the BMNH has seven, the one in the USNM has five, and the one in the NMW has four. New Caledonia as the type locality applies for all the three syntypes. The figured specimen (BMNH 1919.12.31.19) is here designated as the lectotype for H. crebrisculpta Sowerby, 1914. This species has not been recollected and remains known from a single shell. Particularly, no specimen could be found in the old or new holdings of the MNHN, known for its long-standing collection effort at the type locality—New Caledonia—through the ORSTOM-pro- gram. We recognize that the types of some taxa in the Haliotidae are aberrant specimens of other species. How- ever, no similar case can be made for H. crebrisculpta due to lack of indications such as a strong growth mark from an injury, irregularities in the nacre, or a lateral shift of the selenizone. The specimen is not a hybrid, as the characters are not in between any two known species, but occupy a peripheral position in the morphospace of the family. The lectotype is sufficiently distinct to warrant recognition at the species level. It is characterized by its very eccentric apex, oblong shell, and the sculpture con- sisting of scabrous scales on the strong spiral cords (Table 1). Such a combination of characters is not found in any other abalone species. Redescription of lectotype (Figure 1A): Shell small (30 mm), oblong, somewhat convex. Apex eccentric at 1/7.2 from posterior margin. Spire entirely hidden under nar- row columella in ventral view. Tremata large, somewhat oval, moderately raised, seven open. Dorsal surface with numerous spiral cords of variable strength. Spiral cords with fine, tightly spaced lamellae, forming stronger, up- ward directed scales at irregular intervals. Stronger scales in more or less radial rows. Color uniform yellow ochre, dull. Nacre milky, dull. No muscle scar. Comparisons: Haliotis clathrata Reeve, 1846 (Figures 1A, B, 2, Table 1). The spiral cords in H. clathrata are never as broad and elevated, and never bear scrabrous scales such as found in H. crebrisculpta. The apex is more centrally located in H. clathrata: approximately 1/ 3.8 versus 1/7.2 from the posterior margin; this compar- ison is based on shells of similar size (30 mm; Figure 4). Consequently, the shell of H. crebrisculpta is a somewhat more oblong than that of H. clathrata. Whitehead (1982) tentatively synonymized H. dissona (Iredale, 1929) with H. crebrisculpta. As we are unclear on what species concept of H. crebrisculpta had been K. A. Stewart & D. L. Geiger, 1999 Page 87 Figure 1 A. Lectotype of Haliotis crebrisculpta. New Caledonia. BMNH 1919.12.31.19. Length: 30 mm. B. H. clathrata, specimen from syntype series of H. crebrisculpta. New Caledonia. NMW 1955.158.608. Length: 30 mm. C. H. clathrata, specimen from syntype series of H. crebrisculpta. New Caledonia. USNM 341787. Length: 32 mm. used by Whitehead (1982), we could only speculate on the validity of the statement. We agree with Talmadge (1961) that the status of H. dissona is unresolved, because the type specimen is badly worn and fairly small (22 mm). Haliotis dohrniana Dunker, 1863 (Figure 3A). This species has been described from New Caledonia and is here tentatively recognized. It has a broad, scattered dis- tribution in the western Pacific, but is a rare, poorly known species. It has a similar overall shape of the shell, but the apex is somewhat less eccentric. The sculpture of H. dohrniana does not include scabrous scales, but fine spiral cords with some occasional larger bumps, the latter not being found in H. crebrisculpta. Page 88 Table 1 The Veliger, Vol. 42, No. 1 Distinguishing shell characters among Haliotis clathrata, H. crebrisculpta, and H. rubra. Character Post-juvenile shape Shell convexity Aperture Spiral threads Thread thickness Sculpture on threads Apex quotient Spire in ventral view Maximum size Color elements Distribution E-W Distribution N-S Haliotis clathrata oblong flat to moderate + straight fine equal fine to very fine lamellae 1/3 to 1/4.5 mostly visible 4 cm irregular mottling E-Africa—W Pacific Okinawa—Sydney Haliotis crebrisculpta oblong strong + straight thick, square cords alternating strong, scabrous scales 1/7.2 hidden by columella 3 cm ? New Caledonia New Caledonia Haliotis rubra round flat to moderate rounded absent to fine irregular very fine lamellae 1/3.3 fully exposed 16 cm prosocline, off-white flammae S coast of Australia S Queensland—Tasmania Haliotis squamosa Gray, 1826 (Figure 3B). For a re- cent account of this species see Stewart (1984). The very eccentric apex and the lamellae on the spiral cords form- ing elevated scales are common features of H. crebri- sculpta and H. squamosa. However, these scabrous scales appear in an almost random fashion in H. squamosa, whereas they are arranged in more or less radial rows in H. crebrisculpta. The aperture is curved in H. squamosa Figure 2 A. Holotype of Haliotis clathrata. Baclayon, Island of Bohol, Philppines. BMNH. Length: 23.6 mm. B. Haliotis clathrata. Gladstone, Queensland, Australia. Collection R. Pickery, Wilrijk, Belgium. Length: 35 mm. K. A. Stewart & D. L. Geiger, 1999 Page 89 V =< 4 ae ext Figure 3 A. Haliotis dohrniana. No locality. NNW. Length: 25 mm. B. Haliotis squamosa. Dorsal: Holotype. “Australia.” NMW. 76 mm. Ventral. Between Fort Dauphin and Monantenina, Madagascar. DLG. Length: 83 mm. C. Haliotis venusta. Holotype. “‘Eastern Seas.”” BMNH. Length: 38 mm. Page 90 Apex quotient y = 2.6409 + 0.04245x r = 0.68698 p<10%-7 n=70 5 10 115 20 25 30 35 40 45 Shell length (mm) Figure 4 Regression with 95% confidence lines of shell length and apex quotient (shell length/distance of apex from posterior margin of the shell) for Haliotis clathrata. The statistical significance of the regression analysis indicates allometric change in the position of the apex to a more eccentric position. The Veliger, Vol. 42, No. 1 but straight in H. crebrisculpta. Moreover, H. squamosa has been found only in a restricted part of southern Mad- agascar with upwelling conditions, whereas H. crebri- sculpta has a restricted occurrence in tropical New Cal- edonia. Geographic distribution: The distribution of H. crebri- sculpta is limited to the type locality of New Caledonia. Haliotis clathrata Reeve, 1846 (non Lichtenstein, 1794) (Figures 1B, C, 2, 4-7) H. clathrata Reeve, 1846:—Reeve, 1846: species 71, fig. 72.—Weinkauff, 1883: 35-36, pl. 29, fig. 7.—Pilsbry, 1890: 117, pl. 5, fig. 26.—Delhaes, 1909: 28-29, fig. 18.—Ladd, 1966 (fossil): 26, pl. 2, figs. 3—-5.—Gosliner et al., 1996: 125, fig. 428. H. tuvuthaensis Ladd in Ladd & Hofmeister, 1945 (fossil): —Ladd & Hofmeister, 1945: 351, pl. 50, figs. E, E As H. coccoradiata Reeve, 1846:—Salvat et al., 1988: pl. 1, fig. 4. As H. crebrisculpta Sowerby, 1914:—Talmadge, 1963: 137, pl. 14, fig. 1.—Hinton, 1972: 1, fig. 4.—Hinton, 1978: 2, fig. 12.—Anon., 1975: 5.—Abbott & Dance, 1983: p. 22.—Dharma, 1988: pl. 1, fig. 5 Wilson, 1993: 48, pl. 5 figs. 9 A & B.—Pickery & Steppe, 1995: pl. 5, fig. 4. As H. crebrisculpta auct. non Sowerby:—Kaicher, 1981: card no. 2832, holotype; card no. 2879. As H. gemma Reeve, 1846:—Pickery & Steppe, 1995: pl. Dy 1 2. As H. rubra Leach, 1814:—Boone, 1938: 297-298, pl. 113. As H. rubra clathrata Leach, 1814:—Talmadge, 1957: 59— 60. History: Haliotis clathrata is well represented in collec- tions. The holotype (Figure 2A) and two paratypes are in the BMNH. Aaliotis clathrata Lichtenstein, 1794, is an overlooked, senior homonym (Geiger, 1998a, b); we have petitioned the International Commission on Zoological Nomenclature to suppress this name in order to preserve the names of two valid species (Geiger & Stewart, in press). Most records mentioning H. crebrisculpta actually referred to H. clathrata (e.g., Higa, 1983; Stewart, 1986) because all purported illustrations of H. crebrisculpta not involving the here selected lectotype, actually show H. clathrata. The figured specimen labeled H. crebrisculpta in Talmadge (1961) cannot be identified with certainty, but most likely also represents H. clathrata. Kaicher (1981) apparently noted the two different species united under H. crebrisculpta and used two cards (nos. 2878, 2879) for this taxon, one as “‘Haliotis crebrisculpta Sow- erby”’ figuring the here selected lectotype BMNH 1919.12.31.19, and one as “°H. crebrisculpta auct, non Sowerby” figuring a specimen conspecific to the syntype specimens in the USNM and the NMW, i.e., H. clathrata. The synonym Haliotis tuvuthaensis Ladd in Ladd & Hof- meister, 1945, is discussed below under fossil record. Shell (Figures 1B, C, 2A, B): Shell auriform oblong, flat to moderately convex. Maximum size 4 cm. Apex ap- proximately 1/2.4 to 1/5.4 from posterior margin. Apex more eccentric with increasing shell length (Figure 4). Spiral cords three to four, often more prominent, regularly spaced between suture and tremata. Growth lines fine; every four to five one stronger one, often raised as radial lamella. Lamella from suture but never reaching to tre- K. A. Stewart & D. L. Geiger, 1999 Figure 5 Epipodium of Haliotis clathrata. Based on specimen from Com- oros, DLG. Scale bar = 1 mm. mata. Tremata large, slightly oval, somewhat raised, usu- ally four to five open. Spire usually visible in ventral view, sometimes partially obscured by columella of mod- erate width. Base flat. Nacre brilliant, showing grooves of spiral sculpture. No muscle scar. Coloration variable. Red tones from bright orange, dull brick red, Bordeaux red to light brown; some also green (particularly specimens from Stanage Bay, Queensland). Uniformly colored specimens rare. Typically with mark- ings in green and off-white, sharp margins and angles, often serrated, often with additional fine, spiral, stippled lines in contrasting color. Markings in prosocline oblique radial pattern, normally not as flammae. Banding pattern in contrasting coloration always between row of tremata and columella. Width of bands between half and full length between two tremata. For additional color illustra- tions please refer to ‘http://nhm.org/~dgeiger/clathrata. html’. Animal: Epipodium (Figure 5) simple for genus. Only dorsal fingered structures and ventral tentacles. Undulat- ing epipodial fold absent, face bare of any structures. Radula (Figure 6) (for terminology see Geiger, 1996). Cutting edge of rachidian and lateral tooth 1 slightly bent to posterior. Primary ridge of lateral tooth 1 convex, con- tinuous with cutting edge. Secondary ridge attached to main part at approximately % to cutting edge forming angle of approximately 45° with primary ridge. Lateral teeth 3 and 4 with single denticle on outer margin of cusp. Lateral tooth 5 with one to four denticles. Cusps of mar- ginal teeth symmetrically denticulated. Page 91 Habitat: The species has been found from 0 to 75 m depth, most frequently between 2 and 15 m. It is normally associated with coral reefs and lagoons. Comparisons: The distinction between H. clathrata and H. crebrisculpta is discussed above under the latter spe- cies and is summarized in Table 1. Haliotis clathrata is very similar to Haliotis venusta Adams & Reeve, 1848 (Figure 3C), in respect to the over- all shape of the shell, the numerous spiral cords, the large and somewhat elevated tremata, and the type locality be- ing ‘“‘Eastern seas” (Adams & Reeve, 1848: 69). How- ever, no trace of the radial lamellae can be found. Ma- terial resembling the type specimens of H. venusta is very rare, and we have not found any intermediate specimens of H. clathrata and H. venusta. We adopt here a conser- vative position and prefer to keep these two taxa separate until more material of H. venusta becomes available. Hal- iotis venusta has been illustrated in the following publi- cations: Adams & Reeve, 1848: 69, pl. 23, figs. 5 a, b; Weinkauff, 1883: pl. 29, fig. 3; Sowerby, 1882: fig. 55; Kaicher, 1981: card 2843. Haliotis clathrata Reeve, 1846, is not a juvenile H. rubra Leach, 1814; for illustrations of H. rubra see Hin- ton (1978) and Abbott & Dance (1983) [both as H. rub- er]; Hinton (1978), Wells & Bryce (1985), Wilson (1993) [all as H. conicopora]. The spelling H. ruber in some works is incorrect for H. rubra: the adjectival species epithet must be inflected to the feminine of the genus Haliotis. Sowerby (1882:31), Pilsbry (1890:117), Cotton & Godfrey (1933), and Cotton (1959) referred to H. clathrata as a young H. naevosa Martyn, 1784, which is an unavailable synonym (ICZN, 1957: Opinion 456) of H. rubra. Boone (1938) and Wagner & Abbott (1978) repeated the synonymy between H. clathrata and H. rub- ra, and Talmadge (1957, 1963) called the former a variety of the latter. Weinkauff (1883:35) first listed H. clathrata as a juvenile H. rubra (as H. naevosa), but revised his opinion on pages 80 and 83 (‘“‘Nachtrage und Berichtig- ungen’’: postscripts and rectifications), and stated that the two taxa are not synonymous, but valid species. Delhaes (1909:29) explicitly contradicted the opinion of Pilsbry (1890) that H. clathrata represented a juvenile H. rubra, and separated H. clathrata from H. rubra (as H. naevosa) giving a differential diagnosis for the two species. Wilson (1993) also questioned the synonymy between the two species, and agreed with Whitehead (1982) that H. clath- rata needed further study. Haliotis rubra is a temperate species distributed from Jervis in New South Wales to Freemantle in southern Western Australia, and in Tasmania (Figure 7) (Wells & Bryce, 1985; Ludbrook & Gowlett-Holmes, 1989; Prince & Shepherd, 1992; Wilson, 1993); it includes records of H. conicopora Péron, 1816, which has been shown to be only a variation/ecomorph of H. rubra (Brown, 1993; Lee & Vacquier, 1995). Haliotis clathrata, however, is found The Veliger, Vol. 42, No. 1 Figure 6 Radula of Haliotis clathrata. Based on specimen from Comoros, DLG. A. Rachidian tooth and lateral tooth 1. Scale bar = 50 pm. B. Lateral teeth 3 to 5 and innermost marginal teeth. Scale bar = 50 wm. C. Intermediate marginal teeth. Scale bar = 25 pm. D. Outer marginal teeth. Scale bar = 5 wm. mostly in the tropical Pacific (Figure 7). Haliotis rubra is a large species reaching approximately 16 cm in di- ameter, and is commercially harvested (Prince & Shep- herd, 1992); a young specimen (41.3 mm) was described as Haliotis ancile Reeve, 1846. The shell of H. clathrata is much smaller, growing only to approximately 4 cm. The surface of the shell in H. rubra may either be entirely flat, or it may be undulated, but it never has any lamellae as in H. clathrata. The growth lines are always fine and evenly spaced in H. rubra, but H. clathrata shows the pattern described above. Haliotis rubra retains its round form at all growth stages, whereas H. clathrata usually is more or less round as a juvenile (< 15 mm) but gets more elongated as it grows larger (Figure 4). This direc- tional change of shape is an invariant character of abalone with oblong shells. A latitudinal gradient in shell shape is equally unlikely. Species with an extensive range (e.g., H. varia Linnaeus 1758) do not have such a changes in shape. Additionally, no species in the Haliotidae has a distribution spanning from tropical to temperate faunal regions. Thus, there is no evidence that H. clathrata is a juvenile of H. rubra. The distinguishing characters be- tween H. clathrata and H. rubra are summarized in Table 1. Fossil records: Fossil specimens of H. clathrata have been found in the Indo-Pacific, by Ladd (1966) in the Pliocene and Pleistocene of Guam and in the Miocene of Tinian in the Mariana Group, and by Ladd & Hofmeister (1945) in the lower Miocene of Fiji. The latter report described the single specimen at hand as H. tuvuthaensis. It is distinguished by the author through the absence of radial folds in the internal mold; otherwise H. tuvuthaen- sis agrees with H. clathrata (this study). As the stronger growth lines in H. clathrata do not always form radial lamellae (see Figures 1B, C), H. tuvuthaensis is a form of H. clathrata, and is here synonymized with it. Ladd (1966) compared H. tuvuthaensis to Haliotis ovina Gme- lin, 1791. H. ovina is much rounder in its general shape, has a broader columella and more elevated tremata, hence, can be clearly separated from H. clathrata. Talmadge (1963) pointed out the similarity between the fossil Haliotis powelli described by Fleming (1952) from the Pleistocene of New Zealand and H. clathrata (as H. crebrisculpta). We do not agree with Talmadge’s (1963) opinion as H. powelli is a rather flat species and the gen- eral shape of the shell is more rounded; particularly the appertural rim is distinctly curved outward (Lee et al., K. A. Stewart & D. L. Geiger, 1999 Andamanes O ie) Maldives Singapo (eo) OChagos Arch. O Rodrigues Isl. Mauritius B =H. crebrisculpta: type locality Page 93 O Hachijo Isl _ OOkinawa Mariana Isl. OKwajalein Guam vies O Palau Osieronesia O Solomon Isl. American SamoaQ Tonga 8 New Caledonia O = _—-H..clathrata: localities from specimens @ H. clathrata : localities from literature =—=— H.clathrata: distribution from literature cose «=H. rubra: distribution from literature Figure 7 Distribution of Haliotis crebrisculpta, H. clathrata, and H. rubra, based on records herein, and the literature. NG: New Guinea. Localities: based on individual specimen data, either from collections or from illustrations with specific data. Distribution: range indication not based on listed specimens. 1983; Beu & Maxwell, 1990), whereas in H. clathrata it is more or less straight. From this limited evidence, the localities for fossil H. clathrata are in agreement with its present day distribution. Geographic distribution: The species is known from lit- erature records in the shallow water of the central and eastern Pacific region from the Hachijo Jima group off southern Japan, Singapore, Indonesia (Bali, Sulawesi, Ambon), Papua New Guinea, Queensland to Western Australia (Scott Reef), and New Caledonia (Boone, 1938; Talmadge, 1963; Hinton, 1972, 1978; Anon., 1975; Dhar- ma, 1988; Salvat et al., 1988; Wilson, 1993; Baer, 1994; Anon., 1995). Specimens from collections add the follow- ing localities within the central and eastern Pacific: An- daman Islands, Philippines, Java, Borneo, Marshall Is- lands, Federate States of Micronesia, Tonga and Western and American Samoa. New records representing a sub- stantial range extension stem from the Indian Ocean, i.e., from the Maldives, Chagos Archipelago, Rodrigues Is- lands, Aldabra, Comoros, Madagascar, and Kenya. Note that the apparent distributional break for abalone at Cape Comorin, India (cf. Geiger, 1996) is crossed by H. clath- rata. Only one lot has been found from the mainland coast of Africa, none from the Red Sea or the Persian Gulf. Figure 7 shows the distribution of the species. Specimens examined: Specimens are listed below in East-West order. The number after the collection and cat- alog number refers to the number of specimens in the respective lot; preserved animals are marked with ‘‘com- plete.” KENYA: (BMNH 241, 1). MADAGASCAR: 25 km N of Tulear, Mora Mora Village (KAS, 2). Nosi Bé, N Nosy Komba, Pointe Ambarionaombi (ANSP 258629, 1). SW Nosi Bé, between Ambatoloaka and Madirokely (ANSP 259131, 2). COMOROS: (DLG, 1: complete). Sandy Is- land, Mayotte (NM J9665, 1). Anjouan (BMNH no #, 1). ALDABRA ATOLL: Picard Isl. (USNM 836531, 1). SEYCHELLES: (BMNH no #, 1). MAURITIUS: Belle Mare (CASIZ 044963, 2: complete). RODRIGUES ISL.: (BMNH, 1). Pt. Mathwesi (DLG, 1; KAS, 1). CHAGOS ARCHIPELAGO: Ile du Coin, Peros Banhos (BMNH, 1). MALDIVE ISL.: Gan Addu Atoll, south reef (BMNH, 1). Helengeli (A. Faucci collection, 23). Ari Atoll, NE of Feridu Island, Islet 5.5 km NE of Feridu Island (ANSP 303927, 1: complete). ANDAMAN ISL.: 80 km E of S Andaman Island, N end of invisible bank (ANSP 292648, 2). N end Invisible Bank, 45 miles E of S Andaman Island, 11° 23’ N, 093° 31’ E (ANSP 292648, 2). SINGAPORE: (BMNH; ANSP 196380, 1: complete). Raffles Light (ANSP 245726, 3). Sentosa (KAS, 1). SOUTH CHINA SEA: Macclesfield Bank (BMNH, 2). MALAYSIA: Borneo, Sabah, Sipidan (KAS, 1). Borneo, Page 94 The Veliger, Vol. 42, No. 1 Sabah, Sipidan, Sapi Island (KAS, 1: complete). INDO- NESIA: NW Nusa Perida, Toya Pakeh (DLG, 1; KAS, 1). Java, Thousand Island, Palau Pelangi (KAS, 1). Java, Pulau-Pulau Seribu Island, off Jakarta Roya, Pelangi and Putri Islets (LACM 86-163, 8). Sulawesi, off Menado, S side Banuken and Siladen Islet (LACM 88-56, 3: 1 com- plete). Bali (RP, 2). Bali, Lovina Beach, north coast (KAS, 6). Dual Tual (RP, 1). Lesser Sunda Isl., Komodo Isl., Station JEM 87—4 (CASIZ 081123, 1). Ambon, Nus Laut Island (DLG, 1: specimen from Baer, 1994). NEW GUINEA: N coast near Madang, Pig Island (= Tab Is- land) (CASIZ 086544, 1). 2.5 km SW of Biak Dock-Reef (ANSP 206392, 2: complete). New Britain (RP, 1). New Brittain, Rabaul (RP, 3). PHILIPPINES: Bohol (NMW, 1). Luzon Island, Batan- gas (CASIZ 081039, 1). Luzon Island, Batangas, Devil’s Point (CASIZ, 1). Luzon Island, Batangas, Calatagan (DMNH 205368, 1). Luzon Island, Bataan Province (LACM 74724, 1). Palawan, San Pedro Cove, Linapacan Island (LACM 88—285.21, 1). Palawan, Calamian Group, Batunan Island (LACM 74701, 1). Mindanao Island, Zamboanga, Yellow Beach (KAS, 1). Mindoro Island, Aro Point (KAS, 1). Lubang Island (KAS). Maricaibo Island, Devil Point (KAS, 1: complete). JAPAN: Okina- wa, | km WNW of Onna Village (LACM 78-27, 1; LACM 78-29, 1; LACM 79-75, 1). Okinawa, 5 km ESE of Zampamisaki (Bolo Point) (LACM 78-25, 1; LACM 78-100, 1). Okinawa, Serigaki Beach (DLG, 6; AMNH 276888, 3). Okinawa, 1 km S of Kuwae Hospital (USNM 838483, 1). Honshu, Hachijo Island (ANSP 240168, 1). WESTERN AUSTRALIA: Roebuck Bay (AMNH 220132, 1). NE corner of Seringapatam Reef (DLG, 1). Ashmore Reef (DLG, 1; KAS, 1). Cochburn Sound, Woodman Point (ANSP 358590, 1). QUEENSLAND: Lizard Island (LACM 79-53, 1; LACM 79-55, 1). Stan- age Bay (KAS, 4; RP, 1; DLG, 1; SBMNH, 1). Great Barrier Reef, Grub Reef (LACM 83-44, 1). Capricorn Island (CASIZ 102571, 3). Swain’s Reef (CASIZ 102570, 2). Middle Keppel Island (CASIZ 102572, 3). Great Kep- pel Island (KAS, 1). South Keppel Island, off Yepoon (CASIZ 102936). Keppel Group, Conical Island (CASIZ 102937). Keppel Bay, Middle Island (KAS, 3). Keppel Bay, Keppel Island (SBMNH, 1). Keppel Bay (SBMNH, 1). Keppel Bay, Pumpkin Island (KAS, 1). Gladstone (RP, 1). Humpy Island (KAS, 3). Moreton Bay (KAS, 1). Cart- er Reef (DMNH 51383, 1). NEW SOUTH WALES: Bot- any Bay (SBMNH, 2). MARIANA ISL.: W Saipan (DLG, 3; KAS, 1). Guam (KAS, 3; AMNH 220127, 2). Guam, Orote Cliffs (AMNH 220528, 1). SOLOMON ISL.: NE side Vanganu Island, Marovo Lagoon, Kokuana Passage, Matui Island (LACM 89-77, 1). Bunana Island (BMNH, 1). Honiara (CASIZ, 1). PALAU ISL. = BELAU: Koror, Malakal Harbor (ANSP 203083, 1). FEDERATE STATES OF MI- CRONESIA: off Arakabesan Island (ANSP 204544, 1). Upper Mortlocks, Losap Island (DMNH 205366, 1). Ka- pingamarangi Atoll (UCMP loc. # 13107, 1). Helen Reef, Helen Channel, Round Rock (ANSP 399940, 1). NEW CALEDONIA: (CASIZ 102935, 1). Central N side, Bogota Reefs (USNM 693386, 1). Grand Reef of Koumac (MNHN sta. 1316, 3; MNHWN sta. 551, 1). Passe de Koumac (MNHN sta. 1310, 1). Sector of Belep (MNHN sta. 1217, 1; MNHN sta. 1128, 1). Touaourou (USNM 795269, 1). Cook’s Reef (CASIZ 102574). Le des Pins (KAS, 2). Ile des Pins, Kuto Beach (CASIZ 102494). Noumea, [lot Charron, Baulari Bay (ANSP 275419, 1). Noumea, Baie de Citron (ANSP 237557, 1). Noumea, Baie Ouemo (DMNH 19675, 1; ANSP 271204, 1). Noumea, Touho, Koe Reef (DMNH 69885, 1). Nou- mea, Touho (ANSP 238033, 1). Sector of Touho (MNHN sta. 1264, 1; MNHWN sta. 1271, 1). Bourail (AMNH 107198, 1). Ouen Island, Prony Bay (MNHN sta. 232, 1). Sector of Yaté (MNHN sta. 735, 1). N. O. “Allis”? Cam- pagne SMIB 5, 23° 25’ S, 168° 05’ E (MNHN sta. DW99, 1). MARSHALL ISL.: Upper Mortlocks, Losap Island (DMNH 205366, 1). Kapingamarangi Atoll (UCMP loc. # 13107, 1). Kwajalein (KAS, 1). Kwajalein, West Reef (DLG, 2; KAS, 4). Kwajalein, Carlson Island (KAS, 3; KAS, 1). Eniwetok Atoll (AMNH 92485, 1). TONGA: Ha’apai Group, Cornfield and Campbell (CASIZ 102573, 1: complete). Vava’u Group, SW Vava’u Island S end Pangaimota Island cliff at W end of Maugaui (LACM 86— 220, 1). Vava’u Group, N side Nuapupa Island Outside of lagoon (LACM 85-89, 1). Vava’u Group, between Longitau Island and Vaka’eitu Island (LACM 85-90, 1). Eua Island Aa’a Luma Beach (KAS, 1). WESTERN SA- MOA: Savaii Island Mataatu Harbor, Eastern Reef (AMNH 178945, 1). AMERICAN SAMOA: Fagan Bay (KAS, 2). ACKNOWLEDGMENTS We thank Kathie Way and Julia Freeman (BMNH), Ali- son Trew (NMW), Alan Kabat (USNM), Terrence M. Gosliner (CASIZ), James H. McLean (LACM), David R. Lindberg (UCMP), Paula Mikkelsen (DMNH & AMNH), Gary Rosenberg (ANSP), and Philippe Bouchet (MNHN) for accommodating us during our stays at the respective institutions. Ted Baer, Anuschka Faucci, Susan McBride and Roger Pickery kindly made specimens available for study. Melinda Hayes at Hancock Foundation Library and staff at the CASIZ helped with some of the older refer- ences. The visit to DMNH and ANSP were made possible through a Merck DuPont and a Jessup Scholarship, re- spectively, to DLG, and the investigation of the radulae and the epipodia was supported by grants from the West- ern Society of Malacologists, the Hawaiian Malacological Society, and the Lehner Gray Fund for Marine Research (AMNH) to DLG. We thank Lindsey T. Groves, James H. McLean, Barry Roth, and two anonymous reviewers for improvements of the manuscript. K. A. Stewart & D. L. Geiger, 1999 LITERATURE CITED ANONYMOUS. 1975. Some Australian Haliotids. Australian Shell News No. 10:4—5. ANONYMOUS. 1995. Mollusques recoltés a Sulawesi - 1994. Bul- letin de la Societé Internationale de Conchyliologie 17(4): 4-11. ABBOTT, R. T. & S. P. DANcE. 1983. Compendium of Seashells. E. P. Dutton: New York. 411 pp. ADAMS, A. & L. REEVE. 1848. Zoological Voyage of HMS Sa- marang: Mollusca. X, 87 pp., 24 pls. Baer, T. 1994. 4 eme voyage en Indonésie—Automne 1993. Bul- letin de la Societé Internationale de Conchyliologie 16(1): 3-19. Beu, A. G. & P. A. MAXWELL. 1990. Cenozoic Mollusca of New Zealand. New Zealand Geological Survey Paleontological Bulletin 58:1—518. Boone, L. 1938. Scientific results of the world cruises of the yachts “Ara,” 1928-1929, and “Alva,” 1931-1932 Medi- terranean cruise, 1933, and “Alva” South American Cruise, 1935, William K. Vanderbilt, Commanding. Bulletin of the Vanderbilt Marine Museum 7:11—372, pls. 1-152. Brown, L. D. 1993. Biochemical genetics and species relation- ships within the genus Haliotis (Gastropoda: Haliotidae). Journal of Molluscan Studies 59:429—444, Cotton, B. C. & F K. Goprrey. 1933. South Australian shells. South Australian Naturalist 15:14—24, 1 pl. Cotton, B. C. 1959. South Australian Mollusca - Archaeogas- tropoda. W. L. Hakes, Government Printer: Adelaide. 449 PPp- DELHAES, W. 1909. Beitrage zur Morphologie und Phylogenie von Haliotis Linné. Inaugural Dissertation, Bonn. 55 pp. DHARMA, B. 1988. Siput Dan Kerang Indonesia (Indonesian Shells). PT. Sarana Graha, Jakarta. 111 pp. FLEMING, C. A. 1952. Notes on the genus Haliotis (Mollusca). A new subgenus from New Zealand and a new species from the late Cenozoic of Ohope, Bay of Plenty. Transactions of the Royal Society of New Zealand 80:229—232, pl. 50. GEIGER, D. 1991. Einige Aspekte tiber die Biologie von Haliotis. Club Conchylia Informationen 25:87—108. GEIGER, D. 1996. Haliotids in the Red Sea, with neotype desig- nation for Haliotis unilateralis Lamarck, 1822 (Gastropoda: Prosobranchia). Revue Suisse de Zoologie 102:339—354. GEIGER, D. L. 1998a. Recent genera and species of the family Haliotidae (Gastropoda: Vetigastropoda). The Nautilus 111: 85-116. GEIGER, D. L. 1998b. Note on the identity of Haliotis clathrata Lichtenstein, 1794 (not Reeve, 1846). Molluscan Research 19:157-159. GEIGER, D. L. & K. A. STEWART. In press. Case. Suppression of Haliotis clathrata Lichtenstein, 1794 to preserve H. clath- rata Reeve, 1846 and H. elegans Philippi, 1844. Bulletin of Zoological Nomenclature. GOSLINER, T. M., D. W. BEHRENS & G. C. WILLIAMS. 1996. Coral Reef Animals of the Indo-Pacific. Sea Challengers: Monte- rey. 314 pp. HERBERT, D. G. 1990. Designation of lectotype and type locality for Haliotis rugosa Lamarck, 1822 (Mollusca: Gastropoda: Haliotidae). Annals of the Natal Museum 31:207—213. Hica, L. 1983. Recent finds. Hawaiian Shell News 31(4):11. HiInToN, A. 1972. Guide to Shells of Papua New Guinea. Robert Brown & Associates Pty, Port Moresby, Papa New Guinea. 68 pp. [Date from Wilson (1993:18)] Hinton, A. 1978. Guide to Australian Shells. Robert Brown & Page 95 Associates Pty, Port Moresby, Papua New Guinea. 77 pp. [Date from Wilson (1993:18)] Iczn, 1957. Opinion 456: Rejection of the works by Thomas Martyn Published in 1784 with the title ‘The Universal Conchologist” as a work, which does not comply with the requirements of article 25 of the “‘régles’”’ and which there- fore possesses no status in zoological nomenclature and re- jection also of a proposal that the foregoing work should be validated under the plenary powers. Opinions and Declara- tions 15(22):393—418. INTERNATIONAL CODE FOR ZOOLOGICAL NOMENCLATURE, 1985. Adopted by the XX general assembly of the International Union of Biological Sciences. University of California Press: Berkeley. 338 pp. KAICHER, S. D. 1981. Card Catalogue of World Wide Shells, Pack # 28- Haliotidae. Kaicher: St. Petersburg, Florida. Kuropa, T. & T. HABE. 1952. Check List and Bibliography of the Recent Marine Mollusca of Japan. L. W. Stach: Tokyo. 210 pp. Lapp, H. D. 1966. Chitons and gastropods (Haliotidae through Aderobidae) from the western Pacific islands. United States Geological Survey Professional Paper 531:1—98, pls. 1—15. Lapp, H. D. & J. E. HOFFMEISTER. 1945. Geology of Lau, Fiji. Bernice P. Bishop Museum Bulletin 181:3—399, pls. 1-62. Leg, D. E., R. M. Carter, R. P. KING & A. EK Cooper. 1983. An Oligocene rocky shore community from Mt. Luxmore, Fiordland. New Zealand Journal of Geology and Geophysics 26:123-126. Leg, Y.-H. & V. D. VAcQuiER. 1995. Evolution and systematics in Haliotidae (Mollusca, Gastropoda): inference from DNA sequences of sperm lysin. Marine Biology 124:267—278. LICHTENSTEIN, A. A. H. 1794. Catalogus Rerum Naturalium Rar- issimum. Sectio Secunda Continens Conchylia, Item Miner- alia, Ligna Exotica, & Arte Parata. G. F Schniebes: Ham- burg. 118 pp. LupBrook, N. H. & K. L. GOWLETT-HOLMES. 1989. Chitons, gas- tropods, and bivalves. Pp. 501-724 in S. A. Shepherd & I. M. Thomas (eds), Marine Invertebrates of Southern Austra- lia, Part If. South Australian Government Printing Division: Adelaide. PICKERY, R. & L. STEPPE. 1995. General conchology: Haliotidae. Gloria Maris 34: pls. 4—5. Pitssry, H. A. 1890. Manual of Conchology; Structural and Sys- tematic with Illustrations of the Species 12:1—323, 65 pls. PrINcE, J. D. & S. A. SHEPHERD. 1992. Australian abalone fish- eries and their management. Pp. 407—426 in S. A. Shepherd, M. J. Tegner & S. A. Guzman del Proo (eds), Abalone of the World: Biology, Fisheries and Culture. Fishing News Books: Oxford. REEVE, L. 1846. Monograph of the Genus Haliotis. 22 pp., pls. 1-17. SaLvaT, B., C. Rives & P. REVERCE. 1988. Coquillages de Nou- velle-Calédonie. Times Editions/Les Editions du Pacifique: Singapore. 143 pp. SINCLAIR, M. 1963. Studies on the Paua, Haliotis iris Martyn, in the Wellington district 1945—46. Zoological Publications of the Victoria University, Wellington No. 35:1—16. Sowerby, G. B. 11 1882. Thesaurus Conchyliorum. 5, parts 37 & 38:1—-54, pls. 1-14. SowerByY, G. B. m 1914. Descriptions of new Mollusca from New Caledonia, Japan, Philippines, China, and West Africa. Annals and Magazine of Natural History, Series 8, 14:475— 480, pl. 19. Page 96 STATSoFT. 1994. Statistica™ for Macintosh 4.1. StatSoft Inc: Tul- sa, Oklahoma. STEWART, K. 1984. Notes on Haliotis squamosa Gray, 1827. Shells and Sea Life 16:92—95. STEWART, K. 1986. A new Haliotis from Java. Hawaiian Shell News 34(9):11. TALMADGE, R. R. 1957. Proposed revision of Haliotis ruber. The Nautilus 71:57—60. TALMADGE, R. R. 1961. Haliotids and stomatellids from Swain’s Reef, Queensland. The Veliger 3:112—113. TALMADGE, R. R. 1963. Insular haliotids in the western Pacific (Mollusca: Gastropoda), The Veliger 5:129—139, pl. 14. The Veliger, Vol. 42, No. 1 WAGNER, R. J. & R. T. ABBotT. 1978. Family Haliotidae Rafin- esque, 1815 in Standard Catalog of Shells, 3rd ed. American Malacologists: Greenville, 00O—201—00-205. WELLS, FE E., & W. BRycE. 1985. Seashells of Western Australia. Western Australian Museum: Perth. 207 pp. WEINKAUFF, H. C. 1883. Die Gattung Haliotis in Systematisches Conchylien-Cabinet von Martini and Chemnitz, 2(6)B:1-83, pls. 1-30. WHITEHEAD, T. 1982. Annotated list of species of the family Hal- iotidae. Australian Shell News 37—38:5. WILsoNn, B. 1993. Australian Marine Shells 1. Odyssey Publish- ing: Kallaroo. 408 pp. The Veliger 42(1):97—100 (January 4, 1999) THE VELIGER © CMS, Inc., 1999 NOTES, INFORMATION & NEWS Lioconcha (Sulcilioconcha) caledonensis sp. nov., a Species of Veneridae (Bivalvia) from New Caledonia Mary Ellen Harte 1180 Cragmont Ave, Berkeley, California 94708, USA and Kevin L. Lamprell 58 Marsden Road, Kallangur, Queensland 4503 Australia Introduction Careful re-examination of existing collections is some- times the source of new taxa (Lamprell & Stanisic, 1996). Eight lots of specimens labelled Lioconcha (Sulciliocon- cha) melharteae Lamprell & Stanisic, 1996, a venerid species recently described from New Caledonia and taken by extensive sampling programs conducted by the OR- STOM Institute in New Caledonia (Richer de Forges, 1990, 1991), were obtained from the Muséum National d’ Histoire Naturelle, Paris. Some specimens differed from L. (S.) melharteae and are regarded as a new species, described here. Materials and Methods Examination and measurements were done using vernier dial calipers and a 10X magnifying piece. Photographs were prepared by K. Lamprell using a Nikon FM2 cam- era, SB-21 Nikon Speedlight, AF Micro-Nikkor 105 mm Xx f/2.8 lens and copy stand. Abbreviations used in text: lv, left valve; rv, right valve; pv, paired valves, sta., sampling station of OR- STOM Institute in New Caledonia (Richer de Forges, 1990, 1991); MNHN, Muséum National d’Histoire Na- turelle, Paris. Shell length is the greatest distance from anterior to posterior margins. Shell height is the greatest distance from the umbo to the ventral margin. Shell width is the greatest distance between the external surfaces of the conjoined left and right valves. Systematics The systematic arrangement at generic and subgeneric levels follows that of Keen (1969). Genus Lioconcha Morch, 1853 Type species: Venus castrensis Linnaeus, 1758; subse- quent designation by Stoliczka (1870). Subgenus Sulcilioconcha Habe, 1951 Type species: Cytherea philippinarum Hanley, 1844; original designation. Lioconcha (Sulcilioconcha) caledonensis Harte & Lamprell, sp. nov. (Figures la—c, g-i.) Description: Shell trigonally ovate, equivalve, inequila- teral, moderately inflated, lightweight but sturdy, umbo- nes prosogyrous, slightly inflated, lunule well developed, pear-shaped, raised centrally, striate, defined by a faint impressed line; antero-dorsal margin short, slightly con- vex dorsally, sharply sloping, widely rounded terminally; postero-dorsal margin slightly convex, sharply sloping, widely convex posteriorly; ventral margin widely convex, incised. Shell to 21 mm in length. Teleoconch smooth, changing to sculpture on the disc 4.2 mm down from the tip of the umbo of a specimen 17.1 mm in height. Shell with fine, distinct, flattened cords, merging to fine, indis- tinct threads posteriorly, and slightly anastomosing ante- riorly before merging to fine, indistinct threads; interstic- es are narrow and shallow. Periostracum calcified, ara- gonitic, white. Hinge of lv with anterior lateral tooth well developed, knoblike, in height rising above the cardinal teeth from the plain of the hinge plate; anterior cardinal thin, oblique, joined to thick median cardinal forming an inverted v-shape; posterior cardinal long, ridgelike, sep- arated from the median cardinal by a deep pit. Hinge of rv with paired anterior lateral teeth; anterior cardinal short, moderately thick, parallel to the median cardinal; median cardinal bifid, narrowly triangular; posterior car- dinal bifid, elongate, oblique. Pallial line thin. Pallial si- nus small, a slight sinuation at the base of the posterior adductor muscle scar. Exterior of shell white to creamy white, sometimes with sparse, obscure, irregularly spaced Table 1 Dimensions of largest paratypes of Lioconcha (Sulcilio- concha) caledonesis in mm. Sampling Valve(s) station Length Height Width l rv 1103 18.5 17.0 6.6 1 pv 1103 14.1 12.2 8.5 1 pv 1129 18.8 17.3 11.4 1 pv 1117 14.7 12.9 8.7 Page 98 The Veliger, Vol. 42, No. 1 Figure la—c, g—i, d-f. a—c, g-i. Holotype of Lioconcha (Sulcilioconcha) caledonensis, Harte & Lamprell, sp. nov. a. left valve, length 18.5 mm. b. interior of right valve. c. posterior view of conjoined valves, height 16.7 mm, width 11.8 mm. g. interior of left valve. h. right hinge. i. left hinge. d-f. Lioconcha (Sulcilioconcha) melharteae, Lamprell Collection, d. left valve, length 20.5 mm. e. interior of right valve. f. posterior view of conjoined valves, height 17.7 mm, width 13.4 mm. Notes, Information & News Page 99 Table 2 Conchological comparison of Lioconcha (Sulcilioconcha) caledonensis and L. (S.) melharteae. Character Posterior sculpture Color pattern ored. Umbones Posterior shape Anterior lateral tooth (right valve) hinge plane Early teleconch smooth and scaped lines and small triangles; internal color white or cream. Type material: Holotype: MNHN; Nouvelle-Calédonie, Secteur des Belep: 1 pv, sta. 1103, 32 m, 19°43’S, 163°57’E, white muddy sand with oyster shells. B. Rich- er-ORSTOM coll. 25 October 1989. Dimensions of ho- lotype: length 18.5 mm, height 16.7 mm, width of con- joined valves 11.8 mm. Paratypes: MNHN; Nouvelle-Ca- lédonie, Secteur des Belep: 2 pv, 2 rv, 1 lv same data as holotype; 7 pv (+ 1 pv Australian Museum Sydney, AMS C312630), sta. 1129, 40 m, 19°29’S, 163°49’E; 5 pv, sta. 1117, 36 m, 19°38’S, 163°54’E; Lagon Nord: 1 pv, sta. 484, 35 m, 19°00’S, 163°35’E; 2 pv, sta. 517, 42 m, 19°09’S, 163°35'E; 4 pv, sta. 522, 42 m, 19°08’S, 163°38’E. For dimensions of some paratypes, see Table 1. Distribution: Specimens of this species are known only from the Belep Islands of New Caledonia ranging from 12°29’S, 163°49’E to 19°43’S, 163°57’E in depths be- tween 32 and 42 m. Sampling station environments in- clude sta. 1103 (see holotype, above); for sta. 1117, coarse, muddy sand with turritellid shells; and for sta. 1129, white, coarse, shelly sand, with Amusium. Remarks: This species is most similar to Lioconcha (Sulcilioconcha) melharteae Lamprell & Stanisic, 1996. Several conchological characters distinguish Lioconcha caledonensis from L. melharteae (Table 2; Figure la—f) and the other species within Sulcilioconcha. Lioconcha caledonensis has flattened ribs with shallower interstices and is often less colored than L. (S.) philippinarum (Hanley, 1844) or L. (S.) amirantium (Melvill, 1909), an Indian Ocean species very similar to L. philippinarum, the latter two species have rounded ribs and are often colored brown on the shell, escutcheon, and lunule. Lio- concha (Sulcilioconcha) richerdeforgesi Lamprell & Stanisic, 1996, is less trigonal with less inflated umbo- nes, more color patterns and narrower ribs, and gener- ally smaller than L. caledonensis. Lioconcha (Sulcilio- concha) dautzenbergi (Prashad, 1932) is creamier in col- L. caledonensis commarginal; threads merge and be- come indistinct. (Figure Ic) irregular, faint, sparse, zigzag mark- ings; escutcheon and lunule not col- slightly inflated (Figure Ic) slightly more angular rises above the cardinal teeth from the L. melharteae often not commarginal but oblique; threads anastomose and remain distinct. (Figure 1f) a solid posterior radial; occasional com- marginal bands; escutcheon and lunule colored. inflated (Figure 1f) convex lower than the cardinal teeth sculptured or, heavily patterned, and has much wider, rounded ribs. Both Lioconcha (Sulcilioconcha) trimaculata (Lamarck, 1818) and Lioconcha (Sulcilioconcha) polita (Réding, 1798) are more ovate in shape, and more heavily pat- terned and colored, with colored lunules and escutch- eons, and purple or brown colors internally; L. polita is smooth centrally. Literature Cited KEEN, A. M. 1969. Veneridae. Pp. 671—688. in R. C. Moore (ed.), The Treatise on Invertebrate Paleontology. Part N. Mollusca 6, Bivalvia. Geological Society of America and the Univer- sity of Kansas Press: Lawrence. LAMPRELL, K. L. & J. STANISIC. 1996. Callista, Lioconcha and Pitar in New Caledonia and adjacent waters (Mollusca, Ve- neridae). Molluscan Research 17:27—48. RICHER DE ForGEs, B. 1990. Les campagnes d’exploration de la faune bathyale dans la zone économique de la Nouvelle- Calédonie. Mémoires du Muséum National d’Histoire Na- turelle, (A) 145:9—54. RICHER DE ForGEs, B. 1991. Le benthos des fonds meubles des lagons de Nouvelle-Calédonie. Editions de 1 ORSTOM, Col- lection études et theses, Paris 1:1—148. STOLICZKA, F. 1870. The Pelecypods, with a review of all known genera of this class, fossil and recent. Pp. 537 in Geological Survey of India, Palaeontologica Indica, Series 6, Volume 3, Cretaceous Fauna of South India, Geological Survey Of- fice, 1865-1873: Calcutta. International Commission on Zoological Nomeclature The following Application was published on 30 Septem- ber 1998 in Volume 55, Part 3 of the Bulletin of Zoolog- ical Nomenclature. Comment or advice on this applica- tion is invited for publication in the Bulletin and should be sent to the Executive secretary, I. C. Z. N., c/o The Natural History Museum, Cromwell Road, London SW7 5BD, U.K. (e-mail: iczn@nhm.ac.uk). Page 100 Case 3087—HAydrobia Hartmann, 1821 and Cyclostoma acutem Draparnaud, 1805 (currently Hydrobia acuta; Mollusca, Gastropoda): proposed conservation by re- placement of the lectotype of H. acuta with a neotype; Ventrosa Radoman, 1877: proposed designation of Turbo ventrosus Montagu, 1803 as the type species; and HYDROBIINA Mulsant, 1844 (Insecta, Coleoptera): proposed emendation of spelling to HYDROBIUSINA, so removing the homonymy with HYDROBIIDAE Troschel, 1857 (Mollusca). The Veliger, Vol. 42, No. 1 The following Opinion concerning mollusks was pub- lished on 30 September 1998 in Volume 55, Part 3 of the Bulletin of Zoological Nomenclature. Copies of this Opinion can be obtained free of charge from the Execu- tive Secretary at the address given above. Opinion 1905. S. D. Kaicher (1973-1992), Card Cata- logue of World Wide Shells: not suppressed for no- menclature purposes. Information for Contributors Manuscripts Manuscripts must be typed, one side only, on A4 or equivalent (e.g., 842” X 11”) white paper, and double-spaced throughout, including references, figure legends, footnotes, and tables. All margins should be at least 25 mm wide. Text should be ragged right (i-e., not full justified). Avoid hyphenating words at the right margin. Manuscripts, in- cluding 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, in- cluding the year, if possible. Underline scientific names and other words to be printed in italics; no other manipulation of type faces is necessary on the manuscript. Metric and Celsius units are to be used. For aspects of style not ad- dressed here, please see a recent issue of the journal. The Veliger publishes in English only. Authors whose first language is not English should seek the assistance of a col- league who is fluent in English before submitting a manu- script. In most cases, the parts of a manuscript should be as follows: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, literature cited, figure legends, footnotes, tables, and figures. The title page should be a separate sheet and should include the title, authors’ names, and addresses. The abstract should be less than 200 words long and should describe concisely the scope, main results, and conclusions of the paper. It should not include references. 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 (Phillips, 1981), for two authors (Phillips & Smith, 1982), and for more than two (Phillips et al., 1983). The reference need not be cited when author and date are given only as authority for a taxonomic name. The “literature cited” section should include all (and only) references cited in the text, listed in alphabetical order by author. Each citation must be complete, with all journal titles unabbreviated, and in the following forms: a) Periodicals: Hickman, C. S. 1992. Reproduction and development of trochacean gastropods. The Veliger 35:245-272. b) Books: Bequaert, J. C. & W. B. Miller. 1973. The Mollusks of the Arid Southwest. University of Arizona Press: Tuc- son. xvi + 271 pp. c) Composite works: Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135 in R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stanford Univer- sity Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Avoid vertical rules. Figures and plates Figures must be carefully prepared and submitted ready for publication. Each should have a short legend, listed on a sheet following the literature cited. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photo- graphs for halftone reproduction must be of good quality, trimmed squarely, grouped as appropriate, and mounted on suitably heavy board. Where appropriate, a scale bar may be used in the photograph; otherwise, the specimen size should be given in the figure legend. Photographs should be submitted in the desired final size. Clear xerographic copies of figures are suitable for re- viewers’ copies of submitted manuscripts. It is the author's responsibility to ensure that lettering will be legible after any necessary reduction and that lettering size is appropriate to the figure. Use one consecutive set of Arabic numbers for all illus- trations (that is, do not separate “plates” from “text fig- ures’). Processing of manuscripts Each manuscript is critically evaluated by at least two reviewers. Based on these evaluations the editor makes a preliminary decision of acceptance or rejection. The editor's decision and the reviewers’ comments are sent to the author for consideration and further action. Unless requested, only one copy of the final, revised manuscript needs to be re- turned to the editor. The author is informed of the final decision and acceptable manuscripts are forwarded to the printer. The author will receive proofs from the printer. One set of corrected proofs should be mailed promptly to the editor after review. Changes other than the correction of printing errors will be charged to the author at cost. An order form for the purchase of reprints will accom- pany proofs. Reprints are ordered directly from the printer. Authors’ contributions The high costs of publication require that we ask authors for a contribution to defray a portion of the cost of pub- lishing their papers. However, we wish to avoid a handicap to younger contributors and others of limited means and without institutional support. Therefore, we have adopted the policy of asking for the following: $30 per printed page for authors with grant or other institutional support and $10 per page for authors who must pay from their personal funds (2.5 double-spaced manuscript pages normally equal one printed page). This request is made only after the pub- lication of a paper; these contributions are unrelated to the acceptance or rejection of a manuscript, which is entirely on the basis of merit. In addition to this requested contri- bution, authors of papers with an unusually large number of tables or figures will be asked for an additional contri- bution. Because these contributions by individual authors are voluntary, they may be considered by authors as tax- deductible donations to the California Malacozoo- logical Society, Inc., to the extent allowed by law. It should be noted that even at the rate of $30 per page, the CMS is paying well over half the publication costs of a paper. Authors for whom even the $10 per page contri- bution would present a financial hardship should explain this in a letter accompanying their manuscript. The edito- rial board will consider this an application for a grant to cover the publication costs. Authors whose manuscripts in- clude very large tables of numbers or extensive lists of (e.g.) locality data should contact the editor regarding possible electronic archiving of this part of their paper rather than hard-copy publication. Submitting manuscripts Send manuscripts, proofs, books for review, and corre- spondence on editorial matters to Dr. Barry Roth, Editor, 745 Cole Street, San Francisco, CA 94117, USA. CONTENTS — Continued The Giant Amazonian Snail (Pulmonata: Acavidae) beats them all FRANK P. WESSELINGH AND EDMUND GITTENBERGER .......00cceeeeeeeeecees Gastropods and intertidal soft-sediments: the case of Chilina ovalis Sowerby (Pul- monata: Basommatophora) in south-central Chile PEDRO_QUIJON AND) EDUARDO) JARAMILLO) ) "5 sere ye eo ae See renee eee Designation of lectotype for Haliotis crebrisculpta Sowerby, 1914, with a discussion of H. clathrata Reeve, 1846 (non Lichtenstein, 1794) KATHARINE A. STEWART -ANDU DANIEL Lo GEIGER aoe ane eee NOTES, INFORMATION & NEWS Lioconcha (Sulcilioconcha) caledonensis sp. nov., a species of Veneridae (Bivalvia) from New Caledonia MARY ELLEN HARTE AND KEVIN LV: IEAMPREDE “2 cee nt se siclerercs sore roe 67 2. 85 sts VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 42 Se "OA VOX ISSN 0042-3211 April 1999 Number 2 CONTENTS Utilization of artificial diets and effect of protein/energy relationship on growth performance of the apple snail Pomacea bridgesi (Prosobranchia: Ampullar- iidae) ROBERT MENDOZA, CARLOS AGUILERA, JESUS MONTEMAYOR, AND GABINO EX@ DRUGWIE Zeta ee meaen ip age Wei nee ese fSncslic nese locyci'e, abiteraiinng a uae Gnertminrs wean cls abate A new species of Depressigyra? (Gastropoda: Peltospiridae) from cold-seep carbon- ates in Eocene and Oligocene rocks of western Washington PANES lea G OEDERTVANDASMEVEN: REI BENHAM <2 2 -\sj of viele sce es sce sem © Calyptogena diagonalis, a new Vesicomyid bivalve from subduction zone cold seeps in the eastern North Pacific AMES RE BARRYCAND IWANDADIO b. INOGHEVAR: 44a) sidele socio. sere oes 2 sca wd ela ate' ¢ Histological description of the gonad, reproductive cycle, and fertilization of Pisid- ium amnicum (Miiller, 1774) (Bivalvia: Sphaeriidae) Re ARVATO OAT Nee Atrs RAIS ig tuate wayicrecs anes eireutai arepeeceege hora ue Genus actin adele W shepoweue (lv The Eastern Pacific Sportellidae (Bivalvia) JEOREIEINTS WA's (GGA Sle Bi NR a LUN SS CER ee RR a ate ee Laboratory observations of the feeding behavior of the cirrate octopod, Grimpoteu- this sp.: one use of cirri VAINATE Sus @ rag tel WN plige names ets weer tsitae'. ley ata ere Sree Weueyeneterinnes 6 Grete lalate a, ecska eh) elictay ise CONTENTS — Continued 101 12 1 124 132 The Veliger (ISSN 0042-3211) is published quarterly in January, April, July, and October by the California Malacozoological Society, Inc., % Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. Periodicals postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to The Veliger, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. THE VELIGER Scope of the journal The Veliger is an international, peer-reviewed scientific quarterly published by the Cali- fornia Malacozoological Society, a non-profit educational organization. The Veliger is open to original papers pertaining to any problem connected with mollusks. Manuscripts are considered on the understanding that their contents have not appeared, or will not appear, elsewhere in substantially the same or abbreviated form. Holotypes of new species must be deposited in a recognized public museum, with catalogue numbers provided. Even for non- taxonomic papers, placement of voucher specimens in a museum is strongly encouraged and may be required. Very short papers, generally not over 750 words, will be published in a “Notes, Infor- mation & News’ column; in this column will also appear notices of meetings and other items of interest to our members and subscribers. Editor-in-Chief Barry Roth, 745 Cole Street, San Francisco, CA 94117, USA e-mail: veliger@ucmp1.berkeley.edu Production Editor Leslie Roth, San Francisco Board of Directors Michael G. Kellogg, City and County of San Francisco (President) Hans Bertsch, National University, San Diego Henry W. Chaney, Santa Barbara Museum of Natural History Eugene V. Coan, California Academy of Sciences, San Francisco Terrence M. Gosliner, California Academy of Sciences, San Francisco Carole S$. Hickman, University of California, Berkeley EG. Hochberg, Santa Barbara Museum of Natural History Matthew J. James, Sonoma State University David R. Lindberg, University of California, Berkeley James Nybakken, Moss Landing Marine Laboratories David W. Phillips, Davis Peter U. Rodda, California Academy of Sciences, San Francisco Barry Roth, San Francisco Geerat J. Vermeij, University of California, Davis Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (not institutions) interested in any aspect of malacology. New members join the society by sub- scribing to The Veliger. Rates for Volume 42 are US $40.00 for affiliate members in North America (USA, Canada, and Mexico) and US $72.00 for libraries and other institutions. Rates to members outside of North America are US $50.00 and US $82.00 for libraries and other institutions. All rates include postage, by air to addresses outside of North America. Memberships and subscriptions are by Volume only and follow the calendar year, starting January 1. Payment should be made in advance, in US Dollars, using checks drawn from US banks or by international postal order. No credit cards are accepted. Payment should be made to The Veliger or “CMS, Inc.” and not the Santa Barbara Museum of Natural History. Single copies of an issue are US $25.00, postage included. A limited number of back issues are available. Send all business correspondence, including subscription orders, membership applications, pay- ments, and changes of address, to: The Veliger, Dr. Henry Chaney, Secretary, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105, USA. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. Barry Roth, Editor, 745 Cole Street, San Francisco, CA 94117, USA. © This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). THE VELIGER © CMS, Inc., 1999 The Veliger 42(2):101—111 (April 1, 1999) Utilization of Artificial Diets and Effect of Protein/Energy Relationship on Growth Performance of the Apple Snail Pomacea bridgesi (Prosobranchia: Ampullariidae) ROBERTO MENDOZA', CARLOS AGUILERA, JESUS MONTEMAYOR anp GABINO RODRIGUEZ Universidad Aut6noma de Nuevo Leon, Facultad de Ciencias Biolégicas, Aptdo. F-56 Cd. Universitaria. San Nicolas de los Garza, Nuevo Le6n México Abstract. Two feeding experiments were conducted with juvenile apple snails (Pomacea bridgesi.). The first was aimed at determining growth performance and adaptability with a natural vegetable source (dehydrated lettuce, D = 1) compared with a composite diet of high content of fishmeal (D = 2). Snails of three different shell lengths (S1 = 7.5 + 0.92, S2 = 12.6 + 0.29, and S3 = 19.3 + 0.46 mm) were fed at two different percentages of total weight (2% and 6%). The best results (P < 0.05) in terms of specific growth rate (SGR), shell length increase (SLI), feed conversion rate (FCR), and protein efficiency ratio (PER) were obtained with the artificial feed (D2). Snails of smaller length (S1) exhibited the best growth rate, being thus able to assimilate complex diets beginning from this length. Better results were obtained when snails were fed at 6% of their total weight. In the second experiment, a series of eight semi-purified formulations with varying levels of protein (10—40%) and energy (250-350 kcal/100g feed) were fed to triplicate groups of freshwater apple snails for 28 days. Results indicated that the SGR increased, as well as SLI, FCR, and PER with protein levels ranging from 20% to 30%. Low-energy diets (250 kcal/100g feed) were superior to high-energy diets (350kcal/100g feed) for all levels of protein tested. The SGR, FCR, and PER improved as dietary energy level was raised to 85 mg prot/kcal. Further increase of dietary energy had no beneficial effect in each protein level. Growth rates achieved in both experiments with artificial diets (13.83 and 14.16 mm/month, respectively) are far superior to those regularly obtained under laboratory conditions and even to those observed in the wild. The rapid growth rate attained and the ready acceptance of artificial diets suggest that the species could be cultured under intensive culture conditions. INTRODUCTION Apple snails or “tegogolos” (Pomacea) are freshwater mollusks quite common in the tropical lowlands of south- eastern Mexico (Rangel, 1988) and the south of North America (Banarescu, 1990), which present various char- acteristics that make them suitable for culture. They are herbivorous, thus efficient energy converters; they are prolific, reproducing all year round. They can be handled in combination with other species; such as tilapia (Ontiv- eros, 1989), tolerate a wide range of environmental con- ditions, have well-established local markets in some re- gions of the Caribbean and Mexico (Lum-Kong, 1989; Asiain & Olguin, 1995), and under controlled culture conditions, it is possible to avoid transmission of human diseases and parasites, often related to consumption of wild organisms (Asiain & Olguin, 1995). In addition to this, the genus Pomacea appears to posses other desirable qualities, which makes it attractive for culture. The most important feature is the rapid growth rate exhibited by individuals in the wild (13.52 mm/month, attaining a ' Corresponding author: Telephone (83) 52.97.72, fax (83) 10.05.96, e-mail: rmendoza@ccr.dsi.uanl.mx maximum length of 145 mm), which means more meat weight compared with smaller species of cultured snails (Burky, 1974; Lum-Kong, 1989). Its amphibious nature permits it to inhabit water low in dissolved oxygen and to withstand some crowding, indicating the potential for intensive culturing. Fairly high fecundity, high hatchabil- ity, low mortality, short developmental period, and ad- vanced state of hatching also enhance the prospects for culture (Lum-Kong & Kenny, 1989). The fact that the snail can withstand extended periods out of water (Burky, 1973) allows easy transportation to markets, and death as a result of handling is minimal. This results in reduction of labor and transport costs (Lum-Kong, 1989). This ge- nus has also received considerable attention because of its potential as human food (Lum-Kong & Kenny, 1989), source of protein for other aquatic animals (Bombeo-Tub- uran et al., 1995), for its role in biological control of aquatic weeds (Cazzaniga, 1981, 1983), and predation on schisostome-bearing gastropods (Cazzaniga, 1990; Este- benet & Cazzaniga, 1992). Despite all these advantages, the unwise introduction of these snails has had adverse ecologic and economic consequences, and in some cases, these snails have been Page 102 considered as pests in rice paddies because they feed on young rice plants (Bombeo-Tuburan et al., 1995; De Sil- va, 1989). In the wild, apple snails feed preferentially on macro- phytic vegetation (Estebenet, 1995), so under controlled culture conditions they have been traditionally fed with vegetable matter such as lettuce, alfalfa, Chara vulgaris, Potamogeton pectinatus, Psidium guajava, Pistia, strati- otes (Cazzaniga, 1981; Lum-Kong, 1989; Martinez, 1989; Estebenet & Cazzaniga, 1992). In a general way, this type of food is difficult to store, has variable nutritional qual- ity, and its availability is season-dependent. Therefore, development of a cost-effective practical feed is desirable for mass culture of the apple snail. At present, despite some major contributions on different aspects of gastro- pod digestive anatomy (Andrews, 1965), digestive phys- iology (Vonk & Western, 1984), and nutrition (Carefoot, 1982), knowledge of the quantitative and qualitative food requirements of the genus Pomacea has yet to be achieved. The production of a specific aquatic organism can be economical only when these requirements are known. Moreover, most of the experimental diets tested are commercial diets formulated to meet the requirements of other species (Estebenet & Cazzaniga, 1992; Benavi- des, 1994). A major obstacle in formulating a complete diet is the lack of information of the specific nutrient requirements of these animals, particularly the utilization of macronu- trients such as protein, carbohydrates, and lipids. Consid- ering that protein is the most expensive component of prepared feeds, several studies have shown that an ade- quate energy supply with non-protein energy sources can minimize its use, whereas an excess of energy may re- duce total protein intake, as has been shown in other cul- tured species like shrimp (Shiau & Chou, 1991). There- fore, the critical point is to obtain the proper protein/en- ergy (P/E) ratio in a diet for the most economical pro- duction of the apple snail. Reduction of protein in the diet would also reduce the amount of ammonia being ex- creted. Until now, only some attempts have been made to rear apple snails on artificial pellet diets (Benavides, 1994); a formulation which promotes optimal growth in these animals has not yet been reported. Since P/E relationships are basic to defining other nu- tritional requirements (Andrews et al., 1972), the present study was conducted to gain preliminary information on growth response to various levels of dietary proteins by the apple snail. MATERIALS AnD METHODS Two feeding experiments were conducted with juvenile apple snails (Pomacea bridgesi Reeve, 1856). The first experiment was designed to test the growth performance and adaptability of a natural vegetable source (dehydrated lettuce) compared with an artificial diet containing a high The Veliger, Vol. 42, No. 2 percentage of fishmeal. Individuals of three different lengths were fed at two percentages of the total weight of the snails in each aquarium, respectively. In the second experiment, eight semi-purified diets (Lovell, 1980) were used to establish the quantitative P/ E requirements of the snails. Four protein and two energy levels were fed to triplicate groups of freshwater apple snails. Experimental Animals A batch of apple snails was originally obtained from a local aquarium store in 1993 and maintained and repro- duced in captivity. At the start of the growth trials, 30 uniform-sized juvenile snails were selected and allotted at random in three glass aquaria. The experimental diets were fed to triplicate groups of 10 juvenile apple snails stocked in 31.25 L glass aquaria. The length of the shells (from the apex of the shell to the basal extreme aperture) was measured with a caliper to the nearest 0.1 mm. In the first experiment, the snails were sorted in three dif- ferent length classes (S1 = 7.5 + 0.92, S2 = 12.6 + 0.29, and S3 = 19.3 + 0.46 mm) and each group was fed at 2% and 6% of the total weight of the snails of each aquar- ium, respectively with two different diets: Diet 1 = de- hydrated lettuce and Diet 2 = artificial diet for 28 days. As 30 snails were used per treatment, a total of 360 snails was to cover the combination of the above mentioned treatments (3 X 2 X 2). For the second experiment, a different batch of snails of one length only was used. They had an average indi- vidual length of 16.3 mm + 2.3 mm and a mean weight of 1.27g + 0.52 g. A total of 280 snails was randomly divided into eight groups, and fed different diets in trip- licate containing four protein (10, 20, 30, and 40%) and two energy levels (250 and 350 kcal/100g feed), respec- tively, for 28 days. Each group was fed twice a day at a rate of 4% of the total weight of the snails in each aquar- ium daily. In both experiments, the snails were acclimated to lab- oratory conditions for 1 week in a 1500 L fiberglass tank (2.50 X 1.50 X 0.4m; L X W X H). During this period, the snails received fresh lettuce as a maintenance diet. Composition of the Experimental Diets Ingredients, chemical composition, and the P/E content of the experimental diets are summarized in Tables 1, 2, and 3, respectively. Proximate analysis of diets was con- ducted by standard Association of Official Analytical Chemists (AOAC) methods (AOAC, 1984). Moisture was determined gravimetrically considering thermic elimina- tion of water (by drying at 105°C), crude protein by a microKjeldahl method. Crude lipids were either extracted by the Soxhlet method; crude fiber was obtained in a fat- free material sample by dilute acid and alkali treatment; ash content was determined in a muffle furnace by heat- R. Mendoza et al., 1999 Table 1 Ingredients used for the formulation of the artificial diet (D1) tested in experiment 1. Ingredients g/100 g dry diet Fish meal 30.00 Soybean meal 5.60 Wheat meal 47.85 Shrimp meal 4.00 Wheat gluten 5.00 Fish oil 3.00 Lecithin 1.70 Vitamin mix? 2.50 4 Vitamin mix (mg/kg dry diet): thiamin 150, riboflavin 100, pyridoxine 50, pantothenic acid 0.1, niacin 300, biotin 1, B12 100, folic acid 0.1, vitamin E 400, vitamin K 20, vitamin A 15,000 IU/kg, vitamin D 7500 IU/kg. ing at 550°C for 3 h; and nitrogen free extract (NFE) was calculated by difference (NFE = 100 — moisture + pro- tein + lipid + ash). In semi-purified diets the energy levels were adjusted by varying the ratio of dextrin to cellulose; lipid level was kept constant at 25 g/kg of the diet. Since digestible energy values of foodstuffs for apple snails are unknown, physiological values for other invertebrates (Shiu & Chou, 1991) were taken into account for calculation of the energy level (5 kcal/g protein, 9 kcal/g lipids, and 4 keal/g carbohydrates). Feeds and Feeding Feed ingredients were ground, sieved, and mixed. The mixture was pelleted with a meat grinder equipped with a 1.5 mm die. Pelleted diets were dried at 80°C for 1 hr, placed in plastic bags, stored at —20°C, and a 2-day sup- ply was transferred to a refrigerator as needed. In experiment 1, fishmeal was used as the main protein source (Table 1). In the second experiment, only casein was used as protein source (Table 3). In the latter case, the protein content was adjusted by varying the amount of casein. The daily ration was given in two equal por- tions at 08:00 and 16:00 hr, and each morning before feeding, feces and other detritus in each aquarium were siphoned out and mortality was recorded. Experimental Facilities Snails for growth studies were placed in 31.25 L glass aquaria (50 X 25 X 25 cm) filled up to 25 L. The aquaria were provided with aeration and furnished with filtered dechlorinated hard tap freshwater. Two-thirds of the water in the aquaria was exchanged daily to remove impurities and maintain water quality. The dissolved oxygen level was kept at least at 6.0 ppm throughout the experimental period. Water temperature was maintained in the aquaria Page 103 Table 2 Chemical composition of the diets (D1 and D2) tested in experiment 1. Dehydrated Commercial lettuce feed Protein 15.35 34.83 Ether extract DDD, 3.89 Crude fiber 12.91 2.25 Ash 11.34 9.09 N-free extract 58.18 49.94 CME (keal/100 g)? 292.45 408.92 P/E ratio (mg prot/kcal) 45.45 85.4 * CME: Calculated metabolizable energy (kcal/g) based on pro- tein, 5 kceal/g; fat, 9 Keal/g; carbohydrate, 4 kcal/g (Shiau & Chou, 1991). by means of electric heaters, and ranged from 26—28°C during this period. Continuous aeration was provided in each tank throughout the experiment by an air compres- sor. The pH of the water varied between 8.0 and 8.2. Both experiments were conducted on a 12 hr light/dark pho- toperiod cycle. Growth Trial Following a fasting period of 24 hr, the animals were individually weighed and measured to register the initial weight and length. After this, they were weighed in bulk on the 7th and 14th day to adjust feed allowances. At the end of the study, animals were taken from each tank and were again individually weighed and measured. The an- imals were blotted dry before being weighed on a Sar- torious balance to the nearest 0.01 g. Calculation of Snail Performance Specific growth rate (SGR) = 100 (In average final weight. — In average initial weight.)/No. of days, Shell length increase (SLI) = average final length — average, initial length, Feed conversion rate (FCR) = dry weight feed (g)/wet weight gain (g), Protein efficiency ratio (PER) = weight gain (g)/protein fed (g), and Survival = number of individuals at the end of the experiment/num- ber of individuals at the beginning of the experiment X 100, were determined at the end of the experiment. Statistical Analysis The growth differences among the snails reared on the various diets were analyzed in the first experiment by a factorial ANOVA (3 X 2 X 2) with three replicates per treatment combination, while for the second experiment data were subjected to a bifactorial ANOVA (4 X 2). The New Duncan multiple range test (Steel & Torrie, 1980) Page 104 The Veliger, Vol. 42, No. 2 Table 3 Protein and energy content of diets used in the second experiment. Ingredients g/100 g dry diet Di D2 D3 D4 D5 D6 D7 D8 CME (keal/100 g))_ =§ 250 350 Protein (%) 10.00 20.00 30.00 40.00 10.00 20.00 30.00 40.00 Casein 11.10 22.20 33.30 44.40 11.10 22.20 33.30 44.40 Dextrin 43.30 30.80 18.30 5.80 68.30 55.80 43.30 30.80 Fish oil 2.50 2.50 2.50 2.50 2.50 2.50 2.50 2.50 Cellulose 27.10 28.50 29.90 31.30 2.10 3.50 4.90 6.30 Agar 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 Carragenin 3.00 3.00 3.00 3.00 3.00 3.00 3.00 3.00 Maltose 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Vitamin mixture? 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 Mineral mixture‘ 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 P/E (mg prot/kcal) 40.00 80.00 120.00 160.00 28.50 57.10 85.70 114.00 *CME: Calculated metabolizable energy (kcal/g) based on protein, 5 kcal/g; fat, 9 Keal/g; carbohydrate, 4 kcal/g (Shiau & Chou, 1991). > Vitamin mix (mg/kg dry diet): thiamin 300, riboflavin 200, pyridoxine 100, pantothenic acid 0.21, niacin 600, biotin 2, B12 200, folic acid 0.2, vitamin E 800, vitamin K 40, vitamin A 30,000 IU/kg, vitamin D 15,000 IU/kg. was used to identify differences among means at 0.05 level. RESULTS Growth performance of snails fed the experimental diets is presented in Tables 4 and 5 and in Figures 1 to 8. Experiment 1: Specific growth rate. SGR was the variable that reflected the best the effect of the different factors considered (type of feed, feeding level, and initial shell length) as Figure 1 shows. Significant differences (P < 0.05) were ob- served for every interaction of these factors (Table 4). Shell Length Increase. No significant differences were de- tected for this variable when the initial length of snails was compared, or when the interaction between initial length and type of feed was considered. Significant dif- ferences (P < 0.05) were registered only when the kind of feed and the feeding level were considered together. The better performance of both diets when supplied at 6% of the total weight and the marked difference between the artificial diet and the dehydrated lettuce are notable. Table 4 Specific growth rate (SGR), shell length increase (SLI), feed conversion ratio (FCR), and protein efficiency ratio (PER) for the experimental diets of the first experiment 1. Feed Length Diet supply (mm) SGR'! Dehydrated Lettuce 2% 7.50 2.31 + 0.552 12.60 0.82 + 0.225 19.30 0.91 + 0.36 6% 7.50 4.71 + 0.08% 12.60 3:72 =AOU8s 19.30 2.16 + 0.538 Commercial Feed 2% 7.50 55//7/wleS 6S 12.60 3.86 + 0.53 19.30 3.58 + 0.13! 6% 7.50 9.49 + 0.233 12.60 6.90 + 0.50° 19.30 527102924 SLI? FCR? PER* 1.90 + 0.694 ils itey seo) shill Sass) SE 1 Se 1.33 + 0.414 2.45 + 0.58° 2:83) = OVD" 1.76 + 0.114 2.26 + 0.98% 3.33 + 1.30% 5.86 + 0.72° 1.58 + 0.04°4 418 = O11 6.63 + 0.15% 0.98 = 0.1 1¢ 7.27 + 0.40% 5.46 + 2.07° 7210) 350) 7AE 2.60 + 0.63° 5.56 + 1.30° 0.39 + 0.153 8.18 + 3.703 7.00 + 0.36% 0.49 + 0.06%” 5.83 + 0.802%¢ 7.90 + 0.95° 0.48 + 0.01 5.80 + 0.432%¢ 13.83 + 0.61 0.70 + 0.05?” 3.98 + 0.24s¢ P2AO MES O2 0.52 + 0.04?» 5.40 + 0.42°4 12.30 + 0.70? 0.97 + 0.04% 2:93 s10N4e ' Figures with different superscripts in the same column are significantly different from each other (P < 0.05); means of triplicate groups = SD. R. Mendoza et al., 1999 Page 105 Table 5 Specific growth rate (SGR), shell length increase (SLI), feed conversion ratio (FCR), and protein efficiency ratio (PER) for the eight semi-purified experimental diets of the second experiment 1. CME Protein P/E (kcal/100 g) (%) (mg prot/kcal) SGR! SLIP FCR? PER* 250 10.00 40.00 4.02 + 0.06 9.66 + 0.49° 0.90 + 0.04° 11.02 + 0.498 20.00 80.00 5.39 + 0.35% 14.0 + 0.18 0.60 + 0.05 8.28 + 0.77° 30.00 120.00 5.51 + 0.37% 12.96 + 0.68?" 0.63 + 0.062 §.31 = 0.574 40.00 160.00 4.82 + 0.3> ales} ae ish 0.56 + 0.03? 4.43 + 0.31% 350 10.00 28.50 3.6 + 0.234 7.86 + 0.664 1.08 + 0.044 9.2 + 0.36 20.00 57.10 4.72 + 0.26% 11.76 = 0.2 0.77 + 0.04> 6.44 + 0.38° 30.00 85.70 5.77 + 0.63 14.16 + 1.38 0.63 + 0.14 5.31 + 0.944 40.00 114.00 5.49 + 0.722¢ 12.93 + 1.7 0.66 + 0.1% 3.85 + 0.55° ' Figures with different superscripts in the same column are significantly different from each other (P < 0.05); means of triplicate groups + SD. Feed Conversion Rate. The FCR did not seem to be affected were only observed for the source of protein and for the by the feeding level or by the initial weight of the animals. lower FCR obtained with the artificial diet. Nevertheless, there is a slight tendency of smaller animals to show a better FCR. Significant differences (P < 0.05) Protein Efficiency Ratio. No clear tendency for this factor (D1) ARTIFICIAL DIET (6%) (D1) ARTIFICIAL DIET (2%) (D2) DEHYDRATED LETTUCE (6%) (D2) DEHYDRATED LETTUCE (2%) , 19.3 mm (S2) ($3) Figure 1 Specific growth rate of Pomacea bridgesi juveniles of three different lengths fed with an artificial and natural diet supplied at two daily rations. Page 106 7.5mm (S1) 12.6 mm (S2) The Veliger, Vol. 42, No. 2 ARTIFICIAL DIET (6%) (D1) ARTIFICIAL DIET (2%) (D2) DEHYDRATED LETTUCE (6%) (D2) DEHYDRATED LETTUCE (2%) 19.3 mm (S3) Figure 2 Shell length increase of Pomacea bridgesi juveniles of three different lengths fed with an artificial and natural diet supplied at two daily rations. was observed, except that the higher values obtained for the smaller animals fed the diets containing animal protein stand out. Experiment 2: Specific growth rate. Significant differences were found with regard to the protein level. The best performance was obtained with levels ranging from 20-40% (P < 0.05) (Table 5). In this table it is also evident that growth rate increases proportionally with the P/E ratio up to a level of 85 mg protein/kcal after which it decreases slight- ly. Shell length increase. Results were similar to those ob- served for SGR. Higher values were obtained with P/E ratios ranging from 80 to 85 mg protein/kcal (P < 0.05). Feed conversion rate. As for the other variables the best results were obtained with protein levels higher than 10% with the best values in terms of P/E starting from 80 mg protein/kcal up to 160 mg protein/kcal (P < 0.05). Protein efficiency ratio. A clear tendency, although not significant, was observed concerning the protein level. On the other hand, it was observed that the PER decreased simultaneously with the P/E ratio. Survival. The mortality rate was less than 1% for both experiments (0.8% in the first and 0.4% in the second). DISCUSSION The best results in terms of SGR and SLI were obtained with the feed containing animal protein. In spite of the absence of significant differences regarding the length of snails, those of smaller length (7.5 mm) showed the best performance for the growth rate (Table 4, Figures 1, 2). Calcium was unlikely to be limiting, as hard tap water was used for experiments. On the other hand, soft body parts seem to reflect more clearly the nutritional state of gastropods (Brendelberger, 1995). The fact that the small- er snails showed a better growth rate is explained by their better FCR and PER, which would mean that even at this length they are able to assimilate complex diets. The bet- ter performance of artificial feed compared with lettuce can also be appreciated by the results of FCR and PER (Figures 3, 4). This may imply a better digestion of the composite diet, but it is also possible that the lettuce was R. Mendoza et al., 1999 (D1) ARTIFICIAL DIET (6%) (D1) ARTIFICIAL DIET (2%) (D2) DEHYDRATED LETTUCE (6%) us (D2) DEHYDRATED LETTUCE (2%) Page 107 Figure 3 Feed conversion ratio of Pomacea bridgesi juveniles of three different lengths fed with an artificial and natural diet supplied at two daily rations. not as actively consumed by the snails as the artificial diet because of poor palatability. It has been reported for other invertebrates that diets containing a mixture of two or more proteins are better utilized (Alava & Lim, 1983). It was noted that when both diets were supplied at 6% of the total weight, the PER decreased, possibly because the quantity of protein exceeded the digestive capacity of the snails and also because of a higher energetic expenditure to excrete the excess protein. The results obtained in this study are not in agreement with those reported by Este- benet & Cazzaniga (1992) who found a better growth rate in newly hatched snails reared on fresh lettuce than in those with composite diets. However, the lack of data regarding the protein content of the different diets tested and the FCR make it difficult to interpret their results. Lettuce was employed as a reference diet, because it has been used in most of the feeding bioassays with snails of the genus Pomacea (Meenakshi, et al., 1975; Martinez, 1989; Estebenet & Cazzaniga, 1992). Artificial diets were developed as an alternative to the difficulty of harvesting large quantities of aquatic plants, traditionally considered as important for freshwater snails, together with the high storage costs, (Estebenet, 1995). In relation to this, arti- ficial diets have proved effective for the culture and main- tenance of other prosobranchians (Biomphalaria glabra- ta, Standen, 1951; Marisa cornuarietis, Ferguson, 1978), and have also given good results in terms of growth and food conversion rate with other gastropods such as aba- lone (McVeigh, 1994). It was necessary to know the ap- propriate feeding rate for the apple snail because no data were available in the literature. Indeed, in most of the feeding studies carried out with individuals of the genus Pomacea, the animals have been fed ad libitum (Cazza- niga & Estebenet, 1988; Lum-Kong, 1989; Martinez, 1989; Ontiveros, 1989; Asiain & Olguin, 1995). The eventual differences in feed consumption may mask the performance of the diet, as was observed in the first ex- periment where similar growth rates were obtained when animals were fed at 2% of the total weight with an arti- ficial diet or at 6% with the dehydrated lettuce (Figures 1, 2). Similarly, Cazzaniga (1981) and Estebenet (1995) found out that a 6.0 g snail could consume from 4.9 to 12.5 g/day of vegetable matter (with a protein content varying from 0.9 g to 0.18 g), depending on the aquatic plant attractiveness and palatability. This would be equiv- alent to feeding the snails with only 0.3 g to 0.6 g of an artificial feed formulated to have 30% protein. Other re- sults demonstrated that the feeding level did not have a marked influence on the FCR and the PER (Figures 3, 4); this may be because these variables are determined most- ly by the quality and not the quantity of protein, ie., protein from vegetal or animal sources. This led us to consider that the snail can utilize the artificial diet more effectively than the lettuce. This is in agreement with the data reported by Ontiveros (1989) who observed good growth performance when the snails were fed with fish- Page 108 The Veliger, Vol. 42, No. 2 (D1) ARTIFICIAL DIET (6%) (D1) ARTIFICIAL DIET (2%) (D2) DEHYDRATED LETTUCE (6%) (D2) DEHYDRATED LETTUCE (2%) Figure 4 Protein efficiency ratio of Pomacea bridgesi juveniles of three different lengths fed with an artificial and natural diet supplied at two daily rations. meal. Besides, the acceptance of the composite diet con- taining animal protein is supported by the fact that in the wild they eventually fed on other animals. Indeed, unlike the majority of the freshwater snails which are micro- phagous, apple snails belonging to the Ampullaridae fam- ily show three feeding types (not mutually exclusive): microphagous, zoophagous, and macrophytophagous 4 mg Prot./kcal 120 160 350 10/250 20/350 20/250 30/350 40/350 30/250 40/250 Prot.%/kcal (D5) (D1) (D6) (D2) (D7) (D8) (D3) (D4) per 100g feed Figure 5 Specific growth rate of Pomacea bridgesi juveniles fed diets with different protein/energy ratio for 28 days. Bars with the same superscript are not significantly different (P < 0.05). (Cazzaniga & Estebenet, 1984); and even if the macro- phagous type is most common in Pomacea, feeding pref- erentially on macrophytic vegetation (Cazzaniga, 1987; Estebenet, 1995), it has been reported that under labora- tory conditions Pomacea canaliculata ate dead animals even when plant material was available, and that P. pal- udosa was able to depredate fish (Cazzaniga & Estebenet, 29 40 57 80 85 114 120 160 mg Prot./kcal 10/350 10/250 20/350 20/250 30/350 40/350 30/250 40/250 Prot.%/ kcal (D5) (D1) (D6) (D2) (D7) (D8) (D3) (D4) per 100g feed Figure 6 Shell length increase of Pomacea bridgesi juveniles fed diets with different protein/energy ratio for 28 days. Bars with the same superscript are not significantly different (P < 0.05). R. Mendoza et al., 1999 PER 160 mg Prot./keal 29 40 57 80 85 114 120 10/350 10/250 = 20/350 20/250 30/350 40/350 30/250 40/250 Prot.%/ kcal (D5) (D1) (D6) (D2) (D7) (D8) (D3) (D4) per 100g feed Figure 7 Feed conversion ratio of Pomacea bridgesi juveniles fed diets with different protein/energy ratio for 28 days. Bars with the same superscript are not significantly different (P < 0.05). 1984). The same authors indicate that P. insularum fed on dying or dead fish. Other ampullarids, like Marisa cornuarietis, are capable of eating meat from frogs, fish, and crabs (Ferguson, 1978). In terms of energy utilization, overall results showed that the best growth performance and shell length in- crease were obtained with protein levels ranging from 20-30%. When the semi-purified diets are analyzed in- dividually, it can be observed that for those diets with 10% and 20% protein, the energy level of 250 kcal/100g feed resulted in a better growth performance than for di- ets designed to have 350 kcal/100g feed. On the contrary, this relationship is reversed for diets containing 30% and 40% protein, which showed better results with higher en- ergy levels (350 kcal/100g feed) (Figures 5, 6). This re- flects a higher need of energy for a higher protein level. This aspect is also evident when the results of PER and FCR are considered, which indicate that energy levels of 250 kcal/100g feed are better than those of 350 kcal/100g feed for all protein levels and especially for those of 10% and 20% (Figures 7, 8), meaning that the energy content could be limiting the feed ingestion. Indeed, a lack of protein (e.g., 10%), regardless of the energy level, would mean less amino acids available for protein synthesis and deposition, while an excess of protein would imply a higher energetic expenditure for its catabolism and high levels of ammonia excreted, thus being detrimental for intensive culture conditions. In addition, protein would be used for energy and not for growth when inadequate energy is fed (Catacutan & Colosso, 1995); thus, balance of P/E in the feed is important. Considering that there is but a slight difference in growth produced with those di- ets containing 20% and 30%, the better PER of the former and a similar FCR for both, the protein requirement of Pomacea bridgesi appears to be near 20%. When consid- ering the P/E ratio, it is observed that the best perfor- mance in terms of SGR, SLI, and FCR was obtained with Page 109 29 40 57 80 85 114 120 160 mg Prot./kcal 10/350 10/250 20/350 20/250 30/350 40/350 30/250 40/250 Prot.% / kcal (D5) (D1) (D6) (D2) (D7) (D8) (D3) (D4) Per 100g feed Figure 8 Protein efficiency ratio of Pomacea bridgesi juveniles fed diets with different protein/energy ratio for 28 days. Bars with the same superscript are not significantly different (P < 0.05). values between 80 mg and 85 mg protein/kcal (Figures 6, 7 & 8). This value is close to the optimum reported for other aquatic organisms (88-90 mg_ protein/kcal, Haiqing & Xiqin, 1994; 83 mg protein/kcal, Xiqin et al., 19937). The foregoing could explain the difference in growth performance observed for the artificial diet and the dehydrated lettuce in the first experiment—first, by the higher content of protein of the former and second by its energy content. Indeed, it can be appreciated that the P/E ratio of the artificial diet was of 85.40 mg of protein/kcal compared to 45.45 mg protein/kcal for the lettuce. And as can be observed in the second bioassay, in spite of the narrow range of P/E ratio employed in this study, a P/E ratio of 85 mg protein/kcal appears to be the optimum for apple snail juveniles. It should be pointed out that the high fiber content of the lettuce could have resulted in a faster digestive transit and therefore a lower digestibility. The length achieved in these experiments, 13.83 mm/ month in the first experiment and 14.16 mm/month in the second experiment (Figures 2, 6), is far superior to those obtained by other authors under laboratory conditions, as is reflected in the work of Martinez (1989) who reported 5.5 mm/month with P. patula fed with alfalfa; Ontiveros (1989) obtained 5.3 mm/month with P. flagellata fed with Pistia sp., and Benavides (1994) 7.0 mm/month with Po- macea bridgesi with a 30% protein fish feed. Our results suggest that the diets used by this author, besides being higher in protein, also had a high energy level (330 kcal/ 100g feed) and a P/E ratio (90.9 mg protein/kcal), higher than the optimum found in this study. Finally, the results ? Xiqin, H., J. Lizhu, Y. Yunxia & X. Ghohuan. 1993. Studies on the utilization of carbohydrate-rich ingredients and optimal protein: energy ratio in Chinese bream, Megalobrama amblyce- phala Yih. Paper presented at the Fifth Asian Fish Nutrition Net- work Workshop, Thailand. Page 110 obtained are even better than those observed in the wild by Lum-Kong (1989) who registered a shell length in- crease of 13.5 mm/month for Pomacea urceus. This au- thor attributes lower growth rates exhibited by snails in the laboratory to an inadequate diet and high stocking densities. The importance of this study has been centered on the reduction of the protein level, and one way to spare di- etary protein is the utilization of non-protein energy sources, resulting in a reduction of feed cost (Haiqing & Xiqin, 1994). The rapid growth rates observed in this study suggest that the species can be reared in intensive culture systems. However, before apple snails can be intensively cultured, research is required in the production of artificial diets, optimal stocking densities, and in determining other fac- tors limiting growth under artificial conditions as has been pointed out by Lum-Kong (1989). Study concerning the elaboration of practical diets directed to the devel- opment of techniques of intensive culture would lead to the reduction of costs and time involved in manual col- lection from the wild and subsequent transport (Cazza- niga & Estebenet, 1985). ACKNOWLEDGMENTS This research was supported by CONACyT under the Project 4657N. The authors wish to express their grati- tude to Dr. Edna Naranjo for the taxonomic identification of the experimental apple snail individuals used in this study. LITERATURE CITED ALAVA, R. V. & C. Lim. 1983. The quantitative dietary protein requirements of Penaeus monodon juveniles in a controlled environment. Aquaculture 30:53—61. ANDREWS, B. E. 1965. The functional anatomy of the gut of the prosobranch gasteropod Pomacea canaliculata and of some other pilids. Proceedings of the Zoological Society of Lon- don 145:19—36. ANDREWS, J. W., L. V. Sicck & G. J. BAPTIST. 1972. The influence of dietary protein and energy levels on the growth and sur- vival of penaeid shrimp. Aquaculture 1:341—347. ASIAIN, A. & C. OLGuIN. 1995. Evaluation of water spinach (/p- omea aquatica) as feed for apple snail (Pomacea patula). Pp. 51-52. World Aquaculture 95, Book of Abstracts. ASSOCIATION OF OFFICIAL ANALYTICAL CHEMISTS (AOAC). 1984. Official Methods of Analysis. William, S. (ed.), 14th ed. Association of Official Analytical Chemists: Washington, D.C. 879 pp. BANARESCU, P. 1990. General Distribution and dispersal of Fresh Water Animals. Zoogeography of Fresh Water. Editorial Aula-Verlag: GmbH, Wiesbaden, Bucarest-Romania. 517 PP- BENAVIDES, M. 1994. Evaluacion nutricional de tres fuentes pro- téicas en dietas para cultivo de dos lineas de caracol man- zano (Pomacea bridgesi). M. Sc. Dissertation. Universidad Autonoma de Nuevo Leon, México. 50 pp. BOMBEO-TUBURAN, I., S. FUKUMOTO & E. M. RoprRIGUEZ. 1995. The Veliger, Vol. 42, No. 2 Use of the apple snail, cassava, and maize as feeds for the tiger shrimp, Penaeus monodon, in ponds. Aquaculture 131: 91-100. BRENDELBERGER, H. 1995. Dietary preference of three freshwater gastropods for eight natural foods of different energetic con- tent. Malacologia 36:147—153. Burky, A. J. 1973. Organic content of eggs and juveniles of an amphibious snail, Pomacea urceus (Miiller), from the Ve- nezuelan savannah and their ecological significance. (Mol- lusk Seminar, The University of Michigan, Ann Arbor, 1972), Malacological Review 6:59. Burky, A. J., 1974, Growth and biomass production of an am- phibious snail, Pomacea urceus (Miiller), from the Vene- zuelan savannah. Proceedings of the Malacological Society of London 41:127—143. CaAREFOOT, T. 1982. Gastropod nutrition. Pp. 321—337 in G. D. Pruder, C. Landgon & D. Conklin (eds.), Proceedings of the Second International conference. in Aquaculture Nutrition: Biochemical and Physiological Approaches to Shellfish Nu- trition, World Mariculture Society Special Publication. No. 2. Louisiana State University Press: Baton Rouge, Louisi- ana. CatacuTan, M. R. & R. M. CoLoso. 1995. Effect of dietary protein to energy ratios on growth, survival, and body com- position of juvenile Asian seabass, Lates calcarifer. Aqua- culture 131:125—133. CAZZANIGA, N. J. 1981. Evaluacion preliminar de un gaster6podo para el control de malezas acuaticas sumergidas. Pp. 131— 165 in IT Reunion sobre Malezas Subacuaticas en canales de desagtie de CORFO, Buenos Aires, Argentina. CAZZANIGA, N. J. 1983. Apple-snail eating Chara. Aquaphyte 3: 3-4. CazZANniGA, N. J. 1987. Pomacea canaliculata (Lamarck, 1801) en Catamarca (Argentina) y un comentario sobre Ampullaria catamarcensis Sowerby, 1874 (Gasteropoda, Ampullari- idae). IHERINGIA. Serie Zoologica, 66:43—68. CAZZANIGA, N. J. 1990. Predation of Pomacea canaliculata (Am- pullariidae) on adult Biomphalaria peregrina (Planorbidae). Annals of Tropical Medicine and Parasitology. 84:97—100. CAZZANIGA, N. J. & A. L. ESTEBENET. 1984. RevisioOn y notas sobre los habitos alimentarios de los Ampullariidae (Gastro- poda). Historia Natural 4:213—224. CAZZANIGA, N. J. & A. L. ESTEBENET. 1985. Revision de ante- cedentes sobre el uso de caracoles acuaticos (Ampullariidae) en programas de control bioldgico. Malezas (Revista de la Asociacion Argentina para el Control de Malezas) 13:23— 39. CAZZANIGA, N. J. & A. L. ESTEBENET. 1988. Effects of crowding on breeding Pomacea canaliculata (Gasteropoda: Ampul- lariidae). Comparative Physiology and Ecology 13:89—96. De SiLva, S. S. 1989. Exotic Aquatic Organisms in Asian Fish- eries Society, International Development Research Centre of Canada and Australian International Development Assis- tance Bureau Publishers 190 pp. EsTEBENET, A. L. 1995. Food and feeding in Pomacea canali- culata (Gasteropoda: Ampullariidae). The Veliger 38:277— 283. EsTEBENET, A. L. & N. J. CAZZANIGA. 1992. Growth and demog- raphy of Pomacea canaliculata (Gastropoda: Ampullariidae) under laboratory conditions. Malacological Review 25:1—12. FERGUSON, F. EF 1978. The Role of Biological Agents in the Con- trol of Schistosome-Bearing snails. U.S. Department of Health, Education, and Welfare, Public Health Service. Cen- ter for Disease Control: Atlanta. Georgia 107 pp. R. Mendoza et al., 1999 HalQinG, S. & H. XIGIN. 1994. Effects of dietary animal and plant protein ratios and energy levels on growth and body com- position of bream (Megalobrama skolkovii Dyboswski) fin- gerlings. Aquaculture 127:189-196. LovELL, R. T. (1980). Practical fish diets. Pp. 333-350 in Fish Feed Technology. Lectures presented at the FAO/UNDP training course in Fish feed technology, College of Fisheries, University of Washington, Seattle, Washington, USA, 9 Oc- tober—15 December, 1978. United Nations Development Programme. Food and Agriculture Organization of the Unit- ed Nations, Rome. LumM-Kona, A. 1989. The potential of Pomacea urceus as a cul- ture species in Trinidad. Pp. 33-39 in BCPC Monography No. 41 Slugs and Snails in World Agriculture. Lum-Kone, A. & J. S. KENNy. 1989. The reproductive biology of the ampullarid snail Pomacea urceus (Miller). Journal of Molluscan Studies 55:53—65. MarTINEZ, T. 1989. Contribucion a la ecologia y cultivo del car- acol de agua dulce Pomacea patula (Mesogasteropoda: Am- pullariidae). Bachelor’s Dissertation. Instituto Tecnoldgico del Mar, Boca de Rio, Veracruz, Mexico. 40 pp. McVEIGH, S. 1994. South Africans develop new diet for abalone. Fish Farming International 21:15. MEENAKSHI, V. R., P. L. BLACKWELDER, P. E. HARE, M. K. WILBUR & N. WaTABE. 1975. Studies on Shell Regeneration I. Ma- Page 111 trix and Mineral Composition of the Normal and Regener- ated Shell of Pomacea paludosa. Comparative Biochemistry and Physiology 50:347-351. ONTIVEROS, G. 1989. Producci6n semi-intensiva de crias de Po- macea sp. (Caracol dulceacuicola) en estanques de concreto, como apoyo a los programas de recuperacion de los sistemas palustres del municipio de Veracruz. Bachelor’s Dissertation. Instituto Tecnolégico del Mar, Boca de Rio, Veracruz, Méx- ico. 59 pp. RANGEL, L. J. 1988. Estudio morfol6gico de Pomacea flagelata Say, 1827 (Gasteropoda: Ampullaridae) y algunas consider- aciones sobre su taxonomia y distribucién geogrdafica en México. Anales del Instituto de Biologia UNAM, Serie Zoologica, 58:21—34. SHIAU, S. & B. CHou. 1991. Effects of dietary protein and energy on growth performance of tiger shrimp Penaeus monodon reared in seawater. Nippon Suisan Gakkaishi 57:2271—2276. STANDEN, O. D. 1951. Some observations upon the maintenance of Australorbis glabratus in the laboratory. Annals of Trop- ical Medicine and Parasitology 45:80—83. STEEL, R. & J. Torrie. 1980. Principles and Procedures of Sta- tistics. A Biometrical Approach. McGraw-Hill: New York: 633 pp. Vonk, H. J. & J. R. WESTERN. 1984. Comparative Biochemestry and Physiology of Enzymatic Digestion. Academic Press: London. 498 pp. The Veliger 42(2):112—116 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 A New Species of Depressigyra? (Gastropoda: Peltospiridae) from Cold- Seep Carbonates in Eocene and Oligocene Rocks of Western Washington JAMES L. GOEDERT 15207 84th Avenue Court NW, Gig Harbor, Washington 98329, USA AND STEVEN R. BENHAM Department of Geosciences, Pacific Lutheran University, Tacoma, Washington 98447, USA Abstract. Continued study of chemosynthetic marine-invertebrate faunas preserved in carbonates formed by the oxidation of methane at ancient cold-seeps reveals, tentatively, the first fossil record of the gastropod family Peltospiridae and the genus Depressigyra. Depressigyra? statura sp. nov., was found in three cold-seep carbonates within bathyal marine strata in western Washington: the middle Eocene Humptulips Formation; the early Oligocene part of the Makah Formation; and the late Oligocene part of the Lincoln Creek Formation. INTRODUCTION Taxonomic work on minute (< 5 mm height) gastropods from modern chemosynthetic communities such as those found near hydrothermal vents and cold seeps is resulting in the recognition of many new families, genera, and spe- cies (e.g., McLean, 1989; Warén & Bouchet, 1989, 1993). One recently described gastropod, Depressigyra globulus Warén & Bouchet, 1989, is the only known living species of the genus (Warén & Bouchet, 1993). It is one of the most common gastropods in chemosynthetic environ- ments near hydrothermal vents along the Juan de Fuca Ridge (Warén & Bouchet, 1989). Fossils of a new species tentatively referable to the genus Depressigyra have been found in localized, methane-derived carbonates within bathyal siltstones in three different formations in western Washington (Figure 1). This is the first detailed study of a minute archaeogastropod from fossil chemosynthetic communities. The abbreviation used for localities and specimens is LACMIP = Natural History Museum of Los Angeles County, Invertebrate Paleontology Section. Paleoenvironments The fauna preserved in a carbonate within the middle to late Eocene Humptulips Formation (LACMIP loc. 12385) was recognized as a chemosynthetic cold-seep community by Goedert & Squires (1990) and Campbell & Bottjer (1993). Benthic foraminifera indicate bathyal depths of 1500 to 2000 m (W. W. Rau cited in Goedert & Kaler, 1996). Carbonate blocks (LACMIP locs. 8233 and 15911) within bathyal basin-plain turbidites of the early Oligocene part of the Makah Formation are allo- chthonous (Goedert & Campbell, 1995), but they are methane-derived and contain chemosynthetic taxa. This carbonate was precipitated at cold-seeps in a shelf or slope environment, and then broke into blocks up to 2.5 m across when it slid or slumped into deeper parts of a basin (Goedert & Campbell, 1995). A cold-seep carbon- ate (LACMIP loc. 16504) from the late Oligocene part of the Lincoln Creek Formation was first reported by Squires (1995); it contains a diverse chemosynthetic as- semblage that is absent in the surrounding bathyal silt- stone (Squires & Goedert, 1995; Rigby & Goedert, 1996). All of these carbonates differ from other “normal” deep-water carbonates (e.g., nodules and concretions), in that they contain fossils of large numbers of organisms that are not present in surrounding strata, calcite and/or quartz lined vugs, and wavy-laminated carbonate crusts. These deep-water carbonates formed due to the bacterial oxidation of methane at cold-seeps. This interpretation was based on sedimentologic and paleontologic evidence (Campbell & Bottjer, 1993; Goedert & Squires, 1990; Rigby & Goedert, 1996; Squires, 1995; Squires & Goe- dert, 1991, 1995). The faunas contained in these distinc- tive carbonates and their depositional context compare well with western North American ancient and modern cold-seep carbonates described by Campbell & Bottjer (1993), Campbell et al. (1993), and Kulm & Suess (1990). The new species of Depressigyra? is abundant and well preserved in both the Humptulips Formation and the Lin- coln Creek Formation carbonates. Few specimens were found in the Makah Formation; however, most of the shell was lost during preparation because of the indurated na- ture of the micrite. J. L. Goedert & S. R. Benham, 1999 WASHINGTON 0 100 km Page 113 LINCOLN CREEK FM Lu = UJ oO O 4 _ Oo EOCENE Figure | Generalized geographic and chronostratigraphic distribution of localities for Depressigyra? statura Goedert & Ben- ham, sp. nov.; / = Makah Formation, LACMIP locs. 8233 and 15911; 2 = Humptulips Formation, LACMIP loc. 12385, 3 = Lincoln Creek Formation, LACMIP loc. 16504. Table 1 Carbon (65°C) and oxygen (8!%O) stable-isotope analyses of carbonates from the Humptulips Formation (LACMIP loc. 12385) and the Lincoln Creek Formation (LACMIP loc. 16504). All values expressed per mil (%c) relative to PDB standard. Sample 6BC 580 LACMIP loc. 12385: Serpulid? tube wall — 20.8 = 55f/ Serpulid? tube wall — 21.0 = Sni/ Serpulid? tube wall see OD, = Shit Micrite — 24.1 — 6.0 LACMIP loc. 16504: Micrite — 44.33 + 1.6 Micrite! — 46.38 + 2.6 Fibrous splayed calcite! — 46.73 + 2.4 ‘Both samples from same hand specimen of carbonate. Serpulid? tube sample analyses performed by T-M.B. Group, Inc., Miami, Florida; all others by Global Geochemistry Corp., Canoga Park, California. Stable Isotopes Isotopic data, 61°C and 6'8O, for the Makah Formation carbonate confirmed that it had precipitated from meth- ane-enriched fluids (Goedert & Campbell, 1995). Prelim- inary isotopic data (Table 1) indicate that the Humptulips Formation and Lincoln Creek Formation carbonates are also methane derived. Isotopic signatures of serpulid?-tube walls and micrite from the Humptulips Formation (LACMIP loc. 12385) are problematic. The 5'°C values of the tubes (— 20.8 to — 26.2) are similar to some reported for living tube worms from hydrocarbon-seep communities on the Lou- isiana slope (Brooks et al., 1987). The value for the mi- crite is close to one sample reported by Goedert & Camp- bell (1995) from the Makah Formation. These values probably indicate a mixing of methane-derived carbon with less 3'C-depleted sources, perhaps dissolved inor- ganic carbon in seawater and/or particulate and dissolved organic carbon. The unusually low values for 6'3O may represent diagenetic modification, but they could also in- dicate elevated temperatures and/or meteoric influence of pore waters, or 5!8O depletion in marine pore water (Sass et al., 1991). Samples from the Lincoln Creek Formation (LACMIP Page 114 The Veliger, Vol. 42, No. 2 loc. 16504) yielded very negative 6°C values (— 44.33 to — 46.73). As in the Makah Formation carbonate, 64°C values this negative are indicative of precipitation from a methane-enriched fluid source (Goedert & Campbell, 1995, and references therein). Values of 5!8O from the carbonate are positive but they are consistent with pre- cipitation at or near ambient seawater temperatures (K. A. Campbell, personal communication, 1996). SYSTEMATIC PALEONTOLOGY Order ARCHAEOGASTROPODA Thiele, 1925 Suborder NEOMPHALINA McLean, 1990 Superfamily NEOMPHALOIDEA McLean, 1981 Family PELTOSPIRIDAE McLean, 1989 Remarks: Warén & Bouchet (1989) considered the fam- ily Peltospiridae to be polyphyletic. Living peltospirid genera are differentiated by characters that include soft- part anatomy, radular structure, and protoconch sculpture. Additional studies may justify the reassignment of the genus Depressigyra to another family; therefore the cur- rent classification is tentative (Warén & Bouchet, 1993). This family has no previously reported fossil record. The genus Depressigyra was unintentionally referred to the family Hyalogyrinidae by Lewis & Marshall (1996:189). Genus DEPRESSIGYRA Warén & Bouchet, 1989 Type species: Depressigyra globulus Warén & Bouchet, 1989, by original designation. In their diagnosis of the genus Depressigyra, Warén & Bouchet (1989:80) stated that the aperture was ‘‘distinctly opisthocline,’’ whereas it is actually prosocline. This error was confirmed by A. Warén (personal communication, 1997). Warén & Bouch- et (1989) also stated in the diagnosis that the protoconch of Depressigyra has a net-sculpture, but in their descrip- tion of D. globulus they stated that the protoconch sculp- ture was unknown. The original diagnosis of genus De- pressigyra Warén & Bouchet, 1989, is therefore emend- ed. Emended diagnosis: Globular peltospirids of medium size; teleoconch almost smooth except for irregular and slightly sinuous growth lines, aperture round and distinct- ly prosocline; central and lateral teeth of radula unusually slender; no tentacular sexual dimorphism. Remarks: This genus originally included two species, D. globulus Warén & Bouchet, 1989, living only at hydro- thermal vents at various sites on the Juan de Fuca Ridge, and D. planispira Warén & Bouchet, 1989, living at vent sites on the East Pacific Rise. With additional data on shell and soft-part anatomy from new specimens, De- pressigyra planispira subsequently became the type spe- cies of a new genus, Planorbidella Warén & Bouchet, 1993, making the genus Depressigyra monotypic. At least four more gastropod genera living at methane- seeps and having shells similar to D. globulus, but pos- sessing distinctive radulae and protoconchs, await de- scription (A. Warén, personal communication, 1997). The protoconchs of all available specimens of both D. glob- ulus and the new fossil species are too corroded to pre- serve any sculpture that may have been present. There- fore, the new species is tentatively referred to the genus Depressigyra entirely on the basis of similarity of the teleoconch with that of D. globulus. Future studies may warrant reassignment of the new species to another genus. Depressigyra? statura Goedert & Benham, sp. nov. (Figure 2A—G) ““Naticid’’ Goedert & Squires, 1990, p. 1182, fig. 2g; Goe- dert & Kaler, 1996, p. 67, table 1. ““Hyalogyrinid’’ Goe- dert & Campbell, 1995, p. 25, figs. 11, 12. Diagnosis: A Depressigyra? with a spire elevated well above the body whorl. Description: Shell small, globose, thin, nearly smooth except for numerous fine, sinuous, prosocline growth lines; aperture nearly round, prosocline, outer lip thin; whorls convex, suture impressed, spire elevated above body whorl and apex formed by a knoblike protoconch, protoconch surface corroded in all available specimens, appears to be about one whorl; largest shell with 2.25 post-larval whorls. Comparisons: Except for the slightly more inflated whorls and higher spire, the shell of D.? statura sp. nov., resembles that of D. globulus Warén & Bouchet (1989: 80-81, figs. 30, 31, 45-47, 51-52, 78, 83). The sculpture of the protoconch of D. globulus is not known, the apex of all known specimens having been corroded (Warén & Bouchet, 1989). The only measurement for D. globulus is a maximum diameter of 5.4 mm (Warén & Bouchet, 1989:80), and all specimens of D.? statura sp. nov., are smaller (Table 2). Depressigyra? statura sp. nov., somewhat resembles another living chemosynthetic community gastropod in the family Cyathermidae, Cyathermia naticoides Warén & Bouchet (1989: 70-72, figs. 6-10, 15, 16, 18, 21-23, 71, 80), but D.? statura sp. nov., lacks the very distinct and highly diagnostic deep, rounded notch in the lower part of the aperture. Material: Holotype, LACMIP 7892, paratypes LACMIP 7893, 7894, 7895, 7896, 7988, LACMIP loc. 16504, Lin- coln Creek Formation, late Oligocene. Paratype LACMIP 7897, hypotype LACMIP 8343, LACMIP loc. 12385, Humptulips Formation, middle Eocene. Hypotypes LAC- MIP 12318, 12319, LACMIP loc. 8233, Makah Forma- tion, early Oligocene. Additional specimens are stored at Page 115 J. L. Goedert & S. R. Benham, 1999 Figure 2 Depressigyra? statura Goedert & Benham, sp. nov., secondary electron micrographs. All from the Lincoln Creek Formation, LACMIP loc. 16504 unless otherwise noted. Outer lip of aperture broken or partially concealed by matrix in all specimens. A. ho- lotype LACMIP 7892, apertural view, X14; B. paratype LACMIP 7893, back view, X14; C. paratype LACMIP 7988, lateral view showing prosocline aperture, X12; D. paratype LACMIP 7895, bottom view, X14; E. paratype LACMIP 7897, Humptulips For- mation, LACMIP loc. 12385, oblique top view, X14; E paratype LACMIP 7894, top view, X14; G. paratype LACMIP 7896, apex showing growth lines, suture, and corroded protoconch, X90. Table 2 Measurements (in mm) of type specimens of Depressi- gyra? statura sp. nov.; D = diameter, H = height. Specimen D H LACMIP 7892 2.6 2) LACMIP 7893 Pasi Pgs) LACMIP 7894 2.8 3.1 LACMIP 7895 Sul 3.4 LACMIP 7896 2.0 DES) LACMIP 7897 2.4 3.0 LACMIP 7988 2.5 2.6 LACMIP and California State University, Department of Geological Sciences, Northridge (CSUN). Etymology: The species name, statura, Latin meaning Stature, is in reference to the high spire, being contrary with the etymology for genus Depressigyra, alluding to a low spire. ACKNOWLEDGMENTS Gail H. Goedert and Keith L. Kaler assisted with field- work. Isotope analyses were funded and SEM facilities were provided by Pacific Lutheran University (Tacoma, Washington), and Simpson Timber Company (Shelton, Washington) allowed access to one of the localities. We thank Richard L. Squires (CSUN) and Kathleen A. Campbell (NASA/Ames Research Center) for discussions on cold-seep faunas and carbonates. Various drafts of this paper were substantially improved by an anonymous re- viewer, Anders Warén (Swedish Museum of Natural His- tory, Stockholm), Philippe Bouchet (Muséum National d’ Histoire Naturelle, Paris), James H. McLean and Lind- sey T. Groves (LACMIP), Barry Roth, and especially Ri- chard L. Squires (CSUN). LITERATURE CITED Brooks, J. M., M. C. Kennicutt II, C. R. FISHER, S. A. MACKO, K. Cote, J. J. CHILDREss, R. R. BIDIGARE & R. D. VETTER. 1987. Deep-sea hydrocarbon seep communities: evidence for energy and nutritional carbon sources. Science 238:1138— 1142. CAMPBELL, K. A. & D. J. BOTTJER. 1993. Fossil cold seeps (Ju- rassic-Pliocene) along the convergent margin of western North America. National Geographic Research and Explo- ration 9:326—343. CAMPBELL, K. A. C. CARLSON & D. J. BOTTIJER. 1993. Fossil cold seep limestones and associated chemosymbiotic macroin- vertebrate faunas, Jurassic-Cretaceous Great Valley Group, California. Pp. 37-50 in S. Graham & W. Lowe, (eds.), Ad- vances in the Sedimentary Geology of the Great Valley Group. Pacific Section, Society of Economic Paleontologists and Mineralogists, 73. GoeEDERT, J. L. & K. A. CAMPBELL. 1995. An early Oligocene chemosynthetic community from the Makah Formation, Page 116 northwestern Olympic Peninsula, Washington. The Veliger 38:22-29. GoeEDERT, J. L. & K. L. KALER. 1996. A new species of Abys- sochrysos (Gastropoda: Loxonematoidea) from a middle Eo- cene cold-seep carbonate in the Humptulips Formation, western Washington. The Veliger 39:65—70. GOEDERT, J. L. & R. L. Squires. 1990. Eocene deep-sea com- munities in localized limestones formed by subduction-re- lated methane seeps, southwestern Washington. Geology 18: 1182-1185. KuLM, L. D. & E. Sugss. 1990. Relationship between carbonate deposits and fluid venting: Oregon accretionary prism. Jour- nal of Geophysical Research 95(B6):8899-8915. Lewis, K. B. & B. A. MARSHALL. 1996. Seep faunas and other indicators of methane-rich dewatering on New Zealand con- vergent margins. New Zealand Journal of Geology and Geo- physics 39:181—200. MCLEAN, J. H. 1981. The Galapagos Rift limpet Neomphalus: relevance to understanding the evolution of a major Paleo- zoic-Mesozoic radiation. Malacologia 21:291—336. McLean, J. H. 1989. New archaeogastropod limpets from hy- drothermal vents: new family Peltospiridae, new superfamily Peltospiracea. Zoologica Scripta 18:49—66. McLean, J. H. 1990. A new genus and species of neomphalid limpet from the Mariana vents with a review of current un- derstanding of relationships among the Neomphalacea and Peltospiracea. The Nautilus, 104:77—86. RicBy, J. K. & J. L. GOEDERT. 1996. Fossil sponges from a lo- calized cold-seep limestone in Oligocene rocks of the Olym- pic Peninsula, Washington. Journal of Paleontology 70:900— 908. Sass, E., A. BEIN & A. ALMOGI-LABIN. 1991. Oxygen-isotope composition of diagenetic calcite in organic-rich rocks: ev- idence for '8O depletion in marine anaerobic pore water. Ge- ology 19:839-842. Squires, R. L. 1995. First fossil species of the chemosynthetic- community gastropod Provanna: Localized cold-seep lime- stones in upper Eocene and Oligocene rocks, Washington. The Veliger 38:30—36. Squires, R. L. & J. L. GOEDERT. 1991. New late Eocene mollusks from localized limestone deposits formed by subduction-re- lated methane seeps, southwestern Washington. Journal of Paleontology 65:412—416. Squires, R. L. & J. L. GoeperRT. 1995. An extant species of Leptochiton (Mollusca: Polyplacophora) in Eocene and Ol- igocene cold-seep limestones, Olympic Peninsula, Washing- ton. The Veliger 38:47—53. The Veliger, Vol. 42, No. 2 THIELE, J. 1925. Gastropoda der deutschen Tiefsee-Expedition. II Teil: Deutsche Tiefsee-Expedition auf dem Dampfer “Val- divia” 1898-1899. Wissenschaftliche Ergebnisse 17:35- 382. WAREN, A. & P. BOUCHET. 1989. New gastropods from East Pa- cific hydrothermal vents. Zoologica Scripta 18:67—102. WaAREN, A. & P. BOUCHET. 1993. New records, species, genera, and a new family of gastropods from hydrothermal vents and hydrocarbon seeps. Zoologica Scripta 22:1—90. APPENDIX: LOCALITIES CITED LACMIP loc. 8233. Float eroded from bedrock exposed on modern beach terrace at Shipwreck Point, SE1/4 NE1/4 section 36, T. 33 N, R. 14 W, Sekiu River USGS 7.5-minute quadrangle, Provisional Edition 1984, Clal- lam County, Washington. Upper part of Makah For- mation. Age: Early Oligocene. LACMIP loc. 12385. Small hill in abandoned meander of the East fork of the Humptulips River, northwest part of Sec. 4, T. 20 N, R. 9 W, Burnt Hill USGS 7.5 minute quadrangle, Provisional Edition 1990, Grays Harbor County, Washington. Humptulips Formation. Age: Middle Eocene. LACMIP loc. 15911. Jn situ isolated limestone block within thin-bedded sandstone and siltstone deposits, about 30 m stratigraphically above top of Jansen Creek Member, block measures 1.5 m (N-S) by 2.5 m (E— W), and is weathered out 0.75 m higher than surround- ing siltstone; accessible only at low tide. Block is ap- proximately 175 m southeast of tip of Shipwreck Point, SE1/4 NE1/4 Sec. 36, T. 33 N, R. 14 W, Sekiu River USGS 7.5-minute quadrangle, Provisional Edition 1984, Clallam County, Washington. Upper part of Makah Formation. Age: Early Oligocene. LACMIP loc. 16504. Limestone block on north side of sharp bend of the Canyon River, 600 m N and 290 m E of SW corner of Sec. 25, T. 21 N., R. 7 W., Grisdale USGS 7.5 minute quadrangle, Provisional Edition 1990, Grays Harbor County, Washington. Upper part of the Lincoln Creek Formation. Age: earliest late Ol- igocene. This locality was covered by a large landslide in early 1997. The Veliger 42(2):117—123 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 Calyptogena diagonalis, a New Vesicomyid Bivalve from Subduction Zone Cold Seeps in the Eastern North Pacific JAMES P. BARRY Monterey Bay Aquarium Research Institute, PO. Box 628, Moss Landing, California 95039, USA AND RANDALL E. KOCHEVAR Monterey Bay Aquarium, Pacific Grove, California 93950, USA Abstract. A new vesicomyid bivalve species, Calyptogena diagonalis, is described from cold seep communities in the Cascadia subduction zone off the Oregon coast and accretionary wedge sediments along the Pacific coast of Costa Rica. Live bivalves and shells were collected at sulfide seeps near 2021 m depth in Oregon and from 2900 to 3800 m depth in Costa Rica. Shell morphology of C. diagonalis differs considerably from sympatric congeneric and confamilial species of the northeastern Pacific. Shells are large (to 24.0 cm) and elongate (H/L = 0.42), with one or more ridges on the external shell surface extending diagonally from the umbo to near the posteroventral margin. Enlarged, sulfur- colored ctenidia and micrographs of endosymbiotic bacteria held in ctenidia suggest that this species, like other vesi- comyids, is a sulfur-based chemolithoautotroph. INTRODUCTION The bivalve family Vesicomyidae, first established by Dall & Simpson (1901) includes more than 50 species found nearly exclusively in sulfide-rich habitats such as cold seeps, hydrothermal vents, and accumulations of or- ganic debris (e.g., whale carcasses) from 450 to greater than 3000 m depth. All species investigated have been shown to rely nutritionally on sulfide-oxidizing endosym- biotic chemoautotrophic bacteria held in ctenidia (Fiala- Meédioni et al., 1994). Fossil representatives of the Vesicomyidae are known from as early as the Eocene from the Pacific Northwest, and span the Paleogene and Neogene from collections at several locations (Boss & Turner, 1980; Kanno et al., 1989; Niitsuma et al., 1989; Goedert & Squires, 1993). Although several genera have been erected, most extant species fall under two genera (Vesicomya and Calyptogena). Vesico- mya was established in 1886 within the Veneridae (Dall, 1886), and the genus Calyptogena, originally placed in the Carditidae, was described in 1891 (Dall, 1891). Assign- ment of species among genera has resulted in considerable taxonomic confusion within the family, particularly at the generic level (Kojima et al., 1995; Vrijenhoek et al., 1995; Peek & Vrijenhoek, in press). Molecular studies concern- ing taxonomic affinities within the Vesicomyidae may soon resolve the alignment of species among genera (R. Vrijen- hoek, personal communication). Increased exploration and sampling of vent and seep habitats (and other sulfidic environments) since their dis- covery in the late 1970s have greatly expanded our un- derstanding of the natural history and biology of vesi- comyids, including description of many new species. Ear- ly trawl and dredge samplers were deployed most com- monly over soft sediments, thereby undersampling geologically rugged terrain where seep and vent habitats often occur. In addition, these habitats are highly local- ized, further reducing the likelihood of collections using surface-deployed devices. Recent increases in access to these sites by manned submersibles and remotely oper- ated vehicles have allowed focused investigations of en- vironments typically inhabited by vesicomyids, as well as detailed studies of their natural history. In this paper we describe a new species of vesicomyid bivalve collected from cold seeps associated with accretionary sediments along subduction zones off Oregon and Costa Rica. COLLECTION INFORMATION Specimens analyzed for the erection of Calyptogena dia- gonalis sp. nov. were obtained from newly discovered cold seeps in the Cascadia Trough along the Oregon sub- duction zone (D. Orange, unpublished data), and along the Costa Rica accretionary wedge. A total of 15 live clams or articulated shells were collected at the Oregon site (44°40.56"N, 125°7.08"W) during ALVIN dives (# 2644, 2659, and 2663) at a depth of 2021 m. The Cas- cadia fauna was dominated by several species of vesi- comyid clams (mainly C. diagonalis) and bacterial mats, as well as columbellid snails typical of sulfide-rich hab- itats (e.g., Mitrella permodesta). Vestimentiferan worms (Lamellibrachia sp.) were also common, but less abun- Page 118 The Veliger, Vol. 427 Nos Table 1 Paratypes of Calyptogena diagonalis Length (mm) Height Width Site 74.2 35.0 30.4 Oregon 210.0 86.6 Soul Oregon 226.0 91.0 63.0 Oregon 231.0 93.0 58.0 Oregon 201.0 es) 5251 Coasta Rica dant than vesicomyids. Seepage of fluids presumed to be rich in sulfide, methane, or both, appears to be related to dewatering of accretionary sediments during tectonic compression along the Cascadia subduction zone (D. Orange, unpublished). Twenty-six individuals of Calyptogena diagonalis were obtained from seep locations from 2900 to 3800 m depth off Costa Rica, during ALVIN dives # 2715, 2719, and 2728. The Costa Rican site (9°42.28”N, 86°4.38”W) is geographically distant but geologically similar to the Oregon locale, as both are positioned in accretionary complexes undergoing sediment compression owing to tectonic subduction, leading to dewatering of sediments and fluid expulsion at the sea floor (Kahn et al., 1996). The chemosynthetic communities in Costa Rican waters include several species of vesicomyid clams, as well as dense aggregations of serpulid polychaete worms and la- mellibrachid vestimentiferans. Specimens from both sites were compared to vesico- myids housed at the U.S. National Museum of Natural History, the Museum of Comparative Zoology at Harvard University, Los Angeles County Museum of Natural His- tory, and the Santa Barbara Museum of Natural History, and all available published descriptions of vesicomyids. Specimens of Calyptogena diagonalis were also sent to other vesicomyid taxonomists for inspection. Owing to the dissimilarity of these specimens from any described extant or fossil vesicomyid species, we concluded that the erection of a new species within the genus Calyptogena is justified. Assignment of the new species to the genus Calyptogena was based both on its morphological simi- larity to congenerics as well as recent unpublished data from molecular studies confirming the close relationship of C. diagonalis to several congeneric species inhabiting the north Pacific (Vrijenhoek, personal communication). SPECIES DESCRIPTION Calyptogena diagonalis Barry & Kochevar, sp. nov. (Figures 1, 2) Holotype: Length—215.0 mm, height—78.0 mm, width—53.1 mm, sex unknown, collected from Costa Ri- Valves Dive # USNM # Left, Right 2644 (880308) Right 2644 (880309) Right 2644 (880310) Left 2663 (880311) Left, Right 2719 (880312) can cold seep, ALVIN Dive # 2719, 14 February 1994; USNM # 880307, Smithsonian Institution U.S. National Museum of Natural History, Division of Mollusks. Paratypes: See Table 1. Type locality: Cold seeps along the Costa Rica subduc- tion zone (9°42.28”N, 86°4.38”W) from 2980 to 3800 m depth. C. diagonalis occurs in clusters of 10 to hundreds of individuals partially buried in sediment, in association) with other vesicomyid clams and bacterial mats. Description: Shell whitish, chalky, and covered by de- hiscent yellowish brown periostracum. Shell large (to 240 mm long, 95 mm high, and 63 mm wide), elongate, in- equilateral, heavy, solid (Figure 1). Valves strongly in- equilateral, with slightly inflated, incurved umbo posi- tioned far anterior (18-20% of length). Anterior margin short, rounded, slightly gaping due mainly to outward flexure of left valve. Anterodorsal margin short, slightly convex. Umbonal cavity moderate; beaks mildly inflated. Posterior margin subangular, pointed near ventral end, es- pecially in Costa Rican specimens. Lunule short, sublan- ceolate, poorly defined anteriorly. Posterodorsal margin elongate, convex, angular near distal end. Escutcheon in- cised steeply immediately posterior to umbo in some specimens. Margin of incision near umbo forming pos- teriorly directed ridge extending toward postero ventral margin. Ligament deeply embedded, highly inflated, dark brown, lanceolate, calcified along hinge plate in large in- dividuals, encompassing ~16—25% (calcified portion) or ~38-—41% (calcified and uncalcified portion) of postero- dorsal margin. Ventral shell margin nearly straight along midpoint in small individuals, mildly concave in large specimens. Sculpture consisting of strong radial ridge from umbo to posteroventral tip, with similar adjacent ribs on some specimens, poorly defined commarginal lir- ations on shell and periostracum, most crowded near an- terior end. Commarginal ridging suggestive of growth rings weakly evident on some specimens. Viewed ven- trally, slight flexure evident along ventral margin, most notably near posterior end. Large individuals with flaky, mostly dehiscent periostracum, except along shell margin, where periostracum overlaps shell margin to provide J. P. Barry & R. E. Kochevar, 1999 Page 119 Posterior Anterior h Left Hinge Right Hinge Figure 1 Diagnostic shell characteristics of Calyptogena diagonalis Barry & Kochevar, sp. nov. Scale bars = 1 cm. a. External view of left shell valve of holotype (USNM# 880307) from Costa Rican collection. b. External view of right shell valve of holotype. c. Internal view of left shell valve of holotype. d. Internal shell valve of paratype (USNM# 880312) from Costa Rican site, with pallial line and adductor muscle scars highlighted in black. e. Dorsal view of holotype. f. Ventral view of holotype. g. Anterior and posterior views of juvenile specimen (paratype; USNM# 880308) from Oregon seeps. h. Hinge structure of left (paratype; USNM# 880311) and right (holotype) valves. Page 120 Figure 2 Repaired internal surface of shell valve from Oregon specimen (paratype; USNM# 880310). Scale bar = 1 cm. a. Overall view of internal valve, with repaired shell area highlighted. b. Mag- nified view of shell repair. complete seal when shell valves closed. Periostracum in- flated, ruffled along anterior to anteroventral margin. Dis- solution of external shell moderate to extreme in some specimens, principally ventral and posterior to umbones. Fenestrations resulting from dissolution occasionally re- paired by localized calcification of inner shell (Figure 2) in some specimens. Right valve with two cardinal teeth beneath umbo (Fig- ure Id, h). Anterior cardinal tooth strongly protuberant, with parallel to subtrigonal borders, pointing ventrally from umbo, and convex to slightly concave medial sur- face. Posterior tooth dorsal to anterior cardinal, protuber- ant, narrow, and slightly bifid in some. Anterior and pos- terior cardinals joined under beak. Three sockets formed by cardinal teeth and umbonal shell margin to accept car- dinal teeth from left valve, central socket deepest, trian- gular. Posterior hinge plate massive, forming nymph sub- tending and partially enveloping ligament; longest rela- tive to shell length in small specimens. Left valve with three cardinal teeth and two sockets to accept central and dorsal cardinal teeth of right valve (Figure lc, h). Anterior cardinal strongly protuberant, narrow to massive, rounded medially; convex anterior margin merges ventrally with hinge plate, flat posterior face contacts anterior cardinal of right valve. Central car- The Veliger, Vol. 42, No. 2 dinal tooth massive, strongly protuberant, trigonal, point- ed to nearly blunt; anterior surface convex, posterior con- tact surface flat. Posterior cardinal positioned dorsally, small compared to other teeth, long, narrow, produced only slightly above hinge plate, nearly horizontal; medial surface nearly smooth to mildly serrate. Internal shell surface porcellaneous with faintly devel- oped radial internal riblets and minor commarginal un- dulations. Anterior adductor muscle scar recessed dorsal- ly and posteriorly, ovately conic to subelliptical, with mi- nor concentric lirations, extending to anterior shell mar- gin in small individuals (Figure 1d). Posterior adductor muscle scar larger, irregularly ovate, teardrop-shaped, or pear-shaped, pointed dorsally, lacking supportive shell sculpture found in anterior scar. Pallial line weakly evi- dent, broad, with sinuous and irregular margins, mildly convex anteriorly and ventrally, and angular posteriorly, forming small pallial sinus (Figure 1d). Soft anatomy: Our general description of the soft anat- omy of Calyptogena diagonalis is based on dissections of two adult-sized individuals. Soft anatomy is generally similar to that reported for C. pacifica Dall, 1891; C. kil- meri Bernard, 1974; C. magnifica Boss & Turner, 1980; Ectenagena extenta Krylova & Moskalev, 1996; and C. packardana Barry et al., 1997. The most conspicuous fea- tures of all six species are the greatly enlarged and often sulfur-colored ctenidia, large and heavily vascularized foot, reduced digestive system, and red, hemoglobin-rich blood, which all relate to their chemosynthetic life style. Mantle and siphons. Mantle lobes bilaterally symmetri- cal, thickened around shell margin, particularly near an- teroventral margin, attached to shell by thick, broad pal- lial muscles. Mantle cavity opens to create pedal gape from ventral margin of anterior adductor muscle to ven- tral anterior margin of incurrent siphon. Thick folds of inner mantle fused posteriorly to form separate incurrent and excurrent siphons; fusion extends dorsally between adductor muscles. Mantle margin thickened and inflated along anterior margin. Band of sensory papillae along thickened anterior mantle margin, similar to that de- scribed for C. magnifica (Boss & Turner 1980). Incurrent and excurrent siphons formed by fusion of the mantle, conical to cylindrical in side view, ovate in cross section, positioned in pallial sinus formed by folds of thickened mantle musculature. Highly developed pal- lial musculature near posteroventral shell margin in si- phonal region, as in C. magnifica (Boss & Turner, 1980). Incurrent siphon larger and more ovate than excurrent siphon. Distal margin of both siphons uneven, slightly serrate, lacking papillae found in Calyptogena packar- dana (Barry et al., 1997). Densely branched structure near base of incurrent siphon functions as filter to reject large particles. Excurrent siphon smaller in cross section than incurrent siphon, with mildly serrate distal margin, thin collar of tissue lining internal siphonal walls to form J. P. Barry & R. E. Kochevar, 1999 one-way valve similar to other vesicomyids (Bernard, 1974; Barry et al., 1997). Ctenidia: Greatly enlarged ctenidia enveloping body along length, from umbonal cavity ventrally through much of shell cavity. Inner and outer demibranchs on each side of body with ascending and descending lamel- lae. Inner demibranchs fused along distal margins to mid- dle of visceral mass and joined posteriorly, isolating in- current and excurrent pallial chambers. Ctenidia variously colored among specimens, from bright sulfur yellow to purplish red, presumably depending upon content of el- emental sulfur in endosymbiont bacteriocytes (Kochevar & Barry, 1994). We have observed ctenidia of C. pack- ardana, C. pacifica, and C. kilmeri to change gradually from sulfur-colored to deep red in laboratory aquaria, ap- parently due to endobacterial oxidation of elemental sul- fur deposits. Micrographs of ctenidial tissues show en- dosymbiotic bacteria similar to those in related chemo- synthetic vesicomyids (R. Kochever, unpublished data). Foot and visceral mass: Foot large, generally conical, highly muscular and distensible, particularly in its ventral half; highly vascularized, deep red owing to hemoglobin content. Dorsally, foot grading into visceral mass, hous- ing large gonad surrounded laterally and ventrally by foot musculature, and dorsally by stomach, digestive gland, intestinal tract, and heart. Labial palps, stomach, and in- testine greatly reduced, similar to other vesicomyids (Ber- nard, 1974; Boss & Turner, 1980; Barry et al., 1997). Reproductive system: Microscopic inspection of gonad samples from several specimens indicates that Calypto- gena diagonalis is gonochoristic. Ovary or testis found directly dorsal to foot and surrounded by foot muscula- ture. No evidence of sexual dimorphism in shells or soft anatomy other than the gonad was observed. REMARKS Calyptogena diagonalis inhabits seep communities asso- ciated with accretionary complex sediments near 2021 m depth off Oregon and from 2900 to 3800 m off Costa Rica. Owing to its broad latitudinal range, we suspect that this species inhabits other sulfide-rich seeps along conti- nental borderlands of the northeastern Pacific. Observa- tions during ALVIN dives found C. diagonalis in clusters including ~10 to 100 individuals, buried partially in sed- iments presumed to be the locus of seeping sulfide-rich pore fluids. Calyptogena pacifica and other vesicomyid clams cohabit seeps with C. diagonalis. The principal diagnostic shell characters of Calypto- gena diagonalis are its large size, elongate shape, diag- onal ridge along the posterior apex of each valve to near the posteroventral shell margin, and somewhat angular posterodorsal margin. Allometric changes in shell morphology, determined from comparisons of three juvenile shells with five to 10 Page 121 adult-sized shells, is evident in several shell characteris- tics of C. diagonalis. Juveniles are considerably less elon- gate (H/L ~0.55 [juveniles] versus ~0.39 [adults]), more inflated (W/L ~0.41 [juveniles] versus ~0.25 [adults]), and less inequilateral (umbo ~24% along length [juve- niles] versus ~19% [adults]). While direct measures of chemosynthetic physiology in C. diagonalis are lacking, all available evidence indi- cates that this species relies on sulfur-oxidizing endosym- biotic bacteria for most or all of its nutrition. All species of vesicomyid bivalves investigated have been shown to derive their nutrition from thiotrophic endosymbionts (Fi- ala-Médioni et al., 1994). C. diagonalis inhabits seep en- vironments and has morphological (size, soft anatomy, endosymbiotic bacteria, elemental sulfur in ctenidial tis- sues, hemoglobin) and behavioral (inhabits seeps, aggre- gates at sites presumed to have sulfide-rich pore fluids) characteristics very similar to known chemosynthetic ves- icomyids. Analysis of stable carbon isotopic ratios of foot tissues for C. diagonalis also suggest chemosynthesis as the primary nutritional pathway, with values near 36%o, similar to confamilial species known to rely on chemo- synthetic production. Geographic Variation in the Morphology of Calyptogena diagonalis Calyptogena diagonalis from sites off Oregon and Costa Rica differs slightly in shell morphology and may warrant the specification of distinct subspecies for the two groups, though additional collections are required to resolve con- sistent differences among these geographical groups. Shells of Oregon specimens are slightly deeper-bodied than their Costa Rican counterparts, with height/length ratios averaging 0.41 (s.d. = 0.02) and 0.38 (s.d. = 0.03), re- spectively (shells > 150 mm length; t-test = ns). Southern material also has a less inflated ligament, and more prom- inent secondary diagonal ridge dorsal to the primary ridge, leading from near the umbo to the angle in the postero- dorsal margin. Dentition is very similar, with minor vari- ation in shape and orientation of cardinal teeth. The ante- rior cardinal of southern specimens is more protuberant and directed more anteriorly, compared to the nearly ver- tical orientation of northern specimens. Comparison with Other Vesicomyids Calyptogena diagonalis is similar to few extant de- scribed vesicomyids, owing principally to its large size. Ectenagena extenta inhabits seep communities with C. diagonalis, but is considerably more elongate (H/L ~ 0.22), and lacks the characteristic diagonal ridge of C. diagonalis (Table 2). Similarly, Calyptogena phaseolifor- mis Métivier et al., 1986, at present known only from the western Pacific, may be confused with C. diagonalis, but is also highly elongate (H/L ~ 0.24). Morphometric ratios of C. diagonalis are more similar to Calyptogena mag- The Veliger, Vol. 42, No. 2 Table 2 Comparison of morphometric ratios among described extant vesicomyid species similar in morphology to Calyptogena diagonalis Height/Length (H/L) Species Mean S.D. N Calyptogena diagonalis, sp. nov. 0.42 0.07 38 Calyptogena elongata 0.45 0.02 12 Calyptogena kilmeri 0.51 0.03 1805 Calyptogena magnifica 0.44 0.02 14 Calyptogena packardana 0.53 0.03 210 Calyptogena phaseoliformis 0.24 0.01 6 Ectenagena extenta 0.22 0.01 4 Width/Length (W/L) Width/Height (W/H) Mean S.D. N Mean S.D. N 0.29 0.07 36 0.69 0.07 35 0.26 0.08 12 0.58 0.19 12 0.33 0.03 1826 0.65 0.06 1825 0.27 0.02 5 0.61 0.05 5 0.31 0.03 210 0.58 0.04 210 0.16 0.01 4 0.65 0.04 4 0.17 0.01 4 0.78 0.05 4 nifica than any materials examined, but these species dif- fer greatly in shell outline and sculpture, ligament size and shape, and periostracum morphology. Valves of C. magnifica are subelliptical with similarly rounded anterior and posterior margins, and lack either the pointed poste- rior margin or diagonal ridge sculpture characteristic of C. diagonalis. The ligament of C. magnifica is massive and much more extensive than C. diagonalis, extending from the umbo to the posterior pedal retractor muscles (~ 48-50% of posterodorsal margin versus 38-41% in C. diagonalis). The periostracum of both species devel- ops complex and inflated folds along the anterior margin, but these appear to be more extensive in C. magnifica as reported by Boss & Turner (1980). In addition, C. dia- gonalis and C. magnifica inhabit different environments and appear to be endemic to cold seeps and hydrothermal vent sites, respectively. Calyptogena elongata Dall, 1916, is similar in shape, but does not reach the large size of C. diagonalis, is thinner, and lacks a diagonal ridge. Ca- lyptogena packardana is generally similar to small spec- imens of C. diagonalis, but is easily distinguished by its very narrow width to length ratio (0.31) and deeply in- cised escutcheon. Finally, two morphologically similar species, Calyptogena kilmeri and Calyptogena soyoae Okutani, 1957, from the northeastern and northwestern Pacific, respectively, could be confused with small C. dia- gonalis. However, like C. packardana, both species lack a diagonal ridge, and have very different hinge dentition than C. diagonalis. The posterior (dorsal) cardinal tooth of the right valve in these smaller species is directed at nearly 45° toward the posteroventral margin. In C. dia- gonalis, this tooth inclined only about 20 to 30 degrees from parallel with the dorsal shell margin. ACKNOWLEDGMENTS We are grateful to Dr. K. J. Boss, E. V. Coan, and P. H. Scott for providing access to specimens and literature im- portant to this project and for their advice concerning the assignment of a new species. Two referees provided valu- able comments concerning the content and organization of the manuscript. Funding for the project was provided by grants from the National Science Foundation, and the Monterey Bay Aquarium Research Institute. LITERATURE CITED Barry, J. P, R. E. KOCHEVAR, & C. H. BAXTER. 1997. Calyp- togena packardana, a new species of vesicomyid bivalve from cold seeps in Monterey Bay, California. The Veliger, 40(4):341-349. BERNARD, E R. 1974. The genus Calyptogena in British Colum- bia with a description of a new species (Bivalvia, Vesico- myidae). Venus 33:11—22. Boss, K. J. & R. D. TURNER. 1980. The giant white clam from the Galapagos Rift, Calyptogena magnifica species novum. Malacologia 20:161—194. DALL, W. H. 1886. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877-1878) and the Caribbean Sea (1879-1880), by the U.S. Coast Survey steamer “‘Blake’’, Lieutenant-Command- er J. R. Bartlett, U.S.N. commanding. XXIX. Report on the Mollusca, Part 1, Brachiopoda and Pelecypoda. Bulletin of the Museum of Comparative Zoology 12:171—318, pls. 1-9. DALL, W. H. 1891. On some new or interesting west American shells obtained from the dredgings of the U.S. Fish Com- mission steamer Albatross in 1888, and from other sources. United States National Museum, Proceedings, 14:173—191. DALL, W. H. & C. T. Simpson. 1901. The Mollusca of Porto Rico. United States Fisheries Commission Bulletin 20:351—524. FIALA-MEDIONI A., V. PRANAL & J. C. COLOMINES. 1994. Deep- sea symbiotic models chemosynthetic based: comparison of hydrothermal vents and cold seeps bivalve molluscs. Pro- ceedings of the 7th Deep-Sea Biology Symposium, IMBC, Crete. GOEDERT, J. L. & R. L. SQuires. 1993. First Oligocene records of Calyptogena (Bivalvia: Vesicomyidae). The Veliger 36(1):72-77. KAHN, L. M., E. A. SILVER, D. ORANGE, R. KOCHEVAR & B. McApoo. 1996. Surficial evidence of fluid expulsion from the Costa Rica accretionary prism. Geophysical Research Letters 23(8):887—890. KANNO, S., K. AMANO & H. BAN. 1989. Calyptogena (Calypto- gena) pacifica Dall (Bivalvia) from the Neogene system in the Joetsu District, Niigata Prefecture. Transactions and Pro- ceedings of the Palaeontological Society of Japan, New Se- ries 153:25-35. J. P. Barry & R. E. Kochevar, 1999 KOcHEvaR, R. E. & J. P. BARRY. 1994. Physiology of vesicomyid clams from Monterey Canyon cold seeps. Transactions of the American Geophysical Union 75(3):203. Koumma, S., R. SEGAWA, T. KOBAYASHI, T. HASHIMOTO, K. FUJIK- URA, J. HASHIMOTO & S. OnTA. 1995. Phylogenetic relation- ships among species of Calyptogena (Bivalvia: Vesicomyi- dae) collected around Japan revealed by nucleotide sequenc- es of mitochondrial genes. Marine Biology 122:401—407. KryLova, H. M. & L. I. MosKaLey. 1996. Ectenagena extenta, a new species of vesicomyid bivalve from Monterey Bay, California. Ruthenica 6(1):1—10. Page 123 NiitsuMA, N., Y. MATSUSHIMA & D. HIRATA. 1989. Abyssal mol- luscan colony of Calyptogena in the Pliocene strata of the Miura Peninsula, Central Japan. Paleogeography, Paleocli- matology, Paleoecology 71:193—203. PEEK, A & R. C. VRIJENHOEK. In press. Evolutionary relation- ships of Deep-Sea hydrothermal vent and cold seeps. Marine Biology. VRUENHOEK, R. C., S. J. ScHUTZ, R. G. GUSTAFSON & R. A. LUTZ. 1995. Cryptic species of deep-sea clams (Mollusca, Bival- via, Vesicomyidae) in hydrothermal vent and cold-seep en- vironments. Deep-Sea Research 41(8):1171-1189. The Veliger 42(2):124—131 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 Histological Description of the Gonad, Reproductive Cycle, and Fertilization of Pisidium amnicum (Miller, 1774) (Bivalvia: Sphaeriidae) R. ARAUJO AND M. A. RAMOS Museo Nacional de Ciencias Naturales (C.S.LC.). José Gutiérrez Abascal 2. 28006 Madrid. Spain Abstract. A detailed study of the reproductive cycle of a Spanish population of Pisidium amnicum (Miiller, 1774) based on monthly histological gonadal samples is presented. Results of the study of the gonadal cycle perfectly match previously reported data on the dynamics of this population, with mature gametes of both sexes present between July and October. Specimens surviving one reproductive cycle undergo a second gametogenesis resulting in a new brood. Nevertheless, most of them die before birth because of the adults’ limited life span. We suggest that cross-fertilization in these freshwater bivalves occurs in summer in the gills or in the suprabranchial chamber instead of in the gonoduct as has been proposed by other authors. [lustrations of all stages of the gametogenic processes, both male and female, as well as of the first stages of embryonic development, are given. Differences between the reproductive strategies of P. amnicum and other sphaeriid species are also discussed. INTRODUCTION All species of the family Sphaeriidae studied are her- maphroditic and incubatory, retaining fertilized eggs in brood sacs developed in the inner gills. Following Mackie (1978), who reviewed the terms ovoviviparity and vivi- parity, these freshwater bivalves are ovoviviparous. The main literature about reproduction, with histological stud- ies of species of the family Sphaeriidae, deals with the genera Musculium Link, 1807 (Okada, 1935a, b, c, 1936; Heard, 1977) and Sphaerium Scopoli, 1777. (Gilmore, 1917; Woods, 1931, 1932; Thomas, 1959; Heard, 1977). The only authors who studied the gonadal histology of species of the genus Pisidium Pfeiffer, 1821, were Heard (1965), who focused on the reproductive strategies of the North American species, and Meier-Brook (1970) who dealt with several European species, not including Pisi- dium amnicum (Miller, 1774). The population dynamics of P. amnicum, the largest species of the genus, has been studied in Germany (Dan- neel & Hinz, 1976), England (Bass, 1979), Canada (Vin- cent et al., 1981), and Spain (Araujo et al., in press), but no data exists about its gonadal development or its repro- ductive cycle from a histological point of view. Recently, Araujo & Ramos (1997) described the gonadal morphol- ogy and evidence of intrafollicular fertilization in several specimens of this species, discussing the possibility of self-fertilization. This paper describes histologically gametogenesis, the cellular types of germinal lineage, reproductive cycle, and * Corresponding author: R. Araujo. Telephone: 914111328, fax: 915645078, e-mail: rafael@mncn.csic.es fertilization process of a Spanish population of P. amni- cum previously studied by Araujo et al., (in press). It shows that this isolated population, the southernmost of the species in Europe and in the world (except for the North African population cited by Kuiper in 1972), is very well adapted to local conditions as indicated by its breeding success compared with North European popu- lations of the species (Araujo et al., in press). MATERIALS AND METHODS Specimens of P. amnicum were collected monthly be- tween June 1990 and May 1991 in the Mifio River in the NW of the Iberian Peninsula. A description of the sample site, sampling methods, and water physico-chemical char- acteristics are provided by Araujo et al., (in press). In the laboratory, each monthly sample was sorted into 1 mm size classes. Four specimens of 6-7 mm were used to determine which of four protocols was best for fixation: (a) immersion of specimens in hot water (6 seconds, 60°C) and fixation with 70% ethanol; (b) direct fixation in formalin saline solution (100 mL 40% formalin, 9 g sodium chloride and 900 mL distilled water); (c) immer- sion in hot water (6 seconds, 60°C) and fixation with for- malin saline solution; and (d) relaxing specimens with menthol (24—72 hrs), and fixation with formalin saline solution. The second method produced the best results and was used in the study. Fixed specimens were re- moved from their shells, dehydrated in a graded ethanol series (30, 50, 70, 96, and 100%) and embedded in par- affin. Sections were made between 7—10 pm and stained with hematoxylin-eosin and Heidenhain’s azan. Those specimens containing shelled larvae in the gills were sub- merged in a mixture of 70% ethanol and 1% acetic acid R. Araujo & M. A. Ramos, 1999 for 3 days in order to decalcify the embryonic shells, as proposed by Okada (1935a). We were normally success- ful when we decreased this time to 24 hr. As the months in which embryos and/or larvae grow inside the maternal gill, and the month of juvenile release were already known (Araujo et al., in press), the fertil- ization and first stages of cleavage were mainly studied in specimens from August and September. Thus, we stud- ied 13, 14, six, four, and two specimens from September, August, June, July, and October, respectively, and one from the rest. In order to analyze the gametogenetic stag- es, adult specimens of 7—8 mm were observed. Once we knew the months of greater gametogenic activity and the size at which the species reaches sexual maturity, speci- mens of all size classes from these months were ob- served. RESULTS P. amnicum is a simultaneous hermaphrodite, as both male and female gametes develop in the gonad of each sexually mature specimen at the same time. Male and female tissues are organized in follicles, the male fraction being much larger than the female and occupying the an- terior part, while the small ovarian fraction is posterior (Figure 1A, B). Although there are no hermaphroditic fol- licles, both gonadal fractions overlap near the hermaph- roditic ducts, which are exterior and lateral, running par- allel to the cerebro-visceral connectives (Figure 1C). As has been demonstrated by Araujo & Ramos (1997), male and female gametes are commonly found together in this area. Gametogenesis The simultaneous presence of mature spermatozoa and ripe oocytes was restricted to the period from July to October. Specimens born in April-May undergo game- togenesis in summer. The larger gravid specimens found from June to September were the survivors of the pre- vious year in which the gametogenetic activity restarts, although most of them die before the new cycle finishes. Fertilization and settlement of the zygotes occur in late summer (August and September) in both newborn spec- Figure 1 Gonad of P. amnicum. A. Longitudinal section of an 8-9 mm specimen from August. The anterior part is at left. Scale bar = 1 mm. B. Transverse section through the gonad of a 4-5 mm specimen from August. Scale bar = 0.5 mm. C. Transverse sec- tion through the end of the gonad showing the hermaphroditic ducts and the connectives. Scale bar = 50 wm. c, connectives; f, foot; hd, hermaphrodite ducts; i, intestine; ig, inner gill; 0, ovary; r, rectum; s, stomach; t, testis. Res Ee ; "t AS ¢) S Jae Lass y : ee ¢ a \ ~ : zh ‘2 Week Dor 0K: cA coats The Veliger, Vol. 42, No. 2 RED NY, sohase sIer - = a) On Jatt “ “sheystin “SBee4 | Mat. | Ve Phe: Figure 2 Spermatogenesis of P. amnicum. A. Spermatogonia (arrows). Scale bar = 30 wm. B. Spermatocytes I. The arrow shows the anaphase of first division. Scale bar = 20 wm. C. Spermatocytes II. The arrow heads show the metaphase and anaphase of second division. Scale bar = 12 ym. D. Section of testis showing different stages of spermato- genesis. Scale bar = 40 pm. E. Testis of a specimen from January. Scale bar = 75 pm. imens and ones from the previous generation. Gravid an- imals collected during these months still presented many mature oocytes and spermatozoa. Maturation of the male gametes occurs in specimens from May to October, the light in the middle of the fol- licles increasing at the same time as spermiogenesis oc- curs. In October, the follicles are empty, with many sper- matogonia growing from their walls; this evacuation phase is followed by a proliferation phase from Novem- ber to April, with full and compact follicles without light inside. Spermatogonia were present in the testis from April to December. They are large cells (cell diameter = 10-12 ym) with very little cytoplasm, adhering to the follicle walls (Figure 2A). The nucleus is full of chromatin gran- ules so it is difficult to ascertain the number of nucleoli. First order spermatocytes (Figure 2B) are cells with a diameter of 7—8 jm, scant cytoplasm, and the chromatin spreading in a large nucleus. When these cells are found in the middle of the follicle, the chromatin appears at the edge of the nucleus. Second order spermatocytes are smaller (4—6 pm), and we have also found them in meta- phase and anaphase II (Figure 2C) prior to development of the spermatids (2-3 wm). The condensation of the ge- netic material at this last stage makes it easier to see the cytoplasm than at previous stages. The head of the mature spermatozoa is about 5—6 wm, and the tail is very difficult to see. In July all the male cellular lineage is easily ob- served in the same follicle (Figure 2D). In specimens between 4 and 6 mm from August and September (those born in May of the same year), the spermiogenesis is in the latter stages. From July to Sep- tember, spermatogenesis has restarted in the largest (> 7 mm) and old specimens. In October the testis is massive, R. Araujo & M. A. Ramos, 1999 full of polyhedral cells (Figure 2E), with conjunctive tis- sue in the interfollicular space. The testis remains at this stage for the rest of the winter until late spring and sum- mer when spermatogenesis restarts. The process in the female tissues is very similar. Previtellogenetic oocytes are present in August and September in specimens born in May of the same year. After spawning, the ovogenesis restarts in the oldest specimens. The oogonias adhere to the follicle walls. They are cells similar to the sperma- togonias with a diameter of about 20 um. They have a large nucleus that occupies most of the cell, having a refringent light nucleolus and scattered particles of chro- matin, very visible both in azan and hematoxylin-eosin stained slides. A second nucleolus and accompanying cells attached to the oogonias can be observed (Figure 3A). The previtellogenic oocytes are still attached to the follicle wall; they are about 20 pm in diameter with a 15 tum nucleus, showing one or two nucleoli (Figure 3B). Previtellogenic oocytes still maintain accompanying cells (Figure 3C) and sometimes have an anphinucleolus in the nucleus. At the end of the previtellogenic stage, the het- erogeneity of the nucleus increases and the accompanying cells disappear. At this state, spherical corpuscles resem- bling the nuclei of the oocytes can be seen in the female follicles of specimens up to 6—7 mm (Figure 3D). At the beginning of the vitellogenesis, the oocyte size increases and the nucleus still contains the nucleoli and the an- phinucleolus (Figure 3E). The ripe oocytes, now free from the acinar wall, are about 40-60 wm and have a nucleus of 12 wm with one or more nucleoli, the larger ones sometimes having an anphinucleolus (Figure 3F). In specimens from September, the above mentioned spherical corpuscles only appear in specimens over 4 mm. No signs of reabsorption were observed within the Ovarian tissues after ova release, but differentiation of the oogonia immediately follows this event. During August and September, the largest amount of mature male and ripe female gametes appears, often oc- curring together within female follicles near the her- maphroditic duct, allowing the occurrence of self-fertil- ization. P. amnicum becomes sexually mature at a shell length of about 3—5 mm, the male gonad probably maturing first (we detected one mature specimen of 3—4 mm) and the female later (4-5 mm). Testis maturation proceeds from the anterior to the posterior area, and from the center to the periphery. Fertilization and Brooding Only once did we observe a mature oocyte in the her- maphroditic duct, and it had not been fertilized. No gravid specimens appeared among those from June and July that were studied histologically. In the speci- mens from August, three histological observations indi- cated recently fertilized oocytes (zygotes) in the inner Page 127 gills. In one case, the zygote had a nuclear membrane with many nucleoli and the male pronucleus (Figure 4A). In the other two, the zygote had lost the nuclear mem- brane, the male pronucleus was located at the edge of the cell, and the female pronucleus was in meiotic metaphase (Figure 4B, C, D). Embryos at several stages of cleavage were also found. Figure 4E, F shows respectively, one embryo in a stage prior to blastula and one blastula, and Figure 4G shows the arrangement of the embryos within the parental gill. During August and September when the first stages of the embryos were observed, several gravid specimens carried embryos in even more advanced stages, i.e., the ova had been fertilized and had begun the cleavage pro- cess once they fell between the gill filaments. The germ cells from which the gonads develop were observed during all the cleavage of the embryo from the blastula (Figure 5A) to the prodissoconch larvae (Figure 5B) (the one shelled and still covered by the brood sac within the marsupium). DISCUSSION The simultaneous occurrence of mature male and female gametes in P. amnicum specimens over 5 mm collected from July to October, particularly in August and Septem- ber, means that this species, like the other of the family Sphaeriidae studied by Meier-Brook (1970), is a simul- taneous hermaphrodite (see Araujo & Ramos, 1997). This agrees with Meier-Brook’s (1970) point that the simulta- neous occurrence of mature gametes of both sexes in 3 mm specimens of Musculium heterodon (Pilsbry) sug- gests that the species is, for most of its life, a simulta- neous hermaphrodite, although the male fraction probably matures before the female one. This agrees with Okada’s (1935b) concept of protandric maturation. However, fol- lowing Hoagland (1984), the term protandric should de- scribe animals that can change their sex from male to female without reverting to male. Therefore, this term does not apply to the Sphaeriidae. Regarding the reproductive habits of this species, his- tological analysis of the gonads of P. amnicum confirms the reproductive strategy (semelparous and univoltine) postulated by Araujo et al. (in press) on the basis of pop- ulation dynamics. This strategy differs from the rest of the European species of the genus, which, according to Meier-Brook (1970), are iteroparous and multivoltine. After spawning (gamete release), there is a growth pe- riod of the germinal cells in the gonads, which is very long and slow in P. amnicum, especially in the female. In P. lilljeborgii Clessin, 1886, this growth is quicker, allowing two reproductive periods in the same year. The lack of mature female gametes inmediately after spawn- ing (as occurs in P. lilljeborgii) and the long gravidity period in P. amnicum (nearly 9 months) compared with The Veliger, Vol. 42, No. 2 Figure 3 Oogenesis of P. amnicum. A. Oogonia with accompanying cells (arrow head). Scale bar = 20 pm. B. Oocytes. Scale bar = 30 pm. C. Oocyte with accompanying cells (arrow heads). Scale bar = 20 pm. D. Refringent corpuscles in the ovary (specimen from June). Scale bar = 30 ym. E. Oocytes with several nucleoli and anphinucleolus. Scale bar = 30 ym. F. Ripe ova in the ovary of a specimen from August. Scale bar = 40 wm. other species (2 months in P. lilljeborgii) also help to explain this difference. Our results confirm the absence of an extended life span or a successful second brood in P. amnicum from southern areas. Although the gametogenetic processes are reactivated once the single reproductive cycle occurs, this cycle is not completed because the species’ limited life span (15 months, Araujo et al., in press) means specimens die before the brood is fully developed. In other words, those that die in summer are always gravid. This phe- R. Araujo & M. A. Ramos, 1999 Page 129 Figure 4 Fertilization of P. amnicum. A. Zygote in the gill filaments. B. Meiosis of a zygote in the gill filaments. C, D. Consecutive sections of a zygote in the gill filaments. E. Mitosis of an embryo. F. Blastula. From A to FE scale bars = 30 pm. G. Arrangement of the recently formed zygotes in the gill. Scale bar = 120 pm. mp, male pronucleus; fp, female pronucleus; A, B, C. Specimens from September; E, E G, from August. nomenon was cited for the same species in England by North American species. In P. dubium, oogenesis is a Bass (1979). continuous process throughout the year, being more ac- Interesting differences also exist between the gameto- tive at the beginning of the summer, but mature ova are genesis of P. amnicum and P. dubium (Say), its vicariant present throughout the year. Spermatogenesis in this spe- Page 130 The Veliger, Vol. 42, No. 2 Figure 5 A. Blastula with germ cells (arrow heads). Scale bar = 30 pm. B. Prodissoconch larvae with germ cells (arrows). Scale bar = 75 wm. f, foot; pg, pedal ganglia. cies only occurs in summer, with mature male gametes appearing only during a short period in mid summer (Heard, 1965). Other comparisons between the reproduc- tive strategies of Spanish and other European populations of P. amnicum have already been discussed in Araujo et al. (in press). The similarity in the way the testis matures in P. amnicum and Musculium heterodon is also interest- ing, as in both species it occurs from the anterior to the posterior region. In the Japanese species, mature sper- matozoa are present all year, although a smaller number is observed in winter (Okada, 1935a). Lucas (1965) proposed that the number of nucleoli dif- ferentiates spermatogonia with two nucleoli from oogonia with only one. In the Spanish P. amnicum specimens, it was impossible to test this hypothesis because the large amount of chromatin granules in the nucleus of the sper- matogonia makes it difficult to determine the number of nucleoli present. Lucas (1965) also cited the difficulty of observing the spermatocyte II due to the speed of the second meiotic division in mollusks. However, we ob- served this cellular stage in P. amnicum, and Okada (1935a) did so in Musculium heterodon. Regarding the morphology of the spermatozoa, there are conflicting data in the literature. For Monk (1928) the spermatozoa of Sphaerium notatum (Sterki, 1927) lacked a tail, probably due to the difficulty of observing these structures. For Okada (1935a) they were very easy to observe in Mus- culium heterodon. In P. amnicum, the tails of the sper- matozoa are very difficult structures to identify. The existence of primary oocytes in spring and autum in M. heterodon (Okada, 1935a) corresponds to the pe- culiar type of reproduction in this species (and all the Sphaeriinae) in which the embryos are present in the gills of the maternal specimens all year. According to Okada (1935a), the size of the mature ovum (“primary oocyte’’) in M. heterodon is about 40 wm, with an eccentric nucleus of 15-20 pm. Woods (1932) illustrated a mature ovum of about 70 wm in Sphaerium striatinum (Lamarck), while the maximum size detected in Spanish P. amnicum is about 60 ym and corresponds to metaphase I oocytes. Okada (1935a) cited the presence of accessory plasmosomes, a common struc- ture in the nucleus of growing oocytes in mollusks. He suggested that the increase in these structures is transitory due to the increase in nuclear contents; that explains its absence in the first and final stages of oocyte growth. For Stauffacher (1894, in Okada, 1935a), these accessory nu- cleoli arise from the budding of the main nucleolus. How- ever, in P. amnicum these nucleoli are observed in grow- ing and ripe oocytes, and, indeed, in the zygotes recently settled in the gills (Figure 4A), making it difficult to ac- cept such an explanation. As regards the oocyte accom- panying cells, our observations also agree with Okada (1935a) in the sense that their relation with the oocyte become less clear as the oocyte grows, supporting Woods’ (1932) idea that one or several epithelial cells surrounding the growing oocyte are joined to it, totally or partly, aiding oocyte formation. No references have been found in other species of Pis- idium regarding the refringent corpuscles located in the mature female follicles of P. amnicum. According to Ituarte (1997), who studied the oosorption process in Eu- pera platensis Doello Jurado, 1921, a South American sphaeriid, the germinal vesicle of degenerative oocytes are not phagocytosed as occurs with the cell cytoplasm; they remain in the lumen of the follicle. Assuming that the corpuscules may be nuclei of degenerated oocytes, this phenomenon could explain their presence in the go- nad of P. amnicum if other signs of oosorption are found, and their presence indicates a recent spawning process. Page 131 R. Araujo & M. A. Ramos, 1999 Our results suggest that cross-fertilization in P. amni- cum takes place in the inner gills of the parents, where the meiosis of the ova starts. Nevertheless, Araujo & Ra- mos (1997), reported and illustrated one specimen in which most of the oocytes, with a diameter between 40— 60 wm, had lost the nuclear membrane and appeared in meiotic metaphase inside the ovary. These authors also reported several cases suggesting that intrafollicular fer- tilization can occur in P. amnicum. Although no gravid specimens appeared among those histologically studied from June and July in this study, some gravid ones were detected by dissections of survi- vors from a previous cycle (see Araujo et al., in press), suggesting that the fertilization process might begin be- fore August. ACKNOWLEDGMENTS Histological sections were prepared by Purificacion Ar- ribas and Josefina Cavanilles in the laboratory of the Mu- seo Nacional de Ciencias Naturales (Madrid, Spain). We are indebted to Dr. Emilio Roldan and his wife for the facilities provided during the monthly sampling field trips, and to Diego Moreno for his help in the field. Thanks also to Dr. P. Penchaszadeh and S. Jiménez for their comments and help in the preparation of the man- uscript, and to Lesley Ashcroft for reviewing the English version. Two anonymous reviewers made interesting comments which improved the manuscript. This work re- ceived financial support from the Project “‘Fauna Ibérica Il” (SEUI, DGICYT PB89 0081). LITERATURE CITED ARAUJO, R. & M. A. Ramos. 1997. Evidence of intrafollicular fertilization in Pisidium amnicum (Miller, 1774) (Mollusca: Bivalvia). Invertebrate Reproduction and Development 32(3):267—-272. ARAUJO, R., M. A. Ramos & R. MOLINET. In press. Growth pat- tern and dynamics of a Spanish population of Pisidium am- nicum (Bivalvia: Sphaeriidae). Malacologia. Bass, J. A. B. 1979. Growth and fecundity of Pisidium amnicum (Miiller) (Bivalvia: Sphaeriidae) in The Tadnoll Brook, Dor- set, England. Journal of Conchology 30:129—134. DANNEEL, I. & W. Hinz. 1976. Zur Biologie von Pisidium am- nicum O. FE Miiller (Bivalvia). Archiv fiir Hydrobiologie 77(2):213-225. Gitmorg, R. J. 1917. Notes on reproduction and growth in certain viviparous mussels of the family Sphaeriidae. The Nautilus 31:16-30. HEARD, W. H. 1965. Comparative life histories of North Ameri- can pill clams (Sphaeriidae: Pisidium). Malacologia 2(3): 381-411. HEARD, W. H. 1977. Reproduction of fingernail clams (Sphaeri- idae: Sphaerium and Musculium). Malacologia 16(2):421— 455. HOAGLAND, K. E. 1984. Use of the terms protandry, protogyny, and hermaphroditism in malacology. American Malacolog- ical Bulletin 3(1):85-88. IruarRTE, C. E 1997. The role of the follicular epithelium in the oosorption process in Eupera platensis Doello Jurado, 1921 (Bivalvia: Sphaertidae): a light microscopic approach. The Veliger 40(1):47—54. Kuiper, J. G. J. 1972. Une récolte de Pisidium dans le Moyen Atlas. Résultats de la mission biologique au Maroc de l'Université de Gand, Belgique. Publication no. 9. Basteria 36(2—5):189-198. Lucas, A. 1965. Recherche sur la sexualité des mollusques bi- valves. Ph.D. Thesis. Faculté des Sciences de L Université de Rennes. 135 pp. Mackie, G. L. 1978. Are sphaeriid clams ovoviviparous or vi- viparous? The Nautilus 92(4):145—147. MEIER-BROOK, C. 1970. Untersuchungen zur Biologie einiger Pisidium-Arten (Mollusca; Eulamellibranchiata; Sphaeri- idae). Archiv fiir Hydrobiologie/Supplement 38(1/2):73- 150. Monk, C. R. 1928. The anatomy and life-history of a fresh-water mollusk of the genus Sphaerium. Journal of Morphology 45(2):473-503. OKapDA, K. 1935a. Some notes on Musculium heterodon (Pils- bry), a freshwater bivalve I. The genital system and the ga- metogenesis. Science Reports of the Tohoku Imperial Uni- versity, Biology 9:315—-328. Oxapa, K. 1935b. Some notes on Musculium heterodon (Pils- bry), a freshwater bivalve II. The gill, the breeding habits and the marsupial sac. Science Reports of the Tohoku Im- perial University, Biology 9:373-389. OxkapbaA, K. 1935c. Some notes on Musculium heterodon (Pils- bry), a freshwater bivalve III. Fertilization and segmentation. Science Reports of the Tohoku Imperial University, Biology 10:467—483. OxKaDA, K. 1936. Some notes on Sphaerium japonicum biwaense Mori, a freshwater bivalve IV. Gastrula and fetal larva. Sci- ence Reports of the Tohoku Imperial University, Biology 11: 49-68, 2 plates. STAUFFACHER, H. 1894. Eibildung und Furchung bei Cyclas cor- nea Lam. Jenaische Zeitschrfit fiir Naturwissenschaft, 28. Tuomas, G. J. 1959. Self-fertilization and production of young in a sphaeriid clam. The Nautilus 72(4):131—140. VINCENT, B., G. VAILLANCOURT & N. LAFONTAINE. 1981. Cycle de développement, croissance et production de Pisidium am- nicum (Mollusca: Bivalvia) dans le Saint-Laurent (Québec). Canadian Journal of Zoology 59:2350—2359. Woops, FE H. 1931. History of the germ cells in Sphaerium stria- tinum (Lam.). Journal of Morphology 51(2):545—595. Woops, EF H. 1932. Keimbahn determinants and continuity of the germ cells in Sphaerium striatinum (Lam.). Journal of Morphology 53(2):345-365. The Veliger 42(2):132-151 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 The Eastern Pacific Sportellidae (Bivalvia) EUGENE V. COAN Department of Invertebrate Zoology*, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118-4599, USA Abstract. The taxonomy of the eastern Pacific species that have been allocated to the bivalve family Sportellidae is reviewed. All taxa are members of the tropical fauna. The genus Basterotia is represented by five species: B. californica Durham, 1950, here reported from the Recent fauna for the first time; B. obliqua and B. panamica, two new species, the latter the most common eastern Pacific species of Basterotia and here reported to brood its young; B. peninsularis (Jordan, 1936) (of which B. hertleini Durham, 1950, and B. ecuadoriana Olsson, 1961, are synonyms); and B. quadrata (Hanley, 1834) (of which Poromya granatina Dall, 1881, is a synonym). The new genus Basterotina is described, with the new species B. rectangularis as its type species; Basterotina americana (Dall, 1900) from the Plio-Pleistocene of Florida is also a member of this genus. Ensitellops is represented by E. hertleini Emerson & Puffer, 1957 (of which E. pacifica Olsson, 1961, is a synonym). Fabella is represented by F. stearnsii (Dall, 1899) (of which Sportella duhemi Jordan, 1936, is a synonym). Sportella californica Dall, 1899, is an Orobitella (Galeommatoidea: Lasaeidae), and Ani- sodonta pellucida Dall, 1916, is based on a juvenile mactrid, probably Simomactra falcata (Gould, 1850). INTRODUCTION The Sportellidae is one of four bivalve families currently placed into the Cyamioidea, the others being the Cyami- idae (with the Gaimariidae and Perrierinidae regarded as synonyms), the Bernardinidae, and the Neoleptonidae. This complex is much in need of careful analysis to test whether it really represents a clade and whether all of these families, including the Sportellidae, do as well. The purpose of the present study was to review the eastern Pacific taxa that have been allocated to the Spor- tellidae. All the species that remain in the family are members of the Panamic fauna, occurring only south of central Baja California. A short summary of the results of this study was presented in the journal of the San Di- ego Shell Club (Coan, 1997). Very little is known about the anatomy or ecology of the Sportellidae. Like other cyamioideans, they have pos- terior incurrent and excurrent openings, but lack or have only very short siphons. There are only four anatomical accounts of genera that have been placed into this family, and one of them seems instead to represent a galeom- matid. Fischer (1860:23—35, 1886:194) described but did not illustrate the soft parts of Basterotia quadrata (Hanley, 1843) (as “‘Eucharis’’). Key features mentioned were a mostly fused, strongly papillate mantle, with an oval ped- al aperture; separate posterior inhalant and exhalant ap- * Mailing address: 891 San Jude Avenue, Palo Alto, Califor- nia, 94306-2640, USA; also Research Associate, Santa Barbara Museum of Natural History and Los Angeles County Museum of Natural History; gene.coan @sierraclub.org. ertures, the latter forming a short siphon; ctenidia with two demibranchs, the inner larger; and an elongate, ex- tensible, vertically deployed foot, with a linear groove and pit, presumably for a byssus. This account is probably the source of the information Dall (1899:875) provided about this species. The poorly known genus [soconcha Pelseneer, 1911, ex Dautzenberg & Fischer ms, was first made available in an anatomical description and figure of its monotypic species, I. sibogai. It was described as having a single posterior aperture, a long ventral aperture, and an incom- plete anterior inhalent aperture. The foot was described as being pointed and having a strong byssus. The ctenid- ium had only one demibranch, in which were found in- cubating eggs (Pelseneer, 1911:47—48, pl. 16, fig. 12, pl. 17, fig. 1; see also Prashad, 1932:173, pl. 9, figs. 9-12). It was placed in the Sportellidae, with question, by Chav- an (1969:541), probably because of its entirely external ligament, but its anterior inhalent aperture suggests that a better placement might be the Galeommatidae, close to Benthoqueta Iredale, 1930 (type species, by monotypy: Turqueta integra Hedley, 1907:364, pl. 66, figs. 7—10), the soft parts of which were studied by Ponder (1968: 128-131, figs. 5-7). Ponder (1971:127, 129-131, figs. 34, 37, 38) described and figured the soft parts of Anisodonta (Tahunanuia) alata (Powell, 1952:170). The inhalent and exhalent openings are posterior, with very short siphons for both. There is a ventral pedal gape. The foot lacks a byssal gland. The inner demibranch is more than twice the size of the outer, and labial palps were figured as being small. Juveniles were found attached to the inside ventral shell margin, each by a single byssal thread (Figure 1). E. V. Coan, 1999 Page 133 Figure 1 Animal of Anisodonta alata (Powell), after Ponder (1971). Finally, Kay (1979:546, fig. 178C; 551; 548, fig. 179L) described and figured the animal of Basterotia angulata (Dall, Bartsch & Rehder, 1938) (originally described and in Kay [1979] as ‘‘Anisodonta’’)'. She mentioned a me- dium-sized foot without a byssus, posterior inhalent and exhalent apertures, the inhalent with a short siphon (op- posite to the situation in B. quadrata). The inner demi- branch is much larger than the outer (Figure 2). Unfortunately, nothing is known of the anatomy of liv- ing species that have been allocated to Sportella (here to Fabella). FORMAT In the following treatment, each valid taxon is followed by a synonymy, information on type specimens and type localities, notes on distribution and habitat, and an addi- tional discussion. The synonymies include all major accounts about the species, but not most minor mentions in the literature. The entries are arranged in chronological order under each species name, with changes in generic allocation from the previous entry, if any, and other notes given in parentheses. The distributional information is based on specimens I have examined, except as noted. For many species, the available habitat information is sparse; I have summa- rized the data available. References are provided in the Literature Cited for all works and taxa mentioned with dates. The following abbreviations for institutions and collec- tions are used in the text: ANSP—Academy of Natural Sciences of Philadelphia, Pennsylvania, USA; CAS— California Academy of Sciences, San Francisco, Califor- nia, USA; LACM—Natural History Museum of Los An- ' A junior homonym but perhaps also a junior synonym of B. angulata (H. Adams, 1871:789), originally proposed as Eucharis. Figure 2 Animal of Basterotia angulata (Dall, Bartsch & Rehder), after Kay (1979). geles County, California, USA; MCZ—Museum of Com- parative Zoology, Harvard University, Cambridge, Mas- sachusetts, USA; MNHN—Muséum National d’ Histoire Naturelle, Paris, France; NMW-—Naturhistorische Mu- seum, Wien (Vienna), Austria; SBMNH—Santa Barbara Museum of Natural History, Santa Barbara, California, USA; UCMP—University of California Museum of Pa- leontology, Berkeley, California, USA; UF—Florida Mu- seum of Natural History, University of Florida, Gaines- ville, Florida, USA; UMML—University of Miami Ma- rine Laboratory (Rosenstiel School of Marine and At- mospheric Sciences), Miami, Florida, USA; USNM— United States National Museum collection, National Museum Natural History, Smithsonian Institution, Wash- ington, DC, USA; Skoglund Coll.—collection of Carol C. Skoglund, Phoenix, Arizona, USA. DIFFERENTIATING CHARACTERS A comparative listing of key characters is given in Table 1. A few characters merit additional explanation. The position of the beaks is given as a percentage of their distance from the posterior to the anterior end. Thus, the beaks of Ensitellops hertleini are near the anterior end (80%), whereas those of Fabella stearnsii are just pos- terior to the midline (40%). The length of the external ligament is given relative to overall shell length. Aside from Ensitellops hertleini, which lacks an external ligament, most ligaments are of moderate length (9—13%). Ligaments are short, about 7%, only in Basterotia quadrata and Basterotina rectangu- laris. The internal part of the ligament is generally located on the medial surface of the nymph (no nymph present in Ensitellops hertleini). It is fairly broad in Basterotia californica, a narrow band adjacent to the external liga- ment in most taxa, or ‘“‘small,’’ being restricted to an area near the beaks, as in B. quadrata and Fabella stearnsii. The Veliger, Vol. 42, No. 2 Page 134 SI el OV 6C 9ST ae) OTT Tvl OTL LOl CTI so] Oy) = wu -loeg ‘YyS8usy yseq xe yus00y 1aq -wnN poururexg sjo7T [vj0L “pres [ey -ud0 [ews AIOA poaino + ‘pres A[UOAQ ‘Jue ode] “pres yeuss + uUIdIvU poains osuly AJUSAD «UO “pied “jue Iauioo Ad uIsIeuw piemo} osuly poyoyop ‘yue uo IouIOD Ad preMmo} poaino pooayop ‘sunoofoid IouIOD Ad preMmo} MOLIeU poioyop ‘Sunoofoid Iaulod Ad a1e3 premo} -UO]2 “LOA paopop ‘suroafoid IouIOD Ad MOLICU pieMmo} ‘sunoal payoyop -o1id Apystys ajesuojo poeaino “Z110Y A[UdA9 ‘Sunoofoid uly y1991/4}00} Tet[ed jeurpie) AT “pres [en -U99 OBI] + ‘pies ‘jue “pour sunool -oid ApYystys MO] aes -UO]O “VIDA ‘poaino ‘Sunooloid ae3 -UO]O “LIOA ‘Suroafoid 21°38 -UO]O “VIDA ‘Sunoofoid aylsod ‘Bunoofoid a1e3 -UO]a “LIOA ‘sunoofoid 4291/4300} jeulpreD —AY [yeurs MOLIeU ‘qyesuojo MOLECU [jews MO.LICU uinIpour 0} MOLIeU MOIIEU opim quo SI] jeusaquy yysus] o}elopour ouou yoys yoys ysud] ayelopour yysud] a}e1opoul ysud] aye lopoul yisud] ayelopour JUSSI] yeusoyxy Ayuo syooyo yIMoIs dSO] -msnd Ayjasieds asojmsnd Ajasuop asormisnd Ajasuap syeoq eau aso] -njsnd Ayasieds asornisnd Ajasuap soqny -snd osreds AoA A|UO syooyo yIMoIs ainjdjnos ouou ouou ayeuleo 0} poysue Ajdieys ayeuLes 0} poysue Aydreys a[sue popunol o}eu -11e9 0} dieys 0} popunol ajsue syeaq Ieau a[3ue jYSIIS a]sue popunol uoneoreulop odojs Joliojsod /penuad %OV %O08 %OL %OL %SL M8 %O8-SL %0L-09 pua ‘que 0} JANLIOI syeoq jo uonIsog (L'1) 27e8A0 (¢°Z) ayesuoja (S-Z) qejns -ueyDdI-a}eSUOTA (V'1) [euosI.N-3}eA0 (LT) Jeprozoden -9]BAO 0} JJeAO (9'T) [ep -10Z9dv.1-3]8A0 (9'T) anbryjqo-a}e8Ao (9'T) a}eSUOTI-a}BAO (y/1) adeys ususvaqs DI]9q0A LTE YEE sdopjajisuq SLUDINSUDJIAL puyo1ajspg pipaponb DIOAIISDG stuppnsuluad DIJOLIISDG paiuDpupvd DIJOAAISDG pnbyqo DIIOAAISDG DIUAO {IDI DIOLIISDG aeplT[aviods oyloeg Ulo\sey UT SIojORIeYS SuNeNUsIOJIG T 91981 E. V. Coan, 1999 The morphology of the hinge teeth are best understood by comparison to the line drawings (Figures 30—42). The pallial lines are more or less evenly curved in some species (Figures 30, 31, 41, 42), or are deflected toward the postero-ventral margin in most taxa (Figures 32, 34—40). SYSTEMATIC ACCOUNT Cyamioidea Sars, 1878:iv, 65? SPORTELLIDAE Dall, 1899:875 (= Basterotiidae Cossmann, in Cossmann & Peyrot, 1909:25, 133 [of reprint]) Basterotia Mayer, in Hornes, 1859 Harlea Gray, 1842:78, genus without named species. Type species: Corbula quadrata “Hinds, 1843,” by subse- quent designation of Smith, 1890:303. Nomen oblitum; see Discussion. Eucharis Récluz, 1850:167. Type species: Corbula quadrata “Hinds, 1843,” by original designation. Non Eucharis Latreille, 1804:175 (Hymenoptera). Basterotia Mayer, in Hornes, 1859:71—72. Type species: B. corbuloides Mayer, in Hornes, 1859:71—72, by mono- typy. Badenian, Middle Miocene; Mikulov, Czech Re- public. (See also Hornes, 1870:40—41, pl. 3, fig. 11) Basterotella Olsson & Harbison, 1953:97. Type species: Pleurodesma floridana Dall, 1903:1630, pl. 57, fig. 30, by original designation. Late Pliocene or early Pleisto- cene of Florida. Discussion: As discussed by Vokes (1981:157, 160), Harlea was proposed by Gray (1842) and then resurrect- ed by Smith (1890), but it has never actually been used and is thus a “forgotten name” (nomen oblitum). Vokes (1981:157) stated that ‘“‘the International Commission on Zoological Nomenclature has been requested to suppress the name Herlea for purposes of the Law of Priority, but not for the Law of Homonymy,”’ but he evidently never filed a request with the Commission to do so. While a petition to suppress this never-used senior synonym would be called for under the present Code (Art. 79), the soon-expected, newly revised Code will not require pe- titions in such clear-cut situations (P. Tubbs, e-mail, 12 June 1997). As discussed under this species below, the name Bas- terotia quadrata was first made available by Hanley in early 1843, some months before Hinds published it. Basterotella was differentiated from Basterotia, s.s., on the basis that it lacks a sharp angle between the central and posterior slopes, has a longer nymph, and is less pus- tulose externally. However, the holotype of the type spe- cies of Basterotella, Pleurodesma floridana, has a sharp angle near its umbones as do other specimens I have ex- ? Chavan (1969:537) and Ponder (1971:125) credit this family- level name to Philippi (1845), but I find only the genus Cyamium in Philippi (1845:50—51). As far as I have been able to discover, Sars (1878) is the earliest author to use this name. Page 135 amined (UF 9846), and Recent taxa belonging to Baster- otia have every combination and degree of these three characters. For example, B. miocenica Vokes, 1981:161— 163, described from the late Lower Miocene Chipola For- mation of Florida, was unequivocally assigned to Baster- otella, but varies from having a rounded to a sharp angle (USNM 298652, 298653, 198654; UF 77588). Little point would thus be served in attempting to allocate the species of Basterotia to these two subgenera, leaving ei- ther a number of species without subgeneric homes or the subgenera so broadly and complexly defined that they would probably be paraphyletic. Description: Shell ovate, ovate-elongate, ovate-trapezoi- dal, to ovate-trigonal; beaks closer to the posterior end, low to inflated. Central slope set off from posterior slope by an angle in some; angle carinate in some. Surface with irregular growth checks, pustulose in most. Right and left valves with projecting anterior cardinals, that in left valve positioned posterior to that of right valve. External por- tion of ligament of moderate length to very short in some; nymph weak to strong; internal portion of ligament a rel- atively small, triangular area adjacent to external portion, often extending to a pit under beaks. Left valve often with an escutcheon; right valve either lacking or with a much less conspicuous escutcheon. Pallial line entire. Carter & Lutz (1990:7, pl. 61) figured the shell struc- ture of the Hawaiian Basterotia lutea (Dall, Bartsch & Rehder, 1938:124—125, pl. 34, figs. 7, 8) (as ‘“‘Anisodon- gah); Basterotia californica Durham, 1950 (Figures 3—5, 30, 31) Basterotia californica Durham, 1950:94, 170, pl. 25, figs. 9, 13; Keen, 1971:145 (as a synonym of B. hertleini); Ber- nard, 1983:33 (as a synonym of B. hertleini) Type material and locality: UCMP 32668, holotype, left valve, now lost; length, 8.5 mm; height, 4.9 mm; thick- ness, 1.7 mm (Figure 3a, b). UCB Loc. A3582; Bahia Santa Inez, Baja California Sur (27.1°N, 112.0°W); ‘‘from 20-foot terrace level extending from Loc. A3581 to beach’”’; Pleistocene. (Loc. A3581; W. of Punta Santa Inez, about 0.5 miles from beach, at end of hill; Lower Pliocene). Description: Shell ovate-elongate, length/height about 1.6; beaks at approximately 60-70% of distance to an- terior end; anterior end rounded; posterior end subtrun- cate, tilted anteriorly approximately 20° from vertical. Central slope set off from posterior slope by only a rounded angle. Surface with strong, irregular commargin- al growth checks, without conspicuous pustules. Right valve with an elongate, vertical to slightly anteriorly di- rected, projecting anterior cardinal, and a small, horizon- tal process attached to it dorsally; left valve with an oblique, projecting, horizontally elongate anterior cardi- Page 136 The Veliger, Vol. 42, No. 2 E. V. Coan, 1999 nal, and a conspicuous gap under beaks for cardinal of right valve. External portion of ligament of moderate length, separated from internal portion by a ridge; nymph weak; internal portion wide, extending into a pit under beaks. Pallial line even curved, not deflected toward pos- teroventral corner. Left valve with an elongate escutch- eon; escutcheon weaker, most visible posteriorly in right valve. Length to 12.5 mm (SBMNH 143609; Bahia San Luis Gonzaga, Baja California [Norte]). Additional spec- imens are illustrated here (Figures 4, 5, 30, 31). Distribution: Northeastern end of Isla Cedros, Baja Cal- ifornia [Norte] (28.3°N) (LACM 71-152.30), in the Golfo de California from Los Frailes, Baja California Sur (23.4°N) (Skoglund Coll.), north to Bahia Cholla, Sonora (27.9°N) (Skoglund Coll.), south on the Sonoran coast to Bahia San Carlos (27.9°N) (LACM 78-30.10), Mexico. There are records from the intertidal zone to 100 m (mean, 26 m); the only bottom type noted is sand. There are only two live-collected specimens, SBMNH 144173, from 30 m, and SBMNH 143609, for which no habitat information is available. I have seen 29 Recent lots. Also present in the Pleistocene near Bahia Santa Inez, Baja California Sur (type locality). Discussion: It is unfortunate that the unique holotype of this species has been lost, but the original description and illustrations are sufficient to permit its recognition as a distinct species that is also represented in the Recent fau- na of northwest Mexico. It is most similar to Basterotia peninsularis, differing in the following respects: (1) the beaks are closer to the midline; (2) the dorsal and ventral margins are more par- allel; (3) the pallial line is more evenly curved, lacking the sharp bend toward the posteroventral margin present in B. peninsularis; and (4) the anterodorsal shell margin is not flared and pustulose, as it generally is in B. pen- insularis. The hinge also differs significantly: the external portion of the ligament is proportionately longer in B. californica, the nymph is less conspicuous, and the inter- nal portion extends over a broader, more triangular area. In the right valve, the base of the large cardinal tooth is broader in B. californica; in the left valve, the cardinal tooth is more horizontal and less projecting. Juveniles of B. californica tend to be proportionately thicker-shelled and those of the other species. Page 137 Basterotia obliqua, Coan, sp. nov. (Figures!657/5.32- 33) Type material and locality: LACM 2846, holotype, left valve; length, 9.0 mm; thickness, 2.2 mm (Figures 6, 32); LACM 2847, paratype, right valve; length, 10.1 mm; height, 6.2 mm (Figures 7, 33); LACM 2848, paratype, left valve, length, 8.4 mm. LACM 78-30, 3 mi. S. of Las Tetas de Cabra, Bahia San Carlos, Mexico (27.9°N, 111.1°W); 100 m on bottom of shells, cobbles and silt; Roy & Forest Poorman, April 1978. Description: Shell ovate-oblique, thin; length/height about 1.6; beaks approximately 75-80% of distance to anterior end; anterior end narrowed, rounded; posterior end broad, subtruncate, tilted anteriorly about 30° from vertical; dorsal margin oblique to ventral margin; central slope set off from posterior slope by an obscure angle only near beaks. Surface with irregular commarginal growth checks; pustules sparse, restricted to dorsal and anterior margins in some specimens. Right valve with a narrow, projecting, peglike cardinal; left valve with a nar- row, not very projecting cardinal and a gap for cardinal of right valve under beaks. External portion of ligament of moderate length for genus, not separated from internal portion; nymph heavy; internal portion of ligament small, restricted to medial surface of nymph. Pallial line deflect- ed toward posteroventral corner. Left valve with a con- spicuous, elongate escutcheon; escutcheon present but less evident in right valve. Length to 10.7 mm (Skoglund Coll.; Los Frailes, Baja California Sur). Distribution: Type lot, see above (27.9°N); Known from only four lots, from 16—100 m (mean, 57 m): Skoglund Coll.—Los Frailes, Baja California Sur (23.4°N); 50-66 m; left valve; LACM 34-20.13—Caleta Tagus, Isla Isa- bela, Islas Galapagos (0.3°S); 55 m; rock/coral bottom; small, broken left valve; LACM 34-61.16—Bahia Car- tago, Isla Isabela, Islas Galapagos (0.6°S); 15—18 m; sand; left valve. Discussion and comparisons: This species differs from B. peninsularis in being trapezoidal rather than oval, thin- ner shelled, in lacking a produced anterodorsal margin, having narrower beaks, and a sharper angle between the posterodorsal and central slopes. In shape, it is closer to B. panamica, but it is thinner, more trapezoidal, and it Explanation of Figures 3—7 Figures 3-5. Basterotia californica. Figure 3a, b. Holotype, now lost; length, 8.5 mm. Figure 4. Pair, SBMNH 143609; Bahia San Luis Gonzaga, Baja California [Norte]; about 6 m; length, 10.0 mm. Figure 5. Left valve, close- up of hinge; SBMNH 143609; shell length, 5.5 mm. Figures 6, 7. Basterotia obliqua, Coan, sp. nov. Figure 6, holotype, left valve; LACM 2846; length, 9.0 mm. Figure 7, paratype, right valve; LACM 2847; length, 10.1 mm. Page 138 The Veliger, Vol. 42, No. 2 has a narrower escutcheon and a thinner hinge. Basterotia obliqua differs from B. oblonga Smith, 1890 (pp. 303- 304, pl. 22, fig. 5), from St. Helena in having a more pronounced angle between the posterior and central slopes near the beaks and a heavier nymph. It differs from B. lutea (Dall, Bartsch & Rehder, 1938) from Hawaii in having a longer ligament. Etymology: The name is derived from the oblique shape of the shell of this species. Basterotia panamica Coan, sp. nov. (Figures 8—10, 34) ?Basterotia peninsularis (Jordan), auctt. non Jordan, 1936 (see Discussion under B. peninsularis). Durham, 1950: 95, 170; pl. 25, figs. 3, 8; Keen, 1958:106, 107, fig. 218 (in part); Keen, 1971:145, 146, fig. 342 (in part) Type material and locality: SBMNH 144168, holotype, complete pair; length, 7.3 mm; height, 4.5 mm; thickness, 3.5 mm (right valve slightly broken on posterodorsal mar- gin) (Figures 8, 34). SMBNH 144169, paratypes, four pairs containing dried animals, 8.1 mm, 7.7 mm, 7.4 mm, 7.4 mm in length. SBMNH 144170, paratypes, two pairs with dried animals, 8.1 mm, 6.9 mm in length. ‘“Tecuan,”’ Jalisco, Mexico (19.3°N, 104.9°W); in estuary mouth un- der rocks at low tide; Carol C. Skoglund, December 1974. Description: Shell ovate-trapezoidal, length/height about 1.6; beaks approximately 85% of distance to anterior end; anterior end rounded; posterior end subtruncate, tilted an- teriorly about 30° from vertical; central slope generally set off from posterior slope by an angle, varying among lots from rounded to sharp to carinate (angle moderate in type lot). Entire surface with dense fine to coarse granules and with strong, irregular commarginal growth checks. Right valve with an elongate, projecting, nearly vertical anterior cardinal and an obscure, short, thick horizontal process attached to it dorsally; left valve with a narrow, vertically elongate, projecting anterior cardinal and a con- spicuous gap medial to it for cardinal of right valve. Lig- ament of moderate length, its external portion on a nymph of moderate strength, slightly divided from its internal portion, which is narrow to medium in width and extends across anterior end of nymph into a pit under beaks. Pal- lial line deflected toward posteroventral corner. Conspic- uous escutcheon present in left valve, smaller or incon- spicuous in right valve. When valves closed, anterior and ventral margins gaping, the ventral margin gaping two- thirds of distance to posterior slope. Mantle with a short pedal gape. Posterior end with incurrent and excurrent openings, without siphons (openings very small in dried material). This species broods its young along its ventral mantle margin (Figure 10). Length to 11.0 mm (LACM 71—177.30; Punta San Pablo, Baja California Sur), 10.5 mm (SBMNH 144172; south end of Isla San Marcos, 10 Explanation of Figures 8-10 Figures 8—10. Basterotia panamica Coan, sp. nov. Figure 8, ho- lotype, pair; SBMNH 144168; length, 7.3 mm. Figure 9. Left valve, close-up of hinge; SBMNH 144171; Bahta San Luis Gon- zaga, Baja California [Norte]; about 6 m; length, 6.2 mm. Figure 10. Brood on ventral margin of mantle in a left valve; SBMNH 143659, Cabo San Lucas, Baja California Sur; 12 m; adult shell length, 10.5 mm; juveniles, approximately 0.2 mm. Baja California Sur). An additional specimen is illustrated here (Figure 9). Distribution: Punta San Pablo, Baja California Sur (27.2°N) (LACM 71-177.30, a single right valve with a E. V. Coan, 1999 broken hinge, the only specimen seen from the Pacific coast of Baja California), into the Golfo de California as far north as Bahia Cholla, Sonora (31.4°N) (SBMNH 14366; Skoglund Coll.), south along the coasts of Mexico and Central America to Salinas, Guayas Province, Ecua- dor (2.2°S) (CAS 106168), and in the Islas Galapagos at eight stations: Isla Genovesa (0.3°N) (CAS 106384); Isla Isabela (0.3°S) (CAS 106379); Isla Baltra (0.4°S) (LACM 34—46.1); Isla Santa Cruz (0.5°S) (ANSP 154907); Isla Santa Cruz (0.8°S) (CAS 106153); Isla San Cristébal (0.8°S) (LACM 34—43.20); Isla San Crist6bal (0.9°S) (MNHN); Isla Espafiola (1.4°S) (CAS 106153; LACM 34-283.7). This species occurs from the intertidal zone to 119 m (mean, 20 m). Live-collected material has been obtained from the intertidal zone to 11 m. Various bottom types are noted on labels, including mud, sand, and rubble; however, live collected material was obtained under rocks, suggesting a nestling habitat. I have seen 82 Recent lots. Perhaps in the Pleistocene of the southern Golfo de California (see Discussion under B. peninsularis). Discussion and comparisons: This new species differs from Basterotia peninsularis in having a more trapezoidal outline, with a shorter, narrower, more ventrally posi- tioned anterior end and a broader, more truncate posterior end. Its surface is lightly to coarsely pustulose, whereas that of B. peninsularis is scarcely pustulose, with pustules chiefly restricted to the antero- and posterodorsal mar- gins. The posterior slope of this new species is separated from the central slope by an angle, which may be sharp or even carinate in some material; there is no such angle in B. peninsularis. In the right valve, the small horizontal process dorsal to the cardinal tooth is much less conspic- uous; in the left valve, the large cardinal projects more dorsally. The external and internal portions of the liga- ment of the new species are not separated by a conspic- uous ridge medially, as they are in B. peninsularis. The poorly known Basterotia pustula Nowell-Usticke, 1971 (pp. 29-30, pl. 5, fig. 1618), described from St. Croix, Virgin Islands, differs in being thinner and more elongate. Material from the southern end of the distribution of this species is more consistently carinate between the cen- tral and posterior slopes and is also more variable. Even- tually, this southern material may be found to be taxo- nomically distinct. The brooding in this species along the mantle margin is similar to that described for Anisodonta alata (Powell, 1952), by Ponder (1968). Etymology: The name is derived from the Panamic prov- ince, in which this is the most common species of Bas- terotia. Page 139 Basterotia peninsularis (Jordan, 1936) (Figures 11-15, 35, 36) Anisodonta peninsulare Jordan, 1936:147, pl. 18, figs. 11, 12; (the following references all as Basterotia); Dur- ham, 1950:95 (in part), not pl. 25, figs. 3, 8; Hertlein & Strong, 1947:137; Keen, 1958:106 (in part), not 107, fig. 218; Keen, 1971:145 (in part), not 146, fig. 342; Bernard, 1983:33. Basterotia hertleini Durham, 1950:94—95, 170, pl. 25, figs. 4, 11; Emerson & Hertlein, 1964:355, 357, 358, 359, fig. 4g—j; Keen, 1971:145, 146, fig. 343; Hertlein & Grant, 1972:241—242, pl. 57, figs. 6, 11; Bernard, 1983: 33% Basterotia ecuadoriana Olsson, 1961:243, 509, pl. 36, fig. 8, 8a; Keen, 1971: 145 (as a synonym of B. hertleini); Bernard, 1983:33 (as a synonym of B. hertleini). Type material and localities: A. peninsulare—CASGTC 754.05 (originally 5583), holotype, left valve; length, 15.0 mm; height, 10.6 mm; thickness, 3.8 mm (Figure 11). CAS 754.06 (originally 5584), paratype, right valve; length, 11.2 mm (Figure 12). CAS Loc. 754, north of village, Bahia Magdalena, Baja California Sur (24.6°N, 112.2°W); Pleistocene. B. hertleini—_UCMP 32274, ho- lotype, left valve; length, 13.2 mm; height, 7.6 mm; thick- ness, 3.6 mm (Figure 13); UCMP 32328, paratype, left valve; length, 11.5 mm; UCMP 32372, paratype, right valve; length, 10.0 mm; CAS 8581, paratype, right valve; length, 9.9 mm; CAS 8581a, paratype, left valve; length, 11.5 mm. CAS 8581b, paratype, right valve; length, 8.5 mm. UCB Loc. A3670, Puerto Balandra, Isla Carmen, Baja California Sur (26.0°N, 111.2°W); Upper Pliocene; “from sands at left end of outcrop and below base of coral reef.”’ There were also said to be specimens from UCB Locs. A3519 and A3520; Bahia Marquer, Isla Car- men; Upper Pliocene. B. ecuadoriana—ANSP 218892, holotype, left valve; length, 12.3 mm; height, 7.7 mm; thickness, 2.3 mm (Figure 14). Manta, Manabi Province, Ecuador (0.1°S, 80.8°W). Paratype, left valve, length, 15.2 mm, not located. Punta Santa Elena, Guayas Prov- ince, Ecuador (2.2°S, 81.0°W). Description: Shell ovate to ovate-trapezoidal; length/ height about 1.7; beaks approximately 75% of distance to anterior end; anterior end rounded; posterior end subtrun- cate, tilted anteriorly about 30° from vertical; anterodorsal margin often somewhat flared; dorsal margin often den- ticulate; central slope set off from posterior slope by only a rounded angle. Surface with irregular commarginal growth checks; pustules, if present, sparse, restricted to small specimens and along dorsal margin of larger spec- imens. Right valve with a projecting, elongate, vertical anterior cardinal and with a very small anteroposteriorly elongate process attached to it dorsally; left valve with a narrow, vertical, projecting cardinal and a gap under beaks for cardinal of right valve. External portion of lig- ament of moderate length for genus, separated from in- ternal portion by a ridge; nymph of moderate strength; Page 140 The Veliger, Vol. 42, No. 2 15 E. V. Coan, 1999 internal portion of ligament narrow, often with a pit under beaks. Pallial line deflected toward posteroventral corner. Left valve generally with an inconspicuous escutcheon; escutcheon absent or much less evident in right valve. Length to 18.7 mm (MNHN; Punta Santa Elena, Guayas Province, Ecuador). Additional specimens are figured here (Figures 15, 35, 36). Distribution: From Isla Espiritu Santo, Baja California Sur (24.5°N) (CAS 106155), north in the Golfo de Cali- fornia to Bahia Cholla, Sonora (31.4°N) (Skoglund Coll.), Mexico, south to Salinas, Guayas Province, Ecuador (2.2°S) (MNHN; ?CAS 110366), and in the Islas Gala- pagos on Isla Isabela (0.2°S) (LACM 34-30.11) and Isla Baltra (0.4°S) (LACM 34—46.28). Specimens have been obtained from the intertidal zone to 46 m (mean, 16 m); all material has been obtained dead. I have examined 40 Recent lots. The bottom type most often noted is sand, but some labels mention mud or rocks. Also know from the Pliocene of southern California (Hertlein & Grant, 1972) and of the islands in the southern Golfo de Cali- fornia (Durham, 1950; Emerson & Hertlein, 1964) and the Pleistocene of Bahia Magdalena, Baja California Sur (Jordan, 1936). Discussion: The holotype of Anisodonta peninsulare is a very large, oval specimen of the same species that was later named Basterotia hertleini. The paratype of A. pen- insulare is a more typical specimen and is similar to the holotype of Basterotia hertleini. The material referred to and illustrated as B. peninsularis by Durham (1950) con- sists of specimens that are strongly carinate and heavily pustulose—UCMP 32271, 32272, 32273; CAS 66787.01; SBMNH 143667—all from UCB loc. A3548; Pleisto- cene; Isla Coronados, Baja California Sur (26.1°N). These specimens are morphologically intermediate between B. quadrata (see below) and B. panamica. They are thicker- shelled and more regular in shape than B. quadrata, and they have a longer ligament and nymph. On the other hand, they are larger than any known specimens of B. panamica; the largest is 13.9 mm in length, and the spec- imen figured by Durham (1950) is 12 mm in length. Dur- ham’s figures were then reproduced in Keen (1958, 1971), forming a mistaken concept of B. peninsulare among students of the Recent Panamic fauna. These Pleistocene specimens may be assignable to B. panamica, Page 141 or they may represent an evolutionary stage on the way to it. Basterotia ecuadoriana has been generally been syn- onymized with B. hertleini. Olsson differentiated his new species from B. hertleini, saying that B. ecuadoriana dif- fered in being more elongate and less convex. In actual- ity, the holotype of B. hertleini has a length/width ratio of 1.7, whereas the holotype of B. ecuadoriana has a ratio of 1.6. On the other hand, the holotype of B. ecuadoriana is indeed flatter, with a thickness/height ratio of 0.3, whereas the holotype of B. hertleini has a ratio of 0.5. In any event, these differences are within the range of var- iability of B. peninsularis. For comparisons with B. californica, see under that species. Basterotia peninsularis is most similar to the western Atlantic B. elliptica (Récluz, 1850:168—-169) (synonym: Corbula newtoniana C. B. Adams, 1852:240), which differs in being proportionately shorter, more Ovate-trapezoidal, in having a shorter, less produced an- terior end, and in lacking a denticulate dorsal margin. It also has a still shorter external ligament. The western Atlantic Basterotia corbuloidea (Dall, 1899:885-886, 896, pl. 88, fig. 2, as Anisodonta)? is smaller and thinner, and its beaks are closer to the anterior end. Basterotia quadrata (Hanley, 1843) (Figures 16-18, 37, 38) Corbula quadrata Hanley, 1843 (early 1843):7, pl. 12, fig. 36; Hinds, 1843 (Nov.):57; Reeve, 1844:pl. 5, fig. 40; Récluz, 1850:168 (Eucharis); C. B. Adams, 1852:239— 240 (Corbula); Hanley, 1856:345 (Corbula); Fischer, 1860:23—26 (Eucharis); Fischer, 1886:199 (Eucharis); Dall, 1899:875, 877 [Anisodonta (Basterotia)]; Lamy, 1925:505; Olsson, 1961:242 [Basterotia (Basterotia)]; Bernard, 1983:33, 68 (as ‘“‘extralimital’’ to the eastern Pacific). Poromya (?) granatina Dall, 1881:109; Dall, 1886:316, pl. 1, fig. 2, 2a, 2b (as Basterotia quadrata var. granatina), Lamy, 1925;506 (as a variety of B. quadrata). Type material and localities: C. quadrata—Original specimens missing (K. Way, e-mail, 18 September 1996). 3 Not a homonym and not to be confused with the type species of the genus, Basterotia corbuloides. Explanation of Figures 11-15 Figures 11-15. Basterotia peninsulare. Figure 11. Holotype of Anisodonta peninsulare, left valve; CASGTC 754.05; length, 15.0 mm. Figure 12. Paratype of A. peninsulare, right valve; CASGTC 754.06; length, 11.2 mm. Figure 13. Holotype of Basterotia hertleini, left valve; UCMP 32274; length, 13.2 mm. Figure 14. Holotype of Basterotia ecuadoriana, left valve; ANSP 218892; length, 12.3 mm. Figure 15. Left valve, close-up of hinge; SBMNH 143606; Bahia San Luis Gonzaga, Baja California [Norte]; about 6 m; shell length, 8.8 mm. The Veliger, Vol. 42, No. 2 Page 142 E. V. Coan, 1999 Hanley’s figure measures 9 mm in length and 8 mm in height (Figure 16). Hinds gives the length of his specimen as 6 lines (12.7 mm) and its height as 5 lines (10.6 mm) (Figure 17). Original locality unknown. P. granatina— MCZ 8133, holotype, right valve; length, 9.9 mm; height, 6.8 mm; thickness, 3.2 mm (Figure 18). Yucatan Strait; “off Cuba” (a label); 640 fms. (1170 m, probably much too deep to reflect its true habitat). Description: Shell ovate-trigonal, length/height about 1.1; beaks approximately 70% of distance from anterior end, very prominent, inflated, and strongly prosogyrate in some material; anterior end rounded; posterior end round- ed to subtruncate, inclined anteriorly about 30° from ver- tical; central slope divided from posterior slope by a 90° angle near beaks; angle with a carina, most prominent dorsally, broadening toward ventral margin. External sur- face with coarse pustules and heavy, irregular commar- ginal growth checks. Posterodorsal margin pustulose. Right valve with a prominent, ventrally projecting car- dinal; left valve with a prominent, projecting cardinal. External portion of ligament very short, on a short, sturdy nymph; internal portion small, on medial surface of nymph, extending to a pit under beaks. Narrow to wide escutcheon present in left valve; escutcheon not present in right valve. Pallial line not greatly deflected toward posteroventral margin. Length to 14.1 mm (MNHN; Sa- linas, Guayas Province, Ecuador). Additional eastern Pa- cific specimens are illustrated here (Figures 19, 20, 37, 38). Distribution: In the eastern Pacific, from near Isla Par- tida, Baja California Sur (24.5°N) (LACM 60-6.24), throughout the Golfo de California to its head at Bahia Cholla, Sonora (31.4°N) (Skoglund Coll.), south to Sali- nas, Guayas Province, Ecuador (2.2°S) (MNHN), and in the Islas Galapagos on Isla Isabela (0.3°S) (CAS 110365) and Isla Santa Cruz (0.5°S) (ANSP 400163). Stations are from 6 to 119 m (mean, 28 m). No bottom types were noted. I have seen 13 Recent eastern Pacific lots. Also present on the Pleistocene 3rd Terrace at Punta Santa Elena, Guayas Province, Ecuador (Hoffstetter, 1948:75; MNHN). In the western Atlantic, from Cape Lookout, North Carolina (34.3°N) (USNM 94210), on both coasts of Flor- ida, in the Bahamas (C. Redfern, in correspondence, 3 Page 143 April 1997), south to Haiti (USNM 440430 and other lots), Guadalupe (MNHN), and Colombia (UCMP S-10). Discussion: Although overlooked by previous workers, this species was first proposed in Hanley (1843), which was published early in the year, many months before the name was made available by Hinds (1843) (see Literature Cited for collation of Hanley). While Hanley’s figure, an external view, is not unequivocal, interpreting Corbula quadrata Hanley as a nomen dubium would only create a senior homonym of Corbula quadrata Hinds. Specimens of this species from the eastern Pacific are indistinguishable from those from the Caribbean. This species differs from the very similar Basterotia (Baster- otia) ambona Vokes, 1981:160—-161, 163, figs. 1-3, de- scribed from the late Lower Miocene Chipola Formation of Florida, in having a more expanded, denticulate pos- terodorsal margin; in B. ambona, it is more evenly curved and is not denticulate. In the Recent taxon, the poster- oventral corner is more pointed, and the hinge teeth seem somewhat broader (material of B. ambona examined: USNM 298649, 298650, 298651; UF 77591). Indeed, B. quadrata is very close to the Middle Miocene type spe- cies of the genus, B. corbulides, differing in having a denticulate posterodorsal margin, a more curved ventral margin, and on average a slightly more produced anterior end, and in lacking a depressed area just anterior to the carina, which is in part responsible for the straighter ven- tral margin, and perhaps in attaining a larger size (ma- terial of B. corbuloides examined: NMW 1855/XLV/282, 2 paratypes). Dall (1886:316) cited this species from the Pacific coast, but there are no eastern Pacific specimens in the USNM. His record was repeated with question by Olsson (1961:242) and then dismissed by Bernard (1983:33, 68). However, as can be seen, it does indeed occur in the eastern Pacific in addition to the western Atlantic. Dall (1886:316) also mentioned seeing possible material of this species from Korea. I suspect, however, he may have seen specimens of the Indo Pacific Basterotia angulata (H. Adams, 1871:789, 795, pl. 48, fig. 3) (Eucharis), which differs from B. quadrata in being more elongate and having a longer external ligament, and it has no ten- dency to form greatly expanded, prosogyrate beaks. Bas- terotia angulata (H. Adams, 1871) is a senior homonym Explanation of Figures 16—22 Figures 16-20. Basterotia quadrata. Figure 16. Hanley’s (1843) figure; length, 9 mm. Figure 17, Reeve’s (1844) figure, probably of Hinds’ specimen; length, 12.7 mm. Figure 18. Holotype of Poromya granatina, right valve; MCZ 8133; length, 9.9 mm. Figure 19. Right valve; SBMNH 143655; Puerto San Carlos, Sonora, Mexico; 27 m; length, 12.5 mm. Figure 20. Right valve; LACM 72-54.46; Bahia Herradura, Puntarenas Province, Costa Rica; 37 m; length, 9.6 mm. Figures 21—22. Basterotina rectangularis Coan, gen. & sp. nov. Figure 21. Holotype, right valve; SBMNH 144174; length 11.0 mm. Figure 22. Paratype, left valve; SBMNH 144175; length 7.8 mm. Page 144 The Veliger, Vol. 42, No. 2 E. V. Coan, 1999 of B. angulata (Dall, Bartsch & Rehder, 1938:125, pl. 34, figs. 5, 6) (Anisodonta); however, because it may also be a senior synonym of this Hawaiian taxon, it may not need to be renamed. Basterotina, Coan, gen. nov. Type species: B. rectangularis Coan sp. nov. Description: Shell subquadrate; beaks at 70% from pos- terior end; posterior end truncate; central and posterior slopes separated by an angle, often carinate; surface pus- tulose. Hinge teeth not projecting, as in Basterotia. Right valve with a moderate to obscure cardinal, attached dor- sally to an obscure, anteroposteriorly oriented process; left valve with an anterior cardinal formed by hinge mar- gin, fitting anterior to cardinal of right valve. External portion of ligament in a groove, strengthened by a weak nymph most evident posteriorly; internal ligament in an elongate, triangular area. Escutcheon most apparent in left valve. Posterior margin of anterior adductor muscle scar with a radial strengthening rib. Discussion and comparisons: This genus differs from Basterotia in having low, non-projecting hinge teeth, but is similar to many species of that genus in having the posterior and central slope separated by an angle and in being pustulose. It differs from the type species of Ani- sodonta Deshayes, 1858:542-543, A. complanata De- shayes, 1858:543, pl. 22, figs. 1-4, from the Paleocene of France, in being carinate and pustulose, and in having a greater amount of internal ligament, but is similar in having an internal thickening posterior to the anterior ad- ductor (material of A. complanata examined: Sencken- berg Museum). In addition to the following species, Anisodonta amer- icana Dall, 1900, from the Miocene of Florida, also be- longs in this genus (see additional comparison below). Etymology: The name of this genus is derived from Bas- terotia, suggesting its probable relationship to this genus, with the addition of the diminutive, -ina. Basterotina rectangularis Coan, sp. nov. (Figures 21—23, 39, 40) Type material and locality: SBMNH 144174, holotype, right valve; length, 11.0 mm; height, 5.1 mm; thickness, Page 145 2.0 mm (Figures 21, 40). SBMNH 144175, paratype, left valve; length, 7.8 mm (Figures 22, 39). Los Frailes, Baja California, Sur (23.4°S, 109.4°W); 60 m; Pete & Iva Barker, February 1973; ex Skoglund Collection. Description: Shell elongate-rectangular, length/height 2.5; beaks approximately 70% of distance to anterior end; anterior end rounded; posterior end broad, truncate, tilted anteriorly about 40° from vertical; posterodorsal margin slightly concave; central slope set off from posterior slope by an angle, which may be carinate in some material. Surface coarsely pustulose, most conspicuously on ends, and with irregular commarginal growth checks. Right valve with an obscure, slightly anteriorly directed, non- projecting anterior cardinal, with a small, anteroposteri- orly elongate process dorsal to it; left valve with a small, slightly projecting anterior cardinal formed by hinge mar- gin. External portion of ligament elongate, on a conspic- uous nymph; internal portion narrow, separated from ex- ternal portion by an obscure ridge. Pallial line deflected toward posteroventral corner. Left valve with a well-de- fined escutcheon; right valve without one. Anterior ad- ductor muscle scars with a posterior strengthening rib. Length to 11.0 mm (holotype). Two additional specimens are illustrated here (Figure 23). Distribution: East side of Isla Cedros, Baja California [Norte] (28.2°N) (LACM 71—94.27), into the Golfo de California as far north as near Bahia Puertecitos, Baja California [Norte] (30.4°N) (LACM 72-—215.19), and Ba- hia San Carlos, Sonora (27.9°N) (LACM 78-30.9), south to the north side of Isla Salango, Manabi Province, Ec- uador (1.6°S) (LACM 80-65.9), from 9 to 100 m (mean, 47 m). The only bottom type noted on labels was sand. This species is thus far known from only eight lots, all obtained dead. Other referred material: CAS 106154—‘‘Gulf of Cal- ifornia,’’ Mexico; Skoglund Coll. - Punta San Antonio, Sonora, Mexico; SBMNH 143656 - Bahia Las Palmas, Baja California Sur, Mexico. Discussion and comparisons: This species is most sim- ilar to Basterotina americana (Dall, 1900:1133, pl. 36, fig. 7; USNM 107808, holotype, right valve) (Anisodon- ta), described from the Caloosahatchee Formation of Monroe County, Florida (late Pliocene or early Pleisto- cene) and subsequently also figured from beds of similar Explanation of Figures 23—26 Figure 23. Basterotina rectangularis Coan. gen. & sp. nov. (a) Left and (b) right valves, close-up up of hinges; LACM 71-94.27; E. of Isla Cedros, Baja California [Norte]; about 24 m; right valve length, 5.8 mm; left valve length, 4.4 mm. Figures 24-26. Ensitellops hertleini. Figure 24. Holotype of E. hertleini, left valve; UCMP 11243; length, 8.5 mm. Figure 25. Holotype of E. pacifica, left valve; ANSP 218893; length, 5.4 mm. Figure 26, three left valves, showing variability; LACM 40—50.31; Bahia Topoca, Sonora, Mexico; 7.4, 7.4, 7.0 mm. Page 146 age at Shell Creek, Florida (Dall, 1903:pl. 57, fig. 23). The Recent species differs in being more elongate, less produced and angled posterodorsally, and more oblique posteriorly; the posterior end of B. americana is tilted only about 20—30° from vertical. The ventral margin of the anterior hinge plate in the right valve is not as re- cessed in the Recent species as it is in B. americana, in which this margin forms a slot, and the nymph is less conspicuous, as are the hinge teeth. The thickening pos- terior to the anterior adductor muscle scar is heavier in B. americana. Etymology: The name is derived from the rectangular shape of the shell of this species. Ensitellops Olsson & Harbison, 1953 Enstitellops Olsson & Harbison, 1953:93. Type species: Sportella protexta Conrad, 1841:347. Pliocene; North Carolina. (Concerning this species: Campbell, 1993:33, pl. 10, fig. 99.) Description: Shell very elongate, flattened to somewhat inflated, thin, fragile; beaks closer to the posterior end. Surface with irregular commarginal growth checks and scattered pustules. Right valve with a single vertical to anteriorly directed cardinal; left valve with anterior and posterior cardinals. Ligament in a short, posteriorly di- rected resilifer just within shell margin. Pallial line entire. Ensitellops hertleini Emerson & Puffer, 1957 (Figures 24—27, 41) Ensitellops hertleini Emerson & Puffer, 1957:21—22, fig. 2; Keen, 1958:106, 107, fig. 221; Olsson, 1961:242, 509, g. 22 pl. 36, fig. 9; Keen, 1971:145, 146, fig. 344; Bernard, 1983:33. Ensitellops pacifica Olsson, 1961:241—242, 553, pl. 80, fig. 9, 9a; Keen, 1971:145, 146, fig. 345; Bernard, 1983:33. Type material and localities: E. hertleini—UCMP 11243, holotype, left valve; length, 8.5 mm (not 9.5 mm, as originally stated); height, 3.2 mm; thickness, 1.1 mm (Figure 24); UCMP 11325, paratype, right valve, length, 6.2 mm; UCMP 11326, paratype, right valve, length, 4.0 mm. UC Loc. A-3603; Guaymas Harbor, Sonora, Mexico (27.9°N); 4 m; E. pacifica—ANSP 218893, holotype, left valve (not right, as stated in text; posterodorsal margin now chipped); length, 5.4 mm; height, 2.8 mm; thickness, 0.6 mm (Figure 25); ANSP 218894, paratype, left valve; length, 4.7 mm; UMML 30.9906, two right valves and four left valves, from type locality, labeled “‘para”’ [types] by Olsson. El Lagartillo, Las Tablas, Panama (7.8°N); (in fig. caption in error as “‘Ecuador’’). Description: Shell elongate, cylindrical, length/height about 2.5; beaks approximately 80% of distance to an- terior end; overall outline somewhat variable, irregular; typically with a broad, obscure furrow from beaks to cen- tral slope, which narrows the anterior end; anterior end The Veliger, Vol. 42, No. 2 rounded; posterior end slightly pointed to rounded; degree of valve inflation variable, with some specimens fairly flattened, others more inflated, the two valves varying in degree of inflation with no discernable bias as to which valve is more so. Surface with irregular commarginal growth checks and scattered pustules of variable size. Right valve with an anteriorly curved, slightly projecting cardinal; left valve with a vertical cardinal beneath beaks and an anterior cardinal formed by valve margin. Liga- ment in an elongate, narrow resilifer within valve margin. Left valve with an escutcheon; escutcheon not apparent in right valve. Pallial line broad, evenly curved. Posterior margin of anterior adductor with a raised ridge. Length to 9.4 mm (LACM 40—43.1; Bahia San Felipe, Baja Cal- ifornia [Notre]). Additional specimens are illustrated here (Figure 26, 27, 41). Distribution: From Bahia Cholla, at the head of the Gol- fo de California (31.4°N) (SMBNH 143664; Skoglund Coll.), south to La Paz, Baja California Sur (24.2°S) (USNM 554978), and Mazatlan, Sinaloa, Mexico (23.2°N) (USNM 565846, 556376); El Lagartillo, Las Ta- blas, Panama (7.8°N) (type loc. of E. pacifica); Santa Elena, Guayas Province, Ecuador (2.2°S) (Olsson, 1961; specimens not located). Other than beach drift, specimens have been obtained from 2 to 35 m (mean, 12 m). Mud and sand bottoms are noted on some labels. I have seen 32 Recent lots. Discussion: Other than size, there seems to be no basis for separating two eastern Pacific species. As yet, this species is known from only two lots obtained south of Mazatlan, Sinaloa, Mexico. This species differs significantly from the Pliocene to Recent western Atlantic type species of the genus, E. pro- texta (Conrad, 1841), which is flatter, with a more even outline; E. hertleini is more inflated and has a narrower, more produced anterior end. The hinge margin is narrow- er in E. protexta. In the right valve, E. hertleini has a more anterodorsally elongate cardinal, whereas in E. pro- texta it is more vertical. In the left valve, the posterior cardinal of E. hertleini is heavier and more ventrally di- rected, and the anterior cardinal is on the hinge margin and is anteroventrally directed. In the left valve of E. protexta, the posterior cardinal is slightly posteriorly di- rected, and the anterior cardinal is more ventrally direct- ed. Ensitellops protexta has only a slight escutcheon in the left valve, whereas it is more strongly defined in E. hertleini. Ensitellops hertleini has been confused in collections with specimens of Sphenia. Sphenia is more inflated, and it has only a small cardinal tooth in the right valve; its right has a resilifer under the beaks, and the left valve has a projecting resilifer. Ensitellops is still more likely to be confused with small, elongate specimens of Hiatel- la. Small Hiatella have a thicker shell, with conspicuous commarginal folds, a more pointed anterior end, and they E. V. Coan, 1999 often have two external radial rows of spines. The pallial line, if visible, may be seen to consist of a discontinuous series of irregular scars in Hiatella, whereas it is contin- uous in Ensitellops. The hinge of Hiatella is heavier at an equivalent size, the protoconch is larger, and there is a heavy nymph for the external ligament. Fabella Conrad, 1863 Fabella Conrad, 1863:574, 586. Type species (monotypy): Amphidesma constricta Conrad, 1841:347, pl. 2, fig. 15. Pliocene, North Carolina. (Concerning this species: Campbell, 1993:32, pl. 10, fig. 94.). Description: Shell ovate, longer anteriorly. External sur- face with only commarginal growth checks. Right valve with anterior and central cardinal teeth; left valve with a large anterior cardinal and a small to minute central car- dinal. External portion of ligament of moderate length; nymph stout; internal portion of ligament on medial sur- face of nymph. Pallial line entire. The following Panamic species was initially placed in Sportella Deshayes, 1858:593—595 (type species [original designation]: Psammotea dubia Deshayes, 1824 (pp. 76, 6, pl. 10, figs. 13, 14); Middle Eocene, France). Sportella dubia differs from taxa here placed in Fabella in being equilateral, and it has a longer external ligament and a larger, triangular resilifer. Fabella stearnsii (Dall, 1899) (Figures 28, 29, 42) Sportella stearnsii Dall, 1899:879, 885, 896, pl. 87, figs. 9, 12; Hertlein & Strong, 1947:137—138; Keen, 1958:108, 109, fig. 237; Keen, 1971:143, 145, fig. 341; Bernard, 1983:32 (as Neaeromya); Rosewater, 1984:84 (as Pseu- dopythina). Sportella duhemi Jordan, 1936:146—147, pl. 17, figs. 1, 2. Type material and localities: S. stearnsii—USNM 73701, holotype, pair; length, 13.6 mm; height, 10.0 mm; thick- ness, 5.1 mm (Figures 28, 42). Golfo de California; no other information available; S. duhemi—CASGTC 754.04 (originally 5578), holotype, left valve (not right as stated by Jordan); length, 7.5 mm; height, 4.5 mm; thickness, 1.7 mm (Figure 29). CAS Loc. 754, north of village, Ba- hia Magdalena, Baja California Sur (24.6°N, 112.2°W); Pleistocene. Description: Shell ovate; length/height about 0.7; ante- rior end much longer, broadly rounded; beaks at approx- imately 40% of distance from posterior end; posterior end subtruncate. External surface with irregular commarginal growth striae. Right valve with a moderate-sized anterior cardinal and a larger central cardinal. Left valve with a large, oblique anterior cardinal, which fits between car- dinals of right valve, and a very small central cardinal. Escutcheon absent. External portion of ligament of mod- erate length, on a stout nymph; internal portion of liga- Page 147 27a 27b Explanation of Figures 27-29 Figure 27. Ensitellops hertleini. (a) Left and (b) right valves, close-up of hinges; SBMNH 13083; Topolobampo, Sinaloa, Mexico; left valve length, 6.9 mm; right valve length, 7.2 mm. Figures 28, 29. Fabella stearnsii. Figure 28. Holotype of Spor- tella stearnsii, pair, USNM 73701; length, 13.6 mm. Figure 29. Holotype of Sportella duhemi, left valve; CASGTC 754.04; length, 7.5 mm. The Veliger, Vol. 42, No. 2 Page 148 E. V. Coan, 1999 ment on medial surface of nymph. Pallial line evenly curved. Length to 15.6 mm (CAS 106163; Isla Santa Cruz, Islas Galapagos). Distribution: Puertecitos, Baja California [Norte] (30.3°N) (SBMNH 144176), south to two stations in the Islas Galapagos: Isla Santa Cruz (0.8°S) (CAS 106163) and Isla Espanola (1.4°S) (CAS 106387). Records are from 4 to 32 m (mean, 13 m). The only bottom types noted are rock and sand. All the available material was collected dead, except for the holotype, for which no ex- act locality or habitat was recorded. Known from only 15 Recent lots. Also present in the Pleistocene at Bahia Ma- gellan, Baja California Sur (type locality of S. duhemi). Discussion: Comparing his specimen to Dall’s figures of S. stearnsii, Jordan (1936) said that S$. duhemi had a more rectangular shape and more anterior beaks. However, his specimen falls well within the variation of small speci- mens in Recent lots that are now available. For example, it is closely similar to the specimen in CAS 106386 from Bahia Santa Inez. Baja California Sur, Mexico. This species differs from the type species of the genus, Fabella constricta, in having a shorter posterior end, more oval outline, and a smaller cardinal in the left valve. The external portion of the ligament in E. stearnsii is larger, as is the nymph. Fabella stearnsii is most similar to the equally rare Recent western Atlantic F. pilsbryi (Dall, 1899:884—-885, 897, pl. 88, fig. 9), which has a narrower anterior end. EXCLUDED TAXA Sportella californica Dall, 1899 (pp. 885, 897, pl. 88, fig. 5), described from Monterey, California, proves to be an Orobitella (Galeommatoidea: Lasaeidae) (holotype: USNM 159293) (Coan & Scott, 1997:12, 25). Anisodonta pellucida Dall, 1916a (p. 30, nomen nu- dum), 1916b (p. 411), also described from Monterey, Cal- ifornia, is based on a juvenile mactrid (holotype: USNM Page 149 208475), probably Simomactra falcata (Gould, 1850: 216). ACKNOWLEDGMENTS I appreciated the assistance of curators and others, who have allowed me to visit collections, to borrow speci- mens, and to have copies of scarce literature. These in- clude Kenneth J. Boss and Daniel L. Graf, Museum of Comparative Zoology, Harvard University; Philippe Bouchet, Rudo von Cosel, and Philippe Maestrati, Mu- séum d’ Histoire Naturelle, Paris; Raye Germon, Alan R. Kabat, and Thomas R. Waller, National Museum of Nat- ural History, Washington, D.C.; Amanda Diaz, Donald R. Moore, and Nancy Voss, University of Miami; William K. Emerson and Paula Mikkelsen, American Museum of Natural History; Ned S. Gilmore and Gary Rosenberg, Academy of Natural Sciences, Philadelphia; Lindsey T. Groves and James H. McLean, Los Angeles County Mu- seum of Natural History; Ronald Janssen, Natur-Museum Senckenberg, Germany; Elizabeth Kools and Robert van Syoc, California Academy of Sciences; David R. Lind- berg and Karen L. Wetmore, Museum of Paleontology, University of California, Berkeley; Roger Portell, Florida Museum of Natural History; Paul H. Scott, Santa Barbara Museum of Natural History; Ortwin Schultz, Naturhis- torische Museum, Vienna, Austria; Emily H. Vokes, Tu- lane University; and Kathie Way, The Natural History Museum, London. I also acknowledge material and in- formation provided by Lyle Campbell, University of South Carolina-Spartanburg; Colin Redfern of Boca Ra- ton, Florida; Carol C. Skoglund, Phoeniz, Arizona; and Philip Tubbs, The Natural History Museum London. I had photographic assistance from California Academy of Sci- ences SEM technician Darrell Ubick. Sharon Williams provided assistance in preparing the plates. Lyle Camp- bell and Carol C. Skoglund and one anonymous reviewer provided useful comments on the manuscript. Explanation of Figures 30—42 Figures 30, 31. Basterotia californica. Figure 30. Left valve; LACM 71—178.50; Punta San Pablo, Baja California Sur; about 26 m; length, 8.9 mm. Figure 31. Right valve; SBMNH 143609; Bahia San Luis Gonzaga, Baja California [Norte]; about 6 m; length, 12.5 mm. Figures 32, 33. Basterotia obliqua Coan, sp. nov. Figure 32, holotype, left valve; LACM 2846; length, 9.0 mm. Figure 33, paratype, right valve (pallial sinus not visible); LACM 2847; length, 10.1 mm. Figure 34. Basterotia panamica, Coan, sp. nov. holotype, pair; SRMNH 144168; length, 7.3 mm. Figures 35, 36. Basterotia peninsularis. Figure 35. Left valve; MNHN; Punta Santa Elena, Guayas Province, Ecuador; length, 18.6 mm. Figure 36. Right valve; CAS 106385; Corinto, Chinandega Province, Nicaragua; length, 13.0 mm. Figures 37, 38. Basterotia quadrata. Figure 37. Left valve; LACM 69-24.12; Bahia San Luis Gonzaga, Baja California [Norte], Mexico; 9 m; length 13.0 mm. Figure 38. Right valve; SBMNH 143655; Puerto San Carlos, Sonora, Mexico; 27 m; length, 12.6 mm. Figures 39, 40. Basterotina rectangularis Coan, gen. & sp. nov. Figure 39. Paratype, left valve; SBMNH 144175; length 7.8 mm. Figure 40. Holotype, right valve; SBMNH 144174; length 11.0 mm. Figure 41. Ensitellops hertleini, right and left valves; LACM 40—43.1; Bahia San Felipe, Baja California [Norte], Mexico; 5 m; right valve length, 5.7 mm; left valve length, 9.4 mm. Figure 42. Sportella stearnsii, holotype; USNM 73701; length, 13.6 mm. Page 150 LITERATURE CITED ADAMS, C. B. 1852. Description of new species of Corbula from Jamaica. Contributions to Conchology 1(12):233-241. ADAMS, H. 1871. Description of twenty-six new species of shells collected by Report M’ Andrew, Esq., in the Red Sea. Pro- ceedings of the Zoological Society of London, for 1870(3): 788-793, pl. 48. BERNARD, E R. 1983b. Catalogue of the Living Bivalvia of the Eastern Pacific Ocean: Bering Strait to Cape Horn. Canadian Special Publication of Fisheries and Aquatic Sciences 61: viii + 102 pp. CAMPBELL, L. D. 1993. Pliocene Molluscs from the Yorktown and Chowan River Formations of Virginia. Publication of the Department of Mines, Minerals, and Energy, Division of Mineral Resources, Commonwealth of Virginia 127:vii + 259 pp., 43 pls. Carter, J. G. & R. A. Lutz. 1990. Bivalvia (Mollusca). Pp. 5— 28, pls. 1-121 in J. G. Carter (ed.), Skeletal Biominerali- zation: Patterns, Processes and Evolutionary Trends, Vol. 2. Atlas and Index. Van Nostrand Reinhold: New York. CHAVAN, A. 1969. Superfamily Cyamiacea. Pp. 537-543 in L. R. Cox et al. (eds.), Part N (Bivalvia), Mollusca 6, vols. 1 & 2:xxxvil + 952 pp in R. C. Moore (ed.), Treatise on In- vertebrate Paleontology. Geological Society of America & University of Kansas: Lawrence. Coan, E. V. 1997. The eastern Pacific Sportellidae [Bivalvia: Cyamioidea]. The Festivus 29(11):107—112. Coan, E. V. & P. H. Scotr. 1997. Checklist of the Marine bi- valves of the Northeastern Pacific Ocean. Santa Barbara Mu- seum of Natural History. 28 pp. ConrabD, T. A. 1841, 1843. See Hodge & Conrad (1841, 1843). ConraD, T. A. 1863. Catalogue of the Miocene shells of the Atlantic slope. Proceedings of the Academy of Natural Sci- ences of Philadelphia for 1862[14](10—12):559-582. CossMANN, A. E. M. & A. Peyrot. 1909. Conchologie Néogé- nique de |’ Aquitaine 1(1). Actes, Société Linnéenne de Bor- deaux 63(2):73—-144, pls. 1-4, 3 maps; (3):145—232, pls. 5— 7; (4):233-293 (repr.: 1-220, pls. 1-7). DALL, W. H. 1881. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico, and in the Caribbean Sea, 1877-79, by the United States Coast Guard Steamer “‘Blake,”’ ... XV. Preliminary report on the Mollusca. Bulletin of the Museum of Comparative Zoology, Harvard University 9(2):33-144. DALL, W. H. 1886. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877-78) and in the Caribbean Sea (1879-80), by the U. S. Coast Survey Steamer “‘Blake,”’ .. . XXIX. Report on the Mollusca—Part I. Brachiopoda and Pelecypoda. Bulletin of the Museum of Comparative Zoology, Harvard University 12(6):171-318, pls. 1-9. DALL, W. H. 1899. Synopsis of the Recent and Tertiary Lepton- acea of North America and the West Indies. Proceedings of the United States National Museum 21(1177):873-897, pls. 87, 88. DaLL, W. H. 1900. Contributions to the Tertiary fauna of Florida, with especial reference to the silex beds of Tampa and the Pliocene beds of the Caloosahatchie River, including in many cases a complete revision of the generic groups treated of and their American Tertiary species. Part V. Teleodes- macea: Solen to Diplodonta. Transactions of the Wagner Free Institute of Science of Philadelphia 3(5):949—1218, pls. 36-47. DALL, W. H. 1903. Contributions to the Tertiary fauna of Florida, The Veliger, Vol. 42, No. 2 with especial reference to the silex beds of Tampa and the Pliocene beds of the Caloosahatchie River, including in many cases a complete revision of the generic groups treated of and their American species. Part VI. Concluding the work. Transactions of the Wagner Free Institute of Science of Philadelphia 3(6):xiv + 1219-1654, pls. 48-60. Da.LL, W. H. 1916a. Checklist of the Recent Bivalve Mollusks (Pelecypoda) of the Northwest Coast of America from the Polar Sea to San Diego, California. Southwest Museum: Los Angeles. 44 pp., | port. DALL, W. H. 1916b. Diagnoses of new species of marine bivalve mollusks from the northwest coast of America in the collec- tion of the United States National Museum. Proceedings of the United States National Museum 52(2183):393—417. DALL, W. H., P. BARTSCH & H. A. REHDER. 1938. A Manual of the Recent and Fossil Marine Pelecypod Mollusks of the Hawaiian Islands. Bulletin of the Bernice P. Bishop Museum 153:iv + 3—233 pp., 58 pls. (repr.: Kraus: New York, 1971). DESHAYES, G. P. 1824-1837. Description des Coquilles Fossiles des Environs de Paris. L Auteur; Jeune; Freres; Treuttel & Wurz; Levrault: Paris. Vol. 1:392 pp., ?? pls.; Vol. 2:814 pp., ?? pls. (Vol. 1:1-80, 1824; 81-170, 1825; 171-238; 1829; 239-322, 1830; 323-392; 1832; Vol. 2:1-80, 1824; 81-146, 1825; 147-290, 1832; 291-426, 1833; 427-498, 1834; 499-780, 1835; 781-814, 1837). DESHAYES, G. P. 1856—1865. Description des Animaux sans Ver- teébres Découverts dans le Bassin de Paris pour Servir de Supplément a la Descriptions des Coquilles Fossiles des En- virons de Paris Comprenant un Revue Général de Toutes les Espéces Actuallement Connues ... Bailliére et fils: Paris. Vol. 1:912 pp.; Atlas: 87 + 2 pls.; Vol. 2:968 pp.; Vol. 3: 668 pp.; Atlas:107 pls. (collation of Vol. 1:1—80, pls. 1-10, 2 Nov. 1856; 81-160, pls. 11-20, 28 Feb. 1857; 161-240, pls. 21-30, 19 May; 241-312, pls. 31—40, 17 July; 313-392, pls. 41-49, 24 Sept.; 393-480, pls. 50-58, 22 Feb. 1858; 481-552, pls. 59-68, 12 May; 553-624, pls. 69-78, 28 Aug.; 625-704, pls. 79-87, 16 Nov.; 705-912, 10 July 1860). DuruHaM, J. W. 1950. 1940 E. W. Scripps cruise to the Gulf of California, part II: Megascopic paleontology and marine stratigraphy. Memoir of the Geological Society of America 43:vili + 216, 48 pls. EMERSON, W. K. & L. G. HERTLEIN. 1964. Invertebrate megafos- sils of the Belvedere Expedition to the Gulf of California. Transactions of the San Diego Society of Natural History 13(17):333-368. EMERSON, W. K. & E. L. PUFFER. 1957. Recent mollusks of the 1940 “E. W. Scripps” Cruise to the Gulf of California. Nov- itates, American Museum of Natural History, 1825:57 pp. FIscHer, P.-H., I 1860. Du genre Eucharis. Journal de Conchy- liologie 8[(2)4](1):23—26. FISCHER, P.-H., I 1886. Nouvelles observations sur le genre Eu- charis Recluz. Journal de Conchyliologie 34[(3)26](3):193— 203, pl. 11 (part). GouLp, A. A. 1850. (... shells from the Exploring Expedition ...). Proceedings of the Boston Society of Natural History 3(3):214-218. Gray, J. E. 1842. Mollusks. Pp. 48-92 in Synopsis of the Con- tents of the British Museum, 44th ed. British Museum (Nat- ural History): London. [iv] + 308 pp. (concerning: Iredale {1913]). HANLEY, S. C. T. 1842-1856. An Illustrated and Descriptive Cat- alogue of Recent Bivalve Shells. Williams & Norgate: Lon- don. xviii + 392 + 24 pp., pls. 9-24 (pp. 1-32, late 1842; E. V. Coan, 1999 pp. 1—32 (reissue), 33-144, pls. 9-13, early 1843; pp. 145— 272, late 1843; pls. 14—16, late 1844; pls. 17-19, 1846; pp. 273-392 + xviii + 24 pp., pls. 20—24, 1856). HEDLEY, C. 1907. The results of deep sea investigation in the Tasman Sea. II. The expedition of the ““Woy Woy.”’ 2.— Mollusca from eight hundred fathoms. Thirty five miles east of Sydney. Records of the Australian Museum 6(5):356— 364, pls. 66, 67. HERTLEIN, L. G. & U. S. GRANT, Iv. 1972. The geology and paleontology of the marine Pliocene of San Diego, Califor- nia. Part 2B: Paleontology: Pelecypoda. Memoirs of the San Diego Society of Natural History 2:135—409, frontis., pls. 27-57. HERTLEIN, L. G. & A. M. STRONG. 1947. Eastern Pacific expe- ditions of the New York Zoological Society. XXXVI. Mol- lusks from the west coast of Mexico and Central America. Part V. Zoological, New York Zoological Society 31(4): 129-150, pl. 1. Hinps, R. B. 1843. (descriptions of new species of shells col- lected during the voyage of Sir Edward Belcher, C. B., and by H. Duming, Esq., in his late visit to the Philippine Islands ...). Proceedings of the Zoological Society of London for 1843[11](124):55—59. HopcE, J. T., with an appendix by T. A. Conrapb. 1841. Obser- vations on the Secondary and Tertiary formations of the southern Atlantic states. American Journal of Science and the Arts 41(2):332—348, pl. 2 (Conrad appendix: pp. 344— 348, pl. 2; repr.: Hodge [1843]). HopceE, J. T., with an appendix by T. A. CONRAD. 1843. Obser- vations on the Secondary and Tertiary formations of the southern Atlantic states. Reports of the First, Second and Third Meetings of the Association of American Geologists and Naturalists: 94-111, pl. 5. Hornes, M. 1859. (theilte de Charaktere eines neuen Bivalven- Geschlechtes .. .). Verhandlungen der Kaiserlich-K6niglich- en Zoologisch-Botanischen Gesellschaft in Wien 9 (Sit- zungsberichte):7 1—73. Hornes, M. 1870. Die fossilen Mollusken des Tertiaer-Beckens von Wien, II. Band: Bivalven. Abhanglungen der Kaiserlich- KG6niglichen Geologischen Reichsanstalt 4: 479 pp., 85 pls. HOFFSTETTER, R. 1948. Notas sobre el Cuaternario de la Penin- sula de Santa Elena (Ecuador). I. Pelecypoda del Tercer Tablazo. Boletin de Informaciones Cientificas Nacionales (Quito) 2(13/14):67—83. IREDALE, T. 1913. A collation of the molluscan parts of the Syn- opsis of the Contents of the British Museum, 1838-1845. Proceedings of the Malacological Society of London 10(4): 294-309. IREDALE, T. 1930. More notes on the marine Mollusca of New South Wales. Records of the Australian Museum 17(9):384— 407, pls. 62—65. JORDAN, E. K., with introduction by L. G. HERTLEIN. 1936. The Pleistocene fauna of Magdalena Bay, Lower California. Contributions of the Department of Geology, Stanford Uni- versity 1(4):103—173, pls. 17-19. Kay, E. A. 1979. Hawaiian Marine Shells. Special Publication of the Bernice P. Bishop Museum 64(4):xviii + 653 pp. KEEN, A. M. 1958. Sea shells of Tropical West America; Marine Mollusks from Lower California to Colombia. Ist ed. Stan- ford University Press: Stanford, California. xii + 624 pp., 10 pls. KEEN, A. M. 1971. Sea shells of Tropical West America; Marine Page 151 Mollusks from Baja California to Peru. 2nd ed. Stanford University Press: Stanford, California xiv + 1064 pp., 22 pls. Lamy, E. 1925. Note sur le genre Basterotia Mayer 1859 (mol- lusques lamellibranches). Comptes Rendus, Congres des So- ciétés Savantes de Paris et des Départments (Section des Sciences) for 1925:503—508 (repr.: 1—6, 1926). LATREILLE, P. A. 1804. Eucharis. Nouveau Dictionnaire d’Histoire Naturelle Appliquée aux Arts ... 24(Tab.):175. Nowe Lt-UstICcKE, G. W. 1971. A Supplementary Listing of New Shells (illustrated). Revised ed. To be added to the checklist of the marine shells of St. Croix. Nowell-Usticke: St. Croix, U.S. Virgin Islands. 31 pp., 6 pls. Otsson, A. A. 1961. Mollusks of the Tropical Eastern Pacific Particularly from the Southern Half of the Panamic-Pacific Faunal Province (Panama to Peru). Panamic-Pacific Pele- cypoda. Paleontological Research Institution: Ithaca, New York. 574 pp., 86 pls. Otsson, A. A. & A. HARBISON. 1953. Pliocene Mollusca of Southern Florida with Special Reference to Those from North Saint Petersburgh ... with Special Chapters on Tur- ridae by William G. Fargo and Vitrinellidae and Fresh- Water Mollusks by Henry A. Pilsbry. Monograph of the Academy of Natural Sciences of Philadelphia 8:v + 457 pp., 65 pls. PELSENEER, P. 1911. Les lamellibranches de l’Expédition du Si- boga. Partie anatomique in M. Weber (ed.), Siboga-Expe- ditie, Vol. 34(53a)[= livr. 61]:125 + [2] pp., 26 pls. Brill: Leiden. PuHiLipp!, R. A. 1845. Diagnosen einiger neuer Conchylien. Ar- chiv fiir Naturgeschichte 11(1):50—71. PONDER, W. E 1968. Three commensal bivalves from New Zea- land. Records of the Dominion Museum 6(9):125—131. PONDER, W. F 1971. Some New Zealand and subantarctic bi- valves of the Cyamiacea and Leptonacea with descriptions of new taxa. Records of the Dominion Museum 7(13):119— 141, pl. 1. PoweLL, A. W. B. 1952. New Zealand molluscan systematics, part 1. Records of the Auckland Institute and Museum 4(3): 169-185. PRASHAD, B. 1932. The Lamellibranchia of the Siboga Expedi- tion. Systematic part II: Pelecypoda (exclusive of the Pec- tinidae), Vol. 34(53c)[= livr. 118]:353 pp., 9 pls., 1 chart, in M. Weber (ed.), Siboga-Expeditie. Brill: Leiden. RecLuz, C. A. 1850. Monograph d’un nouveau genre de co- quilles bivalves, G. Eucharis. Journal de Conchyliologie 1(2):164-169. REEVE, L. A. 1843-1844. Monograph of the genus Corbula. Vol 2, in L. A. Reeve (ed.), Conchologia Iconica; or, Illustrations of the Shells of Molluscous Animals. 5 pls. (pl. 1, Aug. 1843; pl. 2, Sept.; pl. 3, Jan. 1844; pl. 4, April; pl. 5, May). ROSEWATER, J. 1984. A new species of leptonacean bivalve from off northwestern Peru (Heterodonta: Veneridae: Lasaeidae). The Veliger 27(1):81-89. Sars, G. O. 1878. Bidrag til Kundskaben om Norges Arktiske Fauna. I. Mollusca regionis Arcticae Norvegiae. Brégger: Christiania. xiii + [11] + 466 pp., 34 + 18 pls., 1 map. Situ, E. A. 1890. Report on the marine molluscan fauna of the island of St. Helena. Proceedings of the Zoological Society of London for 1890(2):247-317, pls. 21—24. Vokes, H. E. 1981. Notes on the fauna of the Chipola Forma- tion—XXIII. On the occurrence of the genus Basterotia (Mollusca: Bivalyia). Tulane Studies in Geology and Pale- ontology 16:157—-164. The Veliger 42(2):152—156 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 Laboratory Observations of the Feeding Behavior of the Cirrate Octopod, Grimpoteuthis sp.: One Use of Cirri JAMES C. HUNT Deep Sea Research Department, Japan Marine Science and Technology Center, 2-15 Natsushima-Cho, Yokosuka, 237 Japan Abstract. A single specimen from an undescribed species of Grimpoteuthis (Octopoda: Cirrata) was examined in captivity for 53 days. It was successfully fed both brine shrimp (Artemia) nauplii alone and a mixture of adult brine shrimp with nauplii. Three distinct feeding modes were observed. The first, envelopment, was only exhibited when nauplii alone were presented. The second, entrapment, was exhibited when the mixture of nauplii and adults was presented. The third, cirri-generated current feeding, showed that the cirri in this animal beat in coordinated metachronic patterns to generate currents which transport individual adult brine shrimp under the bell and toward the mouth. Indirect observation suggests these cirral currents function in the other two feeding methods as well. INTRODUCTION Techniques for maintenance, rearing, and culture of shal- low-water cephalopod species have improved greatly over the past 15 years, (see Boletzky & Hanlon, 1983; Hanlon, 1987; Hanlon, 1990). However, there have been no reports of successful long-term maintenance of cirrate octopuses in aquaria. One notable report on the behavior of the cirrate octopod Opisthoteuthis Verrill, 1883, was presented by Pereyra (1965) using specimens housed in shipboard aquaria. Long-term laboratory study has been limited because cirrate octopods are extremely delicate animals which live too deep to be collected by SCUBA divers, while trawl or plankton nets invariably damage them. In order to gather specimens in good condition, a submersible with appropriate collecting chambers must be used. A single cirrate octopod of the genus Grimpoteuthis Robson, 1932, was collected by a submersible and main- tained for nearly 2 mo in an aquarium. Three distinct feeding modes were observed, including direct observa- tion that the cirri beat in metachronic activity patterns to assist transport of prey toward the mouth. METHODS The Monterey Bay Aquarium Research Institute’s (MBA- RI) research vessel R/V Point Lobos, carries a Hysub model ATP 40-1850, a remotely operated vehicle equipped for exploration to depths of 1000 m (see Rob- ison, 1993). A Sony DXC 3000 camera aboard the ROV records images of animals in situ on high resolution Be- tacam videotape. All information, including time-code and scientific observations from the audio track of the videotape, is entered into a computer database. A 7.5-L canister (Youngbluth, 1984) allows animals to be collected without being harmed physically. The col- lector is a clear Plexiglas tube with pivoting top and bot- tom lids that move laterally in unison to open or close the tube. The ROV pilot maneuvers the open tube over an animal without touching it, and then seals the animal and the water surrounding it inside the cylinder. On 8 April 1994, a specimen-collecting dive for the Monterey Bay Aquarium caught a single Grimpoteuthis specimen at a depth of 284 m. The animal was brought to the surface and was moved within 2 hours to a 15 gallon Plexiglas aquarium (44 cm X 34 cm base; 38 cm height) at MBARI’s laboratory in Moss Landing, Cali- fornia. The facility has aquaria in a temperature- and light-controlled laboratory designed for observation of deep-sea specimens. The seawater system is a closed re- circulating type with a 350 gallon reserve tank and sand filtration. Water in the reserve tank is changed monthly. Temperature is maintained at a target of 4°C (+ 0.5°C) and salinity is maintained at a target of 34 ppt (33.7—34.4 ppt). The temperature and salinity of the water in the laboratory tank were similar to the water from the sam- pler and so the octopod was dipped directly into the aquarium. The apparent condition of the animal was ex- cellent upon transfer into the tank. It bobbed near the surface for a short time, then settled to the bottom. The animal was left overnight to acclimate to its new sur- roundings. Observations were made over a period of 53 d. The laboratory was dark except when observations were made with the aid of a red light. Observations also were made using a Litton Model 982 (Gen 2.5 photo- multiplier chip) night-vision monocular with infrared beam. Observations and feeding trials were conducted ev- ery third day for a period of 30 min—1 hr. During the first two trials, pieces of fresh fish and other seafood bought at a local fish market were offered without success. For J. C. Hunt, 1999 the third attempt, live Artemia nauplii were offered, ac- cepted, and subsequently used throughout the study. The brine shrimp were reared in inverted pyramid-shaped aquaria using 2 tbs. of eggs in 4 L of water. The shape of the aquaria allowed nauplii to concentrate near the bot- tom tip where 100 mL was siphoned off for food. New cultures were started weekly. Beginning with the fourth attempt, adult brine shrimp supplied by the Monterey Bay Aquarium were also added to the nauplii. The nauplii were added to adults in a 250 mL beaker such that the resulting mix contained about 10 to 20 adults. RESULTS During the 53 days in captivity, 14 displays of feeding were observed. After death, the animal, a mature female, weighed 39.3 g and had a total length of 78 mm. The animal was not weighed before this time to avoid dis- turbing it, but no appreciable difference in size was ob- served during the period of captivity. This octopod died after an unusual drop in salinity oc- curred in the aquarium (to 33.4 ppt), whereupon the spec- imen was transferred to another aquarium, but by then the fins were small and discolored and the condition of the animal appeared poor. It died the next day, releasing 18 undeveloped eggs through a tear in the mantle wall tissue located behind the siphon. The eggs were imma- ture, lacking the secondary capsule, and did not therefore represent a spawning of the animal. The ovaries were exposed through this tear, and they contained more eggs of varying sizes. Smaller eggs were nearest the mantle apex. Larger eggs were farther inside the mantle, toward the base of the ovary. General Behavior The animal was in excellent condition when placed into the aquarium, with only a small wound on the mantle epidermis, which healed during the first several weeks. The fins of cirrates and midwater squids are indicative of general health (personal observation). With rare excep- tion, specimens with frayed, discolored, or otherwise damaged fins soon die. The fins of this specimen looked completely healthy throughout the observation period. When the animal was active, it bobbed continually at the surface. This action eventually stretched the tissue around the area of the posterior mantle which breached the sur- face, but without ill effect. When observed using night vision, the octopod spent most of its time resting on the bottom of the tank, occa- sionally swimming (once or twice per hour) to the surface and then settling back to the bottom. Under dim red light, it became more active, swimming and bobbing near the surface as soon as the light was turned on, eventually settling back down to the bottom. After it settled, it would resume the behavior exhibited when observed using night vision (only swimming once or twice per hour). During Page 153 the first two weeks; the active swimming and bobbing period under red light lasted from 8-12 min. By the final 2 weeks, that active time had reduced to 4—6 min, sug- gesting the animal was becoming accustomed to the dis- turbances of turning on the red light. When resting on the bottom, the arms were curled forward along the web mar- gin, the tips oriented dorsally (away from the siphon), such that the tips of the first arms nearly touched, and the other three arm pairs oriented similarly along either side. Locomotion The principal means of locomotion was the medusoid- like contractions of the bell-shaped concavity created by the arms and interbrachial webbing. Contractions oc- curred at a rate of about 1.2 per sec. Fins often supported this movement with simultaneous beats, at a rate of 2 to 3 fin strokes for every medusoidlike contraction. A fin stroke almost always occurred at the same time as a con- traction. Yet the fins also acted independently—rotating back and forth to control attitude and direction during swimming and when passively sinking to the tank floor. Feeding Behaviors The Grimpoteuthis exhibited three distinct feeding modes: (1) envelopment, using the arms and web to en- velop prey within the oral surface of the web (Figure 1a); (2) entrapment, using the tank bottom to trap prey be- neath the bell-shaped expansion of the arms and web (Figure 1b); and (3) cirri-generated current feeding, cre- ating currents using the cirri to draw prey beneath the web margin between two arms of the resting cephalopod (Figure 1c). 1. Envelopment: During the first week, only non-living foods were offered, including pieces of fish, crab, shrimp, squid, and fish roe. However, they did not elicit any feed- ing behavior. When live Artemia nauplii were poured di- rectly over the octopus swimming near the top of the tank, the cirrate immediately opened its arms and web wide and enveloped a dense group of brine shrimp within the arm webbing. The octopod sank to the bottom of the tank and the inflated arm-web remained pinched shut, but the enclosed volume proceeded to shrink in size, first dis- tally, working the shrimp toward the beak. This entire behavioral sequence was repeated once more during this feeding session. Cirri occur along the oral surface of the arms and are arranged in two rows. The distal edge of the interbrachial web (the web margin) is what pinches closed, enveloping concentrations of planktonic prey. The volume of water is reduced by gradual contraction of the brachial web and arms, progressing from the arm tips toward the mouth. The result is a continually reducing volume of water with an increasing prey density near the mouth. As water slow- ly escapes through the small aperture created by the arm Page 154 The Veliger, Vol. 42, No. 2 Figure 1 (a) The envelopment feeding behavior of Grimpoteuthis. (b) The entrapment feeding behavior of Grimpoteuthis. (c) The cirri-generated current feeding behavior of Grimpoteuthis. a, Artemia; c, cirri, s, sucker. J. C. Hunt, 1999 tips, it appears there is some mechanism preventing the prey from escaping with the water. Though direct obser- vations were not possible, it appears that the cirri aid in pushing prey toward the mouth. Rippling patterns on the arm-web tissue were highly suggestive of cirri movement underneath that would be similar to the observable cirri activity described below in the cirri-generated current feeding behavior. 2. Entrapment: In order to increase the amount of food ingested, adult Artemia were added to the beaker of nau- plu during the next feeding sequence, with similar en- velopment followed by trapping of food against the bot- tom. After enveloping and eating the shrimp, the Grim- poteuthis swam to the top of the tank, extended its arms such that the webbing became bell-shaped, and sank through the water to the tank floor (see photo in Pereyra (1965, fig. 3e)). The animal controlled its attitude, keep- ing the tips of its extended arms directed downward, by making small adjustments using its fins, but the fins did not actively push the animal down—it sank due to its slightly negative buoyancy. During its descent, it trapped shrimp beneath the rim of the bell-shaped arm-web. When the octopod settled, it forced prey into its mouth by forcing the water beneath the oral surface toward the mouth. To accomplish this, the points located about one- fourth of the arm length from the tip of each arm were pulled together creating an enclosed volume of water. The tips of the arms remained on the bottom; the arms curled inwardly to meet and isolate a volume of water. As the arms contracted slowly beginning where the inner surfac- es of the arms touched, the enclosed volume became pro- gressively smaller. When the volume was confined to the area created within the proximalmost one-third of the arm-web, it stopped reducing further. The animal held this position for several minutes, after which it opened its arms and began moving. Only a few shrimp were seen escaping from the region under the arm-web. It is clear that the vast majority of the prey in the swarm trapped along the bottom were eaten. Once again, I believe the cirri assisted food transport due to the ripples seen in the arm-web, but I have no direct evidence of cirri movement for this feeding behavior. 3. Cirri-generated current feeding: Entrapment against the bottom was exhibited for the remaining weeks, with envelopment observed only once more. Trapping usually was displayed twice at each feeding, followed by cirri displacement feeding behavior where the octopod re- mained on the bottom and drew individual adult brine shrimp into a gap created by lifting one section of its web margin between two adjacent arms and subsequently cre- ating slow currents of water using the cirri. When a por- tion of the margin was lifted high enough, cirri were ob- served aligned toward the edge and moving in metach- ronic waves toward the mouth. Furthermore, individual shrimp were observed being pulled into the opening by Page 155 these waves, often against the direction they were swim- ming. The cirri were clearly observed beating in coordinated metachronic waves to create currents to draw food into the concavity of the oral surface of the arm-web. Currents were generated by the cirri along two adjacent arms in opposite directions such that individual brine shrimp were drawn to the edge of the webbing between these two arms. The beating action of each cirrus includes a quick contraction of muscles along one side, bending the struc- ture into an arc. The cirrus recoils and regains its original shape relatively slowly compared to the contraction, but it is ready to beat again in time for the next wave. I estimated the periodicity of these waves to be 1 per sec, and that the cirrus recoil takes about twice as long as the contraction. DISCUSSION The adaptive significance of the cirri that characterize cir- rate octopods has received little attention in the literature. Young (1977) noted that cirri have large nerves and serve a sensory role, probably to detect the presence of food. Villanueva & Guerra (1991) suggested that microvilli and fusiform structures, detailed using SEM, may serve an “olfactory-like”’ role similar to the role suggested for the cirri of Vampyroteuthis Chun, 1903. It is clear from the current study that at least one functional role of the cirri for this Grimpoteuthis species is to assist in food capture and transport. I note here that the cirri of Vampyroteuthis do not appear to be used in the manner described in this paper and thus may provide a different role for vampy- romorphs (Hunt, personal observation). Many of the moving and resting behaviors described here concur with those given for Opisthoteuthis by Pereya (1965). Specifically, the resting position, swimming, and bobbing behavior, and function of fins in balancing and attitude adjustment are virtually identical in all respects between the specimens from both studies. Some of the behaviors witnessed here had been sus- pected from the morphology of cirrates. Pereyra (1965) speculated that such trapping may be possible although he had no direct evidence. Berry (1952) speculated that cirri may be used to move micro-plankton toward the mouth, but he gave no clear explanation as to how this would be accomplished. Gut-content analysis on one individual indicates that this species feeds on swarming micro-polychaetes (Hoch- berg, personal communication). The stomach contents discussed for three Opisthoteuthis species show that these animals feed on tiny prey, predominantly crustaceans and polychaetes (Pereyra, 1965; Villaneuva & Guerra, 1991). Pereya (1965) noted that very little sediment appeared in the stomach of Opisthoteuthis. Yet this Grimpoteuthis sp. has been observed to feed by sitting on the bottom and stirring up sediments to flush out infaunal microorgan- Page 156 isms (Hochberg, personal communication). It is difficult to determine whether these observations reflect a behav- ioral difference between genera or whether a suite of feeding behaviors is used by various cirrates under dif- ferent circumstances. The issue is further confused be- cause the traits used for classifying these genera are still being debated (Hochberg, personal communication). Nevertheless, for such tiny prey to be ingested in useful numbers, they must be captured in groups. An animal living along the bottom may be able to trap prey using the surface. But there is nothing against which to trap prey in the open space of the water column. Given that Grimpoteuthis has been observed in the water column 50 m above the bottom and reacted immediately to envelop a group of swarming shrimp nauplii, it is likely that the envelopment mode could be used to capture food above the sea bed. Entrapment or cirri-generated current feeding would be useful only along the bottom. Thus it seems reasonable to postulate that the behaviors exhibited in aquaria have natural correlates. It is of course possible that the changing modes of feeding exhibited in this study reflect confinement in a small aquarium. However, one fact suggests the behaviors observed resulted from the food presented and not the tank itself. Food was always introduced in the same man- ner, poured directly over the head of the octopod as it swam near the surface. In the third trial when only Ar- temia nauplii were presented in a dense cloud directly around the octopod, envelopment was observed. Yet on subsequent trials when adult brine shrimp were added to the swarm of nauplii, envelopment was quickly aban- doned for entrapment. During these feeding bouts, en- trapment occurred first (usually twice), followed by cirri- generated current feeding, the latter being used primarily to capture adult brine shrimp. Future experiments intro- ducing a varied feeding regime of brine shrimp adults and/or nauplii should demonstrate clearly whether these feeding behaviors are indeed reflective of a natural re- sponse to various food sizes and concentrations, or an artifact of captivity. Incirrate octopuses have a reputation for being easy to care for in laboratory aquaria. Relatively shallow-water ciurates, such as Grimpoteuthis or Opisthoteuthis may The Veliger, Vol. 42, No. 2 also be good aquarium animals, though they would be somewhat more difficult to maintain than incirrates, as special attention needs to be given to light, temperature, and oxygen conditions. ACKNOWLEDGMENTS This research was supported by the Packard Foundation and the Monterey Bay Aquarium Research Institute. I wish to thank Drs. William Hamner, Eric Hochberg, Bruce Robison, and the reviewers for this journal for their critical reading of the manuscript. Special thanks to Gil- bert Van Dykhausen, the Monterey Bay Aquarium, and the captain and crew of the R/V Point Lobos for their efforts in acquiring the octopus and to K. Carlson for her illustrations. LITERATURE CITED Berry, S. S. 1952. The flapjack devilfish, Opisthoteuthis, in Cal- ifornia. California Fish and Game 38(2):183-188. BOLeTZzky, S. V. & R. T. HANLON. 1983. A review of the labo- ratory maintenance, rearing and culture of cephalopod mol- lusks. Memoires of the Natural History Museum of Victoria 44:147-187. HANLON, R. T. 1987. Mariculture. Pp. 291-305 in P. R. Boyle (ed.), Cephalopod Life Cycles, volume II: Comparative Re- views. Academic Press: New York. HANLON, R. T. 1990. Maintenance, rearing, and culture of Teu- thoid and Sepioid squids. Pp. 35—62 in D. L. Gilbert, W. J. Adelman, Jr. & J. M. Arnold (eds.), Squid as Experimental Animals. Plenum Press: New York. PEREYRA, W. T. 1965. New records and observations of the flap- jack devilfish, Opisthoteuthis californiana Berry. Pacific Science 19:427—441. Rosison, B. H. 1993. Midwater research methods with MBARI’s ROV. MTS Journal 26:32—39. VILLANEUVA, R. & A. GUERRA. 1991. Food and prey detection in two deep-sea cephalopods: Opisthoteuthis agassizi and O. vossi (Octopoda: Cirrata). Bulletin of Marine Science 49(1— 2):288—299. Youna, J. Z. 1977. Brain, behaviour and evolution of cephalo- pods. Symposium of the Zoological Society of London 38: 377-434. YOUNGBLUTH, M. J. 1984. Manned submersibles and sophisticat- ed instrumentation: tools for oceanographic research. Pp. 335-344 in Proceedings of SUBTECH 1983 Symposium. Society of Underwater Technology, London. The Veliger 42(2):157—168 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 Ontogenetic Changes in Boring Behavior by the Rock-Boring Bivalve, Barnea manilensis (Pholadidae) YASUHIRO ITO University Museum, University of Tokyo, Hongo 7-3-1, Tokyo 113, Japan Abstract. Ontogenetic changes in morphology related to boring behavior by Barnea manilensis (Philippi, 1847) were observed. The shell outline changes from round to elongate during the transition from pediveliger to juvenile stages. Along with such a morphological change, the boring style gradually changes from anterior boring in early round-shelled juveniles, where the opening of the anterior valve margin, with rotation around a dorso-ventral axis, abrades the burrow wall, to ventral boring in older and larger long-shelled individuals, which open the ventral margin with rotation around a longitudinal axis (hinge line) for abrasion. This observation, with an examination of the literature, leads to the sug- gestion that early juveniles of all pholads employ anterior boring. Later, many pholads continue anterior boring through- out life, whereas others gradually shift toward ventral boring. In addition, anterior boring is thought to be a primary character of pholads, and ventral boring a specialized character derived from anterior boring. INTRODUCTION Barnea manilensis (Philippi, 1847) is a common bivalve of the Pholadacea which bores into soft rock (e.g., mud- stone and shale) of the intertidal zone and, in Japan, oc- curs from Hokkaido to Okinawa (Habe, 1977). Boring behavior occurs not only in the Pholadacea, but also in the Myacea, Mytilacea, Veneracea, Cardiacea, Gastro- chaenacea, and Hiatellacea (Yonge, 1963; Ansell & Nair, 1969). Among these superfamilies, species of Pholadacea are characterized by an ability to bore into a variety of substrata, such as mud, rock, and wood. Accordingly, the Pholadacea exhibit considerable modifications to shell form as a reflection of its varied boring habits (Nair & Ansell, 1968; Roder, 1977; Seilacher, 1985). The Pholad- idae, including B. manilensis, bore into mud, rock, and other solid substrata and are regarded as less specialized for boring than the wood-boring Teredinidae (also Pho- ladacea) (Nair & Ansell, 1968). The boring mechanism of the Pholadacea has been studied extensively (Miller, 1924; Nair & Ansell, 1968; Roder, 1977; Seilacher, 1985). Pholads are generally con- sidered to be mechanical borers (Miller, 1924; Yonge, 1963; Ansell & Nair, 1969), although chemical boring has also been suggested (Smith, 1969; Morton, 1985, 1986). The boring procedures of pholads have been detailed for the wood borer Teredo (Teredinidae) (Miller, 1924), the rock borer Zirfaea crispata (Pholadidae) (Nair & Ansell, 1968), and the mud borers Cyrtopleura costata and Bar- nea candida (Pholadidae) (Roder, 1977; Seilacher, 1985). The boring behavior of wood borers in their early onto- genetic stage has been outlined with special emphasis on behavior and morphology (Isham & Tierney, 1953; Turn- er & Johnson, 1971). Ontogenetic changes in morphology related to boring behavior in rock-boring pholads have not, however, been well documented. In the present study, ontogenetic changes in shell and tissue morphologies were observed for B. manilensis. Along with these morphological changes, B. manilensis undergoes a conspicuous change in boring style. It changes from anterior boring, where the opening of the anterior valve margin with rotation around a dorso-ventral axis abrades the burrow wall, to ventral boring, or opening the ventral margin with rotation around a longitudinal axis (hinge line). The present study describes this change in boring style. Based on comparision with other boring pholads, the significance of this ontogenetic change in B. manilensis is discussed in relation to specializations in boring style by pholads. MATERIALS AnD METHODS Living specimens of Barnea manilensis were collected from within tuffaceous mudstone at Itsuwa-machi, Ku- mamoto, western Japan, in September 1994. Larger in- dividuals were kept alive in seawater tanks at the Aitsu Marine Biological Station of Kumamoto University. In- dividuals were extracted from the rock without destruc- tion of their shells for later observation of morphology and boring behavior. Their gametes were also obtained from the gonads by dissection and were fertilized in ex- perimental dishes. By culturing their larvae, numerous individuals were obtained for observation of their pedi- veliger larval and post-larval development. Larvae were cultured in seawater (salinity = 31—34%o; temperature = 21—25°C). When the larvae formed a foot and entered the pediveliger stage, fragments of the mudstone in which the adult parents had lived were placed in the culture con- Page 158 The Veliger, Vol. 42, No. 2 tainers as a substratum for their settlement and boring. Crawling and boring behavior were observed with a bin- ocular microscope for more than 50 individuals of pedi- veliger larvae and post-larvae, and recorded with a video camera. More than 50 specimens of pediveliger larvae and post-larvae were fixed in a solution of commercial sugar (10%), formalin (1%), and sodium bicarbonate (0.05%) in seawater (Castro & Le Pennec, 1988) in order to ob- serve the hinge and muscle structures with an optical mi- croscope (Sakai & Sekiguchi, 1990). Further, after fixing, the shells were cleaned and separated using sodium hy- pochlorite, in order to observe the form and structure of the shells with a scanning electron microscope (SEM). Specimens of the post-larval boring stage of B. mani- lensis were also collected for morphological observations from intertidal mudstones at Cape Taito, Chiba, Japan, in June 1991. In order to examine shell development, 30 specimens of pediveliger larvae reared in a culture tank, and 90 specimens of boring-stage shells collected from Cape Taito, were measured. Shell size was measured from SEM photographs, optical photographs, and sketches. All specimens examined are deposited in the Univer- sity Museum, University of Tokyo (UMUT). OBSERVATIONS Ontogenetic Changes in Shell Morphology Besides the clearly apparent change of its shell outline from round to elongate, Barnea manilensis exhibits other ontogenetic changes in shell and soft body morphologies during the transition from pediveliger to juvenile stages. Because these changes are intimately related to the change in its boring behavior, they will be described in detail below. Ontogenetic changes in shell and tissue morphologies may be divided into four stages, termed A, B, C, and D. Stage A corresponds to the pediveliger stage. The begin- ning of stage B is defined by the addition of a dissoconch with formation of the ventral condyle and umbonal re- flection. The extension of the reflection toward the pos- terior over the dorsal condyle (umbo) in turn defines the beginning of stage C. Stage D is defined by the disap- pearance of the ventral condyle. The morphological terms used here are given in Figure 1. Stage A: The round shell with a long provinculum in the pediveliger stage has concentric growth lines and a nearly smooth ventral valve margin, but lacks a ventral condyle (L = 0.23—0.30 mm) (Figures 1A, 2A). Each of the two adductor muscles extends perpendicular to the sagittal plane of the shell (Figure 3A) and their thickness does not vary throughout their length. Stage B: This stage represents a short period during and following metamorphosis (L = approx. 0.3—0.8 mm) (Figures 1B, 2B). The ventral valve margin protrudes ventrally to form a thick and solid ventral condyle: the left condyle has a notch to accommodate the knobby tip of the right condyle. The ventral condyle leaves its trace as an umbonal-ventral sulcus externally and as a ridge internally as the shell grows. The umbonal-ventral sulcus divides the dissoconch into anterior and posterior parts. The anterior external surface has conspicuous growth lines and radial ribs which form undulating denticulate ridges (Turner, 1969) at their intersections. The posterior surface is less ornamented than the anterior and has weak concentric growth lines. The protruding ventral condyles produce two gapes along the ventral valve margin when the valves are closed: the anterior pedal gape and the posterior siphonal gape. The siphonal gape is narrow immediately after metamorphosis begins, then widens as the posterior mar- gins become more concave with growth. The two valves make contact at the dorsal condyles (Figure 3B) and are connected by a ligament, located just posterior to the dorsal condyles, and extending between the recessed resilifer of the rmght valve and the dorsal surface of the chondrophore on the left valve. The liga- ment and the left chondrophore are fully developed in a newly metamorphosed individual, but they become re- duced rapidly toward the end of this stage. The anterior adductor muscle (AAM) is symmetrical between the two valves, and attaches anterior to the dor- sal condyle (umbo). It extends toward the umbo with de- velopment of the outward reflection of the antero-dorsal margin. This reflection is termed the umbonal reflection (Turner, 1969). The posterior adductor muscle (PAM) ex- tends obliquely to the shell’s sagittal plane during this stage (Figure 4B). Its right end attaches to the inner sur- face of the right valve, whereas its left end is located more posteriorly on the slightly outwardly bent and thick- ened postero-dorsal margin of the left valve. The PAM becomes thicker from left to mght, and its scar on the right valve is about two to four times as large as on the left. Stage C: This stage is transitional between stages B and D (L = 0.8—2.0 mm) (Figures 1C, 2C). The shell is lon- gitudinally elongate and oval in shape, but the ventral condyle is still present. Denticulate ridges cover the an- terior external valve surface in the early part of this stage, as they do in stage B. By the end of this stage, these denticulate ridges also appear on the posterior surface, although they are less conspicuous than the anterior ridges. The siphonal gape of the posterior valve margin widens. The pedal gape narrows but becomes longitudinally ex- tended. The ligament and chondrophore become vestigial (Figure 3C). The posterior part of the umbonal reflection, where the AAM< attaches, extends posteriorly beyond the dorsal condyle. This part of the AAM is called the AAM-P herein, and the anterior part of the AAM is termed the Wesito; 1999 Page 159 POST <— ——~»> ANT Figure 1 Morphological characters of four stages of Barnea manilensis. External views of right valves (left side). Internal views of left valves (right side). ANT = anterior; POST = posterior. AAM = anterior adductor muscle scar; PAM = posterior adductor muscle scar; VAM = accessory ventral adductor muscle scar; c = chondrophore; de = dorsal condyle; lh = larval hinge; pd = prodissoconch (larval shell); uvr = umbonal-ventral ridge; uvs = umbonal-ventral sulcus; vc = ventral condyle. The Veliger, Vol. 42, No. 2 Page 160 Y. Ito, 1999 Page 161 AAM-A. The PAM becomes almost equal in diameter at its right and left ends, and again becomes nearly sym- metrical with respect to the sagittal plane. The accessory ventral adductor muscle (VAM) scar becomes apparent in this stage. The VAM attaches to the posterior angle of the pallial line on the umbonal-ventral ridge of each valve. The right and left mantle margins are fused around the VAM. Stage D: This stage is defined by the disappearance of the ventral condyle and is usually represented by larger individuals (L > approx. 2.0 mm) (Figures 1D, 2D). The shell is slender. The umbonal-ventral sulcus is reduced to being merely one of the radial furrows, so that division of anterior and posterior valve surfaces becomes obscure. Denticulate ridges appear on the whole external surface, although more conspicuously anteriorly. Although the ventral condyle disappears, the pedal and siphonal gapes remain. The ligament disappears at this stage. The chondro- phore is reduced to a tiny spine, which protrudes from the dorsal condyle (Figure 3D). The AAM attachment area becomes larger than that of the PAM in this stage, and the PAM becomes symmet- rical. The VAM attaches to a prolongation of the line of the umbonal-ventral ridge (which disappears in this stage). Allometric Shell Growth The round larval shell (stage A) of Barnea manilensis becomes more elongate in larger individuals (stage D). The L/H ratios of pediveliger larval shells (stage A) are about 1.0. These ratios rapidly increase through stages B, C, and the earlier part of stage D (L = approx. 7-10 mm) (Figure 5a), then stay almost constant afterward (L/H = approx. 2.7) (Figure 5b). The dorso-ventral axis passes through the dorsal con- dyle (umbo) and the ventral condyle (or VAM scar). As the axis changes from acline in stage B to prosocline in stage D, the angles (DVA) between the longitudinal axis (or hinge line) and the dorso-ventral axis rapidly decrease from about 80 degrees in stage B to about 40 degrees in stage D at about 10 mm in shell length, and then stay almost constant for the remainder of life (Figure 5c). The DVA and the shell shape (L/H) show a positive correla- tion (r = 0.94), which means that a more slender shell has a more prosocline dorso-ventral axis (Figure 5d). Crawling and Boring Behavior of Barnea manilensis The pediveliger of Barnea manilensis (in growth stage A) does not bore but crawls with a simple set of move- ments (Figure 6A). It extends its foot anteriorly, opens the valves ventrally and attaches the tip of its foot to the substratum. It then closes its valves, its foot contracts, and the resulting motion pulls the body forward. It then relaxes the muscles and the valves gape ventrally. The valve opening in this crawling sequence is only the in- evitable result of extrusion of the foot. It should be mentioned that when growth stage B is reached and boring begins, boring ability is acquired by adding another step to the crawling sequence (Figure 6B), when it opens the anterior margin of the valve. This new- ly added step follows pulling of the body forward by contraction of the foot, in which anterior parts of the closed valves are pressed against the substratum (burrow wall). When the anterior valve margin opens, the anterior valve surfaces abrade the burrow wall (Figure 7B). This is required for boring. The process of boring in stage B is here called anterior boring. This movement is derived from a new muscular movement, which takes place by contracting the PAM and relaxing the AAM (Figure 8B). These muscles work reciprocally to accomplish anterior boring. The boring movement changes with growth, from an- terior boring in stage B to ventral in stage D (Figure 7). An individual in stage B opens only its anterior valve margin (Figure 7B) but, with growth, it tends to open its valves more ventrally. By stage D, ventral valve margin is opened to abrade with both valves’ entire surfaces (Fig- ure 7D). This is called ventral boring. An individual in stage C is transitional between stages B and D, opening its antero-ventral valve margin (antero-ventral boring) (Figure 7C). Development of Boring Behavior Stage A: Boring movement does not occur in this stage, although vigorous crawling activities do. No mor- phological specialization for boring is observed at this Figure 2 Scanning electron micrographs of four stages of Barnea manilensis shells. External views of right valves (upper figure) (A. UMUT RM 27598; B. UMUT RM 27599; C-D. UMUT RM 27597). Internal views of right valves and two valves (lower figure) (A. UMUT RM 27600; B. UMUT RM 27601; C-D. UMUT RM 27596). LV = left valve; RV = right valve. arrow = margin of right PAM scar in stage B. Scale bar: A = 100 pm, B = 100 pm, C = 500 pm, D = 1 mm. Page 162 The Veliger, Vol. 42, No. 2 RV Figure 3 Photographs of the hinge regions of four stages of Barnea manilensis (A. UMUT RM 27600; B. UMUT RM 27602: C. UMUT RM 27596; D. UMUT RM 27594). A-C = Scanning electron micrographs. D = Photographs (shell length = 33 mm). LV = left valve; RV = right valve. c = chondrophore; Ip = ligament pit; dec = dorsal condyle: AAM >= anterior adductor muscle scar. Scale bar: A = 10 pm, B = 50 pm, C = 100 pm, D = | mm. Y. Ito, 1999 Page 163 Figure 4 Photomicrographs showing dorsal views of articulated valves of Barnea manilensis in stage A (left) and stage B (right). In stage A (pediveliger larva), the PAM extends perpendicular to the sagittal plane of the shell. In stage B (metamorphosing individual), the PAM extends obliquely to the sagittal plane. LV = left valve; RV = right valve; arrows indicate attachment positions of the PAM. Scale bar = 100 um. stage. The hinge and musculature enable the valves to rotate around the longitudinal axis (= hinge line), as do the majority of larval bivalves (Figure 7A). The two ad- ductor muscles of the larva are employed only for closing the valves. The relaxation of both adductor muscles leads to a gape along the whole ventral margin. Stage B: The anterior external surfaces of the valves abrade the burrow wall (anterior boring), when the ante- rior valve margin opens widely to rotate around the dor- so-ventral axis by contraction of the PAM (Figure 7B). When the PAM contracts, the AAM is stretched (Figure 8B). This reciprocal movement of the PAM and AAM has also been described for Teredo navalis (Miller, 1924) and Jouannetia cumingii (Morten, 1986). The thick and solid ventral condyles fit tightly together by the notch and knob of their tips, which prevents the valve margin from destruction by concentrating force there during anterior boring. At this stage, the anterior valve margin must open widely even though the posterior margin has a relatively narrow gape. This can be accomplished by the obliquely running PAM. When the PAM contracts, the postero- dorsal margin of the left valve slides inside the right, so that the anterior valve margin can open widely. The over- lapping of the postero-dorsal parts of the two valves ex- tends to an angle of about 30 degrees, but gradually be- comes inconspicuous with growth and is almost invisible by the end of this stage. When the anterior valve margin opens widely due to contraction of the PAM, the AAM located on the um- bonal reflection is stretched. When the PAM relaxes, the AAM also relaxes rather than contracting, which leads to a drawing out of the left valve from inside the right valve. At the same time, the foot is extended and the shell leaves the burrow wall. This is followed by a step in which the two muscles contract simultaneously, and the valves close. During this step, the foot contracts to press the shell against the burrow heading. Stage C: A series of boring movements begins with opening the anterior valve margin by rotation around the dorso-ventral axis due to contraction of the PAM; this movement is identical to that seen in stage B (Figure 7C). This is followed by opening of the ventral valve margin as a result of a change in rotation axis from dorso-ventral to longitudinal. Rotation around the longitudinal axis (hinge line) is effected by contraction of the AAM-P, which is located dorsal to the hinge axis and acts as an abductor muscle (Figure 8C). During this series of movements, the ventral parts of the valves, as well as the anterior parts, abrade the burrow wall (antero-ventral boring). During this boring motion, the AAM-A is stretched. The AAM-A retains the original us @ stage B fe) C) 0 2 4 6 8 Length (mm) 15 20 Length (mm) The Veliger, Vol. 42, No. 2 10. 0 TOMO 66 s0 25 Length (mm) Figure 5 Allometric shell growth in stage B, C, and D of Barnea manilensis from Cape Taito, Chiba, Japan (UMUT RM 27595, 27596). Sa, b = shell length (L)/height (H) ratio plotted against length (L); 5a shows the enlarged dashed area of 5b. 5c = angle (DVA) between longitudinal axis and dorso-ventral axis plotted against length (L). 5d = DVA angle plotted against shell shape (L/H). function of the AAM, i.e., closure of the valves, at this stage and afterward. The ventral shell opening becomes gradually wider with growth. The posterior siphonal gape also becomes wider with growth and enables the anterior margin to open without overlapping the postero-dorsal margin. The ligament becomes vestigial and loses its function during this stage. The AAM-P takes over the function of the ligament to connect the two valves. In stages C and D of B. manilensis, the AAM-P and the PAM are mainly used to open the valves ventrally and anteriorly, respectively. The contraction of the VAM aids valve closure (RGéder, 1977), which in most bivalves is done by contracting the AAM and the PAM synchro- nously. The VAM gradually takes over the role of the ventral condyle to hold the valves together with the dorsal condyle when the valves rotate around the dorso-ventral axis. Stage D: When the ventral valve margin opens by con- traction of the AAM-P, all external surfaces of both valves abrade the burrow wall (ventral boring) (Figure 7D). At the same time, rotation around the prosocline dorso-ventral axis occurs by contraction of the PAM that leads to the valves being somewhat open anteriorly in addition to their being open ventrally. The ventral valve margins, having no ventral condyles, no longer contact each other, in contrast to earlier stages. DISCUSSION Shell Outline and Boring Style This study demonstrates that Barnea manilensis changes its boring behavior throughout ontogeny. Early round-shelled juveniles employ anterior boring, whereas, after going through intermediate shell morphologies and boring styles, the older and larger long-shelled individu- als finally employ ventral boring. Other borers among the Pholadacea also employ anterior and ventral boring. In the pholads, there is a close relationship between boring style and shell morphology (R6der, 1977; Seilacher, Neglitos 1999 Page 165 stage A sequence following sequence Figure 6 Diagrammatic ventral view of crawling cycles in stages A and B of Barnea manilensis. LV = left valve; RV = right valve; f = foot; s = siphon. 1985). In general, pholads having short shells, thick knobby ventral condyles, and large PAMs are anterior borers, because these features open the anterior valve margin for rotation around the dorso-ventral axis to achieve boring. On the other hand, pholads having long shells with narrow pedal gapes and large AAM-Ps are ventral borers because these features open the ventral margin for rotation around the longitudinal axis to pro- duce boring. There are intermediate forms and boring styles between the extremes of anicrior and ventral bor- ing. Anterior Boring vs. Ventral Boring Many species of Pholadacea employ anterior boring throughout life, such as Teredo, and others such as Bar- nea manilensis begin with the early juvenile anterior bor- ing and change toward ventral boring with growth. So far as the author knows, the reverse order of development has never been reported among the Pholadacea. It is prob- able that all Pholadacea species employ anterior boring as early juveniles, although not all have been observed sequence | ale : following : 4 : sea Figure 7 Diagram of boring/crawling styles in the four stages of Barnea manilensis development. Boring movement (sequence 5) changes from anterior boring in stage B to ventral boring in stage D. It does not yet appear in stage A. In stage A, valves open ventrally when adductor muscles relax as in sequence 1. In the stage D, valves open ventrally when the PAM and AAM-P contract as in sequence 5. Circular point = hinge or pivot. LA = longitudinal axis (hinge line); DVA = dorso-ventral axis. to do so. Anterior boring is thought to be the primary character in boring pholads, and ventral boring is consid- ered to be a specialized character derived from anterior boring. This conclusion is based on considerations dis- cussed below. Anterior borers, including juveniles of B. manilensis in stage B, open the anterior valve margin by rotating around the dorso-ventral axis due to contraction of the PAM. The only unique morphological characteristic of anterior boring is the existence of a ventral condyle that defines the ventral end of the dorso-ventral axis. The characteristics of the PAM are essentially those of other The Veliger, Vol. 42, No. 2 stage B s Figure 8 Diagrammatic dorsal view of the boring movement of Barnea manilensis in stages B and C. Circular point = dorsal pivot; s = siphon; f = foot. Closely spaced lines = contracted muscle; widely spaced lines = relaxed or stretched muscle. bivalves, and the slender internal ligament at the dorsal end of the axis is also seen elsewhere in the Myoida. All Pholadacea have a ventral condyle, at least during the juvenile stage immediately after metamorphosis. The ven- tral condyle either disappears or continues to develop de- pending on the boring style of the species. All Teredinidae species are anterior borers that have a prominent ventral condyle (Turner, 1969), which appears immediately after metamorphosis and then develops into a thick knob in larger individuals (Lebour, 1946; Sullivan, 1948; Turner & Johnson, 1971; Fuller et al., 1989; Tan et al., 1993). The members of the Pholadidae, except for the Pholadinae, also possess a ventral condyle (Turner, 1969) throughout life. In three modern genera of Phola- dinae, i.e., Barnea, Cyrtopleura, and Pholas, the ventral condyle is invariably lost in larger shells (Turner, 1969). These genera are ventral borers (R6der, 1977). Although the ventral condyle disappears from B. manilensis in stage D, juveniles in stages B and C still possess it. Zirfaea is an antero-ventral borer in larger individuals (Nair & An- sell, 1968) and is the only genus in the Pholadinae which retains a ventral condyle throughout life. Its umbonal- ventral sulcus becomes invisible with growth (Turner, 1954), which is the same change seen in B. manilensis during its transition from stage C to D. This suggests that the absence of a ventral condyle in other Pholadinae is ascribable to its secondary disappearance during growth and elongation of the shell. On the other hand, some Pholadacea, including B. manilensis in stage D, achieve ventral boring by opening the ventral valve margin with the help of the AAM-P. The AAM-P is located on the dorsal side of the hinge axis, and acts as the abductor muscle (RGder, 1977; Sei- lacher, 1985). In B. manilensis, the umbonal reflection bearing the AAM-P gradually moves to the dorsal side of the dorsal condyle (hinge axis) during the post-larval boring stages. Similar ontogenetic development of the AAM.-P is observed in Zirfaea crispata (Sullivan, 1948; Turner, 1954). The ligament and chondrophore structure in B. manilensis are well developed immediately after metamorphosis when anterior boring is effected, but rap- idly become vestigial with growth, and B. manilensis be- comes a ventral borer. Similar reduction in these struc- tures is observed in the ventral borer Pholas dactylus (Purchon, 1955; Le Pennec, 1980) and the antero-ventral borer Z. crispata (Sullivan, 1948; Purchon, 1955). Ventral boring does not begin until the AAM-P assumes the func- tion of the ligament to connect and open the valves. Pho- lads that have not developed an AAM-P employ anterior boring, as in Teredo (Miller, 1924). The structures needed for ventral boring are more specialized than those for an- terior boring. Diversification of Boring Styles in the Pholadacea Some workers have concluded that anterior-boring pholads specialized from ventral borers (Nair & Ansell, 1968; Seilacher, 1985). This idea is based on the as- sumption that borers primarily use the common burrow- ing mechanism of bivalves for boring in rock as for open- ing and closing the ventral valve margin and rotating around the longitudinal axis of the shell. However, the ventral opening for boring in pholads appears in a later developmental stage than does the opening for normal burrowing and early juvenile movement. In addition, the opening mechanism needed for boring is quite different from that needed for more ordinary movements. In or- dinary burrowing bivalves, when both adductor muscles relax, the ventral valve margin opens by elastic rebound of the ligament and insertion of the foot (Trueman, 1964), which is derived from the crawling sequence of the pedi- veliger stage (Nelson, 1924; Prytherch, 1934; Quayle, Nesta 999 Page 167 1949; Carriker, 1961; Cranfield, 1973). In pholads, how- ever, the opening of the ventral valve margin and rotation around the longitudinal axis is achieved by contraction of the AAM-P, located on the dorsal side of the hinge. The rotation of valves for both anterior and ventral boring is a movement newly added at the time of meta- morphosis to the crawling sequence of the pediveliger stage, as observed in the transition from stage A to B of Barnea manilensis. This new motion is added to the crawling sequence just after contraction of the foot and before relaxation of the muscles. This motion first appears as a rotation around the dorso-ventral axis during the ear- ly juvenile stages, then gradually changes into longitu- dinal motion with growth. All pholads are thought to develop anterior boring im- mediately after metamorphosis, irrespective of the boring style in their later developmental stages. Anterior boring is evidently a primary character common to boring pho- lads. Some pholads, such as Teredo, retain anterior boring throughout life, whereas others, such as Barnea, shift to ventral boring. This study suggests that anterior boring appeared early in the phylogeny of pholads, and that ventral boring evolved later. This interpretation is supported by the fossil record, in which the anterior borers appeared earlier than ventral borers. The oldest pholad, Teredo australis (Moore, 1870), is an anterior borer recorded from the Middle Jurassic. Opertochasma (Martesiinae) and Turnus (subfamily uncertain) probably are anterior borers known from the Upper Jurassic (see Kelly, 1988). Further, the ichnogenus Teredolites, which consists of burrows made in wood by anterior boring (as in modern Teredo or Mar- tesia), is known from the Jurassic (Kelly & Bromley, 1984). The oldest ventral borers, on the other hand, the Pholadinae Pholas? scaphoides (Stephenson, 1952) and Barnea saulae (Kennedy, 1993), are recorded only as far back as the Late Cretaceous. Based on the fossil record, Kelly (1988) suggested that mud borers (ventral borers as used herein) may have been derived secondarily from wood borers (anterior borers). My observations on the ontogenetic development of boring behavior in B. mani- lensis add new insight into the primordial character of anterior boring. ACKNOWLEDGMENTS I wish to thank Professor Kiyotaka Chinzei (Osaka Gak- uin University) and Professor Terufumi Ohno (Kyoto University) for critical reviews of the manuscript and for valuable comments. I am also grateful to Associate Pro- fessor Takao Yamaguchi (Kumamoto University) and As- sociate Professor Satoshi Nojima (Kyushu University) for providing support during the culture of larvae, Mr. Tak- eshi Kitano (Kumamoto Fisheries Research Center) for his helpful advice on culture technique of larvae, and Dr. Louie Marincovich (California Academy of Sciences) for many helpful corrections to the manuscript. Thanks to Professor Takeshi Setoguchi, Associate Professor Haru- yoshi Maeda, Dr. Keiko Yamaguchi, and other colleagues of Kyoto University for many helpful suggestions during the course of this work. LITERATURE CITED ANSELL, A. D. & N. B. Nair. 1969. A comparative study of bivalves which bore mainly by mechanical means. American Zoologist 9:857—868. CARRIKER, M. R. 1961. Interrelation of functional morphology, behavior, and autecology in early stages of the bivalve Mer- cenaria mercenaria. Journal of the Elisha Mitchell Scientific Society 77:168—241. Castro, N. EK & M. LE PENNEC. 1988. Modalities of brooding and morphogenesis of larvae in Oetrea puelchana (D’Orbigny) under experimental rearing. Journal of the Ma- rine Biological Association of the United Kingdom 68:399— 407. CRANFIELD, H. J. 1973. Observations on the behaviour of the pediveliger of Ostrea edulis during attachment and cement- ing. Marine Biology 22:203-—209. FULLER, S. C., Y. Hu, R. A. Lutz & M. CastaGna. 1989. Shell and pallet morphology in early developmental stages of Te- redo navalis Linné (Bivalvia: Teredinidae). The Nautilus 103:24-35. HaseE, T. 1977. Systematics of Mollusca in Japan: Bivalvia and Scaphopoda. Hokuryukan Book Company: Tokyo. [in Jap- anese | IsHAM, L. B. & J. Q. TIERNEY. 1953. Some aspects of the larval development and metamorphosis of Teredo (Lyrodus) ped- icellata De Quatrefages. Bulletin of Marine Science of the Gulf and Caribbean 2:574—589. Ketty, S. R. A. & R. G. BROMLEY. 1984. Ichnological nomen- clature of clavate borings. Palaeontology 27:793—807. KeLLy, S. R. A. 1988. Cretaceous wood-boring bivalves from western Antarctica with a review of the Mesozoic Pholadi- dae. Palaeontology 31:341—372. KENNEDY, G. L. 1993. New Cretaceous and Tertiary Pholadidae (Mollusca: Bivalvia) from California. Journal of Paleontol- ogy 67:397—404. Lesour, M. V. 1946. The species of Teredo from Plymouth wa- ters. Journal of the Marine Biological Association of the United Kingdom 26:381—389. LE PENNEC, M. 1980. The larval and post-larval hinge of some families of bivalve molluscs. Journal of the Marine Biolog- ical Association of the United Kingdom 60:601—617. MILLER, R. C. 1924. The boring mechanism of Teredo. Univer- sity of California Publications in Zoology 26:41—80. Moore, C. 1870. Australian Mesozoic geology and palaeontol- ogy. The Quarterly Journal of the Geological Society of London 26:226—261. Morton, B. 1985. A pallial boring gland in Barnea manilensis (Bivalvia: Pholadidae)?. Pp. 191-197 in B. Morton & D. Dudgeon (eds.), Proceedings of the Second International Workshop on the Malacofauna of Hong Kong and Southern China, Hong Kong, 1983. Hong Kong University Press: Hong Kong. Morton, B. 1986. The biology and functional morphology of the coral-boring Jouannetia cumingii (Bivalvia: Pholada- cea). Journal of Zoology, London, Series A 208:339—366. Nair, N. B. & A. D. ANSELL. 1968. The mechanism of boring Page 168 in Zirphaea crispata (L.) (Bivalvia: Pholadidae). Proceed- ings of the Royal Society of London, Series B 170:155-173. NELson, T. C. 1924. The attachment of oyster larvae. Biological Bulletin of the Marine Biological Laboratory 46:143-151. PRYTHERCH, H. F 1934. The role of copper in the settling, meta- morphosis, and distribution of the American oyster, Ostrea virginica, Ecological Monographs 4:45—107. PurCHON, R. D. 1955. The structure and function of the British Pholadidae (rock-boring Lamellibranchia). Proceedings of the Zoological Society of London 124:859-911. QuaYLeE, D. B. 1949. Movements in Venerupis (Paphia) pullas- tra (Montagu). Proceedings of the Malacological Society of London 28:31-—37. Roper, H. 1977. Zur Beziehung zwischen Konstruktion und Sub- strat bei mechanisch bohrenden Bohrmuscheln (Pholadidae, Teredinidae). Senckenbergiana Maritima 9:105—213. Sakal, A. & H. SEKIGUCHI. 1990. A simple method of exami- nation on hinge apparatus in settled bivalves. Benthos Re- search 39:21—22. [in Japanese] SEILACHER, A. 1985. Bivalve morphology and function. Pp. 88— 101 in D. J. Bottjer, C. S. Hickmann & P. D. Ward (orgs.), Mollusks, Notes for a Short Course. University of Tennes- see, Knoxville, Tennessee. Smit, E. D. 1969. Functional morphology of Penitella conradi relative to shell-penetration. American Zoologist 9:869—880. STEPHENSON, L. W. 1952. Larger invertebrate fossils of the Wood- bine Formation (Cenomanian) of Texas. United States Geo- logical Survey, Professional Paper 242:1—226. The Veliger, Vol. 42, No. 2 SULLIVAN, C. M. 1948. Bivalve larvae of Malpeque Bay, PE.I. Bulletin of the Fisheries Research Board of Canada 77:1— 36. TAN, A. S., Y. Hu, M. Castacna, R. A. Lutz, M. J. KENNISH & A. S. PooLey. 1993. Shell and pallet morphology of early developmental stages of Bankia gouldi (Bartsch, 1908) (Bi- valvia: Teredinidae). The Nautilus 107:63-75. TRUEMAN, E. R. 1964. Adaptive morphology in paleoecological interpretation. Pp. 45—74 in J. Imbrie & N. W. Newell (eds.), Approaches to Paleoecology. John Wiley and Sons, Inc.: New York. TURNER, R. D. 1954. The family Pholadidae in the western At- lantic and the eastern Pacific I. Pholadinae. Johnsonia 3:1— 64. TURNER, R. D. 1969. Superfamily Pholadacea Lamarck. Pp. N702—N742 in R. C. Moore (ed.), Treatise on Invertebrate Paleontology, Part N. Volume 2 (of 3), Mollusca, 6. Bival- via. The Geological Society of America, Inc. and The Uni- versity of Kansas Press: Boulder, Colorado and Lawrence, Kansas. TURNER, R. D. & A. C. JOHNSON. 1971. Biology of marine wood- boring molluscs. Pp. 259-301 in E. B. G. Jones & S. K. Eltringham (eds.), Marine Borers, Fungi and Fouling Organ- isms of Wood. Organisation for Economic Co-operation and Development: Paris. YONGE, C. M. 1963. Rock-boring organisms. Pp. 1—24 in R. E Sognnaes (ed.), Mechanisms of Hard Tissue Destruction. American Association for the Advancement of Science, Pub- lication. The Veliger 42(2):169—174 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 A New Species of Gastropod of the Genus Trophon Montfort, 1810 (Mollusca: Gastropoda: Muricidae) from Subantarctic Waters GUIDO PASTORINO Museo Argentino de Ciencias Naturales, Av. Angel Gallardo 970, 1045 Buenos Aires, Argentina Abstract. Trophon veronicae, a new species of gastropod belonging to the subfamily Trophoninae, is described from deep waters off southern Chile, Argentina, and subantarctic seas. This new species is similar to 7. mucrone Houart from South Brazilian waters. Trophon veronicae sp. nov. can be distinguished from T. mucrone, which is known only from the shell, by its larger size and more slender profile. In addition, the siphonal canal of T. veronicae is very long and curved. The radula and penis of 7. veronicae are described and illustrated with SEM photographs. INTRODUCTION The genus Trophon Montfort, 1810, comprises a group of predatory marine neogastropods that are endemic to South American and Antarctic waters. The genus in- cludes approximately 35 Recent species inhabiting Ant- arctic waters and ranging as far north as Rio de Janeiro, Brazil. Most of these species live in water less than 500 m deep; several range to 1000 m, and very few live deeper. Trophon veronicae sp. nov. is described from bathyal depths of the subantarctic waters off Chile and Argentina. MATERIALS AnD METHODS The holotype and paratypes are from material collected by the United States Antarctic Program on several dif- ferent cruises, plus one additional specimen housed in the collection of the United States National Museum of Nat- ural History (USNM). One paratype has been deposited in the malacological collection of the Departamento de Zoologia Invertebrados, Museo de La Plata, Argentina (MLP-5363). The radulae were prepared according to the method described by Solem (1972) and observed under the scan- ning electron microscope (SEM). The type series contains two specimens with soft parts. These were dissected; the penis was critical point dried, coated with Au-Pd, and photographed under the SEM. Shell ultrastructure data were procured from freshly fractured shell fragments of two specimens. The frag- ments were cut out from the central lip of the last whorl, and also were examined by SEM. SYSTEMATICS Class Gastropoda Cuvier, 1797 Order Neogastropoda Wenz, 1938 Family MurIciDAE Rafinesque, 1815 Subfamily TROPHONINAE Cossmann, 1903 Genus Trophon Montfort, 1810 Type species: Murex magellanicus Gmelin, 1791 (= Tro- phon geversianus (Pallas, 1774)) by original designation. Trophon veronicae Pastorino, sp. nov. (Figures 1—12) Type locality: Eltanin Cruise 25 stat. 325, Blake trawl, off southern Chile 46°00’S, 83°59'W, 742 m, collected on 9 October 1966. Type material: Holotype and 11 paratypes in USNM, 1 paratype in MLP. Material examined: Holotype, USNM 880195, Eltanin Cruise 25 sta. 325, 46°00’S, 83°59'W, Blake trawl, 742 m; 5 paratypes, USNM 880196, Eltanin Cruise 25, sta. 326, 46°04'S, 83°55'W, Blake trawl, collected on 9 Oc- tober 1966, 298 m; 5 paratypes, USNM 870370, Eltanin Cruise 9, sta. 661, 50°32'S, 43°32’W, Menzies trawl, col- lected on 11 August 1963, 1272-1281 m; 1 paratype, USNM 97071, 53°01'S, 73°42'W, 675 m; 1 paratype, MLP 5363, Eltanin Cruise 25 sta. 325, 46°00’S, 83°59'W, Blake trawl, collected on 9 October 1966, 742 m; 2 bro- ken specimens, USNM 880197, Eltanin Cruise 25, sta. 326, 50°32’'S, 43°32'W, Blake trawl, collected on 9 Oc- tober 1966, 298 m. The Veliger, Vol. 42, No. 2 Page 170 G. Pastorino, 1999 Distribution: Known from off southern Chile, the Strait of Magellan, and off South Georgia Islands in 298-1272 m (Figure 13). Etymology: This species is dedicated to Veronica A. Iva- nov. Description: Shell large (to 52 mm), elongate, biconic, fusiform, very slender, chalky; protoconch worn on all but one specimen, of at least two and a half whorls; te- leoconch of seven sharply shouldered whorls, spire very high, about one-third of total shell length. Spire angle about 45°; suture abutting; subsutural ramp steeply in- clined; aperture subovoidal, interior glossy white; anterior siphonal canal very long (same length as aperture), nar- row, curved, open; posterior canal weakly demarcated in some specimens; umbilicus absent; outer lip rounded with reflected edges; inner lip gently curved, adpressed. Axial sculpture of 12—13 regular, weakly lamellose varices on last whorl. Regular growth lines present throughout shell. Spiral ornamentation consisting of three cords, always be- low shoulder, that become obsolete over last whorl; entire shell surface covered by regular, delicate spiral threads. Shell ultrastructure composed of three layers (Figures 5, 6); innermost layer (0.1 shell thickness), composed of collabrally aligned crossed lamellar aragonite, middle lay- er thick (0.45 shell thickness) of crossed lamellar arago- nite with crystal planes oriented perpendicular to growing edge; outer layer thick (0.45 shell thickness) with colla- bral lamellae of crossed lamellar aragonite. Operculum (Figure 11) oval, subpolygonal, with ter- minal nucleus abraded in adult specimens. External sur- face covered by concentric, irregular, often overlapping growth lines. Inner surface with 15—20 regularly rounded and continuous growth lines; very heavily callused, glazed rim present in adult specimens. Animal small relative to shell. Mantle large, mantle roof thin. Cephalic tentacles medium in size, blunt, with rounded large black eyes; mantle edge thickened, smooth, siphon long; pallial organs arranged as in other rachig- lossans; dark osphradium more than half of ctenidium length, thin, slightly asymmetrical, with 75—80 leaflets per side; ctenidium is twice as wide as osphradium, con- taining 140-150 triangular leaflets. Hypobranchial gland brownish and inconspicuous, rectum and large penis to right of hypobranchial gland. Penis large, more than four times length size of ten- Page 171 tacles, wide, flat; papilla conical, flanked by two flaplike extensions of the penis edge. Pleurembolic proboscis short, broad. Radular ribbon small, extending beyond rear of buccal mass. Esophagus loops toward left side, where it receives embedded ducts of salivary glands just anterior to valve of Leiblein. Esophagus joined by brown, well-developed, gland Fram- boisse just posterior to the nerve ring. Large salivary glands envelop retracted proboscis. Accessory salivary glands small, embedded in salivary glands. Gland of Lei- blein conspicuous, overlays esophagus, ends posteriorly in a short blind duct with a small ampulla. Rachiglossan radula (Figures 7, 8, 10) with rachidian teeth very wide (200 wm), central cusp thin, large; lateral cusps half size of central cusp, pointing outward, with inner edge slightly curved; denticle between central and lateral cusp very small, thin, almost obsolete. Base of rachidian tooth strongly curved. Marginal area large, smooth. Lateral teeth with single, long cusps along outer edge of narrow basal plate. Juvenile specimens have proportionally larger inner denticles and thinner rachidian teeth (Figure 10). In ad- dition, the base of the rachidian teeth of juveniles is more curved. Lateral teeth are shorter, thicker. In lateral view both stages have rachidian teeth with a triangular profile. Remarks: The whole complex of the southern South American and Antarctic Trophoninae includes at least 50 described species in addition to several still undescribed. Two groups are clearly differentiated by radular charac- ters. The Patagonian group has complex radulae with the rachidian teeth bearing three cusps, a typical denticle on the inner edge of the lateral cusps, and several denticles along the outer side. Two very developed marginal cusps are also present. In contrast, the Antarctic group has sev- eral types of radulae, but always within the same pattern of a tricuspid, rachidian with an inner, intermediate den- ticle rising from the base of this tooth, and without mar- ginal cusps. Regarding shell characters, the protoconch on most of the Patagonian representatives is in general asymmetrical and paucispiral. The Antarctic species pre- sent a variety of morphologies, including that of 7. ve- ronicae. Compared with the South American group, this new species most closely resembles T. mucrone Houart, 1991, from the Abrolhos Archipelago of southeastern Brazil, but is almost twice the length and width of 7. Figures 1—6 Trophon veronicae Pastorino, sp. nov. 1, a-c holotype USNM 880195 2. Paratype MLP 5363, scale bar = 1 cm; 3. Paratype USNM 880196, protoconch apical view, SEM not coated, scale bar = 200 w. 4. Same specimen side view, SEM not coated, scale bar = 200 pw. 5. Shell ultrastructure, a. innermost aragonitic layer, b. medium aragonitic layer,.c. external aragonitic layer, SEM coated, scale bar = 100 wp. 6. Detail of the layers in Figure 5, scale bar = 2 wp. Page 172 The Veliger, Vol. 42, No. 2 11 2 Figures 7—12 Trophon veronicae Pastorino, sp. nov. 7-8. Radula, side view; scale bar = 80 4; SEM USNM 880195; 9. Penis, scale bar = 300 p. 10. Radula, juvenile specimen, scale bar = 20 jt. 11. Operculum dorsal and ventral view, scale bar = 4 cm. 12. Penis side view, scale bar = 200 w. mucrone (Table 1). Trophon veronicae is also narrower and much more slender than 7. mucrone, although the number of lamellae are similar. The siphonal canal is long and curved in T: veronicae but straight and shorter in J. mucrone. Based on the limited material available, the number of protoconch whorls in 7. veronicae is almost twice that of T. mucrone. In addition, the transition between protoconch and te- leoconch is indistinct in the new species, whereas it is abrupt in 7. mucrone. G. Pastorino, 1999 Page 173 Figure 13 Localities at which Trophon veronicae Pastorino, sp.nov. (cir- cles, white type locality) was collected in subantarctic waters off Chile and Argentina. The radulae and soft parts could not be compared be- cause they remain unknown for T. mucrone. Trophon coulmanensis Smith, 1907 known from Ant- arctic Peninsula and Kerguelen Islands (Dell, 1990) is the only other morphologically similar species in the Antarc- tic. However, differences in size are significant; all the specimens known of T. coulmanensis are not larger than 25 mm. In addition, the protoconch of 7. coulmanensis is asymmetrical and paucispiral, while axial lamellae of the teleoconch usually develop a peripheral spine that never appears in T. veronicae. Harasewych (1984) illustrated the penes of two species belonging to the subfamily Trophoninae: the type species of Trophon, T. geversianus (Pallas, 1774) and Boreotro- phon aculeatus (Watson, 1882). Both species have a dor- soventrally compressed, large penis with a terminal pa- pilla. According to Kool (1993b), T. geversianus also has a vas deferens as an open duct into the mantle cavity, which is very different from the closed duct of T. veron- icae. In a very comprehensive paper about the phylogeny of the Rapaninae, Kool (1993a) described the male re- productive structures of 18 type species of accepted gen- era. Wu (1985) described and illustrated the penis mor- phology of seven species of the genus Acanthina Fischer, 1807. None of the species studied thus far has a penis with lateral folds enveloping the papilla, as here de- scribed. The novel structure of the penis and radula could be indicative of a different generic position. However, the incomplete knowledge of the soft parts of other species of Trophoninae precludes extensive comparisons. ACKNOWLEDGMENTS I express my deep appreciation to M. G. Harasewych (USNM) for all the help, suggestions, and friendship that made this paper possible. The manuscript benefited from reviews by A. Beu and an anonymous reviewer. This work was made during an external scholarship granted by the Consejo Nacional de Investigaciones Cientificas y Técnicas (CONICET), Argentina, to work at the United States National Museum of Natural History, Smithsonian Institution. It was supported in part by a Research Award from the NSF-USAP United States Antarctic program (Contract no. OPP-9509761) and a grant in aid from the Conchologists of America and the Walter E. Sage Me- morial Award. LITERATURE CITED DELL, R. K. 1990. Antarctic mollusca with special reference to the fauna of the Ross sea. Bulletin of the Royal Society of New Zealand, 27:1—311. Table 1 Measurements of 7. veronicae sp. nov. and T. mucrone Houart (in mm). Length T. veronicae sp. nov. USNM 880195 holotype 52.2 MLP 5363 paratype 48.6 USNM 880196 paratype 28.8 USNM 880196 paratype 34.0 USNM 880196 paratype 34.6 Protoconch USNM 880196 1.07 T. mucrone Houart MNHN holotype 26.5 Protoconch 0.86 Lamellae on Width Whorls last whorl 19.6 7 13 19.1 7 13 11.76 V 13 15.0 4 18 15.0 7 13 0.94 2.0—2.5 — 11 7 12 0.71 1.5=1.75 — Page 174 FISCHER DE WALDHEIM, G. 1807. Catalogue systematique et rai- sonné des curiositée de la nature et de l’art. Tome 3. Vé- gétaux et animaux. Museum Démidoff, Moscou. HARASEWYCH, M. G. 1984. Comparative anatomy of four prim- itive muricacean gastropods. Implications for Trophoninae phylogeny. American Malacological Bulletin 1:11—26. Houart, R. 1991. The southeastern Brazilian Muricidae collect- ed by RV Marion-Dusfresne in 1987, with the description of three new species. The Nautilus 105(1):26—37. Kool, S. P. 1993a. Phylogenetic analysis of the Rapaninae (Neo- gastropoda: Muricidae). Malacologia 35(2):155—260. Koo , S. P. 1993b. The systematic position of the genus Nucella (Prosobranchia: Muricidae: Ocenebrinae). The Nautilus 107(2):43-57. MOoNnrTFORT, P. D. DE 1810. Conchyliologie systématique, et clas- sification méthodique des coquilles; offrant leurs figures, The Veliger, Vol. 42, No. 2 leur arrangement generique, leurs descriptions caracteris- tiques, leurs noms; ainsi que leur synonymie en plusieurs langues. Vol. 2. F Schoell: Paris. 676 pp. PaLLas, P. S. 1774. Spicilegia Zoologica quibus novae imprimis et obscurae animalium species iconibus, descriptionibus atque commentariis illustrantur. vol. 1 part 10 Berolini. 41 pp. 4 pls. SoLeM, A. 1972. Malacological application of Scanning Electron Microscopy, I. Radular structure and functioning. The Ve- liger 14(4):327-336. WarTsON, R. B. 1882. Mollusca of H. M. S. ‘Challenger’ Expe- dition.—Part 13. The Journal of The Linnean Society. Zo- ology 16:358-392. Wu, S.-K. 1985. The genus Acanthina (Gastropoda: Muricacea) in West America. Special Publication of the Mukaishima Marine Biological Station Special Contribution, 236:45—66. The Veliger 42(2):175—180 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 Observations on Epithelial Mucocytes in the Sole of Patella Species and Littorina littorea (Linnaeus, 1758) MARK S. DAVIES* Ecology Centre, University of Sunderland, Sunderland, SR1 3SD, United Kingdom Abstract. The sole epithelia of populations of Patella vulgata Linnaeus, P. ulyssiponensis Gmelin, and P. depressa Pennant, were examined by histology. There was no correlation between epithelial mucocyte density and animal size in any population. There was no significant difference in mucocyte density (mean + SE = 87.93 + 0.46 mm!) in P. vulgata populations from different heights on a moderately sheltered shore. Mucocyte density in P. vulgata varied with wave exposure, but not in any consistent pattern. The other patellids examined showed variation in pedal mucocyte density, but it was difficult to identify the causes of this variation. Electron microscopy revealed very similar microvillose epithelial layers in both P. vulgata and Littorina littorea (Linnaeus). Epithelial mucocytes could be seen discharging their contents onto the sole. Goblet cells contained membrane-bound packages (~0.2—1 ym) of mucus (or mucin), similar to those found in terrestrial slugs. INTRODUCTION Histological examination of the gastropod epidermis has attracted much attention (see Simkiss & Wilbur, 1977; Grenon & Walker, 1978; Shirbhate & Cook, 1987 for re- views), as have the functions of the gastropod epidermis, such as respiration (Zaaijer & Wolverkamp, 1958; Jones, 1961), osmoregulation (van Aardt, 1968; Greenaway, 1970), and tenacity (Grenon & Walker, 1981). There has, however, been little attempt to relate such function to the detailed structure of the epidermis. Here I make some observations by microscopy on the structure of the epi- dermis of the foot in intertidal mollusks (Patella species and the periwinkle Littorina littorea [Linnaeus, 1758]) and attempt to relate these observations to the functioning of the mucocytes present. Mucus secretion is a charac- teristic feature of the molluscan epidermis and is impor- tant in a wide range of physiological processes (see Da- vies & Hawkins, 1998). Grenon & Walker (1978) described histologically and biochemically the structure of the foot and pedal glan- dular system of Patella vulgata Linnaeus, 1758, and pro- posed functions for each of nine gland types (P/ to P9) identified. Six of the gland types release their secretions onto the foot sole and three onto the side-wall. The sole epithelium consists of three cell types: non-ciliated cells, ciliated cells, and P9 goblet cells (mucocytes). Of the other pedal gland types, the remainder, except the large anterior pedal gland, are subepithelial and release their contents via necks through the epithelium. The P9 mu- cocytes are randomly distributed throughout the foot, ex- * Telephone UK +191-515-2517; fax UK +191-515-2603; email: mark.davies @ sunderland.ac.uk cept in the peripheral region (Grenon & Walker, 1978) and are by far the most common type of mucocyte present (personal observation). Such epithelial goblet cells are present in many prosobranchs (Fretter & Graham, 1994), and through studies of their secretions (Hunt, 1973; Gren- on & Walker, 1978), their function has been surmised as locomotory (Grenon & Walker, 1978). However, loco- motory mucus is probably secreted mostly by the mar- ginal gland in the anterior marginal groove (Fretter & Graham, 1994; Grenon & Walker, 1978), providing a lay- er of mucus over which the foot can pass. The P9 glands are unlikely to provide a surface lubricant or protective layer as their density is much greater in the sole of the foot than in areas where these functions are more impor- tant, such as the side-wall (Grenon & Walker, 1978). The P9 glands may function in adhesion, and Grenon & Walk- er (1978) suggested some pedal glands secrete a highly viscous mucus for adhesive function while others secrete a less viscous mucus for locomotory purposes. The acid glycosaminoglycan secretion of the P9 glands (Hunt, 1973; Grenon & Walker, 1978) is indicative of high vis- cosity in aqueous solution (Hunt, 1973), and this suggests an adhesive function, although this was not considered so by Grenon & Walker who argued that adhesive was pro- duced by the more acidic mucin of other, subepithelial glands. Denny & Gosline (1980), however, demonstrated plastic viscoelastic properties of a single pedal mucus of Ariolimax columbianus (Gould, 1851), allowing both lo- comotory and adhesive function. MATERIALS anpD METHODS Histology Twenty-five Patella vulgata each were collected from horizontal surfaces at high-, mid-, and low-shore on a Page 176 The Veliger, Vol. 42, No. 2 N Isle of Man | St. Michael's Isiand Bradda 7 Derbyhaven Harbour ( \v Port St. Mary -—4 5 km Figure | Sample sites in the south of the Isle of Man, Irish Sea. moderately sheltered (rated five on Ballantine’s 1961 ex- posure scale) rocky shore at Derbyhaven, Isle of Man (Grid Reference SC 294 685; fig. 1). Further samples of 10 to 15 P. vulgata were collected from horizontal sur- faces each at mid-shore on the semi-exposed shore at Port St. Mary, Isle of Man (Grid Reference SC 208 669; four on Ballantine’s 1961 scale), the very sheltered shore at St. Michael’s Island, Isle of Man (Grid Reference SC 294 674; Ballantine, seven), and the exposed shore at Bradda, Isle of Man (Grid Reference SC 183 699; Ballantine, three) (Figure 1). Fifteen P. ulyssiponensis Gmelin, 1791, were also collected from horizontal surfaces at low-shore at Port St. Mary. An additional five P. ulyssiponensis were collected from shallow rock pools, and 21 P. de- pressa Pennant, 1771, from almost vertical surfaces from mid- to high-shore on an exposed rocky shore at Outer Hope, South Devon (Grid Reference SX 674 403). All limpets were placed in Bouin’s seawater fixative. Voucher specimens have been deposited in the Marine Collection, University of Sunderland (accession numbers MAR-97- 001 to MAR-97-010). On removal from the fixative, the shell length of each animal was recorded and the foot removed. The foot was then cut in half laterally and one half prepared for his- tological examination by dehydration and clearing in xy- lene. Serial transverse sections in wax were made from the cut edge at 8 ym. The second, twenty-fifth, and fif- tieth sections were stained in 1% w/v alcian blue followed by 1% w/v eosin. Sections were examined at 400 under a light microscope and the number of epithelial muco- cytes (P9 goblet cells) were recorded at four fixed stations (determined randomly) across the width of the foot in each section. This procedure was sufficient to give stable means and standard errors of mucocyte density. Electron Microscopy Limpets (Patella vulgata) were collected from mid- shore at Derbyhaven; periwinkles (Littorina littorea) were collected from mid-shore at Port St. Mary harbor. The foot of each animal was cut in half laterally. One half was then fixed in 2.5% v/v glutaraldehyde in sea- water with 0.1 M sodium cacodylate for 1 hr. The tissue was then rinsed for 1 hr in buffered seawater and post- fixed in 1% v/v osmium tetroxide in buffered seawater for 2 hr. After rinsing, the tissue was dehydrated and em- bedded in Spurr’s resin (Spurr, 1969). Sections were cut on a Reichert 4 ultramicrotome and stained for 20 min in 2% w/v uranyl] acetate in 70% v/v alcohol followed by 5 min in 0.3% lead citrate in 0.1 N NaOH. RESULTS anp DISCUSSION Histology An epithelial structure identical to that seen by Grenon & Walker (1978) was observed. There were three distinct cell types: non-ciliated cells, ciliated cells, and mucocytes (Grenon & Walker’s P9 cells). The necks of subepithelial mucus cells (P8 cells), which also discharge their con- tents onto the pedal sole, were also observed within the epithelium. Both types of mucus cells stained with alcian blue, indicating the presence of acid glycosaminoglycan (Grenon & Walker, 1978) indicative of high-viscosity se- cretions (Hunt, 1973). There was no correlation between mucocyte density and shell length for any species measured at any location, indicating that at least for type P9 mucocytes, as the an- imal grows, new mucocytes are produced to maintain a specific density. Mucocyte densities recorded (expressed as mucocytes mm!) are not absolute values with respect to the sole of live Patella. This is because distortion of foot shape may have taken place during histological prep- aration. Nevertheless, mean P9 mucocyte densities re- corded (see below) lie within those recorded by Branch & Marsh (1978) in six species of Patella from South Africa, although Branch & Marsh sectioned at 10 pm rather than 8 wm and counted subepithelial cells. For P. vulgata, differences between mean densities of mucocytes were tested by ANOVA (F = 16.97, P < 0.01) (Figure 2). Subsequent mean separation by SNK test showed no significant difference (at P < 0.05) in mean mucocyte density from high- to low-shore on the mod- erately sheltered shore at Derbyhaven (overall mean = 87.93 mm™! + 0.46 SE) (Figure 2). The relationship be- tween mucocyte density and wave exposure is more com- plex. On the south and east coasts of the Isle of Man there was a significant cline in mucocyte densities from the semi-exposed shore at Port St. Mary (95.51 mm™! + 1.55 SE) through the moderately sheltered shore at Der- byhaven to the very sheltered shore at St. Michael’s Is- land (80.31 mm~! + 1.54 SE). Limpets from the exposed shore at Bradda on the Manx west coast, however, had a mean mucocyte density (80.19 mm™! + 1.86 SE) which was not significantly different to that from St. Michael’s Island. For Patella species from South Africa, limpets secreting more mucus (with a higher density of subepi- thelial mucus glands) have been shown to have lower M. S. Davies, 1999 Page 177 100 ss =~ ak e 90 ** ek E — > eK 2 7) % 80 ® ne) = o 70 je) re) S = > 60 Q 50 an) Syke => eo Sue - © - ® B35 @ el © Gi < ous 1.36 B56 ok = 0 O =- 0O -~ ® O + ® O ® 0 os Ck € oe ¢ op Sn es EnLf n & S29 889 BSG B8S9 ceo Oo % Se wees Wee “wwe 8 ee x 2 wn E O66 ofS oc DD OE we su 2 = ® S Sas Figure 2 Mean P9 pedal mucocyte densities (+SE) from six populations of Patella vulgata on horizontal substrata on the Isle of Man. Means with different numbers of asterisks are significantly different (SNK test, P < 0.05). Exposure grades refer to Ballantine (1961). 100 ; 90 E 2 @ 80 o ae) © 3 70 (Ss) =} = & 60 50 , wy] @ 2 > ‘HI O n| 2 77) S| S S| 2 5|S sls o| © a} Vas ol = ge” ale a0 “lS = 210 =) >I Figure 3 Mean P9 pedal mucocyte densities (+SE) from P. ulyssiponensis and P. depressa populations from Outer Hope (Devon) and Port St. Mary (Isle of Man). tenacities—suggesting mucus acts as a Stefan adhesive— and are found on relatively sheltered shores (Branch & Marsh, 1978), although this relationship did not hold for all species tested. This relationship does not occur for the Manx P. vulgata in terms of P9 cells, except perhaps for those limpets on the exposed shore at Bradda which have the lowest mucocyte density of any P. vulgata population measured in this study. The other sites may not show sufficient variation in exposure for this to be the main factor contributing to mucocyte density, assuming that P9 cells have an adhesive function. Hahn & Denny (1989) have shown that predation of limpets by seabirds is im- portant at some localities and such predation may influ- ence the adhesive capabilities of populations of limpets. Differences in mean mucocyte density also occur in- terspecifically (Figure 3). On the Devon shore P. depres- sa had a higher pedal mucocyte density (77.99 mm7! + 1.52 SE) than did P. ulyssiponensis (62.96 mm~! + 2.34 SE). P. ulyssiponensis on the shore at Port St. Mary had a higher pedal mucocyte density (87.78 mm! + 1.51 SE) than did the conspecifics in Devon. The above means were not tested statistically for significant difference as the populations from which they were derived differ in more than one variable (e.g., species, exposure to wave action, microhabitat). Thus a suggestion of factors pro- ducing these results would be speculative. However, since these limpets are external fertilizers, intraspecific differ- Page 178 The Veliger, Vol. 42, No. 2 M. S. Davies, 1999 Page 179 ences probably arise through post-settlement differential mortality or phenotypic plasticity. Electron Microscopy TEM revealed a foot epithelial layer of P. vulgata con- taining numerous type P9 mucocytes, some of which could be seen discharging their contents onto the epithe- lial surface (Figure 4A). Epithelial mucocytes in L. lit- torea were also visible discharging their contents onto the sole (Figure 4B), and subepithelial mucocytes were also present. For both species the mucus (or mucin) was con- tained within membrane-bound packages of ~0.2—1 wm, similar to those noted by Hunt (1970) and Zylstra (1972), although I am unable to find a reference to this phenom- enon in marine snails. It is well established (e.g., Kapeleta et al., 1996) that the mucus of terrestrial slugs is released in membrane-bound packages, though these are larger, typically 5—10 pm. In slugs, environmental stresses cause the packages to burst and they then absorb water to create a functional mucus. Epithelial gland cells may have an adhesive function and this is consistent with the presence of microvilli on the epithelial surface in both species. The microvilli in- crease foot surface area and suggest an absorptive or re- sorptive function for the epithelial cells. Perhaps adhesion is achieved by active absorption of water or specific mol- ecules from the mucus to increase mucus viscosity (Crisp, 1973), notwithstanding the information supplied by Smith (1991, 1992) that suction is employed by some limpets. Since an increase in adhesion occurs very quickly (e.g., when a limpet is mechanically disturbed) it must be under nervous control. If the mechanism suggested by Crisp (1973) is correct, then a more thorough ultrastructural in- vestigation of the epithelium of intertidal mollusks could reveal these nerve connections. Although some data in this communication support the hypothesis that P9 cells are involved in adhesion, this is by no means confirmed. ACKNOWLEDGMENTS Thanks to M. Haswell and S. Hutchinson for technical assistance and to H. D. Jones, S. J. Hawkins, and J. Ken- naugh for advice. The manuscript was improved by the comments of an anonymous referee. This work was sup- ported by the NERC, United Kingdom. LITERATURE CITED BALLANTINE, W. J. 1961. A biologically defined exposure scale for the comparative description of rocky shores. Field Stud- ies 1:1—-19. BRANCH, G. M. & A. C. MarsH. 1978. Tenacity and shell shape in six Patella species: adaptive features. Journal of Experi- mental Marine Biology and Ecology 34:111—130. Crisp, D. J. 1973. Mechanism of adhesion of fouling organisms. Pp. 691—709 in R. FE Acker (ed), Proceedings of the Third International Congress on Marine Corrosion and Fouling in the Sea. National Bureau of Standards: Gaithersburg, Mar- yland. Davies, M.S. & S. J. HAWKINS. 1998. Mucus from marine mol- luscs. Advances in Marine Biology 34:1-71. DENNY, M. W. & J. M. GOSLINE. 1980. The physical properties of the pedal mucus of the terrestrial slug, Ariolimax col- umbianus. Journal of Experimental Biology 88:375-393. FRETTER, V. & A. GRAHAM. 1994. British Prosobranch Molluscs. The Ray Society: London. xix + 820 pp. GREENAWAY, P. 1970. Sodium regulation in the freshwater mol- lusc Limnaea stagnalis. Journal of Experimental Biology 53: 147-163. GRENON, J.-E & G. WALKER. 1978. The histology and histochem- istry of the pedal glandular system of two limpets: Patella vulgata and Acmaea tessulata (Gastropoda: prosobranchia). Journal of the Marine Biological Association of the United Kingdom 58:803-8 16. GRENON, J.-F & G. WALKER. 1981. The tenacity of the limpet Patella vulgata L.: an experimental approach. Journal of Ex- perimental Marine Biology and Ecology 54:277-308. Haun, T. & M. DENNY. 1989. Tenacity mediated selective pre- dation by oystercatchers on intertidal limpets and its role in maintaining habitat partitioning by ‘Collisella’ scabra and Lottia digitalis. Marine Ecology Progress Series 53:1—10. Hunt, S. 1970. Polysaccharide-Protein Complexes in Inverte- brates. Academic Press: London. 329 pp. Hunt, S. 1973. Fine structure of the secretory epithelium in the hypobranchial gland of the prosobranch gastropod mollusc Buccinum undatum L. Journal of the Marine Biological As- sociation of the United Kingdom 53:59-71. Jones, J. D. 1961. Aspects of respiration in Planorbis corneus and Lymnaea stagnalis (Gastropoda—pulmonata) Compar- ative Biochemistry and Physiology 4:1—29. KAPELETA, M. V., C. JIMENEZ-MALLABRERA, M. J. CARNICER-ROD- RIGUEZ, A. Cook & K. L. SHEPHARD. 1996. Production of mucous granules by the terrestrial slug Arion ater L. Journal of Molluscan Studies 62:251—256. SHIRBHATE, R. & A. Cook. 1987. Pedal and opercular secretory glands of Pomatias, Bithynia and Littorina. Journal of Mol- luscan Studies 53:79—96. SIMKISS, K. & K. M. WiLBur. 1977. The molluscan epidermis Figure 4 Transmission electron micrographs showing the ultrastructure of the pedal sole of (A) Patella vulgata and (B) Littorina littorea. The epithelial layer of columnellar cells lies ventral to the position of the basement membrane. All epithelial cells are microvillose and some are ciliated. Mucus is contained in packages within the vacuoles of epithelial mucocytes (E) (P9 in the limpet) and subepithelial mucocytes (S), and is released (R) onto the sole. Scale bars on (A) = 2 wm; on (B) = 5 um. Page 180 and its secretions. Symposia of the Zoological Society of London 39:35-76. SmitH, A. M. 1991. The role of suction in the adhesion of lim- pets. Journal of Experimental Biology 161:151—169. Situ, A. M. 1992. Alternation between attachment mechanisms by limpets in the field. Journal of Experimental Marine Bi- ology and Ecology 160:205—220. Spurr, A. R. 1969. A low-viscosity epoxy resin embedding me- dium for electron microscopy. Journal of Ultrastructural Re- search 26:31—43. VAN AARDT, W. J. 1968. Quantitative aspects of the water balance The Veliger, Vol. 42, No. 2 in Lymnaea stagnalis (L.). Netherlands Journal of Zoology 18:253-312. ZAAUER, J. J. PR. & H. PB. WOLVERKAMP. 1958. Some experiments of the haemoglobin-oxygen equilibrium in the blood of the ramshorn (Planorbis corneus L.). Acta Physiologica et Phar- macologica Neerlandica 7:56—77. ZYLSTRA, U. 1972. Histochemistry and ultrastructure of the epi- dermis and the subepidermal gland cells of the freshwater snails Lymnaea stagnalis and Biomphalaria pfeifferi. Zeit- schrift fiir Zellforschung und Mikroskopische Anatomie 130:93—124. The Veliger 42(2):181—193 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 NOTES, INFORMATION & NEWS Observations on the Winter Spawning and Larval Development of the Ribbed Limpet Lottia digitalis (Rathke, 1833) in the San Juan Islands, Washington, USA Alan R. Holyoak, Donald J. Brooks and Shawna R. Coblentz Department of Biology, Manchester College, North Manchester, Indiana 46962, USA email: ARHolyoak @ manchester.edu Lottia digitalis (Rathke, 1833) is abundant in rocky in- tertidal communities from Alaska to Baja (Ricketts et al., 1985), but a complete description of its larval develop- ment has yet to be made. Koppen et al. (1996) observed L. digitalis development though the pre-torsional veliger stage, but did not describe later stages of development. In this note we supplement the observations of Koppen et al. (1996) by describing the development of L. digitalis from spawning through the onset of metamorphosis. We collected L. digitalis (n = 265) on 10 January 1997, from False Bay, Cattle Point, Director’s Cove, and Cantilever Pier on San Juan Island, Upper Puget Sound, Washington, USA, and maintained them in a 150 cm X 60 cm running seawater table at the Friday Harbor Lab- oratories. Limpets spawned spontaneously in the seawater table. Eggs were collected via pipette, rinsed, and put into culture in beakers of filtered seawater, one spawn per bea- ker. Most eggs were already fertilized upon collection. We cultured the embryos at 8°C, ambient seawater tem- perature, after the method of Koppen et al. (1996). On 18 January, we added penicillin G (0.06 mg ml~') and streptomycin sulfate (0.05 mg ml~!; after Chia & Koss, 1978) to our cultures in order to control bacterial growth. We used light microscopy to monitor the pattern and tim- ing of larval development twice daily, more frequently during cleavage events. Thirty female L. digitalis spawned in the seawater table between 11 and 17 January 1997. The smallest spawning female was 11.0 mm long and the largest was 20.3 mm long (x = 13.94 mm, SD = 4.25 mm; n = 20 spawning females). The brownish green eggs had a mean diameter of 146.3 pm (SD = 43.8 pm; n = 30 unfertilized eggs). Those eggs were similar in appearance, but were signif- icantly smaller, than those described by Koppen et al. (1996; x = 197.5 pm; SD = 56.6 wm; n = 51 eggs; t = 4.259; v = 79; p < 0.001). We cannot readily explain that size dissimilarity, but we are certain that all limpets in our holding tank were L. digitalis. Perhaps there was a difference in micrometer calibrations between our study and that of Koppen et al. (1996). We also observed two male spawns. Males released pasty white strings of sperm, like those described by Kop- pen, et al. (1996). The spawning males were 13.0 mm and 14.0 mm long, respectively. The proportion of limpets that spawned during our study (32 of 265 limpets—12%) was comparable to that reported by Koppen, et al. (1996; 16 of 140 limpets— 11.4%). The timing and pattern of L. digitalis develop- ment through the pre-torsional veliger stage were similar to those reported by Koppen et al. (1996; see Table 1). Larvae reached the trochophore stage by the 31st hour after spawning. The protoconch was first visible as a shiny spot on the side of trochophore larvae opposite that of the foot rudiment (Figure 1A). Larvae became pre- torsional veligers 2.5 days after spawning (Figure 1B), and completed torsion after 5.3 days (Figure 1C). Once torsion was completed, the velum decreased in size and the foot increased in size. Eye spots appeared on the head 6.5 days after spawning, and tentacles extended through the upper surface of the velum after 8 days. By day 10, larvae crawled rather than swam in culture (Figure 1D). Regression of the velum and crawling were two indica- tors that larvae were undergoing metamorphosis. We were not able to observe the completion of metamorphosis be- cause our scheduled time at the laboratories came to an end. We believe that the L. digitalis we studied would complete metamorphosis within 12—14 days of spawning. Our observations, combined with those of Koppen et Table 1 Timing of larval development for Lottia digitalis, com- pared with data from Koppen et al. (1996). Cumulative Cumulative time time Koppen et al. Developmental stage (hours or days) (1996) Spawning O hr O hr lst cleavage 1-2 hr 1-1.5 hr 2nd cleavage 2—4 hr 2-3 hr 3rd cleavage 3-6 hr 3—4 hr ciliated blastula/gastrula 14 hr 14 hr trochophore 31 hr 24 hr pre-torsional veliger 60 hr 48-72 hr post-torsional veliger 5.3.d = eyespots visible 6.5 d — tentacles visible 7.8d — crawling on the bottom 98d — las ees Figure | Developmental stages of Lottia digitalis. A. Late trochophore stage: a, apical tuft of cilia; tr, trochal band of cilia; f, foot ru- diment; s, shell. B. Pre-torsional veliger: f, foot; v, velum. C. Post-torsional veliger: f, foot; 0, operculum; r, retractor muscles. D. Crawling veliger in mid-metamorphosis: e, eye spot; tn, ten- tacles; r, retractor muscles; 0, operculum. al. (1996), provide the most complete description of lar- val development for this species to date. The two studies also show that the timing and pattern of L. digitalis de- velopment is similar to that of other patellogastropods of the region (Strathmann, 1989), and that a portion of San Juan L. digitalis populations readily spawn in the labo- ratory during the winter. Those characteristics make L. digitalis a good research organism for studying the development and larval ecol- ogy of mollusks with lecithotrophic development, regard- less of time of year. Further studies are needed in order to determine whether San Juan Island L. digitalis spawns in the field during the winter, since L. digitalis reportedly spawns in the field in the San Juan Islands only during spring and summer months (Strathmann, 1987). Literature Cited Cua, ES. & R. Koss. 1978. Development and metamorphosis of the planktotrophic larvae of Rostanga pulchra (Mollusca: Nudibranchia). Marine Biology 46:109-119. Koppen, C. L., J. R. GLAscock & A. R. HOLYOAK. 1996. Spawn- ing and larval development of the ribbed limpet, Lottia dig- italis (Rathke 1833). The Veliger 39:241—243. RICKETTS, E. F, J. CALVIN, J. W. HEDGPETH & D. W. PHILLIPS. 1985. Between Pacific Tides. 5th ed. Stanford University Press: Stanford, California. 652 pp. The Veliger, Vol. 42, No. 2 STRATHMANN, M. FE 1987. Reproduction and Development of Ma- rine Invertebrates of the Northern Pacific Coast. University of Washington Press: Seattle. 670 pp. The Description of a New Species of Favartia (Murexiella) from the South Pacific Ocean Barbara W. Myers! and Carole M. Hertz? Associates, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, California 93105, USA Introduction Dr. Donald R. Shasky of Oceanside, California, made available to us for study some specimens of a muricid he collected offshore at Pointe Taharaa, Papara and Motu Martin, all Tahiti, Society Islands. We have determined them to be an undescribed Favartia (Murexiella) species. There has been considerable difference of opinion in recent years concerning the placement of Favartia Jous- seaume, 1880, and Murexiella Clench & Pérez-Farfante, 1945. Vokes (1968) and Emerson & D’ Attilio (1970) il- lustrated the radula of the type of Murexiella, M. hidalgoi Crosse, 1869, and Ponder (1972) illustrated the radula and operculum of the type of Favartia, F. brevicula Sow- erby, 1834, and he determined that ““Murexiella can be regarded, at best, as being only subgenerically distinct from Favartia.’’ We follow Ponder (1972) in considering Murexiella as a subgenus of Favartia. The following abbreviations for institutions and collec- tions are used in the text: National Museum of Natural History, Smithsonian Institution (USNM); Santa Barbara Museum of Natural History (SBMNH); San Diego Nat- ural History Museum (SDNHM); Shasky Collection (SC); Hertz Collection (HC); Myers Collection (MC). Systematics MuRICIDAE Rafinesque, 1815 MurIcopsINAE Radwin & D’ Attilio, 1971 Genus Favartia Jousseaume, 1880 Subgenus Murexiella Clench & Pérez Farfante, 1945 Favartia (Murexiella) lillouxi Myers & Hertz, sp. nov. (Figures 1—4) Description: Shell small, maximum size 13.8 X 8.4 mm, fusiform, spire elongate. Protoconch with 1% white, bul- ' Mailing address: 3761 Mt. Augustus Ave., San Diego, Cal- ifornia 92111, USA. ? Mailing address: 3883 Mt. Blackburn Ave., San Diego, Cal- ifornia 92111, USA. Notes, Information & News Page 183 bous nuclear whorls somewhat oblique, tip immersed, buttressed on last half whorl. Teleoconch of five whorls, suture moderately impressed, six varices on body whorl, six on penultimate whorl. Leading edge of varix foliose, deeply excavated abaperturally. Aperture ovate, outer lip crenulate reflecting spiral cords, inner lip erect along en- tire length, smooth within, anal sulcus weakly defined. Siphonal canal moderately long, open to right, tubelike, sharply recurved, two-three well-preserved canal termi- nations on siphonal fasciole. Spiral sculpture of two cords on each of first four whorls, body whorl with five strong cords followed by gap and one strong cord on canal, bifid terminally. All cords, webbed between, terminating in Open spines at varices; cords on first and second whorls with two spiral grooves along length, third, fourth and canal cords with one groove. Remnants of appressed scales on cords. Subadult uneroded specimens scaly with fine incised lines covering scales visible under magnifi- cation. Radula and operculum unknown, specimens dead collected. Color ochre to light brown. Type locality: Off Pointe Taharaa, Tahiti, Society Islands (17°45.2’S, 149°30.4’W) in 11—22 meters. Type material: All type material collected within a mile of the type locality (fide D. R. Shasky). Holotype: 12.5 mm X 8.7 mm (SBMNH 144184), off Pointe Taharaa, in 11—22 m, collected from 21—24 October 1996, leg. D. R. Shasky; Paratypes: A, 5.2 x 3.9 mm (USNM 880251), Papara, Tahiti, on coral in 0.6—1.5 m, 16 October 1996, leg. D. R. Shasky; B (broken specimen, spire missing), 7.8 mm width of body whorl (SDNHM 93557); C, 13.8 x 8.4 mm; D, 9.7 X 6.6 mm; E, 8.8 X 6.6 (broken canal); (B-E off Pointe Taharaa, collected in 11—22 m, from 21-— 24 October 1996, leg. D. R. Shasky & P. Lilloux (SC); E 5.2 X 3.3 mm, same data as B—E (HC); G, 3.2 X 2.2 mm, Pointe Taharaa, in 11-22 m, 14—21 October 1996, leg. D. R. Shasky (SC); H, 4.4 X 2.7 mm, same data as G (MC); I, 11.8 mm X 7.3 mm, Pointe Taharaa, in 11-— 22 m, 21-24 October 1996, leg. D. R. Shasky & P. Lil- loux (SC); J, 9.7 X 6.9 mm, in 15 m, Motu Martin, Tahiti, 14 October 1996, leg. D. R. Shasky (SC). Other material studied: Two broken specimens off Motu Martin, Tahiti, in 15 m, 14 October 1996, leg. D. R. Shasky (SC). Distribution: Favartia (Murexiella) lillouxi is known only from Tahiti, Society Islands. Etymology: This species is named in honor of Patrick Lilloux of Mahina, Tahiti, a longtime friend and dive bud- dy of D. R. Shasky, who contributed several of the type specimens. Discussion: This species closely resembles Favartia (Murexiella) rosamiae D’ Attilio & Myers, 1985, but dif- fers in the protoconch, number of varices, and spiral sculpture. The protoconch of F. (M.) lillouxi has 1% bul- bous whorls whereas F. (M.) rosamiae has 3% conical whorls. Favartia (M.) lillouxi has six varices on the body whorl and F. (M.) rosamiae has four. There are five spiral cords on the body whorl and one on the canal in F. (M.) lillouxi and six on the body whorl and two on the canal in F. (M.) rosamiae. Favartia (M.) lillouxi is also similar to F. (M.) voor- windi Ponder, 1972, a species having a shell with a broad- er shoulder and shorter spire. Favartia (M.) lillouxi, with its higher spire, is light brown with straight spines where- as F. (M.) voorwindi has a white shell with recurved spines. Favartia (M.) lillouxi does not closely resemble any other western Pacific species, and our examination of worldwide Favartia species revealed no close congeners. Acknowledgments The San Diego Natural History Museum made its collec- tions in the Scientific Library and Marine Invertebrate Department available to us. David K. Mulliner of San Diego, California, did the photography and Emily H. Vokes of Tulane University, Louisiana, reviewed the manuscript. For this we thank them. We express our ap- preciation to Donald R. Shasky for giving us the oppor- tunity to describe this new Favartia species and for the donation of type material. Literature Cited CLENCH, W. J. & I. J. PEREZ-FARFANTE, 1945. The genus Murex in the Western Atlantic. Johnsonia 1(17):1—58. Crosse, H., 1869. Diagnoses molluscorum novorum, Journal de Conchyliologie 17:408—410. D’Attitio, A. & B. W. Myers, 1985. Two new species of Fa- Explanation of Figures 1 to 4 Figures 1, 2. Favartia (Murexiella) lillouxi Myers & Hertz, sp. nov. Holotype (SBMNH 144184), 12.5 x 8.7 mm. Off Pointe Taharaa, Tahiti, Society Islands, in 11-22 m. Leg. D. R. Shasky, 21—24 October 1996. (1) apertural view (2) dorsal view. Figures 3, 4. Favartia (Murexiella) lillouxi Myers & Hertz, sp. nov. Paratype A, 5.2 X 3.9 mm (USNM 880251). Papara, Tahiti, Society Islands, in 0.6—-1.5 m on coral. Leg. D. R. Shasky, 16 October 1996. (3) apertural view (4) dorsal view. This specimen illustrates the protoconch and the foliaceous nature of the sculpture. Page 184 The Veliger, Vol. 42, No. 2 Notes, Information & News Page 185 vartia from the West Pacific Ocean (Gastropoda: Muricidae). The Nautilus 99(2—3):58—61. EMERSON, W. K. & A. D’ArtTILIO, 1970. Three new species of muricacean gastropods from the Eastern Pacific. The Veliger 12(3):270-274. JOUSSEAUME, F, 1880. Division méthodique de la Famille des Purpuridés. Le Naturaliste 42:335—356. PONDER, W., 1972. Notes on some Australian genera and species of the family Muricidae (Neogastropoda). Journal of the Malacological Society of Australia 2(3):215—248. Sowersy, G. B. (second of name), 1834. Conchological Ilustra- tions (Murex). London, pls. 58—67. VoKES, E. H., 1968. Cenozoic Muricidae of the western Atlantic region Part 1V—Hexaplex and Murexiella. Tulane Studies in Geology 6(3):85—126. High Performance Thin Layer Chromatography Determination of Carbohydrates in the Hemolymph and Digestive Gland of Lymnaea elodes (Gastropoda: Lymnaeidae) Daniel J. Cline*, Bernard Fried** and Joseph Sherma* Departments of *Chemistry and **Biology, Lafayette College, Easton, Pennsylvania 18042, USA Introduction Recent studies in our laboratory have been concerned with high performance thin layer chromatography (HPTLC) analyses of carbohydrates in the hemolymph and digestive gland-gonad (DGG) complex of medically important planorbid snails. Thus, Anderton et al. (1993) reported quantitative values for glucose and trehalose in Biomphalaria glabrata (Say, 1818) snails maintained on various diets, 1.e., leaf lettuce, Tetramin®, and hen’s egg yolk. Perez et al. (1994) studied the effects of larval trem- atode parasitism by Echinostoma caproni Richard; 1964, on sugars in B. glabrata and found that parasitism sig- nificantly reduced the amounts of glucose and trehalose in the hemolymph and DGGs of infected snails. Conaway et al. (1995) provided quantitative data on glucose and trehalose in several strains of Helisoma trivolvis (Say, 1816) snails with and without infection by larval echi- nostomes. Umesh et al. (1996) used HPTLC to analyze the effects of restricted diets on glucose in the hemo- lymph and DGG of B. glabrata and H. trivolvis. Their results showed that glucose levels were not significantly altered in snails maintained on the restricted diets de- scribed in that study. Less information is available on the quantitative anal- ysis of carbohydrates in lymnaeid than in planorbid snails. Lymnaeid snails play a less important role in med- ical malacology than do the planorbids, and this accounts in part for the relative paucity of quantitative biochemical data on the effects of diet and larval trematode parasitism on lymnaeids. A recent study on infection of Lymnaea elodes (Say, 1821) in the USA with a species of Echi- nostoma revolutum (Froelich, 1802) that causes intestinal helminthiasis in wildlife and is a potential foodborne pathogen to man has been reported by Sorensen et al. (1997). Because of that report, there is renewed interest in examining various aspects of the biology and chem- istry of this snail. Moreover, L. elodes is easy to maintain in the laboratory, attains a length of up to 3 cm within 3 months, and is a convenient experimental model for bio- chemical studies. A previous study on this model used HPTLC to determine neutral lipids and phospholipids in whole snail bodies (Frazer et al., 1997). The purpose of the present study was to examine by HPTLC the identity and concentrations of carbohydrates in the hemolymph and digestive gland of L. elodes maintained on a leaf lettuce diet. Materials and Methods Sugar standards were purchased from Sigma (St. Louis, Missouri, USA). Standard solutions of each sugar were prepared at concentrations of 100 ng/wL (standard solu- tion A) and 1.00 pg/L (standard solution B) in 70% ethanol. Stock cultures of L. elodes snails were maintained at 22°C in aerated aquaria containing artificial spring water (ASW) as described in Frazer et al. (1997). Snails were fed ad libitum on boiled leaf lettuce. Most snails were used for analyses immediately after removal from the cul- tures. Some snails were maintained in ASW without food for either 4 or 12 hr prior to use for analyses (referred to below as 4-hr starved or 12-hr starved snails). Analyses were done on pooled samples of hemolymph and digestive glands (DGs) (three snails/pool) from snails ranging between 24 and 30 mm in shell length. For he- molymph analysis, snails were blotted dry with paper towels, gently crushed, and the hemolymph from three snails collected in a 1.5 mL microcentrifuge tube. The sample was centrifuged for 3 min at 8000 g to separate the plasma from the hemocytes. One hundred wL of plas- ma, measured with a 100 pL Drummond (Broomall, Pennsylvania, USA) digital microdispenser, was separat- ed from the amoebocyte pellet and placed in a new mi- crocentrifuge tube with 500 wL of 70% ethanol. The sam- ple was centrifuged for 5 min at 8000 g. The supernatant was combined with two washings of the pellet (100 pL of 70% ethanol for each washing) in a 2 mL vial. The sample was evaporated to dryness in a water bath (S0- 60°C) under a gentle stream of air and then reconstituted in 200 pL of 70% ethanol. Snail DGs were separated from the bodies with for- ceps, taking care to remove and discard the gonads and digestive tract. The wet weight of the three DGs in each pool was determined (about 150 mg) before the tissues were homogenized with 500 wL of 70% ethanol in a 7 Page 186 mL glass homogenizer. The homogenate was quantita- tively transferred to a microcentrifuge tube and centri- fuged at 8000 g for 5 min. The supernatant was removed to a 5 mL vial and combined with two washings of the pellet (100 wL of 70% ethanol for each washing). The sample was evaporated to dryness as described for the hemolymph, but reconstituted in 400 pL of 70% ethanol. Boiled leaf lettuce (Lactuca sativa) samples (150 mg) were prepared as described for the DGs. Qualitative TLC analysis was performed on Merck (Gibbstown, New Jersey, USA) 20 10 cm HPTLC sil- ica gel 60 CF,;, plates with preadsorbent zone and 19 channels (catalog no. 13 153). The layer was precleaned by development with dichloromethane-methanol (1:1), air dried in a fumehood, and impregnated with sodium bi- sulfite and citrate buffer as described earlier (Conaway et al., 1995). Standards of each sugar (2.0 pL of standard solution A) and samples (5.0 and 7.0 wL of reconstituted hemolymph and 2.0 and 6.0 wL of DG) were applied onto the preadsorbent areas of adjacent lanes using a 10 pL Drummond digital microdispenser. Plates were developed three times in the ascending direction for a distance of 7 cm beyond the preadsorbent-silica gel interface in a pa- per-lined Camag (Wilmington, North Carolina, USA) twin-trough chamber with acetonitrile-deionized water (85:15) (mobile phase 1) or ethyl acetate-acetic acid- methanol-water (60:15:15:10) (mobile phase 2). The sol- vent was removed between developments by drying for 2 min with a hair dryer. Sugar zones were detected with 1-naphthol-sulfuric acid reagent as previously described (Anderton et al., 1993). Sugars were identified by comparing R; values (Umesh et al., 1996) between standard and sample zones in both mobile phases. Identities were confirmed by gas chro- matography/mass spectrometry (GC/MS) in the scanning (total ion current) mode. Standards and samples were si- lylated with Tri-Sil® Z (Sigma, St. Louis, Missouri, USA) and chromatographed on a 30 m DB-5MS capillary col- umn (0.25 mm i.d., 0.25 ym film thickness of (5%-phe- nyl)-methylpolysiloxane stationary phase) (J&W Scien- tific, catalog no. 122-5532, Folson, California, USA) in an HP6890 gas chromatograph (Hewlett-Packard Com- pany, Palo Alto, California, USA). Standard and sample solution injection volumes were 1.0 and 2.0 wL, respec- tively, helium carrier gas flow rate was 53 mL/min at 7.5 psi, the column temperature was programmed from 50°C to 280°C at 10°C/min, and the run time was 44 min. Glucose and maltose were quantified on Whatman (Clifton, New Jersey, USA) 20 XK 20 cm TLC LKS5DF silica gel plates with preadsorbent zone and 19 channels (catalog no. 4856-821). The layers were precleaned by development with dichloromethane-methanol (1:1) but were not impregnated with bisulfite and citrate buffer. For each sugar, 6.0 and 8.0 wL of standard solution A and 1.2 and 2.4 wL of standard solution B were applied to the same plate with the sample volumes specified above for The Veliger, Vol. 42, No. 2 qualitative analysis. Plates were developed once for a dis- tance of 18 cm with mobile phase 2. After detection, sam- ple and standard zones were scanned at 515 mm using a Camag TLC Scanner II with a tungsten source, slit di- mension settings of length 4 and width 4, and a scanning rate of 4.0 mm/sec. The maximum absorption wavelength was determined by measuring the in situ spectra of stan- dard sugar zones with the spectral mode of the densitom- eter. The CATS-3 software linear regression program pro- vided a calibration curve relating standard zone weights (0.600—2.40 wg) to their computer-optimized scan areas. The analyte weights in sample zones were determined by the computer from their areas by interpolation from the calibration curve. Only calibration curves with a linear regression correlation coefficient (R) value of at least 0.93 were used for quantitative analysis; R values were typi- cally 0.96—0.99. Sugar concentrations (mg/dL for hemo- lymph and weight percent for DG) were calculated as described by Anderton et al. (1993). For five out of the six 12-hr starved hemolymph sam- ples analyzed, the scan area of maltose in the largest spot- ted aliquot (15 wL) was less than the lowest standard area. The quantity of this sugar was between zero and the ex- perimental detection limit (2.67 mg/dL), but could not be determined since the area of the sample band was not bracketed by the standard areas. These bands were arbi- trarily given a value of 1.33 mg/dL in order to include the data in statistical calculations. This value is one-half the experimental detection limit, calculated from the equation lowest quantifiable hemolymph sugar concentration (mg/dL) = (L X R)/(H X V) X 1/10, where L is the weight of the sugar in the lowest standard aliquot (200 ng), R is the sample reconstitution volume (200 pL), H is the highest sample aliquot spotted (15.0 wL), and V is the original volume of sample (100 pL). Results Sugars were detected as gray or blue bands on a beige background with the naphthol-sulfuric acid reagent. R; values in the two mobile phases used for qualitative anal- ysis of sample bands were reported by Umesh et al. (1996). The glucose standard was found to comigrate with a band in all sample chromatograms (R; = 0.41 in mobile phase | and 0.70 in mobile phase 2). The maltose standard comigrated with bands in chromatograms of all DG samples, fresh fed hemolymph, and 4-hr starved he- molymph sample chromatograms, but not in the 12-hr starved hemolymph or lettuce sample (R; = 0.22 in mo- bile phase 1 and 0.54 in mobile phase 2). A maltose band was not observed in the lettuce or 12-hr starved sample chromatograms above the detection limit of the analysis, 2.67 mg/dL. The fructose standard comigrated with a Notes, Information & News Page 187 Table 1 Quantitative determinations of glucose and maltose in snail digestive gland (DG) and hemolymph samples. Hemolymph mean + SE mg/dL Snail diet Glucose Maltose Fresh fed 33.8 + 4.9 36.1 + 6.0 4-hr starved Als),S) ae 7/24! 40.3 + 14 12-hr starved 43.4 + 9.7 5.00 + 4.6 DG mean + SE weight % Glucose Maltose 0.0065 + 0.0056 0.13 + 0.016 0.084 + 0.012 0.21 + 0.033 0.084 + 0.013 0.26 + 0.017 n = 3-6 for each group. band in the lettuce sample chromatogram at R; = 0.45 in mobile phase | and 0.70 in mobile phase 2. There were two intense bands in DG sample chro- matograms that did not comigrate with any of the 13 sug- ar standards tested (listed in Umesh et al., 1996) (Ry = 0.08 and 0.06 in mobile phase 1 and 0.43 and 0.24 in mobile phase 2). There was also an unidentified band in hemolymph sample chromatograms with an R,; value less than any of the sugars we tested. GC/MS confirmed the identity of glucose and maltose (retention times 17.85 and 25.10 min, respectively) as the major sugars in the hemolymph and DG samples. Other compounds were detected in sample chromatograms, but their mass spectra did not match those of any sugar stan- dards. There was no peak detected at 25.30 min, the re- tention time of trehalose, confirming the absence of this sugar in the samples. GC/MS could not identify the com- pounds present in the unknown TLC bands. Whatman LKSDF plates were more convenient to use for quantitative TLC. The sugar bands were adequately separated with only one development of mobile phase 2, and no pretreatment with sulfite and citrate buffer was necessary. Although the bands were slightly more diffuse compared to the Merck HPTLC plates, they were ade- quately compact and resolved to allow accurate quanti- fication. Additionally, larger volumes of sample could be applied to the preadsorbent zones of the Whatman plates than the Merck plates. This would be important if it was necessary to apply larger volumes in order to have sample band scan areas bracketed by the standard band areas. The R, values of glucose and maltose on LKSDF plates with mobile phase 2 were 0.48 and 0.32, respectively, while the two unidentified bands had R; values of 0.21 and 0.10. Quantitative results for glucose and maltose are listed in Table 1. The following populations were compared us- ing Student’s t-test: fresh fed vs. 4-hr starved, 4-hr starved vs. 12-hr starved, and fresh fed vs. 12-hr starved snails. It was found that there were no statistical differ- ences in hemolymph and DG glucose levels as a result of starvation. However, hemolymph maltose levels were significantly lower between fresh fed and 12-hr starved populations (t = 3.96, P = 0.017). There was also a sig- nificant increase in DG maltose levels between fresh fed and 12-hr starved samples (t = 5.56, P = 0.0009). Glucose and sucrose were quantified in samples of boiled lettuce, and respective percentages of 0.084 + 0.039 and 0.014 + 0.0053 were found (n = 2). Merck HPTLC plates with triple development in mobile phase 1 rather than Whatman TLC plates with a single develop- ment in mobile phase 2 were used for these analyses be- cause mobile phase 2 did not adequately separate glucose and sucrose bands, and sugar bands do not ascend more than half the plate on Whatman TLC plates using mobile phase 1. Sucrose standards A and B were prepared as described above for glucose, and 2.0, 6.0, and 8.0 pL of B and 1.2 and 2.4 pL of A (0.200—2.40 wg) were applied to the plate along with 2.0 and 10.0 pL of reconstituted lettuce sample solution for quantification. Discussion Qualitative and quantitative TLC and GC/MS have shown that glucose and maltose are the major sugars present in L. elodes. These results are similar to those of Umesh et al. (1996) in their analyses of sugars in B. glabrata. The presence of trehalose in L. elodes, as reported in earlier publications from our laboratory in studies on planorbid snails, i.e., Anderton et al. (1993), Perez et al. (1994), Conaway et al. (1995), was not confirmed by TLC or GC/ MS in the present study. Quantitative results indicated a statistical difference in maltose levels of L. elodes that were maintained without food for 12 hr. The loss of maltose from the hemolymph and subsequent increase of this sugar in the DG, as well as the increased weight percent maltose compared to glu- cose in the DG, suggest that maltose is an important car- bohydrate for metabolism in this snail. Regardless of the period of starvation in our study up to 12 hr there were no statistically significant changes in glucose concentrations. Veldhuijzen et al. (1976) found a sharp decline in body tissue glucose levels as a result of starvation for several days of Lymnaea stagnalis (Say, 1821), but glucose levels in the hemolymph remained Page 188 The Veliger, Vol. 42, No. 2 constant. Because their experiment tested starvation over several days, and ours examined starvation only up to 12 hr, direct comparisons of the two studies are not possible. The results of the lettuce analysis suggest that L. elodes may obtain some sugars, e.g., glucose, directly from food. Fructose and sucrose were present in lettuce but not in the hemolymph or DG of the snail; maltose, abundantly present in the snail, was not found in lettuce. These re- sults suggest that this snail can convert carbohydrates found in lettuce into the sugars they need for their met- abolic processes. Acknowledgments The authors thank Dr. Ronald B. Keefe of National Starch and Chemical Company (Bridgewater, New Jersey, USA) for performing the GC/MS analyses. Daniel J. Cline was supported by grants from Research Corporation and the Lafayette College Academic Research Committee (EX- CEL Scholar Program). Literature Cited ANDERTON, C. A., B. FRIED & J. SHERMA. 1993. HPTLC deter- mination of sugars in the hemolymph and digestive gland- gonad complex of Biomphalaria glabrata snails. Journal of Planar Chromatography-Modern TLC 6:51—54. Conaway, C. A., B. FRIED & J. SHERMA. 1995. High performance thin layer chromatographic analysis of sugars in Helisoma trivolvis (Pennsylvania strain) infected with larval Echino- stoma trivolvis and in uninfected H. trivolvis (Pennsylvania and Colorado strains). Journal of Planar Chromatography- Modern TLC 8:184—187. FRAZER, B. A., A. REDDY, B. FRIED & J. SHERMA. 1997. HPTLC Determination of neutral lipids and phospholipids in Lym- naea elodes (Gastropoda). Journal of Planar Chromatogra- phy-Modern TLC 10:128—130. PEREZ, M. K., B. FRIED & J. SHERMA. 1994. High Performance thin-layer chromatographic analysis of sugars in Biomphal- aria glabrata (Gastropoda) infected with Echinostoma cap- roni (Trematoda). Journal of Parasitology 80:336—338. SORENSEN, R. E., I. KANEv, B. FRIED & D. J. MINCHELLA. 1997. The occurrence and identification of Echinostoma revolutum from North American Lymnaea elodes. Journal of Parasitol- ogy 83:169-170. Umesh, A., B. FRIED & J. SHERMA. 1996. Analysis of sugars in the hemolymph and digestive gland-gonad complex (DGG) of Biomphalaria glabrata and Helisoma trivolvis (Colorado and Pennsylvania strains) maintained on restricted diets. The Veliger 39:354-361. VELDHUUZEN J. P. & G. VAN BEEK. 1976. The influence of star- vation and of increased carbohydrate intake on the polysac- charide content of various body parts of the pond snail Lym- naea stagnalis. Netherlands Journal of Zoology 23:106—118. On the Egg Capsules of Epitonium georgettinum (Kiener, 1839) (Gastropoda: Epitoniidae) from Patagonian Shallow Waters Guido Pastorino Museo Argentino de Ciencias Naturales, Av. Angel Gallardo 970, 1045 Buenos Aires, Argentina and Pablo Penchaszadeh INTECMAR, Universidad Simon Bolivar, 1080 Caracas, Venezuela and Museo Arg. Cs. Nat., ““B. Rivadavia,”’ Angel Gallardo 470, 1405 Buenos Aires, Argentina Introduction The family Epitoniidae is represented in the Patagonian littoral by three known species: Epitonium magellanicum (Philippi, 1845); E. tenuistriatum (d’Orbigny, 1839); and E. georgettinum (Kiener, 1839). The record of E. albidum (d’Orbigny, 1842) by Clench & Turner (1951:262) is questionable according to Robertson (1983b). Epitonium georgettinum (Kiener) is the only species of Epitoniidae spanning both the Argentinean and Magellanic malaco- logical provinces, from Rio Grande do Sul in Brazil to Puerto Madryn in Argentina (Rios, 1985). All the species of the family appear to be associated with coelenterates (Robertson, 1983a). Epitonium geor- gettinum is found around the bases of sea anemones that live in intertidal pools and in the adjacent sandy bottom. The purpose of this note is to describe and illustrate egg masses, capsules, and eggs of Epitonium georgettin- um. In addition, the life history of the species is compared with that of other species of the genus Epitonium. Materials and Methods Adult and juvenile specimens and egg masses of E. geor- gettinum were collected from the Patagonian locality of Puerto Pirdmide (42°34'S, 64°17'W) in November, 1995 in the Province of Chubut, Argentina (Figure 1). The specimens and egg masses were collected on the sandy bottom with the string of eggs attached to the animals. However, some small specimens were also collected near (less than 4 cm) the pedal disc of sea anemones. The sea anemones (genus Bunodactis ?) live on a limestone sub- strate in the intertidal and infralittoral zones. The egg masses were housed in the egg collection of the INTECMAR, Universidad Simon Bolivar, Caracas, Venezuela. Results and Discussion Egg masses, egg capsules and eggs: The egg masses were found near the adults; in two cases, the female was still spawning, thus carrying the eggs by means of the string. They are composed of 120 to 238 egg capsules Notes, Information & News Page 189 T 64° W a Golfo San José Peninsula Valdés P. Piramide Golfo Nuevo 10 Figure | Index map showing the locality where the egg capsules and adults of Epitonium georgettinum (Kiener, 1839) were collected. attached to each other by a tough elastic string (Figure 2). The largest egg capsules recorded measured 1.9 X 1.5 mm. The egg capsules are pyramidal or polyhedral in shape and covered with sand grains. The number of eggs per capsule ranged from 122 to 211 (n = 20; x = 152.55 SD = 21.69). The uncleaved egg is 73—78 wm in diameter (n = 13 k = 75.15 SD = 1.46). The diameter of the egg is constant, and seems to be independent of the number of eggs per capsule. The life history of E. georgettinum is apparently sim- ilar to that of other species of the genus despite differ- ences in the adult size. Some of the live specimens col- lected reach up to 31 mm in length, which is 10 mm larger than the other Atlantic species of which the life histories are known. The diameter of the uncleaved egg is 73-78 pm. It is 68 ym for E. albidum (d Orbigny, 1842) 73 wm for E. millecostatum (Pease, 1860), and 78 wm for E. ulu Pilsbry, 1921 (see references in Table 1). Larval lifespan and size in E. ulu (26 days and 0.39 wm according to Bell, 1985) suggests that E. georgettinum spends longer than this as a planktotrophic larva. Acknowledgments Two anonymous reviewers and the editor helped to im- prove this paper. This work was made possible by a scholarship granted by the Consejo Nacional de Investi- gaciones Cientificas y Técnicas (CONICET), Argentina to G. P. and a grant from Fundacion Antorchas, Argentina to P. P. Figure 2 Egg masses of Epitonium georgettinum (Kiener, 1839) from Puerto Piramide, Chubut, Argentina, A. Four egg capsules in life position. B. One broken capsule with the eggs inside. Page 190 The Veliger, Vol. 42, No. 2 Table 1 Measurements of number and size of egg capsule and egg in relation with the adult size of Epitonium species (in mm unless otherwise indicated). Egg capsule size length x width Adult (or length Egg diameter Capsules/egg size Epitonium species only) (wm) mass Eggs/capsule (length) Source E. millecostatum 1.4 X 0.87 aS 90 149-185 OF, Robertson, 1981 (Pease, 1860) x = 165 (m= 3) E. equinaticosta 0.9-1.2 98-106 2-11 28-65 9.5 Robertson, 1983a (d’Orbigny, 1842) x = 102 (n = 10) x = 43 (n = 5) E. albidum (d’Orbigny, 1842) 2.2 X 0.7 68 2,300 248 8-15 Robertson, 1983b E. rupicola (Kurtz, 1860) 2.5 ? 125 400 19.3 McDennott, 1981 E. ulu Pilsbry, 1921 1.76 X 1.32 78 ? 558 13 Bell, 1985 E. georgettinum (Kiener, 1839) LSexX EO 73-78 120-238 122-211 31 this paper k= KX = 152.5 (n = 20) 7.51 (n = 13) 152°5 Literature Cited BELL, J. L. 1985. Larval growth and metamorphosis of a proso- branch gastropod associated with a solitary coral. Proceedings of the Fifth International Coral Reef Congress, 5:159—163. CLENCH, W. & R. TURNER. 1951. The genus Epitonium in the western atlantic part 1. Johnsonia 2(30):249—288. McDermott, J. J. 1981. On the reproduction of Epitonium rupicola Kurtz (Gastropoda: Epitoniidae). The Veliger 24:67—71. Rios, E. DE C. 1985. Seashells of Brazil. Editora da Fundacgao Universidade do Rio Grande. Rio Grande. 328 pp. ROBERTSON, R. 1981. Epitonium millecostatum and Coralliophil- la clathrata: two prosobranch gastropods symbiotic with Indo-Pacific Palythoa Coelenterata: Zoanthidae. Pacific Sci- ence 34:1—17. ROBERTSON, R. 1983a. Observations on the life history of the wentletrap Epitonium echinaticostum in the Bahamas. The Nautilus 97:98—-103. ROBERTSON, R. 1983b. Observations on the life history of the wentletrap Epitonium albidum in the West Indies. American Malacological Bulletin 1:1—12. International Commission on Zoological Nomenclature The following Application was published on 18 Decem- ber 1998 in Volume 55, Part 4 of the Bulletin of Zoolog- ical Nomenclature. Comment or advice on this applica- tion is invited for publication in the Bulletin and should be sent to the Executive Secretary, I.C.Z.N., % The Nat- ural History Museum, Cromwell Road, London SW7 SBD, U.K. (e-mail: iczn@nhm.ac.uk). Case 3036—Haliotis clathrata Reeve, 1846 (non Lich- tenstein, 1794) and H. elegans Philippi, 1844 (Mollus- ca, Gastropoda): proposed conservation of the specific names. Description of a New Species of the Genus Phidiana Gray, 1850 (Nudibranchia: Facelinidae) from Pacific Ocean Waters of Panama Francisco J. Garcia Laboratorio de Biologia Marina; Depto. Fisiologia y Biologia Animal; Facultad Biologia; Avda. Reina Mercedes, 6; Apdo. 1095, 41080 Sevilla, Spain. e-mail: fjgarcia@cica.es and Jesus S. Troncoso Area de Biologia Animal; Facultad Ciencias del Mar; Univ. Vigo; Lagoas-Marcosende; Vigo; Spain. e-mail: troncoso @uvigo.es Introduction In American Pacific waters six species of Phidiana Gray, 1850, have been found (Lance, 1962; Bertsch & Ferreira, 1974; Farmer, 1980), although none were reported from the Pacific coast of Panama. During a scientific expedi- tion around some islands belonging to the National Park of Coiba Island, several specimens of a species of Phi- diana were collected. The color pattern and anatomical features of these specimens allow us to propose a new Phidiana species. In this article, the anatomy of the spec- imens is described and compared with that of other spe- cies of the genus. Materials and Methods The specimens studied in this article were collected during a scientific expedition in February 1997, around the islands belonging to the National Park of Coiba (Pan- ama), located in the Pacific Ocean. The sample locations Notes, Information & News Page 191 Coiba l. \«Jicarita |. Figure 1 Map of the Coiba islands showing the locations where the spec- imens were collected (*). where the specimens were collected are illustrated in Figure 1. The specimens were fixed and preserved in 4% formaldehyde. Systematic Description Phidiana mariadelmarae Garcia & Troncoso, sp. nov. (Figures 2—5) Type material: Holotype, length 16 mm, collected under rocks in the intertidal zone of Coiba Island, Panama (7 Figure 2 Phidiana mariadelmarae Garcia & Troncoso, sp. nov. External view. The holotype. February 1997) deposited in Museo Nacional de Ciencias Naturales de Madrid (Spain) with the code number 15.05/ 27853. Paratype, one specimen of 17 mm in length, col- lected under rocks in the intertidal zone of Jicarita Island, Panama (9 February 1997) and two specimens (9 and 17 mm in length), collected at the same station (14 February 1997), deposited in the same museum with the code num- bers 15.05/27854 and 15.05/30340. Additional material examined: Two specimens, 9 and Figure 3 Phidiana mariadelmarae Garcia & Troncoso, sp. nov. A. Jaws. B. Detail of the masticatory border. Specimens 9 and 13 mm. Page 192 The Veliger, Vol. 42, No. 2 Figure 4 Phidiana mariadelmarae Garcia & Troncoso, sp. nov. Radular tooth. Specimens 9 and 13 mm. 13 mm in length, collected under rocks in the intertidal zone of Jicarita Island, Panama (9 and 14 February 1997) were dissected. Etymology: The name of this species, mariadelmarae, is derived from the name of the wife and eldest daughter of the first author, Maria del Mar. External anatomy (Figure 2): The body is elongated, length was 9, 11, 13, 16 and two of 17 mm, oral tentacles long and cylindrical, rhinophores slightly shorter, and the lower one-third smooth, and two-thirds of the top with 12 annulations. The cerata are two groups of oblique rows, anterior group seven rows, posterior eight rows. Each row has two to seven cerata, the outer shorter than the inner. The genital papilla is on the right side of the body, between the fifth and sixth row of the anterior cerata group. The tail and foot are rounded, propodial tentacles absent. Coloration. The ground color is orange, slightly darker in the cephalic region. The esophagus, jaws, and subra- dular membrane are rose colored, as seen through the body wall. The apex of the oral tentacles and apical third of the rhinophores are white. The eyes are clearly visible behind the base of the rhinophores. A fine white longi- tudinal line extends mid-dorsally from the anterior end of the cardiac region to the level of the rhinophoral base, where it bifurcates. Each branch extends along one side of the head to the apex of the oral tentacle. These lines vary in strength and breakage, depending on the speci- men. Thus, in one specimen of 17 mm length, in which the orange coloration of the body is darker than in the other specimens, the white lines are almost absent. The cerata are hyaline orange. Their orange-brown di- gestive gland branches are visible through the translucent tissue. The cnidosac color is lighter than the rest of the cerata. On the outer surface of the cerata, located in the cardiac region, there is a white band whose extension on each cerata varies according to specimen. The number of cerata with bands also varies, the most numerous being the specimen that lacks white dorsal lines. Imm Figure 5 Phidiana mariadelmarae Garcia & Troncoso, sp. nov. A. Repro- ductive system. B. Detail of the penis. Abbreviations: a, ampulla; ag, albumen gland; dd, deferent duct; hd, hermaphroditic duct; mg, mucus gland; p, penis; sr, seminal receptacle; v, vagina. Specimens 9 and 13 mm. Internal anatomy: The jaws (Figure 3) are rose colored, ovate, and strongly convex on the outer surface. The cutting border has a single row with 22 hooked denticles in a spe- cimen of 11 mm in length. The masticatory border of a specimen of 13 mm only has 15 rounded and worn denti- cles. The subradular membrane and esophagus are rose colored. The radular formula of a 13 mm long specimen is 20 X 0.1.0. The teeth have a large median cusp with three to four hooked denticles on either side, and five hooked lateral cusps. Outer and inner lateral cusps are smaller than median cusps (Figure 4). The reproductive system is illustrated in Figure 5. The hermaphroditic duct widens into a convoluted ampulla, narrowing at its distal end, bifurcating into a deferent duct and a female duct. The deferent duct enlarges into a long and coiled prostatic duct, similar to the ampulla in length and thickness, which connects with a penis enclosed in a muscular and conical penial papilla. The penis is cylin- drical and armed with a curved and pointed apical spine. The common female duct is short and opens into the vagi- nal vestibule. A short and narrow duct connects the in- ternal end of the vagina with the seminal receptacle. Discussion Although the genus Phidiana and other related genera have been broadly discussed (Miller, 1974; Gosliner, 1979; Rudman, 1980), our specimens clearly belong to genus Phidiana. They have long oral tentacles; the rhino- phores are lamellate; the foot is anteriorly rounded; the jaws have a masticatory border with a row of teeth; the radular teeth are provided with lateral denticles on a cen- tral cusp; and the penis is ornamented with a spine. Miller Notes, Information & News (1974) considered the presence of cerata disposed in par- allel rows as a generic characteristic of Phidiana. This feature is present in P. mariadelmarae. The coloration of P. mariadelmarae differs from that of the majority of Phidiana species (Edmunds, 1964; MacFarland, 1966; Rudman, 1980; Willan, 1987). Only P. lascrucensis Bertsch & Ferreira, 1974, and P. pegasus Willan, 1987, have an orange body color similar to that of P. mariadelmarae. However, P. pegasus has the rhino- phores ornamented by pustules and the foot is enlarged anteriorly with two tentacles, approximately equal in length to the rhinophores (Willan, 1987), whereas in P. mariadelmarae the rhinophores are encircled by annula- tions, and the anterior border of the foot lacks elongated propodial tentacles. The coloration of P. mariadelmarae differs from that of P. lascrucensis because this species lacks mid-dorsal longitudinal white lines at the cephalic region, and the dorsal surface of the notum has numerous white spots. Internally, the radular teeth of P. mariadelmarae are similar to those of P. lascrucensis. However, they can be differentiated because on the masticatory border of the jaws, P. lascrucensis has two rows of denticles, with 23— 24 in the first row and five to six in the second. However, P. mariadelmarae has only one row with 15—22 denti- cles. The shape of the denticles differs from other species of Phidiana. The denticles in other species are described as rounded or irregular denticles (MacFarland, 1966; Rudman, 1980), whereas in P. mariadelmarae they are hooked and pointed. P. lynceus Bergh, 1867, is an Atlantic species with a mid-dorsal white line bifurcating at the level of the rhino- phores as in P. mariadelmarae. Both species have a sim- ilar reproductive system. However, P. lynceus has a line of bright vermilion red that runs from one oral tentacle to the other and the mid-dorsal white line extends from the tail to the rhinophores (Edmunds, 1964). The reproductive system of P. mariadelmarae coin- Page 193 cides with that of P. lascrucensis since in both species the penis has a curved apex with a spine at the tip. How- ever, because this system was not described in P. lascru- censis, is not possible to compare them. Acknowledgments We wish to express our gratitude to Agencia Espafiola de Cooperacion Internacional and Instituto Nacional de Recur- sos Naturales Renovables from Panama for their assistance during the expedition in the National Park of Coiba (Pana- ma) and to Engineer Luis Carlos Jiménez Cerrud and to Narciso Bastida (Mali Mali) for their constant help. This research was included in the project Inventario de la Fauna y Flora del Parque Nacional de Coiba, supported by Agen- cia Espanola de Cooperacion Internacional. Our thanks to Diane Avery-Gwynne for reading the paper. Literature Cited BerTSCH, H. & A. J. FERREIRA. 1974. Four new species of nudi- branchs from tropical West America. The Veliger 16:343-353. Epmunpbs, M. 1964. Eolid mollusca from Jamaica, with descrip- tions of two new genera and three new species. Bulletin of Marine Science of the Gulf and Caribbean 14:1—32. FARMER, W. M. 1980. Sea-Slug Gastropods. Wesley M. Farmer Enterprises, Inc. 177 pp. GOSLINER, T. M. 1979. The systematics of the Aeolidacea (Nu- dibranchia: Mollusca) of the Hawaiian Islands, with descrip- tions of two new species. Pacific Science 33:37—77. Lance, J. R. 1962. Two new opisthobranch mollusks from South- ern California. The Veliger 4:155—159. MACFARLAND, F M. 1966. Studies of Opisthobranchiate Mol- lusks of the Pacific Coast of North America. Memoirs of the California Academy of Sciences 6:546 pp. MILLER, M. C. 1974. Aeolid nudibranchs (Gastropoda: Opistho- branchia) of the family Glaucidae from New Zealand waters. Zoological Journal of the Linnean Society 54:31—61. RuDMAN, W. B. 1980. Aeolid opisthobranch molluscs (Glauci- dae) from the Indian Ocean and the south-west Pacific. Zoo- logical Journal of the Linnean Society 68:139-172. WILLAN, R. C. 1987. Description of a new aeolid nudibranch (Mollusca: Opisthobranchia) belonging to the genus Phidi- ana. New Zealand Journal of Zoology 14:409—417. The Veliger 42(2):194—-199 (April 1, 1999) THE VELIGER © CMS, Inc., 1999 BOOKS, PERIODICALS & PAMPHLETS Common and Scientific Names of Aquatic Invertebrates from the United States and Canada: Mollusks. Second Edition by D. D. TuRGEON, J. E QUINN Jr., A. E. BOGAN, E. V. Coan, E G. HOCHBERG Jr., W. G. LYONS, P. M. MIKKEL- SEN, R. J. NEVES, C. E E. Roper, G. ROSENBERG, B. ROTH, A. SCHELTEMA, E G. THOMPSON, M. VECCHIONE & J. D. WILLIAMS. 1998. Common and scientific names of aquatic invertebrates from the United States and Canada: Mol- lusks. 2nd Edition. American Fisheries Society, Special Publication 26, Bethesda, Maryland. ix + 526 pp. (incl. 16 pls.). ISBN 1-888569-01-8 (paper); ISSN 0097-0638. US$ 59.00 (includes CD). In the pages of this journal, informative book reviews typically close with a sentence like “‘of lasting value to a wide spectrum of readers” or ‘‘a fine field companion for work or pleasure on the reefs of the Indo-Pacific.” These concluding sentences typically identify the audi- ence as well as the publication’s value to that audience. As I read this work (hereafter referred to as “the Check- list’’), I looked for similar information to incorporate into my final sentence, but unfortunately a clear user group and purpose were not forthcoming. Therefore, I begin my review with what could have been the final sentence: a commendable and Herculean compilation of scientific and common names of northern American mollusks that has yet to identify its larger audience. Moreover, because of the uneven implementation of the principles that were supposed to guide its production, its overall value and usefulness are seriously compromised. I cannot deny that a handful of colleagues will find this volume a useful source of names for reporting molluscan catch statistics and for legal and regulatory documents. The fact that terrestrial and freshwater taxa are more like- ly to be mentioned in regulatory documents probably ac- counts for the better documentation and preparation of these sections compared to the marine groups. However, I doubt that the value of the Checklist will extend to scientific writing or to the professional shell collectors [sic] (p. 13) who the authors claim will welcome the stan- dardization of scientific and common names provided by this volume. Perhaps Riedl (1983:5) expressed it best [translated here from German], ‘‘Wanting to collect con- stant names is the misleading hope of the dilettante; to become aware of the order itself is the rewarded struggle of the expert.” Much of my conclusion results from my inability to find clear statements as to what the Checklist is supposed to do, and disappointment in how the individual contrib- utors applied (or failed to apply) the American Fisheries Society’s own principles (pp. 14-16) and the AMU/CSM resolutions (pp. 16—17) to their respective sections. In the final wash, a mixture of rank-driven shuffling, fiat, and phylogenetic tree pruning seems to have controlled much of the production of this chimera. In the foreword the AFS charge is clear: “The Com- mittee [on Names of Aquatic Invertebrates] shall be re- sponsible for studying and reporting on matters concern- ing common and scientific names of aquatic invertebrates and shall prepare checklists of names to achieve unifor- mity and avoid confusion in nomenclature”’ (p. vii). Be- cause the International Code of Zoological Nomenclature (ICZN) already provides oversight to achieve uniformity and avoid confusion in scientific names, I assume that the focus of the AFS charge must be directed primarily at common names. In the Checklist’s introduction, the AFS charge is restated and limited as follows: ‘‘Our goal is to keep the scientific nomenclature of this list up to date while achieving uniformity and avoiding confusion in the common names of the mollusks of the United States and Canada”’ (p. 11). Note that both charges focus only on nomenclature—not systematics, taxonomy, or classifica- tion. This is an important distinction and one that is often blurred in our field (and in this Checklist). The proposal of names and resolution of nomenclatural problems traditionally are dealt with using algorithmic procedures such as the application of the ICZN (B. Roth, personal communication). Algorithms are well suited here because they possess three key features: (1) algorithms are substrate neutral; (2) algorithms consist of small, sim- ple steps; and (3) algorithms have guaranteed results (Dennett, 1995:51). Like the Articles and recommenda- tions of the IZCN, many of the AFS principles and the AMU/CSM resolutions provide algorithmic procedures governing the generation and application of common names. Until very recently, algorithmic procedures were rarely used in taxonomy. Instead, taxonomy has been strongly dependent on the researcher and his or her wisdom, judg- ment, and intuition; the results were never guaranteed, even with identical data and training. Recently, the intro- duction of cladistic methods using explicit assumptions and character analyses has provided workers with an al- gorithmic procedure to reconstruct phylogenies. If the method, data, and assumptions are identical, it makes no difference whether the algorithms are executed in a lab- oratory in California or on a veranda in New South Wales. The hypothesis of relationships will be the same. Books, Periodicals & Pamphlets And most importantly, it can be redone, updated or mod- ified, and repeatedly tested. Although taxonomy, nomenclature, and even biodiver- sity itself (see Dennett, 1995) can be viewed as results of algorithmic processes, there is no such validation for the arbitrary assignment of taxonomic ranks or categories to taxa discovered through taxonomic study. Taxonomic ranks are clearly non-algorithmic in their creation and are almost certain to remain so. They are therefore also the most problematic components to apply and justify in sys- tematic studies. The AFS principles and the AMU/CSM resolutions provided a set of algorithmic procedures for the creation and emendation of the Checklist. Examples include, ‘“‘No two species on a list shall have the same primary name,” ‘“‘Names shall not violate the tenets of good taste,” “‘Names intended to honor persons ... are discouraged in that they are without descriptive value,” and ‘‘The most current literature should be used for systematic clas- sification.”’ In addition to these internal Checklist proce- dures for common names and the limited classification format, there is also the ICZN for scientific names. We also have a rapidly expanding, recent literature of phy- logenetic hypotheses of molluscan relationships available for producing meaningful, phylogenetically based classi- fications. Unfortunately, in many cases in the Checklist the algorithmic procedures were either not followed or discarded in favor of the old, comfortable “‘canonical tax- onomy”’ (I thank Barry Roth for coining this very appro- priate term; see, for example, The Veliger 38:81, 1995). In canonical systematics, authorities make subjective and untestable taxonomic decisions by fiat, which are shoe- horned into rank-driven classifications, and compete in the scientific and popular literature for acceptance. In this brand of systematics, algorithmic procedures are restrict- ed to ICZN nomenclature. The Checklist sets the stage for its use of canonical systematics early on by setting up phylogenetic system- atics as a straw man. In the introduction it is stated that scientific names are “‘intended”’ to provide supposed sys- tematic (evolutionary) relationships. This is demonstrably incorrect—historically and even today in many instances. Bartsch, Gould, Pilsbry, Clench, and Keen were all fine taxonomists and together described thousands of taxa. But I do not believe for a minute that they thought they were reconstructing the evolutionary history of groups that they monographed. Keen was explicit about this in her Sea Shells of Tropical West America (Keen, 1971); when her former student James H. McLean organized his contributed sections to reflect evolutionary relationships among taxa, and not alphabetically as in the rest of vol- ume, she provided an advisory notice at the beginning of Dr. McLean’s section (1971:308). With the exception of W. H. Dall (Lindberg, 1998), few North American mal- acologists were interested in studying and incorporating evolutionary relationships into their classifications until Page 195 the late 1960s and early 1970s. Thus, we are burdened with almost 100 years of canonical taxonomic work that likely reflects little in the way of phylogenetic relation- ships among taxa. (See Winsor [1995] for the issues sur- rounding the application of phylogenetic classifications and the history of this debate in England.) The fallacy that current molluscan classifications are phylogenetically based is clearly exposed in the very next sentence where the authors points out that ‘‘... new sys- tematic research and phylogenetic analysis, currently very active areas in malacology, often show that previous ideas of relationships between taxa are wrong and that one or more taxa must be reclassified.”” Why then should this happen so often? The simplest answer is that most current molluscan classifications are not based on evolutionary relationships. Instead, they were built on overall similar- ity or on heavily weighted, personal concepts of ‘‘good’’ characters (e.g., radula characters, shell structure, gill morphology). Only in the last 10 years or so has phylo- genetic systematics begun to provide alternative hypoth- eses of relationships. Phylogenetic studies often contra- dict earlier classifications and can lead to extensive re- classifications of groups. However, the incorporation of published reclassifications of this kind into the Checklist appears to have been uneven. The plan of the list (p. 12) states that the classification used in the Checklist “‘approximates”’ the systematic ar- rangement of taxa advocated by recent phylogenetic an- alyses. However, phylogenetic arrangements cannot be pruned and grafted to conform to political and ecological boundaries or popular sensibilities and remain meaningful representations of relationships. Exclusion of taxa that were included in a phylogenetic analysis is likely to pro- duce paraphyletic groupings that confound relationships, and destroy the classification’s usefulness in estimating biodiversity and biogeographic distributions—two of the stated goals of the Checklist. Examples of ‘‘approximating the systematic arrange- ment of taxa advocated by recent phylogenetic analyses” while maintaining traditional groupings include the pres- ence of the “‘Archaeogastropoda’’—a blatantly paraphy- letic group that refuses to go away in spite of repeated attacks by both evolutionary systematists and cladists alike (Graham, 1985; Salvini-Plawen & MHaszprunar, 1987; Ponder & Lindberg 1997). In the Checklist’s in- carnation of the “‘Archaeogastropoda”’ the Neritopsina are removed from the group, but the taxon Cocculinidae remains grouped within the Archaeogastropoda despite the insightful work of Haszprunar (1988a) and others. While the authors correctly point out that the inclusion of the Cocculinidae within the Neritopsina by Ponder & Lindberg (1997) is not well supported, moving the Coc- culinidae to the end of the list of ““Archaeogastropoda”’ to place them next to the Neritopsina does not reflect this uncertainty in this supposedly phylogenetic arrangement of taxa. Because branch segments can freely rotate at Page 196 their nodes in a cladogram, it is possible to place the terminal branch label Neogastropoda next to the terminal branch label Patellogastropoda in most gastropod phylog- enies. We could then list the taxon names from left to right (or right to left) and have Neogastropoda next to the Patellogastropoda. However, the fact that they are ad- jacent to one another in no way indicates a close rela- tionship unless they are also sister taxa. These nuances cannot be simply mixed in an amalgamation of traditional canonical systematic practices and phylogenetic classifi- cation. However, the real travesty in the Checklist classifica- tion is the absence of the taxon Caenogastropoda. Caen- ogastropoda was proposed almost 40 years ago by Cox (1960) and subsequently appeared in every meaningful study of gastropod systematics (Bieler, 1991). It is men- tioned only once in the Checklist in a footnote to the Gastropoda (p. 56). The absence of the Caenogastropoda from the Checklist seems to hinge on the following state- ment in the footnote. ““Because of the continuing evolu- tion of the higher classification of gastropods, the con- flicts between the existing classifications, and the con- straints imposed by the nature of this list, we have adopt- ed an arrangement that borrows elements from current classifications and phylogenies while maintaining the util- ity of and a degree of familiarity with the list for the nonsystematist.”’ All of these justifications are demonstrably false. In- stability in higher gastropod classification? With the ex- ception of the placement of the hydrothermal vent taxa, Neritopsina, and Cocculinidae, the “‘higher’’ classifica- tion of the Gastropoda has been relatively stable for al- most 10 years (Haszprunar, 1988b; Bieler, 1991:table 1; Ponder & Lindberg, 1997). Prior to Haszprunar’s (1988b) all-out assault on Thiele’s (1925) gastropod classification, Thiele’s system was already suspect with the proposal of Neritopsina by Yonge (1947), Cox’s (1960) proposal of Caenogastropoda, and Golikov & Scarlato’s (1976) clas- sification. Most remaining conflicts are within the larger groupings (i.e., Vetigastropoda, Caenogastropoda), and not questions of monophyly or the relative relationships of the higher taxa used in classification. Constraints im- posed by the list? They must have been unwritten for there is nothing in the AFS principles or AMU/CSM res- olutions that prevents the use of a modern systematic framework. To the contrary, Resolution 20 states that “The most current literature should be used for system- atic classification,”’ and the plan of the list sought to ‘‘ap- proximate the systematic arrangement of advocated by recent phylogenetic analyses, particularly in the gastro- pods” (emphasis added). As argued above, there is no algorithm or procedure for combining canonical and phy- logenetic classifications, and the results of such mischief do not yield practical or utilitarian classifications. Instead, the “‘higher gastropod classification’ used in the Check- list is unique and is not found in any other systematic The Veliger, Vol. 42, No. 2 treatment of the gastropods. It therefore cannot be famil- iar to anyone. Paradigm changes in science often produce a Tower of Babel effect with different groups of practitioners speak- ing languages that are unintelligible to one another. The shift from a canonical to a phylogenetic systematics has had such an effect and its residues are acutely apparent in the Checklist. For example, Mikkelsen’s (1996) phy- logenetic analysis unequivocally supports the demise of the traditional organization of Cephalaspidea. However, her findings are not included in the Checklist because of the “strictures of the organization of this list’’ (again, the mysterious and secret “‘list constraints’? that are not shared with the reader), and “‘pending more explicit state- ments of relationships.”’ Currently, there is no more ex- plicit statement of relationships than the cladogram pro- duced by Mikkelsen’s phylogenetic analysis. Perhaps more data or another outgroup might produce a different tree, but it would certainly not be a “‘more explicit state- ment of relationships,’ just a different one. Another con- fusing rationalization occurs in the footnotes to the Con- idae. Here the authors discuss the classification of Taylor et al. (1993), and concede that it “is better supported by anatomical and radular data than any previous one,” but then go on to suggest that “... a more ‘comfortable’ arrangement would have had these four subfamilies in a family of their own.” Personally, I would be comfortable with four ele- ments—air, water, fire, and earth. I can keep all of the them and their elemental and essential qualities in my head, and easily visualize the transformation of water into air by the addition of fire. I cannot keep 112 elements and associated information like atomic number and weights, and electron configuration in my head, nor can I mentally solve the simplest chemical reactions without an aid called the Periodic Table. This table reflects our current and best understanding of the elements, and more importantly allows us to do superior and more predictive science than the Aristotelian elements. Perhaps our clas- sifications have reached the point where ranks and suffix- es are no longer sufficient to represent our knowledge, and perhaps we require aids like cladograms and indented listings to represent our best understanding of molluscan classification.! One of the most meritorious undertakings of the Checklist framers was the inclusion of Resolution 10— “Justification should be presented when necessary to ex- ' This analogy is not as farfetched as it may initially appear. The conception of the Linnaean classification scheme was guided by Linnaeus’s belief in a Special Creation, perfection of species, and natural groupings that reflected intelligent design. Phyloge- netic classification assumes and seeks to represent descent with modification. The philosophical distance between these two po- sitions is just as great as that between the Aristotelian elements and the Periodic Table. Books, Periodicals & Pamphlets plain inclusion or deletion of a scientific or common name. (This is a procedural requirement [emphasis add- ed] of all editions after the first.).”’ However, this require- Ment is too often ignored or shammed throughout the volume. While many of the Checklist authors provided citations to peer-reviewed, primary literature, others used the footnotes to point to seashell trading cards, privately printed and distributed photocopies, and even other checklists to justify nomenclatural choices. According to my copy of the OED a justification is “‘the action of jus- tifying or showing something to be just, right or proper.” This could be brief, but I assume it would have to contain some explanatory material. The most blatant lack of justifications for nomencla- tural changes is in eastern Pacific bivalves where whole- sale changes are referenced to another checklist and therein to another footnote creating a virtual loop of vagueness (see also Resolution 14). For example, in the AFS Checklist (p. 194) Psephidia stephensae is consid- ered “‘to be a synonym of Nutricola cymata; P. stephen- sae is deleted.’’ Checking the supposed justification for this deletion in the cited reference (Coan & Scott, 1997: 25) we find, ““We regard Psephidia stephensae Hertlein and Grant, 1972, as a synonym of Nutricola cymata.” This is a fiat (OED: ‘‘an authoritative pronouncement, decree, command, order’’) and contains no more infor- mation than the action that it supposedly justifies. It re- mains to be seen whether the long-awaited volume on the marine bivalves of the northeastern Pacific Ocean (Coan & Scott, in preparation) will provide explanations for the multitude of changes made in both checklists. In marked contrast to those who ignored Resolutions 10 and 14, oth- er authors (especially in the terrestrial and freshwater sec- tions) used this resolution to remove and undo unsub- stantiated nomenclatural and distributional changes from the first edition. Other inconsistent applications of the principles and resolutions include the discouraging of patronymics (AFS Principle 6). So while Hemphill lost his slug and Dall, Gould, and Pilsbry their tuskshells, Carpenter kept his carditid, Oldroyd her penshell, and Bartsch his shipworm. There are also some strange biogeographic conventions. Taxa that occur in both the Gulf of Mexico and the trop- ical eastern Pacific (e.g., Aplysia parvula) are listed only as “‘A” (western Atlantic Ocean including the Gulf of Mexico) because the Pacific Ocean that touches the coast of Mexico is outside the area of coverage of the list. How does one use this list to evaluate biodiversity given this kind of data? There are also logic problems with the ex- clusion of Hawaiian taxa from the Checklist. One of the reasons Hawaii is excluded from the Checklist is because “its fauna is of Indo-Pacific origin.’’ Does this mean that the fauna covered in the Checklist must have originated in the US and Canada with the exception of the intro- duced taxa in Appendix 4? Absolutely not: Marincovich (1983), Vermeij et al. (1990), Lindberg (1991), McLean Page 197 (1984), and others have convincingly demonstrated biotic interchange between North America and the temperate regions of Asia and South America. There is also sub- stantial overlap of the Arctic fauna (which is covered in the Checklist) with the faunas of Greenland, Iceland, and Arctic Europe. Was it assumed that widely dispersed Arc- tic taxa originated in North America and subsequently migrated out of the New World? While the AFS charge was clear, the authors’ goals laudable, and the principles and resolutions unambiguous and comprehensible, the 2nd edition of ‘‘Mollusks’’ does not overcome the past and, regrettably, some of the pre- sent practices of molluscan taxonomy. The appendices of endangered and threatened mollusks, extinct mollusks, and introduced mollusks are useful and welcome addi- tions, but the remaining three appendices (‘‘For readers who are relatively new to the field of malacology, .. .’’) seem out of place and passé. They also provide little in- formation for the neophyte. For example, the illustration of chiton anatomy in the appendix “Introduction to North American Mollusks”’ shows only a mouth, anus and gills in addition to the requisite plates and girdle. Evidently, these animals do not reproduce or have other life func- tions. Anatomical illustrations of bivalves, scaphopods, gastropods, and cephalopods show those taxa to be better endowed, but not so the aplacophorans. The coiled mon- oplacophoran protoconch, debunked by Lindberg (1985) and Wingstrand (1985) makes a return appearance in this appendix as well. The Checklist’s introductory materials and many of the appendices are almost identical to the introductory material of the first edition of American Sea- Shells (Abbott, 1954)—Man and Mollusks, Life of [Mol- lusks], Collecting North American Mollusks, Guide to the Molluscan Literature. It’s all there; even the dedication to the esteemed author of two editions of American Sea- shells—R. Tucker Abbott. What about the 3rd edition of the Checklist? A limited view of the future is on the CD that accompanies the Checklist volume. Adobe Acrobat® Reader 3.0.1 is sup- plied on the disc and with it the user can display on- screen facsimiles of the Checklist. The display is in the form of several related documents and each document is searchable. Ten years from now it is unlikely that hard copy of the 3rd edition of the Checklist will need to be produced. The future most likely contains distributed tax- onomic resources, where individual researchers maintain their most recent monographic treatments, data, and clas- sifications on the World Wide Web (or whatever the web becomes). Rather than open a book, we will likely send our electronic assistants to the Checklist URL (e.g., www.IBM.checklist.org) to access a meta-database of distributed taxonomic resources that will then be queried and the results (and supporting data) returned to you in the blink of an eye. For those who cannot wait 10 years, the book/CD combination is available from AFS Publi- Page 198 cation Fulfillment, PO. Box 1020, Sewickley, PA 15143 USA. I thank J. H. McLean for providing pertinent literature, G. Haszprunar for bringing R. Riedl’s quote to my atten- tion, and B. Roth and W. E Ponder for their criticism, insight, and forbearance. However, acknowledgment here in no way indicates their espousal of any of the opinions expressed above. D. R. Lindberg Literature Cited ABBOTT, R. T. 1954. American Seashells. D. Van Nostrand Com- pany: Princeton, New Jersey. 541 pp. BEILER, R. 1991. Gastropod phylogeny and systematics. Annual Review of Ecology and Systematics 23:311-—338. Coan, E. V. & P. H. Scott. 1997. Checklist of the marine bi- valves of the northeastern Pacific Ocean. Santa Barbara Mu- seum of Natural History Contributions in Science 1:1—28. Cox, L. R. 1960. Thoughts on the classification of the Gastro- poda. Proceedings of the Malacological Society of London 33:239-261. DENNETT, D. C. 1995. Darwin’s Dangerous Idea. Touchstone: New York. 586 pp. Go.ikov, A. N. & Y. I. SCARLATO. 1976. Systematics of proso- branch gastropods. Malacologia 15(1):185—232. GraHaM, A. 1985. Evolution within the Gastropoda: Prosobran- chia. Pp. 151-186 in E. R. Trueman & M. R. Clark (eds.), The Mollusca, Vol. 10, Evolution. Academic Press: New York. HASZPRUNAR, G. 1988a. Comparative anatomy of cocculiniform gastropods and its bearing on archaeogastropod systematics. Pp. 64-84 in W. FE Ponder (ed.), Prosobranch Phylogeny. Malacological Review, Supplement 4. HASZPRUNAR, G. 1988b. On the origin and evolution of major gastropod groups, with special reference to the Streptoneura (Mollusca). Journal of Molluscan Studies 54:367—441. KEEN, A. M. 1971. Sea Shells of Tropical West America 2nd ed. Stanford University Press: Stanford, California. 1064 pp. LINDBERG, D. R. 1985. Aplacophorans, monoplacophorans, po- lyplacophorans and scaphopods: The lesser classes. Pp. 230— 247 in T. W. Broadhead (ed.), Mollusks. Notes for a Short Course, University of Tennessee, Department of Geological Sciences, Studies in Geology. 13. LINDBERG, D. R. 1991. Marine biotic interchange between the northern and southern hemispheres. Paleobiology 17:308— 324. LINDBERG, D. R. 1998. William Healey Dall: A Neo-Lamarckian view of molluscan evolution. The Veliger 41:227—238. Marincovicn, L. N. 1983. Asiatic mollusks in Miocene faunas of the Alaska Peninsula. U.S. Geological Survey Profes- sional Paper P1375:178-179. MILKELSEN, P. M. 1996. The evolutionary relationships of Ce- phalaspidea s. 1. (Gastropoda, Opisthobranchia)—a phylo- genetic analysis. Malacologia 37:375—442. McLean, J. H. 1984. Shell reduction and loss in fissurellids: a review of genera and species in the Fissurellidea group. American Malacological Bulletin 2:21—34. PONDER, W. FE & D. R. LINDBERG. 1997. Towards a phylogeny of gastropod molluscs: an analysis using morphological characters. Zoological Journal of the Linnean Society 119: 83-265. The Veliger, Vol. 42, No. 2 RIEDL, R. 1983. Fauna und Flora des Mittelmeeres. Verlag Paul Parey, Hamburg und Berlin. 836 pp. SALVINI-PLAWEN, L. v. & G. HASZPRUNAR. 1987. The Vetigastro- poda and the systematics of streptoneurous Gastropoda (Mollusca). Journal of Zoology, London 211:747-770. TAyLor, J. D., Y. I. Kantor & A. V. Sysoev. 1993. Foregut anatomy, feeding mechanisms, relationships and classifica- tion of the Conoidea (= Toxoglossa) (Gastropoda). Bulletin of the Natural History Museum London (Zoology) 59:125— 170. THIELE, J. 1925. Gastropoda. In Handbuch der Zoologie. Vol. 5. (ed. T. Krumbach) Walter de Gruyter & Co: Leipzig. 275 Pp. VERMEI, G., A. R. PALMER, & D. R. LINDBERG. 1990. Range limits and dispersal of molluscs in the Aleutian Islands, Alaska. The Veliger 33(4):346—-354. WINGSTRAND, K. G. 1985. On the anatomy and relationships of Recent Monoplacophora. Galathea Report 16:7—94. Winsor, M. P. 1995. The English debate on taxonomy and phy- logeny, 1937-1940. History and Philosophy of the Life Sci- ences 17:227-252. YONGE, C. M. 1947. The pallial organs in the aspidobranch Gas- tropoda and their evolution throughout the Mollusca. Phil- osophical Transactions of the Royal Society of London B 232:443-518. Taxonomic Atlas of the Benthic Fauna of the Santa Maria Basin and Western Santa Barbara Channel. Volume 8. The Mollusca Part 1—The Aplacophora, Polyplacophora, Scaphopoda, Bivalvia and Cephalopoda edited by P. VALENTICH ScoTT & J. A. BLAKE. 1998. Santa Barbara Museum of Natural History, Santa Barbara, Cal- ifornia. viii + 250 pp. ISBN 0-936494-13-1. “The Mollusca Part 1”? of the Taxonomic Atlas con- tains treatments of the Aplacophora by Amélie H. Schel- tema, the Polyplacophora by Douglas J. Eernisse, the Sca- phopoda by Ronald L. Shimek, the Bivalvia by Paul Val- entich Scott, and the Cephalopoda by E G. Hochberg; there is a brief general introduction to the Mollusca by Eugene V. Coan. This work completes coverage of the phylum along with the earlier-published Part 2, the Gas- tropoda by James H. McLean and Terrance M. Gosliner (reviewed in Veliger vol. 39, no. 3). The specimen ma- terial described is mainly that collected during U.S. De- partment of the Interior Minerals Management Service (MMS) benthic monitoring in the Santa Maria Basin off central California, from about Point Estero to Point Con- ception, and the western Santa Barbara Channel, south- east of Point Conception, in depths of appoximately 100— 600 m. The relevance of the work, however, extends beyond those geographic limits; and each individual contribution Books, Periodicals & Pamphlets Page 199 includes general information about its taxomic group and advice on collection, preservation, and laboratory meth- ods applicable to the taxon as a whole. An appendix with maps and station lists relates to all articles; except for this, and a general index, each section is in effect free- standing, with its own bibliography and self-contained il- lustrations. The quality of the figures, both line drawings and photographs, ranges from good to superb. Dealing as they do with disparate clades of organisms, the articles differ among themselves as to the characters and the de- tail they address. All, however, are pragmatically oriented and helpful to potential users—the way a taxonomic atlas should be. Five new species and one monotypic new species are described by Scheltema in the Aplacophora, and two new species by Valentich Scott in the Bivalvia. Several new taxonomic combinations occur, of which Enteroctopus dofleini (Wilker, 1910) for the Giant North Pacific Oc- topus will perhaps attract the most notice. The volume and series are available from the Depart- ment of Invertebrates, Santa Barbara Museum of Natural History, 2559 Puesta Del Sol Road, Santa Barbara, Cal- ifornia 93105-2936, U.S.A. B. Roth Land Snails of New Mexico edited by A. L. METCALF and R. A. SMARTT. 1997. Bul- letin 10, New Mexico Museum of Natural History and Science. 111 + 145 pp. This work consists of three related papers. The first, ‘Land snails of New Mexico: a systematic review” (pp. 1—69) by Metcalf and Smartt, surveys snail and slug spe- cies occurring in the state, emphasizing present geograph- ic, altitudinal, and habitat distribution. ‘‘Land snails of New Mexico from a historical zoogeographic point of view” (pp. 71-108) by Metcalf discusses and evaluates efforts to delineate zoogeographic provinces for the land snails in southwestern United States, describes geologic events and fossil faunas of the region from later Mesozoic to Quaternary time, and proposes a model of Cenozoic historical geography of the New Mexico terrestrial mol- lusk fauna. “‘Altitudinal distributions of land snails in some montane canyons in New Mexico” (pp. 109-127) by Timothy J. Dillon and Metcalf focuses in on the results of collecting along six transects in New Mexico mountain ranges. Two appendices present the underlying data on taxa and localities for the first and third articles. The work is available from the New Mexico Museum of Natural History and Science, 1801 Mountain Road NW, Albuquerque, NM 87104 UzS.A. B. Roth Cee eevee, 4 i { i! i iY i , i ; vn noe eae i fi OT nt 1 Information for Contributors Manuscripts Manuscripts must be typed, one side only, on A4 or equivalent (e.g., 842” X 11”) white paper, and double-spaced throughout, including references, figure legends, footnotes, and tables. All margins should be at least 25 mm wide. Text should be ragged right (i-e., not full justified). Avoid hyphenating words at the right margin. Manuscripts, in- cluding 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, in- cluding the year, if possible. Underline scientific names and other words to be printed in italics; no other manipulation of type faces is necessary on the manuscript. Metric and Celsius units are to be used. For aspects of style not ad- dressed here, please see a recent issue of the journal. The Veliger publishes in English only. Authors whose first language is not English should seek the assistance of a col- league who is fluent in English before submitting a manu- script. In most cases, the parts of a manuscript should be as follows: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, literature cited, figure legends, footnotes, tables, and figures. The title page should be a separate sheet and should include the title, authors’ names, and addresses. The abstract should be less than 200 words long and should describe concisely the scope, main results, and conclusions of the paper. It should not include references. 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 (Phillips, 1981), for two authors (Phillips & Smith, 1982), and for more than two (Phillips et al., 1983). The reference need not be cited when author and date are given only as authority for a taxonomic name. The “literature cited” section should include all (and only) references cited in the text, listed in alphabetical order by author. Each citation must be complete, with all journal titles unabbreviated, and in the following forms: a) Periodicals: Hickman, C. S. 1992. Reproduction and development of trochacean gastropods. The Veliger 35:245—272. b) Books: Bequaert, J. C. & W. B. Miller. 1973. The Mollusks of the Arid Southwest. University of Arizona Press: Tuc- son. xvi + 271 pp. c) Composite works: Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135 in R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stanford Univer- sity Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Avoid vertical rules. Figures and plates Figures must be carefully prepared and submitted ready for publication. Each should have a short legend, listed on a sheet following the literature cited. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photo- graphs for halftone reproduction must be of good quality, trimmed squarely, grouped as appropriate, and mounted on suitably heavy board. Where appropriate, a scale bar may be used in the photograph; otherwise, the specimen size should be given in the figure legend. Photographs should be submitted in the desired final size. Clear xerographic copies of figures are suitable for re- viewers copies of submitted manuscripts. It is the author's responsibility to ensure that lettering will be legible after any necessary reduction and that lettering size is appropriate to the figure. Use one consecutive set of Arabic numbers for all illus- trations (that is, do not separate “plates” from “text fig- ures”). Processing of manuscripts Each manuscript is critically evaluated by at least two reviewers. Based on these evaluations the editor makes a preliminary decision of acceptance or rejection. The editor’s decision and the reviewers’ comments are sent to the author for consideration and further action. Unless requested, only one copy of the final, revised manuscript needs to be re- turned to the editor. The author is informed of the final decision and acceptable manuscripts are forwarded to the printer. The author will receive proofs from the printer. One set of corrected proofs should be mailed promptly to the editor after review. Changes other than the correction of printing errors will be charged to the author at cost. An order form for the purchase of reprints will accom- pany proofs. Reprints are ordered directly from the printer. Authors’ contributions The high costs of publication require that we ask authors for a contribution to defray a portion of the cost of pub- lishing their papers. However, we wish to avoid a handicap to younger contributors and others of limited means and without institutional support. Therefore, we have adopted the policy of asking for the following: $30 per printed page for authors with grant or other institutional support and $10 per page for authors who must pay from their personal funds (2.5 double-spaced manuscript pages normally equal one printed page). This request is made only after the pub- lication of a paper; these contributions are unrelated to the acceptance or rejection of a manuscript, which is entirely on the basis of merit. In addition to this requested contri- bution, authors of papers with an unusually large number of tables or figures will be asked for an additional contri- bution. Because these contributions by individual authors are voluntary, they may be considered by authors as tax- deductible donations to the California Malacozoo- logical Society, Inc., to the extent allowed by law. It should be noted that even at the rate of $30 per page, the CMS is paying well over half the publication costs of a paper. Authors for whom even the $10 per page contri- bution would present a financial hardship should explain this in a letter accompanying their manuscript. The edito- rial board will consider this an application for a grant to cover the publication costs. Authors whose manuscripts in- clude very large tables of numbers or extensive lists of (e.g.) locality data should contact the editor regarding possible electronic archiving of this part of their paper rather than hard-copy publication. Submitting manuscripts Send manuscripts, proofs, books for review, and corre- spondence on editorial matters to Dr. Barry Roth, Editor, 745 Cole Street, San Francisco, CA 94117, USA. CONTENTS — Continued Ontogenetic changes in boring behavior by the rock-boring bivalve, Barnea man- ilensis (Pholadidae) YASUPIROSITOY 7 vay carevstaxe incor cP stonerareneke re tele teey ener ence tee eee A new species of gastropod of the genus Trophon Montfort, 1810 (Mollusca: Gas- tropoda: Muricidae) from subantarctic waters GUIDO“PASTORINO: oh coicrt ieee oles cee eee Re OIE ee one eeeee er e e Observations on epithelial mucocytes in the sole of Patella species and Littorina littorea (Linnaeus, 1758) MARK SeTDAVIES.) 3 ie.cts sai: 4: clzns sc den apenas one ese epee ae sae ere Aen ee NOTES, INFORMATION & NEWS Observations on the winter spawning and larval development of the ribbed limpet Lottia digitalis (Rathke, 1833) in the San Juan Islands, Washington, USA ALAN R. HOLyoak, DONALD J. BROOKS, AND SHAWNA R. COBLENTZ ........- The description of a new species of Favartia (Murexiella) from the South Pacific Ocean BARBARA W. MYERS AND ‘CAROLE M. TIERTZ 05. 5. = oe ges i olels apes een High performance thin layer chromatography determination of carbohydrates in the hemolymph and digestive gland of Lymnaea elodes (Gastropoda: Lymnaei- dae) DANIEL J. CLINE, BERNARD FRIED, AND JOSEPH SHERMA ......------2+eeeeeee- On the egg capsules of Epitonium georgettinum (Kiener, 1839) (Gastropoda: Epi- toniidae) from Patagonian shallow waters GUIDO: PASTORINO/ AND: PABLO PENGHASZADEH =... 2. 4.0, en) 05 eee ee Description of a new species of the genus Phidiana Gray, 1850 (Nudibranchia: Facelinidae) from Pacific Ocean waters of Panama FRANCISCO) J... GARCIA AND JESUS /S: TPRONCOSOm. ans 5 se on ee eee BOOKS, PERIODICALS *&, PAMIPENIEE Six ie ss cpetel ee eeer eee eee erie VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 42 julyai e999, OL 4O\ ISSN 0042-3211 vty \\ CONTENTS A new species of Doriopsilla (Nudibranchia: Dendrodorididae) from the Pacific Coast of North America, including a comparison with Doriopsilla albo- punctata (Cooper, 1863) TERRENCE M. GOSLINER, MARIA C. SCHAEFER, AND SANDRA V. MILLEN Observations on the embryonic development of Octopus mimus (Mollusca: Ce- phalopoda) from northern Chile K. WARNKE eilsjlesiviceise)ellejielie: 0: eles) elie) lelieiie)eiie\ieice!.«\(e:1e)/0)/e 0. \e]\e/.e'\s| 6) 0 14) 011s) ee: 0.01161 06 0 je'10.0 10 0! ele) \0 10 10/10 6 6. © Bathymodiolus (Bivalvia: Mytilidae) from hydrothermal vents on the Azores Triple Junction and the Logatchev hydrothermal field, Mid-Atlantic Ridge RUDO VON COSEL, THIERRY COMTET, AND ELENA M. KRYLOVA .............. 218 Shell form and color variability in Alia carinata (Neogastropoda: Columbellidae) SUISSE ANY SRSIRWIREN Sempre ree nelettaunh eatin cn He ss Soar L/S Ra cae span eta cialiaa ellie! cuaieli tard 249 Remains of the prey—recognizing the midden piles of Octopus dofleini (Wiilker) REP DOM EREVAN DED YS CHEE W saat eisiciataatta isos cate nas tathy wese-ncapac alsa ti acd oe ia ease ao 260 Morphometric species recognition in Brachidontes darwinianus and Brachidontes solisianus (Bivalvia: Mytilidae) MARCEL OKAMOTO TANAKA AND CLAUDIA ALVES DE MAGALHAES ...........-. 267 Early development of Fissurella picta (Gmelin, 1791) M. L. GONZALEZ, M. C. PEREZ, D. A. LOPEZ, J. M. URIBE, AND C. A. PINO .. 275 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly in January, April, July, and October by the California Malacozoological Society, Inc., % Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. Periodicals postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to The Veliger, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. Number 3 THE VELIGER Scope of the journal The Veliger is an international, peer-reviewed scientific quarterly published by the Cali- fornia Malacozoological Society, a non-profit educational organization. The Veliger is open to original papers pertaining to any problem connected with mollusks. Manuscripts are considered on the understanding that their contents have not appeared, or will not appear, elsewhere in substantially the same or abbreviated form. Holotypes of new species must be deposited in a recognized public museum, with catalogue numbers provided. Even for non- taxonomic papers, placement of voucher specimens in a museum is strongly encouraged and may be required. Very short papers, generally not over 750 words, will be published in a “Notes, Infor- mation & News” column; in this column will also appear notices of meetings and other items of interest to our members and subscribers. Editor-in-Chief Barry Roth, 745 Cole Street, San Francisco, CA 94117, USA e-mail: veliger@ucmp1.berkeley.edu Production Editor Leslie Roth, San Francisco Board of Directors Michael G. Kellogg, City and County of San Francisco (President) Hans Bertsch, National University, San Diego Henry W. Chaney, Santa Barbara Museum of Natural History Eugene V. Coan, California Academy of Sciences, San Francisco Terrence M. Gosliner, California Academy of Sciences, San Francisco Carole S. Hickman, University of California, Berkeley E G. Hochberg, Santa Barbara Museum of Natural History Matthew J. James, Sonoma State University David R. Lindberg, University of California, Berkeley James Nybakken, Moss Landing Marine Laboratories Peter U. Rodda, California Academy of Sciences, San Francisco Barry Roth, San Francisco Geerat J. Vermeij, University of California, Davis Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (not institutions) interested in any aspect of malacology. New members join the society by sub- scribing to The Veliger. Rates for Volume 42 are US $40.00 for affiliate members in North America (USA, Canada, and Mexico) and US $72.00 for libraries and other institutions. Rates to members outside of North America are US $50.00 and US $82.00 for libraries and other institutions. All rates include postage, by air to addresses outside of North America. Memberships and subscriptions are by Volume only and follow the calendar year, starting January 1. Payment should be made in advance, in US Dollars, using checks drawn from US banks or by international postal order. No credit cards are accepted. Payment should be made to The Veliger or “CMS, Inc.” and not the Santa Barbara Museum of Natural History. Single copies of an issue are US $25.00, postage included. A limited number of back issues are available. Send all business correspondence, including subscription orders, membership applications, pay- ments, and changes of address, to: The Veliger, Dr. Henry Chaney, Secretary, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105, USA. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. Barry Roth, Editor, 745 Cole Street, San Francisco, CA 94117, USA. © This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). The Veliger 42(3):201—-210 (July 1, 1999) THE VELIGER © CMS, Inc., 1999 A New Species of Doriopsilla (Nudibranchia: Dendrodorididae) from the Pacific Coast of North America, Including a Comparison with Doriopsilla albopunctata (Cooper, 1863) TERRENCE M. GOSLINER Department of Invertebrate Zoology and Geology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118, USA MARIA C. SCHAEFER Department of Biology, San Francisco State University 1600 Holloway Avenue, San Francisco, California 94132, USA AND SANDRA V. MILLEN Department of Zoology, University of British Columbia 6270 University Boulevard, Vancouver, B.C., Canada V6T 1Z4 Abstract. Much confusion has surrounded the systematics of the yellow species of Dendrodorididae inhabiting the Pacific coast of North America. Re-examination of Doris albopunctata Cooper, 1863, indicates that it is properly placed in Doriopsilla. Examination of specimens from different localities throughout California, the Pacific coast of Baja California, and within the Gulf of California, indicates that all white-gilled specimens are conspecific with D. albo- punctata, and that D. fulva (MacFarland, 1905) and D. reticulata (Cockerell & Eliot, 1905) are regarded as junior synonyms. A second species with yellow gills is here described as Doriopsilla gemela. It differs from D. albopunctata in aspects of its color pattern, external morphology, digestive system, reproductive anatomy, and developmental biology. The two species also differ in allozyme allelic frequencies. Doriopsilla gemela and D. albopunctata are also compared to other members of the genus present in the eastern Pacific. These comparisons show that, while D. gemela and D. albopunctata are externally similar to each other, their internal anatomy is more similar to other species than to each other. INTRODUCTION Much confusion has surrounded the systematic status of yellow porostomate dorids on the Pacific coast of North America. Cooper (1863) described Doris albopunctata from Santa Barbara, California. MacFarland (1905) later described Doriopsis fulva from Monterey Bay, California, and Cockerell & Eliot (1905) described Doridopsis retic- ulata from San Pedro, California. Steinberg (1961), Roll- er (1970), McDonald (1983), and Valdés & Ortea (1997) considered these three species as synonymous. Behrens (1980, 1991) considered Dendrodoris fulva as distinct from Doriopsilla albopunctata. Behrens (1980, 1991), McDonald & Nybakken (1980), and McDonald (1983) considered another species (as Dendrodoris sp. 1 and Dendrodoris sp. A, respectively) with yellow rather than white gills and different egg ribbon shape as a distinct, undescribed species. Despite the above differences, no detailed examination of anatomy, developmental biology, or genetic differences has been undertaken. This paper examines these taxa in detail to determine systematic re- lationships. MATERIALS AnD METHODS In order to study the anatomy, developmental biology, and genetic variability of the two species of Doriopsilla, numerous specimens of both species were collected and examined. Specimens for anatomical study were collected from five localities along the central and southern Cali- fornia coast and from several localities on both coasts of Baja California. More than 15 individuals of each species were examined to ascertain intraspecific and interspecific anatomical variability. For developmental studies, live specimens of Doriop- silla albopunctata were collected from five localities along the central and southern California coast. Speci- mens of D. gemela were collected from Hill Street, San Diego. Specimens were maintained in the seawater sys- tem at San Francisco State University and at the Steinhart Aquarium of the California Academy of Sciences. Page 202 Allozyme samples were conducted on 54 nudibranchs. Of these individuals 12 were Doripsilla gemela from Hill Street. Forty-two individuals of D. albopunctata were sampled for allozyme comparison, including 12 speci- mens from Hill Street, San Diego, eight from Bird Rock, San Diego, eight from Diablo Canyon, San Luis Obispo County, 11 from Carmel Point, Monterey County, and three from Pillar Point, San Mateo County. Specimens were dissected. The anterior half of each animal was fixed in Bouin’s fixative for anatomical comparison as vouch- ers. The posterior portions of the fresh tissue samples were homogenized in 1:1 volume ratio of tissue to ho- mogenizing buffer. All samples were blotted on Whatman #2 filter paper. Excess moisture was wiped off of samples, and samples were placed in horizontal 11% starch gels. Gels were cooled by placing Blue Ice packs on top of the running gel. Two gel conditions were run at 150 volts on Heathkit model IP-17 power supplies. One gel system was Poulik’s discontinuous buffer (electrode buffer pH 8.2, gel buffer pH 8.7), and was run for 5 hours. The other gel system was an amine citrate continuous buffer system at pH 7.8, and was run for 3.5 hours. Staining of the gels was undertaken by standard procedures. Ten al- lozyme stains were attempted. The allozymes attempted on the Poulik’s gel were Superoxide dismutase (S.O.D), Tri-Peptidase-1 and -2 (Trip-1, Trip-2), Phosphoglucose Isomerase (P.G.I.), Mannose Phosphate Isomerase (M.P.I.), and Phosphoglucomutase (P.G.M.). The allo- zymes attempted on the Amine-citrate gel were Super- oxide dismutase (S.O.D.), Creatin Kinase (C.K.), Aden- alin Kinase (A.K.), Malate Dehydrogenase (M.D.H.), Iso- citrate dehydrogenase (I.D.H.), and 6 Phosphogluconic Acid Dehydrogenase (6P.G.D.H.). These allozymes were chosen because they have a high success rate of staining in a broad array of animal species, and several overlapped with those used in previous nudibranch allozyme inves- tigations (Havenhand et al., 1986 and Morrow et al., 1992). The six allozymes which were scored under these gel conditions were Superoxide Dismutase, Tri-Peptidase 1 and 2, Malate Dehydrogenase, Phosphoglucose Isom- erase, and Phosphoglucomutase. SPECIES DESCRIPTIONS Doriopsilla albopunctata (Cooper, 1863) (Figures 1A,B, 2A,D, 3A,B, 4A) Doris albopuntata Cooper 1863: 58. Doriopsis reticulata Cockerell in Cockerell & Eliot, 1905: 41-42, pl. 7, fig. 5. Doriopsis fulva MacFarland, 1905: 45. (see McDonald, 1983 for complete synonymy) Distribution: Known from Puerto Pefiasco and Bahia de los Angeles, Gulf of California, México; Bahia Tortugas, Baja California Sur, México to Van Damme State Beach, Mendocino County, California (Marcus & Marcus, 1967; Behrens, 1991; present study). The Veliger, Vol. 42, No. 3 Figure 1 Living animals. A. Doriopsilla albopunctata (Cooper, 1863), specimen from Monterey, California showing low dorsal spot density. B. Doriopsilla albopunctata (Cooper, 1863), specimen from San Diego, California, showing high dorsal spot density. C. Doriopsilla gemela Gosliner, Schaefer & Millen, sp. nov., specimen from San Diego, California. Material examined: One specimen, dissected, CASIZ 111388, intertidal zone, Carmel Point, Monterey County, California, 29 July 1996, M. Schaefer, coll.. One speci- men, dissected, CASIZ 111389, intertidal zone, Pillar Point, San Mateo County, California, 3 June 1996, T. T. M. Gosliner et al., 1999 Gosliner, coll. One specimen, dissected, CASIZ 111390, Hill Street, San Diego, California, | August 1996, M. Schaefer, coll. One specimen, dissected, CASIZ 111391, Punta Gringa, Bahia de los Angeles, Baja California, México, March 1997, H. Bertsch, coll. One specimen, dissected, CASIZ 112215, Punta Gringa, Bahia de los Angeles, Baja California, México, March 1997, H. Bertsch, coll. Two specimens dissected, intertidal zone, Pacific Grove, Monterey County, California, June 1976, S. Millen, coll. Two specimens dissected, 23—27 m depth, Scripps Canyon, La Jolla, California, 1 September 1996, M. Miller, coll. Two specimens, CASIZ 071498, intertidal zone, Centro de Aquacultura, Bahia Tortugas, Baja Cal- ifornia Sur, México, 29 June 1984, T. Gosliner, coll. Six specimens, one dissected, CASIZ 072134, 20-23 m depth, Roca Ben, Baja California, México, 20 August 1987, R. Van Syoc, T. Gosliner, coll. External morphology: The living animals (Figure 1A,B) reach a maximum of 60 mm in length. The body color ranges from bright yellow, orange to chestnut brown. The dorsal surface is ornamented with opaque white spots, some of which are present in the center of conical tuber- cles. The tubercles are 0.6—1.0 mm in diameter. The spots in the center of the tubercles are small glands, which are bordered by spicules. The size and density of tubercles varies greatly, within and between localities. Specimens from northern California are usually bright yellow throughout, but may occasionally have a central patch of chestnut brown on the dorsum. Specimens from southern California exhibit much more variation in color than do northern California specimens, but are generally much darker in color. The rhinophores are orange-yellow to yel- low, with 11—30 lamellae. There are five to six bi- or tripinnate gills which are white to pale yellow in color. The notum is densely spiculate. The foot is elongate, but is generally completely covered by the posterior end of the mantle. The head is poorly developed with minute ridges. Internal morphology: The oral tube (the presumed ho- molog of the buccal mass in other dorids that possess a radula) lacks any vestige of a radula. It (Figure 2A) is elongate and tubular and can be wider posteriorly when the tube is not fully extended. It passes through the an- terior nerve ring, and forms a junction with the esopha- gus. Posteriorly, the esophagus is narrower than the oral tube and is uniformly cylindrical and glandular. From the junction of the oral tube and the esophagus, immediately anterior to the nerve ring, a pair of muscles emerges and joins the nerve ring. A second pair of muscles attaches to the base of the buccal ganglia and traverses the length of the glandular portion of the esophagus and joins the muscular portion of the esophagus (gizzard of Marcus & Marcus, 1967). These muscles function as retractor mus- cles for the oral tube. Posterior to the short muscular sec- tion of the esophagus is another glandular segment which Page 203 enters the stomach within the bilobed, highly digitate di- gestive gland. At the junction of the short muscular por- tion of the esophagus and this glandular segment, a sec- ond pair of retractor muscles emerges and connects pos- teriorly with connective tissue near the gills. The intestine emerges between the two lobes of the digestive gland. Here the widest portion of the intestine has several lat- erally directed glandular lobes. At this point, a short rounded pyloric caecum extends from the intestine, where it is situated ventrally. The intestine narrows and contin- ues posteriorly to the anus, which is situated to the far left side, between the left lateral branchial plumes. The reproductive system (Figure 2D) is triaulic. A short preampullary duct widens into an elongate, cylin- drical ampulla. The ampulla divides into a short oviduct, which enters the female gland mass and the more elon- gate vas deferens. The proximal portion of the vas defer- ens expands into a wide, flattened, lobed prostatic portion which envelops most of the bursa copulatrix. From the distal end of the prostatic portion, the vas deferens nar- rows abruptly, then gradually widens and curves into the penial bulb. The penial bulb lacks a distinct penial papilla and contains approximately 16 rows of curved, acutely pointed penial hooks which are approximately 25—35 ym wide at the base and up to 50 wm in length (Figure 3A,B). The vaginal opening, like the penial bulb, is also narrow. The vagina is relatively narrow and straight. At its prox- imal end is a large, thin-walled, spherical bursa copula- trix. The slender receptaculum seminis is extremely elon- gate and extends beyond the proximal end of the bursa copulatrix. The duct of the receptaculum seminis joins the vagina proximally, near the base of the bursa copu- latrix. Near this junction, the uterine duct emerges and joins the female gland mass. The female gland mass is large and completely developed in all specimens exam- ined. Developmental biology: Egg ribbons (Figure 4A) are in the form of a long, narrow, spiral ribbon attached on one edge, consisting of one to three whorls. The ribbon is crenulate along its free edge and may be 2—4 mm in height. This ribbon shape is classified as type A (Todd, 1983). Egg laying in Monterey Bay occurs throughout the year, with an increase in the summer months (Mac- Farland, 1906). The ribbon size, height, and number of whorls are variable, and are dependent on the size of the adult which produced it. Egg ribbon color can be yellow, orange, or off-white. These ribbons almost always have a single egg per capsule. The occurrence of two larvae per egg is rare. Capsules vary from 180 to 240 wm across, with larvae initially measuring 100-150 wm across. Planktotrophic larvae with type B shells (Todd, 1981), previously known as type 1 (Thompson, 1961) hatch after 31 days at 14°C. Genetic variation: To ascertain genetic variability within and between populations of Doriopsilla albopunctata, 12 04 The Veliger, Vol. 42, No. 3 Figure 2 A, Doriopsilla albopunctata (Cooper, 1863), oral tube and esophagus of specimen from Bahia de los Angeles (CASIZ 112215), with blood gland and central nervous system removed, c = caecum; d = digestive gland; g = glandular portion of esophagus; i = intestine; m = muscular portion of esophagus; o = oral tube, scale = 3 mm. B. Doriopsilla gemela Gosliner, Schaefer & Millen, sp. nov., oral tube and esophagus of specimen from Bahia Tortugas (CASIZ 071505), with blood gland and central nervous system removed, c = caecum; d = digestive gland; g = glandular portion of esophagus; i = intestine; m = muscular portion of esophagus; o = oral tube, scale T. M. Gosliner et al., 1999 individuals from Hill Street, San Diego County were compared with 11 individuals from Carmel Point, Mon- terey County; the two localities are 685 km apart. Six loci were examined for allozymatic variation (Schaefer, in preparation). The genetic identity measure, I (Nei, 1972), within this species for the two different popula- tions is 0.950, using allozyme comparisons. The I value ranges from | (if the populations are identical) to 0 (no common alleles). Conspecific populations have I values above 0.9 among varied plants and animals (Thorpe, 1983; Nei, 1987). The genetic distance, D (Nei, 1972), was 0.052 between the two populations. Values for D vary from O (if the populations are identical) to infinity (no common alleles). Allelic frequencies at both sites were within expected parameters as to indicate Hardy- Weinberg Equilibrium, as determined by chi square anal- ysis. Doriopsilla gemela Gosliner, Schaefer & Millen, sp. nov. (Figures 1C, 2B,C,E, 3C—E, 4B) yellow-gilled porostome Behrens, 1980:102, fig. 146. Dendrodoris sp. A McDonald & Nybakken, 1980:54—55, fig. 57; McDonald, 1983:171. Dendrodoris sp. 1 Behrens, 1991:71, fig. 130. Distribution: Known from the Gulf of California, Méx- ico, from Bahia de los Angeles and along the Pacific coast of North America from Bahia Tortugas, Baja California Sur, México to Elkhorn Slough, Monterey County, Cali- fornia (Behrens, 1991; present study). Etymology: The name gemela comes from the Spanish for twin, as this species is externally similar to its sym- patric congener Doriopsilla albopunctata. Type material examined: Holotype, CASIZ 111392, in- tertidal zone, Hill Street, San Diego, California, 1 August 1996, M. Schaefer, coll. Paratypes: One specimen, CASIZ 111393, same locality, date, and collector as holotype. One specimen, CASIZ 111394, same locality, date, and collector as holotype. One specimen, dissected, CASIZ 111395, same locality, date, and collector as holotype. Fifteen specimens, one dissected, CASIZ 071505, inter- tidal zone, Centro de Aquacultura, Bahia Tortugas, Baja California Sur, México, 29 June 1984, T. Gosliner, et al., coll. Three specimens, one dissected. CASIZ 074648, in- Page 205 tertidal zone, Centro de Aquacultura, Bahia Tortugas, Baja California Sur, México, 28 June 1984, S. Klontz, D. Catania, and R. Van Syoc, coll. One specimen, CASIZ 074649, 3-5 m depth, Los Morros, mouth of Bahia Tor- tugas, Baja California Sur, México, | July 1984, T. Gos- liner, coll. One specimen, CASIZ 074647, intertidal zone, Centro de Aquacultura, Bahia Tortugas, Baja California Sur, México, 1 July 1984, T. Gosliner, coll. One speci- men, CASIZ 073523, 7 m. depth, Punta Gringa, Bahia de los Angeles, Baja California México, 20 September 1991, T. Gosliner, coll. Seven specimens, two dissected, CASIZ 074642, intertidal zone, Centro de Aquacultura, Bahia Tortugas, Baja California Sur, México, 2 July 1984, H. Bertsch, coll. One specimen, CASIZ, 071661, 9 m. depth, Punta Gringa, Bahia de los Angeles, Baja California México, 30 June 1987, S. Millen, coll. External morphology: The living animals (Figure 1C) reach a maximum of 40 mm in length. The body color is bright yellow to orange or orange-brown. The dorsal sur- face appears smooth but has some minute tubercles, 0.20—0.24 mm in diameter. The notum is ornamented with small opaque white spots. The rhinophores are or- ange-yellow to yellow, with 7—10 lamellae. There are five to seven bi- or tripinnate gills, which are bright yellow to orange in color. The notum is densely spiculate. The foot is elongate, but is generally completely covered by the posterior end of the mantle. The head is poorly de- veloped with triangular, furrowed tentacles. Internal morphology: The buccal mass lacks any vestige of a radula. The oral tube (Figure 2B, C) is elongate and tubular. It is widest posteriorly at its junction with the narrow, glandular esophagus. The esophagus narrows at the point where it passes through the anterior nerve ring. Posteriorly it is cylindrical, and widens gradually forming one or more loops. The granular surface appears to con- tain glandular cells. Posterior to the elongate glandular segment is an elongate, curved muscular portion of the esophagus, which widens into a short, rounded glandular section immediately anterior to where it joins the stomach within the bilobed digestive gland. The two lobes of the digestive gland are well separated from each other, and their outer edges are partially subdivided by vertical par- titions in the body wall. The intestine emerges between the two lobes of the digestive gland. At this point a short rounded pyloric caecum extends dorsally from the intes- = 2 mm. C. Doriopsilla gemela Gosliner, Schaefer & Millen, sp. nov., oral tube and esophagus of specimen from Hill Street, San Diego (CASIZ 111395) with blood gland removed g = glandular m = muscular portion of esoph- agus; o = oral tube, scale = 1 mm. D. Doriopsilla albopunctata (Cooper, 1863), reproductive system of specimen from Pillar Pt. (CASIZ 111389), a = ampulla; be = bursa copulatix; f = female gland mass; p = penis; pr = prostate; rs = receptaculum seminis; v = vagina, scale = 2 mm. E. Doriopsilla gemela Gosliner, Schaefer & Millen, sp. nov., reproductive system of specimen from Hill Street, San Diego (CASIZ 111395), a = ampulla; be = bursa copulatix; f = female gland mass; p = penis; pr = prostate; rs = receptaculum seminis; v = vagina, scale = 1 mm. Page 206 The Veliger, Vol. 42, No. Figure 3 Penial spines. A. Doriopsilla albopunctata (Cooper, 1863), from Pillar Point, San Mateo County, (CASIZ, 111389), showing entire width of vas deferens, scale = 75 wm. B. Doriopsilla albopunctata (Cooper, 1863), isolated spines, scale = 30 um. C. Doriopsilla gemela Gosliner, Schaefer & Millen, sp. nov., from Bahia Tortugas (CASIZ 071505), entire width of vas deferens, scale = 30 pm. D. Doriopsilla gemela Gosliner, Schaefer & Millen sp. nov., from Bahia Tortugas (CASIZ 071505), vas deferens, scale = 15 wm. E. Doriopsilla gemela Gosliner, Schaefer & Millen, sp. nov., from Bahia Tortugas (CASIZ 071505), isolated penial spine, scale = 7.5 wm. T. M. Gosliner et al., 1999 Figure 4 Egg masses. A. Doriopsilla albopunctata (Cooper, 1863), from Carmel Point, California. B. Doriopsilla gemela Gosliner, Schae- fer & Millen, sp. nov., from Hill Street, San Diego, California. tine. The intestine narrows and continues posteriorly to the anus, which is on the left side, situated between the left lateral branchial plumes. The reproductive system (Figure 2E) is triaulic. A short preampullary duct widens into an elongate, cylindrical ampulla. The ampulla divides into a short oviduct, which enters the female gland mass and the more elongate vas deferens. The proximal portion of the vas deferens widens into a wide, highly digitate, flattened prostatic portion, which envelops most of the bursa copulatrix. From the distal end of the prostatic portion the vas deferens nar- rows abruptly into the elongate, convoluted ejaculatory portion which is highly muscular. The ejaculatory portion of the vas deferens remains relatively narrow throughout its length and does not widen at the penial bulb, which Page 207 is relatively short. The penis contains approximately six rows of curved, acutely pointed penial hooks which are approximately 10—15 «zm wide at the base and 15—20 pm in length (Figure 3C—E). The vaginal opening is enlarged and muscular. The vagina is relatively narrow and straight. At its proximal end is a large, thin-walled, spher- ical bursa copulatrix. The slender receptaculum seminis is extremely elongate and curved. The duct of the recep- taculum seminis joins the vagina distally, near the base of the vaginal duct. From the middle of the receptaculum duct, the uterine duct emerges and joins the female gland mass. The female gland mass is large and completely de- veloped in all specimens examined. Developmental biology: Egg ribbons (Figure 4B) are flat, transparent spirals consisting of three whorls and containing yellow eggs. A 15 mm adult produced an 8 mm ribbon with 2000 eggs. Each capsule has one egg, 240 wm wide. The larvae initially almost fill this capsule. Eggs can vary from 120 to 300 wm wide. The occurrence of two larvae per egg is rare. Lecithotrophic larvae hatch after 31 days at 14°C, with type B shell shape (Todd, 1981) previously known as type 1 (Thompson, 1961). Some species of opisthobranchs exhibit poecilogeny and can change their developmental strategy from lecitho- trophic larvae or direct development to planktotrophic larvae in response to adult starvation or other environ- mental variation (Clark & Goetzfried, 1978). Animals from the Hill Street population were subjected to star- vation but did not demonstrate any change in reproduc- tive strategy. Population genetics: The genetic identity measure, I was 0.374 between the species Doriopsilla albopunctata and Doriopsilla gemela. Both species were collected from Hill Street, San Diego County for the interspecific allo- zyme comparison. Typically, populations of congeneric species have I values from 0.3 to 0.8 (Thorpe, 1983; Nei, 1987). The genetic distance value, D, was 0.983 between these two species. Allelic frequencies were within ex- pected parameters as to indicate Hardy-Weinberg Equi- librium, as determined by chi square analysis, with the exception of MDH, which occurred as three homozygotic alleles with no detected heterozygotes. This is significant at the 0.005 level, as determined by chi square analysis. Details of methodology and banding patterns are pre- sented by Schaefer (in preparation). DISCUSSION The systematics of the genera within the Dendrodorididae has been historically the subject of considerable confu- sion. The names Dendrodoris Ehrenberg, 1831, Doriopsis Pease, 1860, Doridopsis Alder & Hancock, 1864, and Doriopsilla Bergh, 1880, have been applied to various members of the family. Pruvot-Fol (1930) showed that Doridopsis is a junior synonym of Dendrodoris, but con- Page 208 sidered Doriopsis to be a member of the Archidorididae. Subsequently, Doriopsis has been placed in the Dorididae (Kay & Young, 1969). Regardless of its familial place- ment, Doriopsis has a well-developed radula and pecti- nate gills and is clearly not a dendrodorid. Eliot (1906) recognized the distinction between Den- drodoris (as Doridopsis) and Doriopsilla, with the former having elongate buccal nerves and more posteriorly sit- uated buccal ganglia. Valdés & Ortea (1997:240) ques- tioned this distinction and stated that Eliot was incorrect in stating that the buccal ganglia in Doriopsilla were sit- uated anteriorly. However, they did not describe or illus- trate the position of the buccal ganglia of any of the spe- cies they described. The differences in buccal ganglion position noted by Eliot has been confirmed by other au- thors. Marcus (1957:fig.) and Edmunds (1971:figs. 21d, 22d) have shown the more posterior position of the gan- glia in several species of Dendrodoris. Several other au- thors (Marcus, 1961:fig. 19; Marcus & Marcus, 1967:fig. 62; Gosliner, 1991:fig. 8) have indicated that the position of the buccal ganglia in species of Doriopsilla is within the anterior nerve ring. The two species of Doriopsilla examined here also have the buccal ganglia situated in the nerve ring rather than more posteriorly. Most subse- quent workers have considered the two genera as distinct with the exception of Thompson (1975:500) who stated that ‘‘the distinction is based upon several features of the morphology which appear to me inadequate,” but pro- vided no further details. Most recently, Valdés (1996) provided phylogenetic evidence that species of Doriop- silla and Dendrodoris form monophyletic sister clades. The present species possess the synapomorphic features of an eccentic anus, a flattened prostate, and penial spines with an elongate base, which characterize members of Doriopsilla (Valdés & Ortea, 1997). Systematic confusion has surrounded the systematics of the Dendrodorididae from the Pacific coast of North America. Cooper (1863) described Doris albopunctata from Santa Barbara, California. Only the external color- ation was described and no type material is extant (MacFarland, 1905). MacFarland (1905) later described Doriopsis fulva from Monterey Bay, California. Similar- ly, he described only the external anatomy and coloration of the living animal. In his original description, Mac- Farland noted that D. fulva was possibly identical with Cooper’s species, but stated it was difficult to be certain, based on the superficial description and the absence of Cooper’s type material. Subsequently (1906), he de- scribed and illustrated aspects of the reproductive system of D. fulva. Cockerell & Eliot (1905) described Doridop- sis reticulata from San Pedro, California. They described the external morphology and a few aspects of the internal anatomy. They also stated that their species might be identical to Doris albopunctata Cooper. Steinberg (1961) considered these three species as synonymous, the differ- ences being based largely on a darker coloration of spec- The Veliger, Vol. 42, No. 3 imens from southern California. Roller (1970) and McDonald (1983) also considered these three names as synonyms. Behrens (1980, 1991) considered Dendrodoris fulva as distinct from Doriopsilla albopunctata. Our ex- amination of specimens of Doriopsilla albopunctata from Baja California to northern California revealed consider- able variation in the body color, similar to that described by Steinberg (1961). Specimens from northern California are light yellow throughout, but may occasionally have a central patch of chestnut brown on the dorsum. Speci- mens from southern California exhibit much more vari- ation in color than do northern California specimens, but are generally much darker in color. There were no other significant anatomical differences between specimens of different color forms or from different localities. All spec- imens with white gills produced egg masses with a spiral attached to the substrate by its inner edge. All of these egg masses yielded planktotrophic larvae. Analysis of al- lozyme frequencies yielded no significant differences within or between widely separated populations. Analysis of morphological, developmental and genetic data support the conclusions of Steinberg (1961), Roller (1970), and McDonald (1983), that Doris albopunctata Cooper, 1863, Doriopsis fulva MacFarland, 1905, and Doridopsis retic- ulata Cockerell & Eliot, 1905, represent a single species with Cooper’s name having priority as the senior syno- nym. There is no doubt that the material described by Cockerell & Eliot (1905) and MacFarland (1906) are con- specific with the present material. Cockerell & Eliot de- scribed an elongate receptaculum seminis (as “‘sperma- tocyst’’) as in the present material. MacFarland (1906:fig. 38) illustrated part of the reproductive system of Doriop- sis fulva with a receptaculum seminis (as “‘spermato- cyst’’) which exceeds the length of the bursa copulatrix (as “‘spermatotheca’’) and which enters the vagina prox- imally, as in the present material. Material that Behrens (1980, 1991), McDonald & Ny- bakken (1980), and McDonald (1983) considered as a distinct species is conspecific with Doriopsilla gemela. Doriopsilla gemela is morphologically distinct from D. albopunctata. Externally, specimens of D. gemela have deep yellow or yellow-orange gills, whereas those of D. albopunctata are white or pale yellow. There are fewer rhinophoral lamellae (7-10) in D. gemela than in D. al- bopunctata (11-30). The larger tubercles of D. albopunc- tata contain glands, whereas the smaller ones of D. ge- mela do not. The remainder of the external anatomy is extremely similar between the two species. The internal anatomy of the two species differs con- sistently in many significant regards. The glandular por- tion of the esophagus of D. gemela is more elongate and convoluted than in D. albopunctata. More posteriorly, D. gemela has an elongate muscular portion of the esopha- gus, while that of D. albopunctata is short. The digestive gland lobes are well separated and lobed on the outer edges in D. gemela, while they are partially fused in D. T. M. Gosliner et al., 1999 albopunctata. The intestinal caecum is readily visible in D. gemela, but is more ventral and obscured by the glan- dular portion of the intestinal lobes in D. albopunctata. There are also consistent differences in the reproduc- tive anatomy of the two species. The ejaculatory portion of the vas deferens of D. gemela is elongate and consists of many convolutions, whereas in D. albopunctata the ejaculatory segment is shorter and thicker and without convolutions. In D. gemela there are about six rows of small penial spines that are 15—20 wm in length. In con- trast, in D. albopunctata there are about 16 rows of spines that are about 50 ym in length. The receptaculum seminis of both species is fairly elongate, but in D. albopunctata it is more elongate and extends beyond the proximal ex- treme of the bursa copulatrix. More significantly, the duct of the receptaculum seminis of D. gemela enters the distal portion of the vagina, while in D. albopunctata it enters the proximal extreme, near the base of the bursa copu- latrix. The two species differ markedly in their developmental biology. In D. gemela the egg ribbon is flat against the substrate, while in D. albopunctata it is attached by its narrow edge and well elevated from the substrate. In D. gemela the yolky eggs develop into lecithotrophic larvae, while the larvae of D. albopunctata are planktotrophic. Genetic distances in allozyme frequencies are also con- sistent with distinct congeners. Among described species of Doriopsilla, D. gemela is unique in having the receptaculum duct enter the vaginal duct basally. The presence of an elongate muscular por- tion of the esophagus is identical to that described for D. rowena Marcus & Marcus, 1967 (Marcus & Marcus, 1967:206, fig. 62b), rather than a short muscular portion described here for D. albopunctata and previously for D. jJanaina Marcus & Marcus, 1967 (Gosliner, 1991:292, fig. 8). These differences in digestive anatomy between D. gemela and D. albopunctata suggest that despite their similarity in external morphology and coloration, they are both anatomically more similar to other species than to each other. Morphological data from other species of Do- riopsilla is necessary to develop hypotheses of phylogeny and test the suggestion that D. gemela is more closely related to D. rowena than to D. albopunctata and D. jan- aina. Valdés & Ortea (1997) suggested that Doriopsilla row- ena and D. janaina might be possible synonyms of D. areolata Bergh, 1880. However, D. janaina differs from D. areolata in having a short rather than elongate mus- cular portion of the esophagus (Gosliner, 1991). Based on the description of Marcus & Marcus (1967:fig. 62c), D. rowana differs from D. areolata in that the duct of the recepaculum seminis is separated from the bursa copu- latrix, whereas in D. areolata it enters directly at the base of the bursa (Valdés & Ortea, 1997:fig. 4b). The color patterns of these species also differ from the variation described for any of the subspecies of D. areolata. It Page 209 would appear that D. rowena and D. janaina represent distinct species. ACKNOWLEDGMENTS We thank Hans Bertsch for providing specimens and data of both species studied here, especially with respect to populations from the Gulf of California. Dave Behrens also provided field notes and data with respect to these species from central California. Elizabeth Kools kindly assisted in several aspects of the preparation of the man- uscript. Dong Lin of the Photography Department of the California Academy of Sciences kindly provided final scans of several micrographs and several prints were made by Jere Schweikert. Rich Mooi provided assistance with scanning and printing the final line drawings. Robin Lawson of the Osher Molecular Lab kindly assisted in all aspects of the genetic work. LITERATURE CITED BEHRENS, D. W. 1980. Pacific Coast Nudibranchs, a Guide to the Opisthobranchs of the Northeastern Pacific. Sea Challengers: Los Osos, California. 112 pp. BEHRENS, D. W. 1991. Pacific Coast Nudibranchs, a Guide to the Opisthobranchs. Alaska to Baja California. 2nd ed. Sea Challengers: Monterey, California, 107 pp. CLARK, K. B. & A. GOETZFRIED. 1978. Zoogeographic influences on development patterns of North Atlantic Ascoglossa and Nudibranchia, with a discussion of factors affecting egg size and number. Journal of Molluscan Studies 44(3):283-294. COcKERELL, T. D. A. & C. ELtot. 1905. Notes on a collection of Californian nudibranches. Journal of Malacology 12(3):31— 53% Cooper, J. G. 1863. On new or rare Mollusca inhabiting the coast of California. 2. Proceedings of the California Academy of Natural Sciences 3:56—60. Epmunps, M. 1971. Opisthobranchiate Mollusca from Tanzania (Suborder: Doridacea). Zoological Journal of the Linnean Society 50(4):339-396. EvioT, C. N. E. 1906. The genus Doriopsilla Bergh. Journal of Conchology 11(12):366—367. GOSLINER, T. M. 1991. The opisthobranch gastropod fauna of the Galapagos Islands. Pp. 281-305 in M. James (ed.), Gala- pagos Marine Invertebrates. Plenum: New York. HAVENHAND, J. N., J. P. THoRPE & C. D. Topp. 1986. Estimates of biochemical genetic diversity within and between the nu- dibranchs molluscs Adalaria proxima (Alder and Hancock) and Onchidoris muricata (Doridacea: Onchidorididae). Jour- nal of Experimental Marine Biology and Ecology 95:105— 112. Kay, E. A. & D. K. Youna. 1969. The Doridacea (Opisthobran- chia; Mollusca) of the Hawaiian Islands. Pacific Science 23(2):172-231. MaAcFARLAND, F M. 1905. A preliminary account of the Dori- didae of Monterey Bay, California. Proceedings of the Bi- ological Society of Washington 18:35—54. MACFARLAND, FE M. 1906. Opisthobranchiate Mollusca from Monterey Bay, California, and vicinity. Bulletin of the Unit- ed States Bureau of Fisheries, Washington, D.C. 25:109— 151. Marcus, Er. 1957. On Opisthobranchia from Brazil 2. Journal of the Linnean Society of London, 43:390—486. Page 210 Marcus, Er. 1961. Opisthobranchia from North Carolina. The Journal of the Elisha Mitchell Scientific Society 77(2):141— PS Marcus, Er. & Ev. MArcus. 1967. American Opisthobranch Mollusks. University of Miami Institute of Marine Sciences: Miami. 256 pp. McDona_p, G. 1983. A review of the nudibranchs of the Cali- fornia coast. Malacologia 24(1—2):114—276. McDonaLb, G. & J. NYBAKKEN. 1980. Guide to the Nudibranchs of California. American Malacologists: Melbourne, Florida. 72 pp. Morrow, C. C., J. PR THORPE & B. E. PICTON. 1992. Genetic divergence and cryptic speciation in two morphs of the com- mon subtidal nudibranch Doto coronata (Opistobranchia: Dendronotacea: Dotoidae) from the northern Irish Sea. Ma- rine Ecology Progress Series 84:53—61. Nel, M. 1972. Genetic distance between populations. American Naturalist 106:283—292. Ne!, M. 1987. Molecular Evolutionary Genetics. Columbia Uni- versity Press: New York. 512 pp. RoLier, R. A. 1970. A list of recommended nomenclatural changes for MacFarland’s “Studies of opisthobranchiate mollusks of the Pacific coast of North America’. The Ve- liger 12(3):371-374. PRuvot-FoL, A. 1930. Du genre Dendrodoris Ehrenberg et de ses rapports avec le genre Doriopsis Pease et avec quelques autres. Note sur la taxonomie des nudibranches. Bulletin du Museum d'Histoire Naturelle, Paris (2)2(3):291—297. The Veliger, Vol. 42, No. 3 STEINBERG, J. E. 1961. Notes on the opisthobranchs of the West Coast of North America. 1. Nomenclatorial changes in the order Nudibranchia (Southern California). The Veliger 4(2): 57-63. THompson, T. E. 1961. The importance of the larval shell in the classification of the Sacoglossa and the Acoela (Gastropoda: Opisthobranchia). Proceedings of the Malacological Society of London 34(5):233—238. THompson, T. E. 1975. Dorid nudibranchs from eastern Australia (Gastropoda, Opisthobranchia). Journal of Zoology 176(4): 477-517. THORPE, J. P. 1983. Enzyme variation, genetic distance and evo- lutionary divergence in relation to levels of taxonomic sep- aration. Pp. 131—152 in G. S. Oxford & D. Rollingson (eds.), Protein Polymorphism: Adaptive and Taxonomic Signifi- cance. Academic Press: London. Topp, C. D. 1981. Ecology of nudibranch molluscs. Oceano- graphic Marine Biology 19:141—234. Topp, C. D. 1983. Reproductive and trophic ecology of nudi- branch molluscs. Pp. 225—255 in W. D. Russell-Hunter (ed.), The Mollusca. Ecology. Vol. 6. Academic Press: New York. VALDES, A. 1996. Revisio6n de la superfamilia Porodoridoidea Odhner en Franc, 1968 (Mollusca: Nudibranchia) en el Océano Atlantico. Ph.D. thesis, Universidad de Oviedo. VALDES, A. & J. ORTEA. 1997. Review of the genus Doriopsilla Bergh, 1880 (Gastropoda: Nudibranchia) in the Atlantic Ocean. The Veliger 40(3):240—254. The Veliger 42(3):211-217 (July 1, 1999) THE VELIGER © CMS, Inc., 1999 Observations on the Embryonic Development of Octopus mimus (Mollusca: Cephalopoda) from Northern Chile K. WARNKE Zentrum fiir Marine Tropenokologie, Fahrenheitstr. 1, 28359 Bremen, Germany Abstract. The embryonic development of Octopus mimus Gould, 1852, was studied under normal upwelling tem- perature conditions (16°C) and under conditions of medium and strong El Niftio Southern Oscillation (ENSO) events (20°C and 24°C, respectively). The embryonic development under high temperature conditions is faster than at lower temperature. Embryonic development of Octopus mimus under normal upwelling temperature conditions (represented by a constant temperature of 16°C) takes about 35% more time than under conditions of medium ENSO events (at a constant temperature of 20°C), and 62% more on average than under conditions of strong ENSO events (at a constant temperature of 24°C). There were no abnormalities visible on the embryos developed at 24°C. The hatching rate was high (estimated at about 95%). The higher temperature had no adverse effect on the viability of the hatchlings. This suggests genetic fixation of a wide temperature tolerance. The embryonic development of Octopus mimus is very similar to that of O. vulgaris Cuvier, 1797. However, egg and hatchling size, number of gill lamellae per demibranch, and heartbeat frequencies differ between the two species. INTRODUCTION Octopus mimus Gould, 1852, is common along the Chi- lean coast from 18°S to 37°S (Osorio et al., 1979). This octopus is an important resource for artisanal fisheries in the northern part of Chile. From 1978 to 1994 the total catch increased from 4 tons to 3732 tons (Servicio Na- cional de Pesca, 1994). During El Nino Southern Oscillation (ENSO) condi- tions in 1982/1983, the population size of Octopus mimus increased significantly, in contrast to most other inverte- brates of this region which died out or were significantly reduced (Arntz & Fahrbach, 1991). Near Antofagasta (northern Chile) the population of O. mimus increased by a factor of 100 (Tomicic, 1985). Little is known about the life cycle of Octopus mimus. Size at maturity and the reproductive cycle were studied by Arancibia (1984) and Cortez et al. (1995b). Wolff & Perez (1992) investigated aspects of population dynamics, food consumption, and conversion efficiency, and Cortez et al. (1995a) observed feeding dynamics. Octopus mimus was long synonymized with O. vulgaris Cuvier, 1797, and has only recently been recognized again as a separate species (Guerra et al., personal communication). Apart from a few pictures of an egg mass given by Cortez (1995), nothing has been published on the embryonic de- velopment of O. mimus. The present study provides de- tails of the embryonic development, under normal up- welling conditions (16°C) and under conditions of medi- um and strong ENSO influence (temperature 20°C and 24°C, respectively). The aims of this study are (1) the assessment of the influence of ENSO type temperature changes on embryonic development, and (2) to define the differences in embryonic development between O. mimus and O. vulgaris. METHODS Adult Octopus mimus were collected off the coast of northern Chile in the region of Iquique, by SCUBA div- ing at depths of 5—10 m. Dorsal mantle length (measured from the midpoint between the eyes to the apex of the mantle tip), head width, and weight were determined in all adult females. In the field, the main spawning season of O. mimus is between November and March, although egg laying is observed throughout the year. Animals were maintained in 500 L tanks at the Departamento Ciencias del Mar (Universidad Arturo Prat). All octopuses were kept in the laboratory under constant temperature condi- tions of 16°C, 16.5°C, 20°C, and 24°C (+1°C) with a slow continuous flow of clean seawater resulting in a daily renewal of the whole water volume. The nitrate content of the aquarium water was moni- tored following the recommendation of Boletzky & Han- lon (1983). When a high nitrate concentration was ob- served, the aquarium water was partly changed until ni- trate could no longer be detected. Females were kept isolated in covered tanks. Each tank contained a clay flower pot as a hiding place for the fe- male, providing her with a substrate on which she could attach her eggs and brood them as she would do in a den under natural conditions. Brooding females of Octopus mimus were fed daily at least one item from a variety of clams (Venus antiqua, Protothaca thaca, Gari solida, and Tagelus dombii), and crabs (Leptograpsus variegatus and Cancer setosus). Ve- nus antiqua was the dominant food item. The females laid their eggs in long strings or festoons. The number of strings was counted and the string length was measured. The number of eggs in | cm of string was determined, providing an estimation of total egg number per egg mass. At 3-day intervals, 3-10 eggs were re- moved from each egg mass and examined under a dis- secting microscope. Drawings were made, and photo- graphs of the eggs were taken with a camera (Nikon Sys- tem of Microflex HFM-35A-35 mm camera box M35FA) connected to the dissecting microscope (Nikon SMZ-10). A video camera (Sony video color 1 CCD model DXC107A) attached to the microscope was used occa- sionally to monitor embryos. Eggs were fixed in Bouin’s fixative (15 vols. picric acid, 5 vols. formalin, | vol. acetic acid) and preserved in 70% ethanol for later examination. Developmental stages of the embryos were identified according to Naef (1928). The rate of embryonic growth was determined by regular control measurements of egg size, yolk volume, and size of the embryo body. The standard deviation (SD) of size was calculated for each sample size (n). The time required for embryonic development was determined for the different temperatures, along with the frequency of heartbeat at advanced stages. Embryonic mortality within an egg mass was estimated based on the daily observation records. In addition, embryonic development was de- scribed and compared to the observations reported for other species of Octopus (Boletzky, 1967, 1969, 1971a, b, 1987, 1989; Fioroni, 1978; Hochberg et al., 1992; Joll, 1978; Mangold-Wirz, 1983). The size of hatchlings, total length, dorsal mantle length, and head width were measured following Hoch- berg et al. (1992) and standard deviations were calculat- ed. For close observation of behavior, groups of 100 hatchlings each were maintained in six small glass aquar- ia equipped with air bubblers surrounded by fine mesh to avoid damage to the hatchlings. Each aquarium contained 55 L of still seawater, which was changed once a day. The hatchlings were fed various planktonic organisms (mainly larvae of Pagurus sp. and of Cancer setosus). First feeding was observed through the transparent aquar- ium wall and was subsequently validated by close in- spection of the digestive tract of the observed individuals under the dissecting microscope. RESULTS The smallest mature female Octopus mimus found during this study in the field had a mantle length (ML) of 120 mm and a head width of 45 mm, with a total wet weight of 868 g. The ovary weighed 165 g; the capsule length of the mature ovarian eggs measured 1.95 mm. The larg- The Veliger, Vol. 42, No. 3 est female Octopus mimus observed weighed 2818 g with a mantle length of 230 mm, and a head width of 50 mm. Brooding females were fed daily. All females survived for at least 2 weeks after hatching of the last young. Egg and string size. Egg laying was observed nearly throughout the year (see Table 1). From the egg string counts and subsamples of eggs, the average fecundity of a female Octopus mimus was estimated at between 60,000 and 200,000 eggs. Indeed each female laid about 200 strings of eggs ranging from 3-10 cm in length and con- taining an average of 100 eggs in 1 cm. The eggs of Octopus mimus are small in terms of both absolute and relative size. The chorion capsule of freshly laid eggs measured on average 2.03 mm in length (n = 50; SD = 0.25) and 0.9 mm in width (n = 30; SD = 0.08); the stalk measured 5.7 mm (mean of five measure- ments) in length (cf. Mangold-Wirz 1983: fig. 21.2 for O. vulgaris). The yolk mass on average was 1.7 mm long (n = 27; SD = 0.17) and 0.8 mm wide (n = 27; SD = 0.09). The increase of chorion length through embryonic development was similar in all egg masses except “C” and ‘“‘F” (Table 1). The value for egg mass “C”’ is not significant, however, given the small sample size (see Ta- ble 1). In this case, the initial chorion length was rela- tively small compared to the other egg masses. Thus the average increase of chorion length of all egg masses, ex- cluding egg mass ‘“‘C’’, was about 11%. There was no correlation between magnitude of size increase and tem- perature. The hatchlings measured on average 2.34 mm (n = 60 individuals; SD = 0.19) in total length. The range of variation in the hatching size was 2.1—2.6 mm total length. Hatchlings had an average mantle length of 1.85 mm (n = 30; SD = 0.08) and a head width of about 0.84 mm (n = 30; SD = 0.09). Embryonic development. The embryonic development of Octopus mimus (Figure 1A—F) turned out to be very similar to that of O. vulgaris (Boletzky, 1969, 1971a, 1989). Like other cephalopods, O. mimus develops a dis- coblastula at the animal pole of the ovum on the side of the micropyle. Stage I of Naef is defined as the end of cleavage (Figure 1A). Then the prospective yolk sac en- velope grows over the yolk surface toward the vegetal pole. At Stage VII the first relief elevation of the embryo in the mantle region is visible at the animal pole (Figure 1B). By this time, aided by the ciliary beat of the yolk sac envelope, the embryos begin to rotate around their longitudinal axis in a clockwise direction (in apical view). Eventually the direction of the ciliary beat changes and the embryos reverse their position in the chorion, as de- scribed for Octopus vulgaris (Boletzky, 1971a, b). The yolk envelope is completed at about Stage IX and the resulting outer yolk sac begins to pulsate irregularly (about seven beats per minute). At Stage X to XI the arm buds are conspicuous but the mantle rudiment is still flat (Figure 1C). The inner yolk sac shows two posterior lobes at about Stage XII. In the depression between these lobes K. Warnke, 1999 Page 213 Table 1 Duration of embryonic development to the day of hatching and growth of Octopus mimus during embryonic development under the influence of different temperatures (n Duration of embryonic Time develop- to first ment to regular the day of Temperature Egg mass heartbeat hatching CC) (date) (days) (days) 24 A 18 725) 5/8/95-6/2/95 autumn 20 B 23 38 9/18/94-10/26/94 spring 20 Cc 28 43 12/12/94-1/24/95 summer 20 D 22 37 3/20/95-4/26/95 autumn 16.5 13 46 63 4/28/95-6/30/95 autumn 16 F 44 67 2/2/95-4/10/95 summer/autumn = sample size; SD = standard deviation). Length of egg at beginning of embryonic development (mm; mean value) Increase in chorion length during embryonic development Total length of hatchling (mm; mean value) eI APTS} 5.7% n= 10 n = 30 SD = 0.11 SD = 0.20 2.28 DSi 4.0% n 7 n 10 SD = 0.22 SD = 0.44 1.69 1.96 15.98% n= 2 n= SD = 0.58 PL PA 2.42 6.6% n 10 n 9 SD = 0.05 SD = 0.09 2.06 2.15 4.3% n= 10 n= 10 SD = 0.10 SD = 0.04 1.78 2.0 12.3% n= 10 n= 10 SD = 0.16 SD = 0.15 the stomach rudiment has been closed (Figure 1D). The coordination and the first regular pulsation of systemic heart and branchial heart occurs at about Stage XV; it was observed after 18 days (d) at a temperature of 24°C, after 24 d at 20°C (average of two egg masses), and after 44 d at 16°C (Table 1). By stage XV the inner yolk sac is very reduced (Figure 1E) as in other octopods (Boletz- ky, 1975; Joll, 1978); it increases again later (Figure 1F). The outer yolk sac is now clearly demarcated from the body of the embryo. Before hatching, i.e., between Stage XIX and Stage XX when the outer yolk sac has been strongly reduced in size, most of the embryos reverse their position a second time as in O. vulgaris (Portmann, 1933). There are only a few notable differences between the embryos and the hatchlings of Octopus mimus and O. vulgaris. The hatchlings of O. mimus have seven (Figure 1G) gill lamellae per demibranch, while those of O. vul- garis have only five (Boletzky, 1969). The appearance of pigment in the ink sac is somewhat earlier in O. mimus (Stage XVID) than in O. vulgaris (Stage XVIII). The cho- rion stalk length in relation to the chorion capsule length of O. mimus (X2.8) is also slightly higher than in O. vulgaris (X2.5) (Boletzky, personal communication; cf. Mangold-Wirz, 1983: fig. 21.2 for O. vulgaris). Embryonic development in Octopus mimus takes about 65 d under normal upwelling temperature conditions (rep- resented by a constant temperature of 16°C), about 39 d under conditions of medium ENSO events (at a constant temperature of 20°C), and 25 d on average under condi- tions of strong ENSO events (at a constant temperature of 24°C) (Table 1 and Figure 2). Thus development du- ration at 24°C was about 60% of that observed at 20°C, which was about 65% of the development time at 16°C. The time lapse between first and last egg laying and between first and last hatching was between 5 and 14 d, respectively, independent of the rearing temperature. Due to the high room temperature, it was not possible to maintain a constant low water temperature in the Petri dishes used for determination of the heartbeat frequency in embryos when observed under a dissecting micro- scope. For example, an egg mass kept at an average brooding temperature of 20°C contained embryos show- ing between 44 and 26 heartbeats/min. Egg mass ‘“‘E”’ was kept at a temperature of 16.5°C. At Stage XIX, the heartbeat of an embryo of this egg mass showed a fre- quency of about 52 pulsations/min. One day later, at a constant temperature of 22°C, the same embryo had a frequency of 94 pulsations/min. This variation is most likely due to the temperature sensitivity of heartbeat rate. Nevertheless, heartbeat frequency increased with de- velopmental progress in general, until it became relatively Page 214 The Veliger, Vol. 42, No. 3 Yolk envelope Inner yolk sac Central digestive & Outer yolk sac circulatory organs K. Warnke, 1999 Stage XX 7 XV 0 10 20 30 40 50 60 Time in days Figure 2 Course of embryonic development in Octopus mimus at three different temperatures (stages according to Naef, 1928). stable from Stage XV onward. In autumn (March until May 1995), the environmental temperature of 20°C cor- responded to the brooding temperature of the egg mass “TJ” during the entire time of embryonic development. At Stage XIII, the average heartbeat was 28 pulsations/min. At Stage XVIII, the heartbeat was about 77 pulsations/ min. The estimated mortality rate of egg masses reported in this paper was low (about 5-20%). The hatching rate of egg masses ‘“‘B”’ (20°C) and ‘A’ (24°C) was nearly 100%. Egg mass “‘E”’ had fungi on the chorion surfaces, and the mortality rate was more than 50%. After Stage XV, some abnormal stages were visible within this egg mass. The inner yolk sac increased to an abnormally large size. Therefore the data from these embryos are not in- cluded in this paper. Pigmentation. The retina became light orange at Stage X, turned bright red at Stage XIJI-XIV (Figure 1E), and finally became dark brown at Stage XVII/XVIII (Figure 1F). Page 215 Figure 3 Schematic presentation of chromatophore distribution in O. mi- mus (A = dorsal view, B = ventral view). Chromatophores appeared at Stage XV on the head, arms, ventral mantle surface, and on the dorsal surface of the visceral mass (bottom of dorsal mantle cavity). The hatchling chromatophore pattern was nearly complete at Stage XIX (Figure 1F). Hatchlings (Figure 3) have a total of 75-89 chromatophores. Every arm has two to four chromatophores on the outside in a simple row. The dor- sal surface of the digestive complex has six to eight large visceral chromatophores. The dorsal head surface has 9— 10 chromatophores in a 2 + 4 + 4 pattern and one large chromatophore per eye. The ventral head surface has two large chromatophores. The funnel shows five (six) chro- matophores in a 3 (4) + 2 pattern. The ventral mantle is evenly covered with 21—24 chromatophores. Hatchlings. The hatchling size (ML 1.85 mm) is less than 2% (1.15%) of the size of the adult animals (ML 175 mm). Like the hatchlings of other octopodid species producing relatively small eggs (hatchling ML smaller than 8% of adult ML; Boletzky, 1984), the mode of life Figure 1 Living embryos of Octopus mimus (egg size 2 X 1 mm) at different stages of development according to Naef (1928). A. Stage I at the animal pole of the chorion: the end of blastulation is visible (arrow). B. Stage VII: yolk envelope is growing (4/5) over the yolk toward the vegetal pole. At the animal pole the first relief in the mantle region is visible. C. Stage X to XI: dorso-ventral view: after first reversion of the embryo, midgut gland region with huge yolk papilla: clasp of midgut is still open, arm buds conspicuous, mantle rudiment still flat, beginning of retina pigmentation. D. Around Stage XII: arm buds are still rounded, not pointed, inner yolk sac still with two lobes, in the indentation between the lobes the stomach has been closed; the retina is well pigmented, behind the cheek hump the funnel pouches are visible. E. Both animals at Stage XV: lateral view: oval rhomboid retina (red retinal pigment); dorsal view of buccal complex between arms, right dorsal arm with sucker rudiments, inner yolk sac reduced, at this stage regular heartbeats. F Stage X VIII/XIX: dorsal view: shortly before the second reversion, six to eight chromatophores on dorsal surface located on visceral mass (bottom of dorsal mantle cavity), dorsal head with 10 chromatophores in a 2 + 4 + 4 pattern, each arm with a single row of two to four chromatophores. G. Hatchling of Octopus mimus (2.2 X 0.84 mm), note gill with seven lamellae per outer demibranch. Page 216 of hatchlings of O. mimus is initially planktonic. They have relatively short arms. The longest arms (about 0.9— 1.0 mm, n = 3) measure about one half the mantle length of the hatchling. The arms of the hatchlings are subequal in length and every arm has three suckers of similar size. After 4 d of hatching, the inner yolk sac was almost totally absorbed. In the surviving paralarvae, feeding be- gan 5 d after hatching. Larvae of Pagurus sp. and of Cancer setosus were successfully captured by the hatch- lings. Cancer setosus clearly acted as a stimulus for feed- ing as indicated by Villanueva (1994). Pagurus sp. and C. setosus are very abundant littoral species in the north- ern part of Chile and thus appear as an appropriate food source. However, survival of young Octopus mimus was very limited under aquarium conditions. The last paralar- va died 12 d after hatching. DISCUSSION Recent morphological investigations (Guerra et al., per- sonal communication; Hochberg & Mangold, personal communication) and DNA-sequencing results (Sdller et al., work in progress.) indicate that Octopus mimus and O. vulgaris are closely related, but distinct species. It ap- pears that their embryonic development is rather similar. Therefore the staging system of Naef (1928) for embry- onic development of O. vulgaris can be applied to all stages of O. mimus. The first and second reversion, the earliest pulsation of the outer yolk sac, and the beginning of the heartbeats occur at the same development stages. Also the stage when pigmentation begins to be visible in O. mimus and O. vulgaris is the same. In contrast to the chromatophore pattern of O. mimus as described by Cor- tez (1995), I found no clear difference between the chro- matophore patterns observed in O. mimus and those de- scribed by Fioroni (1965) for O. vulgaris, since there is always a relatively high natural variability in the respec- tive patterns. A clear difference was observed in the number of gill lamellae per demibranch at the time of hatching: five for Octopus vulgaris (adult: eight to ten [Boletzky, 1969]) and seven for O. mimus (adult: seven to eight [Cortez, 1995]). The hatchlings of Scaeurgus unicirrhus delle Chiaje, 1830, also have seven lamellae per demibranch of the gill; they are in the same size range, but have four instead of three suckers and more numerous chromato- phores than O. mimus (Boletzky, 1984). Another distinc- tion between O. mimus and O. vulgaris is the earlier ap- pearance of the ink in the sac in O. mimus. The arm length of O. mimus was found to range from 0.9-1.0 mm (measurements from fresh animals), which is somewhat longer than that reported for O. vulgaris (0.7 mm; see Hochberg et al., 1992). This slight difference in the arm length is probably insignificant due to the small number of individuals measured (n = 3). The chorion stalk length in relation of chorion capsule length of O. mimus (X2.8) The Veliger, Vol. 42, No. 3 is also slightly larger than that of O. vulgaris (X2.5) (Bol- etzky, personal communication; cf. Mangold-Wirz, 1983, fig. 21.2 for O. vulgaris). At Stage XVIII, the heartbeat of Octopus mimus was about 77 pulsations/min at a temperature of 20°C. This is comparable to O. tetricus Gould, 1852, with 65—75 beats/ min at a temperature of 19.5°C at Stage X VIII-XIX (Joll, 1978). The stages of the embryonic development of O. tetricus are also similar to O. vulgaris. The embryonic development of Octopus mimus under high temperature conditions is faster than at lower tem- perature. Higher temperatures appeared to be of no dis- advantage to the hatchlings. Indeed there was no visible difference in hatching success between 20°C and 24°C, the hatching rate in egg masses ‘‘B”’ (20°C) and “A” (24°C) being nearly 100%. The low survival rate of the hatchlings could be related to failures in the system used for preparing the seawater for the small culture aquaria, and/or to the limited variety of natural food items available. Whether the rather small average increase of egg size in Octopus mimus (chorion capsule length 11% for O. mimus in these experiments in contrast to 25% for O. vulgaris; Boletzky, 1969) is a nor- mal feature or reflects a less than optimal water quality remains to be seen. In any event, embryonic development appeared perfectly normal (except in egg mass “E”’; these abnormal embryos are not considered in the results of this paper). Compared to the developmental rates in Octopus vul- garis at a variety of temperatures (Boletzky, 1987), the speed of embryonic development of O. mimus is not sig- nificantly different. It is within the normal range deter- mined for other warm water octopods with planktonic young, such as O. cyanea Gray, 1849, O. tetricus and O. bimaculatus Verrill, 1883, (see Boletzky, 1969). Tomicic (1985) described an increase of the population of Octopus mimus by a factor of 100 in northern Chile during the last major ENSO event 1982-1983. Thus, O. mimus, Which is clearly adaptated to life in the cold up- welling waters off Chile, can also live under warm water conditions. This population increase may be due to sev- eral factors—environmental ones such as decrease in numbers of predators or increase of food supply—or to intrinsic ones, i.e., genetic factors making the animals more competitive, perhaps through physiological im- provement of food conversion at higher temperatures. ACKNOWLEDGMENTS This work was supported by grants from the Senator fir Bildung, Wissenschaft und Kunst, Bremen and the Gott- lieb Daimler- und Karl Benz-Stiftung. I would like to thank Dr. S. v. Boletzky for numerous suggestions during preparation of this paper. I am grateful to M. Araya Christie for his help with the experiments carried out in Chile. I also thank Prof. Dr. U. Saint-Paul K. Warnke, 1999 and Dr. E G. Hochberg for their critical reading of earlier versions of the manuscript. I gratefully acknowledge one anonymous reviewer and Dr. Rachel Collin for their con- structive suggestions which helped me to improve the manuscript. R. Fuenzalida Fuenzalida made arrangements at the Departamento Ciencias del Mar (Universidad Ar- turo Prat) in Chile, and Prof. Dr. U. Saint-Paul and S. Kadler made arrangements from Germany. R. Peredo as- sisted in rearing Octopus mimus in Chile. LITERATURE CITED ARANCIBIA, H. 1984. Estudio de talla y peso de pimera madurez sexual en el pulpo Octopus vulgaris. Pp. 1-56 in M. Wolff & H. Perez (eds.), 1992. Population Dynamics, Food Con- sumption and Conversion Efficiency of Octopus mimus Gould 1853 (Cephalopoda: Octopoda), from Antofagasta, Northern Chile. International Council for the Exploration of the Sea: CM. K/Shellfish Commitee: 29:12 pp. ARNTZ, W. E. & E. FAHRBACH. 1991. El Nino Klimaexperiment der Natur. Birkhauser Verlag: Berlin. 264 pp. BOoLeTzky, S. v. 1967. Die embryonale Ausgestaltung der friihen Mitteldarmanlage von Octopus vulgaris Lam. Revue suisse de Zoologie 74:555—562. BOLETZKY, S. v. 1969. Zum Vergleich der Ontogenesen von Oc- topus vulgaris, O. joubini und O. briareus. Revue suisse de Zoologie 76:716—726. BOLeTzky, S. v. 1971la. Rotation and first reversion in the Oc- topus embryo. A case of gradual reversal of ciliary beat. Experientia 27:558—560. BoLeTzky, S. v. 1971b. Zu den Lageveranderungen von Okto- poden-Embryonen (Mollusca: Cephalopoda). Revue suisse de Zoologie 78:538—548. BOoLeTzky, S. v. 1975. A contribution to the study of yolk ab- sorption in the Cephalopoda. Zeitschrift fiir Morphologie der Tiere 80:229—246. BOLeETzkKy, S. v. 1984. The embryonic development of the oc- topus Scaeurgus unicirrhus (Mollusca, Cephalopoda)—Ad- ditional data and discussion. Vie et Milieu 34(2/3):87—93. BOLETZKY, S. v. 1987. Embryonic phase. Pp. 5—31 in P. R. Boyle (ed.), Cephalopod Life Cycles. Vol. Il. Academic Press: London. BOLETZKY, S. v. 1989. Recent studies on spawning, embryonic development, and hatching in the Cephalopoda. Advances in Marine Biology 25:85—115. BOLETZkyY, S. v. & R. T. HANLON. 1983. A review of the labo- ratory maintenance, rearing and culture of cephalopod mol- luscs. Memoirs of the National Museum of Victoria 44:147— 187. Page 217 Cortez, T. 1995. Biologia y ecologia del pulpo comun Octopus mimus Gould, 1852 (Molusca: Cephalopoda) en aguas litor- ales del norte de Chile. Doctoral Dissertation, Universidad de Vigo, Vigo. 292 pp. Cortez, T., B. G. Castro & A. GUERRA. 1995a. Feeding dynam- ics of Octopus mimus (Mollusca: Cephalopoda) in northern Chile waters. Marine Biology 123:497—503. Cortez, T., B. G. Castro & A. GUERRA. 1995b. Reproduction and condition of female Octopus mimus (Mollusca: Cepha- lopoda) in northern Chile waters. Marine Biology 123:505— 510. FIORONI, P. 1965. Die embryonale Musterentwicklung bei einigen mediterranen Tintenfischarten. Vie et Milieu 16, Fascicule 2-A:655—756. FIORONI, P. 1978. Cephalopoda, Tintenfische. Pp. 1-180 in E Sei- del (ed.), Morphogenese der Tiere. Reihe 1, Gustav Fischer Verlag: Stuttgart, New York. HOcCHBERG, F. G., M. Nixon & R. B. TOLL. 1992. Octopoda. Pp. 213-271 in M. J. Sweeney, C. Roper, K. M. Mangold, M. R. Clarke & S. v. Boletzky (eds.), “‘Larval’’ and Juvenile Cephalopods: A Manual for Their Identification. Smithson- ian Contributions to Zoology 513. Jott, L. M. 1978. Observations on the embryonic development of Octopus tetricus (Mollusca: Cephalopoda). Australian Journal of Marine and Freshwater Research 29:19—30. MANGOLD-WIRZ, K. 1983. Octopus vulgaris. Pp. 335-364 in P. R. Boyle (ed.), Cephalopod Life Cycles. Vol. I. Academic Press: London. Naer, A. 1928. Die Cephalopoden (Embryologie). Fauna und Flora des Golfes von Neapel und der angrenzenden Meer- esabschnitte. Monographie 35(2). Verlag von R. Friedlander & Sohn: Berlin. 357 pp. Osorio, R. C., C. ATRIA & FE S. MANN. 1979. Moluscos marinos de importancia economica en Chile. Biologia Pesquera (Chile) 11:3—47. PORTMANN, A. 1933. Observations sur la vie embryonnaire de la pieuvre (Octopus vulgaris Lam.). Archives de Zoologie Ex- perimentale et Génerale 76:24—36. SERVICIO NACIONAL DE Pesca. 1994. Anuario estadistico de pes- ca. Ministerio de Economia y Construccion de Chile. Grafica A & L Impressores: Vina del Mar. 108 pp. Tomicic, J. J. 1985. Efectos del fénomeno El Nino 1982-83 en las comunidades litorales de la Peninsula de Meyjillones. In- vestigacion Pesquera (Santiago, Chile) 32:209—213. VILLANUEVA, R. 1994. Decapod crab zoeae as food for rearing cephalopod paralarvae. Aquaculture 128:143—152. WoLrFr, M. & H. PEREZ. 1992. Population dynamics, food con- sumption and gross conversion efficiency of Octopus mimus Gould 1853 (Cephalopoda: Octopoda), from Antofagasta (Northern Chile). International Council for the Exploration of the Sea: CM. K/Shellfish Commitee: 29:12 pp. The Veliger 42(3):218—248 (July 1, 1999) THE VELIGER © CMS, Inc., 1999 Bathymodiolus (Bivalvia: Mytilidae) from Hydrothermal Vents on the Azores Triple Junction and the Logatchev Hydrothermal Field, Mid-Atlantic Ridge RUDO VON COSEL Muséum National d’Histoire Naturelle, Paris, 55, rue de Buffon, F-75005 Paris, France THIERRY COMTET Centre IFREMER de Brest, Département Environnement Profond. B.P. 70, F-29280 Plouzané, France AND ELENA M. KRYLOVA Institute of Oceanology, Russian Academy of Sciences, Moscow Abstract. Bathymodiolus azoricus n. sp. is described from the Lucky Strike (31°17’N) and the Menez Gwen (37°50'N) hydrothermal fields on the Azores Triple Junction, Mid-Atlantic Ridge, and another species, Bathymodiolus sp. aff. B. puteoserpentis from Logatchev (14°45'N) hydrothermal field is treated but not named. Both species are compared with B. puteoserpentis Cosel, Métivier & Hashimoto, 1994, from the Snake Pit area (23°22'N), also on the Mid Atlantic Ridge. B. azoricus is very variable but well distinguished from Bathymodiolus sp. and B. puteoserpentis by its almost terminal umbo. The three species differ from the type species B. thermophilus principally by the absence of an inner mantle fusion and a very short valvular siphonal membrane. The slight differences between B. puteoserpentis and Bathymodiolus sp. from the Logatchev field do not warrant separation of the latter on species level. The specific ende- mism of Mytilidae along the Mid-Atlantic Ridge is discussed. INTRODUCTION Bathymodiolus species are mytilid bivalves which live in both hydrothermal vent and cold-seep environments using chemoautotrophic processes for their metabolism (Chil- dress & Fisher, 1992). Since the discovery of Bathymo- diolus thermophilus Kenk & Wilson, 1985, on the Gala- pagos Rift in 1977, 11 other species of the genus Bathy- modiolus have been described from a variety of hydro- thermal vent and cold-seep environments of both the Pacific and Atlantic oceans (Cosel et al., 1994; Hashi- moto & Okutani, 1994; Cosel & Olu, 1998; Gustafson et al., 1998). On the Mid-Atlantic Ridge, the vent mytilid Bathy- modiolus puteoserpentis Cosel, Métivier & Hashimoto, 1994, was first collected on the Snake Pit hydrothermal field (23°22'N), in June 1988, during the French HY- DROSNAKE expedition (Cosel et al., 1994). Subsequent- ly, Bathymodiolus-like mussels were found also on other localities of this ridge. The next site on the Mid-Atlantic Ridge (further ab- breviated as MAR) with a mussel population was discov- ered in September 1992 at 37°17.6’'N by an American expedition, and samples were taken by dredge (Van Do- ver, 1995). In 1993, more mytilids were collected at the same locality, then called Lucky Strike hydrothermal field (37°17'N, 1640-1700 m) by the submersible Alvin during Explanation of Figures 1—5 Figures 1—5, Bathymodiolus azoricus Cosel & Comtet, sp. nov. Figure 1. Holotype, MNHN, 111.9 mm. PP11 site, Menez Gwen hydrothermal field Mid-Atlantic Ridge, 37°50.5'N, 31°31.3’W, 866 m, DIVA 2, dive 13. Exterior and interior of both valves, dorsal view of specimen and ventral view of right valve to show the position of foot/byssus retractor scars. Figure 2. Paratype, ZMM, 94.3 mm. Same locality. Exterior of left valve. Figure 3. Paratype, MNHN, 109.7 mm. Same locality. Exterior and interior of left valve. Figure 4. Paratype, MNHN, 103.0 mm. Same locality. Exterior and interior of right valve. Figure 5. Specimen from Menez Gwen, same locality, DIVA 2, dive 11, MNHN, 70.4 mm. Exterior and interior of left valve. R. von Cosel et al., 1999 Page 219 Page 220 The Veliger, Vol. 42, No. 3 the American expedition LUCKY STRIKE 1993 (Van Dover et al., 1996). Additional populations of this mussel were found and specimens collected on a larger scale by the French submersible Nautile during the expeditions DIVA 1 (May 1994) and DIVA 2 (June 1994) of the R/V Nadir. During these cruises, also the newly discovered Menez Gwen hydrothermal field (37°50'N, 844-850 m) was studied and mussels were collected (Desbruyeres et al., 1994; Fouquet et al., 1995). Another mussel population was discovered in June 1993 at 29°10'N (3080 m) on the Broken Spur vent field during the ATLANTIS II cruise (Murton et al., 1995). The two examined shells, brought to our attention by Eve Southward, were provisionally identified as Bathymodi- olus puteoserpentis by the first author. Vent mussels were also found in July 1994 at 14°45'N by the Russian LOGATCHEV-7 expedition on board the R/V Professor Logatchev (Batuyev et al., 1994). A few mussels were sampled by TV-guided bottom grab. More material, in total 15 specimens, was taken in February 1995 by the Russian submersible Mir-2 during the cruise 35 of the R/V Akademik Mstislav Keldysch at 14°50'N. In December 1995, specimens of the same mussel were taken by the French submersible Nautile during the cruise MICROSMOKE at 14°45’N, now called the Logatchev hydrothermal vent field. The two Bathymodiolus species from 37°N (Lucky Strike and Menez Gwen hydrothermal field, Azores Tri- ple Junction) and from the Logatchev hydrothermal field were found to be different from Bathymodiolus puteoser- pentis and are described in this paper, but only one of them is introduced as a new species. Some ecological and biogeographical remarks on mytilids of the MAR are also given. MATERIALS AND METHODS Most of the studied material was collected during the al- ready mentioned French expeditions DIVA 1 and DIVA 2, organized by IFREMER (Institut frangais de Recherche pour l’Exploitation de la Mer) and MICROSMOKE, or- ganized by the CNRS (Centre National de Recherches Scientifiques). The material was sorted by the Centre Na- tional de Tri d’Océanographique Biologique (CENTOB), Brest. Shell lengths and heights were measured using the standards of Kenk & Wilson (1985:fig. 1) in a total of 7832 individuals from both Azores Triple Junction and Logatchev. Anterior part length (i.e., length from the an- terior margin to the umbo) was additionally measured on 159 individuals. Data for Bathymodiolus puteoserpentis were taken from Cosel et al. (1994) and from measure- ments of a few additional specimens. All statistical ana- lyses were carried out using StatView II or Microsoft Excel 5.0. Abbreviations used in the text: LACM—Los Angeles County Museum of Natural History, Los Angeles; MCZ—Museum of Comparative Zoology at Harvard University, Cambridge, Massachusetts; MNHN—Musé- um National d’Histoire Naturelle, Paris, France; NMNZ—National Museum of New Zealand, Wellington, New Zealand; NSMT—Natural Science Museum, Tokyo, Japan; SMF—Natur-Museum und Forschungsinstitut Senckenberg, Frankfurt/M, Germany; USNM—National Museum of Natural History, Smithsonian Institution, Washington, D.C.; ZMM—Zoological Museum of Mos- cow University. sh.—empty shell; spm.—wet preserved specimen(s); R/V—research vessel; MAR—Mid-Atlantic Ridge. SYSTEMATICS Family MYTILIDAE Genus Bathymodiolus Kenk & Wilson, 1985 Bathymodiolus azoricus Cosel & Comtet, sp. nov. (Figures 1-15, 25-33, 36, 37, 39-52, 59, 60, 62) Type material: Holotype, MNHN, Menez Gwen hydro- thermal field, Mid-Atlantic Ridge, DIVA 2 expedition, dive 13, A.-M. Alayse, observer, 15 June 1995. 22 par- atypes with preserved animal, same locality, 14 in MNHN; 1 in MCZ; 1 in NSMT; | in USNM; | in LACM; 1 in SMF; 1 in ZMM; | in Museum Funchal; | in NMNZ. Type locality: PP11 site, 37°50.5'N, 31°31.3'W, 866 m, Explanation of Figures 6—11 Figures 6-11. Bathymodiolus azoricus Cosel & Comtet, sp. nov. Figure 6. Specimen from Lucky Strike hydro- thermal field, 64.0 mm, site Statue of Liberty, 37°17.59'N, 32°16.50’W, 1635 m, “LUCKY STRIKE 1993” expe- dition, dive 2605. Exterior and interior of left valve. Figure 7. Specimen from Lucky Strike hydrothermal field, 44.8 mm. Same locality. Exterior and interior of left valve. Figure 8. Specimen from Lucky Strike hydrothermal field, site Pagoda (PP7), 37°17.63'N, 32°16.96'W, 1629 m, DIVA 2, dive 07, 91.1 mm. Exterior, interior, and ventral inner view of left valve. Figure 9. Specimen from Lucky Strike hydrothermal field. Same locality, 59.3 mm. Exterior of right valve. Figure 10. Specimen from Lucky Strike hydrothermal field, site Eiffel Tower, 37°17.32'N, 32°16.52'W, 1685 m, DIVA 2, dive 08, 84.7 mm. Exterior, interior, and ventral inner view of left valve. Figure 11. Specimen from Lucky Strike hydrothermal field, site Elisabeth, 37°17.63’N, 32°16.87'W, 1640 m, DIVA 2, dive 24, 114.1 mm. Exterior and interior of left valve. All specimens MNHN. R. von Cosel et al., 1999 Page 221 Page 222 Menez Gwen hydrothermal field, Azores Triple Junction, Mid-Atlantic Ridge. Description: Shell up to 119 mm long, more or less elon- gate-modioliform, from thin but solid to rather thick, ex- tremely variable in outline and length/height ratio, but also in tumidity and shell thickness, equivalve, smaller specimens often shorter. Beaks subterminal to almost ter- minal. Anterior margin more or less broadly rounded; ventral margin in juvenile, half-grown, and subadult spec- imens mostly more or less convex or straight, in fully grown specimens straight or more or less concave. Pos- tero-ventral margin evenly rounded, postero-dorsal mar- gin slightly to markedly convex, occasionally straight; postero-dorsal corner rounded; ligament plate usually slightly arched but occasionally almost straight. Exterior with more or less dense, irregular growth lines and growth waves, more or less reflected on interior (see Fig- ures | and 10). Some specimens have about five to six broad and very obscure transverse waves in middle of shell which cause occasionally undulation of concentric striae and may be marked by darker color of periostra- cum; they are reflected as very flat and indistinct waves on interior (see Figure 4). In very few specimens, sculp- ture of faint, broad, radial and sometimes bifurcating un- dulations visible on postero-dorsal slope slightly visible on Figure 6; it may be very slightly reflected on inside. Umbo broad, somewhat flattened. Shell without periostracum dull whitish; interior nacre- ous white. Periostracum strong, warm chestnut brown, in umbonal region and often also lighter brown postero-dorsally; smaller specimens especially often appear more or less two-colored with antero-ventral part dark chestnut brown and postero-dorsal part lighter olive brown with relative sharp limit reaching from umbonal region to postero-ven- tral corner or just in front of it. Surface somewhat dull, with no periostracal hairs; however, byssal endplates of other specimens scattered over whole valve. Hinge edentulous, anterior hinge margin, however, slightly protruding toward ventral. Ligament opisthodetic, strong, extending over whole postero-dorsal margin near- ly to postero-dorsal corner and ending abruptly or in a taper. Subligamental shell ridge hardly marked to obsolete from under umbos to middle of ligament, then missing. The Veliger, Vol. 42, No. 3 Anterior adductor scar long-oval, arched, situated just in front of umbo. Posterior adductor scar united with pos- terior scar of posterior pedal and byssus retractor muscle. Anterior scar of same muscle separated and situated un- der ligament’s end or slighty more forward. Anterior bys- sus retractor muscle scar situated in umbonal cavity, reaching from umbo toward posterior, visible only in pos- terior and ventral view but not in lateral view of interior. Pallial line ventrally slightly concave or straight. Examined larval shells (Figures 45-52) measured be- tween 522 and 527 wm long and were nearly 500 wm high. There is a separation between a very small proto- conch [| (about 110 wm long) and the large protoconch II, which indicates a long planktonic larval phase. Proto- conch II pale beige-salmon and well separated from te- leoconch which in ultra-juvenile specimens is nearly transparent. Surface of protoconch II with fine regular concentric grooves which are more or less densely spaced, protoconch I with irregular sculpture. Animal with very large ctenidia which are nearly four- fifths of shell length; outer and inner demibranch almost of equal size, outer demibranch only slightly shorter an- teriorly. Ascending lamellae of both demibranchs anteri- orly fused to the mantle visceral mass for a very short distance, then becoming free toward posterior. Narrow and well-visible food groove on ventral edge of each de- mibranch; outer surface of ascending lamellae of inner and outer demibranch with grooves just below free edges and parallel to them. No muscular longitudinal ridge on mantle and visceral mass where dorsal edges of ascending lamellae touch mantle lobes. Connection bars between free edges and gill axes absent. Inner mantle folds sepa- rate along whole ventral margin length from anterior ad- ductor to posterior margin. Filaments moderately broad; each fifth to seventh filament with a connecting septum of about half the height of demibranchs. Mantle lobes thin but with strongly muscular mantle margins. Mantle edges with three folds, inner mantle fold frilled but degree of frilling variable. On anterior end, inner mantle folds pass from ventrally over anterior ad- ductor muscle up- and foreward along anterior margin, then fold down- and backward to pass again lower end of anterior adductor muscle or slightly posterior to it to- ward ventral margin. On this ‘‘folding part’’ mantle edge Explanation of Figures 12—15 Figures 12-15 Bathymodiolus azoricus Cosel & Comtet, sp. nov. Figure 12. Specimen from Lucky Strike hydrothermal field, site PPS, 37°17.49’N, 32°16.88'W, 1725 m, DIVA 2, dive 05, 83.0 mm. Exterior, interior, and ventral inner view of right valve and dorsal view of specimen. Figure 13. Specimen from Lucky Strike hydrothermal field. Same locality, 63.1 mm. Exterior, interior, and inner ventral view of left valve. Figure 14. Specimen from Lucky Strike hydrothermal field. Same locality, 61.6 mm. Exterior and interior of left valve, dorsal view of specimen. Note the different tumidity of the specimens on Figures 13 and 14 from the same locality. Figure 15. Specimen from Lucky Strike hydrothermal field. Same locality, 90.3 mm. Exterior and interior of left valve. All specimens MNHN. Page 223 R. von Cosel et al., 1999 Page 224 The Veliger, Vol. 42, No. 3 remains frilled (Figure 33), but occasionally is less or not frilled on passage over anterior adductor from one valve to the other (Figures 31, 32). Valvular siphonal membrane short and rather strong, reaching from postero-ventral corner to exhalent siphonal opening, with more or less developed papilla in middle on anterior edge (see Figures 31-33). Inner siphonal aperture with internal diaphragm with horizontal slit and muscular fold around it. Two very broad and short flattened tentacles ventrally under slit, directed toward anus (see Figures 43, 44). Foot somewhat variable but generally rather small and quite slender, with ventral byssal groove two-thirds to three-fourths length of foot. Foot-byssus retractor muscle complex with rather long anterior retractor; posterior bys- sus retractors consisting of two quite strong, diverging muscle bundles with common base at base of byssus. An- terior bundle very short and broad and arising rather steep- ly toward attachment point on shell inside, posterior bundle very long and thin, passing almost parallel to longitudinal shell axis toward attachment point directly in front of pos- terior adductor. Posterior foot retractor rather thin, arising from base of foot, well in front of base of byssus retractor muscles, passing outer side of anterior retractor toward an- terior bundle of posterior byssus retractor; it reaches inner shell surface closely appressed to anterior bundle over half to two-thirds its length. Labial palps variable in size, gen- erally rather large (Figure 42) but in juveniles more or less small, occasionally also in larger specimens (Figure 41). Posterior labial palps narrow-triangular, anterior two slight- ly smaller than posterior pair and still narrower. Labial palp suspensor muscles present. Mouth transverse, slit-shaped; esophagus a narrow tube with irregular, close-set longitudinal ridges on its inner surface just in front of entrance to stomach. Stomach (Figure 62) small and very elongate for a mytilid, with thin walls, anterior chamber slightly shorter than posterior chamber but both with about equal width. Digestive di- verticula around whole stomach. Style sac and midgut conjoined. Major typhlosole passing from there toward anterior along floor of posterior stomach chamber, and ending in anterior chamber. On left side of posterior chamber, three diverticle ducts open into shallow depres- sion corresponding to left pouch and situated below gas- tric shield. Small grooves leading from every digestive duct opening and joining to form beginning of intestinal groove. This latter following major typhlosole, turning right and running along right side of it toward and into midgut. Minor typhlosole on right side of midgut and ending just after entering posterior chamber. In this cham- ber six openings of digestive diverticula ducts. Stomach of examined specimen contained only some mucus. No crystal style found. Midgut running straight backward to under ventricle, there making very small to moderately large counter- clockwise recurrent loop before entering it just in front of auricular ostiae. Heart with muscular ventricle and very large auricles which are fused posteriorly under in- testine. Selected measurements (length, height, tumidity) in mm with length-height ratios: a) Menez Gwen 111.9 x 47.4 x 36.0 Menez Gwen PI 13 109.7 X 45.8 X 34.7 Menez Gwen PI 13 109.0 x 49.4 x 41.3 Menez Gwen PI 13 108.7 X 49.6 X 38.2 Menez Gwen PI 13 108.0 x 45.8 x 39.1 Menez Gwen PI 13 107.5 X 49.5 X 39.6 Menez Gwen PI 13 103.0 X 44.1 X 35.7 Menez Gwen PI 13 100.3 X 42.3 37.8 Menez Gwen PI 13 98.5 x 44.2 x 33.2 Menez Gwen PI 13 95.8 X 38.0 x 29.4 Menez Gwen PI 13 95.2 * 44.7 X 35.1 Menez Gwen PI 13 95.0 X 40.0 X 34.9 Menez Gwen PI 13 94.3 x 44.1 K 35.5 Menez Gwen PI 13 93.4 X 42.6 X 31.7 Menez Gwen PI 13 93.0 X 44.1 X 36.9 Menez Gwen PI 13 2 holotype MNHN paratype MNHN paratype USNM paratype MNHN paratype MNHN paratype MNHN paratype MNHN paratype MNHN paratype MNHN paratype SMF paratype MCZ paratype MNHN paratype ZMM paratype Funchal Mus. paratype MNHN 92.7 x 40.3 X 33.2 Menez Gwen PI 13 2.3 paratype LACM 90.9 * 38.7 X 34.5 Menez Gwen PI 13 2.3 paratype MNHN 87.6 X 40.4 x 30.0 Menez Gwen Pl 13 2.2 paratype MNHN 83.8 X 36.0 X 28.0 Menez Gwen Pl 13 2.3 paratype MNHN 79.6 X 35.0 X 27.8 Menez Gwen PI 13 2.3 paratype NSMT 77.7 X 32.1 X 27.6 Menez Gwen Pl 13 2.4 paratype MNHN 76.0 X 32.7 X 25.4 Menez Gwen PI 13 2.3 paratype NUNZ 62.9 x 30.3 * 25.7 Menez Gwen PI 13 2.1 paratype MNHN b) Lucky Strike 119.3 X 52.7 X 46.4 mm_ Elisabeth Pl. 24 113.8 X 56.2 * 45.1 mm 101.4 * 45.7 X 38.7 mm 97.0 X 43.5 x 33.4 mm 96.0 x 44.0 x 42.0 mm 94.6 x 45.1 x 40.0 mm pp/7 Pagoda Elisabeth Pl. 24 Eiffel Tower P] 08 2.2 Eiffel Tower Pl] 08 2.2 Eiffel Tower P1 08 2.1 94.5 x 42.6 x 37.4 mm_ pp7 Pagoda wD, 91.1 < 47.4 x 37.0 mm_ pp7 Pagoda 1.9 90.7 X 40.0 X 34.7 mm_ Eiffel Tower P1 08 2.3 89.4 * 43.3 x 32.0 mm 88.4 * 40.4 33.3 mm 85.4 X 37.0 X 37.0 mm 84.1 X 42.7 X 33.6 mm 82.0 38.6 X 32.9 mm pp7 Pagoda pp7 Pagoda pp7 Pagoda Eiffel Tower P1 08 2.1 Eiffel Tower Pl 08 2.0 R. von Cosel et al., 1999 Page 225 81.1 X 37.5 X 31.1 mm_ pp7 Pagoda Dp) 76.0 X 37.2 X 31.3 mm_ pp7 Pagoda 2.0 71.7 X 40.6 X 31.0 mm_ pp7 Pagoda 1.8 67.3 X 29.8 X 28.6 mm_ pp7 Pagoda pips) 61.1 X 28.4 x 24.7 mm_ pp7 Pagoda ee) 52.4 x 28.4 X 25.2 mm_ pp7 Pagoda 1.8 57.8 X 30.3 X 27.7 mm_ pp7 Pagoda 1.9 47.8 X 25.5 xX 24.5 mm _ pp7 Pagoda LS) 44.9 xX 23.1 X 17.5 mm _ PI! 10 Eiffel Tower 1.9 Material examined: Type material; other material: Mid- Atlantic Ridge, Azores Triple Junction, Menez Gwen hy- drothermal field, site PP11, 37°50.5’N, 31°31.3’W, 866 m, DIVA 2, dive 11, M. Biscoito, observer, 14 June 1995, 46 spm.; site Mogued-Gwen (PP10) 37°50.56’N, 31°31.27'W, 877 m, DIVA 1, dive 13, sample 13.6, Y. Fouquet, observer, 21 May 1994, 3 spm.; same locality, DIVA 2 dive 12, D. Desbruyéres, observer, 14 June 1995, 15 spm.; Lucky Strike hydrothermal field, site Statue of Liberty, 37°17.59'N, 32°16.50’W, 1635 m, expedition “LUCKY STRIKE 1993,” dive 2605, D. Desbruyeéres and D. Colodner, observers, 31 May 1993, 29 spm.; site Sintra, 37°17.57'N, 32°16.57'W, 1622 m, DIVA 2, dive 02, Ph. Crassous, observer, 4 June 1995, 26 spm., 10 juv. spm.; site Eiffel Tower, 37°17.32'N, 32°16.52'’W, 1685 m, DIVA 2, dive 08, Th. Comtet, observer, 10 June 1995, 19 spm.; same locality, dive 10, M.-C. Fabri, observer, 12 June 1995, 25 spm, 10 juv. spm.; site Isabel, 37°17.37'N, 32°16.64'W, 1685 m, DIVA 2, dive 01, A.- M. Alayse, observer, 3 June 1995, 15 spm., 3 juv. spm.; same locality, dive 03, 8 juv. spm.; site Pagoda (PP 7), 37°17.63'N, 32°16.96'W, 1629 m, DIVA 2, dive 06, P. Briand, observer 8 June 1995, 11 spm, 8 juv. spm.; same locality, dive 07, P-M. Sarradin, observer; 9 June 1995, 37 spm., 2 v. and numerous juveniles; site PP 5, 37°17.49'N, 32°16.88'W, 1725 m, DIVA 2, dive 05, EF Barriga, observer, 7 June 1995, 15 spm, 13 juv. spm.; site Elisabeth 37°17.63'N, 32°16.87'W, 1640 m DIVA 2 dive 24, A.M. Alayse, observer, 30 June 1995, 15 spm., all MNHN. Biotope: Bathymodiolus azoricus dominates the fauna of both the Menez Gwen and Lucky Strike hydrothermal fields. At Lucky Strike, the mussels live byssally attached to hard substrate and cover the walls of active edifices and flanges (on the Pagoda site), where small specimens reach densities up to 10,000 ind/m? (A. Colaco, personal communications). They also colonize cracks in the sea floor. The species lives at temperatures ranging from about 6°C (i.e., the ambient seawater temperature) to about 30°C. The distribution of the mussels along the thermal and chemical gradient seems to be related to their size, the largest individuals living in the warmest areas (Comtet, unpublished data). At Menez Gwen, the colo- Explanation of Figures 16-21 Figures 16-21 Bathymodiolus sp. Figure 16. Specimen from Logachev hydrothermal field, 14°45'N, 44°58’ W, 2930— 3010 m, cruise 35 R/V Akademik Mstislav Keldysch, sta. 3452, ZMM, 61.5 mm, exterior and interior of left valve, exterior of right valve. Figure 17. Specimen from Logatchev hydrothermal field. Same locality, 83.3 mm. Interior and exterior of right valve. MNHN. Figure 18. Specimen from Logatchev hydrothermal field, Irina site 14°45.10'N, 44°48.60'W, 3040 m, MIKROSMOKE, dive 21, 69.1 mm. Exterior and interior of left valve. Figure 19. Specimen from Logatchev hydrothermal field. Same locality, 122.9 mm. Exterior of both valves, interior of right valve. Figure 20. Specimen from Logatchev hydrothermal field. Same locality, 47.5 mm. Exterior of left valve. Figure 21. Spec- imen from Logatchev hydrothermal field. Same locality, 41.1 mm. Exterior of left valve. All MNHN. Explanation of Figures 22—30 Figures 22—24 Bathymodiolus sp. Figure 22. Specimen from Logatchev hydrothermal field, Irina site, 14°45.10'N, 44°48.60'W, 3063 m, MIKROSMOKE, dive 20, 95.0 mm. Exterior, interior, and inner ventral view of right valve, dorsal view of specimen. Figure 23. Specimen from Logatchev hydrothermal field, Irina site 14°45.10’'N, 44°48.60'W, 3040 m, MICROSMOKE, dive 21, 120.3 mm. Exterior and interior of left valve. Figure 24. Specimen from Logatchev hydrothermal field. Irina site, 14°45.10’N, 44°48.60’W, 3063 m, MICROSMOKE, dive 20, 81.7 mm. Exterior of left valve. All MNHN. Figures 25-30. Bathymodiolus azoricus Cosel & Comtet, sp. nov. Figure 25. Specimen from Lucky Strike hydrothermal field, site Pagoda (PP7), 37°17.63'N, 32°16.96'W, 1629 m, DIVA 2, dive 07, 44.4 mm. Exterior and interior of left valve. Figure 26. Specimen from Lucky Strike hydrothermal field, same locality, 34.6 mm. Exterior and interior of left valve. Figure 27. Specimen from Lucky Strike hydrothermal field, site PP 5, 37°17.49'N, 32°16.88’W, 1725 m, DIVA 2, dive 05, 47.3 mm. Exterior and interior of left valve. Figure 28. Specimen from Lucky Strike hydrothermal field, site Eiffel Tower, 37°17.32'N, 32°16.52'W, 1685 m, DIVA 2, dive 10, 48.3 mm. Exterior and interior of left valve. Note the highly different height of specimens of the same size. Figure 29. Specimen from Lucky Strike hydrothermal field, site Pagoda (PP7), 37°17.63'N, 32°16.96'W, 1629 m, DIVA 2, dive 07, 16.2 mm. Exterior of left valve. Figure 30. Specimen from Menez Gwen hydrothermal field, site Mogued Gwen (PP10) 37°50.56’N, 31°31.27'W, 877 m, DIVA 2, dive 12, 85.0 mm (details of this specimen shown on Figures 31 and 36). All specimens MNHN. The Veliger, Vol. 42, No. 3 Page 227 R. von Cosel et al., 1999 Page 228 The Veliger, Vol. 42, No. 3 nies of Bathymodiolus azoricus are more scattered, which could be due to the presence of soft substrate. Their tem- perature preferences are similar to those of the mussels at Lucky Strike, ranging from about 8°C (i.e., ambient seawater temperature) to about 30°C. The mussels derive their food from intracellular symbiotic chemoautotrophic bacteria, of both sulfide-oxydizing and methanotrophic types (Fiala-Médioni et al., 1996). On Lucky Strike hydrothermal field, many specimens of Bathymodiolus azoricus harbor in their pallial cavity the commensal polynoid polychaete Branchypolynoe see- pensis Pettibone, 1986 This worm was found already in small individuals, from 33 mm length onward. In these small mussels, the polychaete is of course smaller but it can reach up to half the shell length. In the largest studied specimen of 119.3 mm, the worm was 45 mm long. In the Menez Gwen mytilids, Branchypolynoe were never found. A detailed ecological description of the Lucky Strike vent field is given by Van Dover et al. (1996). Distribution: Bathymodiolus azoricus is only known from the Menez Gwen and Lucky Strike hydrothermal fields on the Azores Triple Junction, Mid-Atlantic Ridge. Etymology: The name expresses the proximity of the lo- calities of this species to the Azores archipelago. Remarks: Bathymodiolus azoricus shows an extreme variability, especially in shell shape, tumidity, and length/ height ratio, but also in the position of the anterior scar of the posterior byssus retractor muscle, the form of pos- terior end of the ligament (ending abruptly or tapering), the thickness of the shell, the anterior and posterior man- tle fusion, and the size of the labial palps. The variability of the shell is so that two ‘“‘extreme”’ specimens of B. azoricus suggest two totally different species. The general shell outline can be elongate-trian- gular to oval-oblong or even oval-elongate, almost date- shaped. The anterior margin may be broadly or rather narrowly rounded; the postero-dorsal corner is narrowly rounded to nearly indistinct and more or less integrated into the rounded posterior margin; the postero-dorsal mar- gin is straight to convex. Growth allometry is also vari- able. The surface of the shells varies from smooth and quite glossy with relatively few growth lines to more or less covered with an oxide layer and numerous byssal endplates of other mussels, with strong and dense growth lines and somewhat coarser growth stages. The ventral pallial line is straight to rather markedly curved, this mostly (not always) when the ventral margin is also con- cave. The shell tumidity is also considerably variable (see Figures 12 and 14); extreme forms may be even more tumid than the maximum height of the shell. This variability in shell shape and outline is present already in medium-sized and small specimens (see Fig- ures 25—28); the latter often have a markedly convex ven- tral margin or in some specimens, the margin may be already somewhat concave. Specimens from some local- ities are small but have a more or less adult appearance with broad (high) shell, numerous well-marked growth stages, and oxide layer. Gonads were already present in several specimens of 30—40 mm size; the smallest specimen with clearly vis- ible gonads measures 31.8 X 16.5 mm and is from Menez Gwen; very few gonads were evident in a specimen of 25.4 X 16.5 mm from Isabel site (Lucky Strike) and one of 28.5 X 14.6 from Menez Gwen. Many of the small specimens with gonads are more or less thick-shelled and broad with many growth marks, but not in all cases; the above-mentioned 31.8 mm specimen is smooth, rather narrow, and quite thin-shelled. A few of the larger spec- imens (between 60 and 76 mm) were found with very few gonads or none at all; they all have a rather thin shell. One might expect gonads in the small forms with dense growth rings and thicker shell, which can be viewed as dwarf adults that grew more slowly under less favorable ecological conditions, but the presence of gonads in small, smooth, and thin-shelled juvenile-looking speci- mens from Menez Gwen still needs an explanation. The soft parts also are variable: the papilla on the valv- ular siphonal membrane varies from being hardly visible as a slight curve only, to being large and strongly pro- Explanation of Figures 31—35 Figures 31-33. Bathymodiolus azoricus Cosel & Comtet sp. nov. Figure 31. Specimen from Menez Gwen hydro- thermal field, site Mogued Gwen (same specimen as on Figure 30). Close-up view of anterior and posterior mantle fusion and posterior valvular siphonal membrane. Figure 32. Specimen from Lucky Strike hydrothermal field, site Pagoda (PP7), 37°17.63'N, 32°16.96'W, 1629 m, DIVA 2, dive 07, shell length 56.7 mm. Close-up view of anterior and posterior mantle fusion and posterior valvular siphonal membrane. Figure 33. Specimen from Lucky Strike hydrothermal field, same locality, shell length 59.3 mm. Close-up view of anterior and posterior mantle fusion and posterior valvular siphonal membrane. Note the different width of the “turning back” part above the anterior adductor in the three specimens. Figures 34, 35. Bathymodiolus sp. Figure 34. Specimen from Logatchev hydro- thermal field, Irina site 14°45.10'N, 44°48.60'W, 3040 m, MIKROSMOKE, dive 21, shell length 69.1 mm. Close- up view of anterior and posterior mantle fusion and posterior valvular siphonal membrane. Figure 35. Specimen from Logatchey hydrothermal field, Irina site 14°45.10’N, 44°48.60’W, 3040 m, MIKROSMOKE, dive 21, shell length 71.6 mm. Close-up view of anterior and posterior mantle fusion and posterior valvular siphonal membrane. R. von Cosel et al., 1999 Page 229 Page 230 The Veliger, Vol. 42, No. 3 Explanation of Figures 36—42 Figures 36, 37. Bathymodiolus azoricus Cosel & Comtet, sp. nov. Figure 36. Specimen from Menez Gwen hydro- thermal field, site Mogued Gwen (PP10) 37°50.56'N, 31°31.27'W, 877 m, DIVA 2, dive 12, MNHN, 85.0 mm. Ventral view showing ventral opening. One valve removed. Figure 37. Specimen from Lucky Strike hydrothermal field, site Pagoda (PP7), 37°17.63'N, 32°16.96'W, 1629 m, DIVA 2, dive 07, MNHN, 59.3 mm. Ventral view showing ventral opening. Figure 38. Bathymodiolus sp,. Specimen from Logatchev hydrothermal field, Irina site 14°45.10'N, 44°48.60'W, 3040 m, MIKROSMOKE, dive 20, MNHN, 81.7 mm. Ventral view showing ventral opening. One valve removed. Figures 39—42. Bathymodiolus azoricus Cosel & Comtet, sp. nov. Figure 39. Specimen from Lucky Strike hydrothermal field, same locality, shell length 57.5 mm. Close-up view of exhalent siphon with clearly visible anal papilla. Figure 40. Specimen from Lucky Strike hydrothermal field, site Pagoda (PP7), 37°17.63'N, 32°16.96'W, 1629 m, DIVA 2, dive 07, shell length 56.7 mm. Close-up view of exhalent siphon. Figure 41. Specimen from Menez Gwen hydrothermal field, paratype, MNHN, shell length 107.8 mm. Detail of labial palps. Figure 42, Specimen from Menez Gwen hydrothermal field, paratype, MNHN, shell length 103.0 mm. Detail of labial palps. R. von Cosel et al., 1999 Page 231 Explanation of Figures 43 and 44 Figures 43, 44. Bathymodiolus azoricus Cosel & Comtet, sp. nov., Specimen from Menez Gwen hydrothermal field, site Mogued Gwen (PP10) 37°50.56’N, 31°31.27'W, 877 m, DIVA 1, dive 13, sample 13.6. Shell length 81.1 mm. View into the siphonal cavity under two different angles. Note the anal papilla and the two tentacles on the lower transversal membrane directly under the siphonal opening (excurrent chamber) and on Figure 44 the posterior part of the incurrent chamber. tuberant. Also the passage of the mantle edge over the anterior adductor from one valve to the other is rather variable (Figures 31—33). All these highly variable characters are combined in nearly every sense, even within lots from the same site or dive, so that a clear delimitation of certain “‘morphs”’ is often not possible. However, in some sites, a tendency toward a certain form can be observed, especially in the site Statue of Liberty, where the specimens are excep- tionally slender for a Lucky Strike population (Figures 6, 7). Differences are more clear and seem to be more stable between the specimens of the two hydrothermal fields, Lucky Strike and Menez Gwen, which have a horizontal distance of about 60 km and a depth difference of about 800 m. Menez Gwen mussels have the umbos still some- what more forward (see Figures 1—5) than those from Lucky Strike (for more details, see Biometry). This variability is probably mainly due to abiotic eco- logical factors such as degree of nutrition (availability of nutrients), degree of venting activity (which can change rapidly), physico-chemical conditions, composition and temperature of the water, etc., but also to biotic factors, e.g., population density and competition. All these factors determine growth speed and growth allometry. The more tumid specimens with dense growth lines certainly did not have as favorable conditions as those with a smooth surface, sharp margins, and no incrustations. These spec- imens obviously grew faster and were less disturbed. Some characters might perhaps also be related to genetic factors. In some sites (e.g., Menez Gwen) there is a ten- dency toward a certain homogeneity, but also there a few specimens of other ‘“‘morphs’’ were present in a sample. In other sites, however, all morphs occurred together. Two rather stable characters, however, which distin- guish the shells of B. azoricus from B. puteoserpentis, are the position of the anterior byssus retractor scar and the position of the umbos relative to shell length. In B. puteoserpentis, the anterior byssus retractor scar is situ- ated on the anterior part of the umbonal cavity in front of the umbos, whereas in the new species, the byssus retractor inserts more posterior within the umbonal cavity, normally directly under the umbos. B. azoricus is gen- erally somewhat more elongate, and the umbos are al- ways placed more forward than in B. puteoserpentis, at one-tenth to one-twelfth of shell length or more as op- posed to one-seventh in B. puteoserpentis. Among the known vent mussels, only B. platifrons Hashimoto & Okutani, 1994, has similar almost terminal umbos. We conclude that all ‘“‘morphs”’ belong to a single very variable species and that even genetic differences be- tween the sites (if present) are too small to warrant sep- aration. The Veliger, Vol. 42, No. 3 Bathymodiolus sp. aff. B. puteoserpentis (Figures 16—24, 34, 35, 38, 53-56, 61, 63-67) Description: Shell large, up to 123 mm long, thin but rather solid, modioliform-oval, considerably variable in outline, inflated, equivalve. Juvenile specimens in general somewhat shorter and more oval than adults but also al- ready quite variable. Beaks subterminal. Anterior margin rather broadly rounded; ventral margin straight to slightly convex, in large and fully grown specimens often slightly concave in middle. Postero-ventral margin broadly round- ed, postero-dorsal margin slightly convex to almost straight; postero-dorsal corner rather broadly to very broadly rounded; ligament plate arched, often more in its anterior half. Exterior with well-developed and strong, irregular growth lines and growth waves, which are well reflected on inside. In juveniles and half-grown speci- mens, weak and fine radial sculpture mostly visible in middle part of ventral half of valve as rather dense, ir- regular radial wrinkles (see Figure 19); occasionally also faint, narrow radial lines or waves on posterior slope (see Figure 17). Some faint radial structure also visible on inside of valves within shell material but not sculpturally reflected on the internal surface like growth lines. Umbos broad, very flattened. Shell without periostracum dull whitish; interior nacre- ous white. Periostracum strong, brown with a slight tendency to- ward olive, in umbonal and postero-dorsal region lighter brown, somewhat glossy, with no periostracal hairs; how- ever, byssal endplates of other specimens always scattered over whole valve. Hinge edentulous, anterior hinge margin, however, slightly protruding toward ventral. Ligament opisthodetic, strong, extending over almost whole postero-dorsal mar- gin to postero-dorsal corner. Subligamental shell ridge very faint from under umbos to middle of ligament, then becoming obsolete; under beaks visible only in ventral view and not in lateral view. Anterior adductor scar long- oval, arched, situated in front of umbo. Posterior adductor scar rounded-trapezoid, united with posterior scar of pos- terior foot and byssus retractor muscle. Anterior scar of same muscle separated and situated under ligament, at slightly behind two-thirds of its length (Figure 19). In smaller specimens, this scar is more backward (Figure 22) and in very small (35 mm and smaller) mussels, it is situated just behind ligament’s end. Anterior byssus re- tractor muscle scar on anterior part of umbonal cavity just in front of beaks, visible only in posterior and ventral view but not in lateral view of interior. Pallial line ven- trally straight, in large specimens often slightly concave when ventral margin is also concave. Larval shell 390—400 pm long and 380 pm high (Fig- ures 53-56). Protoconch I 120 wm long, with irregular surface and well separated from Protoconch II, which in- dicates a long planktonic larval phase. Surface of Proto- conch II with very fine, densely spaced and regular con- centric grooves which may not always be visible. Animal with large ctenidia which are about three- fourths of shell length and cover entire visceral mass, each demibranch with descending and shorter ascending lamellae. Outer demibranch anteriorly slightly shorter than inner demibranch. Ascending lamellae of outer and inner demibranchs anteriorly fused to mantle and to vis- ceral mass, respectively, for very short distance; more posteriorly gills entirely free from fusion. No muscular longitudinal ridges on mantle and visceral mass where dorsal edges of ascending lamellae attach. Connection bars between free edges and gill axes absent. Ventral edge with shallow food groove. Outer surface of ascending la- mellae of both demibranchs with folds just below free edges and parallel to them. Filaments wide, fleshy, con- nected with each other by “‘plaquettes” and “‘racquets”’ (Le Pennec & Hily, 1984) on ventral and lateral sides, respectively. Approximately each fourth to seventh fila- ment with septum reaching to half of gill height and con- necting lamellae. In juvenile specimens, ascending la- mellae of demibranchs much shorter than descending la- mellae; in adult specimens, lamellae almost equal-sized. Mantle thin, except for heavily thickened muscular margin and vascularized anterior region. Mantle edges with three folds; posteriorly, inner mantle folds fused dor- sally above exhalent siphon and between exhalent siphon and combined inhalent aperture and pedal gape, forming short, narrow, and rather strong valvular siphonal mem- Explanation of Figures 45-52 Figures 45-52. Bathymodiolus azoricus Cosel & Comtet, sp. nov. Larval shells. Figures 45—48. Specimens from Menez Gwen hydrothermal field, 37°50.54'N, 31°31.30’W, DIVA 2, dive 11. Figure 45. Juvenile specimen. Scale bar: 200 wm. Figure 46. Close-up view of Protoconch II of the same specimen. Scale bar: 100 wm. Figure 47. Protoconch II of another specimen. Scale bar: 100 ym. Figure 48. Protoconch I of another specimen. Scale bar: 20 wm. Figures 49-52. Specimens from Lucky Strike hydrothermal field, site Eiffel Tower, 37°17.32'N, 32°16.52'W, 1685 m, DIVA 2, dive 10. Figure 49. Juvenile specimen. Scale bar: 200 pm. Figure 50. Another specimen showing Protoconchs I and IH. Scale bar: 100 pm. Figure 51. Close-up view of Protoconch I of the same specimen. Scale bar: 10 xm. Figure 52. Protoconch II of another specimen, showing the more or less widely spaced concentric striae. Scale bar: 100 pm. R. von Cosel et al. Page 234 The Veliger, Vol. 42, No. 3 Explanation of Figures 53-58 Figures 53-56. Bathymodiolus sp.. Specimens from Logatchev hydrothermal field, Irina site, 14°45.10'N, 44°48.60'W, 3063 m, MIKROSMOKE dive 20. Figure 53. Juvenile specimen. Scale bar: 1 mm. Figure 54. Close- up view of Protoconch II of the same specimen. Scale bar: 100 pm. Figure 55. Ultra-juvenile specimen with well- distinguished Protoconch I and Protoconch II. Scale bar: 100 im. Figure 56. Close-up view of Protoconch I of the R. von Cosel et al., 1999 brane which reaches from siphonal opening to postero- ventral corner and bears a small papilla. Anteriorly, inner mantle folds fused for very short distance underneath an- terior adductor muscle. Mantle folds passing from ven- trally over adductor muscle up- and foreward along an- terior margin, then folding down- and backward to pass again lower end of anterior adductor muscle toward ven- tral margin. Inner mantle folds frilled over all their length (Figure 38). Pallial muscles and siphonal retractors strong. Exhalent siphon short, inner aperture of it occlud- ed by thin internal diaphragm with narrow horizontal slit. Muscular fold around slit which regulates aperture size. Two small, flattened tentacles directed toward anus situ- ated under slit. As a branchial septum and a fusion of the gills with each other are absent, division of mantle cavity into a ventral incurrent and a dorsal excurrent chamber is not complete. Foot thick, broad, flattened, tapering toward end, with ventral byssal groove three-fourths the length of foot. An- terior byssus retractor moderately long, strong, divided into three small blocks at about half its length. Posterior byssus retractor consisting of two strong, diverging mus- cle bundles of approximately equal width as anterior bys- sus retractor and having common base at base of byssus. Anterior bundle about two times shorter than posterior bundle and arising steeply toward attachment point on shell. Posterior bundle divided into two parallel bundles over all its length and passing at low angle to longitudinal shell axis. Posterior pedal retractor slightly more slender than other muscle bundles, arising from base of foot an- terior to origin of posterior byssus retractors, passing out- er side of anterior retractor toward anterior bundle of pos- terior byssus retractor. Posterior pedal retractors slightly asymmetrical, right one divided into two bundles at about half its length, both bundles inserting on inner and ante- rior sides of anterior bundles of posterior retractors. Left pedal retractor divided into two bundles over all its length, one bundle inserting on outer side of anterior bun- dle of posterior retractor at half its length, and other bun- dle inserting on inner side of anterior bundle. Labial palps narrow, triangular, short but stout, strongly ridged on inner surfaces, anterior palps slightly smaller than posterior ones. Labial palp suspensor muscles pres- ent. From lateral sides of mouth between palps narrow fold running to base of gills. Labial palps of juveniles much shorter and sometimes lacking ridges. Mouth transverse, slitlike, opening into short, thin- walled esophagus. Esophagus entrance at anterior end of stomach surrounded by dark digestive diverticula. Inner Page 235 aa ar ppr pbr (a) pbr (p) f b pa Figure 59 Sketch of foot-byssus retractor muscle complex of Bathymodi- olus azoricus Cosel & Comtet, sp. nov., and its position in the shell (separate slender strand of anterior retractor serving as sup- port for labial palps not drawn); specimen from Menez Gwen DIVA 2, Pl. 13; shell size 98.5 mm; aa, anterior adductor; ar, anterior retractor; ppr, posterior pedal retractor; pbr (p), posterior byssus retractor, posterior bundle; pbr (a), posterior byssus re- tractor, anterior bundle; pa, posterior adductor; f, foot; b, byssus. surface of esophagus near its entrance into stomach bear- ing longitudinal ridges. Stomach (Figure 66) thin-walled, small, elongate, divided into a round anterior, and a more conspicuous posterior chamber. Major typhlosole arising from conjoined style sac and midgut, passing forward along floor of posterior chamber of stomach and termi- nating in shallow depression of floor of anterior chamber, possibly corresponding to food-sorting caecum. Gastric shield on antero-dorsal wall on left side of posterior chamber. Altogether, 11 ducts of digestive diverticula open into stomach. Posterior chamber with six openings. Three ducts of left side open into shallow depression be- low gastric shield, corresponding to left pouch. Small grooves leading from every opening of ducts and joining to form beginning of intestinal groove. Intestinal groove following major typhlosole, turning right and running along right side of major typhlosole into midgut. To right of intestinal groove and parallel to it, shallow groove run- ning on lateral side of posterior chamber. Three ducts of digestive diverticula of right side opening along side of groove. Five ducts of digestive diverticula opening in an- terior chamber near esophagus entrance—two openings from left side and three on floor and from right side. Crystal style present as amorphic mass. Arrangement of same specimen. Scale bar: 20 pm. Figures 57, 58. Bathymodiolus puteoserpentis Cosel, Métivier & Hashimoto, 1994 (for comparison). Snake Pit hydrothermal field, Elan site, 23°22’N, 47°57'W, 3520 m, MIKROSMOKE dive 14. Figure 57. Juvenile specimen. Scale bar: 1 mm. Figure 58. Close-up view of Protoconch II, firmly covered by oxide layer but limits more or less visible. Scale bar: 100 pm. Figure 60 Sketch of foot-byssus retractor muscle complex of Bathymodi- olus azoricus Cosel & Comtet, sp. nov.; specimen from Lucky Strike, DIVA 2, Pl 07, Pagoda. Shell illustrated on Figure 8, size 91.1 mm. Above, lateral view of the complex and its position in the shell (separate slender strand of anterior retractor serving as support for labial palps not drawn); below, ventral view (slightly enlarged as against the lateral view); gap, pedal ganglion. openings of digestive diverticula to anterior stomach chamber may be variable. In very large specimens, inner surface of stomach more smooth and plain. Midgut leaving the postero-ventral end of stomach and running for short distance posteriorly down mid-line, then entering pericardium and making very short counter- clockwise recurrent loop under ventricle before entering it ventrally and slightly anterior to auricular ostia. Rectum running directly down mid-line toward posterior and end- ing in a papillate anus on posterior surface of posterior adductor muscle. Dorsal wall of anus longer than its ven- tral wall, lateral sides bearing flattened ridges running to posterior point of attachment of gills axis to visceral mass. Shape of outgrowths of dorsal wall of anus vari- able. Examined stomach contained mytilid juveniles, sand, and mucuslike material; the rectum contained sand. Heart three-chambered; ventricle rather large (in some preserved specimens), somewhat triangular, muscular, au- ricles very large, fused together posteriorly and with pro- trusions between bundles of posterior byssus retractors. Kidney situated on each side of body below pericar- dium close to longitudinal vein, consisting of a thin- walled lobulate duct. Renopericardial apertures located in antero-lateral extremities of pericardium. Sexes separate, gonads enclosing digestive diverticula and in large specimens extending into mantle. Genital ap- The Veliger, Vol. 42, No. 3 ertures located at tips of small papillae in excurrent cham- bers near byssus. Selected measurements (length, height, tumidity) with length—height ratios: 123.2 X 57.0 X 46.2 mm_ PI 21 DD 120.7 x 54.4 X 46.6 mm_ PI 20 DD, 120.3 X 56.1 x 45.4mm_ PI 21 Api 94.7 X 49.6 X 39.8 mm_ PI 20 1.9 83.2 X 41.2 X 33.2 mm _ PI 21 2.0 81.9 X 39.8 X 31.1 mm_ PI 20 2.1 89.2 < 43.1 * 38.9 mm _ Akademik Mstislav Keldysch 2.1 74.1 * 37.1 X 33.6 mm_ Prof. Logatchev 2.0 72.7 X 36.7 X 30.7 mm _ PI 21 2.0 71.7 X 36.4 X 28.4mm_ PI 20 2.0 71.4 x 38.4 X 30.6 mm _ PI 20 1.9 71.4 X 35.1 X 29.0 mm _ PI 20 2.0 68.0 X 33.4 * 31.4 mm _ Akademik Mstislav Keldysch 2.0 69.3 X 35.7 X 32.2mm_ PI 21 1.9 65.1 X 34.3 X 26.1mm_ PI 21 1.9 61.5 * 32.5 X 27.3 mm _ Akademik Mstislav Keldysch 1.9 54.5 x 30.6 X 23.7 mm_ PI 21 1.8 54.1 X 29.2 X 20.2 mm_ PI 20 1.9 53.5 X 31.2 X 24.7 mm _ PI 21 lod 47.5 X 27.8 X 18.8 mm_ PI 20 1.7 41.2 X 24.2 X 15.2mm_ PI 20 1.7 37.8 * 22.4 X 16.0mm_ PI 20 1.7 31.7 X 19.4 X 13.2 mm_ Pl 20 1.6 22.5 x 14.8 ¥ 9.1 mm = Akademik Mstislav Keldysch 1.5 19.2 * 12.2 X 8.1mm _ = Akademik Mstislav Keldysch 1.6 19.0 X 13.9 * 6.9mm Akademik Mstislav Keldysch 1.4 14.4 x 8.4 x 5.3 mm Akademik Mstislav Keldysch 1.7 12.2 X 7.9 X 4.8 mm Akademik Mstislav Keldysch 1.5 11.5 x 8.0 * 4.5 mm Akademik Mstislav Keldysch 1.4 7.5 < 4.3 mm Akademik Mstislav Keldysch 1.5 x 11.2 X 7.3 X 4.2 mm Akademik Mstislav Keldysch 1.5 x 5.9 x 3.5 mm Akademik Mstislav Keldysch 1.5 8.9 x 5.9 X 3.6 mm Akademik Mstislav Keldysch 1.5 8.7 X 5.3 X 3.2 mm Akademik Mstislav Keldysch 1.6 8.3 * 5.6 X 3.6 mm Akademik Mstislav Keldysch 1.5 Material examined: Mid-Atlantic Ridge, 14°45’N, 44°58'W, Logachev hydrothermal field, cruise 35, R/V Akademik Mstislav Keldysch, sta. 3452, 16 spm. (among R. von Cosel et al., 1999 Page 237 Figure 61 Sketch of foot-byssus retractor muscle complex of Bathymodi- olus sp. from Logatchev and its position in the shell (separate slender strand of anterior retractor serving as support for labial palps not drawn); specimen from PI. 21, shell length 120.3 mm. For explanation, see Figure 59. them 12 juv.), 1 sh., taken by submersible Mir-2, dive 4/ 171, E.S. Chernjev, observer, 23 February 1995, ZMM Moscow. Irina site, 14°45, 10’N, 44°48.60’W, 3063 m, MICROSMOKE, dive 20, D. Prieur, observer, 5 Decem- ber 1995, 11 spm., 20 juv. spm.; same locality, 3040 m, dive 21, Y. Fouquet, observer, 6 December 1995, 7 spm.; same locality, LOGATCHEV-7 cruise, R/V Professor Lo- gatchev, 1 empty shell taken by TV equipped grab, July 1994, MNHN. Biotope: The animal community of Logatchev hydro- thermal field is dominated, in terms of biomass, by Bath- ymodiolus sp., which forms dense aggregations slightly below the zone of shimmering water (Gebruk et al., 1997). A commensal polynoid polychaete, which remains to be identified, was found in the mantle cavity of Bath- ymodiolus sp, but it occurs with a lower frequency than Branchypolynoe seepensis in Bathymodiolus azoricus. In 23 examined specimens, only four were found to host the polychaete, and the worm was also smaller in relation to shell length than in B. azoricus. An ecological description of the Logatchev field is given in Gebruk et al. (1997). Distribution: Mid-Atlantic Ridge, known only from the Logatchev hydrothermal field, 14°45'N, 44°58’'W, 2930— 3063 m. Remarks: The mussels from the Logatchev field popu- lation are extremely close to B. puteoserpentis, and we do not find any really stable distinguishing character which would allow us to describe this Bathymodiolus as a new species separate from the Snake Pit mussels; how- ever, there are some subtle differences in morphology dd m ty min ty mg Figure 62 Stomach of Bathymodiolus azoricus Cosel & Comtet, sp. nov. from Lucky Strike. Halfschematic drawings from the same specimen as on Figure 60. Left, general view: above, anterior chamber; below, posterior chamber; dd, digestive diverticula ducts. Right, stomach opened dorsally; oe, esophagus; dd, digestive diverticula ducts (en- trances); ig, intestinal groove; gs, gastric shield; big, beginning of intestinal groove; Ip, left pouch; m ty, major typhlosole; min ty, minor typhlosole; mg, midgut. Scale: 1 cm. au Vv pbr (a) go = - Ze il NS N Q\ ies r vy mg st Figure 63 Bathymodiolus sp. Above, specimen of Figure 16, general view of soft parts. pa, posterior adductor; pbr (p), posterior byssus retractor, posterior bundle; pbr (a), posterior byssus retractor, an- terior bundle; v, ventricle; au, auricle; go, gonads; k, kidney; ct, ctenidia; ar, anterior retractor; ap, anterior labial palps; aa, ante- rior adductor; f, foot. Below, another specimen, right mantle lobe and ctenidia removed. p, papilla of siphonal membrane; ies, inner aperture of exhalent siphon; r, rectum; mg, midgut; ppr, posterior pedal retractor; st, stomach; gac, cerebral ganglion. Scale: 1 cm. (this paper) and genetics (Jollivet & Comtet, unpublished results). The mussel populations of both hydrothermal vent fields show a considerable variability in shell outline, and they overlap largely in their degree of variability. Bathymodiolus sp. differs from B. puteoserpentis from Snake Pit in its generally slightly thinner shell, which in very large specimens appears somewhat more elongate; however, in B. puteoserpentis rather elongate specimens were also found. The color of the periostracum tends more toward olive green in Bathymodiolus sp., and some The Veliger, Vol. 42, No. 3 Figure 64 Bathymodiolus sp., foot-byssus retractor muscle complex. Above and middle: a juvenile specimen, lateral and dorsal view; below: holotype, lateral view; gap, pedal ganglion; for other explana- tions, see Figure 59. Scale: A—B: 1 mm; C: 1 cm. specimens in coloration and surface with growth lines and growth waves, as well as in shell thickness, resemble freshwater mussels like Anodonta, whereas B. puteoser- pentis is more chestnut brown. The irregular radial wrin- kles on the middle part of the ventral slope are present in both species, but in B. puteoserpentis they are often less pronounced and also often hidden by the layer of oxide. The protoconch II in both populations is approxi- mately the same size. The most important distinctive fea- ture is the intestine, which in the observed Bathymodiolus sp. has a small counterclockwise recurrent loop under the ventricle, but in B. puteoserpentis the intestine changes its direction twice in an S-like manner in more or less the same plane. However, the shape of the intestine coiling seems to be variable within the same species or popula- tions of both Snake Pit and Logatchev (and also in other species, unpublished observation in B. brevior Cosel, Mé- R. von Cosel et al., 1999 Page 239 Figure 65 Bathymodiolus sp. Left, posterior part of soft parts, lateral view; upper right, exhalent siphon, posterior view; lower right, anus. r, rectum; an, anus; ies, inner aperture of exhalent siphon; es, exhalent siphon; p, papilla of siphonal membrane; t, tentacle of the lower transversal membrane. Scale: left: 2 mm; right: 1 mm. tivier & Hashimoto, 1994), and with the few examined specimens of both populations at hand, we cannot for the moment use this character as the only one to separate the Logatchev population on species level. On the other hand, the slight morphological differences, as well as prelimi- nary results of the current genetic research, do not permit us to identify the Logatchev mussel entirely with B. pu- teoserpentis. Bathymodiolus azoricus is easily distinguished by the different shell outline, the almost terminal umbones with a very short anterior part, the considerably larger proto- conch II, and the relationship of protoconch I/protoconch II. The protoconch I of B. azoricus is, in contrast to pro- toconch II, slightly smaller than in Bathymodiolus sp. A genetic study, which is currently in progress by the sec- ond author, will finally determine the genetic distances between the mussels from Logatchev, Snake Pit, and the Azores Triple Junction; these results and examination of further material will also clarify the status of the Logatch- ev mussel. Both B. azoricus and Bathymodiolus sp. are distin- guished from B. thermophilus by the lack of a ventral mantle fusion, a more complicated stomach with left pouch, a coiled intestine, the absence of a lateral muscular ridge on the mantle lobes and visceral mass, and the ab- sence of connecting bars between the free edges of the demibranchs and the gill axes. In comparison with B. thermophilus, the two species here treated are less spe- cialized. They have a much closer affinity to other Bath- ymodiolus-like species from hydrothermal vents, e.g., B. puteoserpentis and an undescribed Bathymodiolus from the Mariana back-arc basin (Craddock et al., 1995). Apart from the intestine coiling, our two species do not differ from other vent Bathymodiolus-like mytilids in anatomi- cal characters used by Craddock et al., 1995. See Table 1 for a comparison of features. Page 240 Comparison of some features of B. thermophilus and MAR Bathymodiolus. General shell form: Tumidity: Shell: Position of umbos: Position of anterior part of posterior byssus retractor muscle scar: Position of anterior byssus re- tractor scar in the umbonal cavity: Ventral pallial line: Intestine: Mantle lobes on anterior half of ventral side: Valvular siphonal membrane: Papilla in valv.s.memb: Muscular longitudinal ridge on mantle lobes and visceral mass: Posterior end of ligament: Subligamental shell ridge: Table 1 The Veliger, Vol. 42, No. 3 B. thermophilus (from 13°N) moderately elongate more or less com- pressed thin but solid subterminal under the end of the ligament slightly behind the um- bos markedly deflected straight fused long and thin present present tapering strong and angular B. azoricus more or less elongate compressed to tumid thin to rather thick almost terminal under the end of liga- ment or more for- ward under the umbos straight to deflected counterclockwise loop separate short, narrow present but variable absent abrupt to tapering obsolete from umbo to middle then missing Bathymodiolus sp. (Logatchev) moderately elongate to stout moderately tumid thin but solid subterminal at 2/3 of the ligament but variable under and in front of the umbos straight counterclockwise loop separate short present, small absent abrupt faint to obsolete B. puteoserpentis somewhat stout to moderately elon- gate moderately tumid thin but solid subterminal under posterior third of ligament, near the end under and in front of the umbos nearly straight changes direction twice separate short present absent abrupt to slightly ta- pering faint to obsolete BIOMETRY Length—Height Relationships Figures 70 and 71 show the allometric relationships between shell height and length for the three species. Al- lometric curves were fitted following the allometric mod- el of Teissier (1948), of the form H = aL, using Micro- soft Excel 5.0. For the three species treated, these curves indicate that length increases faster than height, traducing an elongation of the shell during growth. These results confirm those given by Comtet (1994) from samples col- lected on the sites Sintra and Eiffel Tower during the LUCKY STRIKE 93 cruise. For Bathymodiolus azoricus (Figure 70), the results indicate a great intersite variability in shell shape in individuals larger than 20 mm. The al- lometric relationship for the Menez Gwen population is in the range of those for Lucky Strike, indicating that mussels of both hydrothermal fields cannot be distin- guished by biometrical characteristics. Bathymodiolus sp. and B. puteoserpentis have a similar shape in the ob- served length range (Figure 71). For the statistical comparisons, length/height ratio was used as an index of shape (Cosel et al., 1994). Compar- isons were made on individuals larger than 20 mm. A one-way analysis of variance (ANOVA) was run to compare length/height ratios in Bathymodiolus azoricus from different sites of the Lucky Strike and Menez Gwen hydrothermal fields, in three different length classes (20— 50 mm; 50—70 mm; larger than 70 mm) (Table 2). Due to the small sample size, the site PP5 was not included in the comparison. For the same reason, only four sites were considered for individuals larger than 70 mm. In each size class, length/height ratios are significantly dif- ferent (P = 0.0001) (Table 2). However, pairwise com- parisons using the Fisher PLSD test show no significant difference (significance level 95%) between the sites Is- abel, Pagoda, and Sintra, in the three size classes. Fisher PLSD 20=L(mm) 50=L(mm) <50 <70 70 =L (mm) Isabel vs. Pagoda 0.047 0.050 0.054 Isabel vs. Sintra 0.051 0.050 — Pagoda vs. Sintra 0.041 0.049 — All other pairwise comparisons show significant differ- ences (significance level 95%). R. von Cosel et al., 1999 Page 241 Table 2 Table 3 Bathymodiolus azoricus. Length/height ratios calculated Length/height ratios in Bathymodiolus azoricus, Bathy- for each site of the Menez Gwen and Lucky Strike vent modiolus sp. (Logatchev) and B. puteoserpentis from dif- fields, in three length classes. n: sample size. ferent localities on the MAR, after the subsampling pro- cedure. n: subsample size. Mean Standard n ANOVA 20 = L (mm) < 50 Sian Range Mean = dard n Isabel 1.756 0.136 45 Pagoda 1.781 0.120 108 Bathymodiolus azoricus Eiffel Tower S335) OST 535 ee —10:0001 Eiffel Tower 1.387-2.619 2.057 0.201 100 Sintra 1.763 0.133 67 Isabel 1.518-2.343 1.900 0.199 80 Statue of Liberty 2.051 0.130 592 Pagoda 1.595-2.321 1.973 0.163 102 Menez Gwen 2.009 0.122 83 PPS 1.674-2.374 2.036 0.200 62 50 < L (mm) < 70 Sintra 1.491-2.419 1.909 0.199 74 F 1 p= 9) 9 ) Isabel 1.987 0.119 46 Nae a PpeRy ee Aes Fa Aae 5 Pagoda 1.974 0.099 51 Seer are fi pes ae ate: Eiffel Tower 2.036 0.133 151 P = 0.0001 Bathymodiolus Sintra 1.989 0.102 51 Logatchev 1.515-2.219 1.802 O.171 43 Statue of Liberty 2.289 0.144 47 4 Nene’ G wen 2.162 0.126 114 Bathymodiolus Snake Pit 1.624—2.087 1.930 0.127 19 70 = L (mm) a SS ESS ESSE__==___=_=_—_—_==-_-=___=== Isabel 2.080 0.099 42 Pagoda 2.110 0.109 63 P = 0.0001 Eiffel Tower DSS 0.165 116 Menez Gwen 2.286 0.130 89 Figure 66 Bathymodiolus sp., stomach of specimen on Figure 16. Left,: dorsal-lateral view; middle, ventral view; right, stom- ach opened dorsally. oe, esophagus; dd, digestive diverticula duct (entrance); ig, intestinal groove; mty, major typhlosole; gs, gastric shield; big, beginning of intestinal groove; Ip, left pouch; mg, midgut. Scale: 2 mm. Page 242 The Veliger, Vol. 42, No. 3 wR SE Figure 67 Bathymodiolus sp., upper row: labial palps of adult specimens, viewed from anterior end; below left: labial palps of juvenile specimens; below right: cross-section of ctenidia from the region anterior to pericardium; ap, anterior palps; pp, posterior palps; od, outer demibranch; id, inner demibranch; Scale: 2 mm; ctenidia: 1 mm. A systematical subsampling procedure was realized by dividing the length range into 11 classes (20-30 mm; 110-120 mm; larger than 120 mm) and taking 10 indi- viduals in each class when possible, in order to give the same weight to each size category. Table 3 gives length/ height ratios calculated for each subsample (see also Fig- ure 68). Length/height ratios of B. azoricus and Bathymodiolus sp. were compared by means of a one-way ANOVA. In- dividuals of B. puteoserpentis from Snake Pit were not included in this comparison due to their low numbers. Length/height ratios are significantly different (P = 0.0001). For B. azoricus, pairwise comparisons using the Fisher PLSD test show no significant differences (significance level 95%) between Isabel and Sintra, between Eiffel Tower and PPS, and between Statue of Liberty and Menez Gwen: Fisher PLSD Isabel vs. Sintra 0.059 Eiffel Tower vs. PPS 0.060 Statue of Liberty vs. Menez Gwen 0.064 All other pairwise comparisons show significant dif- ferences at the 95% level. In particular, pairwise com- parisons between each subsample of B. azoricus and the subsample of Bathymodiolus sp. from the Logatchev field indicate that length/height ratios in both species are sig- nificantly different at the 95% level. Length/height ratios of B. puteoserpentis from Snake Pit and Bathymodiolus sp. from Logatchev are signifi- cantly different (Mann-Whitney U test, P = 0.003). How- ever, if we consider only individuals larger than 50 mm in length (because the Snake Pit sample contains exclu- sively such large individuals), mean length/height ratios R. von Cosel et al., 1999 Length/height ratio Isabel Sintra Pagoda Eiffel T. Page 243 PPS S. Liberty Menez Gw. Snake P. Logatchev Figure 68 Length-height ratios of a. Bathymodiolus azoricus Cosel & Comtet, sp. nov. from different sites; b. Bathymodiolus puteoserpentis Cosel, Métivier & Hashimoto, 1994; c. Bathymodiolus sp. from Logatchev. Bars are 1 SD. of B. puteoserpentis and Bathymodiolus sp. (respectively 1.930 + 0.127 and 1.944 + 0.140) do not differ signifi- cantly (Mann-Whitney U test, P = 1). Anterior Part Length Additional measurements of the anterior part length (i.e., the distance from the anterior margin to the umbo) were taken on 130 individuals of Bathymodiolus azoricus (75 specimens from Menez Gwen, 32 from Pagoda, and 23 from Eiffel Tower); 22 individuals of Bathymodiolus sp.; and seven individuals of B. puteoserpentis. Measure- Table 4 Total length/anterior part length ratios in Bathymodiolus azoricus, Bathymodiolus sp. and B. puteoserpentis from different localities of the MAR. n: sample size. Total length/anterior part length ratio Standard Mean deviation n Total length range (mm) B. azoricus Menez Gwen 13.785 3.108 75. 13.10—109.70 Lucky Strike 11.782 3.044 55 38.30—101.40 Bathymodiolus sp. Logatchev 6.610 0.936 22 36.90—-123.40 B. puteoserpentis Snake Pit 7.250 0.799 7 98.20—-136.10 ments from Pagoda and Eiffel Tower were pooled and considered as representing B. azoricus from Lucky Strike. Total length/anterior part length ratios were then calcu- lated and compared between the three species (Table 4 and Figure 69). Bathymodiolus azoricus is clearly distinct from Bathy- modiolus sp. and B. puteoserpentis with a total length/ anterior part length ratio being two times higher for B. azoricus. Total length/anterior part length ratios in B. azo- ricus differ significantly between Lucky Strike and Me- nez Gwen (Student t test P = 0.0004), whereas in Bath- Total length/anterior part length ratio Menez Gwen Lucky Strike Figure 69 Ratios of total shell length/anterior part length of a. Bathymodi- olus azoricus Cosel & Comtet, sp. nov. from Lucky Strike and Menez Gwen; b. Bathymodiolus sp. from Logatchev; c. Bathy- modiolus puteoserpentis Cosel, Métivier & Hashimoto, 1994 from Snake Pit. Bars are 1 SD. Logatchev Snake Pit 0 20 40 60 80 100 120 140 Shell length (mm) Figure 70 Shell height (H) vs. shell length (L) for Bathymodiolus azoricus from several sites of the Lucky Strike and Menez Gwen hydro- thermal fields. Each curve represents the allometric model (y = ax’) (Teissier, 1948) fitted to the observed data (not shown). a Sintra H = 0.6838L°9*° r? = 0.9967 b_ Isabel H = 0.6871L°93!” r> = 0.9968 c Pagoda H = 0.7213L°9!58 r? = 0.9954 d_ Eiffel Tower H = 0.7671L°*”° r> = 0.9969 e PP5 H = 0.7504L°*°75 r> = 0.9969 f Menez Gwen H = 0.7724L°%”*? r? = 0.9958 g Statue of Liberty H = 0.7906L°*% r? = 0.9967 ymodiolus sp. and B. puteoserpentis they are significantly different at the 95% level but not at the 99% level (Mann- Whitney U test, P = 0.0415). Remarks: The biometry study shows a great intersite (and interfield) variability of the shell shape in Bathy- modiolus azoricus. However, from length/height ratio comparisons, three groups can be distinguished, corre- sponding to three different morphs: Isabel/Sintra, Eiffel Tower/PP5, and Statue of Liberty/Menez Gwen, the mus- sels from the latter being the most elongate (1.e., having the highest length/height ratio) (Figure 68). Specimens from Pagoda have a length/height ratio intermediate be- tween the two first groups but can be associated with the first one (Isabel/Sintra). This variability is difficult to ex- plain but could be due to intersite differences in the phys- ico-chemical environment (temperature, fluid composi- tion, etc.). Biotic factors such as mussel density could also affect the shell shape. Despite such variability, B. azoricus of each site could be distinguished from Bathymodiolus sp. by the much higher length/height ratio, i.e., having a more elongate shell. The length/height ratio for B. puteoserpentis, inter- mediate between those of B. azoricus and Bathymodiolus sp., 1s not significantly different from that of Bathymo- The Veliger, Vol. 42, No. 3 60 ab 50 40 30 10 0 20 40 60 80 100 120 140 Shell length (mm) Figure 71 Shell height (H) vs. shell length (L) for Bathymodiolus sp. (a) from the Logatchev hydrothermal field and Bathymodiolus pu- teoserpentis (b) from the Snake Pit hydrothermal field. Each curve represents the allometric model (y = ax’) (Teissier, 1948) fitted to the observed data (not shown). a H = 0.9143L°%879 r> = 0.9928 JES 23210 ie52 r? = 0.9033 b H ll diolus sp., when considering a similar length range (>50 mm). Comparisons of the total length/anterior part length ra- tios confirm the morphological similarity between B. pu- teoserpentis and Bathymodiolus sp., whereas B. azoricus is well distinguished from these two species by the much shorter anterior part, i.e., by an almost terminal umbo (Figure 69). DISCUSSION From the study of the gross anatomy, it is evident that unlike the type species of the genus, Bathymodiolus ther- mophilus, all other large hydrothermal vent or cold seep mussels subsequently described under the generic name Bathymodiolus or still under study (Craddock et al., 1995; Cosel & Olu, 1998; Gustafson et al., 1998) have one ma- jor character complex which is distinct from B. thermo- philus: the absence of an inner mantle fold fusion and the reduction of the valvular siphonal membrane to a short transverse sheet at the posterior end of the animal. In these mussels, the “‘ventral gape” stretches over the whole length of the shell, whereas in B. thermophilus, the fusion encloses the mantle cavity to a large extent, leav- ing only a rather small byssal and inhalent mantle open- ing. In connection with this, the other Bathymodiolus spe- cies lack a lateral muscular ridge on mantle lobes and visceral mass, and a posterior branchial septum, which in B. thermophilus divides more completely the excurrent and incurrent chambers. However, other “‘unusual”’ char- acters found by Kenk & Wilson (1985) in B. thermophilus (e.g., thick and very large ctenidia, very large auricles, R. von Cosel et al., 1999 Page 245 Kurchatov 40° 8 Menez Gwen (850m) eo = Azores Lucky Strike (1700m) Broken Spur (3090m) TAG (3650m) Snake Pit (3500m ‘> E> or, Logatchey (3000m 0 1000km —— Oce aNo, Hayes aD he, ° Pico a re Madeira Canaries 30° o ooo o Cape Verde *o 8 209? (from Needham, 1996) Figure 72 Map of the northern part of the Atlantic showing the known hydrothermal fields and transform faults on the MAR (taken from Needham, 1996). symbiose with chemo-autotrophic bacteria) are present in the other species. The difference in the degree of mantle fusion, already discussed by Cosel et al. (1994), could lead to the introduction of a new genus to separate these species from B. thermophilus. However, if one looks at Mytilidae of other genera of which the anatomy is described, none is known to have a similar ventral mantle fusion like Bathymodiolus ther- mophilus. This feature, only known from a hydrothermal vent mussel, has certainly developed after the separation of the B. thermophilus stock on the East Pacific Rise, and is a derived character, and only by chance, the most apomorphic species of the large mussels from hydro- thermal vents was discovered and described first and is hence the name-bearer of the genus. Species without in- ner mantle fold fusion and very short valvular siphonal membrane are much more plesiomorphic in contrast to the apomorphic character of the fused inner mantle folds in the type species of the genus Bathymodiolus. A new supraspecific taxon cannot be introduced based on this. Moreover, an electrophoretic analysis with 18 enzymes by Craddock et al. (1995) revealed a genetic distance D (Nei, 1978) of 1.865 between Bathymodiolus thermophi- lus and the Snake Pit mussel (now B. puteoserpentis) and 1.871 between B. thermophilus and the Lucky Strike mussel (now B. azoricus); the distance between B. puteo- serpentis and B. azoricus is 1.179. The genetic distances (D values) between these three hydrothermal vent mus- sels are within the values usually found with species-level separation (Craddock et al., 1995). Neither the morphological differences nor the genetic distance are a reason for introducing a new genus for the two species here treated, and as a consequence we main- tain them in the genus Bathymodiolus. Another difference, the more complicated stomach and the intestinal coiling in the species here treated ver- sus the simple stomach without left pouch and with only three pairs of entering digestive gland ducts, and the straight digestive tract in B. thermophilus underlines that the latter is more apomorphic: it shows that the degree of direct filter feeding with digestion via the digestive tract has diminished in favor of nutrition by bacteria, whereas in our species, direct feeding remains more im- portant. Some Zoogeographic Remarks Two species (and one population not recognized as full species) of Bathymodiolus associated with hydro- 37°18'N y= By) ie awe A) Ay oF - Bila 4] 37°17'N NG if SS iY 32°17'W 32°16' W OC) Figure 73 Map of the Lucky Strike hydrothermal field. thermal activity are known to date from the MAR, but only a single species is found at one particular site. Their respective geographic range is limited to a few hydro- thermal fields. The northernmost species is B. azoricus, which occurs at the Lucky Strike and Menez Gwen vent fields (37°17'N and 37°50'N, respectively). B. puteoser- pentis inhabits the Snake Pit field (23°N), and Bathy- modiolus sp. occurs on the Logatchev hydrothermal field (14°45'N). It was found that the morphological differ- ences between the populations of Lucky Strike and Me- nez Gwen, on the one hand, and that of Snake Pit, on the other hand, are the most numerous and most obvious and have led us to distinguish two species. Differences between the Snake Pit population and the Logatchev population are less apparent but observable. Mussels from the Broken Spur hydrothermal field (29°10'N) are morphologically close to B. puteoserpentis, and were provisionally identified with this species by the first au- thor. However, it is necessary to be cautious since the morphological analysis was conducted on only two bro- ken shells. One specimen of a mytilid was collected on the TAG hydrothermal field (26°N, 3640-3680 m), but remains to be identified (P. Rona, personal communica- tion). The endemism at specific level for the Mytilidae of the MAR was explained by a combination of depth effect and the role of transform faults on larval dispersal (Craddock et al., 1995; Van Dover, 1995; Van Dover et al., 1996). The Veliger, Vol. 42, No. 3 The bathymetric ranges observed for each species in- dicate that the MAR mussels occupy two discrete bathy- metric intervals. B. azoricus is restricted to shallower hy- drothermal fields (850 m and 1700 m), whereas B. puteo- serpentis and Bathymodiolus sp., as well as the mussels from Broken Spur and TAG, occur at depths exceeding 3000 m. The examination of ultrajuvenile specimens and larvae in the Protoconch II stage from the Logatchev hy- drothermal field showed that at least two different bivalve species are present, most probably the other also a mus- sel, so it is probable that another mytilid species reaches the site as planktonic larvae but does not find adequate conditions for settling. Transform faults, which are numerous and various along the MAR (Needham, 1996), have been considered as barriers to the dispersal of vent invertebrates (e.g., Craddock et al., 1995; Van Dover, 1995; Van Dover et al., 1996) but their role may depend on the dispersal strategies of each species. Hydrothermal vent mytilids, and especially the species from the MAR, have a plank- totrophic larval development, as inferred from size and morphology of the protoconch II (Lutz et al., 1980; Co- sel et al., 1994; this paper), which allows the larvae to remain in the plankton for at least several weeks and to be widely dispersed by currents, after being driven by the plume to the level of lateral spreading several hun- dred meters above the bottom (Kim et al., 1994; Mul- lineaux & France, 1995; Mullineaux et al., 1995). One might expect that such a dispersal strategy would permit mytilid larvae to colonize vent fields separated by great distances regardless of transform faults or other phys- iographic features. Thus, the 45 km offset of the Pico transform fault (Figure 72), which separates Lucky Strike from Menez Gwen (with a distance of 60 km), does not constitute a major obstacle for the dispersal of B. azoricus. However, if the mussels from Broken Spur are clearly identified as B. puteoserpentis in the future, they would be the northernmost population of the spe- cies. The presence of three major faults (Oceanographer, Hayes, and Atlantis faults) between Lucky Strike and Broken Spur is a possible factor for differentiation be- tween both areas. The Kane fracture zone, with an offset of 145 km, which separates Snake Pit and TAG/Broken Spur, does not seem to prevent dispersal of B. puteoser- pentis, but the Fifteen-Twenty transform fault could ex- plain reduced exchanges between Snake Pit and Lo- gatchev. In addition to the possible role of transform faults, the low spatial frequency of occurrence of hydro- thermal] fields on the MAR, estimated at one field every 175 km by German et al. (1995), could limit the dis- persal by lack of step-by-step processes. However, the occurrence of several other vent species (e.g., the alvi- nocaridid shrimp, Chorocaris chacei (Williams & Rona, 1986), the bythograeid crab, Segonzacia mesatlantica (Williams, 1988), the gastropod, Protolira valvatoides Warén & Bouchet, 1993, and the commensal polynoid R. von Cosel et al., 1999 polychaete, Branchipolynoe seepensis), with a wider geographical range on the MAR (Van Dover et al., 1996; Gebruk et al., 1997; Segonzac, personal communica- tion), suggests that dispersal could occur along the entire MAR despite transform faults, and that these are not the only factors controlling the mytilid geographical distri- bution. Further investigations at intermediate depths and latitudes will allow us to precisely determine the role of depth and topography of the ridge in the distribution of mytilid species along the MAR. ACKNOWLEDGMENTS We would like to express our sincere gratitude to C. Langmuir (Lamont-Doherty Earth Observatory), Y. Fou- quet, D. Desbruyéres and A.-M. Alayse (IFREMER Brest), D. Prieur (Station Biologique de Roscoff), and all the participants of the LUCKY STRIKE 1993, LO- GACHEV-7, Akademik Mstislav Keldysch cruise 35, DIVA 1, DIVA 2, and MICROSMOKE expeditions. We thank M. Segonzac (CENTOB, French National Sorting Center, IFRREMER, Brest) and L. I. Moskalev (P. P. Shir- shov Institute of Oceanology) for placing the collected material at our disposal. We are grateful to S. Gofas and two anonymous referees for critically reading the man- uscript and giving hints to improve it. For help in pre- paring the specimens for SEM photos, we kindly thank S. Gofas; the assistance of Mrs. Grassé (CIME, Dept. MEB, Université Paris-VI) for the SEM photos is grate- fully acknowledged. This work was supported in part by a contribution from the Programme “‘DORSALES” (Contrat IFREMER 961410106), in part by the Euro- pean Community (MAST3-CT95-0040 AMORES), and in part by INTAS (International Association for the Pro- motion of Cooperation with Scientists from the Inde- pendent States of the former Soviet Union) (project number 94-0592). LITERATURE CITED BaTuyeEv, B. N., A. G. Krotov, V. EK Markov, G. A. CHER- KASHEV, S. G. Krasnov & Y. D. Lisitsyn. 1994. Massive sulphide deposits discovered and sampled at 14°45’N, Mid- Atlantic ridge. BRIDGE Newsletter 6:6—10. CuHILbREss, J. J. & C. R. FISHER. 1992. The biology of hydro- thermal vent animals: physioiogy, biochemistry, and auto- trophic symbioses. Oceanography and Marine Biology An- nual Review 30:337—441. ComteT, T. 1994. Etude de la croissance allométrique et de la structure des populations de modioles sur la zone hydro- thermale Lucky Strike (37°17'N sur la ride médio-atlan- tique). Master’s Thesis (DEA, Diplome d’Etudes Approfon- dis), Université de Bretagne Occidentale, Brest. CoseEL, R. von & K. OLu. 1998. Gigantism in Mytilidae. A new Bathymodiolus from cold seep areas on the Barbados accre- tionary Prism. Comptes rendus de |’ Academie des Sciences, Paris, Sciences de la vie, 1998, 321:655—663. CosEL, R. von, B. METIVIER & J. HASHIMOTO. 1994. Three new species of Bathymodiolus (Bivalvia: Mytilidae) from hydro- Page 247 thermal vents in the Lau Basin and the North Fiji Basin, Western Pacific, and the Snake Pit area, Mid-Atlantic Ridge. The Veliger 37(4):374—-392. Crappock, C., W. R. HoeH, R. G. GusTarson, R. A. Lutz, J. HASHIMOTO & R. J. VRIJENHOEK. 1995. Evolutionary rela- tionships among deep-sea mytilids (Bivalvia: Mytilidae) from hydrothermal vents and cold-water methane/sulfide seeps. Marine Biology 121:477—485. DESBRUYERES, D., A.-M. ALAYSE, E. ANTOINE, G. BARBIER, FE BaARRIGA, M. Biscoito, P. BRIAND, J.-P. BRULPORT, T. Com- TET, L. CoRNEC, P. Crassous, P. DANDO, M.-C. Fart, H. FELBECK, FE LALLIER, A. FIALA-MEDIONI, J. GONGALVES, FE MENARD, J. KERDONCUFF, J. PATCHING, L. SALDANHA & P.- M. SARRADIN. 1994. New information on the ecology of deep-sea vent communities in the Azores Triple Junction area: preliminary results of the Diva 2 cruise (May 31—July 4, 1994). Inter Ridge News 3:18-19. FIALA-MEDIONI, A., C. CAVANAUGH, P. DANDO & C. L. VAN Do- VER. 1996. Symbiotic mussels from the Mid-Atlantic Ridge: adaptations to trophic resources. Journal of Conference Ab- stracts 1:788. Fouquet, Y., H. ONDREAS, J-L. CHARLOU, J.-P. DONVAL, J. RAD- FORD-KNOERY, I. Costa, N. LOURENGO & M. K. Tivey. 1995. Atlantic lava lakes and hot vents. Nature 377:201. GEBRUK, A. V., S. V. GALKIN, A. L. VERESHCHAKA, L. I. Mos- KALEY & A. J. SOUTHWARD. 1997. Ecology and biogeogra- phy of the hydrothermal vent fauna of the Mid-Atlantic Ridge. Advances in Marine Biology 32:93—144. GERMAN, C. R., E. T. BAKER & G. KLINKHAMMER. 1995. Regional setting of hydrothermal activity. Pp. 3-15 in L. M. Parson, C. L. Walker & D. R. Dixon (eds.), Hydrothermal Vents and Processes. Geological Society: Boulder, Colorado. GusTAFSON, R. G., R. D. TURNER, R. A. Lutz & R. C. VRUEN- HOEK. 1998. A new genus and five new species of mussels (Bivalvia: Mytilidae) from deep-sea sulfide/hydrocarbon seeps in the Gulf of Mexico. Malacologia 40(1—2):63-112. HasHImMoTo, J. & T. OKUTANI. 1994. Four new mytilid mussels associated with deepsea chemosynthetic communities around Japan. Venus (Japanese Journal of Malacology) 53(2):61-83. Kenk, V. C. & B. R. WILSon. 1985. A new mussel (Bivalvia: Mytilidae) from hydrothermal vents in the Galapagos Rift zone. Malacologia 26(1-2):253—271. Kim, S. L., L. S. MULLINEAUX & K. R. HELFRICH. 1994. Larval dispersal via entrainment into hydrothermal vent plumes. Journal of Geophysical Research C 99:12655—12665. Le PENNEC, M. & A. HiLy. 1984. Anatomie, structure et ultra- structure de la branchie d’un Mytilidae des sites hydrother- maux du Pacifique oriental. Oceanologica Acta 7(4):517— 523. Lutz, R. A., D. JABLONSKI, D. C. RHOADS & R. D. TURNER. 1980. Larval dispersal of a deep-sea hydrothermal vent bivalve from the Galapagos Rift. Marine Biology 57:127—133. MULLINEAUX, L. S. & S. C. FRANCE. 1995. Dispersal mechanisms of deep-sea hydrothermal vent fauna. Pp. 408—424 in S. E. Humphris, R. A. Zierenberg, L. S. Mullineaux, & R. E. Thomson (eds.), Seafloor Hydrothermal Systems: Physical, Chemical, Biological, and Geological interactions. American Geophysical Union: Washington, DC. MULLINEAUX, L. S., P. H. Wiepe & E. BAKER. 1995. Larvae of benthic invertebrates in hydrothermal vent plumes over Juan de Fuca Ridge. Marine Biology 122:585—596. Murton, B. J., C. VAN DOVER & E. SOUTHWARD. 1995. Geolog- ical setting and ecology of the Broken Spur hydrothermal The Veliger, Vol. 42, No. 3 vent field: 29°10’N on the Mid-Atlantic Ridge. Pp. 33-41 in L. M. Parson, C. L. Walker & D. R. Dixon (eds.), Hy- drothermal Vents and Processes. Geological Society: Boul- der, Colorado. NEEDHAM, H. D. 1996. Some features of the North America- Africa plate boundary. Journal of Conference Abstracts 1: 834-835. Ne!, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89: 583-590. TEISSIER, G. 1948. La relation d’allométrie. Sa signification sta- tistique et biologique. Biometrics 4:14—53. VAN Dover, C. L. 1995. Ecology of Mid-Atlantic Ridge hydro- thermal vents. Pp. 257-294 in L. M. Parson, C. L. Walker & D. R. Dixon (eds.), Hydrothermal Vents and Processes. Geological Society: Boulder, Colorado. VAN Dover, C. L., D. DESBRUYERES, M. SEGONZAC, T. COMTET, L. SALDANHA, A. FIALA-MEDIONI & C. LANGMUIR. 1996. Bi- ology of the Lucky Strike hydrothermal field. Deep-Sea Re- search I 43:1509—-1529. Note Added in Proof: After acceptance of this manu- script, two other papers with descriptions of mussels from hydrothermal vents and cold seeps were published: Cosel & Olu (1998) and Gustafson et al. (1998), which unfor- tunately could not be considered here. In total, six new species were described, among them four in the genus Bathymodiolus. This augments the total number of de- scribed and named species of hydrothermal vent and cold seep mussels to 13. The Veliger 42(3):249—259 (July 1, 1999) THE VELIGER © CMS, Inc., 1999 Shell Form and Color Variability in Alia carinata (Neogastropoda: Columbellidae) JEFF W. TUPEN TENERA, Inc., Ecological Services Division, PO. Box 400, Avila Beach, California 93424, USA Abstract. Alia carinata from four different nearshore habitats in coastal central California were analyzed to inves- tigate shell form and color variability. Analysis of select shell dimensions showed that A. carinata from an intertidal red alga, eelgrass, and benthic rock habitats displayed measurably and identifiably distinct forms. Individuals from giant kelp canopies showed considerable form overlap with benthic specimens, and likely represented benthic migrants. In- terhabitat form variability was related to differences in both shape and size, while observed sexual dimorphism appeared to be strictly size-related, with males larger than females. A. carinata from eelgrass were mostly unpatterned and dark in color, while those from the other three habitats were generally patterned and variably colored. Planktonic dispersal of juveniles and the lack of discontinuous shell phenotypes suggest that observed interhabitat form variability was not a result of developmental polymorphism. Rather, phenotypic plasticity and post-settlement selection, potentially resulting from predation and wave exposure differences among sampled habitats, are suspected as important mechanisms in observed intraspecific shell form and color variability. INTRODUCTION Alia carinata (Hinds, 1844) are planktotrophic, determi- nate-growth gastropods which range from Forrester Is- land, Alaska, to southern Baja California (McLean, 1978, 1996). Alia carinata rarely exceed 11 mm in shell height, and are found commonly on intertidal rocks and algae, within the surfgrass Phyllospadix spp., and subtidally on kelp stipes and holdfasts (Abbott & Haderlie, 1980). Jones (1971) noted that A. carinata may sometimes be the most abundant animal living on the kelps Macrocystis spp. Densities of 110 individuals per 0.01 m* have been observed within an intertidal red algae habitat (Tupen, unpublished data). At maturity, this species displays variable shell color- ation, although most are some variation of a light to dark brown base color with fine to large markings or regular patterning (Crane, 1969; Carlton & Roth, 1975; McLean, 1978; Abbott & Haderlie, 1980; Carter & Behrens, 1980). Carter & Behrens (1980) referred to several color variants of A. carinata: a ““common banded form” from the rocky subtidal; a patterned ‘“‘open coast’ form; and a “‘predom- inantly uniform dark brown to black”? form collected from eelgrass (Zostera marina Linnaeus). Bergman et al. (1983) demonstrated that A. carinata from a sheltered harbor environment displayed highly developed shoulder keels, and large shell width to shell height ratios, relative to those collected from an exposed rocky cove. Intraspecific shell color and pattern variability are well known in marine prosobranchs, e.g., Hughes & Mather, 1986, in Littorina sp. (Littorinidae); Etter, 1988, in Nu- cella sp. (Thaididae); Langan-Cranford & Pearse, 1995, in Lacuna spp. (Lacunidae). Although intraspecific shell form variability is not uncommon in gastropods—e.g., Naylor & Begon, 1982, in Littorina sp.; Dillon, 1984, in Goniobasis sp. (Pleuroceridae); Tissot, 1984, in Cypraea sp. (Cypraeidae); Katoh & Foltz, 1994, in Viviparus sp. (Viviparidae)—the literature addressing the nature and extent of, or potential factors controlling, shell form var- lability in A. carinata is limited (i.e., Bergman et al., 1983). The purpose of this study was to identify and describe intraspecific shell form and color variability among Alia carinata from four different habitats, and to generate hy- potheses concerning the controlling factors and adaptive advantages of variable shell forms. This study was not specifically intended to characterize habitat-dependent forms, as this would have required a different study de- sign. Results of this research demonstrated that: (1) both size- and shape-related distinct shell forms existed in three of four sampled habitats; (2) sexual dimorphism was size-related; and (3) shell pattern frequencies were habitat-dependent. MATERIALS AnD METHODS Specimen Collection and Processing Identifications of Alia carinata were based on shell de- scriptions in Carlton & Roth (1975) and McLean (1978). Alia tuberosa (Carpenter, 1864), the only local congener of A. carinata, are distinguished by their flattened whorls and smaller shell heights at maturity (Carlton & Roth, 1975; McLean, 1978). Approximately 75—100 individuals of A. carinata were collected off central California, from Page 250 Morro Bay (35°22'N, 120°52'W) south to Diablo Canyon (35°12'N, 120°51'W), San Luis Obispo County, from each of the following habitats and approximate tidal el- evations: subtidal rocky benthic (—10 m mean lower low water, MLLW, hereafter abbreviated RB in text); intertidal Gastroclonium subarticulatum (Turner) Kttzing (Rho- dophyta, —0.3 m MLLW, hereafter abbreviated GS in text); shallow subtidal eelgrass blades (Zostera marina, —2 m MLLW, hereafter abbreviated ZM in text); and gi- ant kelp (Macrocystis pyrifera (Linnaeus) Agardh, —1.3 m MLLW, hereafter abbreviated MP in text) canopy blades. The first three habitats were selected to represent those in which Carter & Behrens (1980) and Bergman et al. (1983) had noted characteristic A. carinata shell colors and forms. The habitat MP was selected because my qual- itative observations suggested a habitat-specific shell col- or variant. Most A. carinata were collected during the winter and spring of 1995 and 1996, except for specimens from GS, which were collected during the winter, spring, and sum- mer of 1993 and 1994. Bias during collection in the MP and ZM _ habitats was minimized by indiscriminately scraping fronds and leaves, respectively, by hand into a collection jar. Specimens of A. carinata from GS were collected by completely scraping algae masses (0.01 m° patches, collected for a related project) from their rock substrates and later sorting A. carinata from the algae. Collection bias was less easily avoided in the RB habitat, as A. carinata often occurred in depressions and on ir- regular surfaces, which made indiscriminate sampling difficult. In this habitat, rocky substrate was examined closely and all A. carinata observed were collected. Determination of shell patterning and form was some- times prohibited by the presence of epizooic, non-genic- ulate coralline algae (Rhodophyta: Corallinaceae). The majority of A. carinata specimens from MP and ZM hab- itats were encrusted by coralline algae by the middle of each year (May—June), while earlier collections of A. car- inata were rarely encrusted. Attempts at removing the algae by scraping with forceps proved difficult. Conse- quently, analysis of color and form in this study was re- stricted to non-encrusted specimens. In addition, only un- broken individuals were considered for analysis. These field samples therefore resulted in the collection of the following number of eligible, conchologically adult A. carinata (evidenced by a thickened and reflected aperture outer lip) from each of the following habitats: RB (n = 54); GS (n = 66); MP (n = 52); and ZM (n = 50). Fifteen male and 15 female unbroken individuals were randomly selected from each of these four groups for morphometric analysis. Male individuals were identified by the presence of the conspicuous penis. Each snail was drawn with a camera lucida and dis- secting microscope at 15, using the position illustrated in Figure 1 and positioning methods described in Coppois & Glowacki (1983). The following six dimensions were The Veliger, Vol. 42, No. 3 Figure 1 Alia carinata shell dimensions measured for morphometric ana- lyses: ShHt, shell height; ShWd, shell width; SpHt, spire height; SpWd, spire width; ApHt, aperture height; ApWd, aperture width; AA, apical angle; KA, keel angle. See text for landmark descriptions. measured from each magnified drawing using a slide cal- iper (instrument precision 0.1 mm): shell height (ShHt); shell width (ShWd); spire height (SpHt); spire width (SpWd); aperture height (ApHt); and aperture width (ApWd) (Figure 1). All of these dimensions were mea- sured parallel or perpendicular to the central axis of the shell. ApWd did not include the shoulder keel, if present, but was landmarked laterally between the usable, func- tional aperture opening (Figure 1). SpWd was the widest dimension of the penultimate whorl. SpHt was landmar- ked between the apex and the body whorl suture at its intersection with the shell central axis. Caliper measure- ments of magnified drawings were converted to actual dimensions in millimeters referencing an actual shell height determined with a dial micrometer (instrument pre- cision 0.025 mm). Apical angle (AA), in degrees, was calculated by computing the arc tangent of the SpWd: SpHt ratio. This dimension was calculated as an esti- mator of whorl expansion rate. The extent of shoulder carina (keel) development was quantified with the vari- able keel angle (KA), measured from the magnified draw- ing in degrees with a protractor (instrument precision 0.5°). The variable KA was defined as the angle formed between a vertical line drawn through the right marginal body whorl suture (vertex), and a line drawn from this vertex to the widest point of the aperture (Figure 1). Distortion with the camera lucida potentially due to portraying three-dimensional objects in two-dimensional space was minimized by using the same microscope at the same magnification for conchological adults identi- Te Wey tupens 999 Page 251 Table 1 Alia carinata shell variables used in the morphometric analyses. Values reported were obtained by repeated mea- surements (n = 10) on the same individual, see text for methods. Apical and keel angle in degrees, all other di- Mensions in mm. Symbol Mean SD Min Max Variable Shell height ShHt 7.96 0.07 7.9 8.1 Shell width ShWd 4.12 0.05 4.0 4.2 Spire height SpHt Ail 0.05 De 2.9 Spire width SpWd 2.34 0.02 D3) 2.4 Aperture height ApHt 333 0.04 33 3.4 Aperture width ApWd 153) 0.05 1.4 1.6 Apical angle AA 40.18 0.44 39.5 40.9 Keel angle KA Sls 7X0) 1.09 50.0 53.0 cally positioned in the center of the field of view. Mea- surement error was estimated by drawing one test indi- vidual with the camera lucida on 10 separate occasions, removing it from, and repositioning it on, the stage after each drawing. The means, standard deviations, mini- mums, and maximums from these repetitions are pre- sented in Table 1. This exercise demonstrated that the methods used in this study provided repeatable and ac- curate estimations of most shell variables measured. The variable KA was sensitive to slight differences in posi- tioning on the microscope stage, and this was reflected by its relatively low precision (SD = 1.09°). Data Screening Although MANOVA and ANOVA are robust to devi- ations from normality and variance homogeneity if sam- ple sizes are fairly large and equal among groups (Pi- mentel, 1979; Zar, 1996), the presence of univariate or multivariate outliers can cause serious deviations from these assumptions and lead to results that distort statistics and do not generalize to the population(s) being sampled (Tabachnick & Fidell, 1996; Zar, 1996). Univariate out- liers, by habitat and sex, were identified by comparing standardized (z-score) variable values against a critical score of 3.29 at a = 0.001 (2-tailed). Multivariate outli- ers, by habitat and sex, were identified by inspecting Ma- halanobis distances (D*), evaluated against a critical 7 value of 26.12 at a = 0.001, with 8 degrees of freedom (corresponding to the number of variables). Tabachnick & Fidell (1996) recommended that both of these outlier screening procedures be evaluated at a very conservative a = 0.001. Interhabitat Shell Form Univariate size and shape differences among habitats were examined using ANOVA and ANCOVA. Prior to ANCOVA, dependent variables, by habitat, were first tested for significant linear relationships with the covar- iate, and then tested to ensure homogeneity of slopes among groups. The ratio ShWd: ShHt was analyzed with ANOVA to provide comparative data for the only quan- titative reference on A. carinata shell form variability to date (Bergman et al., 1983). Shoulder keel development, using a rating scale (1 to 5, where 1 represented least keeled, and 5 represented most keeled) presented in Berg- man et al. (1983), was analyzed using a Kruskal-Wallis non-parametric ANOVA (H-statistic). Unplanned pair- wise comparisons following significant ANOVA results were Student’s t-tests for the parametric case, and Tukey- type Nemenyi tests (q-statistic) for the non-parametric case (Zar, 1996). Bonferroni adjustments were made to the calculated comparison-wise error rates to protect against Type I error rate inflation. The experiment-wise error rate was maintained at a = 0.05 for all analyses in this study unless noted otherwise. MANOVA was used to test for among-group popula- tion centroid differences using the eight variables from this study. The two variables analyzed in Bergman et al. (1983)—the ordinal scale keel development rating, and the ratio ShWd : ShHt—are arguably and conditionally in- appropriate for multivariate analyses (Pimentel, 1979; Ta- bachnick & Fidell, 1996), and were not included in the MANOVA. If a significant difference existed among groups, canonical variate analysis (CVA) was used to in- terpret these differences. A forward selection, F-to-enter, stepwise CVA procedure was used to determine which subset of the eight variables was optimal in separating group forms. An optimal subset was defined by Klecka (1980) as that set of variables resulting from a stepwise variable selection method (in the present case, F-to-enter), whereby each variable provides a unique (non-redundant) contribution, relative to others in the subset, to discrimi- nate among groups. Costanza & Afifi (1979) recommend- ed a liberal alpha level for F-to-enter of a = 0.15 to ensure entry of important variables into the CVA. This stepwise process minimized the potential for including highly intercorrelated variables in the CVA, as these may complicate the analysis (Klecka, 1980). Derived canoni- cal variates were interpreted if they contributed to de- scribing 90% of the cumulative proportion of total vari- ation among groups. Tissot (1990) noted that when com- bined with biological interpretation, the 90% rule suffi- ciently separates real from trivial variates. Case scores were plotted in discriminant space with 95% confidence ellipses around each group centroid as an estimate of true population differences. The strength of association be- tween groups and interpreted variates was determined with canonical correlation coefficients, R, which, when squared, indicate the proportion of variance shared be- tween groups and variables on that canonical variate. Those variables most important in providing intergroup discrimination were identified with canonical loadings, r, correlations between the interpreted canonical variates Page 252 and variables (Tabachnick & Fidell, 1996). Standardized canonical coefficients revealed the size- or shape-related nature of interpreted variates (Pimentel, 1979). A jackknifed group classification procedure was used on all specimens, where each specimen was classified us- ing variates derived from the other 119 specimens, to test the classification success of the derived variates. Tabachn- ick & Fidell (1996) noted that the entry order of variables during a stepwise CVA may be biased by trivial relation- ships among variables within the sample that do not re- flect true relationships in the population. To test the sta- bility of the jackknifed classification procedure, and to test for variable entry bias, a split-sample validation pro- cedure was used (Tabachnick & Fidell, 1996). In this pro- cedure, canonical variates were recalculated using 96 (80%) randomly selected individuals (learning cases) from the original group of 120 specimens (12 per sex and habitat), and the remaining 24 individuals (test cases) were then classified to habitat using the newly derived variates (SYSTAT, 1996; Tabachnick & Fidell, 1996). Similarities in classification successes among the jack- knifed, learning, and test runs were interpreted as evi- dence of model stability and minimal variable entry bias. Inter- and Intrahabitat Sexual Dimorphism The same statistical procedure described in the previ- ous shell from section was used to determine: (1) if sex- ual dimorphism was evident within A. carinata using all individuals pooled; (2) those variables imporiant in ob- served dimorphism, if present, and; (3) if dimorphism was consistent among all habitat groups. Interhabitat Shell Patterning All individuals were assigned to either “‘variegated”’ or “solid” color categories, based on the presence or ab- sence of any deviation from a uniform, unpatterned ex- terior on the body whorl. Chi-square tests of constructed contingency tables were used to test the null hypothesis that color frequency was independent of group. Rejection of the null hypothesis (demonstrating non-independence) in the among-habitat test was followed with manipula- tions of the contingency tables to identify the disparate habitat(s) (Zar, 1996). All statistics were calculated using version 6.0.1 of SYSTAT for Windows (SYSTAT, 1996). Dry shell vouchers of Alia carinata from each of the four habitats analyzed in this study are deposited at the Los Angeles County Museum of Natural History, Los Angeles, Cali- fornia, catalog numbers LACM 152413-152420. RESULTS Data Screening Two univariate outliers were identified in the sex group screening process, and one case was identified as a mul- The Veliger, Vol. 42, No. 3 tivariate outlier with both the habitat and sex group screening processes. The univariate outliers included a large female (8.4 mm ShHt) with an exceptionally large SpWd (2.7 mm, z = 3.308), and a large male (9.6 mm ShHt) with an exceptionally large ApHt (4.2 mm, z = 3.421). The multivariate outlier case was male (D*? = 26.917) from MP (D? = 28.546) that displayed a very broad form (4.1 mm ShW4d) relative to its unexceptional ShHt (7.5 mm). A log 10(X) transformation of the vari- ables SpWd and ApHt corrected the univariate outlier problems. The multivariate outlier case was deleted from the data set and replaced with a randomly selected indi- vidual from the same habitat and sex. A recheck of the new data set revealed no univariate or multivariate out- liers, by either sex or habitat. Interhabitat Shell Form All variables, except ShWd, differed significantly among habitats (Table 2). Pairwise comparisons revealed that: A. carinata from ZM were generally short, with rel- atively small spires and apertures, and large apical and keel angles; specimens from GS were tall, with relatively large apertures; specimens from MP were tall, with rel- atively tall spires; and A. carinata from the RB habitat displayed features similar to those from MP. The variable ShHt was significantly different among habitats (F = 6.32, P < 0.001, df = 3 and 116) and was designated as a covariate in ANCOVA. With the influ- ence of size extracted or minimized, all variables tested were significantly different among habitats, demonstrat- ing a substantial shape component in among-habitat var- iable differences (Table 3). The variables AA and KA were not analyzed because they did not relate signifi- cantly to the covariate. Pairwise comparisons showed that most variables tested also had substantial size compo- nents, as relationships among habitats differed from those detected with ANOVA. This analysis showed, indepen- dent of size, that: GS forms of A. carinata were narrow with small spires and large apertures; ZM forms were very broad with short spires; MP forms were fairly broad with tall spires; and RB forms displayed large spires and small apertures. The ratio ShWd:ShHt, sensu Bergman et al. (1983), differed significantly among habitats (Table 2). Pairwise comparisons indicated that A. carinata from GS were sig- nificantly narrower than those from the RB and MP hab- itats. Alia carinata from ZM were significantly wider than all other habitat forms (Table 2). Shoulder keel develop- ment in A. carinata, sensu Bergman et al. (1983), was significantly different among the four habitats (Table 4). Pairwise comparisons of rank sums showed that keel de- velopment in specimens from GS was significantly less than that of the ZM (q = 10.1, P < 0.001, df = 4), MP (q = 8.2, P < 0.01), and RB (q = 6.4, P < 0.01) habitats. No other significant differences between paired rank sums J. W. Tupen, 1999 Page 253 Table 2 ANOVA results for Alia carinata from four sampled habitats: subtidal rocky benthic (B), Gastroclonium subarticulatum (G), Macrocystis pyrifera (M), and Zostera marina (Z), n = 30 for each habitat. Means and standard errors for AA and KA in degrees, ShWd:ShHt unitless, all others in mm. Variable means significantly different (* P < .05, ** P < .001), not significantly different (ns), among habitats. PC. = Bonferroni-adjusted t-test. © S Ms 4 ANOVA Variable Mean SE Mean SE Mean SE Mean SE F (3,116) PC. ShHt 7.36 0.09 7.74 0.16 7.59 0.09 7.11 0.08 6.32** G,M > Z ShWd 3.60 0.04 3.67 0.06 B12) 0.05 3.63 0.03 1.14 ns SpHt 2.62 0.04 2.60 0.06 2.68 0.05 2.34 0.04 10.00** M,B,G > Z SpWd 232) 0.03 231 0.04 D382) 0.03 2.21 0.02 2.80* B,M > Z ApHt 3.03 0.03 3.31 0.07 3.14 0.03 3.03 0.03 8.41** G>B,Z ApWd 1.45 0.02 1.60 0.03 1.53 0.02 1.48 0.02 9.73** G>Z,B AA 41.6 0.3 41.8 0.4 41.0 0.4 43.5 0.2 11.54** Z > G,B,M KA 41.8 0.8 29.8 1.1 45.3 162 45.8 0.9 55.15** Z>B>G;M>G ShWd:ShHt 0.489 0.003 0.475 0.003 0.490 0.004 0.512 0.004 16.79** Z>M,B>G were detected. Only four percent (4%) of A. carinata from GS were strongly keeled (rating 3—5). In contrast, 97% of individuals from ZM, 87% from MP, and 80% from the RB habitat were strongly keeled. Significant multivariate differences existed among hab- itat group forms (F = 10.87, P < 0.001, df = 24 and 316). The stepwise CVA revealed that the variables KA, SpWd, AA, ApWd, and ShHt (in decreasing order of rel- ative importance) provided optimal information in dis- criminating among habitat forms; following entry of these five variables into the CVA, no other variables provided significant additional discrimination among groups. The first and second canonical variates (canonical correlations r = 0.82 and r = 0.57, respectively) together summarized almost 93% of the intergroup differences. The third ca- nonical variate was not interpreted because it summarized less than 10% of the variation among groups. Most hab- itat group centroids were significantly different from each other using the five optimal variables on the first and second canonical variates, evidenced by non-overlapping 95% confidence ellipses around habitat centroids in Fig- ure 2. However, no significant difference existed between population forms of A. carinata from MP and RB habitats on the first or second variate. Canonical loadings showed the relative importance of variables to each interpreted canonical variate (Table 5). Variate 1, summarizing 74.9% of intergroup variation, largely represented the variable KA (canonical loading r = —0.81), most important in separating GS forms of A. carinata from the other three habitat groups. Large pos- itive scores on variate 1 represented relatively small shoulder keels. Less important were the variables ApWd (ry = 0.31) and ShHt (7 = 0.19). Large positive scores on variate 1 represented relatively large apertures widths and shell heights. The variables SpWd and AA did not con- tribute substantially to variation described on the first var- iate. The mix of positive and negative canonical coeffi- cients on canonical variate 1 (Table 5) indicated that this variate primarily described shape, rather than size, dif- ferences among groups: GS forms of A. carinata were small-shouldered compared to RB, MP, and ZM habitat forms, independent of size. The second variate, summarizing 17.8% of intergroup variation, largely represented the variable AA (7 = 0.69), Table 3 ANCOVA results for Alia carinata from four sampled habitats, n = 30 for each habitat. Adjusted (for covariate ShHt) means and standard errors in mm. All variable means significantly different (* P < 0.001) among habitats. PC. = Bonferroni-adjusted t-test. KA, AA, and ShWd:ShHt not analyzed with ANCOVA, see text. Habitat abbreviations follow Table 2. ; B G M Vari- able Mean SE Mean SE Mean ShWd 3.63 0.03 3.56 0.03 3.66 SpHt 2.65 0.02 2.49 0.02 2.62 SpWd 2.34 0.02 2.23 0.02 2.28 ApHt 3.06 0.02 3.18 0.02 3.09 ApWd 1.46 0.02 1.56 0.02 1.51 SE 0.03 0.02 0.02 0.02 0.02 i ANCOVA Mean SE F (3,115) PC. 3.76 0.03 10.08* Z>B,G;M>G 2.47 0.02 19.54* B,M > G,Z 2.29 0.02 8.41* B>M,G 3.14 0.02 8.21* G>M,B;Z>B 1.52 0.02 pone: G,Z>B Page 254 The Veliger, Vol. 42, No. 3 Table 4 Number and percent of A. carinata by keel rating and hab- itat, n = 30 for each habitat. Extent of shoulder keel de- velopment evaluated using an assigned rating of 1 (least keeled) to 5 (most keeled). G. subarticulatum (G) rank sum significantly less (H = 60.4, P < 0.001, df = 3) than others at a = 0.05. See Bergman et al. (1983) for illustrations of keel ratings. Habitat abbreviations follow Table 2. Keel Habitat rating G B M Z 1 11 (37%) 0) 0) (0) 2 15 (50%) 6 (20%) 4 (13%) 1 (3%) 3 4 (13%) 12 (40%) 8 (27%) 6 (20%) 4 0 10 (33%) 13 (43%) 14 (47%) 5 0) 2 (7%) 5 (17%) 9 (30%) Rank sum 643.5 1862 2198.5 2556 most important in separating ZM forms of A. carinata from RB and MP forms (Figure 2). Large positive scores on variate 2 represented relatively large apical angles. Less important on canonical variate 2 were the variables KA (r = 0.38), SpWd (r = —0.37), and ShHt (7 = —0.32). Large scores on variate 2 represented relatively large keel angles, and relatively small spire widths and shell heights. The variable ApWd did not load on variate 2 (Table 5). The mix of positive and negative coefficients on canonical variate 2 indicated that this variate also de- scribed shape differences among habitat forms. Specifi- cally, specimens of A. carinata from ZM had larger apical angles, and by definition, greater whorl expansion rates, than specimens from either RB or MP habitats, indepen- dent of size. The CVA and the previous univariate ana- lyses demonstrated that large apical angles in ZM forms of A. carinata were a consequence of shorter spire heights, rather than larger spire widths, relative to overall shell size. Overall jackknifed classification success of habitat forms was moderately high (65.8%), with the following individual habitat forms of A. carinata correctly classi- fied: RB 60%, GS 90%, MP 43.3%, and ZM 70% (Table 6). These percentages, with the exception of that for MP forms, are substantially higher than the 25% classification success that would statistically occur by chance alone. Classification results reflect canonical graph data (Figure 2), as A. carinata from the MP habitat were most often misclassified as RB forms. In the split-sample validation procedure, stepwise CVA of 96 randomly selected specimens resulted in the iden- tification of an optimal subset of variables that included KA, ApWd, AA, and SpWd, in descending order of rel- ative importance. Variates derived from these variables and individuals resulted in the correct jackknifed classi- fication of 62 of the 96 (64.6%) learning specimens. Six- Canonical variate 2 (17.8 %) 4 he 2 3. Al Canonical variate | (74.9 %) Figure 2 Plot of Alia carinata habitat groups on the first and second ca- nonical variates. Letters (B, G, M, Z) represent locations of hab- itat group centroids, and symbols represent individual case scores (circles = benthic hard bottom, + = Macrocystis pyrifera can- opy, X = Gastroclonium subarticulatum, triangles = Zostera ma- rina, n = 30 for each habitat). Ellipses are 95% confidence limits for population group centroids. Relative percentages of total var- lation among groups explained by each variate are shown in pa- rentheses. teen of the 24 (67.7%) remaining test cases were then correctly classified using these derived variates. Individ- ual habitat assignments in both the learning run and the test run closely resembled the habitat assignment per- centages using all specimens (see Table 6). Consistency of results among all three classification runs indicated that the derived canonical variates summarized form variabil- ity among the sampled habitats well. Table 5 Loadings and standardized coefficients by variable and canonical variate. Variate 2 Variate | Variable Loading Coefficient Loading Coefficient KA —0.81 —0.90 0.38 0.29 ApWd 0.31 0.63 0.02 0.07 ShHt 0.19 0.64 = O32 IES) SpWd 0.04 —0.98 = OFT GS AA —0.03 0.24 0.69 1.13 J. W. Tupen, 1999 Page 255 Table 6 Habitat assignment of Alia carinata using jackknifed canonical variate analysis. Habitat abbreviations follow Table 2. Predicted habitat Actual habitat B G B 18 (60%) 1 (3.3%) G 1 (3.3%) 27 (90%) M 7 (23.3%) 3 (10%) UL, 5 (16.7%) 0 Total 31 31 Inter- and Intrahabitat Sexual Dimorphism All variables except AA were significantly larger in males (Table 7). After removing or reducing the effects of size using ANCOVA, no significant differences existed between sexes for every variable tested (Table 8), indi- cating that sexual dimorphism was a result of size differ- ences. The variable ShWd was not analyzed with AN- COVA because it did not meet the assumption of be- tween-group homogeneity of slopes, and the variable KA was not analyzed because it did not display a significant linear relationship with the covariate. Significant differences existed between sex forms with all habitats pooled (F = 2.66, P = 0.01, df = 8 and 111). Stepwise CVA showed that the variable ShWd alone pro- vided optimal information in discriminating among sexes. That is, after its entry into the CVA, no other variables contributed significant additional discrimination between sexes. This result indicated that the variables used in this sex analysis were highly intercorrelated and contained the same discriminating information. Although sex group centroids (based only on ShWd) were significantly dif- ferent, jackknifed classification success was relatively Table 7 ANOVA results for Alia carinata by sex, n = 60 for each sex. Means and standard errors for AA and KA in de- grees, all others in mm. Variable means significantly dif- ferent (* P < 0.05, ** P < 0.001), not significantly dif- ferent (ns), between sexes. Males Females ANOVA Variable Mean SE Mean SE F (1,118) ShHt 7.66 0.09 W225 0.06 13.43** ShWd 3.74 0.03 3.56 0.03 16.14** SpHt 2.65 0.04 2.46 0.03 14.13** SpWd 2.34 0.02 2.24 0.02 IAS ApHt 3.19 0.04 3.07 0.03 6.49* ApWd LSS) 0.02 1.48 0.01 8.08* AA 41.6 0.3 42.3 0.2 0.65 ns KA 41.3 1.1 40.1 ilail 4.25* M Z Total n 7 (23.3%) 4 (13.3%) 30 1 (3.3%) 1 (3.3%) 30 13 (43.3%) 7 (23.3%) 30 4 (13.3%) 21 (70%) 30 25 33 79 (65.8%) low, with males and females correctly classified 63% and 72% of the time, respectively. Within-habitat analysis helped to explain the low ca- nonical correlation coefficient (R = 0.35) for this pooled analysis. No multivariate form differences were detected between sexes within the habitats MP (F = 0.69, P = 0.70, df = 8 and 21) or ZM (F = 0.55, P = 0.81), while significant form differences did exist within the RB (F = 2.43, P = 0.049) and GS habitats (F = 12.22, P < 0.001). The variable ShWd provided optimal information in sep- arating RB habitat population sex centroids, and the de- rived variate (R = 0.60) correctly classified males and females 80% and 73% of the time, respectively. The var- iable SpWd provided optimal information to separate GS forms of A. carinata population sex centroids, and the derived variate (R = 0.71) correctly classified males and females 80% and 93% of the time, respectively. In sum- mary, CVA showed that multivariate models were no more useful than univariate analysis of standardized scores (z-scores) in discriminating between sexes. The sex ratio of A. carinata approximated unity (77 = 0.25, P = 0.62, df = 1) and did not differ among the four habitats (? = 0.57, P = 0.90, df = 3). Table 8 ANCOVA results for Alia carinata by sex, n = 60 for each sex. Adjusted (for covariate ShHt) means and stan- dard errors for AA in degrees, all others in mm. Variable means not significantly different (ns) between sexes. ShWd and KA not analyzed with ANCOVA, see text. Males Females ANCOVA Variable Mean SE Mean SE F (1,117) SpHt ASST 0.02 2.54 0.02 1.24 ns SpWd 2.29 0.01 2.28 0.01 0.50 ns ApHt 3.11 0.02 3.13 0.02 1.72 ns ApWd 1.52 0.01 1.51 0.01 0.20 ns AA 41.9 0.2 42.0 0.2 0.17 ns Page 256 Interhabitat Shell Patterning The distribution of solid and variegated A. carinata in- dividuals within habitats was significantly different among habitat groups (y7 = 12.80, P < 0.01, df = 3). Sixty-eight percent of A. carinata from ZM were solid and dark brown to almost black. In contrast, 60-77% of individuals were variegated in other habitats. Solid spec- imens from non-ZM habitats tended to be lighter in color, ranging from light to dark brown, to bright orange. No significant difference was detected in the pattern frequen- cies among the RB, GS, and MP groups (7 = 1.93, P = 0.38, df = 2). Observed variegations included dark brown to whitish triangular polygons, irregular dots, spiral lines, and spiral and axial tessellations distributed over all or portions of the shell. Many of the specimens from the MP habitat were characterized by a light-colored band on the shoulder carina, as were several of the RB specimens. This pattern was not observed in A. carinata from either GS or ZM habitats. Variegations were more pronounced in A. carinata from GS compared to other habitats. There was no relationship between pattern presence and sex (y” = 1.69, P = 0.19, df = 1). DISCUSSION Alia carinata shell forms were distinct and quantitatively separable in three of four sampled habitats, and observed color variants generally corroborated the observations of Carter & Behrens (1980). Specimens from GS were gen- erally tall and narrow, and often lacked the characteristic shoulder carina that has been used as a key taxonomic feature of this species (Carlton & Roth, 1975; McLean, 1978; Abbott & Haderlie, 1980). In addition, GS forms possessed large apertures and were generally light in col- or with pronounced shell pattern variegations. The ZM forms were short, broad, and stout, and were character- ized primarily by relatively large apical angles. More of- ten than not, they were dark and unpatterned, and dis- played very large shoulder carinas (keel angles). The RB forms were less distinct than the previous two habitat forms, but were characterized by moderate shoulder ca- rina development and small apertures. The RB and MP forms of A. carinata were statistically indistinct, though the latter often possessed characteristic, light-colored shoulder carinas and narrow spires. These results also corroborate those of Bergman et al. (1983), where ShWd: ShHt ratios of 0.526 and 0.469 from protected and ex- posed habitats, respectively, were reported in A. carinata from the Bodega Bay, California region. Comparable ra- tios and habitats from the present study are 0.512 and 0.475 from ZM (protected) and GS (exposed), respec- tively. Bergman et al. (1983) also reported that 9% and 81% of A. carinata from exposed and protected environ- ments, respectively, were strongly keeled (rating 3-5), comparing favorably with the keel rating results from GS (4%) and ZM (97%) habitats in the present study. The Veliger, Vol. 42, No. 3 In general, shell forms documented in this study agree well with causal relationships summarized by other re- search. Wave exposure is correlated with the presence of large apertures (Vermeij, 1978; Rugh, 1977) and narrow shell forms (Bergman et al., 1983; Trussell, 1997b), con- sistent with features of specimens from intertidal GS. Narrow forms may allow individuals to seek refuge in crevices during extreme wave exposure—crevices that may be unavailable to sharply keeled or broad-formed individuals (Trussell, 1997b). The RB and MP specimens of A. carinata from deeper and presumably cooler water had taller spires and smaller AA’s than those from most other habitats, which may indicate slower growth rates and consequent reduced whorl expansion rates (Frank, 1975; Phillips, 1981; Hughes, 1986). Small apertures in RB specimens, and stout shells and short spires in ZM forms of A. carinata may protect against the aperture peeling and spire crushing techniques often used by crabs when preying on gastropods (Vermeij, 1978). Crabs are known to exert considerable predation pres- sure on small gastropods (Vermeij, 1977, 1978; Trussell, 1996), with pressure generally highest in protected, lower energy areas (Trussell, 1996). Bergman et al. (1983) hy- pothesized that large shoulder keels and broad shells in A. carinata from wave-protected habitats (i.e., harbor) were adaptive responses to crab predation pressure. They substantiated this conclusion by noting that shell scars, indicative of failed crab predation attempts (Vermeij, 1978), were more frequent in the protected habitat rela- tive to an exposed habitat. While I did not collect quan- titative data on sympatric durophagous crab species— specifically rock crabs, Cancer spp., and hermit crabs, Paguridae (Spight, 1976; Vermeij, 1977; Gosselin & Chia, 1995)—abundance by habitat, there did appear to be relatively higher numbers of both crab groups in the RB and ZM habitats, and a noticeably greater number of scarred A. carinata in the RB habitat. In one recent study, however, snails only needed to be proximal to, not at- tacked by, crab predators to elicit adaptive morphologic responses (Trussell, 1996, in Littorina sp.). He speculated that chemical or visual cues by prey, or chemical cues by predatory crabs, may have engendered this response. Glo- bose forms of A. carinata from ZM may indicate an adap- tive response to the relatively large numbers of Cancer spp. in this habitat. Bergman et al. (1983) presented pre- liminary observations which indicated that small Cancer sp. were more successful at damaging and killing un- keeled, versus keeled, A. carinata. While breaking A. car- inata shells for sex determinations in this study, I noted that specimens from ZM were much more difficult to crush than those from other sampled habitats. A study of shell compression strengths and crab presence/absence, by habitat, would be a quantitative test of this observa- tion, although Vermeij (1978) noted that studies of this nature often produce highly variable results. Less easily explained using known causal relationships J. W. Tupen, 1999 is the large shell size in forms of A. carinata from GS. Trussel (1997a) showed that larger individuals (e.g., Lit- torina sp.) were often selected against by wave dislodge- ment, while Brown & Quinn (1988) speculated that for- aging efficiency may be reduced in larger individuals (e.g., Collisella spp. and Nucella sp.) trying to maintain their position in the high-energy intertidal zone, resulting in smaller overall sizes. I cannot explain the mechanisms directly responsible for the large shell size of open-coast forms of A. carinata from GS, but speculate that their large foot area, inferred from their large aperture area (Grahame & Mill, 1986; Trussell, 1997b), may increase substrate attachment in a comparatively food-rich envi- ronment (Leigh et al., 1987). It is interesting that the CVA was unable to separate MP from RB forms of A. carinata. Although RB speci- mens displayed unique characteristics relative to other habitats, a large degree of overlap with MP forms con- founded the separation. My recent field observations in- dicate that RB habitat A. carinata sub-adults and adults may migrate from this habitat to the MP canopy to de- posit egg masses, rather than settling on the latter as ju- veniles. This suspected migration appears to coincide with the spring regeneration and growth of MP following annual kelp removal by winter storms. The present mor- phometric analysis suggests that RB habitat forms of A. carinata remain unchanged significantly during their ten- ure on MP. Observed interhabitat sexual dimorphism disparities are also interesting. Purchon (1977) noted that sexual di- morphism in mollusks is rare. However, when present, it is usually size- and not shape-related (Webber, 1977), and is more common in the higher orders of the Gastropoda (Dobberteen & Ellmore, 1986). In cases where dimor- phism is present, and unlike A. carinata, females are gen- erally larger or more globose than males to facilitate the storage or brooding of eggs (e.g., Lindberg, 1985, Mar- garites sp. (Trochidae); Armengol, 1996, Potamolithus spp. (Hydrobiidae). I cannot explain the among-habitat differences in the presence or absence of conchological sexual dimorphism in sampled A. carinata. Sample sizes may have been inadequate to detect subtle differences in size between sexes in A. carinata from MP and ZM hab- itats: males from MP were taller than females (7.69 mm and 7.50 mm ShHt, respectively), though not significantly so, while males and females from ZM were similarly tall between sexes (7.06 mm and 7.15 mm ShHt, respective- ly). Having already speculated that MP specimens of A. carinata may have been RB migrants, I would expect similar sexual dimorphism in individuals from both of these habitats. Shell color and pattern variability in gastropods are of- ten linked to predation-related crypsis (Vermeij, 1978; Dytham et al., 1990, Littorina spp.; Cook & Bridle, 1995, Littoraria sp. Littorinidae; Gardner et al., 1995, Clithon sp. Neritidae;). Analysis of shell patterning in A. carinata Page 257 within and among habitats does not indicate a clear re- lationship with crypsis contributing to shell form vari- ability. Variegated patterning and light base coloration did allow GS specimens of A. carinata to blend well with variably colored algal thalli and entrapped shell frag- ments. However, if predation-induced crypsis was impor- tant, I would expect that individuals from MP, the only homogeneously colored habitat, would be non-variegated and light to dark brown; however, they were not. The distinctive, variegated color form of A. carinata from MP, consisting of a whitish band on the body whorl carina, is quite obvious when observed and is the model color form for the mimic amphipod Pleustes platypa (Crane, 1969). This color form appears to be largely specific to MP, as it was not evident in specimens from GS or ZM habitats. My infrequent observations of this color form on RB hab- itats were more often than not near Macrocystis beds, and these likely represented individuals displaced from the MP canopies. Alia carinata from ZM were also very con- spicuous in their habitat, being solid and dark on a light and variegated background. I have no direct explanation for this apparent lack of crypsis, as durophagous crabs were relatively abundant on the benthic substrate near the ZM habitat. Cook & Bridle (1995) speculated that the lack of crypsis in variably colored Littoraria sp. from canopies of mangroves and the presence of crypsis in the same species found on the trunk potentially resulted from a lack of crab predators and consequent lack of selection pressure in the former habitat. This may be true for A. carinata as well, as I have never observed Cancer spp. or large (> 1 cm) pagurids in either MP or ZM canopies. Shell form variability in mollusks can be genetically based (Crothers, 1984; Grahame & Mill, 1993; Gosselin & Chia, 1995; Wilbur & Gaffney, 1997), a consequence of natural selection and adaptation. Variability can also be due to phenotypic plasticity as a response to the phys- ical and biological environment (Palmer, 1990; Grahame et al., 1990; Chapman, 1994, 1995; Trussell, 1996). In those variable species that demonstrate direct develop- ment life-histories, both mechanisms may be important. However, in those taxa that disperse by settlement of planktonic larvae or juveniles, as do Alia carinata, ge- netic polymorphism is less likely. Rather, form variability should result from phenotypic plasticity or post-settle- ment selection. Bergman et al. (1983) suggested that be- tween-exposure form variability in A. carinata may have represented a polymorphism due to the limited ability of “crawl away larvae’’ to disperse among habitats. Recent literature indicating that A. carinata demonstrate plank- totrophic life histories (McLean, 1996) would suggest ecophenotypic adaptation, rather than genetic polymor- phism. My qualitative and successive observations of form variability following recruitment events, and the lack of discontinuous shell phenotypes (Ford, 1940) also argue against polymorphism. Post-settlement selection, however, remains a likely mechanism in observed inter- Page 258 The Veliger, Vol. 42, No. 3 habitat shell form differences—in particular, selection for hydrodynamic, narrow forms in the wave-swept intertidal (Trussel, 1997b), and selection for stout, highly keeled forms in areas populated with durophagous crab species (Trussel, 1996). Although this study strongly suggests the existence of habitat-specific forms in A. carinata, a sam- pling design that included replication within habitats among regions, through at least a complete year, would be necessary to account for microhabitat (Chapman, 1995), geographic (Chow, 1987), and temporal (Gardner et al., 1995) influences on shell form. This last factor may be particularly important when considering potential among-year variation in growth rates and shell morphol- ogy due to periodic oceanographic events (e.g., El Nino/ ENSO), not uncommon in California. ACKNOWLEDGMENTS I am grateful to all of the following individuals for taking an interest in, and helping to improve, the development and presentation of this research: E. Laman, J. Carter, D. Behrens, and D. Geiger reviewed an earlier version of this manuscript. J. McLean cataloged voucher specimens from this project at the LACM, and commented on a manuscript draft. Two anonymous reviewers and B. Roth edited a later manuscript draft, and identified experimen- tal and analytical approach considerations that I had over- looked. M. DeMaintenon helped with Alia carinata sex determinations, and M. Behrens assisted with specimen collections. C. Clark helped to put the pieces of this effort together. Lab space and processing equipment were pro- vided by Pacific Gas and Electric Company. Partial fund- ing for this research was provided by TENERA, Inc. LITERATURE CITED AssBoTtT, D. & E. HADERLIE. 1980. Prosobranchia: marine snails. Pp. 230-307 in R. Morris, D. Abbott & E. Haderlie (eds.), Intertidal Invertebrates of California. Stanford University Press: Stanford, California. ARMENGOL, M. 1996. Taxonomic revision of Potamolithus aga- petus Pilsbry, 1911, and Potamolithus buschii (Frauenfield, 1865) (Gastropoda: Hydrobiidae). Malacologia 38:1—17. BERGMAN, J., J. GELLER & V. CHow. 1983. Morphological diver- gence and predator-induced shell repair in Alia carinata (Gastropoda: Prosobranchia). The Veliger 26(2):116—-118. Brown, K. & J. QUINN. 1988. The effect of wave action on growth in three species of intertidal gastropods. Oecologia 75:420—425. CarLTon, J. & B. Rotu. 1975. Phylum Mollusca: shelled gastro- pods. Pp. 467-514 in R. Smith & J. T. Carlton (eds.), Lights Manual: Intertidal Invertebrates of the Central California Coast. University of California Press: Berkeley, California. Carter, J. & D. BEHRENS. 1980. Gastropod mimicry by another pleustid amphipod in central California. The Veliger 22: 376-377. CHAPMAN, M. 1994. Small-scale patterns of distribution and size- structure of the intertidal littorinid Littorina unifasciata (Gastropoda: Littorinidae) in New South Wales. Australian Journal of Marine and Freshwater Research 45:635—652. CHAPMAN, M. 1995. Spatial patterns of shell shape of three spe- cies of co-existing littorinid snails in New South Wales, Aus- tralia. Journal of Molluscan Studies 61:141—162. Cuow, V. 1987. Morphological classification of sibling species of Littorina (Gastropoda: Prosobranchia): Discretionary use of discriminant analysis. The Veliger 29:359—366. Cook, L. & J. BRIDLE. 1995. Colour polymorphism in the man- grove snail Littoraria intermedia in Sinai. Malacologia 36: 91-95. Coppois, G. & C. GLOWACKI. 1983. Bulimulid land snails from the Galapagos: 1. Factor analysis of Santa Cruz Island spe- cies. Malacologia 23:209-219. CosTANzA, M. & A. AFiFi. 1979. Comparison of stopping rules in forward stepwise discriminant analysis. Journal of the American Statistical Association 74:777—785. CRANE, J., JR. 1969. Mimicry of the gastropod Mitrella carinata by the amphipod Pleustes platypa. The Veliger 12:200. CROTHERS, J. 1984. Some observations on shell shape variation in Pacific Nucella. Biological Journal of the Linnean Society 21:259-281. DILLON, R., JR. 1984. What shall I measure on my snails? Al- lozyme data and multivariate analysis used to reduce the non-genetic component of morphological variance in Gon- iobasis proxima. Malacologia 25:503-511. DOBBERTEEN, R. & G. ELLMORE. 1986. Embryonic expression of shell dimorphism in Margarites vorticifera (Gastropoda: Trochidae). Malacological Review 19:45—52. DyTHAM, C., J. GRAHAME & P. MILL. 1990. Distribution, abun- dance and shell morphology of Littorina saxatilis (Olivi) and Littorina arcana Hannaford Ellis at Robin Hood’s Bay, North Yorkshire. Pp. 233-240 in K. Johannesson, D. Raf- faelli, & C. Hannaford Ellis (eds.), Progress in Littorinid and Muricid Biology, Hydrobiologia 193. ETTER, R. 1988. Physiological stress and colour polymorphism in the intertidal snail Nucella lapillus. Evolution 42:660— 680. Forp, E. 1940. Polymorphism and taxonomy. Pp. 493-513 in J. Huxley (ed.), The New Systematics. Clarendon Press: Ox- ford. FRANK, P. 1975. Latitudinal variation in the life history features of the black turban snail Tegula funebralis (Prosobranchia: Trochidae). Marine Biology 31:181—192. GARDNER, G., P. MATHER, I. WILLIAMSON & J. HUGHES. 1995. The relationship between shell-pattern frequency and micro- habitat variation in the intertidal prosobranch, Clithon oual- aniensis (Lesson). Malacologia 36:97—109. GossELIN, L. & EF CutA. 1995. Characterizing temperate rocky shores from the perspective of an early juvenile snail: the main threats to survival of newly hatched Nucella emargin- ata. Marine Biology 122:625—635. GRAHAME, J. & P. MILL. 1986. Relative size of the foot of two species of Littorina on a rock shore in Wales. Journal of Zoology 208:229—226. GRAHAME, J. & P. MILL. 1993. Shell shape variation in rough periwinkles: genotypic and phenotypic effects. Pp. 25—30 in J. Aldrich (ed.), Quantified Phenotypic Responses in Mor- phology and Physiology. Proceedings of the Twenty-Seventh European Marine Biology Symposium, Dublin, Ireland, Sep- tember 1992. GRAHAME, J., P Mitt & A. BRown. 1990. Adaptive and non- adaptive variation in two species of rough periwinkle (Lit- torina) on British shores. Hydrobiologia 193:223—231. HuGues, R. 1986. A Functional Biology of Marine Gastropods. J. W. Tupen, 1999 The John Hopkins University Press: Baltimore, Maryland. 245 pp. HuGues, J. & P. MATHER. 1986. Evidence for predation as a fac- tor in determining shell colour frequencies in a mangrove snail Littorina sp. (Prosobranchia: Littorinidae). Evolution 40:68—77. Jones, L. 1971. Studies on selected small herbivorous inverte- brates inhabiting Macrocystis canopies and holdfasts in southern California kelp beds. Pp. 343-367 in W. North (ed.), The Biology of Giant Kelp Beds (Macrocystis) in Cal- ifornia. Beihefte Nr. 32 zur Nova Hedwigia. Katou, M. & D. FoLtz. 1994. Genetic subdivision and morpho- logical variation in a freshwater snail species complex for- merly referred to as Viviparus georgianus (Lea). Biological Journal of the Linnean Society 53:73—90. KLecKA, W. 1980. Discriminant analysis. Sage University paper series On quantitative applications in the social sciences, 07— 019. Sage Publications: Beverly Hills and London. LANGAN-CRANFORD, K. & J. PEARSE. 1995. Breeding experiments confirm species status of two morphologically similar gas- tropods (Lacuna spp.) in central California. Journal of Ex- perimental Marine Biology and Ecology 186:17-31. LEIGH, E. Jr., R. PAINE, J. QUINN & T. SUCHANEK. 1987. Wave action and intertidal productivity. Proceedings of the Na- tional Academy of Sciences (USA) 84:1314—1318. LINDBERG, D. 1985. Shell sexual dimorphism of Margarites vor- ticifera: multivariant analysis and taxonomic implications. Malacological Review 18:1—-8. MCLEAN, J. 1978. Marine Shells of Southern California. Natural History Museum of Los Angeles County, Science Series 24, revised edition. 104 pp. MCLEAN, J. 1996. Gastropoda. Section 1, The Prosobranchia. Pp. 1-160 in P. Scott, J. Blake & A. Lissner (eds.) Taxonomic Atlas of the Benthic Fauna of the Santa Maria Basin and Western Santa Barbara Channel, Vol. 9. Mollusca. Santa Barbara Museum of Natural History: Santa Barbara, Cali- fornia. Nayor, R. & M. BEGON. 1982. Variation within and between populations of Littorina nigrolineata Gray on Holy Island, Anglesey. Journal of Conchology 31:17—30. Patmer, A. 1990. Effect of crab effluent and scent of damaged conspecifics on feeding, growth and shell morphology of the Atlantic dogwhelk Nucella lapillus. Hydrobiologia 193:155— 182. Page 259 PHILLips, D. 1981. Life-history features of the marine intertidal limpet Notoacmaea scutum in central California. Marine Bi- ology 64:95—103. PIMENTEL, R. 1979. Morphometrics: The Multivariate Analysis of Biological Data. Kendall/Hunt Publishing Company: Du- buque, Iowa. 276 pp. PURCHON, R. 1977. The Biology of the Mollusca. 2nd ed. Per- gamon Press: New York. 560 pp. RuGu, N. 1977. Differences in shell morphology between the sibling species Littorina scutulata and Littorina plena (Gas- tropoda: Prosobranchia). The Veliger 40:350—357. SPIGHT, T. 1976. Ecology of hatching size for marine snails. Oec- ologia 24:183—294. SystaT. 1996. SystaT 6.0.1 for Windows: Statistics. SPSS, Inc.: Chicago, Illinois 751 pp. TABACHNICK, B. & L. FIDELL. 1996. Using Multivariate Statistics. 3rd ed. Harper Collins College Publishers: New York. 880 PP- Tissot, B. 1984. Multivariate analysis of geographic variation in Cypraea caputserpentis (Gastropoda: Cypraeidae). The Ve- liger 27:106-119. Tissot, B. 1990. Geographic variation and mass mortality in the black abalone: the roles of development and ecology. Ph.D. Dissertation, Oregon State University, 271 pp. TRUSSELL, G. 1996. Phenotypic plasticity in an intertidal snail: the role of a common crab predator. Evolution 50:448—454. TRUSSELL, G. 1997a. Phenotypic plasticity in the foot size of an intertidal snail. Ecology 78:1033—1048. TRUSSELL, G. 1997b. Phenotypic selection in an intertidal snail: effects of a catastrophic storm. Marine Ecology Progress Se- ries 151:73-79. VERMEI, G. 1977. Patterns in crab claw size: the geography of crushing. Systematic Zoologist 26:138—151. VERMEU, G. 1978. Biogeography and Adaptation, Patterns of Ma- rine Life. Harvard University Press: Cambridge, Massachu- setts 332 pp. WEBBER, H. 1977. Gastropoda: Prosobranchia. Pp. 1-97 in A. Giese & J. Pearse (eds.), Reproduction of Marine Inverte- brates. Volume 4, Molluscs: Gastropods and Cephalopods. Academic Press: New York. WILBUR, A. & P. GAFFNEY. 1997. A genetic basis for geographic variation in shell morphology in the bay scallop, Argopecten irradians. Marine Biology 128:97—105. ZAR, J. 1996. Biostatistical Analysis. 3rd ed. Prentice-Hall: New Jersey. 662 pp. THE VELIGER © CMS, Inc., 1999 The Veliger 42(3):260—266 (July 1, 1999) Remains of the Prey—Recognizing the Midden Piles of Octopus dofleini (Wilker) R. DODGE aAnpb D. SCHEEL Prince William Sound Science Center, Box 705, Cordova, Alaska 99574, USA Abstract. We described the contents and the field signs of 52 midden piles found outside occupied dens of Octopus dofleini (Wilker, 1910) in Prince William Sound and Cook Inlet, Alaska. The contents of midden piles are important data for describing octopus diets; yet the field signs for distinguishing octopus midden piles from remains left by other processes can be subtle. Remains of four crab species, Telmessus cheiragonus (Telesius, 1815), Cancer oregonensis (Dana, 1852), Pugettia gracilis Dana, 1851, and Lophopanopeus bellus (Stimpson, 1860), composed 74% of the prey individuals represented in intertidal middens in Prince William Sound. However, the same species were not typical of other locations: Chlamys hastata (Sowerby, 1843) and C. rubida (Hinds, 1845) were the most common species repre- sented in subtidal middens, while the crab P. gracilis Dana, 1851, and the mussel Mytilus trossulus Gould, 1850, were among the most common in intertidal middens found in Cook Inlet. Drills were found on the hard remains of six species of crabs (75% of eight Crustacea species, 27% of 22 total species). Fifty-six percent of drill marks on crab species were located toward the carapace posterior. Of the crab species sampled in Prince William Sound that were drilled at all, C. oregonensis was the species most often drilled (36%), whereas T. cheiragonus was drilled least often (6%). Drills of O. dofleini on crabs were oblong (2—6 X 1—2.5 mm), and came to a point at one or both ends. Drill marks tapered toward the inside of the shell, and when the final perforation of the inner surface was made, the drill was no more than a pinpoint. A previously undescribed mark in prey remains, the bite mark, occurred on the leg of T. cheiragonus. Bites on weathered prey remains were about 1.2 cm long X 0.5 cm wide, occurring on the inside and outside of the leg. INTRODUCTION > Animal “‘signs,’’ including tracks, scat, wallows, nests or dens, and bones, feathers, or other remains of prey, have been important data in studies of many different species, and have been used to indicate presence or absence, to estimate population size, breeding activity, and foraging ranges, and to examine diet. A great many marine organ- isms also leave “‘sign,”’ particularly the remains of hard- shelled prey. For example, in the northeastern Pacific, sea otters feed on clams and discard the shells (Kvitek et al., 1992). Pycnopodia helianthoides (Brandt, 1835), the sun- flower star, also hunts clams, sometimes in the holes dug by sea otters. Sea stars push the sediment out of the way leaving behind a berm of sand or gravel and the empty shells of their bivalve prey (Kvitek et al., 1992). Octo- puses feed on many different bivalves and clams, includ- ing Saxidomus giganteus Deshayes, 1839, which is also a prey of otters (Riedman & Estes, 1988), and discard the remains in midden piles (Hartwick et al., 1981; Hart- wick & Thorarinsson, 1978; Mather, 1991, 1994; Mather & O’Dor, 1991). Distinguishing whether a clam has been opened by a sea otter, a sea star, or an octopus is useful for determining impacts of predators on invertebrate com- munities, requires specialized knowledge, and can limit research in some studies (e.g., Fotheringham, 1974; Kvi- tek et al., 1992). Determining how animal remains arrived at their pres- ent location on a beach or on the sea floor requires paying attention to sometimes subtle clues (e.g., Fotheringham, 1974). Even so, identifying the source of animal remains can be an important research tool. Descriptions of prey middens have been a primary method of describing oc- topus diets; for example, in a sample of 12 papers inves- tigating the diet of octopuses (Table 1) midden analysis was used in eight. Octopuses are often mobile and leave midden piles behind, so that the absence of an octopus at a midden pile does not necessarily imply that the remains were left by some other predator. Despite this, few de- scriptions of the “‘signs’’ left by octopuses have been published (but see Hartwick et al., 1978; Ambrose, 1983) to assist the beginning octopus researcher or to be used by someone working on species other than octopuses, but interested in attributing marine invertebrate remains to the animals that killed them. In this paper, we report field signs indicative of the presence of Octopus dofleini on beaches in Prince William Sound and Cook Inlet, Alaska, and describe the methods by which middens left by O. dofleini may be recognized. From captive studies, octopuses, including O. dofleini, are known to use three different techniques to gain entry to hard-shelled prey: they may pull it apart, ‘‘drill” through the shell (Nixon, 1979; Hartwick, 1981), or bite it open (Anderson, 1994). The latter two methods leave R. Dodge & D. Scheel, 1999 Page 261 Table 1 Methods used to determine diet in a sample of octopus studies. Method of Species studied determining diet Octopus dofleini stomach contents Japan, review article Location Source Mottet, 1975 O. dofleini midden counts British Columbia Hartwick et al., 1981 O. dofleini midden counts British Columbia Mather et al., 1985 O. dofleini midden counts British Columbia Cosgrove, 1987 O. dofleini observation Washington, captive study Anderson, 1991 O. dofleini midden counts Washington Anderson, 1994 O. dofleini midden counts Alaska Vincent et al., 1998 O. rubescens midden counts California Laidig et al., 1995 O. vulgaris midden counts South Africa Smale & Buchan, 1981 O. vulgaris midden counts Bermuda Mather & O’Dor, 1991 O. vulgaris midden counts Bermuda Mather, 1991 O. vulgaris stomach contents Spanish Mediterranean Sanchez & Obarti, 1993 marks on the prey that may be used to identify remains left by octopuses. The drill mark has frequently been mentioned in the literature (e.g., for O. dofleini, Hartwick et al., 1978, 1981; Hartwick, 1983), but described for O. dofleini only on Saxidomus giganteus, a bivalve (Am- brose et al., 1988). In this paper, we provide the first published descriptions of O. dofleini drills on crab spe- cies. METHODS Three study sites, Port Graham in Cook Inlet (59°21'N, 151°49’W) and Green (60°14'N, 147°14’W) and Monta- gue (60°16’N, 147°26’W) Islands in Prince William Sound were surveyed for Octopus dofleini. The surveys consist- ed of intertidal beach walks and SCUBA dives, to depths of 33 m below mean lower low water (MLLW). Dens were identified by the presence of an octopus. If den litter was present at an occupied den, all bits were collected for later measurement and identification (following Foster 1991 for bivalves and Kozloff 1987 for all other taxa). Remains were judged to be either fresh (without algae growth on inner surfaces and unweathered) or old (either with algal growth or weathered). Old remains were not counted and measured as part of the octopus midden, as these remains may have been buried in the sediment until excavated by the octopus when making its den. All spec- imens collected were inspected for octopus drills or other signs of handling by octopus, and the locations of such marks were noted. Measurements of the lengths and widths were taken of bivalves, crab carapaces, and crab legs. RESULTS Prey remains were collected from 52 intertidal and sub- tidal sites located outside occupied dens and attributed to Octopus dofleini on the basis of the octopus in attendance. Middens characteristically contained crab remains: for example, of the 42 middens found at intertidal dens in Prince William Sound, 40 contained at least one item from a crab. The most common species in these intertidal middens were Telmessus cheiragonus, Cancer oregonen- sis, Pugettia gracilis, and Lophopanopeus bellus, which together composed 74% of the sample of prey (estimated minimum individuals represented by midden items, Table 2). However, middens in other locations often had a dif- ferent composition. Subtidal middens, collected from the same areas in the Sound but from depths between —5 and —33 m MLLW, were dominated by remains of scallops and other bivalves, rather than of crabs (Table 3); while four intertidal middens from Port Graham (in Cook Inlet, Alaska) contained a mixture of crabs and bivalves, but the most abundant species there were not the same as in the Sound (Table 4). Inspection of midden remains sometimes revealed marks left by the octopus: we found both ‘‘drill’’ and “bite’’ marks (Figure 1). The drill mark of O. dofleini on carapaces of Cancer oregonensis and Lophopanopeus bellus (Figure la) was oblong, about 1.5—3.0 millimeters long and 0.25—2.0 millimeters wide, and usually came to a point at one or both ends. The drill gradually tapered from the outside of the shell toward the inside. The final perforation of the inner surface may be no more than a pinpoint. The only other drill marks that we found on shells in this area (not in octopus middens) were moon snail drill marks that were larger and almost perfectly round, and were readily distinguishable by size and shape from the marks of Octopus dofleini. Drills were found on the carapace or chelipeds of six species of prey (27% of 22 species found in middens, Tables 2-5). Drill marks were more frequently encountered on some species than others. Of the crab remains found in inter- tidal middens in the Sound (Table 2), the carapaces of Cancer oregonensis and of Pugettia gracilis were most often drilled, while those of Telmessus cheiragonus were Page 262 The Veliger, Vol. 42, No. 3 Table 2 Prey remains found in 42 intertidal midden piles left by Octopus dofleini on Green and Montague Islands, Prince William Sound, AK. Count found! No. drilled Avg. size? Taxon Species (prop. of sample in parentheses) (sample size) Decapoda Telmessus cheiragonus (Telesius, 1815) 70 (0.30) 4 (0.06) 3.55 (70) Decapoda Cancer oregonensis (Dana, 1852) 58 (0.25) 21 (0.36) 1.99 (55) Decapoda Pugettia gracilis Dana, 1851 23 (0.10) 6 (0.26) 2.77 (21) Decapoda Lophopanopeus bellus (Stimpson, 1860) 20 (0.09) 2 (0.10) 1.57 (18) Bivalvia Macoma inquinata (Deshayes, 1855) 17 (0.07) 0) 2.75 (15) Bivalvia Protothaca staminea (Conrad, 1837) 17 (0.07) 0 2.16 (16) Bivalvia Saxidomus giganteus Deshayes, 1839 8 (0.03) 0) 2.63 (8) Gastropoda Littorina sitkana Philippi, 1845 4 (0.02) 0) Bivalvia Chlamys hastata (Sowerby, 1843) 3 (0.013) (0) 3.83 (3) Bivalvia Pododesmus macroschisma (Deshayes, 1839) 3 (0.01) (0) 6.40 (3) Polyplacophora Tonicella lineata (Wood, 1815) 2 (0.01) 0) 1.20 (1) Bivalvia Mytilus trossulus Gould, 1850 2 (0.01) (0) 1.95 (2) Decapoda Cancer productus Randall, 1839 2 (0.01) 0) 6.00 (1) Decapoda Cryptolithodes sitchensis Brandt, 1853 2 (0.01) 0) Bivalvia Chlamys rubida (Hinds, 1845) 1 (0.004) 0) Decapoda Hapalogaster mertensii Brandt, 1850 1 (0.004) (0) Decapoda Phyllolithodes papillosus Brandt, 1849 1 (0.004) 0) 5.0 (1) Gastropoda Trichotropis cancellata Hinds, 1849 1 (0.004) 0) 10.5 (1) ' For crabs, only the number of carapaces is given, indicating the minimum number of individuals represented in the litter; for bivalves, each count indicates either the left valve, right valve, or both when still attached. ? The mean of the carapace length (cm) for crabs and of valve length for bivalves. Sample size is indicated in parentheses (in some cases, remains were broken and could not be measured, although the fragments could be identified to species). only infrequently drilled. In contrast to crabs, most bi- valves were not drilled (Tables 2—4). Chlamys sp. valves found at both intertidal and subtidal dens were never drilled (n = 19), nor were any other bivalves collected in the Sound (Tables 2 & 3). However, drills were recorded on four bivalves from middens in front of unoccupied dens in Port Graham (Figure 1b). We also examined the locations of drill marks on crab remains, using a sample of 50 drilled carapaces and 21 drilled chelipeds (of all species) collected from midden piles at both occupied and unoccupied dens in Prince Wil- liam Sound (including remains in Tables 2, 3, 5, and ad- ditional remains not listed). Each item was drilled only once. On carapaces, 38% of the drills (n = 19 of 50) were placed over the posterior midline (Figure 2a). Ad- ditional drills were placed immediately to the right (n = 6) or to the left (n = 3) of the posterior midline, so that 56% (n = 28) of the drills were located in the posterior medial portion of the carapaces. The next most common position drilled was over the medial central portion of carapaces (n = 18 or 36%), including drills on the central midline (n = 9) and immediately to the right (n = 4) or left (n = 5) of midline. The remaining 8% of the drills (n = 4) were located in the left posterior section of the Table 3 Prey remains found in six subtidal midden piles left by Octopus dofleini near Green and Montague Islands, Prince William Sound, AK. Taxon Species Bivalvia Chlamys hastata (Sowerby, 1843) Bivalvia Chlamys rubida (Hinds, 1845) Bivalvia Macoma inquinata (Deshayes, 1855) Decapoda Cancer oregonensis (Dana, 1852) Decapoda Pugettia gracilis Dana, 1851 Decapoda Lophopanopeus bellus (Stimpson, 1860) Gastropoda Acmaea sp. Bivalvia Pododesmus macroschisma (Deshayes, 1839) ' Details as in Table 2. No. found! No. drilled Avg. size! 9 (0.24) 0 2.46 (9) 6 (0.16) 0) 2.23 (6) 5 (0.14) 0) 3.58 (5) 5 (0.14) 4 (0.80) 1.68 (5) 4 (0.11) 0) 2.70 (3) 4 (0.11) 0) 1.68 (4) 3 (0.08) 0) 1.73 (3) 1 (0.03) 0 2.60 (1) R. Dodge & D. Scheel, 1999 Page 263 Table 4 Prey remains found in four intertidal midden piles left by Octopus dofleini Port Graham, Cook Inlet, AK. Taxon Species Decapoda Pugettia gracilis Dana, 1851 Bivalvia Mytilus trossulus Gould, 1850 Decapoda Cancer oregonensis (Dana, 1852) Bivalvia Protothaca staminea (Conrad, 1857) Bivalvia Macoma nasuta (Conrad, 1837) Gastropoda Nucella emarginata (Deshayes, 1839) Gastropoda Nucella lima (Gmelin, 1791) Bivalvia Saxidomus giganteus Deshayes, 1839 No. found! No. drilled Avg. size! 7 (0.35) 0) 2.63 (7) 4 (0.20) 0) 2.98 (4) 3 (0.15) 2 (0.67) 2B) 2 (0.10) 0) 1.95 (2) 1 (0.05) 0) 1.80 (1) 1 (0.05) 0) 2.70 (1) 1 (0.05) 0) 2.60 (1) 1 (0.05) 0) 2.60 (1) 13 Details as in Table 2. carapaces. Chelipeds were drilled on both inner (67%) and outer surfaces (33%) (Figure 2b, c), most commonly over the cheliped midline, but occasionally dorsal or ven- tral of the midline. We observed only one drill that was beyond the cheliped joint, but still on the leg of the crab. The Port Graham bivalves (n = 4) were drilled toward the posterior of the umbo (Figure 1b). A previously undescribed mark, which we term the “bite mark’’ (Figure 1c), was found exclusively on the penultimate segment of the Ist (cheliped) leg of Telmes- sus cheiragonus. As implied by the name, we believe the bite mark is made when the octopus uses its beak to bite through the leg of this prey species. Bite marks were oval in shape and greatly varied in size depending on the amount of weathering of the remains. In the freshest spec- imens, bite marks were about 1.2 cm long X 0.5 cm wide, located on the inside of the leg, and extending the entire length of the segment. The bite mark is apparently placed at a weak point in the leg exoskeleton and has a tendency to expand as prey remains break down. Bite marks were not found on the legs of other prey species; however, only five legs from species other than T. cheiragonus were examined. Of the Ist legs examined (n = 112) from Prince William Sound 29% (32) had bite marks. No bite marks were recorded from Port Graham, although only four T. cheiragonus \st legs were found in that area. DISCUSSION We describe remains found in 52 middens found outside the dens of Octopus dofleini in order to assist beginning octopus researchers (or those expert on species other than octopuses) in recognizing an octopus midden. The com- position of the midden may be helpful, as most middens consisted primarily of crab and bivalve remains. Most middens in the intertidal in Prince William Sound con- tained one or more of four crab species (Table 2), so that this species composition was characteristic of middens. However, the characteristic species were different at dif- ferent depths and geographic locations (compare with Ta- bles 3, 4). The local distinctiveness of octopus midden piles has been noted in several papers detailing that oc- topuses in different areas eat different species or handle prey differently (References in Table 1), but may also result from factors influencing the residence time of prey remains in middens under different conditions (Ambrose, 1983; Mather, 1991). Closer inspection of prey remains from middens revealed that octopuses left marks on some of their prey during the process of killing and consuming them. However, some species of prey were more likely to be marked than others, and most items in middens were not marked by the octopus. In our sample, three species of crabs were likely to be drilled on the carapace (Tables 2—4) or a cheliped (Table 5), and a “‘bite’’ mark was often found on the leg of the crab Telmessus cheiragonus. However, most items of any species, and almost all bi- valve remains examined, bore no mark of being handled by O. dofleini. Octopuses typically drill many of their hard-shelled prey. The behavior has been reported for a number of octopus species (Octopus dofleini: Hartwick et al., 1978; Ambrose et al., 1988; O. bimaculatus Verrill and O. bi- maculoides Pickford & McConnaughey: Pilson & Taylor 1961; O. mimus Gould, 1852: Cortez et al., 1998; Eledone cirrhosa (Lamarck): Boyle & Knobloch, 1981; Grisely et al., 1996) and studied in detail in O. vulgaris Cuvier, 1797 (discussed in Nixon & Maconnachie, 1988). The average dimensions of holes drilled by O. dofleini in the butter clam were 1.03 mm on the outer surface of the shell and 0.56 mm on the inner surface of the shell (Ambrose et al., 1988), while drill holes made by O. vulgaris on My- tilus sp. were round with average dimensions of 1.1 mm on the outer surface and only 0.2 mm on the inner sur- face. In the latter species, drill holes were made primarily with the salivary papilla (Nixon, 1980), the secretions of which dissolve the shell (Nixon & Maconnachie, 1988). Multiple drill holes were common on an individual prey item of O. vulgaris (Nixon, 1979). On bivalves, O. vul- garis usually placed holes over the adductor muscle (Nix- on & Maconnachie, 1988); and salivary toxins were se- creted through the hole to paralyze, kill, or partially digest the prey. Some octopuses drill crustaceans as well as bivalves Page 264 The Veliger, Vol. 42, No. 3 aust TTT onaa Lion 2 3 4 Figure 1 Illustrations of the marks found in the prey remains of Octopus dofleini: typical drill mark on (a) the carapace of the Oregon Cancer crab, Cancer oregonensis, and (b) the umbo of the bi- valve, Protothaca staminea (the circle shows the view through a microscope; the scale bar shows one millimeter); and (c) typical bite mark on the leg of the helmet crab, Telmessus cheiragonus. At the time of publication, images of these and other prey re- mains from octopus middens are available at www.pwssc. gen.ak.us/~dls/octopus/specimens/ on the world-wide web. (Octopus vulgaris, Guerra & Nixon, 1987; Eledone cir- rhosa, Boyle & Knobloch, 1981; Grisley et al., 1996). Although drilling as a predatory behavior against mol- lusks has been described in detail (e.g., Pilson & Taylor, 1961; Nixon, 1979; Ambrose et al., 1988; Nixon & Ma- connachie, 1988), work with crustaceans has focused on the toxicity of saliva to this taxa (e.g., Ghiretti, 1959, 1960; Pilson & Taylor, 1961, but for exceptions see Guer- Table 5 Crab chelipeds found in 42 intertidal midden piles left by Octopus dofleini on Green and Montague Islands, Prince William Sound, AK. Count No. Species found! drilled Avg. size? Cancer oregonensis (Dana, 1852) 115 (0.54) 4(0.03) 1.59 (114) Lophopanopeus bellus (Stimpson, 1860) 51 (0.24) 4 (0.08) 1.67 (51) Telmessus cheiragonus (Telesius, 1815) 35 (0.16) 4(0.11) 2.83 (32) Pugettia gracilis Dana, 1851 9 (0.04) 2 (0.22) 1.67 Q) Hapalogaster mertensii Brandt, 1850 3 (0.01) 2 (0.67) 2.57 GB) Cancer productus Randall, 1839 1(0.005) 1(1.00) 4.00 (1) ' The number of crab chelipeds found. This does not indicate the minimum number of prey individuals represented in the litter as both left and right chelipeds were usually found in the same midden. Details as in Table 2. ? Size is the mean of cheliped lengths (cm). Details as in Ta- ble 2. ra & Nixon, 1987; Boyle & Knobloch, 1981; Mather & Nixon, 1995; Grisley et al., 1996). Drilling of crustacean prey has been noted for O. dofleini (Hartwick et al., 1981) although not described in detail. We found that Cancer oregonensis, Pugettia gracilis, and Lophopanopeus bellus remains were often drilled (10-41% of carapaces, Tables 2-4), whereas bivalve prey and some crabs (e.g., Tel- messus cheiragonus) were most often not. We found only a single drill mark per prey item regardless of species, in Figure 2 Illustrations of the position and number of O. dofleini drills (a) on carapaces of the five most abundant crab species in middens (Table 2), (b) on the outer surface, and (c) inner surface of che- lipeds of the same species, plus Hapalogaster mertensii (species in Table 5). R. Dodge & D. Scheel, 1999 Page 265 contrast to Nixon’s (1979) study of Octopus vulgaris which found that some species were drilled only once while others were drilled multiple times. Whether this difference is characteristic of the difference in prey (mostly crabs in this study versus gastropods in Nixon, 1979) or of the octopus species remains to be determined. Drilling was not employed preferentially to open the larg- est prey: Telmessus cheiragonus were typically larger than the other crab species, yet were least frequently drilled. The exoskeleton of 7. cheiragonus is thinner and softer than that of the crabs that were more often drilled (personal observation), and this species probably is easier for the octopuses to pull apart or bite open, techniques that may be attempted in preference to drilling (Hartwick et al., 1981). Bite marks were previously undescribed and little is known about how octopuses utilize their beaks when cap- turing prey. Octopus dofleini are known to “‘chip the edg- es of the (bivalve) shell or break it with their beaks’’ (Anderson, 1994). We observed bite marks on the largest segment of the leg on a soft-shelled crab species Telmes- sus cheiragonus. The marks were 0.3 to 3.0 cm in length. We could find no data on the size of gape in an octopus. However, using Robinson & Hartwick (1983) we esti- mated the beak dimensions of a 2.6 kg octopus (the av- erage size of the octopuses that left the middens described in this study). The pigment upper-lateral-wall length (PULWL) was estimated to be 18.8 mm while the total upper-crest length was 30.3 mm. Assuming an octopus’ gape would be at most half the PULWL, O. dofleini may be able to make bites as large as 9.4 mm and therefore were probably capable of inflicting the marks we de- scribe. In addition, we fed a captive octopus a live 7. cheiragonus crab and collected the crab remains imme- diately after feeding. A fresh mark, similar to worn marks found on remains in middens, was found on the penulti- mate segment of one leg (Figure lc), confirming that Oc- topus dofleini does make this mark when feeding on T. cheiragonus. A captive octopus also left a bite mark on the remains of a Cancer productus crab, the first such mark we have seen on remains of any crab other than T. cheiragonus. Descriptions of prey middens have been a primary method of describing octopus diets (Table 1). Despite this reliance, published information on recognizing when re- mains are left by octopuses is scarce. The composition of midden piles varies from place to place and depends on local habitat as well as other factors (Hartwick et al., 1981; Ambrose, 1984; Vincent et al., 1998), which means that the species that are characteristic of octopus middens must be learned locally. It is not possible to ascribe any single prey remain to a particular cause or process when the remain was found in the absence of an obvious pred- ator and was not marked by the cause of mortality. How- ever, drill marks on one or more items within a midden provide a conclusive indication that the midden contents were left by a foraging octopus. This definite mark of handling left by octopuses allows the beginning research- er to attribute at least some middens to octopuses and learn what prey are characteristic in a locality. Using a suite of factors, including the association of unmarked items with octopuses in their dens or with items marked by drill holes or bites in the same midden, the researcher can become familiar with prey species that, while locally common in octopus middens, are handled in such a way as to leave no physical marks (e.g., Chlamys rubida and C. hastata in this study). While some ambiguity will al- ways remain when middens consist only of unmarked items, this local knowledge of how octopuses typically handle each prey species is the only way that the source of such remains can be inferred. ACKNOWLEDGMENTS This project would not have been possible without the knowledge, assistance, and goodwill of people in Che- nega Bay, Port Graham, Tatitlek, and Cordova. In partic- ular, Mike Eleshansky, Doug Bruck, Simeon Kvashni- koff, Jr., and Jerry Totemoff provided knowledge and as- sistance in the field; and Martha Vlasoff, Walter Meganik, Jr., Jody Seitz, and the fishermen of Cordova shared their knowledge and encouragement. We thank Tania L. S. Vincent, Scott Wilbur, Kathleen Pollet, Kathryn Hough, Roger Trani and Cordova Water Sports, Michael Kyte, Dan Logan, Neal Oppen, Beth Haley, and Tom Kline for assistance with fieldwork; Nora Foster for assistance with museum specimens; and a special thanks to Cap’n Neal- Dawg and Renee Ernster of the F/V Tempest for their support and interest in this project. Tania L. S. Vincent provided the drawings in Figures 1 and 2. We thank two anonymous reviewers for comments on the manuscript; and the Exxon Valdez Oil Spill Trustee Council and Pro- ject AWARE for financial support. LITERATURE CITED AMBROSE, R. E1983. Midden formation by octopuses: the role of biotic and abiotic factors. Marine Behaviour and Physi- ology 10:137-144. AMBROSE, R. F. 1984. Food preferences, prey availability, and the diet of Octopus bimaculatus Verrill. Journal of Experimental Marine Biology and Ecology 77:29—44. AMBROSE, R. F, B. J. LEIGHTON & E. B. HARTWICK. 1988. Char- acterization of boreholes by Octopus dofleini in the bivalve Saxidomus giganteus. Journal of Zoology, London 214:491— 503. ANDERSON, R. C. 1991. The fish-catching ability of Octopus do- fleini. Journal of Cephalopod Biology 2(1):75—76. ANDERSON, R. C. 1994. Octopus bites calm. The Festivus 26(5): 58-59. Boy eg, P. R. & D. KNoBLocu. 1981. Hole boring of crustacean prey by the octopus Eledone cirrhosa (Mollusca, Cephalop- oda). Journal of Zoology London 193:1-10. Cortez, T., B. G. Castro & A. GUERRA. 1998. Drilling behay- iour of Octopus mimus Gould. Journal of Experimental Ma- rine Biology and Ecology 224:193-—203. CosGROoVE, J. A. 1987. Aspects of the natural history of Octopus dofleini, the giant Pacific octopus. M. Sc. Thesis, University of Victoria. Foster, N. R. 1991. Intertidal Bivalves: A Guide to the Common Marine Bivalves of Alaska. University of Alaska Press: Fair- banks. 152 pp. FOTHERINGHAM, N. 1974. Trophic complexity in a littoral bould- erfield. Limnology and Oceanography 19(1):84—91. GHIRETTI, FE 1959. Cephalotoxin: the crab-paralysing agent of the posterior salivary glands of cephalopods. Nature 4669: 1 192— 1193. GHIRETTI, FE 1960. Toxicity of octopus saliva against Crustacea. Annals of the New York Academy of Sciences 90:726—741. GRISLEY, M. S., PB. R. BOYLE & L. N. Key. 1996. Eye puncture as a route of entry for saliva during predation on crabs by the octopus Eledone cirrhosa (Lamarck). Journal of Exper- imental Marine Biology and Ecology 202:225—237. GuERRA, A. & M. NIXon. 1987. Crab and mollusc shell drilling by Octopus vulgaris (Mollusca: Cephalopoda) in the Ria de Bigo (north-west Spain). Journal of Zoology London 211: 515-523. HarRTWICk, E. B. 1983. Octopus dofleini. Pp. 277-291 in P. R. Boyle (ed.), Cephalopod Life Cycles. Vol I. Academic Press: London. Hartwick, E. B. & G. THORARINSSON. 1978. Den associates of the giant Pacific octopus, Octopus dofleini (Wiilker). Ophelia 17(1):163—166. HARTWICK, E. B., G. THORARINSSON & L. TULLOCH. 1978. Meth- ods of attack by Octopus dofleini (Wtilker) on captured bi- valve and gastropod prey. Marine Behaviour and Physiology 5:193—200. HARTWICK, E. B., L. TULLOCH & S. MACDONALD. 1981. Feeding and growth of Octopus dofleini (Wiilker). The Veliger 24(2): 129-138. Koziorr, E. N. 1987. Marine Invertebrates of the Pacific North- west. University of Washington Press: Seattle. 511 pp. KviTEK, R. G., J. S. OLIVER, A. R. DEGRANGE & B.S. ANDERSON. 1992. Changes in Alaskan soft-bottom prey communities along a gradient in sea otter predation. Ecology 73(2):413— 428. Laipic, T. E., P. B. ADAMs, C. H. BAXTER & J. L. BUTLER. 1995. Feeding on Euphausiids by Octopus rubescens. California Fish and Game. Vol. 81:2. MATHER, J. A. 1991. Foraging, feeding and prey remains in mid- The Veliger, Vol. 42, No. 3 dens of juvenile Octopus vulgaris (Mollusca: Cephalopoda). Journal of Zoology, London 224:27-39. MatTuer, J. A. 1994. ‘Home’ choice and modification by juvenile Octopus vulgaris (Mollusca: Cephalopoda): specialized in- telligence and tool use? Journal of Zoology, London 233: 359-368. Martner, J. A. & M. Nrxon. 1995. Octopus vulgaris (Cephalop- oda) drills the chelae of crabs in Bermuda. Journal of Mol- luscan Studies 61:405—406. MatTuer, J. A. & R. K. O’Dor. 1991. Foraging strategies and predation risk shape the natural history of juvenile Octopus vulgaris. Bulletin of Marine Science 49(1—2):256—259. Matuer, J. A., S. RESLER & J. CosGROvVE. 1985. Activity and movement patterns of Octopus dofleini. Marine Behaviour and Physiology 11:301—314. MotrtetT, M. G. 1975. The fishery biology of Octopus dofleini (Wiilker). Technical Report No. 16, Management & Re- search Division, Washington Department of Fisheries. Nixon, M. 1979. Hole-boring in shells by Octopus vulgaris Cu- vier in the Mediterranean. Malacologia 18:431—443. Nrxon, M. 1980. The salivary papilla of Octopus as an accessory radula for drilling shells. Journal of Zoology, London 190: 53-57. Nixon, M. & E. MACONNACHIE. 1988. Drilling by Octopus vul- garis (Mollusca: Cephalopoda) in the Mediterranean. Jour- nal of Zoology, London 216:687-—716. Pitson, M. E. Q. & P. E. TAYLor. 1961. Hole drilling by Octopus. Science 134:1366-1368. RIEDMAN, M. I. & J. A. Estes. 1988. A review of the history, distribution, and foraging ecology of sea otters. Pp. 4—21 in G. R. VanBlaricom & J. A. Estes (eds.), Ecological Studies, 65: The Community Ecology of Sea Otters. Springer-Verlag: New York. RoBINSON, S. M. C. & E. B. HARTWICK. 1983. Relationship be- tween beak morphometrics and live wet weight of the giant Pacific octopus, Octopus dofleini martini (Wiilker). The Ve- liger 26(1):26-29. SANCHEZ, P. & R. OBARTI. 1993. The biology and fishery of Oc- topus vulgaris caught with clay pots on the Spanish Medi- terranean coast. Pp. 477—487 in T. Okutani, R. K. O’Dor & T. Kubodera (eds.), Recent Advances in Cephalopod Fish- eries Biology. Ist ed., Vol. I. Tokai University Press: Tokyo. SMALE, M. J. & P. R. BUCHAN. 1981. Biology of Octopus vulgaris off the east coast of South Africa. Marine Biology 65:1—12. VINCENT, T. L. S., D. SCHEEL & K. HouGH. 1998. Some aspects of diet and foraging behavior of Octopus dofleini in its northernmost range. Marine Ecology 19(1):13-29. The Veliger 42(3):267—274 (July 1, 1999) THE VELIGER © CMS, Inc., 1999 Morphometric Species Recognition in Brachidontes darwinianus and Brachidontes solisianus (Bivalvia: Mytilidae) MARCEL OKAMOTO TANAKA Programa de Pés-Gradua¢gao em Ecologia, IB, Universidade Estadual de Campinas, C.P. 6109, 13083—970 Campinas, SP, Brasil (e-mail: martan @ obelix.unicamp.br) AND CLAUDIA ALVES p—E MAGALHAES Departamento de Zoologia, IB, Universidade Estadual de Campinas, C.P. 6109, 13083—970 Campinas, SP, Brasil Abstract. Shell morphological characters of Brachidontes darwinianus and B. solisianus were investigated in mussels collected in six areas of sympatry in southeast Brazil. Individuals were measured and analyzed for the relation among shell length, height, and width, as well as other systematic characters used to distinguish the two species. All allometric relations varied according to the site sampled, except the height: length ratio, which consistently separated the two species regardless of where they occurred. Values for this ratio change in distinct shell length classes, and 99% confidence intervals are provided for each length class. Internal and external shell characteristics generally used to separate these species varied both during ontogeny and according to environment. Thus, caution is needed when identifying these species, as single characters may not be useful as taxonomic predictors. Combinations of characters are necessary to separate B. darwinianus and B. solisianus with more confidence. INTRODUCTION The mytilids are represented in Brazil by 12 genera. Among these, the genus Brachidontes Swainson, 1840, is the most diverse, with four species divided among three subgenera (Klappenbach, 1965). Mussels belonging to this genus have small lengths (average = 2 cm), sculp- tured with radial ribs, presenting crenulated margins and short ligaments (Rios, 1995). They are common inhabi- tants of the intertidal zone of rocky shores and man- groves, forming dominance belts in the ML WN level. Two species of Brachidontes are studied in this work, B. (Hormomya) darwinianus (d’Orbigny, 1846) and B. (Mytilaster) solisianus (d’Orbigny, 1846). Both species are adapted to stabilize themselves in sites subject to great wave drag forces, presenting a triangular shape, ventral flattening in the antero-posterior plane, and a gregarious way of life (Morton, 1992). B. darwinianus occurs from Rio de Janeiro to the northern limit of Patagonia (Avelar & Narchi, 1984a), being most common in estuaries and attached to intertidal rocks at river mouths (Nalesso et al., 1992). Its valves are fan-shaped, sculptured with rounded radial riblets, and with terminal umbones (Rios, 1995). The smaller B. solisianus is spread throughout the West- ern Atlantic coast from Mexico to Uruguay (Rios, 1995). Its shells are inequilateral, rectangular in their posterior edges, sculptured with very fine radial riblets restricted to their posterior ends, and with subterminal umbones (Avelar & Narchi, 1984b; Rios, 1995). Internally, the two species are distinguished by the position of the median byssal retractor (MBR) muscle scar relative to the liga- ment. In B. darwinianus this scar reaches or extends be- yond the ligament, while in B. solisianus it never reaches the ligament (Klappenbach, 1965). On the coast of Sao Paulo state mixed beds of these mussels are frequent. Because of environmental influences, morphological characters are of low confidence in separating mytilids, presenting high variances as a result of great shell plas- ticity (Seed, 1968). Variation of shell proportions due to density is also common, as demonstrated by Lent (1967) for Modiolus demissus and Brown et al. (1976) for other mytilid species. Nevertheless, it would be very useful to be able to identify species using shell characters. As McDonald et al. (1991) pointed out, to accomplish this goal, “it would be necessary to sample mussels from a wide variety of habitats to determine whether morpho- metric characters can reliably discriminate among spe- cies.” The primary characters used for species distinction within Brachidontes also vary according to environmen- tal conditions. Radial ribs, for example, are associated with shell periostracum, being easily lost mainly in adults from sites most exposed to wave action (Klappenbach, 1965). Variation in shell proportion is found in B. dar- winianus, presenting great phenotypic plasticity in rela- tion to the environment (Nalesso et al., 1992). Page 268 The Veliger, Vol. 42, No. 3 As part of a study of succession in intertidal rocky shores of Sao Paulo, we investigated morphometric char- acters useful for field identification of the two species of Brachidontes. We also tested the congruence of those characters with species diagnoses based on the internal anatomy proposed by Klappenbach (1965), Avelar & Narchi (1984a,b), and Rios (1995). MATERIALS anD METHODS Between January and April 1996 samples of sympatric B. darwinianus and B. solisianus were collected from five rocky shores in SAo Paulo State, southeastern Brazil: Mil- iondrios (23°58'S, 46°22'’W); Barequegaba (23°50’S, 45°26'W); Lagoinha (23°31'S, 45°11'W); Dura (23°30'S, 45°S10'W), and Lazaro (23°31'S, 45°08’W). Further sam- ples of B. solisianus were collected from four additional sites: Cigarras (23°44'S, 45°24'W, State of Sao Paulo); Rasa and Saquarema (22°44'S, 45°24'W and 22°55’S, 42°31'W, respectively, State of Rio de Janeiro), and Pina (8°S, 35°10'W, State of Pernambuco). For initial species separation of B. solisianus and B. darwinianus we relied upon defined characteristics of overall shape (fan or elongated) and sculpture patterns. Mussels were sampled selectively for the whole length range in their dominance zone along each site. Maximum length, width, and height of all animals were measured to the nearest 0.1 mm with vernier calipers. The angle formed by the ligament margin and the ventral axis was estimated to the nearest 5° using a graded angular scale. The valves were opened and the inner side of the right one was examined under a stereoscopic microscope. The distance separating the anterior edge of the median byssal retractor (MBR) scar to the posterior end of the ligament was measured to the nearest 0.001 mm (Figure 1). When the scar passed the resilium, a negative value was attri- buted. The presence and position of the demibranch re- tractor scar (linked or not to the MBR scar) was also noted. When this scar was not markedly visible on the valve, it was registered as absent. To eliminate size ef- fects, B. darwinianus individuals larger than 20 mm were not used for comparisons of taxonomic characters. Our purpose while studying morphometric relation- ships was to find out an objective form to distinguish the two species. For this reason, we investigated attributes that could be linked to general shell shape and express differences in the allometry of B. darwinianus and B. solisianus. Morphometric relations were studied fitting each pair of variables X and Y to the allometric equation, using least squares regression: log Y=a+blogX where a and b are constants (Seed, 1980). Species and localities were compared using analysis of covariance (ANCOVA), according to the following model: Y,=S,+L,+ S*L, + C + S#C + € A MB eo L DRM MBR L DRM height -————_H| length width Figure | Morphological shell characters analyzed in this study. A. Brach- idontes darwinianus. B. Brachidontes solisianus. Abbreviations: L, ligament; MBR, mean retractor muscle scar; DRM, demi- branch retractor muscle scar. where S; denotes species i, L,; the locality 7 effect, C is the covariable being analyzed and e the experimental er- ror (Draper & Smith, 1981). As our goal was not to study differences among different localities, we searched for re- lations that were constant across sites with different hy- drodynamic forces. Thus, when the interaction term (S*L;;) was significant, the relation analyzed was not con- sidered a good taxonomic predictor, as its effect within a species would depend on the beach where the sample was collected. Separate variables studied were analyzed using a two- way ANOVA (Underwood, 1981). For all linear models, assumptions of normality and homogeneity of variances were tested by plotting the residuals against the estimated values. These were visually inspected to detect trends due to violations of the assumptions or any pattern not con- sidered by the models (Box et al., 1978). To analyze variation of the muscle scars, observations were pooled within length classes determined by 5 mm intervals of shell length. Presence of the demibranch scar was analyzed in these classes using log-linear models in a three-way contingency table (Agresti, 1990). In this analysis, the model being analyzed lacks one of the var- iables of interaction terms, and is tested against the sat- urated model (with all variables and interaction terms possible). In this way, it is possible to test the significance of each interaction term. If the interaction is significant, the main effects are also significant, as log-linear models are hierarchical (Agresti, 1990). The model with the least number of variables and interactions that does not signif- icantly differ from the saturated model is the best one to describe the data (Agresti, 1990). Thus, it was possible M. O. Tanaka & C. A. de Magalhaes, 1999 20.1 A B Page 269 C 15.9 11.6 Y =- 0.44 + 0.93 X Y =- 0.29 + 0.86 X =-0.21+0.85X 2 r=0.90 p<.001 r=0.90 p<.001 r=0.90 p<.001 7.4 5.8 3 e. 6p) 43 ° oo. eo « SE a 2.7 AY =-0.48 + 0.88 X Nee Y =- 0.43 + 0.85 X Y =- 0.34 + 0.81 X e 2a os r=0.90 p<.001 Se r=0.90 p<.001 ne) r=0.90 p<.001 £ =—_ — iS Roy L me 20:4 @ is9| D E F D 116 Y =- 0.49 + 0.94 X Y= - 0.63 +1.02 X Y =-0.52 + 0.95 X F=0.90 p<.001 P=0.90 p<.001 P=0.90 p<.001 7.4 5.8 43 2.7 is 24 , Y=-0.51 + 0.95 X Y =- 0.50 + 0.89 X y Y =- 0.54 + 0.90 X 16 of r=0.90 p<.001 r=0.90 p<.001 ee r=0.90 p<.001 12 1620 32 23 4 67 Shell Length (mm) Figure 2 12 1620 32 23 4 67 12 1620 32 Shell height and length relations for B. darwinianus (solid circles, solid lines) and B. solisianus (open circles, dashed lines) within the sites sampled: A. Barequecaba, B. Dura, C. Lagoinha, D. Lazaro, E. Milionarios, E Soares. Specific equations for each relation are presented. Numbers in parenthesis are standard errors for the determination coeffi- cients. All values are plotted in logarithmic scale. to test if the position or presence of these scars were dependent on the species considered, or on the shell length of the animal. All individuals of each species were grouped, as a preliminary survey did not show any pat- tern among localities. All analyses were done with SYS- TAT software (Wilkinson, 1990). RESULTS Analysis of residuals did not detect violations of the as- sumptions made on the linear models used. Our results indicate a great variation of shell characters, with great overlap of shell proportions in the smaller length classes. Shell length was extremely variable among beaches: B. darwinianus varied from 3.0 to 36.0 mm, while B. soli- sianus ranged between 2.2 and 20.0 mm. B. darwinianus had higher and wider valves than B. solisianus, varying from 1.8 to 16.3 mm in height and 0.1 to 12.2 mm in width, while B. solisianus showed height between 0.3 to 7.9 mm and width from 0.1 to 9.2 mm. The relationship between shell height and length was linear in both species (see values for regression coefficients in Figure 2). This relation was determined only by species, without any in- teraction with sampling sites (Table 1). The slopes dif- fered significantly (P < 0.001), with B. darwinianus in- creasing faster in height than B. solisianus of the same shell length (Table 1). This means that the regression lines for the two species cross, and for individuals with shell lengths smaller than 5 mm the values for relative height may overlap. Nevertheless, height: length ratio separates both species (Figure 3). Values for this ratio, however, depend on the relative shell length of the individuals; B. darwinianus maintains fairly constant values at small lengths, decreasing as the individuals attain larger shell lengths, while B. solisianus rapidly decreases this value. Table 1 Analysis of covariance for shell relations between B. dar- winianus and B. solisianus; the underlined variable is the dependent one being tested in the ANCOVA model. ANOVA Sum of Source DF — squares Ee iB Height 7° = 0.99 Localities 5 0.052 2.060 0.070 Species 1 0.003 0.666 0.415 Localities*Species > 0.009 0.339 0.889 Length 1 86.475 17,234.886 <0.001 Length*Species l 0.082 16.317. <0.001 Error 345 1.731 Width 7° = 0.98 Localities 5 0.764 16.868 <0.001 Species 1 0.076 8.390 0.004 Localities*Species 5 0.385 8.496 <0.001 Length 1 105.716 11,668.061 <0.001 Length*Species 1 0.226 24.965 <0.001 Error 344 Sail ly Width 7° = 0.96 Localities 5 0.520 7.424 <0.001 Species l 0.001 0.048 0.826 Localities*Species 5 0.412 5.881 <0.001 Length 1 103.160 7,362.554 <0.001 Length*Species 1 0.533 38.021 <0.001 Error 344 4.820 Both patterns illustrate the changes in the slope of the untransformed relation between height and length, but within the length classes where both species overlap (O— 20 mm), the 99% confidence intervals for each length class indicate a clear separation (Table 2), except for the O—5S mm length class. Shell width and length were positively correlated in both species (Figure 4), but a significant interaction be- tween species and sites was detected (Table 1). This means that differences observed for the two species were dependent on the site considered. Nevertheless, B. soli- sianus seems to become thicker more frequently than B. darwinianus, in individuals greater than 10 mm. B. dar- winianus tended to have taller and narrower shells than B. solisianus, but this was also site-related (see analysis for Width < Height in Table 1). In one site (Soares) the relation was weak for B. darwinianus, with greater vari- ation within shorter individuals. The angle formed by the ventral margin and the liga- ment showed a great variation both within species (ac- cording to shell length) and among beaches, with strong interaction between species and localities (Table 3), and was a poor predictor to distinguish between the two spe- cies. Nevertheless, as a general trend, B. darwinianus in- dividuals exhibited greater angles between shell margins, The Veliger, Vol. 42, No. 3 eB. darwinianus © B. solisianus 0.60 0.55 ¢ , ¢ 0.45 Height:Length Ratio 0-5 5-10 10-15 15-20 20-25 25-30 30-35 35-40 Shell Length Classes (mm) Figure 3 Mean values (+ SE) of the height: length ratio for B. darwini- anus (solid circles) and B. solisianus (open circles) along 5 mm length classes. as indicated by their ranges (B. darwinianus: 30—60°; B. solisianaus: 15—55°). The distance from the median byssal retractor scar to the ligament varied with shell length in B. darwinianus (Figure 5). In individuals smaller than 15 mm, this scar scarcely passed the ligament, while in larger individuals the scar extended beyond the ligament, but there was a great variation within this character. Conversely, in B. so- lisianus the position of the scar almost never reached the ligament, irrespective of shell length (Figure 5), but this species includes only the smaller length classes found in B. darwinianus; thus, individuals from the same size can Table 2 Analysis of variance for the height: length ratio along dif- ferent length classes smaller than 20 mm, and 99% con- fidence intervals for the ratio, showing dependence on length classes for B. solisianus. ANOVA Sum of Source DF _ squares F P Length class 3. 0.118 332512 <0.001 Species 1 0.367 35.62 <0.001 Length class*Species 3 0.068 20.65 <0.001 Error 287 = =0.317 ener B. darwinianus B. solisianus classes Lower 99% Upper 99% Lower 99% Upper 99% 0-5 0.533 0.579 0.509 0.539 5.1-10 0.550 0.576 0.488 0.514 10.1-15 0.536 0.578 0.441 0.461 15.1—20 0.531 0.561 0.407 0.437 M. O. Tanaka & C. A. de Magalhaes, 1999 B Page 271 Cr PEM Ci . 7.41 Y=-1.01+1.09X Y =- 1.021 + 1.050 X \\ ss Y =-0.845+1.023X \ ° 0 oe 5.81 r=0.96 p<.001 r=0.98 p<.001 r=0.99 p<.001 ¢° @ 2 4.3 hz, 2.1 1.6 2 : ; 6° Y =- 0.811 + 0.980 X Y =- 0.655 + 0.880 X ; = - 0.972 + 0.987 X fe) ¢ Fie} AG r= 0.97 p<.001 ie f= 0.99 p<.001 “oe P=0.99 p<.001 11.6] D tee oO E - \ 7.41 Y=-0.909+1.043K \ Soe 5.81 =097 p<.001 4% y 5 Shell Width (mm) 4.3 2.7 2.1 1.6 ° £ = - 0.719 + 0.926 X “valle af r= 0.98 p<.001 ay Pere ae Sy Y =-0.657+0.980X — r=0.89 p<.001 ~ & , Y =- 0.995 + 1.070 X I Y =- 0.486 + 0.854 X =0.97 p<.001 ae r= 0.96 p<.001 203 4 67 12 1620 32 2 3 4 67 12 1620 32 23 4 67 12 1620 32 Shell Length (mm) Figure 4 Shell width and length relations for B. darwinianus (solid circles, solid lines) and B. solisianus (open circles, dashed lines) within the sites sampled. Conventions as in Figure 2. be wrongly assigned to one of the species, if only this character is used. Specimens from the other sites (Cigar- ras, Pina, Rasa, and Saquarema) showed the same pat- terns. Log-linear analysis for the presence of the demibranch retractor muscle scar showed that there is conditional in- dependence between the presence of this scar and the length classes, depending on the species (Table 4). The fitted model was Inf. =- a a; F B ar Ve oF ap, ar aVix where a; is the effect of the species i, 6; is the effect of the length class j, and y, is the response variable, referring to the position k of the demibranch scar (present or not in the shell) (Likelihood Ratio Chi-Square = 5.26, df = 6, P = 0.511). This means that the chance of detecting a scar depends on the shell length of the individual. Both effects were significant, but there was no interaction be- tween them (they are independent). These effects were different for the two species: in B. darwinianus this scar is almost always present, independent of shell length. In B. solisianus almost 40% did not present this scar, with a greater chance of not detecting the scar at all depending on the length class considered. When present, the position of the demibranch scar did not depend on the species or on the length class exam- ined. The final fitted model for these three variables was Inf. = ML a a) ap B; ar Ve a0 ap, (Likelihood Ratio Chi-Square = 11.21, df = 7, P = 0.13). Although there was an interaction between species and length classes, there was no influence of these factors on the position of the demibranch scar, whether linked or not to the MBR scar. DISCUSSION Traditional morphological characters frequently applied in the taxonomy of the species within the genus Brachi- dontes proved to be inconstant over a wide shell length range in this study. Table 3 Mean values for the angle formed by the ventral margin and the ligament and analysis including only sites with both species. Individuals from different length classes were pooled, as not all classes were present in the sites studied. B. darwinianus B. solisianus Mean Mean (std. err.) Range (std. err.) Range Barequegaba 44.1 (0.7) 35-55 36.0 (1.3) 25-55 Dura 38.5 (0.8) 30-45 30.5 (1.0) 20-40 Lagoinha 42.5 (0.8) 30-50 32.5(1.0) 15-40 Lazaro 45.2 (1.3) 35-60 37.0 (0.9) 30-50 Milionarios 39.0(1.7) 30-60 32.5.(1.2) 20-45 Peres 40.3 (0.9) 30-50 37.9 (0.9) 30-45 Cigarras 33.3 (1.6) 15-45 Pina 33.6(1.0) 25-45 Rasa 34.5 (1.3) 25-50 Saquarema 33.7(1.2) 20-45 ANOVA Sum of Source DF squares F P Localities 5) 1,938.567 11.869 <0.001 Species 1 4.585.677 140.381 <0.001 Localities*Species 5 500.917 3.067 0.010 11,269.758 Error 345 Individuals of the species studied showed maximum adult length similar to specimens collected in other sites (Klappenbach, 1965; Avelar & Narchi, 1984a; Nalesso, 1988; Rios, 1995). In these studies, B. darwinianus at- tained a maximum length of 33 mm, while B. solisianus reached a maximum length of 20.1 mm. The sites where we found B. darwinianus individuals with greatest shell lengths had the widest distribution of this species in re- lation to B. solisianus. The former species is found cov- ering the inferior intertidal zone, while B. solisianus is distributed upper in the intertidal (Nalesso, 1988; Tanaka, unpublished data). In the other beaches B. darwinianus is more restricted, occupying patches related to the presence of fresh water flow, or in low proportions of mixed spe- cies beds. All allometric relations examined varied according to the species and locality considered. Individuals of B. dar- winianus become higher within the same length variation of B. solisianus, with shells also becoming more curved. This character varied consistently among the species and the shores sampled, without interaction between them. Seed (1968) showed that in Mytilus edulis shells tend to grow taller and become more curved as individuals attain their maximum length, or when the animals occur in dense populations. This pattern seems to occur in B. dar- winianus, as this species can reach high densities due to The Veliger, Vol. 42, No. 3 the secondary substratum formed by the mussel bed: re- cruitment can occur over the adult individuals, and young mussels can be found with their byssus attached to the shells of the older ones (Tanaka, unpublished data). The best taxonomic character we could find using mor- phometric relationships was the height: length ratio. Al- though at the smallest length class this ratio may overlap (Table 2), this may only reflect similar shell characteris- tics within post-recruited individuals and a greater plas- ticity in adults (Seed, 1980). Larger individuals can be readily separated with 99% of confidence using the values calculated. In the smallest length classes, a good taxonomic pre- dictor is the presence of radial ribs over all the shell sur- face in B. darwinianus, while in B. solisianus these ribs are present only in the shell posterior margin. The number of ribs in B. darwinianus varies depending on the salinity conditions (Nalesso et al., 1992): individuals grown next to rivers present fewer ribs than individuals growing in the sea. In exposed shores, shells are also more eroded, making it more difficult to distinguish very young indi- viduals of the two species. Greater angles between shell margins were also used by Avelar & Narchi (1984a,b) to distinguish B. darwin- ianus from B. solisianus. They proposed a value around 40° for B. darwinianus and 30° for B. solisianus and stat- ed that those values differ for specimens of the same size. We have not found consistency in this character because it exhibited great variation related to shell length, as well as to sample location. The position of the median byssal retractor scar is not a consistent character to separate the species studied. Klappenbach (1965), however, used this character to sep- arate B. exustus and B. solisianus from B. rodriguezi and B. darwinianus. In the latter two species, the scar reaches or extends beyond the ligament, while in the former it falls short of the ligament edge. Within B. solisianus the scar does not reach the ligament, and the distance is con- stant independent of shell length (see Figure 5). In B. darwinianus samples, the scar approaches the resilium as the animal grows, passing it when the shell reaches 15— 20 mm. According to Klappenbach (1965), B. exustus presents terminal umbones, while B. darwinianus has subterminal ones. Two subspecies have been described for B. darwinianus, based on the position of the MBR scar: B. d. darwinianus (with thin radial ribs, MBR scar surpassing the ligament) and B. d. mulleri (with thicker radial ribs, MBR scar only just reaching the ligament). Our data does not support the subdivision of this species based solely on shell characteristics, as they may vary both during the ontogeny of the individuals and according to environmental forces. Although there is a great varia- tion in B. darwinianus, the length of this scar seems to be related to animal size, and in both species individuals smaller than 20 mm present similar scars. Avelar & Narchi (1984a, b) suggested that the position M. O. Tanaka & C. A. de Magalhaes, 1999 Distance from MBR scar to ligament (MM) 0-5 5-10 10-1515-2020-25 25-30 >30 0-5 5-10 10-15 15-2020-25 25-30 >30 Page 273 0-5 5-10 10-15 15-2020-25 25-30 >30 Shell Length Classes (mm) Figure 5 Values of the distance from MBR scar to ligament against shell length classes for B. darwinianus (solid circles) and B. solisianus (open circles) within the sites sampled: A. Barequegaba, B. Dura, C. Lagoinha, D. Lazaro, E. Milionarios, E Soares. of the demibranch retractor muscle scar could be useful to separate B. darwinianus and B. solisianus. However, we found no differences in the position of the scar either be- tween species or among shell length classes. Therefore, no taxonomic value can be assigned for this character. On the other hand, our results show that in B. darwinianus this scar is almost always present, while in B. solisianus a great proportion of shells examined presented no scars at all. This study revealed a great plasticity of shell characters in species of the genus Brachidontes. Such variability was also reported for Mytilus edulis by Seed (1968). He em- phasized that in species with planktonic development the uncertainty of the recruitment site favors wide phenotypic expression. Phenotypic plasticity is adaptive for organ- isms inhabiting unpredictable sites and/or with wide geo- graphical range (Brown, 1985; Etter, 1988). These apply to the species of Brachidontes studied, because of the existence of a planktonic dispersal phase. Morphological characters like shell sculpture and geometry are frequent- ly complex and usually determined by many loci that can themselves be polymorphic (Berger, 1983). Species with wide dispersal capacity are exposed to a great range of different habitats in a heterogeneous environment (Jan- son, 1987). As a consequence, for such cases, Grassle & Grassle (1978) suggest that individuals presenting a great degree of variation in genetic loci related to adaptive characters are favorably selected. McDonald et al. (1991) stated that allozyme characters should be the primary means of distinguishing among species of the genus My- tilus. This may also apply for other mussels such as Brachidontes. Genetic studies in Brachidontes would be very useful to clarify the distinction among species and the selective mechanisms that determine species form and distribution. For ecological purposes, recruits of the two species are not distinguishable using shell characteristics, but the height: length ratio forms a strong character for species discrimination between B. darwinianus and B. so- lisianus adults. Table 4 Frequencies of shells with the demibranch retractor scar linked or not to the median byssal retractor scar. For B. darwinianus only shells with length smaller than 20 mm were considered; there was no difference among the po- sition of scars in longer shells. Position of scar eneth Present Species class (mm) Linked Separated Absent B. darwinianus 0-5 2 3 0 5-10 18 26 1 10-15 13 5 0 15-20 18 13 0) B. solisianus 0-5 4 3 Uf 5-10 20 13 14 10-15 29 11 33 15-20 13 8 12 ACKNOWLEDGMENTS We thank A. L. T. Souza for helping with the measure- ments and discussions on this work, Drs. W. W. Benson, K. S. Brown, B. Roth, and two anonymous referees for providing helpful comments on an earlier draft of this manuscript. MOT was financially supported by a FA- PESP fellowship (proc. 95/2260-0). The early ideas of this work were developed during a course on biosyste- matics by Dr. K. S. Brown. LITERATURE CITED AGrESTI, A. 1990. Categorical Data Analysis. John Wiley & Sons: New York. 558 pp. AVELAR, W. E. P. & W. Narcui. 1984a. Anatomia funcional de Brachidontes darwinianus darwinianus (Orbigny, 1846) (Mollusca: Bivalvia). Papéis Avulsos de Zoologia 35:33 1— 359: AVELAR, W. E. P. & W. NaRcHI. 1984b. Functional anatomy of Brachidontes solisianus (Orbigny, 1846) (Bivalvia: Mytili- dae). Boletim de Zoologia 8:215—237. BERGER, E. M. 1983. Population genetics of marine gastropods and bivalves. Pp. 563—596 in W. D. Russel-Hunter (ed.), The Mollusca. Vol. 6. Academic Press: London. Box, G. E. P, W. G. HUNTER & J. S. HUNTER. 1978. Statistics for Experimenters: an Introduction to Design, Data Analysis and Model Building. John Wiley & Sons: New York. 653 Pp. Brown, K. M. 1985. Intraspecific life history variation in a pond The Veliger, Vol. 42, No. 3 snail: the roles of population divergence and phenotypic plasticity. Evolution 39:387—395. Brown, R. A., R. SEED & R. J. O’CONNoR. 1976. A comparison of relative growth in Cerastoderma (= Cardium) edule, Mo- diolus modiolus and Mytilus edulis (Mollusca: Bivalvia). Journal of Zoology 179:297-315. Draper, N. R. & H. SmitH. 1981. Applied Regression Analysis. 2nd ed. John Wiley & Sons: New York. 709 pp. Etter, R. J. 1988. Asymmetrical developmental plasticity in an intertidal snail. Evolution 42:322—334. GRASSLE, J. EF & J. P. GRASSLE. 1978. Life histories and genetic variation in marine invertebrates. Pp. 347—364 in B. Battag- lia & J. A. Beardmore (eds.), Marine Organisms: Genetics, Ecology and Evolution. Plenum Press: New York. JANSON, K. J. 1987. Allozyme and shell variation in two marine snails (Littorina: Prosobranchia) with different dispersal abilities. Biological Journal of the Linnean Society 30:245— 256. KLAPPENBACH, M. A. 1965. Lista preliminar de los Mytilidae brasilenos con claves para su determinacion y notas sobre su distribucion. Anais da Academia Brasileira de Ciéncias 37(supl.):327-352. Lent, C. M. 1967. Effects of habitat on growth indices in the ribbed mussel, Modiolus (= Arcuatula) demissus. Chesa- peake Science 8:221—227. McDOonaLbD, J. H., R. SEED & R. K. KOEHN. 1991. Allozymes and morphometric characters of three species of Mytilus in the Northern and Southern Hemispheres. Marine Biology 111:323-333. Morton, B. 1992. The evolution and success of the heteromy- arian form in the Mytiloida. Pp. 21-52 in E. Gosling (ed.), The Mussel Mytilus: Ecology, Physiology, Genetics and Culture. Elsevier: Amsterdam. NALEsSO, R. C. 1988. Influéncia da salinidade e exposi¢ao ao ar na distribuicao dos mexlhdes Brachidontes darwinianus e B. solisianus em dois estudarios do litoral do Estado de Sao Paulo. Msc. Thesis. Instituto de Biologia, Universidade Es- tadual de Campinas. Na esso, R. C., L. E L. DUARTE & E. G. MENDES. 1992. Phe- notypic plasticity in Brachidontes darwinianus (Bivalvia: Mytilidae). Revista Brasileira de Biologia 52:245—249. Rios, E. C. 1995. Seashells of Brazil. 2nd ed. Museu Oceano- grafico da FURG, Rio Grande do Sul. cii + 328 pp. SEED, R. 1968. Factors influencing shell shape in the mussel My- tilus edulis. Journal of the Marine Biological Association of the United Kingdom 48:561—584. SEED, R. 1980. A note on the relationship between shell shape and life habits in Geukensia demissa and Brachidontes ex- ustus (Mollusca: Bivalvia). Journal of Molluscan Studies 46: 293-299. UNDERWOOD, A. J. 1981. Techniques of analysis of variance in experimental marine biology and ecology. Oceanography and Marine Biology—an Annual Review 19:513—605. WILKINSON, L. 1990. SYSTAT for DOS. Version 5.0. SYSTAT: Evanston, Illinois. The Veliger 42(3):275—277 (July 1, 1999) THE VELIGER © CMS, Inc., 1999 Early Development of Fissurella picta (Gmelin, 1791) M. L. GONZALEZ, M. C. PEREZ, D. A. LOpPEz, J. M. URIBE AND C. A. PINO Universidad de Los Lagos, Departamento de Acuicultura, Laboratorio de Cultivos Marinos, Casilla 933, Osorno, Chile (e-mail: malugon @ puyehue.di.ulagos.cl) Abstract. The early stages of the embryonic and larval development of the limpet Fissurella picta (Gmelin, 1791) (“‘lapa’’) are described. The period after fertilization and until the initial trochophore stage at 10°C was 74 hr. The principal characteristic of the embryonic development is the presence of a gelatinous coat. Routine methods of spawning induction did not produce positive results. The results suggest that in spite of some critical events, it is possible to obtain larvae until the early trochophore stage under laboratory culture conditions, which marks significant progress in cultivating this species, given the intense ex- ploitation and the high commercial value of the “‘lapa.”’ INTRODUCTION In spite of the growing commercial importance of species of the genus Fissurella (locally referred to as ‘‘lapa’’) in Chile (Bretos et al., 1988a; Oliva & Garrido, 1994), little is known about its embryonic and larval development. The knowledge of early life stages is fundamental to the evaluation of the feasibility of cultivation of species of this genus. The breeding cycle of Fissurella picta will be reported (Pérez et al., in preparation). The aim of this study was to evaluate spawning and fertilization under artificial conditions, as well as to de- scribe the early development of F. picta (Gmelin, 1791), which is one of the principal “‘lapas”’ species from south- ern Chile (Bretos et al., 1988b). MATERIALS anp METHODS During the period of maximum gonadic maturity (July), specimens of Fissurella picta, measuring over 5 cm, were taken from the rocky intertidal zone of Metri Bay, south- ern Chile (41°36’S, 72°42’W). Epibionts were removed and the specimens of F. picta were maintained in the laboratory for an acclimation period of 20 days. Fifteen animals were placed in an aquarium at 10°C, 32%o salin- ity, with a constant air supply, and fed ad libitum, to increase chances of spontaneous spawning, according to the methodology described by Giese & Pearse (1974). The following alternative methods of inducing spawn- ing were also tested (15 specimens for each treatment): (a) thermal shock: animals were maintained for 4 hr at a temperature of 5°C, and subsequently transported to a thermoregulated bath at a temperature of 24°C; (b) elec- tric shock: electric current from a 6V battery was applied to specimens maintained in seawater; (c) osmotic shock: specimens were injected in the siphon with KCI 0.5 M. In addition, gametes, obtained by dissection of 15 spec- imens, were subsequently rinsed in filtered seawater (1 wm), and then fertilized using 15—17 sperm for each oo- cyte. Oocytes were rinsed after fertilization with sterile, filtered seawater (1 wm), and maintained in the dark at 10°C, with a constant air supply, and water change every 2 days during the entire period of development. RESULTS Specimens acclimated in the laboratory and fed ad libi- tum spawned (eggs and sperm) spontaneously into the seawater. The release of oocytes and sperm through the apical hole of the shell indicates that external fertilization occurs in Fissurella picta. Two- and 4-cell embryos were first observed after 3— 4 hr of fertilization; they continued to develop to the trochophore stage within the egg shell after 72 hr (Table 1). The artificial spawning induction methods tested (ther- mal, electric, and osmotic shock) did not produce positive results. From the second half of July to the end of August (wintertime), coincident with the period of maximum ma- turity, we found mobile sperm in dissected gonads. At this time the oocyte gelatinous coat became less dense when rinsed in seawater. This enabled the sperm to pen- etrate through the micropile, into the interior of the oo- cyte, resulting in fertilization. After 15 to 20 min, im- mediately after fertilization, a series of changes occurred in the oocytes. A palear zone was produced in the cyto- plasm of the oocyte; in this area, 1 hr later, the polar corpuscle was eliminated (Figure la; Table 1). Two hours later segmentation began, which occurred, in its entirety, within the capsule which encases the oocyte. Initially blastomeres were produced. At this stage the polar cor- puscle was still present (Figure 1b), followed by the 4- cell embryo (Figure 1c; Table 1). The 8-cell embryo pre- sented macromers and micromers (Figure 1d). Approxi- mately 12 hr after fertilization, the blastula stage was The Veliger, Vol. 42, No. 3 Table 1 Post-fertilization period stages and timing of early devel- opment in Fissurella picta at 10°C, 32%o salinity. Period after fertilization Event (hours) Elimination of polar corpuscule l 2-cell embryo 2 4-cell embryo 3-4 Blastula 12 Gastrula ciliate within the egg shell 19 Initial trochophore in egg shell 72 Length of trochophore swimming 96 Length of maintenance in culture until trochophore dead 144 reached and after 19 hr, the ciliate gastrula was observed, rotating within the capsule, (Figure le). After 72 hr, the initial trochophore stage was observed (Figure 1f). The trochophore had an anterior ring of large cilia in constant motion, producing continuous larval rotation within the capsule. DISCUSSION The results described here show that the artificial routine methods for spawning did not produce positive results. Spawning and fertilization are possible under laboratory conditions in mature specimens of Fissurella picta. In mature specimens maintained in the laboratory under the conditions described in Materials and Methods, sponta- neous spawning occurred. Gametes could be obtained by gonad dissection. Nevertheless, the best results were achieved by maintaining specimens in the laboratory until spontaneous spawning occurred. However, there are prob- lems that must be overcome before culture can be carried out on a larger scale, i.e., larval development until meta- morphosis. Ward (1966) reported oocytes with a gelatinous coat in species of the genus Fissurella. In Chilean species of ve- nerids this has been described (Padilla, 1983; Guerra et al., 1994) where artificial fertilization was possible only under special conditions, such as osmotic shock (Padilla & Olivares, 1986). In spite of this, the components of this gelatine have been associated with the induction of ac- rosomal reaction (Epel, 1975). During periods of maximal maturity in F. picta, rinsing the eggs in seawater was sufficient to obtain fertilization of oocytes. In fact, the loss of gelatinous coat and subsequent fertilization oc- curred spontaneously in the experimental aquaria. The early trochophore is lecithotrophic, and may re- duce early mortality. Lewis (1960) studied trochophore larvae with a short pelagic lifespan in other species of the genus, such as F. barbadensis Gmelin. Similarly, trochophores and veliger larvae that go through gelati- e { . is Figure 1. a. Post-fertilization stages and early development in Fissurella picta at 10°C and 32%c salinity: the polar corpuscule (pc) is elim- inated; b. the first segmentation, formation of two blastomeres with polar corpuscule (pc); c. the 4-cell embryo; d. the 8-cell embryo with macromers (ma) and micromers (mi); e. the ciliate gastrule with rotatory movement at the interior of capsule; f. trochophore larvae scale bars = 50 wm. nous mass enveloping ovocytes and emerge as miniature adults have been reported in Diadora apertura (= F. re- ticulata (Boutan, 1886 in Ward, 1966)). Data for F. picta would indicate that it follows an intermediate pattern, be- tween these two types of development in spite of the fact that the length of the trochophore stage and timing of metamorphosis are not known. Given the intense exploitation and the high commercial value of the “‘lapa’”’ species in Chile (Bretos, 1988a), ba- sic information must be obtained to evaluate the feasibil- ity of intensive cultivation. Our results suggest that is possible to obtain larvae until the early trochophore stage, under artificial conditions. M. L. Gonzalez et al., 1999 ACKNOWLEDGMENTS The support by Fondo de Investigacion Cientifica y Tec- nolo6gica (FONDECYT) for financing this study (Grant 040-93) and by the Departmento de Investigacion, Univ- ersidad de Los Lagos (Grant 304-18) is gratefully ac- knowledged. We thank Dr. Barry Roth and an anonymous reviewer who made valuable comments on the manu- script. The collaboration of Robert Simpfendorfer, Susan Angus, and Sandra Mancilla is also appreciated. LITERATURE CITED Bretos, M., J. GUTIERREZ & Z. ESPINOZA. 1988a. Estudios bio- l6gicos para el manejo de Fissurella picta. Medio Ambiente 9(1):28-34. Bretos, M., V. QUINTANA & V. IBARROLA. 1988b. Bases biol6- gicas para el manejo de Fissurella nigra. Medio Ambiente 9(2):55-62. EpEL, D. 1975. The program of a mechanism of fertilization in the echinoderm egg. American Zoologist 15:507—522. Giese, A. C. & J. S. PEARSE. 1974. Introduction. Pp 1—49 in A. Page 277 C. Giese & J. S. Pearse (eds.), Reproduction of Marine In- vertebrates. Academic Press: New York. GUERRA, R., P. ESpONDA & J. ARRAU. 1994. La cubierta vitelina y el “‘jelly coat’’ en ovocitos de Protothaca thaca y Venus antiqua (Mollusca: Bivalvia). Libro Resimenes XIV Jor- nadas Ciencias del Mar, Puerto Montt, Chile. p. 152. Lewis, J. B. 1960. The fauna of rocky shores of Barbados, West Indies. Canadian Journal of Zoology 38:391—435. Outva, D. & J. GARRIDO. 1994. The impact of artisanal fishermen ““management areas” on the keyhole limpet fishery in Cen- tral Chile. Coustal Zone Canada’94, Halifax 20-23 Septem- ber, pp. 1661-1682. PapILLa, M. 1983. Ultraestructura funcional de los gametos de la almeja. Revista de Biologia Marina, Valparaiso 19(1):47— 62. PapILLA, M. & G. OLIvAREsS. 1986. Evaluacion de la madurez vitelogénica y reproduccion en oocitos extirpados de la al- meja Venus antiqua antiqua. Revista de Biologia Marina, Valparaiso 22(1):61—74. Warp, M. 1966. The breeding cycle of the keyhole limpet Fis- surella barbadensis Gmelin. Bulletin of Marine Science 16(4):685—695. The Veliger 42(3):278—288 (July 1, 1999) THE VELIGER © CMS, Inc., 1999 NOTES, INFORMATION & NEWS Occurrence of the Asian Clam Corbicula fluminea (Miiller, 1774) (Bivalvia: Sphaeriacea: Corbiculidae) in Colorado James R. Cordeiro!* and Sarah MacWilliams? *Zoology Section, University of Colorado Museum, Campus Box 315, Boulder, Colorado 80309-0315, USA *Department of EPO Biology, University of Colorado, Boulder, Colorado 80309, USA The invasive pest species, the Asian clam Corbicula flu- minea (Miiller, 1774) (Bivalvia: Sphaeriacea: Corbiculi- dae), was first reported in the United States in the Colum- bia River, Washington, in the 1930s (Britton & Morton, 1982). Since that time it has spread across the country into at least 35 states (Counts, 1986; McMahon, 1983). The success of this species in colonizing North American fresh waters is mainly due to its high fecundity, high growth rate, and quickly settling, free-living larval stage (McMahon, 1983). In addition, its great adaptability to disturbed environments has allowed it to infest these ar- eas quickly. Possible limitations to its dispersal are spec- ulative but include low water temperatures (French & Schloesser, 1991, 1996; Graney et al., 1980; Mattice & Dye, 1976; McMahon, 1983); exposure during low water (McMahon, 1983); reduced phytoplankton as a food source (French & Schloesser, 1996); low pH, causing shell dissolution (Kat, 1982); overpopulation resulting in massive die-offs (Sickel, 1986), and competition with ze- bra mussels Dreissena polymorpha (Pallas, 1771) (Hebert et al., 1989; Nalepa & Schloesser, 1993). In the United States Corbicula fluminea currently rang- es as far north as the Great Lakes (Clarke, 1981; Counts, 1986; French & Schloesser, 1991, 1996; Scott-Wasilk et al., 1983) and Washington (Britton & Morton, 1982; Counts, 1986), although it has not yet invaded the New England states (Counts, 1986). Southern distribution oc- cupies all southern states including the southern tips of Texas and Florida (Counts, 1986; McMahon, 1983). C. fluminea was first reported in the Platte River as dead shells in Lancaster and Dawson Counties, Nebraska (Freeman & Perkins, 1992). However, Peyton & Maher (1992, 1995) found living populations soon after. Neigh- boring Kansas has populations along the Kansas River drainage and one population on the North Fork of the Ninnescah River (Arkansas River drainage) (Counts, 1991). In 1993, C. fluminea was first discovered in neigh- ' Present Address: Department of Invertebrates, American Mu- seum of Natural History, Central Park West at 79th Street, New York, New York 10024, USA. boring Colorado in Cherry Creek Reservoir, Arapahoe County (Figure 1), part of the Platte River drainage (Nel- son & McNabb, 1994). Since then, Crane et al. (1996) found the species in Highline Lake, Mesa County (Figure 1), in the western part of the state far removed from the Cherry Creek population. Kreiser & Mitton (1995) spec- ulated that C. fluminea could be adapting to colder cli- mates without the benefit of apparent thermal refuges such as power plants, despite high winter mortalities. Pre- viously, it had been reported that severe cold causes heavy mortality in the species (French & Schloesser, 1991, 1996) especially at temperatures below 2°C (Mat- tice & Dye, 1976). We wish to report on the distribution of Corbicula flu- minea in Colorado, including several new localities, all in areas that experience temperatures below 2°C, indicat- ing it can survive and reproduce in colder climates. In 1996, while surveying freshwater mussels along the Arkansas River drainage in Colorado, Cordeiro observed three previously unreported populations of Corbicula flu- minea in Queen’s Reservoir, in the Arkansas River below the dam of John Martin Reservoir, and in Pueblo Reser- voir (Figure 1). MacWilliams surveyed the perimeter of Pueblo Reservoir in November 1996, by canoe and on foot, to determine the approximate density and substra- tum for C. fluminea at that location. Queen’s (Neeskah) Reservoir is a small pond located 5.5 miles north and 5.25 miles east of Big Bend at an elevation of 3875 feet in Kiowa County. It is part of the Colorado State Park System. In Queen’s Reservoir, Cor- deiro found three intact, empty shells plus 30 valves in July 1996. Shell length ranged from 13.8 to 32.5 mm (mean 26.75 mm, n = 33). All the exposed specimens were found at least 10 meters above the water line. The reservoir had receded considerably from the high water mark where bottom sediment was coarse sand to a point where bottom sediment was a very fine, silty, thick mud. The clams were all found on the exposed sediment 10 meters or more from the receded water line, beneath which was black mud. The sandy sediment where the clams were found was less than 12 cm deep atop a much deeper layer of dried mud. Prolonged searching for sev- eral hours revealed no living specimens or dead shells underwater or less than 10 meters from the water’s edge. In October 1996, Cordeiro discovered a second popu- lation west of Queen’s Reservoir in a pool of the Arkan- sas River below the dam to John Martin Reservoir, Bent County. Six valves and five living specimens were col- lected. The species is reportedly surviving in John Martin Yampa and White a } es VEZ Rivers Basin vs ro, cok Flatte River fe Pe \ ) ee ee Ze , 3 ‘a ¢ SIR rman. Ly & > } ay at J yA gj a ~ Be oa Colorado River ~ f aT ses Ney ) ea is a Basin / ~ yaa ) by \ >) UP Gy) \ Arkansas River ~| 2 wa Basin a Sy y a ) S Le yee 3 Rio Grande as, j ae ap , op i hay AE Basin RZ pg va Si ie I~ Niareits / San Juan 7 ~ ies a f River Basin | Sea ea | Figure | Distribution of Corbicula fluminea (Miller) in Colorado. Closed black circles designate localities where C. fluminea has been col- lected in Colorado. A = Highline Lake, Mesa County, summer 1995 (Crane et al., 1996). B = Cherry Creek Reservoir, Arapa- hoe County, June 1993 (Nelson & McNabb, 1994). C = Queen’s Reservoir, Kiowa County, July 1996 (present study). D = Ar kansas River at John Martin Reservoir, Bent County, October 1996 (present study). E = Pueblo Reservoir, Pueblo County, No- vember 1996 (present study). Reservoir (Dr. Scott Herrmann, University of Southern Colorado, personal communication, August 1996). In August 1996, Cordeiro and MacWilliams found Corbicula fluminea in Pueblo Reservoir at Lake Pueblo State Park, Pueblo County, surviving in relatively great numbers. Several individuals were observed in the sandy beach sediment. Few dead shells and many small indi- viduals of less than 5 mm were noted, suggesting that the population is healthy. Samples of the Pueblo Reservoir population were col- lected in November 1996 using a drag net sampler (width 0.305 m) according to the method of Britton & Morton (1982). Individual clams were counted and measured for length with vernier calipers to the nearest millimeter and sorted into three size categories: < 6 mm, 6—13 mm, and 14—27 mm. Clams less than 4 mm were difficult to see within the sample and time-consuming to sort in the field; therefore, the smaller category reflects mainly clams of 4—5 mm length. Categorization of substratum type was determined visually as silt-mud, sand-gravel, or cobble. The mean density of Corbicula fluminea, determined for each of the three substratum types by dividing number of individuals for each substratum type by number of square meter samples taken, was found to be highest in the sand/ gravel substratum (7.5 individuals/m?). Cobble substrata yielded a mean number of 1.75 individuals/m?. Silt/mud substrata had no clams. Sizes ranged from 6 to 27 mm although most clams sampled were in the 6—13 mm size category (87% of gravel-collected clams and 86% of cob- Notes, Information & News Page 279 ble-collected clams). The only substratum to contain all three size categories was gravel. Curiously, no clams of a size less than 6 mm were found in the cobble substra- tum. Mean size of all clams sampled was 8.81 mm (n = 30). Distribution within Pueblo Reservoir was found to be relatively patchy. Although all areas sampled containing gravel or cobble yielded some clams, the density varied considerably. One sample area in particular yielded 51% of all the clams sampled for the entire survey. Density in this area was 19.0/m/?. The presence of all size classes in Pueblo Reservoir, up to the largest clam, 27 mm, is indicative of a healthy population with immature individuals, recently mature adults, and reproductively active adults. A lack of small clams (less than 6 mm) in the cobble substratum may reflect both limitations in sampling methodology, as many small (< 4 mm) individuals were easily overlooked, as well as the difficulties associated with living in cobble substrata due to shell crushing by the cobbles or an in- ability to burrow among them. Patchiness could result from the lack of a planktonic larval stage. Larvae of Cor- bicula fluminea are brooded within the gills resulting in a fully formed juvenile released through the exhalent si- phon in the immediate vicinity of the adult (McMahon, 1983). This causes a patch of clams to form after colo- nization around the original colonizers in an area. The evidence presented here indicates that Corbicula fluminea may be propagating offspring in Pueblo Reser- voir despite seasonal fluctuations in temperature and win- ter ice cover. The presence of the clams below the dam of John Martin Reservoir indicates that C. fluminea may be surviving in the deeper waters of the Arkansas River. The area receives river water from John Martin Reservoir through an opening at the base of the dam. This area is also very close to the bottom of the reservoir where cold- er bottom water flows under the dam. All of these areas (Queen’s Reservoir, John Martin Reservoir, and Pueblo Reservoir) are part of the Arkansas River drainage. Prior to these observations, Corbicula flu- minea had never been documented in the Arkansas River in Colorado. A possible pathway for the spread of the species may be along the Arkansas River drainage. Pop- ulations in Arapahoe, Kiowa, and Bent Counties follow the drainage from the southwest. The population in High- line Lake, west of the Continental Divide, could be an introduction from the pet trade. At least one Denver area commercial pet dealer was selling Corbicula as an aquar- ium novelty (personal observation, August 1996). The University of Colorado Museum also includes one spec- imen (UCM 41599) purchased in 1992 from a Denver area mail order pet dealer. The Cherry Creek population may be indicative of dispersal down the South Platte drainage through Nebraska or another introduction. Cher- ry Creek Reservoir experiences a great deal of boat and Page 280 The Veliger, Vol. 42, No. 3 jet-ski traffic as well as recreational fishing whereby clams could be transported. This species has clearly developed a foothold in two different Colorado drainages, the South Platte and Arkan- sas. Unfortunately, lack of mortality data prevents true measurements of cold tolerance. The presence of Corbic- ula fluminea in Colorado, especially in Pueblo Reservoir at an elevation of 5548 feet and temperatures often below freezing, and Highline Lake in Mesa County at an ele- vation of 4700 feet, is indicative of a rapid spread throughout the rest of the state in the near future. ACKNOWLEDGMENTS We thank Dr. Shi-Kuei Wu of the University of Colorado Museum for reviewing this manuscript, Dr. John Bushnell of the University of Colorado Department of Environ- mental Population and Organismic Biology for use of his sampling equipment, and Mr. Mike French, Park Manager at Lake Pueblo State Park for permission to conduct our research on Pueblo Reservoir. Voucher specimens (UCM Nos. 41575, 41576, 41593, and 41600) were deposited in the University of Colorado Museum, Zoology Section. LITERATURE CITED Britton, K. C. & B. Morton. 1982. A dissecting guide, field and laboratory manual for the introduced bivalve Corbicula fluminea. Malacological Review, Supplement 3:1—82. CLARKE, A. H. 1981. Corbicula fluminea, in Lake Erie. The Nau- tilus 95(2):83-84. Counts, C. L., HI. 1986. The zoogeography and history of the invasion of the United States by Corbicula fluminea (Bi- valvia: Corbiculidae). American Malacological Bulletin, Special Edition (2):7—39. Counts, C. L., Il. 1991. Corbicula (Bivalvia: Corbiculidae). Tryonia: Miscellaneous Publications of the Department of Malacology, The Academy of Natural Sciences of Philadel- phia (21):1—134. CRANE, K. C., D. T. CARDIN & W. R. ELMBLAD. 1996. Summary of 1995 inventory of aquatic mollusks in northwestern Col- orado. Colorado Division of Wildlife: Grand Junction, Col- orado. 8 pp. FREEMAN, P. W. & K. P. PERKINS. 1992. Survey of mollusks of the Platte River. Report of the U.S. Fish and Wildlife Ser- vice: Grand Island, Nebraska. 37 pp. FrencH, J. R. P., HI & D. W. SCHLOESSER. 1991. Growth and overwinter survival of the Asian clam, Corbicula fluminea, in the St. Clair River, Michigan. Hydrobiologia 219:165— 170. FRENCH, J. R. P., HI & D. W. SCHLOESsER. 1996. Distribution and winter survival health of Asian clams, Corbicula fluminea, in the St. Clair River, Michigan. Journal of Freshwater Ecol- ogy 11(2):183-192. GRANEY, R. L., D. S. CHERRY, J. H. RODGERS, JR. & J. CAIRNS, Jr. 1980. The influence of thermal discharges and substrate composition on the population structure and distribution of the Asian clam, Corbicula fluminea, in the New River, Vir- ginia. The Nautilus 94(4):130—135. HEBERT, P. D. N., B. W. MUNCASTER & G. L. MACKIE. 1989. Ecological and genetic studies on Dreissena polymorpha (Pallas): a new mollusc in the Great Lakes. Canadian Journal of Fisheries and Aquatic Science 46:1587—1591. Kat, P. W. 1982. Shell dissolution as a significant cause of mor- tality for Corbicula fluminea (Bivalvia: Corbiculidae) inhab- iting acidic waters. Malacological Review 15:129—134. KREISER, B. R. & J. B. Mitton. 1995. The evolution of cold tolerance in Corbicula fluminea (Bivalvia: Corbiculidae). The Nautilus 109(4):111—112. Matrtice, J. S. & L. L. Dye. 1976. Thermal tolerance of the adult Asiatic clam. Pp. 130-135 in G. W. Esch & R. W. Mc- Farlane (eds.), Thermal Ecology. Vol. 2. U.S. Energy and Development Administration Symposium, Washington, D.C. McMahon, R. F. 1983. Ecology of an invasive pest bivalve, Cor- bicula. Pp. 505-561 in W. D. Russell-Hunter (ed.), The Mol- lusca. Vol. 6. Ecology. Academic Press: London. NALepPA, T. EF & D. W. SCHLOESSER. 1993. Zebra Mussels: Biol- ogy, Impacts, and Control. Lewis Publishers: Florida. 810 Pp. NELSON, S. M. & C. MCNABB. 1994. New record of Asian clam in Colorado. Journal of Freshwater Ecology 9(1):79. Peyton, M. M. & J. L. MAHER. 1992. Bivalves in central Ne- braska irrigation systems and associated reaches of the Platte River. Report of the U.S. Fish and Wildlife Service: Ne- braska. 11 pp. Peyton, M. M. & J. L. MAHER. 1995. A survey of mussels in the Platte River system and associated irrigation and hydro- power canal and lake systems west of Overton, Nebraska. Report of the U.S. Fish and Wildlife Service: Nebraska. 12 Pp: ScoTT-WasILK, J., G. G. Downinc & J. S. LieTzZow. 1983. Oc- currence of the Asian clam Corbicula fluminea in the Mau- mee River and western Lake Erie. Journal of Great Lakes Research 9(1):9-13. SICKEL, J. B. 1986. Corbicula population mortalities: factors in- fluencing population control. American Malacological Bul- letin, Special Edition (2):7-39. Cytotaxonomic Verification of a Non-Indigenous Marine Mussel in the Gulf of Mexico Brenden S. Holland', Daniel S. Gallagher’, David W. Hicks’ and Scott K. Davis? ‘Department of Oceanography, Texas A&M University, College Station, Texas 77843, USA *Department of Animal Science, Texas A&M University, College Station, Texas 77843, USA ’Department of Biology, Box 19498, University of Texas, Arlington, Texas 76019, USA In 1990, invasive populations of the genus Perna of un- known origin(s) were reported from both the Caribbean and Gulf of Mexico; the mussel discovered in the Gulf of Mexico was identified as P. perna (Linnaeus, 1758) (Hicks & Tunnell, 1993), and that from the Caribbean as P. viridis (Linnaeus, 1758) (Agard et al., 1993). Although the identification of P. viridis in the Caribbean has been confirmed by protein electrophoresis (Agard et al., 1993), until now identification of P. perna in the Gulf was based Notes, Information & News solely on morphology. It has been suggested that the Gulf mussel may be a brown morph of P. viridis, and that the two introductions were related. It is difficult to reliably distinguish between P. perna and P. viridis due to their morphological similarity and considerable variation in taxonomically important mor- phological characters (Siddall, 1980; Sadacharan, 1982). With the possible exception of their sympatric occurrence in Sri Lanka and southern India, the mutually exclusive geographic distribution of P. perna and P. viridis has his- torically been of taxonomic value (Siddall, 1980; Agard et al., 1993). In situations where the geographic origin is unknown, as is usually the case with biological introduc- tions, the high degree of phenotypic plasticity within this genus becomes more problematic, and alternative means of identification, such as cytogenetic techniques, are re- quired (Beaumont et al., 1992; Insua et al., 1994). In order to confirm the initial identification of the spe- cies in the Gulf of Mexico as P. perna, other, non-mor- phological characters were required. Cytological charac- ters were selected based on their unambiguous nature and the known karyotypic dimorphism of the two taxa in question. P. viridis has 30 chromosomes and P. perna, 28 (Ahmed, 1974; Jacobi et al., 1990). In this paper we report the karyotype of the non-indigenous mytilid in the Gulf of Mexico for the first time, verifying its identity as Perna perna. Materials and Methods Specimens were collected from the north jetty at Mans- field Pass (26°34'N, 97°17'W), Padre Island National Sea- shore, Texas. Mussels were transported alive to the lab- oratory, then maintained in aerated tanks of seawater. Chromosome preparations were obtained from gill tissue using a modified colchicine-air drying Giemsa technique (Yaseen, 1996). Mussels were incubated in aerated seawater in the pres- ence of 0.05% colchicine at room temperature for 9 hours and the gill tissues dissected and placed in distilled water at room temperature for 30 minutes. Distilled water was replaced with a 0.8% sodium citrate hypotonic solution for 30 minutes. Tissues were then submerged in 50 mL tubes of fresh fixative (3 parts methanol:1 part acetic acid), changed two times for 20 minutes each, and then left in the fixative for 8 hours. About 25 mg of gill tissue were then minced and dis- aggregated in a 70% glacial acetic acid solution to pre- pare a cell suspension. Disaggregated tissues were trans- ferred via pipette back into fresh fixative, from which single drops of the cell suspension were allowed to fall onto glass slides placed at an angle of about 45°, from a height of 15 cm. Cell suspensions on slides were thor- oughly air dried and stained with 5% Giemsa (diluted in phosphate buffer, pH 6.8) for 5 minutes. Black and white Page 281 AC KA KR AR AG AR WRAA AK wa 8O AA BA Aa Figure | Micrograph of metaphase spread (top) and standard karyotype (bottom) of Perna perna (2n = 28) from the Gulf of Mexico. The 14 pairs of autosomes are diagnostic for P. perna, shown grouped by general morphology into two subsets of metacentric and submetacentric (A) and subtelocentric (B). photomicrographs of well-spread mitotic metaphases were taken. Results and Discussion The diploid number of P. perna is 28 (n = 14), whereas that of P. viridis is 30 (n = 15) (Ahmed, 1974; Jacobi et al., 1990). A diploid number of 2n = 28 was scored in twelve mitotic metaphases observed in six individuals of Perna from the northwestern Gulf of Mexico. Chromo- somes were observed to vary from metacentric to sub- telocentric, consistent with chromosome morphology pre- sented by Jacobi et al. (1990) for P. perna. The standard karyotype of the Gulf of Mexico mussel consists of two groups based on overall chromosome morphology; one of 10 metacentric and submetacentric (A), and one of four subtelocentric chromosome pairs (B) (Figure 1). Group- ings were based on qualitative observations. Thus, our observations confirm the original taxonomic identification and verify that the invader is P. perna. Furthermore, based on the largely allopatric natural distributions of P. perna and P. viridis, this result strongly suggests that there were two separate biological introductions involv- The Veliger, Vol. 42, No. 3 ing mussels of the genus Perna in the Western Atlantic Ocean in 1990. Acknowledgments We would like to thank Marcos De Donato for technical assistance in the laboratory, and Yvette Barrios for assis- tance in the field. D. W. Hicks is supported by a grant from the Texas A&M Sea Grant College Program (NAS56RG0388). Laboratory aspects of this study were supported by the TAMU Research Foundation. Literature Cited AGARD, J., R. KISHORE & B. BAYNE. 1993. Perna viridis (Lin- naeus, 1758): first record of the Indo-Pacific green mussel (Mollusca: Bivalvia) in the Caribbean. Caribbean Marine Studies, Vol. 3:59—60. AHMED, M. 1974. Chromosomes of two species of the marine mussel Perna (Mytilidae: Pelecypoda) Boletin del Instituto Oceanografico Universidad de Oriente. 13(1—2):17—22. BEAUMONT, A. R., R. SEED & P. GARCIA-MARTINEZ. 1992. Elec- trophoretic and morphometric criteria for the identification of the mussels Mytilus edulis and M. galloprovincialis. Pp. 251-258 J. S. Ryland & P. A. Tyler (eds.), Reproduction, Genetics and Distributions of Marine Organisms. 23rd Eu- ropean Marine Biology Symposium. Olsen & Olsen Fre- densborg. Hicks, D. W. & J. W. TUNNELL, JR. 1993. Invasion of the south Texas coast by the edible brown mussel Perna perna (Lin- naeus, 1758). The Veliger 36(1):92—94. Insua, A., J. PR. LABAT & C. THIRIOT-QUIEVREUX. 1994. Compar- ative analysis of karyotypes and nucleolar organizer regions in different populations of Myrilus trossulus, Mytilus edulis and Mytilus galloprovincialis. Journal of Molluscan Studies 60:359—370. Jacosi, C. M., C. ROSENBERG & A. M. VIANNA-MorRGANTE. 1990. The karyotype of the brown mussel, Perna perna (L.) (Biv- alvia: Mytilidae). Revista Brasileira de Genetica 30(4):669— 673. SADACHARAN, D. H. 1982. Country report: Sri Lanka in F B. Davy & M. Graham, (eds.), Bivalve Culture in Asia and the Pacific: Proceedings of a workshop Held in Singapore, 16— 19 February, 1982. International Research Development Centre, Ottawa, Canada. SIDDALL, S. E. 1980. A clarification of the genus Perna (Mytil- idae). Bulletin of Marine Science 30(4):858—870. YASEEN, A. E. 1996. The chromosomes of the Egyptian fresh- water snail Melanoides tuberculata (Gastropoda: Prosobran- chia). Journal Molluscan Studies 62:137—141. Rediscovery of the Introduced, Non-Indigenous Bivalve Laternula marilina (Reeve, 1860) (Laternulidae) in the Northeastern Pacific Todd W. Miller'*, Eugene V. Coan? and John W. Chapman! 'Hatfield Marine Science Center, Department of Fisheries and Wildlife, Oregon State University, Newport, Oregon 97365-5296, USA "Department of Invertebrate Zoology, California Academy of Sciences, Golden Gate Park, San Francisco, California 94118-4599, USA The non-indigenous Laternula (Exolaternula) marilina (Reeve, 1860) (Bivalvia: Laternulidae) has been redis- covered in the northeast Pacific in Humboldt Bay, Cali- fornia (40°49’'N, 124°14’W). The first and only previous records of this species from the northeast Pacific are from Coos Bay, Oregon (43°25'N, 124°27’W) where two spec- imens were recovered from high intertidal mud flats in 1963, and a few specimens were recovered from marsh channels in 1966 (Keen, 1969). A photograph of these specimens, identified as Laternula limicola, is in the Cal- ifornia Academy of Sciences (CAS) collections. How- ever, the actual specimens cannot be found. No further records of Laternula from Coos Bay have been made since 1966 even though several systematic intertidal sur- veys of introduced species of Coos Bay were conducted between 1986 and 1997 (James T. Carlton, personal com- munication). Laternula were collected from Humboldt Bay from sa- linities ranging between 28 and 34 psu at four out of 47 high intertidal stations sampled between June 1995 and December 1996 (Figure 1). Both live and empty shelled specimens were found. All live specimens but one (sta- tion 20, Figure 1) were recovered from northeast Hum- boldt Bay. Laternula marilina may be commonly misidentified or overlooked in northeast Pacific bivalve collections due to its omission from major taxonomic references to eastern Pacific marine invertebrates and confusion with other spe- cies. Keen & Coan (1974) include Laternula in their key to marine molluscan genera of western North America. However, Laternula is not included in the bivalve keys of either of the general guides to invertebrates of the northeast Pacific (Smith & Carlton, 1975, and Kozloff, 1997), although it is cited (as L. limicola), but not illus- trated, in the later reference. Laternula limicola (Reeve, 1863) of China is a synonym of Laternula marilina (Zhuang & Cai, 1982). This rediscovery of Laternula was too late for inclusion in the recent checklist of marine bivalves of the northeast Pacific, where it was excluded because it was thought to no longer be present (Coan & Scott, 1997). * To whom all correspondence should be addressed. Notes, Information & News Humboldt Bay, California Figure | Humboldt Bay, California collection sites of the 1995—1996 in- tertidal benthic survey of macro-invertebrates with sites where Laternula were found indicated by crosshatched circles and scaled to the estimated population densities m~’. All presently known northeast Pacific Laternula mari- lina are small relative to Asian populations which attain lengths of 55 mm or more (Zhuang & Cai, 1982). The longest shell of any specimen from Humboldt Bay is 11.6 mm (Table 1), and the maximum length of the original specimens from Coos Bay is approximately 20 mm (Keen, 1969). The illustrations of Zhuang & Cai (1982) indicate a broader, more truncate posterior than occurs in the specimens from Humboldt Bay (Figure 2B). This var- iation appears to be entirely related to differences in size. Laternula marilina has a chondrophore (Figure 2A) and a lithodesma on the ventral surface of its ligament. The external appearances of small individuals of this spe- cies resemble, and could be confused with, the common native clam Cryptomya californica (Conrad, 1837) and the introduced, non-indigenous clam Mya arenaria Lin- naeus, 1758, which occur in similar habitats. Laternula s.s., lack a lithodesma. The presence of a lithodesma in Laternula marilina places it in the subgenus Laternula (Exolaternula). Low densities and patchy local distributions of Later- Page 283 Table 1 Sample sites, numbers collected, estimated densities, and the averages and ranges of valve lengths, measured to +0.1 mm, of live Laternula (Exolaternula) limicola (Reeve, 1860) collected in the Humboldt Bay, California macro-invertebrate survey. Average Sample Density* length Range of site Number (m~?) (mm) lengths (mm) 6 24 160.0 5.6 1.5-11.6 i 1 6.7 8.8 8.8 8 7 46.7 7.8 4.5—7.2 20 6.7 6.5 6.5 “ Density estimated from the remains of 10 4.4 cm diameter by 10 cm deep cores collected at each site and washed on a 0.5 mm mesh sieve. nula populations in estuaries may account for the absence of recent records from Coos Bay and previous records from Humboldt Bay and other northeast Pacific estuaries, including San Francisco Bay. The largest previous macro- invertebrate survey of Humboldt Bay (Barnhart et al., 1992) included samples within channels and at the mouth of the bay, well away from where Laternula were recov- ered in this survey. The small sizes, extremely patchy distribution, and possible restriction of Laternula to high intertidal mud flats, where systematic surveys are seldom conducted, could make this species difficult to find even Figure 2 Hinge (A) and left valve (at slightly different angle) (B), with condrophore, of an 8:5 mm length Laternula (Exolaternula) lim- icola (Reeve, 1860) from station 6, Humboldt Bay, California, California Academy of Sciences, Invertebrate Zoology, accession no. 50388, catalogue no. 111415. in San Francisco Bay, where introduced species are rel- atively well known (Cohen & Carlton, 1995). The variable northeast Pacific records of Laternula over time and among estuaries could also result from re- peated local introductions and extinctions (species turn- over). The small sizes of the northeast Pacific individuals relative to sizes reported from Asian populations may re- flect relatively short life spans. Humboldt Bay and Coos Bay have numerous mechanisms for domestic introduc- tion of Laternula from previously introduced but undis- covered northeast Pacific populations, including larvae carried in coastal currents, or in ballast water from com- mercial shipping (Carlton & Geller, 1995). Adult Later- nula could also be transplanted to Humboldt Bay with domestic oyster transplants (Monroe et al., 1973; Barn- hart et al., 1992) or with internationally transplanted Jap- anese oyster spat (Woelke, 1955) directly from its native range (central to northern Japan [Reeve, 1860] and China [Zhuang & Cai, 1982]). International ballast water traffic (Carlton & Geller, 1995) may also have carried the larvae to Humboldt Bay from Asia. These transport mechanisms could repeatedly carry in new populations which quickly decline to extinction in most areas and most instances. Such sporadic introductions would be difficult to detect. Resolution of these problems will require better defini- tions of the distribution and abundance of this species within and among the northeast Pacific estuaries over time. Acknowledgments We thank J. T. Carlton, Williams College, for his personal communications and assistance in publishing this discov- ery; Andy Cohen, San Francisco Estuary Institute; and an anonymous reviewer for helpful comments on a pre- vious draft of the manuscript. Patrick Clinton, OEO Cor- poration, U.S. EPA Research Laboratory, Newport, Oregon, refined and produced a final copy of the map originally produced for this project by Darien LaBrae of Humboldt State University. We thank the U.S. EPA Re- search Laboratory, Newport, Oregon, for space and lab facilities provided to JWC as a visiting scientist during part of this research. Literature Cited BARNHARDT, R. A., M. J. Boyp & J. E. PEQUEGNAT. 1992. The Ecology of Humboldt Bay, California: an Estuarine Profile. U.S. Fish and Wildlife Service Biological Report 1. 121 pp. CarLTon, J. T. & J. B. GELLER. 1995. Ecological roulette: the global transport of nonindigenous marine organisms. Sci- ence 261:78—82. COHEN, A. N. & J. T. CarLTon. 1995. Nonindigenous aquatic species in a United States estuary: a case study of the bio- logical invasion of the San Francisco Bay and delta. U.S. Fish and Wildlife Service, Washington, D.C. Coan, E. V. & P. H. Scott. 1997. Checklist of the Marine Bi- valves of the Northeastern Pacific Ocean. Santa Barbara Mu- The Veliger, Vol. 42, No. 3 seum of Natural History Contributions in Science No. 1. 28 Pp- KEEN, M. A. 1969. Laternula living on the Pacific coast? The Veliger 11(4):439. KEEN, M. A. & E. CoANn. 1974. Marine Molluscan Genera of Western North America: An Illustrated Key. 2nd ed., Stan- ford University Press: Stanford, California. 208 pp. Koz.orr, E. N. 1997. Marine Invertebrates of the Pacific North- west. University of Washington Press: Seattle. 539 pp. Monrog, G. W., S. I. THompson, P. G. SWARTZELL, B. M. BROWN- ING, J. W. SpETH & R. G. ARneTT. 1973. The Natural Re- sources of Humboldt Bay. California Department of Fish and Game Coastal Wetlands Series No. 6. 160 pp. REEVE, L. A. 1860-1863. Monograph of the genus Anatina. In: L. A. Reeve, (ed.), Conchologia ionica; or, Illustrations of the Shells of Molluscous Animals 14: 4 pls. [pl. 2 Dec. 1860; pls. 1, 3—4, Feb. 1863]. Situ, R. I. & J. T. CARLTON. 1975. Light’s Manual: Intertidal Invertebrates of the Central California Coast. 3rd ed. Uni- versity of California Press: Berkeley, California. 716 pp. WOELKE, C. E. 1955. Introduction of the Kumamoto oyster Os- trea (Crassostrea) giganteus to the Pacific coast. Fisheries Research Papers of the Washington Department of Fisheries 1(3):1-10. ZHUANG, Q. & Y. CAI. 1982. Studies of the Laternulidae off the Chinese coast. Oceanologia et Limnologia Sinica 13(6):553— 561. Chemoattraction of Lymnaea elodes (Gastropoda: Lymnaeidae) to Leaf Lettuce and Tetramin Jason T. Fedok, Bernard Fried* and Aditya Reddy Department of Biology, Lafayette College, Easton, Pennsylvania 18042, USA Introduction Several studies have examined chemoattraction and die- tary preferences of planorbid snails for Tetramin® (fish food) and leaf lettuce (Masterson & Fried, 1992; Mar- copoulos & Fried, 1993; Mancia & Fried, 1995). How- ever, studies using lymnaeid snails are not available. Sor- ensen et al. (1997) identified Lymnaea elodes (Say, 1821) (L. elodes = L. palustris = Stagnicola palustris = S. elo- des) as a vector of the ubiquitous trematode Echinostoma revolutum (Froelich, 1802) in the USA. This species of lymnaeid is easy to maintain in the laboratory and achieves a length of 2 cm within 2 months. Because of its ease of maintanance, its relatively large size, and its role as a vector for an economically important trematode, it is a useful model for laboratory work. This study examines chemoattraction of L. elodes to iceberg leaf lettuce (Lactuca sativa) and to Tetramin and * To whom correspondence should be addressed: telephone: (610) 250-5463; FAX: (610) 250-6557; e-mail: friedb@lafayette.edu Notes, Information & News Table 1 Chemoattraction of Lymnaea elodes snails to leaf lettuce and Tetramin. Percentage of snails in zones Experi- je ae ee SNe ment Choices A B (E square P 1 BI vs Bl 37 24 39 — — p Bl vs L 29 17 54 9.54 0.008 3 Bl vs T 16 13 71 43.22 0.001 4 Lvs T 24 15 61 20.35 0.016 Bl = blank; L = lettuce; T = Tetramin; each experiment was done 10 times; lettuce placed in zone C in Experiment 2 and zone A in Experiment 4; Tetramin placed in zone C in Experi- ments 3 and 4. compares the results with our previous studies on pla- norbid snails. Materials and Methods Stock cultures of Lymnaea elodes were maintained in the laboratory on a diet of iceberg leaf lettuce occasionally supplemented with Tetramin (TetraWerke, Melle, Ger- many) as described by Frazer et al. (1997). The bioassay chamber for the study consisted of 100 < 15 mm petri dishes (Masterson & Fried, 1992). Two parallel lines were drawn 2.8 cm apart on the bottom of each petri dish to produce three zones (A, B, C). The side zones (A and C) each had an area of 14.1 cm?*, and the middle zone (B) had an area of 20.0 cm? (Masterson & Fried, 1992). Fifty- five mL of artificial spring water (ASW; Ulmer, 1970) was added to each dish. The snails were placed into the petri dish 10 min after the food or controls to allow ad- equate diffusion of food throughout the water. The dishes were maintained at 22 + 1°C on a level work bench under overhead diffuse fluorescent light (Marcopoulos & Fried, 1993). Ten petri dishes were used for each trial. Snails of approximately 20 mm shell length were main- tained without food for approximately 3 hr prior to each trial. For each trial, a single snail was placed in the center of zone B in the bioassay chamber described above. Filter paper squares (1 cm*) were used as controls and placed at the edges of zones A and C. They were kept in place with paper clips. In the experimental trials, food items (also about 1 cm’) were used in place of the filter paper squares. Experimental trials matched lettuce versus blanks, Tetramin versus blanks, or lettuce versus Tetramin (see Table 1). For 50 min, at intervals of 5 min, the zone in which the snail was located was recorded (a total of 10 obser- vations per snail). A control group was used in each trial, as described above, to determine random movements of the snails. Random observations based on blank (control) experiments were used as the expected values to calculate Page 285 the chi-square value in each experimental design. A P value of less than 0.05 was considered significant (Man- cia & Fried, 1995). Each experiment was done 10 times for every group. Results and Discussion The results of the chemoattraction trials are presented in Table 1. In trials that matched a food item against a con- trol, snails were significantly attracted to either the lettuce or Tetramin (Experiments 2 and 3). In trials that matched lettuce against Tetramin (Experiment 4), snails were sig- nificantly attracted to the Tetramin. Results of this study are in accord with the earlier find- ings on Biomphalaria glabrata (Say, 1818) (see Master- son & Fried, 1992) and Helisoma trivolvis (Say, 1816) (see Mancia & Fried, 1995) which showed significant chemoattraction to either lettuce or Tetramin in the pres- ence of a blank control. The finding of significant che- moattraction of Lymnaea elodes to Tetramin rather than lettuce in our bioassay is in accord with that of Mancia & Fried (1995) on H. trivolvis, but differs from that of Masterson & Fried (1992) where B. glabrata showed no significant difference in chemoattraction between lettuce and Tetramin in the same bioassay. Unpublished obser- vations by one of us (B.E) suggest that growth and fe- cundity of L. elodes snails in the laboratory are subopti- mal when maintained on Tetramin compared to those reared on leaf lettuce. Thus, snail chemoattraction to a food source is not an indicator that such a food item is optimal for growth and fecundity of that gastropod. Literature Cited FRAZER, B. A., A. REDDY, B. FRIED & J. SHERMA. 1997. HPTLC determination of neutral lipids and phospholipids in Lym- naea elodes (Gastropoda). Journal of Planar Chromatogra- phy 10:128—130. MancliA, M. R. & B. FRieED. 1995. Chemoattraction and dietary preferences of Helisoma trivolvis (Gastropoda: Planorbidae) for leaf lettuce and Tetramin. The Veliger 38:73—74. Marcopou.os, A. & B. FRIED. 1993. Chemoattraction of Biom- phalaria glabrata (Gastropoda: Planorbidae) for lipid stan- dards and lipophilic factors in leaf lettuce and Tetramin. Journal of Chemical Ecology 19:2593—2597. MASTERSON, C. & B. FRIED. 1992. Chemoattraction and dietary preferences of Biomphalaria glabrata (Gastropoda: Planor- bidae) for leaf lettuce, Tetramin, and hen’s egg yolk. Com- parative Biochemistry and Physiology 103A:597—599. SORENSEN, R. E., I. KANEv, B. FRIED & D. J. MINCHELLA. The occurrence and identification of Echinostoma revolutum from North American Lymnaea elodes snails. Journal of Par- asitology 83:169—170. ULMER, M. J. 1970. Notes on rearing snails in the laboratory. Pp. 143-144 in A. J. MacInnis & M. Voge (eds.), Experiments and Techniques in Parasitology. W. H. Freeman: San Fran- cisco. Page 286 Lindeman Lake, British Columbia, Type Locality of Zonitoides randolphi Pilsbry Robert G. Forsyth 2574 Graham Street, Victoria, British Columbia, Canada V8T 3Y7 Records of non-marine mollusks in western Canada are few but much repeated throughout the literature. Such is the case with Lindeman (or erroneously “‘Linderman’’) Lake, which is the northernmost locality (Bequaert & Miller, 1973) of Discus shimekii (Pilsbry, 1890) and the type locality of Zonitoides randolphi Pilsbry, 1898, a ju- nior subjective synonym of D. shimekii according to Pils- bry (1948). Additionally, Lindeman Lake is cited else- where by Dall (1905), Baker (1911), Pilsbry (1948), and Clench & Turner (1962) in connection with this and other species of terrestrial and aquatic mollusks. A series of locality placements which were either imprecise or clear- ly wrong has most recently moved Lindeman Lake to the Yukon Territory (Bequaert & Miller, 1973). The first appearance in the malacological literature of Lindeman Lake, as “‘Linderman Lake, Alaska,” dates from the description of Z. randolphi by Pilsbry (1898); he may have assumed that the locality was in Alaska, or perhaps had been misinformed otherwise. The collector of the new species was P. B. Randolph of Seattle who published (1899) a brief popular account of his travels to the Klondike in 1897-1898. Randolph traveled north up the coast by ship and overland from Dyea, Alaska, via the Chilkoot Pass to the Yukon. Lindeman Lake was a stop en route on the Canadian side of the Chilkoot Pass; Randolph (1899:109) wrote: We laid over one day at Lake Linderman [sic], resting from the past week’s hard work, and I had time to hunt over the flat at the head of the lake where a small stream empties in. Z. randolphi was among the species of terrestrial mol- lusks collected at Lindeman. However, nowhere in Ran- dolph’s account is the location of the lake ever stated. Dall (1905:43) was the first to publish a correction to the earlier errors of Pilsbry and gave the locality as ‘“‘Lake Lindeman, headwaters of the Yukon, British America.” Most subsequent authors, including Pilsbry (1948), fol- lowed Dall who was essentially correct. (Canada was “British America” at that time.) However, more recently Bequaert & Miller (1973:57) placed Lindeman Lake in the Yukon Territory, ‘‘at the head of the Yukon River, ca. 64°30'N, 140°50'W,”’ perhaps not realizing that the head- waters of the river system are in northwest British Co- lumbia. The coordinates given by Bequaert & Miller are clearly erroneous. Thus, the type locality of Z. randolphi and all other references to the locality should be corrected to read Lin- deman Lake, British Columbia, Canada. The terminus of the Chilkoot Trail at the south end of Lindeman Lake is at ca. 59°47'N, 135°05'W (Energy, Mines and Resources The Veliger, Vol. 42, No. 3 Canada, 1984). The small stream mentioned by Randolph (1899) could either be one of the branches of Lindeman Creek or a smaller, unnamed creek to the east. Literature Cited BAKER, FC. 1911. Lymnaeidae of North and Middle America: Recent and Fossil. Chicago Academy of Science, Special Publication, 3. xvi + 539 pp., 58 pls. BEQUAERT, J. C. & W. B. MILLER. 1973. The Mollusks of the Arid Southwest: with an Arizona Check List. University of Arizona Press: Tuscon. xvi + 271 pp. CLENCH, W. J. & R. D. TURNER. 1962. New Names Introduced by H. A. Pilsbry in the Mollusca and Crustacea. Academy of Natural Sciences of Philadelphia, Special Publication 4. v + 166 pp. Da.t_, W. H. 1905. Land and Fresh Water Mollusks. Harriman Alaska Expedition, 8. xii + 171, pls. I-II. ENERGY, MINES AND RESOURCES CANADA. 1984. Homan Lake. Edition 3. National Topographic Series 104 M/14. 1:50,000. PitsBry, H. A. 1898. Descriptions of new species and varieties of American Zonitidae and Endodontidae. The Nautilus 12(8):86-87. PitsBry, H. A. 1948. Land Mollusca of North America (North of Mexico). Academy of Natural Sciences of Philadelphia, Mongraph 3, 2(2):i-xlvii, 521-1113. RANDOLPH, P. B. 1899. Collecting shells in the Klondike Country. The Nautilus 12(10):109—112. A New Species of Gastrocopta (Gastropoda: Pulmonata: Pupillidae) from the Deep River Formation, Late Oligocene or Early Miocene, Montana Barry Roth Department of Invertebrate Zoology, Santa Barbara Museum of Natural History, Santa Barbara, California 93105, USA Roth & Emberton (1994) described an assemblage of land snail fossils from the Deep River Formation, continental deposits ranging from early Oligocene (Chadronian) to middle Miocene (Barstovian) age (Rensberger, 1981; Runkel, 1986), exposed in isolated outcrops in the Smith River Basin between White Sulphur Springs and Fort Lo- gan, Meagher County, Montana. Based on the climatic signatures of extant genera, Roth & Emberton (1994) in- ferred a mesic climate with at least 75 cm/yr precipita- tion. In that paper, the present species was identified as Gastrocopta sp., aff. G. armifera (Say, 1821). Additional study of the material permits its description here as a new species. Notes, Information & News 2 =D Figures 1 and 2 Gastrocopta abyssifluminis Roth, sp. nov. Holotype, SBMNH 110599. Apertural and lateral views. Height 3.57 mm. PUPILLIDAE Turton, 1831 Gastrocopta Wollaston, 1878 Type-species: Pupa acarus Benson, 1856; subsequent designation by Pilsbry (1916). Gastrocopta abyssifluminis Roth, sp. nov. (Figures 1, 2) Gastrocopta sp., aff. G. armifera (Say), Roth & Emberton, 1994: 94. Diagnosis: A large, broadly ovate Gastrocopta with 5.8— 6.0 flattened whorls; suture appressed; base umbilicate, produced and compressed; inner end of angulo-parietal lamella curving toward periphery. Description: Shell broadly ovate, widest above middle of body whorl; apical angle approximately 90°; base narrow- ly umbilicate, somewhat produced, tapering and com- pressed. Whorls 5.8 to 6.0 at maturity, with inconspicu- ous, raised, retractive growth lines; early whorls moder- ately convex; later whorls more flattened; suture ap- pressed. Body whorl not strongly constricted behind aperture; crest absent. Aperture roughly triangular, acute at anterior end; parietal callus effuse, extending well onto face of body whorl. Strong angulo-parietal lamella pre- sent, inner end curving toward periphery; palatal and col- umellar lamellae not detected. Dimensions: Holotype, height 3.57 mm; diameter 2.24 mm; height of body whorl 2.01 mm; whorls 6.75. Para- types, height 3.15—3.99 mm (mean 3.63; n = 11); diam- eter 1.90—2.51 mm (mean 2.35; n = 12); height:diameter Page 287 ratio 1.34—1.80 (mean 1.55; n = 11); whorls 5.8—6.0 (mean 5.94, n = 8). Type material: Holotype, Santa Barbara Museum of Natural History, SBMNH 110599, MONTANA: Meagher County: approximately 19 km northwest of White Sul- phur Springs, 0.4—0.8 km east of White Sulphur Springs- Fort Logan road, in steep, bare north wall of small, me- andering gully tributary to Rabbit Creek; sec. 14, T. 10 N, R. 5 E, Hanson Reservoir Quadrangle (USGS 7.5 Min- ute Series, Topographic, ed. 1971). Deep River Forma- tion, late Oligocene or early Miocene. S. Stillman Berry et al. coll. 24 August 1941. Paratypes (all from same locality as holotype): SBMNH 110298 (2 specimens), A. C. Silberling coll. 21 October 1940. SBMNH 110299 (10 specimens), A. C. Silberling coll. 21 October 1940, and S. Stillman Berry et al. coll. 24 August 1941; SBMNH 111989 (2 speci- mens), collector not stated, 28 August 1954. Referred material: In addition to the type material, 15 specimens from the Berry collection are not designated as types because they are poorly preserved or imperfectly labeled as to locality. Remarks: The type locality of Gastrocopta abyssiflu- minis is the same as that of Euchemotrema occidaneum Roth & Emberton, 1994, and Hendersonia stillmani Roth & Emberton, 1994, and probably equivalent to the Spring Creek 1 locality of Rensberger (1981). The presence of Pupoides montana Pierce, 1992, in the molluscan assem- blage from this locality (Roth & Emberton, 1994) sug- gests correlation with the Cabbage Patch Beds in western Montana (Pierce & Rasmussen, 1992), of Arikareean age. Gastrocopta abyssifluminis is not among the Pupillidae reported from the Cabbage Patch Beds: it is substantially larger than G. obesa, G. oviforma, G. tavennerensis, G. leonardi, and G. minuscula (all, Pierce in Pierce & Ras- mussen, 1992) and has less impressed sutures than any of them. It is relatively broader than G. russelli Pierce, 1992 (height:diameter ratio 1.34—1.80 compared to 1.64— 1.92 for G. russelli), with more tapering anterior end and flatter whorls; the inner end of the angulo-parietal lamella curves toward the periphery rather than toward the col- umella. Etymology: L., abyssus (deep, bottomless) + flumen, flu- minis (stream): of the Deep River Formation. Literature Cited Pierce, H. G. & D. L. Rasmussen. 1992. The nonmarine mol- lusks of the late Oligocene-early Miocene Cabbage Patch fauna of western Montana. I. Geologic setting and the family Pupillidae (Pulmonata: Stylommatophora). Journal of Pale- ontology 66:39—52. Pitssry, H. A. 1916. Pupillidae (Gastrocoptinae) [part]. Manual of Conchology (2) 24 (93):1-112, pls. 1-13. RENSBERGER, J. M. 1981. Evolution in a late Oligocene-early Page 288 The Veliger, Vol. 42, No. 3 Miocene succession of meniscomyine rodents in the Deep River Formation, Montana. Journal of Vertebrate Zoology 1: 185-209. Rot, B. & K. C. EMBERTON. 1994. **Extralimital” land mollusks (Gastropoda) from the Deep River Formation, Montana: ev- idence for mesic medial Tertiary climate. Proceedings of the Academy of Natural Sciences of Philadelphia 145:93-106. RUNKEL, A. C. 1986. Geology and vertebrate paleontology of the Smith River Basin, Montana. Unpublished master’s thesis, Department of Geology, University of Montana. Information for Contributors Manuscripts Manuscripts must be typed, one side only, on A4 or equivalent (e.g., 8%” X 11”) white paper, and double-spaced throughout, including references, figure legends, footnotes, and tables. All margins should be at least 25 mm wide. Text should be ragged right (i-e., not full justified). Avoid hyphenating words at the right margin. Manuscripts, in- cluding 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, in- cluding the year, if possible. Underline scientific names and other words to be printed in italics; no other manipulation of type faces is necessary on the manuscript. Metric and Celsius units are to be used. For aspects of style not ad- dressed here, please see a recent issue of the journal. The Veliger publishes in English only. Authors whose first language is not English should seek the assistance of a col- league who is fluent in English before submitting a manu- script. In most cases, the parts of a manuscript should be as follows: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, literature cited, figure legends, footnotes, tables, and figures. The title page should be a separate sheet and should include the title, authors’ names, and addresses. The abstract should be less than 200 words long and should describe concisely the scope, main results, and conclusions of the paper. It should not include references. 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 (Phillips, 1981), for two authors (Phillips 8 Smith, 1982), and for more than two (Phillips et al., 1983). The reference need not be cited when author and date are given only as authority for a taxonomic name. The “literature cited” section should include all (and only) references cited in the text, listed in alphabetical order by author. Each citation must be complete, with all journal titles unabbreviated, and in the following forms: a) Periodicals: Hickman, C. S. 1992. Reproduction and development of trochacean gastropods. The Veliger 35:245—272. b) Books: Bequaert, J. C. & W. B. Miller. 1973. The Mollusks of the Arid Southwest. University of Arizona Press: Tuc- son. xvi + 271 pp. c) Composite works: Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135 in R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stanford Univer- sity Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Avoid vertical rules. Figures and plates Figures must be carefully prepared and submitted ready for publication. Each should have a short legend, listed on a sheet following the literature cited. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photo- graphs for halftone reproduction must be of good quality, trimmed squarely, grouped as appropriate, and mounted on suitably heavy board. Where appropriate, a scale bar may be used in the photograph; otherwise, the specimen size should be given in the figure legend. Photographs should be submitted in the desired final size. Clear xerographic copies of figures are suitable for re- viewers’ copies of submitted manuscripts. It is the author’s responsibility to ensure that lettering will be legible after any necessary reduction and that lettering size is appropriate to the figure. Use one consecutive set of Arabic numbers for all illus- trations (that is, do not separate “plates” from “text fig- ures’). Processing of manuscripts Each manuscript is critically evaluated by at least two reviewers. Based on these evaluations the editor makes a preliminary decision of acceptance or rejection. The editor’s decision and the reviewers’ comments are sent to the author for consideration and further action. Unless requested, only one copy of the final, revised manuscript needs to be re- turned to the editor. The author is informed of the final decision and acceptable manuscripts are forwarded to the printer. The author will receive proofs from the printer. One set of corrected proofs should be mailed promptly to the editor after review. Changes other than the correction of printing errors will be charged to the author at cost. An order form for the purchase of reprints will accom- pany proofs. Reprints are ordered directly from the printer. Authors’ contributions The high costs of publication require that we ask authors for a contribution to defray a portion of the cost of pub- lishing their papers. However, we wish to avoid a handicap to younger contributors and others of limited means and without institutional support. Therefore, we have adopted the policy of asking for the following: $30 per printed page for authors with grant or other institutional support and $10 per page for authors who must pay from their personal funds (2.5 double-spaced manuscript pages normally equal one printed page). This request is made only after the pub- lication of a paper; these contributions are unrelated to the acceptance or rejection of a manuscript, which is entirely on the basis of merit. In addition to this requested contri- bution, authors of papers with an unusually large number of tables or figures will be asked for an additional contri- bution. Because these contributions by individual authors are voluntary, they may be considered by authors as tax- deductible donations to the California Malacozoo- logical Society, Inc., to the extent allowed by law. It should be noted that even at the rate of $30 per page, the CMS is paying well over half the publication costs of a paper. Authors for whom even the $10 per page contri- bution would present a financial hardship should explain this in a letter accompanying their manuscript. The edito- rial board will consider this an application for a grant to cover the publication costs. Authors whose manuscripts in- clude very large tables of numbers or extensive lists of (e.g.) locality data should contact the editor regarding possible electronic archiving of this part of their paper rather than hard-copy publication. Submitting manuscripts Send manuscripts, proofs, books for review, and corre- spondence on editorial matters to Dr. Barry Roth, Editor, 745 Cole Street, San Francisco, CA 94117, USA. CONTENTS — Continued NOTES, INFORMATION & NEWS Occurrence of the Asian clam Corbicula fluminea (Miller, 1774) (Bivalvia: Sphaer- iacea: Corbiculidae) in Colorado JAMES R: CORDEIRO AND SARAH MACWIELIAMS 22.5 2)2)-)¢ cles ei eee Cytotaxonomic verification of a non-indigenous marine mussel in the Gulf of Mexico BRENDEN S. HOLLAND, DANIEL S. GALLAGHER, DaviD W. HICKS, AND ScoTT K. DINO bon oo dU een oso Doub osS oud vaooobeob boon booKmond jus BPO 0100.6.0 6 Rediscovery of the introduced, non-indigenous bivalve Laternula marilina (Reeve, 1860) (Laternulidae) in the northeastern Pacific Topp W. MILLER, EUGENE V. COAN, AND JOHN W. CHAPMAN Chemoattraction of Lymnaea elodes (Gastropoda: Lymnaeidae) to leaf lettuce and Tetramin JASON T. FEDOK, BERNARD FRIED, AND ADITYA REDDY ......-....se0s2-5eoer Lindeman Lake, British Columbia, type locality of Zonitoides randolphi Pilsbry ROBERT (G. FORSYAIED: 6 5.00.55 dic /ahake: aialale acd Meal aioetes Salar eG ny Ose eeeeee A new species of Gastrocopta (Gastropoda: Pulmonata: Pupillidae) from the Deep River Formation, late Oligocene or early Miocene, Montana BARRY ROTH THE NG MeN VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 42 WL Ser VX October 1, 1999 CONTENTS Another look at the muricine genus Afttiliosa TESTES? “TELLS WAGIRIDS, Sessa N eters crea eer HEE rere UWA net ne tons gree (ee OPE ag SEEN naa A systematic review of the hydrobiid snails (Gastropoda: Rissooidea) of the Great Basin, western United States. Part II. Genera Colligyrus, Eremopyrgus, Flu- minicola, Pristinicola, and Tryonia IROBERGip bel RS EIB Rares y eee cen ora A Mee MaN oe AEM Yes sles tsslsin es ose abet yin iby soesueke Sav Seaeels Land caenogastropods of Mounts Mahermana, Ilapiry, and Vasiha, southeastern Madagascar, with conservation statuses of 17 species of Boucardicus KENNETH C. EMBERTON AND TIMOTHY A. PEARCE ............-0cccceeecees ISSN 0042-3211 The Veliger (ISSN 0042-3211) is published quarterly in January, April, July, and October by the California Malacozoological Society, Inc., % Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. Periodicals postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to The Veliger, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. Number 4 THE VELIGER Scope of the journal The Veliger is an international, peer-reviewed scientific quarterly published by the Cali- fornia Malacozoological Society, a non-profit educational organization. The Veliger is open to original papers pertaining to any problem connected with mollusks. Manuscripts are considered on the understanding that their contents have not appeared, or will not appear, elsewhere in substantially the same or abbreviated form. Holotypes of new species must be deposited in a recognized public museum, with catalogue numbers provided. Even for non- taxonomic papers, placement of voucher specimens in a museum is strongly encouraged and may be required. Very short papers, generally not over 750 words, will be published in a “Notes, Infor- mation & News” column; in this column will also appear notices of meetings and other items of interest to our members and subscribers. Editor-in-Chief Barry Roth, 745 Cole Street, San Francisco, CA 94117, USA e-mail: veliger@ucmp1.berkeley.edu Production Editor Leslie Roth, San Francisco Board of Directors Michael G. Kellogg, City and County of San Francisco (President) Hans Bertsch, National University, San Diego Henry W. Chaney, Santa Barbara Museum of Natural History Eugene V. Coan, California Academy of Sciences, San Francisco Terrence M. Gosliner, California Academy of Sciences, San Francisco Carole S. Hickman, University of California, Berkeley EG. Hochberg, Santa Barbara Museum of Natural History Matthew J. James, Sonoma State University David R. Lindberg, University of California, Berkeley James Nybakken, Moss Landing Marine Laboratories Peter U. Rodda, California Academy of Sciences, San Francisco Barry Roth, San Francisco Geerat J. Vermeij, University of California, Davis Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (not institutions) interested in any aspect of malacology. New members join the society by sub- scribing to The Veliger. Rates for Volume 42 are US $40.00 for affiliate members in North America (USA, Canada, and Mexico) and US $72.00 for libraries and other institutions. Rates to members outside of North America are US $50.00 and US $82.00 for libraries and other institutions. All rates include postage, by air to addresses outside of North America. Memberships and subscriptions are by Volume only and follow the calendar year, starting January 1. Payment should be made in advance, in US Dollars, using checks drawn from US banks or by international postal order. No credit cards are accepted. Payment should be made to The Veliger or “CMS, Inc.” and not the Santa Barbara Museum of Natural History. Single copies of an issue are US $25.00, postage included. A limited number of back issues are available. Send all business correspondence, including subscription orders, membership applications, pay- ments, and changes of address, to: The Veliger, Dr. Henry Chaney, Secretary, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105, USA. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. Barry Roth, Editor, 745 Cole Street, San Francisco, CA 94117, USA. © This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). The Veliger 42(4):289-—305 (October 1, 1999) THE VELIGER © CMS, Inc., 1999 Another Look at the Muricine Genus Aftiliosa EMILY H. VOKES Tulane University (Emerita), New Orleans, Louisiana 70118, USA This paper is fondly dedicated to the memory of the late Anthony D’ Attilio. Abstract. Prior to this study, the muricine genus Aftiliosa comprised 13 (or possibly 14) species, known from tropical waters around the world, and ranging in age from Oligocene to Recent. Of this number, four were known from the fossil record: two (or three) from Europe, and two from the New World. The present study adds seven new species. Of these, three are fossil forms: A. greftae, from the Early Miocene Chipola Formation, northwestern Florida; A. macgintyi, from the Pliocene Tamiami Formation, southern Florida; and A. gibsonsmithi, from the Late Miocene Mataruca Member, Caujaro Formation, Venezuela. The four Recent species are: A. bessei and A. kevani, from the Caribbean; A. perplexa, from Brazil; and A. houarti, from the eastern Indian Ocean (Thailand). INTRODUCTION Seventeen years ago a review of the genus Aftiliosa Emerson, 1968, noted that it “‘is a small muricine group with but a few species” (Vokes & D’ Attilio, 1982:67). In that review, we recognized a total of five or six living species. There was a single eastern Pacific species [A. nodulosa (Adams, 1855)], and two western Atlantic ones [A. aldridgei (Nowell-Usticke, 1969) and A. philippiana (Dall, 1889)]. In the Indo-Pacific area there were two [A. nodulifera (Sowerby, 1841) and A. orri (Cernohorsky, 1976)], and possibly a third [A. caledonica (Jousseaume, 1881)], which might or might not be a synonym of A. nodulifera. However, additional material has shown A. caledonica to be a valid species. At that time, the fossil record in the New World con- sisted of a few specimens of A. aldridgei from the Mio- Phocene Gurabo Formation, Dominican Republic, and a single example of A. nodulosa from the Early Pliocene Esmeraldas beds, Onzole Formation, of northwestern Ecuador. However, the group also has a long history in the Old World where there is an apparently unnamed species in the Oligocene Stampian of France that is markedly sim- ilar to A. attiliosa (compare Figure 1 and Figure 17, for example). There is a second French species of Middle Miocene Burdigalian age, originally named Taurasia sa- cyi by Cossmann & Peyrot (1923:257, pl. 13, figs. 31, 32) that may or may not be a synonym of the contem- poraneous Italian Fusus villae Michelotti, 1847. The orig- inal illustration of the latter (Michelotti, 1847:pl. 10, fig. 11) is poor, but it was somewhat better illustrated by Bel- lardi (1872:pl. 9, fig. 20). On the basis of these illustra- tions, it is probable that A. villae and A. sacyi are the same species, but with no Italian material available for study, this is not certain. Thus, in 1982, there was a total of seven (or nine) known species of Aftiliosa: two (or maybe three) occur- ring in the fossil record of Europe: three in the Recent fauna of the New World (two of these also occurring in the fossil record) and possibly three Recent Indo-Pacific forms. Since that time, there have been five additional living species described: A. goreensis Houart, 1993, from the eastern Atlantic (Senegal); A. glenduffyi Petuch, 1993, from the western Atlantic; A. bozzettii Houart, 1993, from the western Indian Ocean (Somalia); A. ruthae Houart, 1996, from the Philippine Islands; and A. edingeri Houart, 1998, from Western Australia. In the last 17 years much new material has been made available, both fossil and Recent. This paper adds three new fossil species from the western Atlantic, and four new Recent species, three from the western Atlantic and one from the Indian Ocean. Synonymies for those Recent species covered previously (Vokes & D’ Attilio, 1982) are not complete but include only the original references and those citations subsequent to 1982; the reader is referred to that paper for more complete information. Likewise, the European fossil species are not treated systematically due to lack of information. On the basis of morphological similarities, there is one group of species beginning with the French Oligocene Attiliosa sp. (Figure 1) and including the modern A. al- dridgei, A. bessei, and A. kevani, in the western Atlantic. The eastern Atlantic A. goreensis is extremely similar to the Oligocene Aftiliosa sp. and also may be included as a member of the “‘aldridgei complex.’’ The more squa- mose Indo-Pacific A. bozzettii and A. houarti and the spi- nose A. orri, although less similar in morphology are, nevertheless, close enough to indicate that they too should be placed in this group, as are A. nodulifera, A. caledon- ica, and A. ruthae. All of these species share a rounded aperture, with an expanded columellar lip and a surface Page 290 ornamentation that is more or less scabrous, sometimes with small spines developed at the intersection of the spi- ral and axial ornamentation. A second morphological grouping would include: the French Miocene A. sacyi (Figure 2); the two new western Atlantic fossil species A. macgintyi and A. gibsonsmithi; A. nodulosa in the eastern Pacific; A. philippiana in the Caribbean; and perhaps A. perplexa on the coast of Bra- zil. These species share an elongated aperture with a nar- row columellar lip and, with the exception of A. perplexa, no varical spines. The recently described Australian spe- cies A. edingeri bears a sufficient resemblance to the east- ern Pacific A. nodulosa to be included also in this group. A third more distantly related set begins with the Early Miocene A. grefae in northwestern Florida, and includes only the living Caribbean A. glenduffyi. ABBREVIATIONS FoR REPOSITORIES oF FIGURED SPECIMENS AMNH, American Museum of Natural History, New York, New York, USA; IRSNB, Institut Royal des Sci- ences Naturelles de Belgique, Brussels, Belgium; MNHN, Muséum National d’Histoire Naturelle, Paris, France; NMB, Naturhistorisches Museum, Basel, Switzerland; SDSNH, San Diego Natural History Museum, San Diego, California, USA; UF Florida Museum of Natural History, University of Florida, Gainesville, Florida, USA; USNM, National Museum of Natural History, Washington, DC, USA. SYSTEMATICS Family MuriciDAE Rafinesque, 1815 Subfamily MurIcINAE Rafinesque, 1815 Genus Attiliosa Emerson, 1968 Attiliosa Emerson, 1968:380. Type species: Coralliophila incompta Berry, 1960 (= Peristernia nodulosa A. Adams, 1855), by original des- ignation. Explanation Figure la, b. Aftiliosa sp. USNM 377398; height 17.7 mm, di- ameter 10.4 mm; locality: Gaas, France, Stampian. <3. Figure 2a, b. Aftiliosa sacyi (Cossmann & Peyrot, 1923). NMB H-18088; height 13.2 mm, diameter 7.4 mm; locality: St. Paul, Dax, France, Burdigalian. 4. Figures 3 and 4. Attilosa nodulosa (A. Adams, 1855). Figure 3. USNM 418059; height 25.7 mm, diameter 16.6 mm; locality: TU 1399, Esmeraldas beds, Onzole Formation. 2. Figure 4. USNM 859927, height 31.5 mm, diameter 19.0 mm; locality: TU R-487, Guaymas, Sonora, Mexico. 2. Figures 5—7. Attiliosa gretae E. H. Vokes, sp. nov. Figure 5a, b. The Veliger, Vol. 42, No. 4 Diagnosis: Shell stoutly biconic; axial ornamentation of six to 12 rounded ridges, rarely with short spinose pro- cesses developed at intersection of axial and spiral or- namentation; aperture large, with nodules or lirations on inner side of outer lip and two or three nodules at anterior end of columellar lip; siphonal canal short, recurved, forming siphonal fasciole. Discussion: The relatively small (both in number of spe- cies and size of shell) genus Afttiliosa is what one might consider the “‘tag-end’’ of the subfamily Muricinae. These plain shells bear little familial resemblance to the more elaborately varicose groups in the subfamily, such as Murex or Chicoreus. In most cases, the shells are al- most non-varicate, with litle more than axial ridges to mark previous positions of the aperture. On some species there are short spinose processes on the varices, e.g., A. nodulifera (Figure 43), which led to its original definition in the genus Murex, but the majority bear no more than a few spinelets, if any. This non-varicate morphology has led to a great deal of confusion among workers as to the systematic position of the few species named prior to Emerson’s recognition (1968:370) of the true nature of the genus, with its mur- icine radula. As a measure of the dubious appearance of the members of this genus, they have been placed at var- ious times in the following genera: Coralliophila, Calo- trophon, Drupa, Fusus, Latiaxis, Murex, Muricopsis, Muricidea, Ocenebra, Peristernia, Phyllonotus, Poirieria, Ocenebra, Taurasia, Trophon, Typhis, Vasum, and prob- ably others! The shells that have engendered this confusion may be characterized as stoutly biconic and generally small for the subfamily (most specimens are under 30 mm in length). There are from six to 12 axial ridges, which may or may not have small spinose processes, and the shell surface varies from scabrose to smooth. The aperture is relatively large, varying from round to elongate, and an anal channel may be present. The most distinctive generic characters are the strong elongate nodules or lirations on the inner side of the outer lip that may extend well back of Figures | to 9 USNM 498197 (Holotype); height 7.7 mm, diameter 4.9 mm; locality: TU 458, Chipola Formation X6. Figure 6a, b. USNM 498198 (Paratype A); height 5.7 mm, diameter 3.7 mm; locality: TU 819, Chipola Formation. Figure 6a, b X6; 6c X10. Figure 7a, b. USNM 498199 (Paratype B); height 9.3 mm, diameter 7.2 mm; locality: TU 819, Chipola Formation. <6. Figures 8 and 9. Attiliosa glenduffyi Petuch, 1993. Figure 8. USNM 860299; height 13.9 mm, diameter 8.0 mm; locality: Do- minican Republic. <4. Figure 9. USNM 860298; height 14.3 mm, diameter 9.2 mm; locality: Dominican Republic (shell not whitened, to show color pattern). <4. Page 291 E. H. Vokes, 1999 Page 292 into the aperture, and the two or three elongate nodules (reduced to only one in A. perplexa) on the anterior por- tion of the columellar lip, which may or may not be ex- panded at the anterior end. The siphonal canal is short and recurved distally, which gives rise to a siphonal fas- ciole. From the list of ‘‘possible’’ genera cited above, the members of the genus Aftiliosa may be separated from most by the presence of a muricine radula and operculum. Murex and Phyllonotus were used only in the broadest sense and, as defined today, have little resemblance. Cal- otrophon and Poireria (Panamurex) are the two most similar appearing taxa; the morphological similarity is close to some species, especially the Miocene Calotro- phon phagon (Gardner, 1947) and Poirieria (Panamurex) mauryae Vokes, 1970. Both of these species share with Attiliosa the columellar nodules and labral lirations, caus- ing us previously (Vokes & D’ Attilio, 1982:68) to suggest that Aftiliosa originated as a branch of the Poiriera clan. Discovery of the Oligocene species assigned to Aftiliosa pushes the separation further back in time; nevertheless it is to the Poirieria clan that Attiliosa bears the strongest morphological similarity. In the Muricidae, convergence of shell form is a com- mon problem. Thus, one branch of Panamurex, beginning with P. mauryae, in time has moved in the direction of shell simplification, resulting in the Recent P. (P.) velero Vokes, 1970, which becomes morphologicaly much like Attiliosa, differing primarily in the strong spiral orna- mentation, more inflated whorls, and more elongate over- all outline. However, Calotrophon, in time, has gone in the direction of greater shell ornamentation, as well as losing the columellar nodules, so that the living members of Calotrophon are less easily confused with Attiliosa. In the final analysis, one must be reminded once again that the concept of generic separation is completely arti- ficial and is “‘in the eye of the beholder.” We are at- tempting to separate the colors of the spectrum into dis- crete boxes. Which box one places which species in is largely subjective. There are certain forms that are un- equivocal (like red, yellow, or blue) but others are less certain (does turquoise belong with blue or green?). Those species that are here grouped in the concept called Attiliosa share a short ‘‘squatty”’ shell outline, with usually a rounded shoulder. Most are non-varicate. But none of these attributes is absolute: A. macgintyi has a distinctly diamond-shaped outline, and both A. nodulifera and A. orri have spinose varices. The alternative to ac- cepting these exceptions is to create yet more smaller boxes to contain them, and in time this may well happen, if enough additional similar appearing forms are discov- ered. After all, we started with one genus Murex, which has been repeatedly subdivided into first more genera, then into families and subfamilies. The solution, at this time, is to divide the genus into ““species-complexes,”’ which at some future date might The Veliger, Vol. 42, No. 4 well become recognized genus-group taxa. On the basis of shell morphology, the genus Aftiliosa may be separated into three distinct species-complexes, as follows: (1) The aldridgei-complex, characterized by a rounded aperture, expanded columellar lip, more or less sca- brous surface ornamentation, sometimes with small spines developed at the intersection of the spiral and axial ornamentation. This complex includes: the un- named French Oligocene species; A. aldridgei, A. bessei, A. kevani, in the western Atlantic, and A. goreensis, in the eastern Atlantic; A. bozzettii, A. orri, A. houarti, in the Indian Ocean; and A. nodulifera, A. caledonica, and A. ruthae in the Pacific. (2) The nodulosa-complex, characterized by an elongat- ed aperture, narrow columellar lip, more or less smooth surface ornamentation, and no varical spines (with the exception of A. perplexa). This complex includes: A. sacyi (?+A. villae), from the Miocene of Europe; A. gibsonsmithi, A. macgintyi, A. philippi- ana, and A. perplexa, in the western Atlantic; A. no- dulosa in the eastern Pacific; and A. edingeri in the Indian Ocean. (3) The glenduffyi-complex, distinguished from the oth- ers by a marked anal channel and including only the Miocene A. gretae and the Recent western Atlantic A. glenduffyi. FOSSIL SPECIES Attiliosa gretae E. H. Vokes, sp. nov. (Figures 5—7) Description: Shell small for the genus (maximum height approximately 12 mm); protoconch of two large, smooth bulbous whorls, ending at small varix. Axial ornamenta- tion on all teleoconch whorls of eight rounded ridges; on earliest teleoconch whorls small open flanges on abaper- tural side of ridges but these disappearing by about fourth teleoconch whorl. Spiral ornamentation of one strong cord at shoulder, on early whorls several smaller cords anterior to shoulder but most weakening as shell increases in size, leaving just two cords anterior to shoulder; one smaller cord on siphonal canal. Where spiral cords cross axial ridges weak pointed knobs developed; otherwise, shell surface smooth. Suture appressed, sinuated by axial ridges. Aperture oval, marked anal channel; columellar lip smooth except for two small denticles at anterior end; inner side of outer lip with seven strong lirae extending well into aperture. Siphonal canal short, broad: siphonal fasciole increasing in width with increasing shell size. Traces of heavy intritacalx indicating that in life shell surface was covered by this chalky layer. Holotype: USNM 498197; height 7.7 mm, diameter 4.9 mm (Figure 5). Type locality: Chipola Formation; TU 458, east bank of E. H. Vokes, 1999 Chipola River, above Farley Creek (SW % sec. 20, T. 1 N, R. 9 W), Calhoun County, Florida. Paratype A: USNM 498198; height 5.7 mm, diameter 3.7 mm; locality: TU 819 (Figure 6). Paratype B: USNM 498199; height 9.3 mm, diameter 7.2 mm; locality: TU 819 (Figure 7). Occurrence: Chipola Formation, TU localities 458, 547, 548, 817, 819, 999, 1196. Discussion: For some time, we have had a dozen speci- mens of a small species taken from primarily coralline localities in the Chipola Formation, northwestern Florida. Obviously muricid, the exact generic placement was a puzzle until the discovery of A. glenduffyi, living off the shores of the Dominican Republic. This Recent form shares with the Chipola shell a marked anal channel and axial ornamentation consisting of rounded ribs but no true varices. However, A. gretae differs from A. glenduffyi in having a lower spire, a shorter siphonal canal, and a less appressed suture. Both species are small, although the Chipola form is slightly smaller, with a maximum size of 12 mm, but A. glenduffyi attains an adult size of approx- imately 15 mm. Although difficult to see in the photo- graph, paratype B shows typical Aftiliosa nodules at the base of the columellar lip. In addition to the Tulane material, there are four spec- imens in the private collection of Mr. and Mrs. Andrew Murray, of Bradenton, Florida. It is a pleasure to name this new species in honor of Greta (Mrs. Andrew) Mur- ray, for her excellent work on the Chipola fauna, includ- ing collecting paratype B. Attiliosa gibsonsmithi E. H. Vokes, sp. nov. (Figures 10—13) Description: Shell large for the genus (maximum height approximately 25 mm), inflated in outline. Protoconch of one and three-quarters smooth, bulbous whorls; six teleo- conch whorls. Axial ornamentation of nine or 10 swollen axial ridges on each whorl. Spiral ornamentation very faint on early whorls, gradually increasing in strength and becoming three flattened cords on spire whorls; approx- imately six spiral cords on body whorl, that at base of body whorl the largest. Suture appressed, sinuated by ax- ial ridges. Aperture round; inner lip smooth, appressed entire length; two weak nodules at anterior end. Margin of outer lip crenulated by spiral cords, a small projection formed by cord at base of body whorl; nine thin lirae extending far back into aperture. Siphonal canal short, broad; recurved at distal end, forming deep siphonal fas- ciole. Holotype: NMB H-18084; height 19.2 mm, diameter 12.8 mm (Figure 10). Page 293 Type locality: Mataruca Member, Caujaro Formation; NMB 17530, Cementerio de Carrizal, Falcon, Venezuela. Paratype A: NMB H-18085; height 20.4 mm, diameter 11.9 mm (Figure 11). Paratype B: NMB H-18086; height 20.0 mm, diameter 12.3 mm (Figure 12). Paratype C: NMB H-18087; height 14.3 mm, diameter 8.3 mm (Figure 13). Locality of all same as holotype. Occurrence: All material from type locality. Discussion: In the collections of the Naturhistorisches Museum, Basel, Switzerland, there are 35 examples of a second new species, taken from Late Miocene aged beds of the Mataruca Member of the Caujaro Formation, Ven- ezuela. This species also has been a puzzle for many years. The type material was originally sent to me by the collectors, Mr. and Mrs. Jack Gibson Smith, then of Ca- racas, Venezuela, now of Surrey, England. But neither they nor I could come up with any unequivocal generic placement for the form. With the recognition of Coss- mann & Peyrot’s Taurasia sacyi as a member of Attiliosa, similarity to the Venezuelan material indicated placement here might also be appropriate. Morphologically, they both have a strong basal cord, separated from the body whorl, which has rounded axial ribs but no true varices. The suture is sinuated by the axial ribs, and the aperture is strongly lirate within. But A. sacyi is much smaller and more evenly biconic in outline, A. gibsonsmithi having more inflated whorls. In the living fauna the species most nearly akin to A. gibsonsmithi is that found off the west coast of tropical America, A. nodulosa (Adams), which differs from the older species in having a more appressed suture. So, pos- sibly, A. sacyi gave rise to A. gibsonsmithi, which in turn gave rise to A. nodulosa, forming a species-complex long separate from the other Atlantic species. If A. sacyi also gave rise, independently, to the A. philippiana line, this would explain the similarity between the Pacific A. no- dulosa and the Atlantic A. philippiana. Attiliosa macgintyi E. H. Vokes, sp. nov. (Figures 14—16) Description: Shell biconic in outline; six teleoconch whorls in adult, early whorls unknown. Axial ornamen- tation of 11 or 12 rounded ridges on each teleoconch whorl. Spiral ornamentation of low raised cords, only two visible on spire whorls; approximately three on body whorl and three on siphonal canal, but number varying greatly between specimens. Where spiral cords cross ax- ial ridges small nodes developed, otherwise shell surface smooth. Suture appressed, undulated by axial cords; shoulder very sloping. Aperture diamond-shaped, slight anal channel at posterior end. Columellar lip smooth, ap- Page 294 The Veliger, Vol. 42, No. 4 E. H. Vokes, 1999 pressed posteriorly, free-standing anteriorly; two small nodules at anterior end. Inner side of outer lip with seven lirae extending well into aperture. Siphonal canal short, broad, recurved at distal end; deep siphonal fasciole. Holotype: UF 90727; height 24.3 mm, diameter 14.7 mm (Figure 14). Type locality: Tamiami Formation; UF CR007 (= TU 797), material exposed during construction of “‘Alligator Alley,” 13.3 miles east of Florida Highway 29 (T. 49 S, R. 32 E) Collier County, Florida. Paratype A: UF 90728; height 21.8 mm, diameter 12.8 mm (Figure 15). Paratype B: UF 90729; height 20.9 mm, diameter 13.6 mm (Figure 16). Locality: of both same as holotype. Occurrence: All material from type locality. Discussion: In the collection of the late Tom McGinty, Palm Beach, Florida, now located at the Florida Museum of Natural History, there are three examples of a new species originally taken from exposures of the Tamiami Formation available during construction of “Alligator Al- ley” (Florida Highway 84), southern Florida. Unfortu- nately, the three known specimens are badly worn, and details of early whorls are lacking. These shells bear some resemblance to the Recent A. philippiana, but differ in having a striking expansion at the periphery, a less appressed suture, and a somewhat larger size. The new species attains a maximum size of about 25 mm in con- trast to a maximum of under 20 mm for A. philippiana. The overall outline of the shell is much closer to that of the Middle Miocene A. sacyi, suggesting that this new form is intermediate between the latter and A. philippi- ana. Attiliosa aldridgei (Nowell-Usticke, 1969) (Figures 17—25) Vasum aldridgei Nowell-Usticke, 1969:18, pl. 4, fig. 834. Attiliosa aldridgei (Nowell-Usticke). Nowell-Usticke, 1971: Page 295 11, pl. 2, fig. 680; Vokes & D’ Attilio, 1982:69, figs. 6— 9; Vokes, 1989:62, pl. 6, figs. 9, 10; Vokes, 1992:93, pl. 20, figs. 5—8; Houart, 1993a:21, fig. 14. Holotype: American Museum of Natural History, no. AMNH 189620; height 29.4 mm, diameter 20.0 mm. Type locality: Rat Island, Antigua, B.W.I. Fossil occurrences: TU localities 283, 727 (Bermont Formation); 1215 (Gurabo Formation), 1422 (Cercado Formation); 1240 (Moin Formation). Figured specimens: Figure 17. AMNH 168901 (Paratype); height 22.4 mm, diameter 14.0 mm; locality: Antigua, B.W.I. Figure 18. USNM 890890; height 29.7 mm, diameter 18.3 mm; locality: Bimini, B.W.I., 10 meters. Figure 19. USNM 792393; height 12.3 mm, diameter 8.2 mm); locality: Bimini, B.W.I., 10 meters. Figure 20. USNM 890891; height 16.4 mm, diameter 10.5 mm; locality: Discovery Bay, Jamaica, 10 meters. Figure 21. USNM 869515; height 19.1 mm, diameter 12.8 mm; locality: TU R-369, Moin, Costa Rica. Figure 22. USNM 498200; height 20.1 mm, diameter 11.7 mm; locality: TU 1422, Cercado Formation. Figure 23. USNM 890892; height 20.7 mm, diameter 12.4 mm; locality: TU R-109, Bahia de Las Minas, Pan- ama. Figure 24. USNM 890893; height 15.1 mm, diameter 9.2 mm); locality: Cartagena, Colombia. Figure 25. USNM 323924; height 10.8 mm, diameter 6.9 mm; locality: TU 1240, Moin Formation. Discussion: For a complete synonymy of citations prior to 1982, see Vokes & D’ Attilio (1982:69). At that time, the species was known from a relatively few individuals, collected from Bimini, B.W.I. to Panama. Since that orig- inal discussion we have obtained numerous additional Re- cent specimens from other Caribbean Recent localities, and several fossil examples, especially from the coral- reefs of the Cercado and Gurabo formations (Mio-Plio- cene) of the Dominican Republic, as well as the Caloosa- hatchee and Bermont formations of Florida. There is a fair degree of variability in the overall mor- Explanation of Figures 10 to 16 Figures 10-13. Attiliosa gibsonsmithi E. H. Vokes, sp. nov. Fig- ure 10a, b. NMB H-18084 (Holotype); height 19.2 mm, diameter 12.8 mm; locality: NMB 17530, Mataruca Member, Caujaro For- mation. X3. Figure lla, b. NMB H-18085 (Paratype A); height 20.4 mm, diameter 11.9 mm; locality: same as holotype. 3. Figure 12a, b. NMB H-18086 (Paratype B); height 20.0 mm, diameter 12.3 mm; locality: same as holotype. <3. Figure 13. NMB H-18087 (Paratype C); height 14.3 mm, diameter 8.3 mm; locality: same as holotype. x10. Figures 14-16. Attiliosa macgintyi E. H. Vokes, sp. nov. Figure 14a, b. UF 90727 (Holotype); height 24.3 mm, diameter 14.7 mm; locality: UF CRO07 (= TU 797), Tamiami Formation. <2.5. Figure 15a, b. UF 90728 (Paratype A); height 21.8 mm, diameter 12.8 mm; locality: same as holotype. X2.5. Figure 16a, b. UF 90729 (Paratype B); height 20.9, diameter 13.6 mm; locality: same as holotype. 2.5. The Veliger, Vol. 42, No. 4 Page 296 E. H. Vokes, 1999 phology of this species. Some specimens are low-spired and ‘‘chunky”’ (e.g., Figures 21 and 23), and some have a higher spire, with an impressed suture, which gives a ‘““stepped’’ appearance to the shell (e.g., Vokes & D’ Attilio, 1982:fig. 7). Although the two ‘“‘chunky” ex- amples figured here both come from the southern Carib- bean, there does not seem to be any particular geographic distribution to the differences. In the Mio-Pliocene beds of the Dominican Republic, both forms occur together (compare Vokes, 1989:pl. 6, figs. 9, 10). The southern forms have a more lirate aperture and also are more spi- nose in the younger stages (compare Figure 19 with Fig- ures 24, 25). Given the similarity of the adult specimens, there seems little reason to separate the southern form as a distinct species from the more northern typical exam- ples, but rather to accept them as the end members of a single cline. The typical form is ornamented by thin brown lines topping the spiral cords (Figure 18), but some specimens have a single broad color broad at the periphery (Figure 20). This latter color morph was named Muricopsis poeyi by Sarasta & Espinosa (1979:2, fig. 1; holotype refigured by Vokes & D’ Attilio, 1982:fig. 10). The only examples I have seen with such a pattern come from the Greater Antilles, and this may be a geographic variation. Attilosa nodulosa (A. Adams, 1855) (Figures 3, 4) Peristernia nodulosa A. Adams, 1855:313. Coralliophila incompta Berry, 1960:119. Attiliosa nodulosa (Adams). Vokes & D’ Attilio, 1982:69; Vokes, 1988:33, pl. 6, figs. 5, 6. Syntypes: The Natural History Museum, London [British Museum (Natural History)]; see Bullock, 1976:pl. 1, figs. 6, 8. Type locality: ‘‘Australia.”’ Fossil occurrence: Esmeraldas beds, Onzole Formation, TU locality 1399. Page 297 Figured specimens: Figure 3. USNM 418059; height 25.7 mm, diameter 16.6 mm; locality: TU 1399, Esmeraldas beds, Onzole For- mation. Figure 4. USNM 859927; height 31.5 mm, diameter 19.0 mm; locality: TU R-487, Guaymas, Sonora, Mexico. Discussion: For a complete synonymy and discussion of the convoluted nomenclatorial history of this species, which is the type of the genus Aftiliosa, see Vokes & D’ Attilio (1982:69). Discovery of a fossil specimen in the Early Pliocene Esmeraldas beds, Onzole Formation, of northwestern Ecuador (Vokes, 1988:pl. 6, fig. 6; refigured here, Figure 3), reveals a considerable geologic history for the eastern Pacific species. RECENT SPECIES Attiliosa glenduffyi Petuch, 1993 (Figures 8, 9) Attiliosa sp. Vokes, 1992:95, pl. 20, figs. 10, 11. Attiliosa glenduffyi Petuch, 1993:54, figs. 6, 7. Holotype: Carnegie Museum of Natural History; height 13 mm, diameter 9 mm (fide Petuch, 1993:54). Type locality: Samana, Dominican Republic. Figured specimens: Figure 8. USNM 860299; height 13.9 mm, diameter 8.0 mm; locality: Dominican Republic. Figure 9. USNM 860298; height 14.3 mm, diameter 9.2 mm; locality: Dominican Republic. Discussion: This recently described western Atlantic ad- dition to the genus is unique in having a strong color pattern, consisting of a dark brown shell with a white band circling the periphery. As suggested above, it is more similar to the Chipola A. gretae than to the other living forms and may well represent a different species- complex altogether. At this time, the species is known only from the vicinity of the type locality, where it occurs on and under rocks and coral rubble in depths of 1—5 meters (Petuch, 1993:54). Explanation of Figures 17 to 25 Figures 17-25. Attiliosa aldridgei (Nowell-Usticke, 1969). Figure 17. AMNH 168901 (Paratype); height 22.4 mm, diameter 14.0 mm; locality: Antigua, B.W.I. X2.5. Figure 18. USNM 890890; height 29.7 mm, diameter 18.3 mm; locality: Bimini, B.W.I., 10 meters (shell not whitened, to show color pattern). 2. Figure 19a, b. USNM 792393; height 12.3 mm, diameter 8.2 mm; locality: Bimini, B.W.I., 10 meters. Figure 19a <4; 19b X10. Figure 20. USNM 890891; height 16.4 mm, diameter 10.5 mm; locality: Discovery Bay, Jamaica, 10 meters (shell not whitened, to show color pattern). X2.5. Figure 21a, b. USNM 869515; height 19.1 mm, diameter 12.8 mm; locality: TU R-369, Moin, Costa Rica. 2.5. Figure 22a, b. USNM 489200; height 20.1 mm, diameter 11.7 mm; locality: TU 1422, Cercado Formation. X2.5. Figure 23a, b. USNM 890892; height 20.7 mm, diameter 12.4 mm; locality: TU R-109, Bahia de Las Minas, Panama. 2.5. Figure 24a, b. USNM 890893; height 15.1 mm, diameter 9.2 mm; locality: Cartagena, Colombia. 4. Figure 25. USNM 323924; height 10.8 mm, diameter 6.9 mm; locality: TU 1240, Moin Formation. 4. Page 298 The Veliger, Vol. 42, No. 4 E. H. Vokes, 1999 Attiliosa philippiana (Dall, 1889) (Figures 30, 31) Muricidea philippiana Dall, 1889:213; 1902:504, pl. 29, fig. 5. Attiliosa aldridgei (Nowell-Usticke). Vokes, 1976:in part, pl. 8, fig. 10 only. Attiliosa philippiana (Dall). Vokes & D’ Attilio, 1982:69; Vokes, 1992:94, pl. 20, fig. 9. Lectotype: United States National Museum of Natural History, no. USNM 93337; height 14.9 mm, diameter 8.8 mm. Type locality: U.S. Fish Commission Station 2362, off Cabo Catoche, Quintana Roo, Mexico, in 25 fathoms [46 meters]. Figured specimens: Figure 30. USNM 890894; height 14.8 mm, diameter 8.5 mm; locality: San Andres Island, Colombia, 15 meters. Figure 31. USNM 711114; height 10.6 mm, diameter 6.0 mm; locality: TU R-98, Holandes Cay, Panama. Discussion: For a complete synonymy and history of this misunderstood species, see Vokes & D’ Attilio (1982:69). In that discussion it was not noted that one of the spec- imens figured by Vokes (1976:pl. 8, fig. 10; refigured here, Figure 31) as A. aldridgei is actually a juvenile ex- ample of A. philippiana. Perhaps the misidentification was a result of the belief held at the time that A. philip- piana was found only in the Florida- Yucatan portion of the western Atlantic (Vokes, 1976:122). But several larg- er examples (Figure 30) from San Andres Island, taken in 15 meters by SCUBA diver, confirm the presence of this species in the southern Caribbean. Attiliosa bessei E. H. Vokes, sp. nov. (Figures 26—29) Description: Early whorls unknown, all material badly eroded. At least six teleoconch whorls. Axial ornamen- tation of six or seven rounded ridges on each teleoconch Page 299 whorl. Spiral ornamentation very faint, approximately eight cords on body whorl. Suture appressed, sinuated by axial ridges; shoulder sloping. Aperture rounded; colu- mellar lip appressed at posterior end, free-standing at an- terior end; smooth except for two or three elongate nod- ules at anterior end. Outer lip patulous, margin serrated by almost invisible spiral cords, with small adaperturally directed points corresponding to grooves between spiral cords. About seven strong lirae on inner side of outer lip. Siphonal canal short, broad; recurved at distal end, form- ing a deep, wide siphonal fasciole. Shell invariably coated with coralline algae; when removed, color pattern re- vealed as variable brown lines topping spiral cords. Holotype: USNM 880260; height 24.6 mm, diameter 14.5 mm (Figure 26). Type locality: Rosalind Bank, Bay Islands, Honduras, from 30 meters in lobster traps. Paratype A: USNM 880261; height 18.8 mm, diameter 11.0 mm; locality same as holotype (Figure 27). Paratype B: USNM 880262; height 21.6 mm, diameter 12.7 mm; locality: Gorda Banks, Bay Islands, Honduras, from lobster traps (Figure 28). Paratype C: USNM 880263; height 21.0 mm, diameter 11.9 mm; locality: Gorda Banks, Bay Islands, Honduras, from lobster traps (Figure 29). Discussion: From the vicinity of the Bay Islands, Hon- duras, there is a species that is closely related to the more widespread A. aldridgei. However, this new species dif- fers from the latter in having a higher spire and shorter siphonal canal, so that the outline of the shell is biconic with the shoulder knobs at the midpoint of the shell height. It is totally lacking in spines, and the color pattern varies from white with numerous thin brown spiral lines (Figure 28) to brown with thin white spiral lines (Figure 29). This new species is named in honor of Mr. Bruno Bes- Explanation of Figures 26 to 34 Figures 26-29. Attliosa bessei E. H. Vokes, sp. nov. Figure 26a, b. USNM 880260 (Holotype); height 24.6 mm, diameter 14.5 mm; locality: Rosalind Bank, Bay Islands, Honduras. X2.5. Fig- ure 27a, b. USNM 880261 (Paratype A); height 18.8 mm, di- ameter 11.0 mm; locality: Rosalind Bank, Bay Islands, Hondu- ras. X2.5. Figure 28. USNM 880262 (Paratype B); height 21.6 mm, diameter 12.7 mm; locality: Gorda Bank, Bay Islands, Hon- duras (shell not whitened, to show color pattern). *2.5. Figure 29. USNM 880263 (Paratype C); height 21.0 mm, diameter 11.9 mm; locality: Gorda Bank, Bay Islands, Honduras (shell not whitened, to show color pattern). «2.5. Figures 30 and 31. Attiliosa philippiana (Dall, 1889). Figure 30a, b. USNM 890894; height 14.8 mm, diameter 8.5 mm; locality: San Andres Island, Colombia, 15 meters. X4. Figure 31. USNM 711114; height 10.6 mm, diameter 6.0 mm; locality: TU R-98, Holandes Cay, Panama. *4. Figures 32-34. Attiliosa kevani E. H. Vokes, sp. nov. Figure 32a—c. USNM 880264 (Holotype); height 17.9 mm, diameter 11.0 mm; locality: Montego Bay, Jamaica (shell in Fig. 32c not whitened, to show color pattern). <3. Figure 33a, b. USNM 880265 (Paratype A); height 15.2 mm, diameter 10.1 mm; lo- cality: Montego Bay, Jamaica. 3. Figure 34. USNM 880266 (Paratype B); height 7.4 mm, diameter 4.3 mm; locality: Utila, Bay Islands, Honduras, 25 meters. <6. Page 300 se, diver in the Bay Islands, who provided much of the available material. Attiliosa kevani E. H. Vokes, sp. nov. (Figures 32-34) Muricopsis pudicus (Reeve). Humfrey, 1975:138, pl. 16, fig. 7 (not of Reeve). Attiliosa aldridgei (Nowell-Usticke). Vokes, 1992:in part, discussion p. 93 only. Description: Shell small for genus (maximum height un- der 18 mm); spire high. Protoconch of one and one-half flattened and tilted whorls (cf. Radwin & D’ Attilio, 1976: text-fig. 58—Prototyphis angasi); six teleoconch whorls. Axial ornamentation beginning with five small varices on first two teleoconch whorls, increasing to six or seven on later whorls. On first three ornamented whorls each varix with a long, adapically recurved spine at shoulder; spines continuing on later whorls but not as long relative to shell size. No spiral ornamentation visible on spire whorls; on body whorl four faint spiral cords between shoulder and base of whorl; except for varices shell surface almost smooth. Suture appressed, sinuated by varices; shoulder extremely sloping, resulting in spire being almost one- half entire shell height. Aperture rounded, columellar lip expanded and appressed; smooth except for two elongate nodules at anterior end. Outer lip patulous; margin ser- rated with small adaperturally directed points correspond- ing to grooves between spiral cords, in immature speci- mens that one at base of body whorl forming a small sinusigeral projection. Six or seven strong lirae set back from margin of outer lip, extending far into aperture. Si- phonal canal short, broad; reflected at distal end, forming a small siphonal fasciole. Color white, with vague brown lines topping spiral cords; some specimens also with broad brown band on shoulder and base of body whorl. Holotype: USNM 880264; height 17.9 mm, diameter 11.0 mm (Figure 32). The Veliger, Vol. 42, No. 4 Type locality: Montego Bay, Jamaica, 25 meters. Paratype A: USNM 880265; height 15.2 mm, diameter 10.1 mm; locality: Montego Bay, Jamaica, 25 meters (Figure 33). Paratype B: USNM 880266; height 7.4 mm, diameter 4.3 mm; locality: Utila, Bay Islands, Honduras, 25 meters (Figure 34). Discussion: In a previous discussion of A. aldridgei, I stated (Vokes, 1992:93) that the specimen figured by Humfrey (1975:pl. 16, fig. 7) as Muricopsis pudicus (Reeve) was certainly not that species, which is a West African Hexaplex, but was simply a spinose juvenile specimen of A. aldridgei. Furthermore, I noted that in the collection of Kevan and Linda Sunderland, Sunrise, Flor- ida, there were similar juvenile examples taken from 12 to 25 meters depth in the Bay Islands. Since that time numerous adult specimens have been taken by the Sun- derlands in both the Bay Islands and Jamaica, and it is now clear that this is a new species, resembling A. al- dridgei only in the juvenile stages but very different in the adult. According to Mr. Sunderland, this form is al- ways found in old dead reef systems, on algae and always very encrusted with lime, as is typical of other members of the genus Attiliosa. From A. aldridgei the new species differs in being smaller (the holotype at just under 18 mm is by far the largest specimen seen) and more spinose. The outline of the shell is less inflated in A. kKevani and the spire is much higher—almost one-half the total height of the shell. From the other new species, A. bessei, A. kevani differs in much the same ways, with A. bessei being even less spinose than A. aldridgei. Attiliosa perplexa E. H. Vokes, sp. nov. (Figures 35-37) Description: Shell small for genus (maximum height 12 mm), biconic in outline. Protoconch of one and one-half Explanation of Figures 35 to 43 Figure 35-37. Attiliosa perplexa E. H. Vokes, sp. nov. Figure 35a, b. USNM 880257 (Holotype); height 12.6 mm, diameter 6.8 mm; locality: off Guarapari, Espirito Santo, Brazil, under rocks at 20 meters. 4. Figure 36a, b. USNM 880258 (Paratype A); height 10.7 mm, diameter 6.3 mm; locality: same as holotype. x4. Figure 37a, b. USNM 880259 (Paratype B); height 11.4 mm, diameter 6.3 mm; locality: Rasa Island, off Guarapari, Espirito Santo, Brazil, dredged in 30 meters. <4. Figures 38 and 39. Attiliosa houarti E. H. Vokes, sp. nov. Figure 38a, b. USNM 880255 (Holotype); height 23.5 mm, diameter 12.9 mm; locality: Phuket, Thailand, 30 meters. 2.5. Figure 39a, b. USNM 880256 (Paratype); height 15.7 mm, diameter 8.3 mm; locality: Kor Bon Island, western Thailand, 12 meters. 3. Figure 40a, b. Attiliosa goreensis Houart, 1993. MNHN (Holo- type); height 14.2 mm, diameter 9.0 mm; locality: Gorée, Se- negal, 20-25 meters. 3. Figure 41a, b. Attiliosa bozzettii Houart. 1993 IRSNB 1G27.873/ 454 (Holotype); height 17.0 mm, diameter 10.1 mm; locality: Ras Hafun, Somalia, 150—200 meters. 3. Figure 42. Attiliosa orri (Cernohorsky, 1976). USNM 890895; height 30.9 mm; diameter 22.7 mm (including spines); locality: Kantang, Thailand. <2. Figure 43. Attiliosa nodulifera (Sowerby, 1841). SDSHN 78076a; height 29.0 mm, diameter 20.1 mm; locality: Ataa, Ma- laita, Solomon Islands. *2. EH: Vokes, 1999 Page 301 Page 302 smooth, bulbous whorls; six teleoconch whorls. Axial or- namentation of six rounded ridges on each whorl. Spiral ornamentation of flattened cords; four on spire whorls between shoulder and suture; on body whorl two major cords, one at shoulder, one at periphery, approximately 12 secondary cords. Where two major cords cross axial ridges open spines produced, that at shoulder much larg- er; small open flanges at minor cords. Suture incised; un- dulated by axial ridges. Aperture elongate-oval; inner lip smooth, appressed; one small nodule at anterior end. Mar- gin of outer lip extended, scalloped by spiral cords; five strong lirae on inner side. Siphonal canal short, broad; recurved at distal end, forming small siphonal fasciole. Outer surface of shell covered by intritacalx; when re- moved, a single brown spiral band visible at base of body whorl. Holotype: USNM 880257; height 12.6 mm, diameter 6.8 mm (Figure 35). Type locality: Off Guarapari, Espirito Santo, Brazil, un- der rocks at 20 meters. Paratype A: USNM 880258; height 10.7 mm, diameter 6.3 mm; locality same as holotype (Figure 36). Paratype B: USNM 880259; height 11.4 mm, diameter 6.3 mm; Rasa Island, off Guarapari, Espirito Santo, dredged in 30 meters (Figure 37). Discussion: From off the coast of Espirito Santo, Brazil, José and Marcus Coltro have collected a number of spec- imens of a small species that, although obviously muri- cine, seemed to defy placement in any recognized genus. The perplexing question of the generic assignment of this unusual new species was first resolved by Mr. Roland Houart, who recognized that Attiliosa might be the proper assignment (1995, personal communication). However, in terms of other known members of Aftiliosa, there is none that bears more than a generic resemblance. The shell is the most attenuated of all Afttiliosa species, with the height: width ratio more than 2:1. The narrow shell, with a few relatively strong spiral cords, and small size (max- imum height about 12 mm) suggests a relationship with the French Miocene A. sacyi, and therefore, the species is tentatively included with that species-complex. Attiliosa goreensis Houart, 1993 (Figure 40) Attiliosa goreensis Houart, 1993a:20, figs. 11, 12 (holotype), 13 (paratype), 18, 19 (radula), 26, 27 (protoconch). Holotype: Muséum National d’Histoire Naturelle, Paris; height 14.2 mm, diameter 9.0 mm. Type locality: Gorée, Senegal, 20—25 meters. Figured specimen: Holotype (photograph courtesy of Roland Houart). The Veliger, Vol. 42, No. 4 Discussion: As noted in the original description (Houart, 1993a:21), this species is most closely related to A. al- dridgei but differs in having more shouldered whorls, with more regular and equi-sized spiral cords. It is also smaller, attaining a maximum size of approximately 15 mm. Interestingly, the holotype (Figure 40) shows a brown spiral band on the body whorl similar (although narrower) to the color form described as A. poeyana (cf. Figure 20). Attiliosa houarti E. H. Vokes, sp. nov. (Figures 38, 39) Description: Early whorls unknown, seven teleoconch whorls in adult. Axial ornamentation of six or seven rounded ridges on each teleoconch whorl and multiple growth lamellae covering entire surface of shell. No var- ical break visible until adult body whorl. Spiral ornamen- tation of raised cords alternating with smaller threads; approximately 12 major cords on body whorls plus six smaller threads on siphonal canal. Intersection of spiral ornament and growth lamellae giving rise to a scabrous shell surface; on adapertural side of axial ridges lamellae forming small open flanges; that formed by cord at base of body whorl strongest. Suture slightly appressed. Ap- erture rounded; columellar lip greatly expanded and ap- pressed at posterior end, free-standing at anterior end; smooth, with three small nodules at anterior end. Margin of outer lip serrated by spiral ornamentation, seven elon- gate nodules on inner side. Siphonal canal short, broad; extremely recurved at distal end with terminations of ax- ial ridges forming small spurs encircling a deep siphonal fasciole. Color uniformly brown; aperture white. Holotype: USNM 880255; height 23.5 mm, diameter 12.9 mm (Figure 38). Type locality: Phuket, Thailand, in rubble, 30 meters. Paratype: USNM 880256; height 15.7 mm, diameter 8.3 mm; Kor Bon Island, western Thailand, 12 meters (Figure 39). Discussion: Most closely related to A. bozzettii Houart, described from the opposite side of the Indian Ocean, A. houarti differs from the latter in having a larger aperture and shorter siphonal canal. Both differ from the third In- dian Ocean species, A. orri, in lacking the extreme de- velopment of varical spines seen in that form. Neverthe- less, these three Indian Ocean species are more similar to the Atlantic species of the “‘A. aldridgei group”’ than to the species in the Pacific Ocean. Attiliosa bozzettii Houart, 1993 (Figure 41) Attiliosa bozzettii Houart, 1993b:42, figs. 1, 2 (holotype), 3 (paratype), 9 (protoconch). E. H. Vokes, 1999 Holotype: Institut Royal des Sciences Naturelles de Bel- gique, no. IG27.873/454; height 17.0 mm, diameter 10.1 mm. Type locality: Ras Hafun, Somalia, 150—200 meters. Figured specimen: Holotype (photograph courtesy of Roland Houart). Discussion: There is a strong similarity between three Indian Ocean species of Afttiliosa. Of these, A. orri is the most spinose, A. houarti is slightly spinose, and A. boz- zettii is completely non-spinose. All three share an ex- tremely squamose surface ornamentation. A fourth spe- cies, A. edingeri Houart, 1998, recently described from the Indian Ocean side of Australia does not seem to be closely related to these three species but rather is a mem- ber of the A. nodulosa complex. Attiliosa orri (Cernohorsky, 1976) (Figure 42) Muricopsis orri Cernohorsky, 1976:116, figs. 12—20. Attiliosa orri (Cernohorsky). Vokes & D’ Attilio, 1982:71. Holotype: Auckland Institute and Museum, no. TM- 1346; height 27.1 mm, diameter 18.6 (spines excluded). Type locality: Andaman Islands, Indian Ocean, in 55 me- ters. Figured specimen: USNM 890895; height 30.9 mm; di- ameter 22.7 mm (including spines); locality: Kantang, Thailand. Discussion: For a complete synonymy see Vokes & D’ Attilio (1982:71). Attiliosa nodulifera (Sowerby, 1841) (Figure 43) Murex noduliferus Sowerby, 1841a:8, pl. 194, fig. 94; 1841b:147. Murex (Trophon) fruticosus Gould, 1849:143. Murex pagodus A. Adams, 1853:269. Attiliosa nodulifera (Sowerby). Vokes & D’ Attilio, 1982:70, in part, figs. 1-4 (only); D’ Attilio & Myers, 1986:62, figs. 8, 9; Houart, 1996:61, fig. 16. Syntypes: The Natural History Museum, London [British Museum (Natural History)], nos. 1842.5.10 (1618-1619); height of larger (figured by Cernohorsky, 1976: figs. 22, 23) height 20.2 mm, diameter 12.8 mm. Type locality: Masbate, Philippine Islands. Figured specimen: SDSNH 78076a; height 29.0 mm, di- ameter 20.1 mm; locality: Ataa, Malaita, Solomon Is- lands. Discussion: According to Cernohorsky (1976:119), the original illustration of Murex noduliferus Sowerby, 1841, Page 303 is a composite of the two syntypes, a worn mature ex- ample (figured by Cernohorsky) and an immature speci- men, which better shows the spines. Attiliosa caledonica (Jousseaume, 1881) Muricidea caledonica Jousseaume, 1881:349; 1882:345. Murex (Muricidea) caledonica (Jousseaume). Poirier, 1883: 110, pl. 5, fig. 3 ectotype; designated by Fischer-Piette and Beigbeder, 1943:206). “Muricidea”’ caledonica (Jousseaume). Vokes & D’ Attilio, 1982:68, fig. 5 (“‘lectotype’’). Attiliosa caledonica (Jousseaume). D’ Attilio & Myers, 1986:59, figs. 1-7, 10 (fig. 4c = lectotype); Houart, 1996:61, fig. 17; Houart, 1998:96, fig. 31. Lectotype: Muséum National d’ Histoire Naturelle, Paris; height 26.2 mm, diameter 19.0 mm (fide D’ Attilio and Myers, 1986: fig. 4c). Type locality: New Caledonia. Discussion: In our original discussion (Vokes & D Attilio, 1982:68), we noted the differences between A. nodulifera and A. caledonica and concluded that the two forms were synonymous. However, D’ Attilio & Myers (1986) have presented convincing evidence that the two are indeed distinct species. They also demonstrated that the specimen figured as “‘lectotype”’ by Fair (1976:pl. 17, fig. 229) and by Vokes & D’ Attilio (1982:fig. 5) was not the specimen figured by Poirier (1883:pl. 5, fig. 3), which had previously been designated as lectotype by Fischer- Piette & Beigbeder (1943:206); in fact, it is not even part of the type lot. For a complete synonymy of A. caledonica see Vokes & D Attilio (1982:70). Attiliosa ruthae Houart, 1996 Attiliosa ruthae Houart, 1996:61, figs. 15 (paratype), 31-32 (holotype). Holotype: Muséum National d’ Histoire Naturelle, Paris; height 27.5 mm, diameter 18.1 mm. Type locality: Cebu, Philippine Islands. Discussion: This recently described species is most sim- ilar to the sympatric A. nodulifera but has fringed varices rather than spines. Attiliosa edingeri Houart, 1998 Attiliosa edingeri Houart, 1998:96, figs. 1, 2 (holotype), 3, 4 (paratypes), 40 (radula). Holotype: Western Australian Museum, no. WAM S.1101; height 31.9 mm, diameter 18.1 mm. Type locality: Off Esperance, Western Australia, 31—36 meters. Discussion: This recently described species is, as noted Page 304 by Houart (1998:96), unlike any other known from the Indo-Pacific. In its non-varicate morphology it most near- ly resembles the eastern Pacific A. nodulosa but differs in having a more scabrous surface ornamentation. The two are also similar in their relatively large size; the larg- est specimen of A. edingeri measures 35.7 mm in height (Houart, 1998:fig. 4), which is only a bit larger than the specimen here figured (Figure 4) of A. nodulosa. Acknowledgments. This study is largely the result of material provided by other persons and I am grateful to all of the friends who contributed specimens and information. In particular, Kevan and Linda Sunderland, Sunrise, Florida, and Jose and Marcus Coltro, Sao Paulo, Brazil, have always been extremely generous with Caribbean and Brazilian material; without them this study would not have been possible. Mr. and Mrs. Jack Gibson Smith, Surrey, England, originally provided some of the Venezuelan ma- terial, which was later augmented by a loan from Peter Jung, Naturhistorisches Museum, Basel, Switzerland, where the Gibson Smith Collection is now housed. Roger Portell, Florida Museum of Natural History, Gainesville, Florida, sent the unknown spec- imens from the McGinty Collection for my examination, and Andrew and Greta Murray kindly shared their Chipola material for the new species from those beds. Roland Houart, Landen, Belgium, who loaned me the negatives of his new species of Attiliosa that I might reproduce them, over the years has been a valued collaborator in our joint attempts to bring some small degree of order to the Family Muricidae. LITERATURE CITED ADAMS, A. 1853. Descriptions of several new species of Murex, Rissoina, Planaxis, and Eulima, from the Cumingian Col- lection. Proceedings of the Zoological Society of London for 1851 19:267—272. ApbaAMs, A. 1855. Descriptions of twenty-seven new species of shells from the collection of Hugh Cuming, Esq. Proceed- ings of the Zoological Society of London for 1854 22:311— SHU BELLARDI, L. 1872. I Molluschi dei terreni Terziari del Piemonte e della Liguria; pt. 1. Memorie della Real Accademia delle Scienze di Torino (2)27:1—264, pls. 1-15. [issued separately in 1872; journal issued 10 June 1873] BerRRY, S. S. 1960. Notices of new eastern Pacific Mollusca— IV. Leaflets in Malacology 1(19):115—122. BuLLock, R. C. 1976. On the identity of two supposed fascio- lariid gastropods, Fusus mexicanus Reeve, 1848, and Per- isternia nodulosa A. Adams, 1855. Tulane Studies in Ge- ology and Paleontology 12(3):133-—136, 1 pl. CERNOHORSKY, W. O. 1976. The taxonomy of some Indo-Pacific Mollusca; pt. 4. Records of the Auckland Institute and Mu- seum 13:111—129, 43 figs. CossMANN, A. E. M. & A. PEyroT. 1923. Conchologie néogé- nique de |’ Aquitaine. Actes de la Société Linnéenne de Bor- deaux 75:191—318, pls. 12-18. DALL, W. H. 1889. Reports on the dredging, under the supervi- sion of Alexander Agassiz, in the Gulf of Mexico (1877— 78) and in the Caribbean Sea (1879-80), by the U.S. Coast Survey Steamer “Blake,” Lieut.-Commander C. D. Sigsbee, U.S.N., and Commander J. R. Bartlett, U.S.N., commanding. XXIX. Report on the Mollusca; Part Il. Gastropoda and Sca- phopoda. Bulletin of the Museum of Comparative Zoology, Harvard University 18:1—492, pls. 10—40. DaALL, W. H. 1902. Illustrations and descriptions of new, unfi- The Veliger, Vol. 42, No. 4 gured, or imperfectly known shells, chiefly American, in the U.S. National Museum. Proceedings of the United States National Museum 24(1264):499—566, pls. 27—40. D’Attitio, A. & B. W. Myers. 1986. Clarification of Muricidea caledonica Jousseaume, 1881 (= Attiliosa caledonica). Con- chologists of America Bulletin 14(4):59-62, 10 figs. EMERSON, W. K. 1968. Taxonomic placement of Coralliophila incompta Berry, 1960, with the proposal of a new genus, Attiliosa (Gastropoda:Muricacea). The Veliger 10(4):379— 381, pl. 53, 5 text-figs. Fair, R. H. 1976. The Murex Book, an Illustrated Catalogue of Recent Muricidae (Muricinae, Muricopsinae, Ocenebrinae). Published by the author: Honolulu, Hawaii. 138 pp., 23 pls., 67 text-figs. FISCHER-PIETTE, E. & J. BEIGBEDER. 1943. Catalogue des types de Gasteropodes marins conserves au Laboratoire de Mala- cologie. I—Genus Murex. Bulletin du Muséum National d’Histoire Naturelle, Paris (2)15:203—209. GouLp, A. A. 1849. Shells collected by the U.S. Exploring Ex- pedition under the command of Chas. Wilkes. Proceedings of the Boston Society of Natural History 3:140—144. Houart, R. 1993a. Description of three new species and one new subspecies of Muricidae (Muricinae and Muricopsinae) from West Africa. Bollettino Malacologico 29(1—4):17—30, 28 figs. Houart, R. 1993b. Three new species of Muricinae and Muri- copsinae (Gastropoda:Muricidae) from Somalia, Christmas (Line Islands) and the Philippines Islands. Venus (Japanese Journal of Malacology) 52(1):41—46, 11 figs. Houart, R. 1996. Description of new species of Muricidae (Gas- tropoda) from New Caledonia, the Philippine Islands, the northeast Atlantic, and West Africa. Apex 11(2):59—75, 32 figs. Houart, R. 1998. Description of eight new species of Muricidae (Gastropoda). Apex 13(3):95—109, 47 figs. Humrrey, M. 1975. Sea Shells of the West Indies. Taplinger Publishing Company: New York. 351 pp., 32 pls., 20 text- figs. JOUSSEAUME, F. P. 1881. Diagnoses de mollusques nouveaux. Le Naturaliste 3(44):349—350. JOUSSEAUME, FE. P. 1882. Etude des Purpuridae et description d’espéces nouvelles. Revue et Magasin de Zoologie 3(7): 314-348. MICHELOTTI, G. 1847. Déscription des fossiles des terrains Mio- cénes de |’Italie Septentrionale. Naturkundige Verhandelin- gen van de Hollandsche Maatschappij der Wetenschappen te Haarlem (2)3(2):1—409, pls. 1-17. NoweELL-UstIckE, G. W. 1969. A Supplementary Listing of New Shells. Livingston Publishing Company: Wynnwood, Penn- sylvania. 32 pp., 6 pls. NoweELt-UstIcKE, G. W. 1971. A Supplementary Listing of New Shells (illustrated); revised edition. Published by the author: St. Croix, U.S. Virgin Islands. 32 pp., 6 pls. PetucH, E. J. 1993. Molluscan discoveries from the tropical western Atlantic region. La Conchiglia 25(266):51—56, 11 figs. PorrtER, J. 1883. Revision des Murex du Muséum. Nouvelle Ar- chives du Muséum d’Histoire Naturelle (2)5:13—128, pls. 4—6. RApDWIN, G. E. & A. D’AtTiLio. 1976. Murex Shells of the World; An Illustrated Guide to the Muricidae. Stanford University Press: Stanford, California. 284 pp., 32 pls., 192 text-figs. SaRASUA, H. & J. Espinosa. 1979. El Género Muricopsis (Mol- lusca:Gastropoda) en Cuba. Poeyana, no. 193:1—6, 1 fig. SOWERBY, G. B., JR. 1841a. Conchological Illustrations, pls. 187— E. H. Vokes, 1999 Page 305 199; Murex, a Catalogue of Recent Species. Published by the author: London. 9 pp. Sowerby, G. B., JR. 1841b. Descriptions of some new species of Murex, principally from the collection of H. Cuming, Esq. Proceedings of the Zoological Society of London for 1840 8:137-147. VoKEs, E. H. 1976. Cenozoic Muricidae of the western Atlantic region, pt. 7—Calotrophon and Attiliosa. Tulane Studies in Geology and Paleontology 12(3):101—132, pls. 1-7. VoKEs, E. H. 1988. Muricidae (Mollusca: Gastropoda) of the Esmeraldas beds, northwestern Ecuador. Tulane Studies in Geology and Paleontology 21(1):1—50, pls. 1-6, 15 text— figs., 1 table. VoKEs, E. H. 1989. Neogene paleontology in the northern Do- minican Republic. 8. The Family Muricidae (Mollusca: Gas- tropoda). Bulletins of American Paleontology 97(332):5—94, pls. 1-12, 21 text-figs., 3 tables. VoKEs, E. H. 1992. Cenozoic Muricidae of the western Atlantic region, pt. 9—Prterynotus, Poirieria, Aspella, Dermomurex, Calotrophon, Acantholabia, and Attiliosa; additions and cor- rections. Tulane Studies in Geology and Paleontology 25(1— 3):1—108, pls. 1-20, 10 text-figs. VoKEs, E. H. & A. D’AtTILIO. 1982. Review of the muricid genus Attiliosa. The Veliger 25(1):67—71, 10 figs. LOCALITY DATA The following are Tulane University fossil locality num- bers: 283. Caloosahatchee Fm. and Bermont Fm. mixed, spoil banks on cross-canal 1.3 miles southwest of Port Charlotte Railroad Station (formerly Murdock) on south side of Florida Highway 771 and Seaboard Airline Railroad (Sec. 12, T. 40 S, R. 21 E), Charlotte County, Florida. 458. Chipola Fm., east bank of Chipola River, above Farley Creek (SW % Sec. 20, T. 1 N, R. 9 W), Calhoun County, Florida. 547. Chipola Fm., west bank of Chipola River, about 2000 feet above Fourmile Creek (SW % Sec. 29, T. 1 N, R. 9 W), Calhoun County, Florida. 548. Chipola Fm., west bank of Chipola River, at bend about 1800 feet south of mouth of Farley Creek (NW 4% Sec. 29, T. 1 N, R. 9 W), Calhoun County, Florida. 727. Bermont Fm., borrow pits 2.2 miles east of U.S. Highway 27, 15 miles south of South Bay, Palm Beach County, Florida. 817. Chipola Fm., south side of Tenmile Creek, large gully on the property of Mr. A. Sexton (1967) (SE % Sec. 12, T. 1 N, R. 10 W), Calhoun Coun- ty, Florida. 819. Chipola Fm., Farley Creek, 0.2 mile west of bridge of Florida Highway 275 (SW 4% Sec. 21, T. 1 N, R. 9 W), Calhoun County, Florida. 999. Chipola Fm., Farley Creek, about 300 yards downstream from bridge of Florida Highway 275 (SW % Sec. 21, T. 1 N, R. 9 W), Calhoun County, Florida. 1196. Chipola Fm., Farley Creek, north bank about 0.8 mile east of bridge on Florida Highway 275 (NE % Sec. 21, T. 1 N, R. 9 W), Calhoun County, Florida. 1215. Gurabo Fm., Rio Gurabo, bluffs on both sides, from the ford on Los Quemados-Sabaneta road, upstream to approximately 1 km above the ford, Dominican Republic. 1240. Moin Fm., Barrio Los Corales, top of hill at end of road that passes Standard Fruit Company box factory, 1.8 km north of main highway at Pueblo Nuevo, which is 2 km west of Puerto Limon, Costa Rica. 1399. Esmeraldas beds, Onzole Fm., roadcut on west side of village of Camarones, which is 20 km (by road) east of bridge over Rio Esmeraldas at Esmeraldas, Prov. of Esmeraldas, Ecuador. 1422. Cercado Fm., Arroyo Bellaco, which is tributary of Rio Cana from the east, coral reef that is ex- posed for approximately 1 km below the ford at Los Caobas Adentro, 3 km southwest of Las Caobas, Dominican Republic. The following are Tulane University Recent collecting lo- calities: R-98. Anton Bruun Cruise 10, dredged in 40 meters northwest of Holandes Cay, and east-northeast of Cape San Blas (9°37'N, 78°50.3'W), Pana- ma. R-109. Bahia de las Minas, Isla Payardi, Prov. of Co- lon, Panama (7000 YBP). R-369. Moin Bay, north side of Limoén Peninsula; ma- terial dredged from bay for fill to make oil ter- minal (1976), Moin, Costa Rica. R-487. Trawled by shrimpers off Guaymas, Sonora, Mexico. The following is a Florida Museum of Natural History fossil locality number: UF CROO7. Tamiami Fm., material exposed during construction of “‘Alligator Alley,” 13.3 miles east of Florida Highway 29 (T. 49 S, R. 32 E), Collier County, Florida (TU 797 is same locality). The following is a Naturhistorisches Museum, Basel, Switzerland, fossil locality number: NMB 17530. Mataruca Member, Caujaro Fm., Cemen- terio de Carrizal, Falcon, Venezuela. THE VELIGER CMS, Inc., 1999 The Veliger 42(4):306—337 (October 1, 1999) A Systematic Review of the Hydrobiid Snails (Gastropoda: Rissooidea) of the Great Basin, Western United States. Part IJ. Genera Colligyrus, Eremopyrgus, Fluminicola, Pristinicola, and Tryonia ROBERT HERSHLER Department of Invertebrate Zoology (Mollusks), NHB STOP 118, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, USA Abstract. This second and final part of a taxonomic treatment of hydrobiid snails of the Great Basin region in the western United States (based principally on material collected during a recently completed field survey) focuses on fauna other than the genus Pyrgulopsis. A new genus of small amnicoline snails, Colligyrus, is proposed for Hydrobia greggi Pilsbry, 1935, together with a new species from the Harney Lake basin of Oregon. This group is strongly differentiated from other amnicolines by a unique female genitalic groundplan. New records are provided for three species of Flu- minicola, and two new congeners are described from the northwest Great Basin, both of which had previously been confused with F. turbiniformis (Tryon, 1865). A new genus of cochliopine snails, Eremopyrgus, is erected for a new species from Steptoe Valley, Nevada. Eremopyrgus is distinguished from other cochliopines by unique aspects of its glandular penial lobes and other genitalic features. New records are provided for two species of Tryonia, and a new congener is described from thermal springs in central Nevada. Several new records of Pristinicola hemphilli (Pilsbry, 1890) from the extreme northwest Great Basin are provided. INTRODUCTION In the first part of a systematic review of hydrobiid snails of the Great Basin in the western United States (based principally on material collected during a recently com- pleted field survey), 58 new species belonging to the widespread genus Pyrgulopsis were described, and new records were provided for 10 previously described con- geners (Hershler, 1998). In this second and final part of this review, other hydrobiid groups, which are much more modestly represented in the region, are similarly treated. Novelties described herein include two small species of Fluminicola from the northwest Great Basin previously confused with F. turbiniformis (Tryon, 1865), a species of Tryonia from central Nevada, a new genus of cochlio- pine snails from eastern Nevada, and a new genus of small amnicoline snails from northern segments of the Great Basin. The reader is referred to Hershler (1998) for study ra- tionale and methodology. Institutional repositories of ex- amined specimens are indicated by the following abbre- viations: ANSP, Academy of Natural Sciences, Philadel- phia; CAS, California Academy of Sciences, San Fran- cisco; FMNH, Field Museum of Natural History, Chicago; USNM, former United States National Museum, collections now in National Museum of Natural History, Smithsonian Institution, Washington, D.C. Shell parame- ters for new species are summarized in Table 1. SYSTEMATICS Family HyDRoBIIDAE Troschel, 1857 Colligyrus, Hershler, gen. nov. Type species: Hydrobia greggi Pilsbry, 1935. Also in- cluded is Colligyrus depressus, sp. nov. (described be- low). Etymology: From New Latin, collis, hill or high ground; and gyrus, circle or round. Referring to the upland habitat and coiled shell of these snails. Gender masculine. Diagnosis: A northwestern American amnicoline group having a small, globose to conical shell and paucispiral operculum. Female coiled oviduct simple; glandular ovi- duct large, ventrally closed; bursa copulatrix large, pos- teriorly positioned; seminal receptacles, 2. Description: Shell small (up to 3.3 mm in length), thin, globose to conical, umbilicate. Whorls, 3.5—4.5, convex, narrowly shouldered, sutures impressed. Shell clear to white, periostracum thin. Shell apex nearly flat; proto- conch of about 1.5 whorls, sculptured with weak spiral lineations. Teleoconch smooth except for faint growth lines. Aperture medium-sized, ovate or circular; outer lip thin; parietal lip complete across body whorl, thin; colu- mellar lip sometimes slightly thickened. Umbilicus nar- row to perforate. Operculum flat, thin, ovate, paucispiral. Outer margin of operculum without rim; attachment scar and callus weakly developed. Body pigmentation well de- R. Hershler, 1999 Page 307 Table 1 Selected shell parameters for new species. Data expressed as mean with standard deviation given below. Measurements are given in mm. n = number of specimens, » = mean, SD = standard deviation, SH = shell height, SW = shell width, HBW = height of body whorl, WBW = width of body whorl, AH = aperture height, AW = aperture width, SS = shell width/shell height, WH = number of shell whorls. SH SW Colligyrus depressus USNM 860756 pe 1.98 1.78 n= 15 SD 0.05 0.06 Fluminicola insolitus USNM 860757 be 4.55 3.86 n= 13 SD 0.25 0.18 Fluminicola virginius USNM 874902 WL 3.19 2.85 n= 15 SD 0.13 0.08 Eremopyrgus eganensis USNM 874692 bh 3.44 1.89 n= 15 SD 0.21 0.09 Tryonia monitorae USNM 892046 kL 3.87 1.39 n= 15 SD 0.19 0.07 USNM 874882 kL 3.87 1.38 n= 14 SD 0.28 0.08 veloped. Salivary glands long, simple tubes. Radula rib- bon elongate (ca. 15 times longer than wide), coiled be- hind buccal mass. Cutting edge of central teeth straight or weakly concave, bearing 9-15 short cusps. Central cusp pointed, slightly larger than lateral cusps. Basal cusps, 1—2 (sometimes absent on one side), innermost cusp largest. Lateral margins of central teeth angled about 40° to vertical axis of tooth, slightly thickened, distally expanded, projecting slightly beyond V-shaped base of tooth. Lateral teeth with 2—4 inner cusps and 3—5 outer cusps; basal process well developed. Lateral wing of lat- eral teeth rather broad, somewhat longer than cutting edge. Marginal teeth with relatively numerous cusps; cusps on inner marginals larger than those on outer mar- ginals. Dorsal folds of esophagus short, straight. Cephalic tentacles medium length in preserved material. Ctenidium absent, reduced to a vestige, or well developed, with small, triangular filaments. Osphradium medium-sized, narrow. Hypobranchial gland well developed along rec- tum. Renal organ with prominent pallial bulge. Stomach longer than style sac, posterior caecal appendix absent; anterior stomach chamber larger than posterior chamber. Rectum straight in pallial roof. Cephalo-pedal ganglia weakly pigmented; cerebral and pedal commissures short. Testis large. Prostate gland small, walls of medium thick- ness. Penis small to medium-sized relative to head, bi- furcate. Lobe slightly shorter than filament, arising from inner edge at or near base, usually posteriorly oriented, weakly folded along most of length. Lobe containing weakly coiled duct which enters small muscular sac dis- tally and exits as eversible papilla through cup-shaped HBW WBW AH AW SS WH eg} 152 1.05 0.93 0.90 3.60 0.04 0.03 0.05 0.04 0.03 0.13 3.88 3.13 2.59 2.30 0.85 3.78 0.18 0.17 0.12 0.15 0.03 0.09 2.81 2.08 1.88 1.64 0.89 3.48 0.11 0.08 0.10 0.09 0.03 0.22 235 1.71 1.57 0.96 0.55 4.78 11 0.07 0.12 0.05 0.02 0.23 WAZ 1.34 1.04 0.71 0.36 7.02 0.10 0.07 0.05 0.05 0.01 0.20 1.68 1.30 1.01 0.67 0.36 6.98 0.09 0.08 0.06 0.05 0.02 0.32 opening. Duct exits base of lobe into nuchal cavity, broadening to form a large mass of blindly ending glan- dular loops above the salivary glands; gland lined with thin muscular coat. Penial filament straight or coiled to left, tapering to pointed tip. Penial duct medium width, with thick muscular coat, straight or weakly undulating basally, positioned along outer edge of filament. Females Oviparous. Ovary small. Glandular oviduct consisting of sub-equal albumen and capsule glands. Albumen gland with short pallial component. Coiled oviduct a single, posteriorly arched loop opening to anterior portion of al- bumen gland. Bursa copulatrix medium-sized, positioned posteriorly, partly overlapped by albumen gland. Bursal duct ciliated, short to medium length, originating from anterior edge, opening to oviduct slightly behind pallial wall. Posterior seminal receptacle pouchlike, opening to distal arm of coiled oviduct; anterior seminal receptacle ovate to circular, pressed against ventral edge of albumen gland, opening to oviduct just distal to connection with bursal duct. Capsule gland with narrow, vertical lumen. Sperm tube narrow, separated from capsule gland along most of length, but distally fused to form common genital aperture. Remarks: The small anterior female accessory pouch has a cellular structure similar to that of the posterior seminal receptacle and, although sperm was not seen in sectioned material, the pouch nevertheless is interpreted as a sem- inal receptacle. Colligyrus differs from other amnicolines in that females have three sperm pouches: a posterior bur- sa copulatrix and two small seminal receptacles. In con- Page 308 The Veliger, Vol. 42, No. 4 trast, a group composed of Dasyscias, Lyogyrus, and Parabythinella has but a single, posterior sperm pouch (e.g., Thompson & Hershler, 1991:fig. 11 [Dasyscias franzi Thompson & Hershler, 1991]), which represents a bursa copulatrix based on its columnar cell lining and absence of oriented sperm in its lumen (Hershler, unpub- lished); while a second group composed of Amnicola and Marstoniopsis has a posterior bursa copulatrix and a sin- gle large seminal receptacle (e.g., Hershler & Thompson, 1988:fig. 8 [Amnicola limosa (Say, 1817)]). Although the type species of Colligyrus resembles Lyo- gyrus (to which it was most recently allocated) in its di- minutive conical shell, Colligyrus nevertheless is more similar to Amnicola in female genitalic groundplan, al- though it differs in having two (as opposed to a single) seminal receptacles. Colligyrus and Amnicola have a weaker protoconch microsculpture than Lyogyrus and also share a paucispiral (as opposed to multispiral) oper- culum. The anterior seminal receptacle of Colligyrus may be homologous to that of Amnicola, as both of these sacs open to the oviduct distal to the coiled portion at or near the junction of the bursal duct (the posterior seminal re- ceptacle of Colligyrus opens to the distal arm of the coiled oviduct as in most hydrobiids). Note, however, that these sacs otherwise differ in their size, shape, and po- sition. Colligyrus also differs from Amnicola in having diffuse rather than banded pigment on the dorsal surface of the mantle, a much more elongate radula, narrower central cusps on the central and lateral radular teeth, more numerous cusps on the marginal radular teeth, and a more basal position of the penial lobe. Colligyrus lives in cold springs and spring runs in the upper Snake River basin and northeastern (Bonneville ba- sin) and northwestern (Harney Lake basin) portions of the Great Basin. Colligyrus greggi (Pilsbry, 1935) (Figures 1, 2A—D, 3A, 5) Hydrobia greggi Pilsbry, 1935:93—-94, fig. 2.—Henderson, 1936:138, fig. 9.—Baker, 1964:173.—Beetle, 1957: 19.— Beetle, 1961:5.—Beetle, 1989:639. Amnicola greggi (Pilsbry, 1935), Taylor, 1966b:173.—Tay- lor, 1975:90 (literature compilation).—Turgeon et al., 1988:60. Amnicola (Lyogyrus) greggi (Pilsbry, 1935), Burch & Tot- tenham, 1980:124, figs. 292, 303. Diagnosis: Small to medium-sized, with conical shell. Description: Shell (Figure 1A) conical; height, 1.7—3.3 mm; whorls, 3.75—4.5. Protoconch (Figure 1B, C) of about 1.5 whorls, diameter about 0.49 mm; microsculp- ture of numerous weakly incised spiral lineations. Teleo- conch whorls weakly convex, adapical shoulder often well developed. Shell clear to white, often transparent. Periostracum light brown or tan. Aperture ovate or cir- cular, without adapical angulation. Outer lip weakly pro- socline, thin to moderately thick. Parietal lip narrowly adnate to well separated from body whorl. Umbilicus nar- row to perforate; columellar swelling absent to narrow. Operculum (Figure 1D, E) light amber; outer margin without rim, dorsal surface weakly frilled. Attachment scar slightly thickened all around, especially along inner edge. Callus weakly developed. Radula with about 155 rows of teeth; ribbon length, 1.5 mm, ribbon width, 94 wm; central tooth width, 25 ym. Cutting edge of central tooth (Figure 1F) straight or very weakly indented; lateral cusps, 4—7; median cusp pointed, slightly broader and longer than laterals; basal cusps, 1—2; basal tongue narrowly triangular, even with or slightly shorter than lateral margins, basal sockets moderately excavated. Lateral tooth (Figure 1G) with slightly convex dorsal edge; lateral wing longer (150%) than cutting edge; tooth face taller than broad; central cusp weakly pointed, lateral cusps, 2—4 (inner), 3—4 (out- er). Inner marginal teeth (Figure 1H) with 24—27 cusps; outer marginal teeth (Figure 11) with 25—33 cusps. Tentacles light gray to black. Snout light gray. Foot pale, opercular pale or fringed with medium gray pig- ment. Neck pale or pigmented with scattered gray gran- ules. Pallial roof, visceral coil black. Ctenidium well developed, slightly overlapping peri- cardium; filaments about 17, small, about as tall as wide, weakly pleated. Osphradium about 33% of ctenidium length, positioned centrally or slightly posterior to middle of ctenidium. Ovary slightly less than 0.5 whorl, abutting posterior edge of stomach, filling less than 50% of digestive gland behind stomach. Distal female genitalia shown in Figure 2A, B. Bursa copulatrix about 67% of length of albumen gland; narrowly ovate, horizontal, with about half of length overlapped by albumen gland. Bursal duct short (about 33% of length of bursa copulatrix), distinctly nar- Figure | Scanning electron micrographs of shell, operculum, and radula of Colligyrus greggi Hershler, gen. nov., USNM 883531. A. Shell (height 2.3 mm). B, C. Shell apex. Bars = 200 pm, 160 ym, respectively. D. Operculum, outer surface. Bar = 270 jm. E. Operculum, inner surface. Bar = 285 wm. E Central radular teeth. Bar = 12 pm. G. Lateral radular tooth. Bar = 13 ym. H. Inner marginal tooth. Bar = 13 ym. I. Outer marginal tooth. Bar = 12.5 jum. R. Hershler, 1999 Page 309 Page 310 The Veliger, Vol. 42, No. 4 em — Pst Cov Dag Psr Cov R. Hershler, 1999 rower than bursa copulatrix. Posterior seminal receptacle a small, ovate sac (with short duct) overlapping the bursa copulatrix, sometimes overlapped anteriorly by albumen gland. Posterior seminal receptacle abutting the posterior edge of the coiled oviduct, opening to the postero-ventral edge of this duct. Anterior seminal receptacle a somewhat smaller, nearly circular sac; duct absent or very short. Albumen gland with a weak rectal furrow; furrow absent on capsule gland. Albumen gland with short (20%) pallial component. Capsule gland about as long and wide as al- bumen gland. Anterior portion of sperm tube expanded into vestibule; genital opening a small, terminal pore. Testis 1.25 whorls, filling more than 50% of digestive gland behind stomach. Prostate gland bean-shaped, with about 38% of length in pallial roof. Distal vas deferens a broad, thickened tube, slightly undulating in pallial roof and neck. Penis shown in Figure 2C; tubular gland shown in Figure 2D. Penial lobe arising from base. Penial duct narrowing somewhat in base of penis; undulating basally, otherwise straight. Penis pigmented with scattered inter- nal granules; base sometimes darkly pigmented with mel- anin. Type locality: Cliff Creek canyon, a fork of Hoback Can- yon, about 29 miles (46 km) south of Jackson, Wyoming, in the Snake River drainage. The type locality has not been precisely located. A spring in Cliff Creek canyon harboring this species is shown in Figure 4A. Lectotype, ANSP 163812 (Figure 3A); paralectotypes, ANSP 375735. Baker (1964) separated the single figured spec- imen and identified this as the “type by original mea- surement,’ which is construed as a lectotype designation. Remarks: This snail lives in the upper Snake River basin and northeastern corner of the Great Basin (Bonneville basin) (Figure 5). Records from western Montana (Taylor 1966b:173) require confirmation. Populations assigned to this species vary slightly in shell shape, relationship be- tween aperture and body whorl, and thickness of shell lip. Taylor (1966b:173) reported laminate egg capsules (typical of amnicoline snails) for this species. Material examined: IDAHO. Bannock County: Heart Mtn. Spring, Marsh Valley, T. 13 S, R. 39 E, NW % section 2, USNM 883881. Bear Lake County: spring, Page 311 Right Fork Georgetown Canyon, Bear River drainage, T. 11 S, R. 44 E, NW % section 10, USNM 883522.— spring, Home Canyon, Bear River drainage, T. 12 S, R. 45 E, NW % section 32, USNM 883524. Caribou County: Harris Spring complex, Bear River drainage, T. 11 S, R. 41 E, NE % section 9, USNM 883394.—Kackley Spring, Bear River drainage, T. 10 S, R. 40 E, SW % section 21, USNM 883539.—spring, Kelly Park, Soda Springs, Bear River drainage, T. 9 S, R. 42 E, NW % section 5, USNM 883523. Franklin County: spring, Cottonwood Creek, Bear River drainage, T. 12 S, R. 39 E, NE % section 25, USNM 883392. UTAH. Cache County: China Row Spring, Logan Canyon, Cache Valley, T. 12 N, R. 3 E, NE % section 7, USNM 858288, USNM 883393.— spring, east of Porcupine Reservoir, Cache Valley, T. 9 N, R. 2 E, NW % section 17, USNM 883880. WYO- MING. Lincoln County: spring, Sublette Creek, Bear River drainage, T. 24 N, R. 118 W, NW 4% section 8, USNM 883396.—spring, Salt Creek, Bear River drain- age, T. 29 N, R. 119 W, SW % section 24, USNM 883395. Sublette County: spring, Cliff Creek, Snake River drain- age, T. 38 N, R. 114 W, NW % section 23, USNM 883531. Colligyrus depressus Hershler, sp. nov. Harney Basin duskysnail (Figures 2E, F 3B, 5, 6) Etymology: from New Latin, depressus, meaning pressed down, low, and referring to the squat shell of this species. Diagnosis: Small, with globose to low-conic shell. Description: Shell (Figure 6A) low-conic, rarely with eroded spire; height, 1.9—2.1 mm; whorls, 3.5—3.75. Apex often inclined; protoconch (Figure 6B, C) of 1.4—1.5 whorls, diameter about 0.44 mm; microsculpture of nu- merous weak spiral lineations. Teleoconch whorls con- vex, often markedly so, narrowly shouldered. Shell clear to white, translucent. Periostracum tan. Aperture medium- sized, ovate, weakly angled adapically. Outer lip proso- cline, weakly sinuate (adapically advanced). Parietal lip narrowly adnate to or slightly separated from body whorl; columellar lip without swelling. Umbilicus perforate. Figure 2 Genitalia of Colligyrus Hershler, gen. nov., species (A—-D, C. greggi, USNM 883531; E, EK C. depressus Hershler, gen. & sp. nov., USNM 860756). A. Left side of female glandular oviduct and associated structures. Bar = 0.5 mm. B. Bursa copulatrix and associated structures. Scale as in “A.” C. Dorsal surface of penis. Bar = 0.25 mm. D. Tubular gland (coiled within cephalic haemocoel). Scale as in “‘C.” E. Left side of female glandular oviduct and associated structures. Bar = 0.25 mm. EF Bursa copulatrix and associated structures. Scale as in “‘E.” Ag, albumen gland; Asr, anterior seminal receptacle; Bu, bursa copulatrix; Cg, capsule gland; Cov, coiled oviduct; Dag, opening of oviduct to albumen gland; Ga, genital aperture; Pd, penial duct; Psr, posterior seminal receptacle; Pw, posterior wall of pallial cavity; St, sperm tube. Pace 3i2 Figure 3 Type material of Great Basin species of Colligyrus Hershler, gen. nov., Eremopyrgus Hershler, gen. nov., and Tryonia. A. C. greggi, lectotype, ANSP 163812 (shell height 2.6 mm). B. C. depressus Hershler, gen. & sp. nov., holotype, USNM 883876 (1.7 mm). C. E. eganensis Hershler, gen. & sp. nov., holotype, USNM 874692 (3.1 mm). D. T. monitorae Hershler, sp. nov., holotype, USNM 892046 (3.0 mm). The Veliger, Vol. 42, No. Operculum (Figure 6D, E) brown in nuclear region, otherwise clear; outer margin without rim. Attachment scar margin slightly thickened along inner edge. Callus weakly developed in nuclear region. Radula with about 170 rows of teeth; ribbon length, 1.4 mm, ribbon width, 88 wm; central tooth width, 22 um. Cutting edge of central tooth (Figure 6F) weakly to moderately indented; lateral cusps, 6—7; median cusp rounded or weakly pointed, slightly broader and longer than laterals; basal cusps, 1, sometimes absent on one side; basal tongue slightly shorter than lateral margins, basal sockets deeply excavated. Lateral tooth (Figure 6G) with horizontal or slightly convex cutting edge; lateral wing slightly longer (125%) than cutting edge; tooth face Figure 4 Type and other localities for species treated herein. A. Spring, Cliff Creek, Snake River drainage, Sublette County, Wyoming. Habitat of Colligyrgus greggi Hershler, gen. nov. in vicinity of type locality. Photograph, September, 1993. B. Springs, Cricket Creek, Silvies River drainage, Harney County, Oregon. Type locality of C. depressus Hershler, gen. & sp. nov. Photograph (D. Sada), July, 1994. C. Page Springs, Donner und Blitzen River drainage, Harney County, Oregon. Type locality of Fluminicola insolitus Hershler, sp. nov. Photograph (D. Sada), July, 1993. D. “Waterfall” spring, source of Hardscrabble Creek, Pyramid Lake basin, Washoe County, Nevada. Type locality of F. virginius Hershler, sp. nov. Photograph (G. Vinyard), October, 1992. E. Spring northwest of Clark Spring, Steptoe Valley, White Pine County, Nevada. Type locality of Eremopyrgus eganensis Hershler, gen. & sp. nov. Photograph, June, 1992. F Hot Springs, Potts Ranch, Monitor Valley, Nye County, Nevada. Type locality of 7. monitorae Hershler, sp. nov. Photograph (D. Sada), November, 1992. G. Dianas Punch Bowl (Hot Springs), Monitor Valley, Nye County, Nevada. Habitat of 7. monitorae Hershler, sp. nov. Photograph (D. Sada), November, 1992. R. Hershler, 1999 Page 313 Rha op Kee ZN EAN) Page 314 0 40 80 120 160 200 km x Colligyrus depressus e Colligyrus greggi 4 Pristinicola hemphilli The Veliger, Vol. 42, No. 4 Figure 5 Map of northern Great Basin and adjacent regions showing the distribution of Colligyrus Hershler, gen. nov. species and Pristinicola hemphilli. Previously reported localities for Pristinicola hemphilli (see Hershler et al., 1994) are not shown. taller than broad; central cusp rounded, lateral cusps, 3 (inner), 4—5 (outer). Inner marginal teeth (Figure 6H) with 26—30 cusps; outer marginal teeth (Figure 61) with 25-29 cusps. Snout, tentacles, foot light to medium gray. Inner edge of opercular lobe black. Neck unpigmented to medium gray. Pallial roof, visceral coil medium gray to black. Ctendium absent or represented by a few (3-6) small, stubby vestiges; branchial vessel not seen in dissection. Osphradium anteriorly positioned, filling about 20% of pallial cavity length. Ovary about 0.5 whorl, abutting or slightly overlapping posterior stomach chamber, filling less than 50% of di- gestive gland behind stomach. Distal female genitalia shown in Figure 2E, E Bursa copulatrix about 50% of albumen gland length; ovate or clublike, horizontal or obliquely oriented, about 50% overlapped by albumen gland. Bursal duct medium length (50-67% of bursa length), often narrow, sometimes poorly distinguished from bursa. Posterior seminal receptacle small, without duct, overlapping anterior half of bursa copulatrix, posi- tioned near ventral edge of albumen gland. Portions of coiled oviduct adjacent to posterior seminal receptacle of- ten filled with sperm. Anterior seminal receptacle disc- shaped. Albumen gland with weak rectal furrow. Albu- men gland with short (about 22%) pallial section; capsule gland entirely pallial, composed of two distinct glandular units. Genital opening a small terminal slit. Testis 1.0—1.25 whorl, broadly overlapping stomach chambers, filling about 50% of digestive gland behind stomach. Prostate gland ovate, entirely visceral. Pallial vas deferens a broad tube without bends or undulations; Figure 6 Scanning electron micrographs of shell, operculum, and radula of Colligyrus depressus Hershler, gen. & sp. nov. USNM 860756. A. Shell (height 1.8 mm). B, C. Shell apex. Bars = 170 wm, 188 wm, respectively. D. Operculum, outer surface. Bar = 230 xm. E. Operculum, inner surface. Bar = 240 pm. EF Central radular teeth. Bar = 10 pm. G. Lateral radular tooth. Bar = 13 ym. H. Inner marginal tooth. Bar = 12.5 wm. I. Outer marginal teeth. Bar = 11.5 pm. R. Hershler, 1999 Page 315 Page 316 The Veliger, Vol. 42, No. 4 portion of vas deferens in neck straight. Penial lobe aris- ing slightly distal to base. Penial duct narrow throughout. Penis unpigmented. Type locality: Unnamed springs, Cricket Creek, Silvies River drainage, Harney County, Oregon, T. 21 S, R. 28 E, NW % section 12. The type locality is composed of a series of small, cold rheocrenes (11°C, 81 micromhos/cm) (Figure 4B). Holotype, USNM 883876 (Figure 3B); para- types, USNM 860756. Remarks: Colligyrus depressus is thus far known only from the type locality in southeast Oregon (Figure 5). This species differs from C. greggi in its broader shell, absence or vestigial nature of ctenidium, more distal po- sition of penial lobe, longer bursal duct, discoidal shape of the anterior seminal receptacle, division of capsule gland into two distinct units, and slitlike female genital aperture. Material examined: OREGON. Harney County: springs, Cricket Creek, Silvies River drainage, USNM 860756, USNM 883876. Fluminicola Carpenter, 1864 Type species: Paludina nuttalliana Lea, 1838; original designation. Diagnosis: A morphologically diverse group of north- western North American lithoglyphine snails. Remarks: Fluminicola and its species recently were re- viewed by Hershler & Frest (1996). This genus, as cur- rently constituted, is paraphyletic (Hershler & Frest, 1996:fig. 3), but a confident resolution of its systematics must await a more complete study of the type species, for which anatomical material is not available and which may now be extinct owing to urban development along the lower reach of the Willamette River (type locality area). Fluminicola coloradensis Morrison, 1940 (Figures 7, 8A—C) Fluminicola fusca (Haldeman, 1841), Binney, 1865:92 (in part).—Call, 1884:21 (in part).—Pilsbry, 1899:123 (in part).—Hannibal, 1912:187 (in part).—Henderson, 1924:192 (in part).—Chamberlin & Jones, 1929:180— 181, fig. 84 (in part; numerous Utah localities)—Hen- derson, 1936:139.—Jones, 1940:41 (in part).—Baily & Baily, 1951:50 (in part).—Fluminicola seminalis (Hinds, 1842), Chamberlin & Jones, 1929:179—180 (in part). Fluminicola coloradoense Morrison, 1940:125.—Hershler & Frest, 1996:8-9, figs. 1A-F 2, 4A, SA—-D, 6A, 7A, 8A-C, 9A, 10A, 11. Lithoglyphus hindsi (Baird, 1863), Taylor, 1966a:fig. 14 (in part).—Taylor, 1985:306 (Green River).—Taylor & Bright, 1987:249 (in part). Fluminicola hindsi (Baird, 1863), Burch & Tottenham, 1980:102 (in part). Diagnosis: Large, with subglobose to broadly conical shell. Female bursa copulatrix pyriform, with medium length duct. Type locality: Green River, Wyoming (not subsequently restricted). Holotype, USNM 526631. Remarks: Fluminicola coloradensis is the large, globose- shelled species with a prominent adapical shoulder and thickened lip that lives in large springs and streams in the Green River and Bonneville basins. This snail was typically referred to as F. fusca or F. hindsi in the early literature, but in a recent review of the genus (Hershler & Frest, 1996), F. coloradensis was shown to be distinct from F. fuscus (junior synonym, Amnicola hindsi Baird, 1863), which lives in the lower Snake River and Colum- bia River basins. Hershler & Frest (1996) conservatively attributed F. coloradensis solely to the Green River basin, but additional study has shown that Bonneville basin ma- terial conforms to this species (e.g., Figure 8A, B) as earlier suggested by Morrison (1940:125). Variation among populations of F. coloradensis generally is minor, involving slight differences in shell shape, development of an adapical shoulder on teleoconch whorls, thickness of the inner lip, and penis size. Although a few popula- tions closely similar to F. coloradensis are found in the middle portion of the Snake River basin (e.g., Little Wood River), confident confirmation of conspecificity of these with F. coloradensis is beyond the scope of this study. Material from springs in Malad Valley (in the northeast segment of the Bonneville basin), and adjacent portions of the Snake River basin (Arbon Valley, Portneuf River drainage) have a distinctively purple shell, and a higher spire and thinner lip than in characteristic F. coloradensis (Figure 8C). Anatomically this snail is entirely consistent with F. coloradensis, and historical samples of shells from the Malad River also closely conform to typical F. coloradensis. Although I treat the Malad Valley material herein as F. coloradensis, this problem merits further study. Chamberlin & Jones (1929) suggested that a sec- ond species of Fluminicola (identified by them as F. sem- inalis) lives in Utah, but material that I have seen from the localities that they referenced (Utah Lake, Tooele Val- ley) conforms to F. coloradensis. Material examined: IDAHO. Bear Lake County: Bear Lake, Fish Haven, USNM 715883 (subfossil).—Caribou County: Bear River, Soda Springs, USNM 526730.— Kackley Spring, Gem Valley, T. 10 S, R. 40 E, SW % section 21, USNM 883540, USNM 883700, USNM 883890.—spring, southeast of Kackley Spring, Gem Val- ley, T. 10 S, R. 40 E, NW % section 27, USNM 883698. Oneida County: Big Malad Spring, USNM 883479, USNM 883892.—Little Malad Spring, Malad Valley, T. 12 S, R. 34 E, SW % section 14, USNM 883391, USNM R. Hershler, 1999 Page 317 1 Bap ( M2 LN 80 120 160 200 km pon i Oa Fluminicola coloradensis Fluminicola insolitus Fluminicola turbiniformis A v * Fluminicola modoci e # Fluminicola virginius Figure 7 Map of northern Great Basin and adjacent regions showing the distribution of Fluminicola species. The type locality of F. modoci and previously reported localities for F. coloradensis in the Green River drainage (see Hershler & Frest, 1996) are not shown. 883887. UTAH. Elder County: Malad River, ANSP 62606, FMNH 224425, USNM 47873. Cache County: Blacksmith Fork, below Ballard Springs, Cache Valley, FMNH 178420.—Blacksmith Fork, Cache Valley, T. 10 N, R. 2 E, NE % section 9, USNM 883855, USNM 883861.—Murray Spring, Cache Valley, T. 10 N, R. 1 W, SE % section 9, USNM 883475, USNM 883863. Morgan County: East Canyon Creek, Weber River drainage, T. 2 N, R. 3 E, NW % section 35, USNM 883854, USNM 883858.—Weber River, HWY 84, south of Peterson, T. 4 N, R. 2 E, NE % section 6, USNM 874068, USNM 883280, USNM 883862. Rich County: Bear Lake, west shore, FMNH 179555.—Bear Lake, FMNH 178364.— Bear Lake, east shore, FMNH 178365. Salt Lake County: Salt Lake City, USNM 519988.—spring, south of Riv- erton, Jordan River drainage, T. 4 S, R. 1 W, NW 4% sec- tion 5, USNM 883241, USNM 883286, USNM 883859. Tooele County: Tooele Valley, FMNH 178414.—Cotton- wood Creek, Holladay, Jordan River drainage, FMNH 178361. Utah County: Spring Creek, Utah Lake drainage, T.5S,R. 1 E, SW % section 15, USNM 883242, USNM 883860.—Utah Lake, ANSP 27772, ANSP 365332, FMNH 224328, FMNH 224330, USNM 9222, USNM 75452, USNM 31270.—Utah Lake, west shore, FMNH 178857.—Utah Lake, near Saratoga, FMNH 178394, FMNH_ 179556.—Provo Canyon, above Vivian Park, Utah Lake drainage, FMNH 178355, FMNH 178367. Wasatch County: Provo River, below Charleston, FMNH 179221, FMNH 179222. Weber County: 0.8 km below Ogden Canyon, FMNH_ 178391.—just outside Ogden Canyon, Ogden, ANSP 144614.—entrance to Ogden Canyon, Ogden, ANSP 145845.—Ogden River, FMNH 178396. WYOMING. Green River, USNM 526631. Lin- coln County: Smiths Fork, Bear River drainage, T. 24 N, R. 119 W, NE % section 5, USNM 883902. Uinta County: Bear River, HWY 89, northwest of Evanston, T. 16 N, R. Page 318 The Veliger, Vol. 42, No. 4 R. Hershler, 1999 121 W, NW % section 13, USNM 883525, USNM 883864, USNM 883865. Fluminicola insolitus, Hershler, sp. nov. Donner und Blitzen pebblesnail (Figures 7, 8D, 9, 1OA—C) Lithoglyphus turbiniformis (Tryon, 1865), Taylor, 1966a, fig. 9 (in part).—Taylor, 1985:309 (in part; ““headwaters of the Donner und Blitzen River’). Etymology: From New Latin insolitus, meaning unusual or uncommon, and referring to the divergent aspect of this species. Diagnosis: Medium-sized with trochoidal shell. Female bursa copulatrix ovate, with medium length duct. Description: Shell (Figure 9A) trochoidal, rarely with eroded spire; height, 3.6—4.2 mm; whorls, 3.75—4.0. Pro- toconch (Figure 9B) of 1.5 whorls, diameter about 0.76 mm; microsculpture of numerous weak spiral striae. Te- leoconch whorls convex, usually evenly rounded, rarely shouldered. Microsculpture of well-developed collabral growth lines and weak, often eroded, spiral striae. Perios- tracum olive. Shell opaque, dark gray or purple. Aperture large, lunate, weakly angled adapially. Outer lip prosoc- line, thin. Parietal lip thin, complete across body whorl, adnate. Columellar swelling broad, covering near entirety of umbilical region. Shell usually anomphalous, rarely narrowly umbilicate. Operculum (Figure 9C, D) thin, light amber, ovate, paucispiral; outer margin without rim. Attachment scar margin slightly thickened all around. Callus weakly de- veloped. Radula with about 80 rows of teeth; ribbon length, 2.4 mm, ribbon width, 160 wm; central tooth width, 62 pm. Cutting edge of central tooth (Figure 9E, F) weakly in- dented; lateral cusps, 3—6; median cusp narrow U-shaped, slightly broader and longer than laterals; basal cusps ab- sent; basal tongue broad, extending below lateral margins, basal sockets weakly excavated. Lateral margins angled about 50° relative to vertical axis of teeth; margins nar- row, thin. Lateral tooth (Figure 9G) with slightly convex cutting edge. Lateral wing 60% of length of cutting edge; tooth face broader than tall; central cusp U-shaped, lateral cusps, 3—5 (inner), 4—6 (outer). Inner marginal teeth (Fig- ure 9H) with 16-19 cusps; outer marginal teeth (Figure Page 319 91) with13—23 cusps. Cusps on inner marginal teeth larger than those on outer marginals. Salivary glands long, un- pigmented. Stomach longer than style sac. Snout, tentacles, foot dark brown or black. Bases of tentacles around eyes sometimes pale. Neck light gray. Pallial roof and visceral coil dark brown or black, pig- ment somewhat lighter on gonads and genital ducts; pal- lial edge black. Ctenidium positioned slightly anterior to pericardium; filaments about 21, small, taller than wide, without pleats. Osphradium about 33% of ctenidium length. Hypobran- chial gland without anterior swelling. Renal organ with prominent (45%) pallial portion; renal opening simple. Cephalo-pedal ganglia weakly pigmented. Ovary 0.25—0.5 whorl, abutting or slightly posterior to edge of stomach, filling less than 50% of digestive gland behind stomach. Distal female genitalia shown in Figure 10A, B. Coiled oviduct posterior-oblique; proximal arm strongly kinked or with small coil; distal arm swollen with sperm. Coiled oviduct and bursal duct join just be- hind pallial wall (slightly behind posterior edge of cap- sule gland). Bursa copulatrix about 60% of albumen gland length; ovate, transversely oriented, about 50% overlapped by albumen gland. Bursal duct medium length (ca. 50% of bursa length), narrow, originating near ven- tral edge of bursa copulatrix. Seminal receptacle much smaller than bursa copulatrix, positioned just anterior to bursa copulatrix along posterior edge of albumen gland, completely overlapped by albumen gland. Albumen gland with deep rectal furrow; furrow weakly developed on capsule gland. Albumen gland without pallial component; capsule gland with short visceral section. Capsule gland about as long as, but narrow than albumen gland; capsule gland folded over to the right. Ventral channel without anterior vestibule. Genital opening a small terminal slit. Testis 1.0—1.25 whorl, overlapping posterior stomach chamber, filling more than 50% of digestive gland length behind stomach. Prostate gland with 33% of length in pallial roof. Pallial vas deferens with weak proximal kink; portion of vas deferens in neck straight. Penis (Figure 10C) medium to large, broad sickle shape, curved, with- out folds; base sometimes slightly narrowed, medial sec- tion without taper; distal section rounded, with short, nar- row papillalike tip. Penial duct near central, undulating throughout (more pronounced distally). Penis pale or with light dusting of melanin proximally. Figure 8 Type and other shell material for Great Basin species of Fluminicola. A. F. coloradensis, ANSP 27772 (shell height 8.1 mm). B. F. coloradensis, USNM 883280 (8.5 mm). C. F. coloradensis, USNM 883479 (8.5 mm). D. F. insolitus Hershler, sp. nov., holotype, USNM 883466 (3.7 mm). E. F. turbiniformis, USNM 858249 (4.8 mm). EF F. turbi- niformis, USNM 883527 (2.0 mm). G. F. turbiniformis, USNM 858241 (3.2 mm). H. F. virginius Hershler, sp. nov., holotype, USNM 874902 (2.7 mm). The Veliger, Vol. 42, No. 4 R. Hershler, 1999 Type locality: Page Springs, Donner und Blitzen River drainage, Harney County, Oregon, T. 32 S, R. 324% E, NW % section 17. A small rheocrene (11°C, 89 micromhos/ cm) draining west to the Donner und Blitzen River (Fig- ure 4C). This site did not appear disturbed when visited in 1993. Holotype, USNM 883466 (Figure 8D), collected by D. W. Sada, 8 July 1993; paratypes, USNM 860757. Remarks: This snail, endemic to the type locality (Figure 7), is unique in the genus in having a very broad basal process of the central radular teeth and lacking basal cusps on these teeth. Fluminicola insolitus most closely resembles F. turbiniformis, which also lives in the north- west Great Basin (see below) and with which it was pre- viously confused (Taylor, 1966a, 1985), but further differs from this species in the purple tint of its shell, thinner shell parietal lip, stouter lateral radular teeth, and stouter bursa copulatrix with longer duct. Material examined: OREGON. Harney County: Page Springs, Donner und Blitzen River drainage, USNM 860757, USNM 883192, USNM 883466. Fluminicola modoci Hannibal, 1912 (Figure 7) Fluminicola modoci Hannibal, 1912:187, pl. 8: fig. 30.— Turgeon et al., 1988:60.—Hershler & Frest, 1996:13- 14, figs. 4E SH—-J, 6E, 7D, 8J—L, 9D, 10D, 11. Lithoglyphus modoci (Hannibal, 1912), Taylor, 1975:125 (literature compilation). Diagnosis: Small with broadly conical shell. Female bur- sa copulatrix sub-globose, with short duct. Type locality: Fletchers Spring, south end, Goose Lake, California. Lectotype, CAS 60798. Remarks: Hershler & Frest (1996) discussed problems with the types and identity of this species. Dry Creek discharges into the northwestern side of Goose Lake (Fig- ure 7) about 32 km north of the type locality area. Snails from the Dry Creek spring have broad, typically decollate shells closely conforming to F. modoci. Material examined: CALIFORNIA. Modoc County: Fletchers Spring, south end, Goose Lake, CAS 60798. OREGON. Lake County: spring, source of Dry Creek, Goose Lake basin, T. 40 S, R. 17 E, SE % section 36, USNM 883554, USNM 883558. Page 321 Fluminicola turbiniformis (Tryon, 1865) (Figures 7, 8E—G) Amnicola turbiniformis Tryon, 1865:219, pl. 22: fig. 5. Fluminicola turbiniformis (Tryon, 1865), Baker, 1964: 177.—Turgeon et al., 1988:61.—Hershler & Frest, 1996:16-17, figs. SK, L, 6H, 9F 13D, 14, 16B, 17D-F 18B. Lithoglyphus turbiniformis (Tryon, 1865), Taylor, 1966a:24, fig. 9 (in part).—Taylor, 1975:197 (in part; literature compilation).—Taylor, 1985:309 (in part). Diagnosis: Small with ovate to narrow-conic shell. Fe- male bursa copulatrix ovate, with short duct. Type locality: west side of Steens Mountain, Catlow Val- ley, Oregon. The type locality has not been precisely lo- cated. Lectotype, ANSP 27779. Remarks: This species ranges widely throughout the northwest Great Basin, from Lake Abert basin east to Quinn River basin and south to Walker River basin (Fig- ure 7). This range conforms in part to that depicted by Taylor (1966a:fig. 9), although populations in the Donner und Blizten River drainage and the eastern side of the Pyramid Lake basin are herein described as new species, and I do not recognize F. turbiniformis in the Sacramento and Columbia River basins. I am also unable to confirm presence of F. turbiniformis in the Deschutes River drain- age as reported by Taylor (1985:309). Populations of F. turbiniformis vary in size, relative shell height, width of columellar lip, and excavation of umbilical region (Figure 8E-G); and also in size of penis relative to head. Several populations in the Smoke Creek Desert and Honey Lake basin in northeast California are distinguished by their especially large size, rather globose, frequently decollate shell with thin shell lip (Figure 8G), but intergradation with more typical F. turbiniformis is apparent. Material examined: CALIFORNIA. Alpine County: spring, Monitor Creek, Carson River basin, T. 9 N, R. 21 E, section 3, USNM 854752.—springs, northeast side of Bagley Valley, Carson River basin, T. 9 N, R. 21 E, NE % section 15, USNM 858240. Lassen County: spring, Old Marr Ranch, Duck Flat, T. 37 N, R. 17 E, NE % section 31, USNM 858252.—springs, west of Dairy Spring, Grasshopper Valley, T. 34 N, R. 11 E, SW % section 4, USNM 858248.—spring, Painters Creek, Madeline Plains, T. 34 N, R. 17 E, NW % section 30, USNM 858250.—springs, Cold Spring Valley, Madeline Plains, Figure 9 Scanning electron micrographs of shell, operculum, and radula of Fluminicola insolitus Hershler, sp. nov., USNM 860757. A. Shell (height 3.5 mm). B. Shell apex. Bar = 300 pm. C. Operculum, outer surface. Bar = 428 pm. D. Operculum, inner surface. Bar = 444 wm. E, E Central radular teeth. Bars = 26 wm. G. Lateral radular tooth. Bar = 20.5 pm. H. Inner marginal tooth. Bar = 23 wm. I. Outer marginal teeth. Bar = 26 pm. Page 322 The Veliger, Vol. 42, No. Figure 10 Genitalia of Fluminicola species (A—C, F. insolitus Hershler, sp. nov., USNM 860757; D-E F. virginius Hershler, sp. nov., USNM 874103). A. Left side of female glandular oviduct and associated structures. Bar = 0.5 mm. B. Bursa copulatrix and seminal receptacle. Scale as in “‘A.”’ C. Dorsal surface of penis. Scale as in ‘‘A.” D. Left side of female glandular oviduct and associated structures. Bar = 0.5 mm. E. Bursa copulatrix and seminal recep- tacle. Scale as in “‘D.”’ E Dorsal surface of penis. Bar = 0.25 mm. Ag, albumen gland; Bu, bursa copulatrix; Cg, capsule gland, Cov, coiled oviduct; Ga, genital aperture; Pd, penial duct; Pw, posterior wall of pallial cavity; Sr, seminal receptacle; Vc, ventral channel of capsule gland. R. Hershler, 1999 T. 36 N, R. 16 E, SW % section 18, USNM 858251.— Bailey Creek, Madeline Plains, T. 34 N, R. 12 E, NE% section 15, USNM 858249.—spring, south of HWY 36 ca. 4.8 km west of Susanville, Susan River basin, T. 30 N, R. 11 E, SW % section 35, USNM 858244.—Five Springs, Honey Lake basin, T. 31 N, R. 16 E, NE % section 23, USNM 858247.—spring, Willow Creek, Wil- low Creek Valley, T. 32 N, R. 11 E, SE % section 35, USNM 874933.—Shoals Creek, Horse Lake basin, T. 32 N, R. 13 E, NW % section 6, USNM 858245, USNM 858246.—Tule Patch Spring, Honey Lake basin, T. 32 N, R. 15 E, SE % section 10, USNM 854468, USNM 858256, USNM 873394.—spring, east of Sage Hen Spring, Smoke Creek, T. 33 N, R. 16 E, SW % section 25, USNM 874054.—springs, Shinn Ranch, Smoke Creek, T. 33 N, R. 16 E, SW % section 36, USNM 858257, USNM 874100.—Big Spring, Smoke Creek, T. 33 N, R. 16 E, NW % section 3, USNM 858260, USNM 858338.—spring, southwest of Sage Hen Spring, Smoke Creek, T. 33 N, R. 16 E, NE % section 35, USNM 858258.—Sage Hen Spring, Smoke Creek, T. 33 N, R. 16 E, NE % section 35, USNM 858259. Modoc County: springs, 1.1 km north of Fandanga Pass turnoff, Surprise Valley, T. 46 N, R. 16 E, NE % section 31, USNM 858255.—Von Riper Spring, Surprise Valley, T. 39 N, R. 16 E, NW % section 25, USNM 858253.—spring, south- west of Von Riper Spring, Surprise Valley, T. 39 N, R. 16 E, NW % section 36, USNM 858254. Mono County: spring, Silver Creek, Pickel Meadow, West Walker River basin, T. 6 N, R. 22 E, NW % section 24, USNM 873406.—springs, east side West Walker River, T. 6 N, R. 23 E, NE % section 9, USNM 873361.—springs, west side Little Walker River, T. 6 N, R. 23 E, SW % section 15, USNM 873412.—springs, southeast corner Slinkard Valley, West Walker River basin, T. 8 N, R. 22 E, SW % section 15, USNM 873346. Nevada County: spring, Sage- hen Creek, Little Truckee River basin, T. 18 N, R. 15 E, section 1, USNM 892040.—spring, Sagehen Creek, Little Truckee River basin, T. 18 N, R. 15 E, section 3, USNM 892042.—spring, Sagehen Creek, Little Truckee River basin, T. 18 N, R. 16 E, section 6, USNM 892041.—Boca Spring, Truckee River basin, T. 18 N, R. 17 E, NE % section 10, USNM 858242. Sierra County: spring, east side HWY 89, south of Kyburz Flat, Little Truckee River basin, T. 19 N, R. 16 E, NW ¥% section 29, USNM 858241.—spring, Hoke Valley, Little Truckee River ba- sin, T. 19 N, R. 17 E, SW % section 2, USNM 858243, USNM 858357. NEVADA. Humboldt County: North Hell Creek Spring, Virgin Valley, T. 44 N, R. 24 E, NW % section 3, USNM 873213, USNM 873226, USNM 874218.—Boulder Spring, Virgin Valley, T. 44 N, R. 25 E, SW % section 19, USNM 874049.—The Dip (spring), Hell Creek, Virgin Valley, T. 44 N, R. 24 E, SW % section 9, USNM 873207.—spring, Cherry Gulch, Bog Hot Val- ley, T. 45 N, R. 29 E, SW % section 8, USNM 883933.— spring, Fivemile Flat, Summit Lake basin, T. 43 N, R. 25 Page 323 E, SE % section 35, USNM 892030.—spring, Antelope Creek, T. 46 N, R. 30 E, NW % section 28, USNM 874212.—Antelope Springs, Soldier Meadow, T. 41 N, R. 25 E, NW % section 32, USNM 883526.—spring, Bartlett Creek, Black Rock Desert, T. 41 N, R. 27 E, SE % section 5, USNM 883901.—spring, Virgin Creek, Quinn River Valley, T. 46 N, R. 30 E, NE % section 34, USNM 874213, USNM 874738. Lyon County: spring, upper Illinois Canyon, Carson River basin, T. 14 N, R. 22 E, SE % section 7, USNM 874903, USNM 883527.— spring, Sweetwater Mountains, West Walker River basin, T. 8 N, R. 24 E, SE % section 27, USNM 854627 (south- ern spring), USNM 854628 (middle spring). Washoe County: spring, South Catnip Creek, Guano Valley, T. 46 N, R. 22 E, NE % section 24, USNM 874206.—spring, Wall Creek, 4.8 km above reservoir, Duck Flat, T. 38 N, R. 20 E, NE % section 6, USNM 854706.—spring, north- east of Middle Lake, Long Valley, T. 46 N, R. 21 E, center section 30, USNM 854753.—spring, near south- east corner of Boulder Reservoir, Boulder Flat, T. 40 N, R. 20 E, NE % section 32, USNM 874186.—spring, east side of Hays Canyon Range, Boulder Flat, T. 40 N, R. 19 E, SW % section 32, USNM 874266.—spring, south of Garden Lake, Duck Flat, T. 35 N, R. 18 E, SW % section 12, USNM 883927.—Clear Creek, Granite Mountain, Black Rock Desert, T. 34 N, R. 22 E, NE % section 26, USNM 874290.—Clear Creek, Granite Mountain, T. 34 N, R. 22 E, section 26, USNM 854066.—spring (west of road), Bog Hog Ranch Creek, High Rock basin, T. 38 N, R. 23 E, NE % section 19, USNM 874209.—spring (east of road), Bog Hog Ranch Creek, High Rock basin, T. 38 N, R. 23 E, NW % section 20, USNM 874199.—-spring, 3.2 km north-northwest of Little High Rock Reservoir, High Rock basin, T. 39 N, R. 23 E, NW % section 30, USNM 874219.—Cottonwood Spring, High Rock basin, T. 43 N, R. 24 E, SE % section 30, USNM 874215. OREGON. Harney County: west side of Steens Moun- tains, ANSP 27779.—Roaring Springs, Catlow Valley, T. 33 S, R. 30 E, NE % section 31, USNM 883470.— Willow Spring, Catlow Valley, T. 36 S, R. 29 E, NW % section 30, USNM 892029. Lake County: spring near source of Guano Creek, Guano Valley, USNM 883560.—spring, Dairy Creek, Chewaucan River drainage, T. 36 S, R. 17 E, SE % section 12, USNM 883565.—spring, Dairy Creek, Chewaucan River drainage, T. 36 S, R. 17 E, SE % section 11, USNM 883553.—Moss Spring, Lake Abert basin, T. 36S, R. 19 E, SE % section 5, USNM 883882.— springs, Deep Creek Falls, Warner Valley, T. 39 S, R. 23 E, NE % section 23, USNM 883552. Fluminicola virginius Hershler, sp. nov. Virginia Mountains pebblesnail (Figures 7, 8H, 1OD—-E 11) Lithoglyphus turbiniformis (Tryon, 1865), Taylor, 1966a, fig. 9 (in part). Page 324 Etymology: Referring to endemism of this snail in the Virginia Mountains west of Pyramid Lake. Diagnosis: Medium-sized with trochoidal to low conic shell. Female bursa copulatrix ovate-pyriform, with short duct. Description: Shell (Figure 11A) trochoidal to low conic, rarely with eroded spire; height, 2.9—3.4 mm; whorls, 3.25—3.75. Protoconch (Figure 11B, C) of 1.7 whorls, di- ameter about 0.50 mm; microsculpture of numerous spiral striae, often stronger close to periphery. Teleoconch whorls convex, often having peripheral angulation; shoul- der well developed, often forming pronounced, rounded keel. Aperture and last 0.25—0.50 whorl disjunct. Micro- sculpture of well-developed collabral growth lines. Per- iostracum tan. Shell clear, translucent. Aperture large, ovate, angled above. Outer lip weakly prosocline, some- what thickened. Parietal lip complete across body whorl, disjunct, thick. Columellar swelling broad; columellar lip thick. Shell anomphalous or narrowly umbilicate. Operculum (Figure 11D, E) thin, light amber, ellipsoi- dal, paucispiral, nucleus highly eccentric; outer margin without rim. Attachment scar margin slightly thickened along inner edge. Callus weakly developed. Radula with about 85 rows of teeth; ribbon length, 885 ym; ribbon width, 115 wm; central tooth width, 28 wm. Cutting edge of central tooth (Figure 11F) weakly in- dented; lateral cusps, 4—6; median cusp narrow U-shaped, pointed, slightly broader and longer than laterals; basal cusps, 1—2, cusp on outer side smaller; basal tongue V- shaped, even with lateral margins; basal sockets moder- ately excavated; lateral margins narrow, slightly thick- ened, angled about 55° to vertical axis of tooth. Cutting edge of lateral tooth (Figure 11G) horizontal or with weakly indented; cutting edge about 67% of length of lateral wing; tooth face taller than broad; central cusp rounded, narrow U-shaped, lateral cusps, 2—4 (inner), 3— 5 (outer). Inner marginal teeth (Figure 11H) with 28-36 cusps; outer marginal teeth (Figure 111) with 25-30 cusps. Cusps on inner marginal teeth larger than those on outer marginals. Stomach longer than style sac. Snout light gray to black. Tentacles light gray to black, pigment often lighter around eyes. Foot light to medium gray. Opercular lobe black along inner edge. Pallial roof, visceral coil black. Ctenidium positioned slightly anterior to pericardium; filaments about 14, small, about as tall as wide, without pleats. Osphradium about 38% of ctenidium length. Hy- pobranchial gland without anterior swelling. Renal organ with prominent (50%) pallial portion; renal opening slightly thickened. Cephalo-pedal ganglia pigmented. Ovary 0.25—0.4 whorl, abutting or slightly overlapping edge of stomach, filling less than 50% of digestive gland behind stomach. Distal female genitalia shown in Figure 10D, E. Primary loop of coiled oviduct narrowly vertical to circular; distal arm swollen with sperm. Primary loop The Veliger, Vol. 42, No. 4 preceded by smaller, narrowly vertical loop. Oviduct and bursal duct join behind pallial wall (slightly behind pos- terior edge of capsule gland). Bursa copulatrix about 55% of albumen gland length; ovate to pyriform, transversely oriented, about 50% overlapped by albumen gland. Bur- sal duct short (ca. 25—40% of bursa length), narrow, orig- inating near ventral edge of bursa. Seminal receptacle smaller (33%) than bursa copulatrix, positioned slightly anterior or slightly overlapping bursa copulatrix near ven- tral edge of albumen gland, usually completely over- lapped by albumen gland. Albumen gland with well-de- veloped rectal furrow. Albumen gland with moderate (40%) pallial component; capsule gland entirely pallial. Capsule gland about as long, but narrower than albumen gland; capsule gland folded over to the right. Ventral channel without anterior vestibule. Genital opening a small terminal slit. Testis 1.0 whorl, overlapping posterior stomach cham- ber, filling about 50% of digestive gland behind stomach. Prostate gland with 33% of length in pallial roof. Pallial vas deferens very narrow, straight; portion of vas defer- ens in neck straight. Penis (Figure 10F) medium-sized, sickle-shaped, weakly curved, without folds; base some- times slightly narrowed, medium section without taper; distal section sharply pointed. Penial duct near outer edge, straight, surrounded by internal core of melanin dis- tally. External surface of penis pale. Type locality: Unnamed (‘‘Waterfall’’) spring, source of Hardscrabble Creek, Pyramid Lake basin, Washoe Coun- ty, Nevada, T. 24 N, R. 20 E, SE % section 13. A broad (400 m wide), shallow rheocrene (16.1°C, 144 micro- mhos/cm) (Figure 4D). Holotype, USNM 874902 (Figure 8H), collected by G. Vinyard, 31 October 1992; para- types, USNM 860758. Remarks: This species is distinguished from other con- geners by the strongly (adapically) angled shell whorls, large shell aperture, and highly eccentric operculum nu- cleus. Fluminicola virginius otherwise closely resembles F. turbiniformis, although it also differs from this species in having a relatively longer bursal duct. Material examined: NEVADA. Washoe County: Un- named (‘Waterfall’) spring, source of Hardscrabble Creek, Pyramid Lake basin, USNM 860758, USNM 874105, USNM 874902. Pristinicola Hershler et al., 1994 Type species: Bythinella hemphilli Pilsbry, 1890; original designation. Diagnosis: A monotypic northwestern United States ge- nus having small to medium-sized, narrow-conic shell with wrinkled protoconch, smooth whorls, and simple ap- erture. Animal pale. Penis without accessory lobes or glands. Female coiled oviduct of tight loops; glandular R. Hershler, 1999 Page 325 Figure 11 Scanning electron micrographs of shell, operculum, and radula of Fluminicola virginius Hershler, sp. nov. A. Shell, USNM 860758 (height 2.6 mm). B, C. Shell apex, USNM 860758. Bars = 300 wm, 240 wm, respectively. D. Operculum, outer surface, USNM 874105. Bar = 315 wm. E. Operculum, inner surface, USNM 874105. Bar = 300 um. E Central radular teeth, USNM 874105. Bar = 12 pm. G. Lateral radular tooth, USNM 874105. Bar = 13.8 wm. H. Inner marginal tooth, USNM 874105. Bar = 14.6 ym. I. Outer marginal teeth, USNM 874105. Bar = 15 pm. Page 326 oviduct large, ventrally closed; bursa copulatrix large, posteriorly recurved; seminal receptacle pouchlike. Remarks: The single species in this genus was recently reviewed by Hershler et al. (1994). Pristinicola hemphilli (Pilsbry, 1890) (Figure 5) Bythinella hemphilli Pilsbry, 1890:63.—Turgeon et al., 1988:60. Pristinicola hemphilli (Pilsbry, 1890), Hershler et al., 1994: 225-233, figs. 1 (top row), 2-7. Diagnosis: As for genus. Type locality: Near Kentucky Ferry, Snake River, Wash- ington. This locality has not been precisely located (Hen- derson, 1936; Hershler et al., 1994). Lectotype, ANSP 31176. Remarks: The Great Basin populations closely resemble other material of this species, which previously was known from the lower Snake-Columbia River basin and other Pacific coastal drainages of Washington, with shells 3.0—3.5 mm tall and having about 5.0 whorls. The Great Basin localities (Figure 5) are upland waters which drain south to the Harney-Malheur basin. It is noteworthy that the fish faunas of the Great Basin and Columbia River drainage also overlap considerably in this region (Bisson & Bond, 1971). Material examined: OREGON. Harney County: Moun- tain Spring, Silvies River drainage, T. 19 S, R. 32 E, SE % section 3, USNM 883878.—Unnamed springs, Cricket Creek, Silvies River drainage, T. 21 S, R. 28 E, NW % section 12, USNM 883875.—Adams Spring, head of Al- lison Creek, Silver Creek (Harney Lake drainage), T. 19 S, R. 26 E, SW % section 12, USNM 883879. WASH- INGTON. Near Kentucky Ferry, Snake River, ANSP SMUALGA Sy, Eremopyrgus, Hershler, gen. nov. Type species: Eremopyrgus eganensis, sp. nov. Etymology: From Greek, eremos, away from, separate; and pyrgos, tower. Referring to the isolation of this snail in eastern Nevada, and to its elongate shell. Gender mas- culine. The Veliger, Vol. 42, No. 4 Diagnosis: A Great Basin cochliopine group having me- dium-sized, conical shell. Penis ornamented with two squat, glandular lobes, positioned along outer edge me- dially and inner edge distally. Distal portion of penis swollen, pointed, without terminal papilla. Females ovo- viviparous; capsule gland thin-walled, functioning as brood pouch; bursa copulatrix large, seminal receptacle minute; fertilization duct coiled, opening to bursa copu- latrix. Description: Shell (Figure 12A) conical, with pro- nounced spire; height, 3.1—-3.8 mm; whorls, 4.5—5.25, te- leoconch whorls weakly convex, usually evenly rounded, sometimes having sub-sutural angulation. Shell clear- white, periostracum thin, brown. Protoconch (Figure 12B, C) of 1.60—1.75 whorls, diameter about 0.34 mm; initial portion often sculptured with a few, irregular wrinkles, later portion sometimes having a few weak spiral ele- ments. Protoconch followed by distinct, relatively smooth 0.5 whorl corresponding to shell growth within the female brood pouch. Teleoconch microsculpture of faint growth lines and occasional weak spiral striae. Aperture medium- sized, narrowly ovate, strongly angled adapically; outer lip rarely slightly thickened internally, orthocline or weakly prosocline, sometimes weakly sinuate; parietal lip complete across body whorl, thin, broadly adnate; colu- mellar swelling absent or narrow. Shell anomphalous or rimate. Operculum (Figure 12D, E) medium thickness, amber, ellipsoidal, paucispiral, nucleus highly eccentric; outer margin without rim. Attachment scar margin thick- ened all around. Callus sometimes well developed. Sali- vary glands simple, narrow tubes. Radula with about 52 rows of teeth; ribbon length, 557 wm, ribbon width, 90 pm; central tooth width, 22 wm. Central tooth (Figure 12F) with weak dorsal indentation; lateral cusps, 4—5; median cusp pointed, considerably broader and longer than laterals; basal cusp, 1; basal tongue V-shaped, often distinctly separated from remaining base, about even with lateral margins; basal sockets medium indented; lateral margins slightly thickened, inclined about 55° to vertical axis of teeth. Lateral tooth with slightly convex dorsal edge; lateral wings about 150% width of cutting edge; tooth face about as tall as wide; central cusp U-shaped; lateral cusps, 3—4 (inside), 4—5 (outside). Inner marginal teeth with 20—27 cusps; outer marginal teeth with 27-34 cusps. Cusps on inner marginals larger than those on out- Figure 12 Scanning electron micrographs of shell, operculum, and radula of Eremopyrgus eganensis Hershler, sp. nov. A. Shell, USNM 860759 (height 3.3 mm). B, C. Shell apex, USNM 860759. Bars = 150 ym, 230 pm, respectively. D. Operculum, outer surface, USNM 883940. Bar = 333 wm. E. Operculum, inner surface, USNM 883940. Bar = 315 pm. F Central radular teeth, USNM 883940. Bar = 10 pm. G. Lateral radular tooth, USNM 883940. Bar = 11 ym. H. Inner marginal tooth, USNM 883940. Bar jm. 11 pm. I. Outer marginal teeth, USNM 883940. Bar = 12 ey oy oy re Ww o — fa N inl O a0) ~ Page 328 The Veliger, Vol. 42, No. 4 er marginals. Dorsal folds of esophagus long, straight. Cephalic tentacles medium length in preserved material. Tentacles pale or pigmented with scattered gray granules, sometimes forming longitudinal strip. Snout, foot pale or light to medium gray-brown. Opercular lobe sometimes black along inner edge. Neck pale or pigmented with scattered gray-brown granules. Pallial roof, visceral coil variably pigmented, usually light to medium brown, sometimes black. Ctenidium abutting pericardium; fila- ments about 22, well developed, slightly taller than wide, without pleats. Osphradium small (15%), centrally posi- tioned. Hypobranchial gland well developed along rec- tum. Renal organ with large (50%) pallial portion; renal opening simple. Salivary glands small, tubular. Stomach longer than style sac, without posterior caecum. Cephalo- pedal ganglia strongly pigmented; cerebral, pedal com- missures relatively long (ca. 50%). Oviduct terminating as narrow, blind tube just behind stomach. Distal female genitalia shown in Figure 13A, B. Coiled oviduct of a single, small posterior-oblique loop. Seminal receptacle a minute sac positioned along left side of bursa copulatrix near ventral edge. Seminal receptacle duct very short, opening to oviduct along ventral edge of bursa copulatrix just distal (and anterior) to oviduct coil (slightly behind the pallial wall) where also joined by the albumen gland and the narrow, coiled fertilization duct. Fertilization duct opening to left side of bursa copulatrix near anterior edge; duct having several tight coils on right side of bursa co- pulatrix before looping ventrally onto left side, where it coils once more before joining the oviduct. Bursa copu- latrix relatively large, saclike, positioned along the left- ventral side of the brood pouch, extending to near the posterior edge of the brood pouch; duct short, narrow; sperm tube opening to pallial cavity a little anterior to pallial wall. Brood pouch large, posteriorly folded (and greatly narrowed) along right side of bursa copulatrix. Pouch containing relatively few (2-5) embryos having up to 2.5 whorls. Albumen gland very short, narrow, posi- tioned along right edge of bursa copulatrix. Genital ap- erture a slightly muscularized terminal slit. Testis 1.0 whorl, overlapping posterior stomach chamber, filling about 50% of digestive gland behind stomach. Prostate gland strongly recurved, with about 50% of length in pal- lial roof. Anterior vas deferens exiting prostate gland a little behind anterior tip, in front of pallial wall. Pallial vas deferens narrow, with small posterior loop; portion of vas deferens in neck straight. Penis (Figure 13C) medi- um-sized, relatively narrow, weakly curved, gently taper- ing, inner edge somewhat swollen distally; penial tip pointed. Glandular penial lobes, 2, small, cuboidal, slight- ly expanded distally; glands filling distal half of lobes, discharging through rather large terminal openings. Lobe along inner edge distal, lobe along outer edge medial. Penial duct narrow, undulating throughout, positioned near outer edge. Penis pale except for small, dark area near tip. Remarks: Eremopyrgus is assigned to the subfamily Cochliopinae (Hershler & Thompson, 1992) based on its specialized penial glands and female sperm tube separat- ed from the glandular oviduct. This snail is distinguished from all other members of the subfamily by its unique glandular penial lobes, which bear some resemblance to those of members of the “‘Heleobia group” (Hershler & Thompson, 1992), but are cuboidal rather than spherical and do not have a large glandular lumen. Eremopyrgus further differs from the only member of the “‘Heleobia group” that broods young, Mesobia (locally endemic in Honduras), in having spiral (as opposed to wrinkled) pro- toconch sculpture, a distally pointed penis that lacks a terminal papilla, a much larger bursa copulatrix and much smaller seminal receptacle, much shorter ducts of both the bursa copulatrix and seminal receptacle, a more com- plexly coiled fertilization duct, and fewer brooded young. Eremopyrgus does not appear to be closely related to Tryonia, the only other member of the Cochliopinae re- ported from the Great Basin region. Eremopyrgus eganensis Hershler, sp. nov. Steptoe hydrobe (Figures 3C, 12, 13A—-C, 14) Etymology: Refers to distribution of this species along the (eastern) flank of the Egan Range. Diagnosis: As for genus. Description: As for genus. Type locality: spring, northwest of Clark Spring, Steptoe Valley, White Pine County, Nevada, T. 19 N, R. 63 E, NW % section 20. A small rheocrene (19°C, 495 micro- mhos/cm) slightly disturbed by cattle (Figure 4E). Holo- type, USNM 874692 (Figure 3C), collected by R. Hersh- ler and P. Hovingh, 23 June 1992; paratypes, USNM 860759. Remarks: Eremopyrgus eganensis lives in a group of small, closely proximate, warm springs in the southeast segment of Steptoe Valley (Figure 14). Material examined: NEVADA. White Pine County: spring, northwest of Clark Spring, Steptoe Valley, USNM 860759, USNM 874692, USNM 883529, USNM 883940.—springs, Steptoe Ranch, T. 19 N, R. 63 E, SW % section 5, USNM 873219.—“Big Spring,” Steptoe Ranch, T. 19 N, R. 63 E, SW % section 5, USNM 873204.—spring, north of Steptoe Ranch, T. 19 N, R. 63 E, NE % section 5, USNM 873209. Tryonia Stimpson, 1865a Type species: Tryonia clathrata Stimpson, 1865a; orig- inal designation. Diagnosis: Minute to large, with elongate-conic to turri- R. Hershler, 1999 Page 329 Figure 13 Genitalia of Eremopyrgus Hershler, sp. nov., and Tryonia species (A—C, E. eganensis; Hershler, gen. & sp. nov.; D— G, T. monitorae, Hershler, sp. nov. USNM 860760). A. Left side of female glandular oviduct and associated structures, USNM 883940. Bar = 0.5 mm. B. Right side of bursa copulatrix and associated structures, USNM 883940. Scale as in “A.” C. Dorsal surface of penis, USNM 874692. Bar = 0.25 mm. D. Left side of posterior portion of female glandular oviduct and associated structures. Bar = 0.125 mm. E. Left side of anterior portion of brood pouch, showing slightly muscularized opening. Scale as in “D.” E Left side of seminal receptacle and associated structures (coiled oviduct rotated to left). Scale as in “D.” G. Dorsal surface of penis. Bar = 0.25 mm. Ag, albumen gland; Bp, brood pouch; Bpa, opening of brood pouch; Bu, bursa copulatrix; Cg, capsule gland; Cov, coiled oviduct; Fd, fertilization duct; Pd, penial duct; Pl, penial lobe; Pw, posterior wall of pallial cavity; Sr, seminal receptacle; St, sperm tube. Page 330 The Veliger, Vol. 42, No. 4 0 40 80 120 160 200 km ee Lee Ne hu. Eremopyrgus eee: ei 3 ae * e Tryonia clathrata og s Tryonia monitorae ~ ; a eee a Tryonia protea wa << Figure 14 Map of the Great Basin (excluding northernmost portion) and adjacent regions showing the distribution of Ere- mopyrgus Hershler, gen. nov., and Tryonia species. The Mexican locality for T. protea is not shown. R. Hershler, 1999 form shell. Penis ornamented with one or more glandular papillae. Distal portion of penis having blunt, pigmented tip, a sub-terminal swelling along inner edge, and termi- nal papilla through which penial duct opens. Females Ovoviviparous; capsule gland thin-walled, functioning as brood pouch; albumen gland highly reduced; bursa co- pulatrix and seminal receptacle small; fertilization duct coiled, opening to sperm tube. Remarks: The scope and content of this genus remains poorly known as Tryonia has neither been subject to a modern revision nor been shown to be monophyletic. Many of the Recent species now allocated to the genus have not been well studied in terms of their anatomy. Taylor (1966b:196—198) placed numerous Recent-Tertia- ry high-spired species from North, Central, and South America into the genus, whereas Nuttall (1990:184—185) later questioned allocation of South American fossils to this group. Hershler & Thompson (1992) restricted the group to Pliocene-Recent species of North America. Full descriptions of previously reported Great Basin species will be provided in a forthcoming review of this genus. Tryonia clathrata (Stimpson, 1865a) (Figure 14) Tryonia clathrata Stimpson, 1865a:54, pl. 8, fig. 1.—Stimp- son, 1865b:48—49, fig. 29.—Tryon, 1870:67.—Stearns, 1893:281.—Pilsbry, 1899:122.—Stearns, 1901:282.— Walker, 1918:139.—Gregg, 1941:118.—Baker, 1964: 172.—Taylor, 1966b:197.—Taylor, 1975:58 (literature compilation).—Pratt, 1977:7.—Burch & Tottenham, 1980:100, fig. 134.—Williams et al., 1985:36, 45, 48.— Hershler & Thompson, 1987:figs. 1, 11, 12, 13-15, 19, 21—23.—Turgeon et al., 1988:63.—Hershler & Thomp- son, 1992:110, figs. 7la,c—e, 72. Diagnosis: A medium to large species with turriform shell; teleoconch sculpture of numerous, regularly spaced, collabral lamellae. Inner edge of penis ornamented with a single basal and four distal papillae. Type locality: Given as “basin of the Colorado Desert,” but probably in error; subfossil. Lectotype, ANSP 27969. Stimpson (1865b:48) indicated that the type material was collected by Blake during his service on one of the Pa- cific Railroad Surveys. Stearns (1893) disputed the type locality as this species has not been found in numerous other samples from the Colorado Desert, whereas Mer- riam collected living specimens from Pahranagat Valley (Nevada) well to the east. (Note that Pacific Railroad Sur- vey expedition led by Lt. R. S. Williamson, with Blake serving as geologist, explored the Colorado Desert, but did not venture near southern Nevada [Blake, 1857].) Stearns (1901) later suggested that older Colorado Desert collections probably came from near Merriam’s locality. Morrison (1940) reiterated this point and suggested that early usage of the term ‘“‘Colorado Desert’”’ probably re- Page 331 ferred more generally to the Great Basin. Taylor (1966b) suggested that the type material probably came from the Muddy River (Moapa Valley, Nevada). Baker (1964) designated ANSP 27969a as the lecto- type. Although the original label associated with this lot merely identifies it as from Stimpson’s collection, with the locality, ‘Colorado Desert,’ additional material from this lot, ANSP 30778, has a label identifying Blake as the collector, with the locality given as in Stimpson’s de- scription. Remarks: This species lives in warm springs in the White River trough (Moapa, Pahranagat, White River Valleys) in southeastern Nevada (Figure 14). Extant pop- ulations conform to ‘‘Colorado Desert’’ material, with shells varying from about 2.9—7.0 mm, and having 5.75— 8.75 whorls. Shell sculptural development varies from low, riblike ornament (rare) to well-developed, almost spinose lamellae. Whether or not Tryonia clathrata spira- listriata Wessilingh, 1996, from the Pliocene of Guate- mala, is closely related to extant Tryonia clathrata is con- jectural. As noted by Wesselingh (1996), these fossils, although similar to T. clathrata in size, shape and colla- bral shell sculpture, differ in having numerous well-de- veloped spiral lirae on the teleoconch. Material examined: Colorado Desert, ANSP 27969, USNM 27893, USNM 30596, USNM 56403, USNM 121072, USNM 170786. NEVADA. Clark County: 9.6 km northwest of Moapa, Moapa Valley, USNM 791488.—Muddy Spring, Moapa Valley, T. 14 S, R. 65 E, NE % section 16, USNM 873358, USNM 873359.— Muddy Spring, 100 m below source, Moapa Valley, T. 14 S, R. 65 E, NE % section 16, USNM 874346, USNM 874790.—springs, west of Muddy Spring, Moapa Valley, T. 14S, R. 65 E, NW % section 16, USNM 874351.— spring, west of Muddy Spring, Moapa Valley, T. 14 S, R. 65 E, NW % section 16, USNM 874007, USNM 874024.—spring, 0.6 km south of above, Moapa Valley, T. 14 S, R. 65 E, NW % section 16, USNM 850291, USNM 873192.—‘‘Cardy Lamb Spring,’’ Moapa Valley, T. 14 S, R. 65 E, SW % section 16, USNM 874352, USNM 874355, USNM 874788.—‘‘Apcar Springs,” Moapa Valley, T. 14S, R. 65 E, SE % section 16, USNM 874349.—“‘Oasis Spring,’ Moapa Valley, T. 14 S, R. 65 E, NW % section 16, USNM 874010.—Moapa Valley Water District Spring, T. 14 S, R. 65 E, SE % section 16, USNM 874018, USNM 874023.—spring, Moapa Valley National Wildlife Refuge, T. 14 S, R. 65 E, NE % section 21, USNM 873356, USNM 873417, USNM 874343, USNM 874506, USNM 874787.—spring, 14.8 km north- west of Moapa, Moapa Valley, T. 14 S, R. 65 E, NE % section 21, USNM 874080. Lincoln County: Pahranagat Valley, USNM 123621.—warm spring, Pahranagat Val- ley, USNM 107735. Ash Springs, Pahranagat Valley, T. 6S, R. 60 E, NE % section 1, USNM 874011, USNM 874095, USNM 874789.—Crystal Spring, Pahranagat Page 332 Valley, T. 5 S, R. 60 E, NE % section 10, USNM 873157. Nye County: Hot Creek (source), White River Valley, T. 6 N, R. 61 E, NE % section 18, USNM 873196, USNM 874306, USNM 874690.—Moorman Spring, White River Valley, T. 8 N, R. 61 E, SE % section 32, USNM 873178. Tryonia monitorae Hershler, sp. nov. Monitor tryonia (Figures 3D, 13D-—G, 14, 15) Etymology: Refers to endemism of this snail in Monitor Valley. Diagnosis: A medium to large species with turriform shell often weakly sculptured with spiral threads. Penis ornamented with single basal papillae along inner and outer edges, and two distal papillae along inner edge. Description: Shell (Figure 15A) turriform; height, 3.5— 4.6 mm; whorls, 6.25—7.5. Apex flattened, often tilted; protoconch (Figure 15B) of 1.75 whorls, diameter about 0.41 mm; smooth. No obvious zone representing growth during brood period evident. Teleoconch whorls weakly to moderately convex, evenly rounded, without shoulders. Microsculpture of weak growth lines, sometimes strengthened at short intervals. Spiral threads often ob- vious on shells retaining periostracum; threads less ob- vious on cleaned shells. Periostracum brown. Shell clear. Aperture small, ovate. Outer lip thin, orthocline, often sinuate, with abapical portion advanced. Parietal lip often complete across body whorl, thin, broadly adnate. Colu- mellar swelling absent. Shell anomphalous. Operculum (Figure 15 C, D) thin, slightly convex, am- ber, ovate, paucispiral, whorls on outer surface weakly frilled; outer margin having weak rim. Attachment scar margin unthickened on inner surface. Callus absent. Radula with about 47 rows of teeth; ribbon length, 400 wm, ribbon width, 75 wm; central tooth width, 19 pm. Central teeth (Figure 15 E, F) with moderate dorsal in- dentation; lateral cusps, 5—7; median cusp narrowly point- ed, slightly broader and considerably longer than laterals; basal cusps, 1—2, inner cusp larger; basal tongue broad V-shaped, about even with lateral margins; basal sockets medium indented; lateral margins slightly thickened, strongly flared, often with distinct bend along outer edge, inclined about 50° to vertical axis of teeth. Lateral teeth The Veliger, Vol. 42, No. 4 (Figure 15G) with very slight dorsal indentation; lateral wings slightly longer (120%) than width of cutting edge; tooth face slightly taller than wide; central cusp narrowly pointed; lateral cusps 3—4 (inside), 4—6 (outside). Inner marginal teeth (Figure 15H) with 19—25 cusps; outer mar- ginal teeth (Figure 15I) with 18—27 cusps. Cusps on inner marginals larger than those on outer marginals. Snout, tentacles, foot, neck unpigmented to medium gray-brown. Opercular lobe black along inner edge and sides. Pallial roof light gray to near black, visceral coil pale except for black pigment on stomach, to almost en- tirely black on all dorsal surfaces. Ctenidium abutting pericardium; filaments about 35, well developed, about as tall as wide, pleated. Osphradium small (ca. 14%), positioned centrally or slightly posterior to middle of ctenidial axis. Hypobranchial gland not evi- dent in dissection. Renal organ with moderate (30%) pal- lial bulge; renal opening slightly thickened. Salivary glands small, tubular. Stomach as long as style sac; anterior stomach chamber larger than posterior chamber. Cephalo- pedal ganglia unpigmented; cerebral commissure moderate length (ca. 43%); pedal commissure very short. Oviduct terminating as slightly thickened tube a little behind stomach. Distal female genitalia shown in Figure 13 D-F Coiled oviduct of two small, darkly pigmented overlapping loops (initial loop anterior-oblique, second loop posterior-oblique). Seminal receptacle a very small pouch just anterior to the bursa copulatrix (sometimes slightly overlapping left-dorsal surface). Seminal recep- tacle duct about as long as body, opening to coiled ovi- duct at point where latter connects with albumen gland. Fertilization duct of two small, tightly appressed coils opening to sperm tube (a little behind pallial wall) dorsal to coiled oviduct. Bursa copulatrix small, ovate, posi- tioned along left-ventral side of brood pouch, extending to (or slightly posterior to) posterior edge of brood pouch, with narrow duct emerging from anterior edge; duct (to point where joined by fertilization duct) narrow, slightly longer than bursa copulatrix; sperm tube opening to pal- lial cavity a short distance in front of pallial wall. Brood pouch large, posteriorly folded along right side of bursa copulatrix. Pouch containing about 12 variably sized em- bryos; largest embryos of about 2.0 shell whorls. Albu- men gland short, narrow, coursing ventrally across right side of bursa copulatrix and extending onto left side of bursa. Genital aperture broad, slightly muscularized. Figure 15 Scanning electron micrographs of shell, operculum, and radula of Tryonia monitorae Hershler, sp. nov. A. Shell, USNM 860760 (height 2.6 mm). B. Shell apex, USNM 860760. Bar = 133 wm. C. Operculum, outer surface, USNM 883939. Bar = 200 ym. D. Operculum, inner surface, USNM 883939. Bar = 214 wm. E. Central radular teeth, USNM 883939. Bar = 10 ym. F Central radular teeth, USNM 883941. Bar = 8.5 ym. G. Lateral radular tooth, USNM 883939. Bar = 10 wm. H. Inner marginal tooth, USNM 883939. Bar = 11 wm. I. Outer marginal teeth, USNM 883939. Bar = 10 wpm. D oy ox me Ww >) — a N al ) a0) m~ Page 334 Testis 1.5 whorl, occupying more than 50% of diges- tive gland behind stomach. Prostate gland small, ovate, with very short pallial component (13%). Anterior vas deferens exiting from anterior tip of prostate gland, slight- ly in front of pallial wall. Pallial vas deferens straight; portion of vas deferens in neck straight. Penis (Figure 13G) large, narrow, usually strongly curved, gently ta- pering, inner edge with pronounced bulge near terminus; penial tip rounded, with small terminal papilla. Glandular penial lobes, 4, small, conical. Basal lobes positioned on inner curvature and near outer edge. Distal lobes some- what longer and narrower than basal lobes, positioned along inner curvature between middle of penis and distal bulge. Penial duct relatively broad, undulating except for distalmost section, positioned near outer edge. Penis pale or with light gray-brown external pigment; distal portion having small, prominent internal black patch near termi- nus. Type locality: Hot Springs, Potts Ranch, Monitor Valley, Nye County, Nevada, T. 14 N, R. 47 E, NW % section 1. Numerous hot springs are present in this area (Garside & Schilling, 1979:52). Snails were collected from a small thermal (34.5°C) rheocrene (Figure 4F). Holotype, USNM 892046 (Figure 2D), collected by D. W. Sada, 4 August 1996; paratypes, USNM 860760. Remarks: This snail is restricted to the type locality and the warm (31.5°) outflow of Dianas Punch Bowl (Figure 4G). These localities are located a few km apart (Figure 14) along a fault (Garside & Schilling, 1979:52, 53). This species closely resembles 7. margae Hershler, 1989, from Death Valley, in shell shape and penial ornament, but is larger and also differs in having periostracal spiral sculp- ture, fewer opercular whorls, lighter body pigmentation, a narrower basal process on the central radular teeth, and smaller radular cusps. Tryonia monitorae differs from all other congeners that have been studied anatomically in that the albumen gland coils onto the left side of the bursa copulatrix. Material examined: NEVADA. Nye County: Hot Springs, Potts Ranch, Monitor Valley, USNM 860760, USNM 874883, USNM 883530, USNM 883939, USNM 892046.—Dianas Punch Bowl (Hot Springs), Monitor Valley, T. 14 N, R. 47 E, SE % section 22, USNM 874882, USNM 883941. Tryonia protea (Gould, 1855) (Figure 14) Amnicola protea Gould, 1855:129—130.—Gould, 1857:332, pl. XI:figs. 6-9.—Johnson, 1964:132. Melania exigua Conrad, 1855:269 (non Morelet, 1851). Tryonia protea (Gould, 1855), Binney, 1865:71, 72, figs. 140—142.—Tryon, 1870:68.—Berry, 1948:59, 69.— Taylor, 1966a:53—54.—Taylor, 1966b:197.—Russell, 1971:232.—Taylor, 1975:160 (literature compila- The Veliger, Vol. 42, No. 4 tion).—Burch & Tottenham, 1980:100, figs., 136, 137.—Taylor, 1981:153-154 (in part).—Taylor, 1985: 317, fig. 35 (in part).—Turgeon et al., 1988:63.—Her- shler, 1989:207—208, figs. 52-54.—Hershler & Thomp- son, 1992:111. Bythinella protea (Gould, 1855), Stearns, 1893:278—281 (in part). Paludestrina protea (Gould, 1855), Stearns, 1901:277—284, pl. XIX—XXI (in part).—Hannibal, 1912:186 (in part)—Walker, 1918:138-139 (in part).—Henderson, 1924:191, fig. 94 Chamberlin & Jones, 1929:178.— Henderson, 1929:167, fig. 176.—Jones, 1940:44. Hydrobia protea (Gould, 1855), Henderson, 1936:139. Pyrgulopsis imminens Taylor, 1950:28, figs. 1-3. Pyrgulopsis blakeana Taylor, 1950:30, figs. 4—6. Pyrgulopsis cahuillarum Taylor, 1950:31—32, fig. 7. Diagnosis: Large, narrowly turriform, teleoconch often sculptured with spiral ridges and/or collabral ribs. Shell sculpture highly variable within and among populations, ranging from smooth to cancellate. Males unknown. Type locality: Colorado Desert (Gran Jornado); subfos- sil. Syntypes, USNM 121074. Bequaert & Miller (1973: 213) suggested that the type material probably came from Riverside County (California), near Salton View. Remarks: Extant populations tend to have weaker sculp- ture than subfossil material from the type locality area. The species is disjunctly distributed, with populations concentrated in the upper Owens River drainage, lower Colorado River basin, Bonneville basin, and Lahontan ba- sin (Figure 14). Taylor (1966a:53, 54) suggested that liv- ing populations assigned to this species may be compos- ite, but I have not found this evident based on morpho- logic criteria. Late Pleistocene fossil records from Ivan- pah Mountains and Pahrump Valley (Roth & Reynolds, 1990; Taylor, 1986, respectively) and live collections from Gila River drainage, Arizona (Bequaert & Miller, 1973:213) and Meadow Valley Wash, Nevada (Taylor, 1986:fig. 1) require confirmation. Gregg (1941) reported this species from Moapa, Nevada, but I suspect that his material represented relative smooth-shelled variants of Tryonia clathrata. Material examined (exclusive of the numerous sub-fos- sil records from the type locality area): MEXICO. SO- NORA. spring, El Doctor, Colorado River drainage, USNM 873440, USNM 874183. CALIFORNIA. Colo- rado Desert, USNM 121074. Mono County: Hot Creek, Long Valley, T. 3 S, R. 28 E, NE % sec. 25, USNM 857954, USNM 873362, USNM 874182, USNM 883309.—Whitmore Hot Springs, Long Valley, T. 4 S, R. 29 E, NE % sec. 6, USNM 874180.—spring, tributary to Little Alkali Lake, Long Valley, T. 3 S, R. 29 E, NW %4 section 29, USNM 873364, USNM 873365, USNM 874189. Riverside County: warm springs near Salton, USNM 104886.—Dos Palmas Spring, Salt Creek drain- age, T. 8 S, R. 11 E, NW % section 3, USNM 163227, USNM 791494.—spring, ca. 1.0 km WSW of Hunters R. Hershler, 1999 Spring, Salt Creek drainage, T. 8S, R. 11 E, NE % section 14, USNM 873367.—Hunters Spring, Salt Creek drain- age, T. 8 S, R. 11 E, SW % sec. 12, USNM 873443, USNM 874194, USNM 874469.—‘‘Oasis Spring,” Salt Creek drainage, T. 8 S, R. 12 E, NE % section 30, USNM 854744, USNM 873441, USNM 874196. NEVADA. Clark County: Blue Point Spring, Virgin River drainage, T. 18 S, R. 68 E, SW % section 6, USNM 883248, USNM 883884. Washoe County: spring, tributary to Fly Reser- voir, Hualapai Flat, T. 34 N, R. 23 E, SE % section 2, USNM 892080. UTAH. Juab County: Percy Spring, Fish Springs Flat, T. 11 S, R. 14 W, SE % section 26, USNM 854617, USNM 858283, USNM 883474.—South Springs, Fish Springs Flat, T. 11 S, R. 14 W, NE % section 26, USNM 858286. Tooele County: Horseshoe Springs, Skull Valley, T. 2 S, R. 8 W, SE % sec. 26, USNM 858291, USNM 883883.—first spring south of Josepha, Skull Valley, FMNH 178352, FMNH 224336.—spring at Josepha, Skull Valley, FMNH 178411.—spring before Jo- sepha, Skull Valley, FMNH 178379.—Warm Springs, Tooele Valley, T. 2 S, R. 6 W, NE % sec. 16.—spring, Blue Lake, Great Salt Lake Desert, T. 4S, R. 19 N, NW % sec. 7, USNM 883397.—Salt Spring, FMNH 178443, FMNH 224404. Acknowledgments. Collecting permits were provided by the Ne- vada Department of Wildlife, State of Idaho Department of Fish and Game, and U.S. Fish and Wildlife Service. Loan of museum material was facilitated by G. Rosenberg (ANSP), T. Gosliner and E. Kools (CAS), and R. Bieler and J. Slapcinsky (FMNH). T. Frest, D. Giuliani, P. Hovingh, J. Landye, D. Sada, G. Vinyard, and D. Wong assisted with fieldwork and/or shared notes and material. Y. Villacampa (USNM) measured shells; prepared, studied and photographed snail morphology using a scanning electron microscope; and printed photographs. W. Brown and S. Braden facilitated use of the USNM Scanning Electron Micros- copy Laboratory. K. Darrow inked anatomical drawings. M. Ryan (USNM) drew and inked shells and assisted with final map preparation. Terry Frest made helpful comments on a draft of the manuscript. Fiscal support was provided by the Smithsonian In- stitution (awards from the Scholarly Studies Program, and Office of the Provost), Bureau of Land Management (Bureau of Inland Fisheries), National Biological Service, and California Depart- ment of Fish and Game (Contract FG7342). LITERATURE CITED Batty, J. L. & R. I. Batty. 1951-1952. Further observations on the Mollusca of the relict lakes in the Great Basin. The Nau- tilus 65:46—53, 85-93. BairD, W. 1863. Descriptions of some new species of shells, collected at Vancouver Island and in British Columbia, by J. K. Lord, Esq., naturalist to the British North-American Boundary Commission, in the years 1858-1862. Proceed- ings of the Zoological Society of London 31:66—70. Baker, H. B. 1964. Type land snails in the Academy of Natural Sciences of Philadelphia. Part III. Limnophile and thalas- sophile Pulmonata. Part IV. Land and freshwater Prosobran- chia. Proceedings of the Academy of Natural Sciences of Philadelphia 116(4):149-193. Page 335 BEETLE, D. E. 1957. The Mollusca of Teton County, Wyoming. The Nautilus 71(1):12—22. BEETLE, D. E. 1961. A checklist of Wyoming Recent Mollusca. Sterkiana 3:1—9. BEETLE, D. E. 1989. Checklist of Recent Mollusca of Wyoming, USA. Great Basin Naturalist 49(4):637—645. BEQUAERT, J. C. & W. B. MILLER. 1973. The Mollusks of the Arid Southwest, with an Arizona Check List. The University of Arizona Press: Tucson. 271 pp. Berry, E. G. 1948 (*‘1947°’). Snails collected for the schisto- somiasis investigations. United States National Institute of Health Bulletin 189:55—69. BINNEY, W. G. 1865. Land and fresh-water shells of North Amer- ica, IJ: Ampullariidae, Valvatidae, Viviparidae, fresh-water Rissooidae, Cyclophoridae, Truncatellidae, fresh-water Ne- ritidae, Helicinidae. Smithsonian Miscellaneous Collections 7:120 pp. Bisson, P. A. & C. E. BOND. 1971. Origin and distribution of the fishes of Harney basin, Oregon. Copeia 1971(2):268—281. BLAKE, W. P. 1857. Geological report. 370 pages, 13 plates in Vol. 5, Report of explorations and surveys for railroad routes to connect with the routes near the 35th and 32d parallels of north latitude by Lieutenant R. S. Williamson, 1853; in Reports of explorations and surveys, to ascertain the most practicable and economical route for a railroad from the Mississippi River to the Pacific Ocean made under the di- rection of the Secretary of War, in 1853—4, according to acts of Congress of March 3, 1853, May 31, 1854, and August 5, 1854. A. O. P. Nicholson: Washington, D.C. Burcu, J. B. & J. L. TOTTENHAM. 1980. North American fresh- water snails. Species list, ranges and illustrations. Walkerana 1(3):81-215. CALL, R. E. 1884. On the Quaternary and Recent Mollusca of the Great Basin with descriptions of new forms. Introduced by a sketch of the Quaternary lakes of the Great Basin by G. K. Gilbert. United States Geological Survey Bulletin 11: 66 pp., plates I-VI. CARPENTER, P. P. 1864. Supplementary report on the present state of our knowledge with regard to the Mollusca of the west coast of North America. Report of the British Association for the Advancement of Science 33:517—-586. CHAMBERLIN, R. V. & D. T. JONEs. 1929. A descriptive catalog of the Mollusca of Utah. Bulletin of the University of Utah 19:203 pp., map. Conrab, T. 1855. Description of a new species of Melania. Pro- ceedings of the Academy of Natural Sciences of Philadel- phia 7:269. GarsipE, L. J. & J. H. SCHILLING. 1979. Thermal waters of Ne- vada. Nevada Bureau of Mines and Geology Bulletin 91: 163 pp., plate. GouLp, A. A. 1855. New species of land and freshwater shells from western (N.) America. Proceedings of the Boston So- ciety of Natural History 5:127—130. GouLp, A. A. 1857. Catalogue of the Recent shells, with descrip- tions of the new species. Appendix, article 3, pp. 330-336. pl. 11 of pt. 2, Geological report, W. P. Blake, in Vol. 5, Report of explorations and surveys for railroad routes to connect with the routes near the 35th and 32d parallels of north latitude by Lieutenant R. S. Williamson, 1853, in Re- ports of explorations and surveys, to ascertain the most prac- ticable and economical route for a railroad from the Missis- sippi River to the Pacific Ocean made under the direction of the Secretary of War, in 1853-4, according to acts of Page 336 Congress of March 3, 1853, May 31, 1854, and August 5, 1854. A. O. P. Nicholson: Washington, D.C. GreGG, W. O. 1941. Fluminicola avernalis and Fluminicola avernalis carinifera from Nevada. The Nautilus 54(4):117— 118. HALDEMAN, S. 1840-1871. A Monograph of the Limniades or Freshwater Univalve Shells of North America. J. Dobson: Philadelphia. 231 pp. HANNIBAL, H. 1912. A synopsis of the Recent and Tertiary fresh- water mollusks of the Californian Province, based upon an ontogenetic classification. Proceedings of the Malacological Society of London 10:112—166, 167-211. HENDERSON, J. 1924. Mollusca of Colorado, Utah, Montana, Ida- ho and Wyoming. University of Colorado Studies 13:65— 223. HENDERSON, J. 1929. Non-marine Mollusca of Oregon and Wash- ington. University of Colorado Studies 17(2):47—190. HENDERSON, J. 1936. Mollusca of Colorado, Utah, Montana, Ida- ho and Wyoming—Supplement. University of Colorado Studies 23:81—145. HERSHLER, R. 1989. Springsnails (Gastropoda: Hydrobiidae) of Owens and Amargosa River (exclusive of Ash Meadows) drainages, Death Valley system, California-Nevada. Pro- ceedings of the Biological Society of Washington 102:176— 248. HERSHLER, R. 1998. A systematic review of the hydrobiid snails (Gastropoda: Rissooidea) of the Great Basin, western United States. Part I. Genus Pyrgulopsis. The Veliger 41(1):1—132. HERSHLER, R. & T. J. FResT. 1996. A review of the North Amer- ican freshwater snail genus Fluminicola (Hydrobiidae). Smithsonian Contributions to Zoology 583:41 pp. HERSHLER, R. & E G. THOMPSON. 1987. North American Hydro- biidae (Gastropoda: Rissoacea): redescription and systematic relationships of Tryonia Stimpson, 1865 and Pyrgulopsis Call and Pilsbry, 1886. The Nautilus 101:25-32. HERSHLER, R. & E G. THOMPSON. 1988. Notes on morphology of Amnicola limosa (Say, 1817) (Gastropoda: Hydrobiidae) with comments on status of the subfamily Amnicolinae. Malacological Review 21:81—92. HERSHLER, R. & E G. THOMPSON. 1992. A review of the aquatic gastropod subfamily Cochliopinae (Prosobranchia: Hydro- biidae). Malacological Review Supplement 5:140 pp. HERSHLER, R., T. J. FREST, E. J. JOHANNES, P. A. BOWLER & FE G. THompson. 1994. Two new genera of hydrobiid snails (Pros- obranchia: Rissooidea) from the northwestern United States. The Veliger 37(3):221—243. Hinps, R. B. 1842. Descriptions of new shells. Annals and Mag- azine of Natural History 10:81—84. JOHNSON, R. I. 1964. The Recent Mollusca of Augustus Addison Gould. United States National Museum Bulletin 239:182 pp.. 45 plates. Jones, D. T. 1940. Recent collections of Utah Mollusca, with extralimital records from certain Utah cabinets. Proceedings of the Utah Academy of Sciences, Arts and Letters 17:33- 45. Lea, I. 1838. Description of new freshwater and land shells. Transactions of the American Philosophical Society 6:1— 154. MoreLeT, A. 1851. Testacea novissima Insulae Cubanae et Americae Centralis. Pars II. J.-B. Bailliére: Paris. 30 pp. Morrison, J. P. E. 1940. A new species of Fluminicola with notes on “‘Colorado Desert”’ shells, and on the genus Clap- pia. The Nautilus 53:124—127. NUTTALL, C. P. 1990. A review of the Tertiary non-marine mol- The Veliger, Vol. 42, No. 4 luscan faunas of the Pebasian and other inland basins of north-western South America. Bulletin of the British Mu- seum of Natural History (Geology) 45(2):165-371. Pi-ssry, H. A. 1890. Notices of new Amnicolidae. The Nautilus 4:63-64. Pitssry, H. A. 1899. Catalogue of the Amnicolidae of the west- ern United States. The Nautilus 12:121-—127. Pitssry, H. A. 1935. Western and southwestern Amnicolidae and a new Humboldtiana. The Nautilus 48:91-94. PRATT, W. L. 1977. Hydrobiid snails of the Moapa warm spring complex, Nevada. Western Society of Malacologists, Annual Report 10:7. [abstract.] RotH, B. & R. E. REYNOLDs. 1990. Late Quaternary nonmarine Mollusca from Kokoweef Cave, Ivanpah Mountains, Cali- fornia. Bulletin of the Southern California Academy of Sci- ences 89(1):1-9. RUSSELL, R. H. 1971. Mollusca of Fish Springs, Juab County, Utah: rediscovery of Stagnicola pilsbryi (Hemphill, 1890). Great Basin Naturalist 31:223—236. Say, T. 1817. Descriptions of new species of land and fresh water shells of the United States. Journal of the Academy of Nat- ural Sciences of Philadelphia 1(1):123-126. STEARNS, R. E. C. 1893. Report on the land and fresh-water shells collected in California and Nevada by the Death Valley Ex- pedition, including a few additional species obtained by Dr. C. Hart Merriam and assistants in parts of the southwestern United States. North American Fauna 7:269-—283. STEARNS, R. E. C. 1901. The fossil fresh-water shells of the Col- orado Desert, their distribution, environment, and variation. Proceedings of the United States National Museum 24:27 1— 299, pls. XIX—XXIV. Stimpson, W. 1865a. Diagnoses of newly discovered genera of gasteropods, belonging to the sub-fam. Hydrobiinae, of the family Rissoidae. American Journal of Conchology 1:52— 54, pl. 8. STIMPSON, W. 1865b. Researches upon the Hydrobiinae and allied forms; chiefly made upon materials in the museum of the Smithsonian Institution. Smithsonian Miscellaneous Collec- tions 201:59 pp. TayLor, D. W. 1950. Three new Pyrgulopsis from the Colorado Desert, California. Leaflets in Malacology 1(7):27-33. TayLor, D. W. 1966a. Summary of North American Blancan nonmarine mollusks. Malacologia 4:172 pp. TAYLOR, D. W. 1966b. A remarkable snail fauna from Coahuila, México. The Veliger 9:152—228. TayLor, D. W. 1975. Index and bibliography of Late Cenozoic freshwater Mollusca of western North America. University of Michigan Museum of Paleontology, Papers on Paleontol- ogy 10:384 pp., errata (March, 1976). [Claude W. Hibbard Memorial Volume 1.] TAyLor, D. W. 1981. Freshwater mollusks of California: a dis- tributional checklist. California Fish and Game 67(3):140— 163. TAYLOR, D. W. 1985. Evolution of freshwater drainages and mol- lusks in western North America. Pp. 265-321 in C. J. Smiley & A. J. Leviton (eds.), Late Cenozoic History of the Pacific Northwest. American Association for the Advancement of Science: San Francisco. TAYLor, D. W. 1986. Fossil molluscs from the Lake Hill archeo- logical site, Panamint Valley, southeastern California. Pp. 42-54 in E. L. Davis & C. Raven (eds.), Environmental and Paleoenvironmental Studies in Panamint Valley. Contribu- tions of the Great Basin Foundation 2. TAyLor, D. W. & R. C. BriGHT. 1987. Drainage history of the R. Hershler, 1999 Page 337 Bonneville Basin. Pp. 239-256 in R. S. Koop & R. E. Coh- enour (eds.), Cenozoic Geology of Western Utah: Sites for Precious Metal and Hydrocarbon Accumulations. Utah Geo- logical Association Publication 16. THompson, F G. & R. HERSHLER. 1991. Two new hydrobiid snails (Amnicolinae) from Florida and Georgia, with a dis- cussion of the biogeography of freshwater gastropods of south Georgia streams. Malacological Review 24:55-—72. TROSCHEL, F H. 1856-1863. Das gebiss der shnecken zur ber- griindung einer natiirlichen classification. Vol. 1. Nicolaische Verlagsbuchhandlung: Berlin. 252 pp. TRYON, G. W. 1865. Descriptions of new species of Amnicola, Pomatiopsis, Somatogyrus, Gabbia, Hydrobia and Rissoa. American Journal of Conchology 1:219—222 + plate 22. TryYON, G. W. 1870. A Monograph of the Fresh-Water Univalve Mollusca of the United States: Turbidae, Physadae. Part 1. [Conchological Section of] Academy of Natural Sciences of Philadelphia: Philadelphia. 82 pp. TurRGEON, D. D., A. E. BOGAN, E. V. COAN, W. K. EMERSON, W. G. Lyons, W. L. Pratt, C. E E. Roper, A. SCHELTEMA, FE G. THompPSON & J. D. WILLIAMS. 1988. Common and sci- entific names of aquatic invertebrates from the United States and Canada: mollusks. American Fisheries Society Special Publication 16:277 pp., 6 plates. WALKER, B. 1918. A synopsis of the classification of the fresh- water Mollusca of North America, north of Mexico, and a catalogue of the more recently described species, with notes. University of Michigan Museum of Zoology Miscellaneous Publications 6:213 pp. WESSELINGH, F. P. 1996. Geological-paleontological research in the Tertiary and Quaternary of Central America HI. New Pliocene fresh-water gastropods from Guatemala. Documen- ta Naturae 100:23—36. WILLIAMS, J. E., D. B. Bowman, J. E. BRooks, A. A. ECHELLE, R. J. Epwarps, D. A. HENDRICKSON & J. J. LANDYE. 1985. Endangered aquatic ecosystems in North American deserts with a list of vanishing fishes of the region. Journal of the Arizona-Nevada Academy of Science 20:1—62. THE VELIGER © CMS, Inc., 1999 The Veliger 42(4):338—372 (October 1, 1999) Land Caenogastropods of Mounts Mahermana, Ilapiry, and Vasiha, Southeastern Madagascar, with Conservation Statuses of 17 Species of Boucardicus KENNETH C. EMBERTON Molluscan Biodiversity Institute, 110 Old Airport Road, Concord, North Carolina 28025, USA AND TIMOTHY A. PEARCE* Delaware Museum of Natural History, Box 3937, Wilmington, Delaware 19807-0937, USA Abstract. Quantitative, replicated altitudinal transects on the three mountains yielded 25 caenogastropod species in six genera in four families. Madecataulus Fischer-Piette & Bedoucha, 1965, is synonymized under Boucardicus Fischer- Piette & Bedoucha, 1965. Presence is noted of the three large species Hainesia crocea (Sowerby, 1847), Tropidophora sp. 1, and 7. sp. 2. Descriptions are given of the small species Boucardicus albocinctus (E. A. Smith, 1893); B. antiquus sp. nov.; B. carylae sp. nov.; B. culminans (Fischer-Piette, Blanc, Blanc & Salvat, 1993); B. curvifolius sp. nov.; B. delicatus sp. nov.; B. divei Fischer-Piette, Blanc, Blanc & Salvat, 1993; B. esetrae sp. nov.; B. fidimananai sp. nov.; B. fortistriatus sp. nov.; B. magnilobatus sp. nov.; B. mahermanae sp. nov.; B. rakotoarisoni sp. nov.; B. randalanai sp. nov.; B. simplex sp. nov.; B. tridentatus sp. nov.; B. victorhernandezi Emberton, 1998; Cyathopoma randalana sp. nov.; Malarinia calcopercula Emberton, 1994; Tropidophora (Ligatella) vallorzi Fischer-Piette, Blanc, Blanc & Salvat, 1993; Omphalotropis vohimenae sp. nov.; and O. costulata sp. nov. Distributional data were available that allowed evaluation of each of the 17 Boucardicus species for its conservation status, applying the latest IUCN criteria. Four species are proposed as Critically Endangered, 11 as Endangered, and two as Vulnerable. INTRODUCTION Bedoucha, 1965; streptaxids; Sitala H. Adams, 1865; and other pulmonates) (Emberton, 1997); (d) there is evidence that lowlands are richer than highlands in endemic and rare species (Emberton, 1997); and the small land snails (e) are sensitive ecological indicators of mild forest deg- radation from selective cutting or nearby slash-and-burn, and (f) do not seem to compete with congeners via shell size (Pearce & Emberton, unpublished). All of those conclusions were based on undocumented morphospecies. This paper is the first in a series of four papers that identify and describe the species. This paper treats the Mahermana-lIlapiry-Vasiha caenogastropods. In the interests of conservation in Madagascar, it is important to provide as much Red-List data IUCN, 1996) as possible. In this paper, we evaluate the conservation statuses of the 17 Boucardicus species described herein. Recent quantitative sampling of altitudinal transects on Mounts Mahermana, Ilapiry, and Vasiha in southeastern Madagascar yielded 88 species of land snails and slugs (Emberton et al., 1996, 1999; Emberton, 1997). Of these, 81 species are small (“‘micro’’) land snails (< 5 mm in greatest dimension at any collected life stage). Analyses of the distributions of 80 of these species have shown that (a) the best sampling strategy for Madagascar-rain- forest snails is timed searching for micro-snails, while incidentally collecting macro-snails and litter-plus-soil for later picking of the 5.5-1.2 mm and the 1.2—0.85 mm, dry-sieved fractions (Emberton et al., 1996); (b) total land-snail diversity is significantly higher in the unpro- tected Vohimena Mountain Chain than in the protected Anosy Mountain Chain (Emberton et al., 1999); (c) the Vohimena Chain’s greater richness occurs in four of the eight major groups of land snails (charopids; Microcystis Beck, 1837; Kalidos Gude, 1911; and non-Boucardicus ““prosobranchs”’ [caenogastropods]), and does not exist for the other four groups (Boucardicus Fischer-Piette & MATERIALS anp METHODS Collecting methods have been detailed by Emberton et al. (1996). Sixteen stations were collected and numbered in the “Tol” series (for Tolagnaro = Fort Dauphin, the nearest city). These stations have been mapped by Em- * To whom reprint requests should be sent. berton et al. (1996, 1999) and in Emberton (1997). To K. C. Emberton & T. A. Pearce, 1999 shorten the taxonomic descriptions, stations are described briefly below. Station numbers are in the series of the Molluscan Biodiversity Institute (MBI). All stations were restricted to primary forest that had no more than limited selective cutting. Ecological data are given by Emberton (1997:table 1). All stations are in Madagascar: Tulear Province. Mount Mahermana (Vohimena Chain) is north- east of the village of Esetra, apiry (Vohimena Chain) is west of Mahialambo, and Vasiha (Anosy Chain) is west of Malio. Latitude and longitude are given in degrees, minutes, and seconds. MBI 373 (= Tol-1). Summit of Mt. Mahermana, 340 m, 24°26'12”S, 47°13'13’E. MBI 374 (= Tol-2). Slope of Mt. Mahermana, 300 m, 24°26'17"S, 47°13'10"E. MBI 375 (= Tol-3). Slope of Mt. Mahermana, 200 m, 24°26'15"S, 47°13'04"E. MBI 376 (= Tol-4). Valley on Mt. Mahermana, 100 m, 24°26'22”S, 47°12’41"E. MBI 377 (= Tol-5). Summit of Mt. Ilapiry, 540 m, 24°51'40"S, 47°00'20"E. MBI 378 (= Tol-6). Ridge on Mt. Ilapiry, 500 m, 24°51'33”S, 47°00'27’E. MBI 379 (= Tol-7). Ridge, valley, and slope on Mt. Ilapiry, 400 m, 24°51'27"S, 47°00'38’E. MBI 380 (= Tol-8). Slope of Mt. Ilapiry, 300 m, 24°51'36"S, 47°00’40’E. MBI 381 (= Tol-9). Slope of Mt. [lapiry, 200 m, 24°51'39"S, 47°00'46’E. MBI 382 (= Tol-10). Lower summit of Mt. Vasiha, 860 m, 24°55'18"S, 46°44'19"E. MBI 383 (= Tol-11). Slope of Mt. Vasiha, 700 m, 24°55'23”"S, 46°44'27’E. MBI 384 (= Tol-12). Slope of Mt. Vasiha, 500 m, 24°55'19"S, 46°44'45’E. MBI 385 (= Tol-13). Valley on Mt. Vasiha, 400 m, 24°55'25"S, 46°44'45”E. MBI 386 (= Tol-14). Slope of Mt. Vasiha, 300 m, 24°55'37"S, 46°44'49"E. MBI 387 (= Tol-15). Slope of Mt. Vasiha, 200 m, 24°56'13”S, 46°45'13’E. MBI 388 (= Tol-16). Slope of Mt. Vasiha, 100 m, 24°56'20"S, 46°46'07’E. MBI 389 (= Tol-3-—4). Incidental collecting between Tol-3 and Tol-4. MBI 390 (= Tol-1—2). Incidental collecting between Tol-1 and Tol-2. MBI 391 (= Tol-sub-5). Incidental collecting below summit of Mt. Ilapiry, Tol-5. MBI 392 (= Tol-7—9). Incidental collecting between Tol-7 and Tol-9. Species identifications and comparisons were made us- ing Fischer-Piette et al. (1993) and Emberton (1994, 1998). All caenogastropod species were identified, but, as Madagascar’s large caenogastropod species are either fairly well known (Fischer-Piette et al., 1993) or—in the Page 339 case of large Tropidophora—in taxonomic chaos (Em- berton, 1995), descriptions were prepared only for the small species. For each small species, the holotype or a representative shell was photographed in apertural, basal, and side views at either X10, X16, X25, or X40 magnification, and in apical view at X40 magnification. Additional specimens were photographed as needed to illustrate shell variation or ontogeny. Fifty-eight shell characters (Table 1, Figure 1) were measured, or measured and calculated, or scored from the photographs or from the shells themselves. At least one adult male or female anatomy was avail- able for 12 (71%) of the Boucardicus species. From each of these species, one to three reproductive systems were removed and illustrated by photographs and/or camera- lucida drawings as they were turned and progressively dissected to expose characters in the penis and FPSC (fer- tilization pouch-seminal receptacle complex). Only the penis and FPSC were examined because of time con- straints and because these two organs seemed most likely to contain informative characters, based on previous ex- perience. Seventeen reproductive-anatomical characters (Table 1, Figures 26-31) were taken from the drawings or from the dissections themselves. Character matrices were prepared (available from K.C.E. on request) and were used to code character-state data into the DELTA system (Partridge et al., 1993; Dall- witz et al., 1993), which was then used to generate nat- ural-language species descriptions. Computer-assisted taxonomic descriptions and keys have been developed over the years by a number of approaches (e.g., Pank- hurst, 1975; Watson et al., 1986), arguably culminating in the DELTA system (Partridge et al., 1993; Dallwitz et al., 1993). DELTA is ‘‘a flexible data-coding format for taxonomic descriptions, and an associated set of programs for producing and typesetting natural-language descrip- tions and keys, for interactive identification and infor- mation retrieval, and for conversion of data to formats required for phylogenetic and phenetic analysis’ (Par- tridge et al., 1993). For each Boucardicus species, conservation status was evaluated using the latest categories and criteria of the International Union for the Conservation of Nature (IUCN, 1996). Ranges were estimated from distribution data in Emberton (in press). Rainforest extent and decline were assessed using Green & Sussman (1990), Sussman et al. (1994), and the most recently available topographic maps. INFERENCE oF HOMOLOGIES Interpretations of shell homologies were straightforward. Penis width was ruled out as a character, because during mating it can be drastically swollen (Figure 38 versus Figure 37). In the FPSC (fertilization pouch-seminal re- Page 340 Table 1 Shell and reproductive characters used in descriptions. SHELL 1. Diameter (0.1 mm) 2. Height (0.1 mm) 3. Height/Diameter (0.1) 4. Spire angle (degrees) 5. Whorl periphery shape (round, angular, keeled) 6. Whorl shoulder shape (round, flat) 7. Aperture width parallel to parietal callus (% diameter) 8. Aperture height (perpendicular to parietal callus)/width (0.01) 9. Columellar plica (yes, no) 10. Aperture-plane inclination upward from rotational axis (5 de- grees) 11. Apertural anal notch depth (% apertural width) 12. Baso-columellar denticle size (% apertural width) 13. Baso-columellar denticle depth (0.05 whorl) 14. Basal denticle size (% apertural width) 15. Basal denticle depth (0.05 whorl) 16. Upper palatal denticle size (% apertural width) 17. Upper palatal denticle depth (0.05 whorl) 18. Peristome angle of greatest dimension outward from rota- tional axis (5 degrees) 19. Peristome greater dimension/aperture width in same direc- tion (0.01) 20. Peristome greatest dimension/lesser, perpendicular dimen- sion (0.01) 21. Peristome baso-palatal indentation (% basal peristome width) 22. Peristome upper curl forward extension (% upper peristome width) 23. Inner, second peristome (none, projecting up tp 0.01 whorl, projecting up to 0.05 whorl) 24. Umbilicus size pre-constriction (% diameter) 25. Umbilicus size total (% diameter) 26. Whorl number (0.1) 27. Embryonic whorl number (0.1) 28. Embryonic whorl sculpture 29. First whorl diameter (0.01 mm) 30. First three whorls diameter (0.01 mm) 31. Penult-whorl complete spiral grooves depth (0.01 mm) 32. Penult-whorl complete spiral grooves number between su- tures 33. Penult-whorl spiral ridges height (0.01 mm) 34. Penult-whorl spiral ridges number between sutures 35. Penult-whorl spiral grooves or ridges waviness (none, slight, moderate) 36. Penult-whorl transverse ribs height (0.01 mm) 37. Penult-whorl transverse ribs number in last 0.1 whorl 38. Penult-whorl herringbone sculpture (Figures 4, 10) number in last 0.1 whorl 39. Penult-whorl honeycomb sculpture (Figure 19) number per 0.1 whorl 40. Penult-whorl short spiral grooves number between sutures 41. Penult-whorl short spiral grooves length (0.01 mm) 42. Penult-whorl spiral lines of punctae number between sutures 43. Penult-whorl spiral lines of punctae number per 0.1 whorl 44. Pre-apertural constriction distance from aperture (0.1 whorl) 45. Pre-apertural constriction (% whorl diameter constriction) 46. Pre-constriction diminution of sculpture (%) The Veliger, Vol. 42, No. 4 Table 1 Continued. 47. Post-constriction immediate percent diameter swelling (%) 48. Post-constriction secondary constriction (yes, no) 49. Post-secondary constriction swelling (% diameter of first constriction) 50. Pre-apertural transverse ribs height (0.01 mm) 51. Pre-apertural transverse ribs procumbancy angle 52. Pre-apertural transverse ribs number per 0.1 whorl 53. Color general 54. Color apex 55. Spiral color band number 56. Spiral color band color 57. Preapertural constriction color 58. Peristome color (excluding periostracum) PENIS 59. Length (0.1 mm) 60. Length/ shell diameter (0.1) 61. Terminal papilla-ejaculatory pore position (dorsal, central, ventral) 62. Dorsal papilla-ejaculatory pore position (terminal, subter- minal) 63. Papilla protrusion (none, slight, strong) 64. Papilla direction (anterior, posterior) 65. Swelling at tip of penis: swelling width/pre-swelling width (0.1) 66. Gland (present, absent) 67. Gland length/pre-swelling penial width (0.1) 68. Gland proximal-distal position: distance from base of penis to midpoint of gland/penial length (0.1) 69. Gland dorsal-ventral attachment position (dorsal, ventral) 70. Gland free-lobe direction (left, right) FPSC (FERTILIZATION POUCH-SEMINAL RECEPTACLE COMPLEX) 71. Base (present, absent) 72. Base shape 73. Base ducted gland (present, absent) 74. Body-interior muscular funnel (present, absent) 75. Body-and-tube shape ceptacle) great morphological diversity (Figures 27-31, 55-68) called for some judgment. Internal structures of two disparate morphologies (Figures 53, 54) led to the hypothesis of three distinct regions of the Boucardicus FPSC (Figures 27-31). The base appears to be lined with glandular tissue, and it may or may not have additional glandular lobes, or an appendage, or a ducted or ductless gland; the shape of the base varies from globular to thin and elongate; Cyathopoma seems to lack a base. The FPSC body seems to be a muscular tube that may or may not contain a muscular, funnel-shaped organ (Figure 54). Apically (i.e., proximally), the body grades into a thinner- walled tube; the gradation can be abrupt or gradual. SYSTEMATICS Higher classification follows Ponder & Lindberg (1997) and Vaught (1989). Type materials are placed in the Unit- K. €. Emberton & T. A: Pearce, 1999 Page 341 A \