Corea BSUPOROLE rt ? € oc BeeeN HOLE Tye aye, Poli Qeneh san Pipeeyenetsincar acest SOMO Soe ete tame es “Agari Aron natepa dan ens hea ea easter LEDs ihe Os ent tee SMe Qa SW PNFA SNE RONS RGR tly SEM Ati HAMMER, Rome Me Be dsky Va Amina aR DR rces re tane shia iin OWA SUNN Verte! , ao ee crrt ae Vane ive Py rots Cayhey MOR USC CO RR ced een et ty WAS Emy ae Money tale a syte me kaa te sea Onin fase yey Wve ACU yee tenon sive gehen niht ve Uiveaescns “ Pyne Ne Vane Wale Ban. Petar ye tien UR Y Tea trary Spent arora ye saan me Viagne eng sn rh Oh Ae CURR ra anit oo HEPA Ra tal ate het AANZIDAING one ae ees ALANA PN ED SUR MIAUTE NGS Ms AN AFA Y NS ABA Te aOR eno Mita chad thas MINA MEA ER Tinga Netter TINEA NO MEMS CA ANAS za denstak SktibaNvasfegANREAAR OCA AINE ‘ SeLids Ne iMa Nee distns Naame se Tnlauanasuetaatngcaasnaneyaeans 42 ENESCO MANNA Rg Ae NEB IION Melis BEAT STS Ne DRA MUR ES ANSE LIN MOSM REA ML iris fae Hc AMMAN Ranseayenatts alban al MAMA RANencA seg ceae veins: AApannelasnanea WING RE Cane NeAe ARAM stho Sale AE GE, UN Ay katte ang Rent pains os MAZMAQU RS aaeAnasiNs Kam Hyd alre dame IN A tostestuad ic trlam ss eat NaN UCAGMANEN A maya MMAR ABA CA AEM AMEN RST IAA IN Mad deh afi iamemt WPA VIAN ONL wien lh MEATS YH MCA R MD, NVA CME NSATRIM AAT AIEANR CNL NE Na yalnaaeaueceey om eA nce INA N AN ASLAM ATA sar darheeaiccroce MNEMSR IGE RANA, seins IMs ANe mo Nats ROM ALFA ANI nee aA eens LOE M RASA RR Su Maen anveneaenemeas RAY Sith Mabry einnahEA MN AMES MAINED tee Dea Rin Rite” Dias DaEAD NS Csth Mery s 751495 Ve Ga VAM sefhie bo SYirye ania arnt a tAE Nes eam ay italia tee rperrertt! ite Pettis Fin bh a ANE edad Net ae caning Dsl tors ian oN SPL SEDO ROT eC se Eris me tiingins tee None A Nanri om edit sae ona OWN ToaAn dab etm insanhuaams sane tong asiven the agit Ro Mana nsme Wuneniga Motbah one AAMeNs SaloeMed ctantinntiane Vere AL ena Seta ed eee AVE MAMMA Toya ND eRe Vteatan mayne We Neh ® Ee Renan eet ‘1 VEE NNW ean ‘ Hay Segue tw Dike MLaeDeNnaainsAshantnetartytog DON NS Vee! cieawa SAN aay Veni Nine ayA p auto, tants ty, aeneateeont RUG /4GR: Pea nese ase ast gh lat ee ta! pene dato rene Sistatomai ee eVnesy ene 8 1 POP Wy npegend eur eae PobE gt deat ods feta runpanyitear gee paarstoanes axel Sea ea asec AOS Tey aRE Vena Wig wey ga ras ane} Geet aa biyaseseg Hyp AUCH: Setaaees ye aes Vata oeal weeeye gees phoma es Weg aay is Sin evi yoga Meow a Pegs a ie TRANG AS yay ares eVaeigny oat aire PEW EME ery i“ Sy EERO RL ES PAUL QORUC UF ES ee gE KY SHAE ES HEE NS yy izaeege eis Tevet ire tery iieeah er weeeaare yl ae EYRE NEEM Ara Nye TO Uae gsnereayea tenn eee ay weaersee Toe te tat Nee es os Fret reer ace teed: VINER Orne MNES ONS MOUS SATE AY NA Sey bah ae R NA edn Wd se nig aie ba eniba my Sauryen ey, STIRS ASE Mote Nebo mea Sateen aaa eAa Qe acnens & VeAE MN yan es Ae Ad ens SA ein v2 as Sah 4 Ark ster FS IRB a ENE ree risen ee vege Messages eu daa os Bae ye FLUF a Bey peenae sou age NEN aSu eet pe ge eye ee pear rip tipes ea te BERS INCE aaa ERCP Te Ed ped PEAR MED BIN TIES Pe LOPeLy ay el Rie hie Oop r er iere tt) RBH EN Mah unieniss Beg Pea eus ecg ye ye Erne errant sera Usa ec dye phai unre OE aNG Diteironiserereny rraretieri ie neat ane rt Preece geese et sta eD Sat E a EeRT STO SETESPNyROAd eSTETOI TON VEEN DET RESCH HEN E/E CoP S LY PRWR EH IYEE E IENS Si Fyi acy iged FUL UenUN eta ye yeaa pecues* ETE LEpmpLreune reese Lian sm aren tyres ier rN Dad Che Ts PeTistrT teteta ere Tt Seren ie tin taret sear Ee ASG CURBED CYS DTTP ETE ES cas ya gE PH LEED EBIEAS HS WRUNG eins EMEA IYa 9 ge BOY RATA GAP PDT REG ETO EMA Y Les NS Re UTE Dope Gruen we SHRI TY CELE RD OD TG cua RAN ENS erp Pigs eb srat Gea WN WEE Er Ye & ME BREASES FER TTVC DR HE CU ERY SU RUS ea EDC et Wee CR pe age ty WE WEF aCe 34 ELEC tag CVU GRIN Car RE" INI Ci Exenaeanye ree OF BASRA E BE RGERTEE NS Seg ETUC OES E REE AYE NEU SC HIERE CEE POPE BN RENARD easel neers TBE SCOT WEEE ANCA RURYEVER TS C= PUPA we tee rp) wn - 2 ” 77) ty fae = . Oo . Se fo) “ x (@) pao fo) L Vio 7: = Z, = = XY = ie PAE E : >=" >" Se) = = = = 5) = a Zs 2 2 7) zZ 77) NOILALILSNI NVINOSHLINS LIBRARIES SMIT saiuvualy LIBRARIES SMITHSONIAN INSTITUTION LIBRARIES SMITHSONIAK ? = = > 2p) = wo ! & us & G us & am é = oa a cae oc eal fad C st CS = ze Hev7Z = = ay = E ce = fi - QYY ig E cies 3 Pee m 9° * 3 - N = ” $4 = (22) = = w SSIYVYEIT LIBRARIES SMITHSONIAN NOILALILSNI NVINOSHLINS S3IYVYEIT_LIBRARIES SMIT p Zz fen n 2 (7) 2 n z Ree n S ZO NO < 2 SZ = 2 G § 4 til fy = = = Zz 3 Zi eNS Si “Go -F: if =r SS fe) yyy : : : : g ES 2 ae beat: \ 2 a ie alu: 2 ae eae INSTITUTION NOILNLILSNI NVINOSHLINS S3IuNVYSII LIBRARIES SMITHSONIAN _INSTITUTION NOIIMLITSNI NVIN a e ee a es ay Z lu 2 - vx Ww 2 wl = Ww — : % XE g : “g : a Ss 4 < = WS < a < c mS 2 * 5 = 3 aia i) = Z fe) a re} Z = Zz Ne 4 aN = 4 Zz S3INVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IY¥VYsgIT LIBRARIES SMIT 5 : ae ae Sa E 5 = ow = ies] = ow > = (os) = > x = a> = = Ly 5 2 ae = = e SS = pal Ye i es = = Po) = e2 = sing \ \ ae Ge) FE, YY 2) = 22) m ” m Ny 2 m ON Z A Zz wo = ee. = o z INSTITUTION NOILNLILSNI _NVINOSHLIWS S34 tuvudit tl BRARI ES SMITHSONIAN INSTITUTION NOILALILSNI_ NVIN fn Eg IGRI Ohi ®% 1 RR I a, : 2 6 Ee Ny i Ye : z : 2 \gq 3 = ) ~ RA Y yy n wo .7)) 2) DM ~~ We @ an r wo HYG t 3 Be Q 2 WHE B a See z. 2 : ENE E ees A : a 5 ee : SJIYVYUSIT LIBRARIES NOILNLILSNI_NVINOSHLINS SSIYVYEIT LIBRARIES SMIT ”n = 2) S w” Sia nf > ) = ® a Oo 2 a - g = 4 _— ft, — are [ang = . < =A fe ie ee = : _< =| ee = = oc = ac a Nav e S a as rs a. S re) S ~~ ca = nae: = =z a Zz 3 Zz ca =) z a | INSTITUTION NOILNLILSNI NVINOSHLINS SJIYVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOJLOLILSNI NVIN c z E z = z E z c ao = wo — wo = * OF = ow < z E z e z 5 z 5 a WS 2 e 2 = aa = 2 = cay (7p) Oo z ih ez o 2 o z o } SJIYVYGIT_LIBRARIES_ SMITHSONIAN NOILNLILSNI S3IYVYEIT_ LIBRARIES. SMIT no z Ss ” z= w z 2) z ee: n =H 6 = 5 = + fy & =a SWS g Y g a ff 2 gy ZN 2 Z E z = Fa WY E = = = z z 2 3 a a ie : 3 z INSTITUTION NOILNLILSNI NVINOSHLINS S31uvudIT LIBRARIES SMITHSONIAN NOILNLILSNI NVIN . SN Ss “ NOILNLILSNI LIBRARIES NOILNLILSNI LIBRARIES NOILNLILSNI LIB RAR !ES NOILALILSNI LIBRARIES NOILALILSNI Peet Me eae Pe ed ee eee ed ee ee pena NG TCR VY uae oS CIC PW lie Te Ve Pm Belem We bl ll ed ee VO - “=>>=*" SS w ee Sd a —_—_— o S31NVUSIT LIBRARIES SMITHSONIAN INSTITUTION n : z ~w n line 72) = Be Sr = =< = = z = ayes =i NS : 3 5 NYS > = oO aks eK LAX = = 2 E AN 2 > = > =) \ > z 7) 2 7) ae Z INSTITUTION _NVINOSHLINS S3IYUVUSIT z “4 Fahy i z A = n ss 7) 4 w = ox a) = = c es c = oO = a = O = (e) aa fo) Zz = z za Z S31YVYEIT LIBRARIES SMITHSONIAN INSTITUTION 4 c S = 5 = 2 = be E 2 > = > = 3 2 a Vip = 5 = Oo z | O Zz INSTITUTION NOILNLILSNI NVINOSHLINS S31uvuait z oe. (22) za n cS < wie = Hee SS prs Z NN; = oh Pgs =| dir 5 \ ee o at) ro) 2) wn 22) on yg Ww) so fe) z sao O ¥ ls = 2 Hy, = 2 = = > GQ = > = 1?) a (7) 2 w LIBRARIES SMITHSONIAN INSTITUTION tp) > (2p) > (7p) 3 o w AG, 2 — = oc. = oc < = <% a x 3 = oc aj ac n 5 [sa] 5 o 5 z zi s Zz iar INSTITUTION NOILNLILSNI_NVINOSHLINS S3IYVvugit a = es S os oa} — foe] — 2 = 2 = 2 e = Zz E 2 D z Pier vive i pealdWadIT_LIBRARIES SMITHSONIAN INSTITUTION fy, = ~ = ae z : 5 z 5 \w: 2) on WO w G \ . Ww Zz Ee SN 2 E Ns 2 : 3 on ewe Se INSTITUTION NOILNLILSNI_NVINOSHLINS S3/YVYEIT . = G = _ co < = _< @ + ce co ro) 2 J r4 a INSTITUTION NOILNLILSN S31YVYEIT LIBRARIES SMITHSONIAN _INSTITUTION INSTITUTION NOILNLILSNI ae NOILNLILSNI SMITHSONIAN LIBRARIES SMITHSONIAN NOILNLILSNI LIB NOILNLILSN LIBRARIES NOILNLILSNI SMITHSONIAN LIBRARIES SMITHSONIAN NOILALILSNI c z a = wo es = Es] > Es > 2 E 2 Mm 2 ‘Mf m rz) = w | NOILALILSN!I NVINOSHLINS SSIYVYEIT LIBRARIES z 7) z n z oe = = = = ea AN 5 z 5 x S . Ee z E = e = > “@ = > = 77) 2 7} z w _LIBRARIES SMITHSONIAN _INSTITUTION NOJLOLILSNI 4 A er -_ ple = LS $2 72 a& = co = e Puy = e ps S « Ye S a pe ro) og ro) & olf) sma Pag Za ).. NVINOSHLINS S3SIYVYUAIT LIBRARIES SN (2) f Zz = (2) = < = o z 4 / ae (e) see Oo Be RU o \ 2 NN \E > = aN > Z2 wo ” Zi INSTITUTION NOILNLILSNI_ Nt (7p) S w z LW 2 ul wa me 4 pa dy < = < 2 c. S fo = am. aS fa} = a fo) Rs re) _ Zz = 2 NVINOSHLINS S3IYVYSIT LIBRARIES SA Fr : = (im za meee 3 “ S i) = els Ss a 5) z = z i rea w Bs Z ys = NOILNLILSNI N! g z ra) z — = > = 7 Ww) ~ m3 n NVINOSHLINS S31YvVYsIT LIBRARIES Sh Nye dae S a @ = G a — ao _— (odin ae < 2 < = = a ll Oo is ro) =e z J Zz a) INSTITUTION NOILALILSNI Nv 5 — 5 = = _ E = = Be = = a ay E zx z @ z NVINOSHLINS SAlYVddIT_ LIBRARIES SK = crane = = z = L fo) =): Ww 7) , mits o} T+ A Z - x Z > s INN > 2 w % za INSTITUTION NOILNLILSNI_N\ wn S pom 2) : Zz tad rT, td Z w rs ua rs 5 < es , 2 = 2 =” (72) Ww o intr o z SMITHSONIAN NOILALILSNI N' 2 fe Zz w Zz = RN = = = z = z x= g 2 z a) 2 = 2 E = = en @ = 7) a edt 7) _NVINOSHIIWS Saluvudit_LIBRARIES Si 2 Os ud = ud 2 NX = = a = VN 4 << GAY = S B 5 he 5 2 > “ay a vietye THE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 35 January 2, 1992 to October 1, 1992 TABLE OF CONTENTS Number 1 (January 2, 1992) Phenotypic plasticity in the life histories and production of two warm-temperate viviparid prosobranchs KENNETH M. BROWN AND TERRY D. RICHARDSON ... 1 Multiple paternity in Crepidula fornicata (Linnaeus) PATRICK M. GAFFNEY AND BETH MCGEE The egg capsules and early life history of the corvallivorous gastropod Drupella cornus (Roding, 1798) SIREPHANIE Jia pO URNERG goes 6) res eee eer eee etre 16 Mantle-mediated shell decollation increases posterior aperture size in Dentalium rectius (Scaphopoda: Dentaliida) PATRICK: P| REYNOLDS i ascecae alate ene eee 26 Repaired shell damage among soft-bottom mollusks on the con- tinental shelf and upper slope north of Point Conception, California ROYSKAUIKROPP peng ir coer nnciasin te ame bee een 36 Observations on the biology of Turritella gonostoma Valenciennes (Prosobranchia: Turritellidae) from the Gulf of California WARREN D. ALLMON, DOUGLAS S. JONES, AND NIKKI VAUGHAN ree ee ee tre ee eer a Ga co 00 0 00 02 6 52 Simulathena papuensis, a new planaxid genus and species from the Indo-West Pacific RICHARD! S.: HIOUBRICK) (2.60 aera 64 Two new Vitrinellid species from the Gulf of California, Mexico (Gastropoda: Vitrinellidae) CaROLE M. HERTZ, BARBARA W. MYERS, AND JOYCE GEM- MELD 03.25.00 02 8k, ee eee 70 Two giant African land snail species spread to Martinique, French West Indies ALBERT R. MEAD AND LouIs PALCY ............... 74 Reproductive biology of Vermetus sp. and Dendropoma corrodens (Orbigny, 1842): Two vermetid gastropods from the Southern Caribbean PaTRICcIA A. MILOSLAVICH AND PABLO E. PENCHASZADEH 78 Number 2 (April 1, 1992) Anterior inhalant currents and pedal feeding in bivalves R. G. B. Reip, R. F. McMauon, D. O FOIGHIL, AND R. INNIGANU Ps ju oo ees eerie tes Mo nae heen 93 Remarks on Distorsio constricta (Broderip, 1833) and related spe- cies in the eastern Pacific Ocean, with the description of a new species (Gastropoda: Personidae) WILLIAM K. EMERSON AND BETTY JEAN PIECH 105 A re-evaluation of the ontogeny of Cabestana spengleri (Perry, 1811) (Gastropoda: Tonnoidea: Ranellidae) PRANKGRIEDE LM ite eg pee eee eR ee 117 The fossil land snail Helix leidyi Hall & Meek, 1855, a member of a new genus of Humboldtianidae (Gastropoda: Pulmon- ata) EMMETT EVANOFF AND BARRY ROTH New occurrences of the malleid bivalve Nayadina (Exputens) from the Eocene of Jamaica, Mexico, and Washington RICHARD L. SQUIRES 133 An eastern Pacific Mercenaria and notes on other chionine genera (Bivalvia: Veneridae) Mary ELLEN HarTE 137 On the validity, authorship, and publication date of the specific name Ancistrocheirus lesueurtt (Cephalopoda: Ancistrocheir- idae) GIAMBATTISTA BELLO 141 The anatomy, of Arion flagellus Collinge, 1893, present on the Iberian Peninsula (Gastropoda: Arionidae: Terrestria Nuda) JOsE CASTILLEJO 146 Number 3 (July 1, 1992) Humoral immunity: a-macroglobulin activity in the plasma of mollusks PETER B. ARMSTRONG AND JAMES P. QUIGLEY 161 Laevipilina antarctica and Micropilina arntzi, two new monopla- cophorans from the Antarctic ANDERS WAREN AND STEFAN HAIN ................ 165 Comments on and descriptions of eulimid gastropods from trop- ical West America ANDERS WAREN “ini haha eye ene ee Reece hee 177 Geographic and temporal variation in shell morphology of Acan- thina species from California and northern Baja California Gary L. GIANNINY AND DANA H. GEarRy 195 A new genus and species of Facelinidae (Opisthobranchia: Aeo- lidacea) from the Caribbean Sea ll SANDRA V. MILLEN AND JEFFREY C. HAMANN 205 A warm water Atlantic synonymy, Aphelodoris antillensis equals Chromodoris bistellata (Opisthobranchia: Gastropoda) JEFFREY C. HAMANN 215 A new genus and species of polygyrid land snail (Gastropoda: Pulmonata) from Oregon BARRY ROTH AND WALTER B. MILLER 222 Taxonomic re-evaluation and description of Gari radiata (Dunker in Philippi, 1845) (Bivalvia: Tellinoidea: Psammobiidae) RICHARD C. WILLAN 226 Embryonic stages of Loligo bleekeri Keferstein (Mollusca: Ce- phalopoda) GYEONG HUN BAEG, YASUNORI SAKURAI, AND KENJI SHIMAZAKI Number 4 (October 1, 1992) Reproduction and development of trochacean gastropods CAROLE S. HICKMAN 245 Systematic review of the family Choristellidae (Archaeogastro- poda: Lepetellacea) with descriptions of new species JaAMEs H. McLEAN 273 On the anatomy and relationships of the Choristellidae (Ar- chaeogastropoda: Lepetelloidea) (GERHARD) EIASZPRUNAR 950 5522-50) ee ee ee 295 The fine structure of the columellar muscle of some gastropod mollusks M. FRESCURA AND A. N. HODGSON 308 Egg mass and intracapsular development of Cypraea caputdraconis Melvill, 1888, from Easter Island (Gastropoda: Cypraeidae) CECILIA OsoRIO, CARLOS GALLARDO, AND HuGo ATAN 316 New morphologic and geographic data on the neritid gastropod Nerita ( Theliostyla) triangulata Gabb, 1869, from the Eocene of the Pacific coast of North America RICHARD L. SQUIRES 323 A new aeolid (Gastropoda: Nudibranchia) from the Atlantic coasts of the southern Iberian Peninsula J. L. Cervera, J. C. GarciaA-GOMEZ, AND P. J. Lo- REZ= GONZATEZ fein esata Sie airs ano bes it ats 330 Two new species of Helminthoglypta (Gastropoda: Pulmonata) from southern California, with comments on the subgenus Charodotes Pilsbry BARRY ROTH AND F. G. HOCHBERG ill Seasonal variation in the reproductive organs of two populations of Caracolus caracolla (Linné) (Pulmonata: Camaenidae) in Puerto Rico PATRICIA Marcos 347 The ecology of coquina clams Donax variabilis Say, 1822, and Donax parvula Philippi, 1849, on the east coast of Florida ERIK BONSDORFF AND WALTER G. NELSON ......... 358 “Solen rosaceus’—three species RWDOWONEC OSE ae eae es 366 Identification of monosaccharides in hydrolyzed Nautilus shell insoluble matrix by gas chromatography /mass spectrometry MICHAEL J. S. TEVESZ, STEPHEN F. SCHWELGIEN, BETH A. SMITH, Davip G. HEHEMANN, ROGER W. BINKLEY, AND JOSEPH G. CARTER 381 First study on the ecology of Sepia australis in the southern Ben- guela ecosystem M. R. Lipinsk1, M. A. COMPAGNO ROELEVELD, AND C. J. AUGUSTYN 384 Seasonal abundance of the small tropical sepioid /diosepius pyg- maeus (Cephalopoda: Idiosepiidae) at two localities off Townsville, North Queensland, Australia. GEORGERDAACKSONI oes yess a eee 396 Conus striatus survives attack by gonodactyloid! IEAM RINOEING cece siieca trae ele acti Seas cues ls oe ces 398 AUTHOR INDEX AEEMON SS WEUDDE: Sey ginid costa ath ce tle Sion RG ange he iia ere eee anes 52 ARMSTRONG iP Bites Go alura ce cha cals beeen ee heey arora 161 INTRA UED oe dees, oes ire ta seth ears sav eLISAe vet ns) RE Leeda Bee ae 316 AU GUSTYN, Ge Jen iis cstielaya am bos bcos dee RU ee eee et 384 BAEG Gee thle otic tect Cpt) ooeiel ae Uy noes: Late ee ea 234 BREET Ol Ghee iets eer | BD eaiuon 7g ti a Ra a a 141 BRINKER Ya Re Wis tela ses ct cteal ees nthe nua pee eteiinas 7 Moree tee area 381 IBONSDOREES tis. easel cicaselys acon) cet Ec eee ne ee 358 BROWN: 2 ME eh a 5A saciiags re RUA cue wal LD PASS ane Aa Poe een 1 IBUENSUCESOSORS Sli wow re Cos Pete Dreher Len aN Rana 158 GARTERS ee Grow per on ho neetie cui nciasy hae a HE ne censure 381 GASTIELE TOs i 8 ude cs taeusen rage eben eect ee nnereonslana rma renee ar ee 146 GCASTROSIMEAI RESD EB: Siraes ec eae eee een eee 158 GERVER A: Wun gc sce cushy ete eae US aus el aiaiee dans ee Me woe a 330 COANE Gt be ret nce Lan edie MR on ae eit ere ee Oy ORE (243) COMPAGNO ROELEVELD, M.A. ..................... 384 GOSET RE OVOND Secs Bey suc eee EAC UN carmel at a eee 366 DEZGASTR Os Mi sileay Rai seep oie i wee he eee ceneye aaa rc earmtttr ten 158 FEEMERSONA Wie See iste e Seon eb cae. bee Meal ERA oe ers cn 105 SVANOBRS ORE) Sindee iene terns ea een ee pea Ale te ee aye aa DSA Rar a 122 ESINNIGANG (Rie iene aise tc fenced teas foot eee SS eRe Eee 93 FSR'ESGURIA SeIVIN, 6 sti sa te Ee A Ve ine OAR AIC og a, ae RT 308 GAPENEYs Pad MMe ins tien csrat state eicee Aemete Aeate ent eTe aae 12 GAL TAR DON Genie leis aes int. ecyaronie toes fale et Seen ee Res 316 GATEEAR DOL Win (Gilani een ies ay Asie aO ioe hy Ap een 158 GARCIA-GOMEZ) Je Giese cuaises ate aie eae eae 330 RAR Vee TD) SAD ie ane aire, retiree eure ae sve Menten a acon ee 195 GEMMELT, Jeers eat Senate tr acaye marian sachet eee Waite eve 70 GTANNINYitGe OD ie eich eee ie sed caret Deen SM rake, en EN 195 GROVESSIE Ae ois Pe TD ke ARO ee ane a ee ar 157 EATING Se Ste ettey treble isecarn ve even ris testrn eater eet | ae diene EE 165 LAMANNG Up Cry soet crete cae red geese cine ene pelea ets lettre 205, 215 EPARST ES INIGOB icp tec rtcutniye ee nae eae ee ee OES 137 EU NVADQON~NIIGE oe als ch oo blvomesn ps ceeeknbeaas oeciee 295 IEDEHEMANN 1) 81 Gone suc eee ete me eee oe ee ee 381 PRB RG 230 Ol. INA tie cel dol ty ce th meen pT te 70 ERT OK MAN GS Syren eae eg ea 245 FIOCHBER GS HG oe geste rs shut choir recs ey eta ee a 338 FELOD GSONMPA SIN es rare ieescpee Pgs ee ROE Sea ean eRe 308 THOUBRIGK SR iS kas crn ee ae eee eee eee ge 64 JACKSONS | GAD oie. a ciecnad isjein dle ARNE la ee 396 JONES) DoS ethet cota sete idea ee coo eee 52 IKOIN SWAN Je ao AP AU ore LS ed ee eee (89), 398 IKROPPY Rs KG. 8 boca le Ca oe 36 TEIPINSKI) (Mia Ri es Sie ee a oe oe ee 384 WOrEZ=GONZAUEZ, (Boe sn she ee eo eee ee 330 MARCOS, Pe deb acini te se 347 MIGGEE. Bid isos 8 ota ail aeeto cl cn ee eee 12 IMIGIGRAN Jie es 8 se ee Ze) MIiGMAHON, (Re Feo iasi oe eyes i ee 93 MEADS AG Rac soci cd sie ern 5 cidug oS se 74 MULGENG Sie Vis sie Mics arate ecae 0) a ee 205 MOTEERAN Wi BS acti oe ssi ta or ou cinoncei ds ee CR oe D222; MUILOSEAVICH SP iiAy one cots oo 8 mote ee ee 78 MYERS Bi Ws 2) So on Se Sk en ee 70 NELSON} WiitG@eig on fob a SE See 358 OVFOIGHIE; Dio edn ee eee 93 OSORIO; Googe) on ee eee 316 PALGYS Toes ohn a eee oe ae 74 PENCHASZADEHG.P. Ey 2 2 hoe ee eee 78 PHILIPS; De Wa Sain oh 's dard 3 bio ee (160) PIECH) Be Jess 0 edie ee ae 2 ee 105 QUIGLEY, JePee oS 22 8 8 os oe ee 161 REID, Re Gy Bios cet ns ee 93 REYNOLDS), Ps D8) a a cn tec eee ee ee 26 RICHARDSON; “Ts Deo. be de ee eee 1 RIEDEL, Fs) oe 0 elke. DOE on ch ee ee 117 ROELEVELD® Mii As (Gan. 0 3 sins ee ee 384 ROTH! IBY glee iets ce tet oes eee Rea eee 122222338 SAKURAI. Ye Oo sce tile ake ie CS 234 SCHWELGIEN) |S. Fi) sale oo) oe ee eee 381 SHIMAZAKIN Ker ooo) waddle gle og ee 234 SMITH, IB A. 8.0.55 co cae ce cokes eee 381 SQUIRES? ReeTie fe. ryan s ieiees orl ao ee ee IBS}, BS TTREVESZ). Mii Ji!S8cccy dn on 2 sea oe 381 SRURNER NS! Jie csionieioes suautee acme 2S Se 16 VAUGHAN). Nei cscs sale cutis yin se (bn lisnsceet ae Oe ee 52 VON COSEL, (RY. je. feo gke ae a eee 366 WARTEN, (Ag sch. (N00 8 ols e ee de tera ee 165, 177 WIELAN GIRS ©. ce os) ee i 226 Page numbers for book reviews are indicated by parentheses. ISSN 0042-3211 VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 35 January 2, 1992 Number 1 CONTENTS Phenotypic plasticity in the life histories and production of two warm-temperate viviparid prosobranchs KENNETH M. BROWN AND TERRY D. RICHARDSON .................... 1 Multiple paternity in Crepidula fornicata (Linnaeus) RAGRICKOMAGARENEYSAND) BETH MIGGEE) £5)... 0. 5555.5 525 tenses a. 12 The egg capsules and early life history of the corvallivorous gastropod Drupella cornus (Roding, 1798) SIREBEVANUED, | PeONURINE Rumer Vy cp iiiua uioibetc tweet kis eho) Blas Biel oe een, 16 Mantle-mediated shell decollation increases posterior aperture size in Dentalium rectius (Scaphopoda: Dentaliida) ry WER Gea SINE YINOID Sule AG nh Narihaleers Wiest wis ae b4'athe be Mie ale ales 26 Repaired shell damage among soft-bottom mollusks on the continental shelf and upper slope north of Point Conception, California IRONG Is ISIROV ID 5's’ 0: 9 cl er Be A at De cot ye i Oe a 36 Observations on the biology of Turritella gonostoma Valenciennes (Prosobranchia: Turritellidae) from the Gulf of California WARREN D. ALLMON, DOUGLAS S. JONES, AND NIKKI VAUGHAN ........ 52 Simulathena papuensis, a new planaxid genus and species from the Indo-West Pacific RICHARD Otpld OUBRICK sa recgg: side) rimemarashe le dah ie ees dent dimtias Sy latl ores 64 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of January, April, July, and October. Rates for Volume 35 are $28.00 for affiliate membets3 > ,5ic¥ (including domestic mailing charges) and $56.00 for libraries and nonmembers (7n--~ cluding domestic mailing charges). For subscriptions sent to Canada and Mexico, add US $4.00; for subscriptions sent to addresses outside of North America, add US $8.00, which includes air-expedited delivery. Further membership and subscription infor- mation appears on the inside cover. The Veliger is published by the California Ma- lacozoological Society, Inc., % Museum of Paleontology, University of California, Berkeley, CA 94720. Second Class postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to The Veliger, Museum of Paleon- tology, University of California, Berkeley, CA 94720. THE VELIGER Scope of the journal The Veliger is open to original papers pertaining to any problem concerned with mol- lusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, evolutionary, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS’? in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Hans Bertsch, National University, Inglewood, California James T. Carlton, University of Oregon Eugene V. Coan, Research Associate, California Academy of Sciences, San Francisco J. Wyatt Durham, University of California, Berkeley William K. Emerson, American Museum of Natural History, New York Terrence M. Gosliner, California Academy of Sciences, San Francisco Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley David R. Lindberg, University of California, Berkeley James H. McLean, Los Angeles County Museum of Natural History Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Weseuieton Barry Roth, Santa Barbara Museum of Natural History Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (no institutional memberships) interested in any aspect of malacology. As an affiliate member, a person may subscribe to The Veliger for US $28.00 (Volume 35), which now zncludes mailing charges to domestic addresses. There is a one-time membership fee of US $2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription; a reinstatement fee of US $3.00 will be required if membership renewals do not reach the Society on or before October 1 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign members, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $56.00 (Volume 35), which now includes mailing charges to domestic addresses. For subscriptions sent to Canada and Mexico, add US $4.00; for subscriptions sent to addresses outside of North America, add US $8.00, which includes air-expedited delivery. Memberships and subscriptions are by Volume only (January 1 to October 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $25.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: The Veliger, Museum of Paleontology, 3 Earth Sciences Bldg., University of California, Berkeley, CA 94720, USA. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616, USA. THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER. The Veliger 35(1):1-11 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 Phenotypic Plasticity in the Life Histories and Production of ‘Two Warm-Temperate Viviparid Prosobranchs by KENNETH M. BROWN! ann TERRY D. RICHARDSON? Department of Zoology, Louisiana State University, Baton Rouge, Louisiana 70803, USA Abstract. Although prosobranch gastropods are common in southeastern rivers, little is known of the degree of phenotypic plasticity that occurs in their life histories and production ecology. We therefore studied cohort dynamics, life-history variation and secondary production in two viviparid prosobranchs at two sites in southern Louisiana. Bayou Manchac (BM), a slow-flowing flood-plain river, had levels of coarse and fine particulate organic material (potential food resources) twelve and six times greater, respectively, than did Old River (OR), an ox-bow lake, but also had lower average water temperatures, dissolved oxygen levels, pH, and water hardness. Considerable phenotypic plasticity in life histories and production occurred in both species. Viviparus subpurpureus (Say, 1829) dies after reproducing at an age of two years at BM, but grows more rapidly, reaches greater individual sizes, and reproduces and dies after only one year at OR. Campeloma decisum (Say, 1816) females reproduce at an age of two years, and then reproduce again and die at an age of three years at BM, but also grow more rapidly, reach larger shell lengths, and have an annual life-history pattern at OR. Females of both species at any given size brood more young at BM, but average clutch size is similar because females reach larger shell lengths at OR. Both species reached densities on the average four times higher at BM, resulting in higher annual standing stocks and greater annual production. Annual turnover rates, however, were two times greater at OR because of shorter development times and life cycles. Both snail species were found in deeper water at BM (with lower levels of dissolved oxygen) than at OR, suggesting that these snails are well adapted to hypoxia. When sub-adults of V. swhpurpureus from both sources were reared at BM, differences in growth rates and numbers of embryos brooded disappeared, indicating that this intraspecific life-history variation is eco-phenotypic. We suggest that high levels of food resources at BM may promote greater fecundity, population density, and secondary production, but that lower temperatures and dissolved oxygen levels at the same habitat during summer may limit individual growth and result in longer life cycles in both species. INTRODUCTION Populations can often diverge considerably in life-history tactics, and the “ecological templet” (e.g., environmental variables that alter or constrain phenotypic expression) is often important in explaining intraspecific variation (see review in PARTRIDGE & HARVEY, 1988). For example, in freshwater pulmonate snails such “eco-phenotypic” vari- "Send reprint requests to senior author. * Current_address: Box 2008, Bldg. 1504, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6351, USA. ation can be considerable (RUSSELL-HUNTER, 1978; McMauon, 1983), and important environmental vari- ables include periphyton quantity (EISENBERG, 1966; BROWN, 1985) and quality (EISENBERG, 1970; Mc- MAHON et al., 1974), population density (EISENBERG, 1970), and physico-chemical variables such as water tem- perature, dissolved oxygen, calcium concentration, and current velocity (MCMAHON, 1983; LAM & CALow, 1989). Freshwater prosobranch snails also have considerable potential for eco-phenotypic life-history variation. First of all, they can have reduced tolerance adaptation to tem- perature, and greater sensitivity to temperature than pul- Page 2 monates (e.g., greater Q,, values, MCMAHON 1983). Be- cause they cannot rely on aerial respiration, as pulmonates can, and are oxyconformers (7.e., show little regulation of oxygen consumption), they may also be more sensitive to hypoxia than are pulmonates (MCMauon, 1983). Vivipa- rid prosobranchs also show a variety of feeding mechanisms that could promote increased phenotypic variation, in- cluding grazing on periphyton, deposit feeding, and even filter feeding (ALDRIDGE, 1983). Finally, viviparids may show greater levels of population divergence because, un- like pulmonates, which are passively dispersed by birds, viviparids are limited to the relatively poor dispersal abil- ities of adults along river systems (CLARKE, 1981; DAvIs, 1982). Prosobranch populations in separate drainage ba- sins are therefore isolated, and speciation seems to have occurred at greater rates than in pulmonates (RUSSELL-HUNTER, 1978; Davis, 1982). For example, studies of electrophoretic variation in riverine proso- branchs indicate high levels of genetic variation among populations (CHAMBERS, 1980; DILLON & Davis, 1980; see review in BROWN & RICHARDSON, 1988). Divergence in population characters may also alter en- ergy flow through snail populations. Estimates of second- ary production and biomass turnover are tightly linked to life-history traits (WATERS, 1979; BENKE, 1984). Thus, environmental variables affecting phenotypic expression of prosobranch life-history traits may also lead to variation in annual population production and P:B (production to biomass) ratios. Understanding intraspecific life-history variation is thus necessary to understand differences in production that exist among populations. Little is known, however, about the degree of intraspe- cific life-history or production variation exhibited by south- eastern populations of any prosobranch species, despite the fact that the southeastern United States is a center of diversity for the group (CLARKE, 1981; BURCH, 1982). Most recent studies of the life histories of viviparids have, for example, dealt with lentic populations in either the northeastern or midwestern United States (BROWN, 1978; JOKINEN et al., 1982; PACE & SzucH, 1985; BUCKLEY, 1986) or in Europe (RiBI & GEBHARDT, 1986; GEBHARDT & Rist, 1987) and, for the most part, have not included production estimates. ‘Two viviparid species, Viviparus sub- purpureus (Say, 1829) and Campeloma decisum (Say, 1816), are common in tributaries of the southern reaches of the Mississippi River (BURCH, 1982) and have received some attention (BROWN ef al., 1989; RICHARDSON & BROWN, 1989); they are suitable subjects for such a comparative study. Our purpose here is to delineate the range of intraspe- cific variation occurring in both species at two sites in southern Louisiana that differ markedly in several abiotic factors, including detrital abundance (a possible food re- source, ALDRIDGE, 1983), water temperature, hardness, and dissolved oxygen saturation. We report differences in snail density through time, differences in individual growth The Veliger, Vol. 35, No. 1 and cohort dynamics, female reproductive rates, and sec- ondary production and P:B ratios for each species. We also compare the depth distributions of snails at each site, to determine if these snails are limited to shallow depths by hypoxic conditions in deeper waters. We also perform a “common garden” experiment, where sub-adults from both populations of one species are reared together at one site, to determine if life-history differences will disappear, suggesting that the variation is eco-phenotypic. Finally, we speculate on the relative importance of different eco- logical causes of the variation. METHODS Description of Habitats Bayou Manchac (BM, also known as the Iberville Riv- er) flows through Baton Rouge, Louisiana, and is a trib- utary of the Amite River, which flows into Lake Mau- repas. It has an extremely low gradient and slow flow, is approximately 10 m wide, varies from 3 to 6 m in depth, and has a clay substrate with a median particle size of 0.01 mm. It is surrounded with a rich riparian forest, which provides abundant allochthonous detritus each win- ter. Average wet mass of coarse organic detritus (defined as that retained on a 1.0-mm sieve) was 6.02 kg/m? + 0.40 (standard error of the mean, n = 88). Average fine organic content of the remaining sediment (estimated by ashing aliquots of sediment at 550°C overnight) was 5.7 + 0.6% (n = 21). Average water temperature recorded over the sampling interval (October 1986 to October 1988) was 24.4°C + 1.8°C (maximum recorded was 28°C) and average dissolved oxygen level was 2.3 mg/L + 0.5, with a minimum of 0.5 and a maximum of 3.5 mg/L. Average water hardness was 68.3 + 9.9 mg/L (Ca** + Mg?*) and the average pH was 6.8 + 0.1. The sampling site was near the overpass of Highway 61 in southeastern Baton Rouge. Old River (OR, also known as Lake Raccourci) is an ox-bow lake 90 km northwest of Baton Rouge. The ox- bow is connected via a channel to the Mississippi River, and water-level changes are frequent, retarding the growth of trees near the shoreline, and limiting allochthonous in- put. Average coarse organic detritus was only 0.43 kg/m? + 0.04 (n = 59), and percent fine organic content only 0.9 + 0.1% (n = 17). The absence of a riparian canopy resulted in a much higher average water temperature (mean equal to 28.4 + 1.9°C, maximum recorded was 35°C). Dissolved oxygen values were at or near saturation at all sampling dates at OR (mean equal to 9.9 mg/L + 0.6). Average water hardness was much greater at OR (mean equal to 175.9 + 29.7 mg/L), as was pH (mean equal to 8.6 + 0.2). At the sampling site, a sand bar along the north shore of the lake, 90% of the sediment particles were greater than 0.25 mm in diameter. Voucher specimens for each species and site have been deposited at the Los Angeles County Museum of Natural K. M. Brown & T. D. Richardson, 1992 History: Viviparus subpurpureus from BM (LACM 91- 103.1) and from OR (LACM 91-104.1) and Campeloma decisum from BM (LACM 91-103.2) and from OR (LACM 91-104.2). Sampling Methods Both sites were sampled approximately monthly during the summer, but less frequently during winter periods of high water, from fall 1986 (BM) or spring 1987 (OR) to fall 1988. At each date, 10 replicate samples were taken with a 15-by-15-cm Ekman grab. Ten replicate samples were enough for standard errors to be approximately 20% of the mean for snail densities, an acceptable level for benthic studies (ELLIOT, 1976). Counts of snails were con- verted to densities per m’. Samples were sorted through a series of sieves (smallest diameter = 1.0 mm), and the snails removed and sorted to species, with individual shell lengths determined to the nearest 0.1 mm using a caliper. A 1.0-mm sieve retained all snails because embryos are released at sizes =3 mm (BROWN et al., 1989). Life-History Traits and Secondary Production Size-frequency histograms were used to follow cohorts and estimate life-cycle length and individual growth rates. We approximated life-cycle length by determining the pe- riod of time between the appearance of juveniles and dis- appearance of adults in cohorts, and assumed individual growth rates were similar to changes in the median shell size of cohorts through time. To estimate population sex ratios, as well as the size that males and females reached in each population, approximately 30 randomly selected snails (for each species and site) were dissected at each sampling date and sexed, and the number of embryos brooded in each female recorded. Male viviparids are easily distinguished by their larger right tentacle, which is used in copulation. For a subset of these animals, adult dry tissue masses (including eggs in females) were determined on an analytical balance (sensitivity = 0.1 mg) after drying overnight at 60°C. A two-way ANOVA (sex versus site) was performed for each species to detect any site-dependent effects on, or sexual dimorphism in, shell length and dry body mass. For the estimation of secondary production, regressions of snail shell-free dry mass versus shell length were used to estimate sample standing stocks from the shell size- frequency distributions in samples (see discussion of meth- ods in RICHARDSON & BRowN, 1989). We then multiplied these size-specific dry masses times the density of the size classes in samples to calculate total snail biomass/m?. Re- gressions of dry tissue mass versus shell length had coef- ficients of determination (r?) ranging from a low of 0.74 (n = 50) for Campeloma decisum at OR to a high of 0.88 (n = 50) for C. decisum at OR. Production was estimated with the size-frequency method (HYNES & COLEMAN, 1968; HAMILTON, 1969) as modified by KRUEGER & MARTIN Page 3 (1980). Production estimates include embryo dry mass (see methods in RICHARDSON & BROWN, 1989), and are cor- rected for cohort production interval (CPI, BENKE, 1979) and unequal sampling intervals (KRUEGER & MARTIN, 1980). Negative production values for smaller size classes are not included in the overall estimate, following BENKE (1984). Depth Distributions At two separate dates for each site during summer 1989, we determined the vertical distribution of temperature and dissolved oxygen using a YSI meter, and then collected three replicate samples at 0.5, 1.5, and 2.5 m depth at each site. These depths were selected because the maximum depth of BM in summer 1989 was approximately 3 m. Samples were processed as discussed above, and the log- transformed (to remove a mean-variance correlation) den- sities of both species were subjected to a two-way ANOVA at each site to determine, for each species, the effect of sampling date and depth on abundance. Field Growth Experiment Finally, we performed a field experiment to determine if life-history differences between sites in one species had a strong genetic component, or were due more to such environmental differences as food levels. ‘This experiment was performed in BM from April to September 1988, with Viviparus subpurpureus. Sub-adults (15.0 + 1.0 mm ini- tial mean shell length) from both source sites were held at representative field densities in cages at the sampling site in BM. We determined the average shell length and wet total mass (shell plus tissue) attained by snails at the end of the experiment, and compared them between the two source populations with ¢-tests to determine any dif- ferences in growth. We also dissected all females to de- termine any differences in brood size. Cages were plastic trays (52 x 63 x 11 cm, with a bottom surface area of roughly one-third m7’), similar to bread storage trays used in bakeries (Allibert Industries, Quebec, Canada). They were made of impact resistant plastic, open at the top, and had mesh sides and bottoms with openings 5 x 5 cm. The trays were covered and lined with plastic mesh (mesh diameter 2 mm) hot glued to the trays to retain all snails, including embryos. Trays were placed at approximately 0.5 m depth (the depth at which most of the previous sampling occurred), and were allowed to settle into the clay substrate so that detritus would be available to the snails. RESULTS Density Both species were roughly four times as abundant, on the average, at BM than at OR (Figure 1). For Viviparus subpurpureus, populations declined to lows of about 100 Page 4 10007 Campeloma decisum 800F @ BAYOU MANCHAC ai oO OLD RIVER = 600t a 600 WW $ 400 re) S$ 400 = eee. PND cra. MONI resco SFT SAPS LSE (Eu Gnu 2c ac slo 2 2,000 1,600} Viviparus subpurpureus ® BAYOU MANCHAC 1,200 O OLD RIVER No PER M@ So (o) {a} 1986 1987 1988 Figure 1 Temporal changes in the density (+ SE, n = 10) of Viviparus subpurpureus and Campeloma decisum at two warm-temperate sites in southern Louisiana. to 300 snails/m? after winter floods (BROWN et al., 1989), but reached densities as high as 1700/m? after recruitment of embryos in the summer. Corresponding densities at OR ranged from 50 to 250/m?. To determine whether abun- dances differed significantly between sites, we performed a one-way ANOVA with each site and date combination as a separate treatment. Densities were log transformed to remove a mean-variance correlation. Using an orthog- onal contrast, we found that V. subpurpureus was signif- icantly more abundant at BM (Ff = 135.3, P < 0.001). For Campeloma decisum, densities at BM ranged from 30 to 850/m?’, versus peak densities of only 150/m? at OR. A similar one-way ANOVA with an orthogonal contrast between sites revealed these densities to again be highly significantly different (F = 50.6, P < 0.001). Cohort Dynamics and Life-History Variation At BM, adult Viviparus subpurpureus produced a large cohort of juveniles during March and April 1987 (Figure 2). By October, these snails had reached a mean shell The Veliger, Vol. 35, No. 1 length of 13 mm. Since juveniles are released at 3 mm shell length, this represents a growth rate of 1.43 mm per month (e.g., a 10-mm increment over seven months). Mean shell length of this same cohort eventually reached 20 mm in fall 1988, when the snails had reached an age of one and a half years. Some of the original adult cohort probably survived until the fall of 1987, but most died after repro- ducing, explaining why there are few snails in the size classes of 16 to 26 mm during the last six months of 1987 (Figure 2). When sampling began at OR in spring 1987 (Figure 3), Viviparus subpurpureus, probably born earlier that spring, averaged 13 mm in shell length, or about 3 or 4 mm longer than the corresponding cohort in BM. These snails had reached an average shell length of 23 mm by fall 1987, or 10 mm greater than the corresponding cohort in BM. Some juveniles were born to this cohort in fall 1987, but a major release of juveniles occurred just before the first sampling in 1988, probably in April. These snails had again grown to an average size of 23 mm by fall 1988, for an average growth rate of 2.86 mm per month (20- mm increment in seven months) or roughly twice the growth rate in OR. Either these snails, or some surviving older adults, had started releasing embryos by fall 1988 (Figure oy) The sex ratio of Viviparus subpurpureus was biased to- wards females at BM (x? test, n = 275, P < 0.01) but not at OR (n = 212, P = 0.07). Female V. subpurpureus at both sites had significantly greater shell lengths (Table 1, P < 0.0001) and body dry tissue masses (Table 1, P < 0.001) than did males. The greater average tissue masses of females could be due, however, to the embryos they were holding, since embryos were included in these estimates. Viviparus subpurpureus also reached greater average shell lengths (P < 0.0001) and body dry masses (P < 0.001) at OR than at BM. Females started brooding young at a shell length of 16 mm at BM versus a shell length of 19 mm at OR. Clutch- size versus shell-length regressions had similar slopes (P > 0.05) but significantly different intercepts (P < 0.0001) indicating that female Viviparus subpurpureus at any shell length in BM were brooding more embryos than equiv- alent OR females. However, the larger female size at OR explains why females did not differ significantly in mean clutch size between sites (about 10 embryos per female at each site, P > 0.05). Campeloma decisum produced a large cohort of juveniles in March 1987 at BM (Figure 4), and these snails had reached an average size of 15 mm by October, for an average growth rate of 1.71 mm per month (12-mm in- crement in seven months). This cohort reached a maxi- mum shell length of about 20 mm in late 1988 at an approximate age of one and a half years. A second cohort, born in April 1988, had reached a size of 12 mm by October 1988, with an average growth rate of 1.29 mm per month (e.g., 9-mm growth increment divided by seven months). Some snails from the original cohort of adults seen in fall K. M. Brown & T. D. Richardson, 1992 (mm) SHELL LENGTH BAYOU MANCHAC Viviparus subpurpureus | 292 21 689 32 28 B- 10% Hi 100 282 24 162 135 316 413 270 1ONN7 W721 2/13 4/3 7/3 8/27 9/2410/2312/1 IAG 4/22 5/306/20 7/25 8/8 10/19 1986 1987 1988 Figure 2 Percentile size-frequency histograms for cohorts of Viviparus subpurpureus at Bayou Manchac. Sample sizes are given below histograms, and sampling dates are given on the x-axis. OLD RIVER Viviparus subpurpureus 32 B-10% 28 (mm) Sle, LleING Tal VS We WZ" 272) 2x o/3 CAS Wis SEF CSO VA! IIS’ 1988 Figure 3 Percentile size-frequency histograms for cohorts of Viviparus subpurpureus at Old River. Sample sizes are given below histograms, and sampling dates are given on the x-axis. Page 5 Page 6 BAYOU MANCHAC @-10% 28 | fe 24 € 20 Ts i 15 Zea LWW =} ai We gl} LW me 147 144 192 233 13 32 94 The Veliger, Vol. 35, No. 1 Campeloma decisum \O/MI7 11721 2/13 4/3 7/6) 8/27 ~9/24 10/23 12/1 1/16 4/22 5/30 6/20 7/25 8/18 10/19 1986 1987 1988 Figure 4 Percentile size-frequency histograms for cohorts of Campeloma decisum at Bayou Manchac. Sample sizes are given below histograms, and sampling dates are given on the x-axis. 1986 were found in samples as late as mid-summer 1988. Because in 1986 these snails averaged the same size as one-and-a-half year olds in fall 1988 (e.g., 20 to 24 mm, Figure 4), they would be three years old in summer 1988. At OR, two cohorts of Campeloma decisum were present in June 1987 (Figure 5). The younger cohort grew from about 8 mm shell length in June to 16 mm by November, merging with the older cohort. This combined cohort sur- vived the winter, and produced a large cohort of juveniles just before sampling started (again delayed due to high waters) in spring 1988. Assuming these young were pro- duced in April, and since they reached an average size of 18 mm by November 1988, growth rates would average 13 to 15 mm per seven months, or about 2 mm per month. Again, as in Viviparus subpurpureus, cohorts of C. decisum reached larger average sizes and appeared to have greater average growth rates at OR than at BM. The sex ratio of Campeloma decisum was biased towards females at BM (n = 226, P < 0.01), but males at OR (n = 198, P < 0.01). Thus, three of the four populations had biased sex ratios, with females favored in both cases at BM, and sexes either equally abundant or males more common (in C. decisum) at OR. Female C. decisum reached greater shell lengths on the average than males (Table 1, Table 1 Average life-history traits (+ SE) for both sexes of Viviparus subpurpureus and Campeloma decisum at both sites, determined by dissecting individuals sampled from the field. Sample sizes are given in parentheses. Note that only a subsample of snails had shell and body dry masses determined. Viviparus subpurpureus Campeloma decisum Site Sex Shell length Body mass Shell length Body mass Bayou Manchac female 18.8 + 0.2 98.2 + 3.3 20.1 + 0.4 106.6 + 6.6 (n = 182) (n = 92) (n = 168) (n = 84) male 15265-9032 63.7 + 3.8 16.8 + 0.5 82.7 + 10.7 (n = 93) (n = 54) (n = 58) (n = 20) Old River female Post a= (Oil 114.0 + 9.8 22.6 + 0.6 79.1 + 12.6 (n = 119) (n = 48) (n = 72) (n = 22) male 19.0 + 0.4 104.7 + 11.0 21.4 + 0.3 107.1 + 6.7 (n = 93) (n = 36) (n = 126) (n = 44) K. M. Brown & T. D. Richardson, 1992 | OLID. Rien a 24 20 16 =10% (mm) LENGTH Sllelel CHOP 19) 8/2 9724 i712 1987 Page 7 Campeloma decisum wi 136 58 74 92 8 6/13 TIS 8B/14 9/30 114 1988 Figure 5 Percentile size-frequency histograms for cohorts of Campeloma decisum at Old River. Sample sizes are given below histograms, and sampling dates are given on the x-axis. P < 0.0001) but not greater body dry tissue masses (P > 0.05), even though embryos were again included in female tissue mass. However, there was also a significant site x sex interaction for shell length (P < 0.05), indicating that sexual dimorphism in shell length was more obvious at BM, as well as a significant interaction between sex and site for body mass (P < 0.05) with the effect of sex reversed between habitats (Table 1). Finally, C. decisum had sig- nificantly greater shell lengths at OR (Table 1, P < 0.0001) but not greater dry body tissue masses (P > 0.05) than at BM. At Bayou Manchac, female Campeloma decisum start brooding embryos at shell lengths of 19 mm, compared to 21 mm at Old River. Clutch size also increases more rapidly with shell length at BM than at OR (P < 0.001). Thus the larger females at OR again do not have signif- icantly greater average clutch sizes (about 12 embryos per female at each site). Secondary Production These density, life-history, and growth-rate differences among populations can be summarized easily by compar- ing annual secondary production, which in one value in- tegrates density, life-cycle length, reproductive patterns, and growth (BENKE, 1984). Production was significantly higher (comparing 95% C.1.) for Viviparus subpurpureus at BM than at OR, primarily because of higher biomass levels (Table 2). Even though growth rates were greater at OR, lower average biomass resulted in a 79% decrease in production between sites. Complete turnover of biomass, however, occurred almost two times faster at OR, reflecting faster individual growth at this site (Table 2). For Cam- peloma decisum, both production and biomass were also dramatically lower at OR, again because greater growth rates were counterbalanced by lower annual biomasses. Turnover rates (P:B ratios) were again about twice as great at OR (Table 2). Table 2 Annual production (P) and mean biomass (B) in grams of shell-free dry mass/m?, and P:B ratios for Viviparus subpurpureus and Campeloma decisum. Estimates were cal- culated using the size-frequency method. 95% confidence intervals are in parentheses. Bayou Manchac Old River V. subpurpureus 1p 81.8 (19.0) 17.2 (5.9) B 32.3 (8.2) 4.0 (1.7) P:B 2.5 4.3 C. decisum P 31.3 (7.42) 11.1 (5.7) B 15.5 (4.6) Dea @le)) P:B 2.0 4.1 Page 8 The Veliger, Vol. 35, No. 1 Table 3 Comparison of reproductive performance for two source populations of Viviparus subpurpureus reared at Bayou Manchac in cages; see text for methods. Sample sizes are in parentheses. Source Trait Bayou Manchac Old River JP Mean final shell length (mm) 18.9 + 0.2 18.3 + 0.4 >0.05 (55) (20) Mean final total mass (mg) 1767.0 + 59.0 1598.0 + 96.0 >0.05 (55) (20) Brood number 14.5 + 1.2 Neve so DY >0.05 (49) (15) * Significance of t-test. Depth Distributions Both BM and OR are relatively shallow habitats, and neither stratified thermally during summer 1989. How- ever, OR had temperatures averaging 3°C greater than those at BM on the first set of sampling dates, which were within two days of each other. This is due to lack of a riparian shade cover at OR. A substantial difference also occurred in dissolved oxygen profiles, with BM values averaging only 18% of saturation versus 122% at OR. Vertical stratification of dissolved oxygen was more pro- nounced at both sites on the second sampling date. At BM, there was no consistent relationship between the density of either Viviparus subpurpureus or Campeloma decisum and sampling depth (P > 0.05). At OR, however, densities declined much more rapidly for each species with depth. While there was no significant difference in the density of Vivzparus subpurpureus between dates, there was a highly significant effect for sample depth (P < 0.01). Similarly, average densities did not differ between dates for Campeloma decisum (P > 0.05) but did among depths (P < 0.01). Field Rearing Results The results of the field rearing experiment (Table 3) indicated little difference in growth or reproductive per- Table 4 Summary of the relative magnitudes of average physico- chemical and demographic variables for the two vivivarid species at two sites in Louisiana. Variable Bayou Manchac Old River Detrital abundance 6x x Water temperature o x + 4°C Dissolved oxygen 28% saturation 128% saturation Water hardness x LES 26 Snail densities ca. 4x x Life-cycle lengths ca. 2x 5 Individual growth a ca. 1.5-2x Production ca. 3-5x x formance between snails from the two populations when Viviparus subpurpureus were held in a common environ- ment (BM). Both average shell length and wet total mass (shell plus body) were very similar for individuals from the two populations at the end of the experiment (Table 3). Female brood sizes did not differ between the two populations either (Table 3). Since snails had initially similar sizes, we would have expected snails from OR to have reached larger sizes if the growth differences shown by this population were of genetic origin. DISCUSSION Life History Differences Among Sites Considerable intraspecific variation occurred in life-his- tory patterns of both species at both sites (summarized in Table 4). Viviparus subpurpureus reached dramatically greater abundances at BM, but individual growth rates were half those at OR, and both males and females reached smaller final adult sizes than at OR. The size-frequency histograms and dissections of females suggest that V. swb- purpureus is biennial and semelparous at BM, but annual at OR. For example, because year-old females were only 14 mm long on the average at BM, and brooding did not start in this habitat until a shell length of 16 mm, females probably did not mature until an age of one and a half years, in their second fall. Furthermore, because a large cohort of juveniles was not produced in the fall, these females probably brooded their young over their second winter and released them in their third spring at an age of two years. In contrast females at OR reached sizes at which embryos were present in brood pouches by the end of their first summer (at an age of one-half year). These females may have released some young (a few juveniles were present in the fall), but most females probably brood- ed embryos over their first winter, and released them in the spring at an age of one year, dying shortly thereafter. Campeloma decisum also reached higher densities at BM, but again had individual growth rates, about 55% greater, at OR (Table 4). Size-frequency histograms and dissec- tions also suggest that C. deciswm is biennial and semel- K. M. Brown & T. D. Richardson, 1992 parous, or even triennial and iteroparous at BM, but an- nual and semelparous at OR. For example, females at BM evidently do not reach a shell length where they can brood young (18-19 mm) until their second fall, when they are one and a half years old. Thus they probably brood their young over the winter and release them in spring, at an age of two years. A fraction of these females, judging from the size-frequency histograms, may survive and reproduce in a third year. In contrast, females at OR have reached a size where they can brood young by the end of their first summer (at an age of one-half year), and probably brood embryos over the winter, releasing them the next spring. The size-frequency histograms indicate that some females may survive to the end of summer, for a life-cycle length of one and a half years. These life-history differences between BM and OR pop- ulations of both species are also reflected in the production data (Table 4). Because of the higher densities, standing stocks and production were considerably greater at BM for both species. The magnitude of production depends both on standing stocks and the annual P:B ratio (BENKE, 1984), and higher production at BM was thus apparently the result of higher standing stocks. Disparity in produc- tion and biomass estimates are not uncommon between molluscan populations, and these differences for Viviparus subpurpureus and Campeloma decisum fall within the ob- served range (see review in RUSSELL-HUNTER & BUCKLEY, 1983). However, annual turnover ratios (P:B ratios) were ap- proximately twice as high at OR for both Viviparus sub- purpureus and Campeloma decisum. Because turnover ra- tios are dependent on development times (WATERS, 1979; BENKE, 1979) and are also inversely related to the length of the life cycle (RUSSELL- HUNTER & BUCKLEY, 1983), it is understandable that annual P:B ratios were higher at the site with fastest growth and shortest snail longevity (OR). The P:B ratios for the annual populations at OR are within the range of 2.1-9.2 reported for other semel- parous, annual snails. The BM populations, with some- what longer life-spans, have ratios fairly similar to those reported for snails with triennial or biennial, semelparous populations (1.0-2.0, RUSSELL-HUNTER & BUCKLEY, 1983). The life-history and production data are intriguing in comparison to other intraspecific studies of life-history variation in freshwater snails, for a number of reasons. First, the length of the life cycles, especially those at OR, are some of the shortest recorded for freshwater proso- branchs. A more typical pattern for north-temperate pop- ulations of viviparid prosobranchs is for females to mature (e.g., give birth to offspring) at the earliest at two years of age and to be semelparous, or in some cases to reproduce for several years (BROWNE, 1978; JOKINEN et al., 1982; BUCKLEY, 1986; see ALDRIDGE, 1983, for a review). The densities recorded at BM are also much higher than den- sities recorded in even exceptionally abundant north-tem- perate populations (PACE & SZUCH, 1985). However, males Page 9 in these warm-temperate populations still reach smaller shell lengths and/or possibly have shorter life cycles, sim- ilar to the sexual dimorphism seen in more northern pop- ulations. The resulting high standing crops, as well as the short life cycles, result in secondary production estimates that are much higher than those for most north-temperate, gastropod populations (see review in RUSSELL-HUNTER & BUCKLEY, 1983), and indeed are some of the highest known for any lotic organism (see BENKE, 1984). Evidentally the warmer water temperatures, longer growing seasons, and/ or greater food resources may explain the shorter life cy- cles, rapid growth, and greater productivity of these warm- temperate populations (RICHARDSON & BRowNn, 1989). The differences in growth rates and life-cycle lengths between habitats in this study are paradoxical in light of earlier studies. In north-temperate pulmonates and proso- branchs, increased food availability not only causes higher densities and secondary production, but also greater in- dividual growth rates, shorter life cycles, and greater re- productive rates (see reviews in RUSSELL-HUNTER, 1978; McMauon, 1983; ALDRIDGE, 1983). The higher densities and secondary production of both species at BM (the de- tritus-rich site) are as expected since Campeloma decisum is considered a detritivore or scavenger (PACE & SZUCH, 1985) and other species of Viviparus are known to deposit feed (see ALDRIDGE, 1983). The lower growth rates and longer life cycles at BM than at OR are, however, not as expected, and we consider potential causes below. Potential Causes of Intraspecific Life-History Variation We will first discuss those potential causes that seem less likely to be important. The intraspecific life-history variation in these populations does not appear to have a strong genetic basis, since life-history differences disap- peared upon rearing individuals in a common environment. We consider the experiment a fairly conservative test, since the snails were actually sub-adults that had spent about a year in their source habitats before the experiment began. These data suggest that much of the variation is eco-phe- notypic (z.e., of ecological origin). What then are the important ecological variables? McManuon (1983) and ALDRIDGE (1983) reviewed the physiological ecology of pulmonates and prosobranchs, re- spectively, and listed a large number of studies suggesting that temperature, water hardness, and resource levels are important determinants of life-history variation. Of these, the least likely single factor in this study appears to be dissolved oxygen concentration. The lowest oxygen levels occurred at BM (Table 4), where population densities were highest. Furthermore, snails were distributed down to the greatest depths at BM. There was actually more of a decrease in density with depth at the more oxygenated site, OR. However, lower oxygen levels may still have affected the energy available for individual growth in the following way. If most of the respiration at BM was an- Page 10 aerobic, because of low oxygen levels, then the decreased efficiency of anaerobic metabolism might have significantly lowered the amount of non-respired, assimilated energy available for growth. Thus even with the greater food resources, snails at BM might have lower growth rates than snails at OR (if these snails were able to use aerobic respiration and to more efficiently convert assimilated en- ergy into growth). On the other hand, average water temperatures were also higher at OR (Table 4) during summer months be- cause of the absence of a leaf canopy. Water temperature dramatically increases feeding rates and shell deposition, resulting in increased growth and shorter life cycles both in pulmonates (MCMAHON, 1983) and prosobranchs (AL- DRIDGE, 1983). In addition, OR has harder water (Table 4), which may also increase the rate of shell deposition. Thus the warmer, harder, and more oxygenated water at OR may synergistically promote more rapid growth rates. Furthermore, this explanation is not necessarily counter to the observed higher densities and fecundities for females (at any given shell length) at BM. During the winter months, when females are brooding embryos, differences in water temperature and oxygen concentration are not as large between the two habitats. The riparian vegetation along BM looses its leaves, allowing greater absorption of incoming solar radiation by the water, and lower water temperatures overall result in similar dissolved oxygen levels. Two final environmental factors should be considered. First, the overall higher density of snails at BM may have resulted in a density-dependent reduction in individual growth rates, as has been recorded often in freshwater snails (e.g., among others, EISENBERG, 1970; HUNTER, 1975; BRowN, 1985). However, it is hard to understand why density-dependent reductions in fecundity, just as often if not more widely documented, would not occur as well. Finally, size-specific predation by crayfish can alter cohort size distributions, either by the preferential removal of smaller snails, or by causing snails to increase their allo- cation to growth versus reproduction (CROWL & COvICcH, 1990). However, the crayfish Procambarus clarku is com- mon in BM, and is a voracious feeder on the young of both snail species (VARZA, 1989), suggesting that the growth-rate differences between the two habitats were not due to crayfish predation. Regardless of the causal mechanisms, the eco-pheno- typic variation in the life histories of these prosobranch snails has some important implications for prosobranch evolutionary biology. First, considerable phenotypic plas- ticity occurs in the life histories of these populations. Such plasticity is undoubtedly adaptive, since it allows responses to changes in environmental variation (see discussion in RUSSELL-HUNTER, 1978). In this study, for example, fe- males of a given size at the site with apparently lower food resources (OR) did not brood as many embryos. By grow- ing to a larger size, however, they could still produce an The Veliger, Vol. 35, No. 1 equivalent number of offspring, conserving individual fit- ness. ACKNOWLEDGMENTS We would like to acknowledge help of D. E. Varza, E. Haight, P. Pham, and H. Nguyen in sampling and in processing the samples. Vicki Lancaster and Philip Rut- ledge helped with the statistical analysis of the data. Two anonymous reviewers of an earlier manuscript also helped considerably. This research was supported by NSF grant DEB 85-00774 to the senior author. LITERATURE CITED ALDRIDGE, D. W. 1983. Physiological ecology of freshwater prosobranchs. Pp. 329-358. In: W. D. Russell-Hunter (ed.), The Mollusca. Vol. 6, Ecology. Academic Press: Orlando, Florida. BENKE, A. C. 1979. A modification of the Hyne’s method for estimating secondary production with particular significance for multivoltine populations. Limnology and Oceanography 24:168-171. BENKE, A. C. 1984. Secondary production of aquatic insects. Pp. 289-322. In: V. H. Resh & D. M. Rosenberg (eds.), The Ecology of Aquatic Insects. Praegen Publishers: New York. Brown, K. M. & T. D. RICHARDSON. 1988. Genetic poly- morphism in gastropods: a comparison of methods and hab- itat scales. Bulletin of the American Malacological Society 6:9-17. Brown, K. M. 1985. Intraspecific life history variation in a pond snail: the roles of population divergence and phenotypic plasticity. Evolution 39:387-395. Brown, K. M., D. Varza & T. D. RICHARDSON. 1989. Life histories and population dynamics of two subtropical vivipa- rid snails (Prosobranchia: Viviparidae). Journal of the North American Benthological Society 8:222-228. BROWNE, R. A. 1978. Growth, mortality, fecundity, biomass, and productivity of four lake populations of the prosobranch snail, Viviparus georgianus. Ecology 59:742-750. BuCcKLEY, D. E. 1986. Bioenergetics of age-related versus size- related reproductive tactics in female Viviparus georgianus. Biological Journal of the Linnaean Society, London 27:293- 309. Burcu, J.B. 1982. Freshwater Snails (Mollusca: Gastropoda) of North America. Malacological Publications: Ann Arbor, Michigan. CHAMBERS, S. M. 1980. Genetic divergence between popula- tions of Goniobasis (Pleuroceridae) occupying different drain- age systems. Malacologia 20:63-81. CLaRKE, A. H. 1981. The freshwater molluscs of Canada. National Museums of Canada, Ottawa, Canada. CrowL, T. D. & A. P. CovicH. 1990. Predator-induced life history shifts in a freshwater snail. Science 247:949-951. Davis, G. M. 1982. Historical and ecological factors in the evolution, adaptive radiation, and biogeography of fresh- water molluscs. American Zoologist 22:375-395. DILLon, R. T., JR., & G. M. Davis. 1980. The Goniobasis of southern Virginia and northwestern North Carolina: genetic and shell morphometric relationships. Malacologia 20:83- 98. EISENBERG, R. M. 1966. The regulation of population density K. M. Brown & T. D. Richardson, 1992 Page 11 in a natural population of the pond snail Lymnaea elodes. Ecology 47:889-905. EISENBERG, R. M. 1970. The role of food in the regulation of the pond snail, Lymnaea elodes. Ecology 51:680-684. ELuioT, J. M. 1976. Some methods for the statistical analysis of samples of benthic invertebrates. Freshwater Biological Association, Scientific Publication 25. GEBHARDT, M. & G. Risi. 1987. Reproductive effort and growth in the prosobranch snail Viviparus ater. Oecologia 74:209- 214. HAMILTON, A. L. 1969. On estimating annual production. Limnology and Oceanography 14:771-782. Hunter, R. D. 1975. Growth, fecundity, and bioenergetics in three populations of Lymnaea palustris in upstate New York. Ecology 56:50-63. Hynes, H. B. N. & M. J. COLEMAN. 1968. A simple method of assessing the annual production of stream benthos. Lim- nology and Oceanography 13:569-573. JOKINEN, E. H., J. GUERETTE & R. W. KORTMANN. 1982. The natural history of an ovoviviparous snail Viviparus georgianus (Lea) in a soft-water eutrophic lake. Freshwater Inverte- brate Biology 1(4):1-17. KRUEGER, C. C. & F. B. MarTIN. 1980. Computation of con- fidence intervals for the size frequency (Hynes) method of estimating secondary production. Limnology and Oceanog- raphy 25:773-777. LaM, P. K. S. & P. CaLtow. 1989. Intraspecific life history variation in Lymnaea peregra (Gastropoda: Pulmonata) I. Field study. Journal of Animal Ecology 58:571-588. McManon, R. F., R. D. HUNTER & W. D. RUSSELL-HUNTER. 1974. Variation in aufwuchs at six freshwater habitats in terms of carbon biomass and carbon:nitrogen ratios. Hydro- biologia 45:391-394. McManuon, R. F. 1983. Physiological ecology of freshwater pulmonates. Pp. 359-430. In: W. D. Russell-Hunter (ed.), The Mollusca. Vol. 6, Ecology. Academic Press: Orlando, Florida. Pace, G. L. & E. J. SzucH. 1985. An exceptional stream population of the banded applesnail, Viviparus georgianus, in Michigan. Nautilus 99:48-53. PARTRIDGE, L. & P. H. HARVEY. 1988. The ecological context of life history evolution. Science 241:1449-1455. Rist, G. & M. GEBHARDT. 1986. Age specific fecundity and size of offspring in the prosobranch snail, Viviparus ater. Oecologia 71:18-24. RICHARDSON, T. D. & K. M. Brown. 1989. Secondary pro- duction of two subtropical snails (Prosobranchia: Vivipari- dae). Journal of the North American Benthological Society 8:229-236. RUSSELL-HUNTER, W. D. 1978. Ecology of freshwater pul- monates. Pp. 335-383. In: V. Fretter & J. Peake (eds.), The Pulmonates. Vol. 2A, Systematics, Evolution and Ecology. Academic Press: Orlando, Florida. RUSSELL-HUNTER, W. D. & D. E. BUCKLEY. 1983. Actuarial bioenergetics of nonmarine molluscan productivity. Pp. 464- 503. In: W. D. Russell-Hunter (ed.), The Mollusca. Vol. 6, Ecology. Academic Press: Orlando, Florida. VaRZA, D. E. 1989. The effect of ovoviviparous reproduction on juvenile survival in the prosobranch snail, Viviparus sub- purpureus. Master’s Thesis, Louisiana State University. 72 PP- Waters, T. F. 1979. Influence of benthos life history upon the estimation of secondary production. Journal of the Fish- eries Research Board of Canada 36:1425-1430. The Veliger 35(1):12-15 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 Multiple Paternity in Crepidula fornicata (Linnaeus) PATRICK M. GAFFNEY anpb BETH McGEE'! College of Marine Studies, University of Delaware, Lewes, Delaware 19958, USA Abstract. The protandric hermaphrodite Crepzdula fornicata (Linnaeus) is typically found in stacks containing individuals of both sexes as well as immature and transitional forms. Although multiple mating has been observed, it is not known whether it results in multiple paternity. Starch gel electro- phoresis of mothers and their broods demonstrated effective sperm storage and multiple paternity, as well as genetic contributions by individuals not present in the stacks at the time of collection. In order for C. fornicata to be useful as a model system for studies of sex change in the Mollusca, the use of more powerful, currently available methods is recommended to elucidate the ecological dynamics of fertilization and paternity. INTRODUCTION The calyptraeid snail Crepidula fornicata (Linnaeus) is a protandric hermaphrodite in which the change from male to female is influenced by various biotic factors, making it a potentially useful model system for the study of strategies of sex allocation (HOAGLAND, 1978; CHARNOV, 1982; WRIGHT, 1988). In order to understand the evolution of such strategies, we must known more about the “ecological dynamics of fertilization and paternity” (WRIGHT, 1988). Two aspects of particular importance in the case of C. fornicata are (1) whether sperm are stored, and if so, for how long, and (2) whether multiple paternity is possible. Although considerable indirect evidence indicates that many gastropods may store sperm for long periods, this phenomenon has not been studied in detail in proso- branchs. COE (1942) reported that Crepidula onyx was capable of storing sperm for more than a year; a similar observation was made by HOAGLAND (1978) for C. forni- cata. Evidence for multiple paternity in prosobranchs is likewise anecdotal, but it has been reported in several pulmonates (MURRAY, 1964; MULVEY & VRIJENHOEK, 1981). CoE (1936) described multiple insemination of a single female C. fornicata by several young males, and noted that young males “often attach themselves to the sides of groups as ‘supplementary’ males or settle down singly on other objects after having presumably participated in tem- porary matings with one or more females” (COE, 1938). ' Present address: Maryland Department of the Environment, 2500 Broening Highway, Baltimore, Maryland 21224, USA. ORTON (1950) reported that copulation took place between individuals occupying any position in a chain; on the other hand, WILCZYNSKI (1955) argued that copulation in stacks was severely restricted, with motile males probably siring the majority of offspring. ORTON (1952) provided circum- stantial evidence to suggest that isolated individuals may be self-fertilized by their own sperm upon transition to the female phase, but this hypothesis has yet to be tested experimentally. Our understanding of the ecological dynamics of fertil- ization and paternity in Crepidula is limited at present by a lack of empirical data. We report here the occurrence, detected by allozyme electrophoresis, of multiple paternity in a natural population of C. fornicata. METHODS anpD MATERIALS Crepidula fornicata were collected intertidally at Cape Hen- lopen, Delaware, in July 1988. For each substrate (7.e., a single rock, shell, or other hard object), all snails (one “substrate group”) were removed and labelled individually according to substrate, stack, and position in a stack. Sex was determined by examination of gross morphology under a dissecting microscope. Several stacks were dismantled, and females brooding larvae were placed in individual 4-L glass jars containing filtered seawater, where they re- mained until most or all of the larvae were released. The adults were then removed from the jars, and the larvae were fed daily with a mixture of /sochrysis galbana and Chaetoceros calcitrans. Larvae were cultured from a single brooding female from each of six substrate groups. Adults P. M. Gaffney & B. McGee, 1992 A4(m) a3(be)[ AB / AB / BD AB / BB / CC A4(t)| AA / BB / AG A3(f) BB / BC / AC a2£) | | B2(f) Page 13 A2(bf) BB / BB / CC B2(bf)| AC / AB / AC | AA’ / BB) / CD AB / BB / BC | Al(bf) BB / — / CD Bl(bf) | BB / BB / CC | Al(£) | AB / AB / CC | B1(£) BB / BB / CC C1(£) BB / BB / BC 2. 5p A6(j) — BB/BB/— A8(m);——_ BB / BB / BC A5(m)| «B/BB/CC A7(m)|BB / BB /BC Ad (bf) AB / BB / CC AG(E)|IVBCT/ BBY / BC legen AS(f) BB / BD / CD A2(bf) BC / BB / BB ASCE) Bf OS 7 Ge Ala(bf)} BB / BB / CC BC / AD / CC Alb (bf) PKC) BERBERS PCD OS ZS BS A2(f) AD / BB / AC B2(f)| BC / AB / CC Al(£) AC / BB / CD B1(f£) AB / BB / BC A4(m) | BB / AB / BD a3(£) [BC / BB / CC A2(f£) AB / BB / CC Al(£) [ AA / BB / CD | Shell length (cm) am [ a8 7 B87 80] p3(m)[_AB / BB / AC A2(bf) BB / BB / CC B2(f) AB / AB / BC Al (bf) BD / BB / BC Bl(bf) BB / BB / CC l | i ! J 0 1 2 3 4 5 Shell length (cm) Figure 1 Diagrams of six substrate groups showing stack composition and genotypes at three enzyme-coding loci (Gpi/Mpi- 2/Pgm). “ » indicates genotype not obtained. Individual snails are indicated by boxes, with shell length drawn approximately to scale indicated. Key: f, female; bf, brooding female; m, male; t, transitional; j, juvenile. Brood status of females in groups 4 and 5 was not recorded. Double lines mark the female whose brood was sampled. were held in the laboratory for approximately one week before being stored at —76°C for electrophoresis. All adults and a random sample of juveniles from each brood (when they had reached a minimum size of 4-6 mm) were homogenized in equal volumes of 0.1 M Tris pH 8.0 buffer containing 20% glycerol. After centrifugation at 5000 g for 5-10 min a portion of the supernatant from each sample was loaded onto starch gels. Several enzyme systems were stained: glucose phosphate isomerase (GPI, E.C. 5.3.1.9), mannose phosphate isomerase (MPI, E.C. 5.3.1.8), phosphoglucomutase (PGM, E.C. 2.7.5.1), and nonspecific esterases hydrolyzing the fluorescent substrate 4-methylumbelliferyl acetate. Enzymes that were scored reliably in both adults and juveniles were GPI, MPI, and PGM. Three regions of esterase staining (putative loci) were inconsistently scored in the progeny but were not resolved in adults. All enzymes were run on the discon- tinuous lithium hydroxide pH 8.4 buffer of SELANDER ef al. (1971) except for PGM, which was run on the 0.18 M Tris-borate-EDTA pH 8.7 buffer of BOYER et al. (1963). Alleles were designated alphabetically (A to D, in order of decreasing electrophoretic mobility). For two loci, Gpz and PGM, a faster-migrating electromorph was discovered later and called a. The initial null hypothesis was that each brood had been sired by the individual located directly above the brood mother. If the array of progeny genotypes led to the rejection of this hypothesis, we next asked whether an individual located higher in the stack (or one in another stack in that substrate group) could have sired the entire brood. Rejection of this hypothesis implies either that more than one male contributed to a single brood, or alternatively that the responsible male was absent at the time of col- lection. In addition, we estimated our ability to detect sires not present in the stacks, by considering genotypic fre- quencies in the population at large. The latter were derived from a sample of 212 individuals obtained at the same time we collected the snails used for paternity analysis. For example, in substrate group 1, contributions by a male bearing a Gf genotype other than AA, AB, or AB would have been evident, as would contributions from a male with a Mpi-2 genotype other than AA, AB, or AB. The estimated frequency of individuals thus detectable by the use of these two loci is 23%, so in this case our ability to detect the contributions of visiting males to this brood was weak. In some cases (see below), our power of discrimi- nation was better. Although this simple paternity exclusion procedure lacks the resolving power of studies employing either numerous polymorphic gene loci or hypervariable genes (DNA fingerprinting), it nevertheless proved ca- pable of demonstrating multiple paternity. RESULTS The substrate groups and genotypic constitution of indi- vidual snails at three allozyme loci are depicted in Figure Page 14 The Veliger, Vol. 35, No. 1 Table 1. Genotypes of progeny in broods from six females depicted in Figure 1. ND = no data. Substrate group Female Gpi Mpi-2 9 AB 7 AB 1 A2 7 BB 9 BB 37 AB 3 BB 3 AB 2 Ala 1 BC 41 BB 17 AB 6 BB 7 AB 3} A2 1 BC 15 BB 14 BB 4 A3 23 AB 5 BC 2 AB 13 BB 6 AB 5 A4 1 BC 10 BC 10 AB 3 AD 12 BB 6 Al 7 BD 33 BB 1. Genotypic frequencies in progeny are listed in Table 1. In substrate group 1, the distribution of progeny genotypes is consistent with the uppermost individual in stack A (A4) acting as the sole sire. The immediately adjacent snail, A3, was brooding larvae at the time of collection; had it been the sole or primary sire, the Mpz-2 A allele would not have been observed in almost half the progeny. However, since 77% of the population was estimated to possess dilocus genotypes Gpi(AA, AB, or B)/Mpi-2 (AA, AB, or BB), the possibility of a contribution by an undetected sire is substantial. In substrate group 2, the progeny of female Ala were probably sired for the most part by another brooding fe- male (A3), judging from the predominance of the Gp: AB genotype (37 of 41 progeny). The predominance of the BB genotype (33 of 36 progeny) at the Est-2 locus is consistent with this inference, although the parental genotypes are unknown. However, the presence of the Gp: BB and BC genotypes, as well as the unexpected occurrence of the Mpi-2 AB genotype in three individuals, rules out A3 as the sole sire. The Mpi-2 AB genotype suggests the con- tribution of a roving male, since it is unlikely that the large brooding female Alb acted as sire. A significant contri- bution of the only male in the stack, A5, is ruled out by the absence of the Gp a allele in the progeny. This brood is the only one examined that provides unequivocal evi- dence of multiple paternity, as it contains three different Gpi alleles of paternal origin. The 24 progeny of female A2 in substrate group 3 cannot have been sired solely by either A3 or A4. The presence of the Gpz C allele (one individual) is conistent with a Progeny genotypes Pgm Est-1 Est-2 Est-3 9 BB 3 BC ND ND 2CC ND ND 3 AB ND ND 33 BB 15 BB 4BC 9 AA ND 15 BB 1 AA 4 AC 2 AB 13 AB 4CC 5 BB ND 10 BB 11 AA 6 AB 12 BC 1 BB ND 9 BB 4AA 7 BB 7 AB 6 BC 10 BC 3 BB 3 CC ND contribution from A3, while the Mf-2 A allele (seven individuals) could only have come from A4 (or a visiting male). The limited PGM data also point to A4 as a sire. As 42% of the population was estimated to possess the dilocus genotype Gp:(BB or BC)/Mpi-2(AA, AB, or BB), an additional sire or sires may have gone undetected. In substrate group 4, the progeny of female A3 appear to have been sired exclusively by the immediately adjacent individual, A4. All 23 progeny are heterozygotes at the Gpi locus, and genotypes at other loci are in accord with this interpretation. The probability of an additional un- detected sire is negligible, as we estimate that only 6% of the population possessed genotypes [Gp: AA/Mpi-2 BB/ Pgm(AA, AC, or CC)] that would have made their con- tributions undetectable. The progeny of female A4 (Gp: BB) in group 5 may have been sired largely by either individual A6 (Gp: BC) or A7 (Gf BB), or both. However, a small contribution by another male (possibly B3) is indicated by the presence of the Gz A allele. Genotypes found in the progeny of female Al in sub- strate group 6 are compatible with the contribution of a single male, A3. We estimate that 44% of individuals in the population possessed genotypes [Gpz(AB or BB)/Mpi-2 BB/Pgm(AC, BC, BB, or CC)] that would have made their contributions undetectable. DISCUSSION Owing to the small number of polymorphic enzyme loci and the small numbers of progeny scored in this study, P. M. Gaffney & B. McGee, 1992 our ability to detect multiple paternity, and conversely to exclude individuals as potential sires, was limited. Nev- ertheless, these data indicate that (1) typically a single brood is probably sired largely but not exclusively by a single male; (2) sperm can be stored for some time, judging from the apparent male contributions by individuals col- lected as transitional forms or even as brooding females; and (3) contributions are made by individuals not present in a stack, presumably roving males. These results extend the direct observations of ORTON (1950) and HOAGLAND (1978), who noted multiple mating within a stack, including copulation between individuals separated by three or four intervening animals. Multiple mating appears to be effective, resulting in genetic con- tributions from more than one male to single broods. In addition, roving males appear to enjoy some reproductive success. Further study, in the form of detailed field obser- vations coupled with a more powerful form of paternity analysis (e.g., DNA fingerprinting) should greatly improve our understanding of the ecological dynamics of fertiliza- tion and paternity in Crepidida fornicata, a model system for studies of sex change in the Mollusca. ACKNOWLEDGMENTS This project was supported by a grant from the University of Delaware Research Foundation to PMG. We thank two anonymous reviewers for helpful comments. LITERATURE CITED Boyer, S. H., D. C. FAINER & M. A. NAUGHTON. 1963. My- oglobin: inherited structural variation in man. Science 140: 1228-1231. Page 15 CHARNOV, E. L. 1982. The Theory of Sex Allocation. Princeton University Press: Princeton. 355 pp. Cog, W. R. 1936. Sexual phases in Crepidula. Journal of Ex- perimental Zoology 72:455-477. Cog, W. R. 1938. Conditions influencing change of sex in mollusks of the genus Crepidula. Journal of Experimental Zoology 77:401-424. Cog, W.R. 1942. The reproductive organs of the prosobranch mollusk Crepidula onyx and their transformation during the change from male to female phase. Journal of Morphology 70:501-512. HOAGLAND, K. E. 1978. Protandry and the evolution of en- vironmentally-mediated sex change: a study of the Mollusca. Malacologia 17:365-391. Mutvey, M. & R. C. VRIJENHOEK. 1981. Multiple paternity in the hermaphroditic snail, Boomphalaria obstructa. Journal of Heredity 72:308-312. Murray, J. 1964. Multiple mating and effective population size in Cepaea nemoralis. Evolution 18:283-291. ORTON, J. H. 1950. Recent breeding phenomena in the Amer- ican slipper-limpet, Crepidula fornicata. Nature 165:433-434. OrTON, J. H. 1952. Protandry with self-fertilization in the American slipper limpet, Crepidula fornicata. Nature 169: 279-280. SELANDER, R. K., M. H. SMITH, S. Y. YANG, W. E. JOHNSON & J. B. Gentry. 1971. Biochemical polymorphism and systematics in the genus Peromyscus. I. Variation in the old- field mouse (Peromyscus polionotus). Studies in Genetics 6: 49-91 (Univ. Texas Publ. No. 7103). WILczynskI. J. Z. 1955. On sex behaviour and sex determi- nation in Crepidula fornicata. Biological Bulletin 109:353 (abstract). WRIGHT, W.G. 1988. Sex change in the Mollusca. Trends in Ecology and Evolution 3:137-140. The Veliger 35(1):16-25 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 The Egg Capsules and Early Life History of the Corallivorous Gastropod Drupella cornus (Roding, 1798) by STEPHANIE J. TURNER Department of Conservation and Land Management, P.O. Box 51, Wanneroo 6065, Western Australia, Australia Abstract. The corallivorous muricid gastropod Drupella cornus (Roding, 1798) has undergone a marked increase in numbers along the Ningaloo Reef Tract (northwestern Australia) over the last decade. Associated with this population explosion has been a significant increase in the extent of coral predation evident along the reef. The spawning behavior and early life history of D. cornus have been studied in the laboratory. Seven D. cornus were observed to spawn between 4 and 115 capsules over a period of 1-2 weeks. In the laboratory the distinctively shaped capsules, which averaged 2.8 x 3.2 x 1.8 mm in size, were firmly attached to the aquarium walls. Each capsule contained between 311 and 1384 (nm = 50) embryos, which hatched, after 20-37 days, into planktonic veligers. No larvae have been successfully reared through to metamorphosis and settlement in the laboratory; however, the size of the protoconch of juvenile D. cornus suggests that the veligers may spend extended periods in the plankton. INTRODUCTION A number of marine organisms, including species of fish, crustaceans, mollusks, echinoderms and polychaetes, are known to feed, either facultatively or obligately, on scler- actinian corals (see ROBERTSON [1970] for a review). Pre- viously, only the asteroid Acanthaster planci (crown-of- thorns) has been reported to represent a serious threat to coral reefs (see MORAN [1986] for a review). Over the last 5-10 years, however, the predatory activities of the cor- allivorous muricid gastropod Drupella cornus (Réding, 1798) (see Figure 1) have caused conspicuous and signif- icant damage along the Ningaloo Reef Tract (North-West Cape, Western Australia). STODDART (1989) reported that coral cover in the back-reef zone at Ningaloo has been reduced by more than 75% in two-thirds of the reef. Drupella cornus is a common and widespread species found throughout the tropical and subtropical shallow wa- ters of the Indo-West Pacific (WILSON & GILLETT, 1971). Since the early 1980s, the numbers of D. cornus have in- creased dramatically along the Ningaloo Reef, from an estimated 100-200 snails/km of reef to the present 1-2 million snails/km (STODDART, 1989). AYLING & AYLING (1987) estimated that there are approximately 500 million D. cornus in the Ningaloo Marine Park, which encom- passes an area of 224,000 hectares (May e¢ al., 1989). They recorded average D. cornus densities of 5.3-18.5/m?, and maximum densities as high as 175/m?, on the reef flats where D. cornus is most abundant; and they calculated a mean feeding rate of 2.6 cm? of Acropora plate coral/ snail/day, with a range of between 0.6 and 10.1 cm/?. It has been suggested that coral damage previously ascribed to A. planci predation may actually be a result of the activities of Drupella (MOYER et al., 1982). The nature of the damage to the coral colonies themselves, and to the reef communities as a whole, is very similar (MOYER e¢ al., 1982; AYLING & AYLING, 1987). Outbreaks of Drupella species have also been reported around islands in southern Japan (MOYER et al., 1982; FuJIOKA & YAMAZATO, 1983), the Philippines (MOYER et al., 1982), and Enewetak Atoll, Marshall Islands (BOUCHER, 1986). MOYER et al. (1982) recorded densities of D. rugosa as high as 1500/0.5 m? at sites in the Phil- ippines, and they estimated that, between 1979 and 1981, approximately 35% of one of their study reefs in southern Japan, covering an area of approximately 1200 m?, was destroyed by D. fragum predation. Little is known about the biology of Drupella cornus or the effects that large numbers of this species may have on Sa) Gurner, 1992 Figure 1 Adult Drupella cornus (Roding, 1798) collected from Coral Bay, Ningaloo Reef in February 1990. Scale bar = 1 cm. coral ecosystems. This paper presents preliminary obser- vations on the early life history of D. cornus, with a de- scription of its spawn masses and spawning behavior. In- formation on the early life cycle of D. cornus may be of particular importance in understanding the population outbreaks currently being observed on the Ningaloo Reef. FIELD SITES Drupella cornus were collected from shallow-water sites (1-1.5 m depth) at two localities: Coral Bay (14 July 1990 and 17 November 1990) at the southern end of the Nin- galoo Reef Tract and Rat Island (12 February 1990 and 16 March 1990) in the Easter Group of the Houtman Abrolhos Islands (Figure 2). The Ningaloo Reef Tract is the largest fringing reef system in Australia, and extends for 280 km along the western side of the North-West Cape peninsula, between Point Murat and Gnarraloo Bay (21°47'S, 114°00’E and 23°38'S, 113°37’E), 1200 km north of Perth, Western Australia. The Houtman Abrolhos Is- lands are approximately 800 km south of the North-West Cape, and 63 km off Geraldton on the mid-west coast of Australia (between 28°16'S, 113°35’E and 29°S, 114°E). They comprise four groups of islands and reefs (North Island, the Wallabi Group, Easter Group, and Pelsaert Group). More detailed site descriptions and oceanographic information can be found in HEARN et al. (1986), SIMPSON & MasIni (1986), AYLING & AYLING (1987), HEARN & PARKER (1988), VERON & MARSH (1988), and May et al. (1989). MATERIALS anpD METHODS Groups of 6-30 reproductively mature Drupella cornus, ranging in size from 3 to 5 cm shell length, were maintained Page 17 10) 100 200 300 400 km r n 1 i r Port Hedland ee /eExefouth uincatoo / REEF COREE 0) 0 UN nt Tropic of Capricorn WESTERN AUSTRALIA HOUTMAN |: ABROLHOS ©; \.Geraldton Perth 5 i me eee 1 Figure 2 Map showing the field locations mentioned in the text. in covered 2.5 L plastic aquaria held in the University of Western Australia Marine Biological Laboratory in Perth. Each aquarium was supplied with aeration and running, non-recirculated seawater at 20-26°C, which approxi- mates the temperatures recorded in the natural habitat at the time the snails were collected (see SIMPSON & MASINI, 1986; VERON & MarsH, 1988). ‘Each aquarium was checked every 2-3 days for the incidence of egg capsules. Egg capsules were left in stu, attached to the aquarium walls or floor in running aerated seawater until the final developmental stages were at- tained, at which point they were transferred into 100-um filtered seawater. Individual egg capsules were periodically sacrificed and preserved; egg capsule size and the number and size of the developing embryos were recorded. Mea- surements were made with a Zeiss Technival stereomi- croscope and an eyepiece measuring graticule. Page 18 The Veliger, Vol. 35, No. 1 Table 1 Summary of the characteristics of the egg capsules produced by Drupella cornus under laboratory conditions. Mean Days to Spawning date Number Number egg Veli- Mean early Days to (day /month/ of Capsule size? “Exit pore” of eggs/ diameter* gers/ veliger veliger hatch- Female? year) capsules (mm) size (mm) capsule (um) capsule _ size‘ (um) stage® ing 14 2/1V/90 44 — — 311 — — — 3.3 x 3.3 x 1.0 0.5 x 0.4 — — 478 — — — — — 392 = 11 20-29 3 22-23/VII/90 23 — — 908 170 = — 17 26/VII/90 6 — — — = 564 = — — — _ 525 270 x 210 = — — = 548 260 x 200 15 27 4 21/VII/90 if = — 808 160 — — — — — — 451 280 x 230 18 27-28 if 24-26/VI1/90 33 — — 611 180 — — 2.9 x 3.8 x 2.4 0.7 x 0.45 — — 819 — 1.8 x 3.7 x 2.3 0.8 x 0.4 — — 610 — 16 9 20-21/VII/90 23 3.5 x 2.9 x 1.8 0.7 x 0.5 773 — = — — — — — 516 — — — — — 384 — — — — — 885 — — — — — 895 — 19 24/VII/90 8 — — 483 170 — — — — — — 541 — — — — = 528 — — — — — 583 — 17 27-28/VII/90 12 3.0 x 3.2 x 1.6 0.6 x 0.5 580 170 _ —- 3.0 x 2.6 x 1.6 0.6 x 0.4 — — 436 260 x 220 2.9 x 2.8 x 1.7 0.5 x 0.4 — — 497 250 x 210 2.1 x 3.1 x 2.0 0.6 x 0.4 = — 509 — 1.7 x 3.2 x 1.6 0.7 x 0.35 — — 402 — 2.4 x 3.8 x 1.6 0.7 x 0.4 — — 491 — 3.3 x 3.2 x 1.5 0.65 x 0.5 — — 567 — 14 29/VII/90 14 25 Xs 322) xX k8 0.6 x 0.4 380 180 — — 21 30/VII/90 8 2.2 x 4.0 x 2.0 0.7 x 0.4 — — 564 — 2.9 x 3.0 x 1.6 0.6 x 0.5 — — 512 — 31/VI1/90 8 2.7 x 3.0 x 2.0 0.6 x 0.5 484 170 — — 2.1 x 3.3 x 2.4 0.6 x 0.4 — — 554 — 1/VIII/90 9 Be eXe Se OCeIeS 0.7 x 0.5 554 160 — os 3.1 x 3.1 x 1.6 0.6 x 0.5 i — 538 — 3.3 x 2.8 x 1.8 0.7 x 0.5 — — 544 = 3.4 x 3.3 1.6 0.65 x 0.5 — _ 629 — 18 2/VIII/90 12 27 2X! 308x241 0.8 x 0.5 — — 640 — — — — — 496 — 21 3/VII1/90 10 3.0 x 2.9 x 1.6 0.7 x 0.5 378 180 — — oes Pa Ds Ie 0.7 x 0.6 — — 516 — 3.3 x 3.1 x 2.0 0.8 x 0.5 = — 591 — 3.0 x 2.8 x 1.8 0.9 x 0.6 — — 473 — 16 See turner, 1992 Page 19 Table 1 Continued. Mean Days to Spawning date Number Number egg Veli- Mean early Days to (day /month/ of Capsule size? “Exit pore” of eggs/ diameter gers/ veliger veliger _hatch- Female? year) capsules (mm) size (mm) capsule (um) capsule — size‘ (um) stage® ing 4/VIII/90 11 — -- of — —- a 10 29/VII/90 4 — _ 586 160 — — — _ — — 477 — _ — — — 701 — — — — — 475 — 16 et 19-20/VII/90 22 _ —- oo — 1384 — = = — — 1009 — — = — — 1221 — 19 28-37 2 Female 14 was collected from the Abrolhos Islands on 16 March 1990; all the other snails were collected at Coral Bay on 14 July 1990. > Measurements = height x length x width. © The mean diameter of eggs that are less than 12 hours old (n = 20). 4 Length and depth measurements of recently hatched veliger larvae (n = 15). © The approximate developmental times, to the late trochophore/early veliger stage, for embryos developing in the laboratory. Mean water temperature for the capsules produced by the Abrolhos Islands snail was 25°C and for those from Coral Bay individuals was 21°C. ‘No eggs were observed in any of the 11 capsules spawned on this date. Larvae that hatched from the capsules spawned in the laboratory were either reared at 25°C and fed Dunaliella tertiolecta or Isochrysis galbana (Tahitian strain) at a con- centration of 10,000 cells/mL (April 1990) or cultured at 21°C and fed a 1:1:1 mixture of Isochrysis galbana, Chae- toceros gracilis, and Pavlova lutheri at a final concentration of 10,000 cells/mL (July-August 1990). Larvae were test- ed for metamorphic competence by exposing them to small pieces of live Acropora species or coralline-algae-encrusted dead coral. The substratum types selected for initial testing were those suggested from field observations on the dis- tribution of juvenile Drupella cornus. RESULTS Drupella cornus is gonochoristic, and in the populations at Coral Bay and the Abrolhos Islands, the ratio of the two sexes was approximately 1:1 (NARDI, 1991). Although CERNOHORSKY (1969) listed a number of sexually dimor- phic shell characteristics, such criteria are unreliable and the sex of larger living individuals may only be determined with difficulty, by the presence or absence of a penis. Copulation was never observed in the laboratory; however, it appears likely that viable sperm may be retained by the female, since capsule production occurred in the laboratory in isolated females. Spawning generally began within 1- 2 weeks of collection, suggesting that the animals that did spawn were either ready to spawn or were already spawn- ing before being brought into the laboratory. Once spawn- ing had begun it often continued for several days. Under laboratory conditions, spawning has been recorded in April for an Abrolhos Islands individual, and in July for indi- viduals collected at Coral Bay. All the subsequent obser- vations (summarized in Table 1) regarding spawning and embryonic development are based on the egg capsules of the seven individuals that were observed to spawn in the laboratory. The eggs were deposited in capsules that were attached to the sides or floor of the aquarium (Figure 3), to the plastic mesh lid, or in the overflow pipe. Three of the females (Numbers 3, 7, and 9 in Table 1) were observed to affix capsules (21, 31, and 12, respectively) inside the outlet pipes of their tanks, which are 3-cm lengths of plastic tubing with an internal diameter of 0.5 cm. The majority of the capsules were spawned at night, which is also when Drupella cornus is recorded to be most active in the field (FuJIOKA & YAMAZATO, 1983; OXLEY, 1988). The exact length of time it takes for each capsule to be deposited is not known, although preliminary observations indicate that this may take approximately one hour for each capsule. Each female produced a total of 4-115 capsules over a period of 1-16 days. There was no evidence of capsule protection by D. cornus, as the females moved away from the spawn mass once spawning had been completed. The distinctively shaped capsules (Figure 4), which av- eraged 2.8 x 3.2 x 1.8 mm in size (m = 25), were thick- walled and translucent in appearance. Unlike many mu- ricid egg capsules, those produced by Drupella cornus were Page 20 The Veliger, Vol. 35, No. 1 Sas Gurnern 1997 not vase-shaped. Each capsule was fixed directly to the substratum by a flattened base, with no evidence of a basal attachment stalk. None was affixed to previously deposited capsules, but each capsule was joined to adjacent capsules by a confluent basal membrane. In cross section the cap- sules were kidney-shaped, with distinct concave and convex sides, and in general the egg capsules in the same mass were arranged such that the convex side of one capsule was aligned with, and in close proximity to, the concave side of the adjacent capsule. Each capsule had a sealed, oval exit pore (average dimensions = 0.7 x 0.5 mm, n = 25), situated approximately one-third of the way down the concave side of the capsule, through which the veligers left the capsule at hatching. The capsules from each spawning period were generally deposited in discrete, close-packed, irregular groups. How- ever, one female (Number 9 in Table 1) was observed to undergo extensive movement during some of the spawning events; the capsules produced between 29 July 1990 and 4 August 1990 were attached all over the bottom of the aquarium, often singly or in clusters of 2 or 3 capsules. Occasionally the females were observed to add further capsules to an already established mass during subsequent spawning events. It is difficult to determine if these are atypical spawning behaviors, or whether under natural conditions the females will distribute capsules over a rel- atively wide area, rather than attaching all the capsules in one location, with several females contributing to each spawn mass. There was considerable variation between females in the numbers of capsules produced and the numbers of eggs in each capsule. Each capsule contained several hundred embryos, an average of 596 embryos (range = 311-1384) was recorded in the 50 capsules examined. The embryos were distributed around the peripheral areas within the capsules, and were all at approximately the same devel- opmental stage. The eggs were spherical, pale creamy white in color, averaged 170 wm in diameter (n = 200), and appeared to be embedded within a gel-like substance inside the capsules. The general patterns of cleavage, gastrulation, and early development of the veligers were essentially the same as described for other indirectly developing prosobranch gas- tropods (e.g., D’ASARO, 1966; KUME & DAN, 1968). After 2-3 weeks at 21°C, the early veliger stages were evident within the capsules. A small cap-shaped shell developed surrounding the posterior region of the embryo and grad- ually took on the typical spiral form. Simultaneously, two small velar lobes developed at the anterior part of the body. The visceral mass was packed with yolk, which was largely Page 21 absorbed by the time of hatching. The veligers hatched in 27-37 days at 21°C and in 20-29 days at 25°C; therefore, development time seems to vary with temperature. The newly hatched veligers had dextrally coiled shells with 1%- 1% whorls and an average size of 265 x 215 um (length x depth) (n = 75). The veligers were characterized by the presence of a well-developed bilobed ciliated velum, a foot with operculum, prominent eyespots, and a darkly pig- mented anal gland (see D’AsARO, 1966). Characteristic features of the veliger shell included the presence of the larval beak—a prolongation of the outer edge of the shell aperture that extends over the shell opening between the velar lobes—and the distinct grainy appearance of the shell surface (Figure 5). There was a concentration of red/ brown pigmentation at the growing edge of the shell in the region of the larval beak and the developing shell columella. Feeding and shell growth appeared to begin soon after hatching. At 21°C, larvae fed on a 1:1:1 mixture of Isochrysis galbana, Chaetoceros gracilis and Pavlova lutheri at a final concentration of 10,000 cells/mL grew from an average size of 250 x 200 um (n = 11) to 290 x 220 um (n = 16) within the first 7 days. A distinct demarcation was evident between the shell that grew within the egg capsule before hatching (the protoconch I) and the shell that was grown during the planktonic larval period after hatching (the protoconch II) (see LIMA & LuTz, 1990). In the Drupella cornus capsules deposited in the labo- ratory, there was no evidence of significant differences in the number of eggs in the recently spawned capsules and the number of veligers hatching from the capsules at the end of the developmental period (Mann-Whitney U-test, P > 0.05). Thus, D. cornus did not appear to produce food or nurse eggs, and cannibalism did not occur to a significant degree within the capsules spawned in the laboratory. The morphology of molluscan egg capsules is generally regarded as being species specific. Documentation of cap- sule characteristics has, therefore, enabled identification of possible Drupella cornus capsules in the field. During June- July 1990 and October-November 1990, 14 clusters of capsules (varying between 1 and 41 capsules) very similar in appearance and dimensions to those spawned in the laboratory were found attached within small crevices in the dead bases of Montipora and Acropora species sampled at Bundegi Reef in the Exmouth Gulf and at Coral Bay. It is not known whether each of the clusters found on the reef was produced by a single female or whether the fe- males tend to be gregarious, depositing their capsules in the same coral crevice, although not necessarily simulta- neously. Gregarious behavior during spawning has been observed in a number of muricids (e.g., Thais haemastoma Figure 3 Abrolhos Islands Drupella cornus female spawning in the laboratory (2 April 1990). The 44 capsules were attached, in close proximity to each other, on the side wall of a plastic aquarium. Scale bar = 1 cm. Figure 4 Drupella cornus egg capsules. a. Side view of an egg capsule containing the early veliger stages of larval development. Scale bar = 0.5 mm. b. Bottom view of an egg capsule containing the early veliger stages of larval development. Scale bar = 0.5 mm. c. Bottom view of an egg capsule containing the newly deposited eggs. Scale bar = 0.5 mm. d. Side view of an egg capsule showing the capsular plug through which the veligers leave the capsule at hatching. Scale bar = 0.5 mm. e. Side view of an egg capsule. Scale bar = 0.5 mm. Key: e, embryo; m, basal membrane; p, capsular plug; v, developing veliger. S. J. Turner, 1992 floridana by D’ASARO, 1966; Nucella lapillus by FEARE, 1971; Ocenebra poulsoni and Shaskyus festivus by FOTHERINGHAM, 1971). The majority of the veligers had hatched from the capsules, but the number in the capsules still containing veligers varied between 284 and 669 (n = 6). The capsules deposited in the laboratory contained similar numbers of embryos as the capsules spawned in the field (Mann-Whitney U-test, P > 0.05). DISCUSSION The presence of a free-swimming planktonic veliger stage in Drupella cornus is in contrast to many other species of muricid gastropods that undergo direct development, where the veliger stage is retained within the egg capsule until metamorphosis and miniature adults hatch from the cap- sules (see SPIGHT, 1975a, 1976). However, many shallow- water tropical species of muricids hatch as long-term (>1 week) veligers, in contrast to species from high-latitude habitats, which metamorphose before hatching (SPIGHT, 1977a). Although planktonic veligers may be produced under the laboratory conditions employed in this study, SPIGHT (1975a, 1977b) cites incidences of different devel- opmental types in geographically separated populations of gastropods, and within single populations either over time, among females, or even among or within broods of a single female. Observations of developing embryos in egg capsules collected in the field indicate that D. cornus produces plank- tonic larvae in the field, at least under some conditions. Furthermore, the embryos of D. cornus collected from Oki- nawa Island, Japan, and maintained in a laboratory also hatched from the capsules as planktonic veligers (AWAKUNI, 1989). My laboratory observations indicate that the individual fecundity of female Drupella cornus may be high; consid- erable numbers of capsules were deposited by a single female, and each capsule contained several hundred small eggs, all of which appeared to develop into veligers. It is unlikely, however, that fecundity estimates from laboratory observations are truly representative of the potential fe- cundity of females in the field. The lifetime fecundity of Drupella cornus will be deter- mined by the length of the life of D. cornus, the age at which the females start spawning, the frequency of spawn- ing during a female’s lifetime, and the interval between successive spawnings, about which no information is cur- rently available. From a comparison of the reproductive and life-history characteristics of a number of muricid gastropods (see SPIGHT ef al., 1974; SPIGHT, 1975b, 1979; SPIGHT & EMLEN, 1976), the average adult life-span ap- pears to vary between 1 and 12 years, with sexual maturity being attained after 1-5 years, and the number of clutches being produced per year varying between 1 and 12. Be- cause adult D. cornus are relatively large muricids, it may be predicted that they would have a long life expectancy, that they wouid mature relatively late in life, and that each female would produce one large clutch every year for a Page 23 © Figure 5 Veliger larvae of Drupella cornus. a. Recently hatched veliger (left side of photograph) and 48-hr- old veliger (ventral view of the shells). Note the prominent larval beak and the demarcation (indicated by the arrow) between the protoconch I and the protoconch II, which marks the point of hatching from the egg capsule. Scale bar = 100 um. b. An 8-day-old veliger fed on Jsochrysis galbana in unfiltered seawater. Scale bar = 100 um. Key: f, foot; 1, larval beak; s, shell; v, velum. number of successive years. SPIGHT et al. (1974) calculated that the lifetime fecundity of Shaskyus festivus (an intertidal muricid found on Californian shores and of a similar size to D. cornus) was 75,800 embryos, on the basis that each female produced 21 capsules per year, each containing an average of 531 embryos, throughout a reproductive life- span of 6.8 years (FOTHERINGHAM, 1971). This is in con- trast to smaller muricids that mature early, have a short life expectancy, and spawn small clutches several times a year, producing relatively few eggs per female lifetime (SPIGHT et al., 1974). SPIGHT et al. (1974) predicted that reproduction is de- layed until a snail reaches a size at which it is capable of turning its entire annual net energy intake into eggs. Al- though no information is available for Drupella cornus, Page 24 Figure 6 Juvenile Drupella cornus collected from the Ningaloo Reef Tract in October-November 1990. a and b. Shells of juvenile D. cornus (1-2 mm total shell length). The arrows indicate the transition in shell growth between the protoconch and teloconch, which occurs at the time of settlement and metamorphosis. Note the fingerlike projection of the larval beak or sinusigera lip on the outer edge of the protoconch shell. Scale bars = 1 mm (a) and 0.5 mm (b). c. Juvenile D. cornus, total shell length 1.07 cm. Scale bar = 0.25 cm. The Veliger, Vol. 35, No. 1 AWAKUNI (1989) estimated that adults of the related spe- cies D. fraga would reach their maximum size in approx- imately 2.5 years. However, the single D. fraga veliger that survived through to metamorphosis, 17 days after hatching, was only 275 um in shell length, considerably smaller in size than D. cornus veligers of a similar age (AWAKUNI, 1989). The adults of D. fraga are also 1-2 cm smaller than those of D. cornus. The mean egg diameter, veliger hatching size, and de- velopmental periods of Drupella cornus are similar to those recorded for other muricid species whose embryos develop without nurse eggs and hatch as veligers (see SPIGHT, 1975a, 1976). Furthermore, the values recorded for D. cornus collected at Okinawa Island by AWAKUNI (1989) are very similar—the mean egg diameter was 169 um, and embryos hatched as veligers, with a mean shell length of 274 um, after 16-23 days at 24-28°C. No larvae have so far been successfully reared through to settlement in the laboratory (all died within 2-3 weeks of hatching). However, a relatively extended planktonic life (probably of several weeks) can be inferred because hatching occurs at an early veliger stage when the shell has approximately 114 whorls and because the protoconchs of juvenile Drupella cornus collected in the field were be- tween 3 and 4 whorls in size and between 0.7 and 0.95 mm in length (n = 47) (Figure 6). Although his sample sizes were very small, AWAKUNI (1989) found that D. cornus veligers reared in the laboratory grew in shell length at arate of approximately 3.8 wm/day, and reached a shell length of approximately 301 wm in one week and 425 um after 22 days. The well-defined apertural beak is also characteristic of most long-term planktotrophic proso- branch veligers (D’ASARO, 1966). Many marine invertebrates exhibit population out- breaks at irregular intervals (COE, 1956), and although population fluctuations may arise because of varying mor- tality or survival at any of the stages in the life cycle of the organism concerned, it has been suggested that these fluctuations may be primarily attributable to processes affecting the early life-cycle stages. THORSON (1950) found that species with long planktonic larval lives (2 weeks to 3 months) are the most likely to undergo large fluctuations in numbers from year to year because of the vagaries of a planktonic existence, while species with relatively constant populations have either very short planktonic stages (hours or days) in their life cycles or undergo direct development. It is likely, therefore, that an understanding of the early life history of Drupella cornus will contribute towards an explanation for the recent increase in the numbers of the snail along the Ningaloo Reef. Further research into the early life history of D. cornus will need to concentrate on determining the length of the larval life, the identity of larval predators (both pelagic and benthic), and the selec- tivity of the larvae towards different substratum types. — Key: |, larval beak; p, protoconch; t, teloconch. Sse eeurner, 1997 ACKNOWLEDGMENTS This project is funded by the Australian National Parks and Wildlife Service States Cooperative Assistance Pro- gram (Project No. 4465), for which I am grateful. I would also like to thank the University of Western Australia and the Western Australian Fisheries Department for allowing me to use their laboratory facilities. Iam especially grateful for all the assistance received from the staff of the De- partment of Conservation and Land Management at Woodvale and at Exmouth. The advice of Mr. M. Williams with respect to the statistical analysis is also gratefully acknowledged. Special thanks to Mr. T. Cooper, Mr. P. Harding, Mr. S. Lemmens, Ms. S. Mercer, Mr. A. Rock- all, Mr. C. Scott, Ms. E. Stroud, Ms. M. Thornton, and Mr. A. Williams for their unfailing help in the field; and to Dr. J. Stoddart, Mr. K. Nardi, and Mr. B. Marinovic for their helpful advice and discussions. Drs. R. Black, T. Friend, T. Start, and J. Stoddart and two anonymous referees read and commented on the preliminary drafts of this manuscript. LITERATURE CITED AWAKUNI, T. 1989. Reproduction and growth of coral predators Drupella fraga and Drupella cornus (Gastropoda: Muricidae). Unpublished Thesis, University of the Ryukyus, Japan. 25 PP- AYLING, A. M. & A. L. AYLING. 1987. Ningaloo Marine Park: preliminary fish density assessment and habitat survey, with information on coral damage due to Drupella cornus grazing. Unpublished Report, Department of Conservation and Land Management, Western Australia. 87 pp. BoucHer, L. M. 1986. Coral predation by muricid gastropods of the genus Drupella at Enewetak, Marshall Islands. Bul- letin of Marine Science 38:9-11. CERNOHORSKY, W. O. 1969. The Muricidae of Fiji, Part I— Subfamily Thaidinae. The Veliger 11:293-315. Cor, W.R. 1956. Fluctuations in populations of littoral marine invertebrates. Journal of Marine Research 15:212-232. D’Asaro, C. N. 1966. The egg capsules, embryogenesis, and early organogenesis of a common oyster predator, 7hais hae- mastoma floridana (Gastropoda: Prosobranchia). Bulletin of Marine Science 16:884-914. FEARE, C. J. 1971. The adaptive significance of aggregation behaviour in the dogwhelk Nucella lapillus (L.). Oecologia (Berlin) 7:117-126. FOTHERINGHAM, N. 1971. Life history patterns of the littoral gastropods Shaskyus festwus (Hinds) and Ocenebra poulsoni Carpenter (Prosobranchia: Muricidae). Ecology 52:742-757. Fujioka, Y. & K. YAMAZATO. 1983. Host selection of some Okinawan coral associated gastropods belonging to the gen- era Drupella, Coralhiophila and Quoyula. Galaxea 2:59-73. HEARN, C. J., B. G. HaTcuer, R. J. Masini & C. J. SIMPSON. 1986. Oceanographic processes on the Ningaloo Coral Reef, Western Australia. Environmental Dynamics Report ED- 86-171, University of Western Australia. 82 pp. HEARN, C. J. & I. N. PARKER. 1988. Hydrodynamic processes on the Ningaloo Coral Reef, Western Australia. Proceedings of the 6th International Coral Reef Symposium, Australia 2:497-502. Page 25 KuME M. & K. DAN. 1968. Invertebrate Embryology. Chapter 11:485-525. Prosveta: Belgrade. Lima, G. M. & R. A. Lutz. 1990. The relationship of larval shell morphology to mode of development in marine proso- branch gastropods. Journal of the Marine Biological Asso- ciation of the United Kingdom 70:61 1-637. May, R., B. WILSON, S. FRITZ & G. MERCER. 1989. Ningaloo Marine Park Management Plan 1989-1999, Part 2. De- partment of Conservation and Land Management, Western Australia. 74 pp. Moran, P. J. 1986. The Acanthaster phenomenon. Oceanog- raphy and Marine Biology: an Annual Review 24:379-480. Moyer, J. T., W. K. EMERSON & M. Ross. 1982. Massive destruction of scleractinian corals by the muricid gastropod, Drupella, in Japan and the Philippines. The Nautilus 96: 69-82. Narpi, K. 1991. Gametogenesis and reproductive behaviour in Drupella cornus (Roding, 1798) at Ningaloo and Abrolhos Reefs. Unpublished Thesis, Murdoch University, Western Australia. 105 pp. OXLEY, W. G. 1988. A sampling study of a corallivorous gas- tropod Drupella, on inshore and midshelf reefs of the Great Barrier Reef. Unpublished Thesis, James Cook University, Queensland. 84 pp. ROBERTSON, R. 1970. Review of the predators and parasites of stony corals, with special reference to symbiotic proso- branch gastropods. Pacific Science 24:43-54. SIMPSON, C. J. & R. J. Masini. 1986. Tide and seawater temperature data from the Ningaloo Reef Tract, Western Australia, and the implications for coral mass spawning. Department of Conservation and Environment, Western Australia, Bulletin 253:18 pp. SPIGHT, T. M. 1975a. Factors extending gastropod embryonic development and their selective cost. Oecologia (Berlin) 21: 1-16. SpicHT, T. M. 1975b. Ona snail’s chances of becoming a year old. Oikos 26:9-14. SpicHT, T.M. 1976. Ecology of hatching size for marine snails. Oecologia (Berlin) 24:283-294. SpiGHT, T. M. 1977a. Latitude, habitat, and hatching type for muricacean gastropods. The Nautilus 91:67-71. SPIGHT, T. M. 1977b. Is Thais canaliculata (Gastropoda: Mu- ricidae) evolving nurse eggs? The Nautilus 91:74-76. SPIGHT, T. M. 1979. Environment and life history: the case of two marine snails. Pp. 135-143. In: S. E. Stancyk (ed.), Reproductive Ecology of Marine Invertebrates, Belle W. Baruch Library in Marine Science No. 9, University of South Carolina Press. SpPIGHT, T. M., C. BIRKELAND & A. Lyons. 1974. Life histories of large and small murexes (Prosobranchia: Muricidae). Marine Biology 24:229-242. SPIGHT, T. M. & J. EMLEN. 1976. Clutch sizes of two marine snails with a changing food supply. Ecology 57:1162-1178. STODDART, J. A. 1989. Fatal attraction. Landscope, W.A.’s Conservation, Forests and Wildlife Magazine, Winter Edi- tion: 14-20. TuHorson, G. 1950. Reproductive and larval ecology of marine bottom invertebrates. Biological Review 25:1-45. VERON, J. E. N. & L. M. Marsu. 1988. Hermatypic corals of Western Australia: records and annotated species list. Records of the Western Australian Museum, Supplement No. 29:136 pp. WILSON, B. R. & K. GILLETT. 1971. A Field Guide to Aus- tralian Shells: Prosobranch Gastropods. Reed Books Pty. Ltd.: Frenchs Forest, New South Wales. 287 pp. The Veliger 35(1):26-35 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 Mantle-Mediated Shell Decollation Increases Posterior Aperture Size in Dentalium rectius (Scaphopoda: Dentaliida) by PATRICK D. REYNOLDS! Biology Department, University of Victoria, British Columbia, Canada Abstract. In Scaphopoda, incurrent and excurrent water flow through the mantle cavity occurs primarily via the posterior aperture, located at the apex of a tusk-shaped shell. However, maintenance of a sufficiently large posterior opening to the exterior is not accommodated by normal anteriorly directed shell growth, which progressively diminishes the posterior aperture relative to body size. A supplementary increase in posterior aperture size correlated with growth of the organism is therefore necessary to facilitate the passage of respiratory currents, gametes, and waste materials from the mantle cavity. Shell morphometrics, ultrastructure, and direct observations of live specimens of Dentalium rectius confirm that such an increase does take place, and is due primarily to decollation of the shell through dissolution by the posterior mantle. While an increase in posterior aperture size by shell dissolution and truncation had been previously predicted, shell decollation has not been previously recorded from the class. The removal of an apical portion of shell, as opposed to dissolution of the aperture rim, is necessitated in D. rectius and some other Dentaliida by the secretion of a secondary shell by the posterior mantle margin. INTRODUCTION The shell of Dentaliwm rectius Carpenter, 1865, is typical of the Dentaliida, being a slightly curved, translucent cone reaching 7 cm in length. As with all members of the class Scaphopoda it is open at both ends, with extension of the burrowing foot and the feeding captacula through the an- terior aperture (Figure 1) and passage of both inhalant and exhalant respiratory currents through the posterior opening (YONGE, 1937). While the major portion of the scaphopod shell is secreted by the anterior mantle margin (and referred to here as the primary shell), a thin, sec- ondary tube of shell that is produced by the posterior mantle margin often extends from the apex of the primary shell (Figure 1). Such a secondary shell is found only among several dentaliid scaphopod genera (STASEK & McWILLIAMS, 1973; PALMER, 1974). Shell secretion cor- related with soft-tissue growth proceeds anteriorly, by a widening of the anterior aperture accompanied by a slight ' Present address: Institute of Marine Sciences, University of California, Santa Cruz, California 95064, USA. spiral along the shell length, producing the characteristic tusklike shape. The consequences of this shell growth pattern to certain physiological requirements have been noted by La- CAZE-DUTHIERS (1856), PILSBRY & SHARP (1897), and FISCHER-PIETTE & FRANC (1969). With increasing size of the animal, a constant posterior aperture size would become progressively smaller in relation to body size, and eventually be inadequate for the passage of respiratory currents, the release of gametes, and the elimination of feces. While the apical larval shell is lost at an early stage (LACAZE-DUTHIERS, 1856), a continual increase in pos- terior aperture size with growth of the animal has been predicted to occur to ensure the maintenance of ade- quate exchange with the external environment (La- CAZE-DUTHIERS, 1856; PILSBRY & SHARP, 1897; FI- SCHER-PIETTE & FRANC, 1969). FISCHER-PIETTE & FRANC (1969) and STASEK & McWILLIAMS (1973) suggest that, in addition to its role in secondary shell secretion, the posterior mantle margin of scaphopods may increase posterior aperture size by dis- solution or reabsorptive truncation of the shell apex. Lab- oratory observation of shell decollation (the loss of the P. D. Reynolds, 1992 Page 27 Figure 1 Schematic diagram showing measurements taken on all shells. Dorsal is to the top, anterior to the left. Key: a, total length; b, primary shell length; c, anterior aperture height; d, primary shell apex height; e, secondary shell base height; f, secondary shell apex height; g, secondary shell length. Measurements of height represent diameter along the dorsoventral axis. posterior or apical portion of the shell) in Dentaliwm rectius prompted an investigation of the posterior shell growth pattern and an examination of discarded and intact shell apices for evidence of the shell loss mechanism. MATERIALS ano METHODS Live Dentalium rectius for observation were dredged from Imperial Eagle Channel, Barkley Sound (48°52.7'N, 125°11.1’W) near Bamfield, British Columbia, in June 1985, while shell morphometrics were performed on spec- imens collected from Satellite Channel (48°42.6'N, 123°31.9’W) near Victoria, British Columbia, from Oc- tober 1987 to February 1988. Voucher specimens have been deposited at the California Academy of Sciences (San Francisco), catalogue No. 075615. Scaphopods were maintained at Bamfield Marine Sta- tion, British Columbia, over a four-day period in sediment (10-13 cm deep) from the biotope sieved to <1 mm. The sediment and scaphopods (70 specimens in total) were held in plastic beakers of two sizes: 800 mL (10-cm diameter, 2 individuals per beaker, 15 sets) and 1.5L (15-cm di- ameter, 10 and 20 individuals per beaker, 2 and 1 sets respectively). To maintain aeration of sediment, beakers with doubled 1-mm-mesh sides were used and placed in running seawater (10 + 0.5°C). The apical shell of >80% of the specimens (at least 57/70) was seen level with or above the sediment surface at least once, when examined at 1-12 hr intervals. Shell decollation was observed in one of the scaphopods; the discarded shell apex was later frac- tured and examined by scanning electron microscopy (SEM). Shell measurements (Figure 1) were made at 10x mag- nification using a Zeiss dissecting microscope and ocular micrometer. The primary shells of 23% of the specimens were broken during the sampling process and were ex- cluded from subsequent analyses of shell characters; shells with repaired breakages were not excluded. Character re- lationships were described by linear regression using least mean squares (ZAR, 1984). Of the measurements taken, anterior aperture height best represented growth (soft- tissue dry weight) as described by the allometric relation- ship: log(y) = 0.238 log(x) + 0.197, r? = 0.94, P < 0.0001, n = 43 (Figure 2). The absence or presence and the position of a shallow notch at the apex of the primary shell (Figure 3) were noted in a sample of 131 scaphopods (Table 1). These specimens spanned the entire sampled size range (as shown in Figure 2); 93.6% could be sexed by the external appearence of gonads through the semi-trans- parent shell, whereas the remainder were among the small- est specimens and appeared to be juveniles. The rates of decrease in the size of the posterior aperture per unit length, or taper, of the primary and secondary shells were calculated as follows: Table 1 Summary of primary shell morphological characteristics in Dentalium rectus. With Without secondary secondary shell shell Totals All specimens Notched 102 11 113 (86.3%) Not notched 1 4 (3.0%) Repaired breakage 14 — 14 (10.7%) Totals 117 (89.3%) 14(10.7%) 131 (100.0%) Notched shells Ventral notch 98 9 107 (94.7%) Dorsal notch 3 1 4 (3.5%) Both 1 1 2 (1.8%) Totals 102 (90.3%) 11 (9.7%) 113 (100.0%) Page 28 Log (anterior aperture height) (mm) The Veliger, Vol. 35, No. 1 Log (soft tissue dry weight) (mg) Figure 2 Plot of log (anterior aperture height) against log (soft tissue dry weight). log(y) = 0.238 log(x) + 0.197, 7? = 0.94, P < 0.0001, n = 43. Primary shell taper primary shell base (anterior aperture) — primary shell apex primary shell length Secondary shell taper secondary shell base — secondary shell apex (posterior aperture) secondary shell length The means of primary and secondary shell taper were compared using the Student f-test (ZAR, 1984). Shells examined by SEM were rinsed with double-dis- tilled water, dried at room temperature, mounted on al- uminium stubs, and, in some cases, fractured using a fine stainless-steel probe. Specimens were gold coated prior to viewing ina JEOL JSM-35 scanning electron microscope. RESULTS Apical Shell Morphology Observations on apical shell morphology are presented in Table 1. The primary shell apex of most shells (86.3%) bears a notch (Figures 3, 4), which is usually found on the ventral (convex) side (94.7%). Otherwise, specimens have a jagged primary shell apex indicating a repaired break (10.7%) (Figures 5, 6) or a smooth apex with no notch (3%). The primary shell is secreted by the anterior mantle margin, which is reflected by the presence of cir- cular growth lines (Figures 5, 6), although these are often not apparent at the apex of the primary shell owing to erosion (Figures 3, 4). A secondary shell was present in 89.3% of the specimens (Table 1). The secondary shell is secreted by the posterior mantle margin, which produces V-shaped growth incre- ments and a ventral suture line (Figures 3, 4). The apex of the secondary shell was usually broken, although a notch was noted at the apex on intact shells. Secondary shell growth was discontinuous in 13 cases (7.6% of all speci- mens with a secondary shell); nine of these had a notched primary shell with a subsequent break in the secondary shell having been repaired (one with two breaks), three had a notched primary shell with a notched disruption in the secondary shell, and one had a break repaired in the primary shell apex with a notched disruption in the sec- ondary shell. Observation of Shell Decollation and Evidence of Shell Dissolution During observations of live Dentalium rectius, shell de- collation was observed in a specimen approximately 3 cm in length (about 26.7 mm primary shell, 3.3 mm secondary shell), with the posterior shell protruding 7-9 mm from the sediment. The shell apex had detached and fallen over, and was found resting at an angle between the sediment surface and the new aperture rim. The scaphopod had not changed position appreciably over the previous hour. When the container was disturbed several minutes later, the P. D. Reynolds, 1992 Page 29 Explanation of Figures 3 to 6 Figures 3-6. Junction of primary (lower) and secondary (upper) shells. Figure 3. Ventral view of notched primary-secondary shell junction. Note the V-shaped notch at the primary shell apex, the V-shaped growth lines and median suture line (arrowhead) of the secondary shell, and the eroded primary shell. Scale = 0.3 mm. Figure 4. Ventral view of notched primary-secondary shell junction. Note the circular growth lines of the primary shell (arrowheads and lower left corner), V-shaped notch at the primary shell apex, and the V-shaped growth lines, median suture line, and simple prismatic structure (arrow) of the secondary shell. Scale = 0.2 mm. Figure 5. Lateral view of primary-secondary shell junction that lacks a notch (ventral to the right). Note the circular growth lines and jagged apex of the primary shell. Scale = 0.2 mm. Figure 6. Dorsal view of primary-secondary shell junction that lacks a notch. Note the jagged apex of the primary shell. Scale = 0.2 mm. scaphopod burrowed deeper into the sediment, leaving the detached portion of shell on the sediment surface. The animal appeared healthy (7.e., vigorous burrowing move- ments by the foot) when examined two days later. The discarded shell apex measured 5.15 mm in length, of which approximately 1.84 mm was primary shell pos- sessing a jagged apex (Figures 6, 7). At the point of sep- aration the height of the detached shell was 0.67 mm, tapering to 0.38 mm posteriorly (Figure 7). The posterior aperture of the scaphopod therefore increased by 0.29 mm, The Veliger, Vol. 35, No. 1 Explanation of Figures 7 to 9 Figure 7. Dorsal view of discarded shell apex, with lower dorsal portion removed (fracture line indicated by large arrows). Note the V-shaped line of decollation from the primary shell (small arrows). Scale = 0.5 mm. Figure 8. Etched shell layers where decollation from the primary shell occurred. Note the deeply eroded appearance of the two distinct shell layers. Scale = 30 ym. Figure 9. Fractured edge of same shell. In the same orientation as in Figure 8 (cross section). Note the clearly defined shell microstructure. Scale = 30 ym. or 1.8 the original diameter. The edge of the discarded shell, where separation from the primary shell occurred, clearly showed evidence of dissolution by the loss of min- eralized material from the shell layers (Figure 8) when compared to a fractured edge of the same discarded shell (Figure 9). The anterior fractured edge of the discarded shell shows the two layers of aragonitic calcium carbonate of the primary shell, an outer crossed lamellar layer (Fig- ure 10) and an inner simple prismatic layer (Figure 11) that is continuous with the single, prismatic layer of the secondary shell (Figure 4). Comparison with the corre- sponding shell layers at the edge of detachment from the scaphopod (Figures 12, 13) shows a strong etching or erosion; there is a loss of the crossed lamellar crystal ar- rangement accompanied by the production of a rough, irregular surface with deep interstices (Figure 12), where- as the uneven but solid face of the fractured prismatic layer in Figure 11 contrasts with the deeply eroded prismatic layer where detachment occurred, producing layered cav- ities and rounded remains of the crystal structure (Figure 13): The internal surface of the discarded shell apex revealed P. D. Reynolds, 1992 Explanation of Figures 10 to 13 Figure 10. Outer crossed lamellar layer of the decollated shell, fractured region. Scale Figure 11. Inner simple prismatic layer of the decollated shell, fractured region. Scale 5 wm. 5 wm. Figure 12. Outer crossed lamellar layer of the decollated shell at site of decollation, showing evidence of dissolution. Same orientation as in Figure 10. Scale = 5 wm. Figure 13. Inner simple prismatic layer of the decollated shell at site of decollation, showing evidence of dissolution. Same orientation as in Figure 11. Scale = 5 um. partial shell dissolution at a number of locations posterior to the site of detachment (Figures 14, 15), and stereoscopic SEM observations confirmed these markings to be de- pressions into the shell. Close examination revealed a rough surface which differed substantially from the normally smooth, slightly undulating topography of the internal shell (Figure 15). The V-shaped, ventral orientation of these scars corresponded in form and position to the edge where separation of the shell eventually occurred (Figure 14), and with the notches normally found at the primary shell apex. SEM examination of the internal shell surface of several specimens revealed no other markings posterior to the point of retractor muscle insertion. Analysis of Shell Measurements Analysis of shell characters shows that decollation (in- dicated by a notch on the primary shell apex) produces Explanation of Figures 14 and 15 Figure 14. Discarded shell, view of internal ventral surface. Note three sets of V-shaped scars (arrowheads). Double arrowhead, detachment edge. Scale = 0.1 mm. Figure 15. Scar on internal ventral surface of detached shell, produced by shell dissolution. Scale = 20 um. increased posterior aperture sizes (represented at the time of decollation by notched primary shell apex height) with growth (Figure 16). While subsequent secondary shell growth originates from the internal surface of the primary shell apex (Figures 3-6), it does not taper as much as the primary shell (Table 2) and so the increased aperture size is conserved (Figure 17). The Veliger, Vol. 35, No. 1 Primary apex height (mm) 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Anterior aperture height (mm) Secondary apex height (mm) 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Anterior aperture height (mm) ) Explanation of Figures 16 and 17 Figure 16. Plot of anterior aperture height against primary apex height for notched shells (shells which have been decollated), with or without a secondary shell. y = 0.22x + 0.05, 7? = 0.63, n= 110. Figure 17. Plot of anterior aperture height against secondary apex height, notched shells. y = 0.16x + 0.11, r? = 0.51, n = 100. DISCUSSION In order to maintain sufficient exchange between the ex- ternal environment and mantle cavity throughout scapho- pod ontogeny, a secondary modification of shell shape that increases posterior aperture size with growth of the animal must take place. Such an increase occurs in Dentalium rectus, thereby accommodating greater respiratory cur- rents circulating through a larger mantle cavity and fa- cilitating the passage of increasing fecal material and ga- metes. The removal of shell material to achieve this could occur in a number of ways, such as by abrasion by sedi- ments while burrowing or the gradual dissolution of the aperture rim by the mantle. However, the secretion of a secondary shell by the posterior mantle in D. rectius pre- cludes continual dissolution or erosion of the primary shell rim as a mechanism for increasing posterior aperture di- P. D. Reynolds, 1992 Table 2 Comparison of primary and secondary shell taper in Den- talium rectius, for shells that have been decollated (have a notch on the primary shell apex; n = 102). The negative minimum value indicates a secondary shell that increased in width with length. Taper rate (mm/mm) Primary shell Secondary shell Mean + SE* 0.053 + 0.001 0.01 + 0.006 Minimum 0.037 —0.056 Maximum 0.086 0.500 * t-test: 0.001 < P (|t| = 7.289) < 0.0001. ameter. Instead, the entire secondary shell and a posterior portion of the primary shell must be removed. While hap- hazard and potentially fatal to the scaphopod, extrinsic breakage of the primary shell apex, with subsequent repair by secondary shell growth, may increase posterior aperture size in a small percentage of the sampled population at any one time (e.g., 10.69%, Table 1). Shell decollation, by dissolution of the shell anterior to the primary shell apex resulting in the detachment of an apical portion of the shell, is a more effective mechanism for increasing aperture size with growth of the animal. As observed under labo- ratory conditions, D. rectius can decollate the posterior portion of its shell in this way. Repeated periodically, this would account for the observed increase in posterior ap- erture size in this species, and the presence of a ventral notch in the primary shell apex of almost 90% of the study population across the entire sampled size range (Figure 18). The V-shaped internal scars and detachment edge of the discarded shell, producing a notch in the new primary shell apex similar to that found in most specimens, resem- ble surfaces subjected to shell dissolution (DEITH, 1985; SIGNOR, 1985). The consistent orientation, shape, and size of the scars suggest that removal of shell material resulted from secretions of a discrete band of the underlying mantle, in contrast with the large area of outer pavilion epithelium thought to be involved in secondary shell secretion (STASEK & McWILLIAMS, 1973). It is not possible, however, to specify the precise mechanism of shell removal beyond invoking some type of dissolution or chemical erosion of calcium carbonate; shell reabsorption would require active uptake by the mantle of shell materials, as noted by SIGNOR (1985). FISCHER-PIETTE & FRANC (1969) point out that the process by which scaphopods enlarge the shell ante- riorly while progressively shortening it at the posterior end results in the gradual loss of the complete juvenile shell, and the progressive loss of the older portions of the adult shell. This necessitates an anterior shift in the point of retractor muscle insertion as growth proceeds. While no SEM evidence for this was found in the shells examined, it is likely that any scars resulting from such a shift would be obscured by shell secretion of the underlying mantle. Page 33 | | Schematic diagram of shell growth and model for posterior ap- erture enlargement through decollation of the primary shell. The shell is viewed from the ventral side, showing notches in primary and secondary shells. Most recent (arrowhead) and possible fu- ture (double arrowheads) sites of shell decollation are indicated (large arrows, direction of primary shell growth; small arrows, direction of secondary shell growth). Figure 18 These results indicate that the distinctive primary and secondary shell morphology found in Dentaliwm rectius is not simply due to shell repair following breakages or pre- dation by the ratfish Hydrolagus collie: as previously sug- gested (SHIMEK, 1989), although this may account for some proportion of the approximately 10% of the sampled pop- ulation with repaired primary shell apex breakages. In- stead, secondary shell secretion appears to be a normal program of shell growth following mantle-mediated shell decollation. The function of the secondary shell in D. rectius and some other dentaliids (e.g., Episiphon) is open to spec- ulation; perhaps it affords protection to the sensory epi- thelia of the posterior mantle edge or pavilion (REYNOLDS, in press) while maximizing posterior aperture increase. It Page 34 is also important to note that while the original description of D. rectius (Carpenter, 1864, in PALMER, 1958; CAR- PENTER, 1865) makes no mention of the sculpture of the posterior aperture, the description of the subgenus Rhabdus by PrtsBRY & SHARP (1897), which identifies D. rectzus Carpenter, 1865, as the type species, states that a notch, slit, and secondary shell are absent. EMERSON (1962), how- ever, found that the posterior aperture in Rhabdus occa- sionally has the inner shell layer extended as a thin tube. Rhabdus has since been elevated to full generic status by PALMER (1974). Apical shell morphology has been an im- portant shell character in scaphopod taxonomy (e.g., EMERSON, 1962; PALMER, 1974) although the reliability of this and other shell characters has been questioned (SHI- MEK, 1989). In all respects, the study populations fit the original description of D. rectius and those of recent ac- counts of northeastern Pacific scaphopods, which also refer this species to the genus Dentalium (KOZLOFF, 1987; SHI- MEK, 1989). Therefore, it would seem prudent in this report to continue to use the genus-taxon Dentalium pend- ing taxonomic review or redescription of Rhabdus. Decollation in Dentalium rectius results in loss of the secondary shell and usually the oldest portion of the pri- mary shell, the production of a notch in the shell apex prior to secondary shell regrowth, and an increase in pos- terior aperture size. Decollation is likely repeated peri- odically during growth and is an important physiological mechanism for maintaining a sufficiently large opening to the mantle cavity, which would otherwise be progressively restricted by the ontogenetic shell growth pattern of scaph- opods. Decollation of the shell is necessary to achieve this change in shell shape, as direct dissolution of the aperture rim would be ineffective given the continuity of secondary shell secretion in D. rectius. Shell decollation has also been noted in several terrestrial gastropod families such as the Truncatellidae (MorRRISON, 1963), Potamidae, Pleurocer- idae (VERMEIJ, 1974), and Subulinidae (HOCHPOCHLER & KOTHBAUER, 1975; KAT, 1981). Although also occurring in the marine Caecidae, the phenomenon is usually as- sociated with tall-spired terrestrial species, found on hard substrates, in which the shell is wholly supported by the foot (VERMEIJ, 1974; KAT, 1981). The loss of one or more whorls is accomplished by dissolution of the internal shell surface, although eventual separation of the apex may be achieved through erosion of the weakened shell. The new apex is closed off or plugged by subsequent shell secretion (MorrISON, 1963; KAT, 1981). Removal of the shell apex is thought to increase the stability of the terrestrial snail by shifting the center of gravity to lie over the aperture; in contrast, gastropods that live in soft sediments have the shell supported in part by the substrate (VERMEIJ, 1974). In the terrestrial pulmonate gastropod Rumina decollata, the change in shell shape through loss of apical whorls has been related to increased mobility, reduction in shell weight, and reduced water loss, with a resultant increase in body and gonad size contributing to increased fitness (KAT, 1981). Although the immediate consequences of shell The Veliger, Vol. 35, No. 1 decollation in D. rectius differ from those in terrestrial gastropods, it isa mechanism by which an otherwise severe restriction on growth is removed with a similar attendant potential for increasing fitness. In scaphopods of the genus Cadulus and other members of the Gadilida that do not secrete a secondary shell, posterior aperture increase could occur simply by shell dissolution at the aperture rim. In some species, there are several notches in the posterior aperture producing a lobate apex (ABBOTT, 1974), al- though the number of notches can be highly variable (SHI- MEK, 1989). Removal of apical shell material in scapho- pods can be considered one of several growth-related shell shape changes that molluscs achieve through shell disso- lution or reabsorption, such as the enlargement of apical apertures in key-hole limpets (FRETTER & GRAHAM, 1962), removal of apertural spines in muricid gastropods (CaR- RIKER, 1972), internal remodelling of uppermost whorls in a variety of prosobranch gastropods (VERMEIJ, 1974), and surficial dissolution of the penultimate whorl in pro- sobranchs and Nautilus (SIGNOR, 1982, 1985). ACKNOWLEDGMENTS I thank A. R. Fontaine and D. McHugh for critically reading earlier drafts of this manuscript, and R. L. Shimek and an anonymous reviewer for helpful suggestions. D. A. Bright and D. Brand helped collect specimens, for which I am very grateful. This work was funded in part by a Western Canadian Universities Marine Biological Society Scholarship, which allowed the author to attend a Com- parative Ethology course at Bamfield Marine Station in 1985 during which some of the observations reported here were made, and a University of Victoria Graduate Studies Fellowship. LITERATURE CITED ABBOTT, R. T. 1974. American Seashells. Van Nostrand Rein- hold: New York. CARPENTER, P. P. 1865. Diagnoses Specierum et Varietum novarum Molluscorum, pore Sinum Pugetianum a Kenner- lio Doctore, nuper decesso, collectorum. Proceedings of the Academy of Natural Sciences of Philadelphia 1865:54-64. CARRIKER, M. R. 1972. Observations on the removal of spines by muricid gastropods during shell growth. The Veliger 15(2):69-74. DeITH, M. R. 1985. The composition of intertidally deposited growth lines in the shell of the edible cockle, Cerastoderma edule. Journal of the Marine Biological Association of the United Kingdom 65:573-581. EMERSON, W. K. 1962. A classification of the scaphopod mol- lusks. Journal of Paleontology 36(3):461-482. FISCHER-PIETTE, E. & A. FRANC. 1969. Classe des Scapho- podes. Pp. 987-1017. Jn: P.-P. Grassé (ed.), Mollusques, Gasteropodes et Scaphopodes. Masson et Cie: Paris. FRETTER, V. & A. GRAHAM. 1962. British Prosobranch Mol- luscs. Ray Society: London. HOcHPOCHLER, F’. & H. KOTHBAUER. 1975. Der mechanismus der Dekollation bei Rumina decollata (L.) (Gastropoda: Sty- lomatophora). Archiv Molluskenk 106 (1-3):119-121. Kat, P. W. 1981. Shell shape changes in the Gastropoda: shell P. D. Reynolds, 1992 decollation in Rumina decollata (Pulmonata: Subulinidae). The Veliger 24(2):115-119. Koz.orr, E. N. 1987. Marine Invertebrates of the Pacific Northwest. University of Washington Press: Seattle. LacAZE-DUTHIERS, H. 1856. Histoire de l’organisation et du développement du Dentale. Annales des Sciences Naturelles, Quatriéme Series, Paris 6:319-385, pl. 11-13. Morrison, J. P. E. 1963. Cecina from the state of Washington. Nautilus 76:150-151. PALMER, C. P., 1974. Asupraspecific classification of the scaph- opod Mollusca. The Veliger 17(2):115-123. 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. Piussry, H. A. & B. SHARP. 1897. Class Scaphopoda. Manual of Conchology (Series 1) 17:i-xxxii, 1-144. REYNOLDs, P. D. In press. Distribution and ultrastructure of ciliated sensory receptors in the posterior mantle epithelium of Dentalium rectius (Mollusca, Scaphopoda). Acta Zoolo- gica. Page 35 SHIMEK, R. L. 1989. Shell morphometrics and systematics: a revision of the slender, shallow water Cadulus of the north- eastern Pacific (Scaphopoda: Gadilida). The Veliger 32(3): 233-246. SIGNoR, P. W. 1982. Growth-related surficial resorption of the penultimate whorl in Terebra dimidiata (Linnaeus, 1758) and other marine prosobranch gastropods. The Veliger 25(1): 79-82. SIGNoR, P. W. 1985. Surficial shell resorption in Nautilus mac- rophalus Sowerby, 1849. The Veliger 28(2):195-199. STASEK, C. R. & W. R. MCWILLIAMS. 1973. The comparative morphology and evolution of the molluscan mantle edge. The Veliger 16(1):1-19. VERMEYJ, G. J. 1974. Molluscs in mangrove swamps: physi- ognomy, diversity, and regional differences. Systematic Zo- ology 22:609-624. YONGE, C. M. 1937. Circulation of water in the mantle cavity of Dentalium entalis. Proceedings of the Malacological Society of London 22:333-337. ZAR, J. H. 1984. Biostatistical analysis. Prentice-Hall: Engle- wood Cliffs, New Jersey. The Veliger 35(1):36-51 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 Repaired Shell Damage Among Soft-Bottom Mollusks on the Continental Shelf and Upper Slope North of Point Conception, California by ROY K. KROPP Battelle Ocean Sciences, 397 Washington Street, Duxbury, Massachusetts 02332, USA Abstract. The frequencies of repaired shell damage were estimated for univalved mollusks on the inner and outer continental shelf and upper slope in the Santa Maria Basin, north of Point Conception, California. Twenty-one percent of all gastropods and 18.7% of all examined scaphopods had repaired shell damage. The frequency of repaired shells among prosobranchs was significantly lower at upper slope stations than at either inner or outer shelf stations. Conversely, the frequency of repaired shells among scaphopods was significantly greater at upper slope stations than at inner shelf stations. Opis- thobranchs showed no significant depth-related differences in repaired shell damage. The frequencies of shell repairs varied considerably among taxa. For example, the frequency of repaired damage was high for a scaphopod, Cadulus tolme: (47.2%), and for a prosobranch, Alvinia rosana (24.3%), but was low for eulimid gastropods (7.6%) and three species of Epitonium (no repaired shells). INTRODUCTION Within populations of mollusks it may be possible to infer the influence of shell-breaking predators by determining the frequency of repaired shell damage. Although nonle- thal damage to shells may result from non-biological causes (SCHOENER, 1979), unsuccessful predation is thought to provide the most likely explanation for repaired shell dam- age (VERMEIJ, 1987). Repairs are easily recognizable as jagged discontinuities in the normal growth pattern of the shell (VERMEIJ, 1982; VALE & REX, 1988). The frequency of repaired shells in a population is known to be positively correlated with the incidence of unsuccessful attacks by predators (VERMEIJ, 1983). The importance of unsuc- cessful attacks by shell-breaking predators on the evolution of mollusks and the assumptions that must be considered in studies of such predation have been reviewed by VER- MEIJ (1983, 1987). Much of the information available regarding frequency of repaired shell damage among mollusks has resulted from the work of Vermeij on tropical snails (VERMEIJ, 1982; VERME]J e/ al., 1980). From these studies the intensity of shell-breaking predation was inferred to be high and dem- onstrated to have profound effects on the evolution of mol- luscan shell architecture. What is known about the phenomenon among mollusks from the temperate northeast Pacific comes from studies of rocky intertidal areas (FOTHERINGHAM, 1971; GELLER, 1983) or shallow subtidal environments (EDWARDS, 1969; BERGMAN et al., 1983). These studies have shown that repair frequencies are generally lower in the temperate northeast Pacific than in tropical waters. Little is known about the importance of shell-breaking predators in deep-water molluscan communities. The pri- mary source of information comes from studies of shell repairs among deeper water molluscan communities in the northwest Atlantic by VALE & REx (1988, 1989). They found relatively low frequencies of repaired shell damage and suggested that predators on mollusks from deeper water communities were unspecialized. A program designed to monitor the effects of offshore oil and gas development in the Santa Maria Basin, just north of Point Conception, California (HYLAND et al., 1990b) provided the opportunity to investigate the extent of repaired shell damage among several groups of mollusks from the continental shelf and upper slope in the northeast Pacific. Large sample sizes permitted statistical analyses of the relationship of depth, small-scale latitudinal differ- ences, and taxon to the frequency of repaired shell damage. The results of this investigation are presented here. Ree Kropp 1992 MATERIALS anpD METHODS The molluscan material used in this study was obtained during part of a program designed to assess potential long- term impacts of offshore oil and gas development in the southern Santa Maria Basin, California (HYLAND et al., 1990b). Sediment samples were collected with a Hessler- Sandia 0.25-m? box corer partitioned into twenty-five 0.01- m? subcores. From each box-core sample the top 10 cm from 10 subcores was removed and processed for the anal- ysis of the macroinfaunal community. The mollusks from these infaunal samples were used for this study. Details of the infaunal community analyses are reported in HYLAND et al. (1990a, in press). A voucher collection of all macroin- faunal taxa identified during the program, including the molluscan taxa investigated for repaired shell damage, will be deposited in the National Museum of Natural History, Smithsonian Institution, Washington, DC. Additionally, the bulk collection of the macroinfaunal specimens col- lected will be deposited in the Natural History Museum of Los Angeles County, Los Angeles, California. The data were obtained from two sets of collections. The primary set, upon which the depth, transect, and taxonomic comparisons were based, was obtained from the sampling of 10 regional stations (designated R-1 through R-9 and PJ-1) situated along three cross-shelf transects (Figure 1). Three replicate samples were collected from each of the 10 regional stations on eight cruises between October 1986 and May 1989. Stations R-8 and R-9 were not sampled during the October 1986 cruise and Station R-7 was not sampled during the May 1988 cruise. Only two replicates were collected at Station R-3 during the May 1988 cruise. Of the 230 core samples collected, the mollusks from 212 (92%) were available to be examined for repaired shell damage. The second set, used only to supplement the taxonomic comparisons, was obtained from 18 site-specific stations organized in a semi-radial array centered around Station PJ-1 (designated PJ-2 through PJ-23; see HYLAND et al., 1990b). Of the 114 core samples collected from these sta- tions, 99 (87%) were available for examination of the mol- lusks. Because the molluscan fauna of the region has not been characterized, I used the data in appendix F-1 of HYLAND et al. (1990a) to describe briefly the univalved component of the community. Only taxa identified to species were used for these calculations. The top 10 prosobranch species, the top 20 opisthobranch species, and all six scaphopod species were included. A dissection microscope was used to examine each mol- lusk for the presence of repair scars. Molluscan groups examined included prosobranch gastropods, pyramidelli- dan and cephalasipdean opisthobranch gastropods, and scaphopods. All specimens within a sample were examined and scored for shell repair scars except those specimens too damaged by the collection process to evaluate. In the case of one abundant scaphopod species, Cadulus calvfor- Page 37 CALIFORNIA Be Point San Luis 35°00' * j Santa Maria Point Sal ' Purisma Point Point Arguello 34°30! Conception Bathymetry in Meters 121°00' 120°30' Figure 1 Map of the Santa Maria Basin, showing primary sampling lo- cations. nicus Pilsbry & Sharp, 1898, so few specimens were usable that the species was not included in any observations. Re- paired damage in all groups was recognizable as irregular, usually jagged discontinuities in the normal growth pattern of the shell (e.g., VERMEIJ, 1982; VALE & REx, 1988). Following VALE & REx (1988), repairs were recorded as major or minor, although all analyses were done using only the total number of repairs. The location and number of repairs on each shell were noted. For species of Cadulus (scaphopods) with repaired shells, the distance of the repair from the posterior aperture was measured with an ocular micrometer. The measure used as an indicator of shell repair in this study was the frequency of repaired shells (VALE & REx, 1988). Three sets of statistical comparisons were done (chi- square 2 X 2 tables; SOKAL & ROHLF, 1981) using various groupings of the stations listed in Table 1. First, a com- parison of the effect of depth was tested. Three depth categories were established; inner shelf (Stations R-1, R-8, R-4; depth 90-91 m), outer shelf (Stations R-2, PJ-1, R-5; depth 145-161 m), and upper slope (Stations R-3, R-6, R-7, R-9; depth 409-565 m). Second, the effect of “latitudinal” position on the frequency of shell repair was tested by grouping the stations into three transects: north (Stations R-1, R-2, R-3; ca. 35°05'N), middle (Stations R-8, PJ-1, R-9; ca. 34°55'N), and south (Stations R-4, Page 38 The Veliger; Vol} 355) Nom Table 1 Station data for samples used in comparisons of depth and transect effects on the frequency of repaired shells among each taxonomic category. No is the north transect, So is the south transect, Mi is the middle transect, n is the number of specimens examined, % is the percent of the shells having repair scars. Depth Prosobranchs Sta Transect (m) n % R-1 No 91 58 17.2 R-2 No 161 97 16.5 R-3 No 409 30 16.7 R-4 So 92 482 28.4 R-5 So 154 34 11.8 R-6 So 410 233 14.6 R-7 — 565 27 7.4 R-8 Mi 90 78 11S R-9 Mi 410 138 11.6 PJ-1 Mi 145 495 23.4 R-5, R-6; ca. 34°42'N). Although relatively close together, these three transects have been shown to differ somewhat in the structure of their infaunal communities (HYLAND et al., 1990a, in press). Last, two sets of taxonomic com- parisons were made. First, differences in the frequency of repaired shells were compared among five taxonomic cat- egories: an abundant small prosobranch, Alvinia rosana (Bartsch, 1911), all other prosobranchs, pyramidellidan opisthobranchs, cephalasipdean opisthobranchs, and scaphopods. Second, the large sample sizes available al- lowed for comparisons to be done on a finer scale within each of the major groups. Involved were selected compar- isons between families, genera, species, or groups of each taxonomic category. All available data were used for the taxonomic comparisons. RESULTS anp DISCUSSION The Univalved Mollusk Fauna The univalved mollusks collected during the program were mostly small species, having maximum sizes smaller than 10 to 15 mm in length (ABBOTT, 1974). Prosobranch gastropods were the predominant univalves for the two shelf regions, having densities of 93 and 119 individuals/ m? for the inner and outer shelf stations, respectively. Al- vinia rosana (Rissoidae) was the most common gastropod on the shelf, reaching densities of 192 individuals/m? at Station R-4 (inner shelf) and 273 individuals/m? at Station PJ-1 (outer shelf). Alvinia rosana was not found at any upper slope station. A turrid, Kurtziella beta (Dall, 1919), was the second-most abundant species on the inner shelf, reaching a peak density of 23 individuals/m? at Station R-8. The eulimid Balcis rutila (Carpenter, 1864), with a peak density of 12 individuals/m? at Station PJ-1, was the second-ranked prosobranch on the outer shelf. The upper slope was characterized by relatively high abun- dances of Amphissa bicolor Dall, 1892, which reached 65 Opisthobranchs Scaphopods Total fauna n To n To n To 69 21.7 55 9.1 182 16.5 118 11.9 44 29.6 259 16.6 7 Soil 38 18.4 75 21.3 41 36.6 45 13:3 568 27.8 22 45.4 7 28.6 63 25.4 18 11.1 49 26.5 300 16.3 0 — 2 0 29 6.9 174 16.1 37 16.2 289 14.9 0 — 14 57.1 152 15.8 33 12.1 31 322 559 21.6 and 70 individuals/m? at Stations R-9 and R-6, respec- tively, and Bittzwm fetellum Bartsch, 1911, which reached 64 individuals/m? at Station R-6. Opisthobranchs declined in abundance with increasing depth. Most common on the inner shelf were Cylichna diegensis (Dall, 1919), occurring at densities of 23 indi- viduals/m? at Station R-8, and Volvulella panamica Dall, 1919, found at 12 to 15 individuals/m? at Stations R-1 and R-8, respectively. Odostomia pratoma Dall & Bartsch, 1909, and O. phanella Dall & Bartsch, 1909, were the most common opisthobranchs on the outer shelf, occurring at densities of 17 and 18 individuals/m?, respectively. Opis- thobranchs were rare on the upper slope, occurring at only Stations R-3 and R-6 at densities of 2 and 7 individuals/ m?, respectively. Scaphopods were the predominant univalved mollusks on the upper slope, where their densities ranged from 53 to 214 individuals/m?. Common upper slope species were Cadulus californicus (22 to 197 individuals/m?) and C. tol- mei Dall, 1897 (up to 19 individuals/m?). Scaphopods were less common on the shelf, although Szphonodentalium quadrifissatum (Pilsbry & Sharp, 1898) was particularly numerous on the inner shelf (30 individuals/m7?). Repaired Shell Damage The number of specimens examined and the frequency of repair for each major taxonomic category occurring at the primary stations are listed in Table 1. All raw data including the species examined, taxonomic authorities, number of specimens examined, the number repaired, as well as the depth and location for all stations, are listed in the Appendix. Over all depths and transects, 21.0% of the gastropod shells examined and 18.7% of the scaphopods examined were repaired. Major repairs accounted for 93.5% of all repairs counted. Of the 1026 shells having repairs, only 11 (1.1%) had more than one repair. Among gastro- R. K. Kropp, 1992 Page 39 Percent Repaired 30 25 SIEOI o, etetete! 20 ox re $6 OO.@ KOOL soe S058 SRR 0.0.1 $2525 “et ves 25280 -: & es AKA SOS ‘S ores — SS SLX 25 & ve 5 KX? S ox 1.0.0. 15 ox? 1% ?, 2 Ae; es 0. O,@, 0.0, S52 RR o, Me won o SN Se ‘ : < ‘ % ‘ 10 : | & & % ‘ f ee ‘ a 5 Oe, % 3 of oe be £25 RS % ro <0 Rs eS one oes ‘ rates er r ; erete Hl ovens 0 O08 O00! OOO. Prosobranchs Opisthobranchs Scaphopods Total Category [=] Inner Shelf SJ Outer Shelf Upper Slope Figure 2 The percentage of prosobranch, opisthobranch, and scaphopod mollusks with repaired shell damage for each depth category. pods, 21.8% of the prosobranchs and 16.3% of the opis- thobranchs were repaired. The percent repaired found for opisthobranchs may be a conservative estimate because many of the species examined were cephalaspideans, most of which have the early whorls obscured by the body whorl. Most repairs on gastropod shells were located on the body whorl, although the exact number was not determined. Prosobranchs Inner shelf Outer shelf Scaphopods Inner shelf Outer shelf Table 2 Results of chi-square 2 X 2 test comparing the frequency of shell repairs between pairs of depth category for each major taxonomic group. Opisthobranchs Outer Upper Outer shelf slope shelf DS ell Ur Inner shelf 1.69 ZO 1S Outer shelf Total Outer Upper Outer shelf slope shelf 2.02 8.43** Inner shelf 1.46 1.48 Outer shelf ** Significant x* value at a = 0.01. *** Significant x? value at a = 0.001. Depth and ‘Transect Comparisons The percentages of shells having repairs at each of the three depth categories are shown in Figure 2 and the results of the Chi-square tests in Table 2. Although the frequency of repair of prosobranchs decreased with increasing depth, the only significant differences detected were between the Upper slope 0.11 0.94 Upper slope BiO5 eS 3.68 Page 40 35 30 25 Eo} ro) = © Q 20 c — S sy @ 15 4 [S) SS] = So] ) e oO oY 10 oe i oe ee ne) SS 4 oe 5 es sececenes © O SOSH 00,0, 0.0, OOD O” POOL? Prosobranch: o, x Opisthobranchs Category The Veliger, Vol. 35, No. 1 > SSS LLLP 2 2001 6% <5 0,0, x We <> SS 0,0, OO 205 0% , 5 > OS 0S M, oS 220 Scaphopods North J Middle [=] South Figure 3 The percentage of prosobranch, opisthobranch, and scaphopod mollusks with repaired shell damage for each transect. upper slope stations and both the inner shelf and outer shelf stations. However, conclusions regarding the frequency of repair not only reflect the level of predation within a particular depth regime, but they also necessarily reflect the com- position of the resident molluscan fauna. That the repair frequency is relatively low on the upper slope does not mean that it is low for all taxa present. Note that at upper slope stations species of Bzttzwm show a fairly high inci- dence of repair (24.2%, n = 95), whereas Amphissa bicolor and Astyris permodesta (Dall, 1890) had a combined rate of 10.5% (n = 314). Similarly, the relatively high rate of repair at shelf locations might be explained by the high abundance of Alvinia rosana, a species that has a relatively Table 3 Results of chi-square tests comparing the frequency of repaired shells between pairs of transects for each major taxonomic category. Prosobranchs Middle South transect transect North transect 0.20 2.12 Middle transect 2.68 Scaphopods Middle South transect transect North transect <0.01 0.24 Middle transect 0.18 * Significant x? value at a = 0.05. ** Significant x* value at a = 0.01. *** Significant x? value at a = 0.001. Opisthobranchs Middle South transect transect North transect 1.62 7.65** Middle transect 17.06*** Total Middle South transect transect North transect 0.02 6.38* Middle transect 11.36*** R. K. Kropp, 1992 Percent Repaired A. rosana Page 41 Prosobranchia Pyramidellida Cephalaspidea Scaphopoda Taxon Figure 4 The percentage of repaired shells among each of the major taxonomic categories. The category Prosobranchia excludes Alvinia rosana. high repair rate (24.3%, n = 2923). Other shelf taxa, for example species of Epiztonium, had very low repair rates (no repairs, n = 32). The trend found for scaphopods was opposite to that of prosobranchs, as repaired shells increased in frequency with depth, although the only significant difference was between the inner shelf and the upper slope. The repair frequencies observed for scaphopods at upper slope stations reflect the level of predation on a portion of the scaphopod population present because the predominant species oc- curring on the slope, Cadulus californicus, was too damaged by the collection process to evaluate. Opisthobranchs showed no significant depth-related differences in the frequency of repaired shells. The frequencies of shell repairs for each transect are shown in Figure 3. Among the three major taxonomic categories there was a weak trend for the south transect to have higher proportions of shells having repairs than either of the other two transects. However, opisthobranchs showed the only significant differences; the south transect (33.3% repaired) differed from the north transect (18.0%) and the middle transect (13.9%; Table 3). The total uni- valve fauna also showed significant differences between the south transect (24.0%) and either the north transect (18.2%) orthe middle transect (17.9%). There is no im- mediate explanation for the significantly greater repair frequency shown by opisthobranchs on the southern tran- sect as compared to the other two. ‘Taxonomic Comparisons The frequencies of repaired shells found among the five taxonomic categories examined are shown in Figure 4. The percentage of shells repaired for Alvinia rosana (24.3%) was significantly greater than those for any other group (Table 4). As a contrast to A. rosana, the frequency of Table 4 Results of chi-square 2 X 2 tests comparing the frequency or repaired shells between pairs of taxonomic categories. Proso- Pyrami- branchia* Cephala- = Scaph- dellida spidea opoda Alvinia rosana ATO Se eta motes 24.25*** 23 Prosobranchia* 8.03** 0.15 Oa Pyramidellida 4.44* 0.02 Cephalaspidea 4.12* * Excluding Alvinia rosana. * Significant x? value at a = 0.05. ** Significant x? value at a = 0.01. *** Significant x? value at a = 0.001. Page 42 repair was only 13.0% for all other prosobranchs. Among the other possible comparisons, all showed significant dif- ferences except prosobranchs (excluding A. rosana) versus cephalaspideans and pyramidellidans versus scaphopods (18.7%; Table 4). Prosobranchs: The frequency of repair varied consider- ably among prosobranchs, even within the same general sampling regime. For example, species of Epztonium in the study area had no repairs (n = 32), whereas the high- spired, but relatively unarmored, species of Bitttum had 24.8% with repairs (n = 109), including one specimen with four repairs. The two predominant species of the family Eulimidae collected, Balcis rutila and B. micans (Carpenter, 1864), did not differ in the frequency of re- paired shells (x? < 0.01, P = 0.97, n = 165). Similarly, the two predominant species of the family Turridae col- lected, Kurtziella beta and Kurtzia arteaga (Dall & Bartsch, 1910), did not differ in repair frequency (x? = 0.42, P = 0.52, n = 175). However, a comparison between turrids (16.6%, n = 177) and eulimids (7.6%, n = 170) did show a significant difference in the repair frequency (x? = 6.91, P < 0.01). Shell morphology may help explain the dif- ferences. Relatively high repair rates were found among small, thin-shelled species (Alvinia rosana) and some high- spired species (Bittiwm spp.). Rates were low for heavily sculptured species (Epztonium spp.) or very smooth-shelled species (Balcis spp.). VERMEIJ (1987) predicted that several morphological features of gastropod shells may be effective deterrents to predation. These include narrow apertures, heavy sculpture, and tall spires that permit substantial withdrawal of the snail body into the shell. Three of the examples of predation rates mentioned above seem to con- tradict Vermeij’s predictions. However, two of the taxa, Epitonium and Balcis, may be less susceptible to predation because they are symbionts of cnidarians and echinoderms, respectively, which may reduce the likelihood of detection by a predator. Relevant to Balcis, WAREN (1984) observed repair scars on species of eulimids and speculated that the smooth shell may make it difficult for a predator to grasp the shell. Alvinia rosana, a very small rissoid (length up to 2.6 mm; ABBOTT, 1974), would seem not to offer much resistance to a shell-breaking predator. However, small size may limit the maximum size of a potential predator. The relatively high incidence of repaired damage found for Alvinia may indicate that the most influential predators on these snails were relatively small (e.g., crangonid shrimp). The frequency of repair scars found here for Alvinia was similar to the highest values found by VALE & REx (1989) for upper continental slope rissoids in the northwest At- lantic. Opisthobranchs: Among opisthobranchs, pyramidelli- dans (19.1%) had a significantly higher frequency of re- pairs than cephalaspideans (13.8%; Table 4). Two possible explanations may be given for the differences in repair rates between cephalaspideans and pyramidellidans. First, The Veliger, Vol. 35, No. 1 the repair rates for the former may have been underesti- mated because repairs occurring on early whorls may be- come obscured by the body whorl as the snail grows. Sec- ond, the high rates found for the latter group may reflect the relatively high-spired shells found within that group. Among pyramidellid opisthobranchs, the proportion of re- paired shells for species of Turbonilla (27.6%, n = 134) differed significantly from species of Odostomia (12.8%, n = 258; x? = 13.21, P < 0.001). This difference does not appear to be a result of a difference in spatial distribution because the two genera co-occur throughout the Santa Maria Basin. 7urbonilla is a typically higher-spired, more heavily sculptured taxon than Odostomia, features that con- tribute to greater resistance to shell-breaking predators (VERMEIJ, 1983, 1987). Two repair scars were found on each of four pyramidellid specimens. No significant dif- ferences were detected among various comparisons of ceph- alaspidean genera (x? = 0.01 to 3.14, P = 0.08 to 0.93, n = Si stonZ 39): Scaphopods: Differences in the frequency of repaired shell damage between the predominant species found on the inner shelf (Siphonodentalium quadrifissatum; 14.9%, n = 335) and that found on the upper slope (Cadulus tolmez; 47.2%, n = 53) were significant (x? = 30.51, P < 0.001). Both are similar in size, shell surface texture, and mor- phology (SHIMEK, 1989). Of the indices proposed by Shi- mek, only the whorl expansion rate showed a significant difference between the two species. Shimek hypothesized that the shell morphology of these species may allow rel- atively fast burrowing, which may permit escape from potential predators. If the rate of burrowing does affect the ability to escape, then differences in the sediment com- position of the substrate that affect burrowing rate may explain differences in the intensity of predation observed between the two populations. Within the Santa Maria Basin, sediments at the shelf stations generally are coarser than those at the upper slope stations (KINNEY et al., 1990) although Station R-6 is an exception to this trend. The finer sediments on the upper slope may be more difficult to burrow through, thereby reducing the ability of the resident scaphopods to escape predators. A comparison of the genus Dentalium (14.5%, n = 83) with Cadulus/S1- phonodentalium (19.4%, n = 397) showed no significant differences between the two (x? = 1.11, P = 0.29). Most (54%) repair scars on species of Cadulus /Siphono- dentalium occurred within 2 mm of the posterior aperture, although there was a second peak in the frequency of repair at a distance of 9 mm from the aperture (Figure 5). This would indicate that predation intensity is greater on ju- veniles than adults although this requires the assumption that the damage was done to the lip of the anterior aperture. Juveniles are thought to be less capable burrowers than adults (SHIMEK, 1989). The second peak in frequency of repairs may indicate that the risk of a predatory attack may be greater as adults move toward the surface of the sediment to reproduce (SHIMEK, 1989). Reker Nropp, 1992 Number \ \ \IN \\ \\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ N \ NN \ \ \ \ o- NOY WO fF A DN OO — We) (é) aS Page 43 VMMMM@@@@M@q@qww€tt(tu Distance from Posterior (mm) Figure 5 The distance of repaired shell damage from the posterior aperture of species of Cadulus /Siphonodentalium. The frequencies of repaired shell damage reported here differ with the information concerning predation on scaph- opods provided by SHIMEK (1989) although he did not provide quantitative data. Shimek reported that repaired damage is rare for Cadulus aberrans Whiteaves, 1887, but higher for other species, including Dentalium rectius. Shi- mek also found species of Cadulus to be rare in the stomachs of ratfish (Hydrolagus collier), whereas other scaphopods were common. Comparisons with Other Regions The difficulties in making comparisons among regions and studies have been noted by VALE & REX (1988). However, by using only the number of major repairs en- countered, the results presented here can be compared directly with VALE & REx (1988). Also, the results from the Santa Maria Basin, in which only 1% of the repaired shells had multiple repairs, can be roughly compared to the average number of scars per shell used in other studies (e.g.. VERMEIJ, 1982; VERMEIJ et al., 1980). For prosobranchs, the depth-related differences ob- served here, in which the frequency of repaired shells decreases with the increase in depth from shelf stations to upper slope stations, contrasts with the pattern in the northwest Atlantic. VALE & REx (1988) found repair fre- quencies to-increase with depth from shelf to upper slope, though the difference between the two areas was not sta- tistically significant. The range of frequencies of major repairs for shelf habitats was higher in the Santa Maria Basin (10.3—25.9%) than in the northwest Atlantic (4%). For upper slope habitats the ranges found for the two regions were similar: 3.7-13.3% in the Santa Maria Basin and 8-14% in the northwest Atlantic. Comparisons between the results of this study with those focusing on other northeast Pacific mollusks may not be particularly useful because of the differences in the taxa considered by each study. However, it seems that predation by shell-breakers on some members of the soft-bottom Santa Maria Basin fauna is less important as an agent of selection than it is for some intertidal mollusks (FOTHERINGHAM, 1971; GELLER, 1983) or some popula- tions of shallow, subtidal soft-bottom mollusks (BERGMAN et al., 1983). The overall repair frequency found among prosobranchs off California, if calculated as the average number of scars per shell (0.21; 820 single scars + 6 multiple scars, n = 3761), was generally lower than those found for soft-bot- tom habitats in tropical waters. VERMEIJ (1982) found the average number of scars per shell to range from 0 to 2.23 for tropical communities in the Pacific. Literature records of repair scars among opisthobranchs are scanty. VERMEIJ (1982) lists results for two pyrami- dellid genera, Otopleura and Pyramidella, from Guam. Re- pair frequencies (measured as average number of scars per shell) for those taxa ranged from 0.23 to 0.62. These values Page 44 compare to the 0.18 scars per shell recorded for pyrami- dellids from the Santa Maria Basin (70 single repairs + 4 multiple scars = 74 total scars, n = 392). I have not been able to find any literature records of repair frequen- cies for scaphopods from other regions with which to com- pare directly the results from the Santa Maria Basin study. Predation on Univalved Mollusks in the Santa Maria Basin The types of predators most likely to affect the evolution of univalved mollusks have been reviewed by VERMEIJ (1987) and summarized for continental shelf soft-bottom mollusks by VALE & REx (1988). Little information is available for the predators present in the Santa Maria Basin. Most larger predators were rarely sampled during this program (refer to appendix F-1 in HYLAND et al., 1990a, for taxa collected). A few decapods, such as cran- gonid shrimp and juvenile Cancer, that may prey on mol- lusks were collected. Among the fish likely to prey on mollusks, the ratfish (Hydrolagus collier) and several species of sculpins (Cottidae) have geographic and depth ranges at least partially overlapping those of the study area (MIL- LER & LEA, 1972). Ratfish prey on scaphopods (SHIMEK, 1989) and shallow-water cottids feed on small rissoinid snails, inflicting a distinctive pattern of damage to the shell (NorTON, 1988). Several taxa of echinoderms may ingest molluscan prey without damaging the shells (CAREY, 1972; PEARSON & GAGE, 1984). During this program a mollus- civorous asteroid, Astropecten verrilli (Fisher, 1906), was collected rarely. Some species of brittlestars, Ophiura spp., ingest gastropods whole (PEARSON & GAGE, 1984). Ophiura was uncommon in the Santa Maria Basin. One abundant polychaete in the Santa Maria Basin, Chloeia pinnata Moore, 1911, was found to have an intact Alvinia rosana in its digestive tract. Unsuccessful predatory attacks on the molluscan fauna of the Santa Maria Basin do occur. However, the incidence of these attacks, as measured by the frequency of repaired shell damage, is relatively low. It seems likely that in the Santa Maria Basin, as found by VALE & REx (1988) for the northwest Atlantic, the predators on mollusks are gen- eralists and have not had a major influence on the evolution of the resident fauna. ACKNOWLEDGMENTS I thank G. J. Vermeij and J. Hyland for reviewing the manuscript. J. Shrake, Kinnetics Laboratories, identified most of the mollusks used in this study. Funding was provided by the U.S. Department of Interior /Minerals Management Service (Pacific OCS Office) under Contract No. 14-12-0001-30262 to Battelle Ocean Sciences. LITERATURE CITED ABBoTT, R. T. 1974. American Seashells. Van Nostrand Rein- hold Co.: New York. 633 pp. The Veliger, Vol. 35, No. 1 Carey, A. G., JR. 1972. Food sources of sublittoral, bathyal and abyssal asteroids in the northeast Pacific Ocean. Ophelia 10:35-47, BERGMAN, J., J. B. GELLER & V. CHow. 1983. Morphological divergence and predator-induced shell repair in Alia carinata (Gastropoda: Prosobranchia). The Veliger 26:116-118. Epwarps, D.C. 1969. Predators on Olivella biplicata, including a species-specific predator avoidance response. The Veliger 11:326-333. FOTHERINGHAM, N. 1971. Field identification of crab predation on Shaskyus festivus and Ocenebra poulsoni (Prosobranchia: Muricidae). The Veliger 14:204. GELLER, J. B. 1983. Shell repair frequencies of two intertidal gastropods from northern California: microhabitat differ- ences. The Veliger 26:113-115. HYLAND, J., E. BAPTISTE, J. KENNEDY, J. CAMPBELL, R. KRoppP, C. ROBINSON & S. WILLIAMS. 1990a. Macroinfaunal as- semblages in the Santa Maria Basin off the coast of southern California. Pp. 7-1-7-55. In: M. Steinhauer & E. Imamura (eds.), California OCS Phase II Monitoring Program Year- Three Annual Report, Vol. I. HYLAND, J., D. HARDIN, E. CRECELIUsS, D. DRAKE, P. MONTAGNA & M. STEINHAUER. 1990b. Monitoring long-term effects of offshore oil and gas development along the southern Cal- ifornia outer continental shelf and slope: background envi- ronmental conditions in the Santa Maria Basin. Oil & Chemical Pollution 6:195-240. HYLAND, J., E. BAPTISTE, J. CAMPBELL, J. KENNEDY, R. KROPP & S. WILLIAMS. In press. Macroinfaunal communities of the Santa Maria Basin on the California outer continental shelf and slope. Marine Ecology Progress Series. KINNEY, P., D. HARDIN, T. PARR, F. NEWTON & R. KOLPACK. 1990. Sedimentology. Pp. 4-1-4-60. In: M. Steinhauer & E. Imamura (eds.), California OCS Phase II Monitoring Program Year-Three Annual Report, Vol. I. MILLER, D. J.& R.N. LEA. 1972. Guide to the Coastal Marine Fishes of California. California Department of Fish and Game: Sacramento. 235 pp. NorTon, S. F. 1988. Role of the gastropod shell and operculum in inhibiting predation by fishes. Science 241:92-94. PEARSON, M. & J. D. GAGE. 1984. Diets of some deep-sea brittle stars in the Rockall Trough. Marine Biology 82:247- 258. SCHOENER, T. W. 1979. Inferring the properties of predation and other injury-producing agents from injury frequencies. Ecology 60:1110-1115. SHIMEK, R. L. 1989. Shell morphometrics and systematics: a revision of the slender, shallow-water Cadulus of the north- eastern Pacific (Scaphopoda: Gadilida). The Veliger 32:233- 246. SOKAL, R. R. & F. J. ROHLF. 1981. Biometry: The Principles and Practice of Statistics in Biological Research. Freeman: San Francisco. 859 pp. VALE, F. K. & M. A. REx. 1988. Repaired shell damage in deep-sea prosobranch gastropods from the western north Atlantic. Malacologia 28:65-79. VALE, F. K. & M. A. REx. 1989. Repaired shell damage in a complex of rissoid gastropods from the upper continental slope south of New England. The Nautilus 103:105-108. VERMEIJ, G. J. 1982. Gastropod shell form, breakage, and repair in relation to predation by the crab Calappa. Malaco- logia 23:1-12. VERMEJ, G. J. 1983. Shell-breaking predation through time. Pp. 649-669. In: M. J. S. Tevesz & P. L. McCall (eds.), R. K. Kropp, 1992 Page 45 Biotic Interactions in Recent and Fossil Benthic Commu- VERMEI, G. J., E. ZIPSER & E. C. DUDLEY. 1980. Predation nities. Plenum Publ. Co.: New York. in space and time: peeling and drilling in terebrid gastropods. VERME]YJ, G. J. 1987. Evolution and Escalation: An Ecological Paleobiology 6:352-3064. History of Life. Princeton University Press: Princeton. 527 WaREN, A. 1984. A generic revision of the family Eulimidae. pp. Journal of Molluscan Studies, Supplement 13:1-96. APPENDIX Raw shell repair data. Taxonomic authorship is provided at the first occurrence of each species. Major Minor Total Locality and species n repairs repairs repairs Station R-1 [35°05.8’N, 120°49.2’W; 91 m] Alvinia rosana (Bartsch, 1911) 2; Balcis rutila Carpenter, 1864 Caecum crebricinctum Carpenter, 1864 Epitonium Epitonium caamanoi Dall & Bartsch, 1910 Epitonium sawinae (Dall, 1903) Kurtzia arteaga (Dall & Bartsch, 1910) Kurtziella beta (Dall, 1919) 2 Odostomia pratoma Dall & Bartsch, 1909 Odostomia dinella Dall & Bartsch, 1909 Turbonilla Turbonilla raymond: Dall & Bartsch, 1909 Turbonilla newcombe: Dall & Bartsch, 1907 Turbonilla santarosana Dall & Bartsch, 1909 Turbonilla (Chemnitzia) sp. A Turbonilla (Chemnitzia) sp. F Turbonilla (Pyrgiscus) sp. A Cylichna diegensis (Dall, 1919) 13 Volvulella cylindrica (Carpenter, 1864) 10 Volvulella panamica Dall, 1919 16 Sulcoretusa xystrum (Dall, 1919) 9 Siphonodentalium quadrifissatum (Pilsbry & Sharp, 1898) 5D Station R-2 [35°05.5’N, 120°53.4'W; 161 m] Alvinia rosana 5 Amphissa bicolor Dall, 1892 Antiplanes Balcis Balcis micans Carpenter, 1864 Balcis rutila 2 Bittium fetellum Bartsch, 1911 Bittium quadrifilatum Carpenter, 1864 Boreotrophon sp. 1 Epitonium lowe: (Dall, 1906 Kurtzia arteaga Kurtziella beta Odostomia dinella Odostomia jewetti Dall & Bartsch, 1907 Odostomma phanella Dall & Bartsch, 1909 Odostomia pratoma Odostomia tenuisculpta Carpenter, 1864 Turbonilla Turbonilla newcomber Turbonilla raymond Turbonilla (Chemnitzia) sp. F Cylichna diegensis NON NH HN OYE NN = ONKFNONODRrKPOCOOCOOCOFOFNODTWOOOO Ororoocooocdeocoqoowoocnoeoc°coco OWRPNDONDrROCVOVOOCOAONNTOOCOCO SA nn >) eal A NP RP RPHNnNK WOWNPK NK KNOWN RKP KW PrPooorwwoqodoocnedoedorrr rr OF OY oooooonnodcoocoooordqocoooe i OLOUOh cn OE Or Ore Oona = NOt Ore Page 46 The Veliger, Vol. 35, No. 1 APPENDIX Continued. Major Minor Total Locality and species n repairs repairs repairs Cylichnella culcitella (Gould, 1852) 3 1 0) 1 Cylichnella harpa (Dall, 1871) 4 0 0 0 Siphonodentalium quadrifissatum 28 10 0 10 Dentalium rectius Carpenter, 1864 16 0 3 Station R-3 [35°05.3'N, 121°00.9'W; 409 m] Alaba nr. supralirata 1 0 0 0 Amphissa bicolor 30 4 1 5 Balcis rutila 1 0 0 0 Odostomia pratoma 1 0 0 0 Turbonilla 2 2 0 2 Turbonilla (Chemnitzia) sp. A 3 2 0 2 Cadulus tolmei Dall, 1897 11 4 1 5 Dentalium rectius 20 2 0 2 Dentalium vallicolens Raymond, 1904 7 0 0 0 Station R-4 [34°43.0'N, 120°47.4'W; 92 m] Alvinia rosana 428 119 12 131 Balcis rutila 2 0 0) 0 Cancellaria crawfordiana (Dall, 1891) 1 0) 0 0 Epitonium caamanoi 2 0 0 0 Epitonium lowei 1 0 0 0) Epitonium sawinae 6 0 0 0 Kurtzia arteaga 15 2 0 2 Kurtziella beta 20 2 (0) 2, Mitrella 5 1 0 1 Turridae sp. 1 2 1 0 1 Cyclostremella sp. A 16 i 1 8 Odostomia 1 0 1 1 Odostomia dinella 3 2 0 2 Turbonilla 1 1 0) 1 Turbonilla ambusta Dall & Bartsch, 1909 1 1 0 1 Turbonilla newcomber 1 0 0 0 Turbonilla santarosana 1 0 0) 0 Turbonilla (Pyrgiscus) sp. A 2 1 0 1 Cylichna diegensis 2 0 0 0 Cylichnella culcitella 2 0 0 0 Cylichnella harpa 1 1 0 1 Sulcoretusa xystrum 2 0 0 0 Volvulella cylindrica 6 0 0 0 Volvulella panamica 2 0 0 0 Siphonodentalium quadrifissatum 45 6 0 6 Station R-5 [34°42.7'N, 120°50.8’W; 154 m] Alvinia rosana Balcis micans Balcis rutila 1 Epitonium Epitonium sawinae Eulima californica (Bartsch, 1911) Kurtziella beta Simnia sp. 1 Turbonilla Turbonilla ambusta Turbonilla newcomber Turbonilla santarosana Turbonilla (Chemnitzia) sp. F Cylichna diegensis Cylichnella culcitella ooooocooocnooocoqcjo rPrPrePNWONF DCC OCOF OND FPO RW PRP WW AHO PrPFPNWONRF DCC OF OWN R. K. Kropp, 1992 Page 47 APPENDIX Continued. Major Minor Total Locality and species n repairs repairs repairs Cylichnella harpa 2 0 0 0 Volvulella cylindrica 1 0) 0 0 Siphonodentalium quadrifissatum 6 1 0 1 Dentalium vallicolens 1 1 0 1 Station R-6 [34°41.4'N, 120°57.9’W; 410 m] Alaba nr. supralirata 7 2 0 2 Amphissa bicolor 134 9 3 12 Bittium attenuatum Carpenter, 1864 2 2 0 2 Bittium fetellum 87 16 2 18 Bitttum sp. A 3 0 0 0 Turbonilla 5 1 0 1 Turbonilla newcomber 10 1 0 1 Turbonilla (Chemnitzia) sp. A 2; 0 0 0) Volvulella panamica 1 0 0 0 Cadulus tolmei 32 11 2 13 Dentalium rectius 6 0) 0 0) Dentalium vallicolens 11 0) 0) 0 Station R-7 [34°52.9'N, 121°10.3’W; 565 m] Astyris permodesta (Dall, 1890) 27 1 1 2 Dentalium rectius 2 0) 0) 0) Station R-8 [34°55.3’N, 120°45.9'W; 90 m] Alvinia rosana Amphissa bicolor Balcis micans Balcis rutila Epitonium lower Epitonium sawinae Kurtzia arteaga Kurtzielia beta Trochidae sp. 1 Vitrinella oldroydi Bartsch, 1907 Odostomia dinella Turbonilla Turbonilla aepynota Dall & Bartsch, 1909 Turbonilla ambusta Turbonilla raymondi Turbonilla santarosana Turbonilla (Chemnitzia) sp. F Turbonilla (Pyrgiscus) sp. A Cylichna diegensis Cylichnella culcitella Cylichnella harpa Volvulella cylindrica Volvulella panamica Sulcoretusa xystrum Cadulus fusiformis Siphonodentalium quadrifissatum Dentalium sp. 1 Station R-9 [34°53.7'N, 120°59.1’W; 410 m] Admete rhyssa (Dall, 1919) Amphissa bicolor 12 Balcis micans Balcis rutila Bittium fetellum Astyris permodesta Le NWWNNNPRP RKP ANKE NUA WK NY CO aN — i) WN Ww PAW SEORO UN Oe SOOO LOO OL OOO 1 Or ORONO OO, Oro PWNNANONFODARPKRP KP OrP TOF ANTOOCOOPF PWNPHPWONNODKPRP KP OK ODOORKPUNDOCVOCOF WwW (SS) yy WHOA ON Le >) SO wWOONe Le > o ) Page 48 The Veliger, Vol. 35, No. 1 APPENDIX Continued. Major Minor Total Locality and species n repairs repairs repairs Cadulus tolemi 10 6 1 7 Dentalium rectius 1 0 0) 0 Dentalium vallicolens 3 1 1 Station PJ-1 [34°55.8'N, 120°49.9'W; 145 m] Alvinia rosana 455 9 Balcis micans 3 Balcis rutila 22 Bittium sp. A 1 Epitonium 1 Eulima californica 1 Kurtzia arteaga 2 Kurtziella beta 10 Odostomia 1 Odostomia dinella 12 Odostomia phanella 1 Turbonilla ambusta Cylichna diegensis Cylichnella culcitella Cylichnella harpa Diaphana californica Dall, 1919 Volvulella panamica Cadulus fusiformis Siphonodentalium quadrifissatum Station PJ-2 [34°55.3’N, 120°49.6'W; 142 m] Alvinia rosana Balcis rutila Kurtzia arteaga Turbonilla ambusta Turbonilla santarosana Turbonilla (Chemnitzia) sp. F Volvulella cylindrica Siphonodentalium quadrifissatum Station PJ-3 [34°56.3'N, 120°49.6'W; 138 m] Alvinia rosana 2 Balcis micans Balcis rutila Odostomia dinella Volvulella panamica Siphonodentalium quadrifissatum Station PJ-4 [34°56.3'N, 120°50.2'W; 150 m] Alvinia rosana 2 8 Balcis rutila 3 1 1 jo) aN = (=) Poororr ooocoowororoon SFO OLS TORO Or TOTO: Or Ors “Oo -oOWWNWoOVeCoeO AR SOP eo Ae oS N ONNNFAAWND — Ree ePNE SA 2GeDCOCCOW SOOO Cee S) Sloloreleyo oe aeenes oooOrRGA Solero oO oooOorFA Kurtziella beta Cylichna diegensis Station PJ-6 [34°54.7'N, 120°49.9'W; 148 m] Alvinia rosana Balcis micans Bittium fetellum Epitonium ooosF ooor ooounm iw) NG lee} aN Eulima californica Kurtzia arteaga Turbonilla santarosana Turbonilla (Chemnitzia) sp. A Turbonilla (Chemnitzia) sp. F Bee RR EUAN cooooooOoORN coooooco0ooNn eo0oococoH 5 Cylichna diegensis R. K. Kropp, 1992 Page 49 APPENDIX Continued. Major Minor Total Locality and species n repairs repairs repairs Cylichnella culcitella 1 0) 0 0) Diaphana californica 1 0 0 0 Cadulus fusiformis 1 0 0 0 Siphonodentalium quadrifissatum 14 4 1 5 Station PJ-7 [34°55.8’N, 120°48.6’W; 123 m] Alvinia rosana 5 1 0 1 Balcis micans 1 0 0) 0 Balcts rutila 1 0) 0) 0) Eulima californica 1 0) 0) 0 Kurtziella beta 3 0) 0 0 Mitrella tuberosa (Carpenter, 1865) 3} 0) 1 1 Cylichna diegensis 2 0 0) 0 Suphonodentalium quadrifissatum 21 1 0 1 Station PJ-8 [34°56.9'N, 120°49.9'W; 142 m] Admete rhyssa 1 0 0 0 Alvinia rosana 76 18 6 24 Balcis micans 4 0 0 0) Balcts rutila 4 0) 1 1 Epitonium 1 0) 0) 0 Kurtzia arteaga 2 0 0) 0 Kurtziella beta 2 0) 0) 0 Odostomia phanella 1 0 0 0 Turbonilla santarosana 1 0 0) 0 Cylichna diegensis 3 1 0 1 Siphonodentalium quadrifissatum 12 2 0 2 Station PJ-9 [34°55.8’N, 120°51.2’W; 169 m} Alvinia rosana 28 2 1 3 Balcis micans 3 0 0) 0) Balcis rutila 5 1 0) 1 Bittium fetellum 3 0 0 0 Epitonium 2 (0) 0 0 Kurtzia arteaga 3 0 0 0 Odostomia jewetti 3 1 0 1 Odostomia phanella 95 4 5 9 Odostomia pratoma 1 0 0 0 Turbonilla raymondi 2 1 0) 1 Turbonilla santarosana 2 0 0 0) Turbonilla (Chemnitzia) sp. A 2 0 0 0 Cylichna diegensis 1 0 0 0 Cylichnella 2 0 0 0 Cylichnella culcitella 10 0 0 0 Rictaxis punctocaelatus (Carpenter, 1864) 1 0 0) 0 Volvulella cylindrica 1 0 0 0 Siphonodentalium quadrifissatum 2 1 0 1 Dentalium rectius 11 3 0) 3 Station PJ-10 [34°53.6’N, 120°49.9'W; 147 m] Alvinia rosana 218 29 10 39 Balcis micans 2 0) 0 0 Balcis rutila 5 0 0) 0 Bittium fetellum 3 0 0 0 Epitonium 1 0 0 0 Kurtzia arteaga 2 1 0 1 Sabinella bakeri (Bartsch, 1917) 1 1 0 1 Turbonilla 1 0 1 1 Turbonilla paine: Dall & Bartsch, 1909 1 0 (0) 0 Page 50 The Veliger) Vol 35,Nom APPENDIX Continued. Major Minor Total Locality and species n repairs repairs repairs Turbonilla santarosana 1 0 0 0 Turbonilla (Chemnitzia) sp. F 3 1 0 1 Cylichna diegensis 1 0) 0) 0 Cylichnella culcitella 2 0 0) 0 Siphonodentalium quadrifissatum 12 3 0 3 Station PJ-11 [34°58.0'N, 120°49.9'W; 136 m] Alvinia rosana 172 42 6 48 Balcis micans 6 0 0 0 Kurtzia beta 2 0) 0) (0) Odostomia dinella 3 0 0 0 Volvulella panamica 1 0 0 0 Siphonodentalium quadrifissatum 20 1 1 2 Station PJ-12 [34°55.6'N, 120°49.9'W; 145 m] Alvinia rosana 32 5 3 8 Balcis micans 3 0) 0 0 Bittium fetellum 1 1 0 1 Kurtziella beta 1 0) 1 1 Odostomia 2 1 0 1 Odostomia dinella 2 0) 1 1 Dentalium rectius 1 0 0) 0 Station PJ-13 [34°56.0'N, 120°49.9'W; 144 m] Alvinia rosana 46 ilk 3 10 Balcis micans 2 0) 0 0 Kurtzia arteaga 1 0 0 0 Kurtziella beta 1 0) 1 1 Turbonilla raymondi 1 0) 0 0 Siphonodentalium quadrifissatum 3 1 0 1 Station PJ-14 [34°55.8’N, 120°49.3’W; 134 m] Alvinia rosana 19 4 1 5) Balcis micans D, 0 0 0 Balcis rutila 1 0 1 1 Kurtziella beta 1 1 0 1 Siphonodentalium quadrifissatum 6 1 0 1 Station PJ-15 [34°55.8’N, 120°50.6’W; 155 m] Alvinia rosana 514 90 38 128 Balcis micans 2. 0 0 0 Odostomia dinella 5 0 0 0 Odostomia phanella 1 0 1 1 Turbonilla 2 1 0 1 Cylichnella culcitella 2 0 (0) 0 Siphonodentalium quadrifissatum 3 1 0 1 Station PJ-16 [34°55.0’N, 120°49.0’W; 130 m] Alvinia rosana 24 4 0 4 Balcis micans 3 0) 0) 0 Kurtziella beta 2 0 0 0 Stphonodentalium quadrifissatum 7 1 0 1 Station PJ-17 [34°56.6'N, 120°49.0'W; 126 m] Alvinia rosana 45 8 2 10 Epitonium 1 0 0) 0 Odostomia dinella 2 0) 0) 0 Odostomia phanella 1 0 0 0 R. K. Kropp, 1992 Page 51 APPENDIX Continued. Major Minor Total Locality and species n repairs repairs repairs Station PJ-18 [34°56.6'N, 120°50.8’W; 158 m] Alvinia rosana 164 1 Balcis rutila Bittium fetellum Odostomia dinella Odostomia phanella Cylichnella culcitella Siphonodentalium quadrifissatum Dentalium rectius Station PJ-19 [34°55.0'N, 120°50.8’'W; 167 m] Alvinia rosana 4 pas SN Ss PS Ry oOoroooo°cjo SOOO Orolo Or) oroooo°co Balcis micans Balcis rutila Kurtziella beta Odostomia dinella Odostomia phanella Odostomia pratoma Turbonilla santarosana Cylichnella harpa Dentalium rectius Station PJ-20 [34°50.4'N, 120°49.9'W; 148 m] Alvinia rosana Balcis micans Epitonium Kurtzia arteaga Odostomia dinella Turbonilla santarosana Turbonilla (Chemnitzia) sp. F Cylichnella culcitella Siphonodentalium quadrifissatum Station PJ-21 [35°01.2’N, 120°51.2'W; 143 m] Alvinia rosana Balcis micans Suphonodentalium quadrifissatum Station PJ-22 [34°55.2’N, 120°49.9'W; 143 m] Alvinia rosana Balcis micans Kurtztella beta Turbonilla (Chemnitzia) sp. F Cylichna diegensis Cylichnella Cylichnella culcitella Diaphana californica Siphonodentalium quadrifissatum Station PJ-23 [34°56.3’N, 120°49.9'W; 143 m] Alvinia rosana 119 3 Balcis micans 10 Kurtzia arteaga 1 Kurtziella beta Cylichna diegensis Cylichnella culcitella Diaphana californica Volvulella panamica Siphonodentalium quadrifissatum 1 WNANRKF FS NNN LY NoOCcOrOrF OCT OM oOoooooooow NOCGCOF OF OC NP oPrPeran Seonoocoooonw Niooooeaoo°o NONDODOONW = & oon oon ooo nN BPeENPP eR PBPoorPoccoon OLere1O16) SO. moOoFCooCoCORn — NePeENEwW Booococcond Boooooooa NOODDDCOONG The Veliger 35(1):52-63 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 Observations on the Biology of Turritella gonostoma Valenciennes (Prosobranchia: ‘Turritellidae) from the Gulf of California WARREN D. ALLMON Department of Geology, University of South Florida, Tampa, Florida 33620, USA DOUGLAS S. JONES Florida Museum of Natural History, University of Florida, Gainesville, Florida 32611, USA AND NIKKI VAUGHAN Department of Geological Sciences, California State University, Northridge, Northridge, California 91130, USA Abstract. Turritella gonostoma Valenciennes, 1832, is common in several areas in the northern Gulf of California, particularly during the winter months. It may move into shallow water at these times, perhaps following cool, nutrient-rich waters associated with seasonal upwelling, and move off into deeper water when water temperatures rise during the summer. It lays eggs in shallow water in February- April. Eggs hatch as veligers that swim in the plankton for 2-3 weeks. Preliminary feeding experiments suggest that this species does suspension feed, but not at a very high rate compared to other suspension- feeding mollusks. Analysis of oxygen isotopic variation in the shells of three individuals suggests that growth rates are high but variable, and that longevity is around two years. INTRODUCTION Turritelline gastropods (family Turritellidae, subfamilies Turritellinae and Protominae; MARWICK, 1957) are di- verse and abundant components of many fossil and living marine benthic communities, and turritelline-dominated assemblages are common in the geological record; yet very little is known about the biology of individual species (see ALLMON, 1988, for review). Here we present new infor- mation on a single living species, 7urritella gonostoma Va- lenciennes, 1832, from the Gulf of California (expanding and adding to an earlier summary; ALLMON, 1988), and put this information into the wider context of what is known about other turritelline species. Improved knowl- edge of the biology of these organisms may contribute to improved understanding of the mode of origin and envi- ronmental significance of turritelline-dominated fossil as- semblages. Voucher specimens have been deposited in the Los Angeles County Museum of Natural History and the U.S. National Museum of Natural History. GEOGRAPHIC DISTRIBUTION anpb THERMAL RANGE Turritella gonostoma occurs throughout the Gulf of Cali- fornia as far north as Bahia la Cholla, north of the town of Puerto Penasco (31°20'N) (Figure 1) and as far south as the southwestern coast of Colombia (2°20'N); it has not been reported from the west coast of Baja California (re- cords from the literature and collections of the Department of Malacology, Los Angeles County Museum). It is one of the most common turritelline species in the gulf, par- ticularly in the north. In this region it occurs in shallow waters (<5 m) in variable densities (<1-100/m7?), higher W. D. Allmon et al., 1992 values always being observed between December and May (ALLMON, 1988). The location of individuals during the remainder of the year remains unknown. Although Par- KER (1964:98) reported large numbers of turritelline shells freshly deposited on beaches in the gulf, he did not collect a single live individual in extensive grab-sampling of ben- thic macrofauna. At Bahia la Cholla, where most of our observations were made, average monthly water temperatures range from 13.8°C in January to 29.4°C in August (THOMSON, 1987, cited in FuRsSICH & FLESSA, 1987). RODEN & GROVES (1959) reported average surface water temperatures at Puerto Penasco of 14.9°C in January and 31.2°C in Au- gust. PARKER (1964:39) pointed out that the south side of Tiburon Island in the northern Gulf of California is the extreme northern end of the geographic range of many benthic invertebrate species, and suggested that one reason might be the extreme seasonal fluctuations in surface water temperature that characterize the northern gulf. Occa- sional winter cold snaps, during which water temperatures may fall to as low as 8°C, sometimes result in massive mortality of stenothermal algae and macroinvertebrates (FLESSA & EKDALE, 1987). Whether individual turritelline species, or individual animals, are actually adapted to survive and reproduce in a wide range of temperatures is not known. It is possible, however, that at least some species are adapted to thermal instability in another way. If, as discussed by ALLMON (1988), at least some turritelline species are capable of moving significant distances, perhaps on the order of 10?- 10° m in days or weeks, this might allow them to move into shallow water, possibly to reproduce, during times of upwelling and cooler waters, and to move back out to deeper water afterward. SUBSTRATE, LIFE POSITION, Anp MOVEMENT Individuals of Turritella gonostoma at Bahia la Cholla, Sonora, and Mulege, Baja California Sur occur both on and within fine to coarse sands in 0.5-2 m of water at low tide (VAUGHAN, 1983; ALLMON, 1988; HERTZ, 1990) (Fig- ure 2). Some individuals we observed at Bahia la Cholla had their anterior end partly to completely buried (Figure 2A), the location of the aperture marked only by one or a pair of small, subcircular holes in the sand about 1 cm across. Others had only their apex buried, at a very low angle, in the sediment (Figure 2B), and still others were not covered by sand at all. The shell aperture was oriented perpendicular to the substrate surface in all stationary live individuals observed in their natural setting, both buried and exposed. Crawling individuals were oriented aperture down. Animals kept in laboratory aquaria assume similar orientations (ALLMON, 1988). Several species of macro- phytic algae were growing on the apertural (left) side of the shell in a number of live individuals at Baha la Cholla Page 53 CALIFORNIA ARIZONA BAJA CALIF. ® Tucson Puerto Penasco Figure 1 Map of the northern Gulf of California, showing location of Bahia la Cholla. (Figure 2C, D), suggesting that these surfaces had been exposed above the sediment for some length of time. BUCHANAN (1958) and BATHAM (1969) also reported mac- rophytes on the shells of living turritellines. On one oc- casion we observed a single living individual buried ver- tically in the substrate, apex down (Figure 2E). Neither the significance of this orientation nor the means by which it is attained is known. Although active burrowing was observed by individuals in aquaria, some burial in nature seems to be passive by tidal currents washing sediment over the shell. In aquaria, individuals usually remained motionless, partly or wholly buried in the sediment, but on occasion we observed much greater levels of activity. Animals as large as 10 cm in length are able to pull themselves out of the water and crawl up the sides of aquaria. Activity seemed to be greatest at night. SUSCEPTIBILITY TO PREDATION Observed rates of predation on Turritella gonostoma by drilling gastropods is above average for Recent turritellines (DUDLEY & VERMEIJ, 1978; see also ALLMON et al., 1990). We have observed predation on 7. gonostoma in the lab- oratory by Muricanthus nigritus Philippi, 1845, which also lives in the Gulf of California. The following description of an encounter between these two species is representative of several we observed. At approximately 0900 hr on 23 April 1990, one M. nigritus attached itself toa T. gonostoma that was resting completely exposed on the substrate. The turritellid gave no observable reaction. At approximately 1600 hr on 30 April the turritellid became agitated, as indicated by an extrusion of mucus and movement of its cephalopedal mass erratically about the aperture; we pre- sume that the predator had penetrated its shell at this time. When the Muricanthus was removed, we observed that only the visceral mass, and not the cephalopedal mass, had been consumed. Page 54 The Veliger, Vol. 35, No. 1 Figure 2 Orientations of live Turritella gonostoma at Bahia la Cholla. A. Specimen with only posterior (apical) end buried in sediment. Depth ~1 m at low tide. B. Specimen with only anterior (apertural) end buried in sediment. Depression (arrow) marks location of aperture and incurrent and excurrent siphons. Depth *1 m at low tide. In both A and B, apertures are oriented normal to the sediment surface. C and D. Specimens with macrophytic algae growing on shell. C taken in life position at depth *1 m at low tide. D taken in aquarium in the field; specimen pictured with egg mass found next to it in about 1-m depth at low tide. E. Live specimen oriented apex down, exposed on tidal flat at low tide. W. D. Allmon et al., 1992 Page 55 FEEDING Limited circumstantial evidence and direct observations of other species indicate that at least some, and perhaps most, living turritellines are predominantly suspension feeders (e.g., YONGE, 1946; ALLMON, 1988). Turritellines suspen- sion feed by taking water in through an inhalant siphon, trapping particles on the ctenidial surfaces, and trans- porting them to the mouth in a mucous rope (YONGE, 1946; FRETTER & GRAHAM, 1962:571). Water is expelled from an excurrent siphon formed by a fold of mantle tissue. When approximately 1 mL of fluorescein dye was intro- duced into the inhalant siphon of an individual of Turritella gonostoma, it began to be expelled by a smooth laminar flow in approximately 10-12 sec, a relatively rapid rate compared to other gastropods and suggestive of strong water-pumping ability (R. Linsley, personal communi- cation). Other data, however, suggest that at least some turri- tellines may feed by other means (e.g., deposit feeding or surface grazing) at least part of the time (ALLMON, 1988). There evidently have been no measurements of either rates of suspension feeding or the degree to which suspension feeding is obligate in any turritelline species. We therefore undertook simple measurements of filtration rates by 7ur- ritella gonostoma. Methods Three snails were placed in individual glass bowls con- taining a suspension of the naked flagellated alga [sochrysis galbana at concentrations of about 16 x 10% cells/mL. Animals were kept in light or dark conditions at 19—20°C. Algal concentrations were measured hourly using a model ZM electronic particle counter (Coulter Electronics). Fil- tration rates were calculated using the method described in NEWELL (1979:463). Individuals were removed from their shell, rinsed with distilled water, dried at 40°C for 12 hr, and weighed to determine tissue weight. Results Two trials were run with each animal under both light and dark conditions. The highest filtration rates observed are plotted in Figure 3 and summarized in Table 1. In four of the 12 trials, particle concentrations increased, indicating that no net particle clearance was taking place and, probably, that pseudofeces were being produced. Re- sults differed by individual animal; the largest (individual No. 2) showed consistently high filtration rates, while in- dividual No. 1 never showed any measurable particle clear- ance. Filtration rates appeared to be higher in the dark than the light, but the sample size is too small to confirm this pattern. Even the highest filtration rates observed in these tur- ritellids are 1-2 orders of magnitude lower than those commonly observed in other active filter feeders (e.g., Cre- pidula fornicata by NEWELL & KOFOED, 1977; many bi- Table 1 Maximum observed filtration rates of three individuals of Turritella gonostoma in laboratory feeding experiments. Avg. filtration Observed Dry rate shell tissue Duration of | (mL/hr/g) Animal length (mm) mass (g) feeding (hr) _ [no. of trials] 1 90.35 0.26 18.0 (light) 0.00 [2] 5.0 (dark) 0.00 [2] 2 102.10 0.45 3.0 (light) 1.98 [2] 7.0 (dark) 2.94 [2] 3 80.25 0.12 5.0 (light) 0.00 [2] 8.0 (dark) 5.08 [2] valves, see NEWELL, 1979). At least two explanations are possible for these very preliminary results. The animals may not have exhibited their normal feeding behavior in the laboratory setting; although they appeared healthy, they may not have been. Alternatively (or in addition), these turritellines may be very minimal suspension feeders, either because of low metabolic needs or because they obtain food in other ways (e.g., grazing and/or deposit feeding). REPRODUCTION Data from a number of living turritelline species suggest that reproduction is seasonal (ALLMON, 1988). All avail- able anecdotal reports of the occurrence of eggs of Turritella gonostoma in the northern half of the Gulf of California indicate that spawning occurs only in the late winter and early spring (February—April; ALLMON, 1988; HERTZ, 1990). Large, apparently adult individuals are often found associated with egg masses, either individually or in large aggregations of hundreds to thousands (ALLMON, 1988; HERTZ, 1990; Figure 4). Egg masses contain 200 to 300 egg capsules (Figure 2D), which are loosely attached to a central membrane and a simple holdfast, which may be secured to sandy or hard substrates. Egg capsules are 2-3 mm in diameter, and just before hatching contain 1-12 veligers (mean = 3.65, n = 23) (Figure 5). We have observed egg laying in the field once. In April 1987 at Bahia la Cholla a female was observed oriented horizontally at low tide in a small tidal channel in 1-2 cm of water. She had already attached a holdfast at one end of the egg mass to the substrate (a shallowly buried, dead bivalve shell) when observations began. Eggs were ex- truded over the next 15-20 min in several, seemingly peri- staltic surges. A holdfast at the other end of the egg mass was then attached to a nearby rock, the two attachment points being about 4 cm apart. This arrangement allowed the egg mass to form an arch, which moved back and forth with the waves. Following completion of egg laying, the Page 56 Filtration rate (cells/ml x 104) —o— animal 1 —4— animal 2 The Veliger, Vol. 35, No. 1 6 8 10 TIME (hr) 6 8 10 TIME (hr) —O- animal 3 --x-- control Figure 3 Filtration rates of three individuals of Turritella gonostoma in the laboratory. Graphs show highest observed filtration rates (data summarized in Table 1). A. Dark. B. Light. See text for details. female remained at the site as long as observations con- tinued, but was absent the following morning (approxi- mately 12 hr later). Fourteen days after being laid, a portion of the egg mass was transported to a laboratory, and eggs began to hatch en route. This may have been due to disturbance in transit, but is consistent with experience with other egg masses collected in the field and whose laying date was not known but which hatched after 14-21 days in laboratory aquaria. From the egg capsules hatched veligers (Figure 6) that swam in laboratory aquaria for 10-12 days before settling and metamorphosing. The first veligers to settle appeared to prefer the glass sides of the aquaria. Veligers added at least one full whorl to the shell during their planktonic period; newly settled juveniles generally had around 2% whorls (Figure 6; ALLMON, 1988). Eggs and veligers are very sensitive to high temperatures. In one mass left over- night at room temperature (22°C) all veligers died. GROWTH Life-span and age of reproduction have been determined by direct observation in only one turritelline species. In the Australian species Gazameda gunnu (Reeve, 1849), reproduction begins when the adults are 2.5-3 years of age, and is repeated throughout a life-span of 6-7 years (CARRICK, 1980). WRIGHT (1956) observed well-devel- oped gonads in individuals of the northeastern Atlantic species Turritella communis Risso, 1826, as small as 23 mm in length. According to the growth curves calculated W. D. Allmon et al., 1992 Page 57 Figure 4 Several live individuals of Turritella gonostoma (arrows) surrounded by abundant egg masses, exposed at low tide on the tidal flat at Bahia la Cholla. Photo taken in April 1987. by CADEE (1968), this corresponds to an age of less than one year. Life-span is unknown, but maximum reported size is approximately 45 mm, corresponding to an age of Just over two years on Cadée’s curve. Buchanan (personal communication) has attempted to age individuals of 7. communis by counting collabral growth lines, and on this basis suggests a longevity of at least 15 years, with pop- ulation modes of 10-11 years. No such data have ever been presented for Turritella gonostoma. We have been unable to keep large numbers of healthy, feeding adults alive in the laboratory for more than about five months, and so have not made direct ob- servations. The only direct observation of growth we have been able to make in 7. gonostoma was provided by the repair of shell breakage by one individual: slight breakage around the aperture occurred during transport, and by the time the animal died two months later it had laid down approximately 5 mm of thin, translucent white shell around the outside of the aperture (Figure 7). In the absence of other direct observations of growth, we sought to assess the growth history of 7urritella gon- ostoma via analysis of stable isotope profiles of shell car- bonate. Seasonal cycles in the oxygen and/or carbon iso- tope records (18O/'°O, '°C/'?C) have been used to interpret the age and growth rate of many species of fossil and Recent mollusks (e.g., KRANTZ et al., 1987). This technique has found its widest application in work on marine bivalves, but growth histories of marine gastropods have also been reconstructed from stable isotope studies (e.g., WEFER & KILLINGLEY, 1980). Materials and Methods Three specimens of Turritella gonostoma from Bahia la Cholla were used for isotopic study. These specimens had shell lengths of 109, 87, and 69 mm (these lengths will be used to identify the specimens in the discussion that fol- lows). Specimens 87 and 69 were collected alive on 13 February 1987, while specimen 109, collected on the same day, was not alive. The shells of specimens 87 and 69 were complete, whereas the apical end of specimen 109 was abraded. By comparison with the complete shells, we es- timate that the initial six whorls were missing from 109 (Table 2). Figure 5 Egg capsule of Turritella gonostoma encrusted with sand grains. Scale = 1 mm. The outer surface of each specimen was washed to re- move extraneous materials. Using a small dental burr (di- ameter <0.5 mm) mounted in a hand-held drill, samples of aragonitic shell material, each weighing approximately 0.5 mg, were obtained by grinding shallow grooves into the outer shell layer of each whorl in an orientation parallel to the external growth lines (Figure 8). Detailed, serial sampling of specimen 109 yielded 117 separate samples, each spaced about 1-2 mm apart (Figures 8, 9a). After analyzing these samples isotopically and noting the smooth pattern of isotopic change throughout ontogeny, we decided Figure 6 Veligers of Turritella gonostoma approximately 5 days after hatch- ing. Scale bar = 1 mm. The Veliger, Vol. 35, No. 1 Figure 7 Specimen of Turritella gonostoma in laboratory aquarium, show- ing regrowth of broken shell achieved in two months (arrow). that such a dense sampling strategy was not required for the other two specimens. Only one sample per whorl was recovered from specimens 87 (Figure 9b) and 69 (Figure 9c), with the exception that because of small initial whorls in specimen 69, whorls 1 and 2 were combined to produce one sample, as were whorls 3 and 4. This resulted in 12 samples from the 14 whorls of specimen 69 (Figure 9c). Each sample of powdered aragonite was washed in 15% H,O, for 3 hr to remove organic contaminants. The H,O, was then pipetted off and the samples were flushed with distilled H,O, followed by two consecutive washings with methanol (99.9%). ‘The samples were then dried overnight and stored. All samples were analyzed according to stan- dard techniques (WILLIAMS et al., 1977), which involved an initial reaction im vacuo with “100%” orthophosphoric acid at 70°C for 0.25 hr. An on-line, carbonate preparation system facilitated the production and purification of the evolved CO, gas. The isotopic difference between the de- rived sample CO, and the PDB standard was determined with a fully automated, VG Isogas PRISM Series I mass spectrometer equipped with triple collectors and micro- inlet system. All values are reported in the standard (6) notation where: 6'°O = [(18O/'°O)sample/('8O/'°O)standard — 1] x 10° per mil Determinations of 6'°C were made concurrently with 6'8O. Average reproducibility, as evidenced by duplicate analyses and standards run before and after sample strings, was approximately +0.1 per mil (%o). Results The stable isotope records for each specimen are plotted in Figure 9 in standard fashion, with lighter (depleted) values toward the top. This graphical convention derives from paleotemperature interpretations of the isotopic vari- ations (6'%O in particular) in which warmer temperatures, and their correspondingly depleted isotopic values, are W. D. Allmon et al., 1992 Table 2 Specimens of Turritella gonostoma used in isotopic age analysis. Observed Estimated Observed Estimated no. of no. of Estimated length (mm) length (mm) whorls whorls age (years) 109 125 11 17 NGS) 87 87 15 15 15) 69 69 14 14 5) plotted toward the top, while cooler temperatures and heavier (enriched) values toward the bottom. All of these figures plot the record of isotopic change ontogenetically, from the shell apex (left) to the aperture (right). The most detailed isotopic profiles were obtained for specimen 109. Sample spacing remained constant in this individual at 1-2 mm so that with ontogenetic whorl ex- pansion, the number of samples recovered per whorl in- creased steadily toward the aperture. The 117 paired ox- ygen and carbon isotopic analyses exhibited a smooth pattern of variation throughout ontogeny, except for the final few samples (111-117) which indicated an episode of comparatively abrupt change. The 6'8O record in specimen 109 is characterized by a strong cyclicity. Approximately 1.5 cycles are evident, with a Maximum isotopic amplitude of 2.7 per mil (—2.0 to +0.7). The initial, broad cycle, which includes whorls 7 through the beginning of 17, differs sharply from the final one-half cycle, which occurs entirely within the latter half of whorl 17. This pattern almost certainly reflects an on- togenetic decline in shell growth rate (as discussed below). The carbon isotope record, in contrast, shows little or no cyclicity. Following a slight initial enrichment of about 0.5 per mil (from the apex to whorl 11), the remainder of the record reveals a weak ontogenetic trend toward lighter carbon isotopic values. This trend is gradual for the most part but, like the oxygen record, it is abrupt over the last half of the final whorl. The 6'°C values range between +2.4 and +3.6 per mil with a maximum amplitude of 12 The 6'8O and 6°C records for specimens 87 and 69 are broadly comparable to those observed for specimen 109. The record of specimen 87 (Figure 9b) begins with a protracted episode of nearly constant 6'*O values (whorls 1-11). An abrupt enrichment occurs in whor! 12, followed by depletion through whorl 14, completing the first major cycle. The slightly enriched value from the final whorl (15) suggests that a second cycle had begun at the time of capture. The 6'*O values range between —1.9 and —0.3 per mil for an overall isotopic amplitude of 1.6. In specimen 69 (Figure 9c) the 6'8O pattern is similar except that the enrichment phase of the cycle begins sooner (whorl 7-8) than in specimen 87 and is more gradual in nature. The 6'8O values range from —1.5 to +0.5 per mil for a slightly Page 59 Figure 8 Photograph of Turritella gonostoma specimen 109 showing serial isotopic sampling pattern. Specimen length = 109 mm. greater amplitude of 2.0. As in specimen 87, the enriched 6'8O value from the final whorl of specimen 87 suggests a second cycle was initiated prior to capture. As in specimen 109, the 61°C profiles of specimens 87 and 69 do not reveal major cycles. The pattern in specimen 87 is one of gradual enrichment from whorl 1 through 13, followed by a brief episode of depletion across the final two whorls. The 6'°C values vary between +1.9 and +3.1 per mil (isotopic amplitude = 1.2). In specimen 69 the overall 6'°C amplitude is even smaller (0.4), with values ranging from +2.4 to +2.8 per mil, and no clear onto- genetic trends are evident throughout the record. Extreme- ly weak correlations were observed between the 6'*O and 6'C records of any particular specimens, as evidenced by r values of 0.089, 0.032, and 0.035 for specimens 109, 87, and 69, respectively. Discussion The major cycles in the 6'*O profiles of all three spec- imens are best interpreted in terms of the annual cycle of temperature change in the northern Gulf of California. Page 60 The Veliger, Vol. 35, No. 1 Turritella gonostoma (109 mm) a i=) a ° Summer = as) 12 14 Whorl Number a [=) a (6) o r fas) 0 20 40 60 80 100 120 Apex Aperture a Sample Number a a a o a a ' fe} (=) ® ® Turritella gonostoma (87 mm) as Turritella gonostoma (69 mm) a a o. Q ¢ = o (S) ‘a oO Aperture r= ) 5 10 15 b Whorl orl Number c Whorl Number Figure 9 Plots of oxygen and carbon isotopic records, relative to the PDB standard, for three specimens of Turritella gonostoma from Bahia la Cholla. Results are plotted from the apex (left) to aperture (right). a. Specimen 109; 117 paired analyses are plotted in serial, ontogenetic order from left to right. b. Specimen 87; whorl numbers (= sample numbers) increase in ontogenetic order left to right. c. Specimen 69; whorl numbers (= sample numbers) increase in ontogenetic order left to right. Mollusks typically form their shell at or near oxygen iso- topic equilibrium with seawater and, barring significant salinity variations, the influence of temperature upon the 6'°O composition of shell carbonate is well constrained. Strong seasonality, such as obtains in the northern gulf, has been implicated as the principal cause of such cyclical variation in ontogenetic 6'%O profiles in a variety of mol- lusks (e.g., ERLENKEUSER & WEFER, 1981; WILLIAMS et al., 1982; JONES, 1983; JONES et al., 1983, 1989; KRANTZ el al., 1984; CHINZEI et al., 1987; ROMANEK et al., 1987). Because the changes in 6'%O closely track the annual tem- perature cycle, it is possible to calculate the number of years a specimen has lived and to approximate the range of temperatures over which shell formation has occurred. The 6'8O record of specimen 109 (Figure 9a) represents the best data set of the three for isotopic calculations. As indicated, summer 6'%O values reach —2.0 per mil in this specimen while winter values extend to +0.7 per mil, for a total range of 2.7. According to the relationships de- scribing equilibrium oxygen isotopic fractionation between calcium carbonate and water at environmental tempera- tures (summarized by ANDERSON & ARTHUR, 1983), a 1.0 per mil change in carbonate 6'8O corresponds to a change of about 4°C in temperature. Hence, the measured range W. D. Allmon ez al., 1992 Page 61 in 6'8O of 2.7 per mil suggests an annual range in seawater temperature of approximately 11°C. Water samples were not collected as part of this study. Nevertheless, by making the reasonable assumption that the oxygen isotopic composition of the water in which specimen 109 grew approximated the modern global ma- rine average, the aragonite-water fractionation relation- ship of GROSSMAN & Ku (1986) provides a way to estimate the temperature regime under which calcification occurred. Substituting the end-member 6'8O values of —2.0 and +0.7 into this equation yields a maximum (summer) tempera- ture of about 28°C and a minimum (winter) temperature of about 17°C. These estimates match actual measured temperatures from Bahia la Cholla remarkably well (see above). Theoretically, the 6'8O records in specimens 87 and 69 would yield similar temperature estimates if the isotopic sampling scheme had been as detailed as that of specimen 109. The reduced annual isotopic amplitude measured in specimens 87 and 69 results from the coarse sampling pattern, which has the effect of homogenizing the seasonal oxygen isotopic signal and thereby minimizing the extent of the yearly variation. Based on the number of yearly cycles in the 6!8O records of the three specimens of 7urritella gonostoma, only spec- imen 109 was equal to or slightly older than about 1.5 years at the time of death (Table 2). Specimen 109 had experienced two summers and was beginning its second winter when it died (Figure 9a). Because of the coarser sampling scheme adopted for specimens 87 and 69, sub- dividing the yearly cycles into seasons was more difficult. Nevertheless, both individuals appear to have begun their second year of life when they were collected (Figures 9b, c). Neither lived longer than 1.5 years. Using these interpretations of the 6'*O data, a growth curve for each specimen of 7urritella gonostoma was con- structed (Figure 10). Growth rates are clearly fastest in the first year of life, slowing significantly thereafter. In fact, the samples from the apertural end of specimen 109 (Figure 9a) suggest that the decline in growth rate may be rather dramatic during the second year of life. The most rapid seasonal growth seems to occur during the summer, particularly during the first year. This is suggested by the numerous depleted 6'°O values at the beginning of each profile that represent the early shell whorls and record “warm” temperatures. Although the initial six whorls of specimen 109 are missing, their small size and rapid growth rate (based on comparison with the other two specimens) suggest that this animal initiated growth in either late spring or early summer. This is consistent with the ob- served timing of reproduction in 7. gonostoma from the northern Gulf of California, discussed above. Although it is possible that these individuals lived for some unknown length of time without significant shell growth, straightforward interpretation of the data suggests that this species is not long-lived. In its estimated total Whorl Number Turritella gonostoma ) 1 2 Age (years) Figure 10 Growth curves (whorl number vs. age) for Turritella gonostoma specimens 109, 87, and 69, based on interpretations of annual cycles in 6'8O records (see Figure 9). length (Table 2) specimen 109 exceeds reported average adult sizes for Turritella gonostoma (100 mm—ABBOTT & DANCE, 1982; 115 mm—KEEN, 1971). The comparatively young age (<2 years), as well as the dramatic growth rate reduction seen in the last whorl of this large specimen, favor an interpretation of short life-span (2 years or less) for this species. Even if the animals tracked cooler waters to some degree by moving into greater depths in the sum- mer (see below), the strong cyclicity in the isotopic signal virtually demands a short life-span in these specimens. Because the smaller specimens (87 and 69) are approx- imately the same age as 109, the larger size of the latter must have resulted from higher growth rate. This pattern of large specimens growing faster (as opposed to living longer) than their smaller counterparts has been observed in a variety of bivalves (e.g., JONES, 1980; JONES ef al., 1989), and may hold equally well for gastropods. In a study of strombid gastropods from Bermuda, WE- FER & KILLINGLEY (1980) also related cycles in 6'%O pro- files directly to seasonal temperature variation. The 6?C records, in contrast, proved much more difficult to inter- pret, despite exhibiting a fair degree of cyclicity. Two major factors contribute to this situation: (1) carbon iso- topic composition of carbonate skeletons has received less study than the interpretation of oxygen isotopic ratios (WE- FER & KILLINGLEY, 1980), and (2) whereas temperature can often be isolated as the principal cause of oxygen isotopic variation, no such simple cause has been identified for carbon. Possible influences on the 61°C composition of molluscan shell carbonate include: variations in ambient total dissolved carbon (TDC) resulting from changes in Page 62 productivity, runoff, upwelling, mixing, or other factors (e.g., EISMA et al., 1976; KILLINGLEY & BERGER, 1979; ARTHUR ef al., 1983); temperature (GROSSMAN & Ku, 1986); mixing of metabolic carbon derived from various food sources with dissolved inorganic carbon (TANAKA et al., 1986); fractionation due to changes in shell growth rate (z.e., CaCO, precipitation rate; TURNER, 1982); and/or organismal vital effects of various kinds (ROMANEK & GROSSMAN, 1989). The 6'°C profiles of the three turritelline specimens analyzed here (Figure 9) display much less variation than the strombids from Bermuda. If, as suggested by ALLMON (1988), turritellines in the Gulf of California occur in large numbers in areas of seasonal upwelling, it is possible that Turritella gonostoma at Bahia la Cholla moves into shallow water in the winter coincident with upwelling and moves offshore into deeper water during the summer. Following cooler, nutrient-rich water might have the effect of damp- ing the '°C signal in the shell (e.g., KILLINGLEY & BERGER, 1979; ARTHUR et al., 1983). Only more detailed work can confirm this possibility. CONCLUSIONS Although the results presented here are preliminary and based on small sample sizes, they have implications for the interpretation of turritelline paleobiology and the ecology and taphonomy of turritelline-dominated fossil assem- blages. (1) Age and shell size appear to have no direct rela- tionship to one another in Turritella gonostoma. If this applies to other turritellines as well, paleontological studies of heterochrony (which usually use size as a proxy for age) will be limited to conclusions about pattern (7.e., pera- morphosis vs. paedomorphosis) rather than process (e.g., progenesis vs. neoteny) (see JONES, 1988, for further dis- cussion). (2) These gastropods are surprisingly young for their size. If the interpretation of short life-span is correct for these and other turritellines, it suggests that growth rates are very high. If all or most individuals of a species in an aggregation are of the same age, this implies that turri- tellines may be opportunistic, “boom-and-bust” species (e.g., LEVINTON, 1970). (3) Patterns of density in turritelline-dominated assem- blages, Recent or fossil, may be due less to patterns of recruitment (as in many other benthic invertebrates; e.g., BUTMAN, 1987) than to patterns of seasonal movement, migration, and/or aggregation. A great deal of further investigation is needed to test and expand upon the conclusions presented here. (1) We still do not know when, why, and to what degree turritellines burrow. (2) While it is known that at least some turritellines feed by methods other than suspension feeding (see ALLMON, 1988), it remains to be determined to what degree this is the case in different species. The Veliger, Vol. 35, No. 1 (3) Many more data are needed on the interaction be- tween turritellines and their predators. Can they escape? If so, how do they and how often? (4) Are all turritellines as short-lived as Turritella gon- ostoma appears to be? Investigation into these topics will greatly increase our understanding of the history and biology of this important group of gastropods. ACKNOWLEDGMENTS WDA thanks J. Pechenik for assistance and advice on feeding experiments and for commenting on an earlier draft, R. D. Turner for helpful discussions, and R. L. Aiello for assistance in the field and elsewhere. DSJ thanks D. Hodell for assistance with mass spectrometry and I. Quitmeyer for help with sample preparation. NV thanks D. Laidig for field assistance. Work at the Los Angeles County Museum was made possible by a grant from the Lerner Gray Fund for Marine Research to WDA. We are grateful to an anonymous reviewer and especially to C. S. Hickman for helpful comments on an earlier draft. LITERATURE CITED ABBOTT, R. T. & S. P. DANCE. 1982. Compendium of Seashells. E. P. Dutton: New York. 411 pp. ALLMON, W.D. 1988. Ecology of Recent turritelline gastropods (Prosobranchia, Turritellidae): current knowledge and pa- leontological implications. Palaios 3:259-284. ALLMON, W. D., J. C. NIEH & R. D. Norris. 1990. Drilling and peeling of turritelline gastropods since the late Creta- ceous. Palaeontology 33:595-611. ANDERSON, T. F. & M. A. ARTHUR. 1983. Stable isotopes of oxygen and carbon and their application to sedimentologic and paleoenvironmental problems. Pp. 1-1-1-151. Jn: M. A. Arthur (organizer), Stable Isotopes in Sedimentary Ge- ology. SEPM Short Course No. 10. Society of Economic Mineralogists and Paleontologists: Dallas, Texas. ARTHUR, M. A., D. F. WILLIAMS & D. S. JONES. 1983. Sea- sonal temperature-salinity changes and thermocline devel- opment in the mid-Atlantic Bight as recorded by the isotopic composition of bivalves. Geology 11:655-659. BaTHAM, E. J. 1969. Benthic ecology of Glory Cove, Stewart Island. Transactions of the Royal Society of New Zealand (Biological Sciences) 11:74-81. BUCHANAN, J. B. 1958. The bottom fauna communities across the continental shelf off Accra, Ghana (Gold Coast). Pro- ceedings of the Zoological Society of London 130:1-56. BUTMAN, C. A. 1987. Larval settlement of soft-sediment in- vertebrates: the spatial scales of pattern explained by active habitat selection and the emerging role of hydrodynamical processes. Oceanography and Marine Biology Annual Re- view 25:113-165. CaADEE, G. C. 1968. Molluscan biocenoses and thanatocenoses in the Ria de Arosa, Galicia, Spain. Zoologische Verhandel- ingen 95:1-121. Carrick, N. 1980. Aspects of the biology of molluscs on the continental shelf off Sydney, NSW with particular reference of the population biology of Gazameda gunni (Reeve, 1849). Unpublished M.S. Thesis, University of Sydney. CHINZEI, K., H. KoIKE, T. OBA, Y. MATSUSHIMA & H. KITAZATO. 1987. Secular changes in the oxygen isotope ratios of mol- W. D. Allmon et al., 1992 lusc shells during the Holocene of central Japan. Palaeo- geography, Palaeoclimatology, Palaeoecology 61:155-166. DuDLEY, E. C. & G. J. VERMEIJ. 1978. Predation in time and space: drilling in the gastropod Turritella. Paleobiology 4: 436-441. EIsMA, D., W. G. Mook & H. A. Das. 1976. Shell charac- teristics, isotopic composition and trace-element contents of some euryhaline molluscs as indicators of paleosalinity. Pa- laeogeography, Palaeoclimatology, Palaeoecology 19:39-62. ERLENKEUSER, H. & G. WEFER. 1981. Seasonal growth of bivalves from Bermuda recorded in their O-18 profiles. Pro- ceedings of the Fourth International Coral Reef Symposium. Manila. Vol. 2:643-648. FLeEssa, K. W. & A. A. EKDALE. 1987. Paleoecology and ta- phonomy of Recent to Pleistocene intertidal deposits, Gulf of California. Pp. 2-33. In: K. Flessa (ed.), Paleoecology and Taphonomy of Recent to Pleistocene intertidal deposits, Gulf of California. Paleontological Society Special Publi- cation No. 2. FRETTER, V. & A. GRAHAM. 1962. British Prosobranch Mol- luscs. Ray Society: London. 755 pp. FursicH, F. T. & K. W. FLEssa. 1987. Taphonomy of tidal flat molluscs in the northern Gulf of California: paleoen- vironmental analysis despite the perils of preservation. Pa- laios 2:543-559. GrossMAN, E. L. & T. L. Ku. 1986. Oxygen and carbon isotopic fractionation in biogenic aragonite: temperature ef- fects. Chemical Geology (Isotope Geoscience Section) 59:59- 74. HERTZ, C.M. 1990. Turritella gonostoma laying eggs in Choya Bay, Sonora, Mexico. The Festivus (San Diego Shell Club) 22(7):87. Jones, D. S. 1980. Annual cycle of shell growth and repro- duction in the bivalves Sprsula solidissima and Arctica islan- dica. Unpublished Ph.D. Dissertation, Princeton University, Princeton, New Jersey. 238 pp. Jones, D.S. 1983. Sclerochronology: reading the record of the molluscan shell. American Scientist 71:384—-391. Jones, D. S. 1988. Sclerochronology and the size versus age problem. Pp. 93-108. In: M. McKinney (ed.), Heterochrony in evolution. A multidisciplinary approach. Plenum Press: New York. Jones, D. S.,M. A. ARTHUR & D. J. ALLARD. 1989. Sclero- chronological records of temperature and growth from shells of Mercenaria mercenaria from Naragansett Bay, Rhode Is- land. Marine Biology 102:225-234. Jones, D.S.,D. F. WILLIAMS & M.A. ARTHUR. 1983. Growth history and ecology of the Atlantic surf clam Spisula solidis- stma (Dillwyn) as revealed by stable isotopes and annual growth increments. Journal of Experimental Marine Biol- ogy and Ecology 73:225-242. KEEN, A. M. 1971. Sea Shells of Tropical West America. Stanford University Press: Stanford, California. 1064 pp. KILLINGLEY, J. S. & W. H. BERGER. 1979. Stable isotopes in a mollusk shell: detection of upwelling events. Science 205: 186-188. KRANTZ, D. E., D. S. Jones & D. F. WILLIAMS. 1984. Growth rates of the sea scallop, Placopecten magellanicus, determined from the '8O/'%O record in shell calcite. Biological Bulletin 167:186-199. Page 63 KRANTZ, D. E., D. F. WILLIAMS & D. S. JONES. 1987. Eco- logical and paleoenvironmental information using stable iso- tope profiles from living and fossil molluscs. Palaeogeogra- phy, Palaeoclimatology, Palaeoecology 58:249-266. LEVINTON, J. S. 1970. The paleoecological significance of op- portunistic species. Lethaia 3:69-78. Marwick, J. 1957. Generic revision of Turritellidae. Pro- ceedings of the Malacological Society of London 32:144- 166. NEWELL, R. C. 1979. Biology of Intertidal Animals. Marine Ecological Surveys Ltd.: Faversham, Kent, U.K. 781 pp. NEWELL, R. C. & L. H. Koroep. 1977. The energetics of suspension-feeding in the gastropod C7repidula fornicata. Journal of the Marine Biological Association of the U.K. 57:161-180. PARKER, R. H. 1964. Zoogeography and ecology of some mac- roinvertebrates, particularly mollusks, in the Gulf of Cali- fornia and the continental slope off Mexico. Videnkabelige Meddelelser fra Dansk Naturhistorisk Forening 126:1-178. RODEN, G. I. & G. W. Groves. 1959. Recent oceanographic investigations in the Gulf of California. Journal of Marine Research 18:10-35. ROMANEK, C. S. & E. L. GROSSMAN. 1989. Stable isotope profiles of Tridacna maxima as environmental indicators. Pa- laios 4:402-413. ROMANEK, C. S., D. S. JoNnEs, D. F. WILLIAMs, D. E. KRANTZ & R. RADTKE. 1987. Stable isotopic investigation of phys- iological and environmental changes recorded in shell car- bonate from the giant clam 77ridacna maxima. Marine Bi- ology 94:385-393. Tanaka, N., M. C. MONAGHAN & D. M. RYE. 1986. Con- tribution of metabolic carbon to mollusc and barnacle shell carbonate. Nature 320:520-523. THOMSON, D. A. 1987. 1987 tide calendar for the northern Gulf of California. University of Arizona: Tucson. TURNER, J. V. 1982. Kinetic fractionation of '°C during cal- cium carbonate precipitation. Geochimica et Cosmochimica Acta 46:1183-1191. VAUGHAN, D. 1983. A study of the ecology of Turritella gon- ostoma living in the Gulf of California, Mexico. Annual Report (for 1982), Western Society of Malacologists 15:13- 14. WeEFER, G. & J. S. KILLINGLEY. 1980. Growth histories of strombid snails from Bermuda recorded in the O-18 and C-13 profiles. Marine Biology 60:129-135. WILLIAMS, D. F., M. A. ArTHuR, D. S. JONES & N. HEALY-WILLIAMS. 1982. Seasonality and mean annual sea surface temperatures from isotopic and sclerochronological records. Nature 296:432-434. WILLIAMS, D. F., M. A. SOMMER & M. L. BENDER. 1977. Carbon isotopic compositions of recent planktonic forami- nifera of the Indian Ocean. Earth and Planetary Science Letters 36:391-403. WRIGHT, C. A. 1956. Studies on the life-history and ecology of the trematode genus Renicola Cohn, 1904. Proceedings of the Zoological Society of London 126:1-49. YONGE, C.M. 1946. On the habits of Turritella communis Risso. Journal of the Marine Biological Association of the U.K. 26:377-380. The Veliger 35(1):64-69 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 Simulathena papuensis, a New Planaxid Genus and Species from the Indo-West Pacific RICHARD S. HOUBRICK Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, USA Abstract. Simulathena papuensis, sp. nov., is monotypic, has a plain, thin littorinid-like shell with a large aperture and wide shallow anterior canal, and is sculptured with weak spiral incised lines. The periostracum is moderately thick and the lenticular operculum with terminal nucleus is typical of planaxids. The taenioglossate radula is very similar to those of three planaxid genera and also resembles that of Fossarus. A large, extensive subhemocoelic brood pouch of ectodermal origin fills the headfoot, extending anteriorly into the head. A pair of mantle papillae extends from the exhalant siphon. The osphradium is widely separated from the broad, shallow ctenidium. Brooded embryos with large shells hatch from the brood pouch as juvenile snails. Simulathena is the sister group of the Hinea, Holcostoma, Supplanaxis clade, which also includes the Fossarinae. INTRODUCTION Study of an unidentified, marine intertidal prosobranch from Yule Island, Papua New Guinea, has revealed a new genus and species of the planaxid group. The family Plan- axidae Gray, 1850, a relatively small group of taxa, was reviewed by HOUBRICK (1987) and was thought to com- prise six genera and about 20 species throughout the world. HouprIick (1990) subsequently allocated Fossarus Philip- pi, 1841, under the subfamily Fossarinae Troschel, 1861, to the Planaxidae, thereby expanding the family. It is therefore surprising to discover yet another genus and spe- cies to add to the family. The new taxon, lacking the thick shell characteristic of so many other planaxid taxa, has a generalized littorinid shell shape and a squat, rather smooth and thin shell. Radular, opercular, and anatomical inves- tigations indicate that the new taxon shares a number of features with other planaxid taxa. A description and dis- cussion follow. MATERIALS anD METHODS The examined specimens came from the Kanudi Marine Laboratory and were sent by the Papua New Guinea Department of Fisheries to the Australian Museum, Syd- ney, for identification. Five preserved specimens, which constitute the type lot, were studied and two of these dis- sected under a Wild M-5 dissecting microscope for ana- tomical study. One of the specimens was broken and was not used for shell measurements. Although preservation was not optimal, sufficient anatomical features were avail- able for rudimentary character analysis and a description of gross anatomy. However, several important systems, in particular the pallial oviducts, could not be described. Em- bryonic shells and the radula were studied using an Hitachi S-570 scanning electron microscope. RESULTS Simulathena Houbrick, gen. nov. Diagnosis: Shell thin, squat, having inflated whorls sculp- tured with weak spirally incised lines, large, fat body whorl, smooth outer lip and very weak, shallow anterior canal. Operculum lenticular with subterminal nucleus. Radula having triangular rachidian tooth with pair of cusps high on basal plate, beneath cutting edge cusps, and lateral tooth with very wide, long lateral extensions of basal plate. Pair of mantle tentacles emerge from exhalant siphon. Large subhemocoelic cephalic brood pouch containing embryos having direct development, emerging as small snails. Etymology: “like Athena,” a Latin combination of simu, meaning “like,” and Athena, the Greek goddess who sprang forth from the head of Zeus, in reference to direct devel- opment in the cephalic brood pouch. Remarks: This genus appears to be monotypic and differs from other planaxid genera in having a light, thin shell. R. S. Houbrick, 1992 The pair of mantle tentacles emerging from the exhalant siphon is distinctive and does not occur in other members of the family. The large embryos and extensive brood pouch extending forward into the head are also unusual. It is unfortunate that only four specimens were available for study and that their anatomy is only partially known, but it is clear from what has been examined that this species represents an undescribed genus. Simulathena papuensis Houbrick, sp. nov. (Figures 1-13) Description: Shell (Figures 1-3, Table 1): The shell is thin, squat, and globose, comprising about five inflated whorls weakly sculptured with incised spiral lines. The protoconch and embryonic whorls (Figures 4-6) are un- sculptured, having a smooth gradual transition into the incised spiral lines of the adult shell. The adult whorls, exclusive of the body whorl, each have 4 spiral incised lines and numerous microscopic and weak axial striae. The suture is distinct and slightly sunken into each suc- cessive whorl. The body whorl is very large, comprising over 75% of the shell length and is sculptured with about 14 spiral incised lines. The aperture is ovate and large, nearly two-thirds the shell length, and has a concave col- umella with a weak callus and a smooth edged, rounded outer lip, slightly pointed at the shell base. The anterior canal is merely a slight shallow depression in the basal part of the peristome adjacent to the columella. A well- developed, tan-to-olive colored periostracum covers the shell, and also occurs on embryonic shells. Under the periostra- cum, the shell is white with 3 broad light-brown spiral bands. The operculum (Figures 7, 8) is corneous, brown, and lenticular, having a terminal nucleus and many growth lines. The attachment scar is elongate and narrow (Figure 8). Anatomy: The animal (Figure 9) is tightly coiled, having a very large body whorl, which comprises the mantle cav- ity, the pericardial cavity, and part of the kidney. The visceral whorls comprise the large stomach, digestive gland, and gonad. The headfoot is large and muscular with a broad snout (Figure 9, sn), which is enlarged and bilobed at the tip and has short cephalic tentacles, each with a small black eye on the outer side of the tentacular base. A conspicuous birth pore (Figure 9, bp) lies in the right side of the neck in females. The thick columellar muscle (Figure 9, cm) is short but broad, enveloping the ventral side of the body whorl. The large muscular foot is broad and thick, and has a long anterior mucus gland. The mantle skirt is wide and its ventral and dorsal edges are smooth, except for some small papillae (Figure 9, mp) at the in- halant siphon (Figure 9, inh). Emerging from the exhalant siphon are two enlarged, tentaculate papillae (Figure 9, exp) attached to the inner surface of the mantle skirt. The Page 65 Table 1 Shell measurements (mm) and meristics of the holotype (*) and three paratypes of Simulathena papuensis. Length 15.8* 14.1 13.8 13.0 Width 9.4* 8.7 D5 Well Aperture length 9.8* 8.6 9.8 8.2 Aperture width 6.2* 529 6.2 5:3 Number whorls 5% 5 5 5 shells of developing embryos (Figure 9, emb) can be seen through the swollen, thinly stretched cephalic epithelium covering the head of brooding females. The mantle cavity is spacious and extends the depth of the first whorl. A ridegelike osphradium flanked with glan- dular strips on each side is separated from the ctenidium by a wide area of mantle epithelium. The ctenidium (Fig- ure 9, ct) is long and very wide, and has relatively shallow filaments, each comprising a free, fingerlike leading edge and a very broad, shallow attached base. The fingerlike edge is one-half the base length. The rectum (Figure 9, r) is very wide, a little over one-half the ctenidial width, thin- walled, and filled with long, cylindrical fecal pellets stacked in parallel rows and composed of fine sand. The snout tip has paired fleshy lobes (Figure 9, 1) bor- dering a slitlike mouth. A pair of chitinous jaws lie on the inside of the opening to the oral cavity. The buccal mass is small. The radular ribbon (Figure 10) is broad and short, about 4 mm in length. The rachidian tooth (Figure 12) is triangular shaped, having a flat basal plate with a basal central projection and a pair of basal denticles, each located under and adjacent to the lateral edges of the cutting edge of the tooth, which has a large blunt central cusp flanked on each side by two or three pointed denticles. The lateral tooth (Figures 11-13) is bobomerang-shaped, having a basal plate comprising an inner longitudinal pillar and a very long, flat lateral extension, and a cutting edge with a broad, rounded major cusp flanked with an inner pointed denticle and three or four outer pointed denticles. The marginal teeth (Figure 13) have spatulate shafts and broad, sharply folded tips, each with five rounded denticles. The stomach is a large wide organ occupying over one-half the penultimate whorl, and comprising a short style sac and gastric shield. In females, the subhemocoelic brood pouch consists of invaginated, ciliated, ectodermal epithelium that begins at the brood pore (Figure 9, bp) and extends into the neck and head anteriorly to the tentacular lobes and posteriorly and ventrally throughout the headfoot as far back as the posterior mantle cavity. The brood pouch is completely separate from the cephalic hemocoel, but surrounds the nerve ring and mid-esophagus. Internal folding of the inner epithelium of the brood pouch creates many loculae, each accommodating a developing embryo and communicating Page 66 The Veliger, Volh35 Nom Explanation of Figures 1 to 8 Figures 1-8. Simulathena papuensis, sp. nov. Figures 1-3. Holotype AMS C166326, 15.9 x 9.5 mm, Kairuku, Yule Id., Central District, Papua New Guinea. Figures 4-6. Embryonic shells removed from brood pouch showing protoconch, early whorl sculpture, and periostracum (bar = 0.75 mm). Figures 7 and 8. Operculum, 6.7 mm length, showing free (7) and attached (8) sides. with a common, nearly closed lumen. The brood pouch contains 55-60 embryos, most of which are large and of equal size, but smaller embryos are also randomly dis- tributed throughout the pouch. Before hatching, the em- bryos have eyes and an operculum and their shells (Figures 4-6) are in an advanced state, attaining 2.25 mm length and having pigment patterns similar to the adult shell and a well-developed periostracum. R. S. Houbrick, 1992 Holotype: AMS C166326, length 15.9 mm, width 9.5 mm. Paratypes: USNM 859456 (two specimens); length 14.7 mm, width 8.7 mm, and length 13.1 mm, width 9.5 mm. Type locality: Kairuku, Yule Id., Central District, Papua New Guinea (8°50’S, 146°32’E). Etymology: Named after Papua, where this species was found. DISCUSSION The plain, relatively thin shell of Simulathena papuensis superficially resembles those of some thiarid and viviparid freshwater snails, but the anatomy differs substantially from that of the viviparids. My initial impression was that this species was an undescribed thiarid; however, close examination of the shell, operculum, and radula indicate that Simulathena papuensis has more in common with members of the Planaxidae than with thiarid species. In addition, several anatomical characters point to the Plan- axidae as the proper familial assignment. Shell: The shell sculpture of incised spiral lines and the wide, very shallow anterior canal are common concholog- ical characters of other species of Planaxis Lamarck, 1822, Supplanaxis Thiele, 1929, and Holcostoma Adams & Ad- ams, 1853, and a lenticular rather than ovate operculum (Figures 7, 8) is typical of all planaxids (see HOUBRICK, 1987). The shell shape of Simulathena, although unique among planaxids, is most like that of Holcostoma species. The periostracum, while relatively thick, is not hispid as in many other planaxid species (see HOUBRICK, 1987). Anatomy: Simulathena papuensis has a number of char- acters in common with planaxids. The expanded bilobed snout tip is similar to those seen on planaxid species. The radula is closest to those described for species of Supplan- axis, Hinea Gray, 1847, Holcostoma, and Fossarus (see Housrick, 1987, 1990). The broad, shallow, ctenidial filaments are similar to those described in Planaxis sulcatus (Housrick, 1987:36). The pallial gonoducts are open in contrast to the closed systems found in brooding, parthe- nogenetic thiarids. It was not determined if Simulathena is gonochoristic, but this condition is assumed until proven otherwise. Although the two specimens of Simulathena I examined were not preserved well enough to determine the precise pallial oviduct configuration, the large subhe- mocoelic cephalic brood pouch and the birth pore on the right side of the neck are much like those seen in all other examined planaxid genera (see HOUBRICK, 1987), includ- ing Fossarus, subfamily Fossariinae (see HOUBRICK, 1990). Similar cephalic brood pouches also occur in many par- thenogenetic thiarids (MORRISON, 1954), which are prob- ably a sister group of Planaxidae (see HOuBRICK, 1988). As mentioned previously, the pair of mantle tentacles projecting from the exhalant siphon is unusual and not Page 67 pod cm Figure 9 External anatomical features of Simulathena papuensis, sp. nov., showing embryos beneath the thin cephalic epithelum of the anterior brood pouch. Legend: bp, brood pore; cm, columellar muscle; ct, ctenidium; emb, embryo inside brood pouch; exp, exhalant papillae; inh, inhalant siphon; k, kidney; 1, lobe forming lip of mouth; mp, mantle papillae; op, operculum; pod, pallial oviduct; r, rectum containing fecal pellets; sn, snout. seen among other planaxids. The wide separation between the osphradium and ctenidium is also a unique feature of Simulathena. Reproductive biology: The most striking external ana- tomical feature of Simulathena papuensis is the enlarged brood pouch extending forward into the head (Figure 9). The brood pouch has a ciliated epithelium and is formed by an ectodermal invagination, as described in other plan- axids (see HOuBRICK, 1987). It engulfs much of the ce- phalic hemocoel, and comprises numerous, thin-walled loc- ulae, all of which communicate with the lumen of the pouch and with each other. Each locula contains or enfolds a single uncapsulated embryo. Embryos presumably enter the brood pouch as fertilized, uncapsulated eggs, and are probably fed on nutritive liquids secreted by the brood pouch walls. They grow to be quite large (up to 2.25 mm in length), presumably emerging through the brood pore as small snails. Most embryos are very large and it is difficult to imagine how they can maneuver through the interstices of the brood pouch and emerge from the small brood pore. It is possible, but unlikely, that they rupture through the thin epithelium of the head and dorsal surface of the neck when they are ready to hatch, although it is unclear what this would do to other, less-developed em- bryos. Most embryos are roughly the same size, but a number of smaller shelled embryos are present, suggesting that different cohorts are being brooded or that some em- Page 68 The Veliger, Vol. 35, No. 1 Explanation of Figures 10 to 13 Figures 10-13. Scanning electron micrographs of various aspects of the radula of Simulathena papuensis, sp. nov. Figure 10, bar = 175 wm; Figure 11, bar = 65 um; Figure 12, bar = 25 wm; Figure 13, bar = 43 um. bryos do not receive the same nourishment from the brood pouch as do others. It is unlikely that the smaller embryos serve as a food source for larger embryos, because they have well-developed shells. The only other planaxid known to brood very large embryos that hatch out as young snails is the Persian Gulf population of Planaxis sulcatus (see ‘THORSON, 1940; BARKATI & AHMED, 1982; HOUBRICK, 1987), in which nurse eggs have been documented. The large size of the enclosed embryonic snails in St- mulathena papuensis is unusual among most planaxids, but very large embryos are not uncommon among parthe- nogenetic thiarids (Houbrick, personal observations). R. S. Houbrick, 1992 Ecology: Nothing has been recorded about the microhab- itat of this species except that it lives in the intertidal zone. Geographic distribution: Simulathena papuensis is known only from the type locality, but probably occurs in other suitable habitats in New Guinea. Phylogeny: The original phylogeny of the Planaxidae presented by Housrick (1987:fig. 27) is outdated. Angiola Dall, 1926, is now regarded as a synonym of Hinea, be- cause Hinea has been found to exhibit bioluminescence (PONDER, 1988), which was the only character separating the two genera. A preliminary updated phylogenetic analysis of the Planaxidae (18 characters, 9 taxa, consistency index = 74) was run, using many of the same characters originally employed by HousRick (1987:48-50) and some modified, revised ones. The revised analysis was done with the Hennig86 algorithm and included new taxa, the Fossar- inae, and used 7hzara Roding, 1798, as the outgroup. The results suggest that Simulathena is the sister group of the clade comprising Hinea, Holcostoma, Supplanaxis, and Fos- sarus. This is readily seen by the similarity among the radulae of these taxa. However, final resolution of plan- axid phylogeny awaits a more formal cladistic analysis involving many other cerithioidean taxa and a more com- plete reappraisal of the characters. ACKNOWLEDGMENTS I thank Mr. Phil Coleman, of the Australian Museum, Sydney, for sending me the specimens for examination. I am grateful to the staff of the Smithsonian Scanning Elec- tron Microscope facility for assisting me and for providing photomicrographs. I also thank Mr. Victor Krantz, of the Smithsonian’s Photographic Services for photography. Dr. Robert Hershler provided critical comments about the manuscript. Page 69 LITERATURE CITED BARKATI, S. & M. AHMED. 1982. Studies on the reproductive biology of some prosobranchs from the coast of Karachi (Pakistan) bordering the northern Arabian Sea, 1: Obser- vations on Planaxis sulcatus (Born, 1780). The Veliger 24(4): 355-358. Da._, W. H. 1926. New shells from Japan and the Loochoo Islands. Proceedings of the Biological Society of Washington 39:63-60. Gray, J. E. 1850. Systematic arrangement of the figures. Sys- tem of Mollusca. Richard & John E. Taylor: London. Pp. 63-124. Housrick, R. S. 1987. Anatomy, reproductive biology, and phylogeny of the Planaxidae (Cerithiacea: Prosobranchia). Smithsonian Contributions to Zoology No. 445:111 + 57 pp., 27 figs. Housrick, R. S. 1988. Cerithioidean phylogeny. Jn: W. F. Ponder (ed.), Prosobranch Phylogeny. Malacological Re- view, Supplement 4:88-128. Housrick, R. S. 1990. Anatomy, reproductive biology and systematic position of Fossarus ambiguus (Linné) (Fossarinae: Planaxidae; Prosobranchia) Acgoreana, 1990, Supplement: 59-73. Morrison, J. P. E. 1954. The relationships of old and new world melanians. Proceedings of the United States National Museum. 103(3325):357-394, 2 pls. PuHiuippl, R. A. 1841. Zoologische Bemerkungen. Fossarus, ein neues Genus der Kammkiemigen Mollusken. Archiv fur Naturgeschischte 7(1):42-49, pl. 5. PONDER, W.F. 1988. Bioluminescence in Hinea braziliana (La- marck) (Gastropoda: Planaxidae). Journal of Molluscan Studies 54:361. RODING, P. F. 1798. Museum Boltenianum. II. Conchylia. Trappii: Hamburg. 199 pp. TuHorSON, G. 1940. Studies on the egg masses and larval de- velopment of Gastropoda from the Iranian Gulf. Danish Scientific Investigations in Iran, Part 2, Copenhagen. 238 PP- TROSCHEL, F. H. 1856-1863. Das Gebiss der Schnecken zur Begrundung einer Naturlichen Classification, Volume 1: 661. Berlin. The Veliger 35(1):70-73 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 ‘Two New Vitrinellid Species from the Gulf of California, Mexico (Gastropoda: Vitrinellidae) by CAROLE M. HERTZ, BARBARA W. MYERS anp JOYCE GEMMELL Associates, Department of Marine Invertebrates, San Diego Natural History Museum, Balboa Park, P.O. Box 1390, San Diego, California 92112, USA Abstract. Cochliolepis cornis, sp. nov., is described from San Felipe, Baja California, Mexico. This is the first report of the genus in the tropical eastern Pacific. It is compared with C. parasitica Stimpson, 1858, type of the genus, and C’ striata Dall, 1889, both from the tropical western Atlantic. Cyclostremiscus salvatierrensis, sp. nov., is also described from the northern Gulf of California; it ranges south to Costa Rica. It is compared with the tropical eastern Pacific species Cyclostremiscus nummus and Cyclostremiscus major, both of Pilsbry & Olsson, 1952. INTRODUCTION MYERS et al. (1989) gave a preliminary list of the Vitri- nellidae collected by Joyce Gemmell from 1965 to 1976 from San Felipe to San Luis Gonzaga in the northwestern Gulf of California, Mexico. In that report, 25 species were indicated, although five were not identified. Further study of this material has revealed that four of the species were new; two are described in this paper. The other two species have been described by Myers et al. (1991). Other work on the Vitrinellidae from this collection was done by GEM- MELL et al. (1989) and MYERs ef al. (1990). Living spec- imens in this family are seldom collected, and nearly all species descriptions must necessarily be based on shell characters alone. Institutional abbreviations are as follows: LACM, Los Angeles County Museum of Natural History; SBMNH, Santa Barbara Museum of Natural History; SONHM, San Diego Natural History Museum; USNM, National Museum of Natural History, Smithsonian Institution. VITRINELLIDAE Bush, 1897 Cochliolepis Stimpson, 1858 Type species: C. parasitica Stimpson, 1858, by original designation. Cochliolepis cornis Hertz, Myers & Gemmell, sp. nov. (Figures 1—4) Description: Holotype 4.2 mm diameter, 1.4 mm altitude, thin, white, discoidal, flattened. Almost 2 protoconch whorls, 2% teleoconch whorls. Spire low, projecting, tipped lat- erally from aperture. Suture well defined, excavated. Whorls rapidly increasing in diameter and coiled like a ram’s horn. Umbilicus broadly open, narrowing to apex, revealing rounded surface of whorls. Aperture entire, asymmetric, with very faint sulcus posteriorly, slightly pro- duced at periphery; inner lip reflected, appressed to pre- vious whorl. Under magnification entire surface covered by closely spaced spiral threads with irregularly spaced fine axial growth lines. Etymology: The name is Latin, “bearing horns,” sug- gested by the ram’s-horn coiling of the shell. Type locality: Bahia San Felipe, Baja California, Mexico (31°03'N, 114°48’W), in drift. Type material: 36 specimens from type locality, collected by Joyce Gemmell from 1965 to 1976. Holotype: SDNHM 93511. Paratypes: SDNHM 93512, 14 specimens; LACM 2576, 6 specimens; USNM, 8 specimens; SBMNH 35169, 8 specimens. One paratype, dredged by Joyce Gemmell on shrimpboat Chamizal on clay bottom in 25 m just south of Puertecitos, Baja California, Mexico (30°18'30’N, 114°37'24”W), 8-10 July 1969, retained in Hertz collec- tion. Discussion: We have placed this new species in the genus Cochliolepis based on the rapidly increasing diameter of the whorls, coiled like a ram’s horn, and the umbilicus open to the apex (see GARDNER, 1948; ABBOTT, 1974). GARDNER (1948) redefined the genus Cochliolepis and em- phasized the “‘ram’s-horn coiling” and wide umbilical fun- C. M. Hertz e al., 1992 Explanation of Figures 1 to 3 Figures 1-3. Cochliolepis cornis, sp. nov. Holotype, SDNHM 93511. Diameter 4.2 mm, altitude 1.4 mm. Bahia San Felipe, Baja California, Mexico, in drift. Figure 1. Dorsal view. Figure 2. Basal view. Figure 3. Apertural view. nel as diagnostic generic characters. This is the first report of the genus in the eastern Pacific. Cochliolepis cornis resembles the type species, C. par- asitica, in having rapidly expanding whorls and similar size, but C. cornis differs in having the surface covered by closely spaced spiral threads, whereas C. parasitica is smooth except for occasional strong growth lines. Cochliolepis cor- Page 71 Figure 4 Cochholepis cornis, sp. nov. Holotype. Camera lucida drawing showing laterally tipped spire and appressed inner lip. nis has a projecting, tilted protoconch, whereas C. parasitica has a depressed protoconch surrounded by a thin, trans- parent shell layer that extends above the suture of the body whorl, forming what looks like a double suture. Cochlholepis cornis also resembles C. striata Dall, 1889, in shape and its sculpture of closely spaced spiral cords. Cochliolepis cornis differs from C. striata in its smaller size (C. striata attains approximately 6.6 mm diameter) and in having a projecting, tilted protoconch. The protoconch of C. striata is partially covered by the succeeding whorl and the body whorl coils slightly above the periphery. In C. cornis, the body whorl terminates at the periphery. Cochli- olepis corns differs further in having a broadly open um- bilicus instead of the constricted umbilicus of C. striata. Cyclostremiscus Pilsbry & Olsson, 1945 Type species: Vitrinella panamensis C. B. Adams, 1852, by original designation. Cyclostremiscus salvatierrensis Hertz, Myers & Gemmell sp. nov. (Figures 5-8) Description: Holotype 1 mm maximum diameter; pro- toconch of approximately ¥2 whorl, slightly projecting; 2% glassy, somewhat flattened teleoconch whorls; suture deep- ly impressed. Aperture quadrate, with thick parietal callus extending beyond plane of peristome and containing a moderately prominent anal sulcus; lip edge raised and reflected dorsally, slightly produced at termination of first dorsal spiral cord. Body whorl with prominent peripheral cord and three strong, rounded spiral cords above; inter- spaces narrow; remainder of dorsum smooth. Base having two strong spiral cords, with narrow interspaces below and adjacent to periphery; remainder of base smooth, some- what rounded, terminating in a moderately open funnel- shaped umbilicus. Axial sculpture lacking except for a few irregularly spaced growth lines. Etymology: The name derives from Isla Salvatierra, the type locality. Type locality: One km west of Isla Salvatierra (also known as Isla San Luis), Baja California, Gulf of California, Page 72 Explanation of Figures 5 to 7 Figures 5-7. Cyclostremiscus salvatierrensis, sp. nov. Holotype, SDNHM 93513. Diameter 1.0 mm, altitude 0.5 mm. West of Isla Salvatierra, Gulf of California, Mexico. Figure 5. Dorsal view. Figure 6. Basal view. Figure 7. Apertural view. Mexico (29°57'48"N, 114°28’W), in sand with scallop valves in 25 m. Type material: 2 specimens from type locality, collected by Joyce Gemmell on 8-10 July 1969. Holotype: SONHM 93513; paratype: USNM, 1 specimen. Nine paratype lots in LACM as follows: LACM 2581 (ex LACM-AHF 237), The Veliger, Vol. 35, No. 1 Figure 8 Cyclostremiscus salvatierrensis, sp. nov. Holotype. Camera lucida drawing showing the quadrate aperture with thick parietal callus. 1 specimen, Bahia Magdalena, Baja California Sur, Mex- ico (24°19'15”N, 110°37'30”W); LACM 2585 (ex LACM 66-28), 1 specimen, Bahia Partida, Baja California Sur, Mexico (24°25'N, 110°25’W); LACM 2578 (ex LACM 66-22), 1 specimen, Bahia Muertos, Baja California Sur, Mexico (24°55'N, 109°46’W), in 18-55 m; LACM 2583 (ex LACM 66-21), 3 specimens, off Punta Arena, Baja California Sur, Mexico (23°32'N, 109°28'’W) in 18-36 m; LACM 2584 (ex LACM 71-22), 5 specimens, south of Punta Arena (2 km south of Los Tezos Ranch) (23°31'N, 109°W) in 9 m; LACM 2579 (ex LACM 66-19), 1 spec- imen, Bahia Pulmo, Baja California Sur, Mexico (23°22’N, 109°25'W) in 1-6 m; LACM 2580 (ex LACM 66-20), 1 specimen, south end of Bahia Pulmo, under boat anchorage (23°22'N, 109°25'W) in 6 m; LACM 2577 (ex LACM 68-45), 2 specimens, Bahia Cuastocomate, Jalisco, Mexico (19°13'45"N, 104°44'53”W) in 18-36 m; LACM 2582 (ex LACM-AHF 116-33), 1 broken specimen, south of Puerto Culebra, Costa Rica (10°33'35”N, 85°42'30”W) in 4 m. Distribution: The new species is known from scattered records from Isla Salvatierra in the northern Gulf of Cal- ifornia, Mexico, to Puerto Culebra, Costa Rica. Discussion: Cyclostremiscus salvatierrensis is closest to the eastern Pacific species C. nummus and C. major, both of Pilsbry & Olsson, 1952. Cyclostremiscus salvatierrensis has a somewhat flattened shell with a prominent peripheral cord, whereas C. nummus has a tricarinate shape. Cyclos- tremiscus salvatierrensis differs from C. nummus in having three cords immediately above the periphery on the body whorl and two immediately below, with the remaining surface smooth, whereas C. nummus has spiral cords over the entire surface, three cords being stronger. Cyclostremiscus salvatierrensis differs from C. major in its much smaller size, C. major attaining a diameter of 10.9 mm. Cyclostremiscus salvatierrensis has six spiral cords whereas C. major has spiral cords over the entire shell surface. ACKNOWLEDGMENTS We are grateful to James H. McLean and Gale Sphon of the Los Angeles County Museum of Natural History for the loan of specimens. Carol Skoglund of Phoenix, Ari- zona, lent comparative material, and David K. Mulliner photographed the holotypes of the new species. We thank C. M. Hertz et al., 1992 the San Diego Natural History Museum for the use of the collection, library, and facilities in the Department of Marine Invertebrates. James H. McLean kindly reviewed the paper and gave many helpful suggestions. LITERATURE CITED ABBOTT, R. T. 1974. American Seashells. 2nd ed. Van Nos- trand Reinhold Co.: New York. 663 pp. ADAMS, C. B. 1852. Catalogue of shells collected at Panama, with notes on their synonymy, station, and geographical distribution. R. Craighead: New York. vii + 334 pp. Busu, K. J. 1897. Revision of the marine gastropods referred to Cyclostrema, Adeorbis, Vitrinella and related genera with descriptions of some new genera and species belonging to the Atlantic fauna of America. Transactions of the Con- necticut Academy 10:97-144. DALL, W. H. 1889. A preliminary catalogue of the shell-bear- ing marine mollusks and brachiopods of the southeastern coast of the United States, with illustrations of the species. Bulletin of the United States National Museum 37:1-232. GARDNER, J. 1948. Mollusca from the Miocene and Lower Pliocene of Virginia and North Carolina. Pt. 2. Scaphopoda and Gastropoda. United States Geological Survey Profes- sional Paper 199B:179-310. Page 73 GEMMELL, J., C. M. HERTZ & B. W. Myers. 1989. Cyclos- tremiscus nodosus of Pilsbry & Olsson, 1952. The Festivus 21(5):42-45. Myers, B. W., C. M. Hertz & J. GEMMELL. 1989. Prelim- inary report on the Vitrinellidae from San Felipe, Baja Cal- ifornia, Mexico, from the Gemmell collection (1965-1976). Annual Report of the Western Society of Malacology [for 1988] 21:17 [Abstract]. Myers, B. W., C. M. HERTZ & J. GEMMELL. 1990. A dis- cussion of Episcynia bolivarr Pilsbry & Olsson, 1946, and related eastern Pacific species. The Festivus 22(2):14-20. Myers, B. W., C. M. HERTZ & J. GEMMELL. 1991. Two new species of Vitrinellidae from the Gulf of California, Mexico (Gastropoda). Venus, The Japanese Journal of Malacology 50(1):30-36. Piussry, H. A. & A. A. OLSSON. 1945. Vitrinellidae and similar gastropods of the Panamic Province. Part I. Proceedings of the Academy of Natural Sciences of Philadelphia 97:249- 278. Pitssry, H. A. & A. A. OLSSON. 1952. Vitrinellidae of the Panamic Province. Part II. Proceedings of the Academy of Natural Sciences of Philadelphia 104:35-88. Stimpson, W. 1858. A new form of gasteropodous Mollusca, Cochliolepis parasiticus. Proceedings of the Boston Society of Natural History 6:307-309. The Veliger 35(1):74-77 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 Two Giant African Land Snail Species Spread to Martinique, French West Indies ALBERT R. MEAD Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, Arizona 85721, USA LOUIS PALCY Chef, Service Protection des Végétaux Martinique, 97257 Fort de France, Martinique, French West Indies Abstract. The American tropics, long known to be vulnerable to the depredations of the giant African land snails, have recently been penetrated by Achatina fulica, endemic in East Africa, and Limicolaria aurora, endemic in West Africa. Both species can produce extraordinarily heavy infestations, cause serious damage to crops, carry agents of human diseases, and aggressively compete with or replace endemic snail species. Concerted efforts are being made to contain the infestations. For over 40 years, the giant African land snail, Achatina fulica Bowdich, 1822 (normally 9-15 cm in length) has been documented as the world’s largest, most economically important snail pest in agriculture and horticulture (MEAD, 1961). Further, it is known to be a vector of two disease agents in humans: the rat lungworm Anguostrongylus can- tonensis, causing eosinophilic meningoencephalitis (WAL- LACE & ROSEN, 1969; Asst ADOU et al. 1980; PERERA et al., 1983; PEREZ MARTIN ef al., 1984; ANDERSEN et al., 1986), and a gram-negative bacterium, Aeromonas hydro- phila, causing a remarkably wide range of symptoms, par- ticularly in persons on immunosuppressant drugs (DEAN et al., 1970). In prehistoric times, specimens of this species were car- ried from their native home in East Africa to Madagascar, either accidentally or for food. Nearly 200 yr ago, they were introduced from there to the small island groups in the Indian Ocean for medicinal and other purposes. And from there naturalists introduced them to India and Cey- lon. By the 1930s they had spread rapidly through tropical and subtropical East Asia. World War II and the postwar activities spread the infestations to many Pacific islands and interceptions were made in Australian, Canadian, and U.S. ports, including those in Hawaii. It is enigmatic that during all this time, Achatina fulica is still not believed to have become established anywhere in Central or South America, with their wide array of apparently ideal envi- ronments. Now, however, we are seeing the first penetra- tion of the American tropics by this and one other species of the family Achatinidae. In 1984, Achatina fulica (Figure 1) was found established in Guadeloupe, French West Indies, in Cascade aux Ecrevisses, a scenic area in the rain forest of Basse Terre, the volcanic part of the island (FRANKIEL, 1989). Other than the fact that the site is a tourist attraction, there is no indication how, why, or exactly when the species got there. It is most probable that someone who came into possession of one or more specimens in travel, through the mails, or in delivered produce decided the site was suitable for abandonment. The first extension of this infestation was found in 1987 in Sainte Anne on Grande Terre, the calcareous part of the island. Soon after, the snails spread to a number of sites in both parts of the island. A docu- mentary television film was prepared by the local author- ities to alert the people to the problem. In August 1988, this film was aired in Martinique, about 200 km south of Guadeloupe. Responding to this, a teacher in Morne Rouge, a village at the foot of Mt. Pelée at the north end of the island, notified the authorities that Achatina fulica was in a local banana plantation. In- A. R. Mead & L. Palcy, 1992 an Figure 1 Nearly full-grown specimen of Achatina fulica from Morne Rouge, Martinique, FWI, about 12 cm shell length, 7 whorls, black body. spection confirmed the infestation (SCHOTMAN, 1990), but no discernible damage was observed in the plantations or in the cultivated pastures. According to the local people, the first specimens were believed to have been brought to Martinique in July 1988, by a family from an infested area in Les Abymes, Guadeloupe. A television documen- tary film showing this new infestation on Martinique was prepared at the request of the junior author, who urged the people in January 1989 to report any other possible infestation site. Such a site was reported by the directress of a day nursery in Saint Esprit, a village in the south part of the island. The snails, however, appeared different. Specimens were sent to Simon Tillier of the Museum National d’ Histoire Naturelle (Paris) and were forwarded to the senior author. On the basis of the soft anatomy, these proved to be Limicolaria aurora (Jay, 1839), a com- mon, small (4.5-6 cm in length), spotted or completely pale West African species found from Guinea to Gabon (Figure 2). The infestation appears to be limited to about 500 m along the river Saint Esprit, involving about 12 hectares. In contrast to the experience with Achatina fulica in Morne Rouge, at the start of the rainy season in August 1989, L. aurora appeared in considerable numbers, at- tacking yam, bean, pepper, Jerusalem artichoke, cucum- ber, okra, and sweet potato. Inquiries suggest these snails were introduced some time after 1986 by Martinicans who lived in Africa and probably consumed them. This infes- tation is the first record for Limicolaria outside Africa and its coastal islands. In Cameroon, this species causes damage to palm fruits and leguminous cover crops (SPENCE, 1938). Just as soon as this snail was properly identified, the Secretariat of the Caribbean Plant Protection Commission (CPPC) was notified. This ageny in turn notified all mem- ber countries. In addition, at the ninth session of the CPPC in October 1989, in Trinidad and Tobago, the junior au- thor reminded the conferees that Achatina fulica is in Gua- deloupe and that both A. fulica and Limicolaria aurora are in Martinique. Assistance was formally requested by the Federation Departementale des Groupements de Defense Contre les Ennemis des Cultures de la Martinique. Be- cause Morne Rouge is a valuable center for commercial cut-flowers destined for U.S. markets, there is great con- cern that the trade may be in jeopardy of U.S. plant quar- antines. Hence, the Service de la Protection des Végétaux has vigorously pursued a plan of attack for the eradication of the small population of A. fulica of Morne Rouge. The program calls for (1) exhaustive search-and-destruction in the infested area, (2) intensive surveillance in the contam- Page 76 The Veliger, Vol. 35, No. 1 Figure 2 Full-grown specimen of Limicolaria aurora from Saint Esprit, Martinique, FWI, about 6 cm shell length, 9 whorls, striped yellow-tan body. inated, intermediate security, and uncontaminated zones, and (3) molluscicidal attack through the use of metalde- hyde, thiodicarb, and mercaptodimethur. The same strat- egy has been established for L. aurora in Saint Esprit. Biological control through the use of the predatory snails Euglandina rosea (Ferussac) and Gonaxis quadrilateralis (Preston) has not received serious consideration because of the sad experiences of having these and other predators destroy to extinction zoogeographically important endemic snails in Hawaii, ‘Tahiti, New Caledonia, and other Pacific island groups (VAN DER SCHALIE, 1969; HART, 1978; HADFIELD & Kay, 1981; CLARKE et al., 1984; POINTIER & BLANC, 1985). Although both A. fulica and L. aurora are choice items for human consumption in Africa (PILSBRY, 1919; CANSDALE, 1940) and are safe to eat if they are adequately cooked, the authorities have not attempted bi- ological control through encouraging the local people to consume them because it is widely known among the peo- ple that snails are carriers of diseases. Further, counter to the control program, collecting the snails and taking them home to be eaten could enhance dissemination in the island and invite their spread to other islands. The successful establishment of these two pestiferous snails in the Caribbean belatedly demonstrates the pre- dicted vulnerability of this area (MEAD, 1973). It is certain that unless these infestations are contained, further acci- dental and incidental spreading, like that from Guadeloupe to Martinique, will continue to take place, eventually ex- tending to tropical and subtropical Central and South America. Records show the snails have a long list of ac- ceptable food plants. New ones will be added. It is not yet known whether the snails in these new infestations are carrying the rat lungworm Angiostrongylus cantonensis; but this parasite has been reported from West Africa, the Indo- Pacific, and recently Cuba and Puerto Rico (WALLACE & ROSEN, 1969; Assi ADOU et al. 1980; PEREZ MARTIN et al., 1984; ANDERSEN et al., 1986). Therefore, from the public-health standpoint, it should be assumed to be pres- ent in the Martinique and Guadeloupe snail populations. That these two snail pests could serve as novel intermediate hosts for some endemic and introduced parasites is likely. Beyond attacking plants and carrying diseases, the sheer numbers of individuals in their mature populations provide a serious nuisance to the human population. As vigorous, aggressive species, they will provide competition to and even replacement of endemic species. Paradoxically, the smaller Limicolaria aurora may prove eventually to be the worse pest. With the probable exceptions of California and Florida, no region in the world has had greater, more diversified A. R. Mead & L. Palcy, 1992 experience with plant quarantines and control measures then the state of Hawaii. With considerable federal, state, and private funds, and with concerted efforts of profes- sional personnel, that state was able to confine Achatina fulica to the two original infested islands of Oahu and Maui between the years 1936 and 1950. After that, the giant snail made the predicted, relentless progress to all of the other main islands in the archipelago. However, the history of the infestation and final eradication of this snail in southern Florida and in northeastern Australia are in con- trast for a number of important reasons (COLMAN, 1977, 1978; MEAD, 1979). With these earlier examples as guides, the authorities throughout the Caribbean must promulgate carefully considered external and internal quarantines and control measures to contain the spread of these two plant pests. Eradication has been proven possible, but adequate funds are absolutely vital. Regrettably, the inexorable time factor continues to intensify the urgency to take effective measures while the infestations are still somewhat limited. Voucher specimens of Achatina fulica and Limicolaria aurora, respectively are deposited in the Museum National d’Histoire Naturelle, Paris (unnumbered); National Mu- seum of Natural History, Washington, D.C. (USNM 860572, 860571); and Museum of Natural History, Santa Barbara, California (SBMNH 35522, 35523). ACKNOWLEDGMENTS We thank the General and Regional councils of Marti- nique for the financial aid that they have made available to the Federation Departementale des Groupements de Defense Contre les Ennemis des Cultures de la Martinique for the continuing battle against these two snails. LITERATURE CITED ANDERSEN, E., E. J. GUBLER, K. SORENSEN, J. BEDDARD & L. R. AsH. 1986. First report of Angiostrongylus cantonensis in Puerto Rico. American Journal of Tropical Medicine and Hygiene 35:319-322. Asst ADOoU, J., J. K. KOUAME, J. Moreau, M. S. TIMITE, J. Nozais & K. DiEKouabIo. 1980. L’angiostrongyloidose. Mise au point a propos d’un cas. Medecine d’Afrique Noire 27:421-425. CANSDALE, G. D. 1940. Giant forest snails of the Gold Coast. The Field (London) 175:546. CLARKE, B., J. MuRRAY & M. S. JOHNSON. 1984. The ex- Page 77 tinction of endemic species by a program of biological control. Pacific Science 38:97-104. Cotman, P. H. 1977. An introduction of Achatina fulica to Australia. Malacological Review 10:77-78. Cotman, P. H. 1978. An invading giant. Wildlife Australia 15:46-47. DEAN, W. W., A. R. MEAD & W. T. NoRTHEY. 1970. Aero- monas liquefaciens in the giant African snail, Achatina fulica. Journal of Invertebrate Pathology 16:346-351. FRANKIEL, L. 1989. Les Achatines aux Antilles. Centre De- partemental de Documentation Pedegogique, Circular. 10 PR: HADFIELD, M.G. & E. A. Kay. 1981. The multiple villainies of Euglandina rosea (or its human proponents). Hawaiian Shell News 29:5-6. Hart, A. D. 1978. Onslaught against Hawaii’s tree snails. Natural History 87:46-57. Meap, A. R. 1961. The Giant African Snail: A Problem in Economic Malacology. University of Chicago Press: Chi- cago, Illinois. 257 pp. MeEapD, A. R. 1973. A prognosis in the spread of the giant African snail to continental United States. Malacologia 14: 427. Meap, A. R. 1979. Economic malacology: with particular ref- erence to Achatina fulica. Vol. 2-B:150 pp. Jn: V. Fretter & J. Peake (eds.), Pulmonates. Academic Press: London. PERERA, G., M. YONG, J. RODRIGUEZ & D. GALVEZ. 1983. Cuban endemic molluscs infected with Angiostrongylus can- tonensis. Malacological Review 16:87-88. PEREZ MarTIN, O., M. LASTRE GONZALEZ, B. DUMENIGO RIPOLL, P. H. AGUIAR PRIETO & A. AGUILERA. 1984. Infestacion por Angiostrongylus cantonensis en las provincias habaneras. Revista Cubana de Medicina Tropical 36:54—58. Pitssry, H. A. 1919. A review of the land mollusks of the Belgian Congo. Bulletin of the American Museum of Nat- ural History 40:101. POINTIER, J. P. & C. BLANC. 1985. Achatina fulica en Polynesie Francaise: repartition, caracterisation des populations et con- sequences de l’introduction de l’escargot predateur Euglan- dina rosea en 1982-1983. Malakologische Abhandlungen (Dresden) 11:1-15. SCHOTMAN, C. Y. L. 1990. FAO Caribbean Plant Protection Commission Circular Letter PL 31/50, 31/30, 14 March 1990. SPENCE, G. C. 1938. Luimicolaria as a pest. Journal of Con- chology 21:72. VAN DER SCHALIE, H. 1969. Man meddles with nature—Ha- waiian style. The Biologist 51:136-146. WALLACE, G. D. & L. ROSEN. 1969. Studies on eosinophilic meningitis. V. Molluscan hosts of Angiostrongylus cantonensis on Pacific islands. American Journal of Tropical Medicine and Hygiene 18:206-2106. The Veliger 35(1):78-88 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 Reproductive Biology of Vermetus sp. and Dendropoma corrodens (Orbigny, 1842): Two Vermetid Gastropods from the Southern Caribbean PATRICIA A. MILOSLAVICH anp PABLO E. PENCHASZADEH Instituto de Tecnologia y Ciencias Marinas (INTECMAR), Universidad Simon Bolivar, Apartado 89.000, Caracas 1080, Venezuela Abstract. A comparison of the reproduction of two Venezuelan vermetids is presented. Vermetus sp. lives attached to hard substrates, up to 5 m in depth and occurs in decreasing densities with increasing depth, from 30 individuals/m? at sea level to 9 individuals/m? at 5 m. Dendropoma corrodens encrusts calcareous rocks in very shallow waters at densities of 130,000 individuals/m? (mean values). Both species reproduce throughout the year. Vermetus sp. broods up to 54 egg capsules in the mantle cavity, each containing 289 + 114 eggs that measure 240 + 14 um in diameter. Only 188 + 87 embryos developed to veliger larvae, with shells measuring 454 + 20 um in diameter at hatching; the remaining eggs are nurse eggs. Dendropoma corrodens broods up to 8 egg capsules in the mantle cavity, each containing 8 + 1.1 eggs measuring 276 + 25 um. A single egg in each capsule disintegrates, and its yolk is ingested by the developing embryos, which hatch as crawling juveniles with shells measuring 512 + 59 um. A summary of reproductive aspects of the family Vermetidae is given. INTRODUCTION Vermetids are abundant in the intertidal zone of warm- temperate to tropical seas (KEEN, 1961). They constitute a morphologically distinct group among gastropods, char- acterized by an uncoiled shell attached to or buried in the substrate and having mobility only in the early, hatched stages. The larva or juvenile settles to a suitable substrate and attaches; its adult shell then grows in a coil around an axis at a 90° angle to that of the larval shell (KEEN, 1961). As a consequence of sessile life, vermetids show some modifications in feeding and reproduction. The vermetid reproductive system has been described in detail by MORTON (1951, 1965) and HADFIELD (1970). HADFIELD (1969) presented information related to spermatophore structure, and HADFIELD & Hopper (1980) extensively described the male reproductive systems and spermatophoral organs from seven Hawaiian and Californian species in three genera. SCHEUWIMMER (1979, 1981) studied sperm di- morphism and sperm transfer in the vermetid Serpulorbis imbricatus. In the four vermetid genera that have been studied, Dendropoma, Petaloconchus, Serpulorbis, and Vermetus, there are two ways in which egg capsules are brooded: (1) free inside the mantle cavity, as reported by HADFIELD et al. (1972) for D. gregaria, D. platypus, D. ryssoconcha, D. psaro- cephala, P. montereyensis, and V. alu, by HUGHES (1978a) for D. corallinaceum, and by MorRTON (1965) for D. irre- gulare, D. marchadi, and D. tholia; or (2) attached to the internal side of the shell and suspended in the mantle cavity as reported by HADFIELD et al. (1972) for S. variabilis, D. meroclista, and P. keenae, by HUGHES (1978b) for S. aureus, by HUGHES (1978a) for S. natalensis, by HUGHES & LEWIS (1974) for D. maximum, by BANDEL (1976) for P. erectus and P. mcgintyi, by MORTON (1951) for S. zelandicus, and by MorTon (1965) for V. triqueter. BARASH and ZENZIPER (1985) presented a review of the structural diversity and adaptative characters of the Ver- metidae. In Hawaii, HOPPER (1981) studied the ecology and reproductive biology of some vermetid gastropods by comparing the dynamics and reproduction of populations that occupy the same or different habitats and the relation between life-history characteristics and female body size. In the Colombian southern Caribbean, BANDEL (1975a) studied some aspects of the reproduction of two species of Petaloconchus (P. erectus and P. meginty1). HADFIELD (1989) and HADFIELD & IAEA (1989) presented detailed infor- P. A. Miloslavich & P. E. Penchaszadeh, 1992 mation on the reproduction of Petaloconchus montereyensis from two populations (Washington and California) and discussed the effects that these two different latitudes may have on juvenile size and fecundity in the species. SAFRIEL & HADFIELD (1988) also studied in detail some ecological and reproductive aspects of Dendropoma meroclista and an apparent sibling species. In this paper we will describe two Venezuelan vermetids, Vermetus sp. and Dendropoma corrodens, and discuss their reproductive biology concern- ing brooding, type of development, and embryonic nutri- tion. MATERIALS anpD METHODS A population of Vermetus sp. was located in Puerto Ca- bello, Venezuela, living attached to the walls of the Planta Centro power plant cooling-water intake channel (10°30'6"N, 68°9'36”"W). This vermetid is found at the intertidal zone and at all depths in the channel (0-5 m). A Dendropoma corrodens population was located in Mor- rocoy National Park at Punta Mayorquina (10°53'45’N, 68°13'48”W), encrusting dead coral in shallow waters of the intertidal zone, at a maximum depth of 0.70 m (Figure i), From January to December 1986, monthly collections of both species were made at each site. Ten adult (total = 110) specimens of Vermetus sp. were collected and trans- ported to the laboratory in individual plastic bags with seawater, and two to three rocks colonized by Dendropoma corrodens were collected and taken to the laboratory in an ice-chest with seawater. Once in the laboratory, the two species were maintained separately; each individual of Ver- metus sp. was kept in a 300-mL beaker. The following aspects were studied for both species: (1) Localization of egg capsules within the female and brooding type. (2) Number and size of egg capsules per female. (3) Identification of the stages of the embryos inside the egg capsules. (4) Number of eggs and embryos per egg capsule and their size. (5) Time of embryonic development from egg to hatching, by reconstructing experimentally developing series starting from known stages of development. The egg capsules were separated according to the stage of their embryos and kept in 100-mL beakers inside an aquar- ium with Millipore-filtered (Whatman GF B) sea- water. The egg capsules were incubated at 27°C in a Precision (818) Incubator; air was supplied by an air- pump and capsules were kept in complete darkness. Observations of the embryos inside the egg capsules and the substitution of filtered seawater were done daily. At hatching, the type of development was noted (direct to crawling juvenile or indirect to veliger larva). (6) Histological study of the gonads by standard paraffin techniques and hematoxylin-eosin staining. £ - 228, . Oe o PLANTACENTRO yen - PUERTO CABELLO’.-. @ P 6 Q CARIBBEAN SEA VENEZUELA 79° Figure 1 Map of the Caribbean showing study areas: Planta Centro power plant and Punta Mayorquina. Physical parameters recorded monthly at each locality were temperature, salinity, and suspended solids in water (using the standard gravimetric technique described in APHA et al., 1985). RESULTS Habitat Vermetus sp.: This vermetid was found at all depths on the power plant channel wall at population densities rang- ing from 29.9 + 7.1 individuals/m? in the shallow areas (to 1 m depth) to 9.1 + 4.6 individuals/m? in the deeper areas (to 5 m depth). Temperatures varied from 26.4°C (January) to 30°C (April), salinity varied between 34 and 36 ppt, and organic suspended matter varied from 1.7 (February) to 21.8 (April) mg/L. The annual mean value of organic suspended matter was 6.21 + 6.04 mg/L. This high variance is due to an enormous input of organic matter brought by an adjacent stream (Quebrada El Palito) after a strong rain over its headwaters in April, which marks the beginning of the rainy season (April to September). Dendropoma corrodens: This species is entrenched in coral rocks at a population density of 12.6 + 4.8 individ- uals/cm/? at the collection site. Temperatures varied from Page 80 The Veliger, Vol. 35, No. 1 27 to 35°C within a single day, but were generally warmest during May-September and coldest during October—Feb- ruary. Salinity varied between 34 and 36 ppt, and organic suspended matter varied from 0.2 mg/L (February) to 7.5 mg/L (April). The annual mean value of organic sus- pended matter was 1.79 + 2.07 mg/L, and as in the Vermetus area, a considerable input of organic matter was caused by flooding of the stream. Species Description Vermetus sp.: Material has been deposited in the Amer- ican Museum of Natural History, New York (catalogue number 232098). The attached adult shell is 2.5 to 7.0 cm (mean = 4.8 + 1.0 cm, n = 107) in length and 2.0 to 4.4 cm (mean = 2.9 + 0.5 cm, n = 75) in width. The unattached tube grows perpendicular to the substrate and ranges from 1 to 12 cm in length (mean = 3.6 + 1.9 cm, n = 106) with a maximum opening of 1.2 cm in diameter. The coil ranges from 5 to 9 whorls (n = 75). The shell’s exterior is dark gray with thicker lines that are perpendicular to the di- rection of growth; the inside of the shell is smooth and light brown (Figure 2a). The soft bodies are dark red- wine in color, and the cephalic region is distinguished by the presence of small white spots. The adult body length ranges from 5.0 to 14.1 cm (mean = 8.4 + 1.8 cm, n = 102), and the diameter of the anterior region ranges from 0.4 to 1.1 em (mean = 0.7 + 0.1 cm, n = 102). The operculum is concave and brown, and measures 0.3 to 0.6 cm in diameter (mean = 0.5 + 0.1 cm, n = 102). The rachidian tooth of the radula has a trapezoidal shape with two prominent peaks at the base. Dendropoma corrodens: The shell is entrenched in the coral substrate, which is generally covered by calcareous algae (Lithothamnium) (Figure 2b). The shell aperture measures 1.5 + 0.1 mm (n = 102). The inside of the shell is smooth and dark red-wine in color near the aperture and cream-beige in the interior. The diameter of the coil ranges between 3.6 and 3.7 mm. The adult body length ranges from 3.8 to 13.5 mm (mean = 9.2 + 1.8 mm, 7 = 121); the diameter of the anterior region ranges from 0.7 to 1.8 mm (mean = 1.3 + 0.2 mm, n = 118); the anterior region is brown with white and yellow spots on the cephalic region. The operculum is convex and dome shaped, with concentric rings; it is reddish brown and ranges in diameter from 0.7 to 1.8 mm (mean = 1.3 + 0.2 mm, n = 102). The rachidian tooth of the radula has a rectangular shape. Figure 2 Figure 2a. Adult specimen of Vermetus sp. Figure 2b. Specimens of Dendropoma corrodens settled on Lith- othamnium substrate. P. A. Miloslavich & P. E. Penchaszadeh, 1992 Page 81 Table 1 Vermetus sp. Characteristics of embryonic development: stages I, II, III, and IV. Number of developing Capsules embryos per Size of diameter (mm) capsule embryos (um) mean + SD mean + SD mean + SD Stage (n) (n) (n) I nov a= (ahs) 289 + 114 240 + 14 (97) (33) (401) II 2.64 + 0.49 Mh ae [7 340 + 30 (79) (28) (320) III A3)a) 28 (AY ISL as {645) S15) 5 (96) (25) (255) IV 3.42 + 0.45 188 + 87 454 + 20 (51) (4) (55) Reproduction Both species are gonochoristic and reproduce throughout the year. The collected females had brooded egg capsules containing eggs and embryos in every month. Histological analysis of the gonads showed all stages of sexual maturity in both females and males in each collection throughout the year. Vermetus sp.: The female broods up to 54 spherical egg capsules in the mantle cavity (mean = 19.1 + 13.1,” = 49). No relationship was found between female size and the number of egg capsules brooded (Kendall correlation coefficient, P > 0.05, n = 49). The embryos inside the egg capsules were at different stages of development; these stages were classified in order of development from egg to hatching as I, II, III, and IV, or egg, trochophore larva, early veliger larva, and late veliger larva respectively (Ta- ble 1). Each capsule contains embryos at the same stage of development, but one female may be brooding embryos at different stages at the same time. In stage I, the egg capsules measure 2.3 + 0.4 mm in diameter and contain 289 + 114 eggs. The eggs are round and yellow, and measure 240 + 14 wm in length. The time to reach the next stage is four days, but after the first day the embryos are no longer round. In stage II, the egg capsules measure 2.6 + 0.5 mm in diameter and contain 254 + 72 trochophore larvae measuring 340 + 30 um in length. These larvae show signs of movement due to the presence of small cilia in the site where the two lobes of the velum will later develop. Ocular spots are distinguish- stage (from stage I) Number of days to reach each Characteristics of embryos 0 Yellow round eggs. (initial stage) 4 Trochophore larvae. Signs of moving. Small cilia. Ocular spots. 9 Early veliger larvae. Light yellow fragile shell. Packed yolk inside larvae. Small velum. Eyes. Cephalic tentacles. 14 Late veliger larvae. Well developed brown shell. Big velum. Internal yolk almost totally consumed. able at this stage (Figure 3a). The duration of embryonic development in stage II is five days. In stage III, the egg capsules measure 2.9 + 0.4 mm in diameter. These cap- sules contain 181 + 83 early veliger larvae with light yellow, fragile shells measuring 395 + 21 um in length. The velum measures 183 X 94 um in anteroposterior diameter. A large amount of packed yolk can be observed on the inside of the larva. The eyes are distinguishable at this stage (Figure 4a). The duration of stage III is five days. In stage IV, the egg capsules measure 3.4 + 0.5 mm in diameter and contain 188 + 87 late veliger larvae mea- suring 454 + 20 um in length; these larvae have a well developed brown shell with two coils, and the velum mea- sures 279 X 127 um (Figure 3c). At this stage the larval yolk is almost totally consumed (Figure 4b). The veliger larva will remain inside the egg capsule for five days until it hatches. The difference between the number of eggs in stage I and the number of hatching larvae is due to the presence of nurse eggs, which stop their development in an early stage. Early veliger larvae (stage III) and late veliger larvae (stage IV) were observed eating nurse eggs. (The breaking of the nurse eggs is apparently done by the action of oral cilia and it seems that the beating of the velum cilia also helps in the disintegration of the yolk.) The percentage of eggs that develop to veliger larvae ranges from 62 to 66% (100-107 nurse eggs/capsule). Not all the nurse eggs are consumed at hatching and 4 to 30 eggs remain within the capsule. Figure 3c shows the compact mass of nurse eggs Figure 3 Figure 3a. Egg capsule of Vermetus sp. Embryos in stage II. Figure 3b. Egg capsule of Dendropoma corrodens. Embryos in stage II. Figure 3c. Egg capsule of Vermetus sp. Embryos in stage IV. Detail of the compact mass of nurse eggs and veliger larvae. Figure 3d. Egg capsule of Dendropoma corrodens. Embryos between stages III and IV. P. A. Miloslavich & P. E. Penchaszadeh, 1992 that remain at stage IV. The veliger larva that hatches is positively phototropic and settles within 24 hr; once settled, it resorbs the velum and secretes a calcareous tube (Figure 4c). At this stage the pedal and cephalic tentacles are visible. The shell of a newly settled juvenile measures about 465 wm. The period between stage I and settlement is 20 days. In some of the veliger larvae, signs of abnormality were observed as an absence of shell coiling. The rest of the structures were normal, including the size of the velum and its movement. The significance of this shell variation is not clear, and settlement of these larvae under laboratory conditions did not occur. Dendropoma corrodens: ‘The female broods up to eight oval-shaped egg capsules at once in the mantle cavity (mean = 3.6 + 2.0, n = 27). No relationship was found between the number of capsules and the size of the female (Kendall correlation coefficient, P > 0.05, n = 27). As in Vermetus sp., capsules are found containing embryos at different stages of development in the same female. These stages were also classified as I, II, III, and IV, or egg, trochophore larva, early veliger larva, and crawling juvenile, respec- tively (Table 2). In stage I, egg capsules measure 771 + 113 um in length and contain 8.0 + 1.1 eggs; these eggs are round and cream- beige in color and measure 276 + 25 um in diameter. In stage II, egg capsules measure 850 + 127 um and contain 6.7 + 1.4 trochophore larvae measuring 329 + 46 um in length (Figure 3b). In stage III, egg capsules measure 1120 + 58 wm and contain 6.3 + 1.9 larvae with shells mea- suring 465 + 54 um in length; these larvae have a velum, a foot covered by the operculum, and cephalic tentacles (Figure 5a). In stage IV, egg capsules measure 1099 + 132 um and contain 6.3 + 1.8 juveniles measuring 512 + 59 wm; these juveniles have resorbed the velum (Figure 3d). Once the juveniles hatch (Figure 5b), they crawl from their brooding site to the adjacent substrate (Figure 4c). Dendropoma corrodens crawling juveniles present two color morphs: 92.5% are white and 7.5% are brown. The time of development from egg to hatching could not be deter- mined as in Vermetus sp. because the embryos inside the egg capsules die within hours outside the female’s mantle cavity. The existence of nurse eggs was determined for Den- dropoma corrodens. The first piece of evidence is a signif- icant difference (Kruskal-Wallis and Tukey multiple com- parisons, P < 0.05) between the number of embryos per capsule in the first stage as compared to the other three stages. The number of capsules with embryos in stages I, II, III, and IV that were counted for the Kruskal-Wallis test is given in Table 2. The second piece of evidence is that, after stage I and within a few hours of development from egg to trochophore larvae, a small amount of disin- tegrated yolk can be observed (Figure 3b). This yolk has been totally consumed at stage III]. The percentage of individuals that reach the final stage and hatch is 78 to 79%. Page 83 Figure 4 Vermetus sp. a. Early veliger larva, stage III. b. Late veliger larva, stage IV. c. Juvenile, having fixed to the substrate. Key: An, anus; Cm, columella; Cs, cilia of stomach; Ct, cephalic ten- tacles; CVe, cilia of velum; Eso, esophagus; Ey, eye; F, foot; H, heart; Hp, hepatopancreas (digestive gland); In, intestine; M, mouth; Oc, ocellus; Op, operculum; Pc, protoconch; Pt, pedal tentacles; S, stomach; Tu, tube; Ve, velum. Page 84 The Veliger, Vol. 35, No. 1 Table 2 Dendropoma corrodens. Characteristics of embryonic development: stages I, II, III, and IV. Capsules diameter (um) Niele of : developing em- Size of mean + SD bryos per capsule embryos (um) mean + SD mean + SD Stage Length Width (n) (n) Characteristics of embryos 1 CA as WS) 695 + 94 8.0 + 1.1 ZO 25 Light yellow round eggs. (13) (13) (12) (30) II 850 + 127 683 + 97 6.7 + 1.4 329 + 46 Trochophore larvae. (26) (26) (25) (66) Signs of moving. Small cilia. Ocular spots. Ill 1120 + 58 812 + 82 Op as iy 465 + 54 Early veliger larvae. (23) (23) (27) (135) White-yellow fragile shell. Packed yolk inside larvae. Small velum. Eyes. Foot covered by operculum. Cephalic tentacles. IV 1099 + 132 835 + 100 637 == 158 SI2EEI59 Crawling juvenile. (19) (19) (28) (134) Well developed brown or white-yellow shell. No velum. Well developed foot. DISCUSSION inside of the power plant’s machinery, where high tem- Vermetus sp. exploits a very special habitat, a fouling com- munity that has developed on two walls (450 m each and 5 to 6 m depth) of a power plant water intake. The most abundant species in the power plant channel are colonial tunicates, bryozoans, cirripedes, sea anemones, polychaetes (sabellids, serpulids, and spirorbids), hydrozoans, sponges, and algae. In field experiments done to determine the frequency of recruitment of different species established in the power plant channel on Plexiglas® plates of 10 x 10 cm’, LOsADA et al. (1986) found that, for the period July 1985 to March 1986, vermetid larvae settled on 56% of the Plexiglas® plates placed near our collection site at an intensity classified as “frequent” (between 12.6 and 62.5% of total species settled on the plate). However, they could not distinguish between the larvae of Vermetus sp. and Petaloconchus varians, which is another vermetid that lives on the channel wall. The high fecundity of Vermetus probably compensates for the great mortality of the hatching veliger larvae, which settle within a period of 24 hr. No significant difference was found between the shell size of the hatching veliger and the settled juvenile; this indicates, as suggested by SAFRIEL & HADFIELD (1988) for Dendropoma meroclista, that the larva reside too briefly in the plankton for plank- totrophy to be effective. With normal functioning of the power plant, the measured current speed was 0.20 m/sec, so it takes about 37 min to go from the sea entrance to the perature, mechanical stress, and chlorine application cause high mortality in larval populations. Under these condi- tions, the local population of Vermetus sp. can be main- tained by two ways. The first is that vermetid larvae set- tling on the channel walls may be brought by sea from another, relatively close population of Vermetus sp. This is very possible because we noted high densities of this species on the deck of the petroleum refinery El Palito, located about 1 km east and upcurrent of the power plant. The second way is that larvae produced by the vermetid population are able to settle before entering the power plant machinery. LOSADA et al. (1986) mentioned that the spatial heterogeneity generated by the development of a fouling community on the channel walls changes the water circulation pattern near them and, consequently, affects larval settlement; in this case, it probably increases the chance of larvae survival long enough for them to settle. Dendropoma corrodens exploits a habitat whose environ- mental conditions vary little throughout the year. As re- ported by HADFIELD et al. (1972) for Dendropoma mero- clista and Dendropoma rhyssoconcha, and by HUGHES (1979) for Dendropma corallinaceum, this genus preferentially col- onizes patches of the coralline alga Lithothamnium. Dis- persal seems to be very low, and the settlement of several juveniles around a female has been observed, even when the juveniles have the capacity to crawl to surrounding substrates. There is evidence that both Vermetus sp. and Dendro- P. A. Miloslavich & P. E. Penchaszadeh, 1992 poma corrodens reproduce throughout the year, apparently at the same intensity. As in the Hawaiian vermetids (HaD- FIELD ef al., 1972; HOPPER, 1981) and in specimens of Petaloconchus montereyensis from Monterey Bay, Califor- nia, and San Juan Island, Washington, the females we collected had brooding eggs and embryos in every month, and the monthly gonad histological analysis showed grow- ing and mature oocytes in the same individual. This in- dicates that females have the capacity to produce eggs constantly, and because the egg capsules are brooded in the mantle cavity, different stages of embryonic develop- ment are found at the same time in each female. HADFIELD (1989) also found in female specimens of Petaloconchus montereyensis a variety of developmental stages, ranging from uncleaved ova to ready-to-hatch juveniles within a single brood. The different colors observed in the older-shelled ju- veniles of Dendropoma corrodens have been observed also by HADFIELD et al. (1972) in D. gregaria, but in a different proportion. In D. corallinaceum, the embryo develops a dark brown protoconch (HUGHES, 1978a). As observed in Tables 1 and 2, the diameter of the egg capsule increases in both species from the first to the last stages as does the egg capsule volume. This increase is probably due to an increase in permeability to water; in Vermetus sp. the egg capsule breaks easily in the last stage, for example even when the female retracts suddenly in the shell. In Dendropoma corrodens, there are marked differ- ences in the egg capsule volume between stages II and III, with no difference between I and II, or between III and IV. This is probably because the highest growth of the embryos occurs within stages II and III, when the larval shell begins to form. Comparison of some reproductive habits of Vermetus sp. and Dendropoma corrodens with other species of these two genera that have been studied is presented in Table 3. The pattern of reproduction of Vermetus sp. is very similar to that of Vermetus ali (HADFIELD et al., 1972; HOPPER, 1981), that is, in the brooding type, number of egg capsules per female, number of eggs per capsule, size of egg and larvae, and indirect development to veliger larvae. Den- dropoma corrodens, as well as the other species of Dendro- poma studied (HUGHES, 1978a; HADFIELD et al., 1972; Hopper, 1981), broods few egg capsules per female, and it has a relatively small egg diameter (0.276 mm) compared with the rest of the genera (D. maximum, 0.250 mm [HucHEs & Lewis, 1974] and D. meroclista, 0.277 mm [HADFIELD et al., 1972]). The protoconch size (0.512 + 0.059 mm) in D. corrodens is comparable to that of other species of Dendropoma—D. maximum, 0.03 mm (HUGHES & Lewis, 1974) and D. corallinaceum, 0.97 mm (HUGHES, 1978a). Usually in Vermetus and Dendropoma species, capsules are brooded in a slit of the mantle (Table 3): see V. alli, D. gregaria, D. platypus, D. ryssoconcha, and D. psarocephala as reported by HADFIELD et al. (1972) and D. corallinaceum as reported by HUGHES (1978a). Exceptions to the rule Page 85 Figure 5 Dendropoma corrodens. a. Veliger larvae, stage III. b. Crawling juvenile at hatching. c. Juvenile, having fixed to the substrate. See Figure 4 for key to abbreviations. are D. meroclista, D. cf. meroclista, D. maximum, and V. triqueter, which brood their egg capsules attached to the shell (HADFIELD e¢ al., 1972; SAFRIEL & HADFIELD, 1988; HuGHEs & Lewis, 1974; MorTON, 1965, respectively). The species of the genera Petaloconchus and Serpulorbis usually brood their capsules attached to the internal side of the shell: P. keenae and S. variabilis (HADFIELD et al., 1972, S. aureus (HUGHES, 1978b), S. natalensis (HUGHES, Apnys juasaid qSL6l “ISaNVg S961 ‘NOLYOW ZL6L “72 72 ATSIAGVH ZLOL “79 12 ATAIAGVH &8/6| SAHONH IS6] ‘NOLYOW 861 ‘SaHONA IS6I ‘NOLUOW 6861 ‘VAV] 79 ATalddvH 6861 ‘A1dIdavy 6861 ‘VAV] 29 ATaldavy 6861 ‘OL6] ‘ATaldavH 9L61 “THANG The Veliger, Vol. 35, No. 1 ZLOL “72 12 ATSIAAGVH 9L61 ‘4SL6l “THaNVg CLOL “72 12 ATalAaV CLO “79 72 ATaIAAaVH ZLOL “79 72 ATEIAaVH 886 ‘ATaIdavVH 2% Taldavs 8861 ‘ATaIdavVH 2 Taldavs CLO “79 12 ATaUIAAGVH +l6l ‘SIMA'T 7 SAHONY CLO6L “72 12 ATaIAGV Apnys juasoid 28/61 ‘SAHONY JNUIIIJIY Page 86 $839 asinu QQ] s339 osunu s]J20 dANLaNU yJoA Maj s339 asinu s339 osinu $339 osinu yJOA repnotsaa yJOA repNotsaa yJOA repNoIsaa é¥JOA padsasqo jou yJOA repnotsaa 3329 9sinuU | uonynu ajnsdeorsur [euontppy JaBI[IA JeYIUOITIIA /3ulpMeso JaBIOA JIBI[OA Sul[Meo JaBI]oA SuI[MesIO SulpMeso Sul[MeIO Sul[Meso JIBITOA JOBITIA /3ulpMeso IVYIUOITIIA Sul[Meto SuI[MeIO Sul[Meto JaBI[OA JIBIIA I1e] /BulpmMeso Sul[Meso sul[Meto Sul[MeIO Sul[Meto Suryoie fF] YS spueuw 194s Uy]s apueur 14s 184s 14s WIS apueuU WIS opueur Te us Tes IT9us WIS apueur YS apueur WIS apueUr 14s Ileus Tes Ws opueur YS spueuw YS spueuw add} Sutpooig ‘snjaulag = ‘fA ‘siqsojndiag = ‘§ ‘snyouov0}pjag = ‘q ‘bwodoipuaq = -q 4 (pS-1) IZ (AT 881) 682 9-4 SII-S8 Os 0072-08 8+ 74 O¢ (Ib-S1) +7 0€-02 08-OZ 07 O€ = Ol SO'LI (aqtuaanl 1) €¢Z 88'rl (aqtuaant 1) Z¢ v-€ 001-08 (AT O€-LI) = P8I-O€l == O00I< 9 S7-9 S €7-6 S OL 9-2 bel-1Z 9 91-€ SI brr-Lle v-€ (17-9) OT 8-1 (AT L-9) 8 LI I ayeway atnsdeo Jad sajnsder) Jad sokiquiy ‘(aearey = ArT) Ayturey pnawiaa ay} ut syoadse aanonpoidas jo Arewuing CLAD Orc 0 (uur) Ja}aure Ip 334 ‘ds snjaua 4 sajanbiy “4 LD: Se MAI Le: sisuayDjDU DIDIIAG UL snaino §NI10103}] 0D (Aare u0yy) sisuakalaquUouL * purysy uenf{ ues) sisuakasaqUoUL * idquigoue * aDUaaYy * SnJIa41a * Dy IUor0sskYyt vjpyqar0.vsq ° sndajojd ° DIS1JIOLIU yp DISUIOLIUL * WNUAXDUL ° DIUDGILS SUaPOLLOI WNIIDUYIVLOD * {satoadg Z C d 1 S) » Ss iS) Ss Q Q QRaAa Q P. A. Miloslavich & P. E. Penchaszadeh, 1992 1978a), S. daidai and S. medusae (SCHEUWIMMER & NISHI- WAKI, 1982), and P. varians (personal observation). An exception would be P. montereyensis, which broods the egg capsules in a slit of the mantle (HADFIELD, 1970, 1989). Usually, Dendropoma species brood many fewer egg cap- sules than do Vermetus species; D. corrodens broods from 1 to 8 capsules, 3 to 4 being the most common, as in D. gregaria. The maximum number of capsules per female reported for this genus is 15, for D. maximum. Both Ver- metus species, V. alli and Vermetus sp., brood up to 50 and 54 capsules, respectively; but V. ali: produces a few large broods per year (HOPPER, 1981) and in Vermetus sp. there is a continuous production of egg capsules throughout the year. In contrast, V. triqueter broods few egg capsules, between 4 and 6 (BANDEL, 1975b). As noted in Table 3, the egg capsules also differ in the number of embryos they contain according to the genus. Dendropoma species usually have few embryos per capsule (between 1 and 25), with the exceptions of D. platypus, which has up to 70 embryos (HADFIELD et al., 1972) and produces a few large broods per year (HOPPER, 1981), and D. maximum, which has up to 444 embryos (HUGHES & LEwis, 1974). Among Ver- metus species, the range is between 80 and 289 embryos per capsule. At hatching, all the Dendropoma species stud- ied are crawling juveniles with the exception of D. mero- clista, which presents the two hatching modes: crawling juveniles and late veliger larvae (HADFIELD et al., 1972). However, the D. cf. meroclista found on the Sinai coasts of the Gulf of Elat hatches as a typical planktonic veliger larva (SAFRIEL & HADFIELD, 1988). Both Vermetus sp. and V. all: hatch as veliger larvae, and V. triqueter may hatch as a veliconch larva or as a crawling juvenile (BANDEL, 1976). Additional nutrition during embryonic development has previously been reported in other species of the family by Morton (1951) in Serpulorbis imbricata, by HUGHES (1978a) in S. natalensis, and by HADFIELD et al. (1972) and Hopper (1981) in Dendropoma gregaria, D. meroclista, D. platypus, D. rhyssoconcha, D. psarocephala, Petaloconchus keenae, and Serpulorbis variabilis. Of special interest is the case of P. montereyensis, in which each capsule typically produces only a single embryo, with the remainder of the 164-499 eggs in a capsule serving as nurse yolk for the one developing embryo (HADFIELD, 1989; HADFIELD & TAEA, 1989). Both Vermetus sp. and Dendropoma corrodens ingest nurse eggs but in two very different ways. In egg capsules of Vermetus sp., the nurse eggs (about 100 for the develop- ment of 200 veliger larvae) remain inside the egg capsule as a compact mass, until they are almost completely con- sumed by the embryos at the time of hatching. In egg capsules of D. corrodens, one nurse egg provides additional nutrition to the six or seven embryos, and its disintegration into yolk particles occurs within the first hours of devel- opment, that is, after the start of stage I and before stage II; a similar phenomenon was described by PENCHAS- ZADEH (1976) for two species of the genus 77rophon. Page 87 ACKNOWLEDGMENTS We thank Dr. Rudiger Bieler of the Delaware Natural History Museum, who spent a couple of weeks with us in February 1986, for helping us with the identification of the species and for reviewing the manuscript, improving this paper with useful comments. We are especially grate- ful to Eduardo Klein for one year of help on the field trips and the statistical treatment of the data, and to Dr. David Bone of the Universidad Simon Bolivar for reviewing the manuscript. We are indebted to INTECMAR and CON- ICIT, Venezuela, for financial support. LITERATURE CITED A.P.H.A., A.W.W.A. & W.P.C.F. 1985. Standard Methods for the Examination of Water and Wastewater. 16th ed. Washington, D.C. 1134 pp. BANDEL, K. 1975a. Embryoanagehause karibischer Meso- und Neogastropoden (Mollusca). Abhandlung der Mathema- tisch—Naturwissenschaftlichen Klasse, Akademie der Wis- senschaften und der Literatur, Mainz (1):1-133, pls. 1-21. BANDEL, K. 1975b. Das Embryonalgehause Mariner Proso- branchier der Region von Banyuls-sur-Mer. Vie Milieu 25(1):83-118. BANDEL, K. 1976. Observations on spawn, embryonic devel- opment and ecology of some Caribbean Lower Mesogastrop- oda (Mollusca). The Veliger 18(3):249-271. BarasH, A. & Z. ZENZIPER. 1985. Structural and biological adaptations of Vermetidae (Gastropoda). Bolletino Mala- cologico 21(7-9):145-176. HADFIELD, M. G. 1969. Nurse eggs and giant sperm in the Vermetidae. American Zoologist 9:520, 1142. HADFIELD, M. G. 1970. Observations on the anatomy and biology of two California vermetid gastropods. The Veliger 12(3):301-309. HADFIELD, M. G. 1989. Latitudinal effects on juvenile size and fecundity in Petaloconchus (Gastropoda). Bulletin of Ma- rine Science 45(2):369-376. HADFIELD, M. G. & C. N. Hopper. 1980. Ecological and evolutionary significance of pelagic spermatophores of ver- metid gastropods. Marine Biology 57:315-325. HADFIELD, M.G. & D. K. IAEA. 1989. Velum of encapsulated veligers of Petaloconchus (Gastropoda), and the problem of re-evolution of planktotrophic larvae. Bulletin of Marine Science 45(2):377-386. HADFIELD, M. G., E. A. Kay, M. U. GILLETTE & M. C. LLoyp. 1972. The Vermetidae (Mollusca: Gastropoda) of the Ha- waiian Islands. Marine Biology 12:81-98. Hopper, C. N. 1981. The ecology and reproductive biology of some Hawaiian vermetid gastropods. Ph.D. Thesis, Uni- versity of Hawaii, Honolulu. HuGuHeEs, R. N. 1978a. The biology of Dendropoma corallina- ceum and Serpulorbis natalensis, two South African vermetid gastropods. Zoological Journal of the Linnean Society 64: 111-127. HUuGuHEs, R. N. 1978b. A new species of Serpulorbis (Gastrop- oda: Vermetidae) from South Africa. The Veliger 20(3):288- 292. HuGHES, R. N. 1979. Notes on the reproductive strategies of the South African vermetid gastropods Dendropoma coralli- naceum and Serpulorbis natalensis. The Veliger 21(4):423- 427. HuGueEs, R. N. & A. H. Lewis. 1974. On the spatial distri- Page 88 bution, feeding and reproduction of the vermetid gastropod Dendropoma maximum. Journal of Zoology (London) 172: 531-547. KEEN, A. M. 1961. A proposed reclassification of the gastropod family Vermetidae. Bulletin of the British Museum (Natural History), Zoology 7(3):183-212. Losaba, F. S., J. A. LA SCHIAZZA, R. POLEO & S. M. PAULS. 1986. Reclutamiento y desarrollo de las comunidades in- crustantes en Planta Centro. Pp. 127-150. Jn: P. E. Pen- chaszadeh (ed.), Ecologia del ambiente marino costero de Punta Moron. Final report of the third phase of the project CADAFE-USB. Tomo II. Morton, J. E. 1951. The structure and adaptations of the New Zealand Vermetidae. Part I. The genus Serpulorbis. Transactions of the Royal Society of New Zealand 79(1):1- 19. Morton, J. E. 1965. Form and function in the evolution of the Vermetidae. Bulletin of the British Museum (Natural History), Zoology 11(9):585-630. The Veliger, Vol. 35, No. 1 PENCHASZADEH, P. E. 1976. Reproduccion de Gasteropodos Prosobranquios del Atlantico Sur-Occidental. El género 770- phon. Physis 35(90):69-76. SAFRIEL, U. N. & M. G. HADFIELD. 1988. Sibling speciation by life-history divergence in Dendropoma (Gastropoda; Ver- metidae). Biological Journal of the Linnean Society 35:1- 13? SCHEUWIMMER, A. 1979. Sperm transfer in the sessile gastro- pod Serpulorbis (Prosobranchia: Vermetidae). Marine Ecol- ogy Progress Series 1:65-70. SCHEUWIMMER, A. 1981. Spermatozoendimorphismus und Spermatozoenubertragung bei der Wurmschnecke Serpulor- bis imbricatus (Dunker) (Gastropoda: Prosobranchia). Ph.D. Dissertation, University of Vienna, Austria. 73 pp. SCHEUWIMMER, A. & S. NISHIWAKI. 1982. Comparative studies on three Japanese species of Serpulorbis (Prosobranchia: Ver- metidae) with description of a new species. Venus 41(2):85- 101. The Veliger 35(1):89-90 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 BOOKS, PERIODICALS & PAMPHLETS Systematic Revision and Suprageneric Classification of Trochacean Gastropods by CAROLE S. HICKMAN & JAMES H. MCLEAN. 1990. Natural History Museum of Los Angeles County, Science Series No. 35. 169 pp., 100 figs. The order Archaeogastropoda occupies a distinguished, if unappreciated, position in the evolution of life. From small, inconspicuous, Lower Cambrian beginnings, it has survived the vicissitudes of a half billion years of earth history. And its cladogenetic proclivities generated the most spectacular evolutionary radiations in the Mollusca, pro- ducing the remarkably successful opisthobranch and pul- monate subclasses as well as the diverse and more accom- plished caenogastropods within the Prosobranchia. The goal of this monograph was improvement of the family-group level classification of a major and pivotal superfamily of archaogastropods, the Trochacea (the au- thors eschew the more current form, Trochoidea). It results from lengthy and diligent efforts of the most knowledgeable contemporary systematists of the group, it reports a great deal of original research and substantial progress in un- derstanding the taxon, and it succeeds in presenting a richly illustrated, more rational classification, new appreciation of the diversity of trochacean structures and life styles, and a provocative phylogenetic hypothesis. The authors present their systematic approach as “‘con- cept and objectives,’ worth summarizing as their ten com- mandments for revisionary taxonomy. Thou shalt: 1, clear- ly diagnose higher taxa; 2, base diagnoses on apomorphies rather than collections of plesiomorphies; 3, construct large character sets with clearly defined character states; 4, use only characters scored for all taxa; 5, avoid proliferation of higher taxa; 6, provide standardized comparative illus- trations of all character sets for all taxa; 7, include the fossil record and stratigraphic age as evidence of evolu- tionary history; 8, summarize biogeographic and ecological data for each taxon; 9, use natural history data in system- atics; 10, identify currently unresolved problems and prom- ising topics for future research. The monograph clearly reflects adherence to these pol- icies. That it was a formidable task is evident at the outset because, by outgroup comparison with Pleurotomariacea, the major apomorphies of the Trochacea are negative, that is losses of characters: ““The most important innovations of Trochacea include loss of the shell slit, labral emargi- nation, or tremata and loss of the right ctenidium.” In the evolution of prosobranchs, as in primates, the tales of teeth are telling. Probably the most important contribution of this work is its development and synthesis of extensive, largely original comparative data on trocha- cean radular patterns. The radula is a very complex struc- ture as well as a major evolutionary innovation of the Mol- lusca. As in mammals, the numbers and types of teeth, their independent and interdependent modifications, and their positional and functional relationships to each other afford a rich set of characters and states. The authors have discovered many of these, and they illustrate them with the results of excellent SEM studies and a few “in your face” views of radulas at work as well. The table of tro- chacean radular attributes lists an impressive 47 charac- ters, with essentially all states coded as binary. Radular characters are weighted most heavily and de- scribed in greatest detail. A morphological overview section only introduces the topic; most of the basic comparative and functional morphology is in the systematic treatments. Radulas are quite constant in the Turbinidae, so other characters must be used in this family, but the evolution of Trochidae involved several interesting trends, often in the directions of reduced number of teeth per row and specialization of those remaining. The table of non-radular characters contains 71 entries pertaining to shell (26), foot and epipodium (17), oper- culum (9), snout (9), ctenidium (6), eyestalks, and buccal cavity. The overview section discusses some of these co- gently and illustrates them well, e.g., epipodial features and radial and tangential shell aperture orientations and their biomechanical significance. Others get shorter shrift, e.g., the “pseudoproboscis,” operculum secretory region, and ctenidial afferent membrane. The character states of the latter two are given only as long or short, that is they are treated as binary characters although their variation is surely continuous. More explanation would have been particularly welcome here, as these are important char- acters at the subfamily level and subfamilies are particu- larly important in this monograph. While retaining the familiar Turbinidae and Trochidae as the only major fam- ilies of Trochacea, Hickman and McLean accommodate their new taxonomic results primarily by establishing a number of new subfamilies. The authors’ taxonomic approach draws both from evo- lutionary systematics and cladistics. Their methodology seeks to take advantage of the positive contributions of cladistics (reliance on shared, derived characters, objectiv- ity, precision, reproducibility of results, and testability of phylogenetic hypotheses) while avoiding the shortcomings (nonrepresentation of anagenetic change and degree of di- vergence, reliance on parsimony, and difficulties with ac- commodating continuously varying characters and both contemporaneous information and the fossil record). This methodology will irk advocates of purity (that is, cladistics or evolutionary phylogeny or phenetics but please not a mix), or who follow commandments of other gods Page 90 in theoretically framing systematics research, and it will frustrate users who would like the evidence up front, before the interpretation. Here, characters and states are pre- sented, their polarities, apomorphies and convergences as- serted, and they are used to develop the classification, some- times with limited defense. Character state polarities are congently if concisely argued at the family level, but as noted above the authors’ main taxonomic innovation is the establishment of numerous subfamilies. The reader is not given explicit evidence supporting, for example, absence of cephalic lappets in Solariellinae as a secondary loss rather than plesiomorphic, small size as secondary in Col- loniinae, or four pairs of epipodial tentacles as primitive to three pairs. The monograph concludes with a brief but densely packed summary containing diagrams that look suspiciously like cladograms or trees—at least taxa are the terminal buds of a hierarchical branching pattern. However, these are based on and summarize the classifications and their di- agnoses presented in the main text. They are not clado- grams in the usual sense, in which state differences in the selected character sets are allowed to generate tree topol- ogies. The authors characterize the results of such usual cladistic analyses as “analytical cladograms.” They distin- guish the diagrams they present rather as “retrospective cladograms,” that is they are derived a posterior: from the classification. This reviewer is inexpert in both Trochacea and cla- The Veliger Volks a5 Nom distics. The former have made only a few cameo appear- ances in his papers, and he has published only one modest cladistic analysis (and that a decade ago), and last year subtitled a seminar ‘““Why not to be a cladist” (after VAN VALEN, 1978). But he is a bit uneasy that this volume does not give cladistics a fair shake at the problem addressed. Modern computer packages accommodate character weighting (e.g., SANDERSON, 1990), and the authors’ treat- ment of quantitative characters, while questioned above, is suitable for cladistic analysis. One would like to see what alternative trees a cladistic treatment would generate from the characters and states carefully presented, and the like- lihoods of the trees presented here under cladistic con- straints. However, I suspect that Hickman and McLean’s intimate knowledge of the biology and paleontology of trochacean gastropods, and their explicit methodology and presentation of output from the highly effective supercom- puters for systematics in their minds, bring us closer to knowing the true phylogenetic history of the Trochacea. Alan J. Kohn LITERATURE CITED SANDERSON, M. J. 1990. Flexible phylogeny reconstruction: a review of phylogenetic inference packages using parsimony [software review]. Systematic Zoology 39:414-420. VAN VALEN, L. 1978. Why not to be a cladist. Evolutionary Theory 3:285-299. The Veliger 35(1):91-92 (January 2, 1992) THE VELIGER © CMS, Inc., 1992 NOTES, INFORMATION & NEWS First International Latin American Malacological Congress Caracas, Venezuela, 15-18 July 1991 Proposed North-South Collaboration We have recently returned from participating in this important and precedent-setting meeting. The purpose of this memorandum is to convey some of our impressions, and to advance some proposals for future collaboration between malacologists in Latin America and those of us in the United States, Europe, and elsewhere. This Congress was a long time in coming. Plans were first advanced for a Latin American meeting in Costa Rica in 1984, but these never materialized. A second attempt, in Colombia in 1986, was also unsuccessful. Finally, the right set of factors came together—financing, institutional support, and organizational abilities—to bring about the Congress held in Caracas this past July. A great deal of credit goes to Dr. Paulo Penchaszadeh of the Universidad Simon Bolivar, site of the Congress, but especially to Rob- ert Cipriani, who devoted much of his time over several months to organizing the event. Participants stayed in hotels in downtown Caracas, trav- eling by bus each day to the Universidad Simon Bolivar, which nestles in a mountain valley about 1300 m elevation, some 30 minutes from the center of the city. A final banquet was held in a fine restaurant in downtown Caracas. The Congress was a great success, beyond any of our expectations! It was attended by some 170 participants and other malacologists and students from most countries in Latin America and the Caribbean. Particularly well rep- resented were, of course, Venezuela, but also Mexico, Bra- zil, and Chile. Other Latin American countries represented were Peru, Costa Rica, Argentina, Panama, Cuba, Bolivia, Saint Lucia, Martinique, and Turks. Other attendees were present from the United States, including Puerto Rico, Spain, The Netherlands, Italy, and Great Britain. Ap- proximately 90 papers and 9 posters were presented fea- turing land, freshwater, and marine mollusks, and in- cluding paleontology. In part, the Congress attracted so many malacologists and students because of a symposium on Strombus gigas organized by Richard Appeldoorn of the University of Puerto Rico. Fishery for this commercially important spe- cies is on the decline throughout the Caribbean, and sym- posium participants were able to share information on the biology of this species and on measures that can be taken to achieve a more sustainable fishery. A new display of mollusks in the Natural History Mu- seum of the Universidad Simon Bolivar was unveiled at the meeting, featuring displays of Venezuelan mollusks, both Recent and fossil. What of future Congresses? Malacological leaders pres- ent at this Congress concluded that a Latin American association with officers and individual members would not be practical at this time and would be difficult to sustain. Instead, they agreed on, and formed a loose-knit steering committee working with a few participants from each country who would serve as points of information exchange to institutions and amateur groups in their coun- tries. The steering committee, called the Comite Organizador de Congresos Latinoamericanos de Malacologia, was formed 18 July 1991, and consists of: M. Sc. Martha Reguero Reza, Presidente (México) Dr. Antonio Garcia Cubas, Vicepresidente (México) Dra. Maria Villarroel Melo, Secretaria (México) Dr. José Willibaldo Thome, Vocal de Programa (Brazil) Dr. Paulo E. Penchaszadeh, Vocal de Programa (Ven- ezuela) Lic. Roberto Cipriani Fita, Vocal de Programa (Ven- ezuela) The functions of the committee are to: (a) assure the continuity of periodic meetings; (b) establish contact with an advisory group formed by the presidents of the various malacological societies existing in Latin America; (c) pro- mote the establishment of regional malacological societies in those countries where they do not exist, developing lists of malacologists in each area; and (d) approve the fre- quency and sites of congresses, and the constitution of the local organizing committees, with a view to bringing about future meetings. A second congress will be organized in two or three years, probably in Mexico. The Comité suggested that it might be appropriate to consider joint meetings in a few years with the American Malacological Union, the Na- tional Shellfisheries Association, and eventually hosting with Unitas Malacologica. Meanwhile, other measures are necessary to further the progress of malacology in Latin America and the Carib- bean. There is especial need for enhancing communications with workers with common interests and providing access to literature and materials in these countries. These mala- cologists, though having great enthusiasm and ability, meet with considerable frustration because of inadequate re- sources. A number of proposals were advanced by the Comite. For example, the kinds of workshops that Brian Morton and others have run in Hong Kong and elsewhere allow more intensive interaction with skilled professionals. With a little advance planning, visitors can give “‘short courses” on specific topics and, in most cases, these need not be in Page 92 Spanish. Resulting in significant benefits would be some major collaborative projects on particular topics among a number of workers with different skills in different coun- tries; the classification of a particular family, such as the oysters, or the biology of a gastropod like Strombus, are possible examples. The key problem is, of course, resources. While that problem is significant for institutions in the United States, it is acute in Latin America and the Caribbean, where universities with a number of students and faculty inter- ested in malacology lack key literature on the subject. Surely we can do more to help! One final major proposal emerged from the Comité meetings: to form semi-formal, north-south sister insti- tutional linkages. Scientists in the United States and else- where have been very helpful and generous to individual Latin American and Caribbean scientists making specific requests of them for literature or other help. However, this has been very episodic, and our southern colleagues are often unaware of the resources they need or persons with whom they should be in contact. It thus occurred to us that semi-formal matching, or linking, of key institu- tions could result in meaningful mutual benefits. More specifically, we are thinking of taking steps to link one United States institution, such the malacology de- partment of a museum or university, with one or two centers of malacological activity in Latin America or the Caribbean. The United States institution would, at a min- imum, provide a point of first communication for the south- ern institution(s) by suggesting to them literature, persons, or other resources. In return, the Latin American insti- tution would serve the same function for work going on in its area. Beyond this, and at the discretion of the insti- tutions involved, would be exchanging literature, trading literature for specimens needed in northern institutions for research, reciprocal visits, joint projects, and the like. The Comité Organizador de Congresos Latinoameri- canos de Malacologia is undertaking a country-by-country survey of needs and which institutions might be most ap- propriate for such linkages. In return, we have agreed to make such a survey in the United States, and Henry Coo- mans of The Netherlands and Andy Beaumont of Great Britain, who attended the Congress, will make similar inquiries in Europe. The Veligers Vols 355. Noma So, we end with the important question: Would your institution be willing to serve as a contact point for Latin American and Caribbean exchanges? Have you an interest in a particular country or institution? Please drop us a note, mailing it to Gene Coan. Dr. Gene Coan 891 San Jude Ave. Palo Alto, California 94306 Dr. Melbourne R. Carriker College of Marine Studies University of Delaware Lewes, Delaware 19958 Dr. Donald R. Moore RSMAS, University of Miami 4600 Rickenbacker Causeway Miami, Florida 33149 V Reunion Nacional de Malacologia y Conquiliologia [Fifth Mexican Malacological Congress] Once again we have the opportunity to meet to learn about advances in investigations in malacology, conchol- ogy, and related fields. This time the meeting will be in the colonial city of Morelia, which the hosts would also like to show you. They want to surprise you! The Universidad Michoacana de San Nicolas de Hi- dalgo and the Sociedad Mexicana de Malacologia will have a large place for the meeting in the Museo Michoacano for the event, which will occur between the 8th and 12th of December 1992. The organizing committee for the meeting includes Dra. Maria Villarroel and Biol. Ezequiel Gonzales of the Uni- versidad Michoacana, and Dra. Martha Rugero and Dr. Antonio Garcia Cubas, of the Universidad de México. Abstracts are due before 14 March 1992, and full papers before 18 July 1992. The correct form for abstracts and full papers will be the same as for the Fourth Congress. Instructions will be sent with the second notice, together with information about the program and all the infor- mation necessary to begin to plan your trip to Morelia. For more information, write to Dra. Villarroel at. the Universidad Michoacana, Apdo. Postal 59-3, C.P. 58021, Morelia, Michoacan, México. Information for Contributors Manuscripts Manuscripts must be typed on white paper, 82” by 11”, and double-spaced throughout (including references, figure legends, footnotes, and tables). If computer generated copy is to be submitted, margins should be ragged right (v.e., not justified). To facilitate the review process, manuscripts, including figures, should be submitted in triplicate. The first mention in the text of the scientific name of a species should be accompanied by the taxonomic authority, including the year, if possible. Underline scientific names and other words to be printed in italics. Metric and Celsius units are to be used. The sequence of manuscript components should be as follows in most cases: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, lit- erature cited, figure legends, figures, footnotes, and tables. The title page should be on a separate sheet and should include the title, author’s name, and address. The abstract should describe in the briefest possible way (normally less than 200 words) the scope, main results, and conclusions of the paper. Literature cited References in the text should be given by the name of the author(s) followed by the date of publication: for one author (Smith, 1951), for two authors (Smith & Jones, 1952), and for more than two (Smith et al., 1953). The “literature cited” section must include all (but not additional) references quoted in the text. References should be listed in alphabetical order and typed on sheets separate from the text. Each citation must be complete, with all journal titles unabbreviated, and in the following form: a) Periodicals Cate, J. M. 1962. On the identifications of five Pacific Mitra. The Veliger 4:132-134. b) Books Yonge, C. M. & T. E. Thompson. 1976. Living marine molluscs. Collins: London. 288 pp. c) Composite works Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135. In: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stanford Univ. Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Figures and plates Figures must be carefully prepared and should be submitted ready for publication. Each should have a short legend, listed on a sheet following the literature cited. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photographs for half-tone plates must be of good quality. They should be trimmed off squarely, arranged into plates, and mounted on suitable drawing board. Where necessary, a scale should be put on the actual figure. Preferably, photographs should be in the desired final size. It is the author’s responsibility that lettering is legible after final reduction (if any) and that lettering size is appropriate to the figure. Charges will be made for necessary alterations. Processing of manuscripts Upon receipt each manuscript is critically evaluated by at least two referees. Based on these evaluations the editor decides on acceptance or rejection. Acceptable manuscripts are returned to the author for consideration of comments and criticisms, and a finalized manuscript is sent to press. The author will receive from the printer two sets of proofs, which should be corrected carefully for printing errors. At this stage, stylistic changes are no longer appropriate, and changes other than the correction of printing errors will be charged to the author at cost. One set of corrected proofs should be returned to the editor. An order forra for the purchase of reprints will accompany proofs. If reprints are desired, they are to be ordered directly from the printer. Send manuscripts, proofs, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. CONTENTS — Continued Two new Vitrinellid species from the Gulf of California, Mexico (Gastropoda: Vitrinellidae) CAROLE M. HERTZ, BARBARA W. MYERS, AND JOYCE GEMMELL ........ Two giant African land snail species spread to Martinique, French West Indies ALBERT R.) MEAD. AND OUISERALGY: 7c seeks iar omen ee ies ee eee Reproductive biology of Vermetus sp. and Dendropoma corrodens (Orbigny, 1842): Two vermetid gastropods from the Southern Caribbean PaTrRIiciIA A. MILOSLAVICH AND PABLO E. PENCHASZADEH .............. BOOKS, PERIODICALS & PAMPHLETS NOTES, INFORMATION & NEWS SlHE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 35 April, 1992 CONTENTS Anterior inhalant currents and pedal feeding in bivalves R. G. B. ReErp, R. F. McMauon, D. O FOIGHIL, AND R. FINNIGAN ..... 93 Remarks on Distorsio constricta (Broderip, 1833) and related species in the eastern Pacific Ocean, with the description of a new species (Gastropoda: Per- sonidae) WiInPRIAM KO EMERSON AND BEDTEY JEAN PIECH =. ...-: 22 52--2¢e-5-5+: 105 A re-evaluation of the ontogeny of Cabestana spengleri (Perry, 1811) (Gastropoda: ‘Tonnoidea: Ranellidae) IPIRAININS, RSTD] DVDs. aeob a tara ha Popes ae aie eben Ce a a nO aE 117 The fossil land snail Helix leidy: Hall & Meek, 1855, a member of a new genus of Humboldtianidae (Gastropoda: Pulmonata) EMMETT E:.VANOFF ANDY BARRY ROTH |... . = .5 26:65 5G 0s oe be bak at bees 122 New occurrences of the malleid bivalve Nayadina (Exputens) from the Eocene of Jamaica, Mexico, and Washington INICEUARD SIE S OUIRES By top trai tae one ie ea eats hci eM s cecayie race alae aie 2AYin © 133 An eastern Pacific Mercenaria and notes on other chionine genera (Bivalvia: Veneridae) INARI NG WARE: 5 Asus oat ae eel he i sal Suey a isdeanlesoeisyave) Soe iat 137 On the validity, authorship, and publication date of the specific name Ancistroch- eirus lesueuru (Cephalopoda: Ancistrocheiridae) CAMB Ace RIS Acts EMA Obra oe een nye Bui s a Sel cs ean eee that hepckaniefs tea cay Such a gi 141 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of January, April, July, and October. Rates for Volume 35 are $28.00 for affiliate members (including domestic mailing charges) and $56.00 for libraries and nonmembers (zn- cluding domestic mailing charges). For subscriptions sent to Canada and Mexico, add US $4.00; for subscriptions sent to addresses outside of North America, add US $8.00, which includes air-expedited delivery. Further membership and subscription infor- mation appears on the inside cover. The Veliger is published by the California Ma- lacozoological Society, Inc., % Museum of Paleontology, University of California, Berkeley, CA 94720. Second Class postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to The Veliger, Museum of Paleon- tology, University of California, Berkeley, CA 94720. ISSN 0042-3211 Number 2 THE VELIGER Scope of the journal The Veliger is open to original papers pertaining to any problem concerned with mol- lusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, evolutionary, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS”; in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Hans Bertsch, National University, Inglewood, California James T. Carlton, University of Oregon Eugene V. Coan, Research Associate, California Academy of Sciences, San Francisco J. Wyatt Durham, University of California, Berkeley William K. Emerson, American Museum of Natural History, New York Terrence M. Gosliner, California Academy of Sciences, San Francisco Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley David R. Lindberg, University of California, Berkeley James H. McLean, Los Angeles County Museum of Natural History Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Washington Barry Roth, Santa Barbara Museum of Natural History Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (no institutional memberships) interested in any aspect of malacology. As an affiliate member, a person may subscribe to The Veliger for US $28.00 (Volume 35), which now includes mailing charges to domestic addresses. There is a one-time membership fee of US $2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription; a reinstatement fee of US $3.00 will be required if membership renewals do not reach the Society on or before October 1 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign members, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $56.00 (Volume 35), which now includes mailing charges to domestic addresses. For subscriptions sent to Canada and Mexico, add US $4.00; for subscriptions sent to addresses outside of North America, add US $8.00, which includes air-expedited delivery. Memberships and subscriptions are by Volume only (January 1 to October 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $25.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: The Veliger, Museum of Paleontology, 3 Earth Sciences Bldg., University of California, Berkeley, CA 94720, USA. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616, USA. THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER. The Veliger 35(2):93-104 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 Anterior Inhalant Currents and Pedal Feeding in Bivalves by R. G. B. REID, R. F. MCMAHON,! D. O FOIGHIL,? anp R. FINNIGAN Biology Department, University of Victoria, Victoria, British Columbia, Canada V8W 2Y2 ' Biology Department, University of Texas at Arlington, Arlington, Texas 76019, USA ? Department of Biological Sciences, Simon Fraser University, Burnaby, British Columbia, Canada Abstract. The association of anterior inhalant currents and pedal feeding is examined in four bivalve species: Corbicula fluminea, Mysella bidentata, Tridacna gigas, and Patinopecten yessoensis. At some time in the life cycle they all employ the foot for food particle collection. In Corbicula and Mysella this is an adult activity, but in 7ridacna and Patinopecten the behavior is confined to the early juvenile. The position of the inhalant current is determined by the dominant food collecting organ. We argue that pedal feeding is a primitive bivalve function. Where the habit is confined to the early postmetamorphic stage, the use of the foot as a feeding organ spans the period between larval and late juvenile suspension feeding before the gills develop into effective filtration organs. It is probable that this kind of feeding behavior is almost universal in juvenile bivalves and common in the adult forms of small species. This has some importance for bivalve aquaculturists in view of high postmetamorphic mortalities in cultured bivalves that may result from inappropriate feeding regimes. INTRODUCTION The anterior inhalant current found in Protobranchia such as Nuculidae and Solemyidae has long been considered a primitive condition (YONGE, 1939). It is also found uni- versally in the Lucinacea (ALLEN, 1958), and in some minute species belonging to the Galeommatacea (= Lep- tonacea), and Veneracea (OLDFIELD, 1955). OCKELMANN & Muus (1978) suggest that in small bivalves, especially juveniles, the anterior inhalant current is the answer to the functional problem of separating the inhalant from the exhalant currents. They note, ‘““No exception is known to the rule that this flow pattern is found in all bivalve species at least for a time in postmetamorphal life.” Descriptions of such currents are provided by ALLEN (1961) for Pandora inaequivaluis (Pandoridae), by MORTIMER (1962) for three Nucula spp. (Protobranchia: Nuculidae), by CaDby (1969) for Macoma balthica (Tellinidae), by BAYNE (1971) for Mytilus edulis (Mytilidae), by AABEL (1983) for Abra alba (Semelidae), and by KING (1986) for Panope abrupta (Sax- icavidae). Thus the potential for acquiring food through an anterior aperture is general amongst juvenile and small bivalves. The particular aspect that we address here is the association of an anterior inhalant current with the actions of the foot and labial palps, to constitute a feeding habit distinct from suspension feeding and deposit feeding modes involving the posterior inhalant siphon and ctenidial fil- tration. In the lucinacean Fimbria fimbriata, the pedal cilia draw an anterior feeding current from the water column through a thin layer of substrate (MORTON, 1979). When the large foot in this species is protracted into the substrate, its cilia collect deposit particles that enter the mantle cavity and pass them to the ventral marginal food grooves of the gills. When the foot is retracted, its coating of mucus-bound particles is wiped off by the secondarily developed pallial palps, the true palps being vestigial. MORTON (1980) also postulates that the anomalodesmacean Pholadomya candida (Pholadomyidae) uses the foot as a pump for drawing fine sediment and detritus into the mantle cavity. While MI- TROPOLSKIJ (1966) reports particle collection by the foot of the fresh-water sphaeracean Pisidium casertanum, LOPEZ & HOLOPAINEN (1987) state that the particles adhering to the foot are all removed as pseudofeces. Nevertheless, the “interstitial suspension-feeding habit” of Prsidium reported by the latter authors is of some interest in the present context. The intertidal bivalve Geloina erosa may fit this category since MORTON (1976) has demonstrated that at low tide it draws burrow water into the mantle cavity through the pedal gape and ingests particles from it. Page 94 BERNARD (1974) notes that in preserved specimens of the carnivorous Cuspidariidae the tip of the foot is often found inserted into the mouth, and he infers that the live prey, once inside the mantle cavity, are ingested with the aid of the foot. Hitherto, these isolated cases were the only known ex- amples of direct nutritional functions of the foot in mature bivalves. However, there are indications of such a role in some studies of juvenile bivalves. ALLEN (1961) observes that the combined use of the ciliated tip of the foot and the relatively large labial palps produce respiratory, cleansing, and feeding currents. MORTIMER (1962) notes particulate food collection by the ciliated foot of early post- metamorphic juveniles of several Nucula spp. BAYNE (1971) observes that the ciliation of the foot in juvenile Mytilus edulis creates an active feeding current that brings food particles into contact with the inner surfaces of the labial palps. At this stage the gills are not sufficiently developed to create a posterior inhalant current or to collect and pass food to the mouth. Cappy (1969) finds that the pedal ciliation has a similar role in juvenile Macoma balthica (Tellinidae). AABEL (1983) points out that when the foot of juvenile Abra alba contracts during the “digging cycle,” adhering particles are brought into the mantle cavity, along with a sudden influx of water and suspended particles. A definitive study by KING (1986) of juveniles of the geoduck Panope abrupta (Saxicavidae = Hiatellidae) re- veals the full significance of pedal feeding. After meta- morphosis at 4 weeks Panope juveniles use “pedal-palp feeding” exclusively through 6 weeks of byssal plantigrade development, and this feeding mode continues to some extent until deep-burrowing behavior is established about 12 weeks after metamorphosis. The juvenile lies on its side, extends the foot posteriorly, and then sweeps it an- teriorly. Deposit material and mineral particles are bound by mucus and carried adorally by pedal cilia. When the foot reaches the limit of its anterior extension it contracts; the mucus-bound particles are loosened and formed into a bolus by the vortex created by the cilia of the inner faces of the labial palps. Passive contact between the labial palps and the base of the foot, and muscular flicking motions of the palps also free the mucus-bound particles. Occasionally the foot is reflexed and the propodium stuffs the food bolus into the esophagus. This food collecting behavior is ste- reotyped and distinct from digging. If there is no suitable food or substrate, the animals locomote for prolonged periods. Pedal-palp feeding fills the nutritional gap be- tween the suspension feeding mode of the veliger and the suspension feeding mode of the adult. Five weeks after metamorphosis the gill bars extend and reflex, forming a ciliated food groove at their ventral tips, and at this stage they are able to create a weak posterior inhalant current. However, there is still no functional connection between the gills and the mouth, and the posterior current is dis- rupted by the large foot. Not until 12 weeks after meta- morphosis are the gills sufficiently large to filter efficiently, The Veliger, Vol. 35, No. 2 by which time they are functionally connected, via food grooves, with the mouth. The use of oysters as type models for bivalve studies has probably been responsible for the general assumption that the bivalve veliger suspension feeding mode is immediately replaced by the adult filter feeding mode. Oysters rapidly develop functional gills within a day of settlement and attachment. Boring bivalves such as the Teredinidae and Pholadidae develop inhalant siphons at the pediveliger stage and rapidly engage in the mature feeding mode as soon as they form their burrows (WERNER, 1939; JORGEN- SEN, 1946; TURNER & JOHNSON, 1969; CULLINEY & TURNER, 1976). However, there are enough hints in the literature cited above to conclude that the pedal-palp feed- ing behavior of Panope juveniles is widespread, and that oysters and borers are exceptions to a general rule. The present report discusses four cases of pedal feeding in bivalves. The first is in Corbicula fluminea (Corbiculi- dae), a fresh-water species whose members attain 4-5 cm in shell length (McManuon, 1983). The second is Mysella bidentata (Leptonidae), a marine, benthic bivalve that does not exceed 5 mm in shell length. The final two examples are the giant clam 777dacna gigas (Tridacnidae), and the Japanese scallop Patinopecten yessoensis (Pectinidae), both marine epifaunal organisms, where pedal feeding occurs only in the early juveniles. These studies were initiated independently by McMahon and Reid (Corbicula); O Foighil (Mysella); Reid (Tridacna); and Finnigan and O Foighil (Patinopecten). Thus the experimental and obser- vational approaches vary from one species to another. We have collaborated in this joint report in order to emphasize the variety of bivalve types that exploit similar modes of pedal feeding. MATERIALS ano METHODS Corbicula Specimens of Corbicula fluminea (Miller) were collected from the Clear Fork of the Trinity River below Benbrook Lake in Tarrant County, Texas. In preparation for feeding experiments some specimens were kept in filtered water for either 8 or 24 days. This conditioning was for the purpose of a study of the digestive cycle that will be re- ported elsewhere but is nevertheless relevant to our present report. The suspension feeding mode, consisting of filtration of the inhalant siphonal current, was examined by placing the specimens in shallow pans to which cultured Chlam- ydomonas rheinhardi was added. The deposit feeding mode was examined under a variety of conditions. In the first, freshly collected detritus was thinly layered in shallow pans before the animals were introduced. In other instances the pans were thinly layered with ashed and rinsed sand and silt from the biotope and then with soaked G10 Sephadex beads. The mineral particles ranged from 10 to 600 um in diameter and the Sephadex beads ranged from 40 to R. G. B. Reid e¢ al., 1992 120 um but were mainly in the 60 to 80 um range. Corbicula specimens were placed on the deposit layer, care being taken not to resuspend the Sephadex beads. Pedal feeding was also observed when Corbicula specimens were placed in deep gravel in narrow glass tanks. Stomach contents were examined microscopically 1-12 hr after active feeding had commenced. Ciliary currents in the mantle cavity were examined after removal of the upper valve and mantle. Particles used included 280 mesh carborundum (35-75 wm), alumina (ca. 30 wm), and G10 Sephadex beads (40-120 um). Mysella Specimens of Mysella bidentata (Montagu) were col- lected by Van Veen grab from a depth of 18 m in inner Galway Bay, Ireland. They were placed in aerated aquaria maintained at a temperature of 12°C containing a layer of the sandy mud from the biotope, and their behavior was examined through a dissecting microscope within 3 days of sampling. Some specimens fed at the surface of the sediment and others that had dug shallow burrows could be seen through the sides of the aquaria. Stomach contents were determined by histological examination of specimens fixed within 1 hr of sampling. Tridacna Two specimens of early juvenile 77ridacna gigas Linné were examined at the James Cook University Giant Clam Aquaculture Project on Orpheus Island, Great Barrier Reef, Queensland. These were 1 mm and 1.4 mm in shell length and had undergone metamorphosis 6 weeks before they were examined. The specimens were studied under two conditions. In the first case they were placed in seawater in a petri dish along with some of the deposit material from the bottom of the rearing tank. This material consisted of settled phy- toplankton and flocculent organic matter that formed loose clumps, leaving most of the glass bottom of the dish ex- posed. The locomotion and nature of the inhalant currents were observed microscopically. In the second case a 5 mm layer of surface silt, sand, and deposit material from the lower mangrove was placed in a petri dish that was then filled with seawater. After the substrate had settled the behavior of the juveniles was observed microscopically. Patinopecten Spat of Patinopecten yessoensis (Jay) that had been newly set on kinram (a filamentous substrate used for spat col- lection and early development of juveniles) were supplied by the Pacific Biological Station, Nanaimo, B.C. The ju- veniles were between 256 wm and 272 mum in length at metamorphosis. A constant flow of seawater (9.7°C, 32%o0 salinity) was filtered through 125 um nitex and then through 25 um nitex. Once a day the water entering the tanks was Page 95 shut off and Chaetoceros calcitrans and C. gracilis were added to the aquarium at a concentration of 40,000 cells/ mL. Spat were either observed on the kinram with a dis- secting microscope and a videomicrographic camera, or were removed from the kinram and placed in well slides to observe their behavior with an inverted microscope with a cool light source. Some detritus derived from the algal food, and some suspended algae, were present in the sam- ples observed with the inverted microscope. Observations were made periodically until the juveniles had reached a size of 470 um. RESULTS Corbicula In the horizontal position in our experimental trays where the layer of particulate matter was only a few mil- limeters thick, the animals extended the feet horizontally, with the ventral margin somewhat recurved as shown in Figure 1a. Under these conditions only the cilia of the lower side of the foot are active in particle collection. Particles placed on the upper side of the foot are not trans- ported. As the substrate is conveyed to the mantle cavity, the ventral margin of the foot levers the animal across the surface, with the umbones in the leading position, at rates of up to 1 cm/min. In the horizontal position pseudofeces are formed at the lower shell-mantle margins and are discarded in a mucus-bound string. Observations with Sephadex beads demonstrated that in this position deposit feeding effectively transported particles that were ulti- mately found in the stomach. Under the conditions of our experiments there was no resuspension of Sephadex nor uptake through the inhalant siphon. However, the route of uptake is problematical due to the presence of the pedal rejectory tract and the opacity of the shell of Corbicula. Some of the denser clumps of material are brushed off when they touch the edges of the valves and mantle mar- ginal folds. Pedally collected particles that continue into the mantle cavity somehow escape rejection and either are caught up in the marginal ctenidial food grooves (some Sephadex beads are found here in dissected specimens), or are brought into direct contact with the labial palps by the contraction and forward extension of the foot. The rejec- tory tract may be activated only when large quantities of dense matter have built up. The trauma of dissection may activate the rejectory tract when “‘half-shell’ animals are examined. When this is done the pedal collecting mech- anism is deactivated, which may be why the mechanism has been largely overlooked in an otherwise widely ob- served species. Another reason might be that starved Cor- bicula may take a number of hours to begin to feed pedally. When Corbicula fluminea is in a vertical position, with the foot extended down into the substrate, ciliary currents on both sides of the foot conduct particles dorsally into the mantle cavity (Figure 1b). Some of the large particles and mucus-bound clumps are brushed off on to the mantle Page 96 __5mm_, The Veliger, Vol. 35, No. 2 Co Figure 1 Pedal feeding in Corbicula. a. Feeding on a hard surface with a thin sand and detritus layer; large arrow indicates direction of movement of animal. b. Feeding while partially buried in sediment. AG = acceptance tract of gill food groove; D = debris displaced by advance of animal, together with mucus-bound pseudofeces; E = exhalant siphon; F = foot; G = gill; I = inhalant siphon; L = labial palp; R = rejectory groove; V = ventral pedal feeding tract. edges, which produce more mucus that binds this rejected material into pseudofeces. These are left behind in two strings as the animal advances. Animals placed on deep sediment in laboratory aquaria locomote rapidly through the substrate for prolonged periods. In freshly collected field specimens we have repeatedly observed massive ac- cumulations of detritus in an anterior position just below the marginal gill food grooves. We believe that these ac- cumulations result from pedal feeding during locomotion. In the central portion of the foot, there is a narrow anteroposterior band of ciliated grooves. When specimens were observed with the upper valve and mantle tissue removed, the foot was contracted into the mantle cavity and the anteroposterior grooves acted as a rejectory tract, carrying material towards the opening of the inhalant si- phon (Figure 1b; see also BRITTON & MorTON, 1982). Scanning electron microscopy of the contracted foot re- vealed a uniform ciliation over the surface of the organ, with larger cilia near the ventral margin. The rejectory tract is furnished with ciliary tufts similar to those found in the siphons by KRAEMER (1983), who postulates that they are mechanoreceptors. The stomachs of specimens of Corbicula examined with- in 2 hr of collection from the field contained relatively large volumes of green-brown fluid. Algal material in- cluded numerous chains and individual cells of Nitzschia, Melosira, Scenedesmus, and Tabellaria spp., which can be both benthic and planktonic, together with numerous planktonic coccoids and chlorococcum spores. Individuals of all of these algal types fall within a diameter range of R. G. B. Reid et al., 1992 15-50 um. Some of these algal cells were bright green, others were brown, indicating that they either had been detrital or had already been partly digested. The majority of gastric mineral particles were about 20 um in diameter, with a few up to 100 um. Some detrital particles, averaging 50 um, derived from crustacean exoskeletons, were also discovered. Mysella The typical feeding method in Mysella bidentata pro- ceeds as follows. While lying on its side at the substrate surface the animal extends the foot posteriorly to come into contact with the sediment (Figure 2a). The foot is then protracted and moved anteriorly. During this process particles of deposit matter can be seen to ascend the foot. Upon reaching the anterior limit of its extension, the foot is withdrawn. The mantle marginal folds come into contact with the foot at this point and the larger mineral particles are brushed off. Foot protraction and withdrawal are ac- companied by regular valve adductions that expel pseu- dofeces through the exhalant siphon. During feeding a depression is excavated in the substrate by the foot; larger rejected particles are accumulated on the anteroventral valve margins and small rejected particles are deposited posterior to the exhalant opening (Figure 2b). This be- havior, which we will call pedal sweep-feeding, typically lasts for about 1 min and may be repeated for periods of up to 10-15 min. Each pedal sweep-feeding cycle is punc- tuated by a resting period of up to 2 min. This behavior is distinct from locomotion and burrowing behavior in that the individual does not make progress through the sedi- ment. Occasionally, during these observations, the foot could be seen, through the translucent shell, to make contact with the mobile labial palps. This could also be confirmed by removing one valve and the underlying mantle tissue. The vigorously beating cilia of the palps brush the foot, and a bolus of material builds up and can be seen rotating between the palps. Particles of detritus also adhere to the foot during lo- comotion and the animal leaves much of this material behind in a pair of mucus-bound strings of rejecta. It is conceivable that some nutritive matter obtained during locomotion is conducted by the cilia of the foot and accepted by the labial palps. The stomach contents of field-collected specimens of Mysella bidentata consisted largely of detritus and benthic diatoms, with a few forameniferans and flagellates. Most of the diatoms were less than 20 um in length. Tridacna When placed in a petri dish with some flocculent organic debris, locomotion of juvenile 7ridacna gigas proceeds im- mediately, with the anterior protrusion of the foot, and the adhesion of the ventral surface of its anterior portion to the glass (Figure 3a). Contraction of the foot then draws Page 97 Figure 2 a. Pedal feeding in Mysella; large arrows indicate the posteroan- terior motion of the foot. b. Locomotion and pedal currents in Mysella; large arrows indicate direction of movement. D = trail of material displaced by animal together with pseudofeces; DE = depression in substrate created by pedal feeding; F = foot; L = labial palp; P = particles brushed from foot adhering to valve margins; PS = pseudofeces; T = tip of foot appressed to labial palps. the animal forward. During such locomotion an anterior current flows into the mantle cavity through the byssal- pedal gape. Small clumps of detritus are seen to enter the mantle cavity, and through the translucent valves a rotating bolus can be seen in the umbonal region, adjacent to the labial palps. On such a hard surface, locomotion continues until the organism reaches the water-air interface. When placed in a sandy substrate with an average par- ticle size of 100 um, the juvenile lies on its side and engages in pedal sweep-feeding (Figure 3b). The foot is protruded posteriorly and is then swept forward with a slight rotating movement. Strings of mucus are secreted by the foot, and some, with adhering sediment particles, are shed and left behind. At the anterior limit of the foot’s protraction the foot is retracted. This behavior, which lasts about 5 sec, was seen to be repeated 10 times, followed by a resting period of several minutes, after which the activity re- commenced. A second behavioral mode in sediment involves a starting position in which the juvenile is vertical with the siphonal tissues uppermost. The foot is then protruded down into Page 98 The Veliger, Vol. 35, No. 2 Figure 3 Various putative feeding modes of juvenile Tridacna. a. Particle acceptance during forward locomotion over a hard surface. b. Anteroposterior pedal feeding at substrate surface. c. Putative pedal feeding by probing substrate. AF = acceptance tract of foot; B = bolus of food particles in region of mouth; F = foot; J = jet of water created by valve adduction with siphons closed; PI = pedal inhalant current through pedal-byssal gape; RF = retracted position of foot. the sediment (Figure 3c). Partial burrowing results from phasic adduction and the expulsion of jets of water through the byssal-pedal gape. The siphonal openings are con- tracted during this maneuver. It is possible that the pedal cilia gather particulate food in this mode. Patinopecten Newly metamorphosed juvenile Patinopecten yessoensis are approximately 250 wm in valve length and have 3 or 4 small, lobular primary gill filaments that increase in size and number during subsequent ontogeny. By one week after metamorphosis the juveniles exceed 400 um, the gill filaments become physically associated to form supra- and infrabranchial chambers and effective suspension feeding commences. Prior to this, however, if suspended food is available to the juveniles, some collection and ingestion of particles that contact the mantle surfaces does occur. Two pairs of well-developed ciliated labial palps flank the foot and are capable of passing impinging particles to the mouth. The proportionately large foot is well ciliated on the ventral surface, and a prominent tuft of long cilia is present ReiGe bs Reidtetialy 992 Page 99 at the tip of the propodium. The posterior spur or heel of the foot has a duct from which byssus can be secreted. Juveniles locomote actively by means of the foot during the first week of benthic life (Figure 4a), but the level of such activity declines as development proceeds and a more sessile, byssally attached phase ensues (Figure 5). The small juvenile protracts the foot and advances by means of the ventral cilia, the valves held vertically, with the umbones pointing in the direction of movement. In later juveniles the foot is protracted, adheres to the substrate, and then muscular contraction pulls the rest of the body forward. During both types of locomotion the foot remains extended outside the valves and is not withdrawn to contact the labial palps. The tip of the foot with its terminal tuft of cilia is very active and may have mechanosensory and chemosensory functions. In addition to being used for locomotion and byssal attachment, the foot is used by juvenile Patinopecten for pedal probe-feeding. This is distinct from locomotory be- havior: it does not cause forward progression of the animal. In addition, after the foot has been protracted to contact the substrate, it is withdrawn into the mantle cavity where its tip and its adhering particles are brushed against, or occasionally pushed convulsively between, the palps (Fig- ure 4b, c). When juveniles are placed in petri dishes con- taining a deposit of phytoplankton, they begin to feed on this deposit using the foot in such a manner. Particles that are transferred to the palp surfaces are carried by cilia to the mouth and ingested. Several phasic adductions of the valves may occur during this process. Observations of bysally attached juveniles on fouled kin- ram (Figure 5) reveal that individuals can engage in pedal- palp feeding without breaking the anchoring byssal thread. Typically the foot is progressively applied to all of the available surface through 360° of the attachment point. This is achieved by a characteristic rotational movement of the juvenile around the byssal thread, the foot being swept over portions of the substrate and withdrawn to contact the palps, before again being protracted in another direction. On occasion, byssally attached juveniles are seen to open the valves and contract the byssus retractor, bring- ing the labial palps into contact with the kinram. Histo- logical examination of juveniles placed on benthic diatom mats reveal that they are capable of ingesting detrital mat- ter and benthic diatoms (Figure 5d). Small juveniles (less than 350 wm) may feed predominantly by this method, whereas in larger juveniles (350-500 um) the relative im- portance of pedal feeding, in comparison with ctenidial filtration of suspended particular food, is not clear. In the large juveniles there are eight or more pairs of gill filaments that are partly reflexed to divide the mantle cavity into supra- and infrabranchial chambers. A ciliary flow of water enters posteroventrally and is driven by the lateral ctenidial cilia inte the suprabranchial chamber and thence exits posteriorly. Suspended particles that come into contact with the gill filaments are passed dorsally and then 120um_, Figure 4 Early juvenile of Patinopecten. a. Forward pedal locomotion; the “heel” of the foot contains the byssal gland. b. Foot probes into detrital material. c. Foot is retracted and adhering detritus is inserted between labial palps. GB = gill bar; L = labial palp; PD = particles of detritus. carried in a dorsal ciliated tract of the mantle towards the labial palps. Some of this material is accepted by the palp surfaces and ingested, but the rest forms a loose clump in the region of the pedal gape. It may then be expelled by phasic adduction, or picked up by the foot and returned to the palps. The dorsal surface of the foot was seen on one occasion to return potential food that had been lost by the palps towards the mouth. Thus pedal-palp feeding continues to supplement the early suspension-feeding hab- it. DISCUSSION YONGE (1947) concludes that bivalve ctenidia cannot be effective filtration organs until they have reflexed and formed food grooves. ALLEN (1961) suggests that the action of the foot in juvenile Pandora inaequivalvis in creating an anterior feeding current might be found in other juveniles, since there would otherwise be no feeding mechanism be- tween metamorphosis and ctenidial maturation. This is borne out by various reports cited in the Introduction above; in particular, KING (1986) demonstrates that in early Pan- ope juveniles “pedal-palp” feeding bridges that nutritional gap. There are several variations of this habit: Page 100 The Veliger, Vol. 35, No. 2 R. G. B. Reid e¢ al., 1992 (1) Locomotory pedal feeding: the foot comes into contact with deposit material; pedal cilia collect particles and pass them to the labial palps or oral region. (2) Pedal probe-feeding: the animal is stationary and the propodium probes for deposit material that adheres to mucus. Reflection of the foot brings the food into contact with the oral region. (3) Pedal sweep-feeding: the stereotyped posteroan- terior pedal feeding cycle. In most cases relatively large labial palps compact a rotating food bolus that is period- ically ingested. The palps have no sorting function other than rejection of large, dense particles by centrifugation. (4) Interstitial pedal feeding: the foot is thrust into a cavity that it has created in the substrate. Pedal cilia draw resuspended deposit material towards the mouth and labial palps, both on the surface of the foot and in an adoral current. We prefer this term to “interstitial suspension feeding” (LOPEZ & HOLOPAINEN, 1987). In Corbicula fluminea, variations of locomotory pedal feeding and pedal probe-feeding occur when the organism is lying on a hard surface with a thin deposit film, when it is partially buried in sediment, and probably while mov- ing through sediment. Pedal sweep-feeding is absent. There is interstitial pedal feeding, the cilia of both sides of the foot being involved when the organ is protracted into the substrate. The distinction between deposit feeding and sus- pension feeding in such cases becomes blurred. The ex- tensive wandering of Corbicula specimens held in mud in aquaria indicates that locomotory pedal feeding permits this bivalve to continuously work and recycle the organi- cally rich surface layers of sediment. Way et al. (1990), in a report on the dynamics of filter feeding in Corbicula, comment on a pedal ciliary tract that may be capable of detritus feeding, and they conclude that the presence of food particles larger than 20 um in the stomachs of their specimens may be due to other feeding methods than those involving gill filtration of suspended particles. The large amount of detritus that is routinely found in freshly col- lected field specimens is strong, if circumstantial, evidence of the importance of pedal feeding. Corbicula has excited curiosity concerning its ability to grow rapidly under con- ditions that do not appear to have the energetic resources to sustain such growth. Its success has been attributed to its speed of locomotion, adaptability, and the accommo- dating character of its pallial functional morphology (KRAEMER, 1977). These features, along with pedal feed- ing, allow the bivalve to maximize the uptake of particulate matter even when suspended food material is virtually absent. Page 101 In isolation, the case of pedal feeding in Corbicula may seem as exceptional as that of Fimbria. However, it is likely that a number of other mature bivalves, usually minute types, pedomorphically retain a combination of pedal feed- ing and anterior inhalant currents. The retention of a proportionately large foot could be pedomorphic. POPHAM (1940) notes that the proportionately large labial palps of Mysella indicate a significant role in direct particle collec- tion as well as sorting. OCKELMANN & Muus (1978) are of the opinion that feeding in Mysella bidentata is affected by the ciliary resuspension of sediment particles by the immobile extended foot, in the manner proposed by LOPEZ & HOLOPAINEN (1987) for Pisidium. But the pedal sweep- feeding of Mysella described above is a consistent habit that is similar to the mechanism found in juveniles of Panope and Tridacna. The behavior of juvenile Abra alba may also conform to such a pattern (AABEL, 1983). The juvenile pedal induction of an anterior inhalant, feeding current is preserved pedomorphically in minute bivalves such as Lasaea rubra (Galeommatacea) and Turtonia minu- ta (Veneracea) (OLDFIELD, 1955). Current research by the first author and K. Bartlett reveals pedal feeding in the small carditid species Miodontiscus prolongatus, and in the minute mytilid Modtolaria taylori (= Musculus pygmaeus). (We follow the opinion of Coan, personal communication, on these taxonomic revisions.) Moreover, juveniles of the venerid Manila clam Tapes philippinarum are probably exclusively pedal feeders until they attain the size of about 500 um, an observation that has implications for nursery technology for this important commercial species (Reid, Bartlett & Lindsay, unpublished data). The behavioral and neurophysiological capacity for var- ious forms of pedal feeding exist in juvenile T7idacna gigas. Whether they are important in bridging the gap between metamorphosis and the establishment of ctenididal filtra- tion, together with a functional symbiosis, remains to be determined. Peter Lee of James Cook University has un- dertaken a fuller study of the biology of early juveniles of this species. The nutritive significance of pedal feeding in giant clam juveniles is complicated by the existence of symbiotic zooxanthellae. The symbiosis is established in Tridacna gigas within 11 days of metamorphosis (HEs- LINGA et al., 1984). Fitt (personal communication) surmises that the symbiosis may not contribute significantly to the economy of the early juvenile since much of the photosyn- thetic energy goes initially to the reproduction of the zoo- xanthellae. FITT et al. (1984) conclude that high mortality at the transition from pediveliger to juvenile is likely due to changes in the mode of acquiring food, and the nutri- Figure 5 Photomicrographs of Patinopecten juveniles. a. Foot retracted, showing “heel” containing byssal gland. b. Foot probing while juvenile lies on side on hard surface. c. Foot probing fouled surface; four pairs of gill bars are present. d. Stomach contents of juvenile after pedally feeding on benthic diatom mat; arrows indicate diatoms. e. Foot retracted and appressed to labial palps. f. Juveniles byssally attached to fouled kinram. F = foot; GB = gill bar; H = “heel” containing byssus gland; L = labial palp. Page 102 M 90 um, Figure 6. Drawings of generalized Nucula juveniles (after MORTIMER, 1962). a. Pre-ctenidial stage. b. Proto-palpal stage. AC = anterior in- halant current; EC = exhalant current; F = foot; GA = gill anlage; GB = gill bar; LF = lateral flange of foot; LU = upper labial palp; M = position of mouth; PA = pallial arc; PC = posterior inhalant current. tional requirement at this stage would not be likely to differ from the non-symbiotic clams. The case of Patinopecten yessoensis is a caution against over-generalization since pedal sweep-feeding does not ex- ist, and the early juveniles collect some suspended particles even before an efficient ctenidial morphology is in place (O FoIGHIL et al., 1990). The juveniles usually employ pedal probe-feeding, and the foot also creates an anterior inhalant current. Similar behavior has been observed in early juveniles of the purple-hinged rock scallop Crassa- doma gigantea (H. Miller, University of Victoria, personal communication). The absence of pedal sweep-feeding from this scallop juvenile may be correlated with the require- ment of byssal attachment to a stable substrate for epi- faunal inhabitants of turbulent water. Panope abrupta at- The Veliger, Vol. 35, No. 2 taches to several large substrate components by byssus threads, but these are long enough (up to 10 cm) not to impede extensive working of the substrate over a relatively wide area by pedal feeding (KING, 1986). The persistence in the scallop and giant clam of a portion of the spectrum of pedal feeding found in burrowing clams supports the hypothesis that pedal feeding is a primitive component of bivalve functional morphology (KING, 1986). Therefore the feeding behavior of Nucula, which YONGE (1939, 1959) proposes as a model for the primitive bivalve, is of special interest. MORTIMER (1962) studied Nucula sulcata, N. tenuis, and N. turgida. These all acquire pedal feeding ciliary tracts at metamorphosis, before the labial palps and ctenidia have developed. The foot ciliation is supplemented by a pair of “pallial arcs,” narrow, ciliated pathways that pass from the gill rudiments across the dorsal mantle region towards the pedal acceptance tracts (Figure 6a). These species also demonstrate pedal sweep-feeding, described as an in- termittent, back and forth motion, followed by foot re- traction, that the author emphasizes as distinct from lo- comotion. The rudimentary palps provide a ciliary catchment area that directs pedally collected food to the mouth. As soon as the inner gill bars form, they make direct contact with the acceptance tracts of the foot, and particles filtered by them from the mantle water are carried by adoral ctenidial cilia on to the foot. At this stage, until the mature labial palp apparatus is developed, the ctenidia are the most important food-collecting organs. The stron- gest inhalant current during this period is posterior. Thus, Nucula proves the rule that the inhalant and exhalant currents in minute bivalves must be separated into anterior and posterior regions: in the intermediate stages of juvenile development the ctenidia are more important than the palps for feeding, and they create a posterior inhalant current while the mantle margins establish a posterior exhalant current (MORTIMER, 1962). Therefore, even in very small organisms, posterior inhalant and exhalant cur- rents can co-exist. Parsimoniously, the position of the in- halant current should be considered to be an effect of the position and functional morphology of the major feeding organs. Early postmetamorphic pedal feeding is the only thing that Nucula has in common with the other bivalves con- sidered in this study. Nucula juveniles begin their pedal feeding before the labial palps are formed. Therefore, large labial palps are not an essential concomitant of pedal feed- ing. Moreover, the ctenidia bridge part of the gap between metamorphosis and the development of the complex food collecting apparatus of the mature form. STASEK (1961), MORTIMER (1962), and REID & BRAND (1986) agree that the prominent labial palp lamellae of protobranchs arose as primitive suspension feeding organs, and the elongation of the terminal pair of lamellar ridges as the detritus- collecting palp proboscides was a later specialization of the Nuculacea. The gills of mature Nuculidae, as STASEK (1961) notes, retain some particle-collecting role. Thus RaGe Baked etal 1992 nuculids, apart from their early pedal feeding, are excep- tional in both the juvenile and mature forms. The early Nucula juvenile, rather than the adult, could be considered as a model for the primitive minute bivalve. From it, one evolutionary line opted for elaboration of the labial palps for food collection, with a proboscidial deposit feeding habit as the final specialization. The other divergent line opted for specialization of the ctenidia as suspension-filtration organs. In any case pedal feeding remains the only con- sistent common feature. YONGE (1959) is of the opinion that the molluscan an- cestor of the bivalves locomoted with the foot and used small, undifferentiated labial palps to collect and sort de- tritus. The addition of a ciliary pedal feeding function produces the condition found in many of the juvenile and small bivalves that we describe above and that ALLEN (1985) similarly proposes as a primitive bivalve feeding mode. What further light might the examination of pedal feed- ing shed on the original habit and habitat of bivalves: were they infaunal or epifaunal? Contrary to conventional wis- dom the bivalve form is not a necessary correlation of movement through soft substrates. Many bivalves are epi- faunal and other bivalved invertebrates such as brachio- pods, ostracod crustaceans, and gastropod juliids are epi- faunal or planktonic. Most of the range of pedal feeding behaviors functions effectively at the surface of the sub- strate, whether it be particulate or solid. Therefore, prim- itive bivalves may have been epifaunal (STASEK, 1961; ALLEN, 1985; REID & BRAND, 1986), and the infaunal nature of many Recent mature bivalves may be irrelevant to the discussion of the origins of the class. ACKNOWLEDGMENTS Research at the University of Texas at Arlington was supported by a UTA Field Visiting Professorship held by R. G. B. Reid. We are grateful for the assistance of John Barnet with the Corbicula work. Work on Tridacna was supported by an NSERC Operating Grant and a Uni- versity of Victoria travel grant to R. G. B. Reid. We thank Geoff and Sandy Charles and Jeremy Barker for their assistance at the Orpheus Island (Great Barrier Reef) Giant Clam Research Laboratory. We are grateful to Dawna Brand for her assistance with histology and elec- tron microscopy. Dr. J. McInerney provided facilities at Bamfield Marine Station for the observations of Patino- pecten. Neil Bourne’s scallop research group at the Pacific Biological Station, Nanaimo, B.C., provided scallop ju- veniles. In the early stages of our research, discussions with Jeff King and Rick Gustafson were very useful. LITERATURE CITED AABEL, J. P. 1983. Morphology and function in postmeta- morphal Abra alba (Bivalvia: Tellinacea). Sarsia 68:213-219. ALLEN, J. A. 1958. On the basic form and adaptations to habitat Page 103 in the Lucinacea (Eulamellibranchia). Philosophical Trans- actions of the Royal Society, B 241:421-484. ALLEN, J. A. 1961. The development of Pandora inaequivaluis (Linné). Journal of Embryology and Experimental Physi- ology 9:252-268. ALLEN, J. A. 1985. The recent Bivalvia: their form and func- tion. Pp. 337-403. In: E. R. Trueman & M. R. Clarke (eds.), The Mollusca, X. Academic Press: New York. BayNE, B. L. 1971. Some morphological changes that occur at the metamorphosis of the larvae of Mytilus edulis. Pp. 259- 280. Jn: D. J. Crisp (ed.), Fourth European Marine Biology Symposium. Cambridge University Press: Cambridge, U.K. BERNARD, F. R. 1974. Septibranchs of the eastern Pacific (Bi- valvia: Anomalodesmata). Allan Hancock Monographs in Marine Biology 8:279 pp. BRITTON, J. C. & B. MorTON. 1982. A dissection guide, field and laboratory manual for the introduced bivalve Corbicula fluminea. Malacological Review Supplement 3:82 pp. Cappy, J. F. 1969. Development of mantle organs, feeding and locomotion in postlarval Macoma balthica (L.) (Lamel- libranchiata). Canadian Journal of Zoology 47:609-617. CULLINEY, J. L. & R. D. TURNER. 1976. Larval development of the deep water wood boring bivalve Xylophaga atlantica Richards (Mollusca, Bivalvia, Pholadidae). Ophelia 15:149- 161. Fitt, W. K., C. R. FIsHeER & R. K. TRENCH. 1984. Larval biology of tridacnid clams. Aquaculture 39:181-195. HESLINGA, G. A., F. E. PERRON & O. ORAK. 1984. Mass culture of giant clams (F. Tridacnidae) in Palau. Aquacul- ture 39:197-215. JORGENSEN, C.B. 1946. Lamellibranchia. Jn: G. Thorsen (ed.), Reproduction and Larval Development of Danish Marine Bottom Invertebrates. Meddelelser Fra Kommissionen for Danmarks Fiskeri Og Havnandersogelser Serie: Plankton 4:277-311. KING, J. J. 1986. Juvenile feeding ontogeny of the geoduck Panope abrupta (Bivalvia: Saxicavacea), and comparative on- togeny and evolution of feeding in bivalves. M.Sc. Thesis, University of Victoria, Victoria. KRAEMER, L. R. 1977. Aspects of the functional morphology of the mantle/shell and mantle/gill complex of Corbicula (Bivalvia: Sphaeriacea: Corbiculidae). Bulletin of the Amer- ican Malacological Union 1977:25-31. KRAEMER, L. R. 1983. Comparative functional morphology of the cilia of Corbicula (Bivalvia: Corbiculidae): possible cri- teria for effector and putative sensory types. American Mal- acological Bulletin 1:13-20. Lopgz, G. R. & I. J. HOLOPAINEN. 1987. Interstitial suspen- sion-feeding by Pisidium spp. (Pisidiidae: Bivalvia): a new guild in the lentic benthos? American Malacological Bulletin 5:21-30. McManuon, R. F. 1983. Ecology of an invasive pest bivalve, Corbicula. Pp. 505-561. In: W. D. Russell-Hunter (ed.), The Mollusca. Vol. 6. Academic Press: New York. MITROPOLSKIJ, V. I. 1966. O mekhanizme filtratsii i o pitanii steriid (Mollusca, Lamellibranchia). Trudy Instituta Biolo- gii Vodokhranilishcha Akademii Nauk SSSR 12:129-133. Mortimer, J. E. 1962. A comparative study of post-larval feeding mechanisms in the Bivalvia. Ph.D. Thesis, Glasgow University, Glasgow, Scotland. Morton, B. 1976. The biology and functional morphology of the Southeast Asian mangrove bivalve, Polymesoda (Geloina) erosa (Solander, 1786) (Bivalvia: Corbiculidae). Canadian Journal of Zoology 54:482-500. Morton, B. 1979. The biology and functional morphology of Page 104 the coral-sand bivalve Fimbria fimbriata (Linnaeus 1758). Records of the Australian Museum 32:389-420. Morton, B. 1980. The anatomy of the “living fossil” Phola- domya candida Sowerby 1823 (Mollusca: Bivalvia: Anom- alodesmata). Videnskabelige Meddelelser fra Dansk Natur- historik Forening (Kobenhaven) 142:7-102. OCKELMANN, K. W. & K. Muus. 1978. The biology, ecology and behaviour of the bivalve Mysella bidentata (Montagu). Ophelia 17:1-93. O FoicuiL, D., B. KINGZETT, G. O FoIcHIL & N. Bourne. 1990. Growth and survival of juvenile Japanese scallops (Patinopecten yessoensis) in nursery culture. Journal of Shell- fish Research 9:135-144. OLDFIELD, E. 1955. Observations on the anatomy and mode of life of Lasaea rubra (Montagu) and Turtonia minuta (Fa- bricius). Proceedings of the Malacological Society of London 31:226-249. PopHaM, M. L. 1940. The mantle cavity of some of the Ery- cinidae, Montacutidae and Galeommatidae, with special ref- erence to the ciliary mechanisms. Journal of the Marine Biological Association of the U.K. 24:549-587. REID, R. G. B. & D. G. BRAND. 1986. Sulfide-oxidizing sym- biosis in lucinaceans: implications for bivalve evolution. The Veliger 29:3-24. STASEK, C. R. 1961. The ciliation and function of the labial palps of Acila castrensis (Protobranchia, Nuculidae) with an evaluation of the role of the protobranch organs of feeding The Veliger, Vol. 35, No. 2 in the evolution of the Bivalvia. Proceedings of the Zoological Society of London 137:511-538. TuRNER, R. D. & A. C. JOHNSON. 1969. Biology of marine wood-boring molluscs. Pp. 114-129. In: E. B. G. Jones & S. K. Eltringham (eds.), Marine Borers, Fungi and Fouling Organisms of Wood. Organization For Economic Co-op- eration and Development: Seattle. Way, C. M., D. J. HORNBACH, C. A. MILLER-WaAY, B. S. PAYNE & A. C. MILLER. 1990. Dynamics of filter feeding in Corbicula fluminea (Bivalvia: Corbiculidae). Canadian Jour- nal of Zoology 68:115-120. WERNER, B. 1939. Uber die Entwiklung und Artunterscheing von Muschellarven des Nordseeplanktons, unter besonder Berucksichtigung der Schalenentwicklung. Zoologisches Jahrbucher, Abteilen fur Anatomie und Ontogenie der Tiere 66:1-54. YONGE, C. M. 1939. The protobranchiate Mollusca: a func- tional interpretation of their structure and evolution. Philo- sophical Transactions of the Royal Society, B 230:70-147. YONGE, C. M. 1947. The pallial organs in the aspidobranch Gastropoda and their evolution throughout the Mollusca. Philosophical Transactions of the Royal Society, B 232:443- 518. YONGE, C. M. 1959. The status of the Protobranchia in the bivalve Mollusca. Proceedings of the Malacological Society of London 33:210-214. The Veliger 35(2):105-116 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 Remarks on Distorsio constricta (Broderip, 1833) and Related Species in the Eastern Pacific Ocean, with the Description of a New Species (Gastropoda: Personidae) by WILLIAM K. EMERSON Curator of Invertebrates, Department of Invertebrates, American Museum of Natural History, New York, New York 10024, USA AND BETTY JEAN PIECH Associate, Department of Mollusks, Delaware Museum of Natural History, Wilmington, Delaware 19807, USA Abstract. ‘The eastern Pacific species of the tonnacean genus Distorsio are reviewed; four species are recognized. Until recently, only two species of Distorsio were believed present in the eastern Pacific, namely D. decussata (Valenciennes, 1832) and D. constricta (Broderip, 1833). PARTH (1989a, b, 1991a), however, divided the long-held concept of D. constricta (sensu stricto) into three species. In addition to the nominate species, he recognized a second form as a new species, namely D. minoruohnishu Parth (1989b); and he referred a third form to D. ridens (Reeve, 1844a), a taxon recently assigned to the synonymy of D. clathrata (Lamarck, 1816) from the western Atlantic (EMERSON & SAGE, 1990a, b). We reject the attribution of the third form to D. ridens and describe it as a new species. The four species here recognized are D. decussata (Valenciennes, 1832), D. constricta (Broderip, 1833), D. minoruohnishit Parth, 1989, and D. jenniernestae Emerson & Piech, new species. A neotype is selected for D. decussata. INTRODUCTION PARTH (1989a, b, 1991a) has recently commented on the taxonomic status of the eastern Pacific species of Distorsio. In addition to the two west American species previously recognized by KEEN (1971:508), D. decussata (Valenci- ennes, 1832) and D. constricta (Broderip, 1833), he con- cluded that the latter is a complex comprised of three distinct taxa. He recognized, in addition to the nominate form, two other forms. For one of these newly recognized forms, he proposed a new species, D. minoruohnishi Parth (1989b), and he (PARTH, 1991a) referred the third form to D. ridens (Reeve, 1844), which was recently placed in the synonymy of D. clathrata (Lamarck, 1816) by EMERSON & SAGE (1990a, b). We take this opportunity to review the status of these taxa on the basis of specimens from collections newly available to us. Abbreviations The following abbreviations for institutions and expedi- tions are used in the text. AHF—Allan Hancock Foundation Pacific Expeditions, University of Southern California (collection transferred to LACMNH,; for station data, see FRASER, 1943). Page 106 AMNH—American Museum of Natural History, New York. BM(NH)—The Natural History Museum, London [for- merly The British Museum (Natural History)]. CASIZ—California Academy of Sciences, Invertebrate Zoology, San Francisco, California. DMNH—Delaware Museum of Natural History, Wil- mington, Delaware. DSIR, GG—DSIR Geology and Geophysics, Lower Hutt, New Zealand. LACMNH—Los Angeles County Museum of Natural History, Los Angeles, California. NMNH—National Museum of Natural History, U.S. National Museum collection (USNM), Smithsonian In- stitution, Washington, D.C. P-AMNH—Puritan-American Museum of Natural His- tory Expedition to Western Mexico (for station data, see EMERSON, 1958). TCE—Templeton Crocker Expeditions, New York Zoo- logical Society (collection deposited in the AMNH and CASIZ; for station data, see BEEBE, 1937, 1938). Historical Review of the Eastern Pacific Distorsio constricta Complex Before PARTH’s (1989-1991) recent work, only two spe- cies of Distorsio were generally recognized in eastern Pacific waters, namely D. decussata (Valenciennes, 1832) and D. constricta (Broderip, 1833). Owing largely to the failure of Valenciennes to illustrate his taxon, there had been a long period prior to the acceptance of this duospecies con- cept in which these taxa were confused (cf. TRYON, 1880: 35; DALL, 1908:319; PILsBRyY, 1922:359; WOODRING, 1928: 101). These authors either did not recognize D. decussata as a west American species, or, if they did, they mistakenly considered D. constricta to be a junior synonym of D. de- cussata. Eventually, PILSBRY & OLSSON (1941:40, pl. 5, figs. 9, 12) clarified the differences between the two taxa by stating: “It has been commonly believed that but one Distorsio was found living in the Panama Province but actually there are 2 well-marked species, Distorsio decus- satus Val., described from Acapulco, and Distorsio constric- tus Broderip, described from Santa Elena, and well figured by Reeve, (1844, Conch. Icon., Triton, pl. 12, fig. 41). The 2 species are easily separated even when they occur to- gether, as they do at several places along the Ecuadorian and Panamic coast. Distorsio constrictus is a strongly dis- torted shell, the aperture and particularly the inner lip being strongly tuberculate with a short but strongly re- curved anterior canal. Distorsio decussatus is a more slender, regular and thinner shell; the parietal callus is smoother, and the anterior canal is longer and nearly straight. In both species the spiral cord on the shoulder or periphery of the body-whorl is doubled.” The erroneous monospecies concept, however, was still held by some workers as late as KILIAS’ review (1973:207, 208) of the ‘““Cymatiidae.” The Veliger, Vol. 35, No. 2 Actually Dr. Woodring had accepted the duo status (in litt., to W. K. Emerson, 1 May 1954) following the pub- lication of a catalog of the genus Distorsio (EMERSON & PUFFER, 1953:98, 99; see WOODRING, 1959:205) and this concept was held by virtually all workers (cf. LEwIs, 1972: 46). Therefore, until recently, D. constricta and D. decussata were believed to be largely sympatric, the former known from the Gulf of California, Mexico, to Mancora, Ecua- dor, and the latter from the Gulf of California to Manta, Ecuador (KEEN, 1971:508). PARTH (1989a:53, illus. 2nd row, right side) reported two “typical” specimens of the western Pacific species Dis- torsio habei LEWIs (1972:38, figs. 38, 39) in a lot from “Oaxaca,” Mexico, mixed with specimens of D. decussata (VALENCIENNES, 1832:306) and D. constricta (BRODERIP, 1833:5). According to Dr. Beu (in litt., to W. K. Emerson, 25 October 1991), a specimen sent to him by M. Parth as D. haber, from “Oaxaca,” Mexico, is a specimen of D. perdistorta FULTON (1938:55, pl. 3, fig. 3, 3a), again a species not otherwise known from the eastern Pacific. The validity of this record was previously questioned (EMERSON, 1991:68, footnote 30). Until eastern Pacific specimens of D. habe: with unequivocal locality data are forthcoming, we question Parth’s west American records for this taxon. In the same paper, PARTH (1989a:52, illus. 2nd row, right side) figured a specimen from “Oaxaca” as “D. constricta constricta.” This is the same specimen he subsequently designated as paratype 8 of his new species, Distorsio (Dis- torsio) minoruohnishu Parth (1989b), which he shortly thereafter differentiated from D. constricta. Subsequently, PARTH (1991a) called attention to the presence of a third morphological form in the Distorsio constricta complex. He believed that it was referable to D. ridens (Reeve, 1844a), which was recently referred to the synonymy of D. clathrata (Lamarck, 1816), a western At- lantic species (EMERSON & SAGE, 1990a, b). Thus, he split this complex into three taxa, namely D. constricta (Brode- rip, 1833), D. minoruohnishi Parth, 1989, and D. ridens (Reeve, 1844), sensu PARTH (1991a). Before discussing the status of Distorsio ridens and our reasons for rejecting Parth’s assignment of the third form of this complex to D. ridens, we should comment on this newly recognized form and its relation to D. constricta (sensu stricto) and D. minoruohnishu. Fortunately, three syntypes of D. constricta are extant in The Natural History Museum, London (BM[NH] 1989016; Figures 2, 3), and there is no doubt of the concept of this taxon. Both D. constricta and the third form believed to be D. ridens by Parth have larger, more distorted shells than D. mino- ruohnishu, of which the largest known specimen is 51.6 mm in height (AMNH 232209); in contrast, the height of D. constricta and the third form may exceed 60 mm (64.3 mm [LACMNH 70-15.12] and 64.1 mm [AMNH 232215], respectively). PARTH (1991a:10) noted two major differences between Distorsio constricta and the form he believed referable to W. K. Emerson & B. J. Piech, 1992 D. ridens, namely the presence of a “big notch” on the “upper part of the outer lip, almost at the second tooth level”; and, secondly, the “color of the inner and outer lip, subdiscoidal-shaped, ranges from light orange to dark brown.” He did not mention the large white area of the parietal shield that extends into aperture. As noted by Parth, the third form has a distinctive groove on the interior of the outer lip located at the periphery of the body whorl. This “big notch” (Figure 18), however, is not formed in juvenile specimens, for example those with fewer than six postnuclear whorls (Figures 16-18). This groove is a prominent feature in mature specimens and is lacking at all growth stages in D. constricta and D. minoruohnishiv. Also, the color pattern and sculptural details differ mark- edly between D. constricta and the third species, as noted by Parth. We do agree that there are three distinct forms in this complex, but we disagree with Parth’s belief that the third form is referable to D. ridens. Distorsio ridens (Reeve, 1844a, b) was for a long time an enigmatic species, the status of which has been reviewed in recent years by LEWIs (1972), EMERSON & SAGE (1990a, b), and, most recently, by PARTH (1991a). LEwis (1972) could not locate the figured specimen in The Natural His- tory Museum, London, and he selected a lectotype in the American Museum of Natural History (AMNH 6369). He believed D. ridens represented a valid taxon of probable Indo-Pacific origin of which additional specimens had yet to be discovered. EMERSON & SAGE (1990a, b) concluded that Reeve’s taxon was referable to D. clathrata, a common western Atlantic inhabitant. PARTH (1991a) considered the third form in this eastern Pacific complex to be the taxon that REEVE (1844a:fig. 46) described and illustrated as “Triton” ridens from the “Philippine Islands.” We reject this attribution. We describe herein this third form as a new species and reject its allocation to Distorsio ridens for the following reasons: (1) The degree of distortion depicted in REEVE’s illus- tration (1844a:fig. 46) of D. ridens is slight, whereas in the new species, as well as in the illustration by REEVE (1844a:fig. 41) of D. constricta, the specimens are severely distorted (cf. Figures 1 and 9 with Figures 5-8). Fur- thermore, all of the specimens of the new species we have examined, some 40 specimens, as well as the specimens figured by PARTH (1990a:11, 1st row) have similar degrees of distortion. We believe that the artist, G. B. Sowerby, II, certainly would have recognized the distorted nature of the shell, as he did for D. constricta, if he had had before him a specimen of the new species to draw. In contrast, Reeve’s figure of D. ridens depicts a shell with a degree of distortion that is similar to that of Lewis’ lectotype of D. ridens and to specimens of D. clathrata from the western Atlantic Ocean (cf. Figure 9 with Figure 12). (2) The sculptural details of the parietal shield in the drawing of D. ridens (Figure 9) show the axial ribs ex- tending from the suture to the base of the parietal shield. Page 107 In mature specimens of the new species, the axial ribs are posteriorly wanting, as they extend only about half the distance from the suture to the middle portion of the shield (Figures 6, 8). The axial ribs, however, do extend from the suture to the aperture in D. clathrata (cf. Figure 9 with Figures 10, 12). The early plicae on the mid to lower columellar wall are prominent and the first plica is bifid in REEVE’s (1844a) drawing of D. ridens, but they are small and not bifid in the new species. Moreover, they are large and not uncommonly bifid in D. clathrata (see EMERSON & SAGE, 1990a:figs. 1, 14, 15). (3) There is no mention in REEVE’s (1844a, b) descrip- tions of D. ridens of a depression (“‘big notch’’) on the outer lip, nor is one depicted in the REEVE’s illustration (1844a: fig. 46). The wrinkle-like fold above the first anteriorly placed plica on the outer lip as shown in Reeve’s drawing (Figure 9 herein) is not uncommonly found on the outer lip of D. clathrata and is situated above the location of the “big notch” of the new species (cf. Figures 9, 10, 12). (4) The drawing of D. ridens (REEVE, 1844a:fig. 46; Figure 9 herein) shows the orange-brown pigmentation extending over the surface of the parietal shield as is often the case in specimens of D. clathrata (Figure 12). In the new species, the columellar surface immediately above the aperture is a glossy white (Figures 6, 8). PARTH (1991a:10) objected to the placement of Distorsio ridens in the synonymy of D. clathrata largely on the basis of two criteria that do not characterize D. clathrata. These are the single cording of the ribs on the dorsum and the length of the siphonal canal. In the new species, the spiral cords at the periphery of the whorls are “duplicated,” or are even “‘triplicated” in some specimens. REEVE’s (1844a: sp. 46) description of D. ridens states that the “whorls [are] elegantly latticed with prominent, narrow, raised ridges, transverse ridges duplicate.” Actually, the apertural view of REEVE’s illustration (1844a:fig. 46; Figure 9 herein) does not project the presence of “duplicated”’ spiral cords on the periphery of the body whorl. Unfortunately, a view of the dorsal surface of the illustrated specimen was not provided by Reeve. The presence of a double cord would be, of course, in contrast to the single spiral cord on the periphery of D. clathrata (Figures 11, 12). The duplicate sculpture is characteristic of D. decussata as noted by EMERSON & SAGE (1990a:134), and D. ridens was referred to this taxon by BEU (1985:62) and PARTH (1989a:54). Beu (in litt., to W. K. Emerson, 17 January 1990) sub- sequently concluded that D. ridens was referable to D. clathrata based on the identify of the lectotype of D. ridens (Figures 10, 11). PARTH (1991a:10) also believed that Reeve had never personally examined specimens of Distorsio clathrata before he described D. ridens, and, therefore, the specimen illus- trated by REEVE (1844:fig. 46) could not have been refer- able to D. clathrata. It should be noted, however, that Reeve was an associate of Broderip and had access to his collec- tion. BRODERIP (1833:5) in his description of D. constricta Page 108 Distortion Sculpture of body whorl Color Parietal shield Siphonal canal Outer edge of lip Inner edge of lip Prominent colu- mellar plicae Periostracum (out- er surface) actually compared it with D. clathrata. Furthermore, REEVE (1844a:sp. 45) cited “Triton clathratus Lamarck” [= Dis- torsio clathrata] in the synonymy of “Triton cancellinus” [= D. reticularis Linné, fide BEU, 1987:314], and he subse- quently compared D. ridens with “Triton cancellinus” (REEVE, 1844b:115). The status of D. clathrata at that time was poorly understood and specimens of D. clathrata were confused with specimens of D. reticularis from the Indo- Shell comparisons of the four Recent species of eastern Pacific Distorsio. Distorsio constricta Most distorted. 8-13 major axial ribs forming nodules where they cross spi- ral cords, double at the periphery. Body tan, darkest of the 4 species. Shield and outer lip darker than the body. Aperture white. Strong beading. Lower left-hand edge of shield just barely away from the body. Open, angles slightly to the right, recurves to the back. Shortest ca- nal of the 4 species. Thick, straight. 8 plicae, 3rd one slightly larger. 1 in posterior end. 1 on left side above siphon- al canal extending into aperture. Tan-brown color with very numerous soft, short hairs forming a velvet texture. Pacific (PUFFER, 1953:109). Table 1 Distorsio minoruohnishu Slightly less than D. con- stricta. 9-11 major axial ribs forming nodules where they cross spi- ral cords. 12-14 ma- jor spiral cords, dou- ble at the periphery. Body tan to straw color, early whorls darker. Dark band around the periphery of body whorl. Shield and outer lip cream color. Aperture white. Weak beading. Left- hand edge of shield noticeably extends away from the body. Open, angles slightly to the right, recurves to the back. Slightly lon- ger than D. jennier- nestae. Thinner than D. con- stricta flaring at the lower edge. 8 plicae, 3rd one slightly larger. 1 in posterior end. 1-2 on left side above si- phonal canal extend- ing into aperture. Tan-brown color with very numerous soft, short hairs forming a velvet texture. Longer individual hairs at many of the intersec- tions of the axial ribs and spiral cords. The Veliger, Vol. 35, No. 2 Distorsio jenniernestae Almost the same as D. constricta. 10-13 major axial ribs forming nodules where they cross spi- ral cords, double or triple at the periph- ery. Body varying between straw and white. Shield and outer lip have dark brown edge. Aperture white. Smooth to low beading. Left-hand edge of shield noticeably ex- tends away from the body. Open, angles slightly to the right, recurves to the back. Slightly lon- ger than D. constricta. Thinner than D. mino- ruohnishw. Flaring at the lower edge. Lip with a marked de- pression at the pe- riphery of the body whorl, unique to this species. 8 plicae, very prominent 3rd plica. 1 in posterior end. 1 on left side above siphon- al canal extending into the aperture. Tan-brown color with very numerous soft, short hairs forming a velvet texture. Longer darker hairs forming clusters along the axi- al ribs. Distorsio decussata Least distorted. 12-17 major axial ribs forming nodules where they cross spi- ral cords. 13 major spiral cords, single or double at the periph- ery. Body mostly white, some straw color. Shield and lip white with some light tan staining around the edge. Aperture white. Smooth to low beading. Leaf-hand edge of shield noticeably ex- tends away from the body. Open, straight, recurves to the back. Longest canal of the 4 species. Thin like D. jennier- nestae and flaring at the lower edge. 5-8 plicae, very promi- nent 3rd plica. 1 in posterior end. No plicae extending into the aperture above the siphonal canal. Dark-dark brown col- ored (covering a thin straw colored basal layer) with longer, individual hairs at the intersection of the axial ribs and spiral cords. The siphonal canal, Parth’s second point of difference, is indeed longer and less recurved in Distorsio clathrata than in the new species. However, the siphonal canal is shortened by breakage in the lectotype of D. ridens (Figure 10) and this may have been the case of the specimen figured by REEVE (1844a:fig. 46), which is depicted with a narrow siphonal canal, unlike the widely open canal of the new species (cf. Figure 9 with Figures 6, 8). Despite these possible disparities, the gross angular dis- W. K. Emerson & B. J. Piech, 1992 tortion of the whorls that characterizes the new species, together with the other differences (Table 1), serves in our opinion to separate it from Distorsio ridens. We take pleasure in naming the new species for Jennifer Ernest, the daughter of Gladys and James Ernest, who kindly provided us with critical specimens for this review. SYSTEMATIC TREATMENT Superfamily TONNACEA Suter, 1913 Family PERSONIDAE Gray, 1854 Genus Distorsio Roding, 1798 Synonyms: Distortrix Link, 1807; Persona Montfort, 1810; Distorta Perry, 1811; and Rhysema Clench & Turner, 1957; see BEU (1987:310; 1988:89). Type species: Distorsio anus (Linné, 1758) by subsequent designation of PILSBRY (1922:357). Distorsio (Distorsio) constricta (Broderip, 1833) (Figures 1-4, 23) Triton constrictus BRODERIP, 1833:5; REEVE, 1844: Triton sp. 41, pl. 12, fig. 41 [May, 1844]. Distorsio constrictus Brod. [erip]: TRYON, 1880:35, in part, pl. 17, fig. 176 only [copy of REEVE, 1844:fig. 41], not Distorsio cancellinus Roissy, sensu TYRON, 1880:35; PILsBRY & OLSSON, 1941:40, pl. 5, fig. 12, Manta, Ec- uador (Recent specimen); KILIAs, 1973:203, 204, fig. 145, “Peru,” in part, excluding references to Distorsio decussata (Valenciennes, 1832). Type locality: “Hab. ad Montem Christi et Xipixapi” (BRODERIP, 1833:5). ““Monte Christi and Xipixapi, West Columbia (dredged from sandy mud at a depth of from seven to ten fathoms; Cuming” (REEVE, 1844). Restricted by M. SMITH (1944:23) to St. Elena [= Xipixapa], Ec- uador. Type depository: Lectotype, 60.8 x 36.1 mm (Figures 2, 3) and 2 paralectotypes, 60.7 x 34.6 and 54.7 x 31.1 mm, respectively (BM[NH] 198016). Distribution: Islas Murcielago, Guanacaste Prov., Costa Rica, to Manta, Ecuador. Specimens examined: Costa Rica: Off Quepos, Pun- tarenas Prov. (9°22.20'N, 84°09.3’W) in 23 m, 1 specimen, J. McLean leg. (LACMNH 72-59.1); Isla San Pedrito, Islas Murcielago (10°51.5'N, 86°57.95'W), in 2.4-4 m, 1 specimen, J. McLean leg. (LACMNH 72-22.2). PANAMA: Golfo de Panama, Isla Venado, —2.4 tide, 5 April 1981, 1 specimen (AMNH 207600), ex H. DuShane coll.; Isla Venado, 2 specimens (LACMNH 34643); Isla Venado, —1.8 tide, in silty runnels, night 16 March 1980, 2 spec- imens (AMNH 232205), ex H. DuShane coll.; Isla Vena- do, beach, 2 specimens (AMNH 232206), ex A. Marti coll.; Isla Venado, 8 March 1970, 1 specimen, J. McLean leg. (LACMNH 70-15.12); Isla Bono, Islas Otoque, in 9- Page 109 27 m, 2 specimens, J. McLean leg. (LACMNH 65-21.9); Isla Secas, Chiriqui Prov., in 27 m, AHF station 34-125, 1 specimen (LACMNH 34-125.4); Bahia Honda, Vera- guas Prov., in 9-15 m, AHF station 33-120, 1 specimen (LACMNH 33-120.2); “Panama Bay,” intertidally, in sandy mud, 4 specimens (AMNH 232191), ex Abbey Specimen Shells. CoLomsia: Off Isla Gorgona, in 18 m, AHF station 34-98, 1 specimen (LACMNH 34-98.2). Ecuapor: Off Cape San Francisco, in 27 m, AHF station 38-118, 1 specimen (LACMNH 38-118.6); Manta, J. Marks leg. (CASIZ 37339). Remarks: There are in The Natural History Museum, London, three specimens (BM[NH] 198016) that were identified as syntypes by Aileen Blake (A. Beu, in litt., to W. K. Emerson, 25 October 1991). These are large spec- imens (H = 61.2, 60.9, and 54.7 mm), each of which is badly faded, but otherwise well preserved. The second largest, which has some of the periostracum still preserved, appears to be the specimen illustrated by REEVE (1844a: fig. 41) and is here selected as the lectotype (cf. Figure 1 with Figures 2, 3). The old labels accompanying the syn- types cite “Monte Christi and Xipixapi” as the type lo- calities, which are BRODERIP’s (1833:5) citations for the habitat. Distorsio (Distorsio) minoruohnishir Parth, 1989 (Figures 13-15a, b, 25) Distorsio constricta constricta (Broderip): PARTH, 1989a:52, in part, 6 unnumbered figs., specimen in 3rd row, on right side, “Oaxaca, W. messico” [sic]; not Distorsio con- stricta (Broderip, 1833). Distorsio minoruohnishi PARTH, 1989b:8-11, holotype illus- trated on p. 8, holotype and 9 paratypes illustrated on p. 11; PARTH, 1991a:11, 3rd row, first two specimens illustrated on left side of plate. Type depositories: Holotype (BM[NH] 1990025), fide PARTH (1991b:21; here illustrated in Figure 15a, b). Nine paratypes cited and illustrated (PARTH, 1991a:9) from Mexico and Panama. One of these, paratype 2, from the “Bay of Chiriqui, Panama” is deposited in the AMNH (246024). Type locality: “Oaxaca, Mexico” [the Mexican state] cit- ed for the holotype (PARTH, 1989b:9). Here restricted to off “Isla Macapule, [Sinaloa], Mexico” (PARTH, 1989b:9; paratype 9; and 28 topotypes AMNH 174247, 186686, and 232198; Figures 13, 14). Distribution: Golfo de California, Mexico, to off Tumbes, Peru. Specimens examined: MExico: Gulf of California, Baja California Norte: Off Isla San Jose, P-AMNH Station 116, in 67 to 73 m, 2 specimens (AMNH 76180); off Puerto Escondido, PPAMNH Station 138, in 33 to 36 m, 1 specimen (AMNH 76493); Baja California Sur, Punta Page 110 The Veliger, Vol. 35, No. 2 W. K. Emerson & B. J. Piech, 1992 Arena Bank, TCE Station 136-D-17, in 82 m, 4 specimens (AMNH 94097); Gorda Bank, TCE Station 150-D-32, in 165 m, 29 specimens (AMNH 140269). Bahia Los Frailes, PAMNH Station 89, in 36 to 73 m, 2 specimens (AMNH 75807); off Punta Coyote, La Paz Bay, dredged in 45 m, 3 specimens, ex Abbey Specimen Shells (AMNH 183767); south of La Paz, dredged, 5 specimens, ex C. Skoglund coll. (AMNH 186684). Sonora: Bahia Kino, trawled, 2 specimens, A. Luna leg. (AMNH 232210); off Guaymas, shrimp boat, 2 specimens, ex T. Rice coll. (AMNH 180703); Puerto Guaymas, in 33 m, 4 specimens, ex R. Purdy coll. (AMNH 240421); off Guaymas, dredged by fishermen, 1 specimen, G. Eddison coll. (AMNH 232203). Sinaloa, Isla Macapule (south of Bahia de To- polobampo), in 45 m, 22 specimens (together with 4 spec- imens of Distorsio jenniernestae, sp. nov.), A. Luna /eg., ex H. DuShane coll. (AMNH 174247, 232198); same locality and data, ex C. Skoglund coll., 6 specimens (AMNH 186686). Nayarit, Islas Las Tres Marias, off Isla Maria Magdalena, P-AMNH Station 71, in 23 to 27 m, 1 spec- imen (AMNH 75379), Isla Maria Madre, PPAMNH Station 72, in 25 to 27 m, 1 specimen (AMNH 75529); Colima, off Manzanillo, in 15-22 m, 4 specimens, ex S. Bennett coll. (AMNH 232207); off Manzanillo, in 31 m, 2 specimens, ex H. DuShane coll. (AMNH 232204). Costa Rica: Isla Tortuga, 1 specimen, ex K. Vaught coll. (AMNH 214705); Bahia Ballenas, in 64-73 m, TCE Station 213- D-11-17, in 64 m, 3 specimens (AMNH 85336a) (together with 2 specimens of D. jenniernestae); between Bahia Elena and Bahia Juanillo (85°46.13’W), in 26-53 m, 2 specimens (LACMNH 72-12.4); off Bahia Herradura (9°38.8'N, 84°40.8’W), in 37 m, 3 specimens, J. McLean leg. (LACMNH 72-54.4). PANAMA: Golfo de Panama, Palo Seco, 3 specimens, E. Bergeron leg. (AMNH 156620); Isla de los Perlas, 3 specimens (AMNH 123002); Golfo de Montijo, off Isla Gobernadora, 1 specimen, ex B. Piech coll. (AMNH 239212). Golfo de Chiriqui, paratype 2, ex M. Parth coll. (AMNH 246024). CoLomsia: Off Puerto Utria, in 82 m, AHF Station 35-54, 1 specimen (LACMNH 35-54.1). Ecuapor: Off Santa Elena, in 15- 18 m, AHF Station 34-83, 1 specimen (LACMNH 34- 83.6). PERU: Between Caleta La Cruz and Puerto Pizzaro, off Tumbes (3°28’S, 80°36'W), in 9-33 m, 3 specimens, Page 111 J. McLean leg. (LACMNH 72-83.6). (An additional 40+ lots in the LACMNH and AMNH collections from Mex- ico, Costa Rica, and Panama were examined but are not recorded owing to redundancy.) Remarks: The holotype (BM[NH] 1990025; Figure 15a, b herein) is a small, well-preserved, crabbed specimen, with 6 postnuclear whorls, and measures 38.3 mm in height and 21.3 mm in width. Mature examples with 7 post- nuclear whorls attain 51+ mm in height (AMNH 232209, Gulf of California). Some of the more ovate specimens approach Philippine specimens of Distorsio habe in many characters, but differ mostly in the degree of distortion of the whorls, in some of the details of the apertural dentition, and in the possession of a less recurved siphonal canal (cf. Figures 13-15 with Figures 19, 20). Distorsio (Distorsio) enniernestae Emerson & Piech, sp. nov. (Figures 5-8, 16-18, 24) Distorsio (Rhysema) constricta (Broderip): EMERSON & OLD, 1963:26, in part, fig. 24, off Isla Tiburon, Gulf of Cal- ifornia; KEEN, 1971:508, in part, fig. 962, Gorda Bank, Gulf of California (CASIZ coll.); LEwis, 1972:45, in part, fig. 41, off Santa Cruz Id., Galapagos Islands; KERSTITCH, 1989:45, in part, fig. 89 (colored photo- graph of living animal); not Distorsio constricta (Brod- erip). Description: Shell, large (attaining 60+ mm in height), fusiform, very much distorted, spire attenuately acumi- nated (spire produced at an angle of about 45°), with 7 postnuclear whorls angulated at the upper part, and 22 smooth, glossy embryonic whorls (Figure 24). Surface of body whorl sculptured with 10 to 13 major axial ribs and numerous spiral cords forming nodules at the intersections; nodules on periphery of the shoulder largest, crossed by 2 or 3 spiral cords. Aperture large, outer lip thin at edge with 3 distinct plicae (the third being the largest) on the upper (posterior) portion, disjunct from the outer edge, and with 5 or 6 broken denticles on the lower (anterior) portion. Outer lip with a large depression (groove) formed Explanation of Figures 1 to 12 Figures 1-4. Distorsio constricta (Reeve). Figure 1: copy of illustration of Triton constrictus Broderip (REEVE, 1844a: pl. 12, fig. 41). Figures 2, 3: lectotype of 77zton constrictus (BM[NH] 198016). Figure 4: specimen with fully developed apertural morphology (LACMNH 70-15.12). Figures 5-8. Distorsio jenniernestae, sp. nov. Figures 5, 6: holotype (AMNH 232214). Figures 7, 8: paratype (AMNH 232215). Figures 9-12. Distorsio clathrata (Lamarck, 1816). Figure 9: copy of illustration of Triton ridens REEVE (1844a:pl. 12, fig. 46). Figures 10, 11: lectotype of Distorsio ridens (AMNH 6369). Figure 12: specimen from off Punta Patuca, Atlantic Honduras (AMNH 238556). Figures 1-12, x1. Page 112 The Veliger; Vol=35,5No"2 Explanation of Figures 13 to 25 Figures 13-15a, b. Distorsio minoruohnishui Parth, 1989. Figures 13, 14: specimen with fully developed apertural morphology, off Isla Macapule, Sinaloa, Mexico, topotype (AMNH 232198). Figures 15a, b: holotype (BM[NH] 1990025). Figures 16-18. Distorsio jenniernestae, sp. nov., a growth series showing the prominent marginal groove developed in the largest specimen (Figure 18), off Isla Macapule, Sinaloa, Mexico (AMNH 232199). W. K. Emerson & B. J. Piech, 1992 at the periphery of the body whorl (not developed in spec- imens with fewer than 6 postnuclear whorls; see Figures 16-18). Parietal shield extending to the suture, thin with 4 to 6 axial ribs below the suture, replaced by 5 or more weak spiral ribs below, blending into the aperture. Col- umellar inner edge with 8-10 plicae, with the upper 2 (posterior) the largest. Siphonal canal broadly open, short, and recurved slightly upwards. Basic color white and straw- tan, early whorls darker tan, parietal shield and outer lip orange-brown with streaks of white, aperture glossy white. Animal orange-brown with white blotches (KERSTITCH, 1989:fig. 89). Periostracum tan-brown, thin, flaky, with numerous short hairs; darker hairs in clusters along the axial ribs. Operculum small, oblong, terminal with several concentric fine lines on outer surface, and with wide mar- ginal, raised callus and broad bands of uneven rings on inner surface. Type locality: Dredged in 73 m between Isla Cebaco and Isla Coiba, off the Pacific coast of Veraguas, Panama, J. Ernest, 1991. Type depositories: Holotype (AMNH 232214; Figures 5, 6; H = 59.9 mm, W = 32.3 mm) and 4 paratypes (AMNH 232215; Figures 7, 8); 2 paratypes (USNM 860245); 2 paratypes (DMNH 189600); 1 paratype (DSIR-GG WM 15345) and 10 paratypes (B. J. Piech coll.); all from the type locality. Distribution: Golfo de California, Mexico, to the Golfo de Panama, and the Galapagos Islands. Specimens examined: Mexico: Golfo de California, off Isla Tiburon, PPAMNH Station 162, in 73 m (AMNH 77066), illustrated as Distorsio constricta (Broderip) by EMERSON & OLD (1963:27, fig. 24); off Isla Angel de la Guardia, in 93-102 m, AHF station 40-29, 1 specimen (LACMNH 40-29.1). Between Isla Partida and Espiritu Santo, in 73-165 m, AHF Station 60-61, 1 specimen (LACMNH 60-6.9); off Punta Coyote, Bahia de La Paz, Baja California Sur, dredged in 45 m, by commercial fish- ermen, 2 specimens (AMNH 232192) ex Abbey Specimen Shells. Off Cabo Pulmo, in 91 m, AHF 1732-49, 1 spec- imen (LACMNH 49-73.1); off Cabo San Lucas, Baja California Sur, in 137 m, AHF Station 618-37, 1 specimen (LACMNH 37-19.2); Bahia Guaymas, Sonora, dredged in 30 m by fishermen, 1 specimen (AMNH 240421a), ex Page 113 R. Purdy coll.; off Gabo Haro, Sonora, in 183 m, AHF station 60-31, 3 specimens (LACMNH 60-3.4). Isla Ma- capule, Sinaloa, dredged in 45 m by A. Luna, 4 specimens (AMNH 232199), ex S. Bennett coll. Costa Rica: Off Bahia de Ballenas, Golfo de Nicoya, in 64 to 82 m, TCE Station 213-0-11-17, in 64 m, 2 specimens (AMNH 85336) (together with 2 specimens of D. minoruohnishu Parth); off Isla del Cano (8°45'N, 84°0’W) in 73-82 m, 1 specimen, J. McLean leg. (LACMNH 72-67.2). PANAMA: Golfo de Chiriqui, 1 specimen, ex M. Parth coll. (AMNH 232183); between Isla Cébaco and Isla Coiba, in 73 m, type locality, 20 specimens. ECUADOR: Galapagos Islands, SE of Isla Daphne (0°27'S, 90°21.8'W), AHF Station 38-48, in 101 m, 1 specimen (LACMNH 38-48.1). Remarks: Mature specimens of the new species may be distinguished without difficulty from the other three species of Distorsio recognized in the eastern Pacific Ocean (Table 1). The “big notch” on the outer lip is present in specimens with six or more postnuclear whorls (cf. Figures 16-18). The function, if any, of this groove is not known. Joseph Houbrick (personal communication, 24 September 1991) speculated that perhaps it serves as an egg laying sinus or possibly a penile groove, owing to the location of the bursa copulatrix and penis, respectively, in the related families Ranellidae and Bursidae (HOUBRICK & FRETTER, 1969: 417). With the exception of Distorsio decussata, the em- bryonic whorls of all the west American species are similar, consisting of 2% smooth, glossy whorls. This character is not useful in separating the other three taxa (Figures 23- 25). The new species is distributed intertidally to depths of 137 m from near the head of the Gulf of California to the Gulf of Panama. It is also known from the Galapagos Islands in 101 m. Distorsio minoruohnishiu is largely sym- patric in range with the new species in depths to 80+ m, but it also extends southward to northern Peru. In contrast, D. constricta is restricted in distribution to the southern part of the Panamic faunal province, with records from Costa Rica to Ecuador, in tidal waters to depths of 27 m. The interspecific relationships of the new species and the three other west American species are somewhat dif- ficult to interpret. Distorsio constricta in the Pacific and D. macgintyt EMERSON & PUFFER (1953:101; OLSSON & McGinty, 1951:27, pl. 1, figs. 5, 6, 9) in the Atlantic were considered to be geographical subspecies (BEU, 1985: Figures 19, 20. Distorsio habe: Lewis, 1972, off Panglao, Bohol, Philippines (AMNH 232189). Figures 21, 22: Distorsio decussata (Valenciennes, 1832), neotype, off Punta Arena, Gulf of California, TCE Station 136-D-21, in 82 m (AMNH 85335). Figures 23-25. Spires enlarged to show embryonic whorls. Figure 23: D. constricta (AMNH 232205). Figure 24: D. jenniernestae, sp. nov. (AMNH 232199, specimen shown in Figure 16). Figure 25: D. minoruohnishu (AMNH 174247). Figures 13-22, x1; Figures 23-25, x3.5. Page 114 62; EMERSON, 1991:73, table 4) before the discovery of more than one species in the D. constricta complex. OLSSON & McGinty (1951:27), Lewis (1972:46), and others be- lieved that D. simillima (SOWERBY, 1850:48) from the Mio- cene of the Caribbean region was the precursor of this cognate pair. Unfortunately, Sowerby did not provide an illustration of his D. szmillima from the Dominican Re- public and this name has been applied to various species concepts (cf. WOODRING, 1959:206). PFLUG (1961:39-41, pl. 9, figs. 4, 6, 9) selected and illustrated a lectotype of D. simillima. On the basis of this lectotype designation, D. simillima is not referable to the D. constricta complex. Pflug’s lectotype is clearly related to D. decussata (VALENCIENNES, 1832:306). The closeness of the lectotype in shell characters to the Mid-American Pliocene D. gatunensis TOULA (1909: 700, pl. 25, fig. 10; BRown & PIussry, 1911:356, fig. 8; WOODRING, 1959:205, pl. 34, figs. 7, 8; AGUILAR & FIs- CHER, 1986:223, pl. 2, figs. 13, 14) suggests that Toula’s taxon is closely related to D. stmillima. Maury (1917:107, pl. 17, figs. 4, 5) considered D. gatunensis a junior synonym of D. simillima. Beu (zn litt., to W. K. Emerson, 25 October 1991), however, believes D. gatunensis to be identical with D. decussata. This leaves the Pliocene precursor of the D. constricta complex (WOODRING, 1928:pl. 18, fig. 9 and pl. 19, fig. 1) from Panama and elsewhere in the New World tropics without a name. Further study is needed to deter- mine if the fossil populations of D. constricta require spe- cific recognition. Distorsio crassidens (CONRAD, 1848:118, pl. 11, fig. 40; MacNeit & Dockery, 184:121, pl. 31, figs. 5, 6) from the Vicksburg Group of Mississippi (Oli- gocene) and D. simillima (Mio-Pliocene) appear to be lin- ear antecedents of D. decussata. Distorsio clathrata (LA- MARCK, 1816:pl. 413, fig. 4a, b), on the other hand, has a Pliocene presence in the Caribbean region (WOODRING, 1928:pl. 19, figs. 2, 3; RUTSCH, 1930:pl. 17, figs. 4, 5). It is also known from the Ecuadoran Pliocene (OLSSON, 1964: 174, pl. 30, figs. 1-1b), although it did not survive in the Pacific after the closure of the Mid-American seaways. Both Distorsio minoruohnishii and D. jenniernestae may have evolved from the Neogene D. constricta stock in the equatorial waters of Central America, or from some yet unrecognized stocks. On the other hand, the relationship of these two species with the Indo-Pacific D. habe: LEWIs (1972:38, figs. 38, 39) is not clear. Until recently, D. habez was recognized as a geographical subspecies of D. constricta constricta (cf. BEU, 1985:62; LEWIS, 1972:44, figs. 38-39; EMERSON, 1991:68). Perhaps genetic differences deter- mined by molecular studies could shed more light on the relationships of these taxa. Distorsio (Distorsio) decussata (Valenciennes, 1832) (Figures 21, 22) Tritonium decussatum VALENCIENNES, 1832:306; EMERSON & PUFFER, 1953:99; KiLias, 1973:203, in part, name only, excluding references to Distorsio constricta. The Veliger, Vol. 35, No. 2 Distorsio decussatus Valenciennes: PILSBRY & OLSSON, 1941:40, pl. 5, fig. 9; HERTLEIN & STRONG, 1955:265, 266; EMERSON & OLD, 1963:27, fig. 25; KEEN, 1971:508, fig. 963; LEwIs, 1972:43, figs. 36, 37; PARTH, 1991a: 11, 2nd row, four specimens illustrated. Type locality: “Habitat cum praecidente [T7itonium he- mastoma| ad portum Acapulco,” Guerrero, Mexico. Type depository: There are no specimens of this taxon among Valenciennes’ type material in the Muséum Na- tional d’ Histoire, Naturelle, Paris. The types are presumed to be lost (A. Beu, in litt., to W. K. Emerson, 25 October 1991). In the absence of any known type specimens, we here designate as the neotype of D. decussata (Valenciennes, 1832) a specimen dredged from the Arena Bank, Baja California Sur, Mexico (23°29'N, 109°25’'W) in 82 m (AMNH 85335, see Figures 21, 22). Distribution: Golfo de California, Mexico to Manta, Ec- uador (HERTLEIN & STRONG, 1955). Material examined: 34 lots in the AMNH collection, from Mexico, Panama, Colombia, and Ecuador. Remarks: VALENCIENNES (1832:306) did not illustrate his new species from Acapulco, Mexico. He stated Distorsio decussata was intermediate in characters between D. anus and D. clathrata. He compared his species with D. clathrata, noting differences in the anal sinus and the columellar plicae. He described a small (H = 54 mm), weakly dis- torted, white shell with reddish spots on the parietal shield, and he noted the presence of uneven labial plicae of which the third plica was the largest. The siphonal canal was described as elongated and thin edged. For the purpose of nomenclatural stability, we have selected a neotype (see above). This is the largest of the four west American species, attaining more than 85 mm in length (AMNH 226426; dredged off Veraguas, Panama, J. Ernest, 1991). The weak distortion, long siphonal canal, deeply grooved col- umellar notch, and very large third plica on the inner edge of outer lip serve to characterize this species (Table 1). The periostracum on the outer surface is dark brown and covers a tannish basal layer (cf. LEwIs, 1972:37). A delicate prominent single hair occurs on the nodules formed at the intersection of the axial and spiral cords. ACKNOWLEDGMENTS We are indebted to James Ernest of Balboa, Panama, for providing a large series of newly dredged specimens for study. James H. McLean of the Los Angeles County Mu- seum of Natural History lent his holdings (58 lots) of the Distorsio constricta complex. Kathie Way of the British Museum (Natural History) lent type specimens for study. Terrence M. Gosliner of the California Academy of Sci- ences and Henry W. Chaney of the Santa Barbara Mu- seum of Natural History allowed access to their respective collections. Ilse Kaim provided English translations of W. K. Emerson & B. J. Piech, 1992 German texts. We thank Alan Beu of the New Zealand DSIR Geology and Geophysics, Lower Hutt, Hal Lewis of the Academy of Natural Sciences of Philadelphia, Jo- seph R. Houbrick and M. G. Harasewych of the National Museum of Natural History, Washington, D.C., E. Alison Kay of the University of Hawaii, and Manfred Parth of Munchen, Germany, for an exchange of information. We are indebted to our colleagues at the AMNH, Walter E. Sage, III for technical assistance, Andrew S. Modell for photographic services, and Stephanie Crooms for word- processing the manuscript. Alan Beu also kindly reviewed the manuscript and offered many helpful suggestions. We thank him and the other reviewers for their constructive comments. LITERATURE CITED AGUILAR, T. & R. FISCHER. 1986. Moluscos de la Formacion Montezuma (Plioceno-Pleistoceno; Costa Rica). Geologica et Palaeontologica 20:209-241. BEEBE, W. 1937. The Templeton Crocker Expedition. II]. New York Zoological Society, Zoologica 21(1):33-46. BEEBE, W. 1938. Eastern Pacific Expeditions of the New York Zoological Society, XIV. New York Zoological Society, Zoologica 23(3):287-298. Beu, A.G. 1985. A classification and catalogue of living world Ranellidae (= Cymatiidae) and Bursidae. Conchologists of America Bulletin 13(4):55-66. Beu, A. G. 1987 [“1986”]. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part 2. Descriptions of 14 new modern Indo-West Pacific species and subspecies, with revisions of related taxa. New Zealand Journal of Zoology 13(3):273-355 [published 28 January 1987, teste A. G. Beu]. Bru, A. G. 1988. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part 5. Early his- tory of the families, with four new genera and recognition of the family Personidae. Jn: J. Grant-Mackie, A. K. Ma- suda, K. Mouri & K. Ogasawara (eds.), Professor Tamio Kotaka Commemorative Volume on Molluscan Paleontol- ogy, The Saito Gratitude Foundation, Sendai, Japan, Special Volume 2:69-96. BRODERIP, W. J. 1833. [New species . . . accompanied by char- acters by Mr. Broderip.] Proceedings Zoological Society London for 1833 1:4-8 [May 17, 1833]. Brown, A. P. & H. A. PiusBry. 1911. Fauna of the Gatun Formation, Isthmus of Panama. Proceedings Academy of Natural Sciences of Philadelphia 63(2):336-373. ConrRAD, T. A. 1848. Observations on the Eocene formation, and descriptions of one hundred and five new fossils of that period, from the vicinity of Vicksburg, Mississippi; with an Appendix. Journal of the Academy of Natural Sciences of Philadelphia, Second Series 1(2):111-134. DaLL, W. H. 1908. Reports on the dredging operations off the west coast of Central America to the Galapagos, to the west coast of Mexico, and in the Gulf of California.... XIV. The Mollusca and the Brachiopoda. Museum of Compar- ative Zoology, Bulletin 43(6):205-487. EMERSON, W. K. 1958. Results of the Puritan-American Mu- seum of Natural History Expedition to western Mexico. 1. General Account. American Museum Novitates No. 1894: 1-25. Page 115 EMERSON, W. K. 1991. First records for Cymatium mundum (Gould) in the eastern Pacific Ocean, with comments on the zoogeography of the non-tonnacean prosobranch gastropods with Indo-Pacific faunal affinities in west American waters. The Nautilus 105(2):62-80. EMERSON, W. K. & W. E. OLD., JR. 1963. Results of the Puritan-American Museum of Natural History Expedition to western Mexico. 19. The Recent mollusks: Gastropoda, Strombacea, Tonnacea, and Cymatiacea. American Museum Novitates No. 2153:1-38. EMERSON, W. K. & E. L. PUFFER. 1953. A catalogue of the molluscan genus Distorsio (Gastropoda, Cymatiidae). Pro- ceedings of the Biological Society of Washington 66(16):93- 108. EMERSON, W. K. & W. E. SaGE, III. 1990a. Distorsio ridens (Reeve, 1844): a synonym of Distorsio clathrata (Lamarck, 1816) (Gastropoda: Personidae). The Nautilus 103(4):131- 135% EMERSON, W. K. & W. E. SaGE, III. 1990b. Addenda to “Distorsio ridens (Reeve, 1844): a synonym of Distorsio clath- rata (Lamarck, 1816) (Gastropoda: Personidae).” The Nau- tilus 104(3):108-110. FRASER, C. M. 1943. General account of the scientific work of the Velero IIT in the eastern Pacific, 1931-41. Part 3. A ten- year list of the Valero I/I collecting stations. An appendix of collection stations of the Allan Hancock Foundation for the Year 1942. Allan Hancock Pacific Expeditions 1(3):256- 431. University of Southern California Press: Los Angeles. FuLTON, H.C. 1938. Descriptions and figures of new Japanese marine shells. Proceedings of the Malacological Society of London 23(1):55, 56. HERTLEIN, L. G. & A. M. STRONG. 1955. Marine mollusks collected during the ““Askoy” expedition to Panama. Amer- ican Museum Natural History Bulletin 107(2):159-318. Houprick, J. R. & V. FRETTER. 1969. Some aspects of the functional anatomy and biology of Cymatium and Bursa. Proceedings Malacological Society of London 38(5):415- 429. KEEN, A. M. 1971. Sea Shells of Tropical West America. Marine Mollusks from Baja California to Peru. 2nd ed. Stanford University Press: Stanford, CA. 1064 pp. KERSTITCH, A. 1989. Sea of Cortez Marine Invertebrates. A Guide for the Pacific Coast, Mexico to Ecuador. Sea Chal- lengers: Monterey, CA. 114 pp. Kiuias, R. 1973. Das Tierreich Eine Zusammenstelllung und Kennzeichnung der rezenten Tierformen. Gastropoda/Pros- obranchia Cymatiidae. Berlin, Leiferung 92(8):i-viii, 1-235. LAMARCK, J. B. A. M. DE. 1816. Tableau encyclopédique et meéthodique des trois régnes de la nature. Vingt-troisiéme partie. Mollusques et polypes divers, Paris. Liste des objects représentés, 16 pp., Atlas 3, pls. 391-488 [431 bis]. Lewis, H. 1972. Notes on the genus Distorsio (Cymatiidae) with descriptions of new species. The Nautilus 86(2—4):27- 50. MacNEIL, F.S. & D. T. Dockery, II. 1984. Lower Oligocene Gastropoda, Scaphopoda, and Cephalopoda of the Vicksburg Group in Mississippi Department of Natural Resources, Bureau of Geology. Bulletin 124:1-415. Maury, C. J. 1917. Santo Domingo type sections and fossils. Part 1. Bulletins of American Paleontology 5(29):1-251. Ousson, A. A. 1964. Neogene Mollusks from Northwestern Ecuador. Paleontological Research Institution: Ithaca, New York. 256 pp., 38 pls. Otsson, A. A. & T. H. McGinty. 1951. A Distorsio new to the Florida fauna. The Nautilus 65(1):26-28. Page 116 ParTH, M. 1989a. Brief notes on the genus Distorsio Roding, 1798 and description of a new species. La Conchiglia (Rome), Nos. 238-241:52-57 [issue mislabeled on cover as Nos. ““233- 236”). ParTH, M. 1989b. More about the complex of Distorsio con- stricta and description of Distorsio minoruohnishu n. sp. La Conchiglia (Rome), Nos. 246-249:8-11. ParTH, M. 1991a. The rediscovery of Distorsio ridens Reeve, 1844, a valid species. La Conchiglia (Rome), No. 258:8-10. ParTH, M. 1991b. Bursa (Bursa) lucaensis new species from the Philippines. La Conchiglia (Rome), No. 259:21 [“De- rivatio nomines’’]. PriuGc, H. D. 1961. Mollusken aus dem Tertiar von St. Do- mingo. Acta Humboldtiana, Wiesbaden, Series Geologica et Palaeontologica 1:1-107. Pitssry, H. A. 1922. Revision of W. M. Gabb’s Tertiary Mollusca of Santo Domingo. Proceedings Academy of Nat- ural Sciences of Philadelphia 73(2):305-435. Pitssry, H. A. & A. A. OLsSson. 1941. A Pliocene fauna from western Ecuador. Proceedings Academy of Natural Sciences of Philadelphia 93(1):1-79. PuFFER, E. L. 1953. Distorsio reticulata vs. Distorsio clathrata in the West Indies. Proceedings of the Biological Society of Washington 66(17):109-124. REEVE, L. A. 1844a. Monograph of the genus 77zton. Con- chologia Iconica: illustrations of the shells of molluscous animals. Reeve Brothers: London. 2 7riton text and 20 pls. [pl. 12, May 1844]. REEVE, L. A. 1844b. Descriptions of new species of tritons, collected chiefly by H. Cuming, Esq. in the Philippine Is- lands. Proceedings of the Zoological Society of London 12(137):110-122 [Dec. 1844]. RutscH, R. F. 1930. Einige interessante Gastropoden aus dem The Veliger, Vol. 35, No. 2 Tertiar der Staaten Falcon und Lara (Venezuela). Eclogae Geologice Helvetiae, Recueil Périodique de la Société Géo- logique Suisse 23(2):604-614, pl. 17. SMITH, M. 1944. Panamic Marine Shells, Synonymy, No- menclature, Range and Illustrations. Winter Park, FL. 127 pp., 912 figs. SoweErsy, G. B., II. 1850. Descriptions of new species found by J. S. Heniker, Esq. Quarterly Journal of the Geological Society of London 6(1):44-53. TouLa, F. 1909. Eine jungtertiadre Fauna von Gatun am Pan- ama-Kanal. Jahrbuch der Kaiserlich-Koniglichen Geolo- gischen Reichsanstalt 68(4):673-760. Tryon, G. W., JR. 1880. Family Tritonidae. Manual of Con- cholegy, Ser. 1, Philadelphia 3(9):1-64, pls. 1-17 [2 January 1880, fide VANATTA, 1927:97]. VALENCIENNES, A. 1832. Coquilles univalves marines de PAmerique Equinoxiale. Jn: F. H. A. voN Humboldt & A. J. A. Bonpland, Voyage aux régions équinoxiales du Nou- veau Continent, Paris, pt. 2, Recueil d’observations de zoolo- gie et d’anatomie comparée 2:263-339. VANATTA, E.G. 1927. Dates of publication of the parts of the Manual of Conchology, First Series (Cephalopoda, Marine Gastropoda, Polyplacophora, Scaphopoda). The Nautilus 40(3):96-99. WooprInG, W. P. 1928. Marine mollusks from Bowden, Ja- maica. Part II. Gastropods and discussion of results. Car- negie Institution of Washington, Publication No. 385:i-vii + 1-564. WoopRING, W. P. 1959. Geology and paleontology of the Ca- nal Zone and adjoining parts of Panama. Description of Tertiary mollusks (Gastropoda: Vermetidae to Thaididae). U.S. Geological Survey Professional Paper 306-B:i-i11 + 147-239. The Veliger 35(2):117-121 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 A Re-evaluation of the Ontogeny of Cabestana spenglernt (Perry, 1811) (Gastropoda: Tonnoidea: Ranellidae) FRANK RIEDEL Geological-Palaeontological Institut and Museum, Bundesstr. 55, 2000 Hamburg 13, Germany Abstract. ‘The egg mass, hatching, and early veliger of Cabestana spengleri are described. The state- ment by ANDERSON (1959) that C. spengler: has direct development is not confirmed and the contradicting data of KESTEVEN (1901), who figured a larval shell, are confirmed. Cabestana spengleri has a plank- totrophic larva that lives long enough to cross the Tasmanian Sea. INTRODUCTION WILSON & GILLET (1971) wrote: ““However at least one species, Cabestana spengleri, has been reported to have di- rect development (ANDERSON, 1959).” They refer to the Ranellidae Gray, 1854 (= Cymatiidae Iredale, 1913). HOusBRIcK & FRETTER (1969) also cite Anderson’s results: “ANDERSON (1959) [reported] the egg capsules and en- closed veligers of Cymatilesta spengler: (Perry), a species with no free larva, from Queensland.” KILIAs (1973) doc- umented some of Anderson’s figures. Anderson did not mention the publication of KESTEVEN (1901), who figured the protoconch of Cabestana spengleri (Figure 7). The drawing shows a shell that resembles a typical protoconch (for figures of ranellid protoconchs see as examples BANDEL, 1975, 1981, 1991; BEU, 1978, 1986, 1988; BEU & Kay, 1988; BEU & KNUDSEN, 1987; CLENCH & TURNER, 1957; KESTEVEN, 1901, 1902; LAURSEN, 1981; LEwIs, 1974; PILKINGTON, 1974; RICHTER, 1984; SCHEL- TEMA, 1966, 1971; WAREN & BOUCHET, 1990) of a ra- nellid gastropod with a planktotrophic veliger phase. Al- though discussing larval strategies and protoconchs of the Tonnoidea, WAREN & BOUCHET (1990) did not mention the “problem” concerning C. spengleri. If Anderson was right in maintaining that C. spengleri has direct develop- ment, it would not be possible to differentiate protoconchs of species with planktotrophic larvae from those not having this phase in their ontogeny. Consequently a short review of Anderson’s work was necessary. MATERIALS anp METHODS Specimens of Cabestana spengleri and their egg masses were collected during the first week of December 1990 at Long Reef (northern Sydney, Australia) and then transported to the Division of Invertebrate Zoology, Australian Mu- seum. The spawn was kept in 10-L tanks. When hatching started the spawn was moved daily into a new tank in order to keep the older veligers separate from the younger ones. The veligers were observed with the aid of a light microscope. Drawings were made from different stages in different situations. Every day a few hundred veligers were fixed in 70% ethanol. Investigations on living specimens had to be terminated during the fifth day after hatching since facilities were not available to rear the veligers for a longer period of time. The number of egg capsules in a spawn was counted, and ten capsules were removed from one egg mass. Each of these capsules was opened into a dish. The embryos were washed with ethanol and dispersed on the counting frame on the bottom of the dish. When the ethanol had evaporated, the embryos were counted. The different stages of veligers fixed in alcohol were measured (about 20 shells represented every day of development) and photographed with the aid of an SEM (Geological-Palaeontological In- stitut, Hamburg). The material is deposited under M 1035 at the Zoo- logical Institut and Museum (University of Hamburg). RESULTS Cabestana spenglert (Figure 1) lives mainly on rocks ex- posed at low tide (LAXTON, 1969; personal observations) but also occurs down to 30 m depth (POWELL, 1979). Cabestana spengleri feeds on ascidians (LAXTON, 1969; per- sonal observations). Spawning and breeding occurs from October to December (personal communication from Phil Page 118 The Veliger, Vol. 35, No. 23240 Nit MIB PLSEESELADALLR Lap F. Riedel, 1992 Colman; personal observations). This period applies only to the Long Reef region because water temperature influ- ences the start of the breeding season. For example, at Leigh (northern New Zealand), females of C. spengleri spawn and breed from November to January (LAXTON, 1969). The egg mass of Cabestana spengleri (Figure 2) resem- bles a bowl (see also SIMROTH, 1907; LAXTON, 1969; Douc as, 1985—not correct). The size is proportional to the diameter of the shell aperture of the snail that produces the egg mass (LAXTON, 1969; personal observations). ANDERSON (1959) figured a specimen of C. spengleri to- gether with an egg mass, but his photographs show a spawn that is too large for the specimen. Anderson assumed that the egg mass belonged to C. spengler: because the gastropod was attached to it. Actually C. spengleri remains on the spawn until the embryos hatch (LAxTON, 1969; personal observations), a behavior that seems typical for the whole family (BANDEL, 1976; BANDEL & WEDLER, 1987). An- derson cautiously wrote that he suspected the collected egg masses belonged to this species. The spawn of Cabestana spengleri contains between 200 and 350 egg capsules (LAXTON, 1968). The number of egg capsules is not proportional to the diameter of the spawn (LAXTON, 1969). The egg capsule (Figure 3) of a spawn containing 224 capsules holds an average of approximately 4000 embryos, with a range of 3000 to 5000. These num- bers indicate that a single spawn of C. spengleri houses about one million embryos. When hatching commences, an average of one to two veligers per second leave the egg mass, since it takes about one week until all egg capsules are empty. When the ve- ligers start their planktonic life, the shell has reached near- ly one whorl. This part of the whorl is defined as the embryonic shell. Growth can be seen in Figure 4a. This veliger was swimming free for three days. The beginning of growth lines marks the time of hatching. The diameter of the embryonic shell varies around 200 wm. The shape is slightly asymmetrical (Figure 4b), and the reticulate sculpture seems to be typical for the Tonnoidea (BANDEL, 1975). The velum (Figure 5) is bilobate. Each lobe is set with about one hundred 50-um long cilia. The shape of the foot is typical for the Tonnoidea (PILKINGTON, 1976). The foot is also ciliated, but the cilia are much shorter Page 119 than those of the velum. The head possesses two tentacles with eyes at their base. A veliger can reach a speed of 5 mm/sec, which is equivalent to traveling 25 times its own length in one second. Figure 6 shows a two-day-old veliger that has slightly retracted its velum. The visceral mass is rounded and still full of yolk. The amount of yolk quickly decreases during the following days of development. Between the apical part of the visceral mass and the shell, the retractor muscle is visible. The heart is situated beneath the rear end of the pallial cavity. The head is less developed com- pared with that of a veliger one day older (7.e., the tentacles are shorter). The foot bears an operculum that is functional (see also Figure 4b). The completed protoconch (Figures 7, 8) has more than four whorls. DISCUSSION The distribution of Cabestana spengleri includes the south- ern part of eastern Australia as far north as southern Queensland, the coasts of Victoria and South Australia, Tasmania (WILSON & GILLET, 1971) and New Zealand and “surrounding” islands (POWELL, 1979), from the northern Three Kings Islands to southern Stewart Island, and the Chatham Islands. For a gastropod with direct development not known to live deeper than 30 m, it would be difficult to explain such a distribution, on both sides of the Tasman Sea, with populations separated by up to 2000 km of deep ocean. Assuming that the planktotrophic veliger grows as fast as it did during the first days of development (see above) it would need about three months to construct three whorls of larval shell. This is of course a daring assumption be- cause nothing is known about the special circumstances of the veliger on its journey; however, other members of the Ranellidae (SCHELTEMA, 1966, 1971) have larval lives of three months to a year (maybe more), and show that the possibility exists. It seems likely that ANDERSON (1959) either described the early life history of some other gastropod in the egg mass on which Cabestana spengleri happened to be found, or the development within the egg mass of C. spengleri was abnormal in this case (perhaps caused by poor tank con- Explanation of Figures 1 to 6 Figure 1. The teleoconch of Cabestana spengleri. Scale bar: 10 cm. Figure 2. The spawn of C. spengleri in dorsal view and the outline of the side view. Scale bar: 2.5 cm. Figure 3. The egg capsule of C. spengleri, showing the opening on top through which the veligers hatch. Scale bar: 1 cm. Figure 4a, b. Shell of a three-day-old veliger of C. spengleri. Figure 5. Three-day-old veliger of C. spengleri, anterior view. Scale bar: 0.2 mm. Figure 6. Two-day-old veliger of C. spengleri, lateral view. Scale bar: 0.2 mm. Page 120 The Veliger, Vol. 35, No. 2 © Explanation of Figures 7 and 8 Figure 7. Juvenile Cabestana spengleri showing the protoconch, drawn after KESTEVEN (1901). Scale bar: 3 mm. Figure 8. Apical view of the protoconch of a juvenile C. spengleri, showing the number of whorls (drawn from an SEM photo). Scale bar: 1.5 mm. ditions). The second possibility could be supported by An- derson’s figures of the embryo, which show unusual char- acters (e.g., the huge “oral hood”). Of great importance is the fact that a gastropod with a protoconch, indicating a planktotrophic veliger (KESTE- VEN, 1901), actually has this stage in its ontogeny. Galeodea (= Cassidaria) echinophora (Linnaeus, 1758) is one member of the Tonnoidea in which direct development is confirmed (FIORONI, 1966) and its protoconch consists of not more than one whorl (ABBOTT, 1968; WAREN & BOUCHET, 1990). ACKNOWLEDGMENTS I would like to thank Klaus Bandel (GPI Hamburg, Ger- many) and Alan Beu (Geological Survey, Lower Hutt, N.Z.) for constructive criticism. Phil Colman (Australian Museum) guided me at Long Reef and helpfully organized arrangements at the Division of Invertebrate Zoology. Bob Penniket (Warkworth, N.Z.) was very generous and do- nated, for further comparison, shells of juvenile and adult Cabestana spengleri. Winston Ponder (Australian Muse- um) made it possible for me to work at his department and gave much good advice. Patricia Ward did not resign when correcting style and grammar. Finally I thank the Deutsche Forschungsgemeinschaft for financially sup- porting this research project. LITERATURE CITED ABBOTT, R. T. 1968. The helmet shells of the world (Cassidae). Part 1. Indo-Pacific Mollusca 2(9):7-—202. ANDERSON, D. T. 1959. The reproduction and early life history of the gastropod Cymatilesta spengleri (Perry) (fam. Cyma- tidae). Proceedings of the Linnean Society of New South Wales 84:232-238. BANDEL, K. 1975. Embryonalgehduse karibischer Meso- und Neo-gastropoden (Mollusca). Akademie der Wissenschaften und der Literatur/Abhandlungen der Mathematisch-Na- turwissenschaftlichen Klasse 1975, Nr. 1:1-133, pls. 1-21. BANDEL, K. 1976. Die Gelege karibischer Vertreter aus den uberfamilien Strombacea, Naticacea und Tonnacea (Meso- gastropoda) sowie Beobachtungen im Meer und Aqarium. Mitteilungen des Institutes Colombo-Aleman Investiga- ciones Cientificas 8:105-139. BANDEL, K. 1981. Struktur der Molluskenschale im Hinblick auf ihre Funktion. Palaontologische Kursbucher, Funktions- morphologie, Bd. 1:25-47. BANDEL, K. 1991. Ontogenetic changes reflected in the mor- phology of the molluscan shell. Pp. 211-230. In: Schmidt- Kittler and Vogel (eds.), Constructional Morphology and Evolution. Springer: New York. BANDEL, K. & E. WEDLER. 1987. Hydroid, amphineuran and gastropod zonation in the littoral of the Caribbean Sea, Co- lombia. Senckenbergiana maritima 19(1/2):1-129. Bev, A. G. 1978. New records and species of Cymatiidae (Gas- tropoda: Prosobranchia) from the Kermadec Islands, Norfolk Ridge and New Zealand. Journal of the Malacological So- ciety of Australia 4(1/2):29-42. Beu, A. G. 1986. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part 2. Descrip- tions of 14 new modern Indo-West Pacific species and sub- species, with revisions of related taxa. N.Z. Journal of Zo- ology 13:273-355. Beu, A. G. 1988. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part 5. Early his- tory of the families, with four new genera and recognition of the family Personidae. Saito Hoon Kai Special Publication (Prof. T. Kotaka Commem. Vol.):69-96. Beu, A. G. & A. K. Kay. 1988. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part 4. The Cymatium pileare complex. Journal of the Royal Society of N.Z. 18(2):185-223. Beu, A. G. & J. KNUDSEN. 1987. Taxonomy of the families Ranellidae (= Cymatiidae) and Bursidae. Part 3. A review of the trifid-ribbed species of Cymatium (Turnitriton). Journal of the Royal Society of N.Z. 17(1):73-91. epRIedel lO O2 CLENCH, W. J. & R.D. TURNER. 1957. The family Cymatiidae in the Western Atlantic. Johnsonia 3(36):189-244. Douc.tas, N. 1985. Periostracum preservation. Poirieria 14(2): 6-12. FIoRONI, P. 1966. Zur Morphologie und Embryogenese des Darmtraktes und der transitorischen Organe bei Proso- branchiern (Mollusca, Gastropoda). Revue Suisse de Zool- ogie 73:621-876. Houpsrick, J. R. & V. FRETTER. 1969. Some aspects of the functional anatomy and biology of Cymatium and Bursa. Proceedings of the Malacological Society of London 38:415- 429. KESTEVEN, H. L. 1901. The protoconchs of certain Port Jack- son Gasteropoda. Proceedings of the Linnean Society of New South Wales 26(4):709-717. KESTEVEN, H. L. 1902. Notes on Prosobranchiata. Proceedings of the Linnean Society of New South Wales 27(3):443-484. Kiuias, R. 1973. Cymatiidae. Das Tierreich, Lieferung 92:I- VIII, 1-235. Walter de Gruyter: Berlin. LaurRSEN, D. 1981. Taxonomy and distribution of teleplanic prosobranch larvae in the North Atlantic. Dana Report 89: 1-43, pls. 1-3. LAxTON, J. H. 1968. The anatomy, feeding, growth and re- production of some New Zealand Cymatiids. M.S. Thesis, University of Auckland. 99 pp. LaxTon, J. H. 1969. Reproduction in some New Zealand Cymatiidae (Gastropoda: Prosobranchia). Zoological Jour- nal of the Linnean Society of London 48:237-253, pls. 1, 2. Lewis, H. 1974. A new species of Hawaiian Gyrinewm (Cy- matiidae). The Nautilus 88(1):10-13. Page 121 PILKINGTON, M. C. 1974. The eggs and hatching stages of some New Zealand prosobranch molluscs. Journal of the Royal Society N.Z. 4(4):411-431. PILKINGTON, M. C. 1976. Descriptions of veliger larvae of monotocardian gastropods occurring in Otago plankton hauls. Journal of Molluscan Studies 42:337-360. PowELL, A. W. B. 1979. New Zealand Mollusca. Marine, Land and Freshwater Shells. Collins: Auckland. RICHTER, G. 1984. Die Gehauseentwicklung bei den Larven der Cymatiden (Prosobranchia: Tonnacea). Archiv fur Mol- luskenkunde 115(1/3):125-141. SCHELTEMA, R.S. 1966. Evidence for trans-Atlantic transport of gastropod larvae belonging to the genus C'ymatium. Deep- Sea Research 13:83-95. SCHELTEMA, R. S. 1971. The dispersal of the larvae of shoal- water benthic invertebrate species over long distances by ocean currents. Fourth European Marine Biology Sympo- sium:7-28. Cambridge University Press. SIMROTH, H. 1896-1907. Gastropoda Prosobranchia. In: Bronn’s Klassen und Ordnungen des Tierreichs. Dritter Band. Mollusca. Leipzig. 1056 pp., pls. 1-63. WarEN, A. & P. BOUCHET. 1990. Laubierinidae and Pisani- anurinae (Ranellidae), two new deep-sea taxa of the Ton- noidea (Gastropoda: Prosobranchia). The Veliger 33(1):56- 102. WILSON, B. R. & K. GILLET. 1971. Australian Shells. A. H. & A. W. Reed: Sydney. The Veliger 35(2):122-132 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 The Fossil Land Snail Helix leidy Hall & Meek, 1855, a Member of a New Genus of Humboldtianidae (Gastropoda: Pulmonata) EMMETT EVANOFF Geology Section, University of Colorado Museum, Campus Box 315 Boulder, Colorado 80309-0315, USA BARRY ROTH Museum of Paleontology, University of California, Berkeley, California 94720, USA Abstract. The latest Eocene to earliest Oligocene terrestrial snail Helix leidy: Hall & Meek, 1855, is redescribed, based on a re-examination of the holotype and well-preserved specimens from east-central Wyoming. Much early identification of the species was based on a specimen erroneously accepted as the holotype. Helix leidyi is designated the type species of a new genus of Humboldtianidae, Skinnerelix. Skinnerelix differs from the extant genus Humboldtiana von Ihering, 1892, in having a larger number of whorls, a higher spire, and a distinctly everted peristome with a slight constriction behind the basal lip. Humboldtiana may have originated from Skinnerelix progenetically. (Humboldtianidae, Helicidae) probably represents a Laurasian, cratonal clade with a history long independent of the Gondwanan, accretional clade consisting of (Helminthoglyptidae, Bradybaenidae, Xanthonychidae). INTRODUCTION Helix leidyi Hall & Meek, 1855, was the first fossil land snail species described from the western United States and has become one of the most widely cited but poorly un- derstood species of middle Tertiary land snails. The ho- lotype of Helix leidy: was collected in 1853 by F. B. Meek and F. V. Hayden in the badlands of the White River “series” in the Nebraska Territory (now western South Dakota). Subsequent to its description by HALL & MEEK (1855), classification and identification of the species were confused by the designation of an erroneous “holotype” and widespread use of the name for any large helicoid snail fossils from rocks of medial Tertiary age in the mid-con- tinent region. Re-examination of the holotype and discov- ery of well-preserved specimens in east-central Wyoming (Figures 1-7) indicate that Helix lezdy: is a member of a new genus of the Humboldtianidae. Its presence in latest Eocene and earliest Oligocene rocks of the Great Plains and central Rocky Mountains (Figure 8) has significant implications for biogeographic hypotheses concerning Humboldtianidae and Helicoidea. The following abbreviations are used: AMNH, Amer- ican Museum of Natural History, New York; UCM, Uni- versity of Colorado Museum, Boulder; USNM, National Museum of Natural History, Washington, D.C.; USGS, United States Geological Survey. E. Evanoff & B. Roth, 1992 Page 123 Explanations of Figures 1 to 7 Figures 1-7: Skinnerelix leidy: (Hall & Meek, 1855). Figures 1-3. Holotype, AMNH 11174/1, from the Scenic Member, Brule Formation, White River Group (lower Oligocene), near Scenic, Pennington County, South Dakota. Side, apertural, and oblique spire view; specimen coated for photographing; height 24.3 mm. Arrow on Figure 3 indicates point marked by arrow in Figure 9. Figures 4-7. Figured specimen, UCM 30732, from Chadron Member, White River Formation (upper Eocene), near Douglas, Converse County, Wyoming (UCM loc. 90004). Apertural, side, spire, and basal views; specimen coated for photographing; maximum diameter 31.6 mm. Page 124 SYSTEMATIC PALEONTOLOGY Class Gastropoda Subclass Pulmonata Superorder Stylommatophora Order Sigmurethra Superfamily HELICOIDEA Family HUMBOLDTIANIDAE Skinnerelix Evanoff & Roth, gen. nov. Type species: Helix leidyi Hall & Meek, 1855. Generic diagnosis: Shell large (adult shells larger than 2 cm in maximum diameter); globose-conic; height to max- imum diameter ratio greater than 0.8; embryonic shell smooth, consisting of first 1.2 whorls; neanic sculpture of growth rugae and coarse, somewhat crude granulation ar- ranged in diagonal rows; last whorl broadly rounded, tu- mid, descending, constricted basally just behind lip; peri- stome flaring, narrowly turned outward, reflected at base and columella; base narrowly, obliquely perforate. Remarks: Several features indicate assignment of Skin- nerelix to the Humboldtianidae. It shares the following features with the extant genus Humboldtiana von Ihering, 1892: a globose-conic, narrowly perforate shell; fewer than five whorls; an embryonic shell of 1.2 whorls; coarse, rather crude granulation; and a tumid, descending last whorl. The granulation of Skinnerelix consists of close-set, round to ovate granules in diagonal rows relative to the growth lines and collabral rugae (Figures 9-11). Granulation of this type occurs in many species of Humboldtiana, including H. chisosensis Pilsbry, 1927, H. globosa Burch & Thomp- son, 1957, H. palmer: Clench & Rehder, 1930 (Figure 12), and H. texana Pilsbry, 1927. It differs from the fine, reg- ular, fabric-like granulation found in some species of the helminthoglyptid genus Xerarionta Pilsbry, 1913 (e.g., Xerarionta redimita (Binney, 1858); see ROTH, 1984:fig. 32). In some species of Helminthoglypta Ancey, 1895, a still different form of granulation occurs in which close- set collabral rugae are cut into rows of elongate granules by incised spiral striae or shallow, forwardly descending sulci. Skinnerelix differs from Humboldtiana in having more inflated whorls, a slightly larger number of whorls (more than 4.1 whorls compared with typically fewer than 4.1 whorls for Humboldtiana), a higher spire, and a distinctly everted peristome with a slight constriction just behind the basal lip. Features of adult Humboldtiana shells, such as the lack of an everted peristome and the small number of whorls, are characteristics of Skinnerelix shells that have not attained their full, adult growth, and suggest that Hum- boldtiana may have originated from a Skinnerelix-like ancestor by a process of progenesis. Skinnerelix occurs in upper Eocene and lower Oligo- cene rocks of the Big Badlands of South Dakota, the Pine The Veliger, Vol. 35, No. 2 Ridge of northwestern Nebraska, near Douglas, Wyoming (all, S. lecdyr), and in the Keetley Volcanics, near Peoa, northeastern Utah (S. sp., cf. S. leidy:) (Figure 8). Hum- boldtiana is predominantly a genus of the Mexican Plateau (Figure 13), ranging from Mexico City in the south to the Guadalupe Mountains of New Mexico in the north (BURCH & THOMPSON, 1957; BEQUAERT & MILLER, 1973). We have examined the holotype of Humboldtiana? tuck- erae Mansfield, 1937, USNM 495934 (not 4959340, as originally published), from the Tampa Limestone, upper Oligocene of Florida. The outer lip is smoothly rolled outward a short distance, with a small internal varix that is not reflected externally in any constriction of the whorl. The inner lip is triangularly dilated over the umbilicus leaving an open, tubular, oblique perforation. Behind the aperture, the body whorl does not depart from the prior whorl trajectory and in fact descends very little. The embryonic sculpture consists of low, obscure ver- miculation overlain by widely spaced, round, flat-topped papillae. The post-embryonic sculpture consists of indis- tinct collabral ribs, slightly nodulose, and separated by interspaces of about the same width; small patches of short, axially elongated indentations are present. No sculpture of close-set, round to ovate granules in diagonal rows is present. On the basis of these observations, we do not consider Humboldtiana? tuckerae assignable to Skinnerelix. F. G. Thompson (in UNDERWOOD & WILSON, 1974) referred H.? tuckerae to the genus Cepolis Montfort, 1810. UNDERWOOD & WILSON (1974) reported an unnamed species of Humboldtiana from the Garren Group, Hud- speth County, Texas, found in association with Chadron- ian Age land mammals in strata radioisotopically dated at 39-36 Ma. The specimens are crushed and distorted, with little shell surface and no adult apertures preserved. They are probably juvenile. The largest is 17+ mm in diameter, with about 3.3 whorls. Without better material, it is not possible to state whether these specimens represent Hum- boldtiana, Skinnerelix, Xerarionta waltmilleri Roth, 1984, as suggested by ROTH (1984), or another taxon. Etymology: The name Skinnerelix combines the Greek word helix, a spiral, hence a snail, and the name of the late Dr. Morris Skinner, collector for the American Mu- seum of Natural History. Skinner’s collections of White River land snails for the AMNH are the largest in the country, and his stratigraphic studies are the basis for many of our modern concepts of White River correlations. The gender of Skinnerelix is feminine. Skinnerelix leidy: (Hall & Meek, 1855) (Figures 1-7, 9-11) Helix leidyi HALL & MEEK, 1855:394, pl. 3, fig. 12a, b; MEEK, 1876:604-605, pl. 45, fig. 7a, b. ?Helix leidyi Hall & Meek: WHITE, 1877:211 (in part), pl. 21, fig. 3a, b. E. Evanoff & B. Roth, 1992 Page 125 A 100 200 km White River rocks and equivalents Mountain range Ce Sle SNES AN SEAL ~'7, 1 Scenic Epoch oe South Dakota Brule Formation Poleslide Member Scenic Member > | v Chadron Chadron Formation Se 31 o 12) (= ; ow : E| 52 |8 325 6 ) ° fe) G So ee) © = £ 2 ra 2 34 = c = {qv 35 Cc ° S 36 fe : : Figure 8 Distribution (A) and chronostratigraphy (B) of Skinnerelix localities. Helix (Arianta?) leidy: Hall & Meek: WHITE, 1883:455, 475, pl. 21, fig. 3c; COCKERELL & HENDERSON, 1912:232, pl. 32, figs. 32, 33. pl. 22, figs. 1-3; PAMPE, 1974:292, pl. 1, figs. 1-10. [Polygyra] leidyi (Hall & Meek): Hanna, 1920:9. Non Mesodon leidyi (Hall & Meek): Russell, in GOLDICH & Polygyra leidy: (Hall & Meek): TOEPELMAN, 1922:65. ELMs, 1949:1145. Pseudolisinoe leidy: (Hall & Meek): WENZ, 1923:116. Glypterpes leidyi (Hall & Meek): ZILcH, 1960:655. Diagnosis: Shell large, spire broad, whorls inflated and Non Helix leidyi Hall & Meek: WHITE, 1877:211 (in part), shouldered; embryonic shell smooth, regularly increasing Page 126 The Veliger, Vol. 35, No. 2 Explanation of Figures 9 to 12 Figure 9. Skinnerelix leidyi, holotype, AMNH 11174/1. SEM photograph of diagonal granular microsculpture on apical side of last whorl, just before crushed area (see arrow on Figure 3). Bar is 1 mm long; epoxy cast of specimen coated for photographing. Figure 10. Skinnerelix leidyi, figured specimen, UCM 30732. SEM photograph of diagonal granular microsculpture on last and penultimate whorls, about 0.2 whorls behind aperture. Bar is 1 mm long; epoxy cast of specimen coated for photographing. E. Evanoff & B. Roth, 1992 in diameter after nucleus; sculpture of coarse granulations arranged in diagonal rows extending from adapical side to base; last whorl tumid, gradually descending in last 0.2 to 0.25 whorl. Original description: “Shell subglobose, wider than long; spire elevated; volutions four or five, last one large and ventricose; suture distinct; surface unknown; aperture un- known; outer lip reflected; umbilicus small, or perhaps closed. The last volution .65 of whole length. The aperture is ovate, subangular behind” (HALL & MEEK, 1855:394, 411). Description of holotype: Shell large, globose-conic, very narrowly perforate. Spire slightly convex in profile, with apical angle of 121°, sutures moderately impressed, whorls shouldered. Sculpture on penultimate and last whorls in- cluding retractive, moderately prominent growth lines and granulations arranged in diagonal rows. Last whorl tumid, rounded, descending in last 0.2 whorl; base slightly con- stricted upward behind lip. Peristome everted; columellar lip recurved, dilated over umbilical perforation. Type material: Holotype: AMNH 11174/1 (James Hall number 5547/1). South Dakota, Pennington County: near the head of Bear Creek, Mauvaises Terres, turtle and bone bed (HALL & MEEK, 1855:394) collected by Meek and Hayden in 1853. About E %, T. 3 S., R. 13 E. (HARTMAN, 1984:907); from the Scenic Member, Brule Formation, White River Group; Orellan Land Mammal Age. Referred material (all near Douglas, Converse County, Wyoming): UCM 30732 (figured), UCM locality 90004; UCM 30753, UCM locality 87063; UCM 30733, UCM locality 90004; UCM 30734, UCM locality 90005; UCM 30735, UCM locality 83235. Occurring 69.8 to 62.2 m below the top of the Chadron Member, White River For- mation; Chadronian Land Mammal Age. Additional description of referred material: The re- ferred specimens from the Douglas area, Wyoming, are similar in size and identical in shape, sculpture of the penultimate and last whorls, and peristome morphology to the holotype, but are better preserved. The ratio of spire height to shell height in the referred specimens ranges from 0.25 to 0.31 (mean 0.27). The embryonic shells of the referred specimens consist of 1.2 smooth whorls, separated from the neanic whorls by a slight constriction, increasing whorl translation rate, and the beginning of distinct growth Page 127 120 110 100 90 80 CoO rn Skinnerelix spp. . fa Lysinoe breedlovei iN upper Eocene S ae lower Oligocene Seal Figure 13 Distributions of modern and fossil species of Skinnerelix, Lysinoe, and Humboldtiana. Lysinoe distribution from DALL (1897), THOMPSON (1963), ROTH (1984), and unpublished museum re- cords. Humboldtiana distribution from PILSBRY (1927, 1939, 1948), SOLEM (1954, 1955), BURCH & THOMPSON (1957), and BEQUAERT & MILLER (1973). lines. The nucleus has a typical width of 0.5 mm; the embryonic shell is not distinctly inflated after the nucleus. The neanic whorls are rounded, with granulations starting at about whorl 2, initially weak, becoming prominent after 2.75 whorls. The last whorl gradually descends in the last 0.2 to 0.25 whorl, with the lower palatal limb becoming increasingly expanded. The adapical and basal sides are slightly constricted just before the lip. Granulations are coarse and prominent on the adapical and palatal sides, weak on the base, and absent in the shallow umbilical area. The growth lines coalesce to form weak collabral rugae on the last whorl. The aperture is rounded, ovate- lunate; the outer lip is narrowly expanded on the adapical and marginal sides, recurved basally. The columellar lip is dilated and reflected. The parietal wall has a simple callus pad. Figure 11. Skinnerelix leidyi, figured specimen, UCM 30732. SEM photograph of protoconch and coarse granular microsculpture on whorls 2 and 3. Bar is 1 mm long; epoxy cast of specimen coated for photographing. Figure 12. Humboldtiana palmeri Clench & Rehder, 1930, Recent, USNM 408371/1, Davis Mountains, Jeff Davis County, Texas. SEM photograph of protoconch and coarse granular microsculpture arranged in diagonal rows relative to growth lines. Bar is 1 mm long; specimen coated for photographing. Page 128 A. Hall and Meek [1855] Figure 14 Original lithograph illustrations of Helix leidyi. A. Apertural and side views of holotype of Helix leidyi (AMNH 111774/1) as illustrated by HALL & MEEK (1855:pl. 3, fig. 12a, b). B. Ap- ertural and spire views of Helix leidyi(?) (USNM 2102) as il- lustrated by MEEK (1876:pl. 45, fig. 7a, b). Specimen 26.0 mm high, maximum diameter 26.2 mm. Stippled areas represent parts of the specimen reconstructed with beeswax. Remarks: “Helix” leidyi has been the subject of taxonomic confusion. The holotype (AMNH 111774/1) (Figures 1- 3), illustrated by HALL & MEEK (1855; their illustration reproduced here as Figure 14A), is mostly an internal mold of an adult shell, with the embryonic shell and most of the spire poorly preserved. The adapical and marginal sides of the last 0.25 whorl are crushed, but shell is preserved on the adapical side of the last whorl just before the crushed area, and above this on the penultimate whorl. Despite preservational defects, the type specimen has enough fea- tures to distinguish the species from other Tertiary heli- coids. In the National Museum of Natural History is a spec- imen of “Helix” leidyi (USNM 2102) which is labeled “holotype,” but is not the specimen illustrated by HALL & MEEK (1855). This specimen is an internal mold with no original shell and has been so over-prepared that the surface of the spire has been sculpted. The last 0.2 whorl was reconstructed with beeswax (Figure 14B). The wax reconstruction does not represent the descent of the last whorl, indicated by a slight downward reflection of the suture, and does not have a reflected lip, which is clearly preserved on the actual holotype. USNM 2102 was first illustrated by MEEK (1876) without reference to the re- constructed aperture. Unfortunately, this specimen is the one regarded by WHITE (1883), COCKERELL (1915), and subsequent workers as the type specimen of “Helix” lezdyz, The Veliger, Vol. 35, No. 2 producing confusion in the taxonomy of the species. For example, COCKERELL (1915) could not distinguish this specimen from species of Glypterpes Pilsbry, 1892. On this basis, WENZ (1923) subsequently included “#7.” lezdyz in his genus Pseudolisinoe, a junior synonym of Gypterpes. The specimens referred to as “Helix leidyi” by WHITE (1877) include one specimen (USNM 484) that may be assignable to Skinnerelix but has been greatly modified by preparation, and another (USNM 829) that is openly umbilicate and has a dome-shaped shell, features not found in S. leidyz. COCKERELL & HENDERSON (1912:pl. 22, figs. 1-3) il- lustrated three specimens identified as “Helix leidy1.”” One of these (AMNH 43562) has a depressed spire and a distinct umbilicus, and is not a species of Skinnerelix. It is probably a member of an undescribed taxon of Whit- neyan helicoids characterized by a moderately depressed spire and shouldered early whorls. The other two speci- mens (AMNH 43561, 43563) are poorly preserved inter- nal molds and not identifiable. PAMPE (1974) discussed and illustrated specimens he identified as “Helix leidy:” from the Eocene and Oligocene of west Texas, but ROTH (1984) reassigned them to Lysinoe breedlover Roth, 1984, and Xerarionta waltmiller: Roth, 1984. The systematic position of “Helix” lerdy: has long been uncertain. Early workers (HALL & MEEK, 1855; MEEK, 1876; WHITE, 1883; COCKERELL, 1915) placed all large fossil helicoid-shaped snail shells from North America in the Helicidae, under the genera Helix Linnaeus, 1758, or Arianta Leach, 1831. HENDERSON (1935) and LA ROCQUE (1960) continued to refer to fossil helicoids as Helix? or “Helix,” stressing the difficulties in determining generic position from shell features alone. The family Helicidae is now considered to be native to Eurasia and North Africa. Species of Helix and Arianta have domed spires with weak- ly impressed sutures, tumid embryonic shells, and sculp- ture of spiral striations and malleations, features not found in Skinnerelix. HANNA (1920) referred “Helix” leidy: to the Polygyridae (although not explcitly to any one genus) because of its similarity to Polygyra martini Hanna, 1920, of the John Day Formation in Oregon. ‘TOEPELMAN (1922), assigned Skinnerelix leidyi to Polygyra Say, 1818. Russell (in GOLDICH & ELMS, 1949) placed the species in the poly- gyrid genus Mesodon Rafinesque, 1821. Globose-conic polygyrids without apertural barriers, such as certain spe- cies of Mesodon, Neohelix von Ihering, 1892, and Allogona Pilsbry, 1939, have one or more of the following features: a domed spire, spiral striae, distinct radial ridges or mal- leations, a broadly expanded peristome, an ovate-lunate aperture, and a strong basal preapertural constriction that causes the basal wall to rise adapically. Skinnerelix leidy has none of these features. Furthermore, ROTH (1987) demonstrated that P. martini is assignable to Helmintho- glypta, not to the Polygyridae. Helminthoglypta martini E. Evanoff & B. Roth, 1992 differs from Skinnerelix leidy: by being more widely um- bilicate and having sculpture of retractive slanting riblets, spiral striae, and malleation. COCKERELL (1915), examining figured specimens of Skinnerelix leidy: at the National Museum of Natural History, decided that the species was not generically dis- tinguishable from the Eocene taxa “Helix veterna Meek & Hayden, 1861, “Helix” riparia White, 1876, and “He- lix’? veterna veternior Cockerell, 1915. Following COCKERELL (1915) in this respect, WENZ (1923) grouped these taxa into a new genus, Pseudolisinoe, with “H.” ve- terna as the type species. (Pseudolisinoe is an objective syn- onym of Glypterpes Pilsbry, 1892, type species “H.” ve- terna.) From an examination of the holotypes of Glypterpes veterna and G. riparia, we regard Skinnerelix and Glyp- terpes as distinct. Species of Glypterpes have more conical shells, less inflated whorls, spiral striation instead of gran- ulation, and less everted peristomes than species of Skin- nerelix. The modern species of Humboldtiana that most closely resembles Skinnerelix leidy: is H. globosa from the state of Vera Cruz, Mexico. Both species have highly inflated, shouldered whorls, smooth embryonic shells, and granu- lations arranged in diagonal rows relative to the growth rugae; both are of similar size. Distribution and stratigraphic range: Scenic Member, Brule Formation, White River Group, Big Badlands, Pen- nington County, South Dakota (type locality); Orellan Land Mammal Age. Chadron Formation, White River Group, within 6 m below the top of the Chadron, Sioux County, Nebraska (USGS Cenozoic localities 20025, 22672); late Chadronian Land Mammal Age. Chadron Member, White River Formation, between 71.3 to 9.0 m below the top of the Chadron Member near Douglas, Wyoming (EVANOFF, 1990); late Chadronian Land Mam- mal Age. The species ranges from the latest Eocene to earliest Oligocene (late Chadronian to Orellan land mam- mal ages). Skinnerelix leidyi is common and the most obvious land snail fossil of the Chadron Member in the Douglas area. The specimens of Skinnerelix sp., cf. S. leidyi from the upper Eocene Keetley Volcanics (dated between 35 and 34 Ma; HINTZE, 1988) near Peoa, Summit County, Utah (USGS Cenozoic locality 23306) are too incomplete to be referred to S. /ezdy: with certainty. BIOGEOGRAPHIC anp PALEOCLIMATIC SIGNIFICANCE Leaving out the problematic Humboldtiana of UNDERWOOD & WILSON (1974), the fossil record of Humboldtianidae consists of the occurrences of Skinnerelix described above and Lysinoe breedlove: Roth, 1984, from the upper Eocene of Trans-Pecos Texas and the lower Oligocene of Nuevo Leon, Mexico. In western Texas, L. breedlovei ranges from the lower part of the Devils Graveyard Formation (as- sociated with the Uintan Whistler Squat local fauna) to Page 129 the Bandera Mesa Member of the Devils Graveyard For- mation (associated with the early Chadronian Coffee Cup local fauna) (ROTH, 1984). All of these occurrences are older than the occurrences of Skinnerelix in the central Rocky Mountains and western Great Plains. Lysinoe breedlovei also occurs in lower Oligocene rocks of Nuevo Leon, Mexico (ROTH, 1984) that are equivalent to the Vicksburg Group (GARDNER, 1945), and about the same age as the Orellan Land Mammal Age (SWISHER & PRO- THERO, 1990). By the late Eocene, at least two humbold- tianid clades, that of Lysinoe and that of (Humboldtiana, Skinnerelix), were differentiated. Species of Lysznoe now live predominantly south of the range of Humboldtiana (Figure 13), and Lysinoe breedlovei also lived south of Skinnerelix. Both clades have shifted southward since the early Oligocene, but their relative distribution has remained constant. Humboldtiana is a plausible environmental analogue to Skinnerelix. Humboldtiana ranges from Mexico City, Mexico, north to the Guadalupe Mountains in southeast New Mexico (Figure 13). It typically lives in a variety of substrates and woody vegetation, ranging from oak forests on limestone to high coniferous forests and mixed scattered woodlands on volcanic rocks (PILSBRY, 1939). The climate in the range of Humboldtiana is subtropical, as defined by WOLFE (1979), with mean annual temper- atures ranging from 13 to 21°C, and mean annual range of temperatures of 4-20°C (WERNSTEDT, 1972; WORLD METEOROLOGICAL ORGANIZATION, 1979). The presence of a land snail with subtropical affinities in the Rocky Mountains during the late Eocene is consistent with pa- leoclimatic interpretations based on the contemporaneous Florissant flora of Colorado (MACGINITIE, 1953). The classification of the Helicoidea is in a state of flux (along with the rest of the Stylommatophora; see EM- BERTON et al., 1990), with opinions varying as to the re- lationships of the Humboldtianidae (e.g., SCHILEYKO, 1978, 1979, 1991; NORDSIECK, 1987; TILLIER, 1989). A robust phylogenetic hypothesis is only now coming into place for the families of Helicoidea. (The wide-ranging paper of SCHILEYKO [1991] was published almost simultaneously with submission of the present paper; Roth [in prepara- tion] has some rather different ideas about the genera of Helminthoglyptidae.) We regard the Humboldtianidae as a holophyletic group (sensu ASHLOCK, 1971), defined by the synapomorphies of a ring of nodular mucus glands surrounding the vagina at a single level, subtended by a ring of dart sacs sessile on the vagina. In the view of SCHILEYKO (1978, 1991; personal com- munication to Roth, 1990) the condition of multiple mucus glands seated around the vagina and multiple sessile dart sacs is primitive relative to the smaller number of dart sacs (generally one) and more closely adjacent mucus glands found in Helicidae and in secondarily simplified genera such as Leptarionta Fischer & Crosse, 1872. It is mor- phologically closer to the hypothesized ancestral condition Page 130 in which the vaginal wall is extensively glandular and furnished with aragonitic spicules. The branching diagram of SCHILEYKO (1978:fig. 29) contains the clade ((Hum- boldtianidae, Helicidae), (Bradybaenidae, Helmintho- glyptidae)); but the apomorphies (if any) defining the branch segments are not specified. The cladogram of NORDSIECK (1987:fig. 30) includes Humboldtianidae within a heterogeneous family Xantho- nychidae, separated from (Bradybaenidae, (Hygromiidae, Helicidae)) by the single, equivocal character of “dart glands [= mucus glands] not divided/divided.” But in the bradybaenid genus Aegista Albers, 1850, for example, the mucus glands range from single to multiple, with variously one, two, or perhaps more insertions on the nebensack (AZUMA, 1982). In the helminthoglyptid genus Micrarionta Ancey, 1880, the glands are paired; in Helminthoglypta they could be said to be divided—there are two bulbous reservoirs leading into the common duct—even though they ultimately form a single membranous envelope around the dart apparatus. MILLER & NARANJO-GARCIA (1991) pointed out the correspondence between the distribution of the helicoid families Bradybaenidae, Helminthoglyptidae, and Xan- thonychidae and the tectonically accreted terranes around the Pacific Rim. From this pattern they drew the conclu- sion that those families had a common ancestry on a Me- sozoic Gondwanan land mass (‘‘Pacifica’”; see NUR & BEN-AVRAHAM, 1977; JONES et al., 1982) and were dis- persed passively to Asia and the Americas on rafting frag- ments of continental crust. They also noted that the dis- tribution of Humboldtianidae does not correspond to any accretional realm, and therefore excluded the family from the above scenario. Humboldtianidae must have had a history independent from that of the ‘“‘Pacifican”’ families. Recognition of Skinnerelix as a genus of Humboldtianidae does not alter, and in fact reinforces, the model in this respect. (Stratigraphic evidence is not well in accord with the “Pacifica” hypothesis for the origin of Helminthoglypti- dae. The earliest fossil occurrences of Helminthoglypta (H. bozemanensis Roth, 1986), Xerarionta (X. waltmiller:), and Polymita (P. texana Roth, 1984) are an old continent, not accreted terrane. However, a thorough review of the fossil evidence, taking into account the uneven distribution of fossiliferous terrestrial deposits of critical ages, has yet to be made.) Combining the phylogenetic hypothesis of SCHILEYKO (1978, 1991) and the historical zoogeography of MILLER & NARANJO-GARCIA (1991) produces the following model: (Humboldtianidae, Helicidae) represents one (Laurasian, cratonal?) clade, with vicariance between the two families possibly related to the development of the Atlantic Ocean as a dispersal barrier. (Helminthoglyptidae, Bradybaen- idae, Xanthonychidae) represents a second (Gondwanan, accretional?) clade, with possible vicariance related to the breakup of “Pacifica.” Sympatry between members of these The Veliger, Vol. 35, No. 2 two clades (e.g., Recent Sonorella and Humboldtiana in northern Mexico, on old continent) must therefore be the result of dispersal (in this example, presumably of the helminthoglyptid into humboldtianid territory). Dispersal of the Gondwanan clade in North America must have proceeded from accreted to cratonal terrane. Analysis of further characters of the snails themselves can (and should) be used to test the phylogenetic compo- nent of this model. The stratigraphic and geographic dis- tribution of the respective clades can be used to test, and develop a time scale for, the historical component. This model predicts that, in paleoenvironments that could sup- port snails of both clades, the home clade will appear stratigraphically below the first appearance of the dis- persing clade. ACKNOWLEDGMENTS Evanoff’s research was supported by funds from the Uni- versity of Colorado Museum, the Geological Society of America, Sigma Xi, the Smithsonian Institution, and the University of Colorado Department of Geological Sciences. He was assisted in the field by David Cunningham, Greg- ory Cunningham, and Logan Ivy. Barbara Tihen made casts of shells for SEM study. Malcolm McKenna and Norman Newell provided access to the collections at the AMNH. Robert Emry provided access to the collections at the USNM. Anatoly A. Schileyko shared with us many of his views on evolution within the Helicoidea. Tom Wal- ler and Warren Blow loaned the holotype of Humboldti- ana? tuckerae. This study was the result of a portion of Evanoff’s Ph.D. dissertation under the direction of Erle G. Kaufman. To all of these people we extend our ap- preciation. LITERATURE CITED ASHLOCK, P. D. 1971. Monophyly and associated terms. Sys- tematic Zoology 20:63-69. AzuMa, M. 1982. Colored Illustrations of the Land Snails of Japan. Hoikusha Publishing Company, Ltd.: Osaka. 333 PP- BEQUAERT, J. C. & W. B. MILLER. 1973. The Mollusks of the Arid Southwest, with an Arizona Checklist. University of Arizona Press: Tucson. 271 pp. Burcu, J. B. & F. G. THompson. 1957. Three new Mexican land snails of the genus Humboldtiana. University of Mich- igan Museum of Zoology, Occasional Papers, No. 590: 11 PP- COcKERELL, T. D. A. 1915. Gastropod Mollusca from the Tertiary strata of the west. American Museum of Natural History Bulletin 34:115-120. COocKERELL, T. D. A. & J. HENDERSON. 1912. Mollusca from the Tertiary strata of the west. American Museum of Nat- ural History Bulletin 31:229-234. DaLL, W. H. 1897. New species of Mexican land shells. The Nautilus 11:73-74. EMBERTON, K. C., G. S. Kuncio, G. M. Davis, S. M. PHILLIPS, K. M. MonDEREwIcz & Y. H. Guo. 1990. Comparison E. Evanoff & B. Roth, 1992 of recent classifications of stylommatophoran land-snail fam- ilies, and evaluation of large-ribosomal-RNA sequencing for their phylogenetics. Malacologia 31:327-352. EvanorF, E. 1990. Late Eocene and early Oligocene paleocli- mates as indicated by the sedimentology and nonmarine gas- tropods of the White River Formation near Douglas, Wy- oming. Unpublished Ph.D. Thesis, University of Colorado, Boulder. 440 pp. GARDNER, J. 1945. Mollusca of the Tertiary formations of northeastern Mexico. Geological Society of America Memoir 11:332 pp. Go.picH, S. S. & M. A. Etms. 1949. Stratigraphy and pe- trology of the Buck Hill Quandrangle, Texas. Geological Society of America Bulletin 60:1133-1182. HALL, J. & F. B. MEEK. 1855. Descriptions of new species of fossils from the Cretaceous Formations of Nebraska, with observations upon Baculites ovatus and B. compressus, and the progressive development of the septa in baculites, am- monites, and scaphites. American Academy of Arts and Sci- ences Memoirs, n.s. 5:379-411. Hanna, GD. 1920. Miocene land shells from Oregon. Kansas University Science Bulletin 13:3-11. HARTMAN, J. H. 1984. Systematics, biostratigraphy and bio- geography of latest Cretaceous and early Tertiary Vivipar- idae (Mollusca, Gastropoda) of southern Saskatchewan, western North Dakota, eastern Montana, and northern Wy- oming. Unpublished Ph.D. Thesis, University of Minnesota, Minneapolis. 928 pp. HENDERSON, J. 1935. Fossil non-marine Mollusca of North America. Geological Society of America Special Paper 3:313 PP- HINTZE, L. F. 1988. Geologic history of Utah. Brigham Young University Geology Studies, Special Publication 7:202 pp. Jones, D. L., A. Cox, P. CONEY & M. BECK. 1982. The growth of western North America. Scientific American 247:70-84. La RocquE, A. 1960. Molluscan faunas of the Flagstaff For- mation of central Utah. Geological Society of America Mem- oir 78:100 pp. MacaGiniTigz, H. D. 1953. Fossil plants of the Florissant beds, Colorado. Carnegie Institute of Washington, Publication 599: 198 pp. MANSFIELD, W. C. 1937. Mollusks of the Tampa and Su- wannee Limestones of Florida. Florida Department of Con- servation Geological Bulletin 15:334 pp. MEEK, F. B. 1876. Report on the invertebrate Cretaceous and Tertiary fossils of the upper Missouri country. U.S. Geo- logical Survey of the Territories (Hayden Survey), Mono- graph 9:629 pp. MILLER, W. B. & E. NARANJO-GARCIA. 1991. Familial rela- tionships and biogeography of the western American and Caribbean Helicoidea (Mollusca: Gastropoda: Pulmonata). American Malacological Bulletin 8(2):147-153. NorpsiEck, H. 1987. Revision des Systems der Helicoidea (Gastropoda: Stylommatophora). Archiv fir Mollusken- kunde 118:9-50. Nur, A. & Z. BEN-AVRAHAM. 1977. Lost Pacific continent. Nature 279:41-43. PamMPE, W.R. 1974. A preliminary survey of fresh-water early Tertiary invertebrates from Trans-Pecos Texas. Transac- tions of the Gulf Coast Association of Geological Societies 24:291-302. PitsBry, H.A. 1892. Preliminary outline of a new classification of the helices. Proceedings of the Academy of Natural Sci- ences of Philadelphia 44:387-404. Page 131 Pitssry, H. A. 1927. The structure and affinities of Hum- boldtiana and related helicid genera of Mexico and Texas. Proceedings of the Academy of Natural Sciences of Phila- delphia 79:165-192. Piussry, H. A. 1939. Land Mollusca of North America (north of Mexico). Academy of Natural Sciences of Philadelphia, Monograph 3 1(1):1-573. Pitssry, H. A. 1948. Inland mollusks of northern Mexico— 1. The genera Humboldtiana, Sonorella, Oreohelix, and Ash- munella. Proceedings of the Academy of Natural Sciences of Philadelphia 150:185-203. RotuH, B. 1984. Lysinoe (Gastropoda: Pulmonata) and other land snails from Eocene-Oligocene of Trans-Pecos Texas, and their paleoclimatic significance. The Veliger 27:200- 218. Rotn, B. 1987. Anew Recent species of Helminthoglypta (Gas- tropoda: Pulmonata) from the Klamath Mountains, Cali- fornia, and redescription of an Oligo-Miocene species from Oregon. Wasmann Journal of Biology 45:16-25. Rotu, B. In preparation. Phylogeny of the Helminthoglyptidae (Gastropoda: Pulmonata). SCHILEYKO, A. A. 1978. Land mollusks of the superfamily Helicoidea. Fauna of the U.S.S.R., Mollusks 3(6):1-384. SCHILEYKO, A. A. 1979. The system of the order Geophila (= Helicida) (Gastropoda Pulmonata). Pp. 44-69. In: O. A. Scarlato (ed.), Morphology, Systematics and Phylogeny of Mollusca. Proceedings of the Zoological Institute, Academy of Sciences of the U.S.S.R., 80 pp. SCHILEYKO, A. A. 1991. Taxonomic status, phylogenetic re- lations and system of the Helicoidea sensu lato (Pulmonata). Archiv fur Molluskenkunde 120:187-236. SOLEM, A. 1954. Notes on Mexican mollusks. I: Durango, Coahuila and Tamaulipas, with description of two new Humboldtiana. The Nautilus 68:3-10. SOLEM, A. 1955. New and little-known Mexican Helicidae (Mollusca, Pulmonata). The Nautilus 69:40-44. SWISHER, C. C., II] & D. R. PROTHERO. 1990. Single crystal “Ar /°Ar dating of the Eocene-Oligocene transition in North America. Science 249:760-762. THOMPSON, F. G. 1963. New land snails from El Salvador. Proceedings of the Biological Society of Washington 76:19- 32) TILLIER, S. 1989. Comparative morphology, phylogeny and classification of land snails and slugs (Gastropoda: Pulmona- ta: Stylommatophora). Malacologia 30:1-303. TOEPELMAN, W. C. 1922. The paleontology of the area. Pp. 61-73. In: F. Ward (ed.), The Geology of a Portion of the Badlands. South Dakota Geological and Natural History Survey Bulletin 11. UNDERWOOD, J. R., JR. & J. A. WILSON. 1974. Earliest known occurrence of land snail Humboldtiana: from tuff of Garren Group (Oligocene), Trans-Pecos Texas. Journal of Pale- ontology 48:596-597. WENZ, W. 1923. Fossilium catalogus, I. Animalia: Partes 17, 18 et 20, Gastropoda extramarina tertiaria; I, Vorwort, Lite- ratur, Pulmonata I. Junk: Berlin. 884 pp. WERNSTEDT, F. L. 1972. World Climatic Data. Climatic Data Press: Lemont, Pennsylvania. 523 pp. WHITE, C. A. 1877. The invertebrate fossils collected in por- tions of Nevada, Utah, Colorado, New Mexico, and Arizona by parties of the expeditions of 1871, 1872, 1873 and 1874. U.S. Geographic Survey West of the 100th Meridian (Wheeler Survey), 4(1):219 pp. WHITE, C. A. 1883. A review of the non-marine fossil Mollusca Page 132 of North America. U.S. Geological Survey Third Annual Report:403-550. WoLrFE, J. A. 1979. Temperature parameters of humid to mesic forests of eastern Asia and relation to forests of other regions of the northern hemisphere and Australasia. U.S. Geological Survey Professional Paper 1106:37 pp. The Veliger, Vol. 35, No. 2 WORLD METEOROLOGICAL ORGANIZATION (UNESCO). 1979. Climatic atlas of North and Central America. I. Maps of mean temperature and precipitation: scales 1:10,000,000 and 1:5,000,000. ZILCH, A. 1959-1960. Gastropoda, Teil 2, Euthyneura. Hand- buch der Palaozoologie, 6(2):1—-400 (1959), 401-834 (1960). Appendix Locality register of figured and referred specimens. UCM locality number Location USGS 7% min. quadrangle map 83235 260 m W, 260 m N of SW corner, sec. 28, T. 31 N, R. 70 W Dilts Ranch, WY (1949) 87063 380 m E, 1070 m S of NW corner, sec. 27, T. 31 N, R. 70 W Orin, WY (1949) 90004 60 m W, 500 m N of SE corner, sec. 28, T. 31 N, R. 70 W Cedar Hill, WY (1949) 90005 280 m W, 1300 m S of NE corner, sec. 29, T. 31 N, R. 70 W Irvine, WY (1949) The Veliger 35(2):133-136 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 New Occurrences of the Malleid Bivalve Nayadina (Exputens) from the Eocene of Jamaica, Mexico, and Washington RICHARD L. SQUIRES Department of Geological Sciences, California State University, Northridge, California 91330, USA Abstract. New collecting extends the geographic range of two of the three known species of the warm-water, Eocene malleid bivalve Nayadina (Exputens). Nayadina (E.) batequensis Squires, 1990, formerly known only from north-central Baja California Sur, Mexico, is now also known from north- western Jamaica and southern Baja California Sur, Mexico. Nayadina (E.) llajasensis (Clark, 1934), formerly recognized from southern California to central Oregon is now also known from northwestern Washington. INTRODUCTION Nayadina (Exputens) isa warm-water malleid bivalve with Old World Tethyan affinities (PALMER, 1967; GIVENS, 1989). Three species are known: N. (E.) batequensis Squires, 1990, from the lower Eocene part of the Bateque For- mation, Baja California Sur, Mexico; N. (£.) llajasensis (Clark, 1934) from middle lower to lower middle Eocene deposits in southern and central California and central western Oregon; and N. (E.) ocalensis (MacNeil, 1934) from upper Eocene deposits in Florida, Georgia, and North Carolina. These species are reviewed and compared in SQUIRES (1990). New collecting has revealed additional occurrences of N. (E.) batequensis from Jamaica and Baja California Sur, Mexico, and N. (E.) llajasensis from north- western Washington. It is the purpose of this present study to report on these new occurrences. Abbreviations used for catalog and/or locality numbers are: CSUN, California State University, Northridge; IGM, Instituto de Geologia, Universidad Nacional Autonoma Museum de Mexico; LACMIP, Los Angeles County Mu- seum of Natural History, Invertebrate Paleontology Sec- tion; UF, University of Florida, Gainesville. NEW OCCURRENCES oF NAYADINA (EXPUTENS) BATEQUENSIS The author obtained four specimens of Nayadina (E.) bate- quensis from limestones in the Chapelton Formation in northwestern Jamaica. LEwIs & DRAPER (1990) assigned this formation to the lower to middle Eocene. Three of the specimens are from locality UF XJ012 near Montego Bay, and one specimen is from locality UF XJ018 near Chris- tiana. Two of these specimens are illustrated (Figures 1- 3). Specimen UF 37089 (Figure 3) is larger than those previously known for N. (E.) batequensis. Additional col- lecting from the Bateque Formation, Baja California Sur, Mexico (SQUIRES & DEMETRION, in press), however, has yielded comparable large-sized specimens (Figure 4). The author obtained three early adult specimens of Na- yadina (E.) batequensis from sandstones in the upper part of the Tepetate Formation about 75 km north of La Paz in southern Baja California Sur, Mexico. SQUIRES & DE- METRION (1991) showed that these sandstones are coeval with the lower Eocene (““Capay Stage”) part of the Ba- teque Formation that is about 200 km to the north in north-central Baja California Sur. One of the Tepetate Formation specimens is from locality CSUN 1491, and two other specimens are from locality CSUN 1522. One of the specimens from locality CSUN 1522 is illustrated in Figures 5 and 6. Nayadina (Exputens) batequensis is the earliest species of Exputens and can now be shown to have ranged more easterly than the other species of Exputens. The presence of this species in Jamaica and Baja California Sur strongly suggests that either the subgenus emigrated from the Old World into the North America region via the circum- equatorial current that flowed from east to west during Page 134 The Veliger, Vol. 35, No. 2 Explanation of Figures 1 to 10 Figures 1-6. Nayadina (Exputens) batequensis Squires, 1990. Figures 1-3: Chapelton Formation, Jamaica. Figures 1, 2: hypotype, UF 37088, locality UF XJ012, x2.2; Figure 1, left valve; Figure 2, dorsum. Figure 3: hypotype, UF 37089, locality UF XJ018, left valve x1.1. Figure 4: hypotype, IGM 5924, locality CSUN 1470, Bateque Formation, Baja California Sur, Mexico, left valve, x1.1. Figures 5, 6: hypotype, IGM 5925, locality CSUN 1522, Tepetate Formation, Baja California Sur, Mexico, right valve, x1.8; Figure 5, dorsum; Figure 6, exterior. Figures 7-10. Nayadina (Exputens) llajasensis (Clark, 1934). Figure 7: hypotype, LACMIP 11525, locality CSUN 1516, from reworked clast of Crescent Formation in the Aldwell Formation, Pulali Point, Washington, right valve, x1.7. Figures 8-10: locality CSUN 1502, Crescent Formation near Quilcene, Washington; Figure 8, hypotype, LACMIP 11478, right valve, x1.3. Figures 9-10: hypotype, LACMIP 114839, left valve, x2.1; Figure 9, dorsum; Figure 10, interior. the early Eocene or that the subgenus originated in the Jamaica area. NEW OCCURRENCES oF NAYADINA (EXPUTENS) LLAJASENSIS The author obtained ten specimens of Nayadina (E.) lla- jasensis from Eocene rocks about 45 km west of Seattle, Washington. Six specimens are from reworked sedimen- tary clasts in the lower part of the middle Eocene Aldwell (2?) Formation at locality CSUN 1516 at Pulali Point. Some of the reworked clasts consist of very distinctive whitish-colored, calcareous, medium-grained sandstone most likely derived from the underlying lower Eocene up- per part of the Crescent Formation (SQUIRES et al., in press). One of the N. (E.) llajasensis specimens is illustrated in Figure 7. Four other specimens of Nayadina (E.) llajasensis are from the upper Crescent Formation at locality CSUN 1502, about 5 km north of Pulali Point. At CSUN locality 1502, the specimens were found in boulder-sized rocks that are not in place but are in a modern landslide block at the base of a steep cliff mapped by HAMLIN (1962) as Crescent Formation basalt. He did not mention any sed- imentary rocks interbedded with the Crescent Formation in this area, but sedimentary interbeds are present (J. L. Goedert, personal communication). Brachiopods are very abundant at locality CSUN 1502, with numerous speci- mens of Hemithiris reagani Hertlein & Grant, 1944, and common specimens of Terebratulina unguicula weaveri Hertlein & Grant, 1944. There are also a few specimens of a calcareous? sponge, a single specimen of a new anomiid bivalve, and a single large specimen of Ostrea sp. All the associated macrofauna is also present in the upper Crescent Formation at Pulali Point (SQUIRES e¢ al., in press). Ad- ditional evidence from the N. (£.) llajasensis specimens at CSUN locality 1502 are most likely from the upper Cres- cent Formation is that the specimens are in a distinctive whitish-colored, calcareous, medium-grained sandstone identical in lithology to some reworked clasts in the basal part of the Aldwell(?) Formation found by SQUIRES e¢ al. (in press) at Pulali Point at locality CSUN 1516. Two of the N. (E.) llajasensis specimens from CSUN locality 1502 are illustrated (Figures 8-10). Localities CSUN 1502 and 1516 in northwestern Wash- ington are the northernmost occurrences of any species of R. L. Squires, 1992 Page 135 Nayadina (Exputens) and show how extensive warm-water conditions were along the Pacific coast of North America during early to middle Eocene time. Previously, the north- ernmost occurrence of Nayadina (Exputens) was central western Oregon. BALDWIN (1955) reported Exputens alexi (Clark, 1934) from the lower Eocene Siletz River Volcanic Series in central western Oregon. SQUIRES (1990) showed that Exputens alexi is conspecific with N. (E.) llajasensis. ACKNOWLEDGMENTS Roger W. Portell (Florida Natural History Museum) in- formed the author about the Jamaican specimens that he and others collected. He also provided for the loan of the specimens. M. C. Perrilliat (Instituto de Geologia, Univer- sidad Nacional Autonoma Museum de Mexico) arranged for permission for paleontologic collecting in Baja Cali- fornia Sur. She also provided the type-specimen numbers. Robert A. Demetrion helped in collecting the Baja spec- imens. James L. Goedert informed the author about the Washington specimens that he and Keith L. Kaler col- lected. The manuscript was reviewed by E. J. Moore, Oregon State University, Corvallis, and by an anonymous reviewer. LOCALITIES CITED CSUN 1471. Near middle of canyon wall along W side of Arroyo San Juan de Abajo, about 80 m elevation, about 0.75 km W of dirt road from San Jose de Gracia to El Datilon, at 112°44’W and 26°29.5'N, Mexican government 1:50,000 topographic quadrangle map of Punta Santo Domingo (number G12A47), Baja Cali- fornia Sur, Mexico, 1982. Bateque Formation. Age: Middle early Eocene (‘““Capay Stage’’). Collectors: R. L. Squires and R. A. Demetrion, April 1990. CSUN 1491. In a small quarry on S side of Mexico High- way 1, at 74.5 km N of La Paz, coordinates 9.5 and 71.5 of Mexican government 1:50,000 topographic quadrangle map of El Conejo (number G12D81), Baja California Sur, Mexico, 1983. Tepetate Formation. Age: Middle early Eocene (““Capay Stage’’). Collectors: R. L. Squires and R. A. Demetrion, June 1991. CSUN 1502. From boulder-sized rocks not in place but in a modern landslide block, 2 km S of Quilcene on W shore of Quilcene Bay just S of latitude 47°47'30", NE, section 36, T27N, R2W, Quilcene quadrangle (7.5 min- ute), 1953, Jefferson County, northwestern Washing- ton. Upper Crescent Formation. Age: Early Eocene (near the boundary between the ““Capay Stage” and the “Do- mengine Stage’’). Collectors: J. L. Goedert and K. Kal- er, April 1990. (Note: this locality is probably the same as locality University of Washington 353 described by WEAVER (1943:602)). CSUN 1516. About 20 m above the base of the Aldwell(?) Formation, 1380 m N of tip of Pulali Point, in beach cliff along W shore of Dabob Bay, section 18, T26N, R1W, Seabeck quadrangle (7.5 minute), 1953, photo- revised 1968, Jefferson County, northwestern Wash- ington. Aldwell(?) Formation. Age: Middle Eocene (Narizian Stage). Collector: J. L. Goedert, June 1990. CSUN 1522. Ina small arroyo about 0.5 km N of Mexico Highway 1, at 75 km N of La Paz, coordinates 9 and 72 of Mexican government 1:50,000 topographic quad- rangle map of El Conejo (number G12D81), Baja Cal- ifornia Sur, Mexico, 1983. Tepetate Formation. Age: Middle early Eocene (“‘Capay Stage’’). Collector: R. A. Demetrion, July 1991. UF XJ012. Small exposure on W side of road, 5 km S of Johns Hall Quarry, Spring Mount, St. James Parish, Jamaica. Chapelton Formation. Age: Middle Eocene. Collectors: Portell, Bryan, Heller, and Frederick, May 1990. UF XJ018. Along stream between Pump Station and Wait- A-Bit Cave at Wait-A-Bit, Trelawny Parish, Jamaica. Stettin Member of Chapelton Formation. Age: Early Eocene. Collectors: Portell, Bryan, Heller, and Fred- erick, May 1990. LITERATURE CITED BALDWIN, E. M. 1955. Geology of the Marys Peak and Alsea quadrangles, Oregon. U.S. Geological Survey Oil and Gas Investigations Map OM 162. Cuiark, B. L. 1934. A new genus and two new species of Lamellibranchiata from the middle Eocene of California. Journal of Paleontology 8(3):270-272. Givens, C. R. 1989. First record of the Tethyan genus Volu- tilithes (Gastropoda: Volutidae) in the Paleogene of the Gulf Coastal Plain, with a discussion of Tethyan molluscan as- semblages in the Gulf Coastal Plain and Florida. Journal of Paleontology 63(6):852-856. HaMLIN, W. H. 1962. Geology and foraminifera of the Mt. Walker-Quilcene-Leland area, Jefferson County, Washing- ton. M.S. Thesis, University of Washington, Seattle. 127 PP- HERTLEIN, L. G. & U. S. Grant, IV. 1944. The Cenozoic Brachiopoda of western North America. Publications of the University of California at Los Angeles in Mathematical and Physical Sciences 3:1-236. Lewis, J. F. & G. DRAPER. 1990. Geology and tectonic evo- lution of the northern Caribbean margin. Pp. 77-140. In: G. Gengo & J. E. Case (eds.), The Geology of North Amer- ica, Vol. H. The Caribbean Region. Geological Society of America: Boulder, Colorado. MacNEIL, F. S. 1934. The pelecypod genus Vulsella in the Ocala limestone of Florida and its bearing on correlation. Journal of the Washington Academy of Sciences 24(10):428- 431. PALMER, K. V. W. 1967. A comparison of certain Eocene molluscs of the Americas with those of the western Tethys. Pp. 183-193. In: C. G. Adams & D. V. Ager (eds.), Aspects of Tethyan Biogeography. Systematics Association, Publi- cation No. 7. London. Squires, R. L. 1990. First occurrence of the Tethyan bivalve Nayadina (Exputens) in Mexico, and a review of all species of this North American subgenus. The Veliger 33(3):305- 316. Squires, R. L. & R. A. DEMETRION. 1991. Early Eocene Page 136 macrofaunal comparisons between the Tepetate and Bateque Formations, Baja California Sur, Mexico. Geological Society of America, Annual Meeting, San Diego, 23(5):194 (ab- stract). Squires, R. L. & R. A. DEMETRION. In press. Paleontology of the Eocene Bateque Formation, Baja California Sur, Mexico. Los Angeles County Natural History Museum, Contributions in Science. The Veliger, Vol. 35, No. 2 Squires, R. L., J. L. GOEDERT & K. L. KALER. In press. Stratigraphy and paleontology of Eocene rocks at Pulali Point, Jefferson County, northwestern Washington. Wash- ington Division of Mines and Geology, Report of Investi- gations. WEAVER, C. E. 1943. Paleontology of the marine Tertiary formations of Oregon and Washington. University of Wash- ington, Publications in Geology 5 (Parts 1-3):1-789. The Veliger 35(2):137-140 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 An Eastern Pacific Mercenaria and Notes on Other Chionine Genera (Bivalvia: Veneridae) by MARY ELLEN HARTE 1180 Cragmont Avenue, Berkeley, California 94708, USA Abstract. Chione (Lirophora) kelletti, an east Pacific species, is reclassified under Mercenaria. Shells of Mercenaria are large, with prominent nymphs that are often rugosely sculpted, and with subdued, mostly commarginal sculpture. The prominent rugose nymph, as well as affinities in sculpture, size, and shape, ally M. kellettu more closely to Mercenaria, while simultaneously demonstrating the transition between Mercenaria and the taxa Anomalocardia s.s., Lirophora, and Iliochione. This transition supports subsuming Lirophora and Iliochione under Anomalocardia s.s. Additional data, both anatomical and biomolecular, support the reclassification of Anomalocardia s.s. and Mercenaria as subgenera under the senior taxon, Chione. INTRODUCTION Mercenaria Schumacher, 1817 (Veneridae: Chioninae) is distinguished from other chionine genera primarily in hav- ing subdued commarginal sculpture and a prominent, often rugosely sculpted nymph, located posterior to the cardinals. The left posterior cardinal tooth (4b), characteristically a long narrow ridge, is sometimes barely noticeable before the rugose nymph. Whereas two Atlantic species (Mercenaria mercenaria Linnaeus, 1758, and M. campechiensis Gmelin, 1791) and one western Pacific species (M. stimpson: Gould, 1861) are recognized, identifying an eastern Pacific Mercenaria has been elusive. Living Mercenaria observed along the northwest coast of North America were named M. ken- nicotti: Dall, 1871, but were later recognized as human introductions of the Atlantic M. mercenaria (HANNA, 1966; Morris et al., 1980); the type of M. kennicotti, U.S. Na- tional Museum (USNM) 75017, appears to be a specimen of M. campechiensis. The existence of the west tropical American species Mercenaria apodema (Dall, 1902) is du- bious. The extremely worn condition of the single type valve, USNM 6243 (Figure 1), precludes precise identi- fication; the remains of the cardinal teeth are worn flat and devoid of details such as bifidity. OLSSON (1961) spec- ulates that it is a ballast shell from the Caribbean. It lacks a rugose nymph and, in general, its features more closely resemble some species of Protothaca Dall, 1902. The only other reported collection of M. apodema was a pair of dead valves that resemble M. campechiensis, leading KEEN (1971) to conclude that the existence of a Panamic Mercenaria could not as yet be positively proved. Here, I show that Venus kellettu Hinds, 1845, classified under Chione (Li- rophora) by KEEN (1971), is a tropical eastern Pacific Mercenaria, and discuss how this genus is related to sub- genera of Chione and Anomalocardia. MATERIALS anp METHODS Valves of Mercenaria mercenaria, M. campechiensis, M. stimpsoni, Chione (Lirophora) kellettu, and C. (L.) latilirata (Conrad, 1841) were examined from the collections at the Museum of Paleontology, University of California at Berkeley, and the California Academy of Sciences (CAS), using a dissecting scope and a 10x magnifying glass. I also examined the nymphs of species representing the other extant genera and subgenera of Chione and Anomalocardia: of Chione s.s.—C. cancellata (Linnaeus, 1767), C. califor- niensis (Broderip, 1835), C. compta (Broderip, 1835), C. intapurpurea (Conrad, 1849), C. subimbricata (Sowerby, 1835), C. tumens Verrill, 1870, and C. undatella (Sowerby, 1835); of Chione (Austrovenus)—C. stutchburi (Gray, in Wood, 1828); of Chione (Chionista)—C. fluctifraga (Sow- erby, 1853) and C. cortez: (Carpenter, 1864); of Chione (Chionopsis)—C. amathusia (Philippi, 1844), C. gnidia (Broderip & Sowerby, 1829), C. ornatissima (Broderip, 1835), and C. pulicaria (Broderip, 1835); of Chione (dh- ochione)—C. subrugosa (Wood, 1828); of Chione (Liropho- ra)—C. clenchi Pulley, 1952, C. mariae (Orbigny, 1846), and C. paphia (Linnaeus, 1767); of Anomalocardia s.s.— A. brasiliana (Gmelin, 1791) and A. cuneimeris (Conrad, 1846); of Anomalocardia (Anomalodiscus)—A. squamosa Page 138 The Veliger, Vol. 35, No. 2 M. E. Harte, 1992 (Linnaeus, 1758); of Anomalocardia (Cryptonomella)—A. producta (Kuroda & Habe, in Kuroda, 1951). OBSERVATIONS SCHUMACHER (1817) defined the shells of Mercenaria as triangular, heart-shaped and somewhat obese, with the 2a, 2b, 1, and 3b cardinals bifid, and with a prominent, rugose nymph. In addition, the sculpture of the valves consists of closely spaced, sharp, commarginal threads that sometimes become obsolete medially, forming smooth areas; fine, faint radials are often partly evident. The prominent nymph is probably the most consistent trait of this genus. The mature shells are relatively large for chionines (7-15 cm). The nymph of Mercenaria kelletti is clearly more similar to Mercenaria (Figure 3) than Chione (Lirophora): the nymph is prominent, roughly twice the length of the pos- terior cardinal (4b), with a wide, rugosely sculpted area. The nymph of C. latilirata, by contrast, is not prominent, limited to the length of the posterior cardinal, and narrow, with the rugose area barely noticeable between the cardinal and the ligament. The rugose area of the Mercenaria nymph varies from being wide and large in most M. mercenaria to thinner and smaller (some specimens of M. campechiensis). In M. stump- soni, the nymph is prominent but smooth, with the rugose area restricted to a narrow, barely noticeable strip on the posterior flank of the 4b. Within this range, the nymph of M. kellettiu is intermediate, and quite similar to some M. campechiensis. The rugose nymph of M. kelletti is not mentioned in its original description (HINDs, 1845) nor in its descriptions by subsequent workers (e.g., KEEN, 1971; OLSSON, 1961), although a rugose nymph is evident in the illustration of the hinge by OLSSON (1961:pl. 41, fig. 5). The sculpture of the type species of Mercenaria, M. mercenaria, consists of fine concentric threads, which can become obsolete medially to form smooth areas while be- coming slightly lamellar laterally. Some threads can also develop into low, thin, continuous, moderately spaced, con- centric lamellae that are evident on juveniles and some adults; fine, faint radials are sometimes evident on some part of the valve. Except for some of the lateral lamellae that are developed into pronounced flanges, the sculpture of M. kelletti fits that of M. mercenaria; fine concentric threads become obsolete medially forming smooth areas, sometimes over most of the shell. Fine, faint radials are Page 139 sometimes evident on some part of the valve. Mercenaria kellettu differs from other species of Chione (Lirophora), which have strong sculpture over most of the valves (Figure 2), although the sculpture of M. kellettw illustrates the similarities linking Mercenaria to Chione (Lirophora): well spaced intervals of the M. mercenaria valve are demarcated by sulci that are slightly deeper than those separating other concentric threads. In M. kelletti1, these sulci demarcate areas that terminate in pronounced lateral flanges, and in Chione (Lirophora), these areas are pronounced, thickened lirae. In M. kellettiz, the bounded areas sometimes appear slightly swollen, reminiscent of the thickened lirae in Lr- rophora. In other respects, the characters of Mercenaria kelletti are well within the character range of Mercenaria, and include a similar lunule, escutcheon, pallial sinus, cardi- nals, and hinge plate. The differences from other Merce- naria appear slight. The profile of M. kellettu is slightly more elongate than that of most Mercenaria; the average size (7 cm) is intermediate between Chione (Lirophora) (3- 4 cm) and other Mercenaria (10 cm). Unlike other species of Mercenaria, M. kelletti appears to be almost exclusively offshore (50 m or more), and occurs from the Gulf of California to northern Peru (KEEN, 1971). The condition of the nymph appears to be a consistent character within the chionine genera and subgenera ex- amined. Besides Mercenaria, only the type subgenus of Anomalocardia and the Chione subgenera Lirophora and Ihochione have rugose nymphs. In all four taxa, radial sculpture is usually faint, sparse, or absent. Both J/iochione and Anomalocardia have, like Lirophora, thick, well spaced concentric lirae, which become less prominent in some species or specimens. Between the attenuated posterior so distinctive of Anomalocardia and the slightly angular pos- terior of most Mercenaria, specimens of M. kelletti (Figure 2) and species of Livophora and Iliochione form a concho- logical transition in size, shape, and sculpture. [liochione has been classified as Anomalocardia by some workers (e.g., HERTLEIN & STRONG, 1948). Fossil evidence indicates that Chione evolved from Mer- cenaria (STENZEL, 1955); JONES (1979) suggests that the conchological evolution resulted from neoteny. Unlike Mercenaria mercenaria, both the west Pacific C. (Austrove- nus) stutchburu and the west Atlantic C. (Chione) cancellata (type species of the genus Chione) have well spaced com- marginal cords or lamellae overlying prominent radial ribs. Explanation of Figures 1 to 3 Figure 1. The interior (left) and exterior (right) valves of the holotype of Mercenaria apodema, USNM 6243. Figure 2. The exteriors ofMercenaria mercenaria, CAS 2947, top, Chione (Lirophora) kelletti, CAS 17736, middle, left and right, and Chione (Lirophora) latilirata, CAS 47841, bottom. Figure 3. Left hinge plates of Mercenaria mercenaria (CAS 2947), top, with 2a, 2b, and 4b cardinal teeth and nymph (n) indicated; Chione (Lirophora) kellettii (CAS 17736), middle, and Chione (Lirophora) latilirata (CAS 47841), bottom; 2 cm scale bar, right. Page 140 While C. stutchburii is much more similar to C. cancellata than to M. mercenaria in valve sculpture, C. cancellata and M. mercenaria are much more similar in anatomy (JONES, 1979). Additionally, C. (Lirophora) paphia is slightly more similar to C. cancellata than to M. mercenaria in anatomy (JONES, 1979), although one of the few differences between C. paphia and M. mercenaria, size of the palp rugae, is related probably to differences in the turbidity of their environment (JONES, 1979). Anomalocardia was not in- cluded in this comparison. Biomolecular evidence (HARTE, in press) indicates that Anomalocardia s.s., while closely related to both M. mercenaria and C. cancellata, is more closely related to M. mercenaria. CONCLUSIONS The sculpture and rugose nymph of Chione (Lirophora) kelletta: (Hinds, 1845) are diagnostic of the genus Merce- naria, not Chione (Lirophora). Other characters of this spe- cies fit well within Mercenaria, and I propose that the species be reclassified as Mercenaria kelletti (Hinds, 1845). The definition of Mercenaria should be modified to include those species with a prominent nymph and, specifically, those in which the nymph or part of the 4b cardinal is rugosely sculpted. A rugose nymph and predominantly concentric sculpture, as well as transitions in sculpture and profile, ally some Chione subgenera (Lirophora and Ili- ochione) and Anomalocardia s.s. to Mercenaria. The tran- sition from Lirophora and Ilochione to Anomalocardia is relatively slight and continuous, and this might be more accurately indicated by subsuming them both under Anom- alocardia s.s. A more accurate taxonomic classification of Anomalocardia s.s., Mercenaria, and Chione s.s. might be one in which all three taxa are subgeneric members of the senior taxon, Chione. Further studies, both anatomical and biomolecular, are needed to understand how the remaining American subgenera of Chione would be classified within this scheme. The Veliger, Vol. 35, No. 2 ACKNOWLEDGMENTS I am grateful to Dr. Carole Hickman for the use of her photographic equipment, to Dr. Akihiko Matsukuma for the donation of material to the Museum of Paleontology, University of California at Berkeley, for this study, to Ms. Elizabeth Kools for her assistance at the California Acad- emy of Science, and to Drs. Eugene Coan and Thomas Waller for their instructive comments. LITERATURE CITED Hanna, G D. 1966. Introduced mollusks of western North America. Occasional Papers of the California Academy of Sciences 48:1-108. Harte, M. E. In press. A new approach to the study of bivalve evolution. American Malacological Bulletin. HERTLEIN, L. G. & A. M. STRONG. 1948. Eastern Pacific Expeditions of the New York Zoological Society. XXXIX. Mollusks from the west coast of Mexico and Central Amer- ica. Part VI. Zoologica 33:163-197. Hinps, R. B. 1845. Mollusca. Jn: R. B. Hinds (ed.), The Zoology of the Voyage of HMS Sulphur, During the Years 1836-42. Volume II: 72 pp., 21 pls. Smith, Elder and Co.: London. Jones, C. 1979. Anatomy of Chione cancellata and some other chionines (Bivalvia: Veneridae). Malacologia 19:157-199. KEEN, A. M. 1971. Sea Shells of Tropical West America. Stanford University Press: Stanford, CA. 1064 pp., 22 pls. Morris, R. H., D. P. ABBoTT & E. C. HADERLIE. 1980. In- tertidal Invertebrates of California. Stanford University Press: Stanford, CA. 690 pp., 20 pls. Oxsson, A. A. 1961. Mollusks of the Tropical Eastern Pacific, Particularly from the Southern Half of the Panamic-Pacific Faunal Province (Panama to Peru). Paleontological Re- search Institution: Ithaca, New York. 574 pp., 86 pls. SCHUMACHER, C. F. 1817. Essai d’un nouveau systeme des habitations des vers testaces. Copenhagen. 287 pp., 22 pls. STENZEL, H. B. 1955. Ancestors of the quahog. Journal of Sedimentary Petrology 25:145. The Veliger 35(2):141-145 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 On the Validity, Authorship, and Publication Date of the Specific Name Ancistrocheirus lesueurit (Cephalopoda: Ancistrocheiridae) GIAMBATTISTA BELLO Istituto Arion, Casella Postale, 70042 Mola di Bari, Italy Abstract. Analysis of the “Histovre naturelle générale et particuliére des Céphalopodes Acétabuliféres vivants et fossiles” by Férussac & d’Orbigny (1834-1848) indicates that Ancistrocheirus lesueurii (d’Or- bigny, 1842) is the valid name to be applied to the only known species of the genus Ancistrocheirus. INTRODUCTION Traditionally, two nominal species have been attributed to the family Ancistrocheiridae, formerly subfamily An- cistrocheirinae of the family Enoploteuthidae (CLARKE, 1988): Ancistrocheirus lesueurt: (d’Orbigny, 1839) and Thelidioteuthis alessandrinu (Verany, 1851) (e.g., CLARKE, 1966). (The spelling lesweurtz and alessandrini, with a double z ending, is herein adhered to, in compliance with the International Code of Zoological Nomenclature [ICZN, 1985:Art. 33d].) Through a study on juvenile and subadult stages, NESIS (1978) demonstrated that Thelidioteuthis was a synonym of Ancistrocheirus. At present, it is generally agreed that the two names were given to two different phases of the life cycle of a single species, A. lesweurw for the adult and 7. alessandrini for an early juvenile stage. Thus, in accordance with the principle of priority (ICZN, 1985: Art. 23), A. lesweuri is the binomen currently used (e.g., OKUTANI, 1976; ROPER et al., 1984; ROPER et al., 1985; CLARKE, 1986). The 1987 English translation of Nesis’ Cephalopods of the World—the original work in Russian was published in 1982—raised the issue of the correct specific name to be used. According to NEsIS (1978:448; 1987:181) there are two nominal species, 7.e., “Ancistrocheirus alessandrinit (Verany, 1851)” and “‘Ancistrocheirus lesueuri Ferussac et d’Orbigny, 1839,” the latter of which he believes is a doubtful species. However, NEsiIs (1987:171) also states that Ancistrocheirus lesueurii is the type species, so some confusion is present. In another work, NEsIs (1984) lists only Ancistrocheirus alessandrinit. The purposes of this paper are, first, to establish the valid specific name to be applied to the ancistrocheirid squid under discussion and, second, to determine its au- thorship and publication date. SPECIFIC NAME According to NEsIs (1987), the difference between the nominal species Ancistrocheirus lesueuru and Ancistrochei- rus alessandrinu involves the distribution of the large pho- tophores on the ventral side of the mantle. In A. alessan- drinu they are “arranged in alternating groups of fours and twos, not forming unpaired longitudinal row and not extending to tail” (NEsIs, 1987:180-181); in A. lesueuru they form “(in addition to twos and fours) unpaired median longitudinal row and extend to tail” (NEsIs, 1987:181). The total number of photophores in both species is 22. The general distribution of photophores of A. alessandrinu sensu NESIS (1987) is in agreement with that given by OKUTANI (1976) and ROPER et al. (1984, 1985) for A. lesueuru, whereas the photophore arrangement of A. /e- sueuriu sensu NESIS (1987:181) is “44+2+3+3+4+2+2 +3+1+1+1.” Such a photophore distribution is clearly taken from FERUSSAC & D’ORBIGNY (1834-1848: atlas, plate 14 ONYCHOTEUTHE [Onychoteuthis], fig. 4) (see Fig- ure 3, herein). Indeed there is a discrepancy between the number of photophores (22) depicted in this figure and the description given by FERUSsSAC & D’ORBIGNY (1834- 1848:339-340): “le corps est lisse en dessus, marqué en dessous de tubercules saillants [= photophores] au nombre de 21, trés réguliérement disposés.” Such a discrepancy may simply represent a typographical error. Otherwise it can be attributed to the fact that d’Orbigny did not attach a primary taxonomic importance to that character, or did not make a close enough check of the number and position Page 142 of photophores on the specimen or in the figure that depicts the specimen itself. In regard to this, it is noteworthy that the specimen described in FERUSSAC & D’ORBIGNY (1834- 1848:atlas, plate 14 of the genus Onychoteuthis; text, 339- 340) was taken from a dolphin stomach (LU et al., in press), so that the predator digestion processes might have wholly or partly destroyed the integument and its annexes, in- cluding photophores. The above-mentioned specimen, which represents the holotype of the d’Orbigny species and is kept at the Muséum National d’Histoire Naturelle of Paris, catalogue No. MNHN 2-14-614, was also examined by the author. Unfortunately, the specimen is in bad con- dition. The skin is lacerated and largely missing on the ventral surface of the mantle; the “tail” completely lacks the skin. Only a few photophores are present and, possibly, displaced to some extent from their natural position. Ap- parently, none of them lies on the sagittal line, 7.e., in unpaired position. It is not known whether the specimen was already in bad condition when described by d’Orbigny or deteriorated afterwards. Lastly, NEsIS stated (1987:181) that the distribution and habitat of Ancistrocheirus lesueuru Ferussac & d’Orbigny, 1839, are unknown. Indeed d’Orbigny (zn FERUSSAC & D’ORBIGNY, 1834-1848:340) reported the locus typicus: “Hab. Le grand Océan,” which corresponds to the Indo- Pacific, as shown by internal evidence. GRAY (1849) re- ported the Indian Ocean as the locus typicus. To sum up, NEsIs (1987) doubted the identity of An- cistrocheirus lesueurw largely, if not entirely, on the basis of the photophore arrangement depicted by FERUSSAC & D’ORBIGNY (1834-1848:atlas, plate 14 of the genus Ony- choteuthis). In fact, the same work (FERUSSAC & D’ORBIGNY, 1834-1848:339-340) presented a fairly long and sufh- ciently discriminating description of A. lesweuri. There- fore, the nominal taxa “‘Ancistrocheirus alessandriniu (Vér- any, 1851)” and “Ancistrocheirus lesueuri Férussac & d’Orbigny, 1839” accepted by NEsIs (1978:448; 1987:181) are herein considered to be synonymous. On the basis of priority (ICZN, 1985:Art. 23), A. lesweura is the valid name. AUTHORSHIP anp PUBLICATION DATE Determining the correct authorship and publication date of Ancistrocheirus lesueurii is a somewhat more complicated The Veliger, Vol. 35, No. 2 problem. Usually the species is referred to d’Orbigny, 1839 (e.g., CLARKE, 1966, and subsequent authors). The specific name lesweuri: appeared for the first time and was repeatedly referred to in Histoire naturelle générale et particuliére des Céphalopodes Acétabuliferes vivants et fos- siles by FERUSSAC & D’ORBIGNY (1834-1848). WINCK- WORTH (1942) elucidated the publication dates of the 21 parts, issued separately, that make up the work. Additional information on the publication dates of the herein discussed taxa is reported by Lu ef al. (in press) and TILLIER & BOUCHER-RODONI (in press). The chronological order of the appearance of the name lesueurt in FERUSSAC & D’ORBIGNY (1834-1848) is re- ported below, with related comments. (1.) “Onychoteuthis Lesueuri, d’Orbigny,” 1835, plate 4 of the genus Onychoteuthis. This drawing is by d’Orbigny (“A. d’Orbigny pinx.’’), who stated that the plate was printed in 1825 (FERUSSAC & D’ORBIGNY, 1834-1848:331). According to WINCK- WORTH (1942), the heading “Cryptodibranches” proved that the plate was part of a group that was printed in the period 1825-1833 and issued in the years 1834-1835. As stated by WINCKWORTH (1942:35) “these plates may then be definitely dated as 1835.” Lu e¢ al. (in press) and TILLIER & BOUCHER-RODONI (in press) also date the plate as 1835; it was issued in livratsons 4-6, March 1835 (TILLIER & BOUCHER-RODONI, in press). The cephalopod portrayed in the plate under discussion (see Figure 1 herein) is a true Onychoteuthis (cf. banksit). Later, d’Orbigny himself (zn FERUSSAC & D’ORBIGNY, 1834-1848:330-332) listed his O. lesweuri of plate 4 among the synonyms of Onychoteuthis banksi (Leach, 1817): “en 1826, j'ai impose a tort le nom de Onychoteuthis Lesueuru a un exemplaire [of O. banksi] rapporté par M. Lesueur.” LU et al. (in press) remark that “the holotype of this species is also a syntype of Onykia angulatus Lesueur, 1821, which was sent by Lesueur to Ferussac and appeared in Ferussac & d’Orbigny (pl. 4) under the name Onychoteu- this lesueurt.” This type specimen is still extant and is kept at the Museum National d’Histoire Naturelle of Paris: holotype, MNHN 3-1-628, female, ML = 72 mm; MNHN 3-1-643 (gladius, dry) (Lu et al., in press). (2.) “Onychoteuthis Lesueuru, Férussac,” 1835, plate 11 of the genus Onychoteuthis, figs. 1-5. Explanation of Figures 1 to 3 Figure 1. “O. Lesueurii, d’Orbigny”; pl. 4 of the genus Onychoteuthis, figs. 1-7; from FERUSSAC & D’ORBIGNY (1834- 1848). The plate represents a specimen of Onychoteuthis cf. banksw. Key: 1-2, dorsal and ventral views; 3, oral view of arms and tentacles; 4, tentacle club; 5, hook; 6, sucker; 7, sucker ring; not numbered, gladius. Figure 2. “O. Lesueurti, Férussac”; pl. 11 of the genus Onychoteuthis, figs. 1-5; from FERUSSAC & D’ORBIGNY (1834- 1848). The figures represent Ancistrocheirus lesueurii. Key (faint in original): 1, dorsal view; 2-3, gladius; 4-5, arm hook. Figure 3. “Enoploteuthis Lesueuru, dOrbigny”; pl. 14 of the genus Onychoteuthis, figs. 4-10; from FERUssac & D’ORBIGNY (1834-1848). The figures represent Ancistrocheirus lesueuri. Key: 4, ventral view of mantle; 5-6, upper mandible; 7-8, lower mandible; 9-10, “‘tuwbercule”’ [= photophore]. Page 143 G. Bello, 1992 Page 144 According to d’Orbigny (7m FERUSSAC & D’ORBIGNY, 1834-1848:339) this plate was printed in 1835. In addi- tion, it bears Férussac’s name, which supports the as- sumption that it was printed before Feérussac’s death in 1836. WINCKWORTH (1942) included this plate in a group published in the period 1839-1841. He also reported that the plate might have been issued as early as 1835, in the 11th livraison (WINCKWORTH, 1942). In fact, TILLIER & BOUCHER-RODONI (in press) show that plate 11 was issued with livraisons 10-11, in June 1835. There is no doubt that the species portrayed is Ancis- trocheirus lesueuru (Figure 2 herein). The type specimen of Onychoteuthis lesueuri: Ferussac is no longer extant. At the Muséum National d’Histoire Naturelle of Paris there is a gladius that might belong to the type specimen (Boucher-Rodoni, personal communi- cation). (3.) “Enoploteuthis Lesueuru, d’Orbigny,” 1842, plate 14 of the genus Onychoteuthis, figs. 4-10. Despite the fact that d’Orbigny (zn FERuUssACc & D’ORBIGNY, 1834-1848:339) dated this plate as 1835, cer- tain details (reported below) show that it was printed at a later date than plate 11 of the same genus. It is, in fact, part of a batch of plates—some of which may be fairly safely dated as 1838-1839—prepared by a single lithogra- pher and a single printing house. The new genus Enoplo- teuthis, which included the species /esweuri, was established by d’Orbigny (in FERUSSAC & D’ORBIGNY, 1834-1848: 336-337) after he took up the work again in 1837, fol- lowing Ferussac’s death, and completely reviewed the tax- onomy. According to WINCKWORTH (1942:36), the plate publication date is “[1839-]1848” and since “‘in systematic work the dates in brackets should be omitted,” the official publication date should be 1848. However the Enoploteu- this lesueur plate is mentioned in another work by D’ORBIGNY (1845), evidence that it was issued well before 1848. In fact, the plate was issued sometime during the years 1839-1842 (fide TILLIER & BOUCHER-RODONI, in press). There is no clue to ascertain the exact year of publication. Thus, it is herein concluded that the publi- cation date to be used for taxonomic purposes is 1842, in accordance with Art. 21 of the ICZN (1985). The customarily used date of 1839 is attributable to Gray (1849), who assigned that publication date to plate 14, without giving any evidence for it. The dates reported by Gray (1849) are not reliable because of several internal discrepancies. (For instance, GRAY [1849] assigned two different publication dates to plate 11 of the genus Ony- choteuthis of FERUSSAC & D’ORBIGNY’s atlas (1834-1848); e.g., ‘Onychoteuthis leptura Férussac, Céph. Acét. Onycho- teuthis, t. 11. f. 6-14. 1839” (Gray, 1849:47) and “On- ychoteuthis LeSueurii Féruss. & d’Orb. Céph. Acét. Onych. t. 11. f. 1-5, animal, 1835” (Gray, 1849:49). It appears that GRAY (1849) uncritically reported the dates given by d’Orbigny [7m FERUssAC & D’ORBIGNY, 1834-1848)]). Plate 14 represents the species known today by the name Ancistrocheirus lesueuri (Figure 3). The Veliger, Vol. 35, No. 2 Afterwards, d’Orbigny (in FERUSSAC & D’ORBIGNY, 1834-1848:339-340) also gave a fairly complete descrip- tion of “Enoploteuthis Lesueuri.” There is no doubt that the publication date of these text pages is 1848 (WINCKWORTH, 1942; TILLIER & BOUCHER-RODONI, in press). The holotype figured in plate 14 is kept at the Muséum National d’ Histoire Naturelle of Paris: MNHN 2-14-614, male, ML = 130 mm; MNHN 2-14-615 (gladius, dry); MNHN 2-14-616 (beaks). It was collected by Dussumier from the stomach of a dolphin, in 1835; locality unknown (Lu e¢ al., in press). CONCLUSIONS From what has been reported above, it can be deduced that: (a.) Onychoteuthis Lesueuri: d’Orbigny, 1835 in FERUSSAC & D’ORBIGNY, 1834-1848 (divv. 4-6) is an avail- able name (ICZN, 1985:Art. 10a) and is a junior synonym of Onychoteuthis Banksu (Leach, 1817). (b.) Onychoteuthis Lesueuru Ferussac, 1835 in FERUSSAC & D’ORBIGNY, 1834-1848 (divv. 10-11) is a primary junior homonym (ICZN, 1985:Art. 57b) of Onychoteuthis Le- sueuru d’Orbigny, 1835, and as such is permanently invalid (IGZN, 1985:Art. 52b). (c.) Enoploteuthis Lesueuru d’Orbigny, 1842 7n FERUSSAC & D’ORBIGNY, 1834-1848, is the next available published name and must replace the invalid Onychoteuthis Lesueuru Ferussac, 1835 (ICZN, 1985:Art. 60a). The name was proposed in association with illustrations of the species being named, which constitutes an “indication” (ICZN, 1985:Art. 12a[7]). The original spelling of the specific name, Lesweurzz, should begin with a lower case letter (ICZN, 1985:Art. 28) and the double “7” ending should be maintained. The spelling with a single ‘‘z,” 7.e., lesweurt, is incorrect (ICZN, 1985:Arts. 32, 33). In conclusion, and taking into account that the species Enoploteuthis Lesueuru was later assigned to the genus Ancistrocheirus Gray, 1849, the name to be used is Ancis- trocheirus lesueuru (d’Orbigny, 1842 im FERUssAC & D’ORBIGNY, 1834-1848). ACKNOWLEDGMENTS I am grateful to Dr. C. F. E. Roper, Dr. M. J. Sweeney, and Prof. T. Okutani, who kindly reviewed an early ver- sion of the paper and offered criticism, and to an anony- mous reviewer for pertinent comments and suggestions. Thanks are due to Dr. R. Boucher-Rodoni for information and discussion on Férussac and d’Orbigny type specimens and works; she also made the holotype of Ancistrochezrus lesueurw available. Lastly, I wish to thank Mme. C. Fa- varger (Muséum d’Histoire Naturelle of Geneva) who courteously provided the photographs of Feérussac and d’Orbigny’s plates. G. Bello, 1992 LITERATURE CITED CLARKE, M.R. 1966. A review of the systematics and ecology of oceanic squids. Advances in Marine Biology 4:91-300. CLARKE, M. R. (ed.). 1986. A Handbook for the Identification of Cephalopod Beaks. Clarendon Press: Oxford. xiii + 273 PP: CLARKE, M. R. 1988. Evolution of Recent Cephalopods—a brief review. Pp. 331-340. In: M. R. Clarke & E. R. True- man (eds.), The Mollusca. Vol. 12. Academic Press: London and New York. FERussac, A. DE & A. D’ORBIGNY. 1834-1848. Histoire na- turelle générale et particuliére des Céphalopodes Acétabu- liféres vivants et fossiles. J.-B. Bailliére: Paris. 366 pp. (Vol. 1, text); 144 pls. (Vol. 2, atlas). Gray, J. E. 1849. Catalogue of the Mollusca in the Collection of the British Museum. Part I. Cephalopoda Antepedia. British Museum (Natural History): London. 164 pp. INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE. 1985. International Code of Zoological Nomenclature. 3rd ed. International Trust for Zoological Nomenclature: Lon- don. xx + 338 pp. Lu, C. C., R. BoOUCHER-RODONI & A. TILLIER. In press. Cat- alogue of recent Cephalopoda type-specimen in the Muséum National d'Histoire Naturelle, Paris. Bulletin du Muséum national d’Histore naturelle: Paris. NEsISs, K. N. 1978. Podsemejstvo Ancistrocheirinae (Cephalop- oda, Enoploteuthidae). [The subfamily Ancistrocheirinae (Cephalopoda, Enoploteuthidae).] Zoologicheskij Zhurnal 57:446-450. Page 145 Nests, K. N. 1984. Cephalopoda. 182 pp. Jn: O. A. Scarlato (ed.), Polevoj opredelitel’ planctona. [Field card manual of plankton.] Part 3. Zoologicheskij Institut Academii Nauk SSSR: Leningrad. Nesis, K. N. 1987. Cephalopods of the World. Translated from the Russian by B. S. Leviton; edited by L. A. Burgess. T. F. H. Publications Inc.: Neptune City, NJ. 351 pp. OxuTanI, T. 1976. Rare and interesting squids from Japan— V. A gravid female of Ancistrocheirus lesueuri (d’Orbigny, 1839) collected in the Kuroshio area (Oegopsida: Enoplo- teuthidae). Venus 35:73-81. OrRBIGNY, A. D’. 1845. Mollusques vivants et fossiles, ou de- scription de toutes les espéces de coquilles et de mollusques classées suivant leur distribution géologique et geographique. Paris. 605 pp., 35 pls. Roper, C. F. E., M. J. SWEENEY & M. R. CLARKE. 1985. Cephalopods. Pp. 117-205. In: W. Fischer & J. C. Hureau (eds.), FAO Species Identification Sheets for Fishery Pur- poses. Southern Ocean (Fishing areas 48, 58 and 88) (CCAMLR Convention Area). FAO: Rome. Roper, C. F. E., M. J. SWEENEY & C. E. NAUEN. 1984. FAO species catalogue. Vol. 3. Cephalopods of the world. FAO Fisheries Synopsis 125(3):1-277. TILLIER, A. & R. BOUCHER-RODONI. In press. Date of pub- lication of plates and text of Ferussac and d’Orbigny’s His- toire naturelle des Cephalopodes. Nautilus. WINCKWORTH, R. 1942. Notes on the publication of Férussac and Orbigny’s Histoire des Céphalopodes. Proceedings of the Malacological Society of London 25:34-36. The Veliger 35(2):146-156 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 The Anatomy of Arion flagellus Collinge, 1893, Present on the Iberian Peninsula (Gastropoda: Arionidae: Terrestria Nuda) JOSE CASTILLEJO Department of Biology Animal, Faculty of Biology, University of Santiago, 15706 Santiago de Compostela, La Coruna, Spain Abstract. The presence of Arion flagellus in Galicia (NW Spain) is reported. The internal and external morphology (including spermatophore) and copulation of Galician specimens are described, and are discussed with reference to English specimens and to the Portuguese Arionidae described in the last century by Morelet, Mabille, Pollonera, and Simroth. INTRODUCTION Specific identification of the large Arionidae (the large Arionidae are taken to be those that measure at least 80 mm in vivo) of the Iberian Peninsula is difficult if the considered individuals are few in number or incompletely developed. Juveniles may have bands on their mantles and backs that may no longer be present in adults, and body coloring is generally lighter in juveniles than in adults. Many juveniles have a whitish or light gray sole, which in adults, depending on the species, becomes orange, yel- lowish, black, or light in the center and dark laterally. The genitalia of juveniles are easily distinguished from those of subadults or adults, but those of subadults, adults, and seniles are harder to tell apart, all three sets being com- pletely developed and differing only in the relative pro- portions of their parts. Identification based on the genital morphology of a small number of specimens is further complicated by the fact that in specimens that have recently copulated (as demonstrated by the presence of the whole spermatophore in the genitalia) the epiphallus is larger in diameter and shorter than in specimens that have a frag- mented spermatophore or have not mated. One structure that can, in principle, give information about the identity of species is the spermatophore. DAVIES (1987) noted that a distinctive form of spermatophore must indicate the reproductive isolation of a good species, and she went on to assert that if each variable species of Arzon is characterized by the possession of a distinctive, and much less variable, spermatophore, then the organs concerned in the production and exchange of spermatophores must be of considerable taxonomic importance. Clearly the length of the spermatophore must be of some importance, since long ones will take longer to be exchanged than short ones. The sculpture of the spermatophore, which is produced by the lumen of the epiphallus, may also be significant for transfer. The ligula and copulatory behavior ought to be important too, although DAviEs (1987) pointed out that in stabilizing the pair, and perhaps in stimulating the mating process, the position and movements of the ligula may be more important than its changeable shape and size; upon evagination, in some species the ligula is firmly pressed against that of the mate, in others it is rested on the mate’s flanks, and in still others it is used to embrace the mate’s tail. Thus, the identification of species within the Arionidae is not easy, since the sizes and proportions of their organs can vary with both intrinsic factors (developmental stage, cycle phase, etc.) and extrinsic factors (degree of relaxation upon death, preservation, manipulation, etc.). SIMROTH (1889), recognizing the problem, wrote that the dividing lines between the species of the genus Avion are less clear than those between the Limacidae, which is why new species are so easily created. He also felt that what would best enable us to judge what ought to be included under the same name is knowledge of the fauna as a whole, starting from postembryonic development and biology; for Simroth, this investigation would have to take place on the west coast of Europe, the center of the creation of the Arionidae. The situation outlined above, together with method- ological advances in the study and description of new taxa, . Castillejo, 1992 " cat ef a ae “Ty oe 7 2 etka iS be iar ‘ : i Sided a os yo 4, i a Lanes twee Newest ty Pyaepr een 7 Yond : a tes: ! paNeate : of nee ne ney re Page 147 ve Zi veg ie ae: aS Explanation of Figures 1 to 3 Figures 1-3. Arion flagellus, Sierra del Gistral (Lugo). Dorsal, lateral, and ventral views. Scale 1 mm. make it desirable to re-examine the types, and preferably also the topotypes, of all the species described in the last century on the basis of few and often incompletely devel- oped specimens. This re-examination should be supported by the study of whole populations in various biotopes and in all seasons in order to determine the spatial and temporal variability of the species. Only in this way can the lines dividing species be firmly established. Arion flagellus was described by COLLINGE (1893) from Ireland. The large Portuguese Avion species were studied in the last century by MORELET (1845), MABILLE (1868), POLLONERA (1887, 1889, 1890), and SIMROTH (1886, 1891). Morelet described, solely on the basis of external morphology, a series of new species (A. sulcatus, A. tumidus, and A. fuligineus) and re-described several existing species. Mabille included Morelet’s new species in his genus Bau- donia, and created A. lusitanicus with part of A. rufus sensu Morelet, and A. pascalianus with A. fuscatus sensu Morelet. All of Morelet’s and Mabille’s species were accepted by Pollonera, who also described two new species (A. dasilvae and A. nobrez). SIMROTH (1886) created A. hispanicus, but later recognized only A. ater and A. lusitanicus as good species, treating A. hispanicus as a synonym for the latter. Nosre (1941) considered all large Portuguese Avion to be A. ater. SEIXAS (1976), however, reported finding A. ater, A. lusitanicus and A. subfuscus, and CASTILLEJO & RODRIGUEZ (submitted) recently recognized A. ater, A. no- brei, A. lusitanicus, A. fuligineus, and A. intermedius in a revision of the genus Avon in Portugal on the basis of the anatomy of topotypes collected by systematic sampling. Page 148 The Veliger, Vol. 35, No. 2 Explanation of Figures 4 to 6 Figures 4-6. Arion flagellus, Sierra del Gistral (Lugo). Figure 4. Organs in situ. Figure 5. Digestive tract. Figure 6. Pallial complex. Scale 1 mm. Specimens of Arion flagellus (deposited in the collection of the Departamento de Biologia Animal, Facultad de Biologia, Universidad de Santiago, Spain) were collected in Galicia between 1979 and 1984 by standard methods for capture, transport, killing, fixation, and preservation. To record external morphology, the most representative specimens were photographed at the time of capture. Ex- ternal morphology and genitalia were drawn to scale using a camara lucida and a binocular magnifier. Copulation was drawn from slides taken im situ with no scale. The following specimens sent by Dr. S. M. Davies were used for comparison: Arion flagellus from Bramley Bank, Croydon, England; A. lusitanicus collected in the garden of 63 Beechwood Road, South Croydon, Surrey, England; and A. subfuscus from Coulsdon Woods, Surrey, England. We also studied specimens of A. subfuscus collected to the north of Antwerp, Belgium, and sent by Dr. T. Backeljau. In what follows, the morphology of the specimens we found in the northwest of Spain is described and compared with the English topotypes. J. Castillejo, 1992 Page 149 Fo Spo Explanation of Figures 7 to 9 Figures 7-9. Arion flagellus, Sierra del Gistral (Lugo). Various views of genitalia. Key: A, atrium; Ag, albumen gland; Ep, epiphallus; Fo, free oviduct; Hd, hermaphrodite duct; Ot, ovotestis; R, retractor muscle; Sp, spermatheca; Spo, spermoviduct; Vd, vas deferens. Scale 1 mm. ; : tiago de Compostela (La Coruna), U.T.M. = 29TNH34, Asrine, jtegellns Collintge, WSes 14 October “CRY. (23 specimens). Coto del Xirimbao (Ve- Material examined (Map 1): Sobrado de los Monjes (La dra, La Coruna), U.T.M. = 29TNH43, 21 October 1984 Coruna), U.T.M. (Universal Transverse Mercator pro- (6 specimens). Ferreira de Valadouro (Lugo), U.T.M. = yection) = 29TNH76, 8 October 1979 (4 specimens). San- 29TPJ22, 17 September 1984 (20 specimens). Cuadramon Page 150 Fd 10 The Veliger, Vol. 35, No. 2 rR eeerer a pY Explanation of Figures 10 and 11 Figures 10 and 11. Arion flagellus, Sierra del Gistral (Lugo). Figure 10. Ligula. Figure 11. Spermatophore with detail of toothlets. Key: FD, fold; Li, ligula. Scale 1 mm. (Sierra del Gistral, Lugo), U.T.M. = 29TPJ21, 18 Oc- tober 1984 (13 specimens). The species may be very abundant in Galicia, since most reports of Arion lusitanicus may in fact refer to A. flagellus. Specimens were found on acid soil over granite under vegetation dominated by pines, eucalyptus and, to a lesser extent, chestnut and birch. Description: This is a large slug and the fully extended length in vivo can exceed 90 mm. In 70% alcohol, specimens shrink and measure 50 to 70 mm (Figures 1-3). The live body color is dark gray, with a greenish yellow or greenish chestnut flush (the green coloration of the Galician Arion flagellus is striking enough for country folk in some places to refer to this species as the “green slug that lives in the meadows’’); its flanks are lighter, greenish yellow predom- inating over the gray. The back and mantle of some spec- imens have two darker bands, the one on the right sur- rounding the pneumostome; like the flanks, the dorsal area between the bands is lighter. Juveniles are lighter colored than adults, with or without bands. In alcohol, juveniles and adults turn darker and the dorsal bands tend to dis- appear. The dermal tubercules are prominent, as in the Portuguese A. lusitanicus, forming longitudinal keels when the animal contracts. The foot fringe is greenish yellow or greenish orange and the lineoles (vertical lines) are black. J. Castillejo, 1992 Page 151 Explanation of Figures 12 and 13 Figures 12 and 13. Arion flagellus, Santiago de Compostela (La Coruna). Figure 12. Genitalia. Figure 13. Ligula (stimulator organ). Key: Fd, fold; Li, ligula. Scale 1 mm. The tentacles and the back of the head are black. Both adults and juveniles show a uniformly pale greenish yellow or greenish orange sole. The body mucus is whitish or colorless, turning a dirty white upon contact with alcohol; the sole mucus is orange. The limacella (internal shell) is formed of more or less aggregated calcareous grains. Organs in situ (Figure 4). The digestive tract (Figure 5) and pallial complex (Figure 6) have the characteristic to- pography of the genus. The arteries feeding the digestive gland and tract are white in specimens from Ferreira de Valadouro and Cuadramon preserved in 70% alcohol. Genitalia (Figures 7, 8, 9, 12). The ovotestis is volu- minous and is made up of brown acini. The hermaphrodite duct is long and straight, and the albumin gland almond- shaped, somewhat curved in the middle. The spermoviduct is long and has no distinguishing color. The epiphallus measures 20-30 mm in adults containing a whole sper- matophore, while in subadults and seniles it is only 14- 25 mm in length; on the other hand, the vas deferens in these latter classes measures 10-18 mm, while in adults containing a whole spermatophore its length is 10-15 mm. The juvenile epiphallus is up to three times longer than the vas deferens, both structures together measuring 13 mm or less. The junction between the vas deferens and the epiphallus is marked by a constriction, and the epi- phallus has a strong, annular thickening at its entry into the proximal atrium. The distal free oviduct is as long as the vas deferens (half as long as the epiphallus), with a pronounced lateral dilation housing the ligula (stimulator organ); the proximal part of the distal free oviduct can have a lateral elbow. The proximal free oviduct is one- half the length of the distal free oviduct. The ligula (Fig- ures 10, 13) possesses a circular or oblong shape, and is generally located in the proximal third of the distal free oviduct; in some specimens (Figure 13) there are one or two folds that continue to the proximal atrium, where they may give rise to a small dilation at the proximal end of the spermatheca duct. The orifice of the proximal free oviduct may open inside or outside the ligula. The sper- matheca (bursa copulatrix) is spherical or pear-shaped, its duct being shorter than the epiphallus, with, in some spec- imens, a papilla located at the commencement of the duct and pointing towards the proximal atrium (Figure 10). The proximal atrium is small; and the distal atrium is Page 152 The Veliger, Vol. 35, No. 2 16 Explanation of Figures 14 to 16 Figures 14-16. Arion flagellus, Ferreira de Valadouro (Lugo). Copulation. Not to scale. more or less spherical, covered externally by tissue of glan- dular appearance. The oviduct retractor muscle appears strongly united with the retractor of the bursa copulatrix, from which fascicles sometimes extend to the annular thickening of the epiphallus; in the inferior distal free oviduct, parietal muscles are sometimes united. No part of the genitalia has black pigmentation except the distal epiphallus near the annular ring (this pigmentation is not a result of preservation in alcohol, being found in live specimens). Spermatophore (Figure 11). The spermatophore is cy- lindrical, and measures 20 mm long, tapering to a rounded knob at one end and to a point at the other. Upon extraction it resumes the U-shape it has inside the genitalia after transfer. A longitudinal serrate crest with tall, narrow, closely packed teeth runs almost the whole length, dwin- dling at its ends to merge with the body of the spermato- phore. Copula (Figures 14-24). Copulating Arion flagellus were photographed on the night of 17 September 1984 at Ferrei- ra de Valadouro (Lugo); relative humidity was close to 100%. The entire event was not witnessed, but certainly lasted at least 45 min. In the precopulation phase, the two individuals move J. Castillejo, 1992 Page 153 20 Explanation of Figures 17 to 20 Figures 17-20. Arion flagellus, Ferreira de Valadouro (Lugo). Mutual licking phase of copulation. Not to scale. Page 154 The Veliger, Vol. 35, No. 2 24 Explanation of Figures 21 to 24 Figures 21-24. Arion flagellus, Ferreira de Valadouro (Lugo). Copulation, final phase. Not to scale. J. Castillejo, 1992 one behind the other, the rear one licking the other’s caudal area. They then curl into interlocking Cs with their genital orifices opposite each other, and shortly afterwards the posterior part of genitals (atrium and free oviduct) are evaginated or everted (Figure 14) together with the distal parts of the free oviduct and the ligulas, which grip the mate’s tail (Figures 14, 15, 19). Copulation is static (in particular they do not move in circles), except that they often lift their heads, with the tentacles retracted, and lick or scrape the mate with the protracted mandible and radula (Figures 15, 16, 18). At the end of copulation (Figure 24), they begin to extend their tentacles and move away in opposite directions. At this point the amber-colored tips of the spermatophores are visible (Figure 24) and the ligulas are still evaginated. The two individuals separate and in- vaginate their genitalia very rapidly; once separated they curl up and lick the spot where the other had placed its ligula. DISCUSSION The dissected Galician specimens of Arion flagellus resem- ble the English specimens closely, the proportions of the various parts of the genitalia and the size and appearance of the spermatophore being identical. The main difference concerns copulation. DAVIES (1987) states that most of the evaginated anterior part of the genitalia of copulating En- glish specimens lies underneath the animals, and cannot be seen until they start to separate; the evaginated anterior part of the genitalia of Galician A. flagellus lies between and upon the backs of the pair, with the ligulas visibly hugging the mate’s body, a position resembling that of copulatory English A. /usztanicus in DAVIES’ (1987) Figure 3B. The Galician Arion flagellus is very like the Portuguese A. fuligineus. The two species are almost the same length, both may or may not have dorsal bands, and the sole is always white or yellowish. However, A. fuligineus never has the greenish flush that is characteristic of A. flagellus. The genitalia are also very similar (in particular, the rel- ative proportions of the epiphallus and the vas deferens are the same), except that the ligula of A. flagellus is circular whereas that of A. fuligineus is shaped like an inverted V. The spermatophores are of equai length (20 mm), but that of A. fuligineus widens towards the middle, whereas that of A. flagellus has a more uniform cross section; further- more, the spermatophore crest of A. fuligineus is not toothed or has less pronounced teeth than that of A. flagellus, and the spermatophore itself does not adopt the U-shape char- acteristic of A. flagellus. However, the two species differ mainly in their copulation. A. fuligineus places its ligula on the flanks of the mate, whereas the Galician A. flagellus places its ligula upon its mate’s back, gripping it; and the rotation of the pair that occurs in the copulation of the Portuguese A. fuligineus is not performed by the A. flagellus of Galicia (though according to Davigs [1987], English Page 155 specimens of both A. subfuscus and A. flagellus may slowly rotate when copulating). Externally, the Galician Arion flagellus differs in color from the Portuguese A. lusitanicus, which when adult has a dark yellowish chestnut body with two dorsal bands and a sole that is dark gray or black peripherally and lighter centrally; the upper parts of juveniles may be greenish gray or black. Internally, the epiphallus of A. lusitanicus is the same length as the vas deferens, and the spermato- phore is longer than in the Galician A. flagellus (40 mm vs. 20 mm). The two species also differ in their copulation behavior and the shape of the ligula. The Portuguese Arion nobrei is much larger than the Galician A. flagellus, and is olive green or bronze colored. The sole is always black. The genitalia are much larger than in A. flagellus: the epiphallus measures 30-35 mm, and the spermatophore may exceed 65 mm. The Galician Arion flagellus differs from the English and Belgian A. subfuscus in size. Internally, its epiphallus and vas deferens are twice as long as those of A. subfuscus. The Galician Arion flagellus differs in both internal and external morphology from the English A. lusttanicus, which exceeds 100 mm in length and is black or brown, with a sole of the same color. The epiphallus of the English A. lusitanicus is the same length as the vas deferens (both 18- 20 mm), the ligula is elliptical, and the spermatophore is longer than that of A. flagellus and has a different shape. ACKNOWLEDGMENTS We thank Miss S. M. Davies (U.K.) for supplying Arion lusitanicus, A. subfuscus, and A. flagellus for comparison, Dr. T. Backeljau (Belgium) for sending us specimens of A. subfuscus, and Drs. T. Rodriguez and A. Outeiro (Spain) for their help with the drawings of A. flagellus. LITERATURE CITED CASTILLEJO, J. & T. RODRIGUEZ. Submitted for publication. Portuguese Slugs. III. Revision of the Genus Avion Ferussac, 1819. (Gastropoda, Pulmonata: Arionidae). COLLINGE, W. E. 1893. Description of anatomy &c. of a new species and variety of Arion. Annals Magazine of Natural History 12:252-254. Davies, S. M. 1987. Arion flagellus Collinge and Arion lusitan- icus Mabille in the British Isles: a morphological, biological and taxonomic investigation. Journal of Conchology 32:339- 354. MaBILLE, M. J. 1868. Des Limaciens européens. I. Travaux inedites. Revue et Magazin de Zoologie. Pp. 129-145. MORELET, A. 1845. Description des Mollusques terrestres et fluviatiles du Portugal. Paris. 113 pp., 14 pls. Nosre, A. 1941. Fauna malacologica de Portugal, II. Moluscos terrestres e fluviais. Coimbra. 277 pp., 30 pls. POLLONERA, C. 1887. Specie nuove o mal conosciute di Arion europei. Real Accademia della Scienze di Torino 22:1-27. POLLONERA, C. 1889. Nuove contribuzioni allo studio degli Arion europei, 1. Specie portoghesi dell gruppo dell Arion rufus. Real Accademia della Scienze di Torino 24:1-20. Page 156 POLLONERA, C. 1890. Recensement des Arionidae de la Region Paléarctique. Bolletino dei Musei di Zoologia ed Anatomia comparata della Real Universita di Torino 87(5):1-40. Seixas, M. M. P. 1976. Gasteropodes terrestres da fauna Por- tuguesa. Boletim da Sociedade Portuguesa la Ciéncia Na- turais 16:21-46. StMROTH, H. 1886. Weitere Mittheilungen tber palaearktische Nacktschnecken. Jahrbuch der deutschen Malakozoolo- gischen Gesellschaft 13:16-34. The Veliger, Vol. 35, No. 2 SIMROTH, H. 1889. Beitrage zur kenntniss der Nacktschneck- en. Nachrichtsblatt der deutschen Malakozoologischen Ge- sellschaft 11:177-186. StMROTH, H. 1891. Die Nacktschnecken der Portugiesisch- Azorischen fauna. Nova Acta der ksl. Leop.-Carol Deutsch- en Akademie der Naturforscher 56(2):1-224. The Veliger 35(2):157-159 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 NOTES, INFORMATION & NEWS New Range Information for the Banana Slug Ariolimax columbianus (Gould, 1851) by Lindsey T. Groves Malacology Section, Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, California 90007, USA A previously unreported locality for Ariolimax columbianus (Gould, 1851) in Plumas County, California, is herein documented. Los Angeles County Museum of Natural History (LACM) locality 89-38 is in the northernmost Sierra Nevada. Other Sierra Nevada localities have been reported by COOPER (1886) (Placer and Plumas [= Sierra] Counties), MEAD (1943) (Calaveras, El Dorado, Placer [= Cooper, 1886], and Tuolumne [= California Academy of Sciences (CAS) 078011 and 078034] Counties) and Piussry (1948) (Tuolumne County). A generalized lo- cality map in HARPER (1988) included the localities of MEaD (1943) and PILsBRY (1948). COOPER (1886:253) reported “the abundance of some kinds of land molluscs has attracted the attention of the miners at one locality, called “Slug Canon,” in Plumas County” (Slug Canyon is actually located in Sierra County southwest of Down- ieville, California, ca. 1098 m elevation [S. Larson, per- sonal communication, 1991; U.S. Geological Survey Downieville (PR 1975) 7.5’ quandrangle]). Cooper in- cluded this information under the heading of “Ariolimax Californicus, J.G.C.” [sic] [= A. columbianus (Gould, 1851) fide MEAD (1943)] and described it as “this great slug, apparently identical with the coast species.”’ Unpublished Sierra Nevada records from the CAS Invertebrate Zoology collection for A. columbianus include Calaveras, Nevada, and Plumas Counties (E. Kools, personal communication, 1991). The new LACM locality is in NW%, SW%, SEM, section 26, T26N, R9E, U.S. Geological Survey Crescent Mills (1980) 7.5’ quadrangle (40°04.3’N, 120°55.3'W) 2.75 km SW of Crescent Mills, Plumas County, California, at 1097 m elevation, 5 m upslope from Dixie Creek, a south- flowing tributary of Indian Creek on the property of Flora Berridge and Doris and Bruce Livingston. A single living specimen that measured 98 mm in length was collected by the author on 25 August 1989 and is preserved in 70% ethyl alcohol in the LACM Recent mollusk collection (LACM 89-38.4). This locality is significant in that it documents one of the highest elevations (1097 m) that Ariolimax columbianus has been collected and it extends its known range to the northernmost Sierra Nevada. Previ- ously reported high elevation Sierra Nevada localities in- clude: 1128 m near Alta, Placer County and ca. 1098 m near Downieville, Sierra County (COOPER, 1886); and ca. 985 m near Riverton, El Dorado County (MEAD, 1943). Unpublished high elevation localities from the CAS col- lection include CAS 077204 and 077215 1174 m at Rich Gulch, Plumas County. The author was alerted in 1988 to the presence of Arioli- max columbianus in Plumas County by the property own- ers, who had observed this species as early as 1976; how- ever, sightings since 1989 have been rare. Other molluscan taxa associated with this locality are juvenile Pristoloma sp. and Vespericola sierranus (Berry, 1921). A Dixie Creek stream-side leaf-litter sample (LACM 89-37) yielded the introduced species Oxychilus alliartus (Miller, 1822). A subsequent re-collection of these localities was made on 26 November 1989. LACM 87-107 (= LACM 89-38 in lo- cality only) yielded Vitrina alaskana Dall, 1905, and the introduced species Oxychilus draparnaldi (Beck, 1837). A leaf-litter sample from LACM 89-108 (= LACM 89-37 in locality only) yielded Vitrina alaskana Dall, 1905, Ves- pericola sierranus (Berry, 1921), and Helminthoglypta proles (Hemphill, 1892). Acknowledgments Special thanks to Flora Berridge and Doris and Bruce Livingston for reporting the presence of Ariolimax columbi- anus on their property. Thanks to Elizabeth Kools (CAS) and Paul Scott (Santa Barbara Museum of Natural His- tory) for providing timely locality information. Scott Lar- son (Plumas County Museum, Quincy, California) and Lois M. Schenk kindly provided historical information. C. Clifton Coney and James H. McLean (LACM), Barry Roth (University of California, Museum of Paleontology), and an anonymous reviewer critiqued the manuscript. Literature Cited Cooper, J. G. 1886. On fossil and sub-fossil land shells of the United States, with notes on living species. Bulletin of the California Academy of Sciences 1:235-255. Harper, A. B. 1988. The Banana Slug: A Close Look at a Giant Forest Slug of Western North America. Bay Leaves Press: Aptos, California. 32 pp., numerous unnumbered figs. Meap, A. R. 1943. Revision of the giant west coast land slugs of the genus Ariolimax Moerch (Pulmonata: Arionidae). American Midland Naturalist 30(3):675-717, figs. 1-25, pls. 1-4. Pitssry, H. A. 1948. Land Mollusca of North America (North of Mexico). Academy of Natural Sciences of Philadelphia, Monograph 3, 2(2):xlvii + 521-1113, figs. 282-585. Page 158 Use of Gonad Color in Sexing Broodstock of Placuna placenta (Linnaeus, 1758) by Wenresti G. Gallardo, Ma. Teresa R. de Castro, and Robmar T. Buensuceso Aquaculture Department, Southeast Asian Fisheries Development Center (SEAFDEC), 5021 Tigbauan, Iloilo, Philippines Introduction The windowpane oyster, Placuna placenta (Linnaeus, 1758), is a highly valuable bivalve species inhabiting the muddy bottom of coastal bays from the Arabian Sea on the west through the Indian Ocean and Malayan Seas to the coast of China on the east (HORNELL, 1909). In the Philippines, P. placenta is extensively collected from the wild because of the high demand for its translucent shell, which is fash- ioned into various articles exported to the United States and Europe. As a result of overharvesting, P. placenta populations in some natural beds are already depleted. Therefore, there is a need to conserve this resource through aquaculture. Hatchery techniques should be developed to produce seeds for reseeding and farming purposes. At the Aquaculture Department of the Southeast Asian Fisheries Development Center (SEAFDEC/AQD), Pla- cuna placenta has been induced to spawn by water flow manipulation (YOUNG, 1980). Other methods such as the addition of gametes, temperature shock, and salinity shock are presently being tried. With these induced spawning methods, there is a need to have a high degree of certainty as to the sex of the animal being used. Further, it is im- portant that the method of sexing does not involve sacri- ficing the animal. However, it has been reported that sexes in P. placenta can be determined only by gonad histology. ROSELL (1979) stated that male and female P. placenta, which are dioecious, are distinguishable upon histological examination of the gonads, but they are without apparent sexual dimorphism. MaGsucl et al. (1980) reported that the sex of P. placenta cannot be differentiated except by microscopic examination. Nevertheless, if gonad color, as seen through the translucent shell, could be proven to be a highly reliable basis for sexing, then this method would be useful in broodstock selection. Therefore, this study was conducted to investigate the use of gonad color for sexing P. placenta broodstock. Materials and Methods A total of 80 Placuna placenta individuals (64-135 mm) collected in the months of October 1989 and February and May 1990 from Pilar, Capiz, Philippines, by commercial skin divers were used in this study. The color of the gonad was determined by viewing the animal against a light and classified either as orange or cream, for female or male, respectively. The Veliger, Vol. 35, No. 2 Gonads were dissected, fixed in 10% formalin, dehy- drated in different grades of alcohol (70-100%), cleared in toluene, and embedded in 56°C paraffin wax. Sections of 4—5 um thickness were cut and stained with hematoxylin and eosin (BELL & LIGHTNER, 1988). Gonad sections were examined with an Olympus mi- croscope and the sex and gonadal stage determined on the basis of the categories used by ROSELL (1979). Results of the histological sex identification were then compared with gonad color to determine the reliability of gonad color as a basis for sexing. Results and Discussion The number of specimens in which gonad color corre- sponded to sex, or did not correspond, were, for several gonad stages: immature, 10 vs. 4; early active, 6 vs. 1; late active, 25 vs. 0; ripe, 15 vs. 0; partially spawned, 6 vs. 3; spent, 9 vs. 1. Among the 28 samples in October 1989, wherein 50% had immature gonads, 25% partially spawned to spent gonads, and only 3.67% ripe, the level of accuracy in match- ing gonad color with the correct sex was only 78.57%. On the other hand, in the February and May 1990 samples, having no immature individuals but a high percentage (71.97%) of individuals in the late active and ripe gonad stage, the level of accuracy was 90.91% and 96.67%, re- spectively. Histological analysis of the gonad confirmed the sex of the animal. Overall, a high percentage (88.75%) of the samples matched in gonad color and sex, cream for male and orange for female. The unmatched 11.25% of the samples were either in the immature, early active, partially spawned, or spent gonad stages. These findings show that gonad color can be used as a basis for sexing Placuna placenta with ripe gonads. It is not, however, as highly reliable in individuals with im- mature, partially spawned, or spent gonads. Gonad color is also an indication of the animal’s sexual maturity and readiness to spawn. If the color difference is distinct, then the animal is mature or ripe and can be easily induced to spawn. Therefore, in breeding Placuna placenta, those with distinct gonad color (cream or orange) should be chosen. Breeding attempts with this economically im- portant bivalve species at SEAFDEC/AQD now make use of this method of sexing in broodstock selection. This sexing method becomes limited by the thick and opaque shell that develops as a result of old age (MAGSUCI et al., 1980) and makes it difficult to ascertain the gonad color externally. Acknowledgments Thanks are due to Attorney R. Panique and Mrs. N. P. Vardeleon of Polyshell Inc., Philippines, for providing the Placuna placenta samples; G. Erazo, S. Torrento, and F. Torreta for the gonad histological preparation; and Prof. Notes, Information & News Page 159 N. C. Rosell for his helpful comments on the earlier draft of this manuscript. Literature Cited BELL, T. A. AND D. V. LIGHTNER. 1988. A Handbook of Normal Penaeid Shrimp Histology. World Aquaculture So- ciety: Louisiana. 114 pp. HORNELL, J. 1909. Report on the anatomy of Placuna placenta, with notes upon its distribution and economic uses. Report to the Government of Baroda on the Marine Zoology of Okhamandal in Kattiawar 1:43-97. Macsucl, H., A. CONLU & S. MoYANO-AypaA. 1980. The win- dow-pane oyster (kapis) fishery of Western Visayas. Fish- eries Research Journal of the Philippines 5(2):74-80. RosELL, N. C. 1979. A study on the biology and ecology of Placuna placenta Linne. Natural Applied Science Bulletin 31(3-4):203-251. YounG, A. L. 1980. Larval and postlarval development of the window-pane shell, Placuna placenta Linnaeus (Bivalvia: Placunidae) with a discussion on its natural settlement. The Veliger 23(2):141-148. International Commission on Zoological Nomenclature Comment or advice on applications to the ICZN is invited for publication in the Bulletin of Zoological Nomenclature and should be sent to the Executive Secretary, ICZN, % The Natural History Museum, Cromwell Road, London SW7 5BD, U.K. The following applications and opinions were published on 26 March 1991 in Vol. 48, Part 1, of the Bulletin: Case 2736—Haustator Montfort, 1810 (Mollusca, Gas- tropoda): proposed conservation by suppression of Aculea Perry, 1810, an unused senior subjective syn- onym. Case 2769—Laeocochlis Dunker & Metzger, 1874 (Mol- lusca, Gastropoda): proposed conservation as the cor- rect spelling. Case 2732—Ceratites nodosus (Cephalopoda, Ammonoi- dea): proposed attribution of the specific name to Schlotheim, 1813, and proposed designation of a lec- totype. Opinion 1623—Risomurex Olsson & McGinty, 1958 (Mollusca, Gastropoda): Ricinula deformis Reeve, 1846, designated as the type species. The following applications and opinions were published on 30 September 1991 in Vol. 48, Part 3, of the Bulletin: Case 2710—Clavidae McCrady, 1859 (Cnidaria, Hydro- zoa) and Clavinae Casey, 1904 (Mollusca, Gastrop- oda): proposal to remove the homonymy by changing the molluscan subfamily name to Clavusinae. Case 2766—Conus fulmen Reeve, 1843 (Mollusca, Gas- tropoda): proposed conservation by suppression of its unused senior subjective synonym C. modestus Sow- erby, [1833]. And Conus berghausi Michelotti, 1847: proposed precedence over C. demissus Philippi, 1836. Opinion 1650—Cymatiinae Iredale, 1913 (1854) (Mol- lusca, Gastropoda) and Cymatiinae Walton in Hutch- inson, 1940 (Insecta, Heteroptera): homonymy re- moved. Opinion 1651—Mytilus anatinus Linnaeus, 1758 (Cur- rently Anodonta anatina; Mollusca, Bivalvia): neotype designation confirmed. Student Research Grant in Malacology The Western Society of Malacologists and a coalition of western United States shell clubs have announced the availability of grants to support student research in mal- acology. Funds are available for actual research costs, in- cluding but not limited to field and laboratory equipment, chemicals, photographic supplies, computer time and sup- plies, microscope usage fees, and reasonable research travel costs. One or more research grants up to $1500 are avail- able. To be eligible, an applicant must be a full-time student in a formal graduate or undergraduate degree program. The research project must be focused primarily on the systematics, biology, ecology, physiology, biochemistry, or paleontology of marine, terrestrial, or freshwater mollusks. Completed applications must be received no later than 15 May 1992. For more information and an application send a self-addressed, stamped (if residing in U.S.) en- velope to: Malacology Grant, Department of Invertebrate Zoology, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105, USA. The Veliger 35(2):160 (April 1, 1992) THE VELIGER © CMS, Inc., 1992 BOOKS, PERIODICALS & PAMPHLETS Economic Zoology: A Dictionary of Useful and Destructive Animals by B. JANGI. 1991. A. A. Balkema Publishers, P.O. Box 1675, Rotterdam, The Netherlands. Hardback. 216 pp. Price: $63.00 Attracted to this book by an illustration of a giant squid on the cover, I asked for a review copy. The accompanying advertisement described the book as “the first attempt of its kind, resulting in a miniature encyclopaedia on eco- nomic zoology. From amoebas to apes it spans about 1,500 entries.” By “economic zoology” is meant the benefits (uses) and losses to humans caused by animals and animal prod- ucts. The dictionary entries briefly address various animal groups, pathological conditions afflicting them, and eco- nomic practices employed. Although the book is aimed, according to the author, at researchers in biological sciences, the book is more likely to serve a more generalist audience. Someone reading an article on silkworms in China in a popular science mag- azine might be interested in the fact that the silk of bom- bycids is secreted as a single continuous fiber some 800- 1200 m long, or the reader might find useful the definition of “‘frass: silkworm waste used as fish food and fertilizer in China.” A specialist, however, is not likely to find in the book much new in his or her field. Most malacologists probably would not be impressed with the fewer than 30 substantive entries concerning mol- lusks (some additional listings are simple cross-references). Although the entries extend from abalone to Zirfaea, in total the material on mollusks occupies about five pages only. One of these is devoted to “shellfish farming” and most of the rest are found under the class-level headings of Cephalopoda, Gastropoda, and Pelecypoda. Rabbits re- ceive treatment about equal to that of gastropods. A more generally worrisome feature of the book is some out-of-date information and illustrations of poor quality. For example, the cephalopod and pearl-farming entries cite annual catch figures from the 1960s only, and several of the already sparse illustrations are from pre-1960 edi- tions of Ralph Buchsbaum’s Animals without Backbones. Economic Zoology does contain some useful and inter- esting entries, and perhaps larger libraries should have a copy. But the $63 price tag is far too dear for the potential benefits that most individuals might receive—another form of economic zoology. D. W. Phillips Information for Contributors Manuscripts Manuscripts must be typed on white paper, 8/2” by 11”, and double-spaced throughout (including references, figure legends, footnotes, and tables). If computer generated copy is to be submitted, margins should be ragged right (v.e., not justified). To facilitate the review process, manuscripts, including figures, should be submitted in triplicate. The first mention in the text of the scientific name of a species should be accompanied by the taxonomic authority, including the year, if possible. Underline scientific names and other words to be printed in italics. Metric and Celsius units are to be used. The sequence of manuscript components should be as follows in most cases: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, lit- erature cited, figure legends, figures, footnotes, and tables. The title page should be on a separate sheet and should include the title, author’s name, and address. The abstract should describe in the briefest possible way (normally less than 200 words) the scope, main results, and conclusions of the paper. Literature cited References in the text should be given by the name of the author(s) followed by the date of publication: for one author (Smith, 1951), for two authors (Smith & Jones, 1952), and for more than two (Smith et al., 1953). The “literature cited” section must include all (but not additional) references quoted in the text. References should be listed in alphabetical order and typed on sheets separate from the text. Each citation must be complete, with all journal titles unabbreviated, and in the following form: a) Periodicals Cate, J. M. 1962. On the identifications of five Pacific Mitra. The Veliger 4:132-134. b) Books Yonge, C. M. & T. E. Thompson. 1976. Living marine molluscs. Collins: London. 288 pp. c) Composite works Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135. In: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stanford Univ. Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Figures and plates Figures must be carefully prepared and should be submitted ready for publication. Each should have a short legend, listed on a sheet following the literature cited. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photographs for half-tone plates must be of good quality. They should be trimmed off squarely, arranged into plates, and mounted on suitable drawing board. Where necessary, a scale should be put on the actual figure. Preferably, photographs should be in the desired final size. It is the author’s responsibility that lettering is legible after final reduction (if any) and that lettering size is appropriate to the figure. Charges will be made for necessary alterations. Processing of manuscripts Upon receipt each manuscript is critically evaluated by at least two referees. Based on these evaluations the editor decides on acceptance or rejection. Acceptable manuscripts are returned to the author for consideration of comments and criticisms, and a finalized manuscript is sent to press. The author will receive from the printer two sets of proofs, which should be corrected carefully for printing errors. At this stage, stylistic changes are no longer appropriate, and changes other than the correction of printing errors will be charged to the author at cost. One set of corrected proofs should be returned to the editor. An order form for the purchase of reprints will accompany proofs. If reprints are desired, they are to be ordered directly from the printer. Send manuscripts, proofs, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. CONTENTS — Continued The anatomy of Arion flagellus Collinge, 1893, present on the Iberian Peninsula (Gastropoda: Arionidae: Terrestria Nuda) JOSE. CASTILLEJO), csi YRS ee seers cue aes ees od Pet eee eae 146 NOTES, INFORMATION & NEWS New range information for the banana slug Ariolimax columbianus (Gould, 1851) LINDSEY SE-GROVES 9.3 Se ce eS od ee Li Use of gonad color in sexing broodstock of Placuna placenta (Linnaeus, 1758) WENRESTI G. GALLARDO, MA. TERESA R. DE CASTRO, AND ROBMAR T. IBUENSUGESO 255 job sf gb hea rae eae ere ot A a 158 BOOKS,. PERIODICALS &" PAMPHEETS 322. 3a) 4525 oe eee 160 ISSN 0042-3211 BL HE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 35 aibyete e992 Number 3 CONTENTS Humoral immunity: a,-macroglobulin activity in the plasma of mollusks RERERE DW ARMSDRONG ANDEJAMES bP. QUIGLEY 9852552... .5 224.5) 0a. 161 Laevipilina antarctica and Micropilina arntzi, two new monoplacophorans from the Antarctic ZNNIDERSIVVAREN AN DIO DEBAN ILAIN (52) eae as ooh cs Sods ee hs coe 165 Comments on and descriptions of eulimid gastropods from tropical West America ANINTIDIBIRG \AYIRIRIBIN ah oo Se lay a an Slee a ene ET) Geographic and temporal variation in shell morphology of Acanthina species from California and northern Baja California CARY Ges GIANNINY ANDUDANAUH: (GEARY 224.5402. - 60+. .-8s veces oem: 195 A new genus and species of Facelinidae (Opisthobranchia: Aeolidacea) from the Caribbean Sea SANDRAUVAIMITELEN ANDUJERPREY ©) ETAMANN ()) 95)... 2 ae ei a oe 205 A warm water Atlantic synonymy, Aphelodoris antillensis equals Chromodoris bistellata (Opisthobranchia: Gastropoda) BIERDRIE a Cae IANUANINE ert elope ecg n yeas -8 hah nye Rais evils, Caja es ZS A new genus and species of polygyrid land snail (Gastropoda: Pulmonata) from Oregon BARRW@IN ORHVAND VW ALTE RU Bw MINEICERS (ise iis ven S iis te AUN BT 222 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of January, April, July, and October. Rates for Volume 35 are $28.00 for affiliate members (including domestic mailing charges) and $56.00 for libraries and nonmembers (in- cluding domestic mailing charges). For subscriptions sent to Canada and Mexico, add US $4.00; for subscriptions sent to addresses outside of North America, add US $8.00, which includes air-expedited delivery. Further membership and subscription infor- mation appears on the inside cover. The Veliger is published by the California Ma- lacozoological Society, Inc., % Museum of Paleontology, University of California, Berkeley, CA 94720. Second Class postage paid at Berkeley, CA and additional mailing offices. POSTMASTER: Send address changes to The Veliger, Museum of Paleon- tology, University of California, Berkeley, CA 94720. THE VELIGER Scope of the journal The Veliger is open to original papers pertaining to any problem concerned with mol- lusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, evolutionary, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS’ in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Hans Bertsch, National University, Inglewood, California James T. Carlton, University of Oregon Eugene V. Coan, Research Associate, California Academy of Sciences, San Francisco J. Wyatt Durham, University of California, Berkeley William K. Emerson, American Museum of Natural History, New York Terrence M. Gosliner, California Academy of Sciences, San Francisco Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley David R. Lindberg, University of California, Berkeley James H. McLean, Los Angeles County Museum of Natural History Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Washington Barry Roth, Santa Barbara Museum of Natural History Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley Membership and Subscription Affliate membership in the California Malacozoological Society is open to persons (no institutional memberships) interested in any aspect of malacology. As an affiliate member, a person may subscribe to The Veliger for US $28.00 (Volume 35), which now includes mailing charges to domestic addresses. ‘There is a one-time membership fee of US $2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription; a reinstatement fee of US $3.00 will be required if membership renewals do not reach the Society on or before October 1 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign members, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $56.00 (Volume 35), which now includes mailing charges to domestic addresses. For subscriptions sent to Canada and Mexico, add US $4.00; for subscriptions sent to addresses outside of North America, add US $8.00, which includes air-expedited delivery. Memberships and subscriptions are by Volume only (January | to October 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $25.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: The Veliger, Museum of Paleontology, 3 Earth Sciences Bldg., University of California, Berkeley, CA 94720, USA. Send manuscripts, proofs, books for review, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616, USA. THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER. The Veliger 35(3):161-164 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 Humoral Immunity: a,-Macroglobulin Activity in the Plasma of Mollusks PETER B. ARMSTRONG Marine Biological Laboratory, Woods Hole, Massachusetts 02543, and Laboratory for Cell Biology, Department of Zoology, University of California, Davis, California 95616, USA JAMES P. QUIGLEY Marine Biological Laboratory, Woods Hole, Massachusetts 02543, and Department of Pathology, Health Sciences Center, State University of New York, Stony Brook, New York 11794, USA Abstract. The a,-macroglobulins are protease-binding proteins that shield bound proteases from active site protease inhibitors of suitably high molecular mass. The active site of the bound protease is unaffected, and the bound enzyme remains able to hydrolyze low molecular mass amide and ester substrates. The ability to protect the amidolytic activity of trypsin from the high molecular mass inhibitor soybean trypsin inhibitor was used to demonstrate an a,-macroglobulin-like activity in the hemolymph of representatives of bivalve and gastropod mollusks and blood of the cephalopod Loligo peali. This is the first documentation of the presence of a,-macroglobulin-based systems of immunity in mollusks. INTRODUCTION Proteases are involved in physiological processes such as blood clotting and in pathological processes such as neo- plastic invasion, inflammation, and invasion by pathogenic microorganisms and multicellular parasites. Since extra- cellular proteases in the internal milieu may be quite de- structive, animals have evolved regulatory processes for their inactivation and removal. Most important are the peptide protease inhibitors. Members of the a,-macro- globulin class of protease inhibitors act by the unique mechanism of a physical folding of the peptide chains of a,-macroglobulin around the protease molecule to form a “cage” that sterically blocks interaction of the protease with large molecules in the surrounding milieu (STARKEY & BARRETT, 1973). Amide and ester substrates small enough to diffuse into the cage can still be hydrolyzed since the active site of the entrapped protease is unaffected. a,- Macroglobulin is a major protein in mammalian plasma and has been proposed to function as an important scav- enger of proteases from the blood and tissue fluids (for reviews see FEINMAN, 1983; SOTTRUP- JENSEN, 1987). A specific assay for aj-macroglobulin derives from the ability of a@,-macroglobulin to protect trypsin from active site inactivation by soybean trypsin inhibitor (SBTI)! (GANROT, 1966; ARMSTRONG ef al., 1985). To detect a,- macroglobulin in a sample, the sample is reacted sequen- tially with saturating concentrations of trypsin and then with the high molecular mass active site inhibitor SBTTI. SBTI inactivates all free trypsin but fails to interact with a@-macroglobulin-bound trypsin. When a low molecular mass trypsin substrate is then added, the rate of substrate hydrolysis is a measure of the aj-macroglobulin-bound trypsin, and thus of a,-macroglobulin. In this report, we use this assay to document the presence of a,-macroglob- Page 162 ulin in the plasma of members of three classes of mollusks: the gastropod Busycon canaliculatum (Linnaeus, 1758), the bivalve Spisula solidissima (Dillwyn, 1817), and the ceph- alopod Loligo pealiu Lesueur, 1821. MATERIALS anpD METHODS Recently collected animals were obtained from the Marine Resources Center of the Marine Biological Laboratory. Hemolymph was obtained from the different species as follows: Busycon: large individuals were induced to express all seawater from the mantle cavity by repeatedly poking the foot. The margins of the shell were then removed to expose the foot, which was cut with a scalpel blade, and the hemolymph was collected in a beaker. Spisula: he- molymph was aspirated from the marginal sinus with a 10 mL syringe and a 19 ga needle. Loligo: blood was collected from the heart or the post-cardiac sinus with a 3 mL syringe and a 23 ga needle. The hemolymph from all three species was centrifuged to remove the blood cells, and the hemocyanin in the plasma of Busycon and Loligo was removed by precipitation with 3% polyethylene glycol (PEG)!. To avoid the formation of precipitates during the assay, the plasma of Spisula was dialyzed into 50 mM Tris, pH 8.1. The 3% PEG supernatant of Busycon he- molymph was exposed to 10% PEG and the precipitate was redissolved in 50 mM Tris, pH 8.1. No precipitate formed during the assay of the 3% PEG supernatant of Loligo blood, so this was assayed directly. The various samples were measured into individual 1 mL plastic spectrophotometer cuvettes with 50 mM Tris buffer, pH 8.1 as the diluent to a final volume of 1080 uL. Bovine pancreatic trypsin (Sigma cat. No. T-8003: 51% active, as determined by titration with p-nitrophenyl- p’-guanidobenzoate hydrochloride [CHASE & SHAW, 1967]) was then added and the sample was incubated for 10 min, room temperature. Excess SBTI was then added to in- activate all unbound trypsin. The amidolytic activity of the bound trypsin was determined spectrophotometrically by the rate of hydrolysis of Na-benzoyl-DL-arginine p-ni- troanilide (BAPNA). In the absence of a,-macroglobulin, the SBTI completely blocked the hydrolysis of BAPNA. This assay is specific for a-macroglobulin. To determine if a,-macroglobulin from selected mollusks was of the class whose activity requires the presence of a reactive internal thiol ester, inactivation with methylamine was attempted. Samples were treated overnight at room temperature with 0.2 M methylamine in 0.1 M Tris buffer, pH 8.1 as described previously (ARMSTRONG et al., 1985), with par- allel samples exposed to 0.1 M Tris buffer lacking me- thylamine as controls. ' Abbreviations: BAPNA = Na-benzoyl-DL-arginine p-ni- troanilide; PEG = polyethylene glycol; SBTI = soybean trypsin inhibitor. The Veliger, Vol. 35, No. 3 a, -MACROGLOBULIN IN THE PLASMA OF THE SQUID LOLIGO 0.4 = Z 03 —) = + © > 0.2 < re] —— Trypsin alone fo) —o—- Trypsin + 0.37 ml ZB 0.1 Loligo plasma ey —&— Trypsin + 0.02ml Limulus plasma —ts— Trypsin + 0.01 ml Limulus plasma 0.0 0 10 20 30 40 MINUTES a, -MACROGLOBULIN IN THE PLASMA OF THE BIVALVE MOLLUSK SPISULA 0.4 = 03 Z. — = + se 0.2 —— Trypsin alone 3 —t— Trypsin + 0.009 ml a Limulus plasma sa —h— Trypsin+ 1g & on Limulus «2M Dn —O— Trypsin +0.9 ml BI Spisula plasma —®— Trypsin + 0.9 ml MA- treated Spisula plasma 0.0 0 100 200 MINUTES Figure 1 Protection of the amidase activity of trypsin by a@,-macroglobulin in the plasma of mollusks. The plasma of the arthropod Limulus polyphemus was used as an internal standard since the form of a,-macroglobulin of this organism has been extensively charac- terized (ARMSTRONG & QUIGLEY, 1991). The plasma of the squid Loligo pealu (Figure 1a), contained about 7.7% of the a,-mac- roglobulin activity relative to that found in the plasma of Limulus, whereas the plasma of the bivalve Spisula solidissima (Figure 1b) contained lower levels of a,-macroglobulin. All samples of Figure la contained 5 yg of active trypsin and those of Figure 1b con- tained 2.5 ug. RESULTS anp DISCUSSION The three mollusks tested in the present study for a,- macroglobulin using the SBTI protection assay all showed a,-macroglobulin activity in the plasma, with Loligo show- P. B. Armstrong & J. P. Quigley, 1992 Table 1 a>-Macroglobulin activity in the plasma of representative mollusks. ug of trypsin pro- tected/ Genus Treatment of plasma mL Loligo none 6.4 Spisula none 0.20 Spisula methylamine 0.17 Busycon 3-10% PEG cut; x20 concentrated 0.25 ing the highest activity and Busycon the lowest (Table 1, Figure la, b). a,-Macroglobulin activity in whole hemo- cyanin-free plasma of Busycon was present at too low a level for direct detection but could be assayed following a 20-fold concentration of the high molecular mass proteins by precipitation with 10% PEG. Two categories of a,-macroglobulin have been de- scribed: the thiol ester class, which possesses a reactive internal thiol ester bond that is hydrolyzed during reaction with proteases (reviewed in SOTTRUP- JENSEN, 1989), and the ovostatin class, which lacks the thiol ester (NAGASE & Harris, 1983; NAGASE et al., 1983). Both classes of a,- macroglobulin bind proteases, and members of the two classes show significant amino acid sequence homology. A diagnostic feature of the thiol ester class of a,-macroglob- ulin is susceptibility to inactivation by small primary amines such as methylamine (BARRETT et al., 1979; SWENSEN & HowarD, 1979; ARMSTRONG & QUIGLEY, 1987), which act by direct attack on the thiol ester bond (TACK, 1983). The a,-macroglobulin activity of Spisula was resistant to methylamine under conditions where Limulus plasma suf- fered 95% inactivation (Figure 1b, Table 1), suggesting that Spisula a,-macroglobulin may lack a thiol ester bond that is necessary for activity. Unfortunately, the low quan- tity of a,-macroglobulin in the plasma of Spzsula has frus- trated our attempts to purify the protein and subject pu- rified preparations to other tests for the presence of the thiol ester (cf, ARMSTRONG & QUIGLEY, 1987; SPYCHER et al., 1987). Homologues of vertebrate a,-macroglobulin have re- cently been described from chelicerate (QUIGLEY & ARMSTRONG, 1983, 1985) and mandibulate (ARMSTRONG et al., 1985; HERGENHAHN & SODERHALL, 1985; SPYCHER et al., 1987) arthropods. These molecules share numerous functional properties with mammalian a,-macroglobulin and have notable identity at the level of peptide sequence in key functional domains (SPYCHER et al., 1987; HALL et al., 1989; SOTTRUP- JENSEN et al., 1990). Here we report that a,-macroglobulin-like proteins are present also in the blood of representatives of three classes of mollusks. Page 163 The true physiological functions of the a,-macroglob- ulins are unclear even in vertebrates, which have been extensively studied in this regard. It seems likely that one function is the scavenging of proteases (SOTTRUP- JENSEN, 1987). a,-Macroglobulin has also been discovered to bind a variety of growth factors, suggesting that it may serve as a messenger molecule for the delivery of cytokines to the appropriate cellular targets (JAMES, 1990). It also has recently been reported that a,-macroglobulin functions in a complement-like hemolytic pathway in the horseshoe crab, Limulus (ENGHILD et al., 1990). It is anticipated that a comparative study of a,-macroglobulin in a variety of phyla will illuminate its true function. Its presence in mollusks suggests that its involvement in immunity is im- portant enough to have ensured its retention for the one- half billion years of evolution that separates the divergence of the vertebrate and mollusk lineages. ACKNOWLEDGMENTS This research was supported by grant No. GM35185 from the National Institutes of Health. We thank Drs. Kensal Van Holde, John Valois, and Sidney Pierce for instruction on bleeding mollusks. LITERATURE CITED ARMSTRONG, P. B. & J. P. QUIGLEY. 1987. Limulus a)-mac- roglobulin. First evidence in an invertebrate for a protein containing an internal thiol ester bond. Biochemical Journal 248:703-707. ARMSTRONG, P. B. & J. P. QUIGLEY. 1991. a@-Macroglobulin: a recently discovered defense system in arthropods. Pp. 291- 310. In: A. P. Gupta (ed.), Immunology of Insects and Other Arthropods. CRC Press: Boca Raton, Florida. ARMSTRONG, P. B., M. T. ROSSNER, & J. P. QUIGLEY. 1985. An a,-macroglobulinlike activity in the blood of chelicerate and mandibulate arthropods. Journal of Experimental Zo- ology 236:1-9. BARRETT, A. J.. M. A. BROWN & C. A. SAYERS. 1979. The electrophoretically “slow” and “fast” forms of the a@-mac- roglobulin molecule. Biochemical Journal 181:401-418. CuasE, T. & E. SHAW. 1967. p-Nitrophenyl-p’-guanidinoben- zoate HCl, a new active site titrant for trypsin. Biochemical and Biophysical Research Communications 29:508-514. ENGHILD, J. J., I. B. THOGERSEN, G. SALVESEN, G. H. Fey, N. L. FIGLER, S. L. Gontas & S. V. Pizzo. 1990. a-Mac- roglobulin from Limulus polyphemus exhibits proteinase in- hibitory activity and participates in a hemolytic system. Bio- chemistry 29:10070-10080. FEINMAN, R. D. 1983. Chemistry and Biology of a,-Macro- globulin. Annals of the New York Academy of Sciences. Vol. 421. New York Academy of Sciences: New York. 478 pp. GanrROT, P. O. 1966. Determination of a-macroglobulin as trypsin-protein esterase. Clinica Chimica Acta 14:493-501. HALL, M., K. SODERHALL & L. SOTTRUP-JENSEN. 1989. Ami- no acid sequence around the thiolester of a,-macroglobulin from plasma of the crayfish, Pacifastacus leniusculus. FEBS Letters 254:111-114. HERGENHAHN, H.-G. & K. SODERHALL. 1985. a -Macroglob- ulin-like activity in plasma of the crayfish Pacifastacus len- Page 164 iusculus. Comparative Biochemistry and Physiology 81B:833- 835. JaAMEs, K. 1990. Interactions between cytokines and a-mac- roglobulin. Immunology Today 11:163-166. NaGasE, H. & E. D. Harris. 1983. Ovostatin: a novel pro- teinase inhibitor from chicken egg white. II. Mechanism of inhibition studied with collagenase and thermolysin. Journal of Biological Chemistry 258:7490-7498. NacasE, H., E. D. Harris, J. F. WOESSNER & K. BREW. 1983. Ovostatin: a novel proteinase inhibitor from chicken egg white. I. Purification, physicochemical properties and tissue distribution of ovostatin. Journal of Biological Chemistry 258:7481-7489. QUIGLEY, J. P. & P. B. ARMSTRONG. 1983. An endopeptidase inhibitor, similar to mammalian a@,-macroglobulin, detected in the hemolymph of an invertebrate, Limulus polyphemus. Journal of Biological Chemistry 258:7903-7906. QuIGLey, J. P. & P. B. ARMSTRONG. 1985. A homologue of a@-macroglobulin purified from the hemolymph of the horse- shoe crab Limulus polyphemus. Journal of Biological Chem- istry 260:12715-12719. SOTTRUP- JENSEN, L. 1987. a,-Macroglobulin and related thiol ester plasma proteins. Pp. 191-291. Jn: F. W. Putnam (ed.), The Plasma Proteins. Structure, Function, and Genetic Con- trol. 2nd ed., Vol. 5. Academic Press: New York. The Veliger, Vol. 35, No. 3 SOTTRUP- JENSEN, L. 1989. a,-Macroglobulins: structure shape, and mechanism of proteinase complex formation. Journal of Biological Chemistry 264:11,539-11,542. SOTTRUP- JENSEN, L., W. BorTH, M. HALL, J. P. QUIGLEY & P. B. ARMSTRONG. 1990. Sequence similarity between a,- macroglobulin from the horseshoe crab, Limulus polyphemus, and proteins of the a,-macroglobulin family from mammals. Comparative Biochemistry and Physiology 96B:621-625. SPYCHER, S. E., S. AryA, D. E. ISENMAN & R. H. PAINTER. 1987. A functional, thioester-containing a,-macroglobulin homologue isolated from the hemolymph of the American lobster (Homarus americanus). Journal of Biological Chem- istry 262:14,606-14,611. STARKEY, P. M. & A. J. BARRETT. 1973. Human cathepsin B1. Inhibition by a-macroglobulin and other serum pro- teins. Biochemical Journal 131:823-831. SWENSON, R. P. & J. B. Howarp. 1979. Characterization of alkylamine-sensitive site in a-macroglobulin. Proceedings of the National Academy of Sciences, USA 76:4313-4316. Tack, B. F. 1983. The 6-Cys-6-Glu thiolester bond in human C3, C4, and a,-macroglobulin. Springer Seminars in Im- munopathology 6:259-282. The Veliger 35(3):165-176 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 Laevipilina antarctica and Micropilina arntzt, Two New Monoplacophorans from the Antarctic by ANDERS WAREN Swedish Museum of Natural History, Box 50007, S-104 05 Stockholm, Sweden AND STEFAN HAIN Alfred Wegner Institut flr Polar- und Meeresforschung, Columbusstrasse, D-2850 Bremerhaven, Germany Abstract. Two new monoplacophoran species, Laevipilina antarctica and Micropilina arntzi, are described from a depth of 191-742 m in the Lazarev Sea and eastern Weddell Sea. Laevipilina antarctica closely resembles the previously described species of Laevipilina McLean, 1979, and differs mainly in the convexity of the shell and in minor details of radular morphology. The stomach and the intestine of L. antarctica contained fine bottom sediment. The specimens range in length from 1.2 to 3.0 mm and the number of leaflets on the gills increases during ontogeny. Micropilina arntzi is the first live- taken species of its genus, and it is also the smallest known monoplacophoran, with a maximum shell length of 0.92 mm. It broods the young in the distal part of the oviduct and the pallial groove, and they are born at a size of 300 um shell diameter. Both species were found on sediment bottoms with stones and shells, on which they evidently live. INTRODUCTION The number of known Recent species of the class Mono- placophora has slowly but steadily risen since the first discovery of a living species by the Galathea Expedition in 1952. The number has now reached 17, and we list them (Table 1) and plot their distribution (Figure 1). In ad- dition, there are a few records of unidentified specimens that are also mapped. The specimens reported here are the first records of Monoplacophora from Antarctica. ROSEWATER (1970) re- ported an undescribed species from the Scotia Ridge (54°44'S, 55°33’W; Figure 1, number 6), southeast of the Falkland Islands, in 1647-2044 m depth. This locality is, however, separated from the Antarctic continental rise by depths exceeding 3000 m. We have examined this specimen (U.S. National Museum of Natural History, Division of Mollusks, No. 680431): it does not belong to Laevipilina or Micropilina, but seems to be a young specimen, perhaps of Neopilina, since the prismatic layer has a primarily radial orientation, rather than the neatly concentric ar- rangement of conspicuous prisms typical for Laevipilina. FILATOVA et al. (1975) reported Neopilina sp. from the same area (56°29'S, 50°51'W; Figure 1, number 3) at 4664-5631 m depth. This specimen was later described as Neopilina (Lemchephyala) rebainsi Moskalev et al., 1983. That record is, however, also from north of the Antarctic continental rise. Two specimens of the Laevipilina species described herein were found attached to rocks, but most of these and all specimens of the new Muicropilina species were recovered from sorting sediment samples that had been sieved on board the R/V Polarstern, on a 0.5-mm-mesh sieve, and preserved in 95% ethanol. This explains the rather poor preservation of most specimens. Preservation in 10% formalin, buffered with 10 g of sodium-tetraborate per liter sample volume and added to the final sample, would have given a better result, especially if the con- tainer had been shaken a few times during the first two or three days. The discovery that Micropilina arntzi is a brooder does not give any clues about the larval development of other monoplacophorans of which the apical area has been il- Page 166 The Veliger, Vol. 35, No. 3 Table 1 Distribution of Recent Monoplacophora. Condi- No. Species name and referencest tion} 1 Neopilina galatheae Lemche, 1957 MENZIES & LAYTON 1962 l l 2 N. bruunit Menzies, 1968 l 3 N. rebainsi Moskalev et al., 1983 ] 4 N. sp. l MENZIES (1968) l l 5) N. sp. ] MOSKALEV et al. (1983) 6 N. sp. ROSEWATER (1970) Uf Adenopilina adenensis (Tebble, 1967) 8 Vema ewingi (Clarke & Menzies, 1959) MENZIES (1968) 9 Vema bacescui (Menzies, 1968) ] 10 Laevipilina hyalina (McLean, 1979) ] 11 L. rolani Warén & Bouchet, 1990 l 12 L. antarctica Warén & Hain, herein ese > 13 Rokopella oligotropha (Rokop, 1972) ] 14 R. zografi (Dautzenberg & Fischer, 1896) d CESARI et al. (1987) d CESARI et al. (1987) d 15 R. goes (Waren, 1988) d d d 16 R. veleronis (Menzies & Layton, 1963) l 17 Micropilina minuta Waren, 1989* 18 M. tangaroa Marshall, 1990 19 M. arntzi Warén & Hain, herein ld 20 Monoplacophorus zenkewitchi Moskalev et al., 1, r 1983 Depth Size Locality (m) (mm) Off Costa Rica, 09°23'N, 89°32’W 3591 29-37 Off Costa Rica, 10°07'N, 89°50’W 3718 Off Peru, 08°54’S, 80°41'W 4823-4925 15 SE of Falkland Is., 56°29’'S, 50°51’'W 4664-5631 18 Off Peru, 08°46’S, 80°44’W 3909-3970 Off Peru, 11°30’S, 79°25’W 6146-6354 Off Peru, 08°52’S, 80°47'W 6313-6146 Off N Chile, 23°50.0'S, 71°06.0'W 4600 5 SE of Falkland Is., 54°44'S, 55°33’W 1627-2044 D3) Off South Yemen, 13°50'N, 51°47’E 3950-3000 10.7 Off Peru, 07°35’S, 81°24’W 5817-5834 25 07°30'S, 81°25’'W 5841-5854 10°13’S, 80°05'W 6324-6329 12°02'S, 79°08'W 5607-5614 Off Peru, 08°25’S, 81°05’'W 6260-6052 08°20'S, 81°04’W 6260-6364 08°16’S, 81°05’'W 6156-6489 11°30'S, 79°25’W 6146-6354 08°10.5'S, 81°908.1'W 6002 Off Peru, 08°44’S, 80°45’W 5986-6134 28 Off California, 32°41'N, 119°32'W 174-384 3 Off Spain, 42°52’N, 11°51’'W 985-1000 2 Weddell and Lazarev seas, Antarctica 210-644 3.0 Mid-Pacific, 30°05’N, 156°12’W 6065-6079 2.6 Azores 1385-1600 Si Mediterranean, NE of Corsica 180-500 Mediterranean, E of Sardinia 480-900 Caribbean, Virgin Islands 360-540 lei Baja California, 27°52’N, 115°45'W 2730-2769 2.6 Iceland, 63°23'N, 13°25'W 770-926 1.0 N of New Zealand, 31°31'S, 172°50'E 1216-1385 i155) Lazarev Sea, Antarctica 191-765 2.6 W of Hawaii, 20°41.7'N, 170°52.9'W 2000 4.8 + “Author comma date” indicates original description and records therein; ““REFERENCE (date)’’ are additional records. £1—found alive; d—only shells found; r—found attached on rocks. * Also known as a Pleistocene fossil from deep-water deposits in Reggio Calabria, southern Italy (TAVIANI, 1990). lustrated (WAREN 1988, 1989), although it seems likely that M. minuta also is a brooder, judging from the similar shape and size (340 wm) of the apical area. Specimens of the two new species are at present being investigated anatomically by G. Haszprunar (Innsbruck). Recently it has been advocated, especially in the pale- ontological literature (see, for example, PEEL [1991] and references therein), that the name Monoplacophora should be abandoned. The reason given is that the concept of the name has changed considerably since its introduction. The name Monoplacophora was introduced by Odhner in WENZ, 1940, for the superfamily Tryblidioidea. It was intended to be of the same rank as Polyplacophora and conceived to contrast the Polyplacophora. This stabilizes the name as the name of the class containing 7ryblidium Lindstr6ém, 1880, and its type species 7. reticulatum Lind- strém, 1880. The name Monoplacophora has since been extensively used to include Trybliodioidea (sensu Wenz) as the most important taxon. Later some authors have added and removed smaller groups of Paleozoic mollusks. We do not consider such changes in the concept of a taxon to be a convincing reason for changing its name. We have not noticed that the proponents for abandonment have suggested changing the name Gastropoda, another class that has been exposed to similar transfers of Paleozoic taxa. The following abbreviations for institutions are used in the text: SMF—Senckenbergisches Museum und For- schungsinstitut, Frankfurt; SMNH—Swedish Museum of Natural History, Stockholm. A. Warén & S. Hain, 1992 Page 167 150° 120° 90° 60° 30° © 30° 60° 90° 120° 150° 180° 1SOZ 120% 1902602 3027 C2 3025 G02 = 902) 1202” 1502 180? Figure 1 Map showing distribution of Recent Monoplacophora, based on Table 1. Laevipilina McLean, 1979 Vema (Laevipilina) MCLEAN 1979:9. Type species, V. (L.) hyalina McLean, 1979, by original designation. Type locality, off Lower California, 373-384 m. Remarks: MOSKALEV ef al. (1983) considered Laevipilina to be a valid genus and made a new family for it. WAREN (1989) considered Monoplacophorus Moskalev, 1983 (type species M. zenkewitchi Moskalev et al., 1983) to be a syn- onym of Laevipilina, but has since changed his opinion; Monoplacophorus can probably be considered a valid genus, differing from Laevipilina by its more depressed shape. WaREN & BOUCHET (1990) reviewed current knowledge about Laevipilina and described Laevipilina rolani, a new species from off northwestern Spain in 1000 m depth. Laevipilina antarctica Waren & Hain, sp. nov. (Figures 2-5, 6-8, 10-14, 15-16, 19, 27) Type material: Holotype (Figure 3), SMF 309243, and 2 broken paratypes (radula extracted) SMF 309244; 4 paratypes (1 sectioned) from station 248, SMNH 4285, 1 from station 158, SMNH 4352. Type locality: Polarstern Expedition ANT VII/4, station 245, 75°40.4’S, 029°37.2’W, 480 m, 3 specimens. Sand and gravel with stones and rich megafauna (see ARNTZ et al., 1990). Materials examined: The type material and: —Polarstern Expedition ANT VII/4, station 248, 74°39.3'S, 029°34.4'W, 600 m, sand and gravel with stones and rich megafauna (see ARNTZ et al., 1990), 4 specimens (1 specimen serially sectioned, 3 specimens left in alcohol). —Polarstern ANT IX/3, station 158, 72°21.8'S, 16°51.2'W to 72°21.0’S, 16°48.6'W (end of haul), 623-539 m, 1 spec- imen, 1.26 mm diameter, on a stone of 15 cm diameter. —Polarstern ANT IX/3, station 174, 69°43.7'S, 10°44.7'E to 69°42.4'S, 10°47.5'E, 432-432 m, silt with small stones, 1 specimen, 1 shell. —Polarstern ANT IX/3, station 180, 69°57.4'S, 06°19.0'E to 69°57.7'S, 06°21.0'E, 280-298 m, 1 specimen, 2.4 mm diameter. —Polarstern ANT IX/3, station 207, 69°57.4'S, 05°08.4'E to 69°57.5'S, 05°00.4'E, 213-210 m, silt with scattered stones with rich megafauna, large quantities of the bra- chiopod Magellania fragilis Smith, 1907, 1 specimen, 2 shells. —Polarstern ANT IX/3, station 212, 70°00.5'S, 03°56.4'E to 70°00.4'S, 03°57.3’E (end of haul), 568-644 m, 1 spec- imen, 3.0 mm diameter, on a boulder of 80 cm diameter, 4 specimens and 1 shell free in sediment. Description: The shell (Figures 2-5) is small, fragile, depressed, and transparent with a flat peristome. The apex (Figure 2) is slightly mamillate, forms an angle of about 60° with the basal plane, and is situated slightly behind the anterior margin. The apical area measures 230 x 190 um and has no distinct sculpture, only some regularly shaped impressions. Outside this area commences a uni- form, concentric sculpture of low, raised ridges, formed by the concentric arrangement of the prisms of the prismatic Page 168 The Veliger, Vol. 35, No. 3 Explanation of Figures 2 to 5 Figures 2-5. Laevipilina antarctica Waren & Hain, sp. nov., shell. Figure 2. Lateral view of apex. Scale line 50 um. Polarstern station 158. Figure 3. Dorsal view of holotype, periostracum removed. Length 2.06 mm. Figures 4 and 5. Dorsal and lateral view, periostracum left. Length 1.26 mm. Polarstern station 158. layer, which also form indistinct and fragmentary radial ridges. The shell is rather low, with the posterior surface evenly convex and the highest point situated somewhat behind the apex. The prisms (Figures 6-8) are regularly hexagonal and have a diameter of about 30 wm and a height of about 15 wm. Towards the edge of the shell they are larger and less uniformly shaped (Figure 6). The thick- ness of the nacreous layer (Figures 7, 8) is 5-10 um, and it starts only a short distance from the margin of the shell. The periostracum is rather thick and tough. Dimensions. Holotype, length 2.06, breadth 1.76, and height of shell 0.50 mm. Maximum length of the shell 3.0 mm. Soft parts (Figures 15, 16, 19). The velar lobes are well developed and strongly ciliated. The anterior lip is con- spicuous and evidently rather thinly cuticularized. The A. Waren & S. Hain, 1992 Page 169 Explanation of Figures 6 to 9 Figures 6-8. Laevipilina antarctica, shell structure. Figure 6. Interior view of edge of shell (facing lower edge of figure). Scale line 50 um. Figure 7. Fragment of shell, folded double and held together by the periostracum (p). Prismatic layer partly dissolved between periostracum and nacre (n). Scale line 10 wm. Figure 8. Fragment of shell, prismatic layer (p) partly dissolved, nacreous (n) layer intact. Scale line 10 um. Figure 9. Micropilina arntzi Waren & Hain, sp. nov., shell structure. Fragment of shell, exterior surface facing right. Nacreous layer (n) well developed. Scale line 10 um. postoral tentacles are much less developed compared with Laevipilina rolani, and are more similar to those in L. hyalina. They are short and claviform and equipped with about 7 short and stumpy distal appendages. The gills are not well preserved, but the smallest specimen (1.1 mm) has four pairs of gills with O (anterior pair), 0, 1, and 1 (posterior pair) “digits,” respectively. The largest speci- men (from station 212; Figure 19) is well preserved. It has 3, 4, 4, 4, and 4 digits, respectively, on the gills starting with the anterior one. The foot (contracted) measures 1.5 x 0.9 mm. The gonads are visible, by transmitted light, as a large, lobate, dorsal sac along each side of the animal. The anus does not open on a papilla, but is a simple opening in the pallial furrow. The four coils of the intestine could not be discerned by transmitted light. Radula (from a 1.5 mm specimen; Figures 10-14, 27). Length 1.6 mm, width 0.066 mm, with about 65 transverse rows, each of 11 teeth. The outermost tooth (assigned number 6) has no distinct cusp, and an only slightly uneven cutting edge. Tooth number 5 is large, fan-shaped, and Page 170 The Veliger, Vol. 35, No. 3 Explanation of Figures 10 to 14 Figures 10-14. Laevipilina antarctica, radula. Figure 10. Posterior view. Figure 11. Anterior view. Figure 12. Vertical view. Figure 13. Detail of central tooth, posterior view. Figure 14. Newly formed section of radula. Notice the incompletely formed fourth and fifth tooth. Numbers indicate the order of the teeth; the central tooth is 1. Scale lines are 10 um except Figure 13, which is 5 um. A. Waren & S. Hain, 1992 Page 171 Explanation of Figures 15 to 18 Figures 15 and 16. Laevipilina antarctica. Critical point dried bodies. Scale lines 200 um. Polarstern station 173. Figures 17 and 18. Micropilina arntzt. Critical point dried bodies. Fragments of the periostracum are still attached to the pallial margin and the shell. Scale lines 100 um. Polarstern station 173. Key: A, anterior lip; E, embryo; F, foot; P, periostracum (pulled off from shell); PL, posterior lip; PT, postoral tentacles; S, shell; V, velum; 1-5, gills in numerical order. Page 172 The Veliger, Vol. 35, No. 3 Explanation of Figure 19 Figure 19. Laevipilina antarctica, living specimen, diameter 3.0 mm. Polarstern station 212. somewhat similar to a hay-rake, with about 50 lamellar hooks. Tooth number 4 has a truncated and serrated cut- ting edge. Tooth number 3 is slightly smaller than number 4 and has a laterally situated main cusp and about 6 or 7 more central denticles. Tooth number 2 is hand-shaped, with two apically and laterally situated primary cusps and some smaller denticles along the inner side. The central tooth is small, inconspicuous, ridgelike with a small apical cusp. The large and conspicuous tooth number 5 is the last one to be formed during the continuous process of radular formation (Figure 14). Remarks: The prismatic layer is barely visible in incident light, while the concentric sculpture dominates. In trans- mitted light the prisms are clearly visible (Figure 19). The apical area (Figure 2) is probably a larval shell, as is indicated by the presence of a periostracum of which only patches remain. The periostracum was evidently con- tinuous and covered all the shell, but a part of it still covers the transition from the apical area, over to the part of the shell that has concentric and radial sculpture. Laevipilina antarctica differs from L. hyalina in having a less prominent central radular tooth and in having five instead of six pairs of gills, although the specimens are of A. Warén & S. Hain, 1992 the same size. Another difference is that the prisms of L. hyalina are about as high as they are wide, while in L. antarctica they are distinctly shorter than they are wide. Laevipilina rolani differs in having more fully developed postoral tentacles, and in having a more convex shell, its height corresponding to half the length of the shell, while it is only 0.33 of the length in L. antarctica. The specific identification of monoplacophorans is still a problem since very few species are known from more than a single locality and several of the species belonging to genera of small species are very similar to each other. We are therefore not certain about the validity of the criteria used for their separation. It is even possible that Laevipilina antarctica belongs to one of the previously described species, although experience from gastropod lim- pets suggests otherwise. In this connection it is significant to note that the number of digits on the gills varies with the size of the specimens, as mentioned in the description of the soft parts. The postoral tentacles, however, are identical throughout on- togenetic development, as far as could be seen. Most specimens were found free in sediment samples brought home and sorted in the laboratory; only two spec- imens (from stations 158 and 212) were found on stones after intensive search of hundreds of stones of various sizes. This may be because they are scratched off the stones or crushed in the trawl by the surrounding bottom material. We therefore assume that all specimens had been living on stones or old shells, which were common in the bottom material of all stations. One small specimen was serially sectioned to examine the stomach contents, which consisted of numerous mineral particles, unidentified organic material, scattered sponge spicules, radiolarian fragments, a small nematode, and a few polychete bristles. The end of the radula that is in use shows conspicuous signs of wear. Most of the cusps of teeth numbers 1-4 are worn off to form a simple, rounded edge (Figure 27) in- stead of a sharp serration. This supports the assumption that Laevipilina antarctica obtains its food by scraping off the thin layer of sediment, which in these depths covers most hard surfaces. Micropilina Waren, 1989 Micropilina WAREN, 1989:2. Type species, M. minuta Wa- rén, 1989, by original designation. Type locality, off southwestern Iceland, 900-926 m. Remarks: The genus was based on a few empty shells, characterized by the shape, sculpture, and presence of dis- tinct interior muscle scars. MARSHALL (1990) described a second species of the genus from north of New Zealand (Figure 1, number 18), from 1216-1385 m depth, known from a single shell. The new species described here conforms well with a Page 173 position in Micropilina. The shape is similar to that of the previously described species, and the sculpture is of the same construction, although finer. The sculpture of the species of Micropilina bears some resemblance to that of the subapical part of the shell of Rokopella (WAREN, 1988:figs. 3, 8), but it is too early to judge what this means about relationships, since the shell of most monoplacophorans has not been well enough il- lustrated to allow comparisons. Micropilina arntzi Waren & Hain, sp. nov. (Figures 9, 17, 18, 20-26, 28, 29) Type material: Holotype SMF 309780 and 5 paratypes SMF 309781, 20 paratypes SMNH 4380. Type locality: R/V Polarstern ANT IX/3, station 173, 70°00.5'S, 07°09.1'E to 70°00.4'S, 07907.4’'E (end of haul), 739-765 m, 25 specimens, 10 shells, maximum diameter 0.92 mm. Material examined: The type material and: —Polarstern ANT IX/3, station 165, 70°18.9’S, 03°15.8’W to 70°19.2’S, 03°16.8'W, 191-204 m, silt with stones and rich megafauna, 1 shell, 1 specimen. —Polarstern ANT IX/3, station 174, 69°43.7'S, 10°44.7'E to 69°42.4'S, 10°47.5'E, 432-432 m, silt with small stones, 7 specimens, 15 shells, maximum diameter 0.85 mm. —Polarstern ANT IX/3, station 180, 69°57.4'S, 06°19.0’E to 69°57.7'S, 06°21.0'E, 280-298 m, 17 specimens, 1 shell, maximum diameter 0.88 mm. —Polarstern ANT IX/3, station 206, 69°06.9'S, 10°01.0’E to 69°46.8'S, 10°01.6’E, 343-338 m, 1 shell. —Polarstern ANT IX/3, station 207, 69°57.4’S, 05°08.4’E to 69°57.5'S, 05°00.4'E, 213-210 m, silt with scattered stones with rich megafauna, large quantities of the bra- chiopod Magellania fragilis Smith, 1907, 2 shells, 1 spec- imen. —Polarstern ANT IX/3, station 211, 69°58.9'S, 05°08.4'E to 69°57.9'S, 05°00.4'E, 661-742 m, 5 shells. —Polarstern ANT IX/3, station 212, 70°00.5’S, 03°56.4’E to 70°00.4'S, 03°57.3'E, 568-644 m, 4 specimens. Descriptions: The shell (Figures 20-26) is very small, fragile, inflated, almost semiglobular with a large, bulbous apex and flat peristome. The apex (Figure 24) is mamillate and forms an angle of about 45° with the basal plane (Figure 22). There is no distinct sculpture on the slightly worn apical area, apart from occasional small pits, which also were seen in late, brooded young. This area measures 300 x 270 wm. Outside this area commences a fine, ir- regularly concentric striation (Figure 25), not visible with a stereomicroscope. In the furrows between the ridges are numerous small pits, diameter 2-5 wm. Under a stereo- microscope the whole shell has a seemingly granular sur- face, but this is caused by the pits. The shell is unusually Page 174 The Veliger, Vol. 35, No. 3 A. Waren & S. Hain, 1992 Page 175 Explanation of Figures 27 to 29 Figure 27. Laevipilina antarctica, central part of radula with worn edges of teeth number 1 and 2. Figures 28 and 29. Micropilina arntzi, radula, in perpendicular view and with stub tilted 45° to show the posterior surface of the cusps. Teeth numbered 1-6 from central tooth (1). Scale lines 5 ym. convex, with the apex well in front of the anterior edge and the highest point of the shell slightly anterior to the center of the shell. No muscle scars could be discerned on the interior. Slightly more than half the thickness of the shell consists of an interior nacreous layer (Figure 9), the exterior layer does not contain defined prisms. Dimensions. Holotype 0.84 x 0.76 mm, height 0.38 mm. Maximum length of the shell 0.92 mm. Soft parts (Figures 16, 18). The head is unusually large and bulging, with short, tapering, strongly ciliated velar lappets at the sides. The anterior lip seems to be very solid and cuticularized. Postoral tentacles are not present. The Explanation of Figures 20 to 26 Figures 20-26. Micropilina arntzi, shell. Figure 20. Fully developed juvenile from oviduct. Length 305 um. Polarstern station 173. Figures 21-24. Anterior view, lateral view, dorsal view, and apex magnified. Length 0.92 mm. Scale line (Figure 24, only) 50 wm. Polarstern station 173. Figures 25 and 26. Dorsal view and sculpture. Length 0.85 mm.-Scale line (Figure 25 only) 50 um. Polarstern station 211. White arrows indicate border of larval shell. Page 176 foot is round with a thickened rim. Three pairs of small, simple, tubercular gills are situated in the pallial groove and lack appendages. Five small, close-set muscle bundles, diameter 20-50 um, can be seen with transmitted light, situated along the central third of the body and halfway between the midline and the lateral margin. Most speci- mens have one or two embryos under development, evi- dently partly contained in the opening of the gonoduct. The smallest embryos were ovate and of a diameter of about 100 um, the largest ones 300 wm (Figure 20). Radula (Figures 28, 29). The length of the radula slight- ly exceeds that of the shell. The outermost tooth (number 6) is scalelike with a smooth, rounded cutting edge. The large tooth (number 5) has about 20 or 25 hooks in the rakelike comb of teeth. Teeth numbers 4, 3, and 2 resemble each other and have 7, 6, and 4 cusps, respectively. The size of these three teeth diminishes towards the center of the ribbon. The central tooth (number 1) has a fully formed, tricuspidate cutting plate and a sturdy, narrow central supporting ridge. Remarks: Among the monoplacophorans described so far, this one has been found in the greatest numbers, and it is the only one for which the mode of development is known. The young are evidently born at the crawling stage. The increase in size of the embryos must mean that some kind of transfer of nutrients, supplied by the parent, takes place, but the mechanism is not known. The radula of Micropilina arntzt is unusual in that the central and first marginal teeth are less reduced than in all other species for which the radula has been described. Those species have a radula that more closely resembles those of Laevipilina species. ACKNOWLEDGMENTS We want to thank the crew of R/V Polarstern for their assistance during the collecting program and Mme. J. Galeron at the Oceanographic Sorting Centre, Brest, where a part of the material was sorted. K. Rigneus (SMNH) sorted the material from Polarstern Cruise IX /3, A. Hed- strém (SMNH) prepared the serial sections of Laevipilina antarctica, and C. Hammar (SMNH) prepared all pho- tographic prints. Drs. B. A. Marshall (Wellington) and J. H. McLean (Los Angeles) read and gave valuable comments on the manuscript. LITERATURE CITED ARNTZ, W., E. Ernst & I. HEMPEL. 1990. The expedition Antarktis VII/4 (EPOS leg 3) and VII/5 of RV “Polar- stern” in 1989. Berichte zur Polarforschung 68:1-214. The Veliger, Vol. 35, No. 3 CEsaRI, P., F. Grust1 & A. MINELLI. 1987. Recent monopla- cophorans in the Mediterranean Sea: findings of Neopilina zografi (Dautzenberg & Fischer, 1896) off the Isles of Ca- praia, Gorgona, Corsica and Sardinia (Mollusca, Monopla- cophora). Bollettino Malacologico 23:107-118. CLARKE, A. H. & R. J. MENZIES. 1959. Neopilina (Vema) ewingi, a second living species of the Paleozoic class Mono- placophora. Science 129:1026-1027. DAUTZENBERG, P. & H. FISCHER. 1896. Dragages effectués par l’Hirondelle et par la Princesse-Alice: 1. Mollusques Gasteropodes. Memoires de la Societe Zoologique de France 10:139-234. FILATOVA, Z. A., N. G. VINOGRADOVA & N.I. MOSKALEV. 1975. Molluscs (Neopilina) in the Antarctic. Okeanologia 15:143- 145. LEMCHE, H. 1957. A new living deep-sea mollusk of the Cam- bro-Devonian class Monoplacophora. Nature 179:413-416. MARSHALL, B. A. 1990. Micropilina tangaroa, a new monopla- cophoran (Mollusca) from northern New Zealand. Nautilus 104:105-107. McLEan, J. H. 1979. A new monoplacophoran limpet from the continental shelf off southern California. Contributions in Science, Natural History Museum of Los Angeles County 307:1-19. MENZIES, R. J. 1968. New species of Neopilina of the Cambro- Devonian class Monoplacophora from the Milne-Edwards deep of the Peru-Chile Trench, R/V Anton Bruun. Marine Biological Association of India, Proceedings of the Sympo- sium on Mollusca 1:1-9. MENZIES, R. J. & W. LayTon. 1962. A new species of mono- placophoran mollusc Neopilina (Neopilina) veleronis from the slope of the Cedros Trench, Mexico. Annals and Magazine of Natural History 5(13):401-406. MOsKALEV, L. I., YA. I. STAROBOGATOV & Z. A. FILATOVA. 1983. New data on the abyssal Monoplacophora from the Pacific and South Atlantic oceans. Zoologeski Zhurnal 112: 981-996. PEEL, J. 1991. Functional morphology, evolution and system- atics of early Palaeozoic univalved molluscs. Gronlands Geo- logiske Undersogelse, Bulletin 161. 116 pp. Roxop, F. J. 1972. A new species of monoplacophoran from the abyssal North Pacific. The Veliger 15:91-95. ROSEWATER, J. 1970. Monoplacophora in the South Atlantic Ocean. Science 167:1485-1486. TAvIANI, M. 1990. A fossil Coenozoic Monoplacophora. Le- thaia 23:213-216. TEBBLE, N. 1967. A Neopilina from the Gulf of Aden. Nature 215:663-664. WaRrEN, A. 1988. Neopilina goesi, a new Caribbean monopla- cophoran dredged in 1869. Proceedings of the Biological Society of Washington 101:676-681. WarEN, A. 1989. New and little known Mollusca from Iceland. Sarsia 74:1-28. WaREN, A. & P. BouCHET. 1990. Laevipilina rolam, a new monoplacophoran from off southwestern Europe. Journal of Molluscan Studies 56:449-453. WENz, W. 1940. Ursprung und friihe Stammesgeschichte der Gastropode. Archiv fur Molluskenkunde 72:1-10. The Veliger 35(3):177-194 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 Comments on and Descriptions of Eulimid Gastropods from ‘Tropical West America by ANDERS WAREN Swedish Museum of Natural History, Box 50007, S-10405 Stockholm, Sweden Abstract. The author and date of the family name Eulimidae is corrected from H. & A. Adams, 1853, to Philippi, 1853, on the basis of priority. 7urveria pallida sp. nov. is described from the Gulf of California. It is ectoparasitic on the sand dollar Encope grandis L. Agassiz, 1841. Microeulima gen. nov. is described with the type species Alaba terebralis Carpenter, 1857 (Eulima proca de Folin, 1867 = Lewostraca schwengelae Bartsch, 1938 = Strombiformis hemphill: Bartsch, 1917 [new synonyms]). This species occurs from northern Mexico to Ecuador in shallow water. Strombiformis hemphilli Dall, 1883, from Florida, is placed in Microeulima. Scalenostoma babylonica Bartsch, 1917, is a junior synonym of Chemnitzia rangi de Folin, 1867, which is transferred from Scalenostoma to Niso Risso, 1826. Eulimostraca Bartsch, 1917 is discussed and E. macleani sp. nov. is described from Costa Rica. Strombiformis burraget Bartsch, 1917 (= Melanella panamensis Bartsch, 1917 [mew synonym]) and Levostraca linearis Carpenter, 1857, are transferred to Eulimostraca (all from western Mexico and Central America). Eulimetta pagoda gen. et sp. nov. is described from western Central America. Its host species is unknown. Sabinella shaskyi sp. nov. is described from western Central America. It lives in galls in the spines of the cidaroid sea urchin Eucidaris thourars: (Valenciennes, 1846). INTRODUCTION The family Eulimidae contains a large number of species, almost exclusively parasitic on echinoderms. The shell morphology is highly diverse and there are, in addition to species of “typical eulimid” appearance, also limpets and shell-less species in the family (WAREN, 1984b). To some extent the development of the shell depends on the animal’s sex or on the presence or absence of additional individuals of the same species, which in some species determine the sex of newly settled larvae (WAREN, 1984b). This complicates specific classification, but the problem can be overcome by a comparison of larval shells, which are identical within a species. This paper presents some of the results of an exami- nation of the West American eulimid collections in the Los Angeles County Museum of Natural History and the U.S. National Museum of Natural History, made some 15 years ago. During the intervening years I had hoped to obtain further material of the species discussed here, to be able to describe them in more detail and consolidate their systematic position. This has failed, however, except for the new species of Turveria and Sabinella. Nevertheless, I describe them here in an attempt to draw the attention of workers to them. When looking for eulimids, it is always useful to ex- amine echinoderms, which are usually their hosts. This can easily be done by selecting a common echinoderm species and, after a brief examination of each specimen, shaking them in a bucket with brackish water or seawater with some cleansing substance added (e.g., formalin or detergent). Afterwards the residue on the bottom of the bucket is searched for specimens. If any are found, they should be saved together with at least one specimen of the host. This procedure is quite profitable, and frequently yields undescribed species, in addition to invaluable in- formation about species already known. Abbreviations and Conventions Used in Text BMNH—Natural History Museum, London. LACM-—Los Angeles County Museum of Natural His- tory, Los Angeles, California. SMNH-—Swedish Museum of Natural History, Stock- holm. USFC— United States Fish Commission. USNM—National Museum of Natural History, Wash- ington, D.C. In the ennumerations of examined material, “‘shell’’ is Page 178 The Veliger, Vol. 35, No. 3 Explanation of Figures 1 to 10 Figures 1 and 2. Microeulima terebralis, holotype of Leiostraca schwengelae, USNM 127554, 3.12 mm. Figures 3 and 4. Microeulima terebralis, Costa Rica, LACM 72-46.16, front view (2.80 mm) and side view (3.28 mm). Figure 5. Microeulima terebralis, Costa Rica, LACM 72-42.20, front view 2.94 mm. A. Warén, 1992 used for empty shells, whereas “‘specimen” is used for shells containing dried or preserved soft parts. SYSTEMATICS Family EULIMIDAE Philippi, 1853 The family name has usually been ascribed to H. & A. ADAMS (1853:235) (WAREN, 1984b; PONDER & WAREN, 1988). The name was published by the Adams brothers in the eighth section of their monograph, which was issued December 1853. This is later than the publication by PHI- LIPPI (1853:194), which was published before May 1853, since Philippi’s book was reviewed by PETIT (1853) in the May issue of the Journal de Conchyliologie. The correct author of the family is thus Philippi, 1853. Turveria Berry, 1956 Turveria BERRY, 1956:356. Type species, by original des- ignation, Turveria encopendema Berry, 1956, Mexico, Baja California Sur, on sand dollars, Encope spp. (Scu- tellidae). Remarks: The genus was redescribed by WAREN (1991), and two species parasitic on sand dollars in the Gulf of California were included. The second species was wrongly identified as Turveria schwengelae Bartsch, 1938, a mistake which is corrected below. Little is known about the biology of the species of Turve- ria, except that they are regularly found on specimens of sand dollars belonging to the genus Encope L. Agassiz, 1841, a genus endemic to the southeastern United States, the Caribbean, Galapagos, and the American west coast from California to Ecuador (MORTENSEN, 1948). Turveria pallida Waren, sp. nov. (Figures 7, 8, 13) Turveria schwengelae: WAREN 1991:108, figs. 10A, B, 13F, G (not Bartsch, 1938). Type material: Holotype LACM 2425 (from LACM 55554) and 2 paratypes LACM 2426 (from LACM 55553), 2 paratypes SMNH 4122 (from LACM 55554; for locality data see “Material examined”’). Further paratypes, 12 specimens SMNH 4141 and numerous specimens in D. Shasky collection, from the same locality. Type locality: Mexico, Baja California Norte, sand flats at Isla Willard, Bahia San Luis Gonzaga, 29°57'N, 114°17’W, on Encope grandis, 3 specimens, LACM 55554. Page 179 Material examined: The type material and MExIco: Baja California Sur, Bahia Concepcion, 26°42'N, 11°55’W, 1 shell, no host (LACM 63-37.3); Baja California Norte, Isla Willard, Bahia San Luis Gonzaga, 29°57'N, 114°17'W, on Encope sp., 4 specimens (LACM 55553 [paratypes LACM 2426 and SMNH 4122]); Baja California Norte, Bahia San Luis Gonzaga, 29°57'N, 114°17'W, on Encope grandis, intertidal, 36 specimens (D. Shasky collection [12 specimens paratypes SMNH 4141)). Description: (Sex not known.) The shell (Figures 7, 8) is conically lanceolate, solid, smooth, transparent, with brownish markings along the suture and on the outer lip. The larval shell (Figure 13) consists of about 2.7 distinctly convex whorls and is smooth and colorless except for an occasional brownish tint on the initial whorl. The visible height of the larval shell is 260 um, and the total height of the shell of the larva is estimated to about 290 um. The holotype has 7.25 teleoconch whorls of slowly and uni- formly increasing diameter, sculptured by numerous dense and sharp incremental lines. The teleoconch of the holotype has nine incremental scars (1.2, 1.9, 2.8, 3.8, 4.3, 4.8, 5.5, 6.0, and 6.6 whorls from the outer lip) but as usual there is some individual variation in this character. The suture is shallow but distinct and makes a conspicuous bend downwards about 0.3 mm before the outer lip. The ap- erture is constricted in its upper part as a consequence of this and is pear-shaped. The outer lip is distinctly pro- socline, with a shallow sinus below the suture. The color pattern is not as bright as in Turveria encopendema, and consists of a brownish spiral band just below the periphery of the body whorl. This band is concealed under, or visible through, the subsutural zone on earlier whorls. There is also a large, roundedly triangular blotch at the lower part of the outer lip and one less-distinct, sometimes absent, similar spot just below the corner between the outer lip and the suture. Dimensions. Height of holotype 5.23 mm, maximum height 5.8 mm. Remarks: Turveria pallida differs from 7. encopendema by having a regularly conical spire, flatter whorls, and less vivid color pattern, and by being about % taller (shell height 4.91 + 0.27 mm [SD] among 23 mature specimens; compared with 4.18 + 0.27 mm among 26 specimens of T. encopendema). Eulima Risso, 1826 Eulima Risso, 1826:123, type species pending (WAREN, 1992), suggested to be Strombiformis glaber Da Costa, 1778, European. Figure 6. Microeulima sp. from the sea urchin Chaetodiadema granulatum, northwest of Koh-si-Chang, Thailand, 18 m depth, height 3 mm. Zoological Museum of the University of Copenhagen. Figures 7 and 8. Turveria pallida, paratypes, LACM 2426, side view 4.88 mm and front view 4.96 mm. Figures 9 and 10. Microeulima hemphilli, Florida, syntype, USNM 35983, height 3.1 mm. Page 180 The Veliger, Vol. 35, No. 3 A. Warén, 1992 Page 181 Remarks: The genus Eulima was described anatomically by WAREN (1984a). The hosts are known for two Euro- pean species, in both cases ophiuroids (WAREN, 1984a, and unpublished data). Almost all species that at present can be classified in Eulima still remain unknown with respect to anatomy and host choice. The species of Eulima s.s. have a tall (7-20 mm), slender shell, usually with a brownish color pattern, flat whorls, and a tall aperture with a rather straight profile of the outer lip. The animal is not very modified anatomically and retains a small stomach and a buccal mass with a ptenoglossate radula. The type species is a sand dweller, parasitic on ophiuroids, and has large epipodial folds, part- ly covering the base of the shell in order to facilitate the movements in the sand. The generic name Strombiformis Da Costa, 1778, has been used extensively in American literature, but the type species is an European pulmonate (WAREN, 1984b). BARTSCH (1917) revised the western American eulimids and classified several species in Strombiformis. Of the West American species the following seem to belong to Eulima 5.s., judging from shell characters: $. almo Bartsch, 1917; S. barthelowi Bartsch, 1917; S. californica Bartsch, 1917; Eulima fuscostrigata Carpenter, 1864; S. lapazana Bartsch, 1917; S. panamensis Bartsch, 1917; Eulima recta C. B. Adams, 1852; S. townsend: Bartsch, 1917; and E. varians Sowerby, 1834. BARTSCH (1926) also described Strombi- jormis hua, S. salsa, S. inca, and S. paria, all of which belong to Eulima s.s. Not all of these names represent different species, but I am not prepared to present a detailed syn- onymy. A large proportion of all the species of Eulimidae were originally described in Eulima, and are still placed there. Therefore I have frequently placed species provisionally in Eulima, instead of describing new genera for them. Microeulima Waren, gen. nov. Type species: Alaba terebralis Carpenter, 1857, western Central America. Diagnosis: Small eulimids, 2.5-5 mm high, with a slender, lanceolate shell of slowly increasing diameter, flat whorls, and brownish color, either as sutural and collabral bands or uniformly all over shell. Fine, sharp, indistinct axial lines present. Aperture constricted in its right corner, even- ly and broadly rounded at opposite end. Outer lip with distinct subsutural sinus. Parietal callus thick and abruptly demarcated. Etymology: Microeulima, from Greek mikros, meaning “small,” and Eulima, referring to the similarity to species of EFulima s.s. Remarks: I hesitated much before describing this new genus since there is neither a species of which soft parts have been examined, nor a named species for which the host is known. The group of species I classify here is, however, well demarcated, and there are numerous species in tropical areas, almost all of them undescribed. The host is known for a single species, but the shell is in such bad condition that it cannot be described or spe- cifically identified, although the aperture indicates that it belongs to Microeulima. This species was found by T. Mortensen (unpublished data) parasitizing the diadematid sea urchin Chaetodiadema granulatum Mortensen, 1903, in Thailand, northwest of Koh-si-Chang, in 18 m depth. The shell (Figure 6) has a broken apex and the surface is corroded, but the soft parts remain. In this genus belongs Strombiformis hemphilli (Dall, 1883) which was described from Cedar Keys, Levy County, Flor- ida, and later (LYONS 1989:16) recorded from several lo- calities at Hutchinson Island, off Indian River, St. Lucie County, eastern Florida. I figure one of the two syntypes (Figures 9, 10). The shell of this species is uniformly chestnut brown. The genus Euli:mostraca (see below) bears some resem- blance to Microeulima, but the shape of the shell is reg- ularly conical with a larger aperture and it lacks the axial lines typical for Microeulima. Microeulima also resembles Turveria, but species of that genus lack the strongly developed parietal callus. Explanation of Figures 11 to 19 Figure 11. Eulimostraca macleani, paratype, LACM 2371, height of larval shell 390 um. Figure 12. Eulimostraca galapagensis, paratype, USNM 251281, height of larval shell 400 um. Figure 13. Turveria pallida, paratype, LACM 2426, height of larval shell 260 um. Figure 14. Eulimetta pagoda, holotype, LACM 2372, height of larval shell 210 um. Figure 15. Niso interrupta, Mexico, near Guaymas, Bahia Bacochibampo, 9-18 m, LACM 55558. Scale line 0.25 mm. Figure 16. Microeulima terebralis, Costa Rica, LACM 72-42.20, height of larval shell 490 um. Figure 17. Niso rangi, Costa Rica, LACM 72-52.22, height of larval shell 370 um. Figure 18. Niso interrupta, LACM 55558, for collection data see Figure 15, height of larval shell 490 um. Left arrow indicates a teleoconch growth line, right arrow a protoconch growth line. Figure 19. Niso hipolitensis, Mexico. Baja California Sur, Punta Palmilla, intertidal, LACM 66-11.5, height of larval shell 390 um. Page 182 Eulimostraca bartschi Strong & Hertlein, 1937, known from two localities in western Mexico (HERTZ & HERTZ, 1982), probably belongs to Microeulima, but I have not examined any specimens. Microeulima terebralis (Carpenter, 1857) (Figures 1-5, 16) Alaba terebralis CARPENTER, 1857:367. Leiostraca sp. ind. (b): CARPENTER 1857:440. Eulima proca DE FOLIN, 1867:62, pl. 6, fig. 3 (new synonym). Strombiformis hemphilli BARTSCH, 1917:344, pl. 47, fig. 4 (not Dall, 1889) (new synonym). Levostraca schwengelae BARTSCH, 1938:34 (replacement name for Strombiformis hemphilli Bartsch, 1917). Alaba terebralis: BRANN 1966:pl. 40, fig. 427. Levostraca sp. ind. (b): BRANN 1966:pl. 40, fig. 553. Eulima? terebralis: KEEN 1968:424, text fig. 108. Type materials: Alaba terebralis, holotype BMNH 1854.6.4.427; E. proca, 1 syntype BMNH 1868.2.17.13; S. hemphilli, holotype USNM 127554 (Figures 1, 2). Type localities: Alaba terebralis, W Mexico, Sinaba, Ma- zatlan, “‘off Spondylus” (living on?); E. proca, Panama, Achipelago de las Pearlas; S. hemphilli, Baja California Sur, shell drift at Punta Abreojos. Materials examined: The type material and Mexico: Pacific side of Baja California Norte, Isla Cedros, 1.6 km (1 mile) N of Cedros Village, 5-8 m depth, 28°00'N, 115°10'W, 2 shells (LACM 67-65.10); Pacific side of Baja California Sur, Cabo Thurloe, 27°37.5°N, 114°14.9'W, 15-20 m depth, 4 shells (LACM 71-170.9); Pacific side of Baja California Sur, Punta San Pablo, 27°12.5'N, 114°28.9'W, 20-30 m, 1 shell (LACM 71-178.14); Pacific side of Baja California Sur, Isla Asuncion, E Anchorage, 27°06'N, 114°17'W, 8-23 m, 3 shells (LACM 67-66.13); Golfo de California, Baja California Sur, Bahia Concep- cion, 26°42'N, 111°55'W, shallow depth, 17 shells (LACM 63-37.4); Pacific side of Baja California Sur, Bahia Mag- dalena, Man of War Cove, 24°37.5'N, 112°7.5'W, 0-12 m depth, 1 shell (LACM 71-183.10); Sinaloa, Mazatlan, N of Olas Altas Lighthouse, 23°12'N, 106°27’W, intertid- al, 1 shell (LACM 46-9.1); Sinaloa, vicinity of Mazatlan, 23°11'N, 106°26'W, 0-6 m, 4 shells (LACM 63-11.15); Nayarit, 72 km (45 miles) NW of San Blas, Isla Isabela, 21°51'N, 105°55'W, 10 m depth, 1 shell (LACM 67-9.1). Costa Rica: Puntarenas Province, Islas Tortugas, 1 km W of Isla Alcatraz, 09°47'N, 84°53.5'W, 2-8 m depth, 9 shells (LACM 72-46.16); Puntarenas Province, Bahia Ballena, 2 km (1.5 miles) E of Punta Ballena, 09°44.3'N, 84°33.8’W, 3-16 m depth, 8 shells (LACM 72-42.20); Puntarenas Province, Bahia Ballena, 09°44'N, 84°33'W, 1-13 m depth, 65 specimens (D. Shasky collection); Pun- tarenas Province, Bahia Herradura, 09°38.8'N, 84°41'W, 10-12 m depth, 2 shells (LACM 72-52.21); Puntarenas Province, Bahia Herradura, 09°38.0'’N, 84°40.5'W, 23 m depth, 4 shells (LACM 72-53.5); Puntarenas Province, Islets off Punta Quepos, 09°22.7'N, 84°09.7'W, 13-25 m depth, 9 shells (LACM 72-58.21). PANAMA: Canal Zone, The Veliger, Vol. 35, No. 3 off sandspit leading to Isla Venado, 08°53'N, 79°36'W, intertidal, 1 shell (LACM 75-54.12); Fort Amador, Isla Perico, 08°51'N, 79°35'W, intertidal, 1 shell (D. Shasky collection); Isla Vendao, intertidal, 7 shells (D. Shasky collection); Bahia de Panama, Isla Bona, Isla Otoque, 08°36’N, 79°39'W, 10-27 m depth, 1 shell (LACM 65- 21.19); Bahia de Panama, Isla Taboga, 08°35'N, 79°30'’W, 2-5 m depth, 2 shells (LACM 62-25.17); Archipelago de las Perlas, Isla Buyarena, 08.5°N, 79°W, intertidal, 1 shell (D. Shasky collection); Archipelago de las Perlas, Isla Pe- dro Gonzales, 08.5°N, 79°W, intertidal, 1 specimen (D. Shasky collection). ECUADOR: Guayas Province, Santa Elena Peninsula, NW side of Punta Ancon, 02°19.5’S, 80°54.0'W, intertidal, 1 shell (LACM 70-11.6); Guayas Province, Santa Elena Peninsula, W side of Punta Ancon, 02°19’S, 80°54'W,, intertial, 6 shells (D. Shasky collection). Distribution: East Pacific, western Mexico to Ecuador, intertidal to 25 m depth. Remarks: The holotype of Leiostraca schwengelae (Figures 1, 2) has lost most of the characteristic, tall-spired pro- toconch. Examination of the holotype, available only after proofreading the revision of Hypermastus and Turveria (WaREN, 1991), made me realize the mistake and made the synonymy with Alaba terebralis obvious. Alaba terebralis was based on only a fragment with two teleoconch whorls and two whorls of the larval shell left. The syntype of Eulima proca is in good condition. The height of the larval shell is about 500 wm and it consists of slightly more than three whorls. This is almost twice the height of the larval shell of the otherwise some- what similar species of Turveria. These, however, have a larger teleoconch (4-5 mm), and are broader with a rel- atively higher aperture. Like many of the eulimids with a color pattern, this species has regularly appearing thin and sharp collabral lines on the whorls. Scalenostoma Deshayes, 1863 Scalenostoma DESHAYES, 1863:58. Type species, S. carinata Deshayes, 1863, by monotypy, La Reunion, Indian Ocean. Remarks: Species of Scalenostoma inhabit cavities in living specimens of hermatypic corals in shallow water in tropical regions (WAREN, 1980). Presumably they use their long proboscis to parasitize the surrounding polyps of the coral. They are likely to belong to the Eulimidae (but no well preserved specimens have been available for anatomical examination). If this assumption is correct, they differ from most eulimids in not parasitizing echinoderms. The as- sociation with cnidarians indicates a possibility that they are related to the Epitonidae, but the apical whorls (in- cluding the larval shell) are so similar to eulimids that they have frequently been identified as species of that fam- ily. The species are highly variable, large, with a transpar- ent, colorless shell up to 30 mm high. The first 8-10 whorls A. Warén, 1992 look like a specimen belonging to Vitreolina Monterosato, 1884 (Eulimidae), 2-5 mm high, with flat whorls and twisted spire. The growth pattern then suddenly changes and the whorls become fatter and more irregularly coiled. Scalenostoma subulata (Broderip, 1832) has been re- ported from Isla Cascara, Cocos Island, Costa Rica by SHASKY (1983b), who found six primary females, one secondary female, and 10 males in cavities in a piece of living coral from 25 m depth. I take this occasion to figure some of them (Figures 55-59). Two species have been described from western Central America, Chemnitzia rangi de Folin, 1867, and Scalenos- toma babylonica Bartsch, 1917, which usually have been classified in Scalenostoma. They do not conform with the type species, however, except in frequently having the low- er whorls sharply keeled, which by itself is not a diagnostic feature. They are here transferred to Niso Risso, 1826, on the basis of shell characters specified below. Niso Risso, 1826 Niso Risso, 1826:218. Type species, Niso eburnea Risso, 1826, by monotypy, Pleistocene of southern Europe. Remarks: One undescribed species of Niso, from New Caledonia, is known to parasitize a starfish (WAREN, 1984b) and work is underway to describe the anatomy of that species. That species and the type species, NV. eburnea, are very similar to Niso interrupta Sowerby, 1834, a western, Central American species that is used below to exemplify the characters of the genus. EMERSON (1965) revised the West American species and MCcLEAN (1970) described Niso emerson: from Panama. HERTZ & HERTZ (1982) described Eulimostraca attilioi from off La Jolla, California, and commented on the sim- ilarity to Niso, but placed the species in Eulimostraca be- cause it has almost no umbilicus. I consider this character less important than the great similarities in other details, including microsculpture, larval shell, size, and color pat- tern, and transfer it to Niso. HERTZ & HERTZ (1982) also placed N. hipolitensis in Eulimostraca, because of the lack of a well developed umbilicus. I agree with other authors (BARTSCH, 1917; EMERSON, 1965) that hipolitensis shows more affinity to Niso and suggest that it is kept there. Niso rangi (de Folin, 1867) (Figures 17, 20-22, 25-27, 30, 31) Chemnitzia rangi DE FOLIN, 1967:61, pl. VI fig. 1. Scalenostoma babylonica BARTSCH, 1917:338, pl. 45 fig. 2. Type materials: Chemnitza rangi, lost, not in BMNH, MNHN, or Biarritz (Kisco, 1959; P. Bouchet, personal communication); Scalenostoma babylonica, holotype and 1 paratype, USNM 127542 (Figure 20). Type localities: C. rangi, Bahia de Panama, Archipelago de las Pearlas; S$. babylonica, Baja California Sur, Punta San Hipolito. Page 183 Material examined: The type material and MeExIco: Pa- cific side of Baja California Sur, Punta San Pablo An- chorage, 21-24 m, 1 shell (LACM 71-177.5). EL SALva- por: La Union Province, Golfo de Fonseca, Isla Zacatillo, 13°18'N, 87°46'W, 2 m, 2 shells (LACM 73-57.1). COSTA Rica: Guanancaste Province, N of Bahia Potrero, Punta Penca, 10°29.3’N, 85°48.9'W, 8-13 m, 2 shells (LACM 72-38.7); Puntarenas Province, 1 km W of Isla Alcatraz, Isla Trotugas, 09°47.0'N, 84°53.5’W, 1.5-8 m, 3 shells (LACM 72-46.17); Puntarenas Province, Bahia Ballena, 2.4 km E of Punta Ballena, 09°44.3’N, 84°33.8’W, 3-16 m, 1 shell (LACM 72-42.21); Puntarenas Province, W side of Bahia Ballena, 09°44'N, 84°33’W, 6-10 m, 2 shells (D. Shasky collection); Puntarenas Province, Bahia Bal- lena, 09°44'N, 84°33’W, 13-15 m, 2 shells (D. Shasky collection); Puntarenas Province, off Bahia Herradura, 09°38.9'N, 84°40.9'W, 6 m, 1 shell (LACM 72-54.13); Puntarenas Province, Bahia Herradura, reef at N end of bay, 09°38.8'N, 84°40.9'W, 10-18 m, 10 shells (LACM 72-52.22); Puntarenas Province, Bahia Herradura, 09°38.0'N, 84°40.5'W, 23 m, 6 shells (LACM 72-53.6); Puntarenas Province, anchorage inside small islet, 1.5 km S of Punta Quepos, 09°22.7'N, 84°09.7'W, 23 m, 3 shells (LACM. 72-57.5); Puntarenas Province, small islets off Quepos, 09°22.2'N, 84°09.3’W, 25 m, 2 shells (LACM 72-59.5); Puntarenas Province, N side of Isla del Cano, 08°43.3'N, 83°53.1'W, 8-13 m, 3 shells (LACM 72-63.28); Puntarenas Province, 2 km NW of Rincon de Osa, head of Gulfo de Dulce, 08°43.3’N, 83°28.5'W, 2-16 m, 1 shell (LACM 72-71.14); Isla del Coco, 05°33'N, 87°00'W, 1 shell (coll. K. Kaiser). ECUADOR: Guayas Province, N side of Santa Elena Peninsula, E of Salinas, 02°11.5'N, 80°56.5'W, 10 m, 2 shells (LACM 66-114.4); Manabi Province, N side of Isla la Plata, 01°19'S, 81°905’W, 12- 30 m, 2 shells (D. Shasky collection); Manabi Province, Isla Salanga, 01°35'S, 79°50’W, 10-15 m, 1 shell (D. Shas- ky collection). Distribution: Outer coast of Baja Californa Sur from about 27°N to Ecuador, also Cocos Island, in 3-30 m depth. Remarks: Niso rangi is so far known only from empty shells, which makes classification more difficult. For eu- limids, however, the shell is unusually rich in characters, which allows the determination of some relationships. Figures 20-22, 25-27, 30, and 31 show the variation of the shell and that the development of the peripheral keel varies with size and the individual. In addition to what is shown by these figures, it should be mentioned that the shell is dark and dull reddish or yellowish brown. BARTSCH (1917) had access to only two shells when proposing Scalenostoma babylonica and DE FOLIN’s (1867) drawing is erroneous since it shows all whorls as keeled, although the shell was described as having only the last four whorls keeled. That is evidently the reason why BARTSCH (1917) described N. babylonica. Examination of the larval shell (Figure 17) shows that it consists of about 3.5 whorls and has a distinct sculpture Page 184 The Veliger, Vol. 35, No. 3 Explanation of Figures 20 to 27 Figure 20. Niso rangi, holotype of Scalenostoma babylonica, USNM 127542, height 3.0 mm. Figure 21. Niso rangi, Costa Rica, LACM 72-52.22, 2.92 mm. Figure 22. Niso rangi, Costa Rica, LACM 72-59.5, 3.54 mm. A. Warén, 1992 of very fine collabral lines and a height of almost 400 wm. The whole teleoconch is covered by equally sharp but straighter axial lines. Protoconch sculpture, as well as the color, agrees closely with these features in species of Niso. I have exemplified that genus with a young specimen of N. interrupta (Sow- erby, 1834), a Central American species typical for its genus (shell, Figure 28; larval shell, Figure 18; sculpture, Figure 15) and N. hipolitensis Bartsch, 1917 (shell, Figures 23, 24, 29; larval shell, Figure 19). Both Niso hipolitensis and N. rangi are unusual among the species of Niso in their small size, 3-4 mm shell height, whereas most species of the genus have a shell that is 10- 30 mm high, occasionally even higher. Nevertheless I feel satisfied with this systematic position, although it remains to be checked by examination of the soft parts, when such become available. Niso hipolitensis can be distinguished from young spec- imens of N. rangi by having a blunter protoconch and no trace of an umbilicus. A specimen from Isla Taboga, Pan- ama (in the collection of D. Shasky) indicates a larger size than that given by BARTSCH (1917) namely 4 mm. Had it not been that N. hipolitensis was a much more rare species compared with N. rangi, I would not have excluded the possibility of the two names having been based on the male and female, respectively, of the same species. Eulimostraca Bartsch, 1917 Eulimostraca BARTSCH, 1917:333. Type species, by original designation, Eulimostraca galapagensis Bartsch, 1917, Galapagos (Figures 12, 33, 35). Remarks: BARTSCH (1917) placed a single species in Eu- lumostraca when he introduced this generic name, but the genus is represented by several, mostly undescribed species in the Caribbean and western America. Nothing is known about the identity of the hosts of these species. The species of Eulimostraca resemble Microeulima, but species of Eulimostraca have a proportionally larger and expanded aperture, giving the shell a regularly conical appearance. The aperture is, however, less expanded and more similar to Microeulima in young specimens, which indicates that they are related. Also, the larval shells are similar. Eulimostraca macleant, described below, differs from the typical appearance, but has an aperture very similar to that of E. galapagensis and is provisionally included here. Lewostraca linearis CARPENTER, 1857 (p. 440) (from Ma- Page 185 zatlan, Sinaloa, Mexico) was based on a small (1.84 x 0.56 mm) specimen, similar to Eulimostraca galapagensis. The holotype has no trace of color pattern, but is otherwise well preserved and differs from FE. galapagensis by having perfectly flat whorls, by having a slightly more cylindrical shell, and correspondingly by having a smaller aperture. It was figured by BARTSCH (1917:pl. 36, fig. 4), BRANN (1966:pl. 48, fig. 554), and KEEN (1968:text fig. 32, pos- sibly also 1971:fig. 727), and the holotype is in BMNH, Mazatlan collection No. 2025. Eulimostraca attilioi Hertz & Hertz, 1982, was discussed under Niso and transferred to that genus. Eulimostraca galapagensis Bartsch, 1917 (Figures 12, 33, 35) Eulimostraca galapagensis BARTSCH, 1917:333, pl. 43, fig. 1. Eulimostraca galapagensis: SHASKY, 1983a:29. Eulimostraca galapagensis: HERTZ & HERTZ, 1982:74. Type material: Holotype and 7 paratypes, USNM 251281. Type locality: “Galapagos Islands, 72 m.” Material examined: The types and ECUADOR: Galapagos Islands, Isla Isabela, off Tagus Cove, 00°17’S, 91°23’W, 27 m, 1 shell (LACM 34-290.1); Manabi Province, Isla La Plata, 01°16’S, 81°06’W, 30-40 m, 1 shell (D. Shasky collection). Distribution: Only known from the material examined, Galapagos Islands, 27-72 m depth; also Corinto and Ma- nabi Province, Ecuador. Remarks: Eulimostraca galapagensis can be recognized by its distinctly conical shape and evenly yellowish color in fresh specimens. The periphery is encircled by a darker band. I have verified SHASKy’s (1983a) record by direct com- parison with the paratypes. Eulimostraca burragei (Bartsch, 1917) Strombiformis burrage: BARTSCH, 1917:345, pl. 47 fig. 5. Melanella panamensis BARTSCH, 1917:311, pl. 36 fig. 1. Type materials: Melanella panamensis, holotype USNM 251312; Strombiformis burragei, holotype USNM 267582. Type localities: Melanella panamensis, Bahia de Panama, 110 m; Strombiformis burrager, Bahia Concepcion, Golfo de California, 5 m. Figures 23 and 24. Niso hipolitensis, Mexico. Baja California Sur, Punta Palmilla, intertidal, LACM 66-11.5, 2.3 mm. Figure 25. Niso rangi, Costa Rica, LACM 72-38.1, 3.20 mm. Figure 26. Niso rangi, Costa Rica, LACM 72-52.22, 2.68 mm. Figure 27. Niso rangi, Costa Rica, LACM 72-63.28, 3.7 mm. Page 186 The Veliger, Vol. 35, No. Explanation of Figures 28 to 35 Figure 28. Niso interrupta, LACM 55558, same specimen and data as Figure 15, 4.9 mm. Figure 29. Niso hipolitensis, Galapagos Islands, Isla Isabela, off Tagus Cove, 144 m, LACM 34-290.2, 2.4 mm. A. Warén, 1992 Distribution: Only known from the type specimens, from Golfo de California and Bahia de Panama, in 5-110 m. Remarks: Examination of the two holotypes of the two names cited in the synonymy did not reveal any differences between them, except for one being more worn. The ap- erture is less expanded and more ovate than in Eulimostraca galapagensis, but this may be because the specimens are not fully mature. If this is the case, I believe the names to be synonyms of EF. galapagensis. This would not be the first time Bartsch described the same species in three different genera. As first reviser, I prefer to use the name burragez since panamensis invites confusion wth Strombiformis panamensis Bartsch, 1917, a distinct species belonging. to Eulima. Eulimostraca macleani Waren, sp. nov. (Figures 11, 32, 34, 40) Type material: Holotype LACM 2370 and one paratype LACM 2371. Type locality: Costa Rica, Puntarenas Province, Bahia Herradura, 09°38.0'N, 84°40.5’W, 23 m, LACM 72-53. Material examined: Only known from the type lot. Distribution: Only known from the type locality, Costa Rica, in 23 m depth. Etymology: Named after Dr. James H. McLean, LACM, who always has been very helpful during my visits at the museum. Description: The shell (Figures 32, 34) is very tall and slender, cylindrical, colorless(?), and transparent, except for a faint brown line along the outer lip and a more distinct blotch on the lower part of the columella. The larval shell (Figure 11) is pointed, with 2.5 perfectly smooth and evenly convex whorls, distinctly demarcated from the teleoconch. The holotype has 11.6 teleoconch whorls, of which the most apical 2 whorls are almost as convex as those of the larval shell and of more rapidly increasing diameter than later whorls. After these, the whorls grad- ually become flatter and the shell more cylindrical. The body whorl is distinctly angulated at the level of the suture. Starting on the second teleoconch whorl, the surface is covered by a fine spiral striation (Figure 40), barely visible under a good stereomicroscope, and only in patches where incident light is reflected. There are also scattered, occa- sionally sharp and distinct, usually less distinct, incre- mental lines. In addition, there are several incremental scars on the apical 3 whorls, then 4 not very distinct in- Page 187 cremental scars—5.9, 7.8, 10.1, and 10.6 whorls from the larval shell; but in the paratype the positions are different. The aperture is rather broad, with a distinct subsutural sinus in the outer lip. The parietal wall has a thick callus (inner lip). Dimensions. Height of the holotype (largest specimen) 6.97 mm. Remarks: Eulimostraca macleani is probably the most cy- lindrical eulimid known. I am not aware of any similar species from western America. Eulimetta Waren, gen. nov. Type species: Eulimetta pagoda sp. nov. Diagnosis: Very small (ca. 2 mm) eulimid, with a brown- ish shell and a very strong, periodically expanding pe- ripheral keel on the lower whorls. Etymology: Diminutive of Eulima. Remarks: The development of the keels varies. In one shell it starts almost immediately after the larval shell (Figure 37), but in most specimens it seems not to reach full development. This may be because I have failed to recognize that more than one species is involved, or the cause may be that environmental factors direct the devel- opment, as is common in eulimids (see Introduction). Eulimetta must be rather closely related to Eulimostraca and Microeulima, judging from the shape of the larval shell and the aperture, but I prefer to make a new genus for this strangely shaped species. No other eulimid has a similar expansion of the peripheral keel. Eulimetta pagoda Waren, sp. nov. (Figures 14, 36-39, 41) Type material: Holotype LACM 2372 and two paratypes and LACM 2373. Type locality: Mexico, Jalisco, Bahia Cuastocemate, 4.8 km (3 miles) NW of Barra de Navidad, 19°13.8'N, 104°44.9'W, 18-36 m, LACM 68-45. Material examined: The type material and MExIco: Baja California Norte, 16 km S of Bahia San Luis Gonzaga, Punta Final, 29°48’N, 114°17'W, 36 m, 1 shell (LACM 61-6.3); Baja California Sur, off S end of Isla Espiritu Santo, 24°23’N, 110°20’W, 44 m, 1 shell (LACM 36- 140.1); ca. 64 km (40 miles) S of Mazatlan, 22.5°N, 106.5°W, 30-35 m, 2 shells, from shrimp trawler (LACM 60-20.1); Jalisco, E of Punta Mita and off La Cruz, Bahia Figures 30 and 31. Niso rangi, Costa Rica, LACM 72-52.22, 3.8 mm and 3.0 mm. Figure 32. Eulimostraca macleani, holotype, LACM 2370, 6.97 mm. Figures 33 and 35. Eulimostraca galapagensis, syntype, USNM 251281, 3.5 mm. Figure 34. Eulimostraca macleani, paratype, LACM 2371, 6.0 mm. Page 188 The Veliger, Vol. 35, No. 3 Explanation of Figures 36 to 41 Figure 36. Eulimetta pagoda, holotype, 2.0 mm. Figure 37. Eulimetta pagoda, Guatemala, LACM 38-25.10, 1.12 mm. Figure 38. Eulimetta pagoda, Costa Rica, LACM 72-43.4, 1.68 mm. Figure 39. Eulimetta pagoda, paratype, 2.2 mm. Figure 40. Eulimostraca macleani, paratype. Scale line 0.05 mm. Figure 41. Eulimetta pagoda, paratype, keel. Scale line 0.01 mm. Bandera, 20°45'N, 105°25'W, 3-5 m, 4 shells (LACM 65- GUATEMALA: off Punta San Jose, 13°46'’N, 91°14'W, 36 16.47); Jalisco, Bahia Tenacatita, 19°17’N, 104°49'W, 1 m, 1 shell (LACM 38-25.10). Costa Rica: Puntarenas shell (LACM 33-138.1); Jalisco, Bahia Cuastecomate, Province, off Bahia Balena, 10°44.1'N, 84°59.5'W, 12 m, 19°13.8'N, 104°44.9'W, 18-36 m, 2 shells (LACM 2373). 1 shell (LACM 72-43.4). A. Warén, 1992 Distribution: From Baja California Norte, 29°N, to Costa Rica, in 3-44 m depth. Etymology: Named after the pagoda-like apparence of the lower whorls. Description: The shell (Figures 36-39) is very small, very slender, chestnut brown, fairly solid, with the angulated periphery of the whorls often developed into a winglike keel. The larval shell (Figure 14), which is 210 um high, has about 3.5 evenly rounded whorls with indistinct but sharp growth lines. The holotype has 5.5 teleoconch whorls, which except for the regularly appearing growth scars 0.5, 1.0, 1.5, and 2.0 whorls from the outer lip, have no sculp- ture visible with a stereomicroscope. Under SEM there are a few impressed spiral lines paralleling the peripheral keel (Figure 41) and the whole surface is covered by very small and shallow pits. Directly after the rather indistnctly demarcated protoconch starts a rounded keel, formed by a distinct bend in the profile of the whorl. This continues mostly unchanged for 1-4 whorls, after which it becomes more pronounced and keel-like, and transforms into a raised rib. Shortly before the outer lip it becomes lower again, a process that is repeated at each incremental scar; after an incremental scar the keel rapidly reaches maximum de- velopment. As a consequence of this growth pattern the shell looks strongly flattened, when observed from the side. The aperture is pear-shaped, with a small internal dent corresponding to the keel. The parietal callus is thick. Dimensions. Height of holotype 2.02 mm, maximum diameter of body whorl 0.81 mm, minimum diameter of body whorl 0.54 mm; maximum height 2.32 mm. Remarks: I am not aware of any species that can be con- fused with specimens with a developed keel; those with a poorly developed keel may possibly be confused with var- ious species of Microeulima, unless care is taken to look for the fine and sharp (but often indistinct) axial lines of those species. Sabinella Monterosato, 1890 Sabinella MONTEROSATO, 1890:160. Type species, Eulima piriformis Monterosato, 1875 (not Brugnone, 1873) = Eulima bonifaciae Nordsieck, 1974, Mediterranean, par- asitic on test of Cidaris cidaris (Linnaeus, 1758). Remarks: Several species of Sabinella are known to par- asitize cidaroid sea urchins (WAREN, 1984b; BOUCHET & WAREN, 1986; WAREN & MOOLENBEEK, 1989; WAREN & MIFsuD, 1990) in galls in the spines or attached to the test. The following species were placed in Sabinella by BaRTSCH (1917): —Sabinella chathamensis Bartsch, 1917. This species does not belong to Sabinella, but is related to the Caribbean species “Eulima” hians Watson, 1883. It is probably better to provisionally place chathamensis in Eulima, a genus com- prising a very heterogenous mixture of eulimids, and keep Page 189 Sabinella as a monophyletic genus for this small group of cidaroid parasites. —Sabinella baker: Bartsch, 1917. This species probably is an eulimid despite having a rather fragile and irregular shell. I have examined a specimen with dried soft parts, and it has a ptenoglossate radula, similar to species of Eulima. It can provisionally be placed in Eulima. —Sabinella opalina (de Folin, 1867). There is a possible syntype in the Museum of Comparative Zoology, Harvard University, No. 288749, which belongs in the genus Mela- nella. —Sabinella ptilocrinicola Bartsch, 1907. WAREN (1984b) placed this species in Crinolamia Bouchet & Waren, 1979. It lives on deep-sea crinoids. —Sabinella meridionalis Bartsch, 1917, can provisionally be placed in Eulima. Sabinella shaskyi Waren, sp. nov. (Figures 42-46, 48-52) Type material: Holotype LACM 2374, 7 paratypes LACM 2375, 10 paratypes in D. Shasky collection, 12 paratypes SMNH 4378. All paratypes from Baja Cali- fornia Sur, El Pulmo Reef (see below). Type locality: Mexico, Jalisco, 4.8 km (3 miles) NW of Barra de Navidad, 19°13.7'N, 104°44.8'W, 18-36 m, ina gall in a spine of Ewcidaris thourarsi (Valenciennes, 1846), LACM 68-45. Material examined (host Eucidaris thourars1): MEXICO: Baja California Sur, El Pulmo Reef, 1.5-3 m, several host specimens, 11 males, 14 females, 6 specimens left in galls (D. Shasky collection, paratypes in LACM and SMNH); Baja California Sur, off N end of Isla San Pedro Nolasco, 27°59'N, 111°24’W, 10 m, 1 shell (LACM 73-133); Baja California Sur, SE of Isla San Pedro Nolasco, 27°59'N, 111°24’W, 17-23 m, 1 shell (D. Shasky collection); Baja California Sur, Mulege, rocky point in front of La Sereni- dad, 27°00'N, 111°58'W, 3 m, 2 empty galls (+ 1 specimen of Nanobalcis sp. nov.) (D. Shasky collection); Jalisco, Cuastecomate, 19°13.8’N, 104°44.9'W, 5-6 m, 1 apical spine with a gall with 1 male, 2 females, and egg capsules (D. Shasky collection); Nayarit, Isla Maria Madre, 1.6 km (1 mile) S of Puerto Ballena, 21.6°N, 106.5°W, 3-5 m, 3 assumed males from 1 host (D. Shasky collection); Nayrit, Banderas Bay, Islas Tres Mariaas, 20°45’N, 105°30’W, 5-10 m, on host, 1 apical spine with a small, healed gall, 1 dorsal spine with a gall with male, female, and egg capsules (LACM 65-14.7). Costa Rica: Isla del Coco, Baja Alcyone, 05°33'N, 87°00'W, 32-35 m, 2 hosts, each with 2 galls, each with male and female; 1 host with 2 specimens, no gali (D. Shasky collection); Isla del Coco, 05°33'N, 87°00'W, 21 m, 1 host with 1 healed gall, 1 gall with 2 males, 1 female (D. Shasky collection); Isla del Coco, Bahia Chatham, Punta Ulloa, 05°33'N, 87°00’W, 14-20 m, 1 young specimen 1.5 mm, no gall (D. Shasky Page 190 The Veliger, Vol. 35, No. 3 Explanation of Figures 42 to 47 Figure 42. Sabinella shaskyi, Costa Rica, Baja Alcyone, length of spine 6 mm. Figure 43. Sabinella shaskyi, Mexico, Coastocomate, diameter of spine 5 mm. Figure 44. Sabinella shaskyi, Cocos Island, 21 m, length of spine 6 mm. This spine is regenerating, with three new points protruding from the rim of the gall. Figure 45. Sabinella shaskyi, apical view of Figure 43, shell 2.5 mm. Figure 46. Sabinella shaskyi, Mexico, Coastocomate, height of larval shell 450 um. Figure 47. Sabinella troglodytes, from Eucidaris tribuloides, Florida, off Cedar Key, 28°47.5'N, 84°37'W, 43 m, USFC station 2407, height of larval shell 430 um. A. Waren, 1992 Page 191 Explanation of Figures 48 to 54 Figure 48. Sabinella shaskyi, Mexico, Cuastecomate, D. Shasky collection, 3.6 mm. Figures 49 and 50. Sabinella shaskyi, Mexico, El Pulmo Reef, D. Shasky collection, 3.0 mm and 2.9 mm, respectively. Figures 51 and 52. Sabinella shaskyi, holotype, LACM 2379, 3.0 mm. Figures 53 and 54. Sabinella troglodytes, USFC station 2407, for data see Figure 47, 2.2 and 3.1 mm, respectively. collection); Isla del Coco, Isla Manuelita, 05°33’'N, 87°00'W, 13-17 m, 1 shell (D. Shasky collection); Isla del Coco, Isla Manuelita, 05°33’N, 87°00'W, 100-105 m, 1 shell (D. Shasky collection). ECUADOR: Galapagos Islands, Isla Marchena, 0°18’N, 90°30’W, 6 m, 1 shell (LACM 34-285.1); Galapagos Islands, Isla Isabela, Tagus Cove, 0°16'S, 91°23’W, 15 m, 1 shell (LACM 33-165.1); Ga- lapagos Islands, Isla San Salvador, James Bay, 0°12’S, 90°52'W, 45 m, 1 shell (LACM 34-273.1); Manabi Prov- ince, N side of Isla La Plata, 01.1°S, 81.1°W, 10-27 m, in siftings, 2 small males, 1.6-1.7 mm (D. Shasky collec- tion). Distribution: From Ecuador, the Galapagos Islands, and Cocos Island north to Gulf of California, in about 5-45 m depth. Etymology: Named after Dr. D. R. Shasky, who con- tributed much of the material of this study. Page 192 The Veliger, Vol. 35, No. 3 Explanation of Figures 55 to 59 Figures 55-59. Scalenostoma subulata. Costa Rica, Cocos Island, D. Shasky collection. Figure 55. Assumed primary female, lacking penis, height of shell 10.4 mm. Figure 56. Assumed primary female, with penis, height of shell 10.1 mm. Figure 57. Assumed secondary female, with penis, height of shell 11.3 mm. Figures 58 and 59. Assumed male, with penis, height of shell 8.0 mm. Figure 59. Apex enlarged, showing labial scar marking transition from normal eulimid to “‘Scalenostoma shape.” The soft parts were too decayed to check for the presence of a pallial oviduct. Description: Female. The shell (Figures 48-52) is small, grayish white, semitransparent, conical, somewhat irreg- ularly coiled, usually with a slightly twisted or curved spire. The larval shell (Figure 46) is multispiral and dis- tinctly demarcated from the teleoconch. The height is 450 um. It has about 3.3 perfectly smooth whorls of rapidly increasing diameter and protoconch 1 is hardly discernible. The teloconch of the holotype has 4.5 whorls with distinct incremental scars 0.7, 1.4, 2.3, 3.0, 3.5, and 3.9 whorls from the larval shell, but the distance between them is subject to much individual variation. The whorls are quite convex with a shallow but distinct suture. The subsutural zone is not very distinct and occupies 4,-'% of the height of the whorls. The surface of the shell is perfectly smooth under a stereomicroscope, except for more or (usually) less distinct growth lines. Under SEM a fine striation, con- sisting of 4 fine rows of granulae per 10 um, is visible. The aperture is almost triangular, pointed anteriorly and posteriorly. The outer lip is strongly protruding in its adapical part, most so at 4 of its height, counted from the suture. Dimensions. Height of holotype 2.96 mm, maximum height 3.6 mm. Male. The shell resembles that of the female but is slightly more slender, with a straighter spire, and is small- er, % of the height of the female. Galls (Figures 42-45). Variously developed, depending on how long the host has been parasitized. Galls start as a simple lateral or apical depression on a normal-looking spine, inhabited by one or two young specimens. The spine then gradually becomes thicker by a change in its growth pattern and at the same time the depression becomes deep- er. Finally the spine may become completely hollow, with a narrow, apical, or lateral pore, and the diameter of the gall may be twice the diameter of a normal spine. Remarks: DODERLEIN (1887) figured and MORTENSEN (1928:397) mentioned the galls, noticed in specimens of Eucidaris thourarsi from Galapagos and Panama, but they were not aware of the cause. SHASKY (1967) recorded this species under the name “Rosenia nidorum (Pilsbry, 1956),” from Baja California, an identification WAREN (1984b) considered erroneous. A. Warén, 1992 Sabinella shaskyi, however, closely resembles S. troglo- dytes (Thiele, 1925) (= Mucronalia nidorum Pilsbry, 1956) (Figures 53, 54) from the Caribbean area and off West Africa. The two species can be distinguished mainly by the larval shell (Figure 47), which is more slender with flatter whorls in S. troglodytes. Scattered large specimens (Figures 49, 50), of which one was confirmed to be a female, differ in having a less regularly coiled shell and distinct incremental lines. The larval shell is identical with that in normal specimens and I believe this to be individual variation. All the specimens reported above, as well as the hosts, were dried, which made it difficult to make any observa- tions on the soft parts. Furthermore, some of the snails had fallen out of the galls and additional small specimens may have been lost. A few selected specimens from galls were rehydrated, however, and three specimens smaller than 2 mm were confirmed to be males on the basis of a large penis. Three specimens larger than 2 mm but without a penis were assumed to be females. This conforms to the observations on three other species of Sabinella (WAREN, 1984b; BoUCHET & WAREN, 1986). In one case the occurrence of two females and a single male in a gall was confirmed, but no conclusions about the proportions of the sexes or the normal numbers of individuals per gall can be drawn. Two specimens, 1.6 mm high and presumably males, were found free in a sediment sample taken in Ecuador. This indicates that the males may be able to leave the galls for courting. Occasionally the galls are found empty or with empty shells inside, which indicates that the parasite probably has a shorter life-span than the host. (I have noticed this, several times in S. troglodytes also.) Finally, hosts can be found with the spines in a state of repair (Figure 44), with a new point of the spine growing out apically from an empty gall. All these transformations of the spines are possible since the spine is a porously calcified, living en- doskeleton, not a lifeless calcareous structure, as is a spine of a Murex shell. Sabinella shaskyi produces small spherical egg capsules containing about 100 eggs and attaches them to the floor of the gall, where up to a dozen capsules have been found. Examination of 44 specimens of Sabinella shaskyt showed that 25% of the 24 specimens larger than 2 mm had the apex broken off, while not a single (out of 20) specimen smaller than 2 mm had a broken apex. All specimens with a broken apex had lived in incompletely formed galls. This shows two interesting features in the biology of the species: the snails are attacked by shell-cracking predators and the galls have a protective function. Sabinella troglodytes lives in the same way as S. shaskyt, on Eucidars tribuloides (Lamarck, 1816) in the Caribbean area and off West Africa. This host species has been as- sumed to be closely related to and to have diverged from E. thouarsi after the closing of the straits across the Central American isthmus (MORTENSEN, 1928) in the late Plio- Page 193 cene. LEssoIs (1981) investigated the morphologic and ge- netic variation between and within the two sea urchins and confirmed this assumption. It can therefore probably also be assumed that the two species of Sabinella have followed the same allopatric pattern of speciation. ACKNOWLEDGMENTS I thank J. H. McLean, who placed a large part of the eulimid collection in LACM at my disposal, and D. R. Shasky, Redlands, California, who always has been very helpful with material of tropical eulimids and contributed a large part of the specimens used for this study. C. Ham- mar, SMNH, prepared the photographic prints used for illustration. J. H. McLean and D. R. Shasky read and gave valuable comments on the manuscript. LITERATURE CITED ADAMS, H. & A. ADAMS. 1853 (1853-1854). The Genera of Recent Mollusca, Arranged According to Their Organiza- tion. 1. J. van Voorst: London. 484 pp. (Part 8, pp. 225- 256, December 1853). BaRTSCH, P. 1917. A monograph of west American melanellid mollusks. Proceedings of the U.S. National Museum of Nat- ural History 53:295-356. BaRTSCH, P. 1926. Additional new mollusks from Santa Elena Bay, Ecuador. Proceedings of the U.S. National Museum 69(2646):1-20. BARTSCH, P. 1938. Levostraca schwengelae, a new name. Nau- tilus 52:34. Berry, S. S. 1956. A new west Mexican mollusk parasitic on echinoids. American Midland Naturalist 56:355-357. BOUuCcHET, P. & A. WAREN. 1979. The abyssal molluscan fauna of the Norwegian Sea and its relation to other faunas. Sarsia 64:211-243. BOUCHET, P. & A. WAREN. 1986. Revision of the Northeast Atlantic bathyal and abyssal Aclididae, Eulimidae, Epitoni- idae (Mollusca, Gastropoda). Bollettino Malacologico, Sup- plement 2:299-576. BRANN, D.C. 1966. Illustrations to “Catalogue of the Collec- tion of Mazatlan Shells” by Philip P. Carpenter. Paleon- tological Research Institution: Ithaca, N.Y. 111 pp. CARPENTER, P. P. 1857. Catalogue of Mazatlan shells in the British Museum. British Museum: London. i-iv + ix-xvi + 552 pp. CARPENTER, P. P. 1864. Diagnoses of new form of mollusks collected at Cape St. Lucas by Mr J. Xantus. Annals and Magazine of Natural History, Series 3, 14:45-49. DESHAYES, G. P. 1863. Catalogue des mollusques de I’Ile de la Réunion (Bourbon). Jn: L. Maillard, Note sur I’Ile de la Réunion. Dentu: Paris. 144 pp. DODERLEIN, L. 1887. Die Japanischen Seeigel. I. Familie Ci- daridae und Salenidae. Schweitscherbartsche Verlagshand- lung: Stuttgart. 56 pp. EMERSON, W. K. 1965. The eastern Pacific species of Niso (Mollusca: Gastropoda). American Museum Novitates 2218: 1-12. DE FoLin, L. 1867. Les méléagrinicoles, espéces nouvelles. Imprimerie LaPelletier. 74 pp. HERTz, C.M. & J. HERTZ. 1982. A new eastern Pacific species of Eulimostraca (Gastropoda: Eulimidae). The Veliger 25: 72-76. KEEN, A.M. 1968. West American mollusc types at the British Page 194 Museum (Natural History). IV. Carpenter’s Mazatlan col- lection. The Veliger 10:389-439. KEEN, A. M. 1971. Sea Shells of Tropical. West America. Stanford University Press: Stanford, California. 1064 pp. Kisco, B.S. 1959. La collection de Chemnitzidae du Marquis de Folin au Muséum National d’Histoire Naturelle. De- scription de Turbonilla corpulens. Catalogue des espéces pub- liées par de Folin. Journal de Conchyliologie 99:89-112. Lessors, H. A. 1981. Divergence in allopatry: molecular and morphological differentiation between sea urchins separated by the Isthmus of Panama. Evolution 18:618-634. Lyons, W. G. 1989. Nearshore marine ecology at Hutchinson Island, Florida: 1971-1974. XI. Mollusks. Florida Marine Research Publications 1:1-131. McLEan, J. H. 1970. New species of tropical eastern Pacific Gastropoda. Malacological Review 2:115-130. MortTENSEN, T. 1928. A monograph of the Echinoidea. I. Cidaroidea. C. A. Reitzel: Copenhagen. 551 pp. MorTENSEN, T. 1948. A monograph of the Echinoidea. IV:2. Clypeasteroida. C. A. Reitzel: Copenhagen. 471 pp. MonrerosaTo, T. A. DI. 1890. Conchiglie delle profondita di Mare di Palermo. II] Naturalista Siciliano 9:157-166. PETIT, DE LA SAUSSAYE. 1853. Bulletin Bibliographique. Jour- nal de Conchyliologie, Paris 4:216-218 (Issue for May 1853). PHILIPPI, R. A. 1853. Handbuch der Conchyliologie und Mala- cozoologie. E. Anton: Halle. 548 pp. PONDER, W. F. & A. WAREN. 1988. Classification of the Cae- nogastropoda and Heterostropha—a list of family-group names and higher taxa. Malacological Review Supplement 4:288-326. Risso, A. 1826. Histoire naturelle de |’Europe méridionale. 4. Levrault Libraire: Paris. 439 pp. SHASKY, D.R. 1967. Observations on Rosenia nidorum (Pilsbry) and Arene socorroensis (Strong). Annual Report of the Amer- ican Malacological Union 1967:74. The Veliger, Vol. 35, No. 3 SHasky, D. R. 1983a. A preliminary check-list of marine mol- lusks, from Manabi Province, Ecuador. Annual Report, Western Society of Malacologists 16:25-32. SHaASKY, D. R. 1983b. New records of Indo-Pacific Mollusca from Cocos Island, Costa Rica. Nautilus 97:144-145. STRONG, A. M. & L. G. HERTLEIN. 1937. The Templeton Crocker Expedition of the California Academy of Sciences 1932, No. 35. New species of Recent mollusks from the coast of western North America. Proceedings of the California Academy of Sciences, Series 4, 22:159-178. WaREN, A. 1980. Revision of the genera Thyca, Stilifer, Mu- cronalia, Scalenostoma and Echineulima (Mollusca, Proso- branchia, Eulimidae). Zoologica Scripta 9:187-210. WarREN, A. 1984a. An anatomical comparison of Eulima and Pyramidelloides with a revision of the species of Pyramidel- loides (Mollusca, Prosobranchia, Eulimidae). Zoologica Scripta 12:273-294. WaREN, A. 1984b. A generic revision of the family Eulimidae (Gastropoda, Prosobranchia). Journal of Molluscan Studies 13(Suppl.):96 pp. WaREN, A. 1991. Revision of Hypermastus Pilsbry, 1899 and Turveria Berry, 1956 (Gastropoda: Prosobranchia: Eulimi- dae), two genera parasitic on sand dollars. Records of the Australian Museum 43:85-112. WaREN, A. 1992. Balea Gray, 1824 (Mollusca: Gastropoda): proposed conservation. Bulletin of Zoological Nomenclature 49:12-15. WarEN, A. & R. MOOLENBEEK. 1989. A new eulimid gastro- pod, Trochostilifer eucidaricola, parasitic on the pencil urchin Eucidaris tribuloides from the southern Caribbean. Proceed- ings of the Biological Society of Washington 102:169-175. WarEN, A. & C. Mirsup. 1990. Nanobalcis, a new eulimid genus (Prosobranchia) parasitic on cidaroid sea urchin, with two new species and comments on Sabinella bonifaciae (Nord- sieck). Bollettino Malacologico 26:37-46. The Veliger 35(3):195-204 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 Geographic and ‘Temporal Variation in Shell Morphology of Acanthina Species from California and Northern Baja California by GARY L. GIANNINY anp DANA H. GEARY Department of Geology and Geophysics, 1215 W. Dayton Street, University of Wisconsin, Madison, Wisconsin 53706, USA Abstract. Intraspecific patterns of geographic and temporal variation are important to understanding the processes involved in the maintenance and divergence of species. Here we examine patterns of geographic and Pleistocene-Recent temporal variation in species of the neogastropod Acanthina from coastal California and Baja California. A cline of increasing shell height with increasing latitude is present among modern Acanthina punc- tulata. This pattern is similar to that observed in some other mollusk species, and may be related to temperature, growth rate, and individual longevity. In Acanthina spirata, modern samples are consistently different from fossil (Late Pleistocene) samples with respect to shouldering, except that modern samples from Los Angeles appear more similar to the fossils. Modern and fossil samples also differ consistently in relative shell thickness. We believe these differences are more readily explained as genetic, rather than purely ecophenotypic. Range expansions of A. spirata following Pleistocene climate fluctuations may have facilitated the spread of the characteristic modern forms. INTRODUCTION Patterns of geographic and temporal variation within and among species are important to our understanding of how species are maintained and how they give rise to new species. Patterns of variation in organisms inhabiting an area such as the west coast of North America are of special interest because frequent tectonic activity and Pleistocene sea-level and temperature fluctuations have provided a particularly dynamic context in which both organisms and species live out their lives. The degree and type of intra- specific variation present may reflect a species’ response to this dynamic physical environment. This study analyzes data on both geographic and tem- poral variation within species of Acanthina (FISCHER, 1807; Muricacea) from the Pacific coast of California and Baja California. The genus Acanthina includes five extant spe- cies of rocky intertidal carnivores from the west coast of North America (Wu, 1985). We examined the three most northerly (A. spzrata (Blainville, 1832), A. punctulata (Sow- erby, 1831), and A. paucilirata (Stearns, 1872); see Figures 1, 2), and focus here on the former two. In this paper we explore the kinds of variation present in the shell morphology of these species, and examine its geographic and temporal expression. The patterns found reflect the physical variation encountered within species’ geographic ranges, and the dynamic nature of this envi- ronment through time. MATERIALS anp METHODS Pleistocene glacioeustatic sea level highstands left “bathtub rings” of marine terrace sediments along the coast of Cal- ifornia and Baja California. Subsequent tectonism has el- evated these deposits, sparing them from further marine erosion. Marine terraces are formed by the leading edge of the transgressive (and subsequent regressive) belt sander of erosion. These terraces are usually cut into the coastal bedrock (BRADLEY & GRIGGS, 1976), which is often well lithified in this area, creating abundant rocky intertidal habitat. The fossiliferous nearshore sediments on these terraces were probably deposited during a fall of relative sea level (BRADLEY & GRIGGS, 1976). Correlation of marine terrace deposits in California and Baja California has received considerable attention. Many workers have used elevation above current sea level and Page 196 Fossil Xx X x A. Spirata A A. paucilirata QO A. punctulata The Veliger, Vol. 35, No. 3 40° —— San Francisco Bay Point Conception Los Angeles 350 SX San Diego 30° Punta” Eugenia 25° 212 Figure 1 Index maps of the coast of California and Baja California, showing approximate collection localities of modern (left) and fossil (right) samples. Modern latitudinal ranges for species are given at left. Southern end of Acanthina spirata range includes scattered, rare reports as far south as Punta Abreojos, Baja California (Wu, 1985). Baja California Norte (B.C.N.) and Baja California Sud (B.C.S.) are indicated at right. the relative position within a suite of terraces at one locality to correlate terrace deposits (KANAKOFF & EMERSON, 1959; ORTLEIB, 1986; and references therein). VALENTINE (1961) and VALENTINE & MEADE (1961) developed a biostrati- graphic correlation system based on the distribution of climatically sensitive extralimital southern and northern mollusk species. More recently, a variety of other tech- niques have been applied to the problem, including amino acid racemization (WEHMILLER et al., 1977; LAJOIE et al., 1979; KENNEDY et al., 1982), uranium-series dating of corals (KU & KERN, 1974) and mollusks (FANALE & SCHAEFFER, 1965; SZABO & ROSHOLT, 1969), oxygen iso- topic temperature estimates from mollusks (VALENTINE & MEaDE, 1961), and an interdisciplinary approach utilizing the zoogeographic signature (warm, cool) of terrace faunas (KENNEDY, 1978; KENNEDY ef al., 1982, KENNEDY & WEHMILLER, 1986). Amino acid racemization dates in this region are calibrated by uranium-series dating of solitary corals from the 80 ka (= 80,000 years before present) and 125 ka highstand deposits from Oregon to Baja California (VEEH & VALENTINE, 1967; Ku & KERN, 1974; ROCKWELL etal., 1989; MUHS et al., 1990). These relative and absolute dating techniques allow correlation with the oxygen iso- tope curves of SHACKLETON & OPDYKE (1973, 1976) and G. L. Gianniny & D. H. Geary, 1992 sea level curves of BLOOM et al. (1974) and SHACKLETON (1987), placing these deposits in a well-defined temporal framework. Species of Acanthina are generally well represented in the Pleistocene marine terraces of coastal California and Baja California. In addition to personal field collections, we utilized material from the Los Angeles County Mu- seum of Natural History, the San Diego Museum of Nat- ural History, the Santa Barbara Museum of Natural His- tory, the California Academy of Sciences, and the University of California Museum of Paleontology, Berkeley. The fos- sil collections are from localities that have been correlated by other workers on the basis of the temperature affinities of the fauna, calibrated aminostratigraphic age estimates, shoreline-angle elevations, and geomorphic relationships with other deposits of known age (see Table 1). More than 90% of our fossil Acanthina specimens come from the 125 ka highstand deposits of the last interglacial period. The 125 ka transgression probably removed many of the previous terrace deposits (with notable exceptions in the Palos Verdes Hills, San Diego, and Punta Banda, Baja). In the rare pre-125 ka rocky intertidal deposits, Acanthina is seldom preserved (owing to ground-water leaching), and almost never in the large number of well- preserved individuals desirable for morphological analysis. In both modern and fossil samples, the largest individ- uals showing minimal breakage and little or no evidence of shell repair were selected for measurement. Specimens with shell height less than one-third that of the largest individual from that sample were not included, in order to minimize the potential effects of ontogenetic variation. Most modern samples contained 30 measured individuals; fossil samples typically had fewer individuals suitable for measurement (see Table 1). A total of 878 modern and 214 fossil specimens were measured. Shells were mounted in cardboard trays in full apertural view, with the columellar axis horizontal to the tray. Mea- surements were made from a video image of the specimens using a digitizing pad and Bioquant software. In addition, we used modified vernier calipers to measure the thickness of the shell wall. We measured a total of eight variables and calculated an additional six ratios or shape factors (see Appendix). Variables were chosen to describe the size and shape of the shell and the aperture. We employed a variety of univariate and multivariate techniques to investigate potential patterns in the data (using SYSTAT; WIL- KINSON, 1988). We present here only univariate or bivar- iate plots, because they are easiest to interpret, and because the multivariate analyses provide no additional informa- tion or clarity in these particular cases. We examined measurement error by remounting and measuring one specimen 20 times. Measurements subject to the most variability were those that involved tracing of the shell outline, or some portion thereof, such as shell area, or shouldering (see Appendix for standard deviations for the replicates of each variable). For all variables, mea- Page 197 Figure 2 The three species of Acanthina in this study: A. paucilirata, A. punctulata, and A. spirata, from left to right. Scale bar is 1 cm. The characteristic labial spine is used to open the opercular plates on barnacles (YENSEN, 1979; PERRY, 1985), which, with mussels and gastropods, are the typical prey of these carnivores (SLEDER, 1981; MENGE, 1974; PERRY, 1985). surement error is smaller than the differences among sam- ples or species that we discuss here. RESULTS Distinguishing Among Species The three species we studied are distinguished on the basis of radular and other soft-part characters (Wu, 1985), which correlate with differences in shell size, shape, and color and ornamentation pattern. Although distinguishing among species was not our primary objective here, the three species separate fairly well on the basis of shell size and shape (as measured by shell height versus the ratio of shell height to width; Figure 3). In general, specimens of Acan- thina paucilirata are the smallest and relatively most stout, whereas specimens of A. spirata exhibit the largest size and most slender shape. Acanthina punctulata Modern samples of Acanthina punctulata exhibit a cline of increasing shell size with increasing latitude (Figure 4). The southernmost sample and the two northernmost sam- ples do not fit the cline. For all other samples, however, the pattern is robust. Correlation of mean shell height with latitude for these samples yields r = 0.980, P < 0.001; correlation of maximum shell height with latitude yields r = 0.962, P < 0.001; for all samples, correlation of max- imum height with latitude yields 7 = 0.739, P = 0.015. None of the shape variables in Acanthina punctulata shows this latitudinal relationship, nor is shape highly correlated with size in this species. Fossils of A. punctulata Page 198 The Veliger, Vol. 35, No. 3 Table 1 Sample, locality, sample size (7), age, correlation method, source, and latitude for modern and fossil samples of Acanthina used in this study. Method abbreviations: AA, amino acid racimization; BT, biostratigraphic temperature affinities; Phy, physical relationships; Th/U, uranium series dating; VA, volcanic ash. Source abbreviations: 1, T. A. Demere (personal communication); 2, EMERSON (1956); 3, KENNEDY et al. (1982); 4, G. L. Kennedy (personal communication); 5, KERN (1977); 6, KU & KERN (1974); 7, LAJOIE et al. (1979); 8, MuHs (1985); 9, MuHs et al. (1988); 10, ORTLIEB (1986); 11, SARNA-WOJCICKI et al. (1985); 12, WEHMILLER et al. (1977); 13, VALENTINE (1980); 14, VALENTINE & VEEH (1969). Museum abbreviations: CAS, California Academy of Sciences; LACM, Los Angeles County Museum; LACMIP, Los Angeles County Museum, Department of Invertebrate Paleontology; SBMNH, Santa Barbara Museum of Natural History; SDMNH, San Diego Museum of Natural History; UCMP, University of California Museum of Paleontology: UW, University of Wisconsin Geology Museum. Method and Lati- Sample Locality n Age source Museum Catalog no. tude Acanthina paucilirata 102701 Thurloe Bay, BCS 20 modern NA LACM 35179 27.62°N 103101 Punta Piedras, BCN 30 modern NA SBMNH 10461 31.37°N 103102 San Miguel, BCN 10 modern NA LACM 61842 31.90°N 103103 Ensenada, BCN 29 modern NA LACM 11358 31.85°N 103104 Punta Morro, BCN 30 modern NA UW 1861/5 31.86°N 103201 Tijuana Beach, BCN 30 modern NA LACM 62-18 32.50°N 103202 Punta Descanso, BCN 30 modern NA UW 1861/6 32.26°N 103203 La Jolla, CA 30 modern NA LACM 14007 32.85°N 103204 Rosarito, BCN 30 modern NA LACM 127444 32.35°N 701 Punta Loma, CA 6 120 ka BT, AA 5 SDMNH © 12923 32.66°N 702 Punta China, BCN 1 ?120ka BT 2 UCMP A-9002 31.50°N 704 ?Punta Loma, CA 6 120 ka BT 5 LACMIP (UCLA 3605) 32.66°N 705 Punta Loma, CA 19 80 ka BT, AA 9 LACMIP 11701 32.60°N Acanthina punctulata 203101 Santo Thomas, BCN 30 modern NA LACM 66-1 31.55°N 203201 San Diego, CA 27 modern NA LACM 140004 32.71°N 203301 Point Fermin, CA 21 modern NA LACM 63070 33.73°N Point Fermin, CA 9 modern NA LACM 60414 33.73°N 203302 San Nicolas Is., CA 30 modern NA LACM 60700 33.33°N 203401 Santa Cruz Is., CA 10 modern NA LACM 69-11 34.00°N Santa Cruz Is., CA 9 modern NA LACM 60729 34.00°N Santa Cruz Is., CA 10 modern NA LACM 60694 34.00°N 203402 Hobson Point, CA 30 modern NA LACM (T&B Phillips:629) 34.33°N 203501 Shell Beach, CA 30 modern NA LACM 61-11 35.16°N 203502 Piedras Blancas, CA 30 modern NA SBMNH = 02461 35.49°N 203503 Cayucos, CA 8 modern NA LACM 66614 35.35°N Cayucos, CA 22 modern NA SBMNH = 40121 35.35°N 203601 Monterey, CA 7 modern NA SBMNH — 66197 36.62°N Monterey, CA 15 modern NA LACM 64.2 36.62°N Monterey, CA 5 modern NA LACM 14092 36.62°N 601 San Nicolas Is., CA 19 120 ka Th/U, AA 14,8 LACMIP_ 11749 33.33°N Acanthina spirata 3001 Socorro, BCN 24 modern NA LACM 66-3 30.49°N 3201 Rosarito, BCN 30 modern NA SBNHM _— 10080 32.35°N 3302 Newport Beach, CA 30 modern NA SBNHM __ 10082 33.40°N 3303 San Pedro, CA 30 modern NA SBNHM 10082 33.749N 3339 San Pedro, CA 30 modern NA LACM 140001 33.74°N 3401 Summerland, CA 30 modern NA SBMNH 25513 34.60°N 3501 Morro Bay, BCN 28 modern NA SBNHM __ 10078 35.33°N 3601 Del Ray, CA 30 modern NA LACM 60696 36.62°N 3701 San Francisco, CA 30 modern NA SBNHM 10077 37.97°N 3702 San Francisco, CA 23 modern NA LACM 62-2 37.79°N 3801 Tomales Bay, CA 30. modern NA LACM AHF-1629-48 38.17°N 3802 Tomales Bay, CA 30 modern NA LACM AHF-1625-48 38.10°N 901 San Francisco, CA (Merced Fm.) 5 400 ka VA 11 CAS 6039.01 37.73°N San Francisco, CA (Merced Fm.) 5 400 ka VA 11 CAS 59206.01 37.73°N G. L. Gianniny & D. H. Geary, 1992 Page 199 Table 1 Continued Method and Lati- Sample Locality n Age source Museum Catalog no. tude 902 San Francisco (Merced Fm.) 1 400 ka VA 11 CAS 60404.01 37.72°N San Francisco, CA (Merced Fm.) 9 400 ka VA 11 CAS 2318.01 37.72°N 903 Santa Cruz, CA 1 85 ka AA 7 CAS 59647.01 36.95°N 904 Cayucos, CA 1 120ka Th/U, AA 12 LACMIP (UCLA 3393) 35.44°N 905 Ventura (Santa Barbara Fm.) 3 >700 ka VA 7 LACMIP = 50243 34.27°N Ventura (Santa Barbara Fm.) 2 >700 ka VA 7 SDMNH — 29603 34.27°N 906 Sea Cliff, CA 1 5 ka Ccl¥, AA 7 SDMNH © 13910 34.34°N 907 Ventura Terrace, CA 9 45 ka AA 7 LACMIP 5029 34.279°N 908 *“Railroad Depot” 30 =120 ka AA 7 CAS 91.09 33.73°N 909 *San Pedro 14. 120 ka AA 3 LACMIP 131 33.73°N 910 *San Pedro 6 120 ka AA 3 CAS 66025.01 33.73°N 911 *San Pedro 8 120 ka AA 3 CAS 6602.07 33.73°N 912 *San Pedro 7 120 ka AA 3 CAS 66021.01 33.73°N 913 Camp Pendleton, CA 7 120 ka AA 3 LACMIP = 5574 33.40°N 914 La Jolla, CA 6 120 ka AA 6 SDMNH © 12547 32.82°N 915 Camalu, BCN 10 120 ka AA, BT 13. UCSB 1143 30.83°N 916 Camalu, BCN 3 120 ka AA, BT 13 UW 1861/1 30.83°N 917 Pta. San Telmo, BCN 11 ?120 ka Phy 10 UW 1861/2 30.93°N 918 Pta. Baja, BCN 2 ?120ka BT, Phy 2 UW 1861/3 29.96°N 919 San Quintin, BCN 1 ?120 ka 10 LACMIP (UCLA 2411) 30.48°N San Quintin, BCN 3 ?120 ka 10 SBMNH 16066 30.48°N 920 La Fonda, BCN 4 120ka AA, Phy, BT 4 UW 1861/4 32.13°N 921 San Francisco, CA 1 ?400 ka VA 11 UCMP B-4810/37672 37.73°N 922 San Diego, CA (Bay Point Fm.) 6 ?205 ka AA 1 CAS 105.01 32.80°N * Palos Verdes Sands, Los Angeles County, CA. were too rare to provide significant sample sizes for mor- ever, exhibited considerable shell wear, which made ac- phometric analyses. curate comparison of most shell characters difficult. There- fore, we believe our data are inconclusive with respect to Acanthina paucilirata potential trends or patterns within this species. We measured specimens from eight modern and four fossil samples of A. paucilirata. Our fossil specimens, how- 2.0 A. punctulata OA. paucilirata x A. punctulata 1.97 A. spirata 2 £ = £ = ie ah if = 2 = 1.7 2 . - J oO gate i o fs 77) ai maximum height mean height 1.4 31 32 33 34 35 36 37 10 20 30 40 latitude (°N) shell height (mm) Figure 4 Figure 3 Relationship of shell height to latitude among modern samples Sample means for all modern samples for shell height and the of Acanthina punctulata. Both sample means and maximum values ratio of shell height to width. are shown. Page 200 60 50 40 oa 30 Lo] =) 2 = modern £ ee (non-Los Angeles) Ee ° eo ® 2 E 3 Cc 10 1.0 1.2 The Veliger, Vol. 35, No. 3 1.4 1.6 shouldering Figure 5 Histograms for shouldering for all modern and fossil specimens of Acanthina spirata. Shouldering is measured as the trace of the right hand shell margin from apex to body whorl suture, divided by the length of the straight line segment connecting the same two points (see Appendix). Shells with nearly straight shell profiles have lower values for shouldering; shells with pronounced development of shoulders have higher values. Means for the three groups indicated are modern non-Los Angeles (1.12), modern Los Angeles (1.21), and fossil (1.28). Acanthina spirata Comparison of fossil with modern samples of Acanthina spirata revealed at least two consistent morphological dif- ferences: (1) modern samples exhibit less pronounced shouldering than do fossil samples, with the exception of modern samples from Los Angeles, which more closely resemble the fossils; and (2) for any given shell height, modern samples have relatively thicker shells. Differences between modern and fossil samples with respect to shouldering are shown in Figures 5 and 6. Figure 5 plots all individuals with respect to this character; Figure 6 illustrates typical fossil and modern morphologies. ‘The difference between fossil and modern samples is significant with a t-test at P < 0.001. Among modern specimens, those from the Los Angeles area exhibit the highest values for shouldering, and indeed, are intermediate between typ- ical fossil and modern morphologies with respect to this character. The relationship between shell height and shell wall thickness is plotted in Figure 7. We did not measure shell thickness on individuals that showed obvious signs of shell wear or dissolution; all samples for which shell thickness could be accurately measured are plotted in Figure 7. This plot reveals a separation of fossil from modern samples; G. L. Gianniny & D. H. Geary, 1992 Page 201 Figure 6 Typical specimens of Acanthina spirata, showing characteristic differences in shouldering between fossil (right) and modern non- Los Angeles (left) specimens. The modern Los Angeles specimen (center) more closely resembles typical fossil shouldering mor- phology. Scale bar is 1 cm. fossil shells have thinner shells relative to their size than do modern shells. An analysis of covariance (ANCOVA) demonstrates that this difference in relative thickness be- tween fossil and modern samples is significant (P < 0.001). Modern Los Angeles samples are indicated by +’s, but do not differ significantly from other modern samples with respect to relative shell thickness. Samples of Acanthina spirata exhibit no other consistent patterns of geographic variation, such as the size cline observed in A. punctulata. DISCUSSION Two types of patterns are indicated in the data presented above: (1) a latitudinal size cline in Acanthina punctulata; and (2) time-related, and partly geographically related, differences in shell shouldering and thickness in A. spzrata. Cline in Shell Size in Acanthina punctulata Intraspecific clines of increasing size with higher latitude such as that seen over most of the geographic range of Acanthina punctulata have been reported for a number of other mollusks (e.g., WEYMOUTH et al., 1931; NEWELL, 1964; FRANK, 1975; see also HARRINGTON, 1987). This type of clinal variation is thought to stem from slower growth rates but increased longevity at lower temperatures, resulting in larger overall size. We have no information on individual longevity or relative growth rate in these snails, but the similarity of our pattern to that of others suggests a similar cause. The exceptional southernmost and northernmost samples of A. punctulata may result from unusual temperature conditions: proximity to local up- Acanthina_ spirata =~ O° fossil 5 @ modern (non-L.A.) = + modern (Los Angeles) 7) o ) c x 1 = @ S r) £ 7) 20 30 40 shell height (mm) Figure 7 Means of shell height versus shell thickness for samples of Acan- thina spirata. Shell thickness could not be accurately measured on all specimens because of dissolution, wear, or breakage. Only those specimens exhibiting excellent preservation were measured for shell thickness. welling and, therefore, to unusually cool temperatures for the southernmost sample (near Punta Banda), and un- usually warm temperatures at the sheltered northernmost Monterey Bay locality. The sample from Piedras Blancas Point, California (approximately 35.5°N) is not consistent with the overall size pattern; the reasons for this are un- known. Temporal and Geographic Patterns in Acanthina spirata Acanthina spirata exhibits consistent differences in shell shouldering and thickness between fossil and modern sam- ples (Figures 5-7). It appears that these patterns are not temperature related, insofar as these characters do not exhibit any consistent relationship with latitude. Further- more, it seems unlikely that these differences are purely ecophenotypic (non-genetic). A number of authors have reported environmentally correlated variability in gastro- pod shell shape, the basis for which may be at least partially ecophenotypic (e.g., KITCHING ef al., 1966; CROTHERS, 1981, 1983; BROWN & QUINN, 1988). To explain the temporal shifts in A. spirata on a purely ecophenotypic basis, one would have to postulate either: (1) a consistent environ- mental shift across the entire geographic range of the species, or (2) a stable onshore-offshore ecophenotypic gradient, with fossil and modern samples coming from consistently different positions along the gradient. We believe that both of these possibilities are remote. The only mode of dispersal for species of Acanthina is through active movement of individuals. Individuals of this genus have internal fertilization, and juveniles emerge di- rectly from egg cases attached to rocky surfaces (HEWATT, Page 202 1934; SPIGHT, 1977). The limited dispersal ability of spe- cies in this genus raises the question of how the observed differences in shell morphology spread through the range of A. spirata. The pronounced paleoclimatic temperature changes that characterize the Pleistocene, and the latitudinal migrations of species ranges that resulted, may help to explain how the observed morphological shifts occurred. Rather than a direct replacement of fossil morphotypes with modern ones, the shifts in morphology may have occurred as the species spread into its current range after occupying a more re- stricted or only partly overlapping range during cooler (and/or warmer) intervals of the Pleistocene. Marine tem- peratures since 125 ka have been generally cooler than at present, and therefore the northern limit for Acanthina spirata was probably farther south than it is today. The population(s) that spread north to establish the current range of A. spirata most likely represented only a small subset of the variability that was present within the species. Whether the modern phenotype had some selective ad- vantage during recolonization, or came to dominate simply through a kind of founder effect, is unknown. The fact that modern A. spirata from the Los Angeles area resemble fossil samples or are intermediate between typical fossil and modern phenotypes suggests that populations may have been maintained more continuously in this area. This hypothesis requires further testing with distributional data from throughout the Pleistocene. ACKNOWLEDGMENTS We thank the following people, many of whom not only lent specimens from collections in their care, but also gave critical insights into regional stratigraphy and intertidal ecology, and offered an occasional warm roof for the night: Tim Baumgartner, Anamarie Escofe, and M. C. Francisco Suarez Vidal of Centro de Investigacion Cientifica y de Educacion Superior de Ensenada; Thomas A. Demeré and Richard C. Brusca of the San Diego Natural History Museum; George L. Kennedy, Edward C. Wilson, and James H. McLean of the Los Angeles County Museum of Natural History; Paul H. Scott of the Santa Barbara Museum of Natural History; James W. Valentine of the University of California, Berkeley; Michael G. Kellogg of the California Academy of Sciences; L. J. Bryant and David R. Lindberg of the University of California Mu- seum of Paleontology; Shi-Kuei Wu of the University of Colorado Museum; and Barry Roth, San Francisco. This study is part of G. Gianniny’s M.S. Thesis in Geology at the University of Wisconsin, Madison; com- mittee members C. W. Byers and L. J. Maher provided useful comments and discussion of this work. We thank Matthew Colbert, Kathy Beratan, Patrock Williamson, and Chris Smith for logistical help; Susan Werther for drafting Figures 1 and 5; Cynthia Dott for help in the field; Robert Bleiweiss, George L. Kennedy, A. R. Palmer, The Veliger, Vol. 35, No. 3 Barry Roth, and Geerat Vermeij for comments on various versions of the manuscript. Funding for this research was provided by the Graduate School at the University of Wis- consin, Madison. LITERATURE CITED BLoom, A. L., W. S. BROECKER, J. M. A. CHAPPELL, R. K. MatTTHEws & K. J. MESOLELLA. 1974. Quaternary sea level fluctuations on a tectonic coast: new ?°Th/2*Th dates from the Huon Peninsula, New Guinea. Quaternary Re- search 4:185-205. BRADLEY, W. C. & G. B. Griccs. 1976. Form, genesis, and deformation of central California wave-cut platforms. Geo- logical Society of America Bulletin 87:433-449. Brown, K. M. & J. F. QUINN. 1988. The effect of wave action on growth in three species of intertidal gastropods. Oecologia 75:420-425. CROTHERS, J. H. 1981. Shell-shape variation in Faroese dog- whelks (Nucella lapillus (L.)). Biological Journal of the Lin- nean Society 15:327-337. CROTHERS, J. H. 1983. Some observations on shell-shape vari- ation in North American populations of Nucella lapillus (L.). Biological Journal of the Linnean Society 19:237-274. EMERSON, W. K. 1956. Pleistocene invertebrates from Punta China, Baja California, Mexico, with remarks on the com- position of the Pacific Coast Quaternary faunas. American Museum of Natural History Bulletin 3:317-342. FANALE, F. P. & O. A. SCHAEFFER. 1965. Helium-uranium ratios for Pleistocene and Tertiary fossil aragonites. Science 148:312-317. FISCHER, G. 1807. Catalogue systematique et raisonne des curiositee de la nature et de l’art. Tome 3, Vegetaux et animaux. Museum Demidoff: Moscou. FRANK, P. W. 1975. Latitudinal variation in life history fea- tures of the black turban snail Tegula funebralis (Prosobran- chia: Trochidae). Marine Biology 31:181-192. HARRINGTON, R. J. 1987. Skeletal growth histories of Proto- thaca staminea (Conrad) and Protothaca grata (Say) through- out their geographic ranges, northeastern Pacific. The Ve- liger 30:148-158. HewatTt, W. G. 1934. Ecological studies on selected marine intertidal communities of Monterey Bay. Ph.D. Disserta- tion, Stanford University. KANAKOFF, G. P. & W. K. EMERSON. 1959. Late Pleistocene invertebrates of the Newport Bay area, California. Los An- geles County Museum Contributions in Science, No. 31:46 PP- KENNEDY, G. L. 1978. Pleistocene paleoecology, zoogeography and geochronology of marine invertebrate faunas of the Pa- cific Northwest Coast (San Francisco to Puget Sound). Ph.D. Dissertation, University of California, Davis. 824 pp. KENNEDY, G. L., K. R. Lajom & J. F. WEHMILLER. 1982. Aminostratigraphy and faunal correlations of late Quater- nary marine terraces, Pacific Coast, USA. Nature 299:545- 547. KENNEDY, G. L. & J. F. WEHMILLER. 1986. Paleoclimatic implications of Quaternary marine invertebrate faunas from southwestern Santa Barbara County, California. Western Society of Malacologists Annual Report 18:22-23. KERN, J. P. 1977. Origin and history of upper Pleistocene marine terraces, San Diego, California. Geological Society of America Bulletin 88:1553-1566. KITCHING, J. A., L. MUNTZ & F. J. EBLING. 1966. The ecology G. L. Gianniny & D. H. Geary, 1992 of Lough Ine. XV. The ecological significance of shell and body forms in Nucella. Journal of Animal Ecology 35:331- 341. Ku, T.-L. & J. P. Kern. 1974. Uranium-series age of the upper Pleistocene Nestor Terrace, San Diego, California. Geological Society of America Bulletin 85:1713-1716. Lajolgz, K. R., J. P. KERN, J. F. WEHMILLER, G. L. KENNEDY, S. A. MaTuHiEson, A. M. SARNA-WOJCICKI, R. F. YERKES & P. A. McCrory. 1979. Quaternary marine shorelines and crustal deformation, San Diego to Santa Barbara, Cal- ifornia. Pp. 1-15. Jn: P. L. Abbott (ed.), Geological Excur- sions in the Southern California Area. San Diego State Uni- versity. MENGE, J. L. 1974. Prey selection and foraging period of the predaceous intertidal snail, Acanthina punctulata. Oecologia 17:293-316. Muus, D. R. 1985. Amino acid age estimates of marine terraces and sea levels on San Nicolas Island, California. Geology 13:58-61. Munus, D. R., H. M. KeE.tsey, G. H. MILLer, G. L. KENNEDY, J. F. WHELAN & G. W. MCINELLY. 1990. Age estimates and uplift rates for Late Pleistocene marine terraces: south- ern Oregon portion of the Cascadia forearc. Journal of Geo- physical Research 95:6685-6698. Muus, D. R., G. L. KENNEDY & T. K. ROCKWELL. 1988. Uranium-series ages of corals from marine terraces, Pacific coast of North America: implications for the timing and magnitude of Late Pleistocene sea level changes. Programs and Abstracts of the 10th Biennial Meeting, p. 140, Illinois State Water Survey, Urbana. NEWELL, G. E. 1964. Physiological variation in intertidal mol- luscs. Pp. 59-87. In: K. M. Wilbur & C. M. Yonge (eds.), Physiology of Molluscs 1. Academic Press: New York. ORTLEIB, L. 1986. Neotectonique et variations du niveau marin au Quaternaire dans la region du Golfe de Californie, Me- xique. D.Sci. Dissertation, Universite D’Aix Marseille 11. Vol. 1:779 pp., Vol. 2:257 pp. Perry, D. M. 1985. Function of the shell spine in the pre- daceous rocky intertidal snail Acanthina spirata (Prosobran- chia: Muricacea). Marine Biology 88:51-58. ROCKWELL, T. K., D. R. Muus, G. L. KENNEDY, M. E. Hatcu, S. H. Witson & R. E. KLINGER. 1989. Uranium-series ages, faunal correlations and tectonic deformation of marine terraces within the Agua Blanca fault zone at Punta Banda, northern Baja California, Mexico. Pp. 1-16. Jn: P. L. Abbott (ed.), Geologic Studies in Baja California, Pacific Section. Society of Economic Paleontologists and Mineralogists: Los Angeles, California. SaRNnA-WojJcicki, A. M., C. E. MEYER, H. R. Bowman, N. T. HALL, P. C. RusseLt, M. J. WooDWARD & J. L. SLATE. 1985. Correlation of the Rockland Ash Bed, a 400,000 year- old stratigraphic marker in northern California and western Nevada, and implications for middle Pleistocene paleogeog- raphy of central California. Quaternary Research 23:236- DN SHACKELTON, N. J. 1987. Oxygen isotopes, ice volumes and sea level. Quaternary Science Reviews 6:183-190. SHACKELTON, N. J. & N. D. OppyKE. 1973. Oxygen isotope and paleomagnetic stratigraphy of equatorial Pacific core V28-238: oxygen iostope temperatures and ice volumes on a 10° and 10° year scale. Quaternary Research 3:39-55. SHACKLETON, N. J. & N. D. OppykE. 1976. Oxygen-isotope and paleomagnetic stratigraphy of Pacific Core V28-239: Late Pliocene to latest Pleistocene. Geological Society of America Memoirs 45:449-464. Page 203 SLEDER, J. 1981. Acanthina punctulata (Neogastropoda: Muri- cacea): its distribution, activity, diet and predatory behavior. The Veliger 24:172-180. SpIGHT, T. M. 1977. Latitude, habitat and hatching type for Muricacean gastropods. Nautilus 91:67-71. SzaBo, B. J. & J. N. RosHoLT. 1969. Uranium-series dating of Pleistocene shells from southern California—an open sys- tem model. Journal of Geophysical Research 74:3253-3260. VALENTINE, J. W. 1961. Paleoecological molluscan geography of the Californian Pleistocene. University of California Pub- lications in Geological Sciences 37:309-442. VALENTINE, J. W. 1980. Camalu: a Pleistocene terrace fauna from Baja California. Journal of Paleontology 54:1310-1318. VALENTINE J. W. & R. F. MEADE. 1961. California Pleistocene paleotemperatures. University of California Publications in Geological Sciences 40:1-46. VALENTINE, J. W. & H. H. VEEH. 1969. Radiometric ages of Pleistocene terraces from San Nicolas Island, California. Geological Society of America Bulletin 80:1415-1418. VEEH, H. H. & J. W. VALENTINE. 1967. Radiometric ages of Pleistocene fossils from Cayucos, California. Geological So- ciety of America Bulletin 78:547-549. WEHMILLER, J. F. et al. 1977. Correlation and chronology of Pacific Coast marine terrace deposits of continental United States by fossil amino acid stereochemistry—technique eval- uation, relative ages, kinetic model ages, and geologic im- plications. U.S. Geological Survey Open-File Report 77-680, 196 pp. WEYMOUTH, F. W., H. C. MCMILLAN & W. H. RicH. 1931. Latitude and relative growth in the razor clam, Szliqua patula. Journal of Experimental Biology 8:222-249. WILKINSON, L. 1988. SYSTAT: the system for statistics. SY- STAT, Inc.: Evanston, IL. 882 pp. Wu, S.-K. 1985. The genus Acanthina (Gastropoda: Murica- cea) in western America. Special Publication of the Mu- kaishima Marine Biological Station 1985, pp. 45-66. YENSEN, N. P. 1979. The function of the labial spine and the effect of prey size on “switching” polymorphs of Acanthina angelica (Gastropoda: Thaididae). Ph.D. Dissertation, Uni- versity of Arizona, Tucson. 62 pp. APPENDIX Variable descriptions. (The standard deviation of 20 rep- licate measurements of one specimen [with remounting] is given in parentheses.) Size Variables Shell height (SHELLH): The distance (in mm) from base to apex of shell. In cases of extreme shell wear this variable was estimated on the basis of other specimens in the same sample. (0.38) Shell width (SHELLW): The sum of the maximum dis- tance perpendicular to the columellar axis on the right and left sides of the shell (in mm). Snails were mounted on 1-mm graph paper with coiling axis parallel to the y-axis of the paper. Width was measured on the shell image as the maximum delta x. (0.26) Shell area (SHELLA): The area of the shell image cal- culated by Bioquant based on a manual tracing of the Page 204 shell image perimeter with the digitizer mouse (in mm’). (2.63) Shell wall thickness (WALTHIK): The thickness of the shell wall, measured (with modified vernier calipers) just behind the first row of denticles on the body whorl, at approximately the second denticle from the adapical end (in mm). (0.09) Shape Variables Apical angle (ANGLERT): The acute angle measured in degrees between the columellar axis and the body whorl - penultimate whorl suture on the right edge of the shell (shell in standard position). (1.17) Shell shape factor (SHELLSF): 4(z) (shell area)/(shell perimeter). This variable is calculated by Bioquant (unitless). (0.04) Height/width (HWDIV): The ratio of shell height to shell width (unitless). (0.03) Apertural Variables Aperture height (APTH): The distance along the colu- mellar axis measured from the base of the siphonal canal to a point perpendicular to the suture line of the body whorl (in mm). (0.35) The Veliger, Vol. 35, No. 3 Aperture width (APTW): The maximum width of the aperture along a line perpendicular to the columellar axis (in mm). (0.26) Aperture height/aperture width (AHAWDIV): The ratio of aperture height to aperture width. (0.06) Aperture area (APTAREA): The aperture area based on a manual tracing of the aperture image with the digitizer mouse (in mm?). (3.23) Aperture shape factor (APTSF): 4(7) (aperture area)/ (aperture perimeter)’; calculated by Bioquant: a mea- sure of apertural shape (unitless). (0.04) Aperture height/shell height (AHSHDIV): The ratio of aperture height to shell height; a measure of the relative size of the aperture. (0.01) Miscellaneous Shouldering (SHDR): The ratio of: (the traced distance along the right margin of the shell image from the apex to the adapical body whorl suture) divided by (the straight line distance connecting the same two points). This vari- able is a measure of the definition of shell shouldering. (0.14) The Veliger 35(3):205-214 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 A New Genus and Species of Facelinidae (Opisthobranchia: Aeolidacea) from the Caribbean Sea SANDRA V. MILLEN Department of Zoology, University of British Columbia, Vancouver, B.C., Canada VOT 1Z4 JEFFREY C. HAMANN 475 W. Bradley Avenue, El Cajon, California 92020, USA Abstract. A new aeolid species has been found throughout the Caribbean Sea. It possesses the characteristics of the family Facelinidae and the subfamily Favorininae, but its reproductive structures differ from those of all other genera. This species is therefore placed in a new genus and compared to other genera in the subfamily. INTRODUCTION The tropical western Atlantic opisthobranch fauna is in- completely known, and relatively depauperate compared to the Indo-Pacific fauna. Nevertheless, almost 300 species have been described. The warm-water western Atlantic opisthobranch species are primarily endemic, but there are also many amphi-Atlantic species (EDMUNDS, 1968, 1977; JENSEN & CLARK, 1983; ORTEA et al., 1988; TEMPLADO et al., 1991). A number of Caribbean-Panamic species are recognized, and are thought to be representatives of a shared Pliocene fauna (Marcus & MARCUS, 1967; BERTSCH, 1979; GOSLINER & WILLAN, 1991). There are also circumtrop- ical representatives, possibly rernnants of the Tethys Sea fauna. Eveline and Ernst Marcus provided a sound basis for further opisthobranch research, summarized in their checklist of western Atlantic warm-water opisthobranchs (Marcus, 1977). Since then, several synonymies have been made and at least 26 new species have been added (CLARK & GOETZFRIED, 1976; EDMUNDs & JUST, 1983; GOSLINER, 1989; GOSLINER & ARMES, 1984; GOSLINER & GHISELIN, 1987; GOSLINER & KUZIRIAN, 1990; HAMANN & FARMER, 1988; Marcus, 1980, 1983; MEYER, 1977; NUTTALL, 1989; ORTEA & TEMPLADO, 1989; TEMPLADO ef al., 1987; THOMPSON, 1977, 1980). This paper describes a new aeo- lid genus and species belonging to the family Facelinidae. SYSTEMATIC TREATMENT ORDER AEOLIDACEA SUPERFAMILY CLEIOPROCTA FAMILY FACELINIDAE SUBFAMILY FAVORININAE Genus Pauleo Millen & Hamann, gen. nov. Generic diagnosis: Rhinophores smooth or weakly la- mellate. Cerata arranged on pedunculate arches with one or two rows per arch. Foot corners tentacular. Anus cleio- proctic. Nephroproct interhepatic. Salivary glands simple. Oral glands absent. Jaw elongate, notched, and possessing a flange. Masticatory border with single row of denticles. Radula uniseriate with large, cuspidate rachidian teeth. Tooth shape triangular with many short denticles. Re- productive opening below anterior arch of first ceratal cluster. Reproductive system androdiaulic with proximal receptaculum seminis. Vas deferens non-prostatic. Penis with internal glands and proximal, non-glandular pouch containing vas deferens. Penial opening subterminal, un- armed. Type species: Pauleo jubatus gen. & sp. nov., by original designation. Page 206 Pauleo jubatus Millen & Hamann, sp. nov. (Figures 1-8) Etymology: The numerous curly cerata that can bristle forward defensively suggested the generic and species names, which translate as “little lion’s mane.” Material: Holotype: California Academy of Sciences CA- SIZ 077998, 1 specimen, 22 mm long. Specimen collected by J. Hamann on 24 May 1986; Grand Bahama, Baha- mas, outer reef off Taino Beach (26°29'58”N, 78°36'45’W), at 20 m depth on a coral head on sandy substrate. Paratypes: CASIZ 077294, 6 specimens, up to 28 mm long. Collected by T. Gosliner on 6 May 1991; S.W. Point, Grand Cayman, B.W.I., at 25 m depth on the gorgonian Plexaurella dichotoma. United States National Museum of Natural History, USNM 860266, 1 specimen, 19 mm long. Specimen col- lected with the holotype. Other collecting localities: Port Antonio, Jamaica, 2 specimens, largest 31 mm, at 13 m depth. Collected by J. Hamann, 23 August 1990. Discovery Bay, Jamaica, 10 specimens. Collected by J. Hamann on 14 December 1990, at 8 m depth, grazing on downed Plexaurella dichotoma. Photographic records: Jackson Pt., Little Cayman, Cay- man Islands. Photographed by Nancy Sefton, specimen on a coral head on a sandy bottom. Bloody Bay, Little Cayman, photograph by Dr. Marc Chamberlain, November 1987, specimen on a patch reef at 10 m depth. Lighthouse Reef, Belize, photograph by Thomas M. Sullivan, December 1984, of a specimen near the edge of a sandy reef at 12 m depth. Eleuthera Island, Bahamas, photograph in COLIN (1978: 389), on a gorgonian at 20 m depth. Guanaja, Honduras, 42 mm specimen photographed by J. Hamann, 8 August 1991, at 20 m on Plexaurella. Systematic Description External morphology: This translucent aeolid is suf- fused with pale orange on the head, tentacles, rhinophores, cerata, and dorsum. Underlying the orange suffusion is one of a variety of food-derived colors: orange, yellow, tan, pinkish beige, or blue gray. The animal has an elongate, slender, rounded body and a long trailing tail (Figure 1). A median, opaque white stripe extends along the head and down the full length of the body, narrowing between cer- atal clusters. The line varies in width among specimens and may be interrupted. A small, opaque white crescentic patch is present on either side of the head just below the The Veliger, Vol. 35, No. 3 rhinophores. Live specimens measured up to 46 mm in length. The evenly tapered oral tentacles are slightly flat- tened. They are held anterolaterally and have a dorsal white line, wider near the base. The rhinophores are weak- ly lamellate with sloping bars of spicules overlaid with white pigment (Figure 1A, B) or smooth (Figure 1C, D). They are slender and evenly tapered to a fine tip. The rhinophores are much smaller in diameter and one-half the length of the oral tentacles. They end with a minute white tip. The head is rounded; the mouth a vertical slit (Figure 2). The foot has propodial tentacles that recurve inwardly and may be frosted with white coloration dor- sally. The anterior margin of the foot and the propodial tentacles are bilabiate, with a well-developed groove that is wider medially. The foot is slender, with a central groove and a conspicuous flange that is one-third wider than the body itself. The long, slender tail has a slight medial keel. The cerata arise from raised, arched cushions. There are up to eight clearly differentiated clusters of cerata per side (Figure 3). The pre-cardiac arches have a double row of cerata with alternating insertions and more cerata in the anteriormost limbs. The post-cardiac clusters are all in the form of arches with one row of cerata, except for the anterior limb of the first post-cardiac arch which has a double row. Within each arch, the cerata increase in size towards the center. The cerata are curved towards the midline of the body and are up to 6 mm in length in preserved specimens. When disturbed, the animal curls the cerata tips forward in a defensive posture (Figure 4). Each ceras is oblong in cross section and attached on its inner side. The translucent cerata are suffused with the food- derived color. There are small, opaque white tips on the cnidosacs followed by a translucent band through which the long cnidosacs can be seen. The distal one-third of each ceras has opaque white or bluish white spots on an orange ground. On most specimens, the middle one-third has a mottled white or bluish white band, which is more prom- inent on the anterior face of each ceras, on a background color that varies according to the food. The proximal one- third of each ceras lacks opaque white spots and shows the color of the digestive gland. The ceratal branches of the digestive gland are nodular. In each ceras a central branch with rounded lobules extends to the ceratal walls. These lobes are alveolar-like and have a honeycombed appearance. The cerata surfaces are smooth, but appear pustulate because numerous white pigment spots are lo- cated on the lobes. The long, functional cnidosac is lined with large glandular cells. The cerata are easily autoto- mized and move about for a short time after detachment. The genital apertures are located below the anterior Figure 1 Color variation in Pauleo jubatus gen. & sp. nov., Millen & Hamann. A. 32 mm specimen, Discovery Bay, Jamaica. B. 26 mm specimen, Discovery Bay, Jamaica. C. 30 mm specimen, Grand Bahama, Bahamas. D. 42 mm specimen on prey, Plexaurella dichotoma, Guanaja, Honduras. S. V. Millen & J. C. Hamann, 1992 Page 207 Page 208 — HA i mvs a a oy ase 4 4 Smm Figure 2 Pauleo jubatus, ventral view of head and foot. limb of the first ceratal arch; the anal opening is posterior, within the second arch; and the renal pore is located in front of the second arch, in the interhepatic space (Figure 3). Digestive system: The buccal mass has a circular, mus- cular, lip plate. The portion surrounding the mouth slit is convoluted and chitinized. The jaws (Figure 5) are covered by thin, unpigmented epithelium. They are pale golden brown in color, and rectangular, with a vertical flange for muscle attachment, and more posteriorly, a concave dorsal margin. The wing of each jaw is divided by a slight groove into an upper and a lower portion. The upper portion is smaller, lighter in color, and more convex than the lower portion (Figure 5A). The small masticatory process bears, near its tip, one row of approximately 10-18 small den- ticulations (Figure 5B). The radular formula is 20-23 (0-1-0). The teeth have an elongate, bluntly pointed cusp with 9-14 small denticles on each steeply sloping side. A long pair of posterior limbs forms a narrow arch (Figure 6A). In some animals, in- The Veliger, Vol. 35, No. 3 termediate denticles are present (Figure 7). In lateral view (Figure 6B), a prominent knob for articulation projects outward. The teeth are large: 200-460 um long and 77- 240 um wide. Oral glands are absent, and the small salivary glands are located dorsolaterally above the anterior portion of the stomach with long ducts inserting on the buccal bulb just above the buccal ganglia. The short, wide esophagus is lined with cuticle on the dorsal and lateral sides. The S-shaped stomach has three parts. The foreward portion is an elongate oval, curved slightly to the left, with longitudinal papillate striae. The smaller central portion arises dorsally and loops abruptly to the right through 90°, so its longitudinal striae are oriented transversely relative to the animal as a whole. Two anterior and one posterior hepatic ducts branch off this portion. The posterior portion of the stomach loops ventrally and to the left, narrowing gradually into the intestine. The latter runs longitudinally under the posterior hepatic duct and then abruptly bends dorsally and to the right to empty within the second ceratal arch. Each ceratal arch is served by one branch of the digestive gland. Central nervous system: The oval cerebropleural ganglia are completely fused and connected together by a short wide commissure. The rhinophoral ganglia are on long stalks. The eyes are on very short stalks: Small statocysts lie behind the optic ganglia. The oval pedal ganglia are joined to the larger cerebropleural ganglia by short com- missures, and to each other by a longer, wide, circumoe- sophageal commissure. The oval buccal ganglia lie beneath the oesophagus and are attached to each other by a short commissure. Reproductive system (Figure 8): The ovotestis form large, round, grape-like lobules with the female ancini peripheral to the male ancini. There is a long thin ovotestis duct. The duct widens into a sausage-shaped hermaphroditic am- pulla, which coils twice and then extends the length of the female gland mass, narrowing only slightly before bifur- cating into a narrow oviduct and a wide vas deferens. The vas deferens is highly muscular until it enters the penis. The vas deferens does not have a distinct prostatic portion, Figure 3 Right lateral view of Pauleo jubatus showing position of ceratal insertions. Key: a, anus; n, nephroproct; g, genital apertures. S. V. Millen & J. C. Hamann, 1992 Page 209 Figure 4 Pauleo jubatus showing defensive bristling of the cerata with the cnidosacs pointing anteriorly. although it is lined with a single layer of ciliated cuboidal cells, which are probably secretory in function. It loops and then enters a fibrous penial sac, which becomes wrin- kled and deflated looking when the penis is everted. When the penis is withdrawn, the sac envelops the entire vas Figure 5 A. Outer view of right jaw plate of Pauleo jubatus. Scale bar = 0.1 mm. B. Masticatory margin of jaw plate showing denticles. Scale bar = 100 um. deferens and upper portion of the penis. Inside the non- glandular penial sac, the muscular vas deferens enters the proximal portion of the penis. The penis is large, elongate, and flattened in cross section. In preserved specimens it is usually everted, with a small, slightly papillate sheath at its base. The vas deferens opens subterminally and the flattened tip is asymmetrical. The penis has highly mus- cular walls and an internal network of transverse muscle fibres, lacunae, and peripheral glandular cells which are larger on the posterior edge. The glandular cells become more numerous distally and occupy the entire tip of the penis beyond the opening of the vas deferens. No inde- pendent glandular opening was located and no internal duct entered the vas deferens. There is a common atrium for the female gland mass and the vagina ventral to the posterior portion of the male aperture. The vagina forms the anterior face of the common A Figure 6 A. Dorsal view of a radular tooth of Pauleo jubatus. B. Lateral view of a worn radular tooth showing denticles. Page 210 The Veliger, Vol. 35, No. 3 Figure 7 A. SEM view of the dorsal surface of an unused radular tooth of Pauleo jubatus. B. Dorsolateral view showing presence of intermediate denticles. Scale bar = 40 um. S. V. Millen & J. C. Hamann, 1992 duct and continues as a tubular oviduct to the junction with the seminal receptacle. The semiserially arranged receptacle is elongate and irregularly lobate, with a short stalk. The long, tubular oviduct joins the hermaphroditic duct. The female gland mass is composed of a highly con- voluted membrane gland, a small albumen gland, and a large mucous gland. The latter consists of two distinct lobes with the genital organs running down the groove between them. Ecology: Little is known about this species. All specimens have been collected or photographed at night. They are usually found on the side of coral heads, on sandy bottoms, in 8-20 m of water. The darkest specimens, with a bluish gray cast to the body and dark blue-gray ceratal cores, were found feeding on the gorgonian Plexaurella dichotoma, as were some pale pinkish-tan specimens (Figure 1B, D). Functional nematocysts are present in the cnidosacs. The animal varies in color, which probably comes from a varied diet. Most dissected specimens had empty stomachs or fragments of soft polyps, but one specimen was full of opisthobranch eggs, suggesting that they are opportunistic feeders. Their spawn mass has not been seen. DISCUSSION Pauleo jubatus can be distinguished from all other aeolids in the Caribbean by its large body size and the defensive bristling of its cerata. Phidiana lynceus, which is not as large, has some orange coloration, but it can be easily distinguished because the cerata are in rows and have one or two white bands on each ceras. The taxonomy of the aeolid family Facelinidae has re- cently undergone a number of rearrangements, summa- rized by EDMUNDs & JusT (1983) and WILLAN (1987). We favor their use of the family Facelinidae rather than the more all-encompassing Glaucidae proposed by MILLER (1974). The subfamily Favorininae is composed of genera whose cerata are arranged in arches, but this trait is con- sidered polyphyletic so the subfamilies should probably be abandoned (EDMUNDS, 1970; WILLAN, 1987; GOSLINER, 1991). Generic placements are also rather unsatisfactory at present because most genera are separated almost en- tirely on the basis of different penial structures. These differences have been reviewed by MILLER (1974). Since then, however, three new genera have been added, Her- mosita and Bajaeolis by GOSLINER & BEHRENS (1986) and Anetarca by GOSLINER (1991). This new genus Pauleo has both pre- and post-cardiac ceratal clusters arranged in arches on elevated cushions, some of which bear double rows of cerata. A number of genera have the synapomorphy of cerata in arches rather than rows, which is the basis of the polyphyletic subfamily Favorininae. Raised ceratal cushions appear to be a derived state because none of the plesiomorphic reproductive char- acteristics (two bursae, serial receptaculum seminis) occur in these genera. This parallels the situation in the family Page 211 Figure 8 Reproductive system of Pauleo jubatus drawn using a camera lucida. Key: al, albumen gland; ha, hermaphroditic ampulla; me, membrane gland; mu, mucous gland; 0, oviduct; ot, ovotestis duct; P, penis; ps, penial sac; rs, receptaculum seminis; sh, sheath; v, vagina; vd, vas deferens. Flabellinidae (GOSLINER & GRIFFITHS, 1981). Multiple insertions of cerata on the arches are a plesiomorphic fea- ture. All of the genera with raised, arch-shaped cushions are compared in Table 1, except for Amanda, Godiva, Nou- meaella, and Echinopsole, which have apomorphic penial armature. In addition to the arrangement of the cerata, other ex- ternal features sometimes used for generic diagnosis are the shape of the rhinophores and the position of the genital openings. Pauleo jubatus has smooth or lamellate rhino- phores and reproductive openings at the anterior limb of the first ceratal arch. These are both apomorphic condi- tions for the family (WILLAN, 1987) but because they occur in a number of genera in both subfamilies, they are not very helpful for comparative purposes. Internally, the jaw structure of Pauleo jubatus has a number of apomorphies. It has an indented dorsal margin, a dorsal flange, and an upper and lower convex division separated by a groove. Each of these traits can be found in other genera from both subfamilies in the Facelinidae. However, only the genera Dondice Marcus, 1958, Facalana Bergh, 1888, and Sakuraeolis Baba & Hamatani, 1965, and the species “Godiva” banyluensis Portmann & Sand- meier, 1960, share all three apomorphies (‘Table 1). Their jaws all show an intermediate modification towards the extreme type found in the Glaucidae. The jaws of Pauleo jubatus are more elongate than all of the others, and differ from those of Sakuraeolis by having only one row of den- Page 212 The Veliger, Vol. 35, No. 3 Table 1 Genera of Favorininae having multiple rows of cerata on raised arch-shaped pedicles, and having no penial armature. Jaw Genus Rhinophores Notch Flange Austraeolis annulate — + (=) Bajaeolis perfoliate + = Dondice annulate a ate Facalana perfoliate + + Jason papillate = = Pauleo smooth or ts te perfoliate Sakuraeolis smooth st ae “Godiva” banyluensis perfoliate + ate ticles and from those of “Godiva” banyluensis by lacking a flange-like guard. Variations of the basic facelinid radula may be adap- tations to different prey. The steep-sloping sides and many fine denticles on the radula of Pauleo jubatus are char- acteristic of the genus Phyllodesmium Ehrenberg, 1831. Pauleo jubatus was not placed in the genus Phyllodesmium because it has functional cnidosacs, the cerata are in arches on raised pads, and the large penis is highly muscular with internal glands. Most species of Phyllodesmium feed on alcyonacean octocorals and have symbiotic zooxanthellae, but P. horridum (Macnae, 1954) and P. serratum (Baba, 1949) feed on gorgonaceans (GOSLINER, 1987; RUDMAN, 1991). Phyllodesmium horridum has teeth that are almost iden- tical in shape to those of Pauleo jubatus, although they are proportionally much smaller—about one-half the size in similar sized animals. Externally, Phyllodesmium hor- ridum differs from Pauleo jubatus due to its wider, more depressed shape, longer cerata, which are not borne on raised pads, and less tapering ceratal tips, rhinophores, and oral tentacles. Internally, the jaw of Phyllodesmium horridum is not as elongate, and its masticatory denticles are less developed. Both have a two-part jaw with a flange, but in Phyllodesmium horridum the two parts are not sep- arated by a large notch and the smaller flange extends horizontally rather than vertically. The penial structures of the two species also differ greatly. Phyllodesmium ser- ratum has a thin-walled, prostatic vas deferens, which ter- minates in a very tiny, muscular penis tip. The vas deferens of Pauleo jubatus is shorter, narrower, and muscular with a large, muscular penis containing internal glands. Penial elaborations distinguish the various genera of Facelinidae. Table 1 compares species of the subfamily Favorininae that are most similar in ceratal arrangement to Pauleo and that do not have penial armature. The non- Masti- catory border rows Vas deferens Penis single proximal prostatic circle of fleshy filaments several all prostatic conical single non-prostatic basal gland, spiral groove, prostate gland single ? leaflike expansion, fleshy glandular knobs smooth all prostatic distal glands single non-prostatic basal sac, internal glands several all prostatic; 1 fleshy lobes, stalked cell layer thick accessory gland single proximal prostatic basal sac, conical penis glandular penial sac of Pauleo is found only in the species “Godiva” banyluensis (PORTMANN & SANDMEIER, 1960). This thoroughly described species has recently been re- moved from the genus Godiva because it lacks a penial spine (WILLAN, 1987). Previously it was removed from the original genus Dondice because it lacks a separate pros- tate in the penis and a basal penial gland (EDMUNDs, 1964). It is presently without a generic assignment. Al- though the sac around the vas deferens is like that in Pauleo, the rest of the male reproductive system is radically different. “Godiva” banyluensis has a long vas deferens that is prostatic before it enters the sac, but not after. Its conical penis does not contain internal glands. On the other hand, the elaborate stomach of this species, described by GARCIA & Garcia (1984), resembles that of Pauleo jubatus. Similar penial glands to those of Pauleo jubatus are found in the genera Dondice, Jason, and Sakuraeolis. Jason is separated on the basis of unusual jaw and radular mor- phology (MILLER, 1974) but Dondice and Sakuraeolis ap- pear to be closely allied to Pauleo. The penises of both of these latter species are large, with interlacing musculature, blood lacunae, and internal glands. Dondice occidentalis (Engel, 1925) has an accessory penial gland at the base of the penis with a duct extending distally and emptying into the vas deferens either near the penis tip (MARCUS, 1958) or part way along the penis (EDMUNDS, 1964). It also has a prostate gland located inside the penis, which opens independently at its tip. This latter gland is similar to the penial gland found in Pauleo, although no separate duct could be found in histological sections. Pauleo lacks the basal penial gland of Dondice, although it has a non- glandular sac in the same position. Sakureolis has a stalked accessory gland similar to the unstalked gland of Dondice. It does not have a separate prostate gland inside the penis, but the vas deferens is reported to be prostatic throughout its length, even though S. V. Millen & J. C. Hamann, 1992 it consists of only a single cell layer (BABA & HAMATANI, 1965). It appears that the vas deferens of Pauleo is similar. The penis of Sakuraeolis has fleshy flaps at the base of the penial gland, which are not found in either Pauleo or Dondice. The genital openings of both Sakuraeolis and Dondice are more posterior than those of Pauleo, and their teeth are different. In conclusion, although Pauleo is clearly allied to a number of genera in the Favorininae, it possesses a unique combination of characteristics that do not comfortably fit into any other generic diagnosis. The newly created genus Pauleo presently contains only the type species, Pauleo jubatus. ACKNOWLEDGMENTS We would like to thank Ben Rose and Jack Worsfold for pointing out this new species to J.H. We would also like to thank Wes Farmer for his illustration used in Figure 4. We are grateful to all those who provided us with photographic records. This research was partially funded by the Department of Zoology, University of British Co- lumbia. LITERATURE CITED Basa, K. & I. HAMATANI. 1965. The anatomy of Sakuraeolis enosimensis (Baba, 1930), n.g. (= Hervia ceylonica(?) Eliot, 1913) (Nudibranchia—Eolidoidea). Publications of the Seto Marine Biological Laboratory 13:103-113. BertscH, H. 1979. Tropical faunal affinities of opisthobranchs from the Panamic Province (eastern Pacific). Nautilus 93: 57-61. CLARK, K. B. & A. GOETZFRIED. 1976. Lomanotus stauberi, a new dendronotacean nudibranch from central Florida (Mol- lusca: Opisthobranchia). Bulletin of Marine Science of the Gulf and Caribbean 26:474-478. Coin, P. L. 1978. Marine Invertebrates and Plants of the Living Reef. T.F.H. Publications: Neptune City, New Jer- sey. 512 pp. Epmunps, M. 1964. Eolid Mollusca from Jamaica, with de- scriptions of two new genera and three new species. Bulletin of Marine Science of the Gulf and Caribbean 14:1-32. EpMunDs, M. 1968. Eolid Mollusca from Ghana, with further details on west Atlantic species. Bulletin of Marine Science of the Gulf and Caribbean 18:203-219. EpmunpDs, M. 1970. Opisthobranchiate Mollusca from Tan- zania. II. Eolidacea (Cuthonidae, Piseinotecidae and Face- linidae). Proceedings of the Malacological Society of London 39:15-57. Epmunps, M. 1977. Larval development, ocean currents, and origins of the opisthobranch fauna of Ghana. Journal of Molluscan Studies 43:301-308. Epmunps, M. & H. Just. 1983. Eolid nudibranchiate Mol- lusca from Barbados. Journal of Molluscan Studies 49:185- 203. Garcia, J. C. & F. J. Garcia. 1984. Estudio anatomico y algunas resenas ecologicas de Godiva banyulensis (Portmann y Sandmeier) (Gastropoda: Nudibranchia). Cahiers de Biol- ogie Marine 25:49-65. GOsLINER, T. M. 1987. Biogeography of the opisthobranch gastropod fauna of southern Africa. American Malacological Bulletin 5:243-258. Page 213 GosLINER, T. M. 1989. Revision of the Gastropteridae (Opis- thobranchia: Cephalaspidea) with descriptions of a new ge- nus and six new species. The Veliger 32:333-381. GoSLINER, T. M. 1991. Four new species and a new genus of opisthobranch gastropods from the Pacific coast of North America. The Veliger 34:272-290. GosLinER, T. M. & P. T. ARMEs. 1984. A new species of Gastropteron from Florida (Gastropoda: Opisthobranchia). The Veliger 27:54—64. GOSLINER, T. M. & D. W. BEHRENS. 1986. ‘Two new species and genera of aeolid nudibranchs from the tropical eastern Pacific. The Veliger 29:101-113. GosLINER, T. M. & M. T. GHISELIN. 1987. A new species of Tritonia (Opisthobranchia: Gastropoda) from the Caribbean Sea. Bulletin of Marine Science of the Gulf and Caribbean 40:428-436. GosLINER, T. M. & R. J. GRIFFITHS. 1981. Description and revision of some South African aeolidacean Nudibranchia (Mollusca, Gastropoda). Annals of the South African Mu- seum 84:105-150. GOSLINER, T. M. & A. M. Kuzirian. 1990. Two new species of Flabellinidae (Opisthobranchia: Aeolidacea) from Baja California. Proceedings of the California Academy of Sci- ences 47:1-15. GOSLINER, T. M. & R. C. WILLAN. 1991. Review of the Fla- bellinidae (Nudibranchia: Aeolidacea) from the tropical Indo- Pacific, with the descriptions of five new species. The Veliger 34:97-133. Hamann, J. C. & W. M. FARMER. 1988. Two new species of Plocamopherus from the western warm-water Atlantic. The Veliger 31:68-74. JENSEN, K. H. & K. B. CLARK. 1983. Annotated checklist of Florida ascoglossan Opisthobranchia. Nautilus 97:1-13. Marcus, ER. 1958. On western Atlantic opisthobranchiate gastropods. American Museum Novitates 1906:1-82. Marcus, Ev. 1977. An annotated checklist of the western At- lantic warm water Opisthobranchs. Journal of Molluscan Studies 4(Suppl.):1-22. Marcus, Ev. 1980. Review of western Atlantic Elysiidae (Opisthobranchia Ascoglossa) with a description of a new Elysia species. Bulletin of Marine Science of the Gulf and Caribbean 30:54-79. Marcus, Ev. 1983. The western Atlantic Tritonidae. Boletim de Zoologia, Universidade de Sao Paulo 6:177-214. Marcus, Ev. & Er. Marcus. 1967. Opisthobranchs from the Gulf of California. Studies in Tropical Oceanography Mi- ami 6:141-248. MEYER, K. B. 1977. Dorid nudibranchs of the Caribbean coast of the Panama Canal Zone. Bulletin of Marine Science of the Gulf and Caribbean 27:299-307. Miiier, M. C. 1974. Aeolid nudibranchs (Gastropoda: Opis- thobranchia) of the family Glaucidae from New Zealand waters. Zoological Journal of the Linnean Society of London 54:31-61. NuTTALL, T. R. 1989. A new Elysia (Opisthobranchia: As- coglossa) from the Florida Keys. The Veliger 32:302-307. OrTEA, J., A. A. LUQUE & J. TEMPLADO. 1988. Elysia picta Verrill, 1901 and Geztodoris pusae (Marcus, 1955), two am- phiatlantic opisthobranch gastropods. Journal of Molluscan Studies 54:243-247. OrTEA, J. & J. TEMPLADO. 1989. A new species of Cyerce Bergh, 1871 (Opisthobranchia: Ascoglossa) from Cuba [West Indies]. Iberus 8:11-14. PORTMANN, A. & E. SANDMEIER. 1960. Dondice banyulensis, sp. nov., un eolidien nouveau de la Méditeranée. Revue suisse de Zoologie 67:159-168. Page 214 RupMAN, W. B. 1991. Further studies on the taxonomy and biology of the octocoral-feeding genus Phyllodesmium Eh- renberg, 1831 (Nudibranchia: Aeolidoidea). Journal of Mol- luscan Studies 57:167-203. TEMPLADO, J., A. A. LUQUE & J. ORTEA. 1987. A new species of Aegires Loven, 1844 (Opisthobranchia: Doridacea: Ae- giretidae) from the Caribbean Sea: Aegires ortizi spec. nov., with comparative descriptions of the north Atlantic species of this genus. The Veliger 29:303-307. TEMPLADO, J., A. A. LUQUE & J. A. ORTEA. 1991. A com- mented checklist of the amphiatlantic Ascoglossa and Nu- The Veliger, Vol. 35, No. 3 dibranchia (Mollusca—Opisthobranchia). Lauri de la So- cieta Italiana di Malacologia 23:295-326. THompson, T. E. 1977. Jamaican opisthobranch molluscs:I. Journal of Molluscan Studies 43:93-140. TuHompson, T. E. 1980. Jamaican opisthobranch molluscs:II. Journal of Molluscan Studies 46:74-99, WILLAN, R. C. 1987. Phylogenetic systematics and zoogeog- raphy of Australian nudibranchs. 1. Presence of the aeolid Godwa quadricolor (Barnard) in Western Australia. Journal of the Malacological Society of Australia 8:71-85. The Veliger 35(3):215-221 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 A Warm Water Atlantic Synonymy, Aphelodoris antillensis Equals Chromodoris bistellata (Opisthobranchia: Gastropoda) JEFFREY C. HAMANN 475 W. Bradley Avenue, El Cajon, California 92020, USA Abstract. Study of the literature and comparison of specimens shows Chromodoris bistellata (Verrill, 1900), a species previously thought to be endemic to Bermuda, to be a junior synonym of Aphelodoris antillensis Bergh, 1879, a species known from throughout the Caribbean. INTRODUCTION Specimens collected over large geographic distances with minor differences in external appearance or internal anat- omy have often given rise to two or more separate species names. Language barriers and incomplete descriptions make resolution difficult without later works to add to and append the original descriptions and/or a large body of collected material to work with. Both Chromodoris bistellata (Verrill, 1900) and Aphelo- doris antillensis Bergh, 1879, were recognized by later workers, who added important details and extended geo- graphic ranges (see review of literature below). From May 1983 until June 1988, I collected over 25 specimens of Aphelodoris antillensis from locations ranging from the Dominican Republic to Venezuela and including the type locality, the Virgin Islands. I became familiar with the range of morphological variation of Aphelodoris antillensis in the Caribbean, making it immediately ap- parent upon examination of Chromodoris bistellata from Bermuda that they were likely to be conspecific. That opportunity presented itself in July 1988 when I visited Dr. Kerry Clark and Dr. Duane DeFreese in Bermuda. REVIEW oF LITERATURE Aphelodoris antillensis Bergh, 1879 BERGH’s (1879) detailed original description, in Ger- man, was done from preserved material collected in St. Thomas, Virgin Islands. Although comprehensive in con- tent, there were no figures or description of live specimens. Marcus & Marcus (1963) were the first to recognize Bergh’s species in print. They reported conforming exter- nal features, and figured the dorsal view, head, and radula of one specimen from Curacao. The inner denticle on the innermost lateral tooth reported by Bergh was missing but not considered of systematic importance. Internal features were not presented owing to the condition of the preserved material. They later provided figures of the reproductive organs and recorded variations in the color patterns of a specimen collected in Florida (Marcus & Marcus, 1967). Marcus & Marcus (1970) also reported three specimens from Puerto Rico. MEYER (1977) recorded 10 specimens from Panama and reported low tubercles on the notum that disappear upon preservation, accounting for their ab- sence from earlier descriptions. ‘THOMPSON (1980) re- ported two specimens from Jamaica. He noted the swim- ming/escape reaction also reported by GOSLINER (1987) for Aphelodoris brunnea Bergh, 1907. Thompson reported a yellow band on the notal border and confirmed the tu- bercles reported by Meyer. EDMUNDS & JUST (1985) col- lected three specimens in Barbados. They noted further minor variations in color and figured the spawn. Chromodoris bistellata (Verrill, 1900) VERRILL (1900) described Doris bistellata in 74 words. No internal features were reviewed and there were no figures. CLARK (1984) changed the genus to Chromodoris on the basis of the dentition, and figured the teeth. He also published the first photograph in Marine Fauna and Flora of Bermuda (JENSEN & CLARK, 1985). TAXONOMIC TREATMENT Material examined: Aphelodoris antillensis: Malmok, Aruba, 1 specimen, 20 mm, 2 m, December 1980; *Sa- * Specimens deposited in the collection of the California Acad- emy of Sciences. Page 216 mana, Dominican Republic, 6 specimens, 12-28 mm, 1 m, February 1984, CASIZ 075624, one specimen; La Par- guera, Puerto Rico, 2 specimens, to 19 mm, | m, May 1984; Cayo Icacos, Puerto Rico, 1 specimen, 14 mm, 1 m, September 1984; Puerto Manglar, Culebra, 1 specimen, 1 m, December 1984; Lamshur Bay, U.S. Virgin Is., 6 specimens, to 29 mm, 1 m, March 1985; Green Key, Brit- ish Virgin Is., 1 specimen, 17 mm, 1 m, June 1985; Indian Creek, Antigua, 1 specimen, 11 mm, 1 m, January 1986; Deshaies, Guadeloupe, 1 specimen, 19 mm, 2 m, January 1986; Trou Madame, Martinique, 1 specimen, 16 mm, 1 m, June 1986; Pt. Borgnesse, Martinique, 1 specimen, 18 mm, 7 m, June 1986; Piton Bay, St. Lucia, 1 specimen, 9mm, 1 m October 1986; Bequia, St. Vincent, 1 specimen, 10 mm, 1 m, January 1987; Young Island, St. Vincent, 2 specimens, to 19 mm, January 1987; Clarkes Court Bay, Grenada, 1 specimen, 19 mm, 1 m, December 1987; *Chi- mana Grande, Venezuela, 2 specimens, 22 mm and 25 mm, 1 m, July 1989, CASIZ 075638; *Margarita, Ven- ezuela, 1 specimen, 16 mm, 1 m, October 1989, CASIZ 075623; *Grand Cayman Island, British West Indies, 1 specimen, 18 mm, 1.5 m, May 1991, CASIZ 077289. All collections were made by the author. Chromodoris bistellata: *South Ferry Point, Bermuda, 3 specimens, | to9 mm, 2 m, July 1988, collected by Duane DeFreese, CASIZ 075622. Synonymy: Comparison of the specimens of Chromodoris bistellata from Bermuda with the previously described fea- tures of Aphelodoris antillensis and with a varied spectrum of preserved material of same confirm beyond reasonable doubt that they are in fact conspecific. There are no pub- lished or observed characteristics of Chromodoris bistellata that cannot be accommodated within the range of vari- ability of Aphelodoris antillensis. Owing to the priority of Bergh’s name, C. bistellata now becomes a junior synonym of A. antillensis. External morphology: The base color is a translucent beige with widely varying amounts of yellow, brown, and white pigment, the yellow being totally absent in many individuals (Figure 1). The brown pigment spots are con- centrated to form larger blotches on some individuals while others have an almost even density. Still others have such a dense concentration that they appear to have a brown base color. The yellow and white pigments are similarly sprinkled and/or concentrated over the dorsal surface in various degrees. Occasional specimens have brown and/ or yellow dashes, formed by the pigment spots, perpen- dicular to the notal edge. Concentrations of white or brown Ihe Veliger, Vol. 355 Noms often form two more distinct blotches on the dorsum which, in the case of white, gave rise to the name bistellata or “two-starred” (Figure 2). The notal border can be rimmed in yellow but is generally unmarked. MEYER (1977) and Marcus & Marcus (1963) report an orange border. EDMUNDS & JUST (1985) report patches of wine-red color on the dorsal surface. The body is 6 to 8 mm wide on a 20 mm individual and 4 mm high. Individuals tend to be plastic in shape, with some individuals at rest assuming a flatter, wider, body form. The dorsum is generally covered with low tubercles, independent of the color pattern, but these do not extend onto the notal margins. The body is very soft and smooth to the touch. The narrow foot (2 to 4 mm on a 20 mm specimen) is well covered by the mantle and is the same width over most of its length. In front and back the foot is similarly rounded, with the back being slightly smaller. Only the posterior tip of the foot, which is similarly marked as the dorsum, is sometimes visible. Smaller brown and white pigment spots generally fleck the bottom of the foot and to a greater extent the hyponotum. The mouth is round and grooved longitudinally. The tentacles, which BERGH (1879) reported as being very short and cut off, are typical in a live specimen (Figure 2) but shrunken in preserved material, as reported by MARCUS & Marcus (1963). THOMPSON (1980) reported that al- though the tentacles were grooved or in-rolled in life their character was undetectable in his preserved material (Fig- ure 3). The rhinophores are typically shaped and strongly la- mellate. A 20 mm specimen typically has 10-14 obliquely set lamellae sprinkled with brown and white pigment (Fig- ure 4). BERGH (1879) reported 40 lamellae in his material from St. Thomas but my specimens (10-29 mm) from the type locality have 8-12 only. Other material, as well as all other reported data, support a much reduced number of lamellae. Bergh’s specimen was 20 mm preserved and could well have been 40 mm or more in life, which could explain the high number of lamellae. The sheaths are smooth-rimmed and stand almost 1 mm high on a 20 mm specimen. Bergh reported that they are stiffened with thin brown calcified spicules (0.007—0.01 mm in cross section). The simple branchial plume has five bipinnate gills that are sprinkled with brown and white pigment in concen- trations parallel to the individual’s body coloring. The gills are irregularly pinnate rather than featherlike. They also tend to have a more inflated center rib than most chro- modorids. THOMPSON (1980) described them as thick and Figure 1 Aphelodoris antillensis Bergh, 1879. A. 17 mm specimen from British Virgin Islands. B. 14 mm specimen from Puerto Rico. C. 20 mm specimen from Aruba. D. 16 mm specimen from Martinique. E. 19 mm specimen from St. Vincent. F. 25 mm specimen from Venezuela. G. 9 mm specimen from Bermuda. H. 25 mm specimen from Grand Cayman Island. Photographs by Jeff Hamann. Page 217 J. G. Hamann, 1992 Page 218 The Veliger, Vol. 35, No. Figure 2 Aphelodoris antillensis. A. Concentrations of white pigment forming two distinct spots on the dorsum. B. Brown pigment forming larger blotches on the dorsum. C. Note typically chromodorid tentacles and thick and fleshy gills. fleshy (Figure 2). The branchial plume can be withdrawn the posterior mantle edge on a 20 mm specimen. The short into a round branchial pit with a low smooth rim and no (0.5-1 mm) round anal papilla is in the center of the gill prominent markings. The plume is positioned farther back circlet. on the dorsum than on most chromodorids, 2-3 mm from EDMUNDS & JUST (1985) reported that the spawn rib- J. CG. Hamann, 1992 Page 219 Figure 3 Head region of a preserved specimen of Aphelodoris antillensis. Key: T, contracted tentacles; M, mouth. Scale = 5 mm. bon is two eggs thick and makes four and a half spiral turns. MARCUS & Marcus (1970) recorded about 320 eggs in a spiral spawn from Puerto Rico. In spite of the wide variations in colors, patterns, and body shape, specimens of Aphelodoris antillensis are easily recognized. The position of the branchial plume, softness of the body, and color variation within certain limits con- tribute to its widespread recognition. EDMUNDS & JUST (1985) found it unnecessary to examine the radula because “this species is easily recognized from its external fea- tures.” Once some familiarity with the species is estab- lished the variations are quickly recognized. Internal morphology: The jaw plates are constructed of a thick smooth cuticle. BERGH (1879) reported, and I con- firm, that the radula is yellow. The radular formula varies widely with size and among individuals, with 30-52 rows and 30-68 teeth in a half row (Figure 5, Table 1). BERTSCH (1976) examined the variability of the radula in Discodoris evelinae and found 25-69 teeth per half row in 16-35 rows. Although he found positive correlations between size and both the number of teeth per row and the number of rows, he concluded that the number of teeth alone should not be used to establish the validity of a new species. The teeth are all simple hooks, except the outermost two or three, Figure 4 Rhinophore of Aphelodoris antillensis. Scale = 3 mm. Figure 5 Radula of 16 mm Aphelodoris antillensis from Venezuela. A. Lat- eral view. B. Flattened dorsal view. Scale = 0.5 mm. which can develop a projection on top of the hook. The middle teeth of each half row are the largest (Figure 6). BERGH (1879) reported that the innermost tooth had small denticles at its base. None of my specimens had any den- ticles and no other author has confirmed this characteristic. The large size of Bergh’s specimen could explain their presence. Marcus & Marcus (1963) did not consider this absence to be of systematic significance because of the similar confirmed variability of denticles in Discodoris pus- ae Marcus, 1955. The 28 mm specimen of Aphelodoris antillensis from Samana, Dominican Republic, was dissected and com- pared to the 9 mm specimen from South Ferry Point. The reproductive systems were identical in arrangement and were in accordance with the description by MARcus & Marcus (1967) with the exception of the ampullary con- nection, which they showed near the oviduct. The impor- tant features also conformed to the description by BERGH (1879). The genital mass is forward on the right side (Figure 7). A short sperm duct exits the female gland mass near the uterine duct. It connects to the prostate, which is com- posed of many sacs. The remaining vas deferens is con- Table 1 Recorded radular variations in Aphelodoris antillensis. Radular formulas Record Location Size Formula BERGH, 1879 Virgin Islands 20 mmt 52 x 68-0:68 Marcus, 1967 Florida 10 mmf 38 x 67:0-67 Marcus, 1970 Puerto Rico 15mm _ 42 X 66-:0-66 THOMPSON, 1980 Jamaica 16mm _ 36 X 58-0-58 MEYER, 1977 Panama ? 35 x 55:0-55 Marcus, 1963 Curacao 15mm _ 30 x 52-0-52 CLARK, 1984 Bermuda 5mmt 30 x 36-0-36 Present study Bermuda 9mm 32 x 56:0-56 Present study Venezuela 16mm _ 30 X 43-0-43 + Preserved. Page 220 ip 18 29. 43 Figure 6 Teeth of Aphelodoris antillensis from Venezuela. Scale = 100 um. voluted and about the same length as the prostatic portion. The penial sheath is muscular and the penis is unarmed. The bulb-shaped vagina enters a spherical copulatory bursa of approximately equal size. The much longer distal vaginal duct then connects to the pear-shaped seminal receptacle, which is again about the same size. The uterine duct doubles back about half the length of the distal vaginal duct and then branches off to the female gland mass con- necting near the oviduct. The long cylindrical ampulla enters the female gland mass at a location about half the diameter away from the uterine duct. Discussion of the genus: Aphelodoris antillensis is the type species for the genus. Numerous later workers have rec- ognized and/or used the genus, paradoxically all in south- ern temperate waters of Australia and South Africa (e.g., ODHNER, 1924; ELIOT, 1907; BURN, 1966; WILLAN & COLEMAN 1984; GOSLINER, 1987). Aphelodoris was characterized by BERGH (1879) as very close to the Chromodoridae externally, but differing in the short hacked-off tentacles with a groove on the underside, the five plurally pinnate gill leaves, and the much smaller foot. Internally the rhinophores are stiffened with spicules, the jawplates are unarmed, the rachis is naked, the prostate is large, and the penis is unarmed. I found the tentacles to be typically chromodorid in live material, and this opinion was voiced by Marcus & Marcus (1963) as well. Additionally, the gill is positioned farther back on the dorsum than on the typical chromo- dorid, and some species exhibit a head to tail swimming response to stimulation. The body is also soft and smooth to the touch. BuRN (1966) characterized the genus as follows: “glos- The Veliger, Vol. 35, No. 3 Figure 7 Reproductive system of Aphelodoris antillensis. Key: A, ampulla; FG, female gland mass; I, insemination duct; P, penis; Pr, pro- static portion of vas deferens; S, copulatory bursa; SC, seminal receptacle; V, vagina; VD, vas deferens. Scale = 2 mm. sodoridiform” or high, slender, elongate body shape with usually narrow notal brim, smooth notum, high conical rhinophoral and branchial sheaths, laterally grooved oral tentacles, five-branched branchia, smooth labium, hook- shaped radular teeth without denticles, unarmed penial sheath, large prostatic part in male duct, and spermatheca and spermatocyst arranged serially (ODHNER, 1924) or semiserially. ACKNOWLEDGMENTS Many thanks to Dr. Kerry Clark for the opportunity to collect with him in Bermuda and for his critical review of the manuscript. I am also grateful to Wes Farmer for his help with the manuscript and most of the drawings. LITERATURE CITED BERGH, L.S.R. 1879. Die Doriopsen des atlantischen Meeres. Deutsche Malakozoologische Gesellschaft Jahrbucher 6:42- 64. BERTSCH, H. 1976. Intraspecific and ontogenetic radular vari- ation in opisthobranch systematics. Systematic Zoology 25(2): 117-122. Burn, R. F. 1966. Notes on some opisthobranchs mainly from South Australia. Records of the South Australian Museum 15(2):329-352. CxiaRK, K. B. 1984. New records and synomymies of Bermuda opisthobranchs. Nautilus 98(2):85-97. Epmunpbs, M. & H. Just. 1985. Dorid, dendronotoid and arminid nudibranchiate Mollusca from Barbados. Journal of Molluscan Studies 51(1):52-63. EvioT, C. N. E. 1907. Nudibranchs from New Zealand and the Falkland Islands. Proceedings of the Malacological So- ciety of London 7(6):327-361, pl. 28. GOSLINER, T. B. 1987. Nudibranchs of Southern Africa. Sea Challengers: Monterey, California. 136 pp. JENSEN, R. H. & K. B. Ciark. 1985. Gastropoda. Pp. 397- 458. In: W. Sterrer (ed.), Marine Fauna and Flora of Ber- muda. John Wiley & Sons: New York. Marcus, ER. & Ev. Marcus. 1970. Opisthobranchs from J. CG. Hamann, 1992 Curacao and faunistically related regions. Studies on the Fauna of Curagao and Other Caribbean Islands 33:1-129, 160 figs. Marcus, Ev. & Er. Marcus. 1963. Opisthobranchs from the Lesser Antilles. Studies on the Fauna of Curacao and Other Caribbean Islands 19(79):1-76. Marcus, Ev. & Er. Marcus. 1967. American opisthobranch mollusks. Studies in Tropical Oceanography. University of Miami 6:1-256. MEYER, K. B. 1977. Dorid nudibranchs of the Caribbean coast of the Panama Canal Zone. Bulletin of Marine Science 27(2):299-307. Page 221 ODHNER, N.H. 1924. Papers from Dr. Th. Mortensen’s Pacific expedition 1914-16. New Zealand Mollusca. Videnskabe- lige Meddeleser fra Dansk Naturhistorisk Forening 77:1- 90, pls. 1-2. TuHompeson, T. E. 1980. Jamaican opisthobranch molluscs: II. Journal of Molluscan Studies 46(1):74-99. VERRILL, A. E. 1990. The nudibranchs and naked tectibranchs of the Bermudas. Transactions of the Connecticut Academy Arts & Sciences 11(1):15-62, pls. 1-9. WILLAN, R. C. & N. COLEMAN. 1984. Nudibranchs of Aus- tralasia. Australasian Marine Photographic Index: 103 Ca- ringbah, N.S.W., Sydney, Australia. 56 pp. The Veliger 35(3):222-225 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 A New Genus and Species of Polygyrid Land Snail (Gastropoda: Pulmonata) from Oregon by BARRY ROTH anp WALTER B. MILLER Department of Invertebrate Zoology, Santa Barbara Museum of Natural History, Santa Barbara, California 93105, USA Abstract. A new genus and species of polygyrid land snail, Hochbergellus hirsutus, is described from Sisters Rocks, Curry County, Oregon. Hochbergellus resembles Vespericola Pilsbry, 1939, but the penial chamber lacks a verge at the summit and has several internal pilasters that converge subapically to form a fleshy, anteriorly directed protuberance. The penial chamber lacks the prominent, paired dorsal pilasters found in Cryptomastix Pilsbry, 1939. INTRODUCTION The new taxon of land snail described here was collected during the course of a study by the authors (Roth & Miller, manuscripts in preparation) of the West American poly- gyrid land snail genus Vespericola Pilsbry, 1939. The spe- cies was first known to us from shells collected by the junior author in 1952 and subsequently by the late Robert R. Talmadge sometime before 1975. The junior author secured living specimens in 1991. On dissection, the species was found to lack the verge at the summit of the penial chamber that is diagnostic of Vespericola, but instead to bear longitudinal pilasters that fuse subapically into a fleshy, anteriorly directed protuberance. A consideration of anatomical relations among the new taxon, Vespericola, and Cryptomastix Pilsbry, 1939, leads us to propose a new genus for it. MATERIALS ano METHODS Shell height and diameter are vernier caliper measure- ments and exclude the expanded lip of mature shells. Whorls were counted by the method of PILsBRy (1939:xi, fig. B). The density of periostracal setae was estimated by counting the number of setae per square millimeter on the shoulder of the last 0.25 whorl of adult specimens, at 10x mag- nification under a dissecting microscope with an ocular reticle. Three counts were taken per specimen and the mean (to the nearest integer) recorded. Specimens for dissection were prepared by the method of MILLER (1967). Snails were first drowned in water to insure expansion and relaxation, then heated to a tem- perature of 60°C, at which time the bodies could be pulled easily from the shells and dissected. The penis was slit longitudinally to expose the structures on the wall of the penial chamber. Whole mounts of genitalia were prepared by the method of MILLER (1967): stained with hematoxylin, dehydrated and cleared in successive baths of ethanol and toluene, and mounted on slides with Permount mounting medium. Or- gan measurements were taken from mounted specimens. Anatomical drawings were made by projecting the image of the whole mount on paper with an overhead projector. The following abbreviations are used: ANSP, Academy of Natural Sciences of Philadelphia; BR, senior author’s collection, San Francisco, California; CAS, California Academy of Sciences; LACM, Los Angeles County Mu- seum of Natural History; SBMNH, Santa Barbara Mu- seum of Natural History; and USNM, National Museum of Natural History, Smithsonian Institution. SYSTEMATICS Family Polygyridae Pilsbry, 1895 Hochbergellus Roth & Miller, gen. nov. Type species: Hochbergellus hirsutus sp. nov. Polygyridae in which the shell is medium-sized, de- pressed-helicoid to conical, and narrowly umbilicate. The periostracum is matte-surfaced and bears rather sparsely set setae in diagonal rows. The base of the last 0.2 turn of the body whorl is compressed upward; a strong con- striction is present behind the lip. A small parietal lamella is present. The lip is turned outward and reflected. No lamellae are present on the outer or basal lips, but the basal lip is thickened by a low callus. The epiphallus consists of a relatively thick upper section and a narrower lower section, markedly narrower than the apex of the penis. There is a completely enclosed, vestigial epiphallic caecum at the junction of epiphallus and vas deferens. A verge is absent. The upper cavity of the penis bears smooth B. Roth & W. B. Miller, 1992 Page 223 Explanation of Figures 1 to 3 Figures 1-3. Hochbergellus hirsutus Roth & Miller, gen. et sp. nov. Shell, holotype, SBMNH 35554, OREGON: Curry County: Sisters Rocks, ca. 3.8 km N of Euchre Creek at Ophir, W. B. Miller coll., 16 July 1991. Diameter 16.8 mm. to papillose, longitudinal pilasters; four or more dorsal pilasters converge subapically to form a fleshy, anteriorly directed protuberance. There are no diagnostic shell characters that distinguish Hochbergellus at the generic level from other polygyrid genera such as Vespericola and Cryptomastix. The structure and organization of the reproductive sys- tem is much like that of Vespericola, except that Vespericola has a verge through which the seminal duct opens, either terminally or subterminally, into the penial chamber. A fleshy subapical protuberance in the upper penial chamber is not known to occur in Vespericola. The epiphallus is similar in Hochbergellus and Vespericola in consisting of a slender lower section tapering to a thicker upper section. In both genera there is a short, concealed epiphallic caecum at the junction of epiphallus and vas deferens. The sper- mathecal duct is relatively longer in Hochbergellus than in any Vespericola thus far examined. The reproductive system is also similar to that of Cryp- tomastix, except that in Cryptomastix the upper part of the penial chamber bears a pair of large, contiguous, dorsal pilasters, with the seminal duct opening between their upper ends. The duct of the spermatheca is commonly thicker than in Hochbergellus. Etymology: The genus is named for F. G. Hochberg, Jr., Curator of Invertebrate Zoology, Santa Barbara Museum of Natural History, who has keenly and consistently sup- ported our studies of west American land mollusks. Hochbergellus hirsutus Roth & Miller, sp. nov. (Figures 1-6) Diagnosis: A medium-sized polygyrid with depressed-he- licoid to conical, narrowly umbilicate shell, 5.7—6.6 whorls, erect, distant periostracal setae, and usually a small pa- rietal lamella. Penis elongate-conical, mostly enclosed in sheath; with subapical, anteriorly directed, fleshy protu- berance formed by fusion of four or more longitudinal pilasters; spermathecal duct long, slender, cylindrical; spermatheca ovate. Description: Shell (Figures 1-3) medium-sized for the family, depressed-helicoid to conical, narrowly umbilicate, of 5.7-6.6 whorls; base inflated. Spire broadly conic, its sides straight or weakly convex; whorls flattened, suture weakly to moderately impressed. Embryonic whorls 1.5- 1.8, sculptured with coarse, irregularly spaced papillae in diagonal trends and low, crowded, more or less granulose, radiating rugae, strongest below suture. Early teleoconch whorls with irregular, convex-forward retractive growth rugae and distant, erect or curved, acicular setae in pro- tractive, descending rows; 2-3 setae/mm?, approximately 0.7 mm long on spire and body whorl, broad at base, some with basal furcae pointing aperturally, many with finlike basal extension abaperturally. Periostracum between setae radially wrinkled, pebbly to scaly on first four whorls, smoother on whorls five and six, sometimes with a few raised spiral lirae on shoulder of body whorl. Periphery weakly subangulate, grading to rounded on last 0.5 whorl. Base regularly setose, setae smaller than on spire, extend- ing into umbilicus. Last whorl not markedly descending, constricted behind lip. Aperture broadly auriculate, peri- stome shallowly concave in profile, oblique, at angle of about 30° to vertical; lip expanded and reflected, moder- ately thickened submarginally, most strongly turned back- ward at base. Inner lip reflected over narrow umbilicus. Basal lip thickened by low ridge of callus that may reach prominence of a tubercle on inner quadrant. Parietal callus granulose, free edge weakly convex, with shallow re-en- trant below upper limb of peristome. Short, white, straight or upwardly convex parietal lamella usually present, set on upper third of parietal callus, somewhat back from line between upper and lower limbs of peristome. Shell tan, peristome white to pinkish tan. Dimensions of holotype: Diameter (exclusive of expand- ed lip) 16.8 mm, height 11.1 mm, whorls 6.3. Page 224 & af The Veliger, Vol. 35, No. 3 Soft anatomy: The color of the living animal is tan along the foot, darker and grayer on the body-stalk. The mantle over the lung is clear buff, 10-40% maculated with black. The holotype and nine paratypes were dissected. The atrium (Figure 4) is of moderate length. The penis is elongate-conical, somewhat bulbous at the summit, mostly enclosed anteriorly by a thin sheath adnate to the base. The penial retractor muscle is slender for most of its length, widening at its insertion on the epiphallus. A narrow strand of muscle, the “‘retentor,” extends from the penial retractor muscle at its attachment on the epi- phallus to the summit of the penial sheath, from which other thin retentor fibers form connections with parts of the epiphallus and vas deferens. The sheathed part of the penis in the holotype is about 7.5 mm long; the protruding part is about 2.5 mm long. In the paratypes, the sheathed part varies from 6.0 to 8.0 mm (mean, 7.2 mm). The protruding part varies from 3.0 to 4.0 mm (mean, 3.6 mm). The mean ratio of protruding length to sheathed length is 0.5. There is a broad peduncular section of about 1.0 mm between the base of the sheath and the junction with the atrium. The inner surface of the penis (Figures 5, 6) bears 12- 14 long, parallel and anastomosing, longitudinal pilasters, varying from smooth-surfaced near the apex to distinctly papillose near the base. Near the apex, four or five pilasters fuse to form a thick, anteriorly directed fleshy protuber- ance. In one specimen this is a straight, cylindrical struc- ture projecting anteriorly from near the apex of the penis for a length of 2.5 mm and a width of 0.6 mm. In all other specimens dissected, including the holotype, the structure forms a U-shaped, recurved, cylindrical appendage (Fig- ure 6) attached to the dorsal wall of the penial chamber, with the recurved tip projecting anteriorly into the lumen for a length of about 1.2 mm. There is no verge. The epiphallic-seminal duct opens directly into the penial chamber at its apex. The epiphallus consists of a relatively thick (0.8 mm Explanation of Figures 4 to 6 Figures 4-6. Hochbergellus hirsutus Roth & Miller, gen. et sp. nov. Drawings made from projection of stained whole mounts. Scale line = 1 mm. Structures seen in transparency are shown by dotted lines. Figure 4. Anterior part of reproductive system, holotype, SBMNH 35554, OREGON: Curry County: Sisters Rocks, ca. 3.8 km N of Euchre Creek at Ophir, W. B. Miller coll., 16 July 1991. Figure 5. Anterior part of reproductive system with apical portion of penis opened to show fleshy protuberance and pilasters; paratype, SBMNH 35556, same locality data as above. Figure 6. Apical portion of penis opened and magnified to show parallel, anastomosing pilasters and fleshy penial protuberance; paratype, SBMNH 35557. Abbreviations: at, atrium; cp, cut edge of penis; ep, epiphallus; go, genital orifice; ov, oviduct; pe, penis; pi, pilaster; pp, penial protuberance; pr, penial retractor; ps, penial sheath; pt, prostate; re, retentor; sd, spermathecal duct; sp, spermatheca; ut, uterus; va, vagina; vd, vas deferens. B. Roth & W. B. Miller, 1992 diameter) upper section between the vas deferens and the penial retractor muscle, and a narrower tube between the penial retractor and the penis. There is a completely en- closed, vestigial epiphallic caecum at the junction of epi- phallus and vas deferens. The spermathecal duct is long, slender, and cylindrical, about 7.0 mm in length and 1.0 mm in width at its junction with the oviduct, narrowing to a 0.3 mm constriction at the base of the spermatheca. The spermatheca is ovate, about 3.5 mm long and 1.7 mm at its widest diameter in the holotype. In the dissected paratypes, the length varies from 2.5 to 3.5 mm and the width from 0.8 to 1.9 mm. Type material: Holotype: SBMNH 35554 (shell and stained whole mount of genitalia). OREGON: Curry Coun- ty: Sisters Rocks, ca. 3.8 km N of Euchre Creek at Ophir, W. B. Miller coll., 16 July 1991. Paratypes: SBMNH 35556, 35557 (figured paratypes); SBMNH 35555 (30 unfigured paratypes); all from same locality as holotype. Additional paratypes deposited in ANSP, BR, CAS, LACM, and USNM. Referred material (all, OREGON: Curry County): along trail from Ocean View Camp (2.3 km N of Euchre Creek at Ophir), W. B. Miller coll., 4 April 1952 (SBMNH). Sisters Rocks, between Port Orford and Gold Beach. R. R. Talmadge coll., before 1975; under logs (BR). Remarks: In the material at hand, adult shell diameter ranges from 13.7 to 17.2 mm (mean of 54 specimens in- cluding holotype, 15.42 mm); height, 9.0 to 12.5 mm (x = 10.48 mm); height-diameter ratio, 0.62 to 0.80 (« = .680); number of whorls, 5.6 to 6.7 (x = 6.06). Hochbergellus hirsutus differs from other polygyrid species in southwestern Oregon in the reproductive system Page 225 characters specified above for the genus. At the type lo- cality, H. hirsutus is sympatric with Vespericola megasoma (Pilsbry, 1928). The periostracal setae (2-3/mm/?) are much sparser than those of V. megasoma. In addition, in V. mega- soma the inner part of the basal lip is narrowed, then dilated backward so as to enclose the umbilicus from the left side. For purposes of the American Fisheries Society list of the Scientific and Vernacular Names of Mollusks (TurR- GEON et al., 1988), we propose the name “‘Sisters Hespe- rian.” Etymology: Latin, hirsutus, hairy. ACKNOWLEDGMENTS Weare grateful to Ken Emberton for examining specimens and sharing with us his opinions on the systematic position of Hochbergellus hirsutus. LITERATURE CITED MILLER, W. B. 1967. Anatomical revision of the genus Sonorella (Pulmonata: Helminthoglyptidae). Unpublished Ph.D. Dis- sertation, Department of Biological Sciences, University of Arizona. xili + 293 pp. Piussry, H. A. 1939. Land Mollusca of North America (north of Mexico). Academy of Natural Sciences of Philadelphia Monograph 3, 1(1):i-xvii, 1-573, i-ix. TURGEON, D. D., A. E. BOGAN, E. V. CoAN, W. K. EMERSON, W.G. Lyons, W. L. PRATT, C. F. E. ROPER, A. SCHELTEMA, F. G. THompson & J. D. WILLIAMS. 1988. Common and scientific names of aquatic invertebrates from the United States and Canada: mollusks. American Fisheries Society Special Publication 16:1-277, pls. 1-12. The Veliger 35(3):226-233 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 ‘Taxonomic Re-evaluation and Description of Gari radiata (Dunker in Philippi, 1845) (Bivalvia: Tellinoidea: Psammobiidae) by RICHARD C. WILLAN Department of Zoology, University of Queensland, St Lucia, Brisbane, Queensland, Australia 4072 Abstract. A re-examination of type material and literature has established the correct name for one rather common, tropical Indo-west Pacific species of psammobiid—Gari (Psammobia) radiata (Dunker in Philippi, 1845). Despite repeated misidentifications, the only true junior synonym is Psammobia denikei Martens, 1897. Resolution of the correct name required designation of lectotypes for Psammobia virgata Lamarck, 1818, P. radiata Dunker in Philippi, 1845, and P. denike: Martens, 1897. A description of Gari radiata is provided to distinguish it from the closely related, allopatric G. livida (Lamarck, 1818) and G. convexa (Reeve, 1857). INTRODUCTION Iam currently revising the taxonomy of all the Indo-Pacific members of the bivalve family Psammobiidae. A mono- graph treating the 37 species from the Australian and New Zealand region is complete (WILLAN, in press) and re- search on the southeast Asian fauna is now under way. This contribution is intended as an adjunct to the former work, although the species dealt with here actually occurs within the geographical area encompassed by the latter work. This investigation has involved some research with literature on North Atlantic species because of a misiden- tification by LAMARCK (1818). Gari radiata (Dunker in Philippi) has been known by no less than 20 different combinations of genus and species over the years (see synonymy below). Remarkably, nine of these combinations are misidentifications resulting from confusion with four other species. This paper confirms the current use of Gari radiata (Dunker in Philippi, 1845) as taxonomically correct. On the basis of shell morphology Gari radiata belongs to the subgenus Psammobia Lamarck, 1818. It is a tropical species with a considerable resemblance to two temperate southern Pacific (z.e., Australasian) species in the same subgenus—G. livida (Lamarck, 1818) (= Psammotaea zo- nalis Lamarck, 1818) and G. convexa (Reeve, 1857) (= G. hodgei Willan, 1980). The similarities between these three species have caused great taxonomic confusion. This paper provides a description to separate G. radiata from these allopatric species as well as the sympatric G. amethysta (Wood). Gari radiata is briefly compared with G. virgata (Lamarck) to rectify LAMARCK’s (1818) misidentification. I have not had the opportunity to study the anatomy of any of these species. Abbreviations AMS—The Australian Museum, Sydney, Australia BMNH—The Natural History Museum, London, England c—Complete specimen, 7.e., both right and left valves pres- ent in the same lot though not necessarily joined (mea- surements of length for complete specimens in this con- tribution always relate to the right valve) CAS—California Academy of Sciences, San Francisco, USA h—Half valve (either right or left) only contained in lot MNCN—Museo Nacional des Ciencias Naturales, Ma- drid, Spain MHNG—Muséum d’Histoire Naturelle, Geneva, Swit- zerland MNHN—Muséum National d’Histoire Naturelle, Paris, France NHMW-—Naturhistorisches Museum, Vienna, Austria NMV—National Museum of Victoria, Melbourne, Aus- tralia ZMA—Zoologisch Museum, Universiteit van Amster- dam, Netherlands R. C. Willan, 1992 Page 227 ZMB—Museum fir Naturkunde, Humboldt-Universi- tat, Berlin, Germany ZMUC—Zoologisk Museum, Copenhagen, Denmark Gari (Psammobia) radiata (Dunker in Philippi, 1845) (Figures 1-12, 19) Synonymy Solen sp.: BRUGUIERE, 1797:pl. 227, fig. 5. Psammobia radiata DUNKER IN PHILIPPI, 1845:194, pl. 2, fig. 5; DUNKER, 1882:187; PILSBRY, 1895:122; MARTENS, 1897:244, no. 7; HIDALGO, 1903:85; HaBE, 1977:220; HasBeE, 1981:139. Psammobia amethystus Wood: REEVE, 1856, Psammotia pl. 3, species 19 (misidentification, not Solen amethystus Wood, 1815). Psammobia compta Deshayes: REEVE, 1857, Psammobua pl. 4, species 24 (misidentification, not Psammobia compta Deshayes, 1855). Gari compta (Deshayes): TRYON, 1868:73, Gari species 8; BERTIN, 1880:110, no. 34 (misidentification, not Psam- mobia compta Deshayes, 1855). Gari radiata (Dunker): TRYON, 1868:75, Gari species 31; BERTIN, 1880:122, no. 67; KURODA & HABE, 1952:20. Gari (Psammocola) virgata (Lamarck): BERTIN, 1880:125, no. 80 (misidentification, not Psammobia virgata Lamarck, 1818). Psammobia zonalis (Lamarck): E. A. SMITH, 1885:94 (in part); LYNGE, 1909:211 (in part) (misidentification, not Psammotaea zonalis Lamarck, 1818). Psammobia (Gari) amethystina (sic = error pro. amethystus) Reeve (sic = error pro. Wood): PAETEL, 1890:39 (mis- identification, not Solen amethystus Wood, 1815). Psammobia (Psammocola) radiata Dunker: PAETEL, 1890:40; SHIKAMA, 1964:85; SCARLATO, 1965:51, pl. 2, fig. 5. Psammobia denike1 MARTENS, 1897:243, Psammobia no. 5 (new synonym). Psammobia compta Deshayes: HIDALGO, 1903:88, no. 165 (misidentification, not Psammobia compta Deshayes, 1855). Hiatula (Psammotaea) radiata (Dunker): STANDEN & LEICES- TER, 1906:294. Psammobia virgata Lamarck: LAMy, 1914:22 (misidentifi- cation, not Psammobia virgata Lamarck, 1818). Gari (Gari) radiata (Philippi) (sic = error pro. Dunker in Philippi): PRASHAD, 1932:300. Psammocola radiata (Dunker): HABE, 1952:204; Kira, 1959: 153, pl. 59, fig. 2; Kira, 1962:170, pl. 60, fig. 2; HABE & KIKUCHI, 1960:24. Gari sibogai Prashad: HaBE, 1964:197, pl. 61, fig. 3 (misi- dentification, not Gari sibogai Prashad, 1932). Gari (Psammocola) radiata (Philippi) (sic = error pro. Dun- ker in Philippi): SCARLATO, 1965:51, no. 6, pl. 2, fig. 5. Gobraeus radiatus (Dunker): KuRoDA e¢ al., 1971:43 (English part), pl. 97, fig. 10. Psammobia rabiata (sic = error pro. radiata Dunker): H1Go, 1973:368, no. 1150. Gari (Psammobia) radiata (Dunker): MATSUMOTO, 1979:109, no. 1845. Like all the common Indo-Pacific psammobiids, the spe- cies forming the subject of this investigation has had a complicated taxonomic history. Its first appearance was in BRUGUIERE’s (1797) Tableau Encyclopédique et Méthodique de Trois Régnes de la Nature, but no name was supplied for the figured specimen. That shell must have come from Bruguiére’s own collection and it cannot now be located in MNHN. Lamarck (1818:512) listed the Bruguiére figure in the synonymy of his new species Psammobia vir- gata. However, judging by the two syntypes of P. virgata Lamarck (Figure 17) which are in the Lamarck collection, MHNG, P. virgata is the North Atlantic species generally assigned to Gari intermedia (Deshayes, 1855). This same conclusion was reached by DESHAYES & MILNE-EDWARDS (1835) and CHENU (1862). (Obviously Deshayes changed his mind over the identity of P. virgata Lamarck, because earlier he had informed Hanley that it was the same as P. fervensis Gmelin [HANLEY, 1843:59].) Therefore, Bru- guiére’s figured specimen was neither a syntype, nor from the same geographical region, nor indeed even the same species as the syntypes of P. virgata Lamarck. To resolve the matter finally, I hereby designate as lectotype of Psam- mobia virgata Lamarck the larger syntype (probably the shell figured by CHENU, 1862) (complete specimen—48.6 mm) in MHNG (1083/10/1). The lectotype is illustrated here in Figure 17. The smaller specimen in MHNG (38.2 mm, 1083/10/2) becomes a paralectotype. This action fixes P. virgata Lamarck as the senior synonym of both P. costata Hanley, 1842, and P. intermedia Deshayes, 1855. In the Lamarck collection, MNHN, there are two com- plete shells (51.1, 48.5 mm) of the species under consid- eration (Figure 1). On the back of the wooden tablet to which they both remain glued is the label “psammobie vergattae Psammobia virgata var. [c]” written by Lamarck, and on the front is the annotation ‘““Types de Lamarck” written by some person subsequent to Lamarck (possibly Bertin). As LAMy (1914:4) observed, Lamarck never men- tioned a variety “c,” nor does his description of Psammobia virgata correspond with these shells, so they cannot be considered type material. DUNKER (1845) gave a thorough description as well as excellent figures of the present species under the new name of Psammobia radiata. Authorship must be attributed to Dunker in Philippi, because on the title page of Philippi’s Abbildungen und Beichreibungen neuer oder wenig gettann- ter Conchylien 7 is the statement that the work contains contributions by Anton, van Dem Busch, Dunker, Jonas, Koch, Pfeiffer, and Troschel. Further, the letters ‘‘Dkr”’ (= Dunker) appear on page 194 in bold type beside the first introduction of the name Psammobia radiata. Reeve figured two shells of Gaz radiata in his Concholo- gia Iconica. The first (REEVE, 1856, Psammobia species 19) was erroneously called Psammobia amethystus. The actual shell (Figure 2) is in BMNH. The second (REEVE, 1857, Psammobua species 24) was called Psammobia compta De- shayes and localized from ‘“‘Catbalonga, Island of Samar, Philippines.” However, the type locality cited by DESHAYES (1855— publication date authenticated by DUNCAN, 1937) for P. compta was “Van Diemen’s Land” (z.e., Tasmania). Page 228 The Veliger, Vol. 35, No. 3 17 Explanation of Figures 1 to 18 Figures 1 to 12. Gari radiata (Dunker in Philippi). Figure 1. Smaller of two shells labelled Psammobia virgata var. [c] in Lamarck collection, MNHN, 48.5 mm, unknown locality. Figure 2. Specimen figured as ‘““Psammobia amethystus” by REEVE (1856), 66.6 mm, Ceylon, BMNH 1964018. Figure 3. Specimen figured as ‘““Psammobia compta Deshayes” by REEVE (1857), 50.7 mm, Catbalonga, Island of Samar, Philippines, BMNH 1984294/1. Figure 4. Psammobia radiata Dunker in Philippi, lectotype (selected herein), 48.4 mm, Amboina, Java, ZMB 25463. Figure 5. Psammobia denikei Martens, lectotype (selected herein), 21.7 mm, Makassar, ZMB 25069. Figure 6. 41.6 mm, unknown locality, Deshayes collection, MNHN. Figure 7. 47.9 mm, Manila, Luzon Island, Philippine Islands, R. C. Willan, 1992 The syntypic series of P. compta Deshayes, including the shell figured under that name by Reeve, is in BMNH. The series consists of two G. compta (Lamarck) (35.4, 27.4 mm) and one G. radiata (52.4 mm). For conformity with DEsHAYES’ (1855) description, and particularly because of his type locality for P. compta, I have selected the 35.4 mm specimen of G. compta (BMNH 1841.2.6.423) (Figure 14) as lectotype (WILLAN, in press), and thereby relegated G. compta as a junior synonym of G. livida (Lamarck). BERTIN (1880) wrongly construed the two specimens labelled Psammobia virgata var. [c] (now in MNHN) as types and, following REEVE’s (1856) first illustration, reached the erroneous conclusion that Psammobia virgata Lamarck should replace P. amethystus of Reeve (not Wood, 1815). The fact that DAUTZENBERG & FISCHER (1913, 1914) made no mention of Psammobia virgata in their oth- erwise complete account of Lamarck’s psammobiids sug- gests they realized the two specimens labelled var. [c] in MNHN were not types. LAMy (1914) simply noted the existence of these two shells, and reached no conclusion on their status. E. A. SMITH (1885:94) synonymized Psammobia radiata with Psammotaea zonalis Lamarck on the basis of color- ation, concluding: “It [Psammobia zonalis (Lamarck)] var- ies somewhat in painting, some forms being more rayed than others.”” LYNGE (1909) followed E. A. Smith. PRa- SHAD (1932) disputed E. A. Smith’s union of P. radiata and P. zonalis, and he used the name Gari radiata (wrongly attributed to Philippi) for specimens collected on the Siboga Expedition. Following Prashad, the name Gam radiata (Dunker) has subsequently been adopted by most Asian workers except HABE (1964) who incorrectly used G. sv- bogai Prashad. DESHAYES (1855) was apparently unaware of Dunker’s Psammobia radiata when he described Capsa (Capsella) radiata, or if he did know Philippi’s Abbiblungen, he could not have conceived that P. radiata and G. radiata might be placed in the same genus one day. Now, in fact, both taxa are in Gari, and although G. radiata Deshayes (Figure 17) is the secondary homonym, there is no need to replace it Page 229 because that name falls as a junior synonym of G. elongata (Lamarck, 1818) (ICZN, 1985:Article 60a). Psammobia denikei Martens (Figure 5) is based on two juvenile shells of the species under consideration. As such, that name is the one and only true junior synonym of P. radiata Dunker in Philippi. The following sections (Types through Material Ex- amined) all relate specifically to Gari radiata (Dunker in Philippi, 1845). In the descriptive section, terminology related to shell morphology follows Cox (1969:N39-N58) for an equivalve bivalve except I have substituted umbo (plural, umbones) for beak. ‘Types Psammobia radiata Dunker in Philippi: Lectotype, here des- ignated (figured syntype, complete specimen—48.4 mm) in ZMB (Dunker coll. 25463); figured by DUNKER IN PHILIPPI, 1845, Psammotua pl. 2, fig. 5; illustrated here in Figure 4. Paralectotype (complete specimen—46.7 mm) in ZMB (25463). Type locality Amboina, Java. Psammobia denike: Martens: Lectotype, here designated (larger, figured syntype, complete specimen—21.7 mm) in ZMB (25069); figured by MARTENS, 1897:pl. 10, fig. 25; illustrated here in Figure 5. Paralectotype (complete specimen—17.0 mm) in ZMB (25069). Type locality Makassar. Historically Important Figured Specimens Specimen figured as Psammobia amethystus by REEVE (1856) (complete specimen—66.6 mm) in BMNH (1964018); illustrated here in Figure 2. Locality Ceylon. Specimen figured as Psammobia compta Deshayes by REEVE (1857) (complete specimen—50.7 mm) in BMNH (1984294/1); illustrated here in Figure 3. Locality Cat- balonga, Island of Samar, Philippines. Description Maximum length 60 mm. Shell moderately thick, elon- gate; greatest width at level of umbones; nearly equilateral at all stages of growth, although slightly longer posteriorly; moderately inflated; anterior end rounded; ventral margin CAS. Figure 8. 53.7 mm, unknown locality, NHMW G9387. Figure 9. 37.1 mm, 9-16 m, off Cape Liant and Mesam Island, Mellem, Thailand, ZMUC. Figure 10. 46.0 mm, Japan, MNHN. Figures 11 (exterior) and 12 (interior). 53.5 mm, unknown locality, NHMW 347. Figure 13. Gari convexa (Reeve). 54.2 mm, Smokehouse Bay, Port Fitzroy, Great Barrier Island, New Zealand, Willan collection. Figures 14 and 15. Gari livida (Lamarck). Figure 14. Psammobia compta Deshayes, lectotype (selected by WILLAN, in press), 35.5 mm, Tasmania, BMNH 1841.2.6.423. Figure 15. 36.0 mm, Southport, southern Tasmania, WAM 1051-70. Figure 16. Gari virgata (Lamarck). Psammobia virgata Lamarck, lectotype (selected herein), 48.6 mm, “Indian Ocean,” Lamarck collection, MHNG 1083/10/1. Figure 17. Gari elongata (Lamarck). Capsa (Capsella) radiata Deshayes, lectotype, 36.2 mm, Philippine Islands, BMNH 1984267/1. Figure 18. Gari amethysta (Wood). 50.5 mm, Manila, Luzon Island, Philippine Islands, MNHN. Page 230 19 The Veliger,) Vol. 35) Nox Explanation of Figures 19 to 22 Detail of sculpture on posterior slope of right valve of Gari (Psammobia) species. Figure 19. G. radiata (Dunker in Philippi), 66.6 mm, Ceylon, BMNH 1964018. Figure 20. G. convexa (Reeve), 54.2 mm, Smokehouse Bay, Port Fitzroy, Great Barrier Island, New Zealand, Willan collection. Figure 21. G. livida (Lamarck), 39.4 mm, unknown locality, Australia, BMNH 1985182/1. Figure 22. G. amethysta (Wood), 28.8 mm, Philippine Islands, NMV. almost straight; posterior end narrower than anterior end, subacute, with distinct angle at termination of posterior ridge; equivalve (left valve slightly flatter in large adults); commissure at junction of valves’ ventral margins straight (sometimes weakly sinuous); moderate anterior and small posterior gapes. Surface of both valves highly polished, smooth, anterior and middle sections sculptured only with broad, irregular, concentric growth furrows that become more incised near anterodorsal margin. Right valve with a ridge extending from umbo to posterior extremity; nu- merous, fine, raised, concentric striae always extend with- out interruption from posterodorsal margin to this ridge, and, in some specimens, striae cross ridge to become ir- regular, flattened lamellae on posteroventral area of central section; weak ridge but no striae in corresponding position on left valve. Exterior covered with a very thin, dehiscent, pale greenish-brown periostracum. Hinge plate narrow, moderately elongate; nymph moderately thick, high. Right valve with two cardinal teeth, each oblique and of ap- proximately equal strength, the anterior one weakly bifid, the rear one strongly so, diverging from each other at 60°. Left valve with single, weakly bifid anterior cardinal tooth; rear cardinal tooth represented merely by a low lamella on hinge plate, sloping at 65° behind anterior cardinal tooth; distinct, small lunular projection (stronger than pos- terior cardinal) bearing a microscopic lunular ligament also present on left valve. Pallial sinus deep (reaching halfway between hinge plate and rear margin of anterior adductor scar), broad, U-shaped; upper limb gradually descending; anterior margin rounded (more broadly in right valve); lower limb confluent with pallial line for most of its length; ventral extremity of pallial line straight, reaching level with rear of posterior adductor scar (Figure 12). Single, irregularly shaped pedal retractor scar present dorsally in front of hinge plate. Color of exterior bluish lilac, often faintly marbled, interrupted by numerous, ei- ther wide or narrow, reddish rays that emanate from white umbones; rays maintain coloration across shell; rays not symmetrical on right and left valves; central section of interior with purplish glaze and darker, brown-violet patches on either side of white umbonal area; posteriorly, brown-violet color extends onto nymph; reddish rays Page 231 R. C. Willan, 1992 prominent at ventral margin, which is white through lack of overlying purplish glaze. Remarks Gari radiata is closest to G. convexa (Reeve) (= G. hodgei Willan, synonymy discussed by WILLAN, in press) and G. livida (Lamarck). However, there are several characters that consistently separate these three allopatric species. (G. radiata was not compared with G. convexa in an earlier paper [WILLAN, 1980].) Compared with G. radiata, G. convexa (Figures 13, 20) from New Zealand is larger (to 85 mm), heavier, markedly inequilateral when adult and less elongate, the anterodorsal margin descends more steep- ly, the posterior margin is relatively broader (but still angulate), the right valve is more inflated, and the left valve is flatter resulting in a more sinuous commissure, the posterior ridge on the right valve is more rounded, the striae—which are finer and more numerous—never extend to the posterior ridge (Figure 20) let alone cross over it onto the posteroventral area of the valve’s central section, and the upper limb of its pallial sinus is straighter—hence the entire sinus is broader anteriorly; the ground color of G. convexa is never marbled, and its rays (which are rel- atively broader) are generally less distinct from the back- ground, its umbones are flushed with pale cream-yellow, pink or lavender, its internal glaze is thicker and more uniform, and finally there is neither the white umbonal area nor the brownish orange blotches of G. radiata. Gari livida (Figures 14, 15, 21) from temperate southern Australia is smaller (to 45 mm), shorter, and heavier, its posterior margin is relatively broader, its left valve is flatter (but the commissure is straight), its surface bears numer- ous, fine, close, concentric cords, the striae on the posterior slope of the right valve are finer and more numerous (they do, however, reach the posterior ridge and/or cross it to become irregular lamellae like G. radiata) (Figure 21), the upper limb of the pallial sinus is straighter—hence the entire sinus is broader anteriorly; the ground color of G. livida is never marbled, its coloration—both externally and internally—is like that of G. convexa, and its rays are never visible at the ventral margin. REEVE (1856) confused Gari radiata with G. amethysta (Wood) (Figures 18, 22), but G. amethysta is less closely related than either G. convexa or G. livida. Gari amethysta has a broader posterior end, more numerous concentric striae that always cover all the central area as well as the posterior slope on both valves, and its ground coloration is marbled violet-brown with distinct, darker rays. Although it seems unlikely to me any modern author would follow LAMARCK (1818) and confuse Gari radiata with the eastern Atlantic-Mediterranean G. (Gobraeus) virgata (Lamarck), a brief comparison is justified. Gari virgata (Figure 16) is thicker, broader, more truncate pos- teriorly, and differently sculptured, and its pallial sinus is relatively broader. Both species have a purplish internal glaze and brownish violet streak in the escutcheonal area of the posterodorsal margin. Geographical Distribution Within the tropical Indo-west Pacific, Gari radiata has a large, roughly triangular area of distribution. Its north- ern limit is Boso Peninsula, Honsht Island, Japan (KURO- DA et al., 1971). Southwards, it extends through the South China Sea (SCARLATO, 1965) and Philippine Sea (REEVE, 1857; HIDALGO, 1903) to Indonesia (PRASHAD, 1932). Westwards, it extends through the Gulf of Thailand (LYNGE, 1909) and Bay of Bengal. The western extremity is the west coast of the island of Sri Lanka (REEVE, 1856). Thorough searches of collections made recently in the Marianas Islands, Papua New Guinea, New Caledonia, the Kermadec Islands, and northern Australia failed to reveal any specimens of G. radiata. Habitat According to Kuropa et al. (1971), Gari radiata inhabits (presumably clean) sandy substrata from the intertidal zone to 10 m. I have not collected this species myself, so I cannot add any further information on habitat or ecology. Material Examined JAPAN: 6c (AMS C125620; NHMW 347; MNHN); lc, Wakayama-ken, Honsht I. (AMS (74209); 1c, ELWN, Mikawa, Honsht Island (AMS C87763). HONG Kone: 5c,1h (AMS 38771; BMNH 1936.1.8.224-225; CAS; MNHN). PHILIPPINE ISLANDS: 5c (BMNH; MNCN); 1c, Manila, Luzon Island (CAS). INDONESIA: 2c, Makassar (ZMB 25069—lectotype and paralectotype of Psammobia deniket Martens); 1h, Corindon, Makas- sar—01°56'S, 119°917'E (MNHN); 1c, Ambon Island (ZMB 25463—lectotype of Psammobia radiata Dunker in Philippi); 1c, Java (ZMB—paralectotype of Psammobia radiata Dunker in Philippi); 1c, Java (ZMUC); 1c, 10 m, near Onrust (ZMUQ); 1c, Baai van Pidjot, Lombok Island (ZMA); 1c, Toeal, Kei Eilandeu (ZMA). MA.aysia: 1h, on beach, Tanjong Rhu, N Pulau Langkawi, NW Ma- laysia—06°27'N, 99°50'E (AMS) 1c, Penang (MNHN). SINGAPORE: 1h, shallow water (ZMUC). THAILAND: 1c,1h, 9-16 m, off Cape Liant and Mesam Island, Mellem (ZMUC); 1h, 3 m, Koh Kahdat (ZMUC). Sri LANKa: 10c (BMNH 1964018, 1964019; NHMW G9387; MNCN; MNHN); 4c, Gulf of Mannar (BMNH 1953.1.7.163-165). “INDIAN OCEAN”: 2c (MNHN-—spec- imens labelled “Psammobia virgata var. c” by Lamarck). UNKNOWN LocaLity: 4c (MNHN; NHMW 347; G9387). ACKNOWLEDGMENTS The following curators kindly provided access to spec- imens and/or literature: E. Cools (CAS); Y. Finet (MHNG); I. Loch (AMS); R. Kilias (ZMB); R. Moolen- Page 232 The Veliger, Vol. 35, No. 3 beek (ZMA); S. Morris (BMNH); T. Schistte (ZMUC); J. and M. Templado (MNCN); and E. Wawra (NHMW). Funds for travel to Europe in 1989 were provided by The University of Queensland under an Assisted Development Program. LITERATURE CITED BeERTIN, M. V. 1880. Revision des garidées du Muséum d’His- toire Naturelle. Nouvelles Archives du Muséum d’Histoire Naturelle, Paris, 2éme série 3:57-129, pls. 4, 5. BRUGUIERE, J.C. 1797. Planches 190-286. Encyclopédique et Meéthodique des Trois Régnes de la Nature. Part 21. Henri Agasse: Paris. Cox, L.R. 1969. General Features of Bivalvia. Pp. N2-N129. In: R. C. Moore (ed.), Treatise on Invertebrate Paleontology, Part N, Vol. 1, Mollusca 6, Bivalvia. The Geological Society of America and the University of Kansas, Kansas. 489 pp. CHENU, J.C. 1862. Manuel de Conchyliogie et de Paléontolo- gie Conchyliologique. Vol. 2. Victor Masson: Paris. 327 pp. DAUTZENBERG, P. & H. FISCHER. 1913. Sur quelques types de garidés de Lamarck. Bulletin du Muséum d’Histoire Naturelle 19:484-487. DAUTZENBERG, P. & H. FISCHER. 1914. Sur quelques types de garidés de la collection de Lamarck existant au Muséum de Paris. Journal de Conchyliologie:215-228, pls. VI, VII. DeEsHayYEs, G. P. 1855. Descriptions of new shells from the collection of Hugh Cuming, Esq. Proceedings of the Zoo- logical Society of London for 1854, Vol. 22:317-371 [date of publication authenticated by DUNCAN, 1937]. DesHaYEs, G. P. & H. MILNE-EDwWarDs. 1835. Histoire na- turelle des Animaux sans Vertébres, présentant les charac- téres généraux par J.B.P.A. de Lamarck ... ed. 2 revue et augmentée. Vol. 6, Histoire des mollusques. J. B. Bailliére: Paris. iv + 600 pp. Duncan, F.M. 1937. On the date of publication of the Society’s ‘Proceedings,’ 1850-1926. By F. Martin Duncan, F.Z.S., Librarian to the Society. With an Appendix containing the dates of publication of ‘Proceedings,’ 1830-1858, compiled by the late Henry Peavot, originally published in P.Z.S. 1893, 1913. Proceedings of the Zoological Society of London, Series A(1):71-84. DUuNKER, G. 1845. Psammobia radiata Dkr. Page 194. In: R. A. Philippi (ed.). Psammobia. Tab. II. Abbildungen und Beschreibungen neuer oder wenig gekannter Conchylien, unter Mithtfle mehrerer Deutscher Conchyliologen her- ausgegebennew. Vol. 1, Lief. 8. T. F. Fischer: Cassel. Pp. 179-204 + index, pls. viii.1—vilib. DUNKER, G. 1882. Index Molluscorum Maris Japonici Con- scriptus et Tabulis Iconum xvi Illustratus. Theodor Fischer: Cassellis Caltorum. 301 pp., 16 pls. HasBe, T. 1952. Genera of Japanese Shells. Pelecypoda, No. 3. Pp. 187-278. HaBE, T. 1964. Shells of the Western Pacific in Colour, Vol. 2. Hoikusha: Osaka. 233 pp., 66 pls. HaseE, T. 1977. Systematics of Mollusca in Japan: Bivalvia and Scaphopoda. Hokuryukan: Tokyo. 372 pp. HaBE, T. 1981. Family Psammobiidae Fleming. Pp. 138-142 In: Y. Koyama, T. Yamamoto, Y. Toki & H. Minato (eds.), A Catalogue of Molluscs of Wakayama Prefecture. The Province of Kii. 1. Bivalvia, Scaphopoda and Cephalopoda. Special Publications of the Seto Marine Biology Laboratory 7:1-301. Hasse, T. & T. KikucHI. 1960. Fauna and Flora of the Sea around Amakusa Marine Biological Laboratory. Part I. Mollusca. Amakusa Marine Biological Laboratory Kyushu University. 70 pp. HANLEY, S. 1843. An Illustrated and Descriptive Catalogue of Recent Bivalve Shells. Williams & Norgate: London. xviii + 392 pp., 24 pls., suppl. pls. 9-24. H1patco, J. G. 1903. Obras Malcologicas. Parte 1. Estudios preliminares sobre la Fauna Malacologica de las Islas Fi- lipinas, Vol. 2. Memorias de la Real Academia de Ciencias Exactas, Fisicas y Naturales de Madrid. L. Aguado: Madrid. 400 pp., 30 pls. Hie0, S. (ed.). 1973. A Catalogue of Molluscan Fauna of the Japanese Islands and the Adjacent Area. 397 pp. ICZN (INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE). 1985. International Code of Zoological Nomenclature. 3rd ed. International Trust for Zoological Nomenclature and British Museum (Natural History): Lon- don. 338 pp. Kira, T. 1959. Colored Illustrations of the Shells of Japan. 2nd ed. Hoikusha: Osaka. 240 pp., 72 pls. Kira, T. 1962. Shells of the Western Pacific in Colour. Hoiku- sha: Osaka. 224 pp., 72 pls. Kuropa, T. & T. HABE. 1952. Check list and Bibliography of Recent Marine Mollusca of Japan. Leo W. Stach: Tokyo. ii + 210 pp. Kuropa, T., T. HABE & K. OyAMA. 1971. The Sea Shells of Sagami Bay Collected by His Majesty the Emperor of Japan. Maruzen: Tokyo. 741 pp. LAMARCK, J.B. P.A.DE 1818. Histoire naturelle des Animaux sans Vertébres présentant les charactéres généraux ... par J-B.P.A. de Lamarck .. . edition 2 revue et augmentée. Vol. 5, les conchiferes (Conchifer). Verdire, Deterville, and ‘“‘chez Pauteur”: Paris. 612 pp. Lamy, M. E. 1914. Notes sur les espéces Lamarckiennes de Garidae. Bulletin du Muséum d’Histoire Naturelle 20:19- 25, 57-65. LynGE, H. 1909. The Danish Expedition to Siam 1899-1900 IV. Marine Lamellibranchiata. Danske Videnskabernes Sel- skab Skrifter 7. Raekke, Naturvidenskabelig og Mathema- tisk 5. B.L. Bogtrykkeri: Copenhagen. 299 pp. MARTENS, E. VON 1897. Stiss- und Brackwasser-Mollusken des Indischen Archipels. Pp. 1-331, pls. 1-12. In: M. Weber (ed.), Zoologische Ergebnisse Einer Riese in Niederlandisch Ost-Indien Herausgegeben, Vol. 4. E. J. Brill: Leiden. 420 pp., 16 pls. Matsumoto, Y. 1979. Molluscan Shells of Mie Prefecture, Japan. Toba Aquarium: Toba. 179 pp., (4) + 21 pls. PAETEL, F. 1890. Catalog der Conchylien-Sammlung von F. Paetel. Vols. 11-18. Part iii, Acephalen und die Brachio- poden. 256 pp. Pitssry, H. A. 1895. Catalogue of the Marine Mollusks of Japan with Descriptions of New Species and Notes on Oth- ers Collected by Frederick Stearns. Detroit. 196 pp. PRASHAD, B. 1932. The Lamellibranchia of the Siboga Ex- pedition. Systematic Part. II. Pelecypoda (exclusive of the Pectinidae). Siboga Expedition Reports, Monograph 53C: 1-353, pls. 1-9, 1 map. REEVE, L. 1856-1857. Monograph of the genus Psammobia. Conchologia Iconica, Vol. 10. L. Reeve: London. 8 pls. ScARLATO, O. A. 1965. Superfamily Tellinacea (Bivalvia) of Chinese Seas. Studia Marina Sinica 8:27-114, pls. 1-13. SHIKAMA, T. 1964. Selected Shells of the World Illustrated in Colour [II]. Hokurya-kan: Tokyo. 212 pp. SMITH, E. A. 1885. Report on the Lamellibranchia collected by H.M.S. Challenger during the years 1873-76. Reports R. C. Willan, 1992 of the Scientific Results of the Exploratory Voyage of H.M.S. Challenger 1873-1876, Zoology 13, Part 35:1-341, 25 pls. STANDEN, R. & A. LEICESTER. 1906. Report on the molluscan shells collected by Professor Herdman, at Ceylon, in 1902. Ceylon Pearl Oyster Fishery Report 5:267-294. Tryon, G. W. 1868. Description of new species of marine bivalve Mollusca in the collection of the Academy of Natural Sciences, American Journal of Conchology 5:170-172, pl. 16. Page 233 WILLAN, R. C. 1980. A re-evaluation of Gari lineolata (Gray in Yate, 1835) (Bivalvia: Psammobiidae). Journal of the Royal Society of New Zealand 10:173-183. WILLAN, R. C. In press. Taxonomic revision of the family Psammobiidae (Bivalvia: Tellinoidea) in the Australian and New Zealand region. Supplement, Records of The Austra- lian Museum. The Veliger 35(3):234-241 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 Embryonic Stages of Loligo bleeker: Keferstein (Mollusca: Cephalopoda) GYEONG HUN BAEG, YASUNORI SAKURAI anp KENJI SHIMAZAKI Research Institute of North Pacific Fisheries, Faculty of Fisheries, Hokkaido University, Hakodate, Hokkaido, 041 Japan Abstract. For observation of the embryonic development of Loligo bleekeri, a total of 12 fresh egg capsules were kept in a small aquarium at a temperature of 11.7 + 0.4°C, a salinity of ca. 34.00%, and a photoperiodicity of 12L:12D in the laboratory. The period from spawning to hatching ranged from 64 to 67 days. The diameter of eggs ranged from 2.6 to 2.7 mm and mantle length of hatchlings from 3.0 to 3.3 mm. Twenty-eight embryonic stages defined by morphological features are described for Loligo bleeker:. The major developmental pattern of the species was mostly identical to that of L. pealen and L. forbesi. INTRODUCTION A total of about 35 species of squids in the genus Loligo are known to occur in coastal and neritic waters of the world oceans (NESIS, 1987). Loligo bleekeri is distributed around Japan, South Korea, and the northern part of China (NEsIs, 1987). Neritic squids of the family Loliginidae are increasingly used as experimental material for neurophysiological re- search (ROSENBERG, 1973; ARNOLD et al., 1974; MATSUMOTO, 1976), and they are also an important fish- ery resource (Voss, 1973; OKUTANI, 1977; RATHJEN et al., 1979). Previous studies on the embryonic development of Loligo bleekeri (NISHIGAWA, 1868; ISAHAYA, 1934; ISAHAYA & ‘TAKAHASHI, 1934; HAMABE, 1960) have described mor- phological changes with time. For comparative studies of different species, normal developmental stages of squid embryos are required. This study, based on the above short notes, was done primarily to establish a standardized embryonic develop- ment scheme for Loligo bleekeri under stable environmental conditions. MATERIALS anpD METHODS Fresh egg capsules from several captive adult females of Loligo bleekert were collected 8 January 1991 and 13 April 1991, and transported to the laboratory. Twelve egg cap- sules were used in this study. In the laboratory, four egg capsules were suspended in each of three beakers, each containing about 800 mL of seawater, in an aquarium supplied with a running water system. Seawater in the beakers was renewed daily and aerated to prevent abnormal development. The egg capsules were kept at a temperature of 11.7 + 0.4°C, the salinity was ca. 34%o, and the photoperiod was 12L:12D. The light intensity was maintained at 1900 lux during the light period (08:00-20:00) and 5 lux during darkness (20:00-—08:00). Observations were made four times each day from Stage 1 to Stage 9, and once a day from Stage 10 to Stage 28. In each observation, a few eggs were extracted randomly from the egg capsules. For detailed examination, the whit- ish outer coat and chorions were removed. The distribution of the chromatophores of the embryo at Stage 28 was drawn based on color transparencies of hatchlings. The morphological staging criteria follow those of ARNOLD (1965) and SEGAWA et al. (1988). The 30-stage system of ARNOLD (1965) is widely used in cephalopods. Minor variation of stage numbers should be considered when classifying developmental stages, because there are heterochronies of the first appearance of some organ sys- tems used as staging criteria with different species and there are also different authors’ interpretations for the subdivisions. A 28-stage system was appropriate for Loligo bleekeri, and is used in this study. RESULTS anp DISCUSSION Egg development of Loligo bleekeri Egg capsules measured 8-10 cm in length and contained about 40-50 eggs. Individual eggs were enveloped in sev- Gare Bacgiet al 992 Page 235 STAGES 30 ecece E é Lo lk Se ie - ba raw c oes f= iS v5 ; fla fo) wy ® c 6 MCGOWAN (1954); ®° McConaTHY et al. (1980); 7 WorMs (1983); 8 MANGOLD-WIRZ (1963); BOLETZKY (1974); ° SEGAWA et al. (1988); '° SEGAWA (1987). Stage 11 (A-13, 14 days): Blastoderm covers about one- third of egg surface. Stage 12 (A-14/15, 16 days): Blastoderm covers about one-half of egg surface. Stage 13 (A-16, 19 days): Blastoderm covers about two- thirds of egg surface. Stage 14 (A-17, 22 days): Blastoderm nearly covers egg surface. Optic vesicle primordia appear as two thickened placodes on either side of embryo. Shell gland primordium visible at former animal pole. Embryo begins to rotate (very slow gliding movement generated by cilia on the yolk envelope; BOLETZKY, 1971; JOLL, 1978). Slight equatorial constriction appears, which future separates external yolk sac from developing embryo. Stage 15 (A-18, 23 days): Outer yolk sac completely en- veloped by blastoderm. Shell gland begins to invaginate and its border elevates. Peripheral elevation of eye rudi- ment distinct. Mantle primordium first visible. Stage 16 (A-18+, 24 days): Ring folds enclose eye pri- mordia and grow over central part of rudiments. Mouth appears as crescent-shaped invagination on oral surface of dorsal surface. Arm primordia appear as two thickened but not discrete bands of cells lateral to and in front of anterior funnel folds. Anterior and posterior funnel folds present as placodes on ventral surface. Stage 17 (A-19, 26 days): Statocyst primordia appear as slight depression between anterior funnel fold and poste- rior funnel fold. Gill primordia appear as thickened pla- codes along with posterior funnel folds. Unpaired median anal papilla primordium visible between gill primordia. Arm primordia discrete and tentacles elongate. Mantle The Veliger, Vol. 35, No. 3 grows toward both animal and vegetable poles. Mouth invagination distinctly visible. Stage 18 (A-20, 29 days): Anterior and posterior funnel folds elevated and grow toward midline. Eye vesicles nearly closed. Sucker primordia first appear on tentacles. Distinct salivary pit in mouth. Gills apparent. Stage 19 (A-21, 30 days): Shell gland completely closed. Transverse fin primordia appear on mantle but not yet distinct from surrounding mantle tissue. Anterior and pos- terior funnel folds fuse together. Granular particles appear in perivitelline space. First cup-shaped suckers on tentacles visible. Eye lens primordia faintly visible. Light red pig- mentation of retina visible. Stage 20 (A-22, 34 days): Distal edges of anterior funnel folds raise and bend toward midline. Mantle covers one- third of gills. Fins apparent. Suckers on arms III first appear. Eye lenses evident. Stage 21 (A-23, 37 days): Retina cup-shaped. Mantle covers one-half of gills. Funnel folds fuse anteriorly. Stage 22 (A-24, 39 days): Two-thirds of gills covered by anterior-growing mantle. Median margins of funnel folds fuse; funnel retraction muscles still visible. Statocysts fully developed with statoliths. Branchial hearts formed at bases of gills and start to beat irregularly. Suckers on arms IV appear. Iris curved but not pigmented. Gill leaflets visible. Stage 23 (A-25, 40 days): Iris begins to be pigmented. Retina turns dark red-brown. Posterior portion of funnel covered by mantle. Gill leaflets visible through mantle wall. Systemic heart starts to beat. Red chromatophores visible on ventral mantle surface. Mantle contractions ob- served. Posterior lobes of internal yolk sac increase in size. External yolk sac still longer than embryonic body. Stage 24 (A-26, 46 days): Tentacle and lateral arm bases extend as folds that cover about one-half of optic ganglion on each side (cf. NAEF, 1928). Red chromatophores present on tentacles, arms IV, ventral and dorsal head, and surface of dorsal mantle. Stage 25 (A-27, 52 days): Hoyle’s organ evident on pos- terior dorsal mantle between fins. Ventral arm bases cover one-half to two-thirds of eye vesicles. Ink sac begins pro- ducing ink. First red chromatophores observed on arms II. External yolk sac approximately same length as mantle. Stage 26 (A-28, 56 days): Ink sac filled with ink. Second and third row of chromatophores appear on tentacles. Sec- ond row chromatophores yellow; third row chromato- phores small, red. Mid-gut gland and caecum prominent. External yolk sac approximately equal to head length. Olfactory tubercles clearly visible on each side of ventral surface of head. GH Bacsvet al 1992 Page 237 30 30 all is WO w ie HO < l= I = iS HO is HO | 2 Jt 2 ane \ VCH Peek er are ae LO et an cai VCH VCH i] E V pe|W ViVi ire | SGC ec W RP PL | sac PS SGC EC PS 2a cts ips SS SMa = = - = 20 " PSG Vv Vv : PSG PSG 15 -- -------- - oes = ~ - - - - - - --15 | Loligo bleekeri Loligo pealei Loligo forbesi Figure 2 Comparison of chronological appearance of select organs in three species of the genus Loligo, using ARNOLD’s (1965) stage system. Key: EC, eye vesicle closed; EI, eye vesicle invagination begins; FF, funnel formation begins; HO, Hoyle’s organ appears; IS, ink sac appears; PL, primordium of lens visible; PS, primordia of suckers appear; PSG, primordium of shell gland appears; RP, retina pigmentation begins; SGC, shell gland closed; VCH, ventral mantle chromatophores appear. Stage 27 (A-29, 59 days): Chromatophores first appear on arms III. External yolk sac approximately equal to length of tentacles. Stage 28 (A-30, 64-67 days): Hatching. Small external yolk sac dropped and Hoyle’s organ becomes depleted. Internal yolk sac remains as small triangular body. Shortly after hatching, larvae swim toward water surface with mantle pointing upward. Hatchling mantle lengths are 3.0-3.3 mm. Chromatophore patterns relatively regular on ventral and dorsal head and on dorsal mantle, but, among individuals, highly asynchronous on ventral mantle. Comparison of Embryonic Development Among Three Loligo Species Based on the observed sequence of development of the major embryonic features, the developmental pattern of Loligo bleekeri was compared to that of L. pealeu (ARNOLD, 1965). The results are shown in Figure 1. In this figure, for the sake of convenient comparison, developmental stages of L. bleekert were aligned to those of ARNOLD (1965) with the criteria shown in the former section. Within the first five-day period, the eggs of L. bleeker: developed rapidly from Stage 1 to Stage 10. Thereafter, the developmental pattern of the eggs became linear with a gentle slope. This developmental pattern in L. bleeker:1 was almost the same as that observed in L. pealeit by ARNOLD (1965). In con- trast, SEGAWA et al. (1988) found a sigmoid curve in the internal organogenesis of L. forbesv. Appearance of Some Organ Rudiments in Squid Embryos The developmental length and size of hatchlings depend largely on the size of the spawned egg (MANGOLD et al., 1971; SEGAWA et al., 1988). Although about 35 Loligo species are distributed in shallow neritic waters around the world (NEsIs, 1987), embryonic development is poorly known except for a few species. Information on egg size, development time, and hatch- ling size of six species of the family Loliginidae are sum- marized in Table 1. In comparison to other species, L. bleekeri has relatively larger eggs (2.6-2.7 mm) and a longer development time (64-67 days). As seen in Table 1, the development time and mantle length of hatchlings increase as egg size increases. If these differences are valid, then species-specific differences in the first appearance of some organs, relative to stages, may exist. At present three loliginid species—Loligo bleekeri, L. pealeu, and L. forbesi—are available for comparison (see Figure 2). The pattern of chronological appearance of organs is quite similar between the species of Loligo so far examined. However, several differences in the chronology of appear- ance are evident between two or three species (Figure 2): for example, the primordium of the shell gland, PSG; funnel formation, FF; primordia of suckers, PS; retina pigmentation, RP; primordium of lens, PL; ventral mantle chromatophores, VCH; ink sac, IS; Hoyle’s organ, HO. In all instances, differences among species were restricted Page 238 The Veliger, Vol. 35, No. 3 Stage 9 Stage 13 Stage 14 Stage 15 Stage 16 Figure 3 (Part 1) G. H. Baeg et al., 1992 Page 239 Stage 17V Stage 18V Stage 19V Stage 20V Stage 21V Le 22D Stage <3 23V Figure 3 (Part 2) Figure 3 Stages 1-28 in the embryonic development of Loligo bleekeri. Dorsal (D), ventral (V) or laterodorsal (LD) views. Key: A, anal papilla; Ab, arm bases; Aff, anterior funnel fold; Ap, animal pole; Bd, blastoderm; Cd, cell division; Ce, caecum; Ch, chromatophore; F, fin; G, gill; Gl, gill leaflets; Grl, germ-ring layer; Ho, Hoyle’s organ; Ir, iris; Is, ink sac; Iy, internal yolk sac; L, lens; M, mouth; Ma, mantle; Md, marginal division; Mg, mid-gut gland; O, optic vesicle; Of, olfactory tubercle; Pa, primordia of arms; Pb, polar body; Pf, primordium of fin; Pff, posterior funnel fold; Pl, primordium of lens; Pma, primordium of mantle; Po, primordium of optic vesicle; Psg, primordium of shell gland; Pst, primordium of statocyst; Py, posterior lobes of inner yolk; Rp, retina pigmentation; Stl, statolith; Su, sucker; T, tentacle; A2, arm II; A3, arm III; A4, arm IV; HD, dorsal side of head; HV, ventral side of head; MD, dorsal side of mantle; MV, ventral side of mantle; r, red chromatophore; y, yellow chromatophore. Page 240 The Veliger, Vol. 35, No. 3 Stage 28 D Figure 3 (Part 3) G. H. Baeg e¢ al., 1992 Page 241 within a narrow range of not more than three stages. This extreme homogeneity in the morphological development of loliginid squids has been previously reported by HUNTER & SIMON (1975). Recently, NATSUKARI (1984) placed Loligo bleekeri and L. pealei in a separate genus, Heterololigo, because of sev- eral distinct morphological characteristics, including the presence of a well-developed medial and longitudinal crest on the mid-portion of the modified part of the hectocoty- lized arm. A distinct similarity in the developmental patterns of the two Heterololigo species compared to those of Loligo bleekeri and L. forbes: was not found. In Loligo species, the degree of similarity in developmental processes seems to depend solely on egg size. ACKNOWLEDGMENTS The authors are grateful to the staff of the Chiba Fisheries Company, Hakodate, Japan, for providing living material for this study and to the staff of the Usujiri Fisheries Laboratory, Hokkaido University, for their help. This is contribution No. 252 from the Research Institute of North Pacific Fisheries, Faculty of Fisheries, Hokkaido Univer- sity. LITERATURE CITED ARNOLD, J. M. 1965. Normal embryonic stages of the squid, Loligo peale: (Lesueur). Biological Bulletin 128(1):24—32. ARNOLD, J. M., W. C. SUMMERS, D. L. GILBERT, R. S. MANALIS, N. W. Daw & R. J. LASEK. 1974. A Guide to Laboratory Use of the Squid, Loligo peale:. Marine Biological Labora- tory: Woods Hole, Massachusetts. 74 pp. BoLeTzky, S. V. 1971. Rotation and first reversion in the Octopus embryo—a case of gradual reversal of ciliary beat. Experientia 27:558-560. BOLETZKY, S. V. 1974. The larvae of Cephalopoda: a review. Thalassia Jugoslavia 10(1/2):45-76. BoLeTzky, S. V. & R. T. HANLON. 1983. A review of the laboratory maintenance, rearing and culture of cephalopod molluscs. Memoirs National Museum Victoria 44:147-187. FIELDS, W. G. 1965. The structure, development, food rela- tions, reproduction, and life history of the squid Loligo opales- cens Berry. California Department Fish & Game, Fishery Bulletin 131:1-108. HaAMABE, M. 1960. Observations of early development of a squid Loligo bleekeri Keferstein. Annual Report of Japan Sea Regional Fisheries Research Laboratory (6):149-155. [In Japanese with English title and abstract.] Hunter, V. D. & J. L. Simon. 1975. Post-cleavage mor- phology in the squid Lolliguncula brevis (Blainville, 1823). The Veliger 18(1):44-51. IsaHayA, T. 1934. Fertilization and spawning of the egg Loligo bleekert. Hokkaido Suisan Sikenjo ZikyoZunppo (257):9- 10. [In Japanese. ] IsaHAYA, T. & T. TAKAHASHI. 1934. Development of the squid Loligo bleeker. Hokkaido Suisan Sikenjo Zikyo- Zunppo (260):7-8. [In Japanese. ] JOLL, 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. 1963. Biologie des Céphalopodes ben- thiques de la Mer Catalane. Vie et Milieu, supplement 13: 285 pp. MANGOLD, K., S. V. BOLETZKY & D. FROscH. 1971. Repro- ductive biology and embryonic development of Eledone cirrosa (Cephalopoda: Octopoda). Marine Biology 8:109-117. Matsumoto, G. 1976. Transportation and maintenance of adult squid (Doryteuthis bleekeri) for physiological studies. Biological Bulletin 150:279-285. McConatny, D. A., R. T. HANLON & R. F. Hixon. 1980. Chromatophore arrangement of hatchling loliginid squids (Cephalopoda, Myopsida). Malacologia 19(2):279-288. McGowan, J. A. 1954. Observations on the sexual behavior and spawning of the squid, Loligo opalescens, at La Jolla, California, California Department Fish & Game 40:47-54. McMauon, J. & W. C. SUMMERS. 1971. Temperature effects on the developmental rate of squid Loligo peale: embryos. Biological Bulletin 141:561-567. NaEF, A. 1928. Die Cephalopoden. Fauna Flora Golf Neapel 35:1-357. NATSUKARI, Y. 1984. Taxonomical and morphological studies on the loliginid squids—IV. Venus (Japanese Journal of Malacology) 43(3):229-239. Nests, K. N. 1987. Cephalopods of the World. [Translated into English by B. S. Levitov, ed. by L. A. Burgess.]. T. F. H. Publications: Neptune City, New Jersey. 351 pp. NISHIGAWA, T. 1868. Development of the squid Loligo bleekeri. Zoological Magazine 10(115):156-162. [In Japanese. ] OKUTANI, T. 1977. Stock assessment of cephalopod resources fished by Japan. Food and Agriculture Organization, Fish- eries Technical Paper (173):1-62. RATHJEN, W. F., R. F. HIxon & R. T. HANLON. 1979. Squid fishery resources and development in the northwest Atlantic and Gulf of Mexico. Proceedings 31st Annual Gulf & Ca- ribbean Fisheries Institute, Nov. 1978, pp. 145-157. ROSENBERG, P. 1973. The giant axon of the squid: a useful preparation for neurochemical and pharmacological studies. Chapter 3. Jn: R. Fried (ed.), Methods of Neurochemistry, Vol. 4. Marcel Dekker, Inc.: New York. 332 pp. SEGAWA, S. 1987. Life history of the oval squid, Sepzoteuthis lessoniana in Kominato and adjacent waters central Honshu, Japan. Journal of the Tokyo University of Fisheries 74(2): 67-105. SEGAWwA, S., W. T. YANG, H. J. Martuy & R. T. HANLON. 1988. Illustrated embryonic stages of the eastern Atlantic Squid Loligo forbesi. The Veliger 30(3):230-243. SUMMERS, W. C. 1983. Loligo pealer. Pp. 115-142. In: P. R. Boyle (ed.), Cephalopod Life Cycles. Vol. I: Species Ac- counts. Academic Press: London. Voss, G. L. 1973. Cephalopod resources of the world. Food and Agriculture Organization, Fisheries Circular (149):1- i WorMs, J. 1983. Loligo vulgaris. Pp. 143-157. In: P. R. Boyle (ed.), Cephalopod Life Cycles. Vol. I: Species Accounts. Academic Press: London. The Veliger 35(3):242 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 NOTES, INFORMATION & NEWS Erratum, Volume 34, Number 3 A. Solem, Australian Land Snails Incorrectly merged computer files resulted in an error in the originally submitted manuscript by Alan Solem (“‘Dis- tribution and diversity patterns of Australian pupilloid land snails”), published posthumously in the July 1991 issue—34(3):233-252. The error was noticed by Vince Kessner, Darwin, Aus- tralia, for whose diligence we are grateful. According to Margaret Baker, Field Museum of Nat- ural History, Chicago, Illinois, the following entries are correct and should replace the incorrect entry on page 249 (Gyliotrachela napierana). Gyliotrachela napierana Solem, 1981 See SOLEM, 1981:90-91, figs. 1, 2, 9, 13; SOLEM, 1989:502- 503, figs. 91-93. Type locality: 5.9 km NW of Yammera Gap, Napier Ranges, Western Australia. Gyliotrachela ningbingia Solem, 1981 See SOLEM, 1981:91, figs. 3, 4, 10, 14-16, 18, 19; SOLEM, 1989:503-504, figs. 94-90. Type locality: 5.7 km N of No. 8 Bore, Ningbing Ranges, N of Kununurra, Western Australia. International Commission on Zoological Nomenclature: Applications and Opinions The following applications and opinions were published on 26 March 1992 in Volume 49, Part 1 of the Bulletin of Zoological Nomenclature. Comment or advice on the ap- plications is invited for publication in the Bulletin, and should be sent to the Executive Secretary, ICZN, % The Natural History Museum, Cromwell Road, London SW7 5BD, UK. Case 2247—Balea Gray, 1824 (Mollusca: Gastropoda): proposed conservation of name, currently in use for a genus of pulmonate gastropods, which is threatened by the senior objective synonym Strombiformis Da Cos- ta, 1778. Case 2634—Xeromunda Monterosato, 1892 (Mollusca: Gastropoda): proposed designation of Helix candiota Mousson, 1854, as the type species. Opinion 1662—Limax fibratus Martyn, 1784, and Nerita hebraea Martyn, 1786 (currently Placostylus fibratus and Natica hebraea; Mollusca, Gastropoda): specific names conserved, and Placostylus Beck, 1837: L. fi- bratus designated as the type species. Opinion 1663—Fryeria Gray, 1853, and F. rueppelii Bergh, 1869 (Mollusca: Gastropoda): conserved. Opinion 1664—RISSOIDAE Gray, 1847 (Mollusca: Gas- tropoda): given precedence over —TRUNCATELLIDAE Gray, 1840. Opinion 1665—Potamilus Rafinesque, 1818 (Mollusca: Bivalvia): not suppressed. American Society of Zoologists 1992 Meeting The American Society of Zoologists, with the American Microscopical Society, Animal Behavior Society, The Ca- nadian Society of Zoologists, The Crustacean Society, and The International Association of Astacology, will hold its annual meeting in December, in Vancouver, B.C., Canada. The deadline for abstracts is 31 July 1992. Several interesting symposia and workshops are sched- uled. For more information, contact: Mary Adams-Wiley, Executive Officer, ASZ, 104 Sirius Circle, Thousand Oaks, CA 91360. Phone: (805) 492-3585. The Veliger 35(3):243 (July 1, 1992) THE VELIGER © CMS, Inc., 1992 BOOKS, PERIODICALS & PAMPHLETS The Malacofauna of Hong Kong and Southern China, Volumes I and II The Malacofauna of Hong Kong and Southern China, edited by BRIAN MorTon. 1980 [reprinted 1986]. vi + 345 pp. Price: $40. The Malacofauna of Hong Kong and Southern China, II, edited by BRIAN MoRTON & DAviD DUDGEON. 1985. Volume 1:vii + 362 pp.; Volume 2:vill + 363-681 pp. Price: $80. Hong Kong University Press, 139 Pok- fulam Road, Hong Kong. These important volumes contain more useful infor- mation about marine mollusks than most journals do in two or three years. Some non-marine mollusks are also treated. The books are the products of the periodic work- shops on mollusks pioneered by Brian Morton. The session that led to the first volume was held in 1977, the second in 1983. They contain papers that will be of interest and utility for nearly every malacologist. The first volume is divided into sections on taxonomy, ecology, and functional morphology. Highlights of the tax- onomic section include studies on the bivalve families Ve- neridae and Corbiculidae, gastropod limpets and the Nerit- idae, and chitons. The ecology section includes papers on the diets of predatory gastropods and on the correlation of gill function to intertidal distribution of bivalves. The func- tional morphology section has papers on the bivalves Cae- cella, Veremolpa, Coralliophaga, Vulsella, and Arcuatula, and the gastropod Crepidula. The second workshop resulted in two volumes, divided into sections on taxonomy, morphology, ecology, physiol- ogy, and behavior. The taxonomy section has particularly important papers on the Donacidae, Mytilidae, Potami- didae, Caecidae, and Vermetidae. The morphology section has key papers on the Caecidae, Arcidae, Isognomonidae, Veneridae, and bivalve stomachs. The ecology section fea- tures papers on Pinna, Littorinidae, Perna, and several freshwater gastropods, including Biomphalaria. The phys- iology section has papers on several bivalves and on the thermal tolerances of mangrove gastropods. The behavior section includes prey selection by Octopus, naticid preda- tion, nudibranch feeding, and freshwater gastropod for- aging strategies. A third workshop has recently been held and, no doubt, interesting reports will appear in print in two or three years. Gene Coan Two Volumes Noted Manglares y Hombres del Pacifico Colombiano, by HENRY VON PRAHL, JAIME R. CANTERA & RAFAEL CONTRERAS. Colombia (Fondo para la Proteccion del Medio Am- biente “José Celestino Mutis”’). 193 pp., illustrated. April 1990. Covers the distribution and biology of mangrove eco- systems along the Pacific coast of Colombia. Pages 113- 116 list mollusks. Dvustvorchatye molliuski shel'fou 1 kontinental’nogo sklona severnoi Patsifiki [Bivalve Mollusks of the Shelf and Con- tinental Slope of the North Pacific Ocean] , by ALEXANDER I. KAFANOV. Post-4 July 1991. Akad. Nauk. SSSR, Dal’nevostochnoe Otdelenie, Institut Biologii Moria. 198 pp. A checklist covering the North Pacific, south to Japan and central California. Localities are indicated by a nu- merical code keyed to maps, and depths are given. The bibliography is brief and the index is thorough. Gene Coan Information for Contributors Manuscripts Manuscripts must be typed on white paper, 82” by 11”, and double-spaced throughout (including references, figure legends, footnotes, and tables). If computer generated copy is to be submitted, margins should be ragged right (v.e., not justified). To facilitate the review process, manuscripts, including figures, should be submitted in triplicate. The first mention in the text of the scientific name of a species should be accompanied by the taxonomic authority, including the year, if possible. Underline scientific names and other words to be printed in italics. Metric and Celsius units are to be used. The sequence of manuscript components should be as follows in most cases: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, lit- erature cited, figure legends, figures, footnotes, and tables. The title page should be on a separate sheet and should include the title, author’s name, and address. The abstract should describe in the briefest possible way (normally less than 200 words) the scope, main results, and conclusions of the paper. Literature cited References in the text should be given by the name of the author(s) followed by the date of publication: for one author (Smith, 1951), for two authors (Smith & Jones, 1952), and for more than two (Smith ef al., 1953). The “literature cited” section must include all (but not additional) references quoted in the text. References should be listed in alphabetical order and typed on sheets separate from the text. Each citation must be complete, with all journal titles wnabbreviated, and in the following form: a) Periodicals Cate, J. M. 1962. On the identifications of five Pacific Mitra. The Veliger 4:132-134. b) Books Yonge, C. M. & T. E. Thompson. 1976. Living marine molluscs. Collins: London. 288 pp. c) Composite works Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135. In: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stanford Univ. Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Figures and plates Figures must be carefully prepared and should be submitted ready for publication. Each should have a short legend, listed on a sheet following the literature cited. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photographs for half-tone plates must be of good quality. They should be trimmed off squarely, arranged into plates, and mounted on suitable drawing board. Where necessary, a scale should be put on the actual figure. Preferably, photographs should be in the desired final size. It is the author’s responsibility that lettering is legible after final reduction (if any) and that lettering size is appropriate to the figure. Charges will be made for necessary alterations. Processing of manuscripts Upon receipt each manuscript is critically evaluated by at least two referees. Based on these evaluations the editor decides on acceptance or rejection. Acceptable manuscripts are returned to the author for consideration of comments and criticisms, and a finalized manuscript is sent to press. The author will receive from the printer two sets of proofs, which should be corrected carefully for printing errors. At this stage, stylistic changes are no longer appropriate, and changes other than the correction of printing errors will be charged to the author at cost. One set of corrected proofs should be returned to the editor. An order form for the purchase of reprints will accompany proofs. If reprints are desired, they are to be ordered directly from the printer. Send manuscripts, proofs, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. CONTENTS — Continued Taxonomic re-evaluation and description of Gari radiata (Dunker in Philippi, 1845) (Bivalvia: Tellinoidea: Psammobiidae) RICHARD (Gio VWiAILIGAIN 228 5G SiGek cok Skye omen Astle Aces Tec rn Ge fel aa Aen ae 226 Embryonic stages of Loligo bleekeri Keferstein (Mollusca: Cephalopoda) GYEONG HUN BAEG, YASUNORI SAKURAI, AND KENJI SHIMAZAKI ........ 234 NOWESMINEORIMWADEIOING Se INTE WES ee eens greece 242 BOOKS, PERIODICALS & PAMPHLETS QL 401 Ae THE VELIGER A Quarterly published by CALIFORNIA MALACOZOOLOGICAL SOCIETY, INC. Berkeley, California R. Stohler, Founding Editor Volume 35 October 1, 1992 Number 4 / he Pe AA CONTENTS \ ULE | & WYy2 Reproduction and development of trochacean gastropods ~ LipparictS BRANES Le CAROLE? Sse ICKINUAN ceri TOE gt es ae ee ——— 245 Systematic review of the family Choristellidae (Archaeogastropoda: Lepetellacea) with descriptions of new species J[ASNIUSS 1 (SINAC ONES Te tae cite ee er aa ee elt ares En Re Zi On the anatomy and relationships of the Choristellidae (Archaeogastropoda: Lepetelloidea) CERTIARD EEUASZ PRUNARGH bso test Nimwaaieks i es Gan iuhl oadokas ein dmanala ends 295 The fine structure of the columellar muscle of some gastropod mollusks MES ERESCURAPAN DI ALYIN: FTODGSON) 4.00 fs 8 Ste oes La eye le 308 Egg mass and intracapsular development of Cypraea caputdraconis Melvill, 1888, from Easter Island (Gastropoda: Cypraeidae) CECILIA OsORIO, CARLOS GALLARDO, AND HUGO ATAN ................ 316 New morphologic and geographic data on the neritid gastropod Nerita (Thelio- styla) triangulata Gabb, 1869, from the Eocene of the Pacific coast of North America RICHARD leg SOWIRES nn wer earns ass Ken ama iy Goel uA are, a wh avy Meek a cue 323 A new aeolid (Gastropoda: Nudibranchia) from the Atlantic coasts of the southern Iberian Peninsula J. L. CERVERA, J. C. GARCIA-GOMEZ, AND P. J. LOPEZ-GONZALEZ ...... 330 CONTENTS — Continued The Veliger (ISSN 0042-3211) is published quarterly on the first day of January, April, July, and October. Rates for Volume 36 are $32.00 for affiliate members (including domestic mailing charges) and $60.00 for libraries and nonmembers (7n- cluding domestic mailing charges). For subscriptions sent to Canada and Mexico, add US $4.00; for subscriptions sent to addresses outside of North America, add US $8.00, which includes air-expedited delivery. Further membership and subscription infor- mation appears on the inside cover. The Veliger is published by the California Ma- lacozoological Society, Inc., % Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105. Second Class 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 open to original papers pertaining to any problem concerned with mollusks. This is meant to make facilities available for publication of original articles from a wide field of endeavor. Papers dealing with anatomical, cytological, distributional, eco- logical, histological, morphological, physiological, taxonomic, evolutionary, etc., aspects of marine, freshwater, or terrestrial mollusks from any region will be considered. Short articles containing descriptions of new species or lesser taxa will be given preferential treatment in the speed of publication provided that arrangements have been made by the author for depositing the holotype with a recognized public Museum. Museum numbers of the type specimen must be included in the manuscript. Type localities must be defined as accurately as possible, with geographical longitudes and latitudes added. Very short papers, generally not exceeding 500 words, will be published in a column entitled “NOTES, INFORMATION & NEWS”; in this column will also appear notices of meetings, as well as news items that are deemed of interest to our subscribers in general. Editor-in-Chief David W. Phillips, 2410 Oakenshield Road, Davis, CA 95616, USA Editorial Board Hans Bertsch, National University, Inglewood, California James T. Carlton, University of Oregon Eugene V. Coan, Research Associate, California Academy of Sciences, San Francisco J. Wyatt Durham, University of California, Berkeley William K. Emerson, American Museum of Natural History, New York Terrence M. Gosliner, California Academy of Sciences, San Francisco Cadet Hand, University of California, Berkeley Carole S. Hickman, University of California, Berkeley David R. Lindberg, University of California, Berkeley James H. McLean, Los Angeles County Museum of Natural History Frank A. Pitelka, University of California, Berkeley Peter U. Rodda, California Academy of Sciences, San Francisco Clyde F. E. Roper, National Museum of Natural History, Washington Barry Roth, Santa Barbara Museum of Natural History Judith Terry Smith, Stanford University Ralph I. Smith, University of California, Berkeley Wayne P. Sousa, University of California, Berkeley T. E. Thompson, University of Bristol, England Membership and Subscription Affiliate membership in the California Malacozoological Society is open to persons (no institutional memberships) interested in any aspect of malacology. As an affiliate member, a person may subscribe to The Veliger for US $32.00 (Volume 36), which now includes mailing charges to domestic addresses. There is a one-time membership fee of US $2.00, after payment of which, membership is maintained in good standing by the timely renewal of the subscription; a reinstatement fee of US $3.00 will be required if membership renewals do not reach the Society on or before October 1 preceding the start of the new Volume. If a receipt is required, a self-addressed, stamped envelope (or in the case of foreign members, the envelope and two International Postal Reply coupons) should be included with the membership or subscription request. The annual subscription rate to The Veliger for libraries and nonmembers is US $60.00 (Volume 36), which now includes mailing charges to domestic addresses. For subscriptions sent to Canada and Mexico, add US $4.00; for subscriptions sent to addresses outside of North America, add US $8.00, which includes air-expedited delivery. Memberships and subscriptions are by Volume only (January 1 to October 1) and are payable in advance to California Malacozoological Society, Inc. Single copies of an issue are US $25.00 plus postage. Send all business correspondence, including subscription orders, membership applications, payments for them, changes of address, to: The Veliger, 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. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616, USA. THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER. The Veliger 35(4):245-272 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 Reproduction and Development of ‘Trochacean Gastropods by CAROLE S. HICKMAN Department of Integrative Biology and Museum of Paleontology,' University of California, Berkeley, California 94720, USA Abstract. A compilation of comparative data on reproduction and development of trochacean gas- tropods (Prosobranchia: Archaeogastropoda) is presented and analyzed within the framework of the revised suprageneric classification of the superfamily. The traditional portrayal of the male and female reproductive anatomy and gametes as simple and unspecialized is contrasted with the diversity and complexity of elaborations of the epipodium that may function in sperm transfer, elaborations of glandular mantle tissue (urogenital papillae) that produce a variety of primary and secondary gelatinous egg coverings that affect buoyancy and clumping properties of gametes, and evidence of unexpected modi- fications of trochacean sperm. Considerable variation occurs in the periodicity and control of both gametogenesis and spawning, but there is no common or dominant pattern characterizing either higher taxa or ecological settings. In addition to the two common modes, broadcast spawning with pelagic development and spawning of benthic egg masses with benthic development, brooding has arisen in- dependently in at least one turbinid subfamily and three trochid subfamilies. A fourth mode, mixed development, occurs in several taxa that extrude eggs, either as strings that sink to the bottom or that are loosely tacked to the bottom for a benthic phase preceding the pelagic phase. The positions of hatching and settlement within the developmental sequence are highlighted as evolutionarily significant variables in a table and comparative diagram of trochacean developmental stages and events. Hatching of pelagic larvae may occur as early as the trochophore and as late as the post-torsional veliger stage, and settlement may precede the completion of torsion. Popular classifications of molluscan developmental modes and “strategies” are not useful because they underestimate the range of variation in trochacean development. The possibility of feeding by planktic trochacean larvae, as well as the possibility of feeding by benthic larvae between settlement and metamorphosis, merits renewed attention. Problems of interpretation of the trochacean protoconch include documentation of the nature and extent of heterostrophy and investigation of a poorly understood process of asymmetric mechanical deformation of the organic larval shell that precedes mineralization. INTRODUCTION As one of the most primitive groups of living marine gas- tropods, trochaceans have been characterized as having uniformly simple reproductive systems and simple repro- ductive and developmental patterns featuring broadcast spawning, external pelagic fertilization, and a brief, non- feeding (lecithotrophic) planktic larval stage. Although the conventional generalities provide a convenient backdrop for portraying the evolution of caenogastropod reproduc- tive diversity, they mask the wellspring of diversity and ! This is contribution Number 1560 of the Museum of Pale- ontology. flexibility that have served this geologically ancient group throughout 500 million years of evolutionary experimen- tation. Data on reproductive biology and early develop- ment of living trochaceans are scattered in the literature. Because of a recent increase in the number of more detailed studies of individual species, the time is ripe to produce a synthesis and interpretation. Until recently, few reproductive or developmental stud- ies have been experimental or carried out under controlled conditions, and UNDERWOOD (1972b) attributed this to the difficulty of inducing spawning and of rearing of larvae under laboratory conditions. Improved techniques for maintaining reproductively mature adults, inducing spawning, and culturing embryos and larvae (see STRATH- Page 246 MANN, 1987) provide new opportunities to move beyond anecdotal accounts of individual events to characterization of the range of reproductive patterns that have evolved within the superfamily. Most observations to date are of temperate intertidal species, and the taxa that have been studied most inten- sively are trochids in the subfamilies Margaritinae, Tro- chinae (tribes Gibbulini and Cantharidini), and Callios- tomatinae. Little is known about reproduction and development in the turbinid subfamilies, and there are no data for members of the Skeneidae. Studies of individual trochacean species have never been set in a comparative systematic framework, nor has there been an adequate trochacean classification to serve as a basis for comparative study. It is therefore appropriate to evaluate both the strengths and the gaps in our knowledge of trochacean reproduction and development in the context of the new classification and systematic revision of the superfamily (HICKMAN & MCLEAN, 1990), summarized in Table 1. This paper is a supplement to that systematic revision and also a critique of some of the theory in mol- luscan reproduction and development. Controversial issues examined in this paper include the possibility of evolution of a penis or penis-like structures in trochaceans, the possibility of ovoviviparity, the possi- bility of a special type of heterostrophy in trochacean shell growth, and the possibility of feeding in both pelagic and benthic trochacean larvae prior to metamorphosis. Major objectives of this paper are: (1) to document vari- ability in trochacean reproductive and developmental pat- terns within the systematic framework of the superfamily, (2) to identify gaps in comparative data and opportunities for future research, (3) to introduce an expanded set of spawning categories with examples of species in each cat- egory, (4) to present a table of stages in trochacean de- velopment that identifies major events within each stage and emphasizes the events that are variable in their timing, (5) to provide a summary set of illustrations of trochacean developmental stages, and (6) to illustrate the develop- mental patterns in trochaceans in a comparative manner that emphasizes their variability and versatility in contrast to those of caenogastropods, the group that heretofore has provided much of the vocabulary and theoretical frame- work for characterizing prosobranch reproduction and de- velopment. MALE anD FEMALE REPRODUCTIVE ANATOMY Gonads and Reproductive Tract Trochaceans are all dioecious, but males and females are not usually distinguishable by shell features or external features of the animal. When removed from the shell, sexes are distinguishable without dissection, if the gonads are ripe, by the distinctly granular appearance of the ovary in The Veliger, Vol. 35, No. 4 contrast to the smooth appearance of the testis. While color alone is not a reliable guide, female gonads are frequently green or greenish gray and male gonads are frequently creamy white or pink (Hickman, personal observation). The major features of the male and female reproductive systems are shown diagrammatically in Figure 1. PURCHON (1968) characterized this grade of reproductive organi- zation as so simple as to approximate that of the hypo- thetical ancestral mollusk. Even though the trochacean genital tract lacks the degree of distal elaboration seen in the pallial oviduct (and associated glands) of more derived prosobranchs, the reproductive plumbing is more compli- cated than that of patellogastropods. It is made up of the same three basic components identified by FRETTER (1946) as common to all prosobranchs: gonadal, renal, and pallial. In both sexes the ripe gametes pass from the gonad (1) through a short gonoduct to the renopericardial duct (2) through the renopericardial duct to the right kidney, which in turn opens into the mantle cavity (3) through mantle tissue which may form a simple aperture (especially in males), an aperture surrounded by glandular lips, or (es- pecially in females) a greatly enlarged glandular duct of mantle tissue protruding into the mantle cavity as a “‘uro- genital papilla” (FRETTER, 1946, 1955; FRETTER & GRA- HAM, 1962; HADFIELD & STRATHMANN, 1990). Function of Female Urogenital Papillae and Pallial Glands Glandular pallial tissue secretes mucus that constitutes at least part of the secondary jelly coatings in species that spawn eggs in strings or benthic egg masses (FRETTER & GRAHAM, 1962). However, HADFIELD & STRATHMANN (1990) were unable to find a correlation between the amount of jelly produced and the sizes of urogenital papillae, and the function of papillae merits additional study. The pres- ence of a prominent papilla cannot be used as a criterion for inferring benthic development: a large papilla is present in females of species of Calliostoma that undergo predom- inantly pelagic development (HADFIELD & STRATHMANN, 1990:fig. 8), whereas there is little difference between the slightly enlarged lips surrounding the urogenital openings in males and females of cantharidine species that deposit strings of eggs on seagrass blades (Hickman, personal ob- servation). GERSCH (1936) suggested that secretions of the hypo- branchial gland (mucus-secreting glandular pallial tissue) contribute to the formation of secondary jelly coatings, but there are no data to support or refute this hypothesis. The function of the prosobranch hypobranchial gland is spec- ulative, and the histochemistry of its secretions is not well documented (FRETTER & GRAHAM, 1962). Gersch’s in- ference was based on observation that hypobranchial se- cretion was more copious during the breeding season in Gibbula tumida (Montagu, 1803), a species that spawns benthic egg masses. C. S. Hickman, 1992 Page 247 Table 1 Outline classification of trochacean gastropods identifying subfamilies and tribes for which there are published data on reproduction and development (solid circles) and subfamilies and tribes for which no data are available (open circles). Isolated observations or reports of single features do not qualify a taxon for a solid circle, and open circles identify taxa in which full comparative studies of reproduction and development are needed. Superfamily TROCHACEA Data No data Family TURBINIDAE Rafinesque, 1815 Informal Group LIOTINAE + ANGARIINAE Subfamily LIOTIINAE Adams & Adams, 1854 Subfamily ANGARIINAE Thiele, 1921 Informal Group MOELLERIINAE + COLLONIINAE Subfamily MOELLERIINAE Hickman & McLean, 1990 Subfamily COLLONIINAE Cossmann, 1916 Informal Group PRISOGASTERINAE + TURBININAE Subfamily PRISOGASTERINAE Hickman & McLean, 1990 Subfamily TURBININAE Rafinesque, 1815 ® Informal Group GABRIELONINAE + TRICOLIINAE + PHASIANELLINAE Subfamily GABRIELONINAE Hickman & McLean, 1990 O Subfamily TRICOLIINAE Woodring, 1928 @ Subfamily PHASIANELLINAE Swainson, 1840 oO Family TROCHIDAE Rafinesque, 1851 Informal Group TEGULINAE + EUCYCLINAE + MARGARITINAE Subfamily TEGULINAE Kuroda, Habe, & Oyama, 1971 @) Subfamily EUCYCLINAE Koken, 1897 Tribe EUCYCLINI Koken, 1897 @) Tribe CHILODONTINI Wenz, 1938 ® Tribe CALLIOTROPINI Hickman & McLean, 1990 e) Subfamily MARGARITINAE Stoliczka, 1868 Tribe MARGARITINI Stoliczka, 1868 @ Tribe GAZINI Hickman & McLean, 1990 e) Informal Group TROCHINAE + STOMATELLINAE + CALLIOSTOMATINAE + SOLARIELLINAE Subfamily TROCHINAE Rafinesque, 1815 Tribe TROCHINI Rafinesque, 1815 Tribe GIBBULINI Stoliczka, 1868 Tribe CANTHARIDINI Cotton, 1959 Subfamily STOMATELLINAE Gray, 1840 Subfamily CALLIOSTOMATINAE Thiele, 1924 Subfamily SOLARIELLINAE Powell, 1951 O Informal Group HALISTYLINAE + LIRULARIINAE + UMBONIINAE Subfamily HALISTYLINAE Keen, 1958 oO Subfamily LIRULARIINAE Hickman & McLean, 1990 @ Subfamily UMBONIINAE Adams & Adams, 1854 Tribe MONILIINI Hickman & McLean, 1990 Oo Tribe BANKIVIINI Hickman & McLean, 1990 O Tribe UMBONIINI Adams & Adams, 1854 @ Subfamilies of uncertain affinity Subfamily CATAEGINAE McLean & Quinn, 1987 Subfamily TROCHACLIDINAE Thiele, 1928 Subfamily THYSANODONTINAE Marshall, 1988 Family SKENEIDAE Clark, 1851 © O10 OO Oo O00 However, the structure is present in both males and fe- males, and its function is unclear. FRETTER & GRAHAM (1977) referred to it instead as a “‘postoptic tentacle” and Epipodial Elaborations and the Possibility of a Trochacean Penis Isolated reports of a penis in trochaceans have been ignored or treated with skepticism. In the family Skenei- dae, CLARK (1852), JEFFREYS (1865), and HOISATER (1968) described a penis posterior to the right cephalic tentacle. MARSHALL (1988b) as a “suboptic tentacle” that may be compound. A more compelling account of a trochacean penis occurs in an overlooked paper by DALL (1889a), who claimed to Page 248 gd Q(te) ane A rk rko gd g(Ov) rpc B rk rko ugp gd g (Ov) rko The Veliger, Vol. 35, No. 4 rg =v rko Figure 1 Male and female reproductive anatomies of trochacean gastropods. A. Schematic depiction of the male reproductive system (after FRETTER & GRAHAM, 1962). B. Schematic depiction of a female reproductive system with a small urogenital papilla (after FRETTER & GRAHAM, 1962). C. Schematic depiction of the female reproductive system with an enlarged urogenital papilla (after FRETTER & GRAHAM, 1962). D. Adult female of a Prothalotia lehmanni (Menke, 1843) (subfamily Trochinae, tribe Cantharidini) in mantle-cavity dissection with the mantle cut along the left side and laid back to the right to show the position of the urogenital papilla and urogenital opening (rko) relative to the rectum (r), hypobranchial glands (hg), and simple left kidney opening (Iko). In both males and females, gametes pass from the single gonad (g), an ovary (ov) or testis (te), through two sections of duct, the first of gonadal derivation (gd) and the second of nephridial origin (renopericardial duct, rpc), and discharge into the mantle cavity via an opening that is simple in males but which may be further elaborated in females into a glandular duct or urogenital papilla (ugp). have observed an “intromittent male organ” in “several” deep-sea trochaceans, proposing that it was a primitive structure that has been lost in shallow-water species. He gave a detailed description for the eucycline species Cal- liotropis infundibulum (Watson, 1879) of a small tubular structure lying above an elaborately modified right epi- podial flap and stated: “The object of this apparatus is self-evident. The cylinder serves as a conduit for the seminal fluid ejected from the verge.” This remarkable structure is almost certainly the same thing that WAREN & BOUCHET (1989) have described in Bathymargarites symplector Wa- ren & Bouchet, 1989. Although they state (p. 92) that it is the left neck lobe that has been modified, this is clearly a lapsus, because they illustrate it (figs. 105, 107) on the right side of the animal. It is debatable whether the structure described by DALL (1889a) and WAREN & BOUCHET (1989) should be called a penis or copulatory structure, insofar as it is not designed for insertion into the female and internal fertilization. However, any elaboration of the male exhalant right neck lobe that can direct the delivery of sperm is a step away from broadcast spawning in the direction of internal fer- tilization. GAMETOGENESIS, GAMETES, AND REPRODUCTIVE PERIODICITY UNDERWOOD (1972a) provided a detailed description of the stages in spermatogenesis and oogenesis from histo- logical sections of the gonads of three British gibbuline C. S. Hickman, 1992 trochids that he sampled monthly. The most significant result of this program was rigorous documentation of a seasonal pattern of fluctuation in both the proportion of mature oocytes and the quantity of ripe sperm in two of the species [Monodonta lineata (da Costa, 1778) and Gib- bula umbilicalis (da Costa, 1778)] in contrast to constant proportions in the third species (G. cineraria Linnaeus, 1758). Comparative histological data are lacking for rep- resentatives of other trochacean subfamilies and tribes and are required before patterns of gametogenesis can be ex- amined in a systematic context. Eggs and Primary Gelatinous Egg Coverings The typical mature trochacean oocyte is a yolk-rich structure, of 150-300 um diameter, enclosed by a clear membrane (the “envelope” of STRATHMANN [1987], “egg covering” of FRETTER & GRAHAM [1962], and “primary egg membrane,” “‘vitelline membrane,” or “fertilization membrane” of various other authors). The granular, yolky oocyte and the clear, enclosing membrane that is secreted by the oocyte are in turn covered by gelatinous coatings. The complex gelatinous coatings of trochacean eggs may be a unique derived feature of the superfamily. There is no convincing documentation of analogous or homologous coatings in studies of more primitive prosobranch taxa, and they merit more careful study and characterization. Three different kinds of trochacean egg jellies are recog- nized herein: (1) an inner primary (ovarian origin) layer that surrounds mature oocytes, (2) an outer primary (ovar- ian origin) layer that swells upon contact with seawater and may disperse soon after spawning, and (3) secondary (pallial origin) jellies that bind individual oocytes together in strings, clumps, or masses. The composition, properties, and function of the pri- mary and secondary egg jellies are unresolved. Descrip- tions of gelatinous coatings of trochacean eggs have not used a uniform terminology (cf., LEBOUR, 1937; FRETTER, 1955, 1984; FRETTER & GRAHAM, 1962; DESAI, 1966; UNDERWOOD, 1972a, b; HESLINGA, 1981; STRATHMANN, 1987; HOLYOAK, 1988a; HADFIELD & STRATHMANN, 1990), and the primary and secondary jellies have been treated as a single homogeneous unit in some accounts. While there appear to be two distinct primary envelopes under light microscopy (Hickman, personal observation), FRETTER & GRAHAM (1962:322) suggested that they are fundamentally similar, with a line of contact between a denser inner jelly and a thinner superficial layer. Even if the two layers are not chemically distinct or separated by a membrane, they have physical properties that cause them to behave differently. DESAI (1966) observed (in Monodon- ta lineata) that on contact with seawater there was rapid swelling of the outer layer that made the eggs buoyant. He also noted that the outer coating dispersed within 20 minutes of spawning and that eggs sank. UNDERWOOD (1972b) observed (in Gibbula cineraria) not only that the outer coating dispersed soon after spawning, but also that Page 249 it was resistant to the penetration of sperm. This led him to propose an alternative function, that of initially blocking polyspermy. HADFIELD & STRATHMANN (1990) also noted dispersion of the jelly coat following spawning in Mar- garites pupillus (Gould, 1849) but did not speculate on its function. Swelling of jelly coats after spawning has been noted in other taxa. HOLYOAK (1988a) reported expansion in two envelopes, which he described as an outer “gelatinous coat” and a “space” between the egg and the gelatinous coat, in Calhiostoma ligatum (Gould, 1849). He did not indicate how long the swollen gelatinous envelope persisted. HEs- LINGA (1981) reported a single “pitted” jelly coat sur- rounding the eggs of Trochus niloticus (Linnaeus, 1758) that underwent post-fertilization swelling and persisted until hatching, leading him to propose that it functioned to reduce the sinking rate of embryos. Primary gelatinous egg coverings are secreted in the ovaries, as originally suggested by FRETTER & GRAHAM (1962). Direct evidence of ovarian origin is provided by UNDERWOOD (1972a), who observed gradual deposition of the jelly coat late in oogenesis and used presence of the jelly coat to define mature oocytes. UNDERWOOD (1972b) further noted that mature oocytes dissected from the ovary for artificial fertilization showed the same swelling of the outer layer of the jelly coat on contact with seawater. Swelling of mature oocytes on contact with seawater is easily observed in fresh ovarian dissections and is common in trochacean species I have examined. Secondary Egg-Binding Jellies The individual jelly-covered oocytes are sometimes held together by another type of gelatinous substance (herein referred to as egg-binding jellies) at the time of spawning. When egg-binding jellies are secreted by species with planktic development, they are of a soft consistency that holds eggs together loosely and temporarily in strings or clusters that disperse rapidly in moving water. These strings may persist for longer periods and sink to the bottom as a unit when spawned in still water. This soft type of binding jelly is described in Calliostoma ligatum by HaAD- FIELD & STRATHMANN (1990). In trochaceans with benthic embryonic and larval development, binding jellies are of a firmer consistency and are fastened to the substrate, persisting through fertilization and part or all of devel- opment. This firm type of binding jelly is characteristic of the attached egg masses of Calliostoma zizyphinum (Lin- naeus, 1758) and Cantharidus exasperatus (Pennant, 1777) (FRETTER & GRAHAM, 1977; FRETTER, 1984). Egg-binding jellies may be secreted, at least in part, by glandular mantle tissue at the opening of the female uro- genital duct into the mantle cavity (FRETTER, 1955, 1984; FRETTER & GRAHAM, 1962). The glandular tissue ranges from a relatively simple thickening in the mantle cavity surrounding the lips of the urogenital opening to a prom- inent, elongate, protruding papilla. These structures have Page 250 been described and/or illustrated for species of Calliostoma, Margarites, and Lirularia that produce egg-binding jellies (FRETTER & GRAHAM, 1962; HADFIELD & STRATHMANN, 1990). As noted above, it is possible that secretions of the hypobranchial gland also contribute to the formation of egg-binding jellies. Composition and Properties of Primary and Secondary Egg Coverings There is no documentation of the chemical or physical properties of trochacean egg coverings. UNDERWOOD (1972b) reported that the dispersal of the outer layer of primary egg jelly was retarded in water containing sperm, suggesting a chemical interaction. Secondary egg-binding jellies of differing binding ability, differing densities, and differing persistence times could have a significant influ- ence on the probability of fertilization and on the dispersal of eggs following fertilization and prior to hatching. THomMas (1992) challenges the assumption that gametes of broadcast spawning invertebrates mix freely in the water column, and her experimental studies of the fluid me- chanics of spawning in polychaetes and sea urchins show that gamete buoyancy and gamete clumping can have ma- jor effects on fertilization success. The broad range of gamete spawning methods and inferred gamete properties in trochaceans establishes this group as an ideal target for comparative extension of this research. The primary and secondary gelatinous egg coverings of trochaceans are not homologous with the proteinaceous “capsules” and other egg investments of other proso- branchs (neritomorphs and caenogastropods) and opistho- branchs, which are secreted by distinctive, derived capsule and albumen glands that are part of the terminal repro- ductive tract. Classifications of “egg capsules” that have not noted this distinction (e.g., AMIO, 1963:table 7) result in categories that are artificial in their phylogenetic con- stituency. The Problem of Trochacean Egg Size Measurements of trochacean egg sizes in the literature are difficult to interpret if the author has not clearly stated what has been measured. Egg measurements may extend beyond the diameter of the yolky egg to include the thick- ness of the clear envelope surrounding the egg, and the thickness of the inner and outer jelly coatings. Separate measurements of the thicknesses of the various egg cov- erings are likewise difficult to interpret and compare when measurements have been made after spawning or after contact with seawater. This is because individual layers swell and expand differentially both upon contact with seawater and following fertilization. To minimize the larg- est source of error, the ranges of egg diameters reported in this review (Table 2) are based exclusively on studies in which authors differentiated between the measurement of the membrane-bounded egg itself and that of the ge- latinous egg covering(s). The Veliger, Vol. 35, No. 4 Sperm Trochacean sperm are presumed to be of the primitive type (sensu RETZIUs, 1906; FRANZEN, 1956), with a clearly differentiated head (acrosome and nucleus), middle piece, and tail that are traditionally associated with external fer- tilization. There are few comparative data in the literature, however, and this characterization is extrapolated from ultrastructural studies of members of a single trochid sub- family (Trochinae, tribes Trochini and Gibbulini) (see KOHNERT & STORCH, 1983; KOIKE, 1985; AZEVEDO et al., 1985). While more recent comparative study by John Hea- ly (personal communication) confirms the presence of structurally simple sperm in other trochacean subfamilies, there are several unexpected aberrations. HEALY (1989) documents modified spermatozoa in the primitive eucycline trochid Calliotropis glyptus (Watson, 1879), characterized by a deep embedding of the acrosomal complex within the nucleus, effectively eliminating the midpiece region. This is the only report of this form of shortening and invagi- nation in molluscan spermatozoa, although the condition occurs in some polychaetes, arthropods, and echinoderms (see HEALY, 1989:17, for references). HEALY (1990) reports the presence of dimorphic sperm in the skeneid Zalipas laseroni Kershaw, 1958. The pres- ence of both uniflagellate euspermatozoa and multiflagel- late paraspermatozoa, conventionally associated with cae- nogastropods and internal fertilization, provides further evidence of unsuspected diversity within the Trochacea. Periodicity and Control of Gametogenesis In trochaceans with seasonal reproduction, the onset of gametogenesis has been correlated both with rising sea temperatures (LASIAK, 1987) and declining sea tempera- tures (GRANGE, 1976), although it is the seasonal avail- ability of food that is most likely to determine whether gametogenesis can occur. Seasonal ripening of gametes is one mechanism for increasing the probability of repro- ductive success and is common in marine invertebrates (GIESE, 1959), but it is by no means the exclusive pattern in trochaceans, many of which are continuous breeders. Both patterns may be present within a single trochacean genus or within a single habitat. For example, UNDERWOOD (1972a) has reported both seasonal and continuous breed- ing in the two co-occurring species of Gibbula at Plymouth, England. There are two kinds of continuous breeding in trocha- ceans. Some species do not show any seasonal change in the appearance of the gonad and are capable of releasing ripe sperm and eggs continuously throughout the year (UNDERWOOD, 1974). In other species there are reports of synchronized intensified gametogenesis (inferred either from gonadal tissue weight or from histological examination of gonads and determination of proportions of mature ga- metes) that have been interpreted as indicating spawning peaks (JOSKA & BRANCH, 1983; JOLL, 1980). For example, mature individuals of Oxystele variegata (Anton, 1839) un- C. S. Hickman, 1992 Page 251 Table 2 Classification of trochacean spawning methods. Species are listed in alphabetical order within each of the six categories along with literature citations for each species. Species Reference BROADCAST SPAWNERS (range of egg sizes: 75-200 um) Austrocochlea constricta Underwood, 1974 Cantharidus coruscans Simpson, 1977 Gibbula cineraria Robert, 1902; Lebour, 1937; Fretter & Graham, 1977; Underwood, 1972b Gibbula divaricata Bandel, 1982 Gibbula magus Robert, 1902; Lebour, 1937 Gibbula umbilicalis Robert, 1902; Lebour, 1937 Monodonta australis Lasiak, 1987 Monodonta lineata Lebour, 1937; Desai, 1966; Fretter & Graham, 1977 Oxystele tabularis Lasiak, 1987 Oxystele variegata Joska & Branch, 1983; Lasiak, 1987 Subninella undulata Underwood, 1974 Tegula excavata Lewis, 1960 Tricolia pullus Lebour, 1937 Trochus niloticus Moorhouse, 1932; Heslinga, 1981 Umbonium vestiartum Berry, 1986 BENTHIC EGG-MASS SPAWNERS (range of egg sizes: 50-200 um) Astrea caelata Lewis, 1960 Cantharidus exasperatus Robert, 1902; Lebour, 1937; Fretter & Graham, 1977; Bandel, 1982 Cantharidus japonicus Amio, 1963 Cantharidus striatus Robert, 1902; Lebour, 1937 Euchelus gemmatus Duch, 1969 Gibbula adansoni Bandel, 1982 Gibbula drepanensis Bandel, 1982 Gibbula tumida Gersch, 1936; Lebour, 1937 Lirularia succincta Hadfield & Strathmann, 1990 Margarella antarctica Picken, 1979 Margarites helicinus Thorson, 1935; Lebour, 1937 Margarites groenlandicus Thorson, 1935; Fretter & Graham, 1977 Margarites marginatus Hadfield & Strathmann, 1990 Skenea serpuloides Lebour, 1937; Fretter & Graham, 1977 Tricolia pulloides Mooers, 1981 Tricolia speciosa Bandel, 1982 Trochus erythraeus Gohar & Eisawy, 1963 SPAWNERS OF LOOSELY TACKED EGG STRINGS (range of egg sizes: 170-300 wm) Calhiostoma papillosum Robert, 1902; Lebour, 1937; Fretter & Graham, 1977 Calliostoma zizyphinum Lebour, 1937; Fretter & Graham, 1977 SPAWNERS OF UNATTACHED EGG STRINGS (range of egg sizes: 150-225 wm) Calliostoma ligatum Holyoak, 1988a; Hadfield & Strathmann, 1990 Margarites pupillus Hadfield & Strathmann, 1990 SHELL BROODERS (range of egg sizes: no data) Clanculus bertheloti Thorson, 1967 Margarites vorticiferus Lindberg & Dobberteen, 1981 Munditia subquadrata Burn, 1976 Spectamen verum Powell, 1979 MANTLE CAVITY BROODERS (egg size: 300 um) Spectamen gerula Herbert, 1987 Spectamen multistriatum Herbert, 1987 dergo continuous gametogenesis, but male and female go- ruscans (Hedley, 1916) by Simpson (1977), who deter- nadal weights peak in February and again in September— mined from monthly sampling that the proportion of males October (JosKA & BRANCH, 1983). A similar pattern was and females with ripe gonads peaked in summer (Novem- noted in the subantarctic Cantharidus (Plumbelenchus) co- ber—January). Continuous maturation of gametes at a low Page 252 background level coupled with peak maturation periods, provides a potential bet-hedging mechanism in unpredict- able environments. COLMAN & TYLER (1988) found no significant differ- ences in the proportion of developmental stages in the ovaries of females of the deep-sea (>1000 m) eucycline species Calliotropis otto: (Philippi, 1844) collected in Feb- ruary, May, July, and August. This suggests the possi- bility of continuous production of gametes. Data are, how- ever, insufficient for strong inference, and it would certainly be premature to make any generalizations about the pres- ence or absence of reproductive periodicity in deep-sea trochaceans. Information on the neural and hormonal control and coordination of gametogenesis in trochaceans is not avail- able in either the primary literature or in reviews of en- dogenous control of gastropod reproduction (e.g., JOOSE, 1979), SPAWNING METHODS anp BEHAVIORS Although much of the information on trochacean spawning is anecdotal, there are enough detailed accounts to permit a preliminary classification (Table 2). The table lists rep- resentative species and literature citations in each of the categories discussed below. Broadcast Spawning Most trochaceans observed to date are broadcast spawn- ers, shedding their gametes into the water column where external fertilization occurs. In broadcasting species, ga- metes are released in pulses through the right neck lobe as thin streams or milky clouds of sperm and smaller bunches of individual or loosely adhering eggs (Hickman, personal observation; UNDERWOOD, 1972b). Theoretically, the chances of fertilization in broadcast spawning will be increased (1) if males and females are in close proximity when spawning occurs (aggregation) and (2) if males and females release gametes simulta- neously (synchronization). Both mechanisms appear to have evolved independently in different trochacean lineages, al- though neither mechanism is predominant within the su- perfamily. A few broadcasting trochaceans have been observed to aggregate during the spawning season or while engaging in epidemic spawning (DESAI, 1966; SIMPSON, 1977; LasI- AK, 1987; HADFIELD & STRATHMANN, 1990). However, actual pairing of males and females during spawning is more common in trochaceans that secrete benthic egg mass- es or unattached, but initially coherent, egg strings. In the closest form of pairing (contact pairing), the male (some- times a smaller individual) is stereotypically perched on the female shell to bring the regions of both of the apertures that lie immediately above the exhalant right neck lobes into proximity. Contact pairing has not been reported in broadcasting species. The Veliger, Vol. 35, No. 4 The timing and coordination of synchronous or epidemic spawning under natural conditions is not well understood. Spawning can be induced under laboratory conditions in some trochacean species by transferring animals to water that is warmer and still (STRATHMANN, 1987). Exogenous physical factors that have been observed to coincide with, but not necessarily trigger, spawning in the field include temperature increase (GERSCH, 1936; Ducros, 1957; WIL- LIAMS, 1965; UNDERWOOD, 1972b) and vigorous water movement during rough seas caused by strong onshore winds (GRANGE, 1976). A chemical trigger may be involved in the spawning of gametes in some species that engage in epidemic spawning. Little is known about the endogenous spawning rhythms, although daily, tidal, and lunar peri- odicities have been reported for some broadcasting species (e.g., BERRY, 1986). Although synchronization of spawning, like aggrega- tion, is an obvious mechanism for increasing the proba- bility of fertilization in broadcast spawners, there are a number of reports of asynchronous spawning in which gametes are released intermittently as they ripen (e.g., WILLIAMS, 1965; UNDERWOOD, 1972a; LASIAK, 1987). Theoretical considerations of the fates of gametes that are released into the water emphasize the range of physical effects of fluid turbulence and mixing (DENNY, 1988; DENNY & SHIBATA, 1989) and assume that gametes are neutrally buoyant and individually dispersed. Physical properties of some trochacean gametes deviate from these assumptions in ways that will alter their predicted fates. Recognition herein of a separate category for gametes that are shed as strings or aggregates directs attention to a spawning strategy that should be modeled separately. Spawning of Benthic Egg Masses Some trochaceans do not release their eggs into the water column but attach them to the substrate, usually to algae, seagrasses, or the undersides of rocks. In a few instances, individual eggs are surrounded by jelly coating and fixed individually to the substrate [see ElsAwy, 1970, for Tyro- chus dentatus (Forskal, 1875)], but aggregates or masses are more common. In the family Trochidae, benthic development within an attached jelly mass is common in members of the prim- itive subfamilies Eucyclinae and Margaritinae but it should not be interpreted as the primitive mode for the family. Benthic development is most common in high latitude taxa, and it is more probable that this spawning mode has been independently derived numerous times, because it also oc- curs in temperate, plant-associated trochine species of the tribe Cantharidinae as well as in temperate species of two of the most highly derived trochid subfamilies, Calliosto- matinae and Lirulariinae. The pattern is not necessarily consistent within genera, nor is it always correlated with latitude. There are reports of benthic egg masses for species of Gibbula Risso, 1826, from both high and low latitudes [e.g., GERSCH, 1936, for Cc. S. Hickman, 1992 G. tumida (Montagu, 1803) in the North Sea, and BANDEL, 1982, for G. drepanensis (Brugnone, 1873) in the Medi- terranean Sea], but the majority of species reported to date are broadcast spawners (FRETTER & GRAHAM, 1977). Eggs that are spawned in gelatinous strings may be deposited directly onto rocks, algae, or the blades of marine plants in patterns that preserve the form of the original string, such as the sinuous pattern in Cantharidus striatus (Linnaeus, 1758) (FRETTER & GRAHAM, 1977:fig. 49), or the spiral pattern in Gibbula drepanensis (Brugnone, 1873) (BANDEL, 1982:fig. 11). Alternatively, eggs that are spawned in a soft string of jelly may be molded into a compact mass that is fixed to the substrate by the foot as in Margarites marginatus Dall, 1919 [HoLyoAK, 1988b, as M. helicinus (Phipps, 1774); HADFIELD & STRATHMANN, 1990]. THORSON (1935) reported similar compact benthic egg masses for Margarites cinererius (Couthouy, 1839) (now commonly regarded as a synonym of Margarites costalis Gould, 1841) in northeast Greenland. Most compact mass- es are relatively shapeless, although GOHAR & EISAWY (1963:fig. I-2) illustrated an inverted cup-shaped mass, with a central cavity, for 7rochus erythraeus Brocchi, 1821. There are few direct observations of the formation of egg masses, but DUCH (1969) observed that the looped egg masses of Euchelus gemmatus (Gould, 1845) are fixed to the substrate as females crawl in a counter-clockwise di- rection. PICKEN (1979) observed overlapping egg masses attached to the undersides of rocks by Margarella antarctica (Lamy, 1905) at favorable sites in the Orkney Islands. The individual egg masses, which contained as many as 2000 eggs, were deposited in a doughnut shape (PICKEN, 1979: fig. 2A). Fewer descriptions of egg masses are available for the family Turbinidae. The turbinid 7urbo radiatus (Gmelin, 1791) is reported to produce a shapeless benthic egg mass (E1sawy & SORIAL, 1974). BANDEL (1982:fig. 12) re- ported that the tricoliine 77icolia speciosa (Muhlfeld, 1824) deposits eggs in simple rounded gelatinous bands on sea- grass blades, and MOoERS (1981) observed Tricolia pul- loides (Carpenter, 1865) producing sinuous jelly-coated ribbons that are consolidated into disc-shaped egg masses and attached to red algae. Spawning aggregations and contact pairing of individ- uals (see above) occur primarily in those trochaceans that spawn benthic egg masses. The overlapping egg masses of Margarella antarctica described by PICKEN (1979) provide indirect evidence of aggregations of individuals at favorable spawning sites. KOJIMA (1961) described animals of Can- tharidus jessoensis (Shrenk) as “‘copulating” in conjunction with the deposition of egg masses on rocks and Zostera blades during April and May in Japan. HADFIELD & STRATHMANN (1990) observed pairing of males and fe- males under laboratory conditions during the spawning of benthic egg masses by Lirularia succincta. Contact pairing of the small males of Tricolia variabilis (Pease, 1861) on the shells of the larger females (WERTZBERGER, 1968; Kay, Page 253 1979:fig. 17) occurs throughout the breeding season (Oc- tober to January) in Hawaii. Males remain on females during this time, and the pairing is not coincidental solely with the act of spawning (Hickman, personal observation). The stereotypic pairing of males and females described by DucH (1969) for Euchelus gemmatus (Gould, 1845) is in- teresting in the positioning of the smaller male near the left side of the aperture of the female. This is the position of the inhalant current. If sperm are released into the inhalant current, fertilization may occur in the mantle cavity rather than as the eggs are spawned from the right (exhalant) neck lobe. The most recent morphological classification of gastro- pod egg masses by SOLIMAN (1987) underestimates the variation in trochaceans by assigning them to two simple categories: shapeless, gelatinous masses, and cup-shaped masses). Development may proceed at different rates within a single trochacean egg mass, and DuCH (1969) reported more rapid development at the surface of the egg mass than near the center. Spawning of Unattached Egg Strings Strings or aggregates of eggs that are spawned from the right neck lobe in an encasing jelly are recognized here as a distinct category that is intermediate between the broad- cast spawning of individual eggs and the deposition of attached egg masses. Spawning of unattached strings or short chains of eggs is part of a unique trochacean variation (described in great- er detail below) on the pattern of mixed development (sensu PECHENIK, 1979). It is characterized by late hatching of swimming larvae and a brief pelagic phase following pre- dominantly benthic development. It has evolved indepen- dently in two trochid subfamilies (Margaritinae and Cal- liostomatinae) in which other species have totally benthic development and hatch following metamorphosis. In Calliostoma ligatum, unattached chains have been de- scribed by Hunt (1980), HOLYOAK (1988a), and HapD- FIELD & STRATHMANN (1990). Although the egg-binding jelly breaks down rapidly in moving water, egg strings spawned in still water remain together during fertilization and sink to the bottom and undergo early embryonic de- velopment as a unit. As noted above, pairing of males and females during spawning is not uncommon. HADFIELD & STRATHMANN (1990) noted that sperm had penetrated both the egg-binding jelly and the individual jelly coatings in freshly collected eggs from a female that was paired with a male during spawning. BANDEL (1982) reported the spawning of free-floating egg strings by Calliostoma laugiert Payraudeau, 1826, and the spawning of gelatinous floating aggregates of 2-10 eggs by C. granulatum (Born, 1778). In both species, Bandel noted that hatching occurs before metamorphosis and that the veliger is capable of swimming for short distances. Page 254 RAMON (1990) also reported C. granulatum spawning un- attached egg strings (“eggs were attached in long ribbons,” which “remained joined together in amorphous masses that came to rest on the bottom’’), although she did not observe any swimming between hatching and loss of the velum. Spawning of eggs in short chains also occurs in Margarites pupillus according to HADFIELD & STRATHMANN (1990). Although it is tempting to classify species that shed their eggs into the water without fixing them to the substrate as broadcast spawners, the cohesiveness of the eggs at the time of fertilization, their negative buoyancy and settle- ment to the substrate, and their adherence to one another after fertilization is fundamentally different from the sit- uation in broadcasting. Furthermore, the time of hatching has been shifted to very late in development. Calliostoma ligatum does not hatch until after torsion is completed (HOLYOAK, 1988a), whereas the typical trochacean with planktic development hatches as a trochophore or early, pre-torsional veliger. Spawning of Loosely Tacked Egg Strings An additional category forms a link between the spawn- ing of unattached strings of eggs and those that are firmly fixed to the substrate. In this category, a cylindrical, ge- latinous string of unattached eggs is tacked to the substrate at intervals by the front end of the foot. The long, partially floating string of eggs of Calliostoma zizyphinum (Linnaeus, 1758) illustrated by FRETTER & GRAHAM (1977:fig. 56) appears to be tacked to the substrate at only three points. In contrast to the Calliostoma species that spawn unat- tached strings and hatch as late veligers, C. zizyphinum hatches following metamorphosis (ROBERT, 1902; LEBOUR, 1936; FRETTER & GRAHAM, 1977). BANDEL (1982:fig. 13) illustrated the egg string of Cal- liostoma zizyphinum as circular in cross-section, and re- ported individual strings that reached a meter in length and contained 1500 eggs spawned at a rate of 10 to 35 per minute. Brooding Brooding is not common in trochaceans, but it occurs throughout the superfamily, in primitive as well as highly derived groups. It occurs principally in species from high latitudes. ROBERTSON (1985b) tabulated reports of actual or suspected brooding in primitive marine gastropods. There are documented instances of protection of developing em- bryos and larvae in three trochid subfamilies (Margari- tinae, Trochinae, and Solariellinae) and one turbinid sub- family (Liotiinae). LINDBERG & DOBBERTEEN (1981) described umbilical brooding in females of the primitive margaritine trochid Margarites vorticiferus (Dall, 1873). Brooding in this spe- cies is coupled with sexual dimorphism in the shell, with females having a broader umbilical cavity. DOBBERTEEN & ELLMORE (1986) further observed that secretion of the The Veliger, Vol. 35, No. 4 postlarval shell in this species begins while individuals are still confined to the brood space, which they cited as evi- dence of precocious appearance of adult traits. ‘THORSON (1967) reported brood protection within the spiral grooves on the bases of both male and female shells of Clanculus bertheloti d’Orbigny, 1839. An earlier illus- tration, from Thorson’s 1950 Christmas card, of larvae covered with a thin mucus layer, was reproduced by PURCHON (1968:fig. 101). Climo in POWELL (1979:64) reported umbilical brooding in a derived solarielline, Spec- tamen verum (Powell, 1937), and HAYWARD & GRACE (1981:fig. 3) subsequently illustrated a specimen with a brood. BURN (1976) illustrated an individual of the liotiine Munditia subquadrata (Tenison- Woods, 1878) with a mass of developing embryos in the umbilicus. HERBERT (1987) discovered the first instances of tro- chacean brooding of embryos and larvae within the mantle cavity in two South African solarielline species, Spectamen gerula Herbert, 1987, and S. multistriatum (Thiele, 1925). Ovoviviparity True ovoviviparity, retention of developing embryos and larvae within the oviduct, is unknown in trochaceans. The nature of the dimorphism in shell size and shape in two high-latitude trocoliine species, 77icolia gabiniana (Cotton & Godfrey, 1938) and 7. rosea (Angas, 1867), led ROBERTSON (1985b) to suggest the possibility of brooding within the mantle cavity or the oviduct. ARNAUD (1972) reported seeing “nombreux jeunes” through the translu- cent spire of specimens of Margarites refulgens (Smith, 1907) from the Antarctic. However, specimens provided by Arnaud to D. R. Lindberg do not have translucent spires, and it is uncertain what Arnaud observed. Ovoviviparity requires a reliable method of sperm trans- fer to the female. It is highly unlikely to occur in species that do not engage in contact pairing and that do not have a penis or analogous structure for sperm transfer. The fact that there is direct and stereotypic contact of individuals in some trochaceans, however, suggests that sperm transfer and internal fertilization could occur. The as yet poorly understood elaborations of the male right neck lobe re- ported above may represent an evolutionary step in the direction of internal fertilization. SPAWNING PERIODICITY Seasonal Periodicity In trochaceans with seasonal spawning, the periodicity should be related to an underlying seasonal periodicity in gametogenesis, discussed above. Few reports provide direct observations of trochacean spawning periodicity. Deter- mination of spawning time usually is based on indirect evidence, such as seasonal changes in the condition of the gonads, the time of appearance of egg masses, or recruit- ment of newly settled juveniles in an environment (¢.g., UNDERWOOD, 1974; SIMPSON, 1977; JOLL, 1980; MOOERS, GS) Hickman, 1992 1981; JoskA & BRANCH, 1983; LAsIAK, 1987). More in- directly, estimates of spawning time have been extrapolated from seasonal size-frequency data [e.g., WELLS & KEESING, 1987, for Cantharidus pulcherrimus (Wood, 1828)]. The ease with which spawning can be induced under laboratory conditions may provide a more precise indica- tion of when spawning is likely to occur under natural conditions. For example, HADFIELD & STRATHMANN (1990) were unable to induce spawning in Margarites pu- pillus collected in May, early June, and July, whereas they were successful with animals collected between 12 and 26 June in Washington state. Although sampling for the presence of eggs in the plank- ton provides a direct method of monitoring spawning ac- tivity in the field, this is probably not practicable unless populations are extremely dense and reproductive output extremely high. BERRY (1987) used daily plankton egg counts over a 79-day period of observation to estimate the reproductive output of individuals and the entire popu- lation of Umbonium vestiarium (Linnaeus, 1758) on a Ma- laysian sand flat. Unfortunately, single records of gonad condition or egg masses spawned in aquaria have been tabulated and pub- lished as reports of “breeding seasons” (e.g., BOVARD & OSTERUD, 1918; HEwaTT, 1938). Breeding-season data reported in faunal manuals must be interpreted with cau- tion. Diurnal, Tidal, and Lunar Periodicities There are very few data on finer-scale spawning peri- odicities. GRANGE (1976) found no relationship between spawning and time of day (night or day), state of the tides (position in the tidal cycle), or time of month (position in the lunar cycle) in New Zealand trochids and turbinids. Many marine invertebrate groups show a pattern of nighttime spawning on spring tides (KORRINGA, 1957; NAYLOR, 1976) that has been attributed in adaptational terms to nocturnal safety from predation and the greater dispersal potential of spring tides (e.g., BERRY, 1986). HEs- LINGA & HILLMAN (1981) reported a 28-day periodicity of spring tide spawning in 77ochus niloticus in Palau. In spite of the alleged advantage of releasing gametes on nocturnal spring tides, the best documented fine-scale spawning periodicity in a trochacean is the midday spawn- ing of Umbonium vestiarium (Linnaeus, 1758) on high or falling neap tides (BERRY, 1986). Umbonium is classified in a highly derived subfamily and tribe of trochids that are specialized for infaunal suspension feeding (HICKMAN, 1985; HICKMAN & MCLEAN, 1990). Umbonium vestiarium occurs at unusually high densities (12,000/m7?) on tropical intertidal sandflats, where it comprises >99% of the in- tertidal biomass (BERRY, 1986). It undergoes the most rapid development reported for any trochacean, settling and metamorphosing 36-48 hr after fertilization (BERRY, 1986). It is also an annual species with one discrete period of spawning and recruitment each year (BERRY & ZAMRI, Page 255 1983; BERRY, 1987). The adaptive advantages of a repro- ductive pattern that limits dispersal, promotes larval set- tlement in the environment of their origin, and favors settlement of briefly pelagic larvae at times of maximum submergence in quiet water is discussed by NAYLOR (1976) and BERRY (1986). DEVELOPMENTAL STAGES anp TIMING The length of time between fertilization and metamor- phosis (as a benthic juvenile capable of crawling, feeding, and retracting into its shell) of trochaceans varies from 2 or 3 days (HESLINGA, 1981; BERRY, 1986) to 28 days (HADFIELD & STRATHMANN, 1990). Developmental times can be altered under laboratory conditions and generally are longer at lower temperatures and higher latitudes. However, rate of development is not uniform throughout development. Prolonged developmental time results prin- cipally from delayed settling and/or delayed metamor- phosis. In the tropical Trochus niloticus, HESLINGA (1981) found that settlement could be delayed for 7 days and metamorphosis for 13 days in the absence of algal inducers. Delay of settlement is potentially significant to dispersal because of the additional time that larvae spend in the water column. The major stages and events in trochacean embryonic and larval development are summarized in Table 3, and the stages are illustrated in Figure 2. The major stages, for the purpose of this discussion, are: (1) fertilized egg, (2) two-cell cleavage through blastula, (3) gastrula, (4) trochophore, (5) pre-torsional veliger, (6) post-torsional veliger, (7) swimming-crawling demersal veliger, and (8) fully metamorphosed juvenile. These stages are arbitrary, but useful principally because the onset of each is marked by the appearance of a feature (e.g., development of the prototroch or the velum) that is relatively easily recognized, and because the developing animal has a characteristic overall appearance and organizational uniformity throughout each stage even though there may be many important developmental events occurring within each stage. The corresponding developmental events that mark the beginning of each of the above stages are: (1) fertilization, (2) cleavage, (3) gastrulation, (4) prototroch formation, (5) velum formation, (6) torsion, (7) settlement, and (8) meta- morphosis. The order of these main events is the same in all trochaceans, but their timing varies. There are many other events that occur within stages, and there is some variation in the order of appearance of features. The most important events occurring within developmental stages are listed in Table 3. In evolutionary terms, hatching is the most labile event in trochacean development, and its significance will be discussed in greater detail below. Variation in hatching time in six trochaceans with different developmental pat- terns is illustrated in Figure 3. What is important about hatching is not how long it takes, which is strongly cor- related with temperature both across and within species, Page 256 Table 3 Chronological sequence of trochacean developmental stages and events. Early embryonic development and larval de- velopment are subdivided into stages. The event charac- terizing the onset of each stage is listed first, followed by events that usually occur within the stage, although not necessarily in the order listed. The variable events of hatching and early settlement are inserted in parentheses at points where they can occur in the chronological se- quence. Stage Event EARLY EMBRYONIC DEVELOPMENT fertilization: penetration of jelly coat- ing(s) and egg membrane by sperm swelling of egg and associated coatings hardening of jelly coating polar body formation and extrusion Fertilized egg Two-cell cleavage, with increasingly asynchronous through division of macromeres and micro- blastula meres polar flattening of blastula Gastrula formation of blastopore and gastrulation, overgrowth of blastula by ectoderm appearance of locomotor cilia beginning of ciliary beating, with increas- ing coordination of beat beginning of rotation of gastrula within egg membrane differentiation of ectoderm into pretrochal, trochal, and posttrochal cells increasing constriction of blastopore LARVAL DEVELOPMENT prototroch formation by enlargement of trochal cells and elongation of trochal cilia appearance of shell gland appearance of foot rudiment closure of blastopore invagination of stomodaeum (HATCHING) appearance of first shell material ano-pedal flexure Pre-torsional velum formation, through enlargement veliger of prototroch (HATCHING) development of two distinct velar lobes appearance of opercular rudiment and operculum appearance of mantle fold and mantle cavity appearance of digestive gland Post-torsional torsion (first 90 degrees) veliger (HATCHING) completion of torsion (second 90 degrees) (EARLY SETTLEMENT) (HATCHING) asymmetric deformation of unmineralized larval shell mineralization of larval shell first attempts to retract into shell Trochophore The Veliger, Vol. 35, No. 4 Table 3 Continued. Stage Event reduction of velum and elongation of foot appearance of mouth appearance of radula sac (and radula?) appearance of digitate cephalic tentacles appearance of epipodial tentacles appearance of eyes Swimming- settlement or end of exclusively pelagic crawling period stage attempted crawling and retraction into shell; attempted balancing of shell metamorphosis with loss of velum and full development of ability to crawl, feed, and retract into shell (HATCHING) Juvenile snail but the fact that its position is not fixed within the devel- opmental sequence (Table 3 and Figure 3). In species that have exclusively benthic development (e.g., Margarites mar- ginatus), by-passing a planktic larval stage, hatching occurs after metamorphosis (defined here as loss of the velum and assumption of a crawling and feeding mode of life). Across species that have a lecithotrophic planktic stage, there is no fixed point at which larvae become free: hatching may occur as early as the trochophore stage (e.g., Umbonium vestiartum, Trochus niloticus, and many other species) or as late as the post-torsional veliger stage (e.g., Calliostoma ligatum). This is in marked contrast to planktotrophic caenogastropods, which hatch from capsules as well-de- veloped post-torsional veligers (FRETTER & GRAHAM, 1962). Settlement is also a variable event in trochacean devel- opment, occurring as early as the end of the first 90 degrees of torsion (e.g., Gibbula cineraria). At this stage of devel- opment, some other trochaceans have not yet hatched and begun their planktic stage. Completion of the second half of torsion after settlement may be primitive, because it is also reported in haliotids (CROFTs, 1937). Even trocha- ceans that settle immediately following the completion of torsion are leaving the water column at the stage of de- velopment when the typical caenogastropod veliger is hatching and beginning its planktic larval life. Settlement is not always easy to recognize as a discrete event. Larvae gradually may spend less time swimming and more time at or near the bottom during the later stages of torsion. In this swimming-crawling stage, discussed in greater detail below, the larva is no longer capable of efficient swimming and not yet capable of retracting, crawling, or balancing the shell over the head and foot (Figure 2M). Fertilization Sperm penetration of the egg membrane is not readily observed as a discrete event, and the time of fertilization C. S. Hickman, 1992 is usually inferred from visible changes in the egg and its associated structures (Figure 2A, B), or from the onset of cleavage. In species that pair while the female deposits a benthic egg mass or extrudes an unattached egg string, fertilization apparently occurs rapidly. HOLYOAK (1988b) observed ac- tive sperm both within the egg-binding jelly and within the jelly coats of individual eggs following their deposition on the substrate by Margarites marginatus (as M. helicinus). HADFIELD & STRATHMANN (1990) reported that unat- tached strings of eggs extruded by Calliostoma ligatum were fertilized at the time of examination, immediately follow- ing collection from individuals that had been paired as the eggs emerged from the mantle cavity of the female. Fertilization may be equally rapid in broadcasting tro- chaceans, especially those with synchronized spawning. HESLINGA (1981) interpreted the increase in egg diameter that occurred within a few minutes of spawning as indic- ative of fertilization in Trochus niloticus. UNDERWOOD (1972b) noted that a swelling confined to the outer layer of the jelly coat of eggs broadcast by Grzbbula cineraria was impervious to sperm penetration and was a pre-fertiliza- tion event. Fertilization occurred only after the outer layer had dispersed. Accordingly, UNDERWOOD (1972b) defined disappearance of the outer jelly coats as the event marking fertilization. Page 257 Cleavage and the Early Stages of Cell Differentiation The most detailed account of trochacean cleavage and the fate of cell lineages is that of ROBERT (1902) for Gibbula magus (as Trochus magus Linnaeus, 1758). Formation and extrusion of the first and second polar body follow closely on fertilization, and the first cleavage begins soon after appearance of the second polar body. Reports of the times of the first cleavage fall between 1 and 5.6 hr after inferred fertilization (DESAI, 1966; DucH, 1969; UNDERWOOD, 1972b; MooeRs, 1981; HOLyoak, 1988a, b). Few reports continue in detail beyond the first two meridional cleav- ages. Subsequent cleavage follows the classical spiralian pat- tern, and there is nothing unique to trochaceans, gastro- pods, or even mollusks, about the progression of alternating dexiotropic and laeotropic divisions leading to formation of the blastula. After the first two equal and meridional cleavages, spiral cleavage begins with the dexiotropic third cleavage, which produces an asymmetric 8-cell stage with large macromeres and small micromeres (Figure 2C). UNDERWOOD (1972b) is the only recent author to describe cleavage in any detail as far as the fifth cleavage. He noted asynchrony of division of the macromeres and micromeres at the fourth cleavage (producing a transient 12-cell stage Figure 2 Trochacean developmental stages. A. Egg prior to spawning: e, egg; em, egg membrane; igc, inner gelatinous coating; ogc, outer gelatinous coating. B. Unfertilized egg after swelling in seawater. C. Fertilized egg at the 8-cell stage, after the third cleavage (the first spiral, dexiotropic cleavage), with first quartet of micromeres shaded. D. Egg at the 16-cell stage, after the fourth cleavage (laeotropic), with the second quartet of micromeres shaded. E. Early gastrula, following epiboly, with wide open blastopore (b). F. Late gastrula, with blastopore nearly closed. G. Early trochophore larva showing relative cell sizes and shapes: pretr, pretrochal cells; tr, trochal cells; and posttr, posttrochal cells. H. Early trochophore larva: p, ciliated prototroch; shg, position of shell gland invagination; stom, stomodaeal invagination. I. Late trochophore with shell primordium (sp) secreted by shell gland and foot primordium (fp). J. Pre-torsional veliger from right side: v, velum; f, foot; mc, mantle cavity; mf, mantle fold; opp, opercular primordium; Irm, larval retractor muscle; Is, larval shell. K. Planktic veliger in ventral view after first 90 degrees of torsion: f, foot; op, operculum. L. Planktic veliger in left lateral view after first 90 degrees of torsion. M. Extended benthic veliger in left lateral view after first 90 degrees of torsion, with mantle cavity on the right side: m, mouth; ct, cephalic tentacle, e, eye. N. Partially metamorphosed and partially retracted demersal veliger in right lateral view at the completion of torsion: vr, velar remains. O. Fully metamorphosed, benthic juvenile in right lateral view. Figure 3 Comparative development of six trochacean gastropods. The events of fertilization, cleavage, gastrulation, prototroch formation, appearance of the trochophore stage, hatching, appearance of the veliger stage, torsion, settlement, and metamorphosis are depicted by black bands of width corresponding to the reports of time or range of time of observations reported in the literature. The onset of events is correlated by dashed lines. Heavier dashed lines are used to correlate hatching (H), settlement (S), and metamorphosis (M) to emphasize the variation in timing and order of these three events. Species are arranged to emphasize the shift in hatching time to progressively earlier stages of development: the hatching time line crosses the time line of velum formation between Monodonta lineata and Gibbula cineraria. Total development time is correlated with temperature and latitude and is less significant than the order of events. Sources of data are: Margarites marginatus (HOLYOAK, 1988b; HADFIELD & STRATHMANN, 1990); Calliostoma ligatum (HOLYOAK, 1988a, HADFIELD & STRATHMANN, 1990); Monodonta lineata (DESAI, 1966); Gibbula cineraria (UNDERWOOD, 1972b); Trochus niloticus (HESLINGA, 1981); Umbonium vestiarium (BERRY, 1986). Page 258 The Veliger, Vol. 35, No. 4 Ei Gey, (—s 1 year) life in the plankton. This scheme does not recognize the various ways in which benthic and pelagic forms of development have been mixed in trochacean development. The Possibility of Feeding in Pelagic Trochacean Larvae Feeding has not been observed in any trochacean larva to date, but relatively few species have been examined, and those that have been examined are concentrated in a few trochacean subfamilies. ROBERTSON (1985b) urged rejec- tion of the generalization that all archaeogastropods are lecithotrophic (e.g., STRATHMANN, 1978) or nonplankto- trophic (e.g., JABLONSKI & LuTz, 1983) until more data are available, and he noted that INO (1952) has reported feeding in the plankton by larvae of Haliotis discus Reeve, 1846. Even if feeding on particulate matter can be documented in the veligers of some trochaceans, morphological spe- cialization of the velum and a substantial increase in velar area would be prerequisite to significant nutrition. Spe- cifically, the absence of velar structures, such as the meta- troch (post-oral ciliary band) and ciliated food groove to carry particles to the mouth, argue against effective sus- pension feeding. However, particulate matter is not the only available Page 264 The Veliger, Vol. 35, No. 4 Figure 5 Trochacean protoconchs. A. Jlanga lirellata Herbert, 1987 (Trochidae: Solariellinae). Bar = 200 um. B. Calliotropis sp. (Trochidae: Eucyclinae). Bar = 150 um. C. Lodderena minima (Tenison Woods, 1878) (Turbinidae: Liotiinae). Bar = 100 um. D. Gabrielona pisinna Robertson, 1973 (Turbinidae: Gabrieloninae). Bar = 100 um. E. Leptothyra kermadecensis Marshall, 1979 (Turbinidae: Colloniinae). Bar = 150 um, arrow denotes discordant spiral thread. F. Leptothyra benthicola Marshall, 1979 (Turbinidae: Colloniinae). Bar = 150 um, arrow denotes discordant spiral thread. G. Herbertina eos Marshall, 1988 (Trochidae: Thysanodontinae). Bar = 100 um, arrows denote time of hatching and time of settlement. H. Calliostoma (Fautor) consobrina (Powell, 1958) (Trochidae: Calliostomatinae). Bar = 100 um. I. Crosseola concinna (Angas, 1867) (“Skeneidae”’). Bar = 100 um. J. Lirularia succincta (Carpenter, 1864) (Trochidae: Lirulariinae), apical view of the mechanically deformed hyperstrophic larval shell and orthos- C. S. Hickman, 1992 Page 265 source of nutrition to a pelagic larva. JAECKLE & MAN- AHAN (1989b) have reported uptake of dissolved organic matter (DOM) by pelagic larvae of Haliotis rufescens Swainson, 1822, and have shown that the costs of devel- opment in this species cannot be met by the energy supply present in the egg (JAECKLE & MANAHAN, 1989a). Uptake of DOM could play a significant role not only in paying developmental costs, but also in meeting metabolic needs if settlement and metamorphosis are delayed. The Possibility of Feeding in Benthic Trochacean Larvae When larvae are classified according to a planktotro- phic-lecithotrophic dichotomy (e.g., SHUTO, 1974; STRATHMANN, 1978), it is possible to overlook the potential for a nonfeeding planktic larva to begin feeding on the bottom between settlement and metamorphosis. There is no explicit documentation of feeding in this state by tro- chaceans, but demersal veligers that spend time near or on the bottom, alternately swimming feebly and attempting to crawl, before metamorphosis also have been observed making feeding movements (FRETTER, 1967; UNDERWOOD, 1972b). In larvae of Gzbbula cineraria in the ““swimming- attempted creeping” stage, UNDERWOOD (1972b) noted rapid development of the snout prior to shedding of the velum and stated that “the mouth was definitely open and searching movements were made.”’ Whereas there is an abrupt anatomical reorganization associated with the change in feeding mode at metamorphosis in planktotroph- ic species, adult feeding morphology is able to develop gradually and precociously in the larvae of nonfeeding archaeogastropods without impeding function of the velum (FRETTER, 1969). CROFTS (1937) noted development of the radular diverticulum in Haliotis prior to the completion of torsion. As in the case of unconventional feeding in the plankton (uptake of DOM), benthic feeding by a premetamorphic veliger could be substantial enough to increase its chances of survival (e.g., by prolonging larval life and increasing the probability of encountering the proper substrate for final settlement and/or metamorphosis). Poecilogony in Trochaceans HOAGLAND & ROBERTSON (1988) and BOUCHET (1989) reviewed the reports of developmental polymorphism in gastropods and did not find any alleged examples of the phenomenon in trochaceans. The form of poecilogony that has been of primary interest to malacologists and pale- ontologists is that of the restricted definition, in which both feeding (multispiral protoconch) and nonfeeding (pauci- spiral protoconch) larvae are reported either within the same population or within different populations of the same species. Both papers showed that the poecilogony has not been documented conclusively in any marine proso- branch, and that the presence of multispiral and paucispi- ral protoconchs on shells that are “virtually identical” is most likely an indication that cryptic species are involved. Developmental polymorphism need not be recorded in the protoconch, however, and the absence of protoconch differences cannot be used to argue against poecilogony in the broader sense of the term. There is no morphological criterion for distinguishing benthic development from pe- lagic development in the trochacean protoconch (HAD- FIELD & STRATHMANN, 1990) and insufficient evidence to support or refute the possibility of polymorphism of hatch- ing time that would give rise to both types of development within a single species. THE TROCHACEAN PROTOCONCH aAnpb ITS INTERPRETATION Protoconch Morphology Trochacean protoconch form and sculpture are illus- trated in Figure 5A-L. There are relatively few published data on the morphology of the trochacean protoconch, and HICKMAN & MCLEAN (1990) did not use protoconch char- acters in a systematic revision of the superfamily except to diagnose the subfamily Calliostomatinae, species of which share a distinctive, fine, hexagonal protoconch sculpture (Figures 4, 5H). This same pattern of sculpture occurs on the protoconch of members of the Thysanodontinae (Fig- ure 5G) (MARSHALL, 1988a), and the pattern of character state distributions suggests that thysanodontines are de- rived from calliostomatines. The trochacean protoconch shows little variation in form and sculpture relative to the caenogastropod protoconch. This generalization is based on my own survey of repre- sentative species of the trochacean subfamilies and tribes and examination of published scanning electron micro- graphs (RODRIGUEZ BABIO & THIRIOT-QUIEVREUX, 1974; MARSHALL, 1979, 1988a; BANDEL, 1975, 1982; DOBBER- TEEN & ELLMORE, 1986; HERBERT, 1987, 1989; COLMAN & TYLER, 1988; HADFIELD & STRATHMANN, 1990). Pro- trophic early teleoconch of newly hatched juvenile. Bar = 100 um. K. Margarites marginatus Dall, 1919 (Trochidae: Margaritinae), apical view of the mechanically deformed hyperstrophic larval shell and orthostrophic early teleoconch of newly hatched juvenile. Bar = 100 um. L. Margarites marginatus Dall, 1919 (Trochidae: Margaritinae), showing the manner in which the teleoconch of newly hatched juvenile is growing to cover most of the hyperstrophic (apex down) larval shell. Bar = 100 wm. Micrograph A provided by D. G. Herbert, micrographs E-H by B. Marshall, and micrographs J-L by M. G. Hadfield and M. F. Strathmann. Page 266 toconchs are approximately 1.25 whorls. Size is dependent upon egg size and is more closely correlated with water depth than with taxonomic group. The largest protoconchs occur in species with large yolk reserves. Sculpture ranges from smooth at one extreme to the highly ordered hex- agonal network of the calliostomatines, although the most common forms of sculpture are weak spiral threads and disordered or weakly ordered patterns of pitting and gran- ulation. The shape of the large apex is bluntly rounded, and it is partially enveloped by the adult shell, so that it is not possible to locate precisely the position or orientation of an axis of coiling (Figure 5A-L). BANDEL (1982) grouped sculpture of archaeogastropod protoconchs into 10 general types and also concluded that only one, the net-like pattern of Calliostoma, was diagnostic of higher taxa. Many protoconchs are completely smooth (Figure 5C, I) or nearly smooth (Figure 5B). More com- monly, however, the surface has fine relief that forms dis- tinctive patterns at higher magnifications. One of the com- monest patterns of relief is an irregularly anastomosing network of shell material that has a frothy appearance (Figure 5D, J-L). This surface sometimes looks as if it might be the result of secondary dissolution or etching of the protoconch surface (Figure 5D), but it clearly is a primary feature on the shells of larvae reared in the lab- oratory (Figure 5J-L). The most extensive set of protoconch illustrations for a single trochacean subfamily is that of HERBERT (1987), who illustrated 27 solarielline species in three genera. Al- though there is considerable variation in protoconch size (300-800 um diameter), the number of whorls (1.25) is relatively constant, as is the presence of 3-6 fine spiral threads that are not continuous with the more numerous spirals that begin abruptly on the juvenile shell. Such spirals are not, however, restricted to this subfamily (Hick- man, personal observation). Spiral sculpture ranges from a few discrete spiral elements (Figure 5A, F, K, L) to numerous spirals that may be indistinct, wavy, or discon- tinuous (Figure 5E). One of the most distinctive aspects of protoconch sculp- ture is the manner in which spirals appear to be discordant with coiling of the earliest-formed part of the protoconch (e.g., Figure 5E, F). The irregularity of the sculpture figures in arguments about the way the protoconch forms, and has significant bearing on the arguments discussed below about the relationship between the axis of coiling of the protoconch and that of the adult shell. Heterostrophy in Trochacean Shell Growth Heterostrophy is a condition in which the whorls of the protoconch appear to be coiled in the opposite direction from those of the adult shell. Heterostrophy is rare in prosobranchs and is considered by some malacologists to have arisen only once, and that at the caenogastropod grade of organization (e.g., ROBERTSON, 1985a). It is therefore The Veliger, Vol. 35, No. 4 of considerable interest that HADFIELD & STRATHMANN (1990) have observed and illustrated heterostrophic changes in coiling from dextral hyperstrophic protoconchs to dex- tral orthostrophic teleoconchs in species of the primitive trochid subfamily Margaritinae and in species of the de- rived trochid subfamily Lirulariinae. Illustrations of hy- perstrophic protoconchs from their paper are reproduced in Figure 5J-L. There are three important pieces of evidence in the argument for trochacean heterostrophy: (1) the trochacean larval shell is initially bilaterally symmetrical, (2) subse- quent coiling (or deformation) of the larval shell produces an asymmetric condition in which the apex (or at least the majority of the shell) is displaced to the left (Figure 5L), while (3) at metamorphosis the direction of coiling changes to displace the shell to the right. Prior to the work of HADFIELD & STRATHMANN (1990) there were a few unillustrated references to heterostrophy in trochaceans (e.g., DALL, 1889b; MINICHEV, 1971). BANDEL (1975) referred to a “left handed evolute start- off” in the protoconchs of gibbuline and cantharidine tro- chids. I shall argue below that he was describing a phe- nomenon that produces the appearance of heterostrophy in the early portion of the protoconch only. However, it does not arise from a change in coiling direction during accretionary growth and does not seem to provide a com- plete explanation of the coiling pattern illustrated by HaD- FIELD & STRATHMANN (1990). There are two difficulties in evaluating the nature and extent of heterostrophy in trochaceans. In the first place, the bulbous trochacean protoconch is the extension of a large, cup-shaped embryonic shell in which it is difficult to identify an apex or an axis of coiling. In the second place, most of the trochacean protoconch is enveloped in postlarval growth by the adult shell. Only a small portion of the protoconch is visible once the postmetamorphic ju- venile has begun to secrete shell material. It was only by examining larvae and early juveniles (Figure 5L) that HADFIELD & STRATHMANN (1990) were able to detect heterostrophy. The discovery of heterostrophy in trochaceans is con- troversial because it has been considered an evolutionary innovation at a higher grade of prosobranch evolution (e.g., ROBERTSON, 1985a; HASZPRUNAR, 1985). However, it is one thing to show that heterostrophy is present in trocha- ceans and another to show that all heterostrophic shells are homologous. Changes in the direction of coiling be- tween larval and postmetamorphic development may be much more common than heretofore believed, and they may be independently derived in several lineages. Evidence provided by HADFIELD & STRATHMANN (1990) is not suf- ficient to resolve whether trochacean heterostrophy is a special case of heterostrophy. Heterostrophy is a general phenomenon that is recognized by its geometric expression (HIcKMAN, 1980). It may have more than one develop- mental explanation. The question of homology will be CaS ickman 1992 resolved only by studies of the developmental process. I suspect that trochacean heterostrophy is related less to a leftward displacement of the larval shell through a change in the direction of accretionary growth than it is to a leftward displacement by asymmetric mechanical defor- mation of the larval shell prior to its mineralization. Mechanical Deformation of the Trochacean Protoconch During Larval Development An intriguing, but overlooked, observation by BANDEL (1975, 1982) is the change in apparent direction of coiling that takes place within larval development. Although BANDEL (1975) initially described the pointed tip of the protoconch as sinistral in its coiling (“linksgewundenen’’), he subsequently (1982) interpreted this coiling form as a result of deformation of a poorly mineralized shell rather than as a change in coiling direction during accretionary growth. Furthermore, it is only the tip that is sinistral in appearance, and BANDEL’s (1991) interpretation of the larval shell as a whole is that it is mechanically “pulled into the dextral coil.” The most powerful evidence that the initial coil of the protoconch is formed by mechanical deformation rather than accretionary growth is that the discordant spiral sculpture emerges from what appears to be the side of the initial coil rather than from what appears to be the apex (see arrows on Figure 5E, F). In BANDEL’s (1982) inter- pretation, the purely organic, cup-shaped larval shell is bilaterally symmetrical until mechanical deformation oc- curs. Asymmetric deformation forces it into a trochispiral shape, with the apex on the right and the umbilicus on the left. If, however, asymmetric deformation displaced the apex to the left, it could provide a whole or partial expla- nation of the observations of HADFIELD & STRATHMANN (1990). Mechanical deformation has consequences for ecological interpretation and phylogenetic analysis. The amount of mechanical deformation is less in large yolky eggs (BANDEL, 1982), and is correlated with depth. The bulbous, relatively undeformed apex illustrated in Figure 5B is from 250 m. This feature is potentially useful in paleoecological recon- struction. Mechanical deformation is a shared feature of archaeogastropods (sensu HICKMAN, 1988), but its distri- bution among other primitive prosobranch taxa is yet to be determined. It does not occur in neritomorphs. A major unresolved question is whether mechanical de- formation can provide a complete explanation of heter- ostrophy or whether subsequent growth of the larval shell is displaced to the right, producing additional leftward displacement of the spire. Resolution of this question re- quires detailed developmental study of live veligers, with careful observation of the timing and sequence of events that include (1) formation, insertion, and contraction of the larval retractor muscle (or muscles), (2) mechanical deformation of the premineralized shell, (3) the fate of the Page 267 larval retractor muscle (or muscles) before and during metamorphosis, (4) the origin and assumption of function of the columellar muscle, (5) mineralization of the larval shell, (6) and the onset of accretionary growth. Evidence of a pattern of deformation is good. A plausible deformational mechanism is difficult to postulate from ei- ther of the conflicting interpretations of larval musculature and of the attachment points of the animal to the shell. CrortTs (1955) reported a single larval retractor muscle inserted on the pre-torsional right (post-torsional left) side of the shell. BANDEL (1982) states that there is a pair of larval retractors and that the (post-torsional) left is re- sorbed after shell deformation, while the (post-torsional) right becomes the precursor of the columellar muscle. Jan- ice Voltzow (personal communication) notes that the larva is attached at two points, but she observes a difference in the nature of the attachment. The attachment on the post- torsional left side is clearly muscular (the single larval retractor of CROFTS [1955] and the (post-torsional) left larval retractor of BANDEL [1982]), but the attachment on the post-torsional right side appears to be connective tissue rather than muscle. In this interpretation, the single left larval retractor is resorbed at metamorphosis and the col- umellar muscle arises from the post-torsional right me- sodermal band. Ecologic, Biogeographic, and Macroevolutionary Inferences There is considerable debate over the kinds of inter- pretations that can be made from the protoconchs of marine prosobranchs. Long before the advent of scanning electron microscopy, it had been suggested that the protoconch was useful in taxonomic distinctions (e.g., POWELL, 1942) and also could be used to infer developmental modes (e.g., ‘THORSON, 1950). The generalization that species with veligers that feed in the plankton have protoconchs with (1) a small Pro- toconch I (reflecting small egg size), (2) one or more whorls of Protoconch II (= multispiral), and (3) a morphological distinction between Protoconch I and Protoconch II, is not strongly disputed. Likewise, the generalization that species with nonfeeding larvae have protoconchs with (1) a large initial whorl (reflecting large egg size), (2) few post-nu- clear whorls, and (3) lack of a prominent morphological discontinuity has not been questioned. The interpretations that have been made from these generalizations have fared less well. The three major cat- egories of interpretation have involved (1) the geographic ranges of taxa (as a reflection of dispersal potential), (2) the geologic longevity of taxa, and (3) the rates of speciation and extinction of taxa (see JABLONSKI & Lutz, 1983, and references therein). Some predictions, such as the ability of larvae with multispiral protoconchs to achieve broad geographic ranges in a single generation (SCHELTEMA, 1977, 1978, 1979) have strong empirical support. Other Page 268 The Veliger, Vol. 35, No. 4 predictions, such as species with multispiral protoconchs having broader geographic ranges and longer stratigraphic ranges, are supported less strongly by the correlative ob- servations originally provided by POWELL (1942) and SHUTO (1974), and the subsequent, more refined correl- ative analyses of HANSEN (1978, 1980) and JABLONSKI (1986). Anomalies in JABLONSKI’s (1986) Cretaceous data set (nonplanktotrophic species with geographic and strati- graphic ranges exceeding those of planktotrophs) have been noted by HADFIELD & STRATHMANN (1990). There is certainly no theoretical reason why a species that totally lacks planktic development cannot achieve a broad distri- butional range exclusively through the cumulative effects of adult movements over long periods of time. Any prediction that trochaceans (as taxa with paucispi- ral protoconchs) will have restricted geographic and strati- graphic ranges and high rates of speciation and extinction is suspect in terms of the developmental variability dis- cussed herein. By focusing on the lack of larval feeding, it lumps taxa with attached benthic development, demersal benthic development, nonfeeding pelagic development, and several forms of mixed pelagic and non-pelagic develop- ment. The lumping of these categories involves unwar- ranted assumptions that seriously underestimate (1) pas- sive dispersal of floating, unattached eggs and embryos, (2) the length of time that nonfeeding larvae can spend in the water column, (3) the extent to which nonfeeding larvae can delay settlement and/or metamorphosis, and (4) the access that nonfeeding pelagic larvae and demersal larvae may have to fast-moving currents (HADFIELD & STRATH- MANN, 1990). CONCLUSIONS A recurring theme of this paper is the diversity and vari- ability in trochacean reproduction and development that emerges only when comparisons are made in a systematic framework, and only when the traditional expectations of primitive, canalized simplicity are suspended. Most of the published data have been generated outside of a compar- ative or systematic framework as studies of single species, with a bias toward common intertidal species in the vicinity of marine laboratories. There are major gaps in comparative systematic data for the families Skeneidae (no data) and Turbinidae (data for only two of the nine subfamilies). In the Trochidae, data are concentrated in taxa in the three tribes of the subfamily Trochinae and the subfamily Calliostomatinae. Although data have been published for species representing four additional trochid higher taxa, there are 12 remaining higher trochid taxa for which there are no comparative data. Much of the evolutionary novelty in trochacean evolu- tion is based on modifications of the epipodium, and it would be appropriate to look more carefully for dimor- phism in the morphology and behavior of anterior epi- podial structures (especially the right neck lobe). Although previous reports of a trochacean penis have failed to doc- ument the implied function of insertion and internal fer- tilization, the possibility that such structures enhance sperm transfer bears closer scrutiny. Broadcast spawning and the spawning of benthic egg masses are well-established modes of egg release in tro- chaceans, leading to planktic larval development on one hand and benthic larval development on the other. We know less about taxa that brood: the variety of sites where brooding occurs and the number of times brooding has arisen in trochacean evolutionary history. Least well un- derstood are the taxa that spawn discrete strings of eggs: the significance of the alternative patterns of spawning unattached strings that disaggregate in the water column soon after spawning, unattached strings that remain intact and sink to the bottom, and strings that are loosely tacked to the substrate. Do eggs benefit from remaining together in the water column, and for how long? Can sinking (after fertilization but before hatching) provide an additional benefit? Is the loose attachment of egg strings an evolu- tionary step in the direction of benthic development and loss of a planktic stage, or is there an advantage to late hatching and a brief sojourn in the plankton? Mixed ben- thic and planktic development has been modeled as a dis- crete adaptive strategy (PECHENIK, 1979; STRATHMANN, 1985). However, the different ways trochaceans combine larval development in the water column and on the bottom confound the notion of a single mixed strategy, and their ultimate evolutionary significance may lie principally in the range of developmental variability they encompass. Modeling of the mechanics of fluid movements on dif- ferent spatial scales and of the behavior of particles as a function of their properties is a relatively new field (DENNY, 1988), and experimental studies of the fluid mechanics of broadcast spawning (THOMAS, 1992) show that gamete properties can have important functional consequences. Trochacean gametes and larvae comprise a variable system in which experimental tests might be especially fruitful. The questions of heterostrophy and mechanical defor- mation of the trochacean protoconch are especially im- portant to paleontologists because the protoconch is one of the few keys to inferring development in extinct taxa. Interpretation of these phenomena requires additional di- rect observation of events in larval development. Trocha- cean protoconchs are not reliable as sources of information about where development takes place (in the water column or on the bottom) or the length of time between fertilization and metamorphosis. They can, however, be read more carefully for information about events that have occurred during larval development. In summary, it is the unique features of trochacean reproduction and development that seem to provide the most productive avenues for future research. Problems in- viting special attention include: (1) the composition, prop- erties, and function of the primary and secondary egg jellies that are characteristic of trochacean eggs; (2) the compar- C. S. Hickman, 1992 ative structure and functional contributions of female pal- lial glands and elaborations (urogenital papillae) in tro- chacean spawning and reproduction; (3) the effects of different spawning modes on fertilization success and lar- val dispersal; (4) the possibility that trochacean epipodial structures function in sperm transfer and that internal fertilization may occur; (5) the possibility of unconven- tional nutrition (uptake of dissolved organic matter) in planktic trochacean larvae; (6) the possibility of feeding in benthic trochacean larvae between settlement and meta- morphosis; (7) the nature, extent, and evolutionary sig- nificance of the apparent heterostrophy in many trocha- cean protoconchs; and (8) the nature and significance of the widespread apparent mechanical deformation of the tip of the trochacean protoconch during larval develop- ment. ACKNOWLEDGMENTS My perception of the issues in molluscan reproduction and development has grown and changed through lively dis- cussions with D. R. Lindberg, M. G. Hadfield, M. F. Strathmann, W. F. Ponder, R. Robertson, and J. Voltzow. I value both their reinforcement and their dissent. I am especially grateful to D. G. Herbert, B. Marshall, M. G. Hadfield, and M. F. Strathmann for the use of their scan- ning electron micrographs of trochacean protoconchs. In my own survey of trochacean protoconchs, I acknowledge the assistance of G. Avern of the Scanning Electron Mi- croscope Laboratory of the Australian Museum, Sydney, and D. Pardoe of the Electron Microscope Laboratory of the University of California, Berkeley. An anonymous re- viewer provided an heroic editorial critique and upbraiding for which I am especially grateful. I thank R. Bieler, M. G. Hadfield, and D. R. Lindberg for helpful substantive reviews of the manuscript. The Museum of Paleontology has contributed to the payment of page charges. LITERATURE CITED AMIo, N. 1963. A comparative embryology of marine gastro- pods, with ecological considerations. Journal of the Shimo- noseki University of Fisheries 12:229-358 [text in Japanese, tables and figure legends in English]. ARNAUD, P. M. 1972. Invertébrés marins des XII** et XVér expéditions Antarctiques Frangaises en Terre Adélie 8.— Gastéropodes prosobranches. Tethys, Supplement 4:105-134. AZEVEDO, C., A. LoBO-DA-CUHNA & E. OLIVERA. 1985. Ul- trastructure of the spermatozoon in Gibbula umbilicalis (Gas- tropoda, Prosobranchia), with special reference to acrosomal formation. Journal of Submicroscopic Cytology 17:609d- 614. BANDEL, K. 1975. Das Embryonalgehduse mariner Proso- branchier der Region von Banyuls-sur-Mer. Vie et Milieu 25A:83-118. BANDEL, K. 1982. Morphologie und Bildung der frihonto- genetischen Gehause bei conchiferen Mollusken. Facies (Er- langen) 7:1-198. BANDEL, K. 1991. Ontogenetic changes reflected in the mor- phology of the molluscan shell. Pp. 211-230. In: N. Schmidt- Page 269 Kittler & K. Vogel (eds.), Constructional Morphology and Evolution. Springer-Verlag: Berlin. BERRY, A. J. 1986. Daily, tidal, and two-weekly spawning periodicity and brief pelagic dispersal in the tropical inter- tidal gastropod Umbonium vestiarium (L.). Journal of Ex- perimental Marine Biology and Ecology 95:211-223. Berry, A. J. 1987. Reproductive cycles, egg production and recruitment in the Indo-Pacific intertidal gastropod Umboni- um vestiarium (L.). Estuarine, Coastal and Shelf Science 24: 711-723. Berry, A. J. & ZAMRI BIN OTHMAN. 1983. An annual cycle of recruitment, growth and production in a Malaysian pop- ulation of the trochacean gastropod Umbonium vestiarium (L.). Estuarine, Coastal and Shelf Science 17:357-363. Bonar, D. B. 1978. Morphogenesis and metamorphosis in opisthobranch mollusks. Pp. 177-196. In: F. S. Chia & M. E. Rice (eds.), Settlement and Metamorphosis of Marine Invertebrate Larvae. Elsevier /North-Holland Biomedical Press: New York. BoUcHET, P. 1989. A review of poecilogony in gastropods. Journal of Molluscan Studies 55:67-78. BovarD, J. F. & H. L. OsTERuD. 1918. Partial list of the animals yielding embryological materials at the Puget Sound Biological Station. Publications of the Puget Sound Biolog- ical Station 2:127-137. BurRN, R. 1976. Shell with a built-in nest. Australian Shell News 16:3. Cura, F. S. 1974. Classification and adaptive significance of developmental patterns in marine invertebrates. Thalassia Jugoslavica 10:121-130. CLARK, W. 1852. On the Skeneadae [sic]. The Annals and Magazine of Natural History, Ser. 2, 8:44-49. CoLmaN, J. G. & P. A. TYLER. 1988. Observations on the reproductive biology of the deep-sea trochid Calliotropis ottor (Philippi). Journal of Molluscan Studies 54:239-242. Crisp, D. J. 1976. Energy relations of marine invertebrate larvae. Thalassia Jugoslavica 10:102-120. Crofts, D. R. 1937. The development of Haliotis tuberculata, with special reference to organogenesis during torsion. Philo- sophical Transactions of the Royal Society B 208:219-268. Crofts, D. R. 1955. Muscle morphogenesis in primitive gas- tropods and its relation to torsion. Proceedings of the Zoo- logical Society of London 125:711-750. DALL, W. H. 1889a. Notes on the soft parts of Trochus infun- dibulum Watson, with an account of a remarkable sexual modification of the epipodium, hitherto undescribed in Mol- lusca. Nautilus 3(1):2-4. DALL, W. H. 1889b. 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,” Lieut. Commander C. D. Sigsbee, U.S.N., commanding. XXIX. Report of the Mol- lusca. Part 2, Gastropoda and Scaphopoda. Bulletin of the Museum of Comparative Zoology, Harvard University 18: 1-492, pls. 10-40. Denny, M. W. 1988. Biology and the Mechanics of the Wave- Swept Environment. Princeton University Press: Princeton, New Jersey. 329 pp. Denny, M. W. & M.F. Suipata. 1989. Consequences of surf- zone turbulence for settlement and external fertilization. American Naturalist 134:859-889. Desal, N. 1966. The biology of Monodonta lineata (Da Costa). Proceedings of the Malacological Society of London 37:1- Wie DOBBERTEEN, R. A. & G. S. ELLMORE. 1986. Embryonic ex- Page 270 pression of shell dimorphism in Margarites vorticifera [sic] (Gastropoda: Trochidae). Malacological Review 19:45-52. Ducu, T. M. 1969. Spawning and development in the trochid gastropod Euchelus gemmatus (Gould, 1841) in the Hawaiian Islands. The Veliger 11(4):415-417. Ducros, C. 1957. Induction de l’evacuation des produits sexuel chez gibbules. Comptes Rendu Hebdomadaires des Seances de l’Académie des Sciences de Paris 151:285. E1sawy, A. M. 1970. Spawning, development and metamor- phosis of Trochus dentatus Forskal. Bulletin of the Institute of Oceanography and Fisheries, Egypt 1:379-393. Ersawy, A. M. & A. E. SoriaL. 1974. The egg masses and development of 7urbo radiatus Gmelin. Bulletin of the In- stitute of Oceanography and Fisheries, Egypt 4:221-235. FRANZEN, A. 1956. On spermiogenesis, morphology of the sper- matozoon and biology of fertilization among invertebrates. Zoologiska Bidrag fran Uppsala 31:356-482. FRETTER, V. 1946. The genital ducts of Theodoxus, Lamellaria and 77ivia, and a discussion on their evolution in the proso- branchs. Journal of the Marine Biological Association of the United Kingdom 26:312-351. FRETTER, V. 1955. Some observations on Tricolia pullus (L.) and Margarites helicinus (Fabricius). Proceedings of the Mal- acological Society of London 31:159-162. FRETTER, V. 1967. The prosobranch veliger. Proceedings of the Malacological Society of London 37(6):357-366. FRETTER, V. 1969. Aspects of metamorphosis in prosobranch gastropods. Proceedings of the Malacological Society of Lon- don 38:375-386. FRETTER, V. 1984. Prosobranchs. Pp. 1-45. In: A. S. Tompa, N. H. Verdonk & J. A. M. van den Biggelaar (eds.), The Mollusca. Vol. 7. Reproduction. Academic Press, Inc.: Or- lando, Florida. FRETTER, V. & A. GRAHAM. 1962. British Prosobranch Mol- luscs. Ray Society: London. 755 pp. FRETTER, V. & A. GRAHAM. 1977. The prosobranch molluscs of Britain and Denmark. Part 2—Trochacea. Journal of Molluscan Studies, Supplement 3:39-100. GeErRSCH, M. 1936. Die Genitalapparat und die Sexualbiologie der Nordseetrochiden. Zeitschrift fiir Morphologie und Oko- logie der Tiere 31:106-150. GiEsE, A. C. 1959. Annual reproductive cycles of marine in- vertebrates. Annual Review of Physiology 21:547-576. GiEsE, A. C. & J.S. PEARSE. 1974. Introduction: general prin- ciples. Pp. 1-49. In: A. C. Giese & J. S. Pearse (eds.), Reproduction of Marine Invertebrates. Vol. 1. Academic Press: New York. Gouar, H. A. F. & A. M. Etsawy. 1963. The egg masses and development of 77ochus erythraeus Brocchi. Publications of the Marine Biological Station, Al-Ghardagqa, Red Sea 14: 109-148. GRANGE, K. R. 1976. Rough water as a spawning stimulus in some trochid and turbinid gastropods. New Zealand Journal of Marine and Freshwater Research 10:203-216. HADFIELD, M. G. 1978. Metamorphosis in marine molluscan larvae: an analysis of stimulus and response. Pp. 165-175. In: F. S. Chia & M. E. Rice (eds.), Settlement and Meta- morphosis of Marine Invertebrate Larvae. Elsevier /North- Holland Biomedical Press: New York. HADFIELD, M. G. & M. F. STRATHMANN. 1990. Heterostro- phic shells and pelagic development in trochoideans: impli- cations for classification, phylogeny and palaeoecology. Jour- nal of Molluscan Studies 56:239-256. HANSEN, T. A. 1978. Larval dispersal and species longevity in Lower Tertiary gastropods. Science 199:885-887. The Veliger, Vol. 35, No. 4 HANSEN, T. A. 1980. Influence of larval dispersal and geo- graphic distribution on species longevity in neogastropods. Paleobiology 6:193-207. HASZPRUNAR, G. 1985. The Heterobranchia—a new concept of the phylogeny of the higher Gastropoda. Zeitschrift fur Zoologische Systematik und Evolutionsforschung 23:15-37. HayYwarD, B. W. & R. V. GRacE. 1981. Soft bottom macro- fauna and foraminiferal microfauna off Cuvier Island, north- east New Zealand. Tane 27:43-54. HEALy, J. M. 1989. Ultrastructure of spermiogenesis in the gastropod Calliotropis glyptus Watson (Prosobranchia: Tro- chidae) with special reference to the embedded acrosome. Gamete Research 24:9-19. HEALY, J. M. 1990. Euspermatozoa and paraspermatozoa in the trochoid gastropod Zalipais laseroni (Trochoidea: Ske- neidae). Marine Biology 105:497-507. HERBERT, D.G. 1987. Revision of the Solariellinae (Mollusca: Prosobranchia: Trochidae) in southern Africa. Annals of the Natal Museum 82(2):283-382. HERBERT, D.G. 1989. Pagodatrochus, a new genus for Minolia variabilis H. Adams, 1873 (Gastropoda: Trochidae). Journal of Molluscan Studies 55:365-372. HESLINGA, G. A. 1981. Larval development, settlement and metamorphosis of the tropical gastropod T7ochus niloticus. Malacologia 20(2):349-357. HESLINGA, G. A. & A. HILLMAN. 1981. Hatchery culture of the commercial top snail 77ochus niloticus in Palau, Caroline Islands. Aquaculture 22:35-43. HEwatTT, W. G. 1938. Notes on the breeding seasons of the rocky beach fauna of Monterey Bay, California. Proceedings of the California Academy of Sciences (4th Series) 53(19): 283-288. HICKMAN, C. S. 1980. A remarkable case of coaxial heteros- trophy in an Eocene gastropod. Journal of Paleontology 54: 196-199. HICKMAN, C. S. 1985. Comparative morphology and ecology of free-living suspension-feeding gastropods from Hong Kong. Pp. 217-234. In: B. Morton & D. Dudgeon (eds.), The Malacofauna of Hong Kong and of Southern China. Hong Kong University Press: Hong Kong. 681 pp. HICKMAN, C. S. 1988. Archaeogastropod evolution, phylogeny and systematics: a re-evaluation. Pp. 17-34. In: W. F. Ponder (ed.), Prosobranch Phylogeny. Proceedings of the 9th Inter- national Malacological Congress, Edinburgh, 1986. Mala- cological Review, Supplement. Hickman, C. S. & J. H. McLEan.; 1990. Systematic revision and suprageneric classification of trochacean gastropods. Natural History Museum of Los Angeles County Science Series No. 35. VI + 169 pp. HOAGLAND, K. E. & R. ROBERTSON. 1988. An assessment of poecilogony in marine invertebrates: phenomenon or fan- tasy? Biological Bulletin 174:109-125. HOIsATER, T. 1968. Taxonomic notes on the north-European species of ““Cyclostrema” sensu Jeffreys 1883 (Prosobranchia, Diotocardia). Sarsia 33:43-58. Hotyoak, A. R. 1988a. Spawning and larval development of the trochid gastropod Calliostoma ligatum (Gould, 1849). The Veliger 30(4):369-371. Ho yoak, A. R. 1988b. Spawning, egg mass formation, and larval development of the trochid gastropod Margarites hel- icinus (Fabricius). The Veliger 31(1/2):111-113. Hunt, D. E. 1980. Observations on spawning in Calliostoma ligatum (Gould, 1849). The Veliger 22:292. INo, T. 1952. Biological studies on the propagation of Japanese abalone (genus Haliotis). Bulletin of the Tokai Regional C. S. Hickman, 1992 Fisheries Research Laboratory 5. 102 pp. [In Japanese; see ROBERTSON (1985) for reference to a mimeographed trans- lation of this work.] JABLONSKI, D. 1986. Larval ecology and macroevolution in marine invertebrates. Bulletin of Marine Science 39:565- 587. JABLONSKI, D. & R. A. Lutz. 1980. Larval shell morphology: ecological and paleontological applications. Pp. 323-377. In: D. C. Rhoads & R. A. Lutz (eds.), Skeletal Growth of Aquatic Organisms. Plenum Press: New York. JABLONSKI, D. & R. A. Lutz. 1983. Larval ecology of marine benthic invertebrates: paleobiological implications. Biologi- cal Reviews 58:21-89. JAECKLE, W.B. & D. T. MANAHAN. 1989a. Growth and energy imbalance during larval development of a lecithotrophic mol- luscan larva (Haliotis rufescens). Biological Bulletin 177:237- 246. JAECKLE, W. B. & D. T. MANAHAN. 1989b. Feeding by a “nonfeeding” larva: uptake of dissolved amino acids from seawater by lecithotrophic larvae of the gastropod Haluotis rufescens Swainson. Marine Biology 103:87-94. JEFFREYS, G. 1865. British Conchology. Vol. 3. John van Voorst: London. 393 pp. Joti, L. M. 1980. Reproductive biology of two species of Turbinidae (Mollusca: Gastropoda). Australian Journal of Marine and Freshwater Research 31:319-336. Joose, J. 1979. Endocrinology of molluscs. Pp. 107-137. In: S. van der Spoel, A. C. van Bruggen & J. Lever (eds.), The Molluscan Framework. Bohn, Scheltema & Holkema BV: Utrecht, The Netherlands. Joska, M. A. P. & G. M. BRANCH. 1983. The reproductive cycle of the trochid gastropod Oxystele variegata (Anton, 1839). The Veliger 26:47-51. Kay, E. A. 1979. Hawaiian Marine Shells. Bishop Museum Press: Honolulu, Hawaii. 653 pp. KoOuNERT, R. & V. STorcH. 1983. Ultrastrukturelle Unter- suchungen zur Morphologie und Genese der Spermien von Archaeogastropoda. Helgolander Meeresuntersuchungen 36: 77-84. KoIkE, K. 1985. Comparative ultrastructural studies on the spermatozoa of the Prosobranchia (Mollusca: Gastropoda). Science Report of the Faculty of Education, Gumma Uni- versity 34:33-153. Kojima, Y. 1961. On the breeding of Cantharidus callichroa jessoensis (Shrenk). Venus, The Japanese Journal of Mal- acology 21(4):511-515. KorRINGA, P. 1957. Lunar periodicity. Pp. 347-381. In: J. W. Hedgpeth (ed.), Treatise on Marine Ecology and Paleo- ecology. Vol. 1. Ecology. Geological Society of America Memoir 67. Lasik, T. 1987. The reproductive cycles of three trochid gas- tropods from the Transkei Coast, southern Africa. Journal of Molluscan Studies 53:24-32. LEBour, M. V. 1936. Notes on eggs and larvae of some Plym- outh prosobranchs. Journal of the Marine Biological As- sociation of the United Kingdom 20:547-565. LeBour, M.V. 1937. The eggs and larvae of the British proso- branchs with special reference to those living in the plankton. Journal of the Marine Biological Association of the United Kingdom 22:105-166. Lewis, J. B. 1960. The fauna of rocky shores of Barbados, West Indies. Canadian Journal of Zoology 38:391-435. LINDBERG, D.R. & R.A. DOBBERTEEN. 1981. Umbilical brood protection and sexual dimorphism in the boreal Pacific tro- Page 271 chid gastropod, Margarites vorticiferus Dall. International Journal of Invertebrate Reproduction 3:347-355. MARSHALL, B. A. 1979. The Trochidae and Turbinidae of the Kermadec Ridge (Mollusca: Gastropoda). New Zealand Journal of Zoology 6:521-552. MARSHALL, B. A. 1988a. Thysanodontinae: a new subfamily of the Trochidae (Gastropoda). Journal of Molluscan Stud- ies 54:215-229. MARSHALL, B. A. 1988b. Skeneidae, Vitrinellidae, and Orbi- testellidae (Mollusca: Gastropoda). Records of the National Museum of New Zealand 22:949-1004. Matsupa, R. 1987. Animal Evolution in Changing Environ- ments with Special Reference to Abnormal Metamorphosis. John Wiley & Sons: New York. xiv + 355 pp. MILEIKovsky, S. A. 1971. Types of larva development in ma- rine bottom invertebrates, their distribution and ecological significance: a reevaluation. Marine Biology 10:193-213. MINICHEV, Y. S. 1971. On the biology of some pyramidellids (Gastropoda, Pyramidellidae) of the Possjet Bay of the Sea of Japan. Academy of Sciences of the U.S.S.R., Zoological Institute (Leningrad). Explorations of the Fauna of the Seas 8(16):221-229 [in Russian]. MooruHousE, F. W. 1932. Notes on Trochus niloticus. Scientific Reports of the Great Barrier Reef Expedition 3:145-155. Mooers, M. G. 1981. Diet and reproductive biology of the rocky intertidal prosobranch gastropod 7ricolia pulloides. The Veliger 24(2):103-108. Naytor, E. 1976. Rhythmic behaviour and reproduction in marine animals. Pp. 393-429. In: R. C. Newell (ed.), Adap- tion to Environment: Essays on the Physiology of Marine Animals. Butterworths: London. PECHENIK, J. A. 1979. Role of encapsulation in invertebrate life histories. American Naturalist 114(6):859-870. PICKEN, G. B. 1979. Non-pelagic reproduction of some Ant- arctic prosobranch gastropods from Signy Island, South Ork- ney Islands. Malacologia 19(1):109-128. PowELL, A. W. B. 1942. The New Zealand Recent and fossil Mollusca of the family Turridae. Bulletin of the Auckland Institute and Museum, No. 2. 188 pp. POWELL, A. W. B. 1979. New Zealand Mollusca. Collins: Auckland, Sydney and London. xiv + 500 pp. PuRCHON, R. D. 1968. The Biology of the Mollusca. Pergamon Press: Oxford. 560 pp. RAMON, M. 1990. Spawning and development of Calliostoma granulatum in the Mediterranean Sea. Journal of the Marine Biological Association of the United Kingdom 70:321-328. Retzius, G. 1906. Die Spermien der Gasteropoden. Biolo- gische Untersuchungen 13:1-36. RoBERT, A. 1899. Gastéropodes. Le Troque (Trochus turbinatus (Born)). Jn: L. Boutan (ed.), Zoologie descriptive des formes tipiques d’Invertébres. Doin: Paris. RoBERT, A. 1902. Recherches sur le développement des troches. Archives de Zoologie Expérimentale et Générale, 3rd Ser. 10:269-538. ROBERTSON, R. 1985a. Four characters and the higher category systematics of gastropods. American Malacological Bulletin, Special Edition No. 1:1-22. ROBERTSON, R. 1985b. Archaeogastropod biology and the sys- tematics of the genus 77icolia (Trochacea: Tricoliidae) in the Indo-West Pacific. Monographs of Marine Mollusca No. 3: 1-103. RODRIGUEZ Basio, C. & C. THIRIOT-QUIEVREUX. 1974. Gas- téropodes de la région de Roscoff: étude particuliére de la protoconque. Cahiers de Biologie Marine 15:531-549. SCHELTEMA, R. S. 1967. The relationship of temperature to Page 272 The Veliger, Vol. 35, No. 4 the larval development of Nassarius obsoletus (Gastropoda). Biological Bulletin 132:253-265. SCHELTEMA, R. S. 1977. Dispersal of marine invertebrate or- ganisms: paleobiological and biostratigraphic implications. Pp. 73-108. In: E. G. Kauffman & J. E. Hazel (eds.), Concepts and Methods of Biostratigraphy. Dowden, Hutch- inson & Ross: Stroudsburg, Pennsylvania. SCHELTEMA, R.S. 1978. On the relationship between dispersal of pelagic veliger larvae and the evolution of marine proso- branch gastropods. Pp. 303-322. In: B. Battaglia & J. A. Beardmore (eds.), Marine Organisms: Genetics, Ecology and Evolution. Plenum Press: New York. SCHELTEMA, R. S. 1979. Dispersal of pelagic larvae and the zoogeography of Tertiary benthic gastropods. Pp. 391-397. In: J. Gray & A. J. Boucot (eds.), Historical Biogeography, Plate Tectonics, and the Changing Environment. Oregon State University Press: Corvallis, Oregon. SHuTo, T. 1974. Larval ecology of prosobranch gastropods and its bearing on biogeography and paleontology. Lethaia 7: 239-256. Simpson, R. E. 1977. The reproduction of some littoral mol- luscs from Macquarie Island (Sub-Antarctic). Marine Bi- ology 44:125-142. SOLIMAN, G. N. 1987. A scheme for classifying gastropod egg masses with special reference to those from the northwestern Red Sea. Journal of Molluscan Studies 53:1-12. STRATHMANN, M. F. 1987. Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. Univer- sity of Washington Press: Seattle. 682 pp. STRATHMANN, R. R. 1978. The evolution and loss of feeding larval stages of marine invertebrates. Evolution 32:894-906. STRATHMANN, R. R. 1985. Feeding and nonfeeding larval de- velopment and life-history evolution in marine invertebrates. Annual Reviews of Systematics and Ecology 16:339-361. Tuomas, F. I. M. 1992. Effects of adult morphology and gamete physical properties on performance of benthic in- vertebrates: spawning, mass transport, and breakage resis- tance. Doctoral Dissertation, University of California, Berkeley. THORSON, G. 1935. Studies on the egg capsules and develop- ment of Arctic marine prosobranchs. Meddelelser om Grgn- land 100:1-71. TuHorSON, G. 1936. The larval development, growth and me- tabolism of Arctic marine bottom invertebrates. Meddelelser om Gregnland 100:1-155. TuHorson, G. 1950. Reproductive and larval ecology of marine bottom invertebrates. Biological Reviews 25:1-45. TuHorsoNn, G. 1967. Clanculus bertheloti dOrbigny, 1839: eine brutpflegende prosobranchiate Schnecke aus der Brandungs- zone von Teneriffa. Zeitschrift fiir Morphologie und Oko- logie der Tiere 60:162-175. TRAPIDO-ROSENTHAL, H. G. & D. E. Morse. 1986. Regu- lation of receptor-mediated settlement and metamorphosis in larvae of a gastropod mollusc (Haliotis rufescens). Bulletin of Marine Science 39(2):383-392. TuRNER, R. D., J. A. PECHENIK & C. B. CALLOway. 1986. The language of benthic marine invertebrate development patterns: problems and needs. Pp. 227-235. In: Biology of Benthic Marine Organisms—Techniques and Methods as Applied to the Indian Ocean. Oxford & IBH Publishing Company: Bombay. UNDERWOOD, A. J. 1972a. Observations on the reproductive cycles of Monodonta lineata, Gibbula umbilicalis and G. cin- eraria. Marine Biology 17:333-340. UNDERWOOD, A. J. 1972b. Spawning, larval development and settlement behaviour of Gzbbula cineraria (Gastropoda: Prosobranchia) with a reappraisal of torsion in gastropods. Marine Biology 17:341-349. UNDERWOOD, A. J. 1974. The reproductive cycles and geo- graphic distribution of some common eastern Australian prosobranchs (Mollusca: Gastropoda). Australian Journal of Marine and Freshwater Research 25:63-88. WaREN, A. & P. BOUCHET. 1989. New gastropods from East Pacific hydrothermal vents. Zoologica Scripta 18(1):67-102. WEBBER, H. H. 1977. Gastropoda: Prosobranchia. Pp. 1-97. In: A. C. Giese & J. S. Pearse (eds.), Reproduction of Marine Invertebrates. Vol. IV. Molluscs: Gastropods and Cepha- lopods. Academic Press: New York. WELLS, F. E. & J. K. KEESING. 1987. Population characteristics of the gastropod Cantharidus pulcherrimus on intertidal plat- forms in the Perth area of Western Australia. Journal of the Malacological Society of Australia 8:23-35. WERTZBERGER, J. 1968. Behavior of Hiloa variabilis. Hawaiian Shell News 26(6):1. WILLIAMS, E. E. 1965. The growth and distribution of Gzbbula umbilicalis (DaCosta) on a rocky shore in Wales. Journal of Animal Ecology 33:433-442. The Veliger 35(4):273-294 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 Systematic Review of the Family Choristellidae (Archaeogastropoda: Lepetellacea) with Descriptions of New Species by JAMES H. McLEAN Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007, USA Abstract. "The known species in the family Choristellidae Bouchet & Warén, 1979 (= Choristidae of authors) are revised. All occur in continental shelf to abyssal depths and live in spent egg cases of sharks and rays, upon which they feed. The family is assigned to the Lepetellacea by Haszprunar on anatomical characters (1988a, b, c, 1992). Two genera with divergent shell form are recognized: the naticiform Choristella Bush, 1897, and the discoidal Bichoristes, gen. nov. The radula is unique to the family; shell characters are also diagnostic—extremely thin shell, deep suture (except in Bichoristes), complete peristome, sharp umbilical carination, small size, smooth protoconch with bulbous tip, and compressed earliest teleoconch. Previously described species of Choristella are C. tenera (Verrill, 1882) and C. leptalea Bush, 1897, both from the northwestern Atlantic, and C. vitrea (Kuroda & Habe, 1971) from Japan. New species proposed here are C. marshalli from New Zealand, C. nofroni from the Mediterranean, C. ponderi from eastern Australia, and C. hickmanae from Oregon. The monotypic new genus Bichoristes is based on B. wareni from New Caledonia. Bichoristes is considered to be derived from Choristella. Species previously but incorrectly assigned to Choristes Carpenter are discussed in the appendix. Choristes elegans Carpenter, 1872, has already been referred to Naticidae, but is here placed as a synonym of Amauropsis islandica (Gmelin, 1791). INTRODUCTION The family Choristellidae Bouchet & Warén, 1979, com- prises a poorly known group of small, thin-shelled, tro- chiform or naticiform species living offshore at continental shelf to abyssal depths. Living specimens have been col- lected only within the spent egg capsules of sharks and skates, upon which they feed. Shells are paper-thin and easily crushed. The Choristellidae are better known in literature prior to 1979 as the Choristidae Verrill, 1882, the family name having been intended for a species for which some unusual details of the radula, jaw, and external anatomy were originally described. The type species of Choristes, how- ever, is a fossil species that later proved to be a member of the Naticidae. BOUCHET & WAREN (1979) restored the original concept of the Choristidae by substituting the fam- ily name Choristellidae, a name based on Choristella Bush, 1897, another genus proposed in the family. Until recently, the systematic position of the family Choristellidae has been a matter of speculation. The radula provides few direct clues, as it is neither rhipidoglossate nor taenioglossate. VERRILL (1882) said nothing about the possible familial affinity of Choristidae, although BusH (1897) reported that “Professor Verrill placed it among the Tectibranchiata.” THIELE (1929), followed by WENZ (1938) and TayLor & SOHL (1962), placed Choristidae in the Rissoacea; KEEN (1971) placed it near Vitrinellidae; ABBOTT (1974:90) stated that “it may be a tectibranch.” GOLIKOV & STAROBOGATOV (1975:212, 220) placed Cho- ristidae in Naticacea (as order Aspidophora) ‘‘on the basis of shell characters and the shape of the radular teeth.” HICKMAN (1983) considered the choristellid radula to be close to that of the cocculiniform limpet Cocculinella Thiele, 1909. Further evidence in support of affinity be- tween choristellids and cocculiniform limpets was provided by the anatomical investigations of HASZPRUNAR (1988a, b, c), who placed the family in Lepetellacea, noting that Page 274 two of the included families, the Addisoniidae and Choris- tellidae, ‘““have a common ancestry, as revealed by their shared feeding biology (on empty egg-cases of skarks or skates), gill-type (with skeletal rods and mucous zones), and alimentary tract (complete loss of stomach)” (HASZPRUNAR, 1988a:19). Further details on choristellid anatomy and relationships are given by Haszprunar in the accompanying paper (HASZPRUNAR, 1992). This review started as an effort to give a name to the new eastern Pacific species cited by HICKMAN (1983), but it soon became apparent that additional new choristellids have recently been collected and have been awaiting at- tention in other museum collections. Here I update the classification of the family and add one new genus and five new species. Taxa removed from the family are discussed further in the Appendix to this paper. MATERIALS anp METHODS This review is hampered by a shortage of well-preserved material in collections. No material of the first and second named members of the family has been collected in recent years, and none of the original material remains wet-pre- served, making it difficult to verify the early descriptions of soft parts by examination of original material. It is only the newly collected material of the species described here that has made it possible for HASZPRUNAR (1988a, b, c, 1992) to report on the internal anatomy. Radulae were extracted after dissolution of tissue in 10% NaOH at room temperature; dry specimens of Chorvstella tenera were first rehydrated in detergent prior to treatment in NaOH. The radular ribbons were washed in distilled water, dried from a drop of water placed on a stub having a thin smear of rubber cement, and coated with gold or gold/palladium for examination with SEM. Jaws were also extracted with room temperature NaOH and exam- ined with SEM. Preserved specimens were critical point dried and examined with SEM. All depths that were originally cited in fathoms have been changed to meters. Abbreviations of institutions mentioned in the text: AMS, Australian Museum, Sydney; BMNH, Natural History Museum, London; LACM, Los Angeles County Museum of Natural History; MCZ, Museum of Comparative Zo- ology, Harvard University; MNHN, Museum National D’ Histoire Naturelle, Paris; NMNZ, National Museum of New Zealand, Wellington; NZOI, New Zealand Oceanographic Institute, Wellington; USNM, National Museum of Natural History, Washington, D.C. SYSTEMATICS Order ARCHAEOGASTROPODA Thiele, 1925 Suborder COCCULINIFORMIA Haszprunar, 1987 Superfamily LEPETELLACEA Dall, 1892 Although a monofamilial superfamily has been proposed for the Choristellidae (Choristiacea Kuroda & Habe, 1971, The Veliger, Vol. 35, No. 4 emended to Choristelliacea by HICKMAN, 1983), HASZPRUNAR (1988c) united the families Lepetellidae, Bathyphytophilidae, Pyropeltidae, Pseudococculinidae, Osteopeltidae, Cocculinellidae, Addisoniidae, and Choris- tellidae in the Lepetellacea, on the basis of sharing compact shell muscles, two kidneys, separated gonad with simple gonoducts, and statocysts with several statocones. Except for the Choristellidae, the lepetellaceans are her- maphroditic limpets. The family Choristellidae is the only member having a coiled shell and the only member that is gonochoristic. Family CHORISTELLIDAE Bouchet & Warén, 1979 CHORISTIDAE of authors (see below): VERRILL, 1882:540; THIELE, 1929:179; CLARKE, 1961:359; KEEN, 1971:388; ABBOTT, 1974:90; Boss, 1982:1010. CHORISTELLIDAE BOUCHET & WAREN, 1979:225: HICKMAN, 1983:86; HASZPRUNAR, 1988c:66. Included genera: Choristella Bush, 1897, and Bichoristes McLean, gen. nov. Choristella species are defined by dif- ferences in shell proportions, opercular coiling, and ex- ternal anatomy, although knowledge of external anatomy remains incomplete. Bichoristes is monotypic and based on a single specimen for which the shell, radula, opercu- lum, and jaw are known. The description of external anat- omy in the diagnosis that follows is based on that of Choris- tella. Diagnosis: Shell small (maximum dimension not exceed- ing about 10 mm), extremely thin, periostracum thin; whorls 3 to 3.5, rounded or carinate (carinate only in Bichoristes); suture deeply channeled (except in Bichoristes); spire height low to moderate; peristome complete, area of contact min- imal; final lip slightly flared; umbilicus narrow to wide; umbilical wall with sharp descending carina. Protoconch diameter 250-300 um, tip bulbous, surface smooth. Oper- culum of 3-10 whorls, multispiral to paucispiral. Jaw of two prominent, dark brown, finely reticulate plates, fused dorsally, laterally bowed to produce oval mouth opening with jagged edge. Radula. Rachidian tooth triangular, with short base and bluntly pointed overhanging cusp. First lateral tooth with quadrangular shaft, singly cusped in Choristella, bicuspid in Bichoristes. Second lateral tooth with long shaft, bi- cuspid in Choristella, unicuspid in Bichoristes. Third lat- eral tooth with long shaft and pointed cusp. Fourth lateral tooth similar, except reduced and fused to third in Bi- choristes. Fifth lateral tooth vestigial. Remarks: MARINCOVICH (1975, 1977) correctly placed Choristes Carpenter in Dawson, 1872, in the Naticidae (see further notes on Choristes elegans under excluded spe- cies), which left the living species of Choristes, of authors, in limbo. Without citing Marincovich, BOUCHET & WAREN (1979) proposed Choristellidae in a brief note. They wrote: “We want to use this occasion to point out that the genus Choristes Carpenter MS, Dawson, 1872 is a naticid. An J. H. McLean, 1992 examination of the types of Chorvstella leptalea Bush, 1897 (type species of Choristella) and C. tenera Bush, 1897 [ev- idently a lapsus for C. brychia Bush, 1897] has proved that they are synonyms of Choristes elegans var. tenera Verrill, 1882. Verrill’s name therefore has to be used for the type species. Another consequence is that the name Choristidae has to be changed to Choristellidae.” Boss (1982) missed the proposal of Choristellidae and followed MARINCOVICH (1977) in leaving all species de- scribed under Choristes within the Naticidae. Diagnostic shell characters for Choristellidae are the extremely thin shell, smooth protoconch with a bulbous tip, maximum of 3.5 teleoconch whorls, the complete peri- stome, and the sharp carination that descends within the umbilicus. Additionally, Choristella has a deeply channeled suture. Surprisingly, the descending umbilical carination has not previously been noticed, although it provides a consistent shell character for the family. The overall aspect of the radula is similar in the two genera, but differs in having the first lateral bicuspid in Bichoristes and the second lateral bicuspid in Choristella. Shared features are that the lateral teeth are robust and slope away from the rachidian, and that the shafts of the rachidian and first lateral are relatively short, whereas those of the second, third, and fourth laterals are longer and articulate together, and the fifth lateral is vestigial. The choristellid radula cannot be confused with that of any other family. Despite a statement (HICKMAN, 1983: 86) about radular affinity with the Cocculinellidae (“same basic pattern’’), the resemblance is superficial. The coc- culinellid radula, as illustrated by MARSHALL (1983), has the rachidian flanked by a pair of small teeth, followed by a series of stout interlocking teeth of similar morphology with serrate outer edges. Marshall considered the latter to be marginal teeth and the lateral teeth to be represented by the small inner pair. The choristellid radula differs in having the rachidian flanked by massive teeth and none of the succeeding teeth in the row are similar. The bicuspid second lateral tooth of Choristella and the bicuspid first lateral tooth of Bichoristes are evidently fused from the primitive condition for the family, which is not represented in a living genus. The teeth of both families are probably homologous, but I am more inclined to regard the teeth of each family as lateral teeth than as marginals for two reasons: I know of no other examples of massive lateral teeth and the.paired teeth of the choristellid radula could hardly be considered marginals because they have laterally extended shafts, as well as exhibiting partial fusion. The choristellid protoconch has a bulbous tip, similar to that of the Cocculinidae (see MARSHALL, 1986:fig. 5D), but unlike the compressed and laterally pinched tip of the cocculinellid protoconch (MARSHALL, 1983:fig. 11) or the pseudococculinid protoconch (MARSHALL, 1986:fig. 9H). Close affinity with either family is therefore not supported on evidence from the protoconch. Unfortunately, the ad- disoniid protoconch remains unknown (MCLEAN, 1985) and it is not yet possible to confirm with protoconch evi- Page 275 dence the affinity of the two families as advocated by HASZPRUNAR (1992) on anatomical evidence. Genus Choristella Bush, 1897 Choristes Carpenter, of VERRILL, 1882:540; DALL, 1908:328; THIELE, 1929:179; CLARKE, 1961:359; KEEN, 1971:388; ABBOTT, 1974:90. Not Choristes Carpenter in Dawson, 1872 [Naticidae]. Choristella BUSH, 1897:138; THIELE, 1929:179; BOUCHET & WAREN, 1979:225; HICKMAN, 1983:86. Type species (original designation): Choristella leptalea Bush, 1897. Diagnosis: Shell small (maximum dimension about 10 mm), extremely thin (maximum thickness of broken edge 0.05 mm), easily damaged; periostracum thin; whorls 3 to 3.5, rounded; suture deeply channeled, spire height low to moderately high. Peristome complete, contact with pre- vious whorl limited to narrow band; final lip flared but not thickened, reflected near base of columella. Umbilicus narrow to broad, umbilical wall with sharp descending carina that terminates on reflected region of lip at base of columella. Protoconch diameter 250 um, surface smooth; tip bulbous. Outer edge of first quarter turn of teleoconch compressed, not forming regular curve. Operculum thin, up to 5 whorls, multispiral or with final whorl enlarged to give paucispiral effect. External anatomy. Snout prominent, eyes lacking, ce- phalic and epipodial tentacles lacking micropapillae. One to two suboptic tentacles short, posterior to right cephalic tentacle. Gill pectinibranch, leaflets numerous. Sexes sep- arate; male using right cephalic tentacle as copulatory organ; open seminal groove on right tentacle. Jaw. As described for family. Radula. Rachidian tooth relatively small, with trian- gular shaft and small overhanging cusp; base of shaft broadly emerging from ribbon. First lateral tooth massive, shaft quadrangular, overhanging cusp large, triangular, with bluntly pointed tip; base of shaft articulating with tooth below, base of shaft buttressed on inner and outer edges; second lateral tooth separated from third by open channel. Second lateral tooth largest in row, with two large cusps, the innermost with triangular cusp matching that of second lateral, the outermost cusp having a more obtuse angle; position of both cusps descending away from ra- chidian; base with projecting ridge above excavation that accommodates tooth below. Third lateral tooth with long shaft and thick, rounded cusp that projects over the outer cusp of second lateral tooth; base buttressed on inner side by narrow ridge. Fourth lateral tooth with longest shaft and small, beaklike cusp, base buttressed on inner side by projecting ridge. Fifth lateral tooth vestigial, closely ap- pressed to base of fourth lateral tooth. Remarks: Choristella species may be recognized on shell characters alone (thin shell, channeled suture, complete peristome, compression of early teleoconch, and descending umbilical carination). The descending umbilical carination Page 276 may be shared with some skeneiform genera, including Trenchia Knudsen, 1964, as discussed here under rejected species. On shell characters, Choristella may be distin- guished from such genera in having a much more deeply channeled suture and by the compression of the early te- leoconch (for the latter see especially Figure 22). The radulae of all species examined are closely similar. Some differences that may be apparent in the illustrations for each species can be attributed to wear, rather than interspecific differences. The most useful radular char- acters for interspecific discrimination are the morphology and relative size of the rachidian tooth. The bulbous tip of the protoconch is treated under the family heading. In some species the protoconch remains unknown; in all the available specimens of such species it is replaced by a calcareous plug, representing an internal mold of the original protoconch (see Figure 11). In proposing Choristella, BUSH (1897) emphasized a radular difference from Choristes. According to Bush, Cho- ristella leptalea has 13 teeth in the row, as opposed to 11 teeth in Choristes elegans var. tenera. Both CLARKE (1961: 359) and BOUCHET & WAREN (1979) discounted a radular distinction, and attributed the tooth count discrepancy to varying interpretations of the second lateral tooth either as a bicuspidate compound tooth or two separate teeth. | interpret the second tooth as a compound tooth derived by fusion of two separate teeth. BOUCHET & WAREN (1979: fig. 12) provided a drawing of the radula of Choristella tenera that showed the rachidian and five lateral teeth, making a total of 11 teeth in the row. That interpretation of the radula is followed here. Although the radula of Choristella leptalea is not available for SEM study, a generic distinction based on radulae is evidently unfounded. Despite the lack of evidence from the radula, the con- clusion that the taxa proposed separately by Verrill and Bush are the same is not supported here. There are other, more important differences, one of which was well figured in the original accounts: the operculum of Chorvstella tenera is shown with three whorls and expands so rapidly that it looks to be paucispiral (see VERRILL, 1882:pl. 58), whereas the operculum of C. leptalea is shown as multispiral, with five whorls (BUSH, 1897:fig. 8). There are also differences in shell proportions between the two species: C. leptalea is clearly lower-spired than C. tenera, and is smaller. Both have the same number of whorls, which suggests that they are based on mature specimens. There are also differences in the external anatomy that can be detected from a careful reading of the original descriptions. Bush recognized two species and intended to place them in separate genera. I accept that there are two species (contrary to BOUCHET & WAREN, 1979, who recognized only one), but am unable to support a generic distinction. The other species treated here cannot be placed into two separate groups on characters now available. Unfortu- nately, the replacement of Choristes by Choristella changes the type species of the nominate genus to Choristella lep- talea, a species that remains poorly known. The Veliger, Vol. 35, No. 4 On the basis of shell proportions there are two groups of species in Chorvstella, a relatively high-spired group and a relatively low-spired group. Opercular characters do not support generic groupings based on shell proportions, how- ever. High-spired species are C. tenera (Verrill, 1882), C. vitrea (Kuroda & Habe, 1971), C. marshalli sp. nov., and C. nofronu sp. nov. Low-spired species are C. leptalea Bush, 1897, C. ponderi, sp. nov., and C. hickmanae, sp. nov. Choristella tenera (Verrill, 1882) (Figures 1-7) Choristes elegans var. tenera VERRILL, 1882:541, pl. 58, figs. 27 [shell with operculum], 27a [radula]; VERRILL, 1884: 256, pl. 29, figs. 9, 9a, 9b [shells of 3 juvenile specimens]. Choristes tenera: CLARKE, 1961:360; ABBOTT, 1974:90, fig. 865 [copy figs. of VERRILL, 1882]. Choristella tenera: BOUCHET & WAREN, 1979:225, fig. 225 [new drawing of radula, based on paratype, USNM 45151]. Description: Shell (Figures 1-4) large for genus (maxi- mum diameter 10.5 mm), spire height relatively high (height—width ratio of holotype 0.87). Shell wall extremely thin. Surface shiny, brown, periostracum thin, surface fine- ly pitted. Protoconch usually eroded and filled with sec- ondary plug, separated from first teleoconch whorl. Te- leoconch whorls 3.5, rounded, smooth; suture deeply impressed. Umbilicus narrow, deep, not obstructed by re- flection of inner lip. Spiral sculpture represented by fine striae strongest on base and by single narrow ridge deep within umbilicus; axial sculpture lacking except for fine growth increments. Peristome complete, area of contact with previous whorl minimal. Lip flared at base of colu- mella where buttressed by umbilical ridge. Operculum (Figure 4) pale brown, nucleus slightly excentric, 3 whorls, inner edge growing under outer edge of previous whorl (which raises the outer edge of previous whorls), final whorl expanding to produce paucispiral pattern. Dimensions. Height 5.4 mm, width 6.2 mm (holotype); height 9.0 mm, width 10.5 mm (largest specimen, USNM 78902). External anatomy. Because freshly collected, preserved specimens are not available, VERRILL’s (1882) original description of the animal is repeated here: “Head large, short, thick, rounded or truncate, with two short, flat, obtuse anterior tentacles, wide apart, but connected to- gether by a transverse fold; posterior tentacles short, thick, conical, smooth; no eyes visible; proboscis [buccal mass] short, thick, retractile; jaws crescent-shaped, strong, black. Verge situated just below the right posterior [error for anterior?] tentacle, small, papilliform, swollen at base; below this and farther back, a larger and thicker papilla with basal swelling; on each side, between the mantle and foot, at about midlength of the foot, a small mammiform papilla; and two small flat cirri, behind and beneath the operculum. Foot broad, ovate, with two tentacle-like pro- Explanation of Figures 1 to 7 Figures 1-7. Choristella tenera (Verrill, 1882). Figures 1-3. Holotype, USNM 45151, off Martha’s Vineyard Island, Massachusetts, USA. Height 5.4 mm. Apertural, oblique spire, and umbilical views. Figure 4. Largest specimen, showing operculum in place, USNM 78902, USFC Sta. 2730, off Cape Hatteras, North Carolina. Height 9.0 mm. Figure 5. SEM view of jaw, USNM 78902. Scale bar = 200 um. Figure 6. Protoconch, USNM 45253. Scale bar = 100 wm. Figure 7. SEM view of radula, USNM 78902. Scale bar = 40 um. Page 278 cesses in front. Gill large, consisting of numerous thin lamellae, attached to the inner surface of the mantle, over the left side of the neck, and extending obliquely across and over the neck to the right side.” Jaw (Figure 5). Typical for family. Radula (Figure 7). Characteristic for family. Rachidian tooth stout, relatively broad, tip apparently not overhung in present preparation. Type locality: Off Martha’s Vineyard Island, Massachu- setts, USFC Sta. 1031, 466 m, “‘taken from the interior of an old egg-case of a skate (Raia, sp.).” Type material: Holotype, USNM 45151, USFC Sta. 1031, collected in 1881, shell intact, body dried. Eight paratypes in similar condition, USNM 859486. USNM_ 508720, USFC Sta. 1031, 1 paratype same station as type lot. The shell surface of the type lot is dull from prior preservation in alcohol, although other specimens have a shiny surface. Referred material: 7 USNM lots, all dry, most with dried bodies: USNM 45252, USFC Sta. 1096, 580 m off Mar- tha’s Vineyard, 4 broken shells, one loose body attached to operculum. USNM 45253, USFC Sta. 1124, off Mar- tha’s Vineyard, 2 large and numerous small shells. USNM 45254, USFC Sta. 1154, 353 m off Martha’s Vineyard, 1 shell, operculum in place. USNM 45255, USFC Sta. 2234, off Martha’s Vineyard, 1 shell, operculum in place. USNM 40309, USFC Sta. 2262, off Nantucket Shoals, 3 shells, opercula in place. USNM 78902, USFC Sta. 2730, off Cape Hatteras, North Carolina, 1 large and several small shells, all with opercula in place. USNM 78901, USFC Sta. 2731, off Cape Hatteras, North Carolina, 4 small shells, 2 with dried bodies. Remarks: All specimens have the sharp, steeply descend- ing umbilical carination, a diagnostic character that was missed by VERRILL (1882) in the original description and not subsequently noticed. Verrill reported that large spec- imens have 4 to 5 whorls, but this is clearly in error, as the largest specimens do not exceed 3.5 whorls. Verrill compared it to a small specimen of Choristes elegans Car- penter, which he had received from Dawson (VERRILL, 1882:542, pl. 58, fig. 28), considering it “a thin and delicate variety of the ancient type.” Verrill’s description of the external anatomy noted a “‘verge’’ [penis] posterior to the right cephalic tentacle, but this is here regarded as a suboptic tentacle. Choristella marshalli McLean, sp. nov. (Figures 8-15) Description: Shell (Figures 8-10) large for genus (max- imum diameter 8.8 mm), spire height relatively high (height—width ratio of holotype 0.90). Shell wall extremely thin, maximum thickness of broken lip 0.05 mm. Surface shiny, light brown; periostracum thin, surface finely pitted. Protoconch usually etched away and filled with secondary The Veliger, Vol. 35, No. 4 plug, separated from first teleoconch whorl. Teleoconch whorls 3.3, rounded, smooth; suture deeply impressed. Umbilicus narrow, deep, not obstructed by reflection of inner lip. Spiral sculpture represented by fine striae and by single narrow ridge deep within umbilicus; axial sculp- ture lacking except for fine growth increments. Peristome complete, area of contact with previous whorl minimal. Lip flared at base of columella where buttressed by um- bilical ridge. Operculum (Figure 12) pale brown, nucleus slightly excentric, final 3 whorls evenly expanding. Dimensions. Height 7.9 mm, width 8.8 mm (holotype). External anatomy (Figure 14). Right cephalic tentacle of male with open groove. Jaw (Figure 13). Typical for family. Radula (Figure 15). The radula closely approximates that given for the familial description. The shaft of the rachidian is well marked and there is a small overhanging tip. The outermost tooth in the row is unusually well developed. Type locality: SE of Banks Peninsula (44°55.4'S, 174°04.9'E), New Zealand, 1097-1116 m, in empty skate egg case. Type material: 26 specimens—11 intact shells, 15 spec- imens with broken shells and bodies preserved in alecohol— from type locality, R/V James Cook, Sta. J10/37/84, 15 June 1984. The visceral mass has disintegrated in the preserved specimens, which were initially preserved by freezing. Holotype NUNZ M.109053 and 23 paratypes NMNZ M.75210; 1 paratype LACM 2247; 1 paratype AMS. Referred material: NZOI Sta. 132 off Cape Brett, New Zealand (35°11.7'S, 174°49.8'E), 376-450 m, R/V Tan- garoa, 7 May 1975, 2 dried, damaged specimens and 1 small preserved body. NZOI Sta. P292, Tasman Basin (40°42.8'S, 167°56.0’E), 1029 m, 4 preserved specimens, shells broken. NUNZ M.89950, NE of Chatham Island, New Zealand (42°52.3'S, 175°37.3'E), 1032 m in elas- mobranch egg case, F/V Akagi Maru, 9 June 1987, about 15 decalcified or broken-shelled juveniles in alcohol plus about 12 small specimens with dried bodies (SEM of early whorls, Figure 11). Remarks: This species is characterized by its relatively large size and high spire. It resembles Choristella tenera in its size and proportions, but has a less prominent perios- tracum. As in C. tenera, the protoconch of most specimens is etched away, leaving only a plug that is well separated from the first teleoconch whorl (Figure 11). The opercu- lum (Figure 12) is like that of C. tenera, although it has more numerous whorls and the final whorl is not so rapidly expanding. The open seminal groove on the right cephalic tentacle is visible in the critical point dried specimen examined with SEM (Figure 14). Hicks (1986) reported that skate egg cases containing Ee EyMeclean, 119.92 Page 279 Explanation of Figures 8 to 15 Figures 8-15. Choristella marshalli McLean, sp. nov. Figures 8-10. Holotype, NMNZ 75210, SE of Banks Peninsula, New Zealand. Height 7.9 mm. Apertural, oblique lateral, and umbilical views. Figure 11. Early Whorls, showing plug filling protoconch, NUNZ M.89950, NE of Chatham Islands, New Zealand. Scale bar = 200 um. Figure 12. SEM view of operculum, NMNZ 75210, paratype. Scale bar = 1 mm. Figure 13. SEM view of jaw, NMNZ 75210, paratype. Scale bar = 200 um. Figure 14. SEM view of critical point dried paratype, anterior view of body attached to operculum, showing groove on right cephalic tentacle (arrow), NMNZ 75210, paratype. Scale bar = 1 mm. Figure 15. SEM view of radula, NMNZ 75210, paratype. Scale bar = 40 um. The Veliger, Vol. 35, No. 4 Page 280 J. H. McLean, 1992 Choristella species (cited here as the type material of C. marshalli) also yielded type material of the harpacticoid copepod Paramphiascopsis waithonu Hicks, 1986. Harpac- ticoids have been noted to feed on microbiota associated with fecal pellets (HIcKs, 1986). Etymology: The name honors Bruce A. Marshall of the National Museum of New Zealand, Wellington. Choristella vitrea (Kuroda & Habe, 1971) Choristes vitreus Kuroda & Habe in KURODA, HABE & OYAMA, 1971:62, pl. 107, fig. 11. Description (copied from Kuroda & Habe): “Shell rath- er small, thin, translucently white, turbinate in shape. Spire conical and with 5 whorls, rather rapidly increasing their width to the body whorl, well inflated and separated by the deeply impressed sutures. Surface smooth and pol- ished and covered by a thin periostracum and sculptured by the very faint spiral threads and growth lines. Body whorl large and well rounded at the periphery and the base. Aperture wide and semicircular. Outer margin well rounded, thin and slightly expanded. Innermargin [sic] deposited the thin callus on the parietal wall and rather straight [sic]. Columellar margins reflexed and dilated over the widely and deeply perforated umbilicus. Operculum thin, corneus, pale yellowish brown and paucispiral.” Dimensions. Height 10.7 mm, diameter 9.5 mm (holo- type); height 12.2 mm, diameter 9.4 mm (paratype). Type locality: Sagami Bay, Japan, “parasitic on the egg capsules of shark,” depth not indicated. Type material: Holotype and paratype, presumably in Imperial Household Collection, Japan. No other speci- mens are known. Remarks: Although the original material has not been examined and the radula has not been described, the de- scription of this species is compatible with that of the high- spired species group of Choristella. The shell is comparable to C. tenera in size, thinness of shell, and opercular mor- phology, and to C. nofronw in having the height of the shell exceed the breadth. The operculum was said to be paucispiral. The height—width ratio of the holotype is 1.3, compared to 1.13 for C. nofroni. Page 281 Choristella nofronia McLean, sp. nov. (Figures 16-24) Cithna naticiformis Jeffreys, 1883, of GUBBIOLI & NOFRONI, 1986:204 [figures not numbered, size not indicated], non Cithna naticiformis Jeffreys, 1883 Description: Shell (Figures 16-18) medium size for genus (maximum height 6.1 mm), spire height relatively high (height-width ratio of holotype 1.13). Shell wall extremely thin, maximum thickness of broken lip 0.05 mm. Surface shiny, yellowish white, periostracum thin. Protoconch (Figures 21, 22) diameter 250 um, surface smooth. Te- leoconch whorls 2.7 rounded, smooth; suture deeply im- pressed. Umbilicus narrow, deep, partially obstructed by reflection of inner lip. Spiral sculpture of faint striae and single narrow ridge deep within umbilicus, terminating at columellar flare. Base of mature shell rounded, that of immature shell with angulation. Axial sculpture of ex- tremely fine growth increments, sharply raised on umbil- ical slope. Peristome nearly complete. Operculum (Figure 19) pale brown, nucleus slightly excentric, final whorl becoming paucispiral. Dimensions. Height 6.1 mm, width 5.4 mm (holotype). Jaw (Figure 20). Typical for genus. Radula (Figure 24). Typical for the family; the shaft of the rachidian is weakly projecting, the overhanging tip of the rachidian is small but clearly revealed. Type locality: Alboran Sea, westernmost Mediterranean, west of Cabo de Gata, Spain (extending from 01°30'W and 35°30’ to 36°30'N, according to P. Bouchet), 50-100 m. Type material: Holotype (Figures 16-18) MNHN un- cataloged, operculum and radula scanned. Four paratypes MNHN uncataloged (heights 5.2, 3.1, 1.7, 1.2 mm). Two paratypes LACM 2248 (height 3.0 mm, _ protoconch scanned; height 4.1 mm, lip broken). All specimens from the generalized type locality, obtained by I. Nofroni from local fishermen. Referred material: AMS C.167316, Al Hoceima, Mo- rocco (35°14’N, 03°56'W), 50-100 m, with Raja egg cases, August 1986, F. Gubbioli, 2 dry specimens. GUBBIOLI & NOFRONI (1986) wrote: “All our findings, dozens of specimens, many live, come from eggs of Raja Explanation of Figures 16 to 24 Figures 16-24. Choristella nofronia McLean, sp. nov. Figures 16-18. Holotype, MNHN, Alboran Sea, western Mediterranean. Height 6.1 mm. Apertural, spire, and umbilical views. Figure 19. SEM view of operculum of holotype. Scale bar = 1 mm. Figure 20. SEM view of jaw of holotype (elements separated). Scale bar = 200 um. Figure 21. SEM view of larval shell, topotypic material, courtesy A. Waren. The straight diagonal line is an artifact of scanning. Scale bar = 100 um. Figure 22. SEM view of protoconch and first teleoconch whorl of paratype, LACM 2248. Scale bar = 100 um. Figure 23. SEM, oblique umbilical view of juvenile shell showing basal ridge, topotypic specimen, courtesy A. Waren. Scale bar = 1 mm. Figure 24. SEM view of radula, paratype, LACM 2448. Scale bar = 40 um. Page 282 cf. clavata fished in the quadrilateral Marbella, S. Roque (Spain), Tetuan, Al Hoceima (Morocco) at depths between 50 and 100 m.” Remarks: Choristella nofroni is characterized by its rela- tively small size and high profile. In addition, small shells have a weak mid-basal ridge, a feature not observed in any other species. GUBBIOLI & NOFRONI (1986) found this species in 5% of 250 of the egg cases they examined and found that three times as many had specimens of the limpet Addisonia la- teralis (Requien, 1848). Both species were noted in 3% of the examined egg cases. The choristellid affinity was unknown by GUBBIOLI & NOFRONI (1986), who identified it as “Czthna”’ naticiformis Jeffries, 1883. The basal ridge that characterizes small shells (Figure 23) led them to associate the species with Jeffreys’ taxon from 1453 m (795 fm) off the Portuguese coast. However, the basal ridge of that species (syntypes, BMNH 85.11.5.1615-1617, Figures 60, 61) is much more pronounced, and there is a concave rather than convex surface between the umbilical and basal ridges. Jeffreys’ species is treated further in the Appendix. GUBBIOLI & NOFRONI (1986) also suggested that “‘Cy- clostrema” valvatoides Jeffreys, 1883, might also be refer- able to the present species. I have examined the holotype of that species (BMNH 85.11.5.1593). Choristellid affinity is ruled out because it does not have the umbilical ridge characteristic of the family. Etymology: The name honors Italo Nofroni, one of the collectors of the original material. Choristella leptalea Bush, 1897 (Figures 25-29) Choristella leptalea BUSH, 1897:139, text fig. 8 [operculum], text fig. 9 [shell], pl. 23, figs. 16, 16a [radula]. Choristella brychia BUSH, 1897:139, text fig. 10 [spire view of shell}. Description: Shell (Figures 25-29) small for genus (max- imum diameter 4.0 mm), spire height relatively low (height- width ratio of holotype 0.71). Shell wall extremely thin. Shell white, periostracum thin, light brown. Protoconch diameter about 300 um. Teleoconch whorls 3.4, rounded, smooth, suture deeply impressed. Umbilicus narrow, deep, not obstructed by reflection of inner lip, inner extent of umbilicus defined by narrow ridge. Spiral sculpture lack- ing; axial sculpture lacking, except for fine growth incre- ments. Peristome complete, area of contact with previous whorl minimal; lip flared below, broadest at base of col- umella, where meeting umbilical ridge. Operculum of 4.5 whorls, nucleus slightly excentric, final 3 whorls evenly expanding in multispiral pattern. Dimensions. Height 2.5 mm, width 3.5 mm (original measurements of holotype); height 3.1 mm, width 4.0 mm (new measurements of holotype of Choristella brychia). External anatomy. BUSH’s (1897) description is copied The Veliger, Vol. 35, No. 4 here: “The animal has a broad emarginate head with one pair of long slender tentacles; with a rather broad, short, tapered, ciliated verge just beneath the base of the right one. Eyes none. Gill attached to the left side lying across the top of the body just within the mantle edge.” Radula. As noted in the remarks under the genus, the radular illustration and tooth count provided by Bush is incorrect; the radula is probably typical for the genus. Type localities: For Choristella leptalea, off Martha’s Vineyard Island, Massachusetts (USFC Sta. 2547), 713 m, 1885. For C. brychia, off Martha’s Vineyard Island, Massachusetts (USFC Sta. 2234), 1481 m, 1884. Type material: Holotype, Choristella leptalea, USNM 52504 (Figures 25, 26). Although collected alive, the spec- imen is now broken, the final whorl separated. The label reads “‘jaw-radula, operculum mounted,” but these prep- arations could not be located. Holotype, Choristella brychia, USNM 77622 (Figures 27-29). The specimen is intact, although the lip is now broken at the base. Remarks: Choristella leptalea is a relatively small-sized member of the family, having a maximum dimension of only 4.0 mm, compared to 10 mm reached by some species. The number of whorls is equal to that of other species, which suggests that it is based on mature specimens. It occurs sympatrically with C. tenera, from which it differs in its lower spire. Choristella brychia Bush, 1897, was based on a single specimen. It was described briefly: “This is a larger species of firmer texture than the preceding [C. leptalea], although of the same number of whorls. Sculpture none. Color dirty white tinted with brown. Where not worn the surface is slightly lustrous. Interior of aperture very smooth and lustrous, showing a sutural band of delicate rose color.” The size difference of 0.5 mm is not sufficient grounds to recognize C. brychia as a species distinct from C. leptalea. The original figures of the shells are not helpful because an apertural view was used for C. leptalea, whereas a spire view was given for C. brychia. Although Bush did not state that the operculum of Cho- ristella leptalea is multispiral, her fig. 8 clearly shows 4.5 whorls in a multispiral pattern. The diagnosis above in- cludes mention of periostracum, based on my examination of the holotype of C. brychia, although this was not men- tioned by Bush. The remains of the holotype of C. leptalea show an extremely thin, pale periostracum, not as dark as that of C. tenera. The original description of C. leptalea does not include mention of the carination that descends within the umbilicus, which is clearly visible on the ho- lotypes of both C. leptalea and C. brychia. Choristella ponderit McLean, sp. nov. (Figures 30-38) Description: Shell (Figures 30-33) small for genus (max- imum diameter 4.7 mm), spire height relatively low (height- J. H. McLean, 1992 Page 283 Explanation of Figures 25 to 29 Figures 25-29. Choristella leptalea Bush, 1897. Figures 25, 26. Holotype, USNM 52504, off Martha’s Vineyard Island. Original height 2.5 mm (BusH, 1897). Figure 25, broken remains of aperture. Figure 26, broken remains of spire. Figures 27-29. Holotype of Choristella brychia Bush, 1897, USNM 77622, off Martha’s Vineyard Island. Height 3.1 mm. Apertural, basal, and oblique spire views. width ratio of holotype 0.68). Shell wall extremely thin. Surface shiny, white, periostracum thin, colorless. Proto- conch (Figure 36) tip bulbous, surface smooth. Teleoconch whorls 3, rounded, smooth, suture deeply impressed. Um- bilicus narrow, deep, not obstructed by reflection of inner lip, inner extent of umbilicus defined by narrow ridge. Spiral sculpture represented only by umbilical ridge; axial sculpture lacking, except for fine growth increments. Peri- stome complete, area of contact with previous whorl min- imal; lip flared below, broadest at base of columella, where buttressed by umbilical ridge. Operculum (Figure 34) pale brown, nucleus slightly excentric, final whorl rapidly ex- panding to produce paucispiral pattern. Dimensions. Height 3.2 mm, width 4.7 mm (holotype); height 3.5 mm, diameter 4.7 mm (figured specimen, AMS C.155463). External anatomy (Figure 37). The mouth is bordered laterally by projecting oral lappets. No groove on the right tentacle was detected, but the specimen may be female. Jaw (Figure 35). As described for genus. Radula (Figure 38). The radula agrees with that given for the family. The rachidian is unusual in the genus in seeming to have three projecting nubs at the base of the shaft. Type locality: Off Sydney, New South Wales, Australia (33°47.5'S, 151°28.5’E), 124 m, in skate egg case. Type material: 6 specimens from type locality, R/V Ka- pala Sta. K86/14/16, 2 July 1986. Holotype and para- types AMS C.151524, bodies preserved separately. Ten additional paratypes, off Shoalhaven Heads, N.S.W. (34°56'S, 151°9.5'E), 494-585 m, in elasmobranch egg cases, R/V Kapala Sta. K86/23/04, 10 September 1986, small to medium-sized specimens with dried bodies, 3 spec- imens wet-preserved; distribution: 6 paratypes AMS C.167692; 1 paratype LACM 2630, 2 paratypes NMNZ, 1 paratype MNHN. Referred material (arranged north to south): AMS C.155457, NE of North Reef, Queensland (23°08.4'S, 152°12.3'E), R/V Kimbla Sta. 20, 14 December 1977, 1 dead specimen. AMS C.155458, E of North West Island, Queensland (23°19.5'S, 152°35.4’E), 320 m, R/V Kimbla Sta. 23, 14 December 1977, 1 dead specimen. AMS C.155462, E of Lady Musgrave Island, Queensland (23°33.7'S, 152°37.0'E), 339 m, R/V Kimbla Sta. 3, 17 November 1977, 1 dead specimen. AMS C.155459, NE of Lady Musgrave Island, Queensland (23°38.8'S, 152°45.5'E), 365 m, R/V Kimbla Sta. 24, 14 December 1977, 4 small dead shells. AMS C.155461, E of Lady Musgrave Island, Queensland (23°44'S, 152°49'E), 357 m, R/V Kimbla Sta. 2, 17 November 1977, 1 dead spec- imen. AMS C.151990, E of Lady Musgrave Island, Queensland (23°52.2'S, 152°42.2’E), 296 m, R/V Kimbla Page 284 The Veliger, Vol. 35, No. 4 ay Explanation of Figures 30 to 38 Figures 30-38. Choristella ponderi McLean, sp. nov. Figures 30-32. Holotype, AMS C.151524, off Sydney, New South Wales, Australia. Height 3.2 mm. Apertural, spire, and umbilical views. Figure 33. AMS C.155463, off Fraser Island, Queensland, Australia. Height 3.5 mm. Figure 34. SEM view of operculum of paratype, AMS C.151524. Scale bar = 1 mm. Figure 35. SEM view of jaw of paratype, AMS C.151524. Scale bar = 200 um. Figure 36. SEM view of early whorls, showing protoconch and first teleoconch whorl. SEM photo by B. Marshall. AMS C.82431, off Caloundra, Queensland. Scale bar = 100 um. Figure 37. SEM view of critical point dried body, showing cephalic tentacles, oral lappets, and foot with pedal gland, paratype, AMS C.151524. Figure 38. SEM view of radula, paratype, AMS C.151524. Scale bar = 25 um. J. H. McLean, 1992 Sta. 15, 7 July 1984, 3 dry specimens. AMS C.155460, off Frazer Island, Queensland (24°57.9'S, 153°37.3’E), 210 m, R/V Kimbla Sta. 27, 15 December 1977, 1 small dead specimen. AMS C.155463, S end Fraser Island Queensland (27°57'08"S, 153°51'03”E), 201 m, R/V Kim- bla Sta. Q13, 10 November 1976, 1 dry specimen (Figure 33). AMS C€.150125, N of Coolongatta, Queensland (28°07’S, 153°50’E), 146 m, R/V Kapala Sta. K78-17-14, 18 August 1978, 1 small specimen. AMS C.82431, E of Caloundra, Queensland, 91-110 m (50-60 fm), T. A. Garrard Coll., 1 specimen (Figure 37, protoconch). AMS C.150127, off Kiama, N.S.W. (34°46’S, 151°13’E), 387- 552 m, in egg case, R/V Kapala Sta. K86-09-03, 15 April 1986, 2 small specimens, dry shells and wet bodies sepa- rate. Remarks: This species is characterized by its small size, low spire, and relatively few whorls. There are a sufficient number of records to be certain that the specimens are mature. In its small size and low spire it is most similar to Choristella leptalea Bush, a species too poorly known to allow full comparison. Choristella ponderi is broadly distributed on the east coast of Australia. Records are known from Queensland (23°08’S) to New South Wales (34°56’S). Etymology: The name honors Winston Ponder, of the Australian Museum, Sydney. Choristella hickmanae McLean, sp. nov. (Figures 39-45) Choristella n. sp.: HICKMAN, 1983:86, fig. 29 [radula]. Description: Shell (Figures 39-43) large for genus (max- imum diameter 9 mm), spire height relatively low (height- width ratio of holotype 0.72). Shell wall extremely thin, maximum thickness of broken lip 0.1 mm. Surface dull, yellowish white, periostracum not evident, surface finely pitted. Protoconch and earliest teleoconch whorl missing. Remaining whorls 3.5, rounded, smooth; suture deeply impressed. Umbilicus broad, deep, not obstructed by re- flection of inner lip. Spiral sculpture represented only by single narrow ridge deep within umbilicus; axial sculpture lacking, growth increments not apparent. Peristome com- plete, area of contact with previous whorl minimal. Oper- culum (Figure 39) pale brown, nucleus slightly excentric, final 3 whorls evenly expanding in multispiral pattern. Dimensions. Height 6.5 mm, width 9.0 mm (estimated dimension of holotype prior to breakage); height 7 mm, diameter 10 mm (estimated dimension of sectioned para- type). External anatomy. Figure 44 shows the left (umbilical view) side of a paratype specimen prior to sectioning. Four epipodial tentacles are shown adjacent to the operculum. Radula (Figure 45). The radula agrees with the generic description in its overall morphology. The rachidian tooth Page 285 has a weakly projecting shaft, but a small, clearly distinct, overhanging cusp. Type locality: Northern Cascadia Abyssal Plain, at base of continental slope, 95 nautical miles (172 km) west of Strait of Juan de Fuca, Washington (48°38.1'N, 126°58.0'W), 2176 m, gray silty clay. CAREY (1981) de- scribed the bottom conditions for the Cascadia Abyssal Plain. Type material: 3 specimens from type locality, all with damaged shells, collected with beam trawl by A. Carey, Oregon State University (BMT-DWD Sta. 9), 11 Sep- tember 1971. Holotype, LACM 2249 (Figures 42, 43) body used for light microscope preparation of radula. Two paratypes, LACM 2250, one sectioned, shell destroyed, photograph of shell and body prior to sectioning (Figures 39-41), one paratype specimen with badly damaged shell used for SEM preparation of radula by C. Hickman (Fig- ure 45). Remarks: Choristella hickmanae is a relatively low-spired species comparable to C. leptalea and C. ponderi, but is larger than either species (9 mm maximum dimension, compared to 4.0 mm for C. leptalea and 4.7 mm for C. ponder). Each species has 3.5 whorls. The umbilicus of C. hickmaneae is broader than that of C. leptalea and C. ponderi, in which the peristome is slightly reflected over the umbilicus. The fine pitting on the surface of the shell is probably a result of etching due to the original preservation in for- malin. There is no record of association of the type lot with shark or skate egg cases, but the extremely thin shell and damaged condition of all specimens suggest that protection within an elasmobranch egg case would be essential to this species. Etymology: This species is named after Carole S. Hick- man, University of California, Berkeley. Further Records of Chorvstella spp. Four additional lots of Choristella species from the MNHNP collection were received on loan from P. Bouchet subsequent to the initial submission of this paper. All represent immature specimens and I refrain from describ- ing further new taxa from this material because mature examples are unknown. These lots are listed here: MNHN uncataloged, Mozambique Channel (11°44’S, 47°35'E), 3716 m. R/V Suroit, BENTHEDI Expedition, Sat. 87, 3 April 1977. Five specimens, maximum diameter 2.0 mm. Specimens of 1.0 mm in diameter show a basal carination. MNHN uncataloged, Norfolk Ridge (23°03’S, 167°19’E), 503 m. R/V N. O. Vauban, SMIB 3 Expedition, Sta. DW22, 24 May 1987. Two specimens, maximum Page 286 The Veliger, Vol. 35, No. 4 Explanation of Figures 39 to 45 Figures 39-45. Choristella hickmanae McLean, sp. nov. Figures 39-41, 44. Paratype specimen prior to sectioning, Northern Cascadia Abyssal Plain off Washington, LACM 2250. Height 6.5 mm (estimate). Figures 42, 43. Holotype, same locality, LACM 2249. Diameter of broken shell 8.0 mm. Figure 45. SEM view of radula of paratype, LACM 2250. Scale bar = 50 um. diameter 2.0 mm. Shell profile low, not showing basal carination. Radula and jaw examined with SEM, typical for Choristella. MNHN uncataloged, Tanimbar Islands, Indonesia (08°42'S, 131°54’E), 356-368 m. R/V Baruna Jaya 1, KA- RUBAR expedition, Sta. CP69, 2 November 1991. Six specimens, maximum diameter 3.5 mm. Shell profile of medium height; small specimens not showing basal cari- nation. Radula and jaw examined with SEM, typical for Choristella. MNHN uncataloged, Kai Islands, Indonesia (06°08’S, 132°45’E), 390-502 m, R/V Baruna Jaya 1, KARUBAR expedition, Sta. CP35, 27 October 1991. One specimen, maximum diameter 4.0 mm, similar to preceding lot. J. H. McLean, 1992 Bichoristes McLean, gen. nov. Type species: Bichoristes wareni McLean, sp. nov. As the genus is monotypic, the generic diagnosis and remarks are combined in the species treatment below. Bichoristes warent McLean, sp. nov. (Figures 46-53) Description: Shell (Figures 46-48) minute (maximum diameter 3.2 mm), thin, periostracum unknown, whorls 3.2, quadrangular in section, growth form planispiral with two acutely angled, projecting carinations at outer edge, one above and the other below position -of protoconch; upper carina projecting slightly more than lower carina. Suture at position of upper carination of previous whorl, not channeled; whorl extending above suture, forming rounded angulation, base defined at position of broadest possible umbilicus by sharp angulation. Spiral sculpture lacking except for these four carinations. Axial sculpture of exceedingly fine growth increments, prosocline on upper surface of whorl, greatest curvature close to suture, opis- thocline on outer surface between the two keels, prosocline on base, greatest curvature close to umbilical keel. Aperture quadrangular, peristome complete; mature upper lip, outer lip, and lower lip slightly flared. Protoconch (Figure 49) diameter 200 um, tip bulbous, surface smooth, visible equally in spire and umbilical views, similarly recessed in both views. Operculum (Figure 50) thin, multispiral, about 10 whorls visible. Dimensions. Height 1.4 mm, diameter 3.2 mm (holo- type); height 1.2 mm, diameter 3.0 mm (paratype). External anatomy. Unknown. Jaw (Figure 51). Typical for family. Radula. (Figures 52, 53). Rachidian tooth relatively large, with triangular shaft and prominent overhanging cusp; base of shaft with lateral nubs and one central nub. First lateral tooth massive, with two cusps, the innermost small and blunt like that of rachidian, the outermost acutely triangular and with long overhang; shaft base articulating with tooth below, inner edge of shaft articulating with rachidian. Second lateral tooth largest in row, with single large acutely pointed overhanging cusp, its upper profile descending away from rachidian, shaft long and deeply excavated for accommodation of outer lateral teeth. Third lateral tooth with long shaft and pointed cusp that projects over the deeply excavated shaft of second lateral tooth. Fourth lateral tooth small, fused with and emerging from shaft of third. Fifth lateral tooth vestigial, a small flap at the base of the shaft of the third lateral tooth. Type locality: Norfolk Ridge, S of New Caledonia (24°55'S, 162°22’E), 505-515 m. Type material: Holotype MNHN uncataloged, from type locality, R/V Jean Charcot, BIOCAL Expedition Sta. DW66, 3 September 1985. The body of the specimen was Page 287 extracted through a hole filed in the shell by A. Waren, who examined the operculum, radula, and jaw with SEM and provided the prints used here. One paratype, MNHN uncataloged, Norfolk Ridge (23°03’S, 167°19'E), 503 m, R/V N. O Vauban, SMIB 3 Expedition Sta. DW22, 24 May 1987. Remarks: The discovery of a planispirally coiled member of the Choristellidae was unanticipated. Although the shell morphology seems to be completely different from that of other choristellids, there are a number of shared characters: (1) shell is extremely thin; (2) protoconch surface is smooth; (3) teleoconch whorls do not exceed 3.5; (4) contact with the previous whorl is limited to the thin layer that makes the peristome complete where it fuses with the parietal wall, and (5) the umbilicus is as broad as is physically possible, the inner basal keel of Bichoristes corresponding to the sharp umbilical ridge of Chorvstella. Bichoristes adds the outer two keels; these delimit the area of contact for the next whorl, and the result is a planispiral growth form. The radula has the basic choristellid plan, differing from that of Choristella in having the first rather than second lateral tooth the bicuspid tooth. Other distinctions are the nubs at the base of the shaft of the rachidian and the fusion of the fourth lateral tooth with the third. The sculptured shell of Bichoristes has to be interpreted as either derived or primitive in the family. I interpret Bichoristes as derived from the low-spired shell form typ- ified by Choristella ponderi by the not so extreme modifi- cations to the sculpture noted above. Its jaw and radula are so like those of other choristellids that it is difficult to conceive of a differing life mode. Functionally, its qua- drangular shell morphology provides structural support. The narrow planispiral shell form would enable access to the deep crevices at both ends of the elasmobranch egg case, where a planispiral shell could be expected to pen- etrate further than the helically coiled shell form of Choris- tella. Etymology: The name honors Anders Warén, of the Swedish Museum of Natural History, Stockholm, who recognized the familial affinity of the species among ma- terial in the MNHN collection. DISCUSSION Choristellidae and Addisoniidae share many characters of internal anatomy (HASZPRUNAR, 1988c, 1992) and a sim- ilar habitat and feeding specialization on the spent egg cases of elasmobranchs. Members of both families are thin- shelled, affording little protection from predators; instead, protection is provided by the thick walls of the egg cases within which they live. The radula in the Choristellidae and Addisoniidae is relatively large and is provided with robust teeth that are capable of gouging into the walls of the egg cases to provide a direct source of food. Marshall (personal communication) reports that the inner wall of egg cases that contained Page 288 The Veliger, Vol. 35, No. 4 Explanation of Figures 46 to 53 Figures 46-53. Bichoristes wareni McLean, sp. nov. All are SEM views of holotype specimen, MNHN, Norfolk Ridge. Diameter 3.2 mm. Figures 46-48. Apertural, spire, and umbilical views. Figure 49. Protoconch. Scale bar = 100 um. Figure 50. Operculum. Scale bar = 400 um. Figure 51. Jaw. Scale bar = 100 um. Figure 52. Radula. Scale bar = 20 um. Figure 53. Radula. Scale bar = 10 wm. Choristella marshalli were eaten by the limpets. MCLEAN (1985) illustrated radular grazing marks made on the inner wall of an egg case by Addisonia brophy: McLean, 1985. The radula in other cocculiniform families is relatively small with weakly developed teeth in the central field; most of these families also differ in having marginal teeth that are used for sweeping. MARSHALL (1986) has emphasized that the diet in these families is likely the bacteria that are associated with the decomposition of the biogenic sub- strates, rather than the direct food source provided by the substrate. There is no indication that members of either the Choris- tellidae or Addisoniidae occur in the capsules of developing elasmobranchs. The thin-shelled mollusks would be ex- posed to predators during penetration of the egg case. Thus it is incorrect to say that these mollusks are parasitic. Dispersal of these mollusks would necessarily be possible only during the larval stage, at which time the larvae would settle on and enter a spent capsule through the opening from which the young elasmobranch had emerged. Sizes of the egg cases available in the benthos places limits on the maximum size attained by choristellid and addisoniid species. The maximum size of 10 mm in choris- tellids could only be exceeded if the elasmobranch capsule were unusually large. The distribution of each species must depend upon the availability of egg cases of sharks and skates. As noted earlier (MCLEAN, 1985), egg cases are produced by three elasmobranch families, the cat sharks (family Scyliorhyn- idae), with about 85 species in the world, the bullhead or J. H. McLean, 1992 horn sharks (family Heterodontidae) and the skates (fam- ily Rajidae) (ESCHMEYER et al., 1983). Cox (1963) and ESCHMEYER et al. (1983) illustrated the egg cases of the species in these families known from California. If a world- wide study of capsule producing elasmobranchs were avail- able, it would be possible to predict the likelihood of as- sociated species of choristellids and addisoniids. Because too few records are currently known, it is unknown wheth- er choristellid species are host specific. WourRMS (1977) reviewed the literature on elasmo- branch egg-case structure and formation. Egg cases are composed of layers of the structural protein collagen, which exhibits unique chemical and physical properties when deployed in the egg cases. Shark embryos develop within the egg cases for up to nine months, during which there is little evidence of deterioration of the egg cases. The duration of spent egg cases in the benthos is unknown, nor am I aware of their being used as food by other organisms, but the cases undoubtedly persist in the benthos for a number of years. The egg cases should therefore provide a persistent and reliable food source. SUMMARY Additions to knowledge of the Choristellidae that result from this study are: (1) Family-level shell characters are minute to small size, extremely thin shell, complete peristome, deep suture (except Bichoristes), umbilical ridge, smooth bulbous pro- toconch, and maximum of 3.5 teleoconch whorls. (2) The radula is unique to the family. Its resemblance to that of the Cocculinellidae is superficial. (3) The bulbous protoconch tip is unlike the compressed, laterally pinched protoconch tip of Pseudococculinidae and Cocculinellidae. Protoconch characters may yet confirm the affinity to Addisoniidae suggested by anatomical char- acters (HASZPRUNAR, 1992), but have not helped because the protoconch of Addisoniidae is deciduous and remains unknown. (4) Taxa based on shell characters can readily be ex- cluded from the genus if they do not meet all these criteria. In the Appendix, 10 species-level taxa that have previously been assigned to the family are excluded. Skeneiform gen- era with a sharp umbilical ridge may be excluded by lacking the deep suture. (5) Specific characters in Choristella are relative size, relative proportions of height to width, and whether the multispiral operculum looks multispiral or appears to be paucispiral as a result of having only three whorls. Ex- ternal anatomy is too poorly known to be useful at this time. (6) The new genus Bichoristes has a uniquely bicarinate and planispiral shell, although the radula is close to that of Choristella. Most of the shell characters diagnostic for Choristella, including thin shell, 3.5 whorls, smooth pro- toconch, and umbilical ridge, are present. Page 289 (7) Although most species are allopatric, one sympatric pair is known: Choristella tenera and C. leptalea. (8) No species is known to be free living and unasso- ciated with the spent egg cases of elasmobranchs. (9) Shell size is limited by the size of available egg capsules. (10) The family is broadly distributed, having been found in the most extensively sampled regions of the world in temperate zones at continental shelf to abyssal depths. ACKNOWLEDGMENTS Treatment of new species included here would not have been possible had not Philippe Bouchet of the Museum d’ Histoire Naturelle, Paris, Bruce Marshall of the Na- tional Museum of New Zealand, and Winston Ponder of the Australian Museum, Sydney, turned over to me the undescribed choristellid species they had recognized in their museum collections, as well as some scanning work that had been done. SEM micrographs of the radula of Choris- tella hickmanae were provided by Carole Hickman. All illustrations for Bichoristes wareni were provided by An- ders Waren, who also provided two views used for Choris- tella nofronu. Bruce Marshall provided the radular illus- trations for Trenchia agulhasae used in the Appendix. I thank staff members of the Natural History Museum, London, and the U.S. National Museum of Natural His- tory for the loan of pertinent type specimens. I thank Philippe Bouchet, Gerhard Haszprunar, Bruce Marshall, and Anders Waren, and the anonymous reviewers for read- ing the manuscript and offering helpful suggestions. LITERATURE CITED ABBOTT, R. T. 1974. American Seashells. 2nd ed. Van Nos- trand Reinhold: New York. 663 pp., 24 pls, text figs. Boss, K. J. 1982. Mollusca. Pp. 945-1166. In: S. P. Parker (ed.), Synopsis and Classification of Living Organisms. Vol. 1. McGraw-Hill: New York. BOUCHET, P., & A. WAREN. 1979. The abyssal molluscan fauna of the Norwegian Sea and its relation to other faunas. Sarsia 64:211-243. BusH, 1897. Revision of the marine gastropods referred to Cyclostrema, Adeorbis, Vitrinella, and related genera; with descriptions of some new genera and species belonging to the Atlantic fauna of America. Transactions of the Con- necticut Academy of Arts and Sciences 10:97-144, 672. Carey, A. G., JR. 1981. A comparison of benthic infaunal abundance on two abyssal plains in the northeast Pacific Ocean. Deep-Sea Research 28A:467-479. CLARKE, A. H., JR. 1961. Abyssal mollusks from the South Atlantic Ocean. Bulletin of the Museum of Comparative Zoology, Harvard University 125(12):343-387, pls. 1-4. Cox, K. W. 1963. Egg cases of some elasmobranchs and a cyclostome from Californian waters. California Fish and Game 49(4):272-289. DALL, W. H. 1889. Reports on the results of dredging, under the supervision of Alaxander Agassiz, in the Gulf of Mexico (1877-78) and in the Caribbean Sea (1879-80), byt he U.S. Coast Survey steamer “Blake,” Lieut. Commander C. D. Sigsbee, U.S.N., and Commander J. R. Bartlett, U.S.N., Page 290 commanding. XXIX. Report on the Mollusca, Part 2, Gas- tropoda and Scaphopoda. Bulletin of the Museum of Com- parative Zoology, Harvard University 18:1-492, pls. 10-40. DALL, W. H. 1896. Diagnoses of new species of mollusks from the west coast of America. Proceedings of the United States National Museum 18(1034):7-20. Dau, W. H. 1908. Reports on the dredging operations off the west coast of Central America to the Galapagos, to the west coast of Mexico, and in the Gulf of California. ... XIV. The Mollusca and Brachiopoda. Bulletin of the Museum of Com- parative Zoology, Harvard University 43(6):205-487, pls. 1-22. Dawson, J. W. 1872. The post-Pliocene geology of Canada. Canadian Naturalist and Quarterly Journal of Science, New Series, 6(3):369-416, pls. 4-7. ESCHMEYER, W.N., E. S. HERALD & J. HAMMANN. 1983. A Field Guide to Pacific Coast Fishes of North America. Pe- terson Field Guide Series. Houghton Mifflin Company: Bos- ton. 336 pp. Go.ikov, A. N. & Y. I. STAROBOGATOV. 1975. Systematics of prosobranch gastropods. Malacologia 15(1):185-232. GUBBIOLI, F. & I. NOFRONI. 1986. First record of ‘“‘Cithna”’ naticiformis Jeffreys, 1883 in the Mediterranean: ecological and systematic notes. La Conchiglia 18:204-205. HASZPRUNAR, G. 1987. The anatomy of Addisonia (Mollusca: Gastropoda). Zoomorphology 106:269-278. HASZPRUNAR, G. 1988a. Anatomy and relationships of the bone-feeding limpets, Cocculinella minutissima (Smith) and Osteopelta mirabilis Marshall (Mollusca, Archaeogastropo- da). Journal of Molluscan Studies 54:1-20. HASZPRUNAR, G. 1988b. Anatomy and affinities of pseudococ- culinid limpets (Mollusca, Archaeogastropoda). Zoologica Scripta 17(2):161-179. HASZPRUNAR, G. 1988c. Comparative anatomy of cocculini- form gastropods and its bearing on archaeogastropod sys- tematics. Jn: Prosobranch Phylogeny. Proceedings of a sym- posium held at the 9th International Malacological Congress, Edinburgh, 1986. Malacological Review, Supplement 4:64- 84. HaSZPRUNAR, G. 1992. On the anatomy and relationships of the Choristellidae. The Veliger 35(4):295-307. HickMaAN, C.S. 1983. Radular patterns, systematics, diversity, and ecology of deep-sea limpets. The Veliger 26(2):73-92. Hickman, C. S. & J. H. McLean. 1990. Systematic revision and suprageneric classification of trochacean gastropods. Sci- ence Series No. 35:vi + 169 pp. Hicks, G. R. F. 1986. Revised keys to Paramphiascopsis Lang (Copepoda, Harpacticoida, Diosaccidae) including a new species from deep water off New Zealand. Journal of Nat- ural History 20:389-397. JEFFREYS, J. G. 1883. On the Mollusca procured during the Lightning and Porcupine Expeditions, 1868-70. Part VI. Proceedings of the Zoological Society of London [1883]:88- slaty, KaBAT, A. R. 1989. Choristidae Verrill, 1882 (Mollusca, Gas- tropoda) and Choristidae Esben-Petersen, 1915 (Insecta, Mecoptera): a proposal to remove the homonymy. Bulletin of Zoological Nomenclature 46(3):156-160. KABAT, A. R. 1991. The classification of the Naticidae (Mol- lusca: Gastropoda): review and analysis of the supraspecific taxa. Bulletin of the Museum of Comparative Zoology 152(7): 417-449. KEEN, A. M. 1971. Sea Shells of Tropical West America. 2nd ed. Stanford University Press: Stanford, California. xiv + 1064 pp., 22 pls. The Veliger, Vol. 35, No. 4 KNUDSEN, J. 1964. Scaphopoda and Gastropoda from depths exceeding 6000 meters. Galathea Report 7:125-136. Kuropa, T., T. HABE & K. OyAMA. 1971. The Sea Shells of Sagami Bay. Tokyo. 741 pp. [in Japanese], 489 pp. [in English], index, pp. 1-51, pls. 1-12. MaRINCOVICH, L., JR. 1975. New Tertiary and Recent Na- ticidae from the eastern Pacific. The Veliger 18(2):168-173, figs. 1-22. MaRINCOVICH, L., JR. 1977. Cenozoic Naticidae (Mollusca: Gastropoda) of the northeastern Pacific. Bulletins of Amer- ican Paleontology 70(249):165-494, pls. 17-42, 13 text figs. MaRSHALL, B. A. 1983. The family Cocculinellidae (Mollusca: Gastropoda) in New Zealand. National Museum of New Zealand, Records 2(2):139-143. MARSHALL, B. A. 1986. Recent and Tertiary Cocculinidae and Pseudococculinidae (Molluca: Gastropoda) from New Zea- land and New South Wales. New Zealand Journal of Zo- ology 12:505-546. MarRSHALL, B. A. 1988. Skeneidae, Vitrinellidae and Orbites- tellidae (Mollusca: Gastropoda) associated with biogenic substrata from bathyal depths off New Zealand and New South Wales. Journal of Natural History 22:949-1004. McLEan, J. H. 1985. The archaeogastropod family Addison- iidae Dall, 1881: life habit and review of species. The Veliger 28:99-108. MOosKALEV, L. I. 1971. New data on the systematic position of the gastropod mollusks of the order Cocculinida Thiele, 1908. Abstracts of Reports, Fourth Conference on the Investigation of Mollusks, Academy of Sciences, U.S.S.R. Nauka, Len- ingrad, pp. 59-60 [in Russian]. OxkuTANI, T. 1964. Report on the archibenthal and abyssal gastropod Mollusca mainly collected from Sagami Bay and adjacent waters by the R.V. Soyo-Maru during the years 1955-1963. Tokyo University, Faculty of Sciences, Journal, Sec. II, 15(3):374-447, pls. 1-7, text figs. 1-7. RICHARDS, H.G. 1962. Studies on the marine Pleistocene. Part I. The marine Pleistocene of the Americas and Europe. Part II. The marine Pleistocene mollusks of eastern North Amer- ica. Transactions of the American Philosophical Society, New Series 52(3):1-141, pls. 1-21. TayLor, D. W. & N. F. SOHL. 1962. An outline of gastropod classification. Malacologia 1(1):7-32. THIELE, J. 1929-1931. Handbuch der systematischen Weich- tierkunde. Stuttgart. Vol. 1:1-778, figs. 1-782. Turner, R.D. 1978. Wood, mollusks, and deep-sea food chains. Bulletin of the American Malacological Union, Inc. 1977: 13-19. VERRILL, A. E. 1882. Catalogue of marine Mollusca added to the fauna of the New England region, during the past ten years. Transactions of the Connecticut Academy of Arts and Sciences 5:447-587, pls. 45-60. VERRILL, A. E. 1884. Second catalogue of Mollusca, recently added to the fauna of the New England coast and the adjacent parts of the Atlantic, consisting mostly of deep sea species, with notes on others previously recorded. Transactions of the Connecticut Academy of Arts and Sciences 6(1):139- 294, pls. 28-32. WarEN, A. 1980. Marine Mollusca described by John Gwynn Jeffreys, with the location of the type material. Conchological Society of Great Britain and Ireland, Special Publication 1: 1-60, pls. 1-8. WaREN, A. 1991. New and little known Mollusca from Iceland and Scandinavia. Sarsia 76:53-124. Waren, A. 1992. New and little known “Skeneimorph” gas- J. H. McLean, 1992 tropods from the Mediterranean Sea and the adjacent At- lantic Ocean. Bollettino Malacologico 27:149-247. WENZ, W. 1938. Gastropoda. Handbuch der Palaozoologie. Vol. 6, Teil 1: Allgemeiner Teil und Prosobranchia. Berlin. 1639 pp. WourMs, J. P. 1977. Reproduction and development in chon- drichthyan fishes. American Zoologist 17:379-410. APPENDIX—TAXA EXCLUDED FROM CHORISTELLIDAE A number of taxa have been incorrectly allocated to the family Choristellidae (originally assigned to the “Choristi- dae’). The following taxa have not been shown to have the choristellid radular plan and lack some or all of the diagnostic shell characters (small size, extremely thin shell, complete peristome, deep suture, and sharp carination de- scending into the umbilicus). Some of the misallocated taxa are naticids, but many are potential members of the family Skeneidae (superfamily Trochacea). Skeneidae and fam- ilies of similar appearance have been poorly understood but have received recent attention from MARSHALL (1988), HICKMAN & MCLEAN (1990), and WAREN (1991, 1992). It is beyond the scope of this paper to allocate the following taxa, although some suggestions are made. 1. Choristes elegans Carpenter, 1872 (Figure 54) Choristes elegans Carpenter 7n DAWSON, 1872:392, pl. 7, figs. 13, 13a; VERRILL, 1882:542, p. 58, fig. 28 [“I have figured a young fossil specimen for convenient compar- ison’’]; RICHARDS, 1962:79, pl. 17, fig. 15 [““Montreal, Pleistocene’’]; CLARKE, 1961:360 [in list of species under Choristes]; MARINCOVICH, 1977:338 [as valid genus and species of Naticidae]; BOUCHET & WaREN, 1979, fig. 47 [syntype]. Type locality: Pleistocene, St. Lawrence River Estuary, Montreal, Quebec. Lectotype (here designated): USNM 188948; 2 paralectotypes: USNM 56385. Carpenter was uncertain as to the familial relationships of the Pleistocene fossil he described as Choristes elegans: “It is hard to pronounce satisfactorily on its relationships. In its thin, coated shell it resembles Velutina; the striae and loose whirls recall Naticina; the straight pillar lip reminds us of Fossarus; while the umbilicus and rounded base, with entire mouth, best accord with the Natica group.” Although MARINCOVICH (1977) used Choristes for east- ern Pacific naticid species, he cited only the original il- lustration of the type species; it is not clear whether he examined specimens. He did not cite the illustration of RICHARDS (1962:79, pl. 17, fig. 15), who figured a spec- imen identified as Choristes elegans from the Montreal Pleistocene and placed it in Choristidae without comment. BoucHET & WAREN (1979) figured a syntype without citing a catalog number. There are three shells in the USNM collection labeled Choristes elegans Carpenter, “Postpliocene, Montreal, Dawson.” A lectotype (USNM 188948, height 20.1 mm, Figure 54) and two paralecto- types (USNM 56385, heights 16.7 mm and 17.2 mm) are Page 291 here designated. The lectotype (Figure 54) shows irreg- ular spiral sculpture and the inner lip detached from the parietal wall. Carpenter noted the “‘smooth epidermis lin- ing the umbilical chambers, conspicuously preserved, even in these fossil specimens, between the closest part of the parietal region.” The type material of Choristes elegans is here identified as a variation of the naticid Amauropsis islandica (Gmelin, 1791), a morphologically variable species that is broadly distributed in shallow to moderate depths in the North Atlantic and Arctic Oceans. In his rediscription of A. 1s- landica, MARINCOVICH (1977:217, pl. 17, figs. 1-4, pl. 22, fig. 1) described the umbilicus as “open, extremely narrow and slitlike, usually concealed by periostracum of inner lip margin.” This description of the periostracum agrees with that of Carpenter for Choristes elegans. The three specimens have broader umbilici than most specimens of A. islandica, but such a shell form may possibly correlate with the lowered salinity in the estuary of the St. Lawrence River. MaRINCOVICH (1977:338) was the first subsequent au- thor to correctly assign Choristes to the Naticidae. GOLOKOV & STAROBOGATOV’s (1975) assignment to the Naticidae on radular characters cannot be credited because it could only have been based on published illustrations of the radula of “‘Choristes” tenera (which is not a naticid). However, Marincovich did not note the fact that Choristes elegans would have to be considered extinct if recognized as a valid species; he did not compare it to Amauropsis islandica and he did not conclude that Chorvstes tenera Verrill should be assigned elsewhere, despite noting that the radular den- tition of that species differed from that of two Recent naticids he assigned to Choristes. Instead, he stated that “another radular mount should be made to confirm the radular dentition reported by VERRILL (1882).” KaBaT (1989, 1991) was aware that Choristellidae Bouchet & Warén, 1979, solved the nomenclatural prob- lem for the choristellids, but pointed out that Choristidae Verrill, 1882 (Naticacea) presented a problem of hom- onymy for the well-known insect family Choristidae Es- ben-Petersen, 1915 (nominotypical genus Chorista Klug, 1836). Kabat proposed that Choristidae Verrill be emend- ed to Choristeidae Verrill, to conserve Choristidae Esben- Petersen and to retain Choristeidae in the event that it might prove to have utility in the Naticacea. Kabat (per- sonal communication) now agrees with the synonymization of Choristes with Amauropsis. 2. Choristes carpentert Dall, 1896 (Figures 55, 56) Choristes carpentert DALL, 1896:10; DALL, 1908:328, pl. 3, fig. 4; KEEN, 1971:388, fig. 424 [copy fig. of DALL, 1908]; CLARKE, 1961:360 [in list of Choristes species]; MARIN- COVICH, 1977:340, pl. 31, figs. 8, 9, text fig. 11b [radula]. Type locality: Gulf of Panama, 2693 m. Holotype: USNM 123039. Page 292 The Veliger, Vol. 35, No. 4 Explanation of Figures 54 to 63 ee, Figures 54-63. Type specimens of taxa incorrectly referred to Choristellidae. Name combinations as originally proposed. Figure 54. Lectotype, Choristes elegans Carpenter, 1862. USNM--56385. Pleistocene, St. Lawrence River Estuary, Montreal, Quebec. Height 20.1 mm. Figure 55. Holotype, Choristes carpenteri Dall, 1896. USNM 123039. Gulf of Panama, 2693 m. Height 21.0 mm. Figure 56. Choristes carpenteri, second reported specimen, USNM 123038. Gulf of Panama, 2690 m. Figure 57. Holotype, Cyclostrema pompholyx Dall, 1889. USNM 214279. Gulf of Mexico, 1472 m. Diameter 4.1 mm. Figure 58. Holotype, Choristes agulhasae Clarke, 1961. MCZ 224955. Cape Basin off South Africa, 4585 m. Height 2.0 mm. Figure 59. Holotype, Choristes agulhasae argentinae Clarke, 1961. Argentine Basin, 5130 m. MCZ 224956. Height 2.3 mm. Figures 60, 61. Syntype, Czthna naticiformis Jeffreys, 1883. BMNH 85.11.5.1615-1617. Porcupine Expedition of 1870, Sta. 17a, off Portugal, 1353 m. Height 1.8 mm. Specimen is still attached to cardboard mount. Figures 62, 63. SEM views of radula from holotype of Choristes agulhasae Clarke, 1961, courtesy B. Marshall. See text for generic assignment to Trenchia Knudsen, 1964. Scale bar of Figure 62 = 100 um, of Figure 63 = 40 um. Despite the fact that Choristes was based on a shallow- marginal teeth per half row”). Now that Choristes is rel- water type species, MARINCOVICH (1977) retained the ge- egated to the synonymy of Amauropsis, these two species nus for two abyssal, eastern Pacific naticid species, Choris- are in need of generic reassignment in Naticidae. Affinity tes carpentert Dall, 1896, and C. coani Marincovich, 1975, to Amauropsis is ruled out, as its type species has a tri- invoking a unique radular definition (“monocuspate ra- cuspate rachidian tooth. chidian, one monocuspate lateral, and two monocuspate The holotype of Choristes carpenterr (USNM 123039, J. H. McLean, 1992 from USFC Sta. 3382), is 21 mm in height, which is sufficiently large to remove it from consideration as a mem- ber of the Choristellidae. No mention of an operculum was made in the original account and the specimen appears not to have been collected alive. This specimen has the apical area worn. It is illustrated here for the first time (Figure 55). In his subsequent account DALL (1908) mentioned a second specimen, from USFC Sta. 3361, 2690 m, Gulf of Panama. This must have been the specimen to which he referred in reporting that “the animal agrees in general appearance with that of Choristes elegans var. tenera Ver- rill, as described by Verrill.” This specimen, USNM 123038, from USFC Sta. 3361 (Figure 56) is also marked “type”; it measures 15.0 mm in length. It exhibits a char- acteristic sculptural pattern of naticids in having collabral ridges on the upper part of the whorl near the suture. This specimen has the operculum and a dried body, but the body does not have the epipodial tentacles that may be seen on the dried bodies of Choristella tenera. Clearly and inexplicably Dall erred in reporting that the animal agreed with Verrill’s species. Although MARINCOVICH purported to figure the holotype (1977:fig. 8), he actually figured this second specimen mentioned by Dall (USNM 123038), and incorrectly gave the length at 20.5 mm, rather than 15.0 mm. 3. Choristes coani Marincovich, 1975 Choristes coani MARINCOVICH, 1975:169, figs. 2, 6, 7; MARIN- COVICH, 1977:341, pl. 31, figs. 10-12, text fig. 11c [rad- ula]. Type locality: off Central Oregon, 2830 m. Ho- lotype: USNM 741014. This was described by Marincovich in the family Na- ticidae. Like the preceding species, it is in need of generic reassignment. 4. Cyclostrema pompholyx Dall, 1889 (Figure 57) Cyclostrema pompholyx DALL, 1889:394, pl. 28, fig. 9; BUSH, 1897:139; TURNER, 1978:17, figs. 11, 12. Type locality: Gulf of Mexico, 1472 m. Holotype: USNM 214279. Choristes pompholyx: CLARKE, 1961:360 [in list of Choristes species]. DALL (1889) originally stated: “I am in doubt as to the generic place of this species, so simple in its characters and without the soft parts. I had thought of putting it under Choristes or with Vitrinella, and finally in placing it here [Cyclostrema] feel by no means satisfied that the choice is a correct one.”” BUSH (1897) noted that Dall’s species ‘““may prove to be another species of Choristes,” accounting for CLARKE (1961) having placed it in Chorvstes. The shell is sturdy with a broadly inflated lip. It lacks the umbilical ridge of Choristella. There is no evidence to support the allocation of this species to the family Choris- tellidae. Page 293 5. Choristes agulhasae Clarke, 1961 (Figure 58) Choristes agulhasae CLARKE, 1961:361, pl. 3, fig. 1. Type locality: Cape Basin (corrected from Agulhas Basin), SW of Cape Town, South Africa, 4585 m. Holotype: MCZ 224955. No evidence was advanced to support the assignment of this species to the family, although there is an umbilical carination similar to that of Choristella. The radula (Fig- ures 62, 63, SEM photos by Bruce Marshall) is rhipi- doglossate, unlike that of Choristellidae. Marshall (per- sonal communication) has identified it as that of Trenchia Knudsen, 1964 (family Skeneidae), characterized by the elongate and laterally excavated base of the first lateral tooth. 6. Choristes agulhasae argentinae Clarke, 1961 (Figure 59) Choristes agulhasae argentinae CLARKE, 1961:361, pl. 3, figs. 2, 3. Type locality: Argentine Basin, ESE of Buenos Aires, Argentina, 5130 m. Holotype: MCZ 224956. The holotype (Figure 59) is a single empty shell, larger than that of the preceding taxon. No evidence supports assignment to Choristellidae. It may be regarded as a pos- sible member of 77renchia. 7. Cithna naticiformis Jeffreys, 1883 (Figures 60, 61) Cithna naticiformis JEFFREYS, 1883:112, pl. 20, fig. 11; WAREN, 1980:21; GUBBIOLI & NOFRONI, 1986:204, figs. Type locality: Porcupine Expedition of 1870, Sta. 17a, off Cape Mondego, Portugal, 1353 m. Syntypes: 3 shells, BMNH 85.11.5.1615-1617. GUBBIOLI & NOFRONI (1986) incorrectly used this name for Choristella nofroni described here, as detailed in the remarks that follow the new species description. A syntype specimen is illustrated here (Figures 60, 61). Although radular material is not available, it is also a possible species of Trenchia on evidence from shell characters. 8. Cyclostrema valvatoides Jeffreys, 1883 Cyclostrema valvatoides JEFFREYS, 1883:92; WAREN, 1980: 19: GUBBIOLI & NOFRONI, 1986:205. Type locality: Por- cupine Expedition, 1870, Sta. 17a, off Cape Mondego, Portugal, 1353 m. Holotype: BMNH 85.11.5.1593. GUBBIOLI & NOFRONI (1986) also suggested that this name might apply to Choristes nofronu described here. I have examined the holotype, which is in bad condition due to chemical exfoliation; it lacks the umbilical carination of Choristella. Page 294 9. Choristes mollis Okutani, 1964 Choristes mollis OKUTANI, 1964:389. Type locality: off Mi- yake Island, Japan, 1230-1350 m. No evidence supported the original placement in Choris- tes. The granular sculpture, and of most importance, the incomplete peristome as illustrated by Okutani are not characters of the family. The operculum is figured as mul- tispiral with more whorls than in species of Choristella. Reassignment may be possible if the radula is intact in the holotype. Marshall (personal communication) suggests that it be compared to Granigyra Dall, 1889 (family Skeneidae). The Veliger, Vol. 35, No. 4 10. Choristes nipponica Okutani, 1964 Choristes nipponica OKUTANI, 1964:388, pl. 6, fig. 2. Type locality; Sagami Bay, Japan, 1360-1385 m. No evidence was given to support the assignment of this taxon to Choristes. The “shining shell,” sutural shelf rather than channeled suture, produced basal lip, and incomplete peristome are not characters of Choristella. The operculum is illustrated as multispiral. Assignment on radular char- acters may be possible. The Veliger 35(4):295-307 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 On the Anatomy and Relationships of the Choristellidae (Archaeogastropoda: Lepetelloidea) GERHARD HASZPRUNAR Institut fir Zoologie der Leopold-Franzens-Universitat, Technikerstrasse 25, A-6020 Innsbruck, Austria Abstract. The anatomy of three species of the Choristellidae, Choristella marshalli, C. hickmanae, and C. nofronu, which are described by MCLEAN (1992) in the adjoining paper, is described on the basis of reconstructions of serial sections. Choristellids are unique among the lepetelloid Cocculiniformia in having a coiled teleoconch and an operculum as adults. Nevertheless, choristellids can be classified among the Lepetelloidea on anatomical characters. In particular they share the type of gill and the food source (empty egg cases of chondrichthyans) with the addisoniid limpets, although the modifications of the alimentary tract of Choristellidae are unique. Cocculinellidae and Addisoniidae are regarded as the closest relatives of the Choristellidae. Support is given for the hypothesis that teleoconch coiling and gonochorism of the Choristellidae are derived conditions in the Lepetelloidea. INTRODUCTION The Choristellidae have been classified in various positions within the Gastropoda. VERRILL (1882:540) placed them among Tectibranchia because of their aberrant “large gill consisting of numerous thin lamellae, attached to the inner surface of the mantle, over the left side of the neck, and extending obliquely across and over the right side.” In contrast, truncatelloid (= rissoacean) affinities were as- sumed by BusH (1897:139), THIELE (1929), KURODA et al. (1971), and Boss (1982). The family is based on the genus Choristella Bush, 1897 (Choristes Verrill, 1882, non Carpenter, 1872; see BOUCHET & WAREN, 1979:225, and McLEan, 1992). MOSKALEV (1978) and HICKMAN (1983) noted the sim- ilarity of the choristellid radula with the radula of Coc- culinella Thiele, 1909. In the light of preliminary obser- vations on choristellid anatomy, HASZPRUNAR (1988c) strengthened this proposed relationship and classified the family among the cocculiniform Lepetelloidea closest to the Addisoniidae and Cocculinellidae. This view has been accepted by PONDER & WAREN (1988:310). Meanwhile, the anatomy of several lepetelloid families has been out- lined in detail (HASZPRUNAR, 1987a, b, 1988a, b, c; McLEAN & HASZPRUNAR, 1987). Together with the de- tailed description of choristellid anatomy, which is pre- sented here, the previous, preliminary conclusions will be substantiated or slightly modified based on the new data. VERRILL (1882, 1884) first noted the peculiar feeding biology of Choristella. As far as is known all species that have been found on a substrate live and feed in the empty egg cases of sharks (e.g., Scylorhinus) or skates (e.g., Raja) (see, e.g., GUBBIOLI & NOFRONI, 1986). This mode of nourishment is shared solely with the lepetelloid limpet genus Addisonia (VILLA, 1985; RAGOZZI, 1985; MCLEAN, 1985; HASZPRUNAR, 1987a). It will be shown that Addi- soniidae and Choristellidae share not only a similar feeding biology, but also are closely related, despite the differences in their external morphology. This raises questions re- garding the primitiveness of limpets such as Addisoniidae and all other cocculiniform families and coiled forms like the Choristellidae among the Cocculiniformia. In the accompanying paper MCLEAN (1992) provides a complete revision of the family including descriptions of certain new choristellid species and exclusions of taxa pre- viously assigned to the family. Photographs of adult shells and SEM views of protoconchs, opercula, radulae, and jaws are presented there. Anatomical characters of certain species described by MCLEAN (1992) are presented herein. MATERIALS ann METHODS Few specimens of each species were available for anatom- ical investigations and the preservation was poor in most cases. In addition, the content of the alimentary tract (see below) and an excessive amount of debris in the mantle cavity caused ruptures and folds in the sections. Never- theless, I was able to reconstruct at least the gross anatomy Page 296 ~L o- Figure 1 Choristella hickmanae, mantle cavity and gonopericardial system in dorsal view. Anatomy of the heart could not be determined in this species. Abbreviations: es, efferent gill sinus; gg, genital gland; 1, intestine; Ik, left kidney; mc, posterior end of mantle cavity; 0, ovary; od, oviduct; pc, pericardium; r, rectum; rk, right kidney; rs, receptaculum seminis. Scale bar = 1 mm. of the specimens, although uncertainties remain with re- spect to details of the nervous and circulatory system and of histology in general. Abbreviations: LACM—Los Angeles County Museum of Natural History; NUNZ—National Museum of New Zealand, Wellington. Choristella marshalli McLean, 1992: Four males from the original type lot (NMNZ M.75210) have been inves- tigated. The three adult specimens were embedded in para- plast and the 5-um serial sections (one series transverse, one oblique, one sagittal) were stained by Heidenhain’s Azan method. A juvenile male was embedded in plastic (araldite) and the 2-um serial sections were stained by Regaud’s fluid. According to Bruce A. Marshall (NMNZ; The Veliger, Vol. 35, No. 4 personal communication) the specimens were frozen before preservation in alcohol; therefore, the preservation of ex- ternal epithelia was poor. Choristella hickmanae McLean, 1992: A series of sections of a single female was available for anatomical investi- gation. According to McLean (personal communication) the live-collected specimen had been preserved in alcohol, resulting in a good preservation of all tissues. Thus, the histological description is mainly based on this specimen. The nearly transverse, 15-wm-thick serial sections were stained by haematoxylin and eosin. Unfortunately, debris in the left side of the mantle cavity caused many folds in the sections and destroyed most of the cardiac region. Choristella nofrona McLean, 1992: This species was treated by GUBBIOLI & NOFRONI (1986) under the name “Cithna naticiformis Jeffreys, 1883,” but the taxonomy was rectified by MCLEAN (1992). A single juvenile female was available for serial sectioning. After being embedded in paraplast the specimen was sectioned in nearly transversal 5-um serial sections, which were stained by haematoxylin and eosin. It is probable that this specimen died before fixation, because the histological preservation of all tissues was poor. Serial reconstruction was done by means of an ocular micrometer and millimeter paper (by ““hand’’) as described in detail by HASZPRUNAR (1987a). RESULTS External Features The head has a stout snout, and the oral lappets are weakly developed. Whereas the epithelium of the head consists of columnar cells, the region of the mouth opening is formed by tall columnar and cuticularized (true cuticula) cells (Figure 4). Cephalic lappets are lacking; the tentacles are devoid of papillae. The mantle margin forms a simple fold lacking special glands or tentacles. Two thin blood sinuses and a mantle nerve supply the mantle border. The epithelium of the pedal sole consists of tall, colum- nar, densely ciliated cells with elongated nuclei inter- spersed by very few mucous cells. Anteriorly a single, large pedal gland opens via a wide pore, which marks the border between the pro- and mesopodium. Laterally and poste- riorly several smooth epipodial tentacles are present. The posterior dorsal surface of the foot bears an operculum (see also the accompanying paper of MCLEAN, 1992). Many thin blood sinuses are found within the pedal musculature. In addition, a cell type containing many brownish droplets, the so-called “pore-cells,” are interspersed (Figure 7). Similar cells are also present in the wall of the head and many occur in the mantle margin. The single attachment zone and innervation (see HASZPRUNAR, 1985) indicate that there is a single (left) shell muscle, which has its origin on the columella of the shell. G. Haszprunar, 1992 [Buc il ae aa w ill (leet id Mat — > ——— ae es °° a Bom Soa Bor =<2.— $e 20 = 2? aoa Roo . e Figure 2 Choristella marshalli, sagittal section of a gill leaflet, anterior view, slightly schematized. A, B, C, D: zones of gill leaflet (see text). Scale bar = 100 um. Page 297 See te azz, >, at al 2S ae li ca Nedetteny A A pec ve : f 2 lamar : \ u a 2 ia ee | Figure 3 Choristella hickmanae, alimentary tract in dorsal view. Abbrevi- ations: cal, anterior radular cartilage; ca2, posterior radular car- tilage; i, intestine; oe, posterior oesophagus; oep, oesophageal pouch; r, rectum; ra, radula; s1, left cul-de-sac; s2, right cul-de- sac; st, stomach with gastric shield. Scale bar = 1 mm. Mantle Cavity The mantle cavity is moderately deep. Numerous gill leaflets occupy most of the mantle roof, filling the space within the mantle cavity. Additional organs of the pallial roof include the single (left) osphradium to the very left, the anterior portion of the heart (to the left), and the left kidney (ventral). A branch of the right kidney is situated in the center, and a distinct portion of the intestinal sac lies dorsally. To the right of the mantle roof is the distal Page 298 The Veliger, Vol. 35, No. 4 Explanation of Figures 4 to 8 Figures 4-8. Choristella hickmanae, histological details. Figure 4. Jaws (j) and lips (1). Scale bar = 100 um. Figure 5. Radular cartilage (ca) and radular sheath (rs), and anterior oesophagus with asymmetrical oesophageal pouches (oep). Scale bar = 200 um. Figure 6. Oesophageal gland. 1, 2, 3 mark the various cell types (see text). Scale bar = 100 um. Figure 7. Pore cells (po) between muscle fibers and connective tissue (amorphous) in the foot mass. Scale bar = 20 um. Figure 8. Rectum (r) with typhlosole (t) near anus, underlain by hypobranchial(?) gland (h). Scale bar = 100 um. G. Haszprunar, 1992 portion of the rectum, which is underlain by a (hypo- branchial?) gland (Figure 8), and the genital opening, which is also surrounded by a gland (Figure 1). As already noted by VERRILL (1882) the choristellid gill is pectinibranch and consists of numerous leaflets that have a highly characteristic structure (Figure 2): the efferent axis is stiffened by a pair of prominent, elongate skeletal elements, which emerge from the single skeletal rod of the gill axis. Distally the skeletal rods enclose a thin nerve that extends along to the efferent axis of the leaflet. More proximally they enclose the prominent efferent blood sinus. The epithelium at the efferent edge is cuboidal and sparsely ciliated. The following zone (A) is characterized by a squa- mous epithelium, which is still underlain by skeletal ma- terial. This is followed by a short region (B) forming a distinct band of columnar (up to 25 wm), densely ciliated cells with basally situated nuclei. The subsequent zone (C) is characterized by a non-ciliated epithelium, which ap- pears folded because of different cell heights. Near the afferent axis the epithelium of the gill leaflet becomes abruptly much taller (up to 60 wm) and glandular (D). Two types of mucus cells are present, a rare one with a brightly stained amorphous interior, and one that is much more common and shows numerous dark-red stained gran- ules that have concentric rings. The glandular zone of the leaflets occupies the entire dorsal mantle roof. The two juvenile specimens investigated lack the glandular zone of the gill leaflets. To the right, the histology of the glandular zone of the gill changes successively and continuously in that only the bright mucus cells are present. This type of epithelium primarily underlies the rectum. Heart, Circulatory and Excretory Systems The pericardium is situated at the posterior end of the mantle cavity to the left. It is rather large with a single auricle situated anteriorly and to the left, and a ventricle which lies laterally to the left of the proximal portion of the rectum. After passing through one of the two kidneys, the blood is oxygenated either in the gill leaflets or (to a lesser extent) in the mantle edge. Both streams of blood, the ctenidial and the pallial ones, join at the entrance to the auricle. More peripheral pathways of the blood could not be reconstructed. Two kidneys are present (Figure 1). The left one is large and is situated in the pallial roof. It has a simple opening immediately to the left of the anus. A broad, ciliated renopericardial duct is present. The opening of the right kidney is situated immediately to the right of the anus. The right kidney is composed of two large, elongated cavities. One occupies the mantle roof and extends to the left side; the other one extends posteriorly and surrounds the gonoduct (Figure 11). No connection with the peri- cardium could be found, and the histology of the renal epithelium differs considerably from that of the left kidney. Page 299 Genital System There is no indication that choristellid species are her- maphroditic. All observed specimens of Choristella mar- shall: were males, whereas the single specimens of C. hick- manae and C. nofronu were females. The ovary forms a compact mass and occupies the left dorsal portion of the visceral mass of the most apical whorl of the animal. It contains numerous yolk-rich eggs, which are provided with a distinct outer membrane. Eggs in all stages of development are present. There is no trace of any accessory gland for the formation of additional nutrients. The oviduct is a simple ciliated tube of somewhat irregular shape. It starts as a very wide tube, then narrows abruptly and extends forward, being situated at the very right side of the animal and adjacent to or embedded in the posterior lobe of the right kidney (Figure 11). The genital opening is located at the right posterior end of the mantle cavity. A large receptaculum (Figure 1: 7c) is present, the anterior portion of which occupies the posterior pallial roof. The receptaculum is surrounded by a prominent muscular layer and is filled with sperm. The sperm cells are irregular in the center of the vesicle, whereas their heads are more or less oriented outwards near the wall (Figure 12). The receptaculum is provided with a short, narrow, muscular duct, which opens into the mantle cavity. There is no visible connection with the remaining genital system. The region anterior to the genital opening is characterized by a special glandular epithelium (Figure 1: gg) that is clearly distinct from the hypobranchial gland. The females lack an open seminal groove. The testis consists of several lobes that cover the intes- tinal sacs dorsally and at the left side. Sperm cells in all stages of development are found within. Ripe sperm cells have elongated, spiral heads. The vas deferens forms a simple ciliated duct, which is filled with ripe sperm. Again there are no accessory glands. From the genital opening an open seminal groove extends forward along the right of the neck to the right cephalic tentacle, which is used for copulation. In contrast to females, a glandular area around the genital opening and a receptaculum are lacking. Alimentary Tract In Choristella the mouth opening of the alimentary tract (Figure 3) is provided with tall columnar (40 x 3 um) and cuticularized cells with elongated nuclei. Most of the anterior buccal cavity is cuticularized as well. The paired jaws are well developed and consist of numerous elements forming a honeycomblike structure (Figure 4; see also the accompanying paper by MCLEAN, 1992). The sublingual pouch is reduced, and a subradular sense organ is lacking. In general, the choristellid radula consists of 11 teeth per row and resembles that of Cocculinella (compare BUSH, 1897; ‘THIELE, 1909; BOUCHET & WAREN, 1979; HIck- MAN, 1983; MARSHALL, 1983). For detailed descriptions of the radulae of the species investigated see the accom- Explanation of Figures 9 to 13 Figures 9-13. Choristella hickmanae, histological details. Figure 9. Osphradial nerve (os) innervating sensory epi- thelium (arrowhead). Sensory cells have basal round nuclei; supporting cells are characterized by distal elongated nuclei. Scale bar = 20 um. Figure 10. Posterior oesophagus. Scale bar = 200 um. Figure 11. Midgut gland (mg), right kidney (rk), and oviduct (od). Scale bar = 200 wm. Figure 12. Wall of receptaculum with circular muscle sheath (m) and sperm cells (s), which are densely arranged along the wall. Scale bar = 20 um. Figure 13. Huge midgut cell with large mucous droplets (d) and fibers from chondrichthyan egg cases (f). Scale bar = 100 um. G. Haszprunar, 1992 panying paper by MCLEAN (1992). Two pairs of radular cartilages are present. The ventral pair is very large, con- sists of large cartilage cells (Figure 5), and is interconnected by a ventrally inserted horizontal muscle. The second pair is much smaller, consists of much smaller cells, and is positioned dorsally to the first pair (Figure 3). A thick oral sphincter could be observed; otherwise it was not possible to reconstruct the buccal musculature in detail. The small salivary glands are situated dorsally and are simple glandular pouches without a distinct duct. Imme- diately posterior to their opening, the dorsal food channel begins with the appearance of two symmetrical, densely ciliated folds. The emergence point of the oesophagus is marked by the buccal commissure. The anterior oeso- phagus is quite simple and contains only the folds of the dorsal food channel. Laterally it conveys the products of the small oesophageal pouches, which produce very small dark-brown granules. In Choristella marshall: the pouches are of equal size, whereas in C. hickmanae the right one is about three times larger than the left (Figure 5). At this region the folds of the dorsal food channel flatten and disappear. The posterior oesophagus begins as a broad and simple duct of irregular shape without any folds. Its epi- thelium is ciliated and mucus cells are interspersed. After a short distance many large longitudinal folds arise abrupt- ly and characterize the most posterior part of the oesoph- agus (Figure 10), which is provided with a specific gland. This tubular gland surrounds the posterior oesophagus and is composed of three cell types (Figure 6): (1) Very small (diameter about 10 um) cuboidal cells form the blind ends of the tubes and produce mucus that stains bright violet. The lumen of this region is wide (40 um). (2) The epithelial cells of the proximal portion of each tube are much larger (diameter 25-30 um), stain brightly and bear a microvillous border. The lumen is here very narrow, about 5 wm. (3) The cells of the distal region are of a similar size, contain reddish-staining material near their apical surfaces, and bear cilia. The lumen is here somewhat broader, about 10 wm. Numerous tubules enter the pos- terior oesophageal wall at its opening into the stomach. The stomach region is highly modified and specialized. A prominent stomach is lined with ciliary tracts and con- tains a cuticular shield with a large tooth and a very small caecum. Mucus cells are interspersed in the epithelium of the stomach. Near the entrance of the oesophagus a large, densely ciliated fold separates two very large cul-de-sacs from the stomach region. The right sac is smaller, is sit- uated more dorsally, and invades the posterior mantle roof. The left sac is situated ventrally and extends to the upper visceral whorls. The sacs contain large quantities of long pieces of the chondrichthyan egg cases that are in a mucoid matrix. The epithelium of the sacs consists of three distinct cell types: (1) Huge (diameter up to 400 um) vacuolized cells, which are probably used for digestion, contain (phagocytosed ?) food pieces (Figure 13). (2) Much smaller (maximum diameter 50 wm) mucus cells that stain violet. Page 301 (3) Mucus cells of the same size that contain many small, dark-brown granules. The last cell type is usually scattered among the others; it is the only type in a distinct area of the most anterior portion of the right sac. The intestine is demarked from the ventral wall of the stomach by an abrupt change in diameter (from about 600 um to 200 um). The intestine is very short and is continued by the rectum (diagnosed by its position beneath and more distal to the heart) without a distinct change in histology. The rectum extends laterally to the heart, turns to the very right of the mantle roof, then runs forward and releases the feces into the mantle cavity via a simple anus. The rectum has a prominent typhlosole, whose cells are about three times higher than those of the remaining rectal ep- ithelium (Figure 8). Nervous System The nervous system of all choristellid species investi- gated is uniform. It is wide, streptoneurous, and hypoath- roid (7.e., adjacent pleural and pedal ganglia). The cerebral ganglia are situated laterally in the head and are interconnected by a long cerebral commissure. From the labial lobe the very thin buccal connectives emerge, leading to the buccal ganglia, which are situated to the left and right of the emergence point of the oesophagus. A thick but simple tentacular nerve is present; there are no traces of an optic nerve. In addition, the region of the mouth opening is supplied by several nerves. The pleuro-pedal complex is the largest nervous center. Anteriorly situated pedal nerves supply the pedal gland and the anterior portion of the sole. Posteriorly the pedal ganglia give rise to long pedal cords, which are intercon- nected by several commissures. From the pedal cords fine side branches run to the epipodial tentacles. The visceral chain starts with the pleural ganglia, which are located dorsal to the pedal ganglia. A thick pallial nerve emerges from each pleural ganglion and innervates the mantle border. The visceral chain is wide; the sub- and supraoesophageal ganglia are small. From the latter a connective nerve enters the mantle roof at the very left and swells to form an osphradial ganglion. The visceral (left) and genital (right) ganglia are distinct and are sit- uated along the posterior end of the mantle cavity. Sense Organs Although an eye-lobe is present, there is no trace of eyes or optic nerves. The osphradium is situated along the efferent gill axis. The epithelium is taller (about 20 um) than the adjacent one (10-15 wm), lacks any zonation, and is primarily com- posed of two cell types (Figure 9): (1) Supporting cells with proximally situated, spindle-like, quite condensed nu- clei and (2) sensory cells with basally situated, larger and round nuclei. The latter appear to have contact with the underlining osphradial ganglion because the basal lamina Page 302 is often interrupted. A very few large mucus cells are interspersed. A subradular organ could not be detected. The statocysts are situated posteriorly and adjacent to each other and abut the pedal ganglia, although they are supplied by a cerebral nerve, which is partly fused with the cerebropedal connective. The statocysts contain many small statoconia. DISCUSSION Character Analysis Recently I have discussed the evolutionary history of the Cocculiniformia, including preliminary results on the Choristellidae (HASZPRUNAR, 1988c). Therefore the dis- cussion below is restricted to the choristellid characters and to more controversial points of view. Connective tissue of the foot: The cell type with many brownish droplets (Figure 7) calls for some comment. It does not form epithelia or compact mesenchymate tissues, but is always found interspersed, although sometimes in high densities. It occurs in the foot, the wall of the head, and in the mantle margin, in fact at all positions where massive connective tissue is present. Occurrence and cy- tology of this cell type are typical for so-called ‘“pore- cells” (“Blasenzellen” by German authors such as WOLBURG-BUCHHOLZ, 1972), which have been described in a variety of gastropods and bivalves. In the latter group they have been called “brown cells.” As reviewed by SIMKISS & MAson (1983) and MASON et al. (1984), the fine-struc- ture (sievelike structure for ultrafiltration) of pore cells closely resembles that of podocytes, which however form epithelia. There is evidence that pore cells act in hemo- cyanin synthesis, in phagocytosis, in the recycling of re- spiratory pigments, or generally in metal ion metabolism. Among the streptoneuran gastropods, pore cells have been described in the connective tissue or blood sinuses from all parts of the body in Littorina (MARTOJA et al., 1980; MASON et al., 1984; BROUGH & WHITE, 1990), around the ali- mentary tract in Rissoa (MARTOJA & THIRIOT-QUIEVREUX, 1980), in the foot mass of Diodora sp. (Haszprunar, un- published data), and in the osphradial connective tissue of Campanile symbolicum (HASZPRUNAR, 1992). Gill: From the functional point of view the choristellid gill resembles a monopectinate ctenidium of caenogastropods or certain trochoids in serving for respiration as well as for water currents (cf. YONGE, 1947). However, the chor- istellid gill leaflet has two regions where gas exchange could occur, distally (efferent) and proximally (afferent) of the ciliary band, whereas a monopectinate ctenidium has the proximal zone alone. Also the afferent glandular zone is a peculiarity of the choristellid gill that contrasts with the typical monopectinate ctenidium. The choristellid gill resembles closely that of the Ad- disoniidae (cf. HASZPRUNAR, 1987a). Both gill types are The Veliger, Vol. 35, No. 4 monopectinate, spread over the whole mantle roof, and share the presence of paired skeletal support of the leaflets as well as a highly glandular epithelium composing the large proximal portion. Differences occur in that the cho- ristellid gill leaflets have a distinct ciliary band, whereas in addisoniids the ciliary cells are spread over the whole surface of the distal leaflets. Nevertheless, the addisoniid and choristellid gills are very similar in shape and structure and are probably homologous. It has been argued (HASZPRUNAR, 1987a) that the addisoniid gill may also function in brooding, because of the lack of the glandular zone in juveniles. Juvenile choristellids also lack the glan- dular zone; however, the glandular zone is well developed in males, contradicting the brooding hypothesis at least in the Choristellidae. The complete lack of the glandular zone in juveniles also contradicts an interpretation of the rectal portion of this gland as a hypobranchial gland. A hypobranchial gland would be expected to be present in juveniles as well as in adults. Judging from the characteristics of the gill in the more primitive lepetelloid families, I have argued against the ctenidial nature of lepetelloid gills (HASZPRUNAR, 1988c, d). On the other hand, the choristellid gill fits nearly all characters typical for gastropod ctenidia, including the presence of skeletal rods, the presence of a distinct ciliary band and a distinct respiratory zone, innervation, and blood supply. In addition, the presence of specific sensory pockets (bursicles) in the gill leaflets of several lepetelloid families (Lepetellidae, Pseudococculinidae, Pyropeltidae, Bathy- phytophilidae), which are likewise present at the ctenidial leaflets of vetigastropods (e.g., SZAL, 1971; HASZPRUNAR, 1987b), calls for homology. Although I still think that a non-ctenidial nature of the lepetelloid, and thus of the choristellid gill, is more probable than the opposite as- sumption (modified ctenidia), ontogenetic data are needed to confirm this view more convincingly. Circulatory and excretory system: The general orga- nization of the choristellid circulatory and excretory sys- tems is very similar to that of lepetelloids. In particular, the separation of the right kidney from the pericardium, as well as from the genital system, is shared with several lepetelloid families such as Osteopeltidae, Cocculinellidae, and Addisoniidae (HASZPRUNAR, 1987a, 1988a, c; see Ta- ble 3). Outside the Cocculiniformia this condition is found also in the Seguenziidae (HASZPRUNAR, 1988d:fig. 2Q). Genital system: The Choristellidae are unique among the Cocculiniformia (Lepetelloidea) in being gonochoristic. As outlined elsewhere (MCLEAN & HASZPRUNAR, 1987; HASZPRUNAR, 1988c; Table 2), Lepetelloidea show a trend from hermaphroditism to gonochorism, and the Choris- tellidae appear to represent the final step of this trend. Such a postulated trend (in contrast to the opposite direc- tion) largely parallels the increasing specialization of the alimentary tract (see below) and is further supported by G. Haszprunar, 1992 the hermaphroditic condition in the sister group Coccu- linoidea (outgroup comparison). The logical conclusion is that the choristellid gonochorism probably is a secondary phenomenon. Similar trends from hermaphroditism to gonochorism exist in various bivalve subgroups (MACKIE, 1984), and reproductive strategy may vary even within molluscan genera; therefore, the choristellid (lepetelloid) situation is not exceptional. Among the Lepetelloidea, a separated receptaculum with an opening in the center (slightly left) of the posterior mantle roof is present only in Addisoniidae and Choris- tellidae. However, a similar arrangement exists in the archaeogastropod Melanodrymia aurantiaca (see HASZPRU- NAR, 1989), in the seguenziid Carenzia carinata (Haszpru- nar, unpublished data), and in skeneid-like archaeogas- tropods living on sunken wood such as Leptogyra constricta or Xyloskenea costulifera (see MARSHALL, 1988; Haszpru- nar, unpublished data). Thus, convergence cannot be ex- cluded. Because the more primitive lepetelloid families do not have true receptacula, the addisoniid-choristellid re- ceptaculum is unlikely a plesiomorphy of the Lepetello- idea. Like the remaining Lepetelloidea the Choristellidae use the more or less modified right cephalic tentacle for cop- ulation. The Bathysciadiidae in contrast always have a distinct copulatory organ (THIELE, 1908; HASZPRUNAR, 1988c), whereas both states are present in different genera of the Cocculinidae (HASZPRUNAR, 1987c). Alimentary tract: The choristellid jaws are unique for the superfamily and constitute an apomorphy of the family (McLEAN, 1992). The choristellid jaws can be easily de- rived from the teethlike jaw-elements present in other coc- culiniform families such as Pseudococculinidae, Pyropel- tidae, and Osteopeltidae (HASZPRUNAR, 1988c). In general, modification (Choristellidae), reduction (Cocculinidae), or even loss of jaws (Lepetellidae, Bathyphytophilidae, Coc- culinellidae, Addisoniidae) is common among the Coccu- liniformia. Whereas the presence of two radular cartilages is com- mon among primitive Gastropoda (HASZPRUNAR, 1988d), the dorsal position of the posterior cartilage appears unique for the Choristellidae. The similarity of the choristellid and cocculinellid rad- ula has been mentioned by HICKMAN (1983) and is con- firmed by MCLEAN (1992). However, it is possible that this type has evolved independently in the Cocculinellidae and Choristellidae. Again, radular modifications are com- mon among the Cocculiniformia. The choristellid midgut region is difficult to interpret, and in particular the region of the entrance of the oe- sophagus into the stomach offers problems. Here the “cir- cumoesophageal” tubules and the two cul-de-sacs release their products into the alimentary tract. One may regard the “circumoesophageal” tubules as a fused and modified midgut gland, which is possible from the positions of their Page 303 openings. In this case, the homology of the cul-de-sacs would become questionable, because cul-de-sacs as a spe- cialized portion of a stomach are not known among the Archaeogastropoda. The second possibility, which is much more likely in my view, is that the cul-de-sacs represent (paired) aberrant midgut glands. Indeed, the position of the respective open- ings correspond exactly with those of typical midgut glands. In addition, a pallially situated portion of the right midgut gland is also present in the Cocculinellidae (HASZPRUNAR, 1988a). The argument that a molluscan midgut gland does not contain food particles is contradicted by conditions of the Lepetellidae and Bathyphytophilidae, where such mid- gut glands are present (Haszprunar, unpublished data). However, under this assumption the homology of the “circumoesophageal” glands becomes obscure, since there is no equivalent for this structure in other gastropods. The position of the openings (close to the entrance point into the stomach) and the presence of true oesophageal pouches indicate that they are not true oesophageal glands. How- ever, recent investigations on a new species of Bathyphy- tophilus (Haszprunar, unpublished data) also revealed pe- culiar outpockets of the posterior oesophagus, which possibly serve as a reservoir for symbiotic bacteria. Despite the quite different histology, the choristellid circumoesoph- ageal gland might have a similar function. The topic becomes more complicated when the choristel- lids are compared with the remaining lepetelloid families, in particular with the Addisoniidae (identical nourishment and similar gill type, see above). There the stomach is lost and the distal intestine forms a wide saclike tube (HASZPRUNAR, 1987a, 1988a). Although the respective midgut glands are somewhat specialized, there are no food particles within. In any interpretation these conditions contrast with those of the Choristellidae, where certainly non-intestinal (position of openings) cul-de-sacs store the mass of food particles and a “‘circumoesophageal gland” is present. These differences also lead to the question of whether the use of a common food source in the Choristellidae and Addisoniidae is due to convergence or due to a common ancestor (see below). Among the Cocculiniformia in gen- eral a common food source does not necessarily parallel the organization of the alimentary tract: (1) The cocculinid genus 7 euthirostria (known from its radula) and Bathyscia- diidae feed on cephalopod beaks, yet their radulae are entirely different. (2) Osteopeltidae and Cocculinellidae live on bone (but see below), yet the anatomies of the guts are very different. In all cases this may be because of different strategies in nourishment: although food (bone, egg cases, efc.) are usually found in high density in the alimentary systems, nourishment might be either on the decaying bacteria of the respective substrate or on the substrate itself by specific enzymes or with the aid of sym- biotic bacteria. Studies of the ultrastructure of the gut contents are necessary to clear up this point. Page 304 No conclusion about the specific mode of nourishment of the Choristellidae can be reached at the present. Judging from the sections, phagocytosis appears probable, but sym- biosis with bacteria cannot be excluded. It is also not known whether bacterial and/or fungal activity is necessary before the nutrients are available for the animals. Nervous system and sense organs: The choristellid ner- vous system does not show any peculiarities; the hypoath- roid condition once more reflects the archaeogastropod na- ture of the family. Absence of eyes is a common phenomenon among deep- sea gastropods and is thus of minor phylogenetic signifi- cance. Lepetelloidea always have statocysts with several statoconia (HASZPRUNAR, 1988c). On the other hand, stato- conia are present in various archaeogastropod groups such as the Patellogastropoda (Docoglossa), Vetigastropoda, and Seguenziidae, and are therefore of minor phylogenetic im- portance. Phylogenetic Position of the Choristellidae Internal classification: As already noted by MCLEAN (1992), our studies are hampered by the lack of available specimens with soft parts. This also makes it impossible to give clear anatomical support for the species diagnoses presented by MCLEAN (1992). Most probably, characters of the copulatory organ will be useful to distinguish species and possibly also genera as in other lepetelloid families (e.g., Pseudococculinidae; see HASZPRUNAR, 1988b), al- though this cannot be verified at present, because males were available of Choristella marshall: only. Lepetelloid affinities: The lack of a larval shell (7.e., shell produced by the mantle margin prior to metamorphosis), the streptoneurous and hypoathroid central nervous system with pedal cords, the presence of two kidneys, and the type of anterior oesophagus with oesophageal pouches leave little doubt that the Choristellidae belong to the Archaeo- gastropoda (sensu HASZPRUNAR, 1988c, d). Apart from the shell (see below) all remaining choristellid characters fall well within cocculiniform and lepetelloid variability; not a single character contradicts this view. In particular the coelomic conditions (two kidneys and a separated, non- glandular gonoduct), the cocculinellid-like radula (MOSKALEV, 1978; HICKMAN, 1983) and the addisoniid- like gill and nourishment favor an inclusion of the Choris- tellidae in the Lepetelloidea. In addition, cocculiniform groups are well known to be specialized on various ab- errant food in the deep sea (HICKMAN, 1983; MARSHALL, 1986; HASZPRUNAR, 1988c). Relationships to lepetelloid families: HASZPRUNAR (1988c, submitted in 1986) based his conclusions about choristellid affinities on preliminary investigations. In the light of this much more detailed study on choristellid anat- omy, some modifications to the original hypothesis on cho- ristellid relationships are necessary. In particular, it is not possible to derive the choristellid alimentary system with The Veliger, Vol. 35, No. 4 its well-developed stomach from the cocculinellid (reduced stomach) or addisoniid (lost stomach) condition as assumed originally. A cladistic analysis using the software program Hennig’86 (FARRIS, 1988) was made to infer the phy- logenetic relationships of the Choristellidae among the Le- petelloidea. Table 1 provides information concerning char- acter states and the taxa of all higher lepetelloid families. From these, the characters of heart, conditions of kidneys, genital system, and cerebropedal ring have not been con- sidered in the analysis, because they do not provide useful information on the topic (no differences or autapomor- phies). All families are considered to be monophyletic, which is strongly indicated by their diagnostic radular type; all characters were given equal weight. The Pseudococ- culinidae and/or Osteopeltidae were taken as outgroups, with the same result. To infer relationships, estimations of the probability of homology prior to cladistic analysis are necessary. The following basic assumptions were made: (1) Because of their detailed similarity and high complexity I regard as homologous the gills of Addisoniidae and Choristellidae (monopectinate, skeletal rods, distinct proximal glandular zone). (2) Also considered homologous are the distinct modifications of the alimentary tracts of Cocculinellidae and Addisoniidae (stomach reduction or loss, large gran- ules in digestive gland and intestinal sac). (3) In contrast, the identical food source of Addisoniidae and Choristel- lidae cannot be unequivocally interpreted (see above). (4) According to the arrangement of teeth, a derivation of the addisoniid radular type from the cocculinellid-choristellid one appears possible (MARSHALL, 1987). (5) The coccu- linellid gill is known in detail for Cocculinella minutissima (Smith) alone, where it is reduced to two vestigial knobs (HASZPRUNAR, 1988a). However, personal observations on several total specimens of Cocculinella osteophila Marshall (see Acknowledgments) revealed a more prominent pallial gill consisting of several leaflets like those of most lepe- telloid families. Unfortunately the preservation was too poor to study the histology of this gill. Outgroup compar- ison suggests a reduction of the gill in Cocculinella minu- tissima; the basic cocculinellid gill probably resembles the pseudococculinid-osteopeltid condition. (6) Bone feeding of osteopeltids (what bone; cf, MARSHALL, 1987; WAREN, 1989) and cocculinellids (fish bone; cf. MARSHALL, 1983; HASZPRUNAR, 1988a) is considered analogous: the habitat and fauna of large decomposing whales resemble those of the sulfide-rich seeps and vents (SMITH et al., 1989), in contrast to those of smaller decomposing fish. Indeed, there might be an ecological link between the whale-bone feeding Osteopeltidae and the hot-vent living Pyropeltidae (a new species of which has been found recently also on whale bone; McLean, personal communication); both families exhibit radula of the pseudococculinid-like type. In ad- dition, the alimentary tracts of Osteopeltidae and Coccu- linellidae differ largely, and distinct similarities between both families are lacking. Page 305 G. Haszprunar, 1992 ‘saha sey Ajqeqoid pue (€g6l “TIVHSUVJ ‘yidap w Ey] inoqge) 191eM MoTTeYs As9A UT saat] DjLYdGoa}50 *D “(EBBb6 ‘AVNAUdZSVH) sada syou] vuassynurw vjjauynss077 | ‘(suoneasasgo [euosiad) aunjonsjs UMOUYUN Jo sjayee] [18 JUsUTWOId [esraAas sey D/IYdoajso “yD SeIIIYM ‘(LYRE “AVNNUdZSVH{) SIOYeI] [[1B [eISNSIA sey Dwissynurw vjJaUINI90/+D y. (é) so] /Uuasaid (Q) Juasaid (0) 2pm (0) sisejpjoy [esTe/poom (QO) wWeay YSnosyi (0) MosreU ‘Buoy (QO) peznjon.ys ATTews0U (Q) Wuoutu01d (0) sayonod (Q) a1essoZopridrys (Q) Wuasaid (Q) }uasaid jou (0) pesny (Q) 2n1posydeusay (Q) JUsZaIp ‘omy (Q) UeIpses0}0u0Ul (Q) pueys yewtxoid ou (0) spos [e12]24s ou (Q) padoyaaap [am (Q) 1ted auo (Q) quasasd (0) edu yeornowrurds (1) 3S0] (Q) quasaid (0) 9pm (€) saseo B8a aieys/yseysS (1) Weay sassed (1) pauapim doo] ys] “Buoy ({) saynues3 a81e] (Z) 1SO] (1) spurs (Z) 2dAi-prruostppe ({) Buryor] (1) queasaid (1) paresredas (0) ontposydeussy (0) JUssayIp ‘omy (Q) UeIpses0j0Uu0UL (1) pues yeurxoid ym (1) Spor ye19]94s YM (0) padojaaap [Jam (|) auou (1) Surypey (1) syodurty yeorauruAse (1) 380] (Q) quasaid (0) 9pm (€) saseo BSa ayeys /yxreYs (1) Weay sassed (Z) MosIeU 10Ys (Z) payrpour Ajawa.nxa (Q) Juautu0sd (Q) sayonod (1) PAN -PHfeutpNI909 (Q) 1uasaid (1) qasaid (1) paiesedas (1) 9Nstu0yoou08 (Q) JUsTaYIP “omy (Q) ueIpses0}0u0UL ({) puels yeurxoid yim (1) Spor [eI9]948 YM (0) pedojaaap [Jam (Z) sated Aueur (Q) Juasasd (Z) paytoo 4(¢)soy/juasaid ([) saasou Tepod (0) apim (Z) 2u0q ysy (QO) Weay Yysnosyy (L) pouaprm doo] is] ‘Buoy (1) saynues3 331K] ([) 9ZISs UI psonpar (1) spueys ({) ad41-pryfautpno900 (1) Supper] (Q) 1uasaid jou (1) poreredas (0) 2ntposydewssy (Q) WUasaIp om} (Q) uetpses0j}0u0Ur #(¢ ) ia #(c ) «(¢)[etsnsaa /yuautu01d (,) 2uou (1) Surypey (0) yedurty [eotmauruAs (0) uswi3id YIM (Q) Juasaid (1) peie1]U99U09 1VYMIUOS (1) auoq ayeym (Q) Weay ysnoru} (0) Mosreu ‘Buoy (0) paunjonns ATTeusoU (Q) Juautuosd (1) spueys (0) a1essoysopridiys (Q) Juasaid (Q) quasaid jou (1) paresedas (0) 2ntporydewssy (0) WUaTaHIP “OMI (Q) UeIpses0jouoUT (9) uoneuoz yelads ou (QO) Spor [eaJays Buryory (Q) }usutu01d (Q) sted auo (1) Surypey (0) yedurty TeotnoururAs sohq spi0o [epag Sut [epadosqasay (Q) uo Sutpasq winqo2y (A) eunsaiuy (A) purys inspryy (A) yRUOIS snseydosao JomaiUuy (A) enpey smelt uinjnoejdasa.1 payesedag (s)}onpouosy wia}sAs [eUIT) sXoupry Wey] sjayeay TED (Q) wnipodidg sjadde] [eio Tus aeprurpns909 -opnasg aepriuosippy aepry[aistsoyy) Ayrure depl][aurpNdD07+) aepnyedoaso ‘polapioun = ‘orydsowode = ¢ ‘z ‘{ ‘Sorydioworsajd = Q ‘aeprtul[no0000pnasg pue ‘aeprttuosIppy ‘aIepr{[aisuoyy ‘seprfeut~no007) ‘aepnadoaisGQ uo eiep aaneseduiory CLAD Jajoereyy Page 306 A single most parsimonious tree (28 steps, ci = 89, ri = 75) was created with a sequential arrangement of taxa: Pseudococculinidae, Osteopeltidae, Cocculinellidae, Cho- ristellidae, and Addisoniidae. Of course this result depends heavily on the input matrix and on the basic assumption that changes of all characters are of the same likelihood. The result might be different if the remaining lepetelloid families, Lepetellidae and Bathyphytophilidae, the de- tailed anatomy of which remains to be outlined (Haszpru- nar, in preparation), are included in the analysis. There- fore I regard it as a provisional arrangement. Concerning classification I still prefer a sequential ar- rangement of lepetelloid families without the naming of any node in the tree (cf. HASZPRUNAR, 1988c). Shell shape and related characters: Among the Coccu- liniformia, choristellid species are unique in having a coiled teleoconch, and among the Lepetelloidea they are the first example known to be lacking the otherwise typical con- dition that the protoconch tip is partially immersed in the posterior slope of the teleoconch (“fused protoconch tip” of MARSHALL, 1986). Generally it is accepted that gastro- pod limpets are always derived from coiled forms (e.g., YONGE, 1947; FRETTER & GRAHAM, 1962; MCLEAN, 1981); if so, the Choristellidae should represent the most primitive condition among the Cocculiniformia respec- tively among the Lepetelloidea. However, this is in direct contrast to the sequence obtained by the cladistic analysis (see above). According to this and previous phylogenetic analyses (HASZPRUNAR, 1988c; here) the assumption of primitive coiled teleoconchs in the Cocculiniformia would imply the postulation of parallel evolution of limpets, at least twice among the Cocculiniformia: once in the Cocculinoidea (Cocculinidae and Bathysciadiidae) and a second time in the Lepetelloidea (Lepetellidae, Pseudococculinidae, Bathyphytophilidae, Pyropeltidae, Osteopeltidae, Coccu- linellidae, and Addisoniidae). If the Addisoniidae are ac- cepted as a sister-group of the Choristellidae, the addi- soniid condition forces the assumption of another event to evolve limpets. If the sequence revealed by the cladistic analysis above is accepted, even further events of evolution of a limpet shell are to be assumed. Admittedly, the evo- lution of limpets has often occurred among the Archaeogas- tropoda. However, in most cases the assumption of the derived nature of the limpet shell form can be substantiated by a more or less coiled juvenile teleoconch. In contrast, there is no trace of juvenile coiling of the teleoconch in any species of the remaining cocculiniform families. This is also the case in all Patellogastropoda (Docoglossa), which are now regarded as the most primitive gastropod group (WINGSTRAND, 1985; LINDBERG, 1988a, b; HASZPRUNAR, 1988c, d). As outlined in detail elsewhere (HASZPRUNAR, 1988c, d), I consider the uncoiled limpet form to be primary and coiled forms secondary in early gastropod evolution. I frankly admit that this hypothesis needs further confir- mation to be generally accepted. In particular, details of the torsion process of patellogastropods and cocculiniforms The Veliger, Vol. 35, No. 4 should be comparatively reinvestigated by the application of modern techniques (staining methods, video, etc.) to progress with this topic. Nevertheless, also in the light of the new data on Choristellidae and the result of the cladistic analysis above, I think it justified to consider the present dogma of primary coiled forms in the Gastropoda at least with some hesitation and to keep the alternative in mind. ACKNOWLEDGMENTS I am indebted to Bruce A. Marshall (National Museum of New Zealand, Wellington) and James H. McLean (Los Angeles County Museum, California, USA) for providing specimens used in this study and Bruce A. Marshall in particular for the loan of Cocculinella osteophila. James H. McLean made his results on choristellid taxonomy avail- able to me prior to publication. I thank both and Anders Wareén (Naturhistoriska Riksmuseet, Stockholm, Sweden) for critical comments on the manuscript. Sectioning of most specimens was done by Bernhard Ruthensteiner (Institut of Zoology, University of Vienna, Austria); for photo- graphic help I thank Willibald Salvenmoser (Innsbruck). LITERATURE CITED Boss, K. J. 1982. Mollusca. Pp. 947-1166. In: S. P. Parker (ed.), Synopsis and Classification of Living Organisms. Vol. 2. McGraw-Hill Book Company: New York. BOUuCcHET, P. & A. WAREN. 1979. The abyssal molluscan fauna of the Norwegian Sea and its relation to other faunas. Sarsia 64:211-243. BrouGu, C.N. & K. N. WHITE. 1990. Functional morphology of the rectum in the marine gastropod Littorina saxatilis (Oli- vi) (Prosobranchia: Littorinoidea). Journal of Molluscan Studies 56:97-108. Bus, K. J. 1897. Revision of the marine gastropods referred to Cyclostrema, Adeorbis, Vitrinella, and related genera: with description of some new genera and species belonging to the Atlantic fauna of America. Transactions of the Connecticut Academy of Sciences, New Haven 10:97-143, pls. 22-23. Farris, J. S. 1988. Hennig’86, version 1.5: manual, software and MS-DOS program (personally distributed). FRETTER, V. & A. GRAHAM. 1962. British Prosobranch Mol- luscs. Ray Society: London. GUBBIOLI, F. & I. NOFRONI. 1986. First record of “‘Cithna” naticiformis Jeffreys, 1883 in the Mediterranean. La Con- chiglia (Roma) 18(204-205):6-7. HASZPRUNAR, G. 1985. On the innervation of gastropod shell muscles. Journal of Molluscan Studies 51:309-314. HASZPRUNAR, G. 1987a. The anatomy of Addisonia (Mollusca, Gastropoda). Zoomorphology 106:269-278. HASZPRUNAR, G. 1987b. The fine structure of the ctenidial sense organs (bursicles) of Vetigastropoda (Zeugobranchia, Trochoidea) and their phylogenetic significance. Journal of Molluscan Studies 53:46-51. HASZPRUNAR, G. 1987c. Anatomy and affinities of cocculinid limpets (Mollusca, Archaeogastropoda). Zoologica Scripta 16:305-324. HaszpRunar, G. 1988a. Anatomy and systematic position of the bone-feeding limpets, Cocculinella minutissima (Smith) and Osteopelta mirabilis Marshall (Archaeogastropoda). Journal of Molluscan Studies 54:1-20. HaSzpRuNAR, G. 1988b. Anatomy and affinities of pseudococ- G. Haszprunar, 1992 Page 307 culinid limpets (Mollusca, Archaeogastropoda). Zoologica Scripta 17:161-180. HASZPRUNAR, G. 1988c. Comparative anatomy of cocculini- form gastropods and its bearing on archaeogastropod sys- tematics. Jn: W. F. Ponder (ed.), Prosobranch Phylogeny. Malacological Review, Supplement 4:64-84. HASZPRUNAR, G. 1988d. On the origin and evolution of major gastropod groups with special reference to the Streptoneura. Journal of Molluscan Studies 54:367-441. HASZPRUNAR, G. 1989. The anatomy of Melanodrymia auran- tiaca Hickman, a coiled archaeogastropod from the East Pa- cific hydrothermal vents (Mollusca, Gastropoda). Acta Zoo- logica, Stockholm 70:175-186. HASZPRUNAR, G. 1992. The first molluscs—small animals. Bol- lettino di Zoologico 59:1-16. HICKMAN, C.S. 1983. Radular patterns, systematics, diversity, and ecology of deep-sea limpets. The Veliger 26:73-92. Kuropa, T., T. HABE & K. OYAMA. 1971. The Sea Shells of Sagami Bay. Maruzen & Co. Ltd.: Tokyo. 489 pp. LINDBERG, D.R. 1988a. Heterochrony in gastropods—a neon- tological view. Pp. 183-196. In: M. L. McKinney (ed.), Heterochrony in Evolution. A Multidisciplinary Approach. Plenum Press: New York. LINDBERG, D. R. 1988b. The Patellogastropoda. In: W. F. Ponder (ed.), Prosobranch Phylogeny. Malacological Re- view, Supplement 4:35-63. Mackie, G. L. 1984. Bivalves. Pp. 351-418. In: A. S. Tompa, N. H. Verdonk & J. A. M. van den Biggelaar (eds.), The Mollusca, 7: Reproduction. Academic Press: London. MarSHALL, B. A. 1983. The family Cocculinellidae (Mollusca: Gastropoda) in New Zealand. Records of the National Mu- seum of New Zealand 2(12):139-143. MarSHALL, B. A. 1986. Recent and Tertiary Cocculinidae and Pseudococculinidae (Mollusca: Gastropoda) from New Zea- land and New South Wales. New Zealand Journal of Zo- ology 12:505-546. MARSHALL, B.A. 1987. Osteopeltidae (Mollusca: Gastropoda): a new family of limpets associated with whale bone in the deep-sea. Journal of Molluscan Studies 53:121-127. MARSHALL, B. A. 1988. Skeneidae, Vitrinellidae and Orbites- tellidae (Mollusca, Gastropoda) associated with biogenic substrata from bathyal depths off New Zealand and New South Wales. Journal of Natural History 22:969-1004. MartToja, M. & C. THIRIOT-QUIEVREUX. 1980. Les cellules a sulfure cuivrique du Rissoidea, Rissoa ventricosa Desm. (Gasteropode, Prosobranche): répartition, caractéres, inter- prétation écotoxicologique. Archive de Zoologie experimen- tale et generale 122:271-278, 2 pls. Martojé, M., V. T. TUE & B. EKLAIM. 1980. Bioaccumu- lation du cuivre chez Littorina littorea (L.) (Gastéropode Pro- sobranche): signification physiologique et écologique. Jour- nal of Experimental Marine Biology and Ecology 43:251- 270. Mason, A. Z., K. SiMKiss & K. P. RYAN. 1984. The ultra- structural localization of metals in specimens of Littorina littorea collected from clean and polluted sites. Journal of the Marine Biological Association of the U.K. 64:699-720. McLgan, 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. 1985. The archaeogastropod family Addison- iidae Dall, 1882: life habit and review of species. The Veliger 28:99-108. McLEaN, J. H. 1991. Four new pseudococculinid limpets col- lected by the deep-submersible Alvin in the Eastern Pacific. The Veliger 34:38-47. McLeEan, J. H. 1992. Systematic review of the family Choris- tellidae (Prosobranchia: Lepetellacea), with description of new species. The Veliger 35(4):273-294. McLEan, J. H. & G. HASZPRUNAR. 1987. Pyropeltidae, a new family of cocculiniform limpets from hydrothermal vents. The Veliger 30:196-205. MOsKALEV, L. I. 1978. The Lepetellidae (Gastropoda, Pros- obranchia) and forms similar to them. Works of the P. P. Shirshov Institute of Oceanography, Academy of Sciences of the USSR 113:132-146 [in Russian, English translation by G. V. Shkurkin]. PONDER, W. F. & A. WaREN. 1988. Appendix: classification of the Caenogastropoda and Heterostropha—a list of the family-group names and higher taxa. Jn: W. F. Ponder (ed.), Prosobranch Phylogeny. Malacological Review, Supplement 4:288-326. RaGozzi, A. 1985. A proposito di Addisonia lateralis. Notizario C.I.S.M.A. 6:53. SIMKIss, K. & A. Z. MASON. 1983. Metal ions: metabolic and toxic effects. Pp. 101-164. In: P. W. Hochachka (ed.), The Mollusca. Vol. 2: Environmental Biochemistry and Physi- ology. Academic Press: London. SMITH, C. R., H. KUKERT, R. A. WHEATCROFT, P. A. JUMARS & J. W. DEMING. 1989. Vent fauna on whale remains. Nature, London 314:27-28. SZAL, R. A. 1971. New sense organ in primitive gastropods. Nature, London 229:490-492. THIELE, J. 1908. Ueber die Anatomie und systematische Stel- lung von Bathysciadium, Lepetella und Addisonia. Bulletin of the Museum of Comparative Zoology, Harvard 52:81-90, pls. 1-2. THIELE, J. 1909. Cocculinoidea und die Gattung Phenacolepas und Titiscania. Systematisches Conchylien-Cabbinett von Martini & Chemnitz. Vol. 2/11a (Nr. 539):1-48, pls. 1-6. THIELE, J. 1929. Handbuch der Systematischen Weichtier- kunde. Teil 1: Loricata/Gastropoda 1 (Prosobranchier), pp. 1-778. Gustav Fischer Verlag: Jena. VERRILL, A. E. 1882. Catalogue of the marine Mollusca added to the fauna of New England during the last ten years. Transactions of the Connecticut Academy of Sciences, New Haven 5:447-599, pls. 52-53. VERRILL, A. E. 1884. Second catalogue of Mollusca recently added to the fauna of New England coast and adjacent parts of the Atlantic, consisting mostly of deep sea species, with notes on other previously recorded. Transactions of the Con- necticut Academy of Sciences, New Haven 6:139-294, pls. 28-32. VILLA, R. 1985. Note su habitat ed ecologia di Addisonia lateralis (Requien, 1848). Notizario C.1.S.M.A. 5:9-12. WaREN, A. 1989. New and little known Mollusca from Iceland. Sarsia 74:1-28. WINGSTRAND, K. G. 1985. On the anatomy and relationships of recent Monoplacophora. Galathea Report 16:1-94, pls. 1-12. WOLBURG-BUCHHOLZ, I. 1972. Blasenzellen im Bindegewebe des Schlundringes von Cepea nemoralis L. (Gastropoda, Sty- lommatophora). I. Feinstruktur der Zellen. Zeitschrift fur Zellforschung und Mikroskopische Anatomie 128:100-114. YONGE, C. M. 1947. The pallial organs in the aspidobranch gastropods and their evolution throughout the Mollusca. Philosophical Transactions of the Royal Society of London, B 232:443-518. The Veliger 35(4):308-315 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 The Fine Structure of the Columellar Muscle of Some Gastropod Mollusks by M. FRESCURA anpb A. N. HODGSON Department of Zoology and Entomology, Rhodes University, Grahamstown 6140, South Africa Abstract. The ultrastructure of the columellar muscle of five species of prosobranch gastropod with a conical spiral shell, one species (Haliotis spadicea) with a dome-shaped spiral shell, and two pulmonates (Siphonaria capensis and S. concinna) with a conical limpet shell is described. All species have muscle cells (designated Type I) which contain the conventional contractile apparatus of randomly arranged thick and thin filaments, few dense bodies and mitochondria, and a poorly developed sarcoplasmic reticulum. When measured from sections of tissue, the thick filaments range in mean diameter from 26 nm in 7Zurbo sarmaticus to 69 nm in Burnupena cincta. Thick filament diameters measured from prep- arations of isolated filaments are always larger, ranging from 34 nm in 7. sarmaticus to 83 nm in Siphonaria spp. The diameters of the muscle cells and the thick filaments within the cells of spiral- shelled gastropods are, with the exception of the sandy beach whelk Bullia rhodostoma, smaller, when compared to those in the columellar muscle of limpets. All thick filaments have an axial periodicity (about 14 nm) or Bear-Selby net, a feature of filaments that contain paramyosin. A second type of muscle cell (Type II), in the columellar muscle of Burnupena cincta and Bullia rhodostoma, is described. These cells contain bundles of thin filaments which are striated in appearance (periodicity range, 92 to 104 nm). Associated with the muscle is collagenous connective tissue. Whereas the muscle cells of the columellar muscle from H. spadicea and Siphonaria spp. are surrounded by collagen only, those of gastropods with spiral shells are often separated by vesicular spaces, thought to be blood spaces. INTRODUCTION The pedal musculature of gastropods is composed of two distinct regions, the columellar muscle and the tarsic mus- cle, that differ structurally and functionally (VOLTZOow, 1988, 1990; FrEscURA & Hopcson, 1990). The colu- mellar muscle arises in the foot and is inserted onto the columella of the shell. In gastropods with a spiral shell, the principal functions of this muscle are to extend and retract the head and foot into or out of the shell and to rotate the foot (FRETTER & GRAHAM, 1962; BROWN & ‘TRUEMAN, 1982; TRUEMAN & BROwN, 1985; ‘TRUEMAN & Hopcson, 1990). These actions may result in consid- erable changes in the length of the columellar muscle. By contrast, in gastropods with a conical shell (limpets), the principal function of the columellar muscle is to raise the shell off or clamp it firmly against the substratum. Clamp- ing is achieved by powerful contractions of the columellar muscle with minimal change in muscle length (FRESCURA, 1991). The differing functions of the columellar muscle of gastropods with different shell forms suggest that struc- tural differences might exist that would reflect the different roles of the muscle. The structure and arrangement of muscle bundles or fibres within the columellar muscle have been described in detail at the light microscope level for a range of gas- tropods (ROTARIDES, 1941; TTRUEMAN & Brown, 1976, 1985; GAINEY, 1976; BROWN & TRUEMAN, 1982; VOLTzow, 1990; TRUEMAN & Hopcson, 1990; FRESCURA & Hopcson, 1989, 1990). Rather less is known of the fine structure of the columellar muscle. The only studies to date appear to be by PLESCH (1977) on Lymnaea stag- nalis, HUDDART et al. (1977) on Buccinum undatum and Neptunia antiqua, WATABE et al. (1986) on Batillus cor- nutus, and FRESCURA & HODGSON (1989, 1990) on patellid limpets. The aims of this study are to examine the fine structure of the columellar muscle of prosobranch gastropods that have spiral shells, and compare the findings with our pre- vious work on prosobranch limpets (FRESCURA & M. Frescura & A. N. Hodgson, 1992 A Shell Tarsos Tarsos Page 309 Shell R Shell Tarsos Figure 1 Schematic diagrams to illustrate the regions (R) from which columellar muscle was sampled from the limpets (A), gastropods with spiral shells (B), and Haliotis (C). Hopcson, 1989, 1990). In addition we present findings on the structure of the columellar muscle of the pulmonate limpet Stphonaria. MATERIALS anD METHODS Eight species of gastropod with different shell forms were collected from rocky and sandy shores of the east coast of South Africa. Gastropods inhabiting the rocky shores in- cluded four species (Burnupena cincta (Roding, 1798), Buccinidae; Oxystele sinensis (Gmelin, 1791) and O. tagrina (Anton, 1839), Trochidae; Turbo sarmaticus Linné, 1758, Turbinidae) with shells that were spiral and conical, one species (Halvotis spadicea Donovan, 1808, Haliotidae) with a spiral but dome-shaped shell, and two species (Szphonaria capensis Quoy & Gaimard, 1833, and S. concinna Sowerby, 1824; Pulmonata) with a limpet form of shell. One species (Bullia rhodostoma Reeve, 1847, Nassariidae) with a spiral, conical shell was collected from the intertidal zone of a sandy beach. Animals were transported to the laboratory and housed in seawater aquaria until dissected, usually within 24 hr of collection. For light microscopy, portions of the mid-regions of the columellar muscle (Figure 1) were fixed in either 5% for- mal saline or Bouin’s aqueous fixative, embedded in Para- plast, sectioned at 6 wm and stained by two methods: Mil- ligan’s trichrome, omitting orange G stain, and Mallory’s trichrome (HUMASON, 1967). Both methods differentiate between collagenous connective tissue and muscle. For electron microscopy, pieces of tissue about 1 mm? were excised from the mid-region of the columellar tissue (Figure 1) under seawater and fixed in 2.5% glutaralde- hyde in seawater (approximately isosmotic with tissues) for 12 hr at 4°C. Fixed tissue was washed in 0.1 M sodium cacodylate buffer (pH 7.2), postfixed with 1% osmium tetroxide in the same buffer for 90 minutes, dehydrated, and embedded via propylene oxide in an Araldite CY212/ Taab 812 resin mixture (Cross, 1989). Silver/gold sec- tions, cut using glass knives, were stained either with 5% aqueous uranyl acetate for 30 minutes, followed by lead citrate for 10 minutes, or with 10% methanolic uranyl acetate for 30 minutes. Material was viewed on a Jeol JEM CXII eiectron microscope at 80 kV. Isolated filaments were also prepared for transmission electron microscopy from the columellar muscle of Bur- nupena cincta, Oxystele sinensis, Turbo sarmaticus, Bullia rhodostoma, Haltotis spadicea, and Siphonaria concinna. For comparison, isolated filament preparations were made from four species of patellid limpet (Patella granularts, P. oculus, P. cochlear, and P. tabularis), the fine structure of the col- umellar muscle from these limpets having been described in a previous paper (FRESCURA & Hopcson, 1990). For all species, filaments from the dorsoventral bundles of the columellar muscle were homogenized using a Sorvall blender. All mechanical parts and tissue were kept at 0°C. To remove soluble proteins, tissue was transferred to a buffer containing solutions with the final concentrations of 60 mM KCl, 5 mM MgCl, 1 mM NaN,, 0.5 mM EGTA, 10 mM Imidazole, 5 mM DTT, 0.5% Triton X-100, 2 ug/mL leupeptin, pH 7.0 at 4°C. Tissue was washed and spun again before being incubated in a me- dium containing solutions of a final concentration of 80 mM KCl, 5 mM MgCl, 2mM NaN,;,5mM EGTA, 10 mM ATP, 20 mM MES-KOH, 5 mM DTT, 2 ug/mL leupeptin, pH 6.0 at 4°C for 30 minutes on ice to dissociate the thick and thin filaments. Isolated filaments were neg- atively stained with 10% methanolic uranyl acetate on carbon or formvar coated copper grids and viewed using a Jeol JEM CXII transmission electron microscope at 80 or 100 kV. All chemicals were obtained from Sigma Chem- ical Company, USA. All measurements presented in the results were taken from photographs using a Summagraphics digitizing tablet and SigmaScan (Jandel Scientific, USA) software. RESULTS Organization of the Columellar Musculature Light microscopy revealed that the arrangement of mus- cle within the columellar muscle of all species is similar to that previously described for other gastropods (ROTARI- DES, 1941; TRUEMAN & Brown, 1976, 1985; GAINEY, 1976; BROWN & TRUEMAN, 1982; VOLTZOw, 1990; TRUE- MAN & HOoDGsOoN, 1990; FRESCURA & Hopcson, 1990). Therefore only a brief description follows. The majority of muscle fibres are orientated along the long axis of the columellar muscle. These muscle fibres The Veliger, Vol. 35, No. 4 Page 310 M. Frescura & A. N. Hodgson, 1992 are divided into bundles by transverse and oblique muscles. In most species the longitudinal muscle bundles can be up to 150 wm in diameter. The muscle is surrounded by col- lagenous connective tissue, which occupies all of the in- tercellular space in S7phonaria spp. and Haliotis spadicea, and most of the intercellular space in Turbo sarmaticus, Burnupena cincta, Bullia rhodostoma, and Oxystele spp. In the latter case some small intercellular spaces were ob- served. As reported in a previous paper (FRESCURA & Hopecson, 1989), collagen can constitute up to 30% of the columellar tissue (as estimated by area calculations from photographs of sections). Fine Structure of the Columellar Muscle Two types (designated Type I and Type II) of muscle cell have been found in the columellar tissue. Type I cells: Type I cells (Figure 2A) are the predominant muscle cell type in all species. It is probable that the cells are spindle-shaped, and cell diameters range from 3 to 6 um in Oxystele spp. and Haliotis spadicea, 3 to 8 wm in Turbo sarmaticus and Burnupena cincta, 4 to 10 um for Siphonaria spp., and 8 to 13 wm for Bullia rhodostoma (Table 1). Cell components of Type I cells include a ran- dom arrangement of thin and thick filaments (Figure 2B, C). In the Type I cells of all species, the dense bodies, which are sites of thin filament attachment, are relatively electron lucent (Figure 2B) with a loose granular sub- structure and a diameter of about 117 nm (n = 10 mea- surements per species) (Figure 2B). Electron-dense plaques, which may be attachment sites for thin filaments, are some- times seen at the cell membrane. Mitochondria are 1 to 2 wm long on average. In Bullia rhodostoma they occur along the central axis of the cell, whereas in Turbo sarmaticus, Burnupena cincta, and Oxyste- le spp. they are both central and peripheral (Figure 2A). Mitochondria are observed rarely in the muscle cells of Hialiotis spadicea and Siphonaria spp. and seen only at the cell periphery. Subsarcolemmal cisternae are not well de- veloped but are present in all species. Thick filament diameters measured in transverse section vary within and between genera from 26 + 7 nm in Turbo sarmaticus to 69 + 20 nm in Burnupena cincta (Table 1). Some of this variation may be attributable to the filaments being tapered. Measurements of diameters of thick fila- ments in muscle cells of Turbo sarmaticus revealed that Page 311 some Type I cells have filaments with a mean diameter of 26 nm, whereas in others the thick filament diameters are much greater (56 nm) (Table 1). Reliable measurements of thick filament lengths from transmission electron mi- crographs were difficult to obtain but lengths greater than 20 um were observed. Axial striations were seen in lon- gitudinal sections of thick filaments of all genera examined (Figure 2D). The striations had a periodicity of about 14- 15 nm (Table 1). The diameter of the thin filaments in all species was about 6-7 nm (Table 1) as is typical for actin-containing filaments. Length measurements were impracticable be- cause the filaments meandered in and out of the plane of the sections. Type II cells: Type II cells, similar to those described from the columellar muscle of patellid limpets (FRESCURA & Hopcson, 1990), were observed in Burnupena cincta and Bullia rhodostoma only. These cells contain no thick filaments or dense bodies, but mitochondria, subsarcolem- mal cisternae, and glycogen granules are present (Figure 3A, B). The cells have thin filaments that are bundled together by periodic electron-dense regions that give the structures a striated appearance (Figure 3A, B, C). Center- to-center spacings between the electron-dense regions are 92 and 104 nm in Burnupena cincta and Bullia rhodostoma respectively, values that are similar to those reported for patellid limpets (FRESCURA & Hopcson, 1990; Table 2). Intercellular Regions Intercellular regions can separate muscle cells by as much as 9 um. These areas are often packed with collagen (Figure 2A), which is mainly organized in bundles or cross-linked arrays orientated parallel to the long axis of the muscle cells. The collagen fibrils (Figure 2E) have an axial repeat of 62 nm (n = 50). The diameters of the fibrils vary, having an upper limit of 70 nm. In Burnupena cincta, Oxystele spp., Turbo sarmaticus, and Bullia rhodostoma, spaces (presumed to be blood spaces) are also present be- tween the muscle cells (Figure 3A). These spaces were not observed in Szphonaria spp. or Haliotis spadicea. Isolated Filaments All tissue homogenates contained thick and thin muscle filaments as well as collagen fibrils. The diameters of the thick filaments varied among the species (Table 1), ranging Figure 2 Electron micrographs of Type I muscle cells. A. Muscle from Burnupena cincta showing cells with central and peripheral mitochondria (m). Scale bar = 4 wm. B and C. Higher magnifications of muscle cells from Turbo sarmaticus showing the arrangement of thick (T) and thin (t) filaments. Scale bars = 0.5 um. D. Longitudinal section of thick filaments showing axial striations (arrowed). Scale bar = 0.5 um. E. Longitudinal section through intercellular collagen fibrils. Scale bar = 0.5 um. Key: a, attachment plate; c, collagen; d, dense body; dv, dorsoventral muscle; m, mitochondrion; s, sarcoplasmic reticulum; T, thick filaments; t, thin filaments. Page 312 Table 1 The Veliger, Vol. 35, No. 4 Comparison of the morphological features of the columellar muscle from a range of gastropods. Data for Patella sp. from FRESCURA (1991), for Lymnaea from PLESCH (1977). Dm = mean diameter of muscle cells; DT1 = mean diameter of thick filaments measured from sections of tissue; DT2 = mean diameter of thick filaments measured from isolated filaments; P = axial periodicities of thick filaments measured from isolated filaments (7 = number of filaments with 20 periodicities measured per filament); dt = mean diameter of thin filaments. Range (nm) DT2 (nm) P (nm) dt (nm) 11-130 83 + 20 14.2 + 0.2 6-7 (n = 44) 26-94 84 + 29 14°2) E102 6-7 (n = 16) (n = 5) (n = 65) 10-120 80 + 21 14.2 + 0.3 6-7 (n = 10) (n = 5) (n = 25) 11-100 87 + 26 14.3 + 0.4 6-7 (n = 10) (n = 5) (n = 25) 30-180 100 + 30 14.3 + 0.3 6-7 (n = 10) (n = 5) (n = 20) 22-137 SYyae 15) 14.3 + 0.8 6-7 (n = 10) (n = 4) (n = 20) 33-75 63 + 8 15.2 + 0.6 6-7 (n = 10) (n = 5) (n = 20) 33-80 61+9 14.7 + 0.3 6-7 (n = 8) (n = 3) (n = 20) 17-30 3445 14.6 + 0.3 6-7 33-82 79 +6 (n = 20) (n = 12) 50-90 69 + 27 14.9+ 0.5 6-7 (n = 12) (n = 6) (n = 20) 30-50 78 + 15 14.5 + 0.3 6-7 (n = 8) (n = 3) (n = 20) os a 14 7 Species Dm (um) Range (um) DT1 (nm) Patella granularis 9 4-13 Oi 25 Zi) (n = 44) (n = 44) P. oculus 9 4-13 69 + 20 (n = 65) (n = 65) P. miniata 9 4-13 60 + 20 (n = 20) (n = 188) P. cochlear 9 4-13 64 + 20 (n = 20) (n = 58) P. tabularis 9 4-13 SORES 5) (n = 20) (n = 10) Siphonania sp. 9 4-10 oy) ae Alte} (n = 20) (n = 65) Halwotis spadicea 4 3-6 60 + 18 (n = 20) (n = 21) Oxystele sp. 5 3-6 5 Or teat (n = 20) (n = 20) Turbo sarmaticus 6 3-8 PAS a (n = 20) 56+ 15 (n = 24) Burnupena cincta 6 3-8 69 + 20 (n = 20) (n = 21) Bullia rhodostoma 9 8-13 60 + 8 (n = 20) (n = 20) Lymnaea stagnalis _ 3-12 30.4 + 2.5 from 34 nm in Turbo sarmaticus to 100 nm in Patella tabularis. The values obtained from measurements of the isolated filaments were always slightly greater than those measured from transmission electron microscopy of intact tissue. The thick filament lengths for all species ranged between 15 and 30 um, but some of this variability is probably due to filaments breaking during the isolation procedure. Although typical thin filaments were seen, no striated filaments were isolated. The supernatant con- tained only amorphous material. Thick filaments showed either the Bear-Selby net (BEAR & SELBY, 1956) or axial striations (Figure 3D-I) with a periodicity ranging from 14.2 to 15.2 nm (Table 1). Iso- lated collagen fibrils have a striated periodicity of 62 nm. DISCUSSION The columellar muscles of the gastropods examined in this study are predominantly composed of one type of muscle cell (designated Type 1), which has many similar fine structural features in common with Type I cells from the columellar muscle of patellid limpets (FRESCURA & Hopcson, 1990), the shell muscle of Lymnaea stagnalis (PLESCH, 1977), and the anterior byssus retractor muscle (ABRM) of Mytilus edulis (SOBIESZEK, 1973; NICAISE & AMSELLEM, 1983). The similarities include a random ar- rangement of thick and thin filaments, thin filament di- ameters of about 6-7 nm, a paucity of dense bodies and mitochondria, and a poorly developed sarcoplasmic retic- ulum. There are, however, several important differences be- tween the Type I muscle cells of gastropods with spiral shells and the equivalent cells of patellid and siphonariid limpets. With the exception of the sandy beach whelk Bullia rhodostoma, muscle cell diameters are 10 to 12% smaller in the columellar muscle of gastropods with spiral shells (Table 1). In addition, the diameters of the thick filaments of the muscle cells, irrespective of whether they were measured from sections of tissue or isolated filaments, are smaller for most species of spiral-shelled gastropod (Table 1). The exception to this is B. rhodostoma. In all cases, thick filament diameters were greater when mea- sured from the isolated filaments. This discrepancy is prob- ably due in part to the different preparative procedures, M. Frescura & A. N. Hodgson, 1992 Page 313 =F Figure 3 A-C. Sections of Type II muscle cells from the columellar muscle of Bullia rhodostoma. A. Transverse and oblique sections through muscle cells showing the centrally located mitochondria (m) and intercellular vesicular spaces (V) between the cells. Scale bar = 4 um. B and C. Higher magnification of Type II cells showing bundles of striated thin filaments (Sf), sarcoplasmic reticulum (s), and pockets of glycogen (g). Scale bars = 0.5 wm. Figure D-I. Examples of isolated filaments: D, Haliotis spadicea; E, Siphonaria capensis; F, Turbo sarmaticus; G, Patella oculus; H, Patella tabularis; 1, Oxystele sinensis. The Bear-Selby net is apparent in D and F and axial striations in E, G, H, and I. Scale bars = 100 nm. Page 314 Table 2 Center-to-center striation spacings of bundled thin fila- ments from Type II muscle cells from the columellar mus- cle of two gastropods with spiral shells (present study) and four limpets (data from FRESCURA & Hopcson, 1990). 7 = number of independent bundles measured, with 3-8 striations per bundle. Spacing Species (mean + SD) n Patella oculus 135 + 13.0 10 P. vulgata 109 + 7.6 16 P. longicosta 106 + 8.2 10 P. tabularis 89 + 9.0 3 Burnupena cincta 9251510 10 Bullia rhodostoma 104 + 16.0 6 the procedure for isolating filaments causing less shrinkage (BENNETT & ELLIOTT, 1981; CHANTLER, 1983). CHANT- LER (1983) discusses the positive relationship between the diameter of the thick filaments (and hence thick to thin filament ratios) and the force developed by muscle. Thus the combination of thicker muscle bundles and the greater diameter of thick filaments would mean that the columellar muscle of both prosobranch and pulmonate limpets should be more powerful than the equivalent muscle of gastropods with a spiral shell. Clearly this would be advantageous to limpets that inhabit areas where wave activity is most intense. The sandy beach whelk B. rhodostoma, which has a spiral shell, also has relatively large muscle bundles and thick filaments. This gastropod also inhabits areas of in- tense wave activity, where it often extends its large foot, holds it rigid, and uses it to surf up and down the beach (BRowN, 1982). A powerful columellar muscle would be of great advantage for such activity. In Turbo sarmaticus two populations of thick filaments with different diameters were found (Table 1). This finding is not unique, as MorrIsON & ODENSE (1974) also obtained a bimodal distribution for thick filament diameters from the opaque adductors of Arctica islandica and Astarte undata. Although the functional significance of possessing two sizes of thick filament is unknown, the performance characteristics of each muscle cell type may differ. The thick filaments from the columellar muscle of all species have an internal structure that appears to be typical of paramyosin-containing filaments. Furthermore, the ax- ial periodicity of the filaments is about 14 nm, a value that is similar to that of paramyosin-containing filaments from other mollusks (SOBIESZEK, 1973; ELLIOTT & BENNETT, 1982; FRESCURA & Hopcson, 1990). Although the precise function of paramyosin is not known, the molecule is prev- alent in muscle that has a catch mechanism (WATABE & HARTSHORNE, 1990). It is not known whether the colu- mellar muscle of gastropods has catch properties, but such a mechanism would be energetically advantageous for both The Veliger, Vol. 35, No. 4 shell support in spiral-shelled gastropods and clamping in limpets. A further difference in the structure of the columellar muscle of gastropods with coiled shells when compared to limpets (and Haliotis spadicea) is the central and peripheral distribution of the mitochondria in the former and the peripheral distribution only in the latter. Centrally located mitochondria have also been reported in the pedal muscle of Bullia rhodostoma (DA SILVA & HopGson, 1987) and Nassarius krausstanus (TRUEMAN & Hopcson, 1990). Such a distribution may improve the efficiency of ATP avail- ability and may therefore reflect how active the muscle is. In B. rhodostoma and N. kraussianus, the pedal muscle is very active and the centrally located mitochondria are very abundant (DA SILVA & HopcGson, 1987; TRUEMAN & Hopacson, 1990). In the less active columellar muscle de- scribed in this study, mitochondria are not as abundant. FRESCURA & HODGSON (1990) described a second type of muscle cell (designated Type II) in the columellar muscle of patellid limpets. Such muscle cells contained no thick filaments but had bundles of thin filaments with a striated appearance. Type II cells were found in two species of gastropod with coiled shells, namely Burnupena cincta and Bullia rhodostoma. Table 2 shows that the periodicity of the striations in the Type II cells of these two species is similar to that reported by FRESCURA & HODGSON (1990). The function of Type II cells is still unknown and it is perplexing that they are not found in all species. Prelim- inary work using immunocytochemistry suggests that the filaments are composed of actin (FRESCURA, 1991) and, hence, they could also contribute to the contractile prop- erties of the columellar muscle. FRESCURA & HODGSON (1989, 1990) reported that col- lagen is abundant in the columellar tissue of patellid lim- pets where it may function as catch connective tissue. In- tercellular collagen is also abundant in all the species examined in this study, particularly in the pulmonate lim- pet Siphonaria and the “limpet-like” archaeogastropod Haliotis spadicea, where it occupies all the intercellular space. In spiral-shelled gastropods, however, in addition to the intercellular collagen, vesicular spaces are present between the muscle cells. These spaces are probably blood spaces, as demonstrated by VOLTZOw (1985) in the pedal musculature of Busycon contrarium. Although the spaces are relatively small, they may well contribute to the me- chanical properties of the columellar muscle. It is therefore possible that the columellar muscle of some gastropods with a spiral shell does not function as a true muscular hydrostat as has been suggested by TRUEMAN & BROWN (1976) and BROWN & TRUEMAN (1982). ACKNOWLEDGMENTS We would like to thank the staff of the Electron Microscope Unit, Rhodes University, for technical services and Rhodes University and the Foundation for Research and Devel- opment for financial assistance. M. Frescura & A. N. Hodgson, 1992 LITERATURE CITED Bear, R.S. & C. C. SELBY. 1956. The structure of paramyosin fibrils according to X-ray diffraction. Journal of Biophysical and Biochemical Cytology 2:55-69. BENNETT, P. M. & A. ELLIOTT. 1981. The structure of the paramyosin core in molluscan thick filaments. Journal of Muscle Research and Cell Motility 2:65-81. Brown, A.C. 1982. The biology of sandy-beach whelks of the genus Bullia (Nassariidae). Oceanography and Marine Bi- ology Annual Revue 20:309-361. Brown, A. C. & E. R. TRUEMAN. 1982. Muscles that push snails out of their shells. Journal of Molluscan Studies 48: 97-98. CHANTLER, P. D. 1983. Biochemical and structural aspects of molluscan muscle. Pp. 77-154. In: K. M. Wilbur & A. S. M. Saleuddin (eds.), The Mollusca. Vol. 4. Academic Press: New York. Cross, R. H. M. 1989. A reliable epoxy resin mixture and its application in routine electron microscopy. Micron and Mi- croscopica Acta 20:1-7. DA SiLvA, F. M. & A. N. HopGson. 1987. Fine structure of the pedal muscles of the whelk Bullia rhodostoma Reeve: correlation with function. Comparative Biochemistry and Physiology 87A:143-149. ELLioTT, A. & P. M. BENNETT. 1982. Structure of the thick filaments in molluscan adductor muscle. Pp. 11-28. Jn: B. M. Twarog, R. J. C. Levine & M. M. Dewey (eds.), Basic Biology of Muscles: A Comparative Approach. Raven Press: New York. FRESCURA, M. 1991. Aspects of the structure and function of some gastropod columellar muscles (Mollusca). Doctoral Thesis, Rhodes University. 175 pp. FrescurA, M. & A. N. Hopcson. 1989. On collagen and its potential role in the columellar muscle of some gastropods. South African Journal of Science 85:613-614. FREScURA, M. & A. N. Hopcson. 1990. The fine structure of the shell muscle of patellid limpets. Journal of Molluscan Studies 56:435-447. FRETTER, V. & A. GRAHAM. 1962. British Prosobranch Mol- luscs: Their Functional Anatomy and Ecology. Ray Society: London. GaIneEy, L. F. 1976. Locomotion in the Gastropoda: functional morphology of the foot in Neritina reclivata and Thais rustica. Malacologica 15:411-431. HuppartT, H., S. HunT & K. Oates. 1977. Calcium move- ments during contraction in molluscan smooth muscle, and the loci of calcium binding and release. Journal of Experi- mental Biology 68:46-56. Page 315 Humason, G. L. 1967. Animal Tissue Techniques. W. H. Freeman Co.: San Francisco. Morrison, C. M. & P. H. ODENSE. 1974. Ultrastructure of some pelecypod adductor muscle. Journal of Ultrastructure Research 49:228-251. NICAIsE, G. & J. AMSELLEM. 1983. Cytology of muscle and neuromuscular junction. Pp. 1-33. In: K. M. Wilbur & A. S. M. Saleuddin (eds.), The Mollusca. Vol. 4. Academic Press: New York. PLescH, B. 1977. An ultrastructural study of the musculature of the pond snail Lymnaea stagnalis (L.). Cell and Tissue Research 180:317-340. ROTARIDES, M. 1941. Zur Kenntnis der Fussmuskulatur von Nassa mutabilis L. (Gastr. Prosobr.). Annales Historico-na- turales Musei Nationalis Hungarici 34:177-190. SOBIESZEK, A. 1973. The fine structure of the contractile ap- paratus of the anterior byssus retractor muscle of Mytilus edulis. Journal of Ultrastructure Research 43:313-343. TRUEMAN, E. R. & A. C. BRown. 1976. Locomotion, pedal retraction and extension, and the hydraulic systems of Bullia (Gastropoda: Nassariidae). Journal of Zoology, London 178: 365-384. TRUEMAN, E. R. & A. C. BROWN. 1985. The mechanism of shell elevation in Halzotis (Mollusca: Gastropoda) and a con- sideration of the evolution of the hydrostatic skeleton in Mollusca. Journal of Zoology, London 205:585-594. TRUEMAN, E. R. & A. N. HopGson. 1990. The fine structure and function of the foot of Nassarius kraussianus, a gastropod moving by ciliary locomotion. Journal of Molluscan Studies 56:221-228. VOLTzow, J. 1985. Morphology of the pedal circulatory system of the marine gastropod Busycon contrarium and its role in locomotion (Gastropoda, Buccinacea). Zoomorphology 105: 395-400. VOLTzow, J. 1988. The organization of limpet pedal muscu- lature and its evolutionary implications for the Gastropoda. Malacological Review Supplement 4:273-283. VoLTzow, J. 1990. The functional morphology of the pedal musculature of the marine gastropods Busycon contrarium and Haliotis kamtschatkana. The Veliger 33:1-19. WATABE, S. & D. J. HARTSHORNE. 1990. Paramyosin and the catch mechanism. Comparative Biochemistry and Physiology 96B:639-646. WaTABE, S., Y. OCHIAI, Y. KartyA, T. N.-L. Dink, S. Kimura, J. H. PyeunG & K. HasHIMoTo. 1986. Characterization of three types of turban shell Batilus cornutus muscle— ultrastructure and protein composition. Bulletin of the Jap- anese Society of Scientific Fisheries 52:737-744. The Veliger 35(4):316-322 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 Egg Mass and Intracapsular Development of C'ypraea caputdraconis Melvill, 1888, from Easter Island (Gastropoda: Cypraeidae) by CECILIA OSORIO Departamento de Ciencias Ecologicas, Facultad de Ciencias, Universidad de Chile, Casilla 653, Santiago, Chile CARLOS GALLARDO Instituto de Zoologia, Universidad Austral de Chile, Casilla 567, Valdivia, Chile AND HUGO ATAN Departamento de Ciencias Ecologicas, Facultad de Ciencias, Universidad de Chile, Casilla 653, Santiago, Chile Abstract. Egg masses and intracapsular embryos of Cypraea caputdraconis Melvill, 1888, are studied, documenting a discussion about the type of larval development exhibited by this species. The number of capsules per egg mass ranged between 93 and 423 (¥ = 251) and depended on the size of the female. Each capsule contained between 502 and 1302 embryos with a mean of 880; with 251 capsules per egg mass, the average production was about 220,000 larvae per egg mass. The eggs thus produced were quite numerous and small (« = 102 wm). Nurse cells were not observed and development culminated with the hatching of small veliger larvae. Larval development is of the indirect type characteristic of the tropical cypraeids whose development has been previously reported. In spite of possessing the mechanism of larval dispersion here described, Cypraea caputdraconis has a limited geographic distribution and is restricted to the shores of Easter Island and the neighboring Sala y Gomez Island. INTRODUCTION The Pacific Ocean forms a barrier that can be surmounted only by effective mechanisms of dispersion over great dis- tance (KOHN, 1967; STEHLI et al., 1967; VERMEIJ, 1974, 1987; YAMAGUCHI, 1977; SPRINGER, 1982; SCHELTEMA, 1986). Two alternative mechanisms have been cited: (a) planktonic larvae able to disperse by the currents, as is the case of many oceanic benthic species which main- tain contact even with continental populations (QUAYLE, 1964; OkuBOo, 1971; GERDES, 1977; SCHELTEMA, 1968, 1971, 1986); and (b) passive transport on drifting objects or bodies (DELL, 1972; ARNAUD et al., 1976; HIGHSMITH, 1985; O FOIGHIL, 1989) a case in which planktonic larvae are not necessary for dispersal. The mollusk fauna of Easter Island, derived primarily from the Indo-West Pacific (EKMAN, 1953; LADD, 1960; REHDER, 1980), is endemic, due to extreme isolation in the Pacific Ocean (OSORIO & CANTUARIAS, 1989). This raises the question of what dispersal mechanisms are likely to have evolved in these mollusks, and to what extent such mechanisms may be related to their patterns of geographic distribution. An important member of this fauna is the C. Osorio et al., 1992 Page 317 2mm Figure 1 Egg mass of Cypraea caputdraconis. A. Morphology and typical pattern of grouping of egg capsules. B. Morphology of an egg capsule. endemic Eastern Island gastropod Cypraea caputdraconis Melvill, 1888, which has been long used by the islanders in handcrafts or for exchange (McCoy, 1976). To determine the pattern of larval development, and hence the dispersion mechanisms used by this gastropod, we analyze here its egg masses and some embryonic stages of intracapsular development, reporting also some field observations on the reproductive habits of these snails. Cypraeid egg masses have been little documented. OSTERGAARD (1950) described egg masses and larval de- velopment of five Cypraea species from Hawaii, and BANDEL (1973) described egg masses of two Caribbean cypraeids. WILSON (1985) has reported egg masses and incubation periods, and has described hatching stages of seven cy- praeids from southwest Australia with a direct embryonic development. Finally, KATOH (1989) described the life history of C. annulus in Okinawa, including data on fe- cundity, spawning frequency, and the seasonal occurrence of egg masses. MATERIALS anpD METHODS The described material comes from Easter Island (27°09'S, 109°26'W). Field observations and collections of egg mass- es of Cypraea caputdraconis were conducted in January of 1989 and 1990. Only two complete egg masses were col- lected in 1989; most others were destroyed, due to great difficulty in removing them from the substratum. None- theless, the remainder served to illuminate the general structure of the egg capsules, the number of embryos per capsule, and the intracapsular developmental stages. In the second collection, at Tepito Tecura, nine complete egg masses were obtained, allowing estimates of size and em- bryonic productivity in relation to the size of the incubating females. Snails and egg masses were fixed in 10% formalin- seawater. Animals were sexed and then measured with a Vernier caliper to the nearest 0.1 mm. Egg capsules and embryos were examined and counted under a stereoscopic microscope provided with an ocular micrometer. Photo- graphs were taken with a Wild Photoautomat, model MP5 45. A camera lucida and, occasionally, a drawing tube connected to the optical system of the microscope was used for drawings. RESULTS Egg Masses The egg masses of Cypraea caputdraconis are found on low intertidal substrata exposed to strong wave action. They adhere firmly to the rocky substratum in cavities or Page 318 The Veliger, Vol. 35, No. 4 Figure 2 Egg mass of Cypraea caputdraconis. A and B. Group of egg capsules showing the fixation points. C. Early egg mass (stage 1). D and E. Intermediate phase of egg mass (stages 2 and 3). C. Osorio et al., 1992 crevices which offer effective protection, and from which they are difficult to remove. Other organisms found with the egg masses included polychaetes and small sea urchins (Echinometra insularis) and diverse algae and diatoms. The egg masses of Cypraea caputdraconis (Figures 1, 2) consist of many capsules or oothecae, distributed in two to four strata, with a maximum diameter of the egg mass being 25 mm. The same egg mass may contain capsules at different developmental stages, with capsular areas of different colors, varying from white to brown as the de- velopment of the embryos progresses. Egg masses are brooded by the female, which covers them totally with her foot. Capsules are 1.8-2.8 mm long, generally oblong, and sub-rectangular in shape (Figure 1B); their walls are thin, transparent, and colorless. They adhere by means of a short, wide lamellar peduncle. Capsules of the overlying strata adhere to each other at fixation points (Figure 2B). In nine nests found in cavities on 9 January 1990, the female was always attending the egg mass. In three cases, a male was also present; in all three, the egg masses were at the initial cleavage stage. The size of females ranged from 24 to 35.7 mm in length, and that of males from 26 to 31.7 mm. The number of capsules per egg mass (Table 1) ranged from 93 to 423 with a mean of 251.5 (SD = 97.2) and was positively correlated with female size (r = 0.82; P < 0.05). Excluded from these calculations was an egg mass formed by only 82 capsules, since it was still at its initial stage of formation when oviposition was interrupted. For 57 oothecae, the number of embryos per capsule ranged from 502 to 1302 (mean = 879.7; SD = 158.2), but it was not correlated with the ootheca length (r = 0.116). An average egg mass produced by a female contains approx- imately 220,000 embryos. We have been able to verify the existence of egg masses only during the summer months (December-March); ob- servations have not yet been conducted in the remaining months of the year. Intracapsular Development In some egg capsules we observed uncleaved eggs with a mean diameter of 112 wm (SD = 5.1; n = 125). We also distinguished at least five different stages of intracapsular development (Figure 3). In stage 1, the earliest embryonic stage observed (Figure 3A), embryos were spheric or slightly oval and 102-111 ym in diameter (« = 106 um), very similar in size to eggs before initiating development. Stage 2 corresponds to clearly more advanced embryos at the phase termed pre-veliger (Figure 3B); an outline of the shell, though not coiled, covers the lower hemisphere of the embryo like a hood. In the neck region are differ- entiated the so-called “nuchal cells” or cephalic kidneys, transient larval structures observed in the embryonic de- velopment of other prosobranchs (PORTMANN, 1925; GAL- LARDO, 1973, 1977, 1979). Such organs have been pos- Page 319 Table 1 Number of egg capsules per spawn in female Cypraea caputdraconis of different sizes captured at Tepito Tecura (Easter Island) on 9 January 1990. No. of egg capsules Size of female per spawn 24.0 93 24.9 228 26.2 203 DHE? 238 28.3 303 30.6 315 31.5 209 B57 423 tulated to play an excretory function (WIERZEJsKI, 1905; Bloch, cited by KUME & Dan, 1968). The length of stage 2 embryos ranged from 144 to 161 wm, with a mean of 158 um. Stage 3 (Figure 3C) is an intermediate phase of intra- capsular veligers, characterized by a globose shell (111- 128 um; * = 119 wm), which is slightly coiled, and has a pale yellow tinge. A slight outline of the operculum is evident on the foot. The stage 4 veliger (Figure 3D) is characterized by a more pronounced coiling of the shell, which has a slightly brown tinge and ranges in size from 137 to 140 wm in length (* = 142.6 um). The operculum and foot are more developed, and two eye spots are present. In stage 5 (Figure 3E, F) the shell measures 164-178 um in length (x = 170.4 wm). These shells are further coiled and golden brown in color. A dorsocentral beak has formed on the shell over the cephalic region, similar to that observed by DISALVo (1988) in planktonic veligers of Concholepas concholepas. All of the pre-hatching veligers of Cypraea caputdraconis contained in the same capsule were quite uniform in their morphology (Figure 3E). In 18 term oothecae, containing stage 5 larvae and having a mean length of 2.5 mm, the number of larvae per capsule varied from 502 to 1009 (¥ = 788.5 + 129). DISCUSSION The pattern of behavior and reproduction of Cypraea ca- putdraconis resembles that of other cypraeid gastropods (OSTERGAARD, 1950; BANDEL, 1973; WILSON, 1985; KaTOH, 1989). The egg mass is a group of tenuous, trans- lucent capsules distributed in strata and incubated by the female, which protects them under her foot. Though some- what variable in their morphology, the capsules resembled those described by BANDEL (1973) for C. cinerea. The data obtained in the present study suggest a pattern of indirect development with phytoplanktotrophic pelagic larvae. This conclusion is supported by the large number of embryos per egg mass, the relatively small size of the Page 320 The Veliger, Vol. 35, No. 4 Figure 3 Stages of intracapsular development. Cypraea caputdraconis. A. Early embryos (stage 1). B. Pre-veliger phase (stage 2). C. Intermediate veliger phase (stage 3). D. Veliger larvae (stage 4). E and F. Terminal veliger larvae, prior to eclosion (stage 5). Scale = 100 um. egg, the absence of nurse cells, and the small size of the larva at its terminal stage of intracapsular development, prior to hatching. These features roughly coincide with observations on other tropical cypraeid egg masses releas- ing planktonic larvae (OSTERGAARD, 1950; D’AsArRo, 1970; Kay, 1960; BANDEL, 1973). Indirect larval development is the dominant pattern among Cypraea, and particularly characterizes cypraeids inhabiting tropical waters (WILSON, 1985). Species with a planktonic larval stage hatch 10,000 to 500,000 veliger larvae per egg mass, usually with the production of hundreds in each capsule. These species clearly contrast with directly developing species inhabiting temperate seas on the south coast of Australia and in South Africa (WILSON, 1985). The species from southern Australia described by Wilson are char- acterized by the development of a single embryo per oo- C. Osorio et al., 1992 theca, which consumes all other remaining eggs that serve as nurse cells. In each spawn, females produce 50 to 100 offspring in the genus Notocypraea and 150 to 300 in Zoila and Austrocypraea. The same occurs with Cypraeovula in South Africa. According to data reported for tropical cy- praeids with planktonic larvae (WILSON, 1985), we might have expected that the brooding time of Cypraea caput- draconis would be relatively short. In tropical cypraeids it fluctuates between 11 and 18 days, while temperate species with direct development may require from 45 to 55 days to complete intracapsular development. The existence of planktonic larvae in Cypraea caputdra- conis raises the question of capacity for dispersal during their planktonic drift. The open ocean of the tropical Pa- cific forms a barrier for dispersal, which can be overcome only by species possessing the ability for long distance transport. Among the mollusks of the Indo-Pacific there are species with a strictly continental distribution and oce- anic species that can inhabit both continental and oceanic islands (TAYLOR, 1971; REID, 1985). The tendency to lack planktonic larvae is more marked among continental gas- tropods, whereas their oceanic counterparts tend to retain the primitive modality of planktonic larval development (PERRON & KOHN, 1985; VERMEIJ, 1987). The existence of planktonic larvae among gastropods inhabiting oceanic islands is supported by our findings in Cypraea caputdraconis. In turn, direct development has been reported for cypraeids of continental coasts, such as those of Australia and South Africa (WILSON, 1985). Notwith- standing, it must be noted that the distribution of C. ca- putdraconis is limited to Easter Island and Sala y Gomez (OsoRIO & CANTUARIAS, 1989) and the species has not been reported from continental shores. This fact suggest that larvae of C. caputdraconis are not successfully dis- persed over great distances, but rather that dispersal is restricted to a local scale within the oceanic insular area in which the species is found. ACKNOWLEDGMENTS The authors are grateful to Dr. F. Jara and to shell col- lectors of the Pakarati family for valuable assistance in the field, and two anonymous reviewers for suggestions im- proving this manuscript. This work was supported finan- cially by the Departamento Técnico de Investigacion, Universidad de Chile. Proyecto N° 3046-9012. LITERATURE CITED ARNAUD, F., P. M. ARNAUD, A. INTES & P. LE LOEUFF. 1976. Transport d’invertebres benthiques entre l’Afrique du Sud et Sainte Helene per les laminaires (Phaeophyceae). Bulletin Museum Histoire Naturelle, Paris, Ser. 3, 384:49-55. BANDEL, K. 1973. Notes on Cypraea cinerea G. and Cyphoma gibbosum L. from the Caribbean Sea and description of their spawn. The Veliger 15(4):335-337. D’Asaro, C..N. 1970. Egg capsules of prosobranch mollusks from south Florida and the Bahamas and notes on spawning in the laboratory. Bulletin Marine Science 20:414-440. Page 321 DELL, R. K. 1972. Antartic benthos. Advances in Marine Bi- ology 10:1-216. DiSaALvo, L. H. 1988. Observations on the larval and post- metamorphic life of Concholepas concholepas (Brugiére, 1789) in laboratory culture. The Veliger 30(4):358-368. EKMAN, S. 1953. Zoogeography of the Sea. Sidwick & Jackson: London. GALLARDO, C.S. 1973. Desarrollo intracapsular de Concholepas concholepas (Brugiére) (Gastropoda Muricidae). Museo Na- cional de Historia Natural. Publicacion Ocasional 16:3-16. GALLARDO, C. S. 1977. Two modes of development in the morphospecies Crepidula dilatata (Gastropoda Calyptraei- dae) from southern Chile. Marine Biology 39(3):241-251. GALLARDO, C.S. 1979. Developmental pattern and adaptations for reproduction in Nucella crassilabrum and other murica- cean gastropods. Biological Bulletin 157(3):453-463. GERDES, D. 1977. The re-establishment of an Amphiura fili formis (O. F. Muller) population in the inner part of the German Bight. Pp. 277-296. In: B. F. Keegan, P. O’Ceidigh & P. J. S. Boaden (eds.), Biology of Benthic Organisms. HIGHSMITH, R.C. 1985. Floating and algal rafting as potential dispersal mechanisms in brooding invertebrates. Marine Ecology Progress Series 25:169-179. KaTOH, M. 1989. Life history of the golden ring cowry Cypraea annulus (Mollusca Gastropoda) on Okinawa Island. Japan Marine Biology 101:227-233. Kay, E. A. 1960. The functional morphology of Cypraea ca- putserpentis Linn. and an interpretation of the relationships among the Cypraeacea. International Revue Gesamten Hy- drobiologie 45:175-196. Koun, A. J. 1967. Environmental complexity and species di- versity in the gastropod genus Conus on Indo-West Pacific reef platforms. American Naturalist 101:251-259. KumE, M. & K. Dan (eds.). 1968. Invertebrate Embryology. NOLIT Publishing House: Belgrade. 605 pp. Lapp, H.S. 1960. Origin of the Pacific island molluscan fauna. American Journal of Science, Bradley Vol. 258-A:137-150. McCoy, P. C. 1976. Easter Island settlement patterns in the late prehistoric and protohistoric periods. Bulletin 5. Eastern Island Committee, International Foundation for Monu- ments Inc.: New York. O FoicuiL, D. 1989. Planktotrophic larval development is associated with a restricted geographic range in Lasaea, a genus of brooding, hermaphroditic bivalves. Marine Biology 103:349-358. Oxuso, A. 1971. Oceanic diffusion diagrams. Deep-Sea Re- search 18:789-802. Osorio, C. & V. CANTUARIAS. 1989. Vertical distribution of mollusks on the rocky intertidal of Easter Island. Pacific Science 43(4):302-315. OSTERGAARD, J. M. 1950. Spawing and development of some Hawaiian marine gastropods. Pacific Science 4:75-115. PERRON, F. E. & A. J. KOHN. 1985. Larval dispersal and geographic distribution in coral reef gastropods of the genus Conus. Proceedings 5th International Coral Reef Congress 4:95-100 [C. Gabrie, et al. (eds.), Antenne Museum EPHE, Morea, French Polynesia.] PORTMANN, A. 1925. Der Einfluss der Nahreier auf die Lar- venen twicklung von Buccinum und Purpura. Zeitschrift fur Morphologie und Oekologie der Tiere 3:526-541. QuayLe, D. B. 1964. Distribution of introduced marine Mol- lusca into British Columbia waters. Journal of the Fisheries Research Board of Canada 21:1155-1181. REHDER, H. A. 1980. The marine mollusks of Easter Island and Sala y Gomez. Smithsonian Contributions to Zoology 289:1-167. Page 322 Reip, 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. SCHELTEMA, R.S. 1968. Dispersal of larvae by equatorial ocean currents and its importance to the zoogeography of shoal- water tropical species. Nature 217:1159-1162. SCHELTEMA, R.S. 1971. Larval dispersal as a means of genetic exchange between geographically separated populations of shallow-water benthic marine gastropods. Biological Bul- letin 140:284-322. SCHELTEMA, R.S. 1986. Long-distance dispersal by planktonic larvae of shoal-water benthic invertebrates among central Pacific islands. Bulletin Marine Science 39:241-256. SPRINGER, V.G. 1982. Pacific Plate biogeography, with special reference to shorefishes. Smithsonian Contributions to Zo- ology 367:1-182. STEHLI, F. G., A. MCALESTER & C. E. HELSLEY. 1967. Tax- onomic diversity of recent bivalves and some implications for geology. Bulletin of the Geological Society of America 78: 455-465. The Veliger, Vol. 35, No. 4 TayLor, J. D. 1971. Reef-associated molluscan assemblages in the western Indian Ocean. Pp. 509-536. Jn: D. R. Stod- dard & C. M. Yonge (eds.), Regional Variations in Indian Ocean Coral Reefs. Academic Press: New York. VERMEIJ, G. J. 1974. Regional variations in tropical high intertidal gastropod assemblages. Journal of Marine Re- search 32:343-357. VERMEIJ, G. J. 1987. The dispersal barrier in the tropical Pacific; implications for molluscan speciation and extinction. Evolution 41:1046-1058. WIERZEJSKI, A. 1905. Embryologie von Physa fontinalis. Zeit- schrift wissenschafts Zoologie 83 [see KUME & DAN, 1968]. WiItson, B. R. 1985. Direct development in southern Austra- lian cowries (Gastropoda: Cypraeidae). Australian Journal of Marine and Freshwater Research 36:267-280. YAMAGUCHI, M. 1977. Larval behavior and geographic dis- tribution of coral reef asteroids in the Indo-West Pacific. Micronesica 13:183-196. The Veliger 35(4):323-329 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 New Morphologic and Geographic Data on the Neritid Gastropod Nerita (Theliostyla) triangulata Gabb, 1869, from the Eocene of the Pacific Coast of North America by RICHARD L. SQUIRES Department of Geological Sciences, California State University, Northridge, California 91330, USA Abstract. Abundant and exceptionally well-preserved specimens of the neritid gastropod Nerita (Theliostyla) triangulata Gabb, 1869, from brackish-marine deposits of the middle Eocene (““Domengine Stage”) Delmar Formation, northern San Diego County, show previously unreported morphologic features. A tricarinate body whorl is a major characteristic of the species, but on some of the more mature Delmar Formation specimens the three carinae become faint to obsolete as they approach the outer lip. Comparative studies reveal three subjective synonyms of Nerita (Theliostyla) triangulata. They are Nerita triangulata Gabb var. oregonensis Merriam & Turner, 1937, from lower Eocene (“Capay Stage”) strata, southwestern Oregon, Nerita quadrangulata Weaver & Kleinpell, 1963, from upper Eocene (“Tejon Stage’’) strata, southern California, and Nevita n. sp. Clark, 1938, from upper Eocene (“Tejon Stage’’) strata, northern California. The new geographic range of N. (7.) triangulata is San Diego to southwestern Oregon, and the new geologic range is upper lower Eocene (““Capay Stage”’) to upper Eocene (“Tejon Stage’’). Nerita (Theliostyla) triangulata shows close affinities with a few southern European species. 7heliostyla is probably a Tethyan immigrant that arrived on the Pacific coast of North America during the early Eocene. INTRODUCTION The Pacific coast of North America neritid gastropod Neri- ta (Theliostyla) triangulata Gabb, 1869, is a common con- stituent of Eocene molluscan faunas in very shallow marine or brackish-water deposits. By all accounts, it is a species characterized by three strong widely spaced carinae on the body whorl. Recent collecting, however, has yielded ex- ceptionally well-preserved specimens that show a grada- tion from three strong carinae on the early part of the body whorl to faint carinae to no carinae on the late part of the body whorl. It is the purpose of this present study to report on this morphologic variation. All reported Paleogene spe- cies of Nerita from the Pacific coast of North America also were reviewed, and three taxa were detected as subjective synonyms of N. (T.) triangulata. Nerita (Theliostyla) triangulata has no apparent ancestor among indigenous Paleocene or Cretaceous faunas but does have relatives in the lower Eocene of southern Europe. Nerita pentastoma DESHAYES (1866:19, pl. 66, figs. 7-9; COSSMANN & PISsARRO, 1910-1913:pl. 5, fig. 38-5; Sz6Ts, 1953:141, pl. 2, figs. 1, 2), from the lower Eocene of the Paris Basin, France, and of Hungary closely resembles N. (T.) triangulata and should be assigned to the subgenus Thelwostyla. Nerita tricarinata LAMARCK (1804:94; 1806:pl. 14, figs. 4a, b; DESHAYES, 1837:pl. 19, figs. 9, 10; Coss- MANN & PISSARRO, 1910-1913:pl. 5, fig. 38-4; COSSMANN, 1925:pl. 6, figs. 41, 42; PALMER, 1977:vélin 47, figs. 7a, b) from the lower Eocene of the Paris Basin, France, also closely resembles N. (7.) triangulata. GLIBERT (1962:100) assigned N. tricarinata to the subgenus Theliostyla. Nerita hebert: SZOTS (1953:141-142, pl. 2, figs. 3-5) and Nerita hantkeni SZOTs (1953:142, pl. 2, figs. 6, 7), both from the lower Eocene of Hungary closely resemble N. (7.) trian- gulata and appear to be referable to the subgenus The- liostyla. Page 324 Oceanside © Carlsbad 7 ™ Cardiff By The Figure 1 Index map of locality SDSNH 3522 in northern San Diego County. Theliostyla probably originated in the Old World Teth- yan paleobiotic province and immigrated to the Pacific coast of North America during the early Eocene. In ad- dition to the appearance of Nerita (Theluostyla) triangulata on the Pacific coast of North America during the early Eocene, another possible species of Nerita (Theliostyla) appeared in this area during this time. Woops & SAUL (1986) reported a single specimen of N. (7.) n. sp. (?) from the Sepultura Formation near Punta Rosarito, Baja California, Mexico. They questionably assigned this for- mation to the upper Paleocene. FLYNN et al. (1989) as- signed these strata to the lower Eocene (‘“Capay Stage’’). The specimen of Woops & SAUL (1986) seems to be allied to N. (T.) crook: Clark, 1938, known from the upper Eo- cene Markley Formation on Pleasant Creek, Solano Coun- ty, northern California. The molluscan stages used in this report stem from CLARK & VOKES (1936), who proposed five mollusk-based provincial Eocene stages, namely, ‘““Meganos,” “‘Capay,” “Domengine,” “Transition,” and “Tejon.” The stage names are in quotes because they are informal terms. GIVENS (1974) modified the use of the “Capay Stage,” and it is in this modified sense that the “Capay Stage” is used herein. Abbreviations used for catalog and/or locality numbers are: CAS, California Academy of Sciences, San Francisco; LACMIP, Los Angeles County Museum of Natural His- tory, Invertebrate Paleontology Section; SDSNH, San Di- The Veliger, Vol. 35, No. 4 ego Society of Natural History; SUPTC, Stanford Uni- versity Paleontology Type Collection (now housed at CAS); UCMP, University of California Museum of Paleontol- ogy, Berkeley. NEW GEOGRAPHIC OCCURRENCE The new specimens of Nerita (7.) triangulata were found a few kilometers west of the city of San Marcos, northern San Diego County (Figure 1) at locality SDSNH 3522. This locality is 1220 feet (372 m) north and 2180 feet (665 m) east of the southwest corner of section 17, T12S, R3W, U.S. Geological Survey, San Marcos, California, 1968 (photorevised 1983), 7.5-minute topographic quadrangle, at elevation 570 feet (174 m). Approximately 7.5 m of interbedded sandstone and muddy siltstone within a chan- nel complex were temporarily exposed in 1989 by con- struction work in the immediate vicinity of the locality (:.e., the Laurels housing development project). Numerous mol- lusks and abundant leaves were found at locality SDSNH 3522 at the bottom of the exposure in greenish gray, poorly sorted, very fine- to coarse-grained sandstone with clayey matrix. Angular grains of quartz are common. Outcrops of granitic basement are within 200 m of locality SDSNH 3522) The lithology, sedimentary structures, and fossil content at locality SDSNH 3522 agree very closely with the “green mudstone and sandstone” lithology of lithofacies 3 of E1- SENBERG & ABBOTT (1991), who did a detailed study of the various middle Eocene paralic environment rock types that crop out in northern San Diego County. They assigned their lithofacies 3 to the Delmar Formation, which is cor- relative to the upper lower to lower middle Eocene “‘Do- mengine Stage” (GIVENS & KENNEDY, 1979). Mollusks associated with the new specimens of Nerita (T.) triangulata include the gastropods Potamides carbonico- la Cooper, 1894, and Umpquaia oregonensis, Turner, 1938, and the bivalves Acutostrea idriaensis fettkei (Weaver, 1912), Pelecyora aequilateralis (Gabb, 1869), and Corbula (Cu- neocorbula) torreyensis Hanna, 1927. All of these mollusks, including N. (7.) triangulata, are diagnostic of brackish- marine conditions (GIVENS, 1974; GIVENS & KENNEDY, 1976, 1979; SQuiREs, 1991; SQuIRES & DEMERE, 1991). The mollusks at locality SDSNH 3522 were transported a short distance and concentrated within a channel com- plex, along with land-plant remains. The shoreline must have been very nearby because the shells show no evidence of abrasion or sorting. The shoreline was probably irreg- ular with basement crystalline headlands and protected lagoon, and this interpretation would be in keeping with what other workers have reported for the middle Eocene shoreline in the San Diego area (LOHMAR & WARME, 1979; EISENBERG, 1985; SQUIRES & DEMERE, 1991). Nerita (Thelwostyla) triangulata has been previously re- ported from San Diego County. HANNA (1927) and GI- VENS & KENNEDY (1979) reported the species from the Delmar Formation near Torrey Pines State Park. GIVENS R. L. Squires, 1992 Page 325 Explanation of Figures 2 to 9 Figures 2-9. Nerita (Theliostyla) triangulata Gabb, 1869, locality SDSNH 3522, Delmar Formation, northern San Diego County, southern California. Figure 2: hypotype SDSNH 406777, abapertural view, x6. Figure 3: hypotype SDSNH 406787, apertural view, x 4.4. Figures 4, 5: hypotype SDSNH 43472, abapertural and lateral views, x 3.1. Figures 6, 7: hypotype SDSNH 43473, abapertural and oblique apertural views, x 2.9. Figures 8, 9: hypotype SDSNH 43474, slightly crushed, abapertural and lateral views, x 2.7. & KENNEDY (1976) also reported the species from probable middle Eocene (““Domengine Stage’’) rocks near the city of Vista in northern San Diego County. SYSTEMATIC PALEONTOLOGY Family NERITIDAE Rafinesque, 1815 Subfamily NERITINAE Rafinesque, 1815 Genus Nerita Linné, 1758 Type species: By subsequent designation (MONTFORT, 1810), Nerita peloronta Linné, 1758. Subgenus Theliostyla Morch, 1852 Type species: By subsequent designation (KOBELT, 1879), | Nerita albicilla Linné, 1758. Nerita (Theliostyla) triangulata Gabb, 1869 (Figures 2-18) Nerita (Theliostyla) triangulata GABB, 1869:170, pl. 28, figs. 52, 52a; VOKES 1939:182, pl. 22, figs. 31, 33, 34; GIVENS, 1974:61, pl. 5, fig. 4; GIVENS & KENNEDY, 1976:960, 963, pl. 1, figs. 1-4; SQuiREs, 1987:23, fig. 14. Nerita triangulata Gabb: ARNOLD, 1910:14, pl. 14, figs. 12, 12a (figs. repeated in ARNOLD & ANDERSON, 1910:pl. 26, figs. 12, 12a); HANNA, 1927:301, pl. 46, figs. 11, 12, 16, 17; Moore, 1968:28, pl. 12a. Nerita triangulata Gabb var. oregonensis MERRIAM & TURNER, 1937:104, pl. 6, fig. 5; TURNER, 1938:95, pl. 19, figs. 10-12; WEAVER, 1942 [1943]:295-296, pl. 64, figs. 10, 13 Nerita n. sp.: CLARK, 1938:701, pl. 4, fig. 6. Nerita quadrangulata WEAVER & KLEINPELL, 1963:183, pl. 23, fig. 1. Supplementary description: Small sized, broader than high, with rapidly expanding body whorl. Spire very low to flattened, apex usually depressed below gently sloping dorsal surface. Dorsal surface with four to six noded spiral ribs (excluding carina on shoulder) that become coarser toward outer lip. Posterior part of dorsal surface of body whorl elevated and somewhat carinate. Body whorl with three widely spaced, somewhat nodose carinae. On some specimens larger than about 9.5 mm height, carinae on body whorl become faint to obsolete toward outer lip. Degree of obsolescence of carinae increases with size of shell. Medial carina slightly stronger than other two ca- rinae. On a few specimens, body whorl shoulder rounded, and spiral cord in middle of interspace between shoulder and medial carina can approximate or exceed strength of shoulder. Spiral ribs in interspaces of carinae become stronger and more nodose toward outer lip. Juvenile spec- imens faintly nodose, late juvenile specimens nodose, and a few late-mature specimens strongly beaded. Some spec- imens with a fine reticulate sculpture pattern in interspaces near inner lip. Interspace between shoulder and medial carinae with two spiral ribs near inner lip and three to four spiral ribs on late part of body whorl. Interspace between medial and anteriormost carinae with two to three spiral ribs near inner lip and three to five on late part of body whorl. Interspace between anteriormost carina and Page 326 The Veliger, Vol. 35, No. 4 Explanation of Figures 10 to 18 Figures 10-18. Type material of subjective synonyms of Nerita (Theliostyla) triangulata Gabb, 1869. Figures 10- 12: Nerita triangulata Gabb var. oregonensis Merriam & Turner, 1937, Lookingglass Formation, Douglas County, southwestern Oregon, holotype UCMP 33204, abapertural, apertural, and lateral views, x3.3. Figures 13-16: Nerita quadrangulata Weaver & Kleinpell, 1963, undifferentiated Sacate-Gaviota formation, Santa Barbara County, southern California. Figures 13, 14: holotype GAS 66550.01 (ex SUPTC 9175), apertural view, x2.7. Figures 15, 16: paratype CAS 66550.02 (ex SUPTC 9175a), abapertural and lateral views, x 3.3. Figures 17, 18: Nerita n. sp. Clark, 1938, Markley Formation, northern California, hypotype UCMP 30490, abapertural and lateral views, xi2= 5h base of body whorl with two to five (rarely six) fairly prominent spiral ribs with elongate or irregular nodes. Aperture large, quadrate. Outer lip flared, with grooves where intersected by three body-whorl carinae, and with or without shallow grooves where intersected by spiral ribs on the exterior of the body whorl. Outer lip interior with 11 to 15 narrow, elongate teeth not extending to outer lip periphery; interspaces between teeth generally align with grooves along outer lip periphery; teeth usually widely spaced but a few can be paired. Inner lip with six to seven teeth. Two posteriormost teeth stronger than rest, with tooth next to posteriormost tooth strongest. Three to four small, subequal teeth medially. Anteriormost tooth blunt and well developed only in larger specimens. Deck with 16 to 23 small tubercles, arranged very loosely in rows somewhat coincident with inner lip teeth. Two posterior- most teeth can extend onto deck area as ridges, with tooth next to posteriormost the strongest. Original color pattern with alternating axial bands of white and brown becoming chevron shaped on crests of body whorl carinae and forming zig-zag pattern in inter- spaces of carinae, especially in interspace near base of body whorl. Chevrons point toward outer lip on crests of carinae. Color pattern best developed on specimens that are not strongly noded or beaded. Discussion: Of the 61 specimens of Nerita (Theliostyla) trangulata from locality SDSNH 3522, seven have carinae that become faint to obsolete with growth. An example of faint carinae on the late part of the body whorl is shown in Figure 6, and an example of no carinae in that area is shown in Figure 8. Both of these specimens, however, have three strong carinae on the early part of the body whorl (Figures 7, 9, respectively). The late part of the body whorl of the specimen illustrated in Figure 8 resembles the ho- lotype (and only known specimen) of Nerita (Theliostyla) crooki CLARK (1938:700, pl. 4, figs. 1, 2) from the upper Eocene Markley Formation on Pleasant Creek, Solano County, northern California. Nerita (T.) crooki, however, has no carinae on the early part of the body whorl. This species is assigned herein to the subgenus Theliostyla be- cause of the presence of tubercles on the deck. GIVENS & KENNEDY (1976) collected five, exceptionally well-preserved specimens of Nerita (T.) triangulata from a locality in very shallow marine or brackish-water de- posits of probable middle Eocene age (‘““‘Domengine Stage’”’), near the city of Vista in northern San Diego County. Locality SDSNH 3522 is about 3.8 km to the south of their locality. Their specimens are all less than about 9.5 mm in height and have a tricarinate body whorl. MERRIAM & TURNER (1937) reported multiple speci- R. L. Squires, 1992 mens (exact number unknown) of Nerita triangulata var. oregonensis Merriam & Turner, 1937, from the lower Eocene (“‘Capay Stage”) Capay Formation, northern Cal- ifornia, and from the upper Umpqua Formation near Glide, Douglas County, southwestern Oregon. Revisions of strati- graphic nomenclature in southwestern Oregon now place the localities of the Glide section from which N. t. oregonen- sis has been collected in the lower Eocene (““Capay Stage’’) Lookingglass Formation (BALDWIN, 1974, 1975; THoms, 1975; MILEs, 1981). MERRIAM & TURNER (1937) and TURNER (1938) re- ported that the body whorl of Nerita triangulata var. ore- gonensis is less sharply angulate and has weaker spiral ribs than does N. (7.) triangulata. It is not clear from the original description of N. ¢. oregonensis if MERRIAM & TURNER (1937) were describing a morphologic variant or a new subspecies. Their holotype (Figures 10-12), how- ever, is very much like the Delmar Formation specimen shown in Figures 4 and 5. The Oregon specimen is less nodose than the Delmar Formation specimen because the Oregon specimen is a worn shell. In addition, the Oregon specimen is slightly smaller, and nodosity in N. (7.) trian- gulata is weaker on small specimens. I believe that N. triangulata var. oregonensis and N. (T.) triangulata are con- specific. WEAVER & KLEINPELL (1963) reported Nerita quadran- gulata Weaver & Kleinpell, 1963, from the upper Eocene (“Tejon Stage”) undifferentiated Sacate-Gaviota forma- tion, Nojoqui Creek area, Santa Barbara County, southern California. They reported that N. quadrangulata differs from N. (T.) triangulata by having a much weaker body- whorl shoulder, a fourth and nearly equal carina just an- terior, and five rather than four spiral ribs between the two anteriormost carinae. The holotype of N. quadran- gulata (Figures 13, 14) does have a somewhat weak body- whorl shoulder in the vicinity of the outer lip. It is difficult to determine how strong the body-whorl shoulder was in the early part of the whorl because a large portion of shell is missing there. The best preserved paratype of N. quad- rangulata (Figures 15, 16), however, shows the character- istic tricarinate body whorl in both the early and later parts of the body whorl. Also, given the variability in morphology seen in the Delmar Formation specimens of N. (T.) triangulata, the variability in the strength of the shoulder in N. guadrangulata is not sufficient to distinguish this species from N. (7.) triangulata. In addition, N. (7.) triangulata, like N. quadrangulata, can have as many as five spiral ribs between the two anteriormost carinae; GABB (1869) also noted this feature on N. (T.) triangulata. CLARK (1938) reported a single, poorly preserved spec- imen of Nerita n. sp. from the upper Eocene Markley Formation on Pleasant Creek, Solano County, northern California. This worn specimen of Nerita n. sp. has three faint carinae on the body whorl (Figures 17, 18) and is judged to be_N. (7) triangulata. There are three other named species of the genus Nerita Page 327 from Eocene rocks along the Pacific coast of North Amer- ica. Two, Nerita cowlitzensis DICKERSON (1915:58-59, pl. 5, figs. 7a, b) and Nerita washingtoniana WEAVER & PALM- ER (1922:28-29, pl. 11, fig. 4), are from the upper middle Eocene Cowlitz Formation, southwest Washington, and the third, Nerita vokesi DURHAM (1944:156, pl. 17, figs. 11, 12), is from the Molopophorus stephensoni megafaunal zone of northwest Washington. ARMENTROUT (1975) as- signed this zone to his upper Eocene Galvinian Molluscan Stage. Nerita cowlitzensis differs from N. (T7.) triangulata in the following features: smaller, nodose only on the dorsal surface, body whorl with only minute sculpture, and ap- erture more elongate. Nerita washingtoniana differs from N. (T.) triangulata in the following features: much smaller, smooth body whorl, and a possible divaricate color pattern. Nerita vokesi differs from N. (7.) triangulata in the smooth body whorl, only four subequal inner lip teeth, and a much more variable color pattern. It is important to mention that the inner lip teeth of N. (7) triangulata, N. cowlitz- ensis, and N. washingtoniana are very similar. ACKNOWLEDGMENTS Thomas A. Demére (San Diego Natural History Muse- um) informed the author about the well-preserved Delmar Formation specimens and arranged for their loan. Mat- thew Colbert, Daniel J. McGuire, and Bradford. O. Riney (San Diego Natural History Museum) collected these specimens. Colbert also provided field notes and locality information. Jean DeMouth (California Academy of Sci- ences, San Francisco) and Rex Hanger (University of Cal- ifornia, Berkeley) arranged for loans of type material. LouElla R. Saul (Natural History Museum of Los An- geles County) gave valuable comments about fossil nerites. Wesley C. Wehr (Burke Museum, University of Wash- ington) provided information about Nerita vokes. The manuscript benefited from comments by two anonymous reviewers. LITERATURE CITED ARMENTROUT, J. M. 1975. Molluscan biostratigraphy of the Lincoln Creek Formation, southwest Washington. Pp. 14- 128. In: D. E. Weaver, G. R. Hornaday & A. Tipton (eds.), Future Energy Horizons of the Pacific Coast; Paleogene Symposium and Selected Technical Papers. Pacific Sections of American Association of Petroleum Geologists, Society of Economic Paleontologists & Mineralogists, and Society of Exploration Geologists. ARNOLD, R. 1910. Paleontology of the Coalinga district, Fresno and Kings counties, California. United States Geological Survey, Bulletin 396:1-173. ARNOLD, R. & R. ANDERSON. 1910. Geology and oil resources of the Coalinga district, California. United States Geological Survey, Bulletin 398:1-354. BaLpwin, E. M. 1974. Eocene stratigraphy of southwestern Oregon. Oregon Department of Geology and Mineral In- dustries, Bulletin 83:1-40. BALDWIN, E. M. 1975. Revision of the Eocene stratigraphy of Page 328 southwestern Oregon. Pp. 49-64. In: D. W. Weaver, G. R. Hornaday & A. Tipton (eds.), Paleogene Symposium and Selected Technical Papers; Conference on Future Energy Horizons of the Pacific Coast. Pacific Sections of American Association of Petroleum Geologists, Society of Economic Paleontologists & Mineralogists, and Society of Exploration Geologists. Cxark, B. L. 1938. Fauna from the Markley Formation (upper Eocene) on Pleasant Creek, California. Geological Society of America, Bulletin 49:683-730. CLaRK, B. L. & H. E. Vokes. 1936. Summary of marine Eocene sequence of western North America. Geological So- ciety of America, Bulletin 47:851-878. Cooper, J.G. 1894. Catalogue of Californian fossils, parts 2- 5. California State Mining Bureau, Bulletin 4:5-65. CossMANN, A. E. M. 1925. Essais de paléoconchologie com- pareé. Vol. 13. Privately published: Paris. CossMANN, A. E.M. & G. PISSARRO. 1910-1913. Iconographie compléte des coquilles fossiles de !’Eocéne des environs de Paris. Vol. 2 (Gastropodes, etc.). H. Bouillant: Paris. 65 pls. DEsHaAYES, G.-P. 1837. Description de coquilles fossiles des environs de Paris. Atlas (Part 2):pls. 1-101. Chez lauteur et d’autres: Paris. DesHAYES, G.-P. 1866. Description des animaux sans vertébres découverts dans le bassin de Paris. Vol. 3 (Texte):1-667. Atlas (Part 2):pls. 1-107. J.-B.Bailliére et Fils: Paris. DickERSON, R. E. 1915. Fauna of the type Tejon. Its relation to the Cowlitz phase of the Tejon Group of Washington. California Academy of Sciences, Proceedings, 4th Series, 5: 33-98. DurHaAM, J. W. 1944. Megafaunal zones of the Oligocene of northwestern Washington. University of California Publi- cations, Bulletin of the Department of Geological Sciences 27:101-212. EISENBERG, L. I. 1985. Depositional processes in the landward part of an Eocene tidal lagoon, northern San Diego County. Pp. 55-68. In: P. L. Abbott (ed.), On the Manner of De- position of the Eocene Strata in Northern San Diego County. San Diego Association of Geologists, Guidebook. EISENBERG, L. I. & P. L. ABBoTT. 1991. Middle Eocene paralic facies, northern San Diego County, California. Pp. 55-72. In: P. L. Abbott & J. A. May (eds.), Eocene Geologic History San Diego Region. Pacific Section, Society of Economic Pa- leontologists & Mineralogists, Book 68. FLYNN, J. J.. R. M. CIPOLLETTI & M. J. NOvAcEK. 1989. Chronology of early Eocene marine and terrestrial strata, Baja California, Mexico. Geological Society of America, Bulletin 101:1182-1196. Gass, W.M. 1869. Cretaceous and Tertiary fossils. Geological Survey of California. Vol. 2. Palaeontology. Caxton Press: Philadelphia. 299 pp. GIVENS, C. R. 1974. Eocene molluscan biostratigraphy of the Pine Mountain area, Ventura County, California. Univer- sity of California Publications in Geological Sciences 109: 1-107. GIVENS, C. R. & M. P. KENNEDY. 1976. Middle Eocene mol- lusks from northern San Diego County, California. Journal of Paleontology 50:954-975. Givens, C. R. & M. P. KENNEDY. 1979. Eocene molluscan stages and their correlation, San Diego area, California. Pp. 81-95. In: P. L. Abbott (ed.), Eocene Depositional Systems, San Diego, California. Pacific Section, Society of Economic Paleontologists & Mineralogists. GLIBERT, M. 1962. Les Archaeogastropoda fossiles du Cén- ozoique étranger des collections de |’Institut Royal des Sci- The Veliger, Vol. 35, No. 4 ences Naturelles de Belgique. Institut Royal des Sciences Naturelles de Belgique, Série 2, 68:1-124. Hanna, M. A. 1927. An Eocene invertebrate fauna from the La Jolla quadrangle, California. University of California Publications, Bulletin of the Department of Geological Sci- ences 16:247-398. KoBELT, W. 1877-1881. Jn: Martini & Chemnitz, Neue Aus- gabe. Nuremberg. LAMARCK, J. B. P. A. DE MONET DE. 1802-1809. Mémoires sur les fossiles des environs de Paris. Annales du Muséum National d’Histoire Naturelle. Paris. Variously paged. LINNE, C. 1758. Systema naturae per regna tria naturae. Editio 10, reformata 1(1):1-824. Salvii: Holmiae. LouMar, J. M. & J. E. WARME. 1979. An Eocene shelf mar- gin: San Diego County, California. Pp. 165-175. In: J. M. Armentrout, M. R. Cole & H. TerBest, Jr. (eds.), Cenozoic Paleogeography of the Western United States. Pacific Sec- tion, Society of Economic Paleontologists and Mineralogists, Pacific Coast Paleogeography Symposium, No. 3. MERRIAM, C. W. & F. E. TURNER. 1937. The Capay middle Eocene of northern California. University of California Pub- lications, Bulletin of the Department of Geological Sciences 24(6):91-114. MiILEs, G. A. 1981. Planktonic foraminifers of the lower Ter- tiary Roseburg, Lookingglass, and Flournoy formations (Umpqua Group), southwest Oregon. Pp. 85-103. In: J. M. Armentrout (ed.), Pacific Northwest Cenozoic Biostratig- raphy. Geological Society of America, Special Paper 184. Mon TrorT, P. D. 1810. Conchyliologie systématique et clas- sification méthodique des coquilles. Vol. 2. F. Schoell: Paris. 176 pp. Moore, E. J. 1968. Fossil mollusks of San Diego County. San Diego Society of Natural History, Occasional Paper 15:1- 76. Morcu, O. A. L. 1852-1853. Catalogus conchyliorum quae reliquit D. Alphonso d’Aguirra et Gadea Comes de Yoldi. 8 vols. Hafniae. PALMER, K. V. W. 1977. The unpublished vélins of Lamarck (1802-1809): Illustrations of fossils of the Paris Basin Eo- cene. Paleontological Research Institution: Ithaca, New York. 67 pp., 52 vélins. RAFINESQUE, C. S. 1815. Analyse de la nature, ou tabesau de Puniverse et des corps organisées. Palermo. 224 pp. SQUIRES, R. L. 1987. Eocene molluscan paleontology of the Whitaker Peak area, Los Angeles and Ventura counties. Los Angeles County Natural History Museum, Contributions in Sciences 388:1-93. SQuIRES, R. L. 1991. A new middle Eocene potamidid gastro- pod from brackish-marine deposits, southern California. The Veliger 34(4):354-359. Squires, R. L. & T. A. DEMERE. 1991. A middle Eocene marine molluscan assemblage from the usually nonmarine Friars Formation, San Diego County, California. Pp. 181- 188. In: P. L. Abbott & J. A. May (eds.), Eocene Geologic History San Diego Region. Pacific Section, Society of Eco- nomic Paleontologists & Mineralogists, Book 68. Sz6Ts, E. 1953. Mollusques Eocénes de la Hongrie. I. Les Mollusques Eocénes des environs de Gant. Geologica Hun- garica, Series Palaeontologica 22:1-270. TuHoms, R. E. 1975. Biostratigraphy of the Umpqua Group, southwestern Oregon. Pp. 513-562. In: D. W. Weaver, G. R. Hornaday & A. Tipton (eds.), Paleogene Symposium and Selected Technical Papers; Conference on Future Energy Horizons of the Pacific Coast. Pacific Sections of American Association of Petroleum Geologists, Society of Economic R. L. Squires, 1992 Page 329 Paleontologists & Mineralogists, and Society of Exploration Geologists. TurRNER, F. E. 1938. Stratigraphy and Mollusca of the Eocene of western Oregon. Geological Society of America, Special Paper 10:1-130. Vokes, H. E. 1939. Molluscan faunas of the Domengine and Arroyo Hondo Formations of the California Eocene. Annals of the New York Academy of Sciences 38:1-246. WEAVER, C. E. 1912. A preliminary report on the Tertiary paleontology of western Washington. Washington Geolog- ical Survey, Bulletin 15:1-80. WEAVER, C. E. 1942 [1943]. Paleontology of the marine Ter- tiary formations of Oregon and Washington. University of Washington, Publications in Geology 5 (Parts 1-3): 1-789. WEAVER, C. E. & K. V. W. PALMER. 1922. Fauna from the Eocene of Washington. University of Washington Publica- tions in Geology 1:1-56. WEAVER, D. W. & R. M. KLEINPELL. 1963. Mollusca from the Turritella variata Zone and their chronologic and bio- geographic significance. University of California Publica- tions in Geological Sciences 43 (Part 2):81-161. Woops A. J. C. & L. R. SAUL. 1986. New Neritidae from southwestern North America. Journal of Paleontology 60(3): 636-655. The Veliger 35(4):330-337 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 A New Aeolid (Gastropoda: Nudibranchia) from the Atlantic Coasts of the Southern Iberian Peninsula by J. L. CERVERA Laboratorio de Biologia, Facultad de Ciencias del Mar, Universidad de Cadiz, Apdo. 40, 11510 Puerto Real (Cadiz), Spain J. G. GARCIA-GOMEZ anp P. J. LOPEZ-GONZALEZ Laboratorio de Biologia Marina, Departamento de Fisiologia y Biologia Animal, Facultad de Biologia, Universidad de Sevilla, Apdo. 1095, 41080 Sevilla, Spain Abstract. A new species of aeolid nudibranch, Cuthona willani, is described from the Atlantic coasts of the southern Iberian Peninsula. The swellings of the rhinophores and oral tentacles, the shape of cerata, and the coloration separate C. willani from the remaining Lusitanian, Mediterranean, and Mauretanian known species of the genus. INTRODUCTION A new species of the aeolid genus Cuthona Alder & Han- cock, 1855, has been described recently from the Atlantic coasts of the southern Iberian Peninsula by GARCIA et al. (1991). In this paper, we describe another species of the same genus from the same geographical area. Family TERGIPEDIDAE Thiele, 1931 Cuthona Alder & Hancock, 1855 Cuthona willant Cervera, Garcia-Gomez & Lopez-Gonzalez, sp. nov. (Figures 1-3) Material: Holotype: One specimen, 12 mm in length, col- lected intertidally, El Portil (Huelva, Spain) (37°12'40’N, 7°7'50"W), September 1986. This specimen, which was not dissected, has been deposited in the collections of the Museo Nacional de Ciencias Naturales (MNCN) of Ma- drid, catalogue number 15.05/0763. Paratypes: One specimen, 11 mm in length, collected concurrently with the holotype, has been deposited in the Laboratorio de Biologia Marina (LBM), Departamento de Fisiologia y Biologia Animal, Universidad de Sevilla. One specimen, 3 mm in length, collected by scuBA at 20 m depth in Sagres, Portugal (37°N, 8°55’W), during the International Expedition “ALGARVE 88,” May 1988. This specimen has been deposited in the LBM, Departa- mento de Fisiologia y Biologia Animal, Universidad de Sevilla. A color slide of this living specimen of Cuthona willani is on file at the LBM, Universidad de Sevilla. Description: The body is typically aeolidiform (Figure 1A, B) and slightly narrower than the foot, tapering pos- teriorly in a relatively long and pointed tail. The foot corners are square. The oral tentacles are cylindrical, with a slight swelling in their middle part, and long, nearly as long as the rhinophores (Figure 1A, B, E). The rhino- phores are long, smooth, and slightly enlarged at their base. They have a similar swelling to that of the oral tentacles at the apical half (Figure 1A, B). The cerata are arranged in 10 to 13 dorsolaterally oblique rows on either side of the body (Figure 2A). The postcardial rows of cerata on each side are arranged alternately with regard to those on the opposite side. The ceratal half formula is I 1-2, II 2-3, III 3-4, IV 3-5 (precardial), V 4-5, VI 4-5, VII 4- 6, VIII 3-7, IX 2-6, X 1-4, XI 2-4, XII 3, XIII 3 (postcardial). The cerata have a conspicuous subapical globular enlargement, which is more developed in the larg- est (medial) cerata than in the smallest (lateral) ones. The cerata have another less conspicuous swelling before nar- rowing at the base (Figure 1C). The tips of the cerata may appear rounded (when an animal is undisturbed, the apex of the cnidosac is retracted) (Figure 1D, a) or pointed (when an animal is disturbed, the apex of the cnidosac is extended) (Figure 1D, b). The anus is acleioproct and the genital pore is under the second row of cerata on the right side (Figure 2A). ujeple= Cervera: al 1992 Page 331 b Figure 1 Cuthona willani sp. nov. A. Dorsal view of 11-mm adult specimen. B. Lateral view of 3-mm juvenile specimen. C. Detail of a ceras. D. Variability of the apical shape of the cerata—(a) apex of cnidosac extended, (b) apex retracted. E. Detail of a rhinophore. Key: an, anus; drb, dark reddish brown; e, eye; gar(s), garnet (superficial); hye, hyaline yellow; ir, iridescent red; ot, oral tentacle; r, red; vi, violet; ye(i), yellow (internal); yesh, yellowish. Page 332 The Veliger, Vol. 35, No. 4 4608 10¥m NOL O62 26KV Figure 2 Cuthona willani sp. nov. A. Schematic arrangement of cerata. B. Jaw. C. Detail of the masticatory border of jaw. D. Second (a) and eighth (b) radular teeth of the 11-mm specimen. E. Scanning electron micrograph of some radular teeth of the same specimen. Key: an, anus; gp, genital pore. ele Cenverarcual, 1992 Page 333 Figure 3 Cuthona willani sp. nov. A. Reproductive system. B. Detail of junction of the penial gland with the penis. C. Spawn. D. Detail of the cross-section of the same. Key: agl, albumen gland; am, ampulla; dd, deferent duct; hd, hermaph- roditic duct; mgl, mucous gland; pgl, penial gland; pp, penial papilla; pr, prostate; sr, seminal receptacle; vd, vaginal duct. The Veliger, Vol. 35, No. 4 Page 334 (8861) L ‘(S861) a £ ‘(r86l) ad L (0861) a (8861) L ‘(¢861) & 8 L ‘(0861) & “(hS61) Ad (8861) L ‘(+861) d 8 L (0861) & ‘(rS61) Ad (8861) L ‘(861) d 8 L ‘(0861) & ‘(4S61) Ad ‘(Z) (@r81) H BV (8861) L ‘(S861) a £ ‘(y86l) d % L ‘(0861) (8861) L ‘(r861) d % LL ‘(Z861) d 8 S ‘(0861) 4 (8861) L ‘(r861) d % L (2861) d 8 S ‘(1861) Od *(O861) ea “(O861) & “(PS6L) Ad {sa0ua19jay [loo |yews jesids [yews yes -1ds paynjoauos ‘{109 Je[Nodotwaes peoiq (Z) yey & pue usin) 2uo) [109 pauado |jeus [l09 se[NoMOTWIs peo asIMyPopPIUe BSuljins uoqqis [yews umedg ou ou ou ou ou ou ou saporjuay [e410 pue sasoyd -OuTYs UT Bur[Jams (A]]eIStp ua][oms aq Avw) podeys-qnyo a7earyo 9}JRAP]O podeys-qnyo (saytuaanf ur snoq -[Nq 940ul) [eoTApuTTAD podeys-a,purds podeys-a[purds adeys |e1e49") 9 9 Lb mS ve S L-9 poe S-P vI-cl S Savy, Sb DeS v Smit LY ») S [ES eye199 PRI99 sopnuap JO SMO4 JO SMO1 jessie] yerpaeos jerpaeo -\S0g -o1d ve-Sl Oc-SI Sb-SC Oc> vS-SE SOmiS O08-S9 yo} iepnpey ‘puoyjn’) Jo saioads uelurjaineyy pur ‘uvIuRIJa}IpayAy ‘URTUR]ISN'T IY} JO Saunj}eay dUOS Jo UOsTAeduWOT) SiGe (PS8] ‘yooourpy 2 JOply) vIwjnjsng -y (SPS ‘yooourTPy 2 JIp[y) vusowp +) (€p8l yooourH 2 Jop[y) Vuu22U0I “+ (ZBI “x09 -UPF] 2 JOply) vuvu *y (OP8I ‘Sequog) spre “7 (6€81 “1epoor W saqso.f) v7p70f (POST ‘nse -uoW—) DIINLIDI DUOYIN’) saioadg © Page 335 J. L. Cervera et al., 1992 ‘uMOIg 29 UosdUIOY Ty ‘gq 29 J, ‘uosduo ‘Ad ‘spunuipy 2% asnf{ ‘q 2 [ ‘zawon-eniey 2» Za[ezuor-zad oT ‘e ————_———————————— ee ee ee ed “ed “UMorg “gq ‘yooourH 29 Japry ‘H w® Vv 4 eA eS Se 7 oe” aaa a Apnjs juasaid (1661) DD 8 OT ‘D (8861) L (861) d 72 L ‘(0861) ‘(8261) d ® ld (Z861) d B S (8961) § (7861) d 8 S *(8961) S ‘(1861) OA (Z861) d B S (8961) S (9861) ea (7861) d 8 § (9961) § (Z861) d 8 S *(6961) 9S (9961) § (8861) L ‘(861) d 7 L ‘(Z861) d ®% S (0861) @ {sa0uaI9 jay €-Z Jo [too padeys-dno (Du2z0wD ‘D2 JO yep ATasoyo Sul[quiasar) peasy) payioo AyrepnSasit ‘ury) podeys -Aaupry/prosdiya UN} 3UO Jo [100 repno -Jiotwias /padeys-Aaupry sploipAy jo wia}s 24} punore punom “SUIN} [BIIAIS JO [109 podeys -Agupry/prosdiy]2 usin} 9U0 jo [too umedg ou ou ou ou ou ou ou ou Sapor}Ua} e410 pue sasoyd -OUTYI UI suryaMg -1Y431[s pue padeys-usn 6 + [eotapurfAo 9 t dn yuniq e yim podeys-arpurds 6 Z padeys-qnyp S Z (dn ay) spremo} Sut -iade}) padeys-ajpuids € € aqearyo $ € (saoqe pepunol ‘aseq 2y} 1B uUeY) IIpIM Ajjeotde) padeys-qnyo 8 $ pedeys-qnyo 9 @ padeys-a[purds Cae Gal odeys [elesay B1e199 B1e199 JO SMO jo smou [etpses [etpses -1S0g -d1g OI-8 SI-Fl LI-8 6-€ sapnuap [esse] CC 8S Ic PE-Ce LS-9F y99} Jepnpey WL “LL ‘1a81aquauoyps ‘og ‘UURUNIOg 2 [eYIWIYIS ‘gq WS ‘PYPWYS “g ‘uMOI_ 2 UODIg ‘g 29 Ig ‘JOJ-1OANIg DIED ‘ND ® OT ‘5D ‘sog-zepueusay ‘OA ‘sosaisa ‘aou “ds tunqjam “5 166] ‘Z9WOs)-eloIeDy W® Zajezuor)-zadoT ‘eIoIey) 1uosqwuoy) ‘+ 826] ‘UMOIg WW uo suassaqna (8961 ‘JaYPWUYIG) DypsjsorurU (8961 ‘JaYaUIYyIG) apuo7 (8961 ‘124 -IWIYIS) DIDJIUNdoQ]D Ss oO © (9961 ‘JaYPULYIS) 70/290 «D (9961 ‘PYIWIYIS) vsounLs “+ (9761 ‘onys -OU0(].Q) ana0uas *+y saiadg ee ee ee ee ee ‘panunuor IP AGI IE, Page 336 The ground color of the body is hyaline yellowish white. The body exhibits a very delicate yellow pigmentation and, overlying it, another one of garnet color, less dense than the first. Iridescent red patches are present in some regions of the body, principally on the sides of the head and the right side of the pericardial zone. The oral tentacles have the same ground color as the body from base to the swelling, which has an internal red (violet in the 3-mm specimen) ring. From this level up to the tip, the yellow color is present but the garnet becomes less intense (Figure 1A, B). The rhinophores have a coloration similar to that of the oral tentacles. They exhibit some iridescent red patches at their base, and the enlargements also have an internal red (violet in the 3-mm specimen) ring. From this point up to the tip, the garnet color becomes less intense. The surface of the cerata has a yellow internal pigmentation and a red superficial pigmentation, which becomes denser at both ends of the bigger ceratal enlargements, forming two rings of this color. The yellow pigmentation becomes denser from the upper ring towards the ceratal tip, whereas the red becomes fainter. The ground color of the digestive gland is brown, but at the ceratal base it darkens to reddish brown. The garnet pigmentation is rather violet in the 3-mm specimen. This specimen lacks the lateral patches observed in the remaining specimens, and the tips of the cerata also lack red or violet pigmentation (Figure 1C, E). The preserved specimens conserve the color of the internal rings of the rhinophores and oral tentacles, as well as the dark color of the digestive gland at the base of the cerata. The broad jaws of the 3-mm and 11-mm paratypes have a delicately denticulated masticatory border (Figure 2B, C). The radular formula of the two paratypes is 21 x 0.1.0 and 22 x 0.1.0, respectively. The teeth are horseshoe- shaped with a prominent and strong central cusp; the teeth have 8-10 narrow denticles on either side of them in the larger specimen (Figure 2D, E) but only 5 or 6 denticles in the smaller. The larger specimen has 6 teeth with a bifid central cusp (Figures 2D[b], E). The reproductive system of the 11-mm paratype is il- lustrated in Figure 3A. The long ampulla is U-shaped. The prostate is curved, and the deferent duct is thin and relatively long; the penis terminates in a penial spine (Fig- ure 3B). There is an ovoid penial gland (Figure 3A, B). The seminal receptacle is elongate, with an enlarged duct that connects with the mucous gland near the genital pore. The nacreous albumen gland also connects with the mu- cous gland near the genital pore. Biology: Three egg masses were laid in the laboratory by the El Portil specimens (September 1986). These consisted of 2-3 whorls that formed a cup (Figure 3C). They were triangular in section (Figure 3D) and their surfaces were rough. The width of the string was about 1 mm. Each capsule contained one spherical, white egg. The diameter of the capsules was 136.5-165.7 um and that of the eggs was 97.5-117 wm. The Veliger, Vol. 35, No. 4 Discussion: Our specimens are assigned to the genus Cu- thona Alder & Hancock, 1855, on the basis of their jaws and radulae, despite the similarity in ceratal shape to that of many species of the genus Eubranchus Forbes, 1838. Cuthona caerulea (Montagu, 1804) is the only known species from Lusitanian, Mediterranean, and Maureta- nian waters with red ceratal bands. However, despite the high degree of variability of the color pattern of C. caerulea (PRUVOT-FOL, 1954; BROWN, 1980; SCHMEKEL & PORT- MANN, 1982; THOMPSON & BROWN, 1984; THOMPSON, 1988; CATTANEO-VIETTI et al., 1990), our specimens can- not be considered as specimens of this species because C. caerulea lacks the superficial garnet pigmentation, the ir- idescent red patches present in some regions of the body, and the red-violet internal ring of the swellings of the oral tentacles and rhinophores. Moreover, the cerata of C. cae- rulea are spindle-shaped, not urn-shaped and slightly knobbly, and the oral tentacles and rhinophores have no swellings. Other external and internal differences between C. caerulea and C. willani are presented in Table 1. Etymology: The specific name willani is chosen to give recognition to our colleague Dr. R. C. Willan from the University of Queensland (Australia) for his excellent con- tributions to the knowledge of opisthobranch mollusks. ACKNOWLEDGMENTS We deeply thank Dr. P. Bouchet for his kind invitation to participate in the International Marine Biological Ex- pedition “ALGARVE-88,” in which one of the specimens of Cuthona willani was collected, and the Electron Mi- croscopy Service of the University of Cadiz, mainly Mr. Juan Gonzalez, for providing scanning electron micros- copy facilities. This paper has been partially supported by the project “Fauna Ibérica II” DGICYT PB89-0081. LITERATURE CITED ALDER, J. & A. HANCOCK. 1842. Descriptions of several new species of nudibranchous Mollusca found on the coast of Northumberland. Annals and Magazine of Natural History 13:31-36. ALDER, J. & A. HANCOCK. 1845-1855. A monograph of the British Nudibranchiate Mollusca. Ray Society: London. Part 1 (1845); Part 2 (1846); Part 3 (1847); Part 4 (1848); Part 5 (1851); Part 6 (1854); Part 7 (1855). BALLESTEROS, M. 1980. Contribucién al conocimiento de los Sacoglosos y Nudibranquios (Mollusca: Opisthobranchia). Doctoral Thesis, Universidad de Barcelona. Inédita. BALLESTEROS, M. 1986. Presencia y biologia de Cuthona ocellata (Schmekel, 1966) (Mollusca: Opisthobranchia) en la penin- sula Ibérica. Anales de Biologia 7 (Biologia Animal, 2):5-9. Brown, G. H. 1980. The British species of the aeolidacean family Tergipedidae (Gastropoda: Opisthobranchia) with a discussion of the genera. Zoological Journal of the Linnean Society 69:225-255. CATTANEO-VIETTI, R., R. CHEMELLO & R. GIANNUZZI-SAVELLI. 1990. Atlas of Mediterranean Nudibranchs. La Conchiglia: Rome. J. L. Cervera e¢ al., 1992 FERNANDEZ-OVIES,, C. L. 1981. Contribucion a la clasificacién morfologica de las puestas de los Opistobranquios (Mollusca: Gastropoda). Boletin de Ciencias de la Naturaleza I.D.E.A. 28:3-12. Garcia, F. J., P. J. LOPEZ-GONZALEZ & J. C. GARCIA-GOMEZ. 1991. A new species of Tergipedidae (Gastropoda, Nudi- branchia) from the Atlantic coast of Southern Spain. Journal of Molluscan Studies 57 (T. E. Thompson Memorial Issue): 217-222. Just, H. & M. EpmMunpbs. 1985. North Atlantic Nudibranchs (Mollusca) seen by Henning Lemche. Ophelia Supplement 2:1-150. PicTon, B. E. & G. H. Brown. 1978. Anew species of Cuthona (Gastropoda: Opisthobranchia) from the British Isles. Jour- nal of Conchology 29:345-348. PRuvoT-Fo., A. 1954. Mollusques Opisthobranches. 58. Faune de France. Paul Lechevalier: Paris. SCHMEKEL, L. 1966. Zwei neue Arten der Familie Cuthonidae aus dem Golfe von Neapel: 7rinchesia granosa n. sp. und Page 337 Trinchesia ocellata n. sp. (Gastr. Opisth.). Pubblicazioni della Stazione Zoologica di Napoli 35:13-28. SCHMEKEL, L. 1968. Vier neue Cuthonidae aus dem Mittel- meer (Gastr. Nudibr.): 77inchesia albopunctata n. sp., Trin- chesia miniostriata n. sp., Trinchesia wlonae n. sp. und Catriona maua Marcus und Marcus, 1960. Pubblicazioni della Sta- zione Zoologica di Napoli 36:437-457. SCHMEKEL, L. & A. PORTMANN. 1982. Opisthobranchia des Mittelmeeres. Springer-Verlag: Berlin. 410 pp., pls. 1-36. SCHONENBERGER, N. 1969. Beitrage zur Entwicklung und Morphologie von Trinchesia granosa Schmekel (Gastr. Opis- th.). Pubblicazioni della Stazione Zoologica di Napoli 37: 236-292. THOMPSON, T. E. 1988. Molluscs: Benthic Opisthobranchs. Synopses of the British Fauna. New Series, 8 (2nd ed.). E. J. Brill: Leiden. TuHompson, T. E. & G. H. BRown. 1984. Biology of Opis- thobranch Molluscs. Vol. 2. Ray Society: London. 229 pp. The Veliger 35(4):338-346 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 Two New Species of Helminthoglypta (Gastropoda: Pulmonata) from Southern California, with Comments on the Subgenus Charodotes Pilsbry BARRY ROTH anp F. G. HOCHBERG Department of Invertebrate Zoology, Santa Barbara Museum of Natural History, Santa Barbara, California 93105, USA. Abstract. Two new species of helminthoglyptid land snails, formerly regarded as outlying populations of Helminthoglypta trasku (Newcomb), are described: Helminthoglypta (Charodotes) uvasana from along Grapevine Creek, near Fort Tejon, San Emigdio Mountains, Kern County, and H. (C.) vasquezi from Vasquez Rocks and Agua Dulce Canyon, Los Angeles County, California. Early records of H. trasku from Fort Tejon refer to H. uvasana. Helminthoglypta tejonis Berry, recently considered a subspecies of H. traskii, is restored to species rank. The subgenus Charodotes Pilsbry is redefined and removed from the synonymy of Helminthoglypta, sensu stricto. A preliminary list of taxa included in Charodotes is presented. INTRODUCTION In recent years it has become apparent that the land snail species Helminthoglypta trasku (Newcomb, 1861) as con- strued by virtually all authors (e.g., BARTSCH, 1916; PILsBRY, 1939; ROTH, 1973) is a composite taxon. Several of the taxa that PILsBry (1939), the last reviser of the genus, regarded as subspecies or populations of H. traskii are as distinct anatomically and conchologically as other, recognized, valid species of Helminthoglypta. Helminthoglypta trasku is the type species of the sub- genus Charodotes Pilsbry, 1939. MILLER (1981, 1985) showed that Charodotes was founded on a misconception about the reproductive system of H. traski1. PILSBRY (1939) originally had reported that in Charodotes the wall of the upper chamber of the penis was single, rather than double as in all other species of Helminthoglypta. However, in H. trasku and all other species that have been dissected, a double wall is present. As a consequence, Charodotes is currently regarded as a synonym of Helminthoglypta, sensu stricto (e.g., ROTH, 1987). In this paper we redefine the available name Charodotes on the basis of shell and genital characters, remove it from the synonymy of Helminthoglypta, sensu stricto, and apply it to the clade consisting of the taxa formerly included in H. trasku, along with other related species. Land snails identified as Helminthoglypta trasku have been reported from the vicinity of Fort Tejon, Kern Coun- ty, California, since the early days of biological and geo- logical reconnaissance in the West (NEWCOMB, 1865; CoopER, 1869; BARTSCH, 1916; HANNA, 1927; PILSBRY, 1939). HANNA (1927) examined specimens collected by E. C. Van Dyke in 1927 “on the margin of a wet meadow about a quarter of a mile south of old Fort Tejon” but could find no difference between them and H. traski from the Los Angeles area, except for slightly smaller size. PILsBRY (1939:172-174, fig. 85f) illustrated a specimen collected by the geologist W. M. Gabb at Fort Tejon, and commented that the sculpture was typical for the species. Fort Tejon is about 3 km southeast of the type locality of Helminthoglypta tejonis Berry, 1938 (“two miles above Grapevine Station, old State Highway, Tejon Pass, Kern County”). PILsBRY (1939) regarded H. tejonis as a sub- species of H. traski. The circumstance of apparently typical H. trasku being found within a few kilometers of the large, distinctive H. “‘traskw” tejonis went unremarked by all later authors. In March 1987 we collected Helminthoglypta along Grapevine Creek immediately north of the north boundary of Fort Tejon State Historical Park, including living adult material for dissection. Comparisons with H. traski traski and H. tejonis (which we will argue herein should be B. Roth & F. G. Hochberg, 1992 Character Shell sculpture Vagina Lower chamber of penis Upper chamber of penis Table 1 Characters of the subgenera of Helminthoglypta. Helminthoglypta, sensu stricto various combinations of malleations, spiral striae, and collabral rugae more or less beaded by oblique, divaricating sulci opening into atrium near insertion of atri- al sac short, with small papilla or verge at summit in some species long, slender, club- shaped, cylindrical, or inflated Subgenus Charodotes incised spiral striae, col- labral rugae, and papillation of vari- able strength and distribution opening into atrium near insertion of atri- al sac short, with conic or cy- lindrical verge at summit in some spe- cles long, slender, more or less cylindrical; sometimes with swol- len anterior portion Rothelix wrinkle-like collabral rugae and dense overall papillation; spiral sculpture gen- erally faint opening into atrial sac near posterior end large, sausage-shaped, with post-medial constriction; verge absent short, slender, cylindri- cal or narrower at anterior end Page 339 Coyote varying degrees of papillation; spiral striation present in some species opening into atrium near insertion of atrial sac short, more or less cy- lindrical, with short, broad papilla at summit moderately long, more or less conical, ex- panding to promi- nent swelling at restored to species rank) show that the Grapevine Creek taxon is a new species and it is described below as Hel- minthoglypta uvasana, sp. nov. Another outlying charodotean Helminthoglypta occurs at Vasquez Rocks and in Agua Dulce Canyon, northern Los Angeles County. PILSBRY (1939:fig. 85e) illustrated a specimen from Vasquez Rocks. GREGG (1948:3) referred to it as “‘a desert modification of H. traski [sic] and con- sidered it subspecifically distinct. Wendell O. Gregg and Walter B. Miller collected it alive at both localities during the 1950s and 1960s. We secured additional living material at Vasquez Rocks in 1988. The species is described below as Helminthoglypta vasquezt, sp. nov. The following abbreviations are used: ANSP, Academy of Natural Sciences of Philadelphia; BR, senior author’s collection, San Francisco, California; CAS, California Academy of Sciences; LACM, Natural History Museum of Los Angeles County; SBMNH, Santa Barbara Museum of Natural History; UCMP, University of California (Berkeley) Museum of Paleontology; USNM, U.S. Na- tional Museum of Natural History. SYSTEMATICS Family HELMINTHOGLYPTIDAE Pilsbry, 1939 Helminthoglypta Ancey, 1887 Type species: Helix tudiculata A. Binney, 1843, by original designation. lower end (Charodotes) Pilsbry, 1939 Type species: Helix traskii Newcomb, 1861, by original designation. Diagnosis: Shell umbilicate, subglobose to depressed, sculptured with incised spiral striae, more or less promi- nent collabral rugae, and papillation of variable strength and distribution; upper chamber of penis long (double- walled as usual in the genus), slender, more or less cylin- drical, with swollen anterior portion in some species; lower chamber of penis simple-walled, short, sometimes with conic or cylindrical verge at summit; vagina opening into atrium near insertion of atrial sac. Remarks: Table 1 summarizes the diagnostic morpholog- ical characteristics of the subgenera of Helminthoglypta. Malleated sculpture, characteristic of the ‘““Helmintho- glypta tudiculata series” and “‘Helminthoglypta nickliniana series” (PILSBRY, 1939) (groups within the subgenus Hel- minthoglypta, sensu stricto), is absent, or at most weak and localized, in species of Charodotes. A clothlike pattern formed by strong collabral rugae cut into beads or granules by oblique, divaricating sulci, found in many species of the HZ. nickliniana series, is unknown in Charodotes. The subgenus Rothelix Miller, 1985, has a relatively short and narrow upper penial chamber, a large sausage- shaped lower chamber with a post-medial constriction, and a vagina that opens into the atrial sac near its posterior end. In the subgenus Coyote Reeder & Roth, 1988, the upper chamber of the penis is more or less conical, tapering Page 340 The Veliger, Vol. 35, No. 4 (Charodotes) Scale $0 10 0 10 100 Miles Figure 1 Map of California and (inset) Baja California showing distribution of the subgenus Charodotes and type localities of the two new species. Star = type locality of Helminthoglypta uvasana. Triangle = type locality of Helminthoglypta vasquezi. from a slender summit to a prominent swelling at the lower end. The swelling is caused by a pronounced thickening of the walls of the inner tube, usually accompanied by enlarged glandular pilasters along the widening lumen. The swelling projects into the lower chamber of the penis in the form of a short, broad papilla. The swollen anterior portion of the upper chamber that occasionally occurs in Charodotes (e.g., in Helminthoglypta sanctaecrucis) consists of a sudden widening, involving both inner and outer walls, and is not homologous with the swelling in Coyote. PILsBRY (1939:68, 170) originally distinguished Char- odotes from Helminthoglypta, sensu stricto, mainly on the basis of a purported single, thick wall of the upper chamber of the penis. However, H. traskiz, like all other species of Helminthoglypta, actually has a double-walled upper chamber (MILLER, 1981, 1985). PILSBRY (1939) also men- tioned a common duct of the mucus glands as long as the dart sac or shorter, but as MILLER (1985) noted, this character is variable within populations. Charodotes is here redefined primarily on the basis of its striate and papillose shell sculpture and relatively simple, basically cylindrical, upper penial structure. The following species and subspecies (listed in the order in which they were proposed) are assigned to Charodotes: B. Roth & F. G. Hochberg, 1992 Page 341 Explanation of Figures 2 to 4 Figures 2-4. Helminthoglypta (Charodotes) uvasana, sp. nov., shell; holotype SBMNH 35566, top, apertural, and basal views. Diameter 19.8 mm. Helminthoglypta trasku (Newcomb, 1861) H. t. trasku H. t. coronadoensis (Bartsch, 1916) H. t. isidroensis (Bartsch, 1918) H. t. pacoumensis Gregg, 1931 . ayresiana (Newcomb, 1861) carpentert (Newcomb, 1861) walkeriana (Hemphill, 1911) coelata (Bartsch, 1916) phlyctaena (Bartsch, 1916) willett: (Berry, 1920) sanctaecrucis Pilsbry, 1927 field: Pilsbry, 1930 reediana Willett, 1932 misiona Chace, 1937 . tejonis Berry, 1938 . reederi Miller, 1981 . saluiae Roth, 1987 H. s. salviae H. s. mina Roth, 1987 H. uvasana, sp. nov. H. vasquezi, sp. nov. Gert t rset rTtts Figure 1 depicts the distribution of Charodotes, based on these taxa. The following species, which PILsBry (1939) included in Charodotes, may also prove to belong to the subgenus: Helminthoglypta proles (Hemphill in W. G. Bin- ney, 1892); H. ferrissi Pilsbry, 1924; H. euwomphalodes Ber- ry, 1938; H. ingles: Berry, 1938; H. liodoma Berry, 1938; H. stager: Willett, 1938. Helminthoglypta hanna: Pilsbry, 1927, from Isla Guadalupe, Baja California, Mexico, may also belong to Charodotes. Helminthoglypta petricola (Berry, 1916) and its subspecies, included by PILSBRY (1939) in Charodotes, belong to the subgenus Coyote (REEDER & ROTH, 1988). In the species here assigned to Charodotes, incised spiral sculpture is more prominent than papillation. In Hel- minthoglypta avus (Bartsch, 1916), Helminthoglypta callis- toderma (Pilsbry, 1917), and Helminthoglypta orina Berry, 1938, the penis is cylindrical with a globose swelling at the lower end of the upper chamber, and papillation is more prominent than incised spiral striae. Pending addi- tional study, we exclude this group of species from Char- odotes. Helminthoglypta, sensu stricto, as recognized here, is a rather heterogeneous group, which may be subdivided as a result of additional studies now in progress. Helminthoglypta (Charodotes) uwvasana Roth & Hochberg, sp. nov. (Figures 2—5) Epiphragmophora traski trasku (Newcomb): BARTSCH, 1916: 613 (in part). Helminthoglypta trasku (Newcomb): HANNA, 1927:32-34. Helminthoglypta traski [sic] (Newcomb): PILSBRY, 1939:172- 174 (in part), fig. 85f. Non Helix trasku NEWCOMB, 1861:91. Diagnosis: A medium-sized Helminthoglypta with solid, compact, depressed-helicoid, umbilicate shell sculptured with fine spiral striae; granulation present below suture of early whorls, in umbilicus and behind lip; body whorl tightly coiled, scarcely descending. Description—shell of holotype: Shell (Figures 2-4) of medium size for genus, solid, compact, moderately glossy, depressed-helicoid, umbilicate; umbilicus contained 8.25 times in major diameter. Spire low-conic; whorl profile moderately convex; suture distinctly impressed. Embryonic whorls 1.7, narrower than immediately following teleo- conch whorl; nuclear tip smooth, thereafter granulose with low, coarse, irregular collabral rugae and scattered papil- lae; zone below suture densely granulose. Early teleoconch whorls with low, convex-forward, collabral rugae; minor granulation below suture; sparse and inconspicuous pap- illation; and, from third whorl on, fine, incised spiral striae. Striae weak and discontinuous at first, becoming stronger and continuous on later whorls. Striae prominent on body whorl, continuing over base into umbilicus. Base moder- ately inflated, tumid around umbilicus, granulose within Page 342 Figure 5. Helminthoglypta (Charodotes) uvasana, sp. nov., reproductive sys- tem, drawn from projection of stained whole mount; ovotestis and albumen gland region omitted; paratype SBMNH 35567. Abbreviations: as, atrial sac; ds, dart sac; ec, epiphallic caecum; ep, epiphallus; go, genital orifice; Ip, lower chamber of penis; mb, mucus gland bulbs; mg, part of mucus gland membranes; pr, penial retractor muscle; pt, lower part of prostate; sd, sper- mathecal duct; sp, spermatheca; sv, spermathecal diverticulum; up, upper chamber of penis; ut, part of uterus; vd, vas deferens. The Veliger, Vol. 35, No. 4 umbilicus and behind lip. Body whorl tightly coiled, scarcely descending except just before aperture, not constricted be- hind lip. Aperture auricular, moderately oblique; plane of peristome at angle of 35° to vertical; lip turned outward, narrowly expanded, scarcely reflected except at columellar insertion. Upper limb of peristome produced and slightly downturned. Inner lip weakly encroaching on umbilicus. Parietal callus thin, granulose, with sculpture of parietal wall showing through. Shell pinkish tan under a yellowish brown periostracum; with a 1.0-mm-wide russet spiral band on shoulder (prolonging trajectory of suture) with pale zones of equal width (lower zone more conspicuous) on either side of band. Diameter (exclusive of expanded lip) 19.8 mm, height 11.5 mm, width of umbilicus 2.4 mm, whorls 5.7. Measurements and counts of material at hand (n = 38): Range of adult shell diameter 17.4-23.5 mm. Number of whorls 5.3-6.4; number of embryonic whorls 1.7-2.0. Um- bilicus contained 7.5-9.0 times in shell diameter. Soft anatomy: Mantle over lung clear buff, about 30% covered with irregular black spots. Reproductive system (Figure 5) typical of Charodotes. Atrium short and broad. Atrial sac cylindrical, about twice as long as vagina, with spherical dart sac at upper end, lacking a glandular collar. Mucus gland bulbs of moderate size, joined by slender, Y-shaped common duct. Duct of spermatheca slender throughout its length, bearing diverticulum of greater di- ameter, about 1.5 times as long as spermathecal duct above its origin. Penis with short, conical lower chamber (ap- proximately as long as vagina) and long, double-walled upper chamber, cylindrical or slightly wider at summit, leading to epiphallus of same diameter. Verge absent. Epi- phallic caecum about as long as penis plus epiphallus. Type material: Holotype: Santa Barbara Museum of Natural History, SBMNH 35566 (shell, whole mount of mantle tissue, and stained whole mount of reproductive system), CALIFORNIA: Kern County: along Grapevine Creek in Castaic Valley, immediately north of boundary of Fort Tejon State Historical Park (projected SE% sec. 9 to NE% sec. 16, T. 9 N, R. 19 W, San Bernardino Base and Meridian), elevation approximately 940 m (3100 ft); under downed log of Quercus lobata. W. B. Miller, F. G. Hoch- berg, B. Roth coll., 9 March 1987. Paratypes: SBMNH 35567 (10 shells and stained whole mount of reproductive system), from same locality as ho- lotype, under downed oak logs in leaf litter, in brush, and in wood rat nests. Additional paratypes deposited in ANSP, CAS, BR, LACM, and USNM. Referred material: CALIFORNIA: Kern County: Fort Te- jon (ANSP 10697, BR 448, CAS 036312, CAS 051338, UCMP 2497, USNM 58523); near Old Fort Tejon (BR 774, BR 1539, CAS 036330); “Tejon” [sic] (ANSP 10698, W. M. Gabb coll., one specimen figured by PILsBry, 1939: fig. 85f); Grapevine Creek at Fort Tejon (SBMNH 35568); B. Roth & F. G. Hochberg, 1992 Page 343 Explanation of Figures 6 to 8 Figures 6-8. Helminthoglypta tejonis Berry, shell; holotype SBMNH 34216, top, apertural, and basal views. Diameter 30.3 mm. Grapevine Canyon, 0.25 mi [0.4 km] S of Old Fort Tejon (CAS 036290); 0.5 mi [0.8 km] N of Tejon Inn (LACM 114626). Los Angeles County: Oak Flat Ranger Station, 12 mi [19 km] N of Castaic (CAS 036300). Remarks: Helminthoglypta wvasana somewhat resembles presumed topotypic Helminthoglypta traski traskii from Point Fermin, Los Angeles County. The shells of both species are robust and run through about the same range of size and shape, but the incised spiral sculpture of H. uvasana is finer (7 striae/mm on the last % of the body whorl, compared to 4-5 striae/mm at the same location on H. traski). In H. trasku the spermathecal diverticulum is 1.5-2 times as long as the spermathecal duct above its origin. The lower chamber of the penis is longer than that of H. uvasana. Helminthoglypta uvasana differs from H. carpenteri of the southwestern San Joaquin Valley in having a black- spotted mantle when adult. In H. carpenteri the mantle over the lung is uniform brownish gray with a black trans- verse line behind the mantle collar; small juveniles some- times have black spots. The base of the spermathecal duct of H. carpenteri is cavernous; the spermathecal diverticu- lum is only slightly longer than the spermathecal duct above its origin. The collabral rugae on the shell of H. carpenteri are more or less granulose. Helminthoglypta tejonis occurs approximately 3 km to the northwest (the type locality probably is in the projected SE™% of sec. 32, T. 10 N, R. 19 W), and about 21 km west of that, along San Emigdio Creek. The shell of H. tejonis (Figures 6-8) is larger (26.0-31.2 mm in diameter), thin, broadly depressed-helicoid, with 6.25-7.25 whorls. The spire is broadly conic to low-domed, the suture impressed, the whorls shouldered and somewhat flattened. The pe- riphery is broadly rounded, sloping toward the base. In- cised spiral striae first appear on the fourth whorl. Pap- illation is faint to obsolete, confined to the early neanic whorls, and sometimes replaced by minute pits on later whorls. The umbilicus is contained 9-10 times in the shell diameter, about “4 covered by the inner lip. The base of the embryonic whorls is visible in the umbilicus, centered within a regular spiral; in H. uvasana the pit of the um- Explanation of Figures 9 to 11 Figures 9-11. Helminthoglypta vasquezt, sp. nov., shell; holotype SBMNH 35569, top, apertural, and basal views. Diameter 16.4 mm. Page 344 go l1Omm Figure 12 Helminthoglypta vasquezi, sp. nov., reproductive system, drawn from projection of stained whole mount; paratype SBMNH 35570. Abbreviations as in Figure 5. bilicus is oblique and the embryonic whorls are not readily visible. The spermathecal diverticulum in H. tejonis is about as long as the spermathecal duct above its origin. The lower chamber of the penis is rather broadly cylin- drical and approximately twice as long as the vagina; its upper half is nearly filled by a thick, cylindrical verge. Helminthoglypta tejonis has (in common with H. phlyc- taena and H. willetti) a glossy, tumid, broadly depressed- helicoid shell generally more than 25 mm in diameter; the spiral striae are mostly shallow, and papillation is confined The Veliger, Vol. 35, No. 4 to the early neanic whorls. In H. traski the shell is matte to moderately glossy and rarely exceeds 24 mm in diameter. The spiral striae are coarse and strongly impressed. In subspecies H. traski pacoimensis and H. trasku isidroensis, papillation extends onto the body whorl; in H. t. traskzi, papillation fades out by the third or fourth whorl. The spermathecal diverticulum in H. traski is 1.5-2 times as long as the spermathecal duct above its origin. The lower chamber of the penis is cylindrical, relatively long (up to twice as long as the vagina), and sometimes slightly flaring at the base. A verge is absent. The shell and reproductive system distinctions cited above (especially the presence of a verge in Helminthoglypta te- jomis and not in H. traski) lead us to restore H. tejonis to species rank, as originally proposed for it by BERRY (1938). The natural vegetation of the San Emigdio Mountains in the vicinity of Fort Tejon is valley oak (Quercus lobata) savanna, grading locally to chaparral (KUCHLER, 1977). Along Grapevine Creek we found the new species under logs and leaf litter among Quercus lobata, nettle (Urtica holoserica), and poison-oak (Rhus diversiloba). Etymology: Latin, wvasana, pertaining to Canada de las Uvas, a former name for Grapevine Creek (cf. BREWER, 1930). The name “grapevine shoulderband” is proposed for purposes of the American Fisheries Society list of the common names of mollusks (see TURGEON et al., 1988) and other administrative uses. Helminthoglypta (Charodotes) vasquezt, Roth & Hochberg, sp. nov. (Figures 9-12) Helminthoglypta traski [sic] (Newcomb): PILsBRY, 1939:172- 174 (in part), fig. 85e. Non Helix traskii NEWCOMB, 1861:91. Diagnosis: A small Helminthoglypta with thin, delicate, depressed, narrowly umbilicate shell, with fine spiral striae first appearing on last % of penult; body whorl flaring behind lip, scarcely descending. Description—shell of holotype: Shell (Figures 9-11) small for genus, thin and delicate, moderately glossy, de- pressed, umbilicate; umbilicus contained 10.0 times in ma- jor diameter. Spire scarcely elevated, low-domed; whorl profile moderately flattened; suture impressed. Embryonic whorls 1.6; nuclear tip smooth, thereafter granulose with low, coarse, irregular collabral rugae and scattered papil- lae. Early teleoconch whorls with fine, overall, radial wrin- kling and crude, convex-forward, collabral rugae, strongest below suture. From third whorl on, minute, more or less diagonally arranged papillation superimposed, fading out on body whorl except for few scattered papillae. Fine, incised spiral striation first appearing on last % of penult, more prominent on body whorl, continuing over base into umbilicus. Base rather deep, tumid around umbilicus, granulose within umbilicus and behind lip. Body whorl B. Roth & F. G. Hochberg, 1992 expanding rapidly, flared behind lip, scarcely descending except just before aperture, not constricted behind lip. Aperture ovate, moderately oblique; plane of peristome shallowly sinuous in profile, at angle of 45° to vertical; lip thickened but not markedly turned outward, reflected only below columellar insertion. Upper limb of peristome pro- duced and slightly downturned. Inner lip weakly en- croaching on umbilicus. Parietal callus very thin, granu- lose, with sculpture of parietal wall showing through. Shell translucent, pale pinkish tan under a yellowish olive peri- ostracum; with a 0.5-mm-wide russet spiral band on shoul- der (prolonging trajectory of suture), indistinctly bordered by pale zones (upper zone 0.5 mm wide, lower zone 1.0 mm wide). Diameter (exclusive of expanded lip) 16.4 mm, height 9.4 mm, width of umbilicus 1.6 mm, whorls 5.4. Measurements and counts of material at hand (n = 66): Range of adult shell diameter 14.6-19.0 mm. Number of whorls 4.5—5.4; number of embryonic whorls 1.4-1.9. Um- bilicus contained 7.5-10.0 times in shell diameter. Soft anatomy: Mantle over the lung clear buff with black maculation. Reproductive system (Figure 12) typical of Charodotes. Atrium short and broad. Atrial sac cylindrical- conic, approximately twice as long as vagina, with a spher- ical dart sac at upper end, lacking a glandular collar. Mucus gland bulbs rather small, joined by Y-shaped com- mon duct. Duct of the spermatheca slender throughout its length, bearing a diverticulum of greater diameter, about 1.25 times as long as spermathecal duct above its origin. Lower chamber of penis short (slightly longer than vagi- na), narrowly cylindrical, and flaring at base. Double- walled upper chamber moderately long, widening slightly toward summit, leading to epiphallus of same diameter. Verge absent. Epiphallic caecum about as long as penis plus epiphallus. Type material: Holotype: Santa Barbara Museum of Natural History, SBMNH 35569 (shell, whole mount of mantle tissue, and stained whole mount of reproductive system), CALIFORNIA: Los Angeles County: Vasquez Rocks County Park, in small, north-facing amphitheater south of road, west of most prominent outcrops (NE% SW sec. 26, T.5 N, R. 14 W, San Bernardino Base and Meridian); under clump of Yucca whipple:. W. B. Miller, J. D. Good- man, F. G. Hochberg, B. Roth coll., 12 February 1988. Paratypes: SBMNH 35570 (12 shells and stained whole mount of reproductive system), from same locality as ho- lotype. Additional paratypes deposited in ANSP, BR, CAS, LACM, and USNM. Referred material: CALIFORNIA: Los Angeles County: Vasquez Rocks (BR 781, CAS 036791, CAS 036792, CAS 036795, LACM 65520, LACM 114608, SBMNH 35571, SBMNH 35572, SBMNH 35573, SBMNH 35574); Vas- quez Rocks, off Mint Canyon highway, west end about 3 mi [4.8 km] south of highway under roots of yucca (ANSP 157180, one specimen figured by PitsBry, 1939:fig. 85e); Page 345 ridge on N side of Escondido Canyon, Vasquez Rocks County Park, in Yucca whipple: clumps (BR 1611); Agua Dulce Canyon 1.5-2.1 mi [2.4-3.4 km] from junction with Soledad Canyon (SBMNH 35575, SBMNH 35576, SBMNH 35577, SBMNH 35578, SBMNH 35579, SBMNH 35580, SBMNH 35581, SBMNH 35582). Remarks: Helminthoglypta vasquezi differs from H. trasku and H. uvasana in that the shell is thin and delicate with fine spiral striation that does not appear until the last part of the penultimate whorl. The striation in H. traskii is coarser and present by the third whorl. The umbilicus of H. trasku is contained 10-12 times in the shell diameter. The spermathecal diverticulum in H. traskii is 1.5-2 times as long as the spermathecal duct above its origin. Helminthoglypta vasquezi resembles H. salviae from the vicinity of Frazier Park, Kern County, and Quatal and Apache canyons, Ventura County, in having a depressed shell with spire scarcely elevated and a pit-like umbilicus less than one-third covered by the inner lip. The shell of H., salviae is thin but not especially delicate; the collabral rugae are smooth or partly broken up into rows of granules; and the body whorl is tightly coiled throughout, rather than rapidly expanding and flaring behind the aperture as in H. vasquezt. The natural vegetation at the type locality is semi-desert chaparral, including Adenostoma fasciculatum, Juniperus californica, and extensive patches of Yucca whipplet. Etymology: The species is named for the outlaw and folk hero Tiburcio Vasquez (born 1835, hanged 1875), who plied his trade in the Vasquez Rocks area during the 1870s. The name “Vasquez shoulderband” is proposed for pur- poses of the American Fisheries Society list of the common names of mollusks (see TURGEON et al., 1988) and other administrative uses. ACKNOWLEDGMENTS Walt Miller participated in the field work, prepared the whole mounts and drawings of reproductive systems, and discussed helminthoglyptid systematics with us. John D. Goodman assisted with field work and located a population of live adult snails at Vasquez Rocks County Park. Miller and Dick Reeder critically read the manuscript. Gene Coan furnished historical literature. Ken Heartsill provided as- sistance and information on Fort Tejon State Historical Park. Frank T. Hovore issued a collecting permit and Ranger Mike Sharp assisted in the field at Vasquez Rocks County Park, a North Region Natural Areas Park under the jurisdiction of the Los Angeles County Department of Parks and Recreation. LITERATURE CITED BartscH, P. 1916. The Californian land snails of the Epi- phragmophora trasku group. Proceedings of the U.S. National Museum 51(2170):609-619, pls. 114-117. Page 346 Berry, S.S. 1938. Four new Californian helicoid snails. Jour- nal of Entomology and Zoology 30(1):17-25. BREWER, W.H. 1930. Up and Down California in 1860-1864. The Journal of William H. Brewer, Professor of Agriculture in the Sheffield Scientific School from 1864 to 1903. Edited by F. P. Farquhar. Yale University Press: New Haven. xxx + 601 pp. Cooper, J. G. 1869. On the distribution and localities of West Coast helicoid land snails, &c. American Journal of Con- chology 4(4):211-240. Grecc, W. O. 1948. Helicoid snails of the desert regions of California. Minutes of the Conchological Club of Southern California 81:1-7. Hanna, G D. 1927. Helminthoglypta traskii (Newcomb) at “Fort Tejon,” Kern County, California. Nautilus 41(1):32. KUCHLER, A. W. 1977. The map of the natural vegetation of California. Pp. 909-939, map. Jn: M. G. Barbour & J. Major (eds.), Terrestrial Vegetation of California. John Wi- ley and Sons: New York. MILLER, W. B. 1981. Helminthoglypta reederi spec. nov. (Gas- tropoda: Pulmonata: Helminthoglyptidae) from Baja Cali- fornia, Mexico. The Veliger 24(1):46-48. MILLER, W. B. 1985. A new subgenus of Helminthoglypta (Gastropoda: Pulmonata: Helminthoglyptidae). The Veliger 28(1):94-98. NeEwcoms, W. 1861. Descriptions of six new species of shells. Proceedings of the California Academy of Natural Sciences 2:91-94. The Veliger, Vol. 35, No. 4 NEwcoms, W. 1865. Catalogue of helices inhabiting the west coast of America, north of Cape St. Lucas, and west of the Rocky Mountains; together with remarks upon some of the animals, and their special distribution. American Journal of Conchology 1(4):342-350. Pitssry, H. A. 1939. Land Mollusca of North America (north of Mexico). Academy of Natural Sciences of Philadelphia, Monograph 3, 1(1):1-573. REEDER, R. L. & B. ROTH. 1988. A new subgenus of Hel- minthoglypta (Gastropoda: Pulmonata: Helminthoglyptidae) with the description of a new species from San Bernardino County, California. The Veliger 31(3/4):252-257. RoTH, B. 1973. The systematic position of Helminthoglypta trasku fieldi Pilsbry, 1930 (Gastropoda: Stylommatophora). Bulletin of the Southern California Academy of Sciences 72(3):148-155. Rotu, B. 1987. Anew and polytypic species of Helminthoglypta (Gastropoda: Pulmonata) from the Transverse Ranges, Cal- ifornia. The Veliger 30(2):184-189. TurRGEON, D. D., A. E. BOGAN, E. V. COAN, W. K. EMERSON, W.G. Lyons, W. L. PRATT, C. F. E. ROPER, A. SCHELTEMA, F. G. THOMPSON & J. D. WILLIAMS. 1988. Common and scientific names of aquatic invertebrates from the United States and Canada: Mollusks. American Fisheries Society Special Publication 16. viii + 277 pp., 12 pls. The Veliger 35(4):347-357 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 Seasonal Variation in the Reproductive Organs of Two Populations of Caracolus caracolla (Linné) (Pulmonata: Camaenidae) in Puerto Rico by PATRICIA MARCOS! Department of Biology, University of Puerto Rico, Rio Piedras, Puerto Rico 00931 Abstract. Environmental factors such as temperature, water, and day length may influence the reproductive cycle of pulmonates. To determine if environmental conditions influence the reproductive seasonality of the pulmonate Caracolus caracolla (Linné), the histology of the reproductive system from members of two populations was compared. One population of snails was from a rain forest and the other was from a dry coastal plain. Four reproductive organs were studied: the ovotestis, hermaphroditic duct, spermatheca, and albumen gland. The snails were collected monthly and their organs were dissected, measured (size and mass), and sectioned for microscopical examination. Histological and morphological changes indicated seasonal variation in the reproductive organs of Caracolus caracolla from both popu- lations. Statistical analyses show a significant difference in almost all of the morphological measurements of the shells and reproductive organs within and between each population. The gametes were fully developed, in both populations, by April and May. It is possible that the snails from both populations are mating continually throughout the year, since spermatozoa were always present in the hermaphroditic duct. Nevertheless, according to the activity of the spermatheca, there are some peaks of mating activity in December for the dry coastal plain population and in February for the rain forest population. The onset of mating was correlated with an increase in precipitation, especially in the rain forest population. These results incorporate changes that occurred during one year. To determine if these changes occur cyclically, the study would need to be extended by several years. INTRODUCTION Environmental factors such as temperature (BOUILLON, 1956; WOLDA, 1967; RIDDLE, 1983; Tompa, 1984), water (RIDDLE, 1983; SOLEM & CHRISTENSEN, 1984; TOmpPaA, 1984), and day length (HENDERSON & PELLUET, 1960; SOKOLOVE & McCRONE, 1978; Tompa, 1984) can regulate some physiological aspects of terrestrial pulmonates. The possible effects of these external factors on the reproductive cycle has not been studied extensively at the histological level. In the present study the annual reproductive cycles of two populations of the terrestrial pulmonate Caracolus caracolla (Linné, 1758) were analyzed. The morphology and histology of the reproductive system were studied ' Present address: P.O. Box 4135, San Juan, Puerto Rico 00905. throughout the year to determine if the reproductive organs undergo seasonal variation. Studies on terrestrial pulmonates have revealed a sea- sonal variation in the morphology of their reproductive organs. SOLEM & CHRISTENSEN (1984) worked with Aus- tralian camaenids and concluded that there is a seasonal variation in the reproductive activity of mature adults. They determined reproductive activity by analyzing the seasonal variation of the size of the reproductive organs. HEATWOLE & HEATWOLE (1978) studied the annual cycle of reproduction in the camaenid Caracolus caracolla by determining the seasonal occurrence of copulation and ovi- position and the temporal changes in weight of the albumen gland. They compared two Puerto Rican populations, one living in a rain forest and the other in a dry coastal area, and observed two different seasonal patterns in the glan- dular weights of the populations. Page 348 The effect of environmental factors, such as temperature and illumination, on the function of the reproductive or- gans was studied at the histological level by SMITH (1966). He observed that a “critical point” in the maturation of the reproductive tract of the pulmonate Avion ater occurs when spermatozoa from the ovotestis enter the hermaph- roditic duct. This critical point acts as a trigger for the maturation of the rest of the tract; no external factor has an effect on subsequent maturation of the organs, even though before this critical point, during spermatogenesis, external factors may have a large effect. He mentioned that a trigger mechanism of control, probably of neurose- cretory nature, must be operating in the pulmonate. SMITH (1966) observed that the reproductive systems of slugs that were hatched and reared in the laboratory under temper- atures of 18°C, natural illumination, and abundant fresh food matured seven months earlier than the reproductive systems of wild slugs. The timing of the onset of the breeding season for the pulmonate Helminthoglypta arrosa (Binney, 1855) may de- pend on food and water availability but is not dependent on temperature variations (VAN DER LAAN, 1980). Two studies show that high environmental humidity is essential for oviposition (CARRICK, 1942) and mating (RUNHAM & LARYEA, 1968) in the slug Agriolimax reticulatus. The reproductive tract of terrestrial and freshwater gas- tropods is under endocrinological control (PELLUET & LANE, 1961; PELLUET, 1964; BOER & JOosE, 1975; TAKE- DA, 1979). The maturation of oocytes in pulmonates is controlled by a hormone secreted by the dorsal bodies, a structure associated with the cerebral ganglia (GERAERTS & JoosE, 1975; WIJDENES & RUNHAM, 1976). These en- docrinological mechanisms are affected by environmental factors such as photoperiod and humidity (RUNHAM & LARYEA, 1968; SMITH, 1966; SOKOLOVE & MCCRONE, 1978), which have a greater effect on spermatogenesis than on oogenesis. This study presents the first histological comparison of the reproductive system of the camaenids living under dif- ferent environmental conditions of temperature and hu- midity. A pattern of reproductive activity was established for the pulmonate Caracolus caracolla and could be used for the study of other pulmonates. MATERIALS anpD METHODS The endemic terrestrial pulmonate Caracolus caracolla (Linné, 1758) is abundant in the forests of the central mountains as well as in the humid lowland forests of Puerto Rico. The shell is large (55-60 mm in diameter), dextrally coiled, and light to dark brown. Study Sites Two populations were studied. The El Yunque popu- lation lives in a subtropical rain forest (Sierra Palm Forest) located on the Mt. Britton Trail of the Luquillo Experi- mental Forest, Puerto Rico, at an elevation of about 850 The Veliger, Vol. 35, No. 4 m. The relative humidity of the region is over 90% through- out the year, the temperature fluctuates from 13 to 18°C, and the total annual precipitation is over 2.5 m (HEATWOLE & HEATWOLE, 1978). In 1988, the annual temperature range for E] Yunque was from 17 to 20°C, and the total precipitation for the year was 44.95 m (NATIONAL OCEAN- Ic AND ATMOSPHERIC ADMINISTRATION [NOAA], 1989). The Loiza population lives in a dry coastal plain located at sea level on the northern coast of Puerto Rico. Rainfall in this area is sparser and more seasonal than in the moun- tains. The annual temperature range for 1988 was from 26 to 29°C, and the total precipitation for the year was 24.3 m (NOAA, 1989). Collection and Dissection of Specimens Ten mature specimens from each population were col- lected monthly for a period of one year beginning in Sep- tember 1987. The criterion used to determine maturity was the thickness of the lip of the shell (reflected shell lip). A hygrometer was used to measure temperature and rel- ative humidity in the area immediately surrounding the animals each month. The snails were relaxed by placing them in a freezer (5°C) for about half an hour. After being removed from the freezer, the height and width of the shell were mea- sured. The snails were then placed in a dissecting dish that contained a physiological saline solution of 0.75% NaCl to prevent tissue damage. Viewed through an Olympus SZ dissecting microscope, the reproductive system was excised with a scalpel through a ventral lengthwise cut on the body wall. The ovotestis, hermaphroditic duct, spermatheca, and the albumen gland (Figure 1) were removed from the tract and measured (length and width) with a vernier caliper. The ovotestis was difficult to remove because it is embedded in the di- gestive gland. For this reason a scalpel was used to cut the edges of the ovotestis, although parts of the digestive gland remained attached to it. The mass of each organ except the ovotestis was determined by placing it in a separate small petri dish that contained a pre-measured mass of physiological saline and weighed to the nearest 0.001 g using a Sartorius balance, model L420D. The organs were then fixed in Bouin’s fluid. Histological Methods After fixation, the tissues were dehydrated, cleared in benzene, and embedded in Paraplast. Sections were cut at 10 wm and stained with Harris’ modified hematoxylin (Humason, 1979). Some sections of the reproductive or- gans were stained with Milligan Trichrome (HUMASON, 1979), which stains muscular and connective tissue. Histological Examination Sections through different regions of the reproductive organs of each specimen were examined with a light mi- P. Marcos, 1992 Figure 1 The reproductive system of the hermaphroditic snail Caracolus caracolla. Key: ag, albumen gland; dg, digestive system; ep, epi- phallus; go, genital opening; hd, hermaphroditic duct; 0, ovotestis; ov, oviduct; p, penis; s, spermatheca; sp, spermathecal duct; v, vagina; vd, vas deferens (scale = 10 mm). Top of the figure is anterior. croscope. Observations were recorded to compare the his- tology of the monthly samples and to determine if there were seasonal changes in the activity of these organs. Germinal cells (oogonia and spermatogonia) at different stages of development and somatic cells were identified in the acini of the ovotestis. The lumen of the seminal vesicles of the hermaphroditic duct and the lumen of the sper- matheca were analyzed to determine the presence of sper- matozoa, since this is a sign of reproductive activity. The albumen gland was studied only at a morphological level (size and mass) because the active albumen glands are brittle and difficult to section. Statistical Analysis The statistical analysis gave a quantitative measure of the successive changes in the reproductive organs during the year. Thirteen measurements were taken from each snail. These included the lengths and widths of the shell and ovotestis and the lengths, widths, and masses of the hermaphroditic duct, spermatheca, and albumen gland. One-way and two-way analyses of variance (ANOVA), Page 349 Temperature [°C] —— El Yunque — - Loiza CE Mi Ae Md eer Anes. 0" Ne D 550 E 500 2 450 =] 2 400 — 8 350 Q, 300 m 250 A, 200 150 B 100 de MAM doa SS © he 100 Relative Humidity (%] —— El Yunque — - Loiza O Jie MEAs Midi Ay Si ON) iD Figure 2 A. Annual (1988) temperature data for El Yunque and Loiza. B. Annual (1988) precipitation data for El Yunque and Loiza. C. Annual (1988) relative humidity for El Yunque and Loiza. E] Yunque ( ), Loiza (- - -). Page 350 —— Figure 3 Photomicrograph of separate male (ma) and female (fa) acini. The male acini tended to be located near the center of the ovotestis (scale = 1 mm). Principal Component analyses, and Pearson Correlation Coefficient comparisons were performed using the SAS program. RESULTS Climatological Data The patterns of monthly precipitation for 1988 were similar for the two sites, but precipitation was consistently greater in E] Yunque (Figure 2). Monthly temperatures in 1988 were also similar, but the temperature at E] Yunque was always 6-7°C lower than at Loiza. The patterns of relative humidity were similar to the precipitation data. The average annual value for the El Yunque population was 89% and for Loiza 73%. Morphology and Histology of the Reproductive Organs Ovotestis: The ovotestis is embedded in the digestive gland and consists of many irregularly shaped acini (Figure 3). The Veliger, Vol. 35, No. 4 Each acinus is bordered by a layer of cells, the germinal epithelium, and is separated from the other acini by con- nective tissue. The germinal epithelium is one cell thick and consists of columnar cells with basal nuclei. Separate acini in the ovotestis give rise to spermatozoa and ova (Figures 3, 4A-C). The process of oogenesis and spermatogenesis begins in the germinal epithelium and ends in the lumen of the acini, where the mature gametes are liberated. The spermato- gonia develop into spermatocytes (Figure 4B), which be- come attached to the Sertoli cells. Mature spermatozoa are released from the Sertoli cells and remain in the lumen until they are carried to the hermaphroditic duct for stor- age. The oocytes develop in the wall of the acinus (Figure 4C). During some months of the year, yellowish cells were seen near the developing oocytes. These cells were probably follicular cells that surround the oocyte. A mature oocyte is round; its cytoplasm and nucleus stain very dark (Figure 4C). The lumen of each acinus collects the mature gametes and connects to the hermaphroditic duct via collecting tu- bules that transport the mature gametes. Hermaphroditic duct: The hermaphroditic duct has three distinctive parts. The proximal part of the duct to the ovotestis does not store sperm. The next part, the seminal vesicles, have diverticula, which function to store auto- sperm. The distal part of the duct opens into the talon or fertilization pouch, where oocytes are fertilized by allo- sperm. The sections that were analyzed in this study came from the seminal vesicles (Figure 5). About half of the cells comprising the epithelium of the seminal vesicles are ciliated. The ciliated epithelium has narrow cells with basal nuclei and the rest of the epithelium is cuboidal with central nuclei (Figure 6). The vesicles are bordered by circular and longitudinal muscle fibers. Spermatheca: The spermatheca, or bursa copulatrix, has a large lumen and a wall composed of three layers (Figures 7, 8). The layer that borders the lumen consists of two different types of cells, one of which has microvilli at the apical end. The middle layer consists of longitudinal and circular muscle fibers with some pigment cells in between. The third layer is an epithelium that borders the external wall. Albumen gland: The albumen gland is the organ that fluctuates the most in size and color. When inactive the organ is beige and short (15 mm long); when active, it is an intense yellow and two or three times larger. The gland is a tubular, bean-shaped organ that consists of branching tubules separated by connective tissue. The walls of the tubules have secretory cells that have basal nuclei and rounded granules during some months of the year. The tubules lead into a number of small ducts, which unite to form a central duct (Figure 9). The central duct is lined by a ciliated epithelium that varies from cuboidal to co- lumnar. Figure 4 A. Photomicrograph of male acini with spermatogonia and primary spermatocytes, early active stage of the ovotestis (O-1) (scale = 1 mm). B. Acinus with spermatocytes and spermatids attached to the Sertoli cells, late active stage of the ovotestis (O-2) (scale = 0.5 mm). C. Female acinus with mature oocytes, the ripe stage of the ovotestis (O-3) (scale = 0.5 mm). D. Female acini in the release stage of the ovotestis (O-4) (scale = 0.5 mm). E. Irregularly shaped acini of the ovotestis. These acini are not active, atrophy stage (O-5) (scale = 0.1 mm). Key: ct, connective tissue; ge, germinal epithelium; 1, lumen; o, oocyte; ps, primary spermatocytes; s, spermatogonia; Sc, Sertoli cells; sp, spermatids. Page 352 The Veliger, Vol. 35, No. 4 Explanation of Figures 5 and 6 Figure 5. Photomicrograph of the hermaphroditic duct with the lumen full of spermatozoa. Key: ce, ciliated epithelium; s, spermatozoa (scale = 0.5 m). Figure 6. Photomicrograph of the epithelium of the seminal vesicles of the hermaphroditic duct showing the non- ciliated cuboidal cells (cu) (scale = 0.5 mm). Stages of Activity of the (O-1) Early active stage. The walls of the male acini have Reproductive Organs spermatogonia and primary spermatocytes with large nuclei that stain dark purple with hematox- The activity of the reproductive organs of Caracolus car- ylin (Figure 4A). The female acini contain oogonia acolla can be divided into stages as follows for the ovotestis, that are very small and stain dark red with he- spermatheca, and albumen gland. matoxylin. The ovotestis showed five stages of activity: (O-2) Late active stage. The male acini have secondary Explanation of Figures 7 and 8 Figure 7. Photomicrograph of the wall and lumen of the spermatheca. Key: L1, layer of cells that border the lumen; L2, longitudinal and circular muscle fibers; L3, layer of cells that border the external wall. Key: 1, lumen (scale = 1 mm). Figure 8. Photomicrograph of a section of the spermatheca showing the degraded material in the lumen (1) (stage S-3) (scale = 0.5 mm). P. Marcos, 1992 Figure 9 Photomicrograph of a section of the albumen gland in the simple tubular stage (AG-1). Key: cd, central duct (scale = 1 mm). spermatocytes and spermatids. The spermatids de- velop in clusters with their heads attached to the Sertoli cells, which are located near the lumen (Fig- ures 3, 4B). (O-3) Ripe stage. All follicles are expanded and have ma- ture oocytes and spermatozoa in their lumen (Fig- ure 4C). (O-4) Release stage. Mature gametes are still present in the lumen but they are not as numerous as in the ripe stage (Figure 4D). Some spermatids and vi- tellogenic oocytes are present near the lumen. The spaces between the acini are filled with mature gametes. (O-5) Atrophy stage. Most of the acini are empty (Figure 4B). The spermatheca showed three stages of functional ac- tivity: (S-1) Differentiated stage. The walls are differentiated but the muscle fibers are not completely developed. (S-2) Mature stage. The organ is large but still empty (Figure 7). (S-3) Copulation stage. The lumen of the spermatheca is full of residual spermatophore material that comes from the partner at copulation (Figure 8). The histological study of the albumen gland revealed only two stages of activity: Page 353 STAGE — El Yunque — - Loiza A J eM AM lA SO. NeD 5) SPERMATHECA — El Yunque ) — - Loiza cn ire STAGE Ue peMVAEM aU VATS OF NED Figure 10 Activity stages of the ovotestis (A), spermatheca (B), and albumen gland (C) for the populations from El Yunque and Loiza. El Yunque = ( ), Loiza (-- -). Page 354 Table 1 Two-way analysis of variance testing the effects of different months (climate) and populations (location) on shell and reproductive organ measurements in Caracolus caracolla. Key: Ovo, ovotestis; H.d., hermaphroditic duct; Sper, sper- matheca; A.g., albumen gland. Significance levels: * P < 0:05; ** P< 0.01; *** P < 0.001; NS = not significant: Month x Source Month Population population Shell length see a Shai Shell width art ee NS Ovo. length NS 2 Ovo. width eae NS * H.d. length * NS NS H.d. width Eta NS H.d. weight NS NS Sper. length it * He Sper. width ae baa ae Sper. weight eas es ae A.g. length job a a A.g. width KK 2K 2K KK A.g. weight oe 7 KK OK (AG-1) Simple tubular stage. The gland is small with tubules that consist of layers of cells around a central duct (Figure 9). (AG-2) Compound tubular stage. The cells of the tubules are larger but there is no secretion. Two additional stages were recognizable with the aid of the gross morphology data: (AG-3) Secreting stage. The cells start producing secre- tions; the gland starts swelling. (AG-4) Mature stage. The cells are full of secretions; the gland is an intense yellow and two or three times larger than at the start of AG-3. Because the seminal vesicles of the hermaphroditic duct had spermatozoa in the lumen throughout the study period (Figure 5), its activity was not divisible into distinct stages. The timing of the activity stages was different for the populations from El Yunque and Loiza (Figure 10). El Yunque population: The early active stage (O-1) of the male acini was observed in the months of October and November (Figure 10A). Most of the acini were in the late active stage (O-2) in December through January. From February to May spermatids appeared in the acini and in the month of June the lumen was filled with sper- matozoa (stages O-3 and O-4) and remained filled until August. In September the acini were in the atrophy stage (O-5). The reproductive stages of the female acini were similar to those of the male acini, the only difference being that the oocytes began maturing in May. The differentiated stage of the spermatheca was ob- served from March to October (S-1) (Figure 10B). In The Veliger, Vol. 35, No. 4 November and December the mature stage (S-2) was pre- dominant and in January the lumen was completely filled with residual material (stage S-3). The albumen gland showed a maximum size and mass in the month of May (stage AG-4) (Figure 10C). In Au- gust and September the values of size and mass were min- imal (simple [AG-1] and compound [AG-2] tubular stage). The secretory stage (AG-3) probably began in October and lasted until February, when the size and mass of the spermatheca increased. The stages of activity of this gland were reflected in the variation of size and mass throughout the year. Loiza population: By the beginning of the wet season, November, the male acini appeared to be in the early active stage (O-1) and by February they had spermatids, char- acterizing the late active stage, (O-2) (Figure 10A). Oo- gonia and spermatogonia were in the ripe stage (O-3) beginning in April. In July, the ovotestes were releasing gametes (stage O-4). Ovotestes in the atrophy stage (O-5) were found in the months of September and October. The spermatheca were at the differentiated stage (S-1) from February to April, since their lumen was empty and the cells lining the ducts were reduced in size. The mature stage (S-2) was detected because of an increase in the size of this organ, although the lumen was still empty. This stage was observed from April through August. Another characteristic of this stage was that the layer of cells that border the lumen became larger and extended into the lumen. The copulation stage (S-3) was detected because of the presence of material in the lumen; this stage started in September and lasted until January. In this stage the cells that border the lumen were extended and their apices had a material that was continuous with the residual ma- terial found in the lumen. It is possible that these cells were absorbing the residual material. Twice during the year, the albumen gland of snails from the Loiza population had two peaks of activity but no seasonal pattern was evident. Statistical Analysis of Seasonal Variation A two-way analysis of variance (ANOVA) revealed the relative effects of seasonal variation in climate (““Month’’) and location (“Population”), and the interaction of these two (Table 1). The effects were significant for most of the measurements (P = 0.001), especially for the spermatheca and albumen gland measurements. A Principal Component Analysis reduced the number of variables. The variables were divided into two groups: size of the shell (SHELL) and size of the reproductive organs (REPROD). There were two principal components for the SHELL group (SHELL 1 and 2) and three prin- cipal components for the REPROD group (REPROD 1, 2, and 3) accounting for a cumulative percentage variance of 96% for SHELL and 70% for REPROD. SHELL 1 varies seasonally, which means that the average shape of P. Marcos, 1992 Table 2 Pearson Correlation Coefficients between the principal component of shell measurements (SHELL 1) and prin- cipal components of reproductive organs (REPROD 1-4) of Caracolus caracolla. Shell 1 with r P REPROD 1 0.0884 0.2835 REPROD 2 — O55 55 0.0583 REPROD 3 —0.2360 0.0038 REPROD 4 —0.0502 0.5433 the shell of the population is changing throughout the year in both populations. REPROD 1 has positive weighting for the size of the reproductive organs. To determine if the sizes of the reproductive organs were correlated to shell size, Pearson Correlation Coefficients were obtained between SHELL 1 and REPROD 1-4 (Table 2). The results indicate that the only significant correlation observed was between SHELL 1 and RE- PROD 3 (P = 0.0038). Since REPROD 3 did not show significant effects among the factors that were analyzed, the correlation found between REPROD 3 and SHELL 1 will not be considered. These results suggest that the size of the reproductive organs are not correlated with the size of the shell. A two-way ANOVA was performed with REPROD 1, 2, and 3 (Table 3). The first principal component of the reproductive organs, REPROD 1, had significant values. The interaction of the population effect in the first com- ponent is observed during the months of March to June; July to October the interaction is almost overlapping. From November to February there was no interaction between the populations because they followed a similar pattern of increase and decrease of REPROD 1. There was no sig- nificant difference in the effect of month on REPROD 2 (P = 0.1920). There was no effect of month (P = 0.1041) or interaction between month and population on RE- PROD 3 (P = 0.0984). DISCUSSION Histological and gross morphological changes indicate a seasonal variation in most of the reproductive organs of Caracolus caracolla from the El Yunque and Loiza popu- lations. Although the pattern of reproductive activity is similar in the two populations, statistical analyses show significant differences in almost all of the measurements of the shells and reproductive organs within each popu- lation and between populations. The differences between populations indicate that some external factors probably affect the reproductive activity of these snails. Environmental Factors In El Yunque as temperature decreased (October), sper- matogenesis and oogenesis began. When the temperature Page 355 Table 3 Two-way analysis of variance testing the effects of different months and populations on REPROD 1, 2, and 3 mea- surements in Caracolus caracolla. Significance levels: * P < OOS iI << OOS tI << OOO) Source SS DF MS _ F-ratio REPROD 1 Month 89.826 11 8.166 34g 5 Population 90.093 il SOOM Silt Month xX population 99.450 11 9.041 S805 REPROD 2 Month TONS Gi A 6.921 Oana Population 1.227 1 12271 a2 Month x population 42.493 11 3.863 eae REPROD 3 Month Caley wali 3.798 AS OE Population 2.216 1 2.216 2.68 Month x population 14.810 11 1.346 1.63 began to rise (April), the oocytes were already mature and the spermatids began developing into spermatozoa (Figure 10A). This pattern suggests that an elevation of temper- ature may be necessary for the maturation of the sper- matozoa but that they will not necessarily be mature before the oocytes, as other studies suggest (BOUILLON, 1956; Duncan, 1975). In Loiza, although the annual temper- ature pattern is the same as in El Yunque (Figure 2A), the male and female gametes mature in the same month. As monthly precipitation increased in El] Yunque from 250 mm to 500 mm (April to August), copulation began. In Loiza, the peaks of precipitation were in April and August, which suggest that there might be two seasonal peaks of mating activity. Maximum activity of the spermatheca was detected in January and February for the population from El Yunque and in September through December for Loiza (Figure 10C). The activity of the spermatheca should be at a max- imum when copulation occurs. It is possible that once the sperm are mature (April and May), they are stored in the hermaphroditic duct until August (the month of maximum precipitation). Copulation begins during this month, and by December (in Loiza) and February (in El] Yunque) the lumen of the spermatheca is completely filled with de- graded material from the spermatophores. SMITH (1966) described four activity stages for the al- bumen gland of Avion ater. In the present study only the first (AG-1) and second (AG-2) stages could be detected histologically. The secreting (AG-3) and mature (AG-4) stages were determined only by the gross morphology of the gland. The albumen glands of the snails from El Yunque were at the mature stage in March, one month after the spermatheca were at the copulation stage. Since the func- tion of the albumen gland is to secrete the albumen that surrounds the fertilized eggs, it should be active one or Page 356 two months after the snails mate (HEATWOLE & HEATWOLE, 1978; Tompa, 1984). In Loiza, the activity of the albumen gland did not show any seasonal pattern. Although the snails from this dry coastal area showed a seasonal pattern in the activity of the ovotestis and sper- matheca, the albumen gland may be active at various times througout the year so that the fertilized eggs can develop whenever mating occurs. The high standard deviations show that not all snails had the same state of reproductive activity in each month. It is possible that snails from both populations are mating continually throughout the year, since spermatozoa are always present in the hermaphroditic duct. Nevertheless, according to the activity of the spermatheca, there are some peaks of mating activity in February for the El Yunque population and in December for the Loiza population. SOLEM & CHRISTENSEN (1984) found that at the be- ginning of the wet season in Australia there is a marked increase in the size of the ovotestis of camaenids, whereas in this study the size of the ovotestis (length and width) fluctuated throughout the year. In the dry season, es- pecially in March and April, the size of the ovotestis of Caracolus caracolla reached a maximum. Just as I observed in the snails of the Loiza population, SOLEM & CHRISTEN- SEN (1984) found no seasonal change in the size of the albumen gland, probably because both of their sites are arid. The Australian camaenids live in a semi-arid envi- ronment that has extensive wet and dry seasons. The dis- similarity in environmental conditions could explain the different reproductive patterns. HEATWOLE & HEATWOLE (1978) also studied Caracolus caracolla in Puerto Rico and determined that the proportion of animals with functional albumen glands is greater in June and July for both populations, but in the present study the population of El Yunque had a greater propor- tion of functional glands in May. Since the albumen gland is active from March to June, fertilization of the oocytes should have taken place already. Once the sperm are de- veloped in the ovotestis, they are stored in the lobules of the seminal vesicles of the hermaphroditic duct. HEATWOLE & HEATWOLE (1978) observed a greater number of cop- ulations in April (dry season); during the wet season no copulation was observed in the snails from the rain forest. Only three copulations were observed during the wet sea- son in the dry coast population. Their results are in ac- cordance with the results obtained here since there is only one month of difference between copulation and the activity of the albumen gland. An important characteristic observed in this study and not reported previously is that the male and female sex cells developed simultaneously in separate acini of the ovo- testis (Figure 3). Male acini tended to occupy the central position of the ovotestis and the female acini were located in a peripheral position. Previous studies of the ovotestis of pulmonates report that the development of sex cells occurs simultaneously within the same acinus (e.g., The Veliger, Vol. 35, No. 4 BRIDGEFORD & PELLUET, 1952; Lusis, 1961; KUGLER, 1965; JoosE & REITZ, 1969; JONG-BRINK et al., 1976; Tompa, 1984). Since this is the first histological study of the reproductive cycle of this family, it is not clear whether this feature is unique to Caracolus caracolla or represents a characteristic of the family. The shell and reproductive organs are larger in the snails from El] Yunque, but statistical analysis demonstrated no correlation between the size of the shell and the size of the reproductive organs. I expected to find a larger shell in the snails from Loiza since they live in limestone forma- tions (called mogotes) which are an important source of calcium for the construction of shells. This difference in size might be determined by environmental factors that vary between Loiza and El Yunque, such as the type of vegetation, soil characteristics, temperature, or water. The environmental factor that fluctuated the most throughout the year was precipitation (Figure 2B). Pre- cipitation probably determines the seasonality of repro- duction in the camaenids, especially in the Loiza popu- lation. This study also reveals a statistically significant difference in the reproductive activity and shell size be- tween the populations of El] Yunque and Loiza. The dif- ferences could be explained by environmental factors, es- pecially precipitation and relative humidity. The histological analysis of the four reproductive organs shows that copulation begins earlier in the snails of Loiza (September) than in snails from El Yunque (January). Copulation probably starts earlier in Loiza, ensuring that the eggs are deposited in moist soil before the dry season begins. The snails from El Yunque do not confront this problem since their habitat is moist throughout the year. These results document changes that occurred during one year. To determine if these changes occur on an annual cycle, however, the study would need to be extended over several years. To determine the effects of the environmental factors on the reproductive cycle of Caracolus caracolla con- trolled studies of temperature, humidity, and day length should be performed. ACKNOWLEDGMENTS This work originally was completed as partial fulfillment of the M.S. degree at the University of Puerto Rico. It was supported in part by NIH-MBRS RR08102 and NSF-RIMI RII-8903827. I am deeply grateful to Dr. Janice Voltzow for permitting me to use the facilities of her laboratory and for her valuable assistance in the re- daction of this manuscript. I would also like to thank Dr. Ned Fetcher for helping me with the statistical analyses. LITERATURE CITED BOUILLON, J. 1956. Influence of temperature on the histological evolution of the ovotestis of Cepaea nemoralis. Nature 177: 142-143. Boer, H. & J. JoosE. 1975. Endocrinology. Pp. 245-307. In: P. Marcos, 1992 V. Fretter & J. Peake (eds.), The Pulmonates. Vol. 1. Ac- ademic Press: New York. BRIDGEFORD, H. B. & D. PELLUET. 1952. Induced changes in the cells of the ovotestis of the slug Deroceras reticulatum (Miller) with special reference to the nucleolus. Canadian Journal of Zoology 30(2):323-337. CarRRICK, R. 1942. The grey field slug Agriolimax agrestis and its environment. Annals of Applied Biology 29:43-55. Duncan, C. J. 1975. Reproduction. Pp. 309-365. Jn: V. Fretter & J. Peake (eds.), The Pulmonates. Vol. 1. Academic Press: New York. GERAERTS, W. P. & J. JoosE. 1975. Control of vitellogenesis and of growth of female accessory sex organs by the dorsal body hormone (DBH) in the hermaphroditic freshwater snail Lymnaea stagnalis. General and Comparative Endocrinology 27:450-464. HEATWOLE, H. & A. HEATWOLE. 1978. Ecology of the Puerto Rican camaenid tree snails. Malacologia 17(2):241-315. HENDERSON, N. & D. PELLUET. 1960. The effect of visible light on the ovotestis of the slug Deroceras reticulatum. Ca- nadian Journal of Zoology 38:173-178. Humason, G. 1979. Animal Tissue Techniques. 4th ed. W. H. Freeman and Company: San Francisco. 661 pp. Joose, J. & D. Reitz. 1969. Functional anatomical aspects of the ovotestis of Lymnaea stagnalis. Malacologia 9(1):101- 109. JONG-BrRINK, M., A. WiT, G. KRAAL & H. H. Boer. 1976. A light and electron microscope study on oogenesis in the fresh- water pulmonate snail Biomphalaria glabrata. Cell and Tissue Research 171:195-219. KUGLER, O. E. 1965. A morphological and histochemical study of the reproductive system of the slug Philomicus carolinianus. Journal of Morphology 116:117-132. Lusis, O. 1961. Postembryonic changes in the reproductive system of the slug Arion ater rufus. Proceedings of the Zoo- logical Society of London 137(3):433-468. NOAA (NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRA- TION). 1989. Climatological Data, Vol. 34 (4). Puerto Rico and Virgin Islands. Page 357 PELLUET, D. 1964. On the hormonal control of cell differen- tiation in the ovotestis of the slug. Canadian Journal of Zoology 42:195-199. PELLUET, D. & N. LANE. 1961. The relation between neu- rosecretion and cell differentiation in the ovotestis of the slugs. Canadian Journal of Zoology 39:789-805. RIDDLE, W. 1983. Physiological ecology of land snails and slugs. Pp. 431-461. In: W. D. Russell-Hunter (ed.), The Mollusca. Vol. 6. Academic Press: New York. RUNHAM, N. & A. LARYEA. 1968. Studies on the maturation of the reproductive system of Agriolimax reticulatus (Pul- monata: Limacidae). Malacologia 7:93-108. SMITH, B. 1966. Maturation of the reproductive tract of Arion ater. Malacologia 4:325-334. SOKOLOVE, P. & E. McCRONE. 1978. Reproductive maturation in the slug Limax valentianus. Annale Universiteit van Stel- lenbosch, Serie A2, 49:1-46. SoLeM, A. & C. CHRISTENSEN. 1984. Camaenid land snails reproductive cycle and growth patterns in semiarid areas of north-western Australia. Australian Journal of Zoology 32: 471-491. TAKEDA, N. 1979. Induction of egg-laying by steroid hormones in slugs. Comparative Biochemistry and Physiology 62:273- 278. Tompa, A. S. 1984. Land snails (Stylommatophora). Pp. 47- 125. In: A. S. Tompa, N. H. Verdonk & J. A. M. van der Biggelaar (eds.), The Mollusca. Vol.7. Academic Press: New York. VAN DER LaaN, K. L. 1980. Terrestrial pulmonate reproduc- tion: seasonal and annual variation and environmental fac- tors in Helminthoglypta arrosa (Pulmonata: Helicidae). The Veliger 23:48-54. WIJDENES, J. & N. W. RUNHAM. 1976. Studies on the function of the dorsal bodies of Agriolimax reticulatus (Mollusca: Pul- monata). General and Comparative Endocrinology 29:545- BOE Wo pa, H. 1967. The effect of temperature on reproduction in some morphs of the landsnail Cepaea nemoralis. Evolution 21:117-129. The Veliger 35(4):358-365 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 The Ecology of Coquina Clams Donax variabilis Say, 1822, and Donax parvula Philippi, 1849, on the East Coast of Florida ERIK BONSDORFF Department of Biology and Husé Biological Station, Abo Akademi University, SF-20500 Abo, Finland WALTER G. NELSON Department of Oceanography, Ocean Engineering, and Environmental Science, Florida Institute of Technology, Melbourne, Florida 32901, USA Abstract. Studies on the population ecology of the coquina clams Donax variabilis Say, 1822, and D. parvula Philippi, 1849, were carried out at a beach transect on the central east coast of Florida. The transect covered the range from the upper intertidal to the offshore subtidal zone, and sampling was done monthly between February 1982 and August 1983. The seasonality in size-frequency distributions (no clear growth pattern, and narrow size ranges) for the two species were similar at all stations. Large individuals (>12 mm) of both species were absent from the offshore stations. The study illustrates the importance of sampling subtidally, as the peak abundances of juveniles and adults of both species were found offshore. Primary juvenile recruitment occurs subtidally, with specimens subsequently migrating into the swash zone as they increase in size. Small individuals (<3 mm), which constituted a substantial portion (46.8%) of Donax in all seasons in the present study, should be included in future analyses in order to accurately describe Donax populations. INTRODUCTION Bivalves of the genus Donax (Donacidae) are often an abundant component of the intertidal fauna on sand beach- es in many parts of the world (CoE, 1953, 1955; WADE, 1967a, b, c; ANSELL et al., 1972; ANSELL & TRUEMAN, 1973; McLusky et al., 1975; MCLACHLAN et al., 1979). The biology of this genus has recently been reviewed by ANSELL (1983a). Along the east coast of Florida two species are found, namely Donax variabilis Say, 1822, and D. par- vula Philippi, 1849, the separation of which has caused considerable confusion (MoRRISON, 1971). ABBOTT (1974) considered D. parvula to be merely an ecomorph of D. variabilis. The shell morphology of Donax species, includ- ing D. variabilis (CHANLEY, 1969a, b), is known to be quite variable (WADE, 1967c). Comparisons of spatial distri- butions, morphometrics, and, particularly, allozymes (Nel- son et al., unpublished data), support the validity of D. parvula as a separate species. Considerably more information is available for Donax variabilis than for D. parvula. Several studies have exam- ined aspects of burrowing and migratory behavior (PEARSE et al., 1942; TURNER & BELDING, 1957; EDGREN, 1959; TIFFANY, 1971; MIKKELSEN, 1978, 1981, 1985; VEGA & TUNNELL, 1987). Most intertidal populations of D. var- tabilis migrate up and down the beach with the tides, the migratory behavior being stimulated by the acoustic shock of breaking waves. However, two instances of non-migra- tory populations have been reported (EDGREN, 1959; MIK- KELSEN, 1978, 1981). MIKKELSEN (1981) suggests that non-migratory behavior is a local adaptation to deal with a combination of low beach slope and wave energy in areas of irregular semi-diurnal tides and low sand permeability. Other papers on Donax variabilis have examined larval E. Bonsdorff & W. G. Nelson, 1992 development (CHANLEY, 1969a), general ecology (ED- GREN, 1959), shell polychromism (MIKKELSEN, 1978; SCHNEIDER, 1982), and other shell characteristics (ADAMKEWICZ, 1989). The most detailed study of Florida populations of D. variabilis with respect to intertidal dis- tribution, growth rates, and shell color variability is that of MIKKELSEN (1978). Much of the available information on Donax parvula is contained in the taxonomic revision of the genus Donax, by MorrIson (1971), but most often this species has not been separated from D. variabilis. Some additional infor- mation on the abundance, seasonality, and distribution of these two species is available from a number of more recent field studies which have differentiated between D. variabilis and D. parvula (REILLY & BELLIS, 1978, 1983; SPRING, 1981; LEBER, 1982a, b). Most studies of Donax, including those from Florida (EDGREN, 1959; TIFFANY, 1971; MIKKELSEN, 1978, 1981, 1985), have sampled the intertidal zone exclusively. The present study provides data from a transect including sub- tidal stations which indicate that a significant portion of the populations for both Donax species are located below the intertidal zone. Spatial and seasonal changes in density, together with seasonal size-frequency distributions, are presented and compared for both species. MATERIALS anD METHODS Collections of Donax were made as a part of a larger benthic sampling study at approximately monthly intervals between February 1982 and August 1983. The sampling location was a moderately exposed sand beach located on the outer coast of the barrier island in Melbourne Beach, Brevard County, Florida. The precise sampling area was located at coastal construction control line survey marker R-140 (28°2'55”N, 80°34'54”W) of the Florida Depart- ment of Natural Resources. Water temperature measured in the surf on the dates of sampling varied from 14.5 to 29.5°C, while surf salinity ranged between 34.5 and 35.5%o (ALLENBAUGH, 1984; PETERS, 1984). Previous biological sampling at this station has been described by SPRING (1981), GORZELANY (1983), and NELSON (1986). Geological characteristics of the sampling station were extensively studied by STAUBLE et al. (1983). Mean tidal range was approximately 1 m, with mean annual wave height being in the range of 0.7 to 0.9 m (STAUBLE et al., 1983). Beach slope on the foreshore ranged between 15 and 25 degrees, depending on the season. Sediment grain size was a coarse sand on the foreshore (mean grain size, 1.8 phi), generally with a region of very coarse shell frag- ments located in the area of wave break on the shore. Offshore, sediments were a fine sand (mean grain size, 3 phi) (STAUBLE et al., 1983). Organic content of the sedi- ment was approximately 0.5% (Nelson, unpublished data). At the study area, a transect line was established per- pendicular to the shore, and four replicate 20.3-cm-di- ameter cores were taken at the high-tide line, at the base Page 359 of the region of wave run-up on the beach (swash zone, approximately 30 m below the high-tide line), and at dis- tances of 61 m (200 ft) and 91 m (300 ft) from the high- tide line. Water depths at the 61 and 91 m sites were in the range of 1.5-3 m. All samples were taken during low tide, and sieved in the field through a 0.5-mm-mesh screen, fixed in 10% formalin, re-sieved through a 0.5-mm-mesh screen in the laboratory, and stored in a 70% ethanol-rose bengal stain solution until sorted. Small individuals (<3 mm) could not always be assigned to one species, and most small specimens were classed as (“small”) Donax spp. Shell length was measured with an optical micrometer. Dry weight was measured after drying the specimens at 100°C for 24 hr. After weighing, the samples were combusted at 500°C for 3 hr before mea- suring the ash-free dry weight. Weight determinations were done for Donax variabilis only, as the material for D. parvula contained too few individuals to cover the entire range of the total size-frequency distribution of the pop- ulation. Simple linear regression equations relating log-trans- formed shell length and weight (wet-, dry-, and ash-free dry weight) for D. variabilis were computed (SOKAL & ROHLF, 1981). RESULTS Abundance Patterns The seasonal patterns of abundance for Donax variabilis, D. parvula, and Donax spp. (unidentifiable juveniles of D. variabilis and D. parvula) are presented in Figures 1-3. Individuals of both species were only rarely collected from the high-tide line (possibly due to a downshore migration during low tide), and data from this station are not pre- sented. For Donax variabilis in the swash zone, a peak density was observed in February 1982, with relatively lower den- sities being found for the rest of the sampling period. The patterns of abundance at the 61-m and 91-m locations were broadly similar. The highest abundances were found in the April-May period, both in 1982 and 1983. At the 91-m station, there was also a peak in abundance in De- cember 1982. On most sampling dates, densities were greater at the offshore stations than at the swash-zone station (maximum density of 3550 individuals/m? at the 61-m station in May 1983). The pattern for Donax parvula was less consistent among stations. Peaks of density occurred in February and Oc- tober 1982 in the swash zone. At the 61-m location, the maximum density occurred in November 1982, with sec- ondary peaks in August 1982 and 1983. At the 91-m station, abundance was greatest in August 1982 and 1983. The maximum density observed for D. parvula, from the swash zone (February 1982), was 1416 individuals/m?. Donax spp. (<3 mm) were never found at the high-tide station and were only rarely found in the swash zone. At Page 360 2 The Veliger, Vol. 35, No. 4 a= | / MEAN ABUNDANCE m ie. | SWASH ZONE (3536) Figure 1 Spatial distribution of Donax variabilis between February 1982 and August 1983 along the sampled transect (swash zone, 61 m, 91 m). the 61-m station, the greatest abundances were found from May through September. Density at 91 m was variable, with maxima being found in April, September, and No- vember 1982, while abundance in 1983 was uniformly low. The maximum density (59,136 individuals/m* was observed at 61 m during August 1982. Donax variabilis tended to be somewhat more abundant than D. parvula on most sampling dates. Small Donax spp. were generally more abundant than larger specimens of either species, and constituted 46.8% of the total individual count. The small individuals, although pooled for both species (as no distinction in recruitment to the two species can be made), are an important proportion of these pop- ulations and one that has been widely neglected in previous studies on Donax. Size-Frequency Distributions Within the three categories (Donax variabilis, D. parvula, and Donax spp.), size-frequency distributions for the swash zone, 61-m, and 91-m stations were similar in almost all cases. Therefore, data from all three locations were pooled for the size-frequency histograms presented in Figure 4. The only exception involved Donax variabilis, for which a total of 15 specimens of >12 mm length were found in the swash zone during the entire sampling period from February 1982 through August 1983. Because these spec- imens represented only 0.9% of the total number of D. variabilis, and because no specimens of D. parvula of >12 mm length were found (13.5-22.5 mm), these 15 individ- uals are not displayed in Figure 4 in order to standardize the graph axes. The patterns of size-frequency distributions for Donax variabilis were generally similar between 1982 and 1983. In 1982, the mean size decreased in April and shifted to a still smaller size class in August, reflecting the heavy recruitment taking place at this time (Figure 4, August 1982). The size distributions were generally similar from September through December, while January and Feb- ruary 1983 showed a slight shift in the mean to a larger size. Unlike in 1982, when the shift occurred in April, a shift to smaller size classes occurred in March 1983. The mean size remained low through May, and shifted to an even smaller size category in August 1983. E. Bonsdorff & W. G. Nelson, 1992 Page 361 500 -— 400 1 300 —- 200 i. 2 r) oO a ree MEAN ABUNDANCE m — SWASH ZONE Figure 2 Spatial distribution of Donax parvula between February 1982 and August 1983 along the sampled transect (swash zone, 61 m, 91 m). For Donax parvula the annual patterns of changes in the size-frequency distributions were generally similar in both the years sampled (Figure 4). The mean of the size- frequency distribution tended to increase from the late fall (e.g., November 1982) until March, when it reached the annual maximum. From March through the summer, and even until October, the mean size tended to decrease. For the small Donax spp. (Figure 4), there was a slight increase in the mean size from February through May 1982, with a decrease to smaller mean size during June through August. The mean size was somewhat higher from September through December. In January 1983, the size- frequency distribution of the juveniles was strongly skewed to the smallest size class, suggesting the occurrence of re- cruitment (Figure 4). By March, the mean had increased again. During spring and summer 1983 (April-August), mean size was in the size interval of 1.5 mm, as it had been in the period June-September 1982. The shifts to smaller mean sizes suggest recruitment of new individuals to the population. The strong peak of juveniles in the 1.5- mm size class, together with the extremely high density recorded in August 1982 (about 60,000/m7?), suggests that heavy settlement had occurred in July (Figure 4). Thus the size-frequency distributions of the juveniles clearly indicate principal periods of recruitment, but also show a constant input of recruits to the subtidal populations of D. variabilis and D. parvula. Length-Weight Relationships and Growth Rates for Donax variabilis The regressions for size (maximum shell length) and weight (wet weight; y = 0.107x?!°°, r = 0.99, dry weight; y = 0.08x!°°, r = 0.99, and ashfree dry weight; y = 0.041x?°%, r = 0.99) all show highly significant positive correlations, illustrating a population of individuals with active tissue growth throughout the size range collected. This relationship is in good accordance with the general biology of other species of Donax. ANSELL (1983b) showed that many species of Donax have species-specific growth curves. Growth rates of 3.7 mm/month have been reported for D. variabilis from Florida (MIKKELSEN, 1978, 1985). The size distributions and mean sizes per month, however, do not allow for an accurate growth analysis, owing to the Page 362 200 ~— The Veliger, Vol. 35, No. 4 150 —- 2 CS. (7062) (59136) SWASH ZONE vara (5359) MEAN ABUNDANCE m (16139) 91m 0 [opt + JiR IME AY Mie de i VA eS 1982 ce) N D J F M ALM) Ul) eA 1938s Figure 3 Spatial distribution of small Donax spp. between February 1982 and August 1983 along the sampled transect (swash zone, 61 m, 91 m). constant input of small individuals into the population. On the basis of the mean sizes of the individuals, the maximum change in mean size in the present study was 3.43 mm. DISCUSSION Distribution and Abundance Donax variabilis is distributed from the coast of Virginia to the coast of Mississippi (MORRISON, 1971). CHANLEY (1969b) suggested that this species may exhibit summer range extensions as far north as Long Island, and Ja- COBSEN (1955) described such a population, calling it Do- nax fossor. MORRISON (1971) suggested that D. variabilis has a two-year life-span, and that in some cases individuals may survive a third year, whereas ANSELL (1983a) re- ported a one- to two-year life cycle. MIKKELSEN (1978, 1985) estimated that D. variabilis in Florida grow at a rate of 3-3.7 mm/month in the summer months. The present data, with high numbers of recruits and small mean sizes throughout the year, support the previous reports on rapid growth rates and a short life-span, because no senescence was recorded among the individuals. MIKKELSEN (1978) suggested that this species spawns in February with a three-week larval period, resulting in a March settlement. He also suggested that Florida D. variabilis has a second spawning in June. LEBER (1982a) recorded juvenile re- cruitment to a North Carolina population as occurring in February and November. The February settlement was indicative of a winter spawn. He suggested that two-year- old individuals move into the intertidal swash zone in March after overwintering in the shallow offshore zone. MatTTA (1977) observed a June settlement of spat of D. variabilis in Duck, North Carolina. Donax parvula is recorded from Ocracoke, North Car- olina, to St. Lucie County, Florida, and is reported to have a two-year life-span (MORRISON, 1971) possibly spawning somewhat later than D. variabilis. However, LEBER (1982a, b) indicated that recruitment in a North Carolina popu- lation occurs in February, as it did for D. variabilis, al- though the two species have different abundance maxima (D. parvula in May-July, and D. variabilis in August— E. Bonsdorff & W. G. Nelson, 1992 Percent February 1982 Page 363 November 1982 «OD. variabilis D. parvula =] "small" Donax March 1982 December 1982 April 1982 January 1983 May 1982 February 1983 June 1982 March 1983 August 1982 April 1983 September 1982 May 1983 October 1982 August 1983 tl 4A 8 G 6 7 8 9 10 11 12 mm bo 2 8 G2 OFT 859 Figure 4 Size-frequency (percent) distribution (0.5-mm classes) of Donax variabilis (black bars), D. parvula (striped bars), and small Donax spp. (shaded bars) from all sampled dates (data pooled for all stations along the transect studied). The percent values are relative as to the actual numbers of each category to allow for direct comparisons. 10 11 12 mm Page 364 The Veliger, Vol. 35, No. 4 September). REILLY & BELLIs (1978, 1983) also observed young of both Donax species recruiting during March in North Carolina. A review of available seasonal abundance data from a variety of locations (NELSON, 1985) indicated that both Donax variabilis and D. parvula tend to achieve maximum densities during the summer (June-July) in most loca- tions. The present study suggests an earlier period (April- May) for peak abundances of D. variabilis at Melbourne Beach, Florida. Donax parvula generally showed later maxima, in the period from August through November and again in February. The broad period of abundances for small individuals (April-November) suggests an ex- tended period of recruitment, given MIKKELSEN’s (1978, 1981, 1985) estimates of growth rates. The differences observed here as compared with other studies may be in part due to different temperature regimes at different lo- cations and/or differences in food availability for the clams. Also important is that subtidal areas were included in this study, whereas they were not in the other studies. Recorded maximum densities for Donax variabilis range from 166 to 13,114 individuals/m? (NELSON, 1985), whereas estimates of density for D. parvula range from 401 to 1425/m?. Density estimates from the present study fall within the maximum ranges for both species (MIK- KELSEN, 1978, 1981) and the total abundance of Donax spp. during the sampling period in 1982-1983 did not differ from previous records from the same area (GORZELANY, 1983; NELSON, 1986), although the relative proportions between the species has varied over time. Gen- erally the published records of population densities of the species do not include recruits, however, and in the present study such recruits made up a substantial portion of the population throughout the investigated period (average 46.8%), adding almost 60,000 individuals/m? at the peak value. Wide ranges of density values for D. denticulatus at different locations have been shown to depend partially on environmental factors, such as particulate organic carbon (SASTRE, 1984). Size-Frequency Distributions The seasonal patterns of the size-frequency distributions for Donax parvula and D. variabilis were very similar (Fig- ure 4). Neither species showed clear evidence of an increase in average size except for the late winter-early spring period. For both species, the mean size tended to decrease after March until the following fall. This size shift cor- responds to the period when the largest numbers of small Donax spp. were found. The presence of very small clams in the subtidal population virtually year round suggests that although the major period of recruitment may occur in the summer, recruitment to the populations may take place throughout the year. This agrees with ANSELL’s (1983a) observation that the typical spawning pattern for Donax consists of repeated incomplete spawnings by in- dividuals over an extended spawning season, in contrast to the statement by SASTRE (1984) that D. denticulatus from Puerto Rico would be the only known species of Donax with continuous recruitment. Two factors may contribute to the absence of large in- dividuals (>12 mm) from the subtidal populations (sta- tions at 61 m and 91 m offshore). Predation by bottom- feeding fish, which commonly have Donax in their guts (NELSON, 1986), may fall heaviest on the larger size classes. Alternatively, clams may migrate shoreward as they reach larger sizes. LEBER (1982a) suggested an opposite migra- tion pattern with an offshore migration of D. parvula in winter. The size-frequency distributions for D. varzabilis collected from the swash zone at a nearby beach by MIK- KELSEN (1978) suggest that migration may be important. Mikkelsen’s samples (April-September only) indicated that the bulk of the population of the intertidal zone was larger than 12 mm. Since initial recruitment of Donax appears to occur largely subtidally in this area, large intertidal individuals are most likely the result of migration from offshore, with subsequent growth in the swash zone, thus avoiding predation by fish. ACKNOWLEDGMENTS Financial support for the collection of samples was pro- vided by the Florida Sea Grant College program with support from the National Oceanic and Atmospheric Ad- ministration, Office of Sea Grant, U.S. Dept. of Commerce, Grant No. NA80AA-0-0038 to W.G.N. Support for prep- aration of the manuscript was provided by a Fulbright Fellowship to W.G.N. through the Finland-U.S. Educa- tional Exchange Commission. Additional support was pro- vided by grants from the U.S. National Science Foundation (INT-8712670) and the American Scandinavian Foun- dation. Financial support from the U.S. Educational Foun- dation in Finland (the Fulbright Fellowship Programme), the Academy of Finland, and the Maj & Tor Nessling Foundation in Finland to E.B. made possible the analysis of the material and the preparation of the manuscript. The assistance of Tom Allenbaugh and Dennis Peters, who were primarily responsible for sample collection, is grate- fully acknowledged. The assistance of Mr. Paul Mikkelsen in distinguishing between the two species of Donax was essential to this paper. Editorial comments by Ms. Paula Mikkelsen and Dr. Laura Adamkewicz improved the manuscript. LITERATURE CITED ABBoTT, R. T. 1974. American Seashells. 2nd ed. Van Nos- trand Reinhold: New York. 663 pp. ADAMKEWICZ, L. 1989. Differences in the frequencies of several shell characteristics in the clam Donax variabilis around Cape Hatteras, North Carolina. The Veliger 32:21-28. ALLENBAUGH, T. D. 1984. Dynamics of nearshore amphipod assemblages from the central Florida east coast. M.S. Thesis, Florida Institute of Technology, Melbourne, Florida. 46 pp. ANSELL, A. D. 1983a. The biology of the genus Donax. Pp. E. Bonsdorff & W. G. Nelson, 1992 607-635. In: A. McLachlan & T. Erasmus (eds.), Sandy Beaches as Ecosystems. Junk Publishers: The Hague. ANSELL, A. D. 1983b. Species of Donax from Hong Kong: morphology, distribution, behaviour, and metabolism. Pp. 19-47. In: B. Morton & D. Dudgeon (eds.), Proceedings of the Second International Workshop on the Malacofauna of Hong Kong and Southern China. Hong Kong University Press: Hong Kong. ANSELL, A. D., P. StvADAS, B. NARAYANEN & A. TREVALLION. 1972. The ecology of two sandy beaches in South West India. III. Observations on the populations of Donax incar- natus and D. spiculum. Marine Biology 17:318-332. ANSELL, A. D. & E. R. TRUEMAN. 1973. The energy cost of migration of the bivalve Donax on tropical sand beaches. Marine Behavior and Physiology 2:21-32. CHANLEY, P. 1969a. Larval development of the coquina clam, Donax variabilis Say, with a discussion of the structure of the larval hinge in the Tellinacea. Bulletin of Marine Science 19:214-224. CHANLEY, P. 1969b. Donax fossor: a summer range extension of Donax variabilis. Nautilus 83:1-14. CoE, W. R. 1953. Resurgent populations of littoral marine invertebrates and their dependance on ocean currents and tidal movements. Ecology 34:225-229. CoE, W. R. 1955. Ecology of the bean clam, Donax gouldi, on the coast of southern California. Ecology 36:512-514. EDGREN, R. A. 1959. Coquinas (Donax variabilis) on a Florida beach. Ecology 40:498-502. GORZELANY, J. F. 1983. The effects of beach nourishment on the nearshore benthic macrofauna of Indialantic and Mel- bourne Beach, Florida. M.S. Thesis, Florida Institute of Technology, Melbourne, Florida. 114 pp. Jacossen, M. K. 1955. Observations on Donax fossor Say at Rockaway Beach, New York. Nautilus 68:73-77. LEBER, K. M. 1982a. Bivalves (Tellinacea: Donacidae) on a North Carolina beach: contrasting population size structure and tidal migrations. Marine Ecology Progress Series 7:297- 301. LEBER, K. M. 1982b. Seasonality of macroinvertebrates on a temperate, high wave energy sand beach. Bulletin of Marine Science 32:86-98. Matta, J. F. 1977. Beach fauna study of the CERC field research facility, Duck, North Carolina. U.S. Army Corps of Engineers, Coastal Engineering Research Center, Misc. Report 77-6. McLacu. an, A., T. WOOLDRIDGE & G. VAN DER Horst. 1979. Tidal movements of the macrofauna on an exposed sandy beach in South Africa. Journal of Zoology, London 187: 433-442. McLusky, D. S., S. A. Nair, A. STIRLING & R. BHARGAVA. 1975. The ecology of a central Indian beach, with particular reference to Donax incarnatus. Marine Biology 30:267-276. MIKKELSEN, P.S. 1978. A comparison of intertidal distribution, growth rates and shell polychronism between two Florida populations of the coquina clam, Donax variabilis Say, 1822 (Bivalvia: Donacidae). M.S. Thesis, Florida Institute of Technology, Melbourne, Florida. 78 pp. MIKKELSEN, P. S. 1981. A comparison of two Florida popu- lations of the coquina clam, Donax variabilis Say 1822 (Bi- valvia: Donacidae). I. Intertidal density, distribution, and migration. The Veliger 23:230-239. MIKKELSEN, P. S. 1985. A comparison of two Florida popu- lations of the coquina clam, Donax variabilis Say, 1822 (Bi- Page 365 valvia, Donacidae). II. Growth rates. The Veliger 27:308- 311. Morrison, J. P. E. 1971. Western Atlantic Donax. Proceedings of the Biological Society of Washington 83:545-568. NELSON, W.G. 1985. Guidlines for beach restoration projects. Part I. Biological. Florida Sea Grant College, Report No. 76. University of Florida, Gainsville. 66 pp. NELSON, W.G. 1986. Predation and prey population variation in a high energy sand beach macrofaunal community. Pro- ceedings of the 20th European Marine Biological Sympo- sium. Ophelia 26:305-316. Pearse, A.S.,H. J. HUMM & G. W. WHARTON. 1942. Ecology of sand beaches at Beaufort, North Carolina. Ecological Monographs 12:135-140. PETERS, D. R. 1984. Seasonality, residency and spatial distri- bution of juvenile surf zone fishes of the Florida east coast. M.S. Thesis, Florida Institute of Technology, Melbourne, Florida. 66 pp. REILLY, F. J. & V. J. BELLIS. 1978. A study of the ecological impact of beach nourishment with dredged materials on the intertidal zone. Institute for Coastal and Marine Resources, East Carolina University Technical Report No 4. 107 pp. REILLY, F. J. & V. J. BELLIS. 1983. The ecological impact of beach nourishment with dredged materials on the intertidal zone at Bogue Banks, North Carolina. U.S. Army Corps of Engineers, Coastal Engineering Research Center, Miscel- laneous Reports No. 83-3. 74 pp. SaASTRE, M. P. 1984. Relationships between environmental factors and Donax denticulatus populations in Puerto Rico. Estuarine Coastal and Shelf Science 19:217-230. SCHNEIDER, D. 1982. Predation by ruddy turnstones (Arenaria interpres) on a polymorphic clam (Donax variabilis) at Sanibel Island, Florida. Bulletin of Marine Science 32:341-344. SOKAL, R.R. & F. J. ROHLF. 1981. Biometry. W. H. Freeman: San Francisco. 859 pp. SPRING, K. D. 1981. A study of spatial and temporal variation in the nearshore macrobenthic populations of the central Florida east coast. M.S. Thesis, Florida Institute of Tech- nology, Melbourne, Florida. 67 pp. STAUBLE, D. K., M. E. HANSEN, R. L. HUSHLA & L. E. PARSONS. 1983. Biological and physical monitoring of beach erosion control project, Indiatlantic/Melbourne Beach, Florida. Part I: Physical monitoring. U.S. Army Corps of Engineers Dis- trict, Jacksonville, Florida. Unpublished Report, November 1983. TiFFANY, W. J., III. 1971. The tidal migration of Donax var- tabilis Say (Mollusca: Bivalvia). The Veliger 14:82-85. TuRNER, H. J., JR. & D. L. BELDING. 1957. The tidal mi- gration of Donax variabilis Say. Limnology and Oceanog- raphy 2:120-124. VeGA, R.R. & J. W. TUNNELL, JR. 1987. Seasonal abundance, zonation, and migratory behavior of Donax (Donacidae: Bi- valvia) on Mustang and Northern Padre Island, Texas. Mal- acology Data Net (Ecosearch Series) 1:97-136. WabeE, B. A. 1967a. Studies on the biology of the West Indian beach clam, Donax denticulatus Linné. 1. Ecology. Bulletin of Marine Science 17:149-174. WabDE, B. A. 1967b. Studies on the biology of the West Indian beach clam, Donax denticulatus Linné. 2. Life-history. Bul- letin of Marine Science 18:876-901. WabDE, B.A. 1967c. On the taxonomy, morphology, and ecology of the beach clam, Donax striatus Linné. Bulletin of Marine Science 17:723-740. The Veliger 35(4):366-380 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 ‘“Solen rosaceus’’—Three Species RUDO von COSEL Muséum National d’Histoire Naturelle, Paris, 55, rue de Buffon, F-75005 Paris, France Abstract. The small jackknife clams living on the coast of southern California, the outer coast of Baja California, and in the Gulf of California, and known as “‘Solen rosaceus Carpenter, 1864,” belong to three different species: the allopatric species pair S$. (Enszsolen) rosaceus Carpenter, 1864, from southern California, and S. (E.) gemmelli sp. nov. from the upper Gulf of California (San Felipe area), and the variable and widespread S. (E.) rostriformis Dunker, 1862, from southern California to La Paz. The subgenus Ensisolen Habe, 1977, is redefined. Neotypes of Solen rosaceus Carpenter, 1864, and Solen lappeanus Dunker, 1871, are designated. INTRODUCTION During preparative work for a revision of the western American Solenidae, I examined over 150 specimens of the small jackknife clams commonly arranged in collections and published in identification books and faunal lists under the name “‘Solen rosaceus Carpenter, 1864.” In spite of many citations, illustrations of these small bivalves are rather scarce. For example, the clam was not figured in such classic works as KEEP (1887, 1904), GRANT & GALE (1931), OLDROoYD (1925), KEEP & BarLy (1935), and ABBOTT (1954), and also not in some of the smaller and more general field guides, such as ABBOTT (1968); the figures in KEEN (1958, 1971), MCLEAN (1969), and POHLO (1963) are not accurate. There are, in fact, three species involved, which are quite close but differ in several constant characters. One species is described as new; the other two are redescribed and their taxonomy discussed. All belong to the subgenus Ensisolen Habe, 1977, which is redefined herein. Abbreviations used in the text: AMNH—The American Museum of Natural History, New York; BMNH—Brit- ish Museum (Natural History) (now “The Natural His- tory Museum’’), London, Great Britain; CAS—California Academy of Sciences, San Francisco; LACM—Los An- geles County Museum of Natural History, Los Angeles; MCZ—Museum of Comparative Zoology at Harvard University, Cambridge, Massachusetts; MNHN—Mu- séum National d’Histoire Naturelle, Paris, France; SDNHM-—San Diego Natural History Museum, San Di- ego; UCB—University of California, Berkeley; USNM— National Museum of Natural History, Smithsonian In- stitution, Washington, D.C; ZIM—Zoologisches Institut und Museum der Universitat Hamburg, Hamburg, Ger- many; ZMB—Zoologisches Museum der Humboldt- Universitat Berlin, Berlin, Germany. TAXONOMY Genus Solen Linné, 1758 Solen LINNE, 1758:672. Type species: Solen vagina Linné, 1758, by subsequent designation, Schumacher, 1817, central Indo-Pacific (see Figure 62 herein). Approximately 65 Recent species, worldwide, mostly tropical and warm-temperate. Shells small to very large, thin and fragile to solid, in shape variable but in general outline somewhat rectan- gular, from extremely long and narrow to rather short. Dorsal and ventral margin straight and parallel, straight and very slightly tapering, or valves more or less curved with convex ventral margin and dorsal margin concave, straight, or convex. Shells gaping at both ends. Anterior and posterior margin truncated or rounded to semicircular; truncated margins vertical or positively or negatively oblique. Beaks terminal, in species with semicircular an- terior margin appearing slightly subterminal. Hinge and ligamental area straight or slightly bent dorsally. Exterior with or without a more or less pronounced furrow parallel to the anterior margin or with a slight depression only. Anterior adductor scar long-oval to very long and slender. Posterior adductor scar oval to triangular, situated above the posterior pallial line and united with the dorsal limb of the pallial sinus or situated in front of the posterior pallial line and separated from the pallial sinus. Pallial R. von Cosel, 1992 Page 367 Explanation of Figures 1 to 7 Figures 1-7. Solen rosaceus Carpenter, all from California. Scale = 10 mm. Figure 1. San Pedro, Los Angeles (no data), neotype MNHN. Figure 2. San Diego (no data), AMNH 26307 (figured specimen in EMERSON, 1981:pl. 121, fig. 9). Figure 3. San Pedro, leg. E. P. Chace, MCZ 51594. Figure 4. San Pedro, leg. H. N. Lowe, 13 December 1924, MNHN, ex Staadt coll. Figure 5. San Diego, AMNH 26317, ex Oldroyd coll. Figure 6. San Pedro (no data), MNHN. Figure 7. Anaheim Landing, leg. E. P. Chace, MCZ 67379. sinus short to very short, triangular, trapezoid or nearly square. Hinge with one cardinal in each valve, no laterals. Periostracum thin, in fully grown specimens of larger spe- cies rather thick and strong. Subgenus Ensisolen Habe, 1977 Ensisolen HABE, 1977:228 [in Japanese]. Explanation of Figures 8 to 14 Figures 8-14. Solen gemmelli Cosel, sp. nov., all from the San Felipe area, Baja California, Mexico. Scale = 10 mm. Figure 8. San Felipe, between Playa Alicia and El Paraiso, on sand bars at low tide, /eg. Gemmell, holotype SONHM. Figures 9-13. Same locality as in Figure 8, paratypes SDNHM. Figure 14. 32 km S of San Felipe, intertidally, Jeg. F. B. Howard, April 1957, LACM 104778, ex Kanakoff coll. Type species: Solen krusensterni Schrenck, 1867, by orig- inal designation. Northern Japan, northward to Siberia, Ochotsk Sea (see Figure 60 herein). Fourteen to sixteen species, northwestern Pacific, east- ern Pacific, western Atlantic. Shells small to medium-sized, elongate to very elongate, and variable in length-width ratio, straight to somewhat curved, with general aspect like a short Ensis. Anterior margin rounded-truncate to nearly semicircular, posterior Page 368 The Veliger, Vol. 35, No. 4 30 Explanation of Figures 15 to 30 Figures 15-30. Solen rostriformis Dunker, “San Diego,” “San Felipe,” and “Mazatlan” forms. Scale = 10 mm. Figure 15. Santo Domingo, outer coast of Baja California, Mexico, leg. C. R. Orcutt, SDNHM 15442. Neotype of Solen lappeanus Dunker, 1871. Figure 16. Same locality as in Figure 15, SODNHM 15442. Figure 17. Bahia Concepcion, 1.6 km S of Punta Santo Domingo, AMNH 78609. Juvenile with shorter and broader shell. Figure 18. Upper Newport Bay, Orange Co., California, LACM 104777. Figure 19. Santa Barbara, California, Jeg. Forrer, ZMB, ex Dunker coll. Figure 20. Morro Bay, California, leg. M. & E. Caruthers, 1937, ZIM. Figure 21. San Diego Bay, California, MCZ 140200. Figure 22. San Diego Bay, California, 16-20 m, leg. C. A. Kofoid & W. J. Raymond, 13 July 1901, CAS I1Z039968. Figure 23. 32 km S of San Felipe, Baja California, Mexico, leg. F. B. Howard, April 1957, LACM 104778. Figure 24. San Felipe, Baja California, leg. H. N. Lowe, 1933, SDNHM 22576. Figure 25. San Felipe Bay, Baja California, between Playa Alicia and Pete’s Camp, sandbars at edge of low tide, /eg. Gemmell, 1964-1975, SDNHM 90140. Figure 26. 40 km S of San Felipe, Baja California, leg. E. P. Chace, March 1957, MCZ 215041. Figure 27. Same locality as in Figure 23. Figures 28-30. Mazatlan, Sinaloa, Mexico, on beach, leg. L. G. Hertlein, 8 December 1932, CAS 1Z039969. Page 369 R. von Cosel, 1992 Explanation of Figures 31 to 42 Figures 31-42. Solen rostriformis Dunker, southern “La Paz” form. Scale = 10 mm. Figure 31. Holotype BMNH 19771, no locality. Figures 32-35. La Paz, Baja California, Mexico, SDNHM 71460. Figure 36. El Magote, Puerto de La Paz, Baja California, Mexico, 24°10’N, 112°W, intertidal to 2.5 m, leg. McLean et al., 11 April 1966, LACM 66-29. Figures 37, 38. Bahia Magdalena, Baja California Sur (outer coast), intertidal sand flat, 0.8 km S of pier at Puerto San Carlos (24°47.4'N, 112°6.3'W), leg. C. Swift, 3 November 1971, LACM 71-186. Figure 39. Estero de Punta Banda, outer coast of Baja California (31°46.6'N, 116°37.3'W), leg. McLean, 20 December 1964, LACM 64-33. Figures 40, 41. Bahia Cholla, Puerto Penasco, Sonora, Mexico, AMNH 178035. Figure 42. Bahia Cholla, leg. R. B. Beck, MNHN, ex Staadt coll. margin truncate, with rounded corners to well rounded or nearly semicircular. Dorsal margin somewhat concave to slightly convex, ventral margin always slightly to markedly convex. Posterior third or fourth of the valve more or less tapering. Hinge and ligamental area slightly bent dorsally. No sharp furrow parallel to the anterior margin, but some- times with a very shallow, more or less broad depression. Posterior adductor scar always above the pallial sinus and united with its dorsal limb. The short diagnosis by HABE (1977) gives as the dis- tinguishing feature for Ensisolen only the anterior of the shell, which is slightly curved and prolonged towards the anterior; in contrast, Solen has a truncate anterior. Further differences between Ensisolen and all other Solen (at this time maintained under this genus without other subgenera, although a subdivision would be necessary) are the situ- ation of the posterior adductor scar above the pallial sinus and not in front of it, the more or less ensiform outline, and the lack of a deep and pronounced furrow along and parallel to the anterior margin. The anterior margin is indeed never straight and sharply truncate as in several other Solen (see Figures 62 and 63; for more examples, Page 370 Explanation of Figures 43 to 46 Figures 43-46. Solen rosaceus Carpenter. Scale = 10 mm. Figure 43. Neotype MNHN, San Pedro, California, leg. H. N. Lowe, 23 December 1924 (ex Staadt coll.). Interior and exterior of both valves. Figures 44, 45. Terminal Island, Los Angeles, California, leg. W. H. Eshnaur, 28 July 1928, MNHN, ex Staadt coll. Exterior of two left valves. Figure 46. San Diego, California, AMNH 26317, ex Oldroyd coll. Exterior of a left valve. see COSEL, 1989:195, 196, 198) but at least somewhat convex. In Ensisolen, the anterior and posterior corners are always more or less rounded (which makes, together with the often curved shell, the “‘ensiform” outline). The dorsal and ventral margins are never strictly parallel from end to end: the posterior taper is always more or less marked. Solen (Ensisolen) rosaceus Carpenter, 1864 S. sicarius var. rosaceus CARPENTER, 1864:536, 638 (reprinted 1872:22, 124); CARPENTER, 1865:177 (reprinted 1872: 279). The Veliger, Vol. 35, No. 4 “ % a TE oo) nS WORE es 50 Explanation of Figures 47 to 50 Figures 47-50. Solen gemmelli sp. nov., all from the San Felipe area, Baja California, Mexico. Scale = 10 mm. Figure 47. Ho- lotype, SDNHM 90139. San Felipe, between Playa Alicia and El Paraiso, on sandbars at low tide, /eg. Gemmell. Interior and exterior of both valves. Figures 48-50. Three paratypes, SONHM 90139, same locality. External views of right valves. S. rosaceus: MORRIS, 1952 (1960):55, pl. 15, fig. 4; MorrIs, 1966:39, pl. 23, fig. 2; ABBOTT, 1974:495, no. 5634 (fig.); EMERSON, 1981:681, pl. 121, fig. 9. Type material: The material on which Carpenter’s de- scription is based is indicated as coming from E. Jewett (CARPENTER, 1864:536, 1865:177) and J. G. Cooper (CARPENTER, 1865:177). The still present material of these collections is now in USNM, UCB, or the Redpath Mu- seum, McGill University. However, PALMER (1958) was unable to locate the type material of Solen rosaceus in one of these institutions and also not in BMNH (as erroneously stated by OLDROYD, 1925) (PALMER, 1958:25, 115); the material is apparently lost. For nomenclatural stability, a Explanation of Figures 51 to 59 Figures 51-59. Solen rostriformis Dunker. Scale = 10 mm. Figure 51. “San Diego” form. Santo Domingo, outer coast of Baja California, Mexico, leg. C. R. Orcutt, SDNHM 15442. Neotype of Solen lappeanus Dunker, 1871. Interior and exterior of both valves. Figure 52. Second specimen of same lot as in Figure 51. Exterior of left valve. Figure 53. “San Diego” form. San Felipe Bay, Baja California, between Playa Alicia and Pete’s Camp, sandbars at edge of low tide, /eg. Gemmell, 1964-1975, SDNHM 90140. Exterior of left valve, interior of right valve. Figure 54. “San Diego” form. Newport Bay, California, MCZ 63501, ex E. P. Chace coll. Interior and exterior of left valve. Figure 55. Second specimen of same lot as in Figure 54. Interior and exterior of right valve. Figure 56. “San Diego” form. San Pedro, California, leg. A. N. Lowe, December 1924, MNHN, ex Staadt coll. Interior and exterior of right valve. Figure 57. “La Paz” form. Holotype of Solen rostriformis Dunker, 1862. BMNH 19771. Interior and_exterior of both valves. Figure 58. “La Paz” form. La Paz, Baja California Sur, Mexico, SDNHM 71460. Exterior of both valves, interior of left valve. Figure 59. Second specimen of same lot as in Figure 58. Exterior of left valve. Page 372 AM AAS VM VPL The Veliger, Vol. 35, No. 4 DM PAS PPL VL Figure 60 Diagram of Solen krusensterni Schrenck (type species of Enszsolen). Sukhodol Bight, Ussuri Bay, USSR, on beach, leg. K. A. Lutaenko, 2 April 1989, MNHN. Explanation of shell characters: AAS, anterior adductor scar; AM, anterior margin; APL, anterior pallial line; AVC, anteroventral corner; DL, dorsal limb of pallial sinus (here united with the posterior adductor scar); DM, dorsal margin; H, hinge tooth under the beak; L, ligament; PAS, posterior adductor scar; PDC, posterodorsal corner; PM, posterior margin; PPL, posterior pallial line; PS, pallial sinus; PVC, posteroventral corner; VL, ventral limb of pallial sinus; VM, ventral margin; VPL, ventral pallial line. neotype is designated here: MNHN, San Pedro, leg. H. N. Lowe, 23 December 1924. Type locality: “Santa Barbara (Jewett); San Pedro (Coo- per)” (CARPENTER, 1865:177), here restricted to San Pe- dro, California (33°45'N, 118°19’W). Description: Shell small, up to 57 mm long, elongate, somewhat variable in length-width ratio (4.7-5.5:1), thin and fragile. Dorsal margin straight to very faintly concave, rarely somewhat convex; ventral margin generally dis- tinctly convex, often giving the valves as a whole a slightly curved appearance. Anterior margin well rounded and prominent; posterior third or fourth of the valves tapering, ventrally more than dorsally; posterior margin vertically truncate, with well-rounded corners. Broadest part of the valves in front of the posterior muscle scar. Hinge and ligamental area slightly bent upwards. Anterior adductor scar elongate, by % to 4 longer than the ligament. Posterior adductor scar above the pallial sinus and united with its dorsal limb. Pallial sinus short and rather narrow, with the innermost point usually in the middle or the lower part, occasionally in the upper part. Distance between innermost point of pallial sinus and posterior margin relative to the total shell length (“‘pal- lial sinus ratio”; for diagrams see Figure 60 and COsEL, 1989:204) 1:3.6-4.1. Exterior smooth and glossy, with fine, irregular growth lines. Valves whitish, with several concentric pale brownish- red zones parallel to the growth lines. Periostracum light olive greenish. Interior light grayish white with the brown- ish-red zones showing through. Animal not seen. Selected measurements with length—width ratio: 57.3 x 10.5 mm Long Beach, SDMNH 5),0)6 Ik 56.7 x 10.5 mm San Diego, AMNH 5.4:1 Figure 61 Diagram of Solen rostriformis Dunker. Upper Newport Bay, Orange Co., California, LACM 104777. Explanation of shell parameters: a = shell length; b = shell width; a:b = length-width ratio; c = distance from innermost point of the pallial sinus to the posterior margin; c:a = pallial sinus ratio (ratio of the distance between the innermost point of the pallial sinus and the posterior margin to total shell length). R. von Cosel, 1992 ——— 62 63 Explanation of Figures 62 to 63 Figures 62 and 63. Examples of Solen not belonging to Ensisolen. Figure 62. Solen vagina Linné, 1758. Possible syntype, Linnean Society of London. Type species of the genus Solen. Note the sharply truncated anterior margin; an anterior furrow is absent; the posterior adductor scar is united with the dorsal limb of the pallial sinus. Figure 63. Solen sloanu Gray in Hanley, 1842. The inflexion at the anteroventral corner marks the deep anterior furrow; the posterior adductor scar is situated in front of the pallial sinus. 55.6 x 10.4 mm Long Beach, SOMNH So) 8 tk 54.1 x 10.3 mm Long Beach, SOMNH Sol 52.8 x 10.4 mm Long Beach, SOMNH Bell ol 51.6 x 9.5 mm San Diego, AMNH 5.4:1 50.1 x 9.6 mm Long Beach, SOMNH 3/4 91k 50.0 x 9.7 mm Long Beach, SOMNH Seley 47.9 x 9.6 mm San Diego, AMNH 5.0: 1 [Bess3205 [noe Figure 64 Distributions-of Solen rosaceus (circles) and Solen gemmelli Cosel, sp. nov. (squares). The doubtful record from Morro Bay is marked by an empty circle. Page 373 [120° ——T0# Figure 65 Distribution of Solen rostriformis. The doubtful record from Mor- ro Bay is marked by an empty circle. 47.2 x 8.8 mm San Pedro, MCZ 5.4:1 44.8 x 8.8mm San Diego, AMNH pelligal 44.7 x 9.6 mm San Pedro, MCZ 4.7:1 44.4 x 8.7 mm San Pedro (neotype) oelcal 44.0 x 8.8 mm San Diego, AMNH 5.0:1 44.0 x 8.4mm San Diego, AMNH Bp) Il 43.8 x 9.2 mm San Pedro, MCZ 4.8:1 35.1 x 7.1 mm San Pedro, MCZ 4.9:1 30.0 x 6.0 mm San Pedro, MCZ 5.0:1 29.9 x 6.2 mm San Pedro, MCZ 4.8:1 Length / width ratio San Diego San Felipe La Paz Mazatlan Figure 66 Length-width ratio of Solen rostriformis populations from Cali- fornia and the outer coast of northern Baja California (“San Diego’), the San Felipe area in the northern Gulf of California (“San Felipe’’), northern to southern Baja California (“La Paz’’), and Mazatlan. Bars are 1 SD. Page 374 Pallial sinus ratio San Diego San Felipe La Paz Figure 67 Pallial sinus ratio of Solen rostriformis populations from California and the outer coast of northern Baja California (“San Diego’), the San Felipe area in the northern Gulf of California (‘San Felipe”), and northern to southern Baja California (“La Paz’). Bars are 1 SD. Distribution: Restricted to a rather short strip on the California coast. Usually, Santa Barbara (34°N) is given as the northern limit; a mixed lot in ZIM containing both this species and Solen rostriformis is labelled ‘““Morro Bay” (35°20'N), but this record needs confirmation. The species goes southward to San Diego (33°N). Material examined: The neotype; USA, CALIFORNIA: Morro Bay, 1 shell, M. & E. Caruthers, 1937, ZIM; Long Beach, Los Angeles, 7 shells, SDMNH, ex H. N. Lowe coll.; Terminal Beach, Los Angeles, 1 shell, /eg. Tremper, AMNH 206948; Terminal Island, Los Angeles, 4 shells, 5 valves, leg. Mrs. W. H. Eshnaur, 28 July 1928, MNHN, ex Staadt coll.; Anaheim Landing, San Pedro Bay, Los Angeles, 5 shells, leg. E. P. Chace, MCZ 67379; San Pedro, Los Angeles, 10 shells (no more details), MCZ 51594, ex E. P. Chace coll.; 5 shells (no more details), MNHN, ex Denis coll.; San Diego, 6 shells, AMNH 26317, ex I. S. Oldroyd coll., 1 shell, AMNH 26307 (specimen figured in EMERSON, 1981:pl. 121, fig. 9). Biotope: In fine muddy sand in protected bays, from the lower intertidal zone to shallow water, apparently only locally common. Remarks: This species is characterized by its typical pale brownish-red growth zones in combination with the nor- mally slightly curved shell, the rounded anterior margin, and the rounded-truncated posterior margin (one of the studied specimens had a straight shell, Figure 5). The species closest to Solen rosaceus in outline and arrangements of muscle scars and mantle line is S. tazrona Cosel, 1985, from the Colombian Caribbean coast; however, this South American species is a bit more slender, much smaller, thinner, and virtually translucent. The other close Atlantic species, S. viridis Say, 1821, from the U.S. east coast (Rhode Island to Texas) is translucent white (occasionally with a few rose growth lines in southern specimens) and has a pale yellowish green periostracum. The only eastern Pa- The Veliger, Vol. 35, No. 4 cific species resembling S. rosaceus is S. gemmelli sp. nov. (see description below). Solen sicarius, of which S. rosaceus was originally considered a “variety” by Carpenter, is much larger, heavier, and somewhat shorter and broader, with a less rounded posterior margin. This more northern species has a much wider range, from the extreme southeast of Alaska (56°N) (BERNARD, 1983) to north of San Diego (33°N, rare) (BERNARD, 1983); S. rosaceus overlaps with it in the greater part of its range. The type material of Solen rosaceus has never been fig- ured. From the brief descriptions (“‘Solen ?var. rosaceus. Straight, narrower, longer, smaller; glossy, rosy” —Car- PENTER, 1864:638; “‘Solen (?sicarus, var.) rosaceus. S. testa S. sicario simili, sed minore; multo angustiore, elongata, recta, extus et intus rosacea; epidermide tenui, valde ni- tente.”—CARPENTER, 1865:177), it is not clear which of the two sympatrically occurring species Carpenter really had before him. From the “similarity” to S. sicarius and the “rosy” coloration on the interior and exterior, and in spite of the citation “straight,” one could assume that the rather range-restricted California species treated above was concerned (although it is mostly slightly curved), but it is not completely sure. However, the final reason for selecting the neotype of Solen rosaceus from this species is nomen- clatural: using the name vosaceus for this species avoids the introduction of another new name. Solen (Ensisolen) gemmelli Cosel, sp. nov. (Figures 8-14, 47-50, 68, 69) Solen new species A: GEMMELL, Myers & HERTZ, 1987:57. Type material: Holotype SDNHM 90139, San Felipe, Golfo de California, Mexico, between Playa Alicia and El Paraiso, on sandbars at low tide mark, /eg. Gemmell, be- tween 1965 and 1976. Paratypes: Pete’s Camp to Playa Alicia, a coastline of about 50 km, stretching N and S of Bahia San Felipe, 5 specimens (3 partly broken), 3 valves, SDNHM 90139 [new number for paratypes]. Type locality: San Felipe, Golfo de California, Mexico (31°03'N, 114°52'W). Description: Shell rather small, up to 63 mm long, thin and translucent, very elongate, somewhat variable in out- line, slightly curved to occasionally straight, with length- width ratio 5.3-6.3:1. Anterior margin obliquely rounded- truncate, with well-rounded anteroventral corner. Pos- terior margin vertically truncate, slightly convex, with well-rounded dorsal and ventral corners. Dorsal margin somewhat concave to straight, rarely somewhat convex in its posterior part; ventral margin mostly slightly convex. Hinge and ligamental area faintly to markedly bent up- wards. Anterior adductor scar long and narrow, about % to % its length longer than the ligament. Posterior adductor scar united with the pallial sinus and stretching for half of its length above it. Pallial sinus triangular to trapezoid, with R. von Cosel, 1992 the innermost point mostly in the upper part. Distance between that innermost point and the posterior margin relative to the total shell length 1:3.4-4.0. Exterior smooth, with faint growth lines and coarser growth stages. Valves entirely white and lacking any col- oration, periostracum light yellowish green. Live or wet preserved animals not seen. Selected measurements with length-width ratio: 63.0 x 11.1 mm San Felipe (paratype) Soil Sk 60.0 x 9.5 mm San Felipe (holotype) O.2)9 1 57.3 x 9.6 mm San Felipe (paratype) 6.0:1 56.5 x 10.6 mm San Felipe (paratype) Deo) 8 Il 53.4 x 8.8 mm San Felipe (paratype) Gla 49.6 x 8.6 mm San Felipe (paratype) Dye) 3 Al 42.2 x 7.5 mm San Felipe (paratype) 6.2:1 36.8 x 7.0 mm San Felipe (paratype) SP oeul 26.0 x 4.8 mm San Felipe (paratype) are Ik Distribution: At present only known from the San Felipe area, Gulf of California, Pacific coast of Mexico. Material examined: The type material; MEXICO: 32 km S of San Felipe, Baja California, intertidal, 1 valve, leg. Faye B. Howard, April 1957, LACM 104778, ex Kana- koff coll. Etymology: The species is dedicated to Joyce Gemmell, who assembled an extensive marine mollusk collection from the San Felipe area between 1964 and 1975 (GEMMELL et al., 1987) and collected the species here described. Biotope: In fine sand, at lower intertidal zone and low water mark. Remarks: This new species is in outline very close to Solen rosaceus and S. tairona; however, the most conspicuous features that distinguish it from S. rosaceus are the com- plete lack of color and its generally longer and much more slender shell. It is the most slender species of the rosaceus group, only S. tazrona approximates it nearly (5.2:1 for S. tairona versus 5.3:1 for the “shortest” S. (E.) gemmellz). There are no substantial differences in the curvature of the margins and the muscle impressions; however, the anterior muscle scar and the ventral limb of the pallial sinus in S. gemmelli are more prolonged, corresponding to the longer shell. Solen rostriformis (see below), which occurs with S. gemmelli in the same habitat, is shorter and straight, with a rounded posterior margin, more trun- cate anterior margin, and a larger distance between the innermost point of pallial sinus and posterior margin. The southernmost record of Solen rosaceus is San Diego, and no record of this species on the outer coast of Baja California is known to me. This leads to the assumption that S. gemmelli and S. rosaceus might be a pair of allo- patric sibling species, the Caribbean and eastern Atlantic counterpart being S. tairona in the south and S. viridis in the north. Page 375 Solen (Ensisolen) rostriformis Dunker, 1862 Solen rostriformis DUNKER, 1862:421. Solen lappeanus DUNKER, 1871:129-130, pl. 44, fig. 1. Solen new species B: GEMMELL, MYERS & HERTZ, 1987:57. S. rosaceus: WEYMOUTH, 1920:50; pl. 15, fig. 3; Fircu, 1953: 76, fig. 42; MCLEAN, 1969:88, fig. 3; HADERLIE & ABBOTT, 1980:385, pl. 124, fig. 15.62; REHDER, 1981: fig. 615, p. 673. Type material: The holotype of Solen rostriformis is in BMNH (No. 19771). The type material of S. lappeanus has not been located, either in BMNH or in ZMB, and is apparently missing. For nomenclatural stability, a neo- type is designated here: SONHM 15442, Santo Do- mingo, Baja California, leg. C. R. Orcutt. Type locality: Solen rostriformis: not given, here selected as La Paz, Baja California Sur, Mexico (24°10'N, 110°17’W). S. lappeanus: ‘““Mare Antillarum,” here cor- rected to Santo Domingo, Baja California, Mexico (28°10'N, 114°08'W). Description: Shell small to medium-sized, up to 70 mm long, thin and fragile (very large specimens rather solid), elongate, very variable in length-width ratio (4.6-6.0:1), outline and coloration. Dorsal margin straight to faintly convex; ventral margin straight, slightly convex or occa- sionally even slightly concave in the middle part. Anterior margin more or less obliquely truncated, with rounded ventral corner, posterior margin well-rounded to nearly semicircular. Posterior part weakly to markedly tapering from just in front of or behind the level of the posterior adductor scar. Broadest part of the valves in the middle or behind the middle but usually more or less in front of the posterior adductor scar. Hinge and ligamental area slightly bent dorsally. Anterior adductor scar elongate, somewhat variable in length, slightly shorter or longer than the ligament. Pos- terior adductor scar above the pallial sinus and its posterior part united with the dorsal limb of the sinus. Pallial sinus short, with the innermost point mostly at the lower part. Distance between innermost point of the pallial sinus and posterior margin relative to the total shell length somewhat variable but always rather large: 1:2.7-3.6. Exterior with fine irregular growth lines and occasional coarser growth stages. Valve color varies from entirely white, whitish with rosy hue especially between the anterior adductor scar and the ligament plate or around the scar, slightly brownish pink, uniform pale pink or white to pale pink with occasional more intensively colored growth zones. Periostracum greenish to light brownish green, in very large specimens turning to brown. Interior with same coloration as exterior. Animal not seen. Selected measurements with length-width ratio: 69.5 x 13.2 mm Sto. Domingo (neotype of S. lappeanus), SDNHM 15442 3.0) 8 1 Page 376 68.0 x 13.0 mm Sto. Domingo, Baja Cal- ifornia, SONHM 15442 5.2:1 64.5 x 10.8 mm Upper Newport Bay, LACM 104777 6.0:1 64.4 x 10.7 mm Upper Newport Bay, LACM 104777 6.0:1 63.7 x 11.6 mm San Diego, MCZ 74369 5.5:1 59.8 x 10.6 mm Santa Barbara, ZMB 5.6:1 58.9 x 11.2 mm Upper Newport Bay, LACM 104777 53s) 58.4 x 11.6 mm Bahia San Felipe, SDNHM 90140 5:0) 57.5 x 11.0mm San Diego, AMNH 51594 Bi778 Il 56.9 x 12.3 mm San Felipe, SDNHM 22576 4.6:1 55.7 x 10.1 mm Santa Barbara, ZMB 5-5rall 55.5 x 11.8 mm Holotype of S. rostriformis, BMNH 4a 55.5 x 11.1 mm San Diego, MCZ 140200 5.0:1 53.6 x 9.3mm 40 km S of San Felipe, MCZ 215041 5.8:1 51.1 x 10.1 mm 40 km S of San Felipe, MCZ 215041 Sele 49.6 x 8.7 mm Tierra del Fuego Isle, SDNHM 54947 Sisal 49.2 x 9.6 mm Puerto San Carlos, LACM 71-186 Solel 49.1 x 10.5 mm Puerto Penasco, AMNH 178035 4.7:1 48.2 x 10.4mm Mazatlan, CAS 03969 4.6:1 46.7 x 8.8mm Newport Bay, MCZ 63501 Seoral 46.0 x 8.7 mm Mazatlan, CAS 03969 5a 44.0 x 7.9 mm Tierra del Fuego Isle, SDNHM 54947 5.6:1 42.0 x 9.0mm La Paz, SONHM 71460 4.7:1 41.6 x 8.6mm La Paz, SONHM 71460 4.8:1 36.7 x 7.5mm El Magote, La Paz, LACM 66-29 4.9:1 35.8 x 7.4mm Estero de Punta Banda, LACM 64-33 4.8:1 Distribution: Santa Barbara, California (34°N), south- ward to Mazatlan, Sinaloa, Mexico (23°N), and through- out the Gulf of California. As in Solen rosaceus, the mixed lot from Morro Bay in ZIM might suggest a range ex- tension to the north, but that needs confirmation. Material examined: USA, CALIFORNIA: Morro Bay, 1 shell, 1 valve, Jeg. M. Caruthers, 1937, ZIM; Santa Bar- bara, 2 shells, ZMB, ex Dunker coll.; Terminal Beach, Los Angeles, 3 valves, AMHN 206948, ex Tremper coll.; Terminal Island, Los Angeles, 1 valve, leg. Mrs. W. H. Eshnaur, 28 July 1928, MNHN, ex Staadt coll.; Newport Bay, Orange Co., 3 shells, leg. T. Burch, AMNH 131917; The Veliger, Vol. 35, No. 4 7 shells, leg. E. P. Chace, MCZ 63501; Upper Newport Bay, 7 shells, LACM 104777; San Diego Bay (no more details), 30 shells, MCZ 140200, ex Grand Rapids Public Museum; 3 shells, AMNH 51594, ex Constable coll.; 1 valve, AMNH 26917, ex Oldroyd coll.; San Diego Bay, 9-11 fm., 16 specimens, leg. Kofoid & Raymond, 13 July 1901, CAS 039968; San Diego (no more details), 4 shells, 1 valve, MCZ 74369, ex Hemphill coll.; 3 shells; MCZ 74370, ex Button coll.; 4 shells, MCZ 87121, ex Roper coll.; 1 shell, MNHN, ex Denis coll.; Tierra del Fuego Isle, Mission Bay, San Diego, 6 shells, /eg. R. L. Morrison, 11 April 1969, SDNHM 54947; Chula Vista, near San Diego, 1 shell, Jeg. Reed, ZIM; California (no details), 2 x 1 shell, MNHN; Mexico: Estero de Punta Banda, Baja California, 31°46.6’N, 116°37.3’W, intertidal sand and mudflats, 1 specimen, /eg. McLean, 20 December 1964, LACM 64-33; Santo Domingo, Baja California, 2 shells, leg. C. R. Orcutt, SODNHM 15442, ex Baily coll.; Bahia Santa Inez, 1.6 km S of Sto. Domingo Pt., Baja California, 2 shells, leg. C. R. Orcutt, SDNHM 15442, ex Baily coll.; Bahia Santa Inez, 1.6 km S of Sto. Domingo Pt., Baja California, 1 shell, 1 valve, AMNH 78609; Puerto San Carlos, Bahia Magdalena, Baja California Sur, 24°47.4'N, 112°6.3'W,, intertidal sand flat, 3 specimens, /eg. C. Swift, 3 November 1971, LACM 71-186; La Paz, Baja Califor- nia Sur (no more details), 5 shells, 2 valves SONHM 71460; El Magote, La Paz Harbour, 24°10'N, 112°00’W, | specimen, leg. McLean et al., 11 April 1966, LACM 66-29; 40 km S of San Felipe, Baja California, 7 shells, 1 valve, leg. E. P. Chace, March 1957, MCZ 215041; 32 km S of San Felipe, 21 shells, leg. F. B. Howard, April 1957, LACM 104778, ex Kanakoff coll.; Diggs Point, S of San Felipe, 4 shells, Jeg. E. C. Huffman, June 1934, LACM 104775; 3 shells, Jeg. Huffman, MNHN, ex Staadt coll.; San Felipe, in mud just below the surface, 2 speci- mens, leg. M. Rogers, 31 December 1955, LACM 104776; 6 shells, SONHM 225760, ex Lowe coll.; 2 shells, AMNH ex Chace coll.; Bahia San Felipe, between Playa Alicia and Pete’s Camp, sandbars at edge of low tide, 6 specimens, leg. Gemmell, between 1964 and 1975, SDNHM 90140; Bahia Cholla, Puerto Penasco, Sonora, 5 shells, February 1970, AMNH 178035; 1 shell, leg. Mrs. R. B. Beck, MNHN, ex Staadt coll.; Mazatlan, Sinaloa, leg. L. G. Hertlein, 7 valves, 1 fragment, 8 December 1932, CAS 1Z039969. Biotope: In fine sand and fine muddy sand from somewhat above low tide mark to shallow water (10 m). Remarks: Solen rostriformis is distinguished from S. rosa- ceus and §. gemmelli by its broader and less curved or straight shell, with truncated anterior and rounded pos- terior margins, and its white to whitish pink coloration, often with deeper pink in the anterior adductor scar region. This is an extremely variable species, which tends to develop more or less defined “forms” in the different parts of its distribution. Four are identified below. R. von Cosel, 1992 Los Angeles and San Diego area (one isolated lot from Santo Domingo, Baja California) (“San Diego” form): Shells up to about 70 mm long, dorsal margin straight or faintly convex, ventral margin straight or slightly con- vex, occasionally even somewhat concave; length—width ratio 5.2-5.7:1. Posterior part dorsally and ventrally only slightly tapering; broadest part of the valve mostly behind the middle, more or less near the posterior muscle scar. Distance from the innermost point of the pallial sinus to the posterior margin very large, always more than % of shell length, in fully grown specimens to % of total shell length (see Table 1). Valves whitish to beige-whitish, an- terodorsal often tinged with pale pink and occasionally with a few faint pinkish growth zones. Interior white, mostly with pinkish zone above the anterior adductor scar or a rosy hue around the scar. The two specimens from Santo Domingo are pale brownish rosy with a pink hue above the anterior adductor scar. West and south coast of Baja California (Estero de Punta Banda, Bahia Magdalena, La Paz) (‘Baja California” or “La Paz’ form): Shells to about 55 mm long, with the ventral margin convex over the whole length, dorsal margin straight, oc- casionally faintly convex but always less than ventral mar- gin; length—-width ratio 4.7-5.1:1. Posterior part mostly ventrally tapering from in front of the posterior muscle scar; broadest part of the valve in the middle or slightly before the middle, always in front of the posterior adductor scar. Distance between the innermost point of the pallial sinus and the posterior margin often slightly shorter than in the “San Diego” form (see Table 1). Valves entirely pale rosy or whitish with pink growth zones, rarely entirely white. Northwestern part of the Gulf of California (San Felipe area) (“San Felipe” form): Shells up to about 60 mm long, very similar to the San Diego population, but mostly somewhat shorter, with dor- sal and ventral margin straight or slightly convex; length— width ratio 4.2-5.8:1. Posterior part slightly tapering dor- sally and ventrally as in the San Diego specimens. Broadest part of the valves in the middle or behind the middle, slightly in front of the posterior muscle scar. Distance between the innermost point of the pallial sinus and the posterior margin generally shorter as in the San Diego specimens but occasionally as long (see Table 1). Valves entirely white, sometimes with a pale pinkish point near the beak and rarely with a more extended pinkish hue. Mazatlan (““Mazatlan” form): Shells apparently smaller than the more northern spec- imens (see measurements), dorsal and ventral margin faintly convex; length-width ratio 4.6-5.3:1. Posterior part dor- sally and ventrally tapering from the posterior adductor scar onward. Broadest part of the valves just in front of the posterior adductor scar. Distance between the inner- most point of the pallial sinus and the posterior margin Page 377 Table 1 Comparisons of the length—width ratio and the pallial sinus ratio (the ratio of the innermost point of the pallial sinus to total shell length) in the different morphs of Solen ros- triformis. Length-width ratio Ratio Mean SD SE n San Diego 5.2-5.7:1 5.485 0.297 0.082 13 San Felipe 4.2-5.8:1 4.983 0.398 0.083 24 La Paz 4.7-5.1:1 4.63 0.231 0.073 10 Mazatlan 4.6-5.3:1 4.92 0.259 0.116 5 Comparisons (* = significant at 95%) Mean Fisher Scheffe diff. PLSD F-test San Diego vs. San Felipe 0.502 0.234* 6.188* San Diego vs. La Paz 0.855 0.284* 12.205* San Diego vs. Mazatlan 0.565 0.356* 3.404* San Felipe vs. La Paz 0.353 0.256* 2.562 San Felipe vs. Mazatlan 0.063 0.333 0.048 La Paz vs. Mazatlan —0.29 0.37 0.829 Pallial sinus ratio Ratio Mean SD SE n San Diego 1:2.8-3.3 3.07 0.146 0.028 27 San Felipe 1:2.7-3.6 3.2 0.225 0.053 18 La Paz 12S d= 355 3.273 0.135 0.041 11 Mazatlan 1:2.6-3.0 — — — 2 Comparison (* = significant at 957%) Mean Fisher Scheffe diff. PLSD F-test —0.134 0.106* 3.009 —0.202 On25* 5.307* —0.073 0.133 0.599 San Diego vs. San Felipe San Diego vs. La Paz San Felipe vs. La Paz longer than in the Baja California/Gulf of California spec- imens (ratio of this parameter to total shell length in the two measured specimens 1:2.6 and 3.0). Valves pink with more intense coloration around the anterior adductor scar. The length-width ratios and pallial sinus ratios of the different populations are summarized in Table 1 and in Figures 66 and 67; a one-way analysis of variance (ANO- VA) was run to test for significant differences in the means of the two parameters among the four morphs. Looking at the studied material of these populations, the trend towards two main “forms” or “lineages” is ob- vious. A large, straight, usually quite slender, more whitish northern form lives from Santa Barbara, California, to Santo Domingo, Baja California, and is isolated from the outer coast of Baja California, in the extreme northwestern part of the Gulf of California. A smaller, generally some- what shorter, more rosy southern form with convex ventral margin ranges from Estero de Punta Banda, Baja Cali- fornia, southward to La Paz and perhaps along the east Page 378 Table 2 Comparisons of the length-width ratio and the pallial sinus ratio (the ratio of the innermost point of the pallial sinus to total shell length) in Solen rosaceus, S$. gemmelli, and S. rostriformis. Length-width ratio Ratio Mean SD SE n 4: 7-9-9071 | 5.126 | (0!233' (01054. 719 4.6-6.0:1 5.207 0.414 0.081 26 5.3-6.3:1 5.714 0.355 0.118 9 S. rosaceus S. rostriformis S. gemmelli Comparison (* = significant at 95%) Mean Fisher Scheffe diff. PLSD _ F-test S. rosaceus vs. S. rostriformis —0.8 0.213 0.218 S. rosaceus vs. S. gemmelli —0.588 0.285* 8.57* S. rostriformis vs. S. gemmelli = O50 0272) 161992# Pallial sinus ratio Ratio Mean SD SE n 1:3.6-4.1 3.842 0.209 0.045 22 1:2.7-3.6 3.152 0.194 0.026 56 1:3.4-4.0 3.741 0.212 0.08 uf S. rosaceus S. rostriformis S. gemmelli Comparison (* = significant at 95%) Mean Fisher Scheffe diff. PLSD _ F-test S. rosaceus vs. S. rostriformis 0:691 0.1* 94.922* S. rosaceus vs. S. gemmelli 0.101 0.172 0.683 S. rostriformis vs. S. gemmelli =O59 9 ONS 9% —27-255% coast of the Gulf of California northward. According to the studied material, the two forms seem to overlap be- tween Estero de Punta Banda (31°56’N) and Santo Do- mingo (28°10'N, 114°08’W) (see below). The two allopatric populations of the northern (“San Diego” and “San Felipe”) form differ slightly but signif- icantly in their length—width ratios and less in their pallial sinus ratios (see Figures 66, 67). ANOVA (length—width ratio: F = 13.00; df = 3, 47; P < 0.01; pallial sinus ratio: F = 6.32; df = 2, 53; P < 0.01) reveals for the difference in length—width ratio of the two populations a significance at the 95% level in the Fisher PLSD test and the Scheffe F-test; the pallial sinus ratios differ significantly (95% level) only in the Fisher PLSD test. The difference in length-width ratio between the north- ern “San Diego” population and the southern “La Paz” form is significant at 95% in the Fisher PLSD test and Scheffe F-test; the ‘““La Paz” form and the northern “San Felipe” population, however, differed significantly only in the Fisher PLSD. The difference in the pallial sinus ratio is significant at 95% in the Fisher PLSD and the Scheffe F-test between the “San Diego” and the “La Paz” forms; there is no significance at all in this parameter between The Veliger, Vol. 35, No. 4 Length / width ratio S. rostriformis Figure 68 S. rosaceus S. gemmelli Length—width ratios of Solen rosaceus, S. rostriformis, and S. gem- melli. Bars are 1 SD. the “La Paz” form and the “San Felipe” population. The “Mazatlan” form, with very few specimens at hand, was included in the length—width ratio analysis only; it showed a significant difference in comparison with the “San Di- ego” population only. Intergrades between the different forms seem to be not infrequent: the studied specimen from Estero de Punta Banda has the outline of the southern form but is entirely white. The coloration of the specimens from Mazatlan is like that in the southern “La Paz” form, but these spec- imens are straight with a long distance between the pallial sinus and the posterior margin, as in the northern “San Diego” form (for the “Mazatlan” form, this parameter could not be included in the ANOVA). Specimens from Bahia Cholla (Puerto Penasco), in the northeastern part of the Gulf of California, are of the “San Felipe” form; however, they might tend slightly to the “La Paz” form (Figures 40-42). Names are available for both forms: Solen lappeanus Dunker, 1871, for the large and straight northern ‘San Diego-Santo Domingo” variant and Solen rostriformis Dunker, 1862, for the shorter, slightly curved southern “La Paz” form. The holotype of Solen rostriformis most closely resembles specimens from the extreme southern part of the range (La Paz): it is very faintly rosy with weak pink growth zones, and the region of and above the anterior adductor scar is more brownish pink, being most intense just behind the beaks. The original figure of Solen lappeanus closely resembles the northern form and is closest to the maximum-size specimens from Santo Domingo, Baja California, with their light brownish interior and dark periostracum. The very shallow depression parallel to the anterior margin mentioned in the description and seen in the original figure is present in many specimens of this variant (e.g., Figures 51, 54). The neotype is hence selected from a lot from Santo Domingo. It corresponds also to the dimensions given for the figured specimen (72 x 11.5 mm). The two forms are treated here as one species. Analysis R. von Cosel, 1992 Pallial sinus ratio S. rostriformis Figure 69 S. rosaceus S. gemmelli Pallial sinus ratio of Solen rosaceus, S. rostriformis, and S. gem- melli. Bars are 1 SD. of much more material from many still unworked localities, especially on the east coast of the Gulf of California and the outer coast of Baja California (e.g., the “overlapping area’ between Estero de Punta Banda and Santo Domin- go), as well as an electrophoretic analysis of all populations and consideration of the fossil record could finally settle the status of these forms. CONCLUSIONS “Solen rosaceus Carpenter” as understood up to now con- sists in fact of an allopatric species pair, S. rosaceus and S. gemmelli, and another very variable species, S. rostriformis, which has an intermediate range and overlaps in its dis- tribution with both allopatric species. Table 2 and Figures 68 and 69 compare the length-width ratios and pallial sinus ratios of these three species. The principal differences between the species pair S. rosaceus/S. gemmelli and S. rostriformis are the general shell form, with a usually slight- ly curved appearance, posterior taper, rounded anterior margin, and truncate posterior margin in S. rosaceus/S. gemmelli, in combination with a markedly shorter dis- tance between the pallial sinus extremity and the posterior margin in the species pair. Solen rosaceus and S. gemmelli are themselves clearly distinguished by their different length-width ratios. For the length-width ratio (F = 9.215; df = 2, 82; P < 0.001), the difference between Solen gemmelli on one side and S. rosaceus and S. rostriformis on the other side is significant at the 95% level in the Fisher PLSD test and the Scheffe F-test. In S$. rosaceus and S. rostriformis, these parameters are not significantly different. In the pallial sinus ratio (F = 107.53; df = 2, 82; P < 0.001), however, S. rosaceus and S. gemmelli are significantly different from S. rostriformis (95% level), whereas there is no significant difference in this parameter between S. rosaceus and S. gemmelli. Solen rosaceus and S. gemmelli probably have derived from a common ancestor that originally had a continuous distribution. A similar phenomenon is observed in the straight “/appeanus” form of S. rostriformis; here, however, Page 379 the differences between the “San Diego-Santo Domingo” population and the “San Felipe” population are much smaller and concern mainly the length—width ratio. ACKNOWLEDGMENTS For the loan of material I am grateful to W. K. Emerson (AMNH), C. Hertz and A. d’Attilio(SDNHM), R. Kilias (ZMB), J. McLean (LACM), S. Morris (BMNH), B. Roth (CAS), and K. Boss (MCZ). Ch. Chintiroglou (Thessaloniki) was helpful with the statistical analysis. P. G. Oliver (Cardiff), S. Gofas (Paris), G. Coan (Palo Alto) and an anonymous referee are thanked for critically read- ing the manuscript and giving hints for improving it. LITERATURE CITED ApBoTT, R. T. 1954. American Seashells. Van Nostrand Co.: New York. 541 pp., 40 pls. ABBOTT, R. T. 1968. A Guide to Field Identification. Seashells of North America. Golden Press: New York. 280 pp., illus. ABBOTT. R. T. 1974. American Seashells. 2nd ed. Van Nos- trand Reinhold Co.: New York. 663 pp., 24 pls. BERNARD, F. R. 1983. Catalogue of the living Bivalvia of the Eastern Pacific Ocean: Bering Strait to Cape Horn. Cana- dian Special Publication of Fisheries and Aquatic Sciences 61:1-102. 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 Asso- ciation Advancement of Science 1863:517-686 (reprinted 1872:3-172). CARPENTER. P. P. 1865. Diagnoses of new forms of Mollusca from the West Coast of North America, first collected by Col. E. Jewett. Annals and Magazine of Natural History (3) 15:177-182 (reprinted 1872:277-289). CARPENTER, P. P. 1872. The mollusks of western North Amer- ica. Embracing the second report made to the British As- sociation on this subject, with other papers; reprinted by permission, with a general index. Smithsonian Miscellanea Collection 252:325 + 121 pp. CosEL, R. von. 1989. Three new species of Solen (Bivalvia: Solenidae) from the Indian Ocean, with remarks on the Solenidae of Madagascar. Journal of Conchology 33:189- 208. DUNKER, W. 1862. Solenacea nova collectionis Cumingianae. Proceedings of the Zoological Society of London for 1861: 418-427. DUNKER, W. [1871] 1858-1878. Novitates conchologicae. Ab- bildungen und Beschreibungen neuer oder wenig gekannter Conchylien. II. Abtheilung: Meeres-Conchylien. Theodor Fischer: Cassel. 144 pp., 45 pls. EMERSON, W. K. 1981. Mollusks. Jn: J. E. Ransom (ed.), Harper & Row’s Complete Field Guide to North American Wildlife. Western Edition. Harper & Row: New York. Fircu, J. E. 1953. Common bivalves of California. State of California, Department of Fish and Game, Marine Fisheries Branch, Fish Bulletin 90:102 pp., 63 figs. GEMMELL, J., B. W. Myers & C. M. HERTZ. 1987. A faunal study of the bivalves of San Felipe and environs, Gulf of California, from the Gemmell collection (1965-1976). The Festivus 18(Supplement):1-72. GranT, U.S. & H. R. GALe. 1931. Catalogue of the marine Page 380 Pliocene and Pleistocene Mollusca of California and adjacent regions. Memoirs of the San Diego Society of Natural His- tory 1:1036 pp., 32 pls. Hase, T. 1977. Systematics of Mollusca in Japan. Bivalvia and Scaphopoda. Hokuryu-kan: Tokyo. 372 pp., illus. HADERLIE, E. C. & D. P. ABBoTT. 1980. Bivalvia: the clams and allies. Pp. 355-411, pl. 112-128. In: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stanford University Press: Stanford, Califor- nia. KEEN, A. M. 1958. Sea Shells of Tropical West America. Marine Mollusks from Lower California to Colombia. Stan- ford University Press: Stanford, California. 624 pp., 10 pls., illus. KEEN, A. M. 1971. Sea Shells of Tropical West America. 2nd. ed. Marine Mollusks from Baja California to Peru. Stanford University Press: Stanford, California. 1064 pp., 22 pls., illus. KEEP, J. 1887. West Coast Shells. Bancroft: San Francisco. 230 pp., illus. KEEP, J. 1904. West American Shells. Whitaker & Ray: San Francisco. 360 pp., illus. KEEP, J. & J. Batty. 1935. West Coast Shells. Stanford Uni- versity Press: Stanford. 350 pp., illus. LINNE, C. VON. 1758. Systema naturae per regna tria naturae. 10th ed. Vol. 1. Regnum animale. Stockholm. 824 pp. The Veliger, Vol. 35, No. 4 McLean, J. H. 1969. Marine shells of southern California. Los Angeles County Museum of Natural History, Science Series 24, Zoology, 11. 104 pp., illus. Morris, P. A. 1952. A Field Guide to Shells of the Pacific Coast and Hawaii. Houghton Mifflin Co.: Boston. 220 pp., 40 pls. Morris, P. A. 1966. A Field Guide to Shells of the Pacific Coast and Hawaii including Shells of the Gulf of California. Houghton Mifflin Co.: Boston. 297 pp., 72 pls. OLpDROYD, I. S. 1925. The marine shells of the West Coast of North America. Vol. 1. Stanford University Press: Stanford, California. 247 pp., 57 pls. PALMER, K. VAN WINKLE. 1958. Type specimens of marine Mollusca described by P. P. Carpenter from the West Coast (San Diego to British Columbia). Geological Society of America: New York. Memoir 76:376 pp., 35 pls. PoHLo, R. H. 1963. Morphology and mode of burrowing in Siliqua patula and Solen rosaceus (Mollusca: Bivalvia). The Veliger 6(2):98-104. REHDER, H. A. 1981. The Audubon Society Field Guide to North American Seashells. A. Knopf: New York. 896 pp., 705 figs. WEYMOUTH, F. W. 1920. The edible clams, mussels and scal- lops of California. State of California, Fish and Game Com- mission, Contribution 22, State Fisheries Laboratory. 74 pp., 19 pls. THE VELIGER © CMS, Inc., 1992 The Veliger 35(4):381-383 (October 1, 1992) Identification of Monosaccharides in Hydrolyzed Nautilus Shell Insoluble Matrix by Gas Chromatography /Mass Spectrometry by MICHAEL J. S. TEVESZ anp STEPHEN F. SCHWELGIEN Department of Geological Sciences, Cleveland State University, Cleveland, Ohio 44115, USA BETH A. SMITH, DAVID G. HEHEMANN, ano ROGER W. BINKLEY Department of Chemistry, Cleveland State University, Cleveland, Ohio 44115, USA AND JOSEPH G. CARTER Department of Geology, University of North Carolina, Chapel Hill, North Carolina 27514, USA Abstract. This study provides the first survey of the monomeric constituents of hydrolyzed Nautilus pompilius insoluble shell matrix by gas chromatography/mass spectrometry. Identifying the monomeric constituents of hydrolyzed Nautilus insoluble matrix is an important first step in identifying the larger molecules of which the matrix is composed and, ultimately, postulating their function. This study revealed the presence of five monosaccharides not previously reported in the hydrolyzed matrix. These are fucose, xylose, mannose, galactose, and glucose, which together constitute <1% by weight of the matrix. The low concentration of these monosaccharides is not indicative of any structural role for these compounds. Three of these monosaccharides (galactose, mannose, and fucose) are monosaccharide residues often found in glycoproteins. INTRODUCTION Interest in identifying organic molecules in the organic matrix of mollusk shells is widespread. This kind of in- formation has been applied to areas as diverse as absolute age dating, stratigraphic correlation, and understanding biomineralizing processes (GREGOIRE, 1972; CRENSHAW, 1990; WEHMILLER, 1990). Moreover, studying the organic chemistry of surviving members of groups with an im- portant fossil history, such as Nautilus, has added impor- tance because it is the only direct means of understanding the biochemistry of these taxa (LOWENSTAM ef al., 1984). Although there has been progress in determining the gen- eral types of compounds present in Nautilus and other fossil and Recent shells, the identification of the primary struc- tures of particular macromolecules is in its early stages. The most detailed structural information obtained to date has been the determination of amino acids and amino sugars present after acid-catalyzed organic matrix hydro- lysis. This determination often has been done using amino acid analyzers (VOss-FOUCART, 1968; LOWENSTAM et al., 1984). In this study, we have employed gas chromatography / mass spectrometry (GC/MS) as a means of surveying the organic chemical composition of the hydrolyzed insoluble organic matrix of Nautilus pompilius Linné for compounds other than the amino acids and amino sugars known to be present (Voss-FOUCART, 1968; WESTBROEK et al., 1979). Page 382 This analysis is important because identifying the mono- meric constituents of hydrolyzed Nautilus matrix may al- low for the eventual identification of the larger molecules of which these residues are a part. This, then, could lead to a more complete understanding of the matrix and its function. The organic matrix is divided into two parts, that soluble in aqueous solution (soluble matrix) and that insoluble (insoluble matrix). We investigated the insoluble matrix. Use of GC/MS has allowed us to identify and quantify for the first time five monosaccharides not pre- viously reported in Nautilus insoluble matrix. MATERIALS ano METHODS Nautilus Shell Material The material analyzed in this study is the insoluble organic matrix of the shell composing the body chamber of adult Nautilus pompilius. The mineralogy of Nautilus is wholly aragonitic. Its microstructure has been described by several authors (e.g., MUTVEI, 1972; GREGOIRE, 1987; BANDEL, 1990). The outer shell wall has a very thin, inconspicuous periostracum, which is underlain by a layer of granules and spindle-shaped spherulites. The latter grade inward into nearly vertical to slightly reclined, irregular spherulitic prisms, nacreous structure, and then an inner- most layer of irregular simple prismatic to fibrous pris- matic structure. Preparation of Shells for Decalcification Air-dried Nautilus pompilius shells with clean-appear- ing shell surfaces were broken and separated into body chamber fragments. Body chamber fragments then were prepared for decalcification according to the following pro- cedures. The fragments were soaked in 100% Clorox for 22.5 hr, scrubbed with a nylon brush, and soaked again in Clorox for 0.5 hr. The brushing was necessary to remove from the surfaces of the septa material that did not dissolve in Clorox. The shell fragments were then washed and soaked in distilled water for 0.5 hr and then vigorously agitated in order to complete the washing. After drying, these fragments were broken with a mortar and pestle into pieces approximately 2 cm in diameter. Decalcification of Shells and Isolation of the Insoluble Matrix The di-sodium salt (100 g) of ethylenediaminetetraacetic acid (EDTA) was dissolved in 1 L of distilled water. The pH of the solution was adjusted to 7.0 using 1 N sodium hydroxide. Sodium azide (0.2 g), an antibacterial agent, then was added to the solution. Shell fragments (53 g) were placed in the EDTA solution and magnetically stirred for two days, at which time the shell fragments were de- calcified, leaving the brownish matrix as an insoluble res- idue. To ensure complete dissolution, stirring was contin- ued for an additional two days. The stirring then was The Veliger, Vol. 35, No. 4 halted and the contents of the flask allowed to settle over- night. Approximately 800 mL of the supernatant liquid was decanted. The remaining EDTA solution and sus- pended material (insoluble matrix fragments) were trans- ferred to 50-mL centrifuge tubes and the material centri- fuged until the organic matrix collected into a pellet at the bottom of the tube. The supernatant liquid was removed by pipetting and the pellet was shaken with distilled water. The suspension was centrifuged and the supernatant liquid again removed. This washing process was repeated twice. The centrifuge tubes then were placed in a vacuum des- iccator and evacuated using a mechanical pump (ca. 0.1 torr) for 2 hr. The matrix then was left under vacuum in a desiccator containing calcium chloride. Insoluble Matrix Hydrolysis, Derivatization, Analysis, and Quantification The dry insoluble matrix was heated with 4.0 mL of 4 N trifluoroacetic acid for 4 hr at 120°C in a sealed tube and thus hydrolyzed (cf. NESSER & SCHWEIZER, 1984). The volatile material was removed under reduced pressure and the residue was heated at 80°C for 15 min in a mixture of pyridine (1.0 mL) and bis(trimethylsilyl)trifluoroaceta- mide (BSTFA) (0.4 mL) containing a known amount of mannitol as an internal standard. Treatment with BSTFA created trimethylsilylated derivatives of matrix compounds and allowed for their analysis by GC/MS. Analysis was performed using a Finnigan TSQ 45 mass spectrometer operating in positive ion Q1 only mode (source tempera- ture = 120°C, source pressure = 0.03 torr, emission current = 0.3 ma, electron energy = 75 ev). Samples were intro- duced via the gas chromatograph using a 30 m x 0.25 mm Hewlett-Packard HP-5 ultra performance capillary column operating in the unsplit mode (He pressure = 10 psi, injector temperature = 230°C). The column temper- ature was maintained initially at 60°C for 1 min and then increased at 10°C /min to a temperature of 150°C. The temperature was held at 150°C for 5 min, then ramped to a final temperature of 280°C at a rate of 2.5°C/min. The temperature was held at this point for an additional 20 min. Compound identification was performed by compar- ison of both mass spectra and retention times to those obtained from authenic samples. A standard solution containing known amounts of 10 sugars and mannitol, as an internal standard, was deri- vitized according to the procedures described above for matrix analysis. Quantification was accomplished by com- paring peak areas from the matrix sample with those from the standard solution, and then comparing these peaks to those of the internal standard in light of appropriate re- sponse factors. RESULTS Five monosaccharides were identified in the form of silylat- ed derivatives in the matrix after hydrolysis and are listed in Table 1. Clearly, the amount of carbohydrate containing MES ers levesziciall\992 Table 1 Sugar analysis. Monosaccharides in hydrolyzed Nautilus pompulius shell insoluble matrix, by gas chromatography / mass spectrometry. Weight % of Compound Weight % Mole % total sample Fucose 4.1 4.5 0.018 Xylose 6.1 es 0.026 Mannose D3 24.9 0.11 Galactose 12.8 12.6 0.054 Glucose Sle7/ 50.8 0.22 7 weight sugars of the total sample = 0.428%. material present in the insoluble matrix which leads to these sugars is small (<1%). It is a small enough amount to raise the possibility that the carbohydrates may have come from bits of organic matter that were not removed from the shell during the cleaning process. This possibility certainly remains; however, when a second group of shells was cleaned and decalcified by a different person working in a different laboratory, the same sugars were found in the same amounts (after hydrolysis and derivatization). These experiments, taken together, suggest that the sugars listed in Table 1 are a part of the insoluble matrix. This represents the first evidence that these simple sugars are part of the complex macromolecules of which the matrix is composed. Certainly their function at present is unclear; however it seems reasonable that their presence in such low concentrations is not indicative of any structural role for these compounds. In addition to the monosaccharides found in the matrix, the following ones were also derivatized with BSTFA and analyzed as standards: talose, arabinose, altrose, ribose, rhamnose, and lyxose. Because the derivatives of these sugars were readily detected by GC/MS in standards but not in matrix samples, it is unlikely that they are present in Nautilus pompilius insoluble matrix at detectable levels. Of the five monosaccharides detected, three of them (ga- lactose, mannose, and fucose) are important monosaccha- ride constituents of glycoproteins (MONTGOMERY, 1970). Montgomery also reports that the two other important carbohydrate constituents of glycoproteins are the amino sugars glucosamine and galactosamine. In addition to the sugars reported above, we also detected the amino sugar glucosamine when we hydrolyzed the insoluble matrix with HCl, as have several previous workers (e.g., Voss-FOUCART, 1968; LOWENSTAM et al., 1984). While it is generally assumed that much of the glucosamine in mollusk insoluble matrix is in the form of chitin, our results suggest that in addition to studying insoluble matrix car- bohydrates in terms of their relationship to chitin, it may also be interesting to investigate the possibility of their being related to a thus far unreported glycoprotein fraction of Nautilus pompilius insoluble matrix. Although the study of glycoproteins in mollusk matrix is in its infancy, there Page 383 is some evidence of glycoproteinaceous material being pres- ent in mollusk matrix and the matrices of other calcium carbonate secreting invertebrates (CRENSHAW, 1972; LOWENSTAM & WEINER, 1989; COLLINS e¢ al., 1991). ACKNOWLEDGMENTS This project was supported by a Research Challenge Grant, Ohio Board of Regents to R.W.B., M.J.S.T., and D.G.H. The authors thank the Lewis Research Center (NASA) for donation of the mass spectrometer used in this work. LITERATURE CITED BANDEL, K. 1990. Cephalopod shell structure and general mechanisms of shell formation. Pp. 97-134. In: J. G. Carter (ed.), Skeletal Biomineralization: Patterns, Processes and Evolutionary Trends. Vol. I. Van Nostrand Reinhold. Couuins, M. J., G. Muyzer, G. B. Curry, P. SANDBERG & P. WESTBROEK. 1991. Macromolecules in brachiopod shells: characterization and diagenesis. Lethaia 24:387-397. CRENSHAW, M. A. 1972. The soluble matrix from Mercenaria mercenaria shell. Biomineralization 6:6-11. CRENSHAW, M. A. 1990. Biomineralization mechanisms. Pp. 1-9. In: J. G. Carter (ed.), Skeletal Biomineralization: Pat- terns, Processes and Evolutionary Trends. Vol. I. Van Nos- trand Reinhold. GREGOIRE, C. 1972. Structure of the molluscan shell. Chapter 2, pp. 45-102. In: M. Florkin & B. T. Scheer (eds.), Chem- ical Zoology 7, Mollusca. Academic Press: New York. GREGOIRE, C. 1987. Ultrastructure of the Nautilus shell. Pp. 463-486. In: W. B. Saunders & N. H. Landman (eds.), Nautilus—The Biology and Paleobiology of a Living Fossil. Plenum: New York. LOWENSTAM, H. A. & S. WEINER. 1989. On Biomineralization. Oxford University Press. 324 pp. LOWENSTAM, H. A., W. TRAUB & S. WEINER. 1984. Nautilus hard parts: a study of the mineral and organic constituents. Paleobiology 10:268-279. MontTGoMeryY, R. 1970. Glycoproteins. Pp. 628-709. In: W. Pigman & D. Horton (eds.), The Carbohydrates. 2nd ed., Vol. IIB. Academic Press. Mutvel, H. 1972. Ultrastructural relationships between the prismatic and nacreous layers in Nautilus (Cephalopoda). Biomineralization Research Reports 4:81-86. NESSER, J. R. & T. F. SCHWEIZER. 1984. A quantitative de- termination by capillary gas-liquid chromatography of neu- tral and amino sugars (as O-Methyloxime acetates), and a study on hydrolytic conditions for glycoproteins and poly- saccharides in order to increase sugar recoveries. Analytical Biochemistry 142:58-67. Voss-FoucaRT, M. F. 1968. Essais de solubilisation et de fractionnement d’une conchioline (nacre murale de Nautilus pompilius, Mollusque Cephalopode). Comparative Biochem- istry and Physiology 26:877-886. WEHMILLER, J. F. 1990. Amino acid racemization: applications in chemical taxonomy and chronostratigraphy of quaternary fossils. Pp. 583-608. In: J. G. Carter (ed.), Skeletal Biomin- eralization: Patterns, Processes and Evolutionary Trends. Vol. I. Van Nostrand Reinhold. WESTBROEK, P., P. H. VAN DER MEIDE, J. S. VAN DER WeEy-KLopPERS, R. J. VAN DER SLUIS, J. W. DE LEEUW & E. W. DE JONG. 1979. Fossil macromolecules from ceph- alopod shells: characterization, immunological response and diagenesis. Paleobiology 5:151-167. The Veliger 35(4):384-395 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 First Study on the Ecology of Sepia australis in the Southern Benguela Ecosystem M. R. LIPINSKI' Zoology Department, University of Cape Town, Rondebosch 7700, South Africa M. A. COMPAGNO ROELEVELD South African Museum, P.O. Box 61, Cape Town 8000, South Africa C. J. AUGUSTYN Sea Fisheries Research Institute, Private Bag X2, Roggebaai 8012, South Africa Abstract. Sepia australis is most abundant in the eastern South Atlantic between Luderitz and St. Helena Bay (about 27-33°S in 100-200 m). There seems to be no link between the variation in abundance of S. australis and that of its most important predator, the shallow-water Cape hake, Merluccius capensis. The variations in abundance of S. australis and one of its most important prey species, the stomatopod crustacean Pterygosquilla armata capensis, show simultaneous changes, suggesting that both species respond to the same environmental factors. Mantle length, total weight, gonad weight, and sex ratio of Sepia australis vary from year to year and by region off the west coast of southern Africa. Animals from the south coast (eastward of Cape Point) were very different: length-weight relationships were found to be similar in slope and intercept for both sexes and within each sex between years and regions off the west coast, but different for the south coast. INTRODUCTION Sepia australis is one of the most common sepiids in the Benguela ecosystem of southern Africa (in this instance from Luderitz to Agulhas Bank, Figure 1; for definitions and geographic boundaries see SHANNON, 1985). It is par- ticularly abundant off the west coast of southern Africa (Figure 1), as indicated by its frequent occurrence in bot- tom trawls (SANCHEZ & VILLANUEVA, 1989, 1991; LIPIN- SKI et al., 1991) and the abundance of its cuttlebones on some of the South African beaches (ROELEVELD, 1972: 231). Sepia australis is a small cephalopod with a maximum ' Present address: Sea Fisheries Research Institute, Private Bag X2, Roggebaai 8012, South Africa. dorsal mantle length (ML) slightly above 80 mm. Little is known about this species other than its systematic po- sition and distribution range; its importance in the eco- system is only now being assessed (LIPINSKI et al., 1991, and unpublished data; Roeleveld et a/., unpublished data). The species has a known distribution range in southern African waters from southern Namibia (ca. 27° S) to about Rame Head (31°30'S 29°20'E) on the south coast (ADAM & REES, 1966:145; ROELEVELD, 1972:228; SANCHEZ & VILLANUEVA, 1989, 1991; LIPINSKI et al., 1991; Roeleveld et al., unpublished data) and has also been reported from the Red Sea (a single doubtful record in ROCHEBRUNE [1884] repeated by ADAM [1942, 1959]). Its known vertical distribution is 2-457 m (ADAM & REES, 1966:145) but it occurs primarily in the upper 200 m (ROELEVELD, 1972: M. R. Lipinski e¢ al., 1992 277). The species was reported from southern Namibia only recently, with a center of abundance at 28-30°S in 75-250 m of water and a second, smaller center at 27°S in 150-300 m; its absence off northern Namibia was at- tributed to the anoxic conditions and narrow continental shelf (SANCHEZ & VILLANUEVA, 1989). Sepia australis has been caught by the R/S Africana in the course of regular biomass surveys conducted by the Sea Fisheries Research Institute off both the west coast and the south coast of South Africa. In 1986 the abundance of S. australis was merely estimated for catch records but during the latter part of the survey in January 1987 this species was found to form a significant part of the diet of the shallow-water Cape hake (Merluccius capensis) (LIPIN- SKI et al., in press). Since then, catch data for S. australis have been more carefully monitored (Augustyn e¢ al., un- published data) and extensive biological data, part of which form the basis for this study, have been collected. There is no fisheries-related data base, however, because this species is not fished commercially. Sepia australis has recently been found widely distributed along the south coast of southern Africa (east of Cape Point; Figure 1). It was shown to be an important species on the Agulhas Bank (Roeleveld e¢ a/., unpublished data), where it was found to be primarily a cold-water species, most abundant at bottom-water temperatures of 10-11°C. This agrees with the observation that the abundance of S. australis is higher off the west coast than off the south coast (Augustyn e¢ al., unpublished data). This study presents a first assessment of the biology of Sepia australis off the west coast of South Africa, in the southern Benguela ecosystem (Orange River to Cape Agulhas). The distribution, abundance, length/weight re- lationship, distribution of maturity stages, mean gonad weights, and sex ratio of S. australis are described for the first two detailed surveys of the species, in the northern region in 1987 and all three regions in 1988 (Figure 1). Results of the feeding analysis of these samples are de- scribed elsewhere (LIPINSKI et al., 1991). MATERIALS AND METHODS The biomass surveys of the Sea Fisheries Research Insti- tute are conducted by the R/S Africana, usually twice yearly off the west coast of South Africa and once yearly off the south coast. The west coast survey encompasses the southern Benguela ecosystem and extends from the Orange River mouth (ca. 29°S 16°E) to the Agulhas Bank (Figure 1). The south coast survey extends from Cape Agulhas (20°E) to Port Alfred (27°E) (Figure 1). These surveys consist primarily of half-hour bottom trawls during day- light, made in 1.5 X 1.5 m squares selected by the stratified semirandom method (ICSEAF, 1984; PAYNE et al., 1985) and are continuing. All specimens were collected with a 54-m German otter trawl with 1500-kg polyvalent doors, sampling on the sea- bed in depths of 50-500 m. Effective mesh size in the cod Page 385 end was 27.5 mm (using a pilchard liner). For each trawl, all components of the catch were sorted and weighed; key species such as hake, kingklip, monkfish, and squid were measured and subjected to biological analyses (length, weight, sex, maturity stage, stomach contents, etc.). Sepia australis catch data have been recorded in the course of these biomass surveys since 1986, in depths to 345 m, though data for the first year (January and July 1986) are only estimates and should be treated with caution. The actual weight of S. australis caught was recorded from the latter part of the survey (the northern region) in January 1987 and in all subsequent surveys. Almost all S. australis catches were made in the 0-209 m depth stratum, and only these data were further analyzed. Two species that show important interaction with Sepia australis are the shallow-water Cape hake (Merluccius ca- pensis), one of the sepiid’s main predators, and the stomato- pod crustacean Pterygosquilla armata capensis, one of its two main prey species (LIPINSKI et al., 1991). Only limited information is available for the other main prey species of S. australis, the sternoptychid lightfish Maurolicus muelleri (HULLEY & PROSCH, 1987; ARMSTRONG & PROSCH, 1991). The shallow-water Cape hake (Merluccius capensis) is not only one of the most important species in the Benguela ecosystem (both ecologically and commercially) but also one of the most important predators in the 0-209 m depth zone between northern Namibia and Cape Columbine (PAYNE et al., 1985, 1986, 1987, 1988, 1989). The geo- graphical distribution of M. capensis overlaps that of Sepia australis in the area investigated (PAYNE et al., 1985, 1986, 1987, 1988, 1989) and S. australis is both a competitor and a prey item of the shallow-water hake. Sepia australis also forms a link in the food chain between the mesopelagic Maurolicus muellert, which is consumed by small- and me- dium-sized hake only (PAYNE e¢ al., 1987), and large hake, which take S. australis but not M. mueller: (LIPINSKI et al., 1991). Relatively good data are available on the abundance of the stomatopod Pterygosquilla armata capensis (GRIFFITHS & BLAINE, 1988; ABELLO & MACPHERSON, 1990). There- fore, we attempted to compare the abundance of Sepia australis with that of Merluccius capensis and of P. armata capensis caught in the same depth stratum (0-209 m). The abundance indices (as catch per standardized trawl) for S. australis, M. capensis, and P. a. capensis were back-calcu- lated for all R/S Africana benthic biomass surveys (both west coast and south coast) for the years 1986-1990. The abundance of S. australis before January 1987 was roughly estimated (see above). Data from January 1989 were in- complete and are not included; during the latter survey, the data collected for S. australis and P. a. capensis were inadequate, because only 16 trawls could be made in the 0-209 m depth zone. Pterygosquilla a. capensis was not caught in any of the three south coast biomass surveys discussed here. For each survey the mean catch per trawl was calculated for all three species under consideration (Table 1). The Page 386 The Veliger, Vol. 35, No. 4 S 17>- ~ N Vive ~ 1) mn =). SS te =e : ik SA a » Hie N ne fe rd ‘ ni! So) vil oO. : vylt Bet ‘ . 6A) Pea] NS . We" Oo “ily SS tte yh AM oO =i : wylig Ss ANN e@ Al Vin ey FU NSN & e : aN CN \ e@ AN ty; \ Ww \ NON Mate \ 5 4 ee cae eernge: ar & # \ \ Sica! Oo iN a t Q @ ya ae es HN \ (Nec oO Port Nolloth nk = Aa IN ONS Gia VES : Ww Naw yao o iN. NEN 0@ oO x Sita Gs yah WAN eine 30 \ SOURS \oKY e@ VV " Ne Se @ yay | NS \ NENA ja ° } NS ‘A SOS o g au ° \ \ 1 \ ' v4 sab = [6] et 30 See \ ? x ee ‘ ee 1 ) c e ‘ 38° ‘ \ Ne oO ¥ 8 v\ \S © \\ ORTH \. \* ‘===.8 ‘ee Soe H a n Vis oes \ N Nagle se . ’ XO ‘o ‘ \ ~ > N A ee yy oO \ Yh oN x % NSS iY Cc} ORE EAN 1 Ye SA ‘ NS aN 1 \ NO ) VE NY , \ ‘\ NN y] \ oO \ SoS RRR SIN Aeron ee cucuaabacene D pmacsuei BDO Walssamrc erst ornate oc NN x N ‘ \ piNSseeZa 4 s . Wesuets \ oO \ ons \ NSS N \ We SS el et Nc NS = » SIO ' VANS ; RUNS — a hed SOS y Nay NG 5 RS YN Ma An . a \o WN \ ) ay yon a Seo Ne Nel Beers Van \ 8 Konda Awe y \ "oO \ VE \ 2 iy » ‘ nO) ) O: \ Oo , O S=s , He O Bs LN -O y 1 \ S ) st aN ‘ S / \ / oO’ ro yf UN ro einen a 37 CENTRAL 1 a\ a} a aS 2, \ yy > @ SAN K (o) ‘ X ¢ oo [e) NN > o @ < 3 & \ Sy 3 \ ny B i 1 @ \ \ We Oo F ‘\ ‘ Sa [ald Beko. i\ ~ 4 Se ‘ ~ StHelena Bay 7 eae SRG ON on ee a Sy ~-\_ 1Q ¢Cape Columbine yg 41g tSaldanha 2 (e} Ce The | eG can aNTNy 2S C-y Sa) se ) JAN. 1987 FEB.1988 \ CENTRAL 2 Ke \o \ 1 ee Bo ° es NeAait 7CAPE TOWN 34 ‘ \ 43 Bees oo) t & POR 6 EY (2 Pasa VIG Sati MINK ¢ MA Soop Sb ee ao See oa Ses aes ore fates = alr sone Raa ne SANS ic) Sv — ay x Ween .-S) ran \A > Ne ed A ‘ oon SA ANN IN ASN ‘et Cape Agulhas ea Ss WN GRY pees cea ~ SOUTH Ws 2 bey in : : ; We eS @ Trawls with Sepia australis, samples taken SOIR “ : : Ss. <1 4 AGULHAS O Trawls with Sepia australis, samples not taken See : ‘ f ANS OS BANK A Trawls without Sepia australis FASS > x ° NY x \ 36 Win SSO S3ANS Ss “SS ANS ° ° oO 16 18 18 E Figure 1 South African west coast biomass survey sampling area, from the Orange River mouth to Cape Agulhas, and localities of stations from which Sepia australis samples were collected for biological analysis 22-28 January 1987 and 5-23 February 1988 (closed circles). Stations at which S. australis was collected were grouped into northern, central (1 and 2), and southern regions, bounded by broken lines. mean number of animals per trawl was calculated from the mean individual weight for each species. For Sepia australis the mean individual weight was 20.4 g, calculated from regional means for 1987 and 1988 (Table 2). For Pterygosquilla a. capensis the mean weight was 11.9 g, calculated from data of GRIFFITHS & BLAINE (1988). The actual number and total weight of hake for each trawl were determined as part of the routine procedure in these surveys (PAYNE et al., 1985). The biological analysis is based on specimens collected in the northern region between 22 and 28 January 1987 (Figure 1 left, closed circles: 129 specimens from 9 stations) M. R. Lipinski e¢ al., 1992 Kg per trawl A S. australis & P. armata capensis 10 8 oO Pp LF OD Jan July Jan July Feb Aug Aug Jan SOne GO EO eo yOOre OG Number per trawl S. australis & P. armata capensis 500 400 300 Jan July Jan July Feb Aug Aug Jan 186) 166 67) 187) 88 1.88 WEST COAST x S. australis @ P. armata capensis Page 387 Kg per trawl M. capensis 400 300 Sept May May '90 '87 ‘88 '89 Number per trawl M. capensis 4000 3000 2000 1000 0 Sept May May '90 ‘87 ‘88 ‘89 SOUTH COAST O M. capensis Figure 2 Abundance of Sepia australis, Merluccius capensis, and Pterygosquilla armata capensis in southern African waters. Abundance in mean kg/trawl (A) and in mean number of individuals/trawl (B) for each survey off the west coast (see text), sampled in summer (January-February) and winter (July-August) and south coast (see text) in spring or autumn (September or May) in 1986-1990. Data for P. armata capensis were not collected in 1986; this species was absent in south coast surveys. Variance is not given to simplify the drawings; it does not affect the comparisons of relative abundance. and throughout the survey between 5 and 23 February 1988 (Figure 1 right, closed circles: 1150 specimens from 12 stations). These stations may conveniently be grouped into the northern, central 1, central 2, and southern regions, as indicated in Figure 1. The boundary between the north- ern and central 1 regions is arbitrary. There are no pro- found oceanographic differences between these regions. On the other hand, the oceanography and shelf morphology of the two central regions differ substantially (SHANNON, 1985), and the borderline between them at Cape Col- umbine reflects real differences. Unfortunately, no samples of Sepia australis were collected in the central 2 region. ‘sajdwres [enprarpul jo syySiam ues WOJJ payepnoyeo siaquUINN, yy “S[MP] [ENPIAIPUr 10} dJeUINSa UO paseg , The Veliger, Vol. 35, No. 4 0) 0) 0'0 0:0 6CC 0902 LYL CLO GE ole c0 Cy 6 S-8861 “9A UOIsaI UIYyINOG 469 €CL6 £8 LSI +L8 OFL'SI 8°99 L101 Cir lpr C8 COST 81 D-8861 “A worded [eUa1) 181 SPrre CC Olt 8S bLcl C 8h 1901 Cr9 660°F1 OTL CSC CC N-8861 ‘99H WOIBII UIYIION CSI OS6E 81 OLt G8 LOC? 6 8h bOLCcl Cr9 ZIL‘OL LY LT8E 9¢ N-Z861 ‘uel UOIBaI UIYWON SUOIBII ISVOD 1SIAA 0 0 0 0 SSE Ov0'CZ CCOL €6S0 OL bs Ice +0 6'SC o9 6861 Ae 0 0 0 0 Ore 68¢'0€ O'8r1 SLL ZI ©9 89bS SO 9'Cr L8 8861 API 0 0 0) 0) 881 GOc‘SI c 801 CS9L8 ol neko | c 0 Vel 18 L861 ‘dag jseoo YINOS 69€ CLy'sl vr 8'61C CSO 6S‘ LOL £68 SOP 6 Cry 8C1 C7 06 O1Sh OS 066] ‘uel Ss SOLE 90 CCE 9Pbe O87 °ZZLI 8 rIC CCHL OL 8LI 0688 KS bist OS 6861 ‘sny C81 1008 CC c'S6 Lol 8998 L8¢ VcOLl ble 9C0'TI 9S SST tr 8861 “‘sny 9CT ISPel 8c L091 8S¢e SIv 02 +09 CEPye Oly €68°€T CL CIP LS 8861 Ged 86 SPre cl Olt 802¢ COCL 09S F 1961 CL CC9T cl Ses SE L861 Aint 88 LIOLI ve GCOS C81 9CL‘Ol 819 lL or9e L8C L16 91 9°9 PLBE 6S L861 ‘uel 06 PLB8C Vl L9r SOc 6CLEl 9'OL LLEOe 8S B8hC cl *8 0S Cr 986l Ain’ 9CE S09‘91 O€ 9° Lol £9S CCL‘ 8C CHI GCLEL 8 Lol HENS +558 IS 986] ‘uel JSPOD ISOM [men «x ON [men 34 [meng ax ON [men 3y men xx ON men Sy s[Mey aeak eare AdAING /°N /3% /°N /3% / ON /3% jo'ON pure uo sisuaqvo DIDULLD ‘qi swsuadvo "JA suDbAjsnv “S ‘(] ANS Ose 29S) 0661-986] Ssead ay} Ul DUDIIU/pP S/yY UO sAdAINS Ysypunos3 jo synsay “We 6OZ-Q Ul swsuadvs njpwuv vpinbsosduajg pue ‘sisuadva snionp.apy ‘sypsjsnv viday 10J (SsIaquinu pue 1YySIaM) IuUeLpUNQe JO saoIpuUyT Le Page 388 M. R. Lipinski et al., 1992 Mean kg per trawl Jan. 87 Feb. 88 Figure 3 Comparison of abundance of Sepia australis off the west coast by region and for the entire survey in each of two summer surveys (1987 and 1988). Sigma—all trawls in the entire survey (up to 209 m); N, C, S—all trawls in northern, central 1, and southern regions, respectively (up to 209 m). The borderline between the central 2 and southern regions lies at Cape Point (Figure 2). Oceanographically the south- ern region belongs to the south coast, and the borderline between these regions also reflects real differences (SHANNON, 1985). Subsamples of Sepia australis for biological analysis were frozen immediately after capture and processed in the lab- oratory. Subsamples contained up to several hundred in- dividuals taken randomly from the sample (7.e., the whole catch per station per species). Biological analysis consisted of determination of dorsal mantle length (to the nearest millimeter), sex and maturity stage (according to a sim- plified version [ROELEVELD & LILTVED, 1985] of the uni- versal maturity scale for squids [LIPINSKI, 1979]), total weight, gonad weight, and weight of stomach contents Page 389 (results of stomach contents analysis were presented by LIPINSKI et al., 1991). RESULTS Abundance The most northerly station at which Sepia australis was caught during the west coast biomass surveys was at 27°58.8'S, 14°57.6'E (R/S Africana station E46, 190 m, 24 January 1989). During these surveys the species was found at a maximum depth of 345 m, the depth range of main abundance being 140-190 m. Mean catches per trawl of S. australis from nine west coast surveys and three south coast surveys in South African waters (Figure 2A, B) show considerable fluctuation, particularly off the west coast. The abundance off the south coast is more regular (Lipin- ski, unpublished data) but also much lower, by an order of magnitude, than off the west coast (Table 1). Sepia australis was generally more abundant in summer than in winter except in 1988 (Figure 2A, B). This general pattern was confirmed by the results of biomass calculations (Au- gustyn et al., unpublished data). In comparison with the shallow-water hake (Merluccius capensis), the most important fish species in the ecosystem in the 0-209 m depth stratum, the abundance of Sepia australis was sometimes higher, as regards to numbers (Fig- ure 2B), since S. australis is small and M. capensis attains a large size. Compared with the stomatopod Pterygosquilla armata capensis, one of its most important prey species, S. australis was more abundant both by weight and by num- bers (Figure 2A, B). A comparison of variations in the abundance of Sepia australis, Merluccius capensis and Pterygosquilla a. capensis show interesting trends (Figure 2A, B) that will be checked when a longer time series has accumulated. Available data Table 2 Sepia australis mean mantle lengths (ML), total weights (TW), with standard deviations (SD), and sex ratio, by year and region off the west coast (see Figure 1). Area Northern Northern Month, year Jan. 1987 Feb. 1988 Males (M) n 58 167 mean ML + SD 61.24 + 6.46 53.56 + 6.60 mean TW + SD 20.49 + 5.02 15.49 + 4.88 Females (F) n 71 234 mean ML + SD 65.61 + 7.78 58.37 + 7.82 mean TW + SD 24.86 + 7.15 19.84 + 6.50 Sexes combined n 129 401 mean ML + SD 63.64 + 7.51 56.37 + 7.65 mean TW + SD 22.90 + 6.63 18.03 + 6.20 Sex ratio, M:F 1:1.22 1:1.40 Central Southern Feb. 1988 Feb. 1988 Total 299 41 565 Os ae 7/40)) 50.02 + 4.17 VM 2 5.34 AME GSE 2227: 373 38 716 60.43 + 9.13 54.82 + 4.96 22.62 + 8.34 15.33 + 4.26 672 79 1281 58.25 + 8.60 52.33 + 5.14 20.20 + 7.66 13.41 + 3.84 1:1.25 1:0.93 iil 27 Page 390 The Veliger, Vol. 35, No. 4 = ; Most aet a ; Siig es: isa Saye (Figure 2A, B) suggest that variation in hake abundance BS ox eo SS) SS : e Caiealice oo ao! Se is not linked to those of S. australis or P. a. capensis; when Z aa) SiS So : : s & % VV VV VV the last mentioned two species are scarce, hake can ap- & Aw parently switch easily to other prey species. Nor does the = feels © <6 Ss < + abundance of S. australis seem to be influenced by that of cS) = : ) Ss hake. On the other hand, changes in the numbers of P. a. g 6 3 dl Qe ai RRS capensis seem to run parallel with those of S. australis. i 0 FS) coin eon In both 1987 and 1988, the abundance of Sepia australis ? r off the west coast varied greatly among regions (Figure oO BS ‘ Ze se Ze 3). In the southern region the mean catch per trawl was kerr} lon 4 4 4 + . . BE ae GF a g G G G similar to those off the rest of the south coast (Table 1), is) rol : 2 ||c2 whereas in the central 1 and northern regions the abun- a € om a = 23 oo 2090 dance was much higher (Figure 3), especially in the north- E 2 iE g N SS SNS ern region. o . . ~ 2 cal we = 2 © To explore further the relationship between the abun- Ca) ge By eee ne ee sakes dance of Sepia australis and that of hake and stomatopods, G2 the number of times that increases and decreases in oc- Q, < . = = pe, . . . . . oO [eee =5 =5 Ss currences were in the same or opposite directions (Figure ES Es a | & S S é S S S 2A, B) was counted. The results were as follows: Se PX Ss a alll ees Sepia australis Lop || 20 roan) cm ++ Talal =o increase decrease oe aN os Rome) mo) orl . Ko ar! ++ os oS Hake increase 2 1 er leg decrease 1 2 64 = un oO ot — 8 Set |e Pes a oO © © Stomatopods _ increase 3 0 — Z No N+ co J? = O05 o> decrease 0 3} = S = 2 % Ss Ss Ss The Fisher Exact Probability Test fora 2 x 2 contingency o So 2s ca ai GS as table was used to examine these relationships. For hake 3 2 eal Ss the relationship was not significant (P = 0.50) but it was S : 0 Be 2 Seale > So 29 significant for stomatopods (P = 0.05). Gb ae? c|c Ss = aa Sales oe = = ties ee Bes Population Structure S eS mw to conve) Ss 0 omN ar ~ i) : 0 pare) [PZ Nee ae == In all surveys, female Sepia australis were larger and = IS , heavier than males and hence showed greater variation aes eae = 5 Sn 35 about the mean (z.e., larger standard deviation). Dorsal as > 0 | a S6 So Se >elesa SS BS So mantle length (ML) and total weight showed large vari- g =, Sie aay) Mi Yow ations from year to year (Tables 2, 3). In the northern o ¥ Si Res ee ante region, differences of 7.27 mm in mean ML and 4.87 g ise) oo . . fe) . iS ws ies = 5 33 ee in mean weight (both sexes combined) were recorded in € ~|2s8 ae me oat successive years for the same season. Within one year Saline Oy] xs ae Bee eae (1988) the ML and total weight of S. australis in the central ES A 1 region were significantly greater than in the northern 2 = ie. WA) Paiva Byte) and southern regions (Table 3). = O° oo SS) é ; So iis & | a, a) ao) Se The length-weight relationships (Figure 4) were found 3 _ a SS) so ee aes 32 |S e VV VV VV to be similar for both sexes and within each sex between 3 2 ee years and regions, except in the southern region in 1988. Bp) I) Be eh is a a S 3 a a The length-weight relationships for Sepia australis in the a 5 southern region had a somewhat different intercept and (3) el i 2 ae SS % SA slope for both males (4.6 x 10° g and 2.00 g mm', S Z Pree 2 Bes respectively) and females (1.3 x 10° g and 2.34 g mm") is than in the northern and central 1 regions in both years nS is (males: 1.1-1.3 x 10°% g and 2.36-2.38 g mm‘'; females: S) as js 2 = = 0.76-0.91 x 10-3 g and 2.44-2.48 g mm‘|, respectively). = fs 2 fs G Su : The animals in the southern region were also smaller than = @S$S ¢S$8S 23 SS in the other regions (see above). The lower correlation g SFE GEE EEE fficients in th th i robably due pri- S Ss ay & coefficients in the southern region are p y p marily to the smaller sample size. M. R. Lipinski et al., 1992 100 A Males 50 N'88: W = 0.0013 L2.36 (r = 0.96, n = 167) N '87: W=0.0011 L238 (r = 0.95, n = 58) & C '88: W = 0.0012 L2.37 (r = 0.95, n = 299) S '88: W = 0.00 46 L2.00 (r = 0.86, n = 41) Total weight (g) ° Page 391 B Females C '88: W = 0.00085 L2.47 (r = 0.96, n = 373) S '88: W = 0.0013 L2.34 (r = 0.84, n = 38) N '87: W = 0.00076 L2:48 (r = 0.94, n= 71) & N ‘88: W = 0.00091 L2.44 (r = 0.95, n = 234) 20 50 100 20 50 100 Mantle length (mm) Figure 4 Sepia australis length-weight relationships by year and by region off the west coast in summer 1987 and 1988: Male (A) and female (B) regression lines are virtually the same but have been plotted separately to facilitate comparison among regions; appended to the regression lines are the values for each slope and intercept in the relationship W = aL? and the correlation coefficient (r) and sample size (n). N, C, S—north, central 1, and southern region, respectively; W is weight in g; L is length in mm. The distribution of maturity stages varied between years and regions (Figure 5). In the northern region in January 1987, most of the males (56.9%) and females (73.2%) were mature, whereas in February 1988 most animals were completely immature (56.9% and 77.8%, respectively). Mature and maturing males and females were generally bigger than immature animals and there were always slightly more females present than males, except in the southern region, where most of the animals were immature and the sex ratio was about 1:1 (Table 2). In the central 1 region the proportion between mature and immature males was close to parity; there were, however, consider- ably more immature females. Mean gonad weights for the different sexes and maturity stages are presented in Figure 6. In the northern region in January 1987, the mean gonad weight for maturity stages II and III was about the same and considerably bigger than for stage I in both sexes. In February 1988, the mean gonad weight showed a more regular increase in weight from stage I to III in both sexes in all three regions, although the mean gonad weight for stage II was somewhat higher in the central 1 region and there were no fully mature animals of either sex in the southern region. This is in agreement with the observation that in the northern region most of the animals were fully mature or almost so in 1987, whereas in 1988 the animals with the most developed reproductive systems were in the central region, where most of the males and almost half the females were fully mature and stage II animals had somewhat larger gonads (Figures 5, 6). DISCUSSION Abundance Our results confirm that the northern boundary of the distribution range of Sepia australis off the west coast of Page 392 The Veliger, Vol. 35, No. NORTHERN REGION 1987 107 58 Males MI Soe 107 71 Females Fl ISS 9 (45 O%) M \| 12 1% 9 (55.0%) Fll 11.3% 8 M Ill 56.9% 8 F Ill 73.2% iT i 6 6 S) >) 4 4 5 3 2 2 | 1 0) 6) 37 41 45 49 53 57 61 65 69 73 77 81 85 37 41 45 49 53 57 61 65 69 73 77 81 85 NORTHERN REGION 1988 207 167 Males Ml 282% 4g, 254 Females Fl 77.8% eee Mil 10.2% FIl 7.7% M11 32.9% : 16 20 F Ill 145% 14 12 15 10 8 10 6 4 5 2 Se (0) A 4 J a 0) 1a A S Sif Al 452.49) 535, 57 oil 65.69 97S ar7i78il) 85 57 41 45 49 53 57°61 (65) 69 7Sia77esines = ion © CENTRAL REGION 1988 a 30, 299 Males Ml 44.1% 307373 Females Fl S8,1% ; (44.5%) Milli, 18'4% (55.5%) FIl 11.0% 25 M Ill 47.5% F Ill 35.9% 20 57 41) 45) 49) 55° 57 Gill (65) 69" 73" 7/78)! 85: 37 41 45 49 53 57 mos oe) 7S 77 3 SS SOUTHERN REGION 1988 10; 41 Males MI 95.2% 10,38 Females Fl 89.5% g (51.9%) MIl 48% 3 (48.1%) FIl 10.5% 77] Uf 6 6 5 5 4 4 3 5) 9 2 | 1 @) O 37 All 45°49) SSM S7 16165 1692735 7G 85 37 4, 45: 49458) 57-16) 65 CONS aeilines Mantle Length (2 mm classes) | om Zia OF ill Figure 5 Sepia australis maturity stages: frequency of occurrence compared in successive years off the west coast in summer (1987-1988) in the northern region and by region in 1988. Sample size and percentage of each sex are indicated for each region. M—males; F—females. M. R. Lipinski e¢ al., 1992 3 _ MALES M II FEMALES Gonad weight (g), mean and standard deviation F | rl wu Fal F il M Ill F lll Page 393 MI MI Mill; Mi Mil Mil aaa STAGES Fal al sll F ll MATURITY STAGES N 1987 N 1988 C 1988 S 1988 REGIONS AND YEARS Figure 6 Mean gonad weight and standard deviation for Sepia australis by maturity stage and by region in summer 1987 and 1988, off the west coast of southern Africa. southern Africa lies close to 27°00’S in Namibian waters, as established by SANCHEZ & VILLANUEVA (1989). They reported fairly high concentrations of this species, but the units of measure (individuals per mile) are unfortunately not comparable with those presented here. However, their report of a consistently high abundance of S. australis is in accordance with our observations of its great abundance and importance in the Benguela ecosystem (Figure 2; LI- PINSKI et al., 1991). The main concentrations of S. australis are located between St. Helena Bay and Elizabeth Bay, south of Luderitz (SANCHEZ & VILLANUEVA, 1989, 1991; Augustyn e¢ al., unpublished data; our data). Its distri- bution seems to be fairly uniform in the 100-209 m depth zone (our data; Augustyn e¢ a/., unpublished data; SANCHEZ & VILLANUEVA, 1989, 1991). The white-chinned petrel (Procellaria aequinoctialis) seems to be an important consumer of Sepia australis, as most of the Sepioidea reported from the gizzards and stomachs of these birds by JACKSON (1988) probably belonged to this species (Lipinski, unpublished data). The white-chinned petrels, however, most probably scavenge dead S. australis from the sea surface (LIPINSKI & JACKSON, 1989) and therefore the abundance of the birds does not influence the abundance of S. australis. The Cape hake is the most important predator of Sepa australis identified to date (LIPINSKI e¢ al., in press); yet Page 394 there seems to be no link between the changes in abundance of these two species (Figure 2A, B). On the other hand, existing data show parallel changes in abundance for S. australis and Pterygosquilla a. capensis for at least six trawl surveys (see results of Fisher’s Exact Probability Test), which suggests that these two species may respond simi- larly to biological or environmental factors, or some com- bination of these. It is unlikely that these changes reflect predator-prey interactions because there is no time lag (Figure 2A, B), as predicted by the classical Lotka-Vol- terra model and its more sophisticated modifications (e.g., BEGON & MorTIMER, 1981:119-128). The interaction between Sepia australis and its other important prey species, the lightfish Maurolicus muelleri, cannot be assessed at present. Data for M. mueller: are scarce and difficult to interpret (M. Armstrong, personal communication). It would seem, however, that the lightfish also shows seasonal (summer-winter) fluctuations; con- centrations in summer have a higher density, especially in the Cape Columbine area (ARMSTRONG & PROSCH, 1991). The trends in the abundance of Sepza australis reported here are essentially in agreement with the biomass esti- mates found by Augustyn et al. (unpublished data). In addition, there were regional differences (Figure 3A, B), and catches (kg/trawl) were much higher in the northern and central regions than the overall mean value for the entire survey (Table 1). This assessment of abundance for Sepia australis is almost certainly an underestimate of total S. australis biomass, because of the sepiid’s small size, and the relatively large mesh size in the cod end. The catchability coefficient of S. australis must be very different from that of hake, the main target species in the biomass surveys. This suggests that the abundance of S. australis is far greater than reported here and by Augustyn et al. (unpublished data). Population Structure Both length and weight of Sepia australis seem to vary from year to year in the central 1 and northern regions. Animals from the southern region are, however, very dif- ferent (Table 2): they are much smaller and show less variation (smaller standard deviations). This is in agree- ment with the data for the south coast (at a different time of year, May 1988), where most of the S. australis were sexually maturing or mature (Roeleveld et a/., unpublished data). For the squid 7odarodes pacificus, the length-weight re- lationship has been found to be an important indicator of stock conditions (biological characteristics and stock iden- tity) in various areas and different years (MURATA, 1978). The length-weight relationships in Sepia australis were found to be similar in slope and intercept for both sexes and within each sex between years and regions, except in the southern region in 1988 (Figure 4). This similarity (especially between sexes) agrees with observations of ROELEVELD (1972:278) and SANCHEZ & WVILLANUEVA (1991), and was also reported by BELLO (1988) for 5S. The Veliger, Vol. 35, No. 4 orbignyana and S. elegans in the southern Adriatic Sea. This similarity, however, may well have been the result of the relatively uniform size within the samples investi- gated as well as the small data sets available. The analysis of maturity stages revealed considerable variation between years and regions. The general char- acteristics of the animals from the southern region in 1988 most resemble those of animals off the south coast (Roe- leveld e¢ al., unpublished data) but differ in maturity. Most of the Sepia australis off the south coast in May 1988 were either maturing or fully mature (Roeleveld e¢ a/., unpub- lished data); near Cape Point (34°18’S, 18°30’E) in Feb- ruary 1988, most of the specimens were immature (Figure 5). This difference in maturity may be attributed to the three-month time difference. On the other hand, animals from the southern region differed considerably from those of the northern and cen- tral region in abundance, sex ratio, mean length and weight, length-weight relationship, and distribution of maturity stages. All these biological differences indicate that Sepia australis off the south coast may belong to a separate pop- ulation. ACKNOWLEDGMENTS We thank the Director of the Sea Fisheries Research Institute, Cape Town, the Captain and crew of the R/S Africana, and colleagues on the cruises for enabling and assisting with the collection of material. We also gratefully acknowledge Prof. J. G. Field, Marine Biology Research Institute, University of Cape Town, for comments on the manuscript; Prof. L. G. Underhill, Department of Math- ematical Statistics, University of Cape Town, for assistance with statistics; Miss Monique le Roux, University of Cape Town, for technical assistance; Mr. A. van Dalsen and his team at the Sea Fisheries Research Institute and Messrs. C. Hunter and V. Branco and Mrs. M. G. Van der Merwe, South African Museum, for assistance with the figures; and Dr. L. J. V. Compagno of the South African Museum for access to his computer equipment for the preparation of illustrations. We also thank the Trustess and Directors of the South African Museum for financing the participation of M. A. C. Roeleveld in the First International Sepia Symposium, Caen, France, May-June 1989. The contribution by M. R. Lipinski forms part of the Benguela Ecology Programme, sponsored by the South African National Committee for Oceanographic Research of the Foundation for Research and Development. LITERATURE CITED ABELLO, P. & E. MACPHERSON. 1990. Influence of environ- mental conditions on the distribution of Pterygosquilla armata capensis (Crustacea: Stomatopoda) off Namibia. South Af- rican Journal of Marine Science 9:169-175. ADAM, W. 1942. Les Cephalopodes de la mer Rouge. Bulletin de l'Institut Oceanographique, Monaco 822:1-20. ADAM, W. 1959. Les Cephalopodes de la mer Rouge. Jn: Mis- M. R. Lipinski e¢ al., 1992 sion Robert Ph. Dollfus en Egypte (Decembre 1927—Mars 1929). S.S. ’Al Sayad.’ Resultats scientifiques, 3e partie (XXVIII). Centre National de la Recherche Scientifique, Paris, pp. 125-193. ApaM, W. & W. J. REES. 1966. A review of the cephalopod family Sepiidae. Scientific Reports. The John Murray Ex- pedition 1933-34, 11(1):1-165. ARMSTRONG, M. J. & R. M. PRoscH. 1991. Abundance and distribution of the mesopelagic fish Maurolicus mueller: in the southern Benguela upwelling system. South African Journal of Marine Science 10:13-28. Brecon, M. & M. MortTIMeR. 1981. Population Ecology. Blackwell Scientific Publications: Oxford. 200 pp. BELLO, G. 1988. Length-weight relationship in males and fe- males of Sepia orbignyana and S. elegans (Cephalopoda: Se- piidae). Rapport et Proces-verbaux des Reunions. Commis- sion Internationale pour |’ Exploration Scientifique de la Mer Mediterranee 31(2):254. GRIFFITHS, C. L. & M. J. BLAINE. 1988. Distribution, pop- ulation structure and biology of stomatopod Crustacea off the west coast of South Africa. South African Journal of Marine Science 7:45-50. HUuLLey, P. A. & R. M. ProscH. 1987. Mesopelagic fish de- rivatives in the southern Benguela upwelling region. South African Journal of Marine Science 5:597-611. ICSEAF. 1984. Report of the Standing Committee on Stock Assessment (STOCK). Annex 4—Report of the planning group on coordination of research (COORD). Proceedings and Reports of Meetings. 1988(II). International Commis- sion for the Southeast Atlantic Fisheries:83-103. JACKSON, S. 1988. Diets of the white-chinned petrel and sooty shearwater in the southern Benguela region, South Africa. The Condor 90:20-28. LiPINSKI, M. R. 1979. Universal maturity scale for the com- mercially important squids (Cephalopoda: Teuthoidea). In- ternational Commission for the Northwest Atlantic Fish- eries, Research Document 79/II/38:1-40. Lipinski, M. R. & S. JACKSON. 1989. Surface-feeding on ceph- alopods by procellariiform seabirds in the southern Benguela region, South Africa. Journal of Zoology, London 218:549- 563. Lipinski, M. R., A. I. L. PAYNE & B. Rose. In press. The importance of cephalopods as prey for hake and other groundfish in South African waters. Jn: A. I. L. Payne, K. H. Brink, K. H. Mann & R. Hilborn (eds.), Benguela Trophic Functioning. South African Journal of Marine Sci- ence 12. LIPINSKI, M. R., M. A.C. ROELEVELD & C. J. AUGUSTYN. 1991. Feeding studies on Sepia australis, with an assessment of its significance in the Benguela ecosystem. Jn: E. Boucard-Ca- mou (ed.), La Seiche, The Cuttlefish. First International Symposium on the Cuttlefish Sepia, Caen, 1-3 June 1989. Page 395 Centre de Publications de l'Université de Caen, Caen, France: 117-129. Murata, M. 1978. The relation between mantle length and body weight of the squid, Todarodes pacificus Steenstrup. Bulletin of the Hokkaido Regional Fisheries Research Lab- oratory 43:34—-51. [In Japanese, with English summary.] Payn_, A. I. L., C. J. AUGUSTYN & R. W. LESLIE. 1985. Bio- mass index and catch of Cape hake from random stratified sampling cruises in Division 1.6 during 1984. Collection of Scientific Papers 12(II). International Commission for the South East Atlantic Fisheries:99-123. Payne, A. I. L., C. J. AUGUSTYN & R. W. LESLIE. 1986. Results of the South African hake biomass cruises in Division 1.6 in 1985. Collection of Scientific Papers 13(II). International Commission for the South East Atlantic Fisheries:181-196. Payne, A. I. L., R. W. LESLIE & C. J. AUGUSTYN. 1987. Bio- mass index of Cape hake and other demersal fish species in Division 1.6 in 1986. Collection of Scientific Papers 14(II). International Commission for the South East Atlantic Fish- eries:169-191. Payne, A. I. L., R. W. Lestie & C. J. AUGUSTYN. 1988. Re- vised biomass for Cape hake and other demersal fish species in Division 1.6 and the results of the surveys made in 1987. Collection of Scientific Papers 15(II). International Com- mission for the South East Atlantic Fisheries:175-196. Payng, A. I. L., A. BADENHORST, C. J. AUGUSTYN & R. W. LESLIE. 1989. Biomass indices for Cape hake and other demersal species in South African waters in 1988 and earlier. Collection of Scientific Papers 16(II). International Com- mission for the South East Atlantic Fisheries:25-62. ROCHEBRUNE, A. T. DE. 1884. Etude monographique de la famille des Sepiadae. Bulletin de la Societe Philomathique de Paris (7)8:74-122. ROELEVELD, M.A. 1972. A review of the Sepiidae (Cephalopo- da) of southern Africa. Annals of the South African Museum 59(10):193-313. ROELEVELD, M. A. & W. R. LILTVED. 1985. A new species of Sepia (Cephalopoda, Sepiidae) from South Africa. Annals of the South African Museum 96(1):1-18. SANCHEZ, P. & R. VILLANUEVA. 1989. Distribution and abun- dance of three species of cephalopod sepioids in Namibian waters. Collection of Scientific Papers 16(II). International Commission for the South East Atlantic Fisheries:151-160. SANCHEZ, P. & R. VILLANUEVA. 1991. Morphometrics and some aspects of biology of Sepia australis in Namibian waters. In: E. Boucaud-Camou (ed.), La Seiche—The Cuttlefish. Centre de Publications de |’Universite de Caen, Caen, France: 105-115. SHANNON, L. V. 1985. The Benguela ecosystem. Pt. I. Evo- lution of the Benguela, physical features and processes. Jn: Oceanography and Marine Biology Annual Review 23:105- 182. The Veliger 35(4):396-401 (October 1, 1992) THE VELIGER © CMS, Inc., 1992 NOTES, INFORMATION & NEWS Seasonal Abundance of the Small Tropical Sepioid Idiosepius pygmaeus (Cephalopoda: Idiosepiidae) at Two Localities off Townsville, North Queensland, Australia by George D. Jackson! Department of Marine Biology, James Cook University of North Queensland, Townsville 4811, Queensland, Australia Large numbers of the small tropical sepioid Jdiosepius pyg- maeus Steenstrup, 1881, were discovered recently in near- shore and estuarine habitats in North Queensland, Aus- tralia, including along the breakwater systems of the Townsville harbor and marina complex (JACKSON, 1989), but because of the littoral habitat of /. pygmaeus, individ- uals could not be sampled with towed nets. Due to their dark pigmentation and neustonic behavior, individuals of I. pygmaeus could be easily observed and dip-netted by walking along the breakwaters. A visual census technique was therefore used to determine trends in abundance throughout the year. The sepioids collected during the regular sampling episodes were used both for length-fre- quency analysis and for discerning any seasonal patterns in growth (see JACKSON & CHOAT, 1992). This paper reports the results of the visual census technique for mon- itoring the relative abundance of J. pygmaeus at two lo- calities in the Townsville region of North Queensland. Both visual and dip-net census techniques were incor- porated to monitor changes in the numbers of Jdiosepius pygmaeus through time. The study sites were two break- water systems located within 2 km of each other on each side of the Townsville harbor (19°15’S, 146°50’E), which lies within the central Great Barrier Reef province. Al- though the region is tropical, there is a considerable sea- sonal fluctuation in water temperature. Surface water tem- peratures just offshore from Townsville fluctuate between 19.3°C and 30.9°C, with the summer maximum in January and the winter minimum in July (WALKER, 1981). The Townsville western breakwater /marina complex was used for visual monitoring, which consisted of walking along 3.2 km of breakwater and noting the presence of any in- dividuals of /. pygmaeus. The sepioids were left undis- turbed except for a few individuals collected on several occasions for aquarium experiments. Counts were taken approximately monthly from July to December 1987 and then generally fortnightly throughout 1988 and 1989. The Townsville eastern breakwater complex was the Present address: University of Otago, Portobello Marine Lab- oratory, P.O. Box 8, Portobello, New Zealand. site of regular collections of Jdiosepius pygmaeus for the study of seasonal variation in growth. All individuals ob- served were dip-netted along approximately 1.72 km of the breakwater. The sample was obtained by pooling all individuals captured from paired collecting trips, which usually consisted of an afternoon trip followed by one the next morning. Each sample usually consisted of both a low- and high-tide collection. The distance covered during one visual census taken on the western breakwater (3.21 km) was roughly equivalent to the distance covered on the paired sampling trips along the eastern breakwater (3.44 km; z.e., 1.72 km x 2 trips). On the basis of visual counts at the western breakwater, the abundance of Jdiosepius pygmaeus varied considerably, with periods of high abundance interspersed between pe- riods of absence or very low abundance (Figure 1). Gen- erally, very few individuals were observed over the summer months (December to February) during all three years. This pattern is even more accentuated if the total abun- dance from collected specimens on the eastern breakwater is superimposed over the western breakwater visual census (Figure 1). Because sampling was generally taken on alternate weeks at the two breakwaters, monthly abundances are difficult to compare directly. However, during comparable periods, the pattern in abundance was similar at the two break- waters: for example, the drop in IJdiosepius pygmaeus abundance during October-November, followed by a sub- sequent rise in abundance during February 1989. The pattern in abundance for both breakwaters also was similar during the period between July and November 1989. This suggests that J. pygmaeus on both breakwaters is respond- ing in a similar way to some external environmental factor influencing abundance. Although most reports of temperate Jdiosepius species indicate preference for a benthic/seagrass habitat (SASAKI, 1923; BURN, 1959), Idiosepius pygmaeus appears to be predominantly free swimming and littoral. The distribu- tion of I. pygmaeus in shallow water along the Townsville breakwaters bears a close resemblance to the habitat de- scription given by MOYNIHAN (1983:44) for Koror, Palau, in which individuals were captured in shallow water (less than 1 m depth) “over hard and rather bare surfaces, natural rock and coral or artificial concrete and iron” and “floating or swimming high in the shallows near the shore- line on bright sunny days.” Jdiosepius pygmaeus in Aus- tralia was observed most commonly at the surface, and even after considerable disturbance (e.g., missing them with a dip-net) they usually returned to the surface after 1 to 5 min. The regular absence of Jdiosepius pygmaeus in inshore waters during warm periods is not yet fully understood. Notes, Information & News Page 397 160 291 140- Ls 1207 100 5 80 > 60 > 40> NUMBERS OF /. pygmaeus 205 —+— WESTERN BREAKWATER 8 EASTERN BREAKWATER nH (0) I T T T JS IN) 1987 Figure 1 Numbers of individuals of Idiosepius pygmaeus collected along the eastern breakwater and counted along the western breakwater /marina complex. Note that the peak in March 1987 for the eastern breakwater (291 individuals) is not drawn to scale. Idiosepius pygmaeus has been caught in considerable num- bers at night around light traps in 22 m of water, 500 m off Lizard Island (a continental island, northern Great Barrier Reef waters) in early January (McCormick, per- sonal communication) while they are virtually absent from the breakwater habitat at Townsville. The close association of I. pygmaeus with surface waters may be one factor constraining its temporal distribution along the break- waters. High surface water temperature may force /. pyg- maeus to migrate to deeper waters, but since this sampling regime was visual, movement into deeper water would not be detected. Also, juvenile fish were particularly abundant along the breakwaters during the summer periods, and may have increased predation pressures on J. pygmaeus during that period. [diosepius paradoxa in Japan is a prey item for the lancetfish Alepisaurus ferox (KUBOTA & UYENO, 1979). Although J. pygmaeus feeds predominantly on the inshore sergestid shrimp Acetes sibogae australis, which of- ten swarms along the breakwater systems (unpublished data), preliminary monitoring of the seasonal abundance of Acetes did not show any relationship to the seasonal abundance patterns of J. pygmaeus (JACKSON, 1991). Literature Cited Burn, R. 1959. Molluscan field notes—part 3. Victorian Nat- uralist 75:179-181. Jackson, G. D. 1989. The use of statolith microstructures to analyze life-history events in the small tropical cephalopod Idiosepius pygmaeus. Fishery Bulletin, U.S. 87:265-272. Jackson, G. D. 1991. Age, growth and population dynamics of tropical squid and sepioid populations in waters off Townsville, North Queensland, Australia. Ph.D. Thesis, James Cook University of North Queensland, Australia. 221 PP- Jackson, G. D. & J. H. Cuoat. 1992. Growth in tropical cephalopods; an analysis based on statolith microstructure. Canadian Journal of Fisheries and Aquatic Sciences 49:218- 228. KusoTta, T. & T. UYENO. 1979. Food habits of lancetfish Alepisaurus ferox (order Myctophiformes) in Suruga Bay, Japan. Japanese Journal of Ichthyology 17:22-28. MoyninHaNn, M. 1983. Notes on the behavior of Jdiosepius pyg- maeus (Cephalopoda, Idiosepiidae). Behaviour 85:42-57. SasakI, M. 1923. On an adhering habit of a pygmy cuttlefish Idiosepius pygmaeus Steenstrup. Annotationes Zollogicae Ja- ponenses 10:209-213. WALKER, T. A. 1981. Annual temperature cycle in Cleveland Bay Great Barrier Reef province. Australian Journal of Marine and Freshwater Research 32:987-991. Page 398 Conus striatus Survives Attack by Gonodactyloid! by Alan J. Kohn Department of Zoology, University of Washington, Seattle, Washington 98195, USA Gonodactyloid crustaceans (members of the order Sto- matopoda) are masterful predators of gastropod mollusks, despite the strong, protective shells of their prey. To access the gastropod’s soft body, they use highly modified second thoracopods in a remarkably diverse array of shell-frac- turing techniques. These include breaking off the shell apex, holding the shell against the home burrow wall and breaking shell away from the outer lip, and punching a large (5-25-mm across) hole in the dorsum of the shell (CALDWELL & DINGLE, 1976; GEARY et al., 1991). GEARY et al. (1991) document such holes in Recent shells and argue for stomatopods of the genus Gonodactylus and related genera as the cause of similar holes in Pliocene and Pleistocene fossil shells belonging to several prosobranch families. They reported holes of this type most frequently in Strombidae, and search of fossil collections and illus- trations in the literature will likely reveal additional ex- amples. A hole that appears to be about 16 mm across in a Pliocene specimen of 7 ylospira, in the stromboidean fam- ily Struthiolariidae, was recently illustrated (DARRAGH, 1991). Gonodactylus and its relatives punch holes in shells by using the base of the dactylus, which is enlarged, hemi- spherical, and heavily calcified, forming a structure anal- ogous to the ball peen of a hammer head. ‘“‘With this appendage, stomatopods can punch through crab cara- paces, mollusc shells, echinoderm tests, Erlenmeyer flasks, and even the walls of aquaria” (GEARY et al., 1991). I have occasionally observed empty, often worn shells of Recent Conus, particularly C. striatus in Hawaii, with such holes. It was only upon reading the report by GEary et al. (1991) that I understood their likely cause. Of course in the ab- sence of direct observation, the possibility remains of an agent other than a gonodactyloid. Here I report that attacks on gastropods apparently due to gonodactyloids are not inevitably fatal. This is dem- onstrated by a singular specimen of Conus striatus that not only survived such an attack but lived to repair the damage to its shell. It recorded the event by a resulting shell scar of a type that to my knowledge has not previously been The Veliger, Vol. 35, No. 4 reported to occur in nature, and it likely grew by further shell accretion before it was killed and collected by a spec- imen of Homo sapiens. Observations The specimen (Figure 1), in the Staatliches Museum fur Naturkunde, Stuttgart (SMNS), is 76 mm long and 37 mm in maximum diameter. It was collected at Ranong, Thailand, by A. J. da Motta. The shell has about 11 postnuclear or teleoconch whorls. The maximum dimen- sions of the hole are 23 mm in the shell’s axial direction and 19 mm normal to the axis of coiling. It is located in the last whorl, from about 120° to about 185° back from the outer lip or growing edge of the shell when the animal died. The hole was filled by shell material that differs in appearance from the general surface of the normal shell. The outermost layer of normal shell, referred to as layer 1 (KOHN et al., 1979), is usually relatively thin (about 0.6 mm or about 20-25% of total shell thickness in a Conus striatus shell of the size under consideration). Most likely, the hole was filled by shell material secreted by those regions of the mantle responsible for the thickest and strongest middle layer (layer 2) and the inner layer (layer 3) (KOHN ef al., 1979; CURREY & KOHN, 1976). The inner surface of the shell is quite smooth under the repair scar, indicating participation of layer 3 in its se- cretion. On the outer surface of the scar, the shell material is mostly white but contains some irregular brown pigment blotches. The brown pigment in the normal shell along the edge of the hole appears to be in layer 1 and perhaps the outer portion of layer 2. A narrow peripheral band of outer mantle epithelium normally secretes the outermost shell layer (layer 1) (WILBUR & SALEUDDIN, 1983). The absence of its participation in the repair process away from the shell edge may explain the near absence of pigment in the scar. The crystal architecture of the regenerated shell is un- known, as this could not be determined non-destructively. All layers of normal Conus shell are of crossed-lamellar aragonite (KOHN et al., 1979). In other mollusks, repaired shell away from shell edges often differs in structure from normal shell. When repairing such breaks, terrestrial pul- monates replace crossed-lamellar shell with other fabrics, but the only prosobranchs studied, the freshwater meso- Figure 1 Conus striatus Linnaeus, illustrating shell repair following attack attributed to gonodactyloid stomatopod. Specimen from Ranong, Thailand, in SMNS. 76 x 37 mm. A. Dorsal view, showing entire repair scar. B. Ventral view. C. Posterior or apical view, showing absence of shell repairs on shoulders of earlier whorls. D. View into aperture from anterior end, with repair scar (light area) illuminated from below. E. Right laterodorsal view; shell rotated to show jagged collabral repair scar (single arrow) and smooth, collabral line possibly indicating temporary cessation of growth (double arrow). Notes, Information & News Page 399 Page 400 gastropods Viviparus and Pomacea, replace crossed-lamel- lar with crossed-lamellar shell (WATABE, 1983). Although the repair scar I describe here is marked with about 15 growth lines (Figure 1A), it is difficult to discern where it was initiated. The growth lines at the distal limit of the hole (7.e., nearest the outer lip) are small and arcuate, and they overlie the larger, more proximal lines. Because the inner mantle surface may secrete shell material readily only under previously deposited shell, the most likely se- quence of filling of the hole is thus distal to proximal. However, it is possible that repair along the longer, prox- imal lines that parallel the irregular outline of the hole was also initiated at the onset of repair, and that subsequent episodes of parallel shell secretion abutted or slightly over- lapped the earlier shell. Reconstruction initiated from all sides or exclusively from the distal point would likely ex- pose the soft mantle tissue maximally to the environment during the process. However, such exposure may stimulate shell regeneration over the entire exposed mantle (Wa- TABE, 1983). The edges of the repair scar were not filled, so the outside is not flush with the surrounding undamaged shell. This is expected if the repair occurred from the inside by a skirt of mantle tissue too stiff to conform to the edge of the hole. In addition to the repair of the putative stomatopod hole, the Conus striatus shell bears a jagged repair scar about 3- 10 mm (about 4% whorl or 45°) back from the outer lip and generally parallel to it (Figure 1E). It is not possible to ascertain whether this scar occurred before, during, or after the stomatopod attack, or even whether it was caused by a crustacean. However, several aspects suggest that it too could be an outcome of the unsuccessful gonodactyloid attack: (1) Using their 3rd—5th thoracopods, often called maxillipeds, gonodactyloids sometimes wedge their prey against the burrow wall while hammering (CALDWELL & DINGLE, 1976); this could cause damage all along the thin outer lip of the shell. (2) The pattern of concave breaks 7-8 mm long (Figure 1E, single arrow) is not inconsistent with the effect of hammering by a gonodactyloid dactylus. (3) The color pattern of the shell changes abruptly at this scar. The characteristic brown spiral lines (for which Lin- naeus named the species) become more distinct, and gen- erally less pigment is deposited, in shell material secreted after this event (Figure 1E). That this suggests a traumatic experience is supported by two lines of anecdotal evidence. I have frequently observed changes in color pattern in shell secreted after repair of outer lips broken in unsuccessful attacks on Conus by crustaceans. And I have seen similar pattern shifts in C. striatus transported from Guam to Seattle and subsequently fed fishes from Puget Sound (Cot- tidae) rather than their native diet. If this interpretation is correct, the shell removed in the attack would have been about 75-135° from the outer lip at the time of injury. On the other hand, there is in the specimen reported here a smooth collabral line about 15 mm back from the outer lip (Figure 1E, double arrow)—.e., earlier in the The Veliger, Vol. 35, No. 4 animal’s life than the repaired lip just described. The white ground color of shell deposited proximal to this line is densely but irregularly suffused with pink; thereafter, pink pigmentation is sparse, both before and after the jagged collabral scar. This line could possibly indicate the position of the outer lip at the time of the attack, but its uniformity suggests that it represents a growth cessation rather than a response to injury. Following discovery of the shell described here, I ex- amined the entire collection of Conus striatus in the SMNS (n = 107, most also collected by Mr. da Motta) for evidence of shell repairs following attempting predation. I found no other evidence of stomatopod attacks. Three of the 30 additional specimens from western Thailand had repair scars on the last whorl, as did 3 of 76 from other localities throughout the tropical Indo-Pacific region (one each from Samoa, Fiji, and Réunion). These were likely due to at- tacks from crabs, but this cannot be determined with cer- tainty. The average number of 0.06 repairs per last whorl in C. striatus is lower than in most other species surveyed. VERMEIJ (1989) found an overall mean of 0.29 but with wide variation (range 0-1.0; SD 0.25) in 59 samples of 14 other Indo-Pacific Conus species. In two samples of C. striatus in the B. P. Bishop Museum, Honolulu, from Guam (n = 14 shells) and Maui (n = 15 shells), VERMEIJ (in litt.) noted averages of 0.3 and 1.3 repair per shell, respectively (overall average 0.8). Discussion and Conclusion Marine gastropods frequently repair damage to the outer shell lip following unsuccessful predation attempts (e.g., VERMEIJ, 1987), but repair of mid-shell holes is poorly documented and appears to be rare, in contrast to the case in some terrestrial and freshwater forms (WATABE, 1983). Adult Conus frequently survive attacks by crabs that at- tempt to peel off the outer lip but are unable to break the thicker, multi-layered shell farther back. In the direction that the crab claw must apply force (normal to the length of the outer lip), the primary lamels of layer 1 may be pulled apart from each other. However, layer 2 is nearly twice as thick, and its crystals are oriented so that the tightly packed secondary lamels or laths would have to be broken across their long axes. Because of these architec- tural features, the crossed-lamellar structure of C. striatus is highly anisotropic. Its bending strength in the direction of attempted breakage is about 200 MNm * but only about 70 MNm” in the direction at right angles to this. This enables the animal to survive by withdrawing into the shell beyond the limit that the crab can break. It later repairs the shell, leaving a record of the unsuccessful predation as a scar (CURREY & KOHN, 1976). Several factors probably allowed the unusual instance reported here of survival of a gastropod following severe damage to the shell in the form of a large hole punched some distance from the outer lip during a presumed sto- Notes, Information & News matopod attack; all remain speculative but are amenable to experimental test. Whether or not the predator was itself preyed upon or otherwise driven away while sub- duing the Conus is of course unknown. It is unlikely that the Conus retaliated by stinging the predator, because the Conus venom apparatus is an offensive, not defensive weap- on system, and C. striatus in particular usually responds to handling by withdrawal into the shell, not by extension of the proboscis. If the attempted predation was simply unsuccessful, it is unlikely that the Conus had eaten re- cently. Because C. striatus kills and swallows large fishes whole, its foregut is often distended for some hours after feeding (KOHN, 1956). This prevents retraction of the body within the shell and probably would have rendered the Conus completely susceptible to predation by the stomato- pod without shell breakage. In this case, however, the Conus may have been able to withdraw the mantle and body within the shell beyond the position of the stomato- pod’s hammering activities, so that they did not damage its soft tissues. In this case a typical gastropodan outlook on life—when danger threatens, withdraw into your shell; things will improve by and by—appears to have served Conus striatus well and to have caused the “thug of crus- taceandom” (SCHMITT, 1965) to meet its match. Acknowledgments I thank Hans-Jorg Niederhofer for providing research facilities at SMNS, Gary Vermeij for providing unpub- lished data, and both for helpful discussion. Page 401 Literature Cited CALDWELL, R. L. & H. DINGLE. 1976. Stomatopods. Scientific American 234(1):80-89. Currey, J. D. & A. J. KOHN. 1976. Fracture in the crossed- lamellar structure of Conus shells. Journal of Materials Sci- ence 11:1615-1623. DaRRAGH, T. 1991. A revision of the Australian genus 7 ylos- pira Harris, 1897 (Gastropoda: Struthiolariidae). Alcheringa 15:151-175. Geary, D. H., W. D. ALLMON & M. L. REAKA-KuDLA. 1991. Stomatopod predation on fossil gastropods from the Plio- Pleistocene of Florida. Journal of Paleontology 65:355-360. Koun, A. J. 1956. Piscivorous gastropods of the genus Conus. Proceedings of the National Academy of Sciences, USA 42: 168-171. Konn, A. J., E.R. Myers & V. R. MEENAKSHI. 1979. Interior remodeling of the shell by a gastropod mollusc. Proceedings of the National Academy of Sciences, USA 76:3406-3410. ScHMITT, W. L. 1965. Crustaceans. University of Michigan Press: Ann Arbor. VERMEIJ, G. J. 1987. Evolution and Escalation. Princeton University Press: Princeton. VERMEIJ, G. J. 1989. Interoceanic differences in adaptation: effects of history and productivity. Marine Ecology Progress Series 57:293-305. WatTABE, N. 1983. Shell repair. Pp. 289-316. In: K. M. Wilbur (ed.), The Mollusca. Vol. 4. Academic Press: New York. WILBUR, K. M. & A. S.M.SALEUDDIN. 1983. Shell formation. Pp. 235-287. In: K. M. Wilbur (ed.), The Mollusca. Vol. 4. Academic Press: New York. 7 limpontant Notice: ~ Change of Business Address for The Veliger Effective immediately, all business correspondence, in- cluding subscription orders, membership applications, payments for them, changes of address, and reports of missing issues, should be sent to: Dr. Henry W. Chaney Santa Barbara Museum of Natural History 2559 Puesta del Sol Road Santa Barbara, CA 93105 FAX: (805) 963-9679 To our subscribers who might have been inconvenienced while our business affairs were confused or neglected, we offer our apologies. To assist us in meeting your needs while our subscrip- tion and back issue database is being debugged, please send reports of missing issues directly to Dr. Chaney. This will avoid duplicate requests (e.g., from both a subscription agent and you for the same subscription) and speed the processing of your order. This change of address is for the Business Office only. Manuscripts being submitted for publication in The Ve- liger should be sent to the editorial office, as before (see inside cover for more information). Page 402 The Veliger, Vol. 35, No. 4 Collections at the California Academy of Sciences The building housing the Department of Invertebrate Zo- ology and Geology at the California Academy of Sciences is scheduled to undergo structural modifications between Fall 1992 and Summer 1993. This project is intended to strengthen the building against potential earthquake dam- age. During this period most of the non-type holdings of the department will not be accessible for study. However, all type specimens and portions of the non-type collections will be available during the entire project. If you wish to borrow specimens for study, please request them prior to 1 November 1992. Requests for loans of non-type speci- mens made between 1 November 1992 and 1 August 1993 may not be filled due to specimen inaccessibility. Similarly, departmental facilities for visiting scientists will be very limited between 1 November 1992 and 1 August 1993. All questions and requests for specimen loans should be addressed to: Robert Van Syoc Collection Manager, Invertebrates 415-750-7082 Jean DeMouthe Senior Collection Manager, Geology 415-750-7094 Requests may be mailed to either collection manager at the departmental address or sent via FAX (415-750-7090). International Commission on Zoological Nomenclature Comment or advice on applications to the ICZN is invited for publication in the Bulletin of Zoological Nomenclature and should be sent to the Executive Secretary, ICZN, % The Natural History Museum, Cromwell Road, London SW7 5BD, UK. The following applications and opinions were published on 25 June 1992 in Vol. 49, Part 2, of the Bulletin: Case 2801—Potamolithus Pilsbry, 1896 (Mollusca, Gas- tropoda): proposed confirmation of P. rush Pilsbry, 1896, as the type species. Case 2526—Strombiformis albus Da Costa, 1778 (currently Melanella (Balcis) alba; Mollusca, Gastropoda): pro- posed conservation of the specific name. Opinion 1676—Lepidomenia Kowalevsky in Brock, 1883 (Mollusca, Solenogastres): Lepidomenia hystrix Mar- ion & Kowalevsky in Fischer, 1885, designated as the type species. Opinion 1677—Haustator Montfort, 1810 (Mollusca, Gastropoda): conserved. Opinion 1678—Helicarion Ferussac, 1821 (Mollusca, Gastropoda): conserved, and Helicarion cuvieri Férus- sac, 1821, designated as the type species. Opinion 1679—Kobeltia Seibert, 1873 (Mollusca, Gas- tropoda): Arion hortensis Ferussac, 1819, confirmed as the type species. Information for Contributors Manuscripts Manuscripts must be typed on white paper, 82” by 11”, and double-spaced throughout (including references, figure legends, footnotes, and tables). If computer generated copy is to be submitted, margins should be ragged right (7.e., not justified). To facilitate the review process, manuscripts, including figures, should be submitted in triplicate. The first mention in the text of the scientific name of a species should be accompanied by the taxonomic authority, including the year, if possible. Underline scientific names and other words to be printed in italics. Metric and Celsius units are to be used. The sequence of manuscript components should be as follows in most cases: title page, abstract, introduction, materials and methods, results, discussion, acknowledgments, lit- erature cited, figure legends, figures, footnotes, and tables. The title page should be on a separate sheet and should include the title, author’s name, and address. The abstract should describe in the briefest possible way (normally less than 200 words) the scope, main results, and conclusions of the paper. Literature cited References in the text should be given by the name of the author(s) followed by the date of publication: for one author (Smith, 1951), for two authors (Smith & Jones, 1952), and for more than two (Smith et al., 1953). The “literature cited” section must include all (but not additional) references quoted in the text. References should be listed in alphabetical order and typed on sheets separate from the text. Each citation must be complete, with all journal titles unabbreviated, and in the following form: a) Periodicals Cate, J. M. 1962. On the identifications of five Pacific Mitra. The Veliger 4:132-134. b) Books Yonge, C. M. & T. E. Thompson. 1976. Living marine molluscs. Collins: London. 288 pp. c) Composite works Feder, H. M. 1980. Asteroidea: the sea stars. Pp. 117-135. In: R. H. Morris, D. P. Abbott & E. C. Haderlie (eds.), Intertidal Invertebrates of California. Stanford Univ. Press: Stanford, Calif. Tables Tables must be numbered and each typed on a separate sheet. Each table should be headed by a brief legend. Figures and plates Figures must be carefully prepared and should be submitted ready for publication. Each should have a short legend, listed on a sheet following the literature cited. Text figures should be in black ink and completely lettered. Keep in mind page format and column size when designing figures. Photographs for half-tone plates must be of good quality. They should be trimmed off squarely, arranged into plates, and mounted on suitable drawing board. Where necessary, a scale should be put on the actual figure. Preferably, photographs should be in the desired final size. It is the author’s responsibility that lettering is legible after final reduction (if any) and that lettering size is appropriate to the figure. Charges will be made for necessary alterations. Processing of manuscripts Upon receipt each manuscript is critically evaluated by at least two referees. Based on these evaluations the editor decides on acceptance or rejection. Acceptable manuscripts are returned to the author for consideration of comments and criticisms, and a finalized manuscript is sent to press. The author will receive from the printer two sets of proofs, which should be corrected carefully for printing errors. At this stage, stylistic changes are no longer appropriate, and changes other than the correction of printing errors will be charged to the author at cost. One set of corrected proofs should be returned to the editor. An order form for the purchase of reprints will accompany proofs. If reprints are desired, they are to be ordered directly from the printer. Send manuscripts, proofs, and correspondence regarding editorial matters to: Dr. David W. Phillips, Editor, 2410 Oakenshield Road, Davis, CA 95616 USA. CONTENTS — Continued Two new species of Helminthoglypta (Gastropoda: Pulmonata) from southern California, with comments on the subgenus Charodotes Pilsbry BARRY ROTH AND E-G:cHOGHBERG 25. Fut M0070 he ens ee ee ane Seasonal variation in the reproductive organs of two populations of Caracolus caracolla (Linné) (Pulmonata: Camaenidae) in Puerto Rico PATRICIA MARGOS, 25 (592004) circ Hine See oT ete eee Ss ee ose ee a eared ce The ecology of coquina clams Donax variabilis Say, 1822, and Donax parvula Philippi, 1849, on the east coast of Florida ERIK BONSDOREF AND’ WALTER G. NELSON ="4). 365 2455022 eee “Solen rosaceus”—three species RUDO) VON; COSEL!) 9) 52s 2 Ore SoM th aed ae Nene ts Notes ee ae nha a Identification of monosaccharides in hydrolyzed Nautilus shell insoluble matrix by gas chromatography /mass spectrometry MICHAEL J. S. TEVESZ, STEPHEN F. SCHWELGIEN, BETH A. SMITH, DAVID G. HEHEMANN, ROGER W. BINKLEY, AND JOSEPH G. CARTER .......... ' First study on the ecology of Sepia australis in the southern Benguela ecosystem M. R. Lipinski, M. A. COMPAGNO ROELEVELD, AND C. J. AUGUSTYN_.... NOTES, INFORMATION & NEWS Seasonal abundance of the small tropical sepioid Jdiosepius pygmaeus (Cepha- lopoda: Idiosepiidae) at two localities off Townsville, North Queens- land, Australia GEORGE Di JACKSON) (= oe Su) 5 ONE uh Boi oaks oe tees corned ht eee ee Conus striatus survives attack by gonodactyloid! AIAN: \SSIKOHING Wii i se stave gene co hace ee a eee aaa Re ares nis PASH 338 347 358 366 381 384 _— mA — _ =» = = <= % us = Seyi = < = < = < = < > XY =r} z + — 2 \y : 5 = : : z 3 Sf URNS % AOD 77) 22) ped ” 72) 7 | =I WY \S: Oo ae (2) < (2) Ge fe) x \y E AS’ 2 = Zz E Zz = Zz E es s = 5 : 2 2 : a = INS S31 YVYa!I1 LIBRARIES SMITHSONIAN INSTITUTION —NVINOSHLINS - LIBRARIES SN us % ul z & w & = [oa = fac i, ot jag a a at [oad yer) sR) = Jo 1E) ED? WERE: ve 5 = o9 3 co = or = me = Pw Ss o = 4 Ss J S = 2 a 2 a IAN INSTITUTION NOILOLILSNI NVINOSHLINS S3Iy¥vUudtt lel BRARI FS_SMITHSONIAN INSTITOTIG WEN via Lie ae By s Zz = = = Pa) 2 = Ss o = “0 = = cf Cy, hus 5 2 5 = 2A = . 5 a = — = a EK Gz a = 20 im a ree iD m 2 ti m 2 m Bs z ye z wo = 7) = o IWS wo 1Yyvuad Iu BRARI ES SMITHSONIAN _ i NOILALILSNI_ NVINOSHINS Sa luyvuad Mtl B RAR J ES Sh = _< RE = = = <= = = Ke = = 4 car = = \ . z aa : z ay 5 = SQ f, eS 2 g 2 g a gi 2X | 2 E 2 E 2 = 2° = —- = F » 8 = . 8 a S : 7) a IAN_ INSTITUTION NOILALILSNI NVINOSHLINS S3IYVYSIT_ SMITHSONIAN _ INSTITUTION NOILNLILSNI Nb Ss n => Ree w ; = wo z= ” = é : 5 & ° j : 8 : j < % (os i tes © S oc = a c a = = = = fe) = ro) = o = fo) = fe) x =) 2 moreue z a) Zz = Zz WS_SAIlY¥vVud!it LIBRARIES SMITHSONIAN_INSTITUTION NOILALILSNI NVINOSHLINS S3JIYVYUSIT LIBRARIES SN , = 35) = oY, 2 S = Le, =) fal > KE > & Wifi = 5 UR ANREE E = 2 =. UY fp ® ie = SNS F = 5a : o g nee Z Coe ee p 2 IAN INSTITUTION NOILALILSNI aU Ui eS UMASS UA) INS UUMUTLOI e a = wh, = = "Wr, = = 2 z < Je Nei Jas@ri ZR: ; s\n i Ye 3 ) AW S YY flu: = S = 2 = AS S z= \ = re 3 = =e se 5 Ws woz law” G11 LIBRARIES SMITHSONIAN INSTITUTION, NOILALIESNI _NVINOSHLIWS (S30 ¥ Sa) iiia ate ae z 2) P = z z uu — ud 5 uu Oo ad Ww z 4 = VG," : + Qe =z Z E: \ Ps c S Y, Wie = = cS Nw te E a s m Sy ae WS Sal = = . WS = = oe = Zz 4 Z = = 4 2 a IAN INSTITUTION NOILALILSNI NVINOSHLINS S3IYVUYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NV c z a z = 5 = S z ee) 2 2) = *o = wo yx 5 2 ces 3 5 Es 5 2 > = = = > Es > 2 =e D — Po) — ee) a ine A a a = cS at - as [— = = (7p) (22) i Z a pe B EA gine z ae WS SAIYVUYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S31YVYUdIT LIBRARIES SW ” Zz Sek (27) z= Ro (a) Zz w Ce o = ae = = < = = z = Ws § A eh fi dir = z si z = 5 WS = IGM ARS | 2 gj : gy ER 2 Uy * » 2 E 2 E z E » 2a 2 a Ae a 2 a 2 77) % 2 . AN_INSTITUTION NOILALILSNI NVINOSHLINS S3I1YVUdII SMITHSONIAN _INSTITUTION NOJLALILSNI NV e bs Ms a os sh z a S ree ae 2 a 2 a; 2 2 : &@ MS = 2 = 6 : a = z GF QBN =| as = < a ‘ = a a ON is Cc oe Cc = c a s 4 SS es a a fon) a ) a rs pa ) bd Z fe) = re) Fe aad 2 So wap) z i) ae — z= MS JSSTeVUGIT LIBRARIES SMITHSONIAN, INSTITUTION | NOILOLILSNINVINOSHLINS “S31YVUIT LIBRARIES SW 2) Soll =_ san, ro} Wy , = SONG 2)
    Pile by ‘S) VoL See 0 NG 2) = Le > Lapa, £ GY = GAR) 2 LAR= a. 2 A= ® 2 = 14, = i 2 z Nl z 2 Ss = = C =| z =| z =) z 9 -: WX : LS z S z oo SS ° MWS 6 ise ro) Be re) = Ea 2 NOG" 2 E 2 = = = 7 6S Syl : : pe se AE N_INSTITUTION | NOILALILSNI_ NVINOSHLIWS VEGtT_LIBRARIES INSTITUTION | NOLLALILSNI_NVI la u G 3 Go CE a WLS [= ox = = st ow = = a = = cc = oc = 5 rr = = me a 4 ro} = fe) . ro} en 5 os) ro) ;z aj = z a z a 2 S$ SJINVYUSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVUSIT LIBRARIES SMI 2 rc z - 2 ; - i z Sine = ‘iS = = : 0 = 0 ° >» o = r= a > ¢Y a 5 A E Po] [= of e S 2G ff * : 5 2 3 Py ie = rs Gy = - = i= oes EG 7) m no Wy m “” m 7) m wo % NCINSTITUTION | NOILNLILSNI_NVINOSHLINS S3IYVUEIT LIBRARIES SMITHSONIAN INSTITUTION NOILMLILSNI_ NVI < K = z = EG F si 2 2 ne NX: Fj = Sl & SYS o . 5 ee : za EGR 2 : SZH ARS 3 NEEL = » 2 = 2 E See 2 Uy «= 3 1 alle: Se oe We Cis S Saiuvugia LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SaIuvYsIT LIBRARIES SMI ARIES ES wate Ne & \ \N SMITHSONIAN Gy, & NVINOSHLINS S3IYVYE!IT LIBRARIES RNS SK SS Z NOILNLILSNI LIBRARIES LIBRARIES NOILALILSNI N_INSTITUTION NOILNLILSNI NVINOSHLINS S3I1YVYdIT LIBRARIES INSTITUTION NOILALILSNI_NVII ‘ ‘ INSTITUTION NOILNLILSNI S3IYVYEIT_LIBR Saiuvualy Salyvyal INSTITUTION NOILNLILSNI SJ'YVUGIT LIBRARIES SMITHSONIAN INSTITUTION INSTITUTION S S3SI¥VYEIT LIBRARIES SMITHSONIAN we SMI *) NVINOSHLINS S3IYVYEITLIBRARI NVINOSHLIWS NVINOSHLIWS NVINOSHLINS SMITHSONIAN NVINOSHLIWS N_INSTITUTION LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ Nv LIBRARIES SMITHSONIAN LIBRARIES SMITHSONIAN = > > 7) = = ‘ a @ & “ a a ‘2 z = e ae = = 2 - ee 5 S (e} y Oo Oo a 4 2 a $ Zz ST IEVUEIT LIBRARIES _ SMITHSONIAN ~NOHLNLILSNI NVINOSHLINS SJIYVYGIT LIBRARIES SMI Lo 2 ow S) S a Oke aS = 5 ke ~ = = = D => = Rs = > EY D = Bo Ee = - @ - % g mM & ” wo ieee ut = o = cs G z HA z NL INSTITUTION | NOLLOLILSNI_ ss LIBRARIES, SMITHSONIAN INSTITUTION NOILOLILSNI_ NVI SK = = Sy ee = a ae ie z atk & Ye = = = oy ¢ ce = —_ : = ‘4 anne! SQN FE 5 ey Ay ae = a Ae z a5 (o) ri O % Hp Se 8 22) wn . wn 2 ne E 2 “iy 2 i 2 fe a ye B ra ra Se a 2 a S S3iy¥vygly SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IuvVYaIT LIBRARIES SMI (7p) = (7p) = w — = 2 Fe en wo 2) ca wv w 7, 2 5 RSS us a wl w = ow. = oc SB! OYA a ma < z NOILALILSNI _NWINOSHLINS ~°3 luvyd Ja LIBRARI ES_ SMITHSONIAN INSTITUTION “ NOILNLILSNI NVI Sey eo NON ONL Kasim Se OR sta et aN) Ol) eee a en) aNGRO LO) | Seon ce z = : Sous SMITHSONIAN INSTITUTION LIBRARIES a 2 PRIETO EPA NIST IN NIT PETERS iy dae > | be SS Saree sei Mak eT rw 2bay Faaich Dew erare B MOIS He le : ; at aes SSssSeleratasnasensansass too edsscis esau wees : 2 ee Se ees Z : TE Mee Siesta cep r -dyacds RatioPAe MEIN P SEMA SDSTIANES Acad ARAL TES aad Sand sAaArhcorsietsy Suef 1 haz fA we oad +e 4 ere ree at ee ee SPARE SI REATHIOLEM PE PP STA Moes Senta Soe See ae : : rae 3 9088 00842 6991 4.03. ew rie teieny Pitre ey ened ES ~ : : hid th amdahrs mento At DEawaw omnes ee thee ners Send PEE ye "SPAM >: + Ba Se densn oor ea bees ARIE AL = Sopot bes an se eee eaters ap SEAS ANGE So aretahinan ov AOM_ anh =o Bor ceay noe ntesntern Sra spetior eng SHs ARRAN ies shragena stews taMwow ahesezed in aes peer ed Sha lecice totetntientorasoes: Siceneamie abzrrie cab hinay sas PHL SNL ATT mea fia Dat DOaChe Ee ce Deh aL eA eh barns. Sula tots é Dee tnt oehPerent Hate weeed ede nisbbats SRO ere dies LOM tees pene ah te gabette seed Ae Renee ee E Chit Et Saree Far ahern fon parspreet emeretee pete nr Ser sop ae dae ran gedaditn Papen she Roh im aes te Aniflio tase cated ee pelobsindie watefi cw nt eb esehe tad Toa Ord 2 PF eiesiure Cees oh eh Thee Bae YUL a Sekt Ceuae Wi pare resets oh SS tesre eee t Emr eee wey Seve SL atinaehe ce epvade teohead o4L ck miblniate nd tatbdeeete } nenresueas sarge are ree re meaty Sete cetera pyaecage wegeete ayy Fe aN enoretes rire Patera 2 y Cyn yey TUNG Gey era Uw eae a ned ™ % rou rqrareeumary = Pe waay By PUENTE UES Cet MUM geyeye ares tay WIN : : Pieterteirerl (rere ter Bratecne ery Totes asey baveuye pe br eran io \ svrapven Pere roey 8" $ar: - Teor teietay He bey Eww eee om eee Riu ey Te : i hare Taare SUP S PY ERED on eve kr py 9 iy rusute eee tert terest BAO Eres, yiviatwereniefese en ye care Teh Ua rere ne tery oy Vaio wy UN aut Papeete * Perry vib ieee) WINE ive 5 Py mererseeurery sense net eM Gased sca ic watye Soe e WEL ETy 8 ry Fa 4) WOR PNSEMEURTY EWP PTTL BAT K aI Yr Za yaa tag tay Tes eae Tro Ey Aw Urey oer y ary ss wp? WLS Ereae ss Hee re rene ari iter eee treme Te) Tera Ty UENO MOTE DENI wg ORES ED ATEN NTY NG Rate ng reverses yiveye? 5 rita hk bal TA 1 6 ey get) "legate ina EPI IE ILO TILIA TLE VE NTU IC A SEER ETE NS OD pe Tritt ewes tat eae Cte rere 7 61 OGTR TE OTE THOM ITS On TET NY 6 TET OI 16 RET ERTE 1 EO id latte eae leveares A, “7 ier q rhirLtrer i eal ears , ener y ye sT ete eNTy)t Paryheerecysel ola ss rata arzei ry! tale vie & et td ae Leah Ap Mele alan bee teak ok pe wd tad Lee le tk a reaps wy eT FI PLT FIOM TS VFI LE WATTLE GIR VLG re og TENA YOY ¥ < { Payer ned; deer Cae tee ter ee Win cues o-rege ty) t Hr MES GMOTASTES* Fs my Oy EES! prot G8 Tt YL GIES ED” vr PUFEFS WE TIIT GUSTER IF t Darren nereniat (er Fata a a Ak sei WaRrtivsern BW eo EPSP RAWAL TT, PEST 7” weeren wr Ttrese ares Tye ane aleeghgreer POT OD f “699 STEOPOOTSP SY SILT OF! OT ; 47, SUID VIF Ts ry 8 istic ca erties rhestteit SOLPT PULTE COTA DPB OIG PD STN TS wiry te ee erire rs t7ely ; yoy i vee BOC Eagan tad beara ir Srtug a> Takes eons WWte a sevset 4s L4 SDT TEGIE OY ATHY G2 * BLE" STFU A ig erertaysetesart 1 8tF TET EP ATTN TY es RIRI PAT ITO T 177 19 9 7GTE Im rT 2m wai AIS TA: ty brie mera F084 ert eI Te PAIITIIE GAS TMI TNS AMBIT OY TTL = Fists govt epiee see Ls eats eT eNTTIT eee Foamy FOOT AE fie 5 cot buat mp mate Peak 4 Dag thle ax saad